Professional Documents
Culture Documents
/ ' g r o u p s 5 IS
1A , '^
1 J \^ >
H 2
1.00794
2A ?cS a THE UNIVERSITY Of ucEKbU svD ARY
3 4
1'
PHYSICAL SCIENCES &ENGINEERING LIBRARY
Li Be ^ -------------------------------- Transition m etals --
6.941 9.01218
11 12 / 3 4 5 6 7 8 9
Na Mg
22.98977 24.305
3B 4B 5B 6B 7B ^ ------- 8B
19 j 20 y 21 22 23 24 25 26 27
K Ca Sc Ti V Cr Mn Fe Co
39.0983 40.078 44.9559 47.88 50.9415 51.996 54.9380 55.847 58.9332
37 38 39 40 41 42 43 44 45
Rb Sr Y Zr Nb Mo Tc Ru Rh
85.4678 87.62 88.9059 91.224 92.9064 95.94 (98) 101.07 102.9055
55 56 57 72 73 74 75 76 77
Cs Ba *La Hf Ta w Re Os Ir
132.9054 137.33 138.9055 178.49 180.9479 183.85 186.207 190.2 192.22
87 88 89 104 105 106 107 108 109
Fr Ra ^Ac Rf Db Sg Bh Hs Mt
(223) 226.0254 227.0278 (261) (262) (263) (262) (265) (266)
58 59 60 61 62
*Lanthanide series Ce Pr Nd Pm Sm
140.12 140.9077 144.24 (145) 150.36
90 91 92 93 94
^Actinide series Th Pa u Np Pu
232.0381 231.0359 238.0289 237.048 (244)
3 4 0 6 7 0 2 1 7 8 598 4
(269) (272)(277)
63 64 65 66 67 68 69 70 71
Eu Gd Tb Dy Ho Er Tm Yb Lu
151.96 157.25 158.9254 162.50 164.9304 167.26 168.9342 173.04174.967
95 96 97 98 99 100 101 102 103
Am Cm Bk Cf Es Fm Md No Lr
(243) (247) (247) (251) (252) (257) (258) (259) (260)
nirffl
TT
Principles and Applications of
Geochemistry
PRINCIPLES AND APPLICATIONS OF
GEOCHEMISTRY
A Comprehensive Textbook for Geology Students
Second Edition
GUNTER FAURE
T H E O H IO S TA TE U N IV E R S IT Y
Prentice Hall
Upper Saddle River, New Jersey 07458
L ib rary o f C ongress C ataloging -in -P ublication D a ta
Faure, Gunter.
Principles and applications of geochemistry : a comprehensive
textbook for geology students/G unter Faure.2nd ed.
p. cm.
Rev. ed. of: Principles and applications of inorganic
geochemistry. cl991.
Includes bibliographical references and index, ISBN
0-02-336450-5
1. Geochemistry I. Faure, Gunter. Principles and applications
of inorganic geochemistry. II. Title.
QE515.F28
551.9 dc21
1998
L 97-44563
.
C IP
O ( . QL/bOo?
THE UNIVERSITY OF QUEENSLAND LIBRASW
PSCE.
ISBN Q-DS-33b4SD-5
ISBN 0-02-33b450-5
1
What Is Geochemistry? 2 Part II
2
1.1 Early History
PRINCIPLES OF INORGANIC
1.2 G eochem istry in the U.S.S.R. 3 GEOCHEM ISTRY
1.3 V. M. G oldschm idt 4
1.4 M odern G eochem istry 5
R eferences 7
5
2 The Electronic Structure
of Atoms 60
In the Beginning 8
8
5.1 The A tom of Thom son and
2.1 The Big Bang
1 1
R utherford 60
2.2 Stellar Evolution
5.2 B o h rs Theory of the H ydrogen
2.3 Nucleosynthesis 13
2 0 A tom 61
2.4 Sum m ary
2 0 5.3 Em ission of X -rays 63
Problem s
2 0 5.4 Schrdingers M odel of the A tom 64
References
5.5 The A ufbau Principle 67
3
5.6 Sum m ary 69
Problem s 72
The Solar System 22 R eferences 72
2 2
3.1 Origin of the Solar System
Chemical Differentiation
6 .2
The M odern Periodic Table 74
of the Earth 43 6.3 Basic Principles of A tom ic Physics 77
4.1 Internal Structure of the E arth 43 6.4 A tom ic W eights 78
6.5 Sum m ary 81
4.2 The C ontinental Crust: M ajor Problem s 82
Elem ents 46 R eferences 82
VII
V III CO N TEN TS
Adm ission
8.3 Coupled Substitution: Key to the 11
, Thermodynamics 155
1 0 2
Feldspars
8.4 D istribution Coefficients and 103
1 1 .1
Definitions 155
G eotherm om eters
1 . 2 The First Law 157
8.5 G eochem ical Classification of 1
Part V
21
APPLICATIONS O F GEOCHEM ISTRY
Chemical Weathering of Mineral
TO THE SOLUTION OF GLOBAL
Deposits 400
PROBLEMS 2 1.1
M etallic M ineral D eposits 400
2 1
Oxidation of Iron Sulfides and the
.2
19
Role of Bacteria 401
Consequences of Chemical
21.3 E h -pH D iagram for C opper
Weathering 342 M inerals 403
19.1 Changes in Chem ical Com position 21.4 Supergene E nrichm ent of F e -C u
Sulfide D eposits 407
of Rocks 343
21.5 R eplacem ent of Pyrite by
19.2 N orm ative M ineral Com position
Chalcocite 408
of W eathering Products 345 2 1 .6
Oxidation of O re M inerals of
19.3 Susceptibility of M inerals to
O ther M etals 411
W eathering 347
21.7 Geochem ical Exploration 413
19.4 Form ation of Placer D eposits 351 2 1 .8
Production and C onsum ption of
19.5 Provenance D eterm ination by
M ineral Resources 417
Isotopic D ating 353
21.9 Summary 421
19.6 Form ation of Soils 354
Problem s 422
19.7 G eom icrobiology 358
R eferences 423
19.8 Food Production and Population
22
G row th 361
19.9 Summary 363
Problem s 364
R eferences 365 Geochemical Cycles 425
2 2 .1
The Principle of Mass Balance 426
2 2 .2
Mass Balance for M ajor Elem ents
in the Ocean 428
20 22.3 Mass Balance for Trace Elem ents
in the Ocean 431
Chemical Composition of Surface
22.4 The Cycles of C -H -O -N 433
Water 368 22.5 The Sulfur Cycle 443
20.1 Chem ical Analysis of W ater in 2 2 .6
Summary 446
Stream s 368 Problem s 447
20.2 Chem ical Com position of Stream s 370 R eferences 447
CO N TEN TS
25
23
Chemistry of the Atmosphere 449 Effect of Environmental Lead on
450 Human Health 485
23.1 Structure and Composition
23.2 U ltraviolet R adiation 451 25.1 Isotope C om position of
452 Environm ental Lead
23.3 O zone in the Stratosphere 453 25.2 L ead in the E nvironm ent
23.4 O zone H ole over A ntarctica 25.3 L ead in Plants
23.5 Infrared R adiation and the G 455 25.4 L ead Poisoning of Cows and H
orses
reenhouse Effect 457 25.5 H um an B ones and Tissues
23.6 Prediction of the Future G lobal 458 25.6 Lead in the B ones of A ncient
459 Peoples
Climate
459 25.7 Sum m ary
23.7 Sum m ary Problem s R
Problem s R eferences
eferences
Appendix A
24 461 Compilations of Geochemical D ata
Environmental Geochemistry: 462
Disposal of Radioactive Waste 461 465 Appendix B
24.1 The Big Picture
468 Standard Gibbs Free E nergies ( G f)
24.2 H igh-Level N uclear W aste D isposal
24.3 G eological Disposal of Radioactive W 473 and Standard Enthalpies of
aste Form ation (H f )
24.4 G eochem istry of Plutonium 476
24.5 E h - p H Diagrams of N eptunium
478 Author Index 575
and Plutonium
24.6 A nalog Studies: The N atural R
481
eactors at Oklo, G abon
482
24.7 R eactor Accidents: Chernobyl, U
483 Subject Index 589
kraine
24.8 Sum m ary
Problem s R
eferences
Preface
This textbook is intended as an introduction to geo These chapters explain the concept of chemical
chemistry for geology students in their senior year or in equilibrium and use the Law of Mass A ction to study the
their first year of graduate work. A t that time in their dissociation of weak acids and bases and to calculate the
education, students are ready to releam those principles solubility of am orphous silica as a function of the pH .
of chemistry that are especially applicable to the study of Next, the dissociation of salts, the hydrolysis of their
geological processes. G eochem istry can enhance their ions, and the necessity of using the activities of ions
understanding of these processes and can teach them to rather than their molal concentrations in M ass A ction
apply principles of chem istry to the solution of geolog problem s are pre sented. A ll of these concepts com e
ical problem s. Geologists in virtually all branches of the into play in the carbonate equilibria involving calcite, its
science can benefit from an introductory course in ions in aqueous solution, and C 0 2 in the atm osphere.
geochem istry because it can help them to m ake However, the incongruent dissolution of micro-cline to
quantitative predictions about the o u t com e of chem form koalinite plus ions leads to an impasse because the
ical reactions that occur in the con text of m any value of the equilibrium con stant for that reaction has not
geological processes. been m easured.
The subject m atter of this book is presented in five Therefore, C hapter 11 contains a guided tour of the
parts. P art I: P lanet E arth in the Solar System (C hapters principles of therm odynam ics leading up to a derivation
1 through 4) presents the big picture starting w ith the Big of the Law of M ass A ction for reacting mixtures of ideal
Bang, stellar evolu tion, nucleosynthesis, the solar gases at equilibrium . In addition, this chapter dem
system, and the onstrates that equilib rium constants at the standard tem
geochem istry of the E arth . This p art also con perature and pressure can be calculated from the Gibbs
tains a brief history of geochem istry leading up free energy change of chem ical reactions. This skill is im
to a statem ent of its goals and stating the them e of this m ediately used to construct ion -activity dia grams for
book: Studies of geochem istry convert idle speculation alum inosilicate m inerals including zeo lites (C hapter
into understanding. 12) and clay m inerals (C hapter 13).
The second part, Principles of Inorganic
Geochemistry (Chapters 5 through 8 ), starts with the Finally, oxidation -reduction reactions are
electronic structure of atom s and dem on strates how w introduced in C hapter 14, and the previously acquired
ave mechanics and the Aufbau prin ciple lead directly to know ledge of therm odynam ics is used to explain some
the periodicity of the chemical properties of the elements. basic principles of electrochem istry. These insights are
Once the periodic table has been constructed, systematic then used to define the stability limits of w ater on the E
variations in bonding, ionic radii, and the structure of arth and to con struct E h -p H diagram s for the oxides of
crystals becom e apparent, thus preparing the way for a iron and their ions. C hapter 15 deals with the kinetics of
discussion of ionic substitution based on the rules of G chemical reactions, and with the transport of ions by
oldschm idt and Ringwood. Part II concludes w ith diffusion and advection. These concepts are used to
discussions of coupled substitution in feldspar, model the grow th of concretions during diagenesis and
distribution coefficients, and the geo chemical the form ation of m onom ineralic layers at the sedim ent-
classification of the elements. w ater interface.
Part III is entitled A queous Geochemistry and the P art IV is entitled Isotope G eochem istry and M
Stability off M inerals (Chapters 9 through 15). ixing (C hapters 16 through 18). Following
X III
X iv PREFACE
the p resentation of the different m odes o f decay resulting release of m etals into the environm ent is the
and a statem ent of the Law of Radioactivity, basis for geochem ical m ethods of prospecting. The
C hapter 16 continues with the derivation of the chapter ends with a discussion of mineral economics
geochronom etry equation based on the accum u lation of including King H u b b erts prediction of the ultim ate
stable radiogenic daughter nuclides. This equation is then depletion of finite and nonrenew able resources such as
used to explain the principles and assum ptions of dating petroleum .
by the K -A r, R b-Sr, Sm -N d, R e -O s, and U, T h -P b The m igration of ions released by weathering of
m ethods. In addi rocks and m etallic m ineral deposits on the con tinents
tion, the chapter contains descriptions of the leads to a consideration of geochemical cycles in C hapter
2 3 0 T h /238U m ethod of m easuring sedim entation 22. A fter presenting the concept of mass balance, this
rates and of dating by the cosm ogenic radionu chapter contains summaries of the geochem ical cycles C
clides 1 4 C, 1 0 Be, and 2 6 A1. -H -O -N, including water. The links betw een the
C hapter 17 is devoted to the fractionation of the reservoirs of these and other elem ents play an im portant
stable isotopes of H, O, C, and S in nature. This chapter role in sta bilizing the chemical com position of the atm
includes derivations of the basic m athe m atical os phere and the oceans.
relationships used to in terp ret variations
of the isotope com positions of m eteoric water, C hapter 23 on the chem istry of the atm os phere
m arine calcite, clay m inerals, oilfield brines, continues this train of thought by present ing the basic
hydrocarbons, and sulfide minerals. In addition, C hapter facts about the structure and chemical com position of the
17 presents a discussion of the time-dependent variation atm osphere. This inform ation is then used to explain the
of the isotope com positions of Sr and S in m arine form ation of ozone in the stratosphere and its destruction
carbonates of Phanerozoic age and the use of 8 7 S r/86Sr by anthropogenic emissions of C l-bearing gases,
ratios for dating Cenozoic m arine carbonates. especially over A ntarctica. The chapter also con tains a
presentation of the radiation balance of the E arth and the
P art IV concludes with C hapter 18 on mixing and perturbation of this balance by emissions of gases that
dilution of w ater and sedim entary rocks that are m enhance the absorption of infrared radiation in the atm
ixtures of two or three com ponents having different osphere. C hapter 23 ends with a long-range forecast of
chemical compositions. The chapter con tinues with tw o the effects of global warm ing on the tim e scale of 1 0 4
-com ponent mixtures containing elem ents having years based on the w ork of W. S. Broecker.
distinctive isotope com positions (Sr, Pb, O, H , etc.). The
systematics of such mix tures are briefly illustrated using
oilfield brines form ed by mixing two com ponents having The dangers of environm ental contam ination
differ en t ^ S r/^ S r ratios and Sr concentrations. are illustrated in C hapter 24 by consideration of
the problem s associated with high-level radioac
The fifth and last part of this book is entitled A tive waste generated by nuclear reactors. The chapter
pplications of G eochem istry to the Solution of G lobal explains how this waste is generated and how it is (or
Problem s (C hapters 19 through 25). This p art treats the should be) processed prior to perm a nent storage in
consequences of chemical weath ering of rocks, such as underground repositories. The geo chemical properties of
the form ation of soils and their agricultural productivity the transuranium elements, which play an im portant role
(C hapter 19); the chem ical com position of surface w in the long-term safety of such repositories, are briefly
ater, its evolu tion by evaporative concentration, and its discussed using E h -p H diagrams for neptunium and
conta m ination by natural and anthropogenic causes (C plutoni um. Chapter 24 ends with a sum m ary of the
hapter 20); and the w eathering of m etallic min eral events and consequences of the explosion of a nuclear
deposits based on E h - p H diagram s and including a reactor in Chernobyl, U kraine, on A pril 26,1986.
