You are on page 1of 484

CONTAMINANT

HYDROCEOLOGY

C. IV. FETTER

Contaminant Hydrogeology

Contaminant Hydrogeology

C. W. Fetter
Department of Geology
University of WisconsinOshkosh

Macmillan Publishing Company


New

York

Maxwell Macmillan Canada


Toronto
Maxwell Macmillan International
New York

Oxford

Singapore

Sydney

Editor: Robert A. McConnin


Production Editor: Sharon Rudd
Art Coordinator: Peter A. Robison
Text Designer: Debra A. Fargo
Cover Designer: Robert Vega
Production Buyer: Pamela D. Bennett
Illustrations: Maryland CartoGraphics Inc.

This book was set in Garamond by Syntax International and was printed and bound by Book
Press, Inc., a Quebecor America Book Group Company. The cover was stamped by Book Press,
Inc., a Quebecor America Book Group Company

Copyright 1993 by Macmillan Publishing Company, a division of Macmillan, Inc.


Printed in the United States of America
All rights reserved. No part of this book may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopy, recording, or any information storage
and retrieval system, without permission in writing from the Publisher.
Macmillan Publishing Company
866 Third Avenue
New York, New York 10022
Macmillan Publishing Company is part of the
Maxwell Communication Group of Companies.
Maxwell Macmillan Canada, Inc.
1200 Eglinton Avenue East, Suite 200
Don Mills, Ontario M3C 3N1

Library o f C o n g r e s s Cataloging-in-Publication Data


Fetter, C. W. (Charles Willard)
Contaminant hydrogeoIogy/C. W. Fetter,
p.
cm.
Includes bibliographical references and index.
ISBN 0 - 0 2 - 3 3 7 1 3 5 - 8
1. Water, UndergroundPollution. 2. Water, UndergroundPollutionUnited States.
3. Transport theory. 4. Hydrogeology.
I. Title.
TD426.F48
1992
628.1'68dc20
Printing: 1

92-17787
CIP
5

Year: 3

This book is dedicated to my parents,


C. Willard Fetter and Grace Fetter.

Preface

W h e n I c o m p l e t e d the s e c o n d edition of Applied Hydrogeology, I realized that it provided only the barest of introductions to what is o n e of the m o s t fascinating aspects of
hydrogeology, the o c c u r r e n c e and m o v e m e n t of dissolved and n o n a q u e o u s phase contaminants. Consulting w o r k that I was doing also demonstrated that to understand fully
the distribution o f contaminants b e l o w the water table o n e must c o n s i d e r the movem e n t of soil moisture and contaminants in the vadose z o n e . As n o n e of the standard
t e x t b o o k s present advanced topics of solute m o v e m e n t and retardation in b o t h the
saturated and vadose z o n e as well as the o c c u r r e n c e and m o v e m e n t of n o n a q u e o u s
phase liquids, I think that there is a place for an advanced t e x t b o o k on contaminant
hydrogeology.
In a very real s e n s e this n e w b o o k is a sequel to Applied Hydrogeology. T h e r e is
almost no overlap b e t w e e n the two b o o k s ; although s o m e material n e e d e d to be repeated to lay the logical foundation for the advanced c o n c e p t s presented in this b o o k .
Contaminant Hydrogeology is intended to be a t e x t b o o k for a graduate-level c o u r s e in
mass transport and ground-water contamination. Such a c o u r s e might be taught in departments such as geology, civil engineering, geological engineering, or agricultural
engineering. In o r d e r to obtain the fullest benefit from such a c o u r s e , the students
should have c o m p l e t e d a c o u r s e in geohydrology or hydrogeology. B a s i c knowledge of
physics and chemistry is n e e d e d to understand the c o n c e p t s p r e s e n t e d herein.
In addition to its utility as a t e x t b o o k , Contaminant Hydrogeology will be a valuable reference b o o k for the working professional. B o t h solved e x a m p l e p r o b l e m s and
c a s e histories are presented. T h e r e is a mixture of the theoretical and the practical.
Chapter 1 presents an overview of ground-water contamination and a review of basic
mathematics. T h e theory of mass transport in the saturated z o n e is p r e s e n t e d in Chapter 2. T o p i c s include advective-dispersive theory, stochastic transport theory, and description of solute flow using fractals. Retardation and attenuation of dissolved solutes
is covered in Chapter 3, whereas Chapter 4 introduces flow and mass transport in the
vadose z o n e . T h e distribution and m o v e m e n t o f n o n a q u e o u s phase liquids b o t h a b o v e
and b e l o w the water table is discussed in Chapter 5. T h e reactions of inorganic c o m p o u n d s dissolved in g r o u n d water is the topic of Chapter 6. Chapter 7 contains an
overview of organic chemistry and an exhaustive l o o k at biodgradation of organic
c o m p o u n d s in the ground. Chapter 8 contains "how-to" information on conducting

vii

Preface
field investigations to install borings and monitoring wells as well as collecting soil, soilwater, and ground-water samples. T h e latest information on site remediation is found
in Chapter 9.
In a b o o k of this nature there are a very large n u m b e r of variablesfar m o r e
than can be a c c o m m o d a t e d by the 26 letters of the English and the 24 letters of the
G r e e k alphabets. Many variables are indicated by symbols that are a combination of
English a n d / o r G r e e k letters. A variable is defined w h e r e first used in a chapter and
then is listed in a table of notation at the e n d of the chapter. In order to a c c o m m o d a t e
the large n u m b e r of variables in the b o o k , the meaning of s o m e symbols changes from
chapter to chapter. Although this is not a desirable circumstance, it s e e m e d preferable
to such tactics as also utilizing the H e b r e w and Russian alphabets. In many cases, if the
reader g o e s to the original literature cited in the text, the notation of the original article
will not be the s a m e as that used in this text. This was necessary to have consistency
within the text.
Units of measurement have b e e n abbreviated in the text. Appendix E contains a
key to these abbreviations.
1 am grateful to all w h o helped with this project. T h e following individuals provided helpful reviews of chapter drafts: J e a n M. Bahr, University of Wisconsin-Madison;
Robert A. Griffin, University of Alabama; J a m e s I. Hoffman, Eastern Washington University; Martinus Th. van G e n u c h t e n , U.S. Department of Agriculture Salinity Laboratory;
Stephen Kornder, J a m e s River Paper Company; Garrison Sposito, University of California,
Berkeley; and Nicholas Valkenburg, Geraghty and Miller, Inc. Peter Wierenga, University
of Arizona, provided information on measuring soil-moisture tensions and S h l o m o
Neuman, University of Arizona, furnished me with a c o p y of Mualem's Soil Property
Catalogue. Mary D o m m e r prepared the manuscript, and Sue Birch provided s o m e of
the figures.

C. W. Fetter

Contents

Chapter

One

Introduction

1.1 -

Ground Water as a Resource

1.2

Types of Ground-Water Contaminants

1.3

Drinking-Water Standards

11

1.4

Risk and Drinking Water

1.5

Sources of Ground-Water Contamination

14
15

1.5.1

Category I: Sources Designed to Discharge Substances

1.5.2

Category I I : Sources Designed to Store, Treat and/or Dispose of

1.5.3

Category III: Sources Designed to Retain Substances During Transport

1.5.4

Category IV: Sources Discharging Substances as a Consequence of Other

1.5.5

Category V Sources Providing a Conduit for Contaminated Water to

Substances

19

Planned Activities

25
27

Category VI: Naturally Occurring Sources Whose Discharge is Created


and/or Exacerbated by Human Activity

28

1.6

Relative Ranking of Ground-WaterContamination Sources

1.7 '

Ground-Water Contamination as a Long-Term Problem

1.8

Review of Mathematics and the Flow Equation


1.8.1

25

Enter Aquifers
1.5.6

16

Derivatives

29
31

32

32

1.8.2

Darcy's Law

1.8.3

Scaler, Vector, and Tensor Properties of Hydraulic Head and Hydraulic


Conductivity

35
35

1.8.4

Derivation of the Flow Equation in a Deforming Medium

1.8.5

Mathematical Notation

References

37

40

41

ix

Contents

Chapter T w o
Mass T r a n s p o r t in Saturated Media

43

2.1

Introduction

2.2

Transport by Concentration Gradients

2.3

Transport by Advection

2.4

Mechanical Dispersion

2.5

Hydrodynamic Dispersion

2.6

Derivation of the Advection-Dispersion Equation for Solute Transport

2.7

Diffusion versus Dispersion

2.8

Analytical Solutions of the Advection-Dispersion Equation


2.8.1

43

49
51

56

56

2.8.2

Boundary and Initial Conditions

2.8.3

One-Dimensional Step Change in Concentration (First-Type

2.8.4

One-Dimensional Continuous Injection into a Flow Field (Second-Type

2.8.5

Third-Type Boundary Condition

2.8.6

One-Dimensional Slug Injection into a Flow Field

2.8.7

Continuous Injection into a Uniform Two-Dimensional Flow Field

2.8.8

Slug Injection into a Uniform Two-Dimensional Flow Field

Boundary)

56

57
58

2.9

Effects of Transverse Dispersion

2.10

Tests to Determine Dispersivity

60

65

2.10.1

Laboratory Tests

2.10.2

Field Tests for Dispersivity

2.10.3

Single-Well Tracer Test

66
68
69

2.11

Scale Effect of Dispersion

Stochastic Models of Solute Transport

71
77

2.12.1

Introduction

2.12.2

Stochastic Descriptions of Heterogeneity

78

2.12.3

Stochastic Approach to Solute Transport

81

77

Fractal Geometry Approach to Field-scale Dispersion


2.13.1

Introduction

61

66

2.12

2.14

52

54

Methods of Solution

Boundary)

2.13

43

47

85

85

2.13.2

Fractal Mathematics

2.13.3

Fractal Geometry and Dispersion

85

2.13.4

Fractal Scaling of Hydraulic Conductivity

Deterministic Models of Solute Transport


Case Study: Borden Landfill Plume

96

88

93

90

63

61

xi

Contents
2.15

Transport in Fractured Media

2.16

Summary

Chapter Notation
References

Chapter

103

107
109

111

Three

T r a n s f o r m a t i o n , Retardation, and Attenuation of Solutes


3.1

Introduction

3.2

Classification of Chemical Reactions

3.3

Sorption Processes

3.4

1 15
116

117

Equilibrium Surface Reactions

117

3.4.1

Linear Sorption Isotherm

3.4.2

Freundlich Sorption Isotherm

117

3.4.3

Langmuir Sorption Isotherm

3.4.4

Effect of Equilibrium Retardation on Solute Transport

119
122

3.5

Nonequilibrium (Kinetic) Sorption Models

3.6

Sorption of Hydrophobic (Organic) Compounds

3.7

115

129
132

3.6.1

Introduction

3.6.2

Partitioning onto Soil or Aquifer Organic Carbon

3.6.3

Estimating K

132

3.6.4

Estimating K from Solubility Data

3.6.5

Estimating K

3.6.6

Multiple Solute Effects

from K

oc

ow

Data

134

from Molecular Structure

oc

Homogeneous Reactions

140

Introduction

3.7.2

Chemical Equilibrium

3.7.3

Chemical Kinetics

3.7.4

Tenads in Chemical Reactions

Radioactive Decay
Biodgradation

3.10

Colloidal Transport

138

140

3.7.1

3.9

132

133

oc

3.8

123

140
141
141
142

144
144
149

Case Study: Large-scale Field Experiment on the Transport of Reactive and


Nonreactive Solutes in a Scale Aquifer under Natural Ground-Water
GradientsBorden, Ontario
3.1 1

Summary

157

Chapter Notation
References

160

158

150

xii

Contents

Chapter Four
F l o w and Mass Transport in the Vadose Zone

163

4.1

Introduction

4.2

Soil as a Porous Medium

163

4.3

Soil Colloids

4.4

The Electrostatic Double Layer

4.5

Salinity Effects on Hydraulic Conductivity of Soils

4.6

Flow of Water in the Unsaturated Zone


4.6.1
Soil-Water Potential
168

163

164
165
167

168

4.6.2

Soil-Water Characteristic Curves

4.6.3
4.6.4

Hysteresis
175
Construction of a Soil-Water-Retention Curve

4.6.5

Measurement of Soil-Water Potential

4.6.6
4.6.7

Unsaturated Hydraulic Conductivity

4.6.8

Richard Equation
183
Vapor Phase Transport

4.6.9

Buckingham Flux taw

169

177
180

182
184

4.7

Mass Transport in the Unsaturated Zone

4.8

Equilibrium Models of Mass Transport

185
186

4.9

Nonequilibrium Models of Mass Transport

4.10

Anion Exclusion

188

190

Case Study: Relative Movement of Solute and Wetting Fronts


4.11

Preferential Flowpaths in the Vadose Zone

4.12

Summary

193

196

198

Chapter Notation
References

198

200

Chapter Five
Multiphase F l o w

202

5.1

Introduction

202

5.2

Basic Concepts
203
5.2.1
Saturation Ratio
203
5.2.2
Interfacial Tension and Wettability
5.2.3

Capillary Pressure

203

204

5.2.4

Relative Permeability

5.2.5

Darcy's Law for Two-Phase Flow

206

5.2.6

Fluid Potential and Head

212

211

xiii

Contents

S3

Migration ot Light Nonaqueous Phase Liquids (LNAPLs)

5.4

Measurement of the Thickness of a Floating Product

217
225

5.5

Effect of the Rise and Fall of the Water Table on the Distribution of LNAPLs

5.6

Migration of Dense Nonaqueous Phase liquids


5.6.1

Vadose Zone Migration

5.6.2

Vertical Movement in the Saturated Zone

5.6.3

Horizontal Movement in the Saturated Zone

5.7

Monitoring for LNAPLs and DNAPLs

5.8

Summary

231

231

231
233
235

238

239

Chapter Notation
References

240

242

Chapter S i x
Inorganic Chemicals in Ground Water
6.1

Introduction

244

244

6.2

Units of Measurement and Concentration

6.3

Chemical Equilibrium and the Law of Mass Action

6.4

Oxidation-Reduction Reactions

6.5

6.6

6.7

253

253

6.5.2

Relationship of Eh and pH

6.5.3

Eh-pH Diagrams

6.5.4

Calculating Eh-pH Stability Fields

Metal Complexes

253

254
257

267

6.6.1

Hydration of Cations

6.6.2

Complexation

6.6.3

Organic Complexing Agents

267

267
269

Chemistry of Nonmetallic Inorganic Contaminants


6.7.1

6.8

pH

245

249

Relationship between pH and Eh


6.5.1

244

Fluoride

270

6.7.2

Chlorine and Bromine

6.7.3

Sulfur

6.7.4

Nitrogen

6.7.5

Arsenic

272

6.7.6

Selenium

6.7.7

Phosphorus

272
274
276
276

Chemistry of Metals
6.8.1

Beryllium

6.8.2

Strontium

6.8.3

Barium

276
277
277
277

271

270

xiv

6.9

Contents
6.8.4

Vanadium

6.8.5

Chromium

277

6.8.6

Cobalt

278

6.8.7

Nickel

279

6.8.8

Molybdenum

6.8.9

Copper

6.8.10

Silver

6.8.11

Zinc

277

279
279

279
280

6.8.12

Cadmium

280

6.8.13

Mercury

280

6.8.14

Lead

280

Radioact ive Isotopes 281


6.9.1

Introduction

6.9.2

Adsorption of Cationic Radionuclides

281

6.9.3

Uranium

6.9.4

Thorium

6.9.5

Radium

6.9.6

Radon

287

6.9.7

Tritium

288

282

282
285
286

6.10

Geochemical Zonation

6.11

Summary

288

292

Chapter Notation
References

292

293

Chapter S e v e n
Organic Compounds in Ground Water

295

7.1

Introduction

7.2

Physical Properties of Organic Compounds

295

7.3

Organic Structure and Nomenclature


7.3.1

Hydrocarbon Classes

7.3.2

Aromatic Hydrocarbons

7.4

Petroleum Distillates

7.5

Functional Groups
305
7.5.1
Organic Halides
7.5.2

Alcohols

297

305

308

Ethers

7.5.4

Aldehydes and Ketones

7.5.5

Carboxylic Acids
Esters

7.5.7

Phenols

300

301

7.5.3

7.5.6

297

308

312
312

311

311

295

Contents

7.6

xv

7.5.8

Organic Compounds Containing Nitrogen

7.5.9

Organic Compounds Containing Sulfur and Phosphorus

Degradation of Organic Compounds

314
315

316

7.6.1

Introduction

7.6.2

Degradation of Hydrocarbons

316

7.6.3

Degradation of Chlorinated Hydrocarbons

7.6.4

Degradation of Organic Pesticides

318
319

323

Field Examples of Biological Degradation of Organic Molecules


7.7.1

Introduction

7.7.2

Chlorinated Ethanes and Ethenes

7.7.3

Aromatic Compounds

326

326
327

328

7.8

Analysis of Organic Compounds in Ground Water

7.9

Summary

329

334

References

335

Chapter Eight
Ground Water and Soil Monitoring
8.1 /

Introduction

8.2

Monitoring Well Design

8.3

8.4

338

338
338

8.2.1

General Information

8.2.2

Monitoring W e l l Casing

338

8.2.3

Monitoring Well Screens

8.2.4

Naturally Developed and Filter-Packed W e l l s

8.2.5

Annular Seal

339
345

8.2.6

Protective Casing

8.2.7

Screen Length and Setting

8.2.8

Summary of Monitoring Well Design

Installation of Monitoring W e l l s

348
349
351

353

8.3.1

Decontamination Procedures

8.3.2

Methods of Drilling

8.3.3

Drilling in Contaminated Soil

Sample Collection

346

347

353

354
359

360

8.5

Installation of Monitoring Wells

8.6

Monitoring Well Development

364
370

8.7

Record Keeping During Monitoring Well Construction

8.8

Monitoring Well and Borehole Abandonment

8.9

Multiple-level Devices for Ground-Water Monitoring

8.10

W e l l Sampling
378
8.10.1
Introduction

378

375

375
376

xvi

1.11

1.12

>. 13

Contents

8.10.2

Well Purging

8.10.3

Well-Sampling Devices

379

Soil-Gas Monitoring

380

383

8.11.1

Introduction

8.11.2

Methods ol Soil-Gas Monitoring

383

Soil-Water Sampling

385

8.12.1

Introduction

385

8.12.2

Suction Lysimeters

8.12.3

Installation of Suction Lysimeters

Summary

384

385
389

389

References

390

Chapter N i n e
Site Remediation
9.1
9.2

94

Solid Waste

392
392

9.2.2

Removal and Disposal

9.2.3

Containment

9.2.4

Hydrodynamic Isolation

393

393

Pump-and-Treat Systems

399

401

9.3.1

Overview

9.3.2

Capture Zones

9.3.3

Computation of Capture Zones

9.3.4

Optimizing Withdrawal-Injection Systems

9.3.5

Permanent Plume Stabilization

401
403

Treatment of Extracted Ground Water


9.4.1

f^^)

392

Source-Control Measures
9.2.1

f 9.3 J

392

Introduction

Overview

405

416

416

9.4.2

Treatment of Inorganic Contaminants

9.4.3

Treatment of Dissolved Organic Contaminants

Recovery of Nonaqueous Phase Liquids

Removal of Leaking Underground Storage Tanks

9.7

Soil-Vapor Extraction
In Situ Bioremediation

417
417

418

9.6

9.8

414

416

424

427
429

Case Study: Enhanced Biodgradation of Chlorinated Ethenes


9.9

Combination Methods

433

434

Case Study: Remediation of a Drinking Water Aquifer Contaminated with Volatile


Organic Compounds

438

Contents

vii

Case Study: Ground-Water Remediation Using a Pump-and-Treat Technique


Combined with Soil Washing
9.10

Summary

Chapter Notation
References

439

442
443

443

Appendix A
Error Function Values

445

Appendix B
Bessel Functions

446

Appendix C
W(f, B) Values

448

Appendix D
Exponential Integral

450

Appendix E
Unit Abbreviations

451

Index

452

Chapter One

Introduction

1.1

Ground Water as a Resource


G r o u n d water is the s o u r c e for drinking water for many p e o p l e around the world,
especially in rural areas. In the United States g r o u n d water supplies 4 2 . 4 % of the population served by public water utilities. Virtually all the h o m e s that supply their o w n
water have wells and use g r o u n d water. In all, m o r e than half of the population ( 5 2 . 5 % )
of the United States relies u p o n a ground-water s o u r c e for drinking water (Solley, Merk,
and Pierce 1 9 8 8 ) .
T a b l e 1.1 s h o w s the ground-water withdrawals by category of use in the United
States in 1 9 8 5 as well as the p e r c e n t a g e of total use for that category supplied by ground
water. In T a b l e 1.1 public supply refers to water provided by either a public water utility
or a private water c o m p a n y and used for residential, c o m m e r c i a l , and industrial uses,
power-plant cooling, and municipal uses such as fire lighting. All o t h e r categories are
self-supplied, with the user owning the water system. Many of the self supplied systems
rely upon water wells. F r o m 1 9 8 0 to 1 9 8 4 an average of 3 7 0 , 0 0 0 water wells w e r e drilled
in the United States e a c h year (Hindall and E b e r l e 1 9 8 9 ) .
Inasmuch as g r o u n d water provides drinking water to so many p e o p l e , the quality
of g r o u n d water is of paramount importance. Public water suppliers in the United States
are obligated by the Safe Drinking Water Act of 1 9 8 6 to furnish water to their c o n s u m e r s
that m e e t s specific drinking-water standards. If the water d o e s not meet the standards
w h e n it is withdrawn from its s o u r c e , it must be treated. G r o u n d water may not meet
the standards b e c a u s e it contains dissolved constituents c o m i n g from natural sources.
C o m m o n e x a m p l e s of constituents c o m i n g from natural s o u r c e s are total dissolved
solids, sulfate, and chloride. G r o u n d water also may not m e e t the standards b e c a u s e it
contains organic liquids, dissolved organic and inorganic constituents, or p a t h o g e n s that
c a m e from an a n t h r o p o g e n i c source. In such c a s e s the g r o u n d water has b e e n contaminated by the acts of humans.
In the c a s e of self-supplied systems, a s o u r c e of uncontaminated water is of even
greater importance. Such systems are typically tested initially for only a very limited range
of constituents, such as coliform bacteria, nitrate, chloride, and iron. Most times ground
water contamination c a n n o t be tasted, so that with such limited testing it is p o s s i b l e for

Chapter One
T A B L E 1.1

Ground-water usage in the United States, 1985.


Percent of T o t a l U s e

Category
Public water supply
Domestic, self-supplied
Commercial, self-supplied

Ground-water Use

Supplied

(million gallons/day)

by G r o u n d Water

14.600

40.0

3,250

97.9

746

60.7

Irrigation

45,700

33.4

Livestock

3,020

67.6

Industrial (fresh)

3,930

17.6

Industrial (saline)

26

0.7

1,410

52.8

Mining (saline)

626

81.9

Power plant cooling

608

0.5

Mining (fresh)

Source: Solley, Merk, and Pierce, 1988.

a user to have a contaminated s o u r c e and not be aware of it. Additionally, self supplied
systems rarely undergo treatment o t h e r than softening and perhaps iron removal. T h e r e
are limited options available for the h o m e o w n e r w h o wishes to treat contaminated ground
water so that it can be c o n s u m e d .
In addition to providing for the sustenance of human life, ground water has
important ecological functions. Many freshwater habitats are supplied by the discharge
of springs. If the ground water supplying these springs is contaminated, the ecological
function of the freshwater habitat can be impaired.

1.2

Types of Ground-Water Contaminants


A wide variety of materials have b e e n identified as contaminants found in ground water.
T h e s e include synthetic organic chemicals, hydrocarbons, inorganic cations, inorganic
anions, pathogens, and radionuclides. T a b l e 1.2 contains an extensive listing of these
c o m p o u n d s . Most of these materials will dissolve in water to varving degrees. S o m e of
the organic c o m p o u n d s are only slightly soluble and will exist in both a dissolved form
and as an insoluble phase, which can also migrate through the ground. Examples of the
uses of these materials are also given on T a b l e 1.2. T h e s e uses may provide help in
locating the s o u r c e of a c o m p o u n d if it is found in ground water. T h e inorganic cations
and anions o c c u r in nature and may c o m e from natural as well as anthropogenic sources.
S o m e of the radionuclides are naturally occurring and can c o m e from natural sources
as well as mining, milling, and processing ore, industrial uses, and disposal of radioactive
waste. O t h e r radionuclides are man-made and c o m e from nuclear w e a p o n s production
and testing.
T a b l e 1.3 lists the organic contaminants found in ground water at a single hazardous
waste site. Almost 80 c o m p o u n d s were detected at this former organic solvent-recycling
facility.

T A B L E 1.2

Substances known to occur in ground water.


E x a m p l e s of uses

Contaminant
A r o m a t i c hydrocarbons
Acetanilide

Intermediate manufacturing, pharmaceuticals, dyestuffs

Alkyl benzene sulfonates

Detergents

Aniline

Dyestuffs, intermediate, photographic chemicals, pharmaceuticals,

Anthracene

Dyestuffs, intermediate, semiconductor research

herbicides, fungicides, petroleum refining, explosives

Benzene

Detergents, intermediate, solvents, antiknock gasoline

Benzidine

Dyestuffs, reagent, stiffening agent in rubber compounding

Benzyl alcohol

Solvent, perfumes and flavors, photographic developer inks, dye-

Butoxymethyl benzene

NA

Chrysene

Organic synthesis, coal tar by-product

Creosote mixture

W o o d preservatives, disinfectants

Dibenz[a.h.]anthracene

NA

Di-butyl-p-benzoquinone

NA

Dihydrotrimethylquinoline

Rubber antioxidant

stuffs, intermediate

4,4-Dinitrosodiphenylamine

NA

Ethylbenzene

Intermediate, solvent, gasoline

Fluoranthene

Coal tar by-product

Fluorene

Resinous products, dyestuffs, insecticides, coal tar by-product

Fluorescein

Dyestuffs

Isopropyl benzene

Solvent, chemical manufacturing

4,4'-methylene-bis-2-chloroaniline (MOCA)

Curing agent for polyurethanes and epoxy resins

Methylthiobenzothiazole

NA

Naptholene

Solvent, lubricant, explosives, preservatives, intermediate, fungicide.

o-Nitroaniline

Dyestuffs, intermediate, interior paint pigments, chemical

Nitrobenzene

Solvent, polishes, chemical manufacturing

4-Nitrophenol

Chemical manufacturing

n-Nitrosodiphenylamine

Pesticides, retarder of vulcanization of rubber

Phenanthrene

Dyestuffs, explosives, synthesis of drugs, biochemical research

n-Propylbenzene

Dyestuffs, solvent

Pyrene

Biochemical research, coal tar by-product

moth repellant

manufacturing

Styrene (vinyl benzene)

Plastics, resins, protective coatings, intermediate

Toluene

Adhesive solvent in plastics, solvent, aviation and high-octane


blending stock, dilutent and thinner, chemicals, explosives,
detergents

1,2,4-Trimethylbenzene

Manufacture of dyestuffs, pharmaceuticals, chemical manufacturing

Xylenes (m, o, p)

Aviation gasoline, protective coatings, solvent, synthesis of organic


chemicals, gasoline

Oxygenated h y d r o c a r b o n s
Acetic acid

Food additives, plastics, dyestuffs, pharmaceuticals, photographic

Acetone

Dyestuffs, solvent, chemical manufacturing, cleaning and drying of

Benzophenone

Organic synthesis, odor fixative, flavoring, pharmaceuticals

Butyl acetate

Solvent

n-Butyl-benzylphtholate

Plastics, intermediate

chemicals, insecticides
precision equipment

Source: Office of Technology Assessment, rVofecring The Nation's Groundwater from Confortiinotion, 1984, pp. 2 3 3 1 .
a

NA: No information in standard sources.

Chapter One

T A B L E 1.2

Cont'd
Contaminant

E x a m p l e s of uses

O x y g e n a t e d hydrocarbons (cont'd)
Oi-n-butyl phthalate

Plasticizer, solvent, adhesives, insecticides, safety glass, inks, paper

Diethyl ether

Chemical manufacturing, solvent, analytical chemistry, anesthetic,

Diethyl

Plastics, explosives, solvent, insecticides, perfumes

coatings
perfumes
phthalate

Diisopropyl ether

Solvent, rubber cements, paint and varnish removers

2,4-Dimethyl-3-hexanol

Intermediate, solvent, lubricant

2,4-Dimethyl phenol

Pharmaceuticals, plastics, disinfectants, solvent, dyestuffs, insecti-

Di-n-octyl phthalate

Plasticizer for polyvinyl chloride and other vinyls

1,4-Dioxane

Solvent, lacquers, paints, varnishes, cleaning and detergent prepa-

cides, fungicides, additives to lubricants and gasolines

rations, fumigants, paint and varnish removers, wetting agent,


cosmetics
Ethyl acrylate

Polymers, acrylic paints, intermediate

Formic acid

Dyeing and finishing, chemicals, manufacture of fumigants, insecti-

Methanol (methyl alcohol}

Chemical manufacturing, solvents, automotive antifreeze, fuels

Methylcyclohexanone

Solvent, lacquers

Methyl ethyl ketone

Solvent, paint removers, cements and adhesives, cleaning fluids.

Methylphenyl acetamide

NA

Phenols (e.g., p-tert-butylphenol)

Resins, solvent, pharmaceuticals, reagent, dyestuffs and indicators,

Phthalic acid

Dyestuffs, medicine, perfumes, reagent

2-Propanol

Chemical manufacturing, solvent, deicing agent, pharmaceuticals,

2-Propyl-1 -heptanol

Solvent

Tetrahydrofuran

Solvent

Varsol

Paint and varnish thinner

cides, solvents, plastics, refrigerants

printing, acrylic coatings

germicidal points

perfumes, lacquers, dehydrating agent, preservatives

H y d r o c a r b o n s w i t h specific elements
(e.g., w i t h N, P, S, CI, B r , 1, F)
Acetyl chloride

Dyestuffs, pharmaceuticals, organic preparations

Alachlor (Lasso)

Herbicides

Aldicarb (sulfoxide and sulfone; Temik)

Insecticide, nematocide

Aldrin

Insecticides

Atrazine

Herbicides, plant growth regulator, weed-control agent

8enzoyl chloride

Medicine, intermediate

Bromacil

Herbicides

Bromobenzene

Solvent, motor oils, organic synthesis

Bromochloromethane

Fire extinguishers, organic synthesis

Bromodichloromethane

Solvent, fire extinguisher fluid, mineral and salt separations

Bromoform

Solvent, intermediate

Carbofuran

Insecticide, nematocide

Carbon tetrachloride

Degreasers, refrigerants and propellants, fumigants, chemical

Chlordane

Insecticides, oil emulsions

Chlorobenzene

Solvent, pesticides, chemical manufacturing

Chloroform

Plastics, fumigants, insecticides, refrigerants and propellants

manufacturing

Introduction

T A B L E 1.2

Cont'd
Contaminant

E x a m p l e s of uses

H y d r o c a r b o n s w i t h specific elements
(cont'd)
Chlorohexane

NA

Chloromethane (methyl chloride)

Refrigerants, medicine, propellonts, herbicide, organic synthesis

Chloromethyl sulfide

NA

2-Chloronaphthalene

O i l : plasticizer, solvent for dyestuffs, varnish gums and resins.


waxes wax: moisture-, flame-, acid-, and insect-proofing of
fibrous materials; moisture- and flame-proofing of electrical
cable; solvent (see oil)
NA

Chlorpyrifos
Chlorthal-methyl (DCPA, or Dacthal)

Herbicide

p-Chlorophenyl methylsulfone

Herbicide manufacture

Chlorophenylmethyl sulfide

Herbicide manufacture

Chlorophenylmethyl sulfoxide

Herbicide manufacture

o-Chlorotoluene

Solvent, intermediate

p-Chlorotoluene

Solvent, intermediate

Cyclopentadine

Insecticide manufacture

Dibromochloromethane

Organic synthesis

Dibromochloropropane

(DBCP)

Fumigant, nematocide

Dibromodichloroethylene

NA

Dibromoethane (ethylene dibromide, EDB)

Fumigant, nematocide, solvent, waterproofing preparations, organic

Dibromomethane

Organic synthesis, solvent

Dichlofenthion (DCFT)

Pesticides

o-Dichlorobenzene

Solvent, fumigants, dyestuffs, insecticides, degreasers, polishes.

p-Dichlorobenzene

Insecticides, moth repellant, germicide, space odorant, intermediate,

Dichlorobenzidine

Intermediate, curing agent for resins

synthesis

industrial odor control

fumigants

Dichlorocyclooctadiene
Dichlorodiphenyldichloroethane

Pesticides
(ODD,

Dichlorodiphenyldichloroethylene
Dichlorodiphenyltrichloroethane

TDE)

(DDE)
(DDT)

Insecticides
Degradation product of D D T , found as an impurity in D D T residues
Pesticides

1,1 -Dichloroethane

Solvent fumigants, medicine

1,2-Dichloroethane

Solvent, degreasers, soaps and scouring compounds, organic syn-

1,1-Dichloroethylene (vinylidiene chloride)

Saran (used in screens, upholstery, fabrics, carpets, etc.), adhesives.

1,2-Dichloroethylene (cis and trans)

Solvent, perfumes, lacquers, thermoplastics, dye extraction, organic

Dichloroethyl ether

Solvent, organic synthesis, paints, varnishes, lacquers, finish removers.

Dichloroiodomethane

NA

Dichloroisopropylether

Solvent, paint and varnish removers, cleaning solutions

thesis, additive in antiknock gasoline, paint and finish removers

synthetic fibers
synthesis, medicine

drycleaning, fumigants

( = bis-2-chloroisopropylether)
Dichloromethane (methylene chloride)

Solvent, plastics, paint removers, propellants, blowing agent in foams

Dichloropentadiene

NA

2,4-Dichlorophenol

Organic synthesis

2,4-Dichlorophenoxyacetic acid (2,4-DJ

Herbicides

Chapter One

T A B L E 1.2

Conf
Contaminant

E x a m p l e s of uses

H y d r o c a r b o n s w i t h specific elements
(cont'd)
i ,2-Dichloropropane

Solvent, intermediate, scouring compounds, fumigant, nematocide.

Dicyclopentadiene (DCPD)

Insecticide manufacture

additive for antiknock fluids


Dieldrin

Insecticides

Dodomethane

Organic synthesis

Diisopropylmethyl phosphonate (DIMP)

Nerve gas manufacture

Dimethyl disulfide

NA

Dimethylformamide

Solvent, organic synthesis

2,4-Dinotrophenol (Dinoseb, DNBP)

Herbicides

Dithiane

Mustard gas manufacture

Dioxins (e.g., TCDD)

Impurity in the herbicide 2 , 4 , 5 - T

Dodecyl mercaptan (lauryl mercaptan)

Manufacture of synthetic rubber and plastics, pharmaceuticals,

Endosulfan

Insecticides

insecticides, fungicides
Endrin

Insecticides

Ethyl chloride

Chemical manufacturing, anesthetic, solvent, refrigerants, insecticides

Bis-2-ethylhexylphthalate

Plastics

Di-2-ethylexylphthalate

Plasticizers

Fluorobenzene

Insecticide and larvicide intermediate

Fluoroform

Refrigerants, intermediate, blowing agent for foams

Heptachlor

Insecticides

Heptachlorepoxide

Degradation product of heptachlor, also acts as an insecticide

Hexachlorobicycloheptadiene

NA

Hexachlorobutadiene

Solvent, transformer and hydraulic fluid, heat-transfer liquid

2-Hexachlorocyclohexane

Insecticides

(= Benzenehexachloride, or a-BHC)
/f-Hexachlorocyclohexane

(/-BHC)

Insecticides

/-Hexachlorocyclohexane (y-BHC, or Lindane)

Insecticides

Hexachlorocyclopentadiene

Intermediate for resins, dyestuffs, pesticides, fungicides,

Hexochloroethane

Solvent, pyrotechnics and smoke devices, explosives, organic

Hexachloronorbornadiene

NA

Isodrin

Intermediate compound in manufacture of Endrin

Kepone

Pesticides

Malathion

Insecticides

Methoxychlor

Insecticides

Methyl bromide

Fumigants, pesticides, organic synthesis

pharmaceuticals

synthesis

Methyl parathion

Insecticides

Oxathine

Mustard gas manufacture


Insecticides

Parathion
Pentachlorophenol

(PCP)

Insecticides, fungicides, bactericides, algicides, herbicides, wood


preservative

Phorate (Disulfoton)

Insecticides

Polybrominated biphenyls (PBBs)

Flame retardant for plastics, paper, and textiles

Polychlorinated biphenyls (PCBs)

Heat-exchange and insulating fluids in closed systems

Prometon

Herbicides

Introduction

T A B L E 1.2

Cont'd
Contaminant

E x a m p l e s of uses

H y d r o c a r b o n s w i t h specific elements
(cont'd)
RDX (Cyclonite)

Explosives

Simazine

Herbicides

Tetrachlorobenzene

NA

Tetrachloroethanes (1,1,1,2 and 1,1,2,2)

Degreasers, paint removers, varnishes, lacquers, photographic film.

Tetrachloroethylene (or perchloroethylene,

Degreasers, drycleaning, solvent, drying agent, chemical manufac-

organic synthesis, solvent, insecticides, fumigants, weed killer


turing, heat-transfer medium, vermifuge

PCE)
Toxaphene

Insecticides

Triazine

Herbicides

1,2,4-Trichlorobenzene

Solvent, dyestuffs, insecticides, lubricants, heat-transfer medium (e.g..

Trichloroethanes (1,1,1 and 1,1,2)

Pesticides, degreasers, solvent

1,1,2-Trichloroetbylene

Degreasers, paints, drycleaning, dyestuffs, textiles, solvent, refriger-

coolant)

(TCE)

ant and heat exchange liquid, fumigont, intermediate, aerospace


operations
Tricholorfluoromethane (Freon 11)

Solvent, refrigerants, fire extinguishers, intermediate

2,4,6-Trichlorophenol

Fungicides, herbicides, defoliant

2,4,5-Tricholorophenoxyacetic acid (2,4,5-T)

Herbicides, defoliant

2,4,5-Trichlorophenoxypropionic acid (2,4,5-

Herbicides and plant growth regulator

TP or Silvex)
Trichlorotrifluoroethane

Dry-cleaning, fire extinguishers, refrigerants, intermediate, drying

Trinitrotoluene

Explosives, intermediate in dyestuffs and photographic chemicals

agent
(TNT)

Tris-(2,3-dibromopropyl) phosphate

Flame retardant

Vinyl chloride

Organic synthesis, polyvinyl chloride and copolymers, adhesives

Other hydrocarbons
Alkyl sulfonates

Detergents

Cyclohexane

Organic synthesis, solvent, oil extraction

1,3,5,7-Cyclooctatetraene

Organic research

Dicyclopentadiene (DCPD)

Intermediate for insecticides, paints and varnishes, flame retardants

2,3-Dimethylhexane

NA

Fuel oil

Fuel, heating

Gasoline

Fuel

Jet fuels

Fuel

Kerosene

Fuel, heating solvent, insecticides

Lignin

Newsprint, ceramic binder, dyestuffs, drilling fuel additive, plastics

Methylene blue activated substances (MBAS)

Dyestuffs, analytical chemistry

Propane

Fuel, solvent, refrigerants, propellents, organic synthesis

Tannin

Chemical manufacturing, tanning, textiles, electroplating, inks.

4,6,8-Trimethyl-l -nonene

NA

Undecane

Petroleum research, organic synthesis

pharmaceuticals, photography, paper

Metals and cations


Aluminum

Alloys, foundry, paints, protective coatings, electrical industry, pack-

Antimony

Hardening alloys, solders, sheet and pipe, pyrotechnics

aging, building and construction, machinery and equipment

Chapter One

T A B L E 1.2

Cont'd
Contaminant

E x a m p l e s of uses

Metals and cations (cont'd)


Arsenic

Alloys, dyestutts, medicine, solders, electronic devices, insecticides,

Barium

Alloys, lubricant

Beryllium

Structural material in space technology, inertiol guidance systems.

rodenticides, herbicide, preservative

additive to rocket fuels, moderator and reflector of neutrons in


nuclear reactors
Cadmium

Alloys, coatings, batteries, electrical equipment, fire-protection

Calcium

Alloys, fertilizers, reducing agent

Chromium

Alloys, protective coatings, paints, nuclear and high-temperature

Cobalt

Alloys, ceramics, drugs, paints, glass, printing, catalyst, electroplat-

Copper

Alloys, paints, electrical wiring, machinery, construction materials,

Iron

Alloys, machinery, magnets

lead

Alloys, batteries, gasoline additive, sheet and pipe, paints, radia-

Lithium

Alloys, pharmaceuticals, coolant, batteries, solders, propellants

Magnesium

Alloys, batteries, pyrotechnics, precision instruments, optical mirrors

Manganese

Alloys, purifying agent

Mercury

Alloys, electrical apparatus, instruments, fungicides, bactericides,

Molybdenum

Alloys, pigments, lubricant

systems, paints, fungicides, photography

research

ing, lamp filaments


electroplating, piping, insecticides

tion shielding

mildew proofing, paper, pharmaceuticals

Nickel

Alloys, ceramics, batteries, electroplating, catalyst

Palladium

Alloys, catalyst, jewelry, protective coatings, electrical equipment

Potassium

Alloys, catalyst

Selenium

Alloys, electronics, ceramics, catalyst

Silver

Alloys, photography, chemical manufacturing, mirrors, electronic

Sodium

Chemical manufacturing, catalyst, coolant, nonglare lighting for

Thallium

Alloys, glass, pesticides, photoelectric applications

Titanium

Alloys, structural materials, abrasives, coatings

Vanadium

Alloys, catalysts, target material for x-rays

Zinc

Alloys, electroplating, electronics, automotive parts, fungicides.

equipment, jewelry, equipment, catalyst, pharmaceuticals


highways, laboratory reagent

roofing, cable wrappings, nutrition


N o n m e t a l s and a n i o n s
Ammonia

Fertilizers, chemical manufacturing, refrigerants, synthetic fibers,

Boron

Alloys, fibers and filaments, semiconductors, propellants

Chlorides

Chemical manufacturing, water purification, shrink-proofing, flame-

Cyanides

Polymer production (heavy duty tires), coatings, metallurgy.

Fluorides

Toothpastes and other dentrifices, additive to drinking water

Nitrates

Fertilizers, food preservatives

Nitrites

Fertilizers, food preservatives

fuels, dyestuffs

retardants, food processing

pesticides

Introduction

T A B L E 1.2

Cont'd
E x a m p l e s of uses

Contaminant
N o n m e t a l s and a n i o n s (cont'd)
Phosphates

Detergents, fertilizers, food additives

Sulfates

Fertilizers, pesticides

Sulfites

Pulp production and processing, food preservatives

Microorganisms
Bacteria (coliform)
Giardia
Viruses
Radionuclides
Cesium 137

Gamma radiation source for certain foods

Chromium 51

Diagnosis of blood volume, blood cell life, cardiac output, etc.

Cobalt 60

Radiation therapy, irradiation, radiographic testing, research

Iodine 1 31

Medical diagnosis, therapy, leak detection, tracers (e.g., to study


efficiency of mixing pulp Fibers, chemical reactions, and thermal
stability of additives to food products), measuring film thicknesses
Medicine, tracer

Iron 59
Lead 2 1 0

NA

Phosphorus 32

Tracer, medical treatment, industrial measurements (e.g., tire-tread

Plutonium 2 3 8 , 2 4 3

Energy source, weaponry

Radium 2 2 6

Medical treatment,

Radium 2 2 8

Naturally occurring

Radon 2 2 2

Medicine, leak detection, radiography, flow rate measurement

Ruthenium 1 06

Catalyst

Scandium 46

Tracer studies, leak detection, semiconductors

Strontium 90

Medicine, industrial applications (e.g., measuring thicknesses.

Thorium 2 3 2

Naturally occurring

Tritium

Tracer, luminous instrument dials

Uranium 2 3 8

Nuclear reactors

Zinc 65

Industrial tracers (e.g., to study wear in alloys, galvanizing, body

Zirconium 95

NA

wear and thickness of films and ink)


radiography

density control)

metabolism, function of oil additives in lubricating oils)

T h e o c c u r r e n c e o f the s u b s t a n c e s found o n T a b l e s 1.2 and 1.3 c a n b e d e t e c t e d


only if a g r o u n d w a t e r s a m p l e has b e e n c o l l e c t e d and analyzed. In l o w c o n c e n t r a t i o n s
m o s t o f t h e s e s u b s t a n c e s are c o l o r l e s s , tasteless, and o d o r l e s s . A specific analytical
t e c h n i q u e must b e e m p l o y e d t o d e t e r m i n e t h e p r e s e n c e a n d c o n c e n t r a t i o n o f e a c h
s u b s t a n c e . Unless a s a m p l e is c o l l e c t e d and a specific test is p e r f o r m e d , t h e p r e s e n c e
of a c o n t a m i n a n t may n o t be d e t e c t e d . With so many potential c o n t a m i n a n t s , it is p o s s i b l e
that a s a m p l e c o u l d be c o l l e c t e d and tested and a specific c o n t a m i n a n t still not be found
b e c a u s e n o analysis was d o n e for that c o m p o u n d o r e l e m e n t .

10

Chapter One
T A B L E 1.3

Organic compounds detected in ground water at Seymour Recycling

Corporation hazardous waste site, Seymour, Indiana.


Extractable Organics
Phenol

2-Chlorophenol

2,3,6-Trimefhylphenol

2,4-Dimethylphenol

2,3-Dimethylphenol

2,6-Dimefhylphenol

3,4-Dimethylphenol

3,5-Dimethyl phenol

2-Ethylphenol

2-Methyl phenol

3- and/or 4-Methylphenol

Bis(2-ethylhexyl)phthalate

Di-n-butyl phthalate

Isophorone

Benzo(a)anthracene

Chrysene

2-Butanone

2-Hexanone

4-Methyl-2-pentanone

3,3,5-Trimethylhexanol

2-Hexanol

2-Heptanone

Cyclohexanol

Cyclohexanone

4-Methyl-2-pentanol

4-Hydroxy-4-methyl-2-pentanone

2-Hydroxy-triethylamine

Tri-n-propyl-amine

Allcyl amine

1,4-Dioxane

n-n- Dimethylformamide

n-n-Dimethylacetamide

Benzoic acid

4-Methylbenzoic acid

3-Methylbenzoic acid

3-Methyl-butanoic acid

Benzenepropionic acid

Benzeneacetic acid

2-Ethyl-hexanoic acid

2-Ethyl butanoic acid

Octanoic acid

Heptanoic acid

Hexanoic acid

Decanoic acid

Nonanoic acid

Pentanoic acid

Cyclohexanecarboxylic acid

1 -Methyl-2-pyrrolidinone

1-1'-Oxy bis (2-methoxy eth une)

1,2-Dichlorobenzene

1,1,2-Trichloroethane

Tetrachloroethene
Volatile Organics

Benzene

Ethyl benzene

Chloroform

Chloromethane

Chloroethane

1,2-Dichloroethane

1,1 -Dichloroethane

1,1,1-Trichloroethane

1,1,2-Trichloroethane

1,1-Dichloroethene

Trans-1,2-Dichloroethene

Trichloroethene

Tetrachloroethene

Methylene chloride

Vinyl chloride

Dichlorofluoromethane

Tetrahydrofuran

Acetone

2-Butanone

2-Methyl-2-propanol

2-Methyl-2-butanol

2-Propanol

2-Butanol

2-Hexone

4-methyl-2-pentanol

Ethyl ether

m-Xylene

o- and/or p-Xylene

Toluene
Note: Some compounds ore detected in both the extractable and the volatile fractions ond thus appear twice
in the list.
Source: C. W. Fetter, Final Hydrogeologie Report, Seymour Recycling Corporation Hozardous Waste Site,
Report to U. S. Environmental Protection Agency, Region V, September, 1965,

Introduction
T A B L E 1.4

11

Cost of analysis of a single ground-water sample.

Superfund list of 137 synthetic organic compounds


Twenty-three metals
Cyanide

$965
270
40

Radiological compounds

275

Bacterial analysis (fecal coliform and streptococcus}

36

Chloride

10

Fluoride

18

Nitrate

15

Nitrite

15

Ammonia

15

Phosphorous, total

19

Sulfate

16
6

pH

$1700

Total

A great deal of e x p e n s e is involved with a water-quality analysis. T a b l e 1.4 lists the


c o s t of an extensive laboratory analysis (at the 1 9 9 2 list price from o n e independent,
Wisconsin-certified l a b ) . T h i s table d o e s n o t include the c o s t o f collection o f the sample
to be analyzed.
T h e c o s t o f analysis increases a s the d e t e c t i o n l i m i t , the lowest c o n c e n t r a t i o n
that can be reliably detected, d e c r e a s e s . Ground-water contaminants can be routinely
detected at the parts-per-billion level, and with c a r e s o m e c o m p o u n d s can be quantified
at the parts per trillion level. To put that c o n c e n t r a t i o n in perspective, 0.4 mm is o n e trillionth of the distance to the m o o n .

1.3

Drinking-Water

Standards

W h e n measured at the parts-per trillion level, even carefully prepared, triple distilled,
deionized water will be s e e n to contain s o m e dissolved constituents. What d o e s this
mean? We must c o n s i d e r the quality of water with respect to the use to which it will be
placed. Water for many industrial p u r p o s e s n e e d not be as pure as water used for
drinking. In the United States the Safe Drinking W a t e r Act and its a m e n d m e n t s direct
the Environmental Protection Agency to establish m a x i m u m contaminant level goals
( M C L G s ) and m a x i m u m contaminant levels ( M C L s ) for drinking water supplied by public
water agencies.
A m a x i m u m contaminant-level goal is a n o n e n f o r c e a b l e goal set at a level to prevent
known or anticipated adverse health effects with a wide margin of safety. T h e MCI.G for
a c a r c i n o g e n is zero, w h e r e a s for chronically toxic c o m p o u n d s it is b a s e d on an a c c e p t a b l e
daily intake that takes into a c c o u n t e x p o s u r e from air, food, and drinking water. Maximum
contaminant levels are e n f o r c e a b l e standards that are set as c l o s e as feasible to the
MCLGs, taking into a c c o u n t water treatment t e c h n o l o g i e s and cost. Primary MCLs are
based on health risk, and s e c o n d a r y MCLs are based on aesthetics. T a b l e 1.5 contains
the drinking-water standards promulgated by the U.S. Environmental Protection Agency.

T A B L E 1.5

U S E P A drinking-water standards and health goals.

Chemical

MCLG

MCL

SMCL

tra/D

(/ig/L)

(/ig/D

S y n t h e t i c organic chemicals
Acrylamide (1)

0"

Adipates (di(ethylhexyl)adipate)

500'

Treatment technique

Alachlor

0"

Atdicarb

3"

Aldicarb sulfoxide

l"

Aldicarb sulfone

4"
*

Atrazine

3"

Benzene

5"

Benzo[a]anthracene (5)

0'

0.1'

Benzo[a]pyrene

0'

Benzo[b]fluoranthene (5)

Benzo[k]fluoranthene (5)

0'

Butylbenzyl phthalate (5)


Carbofuran

100'
40

Chlorodone

o
o

200'
0'

Dibromochloropropane

p-Dichlorobenzene (9)

75"

1,2-Dichloroethane
1,1 -Oichloroethylene
cis-1,2-Dichloroethylene
trans-1,2-Dichloroethylene
1,2-Dichloropropane
2,4-Dichlorophenoxyacetic acid (2,4-D)
Di(ethylhexyl)phthalate

0.2'

0.3'
d

600

5"

200'

Dibenz[a,h]anthracene (5)

o-Dichlorobenzene (9)

0.2'
0.2'

100'

(DBCP)

0.2'
1

40"

Dalapon

Carbontetrachloride

Chrysene (5}

0.2"
600"
75

10
5

7
70"

100"

70

500'

100"
d

70"
0'

5
70

4'

Note: A pCi (picocurie) is o meosure of the rote of radioactive disintegrations. Mrem ede/yr is o measure of the dose of radiation received by either the whole
body or a single organ.
1. This is o chemicol used in treatment ot drinking water supply. The U S E P A specifies how much moy be used in (he treatment process.
2. Dual numbers were proposed for aluminum because it is o constituent ot a chemicol used in the treatment ot drinking woter and it might not be possible
lor oil treatment systems to meet the lower limit.
3. The total of nitrate plus nitrite cannot exceed 10 m g / L
4. The proposed rule has two levels being considered.
5. The establishment ot MCLGs and MCLs is not required by the Sote Drinking Woter Act tor these compounds; however. MCLGs ond MCLs (or them are
being considered at the indicoted levels.
6. This M C I would replace the current MCL of 5 pCi/L for combined 2 2 6 fta and 228 Ra.
7. There is no MCL for copper and leod. The indicated values are proposed action levels that, under a complicated set of rules, would require treatment of
a woter supply lo reduce potential corrosion of the woter moins and pipes. The usual source of these compounds in public water supplies is primarily from
the corrosion of copper ond leod pipe and solder containing leod,
8. Standard under review as of January 1992.
9. SMCL is a suggested volue only. Concentrations above this level moy couse adverse laste. See Federal Register, January 30, 1991.
Final value. Published in federal Register, April 2, 1986.
b

Finol value. Published in federal Register. July 8, 1987.

' Final value. Published in Federal Register, June 28, 1989.


" Finol value. Published in Federal Register, Jonuary 30, 1 9 9 1 .
Finol volue. Published in Federal Register, July 1, 1 9 9 1 .
' Proposed value. Published in Federal Register, July 25, 1990,
0

Proposed value. Published in Federal Register. July 18, 1991.


Finol value. Published in Federal Register, July 7, 1 9 9 1 .
Proposed value. Published in Federal Register, Nov. 13, 1985.

Proposed value. Published in Federal Register, February, 1978.

T A B L E 1.5

Cont'd

Chemical

MCLG

MCL

SMCL

Og/U

i>g/U

(/tg/L)

Synthetic organic chemicals (cont'd)


Diguat

20'

20'

7'

100'

100'

Endrin

2'

Epichlorohydrin (1)

0"

Dinoseb
Endothall

Treatme nt technique

700"

Ethylbenzene (9)
Ethylene dibromide (EDB|

0.05"

700'

700'

Heptachlor

0"

0.4

Heptachlor epoxide

0.2"

Hexochlorobenzene

0'

1'

50'

50'

Hexachlorocyclopentadiene

[HEX]

0'

Indenopyrene (5)
Methoxychlor

40

0'

Monochlorobenzene

100

Oxamyl (vydate)

200'

0.2
40"

Methylene chloride

5'

100"
200'

PCBs as decachlorobiphenol

0"

Pentachlorophenol

0"

Picloram
Simaze

500'
1'

Styrene (9)

100"

2,3,7,8-TCDD (dioxin]

0'

Tetrachloroethylene

1,2,4-Trichlorobenzene

9'

0.5
1
500'
1'
100"
5 x 110" '
5"
9'

1,1,2-Trichloroethane

3'

5'

1,1,1-Trichloroethane

200

200

Toluene (9)

1000

1000"

Toxaphene

50

2-(2,4,5-Trichlorophenoxy)-

50"

10

Trichloroethylene

(TCE)

8'

0.4'

0.2

Lindane

30

700"

0"

Glyphosate

40

propionic acid (2,4,5-TP, or Silvex)


Vinyl chloride
Xylenes (total) (9)

10,000

10,000"

3'

10/5'

20

Inorganic chemicals
Aluminum (2)
Antimony (4)
Arsenic (8)
Asbestos (fibers per liter)
Barium

50-200
50'
7x10"
2000'

50'
7x10"
2000'

Beryllium

0'

1'

Cadmium

5"

Chromium
Copper (7)
Cyanide
Fluoride (8)
Lead (7)

100"

100"

1,300"

1.300"

200'

200'

4,000

4,000

0"

15"

2,000

14

Chapter O n e

T A B L E 1.5

Cont'd

Chemical

MCLG

MCL

SMCL

(TO/L)

Wa/D

(ra/U

Inorganic chemicals (cont'd)


Mercury
Nickel
Nitrate (as N) (3)
Nitrite (as N) (3)

2"

2"

100'

100'

10,000"

10,000"

1,000"

1,000"

50"

50"

Selenium
Silver

100"

Sulfate (4)

4 x 1 0 - 5 x 10 '

Thallium (4)

4 x 1 0 - 5 x 10 '

0.5'

2/1'

Microbiological parameters
Giardio

lamblia

0 organisms'

Legionella

0 organisms'

Heterotrophic bacteria

0 organisms'

Viruses

0 organisms'

Radionuclides
Radium 2 2 6 (6)

Radium 2 2 6 (6)

20 pCi/L

Radon 2 2 2

0'

3 0 0 pG/L

Uranium

20 pCI/L

20 ug/L
(30 pCI/l)"

Beta and Photon emitters

4 mrem ede/yr*

15 p C i / L

(excluding radium 228)


Adjusted gross alpha emitters

(excluding radium 2 2 6 , uranium,


and radon 2 2 2 )

1.4

Risk and Drinking Water


Cancer-risk levels for varying concentrations of contaminants have b e e n established by
toxicologists using extremely conservative methods. T h e s e methods are so conservative
7

that s o m e have questioned their validity (Ames, Magaw, and Gold 1 9 8 ; Lehr 1 9 9 0 b ) .
Such tests are performed by feeding chemicals in large d o s e s to rodents and then
extrapolating the effects to humans e x p o s e d to low d o s e s by using linear extrapolation
rates. However controversial the methods of establishing c a n c e r risks in drinking water
are, the MCLs obtained from them have the force of law. T h e b a s i c cancer-risk level that
6

the Environmental Protection Agency ( E P A ) uses is the 1 0 ~ levelthat is, o n e additional


c a n c e r death per million people. T h e EPA assumes that the person will c o n s u m e 2 L of
drinking water from the s a m e s o u r c e every day of their lives for 70 yr in arriving at the
6

concentration that has a 1 0 ~ c a n c e r risk. T h e population at large appears to

support

such conservatism, even though about 2 5 % of the population will eventually contract
6

c a n c e r (Wilson and Crouch 1 9 8 7 ) . If you are e x p o s e d to a carcinogen with a 1 0 " risk


level, your personal c h a n c e s of contracting c a n c e r are increased from 2596 to 2 5 . 0 0 1 %
(Lehr 1 9 9 0 a ) . T h e cost to society to support this level of conservatism in purifying

Introduction

15

drinking water is significant. T h e c o s t is even greater w h e n o n e c o n s i d e r s the restoration


of a large n u m b e r of sites w h e r e the g r o u n d water has b e c o m e contaminated with
c h e m i c a l s believed to be c a r c i n o g e n s . In 1 9 8 8 the EPA e x a m i n e d the studies that had
b e e n performed at 1 5 3 Superfund sites. At c l o s e to o n e quarter of the sites, the cleanup
costs w e r e m o r e than $ 1 0 million dollars, and at o n e site they w e r e $ 1 2 0 million ( H a n m e r
1989).
T h e r e is an irreducible risk associated with drinking water. In o r d e r to protect
against pathogenic disease, drinking water is usually chlorinated, especially if the water
c o m e s from a surface s o u r c e . Prior to chlorination of drinking-water supplies, w a t e r b o r n e
disease such as typhoid and c h o l e r a t o o k many lives. B e t w e e n 1 9 2 0 and 1 9 5 0 , a period
when the percentage of the population served by safe drinking water supplies was
increasing, there w e r e 1 0 5 0 deaths in the United States due to w a t e r b o r n e disease,
including typhoid fever, gastroenteritis, shigellosis, and amebiasis. S i n c e 1 9 5 0 there have
only b e e n 20 deaths from similar causes (van d e r Leeden, T r o i s e , and T o d d 1 9 9 0 , T a b l e
7-148).
T h e chlorine reacts with naturally occurring organics in the water to p r o d u c e
trihalomethanes. T h e average chlorinated tap water in the United States is reported to
contain 83 ng/l of c h l o r o f o r m (Ames, Magaw, and G o l d 1 9 8 7 ) . Ames, Magaw, and G o l d
used this as a b a s e with which to c o m p a r e o t h e r potential c a n c e r risks. T a b l e 1.6 contains
c a n c e r risks relative to drinking a liter of chlorinated tap water a day, with tap water
having a risk of 1.0. T h e relative risks w e r e determined as an index o b t a i n e d by dividing
the daily lifetime human e x p o s u r e in milligrams per kilogram of b o d y weight by the
daily d o s e rate for rodents in milligrams per kilogram of b o d y weight. T h e d o s e rate of
rodents is the daily d o s e necessary to give c a n c e r to half the rodents at the e n d of a
standard lifetime. Examination of the table s h o w s that there are n u m e r o u s c a n c e r risks
associated with living and eating. Water from a contaminated well that was c l o s e d in
Santa Clara County, California (Silicon Valley), had 2 8 0 0 / J g / L of trichloroethylene. Drinking 1 L of this water p e r day has a b o u t half the relative c a n c e r risk ( 4 ) as the risk from
nitrosamines ingested w h e n o n e has b a c o n for breakfast ( 9 ) . T h e b a c o n carries additional
risk b e c a u s e high dietary fat is thought to be a p o s s i b l e c o n t r i b u t o r to c o l o n c a n c e r
(Ames, Magaw. and G o l d 1 9 8 7 ) . Water with 2 8 0 0 / t g / L o f trichloroethylene has a 1 0 "
c a n c e r r i s k b a s e d o n the Environmental Protection Agency's Section 3 0 4 ( 1 ) ( 1 ) criteria
(Federal Register, N o v e m b e r 2 8 , 1 9 8 3 ) .
3

However, the c o n s u m e r of b a c o n has m a d e a c o n s c i o u s decision to eat it and


accept the health risks. C o n s u m e r s of tap water have the e x p e c t a t i o n that it is "safe" to
drink and are p r o b a b l y not willing to a c c e p t even very low c a n c e r risks. Society as a
w h o l e places a high value on pure water and is willing to pay to protect it.

1.5

Sources of Ground-Water Contamination


In a
I98H
report, Protecting the Nation's Groundwater from Contamination, the Office
o f T e c h n o l o g y Assessment ( O T A ) o f the U.S. C o n g r e s s listed m o r e than 3 0 different
potential s o u r c e s of g r o u n d water contamination. Although m o s t attention has focused
on waste materials as a s o u r c e of ground-water contamination, there are n u m e r o u s
s o u r c e s that are not associated with solid or liquid wastes. T h e O T A report divides the

16

Chapter

TABLE

1.6

One

Risk of getting cancer relative to drinking chlorinated tap water.

Relative
Risk

Source/Daily Human Exposure

Carcinogen

Water
1.0

Chlorinated tap water, 1 L

4.0

W e l l water, 1 L {worst well in Silicon Valley)

Chloroform, 82 fig
Trichloroefhylene, 2 8 0 0 /ig

Risks in Food
Peanut butter, 1 sandwich

Anatoxin

Mushroom, 1, raw

Hydrazines, etc.

2,800.0

Beer, 1 2 oz

Ethyl alcohol

4,700.0

Wine, 1 glass

Ethyl alcohol

0.3

Coffee, 1 cup

Hydrogen peroxide

Comfrey herbal tea, 1 cup

Symphytine

30.0
100.0

30.0
400.0
2,700.0
90.0
9.0

Bread, 2 slices

Formaldehyde

Cola, 1

Formaldehyde

Shrimp, 1 00 g

Formaldehyde

Cooked bacon, 100 g

Dimethylnitrosamine, diethylnitrosamine

60.0

Cooked fish or squid, broiled in a gas oven, 54 g

Dimethylnitrosamine

70.0

Brown mustard, 5 g

Allyl isothiocyanate

100.0
20.0
200.0

Basil, 1 g of dried leaf

Estragle

All cooked food, average U . S . diet

Heterocyclic amines

Natural root beer, 12 oz. (now banned)

Safrole

Food Additives and Pesticides


Diet soft drink, 12 oz.

Saccharin

0.4

Bread and grain products, average U . S . diet

Ethylene dibromide

0.5

Other food with pesticides, average U . S . diet

PCBs, D D E / D D T

60.0

Risks Around the Home


604.0
2,100.0
8.0

Breathing air in a conventional home, 14 hr

Formaldehyde, benzene

Breathing air in a mobile home, 14 hr

Formaldehyde

Swimming pool, 1 hr (for a child)

Chloroform
Risks at Work

5,800.0

Breathing air at work, U . S . average

Formaldehyde

Commonly Used Drugs


16,000.0

Sleeping pill (Phnobarbital), 60 mg

Phnobarbital

300.0

Pain-relief pill (Phenacetin), 3 0 0 mg

Phenacetin

Source: Joy Lehr, "Toxlcologlcol risk assessment distortions: Port I I I A different look at envlronmentolism," Ground Water 28, no. 3 (1990): 3 3 0 - 4 0 . Based
on a table and dota in Bruce Ames, Renae Magow. ond Lois Gold, "Ranking possible carcinogenic hazards," Science 236 (April 17, 1987): 2 7 1 - 7 9 .

c o n t a m i n a t i o n s o u r c e s into six c a t e g o r i e s . T h e following discussion has added s o m e


s o u r c e s n o t c o n t a i n e d in the O T A report. Figure 1.1 illustrates s o m e of t h e s e c o n t a m
ination s o u r c e s .
1.5.1

Category I: Sources Designed to Discharge Substances

Septic t a n k s and c e s s p o o l s
S e p t i c tanks and c e s s p o o l s are d e s i g n e d t o d i s c h a r g e d o m e s t i c wastewater into the
s u b s u r f a c e a b o v e the water table. W a t e r from toilets, sinks and s h o w e r s , dishwashers,

(mg/L)

Ortho P

(mg/L)

Total P

(mg/L)

Nitrate-Nitrogen

(mg/L)

Ammonia N

(mg/L)

Total N

(no./mL)

Fecal Strep

TSS

(mg/L)

(filtered)

COD (mg/L)

(unfiltered)

COD (mg/L)

BOD

(mg/L)

Effluent quality from six septic tanks.

(g/da)

Average Flow

Site

T A B L E 1.7

One

(no./mL)

Chapter

Fecal Coiiforms

18

75

131

325

249

69

2907

2.7

50.5

34.1

0.68

12.3

10.8

125

176

361

323

44

4127

39.7

57.8

42.5

0.46

14.1

13.6

245

272

542

386

68

27,931

1387

76.3

45.6

0.60

31.4

14.0

315

127

291

217

52

11,113

184

40 2

33.2

0.35

11.0

10.1

860

120

294

245

51

2310

20.7

31.6

20.1

0.16

11.1

10.5

150

122

337

281

48

3246

25.3

56.7

38 3

0.83

11.6

10 5

Source: R. J. Otis, W. C. Boyle, ond D. K. Sauer, Small-Scale Waste Management Program, University of WisconsinModison, 1973.
All values ore means.
b

Includes 340-g/do sewer flow ond 520-g/do from foundation drain.

and washing machines passes from the h o m e into a septic tank, w h e r e it undergoes
settling and s o m e a n a e r o b i c decomposition. It is then discharged to the soil via a drainage
system. In 1 9 7 7 there w e r e an estimated 16.8 million septic systems in use in the United
States (Miller 1 9 8 0 ) . Septic systems discharge a variety of inorganic and organic c o m pounds. T a b l e 1 . 7 contains an analysis of septic-tank effluent. In addition to the domestic
wastewater, septic-tank cleaners containing synthetic organic chemicals such as trichloroethylene, b e n z e n e , and methylene chloride are discharged to the subsurface. An
estimated 4 0 0 , 0 0 0 gal of septic-tank cleaning fluids were used on Long Island, New York,
in 1 9 7 9 ( B u r m a s t e r and Harris 1 9 8 2 ) . Shallow ground water on Long Island is known
to be contaminated by these s a m e chemicals (Eckhardt and Oaksford 1 9 8 8 ) .
Injection w e l l s

Injection wells are used to discharge liquid wastes and o t h e r liquids into subsurface
z o n e s b e l o w the water table. Liquids that are injected include ( 1 ) hazardous wastes, ( 2 )
b r i n e from oil wells, ( 3 ) agricultural and urban runoff, ( 4 ) municipal sewage, ( 5 ) airconditioning return water, (6) heat-pump return water, ( 7 ) liquids used for enhanced
oil recovery from oil fields, ( 8 ) treated water intended for artificial aquifer recharge, and
( 9 ) fluids used in solution mining.
Injection wells can cause ground-water contamination if the fluid being injected
accidentally or deliberately enters a drinking water aquifer. This could happen b e c a u s e
of p o o r well design, p o o r understanding of the geology, fault)' well construction, or
deteriorated well casing. Wastewater correctly injected into subsurface z o n e s containing
unusable water could still migrate to a usable aquifer by being forced through cracks
in a confining layer under unnatural pressures or by flowing through the aquifer to a
nearby well that was improperly constructed or abandoned. Injection wells are nowregulated under the Underground Injection Control Program of the Safe Drinking Water

Introduction

19

Act. T h e 1 9 8 4 a m e n d m e n t s to the R e s o u r c e Conservation and Recovery Act prohibit the


underground injection of certain hazardous wastes.
Land

application

T r e a t e d or untreated municipal and industrial wastewater is applied to the land primarily


via spray irrigation systems. E x p o s u r e to the e l e m e n t s , plants, and m i c r o o r g a n i s m s in
the soil can break d o w n the natural organic matter in the wastewater.
Sludge from wastewater-treatment plants is often applied to the soil as a fertilizer,
as is m a n u r e from farm animals and whey from c h e e s e manufacturing. Oily wastes from
refining operations have b e e n applied to the soil so that they could be b r o k e n down
by soil m i c r o b e s . Nitrogen, p h o s p h o r o u s , heavy metals, and refractory organic c o m p o u n d s are potential ground-water c o n t a m i n a n t s that can leach from soil used for land
applications o f wastes and wastewater.
1.5.2

Category II: S o u r c e s D e s i g n e d t o S t o r e , T r e a t a n d / o r D i s p o s e o f S u b s t a n c e s

Landfills

Landfills are, by definition, designed to minimize adverse effects of waste disposal (Miller
1 9 8 0 ) . However, many w e r e poorly designed and are leaking liquids, genetically t e r m e d
l e a c h a t e , which are contaminating g r o u n d water. Landfills can contain n o n h a z a r d o u s
municipal waste, n o n h a z a r d o u s industrial waste, or hazardous waste as defined by the
R e s o u r c e Conservation and Recovery' Act. Peterson ( 1 9 8 3 ) reported that there w e r e
1 2 , 9 9 1 landfills in the United States, including 2 3 9 5 o p e n dumps. T h e r e are an u n k n o w n
n u m b e r o f a b a n d o n e d landfills.
Materials placed in landfills include such things as municipal garbage and trash,
demolition debris, sludge from wastewater-treatment plants, incinerator ash, foundry
sand and o t h e r foundry wastes, and t o x i c and hazardous materials. Although no longer
permitted, liquid hazardous waste was d i s p o s e d in landfills in the past.
Leachate is formed from the liquids found in the waste as well as by leaching of
the solid waste by rainwater. T a b l e 1.8 contains information on the c h e m i c a l c o m p o s i t i o n
o f leachate from municipal landfills. T o minimize the a m o u n t o f leachate generated,
m o d e r n landfills are built in sections, with a l o w permeability c o v e r placed over the waste
as s o o n as p o s s i b l e to limit the infiltration of rainwater. Modern landfills also have lowpermeability liner systems and collection pipes to remove the leachate that forms so
that it can be taken to a wastewater-treatment plant. A m o d e r n landfill that is properly
sited with respect to the local geology and that has a properly designed and c o n s t r u c t e d
liner, leachate collection system, and low-permeability c o v e r has limited potential to
c o n t a m i n a t e g r o u n d water. However, many landfills do n o t have liners and leachate
collection systems. In the past, landfills tended to be placed in any c o n v e n i e n t h o l e or
low spot, such as a sand pit, quarry, or marsh. Ground-water contamination from such
landfills is highly p r o b a b l e .
Municipal landfills are usually located near urban areas. T h e trend is toward large
landfills that can handle many thousands of tons of waste p e r year. Hazardous-waste
landfills are n o w regulated under the R e s o u r c e Conservation and Recovery Act. T h e r e
is frequently strong local o p p o s i t i o n to the siting of either a municipal or a hazardouswaste landfill. T h i s is referred to as the NIMBY syndrome: Not In My B a c k Y a r d !

Chapter One
T A B L E 1.8

Overall summary from the analysis of municipal solid-waste leachates in Wisconsin.


T y p i c a l Range
(range of

Number of

Range"

site medians)"

Analyses

TDS

584-50,430

2180-25,873

172

Specific conductance

480-72,500

2840-15,485

1 167

Parameter

Overall

Total suspended solids

2-140,900

28-2835

2700

N D - 1 95,000

101-29,200

2905

COD

6.6-97,900

11 2 0 - 5 0 , 4 5 0

467

TOC

ND-30,500

427-5890

52

5-8.9

5.4-7.2

1900

BOD

pH
Total alkalinity (CaCOj)

N D - 1 5,050

960-6845

328

52-225,000

1050-9380

404

Chloride

2-11,375

180-2651

303

Calcium

200-2500

200-2100

12-6010

12-1630

192

Hardness (CaC0 )
3

Sodium
Total Kjeldahl nitrogen

2-3320

47-1470

156

Iron

N D - 1 500

2.1-1400

416

Potassium

ND-2800

ND-1375

19

120-780

120-780

Ammonia-nitrogen

N D - 1 200

26-557

263

Sulfate

N D - 1 850

8.4-500

154

ND-85

ND-85

Zinc

ND-731

ND-54

158

Manganese

ND-31.1

0.03-25.9

67

Total phosphorus

ND-234

0.3-117

454

Boron

0.87-13

1.19-12.3

15

Barium

N D - 1 2.5

ND-5

73
133

Magnesium

Aluminum

ND-7.5

ND-1.65

Nitrate-nitrogen

ND-250

ND-1.4

88

Lead

ND-14.2

ND-1.1 1

142
138

Nickel

ND-5.6

ND-1.0

Antimony

ND-3.19

ND-0.56

76

Copper

ND-4.06

ND-0.32

138

Thallium

ND-0.78

ND-0.31

70

Cyanide

ND-6

ND-0.25

86

ND-70.2

ND-0.225

112
7

Chromium

Arsenic

0.01-1.43

0.034-0.193

Tin

ND-0.16

0.16

Nitrite-nitrogen

ND-1.46

ND-0.11

20

Selenium

ND-1.85

ND-0.09

121

ND-0.4

ND-0.07

158

Silver

ND-1.96

ND-0.024

106

Beryllium

ND-0.36

ND-0.008

76

Mercury

ND-0.01

ND-0.001

111

Molybdenum

Cadmium

All concentronons in milligrams pe lirer except pH (stondord units) and specific conductance (^mhos/cm) ND indicates not detected

Source: Wisconsin Oeportmenl of Notural Resources,

Introduction

21

Open dumps

O p e n d u m p s are typically unregulated. T h e y receive waste mainly from h o u s e h o l d s but


are used for almost any type of waste. W a s t e is frequently burned, and the residue is
only occasionally covered with fill. Such d u m p s do not have liners and Ieachate-collection
systems and by their nature are highly likely to cause ground-water contamination. T h e
u s e of o p e n d u m p s in the United States is no longer p o s s i b l e due to 1 9 9 1 EPA regulations
issued under Subtitle D of the R e s o u r c e Conservation and Recovery Act, which requires
extensive ground-water monitoring at such facilities, requires the p l a c e m e n t of daily
cover, prohibits burning, and will require engineered liners for future e x p a n s i o n s . Most
operators o f o p e n d u m p s did not want the e x p e n s e o f such regulations and s o c l o s e d
the dumps.
Residential disposal

H o m e o w n e r s w h o are not served by a trash collection service must find alternative ways
of disposing of their h o u s e h o l d waste. Included in the h o u s e h o l d waste are hazardous
s u b s t a n c e s such as used e n g i n e oil and antifreeze and leftover yard and garden chemicals
such as pesticides, unused paint, and used paint thinner. In the past these w e r e often
taken to the town dump. However, with the closing of m o s t town dumps, the h o m e o w n e r
must find alternative m e a n s of disposal.
In Wisconsin virtually all town d u m p s w e r e c l o s e d in 1 9 8 9 and 1 9 9 0 . Most, but
not all, counties offer waste disposal in a secure, engineered landfill. However, in large
counties the counts- landfill may be 10 to 20 mi from s o m e parts of the c o u n t y and a
fee is charged, as o p p o s e d to the old town dump, which was c l o s e by and free. In s o m e
situations the residents must drive to a different county to find an o p e n landfill. Unfortunately, this closing of town d u m p s has resulted in an increase in illegal dumping in
state and national forests and a great increase in trash left at roadside rest areas and
parks.
H o m e o w n e r s may p o u r waste liquids into ditches or the sanitary sewer; c o m b u s tibles may be b u r n e d in the backyard. T h e s e are undesirable practices that can easily
result in environmental pollution, including ground-water contamination.
Surface

impoundments

Pits, ponds, and lagoons are used by industries, farmers, and municipalities for the
storage a n d / o r treatment of b o t h liquid nonhazardous and hazardous waste and the
discharge of nonhazardous waste. Prior to the passage of the R e s o u r c e Conservation
and Recovery Act, liquid hazardous wastes w e r e also discharged into pits. T h e s e pits
may be unlined or lined with natural material, such as clay, or artificial materials, such
as plastic sheets, r u b b e r m e m b r a n e s , or asphalt.
I m p o u n d m e n t s are used to treat wastewater by such p r o c e s s e s as settling of solids,
biological oxidation, chemical coagulation and precipitation, and pH adjustment. T h e y
may also be used to store wastewater prior to treatment. W a t e r from surface i m p o u n d
ments may be discharged to a receiving water c o u r s e such as a stream or a lake. Unless
a discharging i m p o u n d m e n t is lined, it will also l o s e water by s e e p a g e into the subsurface.
Nondischarging i m p o u n d m e n t s release water either by evaporation or s e e p a g e into the
ground or a c o m b i n a t i o n of both. Evaporation p o n d s are effective only in arid regions,
w h e r e potential vapotranspiration far e x c e e d s precipitation. Even evaporation p o n d s

Chapter One
that were originally lined may leak and result in ground-water contamination if the liner
deteriorates from contact with the pond's contents.
I m p o u n d m e n t s are used for wastewater treatment by municipalities and industries
such as paper manufacturing, petroleum refining, metals industry, mining, and chemical
manufacturing. T h e y are also used for treatment of agricultural waste, such as farm animal
waste from feedlots. P o w e r plants use surface impoundments as cooling ponds. Mining
operations use surface p o n d s for the separation of tailings, which is waste rock from
the processing of o r e that o c c u r s in a slurry.
Although it is n o w prohibited, until the 1 9 7 0 s lagoons were used for the disposal
of untreated wastewater from manufacturing, o r e processing, and o t h e r industrial uses
into the ground water. B r i n e pits were used for many years in the oil patch for the
disposal of brines p u m p e d up with the oil. Miller ( 1 9 8 0 ) lists 57 cases of ground water
contamination caused by the leakage of wastewater from surface impoundments. In
most of the reported c a s e s water-supply wells had b e e n affected; at the time when use
of such i m p o u n d m e n t s was allowed, ground water monitoring was not required; usually
the only way that leakage was detected was by contamination of a supply well.
In o n e c a s e in Illinois, up to 5 0 0 , 0 0 0 gals per day of mineralized wastewater,
containing high total dissolved solids ( T D S ) , which included chloride, sulfate, and calcium,
from an ore-processing plant were discharged into waste disposal p o n d s excavated in
a glacial drift aquifer for a period of about 40 yr. Concentrations of chloride, sulfate,
T D S , and hardness were elevated in an underlying b e d r o c k aquifer as much as a mile
away from the site (U.S. Nuclear Regulatory C o m m i s s i o n 1 9 8 3 ) .
Wastewater from the manufacturing of nerve gas and pesticides at the Rocky
Mountain Arsenal at Denver was discharged into unlined evaporation ponds from 1 9 4 2
until 1956. In 1 9 5 6 a new p o n d lined with asphalt was constructed; ultimately that liner
failed and the lined p o n d also leaked. Contamination of nearby farm wells was first
detected in 1951 and was especially severe in the drought year of 1 9 5 4 , when irrigated
c r o p s died. Ground-water contamination e x t e n d e d at least 8 mi from the ponds and
was indicated by high chloride content. Ultimately the ground water under and near the
Rocky Mountain Arsenal was found to contain d o z e n s of synthetic organic chemicals,
including two that are especially m o b i l e in the subsurface: diisopropylmethylphosphonate
( D I M P ) . a by-product of the manufacture of nerve gas, and dicyclopentadiene ( D C P D )
a chemical used in the manufacture of pesticides ( K o n i k o w a n d T h o m p s o n 1984; Spanggord, Chou, and Mabey 1 9 7 9 ) . It is estimated that the cleanup of contaminated soil and
ground water at the Rocky Mountain Arsenal will ultimately cost m o r e than $1 billion
(U.S. Water News, March, 1 9 8 8 ) .
T h e Environmental Protection Agency performed a survey of the surface impound
m e n t s located in the United States (U.S. EPA 1 9 8 2 ) . T h e y reported a total of 180,9"'3
impoundments, including 3 7 , 1 8 5 municipal, 1 9 , 4 3 7 agricultural, 2 7 , 9 1 2 industrial, 2 5 , 0 3 8
mining, 6 5 , 6 8 8 brine pits for oil and gas, and 5 9 1 3 miscellaneous. T h e large n u m b e r of
impoundments provides a significant threat to ground-water resources ( O T A 1 9 8 4 ) .
Mine wastes

Mining can p r o d u c e spoils, or unneeded soil, sediment, and rock moved during the
mining process, and tailings, or solid waste left over after the processing of ore. T h e s e
wastes may be piled on the land surface, used to fill low areas, used to restore the land

Introduction

23

to premining c o n t o u r s , or placed in e n g i n e e r e d landfills with leachate-collecuon systems.


Mine wastes can g e n e r a t e leachate as rainwater passes through them. If sulfate or sulfide
minerals are present, sulfuric acid can be generated, and the resulting drainage water
can be acidic. T h i s is likely to o c c u r with coal-mining wastes, c o p p e r and gold o r e s , and
o r e s from massive sulfide mineralization. Mine-waste leachate may also contain heavy
metals and, in the c a s e of uranium and thorium mines, radionuclides. Neutralization of
the mine wastes can prevent the formation of acidic leachate and prevent the mobilization
of many, but n o t all, metallic ions and radionuclides. T h e mine-waste disposal issue is
a large o n e , b e c a u s e an estimated 2.3 billion tons of m i n e wastes are generated annually
in the United States. Leachate p r o d u c e d by unneutralized or u n c o n t a i n e d m i n e wastes
is a threat to surface and ground water.
Material stockpiles

Many bulk c o m m o d i t i e s , s u c h as coal, road salt, o r e s , p h o s p h a t e rock, and building


stone, are stored in o u t d o o r stockpiles. Rainwater percolating through the stockpile can
p r o d u c e leachate similar to that p r o d u c e d by the waste material that resulted from mining
the c o m m o d i t i e s . F o r e x a m p l e , rainwater draining through a coal pile can b e c o m e acidic
from sulfide minerals c o n t a i n e d in the coal. In the northern states road salt is usually
stored indoors, although in the past o u t d o o r storage piles w e r e c o m m o n . Leachate from
the road-salt piles was a c o m m o n s o u r c e of ground-water contamination that has n o w
b e e n mostly eliminated.
Graveyards

If b o d i e s are buried without a casket or in a n o n s e a l e d casket, d e c o m p o s i t i o n will release


organic material. Areas of high rainfall with a shallow water table are m o s t s u s c e p t i b l e
to ground-water contamination from graves. According to B o u w e r ( 1 9 7 8 ) contaminants
can include high bacterial c o u n t s , a m m o n i a , nitrate, and elevated c h e m i c a l o x y g e n demand. Nash ( 1 9 6 2 ) reported that hydrogen sulfide gas in a well was the result of a
seventeenth-century graveyard for black plague victims. T h e well had apparently b e e n
unwittingly b o r e d through the graveyard.
Animal burials

Unless an animal is a famous Kentucky t h o r o u g h b r e d or a b e l o v e d family pet, it is likely


to simply be buried in an o p e n excavation. If large n u m b e r s of animals are buried in
c l o s e proximity, ground-water contamination might o c c u r from the d e c o m p o s i n g carcasses. If the animals had died due to s o m e type of t o x i c poisoning, then additional
opportunities for ground-water contamination would exist if the t o x i c c h e m i c a l w e r e
released as the animals d e c o m p o s e d .
A b o v e - g r o u n d storage tanks

Petroleum products, agricultural chemicals, and o t h e r c h e m i c a l s are stored in aboveground tanks. Ruptures or leaks in the tanks can release chemicals, which then have the
opportunity to s e e p into the ground. A serious c a s e of ground-water contamination
o c c u r r e d in Shelb\T.ille, Indiana, when o n e 55-gal tank of p e r c h l o r o e t h y l e n e was damaged
by vandals and the c o n t e n t s leaked into the ground.

Chapter One
U n d e r g r o u n d storage tanks

T h e Office of T e c h n o l o g y Assessment estimates that in the United States there are s o m e


2.5 million underground storage tanks used to store fuel and o t h e r products ( O T A 1 9 8 4 ) .
T h e r e are at least two tanks, and frequently more, at every gas station. Many h o m e o w n e r s
and farmers have private underground tanks to store heating oil and fuel. Chemicals are
also routinely stored in underground tanks at industrial facilities. Liquid hazardous wastes
can also be stored in underground tanks. Leachate from landfills with leachate-collection
systems may be stored in a tank while it awaits trucking to a treatment facility.
Underground tanks can leak through holes either in the tank itself or in any
associated piping. T h e piping appears to be m o r e vulnerable. Steel tanks are susceptible
to c o r r o s i o n and are being replaced by fiberglass tanks. However, even with fiberglass
tanks, the associated pipes can still leak. Fiberglass tanks do not have the strength of
steel and may crack. A gas-station o w n e r with a leaking tank can e n c o u n t e r tens of
thousands of dollars in costs to remove a leaking tank and associated contaminated soil.
Costs can be even higher if extensive ground-water contamination has occurred. In a 1
yr period a small consulting firm made 28 assessments of sites that contained underground
fuel storage tanks. Even though n o n e of the sites was known to have contamination
prior to the assessments, 22 of the 28 sites ( 7 8 % ) were found to have leaking tanks
( G o r d o n 1 9 9 0 ) . If o n e c o n s i d e r e d the sites being investigated b e c a u s e tanks were known
to be leaking, the percentage of leaking tanks would be even higher.
Even tlie h o m e o w n e r is at risk. O n e purchaser of an older h o m e in the town of
B l a c k Wolf, Wisconsin, had the misfortune to discover an abandoned fuel-oil tank buried
on his property. A total of forty-two 55-gal drums of a mixture of fuel oil and water were
removed from the tank and had to be disposed of at considerable e x p e n s e . Fortunately,
as the tank was mostly b e l o w the water table, the water had leaked into the tank, rather
than the fuel oil leaking out. Had the latter occurred, the costs to remove and dispose
of contaminated soil would have b e e n much higher.
Containers

Many chemical and waste products are stored in drums and other containers. Should
these leak, there is a potential for ground-water contamination.
Open incineration and detonation sites

Sites for the o p e n incineration of wastes are licensed under RCRA. In 1981 there were
2 4 0 such facilities in the United States ( O T A 1 9 8 4 ) . T h e Department of Defense operates
burning grounds and detonation sites for old ammunition. Chemicals released from such
sites can leach into the ground with rainwater.
Radioactive-waste-disposal

sites

T h e disposal of civilian radioactive wastes and uranium mill tailings is licensed under
the Nuclear Regulator)' Commission. High-level radioactive wastes from nuclear power
plants are currently in temporary storage but will eventually go into an underground
repository excavated into rock. T h e first repository is planned for Y u c c a Mountain, Nevada
(U.S. Department of Energy 1 9 8 8 ) . Low-level wastes are buried in shallow landfills. Unless
radioactive wastes are properly buried in engineered sites, there is a potential for radionuclides to migrate from the waste into ground water, as happened at Oak Ridge,

Introduction

25

T e n n e s s e e ; Hanford, Washington; Savannah River Facility, Georgia; and the Idaho National
Engineering Lab.
1.5.3

Category III: Sources D e s i g n e d to R e t a i n S u b s t a n c e s D u r i n g T r a n s p o r t

Pipelines

Included in Category III are sewers to transmit wastewater as well as pipelines for the
transmission of natural gas, petroleum products, and o t h e r liquids such as anhydrous
ammonia. Although the pipelines are designed to retain their c o n t e n t s , many leak to a
greater or lesser extent. T h i s is particularly true of sewers, especially older sections.
Sewers usually have a friction joint that can leak if the pipe shifts position. If the sewer
is a b o v e the water table, leaking sewage can c o n t a m i n a t e the g r o u n d water with bacteria,
nitrogen, and chloride. Steel pipelines are s u b j e c t e d to c o r r o s i o n and can also develop
leaks. Such pipelines have b e e n k n o w n to leak c r u d e oil, gasoline, fuel oil, liquified
petroleum gas, natural gas liquids, jet fuel, diesel fuel, k e r o s e n e , and anhydrous a m m o n i a
(OTA 1984).
Material

transport and

transfer

Material transport and transfer o c c u r s by the m o v e m e n t of p r o d u c t s and wastes \ i a truck


and train along transportation corridors and the associated use of loading facilities. Spills
may result from accidents, and leaks can o c c u r b e c a u s e of faulty e q u i p m e n t . A wide
variety of materials can be released to the environment in this manner. E x p e r i e n c e d and
well-trained crews with the p r o p e r e q u i p m e n t are n e e d e d to clean up such spills. Imp r o p e r actions can result in a spill b e c o m i n g m o r e severe as a result of a misguided
cleanup effort.
1.5.4

Category IV: Sources


Planned

Discharging Substances as a

Consequence of Other

Activities

Irrigation

W h e n c r o p s are irrigated, m o r e water is applied to the field than is n e e d e d for vapotranspiration. T h e e x c e s s water, called r e t u r n f l o w , percolates through the soil z o n e
to the water table. In doing so it can mobilize c h e m i c a l s applied to the fields as fertilizers
and pesticides. Soil salinity and salinity of the shallow g r o u n d water can also increase,
b e c a u s e the evaporation of water c o n c e n t r a t e s the natural salts carried in the irrigation
water. Selenium has b e e n c o n c e n t r a t e d in irrigation return water that has b e e n discharged
to the Kesterson Wildlife Refuge in California's Central Valley.
Pesticide

applications

Chemicals are applied to c r o p s to control weeds, insects, fungi, mites, n e m a t o d e s , and


o t h e r pests. In addition they are used for defoliation, desiccation, and growth regulation
( O T A 1 9 8 4 ) . Approximately 5 5 2 million p o u n d s of active ingredients w e r e applied to
c r o p s in the United States in 1 9 8 2 , and there w e r e 2 8 0 million acre-treatments with
pesticides; s o m e land was treated m o r e than o n c e , so the n u m b e r of a c r e s treated is
actually less than 2 8 0 million acres ( O T A 1 9 8 4 ) .
T h e u s e of pesticides has extensive potential for contaminating g r o u n d water.
Pesticides applied to the soil may migrate through the soil to the water table. Pesticides

Chapter O n e

in use today are usually biodegradable to s o m e extent. However, their breakdown


products ( m e t a b o l i t e s ) can also be found in ground water. T h e potential for contamination is higher at sites w h e r e pesticides are m i x e d and application equipment is loaded
and then rinsed when its use is finished. Soils under such areas may receive a much
greater loading of pesticides than the cropland to which the pesticides are applied.
Application of pesticides by aerial spraying may result in uneven distribution. More than
6 5 % of pesticides are applied by aerial spraying, and the cleanup of the planes and
disposal of associated wastewater p o s e s a special p r o b l e m ( O T A 1 9 8 4 ) .
Atrazine has b e e n used extensively for w e e d control in corn cultivation. In 1985,
3.3 million acres of Wisconsin fannland planted with c o r n was treated with it. A survey
of atrazine in Wisconsin ground waters s h o w e d it o c c u r r e d unevenly in areas where it
was used on fields. Highest concentrations, up to 3.5 parts per billion, were associated
with mixing sites and sandy river-bottom land (Wollenhaupt and Springman 1 9 9 0 ) .
Fertilizer application

Farmers and h o m e o w n e r s alike apply fertilizers containing nitrogen, phosphorous, and


potassium ( p o t a s h ) . P h o s p h o r o u s is not very m o b i l e in soil and thus d o e s not p o s e a
significant threat to ground water. T h e rate of potassium application is generally low
and, although it is mobile, the literature d o e s not indicate that potassium from fertilizers
is a major factor in causing ground-water p r o b l e m s . However, nitrogen from fertilizers
can be a major cause of ground-water contamination.
Farm animal wastes

Farm animal wastes have the potential to contaminate ground water with bacteria, viruses,
nitrogen, and chloride. Animals that are kept on an o p e n range disperse their wastes
over a large area, and the potential for environmental contamination is low. Animals
confined to a small area will concentrate their wastes in the barn, barnyard, or feedlot.
Rainwater infiltrating these wastes can mobilize contaminants, which can be leached into
the soil and eventually into ground water. Manure from farms may be spread o n t o fields
as a fertilizer, whereas large feedlot operations often have wastewater treatment plants.
In northern climates manure spread on frozen fields can have a deleterious effect on
b o t h surface and g r o u n d water during the spring melt. Many farms in northern areas
n o w have c o n c r e t e storage tanks for holding manure during the winter months.
Salt application f o r h i g h w a y deicing

Many states in the snowbelt have a dry-pavement policy that requires the use of highway
deicing salts on city streets, rural highways, and interstate highways. T h e primary deicing
salt is rock salt, consisting mainly of sodium chloride. Additives to improve the handling
of the salt include ferric ferrocyanide and sodium ferrocyanide. Chromate and phosphate
may be added to reduce the corrosiveness of the salt ( O T A 1 9 8 4 ) . T h e salt and additives
eventually are carried from the roadway in runoff and may either wash into surface
streams or s e e p into ground water.
Home water softeners

In areas w h e r e the water supply has high calcium and magnesium content, h o m e water
softeners are used to reduce the hardness. H o m e water softeners are recharged with

27

Introduction

sodium chloride salt. Chlorides from the salt are contained in the backwash water. If
the area is not served by sewers, the backwash water is d i s p o s e d by subsurface drainage
via septic tanks or separate drain fields. Chlorides from this s o u r c e can enter the ground
water reservoir (Hoffman and Fetter 1 9 7 8 ) .
Urban

runoff

Precipitation over urban areas typically results in a greater proportion of runoff and less
infiltration than that falling on nearby rural areas b e c a u s e of the greater a m o u n t of
impervious land surface in the urban area. In addition, the urban runoff contains high
amounts of dissolved and s u s p e n d e d solids from auto emissions, fluid leaks from vehicles,
h o m e use of fertilizers and pesticides, refuse, and pet feces. F o r the most part, the urban
runoff is carried into surface receiving waters, but it may recharge the water table from
leaking storm sewers. T h i s can c o n t r i b u t e to degradation of ground-water quality in
urban areas.
Percolation of atmospheric pollutants

Atmospheric pollutants reach the land either as dry deposition or as dissolved or particulate matter contained in precipitation. S o u r c e s include a u t o m o b i l e emissions, powerplant s m o k e s t a c k s , incinerators, foundries, and o t h e r industrial p r o c e s s e s . Pollutants
include hydrocarbons, synthetic organic chemicals, natural organic c h e m i c a l s , heavy
metals, sulfur, and nitrogen c o m p o u n d s . Infiltrating precipitation may carry these c o m p o u n d s into the soil and g r o u n d water.
Mine drainage

Surface and underground mining may disrupt natural ground water flow patterns and
e x p o s e rocks containing pyrite to oxygenated water. T h i s can result in the production
of acid water, which then drains from the mine. T h e acid m i n e drainage can result in
surface- and ground-water contamination. In o n e very interesting c a s e in Shullsburg,
Wisconsin, a lead and zinc m i n e was active for 25 yr. In o r d e r to w o r k the mine, the
ground water table was lowered b e l o w the m i n e levels by pumping. Sulfide minerals in
the rock were s u b j e c t e d to biologically mediated oxidation along fractures in the rock
and mine workings. Contact of the resulting sulfuric acid with the dolomite host rock
neutralized the sulfuric acid and p r o d u c e d highly s o l u b l e sulfate minerals. W h e n the
mining c e a s e d d u e to e c o n o m i c factors, the dewatering p u m p s w e r e shut down and the
mine workings were Hooded. G r o u n d water in the mine workings dissolved the sulfate
minerals and resulted in high sulfate ( u p to 3 5 0 0 m g / L ) , iron ( u p to 20 m g / L ) , and zinc
( u p to 18 m g / L ) c o n c e n t r a t i o n s . As a result ground-water quality of a n u m b e r of nearby
water supply wells was adversely impacted (Hoffman 1 9 8 4 ) .

1.5.5

Category V: Sources P r o v i d i n g a Conduit for Contaminated Water to


Enter

Aquifers

Production wells

Wells are drilled for the production of oil, gas, geothermal energy, and water. Contaminants can be introduced into the g r o u n d during the drilling of production wells. Improperly constructed wells, c o r r o d e d well casings, and improperly a b a n d o n e d wells can

Chapter One
provide a conduit for the flow of contaminated surface water into the ground or the
movement of contaminated ground water from o n e aquifer into another. Homeowners
may route drainage water from their r o o f and b a s e m e n t drains into abandoned watersupply wells. Old dug wells may b e c o m e receptacles for trash.
Monitoring w e l l s and exploration borings

Many thousands of monitoring wells are being installed in the United States each year.
Exploration borings are installed for the purposes of mineral exploration or construction
design. T h e s e wells and borings have the same potential for c r o s s contamination of
aquifers and introduction of contaminated surface water as production wells.
Construction excavation

Construction activities can strip the soil from bedrock, thus removing much of the natural
protection of b e d r o c k aquifers from ground water contamination. Urban runoff water
can collect in o p e n foundation excavations, which then provide a conduit to aquifers.

1.5.6

Category V I : N a t u r a l l y O c c u r r i n g S o u r c e s W h o s e D i s c h a r g e i s Created a n d / o r
Exacerbated by H u m a n Activity

Ground-water-surface-water

interactions

S o m e aquifers are recharged naturally from surface water if the stream stage is higher
than the water table (Fetter 1 9 8 8 ) . If the surface-water b o d y b e c o m e s contaminated,
then the aquifer being recharged by that water could also b e c o m e contaminated. An
e x c e p t i o n to this might o c c u r if the surface-water contamination is by a material that
could be adsorbed or removed by filtration when it passes through the alluvium under
the stream. Wells located near a stream can induce infiltration from the stream into the
ground-water reservoir by development of a c o n e of depression. Contaminated surface
water can thereby be drawn into an aquifer.
Natural leaching

Dissolved minerals o c c u r in ground water due to natural leaching from rocks and soil.
Naturally occurring ground water may have total dissolved solids in e x c e s s of 1 0 , 0 0 0 to
1 0 0 , 0 0 0 m g / L and may contain undesirable concentrations of various anions and cations.
Human activity that results in acid rain may e n h a n c e the ability of infiltrating rainwater
to leach naturally occurring s u b s t a n c e s from rock and soil.
Saltwater intrusion

Development of freshwater supplies from coastal aquifers may lower the water table
and induce saline ground water that o c c u r s naturally beneath the o c e a n s to move
landward into formerly freshwater aquifers. Upconing of the saltwater-freshwater interface
may also o c c u r if the well field overlies an aquifer containing saline water (Fetter 1 9 8 8 ) .
Ground-water development in areas susceptible to saltwater intrusion should be undertaken with a clear plan that is designed to maximize the amount of fresh water that
can be developed while minimizing the amount of saltwater intrusion and other undesirable effects that can o c c u r (Fetter 1 9 7 2 ) .

Introduction

1.6

29

Relative Ranking of Ground-Water-Contamination Sources


Although there are many potential s o u r c e s o f ground-water contamination, s o m e p o s e
m u c h m o r e of a threat to g r o u n d water than others. Section 3 0 5 ( b ) of the Federal Clean
Water Act requires individual States to submit reports to the Environmental Protection
Agency on the s o u r c e s of ground water contamination in the state and the type of
contaminants observed. T h e data submitted w e r e used to c o m p i l e National Water Quality
/mentor)1988

Report

to

Congress

(U.S.

EPA

1990).

T h e states indicated all the g r o u n d - w a t e r - c o n t a m i n a t i o n s o u r c e s that they considered to be major threats to g r o u n d water in their state. Figure 1.2 s h o w s that m o r e
than half the states and territories listed underground storage tanks, septic tanks, agricultural activities, municipal landfills, and a b a n d o n e d hazardous-waste sites as major
threats to g r o u n d water. O t h e r frequently listed s o u r c e s include industrial landfills, o t h e r
landfills, injection wells, regulated hazardous-waste sites, land application, road salt,
saltwater intrusion, and b r i n e pits from oil and gas wells.
States and territories could also give a priority ranking from 1 to 5, 1 b e i n g the
highest priority, for the various s o u r c e s of g r o u n d water contamination. Figure 1.3 s h o w s
these priority rankings for the s o u r c e s listed on Figure 1.2, w h e r e a s Figure 1.4 s h o w s
these rankings for the individual s o u r c e s g r o u p e d as "other" on Figure 1.2. T h e highestpriority ranking was given to underground storage tanks with 15 states listing this as the

Source
Other
Oil and gas brine pils
Saltwater intrusion

Road sailing
Land application
Regulated hazardous waste sites
Injection wells
Other landfills

Industrial landfills
Abandoned hazardous waste sites
Surface impoundments
Municipal landfills
Agricultural activity
Septic tanks
Underground storage tanks
0

10

20

15

25

30

35

40

Number of states and territories


F I G U R E 1.2

Frequency of various contamination sources considered by states and territories of the

United States to be major threats to ground-water quality. Source: National Water Quality Inventory,
1 9 8 8 Report to Congress, Environmental Protection Agency, 1 9 9 0 .

45

50

Sou r t e
Road Salting

I si Priori!)

2nd Priority

3rd Priority

gl

4th Priority

Industrial Landfills
Sail Waler Intrusion
Land Application
Regulated Hazerdous Waste Sites
Other Landfills
5th Priority

Injection Wells
Oil and Gas Brine Pits
Municipal Landfills
Surface Impoundments
Septic Tanks

1
^ ^ ^ ^ ^ ^

Agricultural Activity
Abandoned Hazardous Waste Sites
Underground Storage Tanks

I
0

10

15

20

25

Number of States and Territories


F I G U R E 1.3

Priority ranking of contamination sources considered by more than 10 states and territories

of the United States to be a major threat to ground-water quality. Source: National Water Quality Inventory,
1 9 8 8 Report to Congress, Environmental Protection Agency, 1 9 9 0 .

Source
Municipal wastewater [.

. .

1st priority

Tankyard
Agricultural well
Sink hole
Fertilizer

and

pesticides

Y///////////7?

2nd priority

3rd priority

pj^jl 4th priority

Waste piles \
U

5th priority

Salt storage
Abandoned wells
Urban runoff
Above-ground storage tanks
Spills

Feed lots
Manufacturing
Construction
Cyanide heaps
Sewer leaks
Mining wastes

Number of states and territories


F I G U R E 1.4

Priority ranking of contamination sources considered by fewer than 10 states and territories

of the United States to be a major threat to ground-water qualify. Source.- National Water Qualify Inventory,
1 9 8 8 Report to Congress, Environmental Protection Agency, 1 9 9 0 .

31

Introduction

Contaminant
Other inorganics

Fluorides

Arsenic
Other agricultural contaminants

Radioactive material

Coliform bacteria

Synthetic organic chemicals

Brine

Metals

Petroleum products

Volatile organic compounds

Pesticides

30

35

Nitrates
0

10

15

20

25

]
I
*

J I
40

45

Number of states and territories


F I G U R E 1.5

Frequency of various contaminants considered by states and territories of the United States

to be a major threat to ground-water quality. Source: National Water Quality Inventory, 1 9 8 8 Report to
Congress, Environmental Protection Agency, 1 9 9 0 .

top p r o b l e m and nine listing it as the s e c o n d most severe p r o b l e m . Rounding out a


"dim- half-dozen" are a b a n d o n e d hazardous-waste sites, agricultural activity, septic tanks,
surface impoundments, and municipal landfills. Figure 1 . 4 s h o w s that s o m e states have
unique p r o b l e m s that are severe at the state level but are not particularly p r o b l e m s a c r o s s
the country. Manufacturing, construction, cyanide heaps, and sewer leaks w e r e listed as
the n u m b e r o n e priority by o n e state but, with the e x c e p t i o n of sewer leaks, w e r e not
even ranked by any o t h e r state.
T h e states also reported the ground-water contaminants o f c o n c e r n . T h e s e data
are reported in Figure 1.5. T h e m o s t frequently reported contaminants w e r e nitrates,
pesticides, volatile organic c o m p o u n d s , petroleum products, metals, brine, and synthetic
organic chemicals. O t h e r parameters of i m p o r t a n c e w e r e bacteria, radioactive materials,
o t h e r agricultural contaminants, arsenic, fluoride, and o t h e r inorganics.