section on the im portance of bacteria in the oxidation of
iron sulfides (C hapter 21). The The final chapter sum m arizes the geochem istry of
Pb in the environm ent and its effect on
PREFACE XV
M iami U niversity (O hio); D onald I. Siegel,
hum an health. Lead is dispersed primarily
through the atm osphere in the form of aerosol Syracuse University, R obert D. Shuster, U niv
particles. Some civilizations have suffered the con ersity of Nebraska, O m aha; and A . Russell Flegal, U
sequences of Pb toxicity because this m etal was used as a niversity of California, Santa Cruz. The second edition has
food additive and in household utensils. T he effects of Pb benefited significantly from com m ents by colleagues
poisoning on hum ans are subtle and difficult to identify who have pointed out errors of various kinds in the first
because all hum ans alive today are contam inated with Pb edition and suggested im provem ents in the presentation
released prim ari ly because of its use as an additive in of the subject m atter. I thank these individuals by nam e
gasoline. The in alphabetical order: W alther M . B arnard, G eorges
im pairm ent of m ental functions as well as the Beaudoin, Phillip Boger, D aniel M arcos Bonotto, Steven
toxic effects of Pb were tragically dem onstrated E . Bushnell, C hristopher Conaway, E than L. G rossm
by the m em bers of Sir John Franklins expedition to the an, G iehyeon Lee, Peter C. Lichtner, James O Neil, R
C anadian A rctic in 1845-47. isto Piispanen, R obert D. Shuster, and Terri L. Woods. I
The presentation of the subject m atter in this book also thank R obert A. M cConnin, Executive E ditor of
starts with basic principles and emphasizes quantitative m Prentice Hall, for his interest in this book, Elisabeth H . B
ethods of problem solving in order to gain better elfer for
understanding of natural phenom e na. A lthough this
book is intended to be an intro duction to geochemistry, her role in its production, and Susan M.
the references at the end of each chapter will perm it M cM ullen for drafting the diagram s of the first
students to pursue topics of their choice in the library and edition. I am grateful to B etty H eath for typing
thereby approach the state o f the art. In addition, the end- part of the m anuscript for the second edition.
of-chapter problem s enable students to test their The ultim ate purpose of this book is to
understanding of the principles. The book ends encourage students of the E arth to adopt the
quantitative approach to problem solving that is exem
with a com pilation of geochemical data in plified by geochem istry, and to m ake them aw are of the
A ppendix A and of therm odynam ic data for a im portant role E arth scientists m ust play in the effort to
large num ber of elem ents and their com pounds in A preserve our h ab itat on the E arth .
ppendix B. In addition, m ost chapters contain data tables
and diagram s that will m ake the book a useful source o f G U N T E R FA U R E
geochemical data. The diagram s have lengthy captions Postscript: A m anual with solutions to the
that should encourage browsing and will help users to
extract inform a tion from them w ithout necessarily
having to reread the entire chapter. end-of-chapter problem s is available to instruc
tors from the publisher by writing on school sta tionery to
This geochem istry text dem onstrates th at the the Geology E ditor, Prentice Hall, 1 Lake Street, U pper
principles and applications of geochem istry are internally Saddle River, NJ 07458.
consistent, useful for understanding the world around us,
and accessible to anyone willing to m ake th e effort. In
addition, this book m akes the point th at geochem ists
have an im portant task to perform in society by assisting
in the prudent use of natural resources and in the safe
disposal of the resulting waste products.
F uture generations may rem em ber the 20th century prim arily for the advances in tech nology that
have enabled us to begin the exploration of the solar system . O ur result ing aw areness of the celestial
environm ent of the E arth has raised questions about the origin and evolution of the solar system, of
the M ilky Way Galaxy, and of the U niverse at large. The start of an introductory geochem istry
course is an appropriate tim e to sat isfy our desire to know the origins of the world in which we live.
1
What Is Geochemistry?
Geology began when early man first picked up a stone, considered its qualities,
and decided that it was better than the stone he already had. Good stones were
useful and they were collected, mined, and traded.
stones. Fersm an earned this accolade by writing books ir : -asingly relevant as we attem pt to predict the
intended for the layman: M ineralogy fo r cilimntic consequences of the greenhouse effect caused
Everyone (1928), Recollections about a Stone (1940), M by the discharge of carbon dioxide and o ther gases into
y Travels, Stories about Precious Stones, the atm osphere. We are carrying omt an experiment on a
and Geochemistry fo r Everyone (1958). The last global scale the outcome o f which could be detrim ental
two w ere published posthum ously. F ersm ans to life. The response of the biosphere to this perturbation
conviction that geochem istry should serve his of
country is reflected in his words: We cannot be m erely tlhe environm ent may be crucial to the ultimate
idle adm irers of our vast country; we m ust actively help outcom e of this experim ent.
reshape it and create a new life (Shcherbakov, 1958). V. I. Vernadsky is revered in the U.S.S.R. as
oine of the giants of science in the 2 0 th cen
The scientific legacy of Fersm ans form er teacher, tury. His teachings still m otivate the geochem
Vladim ir Ivanovich Vernadsky (1863-1945), is similarly ists working at the V ernadsky Institute 'of
impressive, but it has not been G eochem istry and A nalytical C hem istry in
appreciated in the w estern world. Vernadsky Moscow, which is am ong the forem ost geochem i cal
em phasized the im portance of the activities of research institutes in the world.
living organisms in geological and geochemical processes.
In fact, he regarded living m atter as the m ost pow erful
geological force and w rote that the E a rth s crust
1.3 V. M. Goldschmidt
originated from the biosphere. These convictions about T he Institute of G eochem istry at the University
the im portance of life on E arth were not widely accepted o>f G ttingen in G erm any was m ade fam ous by
by geologists in N orth A m erica, partly because tlhe work of Victor M. G oldschm idt (1888-1947). G
Vernadsky wrote in m any languages (Russian, German, oldschm idt earned his d o cto rate at the U niversity of
French, Czech, Serbo-Croatian, and Japanese). His most Oslo in 1911 with a study of contact nnetam orphism
im portant books (Vernadsky, 1924, 1929) were written in based on the therm odynam ic p'hase rule. In the
French and published in Paris. However, Vernadsky was following year Max von Laue discovered diffraction of x-
also far ahead of his contem po raries by concluding that rays by crystals and thereby provided a m ethod for
the biosphere may have controlled the environm ent on determ ining the crystal structure of a m ineral and the
the surface of the E arth from the very beginning, long radii of the ions of which it is com posed. From 1922 to
before the existence of life in Early Precam brian time was 1926 G oldschm idt and his associates at the University of
known (Lapo, 1986). Vernadskys views on this subject Oslo used x-ray diffraction to determ ine the
are now reflected in the Gaia Hypothesis of Lovelock
(1979), made popular on American television in the crystal structures of many m inerals (Goldschmidt,
Planet E arth series by the Public Broadcasting System. 1930). In 1930 G oldschm idt m oved to the
Vernadsky also iden tified the biosphere as the principal Unrversity of G ttingen, w here he studied the
transform er of solar energy into chemical energy. This distribution of the chemical elem ents using an
idea is now widely accepted. For exam ple, Cloud (1983) optical spectrograph. From this body of data he deduced
described the biosphere as a huge metabolic device for that the chemical com position of m iner als is determined
the capture, storage and transfer of energy. Fossil fuels by the requirem ents of closest packing of ions. M
(coal, oil shale, petroleum , and natural gas) are now oreover, the substitution of the ions of a major elem ent by
regarded as deposits of solar energy collected by the the ions of a trace ele m ent depends on the similarity of
biosphere. their radii and charges. These generalizations provided a
rational explanation for the observed distribution of the
elem ents in the minerals of the crust of the E arth .
Goldschmidt returned to Oslo in 1935 b u t
V ernadskys views on the im portance of the w as unable to continue his w ork after th e
biosphere (V ernadsky, 1945) are becom ing
1.4 MODERN GEOCHEMISTRY 5
G erm an invasion of Norway in 1940. H e fled to N. L. Bowen and his colleagues, and by their suc cessors,
Sw eden in 1942 and ultim ately made his way to at that laboratory have becom e the foun dation of m
E ngland, w here he w orked at the M acaulay Institute for odern igneous petrology. B ow ens book, The Evolution o
Soil R esearch. His health had seri ously deteriorated as a f the Igneous Rocks, pub lished in 1928, turned igneous
consequence of im prison petrology from a preoccupation with the description and
m en t in co n cen tratio n cam ps in Norway. classifica tion of igneous rocks tow ard a concern for their
G oldschm idt returned to Oslo in 1946 but died origin and the geochemical differentiation of the E arth by
th ere prem aturely in 1947 at the age of 59. H e left m agm atic activity.
behind the m anuscript of a partially com pleted book
entitled G eochem istry, which was com plet ed by A lex In 1952 B rian M ason published the first edi
M uir and published posthum ously in 1954. A special tion of Principles o f Geochemistry, which was
com m em orative issue of A p p lie d widely used as a textbook at universities and helped to
G e o c h e m istry (1988) was devoted to establish geochem istry as a legitim ate com ponent of E
G oldschm idt on the 100th anniversary of his arth Science. In the next decade R obert M. Gar'rels and K
birthday. onrad B. K rauskopf used therm odynam ics and solution
G oldschm idts principal contribution to geo chem istry to determ ine the stability of m inerals and the
chem istry is the rational explanation of isomor- m obil ity of their ions at the surface of the E arth . They
phous substitution in crystals based on the com patibility trained the m odern generation of geochem ists by their
of the radii and charges of the ions. H e thereby achieved own research and through their popular
one of the m ajor goals of geo chemistry. Nevertheless,
the urge to know and to understand the distribution of the textbooks (G arrels, 1960; G arrels and Christ,
chemical ele m ents has continued to m otivate the work 1965; Krauskopf, 1967).
of sev G eochem ical prospecting, as we know it
eral outstan d in g geochem ists, notably L. H. today, was initiated in 1947 by the U.S. Geological
A h ren s in South A frica, S. R. N ockolds in Survey (Hawkes and Lakin, 1949) and was based on
England, S. R . Taylor in Australia, K. R ankam a in colorimetry. The techniques developed in the
Finland, K. H . W edepohl in Germany, A . R U nited States w ere used in 1953 for soil surveys
V inogradov in the U.S.S.R., D. M. Shaw in in the U nited Kingdom and in A frica (Webb,
Canada, and K. K. Turekian in the U nited States. 1953). Subsequently, the Geochemical Prospecting
R esearch C enter at the Im perial College of
Science and Technology in London becam e a cen ter of
activity in this aspect of geochemistry. Geochemical
1.4 Modern Geochemistry prospecting continues to be one of the most im portant
Beginning in the 1950s, geochem ists tu rn ed practical applications of geo chemistry and now relies on
increasingly to the study of chemical reactions and a wide range of sophisticated analytical techniques.
processes. The roots of this line of research can be found
in the work of A . Fersman, who used the concepts of E nvironm ental geochem istry has arisen recently
therm odynam ics to study the stability of m inerals in because of the need to m onitor the dis persion of m etals
their natural environment. In addi tion, J. H . vant H off and various organic com pounds that are introduced into
(1852-1911) studied the crys tallization of salts by the environm ent as anthropogenic contam inants. This
evaporating seaw ater in Berlin and, in 1904, the new applica tion of geochem istry to the welfare of hum
Geophysical Laboratory of the Carnegie Institution of W ankind is closely related to pollutant hydrogeology and
ashington was founded in the U nited States. The principal medical geochemistry. Geochem ical prospecting and
objec tive of the Geophysical Laboratory has been to environmental geochemistry are of param ount im
study the origin of igneous rocks and ore deposits by portance because they contribute to the con tinued well-
experim ental methods. The results achieved by being of the hum an species on Earth .
6 W HAT IS GEOCHEMISTRY?
Since about the middle of the 20th century In fact, m ost E arth Science journals now publish papers
geochemistry has becom e highly diversified into many in which geochemical data or principles are used to
subdivisions, among them inorganic geo chemistry, explain a geological process or to solve a particular
organic geochemistry, isotope geochem istry, geochem problem . The strength of geo chem istry lies in the fact
ical prospecting, m edical geo chemistry, aqueous that most geological processes involve chemical reactions
geochem istry, trace-elem ent geochemistry, and in some sig nificant way. G eochem istry really is for
cosmochemistry. Each of these subdivisions has its own everyone, as Fersm an claimed in his popular book.
rapidly growing body of literature. However, the geochem
ists of the world have organized scientific societies that W hat then is geochemistry? There is no easy answ
transcend the boundaries of specialization and help to er to this question because geochemists con cern them
unify the field. These societies also publish journals that selves with a wide range of natural phe nom ena and use
prom ote the free flow of inform ation and ideas am ong many different techniques to study them . According to
their m em bers. For example, the Geochem ical Society Fersman, Geochemistry studies the history of chemical
and the M eteoritical Society both sponsor Geochimica et elements in the
Cosmochimica Acta. E a rth s crust and their behavior under different
therm odynam ic and physicochem ical natural
The E uropean A ssociation for G eochem istry has conditions. G oldschm idts definition of geo
adopted the journal Chemical Geology, and the chem istry was G eochem istry is concerned with the
International A ssociation of G eochem istry and Cosm laws governing the distribution of the chem i cal elem ents
ochem istry sponsors A p p lied Geochemistry. A m ong and their isotopes throughout the E a rth (Correns,
other journals that publish papers dealing with geochem 1969). We could go on and on. G eochem istry is a highly
istry are diversified and con stantly evolving subject that can be
described in m any different ways.
Earth and Planetary Science Letters
Contributions to M ineralogy and Petrology For our purposes the m ajor goals of geo chem ists
Journal o f Geochemical Exploration are
E conom ic Geology
1. To know the distribution of the chemical elem
A m erican Journal o f Science
ents in the E arth and in the solar system .
Nature
Science 2. To discover the causes for the observed chemical
composition of terrestrial and extraterrestrial
Some journals are closely associated with certain regions,
materials.
for example,
3. To study chemical reactions on the surface of the
G eokhim ia Earth, in its interior, and in the solar system
D oklady (E arth around us.
Science Section) 4. To assemble this inform ation into geo chemical
Lithology and Russia cycles and to learn how these cycles have
M ineral Resources operated in the geologic past and how they may
Geochemistry be altered in the future.
International These are the lofty goals of a scientific disci pline. T
Geochemical Journal Japan heir intrinsic m erit is apparent. How ever, geochem ists
Chinese Journal P eoples Republic practicing their craft today also have an obligation to hum
o f Geochemistry of China ankind to assist in the devel
REFERENCES 7
O
WEN
C ertain questions about our existence on E arth are so Form ation of atomic nuclei began about
fundam ental that they have been incorpo rated into 13.8 sec after the Big Bang when the tem pera
religious mythologies. These questions not only concern ture of the universe had decreased to 3 X 109 K. This
the origin of the E arth and the evolution of life but also process continued for about 30 min, but did not go
extend to the origin of the universe and to the nature of beyond helium because the nuclear reac tions could not
space and time. bridge a gap in the stabilities of the nuclei of lithium,
D id the universe have a beginning and will it beryllium, and boron. A t that time the universe was an
ever end? W hat existed before the universe intensely hot and rapidly expanding fireball.
form ed? D oes the universe have limits and what exists
beyond those limits? It is proper to raise these questions Some 700,000 years later, when the tem per ature
at the beginning of a geochem istry course because they had decreased to about 3 X 103 K, elec trons becam e
are within the scope of cosmochemistry. attached to the nuclei of hydrogen and helium . M atter
and radiation were thereby separated from each other,
and the universe becam e transparent to light.