1.7

Ground-Water Contamination as a Long-Term Problem


O n e of the factors of ground water contamination that m a k e s it so serious is its longterm nature. Wastes buried long a g o may c a u s e ground-water contamination that takes
d e c a d e s to be discovered. Although many ground w a t e r - c o n t a m i n a t i o n sites are small,
s o m e of the long-term sites are fairly extensive d u e to the long time period over which
contamination has b e e n migrating away from the s o u r c e .
In the 1 9 3 0 s poison baits utilizing arsenic w e r e used in the Midwest to c o u n t e r a
g r a s s h o p p e r infestation. Apparently, leftover p o i s o n bait was buried w h e n the infestations

32

Chapter One
ended. In 1 9 7 2 a water-supply well was drilled for a small business. In short order, 11
of 13 e m p l o y e e s b e c a m e ill with arsenic poisoning. Tests of the well showed it contained
21 m g / L of arsenic and soil at the site had 3 0 0 0 to 1 2 , 0 0 0 m g / L of arsenic. This was
apparently a mixing or burial site for arsenic-laden g r a s s h o p p e r bait (American Water
Resources Association 1 9 7 5 ) .
Beginning in 1 9 1 0 waste fuel oil and solvents from a railroad yard were discharged
into the dry b e d of the Mojave River near Barstow, California. A study in 1 9 7 2 showed
that a z o n e of contaminated ground water e x t e n d e d nearly 4.25 mi from the site and
was 1 8 0 0 ft wide ( H u g h e s 1 9 7 5 ) .
Starting in 1 9 3 6 a seepage lagoon was used for the disposal of treated domestic
sewage at the Otis Air F o r c e base, C a p e Cod, Massachusetts. Over a 50-yr period about
2.5 billion gals of treated sewage was discharged into the rapid-infiltration ponds.
T h e sewage percolated through the unsaturated z o n e and recharged a shallow sand and
gravel aquifer. B e c a u s e of the high rate of ground-water flow, about 1.0 to 1.5 ft per day,
the plume has migrated m o r e than 2 mi downgradient. T h e plume can be traced by
elevated c o n c e n t r a t i o n s of chloride, b o r o n , nitrate, detergents, and volatile organic c o m
pounds. T h e plume is narrow and thin due to limited transverse dispersion ( H e s s 1 9 8 8 ) .
A coal-tar distillation and wood-preservative plant was operated from 1 9 1 8 to 1972
at St. Louis Park, Minnesota. Coal tar, which is obtained by heating coal in the a b s e n c e
of air, is a c o m p l e x mixture of hundreds of organic c o m p o u n d s , including polynudear
aromatic hydrocarbons ( P A H ) . T h e coal tar was distilled to form creosote, which was
then used as a w o o d preservative. Coal-tar chemicals and c r e o s o t e entered the environment by spills and drippings at the wood-preservative facility as well as via plant p r o c e s s
discharge water, which went into ponds. Coal tar is denser and m o r e viscous than water
and is only slightly soluble. T h e coal-tar c o m p o u n d s migrated downward into the underlying glacial drift aquifer. Several old, d e e p wells on the site had defective casings,
which allowed coal tar to migrate downward into deep, b e d r o c k aquifers. O n e 595-ftd e e p well on the site was found to contain a c o l u m n of coal tar 100 ft long. About
1 5 0 gal/min of contaminated water was entering this well from the glacial drift aquifer
through a leak in the casing. T h i s water then drained downward into the d e e p b e d r o c k
aquifers, carrying contamination with it. After 60 yr of leakage the contamination had
spread m o r e than 2 mi from the plant site in several directions. Water supply wells located
outside of the area of contamination have drawn contaminated water into the b e d r o c k
aquifers up gradient of the site in terms of the regional ground-water-flow direction
(Hult and Stark 1 9 8 8 ) .

1.8

R e v i e w of Mathematics and the F l o w Equation


1.8.1

Derivatives

Soil-moisture movement, ground water flow, and solute transport may be described by
m e a n s of partial differential equations. Thus, a b r i e f review is in order.
If a bicyclist is traveling down a highway, we can measure the time that it takes
the rider, w h o has a flying start, to go from a starting point ( Si / , ) . or the location at the
starting time, / , ) to a point s o m e w h e r e down the highway ( 5 ( r ) , or the location at
2

Introduction

33
w

elapsed time / ) - I f e wish t o k n o w the average s p e e d o f the rider over this distance,
2

we divide the distance from point S ( r , ) to point 5 ( / ) by the elapsed time, t fj.
2

AS

_S0 )-SQ )

At

h - h

T h e rider will be going m o r e slowly uphill and faster downhill. T h e average s p e e d will
thus include a lot of variation. If we w e r e to m e a s u r e the rider's s p e e d over a shorter
part of the c o u r s e , there would be less variation in s p e e d . As the length of time over
which the distance traveled is m e a s u r e d b e c o m e s shorter and shorter, the variation in
s p e e d d e c r e a s e s . If the time b e c o m e s infinitesimally s m a l l f o r e x a m p l e , the time that
it takes the rider to travel a few m i c r o n s w e obtain an instantaneous speed. T h i s is
known a s the f i r s t d e r i v a t i v e o f d i s t a n c e w i t h r e s p e c t t o t i m e and i s defined b y
dS(h)
dt

,.
= lim

SO)-SO,)

(1-2)

<-<*,

w h e r e t is any arbitrary time. Figure 1.6 s h o w s a graph of distance traveled by o u r


bicyclist as a function of time. T h e s l o p e of the line from time r, to time t is the average
2

s p e e d over that p a n of the highway and is e x p r e s s e d as A SI A t. T h e instantaneous s p e e d

F I G U R E 1.6

Graph of distance traveled versus time graphically showing speed, which is the first derivative

of distance with respect to time.

Chapter One

8 6

F I G U R E 1.7

4 I

10

15
Time (sec)

20

25

30

35

Graph of speed versus time graphically showing acceleration, which is the second derivative

of distance with respect to time.

at time
is the s l o p e of the tangent to the curve at that point, which is e x p r e s s e d as
dS/dt.
Note that the s l o p e of distance versus time on Figure 1.6 keeps changing. This
reflects the changes in speed that o c c u r as the rider g o e s up and down hills. As the rider
g o e s over the crest of a hill, he or she will perhaps be going rather slowly. As the rider
g o e s downhill, the velocity will increase. We can c o m p a r e the crest-of the hill velocity
with the bottom-of-the-hill velocity and see that it has increased. This is a measure of
the acceleration that o c c u r s as gravity and the leg muscles of the bicyclist c o m b i n e to
increase speed. Figure 1.7 shows the speed of the rider as he or she g o e s over a hill.
At t = 0 the rider is coming over the crest of the hill and the speed is 10 mi/hr. At t = 30
s e c , w h e n the rider is near the b o t t o m of the hill, the speed is 26 mi/hr. T h e average
rate of c h a n g e in speed is ( 2 6 m i / h r 10 m i / h r ) / 3 0 sec, or 0.53 m i / h r / s e c . T h e rate of
c h a n g e is faster near the top of the hill where the slope is steeper and there is less wind
resistance, since the rider is moving m o r e slowly. From 0 to 5 sec the speed changes
from 10 to 15 mi/hr, or 1.0 m i / h r / s e c . Acceleration is the rate of change of speed with
time, which is a s e c o n d derivative. It is the slope of a tangent to the curve at a given
time. It can be e x p r e s s e d as

dS

35

Introduction

T h e tangent at 5 s e c can be seen to be steeper than the tangent at 30 sec, w h e r e the


rate of c h a n g e is less.
In hydrogeology we have many parameters that are a function of m o r e than o n e
independent variable. F o r e x a m p l e , hydraulic head is a function of the three s p a c e
variables: b = b(x, y, z). We frequently differentiate head with respect to o n e of the
s p a c e variables while holding the o t h e r two variables constant. Such derivatives of a
parameter with respect to a single variable are called p a r t i a l d e r i v a t i v e s . T h e s e c o n d
derivative of hydraulic head with respect to the s p a c e variables is
2

8h
8x
1

1.8.2

Dorcy's

db
Iy
1

dh
8z
1

Law

T h e first e x p e r i m e n t a l study of ground-water flow was p e r f o r m e d by Henry Darcy ( D a r c y


1 8 5 6 ) . He found that the one-dimensional flow of water through a pipe filled with sand
was proportional to the c r o s s sectional area and the head loss along the pipe and inversely
proportional to the flow length. Darcy's law can be e x p r e s s e d as

Q=-KA

<*
al

(1.3)

where
Q = volumetric discharge
K= proportionality c o n s t a n t k n o w n as hydraulic conductivity
A = cross-sectional area

dhldl =

gradient o f hydraulic head

T h i s equation can also b e e x p r e s s e d i n terms o f s p e c i f i c d i s c h a r g e , o r D a r c y


flux, q, which is the volumetric flow rate, Q, divided by the cross-sectional area, A.
, = - K -

(1.4)

Darcy's law was obtained for one-dimensional flow. However, as was previously
stated, head is a function of all three dimensions: h = h(x, y, z).
T h e hydraulic conductivity is the measure of the ability of the fractured or p o r o u s
media to transmit water. It can have different values, depending u p o n the actual direction
that the water is flowing through the p o r o u s media. In such a c a s e the medium is said
t o b e a n i s o t r o p i c . T h e value o f the hydraulic conductivity can b e measured i n three
principle directions, K , IC and AT.. If the hydraulic conductivity is the s a m e in all direc
x

tions, then IC. = Ky = K = AT and the medium is said to be i s o t r o p i c .


2

1.8.3

Scaler, Vector, and T e n s o r P r o p e r t i e s of H y d r a u l i c Head and


Hydraulic Conductivity

We first need to define s o m e terms relating to t e n s o r s . A zero-order tensor, also called


a s c a l e r , is a quantity characterized only by its size or magnitude. E x a m p l e s in hydro
geology include hydraulic head, chemical concentration, and temperature. A first-order

Chapter One
tensor, or v e c t o r , is a quantity that has both a magnitude and a direction. Vectors
require three c o m p o n e n t s , each having a magnitude and direction. Velocity, specific
discharge, mass flux, and heat flux are e x a m p l e s . A s e c o n d order t e n s o r o r . simply,
t e n s o r a c t s like the product of two vectors, requiring nine c o m p o n e n t s to account
for all p o s s i b l e products of the three c o m p o n e n t s of e a c h vector. E x a m p l e s in hydrogeology are intrinsic permeability, hydraulic conductivity, thermal conductivity, and hydrodynamic dispersion.
T h e hydraulic head is a scaler. However, the gradient of the head is a vector as it
has b o t h a magnitude and a direction. T h e gradient of h is designated as grad h:

dh
ox

dh . dh
oy
oz

grad i = i - + ) + k

(1-5)

w h e r e i, j, and k are unit vectors in the x, y, and z directions. An equivalent notation


is the use of the vector differential operator, del, which has the symbol V. This operator
is equivalent to

(1.6)

dy

dx

Another vector is the specific discharge, q. It has three c o m p o n e n t s , q , q , and


x

q., when measured along the Cartesian coordinate a x e s . Associated with any vector is
a positive scaler with a value equal to the magnitude of the vector. If q is the magnitude
of the vector q, this can be e x p r e s s e d as

?=|q|

(1-7)

A s e c o n d - o r d e r tensor, such as K, hydraulic conductivity, can be described by nine


c o m p o n e n t s . In matrix form they are e x p r e s s e d as:
Kix

Key

Ky

Ky

Ky

K,

K,

A"_.

Ktz

(1.8)

If the t e n s o r is symmetric, Ay = A},; then inspection of ( 1 . 8 ) shows that there are


only six independent c o m p o n e n t s of K.
If the c o o r d i n a t e system is oriented along the principal axes, the tensor b e c o m e s
0

0
0

(1-9)

K
F o r the special c a s e of an isotropic mediathat is, the value of K d o e s not d e p e n d
upon the direction in which it is m e a s u r e d t h e tensor b e c o m e s
A"

A'

0
0
AT

(1.10)

Introduction

37

T h e three c o m p o n e n t s of the specific discharge vector, q, are

dh
"z
dh

dh
dh
dx
dh
dh
>*ex~ *dy~
dh
dh
"dx~

(.ii)

dh
"Tz

F o r the special c a s e w h e r e we orient the a x e s of the x, y, and z c o o r d i n a t e system


with the three principal directions of anisotropy, K is the matrix shown in ( 1 . 9 ) and the
three c o m p o n e n t s o f the specific discharge vector are

dh
-KxxT
ox
dh

fx =
q,=

-K,,-f

fl-M)

dh
K

*>=- Tz
F o r an isotropic material, K is represented by the matrix in ( 1 . 1 0 ) and

q= -K-

dh
dh
dh
KK
ox
oy
oz

q= -ATgrad h

.
(1.13)

(1.14)

If we multiply two vectors together and the result is a scaler, then the product is
called a d o t p r o d u c t , o r i n n e r p r o d u c t . F o r e x a m p l e , the del o p e r a t o r dotted into
a vector yields a scaler, called the d i v e r g e n c e . B a s e d on grad h, we can find a velocity
vector v such that the magnitude and direction vary throughout the p o r o u s media. If
we apply the del o p e r a t o r to v, we obtain the following:

dv, dv
dx oy

dv,
dz

V v = divv = -^ + ^ + -^

(1.15)

If we apply the del o p e r a t o r to grad h, the result is the s e c o n d derivative of head:


2

dh dh Bh
V-gdh= + +
dx
dy Yz
2

1.8.4

(1.16)

D e r i v a t i o n of the F l o w Equation in a D e f o r m i n g M e d i u m

T h e l a w o f m a s s c o n s e r v a t i o n states that there can b e n o net c h a n g e i n the mass


of fluid in a small representative elementary volume ( R E V ) of a p o r o u s medium. In o t h e r
words, the mass entering the REV less the mass leaving the REV is equal to the c h a n g e
in mass storage with time.

38

Chapter

One

z
F I G U R E 1.8

Representative elementary volume for fluid flow.

T h e representative elementary volume is shown on Figure 1.8. T h e three sides


have length dx, dy, and dz, respectively. T h e area of the two faces normal to the x axis
is dy dz, the area of the faces normal to the y axis is dx dz, and the area of the faces
normal to the z axis is dx dy.
T h e c o m p o n e n t of mass flux into the REV parallel to the praxis is the fluid density
times the flux r a t e
Mass influx along x axis = p q dy dz
w

(1.17)

where
p

= f l u i d density ( M / L )

q = specific discharge or volume of flow per c r o s s sectional area ( L / T )


x

dydz = cross-sectional area ( L )


T h e units of mass inflow are mass per unit time ( M / T ) . *

* The units of a variable can be expressed in terms of their fundamental dimensions. These are length, L
mass, M, and time, T. T h e fundamental dimensions for density' are mass per unit volume. Volume is length
cubed, so the shorthand expression for the fundamental dimensions of density is M t. . Specific discharge
has the dimensions of velocity, so the fundamental dimensions are I. T, and area has fundamental dimensions
3

off. .

39

Introduction

T h e mass outflow rate will be different than the inflow rate and can be given as:

8(p q ) dx
dx
w

Mass outflow rate parallel to x axis

dydz

(1.18)

T h e net mass accumulation within the control volume due to the flow c o m p o n e n t
parallel to the x axis is the mass inflow minus the mass outflow, or

d(p q ) dxdydz
dx
w

Similar terms exist for the net mass accumulation due to flow c o m p o n e n t s parallel
to the y and z axes:

d(p q ) dydxdz
8y
d(p q.) dzdxdy
~d~z
w

T h e s e three terms can be s u m m e d to find the total net mass accumulation within
the control volume.

dxdydz

( P ^ ) + ~ ( P ^ ) + ( P W = )

(1-19)

T h e mass of water in the REV, M, is the density of water, p, rimes the porosity,

n,

times the volume,

dxdydz.

T h e c h a n g e in mass with respect to time is

dM d
(p n dxdydz)
dt dt

(1.20)

F r o m the law of conservation of mass, Equation 1.19 must equal Equation 1.20.
( P ^ ) + ^ ( P ^ ) + ( P ^ )

dxdydz

= (p )
w

dxdydz

(1.21)

dt
We can a s s u m e that although density of the fluid may c h a n g e with time, at any
given time it will be the s a m e everywhere in the REV. U n d e r this assumption Equation
1.21 can be simplified to

dq
dx

dq dq
dy
dz\

- t

d
(Pw)
p dt

(1.22)

We may substitute Darcy's law for the specific discharge c o m p o n e n t s given on the
left side. If the xyz c o o r d i n a t e system is aligned with the principal a x e s of anisotropy,
then Equation 1.12 may be used, and the left side of Equation 1.22 b e c o m e s

d_l dh\ d_l dh\


dx\ dx)
dyX^ dy)
Kxx

d_
dz

db\

(1.23)

T h e c h a n g e in mass within the REV is due to c h a n g e s in the porosity and the


density of water as the head c h a n g e s with time. T h u s the c h a n g e in the volume of water
in storage is proportional to the c h a n g e in head with time. T h e right side of Equation

Chapter O n e

1.22 can be e x p r e s s e d as a proportionality constant, 5 the specific storage, times the


change in head with time.
I d ,
-

dh

^ = S.J

(1-24)

C o m b i n i n g Equations 1.22, 1.23, and 1.24 we obtain the main equation for tran
sient flow in an anisotropic medium when the coordinate system is oriented along the
principal a x e s o f anisotropy:
d I
k

dh\

t \ - ai)
1.8.5

d (

dh\

d (
+

dh\

dh

= s

( 1

Ty) T [ " Fz) 'J


Z

'

2 5 )

Mathematical Notation

In del and tensor notation Equation 1.25 b e c o m e s


V-K-Vh = S

(1.26)

Another form of e x p r e s s i o n is called Einstein's summation notation. For example.


Darcy's law in the familiar, o n e dimensional form is
dh
C = - K -

(1.27)

It is implied in the preceding equation that the specific discharge is parallel to the
direction of dlj dl and that the medium is isotropic. In a m o r e general form, specific
discharge, q, is a vector with c o m p o n e n t s ty,, q , and <y . Grad h is a vector that we will
call h. This vector also has c o m p o n e n t s h h , and h . Hydraulic conductivity. K. is a
tensor with nine c o m p o n e n t s . To d e s c r i b e Darcy's law in the most general form, we
need three equations.
2

lt

1x = Knhi + K h + A-,3*.

(1.28a)

q = Kb + Kh +

(1.28b)

i2

2l

22

q, = K h, + K h + A- *,
3l

32

33

(1.28c)

T h e inner product can be e x p r e s s e d in index notation as


1 i = I,K,jb
i

(i,y=l,2,3)

(1.29)

In Einstein's summation notation, the is dropped with the understanding that


the summation is over the repeated indices:
q = Kh
i

tl

0,7=1,2,3)

(1-30)

In vector notation this can be e x p r e s s e d as either


q = K-grad/>

(1-31)

or
q = Kh

(1.32)

41

Introduction

In del notation this is

q=K-Vh

(1.33)

In general, we will use the standard form of differential equations rather than any
of the shorthand notation. However, the literature cited in this text often uses the c o m p a c t
forms and the reader should be aware of them.

References
American Water Resources Association. 1975. Status of waterb o r n e diseases in the U.S. and Canada. Journal of
American Water Works Association 67, no. 2 : 9 5 - 9 8 .
Ames, B. N., Renae Magaw, and L S. Gold. 1987. Ranking
possible carcinogenic hazards. S c i e n c e 2 3 6 (April 17,
1987):271-77.
Bouwer, Herman. 1978. Groundwater hydrology. New York:
McGraw-Hill B o o k Company, pp. 4 2 3 - 2 4 .
Burmaster, D. R. and R H. Harris. 1982. Groundwater contamination, an emerging threat. Technology Review
85, no. 5 : 5 0 - 6 2 .
Darcy, Henry. 1856. Les fontaines publiques de la ville de
Dijon. Paris: Victor Dalmont, 647 pp.
Eckhardt, D. A, and E. T. Oaksford. 1988. "Relation of land
use to ground-water quality in the upper glacial aquifer, Long Island, New York." In National Water Summary 1986Hydrologic Events and Ground Water
Quality- U.S. Geological Survey Water Supply Paper
2325, pp. 1 1 5 - 1 2 1 .
Fetter, C. W. 1972. T h e c o n c e p t of safe groundwater yield in
coastal aquifers. Water Resources Bulletin 8, no.
5:1173-76.
. 1988. Applied' Ijydrogeology. 2 d ed. Columbus, Ohio:
Merrill Publishing Company, 5 8 8 pp.
Gordon, J a m e s . 1990. OMNI Engineers, personal communication.
Hanmer, Rebecca. 1989. Environmental Protection Agency,
Testimony in a hearing on the seriousness and extent
of ground water contamination before the Senate Subcommittee on Superfund, Ocean and Water Protec
tion of the Committee of Environment and Public
Works, August 1, 1989.
Hess, K. M. 1988. "Sewage Plume in a sand and gravel aquifer,
Cape Cod, Massachusetts." In National Water Sum
mary'. 1986,87-92. U.S. Geological Survey Water Sup
ph/ Paper 2325.
Hindall, S. M and Michale Eberle. 1989. National and regional
trends in water well drilling in the United States,
1 9 6 4 - 8 4 . U.S. Geological Survey Circular 1029,
15 pp.
Hoffman, J. I. 1984. "Geochemistry of acid mine drainage on
the aquifers of southeastern Wisconsin and regulatory
implications." In Proceedings of the National Water
Well Association Conference on the Impact of Mining
on Ground Water, 1 4 6 - 1 6 1 . August 2 4 - 2 7 , 1984,
Denver, Colorado, National Water Well Association,
Dublin, Ohio.

Hoffman, J. I., and C. W. Fetter. 1978. Water softener salt: a


major source of groundwater contamination. Geological Society of America, Abstracts with Programs
10, no. 7 : 4 2 3 .
Hughes, J. L 1975. Evaluation of ground-water degradation
resulting from waste disposal to alluvium near Bar
stow, California. U.S. Geological Survey Professional
Paper 8 7 8 .
Hult, M. F., and J. R. Stark. 1988. Coal-tar derivatives in the
Prairie du C h i e n - J o r d a n Aquifer, St. Louis Park, Minnesota. In National Water Summary, 1986. U.S. Geological Survey Water Supply Paper 2325, pp. 8 7 - 9 2 .
Konikow, L F., and D. W. T h o m p s o n . 1984. Groundwater
contamination and aquifer reclamation at the Rocky
Mountain Arsenal, Colorado. In Groundwater Conlamination. Washington, D.C.: National Academy
Press. 9 3 - 1 0 3 .
Lehr, J. H. 1990a. Toxicological risk assessment distortions:
Part I. Ground Water 28, no. 1 : 2 - 8 .

. 1990b. Toxicological risk assessment distortions:


Part I I T h e d o s e makes the poison. Ground Water
28, no. 2 : 1 7 0 - 5 .
Miller, David. 1980. Waste disposal effects on ground water.
Berkley, Calif.: Premier Press, 512 pp.
Nash, G. J. C. 1962. Discussion of a paper by E. C. Wood.
Proceedings of the Society of Water Treatment and
Examination
11:33.
Office of Technology Assessment. 1984. Protecting the na
lion's groundwater from contamination.
Washington, D.C: U.S. Congress, OTA 0 - 2 7 6 , two volumes.
Peterson, N. M. 1983- 1983 survey of landfills. Waste Age
(March 1 9 8 3 ) : 3 7 - 4 0 .
Solley, W. B . , C. F. Merk, and R. R. Pierce. 1988. Estimated
use of water in the United States in 1985. U.S. Geological Survey Circular 1004, 82 pp.
Spanggord, R J . , Tsong Wen Chou, and W. R Mabey. 1979.
Studies of environmental fates of DIMP and DCPD.
Contract report by SRI International, Menlo Park,
Calif., for U.S. Army Medical Research and Develop
ment Command, Fort Detrick, Fredrick, Md, 65 pp.
U.S. Department of Energy. 1988. Site characterization plan
overview. Yucca Mountain Site, Nevada Research
and Development Area. Washington, D. C: Office of
Civilian Radioactive Waste Management, 164 pp.
U.S. Environmental Protection Agency. 1982. Surface impoundment assessment national report. Office of
Drinking Water.

42

Chapter One

. 1990. National u aler quality inventory. 1988 Report


to Congress, EPA 4 4 0 4-90-003, 187 pp.
U.S. Nuclear Regulatory Commission. 1983. Final environmental statement related to the decommissioning of
the Rare Earths Facility, Vi'est Chicago, Illinois.
NUREG-0904.
U.S. Water News. 1988. At Jl billion, Colorado arsenal is
costliest cleanup yet. 4, no. 9 (March 1 9 8 8 ) : 1.
ran der Leeden, Fritz, F. L Troise, and D. K. Todd. 1990. The
water encyclopedia, Chelsea, Mich: Lewis Publishers.

Wilson, Richard, and E. A. C. Crouch. 1987. Risk assessment


and comparisons: An introduction. Science. 236 (April
17, 1 9 8 7 1 : 2 6 7 - 7 0 .
Wollenhaupt, N. C, and R. E. Springman. 1990. Atrazine in
groundwater: A current perspective. University of
WisconsinExtension, Agricultural Bulletin G3S2S,
17 pp.

Chapter T w o

Mass Transport in Saturated Media

2.1

Introduction
In this c h a p t e r we will c o n s i d e r the transport of solutes dissolved in g r o u n d water. This
is known as m a s s or s o l u t e t r a n s p o r t . The m e t h o d s presented in this c h a p t e r are
b a s e d on partial differential equations for dispersion that have b e e n developed for
h o m o g e n e o u s media ( O g a t a and B a n k s 1 9 6 1 ; Ogata 1 9 7 0 ; B e a r 1 9 7 2 ; B e a r and Verruijt
1 9 8 7 ) . T h e s e equations are similar in form to the familiar partial differential equations
for fluid flow. In recent years much work has b e e n d o n e on the theories of mass transport
in r e s p o n s e to the great interest in p r o b l e m s of ground-water contamination. O n e of
the o u t c o m e s of this has b e e n the d e v e l o p m e n t of what is essentially a n e w b r a n c h of
subsurface hydrology, w h e r e the flow of fluid and solutes is treated by statistical models;
these m o d e l s can a c c o u n t for the role of varying hydraulic conductivity that a c c o m p a n i e s
aquifer heterogeneity. Very recently fractal g e o m e t r y has b e e n used to d e s c r i b e the solute
transport b a s e d on the c o n c e p t that aquifer heterogeneities have repeating patterns.

2.2

T r a n s p o r t by Concentration Gradients
A solute in water will m o v e from an area of greater c o n c e n t r a t i o n toward an area w h e r e
it is less c o n c e n t r a t e d . T h i s p r o c e s s is k n o w n as m o l e c u l a r d i f f u s i o n , < >r d i f f u s i o n .
Diffusion will o c c u r as long as a concentration gradient exists, even if the fluid is not
moving. T h e mass of fluid diffusing is proportional to the c o n c e n t r a t i o n gradient, which
can be e x p r e s s e d as F i c k ' s first l a w ; in o n e dimension, Fick's first law is
F=

(2.1)

-D (dC/dx)
d

where
F = m a s s flux of solute per unit area p e r unit time
D

= diffusion coefficient ( L / T )
3

C= solute c o n c e n t r a t i o n ( M / L )
3

dC/dx = c o n c e n t r a t i o n gradient ( M / L / L )

43

44

Chapter Two

T h e negative sign indicates that the movement is from areas of greater concentration to
t h o s e of lesser concentration. Values for D are well known and range from 1 x 1 0 ~
to 2 x 1 0 " m / s e c at 2 5 C . They do not vary much with concentration, but they are
somewhat temperature-dependent, being about 5 0 % less at 5C ( R o b i n s o n and Stokes
9

1965).
For systems w h e r e the concentrations are changing with time, F i c k ' s s e c o n d
l a w applies. In o n e dimension this is
2

(2.2)

8C/8t= D 8 C/8x
d

where

8C/8t=

c h a n g e in concentration with time ( M / L / T ) .

In p o r o u s media, diffusion c a n n o t p r o c e e d as fast as it can in water b e c a u s e the


ions must follow longer pathways as they travel around mineral grains. To a c c o u n t for
this, an effective diffusion coefficient, >*, must be used.

(2.3)

D' = 0)D

w h e r e to is a coefficient that is related to the tortuosity ( B e a r 1 9 7 2 ) . T o r t u o s i t y is a


measure of the effect of the shape of the flowpath followed by water molecules in a
p o r o u s media. If L is the straight-line distance between the ends of a tortuous flowpath
of length l , the tortuosity, T, can be defined as 7 " = LjL. Tortuosity in a p o r o u s media
e

is always greater than 1, b e c a u s e the flowpaths that water molecules take must diverge
around solid particles. Flowpaths across a representative sample of a well-sorted sediment
would tend to be shorter than those a c r o s s a poorly sorted sediment in which the smaller
grains were filling the voids between the larger grains. Thus the well-sorted sediment
would tend to have a lower value for tortuosity than the poorly sorted sediment. ( T o r 2

tuosity has also b e e n defined as (L/L ) (Carman 1937; B e a r 1 9 7 2 ) . With this definition,
e

tortuosity always has a value less than 1. This definition will not be used in this text.)
T h e value of to, which is always less than 1, can be found from diffusion experiments
in which a solute is allowed to diffuse a c r o s s a volume of a p o r o u s medium. Perkins
and J o h n s o n ( 1 9 6 3 ) found that <o was equal to 0.7 for sand column studies using a
uniform sand. According to F r e e z e and Cherry ( 1 9 7 9 ) , to ranges from 0.5 to 0.01 for
laboratory studies using g e o l o g i c materials.
Diffusion will c a u s e a solute to spread away from the place where it is introduced
into a p o r o u s medium, even in the a b s e n c e of ground-water flow. Figure 2.1 shows the
distribution of a solute introduced at concentration Q, at time <o, over an interval (x a)
to (x + a.) At succeeding times r, and / , the solute has spread out, resulting in a lower
2

concentration over the interval (x a) to (x + a) but increasing concentrations outside


of this interval.
T h e solute concentration follows a normal, or Gaussian, distribution and can be
2

d e s c r i b e d by two statistical properties, the mean, C, and variance, c r , which are defined
c

in Section 2.12.2.
T h e effective diffusion coefficient, D*', can be defined ( D e Josselin D e j o n g 1 9 5 8 )
as
(2.4)

M a s s T r a n s p o r t i n Saturated M e d i a

45

T h i s is an alternative definition of effective diffusion coefficient to the o n e given in Equation


2.3.
T h e p r o c e s s of diffusion is c o m p l i c a t e d by the fact that the ions must maintain
electrical neutrality as they diffuse. If we have a solution of NaCl, the N a

c a n n o t diffuse

faster than the CI~ unless there is s o m e o t h e r negative ion in the region into which the
Na

is diffusing

If the solute is a d s o r b e d o n t o the mineral surfaces of the p o r o u s

medium, the net rate of diffusion will be obviously less that for a n o n a d s o r b e d species.
Diffusion can o c c u r w h e n the c o n c e n t r a t i o n of a c h e m i c a l s p e c i e s is greater in
o n e stratum than in an adjacent stratum. F o r e x a m p l e , solid waste containing a high
c o n c e n t r a t i o n of chloride ion may be placed directly on the clay liner of a landfill. T h e
c o n c e n t r a t i o n of chloride in the leachate contained in the solid waste is so much greater
than the c o n c e n t r a t i o n of chloride in the p o r e water of the clay liner that the latter may
be c o n s i d e r e d to be z e r o as a simplifying assumption in determining a conservative
estimate of the m a x i m u m diffusion rate. If the solid waste and the clay are b o t h saturated,
the chloride ion will diffuse from the solid waste, w h e r e its c o n c e n t r a t i o n is greater, into
tlie clay liner, even if there is no fluid flow. T h e c o n c e n t r a t i o n of chloride in the solid
waste, CQ, will be assumed to be a constant with time, as it can be replaced by dissolution
of additional chloride. T h e c o n c e n t r a t i o n of chloride in the clay liner, Q (x, t), at s o m e
distance x from the solid waste interface and s o m e time t after the waste was placed,
can be determined from Equation 2.5 (Crank, 1 9 5 6 ) . This is a solution to Equation 2.2
for the appropriate boundary and initial conditions.
t

er[c

d(x, ) = o , . o.s
2 D l)

(2.5)

Chapter T w o

where
Q = the concentration at distance x from the s o u r c e at time t since diffusion began
Qi = the original concentration, which remains a constant
erfc = the complementary error function (Appendix A)
T h e complementary e r r o r function, erfc, is a mathematical function that is related
to the normal, or Gaussian, distribution. This m e a n s that the solution described by
Equation 2.5 is normally distributed, as is e x p e c t e d for a diffusional process. Figure 2.2
shows the profile of relative concentration for a solute diffusing from a region where
the concentration is Q to a region where it was initially zero. B e c a u s e the profile is
normally distributed, 8 4 % of the values will be less than the value that is o n e standard
deviation m o r e than the mean and 1 6 % of the values will be less than the value that is
o n e standard deviation less than the mean. T h e standard deviation is the square root of
the variance.
T h e complementary error function is tabulated in Appendix A. It is related to the
e r r o r function, erf, by
erfc(S) = 1 - erf(B)
T h e value of e r f c ( B ) is 0 for all positive values of B greater than 3.0 and 1.0 for
a B of 0. F o r s o m e applications it may be necessary to find erfc of a negative number.
Appendix A d o e s not give values for e r f c ( B ) for negative values of B. T h e s e must be
c o m p u t e d from the relationship
e r f c ( - B ) = 1 + erf B

+a

- a

FIGURE 2.2

The profile of a diffusing front as predicted by the complementary error function.

47

Mass Transport in Saturated Media

T h u s values of e r f c ( B ) range from 0 to + 2, since the m a x i m u m value of e r f ( B )


is T.O for 3.0 and all greater n u m b e r s .

EXAMPLE
PROBLEM

Assume a D o f 1 x 1 0 ~ m / s e c and an to o f 0.5, to give a D* o f 5 x 10

1 0

m /sec.

Find the value o f the c o n c e n t r a t i o n ratio, CJCQ, at a distance o f 5 m after 1 0 0 yr o f


diffusion.
1. Convert 1 0 0 yr to s e c o n d s :
9

1 0 0 yr x 3 6 5 days/yr x 1 4 4 0 m i n / d a y x 60 s e c / m i n = 3.15 x 1 0 s e c
2. Insert values into Equation 2.5:

SL = erfc
q,

2(5 x 1 0 "

1 0

m / s e c x 3.15 x 1 0 s e c )

0 5

3. Solve:

= erfc ( ) = erfc 1.99 = 0 . 0 0 5

In 100 yr, diffusion over a 5-m distance would yield a c o n c e n t r a t i o n that is 0 . 5 % of the
original.

F r o m the p r e c e d i n g e x a m p l e p r o b l e m it is obvious that diffusion is not a particularly


rapid m e a n s of transporting dissolved solutes. Diffusion is the p r e d o m i n a n t m e c h a n i s m
of transport only in low-permeability hydrogeologic regimes. However, it is p o s s i b l e for
solutes to m o v e through a p o r o u s or a fractured medium by diffusion even if the ground
water is not flowing.

2.3

Transport by Advection
Dissolved solids are carried along with the flowing g r o u n d water. T h i s p r o c e s s is called
a d v e c t i v e t r a n s p o r t , < >r c o n v e c t i o n . T h e a m o u n t of solute that is b e i n g transported
is a function of its c o n c e n t r a t i o n in the g r o u n d water and the quantity of the ground
water flowing. F o r one-dimensional flow normal to a unit cross-sectional area of the
p o r o u s media, the quantity o f water flowing is equal to the
the

effective porosity.

Average linear velocity,

v,
x

average linear velocity

times

is the rate at which the flux o f

water a c r o s s the unit cross-sectional area of p o r e s p a c e o c c u r s . It is n o t the average rate


at which the water m o l e c u l e s are moving along individual Ilowpaths, which is greater
than the average linear velocity due to tortuosity. T h e e f f e c t i v e p o r o s i t y , n , is the
e

porosity through which flow can o c c u r . N o n i n t e r c o n n e c t e d and dead-end p o r e s are not


included in the effective porosity.

Chapter T w o

where
r = average linear velocity (L/T)
A' = hydraulic conductivity (L T)

n , effective porosity
dhiil =

hydraulic

gradient

(LI)

T h e one-dimensional mass Bux, l ' , d u e to advection is equal (o t h e quantity o f


x

water Bowing times the c o n c e n t r a t i o n of dissolved solids and is given by Equation Z.":
h, = v n,C

(2.7)

T h e o n e dimensional advective transport equation is


BC

BC

Bt

ox

( T h e derivation of this equation is given in Section 2.6.)


Si ilution of the advective transport equation yields a sharp concentration front. On
the advancing side o f the front, the concentration is equal t o that o f the invading ground
water, whereas on the o t h e r side of the front it is unchanged from the background value
I hi-, is known as p l u g H o w , with ill the | x i r e fluid being replaced by the invading
solute front. The sharp interface that results from plug flow is shown in Figure 2.3 T h e
vertical dashed line at I represents an advancing solute front due to advection .done
Due to the heterogeneity o f g e o l o g i c materials, advective transpon in different
Strata can result in solute fronts spreading at different rates in each strata. If o n e obtains
a sample of water for p u r p o s e s of monitoring the spread of a dissolved contaminant
from a b o r e h o l e that penetrates several strata, the water sample will be a c o m p o s i t e of
the water from e a c h strata. Due to the fact that advection will transport solutes at different
rales in e a c h stratum, the c o m p o s i t e sample may be a mixture of water containing the
transported solute c o m i n g from o n e stratum and uncontaminated ground water c o m i n g
from a different stratum w h e r e t h e average linear velocity is lower. T h e concentration
of the contaminant in the c o m p o s i t e sample would thus be less than in the source.

V position ot input
j

Tracer front if """"^


diffusion onl>

water ut lime /

Dispelled
tracer front

F I G U R E 2.3

Distance i

Advective transport and the influence of longitudinal dispersion and diffusion on the transport

of a solute in one-dimensional flow. Source-. C W. Fetter, Applied Hydrogeology, 2d ed. (New York:
Macmillan Publishing Company, 1 9 8 8 ) .

Mass Transport in Saturated Media

2.4

49

Mechanical Dispersion
G r o u n d water is moving at rates that are b o t h greater and less than the average linear
velocity. At the m a c r o s c o p i c s c a l e t h a t is, over a d o m a i n including a sufficient volume
that the effects of individual p o r e s are averaged ( B e a r 1 9 7 2 ) t h e r e are three b a s i c
causes of this p h e n o m e n o n : ( 1 ) As fluid moves through the pores, it will move faster
in the c e n t e r of the p o r e s than along the edges. ( 2 ) S o m e of the fluid particles will travel
along longer flow paths in the p o r o u s media than o t h e r particles to go the s a m e linear
distance. ( 3 ) S o m e p o r e s are larger than others, which allows the fluid flowing through
these p o r e s to m o v e faster. T h e s e factors are illustrated in Figure 2.4.
If all g r o u n d water containing a solute were to travel at exactly the s a m e rate, it
would displace water that d o e s n o t contain the solute and create an abrupt interface
b e t w e e n the two waters. However, b e c a u s e the invading solute-containing water is n o t
all traveling at the s a m e velocity, mixing o c c u r s along the flowpafh. T h i s mixing is called
m e c h a n i c a l d i s p e r s i o n , and it results in a dilution of the solute at the advancing

Pore
size

Path
length

Friction
in pore

F I G U R E 2.4

Factors causing longitudinal dispersion at the scale of individual pores. Source: C. W. Fetter,

Applied Hydrogeo/ogy, 2d ed. (New York: Macmillan Publishing Company, 1 9 8 8 ) .

Chapter T w o

edge of flow. T h e mixing that o c c u r s along the direction of the flowpath is called
longitudinal dispersion.
An advancing solute front will also tend to spread in directions normal to the
direction of flow b e c a u s e at the p o r e scale the flowpaths can diverge, as shown in Figure
2.5. T h e result of this is mixing in directions normal to the flow path called t r a n s v e r s e
dispersion.
If we assume that mechanical dispersion can be described by Fick's law for diffusion
(Equations 2.1 and 2 . 2 ) and that the amount of mechanical dispersion is a function of
the average linear velocity, then we can introduce a coefficient of mechanical dispersion.
T h i s is equal to a property of the medium called dynamic dispersivity, or simply dispersivity, a, times the average linear velocity. If i is the principle direction of flow, the
following definitions apply:
Coefficient of longitudinal mechanical dispersion = x Vj
t

(2.9)

where
v = the average linear velocity in the i direction ( L / T )
t

or, = the dynamic dispersivity in the (direction ( L )


and
Coefficient of transverse mechanical dispersion = a,- v

(2.10)

where
f = the average linear velocity in the /'direction ( L / T )
f

y.J the dynamic dispersivity in the j direction ( L )

F I G U R E 2.5

Flowpaths in a porous medium that cause lateral hydrodynamic dispersion. Source: C. W .

Fetter, Applied Hydrogeology. 2d ed. (New York: Macmillan Publishing Company,! 9 8 8 ) .

Mass Transport in Saturated Media

2.5

51

Hydrodynamic Dispersion
T h e p r o c e s s o f m o l e c u l a r diffusion c a n n o t b e separated from mechanical dispersion i n
flowing g r o u n d water. T h e two are c o m b i n e d to define a parameter called the h y d r o d y n a m i c d i s p e r s i o n c o e f f i c i e n t , D. It is represented by the following formulas:
D = <x v + D*

(2.11a)

D = a v + D*

(2.11b)

where
D = hydrodynamic dispersion coefficient parallel to the principal
L

direction o f f l o w (longitudinal)
D = hydrodynamic dispersion coefficient perpendicular to the principal
T

direction o f f l o w ( t r a n s v e r s e )
a = longitudinal dynamic dispersiviry
t

a = transverse dynamic dispersiviry


r

Figure 2.3 s h o w s the effect of diffusion and mechanical dispersion on the relative
c o n c e n t r a t i o n (C/C ) o f a solute acting as a tracer that has b e e n injected into a p o r o u s
0

medium under one-dimensional flow conditions. T h e vertical line at V represents the


advective transport without dispersion. Effects of diffusion and mechanical dispersion
are shown.
T h e p r o c e s s of hydrodynamic dispersion can be illustrated by Figure 2.6. A mass
of solute is instantaneously introduced into the aquifer at time /<, over the interval
x = 0 + a. T h e resulting initial c o n c e n t r a t i o n is Q,. T h e advecting g r o u n d water carries
the mass of solute with it. In the p r o c e s s the solute slug spreads out, so that the m a x i m u m
c o n c e n t r a t i o n d e c r e a s e s with time, as shown for times r and t . T h e diffusional m o d e l
t

of hydrodynamic dispersion predicts that the c o n c e n t r a t i o n curves will have a Gaussian


distribution that is d e s c r i b e d by the m e a n and the variance. With this distribution the

FIGURE 2.6

Transport and spreading of o solute slug with time due to advection and dispersion. A slug

of solute was injected at x = 0 + a at time f with a resulting concentration of C . The ground-water flow
0

is to the right.

52

Chapter T w o

coefficients of longitudinal and transverse hydrodynaniic dispersion can be defined as


L = -^

(2.12a)
2

r
t = ^ J

(2.12b)

where
t = time

2.6

o>

= variance of the transverse spreading of the plume

<r

= variance of the longitudinal spreading of the plume

Derivation of the Advection-Dispersion Equation for Solute Transport


This derivation of the advection dispersion equation is based on work by Freeze and
Cherry ( 1 9 7 9 ) , B e a r ( 1 9 7 2 ) , and Ogata ( 1 9 7 0 ) Working assumptions are that the p o r o u s
medium is h o m o g e n e o u s , isotropic, and saturated with fluid and that flow conditions
are such that Darcy's law is valid.
T h e derivation is based on the conservation of mass of solute flux into and out
of a small representative elementary volume ( R E V ) of the p o r o u s media. T h e flow is at
a m a c r o s c o p i c scale, which m e a n s that it a c c o u n t s for the differences in flow from pore
to pore. A representative elementary volume is illustrated in Figure 1.8.
T h e average linear velocity, r, has c o m p o n e n t s v

xl

v and v . T h e concentration
r

of solute, C, is mass per unit volume of solution. Mass of solute per unit volume of
aquifer is the product of the porosity, n , and C. Porosity is considered to be a constant
e

b e c a u s e the aquifer is h o m o g e n e o u s .
T h e solute will be transported by advection and hydrodynamic dispersion. In the
/ d i r e c t i o n the solute transport is given by
Advective transport = i\n CdA
t

Dispersive transport = n,D, dA

(2.13)
(2.14)

w h e r e dA is the c r o s s sectional area of the element and the / direction is normal to that
c r o s s sectional face.
T h e total mass of solute per unit c r o s s sectional area transported in the / direction
per unit time, /;, is the sum of the advective and the dispersive transport and is given
by
F, = v,n.C- n.D,

dC

(2.15)

ci

T h e negative sign indicates that the dispersive flux is from areas of greater to areas of
lesser concentration.
T h e total a m o u n t of solute entering the representative elementary volume is
F dzdy-Y

l'

dzdx+

h.

dx dy

M a s s T r a n s p o r t i n Saturated M e d i a

53

T h e total a m o u n t of solute leaving the representative elementary volume is


SP
F

"

, \

~Bx

SF

~8^

(
d

8F

~Sz

T h e difference b e t w e e n the mass of the solute entering the representative elementary volume and the a m o u n t leaving it is
8F
-r +
By

BF,

+
Bx

BF \

BzJ

dxdydz

T h e rate of mass c h a n g e in the representative elementary volume is


BC
n,
Bt

dxdydz

By the law of mass conservation, the rate of mass c h a n g e in the representative


elementary volume must be equal to the difference in the mass of the solute entering
and the mass leaving.
BF

BF,

BF.

BC

_
(2 l6)

tx ty Tz-- 'T

Equation 2.15 can be used to find the values of F , F , and F . T h e s e are substituted
x

in Equation 2.16, which b e c o m e s , after cancellation of n from b o t h sides,


c

8 ( 8C\
Tx\ *Tx)
D

8 ( 8c\
Ty{ >Ty)

d ,
-{v C)
Bx

By

8 ( BC
irz{ <-8

. ,
{v C){v C)
*
Bz
y

(2.17)

Bt

Equation 2.17 is the three dimensional equation of mass transport f o r a conservative


s o l u t e t h a t is, o n e that d o e s not interact with the p o r o u s media or undergo biological
or radioactive decay.
In a h o m o g e n e o u s medium, D , D , and D. do not vary in space. However, b e c a u s e
x

the coefficient of hydrodynamic dispersion is a function of the flow direction, even in


an isotropic, h o m o g e n e o u s medium, D ^ D = D . F o r t h o s e d o m a i n s w h e r e the avx

erage linear velocity, v , is uniform in space, Equation 2.17 for one-dimensional flow in
x

a h o m o g e n e o u s , isotropic p o r o u s media is
2

8C
BC
BC
^ n - v =
Bx
' dx
Bt

,
(2.18)
'

In a h o m o g e n e o u s medium with a uniform velocity field, Equation 2.17 for twodimensional flow with the direction of flow parallel to the x axis is
z

dC
D

3C

^ -

8C

BC

.
( 2

^ T x= Tt

where
2

D = the longitudinal hydrodynamic dispersion ( L / T )


L

D = the transverse hydrodynamic dispersion ( L / T )


T

1 9 )

54

Chapter T w o

Equation 2.17 for radial flow from a well can be written in polar coordinates (Ogata
1 9 7 0 ) as

d_l

dc\

dr\

dr

DdC

BC_dC

r dr

dr

(2.20)

dt

where
r = radial distance to the well

u = average p o r e velocity of injection, which is found from

where

Znn^Rr

Q = the rate of injection into the well


n = effective porosity
e

R = length of well screen or o p e n b o r e h o l e

2.7

Diffusion versus Dispersion


In the previous section the mass transport equation was derived on the basis of hydrodynamic dispersion, which is the sum of mechanical dispersion and diffusion. It would
have b e e n p o s s i b l e to separate the hydrodynamic dispersion term into the two c o m ponents and have separate terms in the equation for them. However, as a practical matter,
under most conditions of ground-water flow, diffusion is insignificant and is neglected.
It is p o s s i b l e to evaluate the relative contribution of mechanical dispersion and
diffusion to solute transport. A P e c l e t n u m b e r is a dimensionless n u m b e r that can
relate the effectiveness of mass transport by advection to the effectiveness of mass
transport by either dispersion or diffusion. Peclet n u m b e r s have the general form of
v d/D or v L/D , w h e r e v
x

is the advective velocity, d and L are characteristic flow

lengths, D is the coefficient of molecular diffusion, and D is the longitudinal hydro


d

dynamic dispersion coefficient. T h e Peclet number, P, which defines the ratio of transport
by advection to the rate of transport by molecular diffusion in column studies, is a
dimensionless parameter defined as v d/D , where d is the average grain diameter and
x

D is the coefficient of molecular diffusion. A plot of the ratio of DJD versus the Peclet
d

n u m b e r is given in Figure 2.7a. Shown on this figure are the results of a n u m b e r of


experimental measurements using sand c o l u m n s and tracers as well as s o m e experimental
curves from several investigators (Perkins and J o h n s o n 1963).
At zero flow velocity D is equal to D', since D = i v + D*. In this manner the
L

value of co, the tortuosity factor, can be experimentally determined as D* = wD . At very


d

low velocities, the ratio of DJD is a constant with a value of about 0.7, which is the
d

experimentally determined value of to for uniform sand. This shows up on the left side
of Figure 2 . 7 ( a ) as a horizontal line. In this z o n e diffusion is the predominant force, and
dispersion can be neglected. B e t w e e n a Peclet n u m b e r of about 0.4 to 6 there is a
transition zone, w h e r e the effects of diffusion and longitudinal mechanical dispersion
are m o r e or less equal.

55

Mass Transport in Saturated Media

= longitudinal dispersion coefficient, cm /sec


2

D = molecular diffusion coefficient, c m / s e c


d

v = average solution velocity, cm/sec


d = average diameter of the particles, cm

Advcctivc
dispersion
controls

Diffusion
controls
I

I I MMI

0.01

I I INII

I I IMN

I I INII

I I Mil

0.1

vd
x

(a)

D = transverse dispersion coefficient, c m / s e c


T

D = molecular diffusion coefficient, c m / s e c


d

= average solution velocity, cm/sec

d = average diameter of the particles, cm

Diffusion
controls

F I G U R E 2.7

Graph of dimensionless dispersion coefficients versus Peclet number, P = v d/D . (a) DJD
t

versus P and (b) D / D versus P. Source-. T. K. Perkins and O. C. Johnson, Society of Petroleum Engineers
r

Journal, 3 (1 9 6 3 ) : 7 0 - 8 4 . Copyright 1 9 6 3 , Society of Petroleum Engineers.

Figure 2 . 7 ( b ) s h o w s the plot of D /D as a function of Peclet n u m b e r . Although


the curve has the s a m e shape as in ( a ) , it o c c u r s at Peclet n u m b e r s roughly 1 0 0 times
greater. T h i s m e a n s that diffusion has m o r e c o n t r o l over transverse dispersion at higher
Peclet n u m b e r s than it d o e s for longitudinal dispersion. Higher Peclet n u m b e r s o c c u r
with higher velocities a n d / o r longer flow paths. At higher Peclet n u m b e r s mechanical
dispersion is the predominant cause of mixing of the contaminant plume (Perkins and
T

56

Chapter T w o

J o h n s o n 1963; B e a r 1972; B e a r and Verruijt 1 9 8 7 ) and the effects of diffusion can be


ignored. Under these conditions D , can b e replaced with v , i n the advection-dispersion
equations.

2.8

Analytical Solutions of the Advection-Dispersion Equation


2.8.1

Methods of Solution

T h e advection-dispersion equations can be solved by either numerical or analytical


methods. Analytical m e t h o d s involve the solution of the partial differential equations
using calculus based on the initial and boundary value conditions. They are limited to
simple geometry and in general require that the aquifer be h o m o g e n e o u s . A n u m b e r of
analytical solutions are presented in this chapter. They are useful in that they can be
solved with a calculator and a table of e r r o r functions or even a pencil and paper, if
o n e is so inclined
Numerical m e t h o d s involve the solution of the partial differential equation by
numerical m e t h o d s of analysis. T h e y are m o r e powerful than analytical solutions in the
s e n s e that aquifers of any geometry can be analyzed and aquifer heterogeneities can be
a c c o m m o d a t e d . However, there can be other p r o b l e m s with numerical models, such as
numerical errors, which can cause solutions to s h o w e x c e s s spreading of solute fronts
or plumes that are n o t related to the dispersion of the tracer that is the subject of the
modeling. B e a r and Verruijt ( 1 9 8 7 ) present a g o o d introduction to the use of numerical
m o d e l s to solve mass transport equations. T h e s e solutions are normally found by meth
o d s of c o m p u t e r modeling, a topic b e y o n d the s c o p e of this text.
2.8.2

B o u n d a r y and Initial Conditions

In o r d e r to obtain a unique solution to a differential equation it is necessary to specify


the initial and the boundary conditions that apply. T h e i n i t i a l c o n d i t i o n s describe
the values of the variable under consideration, in this case concentration, at s o m e initial
time equal to 0. T h e b o u n d a r y c o n d i t i o n s specify the interaction between the area
under investigation and its external environment.
T h e r e are three types of boundary conditions for mass transport. T h e boundary
condition of the first type is a f i x e d c o n c e n t r a t i o n . T h e boundary condition of the
s e c o n d type is a fixed g r a d i e n t . A v a r i a b l e flux boundary constitutes the boundary
condition of the third type.
B o u n d a r y and initial conditions are shown in a shorthand form. For one-dimensional flow we n e e d to specify the conditions relative to the location, x, and the time,
I. By convention this is shown in the form

eu

= at)

w h e r e C{t) is s o m e known function.


F o r e x a m p l e , we can write
C(0, / )

Co

/ > 0

C U 0)

o,

x>

0,

t >o

C(oo,

t)

57

M a s s T r a n s p o r t i n Saturated M e d i a

T h e first statement says that for all time t equal to or greater than zero, at x = 0
the c o n c e n t r a t i o n is maintained at Q. This is a fixed-concentration boundary condition
located at jc = 0 (first-type b o u n d a r y ) . T h e s e c o n d statement is an initial condition that
says at time / = 0, the concentration is z e r o everywhere within the flow domain, that is,
w h e r e x is greater than or equal zero. As s o o n as flow starts, solute at a concentration
of Q, will c r o s s the x = 0 boundary.
T h e third condition s h o w s diat the flow system is infinitely long and that no matter
h o w large time gets, the concentration will still be z e r o at the e n d of the system (firsttype boundary' condition at x= c o ) .
We could also have specified an initial condition that within the domain the initial
solute concentration was q. This would be written as
C(x, 0)

= q,

x> 0

O t h e r e x a m p l e s of concentration (first type) boundary conditions are exponential decay


of the s o u r c e term and pulse loading at a constant concentration for a period of time
followed by another period of time with a different constant concentration.
Exponential decay for the s o u r c e term c a n be e x p r e s s e d as
C(0,

t)

0,1-'"

w h e r e i = a decay constant.
Pulse loading w h e r e the concentration is q, for times from 0 to tg and then is 0
for all time m o r e than tg is e x p r e s s e d as

c(o, r) = q> o <


C(0, /) = 0

t < <o

/ > <o

Fixed-gradient boundaries are e x p r e s s e d as

= A0

dC
dx

dC

or

AO

dx

w h e r e fit) is s o m e k n o w n function. A c o m m o n fixed-gradient condition is dC/dx= 0,


or a no-gradient boundary.
T h e variable-flux boundary, a third type, is given as
dC
D
ox

v C=
x

v Cit)
x

w h e r e Cit) is a k n o w n concentration function. A c o m m o n variable-flux boundary is a


constant flux with a constant input concentration, e x p r e s s e d as
/

dC
D

2.8.3

\
= vCo

+VC

\- lx )

x=0

O n e - D i m e n s i o n a l Step Change in Concentration ( F i r s t - T y p e B o u n d a r y )

Sand c o l u m n e x p e r i m e n t s have b e e n used to evaluate b o t h the coefficients of diffusion


and dispersion at the laboratory scale. A rube is filled with sand and then saturated with
water. Water is m a d e to flow through the tube at a steady rate, creating, in effect, a
permeameter. A solution containing a tracer is then introduced into the sand c o l u m n

Chapter T w o

in place of the water. T h e initial concentration of the solute in the column is zero, and
the concentration of the tracer solution is Q. T h e tracer in the water exiting the tube
is analyzed, and the ratio of the tracer concentration at time /, over Q, the injected tracer concentration, is plotted as a function of time. This is called a fixed-step
function.
T h e boundary and initial conditions are given by
C(x, 0) = 0

x> 0

Initial condition

C ( 0 , t) = Co

t > 01
>
C ( o o , t) = 0 / > 0 J

Boundary conditions

T h e solution to Equation 2.18 for these conditions is (Ogata and B a n k s 1 9 6 1 )


Co

C=

( ^ ) M )

(2.21)

This equation may be e x p r e s s e d in dimensionless form as

' (1 - <)

CjtOji, P) = 0 . 5 ^ e r f c

exp(P )
e

erfc

+ ')
(2.22)

where
t

v t/L
x

C = CI Co
R

P, = Peclet n u m b e r when flow distance, I, is c h o s e n as the reference length


(P =v L/D )
e

erfc :
2.8.4

c o m p l e m e n t a r y e r r o r function

O n e - D i m e n s i o n a l C o n t i n u o u s Injection into a F l o w Field


(Second-Type Boundary)

In nature there are not many situations w h e r e there would be a sudden change in the
quality of the water entering an aquifer. A much m o r e likely condition is that there would
be leakage of contaminated water into the ground water flowing in an aquifer. F o r the
one-dimensional case, this might be a canal that is discharging contaminated water into
an aquifer as a line s o u r c e (Figure 2 . 8 ) .
T h e rate of injection is c o n s i d e r e d to be constant, with the injected mass of the
solute proportional to the duration of the injection. T h e initial concentration of the
solute in the aquifer is zero, and the concentration of the solute being injected is CoT h e solute is free to disperse both up gradient and down gradient.
T h e boundary and initial conditions are
C(x, 0) = 0,

oo < . * < + oo

f " n C(x, t)dx= Q,n v t,


J-
C(oo, r) = 0
e

r>01
>
r>0j

Initial condition
Boundary condtuons

59

M a s s T r a n s p o r t i n Saturated Media
Plan view

77777777777777777.
F I G U R E 2.8

Leakage from a canal as a line source for injection of a contaminant into an aquifer. Source:

J. P. Saufy, Wafer Resources Research 16, no. 1 (1 9 8 0 ) : 1 4 5 - 5 8 . Copyright by the American Geophysical
Union.

T h e s e c o n d b o u n d a r y c o n d i t i o n states that t h e injected mass o f c o n t a m i n a n t over the


d o m a i n from go to + oo is p r o p o r t i o n a l to the length of time of the injection.
T h e solution to this flow p r o b l e m (Saury 1 9 8 0 ) is
(vxL\

erfc

(L+

v,t

2JD.I

(2.23)

In d i m e n s i o n l e s s form this is
C(fR,/3(,) = 0 . s | e r f c ^ ^

'

e x p f P , . ) erfc

(1 + ' * )

(2.24)

It c a n be s e e n that Equations 2.21 and 2.23 are very similar, the only difference
being that the s e c o n d term is subtracted rather than added in 2.23.
Saury ( 1 9 8 0 ) gives an a p p r o x i m a t i o n for the o n e dimensional dispersion equation
as
C

(2.25)

2V//J

In d i m e n s i o n l e s s form this is
CrUR,P,)

0.5

erfc

(1

Ir)

(2.26)

60

F I G U R E 2.9

Chapter T w o

Dimensionless-type curves for the continuous injection of a tracer into a one-dimensional

flow field. Source: J. P. Sauty, Wafer Resources Research 16, no. 1 ( 1 9 6 0 ) : 1 4 5 - 5 8 . Copyright by the
American Geophysical Union.

This approximation c o m e s about b e c a u s e for large Peclet numbers, the s e c o n d


term of Equations 2.21 and 2.23 is much smaller than first term and can be neglected.
Figure 2.9 demonstrates under what conditions this approximation is valid. In Figure
2.9 the dimensionless concentration, C, is plotted as a function of dimensionless time,
/ , for continuous tracer injection using the fixed-step function, Equation 2.22, the
continuous-injection function, Equation 2.24, and the approximate solution, Equation
2.26. Curves are plotted for three Peclet numbers, 1, 10, and 100. Sauty ( 1 9 8 0 ) defined
a Peclet n u m b e r as P = v L/D , where L is the distance from the point of injection of
the solute to the point of measurement and D is the coefficient of hydrod>namic
dispersion. This Peclet n u m b e r defines the rate of transport by advection to the rate of
transport by hydrodynamic dispersion. For Peclet n u m b e r 1, the fixed step function and
the continuous-injection function give quite different results, whereas for Peclet n u m b e r
1 0 0 they are almost identical. T h e approximate solution lies midway between the other
two. This figure suggests that for Peclet numbers less than about 10, the exact solutions
n e e d to be considered, whereas for Peclet numbers greater than 10, the approximate
solution is probably acceptable, especially as the Peclet n u m b e r approaches 100. This
Peclet n u m b e r increases with flow path length as advective transport b e c o m e s m o r e
dominant over dispersive transport. Thus for mass transport near the inlet boundary, it
is important to use the correct equation, but as o n e g o e s away from the inlet boundary,
it is less important that the correct form of the equation is employed.
R

2.8.5

T h i r d - T y p e Boundary Condition

A solution for Equation 2.18 for the following boundary condition was given by van
Genuchten ( 1 9 8 1 ) .

61

M a s s T r a n s p o r t i n Saturated M e d i a

Initial condition

C(x, 0 ) = 0
BC
- D +
ox

vC
x

B o u n d a r y conditions

BC
(finite)

Bx

T h e third condition specifies that as A' a p p r o a c h e s infinity, the concentration gra


dient will still be finite. Under these c o n d i t i o n s the solution to Equation 2 . 1 8 is:

C =

erfc

2V/J

\nDj

4D I
L

1 /
vL
1 +
2 \
D,
x

(2.27)

T h i s equation also reduces to the a p p r o x i m a t e solution, Equation 2 . 2 5 , as the flow


length increases.
2.8.6

O n e - D i m e n s i o n a l S l u g Injection into a F l o w Field

If a slug of contamination is instantaneously injected into a uniform, o n e dimensional


flow field, it will pass through the aquifer as a pulse with a peak concentration, ^ , at
s o m e time after injection, t . T h e solution to Equation 2 . 1 8 under these c o n d i t i o n s
m

(Saury 1 9 8 0 ) is in dimensionless form:

" '

' ^

,)" e x p ( - ^ (1 - W )

( ^

( " 4 ^

" ' '

with
F- = C

R m a

(2.29)

where
2

112

'jtmax = (.1 + P,~ ) P ' ( d i m e n s i o n l e s s time at which peak


concentration o c c u r s )
e

Cr = C/Cn,,
In Figure 2.10, C

(C/C^^) tor a slug injected into a uniform one-dimensional

flow field is plotted against dimensionless time, t , for several Peclet n u m b e r s . It c a n be


R

seen that the time for the peak c o n c e n t r a t i o n ( Q ,

a x

) to o c c u r increases with the Peclet

n u m b e r , up to a limit of / = 1. Breakthrough b e c o m e s m o r e symmetric with increasing

2.8.7

C o n t i n u o u s Injection into a U n i f o r m T w o - D i m e n s i o n a l F l o w Field

If a tracer is continuously injected into a uniform flow field from a single point that fully
penetrates the aquifer, a two-dimensional plume will form that l o o k s similar to Figure

Chapter T w o

FIGURE 2.10

Dimensionless-fype curve for the injection of a slug of a tracer into a one-dimensional

flow field. Source: J. P. Saury, Wafer Resources Research 16, no. 1 (1980): 1 4 5 - 5 8 . Copyright by the
American Geophysical Union.

2 . 1 1 . It will spread along the axis of flow due to longitudinal dispersion and normal to
the axis of flow due to transverse dispersion. This is the type of contamination that
would o c c u r due to leakage of liquids from a landfill or lagoon.
T h e mass transport equation for two-dimensional flow, Equation 2.19, has b e e n
solved for several boundary conditions. T h e well is located at the origin (x = 0, y = 0)
and there is a uniform velocity at a rate v parallel to the x axis. T h e r e is a continuous
injection of a solute of concentration, Q>, at a rate Q at the origin.
B e a r ( 1 9 7 2 ) gives the solution to Equation 2.19 for the condition where the growth
of the plume has stabilizedthat is, as time a p p r o a c h e s infinityas
x

CO,

!1

\2n{D D Y )
L

e x p p ^ ) J ( 3 L f f ! S\)' ]
V

\2D D )
L

l\4D \D
L

D ))
T

( 2

.30)

where
K = the modified B e s s e l function of the s e c o n d kind and zero order
0

(values are tabulated in Appendix B)


Q = rate at which a tracer of concentration Q, is being injected

Ground-water flow

FIGURE 2.11

Plume resulting from the continuous injection of a tracer into a two-dimensional flow field.

Source: C. W. Fetter, Applied Hydrogeology. 2nd ed. (New York: Macmillan Publishing Company, 1 9 8 8 ) .

Mass Transport in Saturated Media

63

T h e two-dimensional growth of a p l u m e from a c o n t i n u o u s s o u r c e can be tracked


through time using a solution to Equation 2.19 developed by Y. Emsellem ( s e e Fried
1 9 7 5 ) . T h e solution has the form
>*y< 0

c(

, _ , ^ m / 2

4n(.D D y
L

e x pF ( ^ \ w ( 0 , B) - WU, B)]

(2.31)

V2D,

where
2

\v x)

(r y)

B--

4D

4D,D

I = time
\\"{t, B] = a function derived by Hantush and tabulated in Appendix C
(In well hydraulics this function is k n o w n as the leaky well
function
2.8.8

W[u,

r/b].)

S l u g Injection into a U n i f o r m T w o - D i m e n s i o n a l F l o w Field

If a slug of contamination is injected over the full thickness of a two-dimensional uniform


flow field in a short period of time, it will move in the direction of flow and spread with
time. T h i s result is illustrated by Figure 2.12 and represents the pattern of contamination
at three increments that result from a one-time spill. Figure 2.12 is b a s e d on the results
o f a laboratory e x p e r i m e n t c o n d u c t e d b y B e a r ( 1 9 6 1 ) . Figure 2.13 s h o w s the spread o f
a plume of chloride that was injected into an aquifer as a part of a large-scale field test
(Mackay et al. 1 9 8 6 ) . T h e p l u m e that resulted from the field test is m o r e c o m p l e x than
the laboratory p l u m e due to the heterogeneities e n c o u n t e r e d in the real world and the
fact the p l u m e may not be following the diffusional model of dispersion.
De J o s s e l i n De J o n g ( 1 9 5 8 ) derived a solution to this p r o b l e m on the basis of a
statistical treatment of lateral and transverse dispersivities. B e a r ( 1 9 6 1 ) later verified it
experimentally. If a tracer with c o n c e n t r a t i o n Q, is injected into a two-dimensional flow
field over an area A at a point (XQ, y ), the c o n c e n t r a t i o n at a point (x, y), at time /after
0

the injection is
n

(x-

(x

v t))

(y-y ) '

(2.32)

y(cm)
Initial point injection
/ C7C = 100%
0

1 7

o.i%_.

0.5%.

-0.5%--..

o
-1

1
5

10
Distance x from
injection point ( c m )

F I G U R E 2.12

Injection of a slug of a tracer into a two-dimensional flow field shown at three time

increments. Experimental results from J. Bear, Journal of Geophysical Research 66, no. 8 (1 9 6 1 ) : 2455
6 7 . Copyright by the American Geophysical Union.

Chapter T w o

FIGURE 2.13

Vertically averaged chloride concentration at 1 day, 8 5 days, 4 6 2 days, and 6 4 7 days

after the injection of a slug into a shallow aquifer. Source-. D. M. Mackay et al. Wafer Resources Research
2 2 , no. 13 (1 9 8 6 ) : 2 0 1 7 - 2 9 . Copyright by the American Geophysical Union.

Mass Transport in Saturated Medio

2.9

65

Effects of Transverse Dispersion


T h e ratio of longitudinal to transverse dispersivity ( a / a ) in an aquifer is an important
L

control over the s h a p e of a contaminant plume in two dimensional mass transport. T h e


lower the ratio, the b r o a d e r the shape of the resulting plume will be. Figure 2.14 s h o w s
various two-dimensional shapes of a contaminant plume, w h e r e the only factor varied
was the ratio of longitudinal to transverse dispersivity. T h i s illustrates the fact that it is
important to have s o m e k n o w l e d g e of the transverse dispersivity in addition to the

FIGURE 2.14

The effect of changing dispersivity ratio on the spread of a contaminant plume from a

continuous source. Source: Robert L. Stoilar.

66

Chapter T w o

longitudinal dispersivity. T h e r e is a paucity of data in the literature on the relationships


of longitudinal to transverse dispersivities. From the few field studies available, a /a is
in the range of 6 to 20 (Anderson 1 9 7 9 , Klotz et al. 1 9 8 0 ) . In addition, dispersivity ratios
based on field studies are b a s e d on fitting the diffusional model of dispersion to cases
where it might not be applicable.
L

2.10

Tests to Determine Dispersivity


2.10.1

Laboratory T e s t s

Diffusion and dispersivity can be determined in the laboratory- using c o l u m n s packed


with the p o r o u s media under investigation. T h e results of column studies are often
reported in terms of p o r e volumes of fluid that is eluted. O n e p o r e volume is the crosssectional area of the c o l u m n times the length times the porosity (ALn). T h e unit discharge
rate from the column is the linear velocity times the porosity times the cross-sectional
area (v nA). T h e total discharge over a period of time is the product of time and the
discharge rate (t\>iAl).
x

T h e total n u m b e r of p o r e volumes, U, is the total discharge divided by the volume


of a single p o r e volume:
I'nAI
"

v.t
R

It can be s e e n that the n u m b e r of p o r e volumes is equivalent to a dimensionless time,


Ir

With this equivalency Equation 2.25, the approximate one-dimensional dispersion


equation, can be rearranged to yield (Brigham 1 9 7 4 )
C_

(2.34)

WHERE

U= the n u m b e r of effluent p o r e volumes, w h e r e a p o r e volume is equal to


the total column volume times the porosity
L = the length of the column
Equation 2.34 can, through appropriate substitution, be m a d e equivalent to Equation
2.26.
T h e concentration of the tracer in the effluent, C is measured for various values
of U, and then C/Q, is plotted as a function of [ ( 7 - l ) / / " ] on linear probability
paper. If the data plot as a straight line, they are normally distributed, the diffusive form
of the advection-dispersion equation is valid, and the slope of the line Is related to the
longitudinal hydrodynamic dispersion.
T h e value of D, can be found from
2

a.s.-yo.ta)

(2-35)

Mass Transport in Saturated Media

67

where
2

/o.84 =
/o.i =

D / f / " ] w h e n C / Q i s 0.84
1 2

K.U- D/U ' ] w h e n C/Q, is 0 . 1 6

S i n c e > = a v + D*, then


t

T h e average linear velocity in the c o l u m n can be found from the quantity of water
discharging per unit time divided by the product of the cross-sectional area and the
porosity. T h e effective diffusion coefficient can either be m e a s u r e d in a c o l u m n test or
estimated.
EXAMPLE

Pickens and Grisak ( 1 9 8 1 ) c o n d u c t e d a laboratory study o f dispersion in sand c o l u m n s

PROBLEM

with the following characteristics:


Chloride

Tracer
C o l u m n length

30 cm

C o l u m n diameter

4.45

cm

Mean grain size

0.20

mm

Uniformity coefficient of sand

2.3
0.36

Porosity
F l o w rate
Test Rl

5.12 x 1 0 "

T e s t R2

1.40 x 1 0 "

mL/sec

T e s t R3

7.75 x 1 0 "

mL/sec

Test Rl

9.26 x 1 0 "

T e s t R2

2.53 x 1 0 "

cm/sec

T e s t R3

8.60 x 1 0 "

cm/sec

11)1 SCI

Average Linear Velocity


cm/sec

T e s t Rl was run using chloride at 2 0 0 m g / L , followed by test R2, in which the


saline solution was flushed out of the c o l u m n using deionized water, and then test R3,
w h e r e the 2 0 0 - m g / L chloride solution was again introduced into the column.
T h e results of the three tests are plotted in Figure 2.15. T h e results of test R2 have
a reverse s l o p e as deionized water replaced the saline solution. It can be s e e n that the
results form a straight line.
F o r chloride in water at 25C, the molecular diffusion coefficient is 2.03 x
5

1 0 " c m / s e c . B a s e d o n this, Pickens and Grisak estimated the effective diffusion c o e f


5

ficient to be 1.02 x 1 0 " c m / s e c . T h e hydrodynamic dispersion coefficients are b a s e d


on the s l o p e of the straight lines. T h e following values were o b t a i n e d for the three tests:
Test

Hydrodynamic Dispersion

Rl

4.05 x 1 0 " ' c m / s e c

R2

8.65 x 1 0 " c m / s e c

R3

3.76 x 1 0 ~ c m / s e c

Dispersivity

0.033 cm

0.030 cm

0.043 cm

T h e replicate values are not equal b e c a u s e of experimental error.

68

Chapter T w o

-0.04

FIGURE 2.15

Plot of (U 1 ) / U

W 2

versus C / C on probability paper for determination of dispersion in


0

a laboratory sand column. Source: S. F. Pickens and G. E. Grisak, Water Resources Research 17, no. 4
(1 9 8 1 ) : 1 191 121 1. Copyright by the American Geophysical Union.

T h e computation of dispersivity for test Rl is illustrated here:


Dr. ~ D'

x
s

( 4 . 0 5 x 1 0 " c m 7 s e c - 1.02 x 1 0 " c m V s e c )

= 0.033 cm

9.26 x 1 0 ~ c m / s e c

2.10.2

Field T e s t s for Dispersivity

Dispersivity can he determined in the field bv two means II there is a contaminated


aquifer, the plume of known contamination can be mapped and the advection dispersion
equation solved with dispersivity as the unknown. Pinder ( 1 9 7 3 ) used this approach in
a groundwater modeling study of a plume of dissolved chromium in a sand and gravel
aquifer on k i n g Island, New York. He started with initial guesses of a and a and then
varied them during successive model runs until the c o m p u t e r model yielded a reasonable
reproduction of the o b s e r v e d contaminant plume. O n e of the difficulties of this approach
is that the concentration and volume of the contaminant s o u r c e are often not known.
L

Mass Transport in Saturated Media

69

A m u c h m o r e c o m m o n a p p r o a c h is the use of a tracer that is injected into the


g r o u n d via a well. T h e r e are a variety of variations to this approach. Natural gradient
tests involve the injection of a tracer into an aquifer, followed by the m e a s u r e m e n t of
the p l u m e that d e v e l o p e d under the prevailing water table gradient (e.g., Sudicky and
Cherry 1 9 7 9 ; Gillham et al. 1 9 8 4 ; Mackay et al. 1 9 8 6 ; LeBlanc et al. 1 9 9 1 ; Garabedian et
al. 1991 ) T h e plume is m e a s u r e d by m e a n s of small a m o u n t s of water withdrawn from
down-gradient observation wells and multilevel p i e z o m e t e r s . O n e - and two-well tests
have also b e e n used in which a tracer is p u m p e d into the g r o u n d and then groundwater
containing the tracer is p u m p e d b a c k out of the g r o u n d (e.g., Fried 1 9 7 5 ; G r o v e and
B e e t e m 1 9 7 1 ; Saury 1 9 7 8 ; P i c k e n s et al. 1 9 8 1 ; P i c k e n s and Grisak, 1 9 8 1 ) .

2.10.3

Single-Well Tracer T e s t

A single-well tracer test involves the injection of water containing a conservative tracer
into an aquifer via an injection well and then the s u b s e q u e n t pumping of that well to
recover the injected fluid. T h e Huid velocities of the water b e i n g p u m p e d and injected
are m u c h greater than the natural ground-water gradients.
Equation 2 . 2 0 can b e written ( H o o p e s and Harleman 1 9 6 7 ) a s

C
^ 7
tit

dC
" l ~
tir

SC
t-"TT
or

D* d
T
r or

(2.37)

Gelhar and Collins ( 1 9 7 1 ) derived a solution to Equation 2.37 for the withdrawal
p h a s e of an injection-withdrawal well test in which the diffusion term is neglected b e c a u s e
it is very m u c h smaller than the dispersion term. T h e relative c o n c e n t r a t i o n of the water
b e i n g withdrawn from the injection well is

= - erfc
Co
2

(U p

{lf(* /R,)L2
L

- (1 - {/,//,)]

- 1
[1 1/2

(2.38)

where
cumulative volume of water withdrawn during various times
u,

Re >

total v o l u m e of water injected during the injection p h a s e


average frontal position of the injected water at the e n d of the injection
period, which is defined by

Q<
nbn
where
Q = rate of injection
t = total time of injection
b = aquifer thickness
n = porosity

(2.39)

F I G U R E 2.16

Comparison of measured C / C values for a single-well injection-withdrawal test versus


0

an analytical solution. Source: S. F. Pickens and G. E. Grisak, Water Resources Research 17, no. 4
(1 9 8 1 ) : 1 1 9 1 - 1 2 1 1 . Copyright by the American Geophysical Union.

Mass Transport in Saturated Media

EXAMPLE
PROBLEM

71

Pickens and Grisak ( 1 9 8 1 ) performed a single-well injection-withdrawal tracer test into


a confined sand aquifer a b o u t 8.2 m thick with an average hydraulic conductivity o f
1 . 4 x 1 0 " c m / s e c and a porosity of 0 . 3 8 . T h e sediment tested in the c o l u m n study
d e s c r i b e d in the previous e x a m p l e p r o b l e m c a m e from this aquifer.
2

T h e injection well was 5.7 cm in diameter and the full thickness of the aquifer was
s c r e e n e d . Clear water was injected at a constant rate for 24 hr prior to the stan of the
test to establish steady-state conditions. T h e tracer used during the tests was
I, a
radioactive iodine, which was added to the injected water. All m e a s u r e m e n t s w e r e corrected for the radioactive decay that o c c u r r e d during the test.
1 3 1

T w o tests w e r e performed on the well. T h e first test, S W 1 , had an injection rate


of 0 . 8 8 6 L / s e c and injection c o n t i n u e d for 1.25 da. A total volume of 95.6 m of water
was injected, and the injection front reached an average radial distance away from the
well of 3 1 3 m. W a t e r was then p u m p e d for 2.0 da at the s a m e rate, so that a total of
1 5 3 m of water was withdrawn. T h e s e c o n d test, S W 2 , was longer. Water with the tracer
was added at a rate of 0 . 7 1 9 L / s e c for 3-93 da. A total of 2 4 4 m of water was added,
and the average position of the injection front reached to 4 . 9 9 m from the well. During
the withdrawal phase a total of 8 8 6 m of water was p u m p e d over a period of 16.9 da
at an average rate of 0 . 6 0 6 L / s e c .
3

T h e results of the test are shown in Figure 2 . 1 6 . Relative concentration, C/CQ, is


plotted against UI Uj. T h e dots represent field values and the solid lines are curves, which
w e r e c o m p u t e d using Equation 2.38. Various curves were c o m p u t e d for different values
of a , and the curves with the b e s t fit to the field data w e r e plotted on the graphs. In
Figure 2 . 1 6 ( a ) the calculated curve was b a s e d on a longitudinal dispersivity of 3.0 cm,
whereas for curve 2 . 1 6 ( b ) the best-fit curve was b a s e d on a longitudinal dispersivity of
9.0 c m . This test illustrates the scale d e p e n d e n t nature of dispersivity. T h e s e c o n d test,
in which a larger volume of water was injected, tested a larger volume of the aquifer
than the first test and yielded a higher dispersivity value.
L

2.11

Scale Effect of Dispersion


T h e two e x a m p l e p r o b l e m s derived from Pickens and Grisak ( 1 9 8 1 ) illustrate what has
b e e n called the s c a l e e f f e c t o f d i s p e r s i o n (Fried 1 9 7 5 ) . A t the laboratory scale the
m e a n value of a was determined to be 0 . 0 3 5 cm w h e n the flow length was 30 cm. With
the single well injection withdrawal test, a was 3 cm w h e n the solute front traveled
3.1 m and 9 cm w h e n the solute front traveled 5.0 m. In a two-well recirculating withdrawal-injection tracer test with wells located 8 m apart, x was determined to be 50 cm.
All these values were o b t a i n e d from the s a m e site. T h e greater the flow length, the larger
the value of longitudinal dispersivity n e e d e d to fit the data to the advection-dispersion
equation.
L

Lallemand-Barres and P e a u d e c e r f ( 1 9 7 8 ) published a graph on which dispersivity,


as measured in the field, was plotted against flow length on log log paper (Figure 2 . 1 7 ) .
This graph suggested that the longitudinal dispersivity could be estimated to be about

Chapter T w o

FIGURE 2.17

Field-measured values of longitudinal dispersivity as a function of the scale of measurement.

Source-. P. Lallemand-Barres and P. Peaudecerf, Bulletin, Bureau de Recherches Gologiques et Minires,


Sec 3 / 4 ( 1 9 7 8 ) : 2 7 7 - 8 4 . Editions BRGM B P 6 0 0 9 4 5 0 6 0 O R L E A N S CEDEX 2.

Mass Transport in Saturated Media

73

10,000

0.01

o
100

00,000

1000
Scale (m)

F I G U R E 2.1 8

Field-measured values of longitudinal dispersivity as a function of the scale of measurement.

The largest circles represent the most reliable data. Source: t. W. Gelhar, Water Resources Research 2 2 ,
no. 9 (1 9 8 6 ) : 1 3 5 S - 1 4 5 S . Copyright by the American Geophysical Union.

0.1 of the flow length. Gelhar ( 1 9 8 6 ) published a similar graph (Figure 2 . 1 8 ) , which
contained m o r e data points and was e x t e n d e d to flow lengths m o r e than an o r d e r of
magnitude greater than the Lallemand-Barres and P e a u d e c e r f figure. T h e additional data
on the Gelhar graph suggest that the relationship b e t w e e n a and flow length is m o r e
L

c o m p l e x than a simple 1 to 10 ratio.


T h e longitudinal dispersivity that o c c u r s at field-scale flow lengths can be called
m a c r o d i s p e r s i o n . In a flow domain that e n c o m p a s s e s a few p o r e lengths, mechanical
dispersion is caused by differences in the fluid velocities within a pore, b e t w e e n p o r e s
of slightly different size, and b e c a u s e different flow paths have slightly different lengths.
However, at the field scale, even aquifers that are c o n s i d e r e d to be h o m o g e n e o u s will
have layers and z o n e s of somewhat different hydraulic conductivity. If mechanical dis
persion can be caused by slight differences in the fluid velocity within a single pore,

Chapter T w o

imagine tin 11 i n li.inu ,il dispcrsii in th.it will result . i s t h e HUH I passes thri itigh regii i n s i ,1
the aquifer with different conductivity values and corresponding different velocities.
Hydraulic conductivity is frequently determined on the basis of a pumping test,
w h e r e water is removed from a large volume of the aquifer. As a result, the hydraulic
conductivity that is obtained is an average value over the entire region of the aquifer
contributing water to the well. This averaging will c o n c e a l real differences in hydraulic
conductivity across the aquifer. T h e s e differences exist in both vertical and longitudinal
sections.
Figure 2 . 1 9 ( a ) shows the variation of laboratory-determined values of intrinsic
permeability, e x p r e s s e d as the logarithm of the value in millidarcies, versus depth for

FIGURE 2.19

Permeability in millidarcies and porosity data from laboratory tests of cores from the Mt.

Simon aquifer in Illinois. Source: t. W. Gelhar, Water Resources Research. 22, no. 9 (1986): 1 3 5 S - 1 4 5 S .
Copyright by the American Geophysical Union.

75

M a s s T r a n s p o r t i n Saturated M e d i a

Hydraulic conductivity (crn/sec)


FIGURE 2.20

Hydraulic conductivity (cm/sec)

Hydraulic conductivity as determined by permeameter tests of remolded sediment samples

from a glacial drift aquifer. The borings from which the cores were obtained are separated by one meter
horizontally. Source: E. A. Sudicky, Wafer Resources Research 2 2 , no. 13 (1 9 8 6 ) : 2 0 6 9 - 8 2 . Copyright by
the American Geophysical Union.

rock c o r e samples from a well in the Mt. Simon aquifer in Illinois ( G e l h a r 1 9 8 6 , b a s e d


o n B a k r 1 9 7 6 ) . Figure 2 . 1 9 ( b ) s h o w s the variation o f porosity with depth for the s a m e
c o r e samples. Figure 2.20 s h o w s profiles of the vertical variation in hydraulic conductivity
b a s e d o n p e r m e a m e t e r tests o f r e p a c k e d c o r e samples o f sediment, from two borings
located 1 m apart (Sudicky 1 9 8 6 ) . Figure 2.21 s h o w s the distribution of the log of hydraulic
conductivity of a c r o s s section in a stratified sandy outwash aquifer with layers of primarily
medium-grained, fine-grained, and silly, fine-grained sand. T h e c r o s s s e c t i o n is 1.75 m
d e e p by 19 m long (Sudicky 1 9 8 6 ) .
Figures 2.19 through 2.21 illustrate the natural variation of b o t h hydraulic conductivity and porosity. Even aquifers that are usually c o n s i d e r e d to be h o m o g e n e o u s
still have variations in porosity and hydraulic conductivity. Hydraulic conductivity of
g e o l o g i c materials varies over a very wide range of values, up to nine orders of magnitude.
Porosity varies over a much, m u c h smaller range: approximately from 1 to 6 0 % or less
than two orders of magnitude. From the standpoint of describing aquifers mathematically,
it is s o m e t i m e s useful to a s s u m e that hydraulic conductivity follows a lognormal distribution, which m e a n s that the logarithms of the conductivity values are normally distributed, whereas porosity is normally distributed ( F r e e z e 1 9 7 5 ) . S i n c e dispersion d e p e n d s
upon variations in the fluid velocity and from Darcy's law [ v = {K/ n)( dhl dl)\, it is obvious
that variations in b o t h hydraulic conductivity and porosity play a role. However, s i n c e
hydraulic conductivity varies over a much larger range, it is the m o r e i m p o r t a n t

FIGURE 2.21

8
9
10
11
Longitudinal distance (m)

Distribution of the hydraulic conductivity along a cross section through a glacial drift

aquifer. Hydraulic conductivity is expressed as a negative log value. (If K = 5 x 1 0

cm/sec, then -log

K is 1.3.) Sample locations are every 5 cm vertically and every 1 m horizontally. Hydraulic conductivity
was less than 10

cm/sec in the stippled zones. Source: E. A. Sudicky, Water Resources Research 22,

no. 13 ( 1 9 8 6 } : 2 0 6 9 - 8 2 . Copyright by the American Geophysical Union.

12

13

14

15

16

17

18

19

M a s s T r a n s p o r t i n Saturated M e d i a

77

This leads us to an explanation for the scale factor. As the flow path gets longer,
ground water will have an opportunity to e n c o u n t e r greater and greater variations in
hydraulic conductivity and porosity. Even if the average linear velocity remains the same,
the deviations from the average will increase, and h e n c e the mechanical dispersion will
also increase. It is logical that the flow path will eventually b e c o m e long e n o u g h that all
p o s s i b l e variations in hydraulic conductivity will have b e e n e n c o u n t e r e d and that the
value of mechanical dispersion will reach a maximum. If o n e assumes that the distribution
of hydraulic conductivity has s o m e definable distribution, such as normal or lognormal,
and that transverse dispersion is occurring, it c a n be s h o w n that apparent macrodispersivity will a p p r o a c h an asymptotic limit at long travel distances and large travel times
( M a t h e r o n and de Marsily 1 9 8 0 ; Molz, Guven, and Melville 1 9 8 3 ; Gelhar and A x n e s s
1 9 8 3 ; Dagan 1 9 8 8 ) . W h e n the asymptotic limit is reached, the p l u m e will c o n t i n u e to
spread. In this region the variance of the p l u m e will g r o w proportionally to the time or
m e a n travel distance, as it d o e s at the laboratory c o l u m n scale. T h e advective-dispersion
m o d e l is b a s e d on the assumption that dispersion follows Fick's law. S o m e authors
c o n t e n d that dispersion follows F i c k ' s law only at the laboratory scale, w h e r e it is c a u s e d
by local mechanical dispersion, and for very long flow paths, w h e r e the effects of advection
through h e t e r o g e n e o u s materials and local transverse dispersion create m a c r o s c a l e dispersion that follows F i c k ' s law (e.g., G e l h a r 1 9 8 6 ; Dagan 1 9 8 8 ) . T h e c o n t e n t i o n that
m a c r o s c a l e dispersion b e c o m e s Fickian (i.e., follows F i c k ' s law) at long travel times and
distances is s o m e w h a t controversial, especially if the flow is through geological formations
that are h e t e r o g e n e o u s at different scales (Anderson 1 9 9 0 ) .

2.12

Stochastic Models of Solute T r a n s p o r t


2.12.1

Introduction

T h e normal m a n n e r of determining a field-scale dispersion coefficient is to l o o k for a


natural tracer or inject a tracer into an aquifer and o b s e r v e the resulting development
of a plume. A solute-transport m o d e l is then c o n s t r u c t e d and the c o m p u t e d solute
distribution is fitted to the o b s e r v e d field data by adjusting the dispersion coefficients.
Dispersion coefficients o b t a i n e d in this m a n n e r are fitted curve parameters and do n o t
represent an intrinsic property of the aquifer. T h i s is especially true w h e n the aquifer is
a s s u m e d to be h o m o g e n e o u s and is d e s c r i b e d by a single value for hydraulic conductivity
and porosity. It is apparent that flow and transport modeling b a s e d on a single value
for porosity and hydraulic conductivity is a g r o s s simplification of the c o m p l e x i t y of
nature. F o r analytical solutions, we are constrained to u s e of a single value for average
linear velocity, and for numerical m o d e l s we often use a single value b e c a u s e that is all
we have.
A d e t e r m i n i s t i c m o d e l is o n e w h e r e a partial differential equation is solved,
either numerically or analytically, for a given set of input values, aquifer parameters, and
boundary' conditions. T h e resulting output variable has a specific value at a given place
in the aquifer. It is assumed that the distribution of aquifer parameters is known. T h e
equations given earlier in this c h a p t e r are e x a m p l e s of deterministic models.

Chapter T w o

A s t o c h a s t i c m o d e l is a model in which there is a statistical uncertainty in the


value of the output variables, such as solute distribution. T h e probabilistic nature of this
o u t c o m e is due to the fact that there is uncertainty in the value and distribution of the
underlying aquifer parameters, such as the distribution and value of hydraulic conductivity
and porosity ( F r e e z e 1 9 7 5 ; Dagan 1 9 8 8 ) .
T h e idea behind stochastic modeling is very attractive. It is obvious that it takes a
great effort to determine hydraulic conductivity and porosity at m o r e than a few locations
in an aquifer system. If we could determine the distribution of aquifer properties with
a high d e g r e e of detail, then a numerical solution of a deterministic model would yield
results with a high degree of reliability. However, with limited knowledge of aquifer
parameters, a deterministic model makes only a prediction of the value of an output
variable at a given point and time in the aquifer. T h e stochastic model is based on a
probabilistic distribution of aquifer parameters. At the outset it is recognized in the
stochastic m o d e l that the result will be only s o m e range of possible o u t c o m e s . It cannot
tell us what the concentration of a solute will be at a particular point in the aquifer at
a given time. T h e stochastic model thus recognizes the probabilistic nature of the answer,
whereas the deterministic model suggests that there is only o n e " c o r r e c t " answer. Of
course, the e x p e r i e n c e d hydrogeologist recognizes the uncertainty even in the deterministic answer. T h e r e have b e e n literally hundreds of papers written since 1 9 7 5 on
various aspects of stochastic modeling of ground-water flow and solute transport (e.g.,
F r e e z e 1 9 7 5 ; Gelhar, Gutjahr, and Naff 1 9 7 9 ; Gelhar and Axness 1 9 8 3 ; Gelhar 1986;
Dagan 1 9 8 2 , 1 9 8 4 , 1 9 8 6 , 1 9 8 8 ; Neuman, Winter, and Newman 1 9 8 7 ) . T h e stochastic
transport m o d e l s predict the m o v e m e n t of the c e n t e r of mass of the solute plume
( S p o s i t o , Jury, and Gupta 1 9 8 6 ) and the average m o m e n t of inertia or s e c o n d spatial
m o m e n t of the solute with respect to its center of mass (Dagan 1 9 8 8 ) .
Stochastic m o d e l s have reached the stage of development where their accuracy
has b e e n tested by c o m p a r i s o n of model-predicted results with the movement of a tracer
in field tests ( S p o s i t o and Barry 1987; Barry, Coves and Sposito 1 9 8 8 ) .
2.12.2

Stochastic Descriptions of Heterogeneity

T h e greatest uncertainty in the input parameters of a model is the value of hydraulic


conductivity, b e c a u s e it varies over such a wide range for geologic materials, if we make
a measurement of hydraulic conductivity at a given location, the only uncertainty in its
value at that location is d u e to errors in measuring its value. However, at all locations
w h e r e hydraulic conductivity is not measured, additional uncertainty exists. If we make
a n u m b e r of m e a s u r e m e n t s of the value of hydraulic conductivity, we can estimate this
uncertainty using certain statistical techniques.
Let us define Y as the log of the hydraulic conductivity, K, and assume that the
log value Y is normally distributed. We will assume a o n e dimensional series of Y values
{Yu ^ 2 . Y , K
3

Y] ( F r e e z e et al. 1 9 9 0 ) . Therefore,
>i = log K

(2.40)

T h e population that consists of all of the values of Y has a mean value, n , and a
Y

standard deviation, oy T h e only way to obtain precise values of /ty and a would be to
Y

sample the aquifer everywhere, clearly an impossible task, but we can find estimates of

Mass Transport in Saturated Media

V,

y,

79

(a)

Hy= y

oi
0

Lag (k) or separation (H)


(c)
FIGURE 2.22

(a) One-dimensional sequence of log hydraulic conductivity values, Y; (b) probability

distribution function for Y; (c) autocorrelation function for Y. Source: R. A. Freeze et al., Ground Water
2 8 , no. 5 ( 1 9 9 0 ) : 7 3 8 - 6 6 . Used with permission. Copyright 1 9 9 0 Water W e l l Journal Publishing Co.

their values based on the locations w h e r e we have actually m e a s u r e d K. If we have a


series of Y values {Y

lt

Y , Y ,. . . , Y}, as in Figure 2 . 2 2 ( a ) , b a s e d on m e a s u r e d values


2

of K, then o u r estimate of the population mean is o b t a i n e d from the m e a n value of the


sampled values, Y, which can be found from
(2.41)

Chapter T w o

T h e estimate of the variance of the population is also obtained by the variance of the
2

sampled values, Sy , which is found from the following equation:


1 "

I (.Y,-Y)(Y,-Y)

V= -

_
(2.42)

N 1=I

F o r a normally distributed population, the probabilistic value is called a probability


density function ( P D F ) and is d e s c r i b e d by the mean and the variance. T h e variance is
a measure of the degree of heterogeneity of the aquifer. T h e greater the value, the m o r e
h e t e r o g e n e o u s the aquifer. T h e PDF can be represented as a bell-shaped curve with the
peak equal to the mean, as in Figure 2 . 2 2 ( b ) , and the spread of the bell can be defined
by either the variance or the standard deviation, S , which is the square root of the
Y

variance.
If we have measured the value of Y at a n u m b e r of locations and wish to estimate
t

the value Yj at s o m e o t h e r location j that is not c l o s e to any of the measured values,


h o w can we estimate the value of K? O n e approach is to say that the most likely estimate
of Yj is the mean of the measured values of Y , and the uncertainty in this value is
t

normally distributed with a standard deviation equal to the standard deviation of the
measured values, S . In doing so we have a c c e p t e d the e r g o d i c h y p o t h e s i s . This
t

m e a n s that there is a 1 6 % c h a n c e that the value of Yj is greater than Y + Sy, a 5 0 %


c h a n c e that it is greater than Y, and an 8 4 % c h a n c e that it is greater than Y S .
Y

Hydraulic conductivity values measured at locations c l o s e to each other are likely


to be somewhat similar. T h e farther apart the measurements, the less likely that the
values will be similar. This is due to the fact that as distances b e c o m e greater, the c h a n c e
that there will be a c h a n g e in g e o l o g i c formation increases. T h e function that describes
this is the a u t o c o r r e l a t i o n f u n c t i o n , p . T h e value of the autocorrelation function
Y

decreases with the distance b e t w e e n two measurements. An estimate of the autocorrelation function, r , can be obtained from the measured sample values by the following
r

equation:

* = T2- t
Oy

(.Yt-YXY^-Y)

(2.43)

= 1

with k, the l a g , being a whole n u m b e r representing a position in the s e q u e n c e away


from the i position. Figure 2 . 2 2 ( c ) s h o w s an autocorrelation function plotted against
2

the lag. If the lag is zero, then Equation 2.43 reduces to r = S /S


Y

= 1. This means

that a Y value is perfecdy correlated with itself.


T h e autocorrelation factor can be e x p r e s s e d in terms of either lag, p , or distance,
Yk

p (H).
Y

W h e n a m e a s u r e m e n t o f Y is made at position
{

is m a d e at position

the absolute value of X -

and a measurement o f Y^

is called the s e p a r a t i o n ,

H.
If the autocorrelation function has an exponential form, then it can be e x p r e s s e d
as
p (.H)
Y

exp[-\H\/A ]
Y

(2.44)

Mass Transport in Saturated Media

81

w h e r e l , the c o r r e l a t i o n l e n g t h , i s representative o f the length over which Y i s


Y

correlated. It is the distance over which p (H) decays to a value of e~ . T h e i n t e g r a l


Y

s c a l e , ty. is the area under the curve.


(2.45)
Integration o f Equation 2 . 4 5 will s h o w that c = l / l , s o that the correlation
y

structure can be d e s c r i b e d by either the correlation length or the integral scale.


The a u t o c o v a r i a n c e , x

or x (H), is equal to the autocorrelation times the

Yk

variance.
x {H)
Y

<J p (H)
Y

(2-46)

W e can d e s c r i b e the distribution o f heterogeneity o f Y b y the u s e o f three stochastic


functions, p , r r ( o r r j
Y

2
y

) , and X . I f a stochastic p r o c e s s i s said t o b e stationary, the


Y

values o f p , <7y(or o " ) , and X d o not vary in s p a c e in the region b e i n g studied. I f the
2

hydraulic conductivity of an aquifer can be d e s c r i b e d as a stationary stochastic p r o c e s s ,


the aquifer is uniformly h e t e r o g e n e o u s .
2.12.3

Stochastic Approach to Solute T r a n s p o r t

If we a c c e p t the idea that we don't k n o w the value of the hydraulic conductivity and the
porosity everywhere, then we must a c c e p t the idea that it is not p o s s i b l e to predict the
actual c o n c e n t r a t i o n of a solute that has u n d e r g o n e transport through an aquifer. T h e
best estimate o f the c o n c e n t r a t i o n i s the e n s e m b l e m e a n c o n c e n t r a t i o n , < C ) , o r
the m e a n of all the m e a n s of an e n s e m b l e of all p o s s i b l e random but equivalent populations, and the associated variance. T h e m o v e m e n t of a solute b o d y may be d e s c r i b e d
b y the m o t i o n o f the c e n t e r o f mass o f the b o d y and the s e c o n d - o r d e r spatial m o m e n t ,
o r the m o m e n t o f inertia (Dagan 1 9 8 8 ) . However, it is important to n o t e that the p r o c e s s
of advective transport d o m i n a t e s macrodispersion. T h i s m e a n s that w h e t h e r o n e uses a
deterministic m o d e l or a stochastic m o d e l , the large picture of solute transport will
emerge, since b o t h a c c o u n t primarily for advective transport, with the dispersion factor
tending to s m e a r the leading e d g e of the plume.
Dagan ( 1 9 8 7 , 1 9 8 8 ) has derived a linear m o d e l o f stochastic transport. T h i s m o d e l
unites the w o r k o f Dagan ( 1 9 8 2 , 1 9 8 4 ) , Gelhar and Axness ( 1 9 8 3 ) and Neuman, Winter,
and Newman ( 1 9 8 7 ) . In o r d e r to d o s o , h e neglected all nonlinear terms, such as t h o s e
arising from the deviation of solute particles from their mean trajectory. Neuman and
Zhang ( 1 9 9 0 ) and Zhang and Neuman ( 1 9 9 0 ) have derived a quasilinear stochastic m o d e l
that is m o r e general than Dagan's linear model, which is applicable only for solute
transport domains with a large Peclet n u m b e r t h a t is, t h o s e with a long flow distance.
However, as solute transport over long flow distances represents a practical p r o b l e m
and Dagan ( 1 9 8 8 ) presents s o m e useful closed-form analytical solutions, we will e x a m i n e
his results.
T h e c a s e that is presented is for the m o v e m e n t of a slug of contamination such
that at t = 0 a solute of total mass M is injected into the aquifer at c o n c e n t r a t i o n Cq.
We wish to k n o w the concentration C(x, 0 at s o m e future time, /, and distance from
the point of injection, x

Chapter T w o

The velocity vector, V, can be determined from Darcy's law:


v /

V=

>

(2-47)

L e t V = U + u . w h e r e U = <V>, the e n s e m b l e average for V, and u is the fluctuation


in V. U is assumed to be constant and its covariance u = < / x ) , ( x ' ) ) is a function
of the separation vector H = x x ' , w h e r e /, / = 1, 2, 3 and x ( j c , , j r , * ) is the
coordinate vector.
T h e position of the center of mass < X > is defined as
jt

< X > = f-M J x <C>

rfx

(2.48)

T h e total mass M is defined as


M=(nCdx

(2.49)

T h e average m o m e n t of inertia or s e c o n d spatial m o m e n t of the solute b o d y is

XjM = (jjj / ( * , - - <Xj( , - < X , < C >


X

rfx

(2.50)

F o r a conservative solute the position of the center of mass of the solute slug can
be obtained from the advective equation.
<X>=Uf+X

(2.51)

where
< X > = the e n s e m b l e mean average position of the center of mass
U = <V> the e n s e m b l e average of ground-water velocity
X = the starting position o f the contamination
0

I = time
T h e value of U can be estimated from Darcy's law:
V= -(~^j(db/dl)

(2.52)

w h e r e K is the geometric mean of the hydraulic conductivity values measured in the


area over which U is being estimated. T h e geometric mean of K is found by taking
the natural log of each value, finding the mean of the natural logs, and then finding the
exponential of the m e a n of the natural logs.
0

T h e total particle displacement, X is defined as


X, = VI + X + X
with

dX'/dt= u ( X , ) and

dX /dl= u , w h e r e u is the
d

(2.53)

fluctuation

in the convective

velocity and u is velocity of a Browian motion of zero mean. X' is the residual of the
d

Mass Transport in Saturated Media

83

displacement of a particle defined as dX'/dt= u ( X , ) , which is mechanical dispersion


plus h e t e r o g e n e o u s advection.
Dagan ( 1 9 8 7 , 1 9 8 8 ) s h o w e d that
2

d Xj,(t)

dt

(2n)

J Y < / ) * r ' ) exp(/{' X')>

exp?

Vt -

Dt^&i) dl d

0 ; 1 = 1,2, 3 )

(2.54)

w h e r e and {' are F o u r i e r transform wave v e c t o r n u m b e r s , * is the c o m p l e x conjugate


of | , and i stands for the imaginary unit. T h i s equation has b e e n solved for b o t h
longitudinal and transverse dispersion. T h e longitudinal spread of the solute b o d y is
characterized by X^it)
Dimensionless
2

22

= X /e ,
22

and the transverse spread,

forms

of

f = tU/ , w h e r e e
ll

X (/),

X (t),

22

by X (t).
22

and

are

given

by

X^' = X / ,
11

is the integral scale for the log conductivity for

horizontal flow.
Solutions of 2.54 have b e e n o b t a i n e d for conditions of average flow parallel to the
plane of anisotropy and for large Peclet n u m b e r s . C l o s e d forms of these equations are
available for two conditions: ( 1 ) isotropic conditions with three-dimensional heterogeneity and ( 2 ) anisotropic conditions w h e r e the log conductivity correlation scale for
vertical flow is m u c h less than the log conductivity correlation scale for horizontal flow.
T h e anisotrophy ratio, je , is indicated by fi.
h

Longitudinal dispersion with isotropic conditions:

Longitudinal dispersion with two-dimensional anisotropic conditions:

(2.56)

Transverse dispersion with three-dimensional isotropic heterogeneity:


1 1

3-7

/ 4

73-^73+

A ,.'

(2-57)

Transverse dispersion with two-dimensional isotropic heterogeneity.