Subsequently, m at ter began to be organized into stars,
2.1 The Big Bang
galaxies, and galactic clusters as the universe continued
The universe started like a bubble in a stream . At first it to expand to the present time (W einberg, 1977).
was not there, and suddenly it form ed and
expanded rapidly as though it w ere exploding But how do we know all this? The answ er is that
(G ott, 1982). Science has its share of practical the expansion of the universe can be seen in the red
jokers who im m ediately referred to the start of the shift of spectral lines of light em itted by distant
expansion of the universe as the Big Bang (Gamow, galaxies, and it can be heard as the cosmic microwave
1952). From the very beginning the universe had all of radiation, which is the rem nant of the fireball, that still
the mass and energy it con tains today. As a result, its fills the universe. In addition, the properties of the
pressure and tem per ature, say 1CT3 2 sec after the Big universe im m edi ately after the Big Bang were similar
Bang, w ere so high that m atter existed in its m ost to those of atom ic nuclei. Therefore, a very fruitful
fundam ental form as quark soup. As the universe collabo ration has developed among nuclear physicists
expanded and cooled, the quarks com bined to form more and cosmologists that has enabled them to reconstruct
fam iliar nuclear particles th at ultim ately becam e the history of the universe back to about 1CT3 2 sec after
organized into nuclei of hydrogen and helium . the Big Bang. These studies
8
2.1 THE BIG BANG 9
have shown that the forces we recognize at low tem characteristic spectral lines of light they emit. This red
perature are, at least in p art, unified at extrem ely high shift is related to the recessional veloc ity by an equation
tem peratures and densities. There is hope that a G rand derived in 1842 by Johann Christian D oppler in Prague:
Unified Theory (G U T ) will eventually em erge th at
may perm it us to approach even closer to understanding
the start of the universe.
where A' is the wavelength of a spectral line of light em
W hat about the future? Will the universe continue to itted by a moving source, A is the wave length of the same
expand forever? The answer is that the future of the line em itted by a stationary source, c is the velocity of
universe can be predicted only if we know the total am light, and v is the reces sional velocity. H ubbles estim
ount of m atter it contains. The m atter th at is detectable ates of the dis tances to the galaxies were based on the p
at the present tim e is not sufficient to perm it gravity to roper
overcom e the expansion. If expansion continues, the ties of the C epheid Variables studied previously
universe will becom e colder and em ptier with no by H . S. Leavitt and H. Shapley at H arvard
prospect of an end. However, a large fraction of the mass University. The C epheid Variables are bright stars in the
of the universe is hidden from view in the form of gas and constellation C epheus whose period of variation depends
dust in interstellar and intergalactic space, and in the on their absolute luminosity, which is the total radiant
bodies of stars that no longer em it light. In addition, we energy em itted by an astronom ical body. H ubble found
still cannot rule out the pos sibility that neutrinos have such variable stars in the galaxies he was studying and
mass even when they are at rest. If the mass of the deter m ined their absolute luminosities from their peri
universe is sufficient to slow the expansion and ultim ods. The intensity of light em itted by a star decreases as
ately to reverse it, then the universe will eventually the square of the distance increases.
contract until it disappears again in the stream of time.
Therefore, the distance to a star can be deter
m ined from a com parison of its absolute and its
Since the universe had a beginning and is apparent luminosity, w here the latter is defined as the
still expanding, it cannot be infinite in size. radiant pow er received by the telescope per square
How ever, the edge of the universe cannot be centim eter. In this way, H ubble deter m ined the
seen with telescopes because it takes too long for the light recessional velocities and distances of the galaxies in the
to reach us. As the universe expands, space expands with Virgo cluster and expressed their relationship as:
it. In other words, it seems to be im possible to exceed the
physical limits of the universe. We are trapped in our v=Hd (2.2)
expanding bu b ble. If other universes exist, we cannot
com m uni cate with them . where v is the recessional velocity in km /sec, d is the
distance in 106 light years, and H is the H ubble constant
Now that we have seen the big picture, let us review (H ubble, 1936).
certain events in the history of the stan dard m odel of The H ubble constant can be used to place a limit on
cosmology to show that progress in the age of the universe. If two objects are moving apart
Science is som etim es accidental. with velocity v, the tim e (t) required for them to becom e
In 1929 the Am erican astronom er Edw in separated by a distance d is:
H ubble reported that eighteen galaxies in the
Virgo cluster are receding from E arth at different rates th
at increase with their distances from Earth. H e calculated
the recessional velocities of these galaxies by m eans of The initial results indicated that the H ubble con stant had
the D oppler effect from observed increases of the a value of 170 k m /sec/1 0 6 light years,
wavelengths of
10 IN THE BEGINNING
which corresponds to an expansion tim e of less than 2 X was agreed that they would publish two com pan ion
109 years. This result was very awkward because age letters in the Astrophysical Journal. Penzias and Wilson
determ inations based on radioactivi ty had established announced the discovery, and Dicke and his colleagues
that the E arth is older than this date. Eventually W alter B explained the cosmological significance of the microwave
aade discovered an error in the calibration of the C epheid background radia tion (Penzias and Wilson, 1965). In
Variables, and the value of the H ubble constant was 1978 Penzias and Wilson shared the N obel Prize in
revised (B aade, 1968). The presently accepted value is physics for their discovery.
15 k m /sec/1 0 6 light years, which indicates an The radiation discovered by Penzias and
expansion time for the universe of less than Wilson is a rem nant of the radiation that filled
20 X 109 years. This date is com patible with inde the universe for about 700,000 years w hen its
p endent estim ates of its age based on considera tion of tem perature was greater than about 3000 K.
nucleosynthesis and the evolution of stars. By com bining During this early period, m atter consisted of a
all three m ethods H ainebach et al. m ixture of nuclear particles and photons in ther
(1978) refined th e age of the universe to mal equilibrium with each other. U nder these
(14.5 1.0) X 109 years. conditions the energy of radiation at a specific
The Big Bang theory of cosmology was not w avelength is inversely p ro p o rtio n al to the
accepted for m any years for a variety of reasons. absolute tem perature. A ccording to an equation derived
The turning point came in 1964 w hen A rno A. by Max Planck at the start of the 20th century, the energy
Penzias and R o b ert W. W ilson discovered a of blackbody radiation at a particular tem perature
m icrow ave background rad iatio n th at corre increases rapidly with increasing wavelength to a m axim
sponds to a blackbody tem perature of about 3 K. The um and then decreases at longer wavelengths. R adiation
discovery of this radiation was accidental, in therm al equilibrium with m atter has the same
even though its existence had been predicted
properties as radiation inside a black box with opaque
tw enty years earlier by G eorge G am ow and his
walls. Therefore, the energy distribution of radiation in the
colleagues R alph A . A lpher and R obert Herm an.
early universe is related to the wavelength and to the
Because they w ere unaw are of G am ow s work, Penzias
absolute tem perature by Plancks equation . The
and W ilson w ere skeptical about the phenom enon they
wavelength near which most of the energy of blackbody
had discovered and took great care to elim inate all
radiation is con centrated (Amax) is approxim ately equal
extraneous sources of the background radiation. For
to:
example, they noticed th a t two pigeons had been nesting
in the throat of the antenna they were using at H olm del,
New Jersey. The pigeons w ere caught and taken to a
distant location, but prom ptly returned. Therefore, they w
ere caught again and dealt with by more decisive where Amax is m easured in centim eters and T is in
means. The pigeons had also coated the antenna with a Kelvins (W einberg, 1977).
white dielectric m aterial, which was carefully removed. The original m easurem ent of Penzias and Wilson
However, the intensity of the background radiation rem was at a wavelength of 7.35 cm, which is much greater
ained constant and independent of time in the course of a than the typical w avelength of radi
year. ation at 3 K. Since 1965, m any additional m ea
surem ents at different wavelengths have con
W ord of this phenom enon reached a group of
firm ed th at the cosmic background radiation
astrophysicists at nearby Princeton University who w ere
does fit Plancks form ula for blackbody radiation . The
working on models of the early history of the universe
characteristic tem perature of this radiation is about 3 K,
under the guidance of R obert H. Dicke. Eventually,
indicating th at the typical w avelength of photons has
Penzias called Dicke, and it
increased by a factor of about 1 0 0 0
2.2 STELLAR EVOLUTION 11
because of the expansion of the universe since its tem clouds of gas and dust may contract to form new stars
perature was 3000 K (Weinberg, 1977). whose evolution depends on their masses and on the H /H
e ratio of the gas cloud from which they form ed.
2.2 Stellar Evolution The evolution of stars can be described by specifying
M atter in the universe is organized into a hier archy of their luminosities and surface tem pera tures. The lum
heavenly bodies listed below in order of decreasing size. inosity of a star is proportional to its mass, and its surface
tem perature or color is an indicator of its volume. W hen
a cloud of interstel lar gas contracts, its tem perature
clusters of galaxies comets
increases, and it begins to radiate energy in the infrared
galaxies asteroids and visible parts of the spectrum . A s the tem perature in
stars, pulsars, and m eteoroids the core of the gas cloud approaches 20 X 10 6 K, energy
black holes dust particles production by hydrogen fusion becomes possible, and a
planets molecules star is born. M ost of the stars of a typical galaxy derive
energy from this process and therefore plot in a band,
satellites atom s of H and H e
called the main sequence, on the H ertzsprung -R ussell
O n a subatom ic scale, space between stars and galaxies diagram shown in Figure 2.1. M assive stars, called blue
is filled with cosmic rays (energetic nuclear particles) and giants, have high luminosities and high surface tem
photons (light). peratures. The Sun is a star of interm ediate mass and has
Stars are the basic units in the hierarchy of heavenly a surface tem perature of about 5800 K. Stars that are less
bodies within which m atter continues to evolve by massive than the Sun are called red dwarfs and plot at the
nuclear reactions. M any billions of stars are grouped low er end of the main sequence.
together to form a galaxy, and large num bers of such
galaxies are associated into galac tic clusters. Stars may
have stellar companions or they may have orbiting As a star five times m ore massive than the
planets, including ghostly comets that flare briefly when Sun converts hydrogen to helium while on the
they approach the star on their eccentric orbits. The m ain sequence, the density of the core increases, causing
planets in our solar system have their own retenue of the interior of the star to contract. The core tem perature
satellites. The space betw een M ars and Jupiter contains therefore rises slowly during the hydrogen-burning phase.
the asteroids, m ost of which are fragments of larger This higher tem perature accelerates the fusion reaction
bodies that have been broken up by collisions and by the and causes the outer envelope of the star to expand.
gravitational forces of Jupiter and Mars. Pieces of the However, when the core becom es depleted in hydrogen,
asteroids have impacted as meteorites on the surfaces of the rate of energy production declines and the star
the planets and their satellites and have left a record of contracts, raising the core tem perature still fur ther. The
these events in craters. site of energy production now shifts from the core to the
surrounding shell. The result ing changes in lum inosity
O n an even sm aller scale, space betw een stars and surface tem pera ture cause the star to move off the
contains clouds of gas and solid particles. The gas is com main sequence tow ard the realm of the red giants (Figure
posed prim arily of hydrogen and of helium th at w ere 2.1).
produced during the initial expansion of the universe. In
addition, the interstellar m edi um contains elem ents of The helium produced by hydrogen fusion in the shell
higher atomic num ber th at w ere synthesized by nuclear accum ulates in the core, which continues to contract and
reactions in the interiors o f stars that have since therefore gets still hotter. The resulting expansion of the
exploded. A third com ponent consists of com pounds of envelope lowers the surface tem perature and causes the
hydrogen and carbon th at are the precursors of life. These color to turn red. A t the sam e tim e, the shell in which
hydrogen
12 IN THE BEGINNING
Figure 2.1 Stellar evolution on a Hertzsprung-Russell diagram for stars ranging from 1 to 9 solar masses. When a star
has used up the hydrogen in its core, it contracts and then moves off the main sequence and enters the realm of the red
giants, which generate energy by helium fusion. The evolutionary track and the life expectancy of stars are strongly
dependent on their masses. Stars five times more massive than the sun are nearly 1000 times brighter, have surface
temperatures of about 18.000 Kcompared to 5800 K for the Sunand remain on the main sequence only about 68
million years. Their evolution to the end of the major phase of helium burning takes only about 87 million years (Iben,
1967).
is reacting gradually thins time, hydrogen fusion in
as it moves toward the the shell around the core
surface, and the continues. The
luminosity of the star luminosities and surface
declines. These changes tem pera tures (color) of
transform a m ain- red giants become
sequence star into a increasingly variable as
bloated red giant. For they evolve, reflecting
example, the radius of a changes in the rates of
star five tim es m ore energy production in the
massive than the Sun core and shell. The
increases about 30-fold evolutionary tracks in
just before helium burn Figure 2.1 illustrate the
ing in the core begins. importance of the mass of
a star to its evolution. A
W hen the core tem perature approaches
star five times as massive
1 0 0 X 106 K, helium fusion byas
m the
eansSun
of the
is triple
1000 times
alpha process begins and brighter while on the main
converts three helium sequence and has a m ore
nuclei into the nucleus of eventful life as a red giant
carbon-12. A t the same than stars below about two
solar masses (Iben,
1967,1974).
2.3 NUCLEOSYNTHESIS13
The length of tim e a star spends on the main view as their lum inosities and surface tem pera tures
sequence depends primarily on its mass and to a lesser diminish with time. Stars that are apprecia bly m ore
extent on the H /H e ratio of its ancestral gas cloud. In massive than the Sun develop dense cores because of the
general, massive stars (blue giants) consum e their fuel synthesis of heavy chemical elem ents by nuclear
rapidly and may spend only 10 X 106 years on the main reactions. Eventually, such stars becom e unstable and
sequence. Small stars (red dwarfs) have much slower m explode as supernovas. The core then collapses until its
etabolic rates and rem ain on the main sequence for very radius is reduced
long periods of tim e exceeding 10 X 109 years. The Sun, to about 1 0 km and its density is of the order of
being a star of m odest magnitude, has enough hydrogen in 10n to 101 5 g/cm 3. Such stars are com posed of a
its core to last about 9 X 109 years at th e present rate of neutron gas because electrons and protons are forced to
consum ption. Since it form ed about 4.5 X 109 years ago, com bine under the enorm ous pressure and the abundance
the Sun has achieved m iddle age and will provide energy of neutrons greatly increases as a result. N eutron stars
to the planets of the solar system for a very long tim e to have very rapid rates of rotation and em it pulsed radio
come. H ow ever, ultim ately its lum inosity will increase, waves that were first observed in 1965 by Jocelyn Bell, a
and it will expand to becom e a red giant, as show n in graduate student working with A . Hewish at the
Figure 2.1. The tem perature on the sur face of the E arth Cavendish Laboratory of Cam bridge University in
will then rise and becom e intol erable to life forms. The England (Hewish, 1975). The Crab N ebula contains such
expansion of the Sun m ay engulf the terrestrial planets, a pulsar, which is the rem nant of a supernova observed
including Earth, and vaporize them . W hen all of its by Chinese astronom ers in 1054 A .D . (Fowler, 1967). The
nuclear fuel has been consum ed, the Sun will assum e the cores of the most massive stars collapse to form black
end stage of stellar evolution th at is appropriate for a star holes in accordance with Einsteins theory of general
of its mass and chemical com position. relativity. Black holes have radii of only a few kilom eters
and densities in excess of 10 1 6 g/cm 3. Their gravitational
field is so great that neither light nor m atter can escape
from them, hence the nam e black hole. O bservational
Toward the end of the giant stage, stars becom e evidence supporting the existence of black holes is
increasingly unstable. W hen the fuel for a particular growing and they are believed to be an im portant phenom
energy-producing reaction is exhaust ed, the star contracts enon in the evolution of galaxies.
and its internal tem perature rises. The increase in tem
perature may trigger a new set of nuclear reactions. In
stars of sufficient mass, this activity culminates in a Stars, it seems, have predictable evolutionary life
gigantic explo sion (supernova) as a result of which a cycles. They form, shine brightly for a while, and then
large frac tion of the outer envelope of the star is blown die. H ans Bethe (1968, p. 547) put it this way:
away. The debris from such explosions mixes with hydro
gen and helium in interstellar space to form clouds of gas If all this is true, stars have a life cycle much like
and dust from which new stars may form. animals. They are born, they grow, they go through
a definite internal development, and finally die, to
give back the material of which they are made so
A s stars reach the end of their evolution they
that new stars may live.
turn into white dwarfs, or neutron stars (pulsars),
or black holes, depending on their masses
(W heeler, 1973). Stars whose mass is less than 2.3 Nucleosynthesis
about 1 . 2 solar masses contract until their radius is
only about 1 X 104 km and their density is The origin of the chem ical elem ents is intim ately linked
betw een 104 and 108 g/cm 3. Stars in this configu to the evolution of stars because the ele m ents are
ration have low luminosities but high surface tem synthesized by the nuclear reactions from which stars
derive the energy they radiate into space. O nly helium and
peratures and are therefore called white dwarfs (Figure
deuterium , the
2.1). They gradually cool and fade from
14 IN THE BEGINNING
heavy isotope of hydrogen, w ere synthesized during the Science (an unlikely place for a nuclear astro physics
initial expansion of the universe. The entire theory of paper) on the buildup of heavy elements
nucleosynthesis was presented in a detailed paper by B by neutron capture and subsequently cham pi
urbidge, B urbidge, Fowler and H oyle (1957) referred to oned the idea th at the chemical elem ents were
affectionately as B2 FH . Subsequent contributions (e.g., synthesized during the first 30 min after the Big Bang. In
Schram m and A rn ett, 1973) have been concerned with the mid-1940s A lpher and Gamow wrote a paper
spe cific aspects of the theory and its application to the detailing the origin of the chemical ele
evolution of stars of different m asses and ini tial m ents based on that assumption. A t G am ow s
compositions. suggestion, H ans B ethes name was added in
absentia, thus creating the famous trium virate
The theory p resen ted by B 2F H (1957) evolved A lpher, Bethe, and Gam ow (1948).
from the work of several other scientists, am ong whom G The abundance of the chemical elem ents and
eorge G am ow deserves special recognition. G am ow their naturally occurring isotopes is the blueprint for all
received his doctorate from theories of nucleosynthesis. For this rea son, geochem
the U niversity of Leningrad and cam e to the ists and stellar spectroscopists have devoted m uch tim e
U niversity of G ttingen in th e spring of 1928 and effort to obtaining accu rate analytical data on the
for som e postdoctoral studies. T here he m et concentrations of the elem ents in the Sun and other
Fritz H outerm ans, with w hom he spent many nearby stars from the wavelength spectra of the light they
hours in a cafe making calculations (Gamow, emit. Inform ation on the abundances of nonvolatile elem
1963). Subsequently, H outerm ans m oved to the ents has com e also from chemical analyses of stony m
U niversity of Berlin w here he m et the British eteorites, especially the carbonaceous chondrites, which
astronom er R o b ert A tkinson. H outerm ans and are the most undifferentiated
A tkinson used G am ow s theory of alpha decay to
propose that stars generate energy in their interi ors by the sam ples of m atter in the solar system available
form ation of helium nuclei from four protons captured by to us (M ason, 1962). Table 2.1 lists the abun
the nucleus of another light elem ent. The initial title of dances of th e elem ents in the solar system com piled by
their paper was Wie kann m an einen H elium kern im A nders and E bihara (1982). The abun dances are
Potentialtopf kochen. (H ow one can cook a helium expressed in term s of the num ber of atom s relative to
nucleus in a pressure cooker.) Needless to say, the editors 106 atom s of silicon. Figure 2.2 is a plot of these data and
of the Zeitschrift f r P hysik insisted that the title be illustrates several im portant observations about the
changed (A tkinson and H outerm ans, 1929). abundances of the elem ents.
The terrestrial abundances of the radioactive daughters of uranium and thorium are very low and are therefore indicated as ~0 . In
addition, neptunium and plutonium are produced in nuclear reactors and therefore occur on the Earth. However, the transuranium
elements having atomic numbers of 95 or greater do not occur in the solar system. Their abundances are therefore stated as 0. S O U R C E :
Anders and Ebihara (1982).
16 IN THE BEGINNING
Figure 2.2 Abundances of the chemical elements in the solar system in terms of atoms per 106 atoms of Si. The data
were derived primarily by analysis of carbonaceous chondrite meteorites and by optical spectroscopy of light from the
Sun and nearby stars (Anders and Ebihara, 1982).
5. T he abundances of ters of long-
lithium , lived radioactive
beryllium, and isotopes of
boron are anom uranium and
alously low com thorium .
pared to o ther
The nucleosynthesis
elem ents of low
m odel of B 2F H (1957)
atom ic num ber.
includes eight different
6 . The abundance of
kinds of nuclear
iron is notably
reactions that occur at
higher than those specified tem peratures
of other elem ents in the course of the
with similar atom evolution of a star.
ic numbers. Several of these reac
tions m ay take place sim
7. Two elem ents, ultaneously in the cores
technetium and and outer shells of
prom ethi um , do massive stars. As a
not occur in the result, the core of a star
solar system m ay have a different
because all of their chemical com position
isotopes are than the shell
surrounding it. M
unstable and decay
oreover, not all of the
rapidly.
nuclear processes take
place in all stars.
8 . T he elem ents
Consequently, other stars
having atom ic
in the M ilky Way
num bers greater
Galaxy do n o t
than 83 (Bi) have
necessarily have the
no stable iso topes,
same chem i cal com
but occur naturally position as the Sun and
at very low her planets.
abundances
because they are
the daugh
2.3 NUCLEOSYNTHESIS17
A ll stars on the main sequence generate ener gy by ed, the interstellar gas clouds contained elem ents of
hydrogen fusion reactions. This process results in the higher atom ic num ber. The presence of carbon-12 (gC)
synthesis of helium either by the synthesized by the ancestral stars has m ade it easier for
direct p ro to n -p ro to n chain (equations 2.5-2.8) subsequent generations of stars to generate energy by
or by the CN O cycle (equations 2.9 -2.14). The hydrogen fusion. This alter native m ode of hydrogen
proton -proton chain works as follows. Two nuclei fusion was discovered by H ans B ethe and is know n as
of hydrogen, consisting of one proton each, collide to the CN O cycle:
form the nucleus of deuterium ((H ) plus a positron ( +) 12r + !H 13N + y (2.9)
and a neutrino (v). (The designation of atomic species is 0
'(N >13r + + + V (2 . 1 )
presented in Chapter 6 .) Each reaction of this kind
1
liberates 0.422 million electron volts (M eV) of energy. l3r + IH >(N + y (2 . 1 )
The positron (positively charged electron) is annihilated 2
'(N + 15o + y (2 . 1 )
by interacting with a negatively charged electron giving
S'* +
off additional energy of 1.02 MeV. The deuterium nucleus N
| o 13 ++ V (2.13)
col lides with another proton to form the nucleus of
13N + + (H e (2.14)
helium-3 QHe) plus a gamma ray (y) and 5.493 M eV of
!H >12r
energy. Finally, two helium-3 nuclei must collide to form
The end result is that four protons are fused to form one
helium-4 ((H e), two protons, and 12.859 MeV. The end
nucleus of (H e, as in the direct proton - proton chain. The
result is that four hydrogen nuclei fuse to form one
nucleus of l\C acts as a sort of catalyst and is released at
nucleus of helium-4, a gamma ray, a neutrino, and 19.794 the end. It can then be reused for another revolution of the
M eV of energy. CN O cycle.
The Sun contains elem ents of higher atomic num ber
The entire process can be described by a series of
than helium including *gC and therefore carries on
equations in which the nuclei of hydrogen and helium are
hydrogen fusion by th e C N O cycle. In fact, m ost stars
represented by the symbols of the appropriate isotopes
in our M ilky W ay G alaxy are second-generation stars
(see C hapter 6 ), even though these isotopes do not
because our G alaxy is so old that only the very sm allest
actually exist in atomic form in stellar interiors w here
first-generation stars could have survived to the present
their electrons are rem oved
time.
from them because of the high tem perature:
The low reaction cross section of the p ro to n -
|H + JH -> (H + + + v + 0.422 M eV (2.5)
proton chain by which the ancestral stars gener ated
+ + ~ 7 -> 1.02 M eV (annihilation) (2.6) energy has been a source of concern to nuclear
astrophysicists. W hen this difficulty was pointed out to
(H + }H -> (H e + y + 5.493 M eV (2.7) Sir A rthur E ddington, who pro posed hydrogen fusion in
(H e + (H e - (H e + }H + |H + 12.859 M eV stars in 1920, he replied (Fowler, 1967):
(2.8)
We do not argue with the critic who urges that the
The direct proton -proton fusion to form helium -4 stars are not hot enough for this process; we tell him
can only take place at a tem perature of about to go and find a hotter place.
1 0 X 106 K, and even then the probability of its
occurrence (or its reaction cross section) is very small. A fter the hydrogen in the core has been convert ed to
Nevertheless, this process was the only source of nuclear helium ash, hydrogen fusion ends, and the core
energy for first-generation stars th at form ed from the contracts under the influence of gravity. The core tem
prim ordial m ixture of hydrogen and helium after the Big perature rises tow ard 100 x 10 6 K, and the helium ash
Bang. becom es the fuel for the next set of energy-producing
Once the first generation of stars had run through nuclear reactions. The critical reaction for helium burning
their evolutionary cycles and had explod is the fusion of three
18 IN THE BEGINNING
alpha particles (triple-alpha process) to form a nucleus of are the neutron-capture reactions, which produce a large
l\C\ num ber of the atoms having atomic numbers greater than
26 (Fe). These reactions involve the addition of a neutron
jH e + gHe - gBe (2.15)
to the nucleus of an atom to produce an isotope having
IjBe + gHe - l2C + y (2.16)
the same atomic number but a large mass number. For
This is the critical link in the chain of nucleosyn thesis example, the equation:
because it bridges the gap in nuclear stabili ty of the ggNi + in -> Ni + y (2.21)
isotopes of lithium, beryllium, and boron. The crux of the neutron
problem is that the nucleus of gBe is very unstable and indicates that the nucleus of ggNi absorbs a neu tron,
decays quickly with a half-life of about 10 " 1 6 sec. which changes it to an excited state of ggNi, which then
Therefore, gBe m ust absorb a third helium nucleus very deexcites by em itting a gam m a ray. Nickel-63 is
soon after its form ation to make it safely to stable 'gC radioactive and decays to stable ggCu by em itting a ~
(Fowler, 1967). An alternative reaction involving the particle:
addition of a pro ton to the nucleus of gHe has an even
smaller chance to succeed because the product, gLi, has a ggNi -> ggCu + - + v + 0.0659 M eV (2.22)
half-life of only 1 (T 2 1 sec and decom poses to form helium w here v is an antineutrino. ggCu is a stable iso
and hydrogen: tope of copper and can absorb another neutron to form
IgCu:
jH e + }H - 53Li (2.17) ggCu + in -> IgCu + y (2.23)
gLi - gHe + |H (2.18)
Copper-64 is radioactive and undergoes branched decay
The triple-alpha process is indeed the key to to form g^Zn and ggNi, both of which are stable:
the synthesis of all elem ents beyond helium .
W ithout it, stellar evolution w ould be short- ggCu -> ggZn + ~ + v + 0.575 M eV (2.24)
circuited, and the universe would be com posed ggCu -> g N i + + + v + 1.678 MeV (2.25)
only of hydrogen and helium .
H elium burning sustains red giants only for a few This process of successive additions of neutrons is
tens of millions of years or less. W ith increas illustrated in Figure 2.3. It takes place during the red giant
ing tem perature in the core, alpha particles fuse stage of stellar evolution when the neu tron flux is low
with the nuclei of 'gC to produce nuclei of still enough to perm it the product nucleus to decay before the
higher atomic num ber: next neutron is added. This process therefore is
12C + gHe -> 'gO (2.19) characteristically slow and is therefore referred to as the s-
'gO + gHe -> 2"Ne (2.20) process.
By exam ining Figure 2.3 closely we see that the
etc.
track of the s-process bypasses Zn, which is
However, electrostatic repulsion betw een posi tively one of the stable isotopes of . zinc. In order to
charged nuclei and alpha particles limits the size of the m ake this isotope by neutron capture reactions,
atom s th at can form in this way. The heaviest atom the pace m ust be speeded up so that unstable
produced by the addition of alpha particles is ggNi, which goZn can pick up a neutron to form ggZn before it decays
decays to ggCo and then to stable ggFe. These nuclear
to stable 3 ?Ga:
reactions there fore cause the enhanced abundance of the
ele m ents in the iron group illustrated in Figure 2.2. goZn + In - ]0Z n + y (2.26)
A n even m ore rapid rate of neutron capture is
During the final stages of the evolution of red giants, required to m ake ggZn from ggCu by addition of five
several other kinds of nuclear reactions occur (B 2 FH , neutrons in succession to form 2 9 CU, which then decays
1957). The most im portant of these by ~ emission to stable ggZn:
2.3 NUCLEOSYNTHESIS 19
Figure 2.3 Nucleosynthesis in red giants by neutron capture on a slow time scale (s-process) followed by
beta decay. The dark squares are stable isotopes, whereas the unshaded ones are radioactive. The process
starts with stable (|Ni, which absorbs a neutron to form unstable Ni, which decays to stable ))Cu by
emitting a ~ particle. The main line of the s-process, as indicated by arrows, proceeds from N i to ((Se
and beyond but bypasses ([}Zn and ((Se. (jjZn is produced by neutron capture on a rapid time scale (r-
process) from fgZn via unstable (fjZn and from ))Cu, which captures five neutrons in rapid succession to
form (Cu followed by decay to stable (JjZn. ((Se is a proton-rich nuclide that cannot form by either the
s-process or the r-process and requires the addition of two protons (p-process) to stable ((Ge.
20 IN THE
BEGINNING
tem peratures are
65Cu + 5 in -> JgCu + IjGe + 2 }H
29 However, pulsars and sufficient to cause
70
lC u ~ + v + black holes have high hydrogen fusion. Stars
3oZn + This process also takes
C apture of neutrons at place at the very end of inter nal pressures and tem evolve through predictable
such rapid rates is charac the giant stage of stellar peratures that cause stages depending on their
teristic of the r-process, evolution. atomic nuclei to masses and initial
which requires a much The system of disintegrate into m ore compositions. They
greater neutron flux than nuclear reactions primitive con stituents. generate energy by nuclear
the s-process and th ere originally pro posed by B The relative proportions of reactions that synthesize
fore takes place only 2F H (1957) can account the chemical elements in other elem ents from prim
during the last few m for the observed other stars are different ordial hydro gen and
inutes in the life of a red abundances of the because local conditions m helium . In the end, stars
giant w hen it explodes as chemical elem ents in the ay affect the yields of the explode, and the rem nants
a supernova. solar system and in nearby many nuclear reactions becom e solid objects of
stars. Nucleo synthesis is that contribute to their great density.
How ever, no neutron taking place at the present synthesis. The abundances of
capture on any tim e scale time in the stars of our the chemical elem ents in
the solar system can be
can account for the form galaxy and in the stars of 2.4 Summary explained by the nuclear
ation of some atom s such other galaxies throughout
as stable ((Se shown also the universe. We have We live in an expanding reactions that energize the
in Figure 2.3. This nuclide good evidence in the universe whose future is stars. These reactions
is synthesized by the wavelength spectra of uncertain. The universe progress from fusion of
addition of two p ro tons light from distant galaxies started with a Big Bang hydrogen and helium to
to stable ((G e in the so- that the chemical elements about 15 X 109 years ago neutron capture and to
called p-process: we find on E arth also and has evolved since then other reactions, m ost of
occur everywhere else in in accordance with the which occur only for a
the universe. laws of physics. short time at the end of the
Stars are the basic active life of a star. The
units in the hierarchy of chemical elem ents we
heavenly bodies. They know on E arth occur
form by contraction of throughout the universe,
clouds of interstellar gas but their abundances vary
and dust until their core because local conditions
affect the yields of the
nucleosynthesis processes.
Problems*
1. How has the abundance of hydrogen (H) in the 7. Why is lead (Pb) more abundant than we might have
universe changed since the Big Bang? expected?
2. Why are elements with even atomic numbers more 8. Check the abundance of argon (Ar) and deter mine
abundant than their neighbors with odd atomic numbers? whether it is greater than expected. If so, suggest an
explanation.
3. How do we know that the Sun is at least a sec ond- 9. How did lithium (Li), beryllium (Be), and boron
generation star? (B) form?
4. Why do technetium (Tc) and promethium (Pm) lack 10. Compare the abundances of the rare earths
stable isotopes? (lanthanum to lutetium) to such well-known metals as Ta,
5. What other elements also lack stable isotopes? W, Pt, Au, Hg, and Pb. Are the rare earths really all
thatrare?
6. Why do the elements of Problem 5 exist on the
Earth, whereas Tc and Pm do not?
*Not all of these questions have unequivocal answers.