1 - ( 1 + Oil-''

(2.58)

In Equations 2 . 5 6 and 2 . 5 8 , E is the Euler n u m b e r ( 0 . 5 7 7 . . .) and Ei is the


exponential integral, which is tabulated in Appendix D. Figure 2.23 s h o w s the relationship

Chapter T w o

10

15

20

25

t> = tUI

FIGURE 2.23

Dependence of longitudinal spatial moment, X , on travel time, t. Curve for il = 1 is for


n

isotropic three-dimensional heterogeneity and curve for Q = infinity is for two-dimensional transport in the
horizontal plane for isotropic heterogeneity. Source: G. Dagon, Wafer Resources Research 24, no. 9
(1988): 1 4 9 1 - 1 5 0 0 . Copyright by the American Geophysical Union.

of the dimensionless longitudinal spatial moment, X,


to dimensionless time, f. Curves
are given for various anisotropy ratios, $2, w h e r e Q = eje . T h e curve for Q = 1.0 is the
solution to Equation 2.55, whereas the curve for 7 = oo is the solution to Equation 2.56.
Dagan ( 1 9 8 8 ) defined an apparent longitudinal macrodispersivity coefficient as "the
value of the constant dispersivity that would lead to the solution of the convection
dispersion equation for the s a m e X , as the actual, time-dependent o n e . " He indicated
that this was equal to X, J2t. Figure 2.24 shows the relationship of dimensionless apparent
longitudinal macrodispersivity to dimensionless time for various anisotropy ratios. It can
be seen that this parameter has limited d e p e n d e n c e on the anisotropy ratio. T h e apparent
macrodispersivity a p p r o a c h e s an asymptotic value with increasing travel time, which
c o r r e s p o n d s to increasing travel distance. This figure demonstrates that at all times
longitudinal macrodispersion is dominated by advective transport. T h e asymptotic value
h

for D is a Uand is independent of CI.


L

Since e = \/X , the asymptotic value o f D is also equal to (a U)/X and the
asymptotic macrodispersivity, <x , is equal to a IX .
Figure 2.25 shows the d e p e n d e n c e of the horizontal, transverse spatial moment,
X , on time and Figure 2.26 shows the s a m e thing for the vertical, transverse spatial
moment, X . It can be seen that the effect of the anisotropy ratio is significant for both
these moments.
t

22

i}

Mass Transport in Saturated Media

1.0 r

FIGURE 2.24

85

n = 0.2

Dependence of the apparent longitudinal macrodispersivity upon travel time. Intermediate

curves are for Q = 0.4, 0.6, and 0.8. Source: G. Dagan, Water Resources Research 2 4 , no. 9 (1 9 8 8 ) : 1 4 9 1 1 5 0 0 . Copyright by the American Geophysical Union.

2.13

Fractal Geometry Approach to Field-scale Dispersion


2.13.1

Introduction

A recently developed field of mathematics is fractal g e o m e t r y ( M a n d e l b r o t 1 9 8 3 ) . Fractal


geometry is a way of looking at irregular o b j e c t s , such as coastlines or aquifers. O n e of
the p r e c e p t s of fractal g e o m e t r y is that irregular o b j e c t s in nature tend to have patterns
that repeat themselves at different scales, a p h e n o m e n o n k n o w n as self-similarity. F o r
e x a m p l e , in a sedimentary aquifer the relationship of individual p o r e s to e a c h o t h e r may
be similar to the relationship of laminae to e a c h other, which may be similar to the
relationship of b e d s to e a c h other, which may be similar to the relationship of g e o l o g i c
formations to each other.
2.13.2

Fractal

Mathematics

In a classic paper Mandelbrot ( 1 9 6 7 ) d e m o n s t r a t e d that the measured length of an


irregularly shaped o b j e c t , in that c a s e the coastline of Great Britain, d e p e n d s upon the

Chapter T w o

FIGURE 2.25

Dependence of the horizontal plane transverse spatial moment, X , on travel time. Source:
72

G. Dagan, Water Resources Research 2 4 , no. 9 (1 9 8 8 ) : 1 4 9 1 - 1 5 0 0 . Copyright by the American Geophysical


Union.

scale o f measurement. T h e degree o f irregularity o f the coastline i s independent o f the


scale at which the measurement takes place. A coastline has a similar irregular shape if
viewed from the b e a c h , an airplane, or a space station.
If we measure a straight line with a ruler, the length is constant and equal to the
n u m b e r of units times the unit length in which the measurement is made. If the unit
length is halved, the n u m b e r of units is doubled, but the overall length remains the
same.
If we measure an irregular line, the accuracy of the measurement is a function of
the scale of the measuring device. In the c a s e of our coastline, if we used a ruler that
had a minimum scale of 1 0 0 km, we would get a certain approximate length. If we then
used a ruler that had a minimum scale of 1 km, we would get a different, m o r e accurate
measurement, which would be longer b e c a u s e we could m o r e accurately trace the
irregularities of the coast. If we made yet another measurement using a ruler with a 1
m scale, we would obtain a third, even m o r e accurate length, that was longer still. Even
so, we have yet to measure the curve of the coastline as it b e n d s around individual rocks,
much less the curve of the coastline around grains of sand. Using conventional geometry,
the measured length is a function of the minimum scale of measurement.

Mass Transport in Saturated Media

I
5

I
10

I
15

87

I
20

25

30

I
35

40

L
45

= nV/e,
FIGURE 2.26

Dependence of the vertical plane transverse spatial moment, X

3 3

, on travel time. Source:

G. Dagon, Water Resources Research 24, no. 9 (1 9 8 8 ) : 1 4 9 1 -1 5 0 0 . Copyright by the American Geophysical
Union.

Mandelbrot ( 1 9 6 7 ) s h o w e d that for the c a s e of a coastline, there was a constant


length that was i n d e p e n d e n t of the unit of measure. This was e x p r e s s e d as
J = Nqf = constant

(2.59)

where
J = constant length
N = n u m b e r of units
r) = unit of m e a s u r e m e n t
/ = fractal dimension
T h e fractal dimension, /, is a n u m b e r that, for a o n e dimensional irregular o b j e c t ,
lies b e t w e e n 1 and 2. If / is 1, then the line is straight, and if it is 2, the line is so irregular
that it fills a plane.
T h e length measured at m e a s u r e m e n t scale n, L(t]) can be related to / by
('/)=yy"

(2.60)

Chapter T w o

2.13.3

Fractal Geometry and D i s p e r s i o n

Wheatcraft and Tyler ( 1 9 8 8 ) devised an approach to understanding dispersion based


on the application of fractal geometry. T h e path taken by a particle of water traveling
through a p o r o u s medium is not straight and thus is longer than the straight-line distance
between the two endpoints. This p h e n o m e n o n has already b e e n discussed as tortuosity.
If the tortuosity increases with increasing length of the flow path, the streamtube can
be c o n s i d e r e d to be a fractal path.
O n e p r o b l e m in using fractal mathematics in real-world p r o b l e m s is that, from
Equation 2.60, as r\ tends toward zero, ( ( ) ) grows toward infinity. T h e r e are no infinitely
long flowpaths in an aquifer; therefore, there must be s o m e lower b o u n d to the length
of r\. T h i s is on the o r d e r of the size of the average p o r e radius, as at a scale lower than
this length the water particles can simply flow across the p o r e and don't necessarily
follow a fractal path. This lower limit is called the f r a c t a l c u t o f f l i m i t , t\ .
c

If we c o n s i d e r the fractal path taken by a particle of water to have a length L and


the straight-line length from the start of the fractal streamtube to the e n d to be 4, the
following relationship holds:
s

L = ^-'L/

(2.61)

Here t\ c a n n o t be less than the fractal cutoff limit, nor can it be longer than the straightline length of the streamtube. In a m o r e general sense, the length of any part of the
streamtube, x , can be found from
f

x = n '~ x/
f

(2.62)

w h e r e x is the straight line length from the origin of the streamtube to the coordinate
s

of the e n d of x and r\ has a length less than x but greater than zero.
s

F r a c t a l t o r t u o s i t y , 7 } , i s defined a s
7} = ^ =

(2-63)

As the travel distance b e c o m e s greater, the fractal tortuosity also increases. This
is a result of the water particle encountering m o r e and m o r e heterogeneity as the flow
path e x t e n d s across heterogeneity boundaries.
T h e average linear velocity, v , is the average velocity determined by dividing the
x

straight-line distance, x , by the travel time. T h e fractal velocity, v , is the fractal distance,
s

x , divided by the travel time.


f

ax,
Although we measure the value of

dx

in the field, particles of water containing

solute actually are traveling at v , the fractal velocity. This has an effect on dispersion.
f

If a tracer e x p e r i m e n t is performed, a value of the coefficient of longitudinal


dispersivity can be obtained. We will call this the field-measured dispersion coefficient,
D. T h e concentration variance, a

2
c

, for the time when C/Q, = 0.5 and for distance x,

is equal to
a

2
c

= 2D x
m

(2.64)

89

Mass Transport in Saturated Media

S i n c e the tracer particles have moved along the fractal streamtube, they have actually
g o n e a distance x , which is longer than x . If the e x p e r i m e n t is c o n t i n u e d over longer
f

flow paths, the fractal s t r e a m t u b e length will g r o w faster than the straight line distance.
As a result the tracer will disperse m o r e than is predicted by 2.64. T h e variance in the
fractal streamtube is
<r

= 2D Xj

(2.65)

w h e r e D is the coefficient of p o r e scale longitudinal dispersion not affected by the scale


L

of measurement. It is typically on the o r d e r of 0.001 to 1 cm. T h e measured coefficient


o f f i e l d dispersion, which i s scale dependent, can b e related t o the coefficient o f p o r e
scale dispersion b y
i

D =
m

D t, -W~

(2-66)

Equation 2 . 6 6 suggests that the coefficient of field dispersion will increase as the
l

scale of observation increases and that is it is proportional to x/~ . However, for a


nonfractal p o r o u s medium, / = 1 and D = D .
m

In a real aquifer a particle of water will follow o n e fractal streamtube, but the actual
path it will follow is uncertain. T h e r e is an e n s e m b l e of fractal streamtubes, e a c h with
the s a m e origin and e a c h with the s a m e straight line distance, x , but e a c h with a different
s

fractal length, x . T h e r e f o r e , there are many realizations of the fractal s t r e a m t u b e length,


f

and we can c o m p u t e a mean and a variance for each.


If we have an e n s e m b l e of fractal streamtubes, e a c h with a different fractal length,
x , then we can find a variance for the e n s e m b l e . F r o m Equation 2.62 we have
s

v a K * , ) = v a r < V ~ x/)

(2.67)
2

w h e r e w e use the notation var(A-) a s equivalent t o a .


If x , the straight-line distance, is fixed and the fractal dimension, / is constant over
s

the scale that we are studying, the only s o u r c e of variation in x is variation in the fractal
f

cutoff limit, n . T h e r e f o r e , Equation 2.67 b e c o m e s


c

varU,) = x

' var(t7 '"')

(2.68)

Equation 2 . 6 8 c o n t a i n s an e x p r e s s i o n for variation at the m i c r o s c o p i c scale


(vaK^

1 _

) ) b a s e d o n the fractal cutoff limit. T h e term x

a c c o u n t s for aquifer het-

erogeneities, b e c a u s e it provides for variation in the travel-path length as the scale of


observation increases.
T h e variation in travel length on the basis of advection dispersion (diffusional)
theory can be found from
varUf) = 2Dx

(2.69)

w h e r e D is the field-measured ( c a l c u l a t e d ) dispersion coefficient and x is the straight


m

line travel distance, which is equivalent to the scale of measurement. Equations 2 . 6 8 and
2.69 can be c o m b i n e d to yield an e x p r e s s i o n for scale-dependent dispersion coefficient
b a s e d on fractals.
D

= 0.5 v a r ( V - - 0 *

2 /
s

~'

(2.70)

Chapter T w o

2.13.4

Fractal Scaling of Hydraulic Conductivity

Neuman ( 1 9 9 0 ) has presented a fractal analysis of the scale effect of dispersion, previously
discussed in Section 2 . 1 1 . Neuman refers to dispersivities that are measured in the field
as apparent dispersivities b e c a u s e they are obtained by calculations that depend upon
the theory that the o b s e r v e r was using. Figure 2.27 contains a logarithmic plot of apparent
longitudinal dispersfvity as measured in field and lab studies, tx , as a function of the
m

travel distance,

or apparent length scale. All data with an apparent length scale greater

than 3 5 0 0 m w e r e e x c l u d e d for theoretical reasons. Regression analysis s h o w e d that


although the data are widely scattered, a best fit line with narrow 9 5 % confidence bands

Apparent length scale I ml


FIGURE 2.27

Apparent longitudinal dispersivity from field and laboratory studies as a function of the

scale of the study. Results from the calibration of numerical models are not included. Source: S. Neuman,
Water Resources Research 2 6 , no. 6 (1990): 1 7 4 9 - 5 8 . Copyright by the American Geophysical Union.

91

Mass Transport in Saturated Media


c o u l d be obtained. T h e equation for the line is
a

4 6

= 0.01754'

(2.71)

T h i s line of b e s t fit has a regression coefficient, r , of 0 . 7 4 , which m e a n s that it a c c o u n t s


for 7 4 % o f the variation a b o u t the mean. T h e o t h e r 2 6 % may b e due t o e x p e r i m e n t a l
and interpretive errors or may represent deviation of the real system from that d e s c r i b e d
b y Equation 2.71. T h e 9 5 % c o n f i d e n c e intervals a b o u t the coefficient o f 0 . 0 1 7 5 are 0 . 0 1 1 3
and 0 . 0 2 7 2 and the 9 5 % c o n f i d e n c e intervals a b o u t the e x p o n e n t of 1.46 are 1.30 and
1.61.
T h e apparent longitudinal dispersivity can be calculated from the variance of the
log hydraulic conductivity by the equation ( N e u m a n and Zhang 1 9 9 0 )
a = c W

(2.72)

w h e r e c is a constant related to anisotropy that is approximately 0.5 for isotropic media


( Neuman and Zhang 1 9 9 0 , Figure 6) and c a n be 1 or greater in anisotropic media (Zhang
and Neuman, 1 9 9 0 , Figure 1 ) .
C o m b i n i n g Equations 2.71 and 2.72, we obtain
2

o-, = ( 0 . 0 1 7 5 / O 4 -

4 6

(2.73)

T h e log hydraulic conductivity, Y = In K has a semivariogram, y , at travel distance,


Y

s, given by
2 v

y 's) = 4 s

(2.74)

w h e r e Z is a c o n s t a n t and v is a coefficient k n o w n as the Hurst coefficient such that


0

0 < 2v < 1. It c a n be s h o w n ( N e u m a n 1 9 9 0 ) that such a semivariogram characterizes a


self-similar random field with h o m o g e n e o u s i n c r e m e n t s that can be d e s c r i b e d by the
fractal d i m e n s i o n
/= G+ 1 - v

(2.75)

w h e r e G = 1 for a o n e dimensional field, 2 for a two-dimensional field, and 3 for a threedimensional field.
T h e variance of Y has the semivariogram
N

A).0175 \

F r o m the e x p o n e n t i a l value o f 4 . i t

t>

(2.76)
th

a t

value o f 0 . 2 3 , indicating

that the regression line of Figure 2.27 represents a log hydraulic conductivity field that
is self-similar with a fractal dimension f = G + 0.77. Neuman ( 1 9 9 0 ) performed a similar
analysis on only the data representing flowpaths less than 1 0 0 m long. He o b t a i n e d the
following results on this data subset:
<x = 0 . 0 1 6 9 4 '
m

M4) = ( ^ ) C "
/=G+0.74

53

(2-77)
(2-78)
(2.79)

Chapter T w o

On this basis Neuman has p r o p o s e d that the log hydraulic conductivity values of the
geological materials represented by Figure 2.27 can be characterized by the following,
which form a universal scaling rule:

(2.80)

Vy(4) =
f=G+0.75

(2.81)

T h e preceding scaling rule can be considered to account for the self similarity of
log hydraulic conductivities from a wide variety of geologic materials over a large range
of flowpath lengths and under diverse flow conditions. It would not necessarily describe
any particular location.

Columns
I 2 3 4 5 6 7 8 9 10 11 12 13
Explanation

S.

Solute source

4
5
I

Impermeable boundary

Constant-head cells

FIGURE 2.28

Finite-difference grid and boundary conditions for a deterministic model of solute transport.

Source-. A. D. Davis, Ground Water 24, no. 5 (1 9 8 6 ) : 6 0 9 - l 5 . Used with permission. Copyright 1986
Water W e l l Journal Publishing Co.

Mass Transport in Saturated Media

2.14

93

Deferministic Models of Solute T r a n s p o r t


Although w o r k e r s in stochastic theory have asserted that the theoretical basis for the
deterministic advective-dispersive solute transport equation is suspect e x c e p t for long
times and large distances (Anderson, 1 9 8 4 ) , it has b e e n used with a great deal of s u c c e s s
in many field and m o d e l applications. O n e of the p r o b l e m s raised with field scale
application of the advective-dispersive equation is that it requires an unrealistically high
value for the coefficient of longitudinal dispersion.
A model study by Davis ( 1 9 8 6 ) demonstrates that deterministic m o d e l s can be
developed that i n c o r p o r a t e heterogeneities. He m o d e l e d two aquifers with identical
boundary conditions (Figure 2 . 2 8 ) . O n e was uniform (Figure 2 . 2 9 ( a ) ) and o n e had

1,

o
Flow
(a)

FIGURE 2.29

Flow
Scale

(b)

Model areas for finite difference solute transport model with (a) uniform transmissivity and

(b) with heterogeneous transmissivity. Source-. A, D, Davis, Ground Wafer 2 4 , no. 5 (1 9 8 6 ) : 6 0 9 - 1 5. Used
with permission. Copyright 1 9 8 6 Water W e l l Journal Publishing Co.

Chapter T w o

variable transmissivity in the form of m o r e permeable channels (Figure 2 . 2 9 ( b ) ) . T h e


deterministic model, based on the two-dimensional solute transport equation, was used
with small values of <X and a , 0 . 0 0 0 3 m, and 0 . 0 0 0 0 9 m, respectively. T h e resulting solute
L

plume in the uniform media is very long and narrow. See Figure 2 . 3 0 ( b ) . If larger values
of a and a are u s e d 3 m and 1 m, respectivelythen a much broader plume results.
L

S e e Figure 2 . 3 0 ( a ) . However, if the h e t e r o g e n e o u s aquifer is used with the small values


of dispersivity, the resulting plume, shown in Figure 2 . 3 0 ( c ) has a size very similar to
that created in the uniform media by using large values of dispersivity. This demonstrates
that if deterministic models include the aquifer heterogeneities, then it may be possible
to u s e dispersivity values that are m o r e on the order of lab-scale values. Davis ( 1 9 8 6 )
used the advective-dispersion equation in a model with varying transmissivities and with

Flow

Flow

Flow

a = 3 m " i = I in
L

T= 0.0363 m /sec
Q = 0.165 m-Vsec
; = 0.01 yr

a = 0.0003 m

a = 0.00009 m
r

7 = 0.0363 m-/sec
Q = 0.165 m /sec
I = 0.01 yr
3

(b)

(a)

a = 0.0003 m
o = 0.00009 m
7, = 0.223 m /sec 7, = 0.0223 rrr/sec
Q = 0.165 m '/sec
I = 0.01 yr
(c)

Scale
0

FIGURE 2.30

200 m

Model results for finite-difference solute-transport model, (a) Uniform media with large

dispersivity values, (b) uniform media with small dispersivity values, and (c) heterogeneous media with small
dispersivity values. Source: A. D. Davis, Ground Water 24, no. 5 ( 1 9 8 6 ) : 6 0 9 - l 5. Used with permission.
Copyright 1986 Water W e l l Journal Publishing Co.

Moss Transport in Saturated Media

95

a value of a of only 0 01 m was a b l e to r e p r o d u c e a solute plume that e x t e n d e d over


L

a flow length of about S 0 0 m. He found that a fine mesh for the finite difference model
grid was necessary for accurate results. Figure 2.31 c o m p a r e s the results of his model
results with the field data.
F r o m a practical standpoint, the advection dispersion solute transport equation
has b e e n shown to give e x c e l l e n t results when applied to field situations. T h e K o n i k o w

Bredehoeft solute transport m o d e l ( M O C ) is o n e o f the most widelv used solute transport


m o d e l s and is b a s e d on the advection dispersion equation ( K o n i k o w and B r e d e h o e f t
1 9 ^ 8 ) . It was used with great s u c c e s s in modeling the m o v e m e n t of conservative g r o u n d
water c o n t a m i n a n t s at the Rocky Mountain Arsenal ( K o n i k o w 1 9 7 7 ; K o n i k o w and
T h o m p s o n 1 9 8 4 ) . However, even if the advection dispersion m o d e l can be fitted to data

Explanation
L i m i t of aquifer
25 Contours of equal chloride concentration (mg//)
(dashed where approximated)
F I G U R E 2.31

Comparison of (a) field observations ot solute plume in on aquifer and (b) solute plume

os computed by finite-difference solute-transport model for a heterogeneous aquifer. Source: A. D. Davis,


Ground Water 2 4 , no. 5 { 1 9 8 6 1 : 6 0 9 - 1 5 . Used with permission. Copyright 1 9 8 6 Water W e l l Journal
Publishing Co.

96

Chapter T w o

on the past m o v e m e n t of a solute plume, this d o e s not necessarily mean the model will
be accurate in predicting future movement of the plume.

Case Study: Borden Landfill Plume


An abandoned landfill in a s h a l l o w sand aquifer at Canadian Forces Base in Borden, O n t a r i o , has been e x t e n s i v e l y studied (Cherry, 1 9 8 3 ; M a c F a r l a n e et al., 1 9 8 3 ) . F r i n d and
H o k k a n e n ( 1 9 8 7 ) made a very interesting study of the plume based on a deterministic model.
T h e landfill w a s active from 1 9 4 0 to 1 9 7 6 and covers about 5.4 ha to a depth of 5 to
10 m. F i g u r e 2 . 3 2 s h o w s the location of water table w e l l s and multilevel sampling devices.
T h e multilevel sampling devices are concentrated along the long a x i s of the plume of groundwater contamination. T h e vertical location of the sampling points along cross section A - A '
are s h o w n in F i g u r e 2 . 3 3 . T h e aquifer is about 20 m thick beneath the landfill and thins to
about 9 . 5 m in the direction of ground-water flow. T h e aquifer consists of laminated fine to
medium sand. An average hydraulic conductivity of 1.16 X 1 0 ~
5.8 x 1 0

cm/sec h o r i z o n t a l l y and

cm/sec vertically w a s used in the model with a p o r o s i t y of 0 . 3 8 . In 1 9 7 9 a very

extensive study of the water quality of the plume w a s conducted. F i g u r e 2 . 3 4 s h o w s the


plume of chloride contamination along c r o s s section A - A ' . In 1 9 7 9 the plume extended about
7 5 0 m from the landfill and had sunk to the bottom of the aquifer and then moved laterally
with the f l o w i n g ground water. T h e s i n k i n g of the plume is believed to be caused by r e charge concentrated in a sand pit to the north of the l a n d f i l l , which is in the direction of
flow.

o Multilevel sampler
<D Bundle piezometer
FIGURE 2.32

Location of landfill at Canadian Forces Base, Borden, Ontario, showing location of cross

section and monitoring network. Source-. E. O. Frind and G. E. Hokkanen, Wafer Resources Research 23,
no. 5 (1 9 8 7 ) : 91 830. Copyright by the American Geophysical Union.

97

M a s s T r a n s p o r t i n Saturated M e d i a

FIGURE 2.33

Cross section of aquifer at the Borden landfill showing the location of multilevel monitoring

devices. Source-. E. O. Frind and G. E. Hokkanen, Water Resources Research 2 3 , no. 5 (1987): 9 1 8 - 3 0 .
Copyright by the American Geophysical Union.

Distance (m)
FIGURE 2.34

Chloride plume along the Borden landfill cross section in 1 9 7 9 . Values are in milligrams

per liter. Source-. E. O. Frind and G. E. Hokkanen, Wafer Resources Research 2 3 , no. 5 (1987): 91830.
Copyright by the American Geophysical Union.

T h e finite-difference g r i d system for the cross-sectional model is s h o w n in F i g u r e 2 . 3 5 .


Equipotential lines for observed conditions were e s s e n t i a l l y vertical (Figure 2 . 3 6 ) . T h e model
w a s calibrated against the water-table contours for steady-state conditions.
S e n s i t i v i t y a n a l y s e s were performed to determine the impact of v a r y i n g y. and oc .
L

F i e l d tests had indicated that the value of y. at the site is on the order of 5 to 10 m
L

(Sudicky, Cherry, and F r i n d 1 9 8 3 ) . F i g u r e 2 . 3 7 s h o w s the sensitivity of the plume to the value

y,
40-

200

400

600
Length (in)

F I G U R E 2.35

Finite-difference grid for Borden landfill solute transport model. Source: E. O. Frind and

G. E. Hokkanen, Water Resources Research 2 3 , no. 5 (1 9 8 7 ) : 91830. Copyright by the American Geophysical Union.

800

10(H)

Mass Transport in Saturated Media

100

200

99

300

400

500

600

700

800

900

1000

Distance (m)
FIGURE 2.36

Equipotential lines from the calibration of the Borden landfill solute-transport model; values

in meters above datum. Source-. E. O. Frind and G. E. Hokkanen, Water Resources Research 2 3 , no. 5
(1 9 8 7 ) : 91 830. Copyright by the American Geophysical Union.

of a . The value of oc was kept at 10 m and y. was varied from 0.005 m to 1.0 m. It can be
T

seen that the shape of the plume is very sensitive to the value of a . With a high value of a ,
T

the plume spread through the entire vertical thickness of the aquifer, whereas with a low
value it tended to sink toward the bottom. Figure 2.38 illustrates the fact that the plume was
not very sensitive to changes in the value of a over the range tested. The value of a was
(

kept constant at 0.01 m, whereas i

varied from 2.5 to 20 m. This figure is slightly misleading

in that there is a 1 0 : 1 vertical exaggeration, so that the vertical spreading is more obvious
than the horizontal. Also, the value of y. was varied by a factor of 200, whereas X was
T

varied only by a factor of 8.


Additional sensitivity analyses were conducted with respect to the water table boundary conditions and the concentration, size, and growth pattern of the source. The authors
found that in order to reproduce the observed distribution, a source history that included
multiple periods of high concentration was needed. Figure 2.39(a) shows the shape of the
observed plume. Figure 2.39(b) illustrates the shape of a plume generated by a source with
a history in which the concentration gradually increased (smooth source concentration), and
Figure 2.39(c) contains the computed plume with the best match to the observed plume. It
was generated by a run of the model in which the source concentration had two different
periods of peak concentration. Although the solution was not uniquethat is, several different combinations of model inputs might yield the same outputthe shape of the plume could
be reproduced with good accuracy. This was especially true at the leading edge of the
plume, which is the most important part from the standpoint of predicting the movement of
the plume into uncontaminated areas of the aquifer.

100

Chapter T w o

FIGURE 2.37

Sensitivity analysis of the Borden landfill solute-transport model with respect to transverse

dispersivity. Source-. E. O. Frind and G. E. Hokkanen, Water Resources Research 23, no. 5 (1 987): 91 8
3 0 . Copyright by the American Geophysical Union.

Mass Transport in Saturated Media

FIGURE 2.38

101

Sensitivity analysis of the Borden landfill solute-transport model with respect to longitudinal

dispersivity. Source-. E. O. Frind and G. E. Hokkanen, Water Resources Research 23, no. 5 {1 9 8 7 ) : 918
3 0 . Copyright by the American Geophysical Union.

F I G U R E 2.39

Comparison of (a) the observed chloride plume at the Borden landfill with (b) the chloride

plume simulated by the solute transport model with a smooth source concentration and (c) the chloride
plume simulated by the solute transport model with a doubly peaked source concentration. Source-. . O.
Frind and G. E. Hokkanen, Wafer Resources Research 2 3 , no. 5 (1987): 91 8 - 3 0 . Copyright by the American
Geophysical Union.

M a s s T r a n s p o r t i n Saturated M e d i a

2.15

103

Transport in Fractured Media


Solute transport in fractured rock media is as important a p r o c e s s as transport in porous media. However, less research has b e e n d o n e on this topic than on transport in
p o r o u s media. O n e reason may be the c o m p l e x i t y of solute transport in fractured media.
T h e rock in which fractures exist is porous. Hence, fluid moves in the fractures as well
as in the rock matrix Solutes in the fractures can diffuse into the fluid contained in the
rock matrix and vice versa (Neretnieks 1 9 8 0 ) . T h e fractures themselves are not s m o o t h
channels but contain dead e n d passages that hold nonmoving water into which solutes
can diffuse (Raven, Novakowski, and Lapcevic 1 9 8 8 ) .
Berkowitz, Bear, and Braester ( 1 9 8 8 ) suggested that solute transport in fractured
media can be c o n s i d e r e d at a n u m b e r of different scales. A very-near-field scale would
be a single fracture near the source. A near-field scale would include a few fractures
near the source. At a larger scale, the far field, the fracture network and the p o r o u s
media matrix would have separate, discernable impacts on flow. At a very-far-field scale,
which exists at c o n s i d e r a b l e distance from the s o u r c e , the entire flow domain can be
c o n s i d e r e d as an equivalent p o r o u s medium in which the repeating fractures b e c a m e
large pores.
A n u m b e r of different a p p r o a c h e s to solute transport in fractured media have b e e n
attempted. T h e s e include analysis of transport in a single fracture in which effects of the
transport in the fractures as well as interactions with a p o r o u s matrix are c o n s i d e r e d
(e.g., Grisak and Pickens 1 9 8 0 , 1 9 8 1 ; Tang, Frind, and Sudicky 1 9 8 1 ; Rasmuson and
Neretnieks 1 9 8 1 ; Rasmussen 1 9 8 4 ; Sudicky and Frind 1 9 8 4 ) . Sudicky and Frind ( 1 9 8 2 )
and B a r k e r ( 1 9 8 2 ) e x a m i n e d transport in a media that consists of equally s p a c e d fractures
in a p o r o u s media. E n d o and o t h e r s ( 1 9 8 4 ) m a d e a deterministic study of flow in an
irregular network of fractures contained in an i m p e r m e a b l e host rock, whereas Schwartz
et al. ( 1 9 8 3 ) and Smith and Schwartz ( 1 9 8 4 ) a p p r o a c h e d the s a m e p r o b l e m using a
stochastic model. Berkowitz, Bear, and Braester ( 1 9 8 8 ) and Schwartz and Smith ( 1 9 8 8 )
e x a m i n e d the conditions under which the p o r o u s media matrix and the fractures can
be c o n s i d e r e d to be a continuum that is representative of an equivalent p o r o u s media.
Raven, Novakowski, and Lapcevic ( 1 9 8 8 ) m a d e a field study of flow through a single
fracture to test a model that incorporates the effects of nonflowing water in the fractures.
Tsang et al. ( 1 9 8 8 ) and M o r e n o et al. ( 1 9 8 8 ) e x a m i n e fracture flow on the basis of the
assumption that m o s t of the flow is c o n c e n t r a t e d in a few channels.
O n e of the first considerations in dealing with fracture flow is deciding h o w to
treat flow in a single fracture. S o m e authors (e.g., Tang, Frind, and Sudicky, 1 9 8 1 ; Schwartz
and Smith 1 9 8 8 ) a s s u m e that the fluid in a fracture is all moving at a constant velocity.
Conversely, E n d o et al. ( 1 9 8 4 ) treated flow in a fracture to be two-dimensional, with a
parabolic velocity profile a c r o s s the width of the fracture, as shown in Figure 2 . 4 0 .
Transport within a single fracture is due to advection, which o c c u r s at different rates,
depending upon the position between the parallel walls of the fracture, and molecular
diffusion, both normal and parallel to the flow direction.
Hull, Miller, and C l e m o ( 1 9 8 7 ) e x a m i n e d the conditions w h e r e b y diffusion within
the fracture n e e d s to be considered. In a fracture with parallel sides, the solute transport
within the fracture is d e s c r i b e d by

104

Chapter T w o

FIGURE

2.40

Horizontal distribution of flow in a vertical fracture and diffusion into the porous media

matrix.

BC
dt

= D*

By

6l/[(r)-(r) ]

BC
dx

(2.82)

where
V = average fluid velocity in a fracture
T = fractional transverse position in a fracture
At high flow rates, advection will dominate and the concentration will follow the
velocity profile of Figure 2.40. At low velocities, diffusion will be important, since the
concentration gradient at the solute front will be high and the distance will be short.
Under these conditions, diffusion will h o m o g e n i z e the solute across the width of the
fracture.
If L is the length of the fracture between c r o s s fractures and /? is the aperture of
the fracture, the fracture residence time is L/V. This can be c o m p a r e d with (f)/2) /D to
determine if diffusion needs to be considered (Crank 1 9 5 6 ) . If diffusion induces a change
in the tracer concentration of less than 2% over a distance of 1 0 % of the width of the
fracture, the diffusion can be considered negligible, and the residence time in the fracture
will be
2

^ < 0 . 0 0 3 ^
V
D

(2.83)

Mass Transport in Saturated Media

105

If diffusion affects the tracer c o n c e n t r a t i o n to the e x t e n t that the tracer front is at


98% of the equilibrium value at all points a c r o s s the fracture, the diffusion has homogenized the front, and the residence time in the fracture will be
i > 0 . 0 5
V
D

(2.84)

Figure 2.41 indicates the c i r c u m s t a n c e s u n d e r which fracture flow can be c o n s i d e r e d


to be o n e - or two-dimensional. Fracture r e s i d e n c e time ( L / V) is plotted against fracture
9

aperture on this figure, which is b a s e d on a diffusion coefficient of 1.7 x 1 0 ~ m V s e c .


T h e figure s h o w s the c o n d i t i o n s under which diffusion will h o m o g e n i z e the flow so that
the transport within the fracture can be treated as one-dimensional (uniform c o n d i t i o n s
a c r o s s the a p e r t u r e ) . However, diffusion will still spread the tracer in advance of the
advei ling water Ft ir even large I'rai tures ol I mm aperture, this will o i c t i r with a resident c
time of 1 min or m o r e . T h i s suggests that for m o s t flow situations, o n e d o e s n o t n e e d
to c o n s i d e r the velocity distribution a c r o s s the fracture.
W h e n the flow in a fracture is h o m o g e n e o u s , the mass transport can then be
d e s c r i b e d by the o n e dimensional advection-dispersion equation with the longitudinal
dispersion coefficient equal to (Hull, Miller, and C l e m o 1987)
D =

(2.85)

210D

O n e a p p r o a c h to solute transport modeling is to d e t e r m i n e the flux of water


through the fractures and then use a numerical t e c h n i q u e k n o w n as a random walk
m o d e l to simulate diffusion of the solute (Hull, Miller, and C l e m o 1987). This ignores
any diffusion into the p o r o u s media matrix. According to W i t h e r s p o o n et al. (1980),

lu.ni

lOjum

LOO/tm

I mm

I cm

Fracture aperture
FIGURE 2.41

Fracture residence time necessary for homogenization of the tracer across the fracture

width by molecular diffusion. Source. Modified from f. C. Hull, J. D. Miller, and T. M. Clemo, Water
Resources Research 2 3 , no. 8 (1987): 1 5 0 5 - 1 3 . Copyright by the American Geophysical Union.

106

Chapter T w o

flow through a fracture can be d e s c r i b e d by Darcy's law using an equivalent hydraulic


conductivity for a fracture, K , given by
f

T h e quantity of flow, Q can be found from the c u b i c law

6=

7 g / r #

(2.87)

where
g = acceleration of gravity
7 = hydraulic gradient along the fracture
a = width of the fracturethat is, the third dimension after length and
aperture
H = viscosity of fluid
If the velocity in the channel needs to be d e s c r i b e d in two dimensions, this can
be d o n e with three equations: o n e for the maximum velocity in the c e n t e r of the fracrure,
o n e for the flow velocity profile across the aperture, and o n e for the vertical velocity
profile in the fracture.
T h e m a x i m u m velocity can be found from (Hull, Miller, and C l e m o 1 9 8 7 ) :

in

1.5 + 1.1664

l/ (max) =
x

(2.88)

T h e velocity profile across the aperture is given by:


2

V (y) = 4 ( T - T )

(2.89)

w h e r e T = fractional transverse position in a fracture, y/fl.


T h e vertical velocity profile is given by
V (0 = 15.56C - 9 7 . 7 2 C
x

+ 308{

- 513C

+ 43K

- 143.7C

(2.90)

w h e r e = fractional vertical position in a fracture, z/a.


Raven, Novakowski, and Lapcevic ( 1 9 8 8 ) pointed out that the fractures through
which flow o c c u r s are not s m o o t h , parallel plates but have irregular walls that p r o m o t e
the formation of z o n e s along the e d g e of the fracture where the water is immobile
(Figure 2 . 4 2 ) . T h e fluid moves through the m o b i l e zone, but the solutes can diffuse into
the i m m o b i l e fluid zones. T h e solute would be stored in the immobile fluid during the
early part of solute transport and would be released from storage if the solute c o n c e n tration in the m o b i l e fluid would d e c r e a s e f o r e x a m p l e , as might happen during the
latter part of a slug injection test. T h e y derived an advection dispersion equation for
mass transport in the fracture with "transient solute storage" in the immobile fluid z o n e
(ADTS m o d e l ) . (An even m o r e c o m p r e h e n s i v e model would have provision for transient
solute storage in the i m m o b i l e fluid z o n e as well as diffusion into a p o r o u s media matrix.)
A field test was performed on the flow through a single fracture that had b e e n
isolated by packers in the b o r e h o l e . Water was injected into o n e b o r e h o l e and withdrawn

107

Mass Transport in Saturated Media


Fluid
velocity
proli le

Monile
fluid /.one

F I G U R E 2.42

Vortices
and eddies

"Inerial
core"

Immobile
fluid /one

Zones of mobile and immobile water in a fracture. Source: K. G. Raven, K. S. Novakowski,

and P. A. Lapcevic, Water Resources Research 24, no. 12 (1988): 2 0 1 9 - 3 2 . Published by the American
Geophysical Union.

from another. T h e water c o n t a i n e d a t r a c e r for the first tew h o u r s of the test, and then
water without the tracer was again injected. Figure 2.43 c o n t a i n s circles representing the
field data, in t e r m s of relative c o n c e n t r a t i o n , plotted versus e l a p s e d time. Also s h o w n
on this figure a r e the results of a conventional advection-dispersion ( A D ) m o d e l and an
advection-dlspersion transient storage ( A D T S ) m o d e l . B o t h m o d e l s m a t c h e d the o b s e r v e d
data for the first few h o u r s of t h e test. However, t h e ADTS m o d e l was far s u p e r i o r in
matching the f i e l d data over the entire c o u r s e o f t h e test. T h e effect o f transient storage
was to r e d u c e the p e a k c o n c e n t r a t i o n and to i n c r e a s e the c o n c e n t r a t i o n s a b o v e what
would be p r o d u c e d by advection dispersion a l o n e during t h e later p e r i o d s of t h e test.

2.16

Summary
Solutes dissolved in g r o u n d water are transported in two ways. Diffusion will c a u s e
solutes to m o v e in the direction of t h e c o n c e n t r a t i o n g r a d i e n t t h a t is, from areas of
higher to lower c o n c e n t r a t i o n . T h i s transport c a n o c c u r even if the g r o u n d water is n o t
flowing and may be the major factor in mass transport in g e o l o g i c materials of very low
permeability.
Solutes a r e also transported by the p r o c e s s of advection, which is also k n o w n as
c o n v e c t i o n . T h i s o c c u r s as t h e flowing g r o u n d water carries the dissolved solutes with
it. At t h e scale of a few p o r e diameters, g r o u n d water will m o v e parallel to the flow path
at different rates due to differences in p o r e size. T h i s c a u s e s t h e solute p l u m e to spread
along the direction of the flow path, a p r o c e s s called longitudinal dispersion. T h e solute
p l u m e will also spread laterally as flow paths diverge a r o u n d mineral grains, a p r o c e s s

108

Chapter T w o

o Ground-water samples
-

AD model simulation

A D T S model simulation

12

15

Elapsed time (hr)


FIGURE 2.43

Comparison ot field data from a tracer test in fractured rock with results of model simulation

using an advection-diffusion (AD) model and an advection-diffusion transient storage (ADTS) model. Source:
K. G. Raven, K. S. Novakowski, and P. A. Lapcevic, Wafer Resources Research 24, no. 12 (1988): 201 9
3 2 . Published by the American Geophysical Union.

k n o w n as transverse dispersion. At the laboratory column scale, the movement of a


contaminant through a uniform p o r o u s media can be described by the advection
dispersion equation, which a c c o u n t s for advection, diffusion, and p o r e scale dispersion.
In field-scale studies it has b e e n found that the coefficient of longitudinal dispersion
o b t a i n e d from the advection dispersion equation increases with the length of the flow
path. This is due to the h e t e r o g e n e o u s nature of aquifer materials. As the length of the
flow path increases, the range of permeability values that affect the rate of g r o u n d w a t e r
flow also increases. This causes the resulting solute plume to spread out m o r e and m o r e
This can b e called macrodispersion. For flow paths less than 3500 m an apparent dis
persion coefficient can be statistically correlated to the length of the flow path by the

Moss Transport in Saturated Media


1

109

4 6

equation a = 0 . 0 1 7 5 Z , ' - Eventually the apparent dispersivity appears to reach a maximum value.
Stochastic m e t h o d s of analysis have also b e e n developed to analyze solute transport
at the field scale. Stochastic m e t h o d s are b a s e d on the variation in the hydraulic conductivity values b e c a u s e it is that variation that causes the solute plume to spread. T h e
ground-water velocity d e p e n d s u p o n the porosity as well as the hydraulic conductivity,
but the hydraulic conductivity varies over a m u c h greater range than porosity.
m

At the field scale the spreading due to hydraulic conductivity variation is much
greater than that due to pore-scale dispersion. B o t h stochastic and advection-dispersion
m o d e l s demonstrate that the primary m o v e m e n t of the solute plume is due to advection.
T h e stochastic m o d e l yields the m o v e m e n t o f the c e n t e r o f mass o f the solute plume
from the average rate of m o v e m e n t of the ground water. T h e variance of the solute
concentration a b o u t the mean position, or the s e c o n d spatial m o m e n t , is also o b t a i n e d
from stochastic models.
If o n e has sufficient knowledge of the distribution of hydraulic conductivity in an
aquifer, then a numerical advection dispersion model of ground-water flow can be developed that uses a pore-scale dispersion value. This type of model has theoretical validity,
b e c a u s e the necessary coefficient of longitudinal dispersion d o e s not c h a n g e with flow
path length. It can be used to predict future solute c o n c e n t r a t i o n s at specific places and
times. Narurally, such predictions will not be 1 0 0 % accurate, b e c a u s e o n e can obviously
never k n o w the value of the hydraulic conductivity every place in the flow field.

Chapter Notation
Cross-sectional area
Width of a fracture
Aquifer thickness

a
b

[0,x) /(2D )

Q
Co
CR
<C)

d
dli/dl
D*
D
D,
D
d

D
D
D
E

LM

+ (v yY/{4D D )V
x

Solute concentration
Concentration at s o m e point x and time /
Concentration at time 0
Dimensionless solute concentration (C/C )
E n s e m b l e mean concentration
Constant related to anisotropy
Characteristic flow length for Peclet n u m b e r , P
Hydraulic gradient
Effective diffusion coefficient
Molecular diffusion coefficient
Coefficient of hydrodynamic dispersion in the i direction
Coefficient of longitudinal hydrodynamic dispersion
Coefficient of longitudinal macrodispersivity at the asymptotic limit
Field-measured (calculated) coefficient of hydrodynamic dispersion
Coefficient of transverse hydrodynamic dispersion
Euler n u m b e r ( 0 . 5 7 7 . . .)
Exponential integral
0

Chapter T w o

I
F
g
G
h

It
i
1

J
k

K
f
KG

*o
L
4
h
M
N
II

n.
P
Pe
Q
r
R

Fractal dimension
Mass flux of solute per unit area per unit time
Acceleration of gravity
Topological dimension
Hydraulic head
Separation of autocorrelation function
Decay constant
Hydraulic gradient along a fracture
Constant length
Lag in autocorrelation function
Hydraulic conductivity
Equivalent hydraulic conductivity of a fracture
G e o m e t r i c mean of hydraulic conductivity
Modified Bessel function of s e c o n d kind and zero order
Straight-line distance between ends of a flowpath
Length of a tortuous flowpath
Length of a fractal flowpath
Straight line length between ends of a fractal flowpath
Total mass o f solute
N u m b e r o f units
Porosity
Effective porosity
Peclet
Peclet

number
number

(v d/D )
(v L/D )
x

Rate at which a tracer is being injected into an aquifer


Radial distance to a well
Length of well screen or o p e n b o r e h o l e

Tf
u
u

Average frontal position of water injected into a well


Autocorrelation of sampled values of Y
Standard d e r a t i o n of sampled values of Y
Variance of sampled values of Y
Time
Dimensionless time (tU/e )
Dimensionless time (v t/L)
Tortuosity
Fractal tortuosity
Average velocity of injection of water into a well
Fluctuation in the velocity vector

Covariance
<V> e n s e m b l e mean of velocity vectors

Kf
R

Sy
t
f
IR

u
U

Vx

P o r e volume
Total volume of water injected into a well
Cumulative volume of water withdrawn from a well
Velocity along a fractal flowpath
Average linear velocity in the x direction
Average fluid velocity in a fracture

M o s s T r a n s p o r t in Saturated M e d i a
V

Velocity v e c t o r

<v>

E n s e m b l e mean o f the velocity v e c t o r s

W\I,B]

Hantush leaky well function

Coordinate vector

Length of fractal flowpath

Oo,

yo)

Origin of an xy field

Straight-line distance

Residual of the d i s p l a c e m e n t of a particle

<x>

E n s e m b l e mean o f the c e n t e r o f m a s s

S e c o n d spatial m o m e n t of the solute m a s s at time / and location /', /


x,
Y
y,

Total particle d i s p l a c e m e n t

Constant related to a semivariogram

a,

Transverse dynamic dispersiviry

Aperture of a fracture

Mean o f s a m p l e values o f Y
log A:,
Dynamic dispersiviry
Longitudinal dynamic dispersiviry
Apparent dispersiviry
Correlation length for horizontal hydraulic conductivity

e,,

Correlation length for vertical hydraulic conductivity


Fractal unit of m e a s u r e m e n t
Fractal cutoff limit
Semivariogram of Y

t\
A,

Correlation length o f autocorrelation

/'

Viscosity of a fluid
Mean of population of Y

Hurst coefficient for fractal d i m e n s i o n s

CO

Coefficient related to tortuosity

Anisotropy ratio (e/e )

Density of a fluid

Pr
(Jy

Autocorrelation o f the population o f Y

Standard deviation o f population o f Y


2

Variance of population of Y
Fractional transverse position in a fracture
Autocovariance

Angle in polar c o o r d i n a t e system

Fractional vertical position in a fracture

Fourier transform wave v e c t o r n u m b e r

References
Anderson, Mars' P. 1979. "Using models to simulate the m< >ve
ment of contaminants through groundwater flow systems." Critical Reviews in Environmental Controls'),
no. 2 : 9 7 - 1 5 6 .

-. 1984, "Movement of contaminants in groundwater:


Groundwater transportadvection and dispersion."
In Groundwater Contamination, 3 7 - 4 5 . Washington,
D.C.: National Academy Press,

112

Chapter T w o

. 1990. Aquifer heterogeneitya geological perspective. Proceedings. Fifth Canadian/American Confer


ence on
Hydrogeology.
National
Water Well
Association, pp. 3 - 2 2 .
Ayra, A 1986. Dispersion and reservoir heterogeneity. Ph.D.
Diss. University of Texas, Austin.
Bakr, A A 1976. Effect of spatial variations of hydraulic conductivity on groundwater flow", Ph.D. diss. New Mexico Institute of Mining and Technology, Socorro.
Barker, J. A 1982. Laplace transform solutions for solute
transport in fissured aquifers. Advances in Water Re
sources 5, no. 2 : 9 8 - 1 0 4 .
Barry, D. A, J. Coves, and Garrison Sposito. 1988. On the
Dagan model of solute transport in groundwater: ap
plication to the Borden site. Water Resources Research
24, no. 1 0 : 1 8 0 5 - 1 7 .
Bear, J a c o b . 1961. Some experiments on dispersion. Journal
of Geophysical Research (3d, no. 8 : 2 4 5 5 - 6 7 .
. 1972. Dynamics offluids in porous media. New York:
American Elsevier Publishing Company, 764 pp.
Bear, J a c o b , and Arnold Verruijt. 1987. Modeling ground
water flow andj'dilution, Dordrecht, Netherlands: D.
Rcidel Publishing Company, 414 pp.
Berkowitz, Brian, J a c o b Bear, and Carol Braester. 1988. Con
tinuum models for contaminant transport in fractured
porous formations. Water Resources Research 24, no.
8:1225-36.
Brigham, W. E. 1974. Mixing equations in short laboratory
columns. Society of Petroleum Engineers Journal,
14:91-99.
Carman, P. C. 1937. Fluid flow through a granular bed. Trans
actions, Institute of Chemical Engineers London.
15:150-56.
Cherry, J o h n . A. 1983. Migration of contaminants in groundwater at a landfill: a case study. Journal of Hydrology
63, no. 1/2.
Crank, J. 1956. Lhe mathematics of d^fusion. New York. Ox
ford University Press.
Dagan, Gedeon. 1982. Stochastic modeling of groundwater
flow by unconditional and conditional probabilities,
2, T h e solute transport. Water Resources Research 18,
no. 4 : 8 3 5 - 4 8 .
. 1984. Solute transport in heterogeneous porous for
mations. Journal of Fluid Mechanics 1 4 5 : 1 5 1 - 7 7 .
. 1986. Statistical theory o f ground water f l o w and
transport: Pore to laboratory, laboratory to formation,
and formation to regional scale. Water Resources Re
search, 22, no. 9 I 2 0 S - 1 3 4 S .
. 1987. Theory of solute transport in water. Annual
Reviews of Fluid Mechanics, 1 9 : 1 8 3 - 2 1 5 .
. 1988. Time-dependent macrodispersion for solute
transport in anisotropic heterogeneous aquifers.
Water Resources Research 24, no. 9: 1 4 9 1 - 1 5 0 0 .
Davis, A. D. 1986. Deterministic modeling of dispersion in
heterogeneous permeable media. Ground Water 2*,
no. 5 : 6 0 9 - 1 5 .
De Josselin De Jong, G. 1958. longitudinal and transverse
diffusion in granular deposits. Transactions, Ameri
am Geof)l.yysical Union 39, no. 1:67.
Endo. H. K . J . C. S. Long, C. R. Wilson, and P. A Witherspoon.
1984. A model for investigating mechanical transport

in fracture networks. Water Resources Research 20,


no. 1 0 : 1 3 9 0 - 1 4 0 0 .
Fetter, C. W. 1988 Applied hydrogeology. 2d ed. New York:
Macmillan Publishing Company, 592 pp.
Freeze, R. Allen. 1975. A stochastic-conceptual analysis of one
dimensional groundwater flow in a nonuniform ho
mogeneous media. Water Resources Research 11, no.
5:725-41.
Freeze, R. Allen, and John A Cherry. 1979, Groundwater
Englewood Cliffs, N.J.: Prenuce Hall, 604 pp.
Freeze, R. A , J . Massmann, L Smith. T. Sperling, and B.James,
1990, Hydrogeological decision analysis: 1. A frame
work. Ground Water 28, no. 5 : 7 3 8 - 6 6 .
Fried, J. J. 1975. Groundwater Pollution Amsterdam: Elsevier
Scientific Publishing Company,
Frind, E. O., and G. E. Hokkanen. 1987. Simulation of the
Borden plume using the alternating direction Galerkin
technique. Water Resources Research 23. no. 5 : 9 1 8 30.
Garabedian, S. P., D. R. LeBlanc, L W. Gelhar, and M A Celia.
1991. Largescale natural gradient tracer test in sand
and gravel, Cape Cod, Massachusetts, 2. Analysis of
spatial moments for a nonreactive tracer. Water Re
sources Research 27, no. 5 : 9 1 1 - 2 4 .
Gelhar, L W. 1986. Stochastic subsurface hydrology from
theory to applications. Water Resources Research 22,
no. 9 : 1 3 5 S - 1 4 5 S .
Gelhar. L W., and C. L Axness. 1983. Three dimensional
stochastic analysis of macrodispersion in aquifers.
Water Resources Research 19, no. 1:161 - 8 0 .
Gelhar, L W., and M. A Collins. 1971. General analysis of
longitudinal dispersion in nonuniform flow. Water
Resources Researcijl, no. 6 : 1 5 1 1 - 2 1 .
Gelhar, L W . . A . L Gutjahr, and R L Naff. 1979. Stochastic
analysis of macrodispersion in a stratified aquifer.
Water Resources Research 15, no. 6 : 1 3 8 7 - 9 1 .
Gillham, R. W., E. A Sudicky, J. A Cherry, and E. O. Frind.
1984. An advection diffusion concept for solute trans
port in heterogeneous unconsolidated geological de
posits. Water Resources Research 20, no. 3 : 3 6 9 - 7 8 .
Grisak. G. E., and J. F. Pickens. 1980. Solute transport through
fractured media, 1. T h e effects of matrix diffusion.
Water Resources Research 1 6 , no. 4 : 7 1 9 - 3 0 .
. 1981. An analytical solution for solute transport
through fractured media with matrix diffusion. Journal of Hydrology 5 2 : 4 7 - 5 7 .
Grove, D. B., and W. A Beetem, 1971. Porosity and dispersion
constant calculations for a fractured carbonate aquifer
using the two well tracer method. Water Resources
Research!, no. 1 : 1 2 8 - 3 4 .
Hoopes, J. A, and D. R. F. Harleman. 1967. Dispersion in
radial flow from a recharge well. Journal of Geophysical Research 72, no. 1 4 : 3 5 9 5 - 3 6 0 7 .
Hull. L C, J D. Miller, and T. M. Clemo. 198T Laboratory
and simulation studies of solute transport in fracture
networks Water Resources Research 23. no. 8 : 1 5 0 5 13.
Klotz, D, K. P. Seiler, H. Moser. and F. Neumaier. 1980.
I iispershiry and velocity relationship from laboratory
and field relationships. Journal of Hydrology -IS. no,
3:169-84.

Mass Transport in Saturated Media


Konikow, L F. 1977. Modeling chloride movement in the
alluvial aquifer at the Rocky Mountain Arsenal, Colorado. US. Geological Survey Water Supply Paper
2044.
Konikow, L F., and J. D. Bredehoeft. 1978. Computer model
of two-dimensional solute transport and dispersion
in ground water. U.S. Geological Survey Techniques
of Water Resources Investigations, B o o k 7, Chapter
C2.
Konikow, L F., and D. W. T h o m p s o n . 1984. "Groundwater
contamination and aquifer restoration at the Rocky
Mountain Arsenal, Colorado." In Groundwater Contamination 9 3 - 1 0 3 . Washington D.C.: National Academy Press.
Lallemand-Barres, P., and P. Peaudecerf. 1 9 7 8 . Recherche des
relations entre la valeur de la dispersivite macroscopique d u n milieu aquifere, ses autres caractristiques
et les conditions de mesure, etude bibliographique.
Bulletin, Bureau de Recherches Gologiques et Minires. Sec. 3 / 4 : 2 7 7 - 8 7 .
LeBlanc, D. R., S. P. Garabedian, K. M. Hess, L W. Gelhard,
R. D. Quadri, K. G. Stollenwerk, and W. W. W o o d .
1991. Large-scale natural gradient tracer test in sand
and gravel, Cape Cod, Massachusetts, 1. Experimental
design and observed tracer movement. Water Re
sources Researci} 21, no. 5 : 8 9 5 - 9 1 0 .
MacFarlane, D. S, J. A. Cherry, R. W. Gillham, and E. A. Sudicky.
1983. Migration of contaminants in groundwater at a
landfill: A case study, 1. Groundwater flow and plume
definition. Journal of Hydrology 6 3 , no. 1 / 2 : 1 - 3 0 .
Mackay, D. M., D. L Freyberg, P. V. Roberts, and J. A. Cherry.
1986. A natural gradient experiment on solute transport in a sand aquifer, 1. Approach and overview of
plume movement. Water Resources Research 22, no.
13:2017-29.
Mandelbrot, B. B. 1967. How long is the coastline of Great
Britain? Statistical s e l f similarity and fractional dimen
sion. Science 1 5 5 : 6 3 6 - 3 8 .
. 1983. The fractal geometry of nature. New York: W.
H.
Freeman.
Matheron, G., and G. de Marsily. 1 9 8 0 . Is transport in porous
media always diffusive? A counterexample. Water Resources Research 1 6 , no. 5 : 9 0 1 - 1 7 .
Molz, Fred J . , Oktay Guven, and J o e l G. Melville. 1 9 8 3 . An
examination of scale-dependent dispersion coefficients. Ground Water21, no. 6 : 7 1 5 - 2 5 .
Moreno, L, Y. W. Tsang, C. F. Tsang, F. V. Hale, and I. Neretnieks. 1988. Flow and tracer transport in a single
fracture: A stochastic model and its relation to s o m e
field observations. Water Resources Research 24, no.
12:2033-48.
Neretnieks, I. 1980. Diffusion in the rock matrix: An important
factor in radionuclide migration. Journal of Geophysical Research 8 5 , no. B 8 : 4 3 7 9 - 9 7 .
Neuman, S. P. 1990. Universal scaling of hydraulic conductivities and dispersivities in geologic m e d i a Water
Resources Research 26, no. 8 : 1 7 4 9 - 5 8 .
Neuman, S. P., C. L Winter, and C. N. Newman. 1987. Stochastic theory of field-scale Fickian dispersion in an
isotropic porous media. Water Resources Research
2 3 , no. 3 : 4 5 3 - 6 6 .

113

Neuman, S. P., and Y.-K. Zhang. 1990. A quasi-linear theory


of non-Fickian and Fickian subsurface dispersion, 1.
Theoretical analysis with application to isotropic media. Water Resources Research 26, no. 5 : 8 8 7 - 9 0 2 .
Ogata, Akio. 1970. Theory of dispersion in a granular medium. U.S. Geological Survey Professional Paper
411-1.
Ogata, Akio, and R. B. Banks. 1 9 6 1 . A solution of the differential equation of longitudinal dispersion in porous
media. U.S. Geological Survey Professional Paper
411-A.
Perkins, T. K., and O. C. J o h n s o n . 1963. A review of diffusion
and dispersion in porous media. Society of Petroleum
Engineers Journal
3:70-84.
Pickens, J. F., R. E . J a c k s o n , K . J . Inch, and W. F. Merritt. 1 9 8 1 .
Measurement of distribution coefficients using a radial
injection duel tracer test. Water Resources Research
17, no. 3 : 5 2 9 - 4 4 .
Pickens, J o h n F., and Gerald E. Grisak. 1 9 8 1 . Scale dependent
dispersion in a stratified granular aquifer. Water Re
sources Research 17, no. 4 : 1 1 9 1 - 1 2 1 1 .
Pinder, George. 1973. A Galerkin-finite element simulation of
groundwater contamination on Long Island, New
York, Water Resources Research 9. no. 6 : 1 6 5 7 - 6 9 Rasmuson, A 1984. Migration of radionuclides in fissured
rock: Analytical solutions for the case of constant
source strength. Water Resources Research 20, no.
10:1435^*2.
Rasmuson, A., and I. Neretnieks. 1 9 8 1 . Migration of radio
nuclides in fissured rock: T h e influence of micropore
diffusion and longitudinal dispersion. Journal of Geo
physical Researci) 8 6 , no. B 5 : 3 7 4 9 - 5 8 .
Raven, K. G., K. S. Novakowski, and P. A. Lapcevic. 1988.
Interpretation of field tests of a single fracture using
a transient solute storage model. Water Resources
Research 24, no. 1 2 : 2 0 1 9 - 3 2 .
Robinson, R. A, and R. H. Stokes. 1965. Electrolyte solutions.
London: Butierworth Press.
Saury, Jean-Pierre. 1978. Identification des paramtres du
transport hydrodispersif das les aquiferes par inter
pretation de traages en coulement cyclindrique convergent ou divergent. Journal of Hydrology' 3 9 , no.
3/4:69-103.
. 1980. An analysis of hydrodispersive transfer in aquifers. Water Resources Research 1 6 , no. 1 : 1 4 5 - 5 8 .
Schwartz, F. W., and L Smith. 1988. A continuum approach
for modeling mass transport in fractured media. Water
Resources Research 24, no. 8 : 1 3 6 0 - 7 2 .
Schwartz, F. W., L Smith, and A. S. Crowe. 1983. A stochastic
analysis of macroscopic dispersion in fractured media.
Water Resources Research 19, no. 5 : 1 2 5 3 - 6 5 .
Smith, L, and F. W. Schwartz. 1984. An analysis of the influence
of fracture geometry on mass transport in fractured
media. Water Resources Researci} 2 0 , no. 9 : 1 2 4 1 - 5 2 .
Sposito, Garrison, and D. A. Barry. 1987. On the Dagan model
of solute transport in groundwater: foundational as
pects. Water Resources Research 23, no. 1 0 : 1 8 6 7 - 7 5 .
Sposito, Garrison, W. A. Jury, and V. K. Gupta. 1986. Fun
damental problems in the stochastic convection-dispersion model of solute transport in aquifers and field
soils. Water Resources Research 22, no. 1 : 7 7 - 8 8 .

1 14

Chapter T w o

Sudicky, F. A. 1986. A natural gradient experiment on solute


transport in a sand aquifer: Spatial variability of hydraulic conductivity and its role in the dispersion pro-

cess. Water Resources Research 22, no. 13:2069-82.


Sudicky, E. A., and J. A Cherry. 1979. Field observations of
tracer dispersion under natural flow conditions in an
unconfined sandy aquifer. Water Pollution Researcfy,
Canada 1 4 : 1 - 1 7 .
Sudicky, E. A., J. A. Cherry, and E. O. Frind. 1983. Migration
of contaminants in groundwater at a landfill: A case
study, 4, A natural gradient dispersion test. Journal
of Hydrology 6 3 , no. 1 / 2 : 8 1 - 1 0 8 .
Sudicky, F. A., and F. O. Frind. 1982. Contaminant transport
in fractured porous media: Analytical solution for a
system of parallel fractures Water Resources Research
18, no. 6 : 1 6 3 4 ^ 2 .
.1984. Contaminant transport in fractured porous media: Analytical solution for a two-member decay chain
in a single fracture. Water Resources Research 20, no.
7:1021-29.
Tang, D. H., E. O. Frind, and E. A Sudicky. 1981. Contaminant
transport in fractured porous media: Analytical solu-

tion for a single fracture. Water Resources Research?


17, no. 3 : 5 5 5 - 6 4 .
Tsang, Y. W., C. F. Tsanf, I. Neretnieks, and L Moreno. 1988.
Flow and tracer transport in fractured media: A van
able aperture channel model and its properties. Water
Resources Research 24, no. 1 2 : 2 0 4 9 - 6 0 .
Van Genuchten, M. Th. 1981, Analytical solutions for chemical
transport with simultaneous adsorption, zero-order
pnxluciion, and first-order decay. Journal of Hy
drotogy . 4 9 : 2 1 3 - 2 3 3 .
Wheatcraft, S. W., and S. W. Tyler. 1988. An explanation of
scale-dependent dispersrvity in heterogeneous aquifers using concepts of fractal geometry. Water Resources Research 24, no. 4 : 5 6 6 - 7 8 .
Witherspoon, P. A, J. S. Y. Yang, K. Iwai, and J. E. Gale. 1980.
Validity of the cubic law for fluid flow in a deformable
rock fracture. Water Resources Research 16, no.
6:1016-24.
Zhang, Y.-K., and S. P. Neuman. 1990. A quasi linear theory
of non Fickian and Fickian subsurface dispersion, 2.
Application to anisotropic media and the Borden site.
Water Resources Research 26, no. 5 : 9 0 3 - 1 3 .

Chapter Three

Transformation, Retardation,
and Attenuation of Solutes

3.1

Introduction
Solutes dissolved in g r o u n d water are s u b j e c t to a n u m b e r of different p r o c e s s e s through
which they can be removed from the g r o u n d water. T h e y can be s o r b e d o n t o the surfaces
of the mineral grains of the aquifer, s o r b e d by organic c a r b o n that might be present in
the aquifer, undergo chemical precipitation, be s u b j e c t e d to abiotic as well as biodgradation, and participate in oxidation reduction reactions. Furthermore, radioactive
c o m p o u n d s can decay. As a result of sorption p r o c e s s e s , s o m e solutes will move much
m o r e slowly through the aquifer than the g r o u n d water that is transporting them; this
effect is called r e t a r d a t i o n . Biodgradation, radioactive decay, and precipitation will
d e c r e a s e the c o n c e n t r a t i o n of solute in the plume but may not necessarily slow the rate
o f plume movement.
Equation 2.18, the one-dimensional advection-dispersion equation, can be modified
to include sorption and decay. This can be e x p r e s s e d as (Miller and W e b e r 1 9 8 4 ) :
2

C
Tt~

dC
DL

CT>

C
v

(dispersion)

'Tx

BidC
~

(advection)

TIT

(6C\
+

(sorption)

(j-tL

,
(

(reaction)

where
C = c o n c e n t r a t i o n of solute in liquid p h a s e
t= time
D = longitudinal dispersion coefficient
L

v = average linear groundwater velocity


x

B = bulk density o f aquifer


d

8 = volumetric moisture c o n t e n t or porosity for saturated media


C* = a m o u n t of solute s o r b e d per unit weight of solid
rxn = subscript indicating a biological or chemical reaction of the solute
( o t h e r than s o r p t i o n )

115

116

Chapter T h r e e

T h e first term on the right side of Equation 31 represents the dispersion of the
solute, the s e c o n d term is the advection of the solute, the third term is the transfer of
the solute from the liquid phase to the solid particles by sorption, and the last term
simply indicates that there may be a c h a n g e in concentration of the solute with time
due to biological or chemical reactions or radioactive decay.

3.2

Classification of Chemical Reactions


Rubin ( 1 9 8 3 ) listed six different classes of chemical reactions that can o c c u r in solute
transport (Figure 3 . 1 ) . At the highest, or A, level reactions are either ( 1 ) "sufficiently
fast" and reversible or ( 2 ) "insufficiently fast" a n d / o r irreversible. Sufficiently fast reactions
are reversible reactions that are fast relative to ground-water flow rates and are faster
than any o t h e r reactions that act to change solute concentration. With these reactions
o n e can a s s u m e that locally the solute is in chemical equilibrium with the surroundings
(local equilibrium assumption, or LEA, system). If the reaction is not sufficiently fast for
local equilibrium to develop or if it is irreversible, then it falls into the s e c o n d major
grouping.
At the s e c o n d , or B, level reactions are either ( 1 ) h o m o g e n e o u s or ( 2 ) heterog e n e o u s . H o m o g e n e o u s reactions take place within a single phase, the dissolved phase,
whereas h e t e r o g e n e o u s reactions involve both the dissolved phase and the solid phase.
Level C reactions, representing the greatest specification, apply only to heterogeneous

Level B

Homogeneous

F I G U R E 3.1

Heterogeneous

Classification of chemical reactions useful in solute transport analyses. Source: J, Rubin,

Wafer Resources Research 19, no. 5 (1 9 8 3 ) : 1 2 3 1 - 5 2 . Published by the American Geophysical Union.

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

117

reactions. T h e s e can b e either ( 1 ) surface reactions, such a s h y d r o p h o b i c adsorption


o f neutral organic c o m p o u n d s and ion e x c h a n g e o f charged ions, o r ( 2 ) classical chemical
reactions such as precipitation and dissolution.

3.3

Sorption Processes
S o t p t i o n p r o c e s s e s include adsorption, c h e m i s o r p t i o n , absorption, and ion e x c h a n g e .
A d s o r p t i o n includes the p r o c e s s e s by which a solute clings to a solid surface. Cations
may be attracted to the region c l o s e to a negatively c h a r g e d clay-mineral surface and
held there b y electrostatic forces; this p r o c e s s i s called c a t i o n e x c h a n g e . Anion
e x c h a n g e c a n o c c u r at positively charged sites on iron and aluminum o x i d e s and the
b r o k e n e d g e s o f clay minerals. C h e m i s o r p t i o n o c c u r s w h e n the solute i s i n c o r p o r a t e d
on a sediment, soil, or rock surface by a c h e m i c a l reaction. A b s o r p t i o n o c c u r s w h e n
the aquifer particles are p o r o u s so that the solute can diffuse into the particle and be
s o r b e d o n t o interior surfaces ( W o o d , Kramer, and Hern 1 9 9 0 ) .
In this c h a p t e r we will n o t attempt to separate these p h e n o m e n a but will simply
u s e the term sorption to indicate the overall result of the various p r o c e s s e s . F r o m a
practical view the important aspect is the removal of the solute from solution, irrespective
of the p r o c e s s . T h e p r o c e s s by which a contaminant, which was originally in solution,
b e c o m e s distributed b e t w e e e n the solution and the solid p h a s e is called p a r t i t i o n i n g .
Sorption is d e t e r m i n e d experimentally by measuring h o w m u c h of a solute can
be s o r b e d by a particular sediment, soil, or r o c k type. Aliquots of the solute in varying
c o n c e n t r a t i o n s are well m i x e d with the solid, and the a m o u n t of solute removed is
determined. T h e capacity of a solid to remove a solute is a function of the c o n c e n t r a t i o n
of the solute. T h e results of the e x p e r i m e n t are plotted on a graph that s h o w s the solute
c o n c e n t r a t i o n versus the a m o u n t s o r b e d o n t o the solid. If the sorptive p r o c e s s is rapid
c o m p a r e d with the flow velocity, the solute will reach an equilibrium condition with the
s o r b e d phase. This p r o c e s s can b e d e s c r i b e d b y a n e q u i l i b r i u m s o r p t i o n i s o t h e r m .
It is an e x a m p l e of a sufficiently fast, h e t e r o g e n e o u s surface reaction. If the sorptive
p r o c e s s is s l o w c o m p a r e d with the rate of fluid flow in the p o r o u s media, the solute
may n o t c o m e t o equilibrium with the s o r b e d phase, and a k i n e t i c s o r p t i o n m o d e l
will be n e e d e d to d e s c r i b e the p r o c e s s . T h e s e are insufficiently fast, h e t e r o g e n e o u s
surface reactions. Travis and Etnier ( 1 9 8 1 ) give a c o m p r e h e n s i v e review of sorption
isotherms and kinetic models.

3.4

Equilibrium Surface Reactions


3.4.1

Linear Sorption Isotherm

If there is a direct, linear relationship b e t w e e n the a m o u n t of a solute s o r b e d o n t o solid,


C*, and the concentration of the solute, Q the adsorption isotherm of C as a function
of C . w i l l plot as a straight line on graph paper (Figure 3 . 2 ) . T h e resulting linear sorption
isotherm is d e s c r i b e d by the equation
C*

KC
D

(3.2)

118

Chapter T h r e e

where
C* = mass of solute s o r b e d per dry unit weight of solid ( m g / k g )
C= concentration of solute in solution in equilibrium with the mass of
solute s o r b e d o n t o the solid ( m g / L )
K = coefficient ( L / k g )
t

T h e coefficient K is known as the d i s t r i b u t i o n c o e f f i c i e n t . It is equal to the


s l o p e of the linear sorption isotherm.
T h e linear sorption isotherm is very appealing from the standpoint of mathematical
manipulation. If Equation 3.2 is substituted into Equation 3.1, the resulting advection
dispersion equation is
d

dC

Tt

SC

= DL

dC
v

B dUCjC)

- - 'T -J-dr
x

....
( 3 3 )

119

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

T h i s can be reorganized as
2

8C

d_C

8x

8x

(3.4)

What has b e e n t e r m e d the r e t a r d a t i o n f a c t o r , r is given by


fl

(3.5)

+ ^

If the average linear ground-water velocity is v , the average velocity of the solute
x

front w h e r e the c o n c e n t r a t i o n is o n e - h a l f of the original, v , is given by


e

(3.6)
Equations 3.3 through 3.6 are c o n v e n i e n t to solve mathematically and have b e e n
used in a n u m b e r of studies to predict the rate of m o v e m e n t of a solute front (Anderson
1 9 7 9 ; Faust and M e r c e r 1 9 8 0 ; Prickett, Naymik, and Lonnquist 1 9 8 1 ; Srinivasan and M e r c e r
1988).
T h e r e are two limitations of the linear-sorption isotherm model. O n e is that it d o e s
not limit the amount of solute that can be s o r b e d o n t o the solid. T h i s is clearly n o t the
case; there must be an u p p e r limit to the mass of solute that can be s o r b e d . In addition,
if there are onfy a few data points, what is actually a curvilinear e x p e r i m e n t a l plot of C
versus C* might be misinterpreted to be a linear relationship. Figure 3.3 illustrates h o w
it is important never to extrapolate from a limited data set to a range outside the data
set and a s s u m e that a linear relationship exists in that region. T h e s u b s e t of the sorption
data on Figure 3.3 marked by triangles can be used with the origin to form a linear
relationship. T h e subset of the sorption data marked with squares can also be used with
the origin to create a different linear relationship. However, if all the data are included,
o n e can s e e that the sorption isotherm is not linear at all.
3.4.2

Freundlich Sorption Isotherm

A m o r e general equilibrium isotherm i s the F r e u n d l i c h s o r p t i o n i s o t h e r m . T h i s i s


defined by the nonlinear relationship
C* = KC

(3.7)

w h e r e K and N are constants.


If the sorption characteristics can be d e s c r i b e d by a Freundlich sorption isotherm,
w h e n C is plotted as a function of C* on graph paper the data will be curvilinear ( F i g u r e
3 . 4 ( a ) ) . However, the data can be linearized by use of the following equation:
log C* = log K + N log C

(3.8)

If log C is plotted against log C on graph paper, the result will be linear with a
s l o p e of A ' a n d an intercept of log K. This is illustrated in Figure 3 . 4 ( b ) .

120

Chapter

Three

C* -

F I G U R E 3.3

Nonlinear sorption isotherms can be misinterpreted as linear sorption isotherms if a small

data set is extrapolated out of its range. The subset of the data represented by triangles can be interpreted
as a linear sorption isotherm, as can the data subset consisting of squares. However, if the complete data
set, which includes the triangles, circles, and squares, is used, it can be seen that the isotherm is nonlinear.

If Equation 37 is substituted into Equation 3 . 1 , the result is

dC

dC

dC

B d(KC )

(3 9)

H - ^ W M r i

After differentiation and reorganization, Equation 3.9 b e c o m e s


N

dC(

B KNC ~ \

dC

1+

bC

(3 10)

;-

c
(a)

F I G U R E 3.4

log C
<b)

(a) Nonlinear Freundlich sorption isotherm with C" versus C plotted on cross-section paper,

(b) The Freundlich sorption isotherm can be made linear by plotting log C* versus C.
121

122

Chapter T h r e e

T h e retardation factor for a Freundlich sorption isotherm, r , is


f!

B KNC ~
i

1+

=>/

(3.11)

If A is greater than 1, Equation 3.10 will lead to a spreading front, whereas if A is


less than 1, the front will be self-sharpening. If A'is equal to 1, the Freundlich sorption
isotherm b e c o m e s the linear sorption isotherm.
T h e Freundlich sorption isotherm is o n e that has b e e n widely applied to the
sorption by soils of various metals and organic c o m p o u n d s such as sulfate ( B o r n e m j s z a
and Llanos 1 9 6 7 ) , cadmium (Street Lindsay, and Sabey 1 9 7 7 ) , c o p p e r and zinc (Sidle,
Kardos, and van G e n u c h t e n 1 9 7 7 ) , molybdenum (Jarrell and Dawson 1 9 7 8 ) , organophosphorous pesticides (Yaron, 1 9 7 8 ) , p-chloroanaline residues (Van Bladel and Moreale
1 9 7 7 ) , and parathion and related c o m p o u n d s ( B o w m a n and Sans 1 9 7 7 ) . T h e Freundlich
sorption isotherm suffers from the s a m e fundamental p r o b l e m as the linear sorption
isotherm; there is theoretically no upper limit to the amount of a solute that could be
sorbed. O n e should be careful not to extrapolate the equation beyond the limits of the
experimental data. T h e Freundlich sorption isotherm is usually obtained by an empirical
fit to experimental data.
3.4.3

Langmuir Sorption Isotherm

T h e L a n g m u i r s o r p t i o n i s o t h e r m was developed with the c o n c e p t that a solid


surface p o s s e s s e s a finite n u m b e r of sorption sites. W h e n all the sorption sites are filled,
the surface will no longer s o r b solute from solution. T h e form of the Langmuir sorption
isotherm is

i * ^

312

< >

where
a = an absorption constant related to the binding energy ( L / m g )
= the maximum amount of solute that can be a b s o r b e d bv the solid
(mgAg)
T h e Langmuir sorption isotherm can also be e x p r e s s e d as

W h e n Equation 3 . 1 3 is substituted into Equation 3 . 1 , the following equation is


obtained:

sc

sc

sc

n ' ^ - ^ T x - i a

123

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

I >i11L ix 1111.111< IN .mil reorganization nl Equation S 1 i yields

*P

Be (

0 \ ( \ +

dC

dC

T h e retardation factor for the Langmuir sorption isotherm, r

Bit

1 + ^e

(3.15)

aC)\

*fi
\(i + z o

= r,

f h

is

(3.16)

If the sorption of a solute o n t o a solid surface follows a Langmuir sorption isotherm,


w h e n e x p e r i m e n t a l data of C versus C are plotted on graph p a p e r they will have a
curved s h a p e that r e a c h e s a m a x i m u m value ( F i g u r e 3 . 5 ( a ) ) . If C/C* is plotted versus
C o n graph paper, the data will follow a straight line. T h e m a x i m u m ion sorption, /?, is
the reciprocal of the s l o p e of the line, and the binding energy constant, a, is the s l o p e
o f the line divided b y the intercept ( F i g u r e 3 5 ( b ) ) .
In studies of the sorption of p h o s p h o r o u s on soils, it has b e e n found that a plot
of C/C* versus C will yield curves with two straight line s e g m e n t s ( F e t t e r 1 9 7 7 ; Munns
and F o x 1 9 7 6 ) . T h i s has b e e n interpreted t o m e a n that there are two types o f s o r p t i o n
sites which differ i n their b o n d i n g energy. T h e L a n g m u i r t w o - s u r f a c e s o r p t i o n
i s o t h e r m is
C*

a /?
2

1 + a,C

1 + a C

(3.17)

where
or, = the b o n d i n g strength at the type 1 sites
a, = the b o n d i n g strength at the type 2 sites
/ ? , = the m a x i m u m a m o u n t of solute that can be s o r b e d at the type 1 sites
f} = the m a x i m u m a m o u n t of solute that c a n be s o r b e d at the type 2 sites
2

3.4.4

Effect o f E q u i l i b r i u m R e t a r d a t i o n o n S o l u t e T r a n s p o r t

T h e effects o f equilibrium retardation can b e illustrated through u s e o f a c o m p u t e r


m o d e l , B I O I D . T h i s m o d e l was d e v e l o p e d b y Srinivasan and M e r c e r ( 1 9 8 8 ) and simulates
b o t h sorption p r o c e s s e s and biodgradation in mass transport. It is very flexible and
can simulate linear. Freundlich. and Langmuir adsorption as well as a e r o b i c and a n a e r o b i c
biodgradation.
T h e situation b e i n g m o d e l e d is o n e - d i m e n s i o n a l m a s s transport through a saturated
p o r o u s medium that is in a c o l u m n 16 cm long. T h e p o r e water velocity is 0.1 c m / s e c ,
2

the dispersion coefficient is 0.1 c m / s e c , and the porosity is 0.37. T h e initial solute
c o n c e n t r a t i o n is 0.0 mg I

For 2 nun a solute with a c o n c e n t r a t i o n ol 0.05 mg I is

injected into the top of the soil c o l u m n and allowed to drain from the b o t t o m . After 2

c*

F I G U R E 3.5

(a) Nonlinear Langmuir sorption isotherm will reach a maximum sorption value when C* is

plotted versus C. (b) The Langmuir sorption isotherm can be made linear by plotting C/C* versus C
124

Transformation, Retardation, and Attenuation of Solutes

125

Description

Distance ( c m )

BIOID
Version 1.2
GeoTrans. Inc.

Case 1 , none

Case 2, linear

No retardation versus linear retardation

3.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

Pore volume

FIGURE 3.6

Illustration of the effect of retardation by comparing the breakthrough curve of a solute

which isn't retarded with the breakthrough curve of a solute that undergoes linear-type retardation. Model
simulation using B I O I D from Geotrans, Inc.

min the c o n c e n t r a t i o n of the solute in the water entering the c o l u m n is set b a c k to 0 . 0 0 .


T h e model yields the solute c o n c e n t r a t i o n in the water draining from the soil c o l u m n
as a function of the n u m b e r of p o r e volumes that have b e e n drained.
Figure 3.6 shows the general effect of retardation. O n e of the two curves, the solid
o n e , is the solute breakthrough curve with no retardation ( a n d no degradation). T h e
dashed curve s h o w s the breakthrough of a solute that is undergoing retardation, which
follows a linear sorption isotherm, and has a K value of 0 . 4 7 6 /tg/g. It can be s e e n that
d

the retarded s u b s t a n c e ( d a s h e d c u r v e ) has a lower peak value and that the peak c o m e s
later; i.e., it takes m o r e p o r e volumes for it to o c c u r than the unretarded peak (solid
line).
Figure 3.7 illustrates the effect of different N values on the Freundlich sorption
isotherm. T h e model is simulating exactly the s a m e situation as before, e x c e p t that there
is Freundlich-type retardation. T h e s a m e K value is used, with the solid line illustrating
d

an N value of 1.3 and the dashed curve representing an N value of 0.7. ( T h e linear

1 26

Chapter T h r e e

Dislance (cm)

Description

mom
Case 1, N = 1.3
Case 2. N = 0.7

Version 1.2
GeoTrans. Inc.

Freundlich isotherm retardation

Pore volume

F I G U R E 3.7

Illustration of the effect of the value of the constant N in the Freundlich sorption isotherm.

The solid curve has an N greater than 1, whereas the dashed curve has an N less than 1. Model simulation
using B I O I D from Geotrans, Inc.

sorption isotherm is a special c a s e of the Freundlich sorption isotherm with an N value


of 1.0.) Figure 3-7 s h o w s that with an TV value greater than 1, the breakthrough curve
arrives earlier (i.e., takes fewer p o r e v o l u m e s ) and has a greater peak value than the
breakthrough curve with an N value less than 1.
Caution should be used if experimental absorption studies indicate an N value
greater than 1.0 for a Freundlich sorption isotherm. T h e r e is no theoretical reason why
the exponential constant should be greater than the linear value of 1.0. S o m e researchers
believe that N values greater than 1.0 are a result of a combination of sorption and
precipitation that is occurring b e c a u s e the experimental concentrations are e x c e e d i n g
the water solubility of the c o m p o u n d (Griffin 1 9 9 1 ) .
Figure 3.8 c o m p a r e s the linear sorption isotherm with a Langmuir sorption isotherm.
T h e Langmuir sorption isotherm has a maximum binding energy of 0.3-t5 /tg g and a
maximum sorption of 0 . 4 7 5 Mg/g- T h e Langmuir sorption isotherm results in a higher
peak value at breakthrough, which arrives at an earlier time than the linear sorption
isotherm. In this particular c a s e the Langmuir isotherm is not very different than the
linear sorption isotherm.

127

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

Description

Distance (cm)

BIOID
Version 1.2
GeoTrans. Inc.

Case 1. linear
Case 2, langmuir

Linear versus Langmuir isotherms

0.03

I m 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 T i T n T I 111111111
0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

Pore volume

F I G U R E 3.8

Illustration of the effect of different sorption isotherms in modeling solute transport. The

solid curve is for a linear sorption isotherm while the dashed curve is for a Langmuir sorption isotherm.
Model simulation using B I O I D from Geotrans. Inc.

EXAMPLE
PROBLEM

Sorption o f p h o s p h o r o u s b y a c a l c a r e o u s glacial outwash was studied by m e a n s o f a


batch sorption test. T h e outwash was air-dried and then sieved to segregate the fraction
that was finer than 2 mm. T h e c o a r s e r material was discarded. Ten-gram samples of the
sediment w e r e added to flasks containing 1 0 0 mL of 0.1 M NaCl and disodium p h o s p h a t e
in c o n c e n t r a t i o n s ranging from 0.53 to 12.1 m g / L . T h e flasks w e r e shaken for 4 da on
an autoshaker. T h e samples w e r e then filtered and the filtrate analyzed for o r t h o p h o s phate. T h e sediment was e x t r a c t e d with dilute HC1 and the extract was analyzed to
determine the amount of p h o s p h o r o u s s o r b e d to the sediment prior to the test. T h i s
a m o u n t was 0 . 0 1 6 m g / g .
T h e initial concentration of p h o s p h o r o u s in solution was known and the equilibrium concentration was determined by analysis. By knowing the v o l u m e of solution and
the initial concentration, the mass of p h o s p h o r o u s c o u l d be c o m p u t e d . F o r e x a m p l e a
100-mL sample with a c o n c e n t r a t i o n of 3.85 m g / L has 0 . 3 8 5 mg of P. At equilibrium
with the sediment, this aliquot had 2.45 m g / L of P, or 0 . 2 4 5 mg, still in solution. T h e
amount s o r b e d was 0.14 m g ( 0 . 3 8 5 m g 0 . 2 t 5 m g ) , o r 0 . 0 1 4 m g / g o f sediment. Prior
to the sorption test the sediment had b e e n e x t r a c t e d with dilute HC1 and the extract
tested for P. It was found to contain 0 . 0 1 6 m g / g of P. T h i s a m o u n t already o c c u p i e d

128

Chapter T h r e e

s o m e of the sorption sites and had to be added to the amount s o r b e d during the test.
T h e following table lists the initial and equilibrium concentrations for P, the amount
s o r b e d o n t o the soil, and the value of C/C*. It is interesting to n o t e that for the lowest
initial concentration, the equilibrium concentration is greater than the initial concentration. This is d u e to P desorbing from the sediment.
Amount

Amount
Sorbed in
0.016 mg/g

(mg/L)

(mg/g)

Initial

Equilibrium

Sorbed
per Gram

Concentration

Concentration

during Test

(mg/L)

C(mg/L)

(mg/g)

0.53
1.95

0.55

-0.002

1.25

0.007

3.85

2.45

6.05
8.0
12.1

Test Plus
(mg/g)

C/C

0.014

39

0.014

0.023
0.030

54.5
81

3.85
5.00

0.022

0.038

0.030

0.046

103
108.5

7.70

0.044

0.060

127.5

20 -

F I G U R E 3.9

Ol

4
C (mg/L)

A linear Langmuir two-surface sorption isotherm for the sorption of phosphate on calcareous

glacial outwash. Source: C. W. Fetter, Ground Water 15, no. 5 (1977): 3 6 5 - 7 1 . Used with permission.
Copyright 1977 Water W e l l Journal Publishing Co.

129

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

Figure 3.9 s h o w s the plot of C/C* versus C. T h i s is clearly a Langmuir two surface
sorption isotherm. T h e sorption m a x i m a for low c o n c e n t r a t i o n s is 0.05 mg P per gram
of sediment, and for the higher c o n c e n t r a t i o n s it is 0 . 1 6 mg P per gram of sediment.

3.5

Nonequilibrium (Kinetic) Sorption Models


All the equilibrium m o d e l s a s s u m e that the rate of c h a n g e in c o n c e n t r a t i o n d u e to
sorption is much greater than the c h a n g e due to any o t h e r cause and that the flow rate
is l o w e n o u g h that equilibrium can be reached. If this is not the c a s e and equilibrium
is not attained, a kinetic m o d e l is m o r e appropriate. In a kinetic m o d e l the solute transport
equation is linked to an appropriate equation to d e s c r i b e the rate that the solute is
s o r b e d o n t o the solid surface and d e s o r b e d from the surface.
T h e m o s t simple nonequilibrium condition is that the rate of sorption is a function
of the c o n c e n t r a t i o n of the solute remaining in solution and that o n c e s o r b e d o n t o the
solid, the solute c a n n o t be d e s o r b e d . T h i s is an irreversible reaction and the p r o c e s s
leads to attenuation of the solute ( n o t retardation which by definition is reversible). T h e
i r r e v e r s i b l e f i r s t - o r d e r k i n e t i c s o r p t i o n m o d e l that d e s c r i b e s this consists o f the
following pair of equations:
SC*

dt

k-^C
2

bC
-T =

bC

bC
v

dx

dt

where

(3.18)

dx

B bC

- -

9 dt

(3.19)

= a first-order decay rate constant.

If the rate of solute sorption is related to the a m o u n t that has already b e e n s o r b e d


and the reaction i s reversible, then the r e v e r s i b l e l i n e a r k i n e t i c s o r p t i o n m o d e l
can be used. T h i s c o n s i s t s of Equation 3 . 1 9 and the following e x p r e s s i o n for the rate
o f sorption:
dC*

= k C-k C*
dt
2

(3.20)

where
k = forward rate constant
2

k = backward rate constant


3

If sufficient time is available for the system to reach equilibrium, then there is no
further c h a n g e in

C* with time and

dC*/dt= 0, so that

k C= k C*. T h i s c a n be
2

rearranged to C* = (k /k )C, which is a linear equilibrium sorption isotherm.


2

Equation 3.20 is s o m e t i m e s written in a slightly different form (Nielsen, van Genuchten, and Biggar 1 9 8 6 ) :
dC*

= y(k C-C)

(3.21)

where
y = a first-order rate coefficient
rfe = a constant equivalent to K
4

130

Chapter T h r e e

Equation 3.21 d e s c r i b e s a situation where reversible linear sorption is limited by a firstorder diffusion process.
This model has b e e n used to d e s c r i b e the sorption of pesticides (Leistra and
Dekkers 1977; H o r n s b y and Davidson 1 9 7 3 ) as well as s o m e organics (Davidson and
Chang 1 9 7 2 ) .
A third kinetic model i s the r e v e r s i b l e n o n l i n e a r k i n e t i c s o r p t i o n m o d e l .
T h i s couples Equation 3.19 with
8C*
= k C - k C*
5

(3.22)

w h e r e k$, k , and Nare constants. This model describes a situation where the forward
6

( s o r p t i o n ) reaction is nonlinear, whereas the backward ( d e s o r p t i o n ) reaction is linear.


This equation has b e e n used, with a value of N less than 1, to describe the sorption of
P ( F i s k e I l et al. 1 9 7 9 ) and herbicides (Enfield and B l e d s o e 1 9 7 5 ) .
At the equilibrium condition for the reversible nonlinear model, dC*ldi = 0 and
N

k C = k C*, which can be rearranged as


5

C* = (.k /k )C , which is the Freundlich


s

sorption isotherm.
The b i l i n e a r a d s o r p t i o n m o d e l is tire kinetic version of the Langmuir sorption
isotherm. This m o d e l has the form
dC*
= fc C(0 - C) - k C*
7

(3.23)

ol

where
fi = the maximum amount of solute that can be s o r b e d
k-i = the forward rate constant
k = the backward rate constant
a

In s o m e cases the sorption of ions may be controlled by the rate at which the
ions are transported to the e x c h a n g e sites by diffusion, even though the sorption may
be instantaneous o n c e the ions reach the sorption or e x c h a n g e sites. In this situation a
d i f f u s i o n - c o n t r o l l e d r a t e l a w must b e employed (Nkedi-Kizza and e t al. 1 9 8 4 ) . T h e
liquid has a m o b i l e phase, through which advective flow occurs, and an immobile phase
near the solid surfaces. Transfer of solutes across the immobile water to the solid surfaces
o c c u r s by diffusion. T h e rate of solute transfer across the immobile water is assumed
to be proportional to the difference in concentration between the two regions. T h e
equations that are given are applicable to both saturated and unsaturated flow. F o r
unsaturated flow, Q is the volumetric water content and for saturated flow, 6 is the
porosity. T h e equations can a c c o u n t for a system w h e r e s o m e of the solid is in direct
contact with the m o b i l e phase and s o m e is in direct contact with the immobile phase.
This system requires a pair of equations:

8C

dC

dC

dC*

131

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

(1 -f)B ~^ = x{C - C ) - e
d

im

(3.25)

im

where
9

= porosity o c c u p i e d by m o b i l e p h a s e
= porosity o c c u p i e d by i m m o b i l e phase

lm

C = solute c o n c e n t r a t i o n in the m o b i l e phase


Q

= solute concentration in the i m m o b i l e phase

Q* = a b s o r b e d c o n c e n t r a t i o n in c o n t a c t with the m o b i l e phase


CJ* = a b s o r b e d c o n c e n t r a t i o n in c o n t a c t with the i m m o b i l e phase
v = velocity of the m o b i l e p h a s e
m

/ = fraction of the solid surfaces in c o n t a c t with the m o b i l e phase


D = apparent diffusion c o n s t a n t for m o b i l e phase
m

T = first-order, mass transfer coefficient


If the sorption of the solute is in equilibrium and reversible and follows a linear
sorption isotherm, then

Ql

K C
d

and

C* = fGQ
m

(3.26)

T h e total sorption of solute from b o t h the m o b i l e and i m m o b i l e regions is

C*=/C

+ (l-/)Ci;

(3-27)

With these sorption conditions, Equations 3 . 2 4 and 3.25 can be written as

(0

+ BJK) = 0 D
m

- 0V

|* -

[6

im

+ (1

-f)B K \
d

-jf

(3.28)
(

An analytical solution to Equations 3 . 2 8 and 3.29 is available (van G e n u c h t e n and Wierenga


1976).
T h e r e are a n u m b e r of additional nonlinear sorption m o d e l s d e s c r i b e d in the
literature (e.g., Travis and Etnier 1 9 8 1 ) . T h e sorption of solutes by solids is c o m p l e x ,
and there d o e s not appear to be a single universal model. T h e b e s t approach is to m a k e
an experimental study of the sorptive capacity and rate of the particular solute and solid
that is of c o n c e r n . O n e can then search the literature for a m o d e l that adequately d e s c r i b e s
the experimental results. T h e s e m o d e l s may be c o m p l e x if two equations must be solved
simultaneously. S o m e of the models, such as the bilinear adsorption model, do not have
an analytical solution when c o u p l e d with the advection-dispersion equation. Most of the
readily available c o m p u t e r programs, such as B I O I D , are not c a p a b l e of addressing
nonequilibrium sorption.

132

Chapter T h r e e

3.6

Sorption of Hydrophobic (Organic) Compounds


3.6.1

Introduction

Many organic c o m p o u n d s dissolved in ground water can be adsorbed o n t o solid surfaces


by what is called the hydrophobic effect (Roy and Griffin 1 9 8 5 ) . T h e s e c o m p o u n d s exist
as electrically-neutral species with differing degrees of polarity T h e solubility of organicc o m p o u n d s in water is a function of the degree to which they are attracted by the polar
water molecule. This attraction d e p e n d s upon the polarity of the organic molecule itself.
Hydrophobic c o m p o u n d s can be dissolved in many nonpolar organic solvents but have
a low solubility in water. W h e n dissolved in water, these molecules tend to be attracted
to surfaces that are less polar than water. T h e r e is a small but limited amount of adsorption
of organics on pure mineral surfaces (Ciccioli et al. 1 9 8 0 ; Rogers, McFarlane, and Cross
1 9 8 0 ; Griffin and Chian 1 9 8 0 ) . However, the primary adsorptive surface is the fraction
of organic solids in the soil or aquifer (Karickhoff, B r o w n , and Scott 1979; Schwarzenbach
and Westall 1 9 8 1 ; D z o m b a c k and Luthy 1 9 8 4 ) .
3.6.2

Partitioning onto Soil or Aquifer Organic Carbon

T h e partitioning of a solute o n t o mineral surface or organic c a r b o n content of the soil


or aquifer is almost exclusively o n t o the organic c a r b o n fraction, / , if it constitutes at
least 1% of the soil or aquifer on a weight basis (Karickhoff, B r o w n , and Scott 1 9 7 9 ) .
Under these circumstances a partition coefficient with respect to the organic fraction,
K , can be defined as
OC

(3.30)

A partition coefficient based on soil or aquifer organic matter, K , is also used.


OM

B e c a u s e the weight of the organic matter is greater than that of the organic carbon
alone, K

OC

to K

OC

will be larger than K . B a s e d on lab studies K


OM

OM

can be approximately related

by the equation ( O l s e n and Davis 1 9 9 0 )


K. = 1.724/C
E

(3.31)

II the organic fraction is less than 1%, then it is not automatic that the soil or
aquifer organic c a r b o n will be the primary surface o n t o which the organic c o m p o u n d s
will partition. T h e r e is s o m e critical level of soil or aquifer organic carbon at which the
sorption o n t o the organic matter is equal to the sorption o n t o the mineral matter. B e l o w
this critical level, f* , the organic molecules will be primarily s o r b e d o n t o the mineral
c

surfaces. McCarty, Reinhard, and Rittman ( 1 9 8 1 ) have shown that this critical organicc a r b o n level d e p e n d s upon two variables, the surface area of the soil or aquifer, S,
which is related to the clay content, and a property of the pure organic c o m p o u n d called
the octanol-water partition coefficient.
T h e o c t a n o l - w a t e r p a r t i t i o n c o e f f i c i e n t , A', i s o n e measure o f how hydrop h o b i c a c o m p o u n d is. T h e organic c o m p o u n d is shaken with a mixture of n-octanol
and water and the proportion dissolving into each phase is measured. T h e octanol-water

T r a n s f o r m a t i o n , Retardation, and Attenuation o f S o l u t e s

T A B L E 3.1

Representative

133

values for different organic compounds.


Minimum

Chemical

'C.w

Dichloroethane

62

0.002

2,000

Benzene

135

0.001

1,000

Trichloroethylene

195

0.0007

700

Perchloroethylene

760

0.0002

200

Naphthalene

0.00009

2,350

Pyrene

209,000

Soil

O r g a n i c C a r b o n ( m g kg)

90

0.000002

partition coefficient is the ratio of the concentration in the o c t a n o l to the concentration


in the water- C
|/C
, . It is usually e x p r e s s e d as a log value in reference b o o k s .
According to McCarty, Reinhard, and Rittman ( 1 9 8 1 ) the value of f* can be found
o c l a n o

w a l <

from

C =

2OO(A;1) -

84

( 3 3 2 )

This equation suggests that soils or aquifers with low organic c a r b o n content would
retain organic c o m p o u n d s with high K values but might not retain t h o s e with low K
values. Assuming a surface area of 12 m / g , which would be found with a typical kaolinite
clay soil, T a b l e 3.1 contains f* values and the c o r r e s p o n d i n g minimum soil organic
c a r b o n c o n t e n t necessary b e f o r e organic c o m p o u n d s with different K
values will s o r b
primarily on the organic carbon.
W

oc

3.6.3

Estimating K

from

ow

Data

A n u m b e r of researchers have found that there is a relationship between the o c t a n o l


water partition coefficient and the K value for various organic c o m p o u n d s . T h e use of
such a relationship is predicated upon the following (Karickhoff 1984): ( 1 ) Sorption is
primarily on the organic c a r b o n in the soil or aquifer. ( 2 ) Sorption is primarily hydrop h o b i c , as c o m p a r e d with polar g r o u p interactions, ionic bonding, or chemisorption.
( 3 ) T h e r e is a linear relationship between sorption and the concentration of the solute.
O C

A n u m b e r of different organic c o m p o u n d s have b e e n studied, with the result that


a n u m b e r of different relationships have b e e n developed. Olsen and Da\ls ( 1 9 9 0 ) listed
a total of nine different equations that have b e e n developed. Karickhoff ( 1 9 8 4 ) lists four
equations that have published least squares regression correlation coefficients ( r values)
that e x c e e d 0.9. (A correlation coefficient of 1.00 would mean that there is a perfect
correlation between K and AT. A correlation coefficient of 0.9 m e a n s that 9 0 % of the
variation is a c c o u n t e d for by the equation.)
1

T a b l e 3.2 lists a n u m b e r o f equations that relate K


to K
. W h e r e known, correlation coefficients are listed b e l o w the equations.
T h e equations in T a b l e 3.2 have b e e n derived for many different organic c o m pounds. S o m e have utilized related c o m p o u n d s , whereas others are b a s e d on a mixture
of different organic molecules. The hydrogeologist or e n g i n e e r w h o wishes to estimate
a K value is placed in the situation of deciding which equation to use. T h e best c h o i c e
A C

O L

134

TABLE

Chapter T h r e e

3.2

Equations for estimating

from K

o w

Equation
Number

Equation

(Tl)

log K

(T2)

log < = 1 0 0 log K

Chemicals Used

= 0.52 log K

o m

ow

+ 0.62

72 substituted benzene

- 0.21

10 polyaromatic

Reference
Briggs, 1981

pesticides
o w

Karickhoff, Brown, and Scott 1979

hydrocarbons
(T3)

K^ = 0.63K

(T4)

log

OW

= 0.544 log K

+ 1.377

ow

Miscellaneous organics

Karickhoff, Brown, and Scott 1979

45 organics, mostly

Kenaga and Goring 1980

pesticides
(T5)

log
2

r
(T6)

= 1.029 log K

o w

-0.18

1 3 pesticides

Rao and Davidson 1980

s-trizines and

Rao and Davidson 1980

= 0 . 9 1 ; n = 13

log

= 0-94 log K

+ 0.22

ow

dinitroanalines
(T7)

log K

= 0.989 log K

- 0.346

o w

5 polyaromatic

r = 0.991; n = 5
(TB)

log K

Karickhoff 1981

hydrocarbons

= 0.937 log K - 0.006

Aromatics, polyaromat-

Lyman 1982

ics, triazines
(T9)

In K = In K
x

o l v

- 0.7301

D D T , tetrachlorobi-

McCall, Swann, and Laskowski 1983

phenyl, lindane.
2,4-D, and
dichloropropane
(T10)

log
r

(TU)

= 0-904 log K

- 0.779

ow

= 0.989; n=

Benzene, chlorinated

12

log K ^ = 0.72 log K

ow

+ 0.49

Methylated and chlori-

r = 0.95; n = 1 3

m 2)

log K

Chiou, Porter, and Schmedding 1983

benzenes, PCBs
Schwarzenbach and Westoll 1981

nated benzenes

= 1.00 log K

o w

- 0.317

22 polynuclear

r = 0.98; n = 22

Hassert et al. 1980

aromatics

is an equation that was derived on the basis of chemicals similar to the o n e under study.
T h e various equations tend to yield similar results for many c o m p o u n d s . T a b l e 3.3 shows
log K

AC

values c o m p u t e d from the equations in T a b l e 3.2 for several different organic

c o m p o u n d s that cover a wide range of K

OW

values.

Although there are a n u m b e r of equations, m o s t of the c o m p u t e d log K

OC

values

for the e x a m p l e chemicals fall close to or within o n e standard d e l a t i o n of the geometric


means. T h e r e are s o m e data i n the literature o n actual measured values o f log K . T a b l e
3.4 gives s o m e experimental K

BC

values.

T h e values in T a b l e 3.4 fall c l o s e to the m e a n s listed on T a b l e 3.3. T h e equations


that yield the maximum or minimum values for a particular c o m p o u n d may not be the
most appropriate to use for that c o m p o u n d . It appears that there is no universal equation
that relates K

OC

3.6.4

to K

OW

Estimating

for all classes of organic c o m p o u n d s .


f r o m Solubility Data

T h e value of A^ can also be estimated from the solubility, 5, of a particular c o m p o u n d .


c

Several different equations describing this relationship have b e e n published and are
listed in T a b l e 3.5.

mrvocNCNcoorv-cNO
m m co
COCNCNCNO^O^UCNCOO

LO co
oIV. oo rN-VO ro
r-*. o oT r-Vo o
co o
TT TT TT co co

15

TT ->J- (T)

nI (N
N n
o- m .
O CN d

CN O TnCNtNCNCOKO-Tm

T J - O O O O N O f O N 00

mOtCNCOKCNt-On^N
IJ~> r O*-

CD
OO
I N n r00 CN O O

CN
r-V
i oo nco

& r ^ ^ ^ & l>- r~ U~>

i-"

CN^OOOOCN^

CN oCN
c NCN
rv.o->TrrooaD

O- CTT
N -O
t>. O
-o co
O
OC
r-O
R-<fV

CN
r CNm-^i-i/i-Or-.coO' ^

CN

-o m o

(M N CD T

VO u

135

136

Chapter T h r e e

T A B L E 3.4

Experimentally derived K

QC

values.

Compound

Reference

Benzene

Ethylbenzene

1.50

Chiou, Porter, and Schmedding 1983

1.92

Karickhoff, Brown, and Scott 1 9 7 9

1.98

Rogers, McFarlane, and Cross 1980

2.22

Chiou, Porter, and Schmedding 1983

2,2'-Dichlorobiphenyl

3.92

Chiou, Porter, and Schmedding 1 9 8 3

Tetrachloroethene

2.32

Chiou, Peters, and Freed 1 9 7 9

Napthalene

3.11

Karickhoff, Brown, and Scott 1 9 7 9

Pyrene

4.92

Karickhoff, Brown, and Scott 1 9 7 9

4.80

Means et al. 1980

T A B L E 3.5

Empirical equations by which

can be estimated from S .

Equation
Number
(T13)

Equation

Reference

log K = 0.44 - 0.54 log S


2

(T14)

Karickhoff, Brown, and

S in mole fraction, r = 0.94

Scott 1 9 7 9

log K = 3.64 - 0.55 log 5

Kenaga 1980

S in mg/L
[ I T 5)

log K = 4 . 2 7 3 - 0.686 log S

Means et al. 1980

S in mg/L
(T16)

log K = 3.95 - 0.62 log S


x

Hassett et al. 1983

S in mg/L
(T17)

log K

om

= 0.001
!

- 0.729 log S

S in moles/L, r = 0.996

Chiou, Porter, and


Schmedding 1 9 8 3

A q u e o u s solubility can be e x p r e s s e d in several ways. T h e most c o m m o n is a m a s s


p e r v o l u m e unit, such as milligram per liter. Equation T 1 3 in T a b l e 3.5 uses the c o n c e p t
of m o l e f r a c t i o n . This is the ratio of the m o l e s of a substance to the total n u m b e r of
m o l e s of solution. A m o l e of a substance is equal to its formula weight in grams. In
Equation T 1 7 ( T a b l e 3 . 5 ) solubility is in terms of m o l e s of solute per liter of solution,
a unit known as m o l a r i t y . F o r dilute solutions, to c o n v e n molarity to m o l e fraction,
divide the molarity by 55.6, the n u m b e r of moles of water in a liter.
T a b l e 3.6 contains log K estimated from the solubility for the s a m e c o m p o u n d s
that are listed in T a b l e 3.3. A c o m p a r i s o n of the results of T a b l e 36 with the experimentally
derived values for K found in T a b l e 3.4 shows that all the equations yield an estimate
that is within an o r d e r of magnitude of the experimental result.
T h e use of solubility data is complicated by temperature and ionic strength effects
on solubility. Published solubility data s o m e t i m e s do not indicate the temperature at
which the m e a s u r e m e n t was made. F o r a n u m b e r of reasons, such as the puriry of the
organic chemical, the ionic strength of the water, the temperature, and the experimental
procedure, there can be a range in the reported solubility data in the literature. For
similar reasons there can be a range of octanol-water partition coefficient values reported
for the s a m e c o m p o u n d (Sabljic 1 9 8 7 ) . B e c a u s e of this, o n e should recognize that K

OC

values o b t a i n e d from K

OW

or solubility data are truly estimates.

T A B L E 3.6

values estimated from the aqueous solubility.

Compound:

Dichloroethane

Trichloroethene

Benzene

Ethyl

Tetrachloroethene

Napthalene

2,2'-Dichlorobi phenyl

Pyrene

106.18

128.18

223.10

202.26

31

1.86

0.032

1.49

0.269

-1.50

Benzene
Molecular weight:

78.12

98.96

165.82

131.3B

** Il 1 h i l l LJ I L U I /I

Log S :

5500

1780

1100

140

3.74

3.25

3.04

2.15

5.56 x 10
Log S :

1.00 x

2.28 x

10"

-1.64

-1.25
10~

4.10 x

8.37 x 1 0 ~

-2.08

10"

1.51

x 10"

Solubility
|mg L)
150
2.18

Ci.ikill.i
1 lu. I\ 1
s o l u b i l i t y (moles L)

8.44 x 10

-3.07
<

1.41

x 10

2.42 x

10""

-3.62

-2.85

8.32 x

S o l u b i l i y imoie r r a c r i o n s j
2.54 x 1 0 ~
1.52 x 1 0 ~

4.35 x 1 0 ~
-5.36

-6.83

1.49 x

1.58 x 10

10"

2.84 x

-3.39

-3.82

-4.82

tri 3)

2.06

2.27

2.50

3.04

2.87

3.33

4.13

5.06

(TU)

1.58

1.85

1.97

2.46

2.44

2.82

3.79

4.47

( T I 5)

1.67

2.01

2.15

2.76

2.74

3.21

4.46

5.27

( T I 6)

1.63

1.94

2.07

2.62

2.60

3.03

4.12

4.88

( T I 7)

1.15

1.43

1.75

2.47

2.31

2.88

3.93

5.19

10"'

-8.55

Equation
Number"

Estimated

log

Range

1.15-2.06

1.43-2.27

175-2.50

1.80-3.04

2.31-2.87

2.82-3.33

379-4.46

4.47-5.27

Mean

1.62

1.90

2.09

2.67

2.59

3.05

4.09

4.97

St. dev.

0.32

0.31

0.27

0.24

0.22

0.22

0.25

0.32

Coef. var

0 08

0.07

0.06

0.05

0.04

0.04

0.05

0.08

The equation numbers in this table refer to Table 3.5.

-6.80

-3.00

Log S :

-4.60

1 0

-5.08

138

Chapter T h r e e

3.6.5

Estimating K

from Molecular Structure

T h e applicability of using octanol-water partition coefficient and aqueous solubility data


to estimate K has b e e n questioned b e c a u s e s o m e organic c o m p o u n d s have similar
a q u e o u s solubility but very different octanol solubility (Ellgehausen, D'Hondt, and Fuerer
1 9 8 1 ; Mingelgrin and Gerstl 1983; Olsen and Davis 1 9 9 0 ) . A m o r e fundamental approach
has b e e n suggested on the basis of molecular topology ( K o c h 1 9 8 3 ; Sabljic 1984, 1987;
Sabljic and Protic 1 9 8 2 ) .
oc

M o l e c u l a r t o p o l o g y refers t o the shape o f the organic molecule. T h e particular


parameter of molecular structure that has b e e n related to K is the first-order molecular
connectivity index, 'jr. T h e first-order molecular connectivity index is calculated on the
basis of the nonhydrogen part of the molecule. Each nonhydrogen atom has an atomic
<5 value, which is the n u m b e r of adjacent nonhydrogen atoms. A connectivity index is
then calculated for the molecule by the following formula:
I

= I(^)--

(3.33)

where d and Sj are the delta values for a pair of adjacent nonhydrogen atoms and the
summation takes place over all the b o n d s between nonhydrogen atoms.
L

Sabljic ( 1 9 8 7 ) made a regression analysis between the molecular connectivity and


observed A ' values for 72 organic molecules, including c h l o r o b e n z e n e s , polyaromatic
hydrocarbons, alkylbenzenes, chlorinated alkanes and alkenes, chlorophenols, and heterocyclic and substituted polyaromatic hydrocarbons. T h e relationship between K and
X, which had an r value of 0.95, is given by Equation 3-34. In order to convert K
to K , multiply by 1.724.
om

om

om

(3.34)

log A-, = 0 . 5 3 ' / ; 4 - 0 . 5 4

Equation 3-34 was derived on the basis of nonpolar organic compounds. Sabljic
( 1 9 8 7 ) gave an empirical method of extending this equation to classes of polar and even
ionic organic c o m p o u n d s . W h e n Equation 3.34 is used for polar and ionic organic
c o m p o u n d s , it predicts a K that is higher than observed values reported in the literature.
T h e nonpolar organics from which Equation 3.34 was derived are m o r e strongly sorbed
to soil organic matter than polar organic c o m p o u n d s . Sabljic introduced a polarity
correction value, P , for each of 17 different groups of polar organics ( T a b l e 3 . 7 ) . Equation
3.35 can be used to predict the value of K for polar organics:
om

om

log/c;_, = 0 . 5 3 > : 4 - 0 . 5 4 - / >

(3.35)

As these polarity correction factors are empirically determined, this method cannot
be e x t e n d e d to other classes of polar organic c o m p o u n d s until additional experimental
work is done. In addition, organic phosphates fall into two groups rather than o n e .
However, this correction e x t e n d s o u r ability to estimate K and by extension K, to
a n u m b e r of additional organic c o m p o u n d s . T h e molecular topology method has several
advantages over the estimation of K from octanol-water partition coefficients and
aqueous solubility:
omi

oc

1. T h e r e is a theoretical basis to the molecular topology method for nonpolar organic


compounds.

T r a n s f o r m a t i o n , Retardation, and Attenuation o f S o l u t e s


T A B L E 3.7

139

Polarity correction factors for classes of polar organic

compounds.
P o l a r i t y Correction
Class of C o m p o u n d s

Factor, P,

Substituted benzenes and pyridines

1.00

Organic phosphates (group 1)

1.03

Carbamates

1.05

Anilines

1.08

Nitrobenzenes

1.16

Phenylureas

1.88

Triazines

1.88

Acetanilides

1.97

Uracils

1.99

Alkyl-N-phenylcarbamates

2.01

3-Phenyl-l -methylureas

2.07

3-Phenyl-l -methyl-1 -methoxyureas

2.13

Dinitrobenzenes

2.28

3-Phenyl-l ,1 -dimethylureas

2.36

Organic acids

2.39

3-Phenyl-1 -cycloalkylureas

2.76

Organic phosphates (group 2)

3.19

2. T h e literature contains a range of experimentally derived values for b o t h the octanol


water partition coefficient and the a q u e o u s solubility for a n u m b e r of n o n p o l a r organic
c o m p o u n d s . O n e has no way of knowing which is the c o r r e c t value.
3. S o m e c o m p o u n d s with similar a q u e o u s solubility values have quite different octanolwater partition coefficients.
4. T h e r e are a n u m b e r of c o m p e t i n g equations that can be used for b o t h the K

OC

and

solubility methods. O n e is never quite sure which to select.


5. T h e solubility and K

OW

m e t h o d s w e r e devised strictly for n o n p o l a r organic c o m p o u n d s .

T h e r e is no way to e x t e n d them to polar organic c o m p o u n d s .


However, the molecular topology m e t h o d d o e s not a c c o u n t for the ionic strength,
pH, and temperature of the solution.

EXAMPLE
PROBLEM

C o m p u t e the value o f x for trichloroethylene. T h e structure o f trichloroethylene is as


shown:
CI

CI

= C
\

/
CI

T h e r e are four nonhydrogen pairs: C a r b o n 1 is b o n d e d to two c h l o r i n e atoms


and c a r b o n 2. C a r b o n 2 is b o n d e d to o n e c h l o r i n e atom. T h e following table shows
the <5, and <5, values as well as the c o m p u t e d ( c i ^ )

'

value for each pair.

140

Chapter T h r e e

Nonhydrogen pair

(6,*,)-

Cl-C(l)
Cl-C(l)

C(l)-C(2)

3
2

C(2)-C1

3
3
2

0.577
0.577
0.408
0.707
l

2.269 = x

3.6.6

Multiple Solute Effects

Many hazardous waste sites contain m o r e than o n e organic solvent in aqueous solution.
It has b e e n shown that the solubility of structurally similar hydrophobic organic liquids
in a q u e o u s solutions is dependent upon the mixture of solutes present ( B a n e r j e e 1 9 8 4 ) .
T h e s e liquids behave in a nearly ideal fashion, which is described by the following
equation:

org

(3.36)

where
Q the equilibrium molar concentration of the r'th c o m p o n e n t in
the mixture
5,- = the water solubility of the c o m p o n e n t in its pure form

(XiXrg

m o l e fraction o f the

ith

c o m p o u n d in the organic phase

T h e concentration of a solute in a solution that is saturated with several structurally


similar c o m p o u n d s is less than it would be in water alone. Mixtures of dissimilar liquids
and of organic solids are m o r e c o m p l e x ( B a n e r j e e 1 9 8 4 ) . In addition to having mutual
effects on solubility, organic mixtures will also c o m p e t e for sorption sites.
In o n e study the p r e s e n c e of organic solvents, such as methanol, ethanol,
2-propanol, and butanol, in concentrations of 5 to 1 0 % reduced the soil water partition
coefficient for an organic molecule, k e p o n e . This increased the rate at which k e p o n e
migrated through a soil column (Staples and Geiselmann, 1 9 8 8 ) .
If a chemical analysis indicates that an aqueous sample contains a hydrophobic
organic c o m p o u n d in amounts in e x c e s s of its solubility, it is likely that part of the
c o m p o u n d is present as a n o n a q u e o u s phase liquid. Under such conditions n o n a q u e o u s
p h a s e liquid transport theory must be considered (Hunt, Sitar, and Udell 1 9 8 8 ) .

3.7

Homogeneous Reactions
3.7.1

Introduction

According to the classification system of Rubin ( 1 9 8 3 ) h o m o g e n e o u s reactions are o n e s


that take place entirely within the liquid phase. If the reactions are revesible and p r o c e e d
rapidly enough, the reaction can be described as being in local chemical equilibrium

Transformation,

Retardation, and

Attenuation of S o l u t e s

141

(Walsh et al. 1 9 8 4 ) . If the reaction either d o e s n o t reach equilibrium or is nonreversible,


then it is treated as a h o m o g e n e o u s , nonequilibrium reaction.
3.7.2

Chemical E q u i l i b r i u m

If two c o m p o u n d s in solution, A and B, react to form product C, which can dissociate


into A and B, then the reaction is reversible and is e x p r e s s e d as:

ah + be,

cC

(337)

w h e r e a, b, and c represent the n u m b e r of m o l e c u l e s of each c o m p o u n d that are n e e d e d


to b a l a n c e the reaction. W h e n the reaction has p r o g r e s s e d to the point that no further
net production of C o c c u r s , then it has r e a c h e d equilibrium. T h e reaction continues,
but the forward rate and the reverse rate have b e c o m e equal. At that point, we can
m e a s u r e [ A ] , [ B ] , and [ C ] , the c o n c e n t r a t i o n s o f the reactants and the product. T h e
relationship b e t w e e n them is e x p r e s s e d as an equilibrium constant, K .
eq

[CY
K

~WW

(3 38)

'

T h e reaction must have sufficient time to p r o c e e d to the point of equilibrium


b e f o r e Equation 3.38 is valid. T h i s represents a Class I reaction according to Figure 3 . 1 .
3.7.3

Chemical Kinetics

If reaction 3-37 w e r e to p r o c e e d so slowly that it would not have time to c o m e to


equilibrium in the framework of the ground'-water flow system, then local equilibrium
c a n n o t be a s s u m e d and we must c o n s i d e r it in the framework of chemical kinetics. We
can look at the reaction from the standpoint of either the disappearance of the reactants,
A and B, or the a p p e a r a n c e of the product, C. We have three rates to consider, two that
d e s c r i b e the disappearance o f the reactants and o n e that d e s c r i b e s the p r o d u c t i o n o f
the product:

= -

^ = *[A]'[B]

(3.39)

d[B]
RB=

- - ^ -

= K'[AJ"[B]

(3.40)

at
R C = ^ - = K"[CY
where
R

and R = the reaction rates for the disappearance of A and B


B

R = the reaction rate for the a p p e a r a n c e of C


c

[A], [B], and [ C ] = the m e a s u r e d c o n c e n t r a t i o n s o f A , B , and C


K, K', and K" = reaction rate constants
p, q, and r = reaction o r d e r with respect to the indicated reactant
o r product

(3.41)

142

Chapter T h r e e

If o n e of p, q, or r in Equations 3 3 9 , 3.40, and 3.41 is equal to zero, then the


reaction rate is not a function of that product or reactant and the reaction is said to be
zero o r d e r with respect to that reactant or product.
If o n e of p, q, or r is equal to 1, then there is a linear relationship with respect to
that product or reactant, and the reaction rate is said to be first order.
If n o n e of p, q, or r is 0 or 1, then there is a m o r e c o m p l e x relationship with
respect to that product or reactant. Such a reaction is m o r e difficult to analyze mathematically than zero- or first-order kinetics.
If a c o m p o u n d is present in a system in great e x c e s s , then the reaction rate may
well be independent of the concentration. If we consider the depletion of reactant A in
such a system, it could be described by the zero-order equation

[A] = [ A ] - t c r

(3.42)

where
K = the reaction rate constant
[A] o = the initial concentration
[A] = the concentration at s o m e time /
In a first order system, the rate at which the reactant disappears is described by
[A I

ln^f=-e/

(3.43)

[A] = [ A l o e - " '

(3.44)

or

Reaction-rate constants and the order of the reactions must be determined experimentally. T h e s e are not equal to the equilibrium constant of the reaction.

3.7.4

T e n a d s in Chemical Reactions

Rubin ( 1 9 8 3 ) introduced the c o n c e p t of tenads m chemical reactions. This classification


s c h e m e is useful as a means of formulating transport equations that incorporate chemical
reactions. A t e n a d is a reacting or nonreacting chemical entity w h o s e global mass in
the system is reaction independent. A chemical entity may be either an uncharged atom
or group of atoms, an ion, or an electron. Chemical species may be c o m p o s e d of o n e
or m o r e tenads. T h e global mass of a tenad is the sum of the mass of the tenad wherever
it occurs. Reaction-independent means that the global mass d o e s not depend upon the
extent o f progress o f the reaction.
Reactions can have alternative sets of tenads. Consider a system with solid C a S 0
in water. T h e reaction that o c c u r s is
4

CaS0

Ca

2+

+ S0

2
4

"

(3.45)

T h e two alternative sets of tenads are ( 1 ) the atoms Ca, S, and O, and ( 2 ) the ions
Ca
and S 0 " . C a S 0 ( s o l i d ) cannot b e a tenad b e c a u s e the mass depends upon
the reaction progress, but the C a
in C a S 0 ( s o l i d ) Is p a n of the tenad { C a }.
2 +

2 +

2+

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

143

We will c o n s i d e r the chemical c h a n g e s that can o c c u r during o n e dimensional Bow


of solute. T h e solute flux equation is

F,= - "*DJ^- + n v [C]i


e

(3.46)

where
Fj = the x-direction mass flux of a solute
[ C ] , = the concentration of the solute
D = the coefficient o f longitudinal dispersion
L

n = the effective porosity


e

T h e one-dimensional continuity equation can be written as


T

dm

dF

-JT=-^

+s

. _

(3 47)

where
m

= global mass of the given chemical entity stored in a unit volume of the
p o r o u s media (solid and solute p h a s e s )

F = total mass flux of the entity into or out of the unit volume
T

S = total s o u r c e / s i n k term for the entity


T

F o r a tenad that exists only in the form of solute ;', m is equal to n.[C|,-, and
/'' =
T h e total s i n k / s o u r c e term, S , represents s o u r c e s and sinks from external
sources, S biological growth and decay, 5^,,, and chemical reactions, S
.
T

cttl

c h c m

fdm \

(dm \

St / ,

V st j

(dm
b i o

\ dt

(3.48)
1

In o r d e r to illustrate the use of the tenad method, we will e x a m i n e the class of


sufficiently fast, reversible, h o m o g e n e o u s reactions, indicated by / o n Figure 3.1. F o r the
c a s e of reversible h o m o g e n e o u s reactions, there is no s o u r c e / s i n k term. In this c a s e
Equation 3.46 can be written as
S

n -^ =W[C\
e

(3.49)

w h e r e W is a linear o p e r a t o r defined as
w=n

< ^-"< >

For the c a s e of flow in the vadose zone, the volumetric water content, 8, should
be used in place of the effective porosity, n .
In the e x a m p l e we will have the ground water containing two solutes, E and F,
which react to form a product, EF. T h e invading a q u e o u s solution contains solutes E
and G, which react to form product EG. T h e reactions are reversible and sufficiently fast
for the local equilibrium assumption to apply.
e

144

Chapter T h r e e

T h e two reactions that o c c u r are

E F ^ ^ E+F

(3.51)

EG ^ E + G

(3.52)

T h e reaction has three tenads, E, F, and G. It could represent oxidation, reduction,


or c o m p l e x formation. T h e r e are a total of five participants in the reaction, the three
tenads, EF, and EG. We need five equations in order to be able to solve for these five
unknowns. T h e s e equations can be obtained by writing a convective transport equation
for each of the three tenads, E, F, and G, and by writing the equilibrium equation for
c.u li . >l I h e IWI pre idui Is I I and I < I

8[EF]

3[E]

d[EG]
1

n +n - + n, - = W[E] + W [ E F ] + W[EG]
e

8t

'

'

Bt

fl[F]

dt

= W[F] + W{EE]

B]EG]

n -jy-+n -~-=W]G] + W[EG]


r

(3.53)

3[EF]

- j f + n,

BIG]

[E](F]
[EF]
[E][G]
[EG]

(3.54)
(3.55)

(3.56)
(3.57)

Rubin ( 1 9 8 3 ) presents methods of getting similar sets of equations for all six of
the classes of chemical reactions listed in Figure 3.1.

3.8

Radioactive Decay
If radionuclides enter the ground water system, those which are cations are subjected
to retardation on soil surfaces. In addition they will undergo radioactive decay, which
will reduce the concentration of radionuclides in both the dissolved and sorbed phases.
A factor for radioactive decay can be substituted for the last term of Equation 3.1 in the
following form:
3C\

ln^

(3 58)

w h e r e /. is the half life of the radionuclide.

3.9

Biodgradation
T h e degradation of dissolved organic molecules in ground water is of great interest to
practicing contaminant hydrogeologists. As was shown in Chapter 1, much of the ground
water contamination is due to organic chemicals, including hydrocarbons. T h e mechanism of biodgradation will be covered in detail in Chapter 7. In this section we will

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

145

e x a m i n e the transport and decay equations that can be used to d e s c r i b e the p r o c e s s .


Although a wide range of organic m o l e c u l e s can be degraded, for the sake of this
discussion we will refer to them simply as hydrocarbons. T h e h y d r o c a r b o n s form a
substrate for microbial growththat is, they provide the energy s o u r c e for the m i c r o b e s
which form a b i o f i l m on the solid surfaces in the aquifer.
If the m i c r o b e s require o x y g e n in their metabolism, then the p r o c e s s is called
a e r o b i c b i o d g r a d a t i o n . T h e removal o f the hydrocarbons, the c o n s u m p t i o n o f
oxygen in the process, and the growth of the m i c r o b e s in the aquifer, ignoring transport
through the biofilm, can be d e s c r i b e d by the following equations, which are modifications
o f the M o n o d f u n c t i o n (also k n o w n a s Michaelis-Menten function) ( B o r d e n and
Bedient 1 9 8 6 ) :
dH

\(

.'...

'

"

\K

(-JL-)(-J)

MAY

(3.60)

= MJij\[\
dt

+ Oj\K + O)
0

k YC

0C

(3.61)

- bM,

where
H = hydrocarbon concentration in p o r e fluid ( M L
0 = oxygen concentration i n p o r e fluid ( M L

- 3

- 3

)
- 3

M, = total a e r o b i c microbial c o n c e n t r a t i o n ( M L

/J = m a x i m u m hydrocarbon utilization rate p e r unit mass of


aerobic microorganisms ( T

Y = microbial \ i e l d coefficient (g cells/g h y d r o c a r b o n )


K = hydrocarbon half-saturation constant ( M L

- 3

K = oxygen half saturation constant ( M L


0

- 3

k = first-order decay rate of natural organic c a r b o n


c

C natural organic c a r b o n concentration ( M L


oc

b = microbial decay rate ( T

- 3

G = ratio of oxygen to hydrocarbon c o n s u m e d


T h e microorganisms will grow on both naturally occurring organic c a r b o n as well
as hydrocarbon contaminants. T h e microorganisms tend not to m o v e in the aquifer
b e c a u s e they generally adhere to aquifer materials (Harvey, Smith, and G e o r g e 1 9 8 4 ) .
Even if the m i c r o b e s are free to move, the natural tendency of the aquifer matrix will
be to filter them out. T h e r e will be s o m e tendency for m i c r o b e s to transfer from the
solid surface to solution. As a first approximation this can be c o n s i d e r e d to be a linear
function of the total mass of microorganisms.
We can c o m b i n e Equations 3.59, 3 . 6 0 , and 3.61 individually with Equation 3.1 to
obtain solute transport equations for hydrocarbon, oxygen and microorganisms. T h e
hydrocarbon is assumed to s o r b o n t o the solid surfaces following a linear sorption

(3 59)

j-

146

Chapter Three
isotherm. T h e resulting equations are ( B o r d e n and Bedient 1 9 8 6 )
dH

if

dH

dO

B0

dH\

h,M,(

\
( 3

F - ^ ? - ^- ( l w X * T j
^ _ i u * _ a
p q ( ^ M
w

dt

r \
m

dx

dxJ

_ ^

\K + H)\K + OJ

6 3 )

'

O M )

'.

'

where

M = concentration of a e r o b i c m i c r o b e s in solution
s

r = retardation factor for hydrocarbon


h

r = microbial retardation factor


m

v = average linear ground-water velocity


x

S o m e microorganisms can degrade hydrocarbons in the a b s e n c e of oxygen. T h e s e


m i c r o b e s use another electron acceptor, such as nitrate (Major, Mayfield, and Barker
1 9 8 8 ) . Anaerobic d e c o m p o s i t i o n o f hydrocarbons can b e described b y another variation
of the M o n o d function, which describes two-step catalytic chemical reactions ( B o u w e r
and McCarty 1 9 8 4 ) . This function is
dH
dt

(
= - h M\
ua

\K

Hj

(3.65)

where
M = total mass of a n a e r o b i c m i c r o b e s
a

u a

= maximum hydrocarbon utilization rate per unit mass of anaerobic


microbes

K = half maximum rate concentration of the hydrocarbon for anaerobic


a

decay
T h e solute transport and decay equation for a n a e r o b i c biodgradation in the
a q u e o u s phase is
dH

1 /

dH

dh\

K-M.l

ferJ

( 3

6 6 )

If the concentration of the hydrocarbon, H, is much less than K , the half maximum
rate concentration, then Equation 3.65 can be simplified to a linear form by neglecting
H in the denominator ( B o u w e r and McCarty 1 9 8 t ) . This results in a first-order decay
term:
a

dH
dt

(hju:
lia
'.i

tl

(3.67)

Under these conditions the solute transport equation with anaerobic biodegradation b e c o m e s

dt

\ dx * dx J \

r K,
H

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

147

A single microbial growth substrate (that is, a hydrocarbon or o t h e r organic chemical that can serve as an energy s o u r c e for the m i c r o b e s ) cannot be reduced for e x t e n d e d
periods of time b e l o w the minimum concentration n e e d e d to maintain the microbial
population. In o t h e r words, m i c r o b e s cannot degrade a s u b s t a n c e b e l o w the c o n c e n tration that the)' n e e d to c o n t i n u e to exist. This minimum concentration, Hmi, is a
function of the hydrocarbon, the electron acceptor, and the microorganism. It can be
e x p r e s s e d a s ( B o u w e r and M c C a n y 1 9 8 4 )
(3.69)
In the c a s e of a e r o b i c microbial decay, there is also s o m e minimum oxygen
concentration b e l o w which a e r o b i c decay will not o c c u r . Although there is a minimum
hydrocarbon concentration for e a c h hydrocarbon that might be present, m i c r o b e s may
utilize m o r e than o n e hydrocarbon as a growth substrate. As a result, if there is a mixture
of h y d r o c a r b o n s present, the m i c r o b e s could degrade the separate h y d r o c a r b o n s to
c o n c e n t r a t i o n s that are lower than if the individual hydrocarbon w e r e the only o n e
present ( M c C a n y , Reinhard, and Rittman, 1 9 8 1 ) . T h e primary substrate that is supporting
the m i c r o b e s can be a single s u b s t a n c e or a mixture of substances. Although the primary
substrate is supporting the growth of the m i c r o b e s , s u b s t a n c e s that are present in trace
a m o u n t s can be c o n s u m e d by the microbial population through a p r o c e s s k n o w n as
s e c o n d a r e utilization. T h e decay of a hydrocarbon that is undergoing secondary utilization
is d e s c r i b e d by Equation 3.67.
Figure 3.10 s h o w s the results of a laboratory e x p e r i m e n t in microbial decay ( B o u w e r
and McCarty 1 9 8 4 ) . An a e r o b i c soil c o l u m n was established with a microbial population.
A solution with 10 m g / L of acetate as the primary substrate along with 10 /tg/L of
c h l o r o b e n z e n e and 10 /rg/L of 1,4-dichlorobenzene as secondary substrates was introduced into the soil c o l u m n as a constant flow. T h e utilization of the primary and the
secondary substrates o c c u r r e d simultaneously in the first 7 cm of the soil column. After
that flow distance, all three c o m p o u n d s r e a c h e d an irreducibly low c o n c e n t r a t i o n that
would not support further microbial growth.
Srinivasan and Mercer ( 1 9 8 8 ) developed a c o m p u t e r model, B I O I D , which simulates one-dimensional solute transport with biodgradation. In their m o d e l they relied
upon the observations of B o r d e n et al. ( 1 9 8 6 ) that microbial growth reaches equilibrium
rapidly with respect to the rate of ground-water flow and therefore the microbial population can be a s s u m e d to be constant. This m e a n s that Equation 3-64 is not needed,
since dMjdt = 0. T h e m o d e l solves the equivalent of Equations 3.62 and 3.63 for a e r o b i c
biodgradation and Equation 3.66 for a n a e r o b i c d e c o m p o s i t i o n . T h e c o m p u t e r c o d e
automatically switches from a e r o b i c to a n a e r o b i c d e c o m p o s i t i o n if the o x y g e n levels
d r o p b e l o w the minimum to support a e r o b i c decay.
If the substrate concentration, H, is less than 0.25Af and conditions are anaerobic,
then the equation used is ( 3 . 6 8 ) . If H K, the Michaelis-Menten function and the firstorder decay function will yield similar results. Figure 3.11 s h o w s the modeling of the
a n a e r o b i c biodgradation of trichloroethylene (Srinivasan and Mercer 1 9 8 8 ) . A pulse of
contamination of 2 7 8 0 - ^ g / L trichloroethylene was injected into an aquifer for 1 5 0 hr,
followed by c o n t i n u e d injection at 0 ng/L. Figure 3.11 s h o w s the concentration of
trichloroethylene in the plume after 2 5 0 0 hr and again after 3 7 5 0 hr. B o t h MichaelisMenten and first-order decay functions w e r e used in the modeling, with very similar

o Measured data

Chlorobenzene

1.4 - dichlorobenzene

10

15

20

Distance along column (cm)


FIGURE 3.10

Measured steady-state profiles of the biodegradation of acetate, chlorobenzene and 1,4-

dichlorobenzene in an aerobic biofilm reactor. Source: Modified from E. J. Bouwer and P. L. McCarty,
Ground Wafer 2 2 , no. 4 (1 9 8 4 ) : 4 3 3 - 4 0 . Used with permission. Copyright 1 9 8 4 Water W e l l Journal
Publishing Co.

T C E depletion in VB sand

Michaelis-Menten kinetics
First-order degradation
300

/!r
\ t ^ T = 2500 h r
1
1
II

ft
\\
\\

T = 3750 hr

\
\
\

II
II

\
\\

[
I

/'

II

ft
II

50
Distance (ft)
FIGURE 3.11

Modeling of the movement and anaerobic biodgradation of a trichloroethylene plume

in a sand aquifer using both Michaelis-Menten and first-order decay functions. Source: P. Srinivasan and
J. W. Mercer, Ground Water 26, no. 4 (1 9 8 8 ) : 4 7 5 - 8 7 . Used with permission. Copyright 1988 Water
W e l l Journal Publishing Co.

148

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

149

0.05

0.04

Distance (m)
FIGURE 3.12

B I O l D model showing the position of a solute plume with (a) no retardation and no decay,

(b) no retardation but biodgradation, (c) retardation that follows a linear sorption isotherm but no decay,
and (d) retardation that follows a linear sorption isotherm and biodgradation.

results. T h e reduction in the peak c o n c e n t r a t i o n of the trichloroethylene p l u m e with


travel time can also be s e e n on this figure.
T h e B I O I D m o d e l can b e used t o illustrate the effects o f retardation and biodgradation. Figure 3.12 s h o w s the position of a solute p l u m e with ( a ) no retardation and
no decay, ( b ) no retardation but biodgradation, ( c ) retardation that follows a linear
sorption isotherm but no d e c a y and ( d ) retardation that follows a linear sorption isotherm
and biodgradation. T h i s m o d e l represents the m o v e m e n t of a slug of solute introduced
into an aquifer over a 160-da period. T h e position of the solute front is s h o w n after 8 0 0
da of travel. Biodgradation is m o d e l e d as first-order decay with a half-life of 4 0 0 da.
Retardation delays the advance of the solute front, whereas biodgradation reduces the
peak value.

3.10

Colloidal Transport
C o l l o i d s are particles with diameters less than 1 /rm. Colloidal-size particles include
dissolved organic m a c r o m o l e c u l e s , s u c h as h u m i c s u b s t a n c e s , microorganisms, tiny
droplets of insoluble organic liquids, and mineral matter (McCarthy and Zachara 1 9 8 9 ) .
S o m e colloids may be small e n o u g h to flow through the p o r e s of an aquifer. If dissolved
solutes partition o n t o a colloid, this can c r e a t e a s e c o n d m o b i l e phase. T h e solute can
then be found in three regions: dissolved, s o r b e d o n t o m o b i l e colloids, and s o r b e d o n t o

150

Chapter Three
immobile surfaces. T h e study of colloids in ground water is greatly complicated by the
fact that the p r o c e s s of installing monitoring systems such as wells and piezometers may
introduce colloids that were not originally present. Sampling p r o c e s s e s may also create
colloids, such as the precipitation of colloidal iron due to oxygenation of water and the
dislocation of stable colloids due to t o o rapid pumping.
In order for colloids to participate in contaminant transport, they must first be
released to the ground water. This can o c c u r due to chemical precipitation, biological
activity, or disaggregation of stable aggregates. T h e y can also be carried into the aquifer
by infiltrating water, especially through cracks and m a c r o p o r e s .
Colloids are m o b i l e if their surface chemistry is such that the individual colloids
are repulsed so that they remain disaggregated, rather than being attracted to form larger
particles. In addition, the pore-size geometry must be such that the colloids are not
filtered from the suspension.
T h e r e is ample evidence in the literature that colloids can migrate in aquifers.
Keswick, Wang, and Gerwa ( 1 9 8 2 ) report that bacteria have migrated up to 9 0 0 m in
an aquifer and viruses have migrated up to 9 2 0 m. Layered silicate clays from surface
soils have b e e n found to travel up to several hundred meters to wells (Nightingale and
Bianchi 1 9 7 7 ) . Asbestos fibers have b e e n found in an aquifer recharged with surface
water containing the mineral (Hayward 1 9 8 4 ) .
Colloids have b e e n implicated in the u n e x p e c t e d movement of plutonium and
americum, radioactive elements that are normally believed to be relatively immobile in
the soil due to a high distribution coefficient (McCarthy and Zachara 1 9 8 9 ) .
An a q u e o u s solution may contain dissolved organic macromolecules. Hydrophobic
organic solvents in a q u e o u s mixtures may partition o n t o these m a c r o m o l e c u l e s rather
than soil organic c a r b o n (Enfield 1 9 8 5 ) . W h e n this happens the mobility of the organic
solvent is greatly e n h a n c e d , especially if it has a low mobility (high fC value). In fine
oc

grained soils or aquifers the m a c r o m o l e c u l e s may even have a velocity greater than the
average linear ground-water velocity (Enfield and Bengtsson 1 9 8 8 ) . This is due to the
s i z e - e x c l u s i o n e f f e c t , which o c c u r s when molecules or ions are so large that they
c a n n o t travel through the smaller pores. As a result, they are restricted to the larger
pores, in which the ground-water velocity is greater than average. Thus these molecules
will travel at a rate greater than the average linear ground water velocity. This effect is
m o r e prevalent in fine-grained soils and aquifers with s o m e pores small enough to
e x c l u d e s o m e molecules. Organic m a c r o m o l e c u l e s are likely to be produced in municipal
landfill leai hate, which has a high ami runt of d i s s , iked i irganit i art* in f h i s i s o n e teasi in
codisposal of toxic organic liquids and municipal refuse is not wise.

Case Study: Large-scale Field Experiment on the Transport


of Reactive and Nonreactive Solutes in a Sand Aquifer Under
Natural Ground-water G r a d i e n t s B o r d e n , Ontario
3

In August, 1982, an experiment was begun by injecting about 1 2 m of water containing a


number of solutes into a sand aquifer in which 275 multilevel ground-water samplers had
previously been installed (Mackay et al. 1986; Freyberg 1986; Roberts, Goltz, and Mackay
1986). Each multilevel sampler had from 1 4 to 18 sampling ports vertically separated by

Transformation, Retardation, and Attenuation of Solutes

15

about 0.2 to 0.3 m. Figure 3.13 shows the distribution of sampling points. The average porosity of the sand is 0.33, the geometric mean of hydraulic conductivity is 7.2 x 1 0

- 5

m/sec,

and the mean annual horizontal gradient is 0.0043. The average linear ground-water velocity computed from these values is 29.6 m/yr. The direction of ground-water flow at the site is
to the northeast in the direction indicated by line A-A' on Figure 3.13.
The injected water contained the following solutes:
Concentration
Solute

(mg/L)

Mass
(g)

Chloride ion

892

10,700

Bromide ion

324

3,870

Bromoform

0.032

Tetrachloroethylene

0.030

0.38
0.36

Carbon tetrachloride

0.031

0.37

1,2-dichlorobenzene

0.332

4.0

Hexachloroethane

0.020

0.23

From August 24, 1982, to June 2, 1984, synoptic monitoring was accomplished as
water was withdrawn from a large number of the monitoring devices and analyzed for the
ionic tracers and organics on 18 different occasions. This was done in order to assess the
overall movement of the plume. In addition, at selected points along the flowpath time-series
monitoring was done by sampling on a much more frequent basis than the synoptic monitoring. In all, 14,465 samples were analyzed during this time period for the synoptic monitoring
program, and 1246 samples were analyzed for the time-series monitoring.
Figure 3.14 shows the breakthrough curves for chloride, carbon tetrachloride, and tetrachloroethylene at a monitoring point in the center of the plume located 5.0 m from the
injection wells. Values are shown as relative concentration, which is the observed concentration for a parameter divided by the injected concentration. At 100 da the chloride slug has
just about passed the observation point, the carbon tetrachloride plume has just about
reached a peak, and the tetrachloroethylene plume has yet to reach it. By 200 da both the
carbon tetrachloride and the chloride plumes have passed the monitoring point, whereas the
tetrachloroethylene is near its peak value. The behavior of bromoform was very close to that
of carbon tetrachloride and is not shown.
Figure 3.14 illustrates the chromatographic effect of retardation. The chloride ion is
essentially unaffected by travel through the aquifer, whereas the carbon tetrachloride and the
tetrachloroethylene are traveling at slower rates. The result is a separation of the components
of the plume, a phenomenon known as the c h r o m a t o g r a p h i c effect.
Figure 3.15 shows the arrival at the monitoring point of chloride ion and two other
organics, dichlorobenzene and hexachloroethane. The occurrences of both of these compounds are sporadic, and they have relative concentrations much less than carbon tetrachloride and tetrachloroethylene. Analysis of the area under the time-concentration curves shows
that the relative areas of carbon tetrachloride, bromoform, and tetrachloroethylene are about
the same size as the chloride curve. This indicates that the mass of these organics being
measured is about the same as the mass that was introduced into the aquifer. However, for
dichlorobenzene and hexachloroethane the relative areas are much less than that of chloride,
indicating that the mass being measured is much less than the mass that was introduced. The

Time (days)
FIGURE 3.14

Arrival times of chloride, carbon tetrachloride, and tetrachloroethylene at a measuring

point 5.0 m downgradient from the injection well at Borden, Ontario. Source: P. V. Roberts, M. N. Goltz,
and D. M. Mackay, Wafer Resources Research 22, no. 13 {1 9 8 6 ) : 2 0 4 7 - 5 9 . Copyright by the American
Geophysical Union.

Time (days)
F I G U R E 3.15

Arrival times of chloride, dichlorobenzene, and hexachloroethane at a measuring point

5.0 m downgradient from the injection well at Borden, Ontario. Source-. P. V. Roberts, M. N. Goltz, and
D. M. Mackay, Water Resources Research 2 2 , no. 13 ( 1 9 8 6 ) : 2 0 4 7 - 5 9 . Copyright by the American
Geophysical Union.

153

> I

(m)

(i

-10
-10
FIGURE 3.16

10

20

30

y(m)

Plumes of chloride, carbon tetrachloride, and tetrachloroethylene at the end of the ex-

perimental period. The plumes are based on depth-averaged values. Source: P. V. Roberts, M. N. Goltz,
and D. M. Mackay, Water Resources Research 2 2 , no. 13 (1986):204759. Copyright by the American
Geophysical Union.

154

Transformation, Retardation, and Attenuation of Solilutes


1

155

(m)

60

50

40

633 days

30

20
633 days
m )

Kj/$y)
10

da

>'

days

-10
-1 i

<f

V. 16 days

I
10

I
y(m)

-10

10

y (m)

(a)
FIGURE 3.17

(b)

(a) Growth of carbon tetrachloride plume with time; (b) growth of tetrachloroethylene

plume with time. Source.- P. V. Roberts, M. N. Goltz, and D. M. Mackay, Water Resources Research 22,
no. 13 ( 1 9 8 6 ) : 2 0 4 7 - 5 9 . Copyright by the American Geophysical Union.

missing mass was presumably removed by biodgradation or abiotic pathways such as hydrolysis (Chapter 7).
Similar results were obtained during the synoptic sampling. Figure 3.16 shows the
plumes of chloride (CI), carbon tetrachloride (CTET) and tetrachloroethylene (PCE) at the end
of the experimental period (after 633 days for the organics and 647 days for the chloride).
The chloride can be seen to have moved significantly beyond the organics and carbon tetrachloride has moved farther than tetrachloroethylene. Figure 3.17(a) shows the growth of the
carbon tetrachloride plume and 3.17(b) shows the growth of the tetrachloroethylene plume.
The relative velocities of the various solutes are indicated by the positions of the centers of mass of the plumes at the end of the experiment. The chloride plume was measured
after 647 days of travel and the organic plumes were measured after 633 days. The distance from the center of the injection zone to the center of mass of the plume as well as the

1 56

Chapter Three
average velocities are given in the following table:
Distance to
Center o f M a s s

Average Velocity

(m)

(m/day)

Chloride

58.21

0.0900

Carbon tetrachloride

24.82

0.0392

Bromoform

21.51

0.0340

Tetrachloroethylene

12.33

0.0195

Compound

Dichlorobenzene

8.09

0.0128
None detected after 633 da

Hexachloroethane

The distances traveled by each of the compounds as a function of time are plotted n
Figure 3.18. The distance traveled by chloride is linear with time, indicating a constant advective rate. On the other hand, the organic solutes indicate decreasing velocities with increasing time.
The organic compounds have the following aqueous solubilities, octanol-water partition
coefficients, and first-order molecular connectivity indices:
Solubility

1,

Bromoform

3190 mg/L at 3 0 C

Carbon tetrachloride

805 mg/L at 20C

2.83

2.70

'X
2.00
2.00

Tetrachloroethylene

1503 mg/L at 25 C

3.40

2.60

2.64

3.38
3.82

3.40

3.83
3.25

log

2.30

150 mg/L at 2 5 C
1,2-dichlorobenzene

156 mg/L at 25C

Hexachloroethane

50 mg/L at 22.3C

3.60

Value from P. H. Howard, Fate and Exposure Data (or Organic Chemicals (Chelsea, Mich.: Lewis Publishers, 1990).
" Volue from Mockoy el al. (1986).
1

Value from Korel Verschueren, Handbook of Environmental Dato on Organic Chemicals, 2d ed. (New York: Von Noslrond
Reinhold Co., 1983).

This case history illustrates the problems with predicting the transport of organic compounds in ground water. The organic carbon content of the aquifer is low (0.02%), so absorption is limited. The observed order of relative velocity from greatest to least was carbon
tetrachloride, bromoform, tetrachloroethylene, and dichlorobenzene. The published solubility
data were collected at three different temperatures and thus cannot be directly compared.
Two solubilities for tetrachloroethylene are published for the same temperature. Other than
for 1,2-dichlorobenzene, the published K values are not particularly close. Moreover, the
ow

relative order of the K values are not consistent from one reference source to another nor
ow

does either correspond to the observed relative velocities. However, the computed first-order
molecular connectivity indices have a rank order that corresponds to the observed movement
of the different plumes. This suggests that molecular topology might be more useful than the
other methods in predicting relative plume movements. There appears to be increased retardation with time, which indicates that nonequilibrium sorption may be occurring. Further research on physical properties of organic chemicals may resolve the inconsistencies in the
published literature and make the use of K values more reliable.
ow

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

157

T i m e (days)
FIGURE 3.18

Distances traveled by the centers of mass of various plumes as a function of time since

injection. Source-. P. V. Roberts, M. N. Goltz, and D. M. Mackay, Water Resources Research 2 2 , no. 13
(1 9 8 6 ) : 2 0 4 7 - 5 9 . Copyright by the American Geophysical Union.

3.11

SUMMARY
T h e advection-dispersion equation can be modified to reflect the effects of solutes that
are removed from solution by sorption, chemical reaction, and biological and radioactive
decay. Sorption may o c c u r due to adsorption, chemisorption, ion e x c h a n g e , and absorption. Sorption of inorganic solutes o c c u r s primarily on mineral surfaces and is a
function of the surface area available for sorption as well as the n u m b e r of i o n - e x c h a n g e

158

Chapter T h r e e

sites provided by clay and o x i d e minerals. T h e sorption of inorganic ions and organic
chemical solutes can be quantified by experimentali)- derived adsorption isotherms.
Organic solutes may be s o r b e d on either soil mineral surfaces or by soil or aquifer
organic matter. Nonpolar organic c o m p o u n d s are s o r b e d to a greater extent by soil
organic matter than by mineral surfaces. T h e affinity of a particular organic molecule to
be s o r b e d by soil or aquifer organic matter can be estimated from either the octanolwater partition coefficient, the aqueous solubility, or the molecular structure.
T h e advection-dispersion equation can also be modified to account for the disappearance of solutes or the appearance of a product due to other chemical reactions.
S o m e of these reactions can be described by equilibrium reactions; others need to be
dealt with on the basis of the kinetics of the reaction. A term to describe the disappearance
of a solute due to radioactive decay can also be appended to the advection-dispersion
equation. T h e biological degradation of organic c o m p o u n d s can o c c u r under both
a e r o b i c and anaerobic conditions. Biodegradation terms can be joined to the advectiondispersion equation.

Chapter Notation
[A\

Concentration of ion in solution


Bulk density of soil

b
C
C*
Q
Q

Microbial decay rate


Concentration of solute in liquid phase
Amount of solute s o r b e d per unit weight of soil
Equilibrium molar concentration of the ith c o m p o n e n t
Solute concentration in immobile phase
Solute concentration in mobile phase
A b s o r b e d concentration in contact with mobile phase
A b s o r b e d concentration in contact with immobile phase

Q,
C
<S

oc

D
D

Fi
F
T

f
L

Natural organic c a r b o n concentration


Longitudinal dispersion coefficient
Apparent diffusion constant for mobile phase
x-direction mass flux of a solute
Total mass flux of chemical entity
Fraction of solid surfaces in contact with mobile phase
Fraction of soil that consists of organic c a r b o n

Hmin

Critical level of soil organic c a r b o n


Ratio of oxygen to hydrocarbon c o n s u m e d
Hydrocarbon concentration in p o r e fluid
Minimum hydrocarbon concentration to support microbial growth

K
K

Maximum hydrocarbon utilization rate by a e r o b i c m i c r o b e s


Maximum hydrocarbon utilization rate by anaerobic m i c r o b e s

Coefficient in Freundlich sorption isotherm


Half maximum rate concentration for hydrocarbon for anaerobic

G
II

decay

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

Distribution coefficient
H y d r o c a r b o n half-saturation ratio

O x y g e n half-satu ration ratio

Koc

Distribution coefficient for soil organic c a r b o n


Distribution coefficient for soil organic matter

Kw

O c t a n o l water partition coefficient

K
k

First-order decay rate for natural organic c a r b o n

Kinetic rate constants (i = 1 to 8)


Liter

m
K

G l o b a l mass o f chemical entity


Total mass o f a n a e r o b i c m i c r o b e s

M
M,
N
O
Pf
Q

C o n c e n t r a t i o n o f a e r o b i c m i c r o b e s i n solution

PA

Reaction rate for disappearance of A

>y
f/
fi
h
>m
S
Sa

Retardation factor for linear sorption isotherm

Total a e r o b i c microbial mass


Coefficient in Freundlich sorption isotherm
O x y g e n c o n c e n t r a t i o n in p o r e fluid
Polarity c o r r e c t i o n factor
A--direction f l u x o f water

Retardation factor for Freundlich sorption isotherm

Retardation factor for Langmuir sorption isotherm

Retardation factor for h y d r o c a r b o n


Retardation factor for m i c r o b e s
Solubility of chemical in water
Surface area o f soil

5,
5

Water solubility of t'th c o m p o u n d

Time

S o u r c e / s i n k term for c h e m i c a l entity


Average linear velocity of solute front

Ground-water velocity in m o b i l e p h a s e

Average linear g r o u n d water velocity

Linear o p e r a t o r defined in Equation 3 . 5 0


Mole fraction of t'th c o m p o u n d in organic p h a s e

Microbial yield coefficient

y.

Absorption constant for Langmuir sorption isotherm

Maximum solute sorption from Langmuir sorption isotherm

First o r d e r rate coefficient

First-order mass-transfer coefficient

1 lalf life of radionuclide


N u m b e r of adjacent nonhydrogen atoms in an organic m o l e c u l e

u
l

X
9

Volumetric moisture c o n t e n t

Porosity o c c u p i e d by m o b i l e p h a s e

First o r d e r molecular connectivity index

Porosity o c c u p i e d by i m m o b i l e p h a s e
K,

K\

K"

Reaction-rate constants

159

160

Chapter T h r e e

References
Anderson, M. P. 1979. Using models to simulate the movement of contaminates through groundwater flow systems. CRC Critical Ret iews in Ent ironmental Control
9, no. 2 : 9 7 - 1 5 6 .
Banerjee, Sujit. 1984. Solubility of organic mixtures in water.
Environmental Science
and
Technology
18,
no.
8:587-91.
Briggs, G. G. 1981. Theoretical and experimental relation
si ,ips between soil adsorption, octanol-water partition
coefficients, water solubilities, bioconcentration factors, and the parachor. Journal of Agriculture and
Food Chemistry. 2 9 : 1 0 5 0 - 5 9 .
Bomemisza, E and R. Llanos. 1967. Sulfate movement, sorption and desorpuon in three Costa Rica soils. Soil
Science Society of America Proceedings 3 1 : 3 5 6 - 6 0 .
Borden, R. C. and P. B. Bedient. 1986. Transport of dissolved
hydrocarbons influenced by oxygen-limited biodegradation, 1. Theoretical development. Water Re
sources Research 22, no. 1 3 : 1 9 7 3 - 8 2 .
Borden. R. C.. P. B. Bedient, M. D. Lee, C. H. Ward, and J. T.
Wilson. 1986. Transport of dissolved hydrocarbons
influenced by oxygen-limited biodegradation, 2. Field
applications. Water Resources Research 22, no.
13:1983-90.
Bouwer, E. J . , and P. L McCarty. 1984. Modeling of trace
organics biotransformation in
the subsurface.
Ground Water 22, no. 4 : 4 3 3 - 4 0 .
Bowman, B. T., and W. W. Sans. 1977. Adsorption of para
thion, fenitrothion, methylparathion, aminoparathion
and paraoxon by Na , C a , and F e
montmorillonite suspensions. Soil Science Society of America
Journal 4 1 : 5 1 4 - 1 9 .
+

2 +

3 +

Ciccioli, P., W. T. Cooper, P. M. Hammer, and J. M. Hayes.


1980. Organic solute mineral surface interactions:
A new method for the determination of groundwater velocities. Water Resources Research 1 6 , no.
1:217-23.
Chiou, C. T., L. J. Peters, and V. H. Freed. 1979. A physical
c o n c e p t of soil-water equilibria for nonionic organic
compounds. Science 206, no. 1 6 : 8 3 1 - 3 2 .
Chiou, C. T., P. E. Porter, and D. W. Schmedding. 1983.
Partition equilibrium of nonionic organic compounds
between soil organic matter and water. Environmen
tat Science and Technology 17, no. 4 : 2 2 7 - 3 1 .
Davidson, J. M., and R. K. Chang. 1972. Transport of picloram
in relation to soil physical conditions and pore water
velocity. Soil Science Society of America Proceedings
36:257-61.
Dzombak. D. A, and R. G. Luthy. 1984. Estimating adsorption
of porycyclic aromatic hydrocarbons on soils. Soil Sci
ence 137, no. 5 : 2 9 2 - 3 0 8 .
Ellgehausen, H., C. D'Hondt, and R. Fuerer. 1981. Reversedphase chromatography as a general method for determining octan-1 -oL water partition coefficients.
Pesticide Science 1 2 : 2 1 9 - 2 7 .
Enfield, C. G. 1985. Chemical transport facilitated by multi
phase flow systems. Water Science and Technology
17:1-12.

Enfield, C. G., and B. E. Bledsoe. 1975. Fate of wastewater


phosphorus in soil. Journal of Irrigation and Drainage Dit ision, American Society of Cit il Engineers 101,
no. I R 3 : 1 4 5 - 5 5 .
Fntield, C. G., and G. Bengtsson. 1988. Macromolecular trans
port of hydrophobic contaminants in aqueous envi
ronments. Ground Water 26, no. 1 : 6 4 - 7 0 .
Faust, C. R., and J. W, Mercer. 1980. Ground water model
ing: recent developments. Ground Water 18, no.
6:569-77.
Fetter, C. W. 1977. Attenuation of waste water elutriated
through glacial outwash. Ground Water 15, no.
5:365-71.
F i s k e l l J . G. A, R. S. Mansell, H. M. Selim, and F. G. Martin
1979. Kinetic behavior of phosphate sorption by
acid, sandy soil. Journal of Environmental Qttality
8:579-84.
Freyberg, D. L 1986. A natural gradient experiment on solute
transport in a sand aquifer, 2. Spatial movements and
the advection and dispersion of nonreactive tracers.
Water Resources Research 22, no. 1 3 : 2 0 3 1 - 4 7 .
Griffin, R. A 1991. Personal communication.
Griffin, R. A, and E. S. K. Chian. 1980. Attenuation of watersoluble polychlorinated biphenyls by earth materials.
Environmental Geology Note 86, Illinois State Geological Survey.
Harvey, R. W, R L Smith, and L George. 1984. Effect of
organic contamination upon microbial distribution
and heterotrophic uptake in a Cape Cod, Mass., aquifer.
Applied
and
Environmental
Microbiology
48:1197-1202.
Hassett, J. J . , J. C. Means, W. L Banwart, and S. G. Wood.
1980
Sorption properties of sediments and energyrelated pollutants. U S . Environmental Protection
Agency, EPA-600/3-80 0 4 1 .
Hassett, J. J . , W. L Banwart, and R. A Griffin. 1983- Correlation
of compound properties with sorption characteristics
of nonpolar compounds by soils and sediments: Concepts and limitations. In Environment and Solid
Wastes.- Qjaracterization. Treatment and Disposal,
ed. C. W. Francis and S. I. Auerback, 1 6 1 - 1 7 8 . Boston:
Butterworth Publishers.
Havward, S. B. 1984. Field monitoring of chrysotile asbestos
in California waters. Journal, American Water Works
Association 76, no. 3 : 6 6 - 7 3 .
Hornsby, A G., and J. M. Davidson. 1973. Solution and ad
sorbed tluometuron concentration distribution in a
water saturated soil: Experimental and predicted evaluation. Soil Science Society of America Proceedings.
37:823-28.
Hunt, J. R., N. Sitar, and K. S. Udell. 1988. Nonaqueous phase
liquid transport and cleanup. 1. Analysis of mecha
nisms. Water Resources Research 24, no. 8 : 1 2 4 7 - 5 8 .
Jarrell, W. M. and M. D. Dawson. 1978. Sorption and availability of molybdenum in soils of western Oregon.
Soil Science Society of America Journal -12:412-15.
Karickhoff, S. W, D. S. Brown, and T. A Scott. 1979- Sorption
of hydrophobic pollutants on natural sediments.

T r a n s f o r m a t i o n , Retardation, and Attenuation of S o l u t e s

Water Research 1 3 : 2 4 1 - 4 8 .
KarickhotT, S. W. 1981. Semi empirical estimation of sorption
of hydrophobic pollutants on natural sediments and
soils. atmosphere 1 0 : 8 3 3 - 4 6 .
Karickhoff, S. W. 198-4. Organic pollutant sorption in aquatic
systems. Journal of Hydraulic Engineering 110, no.
6:707-35.
Kenaga, E. E., and C. A I. Goring. 1980. "Relationship between
water solubility, soil sorption, octanol-water partitioning and bioconcentration of chemicals in biota." In
3rd Aquatic
Toxicology Symposium,
Proceedings
of tite American Society of Testing and Materials,
7 8 - 1 1 5 . No. STP 707.
Kenaga. E. E. 1980. Predicted bioconcentration factors and
soil sorption coefficients of pesticides and other
chemicals. Ecotaxicology and Environmental Safety
4:26-38.
Keswick, B. H., D. S. Wang, and C. P. Gerba. 1982. T h e use
of microorganisms as ground water tracers: A review.
Ground Water 20, no. 2 : 1 4 2 - 4 9 .
Koch, R. 1983. Molecular connectivity index for assessing
ecotoxicological behavior of organic compounds.
Toxicotogical
and
Environmental
Qwrnistry
6:87-96.
Leistra, M., and W. A Dekkers. 1977. S o m e models for the
adsorption kinetics of pesticides in soils. Journal of
Environmental Science and Health B 1 2 : 8 5 - 1 0 3 .
Lyman, W. J. 1982. "Adsorption coefficient for soils and sediment.'
In
Handbookof ChemicalPropertyEstimation
Methods, ed. W. J. Lyman et a l , 4 . 1 - 4 . 3 3 . New York:
McGrawHill.
Mackay. D. M., D. L Freybetg, P. V. Roberts, and J. A. Cherry.
1986. A natural gradient experiment on solute transport in a sand aquifer, 1. Approach and overview of
plume movement. Water Resources Research 22, no.
13:2017-30.
Major, D. W., C. I. Mayfield, and J. F. Barker. 1988. Biotrans
formation of benzene by denitification in aquifer sand.
Ground Water 26, no. 1 : 8 - 1 4 .
McCall, P. J . , R. L Swann, and D. A. Laskowski. 1983. Partition
Models for Equilibrium Distribution of Chemicals in
Environmental Compartments. In Fate of Chemicals
in the Environment, ed. R. L Swann and A. Eschen
roder, 1 0 5 - 2 3 . American Chemical Society.
McCarthy, J. F.. and J. M. Zachara. 1989. Subsurface transport
of contaminants. Environmental Science and Technology 23, no. 5 : 4 9 6 - 5 0 2 .
McCarty, P. L, M. Reinhard, and B. E. Rittman. 1981. Trace
otganics in groundwater. Environmental Science and
Technology 15, no. 1 : 4 0 - 5 1 .
Means, J. C, S. G. Wood, J. J. Hassett, and W. L Ban wan.
1980. Sorption of polynuclear aromatic hydrocarbons
by sediments and soils. Environmental Science and
Technology 14, no. 1 2 : 1 5 2 4 - 2 8 .
Miller, C. T., and W. J. Weber, Jr. 1984. Modeling organic
contamination partitioning in ground-water systems.
Ground Water 22, no. 5 : 5 8 4 - 9 2 .
Mingelgrin. {'., and Z. Gerstl. 1983- Reevaluation of partition
ing as a mechanism of nonionic chemicals adsorption
in soils. Journal of Environmental Quality 1 2 : 1
11.
Munns, D. N., and R. L F o x . 1976. T h e slow reactions that

161

continue after phosphate adsorption: kinetics and


equilibrium in tropical soils. Soil Science Society of
America
Proceedings 4 0 : 4 6 - 5 1 .
Nielsen, D. R., M. Th. van Genuchten and J. W. Biggar. 1986.
Water flow and solute transport in the unsaturated
zone. Water Resources Research 22, no. 9 : 8 9 S - 1 0 8 S .
Nightingale, H. I., and W. C. Bianchi. 1977. Ground-water
turbidity resulting from artificial recharge. Ground
Water 15, no. 2 : 1 4 6 - 5 2 .
Nkedi Kizza, P., J. W. Biggar, H. M. Selim, M. Th. van Genuchten, P. J. Wierenga, J. M. Davidson, and D. R.
Nielsen. 1984. On the equivalence of two conceptual
models for describing ion exchange during transport
through an aggregated oxisol. Water Resources Re
search 20, no. 8 : 1 1 2 3 - 3 0 .
Olsen, R. L, and A. Davis. 1990. Predicting the Fate and
Transport of Organic Compounds in Groundwater:
Part 1. Hazardous Materials Control 3, no. 3 : 3 8 - 6 4 .
Prickett, T. A., T. C. Naymik, and C. G. Lonnquist. 1981. A
"random walk" solute transport model for selected
groundwater quality evaluations. Illinois State Water
Survey, Bulletin 6 5 , 103pp.
Rao, P. S. C, and J. M. Davidson. 1980. Estimation of pesticide
retention and transformation parameters required in
nonpoint source pollution models. In Environmental
Impact of Nonpoint Source Pollution, ed. M. R. Overcash and J. M. Davidson, 2 3 - 6 7 . Ann Arbor, Mich.:
Ann Arbor Science Publishers, Inc.
Roberts, P. V., M. N. Goltz, and D. A. Mackay. 1986. A natural
gradient experiment on solute transport in a sand
aquifer, 3. Retardation estimates and mass balances
for organic solutes. Water Resources Research 22, no.
13:2047-59.
Rogers, R. D., J. C. McFarlane, and A. J. Cross. 1980. Adsorp
lion and desorption of benzene in two soils and
montmorillonite clay.
Environmental Science and
Tedmology 14, no. 4 : 4 5 7 - 6 0 .
Roy, W. R and R. A. Griffin. 1 9 8 5 . Mobility of organic solvents
in water-sat united soil materials. Environmental Geology and Water Sciences l:24\-47.
Rubin, J a c o b . 1983. Transport of reacting solutes in porous
media: Relationship between mathematical nature of
problem formation and chemical nature of reactions.
Water Resources Research 19, no. 5 : 1 2 3 1 - 5 2 .
Sabljic, A., and M. Protic.
lecular connectivity
cients of poh/cyclic
of Environmental
28:162-65.

1982. Relationship between moindices and soil sorption coeffiaromatic hydrocarbons. Bulletin
Contamination
and
Toxicology

Sabljic, A. 1984. Predictions of the nature and strength of


soil sorption of organic pollutants by molecular topology. Journal of Agricultural and Food Chemistry
32:243-46.
Sabljic, A. 1987. On the prediction of soil sorption coefficients
of organic pollutants from molecular structure; Application of molecular topology model. Environmental Science and Technology 2 1 , no. 4 : 3 5 8 - 6 5 .
Schwar/.enbach, R. P., and J. Westall. 1981. Transport of nonpolar organic compounds from surface water to
groundwater: Laboratory sorption studies. Environmental Science and Tedmology 15, no. 1 1 : 1 3 6 0 - 6 7 .

1 62

Chapter T h r e e

Sidle, R. C, L T. Kardos, and M. Th. van Genuchten. 1977.


Heavy metal transport model in a sludge treated soil.
Journal of Environmental Quality 6:43843.
Srinivasan, P.. and J. W. Mercer. 1988. Simulation of biodeg
radation and sorption processes in ground water.
Ground Water 26, no. 4 : 4 7 5 - 8 7 .
Staples, C. A., and S.J. Geiselmann. 1988. Cosolvent influences
on organic solute retardation factors. Ground Water
26, no. 2 : 1 9 2 - 9 8 .
Street, J. J . , W. L Lindsay, and B. R Sabey. 1977. Solubility
and plant uptake of cadmium in soils amended with
cadmium and sewage sludge. Journal of Environ
mental Quality 6 : 7 2 - 7 7 .
Travis, C C.. and E. L Etnier. 1 9 8 1 . A survey of sorption
relationships for reactive solutes in soil. Journal of
Environmental Quality 10, no. 1 : 8 - 1 7 .
Van Blade), R. and A. Moreale. 1977. Adsorption of herbicide
derived p chloroaniline residues in soils: A predictive
equation. Journal of Soil Science 2 8 : 9 3 - 1 0 2 .
Van Genuchten, M. Th., J. M. Davidson, and P. J. Wierenga.

1974. An evaluation of kinetic and equilibrium equa


tions for the prediction of pesticide movement
through porous media. Soil Science Society of 'America Proceedings 3 8 : 2 9 - 3 5 .
Van Genuchten, M. Th. and P.J. Wierenga. 1976. Mass transfer
studies in sorbing porous media, 1. Analytical solutions.
Soil Science Society of America Journal
40:473-80.
Walsh. M. P., S. L Bryant, R. S. Schechier, and L W. Lake.
1984. Precipitation and dissolution of solids attend
ing How through porous media. American Institute
of Chemical Engineers Journal 30, no. 2 : 3 1 7 - 2 8 .
Wood, W. W.. T. P. Kramer, and P. P. Hem, J r . 1990. Inter
granular diffusion: An important mechanism influencing solute transport in clastic aquifers. Science2-V?
(March 3 0 ) : 1 5 6 9 - 7 2 .
Yaron, B. 1978. Some aspects of surface interactions of clays
with organophosphorous pesticides. Soil Science
125:210-16.

Chapter Four

Flow and Mass Transport


in the Vadose Zone

4.1

Introduction
Flow through the vadose z o n e is a topic that m o s t hydrogeology texts tend to c o v e r in
a c u r s o r ) ' manner. Classical hydrogeology is c o n c e r n e d primarily with obtaining water
from wells. T h e v a d o s e , or water unsaturated z o n e , is s e e n as a s o m e w h a t mysterious
realm through which recharge water must pass on the way to the water t a b l e a watery
purgatory. Soil scientists have b e e n the primary force behind developing an understanding
of unsaturated z o n e flow. Historically, soil scientists w e r e c o n c e r n e d with such topics
as the passage of water and solutes to roots of plants, water that flows primarily in the
vadose zone. More recently soil scientists have studied the transport and fate of contaminants in the vadose zone. With the development of the s c i e n c e of contaminant
hydrogeology, hydrogeologists have b e c o m e m u c h m o r e interested in the mysteries of
the vadose z o n e . Many releases of contaminants to the subsurface o c c u r within or a b o v e
the vadose z o n e . Contaminants are u n d e r s t o o d to include materials applied deliberately
to the soil, such as fertilizers and pesticides, as well as t h o s e released accidentally. T h e
hydrogeologist is suddenly faced with the daunting task of understanding the transport
o f dissolved contaminants through the vadose zone. T o this end, m u c h can b e learned
from the w o r k of soil scientists. Transport in the vadose z o n e may also o c c u r by flow
of a pure n o n a q u e o u s phase liquid or g a s e o u s phase.
T h e vadose z o n e e x t e n d s from the land surface to the water table. It includes the
capillars' fringe, w h e r e p o r e s may actually be saturated. T h e main distinguishing feature
of the vadose z o n e is that the p o r e water pressures are generally negative. Areas of the
vadose z o n e a b o v e the capillary fringe may temporarily be saturated due to surface
ponding o f water o r b e c a u s e o f the d e v e l o p m e n t o f p e r c h e d water tables a b o v e relatively
low permeability soil layers.

4.2

Soil as a Porous Medium


T h e vadose z o n e includes the soil layers at the Earth's surface. It may also include
sediment a n d / o r consolidated rock. I lowever, the p r e s e n c e of soil c o m p l i c a t e s the study
of vadose z o n e hydrology, also known as soil physics.
163

164

Chapter F o u r

Soil is a c o m p l e x material. In physical form it consists mostly of mineral grains of


varying size as well as varying amounts of organic matter. T h e mineral grains are arranged
in such a fashion that the soil has structure; that is, there is a specific orientation and
arrangement of the individual grains. T h e individual grains usually form larger units called
1

a g g r e g a t e s , " p e d s , which are b o u n d b y organic matter (e.g., Hillel 1 9 8 0 ) . T h e porosity


and permeability of the soil is a function of both soil texture and the soil structure. T h e
stnicture is a function of the physical shape and size of the aggregates. Moreover, it may
be vastly influenced by the soil chemistry, since soil minerals have an electrical charge
on their surface. T h i s surface charge, which is primarily due to the clay minerals, affects
the stability of soil structural units. Soil contains mineral matter, organic matter, water
containing dissolved solutes, and gases. T h e soil also has macropores, such as root casts
and w o r m h o l e s , and drying cracks in fine-textured soils. T h e s e form preferential channels
for water movement.
T h e amount of moisture in a soil can be e x p r e s s e d as the gravimetric water content,
which is the weight of the water as a ratio to the weight of the dry soil mass. T h e
moisture state can also be e x p r e s s e d as the volumetric water content, 8, which is the
volume of water as a ratio to the total volume of the soil mass O n e must be careful in
measuring volumetric water content, since in many soils (especially those with fine
t e x t u r e ) the volume changes as water is i m b i b e d or drained. This is due to the interactions
b e t w e e n the charged soil particles and the polar water molecules.

4.3

Soil Colloids
T h e clay fraction of the soil consists of mineral particles that are less than 2 ^tm in
diameter (Hillel 1 9 8 0 ) . Clay particles consist primarily of secondary minerals that have
b e e n formed by weathering. Clay minerals have an unbalanced negative electrical charge
at the surface Electrostatic attraction exists between the surface of the clay particles,
the polar soil-water molecules, and solutes dissolved in the soil water. Fine-grained
materials with an electrostatic surface charge are called c o l l o i d s .
Clay minerals have a definite crystal structure, consisting primarily of aluminum,
silica, and oxygen. K a o l i n i t e (e.g., A l S i O
4

1 0

( O H ) ) is a clay mineral with a low specific


8

surface (surface area per unit m a s s ) ; it ranges from 5 to 20 m / g . T h e low specific surface
m e a n s that kaolinite is not particularly reactive. M i t e (e.g., A I S i A l O ( O H ) K , with
4

2 0

0 8

the potassium occurring between layers) has a larger specific surface area, ranging from
2

8 0 t o 1 2 0 m / g . M o n t m o r i l l o n i t e (e.g., A l M g
3 5

0 5

Si O
8

2 0

( O H ) ) i s the most reactive


4

clay with a specific surface area of 7 0 0 to 8 0 0 m / g . T h e reactive clays can a b s o r b large


amounts of water and ions between their sheetlike mineral grains. This property gives
soils high in reactive clays the ability to swell as water is absorbed. It also means that
they shrink and crack when dried. Montmorillonite has the largest shrink-swell behavior,
and kaolinite has the least. C h l o r i t e (e.g., M g S i A l O ( O H ) , with M g ( O H )
6

2 0

1 2

oc

curring between the layers) is a n o t h e r c o m m o n clay mineral with a behavior similar to


illite.
Clay-size particles may also include s e s q u i o x i d e s , which are hydrated aluminum
and iron o x i d e s ( A 1 0 - n H 0 and F e 0 - n I 1 0 ) . Limonite, goethite, and gibbsite are
2

Flow and Mass Transport in the Vadose Zone

165

e x a m p l e s . T h e s e s u b s t a n c e s are generally a m o r p h o u s and have less electrostatic properties than the silicate clay minerals. S e s q u i o x i d e s often act as c e m e n t i n g agents for soils.
Soil may also contain d e c o m p o s e d organic material, which is s o m e t i m e s called
h u m u s . Humus is a c o m p l e x mixture of organic m o l e c u l e s that are aggregated into
colloidal-size particles. Humus particles are also negatively charged. Humus is found
primarily in the A horizon of soils. T h e h u m u s c o n t e n t of mineral soils can range from
0 to about 1 0 % by weight. T h e organic c o n t e n t of organic soils, such as peat, can be
u p t o 5 0 % o r m o r e b y weight.

4.4

T h e Electrostatic Double Layer


T h e colloidal p a n i c l e s of the soil have an unbalanced, negative surface charge. T h i s
negative charge is balanced by positively charged cations that are attracted to the surface
of the colloid T h e s e cations exist as solutes in water. W h e n the colloid is dry, the layer
of water held to the surface will be thin, and the neutralizing cations will be closely held
to the particle surface. As the colloid b e c o m e s m o r e hydrated, the cations will dissociate
from the surface and form a swarm of ions near the negatively charged surface layer.
T h e cations have a m o r e or less fixed position near the negatively charged particle
surface. T h e particle surface and the cation swarm form what is k n o w n as an e l e c t r o static double layer.
T h e negatively charged particle surface will tend to repel anions. Flence, the region
near the particle surface will have an a b u n d a n c e of cations and relatively few anions.
Figure 4.1 s h o w s the distribution of monovalent cations and anions near the surface of
a montmorillonite particle. T h e effects of the electrostatic d o u b l e layer e x t e n d to the
distance at which the n u m b e r of cations in solution equals the n u m b e r of anions. Figure
4.1 shows that the electrostatic double-layer effect is a function of the solute concentration
and e x t e n d s farther in m o r e dilute solutions. T h e e x t e n t of the electrostatic d o u b l e layer
is less for divalent cations as o p p o s e d to monovalent cations. T h e thickness of the
electrostatic d o u b l e layer can be c o m p u t e d from:

(4.1)
where
z = the characteristic thickness of the d o u b l e layer
0

e = the elementary charge of an ion, 4.77 x 1 0 "

1 0

esu

e = the dielectric constant


k = the Boltzmann constant ( t h e ratio of the gas constant to Avagadro's
B

number)
V = the valence of the ions in solution
3

Ho = the c o n c e n t r a t i o n of ions in bulk solution ( i o n s / c m )


T = temperature in Kelvins

166

Chapter F o u r

T h e electrostatic d o u b l e layer is very important in contaminant hydrogeology,


b e c a u s e the cations associated with this layer can be replaced by other cations in sc ilutii m.
This replacement p r o c e s s i s known a s c a t i o n e x c h a n g e . T h e total n u m b e r o f positive
charges that can be e x c h a n g e d in a soil is independent of the cation species and is
e x p r e s s e d as the cation e x c h a n g e capacity. Cation e x c h a n g e affects the transport of ions
in solution. T h e type and concentration of cations involved in the e x c h a n g e process can
also affect the hydraulic conductivity of the soil.
S o m e colloids can also attract anions. Although the surface of clay particles carries
a negative charge, the edges usually carry a net positive charge. Since the surface area
far e x c e e d s the e d g e area, the cation-exchange capacity for most soils far e x c e e d s the
anion-exchange capacity.

Kaolinite

and

humus

have

the greater

anion e x c h a n g e

capacity.
Different cations are held with greater tenacity by the colloids. T h e larger the ionic
radius and the greater the valence charge, the m o r e tightly the cation is held. T h e ionic

F l o w ond M a s s T r a n s p o r t in the V a d o s e Z o n e

167

radius of the cation is affected by hydration, b e c a u s e polar water m o l e c u l e s are attracted


to the ion. It is the radius of the hydrated ion that is important. T h e normal o r d e r of
preference for cation e x c h a n g e is
Al

+ 3

Ca

+ 2

> Mg

+ 2

NH

+
4

> K

> H 0

> Na

> Li

However, if a soil is flooded with a solution containing a large c o n c e n t r a t i o n of o n e


cation, the normal c a t i o n - e x c h a n g e o r d e r can be reversed.

4.5

Salinity Effects on Hydraulic Conductivity of Soils


T h e hydraulic conductivity of a soil can be affected by the strength and type of cations
contained in the soil water (Nielsen, van G e n u c h t e n , and Biggar 1 9 8 6 ) . T h e impact of
the solute increases with the a m o u n t of colloidal particles in the soil. Soil swelling c a u s e d
by increased salinity can reduce hydraulic conductivity. As the electrostatic d o u b l e layer
grows thicker, the hydraulic conductivity decreases, b e c a u s e clay minerals tend to swell
and e x p a n d into the p o r e space. Sodium is especially important in this p r o c e s s . T h e
electrostatic d o u b l e layer is thicker w h e n it contains the monovalent sodium ions, and
as such sodium tends to w e a k e n the b o n d s b e t w e e n clay particles. T h e effect of swelling
is reversible if the saline water is flushed from the p o r e s . However, if smaller particles
break l o o s e from the soil structure, they can be transported by flowing water until they
are carried into small p o r e throats, w h e r e they can lodge. T h i s causes a m o r e or less
irreversible reduction in hydraulic conductivity ( D a n e and Klute 1 9 7 7 ) .
Examination of Equation 4.1 s h o w s that the electrostatic d o u b l e layer grows with
decreasing c o n c e n t r a t i o n of cations in the soil water. It will also d e c r e a s e with an increase
in the ratio of the c o n c e n t r a t i o n of monovalent to divalent cations in the soil water. T h e
principal cations in m o s t natural waters are sodium, calcium, and magnesium. T h e sodium
adsorption ratio (SAR) is a m e a s u r e of the ratio of the c o n c e n t r a t i o n of monovalent
sodium to divalent calcium and magnesium:
Na
SAR =

(4.2)
<Ca + Mg

where
Na = c o n c e n t r a t i o n of sodium in milliequivalents per liter
Ca = concentration of calcium in milliequivalents per liter
Mg = concentration of magnesium in milliequivalents per liter
T h e greater the SAR of the soil water and the lower the total solute concentration,
the lower the hydraulic conductivity of the soil, especially if it contains the expansive
clays illite and montmorillonite. Figure 4.2 s h o w s the effect for o n e soil of decreasing
solute concentration on the unsaturated hydraulic conductivity of a soil. During these
tests the SAR was kept at 4 0 , so that the relative p r o p o r t i o n s of sodium, calcium, and
magnesium did not change. An increase in t h e p H of soil water has also b e e n demonstrated
to result in a reduction of hydraulic conductivity for s o m e soils (Suarez 1 9 8 5 ) .

168

Chapter Four

0.35
1

I (cmVcm- )
F I G U R E 4.2

Hydraulic conductivity as a function of volumetric water content for solutions of varying

solute concentration but with a constant SAR value of 4 0 . Source: J. H. Dane and A. Klute, Soil Science
Society of America Journal 4 1 , no. 6 (1 977): 1 04357.

4.6

F l o w of Water in the Unsaturated Zone


V a d o s e z o n e hydrology is different from saturated zone hydrology b e c a u s e of the prese n c e of air in the p o r e space. T h e relative proportion of air and water in the pores can
vary, and with it can vary the hydraulic properties of the p o r o u s media.
4.6.1

Soil-Water Potential

In saturated flow the driving potential for groundwater flow is due to the pore-water
pressure and elevation a b o v e a reference datum (Fetter 1 9 8 8 ) . However, in unsaturated
flow the p o r e water is under a negative pressure caused by surface tension. Soil physicists
call this the c a p i l l a r y p o t e n t i a l , or m a t r i c p o t e n t i a l , t/r; it is a function of the

Flow and Mass Transport in the Vadose Zone

169

volumetric water c o n t e n t of the soil, 0. T h e lower the water content, the loweri.e.,
m o r e negativethe value of the matric potential.
T h e total soil-moisture potential, <p, is the sum of the matric potential, i/r, a pressure
potential, the gravitational potential, Z, an o s m o t i c potential, and an electrochemical potential. However, we will a s s u m e that the o s m o t i c potential and the e l e c t r o c h e m i c a l
potential do not vary within the soil and that the pressure is equal to atmospheric. S i n c e
we will eventually want to find the gradient of the potential, we can neglect o s m o t i c
pressure and e l e c t r o c h e m i c a l potentials, b e c a u s e their gradient will be zero. Total soil
moisture potential is therefore reduced to the sum of the matric and gravitational
potentials:

4> = 4,(6) + Z

(4.3)

Matric potential may be m e a s u r e d as a capillary pressure, P , which has the units


c

of newtons p e r square meter, which are equivalent to joules per c u b i c m e t e r or energy


2

per unit volume ( L M ~ ' T ~ ) . If the matric potential is measured on a pressure basis,
then the gravitational potential, Z, is equal to p gz, w h e r e g is the acceleration of gravity,
w

p is the density of water, and z is the elevation a b o v e a reference plane. T h e total soil
moisture potential in terms of energy per unit volume can thus be found from

4>EV = P + Pgz

(4-4)

If Equation 4.4 is divided by pg, the result is the soil moisture potential e x p r e s s e d
as energy per unit weight, which also has units of length ( L ) . This is equivalent to h e a d
in saturated flow. T h e matric potential is also e x p r e s s e d in units of length, typically
centimeters o f water:

4> = +z = b + z
EW

(4.5)

Pg
w h e r e h, pressure potential, is the matric potential in units of length.
Dividing Equation 4.4 b y p gives the soil moisture potential e x p r e s s e d a s energy
w

per unit mass, with units o f joules per kilogram ( L T ~ ) :

4>EM = + gz
Pw

(4.6)

C o m m o n units for total potential and pressure potential include a t m o s p h e r e s of


pressure and centimeters of water. O n e a t m o s p h e r e is equivalent to about 1 0 0 0 cm of
s

water. Also, 1 0 pascals of pressure is equal to about 1 a t m o s p h e r e ( a t m ) .


4.6.2

S o i l - W a t e r Characteristic Curves

T h e relationship b e t w e e n matric potential or pressure head and volumetric water c o n t e n t


for a particular soil is known as a s o i l - w a t e r c h a r a c t e r i s t i c c u r v e , or a soil water
retention curve. Figure 4.3 s h o w s an idealized soil-water characteristic curve.
At a t m o s p h e r i c pressure the soil is saturated, with the water c o n t e n t equal to 0 .
S

T h e soil will remain saturated as the matric potential is gradually decreased. Eventually,
the matric potential will b e c o m e negative e n o u g h that water can begin to drain from
the soil. T h i s matric potential is known as the b u b b l i n g p r e s s u r e . It is marked h on
b

170

Chapter F o u r

0.2

0.3

0.4
3

0.5

0.6

Water content. 6 (cm /cm )


FIGURE 4.3

Typical soil-water retention curve. Source: M. T h . van Genuchten, Soil Science Society of

America Journal 44 (1 9 8 0 ) : 8 9 2 - 9 8 .

Figure 4.3. T h e moisture c o n t e n t will continue to decline as the matric pressure is lowered,
until it reaches s o m e irreducible minimum water content, 9 . Should the matric potential
r

be further reduced, the soil would not lose any additional moisture.
T h e soil-moisture characteristic curve also shows the pore-size distribution of the
soil. Figure 4.4 shows idealized soil-moisture characteristic curves for two soils, o n e of
which is well sorted and o n e of which is poorly sorted.
T h e well sorted soil has a narrower range of matric potential over which the water
content changes than the poorly sorted soil. In a well-sorted soil, most of the grains are
in a narrow size range, and h e n c e the p o r e s also have a narrow size range. T h e poorly
sorted soil has a wider size range for both grains and pores. T h e well-sorted soil has a
higher bubbling pressure b e c a u s e it has larger pores. However, o n c e the well-sorted
soil begins to unsaturate, it d o e s so rapidly, again b e c a u s e most of the pores are large.

F l o w and M a s s T r a n s p o r t in the Vadose Z o n e

171

172

Chapter F o u r

T h e r e are s o m e simple empirical e x p r e s s i o n s that can be used to relate the water


c o n t e n t of a soil to the matric potential. B r o o k s and Corey ( 1 9 6 6 ) used the relationship
(4.7)
where
9 = volumetric water content
0 = volumetric water content at saturation
S

6 = irreducible minimum water content


r

t/r = matric potential


h = bubbling pressure
b

X = experimentally derived parameter


B r o o k s and Corey ( 1 9 6 6 ) also defined an e f f e c t i v e s a t u r a t i o n , S , as
e

(4.8)
w h e r e S = the saturation ratio, 9/0 .
w

A graph of capillary pressure divided by the specific weight (P /y) of the fluid
c

versus the effective saturation ( 5 ) is shown for water in four different porous media in
e

Figure 4 . 5 ( a ) . Figure 4 . 5 ( b ) shows P /y plotted versus S on log-log paper for the same
c

data sets. Note that on Figure 4 . 5 ( b ) the data sets plot mainly as straight lines, e x c e p t
c l o s e to the point w h e r e S is equal to 1.0. (At S = 1.0, the saturation ratio, 5 is equal
e

to 1.0). T h e negative s l o p e of the line is called /., the Brooks-Corey pore-size distribution
index, and is o n e of two constants that characterize the media. T h e o t h e r constant is
the intercept of the e x t e n s i o n of the straight line with the S = 1.0 axis. This constant
e

b , the bubbling pressure, has already b e e n defined. T h e capillary behavior of a porous


b

medium can thus be defined on the basis of these two constants, / and h . In Figure
b

4 . 5 ( b ) curve 1 for volcanic sand has a x value of 2.29 and an b of about 15 cm.
b

T h e volcanic sand, fine sand, and glass beads shown in Figure 4.5 have a fairly
narrow grain-size distribution, which results in a narrow range of pore-size distribution.
T h e y are not representative of most natural soils. T h e calculated X values for these soils
are also higher than o n e finds in m o s t natural soils. T h e T o u c h e t silt loam is m o r e
representative of a normal soil than the o t h e r three.
Van G e n u c h t e n ( 1 9 8 0 ) also derived an empirical relationship between matric
potential and volumetric water content. He defined the relationship by the expression

o = e,+

e - e.
s

[1 + ( # ) " ]

n=

(4.9)

(4.10)

1 - m

a = (2" - 1 ) ' w h e r e m is a parameter estimated from the soil-water retention curve

(4.11)

173

F l o w and M a s s T r a n s p o r t in the Vadose Z o n e

FIGURE 4.5

Capillary pressure head as a function of effective saturation for porous materials with

various pore sizes, (a) Plotted on arithmetic paper and (b) plotted on log-log paper. Source: R. H. Brooks
and A. T. Corey, Proceedings, American Society of Civil Engineers, Irrigation and Drainage Division 9 2 ,
no. IR2 ( 1 9 6 6 ) : 6 1 - 8 7 .

To find the van G e n u c h t e n soil parameters, a soil-water retention curve ranging


from a matric potential of 0 to a matric potential of 1 5 , 0 0 0 cm is constructed. T h e
value of 9 is found at a matric potential of 0 and the value of 9 is that c o r r e s p o n d i n g
S

to a matric potential of 1 5 , 0 0 0 cm. Figure 4.3 s h o w s such a plot. T h e point P o n the


curve c o r r e s p o n d s to a water c o n t e n t 9 , which is found from
P

0,-1-1

(4.12)

T h e slope, 5, of the line at point P is determined graphically from the experimental


soil-water retention curve. A d i m e n s i o n l e s s slope, S , is then found from the relationship:
p

(4.13)

S = T h e parameter m can then be determined from the value of S

using o n e of these

formulas:
fl - e x p ( - 0 . 8 S )

(0 < S<

1 )

(4.14)
0.5755

0.1

0.025

174

Chapter F o u r

11

i.l il ri i
limili

-8000
-7000
-6000
-5000
-4000
-3000

-2000

_
Drying cycle
o Welling cycle

Ridcall clay loam


3101
o

om

om
o
0

-1000
-900
-800
-700
-600
-500
-400
-300

--

S i l i of Nave-Yahr
cm 2003
o

o
o
o

cm

o*
om

>.

Sand

-200

o
o

o
o

om
m
m
m

o
o

4106

om

o m

o
o

-100
-90
-80
-70

om

o m

-60

-50

-40
a

-30

-20

-10

Volumetric water conteni


FIGURE 4.6

Soil-moisture-retenn'on curves for three soils for both drying and wetting cycles. Source-.

Yechezkel Mualem, Catalogue of Hydraulic Properties of Unsaturated Soils (Haifa, Israel: Technion, 1 9 7 6 ) .

Flow and Mass Transport in the Vadose Zone

175

We can find the values of m and a from Equations 4 . 1 4 and 4 . 1 1 , respectively, by


using tlie b u b b l i n g pressure o b t a i n e d from the soil-water retention curve.
F r o m Figure 4.3 the s l o p e of the curve at P'\s a b o u t 0 . 3 4 . T h e dimensionless s l o p e
determined from Equation 4 . 1 3 is a b o u t 0 . 8 5 . F r o m Equation 4 . 1 4 m is determined to
be a b o u t 0.5 and from Equation 4 . 1 0 , n is 2.0. To estimate a from Equation 4 . 1 1 , we
need the value of h , the b u b b l i n g pressure. F r o m Figure 4.3 the value of h is a b o u t
b

- 3 5 5 , so log h ~ 2.55 and a ~ 0 . 0 0 5 .


b

4.6.3

Hysteresis

If o n e constructs a soil-water retention curve by obtaining data from a sample that is


initially saturated and then applying suction to d e s o r b water, the curve is k n o w n as a
d r y i n g c u r v e . If the sample is then resaturated by decreasing the suction, it will follow
a w e t t i n g c u r v e . Figure 4 . 6 s h o w s drying and wetting curves for three different soils.
Typically, the drying curve and the wetting curve will not be the same. T h i s p h e n o m e n o n
i s called h y s t e r e s i s . T h e causes o f hysteresis include (Hillel 1 9 8 0 )
1. T h e g e o m e t r i c effects of the s h a p e of single pores, which give rise to the so-called
ink bottle effect. This effect is illustrated in Figure 4.7. T h e p o r e has a throat radius
of r and a m a x i m u m radius of R. T h e matric potential w h e n the air-water interface
is at the p o r e throat, t/r , is equal to 2oyV, w h e r e a is the interfacial tension b e t w e e n
r

the p o r e water and the mineral surface ( s e e S e c t i o n 5 . 2 . 3 ) . T h e p o r e will drain abruptly


w h e n i/r has a m o r e negative pressure than ip . T h e p o r e c a n n o t then rewet until ip
r

falls b e l o w t/r . S i n c e R > r, then i/r > t/r; that is, it takes a lower matric potential
r

( m o r e negative p r e s s u r e ) to drain a p o r e than to fill it


2. T h e c o n t a c t angle b e t w e e n the water and the mineral surface is greater during the
period w h e n a water front is advancing as o p p o s e d to when it is retreating. T h e
advancing m e n i s c u s that forms during wetting will have a greater radius of curvature
and h e n c e a lower matric potentialthan that e x h i b i t e d by a m e n i s c u s that forms
during a drying cycle.

FIGURE 4.7
wetting.

Pore geometry affects equilibrium height of capillary water during (o) drainage and (b)

176

Chapter F o u r

3. Air that is trapped in pores during a wetting cycle will reduce the water content of
soil as it is being wetted. Eventually that trapped air will dissolve.
T h e hysteresis effect may be augmented by the shrinking and swelling of the clays
as the soil wets and dries and may also be affected by the rates of wetting and drying
(Davidson, Nielsen, and Biggar 1 9 6 6 ) .
If the soil is not dried to the maximum extent possible (greatest negative pressure
h e a d ) , when it is rewet, the soil will follow an intermediate curve known as a wetting
( o r drying) scanning curve. T h e r e are many wetting and drying scanning curves, depending upon the point on the main wetting or drying curve w h e r e the scanning curve
starts.
4.6.4

Construction of a Soil-Water-Retention Curve

Laboratory measurements of matric potential as a function of water content are made


to construct a soil-water-retention curve. In the wet-soil range ( 1 to 3 0 0 cm of waterpressure h e a d ) , a tension plate assembly is used. A saturated soil sample of known water
c o n t e n t is placed on a p o r o u s plate in a B u c h n e r funnel. T h e p o r o u s plate is saturated
and c o n n e c t e d to a water column that ends in a burette (Figure 4 . 8 ) . T h e position of
the burette can be changed to decrease the pressure head. As the pressure head b e c o m e s
m o r e negative, water is drained from the soil sample and the amount is measured in
the burette o n c e equilibrium has b e e n reached. Care must be taken to avoid evaporation

Buchner funnel

Burette
Porous plate

Tygon tubing
FIGURE 4.8

Equilibrating the water content of a soil sample with a known matric potential using a

tension plate assembly.

F l o w and M a s s T r a n s p o r t in the V a d o s e Z o n e

177

of water from the soil sample and the burette. A n u m b e r of m e a s u r e m e n t s are m a d e at


progressively m o r e negative pressure heads to determine the drying curve. O n c e the
practical limit of the tension plate assembly is reached ( 3 0 0 c m ) , the burette is raised
in a n u m b e r of steps to construct a wetting curve. T h i s is a wetting scanning curve, as
the soil is not fully drained at a pressure head of 3 0 0 cm of water.
F o r soils in the dry range ( 3 0 0 to 1 5 , 0 0 0 cm of water), a pressure-plate assembly
is used. T h e soil samples are placed on a saturated p o r o u s plate that is in a pressure
c h a m b e r . T h e pressure b e l o w the p o r o u s plate is kept at a t m o s p h e r i c pressure and the
pressure a b o v e die p o r o u s plate can b e set b e t w e e n 0.3 atm ( 3 0 0 c m o f w a t e r ) and 1 5
atm ( 1 5 , 0 0 0 c m o f w a t e r ) . T h e pressure a c r o s s the soil sample and p o r o u s plate causes
water to flow from the soil s a m p l e a c r o s s the p o r o u s plate into a lower reservoir.

4.6.5

Measurement of Soil-Water Potential

Matric potential is measured in the field with a tensiometer. T h i s apparatus c o n s i s t s of


a p o r o u s c e r a m i c c u p attached to a tube, which is buried in the soil. T h e t u b e is filled
with water and attached to a device such as a vacuum gauge, m a n o m e t e r , or pressure
transducer, which can m e a s u r e the tension. T h e matric potential of the soil tries to draw
water from the water-filled p o r o u s cup, and the resulting tension is measured. T e n s i o m eters can measure soil moisture tensions up to a b o u t 8 0 0 cm.
Figure 4 . 1 0 s h o w s the operation of two tensiometers in determining the gradient
of the matric potential. T e n s i o m e t e r A has the p o r o u s c e r a m i c c u p at a depth of 1 0 0 cm,

EXAMPLE

T h e following data have b e e n o b t a i n e d for a soil s a m p l e using a tension plate assembly.

PROBLEM

Construct a water-retention curve.

Volumetric Water Content


P r e s s u r e head (cm)
0

W e t t i n g Cycle

D r y i n g Cycle

0.447

0.448

-30

0.431

0.448

-60

0.411

0.443

-90

0.400

0.437

-120

0.392

0.424

-150

0.385

0.408

-180

0.379

0.391

0.377

0.377

-210

Wetting S c a n n i n g Curve
-120

0.411

-90

0.418

-60

0.423

-30

0.436

T h e data are plotted in Figure 4.9.

0.477

178

Chapter F o u r

0.36

0.38

0.40

0.42

0.44

0.46

Volumetric water content


FIGURE 4.9

Matric potential as a function of volumetric water content for the example problem.

whereas tensiometer B is at a depth of 30 cm. T h e total potential measured for ten


s i o m e t e r A is 1 2 6 cm, whereas for tensiometer B it is 88 cm. Since total potential
is the sum of pressure head and elevation head, o n e must subtract the elevation head
from total head to get pressure head. T h u s the pressure head measured in tensiometer
A is 2 6 cm, whereas for tensiometer B it is 5 8 cm. However, the total potential
gradient is downward, b e c a u s e the total head at A is m o r e negative than that at B. T h e
gradient is c o m p u t e d by finding the difference between the two total heads and dividing
the difference by the distance b e t w e e n the two p o r o u s cups. In Figure 4 . 1 0 the gradient
is
- 1 2 6 - ( - 8 8 )
-70

= ().->-*

(downward)

FIGURE 4.10

Tensiometer used to measure soil-water potential in the field.

180

Chapter F o u r

4.6.6

Unsaturated Hydraulic Conductivity

W h e n a rock or sediment is saturated, all the pores are filled with water, and most of
them transmit water. Only "dead-end" p o r e s do not participate in the transmittal of
water. Unsaturated soils have a lower hydraulic conductivity b e c a u s e s o m e of the pore
s p a c e is filled with air and thus can't transmit water. Soil moisture in the vadose z o n e
travels through only the wetted c r o s s section of p o r e space. As a saturated soil drains,
the larger p o r e s empty first, especially in soils that are structured. B e c a u s e these have
the greatest pore-level hydraulic conductivity, there is an immediate large drop in the
ability of the soil to transmit water. T h e unsaturated hydraulic conductivity is a function
of the water c o n t e n t of the soil: K = K(6). Unsaturated hydraulic conductivity can also
b e considered to b e a function o f the matric potential: K = K(\ji). Figure 4 . 1 1 is a graph
of unsaturated hydraulic conductivity as a function of matric potential. It can be seen
that this relationship also exhibits hysteresis.
T h e flow of soil moisture is influenced by temperature. Figure 4 . 1 2 shows the
influence of water temperature on curves of unsaturated hydraulic conductivity versus

-250 -

Flow and Mass Transport in the Vadose Zone

FIGURE 4.12

181

Unsaturated hydraulic conductivity as a function of water content for three temperatures.

Source: J. Constantz, Soil Science Society of America Journal 4 6 , no. 3 (1 982):46670.

water content. A c h a n g e from 2C to 2 5 C can cause unsaturated hydraulic conductivity


to increase by as m u c h as an o r d e r of magnitude. Constantz ( 1 9 8 2 ) w r o t e the following
e x p r e s s i o n for the relationship of unsaturated hydraulic conductivity to intrinsic permeability of the soil:
m

where
k,(9) = the relative conductivity, which is a n u m b e r from 0 to 1.0 that is
the ratio of the unsaturated hydraulic conductivity at a given 9 to
the saturated hydraulic conductivity
k = the intrinsic permeability
p. = the density of water at a given temperature
g = the acceleration of gravity
/t = the dynamic viscosity of soil water at a given temperature
Constantz ( 1 9 8 2 ) reports that the effect of temperature on unsaturated hydraulic conductivity is primarily a function of the effect of temperature on dynamic viscosity.
Unsaturated hydraulic conductivity c a n be determined by b o t h field m e t h o d s
( G r e e n , Ahuja, and C h o n g 1 9 8 6 ) and laboratory techniques (Klute and Dirksen 1 9 8 6 ) .
However, b o t h field and laboratory m e t h o d s are time c o n s u m i n g and tedious, with
n u m e r o u s practical limitations (van G e n u c h t e n 1 9 8 8 ) . As a result, unsaturated hydraulic

182

Chapter Four
conductivity is often estimated from soil parameters obtained from soil-water retention
curves.
Van G e n u c h t e n ( 1 9 8 0 ) derived e x p r e s s i o n s that relate the unsaturated hydraulic
conductivity to both the water c o n t e n t and the pressure head. T h e relationship between
the unsaturated hydraulic conductivity and the water c o n t e n t is
1/2

K{8) = AT,5, [1 - (1 - 5 , " T ]

(4.16)

where

s = (e - e )/(e, - e )
e

K(8) = unsaturated hydraulic conductivity at water content 8


K = saturated hydraulic conductivity
s

m = van G e n u c h t e n soil parameter


T h e equivalent relationship between unsaturated hydraulic conductivity and pressure head is
{ l - ( a * ) " - ' [ l + (aA)"]""'}

"

where
K(h) = the unsaturated hydraulic conductivity at pressure head h
h = pressure head
m = van G e n u c h t e n soil parameter
n = van G e n u c h t e n soil parameter
a = van G e n u c h t e n soil parameter
Figure 4.13 s h o w s observed values ( o p e n circles) and calculated curves (solid
lines) based on Equation 4.17 for relative permeability ( K = K(h)/K ) as a function of
r

pressure head for ( a ) a sandstone and ( b ) a silt loam. T h e predictive equation quite
closely follows the o b s e r v e d values.

4.6.7

Buckingham Flux

Law

T h e first to recognize the basic laws for the flow of water in soil was Buckingham ( 1 9 0 7 ) .
He recognized that the matric potential, t/r, of unsaturated soils was a function of the
water content, 8, temperature, and bulk density of the soil. He also realized that the flow
of water a c r o s s a unit cross-sectional area was proportional to the gradient of the soil
water matric potential. T h e proportionality constant, K(8), was recognized to be a
function of the water content. Buckingham, a physicist, appears to have had no knowledge
of Darcy's work on saturated flow s o m e half century before ( S p o s i t o 1 9 8 6 ) .
What is n o w known as the Buckingham flux law was formalized by Richards ( 1 9 2 8 ) ,
w h o e x t e n d e d the c o n c e p t of the potential gradient to include the total soil moisture
potential. (j>. Written in vector notation, the Buckingham flux law is
q=-JvWV(d>)

(4.18)

F l o w and M a s s T r a n s p o r t in the V a d o s e Z o n e

FIGURE 4.13

183

Observed values (open circles) and calculated curves (solid lines) for relative hydraulic

conductivity of (a) Hygiene sandstone and (b) Touchet silt loam G.E.3. Source: M. T h . van Genuchten, Soil
Science Society of America Journal 44 (1 9 8 0 ) : 8 9 2 - 9 8 .

where
3

q = the soil moisture flux ( L L " L

_ 1

K(\fi) = the unsaturated hydraulic conductivity ( L T ~ ' ) at a given ip


V(<A) = the gradient of the total soil water potential, <p, w h e r e <p = t/r + Z
(a- )
1

4.6.8

Richards Equation

T h e continuity equation for soil moisture through a representative elementary volume


of the unsaturated z o n e can be stated as the c h a n g e in total volumetric water c o n t e n t
with time and is equal to the sum of any c h a n g e in the flux of water into and out of
the representative elemental volume. T h e continuity equation can be e x p r e s s e d as
38

(3q,

3<7

3q\

w h e r e q , q , and q are soil moisture fluxes.


x

In vector notation Equation 4.19 is

38
- = - V q

(4.20)

C o m b i n i n g Equations 4 . 2 0 with 4 . 1 8 , we obtain the Richards equation (Richards

1931):
y = V [K(ip)V<j)\

(4.21)

Chapter Four

T h e value of Vz is 0 in the xand j directions and is 1 in the z direction. In addition,


d> is equal to ip + Z T h e r e f o r e , if z is taken to be positive in a downward direction,
Equation 4.21 can be rewritten as

(4.22)

F o r one-dimensional flow, Equation 4 . 2 2 reduces to

( 4

2 3 )

If the matric potential is m u c h greater than the gravity gradient, then the last term
of Equation 4 . 2 3 can be dropped; the resulting equation is

T h e preceding equations a s s u m e a constant temperature and air pressure, a non


d e f o r m a b l e soil matrix, i n c o m p r e s s i b l e water, and that soil water density is independent
of solute c o n c e n t r a t i o n and d o e s not vary throughout the flow domain. Furthermore,
these equations a s s u m e that the p r e s e n c e of air can be ignored, e x c e p t as it affects the
value of K. Equation 4.23 is nonlinear and difficult to solve by analytical means. However,
numerical m e t h o d s of solution of this equation have b e e n developed (Nielsen, van
G e n u c h t e n , and Biggar 1 9 8 6 ) .
4.6.9

Vapor Phase Transport

Under special c i r c u m s t a n c e s vapor can m o v e through the soil under air pressure gradients
as the a t m o s p h e r i c pressure fluctuates. However, water vapor normally moves by diffusion
from areas w h e r e the vapor pressure in the unsaturated p o r e s is higher to areas w h e r e
it is lower.

Diffusion of soil moisture vapor is given by

where
q = the vapor flux
v

p = the vapor concentration i n the g a s e o u s phase


v

D = the diffusion coefficient for water vapor


v

Assuming no convective transport, the nonsteady, one-dimensional transport of


soil moisture in the vapor phase is given by ( J a c k s o n 1 9 6 4 ) as

185

F l o w and M a s s T r a n s p o r t in the Vadose Z o n e

Simultaneous transport of water in b o t h the liquid and vapor p h a s e is d e s c r i b e d


by the following equation, which is obtained by c o m b i n i n g Equations 4 . 2 4 and 4 . 2 6 :

dt

4.7

_d_
x

(4.27)

D,

Mass Transport in the Unsaturated Zone


T h e steady-state diffusion of a solute in soil moisture is given by (Hillel, 1 9 8 0 )
J

D*(ff)dCldz

(4.28)

where
J = the mass flux of solute per unit area p e r unit time
>*((?) = the soil diffusion coefficient, which is a function of the water
content, the tortuosity of the soil, and o t h e r factors related to
the electrostatic d o u b l e layer
dCldz = the concentration gradient in the soil moisture
T h e s e c o n d - o r d e r diffusion equation for transient diffusion of solutes in soil water
is
8_C__d_
dt~

Sz

cz

(4.29)

Soil moisture traveling through the unsaturated z o n e moves at different velocities


in different p o r e s due to the fact that the saturated p o r e s through which the moisture
moves have different sized p o r e throats. In addition, velocities within each saturated p o r e
will vary across the width of the pore. As a result, soil water carrying a solute will m i x
with o t h e r soil moisture. This is analogous to the mechanical mixing of saturated flow.
Mechanical mixing is found from the following equation (Nielsen, van G e n u c h t e n , and
Biggar 1 9 8 6 ) :
Mechanical mixing =

(4.30)

where
C = an empirical soil moisture dispersivity
v = the average linear soil moisture velocity
T h e soil moisture dispersion coefficient, D , is the sum of the diffusion and mechanical mixing:
s

D, = D* + C|A|

(4.31)

T h e total o n e dimensional solute flux in the vadose z o n e is the result of advection,


diffusion, and hydrodynamic dispersion. With diffusion and hydrodynamic dispersion
c o m b i n e d as the soil-moisture dispersion coefficient, this result can be e x p r e s s e d as
J = vdC - Dfi dC/dz

(4.32)

1 86

Chapter

Four

where
J = the total mass of solute a c r o s s a unit cross-sectional area in a unit
time
v = the average soil moisture velocity
C = the solute c o n c e n t r a t i o n in the soil moisture
8 = the volumetric water c o n t e n t
dCldz

the solute gradient

D s = the soil moisture diffusion coefficient, which is a function of both

and v

T h e continuity equation for a solute flux requires that the rate of c h a n g e of the
total solute mass p r e s e n t in a representative elemental volume be equal to the difference
b e t w e e n the solute flux going into the REV and that leaving the REV. T h e total solute
m a s s is the sum of t h e dissolved solute mass and the mass of any solute associated with
the solid p h a s e of the soil. T h e dissolved solute mass is equal to the product of 8 and
C. T h e solute mass b o u n d to the soil is the product of the soil bulk density, Bd, and the
c o n c e n t r a t i o n of the solute p h a s e b o u n d to t h e soil, C T h e continuity equation for
the solute is:
d(BdC)

d(8C)

dj

-^r+=-rz

(433)

By c o m b i n i n g Equations 4.32 and 4.33, we obtain

bt

' 7 ~
dt

Tl

dz

^ TZI A " T ~ I

(4.34)

T h e convective soil moisture flux, q, is the product of the volumetric water content,
f?, and the average soil moisture velocity, v. T h e r e may also be s o u r c e s and sinks of the
solute not a c c o u n t e d for by the a b s o r b e d concentration, C*. F o r example, plants may
remove nutrients from solution and solutes might be created by biological decay as well
as microbial and c h e m i c a l transformation and precipitations. T h e s e can be added to
Equation 4.34 by m e a n s of a term for the summation of Y w h e r e T represents o t h e r
s o u r c e s and sinks. With a slight rearrangement Equation 4.34 b e c o m e s the fundamental
m a s s transport equation for the vadose z o n e :

If Bd, Ds, 0, and q are assumed to be constant in time and space, Equation 4.35 reduces
to Equation 3.1, the basic one-dimensional advection-dispersion equation.