References
A L P H E R , R. A., H. B E T H E , and A T K IN S O N , R. dE., and F. G. H
G. G A M O W , 1948. The origin O U T E R M A N S , 1929. Zur
of
Frage der
the chemical elements. Phys. Rev.,
Aufbaumglichkeit der
A N D E R S , E ., and M. E B I H A R A , 1982. Solar-system
Elemente in Sternen. Zeit.
dances of the elements. Geochim. Cosmochim.
Phys., 54:656-665.
46:2363-2380.
B A A D E , W., 1968. Evolution o f
Stars and Galaxies. Harvard
University Press,
Cambridge, MA.
REFERENCES 21
B E T H E , H. A., 1939. Energy production in stars. Phys. Rev., H E W IS H , A., 1975. Pulsars and high density physics. Science,
55:434456. 188:1079-1083.
B ethe, H. A., 1968. Energy production in stars. Science, H U B B L E , E., 1936. The Realm o f the Nebulae. Yale University
161:541-547. Press, New Haven, CT. Reprinted 1958 by Dover Pub., Inc.,
B U R B R I D G E , E . M., G. R. B U R B R I D G E , W. A. F O W L E R , and F. New York.
H O Y L E , 1957. Synthesis of the elements in stars. Rev. Mod. I B E N , I., Jr., 1967. Stellar evolution: Comparison of theory with
Phys., 29:547-650. observation. Science, 155:785796.
F O W L E R , W. A., 1967. Nuclear astrophysics. Amer. Phil. Soc. I B E N , I., Jr., 1974. Post main sequence evolution of single stars.
Mem. 67,109 pp. Ann. Rev. Astron. and Astrophys., 12:215-256.
Gamow, G., 1935. Nuclear transformations and the origin of M A S O N , B., 1962. Meteorites. Wiley, New York, 274 pp.
the chemical elements. Ohio J. Sei., 35:406-413. P E N Z IA S , A. A., and R. W. W IL S O N , 1965. A measurement of
Gamow, G., 1952. The Creation o f the Universe. Viking Press, excess antenna temperature at 4,080 Mc/s. Astrophys. J.,
New York, 144 pp. 142:419.
G A M O W , G., 1963. My early memories of Fritz Houtermans. In S C H R A M M , D . N., and W. D. A R N E T T , 1973. Explosive
J. Geiss and E. D. Goldberg (Eds.), Earth Science and Nucleosynthesis. University of Texas Press, Austin and
Meteoritics. North-Holland Publ. Co., Amsterdam, 312 pp. London, 301 pp.
G O T T , J. R., Ill, 1982. Creation of open universes from de W E IN B E R G , S., 1977. The First Three Minutes. Bantam Books,
Sitter space. Nature, 295:304-307. New York, 177 pp.
H A IN E B A C H , K ., D . K A Z A N A S , and D . N. SC H R A M M , 1978. A W H E E L E R , J. C., 1973. After the supernova, what? Amer.
consistent age for the universe. Short papers of the Fourth Scientist, 61:42-51.
International Conference, Geochronology, Cosmochronology
and Isotope Geology. U.S. Geol. Sur. Open-File Rept., 78-
701,159-161.
3
The Solar System
The Sun form ed from a cloud of gas and dust par ticles, stellar gas and dust know n as the solar nebula. It had
as did all other stars in the Milky Way G alaxy and form ed about six billion years ago as a result
elsewhere in the universe. Sometimes the process leads to of the term inal explosions of ancestral stars,
the form ation of two com panion stars. However, in the which added the elem ents they had synthesized
case of the Sun a small fraction of the original cloud to the prim ordial hydrogen and helium that orig inated
accreted to form a set of nine planets, including the Earth. from the Big Bang. The chemical composi tion of the
C om pared to stars, these planets are insignificant solar nebula was given in Chapter 2 (Table 2.1). The dust
objects, but to us they are the very basis of exis tence. cloud was rotating in the same sense of the M ilky Way
The hum an race appeared on the Earth only two or three Galaxy and was acted upon by gravitational, magnetic,
million years ago and, after a slow start, learned to fly and elec trical forces.
and to explore the solar system within a span of a m ere
century. As soon as the m ain mass of the solar nebu la
began to contract, order began to be imposed on it by the
G eochem istry today not only encompasses physical and chemical conditions that existed during this
the study of the com position and chemical phase of star formation. These included the developm ent
processes occurring on the E arth, but is also con cerned of pressure and tem perature gradients and an increase in
with all of the planets and their satellites. Inform ation the rate of rotation. Certain kinds of solid particles that
for geochemical studies of the solar system is derived by had form ed in the nebula evaporated as the tem pera
analysis of m eteorites and rock samples from the M oon
and by remote sensing of planetary surfaces. The ture increased in order to m aintain equilibrium
exploration of the solar system has expanded our horizon betw een solids and gases. As a result, only the
and has provided the basis for com parative planetary m ost refractory particles (Fe-N i alloys, A12 0 3,
geochemistry. The satellites of the large gaseous planets CaO, etc.) survived in the hottest part of the con
are of special interest in this new field of study because tracting nebula, w hereas in the cooler outer regions a
som e of them are larger than our M oon and have very larger variety of com pounds remained in the solid state.
different chemical compo sitions and surface features The increase in the rate of rota tion caused p art of the
than the E arth . nebula outside of the pro-tosun to form a central disk.
The solid particles congregated in this disk and m ade it
We therefore need to becom e acquainted with the sufficiently opaque to absorb infrared radiation. The tem
new worlds we m ust explore before we concentrate our per ature in the central disk therefore increased until it
attention on the conventional geochem istry of the E arth. ranged from about 2000 K at the center to about 40 K at
approxim ately 7.5 X 109 km from the protosun. The
pressure ranged from less than 0 . 1 atm to about 1 0 - 7 atm
near the edge of the disk (Cam eron, 1978; Cam eron and
3.1 Origin of the Solar System
Pine, 1973).
The origin of the planets of the solar system is intim ately
linked to the form ation of the Sun. In the beginning there The developm ent of pressure and tem pera ture
was a diffuse mass of inter gradients within the disk caused the first
22
3.1 ORIGIN OF THE SOLAR SYSTEM 23
m ajor chemical differentiation of the solar nebu la. C om hydrogen fusion was less than 100,000 years. The Suns
pounds with low vapor pressures persisted throughout the initial luminosity was two to three times greater than is
nebula and form ed dust parti cles, w hereas com pounds consistent with the main sequence because the Sun
with high vapor pres sures could exist only in the cooler contained excess therm al energy g enerated during the
outer regions. The condensation tem peratures of various initial contraction . This superlum inous phase of the
com pounds that existed in the solar nebula are listed in Sun lasted about 1 0 million years and resulted in the
Table 3.1. The condensates accreted to form larger bodies expulsion of about 25% of its original mass in the form of
as a result of selective adhesion caused by electrostatic and solar w ind com posed of a proton/electron plas ma.
magnetic forces. The resulting solid bodies, called This is called the T-Tauri stage of stellar evolution after
planetesimals, had diam eters ranging from about 1 0 m to the star th at is the prototype for this process. All gaseous
m ore than 1 0 0 0 km, and their chemical compositions varied m atter in the vicinity of the Sun was blown away during
with distance from the center of the planetary disk.The this period, and only the solid planetesim als having diam
planetesim als close to the protosun were com posed of eters larger than about 1 0 m rem ained.
refractory com pounds dom inated by oxides and m etallic
iron and nickel; farther out were Mg and Fe silicates and
farthest out, ices com posed of water, ammonia, m ethane, The planetesim als in the inner region of the
and other planetary disk subsequently accreted to form the
so -called earthlike planets M ercury, Venus,
E arth, and M ars and the parent bodies of the
volatiles. m eteo rites now rep resen ted by asteroids.
The initial rate of evolution of the solar sys Instabilities in the outer p art of the disk resulted in the
tem was rem arkably fast. The tim e required for form ation of the gaseous outer planets: Jupiter, Saturn, U
the Sun to reach the ignition tem perature for
ranus, N eptune, and Pluto.
Table 3.1 Condensates from the Solar Nebula at Different Temperatures
Temperature, C Condensates
1325 refractory oxides: CaO, A120 3, T i02, REE oxides
1025 metallic Fe and Ni
925 enstatite (MgSiO,)
925-220 Fe forms FeO, which reacts with enstatite to form olivine [(Fe,Mg)2S i0 4] Na
725 reacts with A120 3 and silicates to form feldspar and related minerals.
Condensation of K and other alkali metals F12S
400 reacts with metallic Fe to form troilite (FeS)
280 H20 vapor reacts with Ca-bearing minerals to make tremolite
150 H20 vapor reacts with olivine to form serpentine
- 100 HzO vapor condenses to form ice
- 125 NH3 gas reacts with water ice to form solid NH3 H20
- 150 CH4 gas reacts with water ice to form solid CH4 7H20
1N>O Ar and excess CH4 condense to form Ar and CH4
-250" Ne, H, and He condense
These reactions probably did not occur because the temperature in the planetary disk did not decrease to such
low values.
SOURCE: Glass (1982), based on Lewis (1974).
24 THE SOLAR SYSTEM
The origin and chem ical com position of Pluto are The descriptive physical properties of the solar
still not well known because this planet is difficult to system listed in Table 3.2 indicate that 99.87% of the total
observe from E arth . Pluto also does not conform to the mass of the solar system
so-called Titius-B ode law, which appears to govern the (2.052 X 103 3 g) is concentrated in the Sun. The
distances of the planets from the Sun when these are rem aining 0.13% is distributed among the nine
expressed in astronomical units (A . U.), defined as the m ajor planets, among which Jupiter is by far the largest
average with about 71% of the planetary masses. The densities and
distance betw een the E a rth and Sun. The sizes of the planets, shown in Figure 3.1, vary widely and
T itius-B ode law was published in 1772 by J. E. imply the existence of large differences in their chemical
Bode, director of the A stronom ical O bservatory in Berlin, compositions. The inner planets (M ercury, Venus, E arth,
and is based on a series of num bers dis covered by J. D. and M ars), as well as the M oon and the asteroids, are
Titius of W ittenberg in 1766. The series is com posed of solid objects com posed prim arily of elem ents and com
the num bers 0.4, 0.7, 1.0, 1 .6 . 2 .8 . . . . , which are obtained pounds having low vapor pressures. The outer planets
by writing 0 ,3 , 6 , 1 2 .2 4 ...., adding 4 to each num ber, and (Jupiter, Saturn, Uranus, and N eptune) have low densities
dividing by 10. The resulting num bers m atch the dis and are essentially gaseous, although all of them probably
tances of the planets from the Sun rem arkably well up to have condensed cores. The inner planets resem ble the E
and including U ranus (M ehlin, 1968). However, the arth in chemical com position and are therefore referred to
radius of the orbit of N eptune is only 30.1 A .U ., w hereas as the terrestrial or earthlike planets. The outer planets
the T itius-B ode value is 38.8 and the discrepancy in the consist prim arily of hydrogen and helium
case of Pluto is even greater (Table 3.2). The T itius-B ode
law predicts
with small am ounts of the other elem ents and
a value of 77.2, but the actual orbital radius of resem ble the Sun in chemical com position.
P luto is only 39.4 A .U . The discrepancy may The earthlike planets (plus the M oon and
suggest th at P luto did n o t form in th e orbit it asteroids), taken together, m ake up only 0.0006% of the
now occupies. total mass of the solar system and only
Table 3 .2 Properties of the Sun and H er Planets
Mean solar distance
Equatorial Density Number o f
Object 106km A.U. diameter, km Mass, g g/cm3 satellites
Sun 1,391,400 1.987 x 1033 1.4
SOURCE: Carr et al. (1984), Glass (1982), and fact sheet, Jet Propulsion Laboratory.
3.2 ORIGIN O F THE EARTHLIKE PLANETS 25
0.44% of the planetary masses. A ll of these objects are so 3.2 Origin of the Earthlike
close to the Sun that the orbital radius of the outerm ost
Planets
asteroid is only about 7% of the total radius of the solar
system. Evidently, the terrestrial planets are not typical o f O ur current understanding of the origin of the solar
the solar system and owe their existence to the special system indicates that the earthlike planets w ere hot w hen
conditions in the planetary disk close to th e Sun. E arth is they form ed and that their inter nal geochem ical
the largest of the inner planets w ith about 50.3% of the differentiation may have begun with the sequential
mass, followed by Venus (40.9% ), M ars (5.4% ), and M accretion of planetesim als of differing com positions (M
ercury (2.8%). urray et. al., 1981). Planetesim als com posed of m etallic
iron and oxides accreted first to form a core that was sub
The earthlike planets, viewed in Figure 3.1 in th e sequently buried by the planetesim als com posed of
perspective of the entire solar system, are a physical and silicates. The earthlike planets w ere initially m olten
chemical anomaly. The E arth is unique am ong its because of the heat generated by the rapid capture of the
earthlike neighbors in having about 71% of its surface planetesim als and because of radioactive heating.
covered by liquid w ater w ithin which life developed
early in its history and evolved into the m ultitude of
species of the p lan t and anim al kingdoms. Insofar as we The last phase of form ation of E arth, Venus, and M
know, life form s do not exist at the present tim e any w ars involved the capture of planetesim als com posed of
here else in the solar system. volatile com pounds that had form ed in the outer reaches
of the planetary disk, perhaps
M ean Distance from th e C enter of th e Sun, 10s km
Figure 3.1 A: Variation of density of the planets with mean distance from the Sun. Note that the Earth has the
highest density among the earthlike planets, which, as a group, are more dense than the outer gaseous
planets. B: The planets of the solar system magnified 2000 times relative to the distance scale. The
earthlike planets are very small in relation to the Sun and the gaseous planets of the solar system.
26 THE SOLAR SYSTEM
beyond the orbit of Jupiter. These volatile-ric )
to form m ore than 4 X 109 years ago by conden sation of
planetesim als, also know n as com etesim ai-. w ater vapor in the atm osphere. The presence of a large
deposited solids com posed of water, ammonia, volume of w ater on the sur face of E arth perm itted
m ethane, and other volatiles on the surfaces of the geological processes to operate and created conditions
planets.The w ater and other volatiles deposited on conducive to the developm ent and evolution of life. N
the E arth prom ptly evaporated to form a dense either the E arth nor any of the earthlike planets ever had
atm osphere from which w ater ultim ately con atm ospheres com posed of the hydrogen and heli um of
densed as the surface of the E arth cooled. Mercury and the solar nebula because these gases were expelled from
the M oon do not have atm ospheres partly because they the inner region of the solar system during the T-Tauri
are too small to retain gaseous ele m ents of low atom ic stage of the Sun.
num ber and their compounds.
A ccording to this scenario, the earthlike planets
have been cooling since the time of their 3.3 Satellites of the Outer Planets
form ation. M ercury and the M oon have cooled
sufficiently to becom e geologically inactive in All of the outer planets have satellites, som e of which are
the sense th at their interiors no longer interact larger than the M oon and the planet Mercury. In addition,
w ith their surfaces. Venus and E arth, being the these satellites have a wide range of chem ical com
largest of the earthlike planets, have retained positions and have responded in very different ways to the
m ore of the initial heat than their neighbors in forces acting on them . The large satellites of Jupiter were
the solar system and are still active. M ars is inter m ediate seen by Galileo Galilei on January 7,1610, with his newly
in size and has had volcanic eruptions in the not-too- built telescope. Actually, they m ay
distance geologic past. However, the age of the last m
artian volcanism is not known. have been observed even earlier by Sim on
In spite of the similarity in size and overall com Marius. However, detailed images of their su r
position of Venus and E arth, their surfaces have evolved faces were obtained only recently during flybys of the A
very differently. Venus has a dense atm osphere com m erican Pioneer and Voyager space probes. A chronology
posed of C 0 2 that has caused its surface to becom e of these events in Table 3.3 reveals the sudden burst of
extrem ely hot and dry. The surface of E arth cooled activity in the explo ration of the solar system betw een
rapidly, allowing oceans 1973 and 1989.