4.8

Equilibrium Models of Mass Transport


In o r d e r to a c c o u n t for a solute, which can be in either a dissolved form or a b s o r b e d
by the soil, we n e e d to k n o w the relationship between the c o n c e n t r a t i o n in solution. C
and the a b s o r b e d concentration, C*. If the solute reaches equilibrium rapidly between

187

F l o w and M a s s T r a n s p o r t i n the Vadose Z o n e

the dissolved and a b s o r b e d phase, then the relationship can be d e s c r i b e d by an absorption isotherm. Sorption isotherms are discussed m o r e thoroughly in Chapter 3. F o r
illustration in this chapter, we will use the linear isotherm

C* = K C

(4.36)

where K , the distribution coefficient, is the slope of the plot of C* as a function of C.


Under certain conditions the s o u r c e and sink function, T,-, may be approximated
by zero- and first-order decay and production terms (Nielsen, van G e n u c h t e n , and Biggar
1 9 8 6 ) . If t;, and t] are rate constants for first-order decay in the liquid and solid phases,
respectively, and
and are zero-order rate terms for production in the liquid and
solid phases, then
d

r, = - r,flc - n B c* + i,e +
s

iB
s

(4.37)

B y substitution o f Equations 4 . 3 6 and 4.37 into 4.35 and simplification o f terms,


we can obtain the following equation:
.d_

(PC

dC

- = A t st - " ^ - ? C + {

~Tt ~ IP ~ " ~dz '

(4.38)

w h e r e R is the retardation factor, which is given by

R = 1 +^

(4.39)

and t] and are consolidated rate factors, given by

1 = 1, + ^

(4.40)

{ = i, +

(4-41)

Equation 4.38 is b a s e d on steady-state flow, as the volumetric moisture content


and the fluid velocity are taken to be constants.
Van G e n u c h t e n ( 1 9 8 1 ) solved equation 4 . 3 8 for a n u m b e r of different boundary
conditions. In general, at time equals 0 and at s o m e place in the soil column, z, the
solute concentration is QlC(z, 0) = Q(z)]. T h e concentration introduced into the top
of the soil c o l u m n w h e r e z = 0 at s o m e time t is C o l C ( 0 , t) = Q ( r ) ] . T h e rate that
solute is introduced into the top of the soil c o l u m n by b o t h advection and diffusion is
equal to the p o r e water velocity, v, times Q,. ( T h e usual boundary condition adopted
for this is (- D dC/dz + vC) = vQ,.)
s

If a pulse of solute is introduced into the soil c o l u m n for a time period of 0 to


/, during that time the rate is vC . After time t , the pulse is e n d e d and the rate at which
solute is introduced is 0. ( F o r 0 < t < t , (-D,dC/dz + vC) = vC\; for / > t ,
(-D dC/dz + vC) = 0 ) . T h e soil column is infinitely long (dC/dz{co, t) = 0 ) .
In this c a s e the solution to Equation 4 . 3 8 for times when the pulse is being injected
( 0 < t < to) is
0

C(z,

0=

(Co -

A(z, t) + B(z, t)

(4A2)

188

Chapter Four

F o r times greater than


solution to Equation 4 . 3 8 is

that is, after injection of the pulse has s t o p p e d t h e

(4.43)

C(z, T) = (CO- A A(z, T) + B(z, T) - CoAix, T - T )


0

In b o t h Equations 4.42 and 4 . 4 3 , the following arguments are used:


'i)z
[~Q - u,

A(z, t) =
(v

eXP

u) \_

erfc

2D

~ Rz - ut
l2

2(D RtV _

\(v

u)z~

Rz + ut

erfc

(4.44)

2(D Rt)"
s

e x

+ ^
2r,D,
B(z,

t)

VA

c exp

r)

2(D Rt)"

R)

"E

(Rz -

Rz

vt

v tVi

erfc
li2

2(D Rt) \

\KRDJ

R
x exp

Rz + vt

erfc

1/

vt)

2\

vz
+

D,

J j \
+

D R)
S

(vz

\ D

AD,Rt
x erfc

+-+ I Q

Rz Rt/+ vt
2(D
12

exp

(4.45)

and

(4.46)
F o r the steady-state case, Equation 4 . 3 8 may be written as

_ 8C
2

dz

v-rtC

t\'

(4.47)

dz

T h e solution to this is
C(z) = i +

U)Z

iJv'

+ a

exp

2D

(4.48)

w h e r e u is defined in Equation 4.46.


Equations 4.42, 4 . 4 3 , and 4 . 4 8 are only s o m e of the analytical solutions that van
G e n u c h t e n ( 1 9 8 1 ) has o b t a i n e d for Equation 4 . 3 8 . Solutions for other boundary and
initial conditions include variable initial solute concentration in the soil column and an
exponentially decaying s o u r c e term. T h e reader is directed to van G e n u c h t e n ' s original
paper for additional solutions.

4.9

Nonequilibrium Models of Mass Transport


T h e soil moisture may move at such a quick rate that a solute may not be able to reach
an equilibrium position with respect to chemical reactions that are occurring. O n e
nonequilibrium formulation arises when the adsorption p r o c e s s can be described by a

189

F l o w and M a s s T r a n s p o r t i n the Vadose Z o n e

first-order linear rate equation. Under this condition, assuming steady-state flow and
ignoring the s o u r c e / s i n k term, Equation 4.34 can be written as a c o u p l e d system (Nielsen,
van G e n u c h t e n , and Biggar 1 9 8 6 ) as follows:
B 8C*

SC

J^T T

SC

= D

^ -

SC

,,
( 4

4 9 a )

and

^ - = C (.K C - C )
at

(4.49b)

w h e r e C is a first-order rate coefficient. Equations 4 . 4 9 a and b have b e e n used by many


to d e s c r i b e nonequilibrium transport in soils. As s h o w n by van G e n u c h t e n , Davidson,
and Wierenga ( 1 9 7 4 ) , the first-order rate m o d e l has not materially improved the description of nonequilibrium transport in soils. An alternative m o d e l of nonequilibrium
transport arises w h e n soil water is assumed to consist of a m o b i l e phase and an i m m o b i l e
phase. T h e m o b i l e phase o c c u p i e s the c e n t e r o f saturated pores. I m m o b i l e water consists
of thin coatings on soil particles, dead-end pores, and water trapped in small unsaturated
p o r e s (Coats and Smith 1 9 6 4 ) . E x c h a n g e o f solute b e t w e e n the m o b i l e and i m m o b i l e
phases o c c u r s due to diffusion. In addition, solute in b o t h the m o b i l e and i m m o b i l e
p h a s e s can participate in adsorption-desorption reactions. F o r Freundlich-type linear
equilibrium, this conceptualization can be d e s c r i b e d by the following equations (van
G e n u c h t e n and Wierenga 1 9 7 6 ) :
r

+ BR
at
lm

im

% = 0 D 5%
St

- 6v
Sz

Sz

* K,J-T7 = (C -C )
im

St

im

(4.50a)

(4.50b)

where
= a mass transfer coefficient
.

R = a retardation factor for the m o b i l e water


m

Jt

im

= a retardation factor for the i m m o b i l e water

C = solute concentration in the m o b i l e water


m

im

= solute concentration in the i m m o b i l e water

6 = volumetric m o b i l e water c o n t e n t
m

im

sm

= volumetric i m m o b i l e water c o n t e n t
= soil moisture dispersion coefficient for the m o b i l e water

F o r saturated soils, the a m o u n t of i m m o b i l e water is a function of the soil-water


flux, the size of soil aggregates, and the concentration of the ionic solute. T h e mass
transfer coefficient, , is also a function of t h e s e s a m e factors as well as the s p e c i e s
being transferred (Nkedi-Kizza et al. 1 9 8 3 ) .
T h e preceding model with m o b i l e and i m m o b i l e water z o n e s has b e e n successfully
used at the laboratory scale (Nkedi-Kizza, et. al. 1 9 8 3 ) ; Nielsen, van G e n u c h t e n , and
Biggar 1 9 8 6 ) . Figure 4.14 shows a breakthrough curve for a c o l u m n study of the transport

190

Chapter

Four

*Vrmrr
0

10

15

Pore volume V / V
FIGURE 4.14

20

Observed and calculated breakthrough curves for a solution of boron passing through a

column filled with an aggregated clay loam soil. Source: D. R. Nielsen, M. T h . van Genuchten, and J. W .
Biggar, Water Resources Research 2 2 , no. 9 (1 9 8 6 ) : 8 9 S - 1 0 8 S . Copyright by the American Geophysical
Union.

of a solution of b o r o n in an aggregated clay loam soil. T h e curve labeled "1-region


m o d e l " is based on Equation 4.38, and the curve labeled "2-region model" is based on
Equations 4.50a and b. In this c a s e the 2-region model d o e s a better j o b of matching
the experimental data.

4.10

A n i o n Exclusion
F o r a nonreactive solute, Equation -t.35 can be written as
2

dC

dC

dC

Many solutes are considered to be nonreactive in the s e n s e that they do not sorb
o n t o particle surfaces. Included in this category are anions such as chloride. However,
chloride carries a negative charge, and if there are many clay particles in the porous
medium, the electrostatic d o u b l e layer will repel anions. Consequently, there is a region
around each colloidal particle from which the anions are repelled, with the resulting
distribution shown on Figure 4 . 1 . As a first approximation we can assume there is a
two-phase distribution of the solute in the soil water. Within the exclusion volume the
concentration is zero, with all of the solute thus being concentrated in the p o r e water
outside o f the exclusion volume ( J a m e s and Rubin 1 9 8 6 ) . T h e volume o f the exclusion
z o n e is 8 .
T h e anion concentration in the p o r e water under conditions of anion exclusion
is given by ( B r e s l e r 1 9 7 3 )
ex

rir

r) f

BC

- * 4 ? - * s

( 4

5 2 )

Flow and Mass Transport in the Vadose Zone

191

w h e r e C i s the c o n c e n t r a t i o n of the solute in the bulk p o r e solution (that is, including


the water in the e x c l u s i o n z o n e ) .
T h e value of ( ? can be found experimentally from a soil c o l u m n test. Water with
an initial concentration of a single anion of CQ is introduced into a soil c o l u m n that has
a known water content. T h e water c o n t e n t is held constant, and sufficient solute of
c o n c e n t r a t i o n Q, is introduced so that the c o n c e n t r a t i o n of the water leaving the soil
column, Q is equal to Q,. T h e mass of the anion contained in the soil c o l u m n is
u

determined, and a c o n c e n t r a t i o n is calculated based on the total water c o n t e n t in the


soil column, C

c a k

. This calculated c o n c e n t r a t i o n of anionic solution is less than C ,


ou

b e c a u s e water in the e x c l u s i o n volume, which contains no anions, was used in the


c alt uljiion. T h e value of the e x c l u s i o n volume can be found from ( B o n d , Gardiner, and
Smiles 1 9 8 2 ) :

(4.53)
In addition to decreasing the o b s e r v e d c o n c e n t r a t i o n of the anion in the soil
column, anion e x c l u s i o n causes the anions to travel faster than the average rate of moving
p o r e water. T h e average rate of pore-water movement, r, is equal to the rate of the fluid
flux, q, divided by the water content, 9. T h e anions c a n n o t travel through the e x c l u d e d
part of the volumetric water content, which is c l o s e to the mineral surfaces and has a
l o w or z e r o velocity. T h e r e f o r e , they must move in the part of the p o r e water that is
available to themi.e., the c e n t e r of the p o r e s , w h e r e the fluid velocity is greater than
average. As a result, e x c l u d e d anions will travel further in a given period of time than
they would in the a b s e n c e of anion e x c l u s i o n .
An a p p r o x i m a t e solution to Equation 4.52 tor conditions of uniform water c o n t e n t
and a steady flux of a dissolved anion of c o n c e n t r a t i o n Q, into a semi-infinite soil c o l u m n
with an initial concentration of the dissolved anion, C is
h

(4.54)

EXAMPLE
PROBLEM

J a m e s and Rubin ( 1 9 8 6 ) p e r f o r m e d an e x p e r i m e n t in which they measured the concentration profiles o f chloride introduced into soil c o l u m n s containing Delhi sand. T h e
sand was 9 0 % sand, 7% silt, and 5% clay by weight with a c a t i o n - e x c h a n g e capacity of
0.05 mol kg and an organic c a r b o n c o n t e n t of 0 . 0 0 3 g of c a r b o n per gram of soil. T h e
soil c o l u m n was constructed with a suction line on the b o t t o m to create a soil w a t e r
tension on the entire c o l u m n . T h e soil c o l u m n s w e r e initially leached with a nitrate
solution to remove any chloride. Water c o n t e n t (0) was kept uniform throughout the
soil c o l u m n and the volumetric water flux (<r/) was numerically equal to the unsaturated
hydraulic conductivity evaluated at 6. A solution containing chloride was introduced into
the soil column at the s a m e constant volumetric water flux established during the leaching
phase. T h e soil c o l u m n s w e r e constructed as a series of s e c t i o n s either 1.14 cm or
2.28 cm long so that at the end of the injection phase they could be disassembled and
the chloride mass and water c o n t e n t o f each section could b e measured.
Several e x p e r i m e n t s w e r e c o n d u c t e d in which the water c o n t e n t ranged from 0 . 1 6 7
to 0 . 2 2 5 and the volumetric water fluxes ranged from 0 . 0 3 9 3 cm hr to 0 . 3 9 7 c m / h r .
Figure 4 . 1 5 s h o w s the results of e x p e r i m e n t s A and B. which had water c o n t e n t s of

192

Chapter F o u r

25 -

30
FIGURE 4.15

Chloride concentration profiles in a soil column affected by anion exclusion. Source:

R. V. James and S. Rubin, Soil Science Society of America Journal 50 (1986): 1 1 4 2 - 5 0 .

0.221 and 0 . 2 2 5 , respectively, while both had a water flux of 0.397 cm/hr. T h e r e are two
curves on the figure, o n e c o m p u t e d using Equation 4 35 and o n e using Equation 4.54.
Equation 4.35 is applied for a nonsorbing, nonreactive solute. T h e chloride concentrations
are shown as dimensionless chloride, that is

C^JCg.

No measured chloride values were

equal to Q, illustrating the results of anion exclusion. T h e experimental results can be


seen to follow closely the curve c o m p u t e d from Equation 4.54.
T h e excluded water c o n t e n t was c o m p u t e d from the experimental results at the
top of the soil c o l u m n s using Equation 4.53. It was found to be 0 . 0 1 9 for both saturated

Flow and Mass Transport in the Vadose Zone

193

and unsaturated conditions. T h e dispersion coefficient was determined by solving Equation 4.52, with appropriate boundary conditions as
erfc"

2 ( c - q)"

(0 - fc>)

(Cb - q).

2s/(0 - 0 ) A '

2sJ(e - e )D ei
ex

(4-55)

T h e left side of Equation 4.55 was plotted against the depth, z, and a straight line was
determined by the m e t h o d of least squares regression. T h e value of D was then calculated
from the slope of the line. Inspection of Equation 4.55 s h o w s that D is the only unknown.
T h e curves on Figure 4.15 w e r e calculated using the experimentally determined values
of As

Case Study: Relative Movement of Solute and Wetting Fronts


Pickens and Gillham (1980) studied the relative motion of the wetting front and the solute in
an unsaturated soil column by use of a finite-element model. The model allows for the use of
either nonhysteretic or hysteretic water-contentpressure-head relationships. Previous work

-80

0.0

0.1

0.2

0.3
3

Water content (cm /cm )


FIGURE 4.16

A water-content-pressure-head hysteresis loop used in a model study of solute infiltration

in a soil column Source: J. F. Pickens and R. W. Gillham, Water Resources Research 16, No. 6 (1980): 1 0 7 1 78. Copyright by the American Geophysical Union.

194

Chapter F o u r

Time = 30 min

FIGURE 4.17

Model results of matric potential, ((//), water content {()), pore-water velocity (v), relative

solute concentration (C/C ), and volumetric solute concentration (OC/C ). Source-. J. F. Pickens and R. W.
0

Gillham, Water Resources Research 16, no. 6 (1 9 8 0 ) : 1 0 7 1 - 7 8 . Copyright by the American Geophysical
Union.

( G i l l h a m , Klute, and Heermann 1 9 7 9 ) had s h o w n that it w a s very important to include hystere s i s in the model because the equation for computation of the pore-water velocity and the
advection-dispersion equation contains a term for the water content.
T h e modeled s o i l column had a saturated hydraulic conductivity of 0 . 2 9 cm/mn, a saturated water content of 0 . 3 0 1 , and a longitudinal d i s p e r s i v i r y of 0 . 5 cm. T h e initial s o i l - m o i s ture condition in the model had a linear p r e s s u r e head variation between 0 at the bottom to
1 50 cm at the top of the s o i l column. T h e corresponding initial water-content values w e r e
obtained from a water-contentpressure-head h y s t e r e s i s loop (Figure 4 . 1 6 ) and f o l l o w e d the
main drainage curve, so that the model represented conditions that would develop if a satur a t e d - s o i l column w e r e d r a i n e d . T h e s o i l column did not contain any solute p r i o r to infiltrat i o n . A slug of water containing a solute at concentration C w a s a l l o w e d to infiltrate at a
0

at a rate of 0 . 1 7 cm/min for a 3 0 - m i n period, resulting in a total depth of infiltration of 5 cm.


T h e model computed the p r e s s u r e head, water content, vertical pore-water velocity, relative

Flow and Mass Transport in the Vadose Zone

195

T i m e = 300 min

F I G U R E 4.17

Continued

solute concentration (C/Q), and volumetric solute concentration

[()C/C )
Q

after 30 min, 60 min,

300 min, and 5760 min. Figure 4.17 shows the results of the model study. At 5760 min the
head distribution had returned to the preinfiltration conditions, which indicates that equilibrium had been reached. Inspection of Figure 4.17 shows that although the water in the bottom of the soil column is above the irreducible water content at 5760 min, the solute has
remained near the top of the soil column. This can occur because the water found at the
bottom of the soil column was displaced downward from the top of the soil column by
the infiltrating water containing the solute, which did not penetrate past the depth where
C/C = 0.
0

The model was run under three conditions. In the hysteretic mode the hysteretic relationship between water content and pressure head was used. As the slug of water infiltrated,
the wetting relationships were used, followed by drying curves when the infiltrating slug had
passed. This model run is represented as a solid line on Figure 4.17. In the nonhysteretic
wetting-curve mode, the pressure-headwater-content relation was based on the main wetting curve, and in the nonhysteretic drying-curve mode, it was based on the main drying
curve. The results of these two modes are represented by circles and triangles, respectively.

196

Chapter Four

T h e importance of using the hysteretic mode appears to be greater for the pressure head
and the water content than for the pore velocity and the solute front movement.

4.11

Preferential Flowpaths in the Vadose Zone


T h e preceding analyses all treat the unsaturated z o n e as a h o m o g e n e o u s , p o r o u s medium.
However, this is certainly not the case. In the root z o n e there are numerous large pores
and cracks formed by such agents as plant roots, shrinkage cracks, and animal burrows.
T h e s e m a c r o p o r e s can form preferential pathways for the movement of water and
solute, both vertically and horizontally through the root z o n e ( B e v e n and Germann
1 9 8 2 ) . This situation can lead to "short circuiting" of the infiltrating water as it moves
through the m a c r o p o r e s at a rate much greater than would be e x p e c t e d from the hydraulic
conductivity of the soil matrix; s e e Figure 4 . 1 8 ( a ) .
A s e c o n d type of preferential flow is f i n g e r i n g , which o c c u r s when a uniformly
infiltrating solute front is split into downward reaching "fingers" due to instability caused
by p o r e scale permeabilii,' variations. Instability often o c c u r s when an advancing wetting
front reaches a boundary w h e r e a finer sediment overlies a c o a r s e r sediment, s e e Figure
4 . 1 8 ( b ) (Hillel and Baker 1 9 8 8 ) .
A third type of preferential flow is t u n n e l i n g (Kung 1 9 9 0 b ) . Funneling o c c u r s
in the vadose z o n e b e l o w the root z o n e and is associated with stratified soil or sediment
profiles. Sloping coarse-sand layers e m b e d d e d in fine sand layers can impede the down
ward infiltration of water. T h e sloping layer will collect the water like the sides of a
funnel and direct the flow to the e n d of the layer, w h e r e it can again percolate vertically,
but in a c o n c e n t r a t e d volume, as shown in Figure 4 . 1 8 ( c ) . Field studies using water
containing dye placed in furrows indicate that the water is moving in the fine-sand layer
a b o v e the discontinuity of the coarse-sand layer (Kung 1 9 9 0 a ) . T h e s e s a m e dye studies
s h o w e d that b e c a u s e of funneling, the volume of the soil containing dye decreased with
depth. T h e dyed soil region o c c u p i e d about 5 0 % of the soil volume at 15 to 2.0 m; from
3.0 to 3-5 m, it o c c u p i e d only 1 0 % of the soil volume, and by 5.6 to 6.6 m, it was found
in about 1% of the soil volume. At this depth a single column of dyed soil was found,
obviously formed by funneling of flow of dyed water from above ( K u n g 1 9 9 0 a ) .
T h e s e o c c u r r e n c e s of preferential flow in particular and soil heterogeneity in general
have disturbing implications for monitoring solute movement in the unsaturated zone.
S o m e studies have recorded seemingly anomalous results, with d e e p e r soil layers having
greater c o n c e n t r a t i o n s o f solute than m o r e shallow layers (Kung 1 9 9 0 b ) . T h e s e anomalies
can be explained by preferential flow patterns, with infiltrating solute being directed to
certain regions of the vadose z o n e by short circuiting, fingering, and funneling. This
suggests that a large n u m b e r of sampling devices in the vadose z o n e might be needed
if a reasonably accurate picture of the distribution of a contaminant is to be obtained
As contaminants reaching the water table may need to pass through the unsaturated
zone, preferential flow paths in the unsaturated z o n e also have implications for ground
water monitoring. In the c a s e of a contaminant that was evenly spread on the land
surfacefor e x a m p l e , an agricultural c h e m i c a l o n e would e x p e c t that there would bean evenly distributed solute load reaching the water table via the vadose zone. However,
due to preferential flow paths the mass of solute may be concentrated in s o m e locations,

197

198

Chapter Four

resulting in an uneven distribution in the shallow ground water beneath the site. Monitoring wells beneath the site may s h o w varying solute concentrations, depending upon
h o w c l o s e they are to an up-gradient point of concentrated recharge.

4.12

Summary
T h e \ a d o s e z o n e e x t e n d s from the land surface to the water table; moisture in the vadose
z o n e is under tension. Soil particles can have charged surfaces, which can attract or
repel anions and cations. Moisture moves through the vadose z o n e due to a potential
that is the sum of the elevation potential and the matric potential. Matric potential is a
function of the volumetric water c o n t e n t and d e p e n d s upon whether the soil has previously u n d e r g o n e wetting or drying. T h e unsaturated hydraulic conductivity is also a
function of the volumetric water content. T h e flux of moisture through the soil can be
calculated from Darcy's law or the Buckingham flux law. T h e nonsteady flow of soil
moisture is d e s c r i b e d by the Richards' equation. T h e nonsteady movement of soil vapor
and soil moisture can also be d e s c r i b e d by a partial differential equation.
Solute m o v e m e n t through the vadose z o n e p r o c e e d s by both advection and diffusion. T h e r e is an advective-dispersive equation for solute transport in the vadose z o n e
that a c c o u n t s for retardation through sorption o n t o soil particles. Analytical solutions to
this equation exist. Solute movement in the vadose z o n e is affected by regions of
i m m o b i l e water found in dead e n d pores. It is also affected by an anion-exclusion z o n e
w h e n clays are present in the soil. Preferential pathways of solute movement may be
present that create pathways for lateral movement of water in the vadose z o n e and
c o n c e n t r a t e infiltrating water into certain regions. Preferential pathways of water move
ment may make monitoring of the vadose z o n e and shallow water table difficult.

Chapter Notation

A(z, t)
B(z, t)

T e r m defined by Equation 4.44


T e r m defined by Equation 4.45
Bulk density of soil

Solute concentration

Solute phase b o u n d to soil

Initial solute concentration in soil water


Solute concentration in immobile water

On

Solute concentration in m o b i l e water

Co

Solute concentration in injected water


Concentration of solute calculated in soil if the water in the anion
exclusion z o n e is included
Diffusion coefficient for water vapor
Soil diffusion coefficient, which is a function of 6
Dispersion coefficient for soil moisture, which is a function of 9
Dispersion coefficient for m o b i l e soil moisture

Elementary charge of an ion

Flow and Mass Transport in the Vadose Zone

g
h

k
k
te)
B

Ki
m

n
0

Pc
1
1
II

R
Rm
Rim
S
S
S
P

1
'O
V

V
Z
Z
Zo

a
N
e

c
c

'/
11

1,

e
e.,
X

Ci
L

Acceleration of gravity
Pressure potential in units of length
Mass flux of solute
B o l t z m a n constant
Intrinsic permeability
Unsaturated hydraulic conductivity, which is a function of
Relative permeability as a function of 6
Distribution coefficient
Van G e n u c h t e n soil parameter
Van G e n u c h t e n soil parameter
Concentration of ions in bulk solution
Capillary pressure
Volumetric soil water flux (specific d i s c h a r g e )
V a p o r flux
T e r m defined b y Equation 4 . 4 6
Retardation factor
Retardation factor for m o b i l e water
Retardation factor for i m m o b i l e water
Slope
Dimensionless s l o p e
Saturation ratio (0/6 )
T e m p e r a t u r e in Kelvins
Time
T i m e w h e n pulse o f injection o f solute c e a s e s
V a l e n c e of ions in solution
Average linear soil moisture velocity (Q/ff)
Elevation
Gravitational potential
Characteristic length or e x t e n t of the d o u b l e layer
Van G e n u c h t e n soil parameter
Mass transfer coefficient
Dielectric constant
Empirical soil moisture dispersivity
First-order rate coefficient
Consolidated first-order decay term
First-order decay rate constant for liquid p h a s e
First-order decay rate constant for solid p h a s e
Volumetric water c o n t e n t
V o l u m e o f e x c l u d e d z o n e due t o anion e x c l u s i o n
Volumetric water c o n t e n t of m o b i l e water
Volumetric water c o n t e n t of i m m o b i l e water
B r o o k s - C o r e y soil parameter
Dynamic viscosity of water
Consolidated z e r o o r d e r s o u r c e term
Zero-order s o u r c e term for liquid phase
Zero-order s o u r c e term for solid p h a s e
s

199

200

Chapter F o u r

Po

Vapor concentration in vapor phase

P,v
0"

Density of water
Interfacial tension

Various s o u r c e s and sinks for solute

<t>EV
<t>EW

Soil moisture potential in terms of energy per unit volume


Soil moisture potential in terms of energy per unit weight

<t>EM

Soil moisture potential in terms of energy per unit mass

<P

Matric potential of soil

References
Beven, Keith, and Peter Germann. 1982. Macropores and
water flow in soil. Water Resources Researcl) 18, no.
5:1311-25.
Boersma, L, et al. 1972. "Theoretical research." In Soil Water,
edited by Nielsen, D. R., R. D. Jackson, J. W. Cary, and
D. D. Evans, 2 1 - 6 3 - Madison, Wis. American Society
of Agronomy, 175 pp.
B o n d . W. J . , B. N. Gardiner, and D. E. Smiles. 1982. Constant
flux adsorption of a tritiated calcium chloride solution
by a clay soil with anion exclusion. Soil Science Society
of America Journal 46:1133-37.
Bresler, E. 1973. Anion exclusion and coupling effects in
nonsieady transport through unsaturated soils: I. Theory. Soil Science Society of America Proceedings

37:663-69.
Brooks, R. K, and A. T. Core)'. 1966. Properties of porous
media affecting fluid flow. Proceedings, American Society of Civil Engineers, Irrigation and Drainage Divisional, no. I R 2 : 6 l - 8 7 .
Buckingham, E. 1907. Studies on the movement of soil moisture. Bureau of Soils Bulletin 38. Washington, D.C.:
U.S. Department of Agriculture.
Coats, K. H., and B. D. Smith. 196-*. Dead end pore volume
and dispersion in porous media. Society of Petroleum
Engineer Journal
4:73-84.
Constants, J. 1982. Temperature dependence of unsaturated
hydraulic conductivity of two soils. Soil Science Society
of America Journal 46, no. 3 : 4 6 6 - 7 0 .
Dane, J. H., and A. Klute. 1977. "Salt effects on the hydraulic
properties of a swelling soil." Soil Science Society of
America Journal'41, no. 6 : 1 0 4 3 - 5 7 .
Davidson, J. M., D. R. Nielsen, and J. W. Biggar. 1966. T h e
dependency of soil water uptake and release on the
applied pressure increment. Soil Science Society of
America Proceedings 30, no. 3 : 2 9 8 - 3 0 4 .
Fetter, C. W. 1988, Applied Hyarogeology, 2d ed. New York:
Macmillan Publishing Company, 5 8 8 pp.
Gillham, R. W., A. Klute, and D. F. Heermann. 1979. Measurement and numerical simulation of hysteretic flow
in a heterogeneous porous medium. Soil Science
Society of America Journal 4 3 : 1 0 6 1 - 6 7 .
Green, R. E L R. Aliuja, and S. K. Chong. 1986. "Hydraulic
conductivity, diffusiviry, and sorptivity of unsaturated
soils: Field methods." In Methods of Soil Analysis, 1.
Physical and Mineralogical Methods 2d ed., edited

by A. Klute. Agronomy 9 ( 0 : 7 7 1 - 9 8 . Madison, Wis.:


American Society of Agronomy.
Hillel. Daniel. 1980. Fundamentals of Soil Physics NewYork:
Academic Press, Inc., 413 pp.
Hillel, D., and R. S. Baker. 1988. A descriptive theory of fingering during infiltration into layered soils. Soil Science
146:51-56.
Jackson, R. D. 1964. Water vapor diffusion in relatively dry
soil: 1. Theoretical considerations and sorption experiments. Soil Science Society of America Proceedings 28.: 1 7 2 - 7 6 .
James, R. V., and J a c o b Rubin. 1986. Transport of chloride
ion in a water-unsaturated soil exhibiting anion exclusion. Soil Science Society of America Journal
50:1142-50.
Javandel, I., C. Doughty, and C. F, Tsang. 1984. Groundwater
Transport: Handbook of Mathematical Models Water
Resources Monograph 1 0 . Washington, D. C: American Geophysical Union. 228 pp.
Klute, A., and C. Dirksen. 1986. "Hydraulic conductivity and
diffusivity: Laboratory methods." In Methods of Soil
Analysis, I. Physical and Mineralogical Methods. 2d
ed., edited by A. Klute. Agronomy 9 ( 0 : 6 8 7 - 7 3 4 .
Madison, Wis.: American Society of Agronomy.
Kung. K J. S. 1990a. Preferential flow in a sand)' vadose zone:
1. Field observation. Geoderma 4 6 : 5 1 - 8 .
. 1990b, "Preferential flow in a sand)' vadose zone:
2. Mechanism
and
implications."
Geoderma
46:59-71.
Nkedi-Kizza, P., J. W. Biggar, M. Th. van Genuchten, P. J.
Wierenga, H. M. Selim, J. M. Davidson, and D. R. Niel
sen. 1983. Modeling tritium and chloride 36 transport
through an aggregated oxisol. Water Resources Re
search 19, no. 3 : 6 9 1 - 7 0 0 .
Nielsen, D. R., M. Th. van Genuchten, and J. W. Biggar. 1986.
Water flow and solute transport processes in the unsaturated zone. Water Resources Research 22, no.
9:89S-108S.
Pickens, J. F and R W. Gillham. 1980. Finite element anal
ysis of solute trans pod under hysteretic unsaturated
flow conditions. Water Resources Research 1 6 , no.
6:1071-78.
Richards, L A. 1928. T h e usefulness of capillar)- potential to
soil moisture and plant investigators. Journal of Agricultural Research 3 7 : 7 1 9 - 4 2 .

F l o w and M a s s T r a n s p o r t in the Vadose Z o n e

. 1931. Capillar,-conduction o f liquids through porous


mediums. Physics 1 : 3 1 8 - 3 3 .
Sposito, Garrison. 1986. T h e physics of soil water physics.
Water Resources Research 22, no. 9 : 8 3 S - 8 8 S .
Suarez, D. L 1985. Chemical effects on infiltration. In Pro
ceedings National Resources Modeling Symposium,
Edited by D. G. DeCoursey, 4 1 6 - 1 9 . Washington,
D. C. U. S. Department of Agriculture. Agricultural
Research Service.
Suarez, D. L, J. D. Rhoades, R. Lavado, and C. M Grieve. 1984.
Effect of pH on saturated hydraulic conductivity and
soil dispersion. Soil Science Society of America Journal 48, no. 1 : 5 0 - 5 5 .
Van Genuchten, M. Th. 1980. A closed form equation for
predicting the hydraulic conductivity of unsaturated
soils.
Soil Science Society
of America Journal
44:892-98.
. 1981. Analytical solutions for chemical transport with

201

simultaneous adsorption, zero-order production, and


first-order decay. Journal of Hydrology 4 9 : 2 1 3 - 3 3 Van Genuchten, M. Th., J. M. Davidson, and P. J. Wierenga.
1974. An evaluation of kinetic and equilibrium equa
tions for the prediction of pesticide movement in
porous soils. Soil Science Society of America Proceedings 38, no. 1 : 2 9 - 3 5 .
Van Genuchten, M. Th., F. Kaveh, W. B. Russell, and S. R.
Yates. 1988. "Direct and indirect methods of estimating the hydraulic properties of unsaturated soils."
In Land Qualities in Space and Time, edited by J.
Bouma and A. K. Bregt, 61 - 7 2 . Wageningen, the Neth
erlands: International Society of Soil Science.
Van Genuchten, M. Th., and P. J. Wierenga. 1976. Mass transfer studies in sorbing porous media, I, Analytical
solutions. So/7 Science of America Journal 4 0 , no.
4:473-80.

Chapter Five

Multiphase Flow

5.1

Introduction
T h e m o v e m e n t of liquids that are immiscible with water through the vadose z o n e as
well as b e l o w the water table is an important facet of contaminant hydrogeology. Such
liquids are often called n o n a q u e o u s p h a s e l i q u i d s . T h e y may have densities that are
greater than water ( d e n s e n o n a q u e o u s phase liquids, or D N A P L s ) or densities that are
less than water (light n o n a q u e o u s phase liquids, or L N A P L s ) . They may be partiallysoluble in water, so that a dissolved phase as well as a n o n a q u e o u s phase may be present
(Schwille 1 9 8 1 , 1 9 8 4 , 1 9 8 8 ) . T w o - p h a s e flow may o c c u r b e l o w the water table with water
and a DNAPL ( M c W h o r t e r and Sunada 1 9 9 0 ) . Three-phase flow may o c c u r in the vadose
z o n e with air, water, and an NAPL (Abrila and Pinder 1985a, 1 9 8 5 b ) . In the vadose
z o n e the NAPL may partition into the air as a vapor phase ( B a e h r 1 9 8 7 ) . T h e r e may be
multiple c o m p o u n d s in the n o n a q u e o u s phase, each with different properties ( C o r a p
cioglu and B a e h r 1987; B a e h r and Corapcioglu 1 9 8 7 ) . F l o w is dependent upon the
densities, viscosities, and interfacial tensions of the liquids. In addition to dispersion and
diffusion, c o m p o u n d s can undergo adsorption and chemical and biological degradation.
T h e s e p r o c e s s e s p o s e extremely c o m p l e x hydrogeological c h a l l e n g e s s o m u c h s o that.
Pinder and Abrila ( 1 9 8 6 ) had to utilize 27 independent equations in the development
of a c o m p r e h e n s i v e model for the flow of n o n a q u e o u s phase liquids.
T h e b a s i c theory of n o n a q u e o u s phase liquid transport has b e e n worked out bypetroleum reservoir engineers, w h o are c o n c e r n e d with the movement of petroleum
through reservoirs that also contain water (e.g., Buckley and Leverett 1 9 4 2 ) . This theory
has b e e n e x t e n d e d to the vadose zone, since most n o n a q u e o u s phase liquid p r o b l e m s
start with a spill or leak that creates a release a b o v e the water table. In addition, soil
moisture flow in the vadose z o n e may be viewed as a subset of multiphase flow.
C o m m o n LNAPLs include gasoline and diesel fuel. DNAPLs include the chlorinated
h y d r o c a r b o n s , such as trichloroethylene and pentachlorophenol. S o m e industrial ap
plications involve solutions of DNAPLs in mineral oils; the resulting solution is an LNAPL,
even though the pure formulation o f the c o m p o u n d o f c o n c e r n might b e m o r e dense
than watet. F o r e x a m p l e w o o d preservatives are sometimes formulated by dissolving

202

203

Multiphase Flow

p e n t a c h l o r o p h e n o l in a carrier oil. Although p e n t a c h l o r o p h e n o l has a specific gravity of


1.978, the c o m m e r c i a l solution is less d e n s e than water.

5.2

Basic Concepts
5.2.1

Saturation Ratio

T h e s a t u r a t i o n r a t i o of a fluid is the fraction of the total p o r e s p a c e filled with that


liquid. T h e total of the saturation ratios for all the fluids present, including air, add up
to 1.0. Saturation ratio can also be e x p r e s s e d as a p e r c e n t saturation.
5.2.2

Interracial T e n s i o n and Wettability

A liquid in c o n t a c t with a n o t h e r s u b s t a n c e , which c a n be a solid, an i m m i s c i b l e liquid,


or gas, p o s s e s s e s interfacial energy. T h i s energy' is the result of the difference in the
d e g r e e of attraction for the m o l e c u l e s of the s u b s t a n c e at the liquid surface to each
o t h e r c o m p a r e d with their d e g r e e o f attraction for m o l e c u l e s o f the o t h e r substance.
T h i s p h e n o m e n o n is called i n t e r f a c i a l t e n s i o n . It is defined as the a m o u n t of w o r k
necessary to separate a unit area of o n e s u b s t a n c e from another. Units are dynes per
centimeter. T h e interfacial tension b e t w e e n s u b s t a n c e s i and j is designated as a .
uj

Figure 5.1 s h o w s the interfacial tension b e t w e e n two liquids, G and L, and a solid
surface. T h e interface angle b e t w e e n the two liquids is indicated by 8. Equation 5.1 gives
the relationship b e t w e e n 8 and the interface tensions for the three interfaces: liquid
G/solid, liquid i / s o l i d , and liquid G/liquid L .
cos 8 = "

S G

~ "

S L

(5.1)

By convention, 9 is measured through the d e n s e r fluid. In general, o n e or the o t h e r


of the fluids will preferentially spread over, or wet, the entire solid surface. If 9 is less
than 9 0 , then liquid L will preferentially wet the surface. If 8 is m o r e than 9 0 , then

F I G U R E 5.1

Interfacial tensions between a solid surface, a wetting liquid phase, L, and a nonwettlng

gas or liquid phase, G.

204

Chapter Five

liquid G will preferentially wet the surface. That is, if we have two liquids competing for
a surface, o n e dominates and coats the solid surface. In oil-water systems, water tends
to preferentially wet the surface. However, if the surface is dry and first b e c o m e s coated
with oil, then the system is oil-wet, as the water d o e s not c o m e into contact with the
surface. An a u t o m o b i l e that has b e e n w a x e d is an e x a m p l e of an oil-wet surface. Water
will form beads and flow in rivulets rather than move uniformly over the surface, as it
d o e s for an unwaxed car.
Aquifers are naturally water-wet b e c a u s e they contain water before any NAPLs are
discharged to them. T h e vadose z o n e may be either water-wet or oil-wet, depending
u p o n w h e t h e r the soil is moist or dry w h e n the oil is discharged. However, even soil
in the vadose z o n e that appears to be dry will have water held to it by capillary pressures.
At very low water content, water forms p e n d u l a r r i n g s around the grain contact points,
with a thin film of water coating the rest of the grains. This water cannot flow, but it
still coats the mineral grains, making the vadose z o n e water-wet.
5.2.3

Capillary Pressure

If two immiscible liquids are in contact, a curved surface will tend to develop at the
interface. By measuring the p o r e pressure near the interface in each phase, o n e will find
that the pressures are not the same. T h e difference is the c a p i l l a r y p r e s s u r e .
In the vadose z o n e capillary pressure has a negative value. We can also refer to
capillary pressure as a tension, in which case it would have a positive value. If P is the
w

pressure of die wetting fluid and P is the pressure of the nonwetting fluid, then P ,
e

the capillary pressure, is found from Equation 5.2.

p = p - p
c

(5.2)

Figure 5.2 shows the radius of curvature, r', for a spherical air-water interface.
Equation 5.3 gives the relationship between the capillary pressure, P , the interfacial
c

tension, a, and the radius of curvature.

Capillar)' pressure is directly proportional to the interfacial tension and inversely


proportional to the radius of curvature. T h e radius of curvature is dependent upon the
p o r e size and the amount of e a c h fluid present. This means that the capillary' pressure
is a function of the properties of the two immiscible liquids present and is different for
differing p r o p o r t i o n s of water and NAPL in the s a m e p o r o u s media. Further, P is a
c

property of the m a c r o s c o p i c geometry of the void spaces in the p o r o u s media, which


cannot easily be described mathematically.
II" we wish to k n o w the capillary pressure in a real p o r o u s media, we must consider
an analogous medium with known geometry. F o r e x a m p l e , we could consider the real
p o r o u s media to be similar to a bundle of thin glass tubes of radius r. T h e capillary
pressure in a thin tube, P can be determined from Equation 5.4:
(5.4)

Multiphase Flow

205

Solid

Air
/>

Fluid

X e ^ -

Solid

F I G U R E 5.2

Radius of curvature for a spherical capillary interface.

F o r a given p o r o u s medium the relationship of the capillary pressure to the saturation ratio can be determined experimentally in the laboratory. Figure 5.3 s h o w s
capillary pressure curves. If the p o r o u s medium starts off saturated with the wetting fluid
and the wetting fluid is slowly displaced by a nonwetting fluid, thus reducing the wettingfluid saturation ratio, S, the result is a d r a i n a g e , or d r y i n g , c u r v e . As the wettingfluid saturation ratio drops, the capillary pressure will b e c o m e m o r e negative. Eventually
no m o r e wetting fluid will be displaced by the nonwetting fluid, even with further d e c r e a s e s
in capillary pressure. T h i s saturation value is k n o w n as the i r r e d u c i b l e w e t t i n g - f l u i d
s a t u r a t i o n , S , also known a s r e s i d u a l w e t t i n g s a t u r a t i o n . Next, w e will displace
wi

the nonwetting fluid by forcing the wetting fluid into the sample. T h e result is called an
i m b i b i t i o n , o r w e t t i n g , c u r v e . Notice o n Figure 5.3 that the imbibition curve d o e s
not follow the s a m e pathway as the drainage curve. Recall that this p h e n o m e n o n is called
h y s t e r e s i s . T h e r e is no unique relationship b e t w e e n capillary pressure and saturation
ratio; the relationship d e p e n d s upon which fluid is being displaced. W h e n the imbibition
curve r e a c h e s z e r o capillary pressure, s o m e of the nonwetting fluid will remain in the
p o r o u s media. T h i s saturation value i s k n o w n a s the i r r e d u c i b l e , o r r e s i d u a l , n o n w e t t i n g f l u i d s a t u r a t i o n , S .
r

Notice that on Figure 5.3 the drainage curve starts off at a wetting fluid saturation
ratio of 1.0 with a n o n z e r o capillary pressure, P . T h i s is the t h r e s h o l d v a l u e , also
t

called the d i s p l a c e m e n t , i m b i b i t i o n , b u b b l i n g p r e s s u r e , o r a i r - e n t r y , v a l u e .
In order for the nonwetting fluid to start to displace the wetting fluid, this pressure must
be exceeded.
Figure 5 . 4 ( a ) s h o w s an experimentally derived capillary-pressure curve for air and
water in a medium sand. Figure 5 . 4 ( b ) s h o w s a capillary pressure curve for air and

206

Chapter F i v e

F I G U R E 5.3

Capillary-pressurewetting-fluid saturation curves for two-phase flow.

trichloroethylene in the s a m e medium sand. Figure 5 . 4 ( c ) contains an air and water


capillary-pressure curve for fine sand and Figure 5 . 4 ( d ) shows an air and trichloroethylene
(TCE) curve for the s a m e fine sand (Lin, Pinder, and W o o d 1 9 8 2 ) . T h e s e diagrams
illustrate that the capillary-pressure relationships are unique for a given porous media
and d e p e n d as well on the specific immiscible fluid.
5.2.4

Relative Permeability

During simultaneous flow of two immiscible fluids, part of the available p o r e space will
be filled with o n e fluid and the remainder will be filled with the o t h e r fluid. Figure 5.5
shows p o s s i b l e fluid saturation states for water and oil with differing ratios of each and
for both water-wet and oil wet circumstances.

Multiphase Flow

207

B e c a u s e the two fluids must c o m p e t e for s p a c e in which to flow, the c r o s s sectional


area of the p o r e s p a c e available for each fluid is less than the total p o r e space. T h i s
leads to the c o n c e p t of relative permeability. Relative permeability is the ratio of the
intrinsic permeability for the fluid at a given saturation ratio to the total intrinsic permeability of the rock. A relative permeability exists for b o t h the wetting and the nonwetting

Water wet sand

Oil wet sand

Multiphase Flow

209

phase. Figure 5.6 s h o w s two-phase relative-permeability curves for b o t h wetting and


nonwetting liquids.
T h e irreducible water saturation is the water content at which no additional water
will Bow. In Figure 5.6, water is the wetting fluid, and the irreducible water (wettingfluid ) saturation is shown on the left side. T h u s water won't flow at all until the irreducible
wetting fluid saturation, 5. is e x c e e d e d . T h e nonwetting fluid won't begin to flow until
the residual nonwetting-fluid saturation, Sr, is e x c e e d e d . This is shown o n the right
side of Figure 5.6. F o r a two-phase oil-water system, if the water content is less than the
irreducible water saturation, oil c a n flow, but water will be held by capillary forces.
Likewise, at an oil c o n t e n t less than the residual oil saturation, water can flow, but oil
cannot, at least as a separate phase. Oil droplets dispersed in the water can still migrate.
Relative permeability is normally determined by laboratory tests of r o c k - c o r e samples. T h e r e appears to be a hysteresis effect for relative permeability. This is not surprising,
since such an effect exists for capillary pressure. Figure 5 . 7 ( a ) s h o w s experimentally
derived relative-permeability curves in a fine sand for water in the p r e s e n c e of T C E (Lin,
Pinder, and W o o d 1 9 8 2 ) . T h e left curve was measured in a c o r e that initially had a high
saturation ratio of T C E and then had varying amounts of water injected into it to increase
the water saturation ratio. T h e right curve was measured for a c o r e that was saturated
with water and then was injected with T C E . Note that the two curves have greatly different
shapes and that the relative permeability for water can vary substantially, depending
upon whether the particular water-saturation value was r e a c h e d by water displacing T C E
or T C E displacing water. Figure 5 . 7 ( b ) s h o w s experimentally derived relative-permeability
curves for T C E in the p r e s e n c e of water for the s a m e fine sand as Figure 5 . 7 ( a ) (Lin,
Pinder, and W o o d 1 9 8 2 ) . T h e two curves, representing water displacing T C E and T C E

Nonwetting fluid saturation ralio


.r 0.0

1.0

0
0.0

I ii

Wetting fluid saturation ratio


F I G U R E 5.6

Typical relative permeability curves for a two-phase system.

F I G U R E 5.7

Experimentally derived relative permeability curves for (a) water with respect to TCE a

(b) for TCE with respect to water. Source: C. Lin, G. F. Pinder, and E. F. Wood, Water Resources Prog
Report 8 3 - W R - 2 (Princeton, N. J . Princeton University, 1982).
:

Multiphase Flow

211

displacing water, are m u c h c l o s e r than the curves for relative permeability of water.
However, they are not exactly the same. Hysteresis is an added c o m p l e x i t y to analytical
treatment of multiphase flow. Parker and L e n h a r d ( 1 9 8 7 ) and Lenhard and Parker ( 1 9 8 7 a )
have p r o p o s e d m o d e l s to d e s c r i b e the hysteretic relations of saturation-pressure ratios
and permeability saturation relations.
Relative permeabilities o f three phase systems o f a i r - w a t e r - n o n a q u e o u s phase
liquids are m o r e c o m p l e x . A m e t h o d of estimating these values from data developed for
two-phase systems was devised by S t o n e ( 1 9 7 3 ) . First the relative permeability of water,
k, , as a function of the saturation ratio for water, S , is o b t a i n e d for a w a t e r - n o n a q u e o u s
phase system. Next, the relative permeability of air, <!., as a function of the saturation
ratio for air, S, is determined for an a i r - n o n a q u e o u s phase system. Finally, the relative
permeability of the n o n a q u e o u s p h a s e liquid, k, , in the three-phase system is determined
from Equation 5.5 (Faust 1 9 8 5 ) :
w

(5.5)

ft,

where
l$ = the relative permeability of the n o n a q u e o u s p h a s e at the residual
v

saturation of water in a w a t e r - n o n a q u e o u s phase system


k,

= the relative permeability of the n o n a q u e o u s phase system as a

wn

function of S

k, = the relative permeability of the n o n a q u e o u s p h a s e in an


a

a i r - n o n a q u e o u s p h a s e system as a function of S .
a

Figure 5.8 is a ternary relative permeability diagram for the n o n a q u e o u s phase


liquid in an air w a t e r - n o n a q u e o u s phase systemi.e., the vadose z o n e w h e r e the
water is the wetting fluid (Faust 1 9 8 5 ) .

5.2.5

Darcy's L a w For T w o - P h a s e F l o w

Darcy's law for the steady-state saturated flow of water in the p r e s e n c e of a n o n a q u e o u s


phase liquid is given (Schwille 1 9 8 4 ) as

k kp
/*
rw

dh
dl

where
Q = volume of water flowing
w

k, = relative permeability of water in the p r e s e n c e of the nonwetting fluid


w

kj = intrinsic permeability of the rock


p. = density of the water
p = dynamic viscosity of the water
w

A = c r o s s sectional area of flow


dhjdi = gradient of the head of the water

212

Chapter Five

F I G U R E 5.8

Ternary diagram showing the relative permeability of the NAPL phase in an air-water-

NAPL system as a function of phase saturation. Source: C. R. Faust, Water Resources Research 2 1 , no.
4 ( 1 9 8 5 ) : 5 8 7 - 9 6 . Copyright by the American Geophysical Union.

A similar e x p r e s s i o n for the nonwetting fluid is

H
5.2.6

at

Fluid Potential and Head

l l u h b e r t ( 1 9 5 3 ) defined fluid potential, <t>, as the amount of work needed to move a


unit mass of fluid from s o m e standard position and condition to a different position
and condition. Position represents the potential energy of the fluid or elevation a b o v e
the standard datum. T h e condition can be represented by the difference in pressure
b e t w e e n the position under consideration and the standard pressure.
T h e fluid potential is thus defined as
<t> = g(z-z,) + (.P-P,)v

where
g = acceleration of gravity
z = elevation
z, = standard elevation
P pressure
P = standard pressure
s

v = volume per unit mass


m

(5.8)

Multiphase

Flow

213

S i n c e volume p e r unit mass is the reciprocal of density, p. Equation 5.8 can be


e x p r e s s e d as
<t> = g(z-z ) + -

(5.9)

P
If the standard pressure is taken as a t m o s p h e r i c and z i s defined as elevation above
a convenient datum, such as sea level, then Equation 5.9 b e c o m e s
P

<S> = gz + -

(5.10)

P
If a pipe with an o p e n b o t t o m is inserted into an aquifer to a point at distance z
a b o v e the datum, the fluid pressure at that location will cause the fluid in the aquifer to
rise to a height h a b o v e the datum. T h e fluid pressure is equal to the weight of the fluid
in the pipe p e r unit cross-sectional area:

(5.11)

P=pg(h-z)
This can be substituted into Equation 5.10 to yield

e s v ^ i l

g h

p
w h e r e h is the total head.
Fluid will flow from an area of higher fluid potential, <1> + A<1>, to an area of lower
fluid potential, <t>. T h e force p e r unit mass e x e r t e d on the fluid by its environment is a
vector, E. This force vector is perpendicular to the equipotential surfaces and in the
direction of decreasing potential. It has a magnitude equal to the c h a n g e in potential,
A<t>, divided bv the distance over which the c h a n g e in potential is measured, An (Figure
5.9):

F I G U R E 5.9

Relation between force field and potential gradient. Source-. Modified from M. K. Hubbert,

American Association of Petroleum Geologists Bulletin 37 (1 9 5 3 ) : 1 9 5 4 - 2 0 2 6 ,

Chapter Five

(5.13)
T h e force vector E can be e x p r e s s e d several other ways:

E = -grad<D

(5.14)

E=-ggrad*

(5.15)

(5.16)

= g--gradP
P

Equation 5.16 shows that at a point a unit mass of a fluid will be acted upon by
a force E, which is the vector sum of gravity and the negative gradient of the pressure
divided by the fluid density. Figure 5.10 shows the vector c o m p o n e n t s of the force vector
E for ( a ) the hydrostatic c a s e where E = 0 and ( b ) the hydrodynamic case where E # 0.
Equation 5.16 shows that the direction of the force vector is a function of the fluid
density. T h u s for the same point in the aquifer, different fluids will have different force
vectors and, h e n c e , different flow directions in the same potential field. Consider Figure
5.11, which has a force vector for water, E, a force vector for an LNAPL, E L N A P L , and
a force vector for a DNAPL, E D N A P L . F o r convenience, the water is shown to b e flowing
horizontally; that is, the force vector E is horizontal, although it could be going in any
direction. Since the density of the LNAPL is less than the density of water, the vector
grad / y p p L longer than the vector grad P/pw and the resulting vector E L N A P L
is angled upward c o m p a r e d with E w . T h e vector E D N A P L is angled downward because
the vector grad P / P D N A P I . ' shorter than grad P/p \PLi s

L N A

FIGURE 5.10

Vector components of the force vector for (o) hydrostatic case and (b) hydrodynamic case.

Source: M. K. Hubbert, American Association of Petroleum Geologists Bulletin 37 ( 1 9 5 3 1 : 1 9 5 4 - 2 0 2 6 .

Multiphase Flow

215

LNAPL

F I G U R E 5.11

Force vectors for a DNAPL, water, and an L N A P f in the same potential field. The DNAPL

sinks and the 1NAPL rises with respect to the direction of ground-water flow. Source: Modified from M. K.
Hubbert, American Associofion of Petroleum Geologists Bulletin 37 (1953): 1 9 5 4 - 2 0 2 6 .

T h i s figure illustrates why a DNAPL will sink a n d an LNAPL will rise with r e s p e c t
to the d i r e c t i o n of g r o u n d w a t e r flow in the s a m e potential field.
T h e fluid potential of a n o n w e t t i n g fluid, e i t h e r an LNAPL or a DNAPL, is given by

* =

(5.17)

216

Chapter

Five

and the fluid potential for water is


p

<D = g z +

(5.18)

If we solve Equation 5.18 for P a n d substitute it into 5.17, we obtain


P

- " ~ " gz

(5.19)

This e x p r e s s i o n relates the fluid potential of a nonwetting fluid to the fluid potential
of water at the s a m e location.
From Equation 5.12, <b =
nw

gh
w

and <I> =

>

gh ,
w

- e ^ j ^

Equation 5.19 can b e written as


z

Pnw

In Equation 5.20, z is the elevation of the point in the aquifer, h is the height
a b o v e the datum that water would stand in an o p e n pipe terminating at the point, and
h is the height that a nonwetting fluid of density p would stand. Figure 5.12 illustrates
K

nw

FIGURE 5.12

Total head, h, pressure head, P/pg, and elevation head, z, for open pipes filled with (a)

water, (b} an LNAPL, and (c) an NAPL. All pipes have the same pressure at the open end.

Multiphase Flow

217

the relationships b e t w e e n b , 6
, and 6
. T h e fluid elevation i n the pipe filled
with LNAPL will be higher than the pipe filled with water, w h e r e a s the fluid elevation of
the pipe filled with DNAPL will be lower.
w

5.3

L N A P L

D N A P L

Migration of Light Nonaqueous Phase Liquids ( L N A P L s )


Light n o n a q u e o u s p h a s e liquids are less d e n s e that water. W h e n spilled at the land
surface, they migrate vertically in the vadose z o n e under the influence of gravity and
capillary forces, just as water does. Unless the vadose z o n e is e x t r e m e l y dry, it will be
water-wet, and the LNAPL will be the nonwetting phase.
Figure 5.13 s h o w s the distribution of water in the vadose z o n e (Abdul 1 9 8 8 ) . Notice
that at the top of the vadose z o n e the water is held at the irreducible water saturation.
T h e water held h e r e is called p e n d u l a r water. B e l o w that is a z o n e w h e r e the water
c o n t e n t is a b o v e the irreducible saturation; this is s o m e t i m e s called f u n i c u l a r water.
W h e n c l o s e to 1 0 0 % water saturation is reached, we find the capillary fringe. T h e air
water relationship of the vadose z o n e behaves as a two-phase i m m i s c i b l e flow, so there
is residual air saturation in the capillary z o n e . T h i s is s h o w n in Figure 5.13; however,
we will ignore the residual air saturation in further analysis b e c a u s e it usually is a small
value.
T h e capillary fringe is n o t a regular surface, such as the water table. T h e height of
the capillary rise will be different in e a c h vertical set of i n t e r c o n n e c t e d pores, depending

F I G U R E 5.13

Vertical distribution of water in the vadose zone in the absence of nonaqueous phase

liquids. Source: A. S. Abdul, Ground Water Monitoring Review 8, no. 4 (1988): 7 3 - 8 1 . Copyright 1 9 8 8
Water W e l l Journal Publishing Co.

218

Chapter F i v e

upon the mean p o r e diameter of the set. T h u s the capillary fringe has a ragged upper
surface. However, for the sake of simplicity in diagrams of the capillary fringe, we will
s h o w it as a level surface. We can use the capillary tube model to estimate the average
height of the capillary fringe. Equation 5.4 can give us the capillary pressure, P b a s e d
on a mean p o r e radius, r. T h i s is equal to the weight of the water in the capillary tube,
which is found by multiplying the height of the water in the tube, h , by the specific
weight of water, y.
c

(5.21)

(5.22)
For pure water in a clean glass tube, 9 can be taken as zero, and c o s 0 is 1.0. T h e
value o of for water at 2 0 C is 0 . 0 7 4 g/cm. With these values, Equation 5.22 b e c o m e s
0.15
4 =
r

centimeter

(5.23)

T a b l e 5.1 shows the heights of the capillary fringe that were observed experimentally
in various materials. T h e visual capillary height is the level where the water saturation
ratio is c l o s e to 1.0. T h e capillary water in the funicular z o n e is above this height, although
it is not visible.
Figure 5.13 shows the capillary fringe as extending to the height where the watersaturation ratio begins to decline. This height is based on the larger pores, in which the
capillary rise would be least. F o r the smaller pores, the capillary rise would be greater,
e x t e n d upward into what is labeled the funicular zone. Water in that zone is not moving
downward but is being held in place by capillary forces. This illustrates the irregularity
o f the capillary fringe. W e use the phrase c a p i l l a r y z o n e t o mean the part o f the
capillary fringe w h e r e the water saturation ratio is at or c l o s e to 1.0.
T h e LNAPL will travel vertically in the vadose zone. If a sufficient quantity is present
so that the residual LNAPL saturation is e x c e e d e d , it will eventually reach the top of the
capillar)' zone. However, m u c h of the LNAPL may remain behind, trapped in the vadose
zone. E c k b e r g and Sunada ( 1 9 8 4 ) studied the distribution of oil in the vadose zone.

T A B L E 5.1

Visual capillary rise in unconsolidated materials (porosity of alt samples

is about 4 1 % ) .
Material
Fine gravel

Grain S i z e (mm)
2 5

Very coarse sand

Capillary R i s e (cm)
2.5
6.5
13.5

Coarse sand

0.5

Medium sand

0.2

0.5

24.6

Fine sand

0.1

0.2

42.8

Silt

0.05

0.1

105.5

Fine silt

0.02

0.05

200 +

Source: Lohman (1972).

219

Multiphase Flow

Figure 5.14 s h o w s the c h a n g e s in the distribution of water and oil in a sand c o l u m n


into which a quantity of oil was added. Note that much of the oil remains throughout
the thickness of the vadose z o n e as a residual oil.
In moving downward, an LNAPL may displace s o m e of the capillary' water in the
vadose zone, causing it to m o v e ahead of the advancing LNAPL front. O n c e the capillary
z o n e is reached, LNAPL will begin to accumulate. Initially, the LNAPL will be under
tension, just as the water in the vadose z o n e is under tension. As additional LNAPL
accumulates a b o v e the capillary z o n e , an "oil table" will develop, with s o m e LNAPL having a positive p o r e pressure. T h e capillary z o n e will b e c o m e thinner, and m o b i l e , or "free,"
LNAPL will accumulate. Eventually, the capillary z o n e may disappear altogether and the
oil table will rest directly on the water table. In the c o r e of a thick z o n e of m o b i l e LNAPL,
the water table may be d e p r e s s e d by the weight of the LNAPL.
Abdul ( 1 9 8 8 ) c o n d u c t e d an e x p e r i m e n t to o b s e r v e the d e v e l o p m e n t of an oil
table. Vertical c o l u m n s w e r e p a c k e d with sand and partially filled with water. T h e c o l u m n s
had m a n o m e t e r / t e n s i o m e t e r s installed at various heights to m e a s u r e the pore-water

O i l or water content
F I G U R E 5.14

O i l or water content

Changes in the vertical distribution of oil with time after a slug of oil is added to the top

of a column of sand. Oil content and water content are expressed as a fraction of the total volume of the
porous media. Source: D. K. Eckberg and D. K. Sunada, Water Resources Research 20, no. 1 2 (1 9 8 4 ) : 1 8 9 1 97. Copyright by the American Geophysical Union.

F I G U R E 5.15

Hydrostatic pressure head/tension distribution in a sand column to which oil is being added

to the top of the column, (a) Before the addition of the oil, (b) after addition of the oil showing the
development of an oil fringe, (c) after addition of sufficient oil for an oil table to form, and (d) after sufficient
mobile oil has accumulated to eliminate the water capillary fringe. Source: A. S. Abdul, Ground Water
Monitoring Review 8, no. 4 (1 9 8 8 ) : 7 3 - 8 1 . Copyright 1988 Water W e l l Journal Publishing Co. Used
with permission.
220

Multiphase Flow

221

pressure b e l o w the water table and tension a b o v e the water table. T h e elevation w h e r e
the gauge pressure is z e r o is the water table. Figure 5 . 1 5 ( a ) s h o w s the distribution of
p o r e pressures b e f o r e any oil was added. Oil was then added to the t o p of the soil
column, and the system was allowed to c o m e into equilibrium. Figure 5 . 1 5 ( b ) r e c o r d s
the conditions after oil was added. T h e capillary fringe thinned, and the water table r o s e
as the advancing oil displaced capillary water downward. Oil under tension accumulated
above the capillary z o n e . Figure 5 . 1 5 ( c ) s h o w s that further addition of oil resulted in
the formation of an oil table with positive p o r e pressure a b o v e the water capillar)' zone,
which was still u n d e r tension. Eventually e n o u g h oil was added so the water capillary
fringe disappeared and the oil table rested directly on the water table. An oil capillary
fringe existed a b o v e the oil table (Figure 5 . 1 5 ( d ) ) .
T h e m o b i l e LNAPL can migrate in the vadose z o n e following the s l o p e of the water
table. Figure 5 . 1 6 s h o w s the shape of a migrating spill of an LNAPL. Note that there is
residual LNAPL in the unsaturated z o n e a b o v e the m o b i l e LNAPL. In addition s o m e of
the LNAPL c o m p o n e n t s can dissolve in the g r o u n d water and m o v e by diffusion and
advection with the g r o u n d water. F o r e x a m p l e , gasoline will release significant a m o u n t s
o f b e n z e n e , toluene, ethylbenzene, and xylene ( B T E X ) a s soluble fractions.
T h e residual NAPL material in the vadose z o n e can partition into the vapor phase
as well as a soluble p h a s e in capillary water. T h e d e g r e e of the partitioning will d e p e n d
upon the relative volatility of the material and its solubility in water.

223

Multiphase Flow

H e n r y ' s l a w states there is a linear relationship b e t w e e n the vapor pressure of


a solute a b o v e its a q u e o u s solution and the c o n c e n t r a t i o n in solution. T h e proportionality
constant b e t w e e n the two i s called a H e n r y ' s l a w c o n s t a n t , which can b e e x p r e s s e d
i n units o f a t m o s p h e r e s / ( m o l e s / c u b i c m e t e r w a t e r ) ( S e c t i o n 7 . 2 ) . T h e proportionality
constant from Henry's law has also b e e n e x p r e s s e d as a water-air partition coefficient.
This is the ratio of the a q u e o u s solubility of a substance, e x p r e s s e d in milligrams per
liter at a given temperature to the saturated vapor c o n c e n t r a t i o n of the pure p h a s e of
the substance, also e x p r e s s e d in milligrams per liter ( B a e h r 1 9 8 7 ) .
T h o s e c o m p o u n d s with low water-air partition coefficients, such as the alkanes,
favor the vapor phase, whereas t h o s e with high water air partition coefficients, such as
b e n z e n e , favor the a q u e o u s phase. H y d r o c a r b o n s such as gasoline are a mixture of up
to 2 0 0 different organic c o m p o u n d s ; therefore, various water-air partition coefficients
are n e e d e d t o d e s c r i b e the behavior o f the various constituent c o m p o u n d s . T h e diffusive
properties of the soil are also very important in controlling vapor p h a s e transport ( B a e h r
and Corapcioglu 1 9 8 7 ) .
B a e h r ( 1 9 8 ^ ) developed a model to d e s c r i b e the vapor p h a s e and a q u e o u s transport
of residual h y d r o c a r b o n s in the vadose z o n e . Figure 5.17 s h o w s the partitioning of
h y d r o c a r b o n mass from gasoline into vadose z o n e water as a function of time. This
figure shows that the aromatic c o m p o u n d s , t h o s e b a s e d on the b e n z e n e ring, partition
into water at a higher rate and for a longer time period than the n o n a r o m a t i c c o m p o u n d s .
T h i s i s t o b e e x p e c t e d , s i n c e the n o n a r o m a t i c h y d r o c a r b o n s studied, C

alkenes, C 5

(. alkanes .iml ( uapihenes h a w nnii h l< >v\ci waiei aii p.irlilii in o lefiii ients than the
aromatic constituents, b e n z e n e , toluene, ethylbenzene, xylene, etc. T a b l e 5.2 gives waterair partition coefficients for selected gasoline constituents. T h e selective partitioning of
b e n z e n e , toluene, ethylbenzene, and xylene in the a q u e o u s capillary p h a s e helps to
explain why these c o m p o u n d s are so diagnostic of a gasoline spill. T h e y can reach the
water table via infiltration of capillary water through a z o n e of residual gasoline, even if
no gasoline itself reaches the water table. Figure 5.18 illustrates this p h e n o m e n o n .

T A B L E 5.2

Water-oi partition coefTicie nts for selectee organic compounds.

Compound

Formula

Molecular

Wafer-Air

Weight

P a r t i t i o n Coefficient

Aromatics
Benzene

QH

Toluene

C H

o-Xylene

C H,

Ethylbenzene

C Hi

78

5.88

92

3.85

106

4.68

106

3.80

Nonaromatics
Cyclohexane

C H,j

84

0.15

1-Hexene

CH

84

0.067

n-Hexane

C.H

86

0.015

n-Octane

1 4

114

0.0079

Source: A. I. Baehr. Wofer Resources Research 23, no. 10; 1928. Published 1987 by Americon Geophysical
Union. Used with permission.

224

F I G U R E 5.18

Chapter Five

Process of ground water being contaminated by gasoline constituents from residual

in the vadose zone.

Multiphase Flow

5.4

225

Measurement of the Thickness of a Floating Product


T h e m e a s u r e m e n t of the a m o u n t of m o b i l e LNAPL a b o v e the water table is not straight
forward. Figure 5.19 s h o w s the distribution of an LNAPL a b o v e the water table for the
condition w h e r e a water capillary z o n e exists. T h i s diagram s h o w s that there is a z o n e
of i m m o b i l e LNAPL a b o v e the capillary z o n e w h e r e the LNAPL c o n t e n t is less than the
residual LNAPL saturation. W h e n the LNAPL c o n t e n t e x c e e d s the residual LNAPL saturation
and the sum ot the water saturation and the LNAPL saturation is 1 0 0 % , there will be
positive p o r e pressures. In this z o n e LNAPL will be m o b i l e and can flow laterally into a
monitoring well. T h e s c r e e n z o n e o f the monitoring well must thus e x t e n d a b o v e the
top of the z o n e of free or m o b i l e LNAPL. T h e water level in the monitoring well will
initially be at the water table, which is b e l o w the level of the b o t t o m of the m o b i l e LNAPL
z o n e . T h e LNAPL will flow down the monitoring well to the water table. T h e weight of
the LNAPL will then d e p r e s s the surface of the water in the monitoring well b e l o w that

FIGURE 5.19

Comparison of distribution of mobile oil in an aquifer with the thickness of floating oil in

a monitoring well for the case where a water capillary fringe exists below the zone of mobile oil.

Chapter Five

of the water table. As a result the thickness of LNAPL measured in a monitoring well is
greater than the actual thickness of the m o b i l e LNAPL in the vadose zone. This effect is
greater for thin z o n e s of free LNAPL, w h e r e the capillary z o n e may be much thicker,
than for a thick z o n e of free LNAPL, w h e r e the capillary z o n e may be thin or missing.
It is also greater in fine-grained materials, w h e r e the capillar)' fringe may be thicker.
T h e depth b e l o w the water table at which the LNAPL will reach equilibrium in a
monitoring well can be calculated. At equilibrium, the pressure in the monitoring well
on the LNAPL side of the interface is P , and the pressure on the water side of the
0

interface is P . T h e two pressures must be equal:


w

P = P

(5.24)

P is equal to the density of the oil, p, times the thickness of the oil layer, T.
0

P = pj

(5.25)

P is equal to the density of the water, p , times the distance from the water table to
the interface, W.
w

P = pW

(5.26)

P T= p W

(5.27)

W= (j^J 7'

(5.28)

Since P = P, then
0

Solving for ll yields

Farr, Houghtalen, and McWhorter ( 1 9 9 0 ) and Lenhard and Parker ( 1 9 9 0 ) developed


two m e t h o d s to estimate the volume of recoverable LNAPL in an aquifer based on the
thickness of the LNAPL floating in a monitoring well. T h e s e methods are based on the
capillar)- soil properties. O n e of the two m e t h o d s is based on the determination of soil
properties as reported by B r o o k s and Corey ( 1 9 6 6 ) . We will look at this method in
s o m e detail using the derivation of Farr, Houghtalen, and McWhorter.
7" as shown in Figure 5.19 is the difference between the depth to the water o i l
interface in the well, DJ and the depth to the oil-air interface. D". T h e values of the
w

depth to the oil table in the aquifer, D" , and the depth to the top of the capillar)- fringe,
a

D, can be computed.
D" = D " ' -
Pog

(5.29)

D7 = D7 - ,

(5-30)

s
(p* - Po)g
PT

where
PT = the B r o o k s - C o r e y air-organic displacement pressure
P J

= the B r o o k s - C o r e y organic-water displacement pressur

g = the acceleration of gravity

Multiphase Flow

227

Equation 5.30 may be rewritten as

D"' = D" +T- ,

" ,
(Pw - P)g

(5.31)
w

If any of the organic liquid exists at a positive p o r e pressure, then D


and from Equation 5 . 3 1 ,

will be

greater than

T > , .
(P - P)g

(5.32)

If the organic liquid is all under tension in the capillary zone, then there will be
no m o b i l e organic layer and no organic liquid will collect in the monitoring well. Under
these conditions, Equations 5.29, 5.30, 5.31, and 5.32 are not applicable. However, as
s o o n as free organic liquid appears in the aquifer, it will collect to a depth of at least
PT/(P

~ Po)gT h e total volume of nonresidual organic liquid in the vadose z o n e is given by

K = " | / o l d " SJ dz - J ; j ( l - (S + S,)] dz^


D

(5.33)

where
1 / = the volume o f organic liquid per unit area
n = the porosity
S = the water-saturation ratio
w

S = the organic liquid saturation ratio


z = the vertical coordinate measured positively downward
w

D, = a value determined from Equation 5.30


D" = a value determined from Equation 5.29
fl'" = the top of the z o n e w h e r e nonresidual oil o c c u r s
B a s e d on w o r k by Lenhard and Parker ( 1 9 8 7 , 1 9 8 8 ) , the fluid-content relations

S -S
0

= ( l - S )
t

S, + S =
w

1,
/P"

" H i *
5 = 1,

P >P7

(5.34a)

P" <P7

(5.34b)

p'; > PT

(5.35a)

<P7

(5.35b)

where
S = the irreducible water saturation
wi

X = the B r o o k s - C o r e y pore-size distribution index


In addition,

P" = p g(D: - (P'f/Pog) - z) + P"


c

P7 = g ( p . -

(5.36)

Po)

"'

PT
(Pw - P,)g

+ PT

(5.37)

228

Chapter Five

ow

Integration o f Equation 5.33 for D > 0, using Equations 5.34, 5.35, 5.36, and 5.37,
yields the following. F o r X n o t equal to 1,
<3>(1
V, =

~S )D
wi

X + (1

(5.38a)

-X)

1 X

F o r X equal to 1,

V = n(\ - S )U - D(\ + In D]
0

wi

(5.38b)

where

PT

D =

(Pw - P,)g
T =

D -

DJ

Po.

PT

>

(Pw - Po)g
If organic liquid a b o v e the residual saturation exists all the way to the land surface,
w

then D " d o e s not exist. U n d e r this condition integration o f Equation 5.33 yields the
a

following. F o r X not equal to 1,


V =
0

(1

-S )^(T-D)
wi

Pogd
PT

(Pw-

( p - P)i?(l

(5.39a)

F o r X equal to 1,
V; = n ( l - S,) ( T - D )

lnr + l n C

(5.39b)

(Pw - Po)i?
Pg
T h e B r o o k s - C o r e y soil parameters can thus be used to estimate the volume of
recoverable organic liquid in an aquifer b a s e d on the thickness of the organic liquid in
the well, the measured depths in the well to the air-organic interface, and the organic
water interface, along with the densities of the organic liquid and the water. O n e must
measure the B r o o k s - C o r e y soil parameters and the densities in the lab. T h e weakness
in the B r o o k s - C o r e y approach is that it may not be accurate for very small volumes of
m o b i l e organic liquid in the soil. F r o m Equation 5.32 the thickness of organic liquid in
the well is at least

PT/Kp p )g\,
w

even for very small volumes o f m o b i l e LNAPL.

Farr, Houghtalen, and McWhorter ( 1 9 9 0 ) also presented an alternative m e t h o d of


analysis b a s e d on soil parameters developed by van G e n u c h t e n ( 1 9 8 0 ) . However, the
equations b a s e d on the van G e n u c h t e n soil parameters are nonlinear and can't be solved
analytically; they must be solved numerically. T h u s this model is not as convenient as
the o n e b a s e d on a B r o o k s - C o r e y soil. However, under certain conditions, such as thin
layers of m o b i l e organic liquid in the soil, the van G e n u c h t e n model may be m o r e
accurate than the B r o o k s - C o r e y model.
Figure 5 . 2 0 ( a ) s h o w s a graph of the volume of gasoline in a sandstone versus the
measured thickness in a well c o m p u t e d by both the B r o o k s - C o r e y soil model and the
van G e n u c h t e n soil model. Figure 5 . 2 0 ( b ) s h o w s the s a m e thing for a different organicliquid, Soltrol (Farr, Houghtalen, and McWhorter 1 9 9 0 ) . It can be s e e n that for gasoline
in sandstone, the van G e n u c h t e n and B r o o k s - C o r e y m o d e l s yield quite similar results.

Depth of Solirol in well (cm)


Van Genuchten

+ Brooks-Corey
(b)

F I G U R E 5.20

Volume of recoverable LNAPL in a porous media as a function of the depth of product

floating in a monitoring well as computed by the method of Farr et al. (1 990) based on both van Genuchten
and Brooks-Corey soils models for (a) gasoline and (b) Soltrol. Source-. A. M. Farr, R. J. Houghtalen, and
D. B. McWhorter, Ground Water, 2 8 , no. 1 ( 1 9 9 0 ) : 4 8 - 5 6 . Copyright 1 9 9 0 Water W e l l Journal
Publishing Co. Used with permission.

229

F I G U R E 5.21

Volume of recoverable LNAPL in a porous media as a function of the depth of product

floating in a monitoring well as computed by the method of Farr et al. (1990) based on a van Genuchten
soil model (a) for various L N A P L s in a sandstone and (b) for gasoline in various porous media. Source:
A. M. Farr, K. J. Houghtalen, and D. B. McWhorter, Ground Water 28, no. 1 (1990): 4856. Used with
permission. Copyright 1990 Water W e l l Journal Publishing Co.

230

Multiphase Flow

23)

However, for Soltrol, they do not, b e c a u s e the B r o o k s - C o r e y m o d e l has a s o m e w h a t


greater thickness of organic liquid in the well for very small volumes in the soil. Figure
5 . 2 1 ( a ) s h o w s the c o m p u t e d volume-thickness relationships, b a s e d on the van G e n u c h t e n
soil model, for several LNAPLs in a sandstone. Figure 5 . 2 1 ( b ) s h o w s the c o m p u t e d
volume-thickness relationships, again based on the van G e n u c h t e n soil model, for gasoline
in a n u m b e r of different p o r o u s media (Farr, Floughtalen, and M c W h o r t e r 1 9 9 0 ) . T h e s e
diagrams illustrate the fact that there is no simple, c o n s t a n t relationship b e t w e e n the
volume of an LNAPL in a soil and the thickness as m e a s u r e d in a monitoring well. It is
a function of the properties of b o t h the soil and the organic liquid.

5.5

Effect of the Rise and Fall of the Water Table on the Distribution of
LNAPLs
T h e flow of LNAPLs is c o m p l i c a t e d by the rise and fall of the water table with the seasons.
Figure 5 . 2 2 ( a ) s h o w s a layer of oil floating on the surface of the capillary z o n e . As the
water table falls, the layer of m o b i l e oil also falls. Residual oil is left in the vadose z o n e
a b o v e the oil table as it falls. T h i s is illustrated in Figure 5 . 2 2 ( b ) . W h e n the water table
rises, the oil table also rises. However, as Figure 5 . 2 2 ( c ) illustrates, residual oil is left
b e h i n d in the saturated z o n e . If the water table rises faster than the oil table c a n rise,
" p o c k e t s " of free oil might b e c o m e left b e l o w the water table. T h e flow of water and
h y d r o c a r b o n s is c o n t r o l l e d by Darcy's law and d e p e n d s upon the effects of density,
viscosity, and relative permeability. Depending upon these factors, either the hydrocarbon
or the water could have a greater velocity as the water table rises and falls.
In cleaning up LNAPL spills, the m o b i l e LNAPL can be removed by skimming wells
or trenches. However, a c o n s i d e r a b l e a m o u n t of LNAPL will be left as a residual on the
soil. Volatile LNAPLs can be removed by a soil-vapor extraction system. However, nonvolatile products will remain b e h i n d in the soil. T h e a m o u n t d e p e n d s upon the properties
of the LNAPL and the texture of the soil. T h e oil retention capacity of soil is estimated
3

to range from 5 L / m for gravel to 40 L / m for silry sand ( T e s t a and Paczkowski 1 9 8 9 ) .


Many h y d r o c a r b o n s can be degraded by soil bacteria, especially if the soil is a e r o b i c .
Systems that diffuse air into the soil have b e e n effective in bioremediation of h y d r o c a r b o n
spills. T h i s p r o b l e m is addressed in Chapter 9.

5.6

Migration of Dense Nonaqueous Phase Liquids


5.6.1

Vadose Zone Migration

D e n s e n o n a q u e o u s phase liquids (DNAPLs) have a specific gravity greater than 1. W h e n


spilled on the land surface or discharged to the subsurface, o n c e the residual saturation
value is e x c e e d e d they m o v e vertically in the vadose z o n e under the influence of gravity.
Since it is the wetting liquid, water o c c u p i e s the smaller p o r e s and capillary channels in
the vadose z o n e . T h e DNAPL migrates through the larger p o r e openings, which initially
have water coating the mineral grains, with air filling the remainder of the p o r e . T h e
DNAPL displaces the air, so the p o r e b e c o m e s filled with the small a m o u n t of water
wetting the mineral surface and the DNAPL. T h e vadose z o n e permeability for die DNAPL

/( Oil saturation
l(K) X 6 0 4 0 20 0
Vadose
zone

Capillary fringe

Saturated
zone

0 2 0 4 0 6 0 80 100
% Water saturation

','( Oil saturation


100 St) 6 0 4 0 20 0

\ atlose
zone

Mobile oil

Capillar)' fringe
Saturated zone

FIGURE 5.22

Effect of a falling and then rising water table on the distribution of mobile and residual

phases of an LNAPL.
232

0 20 4 0 6 0 8 0 100
% Water saturation

Multiphase Flow

233

is greater than for water b e c a u s e the p o r e s through which the DNAPL, as the nonwetting
fluid, is migrating are larger than the p o r e s through which the water, as the wetting fluid,
migrates.
W h e n the DNAPL reaches the capillary zone, w h e r e the p o r e s are all filled with
water, it must start to displace water in o r d e r to migrate downward. T h i s displacement
of water with DNAPL also o c c u r s b e l o w the water table.
5.6.2

Vertical M o v e m e n t in the Saturated Z o n e

If water in the unsaturated z o n e is static, the DNAPL will c o n t i n u e to migrate downward


under the force of gravity. In o r d e r to displace the water filling the p o r e s , the DNAPL
must have sufficient mass to o v e r c o m e the capillary forces that hold the water in the
p o r e . Vertical stringers of DNAPL can o c c u p y vertically c o n n e c t e d pores. W h e n the vertical
stringer has a sufficient height, its weight can displace the water in the pore. T h e critical
height, b , can be determined from H o b s o n ' s formula ( B e r g , 1 9 7 5 ) :
c

20-cos
h

9(1/r,

g(p

!/>,)

(5.40)

~ Po)

where
cr = interfacial tension b e t w e e n the two liquids

6 = wetting

angle

r, = pore-throat radius
r

p o r e radius

g = acceleration of gravity
Pw = density o f water
Po = density of the DNAPL
F o r a well-rounded, well-sorted sediment of diameter d with r h o m b o h e d r a l packing, the p o r e throat radius, r and the p o r e radius, r , can be estimated from the following
p

formulas:
r

(5.41)
(5.42)

0.2Ud

r, = 0.011 d

T h e smaller the sediment diameter, the smaller b o t h the pore-throat radius and
p o r e radius. Equation 5.40 demonstrates that the value of h is inversely related to the
p o r e diameter. F o r finer grained materials, the DNAPL stringer must be longer than for
c o a r s e r materials. Even thin, fine-grained layers could act as confining layers for a DNAPL.
0

If sufficient a m o u n t s of the DNAPL are present to o v e r c o m e the capillary pressure


and e x p e l water from pores, DNAPL will continue to migrate downward under the force
of gravity until it r e a c h e s an aquitard layer, w h e r e the p o r e o p e n i n g s are so small that
the DNAPL c a n n o t o v e r c o m e the capillary forces binding the water in the pores. A layer
of DNAPL then accumulates on the surface of the aquitard. T h i s z o n e also has irreducible
water present. Fluid p u m p e d from this z o n e is exclusively DNAPL. Above the layer of
DNAPL and irreducible water, there is a z o n e with DNAPL and water c o n t e n t a b o v e the
irreducible saturation. Fluid p u m p e d from this level includes b o t h DNAPL and water.

234

Chapter F i v e

From the top of this z o n e to the water table, the pores contain residual DNAPL and
water. Fluid p u m p e d from this z o n e is water. Since many DNAPLs are slightly to moderately
soluble in water, water p u m p e d from any of these z o n e s can also contain dissolved
organics. Figure 5.23 shows the DNAPL zones described previously.
Monitoring wells to detect DNAPLs should be placed at the bottom of the aquifer,
just at the top of the confining layer. DNAPL from the z o n e of m o b i l e DNAPL and
irreducible water flows to the monitoring well, as will both water and DNAPL from the
z o n e w h e r e both are mobile. T h e water and DNAPL from this z o n e will separate in the
monitoring well, with the DNAPL sinking and the water rising. Well A in Figure 5.23
shows that the total depth of the DNAPL in the monitoring well is b e l o w the top of the
z o n e w h e r e b o t h water and DNAPL are present. T h e e x a c t position depends upon the
average DNAPL saturation in this zone. If a monitoring well extends b e l o w the top of
the confining layer, a false thickness of the DNAPL will be measured, as it will fill the
monitoring well b e l o w the level of the confining layer. Monitoring well B in Figure 5.23
shows this circumstance.
T h e relative thickness of the various z o n e s described here depends upon the grainsize distribution, which is reflected in the permeability of the saturated zone. A low
permeability aquifer (small p o r e s ) will have a thin layer of DNAPL collect on the bottom,

Percent DNAPL saturation


Well
B

Well
A

100 80

60

40

20

t
-ao a5
a
>

__

Capillary fringe I

I
Water table
Zone of mobile water
and residual DNAPL

j
5
3

Zone of mobile water


and DNAPL

ci

Zone of mobile DNAPL


and irreducible water

20 40 60 80 100

Pereenl water saturation

F I G U R E 5.23

Zones of a DNAPL and the relationship of mobile DNAPL and nonmobile DNAPL to the

DNAPL saturation; relationship of mobile DNAPL thickness to thickness of DNAPL is measured in a monitoring
well.

Multiphase Flow

235

while a m o r e p e r m e a b l e aquifer (large p o r e s ) will have a thicker z o n e of m o b i l e NAPL


on the b o t t o m and a thinner z o n e w h e r e b o t h DNAPL and water are m o b i l e . Figure 5.24
illustrates this p h e n o m e n o n (Villaume 1 9 8 5 ) .
If the DNAPL is not spilled in sufficient a m o u n t s to o v e r c o m e the residual saturation
in the vadose zone, ground-water contamination by the dissolved p h a s e can still occur.
T h e residual DNAPL forms a s o u r c e that, although not mobile, can slowly partition into
both the vapor p h a s e and the a q u e o u s p h a s e of the water infiltrating through the vadose
z o n e . If the DNAFL is volatile, as many are, the vapor can diffuse through the vadose
z o n e as well. T h e vapor phase can partition into the p o r e water so that the area of soil
water contamination can spread via the vapor phase.
5.6.3

H o r i z o n t a l M o v e m e n t in the Saturated Z o n e

If the DNAPL exists in a c o n t i n u o u s phase, it will m o v e b e l o w the water table according


to the force vector

D N A P L

d e s c r i b e d in Equation 5.16. If the DNAPL is in the form o f

discontinuous stringers that are sinking in the aquifer, flowing g r o u n d water will tend
to displace the DNAPL stringers in the direction of flow. J u s t as the sinking DNAPL
stringers displace water by o v e r c o m i n g the capillary pressure holding the water in a
p o r e , the laterally moving water must o v e r c o m e the capillary pressure of the DNAPL
stringer to displace it sideways.
Villaume ( 1 9 8 5 ) indicates that the lateral pressure gradient n e e d e d to displace a
DNAPL sideways, grad P, is given by

g r a d P

Ground water

( 5

= 4 ( l A -l/r,)

"

4 3 )

N A P L saturation {%)

Ground water

100

X'100

Water saturation (%)


F I G U R E 5.24

Effect of high and low permeability (and porosity) on the distribution of mobile D N A P L

at the bottom of an aquifer; the arrows indicate level of original injection of the D N A P L . Source-. J. F.
Villaume,

Ground Water Monitoring Review 5,

Journal Publishing Co.

no. 2 ( 1 9 8 5 ) : 6 0 - 7 4 . Copyright

1 9 8 5 Water W e l l

236

Chapter

Five

where
a = interfaeial tension
L = length of the c o n t i n u o u s DNAPL phase in the direction of flow
0

r, = pore-throat radius
r = p o r e radius
p

O n c e the percolating DNAPL reaches the aquitard layer, it can begin to move
laterally, even in the a b s e n c e of a hydraulic gradient on the water table. It migrates down
the dip of the aquitard. DNAPLs can collect in low spots on the surface of an aquitard.
It is p o s s i b l e for the DNAPL to migrate downdip, even if the hydraulic gradient and
ground-water flow are in the o p p o s i t e direction (Figure 5 . 2 5 ) .
If a p o c k e t of static DNAPL collects in a low spot on the surface of an aquitard
and ground water is flowing in the aquifer a b o v e the DNAPL, the interface between the
flowing ground water and the static DNAPL will be sloping. This angle can be found
from ( H u b b e r t 1 9 5 3 )
T =

Pw

F I G U R E 5.25

dh/dl

(5.44)

PDNAPL

Distribution of a dense nonaqueous phase liquid in the vadose and saturated zone.

Multiphase Flow

237

w h e r e T is the angle of the interface and dh/dl is the s l o p e of the water table (Figure
5 . 2 6 ) . A negative value for x means that the angle has a s l o p e o p p o s i t e to the direction
of the water table.
B u c k l e y and Leverett ( 1 9 4 2 ) derived the following equation for the one-dimensional
flow of two immiscible, i n c o m p r e s s i b l e fluids:

dx

df

IG

dS

dx

dS

dS

dx

'17

(5.45)

where
q, = total volume flux
S = relative saturation of the wetting p h a s e
n = porosity
/ = a function that d e p e n d s upon the value of S and is defined as:
ASJ

(5.46)

and G is a function that d e p e n d s upon the value of S and is defined as


KJ
ft

dP

(5.47)

dS

where
k^, = permeability to wetting fluid
= permeability to nonwetting fluid
H = dynamic viscosity of wetting fluid
p = dynamic viscosity of nonwetting fluid
P = capillary pressure
c

T h e volume flux of the wetting phase, cy , is given by


w

(5-48)

FIGURE 5.26

Sloping interface between a static layer of DNAPL and flowing ground water.

238

Chapter Five

(5.49)
Buckley and Leverett ( 1 9 4 2 ) solved these equations by assuming that the force
d u e to the rate at which the fluid is added at the boundary, </,, is much greater than the
capillary pressure force, P . T h u s the s e c o n d term in Equation 5.48 can be ignored.
c

However, this may not be accurate. McWhorter and Sunada ( 1 9 9 0 ) have developed an
e x a c t integral solution to this two p h a s e flow p r o b l e m that accounts for both forces.
T h e y have solutions for either the displacement of the wetting fluid by the advance of
a nonwetting fluid or the displacement of a nonwetting fluid by an invading wetting fluid
T h e former solution can be applied to the p r o b l e m of the displacement of water in an
aquifer by the lateral flow of a DNAPL.

5.7

Monitoring for L N A P L s and D N A P L s


Special consideration must be given to the design of monitoring wells and the collection
of ground w a t e r samples to test for the p r e s e n c e of LNAPLs and DNAPLs (floaters and
sinkers). Naturally, different types of wells are used for each separate phase.
B e c a u s e LNAPLs float on the capillary layer, a monitoring well for LNAPL detection
should e x t e n d from a b o v e the capillary z o n e to b e l o w the water table. If an LNAPL is
present, it will be floating at the surface of the liquid column in the monitoring well
Prior to any purging of the well, a top loading bailer should be carefully lowered to just
below the liquid surface so that the t o p layer of liquid drains into the bailer If an LNAPL
is present, the top loading bailer should capture it. T h e LNAPL should be analyzed
qualitatively. If there is a mixture of c o m p o u n d s present, it might be necessary to
determine the proportion of each in the LNAPL. Special p r o b e s are available to measure
the thickness of LNAPLs floating in a monitoring well.
To sample a I >\API.. a m< initi iring well should be constructed with a si recti at the
very b o t t o m of the aquifer It may be helpful to have a length of solid pipe as a sump
at the b o t t o m of the screen so that if even a thin layer of m o b i l e DNAPL is present, it
can collect in the s u m p in a sufficient thickness to sample. If a sump is used, the thickness
of any measured product thickness must be reduced by the length of the sump. A bottom
loading bailer is used to collect the liquid from the b o t t o m of the sump prior to any
well purging. The bailer is slowly lowered all the way to the b o t t o m of the monitoring
well and then slowly raised. T h e collected sample should be placed into a glass jar to
s e e if there is a d e n s e layer on the bottom. In c a s e the entire bailer is filled with a clear
liquid, part of the c o n t e n t s should be placed in a jar partially filled with water to s e e if
a separate phase forms.
II the chemical analysis of an a q u e o u s water sample indicates that an organic
c o m p o u n d is present in a m o u n t s greater than its published solubility value, the c o m
pound mav be p r e s e i it . i s a in inaquei >us phase that was emulsified i n the sample collection
process.
Monitoring for DNAPLs in areas of fractured b e d r o c k geology is much m o r e difficult
than in sand aquifers. Figure 5.27 shows a buried barrel from which a DNAPL has leaked

Multiphase Flow

239

Sand
and
gravel
aquifer
Fractured
bedrock
aquifer
Water
table
Dissolved DNAPL

FIGURE 5.27

Movement of a DNAPL into a fractured bedrock aquifer that underlies a sand and gravel

aquifer.

T h e DNAPL moves vertically through the vadose z o n e and the saturated sand and gravel
aquifer that overlies the fractured b e d r o c k . S o m e of the DNAPL dissolves in the flowing
ground water and is transported by advection. W h e n the DNAPL reaches the b e d r o c k
surface it flows d o w n s l o p e , in this c a s e in a direction o p p o s i t e to the flow of ground
water. T h e DNAPL moves vertically into fractures in the b e d r o c k b e l o w the p o o l that
collects on the b e d r o c k surface. T h e DNAPL can also migrate horizontally b e t w e e n
vertical cracks. As a result, the DNAPL can spread in u n e x p e c t e d and unpredictable
directions from the leaking barrel. Many b e d r o c k monitoring wells would be n e e d e d
even to find the portions of the DNAPL plume in the bedrock; it might be impossible
to locate all of it (Mackay and Cherry 1 9 8 9 ) .

5.8

Summary
Many c a s e s of ground-water contamination involve organic liquids that are either insolub l e or only partially soluble in water. T h e s e liquids may be present b o t h above and
b e l o w the water table as separate n o n a q u e o u s p h a s e liquids (NAPLs). A NAPL that is
less d e n s e than water (LNAPL) will float on the water table or the top of the capillary
fringe. A NAPL that is m o r e d e n s e than water (DNAPL) may sink into the aquifer b e l o w
the water table.
Water and a NAPL may b o t h be present in the ground. D e p e n d i n g upon their
relative proportions, only the water or only the NAPL or b o t h may be mobile. Depending
upon the surface tension of the liquids, o n e will be a wetting fluid and o n e a nonwetting
fluid. T h e term relative permeability refers to the permeability of the soil for o n e fluid
in the p r e s e n c e of a given volumetric c o n t e n t of a s e c o n d fluid. Darcy's law can be
written in terms of the relative permeability for b o t h a nonwetting and a wetting fluid.

240

Chapter Five

If a sufficient depth of an LNAPL collects on the surface of the capillary fringe, it


can How into a shallow monitoring well. T h e thickness of LNAPL measured in a monitoring
well is greater than the thickness of the free LNAPL in the subsurface. If certain soil
parameters are known, o n e can c o m p u t e the thickness of the mobile LNAPL based on
die thickness that accumulates in a monitoring well.
A DNAPL may sink in an aquifer until it reaches a fine-grained layer. DNAPL may
accumulate in a m o b i l e layer at the bottom of the aquifer. A monitoring well screened
at the b o t t o m of the aquifer may be used to detect the p r e s e n c e of a DNAPL.

Chapter Notation
A

Cross-sectional area of flow

P o r o u s media grain diameter

dhjdl

Gradient of fluid potential for wetting fluid

dhjdl
D
EC
D
ryOV/a
a
D

Gradient of fluid potential for nonwetting fluid


Argument = (,PT/(p - p)g) - (P" - pg)
Depth in the aquifer to the top of the capillary fringe
w

Depth in the aquifer to the top of the oil table


Depth in the aquifer to the top of the z o n e where nonresidual organic
fluid occurs
Depth in a well to the air organic interface

Depth in a well to the organic water interface

F o r c e vector

F o r c e vector for water

''DNAPL

F o r c e vector for a DNAPL

^LNAPL

F o r c e vector for a LNAPL

A Buckley-Leverett function

A Buckle}-Leverett function

Acceleration of gravity

grad P

Lateral pressure gradient


Height of capillary' rise in a tube

Head ( p o t e n t i a l ) of wetting fluid

^fiw
A.

Head ( p o t e n t i a l ) o f nonwetting f l u i d
Critical height for a DNAPL stringer before it can displace water in the

saturated z o n e
Intrinsic permeability

Relative permeability
Relative permeability of water

Ka

*
K

rnw

Relative permeability of
Relative permeability of
Relative permeability of
saturation of water in a

air
a n o n a q u e o u s phase liquid
n o n a q u e o u s phase liquid at the residual
w a t e r - n o n a q u e o u s phase system

Relative permeability of the nonaqueous phase system as a function of

Multiphase

k,

4
n
P
P "
d

Flow

241

Relative permeability of the n o n a q u e o u s phase in an a i r - n o n a q u e o u s


phase system as a function of S
Vertical length of a DNAPL stringer
Porosity
Fluid pressure
B r o o k s - C o r e y air-organic displacement pressure

P
P
P,
P

B r o o k s - C o r e y organic water displacement pressure


P o r e pressure for a wetting fluid
P o r e pressure for a nonwetting fluid
Capillary pressure

P
p

T h r e s h o l d or displacement pressure
Pressure in an LNAPL at an LNAPL-water interface in a monitoring well

Reference pressure

P,
P
Q
0,,

Capillary pressure in a thin tube


Pressure in water at an LNAPL-water interface in a monitoring well
V o l u m e of wetting fluid flowing in a two-phase system
Vol ume of nonwetting fluid flowing in a two-phase system

<j
q

Total volume flux


V o l u m e flux of wetting fluid

;/,
r

V o l u m e flux of nonwetting fluid


Radius of a thin tube

r
r

Radius of curvature for an air-water interface


P o r e radius
P o r e throat radius
Saturation ratio for air
Effective saturation
Organic liquid saturation ratio

S
S
S

S
S
S
T
V
w

wi

Wetting-fluid or water saturation ratio


Irreducible wetting fluid saturation
Residual nonwetting-fluid saturation
T h i c k n e s s of oil in a monitoring well
V o l u m e of organic liquid per unit area

Depth from the water table to an organic fluid-water interface in a


well

z
z

Vertical c o o r d i n a t e measured positively downward


Reference elevation

6
a, j
y

Interfacial angle b e t w e e n two nonwetting fluids


Interfacial tension b e t w e e n s u b s t a n c e s ('and j
Specific weight

/
p
p

B r o o k s - C o r e y pore-size distribution index


Density
Density of water

p
p.

Density of an organic fluid


Density of a nonwetting fluid

PDNAPL

Density o f a DNAPL

242

Chapter F i v e

Density of a LNAPL

Plnapl

Dynamic viscosity of water


a>

Fluid potential

*.

Fluid potential for nonwetting fluid

Fluid potential for wetting fluid

References
Abdul. Abdul S. 1988. Migration of petroleum products
through sandy hydrogeologic systems. Ground Water
Monitoring Review 8, no. 4 : 7 3 - 8 1 .
Abdul, Abdul S., Sheila F. Kau, and Thomas L Gibson. 1989.
Limitations of monitoring wells for the detection and
quantification of petroleum products in soils and
aquifers. Ground Water Monitoring Review 9, no.
2:90-99.
Abriola, Linda M., and George F. Finder. 1985a. A multiphase
approach to the modeling of porous media contamination by organic compounds: 1. Equation
development. Water Resources Research 2 1 , no.
1:11-18.
. 1 9 8 5 b . A multiphase approach to the modeling of
porous media contamination by organic compounds:
2. Numerical simulation. Water Resources Research
2 1 , no. 1 : 1 9 - 2 6 .
Baehr, Arthur L 1987. Selective transport of hydrocarbons
in the unsaturated zone due to aqueous and vapor
phase partitioning. Water Resources Research 23, no.
10:1926-38.
Baehr, Arthur L, and M. Yavuz Corapcioglu. 1987. A compositional multiphase model for groundwater contamination by petroleum products: 2. Numerical solution.
Water Resources Research 23, no. 1 : 2 0 1 - 1 3 .
Bear, J a c o b . 1972. Dynamics of fluids in porous media.
New York: American Elsevier Publishing Company,
7 6 4 pp.
Berg, R, R 1975. Capillar)' pressures in stratigraphie traps.
Bulletin. American Association of Petroleum Geolo
gists 59, no. 6 : 9 3 5 - 5 6 .
Brooks, R H., and A T. Core) . 1966. Properties of porous
media affecting fluid flow. Proceedings, American So
ciety of Civil Engineers. Irrigation and Drainage Divisional, no. I R 2 : 6 l - 8 7 .
Buckle)', S. E., and M. C. Leverett. 1942. Mechanism of fluid
displacement in sand. Transactions. American Institute of Mining, Metallurgical, and Petroleum Engineering 1 4 6 : 1 0 7 - 1 6 .
Corapcioglu, Yavuz M., and Arthur L Baehr. 1987. A compositional multiphase model for groundwater contamination by petroleum products: 1. Theoretical
considerations. Water Resources Research 23, no.
1:191-200.
Eckberg, David K., and Daniel K. Sunada. 1984. Nonsteady
three-phase immiscible fluid distribution in porous media. Water Resources Research 20, no.
12:1891 - 9 7 .
Farr, A M., R J. Houghtalen, and D. B. McWhorter. 1990.
-

Volume estimation of light nonaqueous phase liquids


in porous media. Ground Water 28, no. 1 : 4 8 - 5 6 .
Faust, Charles R 1985. Transport ot immiscible fluids within
and below the unsaturated zone: a numerical model.
Water Resources ResearcJ) 21, no. 4 : 5 8 7 - 9 6 .
Hall. Robert A, Steven B. Blake, and Stephen C. Champlin,
Jr. 1984. Determination of hydrocarbon thicknesses
in sediments using borehole data. Proceedings of the
Fourth National Symposium and Exposition on Aquifer Restoration and Ground Water Monitoring Na
tional Water Well Association, pp. 3 0 0 - 1 0 .
Hochmuth, D. P., and David K. Sunada. 1985. Groundwater
model of two phase immiscible flow in coarse material. Ground Water 23, no. 5 : 6 1 7 - 2 6 .
Hubbert, M. King. 1953. Entrapment of petroleum under
hydrodynamic conditions. American Association of
Petroleum Geologists Bulletin 37, no. 8:1954 - 2 0 2 6 .
Lenhard, R J . , and J. C. Parker. 1987a. A model for hysteretic
constructive relations governing multiphase flow: 2.
Permeability saturation relations. Water Resources Re
searcJj25, no. 1 2 : 2 1 9 7 - 2 2 0 6 .
. 1987b. Measurement and prediction of saturation
pressure relationships in three-phase porous media
systems.
Journal
of Contaminant
Hydrogeology
1:407-24.
. 1988. Experimental validation of the theory of extending two-phase saturation-pressure relations to
three-fluid phase systems for monotonic drainage
paths. Watet Resources Research 24, no. 3 : 3 7 3 - 8 0 .
. 1990. Estimation of free hydrocarbon volume from
fluid levels in monitoring wells. Ground Water 28,
no. 1 : 5 7 - 6 7 .
Lohman, S. W. 1972. Ground water hydraulics U.S. Geological Survey Professional Paper 7 0 8 : 7 0 pp.
Lin, C. George F. Pinder, and E. F. Wood. 1982- Water resources program report 83 WR-2. Water Resources
Program. Princeton. N. J.: Princeton University.
Mackay, D. M,, and J. A Cherry. 1989. Groundwater contamination: pump and treat remediation. Environmental
Science and Engineering 23. no. 6 : 6 3 0 - 3 7 .
McWhorter, David B, and Daniel K. Sunada. 1990. Exact integral solutions for two-phase flow. Water Resources
Research26, no. 3 : 3 9 9 - 4 1 3 .
Parker, J. C, and R J. Lenhard. 1987. A model for hysteretic
constructive relations governing multiphase flow: 2.
Saturation pressure relations. Water Resources Re
search 23, no. 1 2 : 2 1 8 7 - 9 6 .
Pinder. George F . and UIKI.I M. Abriola. 19H6. On tlx- sim
illation of nonaqueous phase organic compounds in

Multiphase Flow

the subsurface. Water Resources Research 22, no.


9:109S-119S.
Schwille, F. 1981. "Groundwater pollution in porous media
by fluids immiscible with water." The science of the
total environment 2 1 : 1 7 3 - 8 5 .
. 1984. "Migration of organic fluids immiscible with
water in the unsaturated zone." In Pollutants in porous media, edited by B. Yaron, G. Dagen and J.
Goldshmid. 2"?-48. Berlin: Springer Verlag.
Schwille, F. 1988. Dense chlorinated solvents in porous and
fractured media. Translated by J a m e s F. Pankow.
Chelsea. Mich.: Lewis Publishers, 146 pp.
Stone. H. L 1973. Estimation of three phase relative perme-

243

ability and residual oil data. Journal of Canadian


Petroleum Technology 1 2 , no. 4 : 5 3 - 6 1 .
Testa, Stephen M.. and Michael T. Paczkowski. 1989. Volume
determination and recovery of free hydrocarbon.
Ground Water Monitoring Review 9. no. 1 : 1 2 0 - 2 7 .
Van Genuchten, M. Th. 1980. A closed-form equation for
predicting the hydraulic conductivity of unsaturated
soils. Soil Science Society of America Journal44:892 98.
Vdlaume, J a m e s F. 1985. Investigations at sites contaminated
with dense nonaqueous phase liquids (NAPLS).
Ground Water Monitoring Review 5, no. 2 : 6 0 - 7 4 .

Chapter S i x

Inorganic Chemicals in Ground Water

6.1

Introduction
G r o u n d water is a solvent that is in contact with various earth materials. As a result,
ground water naturally contains dissolved cations and anions as well as s o m e nonionic
inorganic material, such as silica ( S i 0 ) . Naturally occurring ground water can contain
2

dissolved solids that range in concentration from less than 1 0 0 m g / L to m o r e than


5 0 0 , 0 0 0 m g / L ( H e m 1 9 8 5 ) . T h e major ion constituents of natural water include calcium,
magnesium, sodium, potassium, chloride, sulfate, and b i c a r b o n a t e / c a r b o n a t e . Dissolved
gasses can include nitrogen, c a r b o n dioxide, methane, oxygen, and hydrogen sulfide.
T h e r e are a n u m b e r of ions that can be naturally present in small amounts that can
affect the water quality. In addition, inorganic ions that impact upon water quality can
be released to the subsurface via human activity.
We have already seen in Chapter 3 that ions can be removed from solution by ion
e x c h a n g e and sorption. In this chapter we will e x a m i n e other chemical p r o c e s s e s that
act to remove inorganic ions from solution. We will also e x a m i n e the chemical properties
of a n u m b e r of inorganic materials frequently found in ground water. T h e geochemical
zonation that can o c c u r near landfills that have received municipal waste will be used
to illustrate s o m e basic principles.

6.2

U n i t s of Measurement and Concentration


Chemical analyses are usually reported on the basis of weight of solute per volume of
solvent. C o m m o n units are m i l l i g r a m s p e r l i t e r ( m g / L ) and m i c r o g r a m s p e r l i t e r
( / t g / L ) . Equivalent weight units are frequently used when the chemical behavior of a
solute is being considered. T h e e q u i v a l e n t w e i g h t of an ion is the formula weight
divided by the electrical charge. If the concentration of the ion in milligrams per liter is
divided by the formula weight, the resulting concentration is e x p r e s s e d in terms of
m i l l i e q u i v a l e n t s p e r l i t e r . O n e m o l e of a substance is its formula weight in grams.
A 1 m o l a l solution has 1 m o l e of solute in 1 0 0 0 g of solvent. A 1 m o l a r (1 . 1 / ) solution
has 1 m o l e of solute in a liter of solvent.

244

Inorganic Chemicals in G r o u n d W a t e r

245

If a solution is dilute and there is no n e e d to m a k e density c o r r e c t i o n s , the molality


can be determined from the c o n c e n t r a t i o n by the following equation:
Molality =

6.3

milligrams per liter x 10 ~


,,
formula weight in grams

(6.1)

Chemical Equilibrium and the L a w of M a s s Action


T h e law of mass action states that the rate of a chemical reaction will be proportional
to the active m a s s e s of the participating s u b s t a n c e s ( H e m 1 9 8 5 ) . If there are two s u b
stances, A and B, reacting to form two o t h e r substances, C and D, and if the p r o c e s s is
reversible, then the reaction can be written as

(6.2)

aA + bB ; = ! cC + dD

T h e rate of the forward reaction, / ? , , is


/?, = * ; [ A ] " [ B ] '

(6.3)

w h e r e a s the rate of the reverse reaction, R , is


2

R = *[C]'[D]'

(6.4)

where:
[A] = active c o n c e n t r a t i o n of s u b s t a n c e A
k[ = proportionality constant for the forward reaction
k' = proportionality constant for the reverse reaction
2

If the reaction p r o g r e s s e s to a point w h e r e the forward reaction rate is equal to


the reverse reaction rate, then
ti[A!'[B]'^4i[C]'[D]

(6.5)

Equation 6.5 c a n be rearranged to yield the following e x p r e s s i o n :


[C]'[D]'

k\

[A]"[B]

k'

(6.6)

w h e r e AT is the e q u i l i b r i u m c o n s t a n t .
eq

If two or m o r e ions react to form a solid precipitate and the reaction is reversible,
then it can be represented as
aA + m == CAB

(6.7)

T h e equilibrium relationship of this reaction is:

K
K

[ A l

'

"-1W

t B l

<(,K
(6

8)

246

Chapter S i x

w h e r e K is called a s o l u b i l i t y p r o d u c t . T h e activity of the solid together with the


sp

water is defined as unity. Solubility products can be used to c o m p u t e the concentration


of a solute in equilibrium with a solid phase, either via dissolution of the solid into an
undersaturated solution or following precipitation of the solid from a saturated solution.
If o n e is dealing with a very dilute aqueous solution, then molal concentrations
can be used to determine chemical equilibrium. However, for the general case, o n e must
use c h e m i c a l a c t i v i t i e s to employ the law of mass action.
T h e chemical activity of ion X, [X], is equal to the molal concentration of X, m
times a factor known as an a c t i v i t y c o e f f i c i e n t , y :
x

(6.9)

IX] = m,y

T h e activity coefficient varies with the total amount of cations and anions in solution.
T h e concentration and charge of the various ions in a solution determine its i o n i c
s t r e n g t h . Ionic strength can be c o m p u t e d from the following formula:
m z

I=2-Y. i ?

(6.10)

where
/ = ionic strength
= molality of the ith ion
z, = charge of the /th ion
O n c e the ionic strength is determined, the activity coefficient can be calculated
using the D e b y e Huckel equation:
(6.11)

-log 7,=
where
y, = the activity coefficient for ionic species i
Zj = the charge on ionic species i
I = ionic strength of the solution
A = constant equal to 0 . 5 0 8 5 at 2 5 C
B = constant equal to 0.3281 at 25C
, = the effective diameter of the ion from T a b l e 6.1

T h e Debye-Hiickel equation can be used with solutions that have an ionic strength
of 0.1 or less (approximately 5 0 0 0 m g / L ) . Figure 6.1 is a graph showing the relationship
of activity coefficient to ionic strength for specific ions; it was calculated using the Debye
Huckel equation. Specific curves are for ions with the same effective diameter and charge
as listed in T a b l e 6.1. Not all the ions to which a curve applies are listed on the figure.
F o r e x a m p l e , the curve labeled C a
2+

2+

M n , N i , and C o

2+

2 +

and F e

2 +

2 +

2 +

2 +

can also b e used for C u , Z n , S n ,

, b e c a u s e all these ions have the s a m e effective diameter and

charge.
Chemical equilibrium is a useful c o n c e p t in studies of contaminant hydrogeology.
Ionic contaminants discharged into ground water may react with naturally occurring ions

Inorganic Chemicals in G r o u n d W a t e r

T A B L E 6.1

247

Values of the parameter o, in the Debye-Huckel equation.

a,

Ion
4 +

Th

AI

Mg

Co

Fe(CN)

PO ~, F e ( C N ) " , H g

O H " , F " , CNS", CNO", H S " , CI0 ", K , CI", Br", I " , CN", N 0 " , N 0 " , Rb ,

3 +

Sn'

11

3 +

,Fe
2 +

2 +

, Cr *, H

, Be

, Cu \ Zn
4 _
6

2 +

2 +

, Sn

, Sr *, Ba

2 +

2 +

2 _
3

, Mn

, Cd

2 +

, S0

2 +

, Fe

, Hg
2
4

2 +

2 +

, Ni
2

, S

" , Se0

2
4

2 +

, Co

, Pb

2 +

2 +

~, Cr0

3
4

, Li*

, C0

2
3

", S 0

~, H P 0

2
4

2 _
3

, Mo0

2 _
4

" , Na , HC0 ~, H;P0 ~


3

Cs

, NH

+
4

, Ag

Source: Reprinted with permission from J. Kielland, "Individual Activity Coefficients of Ions in Aqueous Solutions," American Chemical
Society Journal 59 (1937): 1 6 7 6 - 7 8 . Published 1937 by the Americon Chemical Society.

Activity coefficient (y)


F I G U R E 6.1

Relationship of activity coefficients of dissolved ions as a function of the ionic strength of

a solution at 25 C. Source: J. D. Hem, "Study and interpretation of the chemical characteristics of natural
waters," Water Suppiy Paper 2 2 5 4 , U . S . Geological Survey, 1985.

248

Chapter S i x
T A B L E 6.2

Solubility products for selected minerals and compounds.

Compound

S o l u b i l i t y Product

Minerai Nome

Chlorides
CuCI

10"

PbCI,

IO"'

Hg CI
2

4 7

IO"'"

AgCI

10""

Fluorides
BaF

SrF

to-""

MgF
PbF

10-"

CaF

8 2

10"
10""
10""

Fluorite
Sellaite

Sulfates
BaSO

) 0

-m.o

PbSO.

10""
10""
10"

Ag SO,

IO"

SrSO,

IO""

CaSO,

CaSGy2 H 0
2

7 8

Borite
Anhydrite
Gypsum
Anglesite

4 8

Celestite

Sulfides
Cu S

io- "

CuS

IO

FeS

IO""-'

PbS
HgS

IO" "
, Q-533

ZnS

IO" "

ZnS

"

Golena

Wurtzite

IO" "

Sphalerite

10""
, 0-137

Witherite

Cinnebar

Carbonates
BaCOj
CdCOj

- 8 3 5

CaCOj

IO

CaCOj

IO"

C0CO3

IO"
, -107

FeC0

8 2 2

Aragonite

1 0 0

Siderite

,0-131

PbCOj

MgC0

Calcite

10-"

Magnesite

MnCOj

IO""

Rhodochrosite

IO"'

Variscite

Phosphates

AIPCV2 H 0
2

CaHPO,'2 H 0
2

Ca (POJ

Cu,(POJ

10""
IO"

2 8 7

3 6

IO" *
, -2l

FePO,

FePGy2 H 0
2

10 -

2 6 4

Source: K. 8. Krouskopf, Introduction to Geochemistry, 2d ed. (New York: McGrow-Hill, 1979)

249

Inorganic Chemicals in G r o u n d W a t e r

in the ground water to form a precipitate or they may mobilize ions s o r b e d on solid
surfaces. T h e y may also u n d e r g o oxidation or reduction. B o t h these p r o c e s s e s are
reversible and can be d e s c r i b e d by chemical equilibrium. Many g e o c h e m i c a l p r o c e s s e s
in ground water are not readily reversible, such as weathering of silicate minerals. T h e s e
reactions must be treated using kinetics. However, as this type of reaction is n o t of
significant interest in contaminant hydrogeology, we do not c o n s i d e r kinetic models.
T a b l e 6.2 contains the solubility products for a large n u m b e r of minerals, including
many that can be formed from trace metals that can be ground-water contaminants.

6.4

Oxidation-Reduction Reactions
In s o m e chemical reactions the participating elements c h a n g e their valence state through
the gain or loss of e l e c t r o n ( s ) . If an electron is gained, there is a loss of positive valence
called a reduction. A loss of negative valence is called an oxidation. T o g e t h e r , these are
referred t o a s o x i d a t i o n - r e d u c t i o n , o r r e d o x , reactions. I n environmental systems
they may be controlled by m i c r o o r g a n i s m s that do not participate in the reaction but
act as catalysts. T h e m i c r o b e s o c c u r as a biofilm on the surfaces of the aquifer materials.
T h e y obtain energy b y oxidation o f organic c o m p o u n d s o r hydrogen o r r e d u c e d inorganic
forms of iron, nitrogen, and sulfur. Electron a c c e p t o r s are necessary for these biologically
mediated r e d o x reactions to o c c u r . U n d e r a e r o b i c c o n d i t i o n s o x y g e n is the electron
acceptor, w h e r e a s u n d e r a n a e r o b i c conditions nitrate, sulfate, and c a r b o n d i o x i d e are
the e l e c t r o n a c c e p t o r s (McCarty, Rittman, and B o u w e r 1 9 8 4 ) .
An e x a m p l e of a reduction is
Fe

2 +

4 - 2 e " ^ = Fe

(6.12)

In this e x a m p l e , ferrous iron is r e d u c e d to metallic iron by the addition of two


electrons. T h i s is a half-reaction, since the e l e c t r o n s must be supplied either by an
electrical current or by a simultaneous reaction in which a n o t h e r e l e m e n t is o x i d i z e d
and releases the requisite n u m b e r of electrons. T h e standard electrical potential of a
half-reaction is the voltage represented by the flow of e l e c t r o n s w h e n the reaction is at
equilibrium. Under standard c o n d i t i o n s ( 2 5 C and 1 atm p r e s s u r e ) the standard potential
is represented by the s y m b o l E. T h e potential is in volts, with a negative value representing reducing conditions and a positive value representing oxidizing c o n d i t i o n s ( H e m
1 9 8 5 ) . B y convention, the standard potential for the reduction o f H

t o hydrogen gas

is 0:
2 H

+ 2 e " ^ H

(gas)

(6.13)

An e x a m p l e of oxidation o c c u r s w h e r e ferrous iron l o s e s an electron to form ferric iron:


Fe

2 +

^ Fe

3 +

+ e"

(6.14)

Oxidation-reduction reactions involve elements that can o c c u r in m o r e than o n e


valence state. In Equations 6.12 and 6.14, iron o c c u r r e d in the metallic ( 0 ) as well as
the 4- 2 and 4- 3 states. Metals can usually o c c u r in the metallic state with a z e r o valence
and at least o n e o t h e r valence state. S o m e elements that are environmentally important

250

Chapter S i x

can o c c u r in several different valence states. T a b l e 6.3 lists several elements that o c c u r
in different valence states and e x a m p l e s of c o m p o u n d s and ions formed from those
elements.
In order for oxidation or reduction to o c c u r in a chemical reaction, o n e element
must be reduced while a s e c o n d element is being oxidized. F o r example, the c o m p l e t e
equation for the oxidation of ferrous iron to ferric iron is
4Fe

2 +

+ 0

+ 4H

3 +

;=t 2 H 0 + 4 Fe

(6.15)

This c o m p l e t e reaction is c o m p o s e d of two half-reactions:


4Fe

2 +

^=^4Fe

Oj + 4 H * + 4 e " ^ 2 H

3 +

+4e"

[oxidation]

(reduction]

(6.16)
(6.17)

An aqueous solution has an oxidation potential indicated by the symbol Eh. This
can b e calculated from the N e r n s t e q u a t i o n :

KT, [products]
In -r
7
reactants

Eh = E

(6.18)

nF

T A B t E 6.3

Selected elements that can exist in more than one

oxidation state.
Element
Carbon

+ 4
0

Chromium
Copper
Mercury
Iron
Nitrogen

COj ~

-4

CH,
CrCV". C r 0 ~

+ 3

C r * , Cr(OH)

+ 1

CuCI

+ 2

CuS

+ 1

Hg,CI

+ 2

HgS

+ 2

Fe

Fe

4-5

N0 "

+ 3

NO,"

4-3

- 3

Sulfur

HC0 ~,

4-6

0
Oxygen

Examples

Valence State

, FeS

3 +

, FelOHIj

N
N H / , NHj
O

- 1

H 0

- 2

H 0, O "

- 2

H S , S ~ , PbS

-t-2

s cVso -

4-5
+ 6

2
2

251

Inorganic Chemicals in G r o u n d W a t e r

where:
Eh = oxidation potential of the a q u e o u s solution in volts
E = standard potential of r e d o x reaction in volts
R = gas constant, 0 . 0 0 1 9 9 K c a l / ( m o l e K )
T = temperature in Kelvins
F = Faraday constant, 2 3 . 0 6 K c a l / V
n = n u m b e r of electrons in half-reaction
1

] = activity of products and reactants

T h e standard potential for a reaction c a n be determined from the relationship


o

-AGS

(6.19)

nF

w h e r e AGS in volts is the f r e e e n e r g y , or G i b b s f r e e e n e r g y , of the reaction.


T h e free energy of a reaction is the sum of the free energies of the products minus
the sum of the free energies of the reactants. F o r the reaction
aA + bB

cC + rfD

the free energy can be found from:


AGS = c A G + dAG - aAG - bAG
c

(6.20)

Values of free energy for many elements, ions, and c o m p o u n d s are found in
standard reference works. T a b l e 6.4 contains values for a n u m b e r of species.
T h e equilibrium constant for a reaction is related to the free energy of the reaction
by
0

A G , = -RT\t\K

(6.21)

eq

At standard temperature and pressure and with AGS ' kilocalories, Equation 6.21
n

c a n be rewritten as
AG
l 0 8 / f

-=T ^ f

( 6

"

2 2 )

T h e oxidation potential of an a q u e o u s solution can be measured using a specific


ion electrode. If the value is positive, the solution is oxidizing, and if it is negative, the
solution is reducing. Oxidation potential is measured in volts relative to the hydrogen
electrode, which is at zero. Commercially available Eh meters are available that can be
attached to a ground-water sampling pump. T h e ground-water sample is p u m p e d under
positive pressure into the flowthrough cell w h e r e the e l e c t r o d e is located. T h e water
sample is never subjected to a vacuum, which could c a u s e degassing. Moreover, it is
not e x p o s e d to the atmosphere, w h e r e it can c o m e into c o n t a c t with a t m o s p h e r i c oxygen.
T h i s has simplified the accurate and precise m e a s u r e m e n t of Eh in g r o u n d water.

T A B L E 6.4

Standard Gibbs free energy of formation for selected species.

Species

\ G kcal/mole

Arsenic

Species

5 kcal/mole

Manganese

Species
U0 (c)

(uranite)

As 0 (c)

-187.0

Mn0 (c) (pyrolusite)

-111.18

As,0 (c)

-275.46

Mn 0

(bixbyite)

-210.6

U0

As S (c)

-40.3

Mn 0

(hausmannite)

-306.7

(U0 ) (OH)
2

M n ( O H ) (c) amorphous

-147.0

(U0 ) (OH)

-183.1

MnC0 (c) (rhodochrosite)

-195.2

(U0 ) (OH) -

-180.04

Mn

FeAsO(c)

H AsCUaq)
3

H AsO (aq)
2

HAs0
As0

(aq)

3 _

185.13

-155.0

HAsOj(aq)

170.82

(aq)

(aq)
+

M n O H (aq)

Mo0 (c)

As0 -(aq)

-83.66

Cr 0 (c)
2

HCr0

Cr,0 -(aq)
7

-367.07

U0 (COj) '-(aq)

-635.69

Miscellaneous species
-254.2
-114.7

-199.9

CuFe 0 (c)

-205.26

NiFe 0 (c)

-232.6"

Ag 0(c)

-2.68"

AgCI(c)

-26.24

Ag S(c)

-9.72

Ag C0 (c)
2

-311.0'

-104.4

HjO(l)

- 56.687

OhT(aq)

- 37.594

0 (aq)

Ag*(aq)

18.43

-3.9

HSO~(aq)
2

-434.5

-503.2

ZnFe 0 (c)

(brochantite)

-1037.5

CuFe0 (c)

CuSO,-3Cu(OH) (c)

-560.99
-945.16

-127.40

Silver

Mo0 "(aq)

31.0

(aq)

-182.8'

CuO(c)

(aq)

-252.9'

Copper

-233

-173.96'

U0 C0 (c)
U0 (C0 ) -(aq)
2

-96.8

-227.68

-54.5

(aq)

-246.61
-229.69

FeMoO.(c)

Cr0 '(aq)

Mo0 (c)
2

Chromium

-159.66

(aq)

Molybdenum

-96.25

U0

AG kcal/mole

S0 ~(aq)
4

AgOH(aq)

-22.0

H S(aq)

-180.69
-177.97
-6.66

Cu 0(c)
2

-34.9

Ag(OH) -(aq)

-62.2

HS~(aq)

Cu S(c)

-20.6

AgCI(aq)

-17.4

S -(aq)

20.5

-51.5

C0 (g)

- 94.254

Cu

(aq)

CuS0 (aq)
4

15.67"

AgCI (aq)

-165.45

Vanadium

HCuCV(aq)
2

Cu0 "(aq)
2

-61.8

H.VO<

-43.9

H VO
3

Cu

(aq)

11.95

H V0
2

Iron
Fe(OH) (c)

ppt.

FeC0 (c)

(siderite)

-166.0

(hematite)

Fe

(aq)

CH (aq)
4

-8.22

HPO, ~(aq)

-260.34

-57.8'

H P0 (aq)

-270.14'

H P0 (c)

-273.10

No * (aq)

-62.59'

-54.83
-106.7'

U *(aq)
UOH

FeO(c)

-60.03

Fe S(c) (pyrite)

-38.3

FeS(c)

-24.22

(aq)

Cr(aq)
3

Uranium

-18.85
147.0

-212.9'

P0 "(aq)
4

-62.58'
198.4'

oq = aqueous solution

-12.13

- 1 55.65'

Fe(OH) (aq)

c = solid

-31.37

CH (g)

-163.2'

Criaq)

VOOH*

(aq)

-106.7'

140.26

V(OH) *

-177.4

Fe(OH) "(aq)
4

-126.17

-111.41'

(aq)

VOH *(aq)

(aq)
+

HCCV(aq)
COj -(aq)

- 1 48.94

V(OH) (aq)

FeOH ' (aq)

252

-214.9'

-39.9
-1.1

2 +

Fe(OH)
+

-233.0'

-159.35

(aq)

FeOH
2

V0 -(aq)

H C0 (aq)

V0

-253.67'
-249.2'

-92.26

-244'

_
4

C0 (aq)

-116.3'

FeS (c) (pyrite)


3

0
4

ppt.

Fe

HV0 ~(aq)

Fe(OH) (c)

Fe 0

2.88

(aq)

U(OH)(c)

-0.00
-31.38
-243.5

-126.44'

(oq)

-67.51

-182.24

NH *(aq)

-18.99

-347.18

Pb *(aq)

-5.83

Inorganic Chemicals in G r o u n d W a t e r

6.5

R e l a t i o n s h i p between pH and Eh
6.5.1

pH

W a t e r u n d e r g o e s a dissociation into two ionic species:

0 ; = H

+ O H "

(6.23)

T h e equilibrium c o n s t a n t for this reaction is


+

[H ] [ O H - ]

(6.24)

[H 0]
2

T h e value of this equilibrium constant d e p e n d s u p o n the temperature, but at 2 5 C


it is 1 x 1 0 ~ . W a t e r that is neutral has the s a m e n u m b e r of H and O H " ions. If
there are m o r e H ions, water is acidic, and if there are m o r e O H ~ ions, it is basic.
1 4

T h e p H o f a n a q u e o u s solution i s a m e a s u r e o f the n u m b e r o f hydrogen ions o r


p r o t o n s present. T h e definition of pH is the negative logarithm of the hydrogen-ion
activity. It ranges from 0 ( m o s t a c i d i c ) to 14 ( m o s t b a s i c ) , and at 2 5 C a pH of 7 m e a n s
that the solution is neutral. B e c a u s e [ H O ] is unity, from Equation 6.24 we have the
relationship [H J [ O H " ] = K = 1 0 "
T h e pH of a solution is m e a s u r e d with a pH
m e t e r and an electrode. It should be m e a s u r e d in the field, preferably in a flowthrough
cell so that dissolved gas isn't e x c h a n g e d with the a t m o s p h e r e prior to the measurement.
T h e pH of a solution is especially sensitive to the a m o u n t of dissolved C 0 .
z

1 4

eq

6.5.2

Relationship of Eh and pH

We thus have two ways to characterize a solution. T h e pH d e s c r i b e s the n u m b e r of


p r o t o n s present and the Eh is related to the n u m b e r of electrons. Eh and pH can be
related through the Nernst equation for a reaction that contains water and H

ions.

Such a reaction can be written ( R o b e r t s o n 1 9 7 5 )

bB + m H

+ ne~

ah + wU 0
2

(6.25)

Sources: for Toble 6.4 (opposite)


0

Wogemon, D. D., W. H. Evans, V. B. Parker, I. Halow, S. AA. Baily, and R. H. Schumm. 1968. Selected values of chemical thermodynamic properties. Notional

Wogemon, D. D., W. H. Evans, V. B. Paker, I. Halow, S. M. Baily, and R. H. Schumm. 1969. Selected volues of chemical thermodynamic properties. National

Bureou of Standards Technical Note 270-3; 264 pp.


Bureou of Standords Technical Note 270-4; 14) pp.
CODATA Task Group on Key Values for Thermodynamics. 1976. Recommended key values for thermodynamics 1975. Journal of Chemical Thermodynamics
8:603-5.
0

C O D A T A Task Group on Key Values for Themodynamics. 1977. Recommended key values for thermodynamics 1976. Journal of Chemical Thermodynamics
9:705-6.

* Giridhor J., and Donald Longmuir.


+4 H

= U *

+ Cu

I +

1 9 9 1 . Determination of E" for the U 0 2 * / U '

couple from measurement of the equilibrium: UO?'* + Cu(s)

+ 2 H 0 ot 25" C and some geochemicol implications. Rodiochemico Acta 54:13338.


2

' Hondbook of Chemistry ond Physics. Selected Values of Chemicol Thermodynamic Properties. Boco Roton, Flo.: CRC Press.
9

Robie, R. A.. B. S. Hemingway, and J. R. Fisher. 1978. Thermodynamic properties of minerals and related substances ot 298.15 K ond 1 bar (105 pascals)

Feitknecht, Walter, ond P. W. Schindler. 1963. Solubility constonts of metol oxides, metal hydroxides and metal salts in oqueous solution. Pure ond Applied

pressure ond higher temperatures. U. S. Geological Survey Bulletin 1452; 456 pp.
Chemistry 6 : 1 3 0 - 5 7 .
' Baes, C. F., Jr., and R. E. Messmer. 1976. The Hydrolysis of Cations. New York: Wiley, 489 pp.
' Wageman, D. D., W, H, Evans, V. B. Porker, I. Halow, S. fv\. Baily, and R. H. Schumm. 1968. Selected values of chemicol thermodynamic properties. Notional
Bureou of Standards Technical Note 270-5.
1

Longmuir, Donold. 1977. Uranium solution minerol equilibria of low temperatures. Geochimico et Cosmochimico Acta 42:54769.

254

Chapter S i x

where
A = reactant
B = product
n = n u m b e r of electrons released
a = moles of reactant
w = m o l e s of water
b = moles of product
m = m o l e s of hydrogen ions
T h e Nernst equation for Reaction 6.25 is
RT

E h = /

-^

[A][H Or
2

( 6

lijWr

2 6 )

T h e activity of water is unity. F o r a particular reaction, is given and R T, and


F a r e constants. T h e significant variables are the Eh and the activities of the reactant, the
product, and the hydrogen-ion activity, which can be e x p r e s s e d as a pH. Equation 6.26
can be rearranged and e x p r e s s e d in base 10 logs as either

Eh

3 0 3

= - -

1 O 8

( 6

iir^r

2 7 )

or
Eh = E - 2.303 log
nt

+ 2.303 log[H ]
nF

|BJ

(6.28)

At 2 5 C and 1 atm of pressure, Equation 6.28 can be e x p r e s s e d as

o _2^92

E h =

n
6.5.3

W_
IB

Q 0 5 9 2

Eh-pH Diagrams

The Eh pH relationship is particularly useful when applied in the form of an Eh-pH


diagram, with Eh the ordinate and pH the abscissa. If a solution has several ions present
that can react to form different products or o c c u r in different valence states, the stable
product or valence state at a given concentration of reactants will be a function of the
pH and Eh of the solution.
Figure 6.2 is a basic Eh-pH diagram. T h e range of pH is 0 to 14. F o r Eh, it is
convenient to specify a range of about 4 - 1 . 4 to 1.0 V. In certain regions of the EhpH field, water will be oxidized to 0 , and in other regions water will be reduced to
2

1!,. We will calculate these regions as an e x a m p l e problem.

EXAMPLE
PROBLEM

Calculate the stability field for water at standard conditions.


T h e oxidation o f water is given by
0 ( g ) -I- 4 H
2

4- 4 c"

2 H 0(/)
2

255

Inorganic Chemicals in G r o u n d W a t e r

F r o m T a b l e 6.4,
A G g o , = - 5 6 . 6 9 kcal
2

A-Co

= 0

2(8l

A G ^ = 0
F r o m Equation 6.20,

2Aca

AG?, =

20(1)

- AGg

- 4 AG,

; ( g )

A GR = 2 ( - 5 6 . 6 9 ) - 0 - 4 ( 0 ) = - 1 1 3 . 3 8 kcal
T h e value of A G ^ in kilocalories is converted to a standard potential by use of
Equation 6.19:
e

-AGI

- ( - 1 1 3 38)

4 23.06

T h e Nernst equation ( E q u a t i o n 6 . 2 4 ) can b e e x p r e s s e d as:


E h

o _ ^

.
nF

[H 0]
" [0 ][H ]

'

T h e activity of dissolved g a s e o u s o x y g e n is e x p r e s s e d as a partial pressure, P q . .


At standard conditions it has a value of 1 atm. T h e activity of water is unity. T h e Nernst
equation thus reduces to
E h

0.00199 - 2 9 8
,
.
/
Zrs
- 3 0 3 log[H ] "
4 23.06
2

9 -

This e x p r e s s i o n can be reduced to


Eh

( v o l l s )

= 1 . 2 2 9 - 0.0592pH

T h i s equation defines the upper boundary of stability for water, a b o v e which


oxidation would break apart the water molecule.
T h e reduction of hydrogen ions to form g a s e o u s hydrogen is
2H

+ 2 e "

2 ( g a s )

F r o m T a b l e 6.4,

ACS* = 0
A^lgas) = 0
T h e value o f A G

for the formation o f hydrogen gas i s obviously zero. Therefore,

the value of E is also zero.


F r o m the Nernst equation,
-

0.00199 298

2}0i
E

2-23.06

^WV

T h e value of P is 1 atm and the calculated value of E is 0, h e n c e the preceding


2

e x p r e s s i o n can b e reduced t o
Eh,
v 0

s )

= 0 . 0 0 0 - 0.0592pH

256

Chapter S i x

pH
F I G U R E 6.2

Eh-pH diagram showing the stability field for water.

Inorganic Chemicals in G r o u n d W a t e r

257

This equation forms the lower boundary of the stability held for water. T h e s e
boundaries are plotted in Figure 6.2.

6.5.4

Calculating E h - p H Stability Fields

T h e stability fields within the Eh-pH diagram for various forms of an element can be
c o m p u t e d using chemical thermodynamics. B a s i c s o u r c e s of thermodynamic data include
Wagman et al. ( 1 9 6 8 , 1 9 6 9 , 1 9 7 1 ) and R o b i e , Hemingway, and Fisher ( 1 9 7 8 ) .
B o u n d a r i e s for an element b e t w e e n dissolved s p e c i e s that have different valence
states are c o m p u t e d using the Nernst equation (Equation 6.27, 6.28, or 6 . 2 9 ) . If b o t h
ions are at the s a m e valence state, then the equation for chemical equilibrium is used.
If o n e is calculating the boundary b e t w e e n a solid s p e c i e s and a dissolved form, the
chemical activity of the solid s p e c i e s is 1. F o r s o m e of the boundaries of solid species,
it will be necessary to assume an activity for the dissolved species.

EXAMPLE

Calculate an Eh-pH diagram for iron in which the solid s p e c i e s are F e ( O H )

PROBLEM

and the activity o f dissolved iron is 56 fig/L ( 1 C T M ) .

and F e O

Soluble forms of the ferrous ion and the ferric ion include F e

2 +

, Fe

3 +

, FeOH

2 +

and Fe( O H ) . Transformations b e t w e e n these ions are determined b y r e d o x equations:


2

FeOH
Fe(OH)

+
2

2 +

+ H
+ 2 H
Fe

+ e"
+

Fe

+ e" ^ F e

3 +

+ e"

2 +

2 +

+ H 0
2

+ 2 H 0
2

Fe

2 +

(6.30)
(6.31)
(6.32)

T h e free energy and the standard potential for these reactions can be determined
from Equations 6.20 and 6.19, respectively.
From T a b l e 6.4 free energies are as follows.
FeOH
Fe(OH)

2 +

+
2

Fe

2 +

Fe

3 +

= - 5 4 . 8 3 kcal/mol

FeO = - 6 0 . 0 3 kcal/mol
F e ( O H ) = - 1 6 6 kcal/mol

= - 1 0 6 . 7 kcal/mol

F e ( O H ) ~ = - 1 9 8 . 4 kcal/mol

= - 1 8 . 8 5 kcal/mol

= - 1 . 1 kcal/mol

=0

H 0 = - 5 6 . 6 9 kcal/mol
2

F o r Reaction 6.30 ( F e O H

2 +

+ H

AG" = [AGS o + 4 C
2

AG , =

F t !

+ e" ^ F e

+ H 0):
2

. ] - [AGp QH2+ + AGg.]


C

-56.69 + (-18.85) - (-54.83) - 0

AG?, = - 2 0 . 7 1 k c a l / m o l
p0

2 +

AG ,
-(-20.71)
_ :
1 y
nF

E = + 0 . 8 9 8 V

1 23.06

258

Chapter S i x

F o r Reaction 6.31 ( F e ( O H )

+
2

+ 2 H

+ e" ^ F e

AGS = 2AG o + A C , V - A C ?
2

t ( O H 1 :

2 +

+ 2 H 0):
2

. - 2AC&.

AGg = 2( - 5 6 . 6 9 ) + ( - 1 8 . 8 5 ) - ( - 1 0 6 . 7 ) - 2(0)
AG?, = - 2 5 . 5 3 k c a l / m o l
E

- A G _ -(-25.53)

nF

1 23.06

E = + 1.107 V
3

+ e

ACS =

^ F e

):

4Cf,

- AGp j.

agS =
agS = - 1 7 . 7 5

- (-1.1)

-18.85

E =

-ACS

-(-17.75)

nF

1 23.06

E = + 0.770
T h e boundaries between the stability fields are determined from the Nernst equa
tion. At the boundary between two fields, the activities of the iron species on the left of
the reaction is equal to the activity of the iron species on the right of the equation
i.e., the two species are at equilibrium.
F o r Reaction 6.30 ( F e O I I

2 +

+ H

+ e" ^ F e

2 +

+ H 0):
2

2+

^
Eh = E

0.0592,
[Fe ]
m
log -.
trr - 0 . 0 5 9 2 pH
n
(FeOH ]
n
6

2 +

2 +

2 +

Since | F e O H ] = ( F e ), m ( t h e n u m b e r of hydrogen i o n s ) = 1, ; ; ( t h e n u m b e r
of e l e c t r o n s ) = 1, E - + 0 . 8 9 8 V, and log 1 = 0:
Eh

( v o l I S )

F o r Reaction 6.31 ( F e ( O H )

+
2

= 0 . 8 9 8 - 0 . 0 5 9 2 pH
+ 2 H

(6.33)
2

+ e" ? F e * + 2 H 0 ) :
2

2+

0.0592

Eh = E

[Fe ]

m
r - 0.0592 - pH

og
+

n
*[Fe(OH) )
n
] = [ F e ], m= 2, n= 1, and E = + 1.107V:
2

Since [ F e ( O H )

+
2

2 +

Eh(voi. ) = 1 1 0 7 - 0 . 1 1 8 4 pH

(6.34)

For Reaction 6.32 ( F e

3 +

Eh =

+ e" ^ F e

2 +

):
2 +

0.0592
|Fe ]
m
log 1 ^ - 0 - 0 , 9 2 - p H
+

This reaction is independent of p l l b e c a u s e neither |F1 ] nor | O H J appears in


the reaction. H e n c e the value of tn is 0. B e c a u s e as I F e

3 +

2 +

] = I F e 1 and log 1 = 0, Eh

is a constant equal to E, which is 0 . 7 7 0 V:


Eh = 0 . 7 7 0 V

(6.35)

259

Inorganic Chemicals in G r o u n d W a t e r

T h e boundary between two dissolved species that are at the s a m e valence state
can be determined from chemical equilibrium.
F o r iron there are two boundaries b e t w e e n dissolved ions of ferric iron. T h e s e
boundaries are represented by these reactions:
Fe
FeOH

3 +

2 +

+ H 0

FeOH

+ H 0

Fe(OH)

2 +

3 +

+ H
+
2

2 +

(6.36)

+ H

(6.37)

F o r Reaction 6.36 ( F e
+ H 0 ^ F e O H
+ H ) , the equilibrium constant can
be o b t a i n e d from the free energy of the reaction. T h e first step is to find the free energy
of the reaction using Equation 6.20.
2

AG

AGg 2

AG

-54.83 + 0 - ( - 1 . 1 ) - ( - 5 6 . 6 9 )

AG

e0

+ AG&

- AG?,., - AC o

+ 2.96 k c a l / m o l

T h e n e x t step is to determine the equilibrium constant using Equation 6.22.


AGS
b

2.96

^ = - l 3 6 4
K.. = 10

= - U 6 4 = -

-2.17

F r o m Equation 6.6,
2 +

_ [FeOH
[Fe

3 +

][H + ]

][H 0]
2

S i n c e [ H 0 ] = 1 and at the boundary [ F e O H

2 +

[H ] = 1 0 -

3+

] = [Fe

],

2 1 7

(6.38)

This m e a n s that a vertical line at a pH of 2.17 separates these two stability fields.
F o r Reaction 6.37 ( F e O H
4- H 0 ^ F e ( O H )
+ H ), find the free energy of
the reaction:
2 +

AG% = AG?e(OH)i+

+ AGS+

- AGp i*

- AGft o

MH

AGS = - 1 0 6 . 7 4- 0 - ( - 5 4 . 8 3 ) - (56.69)
AG

4-4.82 k c a l / m o l

Next find the value of K


-AGg
l

-4.82

T364- U64"

K , = 10

3.53

Finally, from Equation 6.6,


2

_ [ F e ( O H ) + ][H + ] _
"

[FeOH

2 +

][H 0]
2

260

Chapter S i x

2+

2+

S i n c e [ H O j = 1 and [ F e ( O H ) ] = [ F e O H ],
2

[H ] = 1 0 - '

5 3

(6.39)

Lines that delineate the stability field for solids can be obtained by similar reasoning.
R e m e m b e r that the activity of a solid in equilibrium with dissolved species is 1. T h e
location of the boundaries of solid species is a function of the amount of dissolved iron
present.
In this situation there are two stable iron precipitates, F e ( O H )
reactions at the boundaries include
Fe(OH) + H

Fe(OH) + 3 H

Fe(OHV + H

FeO + 3 H 0
+

Reaction 6.40 ( F e ( O H ) 4- H ^ F e ( O H )
rium approach:
3

A t j

H)

F e ( 0

4" A g S o

.=-Fe

4-H 0

(6.43)
(6.45)

A G F

(6.42)

4- H 0 ) is solved using an equilib

(6.40)

(6.44)

+ e"

and FeO. T h e

(6.41)

FeO 4 - 2 H

AC =

+ 3 H 0

FeO + H 0 + O H "

+ H 0

Fe(OH) + H 0

F e ( O H ) + e~
Fe(OH) " + 2 H

2 +

+ e" = Fe

Fe(OH)

AGS

AG = - 1 0 6 . 7 + ( - 5 6 . 6 9 ) - (-166) - 0 = 2.61
ag_
2.61

log

r =

Keq =

1.364

- t l 6 4 = -

-1.91

l f J

K = [ F e ( O H )

|H 0]

,.

(Fe(OH)

9 1

[H ]

S i n c e ( F e ( O H ) ] = 1 and ( H 0 ) = 1,
3

[H 1=^SP
Reaction 6.41 ( F e ( O H ) + 3 H * 4 e " ^ F e
Nernst equation:

AG = 4 G , + 3 AG o - A G ?
2

(6-46)
4 - 3 H 0 ) i s solved using the
2

e ( O H l j

- 3 AG.

AG = - 1 8 . 8 5 + 3 C - 5 6 . 6 9 ) - ( - 1 6 6 ) - 0
AG = - 2 2 . 9 5 kcal/mol
g

-AGS
?

nF
_

E h

-(-22.95)

1 23.06

0.0592

^ _ _

= 4-0.994 V

2+

[Fe ]

_ L _ i _ _ 0 . 0 5 9 2 - p H

Inorganic Chemicals i n G r o u n d W a t e r

261

S i n c e there are three hydrogen ions (m = 3) and o n e e l e c t r o n (n = 1) and


l F e ( O H ) ] = 1, then
3

Eh,voi. ) = 0.994 - 0 . 0 5 9 2 l o g ( F e

2 +

Reaction 6.42 ( F e ( O H ) ~ + H

^ F e ( O H )

] - 0.178pH

(6.47)

+ H 0 ) i s solved b y using a n equi-

librium approach:
AG'S

AG

AG?, =
,
1

AGf

AG?.

AC?i?

e ( O H

AG +

) 4

-24.29
24.29

^ = - 1 3 6 4 =
A:.,

- 1 6 6 - 56.69 - ( - 1 9 8 . 4 ) - 0

h)J

c ( 0

io

1 7

1^64 =

[Fe(OH) ][H Q]
3

"

(Fe(OH) "][H ]
4

S i n c e [ F e ( O H ) ] = 1 and [ H 0 ] = 1 ,
3

-i7.

[H

] = ^ r

(6.48)

[Fe(OH) -]
4

Equation 6.43 ( F e ( O H )

+ e ~ ^ F e O + H 0 + O H ) i s solved using the Nernst

equation:

AG?, = AG? + AGSp + AGg- - AG? o ,


0O

AG?. =

AGS

E h

e(

H 3

-60.03 + (-56.69) + (-37.59) - (-166)

= + " 6 9

-AGS

-11.69

nF

1 23.06

o _^0592
n

-0.507 V

[FeO][H Q][OH-]
2

'

[Fe(OH) ]
3

S i n c e [ F e ( O H ) ] = 1, [ F e O ] = 1, [ H O j = 1, a n d n - 1,
3

Eh = - 0 . 5 0 7 - 0 . 0 5 9 2 l o g [ O H ~ ]
B e c a u s e the diagram uses pH as a variable,
terms
log 1 0 "

of
1 4

pH.

By

definition,

[OH ] = 10

_ 1

[OH ]
+

* / [ H ];

must be e x p r e s s e d in

therefore,

log [ O H ] =

- log [H ], so that
Eh =

- 0 . 5 0 7 - 0.0592(pH - 14)

(6.49)

262

Chapter S i x

Reaction 6.44 ( F e ( O H )
Nernst equation:

+ 2 H ' + e

^ F e O + 3 H 0 ) i s solved using the


2

AG = AG^o + 3 AG o - A C ? . o , - - 2 AGS*
K

AG?, = - 6 0 . 0 3 + 3 ( - 5 6 . 6 9 ) - ( - 1 9 8 . 4 ) - 0
AG

-31.7kcal/mol
AG?

-(-31.7)

1-23.06
0.0592,
_ ^

0
E

'

[FeO)[H Q]
_ _ _ _
2

Since [ F e O ] and [ H 0 ] are 1 and n = 1,


2

E h = + 1 . 3 7 5 - 0 . 0 5 9 2 log 7 - .
[Fe(OH) -][H
6

This can be e x p r e s s e d in terms of pH as


+

Eh = 1.375 + 0.0592 l o g l F e ( O H V ] + 2 ( 0 . 0 5 9 2 ) log[H ]


Eh,,,,,) = 1.375 + 0 . 0 5 9 2 l o g [ F e ( O H ) - ] - 0.118pH
4

Reaction 6.45 ( F e O + 2 H

^ Fe

AGS =

2 +

(6.50)

+ H 0 ) is solved as an equilibrium reaction:


2

+ AGS o - AGp
2

t0

- 2AG.

AG?, = - 1 8 . 8 5 + ( - 5 6 . 6 9 ) - ( - 6 0 . 0 3 ) - 2 ( 0 )
AG? = - 1 5 . 5 1 kcal/mol
l

AG
T36T

'

1 1 3 6

AT = l O
e

2 +

][l

[FeO][H

Since [ H O l and [ F e O ] = 1,
2

[Fe
10

2 +

[ + 1 /

[ F e 2

IH 1 =

/^Ti36

(6.51)

Several of the equations, including Equations 6.46, 6.47, 6.48, 6.50, and 6.51, depend
upon the activity of the dissolved iron.
T h e following equations, which are independent of dissolved iron activity, have
b e e n derived.

Inorganic Chemicals in G r o u n d W a t e r

263

Equation
Number

Boundary
FeOH

2 +

-Fe
+

Fe(OH)
Fe
Fe

3 +

2
2

-Fe

FeOH

2 +

Equation

6.31
6.32

-Fe
-FeOH

3 +

2 +

Eh

( v o l l s )

Eh(voiis)

-Fe(OH)

pH = 3.53
= 0.322 - 0 . 0 5 9 2 p H

6.39

Fe(OH) -FeO

6.49

= 1.107 - 0.118pH

= 0.770
pH = 2.17

6.33
6.38

2 +

= 0 . 8 9 8 - 0.0592pH

E r i

(olls]

T h e s e equations n e e d a dissolved iron activity, which is set at 10

6 00

mol:

Equation
Boundary

Number

Equation

6.46
Fe(OH) -Fe(OH)
3

[H ] = [ 1 0 - ] / 1 0 - ' =

pH-

9 1

4 0 9

6.47
Fe(OH) -Fe

10"

Eh

(ohs)

4.09
6

0.994 - 0.0592 l o g [ l 0 " ]

- 0.0178pH

Eh

Fe(OH) --Fe(OH)
4

( v o l l s )

= 1.349 - 0 . 1 7 8 p H
6

(H ] = i o - ' 7 [ i o - ]
+

6.48

1 7

IO"

1 1 8

pH = 11.8
6

Efy,i, , = 1 3 7 5 + 0 . 0 5 9 2 l o g [ l 0 " ]

6.50
Fe(OH) -

FeO

- 0.118pH

Eh vi,s) = 1 0 2 0 2 - 0 . 1 1 8 p H
(

FeO-Fe

[H ] = ( [ l 0 - ] / 1 0 " -

6.51

[H ] = I O "

3 6

) -

8 6 8

pH = 8.68

O n c e the equations have b e e n developed for the desired molar c o n c e n t r a t i o n of


dissolved iron, the lines represented by the equations are plotted on an Eh pH field.
This has b e e n d o n e in Figure 6.3. T h e equation n u m b e r s are on the lines.
In order to finish the stability field diagram, we n e e d to decide which segment of
3+

each line is needed. If we start with F e , it participates in two reactions, o n e characterized


by Equation 6.33 and the o t h e r by Equation 6 . 3 8 . Equation 6.33 is a horizontal line at
0.77 V and Equation 6 . 3 8 is a vertical line at pH = 2.17. T h e s e two lines define a c o r n e r
at the upper left of the diagram in which F e
region from the F e O H

2 +

3 +

can exist. Equation 6 . 3 8 divides the F e

3 +

region. T h i s region is also b o u n d e d by Equations 6.31 and

6.39. T h e s e g m e n t s of these lines that intersect are used to define the F e O H

2 +

region.

In a similar m a n n e r of analysis, the line s e g m e n t s surrounding each of the stability fields


are determined. T h e resulting Eh-pH diagram for the iron system is thus determined
and is illustrated in Figure 6.4.

264

Chapter S i x

F I G U R E 6.3

Equations for an Eh-pH diagram for dissolved iron with dissolved iron activity of 10

mol

under standard conditions. (Numbers adjacent to dotted lines refer to equations discussed in the text.)

266

Chapter S i x

F I G U R E 6.5

Eh-pH diagram showing fields of stability for dissolved iron under standard conditions.

Activity of dissolved iron is 10

mol (56 ^ g / L ) , of sulfur species is 96 mg/L as S 0

dioxide species is 61 mg/L as H C 0

z
4

~ , and of carbon

. Source: J. D. Hem, Study and interpretation of the chemical

characteristics of natural waters, U. S. Geological Survey Water Supply Paper 2 2 5 4 , 1985.

Inorganic Chemicals in G r o u n d W a t e r

267

T h e iron Eh-pH diagram of Figure 6.4 is for a system that contains only dissolved
iron. If o t h e r e l e m e n t s are present, such as sulfur, then additional iron c o m p o u n d s arcp o s s i b l e . Figure 6.5 s h o w s an Eh-pH diagram for a system with an iron activity of
6

5 6 ng/L ( 1 0 ~ m o l ) , sulfur o f 9 6 m g / L a s S 0

~ , and c a r b o n d i o x i d e o f 6 1 m g / L a s

2
4

H C 0 ~ . Solids in the shaded area are thermodynamically stable. Under the conditions
3

specified in this diagram, iron c a r b o n a t e ( F e C 0 ) saturation was n o t reached and n o n e


3

is r e c o r d e d as a solid phase.
T h e area of the region in which iron is precipitated rather than dissolved is a
function of the c o n c e n t r a t i o n of dissolved iron. T h e m o r e dissolved iron that is present,
the greater the size of the stability field for the precipitates. This is illustrated in Figure
6.6. I n this diagram the sulfur i s 9 6 m g / L a s S 0

2
4

and the c a r b o n a t e i s 6 1 m g / L a s

H C 0 ~ . Dissolved iron ranges from 5.6 / t g / L to 56 mg/L.


3

Eh-pH diagrams have b e e n used in the ground-water literature to explain such


p h e n o m e n a as the solubility of ferric o x y h y d r o x i d e s ( W h i t t e m o r e and Langmuir 1 9 7 5 ) ,
hexavalent c h r o m i u m ( R o b e r t s o n 1 9 7 5 ) , m a n g a n e s e ( H e m 1 9 8 5 ) , iron, c o p p e r , silver,
chromium, manganese, vanadium, m o l y b d e n u m , and arsenic ( H e m 1 9 7 7 ) , uranium (Lang
muir 1 9 7 8 ) , thorium (Langmuir and Herman 1 9 8 0 ) , and arsenic (Matisoff et al. 1 9 8 2 ) .

6.6

Metal Complexes
6.6.1

Hydration of Cations

Although we c o n s i d e r that metallic ions exist in solution as an isolated ion, such as


Cu

2+

. in fact that is not the case. T h e C u

2 +

ion is surrounded by polar water a t o m s

that are chemically b o u n d to the ion. Metallic ions, in general, have six water m o l e c u l e s
surrounding them. T h e hydrated cupric ion is C u ( H 0 )
2

2+
6

. Even outside the shell of

chemically b o u n d water m o l e c u l e s , there is a region w h e r e the polar water m o l e c u l e s


are o r d e r e d by the electrostatic charge of the metallic ion. Anions in c l o s e association
with a metal cation are called l i g a n d s ; together they form a coordination c o m p o u n d .
Water is c o n s i d e r e d to be a l i g a n d that is b o u n d to the metal ion. If o t h e r ligands bind
to the metal, they must replace s o m e of the water m o l e c u l e s acting as ligands. T h e
stability of a c o m p l e x relative to cation or ligand e x c h a n g e can be d e s c r i b e d by equilibrium constants for the reaction.
6.6.2

Complexation

T h e following inorganic anions act a s simple ligands i n natural waters: O H ~ , C 0


S0

2
4

" , OT, Br", F~, N 0 ~ , S i 0


3

2
3

", S ", S 0 " , P0


2

3
4

~, P 0
2

4
7

~, P O
3

5
l 0

" and CN~.

A m m o n i a ( N H ) is a polar m o l e c u l e that can also act as a ligand. Ligands can b o n d


3

either covalendy or electrostatically with a metal to form a c o m p l e x ion or c o m p o u n d .


We have already l o o k e d at the c o m p l e x forms of the ferric ion and the hydroxyl ion.
2+

T h e y form a series of c o m p l e x ions: F e O H , F e ( O H ) , F e ( O H )


2

and F e ( O H ) " .
4

C o m p l e x formation is involved with c h e m i c a l equilibrium of ionic c o m p o u n d s and


oxidation-reduction reactions.

.?

In the c a s e of monovalent ions, there is ranHyone site w h e r e the ligand b o n d s to


the metal ion. If the ligand has m o r e than o n e site that can b o n d , then it forms what is
known a s a c h e l a t i n g a g e n t . F o r e x a m p l e , the p y r o p h o s p h a t e ion, P 0
2

4
7

" , can b o n d

268

Chapter S i x

_ [ oo
0

I I

I
4

'

10

12

14

pH
FIGURE 6.6

Equilibrium activity of dissolved iron as a function of Eh and pH under standard conditions,

sulfur activity of 96 mg/L as S O ,


Hem,

2 -

and activity of carbon dioxide of 61 mg/L as H C 0 ~ . Source-. J. D.

Study and interpretation of the chemical characteristics of natural waters,

Water Supply Paper 2 2 5 4 , 1 9 8 5 .

U . S . Geological Survey

Inorganic Chemicals in G r o u n d W a t e r

269

to a metal ion, such as cadmium, at two locations:


2

o ~

o = p o P = O
x

6.6.3

c d

Organic Complexing Agents

B o t h natural waters and wastewaters contain a n u m b e r of organic c o m p o u n d s that c a n


act as chelating agents. In general, t h e s e organic c o m p o u n d s have a functional g r o u p that
c o n t a i n s oxygen, nitrogen, p h o s p h o r o u s , or sulfur. If R symbolizes o n e or m o r e c a r b o n
a t o m s with the appropriate n u m b e r o f hydrogens, then organic c o m p l e x e s can include
functional g r o u p s such as
O

II

CT

OH

n:

\
Alcohol

I
H

Carboxylale

Amine

O
Phosphate

T h e r e are a n u m b e r of organic c o m p l e x i n g agents that o c c u r in nature. T h e y are


associated with h u m i c s u b s t a n c e s that form from the d e c o m p o s i t i o n o f vegetation.
T h e s e are c o m p l e x organic m o l e c u l e s with m o l e c u l a r weights ranging upward into the
tens of thousands. If a h u m i c s u b s t a n c e is e x t r a c t e d with a strong b a s e and then acidified,
there are three products. T h e n o n e x t r a c t a b l e organic material is called l i u i n i n . Substances called f u l v i c a c i d s remain in the acidic solution, and o t h e r s u b s t a n c e s called
h u m i c a c i d s precipitate from the acidified extract. T h e s e represent classes o f c o m p o u n d s that contain many different individual organic m o l e c u l e s . Humic and fulvic acids
contain many functional g r o u p s that can c h e l a t e to metals. Metals may be kept in solution
by chelation with s o l u b l e fulvic acids or they may bind to the i n s o l u b l e h u m i c s u b s t a n c e s
b y cation e x c h a n g e (Manahan, 1 9 8 4 ) .
Synthetic organic c o m p l e x i n g agents are used in a n u m b e r of industrial p r o c e s s e s .
T h e y can be used as cleaning c o m p o u n d s , as constituents of detergent, in metal plating
baths, and in water conditioning. T h e s e c o m p o u n d s include sodium tripolyphosphate,
sodium ethylenediaminetetraacetate ( E D T A ) , citric acid, and sodium nitrilotriacetate
( N T A ) . T h e structures o f s o m e o f t h e s e c o m p o u n d s are given i n Figure 6.7. Synthetic
chelating agents may k e e p metals in solution under c o n d i t i o n s w h e r e the unchelated
metal would precipitate or u n d e r g o cation e x c h a n g e .
EDTA in wastewater can vastly i n c r e a s e the mobility of associated metals in the
subsurface. Monitoring wells near radioactive waste disposal t r e n c h e s at the O a k Ridge
( T e n n e s s e e ) National Laboratory c o n t a i n e d significant levels of sodium EDTA, which
was used as a cleaning agent. T h e s a m e wells also c o n t a i n e d radioactive

6 0

C o , a metal

that is normally not e x p e c t e d to migrate very far due to cation e x c h a n g e . T h e

6 0

C o had

b e e n chelated by the EDTA and h e n c e had gready increased mobility in the subsurface
( M e a n s , Crerar, and Duguid 1 9 7 8 ) .

270

Chapter S i x

CO,H

I
H0 C CH
2

C CHj C 0 H
2

OH
Citric acid

H
Nirrilotriacelore (NTA)

F I G U R E 6.7

Structure of chelating agents including (a) citric acid, (b) nitrilotrlacetate (NTA), and (c)

ethylenediaminetetroacetate

6.7

(EDTA).

Chemistry of Nonmetallic Inorganic Contaminants


6.7.1

Fluoride

Fluoride o c c u r s in water as the F" ion. In natural waters the amount of fluoride present
is generally less than 1.0 m g / L , although concentrations as great as 67 m g / L have b e e n
reported ( H e m 1 9 8 5 ) . Fluoride is present in minerals such as fluorite ( C a F ) and apatite
2

( C a ( C I , F , O H X P 0 ) ) . Weathering of these minerals may release fluoride. It may be


5

released as a contaminant from industrial p r o c e s s e s utilizing hydrofluoric acid. Cryolite


( N a A l F ) is used as a flux in the electrolytic production of aluminum. T h e manufacture
3

of p h o s p h a t e fertilizer from p h o s p h a t e rich rock may also release fluoride. Effluent from
a Florida fertilizer plant had fluoride ranging from 2 8 1 0 to 5 1 5 0 mg L ( C r o s s and Ross
1970),

Inorganic Chemicals in G r o u n d W a t e r

271

Fluoride can form c o m p l e x e s in water with a n u m b e r of cations, including aluminum, beryllium and ferric iron ( H e m , 1 9 8 5 ) . Dissolved fluoride can react with calcium
t o form fluorite. T h e solubility product for fluorite i s 1 0 "

1 0 A

. Precipitation o f fluorite

can act as a control on the amount of dissolved fluoride in solution if dissolved calcium
is present. T a b l e 6.5 s h o w s the equilibrium a m o u n t of dissolved fluoride calculated for
various activities of calcium. Actual activities of fluoride are likely to be somewhat higher
due to the effect of the ionic strength of the solution as well as the effect of any c o m p l e x e s
that might form with the fluoride ion.
C o r b e t t and Manner ( 1 9 8 4 ) have reported on the distribution of fluoride in water
from b o t h unconsolidated and b e d r o c k aquifers of northeastern O h i o . T h e y found that
2 3 9 out of 2 5 5 wells had fluoride c o n c e n t r a t i o n s less than 1 mg/L. However, 14 of the
wells had fluoride ranging from 1 to 5.9 m g / L . All these high-fluoride wells w e r e associated
with a specific b e d r o c k formation. Such information is useful from a public health
standpoint b e c a u s e there is a 2 . 0 - m g / L drinking-water criteria for fluoride. S o m e fluoride
is n e e d e d to build strong teeth in growing children; however, fluoride in e x c e s s of
2.0 m g / L will c a u s e teeth to discolor.
6.7.2

Chlorine and B r o m i n e

T h e halides c h l o r i n e and b r o m i n e have similar chemistry, although c h l o r i n e is far m o r e


abundant in nature than b r o m i n e . Even though the e l e m e n t s can exist in a n u m b e r of
oxidation states, the chloride and b r o m i d e ions ( C P , B r " ) are the only o n e s o f significance in natural waters. Chlorine gas is used an a disinfectant for purification of water
and is a strong oxidizing agent w h e n dissolved in water. T h e chloride ion o c c u r s in
natural waters in fairly l o w c o n c e n t r a t i o n s , usually less than 100 m g / L , unless the water
is brackish or saline. Chloride is used by humans in many applications and c a n be added
to the subsurface via industrial discharges, sewage, animal wastes, and road salting.
Commercial fertilizers can contain chloride as KC1. Chlorine and b r o m i n e are c o m p o n e n t s
of halogenated organic c o m p o u n d s used for industrial solvents and pesticides. T h e s e
c o m p o u n d s have b e e n released to the environment b o t h intentionally through the use
of pesticides and accidentally through spills and leaks.
Chloride and b r o m i d e ions are not reactive. T h e y don't participate in r e d o x reactions, aren't s o r b e d o n t o mineral or organic surfaces, and don't form insoluble pre
cipitates. Chloride is s o m e t i m e s used as a tracer in ground-water studies b e c a u s e it is
conservative.

T A B L E 6.5

Equilibrium fluoride concentrations as a function of calcium activity.


Calcium

Fluoride

Activity

Concentration

Activity

(mol)

(mg/L)

(mol)

800

4.48 x 1 0 "

400

6.31 x 1 0 ~

2 x 1(T
1CT

io

s x

icr

- 3

5 x

io-

10-

200

8.92 x 10

Concentration
(mg/L)
5

0.85

1.20

1.70

40

2.00 x 1 0 -

20

2.82 x 1 0 "

5.36

6.31 x 10

11.99

3.79

272

Chapter S i x

6.7.3

Sulfur

Sulfur is released to the environment by the weathering of minerals containing the


element. Rock containing pyrite can be oxidized to release sulfur, with microorganisms
acting as a catalyst and mediating the oxidation. This is the s o u r c e of the acidic water
that drains from many areas that have b e e n mined. Sulfuric acid is widely used in industrial
p r o c e s s e s . Sulfur can be released to the environment by the processing of sulfide o r e s
and by the burning of fossil fuels, all of which contain sulfur to s o m e degree.
2

Sulfur can exist in valence states ranging from S " to S

+ 6

. Figure 6.8 is an Eh pH

diagram showing the stability o f the two oxidized forms o f sulfur, H S 0 ~ and S 0
4

~,

and the three reduced forms, S " , H S " , and H S ( a q u e o u s ) . T h e field o f stability for
2

elemental sulfur is also shown. T h e total sulfur activity used in computing the diagram
3

is 1 0 " m o l / L o r 9 6 m g / L a s S 0

2
4

" . I f a greater total sulfur activity were used, the stability

field for elemental sulfur would be larger. Although this is a very useful diagram for
understanding the equilibrium conditions for dissolved sulfur, the r e d o x reactions can
be slow if m i c r o b e s are not mediating the reactions. Hence, it may take a long time for
the system to reach equilibrium.
Gypsum (calcium sulfate) is quite soluble in water (K = 1 0 ~
eq

4 , 6

) and, e x c e p t for

waters with extremely high sulfate, would not be a sink for sulfate. Strontium sulfate is
6

sparingly s o l u b l e (K = 1 0 ~ ' ) , whereas barium sulfate is nearly insoluble (K =


eq

jQ-io.oj

o w e v e r

eq

strontium and barium are not found in much a b u n d a n c e in natural

waters. Sulfate could act as a sink for strontium and barium.


6.7.4

Nitrogen

Nitrogen is another element that can o c c u r in both oxidized and reduced forms as well
a s the elemental state. T h e c o m m o n forms o f inorganic nitrogen include nitrate, N 0 " ,
3

nitrite, N 0 ~ , nitrogen gas, N , ammonium, N H


2

+
4

and cyanide, C N " . Nitrogen i s also

a major constituent of organic matter in the form of a m i n o acids. T h e majority ( 7 8 % )


of the Earth's a t m o s p h e r e is nitrogen gas. Atmospheric nitrogen can be "fixed", or
converted to nitrate, by cyanobacteria in lakes and the o c e a n and by bacteria living on
the roots of plants such as legumes and lichens. Atmospheric nitrogen can also be
converted to oxidized and reduced forms via fertilizer production and by heating it to
high temperatures in internal c o m b u s t i o n engines, power plants, lightening discharges
and forest fires. Rainwater contains dissolved nitrate and ammonia. Nitrogen is released
to the subsurface from sewage, animal wastes, and fertilizers.
In soil and ground water, oxidation and reduction of nitrogen species is accomplished by microorganisms. Under oxidizing conditions a m m o n i a is c o n v e n e d to nitrite,
which is converted to nitrate. Nitrite is a very reactive ion and is almost immediately
converted to nitrate, so that little nitrite is normally found in the environment. Under
reducing conditions nitrate is converted primarily to nitrogen gas, a p r o c e s s known as
d e n i t r i f i c a t i o n . Organic matter will decay to ammonia under reducing conditions.
Septic tank effluent, for e x a m p l e , normally has high a m m o n i a and very little nitrate. If
the receiving ground water is reducing, the nitrogen will stay in the ammonia form. If
it is oxidizing, bacteria will convert the a m m o n i a to nitrate (Feth 1 9 6 6 ) .
Nitrate contamination of ground water has b e e n d o c u m e n t e d in a n u m b e r of areas
(e.g. Hill 1982; Flipse et al. 1 9 8 4 ; and Silver and Fielden 1 9 8 0 ) . Hill studied the distribution

Chapter S i x

of nitrate in ground water from a shallow unconsolidated sand aquifer. It was found
mat the ground water b e n e a t h areas of forest or permanent pasture has less than 1.0
m g / L of nitrate as nitrogen. T h e g r o u n d water beneath heavily fertilized potato fields
typically contained in e x c e s s of 10 mg/L nitrate as nitrogen. Gray and Morgan-Jones
( 1 9 8 0 ) found that the nitrate c o n t e n t of g r o u n d water in a study area increased over
the past 40 yr and that the use of fertilizers in this c a t c h m e n t area also increased over
the s a m e time period.
Nitrogen o c c u r s as two isotopes,

1 4

N and

1 5

N . Of the two,

1 4

N is by far the most

1 5

abundant in the atmosphere. T h e relative a b u n d a n c e of N t h a t is, the

l 5

N / ' * N ratio

in nitrate may be used to distinguish nitrate that c o m e s from animal and human waste
from nitrate that c o m e s from mineral fertilizers (Flipse et al. 1 9 8 4 ) .
The

1 5

N/

1 4

N ratio is usually e x p r e s s e d as a S' N value, which is defined as

, ,
<5 N (%o) =

1 5

N/

15

1 4

1 5

l 4

N ) s a m p l e - ( N / N ) standard
ri5M/i*xA
jj
x 1000
( N / N ) standard
1 5

1 4

w h e r e %o stands for parts per thousand.


15

If the i 5 N is positive, then the nitrate of the sample has b e e n enriched in

1 5

with respect to the standard. F o r nitrogen, the standard is the atmospheric composition.
15

Nitrate from animal and human waste typically has a i 5 N in e x c e s s of 4- 10%o.


Flipse and B o n n e r ( 1 9 8 5 ) found that mineral fertilizers used on Long Island had
15

15

<5 N values that averaged 0.2%o at o n e site and 5.9%o at another. However, the <5 N
of the g r o u n d water b e n e a t h the sites that had b e e n fertilized was about 4- 6%o. This
15

increase in <5 N from the mineral fertilizer was attributed to fractionation that o c c u r r e d
15

during infiltration of the nitrogen. However, the resulting <5 N was still clearly lower
than that e x p e c t e d from animal and human waste.
6.7.5

Arsenic

Arsenic can o c c u r in valance states of 4 - 5 , 4-3, 4 - 1 , 0 and 3 . However, the important


states of dissolved arsenic in water are the arsenate H A s 0

3 _ n
4

, with a valance state of

4 - 5 , and the arsenite H A s 0 ~ " , with a valance state of 4-3. An Eh-pH diagram for
3

arsenic that s h o w s the fields of stability for the arsenates and arsenites is given in Figure
6.9- Dissolved arsenic s p e c i e s can be a b s o r b e d by ferric hydroxides. Arsenic ( 4 - 5 ) is
m o r e strongly s o r b e d than arsenic ( 4 - 3 ) . Ferric hydroxides are stable over a wide Eh
pH range, so this fact limits the mobility of arsenic. However, conditions that reduce
Fe

3+

to F e

2 +

and A s

5 +

to A s

3 +

increase the mobility o f arsenic i n the environment,

b e c a u s e the precipitated ferric hydroxides b e c o m e soluble ferrous hydroxides (Matisoff,


et al. 1 9 8 2 ) . In an oxidizing environment with a pH a b o v e 4 . 0 9 , we will find colloidal
ferric iron hydroxides, which will s o r b arsenic and would thus e x p e c t to have little
arsenic in solution. Under strongly reducing conditions, if b o t h iron and hydrogen sulfide
are present, arsenic sulfide coprecipitates with iron sulfide. Mildly reducing conditions
that lack hydrogen sulfide present conditions under which o n e would e x p e c t to find the
most m o b i l e arsenic, as iron would be in the soluble ferrous state and arsenic would
be in the arsenite form ( H o u n s l o w 1 9 8 0 ) .
Arsenic has b e e n released to the environment through the burning of coal and
the smelting of ores. In the past it was used in the formulation of insecticides and

Inorganic Chemicals in G r o u n d W a t e r

275

10

12

pH
FIGURE 6.9

Eh-pH diagram for arsenic species under standard conditions. Source: A. H. Welch, M. S .

Lico, and J. L. Hughes, Ground Water 2 6 , no.3 (1 9 8 8 ) : 3 3 3 - 4 7 . Used with permission. Copyright 1 9 8 8
by Water W e l l Journal Publishing Co.

e m b a l m i n g c o r p s e s . Starting at the time of the Civil W a r in the United States ( 1 8 6 0 1 8 6 5 ) , arsenic was an ingredient in a popular e m b a l m i n g fluid. As m u c h as to 3 lb of
arsenic could have used p e r c o r p s e . T h e u s e of arsenic in e m b a l m i n g fluids was b a n n e d
by the federal government in 1 9 1 0 b e c a u s e its use interfered with the investigation of
suspected arsenic poisonings. However, graveyards from the Civil W a r and the late
nineteenth century may be a s o u r c e of arsenic contamination ( K o n e f e s 1 9 9 0 ) . It has
s o m e m o d e r n industrial uses. G r o u n d water has b e e n found to have high ( u p to
96 ng/L) concentrations from natural s o u r c e s in northeastern O h i o (Matisoff et al. 1 9 8 2 ) .

276

Chapter S i x

Elevated arsenic ( u p to 5 m g / L ) in ground water in Nova Scotia, Canada, was reportedly


d u e to the weathering of piles of mining waste that contained arsenopyrite (Grantham
and J o n e s 1 9 7 7 ) . In the western United States high ( > 5 0 fig/l) concentrations o f arsenic
are c o m m o n in ground water. T h e s e are associated with areas of sedimentary rocks
derived from volcanic areas, geothermal systems and gold and uranium mining districts.
Irrigation in s o m e areas has liberated arsenic to the extent that concentrations of up to
1 m g / L are found in shallow ground water beneath irrigated fields (Welch, Lico, and
Hughes 1 9 8 8 ) .

6.7.6

Selenium

Selenium o c c u r s in oxidizing solutions as selenite, S e 0


2

selenate, S e 0

2
3

~ , with a +4 valance and as

" , with a + 6 valence. It can be reduced to the insoluble elemental form,

S e . It may also form a precipitate ferroselanite, F e S e , under reducing conditions.


2

Selenate may be s o r b e d o n t o a m o r p h o u s ferric hydroxides. Selenium has a n u m b e r of


industrial uses, such as the manufacture of pigments, stainless steel, and rubber compounds. It is contained in phosphate fertilizers. Selenium has b e e n known to concentrate
in irrigation return water draining from land that has soil high in selenium.

6.7.7

Phosphorus

P h o s p h o r u s can o c c u r in a n u m b e r of valance states, but in natural water it is really


significant only in the + 5 state. Dissolved p h o s p h o r u s in water o c c u r s as phosphoric
acid ( H P 0 ) and its dissociation products, the o r t h o p h o s p h a t e ions: H P 0 ~ , H P 0
3

and P 0

3
4

2
4

~. T h e proportion of each present in an aqueous solution is a function of pH.

Dissolved p h o s p h o r u s is readily s o r b e d o n t o soil and has a very low mobility in ground


water. In alkaline soils it can react with calcium carbonate to form a mineral precipitate,
hydroxyapatite.
3 HP0

2
4

" + 5 CaC0 + 2 H 0
3

C a ( P 0 ) ( O H ) + 5 H C 0 " + OH"
5

Phosphate is released to the environment from mineral fertilizers, animal wastes, sewage,
and detergents.

6.8

Chemistry of Metals
Metals are cations, and most have fairly limited mobility in soil and ground water because
of cation e x c h a n g e or sorption on the surface of mineral grains. They can also form
precipitates of varying solubility under specific Eh pH conditions. Metals are mobile in
ground water if the Eh-pH range is such that soluble ions exist and the soil has a low
cation-exchange capacity ( D o w d y and Volk 1 9 8 3 ) . T h e y can also be m o b i l e if they are
chelated or if they are attached to a m o b i l e colloid. Conditions that p r o m o t e mobility
include an acidic, sandy soil with low organic and clay content. Discharge of a metal in
an acidic solution would k e e p the metal soluble and p r o m o t e mobility.

Inorganic Chemicals in G r o u n d W a t e r

6.8.1

277

Beryllium

Beryllium o c c u r s only in the + 2 valence state. In natural waters we can have B e


+

B e ( O H ) , B e ( O H ) and B e ( O H )
2

2+

. Beryllium o x i d e and hydroxide have low solubilities

and can act as a c o n t r o l on beryllium concentration. At equilibrium with B e ( O H ) , the


2

dissolved form would have an activity of about 1 0 0 /tg/L at a pH of 6 ( H e m 1 9 8 5 ) .


6.8.2

Strontium

Strontium also o c c u r s in the + 2 valence state and has a chemistry similar to that of
calcium. T h e solubility product for strontium sulfate, S r S 0 , i s 1 0 ~

6 , 4

. This suggests that

there might be an equilibrium c o n t r o l on strontium c o n c e n t r a t i o n if sulfate is present


in the water. Strontium c a r b o n a t e , S r C 0 , has a solubility product of 1 0 "
3

1 0

. In general,

strontium is present in ground waters in c o n c e n t r a t i o n s of less than 1 m g / L .


6.8.3

Barium

T h i s alkaline earth e l e m e n t also has a valance of + 2 . Its distribution is controlled by


the solubility o f barite, B a S 0 . Barite has a solubility product o f 1 0 "

1 0

. I f the activity o f

sulfate is 96 mg L ( 1 0 ~ M), then the activity of barium is 1 0 ~ M, or 0 . 0 1 4 mg/L.


6.8.4

Vanadium

T h i s transition metal has oxidation states o f + 3 , + 4 , and + 5 . I n a q u e o u s solutions i t


forms 10 different o x i d e s and hydroxides. Dissolved iron can react with vanadium to
form an insoluble ferrous vanadate, which can act as a c o n t r o l on vanadium in natural
water ( H e m 1 9 7 7 ) .
FeOH
6.8.5

+ 2 H V0 " + H
2

=== F e ( V 0 )
3

+ 3 H 0
2

Chromium

C h r o m i u m in natural waters o c c u r s in a +3 and a +6 valance state. Stable ionic forms


3 +

2 +

i n a q u e o u s systems i n c l u d e C r , C r O H , C r ( O H ) , C r 0
2

2 +
7

and C r 0

2
4

" . Chromous

hydroxide, C r ( O H ) is a p o s s i b l e precipitate under reducing conditions. Figure 6 . 1 0 is


3

an Eh-pH diagram for the stability- field for c h r o m o u s hydroxide. Under s o m e conditions
c h r o m a t e might react with ferrous iron to p r o d u c e a c h r o m o u s hydroxide precipitate
(Robertson 1975).
Cr0

2
4

" + 3 Fe

2 +

+ 8 H 0
2

3 Fe(OH)

+ Cr(OH)

+ 4 H

In general the hexavalent c h r o m i u m in g r o u n d water is s o l u b l e and m o b i l e and


trivalent c h r o m i u m will be insoluble and immobile. Industrial discharges of hexavalent
c h r o m i u m are c o m m o n from metal-plating industries. This material may be quite m o b i l e
in ground water. A hexavalent c h r o m i u m spill on Long Island, New Y o r k , traveled m o r e
than 3 0 0 0 ft from a waste-discharge p o n d to a stream (Perlmutter, Lieber, and Frauenthal
1963). Hexavalent c h r o m i u m from a natural s o u r c e has b e e n found in g r o u n d water in
Paradise Valley, Arizona ( R o b e r t s o n 1 9 7 5 ) .

278

Chapter S i x

6.8.6

Cobalt

Cobalt o c c u r s with valence states of +2 and + 3 . In the Eh and pH range of natural


waters, only the + 2 valence state is stable. It is thought that cobalt can coprecipitate or
be a b s o r b e d by manganese and iron oxides. Cobalt carbonate has a solubility product
1

of 10 " . At a pH of 8.0 with 100 mg/ L of carbonate, the equilibrium solubility of cobalt
i s 6 / j g / L ( H e m , 1 9 8 5 ) . T h e solubility product o f cobalt sulfide i s very low, 1 0

21

Virtually no cobalt would be in solution in a reducing environment. Radioactive cobalt

Inorganic Chemicals in G r o u n d W a t e r

279

is a waste product of certain d e f e n s e activities ( M e a n s , Crerar, and Duguid 1 9 7 8 ) . C o b a l t


o c c u r s in nature as smaltite ( C o A s ) , and c o b a l t i t e ( C o A s S ) .
2

6.8.7

Nickel

This metal o c c u r s in a q u e o u s solutions in the +2 valence state. Nickel o r e s include a


variety of minerals, consisting of nickel, antimony, sulfur, and arsenic: NiSb, NiAs , NiAsS,
2

and NiSbS. Nickel c a r b o n a t e is m o r e s o l u b l e that c o b a l t c a r b o n a t e (K

sp

= 10~ ' ),

w h e r e a s the sulfide has a similar solubility (K

19i

sp

= lQ~ ). Nickel is widely used in

industry.

6.8.8

Molybdenum

Molybdenum o c c u r s a s the o r e mineral molybdenite, M o S . T h e m o s t c o m m o n oxidation


2

states are 4 - 4 and 4 - 6 . U n d e r oxidizing c o n d i t i o n s the M o


p H 1.8 o n e finds H M o 0
2

6 +

state dominates. B e l o w

( a q u e o u s ) . B e t w e e n p H 1.8 and p H 5.3, H M 0 O 4 " o c c u r s ,

w h e r e a s a b o v e pH 5.3 the molybdate ion, M o 0

2
4

" , is stable. If ferrous iron is present,

ferrous molybdate ( F e M o 0 ) p r e s e n t s a p o s s i b l e solubility c o n t r o l , s i n c e this has a


4

solubility p r o d u c t o f 1 0 ~

1 0

4 5

'

a t a p H range o f 5.3 t o 8.5 ( H e m 1 9 7 7 ) . M o l y b d e n u m

may also s o r b o n t o a m o r p h o u s ferric hydroxide ( K a b a c k and Runnels 1 9 8 0 ) . T h e


8

solubility p r o d u c t o f calcium molybdate, C a M o 0 , i s 1 0 ~ ' ( H e m , 1 9 8 5 ) . Molybdenum


4

is used as an alloy in steel and as an additive to lubricants. W a s t e s o u r c e s include mining


and smelting o f ore.

6.8.9

Copper

C o p p e r o c c u r s in either a 4-1 or a 4 - 2 valence state. Dissolved c o p p e r s p e c i e s in water


include C u

2 +

, H C u 0 ~, C u 0
2

2 _
2

, and C u . Cupric c o p p e r and ferrous iron can undergo

an oxidation-reduction:
Cu

2 +

4- 2 F e

2 +

4- 7 H 0
2

C u 0 4- 2 F e ( O H ) 4- 8 H
2

B o t h cupric and c u p r o u s sulfide have very low solubility products. C o p p e r c o n c e n t r a t i o n s


can be very high in acid mine drainage from metal mines, up to several hundred milligrams
p e r liter. C o p p e r can be l e a c h e d from c o p p e r water-supply pipes and fixtures, especially
by waters that have a pH of less than 7 ( H e m , 1 9 8 5 ) .

6.8.10

Silver

Silver, a rare element, is widely used in industry, especially in photography. It o c c u r s in


the 4-1 valence state. Silver chloride, AgCl, has a solubility p r o d u c t of 1 0 ~

9 7

, which

limits the solubility of silver in waters with c h l o r i d e ion. Silver can also be naturally
reduced to the metallic state by ferrous iron:
Ag

4- F e

2 +

4- 3 H 0 , = - F e ( O H )

4- Ag 4- 3 H

Silver sulfide has a low solubility. Thus, in water with chloride, iron, and sulfur present,
stable solid forms of silver o c c u r over the entire Eh-pH field ( H e m , 1 9 7 7 ) . As a result,
there is very little s o l u b l e silver in natural waters.

280

Chapter S i x

6.8.11

Zinc

Zinc is a fairly c o m m o n metal and is extensively used in metallurgy and as a pigment,


zinc oxide, which is often worn on the n o s e s of lifeguards and other people in the sun.
It o c c u r s in the +2 valence state. Zinc c a r b o n a t e has a rather low equilibrium constant,
1 0 ' , which would limit the solubility at pH ranges where the carbonate ion predom
inates. In a pH range of 8 to 11 and with 6 1 0 m g / L of H C 0 ~ , there should be less
than 1 0 0 /tg/L o f dissolved zinc ( H e m 1 9 8 5 ) .
1 0

6.8.12

Cadmium

Cadmium has a very low m a x i m u m contaminant level ( M C L ) in drinking water


10 ng/Ldue to its toxicity. It exists in aqueous solution in the +2 valence state.
Cadmium c a r b o n a t e has a very low solubility product, 1 0 "
. Although this could serve
as a control on solubility under s o m e conditions, cadmium can be mobile in the environment. On Long Island, New York, a metal plating waste containing cadmium and
chromium traveled about 3 0 0 0 ft in a shallow aquifer (Perlmutter, Lieber, and Frauenthal
1 9 6 3 ) . Cadmium has b e e n implicated in an outbreak of a disease in Japan resulting in
a softening of the b o n e s of the victims that resulted in e x t r e m e b o n e pain. T h e cadmium
was traced to rice and soybeans grown in soil contaminated by airborne cadmium that
c a m e from a nearby lead and zinc-smelting operation ( E m m e r s o n 1 9 7 0 ) .
1 3 , 7

6.8.13

Mercury

Mercury has the lowest MCL for any inorganic chemical, 2 / J g / L It is considered to be
very toxic. It has b e e n known to concentrate in the food chain, especially in fish. Several
outbreaks of mercury poisoning have b e e n confirmed in Japan. Local discharges of
mercury from industrial p r o c e s s e s into surface-water b o d i e s resulted in high mercury
levels in fish. Inhabitants of fishing villages ate fish up to three times a day. Mortality of
t h o s e affected was a b o u t 4 0 % , and the poisoning was passed to unborn babies by
apparently healthy mothers (Waldbott 1 9 7 3 ) . Mercury o c c u r s as a metal and in the
valance states +1 and + 2 . Most of the inorganic mercury c o m p o u n d s have a low
solubility. T h e solubility product of H g C l is 1 0 " ' , and for HgS it is about 1 0 " .
Under m o s t natural conditions there is little soluble inorganic mercury. However, meth
ane generating bacteria can convert metallic mercury to organic forms such as methyl
mercury, H g C H . T h e monomethyl mercury ion is soluble in water. Bacteria can also
p r o d u c e dimethyl mercury, H g ( C H ) , which is volatile. Other organic forms of mercury,
such as ethylmercuric chloride ( C H H g C l ) , are manufactured and used as fungicides.
1 7

5 0

6.8.14

Lead
2 +

Lead o c c u r s in aqueous solution as P b and in various hydroxides. Various lead compounds have solubility products that indicate that under the right Eh-pH conditions, lead
solubility would be limited in natural waters: P b C l , K = 1 0 " ; P b F , K = 1 0 " ;
PbSO, K = 1 0 " ; P b C 0 , K = 1 0 "
a n d P b S , K, = 1 0 "
. Lead and the other
metals are cations that can be e x p e c t e d to undergo cation e x c h a n g e with clays. Hence,
the mobility of lead in ground water is limited. This was born out by a study at a storage
battery manufacturing facility at Medley, Florida. Soil near the facility was contaminated
4 8

7 8

sp

1 3

sp

sp

2 7

sp

281

Inorganic Chemicals in G r o u n d W a t e r

with lead in a m o u n t s of up to 9 8 , 6 0 0 mg/kg. However, shallow g r o u n d water immediately


b e n e a t h the c o n t a m i n a t e d soil averaged less than 10 /rg/L of lead, with t h e m a x i m u m
b e i n g 31 /tg/L T h e soils at the site are high in c a r b o n a t e , with s o m e clay; however, the
m e c h a n i s m of lead removal by t h e soils is n o t known. Lead l e a c h e d from lead pipes
and s o l d e r u s e d to join c o p p e r pipes is a potential threat for users of drinking water
that is acidic or p o o r l y buffered ( H e m , 1 9 8 5 ) .

6.9

Radioactive Isotopes
6.9.1

Introduction

Certain i s o t o p e s o f e l e m e n t s u n d e r g o s p o n t a n e o u s decay, resulting i n t h e release o f


energy and e n e r g e t i c particles a n d c o n s e q u e n t formation o f different i s o t o p e s . S o m e o f
t h e s e radioactive i s o t o p e s a r e naturally o c c u r r i n g and o t h e r s a r e c r e a t e d b y t h e b o m
b a r d m e n t o f t h e Earth b y c o s m i c radiation. H u m a n s have c r e a t e d n u c l e a r i s o t o p e s through
t h e d e t o n a t i o n o f n u c l e a r w e a p o n s and the c o n s t r u c t i o n o f n u c l e a r reactors. T a b l e 6.6
lists the s o u r c e s o f environmentally important i s o t o p e s .
Radionuclides emit ionizing r a d i a t i o n a l p h a particles, b e t a particles, and g a m m a
r a y s w h e n they decay. An alpha particle is a helium nucleus with a t o m i c m a s s 4 a n d
a t o m i c n u m b e r 2. A beta particle is either a negative e l e c t r o n or a p o s i t r o n (positive
e l e c t r o n ) . G a m m a radiation c o n s i s t s of e l e c t r o m a g n e t i c radiation similar to X rays but
m o r e e n e r g e t i c (i.e., it has a s h o r t e r wave l e n g t h ) . G a m m a radiation is m o r e destructive
to tissue than X rays. T h e primar,- effect of t h e s e p a n i c l e s is to p r o d u c e ions, h e n c e t h e
n a m e ionizing radiation. Alpha particles do n o t p e n e t r a t e very far i n t o matter d u e to
their large size, but they p r o d u c e a lot of i o n s along their s h o r t path. B e t a particles
p e n e t r a t e to a greater d e p t h but p r o d u c e fewer i o n s p e r unit path length.
Radionuclide c o n c e n t r a t i o n s c a n b e r e p o r t e d i n t e r m s o f their m a s s p e r v o l u m e
c o n c e n t r a t i o n e . g . , milligrams p e r liter. However, they are m o r e frequently r e p o r t e d in
terms of a standard unit of radioactivity, the c u r i e ( C i ) . A curie is 3.7 x 1 0 1 0 disinte
grations p e r s e c o n d . In water we use the p i c o c u r i e ( p C i ) , which is 1 x 1 0 ~ 1 2 Ci, or

T A B L E 6.6

Sources of environmentally important radioactive isotopes.


Radionuclides

Source
Naturally occurring

40

Cosmic irradiation

H , 7 Be,

Nuclear weapons tests

90

Mining wasteuranium, phosphate, coal

222

Industrial wastese.g., nuclear power


plants, weapons manufacturing,
research and medical waste

K,

H,

222

226

Rn,

Sr,

|4

C,

l37

Ra,
22

Cs,

230232

235238

Th,

Na
239 340

Pu

Rn
Ra, Th
' U
59,63 N l 6 0 C O j 90 S( . 93.99^ 9 9 c 107pd_
226

, 2

'l,

237

Np,

,37

Cs,

235 23S

,44

Ce,

,5,

239,2-0>2pU(

Sm,

l52

1M

< Eu,

24l.243 A m

Source: G. W. Gee, Dhanpar Roi, ond R, J. Seme, "Mobility of radionuclides in soil." In D. W. Nelson et al. (ed.}
Chemical Mobility ond Reactivity in Soil Systems, 203. (Madison, Wis.: Soil Science Society of America Spec. Pub).
1 ) , 1983).

282

Chapter S i x

3.7 x 1 0 "

disintegrations per s e c o n d . In the SI system the unit of radioactivity is the

b e c q u e r e l ( B q ) , which is 1 disintegration per s e c o n d .


Radiation d o s e s are measured in terms of r a d s , which are a measure of the
absorption by the body of ionizing radiation of any type. A rad is equivalent to 1 0 0 ergs
of energy from ionizing radiation a b s o r b e d per gram of soft tissue. In the SI system the
unit of d o s e is a g r a y ( G y ) , which is equal to 1 0 0 rads.
T h e effect of ionizing radiation d e p e n d s upon the type of particle and the body
tissue with which it interacts. Therefore, the absolute measurement of d o s e must be
converted to a d o s e e q u i v a l e n t . T h e unit of d o s e equivalent is the r e m . Rads are
converted to rems by multiplying by a factor that d e p e n d s upon the type of ionizing
radiation and its biological effect. F o r e x a m p l e , with g a m m a radiation the factor is 1 and
a rad is equal to a rem. In the SI system the unit of d o s e equivalent is the s e i v e r t (Sv),
and it is equal to a gray times the d o s e factor. A seivert is 100 rem.
6.9.2

A d s o r p t i o n o f Cationic R a d i o n u c l i d e s

T h e cationic radionuclides may be subjected to ion e x c h a n g e and other processes that


s o r b the radionuclide o n t o mineral or organic surfaces in the soil. T h e following transition
metals and lanthanides have large distribution coefficients and h e n c e low mobilities in
waters that are i n the neutral range:
1 4 7

1 5 1

Pm,

1 5 2

Sm,

E u , and

1 5 4

6 0

Co,

S 9

Ni,

6 3

6 5

Ni,

Zn,

9 3

Zr,

1 0 7

P d , "Ag, " C e ,

E u . Many o f them d o not d e s o r b significantly. T h e degree

of sorption is strongly related to the pH of the solution. Insoluble metal hydroxides may
also be formed. T e c h n e t i u m ( T c ) solubility depends strongly upon the Eh of the solution,
b e c a u s e under oxidizing conditions it forms the soluble pertechnetate ion ( T c 0 ~ )
4

( G e e , Rai, and S e m e 1 9 8 3 ) .
9 0

Sr,

1 3 7

C s , and

2 2 6

R a undergo cation e x c h a n g e i n a fashion similar t o other

e x c h a n g e a b l e cations, such as C a

2 +

2+

and M g . T h o r i u m and lead also have high

distribution coefficients and limited mobility in neutral to alkaline soil. Lead is s o r b e d


on hydrous o x i d e s of iron, aluminum, and most likely manganese. Thorium hydroxides
are of very limited solubility ( G e e , Rai, and S e m e 1 9 8 3 ) .

6.9.3

Uranium

Cranium o c c u r s primarily a s
disintegration products o f
isotope.

2 3 S

2 3 8

2 3 8

U , with

U is

U can decay t o form

2 2 6

2 2 3

2 3 5

U being much m o r e rare. O n e o f the


2 0 6

R a . This decay series ends with

Ra.

2 3 8

P b , a stable
9

U has a very long half-life, 4.5 x 1 0 years,

which indicates that it is not very radioactive.


T h e chemistry of dissolved uranium is somewhat c o m p l e x (Giridher and Langmuir
1 9 9 1 ; Langmuir 1 9 7 8 ) . I t has three valence states, + 4 , + 5 , a n d + 6 . Uranium can undergo
oxidation reduction reactions such a s oxidation from the + 4 t o the + 6 state:
U

4 +

+ 2 H 0 === U 0
2

2 +
2

+ 4 H

+ 2 e"

In a system with just uranium and water, stable species include ( 1 ) +4 valance:
4+

U , UOH

3+

, and U ( O H ) ; and ( 2 ) + 6 valance: U 0


4

2+
2

, U0 H , ( U 0 ) ( 0 H )
2

+
5

and

( U 0 ) ( O H ) " . T h e stability fields for these ions and precipitates are shown in Figure
2

6.11 for an aqueous solution with a total uranium activity of 1 0 ~ m o l / L T h e U ( 6 + )

283

Inorganic Chemicals in G r o u n d W a t e r

s p e c i e s has a tendency to form c o m p l e x e s with a wide variety of inorganic anions,


including c a r b o n a t e , hydroxide, phosphate, fluoride, and sulfate. This can be illustrated
6

with an Eh-pH diagram for the s a m e 1 0 " m o l / L solution of U but in c o n t a c t with


c a r b o n dioxide at a partial pressure of 1 0 ~ atm. T h i s is represented by Figure 6.12. In
2

Figure 6.11 die U 0

2
2

formed a series o f c o m p l e x e s with O H " , starting with U 0 O H ,


2

0.40 -

-0.40

FIGURE 6.11

Eh-pH diagram for dissolved species of uranium under standard conditions for the system
6

U 0 H 0 with 1 0 " mol/L U. The stability field for solid U 0


2

and Donald Langmuir, Radiochimica Acta 5 4 : 1 3 3 3 8 , 1 9 9 1 .

(uranite) is shaded. Source: J. Giridhar

284

Chapter S i x

at a pH above about 5.2. With c a r b o n present, as in Figure 6.12, U 0

can form a

series of c a r b o n a t e c o m p l e x e s that replace the hydroxyl c o m p l e x e s .


T h e soluble c o m p l e x e s of oxidized uranium depend upon the pH of the water.
Most natural water contains fluoride, phosphorus, c a r b o n dioxide, and sulfur. Figure
6.13 shows the distribution of uranyl c o m p l e x e s for a ground water under standard
conditions

with

C O ;

= 1 0

atm,

F " = 0.3 mg/L,

C I " = 1 0 mg/L,

S 0

2
4

pH
FIGURE 6.12

Eh-pH diagram for dissolved species of uranium under standard conditions for the system
6

U C 0 0 H 0 with 1 0 ~ mol/L U. The stability field for solid U 0 luranite) is shaded. Source: J.
2

Giridhar and Donald Langmuir, Radiochimica Acta 5 4 : 1 3 3 - 3 8 , 1 9 9 1 .

285

Inorganic Chemicals in G r o u n d W a t e r

U0 (C0 ) - 2

U0 (C0 ) 2

FIGURE 6.13

10

Distribution of uranyl complexes as a function of pH for ground water that contains 0.3
2

mg/L F " , 1 0 mg/L C I " , 100 mg/L S O , " , 0.1 mg/L P 0

2
4

" and P

2 5

C O l

o f 1 0 ~ a t m . Source: Reprinted from

Geochimtca et Cosmochimica Acta, 4 2 , D. Langmuir, "Uranium solution-mineral equilibrium at low temperatures with applications to sedimentary ore deposits." Copyright 1 9 7 8 , Pergamon Press pic.

1 0 0 mg/L, and P 0 " = 0.1 m g / L . In different pH ranges the m o s t prevalent stable species include U 0 F , U 0 ( H P 0 ) " , U 0 ( C 0 )
, and U 0 ( C 0 ) " .
4

2 _

I f reduced s p e c i e s o f iron o r sulfur are present, they c o u l d reduce U ( 6 + ) t o


U ( 4 + ) and precipitate the nearly insoluble mineral uranite, U 0 . This reduction c o u l d
o c c u r b y oxidation o f H S ~ t o S 0 ~ :
2

4 U0 (C0 )
2

4
3

~ 4- HS" 4- 15 H

4 U0

(s) 4- S 0

2
4

" 4- 12 C 0

(g) 4- 8 H 0
2

T h e s a m e reduction could b e a c c o m p l i s h e d b y oxidation o f ferrous iron t o ferric


hydroxide:
U0 (C0 )
2

4
3

" 4- 2 F e

2 +

4- 3 H 0
2

U0

(s) 4- 2 F e ( O H ) 4- 3 C 0
3

B e c a u s e sulfur and iron are c o m m o n in ground-water systems, under reducing


conditions o n e could e x p e c t the formation of uranite, which would remove uranium
from solution. Figures 6.11 and 6.12 s h o w the stability field for uranite ( U 0 ) . An MCL
of 30 pCi/L for uranium has b e e n p r o p o s e d by the EPA (FederalRegister, July 1 8 , 1 9 9 1 ) .
2

6.9.4

Thorium

Thorium is a naturally occurring element with a principal i s o t o p e of


a half-life of 1.39 x 1 0
228-

1 0

2 3 2

T h , which has

years. Daughter products of thorium decay include

2 2 4

R a and

K.i

T h e chemistry of thorium is much simpler than that of uranium. T h o r i u m o c c u r s


only in a 4- 4 valance, so it d o e s not undergo oxidation-reduction. T h o r i u m oxide, T h 0 ,
has a very low solubility. T h e primary thorium o r e is monazite, which contains o x i d e s
of thorium, phosphorus, and the rare earths yttrium, lanthanum, and cerium.
2

286

Chapter S i x

T h e mobility of thorium is greatly e n h a n c e d if ligands are present to form complexes. Figure 6.14 shows the inorganic thorium c o m p l e x e s that form as a function of
the p H o f the solution. T h e solution contains 0.3 m g / L F ~ , 1 0 m g / L C I " , l O O m g L
S 0 " , and 0.1 m g / L P 0 " . I t can b e seen from this figure that the most abundant
a q u e o u s s p e c i e s i n order o f increasing p H are T h ( S O ) , T h F , T h ( H P O ) ,
T h ( H P 0 ) ~ and T h ( O H ) . However, the mobility of thorium c o m p l e x e s formed byorganic ligands such as EDTA and citric acid are much greater than those formed by
inorganic ligands (Langmuir and Herman, 1 9 8 0 ) .
2

2+

Adsorption of dissolved thorium increases with increasing pH above pH 2. T h e


sorption of thorium o n t o clays, o x i d e s and soil organic matter is nearly total by a pH
of 6.5. Strongly c o m p l e x i n g organic ligands such as EDTA can retard sorption or even
p r o m o t e desorption (Langmuir and Herman 1 9 8 0 ) . Thorium in natural waters and soil
should be nearly immobile due to the low solubility of the minerals and the strong
tendency for dissolved forms to be s o r b e d only by clays, mineral oxides, and soil organic
matter.

6.9.5

Radium

Radium o c c u r s naturally i n four isotopes:


into both
2 3 0

2 2 8

R a and

2 2 1

R a , whereas

T h , which in turn decays to

2 2 6

2 3 5

2 2 3

Ra,

2 2 4

Ra,

U decays t o

R a . O n e isotope,

2 2 6

2 2 3

2 2 6

R a and

R a and

2 2 8

2 3 8

Ra.

2 3 2

T h decays

U disintegrates t o

R a , has a much longer half life

than any of the others, 1 5 9 9 years. B e c a u s e of their short half-lives, the radium isotopes
are strongly radioactive ( H e m 1 9 8 5 ) .

287

Inorganic Chemicals in G r o u n d W a t e r

K n o w l e d g e of the a q u e o u s chemistry of radium is sketch}'. It is reportedly similar


in L heniu al beha\ ior to barium (1 iem, 1 9 8 5 ) and calcium (Kathren, 1 9 8 4 ) li is more
s o l u b l e than uranium or thorium and can be b i o c o n c e n t r a t e d by plants (Brazil nuts
have an especially high radium c o n t e n t ) . Radium can be strongly e x c h a n g e d in the cation
e x c h a n g e series. According to Kathren ( 1 9 8 4 ) . the cation e x c h a n g e s e q u e n c e for soils
is
Sr

2 +

2 2 8

< Ra

2 +

< Ca

2 +

< Mg

2 +

< Cs

2 +

< Rb

2 +

R a has a m u c h shorter half life, ( 5 . 8 y r ) than

is m o r e abundant in nature than

2 3 8

U , the parent of

< K

< NH

+
4

< Na

< Li

2 2 6

2 3 2

R a . However, its parent,

2 2 6

R a . As a result, b o t h i s o t o p e s

are found in g r o u n d water. T h e U.S. EPA has p r o p o s e d MCLs of 20 pCi/L for b o t h


and

2 2 8

Th,

2 2 6

Ra

R a (Federal Register, July 18, 1 9 9 1 ) . Wells with high radium levels in g r o u n d

water have b e e n discovered to be c o n c e n t r a t e d in two areas of the United States: the


Piedmont and coastal plain of the Middle Atlantic states and the u p p e r Midwestern states
of Minnesota, Iowa, Illinois, Missouri, and W i s c o n s i n ( H e s s et al. 1 9 8 5 ) . T a b l e 6.7 summarizes the distribution o f

2 2 6

R a and

2 2 8

R a i n the Atlantic coastal plain and Piedmont

region.
T h e radium c o n t e n t of g r o u n d water is a function of the rock type of the aquifer.
Igneous rocks, such as granites, contain the highest p r o p o r t i o n of uranium and thorium,
the parent i s o t o p e s of radium. Granitic rock aquifers and sands and s a n d s t o n e s derived
from the weathering of granites have the potential to have high radium. P h o s p h a t e rock
is also very high in uranium. Radium is not only a p r o b l e m that is naturally occurring,
but there are localized areas of radium contamination from industrial operations. T h e s e
are associated with uranium mill tailings as well as facilities w h e r e radioluminescent
paints were prepared and used. F o r e x a m p l e , from W o r l d W a r I up until 1 9 6 8 wrist
watches with radium dials that glowed in the dark were sold in the United States.

6.9.6

Radon

T h e r e are several i s o t o p e s of radon, but

2 2 2

R n is the only o n e that is important envi-

ronmentally T h e o t h e r i s o t o p e s have half-lives of less than 1 min. T h e half-life of

T A B L E 6.7

Distribution of

2 2 6 6

R a and

? 2 8

2 2 2

Rn

R a b> aquifer type in the Atlantic coastal plain and Piedmont

provinces.
R a - 2 2 8 (pCi/L)
Aquifer Type
Igneous (acidic)
Metamorphic

Number of

Geometric

Samples

R a - 2 2 6 (pCi/L)
Geometric

Mean

Range

Mean

Range

42

1.39

0.0-22.6

1.8

0.0-15.9

75

0.33

0.0-3.9

0.37

0.0-7.4

143

1.05

0.0-17.6

1.36

0.0-25.9

Arkose

92

2.16

0.0-13.5

2.19

0.0-23.0

Quartzose

50

0.27

0.0-17.6

0.55

0.0-25.9

16

0.06

0.0-0.2

0.12

0.0-0.3

Sand

Limestone

Source: C. T. Hess, J. Michel, T. R. Horton, H. M. Prichord, and W. A. Coniglio, "The occurrence of radioactivity in public water supplies
in the United States," Heaffh Physics 48 ( 1 9 8 5 ) : 5 5 3 - 8 6 .

288

Chapter S i x

2 2 2

2 2 6

is 3.8 days.
R n is produced by the decay of R a , so that it is associated with rocks
that are high in uranium. Radon can be associated with water that is low in dissolved
R a , b e c a u s e it c o m e s primarily from the decay of the radium in the rock. Radon is
a n o b l e gas and d o e s not undergo any chemical reactions, n o r is it s o r b e d o n t o mineral
matter. Radon is lost from water by diffusion into the atmosphere and by radioactive
decay through a series of short-lived daughter products to
P b , which has a half life
o f 21.8 yr.
2 2 6

2 1 0

T h e EPA has p r o p o s e d an MCL standard of 3 0 0 pCi/L for radon in drinking water


(Federal Register, July 18, 1 9 9 1 ) . However, there is also a health c o n c e r n for excessive
radon accumulation in h o m e s . Radon can enter h o m e s through emanations from the
soil as well as by diffusion from tap water with a high radon content.
O w n e r s of private water systems are m o s t at risk from radon in drinking water.
Public water-supply systems normally have storage facilities to supply water during fires.
T h e residence time for the water in these facilities allows the radon to both diffuse and
decay. Private water systems rely upon wells and usually only have a very small storage
facility used to maintain pressure.
Brutsaert et al. ( 1 9 8 1 ) studied radon in ground water in Maine. They found radon
levels in private wells of up to 1 2 2 , 0 0 0 pCi/L. T h e average
R n content of wells obtaining water from granites was 2 2 , 1 0 0 pCi/L, from sillimanite-grade metasedimentary
rocks, it was 1 3 , 6 0 0 pCi/L, and from chlorite-grade metasedimentary rocks, it was
1 1 0 0 pCi/L. T h e high radon values in the high-grade metamorphic terrains were believed
to be due to metamorphic pegmatites and associated uranium mineralization.
2 2 2

6.9.7

Tritium
3

Tritium. H , is produced naturally by cosmic-ray b o m b a r d m e n t of the atmosphere, by


thermonuclear detonations and in nuclear reactors. As it is an isotope of hydrogen, it
can form a water m o l e c u l e or be incorporated into living tissue. Tritium has a half life
of 12.6 yr. Much of the radiation at low-level radioactive waste sites is due to tritium
(Kathren 1 9 8 4 ) . Low-level radioactive waste is disposed by shallow burial on land. If
waste packages at low-level radioactive waste sites leak and if the landfill is not secure,
tritium is likely to escape. Tritium migration via ground water flow from low level waste
disposal has b e e n detected at sites at the Savannah River facility, Los Alamos National
Laboratory ( O v e r c a m p 1 9 8 2 ) , and the Sheffield, Illinois, commercial disposal site (Foster,
Erickson, and Healy 1 9 8 4 ) , and the Hanford, Washington, site (Levi, 1 9 9 2 ) .