Table 3.3 Chronology of Exploration of the Satellites of Jupiter and S a tu rn
Planet Encounter date Spacecraft Comments
Jupiter Dec. 3,1973 Pioneer 10 left solar system on June 14,1983
Data from Glass (1982), Greeley (1985), Hunt and Moore (1981), Hamblin and Christiansen (1990),
and fact sheet, Jet Propulsion Laboratory.
addition, less heat is of other gases including
generated by radioactivity ethane, acetylene, ethyl
because its rocky core is ene, and hydrogen
relatively small. cyanide. Like the Gallilean
The satellites of satellites of Jupiter, each
Saturn, U ranus, and N of the satellites of Saturn,
eptune are identified in U ranus, and N eptune is a
Table 3.4. A ll of them unique body in the solar
were seen during flybys system th at has recorded
of Voyager 2. We will m on its surface a history of
en tion only Titan, the events caused by internal
largest of the satellites of and external processes. Im
Saturn. It has a dense atm ages and rem ote-sensing
osphere com posed prin data of these satellites w
cipally of m ethane, ere received only recently
nitrogen, and sm aller am during
ounts
3.4 PICTURES OF O UR SOLAR SYSTEM 29
Figure 3.2 A: Variation of the density of the Galilean satellites of Jupiter with increasing distance from the planet.
The decrease in density is caused by increases in the proportion of water relative to silicate material. B: The Galilean
satellites magnified 50 times relative to the distance scale. Amalthea is much smaller than Io but appears to be a
silicate object.
the flyby of the Voyager The exploration of
space probes in 1981 the new worlds, which are
(Saturn), 1986 (U ranus), suddenly w ithin our
and 1989 (N eptune). reach, will becom e an im
por tant task of the com m
unity of E arth Scientists.
3.4 Pictures A lthough an
of Our understanding of these
new worlds m ust ultim
Solar ately be based on studies
System of the chem ical com
positions of m atter and of
The exploration of the reactions and processes
solar system relies prim that take place on them ,
ar ily on unm anned space images of the surfaces of
probes th at have the planets help to identify the
capa bility of taking problem s that need to be
pictures of planetary solved.
surfaces and of returning
the images to E arth . The For this reason, we
two Viking landers and now exam ine som e of
the Pathfinder mission on the images of our solar
Mars, as well as the two system to confront the
Voyager spacecraft and chal lenge that lies ahead .
the space craft G alileo in Excellent photographs of
the Jupiter system are landform s on planetary
spectacular examples of surfaces appear in text
this technique, w hereas books by H am blin and
the M oon has been Christiansen (1990) and
explored prim arily by A by G reeley (1985).
m erican astro nauts who
actually w alked on its
surface.
30 THE SOLAR SYSTEM
The Moon is a familiar image in the sky. Its surface is pockmarked with craters formed by impacts of meteoroids, which
continue to fall, though at a greatly reduced rate. The lunar landscape consists of dark plains, called mare (singular) and
maria or mares (plural), and bright mountainous highlands. The mare basins, formed by impacts of large objects during
the early history of the Moon, are filled with sheetlike basalt flows. The highlands are older than the mare basins and are
composed of anorthositic gabbro. The surface of the Moon is covered by a layer of regolith (colloquially called soil) that
consists of rock and mineral particles, beads of impact glass, and chunks of regolith breccia.
This view of the Moon was taken from space by the astronauts of Apollo 17 in December 1972. It shows the roughly
circular, black Mare Crisium in the left upper quadrant. Southwest of Mare Crisium is the Mare Fecunditatis, and
directly west of Crisium is the Mare Tranquilitatis. The Mare Serenitatis is located northwest of Tranquilitatis and
extends northwesterly beyond the horizon. (Photo by NASA.)
3.4 PICTURES OF OUR SOLAR SYSTEM 31
In the late evening of July 20,1969 (EDT on Earth), Neil A. Armstrong and Edwin E. Aldrin descended from their
spacecraft and set foot on the surface of the Moon. They had landed near the southwest margin of the Sea of Tranquility
(Mare Tranquilitatis). The picture shows Edwin Aldrin on his way down just before he stepped onto the lunar surface.
The dusty plain in the background contains scattered boulders ejected from craters exca vated by impact of meteoroids.
(Photo by NASA.)
32 THE SOLAR SYSTEM
The surface of the planet Mars comes closest among all the planets in the solar system to the landscapes of Earth. Mars
has an atmosphere composed of N2 and CO, with a small amount of water. It also has roughly circu lar plains called
planitia and highly cratered highlands resembling those of the Moon. Mars has been an active planet, as indicated by
large shield volcanoes and rift valleys. In some places on Mars the surface is dissected by valleys in dendritic patterns
similar to stream valleys on Earth. Therefore, there is reason to believe that liquid water has existed on the surface of
Mars and that ice, in the form of permafrost, may still occur on Mars at the present time.
The picture shows the summit of the volcano Olympus Mons protruding through clouds on a frosty morning on Mars
much like Mauna Loa on the island of Hawaii. The summit contains several overlapping calderas whose presence
suggests a long history of volcanic activity. The volcano is 550 km in diameter at its base, and it rises 25 km above the
surrounding plain far higher than any mountain on Earth. (Photo by NASA.)
3.4 PICTURES OF O U R SOLAR SYSTEM
The Valles Marineris on Mars, which are probably rift valleys, extend more than 2400 km in
an east-west direction near the martian equator. Some of the valleys are up to 200 km wide
and 7 km deep. The walls of the valleys have been extensively modified by slides and by
erosional channels. The rocks into which the valleys are cut are layered and may be sheetlike
flows of basalt. (Photo by NASA.)
34 THE SOLAR SYSTEM
The first Viking lander touched down on Mars in Chryse Planitia on July 20,1976. As shown, this area is a stony desert
containing small sand dunes and angular boulders. The view at the landing site of Viking 2 in the Utopia Planitia is quite
similar, indicating the importance of wind in shaping the surface of Mars. Some of the boulders are pitted or vesicular,
perhaps because they originated from underlying lava flows. (Photo by NASA.)
3.4 PICTURES OF OUR SOLAR SYSTEM 35
Jupiter is by far the largest and most massive of the planets in the solar system. It has a turbulent atmosphere that
contains several cyclonic storm centers, including the Great Red Spot. The atmosphere is composed pri marily of
hydrogen and helium with small amounts of other elements of higher atomic number. Jupiter is not massive enough to
initiate hydrogen fusion in its core. Nevertheless, it has several sets of satellites, and some of the satellites are
comparable in size to the Moon of the Earth. (Photo by NASA.)
36 THE SOLAR SYSTEM
^ 4 fit
<& if
Io is the innermost of the so-called Galilean satellites of Jupiter. It is probably composed of silicate rocks like the Moon
and Mars, but its many active volcanoes emit sulfur-rich lavas and gases. The surface of Io is not cratered because impact
craters are quickly buried by lava flows and volcanic ash. The heat that causes the volcanic activi ty is generated by tidal
forces caused by the gravitational fields of Jupiter and Europa, the neighbor of Io. (Photo by NASA.)
3.4 PICTURES OF O UR SOLAR SYSTEM 37
Europa is the second of the Galilean satellites of Jupiter. Unlike Io, Europa is encased in a layer of water ice beneath
which liquid water may be present. The icy surface is cut by sets of lines that may be fractures filled with ice-dikes
from below. Impact craters are present but are not common, presumably because of periodic renewal of the surface of
extrusion of water through the system of fractures. The interior of Europa consists of silicate rocks and a small dense
core. Ganymede and Callisto, the other two Galilean satellites, are also ice-covered, but each is distinctly different from
its neighbors. (Photo by NASA.)
38 THE SOLAR SYSTEM
Saturn is justly famous for its intricate rings, although all of the gaseous planets are now known to have rings. Saturn,
like Jupiter, has a retinue of satellites, the largest of which is Titan. This satellite is only slightly smaller than Ganymede
and has an atmosphere composed mainly of nitrogen with some methane and other hydrocar bon gases. Because of the
low surface temperature on Titan, most of the methane (or ethane, CfjHg) may form oceans of liquid hydrocarbons.
(Photo by NASA.)
3.4 PICTURES OF O UR SOLAR SYSTEM 39
Neptune, the last of the great gaseous planets, was photographed in August 1989 by Voyager 2. It is blue in color and has
a turbulent atmosphere like that of Jupiter in spite of its great distance from the Sun (more than 4 x IO9 km) and its
much smaller size compared to Jupiter (about 6%). Neptune has at least eight satellites, Triton being the most massive.
The surface temperature of Triton is only 37 K or -2 3 6 C. Nevertheless, cryovol-canic activity is occurring on Triton
involving the eruption of liquid N2 and perhaps methane. Consequently, the surface of Triton contains few impact
craters and is composed of landforms never seen before on any satellite. (Photo by NASA.)
40 THE SOLAR SYSTEM
Planet Earth is our home in the solar system. It is the only planet or satellite with liquid water on its surface and with an
atmosphere containing molecular oxygen. Earth is also the only place in the solar system that can sustain life as we know
it.
This view of the southern hemisphere showing Africa, the Indian Ocean with Madagascar, and the Atlantic Ocean
was taken from space by the astronauts of Apollo 17, December 1972. (Photo by NASA.)
3.4 PICTURES OF O UR SOLAR SYSTEM 41
Manicouagen-Mushalagen Lakes, north shore, St. Lawrence River, Quebec, Canada. These two lakes form a cir cular
structure because they are the remnants of a deeply eroded impact crater. This and hundreds of other such impact craters
testify to the fact that the Earth has been bombarded by meteoroids and comets just as have the Moon, Mars, and all
other bodies in the solar system. In fact, such impacts are now recognized as an important geological process that has
disrupted the surface environment by causing short-term catastrophic climate changes and biological extinctions. (Photo
by NASA.)
42 THE SOLAR SYSTEM
plate boundaries. E stim ates of the chem ical basaltic m agm as by partial m elting under the
com position of the m antle m ust m eet certain physical conditions likely to prevail in th at
geophysical req u irem en ts regarding density, region of the E arth .
seism ic velocities, and h eat production by the These requirem ents are generally satisfied by
decay of U, Th, and K. In addition, the m antle a rock calledp y r o lite invented by Ringw ood (1966)
m ust be capable of form ing th e m ajor types of by combining peridotite and basalt in the ratio of 3
S i0 2
Table 4 .2 E s tim a te s o f th e C h e m ic a l
C o m p o sitio n o f th e M a n tle o f th e E a r th
MgO
FeO
A I 2O 3
CaO
NazO
Cr20 3
MnO
P2Os
K 2O
T i02
NiO
Sum
Whole Earth
Core
Mantle
Crustb
Hydrosphere
Atmosphere
Radius.
bContinental crust.
SO U R C E : Parker (1967) and Taylor and McLennan (1985).
Sample 13
45.2
37.5
8.0
3.5
3.1
0.57
0.43
0.14
0.06
0.13
0.17
98.8
46 CHEMICAL DIFFERENTIATION OF THE EARTH
up the continents but also addition, transport at the
upper mantle could be, and occurs under the oceans. but their occurrence at th e base of the ice sheet and
in fact had been, recov The oceanic crust is only surface of the conti subsequent deposition by
5 - 8 km thick nents enhances their im portance.rnelt-w ater caused glacial
ered by dredging.
and differs from th e The chem ical
continental crust in chem i com position clay to be well mixed,
cal com position and origin . T he tal crust has been estim making it a representative
ated
large num bers of chem sam ple of the rocks in the
ical
4.2 The crust is com posed prim arily of igneous
path of the ice sheet.The
m etam o rp h ic rocks. bining
M etam o rp h ic chem ical analyses
Co concentrated in orogenic types w eighted in term s average chemical com
of
nti belts, but the distinc (3) analyzing sedim ent derived position of 77 sam ples of
tion betw een igneous nent, (4) com bining the com positions glacial clay from sou th
nen is not always obvious and m afic rocks in varying ern Norway (colum n 3,
tal igneous rocks in the (5) by m odeling. The Table 4.3) is, in fact, quite
results
continental crust m ethods are surprisingly sim ilar to the estim ates
Cru have of C larke and W
sim ilar as indicated in
st: been m etam orphosed . S edim Tableentary
4.3. ashington. We note,
canic rocks form at th e however, that the
Maj and th erefore tend to
C larke and W
concentrations of C aO
ashington (1924) estim
or igneous and m etam orphicated the chemical com and
and volcanic rocks m ay
Ele position of the crust not
into elongated m asses by only by averaging a large
me to n ic basins in w hich num ber of chemical
deposited . C larke and W ashington Na20 of the glacial clay are som ew h
nts m ated that the continental
analyses of igneous rocks
crust their values in colum ns 1 and 2
from all continents and
The crust of the E arth is 95% igneous and m etam o rp h ic because these elem ents w ere parti
from the ocean basins, but
im portant because it con 5% sed im en tary rocks. also by com bining the the clay by leaching. In
tains all of the natural 4% shale, 0.75% sandstone, and 0.25%
average chemical com addition, the glacial clays
resources th at sustain us. stone. Evidently, sedim entary
position of igneous rocks have a low er T i0 2
It only a m inor p o rtio n of with those of shale, content than the crustal
includes the atm osphere, h sandstone, and lim estone aver ages of Clarke and W
ydrosphere, and bio s ashington, which m ay be
in accordance with their
phere, as well as part of the a regional characteristic of
observed abundances. The
lithosphere. The term the rocks of Scandinavia.
results in columns 1 and 2
lithosphere is now defined
of Table 4.3 are very
as the rigid outer por tion
similar. An enorm ous am
of the E arth consisting of
ount of work was required
the crust and upper m antle
to collect and analyze the
in contrast to the
well over 5000 rock sam
underlying asthenos-phere,
ples on which th e estim
which deform s plastically
ate in colum n 1 is
in response to tectonic
based.Therefore, V. M. G
stress. The study of the
oldschm idt decided in the
chem ical com po sition of
early 1930s to analyze the
the continental crust, and
clay-size fraction of till in
of the different types of
southern Norway because,
rocks of which it is com
he argued, glacial sedim
posed, has been a
ent of Pleistocene age was
m ajo r objective of geochem istry
produced by m echanical
1969; Taylor and M cLennan, 1985).
ero sion of bedrock in
The crust is the o u ter shell of
Scandinavia with little
w hich lies above the m antle. It not only
chemical w eathering. In
4.2 THE CONTINENTAL CRUST: MAJOR ELEMENTS47
Table 4.3 E s tim a te s o f th e A v e ra g e C h em ical C o m p o sitio n o f th e C o n tin e n ta l C ru st in
W e ig h t P e rc e n ts
r 2b 3C 4d 5e 6< 7% <Sh
Clarke and Washington (1924) based on the average of 5159 analyses of igneous rocks from all continents and the oceans. bClarke
and Washington (1924) based on 95% igneous rocks, 4% shale, 0.75% sandstone, 0.25% limestone.
'Goldschmidt (1954) based on 77 analyses of glacial clay from southern Norway. dDaly
(1914) based on 1 :1 mixture of average granite and basalt.
'Poldervaart (1955), based on the average of young folded belts and continental shield region shown in Table 4.4. fRonov
and Yaroshevsky (1976) based on a detailed model like that of Poldervaart (1955).
eTaylor and McLennan (1985,Table 3.4, No. 10) based on 75% Archean crust and 25% andesite crust.
hSederholm (1925) based on the Precambrian rocks of Finland.
S O U R C E : Parker (1967);Taylor and McLennan (1985).
continental crust
Columns 4 -8 of crust as a whole.
Table 4.3 contain Poldervaarts estimates of
estimates of the chemical the chemical com position
composition of the of each of the four
continental crust derived structural units are shown
by various methods. Daly separately in Table 4.4.