6.10

Geochemical Zonation
Landfills have proven to be a s o u r c e of ground water contamination in many different
geologic terrains, climates, and hydrogeologic settings. Landfill l e a c h a t e is the liquid
that is the product of the liquid content of the waste, infiltrating precipitation, and ground
water if the waste is b e l o w the water table. T h e s e liquids mix with the waste and dissolve
both inorganic and organic constituents. Leachate is a c o m p l e x mixture of dissolved
and colloidal organic matter and inorganic c o m p o u n d s and ions. Many of the chemical

Inorganic Chemicals in G r o u n d W a t e r

289

p r o c e s s e s are controlled by microbial activity. In this section we will e x a m i n e the geochemical zonation that o c c u r s in ground water affected by leachate from a landfill. T h i s
will serve to illustrate s o m e of the reactions discussed in this chapter.
Organic matter is a major constituent of municipal landfills, w h e r e it is d e c o m p o s e d
by bacteria and o t h e r m i c r o b e s . T h e initial d e c o m p o s i t i o n is under a e r o b i c conditions.
O n c e the oxygen in the landfill is c o n s u m e d , a n a e r o b i c d e c o m p o s i t i o n b e c o m e s prevalent. T h e c o m p l e t e a e r o b i c d e c o m p o s i t i o n of glucose, a sugar found in organic waste,
yields c a r b o n d i o x i d e and water as products ( B a e d e c k e r and B a c k 1 9 7 9 a ) :
C H
6

1 2

4-6 0

6 C0

4- 6 H 0
2

T h e c a r b o n d i o x i d e thus p r o d u c e d forms c a r b o n i c acid in the leachate and can also


e s c a p e from the waste as a gas.
Anaerobic d e c o m p o s i t i o n p r o d u c e s a variety of organic acids as intermediate steps
i n the formation o f methane. T h i s p r o c e s s c a n b e represented b y the fermentation o f
glucose:
2 C H 3 C O C O O H (pyruvic acid) 4- 2 H
2

CH3COOH

HCOOH
C0

+ 4 H

C H 3 C O O H (acetic a c i d ) + H C O O H (formic a c i d )

CHjCOCOOH + H 0

CH

C0

C0

+ H

C H 4- 2 H 0

Any hydrogen gas p r o d u c e d is utilized by the methane-forming bacteria in their reduction


of organic c o m p o u n d s In addition, oxidized forms of nitrogen and sulfur c o m p o u n d s
are also reduced t o form N H

and H S ( B a e d e c k e r and B a c k 1 9 7 9 b ) . T h e m e t h a n e
2

formed can e s c a p e from the landfill as a gas and can also dissolve in the leachate.
T h e pH of landfill leachate is generally in the range of 6.5 to 7.0. T h i s is due to
the buffering offered b y the generation o f large amounts o f C 0 , which dissociates t o
2

H C O 3 " , to a lesser e x t e n t the formation o f N H , which forms N H


3

reduction o f S 0

2
4

+
4

in water, and the

~ t o H S . A s has b e e n s h o w n earlier i n this chapter, the solubility o f


2

mineral s p e c i e s is a function of the Eh and pH of the environment. T h e Eh of landfill


leachate is controlled by the reduction of organic c o m p o u n d s and oxidized forms of
nitrogen, sulfur, iron, and manganese. W h e n reduced leachate m i x e s with oxygenated
ground water, the Eh of the resulting solution can b e c o m e m o r e oxidizing.
B a e d e c k e r and B a c k ( 1 9 7 9 b ) have c o m p a r e d the r e d o x zonation of a landfill with
that of marine sediments, a g e o c h e m i c a l environment that has b e e n extensively studied.
T h e oxygenated water at the b o t t o m of the sea represents an a e r o b i c environment. T h e
sediments at the b o t t o m are a n a e r o b i c . In the transition from a e r o b i c to a n a e r o b i c
conditions, there is a sulfate-reducing zone, w h e r e d e c o m p o s i t i o n of organic matter
reduces sulfate to sulfide. Ferric iron will also be reduced to ferrous iron in this z o n e .
Ferrous iron reacts with the hydrogen sulfide form insoluble sulfide precipitates. T h e
most reduced z o n e in marine sediments is an a n a e r o b i c c a r b o n a t e reducing z o n e that
+

has b e c o m e depleted in sulfur. In this z o n e we find the production of C H and N H .


4

In the study of leachate plumes at landfills, a three-part zonation has b e e n found.


T h e landfill itself represents an a n a e r o b i c zone. T h i s a n a e r o b i c z o n e e x t e n d s along the

290

Chapter S i x

plume of leachate mixing with ground water. In this z o n e we find the production of
m e t h a n e and ammonia as microorganisms d e c o m p o s e organic matter and obtain oxygen
from the reduction of sulfate and nitrate. Reduction will liberate soluble ferrous iron.
Manganese may also b e c o m e soluble due to dissolution of native minerals. Ferrous
sulfide may precipitate in this zone.
As the leachate mixes with oxygenated ground water, it b e c o m e s less reducing
and forms a transition zone. Most of the soluble organic matter has already b e e n
d e c o m p o s e d in this zone. In the transition z o n e there is coprecipitation of trace metals
with iron and manganese hydroxides. Ahead of the transition zone, there is an aerobic
zone, w h e r e the leading e d g e of the leachate has changed the native ground water quality,
but not enough to deplete the oxygen. T h e a e r o b i c z o n e contains nitrate and sulfate at
the b a c k g r o u n d levels of the aquifer.
B a e d e c k e r and B a c k ( 1 9 7 9 a , 1 9 7 9 b ) have defined the boundaries o f these z o n e s
at a landfill on the basis of ( 1 ) dissolved oxygen content, ( 2 ) the ratio of reduced nitrogen
t o nitrate, ( 3 ) the p r e s e n c e o f methane gas, and ( 4 ) the ratio o f dissolved manganese
to iron.
T o t a l r e d u c e d n i t r o g e n , the sum o f organic nitrogen and ammonia, i s measured
by a test called Kjeldahl nitrogen (Kjl N). B a e d e c k e r and B a c k ( 1 9 7 9 a , 1 9 7 9 b ) examined
the ratio Kjl N n o ,

and lound that it was greatest in the anaerobic zone where arnmi inia

was present, d e c r e a s e d in the transition zone, and was quite low in the a e r o b i c zone,
w h e r e the nitrogen was primarily in the form of nitrate.
If there is limited sulfur present, there will be soluble iron and manganese in the
a n a e r o b i c zone. As the iron traverses to the transition zone, it will begin to be oxidized
and precipitate as ferric hydroxide. Manganese is soluble over a much larger Eh pH
range than iron and will remain in solution longer as the plume moves into oxygenated
ground water. B e c a u s e iron is much m o r e abundant than manganese, there will be a
large ratio of dissolved iron to dissolved manganese in the anaerobic zone. This ratio
will gradually decrease in t h e transition z o n e as the iron is preferentially precipitated.
Eventually both dissolved iron and manganese will disappear as the aerated z o n e is
reached.
Methane gas and ammonia are found only in the anaerobic z o n e and dissolved
oxygen is found only in the a e r o b i c zone. Naturally, the preceding model of geochemical
zonation is valid only for circumstances w h e r e the receiving ground water contains
dissolved oxygen. Figure 6.15 s h o w s an idealized leachate plume and the geochemical
zonation.
If landfill leachate is discharged into a c a r b o n a t e rock aquifer, or an unconsolidated
aquifer with large amounts of carbonate, reactions between the acids of the leachate
and the c a r b o n a t e can o c c u r ( K e h e w and Passero 1 9 9 0 ) . B o t h organic acids produced
by the reduction of organic c o m p o u n d s and c a r b o n i c acid from the production of
c a r b o n d i o x i d e are present in leachate. It will react with the c a r b o n a t e rock, releasing
calcium and magnesium:
CaC0
C0

2
3

~ + FI

CaMg(C0 )
3

2C0

2
3

"

+2FI

Ca

2 +

+ C0

2
3

"

HC0 "
3

Ca *

+Mg

2 HC0 "
3

2 +

+ 2 C 0

23

292

Chapter Six

T h e s e reactions liberate calcium and magnesium, and the resulting leachate has a
high hardness. T h e increase in b i c a r b o n a t e buffers the leachate to nearly neutral and
also p r o m o t e s the precipitation o f c a r b o n a t e minerals such a s siderite ( F e C 0 ) .
3

6.11

Summary
T h e inorganic chemistry of contaminated ground water can be studied from the stand
point of chemical thermodynamics. T h e first step is to obtain a representative chemical
analysis of the water. T h i s should contain the major anions and cations: C a
2

Na , K , H C 0 " , CI", P O 4 " , S 0


+

2
4

2 +

, Mg

2 +

" , N 0 ~ , and F " . In addition, pH and Eh should


3

be measured in the field in a flowthrough cell so that the sample d o e s not c o m e into
c o n t a c t with the atmosphere. If the Eh is positive, dissolved oxygen should be measured;
if it is negative, hydrogen sulfide and a m m o n i a should be measured. T h e s e data should
be o b t a i n e d to aid in interpretation of s p e c i e s analyses for suspected inorganic contam
inants.
Inorganic contaminants in g r o u n d water can be removed by precipitation. T h e law
of mass action can be used to study chemical equilibrium. Thermoch/narnically stable
s p e c i e s may precipitate as the inorganic contaminant m i x e s with n a m e ground water.
Many inorganic e l e m e n t s can exist in different valence states. T h e elements can
participate in oxidation reduction reactions. S o m e of the redox reactions are mediated
by microorganisms. T h e Nernst equation relates the Eh and pH of an aqueous solution.
T h e Nernst equation and chemical equilibrium can be used to construct Eh pH diagrams,
which s h o w the fields of stability for various chemical spills.
Organic matter can be d e c o m p o s e d by m i c r o b e s in a landfill. Under oxidizing
conditions o x y g e n is the e l e c t r o n acceptor, whereas under reducing conditions the
e l e c t r o n a c c e p t o r can be sulfate, nitrate, ferric iron, or c a r b o n dioxide. T h r e e redox
z o n e s have b e e n identified in the leachate plumes at landfills: an a n a e r o b i c zone, where
the waste is being d e c o m p o s e d , m e t h a n e is being formed, and iron, sulfur, and nitrogen
exist in reduced forms; a transition zone; and an a n a e r o b i c zone, with oxidizing conditions
and oxidized forms of iron, sulfur, nitrogen, and c a r b o n .

Chapter Notation
A

Constant equal to 0 . 5 0 8 5 at 25C

[A
a

Activity of reactant A

Constant equal to 0 . 3 2 8 1 at 25C

[B]

Activity of product B

Eh

Oxidation potential of the a q u e o u s solution in volts

Moles o f reactant
Effective diameter of an ion from ( T a b l e 6 . 1 )

Moles o f product
Standard potential of r e d o x reaction in volts

Faraday constant

Ionic strength

293

Inorganic Chemicals in G r o u n d W a t e r

Equilibrium constant

Ksp

Solubility product

Moles o f hydrogen ions

Molality of the /th ion

Molal c o n c e n t r a t i o n

ii

N u m b e r of electrons in half-reaction

R
T

Gas constant

T e m p e r a t u r e in Kelvins

Moles o f water

Charge on ionic s p e c i e s /'

AG

Free energy of a reaction

7i

Activity coefficient for ionic s p e c i e s i


Activity coefficient

y*

References
Baedecker, Mary J o , and William Back. 1979a. Hydrogeological processes and chemical reactions at a landfill.
Ground Water 17, no. 5 : 4 2 9 - 3 7 .
. 1979b. Modem marine sediments as a natural analog
to the chemically stressed environment of a landfill.
Journal of Hydrology 4 3 : 3 9 3 - 4 1 4 .
BruLsaert, W. F., S. A. Norton, C. T. Hess, and J. S. Williams.
1981. Geologic and hydrologic factors controlling ra
don 2 2 2 in ground water in Maine. Ground Water 19,
no. 4 : 4 0 7 - 1 7 .
Corbett, R. G., and B. M. Manner. 1984. Fluoride in the ground
water of Northeastern Ohio. Ground Water 22, no.
1:13-17.
Cross, F. L, and R. W. Ross. 1970. Fluoride uptake from
gypsum ponds. Fluoride 3 : 9 7 - 1 0 1 .
Dowdy, R. H., and V. V, Volk. 1983. Movement of heavy metals
in soil. In Qiemical Mobility and Reactivity in Soil
Systems, 2 2 9 - 3 9 . Madison, Wis.: Soil Science Society
o f America.
Emmerson, B. T. 1970. "Ouch-ouch" disease: T h e osteomalacia of cadmium nephropathy. Annals of Internal
Medicine 7 3 ; 8 5 4 .
Feth, J. H. 1966. Nitrogen compounds in natural watera
review. Water Resources Research 2, no. 1 : 4 1 - 5 8 .
Flipse, W. J . , Jr., B. G. Katz, J. B. Linder, and R. Markel. 1984.
Sources of nitrate in ground water in a sewered housing development, central Long Island, New York.
Ground Water 22, no. 4 : 4 1 8 - 2 5 .
Flipse, W. J . , Jr., and F. T. Bonner. 1985. "Nitrogen-isotope
ratios of nitrate in ground water under fertilized fields,
Long Island, New York. Ground Water 2 3 , no.
1:59-67.
Foster, J. B . , J. R. Erickson, and R. W. Healy. 1984. Hydro
geology of a low level radioactive-waste disposal site
near Sheffield, Illinois. U. S. Geological Survey, Water
Resources investigations Report 8 3 - 4 1 2 5 . 87 pp.
Gee, G. W., Dhanpat Rai, and R. J. S e m e . 1983. "Mobility of
radionuclides in soil." In Chemical Mobility and Reac-

tivity in Soil Systems, 2 0 3 - 2 7 . Madison, Wis.: Soil Scie n c e Society of America.


0

Giridhar, J . , and D. Langmuir. 1991- Determination of E


for the U 0
/U
couple from measurement o f
the equilibrium: U 0
+ Cu(s) + 4 H = U
+
Cu
+ 2 H 0 at 25C and s o m e geochemical implications. Radiochimica Acta 54: 1 3 3 - 3 8 .
2 +

4 +

4 +

2 +

Grantham, D. A., and J. F. J o n e s . 1977. Arsenic contamination


of water wells in Nova Scotia. JournalAmerican Water
Works Association 6 9 : 6 5 3 - 5 7 .
Gray, E. M., and M. Morgan-Jones. 1980. A comparative study
of nitrate levels at three adjacent ground-water sources
in a chalk catchment area west of London. Ground
Water 18, no. 2 : 1 5 9 - 6 7 .
Hem, J o h n D. 1977. Reactions of metal ions at surfaces of
hydrous iron oxide. Geochimica et Cosmochimica
Acta41, no. 4 : 5 2 7 - 3 8 .
. 1985. Study and interpretation of the chemical characteristics of natural water. U. S. Geological Survey
Water Supply Paper 2 2 5 4 , 2 6 3 ppHess, C. T., J. Michel, T. R. Horton, H. M. Prichard, and
W. A. Coniglio. 1985. T h e occurrence of radioactivity
in public water supplies in the United States. Health
Physics 4 8 : 5 5 3 - 8 6 .
Hill, A. R. 1982. Nitrate distribution in the ground water of
theAlliston region of Ontario, Canada. Groundwater
20, no. 6 : 6 9 6 - 7 0 2 .
Hounslow, A. W. 1980. Ground-water geochemistry: Arsenic
in landfills. Ground Water 18, no. 4:331 - 3 3 .
Kaback, D. S., and D. D. Runnels. 1980. Geochemistry of
molybdenum in s o m e stream sediments and waters.
GeocJjimica et Cosmochimica Acta 4 4 : 4 4 7 - 5 6 .
Kathren. R. L 1984. Radioactivity in the environment. Chur,
Switzerland: Harwood Academic Publishers, 397 pp.
Kehew, A. E., and R. N. Passero. 1990. pH and redox buffering
mechanisms in a glacial drift aquifer contaminated by
landfill leachate. Ground Water 28, no. 5 : 7 2 8 - 3 7 .

294

Chapter S i x

Konefes, John. 1990. Cited in Ground Water in ihe News.


Ground Water 28, no. 6:997.
Langmuir, Donald. 1978. Uranium solution-mineral equilibria
at low temperatures with applications to sedimentan
ore deposits. Geocbimica et Cosmocbimica Acta 42,
no. 6 : 5 4 7 - 7 0 .
Langmuir, Donald, and J. S. Herman. 1980. T h e mobility of
thorium in natural waters at low temperatures. Geo
chimica et Cosmocbimica Acta 4 4 : 1 7 5 3 - 6 6 .
Levi, B. G. 1992. Hanford seeks short and long-term solutions
to its legacy of waste. Physics Today 45: no. 3 (March):
17-21.
Man ah an, S. E. 1984. Environmental chemistry, 4th ed. Boston: PWS Publishers, 6 1 2 pp.
Malison", Gerald, C. J. Khourey, J. F. Hall, A. W. Varnes, and
W. H. Strain. 1982. T h e nature and source of arsenic
in Northeastern Ohio ground water. Ground Water
20, no. 4 : 4 4 6 - 5 6 .
McCarty, P. L, B. E. Rittman, and E. J. Bouwer. 1984. "Microbiological Processes Affecting Chemical Transformations in Groundwater." In Groundwater pollution
microbiology. Edited by Gabriel Bitton and C. P.
Gerba, 8 9 - 1 1 5 . New York: J o h n Wiley and Sons.
Means, J. L, D. A. Crerar, and J. O. Duguid. 1978. Migration
of radioactive wastes: radionuclide mobilization by
c o m p l e t i n g agents. Science 2 0 0 : 1 4 7 7 - 8 1 .
Overcamp, T . J . 1982. "Low-level radioactive waste disposal
by shallow land burial." In Handbook of environ
mental radiation. Edited by A W. Klement, Jr., 2 0 7 67. B o c a Raton, Fia.: CRC Press.
Perlmutter, N. M., Maxim Lieber, and H. L Erauenthal. 1963Movement of waterbome cadmium and bexavalenl
chromium wastes in South Earmingdale, Nassau
County, Long island. U. S. Geological Survey Professional Paper 475C, pp. C 1 7 0 - C 1 8 4 .

Robertson, F. N. 1975. Hexavalent chromium in the ground


water in Paradise Valley, Arizona. Ground Water 13,
no. 6 : 5 1 6 - 2 7 .
Robie, R. A., B, S. Hemingway, and J. R. Fisher. 1978. Thermodynamic properties of minerals and related sub
stances at 298.15 K and \ bar( 1 0 pascals) pressure
and at higher temperatures U. S. Geological Survey
Bulletin 1-452, 456 pp.
Silver, B. A., andj. R. Fielden. 1980. Distribution and probablesource of nitrate in ground water of Paradise Valley,
Arizona. Ground Water 18, no. 3 : 2 4 4 - 5 1 .
Wagman, D. D., W. H. Evans, V. B. Parker, I. Halow, S. M.
Bailey, and R. H. Schumm. 1968, Selected values of
chemical thermodynamic properties (tables for ele
ments 1 -3-t in standard order of arrangement).
National Bureau of Standards Technical Note 270-3,
26-i pp.
. 1969. Selected values of chemical thermodynamic
properties (tables for elements 3 5 - 5 3 in standard
order of arrangement). National Bureau of Standards
Technical Note 270-4, 141 pp.
Wagman, D. D., W. H. Evans, V. B. Parker, I. Halow, S. M.
Bailey, R. H. Schumm, and K. L Chumey. 1971. Selected values of chemical thermodynamic properties
(tables for elements 5 4 - 6 1 in standard order of ar
rangementI National Bureau of Standards Technical
Note 270-5, 41 pp.
Waldbott, G. L. 1973. Health Effects of Environmental Pol
lutants, St. Louis, Mo.: C. V. Mosby Co., 3 1 6 pp.
Welch, A. H., M. S. Lico, and J. L Hughes. 1988. Arsenic in
ground water of the western United States. Ground
Water 26, no. 3 : 3 3 3 - 4 7 .
Whittemore. D. O. and Donald Langmuir, 1975. T h e solubility
of ferric oxyhydroxides in natural waters. Ground
Water 13, no. 4 : 3 6 0 - 6 5 .
5

Chapter Seven

Organic Compounds in Ground Water

7.1

Introduction
Organic c o m p o u n d s can o c c u r in the g r o u n d either p u r e as c o m p o u n d s , a mixture of
c o m p o u n d s , or dissolved in water. In this c h a p t e r we will first e x a m i n e the physical
properties of organic c o m p o u n d s and the way that physical properties affect the behavior
of organic chemicals in the subsurface. We will then learn the structure of c o m m o n
organic c o m p o u n d s , their n o m e n c l a t u r e , and their p o s s i b l e s o u r c e s in the subsurface.
Finally, we will s e e h o w organic chemicals can undergo transformations by b o t h chemical
reactions and microbial degradation.

7.2

Physical Properties of Organic Compounds


T h e r e are several important physical properties of organic c o m p o u n d s that help us to
understand h o w they will behave. T h e s e c o m p o u n d s can e x i s t as gases, liquids, and
vapors. W e c a n use the m e l t i n g p o i n t s and b o i l i n g p o i n t s t o evaluate i f a particular
c o m p o u n d will be a gas, liquid, or vapor at a certain temperature. If the specified
temperature is b e l o w the melting point, the c o m p o u n d will be a solid. If the temperature
falls b e t w e e n the melting point and the boiling point, the c o m p o u n d will be a liquid,
and if the temperature is a b o v e the boiling point, the c o m p o u n d will be a gas. Boiling
points are usually given at a pressure of ^60 mm mercury' (1 a t m ) . Boiling points for a
h o m o l o g o u s series of c o m p o u n d s will increase with increasing molecular weight. (A
h o m o l o g o u s s e r i e s i s o n e with the s a m e basic structure but increasing n u m b e r s o f
carbon atoms.)
T h e s p e c i f i c g r a v i t y o f a s u b s t a n c e (liquid o r s o l i d ) i s the ratio o f the weight
o f a given v o l u m e o f that s u b s t a n c e t o the weight o f the s a m e v o l u m e o f water. T h e
water weight is usually m e a s u r e d at 4 C , w h e r e a s the organic liquid weight may be
measured at s o m e o t h e r temperature, which is typically 20C. If the specific gravity of
the pure s u b s t a n c e is less than 1.0, the s u b s t a n c e will float on water, w h e r e a s if the
specific gravity is greater than 1.0, the s u b s t a n c e will sink in water.
W a t e r s o l u b i l i t y is an important property of organic s u b s t a n c e s . F o r a gas this
must be m e a s u r e d at a given vapor pressure. F o r a liquid it is a function of the temperature
of the water and the nature of the substance. Solubilities of organic materials can range
295

296

Chapter Seven

from completely miscible with water to nearly insoluble. More soluble materials have a
greater potential mobility in the environment. Laboratory solubility is measured using
distilled water, which may not have the same effect as natural waters.
T h e o c t a n o l - w a t e r p a r t i t i o n c o e f f i c i e n t is a measure of the degree to which
an organic s u b s t a n c e will preferentially dissolve in water or an organic solvent. T h e
s u b s t a n c e is m i x e d with equal amounts of two immiscible fluids, water and octanol (an
eight-carbon chain a l c o h o l ) . T h e coefficient is the ratio of the equilibrium concentration
of the s u b s t a n c e in octanol to the equilibrium concentration in water:

^, = 7

(7.1)

C'waler

This is usually given as a logarithm. T h e greater the value, the greater the tendency
to dissolve in the organic liquid rather than the water. T h e greater the octanol-watei
partition coefficient, the less m o b i l e the c o m p o u n d tends to be in the environment.
V a p o r p r e s s u r e is a measure of the tendency' of a substance to pass from a solid
or a liquid to a vapor state. It is the pressure of the gas in equilibrium with the liquid
or the solid at a given temperature. T h e greater the vapor pressure, the m o r e volatile
the substance.
T h e v a p o r d e n s i t y of a gas indicates if it will rise or sink in the atmosphere. If
the gas is lighter than air. it will rise; if it is denser than air, it will sink. T h e vapor density.
V is related to the equilibrium vapor pressure, the gram molecular weight of the gas,
it

and the temperature by Equation 7.2.


PM
V =

(7.2)

where
P = equilibrium vapor pressures in atms
M = gram molecular weight
R = gas constant ( 0 . 0 8 2 L - a t m s / m o l / K )
T = temperature in K
H e n r y ' s l a w states that there is a linear relationship between the partial pressure
of a gas above a liquid and the m o l e fraction of the gas dissolved in the liquid. It is
given as Equation 7.3.

Hl =

(73)

where
P = partial pressure of gas (atms at a given temperature)
x

C = equilibrium concentration of the gas in solution ( m o l ' m water)


x

H = Henry's law constant in a t m / m o l / m water)


L

Henry's law is valid if the gas is sparingly soluble, the gas phase is reasonably ideal,
and the gas will not react with the solute. It can also be applied to organic c o m p o u n d s

O r g a n i c Compounds i n G r o u n d W a t e r

297

that are volatile liquids w h e n they are dissolved in water. T h e greater the Henry's law
constant, the greater the rate of volatilization from soil or water.

7.3

Organic Structure and Nomenclature


7.3.1

Hydrocarbon Classes

Organic chemistry is b a s e d on the behavior of c a r b o n atoms. T h e m o s t simple organicc o m p o u n d s are h y d r o c a r b o n s , which consist solely o f c a r b o n and hydrogen. C a r b o n
has four b o n d i n g locations and can b o n d with such e l e m e n t s as oxygen, nitrogen, sulfur,
p h o s p h o r u s , chlorine, b r o m i n e , and fluorine as well as hydrogen. C a r b o n can form
single, d o u b l e , and triple b o n d s .
H y d r o c a r b o n s can b e divided into two classes, a r o m a t i c h y d r o c a r b o n s , which
contain a b e n z e n e ring, and a l i p h a t i c h y d r o c a r b o n s , which don't contain a b e n z e n e
ring. T h e b e n z e n e ring c o n t a i n s six c a r b o n atoms j o i n e d in a ring structure with alternating
single and d o u b l e b o n d s . T h e single and d o u b l e b o n d s c h a n g e positions o n the r i n g
so that they are c o n s i d e r e d to be equal. A berrzene ring is r e p r e s e n t e d by the s y m b o l
in Figure 7 . 1 ( a ) .
Aliphatic hydrocarbons

T h e c a r b o n s o f aliphatic h y d r o c a r b o n s with m o r e than o n e c a r b o n atom can b e j o i n e d


b y single b o n d s ( a l k a n e s ) , d o u b l e b o n d s ( a l k e n e s ) , o r triple b o n d s ( a l k y n e s ) .
If there is a c o m b i n a t i o n of single and multiple b o n d s , the c o m p o u n d is classified on
the basis of the multiple b o n d .
Alkanes

Alkanes are also k n o w n a s s a t u r a t e d h y d r o c a r b o n s , o r p a r a f f i n s . T h e general


formula is CH + 2- T h e first six straight-chain alkanes are as follows:
2

I
Methane

CH.

H CH

I
H
H

I
Ethane

CH,CH,

-C-

I
H

I
H C C -C H

I
II

298

Chapter Seven

Butane

II

II

I lexane

1
1

1
1

1
H

1
H

11
H

II

1
1

1
1
H CC
1
11
1
H
H

CHXHXHXHXH,

Pentane

1
1
1
H C C C

CH CH CHXH,
3

II

II
1

1
C

1
H

11

CHXHXHXHXHXH,

X
H

C C H

II

Organic c o m p o u n d s with the s a m e formula may have different structural relationships. T h i s i s b e c a u s e alkanes can also have b r a n c h e d c h a i n s . T h e s e c o m p o u n d s are
s t r u c t u r a l i s o m e r s , and although they have the s a m e formula, they are different
c o m p o u n d s with different properties. F o r e x a m p l e the formula C H represents three
isomers:
5

l 2

Pentane

CHXHXHXHXHj

II

II

11

cccc

H C-

I
H

I
11

2-Methylbutane
11

HC H
H
I H

CH,

xccCHXCHXH,
H

C H

II

II

II

II

2,2-Dimethylpropane
H

I
CH,

HCH

I
i II ( i II

CH,

H
II

H C - -C

I
II
H-

C-

I
H
C H

I
H

299

O r g a n i c Compounds i n G r o u n d W a t e r

Note that there are two ways of illustrating the structural formulas. On the left is
a c o n d e n s e d form, w h e r e only s o m e of the b o n d s are shown, w h e r e a s on the right all
the b o n d s b e t w e e n atoms are shown. In m o s t c a s e s we will use the c o n d e n s e d form as
it is m u c h m o r e convenient.
T h e s e isomers have b e e n n a m e d using the rules o f the International Union o f Pure
and Applied Chemistry ( I U P A C ) . F o r alkanes these rules are as follows:
1 . T h e b a s e n a m e o f the c o m p o u n d i s the n a m e o f the longest straight-chain alkane
that is present.
2. Any chain that b r a n c h e s from the straight chain or any functional group that is attached
to the straight chain is named. C a r b o n chains are n a m e d as alkyl groups. T h e following
are important alkyl groups:
Methyl

CH,

Ethyl

CH,CH,

Propyl

CH,CH CH

Isopropyl

CH3CHCH3

BUTVL

CH,CH CH,CH,
2

CH,

/-Butvl

CHXCH,

I
O t h e r functional g r o u p s include such things as c h l o r i d e ions, indicated by the
prefix c h l o r o .
3. T h e location of the functional group is indicated by n u m b e r i n g the c a r b o n atoms in
the longest straight chain, with the e n d c a r b o n closest to the position of the first
functional group being c a r b o n 1.
4. F o r m o r e than o n e of the s a m e functional group, the prefixes di-, tri-, and tetra- are
used.
Consider the following branched-chain alkane:
CH,

CHXH,

CH,CHCH,CHCHXH,CH,
I

T h e longest straight chain has seven c a r b o n s , so this is a heptane. T h e c a r b o n s


are n u m b e r e d starting with the e n d closest to an attached functional group. A methyl
group is attached to the s e c o n d c a r b o n and an ethyl group is attached to the fourth
c a r b o n . T h e n a m e o f the c o m p o u n d i s 2-methyl-4-ethylheptane.
Alkanes can also have a cyclical structurethat is, the two ends of the c a r b o n
chain can be j o i n e d together. T h e shortest chain that can form a c y c l o a l k a n e is propane.
Cycloalkanes are also k n o w n as cycloparaffins, or naphthenes, and have a formula of
C H . T h e structure o f c y c l o b u t a n e i s
2

H C C H

I
II

c C

II

300

Chapter Seven
Alkenes

Alkenes have a c a r b o n - c a r b o n d o u b l e bond, with the general formula CH . Alkenes


are also known a s u n s a t u r a t e d h y d r o c a r b o n s , o r o l e f i n s . They are named b y
finding the longest chain that contains the d o u b l e b o n d . T h e base name ends in ene
( o r -ylene) rather than -arte. T h e c a r b o n atoms in the straight chain are numbered
starting at the end nearest the d o u b l e b o n d . T h e simplest alkene is ethene (ethylene);
it has the formula C H = C H . If the d o u b l e b o n d can o c c u r in m o r e than o n e position,
that position is indicated by a numerical prefix, which is the n u m b e r of the first c a r b o n
atom containing the d o u b l e b o n d . F o r example, b u t e n e can have the d o u b l e b o n d in
2

t\Vi i

|li 'Mil. i l l s

CH = CHCH CH
2

CH CH=CHCH
3

l-Burene

2-Buiene

If there are functional groups present, then their position is indicated by the carbon
atom to which they are b o n d e d . S o m e of the alkenes can exist as structural isomers
b e c a u s e of the d o u b l e b o n d . C a r b o n molecules in alkanes can rotate around the single
b o n d s so that a n u m b e r of structural forms are equivalent. With d o u b l e bonds, such
rotation is not possible. F o r e x a m p l e there are 2 isomers of 1,2-dichloroethene:
CI

CI

CI

C = C

/
II

/
II

Cjts-l,2-Dichloroethene

H
/rflr-l,2-Dichloroctheni.'

In addition to the formal names derived from the IUPAC system, many organic
c o m p o u n d s have c o m m o n names as well. C o m m o n names will be given as synonyms.

7.3.2

Aromatic Hydrocarbons

Aromatic hydrocarbons are based on the b e n z e n e ring. Figure 7.1(a) shows the cond e n s e d form of the b e n z e n e ring. O t h e r molecules can be formed by joining functional
groups to the b e n z e n e ring. Figure 7 . 1 ( b ) shows methylbenzene, which has the c o m m o n
n a m e toluene. If only o n e functional group is attached, then a position is not specified,
since all six c a r b o n atoms are equivalent. If there are two functional groups attached,
then there are three isomers. If the two functional groups are the same, they may be
distinguished by the prefixes ortbo- ( o ) , meta- (m) and para ( p ) , as shown in Figure
7 . 1 ( c ) for dimethylbenzene, which is c o m m o n l y called xylene. A numbering system
for the six c a r b o n atoms may also be used, as illustrated in Figure 7 . 1 ( d ) for 1,3,5trimethylbenzene.
T w o o r m o r e b e n z e n e r i n g s may b e joined together. T h e most simple p o l y c y c l i c
a r o m a t i c h y d r o c a r b o n ( P A H ) i s naphthalene, which consists o f two b e n z e n e r i n g s .
T h r e e or m o r e b e n z e n e rings can also join. T h e structure and s o m e properties of a
n u m b e r of PAH c o m p o u n d s are shown in Figure 7 . 2 . In these c o m p o u n d s the b e n z e n e
rings share c a r b o n atoms. PAHs are found in heavy fractions of petroleum distillation.

301

O r g a n i c Compounds in G r o u n d W a t e r

F I G U R E 7.1
(toluene); (c)

Structure and nomenclature of the benzene ring, (a) Benzene ring; (b) methyl benzene
o-dimethyl

benzene,

m-dimethyl

benzene, and p-dimethyl

benzene;

(d)

1,3,5,-trimethyt

benzene.

asphalt, coal tar, and c r e o s o t e . T h e y also form from the i n c o m p l e t e c o m b u s t i o n of fossil


fuels.
If the b e n z e n e ring is j o i n e d to a n o t h e r group, it may be n a m e d as a functional
group, phenyl-. F o r e x a m p l e , Figure 7.3 s h o w s the structure of biphenyl and diphe
nylmethane. If biphenyl is chlorinated, it yields a mixture of i s o m e r s containing 1 to 10
chloride ions. T h e s e c o m p o u n d s are called p o l y c h l o r i n a t e d b i p h e n y l s , o r P C B s ,
and always contain a mixture of isomers. T h e y are quite resistant to chemical, thermal,
or biological degradation and tend to persist in the environment.

7.4

Petroleum Distillates
Crude oil consists of a mixture of h y d r o c a r b o n s of varying molecular weight and on
the average contains about 8 4 . 5 % c a r b o n , 1 3 % hydrogen, 1.5% sulfur, 0 . 5 % nitrogen,
and 0 . 5 % oxygen. A typical crude oil might consist of a b o u t 2 5 % alkanes (paraffins),
5 0 % cycloalkanes ( n a p h r h e n e s ) , 1 7 % aromatics, including polycyclic aromatics, and 8%
asphaltics, which are molecules of very high molecular weight with m o r e than 40 c a r b o n

Soil-Water
Name

Structure

Molecular

Solubility

Partition

Weight

in Water

Coefficient

Benzene

Toluene

Xylene, ortho

Ethyl benzene

Naphthalene

Acenaphthene

Acenaphthylene

Fluorene

Fluoranthene

Phenanthrene

Anthracene

FIGURE 7.2

Structure and properties of some aromatic and polycyclic aromatic hydrocarbons.

Soil-Water

Structure

Name

Pyrene

Benzo[o]anthracene

Benzo[o]pyrene

Chrysene

oo?
uOO
^00

Benzo[b]fluoronthene

Benzo[g,fi,i]perylene

Que

Dibenz[a,h]anthracene

Benzo[ fc]fluoranthene

F I G U R E 7.2

Continued

0@&h

Molecular

Solubility

Partition

Weight

in Water

Coefficient

202.26

0 . 1 3 5 mg/L

63,000

228

0 . 0 1 4 mg/L

125,719

252.3

0.0038

282,185

228.2

0.006 mg/L

252

0.0012

276

0 . 0 0 0 2 6 mg/L

1,488,389

278.35

0 . 0 0 2 4 9 mg/L

1,668,800

252

0 . 0 0 0 5 5 mg/L

2,020,971

mg/L

mg/L

420,108

1,148,497

Chapter Seven

3'

2'

There can be from 1 to 10 chloride ions bonded to the biphenyl


Polychlorinoted biphenyl

cci

Dichloro-diphenyl-trictiloroethane (DDT)

F I G U R E 7.3

Structure of some phenyl compounds.

atoms. T h e r e have b e e n m o r e than 6 0 0 hydrocarbon c o m p o u n d s identified in petroleum


(Hunt, 1 9 7 9 ) .
Petroleum is separated into fractions by distillation. T h e boiling point of hydro
c a r b o n s is correlated to the n u m b e r of c a r b o n atoms. For example, T a b l e 7 . 1 lists several
h y d r o c a r b o n s with six c a r b o n atoms and their boiling points. Although all have different
structures, the boiling points are similar.
In a h o m o l o g o u s series, the boiling point of the hydrocarbon will rise with the
n u m b e r of c a r b o n atoms. T a b l e 7.2 shows the boiling points for s o m e of the normal
alkanes. B e c a u s e of these two characteristics, during distillation petroleum is separated

T A B L E 7.1

Boiling points of several hydro-

carbons with six carbon atoms.


B o i l i n g Point
Hydrocarbon

(C)

Benzene

B0.1

Hexane

68.7

Cyclohexane

81.0

2-Methylpentane

60

Methylcyclopentane

72

Organic

Compouni s in G r o u n d W a t e r

TABLE

7.2

Boiling points of norma alkanes.


B o i l i n g Point

Alkane

Carbon A t o m s

(C)

Butane

Pentane

36

Hexane

69

Heptane

98

Octane

126

Nonane

151

Decane

10

174

into the h y d r o c a r b o n fractions with similar n u m b e r s of c a r b o n atoms according to T a b l e


7.3.
Although the c o m p o s i t i o n of each fraction is c o m p l e x and is subject to a great
deal of variation depending u p o n the crude oil and the refiner)', the h y d r o c a r b o n s in
each fraction will have similar n u m b e r s of c a r b o n atoms and boiling points. Gasoline,
for e x a m p l e , contains over 1 0 0 separate organic c o m p o u n d s . T a b l e 7.4 gives an analysis
of two different brands of gasoline. In addition to the listed chemicals, there are many
c o m p o u n d s present that constitute less than 0 . 0 1 % of the mixture.

7.5

Functional Groups
7.5.1

Organic Halides

We have already s e e n that c h l o r i n e atoms c a n be substituted for hydrogen a t o m s at


various places on organic m o l e c u l e s . T h i s is also true for b r o m i n e and fluorine atoms.
Such c o m p o u n d s are n a m e d by prefixing the n a m e of the basic m o l e c u l e with the term
cbloro-, brotno-, or fluoro- and specifying the n u m b e r and position. Figure 7.4 shows

T A B L E 7.3

Hydrocarbon fractions separated by distillation.


R a n g e of C

Fraction

A t o m s per

Boiling Point

Molecule

Range

Uses

Gas

1 to 4

20 C

Cooking, home heating,

Gasoline

5 to 10

20-190C

Fuel, benzene for chemical

Kerosene

11 to 1 3

190-260C

Diesel

14 to 18

260-360C

Diesel fuel and fuel oil

Heavy gas and

19 to 40

360-530C

Lubricating oil, greases.

chemical feed stock


feed stock

lubricating

Fuel, jet fuel

waxes

oils
Residuum

>40

> 560C

Asphalt

T A B L E 7.4
gasolines.

Chemcial composition of 8 7 . ) octane Union 76 unleaded regular and 9 2 . 9 octane Amoco premium unleaded

Hydrocarbon*

Percent

Percent

Composition

Composition

Union 76

Amoco

Union 76

Amoco

No-lead

Premium

No-lead

Premium

Regular

No Lead

Regular

No-Lead

Hydrocarbon"

Isobutane

1.86

1.40

3-Methylheptane

0.70

n-Butane/1,3-butodiene

7.75

3.52

2,2,5-Trimethylhexane

0.81

0.23
0.76

trans-2-Butene/2,2-dimethylpropane

0.25

0.13

n-Octane

0.76

0.20

cis-2-Butene/1 -butyne

0.25

0.13

2,3,5-Trimethylhexane

0.18

0.13

3-Methyl-l -butene

0.10

0.07

2,4-Dimethylheptane

0.14

0.08

Isopentane

6.16

7.12

2,5- and 3,5-

0.24

0.09

0.32

0.18

n-Pentane

3.06

2.37

1.17

0.94

frans-2-Pentene

0.89

0.73

cis-2-Pentene

0.51

0.41

p- and m-Xylene

4.58

2.60

2-Methyl-2-butene

1.22

1.50

2,4,5-Trimethylheptane

0.37

0.10

2,2-Dimethylbutane

0.41

0.08

o-Xylene/unknown C

2.46

1.61

Cyclopentene

0.37

0.31

Cyclopentane/3- and

0.48

0.42

2,4-Dimethyloctane

0.14

0.05

n-Nonane

0.27

0.18

-Pentane/2-butyne

4-methyl-1 -pentene

Dimethylheptane
Ethylbenzene/2,3Dimethylheptane

paraffin

2,3-Dimethylbutane

0.86

0.78

C,

paraffin

0.16

0.32

2-Methylpentane/2,3-

2.76

2.76

n-Propylbenzene

0.70

0.90

1,3,5-Trimethyibenzene

2.74

3.35
1.42

dimethyl-1-butene

3-Methylpentane

1.76

1.47

3,4-Dimethyloctane

1.12

0.64

0.64

-Methyl-3-ethylbenzene

1.52

1.53

n-Hexane/cis-3-hexene

1.32

0.83

-Methyl-2-ethylbenzene

0.28

0.07

frans-3-Hexene

0.80

0.73

1,2,4-Trimethylbenzene

3.75

4.59

2-Methyl-2-pentene

0.61

0.65

sec-Butyl benzene

0.25

0.17

2-Hexene (cis and trans)

0.33

0.27

1,2,3-Trimethylbenzene

1.21

1.26

Methylcyclopentane/3-methyl-

1.17

0.77

Indane

0.62

0.66

Isobutylbenzene

0.42

0.48

2,4-Dimethylpentane

1.15

0.86

1.15

1.32

Benzene/cyclohexane

1.76

1.96

1,3-Diethylbenzene

0.67

0.78

Cyclohexene/2,3-dimethylpentane/

2.73

1.31

-Hexene/2-ethyl-butene

trans-2-pentene

2-methylhexane

-Methyl-3-n-propyibenzene

0.82

0.99

1,2-Diethylbenzene

0.57

0.68
1.38

-Methyl-3-isopropylbenzene

3-Methylhexane

1.91

1.04

2-Methyldecane

1.83

2,2,4-Trimethylpentane

3.75

2.07

aromatic

1.53

1.62

n-Heptane

1.23

0.42

C | aromatic

0.51

0.67

Methylcyclohexane

1.57

0.33

n-Undecane

0.75

0.69

Dimethylhexene

0.28

0.25

0.77

1.03

2,2-Dimethylhexane

0.12

0.18

Unknowns

7.90

10.17

2,4- and 2,5-Dimethylhexane

1.14

0.84

2,3,3-Trimethylpentane

2.26

1.82

Percent aromatics

31.23

44.20

Toluene/2,3-dimethylhexane

5.54

20.25

Percent olefins

10.54

9.33

2-Methylheptane/l -methyl-

0.37

0.10

Percent paraffins

58.23

46.47

1.20

0.25

1 0

1 0

aromatic

cyclohexene
4-Methylheptane

Source: Reprinted with permission from J. E. Sigsby, Jr., Silvestre Tejada, ond William Roy, "Volatile Orgonic Compound Emissions from 46 In-Use Passenger
Cars," Environmertfo/ Science and Technology 2 1 , no. 5 (1987):467. Copyright 1987 American Chemicol Society.

306

U s e s and Other Sources

Structure

Name

CI

Trichloromethane

Liquid used in manufacture of

anesthetics, pharmaceuticals,

CI C CI

fluorocarbon refrigerants and

plastics. Used as solvent and

(chloroform)

insecticide. Formed from methane

when chlorinating drinking water.


Vinyl chloride
(chloroethene)

H
\

/
C

\ ,
Cl

H
Chloroethane

polyvinyl chloride. End product of

=C

/
H
1
1

H
1
1

1
1
H

1
1
H

H
1
1

H
1
1

Trichloroethene
(Trichloroethylene)

(perchloroethene)
(perchloroethylene)

chlorinated ethanes.

Liquid used to manufacture vinyl


chloride. Degradation product of
trichloroethane.

11
H
Cl

CL

cl
\

c= c
\

Cl

Cl

Solvent used in dry cleaning and metal


degreasing. Organic synthesis.

c = c
/
\
H
Cl
Tetrachloroethene

chlorinated ethenes.

lead. Degradation product of

CL C - C C L

1
1
H

microbial degradation of

Liquid used to manufacture tetraethyl

H C C Cl

,2-Dichloroethane

Gas used in the manufacture of

Degradation product of
tetrachloroethene.
Solvent used in dry cleaning and metal
degreasing. Used to remove soot
from industrial boilers. Used in

cl

manufacture of paint removers and


printing inks.

,2-Dibromo-3-chloropropane
(DBCP)

Br

Br

CL

I
I

I
I

I
I

Soil fumigant to kill nematodes.


Intermediate in organic synthesis.

H C C C -

H
o-Dichlorobenzene
(1,2-dichlorobenzene)

On"

Chemical intermediate. Solvent.


Fumigant and insecticide. Used for
industrial odor control. Found in
sewage from odor control chemicals
used in toilets.

FIGURE 7.4

Organic halides found in hazardous waste.

307

308

Chapter Seven

the structure and uses o f s o m e o r g a n i c h a l i d e s that are s o m e t i m e s found a s groundwater contaminants. T a b l e 7.5 gives the physical properties of methane, ethane, ethene,
and their chlorinated forms. T h e s e are widely used organic c o m p o u n d s that have often
b e e n found as contaminants in ground water. T e t r a c h l o r o e t h e n e ( p e r c h l o r o e t h e n e ) ,
trichloroethene, and 1,1,1-trichloroethane have a low flammability and a high vapor
density, which m a k e them very useful as solvents. T h e y are frequently used for degreasing
metal parts and as the fluid in dry cleaning. T h e y are d e n s e r than water; if spilled on
the g r o u n d in quantities great e n o u g h to o v e r c o m e the residual saturation, the pure
phase may migrate vertically downward through an aquifer. T h e y are also quite soluble
in water and can migrate as a dissolved phase with flowing ground water.
7.5.2

Alcohols

An a l c o h o l has o n e or m o r e hydroxyl groups, O H , substituted for hydrogen atoms


on an aliphatic hydrocarbon. Alcohols are named by finding the longest hydrocarbon
chain that includes the c a r b o n atom to which the hydroxyl group is attached. That
b e c o m e s the b a s e name, to which suffix

ol is added. T h e position of the c a r b o n atom

to which the hydroxyl g r o u p is attached is indicated by a numerical prefix. If there are


two hydroxyl groups, then the suffix

diol is used, and so forth. Figure 7.5 lists s o m e

c o m m o n alcohols that may find their way to a hazardous waste site. Alcohols are miscible
with water and h e n c e have a potential for significant mobility in ground water. However,
many are also readily biodegraded.
7.5.3

Ethers

E t h e r s have a n o x y g e n atom b o n d e d b e t w e e n two c a r b o n atoms. T h e c o m m o n name


is o b t a i n e d from the n a m e s of the two h y d r o c a r b o n s followed by the word ether. Methyl
ethyl e t h e r is:
CH OCH CH
3

In the IUPAC system, the methoxy functional group is C H 0 , so the preceding


3

c o m p o u n d is also known as 1 - m e t h o x y e t h a n e .
T w o cyclical ethers have b e e n found to be rather persistent contaminants in ground
water. Tetrahydrofuran is used in large quantities as a chemical intermediate and as a
solvent for resins such as PVC, adhesives, and various coatings. It is miscible with water,
is resistant to biodgradation, and has the following structure:
/ \
CH,
CH,

CH, C H ,
1 , 4 - D i o x a n e is also miscible with water and is extremely resistant to biodgradation.
At the Seymour Recycling Corporation Hazardous Waste Site in Indiana, m o r e than 80
c h e m i c a l s have b e e n detected in the plume of contaminated g r o u n d water. T h e comp o u n d that has traveled the greatest distance is l . t - d i o x a n e , with tetrahydrofuran being
the s e c o n d - m o s t m o b i l e c o m p o u n d . 1 , 4 - D i o x a n e is used as a solvent in a wide variety
of lacquers, paints, varnishes, inks, dyes, emulsions, and so on. It is added to chlorinated
solvents as a stabilizer to prevent degradation. Note that it is not a dioxin, which is a

T A B L E 7.5

Common chlorinated alkanes and alkenes with their physical properties.

Alkanes

Synonym

Specific
Gravity

Structure

Melting
Point"

Boiling
Point"

Vapor
Pressure"

Water
Solubility"

Henry's
Law Constant
(m x atm/mol)

0.00584 at 17.5C
0.00131 at 17.5C
0.00246 at 17.5C
0.0211 at I7.5C

Methanes
Methane
Methyl chloride
Methylene chloride
Chloroform
Corbontetrachloride
Ethane
Chloroethane
1,1-Dichloroethone
1,2-Dichloroethane
1,1,1 -Trichloroethane
1,1,2-Trichloroethone
1,1,2,2-Tetrochloroethane
1,2,2,2-Tetrachloroethane
Pentochloroethane
Hexochloroethane

(Chloro methane)
(Dichloromethane)
(Trichloromethone)
(Tetrochloromethane)
(Dimethyl)
(Ethylchloride)
(Ethylenedichloride)
(Methylchloroform)
(Vinyltrichloride)

(Perchloroethane)

CH,
CHjCI
CH,CI,
CHCIj
CCI,

1.489 at 20C

CHJCHJ
CH,CICH,
CHCIjCH,
CH,CICH,CI
CCI3CH3
CHCI,CH,CI
CHCIjCHCI,
CHJCICCIJ
CCIjCHCI,
CCIJCCIJ

-97.7C
-97C
-64C
-23C

0.92 at 20C
1.174 at 20C
1.25 at 20C
1.35 at 20C
1 44 at 20C
1.60 at 20C

Ethanes
-172C
-138.3 C
-97.4C
-35.4C
-32C
-35C
-42.5C

1.60
1.67 at 25C
2.09 at 20C

-29C
1B7.4C

-24C
40-52 C
62C
76.7C

5.0 aim at 20C


3 4 9 mm at 20C
160 mm at 20"C
90 mm at 20C

4,000 cu cm/L
20,000 mg/L at 2 0 C
8,000 mg/l at 20C
8 0 0 mg/t at 20 C

-89C
12.4C
57.3C
83.5C
71/81"C
I13.7X
146.4C
138"C
162C

38.5 atm at 20"C


1.000 mm at 20C
180 mm at 20C
61 mm at 20C
100 mm at 20C
1 9 mm at 20C
5 mm at 20C

60.4 mg/t al 20 C
5,740 mg/t at 2 0 C
5,500 mg/t at 20C
8,690 mg/t at 20C
4,400 mg/t at 20C
4,500 mg/t at 20C
2,900 mg/t at 20 C

0.00846 at 17.5 C
0.00389 at 17.5C
0.0120 at 17.5C

3.4 mm at 2 0 X
50 mg/L at 22C

Ettienes
Ethene
Vinyl chloride
1,1 -Dichloroethene
cis-1,2-Dichloroethene
trans-1,2-Dichloroethene
Trichloroethene
Tetrachloroethene

(Ethylene)
(Chloroethene)

(TCE)
(Perchloroethene)

CH,=CH,
CH,=CHCI
CCIJ=CHJ

1.218 at 20C
1.28
1.26
1.46 at 20C
1.626 at 20C

CHCI=CHCI
CHCI=CHCI
CHCI=CCIj
CCI,=CCI
2

-169C
-153C
-122.5C
-81C
-50C
-87C
-22.7"C

-104C
-13.9C
31.9C
60C
48C
86.7"C
121.4'C

> 4 0 atm at 20"C


2,660 mm at 25C
5 0 0 mm at 20C
2 0 0 mm at 2 5 C
2 0 0 mm at 14C
60 mm at 20C
14 mm at 2 0 C

131 mg/L at 20C


1.1 mg/L at 25C
8 0 0 mg/L at 20 C
6 0 0 mg/L at 20 C
1,100 mg/L al 25C
150 mg/L at 2 5 C

Sources: K. Verschueren, Handbook of Environmental Data on Organic Chemicals, 2d ed. (New York: Van Noslrond Reinhold Company, 1983), 1310 pp.
b

J. M. Gossett, "Measurement ot Henry's Low Constants for C, and C Chlorinated Hydrocarbons," Environmental Science and Engineering 2 1 , no. 2 (1987):202-9.
;

0.0193 at 17.5 C
0.0191 at 17.5C
0.00265 at 17.5C
0.00660 al 17.5C
0.00632 at 17.5C
0.01 17 at 1 7 . 5 X

Name

Structure

Methanol

U s e s and

(wood alcohol)

Sources

Solvent. May be added to gasoline.

Manufacture of formaldehyde and

H C H

methyl halides.

OH
Ethanol
(grain alcohol)

Preparation of distilled spirits.


Solvent, Manufacture of acetaldehyde,

H C C H
1
1
1
1
H
OH

(ethylalcohol)

acetic acid, ethyl ether, etc.


Preparation of lacquers, perfumes,
cosmetics, over-the-counter medicines.
Degradation of ethyl acetate in
excess of water.

Ethylene glycol

Antifreeze (engine coolent) compound.

1 1

(1.2-ethanedlol)

Manufacture of polyester fiber and

H C C H

film. Deicing compound for airplanes


at gate. Solvent base.

OH OH
Propanol
H

1
c

1 1
c c

and during sewage treatment and


H

1
H

Released from fermentation of whisky

1
OH

decomposition of organic matter.


Solvent in printing, used in nail polish.
brake fluid, lacquers, cleaners,
polishes.

1,2-Propanediol

OH

OH

H
CH, H
H
1
I I
1
1
1
1
1
H C c c c H
I
1
1
1

OH

f-Butanol
(2-methyl-2-propanol)
H

H
1
1
H C
|

310

and perfumes. Solvent, Paint removers.


H

OH

C H
1

c c c

1
1

1
1

11

CH

Octane booster in gasoline. Lacquer.


Solvent for pharmaceuticals.

OH

Alcohols found in hazardous waste.

Solvent.

Manufacture of flotation agents, flavors,

CH, H

1 1 1
C C C
1

F I G U R E 7.5

coatings. Antifreeze formulations.


H

c c c

4-Methyl-2-pentanol

Solvent used in paints, inks, and

1 1 1

(propylene glycol)

2-Methy-2-butanol

Solvent.

O r g a n i c Compounds i n G r o u n d W a t e r

311

class o f c o m p o u n d s thought t o b e highly toxic. T h e structure o f 1 , 4 - d i o x a n e i s


/ \
CH,
CH,
CH,
x

7.5.4

CH,

Aldehydes and Ketones

Aldehydes and k e t o n e s contain the c a r b o n y l g r o u p , which has this structure:

\
c=o

/
An a l d e h y d e has at least o n e hydrogen atom b o n d e d to the c a r b o n . T h e IlIPAC
n a m e is o b t a i n e d by finding the n a m e of the h y d r o c a r b o n and adding the suffix -al. T h e
two m o s t simple aldehydes are methanal ( f o r m a l d e h y d e ) and ethanal ( a c e t a l d e h y d e ) :
CH,

\
c=o

c=o

/
H

Formaldehyde
(methanal)

Acetaldehyde
(ethanal)

K e t o n e s have the carbonyl g r o u p b o n d e d t o two h y d r o c a r b o n s . T h e y are n a m e d


by indicating the n a m e of the h y d r o c a r b o n and ending with the suffix -one. T h e m o s t
simple k e t o n e is p r o p a n o n e , which is c o m m o n l y k n o w n as a c e t o n e . A c e t o n e is a widely
used solvent, as are o t h e r k e t o n e s such as 2 - b u t a n o n e (methyl ethyl k e t o n e ) , 4 methyl2 - p e n t a n o n e (methyl isobutyl k e t o n e ) and 2 - p e n t a n o n e (methyl propyl k e t o n e ) . Structures o f s o m e o f t h e s e k e t o n e s include the following:
CH,CH,CH,

CH^
2-Pentanone
7.5.5

CH,CH,

V
c=o

CH,

\
c=o

CHj
2-Buranone

c=o

/
CH^
Propanone (acetone)

Carboxylic Acids

C a r b o x y l i c a c i d s have the c a r b o x y l g r o u p , C O O H . T h e y are n a m e d b y taking


the stem n a m e and adding the suffix o i c acid. T h e structure of the functional g r o u p is

C OH
Acetic acid ( e t h a n o i c a c i d ) is a c o m m o n industrial chemical, which isn't thought
to p o s e health threats. O n e c a r b o x y l i c acid that is widely used is 2 - p r o p e n i c acid (acrylic

31 2

Chapter Seven

acid). Large amounts are used during the manufacture of acrylic esters. Its structure is
H

II

H C = C C OH
Another carboxylic acid used in large quantities for the manufacture of .synthetic
fibers, plasticizers, resins, and foams is adipic, or 1,6-hexanedioic acid, which has two
carboxyl groups
O

II

II

HO C C C C C C O H

Carboxylic acids are weak acids and do not strongly dissociate in water.
7.5.6

Esters

E s t e r s are the result of the c o m b i n a t i o n of a carboxylic acid with an alcohol. T h e tunc


tional group is:
O

II
R C OR'
w h e r e R represents the remainder of the carboxylic acid and R' is the alcohol. T h e ester
is n a m e d by the n a m e of the alcohol group followed by the n a m e of the carboxylic acid
group with the suffix

ate. Ethanol and formic acid join to create ethyl formate. S o m e

esters have o d o r s that are associated with many everyday substances. F o r example, ethyl
formate smells like rum.
O

II

H C OCCH

Ethyl formate

Esters that are used in industry for flavorings, in perfumes, and as solvents, especially
for paints, include ethyl formate, penryl acetate (n-amyl acetate), butyl acetate, ethyl
acetate, isobutyl acetate, and 3-methylbutyl acetate (isoamyl acetate).
O n e class of esters that is very c o m m o n in industrial use is the phthalates. They
are used as plasticizers to improve the flexibility of various plastics. S o m e are used as
solvents and in insect repellents. Figure 7.6 gives the structural formulas for several
c o m m o n phthalates. Figure 7 . 6 ( a ) shows the b a s e structure of the phthalates, which is
derived from phthalic acid.
7.5.7

Phenols

P h e n o l s are based on a hydroxyl radical b o n d e d to a b e n z e n e ring. Phenol can o c c u r


naturally in ground water, usually in low amounts associated with d e c o m p o s i n g organic

313

O r g a n i c Compounds i n G r o u n d W a f e r

[dl

FIGURE 7.6

Structure of some phthalates. (a) Dimethyl phthalate; (b) diethyl phthalate; (c) dJ-n-butyl

phthalate; (d) butyl benzyl phthalate.

matter. P h e n o l is also a c o m m o n ground-water contaminant, d u e to its use in many


industrial p r o c e s s e s .
T h e r e are many p h e n o l - b a s e d c o m p o u n d s used in industry that can o c c u r in g r o u n d
water. C r e s o l s have a methyl g r o u p attached to the b e n z e n e ring of toluene. T h e y are
used in industry and are also released during coal-tar refining. C h l o r o p h e n o l s have o n e
or m o r e chloride ions on the b e n z e n e ring of a phenol. C h l o r o p h e n o l is a synthetic
intermediate for the manufacture of dye and m o r e highly chlorinated phenols. Tric h l o r o p h e n o l is used to preserve w o o d and leather and as a b i o c i d e and antimildew
agent. P e n t a c h l o r o p h e n o l is used as a w o o d preservative. It is a solid and is usually
dissolved in a carrier solvent, such as diesel fuel, in making treated w o o d . Nitrophenols

314

Chapter

Seven

have a nitrate group b o n d e d to the b e n z e n e ring of the phenol. S o m e phenols found


in hazardous waste include phenol, p chloro-m-cresol, 2 chlorophenol, 2,4-dichloro
phenol, 2,4-dimethylphenol, 4 . 6 d i n i t r o - o c r e s o l , 2,4-dinitrophenol, 2 nitrophenol, 4-ni
trophenol, 2,4,6-trichlorophenol, and pentachlorophenol. T h e structure o f s o m e o f these
c o m p o u n d s is found in Figure 7.7.

7.5.8

Organic Compounds Containing Nitrogen

We have already seen that organic c o m p o u n d s can include nitrogen by attaching a nitrate
ion to a c a r b o n atom, for e x a m p l e nitrophenols. Nitrotoluenes are also c o m m o n industrial
chemicals and include 2- and 4-nitrotoluene, 2,4-dinitrotoluene, and 2,4,6 trinitrotoluene
( T N T ) . T N T is an explosive and has b e e n reported as a soil contaminant in areas of
waste disposal from manufacture of munitions and explosives.

OH

N0
2-Nitrophenol

F I G U R E 7.7

Structure of phenols.

2,4 Dinitrophenol

N0

4.6-Dinltro-o-cresol

O r g a n i c Compounds i n G r o u n d W a t e r

315

A m i n e s are organic c o m p o u n d s w h e r e the b a s e m o l e c u l e i s a m m o n i a and o n e


or m o r e of the hydrogen a t o m s is replaced by a h y d r o c a r b o n group. S o m e simple a m i n e s
include the following:
CH NH
3

Methyl amine

CH CH NH
3

CH NHCH

Ethyl amine

Dimethyl amine

A m i d e s have the following g r o u p present:


H

R' N C H
Acrylamide ( 2 - p r o p e n e a m i d e ) is used in the synthesis of dyes and the manufacture
of polymers, adhesives, and p e r m a n e n t - p r e s s fabrics. It has this structure:
H

II

H C = C C NH,
N i t r i l e s have a cyanide group, in which there is a triple b o n d b e t w e e n a c a r b o n
atom and a nitrogen atom. An e x a m p l e of a nitrile is acrylonitrile (2-propenenitrile).
T h i s c h e m i c a l is p r o d u c e d in very large a m o u n t s and is used in the manufacture of
acrylic fibers, polyacrylonitrile plastics, ABS resins, and o t h e r products. It has this structure:
C ^ N

/
H C = C

\
H
Atrizine, a c o m p l e x m o l e c u l e containing a six-sided ring structure with alternating
1

c a r b o n and nitrogen atoms, is a ver) wider)' used herbicide. It is used as a p r e e m e r g e n t


w e e d c o n t r o l on c o r n in the midwestern United States and for w e e d c o n t r o l on sugar
c a n e and p i n e a p p l e fields in Hawaii. Its structure is given in Figure 7.8. Although atrazine
is fair!)' quickly degraded in the environment by c h e m i c a l and microbial p r o c e s s e s it has
b e e n found in a n u m b e r of wells in rural areas of W i s c o n s i n . During biodgradation
the parent c o m p o u n d i s transformed t o o t h e r c o m p o u n d s , called m e t a b o l i t e s . T w o
pathways by which the degradation of atrizine b e g i n s are s h o w n in Figure 7.8. B i o d gradation is discussed in detail later in this chapter.
7.5.9

Organic Compounds Containing S u l f u r and P h o s p h o r u s

Organic c o m p o u n d s may also contain sulfur and p h o s p h o r u s m o l e c u l e s . Organosulfur


c o m p o u n d s tend t o have offensive o d o r s . E x a m p l e s o f simple organosulfur c o m p o u n d s
include mercaptans, the active ingredients in the spray of a skunk.
H
H

I
C
I

H
S

H
Methyl mercaptan

I I
HC CS H
I I
II

Ethyl mercaptan

H C S C H

Dimethyl mercaptan

316

Chapter Seven

CI
H

N ^ N
C,N NH

J L

NH

CH
CH

Atrazine structure
CI

OH
CH,

- N H - U

N ^ N

J - N H CH
CH

H C NH - ^
5

3
H

J - N H - H
CH

Chemicol hydrolysis to hydroxyatrazine


CI

CI
CH

N ^ N

H C -NH-l*, J5

NH

CH
1
CH

..
> H,N

U^ J U h - C H
1

CH

Microbial dealkylofion to
2-chloro-4-arnino-6-isopropylamino-s-triozin

F I G U R E 7.8

Structure of otrizine and metabolic pathways leading to degradation.

Man)' pesticides include sulfur, p h o s p h o r u s , or both in their composition. Figure


7.9 gives the structure of s o m e c o m m o n o r g a n o p h o s p h o r u s and organosulfur pesticides.
Most of these have b e e n associated with ground-water contamination either in areas of
their use or w h e r e the)' w e r e manufactured.

7.6

Degradation of Organic Compounds


7.6.1

Introduction

It is well known that straight chain and aromatic h y d r o c a r b o n s associated with petroleum
products can undergo biological degradation ( B a r k e r , Patrick, and Major 1 9 8 ~ J. T.
Wilson et al. 1 9 8 6 ) . Hydrogeologists have recently observed that halogenated organic
solvents dissolved in g r o u n d water undergo transformations under natural conditions
with the c o m p o u n d s undergoing progressive dehalogenation ( R o b e r t s , Schreiner, and
Hopkins 1 9 8 2 ; Parsons, W o o d , and De Marco 1 9 8 4 ; CLine and Viste 1 9 8 5 ) . B e n z e n e ,
toluene and xylene have also b e e n o b s e r v e d to undergo degradation in aquifers (Chiang
et al. 1 9 8 9 ; B a r k e r and Patrick 1 9 8 5 ; Barker, Patrick, and Major 1 9 8 7 ) .

Organic Compounds in Ground Water

Name

317

Use

Structure
O

Captan

Fungicide used on foliage

C
N S CCI,

II
o
Systemic insecticide used on

Disu Itone
CH CH O
3

plants

II

CH,CH,0
P S CH CH S CH CH
CH, H
2

Aldicarb

Systemic insecticide and


nematicide

O N
H CH,
H,C S C C = N O C

I
CH,
Endosulfan

s=o

Malathion
I C H j O l j P S CH C O C H
2

CH, C OC,H

Parathion
C H O
2

PO

Asulam

Insecticide and acaricide


NO,

II
II

II

Herbicide

S NH C OCH,
O
F I G U R E 7.9

Structure and uses of organosulfur and organophosphorus pesticides.

In this section we will e x a m i n e the types of organic c o m p o u n d s that can undergo


degradation, the m e a n s by which such degradation o c c u r s , and the conditions under
which it o c c u r s . Degradation is defined as the p r o c e s s of an organic m o l e c u l e b e c o m i n g
smaller by chemical or biological means. A m o l e c u l e might have a halide ion replaced
with a hydrogen ion, thus being transformed into a c o m p o u n d with a lower molecular
weight. C a r b o n atoms can be b r o k e n off the m o l e c u l e , leaving it with fewer atoms. T h e

318

Chapter Seven

OH

OH

la)
OH

i )H

lb]

FIGURE 7.10

Degradation of a benzene ring, (a) Aerobic degradation of benzene in the presence of

oxygen, (b) Anaerobic degradation of phenol in the presence of nitrate.

ultimate product of degradation of a hydrocarbon is methane or c a r b o n dioxide and


water.
7.6.2

Degradation of Hydrocarbons

Alkanes can be degraded under a e r o b i c conditions by m i c r o b e s (Manahan, 1 9 8 4 ) . Bacteria that can do this include Micrococcus, Pseudomonas, Mycobacterium, and Nocardia.
T h e first step in the p r o c e s s is the conversion of a terminal C H g r o u p to a carboxyl
3

group. T h e m i c r o b e s then attack the s e c o n d c a r b o n of the newly formed carboxylic


acid and remove a two-carbon fragment, forming c a r b o n dioxide. T h e oxidation of
n-hexane follows these steps:
CH CH CH CH CH CH 4- 0
3

C H C H C H C H C H C 0 H 4- 3 0
3

CH CH CH CH CH C0 H
3

C H C H C H C 0 H 4- 2 C 0

C H C H C H C 0 H 4- 3 0
3

C H C 0 H 4- 2 C 0
3

4- f l 0
2

4- H 0
2

T h e resulting acetic acid can easily be further degraded to c a r b o n dioxide and


water. B r a n c h e d chain hydrocarbons are m o r e resistant to microbial degradation than
normal alkanes.
Aromatic h y d r o c a r b o n s can also be degraded under a e r o b i c conditions. T h e first
step in the cleavage of the ring is to replace two hydrogens on adjacent c a r b o n atoms
with hydroxyl groups. T h e b e n z e n e ring is then cleaved between these two rings to
form a bicarboxylic acid. Figure 7 . 1 0 ( a ) illustrates the ringcleavage process, which
requires molecular oxygen.
It is also p o s s i b l e for aromatic hydrocarbons to undergo anaerobic degradation
(Evans 1 9 7 7 ) . T h e b e n z e n e ring can be degraded in the p r e s e n c e of nitrate by Pseu
domonas sp. and Moraxella sp.* T h e b e n z e n e ring is first saturated to cyclohexane,

* The designation Pseudomonas sp. means that the bacterium has been identified to the level of the genus
Pseudomonas. but not to the species level.

O r g a n i c Compounds in G r o u n d W a t e r

319

then oxidized to a ketone, and finally cleaved by hydrolysis to form a carboxylic acid.
Figure 7 . 1 0 ( b ) s h o w s the degradation of phenol, first to c y c l o h e x a n o l . then an oxidation
to c y c l o h e x a n a t e and then ring cleavage to form h e x a n o i c acid. During this p r o c e s s the
nitrate is reduced to nitrogen.
Major et al. ( 1 9 8 8 ) m a d e a study of biodgradation of b e n z e n e , toluene, o-xylene,
and m-xylene in the laboratory using nitrate as the electron a c c e p t o r under denitrifying
conditions. T h e y found that degradation of all four c o m p o u n d s o c c u r r e d and suggested
that bioremediation using added nitrate to g r o u n d water might be possible.
Polycyclic aromatic h y d r o c a r b o n s ( P A H s ) have also b e e n reported to be degraded
by m i c r o b e s . B u m p u s ( 1 9 8 9 ) demonstrated that under a e r o b i c conditions in the laboratory, a fungus, Phanerochaete chtysosporium. was able to degrade all the major PAHs
in a n t h r a c e n e oil, which is a c o m p l e x mixture of PAH c o m p o u n d s o b t a i n e d from the
fractional distillation of coal oil. PAHs reported to be degraded included fluorene, phenanthrene, anthracene, carbazole, fluoranthene, pyrene, b e n z o l a] fluorene, 1-methylfluo r e n e , and acenaphthalene.
T h e initial steps in ring cleavage of PAH c o m p o u n d s naphthalene, phenanthrene,
anthracene. 1,2-benzanthracene, and b e n z o l a] pyrene are s h o w n on Figure 7.11 ( G u e r i n
and J o n e s 1 9 8 8 ) . Heitkamp e t al. ( 1 9 8 8 ) have identified s o m e o f the products o f r i n g
oxidation and ring fusion of pyrene, a four-ring PAH, that w e r e produced by the bacterium
Mycobacterium sp. in an oxygenated lab environment that had additional organic nutrients. T h e identified metabolites of pyrene are s h o w n in Figure 7.12. Heitkamp and
Cerniglia ( 1 9 8 9 ) have s h o w n that this bacterium can also degrade a mixture of 2methylnaphthalene, phenanthrene, pyrene, and b e n z o l a ] a n t h r a c e n e in a laboratory culture m a d e with natural sediment and ground water, demonstrating that the bacteria could
c o m p e t e with native soil bacteria for available nutrients.
T h e s e laboratory studies are valuable for identifying m i c r o b e s that can degrade
organic c o m p o u n d s . T h e m i c r o b e s themselves are initially found at sites w h e r e PAH
c o m p o u n d s have b e e n released to the soil. T h e Mycobacterium sp. bacterium was isolated
from hydrocarbon contaminated sediment samples from an oil field near Port Aransas,
T e x a s (Heitkamp and Cerniglia 1 9 8 8 ) . Mueller et al. ( 1 9 9 0 ) isolated a bacterium, Pseudomonaspattcimoblis, that is c a p a b l e of using fluoranthene as the s o l e s o u r c e of c a r b o n
for energy and growth. This strain was discovered at a site w h e r e c r e o s o t e , which is rich
in PAH, had b e e n discarded. T h e ability- of these naturally occurring strains of m i c r o b e s
to degrade PAH c o m p o u n d s b o d e s well for bioremediation efforts at PAH waste sites.
T h e previously cited studies w e r e c o n d u c t e d u n d e r a e r o b i c conditions. However,
many sites with organic wastes have a n a e r o b i c conditions, b e c a u s e the soil bacteria
c o n s u m e o x y g e n in degrading the organic waste. Mihelcic and Luthy ( 1 9 8 8 ) found that
microbial populations could degrade a c e n a p h t h e n e and naphthalene under a n a e r o b i c
conditions if sufficient nitrate was present to act as a s o u r c e of oxygen. T h e PAH
c o m p o u n d s and soil organic c a r b o n w e r e mineralized with the production of nitrogen.
However, no mineralization o c c u r r e d after the nitrate was depleted.
7.6.3

Degradation of Chlorinated Hydrocarbons

T h e following classes of organic c o m p o u n d s have b e e n found to undergo either biotic


or abiotic degradation: chlorinated methanes, chlorinated ethanes, chlorinated propanes,

320

Chapter Seven

OH

on

OH

OH

CO,H

Naphthalei

OH

. CO.H

O
a
1,2-Etenzantfiracene

Benzo[o]pyrene

F I G U R E 7.11

Cleavage of ring structure of some polycyclic aromatic hydrocarbons.

chlorinated butanes, chlorinated ethenes, brominated methanes, brominated ethenes,


b r o m o c h l o r o p r o p a n e s , chlorinated phenols, chlorinated benzenes, and halogenated acetates (e.g. Mikesell and B o y d 1 9 9 0 ; Harker and Kim 1 9 9 0 ; T o p p and Hanson, 1990;
Kohring, Rogers, and Wiegel 1989; Galli and McCarty 1989a, 1 9 8 9 b ; Freedman and Gossett
1989; Tsien et al. 1989; little et al. 1988; Fliermans et al. 1 9 8 8 ; Nelson, Montgomery, and
Pritchard 1 9 8 8 ; Haigler, Nishino, and Spain 1 9 8 8 ; Fathepure and B o y d 1988; Egli et al.

O r g a n i c Compounds in G r o u n d W a t e r

321

JQQ

Ou
Pyrene

Ring-oxidation products

JQO
H

OH

os-4,5-Pyrenedihydrodiol

OH

fransd, 5-Pyrened i hyd rod iol

Pyrenol

Ring-fission products:

r j
COOH
4-Hydroxyperinaphrhenone

r~\

a il

4-Phenantfiroic acid
u
/

"~-OH

Phthalic acid

FIGURE 7.12

Cinnamic ac d

Structure of identified pyrene metabolites produced by Mycobacterium sp. Source: M. A.

Heitkamp et al., Applied and Environmental Microbiology 54, no. 10, ( ! 9 8 8 ) : 2 5 5 6 - 6 5 .

1 9 8 8 ; Vogel, Criddle, and McCarty 1 9 8 7 ; Vogel and McCarty 1 9 8 7 a , 1 9 8 7 b , 1 9 8 5 ; BarrioLage, Parsons, and Nassar 1987; Barrio-Lage, Parsons, Nassar, and Lorenzo 1 9 8 6 , 198"";
Nelson et al. 1 9 8 6 , 1 9 8 7 ; Vogel and Reinhard 1 9 8 6 ; G r b i c Galic and Vogel 1987; Pignatello
1 9 8 6 ; Strand and Shippert 1 9 8 6 ; B. H. Wilson, Smith, and Rees 1 9 8 6 ; J a n s s e n et al. 1 9 8 5 ;
S c h w a r z e n b a c h , Giger, Schaffner 1 9 8 5 ; Wilson and Wilson 1 9 8 5 ; Kleopfer et al. 1985;
LaPat-Polasko, McCarty, and Z e h n d e r 1 9 8 4 ; B o u w e r and McCarty 1983a,

1983b;

Burlinson, Lee, and Rosenblatt 1 9 8 2 ) .


T h e chlorinated e t h a n e s and e t h e n e s have b e e n well studied and are c o m m o n
ground water contaminants. T h e r e are a n u m b e r of reactions that can o c c u r abiotically

322

Chapter Seven

to b r e a k them into lower-molecular weight c o m p o u n d s (Vogel, Criddle, and McCarty


1987).
S u b s t i t u t i o n is a reaction w h e r e water reacts with the halogenated c o m p o u n d
to substitute an O H " for an X " , creating an alcohol. This can o c c u r in water without
either inorganic or biological catalysts, but the reaction rates are slow. T h e reaction is
also called h y d r o l y s i s and can be illustrated with 1 b r o m o p r o p a n e :
CH CH CH Br + H 0
3

C H C H C H O H + HBr
3

Substitution reactions p r o c e e d most rapidly for m o n o h a l o g e n a t e d c o m p o u n d s .


M o n o h a l o g e n a t e d c o m p o u n d s have reaction half-lives of about 1 mo. As the n u m b e r of
halogen ions increases, the half-life for reactions due to substitution increases rapidly
into the range of years to hundreds of years.
O t h e r groups can be substituted as well, such as the reaction with an H S " radical
under reducing conditions to release an X~ ion and form a mercaptan;
CH CH Br + HS"
3

CH CH SH + Br"
3

D e h y d r o h a l o g e n a t i o n is a reaction w h e r e an alkane loses a halide ion from


o n e c a r b o n atom and then a hydrogen ion from an adjacent c a r b o n . T h e result is the
formation of a d o u b l e b o n d b e t w e e n the c a r b o n atoms, thus creating an alkene. Dehydrohalogenation can transform 1,1,1 trichloroethane to 1,1-dichloroethene:
CC1 CH
3

C1 =CH
2

+ HCI

Tlte rate of dehydrohalogenation increases with increasing n u m b e r s of halogen


ions; h e n c e c o m p o u n d s that undergo substitution most slowly undergo dehydrohalogenation most rapidly. B r o m i n e ions are m o r e rapidly removed than chlorine ions in
these abiotic reactions. Burlinson, Lee, and Rosenblatt ( 1 9 8 2 ) found that b r o m i n e is
removed from d i b r o m o c h l o r o p r o p a n e six time faster than chlorine.
Oxidations and reductions are typically biologically mediated and require external
electron d o n o r s or acceptors. In s o m e cases polyhalogenated aliphatic c o m p o u n d s will
act as electron a c c e p t o r s and b e c o m e reduced.
O x i d a t i o n s include a-hydroxylation, which is the addition of an O H ~ radical to
an alkane in place of an H atom on a c a r b o n that also contains a halogen ion. T h e result
is the formation of a chlorinated alcohol. T h e a-hydroxylation of 1,1-dichloroethane
forms 1,1-dichloroethanol:
CH CHC1
3

+ H 0
2

CH CCl OH + 2 H
3

+ 2e"

T h e halogenated alcohol can then undergo the further loss of hydrogen from the
hydroxy! g r o u p and a halide to form an aldehyde:
H

H
H C
11

C OH
CI

CI

H C C = 0 + H* + ClII

Organic Compounds in Ground Water

323

Oxidation of a c a r b o n - c a r b o n d o u b l e b o n d can result in the formation of an e p o x y ;


the p r o c e s s i s called e p o x i d a t i o n :
CI

CI

CI

/
C=C

+ H,0

>

Cl

CI

\ / \ /
C C

/
H

+2H

+ 2e-

CI

T h e e p o x y is short-lived and under neutral pH c o n d i t i o n s can be o x i d i z e d to a


c a r b o x y l i c acid:
Cl

Cl

OH

\ / \ /
I I
cc + H,O C I c c = o + H* + cr
/
\
I
Cl

Cl

R e d u c t i o n s start with the removal of a halide ion by a r e d u c e d s p e c i e s , such as


a reduced transition metal or a transition metal c o m p l e x . T h e reduced s p e c i e s is thus
oxidized. T h e alkyl radical thus formed can react with a H

the departed X ~ . T h i s p r o c e s s i s called h y d r o g e n o l y s i s .


C H C C 1 + H + e"
CH CHCI

ion, which substitutes for

+ CP

Reductions c a n also o c c u r if there are halides on adjacent c a r b o n atoms. In this


c a s e the loss of a halogen from each c a r b o n atom creates an a l k e n e by formation of a
d o u b l e b o n d between the c a r b o n atoms. T h i s reduction i s called d i h a l o e l i m i n a t i o n
and can transform a chlorinated alkane into a chlorinated alkene. T h e p r o c e s s is illustrated
b y the transformation o f h e x a c h l o r o e t h a n e t o t e t r a c h l o r o e t h e n e
CC1 ~CC1
3

+ 2 e~

CCI =CCI
2

+ 2 CI"

Figure 7.13 illustrates the biologically mediated reductions that can o c c u r with
chlorinated ethane and e t h e n e s under hydrogenolysis and dihaloelimination. Also s h o w n
in the figure are the relative half-reaction reduction potentials calculated from the free
energy of the reaction by using the Nernst equation.
Environmental c o n d i t i o n s influencing the type and rate of the preceding reactions
include pH, temperature, state of oxidation or reduction, m i c r o o r g a n i s m s present, and
types of o t h e r chemicals present. Reaction kinetics also play an important role in the
determination of the abiotic and biotic fate of organic contaminants. Theoretically, the
end products o f the abiotic reactions are ethane and e t h e n e , which should b e a m e n a b l e
to further biodgradation. However, under field c o n d i t i o n s such a favorable o u t c o m e
might require many years to o c c u r . B e f o r e it d o e s , the contaminant might well How from
the point of origin to c o n t a m i n a t e a large area of the aquifer system.
7.6.4

Degradation of Organic Pesticides

O n e of the first organic pesticides developed was D D T , or dichloro-diphenyl-trichloroethane. This proved to be a very effective agent against a wide variety of insects, and
it was inexpensive. However it was found to be very resistant to degradation and h e n c e
persisted in the environment. D a m a g e to nontargeted wildlife, especially birds, was

IIDCE = 1,1 -dichloroethene


IH2-TECA = 1,1,1,2-tetrachloroethane
TCA = trichloroeihane
ll-DCA = 1.1 -dichloroelhane
HCA = hexachloroelhane
PCA = pcntachloroclhanc
I 22-TECA = 1,1,2,2,-lelrachlorocihanc
U2-TCA = 1,1,2-irichloroclhanc
12-DCA = 1,2 dichloroelhane
F I G U R E 7.13

CA = chloroeihane
A = ethane
PCE = perch loroethene
TCE = trichliirocihenc
tDCE = c/.v-dichlorocthcnc
tDCE = trans;-dichlor(K'lhcne
IIDCE = 1,1 -dichlorwihenc
VC = vinyl chkiridc
= ethylene

Pathways of chlorinated alkane and alkene reduction and estimated relative half-life

reduction potentials in volts. Source: Reprinted with permission from T. M. Vogel, C. S. Criddle, and P. L.
McCarty, Environmental Science and Technology 2 1 , no. 8, (1 9 8 7 ) : 7 2 2 - 3 6 . Copyright 1987 American
Chemical Society.

324

325

O r g a n i c Compounds i n G r o u n d W a t e r

reported, and the use of this c o m p o u n d was eventually b a n n e d in the United States.
Although D D T d o e s degrade, the first metabolite formed, dichloro-diphenyl dichloroe t h a n e ( D D D ) , proved to be at least as t o x i c as D D T . D D D forms from D D T by the
removal of a c h l o r i n e atom by hydrolysis. Figure 7.14 illustrates a m e t a b o l i c pathway for
the degradation o f D D T .

CIC CI

I
CI
Dichloro-diphenyl-trichloroel+iane

-K>
CI C C I

I
H
Dichloro-diphenyl-dichloroethane

O-rO
H C H

I
H
Dichloro-di phenyl-ethane

H C H

I
OH
Dichloro-diphenyl-ettianol

C = O

I
OH
Dichloro-diphenyl-acetic acid

FIGURE 7.14

Metabolic pathway for the degradation of dichloro-diphenyl-trichloroethane (DDT).

326

Chapter Seven

Newer pesticides are formulated so that they will not persist in the environment.
T h e y have structures that will m o r e readily undergo abiotic, or biologically mediated,
degradation. Smith ( 1 9 8 8 ) lists the following pathways by which herbicides are degraded.
Naturally, the s a m e reactions will apply to o t h e r pesticides.
D e a l k y l a t i o n : T h e biological removal of a methyl or other alkyl group from a
nitrogen atom ( s e e Figure 7 . 8 )
D e a l k o x y l a t i o n : T h e biological removal of a m e t h o x y (methyl e t h e r ) group
from a nitrogen atom.
D e c a r b o x y l a t i o n : The biological o r abiotic removal o f o n e c a r b o n and two
o x y g e n atoms from a carboxyl group.
D e h a l o g e n a t i o n : Biological replacement of a chlorine atom with a hydrogen
atom. This is especially important in degradation of insecticides, many of which
are chlorinated hydrocarbons.
E t h e r c l e a v a g e : T h e biological cleaving o f a n ether b y breaking the b o n d
b e t w e e n oxygen and c a r b o n atoms.
H y d r o l y s i s : T h e chemical o r biological cleavage o f molecules b y the addition
of water. T h e products usually include an alcohol or carboxylic acid.
H y d r o x y l a t i o n : T h e biological introduction o f hydroxyl groups into either
aliphatic or aromatic c o m p o u n d s .
M e t h y l a t i o n : T h e biological addition of a methyl group to an alcohol or
p h e n o l to form a methyl ether.
O x i d a t i o n : Oxidation of an alcohol to an aldehyde, which can be further
oxidized to a carboxylic acid, d o n e by either chemical or biological processes.
E p o x i d e formation is a major step in many oxidation p r o c e s s e s .
B e t a - o x i d a t i o n : Biological removal of two c a r b o n atoms from an alkane chain
linked to an aromatic ring structure.
R e d u c t i o n : Biological reduction of a nitrate group linked to a herbicide to
form an amine group.
R i n g c l e a v a g e : O n e o f many p r o c e s s e s b y which m i c r o b e s can break the
structure or an aromatic string.

7.7

Field Examples of Biological Degradation of Organic Molecules


7.7.1

Introduction

A vast b o d y of information of biological and abiotic degradation of organic molecules


has accumulated, based on laboratory studies using m i c r o c o s m s , the microbial equivalent
of an aquarium. Under such carefully controlled conditions, the disappearance of a
c o m p o u n d can be determined and the appearance of metabolites noted. However, there
have b e e n far fewer field studies of microbial degradation. It is far m o r e difficult to make
field studies and interpret the results. At the most basic level, is a c o m p o u n d disappearing
due to degradation, or is s o m e o t h e r p r o c e s s s u c h as volatilization, sorption or dilu

327

O r g a n i c Compounds i n G r o u n d W a t e r

tioninvolved? However, e n o u g h of a b o d y of information has accumulated to d o c u m e n t


the natural degradation of a variety of organic c o m p o u n d s in soil and g r o u n d water. In
Chapter 9 we will e x a m i n e h o w e n h a n c e d biodgradation can be used as a m e a n s of
remediating aquifers and soils contaminated with organic c o m p o u n d s .
7.7.2

Chlorinated Ethanes and Ethenes

Cline and Viste ( 1 9 8 5 ) n o t e d that at a field study of a solvent-recovery facility in Wisconsin,


wells downgradient from the site contained di- and m o n o c h l o r o ethanes and e t h e n e s ,
including \inyl chloride. However, the facility had never a c c e p t e d these products, which
w e r e interpreted t o b e the metabolites o f trichloroethane and trichloroethene.
Fetter ( 1 9 8 9 ) reported the distribution of chlorinated e t h a n e s and e t h e n e s in
g r o u n d water at the S e y m o u r Recycling C e n t e r Superfund site, a former solvent-recycling
facility in Indiana. T a b l e 7.6 contains the analyses of g r o u n d water from three wells:
203a, located b e n e a t h the facility; 2 0 6 a , located 50 ft down-gradient from the site b o u n d ary; and 2 0 ^ b , located 3 0 0 ft down-gradient from the site boundary. ( D a t a from Fetter
( 1 9 8 9 ) have b e e n updated in T a b l e 7 . 6 . ) T h e well beneath the site c o n t a i n e d trichloroethene, 1,1,1 trichloroethane, 1,2-dichloroethane, 1.1-dichloroethane, 1,1-dichloroethene,

rram-l,2-dichIoroethene,

chloroethane,

and

vinyl

chloride.

Chloroethane,

1.2 dichloroethane, 1,1-dichloroethane and r r a n s - l , 2 - d i c h l o r o e t h e n e have similar K

oc

values, ranging from 36 to 4 5 . T h e y would migrate at just a b o u t the s a m e rate. Well 2 0 6 a ,


located 50 ft from the site boundary, had higher c h l o r o e t h a n e , lower 1,1-dichloroethane
and trans -1.2 d i c h l o r o e t h e n e , variable vinyl chloride, and no detectable hits on the o t h e r
c o m p o u n d s . Well 2 0 7 b , located 2 5 0 ft down-gradient from well

206a, had elevated chlo-

roethane and no significant hits on any of the o t h e r c o m p o u n d s . Over a 6-yr monitoring


period the c h l o r o e t h a n e in this down gradient well increased from 41 /tg/L to a m a x i m u m
of 1 8 , 0 0 0 / r g / L T h i s is convincing evidence that the chlorinated ethanes and e t h e n e s
are being degraded, with c h l o r o e t h a n e being o n e e n d product. T h e m e c h a n i s m o f
degradation is not known, but the g r o u n d water at the site is reducing, having large
amounts of dissolved iron. T h e fate of the vinyl c h l o r i d e is unclear. It may be being
mineralized, it may be escaping through volatilization, or it may not be migrating d u e
to a high K

oc

value.

J a c k s o n , Priddle, and Lesage ( 1 9 9 0 ) investigated the transport and fate of CFC


113 (l,l,2-trichloro-l,2,2-trifluoroethane, or F r o n " ), a c o m p o u n d used as a solvent to
clean circuit b o a r d s and s e m i c o n d u c t o r s in the e l e c t r o n i c s industry. Field studies of a
landfill w h e r e this c o m p o u n d had b e e n d i s p o s e d s h o w e d that it was present in the
g r o u n d water along with two metabolites, indicating that biotransformation was occurring.
Reductive dechlorination transforms it to l,2-dichloro-l,2,2-trifluoroethane ( C F C - 1 2 3 a ) ,
which can undergo dihalide elimination to form l-chloro-l,2,2-trifluoroethene ( C F C 1 1 1 3 ) . T h e latter c o m p o u n d is quite toxic, a property not shared with the m o r e halogenated forms.
CI

CI

Cl

1 1

Cl

Cl

1 1

C C F
F

CFC-113

H
>

F C C F

CFC-123a

/
c=c

>

Cl
CFC-1113

328
TABLE

Chapter Seven
7.6

Seymour Recycl ng Corporation summary of well analyses for chlorinated ethanes and ethenes.

Compound

TCE

111TCA

152

155

11DCE

11 D C A

12DCA

217

45

36

39

42

f-DCE

CA

VC
8,400

W e l l 203a (Monitoring we II under site)


Aug. 84

nd

1.500

500

9,600

nd

16,000

2,300

1,100

Dec. 84

nd

nd

nd

800

200

200

4,000

2,700

Jun. 85

500

16,000

600

13,000

400

1 5,000

1,500

1,000

Jun. 9 0

54

690

190

4,500

<250

6,500

1,600

300

W e l l 2 0 6 a (Monitoring well 50 ft down gradient from site)


Aug. 84

nd

nd

nd

330

nd

75

9,500

nd

Dec. 84

nd

nd

nd

nd

nd

60

3,000

<60

Jun. 8 5

nd

nd

nd

5,600

nd

1,700

17,000

3,500

Feb. 88

nd

nd

nd

51

nd

52

1,600

nd

Jan. 8 9

nd

nd

nd

12J

nd

nd

800

nd

Aug. 8 9

nd

nd

nd

nd

nd

12J

6,200

nd

Jun. 90

nd

nd

nd

28

nd

22

1,500

nd

W e l l 207b (Monitoring well 3 0 0 ft down-gradient fron site)


Aug. 84

nd

nd

nd

nd

nd

41

nd

Dec. 84

nd

nd

nd

nd

nd

nd

200

nd

Jun. 8 5

nd

nd

nd

nd

nd

nd

800

nd

Feb. 88

nd

nd

nd

nd

nd

nd

5,800

nd

Jan. 8 9

nd

nd

nd

nd

nd

nd

1,400

nd

Aug. 8 9

nd

nd

nd

nd

nd

2J

18,000

nd

Jun. 90

nd

nd

13J

nd

nd

nd

13,000

nd

Data from U.S. EPA.


Key: TCE = trichloroethene
111TCA = 1,1,1-trichloroethane
11OCE = 1,1-dichloroethene
11DCA = 1,1-dichloroethane

7.7.3

12DCA = 1,2-dichloroelhone
t-DCE = frons-l,2-dichloroelhene

All analyses in microgfams/liler.


nd = not detected

CA = chloroethone
VC = vinyl chloride

J = estimated value

Aromatic Compounds

T h e r e is special interest in the biodgradation of aromatic c o m p o u n d s , b e c a u s e they


are the most water-soluble fraction of gasoline, which has leaked from underground
storage tanks into soil and ground water all over the world. B e n z e n e , toluene, o-, m-,
and p-xylene, trimethylbenzene, and ethylbenzene constitute the major aromatic components o f gasoline ( K r e a m e r and Stezenbach 1 9 9 0 ) .
B a r k e r and Patrick ( 1 9 8 5 ) and Barker, Patrick, and Major ( 1 9 8 7 ) injected ground
water spiked with 2.36 mg L b e n z e n e , 1.75 mg L toluene. 1.08 mg L p-xylene. 1 09 mg L
m-xylene, 1.29 m g / L o-xylene, and 1 2 8 0 mg L chloride into a sand aquifer. T h e aquifer
was aerobic, with 7 to 8 m g / L of dissolved oxygen at the start of the experiment. T h e
plume was sampled using a dense network of monitoring wells over a period of 4 3 4
days. All the aromatic c o m p o u n d s degraded, with the m and p-xylenes being the most
readily degraded, then the o-xylene, then toluene, and finally the benzene. Complete
removal of b e n z e n e occurred within 1.2 yr.

O r g a n i c Compounds i n G r o u n d W a t e r

329

Rifai et al. ( 1 9 8 8 ) o b s e r v e d that at a site w h e r e there was a spill of aviation gasoline,


t h o s e monitoring wells with high c o n c e n t r a t i o n s of dissolved o x y g e n in the g r o u n d
water had low a m o u n t s o f dissolved b e n z e n e , toluene, and xylene ( B T X ) , and t h o s e
with high B T X had low dissolved oxygen. This was interpreted to m e a n that a e r o b i c
d e c o m p o s i t i o n of the B T X was occurring, with the dissolved o x y g e n in the ground water
being c o n s u m e d in the p r o c e s s . Chiang et al. ( 1 9 8 9 ) also m a d e a study of the relationship
of dissolved o x y g e n in g r o u n d water to B T X in ground water. T h e y found a statistically
significant relationship b e t w e e n the a m o u n t o f dissolved o x y g e n and the c o n c e n t r a t i o n
o f dissolved B T X . T h e y o b s e r v e d that a minimum dissolved-oxygen c o n c e n t r a t i o n o f
0.9 m g / L i n g r o u n d water was necessary for c o m p l e t e a e r o b i c degradation o f B T X .
Degradation of p h e n o l and naphthalene in g r o u n d water has b e e n reported by
Ehrlich et al. ( 1 9 8 2 ) . C o n c e n t r a t i o n s of p h e n o l i c c o m p o u n d s in a contaminant plume
d e c r e a s e d from 30 m g / L to less than 0.2 m g / L over a distance of 4 3 0 m, w h e r e a s naphthalene d e c r e a s e d from 20 m g / L to 2 m g / L over the s a m e distance. Sodium, a conservative element, d e c r e a s e d only from 4 3 0 to 1 2 0 m g / L along the flow path, indicating
that dilution was r e s p o n s i b l e for s o m e b u t by no m e a n s a l l o f the decline. Methane
formation was occurring in the c o n t a m i n a t e d portion of the aquifer and m e t h a n o g e n i c
bacteria w e r e isolated there, indicating a n a e r o b i c d e c o m p o s i t i o n was a factor.
A e r o b i c degradation of p h e n o l and naphthalene has also b e e n o b s e r v e d in the
field ( K l e c k a et al. 1 9 9 0 ) . G r o u n d water b e n e a t h a site used for the disposal of waste
from a charcoal-making operation c o n t a i n e d 2 2 0 /tg/L p h e n o l , 5 7 0 tig/L 2-methylphenol
( c r e s o l ) , 8 6 0 /tg/L 2,4-dimethyl p h e n o l , and 2 2 0 / J g / L naphthalene. T h e native g r o u n d
water was a e r o b i c , with dissolved o x y g e n present in all areas e x c e p t the m o s t contaminated part o f the c o n t a m i n a n t plume. C o n c e n t r a t i o n s o f p h e n o l s and naphthalene
d e c r e a s e d to b a c k g r o u n d levels within 100 m down-gradient of the site. Laboratory studies
in m i c r o c o s m s using sediment from the aquifer d e m o n s t r a t e d that biodgradation of
phenol, c r e o s o l , and naphthalene could o c c u r under a e r o b i c conditions. C o m p u t e r
modeling confirmed that the major factor in reducing the p h e n o l s over a 1 0 0 m flowpath was biodgradation and not sorption.

7.8

A n a l y s i s of Organic Compounds in Ground Water


T h e r e are a n u m b e r of m e t h o d s by which organic c o m p o u n d s dissolved in g r o u n d water
can be identified and quantified. In g a s c h r o m a t o g r a p h y a mixture of volatile materials
is transported by a carrier gas through a c o l u m n p a c k e d with either an a b s o r b i n g solid
p h a s e or an a b s o r b i n g liquid p h a s e that is c o a t e d on a solid material. T h e volatile
c o m p o n e n t will partition b e t w e e n the carrier gas and the a b s o r b i n g phase, and the
length of time that it takes for the c o m p o n e n t to traverse the c o l u m n will be characteristic.
If there are several c o m p o u n d s in a mixture, the c o l u m n will separate t h e m and they
will arrive at different times. A detector, such as a p h o t o i o n i z a t i o n d e t e c t o r or a
f l a m e i o n i z a t i o n d e t e c t o r , located a t the e n d o f the c o l u m n can d e t e r m i n e the
quantity and identity o f e a c h volatile c o m p o n e n t . I n m a s s s p e c t r o m e t r y the c o m p o u n d
is ionized by an electrical discharge and the ions are then separated b a s e d on their
charge-to-mass ratio. T h e output i s a m a s s s p e c t r u m , which can b e c o m p a r e d t o the
mass spectra of a large n u m b e r of standard c o m p o u n d s stored in an e l e c t r o n i c data

330

Chapter Seven

b a s e called a mass-spectra library. T h e most c o m m o n l y used laboratory method is gas


chromatography c o m b i n e d with mass spectrometry, frequently known as G C / M S . In
G C / M S the mass spectrometry is p r e c e d e d by gas chromatography to separate the
c o m p o u n d s . Figure 7.15 shows a library' spectrum of b e n z e n e matched against the mass
spectrum of a sample identified as benzene.
T h e U.S. EPA has developed a series of standard m e t h o d s for the analysis of organicchemicals dissolved in water. T h e s e are known as the 600 series m e t h o d s ( 4 0 CFR Part
136, Federal Register. O c t o b e r 26, 1 9 8 4 ) . T a b l e 7.7 lists the 600 series and the target
compounds.
Method 601 for purgeable h a l o c a r b o n s and method 602 for purgeable aromatics
are gas-chromatograph m e t h o d s using a purge and trap p r o c e d u r e to isolate volatile
organics. Nitrogen is b u b b l e d through a column of water containing the dissolved
organics, so that the organics are purged from the water and carried with the gas.
T h e organics are trapped on T e n a x , a solid sorbent. T h e T e n a x is then heated, and the
volatile organics are swept into a gas chromatograph for separation and then detected.
Method 601 uses a halide-specific detector and m e t h o d 602 specifies a photoionization
detector. T h e results are reported as volatile organics or purgeable organics.
Methods 6 0 3 through 6 1 2 are gas-chromatographic methods for substances other
than purgeable hydrocarbons and aromatics. Method 6 1 3 is a GC MS method fordioxin.

Sample spectrum

r'i

ti

'

' ' '

Library spectrum of benzene

F I G U R E 7.15

Comparison of the mass spectrum of an unknown with the mass spectrum of benzene

from the library of mass spectra. Source: B. L. Roberts, Ground Water Monitoring Review 5, no. 4,
( 1 9 8 5 ) : 4 1 - 4 3 . Used with permission. Copyright 1985 Water W e l l Journal Publishing Co.

O r g a n i c Compounds i n G r o u n d W a t e r
TABLE

7.7

331

U . S . EPA 6 0 0 series analytical methods for organic compounds.

Method

Analytical

Number

Technique

Target Compounds

601

GC

Purgeable halocorbons

602

GC

Purgeable aromatics

603

GC

Acrolein and acrylonitrile

604

FIDGC

Phenols

605

HPLC

Benzidines

606

GC

Phthalate ester

607

GC

Nitrosamines

608

GC

Organochlorine pesticides and PCBs

609

GC

Nitroaromatics and isophorone

610

GC 8. HPLC

Polycyclic aromatic hydrocarbons

611

GC

Haloethers

612

GC

Chlorinated hydrocarbons

613

GC/MS

2,3,7,8-TCDD (dioxin)

624

GC/MS

Purgeable organics

625

GC/MS

Acid and base/neutral extractable organics

Key: GC = gos chromatography


FIDGC flame ionization detector gos chromotogrophy
HPLC = high-performance liquid chromatography
G C / M S gos chromatography/moss spectrometry

Many target c o m p o u n d s can be determined by either a GC or a G C / M S method. Method


6 2 4 is a G C / M S m e t h o d for purgeable organic c o m p o u n d s that includes c o m p o u n d s
also determined by GC m e t h o d s 6 0 1 , 6 0 2 , 6 0 3 , and 6 1 2 . M e t h o d 6 2 5 is a G C / M S m e t h o d
for semivolatile organics that includes c o m p o u n d s also detected by GC m e t h o d s 6 0 4 ,
6 0 6 , 6 0 7 , 6 0 9 , 610, 6 1 1 , and 6 1 2 .
B e c a u s e o f the larger n u m b e r o f target c o m p o u n d s , m e t h o d s 6 2 4 and 6 2 5 are
usually specified for organic analysis of contaminated water. T h e s e are s u p p l e m e n t e d
by m e t h o d 6 0 8 for pesticides and polychlorinated b i p h e n o l s and m e t h o d 6 1 3 for 2,3,7,8T C D D ( d i o x i n ) . If certain c o m p o u n d s are d e t e c t e d by m e t h o d 6 2 5 , they should then
be confirmed with a GC m e t h o d specific to that c o m p o u n d .
In m e t h o d 6 2 4 an inert gas, s u c h as nitrogen, is b u b b l e d through a 5 mL water
sample so that the volatile organics are transferred from the water phase to the vapor
phase. T h e vapors are swept to a s o r b e n t trap, w h e r e they are trapped on T e n a x . After
purging is c o m p l e t e , the T e n a x is heated and flushed with an inert gas to d e s o r b the
volatiles and drive them into a gas-chromatographic column. T h e purgeables are separated
in the c o l u m n and then d e t e c t e d with a mass s p e c t r o m e t e r .
Not all the organic c o m p o u n d s are purged from the water by the nitrogen gas.
T h e s e are the semivolatile c o m p o u n d s . Method 6 2 5 is for the analysis of e x t r a c t a b l e
semivolatiles. T h e pH of a 1 -L sample is adjusted so that it is greater than 11. T h e sample
is then extracted with methylene chloride. T h i s is k n o w n as the b a s e / n e u t r a l f r a c t i o n .
T h e pH of the sample is then adjusted to less than 2. T h e sample is again extracted with
methylene chloride to form the acid fraction. Each fraction is chemically dried to remove
the water and c o n c e n t r a t e d by distillation to 1 mL. T h e fractions are then analyzed
separately by G C / M S .

332

Chapter Seven

T h e U.S. EPA has created a target c o m p o u n d list of organic chemicals for its
contract laboratory program. T h e 1 2 6 c o m p o u n d s on this list are also known as p r i o r i t y
p o l l u t a n t o r g a n i c c o m p o u n d s . l a b o r a t o r i e s are required t o identify and quantify
any c o m p o u n d s on this list through m e t h o d s 6 0 8 , 6 2 4 , and 6 2 5 . T a b l e 7.8 presents the
target c o m p o u n d list. T h e first 34 c o m p o u n d s are volatiles determined by method 6 2 4 .
C o m p o u n d s 35 to 99 are semivolatiles determined through method 6 2 5 . Pesticides and
P C B s constitute c o m p o u n d s 100 to 1 2 6 and are detected through m e t h o d 6 0 8 .
T h e r e are a n u m b e r of additional EPA m e t h o d s for organic c o m p o u n d s . O n e
e x a m p l e is t o t a l p e t r o l e u m h y d r o c a r b o n s by m e t h o d 4 1 8 . 1 . In this method a liter
of sample is extracted with fluorocarbon 113 at a pH less than 2. Interfering c o m p o u n d s
are then removed with a silica gel a b s o r b e n t . An infrared analysis of the sample is made
by direct c o m p a r i s o n to a calibration curve made from prepared samples with a known
concentration.
T h e r e are also specific EPA m e t h o d s for analysis of organic c o m p o u n d s on a soil
matrix. This involves an extraction p r o c e d u r e followed by G C / M S analysis.
T h e analysis of water samples will frequently result in the analyst determining that
a particular c o m p o u n d was not found. T h e answer that is reported, however, is not zero.
It is instead reported as less than detection, with the d e t e c t i o n l i m i t given. T h e detection
limit is the lowest concentration level that can be determined to be statistically different
than a blank sample (American Chemical Society, 1 9 8 3 ) . A m e t h o d d e t e c t i o n l i m i t
i s specified for each o f the EPA 6 0 0 series analytical procedures. T h e l i m i t o f q u a n t i f i c a t i o n is the level above which the quantitative results may be e x p r e s s e d with a
specified d e g r e e of confidence. Values between the detection limit and the limit of
quantification may b e reported a s e s t i m a t e d v a l u e s .
Swallow, Shifrin, and Doherty ( 1 9 8 8 ) point out s o m e pitfalls in the analysis of
organic c o m p o u n d s in ground water. T h e G C / M S instruments are designed for automatic
operation, with the mass spectra matched electronically to a library spectrum. This match
may not be accurate. Swallow, Shifrin, and Doherty report o n e c a s e where the same
c o m p o u n d was reported as eight different c o m p o u n d s in 13 samples, even though the
spectra were virtually identical. T h e only differences were small and could be attributed
to b a c k g r o u n d interferences. If an e x p e r i e n c e d analyst had not e x a m i n e d the spectra,
the e r r o r would not have b e e n found. Reference standards are run on a limited set of
potential c o m p o u n d s , and the accuracy of matching is high with c o m p o u n d s that have
b e e n used as a reference. However, for many c o m p o u n d s for which a reference has not
b e e n run, the automated p r o c e d u r e may not get a g o o d match between the sample
spectrum and a library spectrum. If no match is found, then the sample will be reported
as an unknown. However, this could be a significant pollutant at the site under investigation and very important, even though a library match could not be made. In that
c a s e reference standards of unusual c o m p o u n d s that might have b e e n used in the area
would n e e d to be prepared to obtain additional spectra for the library.
T h e instruments are sensitive, and large concentrations of a c o m p o u n d may not
b e tolerated. S a m p l e d i l u t i o n may b e necessary for samples where o n e o r m o r e
c o m p o u n d s are in high concentration. T h e s e are reported as diluted samples. However,
dilution raises the detection limits for all c o m p o u n d s . A sample may require dilution
b e c a u s e of a high concentration of a m o r e or less benign c o m p o u n d , and the resulting

TABLE 7.8
Number

Torget compound list.


Chemical Name

Number

4-Methylphenol

45
46

n-Nltrosod in-Propylamine

86
87

47

Hexachloroethone
Nitrobenzene

88
89

48
49

Isophorone
2-Nitrophenol

90
91

50
51

2,4-Dimethylphenol

92

Chrysene
bis(2-Ethylhexyl) phthalate

Benzoic acid
bis(2-Chlofoethoxy) methane

93
94

di-n-Odyl phthalate
Benzol bjfluoranthene

2,4-Dichlorophenol
1,2,4-Trichlotobenzene

95
96

Benzol klfluoranthene
8enzo[alpyrene

55

Naphthalene

97

lndeno(l,2,3.cd)pyrene

4-Chloroaniline

98
99

Dibenz[a,h]anthracene

Bromomethane

3
4

Vinyl chloride
Chloroethane

Methylene chloride
Acetone

10
1 1
12
13
14

Corbon disulfide
1,1 -Dichloroethene
1,1 -Dichlotoethane
1,2-Dichloroethene
Chloroform
1,2-Dlchloroethane
2-Butanone

Chemical Name

44

43

8
9

Number
85

Chloromethane

6
7

Chemical Name
bis(2-Chloroisopropyl) ether

1
2

52
53
54

di'-n-Butyl phthalate
Fluoranthene
Pyrene
Butyl benzyl phthalate
3,3-Dichlorobenzidine
Benz[olanthracene

15

Corbon tetrachloride

56
57

16
17

Vinyl acetate
Bromodichloromethane

58
59

18
19

1,2-Dichloropiopone

60

Hexachlorocyclopentodiene

cis-1,3-Dichloropropene

61

2,4,6-Trichlorophenol

Trichloroethene

62

Chlorodi bromomethane
1,1,2-Trichloroethane

63
64

2-Nitroaniline

106

Benzene

65

Dimethyl phthalate

107

Endosulfan 1 (alpha)

trons-1,3-Dichloropropene
Bromoform

66

Acenaphthylene
2,6-Dinitrotoluene

108
109

Dieldrin
4,4-DDE

20
21
22
23
24

1,1,1 -Trichloroethane

Hexochlorobutadiene
4-Chloro-m-cresol
2-Methylnaphthalene

100
101
102

Benzo[g,h,i]perylene
BHC-alpha
BHC-beto
BHC-gamma (lindane)
BHC-delto

2,4,5-Trichlofophenol

103
104

2-Chloronophthalene

105

Aldrln
Heptochlor epoxide

Heptochlor

26
27

4-Methyl-2-pentanone

67
68
69

3-Nitroaniline
Acenaphthene

110
111

Endrin

2-Hexanone

28
29

Tetrachlotoethene

70

2,4-Dinitrophenol

112

4,4-DDD

Toluene

71

30

1,1,2,2-Tettochloroethane

72

4-Nitrophenol
Dibenzofuron

113
114

Endosulfan sulfate
DDT-P.P' (4,4'-DDT)

31
32

Chlorobenzene

73
74

2,4-Dinltrotoluene

Methoxychlor

33
34

Styrene
Xylene, total

75

35

Phenol

76
77

4-Chlorophenyl phenyl ether


Fluorene

115
116
117

36

bis(2-Chloroethyl) ether

78

37

2-Chlorophenol

79

38
39

1,3-Dichlorobenzene

80

1,4-Dichlotobenzene

40

Benzyl alcohol

41

1,2-Dichlorobenzene

42

2-Methylphenol

25

Ethyl benzene

Diethyl phtholote

4-Nitrooniline
4,6-Dinitro-2-methylphenol

118
119

Endosulfan 2 (beta)

Endrin ketone
Chlordane, alpha
Chlordone, beta
Toxaphene

120

PCB-1016 (Aroclor 1016)

N-Nittosodiphenylamine
4-Bromophenyl phenyl ether

121

PCB-1221 (Atoclor 1221)

122

PCB-1232 (Aiocloi 1232)

81
82

Hexochlorobenzene
Pentochlorophenol

123
124

PCB-1242 (Aroclor 1242)

83
84

Phenonthfene

125

PCB-1248 (Aroclor 1248)


PCB-1254 (Aroclor 1254)

Anthrocene

126

PCB-1260 (Aroclor 1260)

334

Chapter Seven

high detection limits may then mask the p r e s e n c e of an important c o m p o u n d . A


10:1 dilution will result in a 10-fold increase in the detection limit.
Quality assurance and quality control ( Q A / Q C ) are important parts of any analytical
program for organic chemicals. A quality-assurance and quality control program aims to
determine the a c c u r a c y , or c o r r e c t n e s s , of the data as well as the p r e c i s i o n , or
repeatability, of the analyses.
F i e l d and m e t h o d b l a n k s are used to detect if any organics are being inadvertently introduced during the sampling and analysis procedures. A field blank is a
sample of very pure water that is run through the sampling equipment and then put
into a bottle and returned to the lab for analysis. If all g o e s well, the field blank should
n o t contain any organic c o m p o u n d s . Method blanks are also used to determine the
purity of the solvents and reagents used in the analysis. Method blanks are samples of
distilled water that are analyzed using the solvents and chemicals called for in the
procedure.
If a field and or method blank contains organic c o m p o u n d s , these c o m p o u n d s
w e r e introduced during the sampling or analytical procedures. Certain organic comp o u n d s that are used in laboratories are frequently detected in the method blanks. T h e s e
c o m p o u n d s include a c e t o n e , methylene chloride, toluene, 2-butanone, d i butyl phthal
ate, di--octyl phthalate, and Ws(2-ethylhexyl) phthalate. If o n e of these c o m p o u n d s is
present in a field or m e t h o d blank, then the U.S. EPA's Laboratory Data Validation
Functional Guidelines specify that the detection limit for that c o m p o u n d be set at 10
times the greatest amount found in a blank For all other organic c o m p o u n d s that might
be found in a m e t h o d or field blank, the detection limit is set at 5 times the greatest
amount found in the blank.
S p i k e d s a m p l e s have a known amount of an organic c o m p o u n d added to water
and then run through the analytical process. If possible, uncontaminated ground water
from the site is used to prepare the spike. This b e c o m e s a standard solution; since the
initial concentration is known, a p e r c e n t r e c o v e r y of the analyte can be determined.
A perfect analysis will have a 1 0 0 % recovery. T h e Q A / Q C specifications for the lab
contract will determine an a c c e p t a b l e percent recovery for valid data. If too much or
t o o little of the spiked c o m p o u n d is recovered, the reported data are not valid.
Duplicate samples are used to validate the precision of the analysis. T h e lab may
take a sample from the field, split it into two or m o r e aliquots, and analyze them to
m a k e a l a b d u p l i c a t e . T h e hydrogeologist may also collect a duplicate or triplicate
sample in the field and submit it as a field duplicate. Most often field duplicate-, are
submitted as blind duplicates so that the analyst doesn't k n o w that a split sample was
submitted. Standard procedures generally call for 1 split sample to be collected for every
10 field samples.

7.9

Summary
T h e r e are a n u m b e r of important physical properties of organic chemicals that influence
their behavior. T h e s e include the melting point and boiling point, density, vapor pressure,
vapor density, octanol-water partition coefficient, and Henry's law constant. Organic
c o m p o u n d s are based on c a r b o n , which may be b o n d e d to hydrogen, halides, oxygen,

335

O r g a n i c Compounds i n G r o u n d W a t e r

nitrogen, p h o s p h o r u s , and sulfur. T h e m o s t s i m p l e o r g a n i c c o m p o u n d s are hydrocarb o n s , w h i c h have o n l y c a r b o n and hydrogen. C a r b o n a t o m s have four b o n d i n g l o c a t i o n s
and may form single, d o u b l e , and triple b o n d s with o t h e r c a r b o n a t o m s . T h e r e is a
formal system for naming o r g a n i c c o m p o u n d s , although many c o m p o u n d s have c o m m o n
n a m e s that a r o s e b e f o r e the formal system w a s d e v e l o p e d . Many o r g a n i c c o m p o u n d s
m a y b e d e g r a d e d b y c h e m i c a l and m i c r o b i a l m e a n s that r e d u c e the m o l e c u l a r weight
and c o m p l e x i t y o f t h e c o m p o u n d s . T h e e n d p r o d u c t o f degradation o f h y d r o c a r b o n s
i s c a r b o n d i o x i d e o r m e t h a n e , d e p e n d i n g u p o n t h e c o n d i t i o n s u n d e r w h i c h degradation
o c c u r s . O r g a n i c c o m p o u n d s dissolved i n w a t e r are m o s t c o m m o n l y analyzed b y gas
chromatography, mass spectrometry, or a combination of both.

References
American Chemical Society Committee on Environmental Improvement. 1983. Analytical Chemistry, vol. 55, pp.
2210-18.
Barker, J. F., and G. C. Patrick. 1985. Natural attenuation of
aromatic hydrocarbons in a shallow sand aquifer. Proceedings of the Conference on Petroleum Hydrocar
bons and Organic Chemicals in Ground Water:
Prevention, Detection and Restoration, 1 6 0 - 1 7 7 .
Dublin, Ohio: National Water Well Association.
Barker, J. F., G. C. Patrick, and D. Major. 1987. Natural attenuation of aromatic hydrocarbons in a shallow sand
aquifer. Ground Water Monitoring Review 7, no. 1;
64-71.
Barrio-Lage, G., F. Z. Parsons, and R. S. Nassar. 1987. Kinetics
of the depletion of trichloroethene. Environmental
Science and Technology 21, no. 4 : 3 6 6 - 7 0 .
Barrio-Lage, G., F. Z. Parsons, R. S. Nassar, and P. A. Lorenzo.
1986. Sequential dehalogenation of chlorinated eth
enes.
EnvironmentalScience and'Technology20, no.
1:96-99
. 1987. Biotransformation of trichloroethene in a variety of subsurface materials. Environmental Toxicology and Chemistry 6:571 - 7 8 .
Bouwer, E. J . , and P. L McCarty. 1983a. Transformations of
halogenated organic compounds under denitrification
conditions. Applied and Environmental Microbiology
.45, no. 4 : 1 2 9 5 - 9 9 .
. 1983b. Transformations o f 1 and 2-carbon aliphatic
organic compounds under methanogenic conditions.
Applied and Environmental Microbiology 4 5 , no.
4:1286-94.
Bumpus, J. A.. 1989. Biodgradation of polycydic aromatic
hydrocarbons by Pbanerochaete cJynsosporinm. Ap
plied and Environmental Microbiology 55,
no.
1:154-58.
Burlinson. N. E., L A. Lee, and D. H. Rosenblatt. 1982. Kinetics
and products of hydrolysis of l,2-dibromo-3-chloropropane. Environmental Science and Technolog}' 16,
no. 9 : 6 2 7 - 3 2 .
Chiang, C. Y., J. P. Salanitro, E. Y. Chai, J. D. Colhart, and C.
L Klein. 1989. Aerobic biodgradation of benzene,
toluene, and xylene in a sandy aquiferdata analysis

and computer modeling. Ground Water 27, no.


6:823-34.
Cline, P. V., and D. R. Viste. 1985. Migration and degradation
patterns of volatile organic compounds. Waste Management & Research 3 : 3 5 1 - 6 0 .
Egli, C, T. Tschan, R. Scholtz, A. M. Cook, and T. Leisinger.
1988. Transformation of tetrachloromethane to dichloromcthane and carbon dioxide by Acetobacterium
woodi.
AppliedandEnvironmentalMicrobiology
54, no. 1 1 : 2 8 1 9 - 2 4 .
Ehrlich, G. G., D. F. Goerlitz, E. M. Godsy, and M. F. Hult,
1982, Degradation of phenolic contaminants in
ground water by anaerobic bacteria: St. Louis Park,
Minnesota. Ground Water 20, no. 6 : 7 0 3 - 1 0 .
Evans, W. C. 1977. Biochemistry of the bacterial catabolism
of aromatic c o m p o u n d s in anaerobic environments.
Nature 2 7 0 : 1 7 - 2 2 .
Fathepure, B a b u Z., and S. A. Boyd. 1988. D e p e n d e n c e of
tetrachloroethylene dechlorination on methanogenic
substrate consumption by Methanosarcina sp. Strain
DCM. Applied and Environmental Microbiology 54,
no. 1 2 : 2 9 7 6 - 8 0 .
Fetter, C. W, 1989- Transport and fate of organic compounds
in ground water, in Recent Advances in GroundWater Hydrology Edited by J. E. Moore, A. A. Zaporozec, S. C. Csallany, and T. C. Vamey, 1 7 4 - 8 4 . St.
Paul, Minn.: American Institute of Hydrology.
Fliermans, C. B . , T . J . Phelps, D. Ringelberg, A. T. Mikell, and
D. C. White. 1988. Mineralization of trichloroethylene
by heterotrophic enrichment cultures. Applied and
Environmental Microbiology 54, no. 7 : 1 7 0 9 - 1 4 .
Freedman, D. L and J. M. Gossett. 1989. Biological reductive
dechlorination of tetrachloroethylene and trichloroethylene to ethylene under methanogenic conditions.
Applied and Environmental Microbiology 55, no.
9:2144-51.
Galli, Rene, and P. L McCarty. 1989a. Biotransformation of
1,1,1-trichloroethane, trichloromethane and tetrachloromethane by a Clostridium sp. Applied and
Environmental Microbiology 55, no. 4 : 8 3 7 - 4 4 .
. 1 9 8 9 b . Kinetics o f biotransformation o f 1,1,1iri chloroethane by Clostridium sp. strain TCAAIIB.

336

Chapter Seven

Applied and Environmental Microbiology 55, no. 4:


844-51.
Grbic-Galic, D., and T. M. Vogel. 1987. Transformation of
toluene and benzene by mixed methanogenic cultures. Applied and Environmental Microbiology 53,
no. 2 : 2 5 4 - 6 0 .
Guerin, W. F.. and G. E.Jones. 1988. Two-stage mineralization
of phenanthrene by estuarine enrichment cultures.
Applied and Environmental Microbiology 54, no.
4:929-36.
Haigier, B. E., S. F. Nishino, and J. C. Spain. 1988. Degradation
of 1,2-dichlorobenzene by a Pseudomonassp. Applied
and Environmental Microbiology 54, no. 2 : 2 9 4 - 3 0 1 .
Hall, A. E. 1988. Transformations in soil. In Environmental
Chemistry of Herbicides Edited by R Glover, vol. I,
1 7 1 - 2 0 0 . Boca Raton, Fla.: CRC Press.
Harker, A. R., and Young K i m . 1990. Trichloroethylene degradation by two independent aromatic-degrading
pathways in Alcaligenes eutropbus JMP134. Applied
and Environmental Microbiology 56, no. 4 : 1 1 7 9 - 8 1 .
Heitkamp, M. A., and C. E. Cemiglia. 1988. Mineralization of
porycyclic aromatic hydrocarbons by a bacterium isolated from sediment below an oil field. Applied and
Environmental Microbiology 54, no. 6 : 1 6 1 2 - 1 4 .
. 1989. Pofycyclic aromatic hydrocarbon degradation
by a Mycobacterium sp. in a microcosm containing
sediment and water from a pristine ecosystem. Applied
and Environmental Microbiology 55, no. 8 : 1 9 6 8 - 7 3 .
Heitkamp, M. A., J. P. Freeman, D. W. Miller, and Carl Cerniglia.
1988. Pyxene degradation by a Mycobacterium sp.:
Identification of ring oxidation and ring fission products. Applied and Environmental Microbiology 54,
no. 10:2556-65Hunt, J. M. 1979. Petroleum Geochemistry and Geology. New
Y o r k : W. H. Freeman and Company, 617 pp.
Jackson, R E., M. W. Priddle, and S. Lesage. 1990. Transport
and fate of CFC-113 in ground water. Proceedings of
Petroleum Hydrocarbons and Organic C/jemicals in
Ground Water. Prevention, Detection and Restoration, 1 2 9 - 4 2 . Dublin, Ohio: National Water Well Association.
Janssen, D. B . , A. Scheper, L Dijkhuizen, and B. Witholt. 1985.
Degradation of halogenated aliphatic compounds by
Xantlxibacter autotropbicus GJ10. Applied and Environmental Microbiology 49, no. 3 : 6 7 3 - 7 7 ,
Klecka, G M.. J. W. Davis, D. R Gray, and S. S. Madsen. 1990.
Natural bioremediation of organic contaminants in
ground water: Cliffs-Dow superfund site. Ground
Water 28, no. 4 : 5 3 4 - 4 3 .
Kleopfer, R D., D. M. Easley, B. B. Hass, T. G. Deihl, D. E.
Jackson, and C. J. Wurrey. 1985. Anaerobic dgrada
on of trichlorocihyiene in soil. Environmental Science and Technology 19, no. 3:277 - 8 0 .
Kohring, G-W, J. E. Rogers, and J. Wiegel. 1989. Anaerobic
biodgradation of 2,4-dichlorophenol in freshwater
lake sediments at different temperatures. Applied and
Environmental Microbiology 55, no. 2 : 3 4 8 - 5 2 .
Kreamer, D. K., and K . J . Stezenbach. 1990. Development of
a standard, pure-compound base gasoline mixture for
use as a reference in field and laboratory experiments.
Ground Water Monitoring Review 10, no. 2 : 1 3 5 - 4 5 .

laPat Polasko, L T . , P. L McCarty, and A. J. B. Zehnder. 1984.


Secondary substrate utilization of methylene chloride
by an isolated strain of Pseudomonas sp. Appliedand
Environmental Microbiology 47, no. 4 : 8 2 5 - 3 0 .
Little, C. D., A. V. Palumbo, S. E. Herbes, M. E. Lidstrom, R
L Tyndale, and P. J. Gilmer. Trichloroethylene biode
gradation by a methane oxidizing bacterium. Applied
and Environmental Microbiology 54, no. 4 : 9 5 1 - 5 6 .
Major, D. W., C. I. Mayfield, and J. F. Barker. 1988. Biotransformation of benzene by denitrification in aquifer
sand. Ground Water 26, no. 1:8-14.
Manahan, S. E. 1984. Environmental Chemistry, 4th ed. Boston: PWS Publishers, 6 1 2 pp.
McCarty, P. L, M. Reinhard, and B. E. Rittman. 1981. Trace
organics in groundwater. Environmental Science and
TecJmology 15, no. 1 : 4 0 - 5 1 .
Mihelcic, J. R, and R G. Luthy. 1988. Microbial degradation
of acenaphthene and naphthalene under denitrifica
tion conditions in soil-water systems. Applied and
Environmental Microbiology 54, no. 5 : 1 1 8 8 - 9 8 .
Mikesell, M. D., and S. A. Boyd. 1990. Dechlorination of chloroform by Metbanosarcina strains. Applied and En
vironmentat Microbiology 56, no. 4 : 1 1 9 8 - 1 2 0 1 .
Mueller, J. G., P. J. Chapman, B. O. Blattmann, and P. H.
Pritchard. 1990. Isolation and characterization of a
lluoranthene utilizing strain of Pseudomonas pauci
tnobis.
Applied and Environmental Microbiology
5 6 , no. 4 : 1 0 7 9 - 8 6 .
Nelson, M. J. K., S. O. Montgomery, W. R Mahaffey, and P.
H. Pritchard. 1987. Biodgradation of trichloroethylene and involvement of an aromatic biodegradatrve
pathway. Applied and Environmental Microbiology
5 3 . no. 5 : 9 4 9 - 5 4 .
Nelson, M. J. K., S. O. Montgomery, E. J. O'Neill, and P. H.
Pritchard. 1986. Aerobic metabolism of trichloroethylene by a bacterial isolate. Applied and Environmental Microbiology 52, no. 2 : 3 8 3 - 8 4 .
Nelson, M. J. K., S. O. Montgomery, and P. H. Pritchard. 1988.
Trichloroethylene metabolism by microorganisms
that degrade aromatic compounds. Applied and En
vironmental Microbiology 54, no. 2:604 - 6 .
Parsons, F., P. R Wood, and J. DeMarco. 1984. Transformations of tet rachloroethene and trichloroethene in
microcosms and groundwater. Journal of American
Water Works Association 76. no. 2 : 5 6 - 5 9 .
Pignatello. J. J. 1986. Ethylene dibromide mineralization in
soils under aerobic conditions. Appliedand Environmental Microbiology 5 1 , no. 3:588 - 9 2 .
Rirai, H. S . . P. B. Bedient, J, T. Wilson, K. M. Miller, andj. M.
Armstrong. 1988. Biodegradation modeling at aviation
fuel spill site. Journal of Environmental Engineering
114, no. 5 : 1 0 0 7 - 2 9 .
Roberts, P. V., J. Schreiner, and G. D. Hopkins. 1982. Field
I >! < irganii vvatei quality i hanges during gn iund
water recharge in the Palo Alto baylands. Water Research 1 6 : 1 0 2 5 - 3 5 .
Schwarzenbach, R P., W. Giger, C. Schaffner, and O. Wanner.
1985. Groundwater contamination by volatile halo
genated alkanes: Abiotic formation of volatile sulfur
compounds under anaerobic conditions. Environ
mental Science and Technology 19. no. 4 : 3 2 2 - 2 7 .

stud)

O r g a n i c Compounds i n G r o u n d W a t e r

Smith, Allen F. 1988. Transformations in soil. In Environ


menial Chemistry of Herbicides, edited by R. Grover.
171 - 2 0 0 . B o c a Raton, Fla.: CRC Press.
Strand. S, F.., and L Shippert. 1986. Oxidation of chloroform
in an aerobic soil e x p o s e d to natural gas. Applied and
Environmental Microbiology 52, no. 1 : 2 0 3 - 5 .
Swallow. K. C, N. S. Shifrin, and P.J. Doherry. 1988. Hazardous
organic c o m p o u n d analysis. Environmental Science
and Technology 22, no. 2 : 1 3 6 - 4 2 .
T o p p , Edward, and R. S. Hanson. 1990. Degradation of pen
tachlorophenol by a Flavobacterium species grown
in continuous culture under various nutrient limitations. Applied and Environmental Microbiology 56,
no. 2 : 5 4 1 - 4 4 .
Tsien, H C . G. A. Brusseau, R. S. Hanson, and L P. Wackett.
1989. Biodegradation of trichlortx-thylene by Methyl
osinus trichosporium O B 3 b . Applied and Environ
mental Microbiology 55, no. 1 2 : 3 1 5 5 - 6 1 .
Vogel, T. M, C. S. Criddle, and P. L McCarry. 1987. Trans
formations of halogenated aliphatic compounds. En
vironmental Science and Technology 2 1 , no. 8:
722-36.
Vogel, T. M, and P. L McCarry. 1985. Biotransformation of
tetrachlorthylene to trichloroethylene, dichloro
ethylene, \inyl chloride, and carbon dioxide under
methanogenic conditions. Applied and Environmental Microbiology 49, no. 5 : 1 0 8 0 - 8 3 .

337

. 1987a. Abiotic and biotic transformations of 1,1,1


trichlomethane under methanogenic conditions. En
i ironmental
Science
and
Technology
21,
12:1208-13.
. 1987b. Rate of abiotic formation of 1,1 dichloro
ethylene from 1,1,1-trichloroethane in groundwater.
Journal of Contaminant Hydrology 1:299 - 3 0 8 .
Vogel, T, M. and M. Reinhard. 1986. Reaction products and
rates of disappearance of simple bromoalkanes, 1,2
d i b r o m o p r o p a n e and 1,2-dibromoethane in water.
Environmental Science
and
Technology 20,
no.
10:992-97.
Wilson, J. T., L E. Leach, M. Henson, and J. N. J o n e s . 1986.
In situ biorestoration as a ground water remediation
technique. Ground Water Monitoring Review 6, no.
4:56-64.
Wilson, B. H., G. B. Smith, and J. F. Rees. 1986. Biotransfor
mations of selected alkyl benzenes and halogenated
aliphatic hydrocarbons in methanogenic aquifer material: A microcosm study. Environmental Science
and TecJmology 20, no. 1 0 : 9 9 7 - 1 0 0 2 .
Wilson, J. T. and B. H. Wilson. 1985. Biotransformation of
trichloroethylene in soil. Applied and Environmental
Microbiology 4 9 , no. 1 : 2 ^ 2 - 4 3 .

Chapter Eight

Ground Water and Soil Monitoring

8.1

Introduction
Methods of installing monitoring wells and collecting ground water samples have b e e n
developed with the specific intention of obtaining a representative sample of water from
an aquifer. T h e s e m e t h o d s minimize the potential for the introduction of contaminants
into the ground through the p r o c e s s of installing a monitoring well Wells and sampling
devices can be constructed of materials that have a minimum tendency to leach materials
into and s o r b c o m p o u n d s from the water sample. G r o u n d water samples can be collected
in such a m a n n e r that dissolved gases are not lost or e x c h a n g e d with the atmospheric
gases. Soil samples can also be collected for classification and chemical analysis.
Methods of collecting samples of soil water are also available. Soil gas sampling
can be d o n e to give an indication of areas w h e r e volatile organic c o m p o u n d s are contained
in the soil or ground water.

8.2

Monitoring Well Design


8.2.1

General Information

Monitoring wells are installed for a n u m b e r of different purposes. During the installation
of a monitoring well, a soil boring may be made or rock-core samples may be collected
to determine the basic geology of the site. Prior to the design of a well, it is necessary
to determine what its use will b e . S o m e purposes of monitoring wells include the
following:
Measuring the elevation of the water table
Measuring a Potentiometrie water level within an aquifer
Collecting a water sample for chemical analysis
Collecting a sample of a n o n a q u e o u s phase liquid that is less dense than water
Collecting a sample of a n o n a q u e o u s phase liquid that is m o r e d e n s e than water
Testing the permeability of an aquifer or aquiclude
Providing a c c e s s for geophysical instruments
Collecting a sample of soil gas
338

G r o u n d - W a t e r and S o i l M o n i t o r i n g

339

T h e use for which the well is intended will dictate the design. F o r e x a m p l e , if a
well is to be used for the collection of water samples, the casing must be large e n o u g h
to a c c o m m o d a t e the water-sampling device. However, the diameter should not be much
larger than the minimum size, b e c a u s e prior to the sampling of a well, stagnant water
must be removed from the casing; the larger the diameter of the casing, the greater the
volume of water that must be p u m p e d and properly disposed. T h e factors that should
be included in the design of a monitoring well include
T \ p e of casing material
Diameter of the casing
If there will be a well s c r e e n or an o p e n b o r e h o l e
Length of casing
Depth of the well
Setting and length of the well s c r e e n
Diameter of well s c r e e n
T y p e of material for well s c r e e n
Slot o p e n i n g of well s c r e e n
If an artificial filter pack (gravel p a c k ) is necessary
Gradation of filter pack (gravel p a c k ) material
Method of installation of well and s c r e e n
Material used to seal annular s p a c e b e t w e e n casing and b o r e h o l e wall
Protective casing or well vault

8.2.2

Monitoring Well Casing

All monitoring wells have a c a s i n g , w h e t h e r they have a s c r e e n or terminate in an o p e n


b o r e h o l e in b e d r o c k . T h e casing is a p i e c e of solid pipe that leads from the g r o u n d
surface to the well s c r e e n or o p e n b o r e h o l e and is intended to k e e p b o t h soil and water
from entering the well o t h e r than through the s c r e e n or o p e n b o r e h o l e . Casing also
prevents water from flowing from o n e aquifer horizon to another.
T h e diameter of the casing for a monitoring well is d e t e r m i n e d by the use for
which the monitoring well is planned. If the only p u r p o s e of the monitoring well is to
measure water levels, then a 1 in-inside diameter casing is all that is needed. An electric
p r o b e to m e a s u r e water level or a pressure transducer will fit inside the 1-in. casing.
Figure 8.1 s h o w s an electric p r o b e being lowered into a 2-in. casing.
If a well is to be used to collect a ground-water sample, the diameter of the well
needs to be such that standard well-sampling e q u i p m e n t can fit inside. T h e c o m m o n
standard for well-sampling e q u i p m e n t is a nominal 2-in. diameter. T h i s can a c c o m m o d a t e
a wide variety of p u m p s that can withdraw water at rates of 0.5 to 2 or 3 gal/min. Specially
designed b o r e h o l e geophysical e q u i p m e n t can also fit inside a 2-in. diameter casing.
S o m e states mandate the casing diameter for monitoring wells. F o r e x a m p l e , the Wisc o n s i n Department of Natural R e s o u r c e s requires a minimum inside diameter of 1.9 in.
and a m a x i m u m inside diameter of 4.0 in., w h e r e a s the New J e r s e y D e p a r t m e n t of
Environmental Protection requires a 4-in.-diameter well under all conditions.
F o r s o m e applications, monitoring wells may be intended for several functions such as measuring water levels, collecting water samples, pumping to remove

340

Chapter Eight

F I G U R E 8.1

Electric probe used to measure water levels in monitoring wells. Photo credit: Jim Labre.

contaminated water, and perhaps floating n o n a q u e o u s phase liquids and as a pan of a


vapor-extraction system. T h e s e wells generally have diameters larger than 2 in. to acc o m m o d a t e pumping equipment with a higher-flow capacity. T h e actual equipment to
be used determines the casing diameter.
Casing diameter can also be influenced by the depth of the well. T h e deeper the
well, the stronger the casing and screen must be to resist the lateral pressure at the final
depth and the crushing force of the weight of the length of casing. Larger diameter
casing can be made with thicker walls to have greater strength. It is easier to have a
straight well with stronger casing. Straight wells are important in accommodating bailers
and pumps.
T h e outside diameter of casing is standard; however, the inside diameter is a
function of the wall thickness. T a b l e 8.1 lists the wall thickness and inside diameter for
various schedules of casing. Heavier-schedule casing is stronger b e c a u s e it has a thicker
wall. T h e strength of a casing also d e p e n d s upon the material from which it is constructed.
A schedule 5 casing made of stainless steel is stronger than a schedule 40 casing made
of polyvinyl chloride ( P V C ) , yet leaves a greater inside diameter.

TABLE

8.1

Dimensions of inside and outside dian eters of well cas ngs.


Schedu e 5

Sched ule 1 0

Schedule 4 0

Schedule 8 0

Outside

Wall

Inside

Wall

Inside

Wall

Inside

Wall

Inside

Pipe S i z e

Diameter

Thickness

Diameter

Thickness

Diameter

Thickness

Diameter

Thickness

Diameter

Nominal 2 "

2.375"

0.065"

2.245"

0.109"

2.157"

0.154"

2.067"

0.218"

1.939"

Nominal 3"

3.500"

0.083"

3.334"

0.120"

3.260"

0.216"

3.068"

0.300"

2.900"

Nominal 4 "

4.500"

0.083"

4.334"

0.120"

4.260"

0.237"

4.026"

0.337"

3.826"

Nominal 5'

5.563"

0.109"

5.345"

0.134"

5.295"

0.258"

5.047"

0.375"

4.813"

Nominal 6'

6.625"

0.109"

6.407"

0.134"

6.357"

0.280"

6.065"

0.432"

5.761"

342

Chapter Eight

T h e r e are a n u m b e r of materials used to make well casings and screens. T h e s e


materials vary in chemical inertness, strength, durability, ease of handling, and cost. O n e
must always consider the intended use of the monitoring well before selecting a material.
What is the chemistry of the ground water and associated contaminants? Will any compounds present in the ground water react with any of the possible casing materials?
H o w d e e p will the well be; what are the strength requirements? Is the well intended for
a short term monitoring project or will it remain in service for many years?
Well casings are available in the following materials: fluoropolymers, such as PTFE,
or polytetrafluoroethylene ( T e f l o n * is the brand name of o n e manufacturer of P T F E ) ,
mild steel, stainless steel, galvanized steel, fiberglass, PVC, and polypropylene. Mild or
galvanized steel is often used for water-supply well casings but is not as frequendy found
in monitoring wells b e c a u s e it may react with the ground water to leach metals from
the casing ( B a r c e l o n a , G i b b , and Miller 1 9 8 3 ) . Polypropylene is not widely available.
Most monitoring wells are made of stainless steel or PVC, with PTFE being less c o m m o n .
PVC casing is the least expensive. Relative casing costs for other materials, c o m p a r e d
with PVC, are mild steel = 1.1, polypropylene = 2.1, type 3 0 4 stainless steel = 6.9, type
3 1 6 stainless steel = 11.2, and P T F E = 20.7. T y p e 3 1 6 stainless steel is m o r e resistant
to c o r r o s i o n than type 3 0 4 under reducing conditions (Aller et al. 1 9 8 9 ) .
Stainless steel has the greatest strength, followed by mild steel. B o t h are also resistant
to heat, but they are heavier than the plastics and are, therefore, m o r e difficult to install.
T h e lower strength of the plastics is c o m p e n s a t e d for by using a heavier schedule casing
that necessary with steel. Most monitoring wells are shallow enough that schedule 40
or 80 PVC has sufficient strength. PTFE is m o r e brittle and has less wear resistance than
PVC or polypropylene and is h e n c e less durable. PTFE also has a low tensile strength
and high weight per unit length, which limits its use to shallow depths. Even there, PTFE
casing tends to b o w under its weight when installed in monitoring wells and may not
be straight and plumb. Although its nonstick properties are g o o d in frying pans, the
neat c e m e n t grout used to seal the annular s p a c e between the casing and the b o r e h o l e
may not b o n d to the P T F E casing (Nielsen 1 9 8 8 ) .
In the selection of casing material for ground-water monitoring wells, we must
c o n s i d e r the potential chemical reactions between the casing material and the ground
water. Ideally, casing material should neither leach matter into water n o r s o r b chemicals
from water.
Reynolds and Gillham ( 1 9 8 5 ) studied the sorption from aqueous solution of five
halogenated organic c o m p o u n d s by several polymer materials. T h e organic c o m p o u n d s
used were 1,1,1-trichloroethane, 1,1,2,2-tetrachloroethane, hexachloroethane, perchlo
roethene, and b r o m o f o r m . T h e materials tested were PVC, PTFE, nylon, polypropylene,
polyethylene, and latex rubber. Nylon, polypropylene, polyethylene, and latex rubber
rapidly a b s o r b e d all five c o m p o u n d s . PVC a b s o r b e d all the c o m p o u n d s but 1.1.1 trichlo
roethane, although the rate of absorption was low. PTFE a b s o r b e d all the c o m p o u n d s
but b r o m o f o r m ; although the rate of adsorption of three of the four c o m p o u n d s was
low, PTFE a b s o r b e d 5 0 % of the perchloroethylene in 8 hr.
Parker, Hewitt, and J e n k i n s ( 1 9 9 0 ) evaluated the suitability of PVC, PTFE, stainless
steel type 3 0 4 (SS 3 0 4 ) , and stainless steel type 3 1 6 (SS 3 1 6 ) as casing material for
monitoring metals in ground water. They evaluated the interaction of four trace elements

G r o u n d - W a t e r and

Soil

343

Monitoring

that are of c o n c e r n in ground-water studies: arsenic, cadmium, chromium, and lead. T h e


metals w e r e tested at c o n c e n t r a t i o n s of 50 and 1 0 0 ng/L dissolved in ground water.
Figure 8.2 s h o w s the results of this study. If the concentration relative to control remains
at 1 . 0 , there is no interaction; if it d r o p s to less than 1 . 0 , then the e l e m e n t is sorbing
o n t o the casing material; and if it rises a b o v e 1 . 0 , the e l e m e n t is being leached from the
casing. T h e P T F E was the m o s t inert with respect to the metals, and the PVC was m u c h
better than either SS 3 0 4 or SS 3 1 6 .

1.2

'

1
PVC.

PTFE

S S 304

~~~ ^z
M

0.8

S S 316

0.8

0.4

Time (hr)

40
Time (hr)
(b)

(a)

S S 304

=-0
PVC,

PTFE

PTFE

o
PVC

S S 316

S S 304
6 ,

0.4

1
40

1
40

T i m e (hr)

T i m e (hr)

(c)

(d)

F I G U R E 8.2

Sorption and leaching of arsenic, cadmium, chromium, and lead by well casings made from

PVC, P T F E , type 3 0 4 stainless steel, and type 3 1 6 stainless steel. Source: L. V. Parker, A. D. Hewitt, and
T. F. Jenkins, Ground Water Monitoring Review 10, no. 2 (1 9 9 0 ) : 1 4 6 - 5 6 . Used with permission. Copyright
1990 Water W e l l Journal Publishing Co.

344

Chapter Eight

T h e interaction of several organic c o m p o u n d s with the same well casing materials


was also studied by Parker, Hewitt, and J e n k i n s ( 1 9 9 0 ) . Ten organic c o m p o u n d s were
tested, including chlorinated ethenes, c h l o r o b e n z e n e s , nitrobenzenes, and nitrotoluenes.
None of the c o m p o u n d s was s o r b e d o n t o either type of stainless steel. Many of the
c o m p o u n d s were s o r b e d by the plastic casings, with the PTFE sorbing at a greater rate
than the PVC. T h e amount and rate of sorption varied by c o m p o u n d . Figure 8.3 shows
the sorption of trichloroethene by the four casing types. Clearly, stainless steel is the
material of c h o i c e for monitoring organics, and PTFE is to be avoided. F o r a c o m p r o m i s e
material for monitoring both organics and inorganics, PVC appears to be the best. It
also has the appeal of having the lowest cost. PVC manufactured specifically for well
casing should be used, and it should carry the designation NSF wc, which indicates that
the casing c o n f o r m s to National Sanitation Foundation Standard 14 for potable water
supply (National Sanitation Foundation 1 9 8 8 ) .
However, PVC should be avoided if certain organic c o m p o u n d s are present in the
ground as n o n a q u e o u s phase liquids. It is reportedly soluble in low molecular-weight
ketones, aldehydes, amines and chlorinated alkanes, and alkenes ( B a r c e l o n a , G i b b , and
Miller 1 9 8 3 ) . Hkewise, PVC casing should also never be joined with solvent glued joints.
T h e s e solvents include c o m p o u n d s such as methylethylketone and tetrahydrofuran and
they may leach into ground water samples. Threaded joints that are machined directly
o n t o the PVC are the preferred m e t h o d of joining casing sections and casing to screen.
J o i n t s should be flush on the inside of the casing to prevent equipment being lowered
into the casing from hanging up in a projecting joint.

F I G U R E 8.3

Sorption of trichloroethene from ground water by PVC, PTFE, type 3 0 4 , and type 3 1 6

stainless steel well casings. Source: L. V. Parker, A. D. Hewitt, and T. F. Jenkins, Ground Water Monitoring
Review 10, no. 2 (1 9 9 0 ) : 14656. Used with permission. Copyright 1990 Water W e l l Journal Publishing Co.

G r o u n d - W a t e r and

8.2.3

Soil

Monitoring

345

M o n i t o r i n g Well Screens

If the monitoring well terminates in an unconsolidated formation, a s c r e e n is necessary


to allow the water to enter while keeping the sediment out. In m o s t monitoring well
applications, the well s c r e e n is the s a m e diameter as the casing to which it is attached
by a threaded coupling. Likewise, the well s c r e e n is normally m a d e of the s a m e material
as the casing. T h e considerations that go into deciding the material to use for the casing
also apply to the s c r e e n .
T h e s c r e e n will have o p e n i n g s to permit the water to enter. Manufactured well
s c r e e n should always be used rather than hand-cut slots or drilled h o l e s in plastic pipe.
T h e two c o m m o n s c r e e n s for monitoring wells are slotted pipe, which is available in
PVC and P T F E , and c o n t i n u o u s wire wrap, which is available in stainless steel. Figure
8 t illustrates these two s c r e e n types.
T h e width of the slot or wire-wrap o p e n i n g is precisely controlled during the
manufacture of the screen; the s c r e e n is available in a variety of o p e n i n g sizes, generally
ranging from 0 . 0 0 8 to 0 . 2 5 0 in. A screen with an o p e n i n g of 0 . 0 1 0 is referred to as a 10-

F I G U R E 8.4

Slotted and continuous wire-wrapped monitoring well screens.

346

Chapter Eight

slot screen. Many manufacturers carry only a limited n u m b e r of slot sizes in s t o c k f o r


e x a m p l e , 10- and 20-slot. Since the casing and screen are typically ordered in advance
of the well construction, the hydrogeologist usually has settled on a standard design
prior to going on the j o b .

8.2.4

Naturally Developed and Filter-Packed W e l l s

T h e casing and screen may be placed in the b o r e h o l e and the native sediment allowed
to cave around the screen. T h i s is called a n a t u r a l l y d e v e l o p e d w e l l and is often
used in sandy sediment with very limited amounts of silt and clay present. At least 9 0 %
of the sediment should be retained on a 10-slot screen b e f o r e a naturally developed
well is c o n s i d e r e d (Aller et al. 1 9 8 9 ) . W h e n water is withdrawn from such a well, it may
initially be cloudy due to suspended silt and clay, but the water should eventually clear
as the fines near the s c r e e n are removed by a p r o c e s s called well development. In a
naturally developed well the slot size is selected to allow s o m e of the fine sediment to
enter the well during development; this leaves only the c o a r s e r sediment outside the
screen.
In designing a water well, it is very important that the well be hydraulically effective
i.e., there should be a minimal loss of energy as the water flows into the well. T h e
selection of the slot o p e n i n g for naturally developed water wells is very important and
is b a s e d on a grain-size distribution curve of the sediment o p p o s i t e the well screen.
Monitoring wells are designed to retain much m o r e of the natural formation than water
wells b e c a u s e they are m u c h m o r e difficult to develop (Driscoll 1 9 8 6 ) . Monitoring wells
are not usually designed with the precision necessary for a water supply well. T h e well
should be hydraulically efficient as well as being as clear of silt and clay as possible. If
preliminary investigations indicate that the aquifer to be monitored has reasonably c o a r s e
sand or gravel and few fines, a standard slot size may be preselected for all the monitoring
wells. Ten-slot s c r e e n is frequently used under these conditions.
If the formation is c o h e s i v e t h a t is, has a high clay c o n t e n t o r if it is sandy with
a high silt content, it will be necessary to use an a r t i f i c i a l f i l t e r p a c k . Filter pack
material is medium to c o a r s e sand that is predominate!)' silica with no carbonates. It is
mined and graded to have a specific grain-size distribution. Manufactured filter pack
material c o m e s washed and bagged and is far preferable to native sand as artificial filter
pack. T h e filter-pack material is placed in the b o r e h o l e opposite the well screen. Its
p u r p o s e is to stabilize the natural formation and k e e p it out of the screen. This will
reduce the amount of silt and clay that enters the well when it is developed.
T h e grain size of the filter pack material is based on the nature of the formation
o p p o s i t e the screen. If the formation is fine sand, then the grain size distribution is
determined. T h e filter pack material should have an average grain size that is twice the
average grain size of the formation and have a uniformity coefficient (ratio of 4 0 % retained
size to 9 0 % retained s i z e ) between 2 and 3 (Driscoll 1 9 8 6 ) . T h e screen-slot opening is
then selected to retain 9 0 % of the filter pack. T h e minimum practical slot size for
monitoring well screens is 0 . 0 0 8 in. Figure 8.5 shows a grain-size distribution curve for
a filter-pack material designed for an eight-slot screen. If the monitoring well is in silt
or clay, all o n e can do is install an 8 slot screen and appropriate filter pack.

G r o u n d - W a t e r and

Soil

Monitoring

347

Slot opening and grain size (thousandths of an inch)


F I G U R E 8.5

Grain-size distribution curve used to select an eight-slot screen for a monitoring well.

T h e filter-pack material should be 2 to 3 in. thick. T h i s m e a n s that a 2-in.-diameter


well s c r e e n should be installed in a b o r e h o l e 6 to 8 in. in diameter. T h e filter pack material
is normally e x t e n d e d 2 or 3 ft a b o v e the top of the well s c r e e n to allow for settlement
of the material during development.

8.2.5

A n n u l a r Seal

T h e a n n u l a r s p a c e i n the b o r e h o l e a b o v e the f i l t e r pack must b e sealed t o prevent


the m o v e m e n t of surface water downward to the filter pack. It may also be sealed to
prevent vertical m o v e m e n t of g r o u n d water from o n e z o n e to another or to isolate a
discrete sampling zone. T h e seal should be m a d e of a material that has a low permeability,
b o n d s well to the natural formation and the casing, and e x p a n d s after it has b e e n
e m p l a c e d to e n s u r e a tight seal. It should set up within a day or so and be durable and
permanent.
Materials typically used for an annular seal are b e n t o n i t e pellets, granular b e n t o n i t e
slurry, neat c e m e n t grout, bentonite-sand slurry, and neat c e m e n t grout with a powdered
b e n t o n i t e additive.
Neat c e m e n t grout is a mixture of 94 lb of type I Portland c e m e n t with 5 to 6 gal
of water. Granular b e n t o n i t e slurry is a mixture of 30 lb of untreated b e n t o n i t e p o w d e r
m i x e d with 125 lb of untreated b e n t o n i t e granules with 1 0 0 gal of water. Bentonitec e m e n t grout is a mixture of 5 lb of untreated p o w d e r e d b e n t o n i t e with 94 lb of type

348

Chapter Eight

I Portland c e m e n t and 5 to 6 gal of water. B e n t o n i t e sand slurry is a mixture of 55 lb


of untreated powdered bentonite with 100 gal of water and 10 to 2 5 % sand by volume
to make a slurry that weighs 12 lb/gal. All water used to make these slurries should be
from a s o u r c e that is fresh and known to be uncontaminated and free from floating oil.
B e n t o n i t e is a clay containing at least 8 5 % sodium montmorillonite; it will swell
to several times its original volume when thoroughly hydrated. This hydration takes place
b e l o w the water table. However, bentonite has a high cation-exchange capacity and can
affect the chemistry of water that c o m e s into contact with it. Portland c e m e n t is used
to make c e m e n t grout. W h e n Portland c e m e n t cures, it is highly alkaline and can affect
the pH of ground water that c o m e s into contact with it. Neat c e m e n t grout will shrink
by at least 1 7 % when it cures. T h e addition of bentonite to make a bentonite-cement
grout significantly reduces the shrinkage problem. If neat c e m e n t grout or bentonite
c e m e n t grout is used, the casing material should be either stainless steel or schedule 80
PVC d u e to the heat generated as the c e m e n t cures.
T h e materials available for an annular seal are not ideal. Although they can be used
to m a k e an i m p e r m e a b l e seal, there is a c h a n c e they might affect ground-water quality
in their immediate vicinity. This p r o b l e m is mitigated if 2 ft of fine sand is placed in the
annular s p a c e above the filter-pack material or native sand opposite the screen. This
keeps the annular seal material from c o m i n g into contact with the water entering the
well screen.
Many hydrogeologists place a 2- or 3-ft layer of bentonite pellets above the fine
sand if the pellets will be b e l o w the water table. T h e pellets will swell and keep the
grout material from entering the filter pack material. If the top of the 2-ft fine-sand seal
is above the water table, then 2 ft of granular bentonite may be placed prior to the
addition of the annular seal.
8.2.6

Protective Casing

In o r d e r to provide physical protection for the investment in a costly monitoring well,


as well as to protect from vandalism by individuals accidentally or intentionally putting
foreign fluids and o b j e c t s into a monitoring well, a locking protective steel casing or
well vault is needed.
A protective casing e x t e n d s several feet above the ground surface. It extends above
the t o p of the monitoring well and has an inside diameter sufficiently large so that the
hydrogeologist can reach inside and unscrew a cap from the monitoring well. It is set
into a surface c e m e n t seal. For monitoring wells installed in freezing climates, a drain
h o l e at the bottom of the surface casing is desirable to prevent accumulation of moisture
that could freeze in the annular s p a c e between the protective casing and the monitoring
well. ( T h e author has seen a stainless-steel monitoring well casing pinched shut by water
that accumulated in a protective casing without a drain h o l e and then froze!)
In s o m e applications, it is not practical to have a monitoring well that extends
above g r o u n d f o r e x a m p l e , in the driveway at a gas station. T h e r e are small well vaults
available that can be used for protection for monitoring wells. However, they should be
in places that are not going to flood; otherwise floodwaters could enter the aquifer via
the monitoring well. If a well vault is used in a gas station or similar location, it should
be clearly marked and should be distinctive from the fillers for underground storage

G r o u n d - W a t e r and

Soil

Monitoring

349

tanks so that an inattentive p e r s o n d o e s n ' t try to fill it with gasoline! A locking well cap
without a vent h o l e should also be used.
8.2.7

Screen Length and Setting

T h e hydrogeologist must d e c i d e on the length of the screen and the depth to which it
will be set, b a s e d on the objectives of the monitoring program. Objectives could include
monitoring the position of the water table, measuring the potentiometric head at s o m e
depth in the aquifer, collecting representative water samples from various d e p t h s in the
aquifer, and detecting both light and d e n s e n o n a q u e o u s phase liquids. Moreover, monitoring might be intended to detect the migration of g r o u n d water containing c o n t a m
inants into an aquifer or evaluating the effectiveness of removing contaminants from an
aquifer. All might require different a p p r o a c h e s .
T o m o n i t o r the position o f the water table o r t o detect the p r e s e n c e o f LNAPLs,
the s c r e e n must be set so that it intersects the water table. T h e s c r e e n must be long
enough to intersect the water table over the range of annual fluctuation. In addition, the
s c r e e n must be long e n o u g h so that when the water table is at its greatest depth b e l o w
the land surface, there is e n o u g h of the s c r e e n remaining b e l o w the water table to
contain sufficient water for a water sample. A water table monitoring well will also be
able to detect the p r e s e n c e of light n o n a q u e o u s phase liquids. In m o s t applications the
minimum length of the s c r e e n for a water t a b l e - m o n i t o r i n g well is 10 ft with 5 ft a b o v e
and 5 ft b e l o w the water table. If the water table has m o r e than 5 ft of annual fluctuation,
a longer well s c r e e n is needed. However, s o m e states specify a m a x i m u m s c r e e n length
o f 1 0 ft. Figure 8.6 s h o w s e x a m p l e s o f incorrect ( a and b ) and c o r r e c t ( c ) p l a c e m e n t
of a multipurpose monitoring well intended to m e a s u r e the position of the water table,
detect floating n o n a q u e o u s phase liquids, and collect water samples from the upper part
o f the aquifer.
If the p u r p o s e of a monitoring well is to m e a s u r e the potentiometric pressure at
s o m e depth in the aquifer, then the well is called a p i e z o m e t e r . A p i e z o m e t e r should
have a relatively short s c r e e n length, 2 to 5 ft, so that the pressure that is r e c o r d e d is
representative of only a small vertical section of the aquifer. A p i e z o m e t e r can also be
used to collect ground-water samples that are representative of a small vertical section
of the aquifer.
Monitoring wells utilized to collect g r o u n d water samples should be designed with
respect to a specific ground-water monitoring goal. T h e concentration of g r o u n d water
contaminants can vary vertically If a monitoring well has a long well screen, it has a
greater probability of intersecting a plume of contamination. However, a water sample
taken from such a well may draw water from b o t h contaminated and uncontaminated
parts of the aquifer, resulting in a reported concentration that is less than that of the
ground water in the plume. T h i s is illustrated in Figure 8.7.
T h e collection of such unrepresentative water samples may have serious implications for the implementation of ground water regulations. In monitoring ground water
in o r d e r to find the actual concentration of contaminants in a plume, it may be necessary
to u s e several p i e z o m e t e r s s c r e e n e d at different depths at the s a m e location. T h i s is
expensive, not only due to the initial c o s t of the wells but also due to the c o s t s of
multiple chemical analyses for each round of sampling. However, such a configuration

350

Chapter Eight

Seasonal
high water
table

Seasonal
low water
table ' . "

Not to scale

" .

'

' .
(a)

F I G U R E 8.6

* .

' . \ ' . \ ' . \ '

", "

. *,
(b)

". * . ". " . ". * - ".


(c)

(a) Incorrect placement of wafer table-monitoring well screen. Seasonally high water table

is above the top of the screen and floating, nonaqueous phase liquids would be above the screen and
not detected, (b) Incorrect placement of water tablemonitoring well screen. Seasonally low water table
is so far down in well that there is not enough water in well to collect a sample for chemical analysis.
(The water table elevation could still be determined.) (c) Correct length and placement of water tablemonitoring well screen.

will yield the greatest a m o u n t of information a b o u t the hydraulic head as well as the
water quality.
If a m o n i t o r i n g well is intended to serve a.s warning that a p l u m e of contamination
is e s c a p i n g from a potential s o u r c e , then it should be s c r e e n e d in the m o s t p e r m e a b l e
parts of the aquifer. G r o u n d water and c o n t a m i n a n t s that it may be carrying not only

G r o u n d - W a t e r and

Soil

Monitoring

351

Waste source

F I G U R E 8.7

Effect of monitoring well-screen length on water-quality sampling. Monitoring well A is

fully screened through the thickness of the aquifer. It intersects the plume of leachate but the reported
concentration will be less than the actual concentration as water is withdrawn from both contaminated
and uncontaminated parts of the aquifer. Piezometer B is also screened to intersect the plume of leachate.
The reported concentration will be representative of the leachate. Piezometer C and water table monitoring
well D don't intersect the plume, indicating that it is deep in the aquifer.

preferentially travel through the most p e r m e a b l e material but travel faster there as well.
H e n c e , the leading e d g e of a p l u m e of contamination will follow the m o s t p e r m e a b l e
pathway.
If the p l u m e of contaminated water is following a z o n e or direction of high hydraulic
conductivity', it may flow in a direction that is not parallel to grad h. This may mean that
the location of the plume is not exactly down-gradient from the source.
On the o t h e r hand, if an aquifer is contaminated and a monitoring well has b e e n
installed to m o n i t o r the progress of a remediation effort, the well should not be s c r e e n e d
in the m o s t p e r m e a b l e part of the aquifer. In p u m p and treat systems, the water will
preferentially travel through and flush out the m o r e p e r m e a b l e zones. A well s c r e e n e d
in a p e r m e a b l e z o n e may indicate that the aquifer is rapidly being cleaned, but in fact
less p e r m e a b l e z o n e s located nearby may still have high c o n c e n t r a t i o n s of contaminants
that have yet to be removed.
8.2.8

Summary of Monitoring Well Design

Figure 8.8 illustrates details of the final design of a water table observation well and a
p i e z o m e t e r illustrating all the design e l e m e n t s discussed in this section.

Water table
observation well

Vented cap

F I G U R E 8.8

Construction details of a water table observation well and piezometer. Source: Wisconsin

Department of Natural Resources.

352

G r o u n d - W a t e r and

8.3

Soil

Monitoring

353

Installation of Monitoring Wells


8.3.1

Decontamination Procedures

B e c a u s e the p u r p o s e of drilling a monitoring well is to collect a sample of water and


analyze it for very small c o n c e n t r a t i o n s of c h e m i c a l s , it is highly desirable not to introd u c e any c h e m i c a l s into the aquifer as a part of the well-drilling and installation p r o c e
dure. T h e p r o c e s s of cleaning the e q u i p m e n t and supplies that will be used is called
decontamination.
W h e n materials are manufactured, they may b e c o m e c o a t e d with s u b s t a n c e s such
as grease and oil. T h e r e f o r e , unless the manufacturer specifically guarantees that the
article has b e e n d e c o n t a m i n a t e d and has shipped it in a well sealed wrapper, it should
be decontaminated. E q u i p m e n t that has b e e n used at a c o n t a m i n a t e d site should be
a s s u m e d to be contaminated and should be d e c o n t a m i n a t e d b e f o r e it is used at a n o t h e r
site. Even at the s a m e site, if a drill rig or a bailer is used at different wells, d e c o n t a m i n a t i o n
is required to prevent c r o s s contamination ( c o n t a m i n a t i o n from o n e area being introd u c e d into a clean a r e a ) that could o c c u r .
T h e r e is wide variability in required and r e c o m m e n d e d decontamination p r o c e
dures b e t w e e n the USEPA and the various states ( M i c k h a m , Bellandi, and Tifft 1 9 8 9 ) .
T h e hydrogeologist must consult with the appropriate regulator)' authority to d e t e r m i n e
if a specific d e c o n t a m i n a t i o n p r o c e d u r e is required. In the a b s e n c e of a specific require
ment, the following g e n e r i c p r o c e d u r e should adequately clean e q u i p m e n t and supplies.
In s o m e c a s e s not all the steps are required. F o r cleaning large e q u i p m e n t such as a
drilling rig, a specific area must be set aside and a d e c o n t a m i n a t i o n pad must be
c o n s t r u c t e d to capture all the fluids used in the p r o c e s s . If the rig is contaminated, wash
water from it may also b e c o m e contaminated. Be careful that any solvents used aren't
accidentali)' released to the e m i r o n m e n t . Small tools such as bailers and S h e l b y t u b e s
c a n be cleaned in b u c k e t s set on a polyethylene sheet. Sampling p u m p s can be cleaned
by running various wash solutions through them, as well as washing the exterior.
T h e following steps are used to clean drilling and soil sampling e q u i p m e n t including drill rig, augers, drill rod, tools, sampling tubes, etc.:
1. U s e a wire brush or similar e q u i p m e n t to r e m o v e all dried sediment and thick
accumulations o f grease.
2. Wash the e q u i p m e n t with a soft brush and water with p h o s p h a t e free detergent.
3. In e x t r e m e c o n d i t i o n s organic residues can be removed by washing the equipment
with an organic solvent such as m e t h a n o l or propanol. Don't use solvents such as
t r i c h l o r o e t h e n e that might be e x p e c t e d to be found at a hazardous waste site.
4. Clean and rinse the e q u i p m e n t with p o t a b l e water.
5. Rinse the e q u i p m e n t with deionized water.
Steam cleaning with a pressure sprayer can be used in step 4 for e q u i p m e n t that
can withstand the heat and force of the spray.
W h e n e q u i p m e n t has b e e n decontaminated, it should not be placed on the ground.
It can be wrapped in clean p a p e r or aluminum foil or set on polyethylene sheets.
Sampling e q u i p m e n t should also be d e c o n t a m i n a t e d b e t w e e n uses. If the e q u i p
m e n t has not c o m e into c o n t a c t with n o n a q u e o u s p h a s e liquids, rinsing with pota-

354

Chapter Eight
b l e water and washing thoroughly with phosphate free detergent, including scrubbing
the inside of tubes with a bottle brush, followed by a potable-water wash and then a
deionized water rinse, should suffice. If the equipment has c o m e into contact with
n o n a q u e o u s phase organic liquids, then an initial solvent wash may be necessary.
T h e c o s t of decontamination of sampling equipment and the uncertainty introduced
in solvent washing h a s led m a m hydn igei >l< igists to specify i lei In ated sampling equip
ment in each well. Disposable bailers are also available that are less expensive than the
cost of labor involved in cleaning reusable bailers.
8.3.2

Methods of Drilling

T h e r e are a n u m b e r of m e t h o d s of drilling that are appropriate for installation of


monitoring wells. W h e n working in shallow unconsolidated formations, hollow-stem
augers are c o m m o n l y used. If a well is to be drilled d e e p e r than about 100 ft or into
b e d r o c k , a rotary drilling m e t h o d may be appropriate. Cable-tool drilling is an excellent
way of installing monitoring wells in both unconsolidated and consolidated formation,
but it is slow and may be expensive. Hollow-stem augers with a bit that contains carbide
teeth can also be used in weak, indurated rock.
Hollow-Stem

Augers

Hackett ( 1 9 8 7 , 1 9 8 8 ) presents an authoritative discussion of all aspects of drilling with


hollow stem augers. A hollow-stem auger l o o k s a little like a large, untapered screw
(Figure 8 . 9 ) . T h e auger flights are constructed around a hollow pipe A drilling rig rotates
the augers and a bit on the e n d of the auger l o o s e n s the sediment, which is then brought
up to the surface by the rotating auger flights. T h e cuttings accumulate at the surface
and must be shoveled away from the augers. Figure 8.10 pictures an auger drilling rig
T h e auger is advanced into the ground as it is rotated. A plug on the e n d of a rod
inserted through the hollow stem may be screwed into the bit to seal the e n d of the
opening and prevent sediment from c o m i n g up inside the hollow stem. Alternatively, a
nonretrievable plug can be placed in the end of the bit. This plug is k n o c k e d out of the
end of the augers when the final depth is reached and it is no longer needed. However,
knock-out plugs preclude the collection of soil and water samples during drilling.
O n e advantage of drilling with hollow-stem augers is that drilling fluids and mud
are normally not required. Circulation of drilling fluids has the potential to spread
contaminants throughout the b o r e h o l e . Drilling mud is a viscous liquid needed in mudrotary drilling that can line the b o r e h o l e and partially seal it. However, when drilling in
formations with cohesive layers, the auger bit may smear clay from the cohesive layers
so that it m i x e s with sand and gravel layers at the perimeter of the b o r e h o l e . Auger
drilling typically can advance about 50 to 1 0 0 ft per day if samples are being taken.
Augers usually c o m e in lengths of 5 ft. O n e flight is advanced into the ground and
then the drill stem is d i s c o n n e c t e d and another flight is attached to the augers in the
ground. At this time samples of the formation ahead of the auger bit may be taken. T h e
plug on the e n d of the bit must be removed before sampling can occur. T h e maximum
depth at which hollow-stem auger drilling can normally be used is 1 5 0 ft (Hackett 1 9 8 7 ) ;
as a practical matter, it rarely e x c e e d s 100 ft.

G r o u n d - W a t e r and

FIGURE 8.9

Soil

Monitoring

355

Hollow-stem auger drill rod and bit. Source: M. L. Scalf et al. Manual of Ground Wafer

Sampling Procedures, 1 9 8 1 . National Water W e l l Association. Used with permission.

S o m e loosely consolidated sands, called heaving sands, c a n e n t e r into the b o r e h o l e


w h e n the plug is being removed. This p r o b l e m can be avoided by keeping the hollow
stem tilled with p o t a b l e water when the plug is removed. T h e p o t a b l e water should not
contain any contaminants that might be found in the g r o u n d water. If no formation
sampling is planned, then a k n o c k out plug c a n be used to prevent the introduction of
heaving sands into the hollow stem during drilling. T h e knock-out plug can be removed
after the final depth is reached by pushing down with the well casing if it is stainless
steel; otherwise it will be necessaiy t< > use a n id prii >r u i the time that the well is installed.

356

Chapter Eight

FIGURE 8.10

Hollow-stem auger drilling rig. Photo credit: C. W . Fetter.

Hollow stem auger drilling can also be used to sample water quality at various
depths during drilling. At a selected horizon the plug at the e n d of the hollow stem is
removed. A well point on the end of a rod is lowered to the bottom of the augers and
then is driven ahead of the bit by hammering or hydraulic pressure. T h e well point is
developed by pumping until clear water is obtained. A sample of ground water at that
depth is then obtained. It is best if p o t a b l e water is not used in the b o r e h o l e during
this procedure, b e c a u s e it could interfere with the ground water quality. If there is heaving
sand inside the augers, the well point can be driven through it.
Keeley and B o a t e n g (1987) suggested a modification o f the hollow-stem auger
drilling technique, in which a temporary casing that is larger in diameter than the auger

G r o u n d - W a t e r and

Soil

Monitoring

357

bit is employed. T h e auger is advanced several feet and then the temporary casing is
driven to the depth of the auger bit by repeatedly dropping a heavy weight on the t o p
of the casing. T h e advantage of this modification is that it prevents mixing of soil horizons
as the augers rotate.
Mud-Rotary

Drilling

Mud-rotary drilling can be used in b o t h unconsolidated and consolidated formations. It


is fairly rapid, up to 100 ft/da)-, and can be used to depths far in e x c e s s of any that might
be required for g r o u n d w a t e r - c o n t a m i n a t i o n studies. A heavy drilling fluid, m a d e by
mixing various additives to water, is circulated in the b o r e h o l e by pumping it down the
inside of hollow drill rods. T h e m u d rises b a c k to the surface in the annular s p a c e
b e t w e e n the b o r e h o l e wall and the drill pipe. T h e rising mud carries with it the drill
cuttings, which settle out in a mud tank at the surface. Figure 8 . 1 1 s h o w s the circulation
pattern for mud-rotary drilling.
O n e advantage of mud rotary drilling is that the b o r e h o l e will remain o p e n after
the string of drill pipe and the bit are removed. T h i s m e a n s that a c o m p l e t e suite of
geophysical logs can be run on the hole, which is kept o p e n by the weight of the mud
inside it. However, the fluid in die drilling mud can penetrate the native formations and
alter the ground-water geochemistry. T h e coating of drilling mud on the b o r e h o l e walls
may be difficult to remove. This can i m p e d e the hydraulic c o n n e c t i o n b e t w e e n the well
and the formation. B e n t o n i t e - b a s e d drilling muds may remove metals from the ground
water and affect the chemistry. B e c a u s e of the potential p r o b l e m s with drilling mud,
rotary drilling may not be as suitable for ground-water contamination studies as hollowstem augers. However, under many hydrogeologic conditions it is the only drilling m e t h o d
that is practical. G o o d well development to remove the residual drilling mud in the
screen z o n e is very important.
Air-Rotary

Drilling

If the monitoring well is to be installed in b e d r o c k , then air rotary drilling may be


considered. First a surface casing n e e d s to be installed through any unconsolidated
material. Typically this is d o n e by using mud-rotary drilling. T h e surface casing is large
enough that the air rotary bit can fit inside of it.
T h e fluid used in air rotary is c o m p r e s s e d air, which is blown down the inside of
the drill pipe. T h e air then blows the cuttings b a c k up the annular space, w h e r e they
accumulate around the b o r e h o l e . W h e n the water table is e n c o u n t e r e d , the air may b l o w
ground water out of the b o r e h o l e as well. If this o c c u r s , it is p o s s i b l e to determine w h e n
the water table is e n c o u n t e r e d and the relative yield of the well. However, the air may
also force the water back into the formation. Air-rotary drilling using a down-hole
percussion bit can drill up to 60 ft/hr. Samples are c o l l e c t e d as chips, which are brought
to the surface with the return flow of air and water.
Air-rotary drilling is fast and can go to depths of a thousand feet of m o r e . B e c a u s e
air is the drilling fluid, contamination p r o b l e m s are minimized. However, the drillers may
want to add a foaming agent to the air as it g o e s down the hole. T h e foam helps to
float the chips to the surface, but it c o n s i s t s of organic chemicals such as isopropyl
alcohol, ethyl alcohol, and alcohol ether sulfate, and its use should be avoided. T h e air

G r o u n d - W a t e r and

Soil

Monitoring

359

c o m p r e s s o r s have air filters, which need to be in g o o d working condition; otherwise,


lubricating oil and o t h e r contaminants may be introduced into the b o r e h o l e with the
c o m p r e s s e d air. Percussion h a m m e r s used for air-rotary drilling may also n e e d lubricants.
Air-rotary drilling may introduce volatile organic c o m p o u n d s into the a t m o s p h e r e
as well as blowing contaminated dust out of the b o r e h o l e .
Reverse-Rotary

Drilling

In reverse-rotary drilling the circulating drilling fluid drains down the annular s p a c e and
then is pulled up the c e n t e r of the drill stem by a suction p u m p located on the drill rig.
B e c a u s e the drilling fluid rises with a much greater velocity in reverse-rotary than in mudrotary drilling, a m u c h less viscous drilling fluid is used. In many c a s e s clear water m i x e d
with the drill cuttings is all that is necessary. T h i s gives the reverse-rotary m e t h o d an
advantage over mud-rotary drilling, since it is m u c h easier to develop the well b e c a u s e
there is no mud wall on the b o r e h o l e to b r e a k down. However, reverse-rotary drilling
is m o r e e x p e n s i v e than the mud-rotary m e t h o d , and the minimum b o r e h o l e diameter
is 12 in.
Cable-Tool

Drilling

Cable-tool drilling is o n e of the oldest drilling m e t h o d s and has b e e n used widely for
the installation of water wells. Although the drilling e q u i p m e n t is less e x p e n s i v e than
for s o m e o t h e r m e t h o d s , the drilling is slow and overall c o s t s may be e x p e n s i v e due to
high labor costs.
In cable-tool drilling a heavy bit is located at the e n d of a tool string hanging from
a cable. T h e drill rig repeatedly lifts and drops the hammer, which b r e a k s up consolidated
r o c k or l o o s e n s unconsolidated sediment. A steel casing is driven into the formation
behind the bit. W h e n the b o t t o m of the casing fills with b r o k e n r o c k and sediment, the
tool string and bit are removed, and a bottom-loading bailer is used to remove the
accumulated cuttings. B e l o w the water table, the g r o u n d water and cuttings m a k e a
slurry. Above the water table, water must be added to m a k e a slurry so that the bailer
c a n be used. Drive casing is n e e d e d only until b e d r o c k is reached. In m o s t b e d r o c k
formations the h o l e will stay o p e n without drive casing. Figure 8.12 s h o w s the tools
used for cable-tool drilling.
Advantages of cable-tool drilling include the fact that no drilling fluids are used
and that nothing is circulated through the well. B o t h factors serve to limit contamination
p r o b l e m s . It is easy to collect representative samples of the formation during bailing of
the casing. Well points can be driven ahead of the casing in unconsolidated formations
for the collection of water quality samples. Cable-tool drilling can be used to depths in
e x c e s s of 1 0 0 0 ft.
8.3.3

Drilling

in

Contaminated Soil

When drilling at a contaminated site, the cuttings that are brought to the surface may
be contaminated. Drilling personnel should wear appropriate protective clothing and,
if necessary, use breathing apparatus. A large, heavy sheet of plastic should be placed
in the work area, and the drill bit should be advanced through a h o l e in the c e n t e r of
the plastic sheet.

360

Chapter Eight

Flapper valve

Drilling hit
FIGURE 8.12

Bailer

Tools used for cable-tool drilling.

T h e cuttings augered to the surface can be collected from the plastic sheet and
put into containers for p r o p e r disposal.

8.4

Sample Collection
Samples of unconsolidated materials and rock are n e e d e d to delineate the geology- of a
site. T h e y are c o l l e c t e d by drilling b o r i n g s . Borings may be made just for sample
collection, or they may be made as a part of the p r o c e s s of installing a monitoring well.
Borings may be made by any of the m e t h o d s of drilling discussed in Section 8.3.
During the drilling process, earth materials are brought to the surface. During the
augering p r o c e s s , soil and sediment ride up the augers; in the mud rotary process, earth
materials c o m e up m i x e d with the mud; in air-rotary drilling there is a slurry of rock
and water brought to the surface by air pressure; and cuttings are brought to the surface
with a bailer in cable-tool drilling. In all cases the samples are disturbed, s o m e (such
as the samples on the augers) m o r e than others. In addition, it may be difficult to tell
the e x a c t depth represented by the sample. Fine layering of sedimentary materials cannot
be distinguished in such samples. An imprecise model of the geology can be constructed
from these samples, but for m o r e details, undisturbed samples should be collected.

G r o u n d - W a t e r and

Soil

Monitoring

361

A c o r e s a m p l e is c o l l e c t e d in a special sampling t u b e that is driven into unconsolidated formations or by the drilling of a r o c k - c o r e s a m p l e with a special r o c k - c o r e
bit studded with industrial diamonds.
T h e r e are two main types of sampling t u b e s for unconsolidated samples. B o t h
types of t u b e s can be used with hollow-stem augering and mud-rotary drilling if the drill
rod and bit have a provision for a sample t u b e to be e x t e n d e d through them into the
formation ahead of the bit. Figure 8.13 s h o w s the s e q u e n c e for the e x t e n s i o n of a sampler
through the e n d of the bit of a hollow-stem auger. T h e sampler is driven or pushed
into undisturbed formation ahead of the bit. In cable-tool drilling the drill bit is removed
from the b o r e h o l e and the sampler is lowered on a rod or c a b l e .
A S h e l b y t u b e is a thin-walled t u b e that can be s c r e w e d to the e n d of a rod,
lowered to die b o t t o m of the drilled hole, and p r e s s e d into c o h e s i v e sediments by using
hydraulic pressure reacting against the weight of the drill rig. T h e s e samples are said to
be undisturbed, although they are in fact minimally disturbed. T h e precise m e t h o d of
collecting a S h e l b y t u b e sample is d e s c r i b e d in m e t h o d ASTM D 1 5 8 7 (American Society
for Testing and Materials 1 9 8 3 ) . T h e sample can be e x t r u d e d from the Shelby t u b e in
the lab and trimmed into a p e r m e a m e t e r for a permeability test. Details of the microstratigraphy c a n be e x a m i n e d as well. S h e l b y t u b e sampling d o e s not work with nonc o h e s i v e sediments.
A s p l i t - s p o o n s a m p l e r can b e used for the collection o f samples o f b o t h cohesive
and n o n c o h e s i v e sediments. T h e split-spoon sampler c o n s i s t s of a split t u b e with thicker
walls than a Shelby tube. T h e two halves are placed together and j o i n e d by screwing a
circular drive s h o e on the b o t t o m and a head assembly on the top. T h e a s s e m b l e d splits p o o n sampler is s c r e w e d to a rod and lowered to the b o t t o m of the drill hole. A pipelike weight of 1 4 0 lb is placed on the top of the rod. T h e weight is repeatedly raised
and d r o p p e d a distance of 30 in. in o r d e r to drive the split s p o o n sampler into the
formation. T h e n u m b e r of blows necessary to drive the sampler every 6 in. is r e c o r d e d
as the sampler is driven 18 in. into the formation. T h e m o r e d e n s e the formation, the
greater the n u m b e r of b l o w s n e e d e d to drive it 6 in. T h e p r o c e s s is d e s c r i b e d by ASTM
1 5 8 6 (American Society for Testing and Materials 1 9 8 4 ) .
After the split-spoon sampler is driven 18 in., or after refusal, it is brought to the
surface and o p e n e d . Frequently, less than 18 in. of sediment have b e e n collected. Fine,
n o n c o h e s i v e sediment such as sand may fall out of the sampler as it is b e i n g retrieved.
Sediment greater in diameter than one-third of the diameter of the sampler may not
enter it at all. A p e b b l e may lodge in the barrel and not allow any o t h e r sediment to
enter. Sediments may c o m p a c t in the sampler so that a full 18-in. sample may actually
o c c u p y less than 18 in. of the c o r e barrel. T h e hydrogeologist examining the split-spoon
sample must use his or her judgment in making a log b a s e d on the split-spoon samples.
Figure 8.14 is a p h o t o g r a p h of a split s p o o n sample.
A standard for c o r e samples is to collect o n e 18-in. sample every 5 ft. T h i s frequency
of sampling is suitable for relatively h o m o g e n e o u s formations. However, if the microstratigraphy of the formations is i m p o r t a n t f o r e x a m p l e , if there are p e r m e a b l e sand
s e a m s in a clay f o r m a t i o n t h e n c o n t i n u o u s - c o r e samples should be collected. Continuous c o r e s are m a d e by advancing the drill bit to the full depth that was sampled by a
split s p o o n and then immediately taking a n o t h e r c o r e s a m p l e of the fresh formation

362

Chapter Eight

Auger - A ^ J
column V > v

Auger
colum]

(} O.?

oP.'0:-?':'^,.

Center rod

O
0.cS,C;.oc^.^-.C*.
3'([ Sampling

>i<4>

(/^Ho-C'tj
C Open axis
of auger
column with
C pilot
C 0 assembly
C and center
i' rod

o.

Pilot ^ ;
assembly ^
3

c 2 f ;

: 0 -':0 > *

.>. Split barrel-j


o or Ihin5
C walled tube ^ .
sampler
5

r\o

mM
r o o ^ o O

cJ&rvp

c^'.c
:0

Qf.? c^.^o Qr^


j6

QCTTO

O':o V

F I G U R E 8.1 3

Sequential steps for the collection of a core sample through a hollow-stem auger. Source:

Glen Hackett, Ground Water Monitoring Review 7, no. 4 (1987): 51 - 6 2 . Used with permission. Copyright
1 9 8 7 Water W e l l Journal Publishing Co.

G r o u n d - W a t e r and S o i l M o n i t o r i n g

F I G U R E 8.14

363

Hydrogeologist describing a split-spoon sample. Photo credit: C. W. Fetter.

ahead of the previous c o r e . C o n t i n u o u s - c o r e samples can also be collected in cohesive


soils by using a special c o r e barrel that collects a 5-ft-long sample inside a hollow-stem
auger as the auger is being advanced. If a s e q u e n c e consisting of a 3-in split s p o o n
followed by a 2.5-in split s p o o n followed by a 2 in split s p o o n is used, 4.5 to 5 ft of
continuous sample c o r e can be collected b e f o r e the augers n e e d to be advanced.
In consolidated formation a r o c k - c o r e sample is c o l l e c t e d by use of a c o r e barrel
with a diamond studded bit. T h e rotating bit grinds up rock in an annular pattern, leaving
an undisturbed c e n t e r of rock that enters the c o r e barrel. T h e r e is a c o r e lifter just
behind the bit to k e e p the c o r e from falling out of the c o r e barrel w h e n the drill rods
and bit are retrieved from the b o r e h o l e .

364

8.5

Chapter Eight

Installation of Monitoring Wells


Following the collection of samples during the installation of borings, a monitoring well
can be installed in the b o r e h o l e . B o r e h o l e s drilled with mud should stay o p e n with the
drill rod removed. Hollow-stem augers are generally left in the ground and the well is
installed through them, as is drive casing in cable-tool drilling. T h e r e must be a sufficient

Threaded, flushjoint casing ^


and intake

~mSM
Inside diameter of
hollow-stem auger

.
~
CO
Maximum
working space

-- e
Hollow-siem
auger
Well casing
inserted through
hollow-stem auger

Auger column

few

earn

Plan view

;3f.G.

Inatto
o ti:Q:C
0:oS-<j

' o "C/e &

X-viY

S E

. .Oes o ' , Q . Q a -

go

OTO-

Cross-sectional view
F I G U R E 8.15

Cosing offset inside hollow-stem auger to give greatest working opening. Source: Glen

Hackett, Ground Water Monitoring Review 8, no. 1


1988 Water W e l l Journal Publishing Co.

(1 9 8 8 ) : 6 0 - 6 8 . Used with permission. Copyright

G r o u n d - W a t e r and S o i l M o n i t o r i n g

365

working o p e n i n g inside the casing or augers. F o r a 2-in. nominal monitoring well, this
m e a n s a minimum 4^ in. o p e n i n g is needed. T h e casing can be offset within the auger
to give the largest working o p e n i n g (Figure 8 . 1 5 ) . With a 4^-in. inside diameter auger
and a 2-in. nominal casing, this creates a 1.875-in. working opening.
T h e first step in the installation of a monitoring well is to s c r e w the well s c r e e n
to the casing and then lower the assembly through the inside of the augers or temporary
casing. Prior to installation the casing and s c r e e n should be thoroughly decontaminated.
T h e casing and s c r e e n may be wrapped in white b u t c h e r paper after it has b e e n de
contaminated and then kept w r a p p e d until just b e f o r e it is lowered into the augers.
Figure 8 . 1 6 s h o w s a 20-ft casing and s c r e e n being lowered by hand. Longer casings n e e d
to be lowered on a c a b l e using the drilling rig.

FIGURE 8.16

Lowering well screen and casing Into hollow-stem augers. Note the white wrapping paper

around the decontaminated casing. Photo credit: C. W. Fetter.

366

Chapter Eight

O n c e the casing ami

si r e c t i

have been

It

iwcrcil into the well, the filter pack material

needs to be placed. T h e volume of filter-pack material necessary to fill the annular space
between the si recti and easing and the b o r e h o l e wall from the bottom til the b o r e h o l e
to a point 2 ft a b o v e the top of the screen should be c o m p u t e d . At least this much
material must be on hand b e f o r e starting the filter pack installation. A weighted measuring
tape is lowered into the working o p e n i n g between the casing and the hollow stem auger
and the total depth of the b o r e h o l e is measured and recorded.
If the formation is cohesive and can stand o p e n for a short while, the augers are
withdrawn 1 or 2 ft from the bottom. Filter pack material is then poured into the working
opening, and the annular s p a c e is filled to the levels of the auger bit. Care should be
taken that the filter pack sand doesn't fill the s p a c e between the casing and the augers,
b e c a u s e it can lock the casing and hollow-stem augers together. T h e weighted tape is
used to determine the position of the top of the filter pack. T h e augers are then withdrawn
another 1 to 2 ft, and the p r o c e s s is repeated until the entire filter pack is placed. Figure
B . P illustrates what is known as the free-fall m e t h o d of filter pack emplacement.
W h e n filter pack material d r o p s through a water column, it may separate according
to size. It may also bridge the s p a c e between the casing and the auger and create a void
below. T o avoid these p r o b l e m s , a t r e m m i e p i p e should b e used wherever possible
(Figure 8 . 1 8 ) . A tremmie pipe is a pipe that e x t e n d s from the surface and through which
the filter-pack sand may be poured. After the augers are withdrawn a few feet, the annular
s p a c e is filled with sand being poured down the tremmie pipe. T h e tremmie pipe is
raised as the level of sand rises. T h e tremmie pipe can be used to tamp down the sand,
and the weighted tape is used to measure the position of the top of the filter pack.
If the formation is n o n c o h e s i v e , it will collapse as the augers are withdrawn. Under
these conditions, the withdrawal of the augers and the addition of the filter-pack material
must o c c u r simultaneously (Hackett 1 9 8 8 ) . A c a b l e must be attached to the top of the
auger string so that the working o p e n i n g is a c c e s s i b l e at all times. T h e hollow stem of
the augers is filled with clean water, and a positive hydraulic head is thus maintained
throughout the operation. T h e augers are very slowly withdrawn, and at the same time
filter pack material is added so that the top of the filter pack is within an inch or so of
the b o t t o m of the augers. This requires precise coordination of the rate of addition of
filter pack sand and the rate of withdrawal of the augers.
T h e final depth of the top of the filter pack is confirmed by measurement with
the weighted tape. It should be 2 ft a b o v e the top of the screen. T h e augers are then
withdrawn a n o t h e r 2 ft and 2 ft of pure bentonite clay are placed by free fall through
the working space. If the area is b e l o w the water table, bentonite pellets are used; if the
area is a b o v e the water table, granular bentonite is used. Pellets should be dropped a
few at a time so that they aren't caught in the working s p a c e as they start to swell by
hydration. T h e weighted tape is used to confirm the final thickness of the bentonite layer
after enough time has elapsed to allow the bentonite to hydrate.
Placement of die annular seal should take place by use of a tremmie pipe. T h e
tremmie pipe should be lowered to the t o p of the bentonite seal. T h e augers are
withdrawn 2 ft or so .mil the annular space is filled from the b o t t o m with grout, which
is either p u m p e d down the tremmie pipe or is fed by gravity (Figure 8 . 1 9 ) . T h e weighted
tape is used to confirm the position of the top of the grout. T h e augers are then repeatedly

\\ eighied
measuring

s f ^ C Well

cas

Hollow
Holl. -stem

HX.

auger

Pinn view

Weighted
measuring tape

Auger column
retracted
1 to 2 ft
from borehole

measuring tape

U eighted
measuring tape

Cj

:'ps,-<ok)

O'B.ns ' % y
. M ' M .
pc>S?q:o

.-an. Q
pkJ'Spq'.O

SgtSSP.

:*?e>VQ
&%.'.q':

P.CJ'<?
."rS'O,

in-^Cf
W'S '
:

{ f e s a
V.-0.'(Q:O

.'O

tes

Qg)3'
pi(>3?o'.0

\-.:v.C .<
:/

-sis*

:-.o..'o .?0
;

Q..a-

b>&*

\(7..^0.

.r-t C\C-A
ti'PO'U

S .
'fc'PT'M'
ti-.RY.O-Y

'OVO.'/}':;. ..

a'0<?'S<JKi.

G'pS-Qirj

i Q

amasti.

:-p,'W-?'

feg?

fc>>.'

M'0:P:.';>.'.3:

JK3

c i

Cross-sectional view
Placement of weighted
measuring tape

FIGURE 8.17

-:P.te>

;-:--.r\'.o-(>

,-5:\>s;.c

y.c-F-':-'.''/-f>is>

6 a > : 0

u*-* Filter pack

^ / r C

0^

Auger column retracted

Free-fall method of filter-pack emplacement with a hollow-stem auger. The method also

works with drive casing. Source: Glen Hackett, Ground Water Monitoring Review 8, no. 1 {1 9 8 8 ) : 60
6 8 . Used with permission. Copyright 1 9 8 8 Water W e l l Journal Publishing Co.

W O P : , a ON
>o- i . : P o v
Cross-sectional view
Filter pack free-falls through
working space between casing
and auger

368

Chapter Eight

Well C.IMT1L!

i l l k - l ["M.. ,
material poured
down trmie
Weightedmeasuring tape
Weighted
measuring lape

Auger column
re traded
1 (o 2 ft
from borehole

Trmie pipe
slowly raised i
filter pack is
poured

>(>?o.O

WO?-

l'pV.Si&P
Vio "

.-o-'O.-'o-

T.>

'

few

a:r:0
t>i(/-vS?o:rj

fe
A

'o-.-o/n'.-.--;

... 'Yc-.o:
>- r>S?o:o

s%--:.Q-:

:-p-0 .M

o OY).>-

0:p&<bK.
:

:-P..q P

F i l l e r pack
!

Oj^.iiS)

j;os..f :b: -\)0

Cross-sectional view
Weighted measuring tape and
iremic pipe in retracted
auger column
FIGURE 8.18

F i l l e r pack poured through bottomdischarge tremie pipe

Use of a tremmie pipe for emplacement of filter-pack material. Source: Glen Hackett,

Ground Wafer Monitoring Review 8, no. 1 (1988|: 6 0 - 6 8 . Used with permission. Copyright 1988 Water
W e l l Journal Publishing Co.

G r o u n d - W a t e r and

Soil

369

Monitoring

Auger

Tremmie pipe'

Casing

;C7.- <=r-:
0

- Bentonite
pellets
^r.
0

-..-o^r

- Filter pack

tit?
c>-S

rete

<oo..%-

. . o

o -OC?
-o^
pp.?e?Ci '

'

'
.

Co?

.LJ

FIGURE 8.19

Use of a tremmie pipe for emplacement of grout above the bentonite pellets.

370

Chapter Eight

withdrawn and grout is e m p l a c e d until it is brought c l o s e to the surface. T h e tremmie


pipe can be left at the b o t t o m until the grout is brought to the surface, or it can be
raised as the augers are withdrawn. F o r slumping sediments, the grout must be emplaced
simultaneously and at die s a m e rate as the augers are withdrawn.
T h e final step is the installation of a locking protective cap. Bentonite-cement can
be brought all the way to the surface. If b e n t o n i t e grout or bentonite sand grout is used,
the final few feet filling the annular s p a c e must be neat c e m e n t or bentonite c e m e n t
grout. T h e locking protective casing can be pushed into the c e m e n t grout when it is
still soft. If the grout setdes overnight, its level should be brought to the surface with
additional material. A stronger surface seal can be obtained if the top 2 ft of the annular
s p a c e are filled with c o n c r e t e , as o p p o s e d to bentonite-cement grout.

8.6

Monitoring Well Development


O n c e a monitoring well is constructed, it is necessary that it undergo development This
is a p r o c e s s of removing fine sand, silt, and clay from the aquifer around the well screen.
If drilling mud was used, then vigorous development may be n e e d e d to break down the
mud pack on the b o r e h o l e wall. Development is n e e d e d to create a well that ideally will
not pump silt and clay when it is sampled. It also may create a z o n e around the well
s c r e e n that is m o r e p e r m e a b l e than the native soil and that stabilizes the native soil so
that the fine sediments do not enter the filter pack.
Aller et al. ( 1 9 8 9 ) made the following observations about monitoring well development:
1. Using c o m p r e s s e d air for well development may alter native water chemistry, crack
the casing or b l o w the b o t t o m cap off the screen.
2. Adding water to the well for flushing the well or surging can alter the ground-water
chemistry, at a minimum by dilution.
3. Breaking down a mud wall left in the b o r e h o l e from mud-rotary drilling is very difficult
4. Developing a well when the screen is in a clean, h o m o g e n e o u s , high permeability
aquifer is relatively easy.
5. Developing a well when the s c r e e n is in a fine-grained, stratified, low-permeability
formation is difficult.
6. Developing a large diameter well is easier than developing a small-diameter well.
7. Shallow monitoring wells are easier to develop than d e e p e r monitoring wells.
8. Monitoring wells that can be bailed dry tend to be turbid b e c a u s e of the steep hydraulic
gradients that are developed.
9. In the final analysis, many monitoring wells c a n n o t be developed to the point where
a nonturbid, ground-water sample can be collected. T h i s is especially true if the
formation d o e s not yield very much water, so that extensive development is not
possible.
If the b o r e h o l e is drilled into a stable, consolidated formation, especially if mud
is used during the drilling, it may be advantageous to flush the b o r e h o l e with potable,
fresh water to wash out as much of the mud as possible prior to installation of the well

G r o u n d - W a t e r and S o i l M o n i t o r i n g

371

and filter pack. T h i s will greatly cut the time n e e d e d for well development. In s o m e c a s e s
it will not be permissible to add water to the b o r e h o l e , since this might alter the g r o u n d
water chemistry.
T h e r e are three p r o c e d u r e s used for monitoring well development: bailing, surgeb l o c k surging, and p u m p i n g / o v e r p u m p i n g / b a c k w a s h i n g . T h e s e may b e used a l o n e o r
in c o m b i n a t i o n .
A b a i l e r for a monitoring well is a section of pipe that is o p e n on the top e n d
and has a foot valve on the b o t t o m end. It is attached to a line so that it can be lowered
into the well. Water fills it from the b o t t o m ; then w h e n it is raised by the line, the foot
valve c l o s e s and the water inside is trapped. Figure 8.20 is a diagram of a bailer.
W h e n developing a well by bailing, the bailer should be allowed to free-fall to the
water surface. W h e n it strikes the water, a pressure wave results, which p u s h e s water
from the well s c r e e n out into the formation. After the bailer is filled with water, it is
withdrawn and water from the formation e n t e r s the screen. T h i s b a c k and forth m o t i o n
of water through the filter pack l o o s e n s fine sediment so that it can be drawn into the
well and removed by the bailer. T h e bailer should be allowed to sink to the s c r e e n area
so that the water that fills it contains the fine sediment that is brought into the well from
the filter pack. T h e bailer can also be raised and lowered w h e n it is s u b m e r g e d to force
water to move b a c k and forth through the s c r e e n area. Bailing can take s o m e hours to
develop a monitoring well effectively; this can translate into significant labor cost. Bailing
for development can be undertaken by hand or by using a c a b l e attached to a powero p e r a t e d drum on a drill rig or truck. S o m e well-development outfits have an arm that
can go up and down like a walking b e a m to create a surging action. Care should be
taken so that die surging action is not vigorous e n o u g h to collapse the well screen.
A s u r g e b l o c k is a device that fits inside the well with a flexible gasket that is
c l o s e in size to the inside diameter of the well. Figure 8.21 s h o w s the design of a surge
b l o c k for small-diameter monitoring wells. It is attached to a r o d that is raised and
lowered with a stroke of a b o u t 3 ft. Most of the water is moved up or down by the
action o f the surge b l o c k , although s o m e fraction o f the water bypasses the surge block.
T h e surging is initiated with the surge b l o c k at the top of the well screen, and the b l o c k
is gradually lowered until the entire s c r e e n e d area has b e e n surged. Every so often the
surge b l o c k is removed and the well bailed to remove the sediment that has b e e n brought
into the well. If t o o m u c h sediment accumulates a b o v e the surge b l o c k , it can bind
b e t w e e n the surge b l o c k and the casing wall and l o c k up the surge b l o c k . To avoid this,
the surge b l o c k must be removed and the well bailed frequently. Surging b e g i n s with a
gende action; as d e v e l o p m e n t p r o g r e s s e s it b e c o m e s m o r e vigorous, with a m o r e rapid
plunging action. Again, t o o vigorous a surging action might c a u s e the s c r e e n to collapse.
Pumping the well can aid in development. A n u m b e r of different types of p u m p s
are suitable. However, s o m e p u m p s might be damaged by the sediment that must
necessarily be removed during development. T h e ideal p u m p for development is c a p a b l e
of a wide range of flow velocities and d o e s n ' t have a valve that prevents backflow. T h e
pump intake should be in the s c r e e n z o n e so that it will immediately pick up sediment
that is brought into the well. T h e p u m p is started at a l o w velocity and is shut off every
so often. W h e n it is shut down, the water in the p u m p c o l u m n b e t w e e n the water surface
and the p u m p will flow b a c k into the well and out into the filter pack. W h e n the p u m p
is started again, this water will be drawn into the well and will l o o s e n fine sediment in

372

Chapter Eight

FIGURE 8.20

Diagram of a bottom-loading bailer.

G r o u n d - W a t e r and

Soil

373

Monitoring

Polypropylene tube ( 0 . 3 7 5 - i n . O.D.)


Stainless steel cable ( 0 . 0 6 3 - i n . O.D.)
Fernile
Water ports ( 0 . 2 5 - i n . O.D.)
Stainless steel hex nut (0.63 in.)

O
o

o
o

Viton discs ( 0 . 0 5 - i n . thick, 2.1 - i n . O.D. and


0 . 6 8 - i n . I.D.)

S C H 80 P V C pipe ( 1 . 9 0 - i n . O.D.)
top Fitting

Oy

A i r vent passage
(0.375-in. O.D.)

N P T threading

Cross section
A - A'

- Stainless steel coupling


(1.325-in. O.D.)

stainless steel pipe


(1.067-in. O.D.)

S C H 80 P V C pipe ( 1 . 9 0 - i n . O.D.)
bottom fitting

Stainless steel hex nut (0.63-in.)


: A i r vent ports ( 0 . 1 2 5 - i n . O.D.)
Stainless steel tube (0.375-in. O.D.)
- Swage block

FIGURE 8.21

Design of a specialized surge block for monitoring wells. Source: Ronald Schalla and R.

W. Landick, Ground Water Monitoring Review 6, no. 2 (1 9 8 6 ) : 7 7 - 8 0 . Used with permission. Copyright
1 9 8 6 Wafer W e l l Journal Publishing Co.

t h e filter pack. With time the rate at w h i c h t h e well is p u m p e d is i n c r e a s e d so that the


w a t e r velocity through the filter p a c k into the s c r e e n is i n c r e a s e d . Eventually, t h e well
will be o v e r p u m p e d t h a t is, p u m p e d at a rate that is g r e a t e r than the flow i n t o the
well through the well s c r e e n . T h e r e will be a rapid d e c r e a s e in t h e w a t e r level in the
well during o v e r p u m p i n g , and it c a n ' t be s u s t a i n e d for very long. Eventually the w a t e r
from the well s h o u l d clear. Monitoring wells s h o u l d be periodically r e d e v e l o p e d .

374

Chapter Eight

Facility/Project Name

Date Well Installed

Section Location

Well Installed By: (Person's Name and Firm)

mm

1/4 of

1/4 of Section

d d

Well Name

Proiective pipe. top elevation

ft M S L

I.
2

W e l l casing, top elevation

. ft M S L

Land surface elevation

ft M S L

Surfacc seal, bottom

Cap and lock?


Protective cover pipe:
a

Inside diameter:

Length:

Material:

Steel
.

._ fl
d

U S C S classification of soil near screen:


GP GM
S M SC
Bedrock

GC G W S W
M L MH CL

Sieve analysis attached?

SP
CH

D Yes

Other

Additional protection?
If yes, describe:

Surface seal:

j y

Bentonite

Concrete

Other

Material between well easing and protective pipe:

Bentonite

No

Annular space seal O

Air
None

Drilling method used:

Rotary

Other

Hollow Stem Auger

Annular space seal:

Other

Drilling fluid used: Water d 0 2


Drilling Mud

Drilling additives used?

O03

0 Yes

Granular Bentonite

Lbs/gal mud weight . . Bentonile-sand slurry


Lbs/gal mud weight
% Bentonite

Bentonite slurry D
Bentonite-cement grout D

Ft" volume added for any of the above


How installed:

Tremie D

No

Tremi e pumped
Gravity D

Describe

Bentonite seal:

Source of water (attach analysis):

1/4 in.

Bentonite granules

3/8 in.

1/2 in.

Bentonite pellets
Other Q

7.

Fine sand material: Manufacturer, product name


and mesh size

&

Filter pack material: Manufacturer, product namc


and mesh size

Volume added _
H. ni. mi. seal, lop

ft M S L or

Fine sand, top

ft M S L or _ _ . . . f>

Filter pack, top

fl M S L or _ _ _ . _ f l

W e l l screen, lop

ft M S L or _ _ _ . _

Well screen, bottom

fl M S L or

Filter pack, bottom

ft M S L or

Borehole, bottom

fi M S L r _ _

Borehole, diameter

in.

O.D. well casing

in.

ft

Volume added _
9.

W e l l casing:

Flush threaded P V C schedule 40


Flush threaded P V C schedule 80

ft

ft
ft

Other
10.

Screen material:
Factory cut

Screen type:

Continuous slot
Other

...ft

Manufacturer _

Slotted length;
NoneG

Backfill material (below filter packl:

I.D. well casing

FIGURE 8.22

0..

Slot size:

II.

Form for recording information about construction details of a monitoring well. Source:

Modified from Wisconsin Department of Natural Resources.

Other

G r o u n d - W a t e r and

8.7

Soil

Monitoring

375

Record Keeping During Monitoring Well Construction


Many states have detailed record-keeping requirements for monitoring well construction.
Even w h e n detailed records aren't required by statute, they should be kept as a matter
of s o u n d professional practice. Figure 8.22 s h o w s a one-page form that can be used to
record important information on monitoring well construction. Many firms have their
o w n forms, and many states have a required form. T h e form should be filled out in the
field as the information is collected.
R e c o r d s should also be kept of well development, including the date that it o c c u r r e d ,
the m e t h o d used, the water level at the start of development, the water level at the e n d
of development, the time spent developing the well, and the volume of water removed.
T h e thickness o f sediment o n the b o t t o m o f the well can b e determined b y measuring
the depth to the b o t t o m of the well with a weighted tape. This measured distance is
subtracted from the measured length of well casing and s c r e e n that was installed. T h e
difference b e t w e e n the two d i m e n s i o n s is the thickness of sediment inside the well. T h i s
thickness should be measured and r e c o r d e d b o t h b e f o r e and after development. If
possible, sediment should be removed with a bottom-loading bailer.

8.8

Monitoring Well and Borehole Abandonment


S o m e t i m e s difficulties are e n c o u n t e r e d during the construction of a monitoring well that
prevent its c o m p l e t i o n . Monitoring wells may be installed for a specific time period, after
which they must be removed. T e s t borings may be m a d e with no intention of using the
b o r e h o l e for construction of a monitoring well. In all such cases, p r o p e r a b a n d o n m e n t
of the well should be undertaken. Many states have specific well a b a n d o n m e n t c o d e s .
In the a b s e n c e of specific requirements, monitoring wells and b o r e h o l e s should be
a b a n d o n e d in such a m a n n e r that surface water c a n n o t drain into the aquifer. Otherwise
a direct c o n n e c t i o n for contaminated water from the surface to the aquifer can result.
If a casing and s c r e e n have b e e n installed, they should be removed if possible.
This can be a c c o m p l i s h e d by pulling if the annular seal has not b e e n filled with a cementtype grout. If a plastic casing breaks while being pulled, it can be removed by drilling
it out with hollow-stem augers. Following the removal of the casing, a tremmie pipe
should be used to fill the resulting b o r e h o l e from the b o t t o m with an appropriate
material, such as neat c e m e n t or b e n t o n i t e grout. T h e grout is placed while the augers
are being pulled out of the hole. Material removed from a monitoring well may be
contaminated and should be properly d i s p o s e d of.
If the casing and s c r e e n have b e e n grouted into place, it may not be p o s s i b l e to
remove them. If this is the c a s e , the casing should be cut off b e l o w grade; then the
s c r e e n and casing must be filled from the b o t t o m using a tremmie pipe and an appropriate
material such as neat c e m e n t or b e n t o n i t e grout.
B o r e h o l e s in sediments can be filled with grout or native soils m i x e d with bentonite.
B o r e h o l e s into b e d r o c k should be grouted with a cement-type grout.

376

Chapter Eight

8.9

Multiple-level Devices for Ground-Water Monitoring


Usually monitoring wells are installed in nested configurations, with a water table monitoring well and o n e or m o r e piezometers s c r e e n e d b e l o w the water table. T h e best
situation is to install each monitoring well or piezometer in an individual borehole. Under
these circumstances it is possible to obtain an excellent seal to prevent vertical movement
of water along the casing. T h e s e individual wells can be placed within 5 ft of each other.
Under s o m e circumstances it may be possible to begin with a large-diameter
b o r e h o l e , place two or m o r e monitoring wells or piezometers in the b o r e h o l e , and
provide a seal b e t w e e n them. However, the construction of such a well nest might be
quite expensive. F o r example, it could be d o n e in an unconsolidated formation by ( 1 )
driving a 10- or 12-in.-diameter casing with a c a b l e tool rig to the depth of the bottom
piezometer, ( 2 ) installing the deepest piezometer, ( 3 ) pulling back the casing and
installing the filter pack and seal for the deepest piezometer, ( 4 ) pulling back the casing
and grouting up to the level of the next piezometer, ( 5 ) allowing the grout to harden,
( 6 ) installing the next piezometer, ( 7 ) pulling back the casing and installing the filter
pack and seal for the s e c o n d piezometer, ( 8 ) pulling back the casing and grouting up
to the level of the third piezometer, etc. Moreover, it is difficult to install a seal between
piezometers in the s a m e b o r e h o l e that can positively prevent vertical movement of ground
water along the casing. Such a design may make s e n s e when multiple piezometers are
to be put into a b e d r o c k b o r e h o l e hundreds of feet d e e p where the drilling costs are
very high. F o r shallow wells, especially in sediment, the design usually is not cost effective
It is possible to install inexpensive multilevel sampling devices in a sandy aquifer.
O n e such device is shown in Figure 8.23. It consists of a rigid PVC tube, inside of which
are multiple tubes of flexible tubing. Each tube leads from the surface and ends at a
different depth. Each has a port into which a ground-water sample can be drawn. Sampling
ports can be spaced vertically at distances as c l o s e as 1 ft or less, so that very detailed
vertical sampling can be accomplished. Water is withdrawn from the tubing by applying
a suction, so the water table must be less than 25 ft b e l o w the surface.
Multilevel samplers also can be constructed by fastening a bundle of flexible tubes,
each of a different length, to the outside of a rigid PVC pipe that acts as a spine.
T h e multilevel sampler is installed by using a hollow-stem auger. T h e device is
constructed at the surface and lowered to the desired depth through the augers. It is
not possible to develop the sampling ports, so this device can be installed only in clean,
sandy sediment. No filter pack or grout is used in this construction. T h e augers are
withdrawn and the native sand is allowed to slump around it. O n e disadvantage of this
device is that it is usually not possible to measure water levels with it.
A s e c o n d type of multilevel sampler can be used in b e d r o c k boreholes. In this
design (Figure 8 . 2 4 ) packers are located above and b e l o w each sampling port; when
they are inflated, they seal off that part of the b o r e h o l e . This is a permanendy installed
device, but it can be removed and reused if desired.
B a r k e r et al. ( 1 9 8 7 ) evaluated the bias in samples that can be introduced by the
use of multilevel piezometers constructed out of flexible tubing. Leaching of plastics
from the plastic tubing is o n e problem. Another problem is the sorbing of organics by
the tubing. B o t h these p r o b l e m s can be minimized by using T e f l o n * tubing and thor

G r o u n d - W a t e r and

Soil

Monitoring

377

Field installation

s End cap