(1914) estim ated its These results illustrate the
composition by combining differences in the
chemical composi
analyses of granite and tion
(continental
of the
th e L ith ic C ru st o f th e E a r th in W eight P e rc e n t
Poldervaart Ronov and Yaroshevsky
1 2 3 4 5
Continental Young Suboceanic Deep Oceanic
shieldsa folded beltsb regions: oceanicd crust
S i0 2 59.8 58.4 49.4 46.6 49.4
Sedimentary cover 0.5 km: 41% shale, 43% sandstone, 16% limestone. Igneous rocks vary in composition with depth
such that there are 22 km of granodiorite, 3 km of diorite, and 10.5 km of basalt.
bSedimentary cover 5 km: 52% shale, 13% sandstone, 22% limestone, 5% greywacke, 6% andesite, 2% rhyolite. Igneous
rocks: 22.5 km of granodiorite, 2.7 km of diorite, and 7.3 km of basalt. In addition, 40% of the sediment layer is assumed to be
granodiorite based on the outcrop area of granitic batholiths.
cShelf sediment 4 km, similar in composition to that of folded mountain belts. In addition: 74% terrigenous mud, 22%
coral mud, 4% volcanic mud. Igneous rocks: 4 km of diorite, 2 km of tholeiite, and 5 km of olvine basalt.
dSediment thickness 0.3 km: 72% calcareous sands and oozes, 19% red clay, 9% siliceous ooze. Igneous rocks: 5.57
km of olivine basalt.
S O U R C E : Poldervaart (1955);Ronov and Yaroshevsky (1976).
summarized by M oores rocks can be duplicated by
(1982), indicates that the combining chemical
oceanic crust is actually analyses of mafic and
com posed of mafic and felsic igneous rocks in the
ultramafic rocks, whose proportions 1:5. The
average composition may resulting chemical com
approach that of basalt, posi tion resembles that of
although trace elem ent diorite or andesite, which
con centrations have been is
altered by interaction with why this crustal m odel is generally referred
heated seawater. Ronov the andesite m odel. Subsequently,
and Yaroshevsky (1976) M cLennan (1985) proposed that the
also m odeled the crust consists of 75 %
continental and oceanic average A rchean rocks
crust and arrived at and 25% andesite crust.
compositions that are The resulting com position
similar to those of is shown in column 7 of
Poldervaart (1955). Their Table 4.3. It differs from
estim ate of the chemical all other estim ates listed
composition of the in Table 4.3 by having som
continental crust is listed e
in col um n 6 of Table 4.3, what lower concentrations of S i0
w hereas column 5 of K20 and higher concentrations
Table 4.4 contains their FeO, MgO, and CaO.
estim ate for the oceanic Accordingly, the chemical
crust. composition of the
continental crust proposed
Taylor (1964) revived by Taylor and M cLennan
D alys m ethod of esti (1985) is som ew hat m
mating the chemical com ore
position of the continental
crust by dem onstrating
that the average concentra
tions of rare earth elem
ents (R E E s) in sedim
entary
4.3 DIFFERENTIATION OF IGNEOUS AND SEDIMENTARY ROCKS 49
45.5 high-Ca granites
mafic than those of their
o on the extent of melting, represent igneous rocks of
Si 26.8
predecessors. However, it which is controlled by the interm ediate com
Al 8.40
does m eet the following pressure and tem perature position, which may not
Fe 7.06 at the source. The com actu ally be granites in the
im portant constraints.
Ca 5.3 position of magmas may strict sense of th at term .
1. It satisfies the All of the rem aining elembe
entsmodified subsequently The sedim entary
observed rate of by assimilation of country rocks are just as highly
heat flow based rocks or by mixing with diversified in th eir chem
very small fraction of the
on the decay of crust of the Earth. magmas derived from ical com positions as
U ,Th, and K. different sources. The igneous rocks. H ow ever,
2. It is capable of crys tallization of magmas the processes th at cause
generating the m depends on their chemical this diversification operate
ore highly
4.3 Differ compositions and on the on the surface of the E arth
differentiated entiat physical conditions in the rath er than at depth and
include weather
granodiorites of ion of magm a chamber. The
ing, transport,
the upper crust by rocks produced by
partial melting.
Igneo fractional crystallization of
deposition, and
3. It restricts crustal us m agm a may vary widely lithification. The
in their m ineral com
grow th by and position from ultram afic
study of these processes encom pas
andesitic island portion of the subject of geochem is
-arc volcanism to
Sedim cum ulates of early-form occupy the rem ainder of this book.
ed ferrom agnesian m
post-A rchean entar Sedim entary rocks can be divided
inerals (olivine or groups depending on w
time. y pyroxene) to quartz-rich hether they are com posed
4. It recognizes th at
75% of th e Rocks late-stage differentiates. of m ineral and rock
continental particles of preexisting
The chemical The most com rocks or w hether they
crust is com compositions of igneous prehensive com pilation of precipitated from aqueous
posed of rocks rocks vary widely because the chem ical com solu
th at are m ore of geochemical positions of the highly tion. This division is n o t com pletely
than 2500 differentiation that takes diverse igneous rocks can because clastic sedim entary rocks ma
million years place during their form be found in the H a n chemically precipitated m
ation. The differentia tion dbook o f G eochemistry,
old, which did ineral cements, w hereas
starts during m agm a edited by K. H . W
not the chemical precipitates m
form ation by partial m edepohl from 1969
necessarily form by elt ing of rocks in the ay contain a com po nent
to 1978. The com pilation of data
magm atic processes of detrital sedim ent. M
lower crust or upper based on earlier com pilations by
tant in post-A rchean oreover, the chemical com
mantle. The chemical com W edepohl (1961) andV
positions of rocks in both
In conclusion, we position of the resulting inogradov (1962).The data
are grouped into ultram groups are modified
note that the chemical magm a depends on the
com position of the composition of the source afic rocks, basalt, high-Ca
continental crust of the E rocks and granite, and low -Ca
arth is very different from granite and therefore rep
th at of the solar nebula re sent m ost of the com
and positional spectrum of
stony m eteorites. The nine m ost abundant igneous rocks. The
m ents and their average concentrations category basalt includes
continental crust are: both volcanic and plutonic
Wt.% rocks of basaltic com
position, w here as the
50 CHEM ICAL DIFFERENTIATION OF THE EARTH
by ion exchange processes associated with electri cal the E arth so rt the elem ents on the basis of their chem
surface charges of m ineral grains regardless of w hether ical properties. We can think of the surface environm ent
they are detrital or form ed as chemical precipitates. of the E arth as a giant m a chine that processes igneous
and m etam orphic rocks into a wide variety of sedim
The com pilation of data in Table 4.5 includes two entary rocks. The products of this m achine are highly
rock types th at are predom inantly detrital (shale and diversified because the chem ical elem ents, and the m
sandstone), one that is predom iriantly precipitated inerals they form , respond differently to the treatm ent
(carbonate rocks), and deep-sea clay, which is a m ixture they experi ence as they pass through the machine.
of detrital grains and chemical precipitates. The selection
of these rock types has an additional virtue because it The representation of geological processes
includes rocks that form in shallow m arine basins in the operating on the surface of the E arth as a
vicinity of continents as well as sedim ent deposited in m achine is helpful because it em phasizes the
deep ocean basins far from land. coherence of these processes and because it pro
m otes a global view of the geochemistry of the
The ap p aren t geochem ical differentiation indicated E a rth s surface. We return to this them e in P art V of this
by these data reflects the way in which geological book, w here we consider the geochemical cycles of
processes operating on the surface of selected elem ents (C hapter 22).
52 CHEMICAL DIFFERENTIATION OF THE EARTH
w ater (0 .38% ).T hat oxygen bonds in the
leaves a very small minerals.
4.4 D fraction of the w ater in O nce an elem ent has
Table 4 .6 In v e n to ry o f W a te r in th
H y d ro s p h e re
if the hydrosphere in entered the hydros phere, it
Volume,
shallow ground is subjected to other kinds
f w ater (0 .30% ), lakes (0 of.01% ),
processes th at m ay Reservoir 106 few3
e (0 .0 0 0 1 % ) w here it is accessible to concentration .
affect its Oceans 1370
m unicipal and industrial use (see alsoexample,
C hapter som e ions
r For Ice sheets and glaciers 29
W hen igneous, m are selectively adsorbed on
e etam orphic, or sedim the charged surfaces of
Deep groundwater 5.3
(750-4000 m)
n entary rocks come in clay m inerals or particles
Shallow groundwater 4.2
contact w ith w ater at or of oxides and hydroxides
ti near the sur face of the E of Fe, M n, and A l. O th er (<750 m)
a arth, different kinds of elem ents m ay enter th e Lakes 0.125
chemical reac tions occur, biosphere as nutrients and Soil moisture 0.065
ti which collectively Atmosphere3 0.013
becom e associated with
o constitute chem ical w organism s and biogenic Rivers 0.001 7
eathering or w ater-rock
n interaction. The outcom e
carbon com pounds. The Biosphere 0.000 6
noble gases are released Total 1408.7
o of these reactions depends into the atm osphere, w Liquid equivalent of water vapor.
not only on the chemi cal
f hereas oxygen, carbon
Berner and Berner (1987).
and physical conditions at dioxide, and
SOURCE:
Constituent % ppm
Alkaline earths 151 (27.5-315)
N2 78.084 (except Be and Ba)
o2 20.946 Halogens 1790 (8-3350)
0.033
co2
Ar 0.934 In addition, B and S, both of which are m ajor con
Ne 18.18 servative elem ents, have large enrichm ent factors
He 5.24 indicating that they too tend to accum ulate in the
Kr 1.14 oceans. They are follow ed by P, V, Ni, Ge, Se, Mo,
Xe 0.087 Sb, W, Au, and U, w hose enrichm ent factors are all
H2 0.5 < 20 but > 1 . Only M o and U are conservative
CH4 am ong the elem ents in this group.
2
The enrichm ent factors of the rem aining non
N 2O 0.5
conservative elem ents are all < 1 , an indication that
Data from the CRC Handbook o f Chemistry and they are rapidly rem oved from seaw ater by geo
Physics (Weast et al., 1986). chemical processes.The concentrations of these ele
m ents in seaw ater are controlled prim arily by ion
of the nonconservative elem ents vary with depth exchange reactions with particulate m atter, by ionic
as well as regionally w ithin the oceans, and their substitution in calcium carbonate, calcium phos
concentration ratios w ith chloride are n o t con phate, and am orphous silica, which are in large part
stant. The nonconservative elem ents in seaw ater biogenic in origin, by direct precipitation caused by
include m ost of the trace elem ents along with dis evaporation of seawater, and by exchange reactions
solved nitrate, bicarbonate, silicic acid, phosphate, w ith volcanic rocks along spreading ridges (H art
and dissolved oxygen. T he concentrations of m any and Staudigel, 1982). The concentrations of most
nonconservative constituents vary w ith depth elem ents in seaw ater are n o t controlled by the solu
because they are involved in the biological activity bilities of their compounds.
of the oceans. A s a result, they have low concen In contrast to the conservative elem ents (and
trations in the surface layer of the oceans where a few nonconservative ones), the m ajority of the
m ost of th e biological activity is concentrated. nonconservative elem ents do n o t accum ulate in
W hen the organism s die, their bodies sink and the ocean but are rapidly rem oved from it by one
decom pose, thereby returning nutrients to the or several of the geochemical processes m en
water. Consequently, deep ocean w ater is enriched tioned earlier. The adsorption of ions on the
in nutrients com pared to surface water. charged surfaces of both inorganic and organic
The conservative constituents have large particles plays a very im portant role in this regard
enrichm ent factors, indicating that they are unreac and was described by B roecker and Peng (1982)
tive and therefore becom e concentrated in seawa as the great particulate sweep. O ther im portant
ter com pared to river water. The average enrich books on the geochem istry of seaw ater are by
m ent factor of the m ajor (conservative) con B ern er and B ern er (1987), H olland (1978),
stituents of seaw ater is 775. These elem ents are G oldberg (1972), and Riley and C hester (1971).
generally joined by their congeners, m any of which The wide variation of the enrichm ent factors
are trace elements. For example, the alkali metals, we have been discussing im plies th at the unreac
alkaline earths (except Be and Ba), and halogens tive conservative elem ents rem ain in the oceanic
all have higher concentrations in seaw ater than in reservoir m uch longer than the reactive noncon
56 CHEM ICAL DIFFERENTIATION OF THE EARTH
dence tim es (M O R T ) Evidently, the oceans
servative elem ents. This listed in Table 4.7 w ere geology because it play a very im portant role
cal culated by this m ethod contradicted Ford K in the geochem ical
phenom enon is reflected
based on sedim entation differentiation of the crust
by the residence time elvins calculations based
rates provided by of the E arth by storing
defined as: Fisitsyn et al. on th e cooling of the
som e elem ents and letting
(1982). The Earth, which indicated that others pass rapidly to the
M O R Ts calculated from its age was less than 40 m sedim ent accu
the output to deep-sea il lion years. m ulating at the bottom of the ocea
w here t is the residence sedim ent differ from m ents that are rem oved from the oc
tim e in years, A x is the those obtained from river- The M O RTs of the form of sedim ent may
total am ount of an elem input data, but the two sets elem ents in Table 4.7 vary ultim ately reenter the
ent (x ) in solution in the of results are well over nine orders of m rock cycle and becom e
oceans, and d X /d t is the agnitude from 0.69 years
correlated . the input for another pass
average annua] input of for Fe to 7.9 X 10s years
through the geochem ical
th a t elem ent into the oceans. TheActually,
am ount the residence for Br. As expected, the
an elem ent in the oceans is the
assum ed to was
be con machine.
oceans first used con servative elem ents
stant, which is probably the E arth . The idea was have long residence times,
valid for the open ocean 1715 by E dm und Hailey, w hereas m any 4.5 Summary
but may not apply to the but nothing was done nonconservative elem ents The E arth differen tiated
continental shelves and about it for nearly 200 have sig nificantly shorter very early in its history
estuaries. The annual years. Finally, in 1899 residence times. The into a core, m antle, and
input into the oceans can John Joly took up H average residence tim es in crust w ith characteristic
be estim ated from the discharge the oceans of the
aileys suggestion and chem ical com positions. T
com position of the m ajor rivers
calcu lated the residence conservative elem ents are he core consists of m
Such estim ates are uncertain, how ever, because
tim e of N a in the oceans. etallic Fe, Ni, and re la te
both discharge and elem ental concentrations of
His result was 90 m illion d elem ents. T he m an tle
stream s vary seasonally
years, which is less than is com posed prim arily of
and are not know n for all Alkali metals
half the 200 million years silicates and oxides of M g
rivers with equal certainty
calculated by Taylor and Alkaline earths (except Be and Ba)
and Fe. The chem ical com
(see also C hapter 20). In
M cFennan (1985). N Halogens position of the continental
addition, the ru n o u t of
evertheless, Jolys result crust has b e e n estim ated
groundw ater into the
was an im portant m in different ways with
oceans is disregarded The other two m ajor elem
ilestone in the history of generally co n co rd an t
because of lack of inform ents (B and S) have results. T he nine
a tion. Furtherm ore, the
residence tim es of 16 X
discharge of industrial and
106 and 500 X 106 years,
m unicipal waste w ater
has significantly altered respectively.
the chem ical com
The lithophile elem
position of m any rivers.
ents have widely different
For this reason Taylor and
enrichm ent factors and
M cL ennan (1985) estim
residence times, w hereas
ated the average annual
a m ajority of the
input to the oceans from
chalcophile and
the rate of deposition and
the chem ical com siderophile ele m ents are
position depleted in seaw ater and
have short residence
of deep -sea sedim ent as originally times.
Fi (1982). Accordingly, the m ean
REFERENCES 57
igneous rocks. They form by chemical w eather
m ost ab u n d an t chem ical elem ents in th e co n ti nental
crust of the E arth are O, Si, A l, Fe, Ca, Mg, Na, K, and Ti. ing of igneous, m etam orphic, and sedim entary
A ll o th er elem ents to g eth er m ake up only a very sm rocks and by the transport, deposition, and lithi-fication of
all fraction o f the co n ti nental crust. the resulting products. M any elem ents
becom e concentrated in deep-sea clay and may
The distribution of trace elem ents am ong dif ferent reenter the rock cycle w hen deep-sea sedim ent
kinds of igneous rocks depends on their ability to replace melts in subduction zones to form magma.
m ajor elem ents in the ionic crys tals that form during The oceans play an im p o rtan t role in the
crystallization of m agm a. A s a result, different kinds of chem ical differentiation of the crust because
igneous rocks are enriched in specific trace elem ents, som som e elem ents accum ulate in seaw ater, w h ere
e of which are im portant industrial metals. C ertain kinds as others pass through it rapidly. The geological and
of igneous rocks may therefore host ore deposits of these geochem ical processes associated w ith chem ical w
metals. eathering of m inerals, tran sp o rt of w eathering products,
and their ultim ate deposi
The chem ical com positions of sedim entary rocks tion in th e oceans are the subjects of P art III
are just as diversified as those of the of this book .
Problems
1. Recalculate the chemical composition of pyrolite 4. Compare the results of Problem 3 to the esti mates
(Table 4.2, column 1) in terms of the weight percent of the average chemical composition of the con tinental
concentrations of the elements and rank them in terms of crust (Table 4.3). Take note of discrepancies and
their abundance. similarities and relate them to the geochemical properties
of the individual elements.
2. Recalculate the concentrations of Si, Al, Fe, Ca, Mg,
Na, K, Ti, Mn, and P of shale, sandstone, and carbonate 5. Assess the validity of the hypothesis that the sedi
rocks (Table 4.5) in terms of the appro priate oxides. mentary rocks (shale, sandstone, and limestone) collec
tively represent the weathering crust on the continents.
What additional sources or sinks are needed to ade quately
3. Combine the concentrations calculated in Problem 2 describe the redistribution of elements from the
in proportions of 4 parts shale, 0.75 parts sandstone, and continental crust into different kinds of marine sediment?
0.25 parts limestone.
References
G O L D S C H M ID T , V. M., 1954.
B E R N E R , E. K., and R. A. B E R N E R ,
Cycle. Prentice-Hall, Upper Saddle Geochemistry.
River, NJ, 397 Oxford
pp.
University
B R O E C K E R , W. S., and T.-H. Peng, 1982.
Eldigio Press, Palisades, NY, 690 pp.Press, 730 pp.
H A R T , S. R., and H. ST A U D IG E L ,
B U L L E N , K. E., 1963. An Introduction
1982. The control of
Seismology, 3rd ed. Cambridge
alkalies and uranium in
Cambridge, England, 381
pp. seawater by ocean crust
C L A R K E , F. W ., a n d H. S. W A S H alteration. Earth Planet.
IN G T O N , 1924. The Composition Sei. Lett., 58:202-212.
o f the Earths Crust. U.S. H E N D E R S O N , P., 1982.
Geol. Surv. Prof. Paper Inorganic
127,117 pp. Geochemistry.
D A L Y , R. A ., 1914. Igneous Pergamon
Rocks and Their Origin. Press, Oxford, England,
McGraw- 353 pp.
Hill, New York, 563 pp. H E R R M A N N , A. G., 1970.
G O L D B E R G , E. D. (Ed.), 1972. Yttrium and the lanthanides.
The Sea. Wiley, New York, 5 In K. H.
vols. Wedepohl (Ed.), Handbook
o f Geochemistry, vol. 2,
part 5, ch. 39, sec. E, 57-71.
Springer-Verlag, Berlin.
H O L L A N D , H. D., 1978. The
Chemistry o f the Atmosphere
and
Oceans. Wiley, New York,
351 pp.
58 CHEM ICAL DIFFERENTIATION OF THE EARTH
S EDERHOLM
of the Earths crust.
Geochem. Int.,
H U T C H IS O N , R., 1974. The , J. J., 1925. The
13(6):89-121.
formation of the Earth.
Nature, average
250:556-568. composition of
K R A U S K O P F , K . B., 1979. the
Introduction to Geochemistry, earths crust in Finland.
2nd ed. Comm. geol.
McGraw-Hill, New York, Finlande Bull.,
617 pp. 12(70), 20 pp.
Li, Y.-H., 1982. A brief T A Y L O R , S. R., 1964. The
discussion on the mean abundance of chemical
oceanic residence time of elements in the continental
elements. Geochim. crust a new table.
Cosmochim. Acta, 46:2671- Geochim. Cosmochim.
2675. Acta, 28:1273-1285.
L IS IT S Y N , A. P., et al., 1982. T A Y L O R , S. R., and S. M. M C L
The relation between E N N A N , 1985. The Continental
A n im portant task of geochem ists is to understand the chem ical com positions of m inerals and
rocks based on the physical and chem ical properties of the elem ents and their atoms. The regular
pattern of variation of the electronic structures of atom s d eter m ines the chem ical properties of the
elem ents and hence their positions in the period ic table. Consequently, our understanding of the
distribution of chem ical elem ents am ong the m inerals and rocks of the E arth rests securely on the
principles of physics and chem istry em bodied in the periodic table.
5
The Electronic Structure
of Atoms
We now end our tour of th e stars and the solar that the plum -pudding m odel of the atom could not be
system and turn to som e basic principles of correct because it did not provide for the quantization of
chem istry th at will help us to understand the electrom agnetic radiation that had been discovered by M
chemical processes taking place on the earthlike planets. ax Planck (1858-1947) in G erm any (Planck, 1900). H
The intellectual excitem ent generated by the exploration ow ever, Thomson
of the solar system cannot hide the fact that only on the E was in no m ood to argue with th e 26-year-old
arth can we live w ith out elaborate protection against the D ane who had little know ledge of the English
harsh envi ronm ents of the other planets. This lesson of language at the time. Sir John, w ho was the direc
planetary exploration should increase our in ter est in our tor of the Cavendish Laboratory, had won the
own planet and in the natural process es th at take place N obel Prize in 1906, and had been knighted in
on its surface and in its interior. We need to understand 1908, was not interested in B ohrs ideas about
these processes well in order to assure adequate living how to im prove his electron theory. Besides, he was then
conditions for working on his positive-ray apparatus, which would
soon lead to the discovery of iso topes and the developm
future generations. For these reasons w e now
ent of mass spectrom e ters (H olton, 1986).
m ove to a description of atom s whose orbiting
electrons resem ble the orbiting planets and their satellites B ohr therefore decided to leave Cambridge, and
of the solar system . went to M anchester where E rnest R utherford (1871-
1937) was the Langw orthy Professor of Physics. Perhaps
as a test of T hom sons m odel of the atom , R utherford
5.1 The Atom of Thomson and had fired a beam of alpha particles at a piece of gold foil
Rutherford and had observed to his great surprise th at the beam did
not simply pass through the foil as expected. Instead, a
Throughout the 19th century physicists accepted Sir Isaac sig nificant num ber of alpha particles w ere scattered and
N ew tons opinion th at atom s are solid, som e even seem ed to bounce straight back as though
massy, hard, im penetrable, m oveable particles they had struck a heavy object. H e said later It was
(Jastrow, 1967, p. 9). H ow ever, in 1897 J. J. quite the m ost incredible event that has ever happened to
Thom son (1856-1940) discovered the electron m e in m y life. It was alm ost as incredible as if you fired
and used it to develop a new m odel of the atom in which a 15-inch shell at a piece of tissue paper and it cam e back
negatively charged electrons w ere thought to be em and hit you. (Jastrow, 1967, p. 9). A fter thinking about
bedded in a positively charged m atrix like raisins in a it, R utherford decided that the m assive objects had to be
plum pudding. In 1911 Niels B ohr (1885-1962) cam e to very small because they w ere struck only rarely by alpha
the C avendish Laboratory at Cam bridge as a particles and concluded that most
postdoctoral fellow to work with Thom son on his
electron theory. H e argued
60
5.2 BOHR'S THEORY OF THE HYDROGEN ATOM 61
(H olton, 1986, p. 240).
of the m ass of the atom
y
Bohr did object and came velocity v. The condition
was concentrated in this back to M anchester from for stability of such an
small body, which he nam Copenhagen to discuss his o atom is that the force of
ed the nucleus. paper with R utherford electrostatic attraction m
R u th erfo rd s m point by point. In the end, f
ust be equal to the
odel of th e atom was R utherford agreed to t centrifugal force:
inherent ly unstable submit the entire
because in classical manuscript, but grumbled
h mv
(5.1)
physics an electron that B ohr had been so e
orbiting a positively obstinate. The energy of such an
charged nucleus loses
energy by em itting an H atom is the sum of the
kinetic energy and th e
electrom agnetic wave. As W hen this paper and y potential energy.
a result, the electron m ust a second one w ere pub
speed up and the radius of lished (B ohr, 1913a, b), d
1 2
its orbit m ust decrease they were not well r m v
until the electron crashes received. The prom inent
into the nucleus. G erm an physicist O tto S o The potential energy term
Nevertheless, B ohr liked tem m ay have spoken for g has a negative sign
R u th erfo rd s m odel m any of his colleagues w because the electron is
because he had a hunch th hen he reportedly said, If
e attracted tow ard the p ro
at he could im prove it by this nonsense is cor rect, I n ton. From equation 5.1 we
extending the concept of will give up being a obtain:
quantiza tion of electrom physicist. (H olton, 1986,
agnetic energy to the m p. 240). It was the end of A m
rv
echanical energy of an an era. Physics was about t
electron orbiting the to be shaken to its roots by Substituting into equation
nucleus. quantum mechanics and o 5.2 yields:
the uncertainty principle. m \( e
We all know that B (5.4)
ohr succeeded brilliantly, The instability of an atom E=t-r)~
received the N obel Prize com posed of one elec tron 7
in physics in 1922, and 5.2 B orbiting the nucleus can be
becam e one of the leaders dem onstrated by m eans =
in science in the 2 0 th cen
o of N ew tonian physics.
h
21
tury. However, he had a We assume that the
hard tim e convincing even electron orbits the proton
r r
R utherford that his model at a distance r and a
of the atom was right. W ' We see from equation 5.4
that the energy of the atom
hen Bohr, who had briefly s is negative and that it is
returned to D enm ark, sent
a m anuscript copy of his inversely related to the
radius of the orbit. In the
paper, R utherford agreed T limit, the energy approach
to subm it it for
publication but added, I h es minus infinity as the
radius goes to zero.
suppose you have no e
The instability arises
objection to my using my
judgm ent to cut out any m
o because the atom is an
atter I may consider un r energy well. Energy is
necessary in your paper? liberated as the radius of
the electron orbit
decreases. Since there are
no restrictions on the m
agnitude of the electronic
radius in this m odel, atom
s will tend to achieve a
condition of m inim um
energy and m axim um sta
bility by reducing the
radius to zero.
inside back
cover).Therefore,
according to Bohr:
m vr =
w here n = 1, 2, 3, . . . ,
and is know n as the first
quantum num ber. H ow
ever, the atom m ust still
62 THE ELECTRONIC STRUCTURE OF ATOMS
e an atom as a result of an
2
satisfy the condition square of the electronic
first quantum numtransition is:
ber. Evidently,
T he sm allest radius perm
expressed by equation 5 .1 . the quantum num ber plays a decisive role a it hc
itted by B o h rs theory is
Accordingly, the angular because it controls both the A=t
obtained from equation
m om entum of the elec radius of the elec tronic
5.10 by setting n = 1, in In other words, the energy
tron m ust also be: orbit and the energy of the
which case: of the radiation is
atom .
inversely proportional to
m vr = r = aQ = 0.529 X 10 In B ohrs m odel of
its wavelength.
the atom the electron can
w here a0 is the so-called W hen B ohr
and a stable atom m ust B o h r radius. occupy only certain orbits
developed this m odel of
satisfy both equations 5 . 5 whose radii depend on the
The energy of the the atom betw een 1912
and 5.6; thus: first quantum num ber. The
atom can now be and 1913, it was already
energy of the atom varies
e2 _ calculat ed by know n that hydrogen
v in increm ents depending
substituting equation atom s absorb and em it
on which orbit the electron
5.10 for the radius occupies. W hen the light at discrete
The velocity of the
wavelengths. These
electron is derivable from into equation 5.4 for the energy electron jum: ps from a
of th e atom
E = - f e 2 (Air2m e2\ higher orbit to a low er wavelengths had been ex
equation 5.5: pressed by the equation:
21,2 one, the energy of the atom
v= decreases. The energy
nh
m r see th at in B o h rs m odel liberated
2 v We the energyin this way is
Substituting into equation atom no longer depends on radiated away as an 4
(5.7) yields: the radius of the elec tronic electrom agnetic wave. B
w here al and a2 are
ohr used an equation
2m nre2 _ orbit b u t varies as the whole num bers, ax > a2,
derived by A lb e rt
nh reciprocal of the and R is the R ydberg
Einstein to represent this
constant, whose value was
which leads to: energy change.
know n to be 109,678.18
,21,2
nz If E 1 is the energy of cm -1. (N ote th at we use
h the atom w hen the a in equation 5.17 instead
r=
(5.10 elec of n in order to avoid con
)
4
7
tron is in a particular orbit
fusionand with the first
r w hen the electron is in a
quantum num ber.) By
2 change in the energy of the
m apply
atom as the electron
e moves from orbit 1 to orbit
2 - E2=
E quation 5.10 indicates th
w here v is the frequency
at the radius of the electron
of the electrom agnetic
now depends on the square
radiation and h is Plancks
of the first quantum num
constant. The velocity of
ber m ultiplied by a
electrom agnetic radiation
constant a w hose m
c is related to its w
agnitude is:
avelength A and the
frequency v by:
h2
(5.11) c = Xv
4
7
r Therefore, the energy of
2
m the radiation em itted by
5.3 EMISSION OF X-RAYS 63
ing quantum m echanics to R u th erfo rd s m odel of the (1887-1961), som e aspects of it are still in use in the
hydrogen atom B ohr was able to calculate the w physics of x-rays.
avelengths of light th at hydrogen atom s can em it as a In a com m ercial x-ray tube a stream of elec trons is
result of transitions am ong th e orbits the electron can accelerated by a voltage difference in a vacuum before
occupy. By com bining equations 5.13, 5.14, and 5.16 we striking a target com posed of a m etal such as chrom ium ,
have: copper, m olybdenum , or tungsten. The energetic
27r2me4/ 1 1 he electrons interact with the electrons of the target atom s
(5.18) and m ay knock them out of their orbits. X -rays are then
T generat ed when the resulting vacancies are filled by other
where n x and n 2 are th e quantum num bers, and n2 > n x electrons.
as required by the R ydberg form ula. It follows that:
The wavelengths of these x-rays depend on the
1 difference in the energies of the orbit in which a vacancy
1 _ 27r2me4 / 1
has been created and the orbit from which the replacing
A h3c (5.19)
electron originated, as re quired by equation 5.16. These
By com paring equations 5.19 (B ohrs m odel) and 5.17 x-rays therefore have discrete wavelengths and give rise to
(R ydberg form ula) we see that: a wave length spectrum th at is characteristic of the target
elem ent. The characteristic x-ray spectra w ere dis
R= 2iT2m e4
(5.20) covered in 1911 by Charles G. B arkla (1877-1944), who
h3c received the N obel Prize for physics in 1917. The
relationship betw een the characteristic x-ray spectrum and
B ohr substituted appropriate values into equation 5.20
the atom ic num ber of the target ele m ent was discovered
and calculated a value of 109,677.76 cm - 1 for the R
in 1913 by H . G. J. M oseley (1877-1915).
ydberg constant. The result agreed very well with the
observed value and thereby confirm ed B ohrs m odel for
the hydrogen atom . Niels B ohr was justifiably pleased
In addition to the characteristic wavelength spectrum ,
with this result and subse quently devoted him self w
a continuous x-ray spectrum is p ro duced by electrons th
holeheartedly to the study of quantum mechanics.
at pass through the target atom s w ithout actually colliding
with any of their electrons. N evertheless, som e energy is
The R oyal D anish A cadem y of Science, which
trans ferred in this type of interaction, and this energy is
receives a large portion of its funding from the C arlsberg
em itted as x-rays having continuously varying
Brewery, gave B ohr a grant of m oney to build an
wavelengths.
Institute for Theoretical Physics in C openhagen. B ohrs
institute becam e the center for research in quantum m
The characteristic x-ray spectrum of the tar get atom s
echanics and nuclear physics. It was fam ous n o t only for
is divided into several series of w ave lengths referred to by
its intellectual excitem ent but also for its relaxed atm
the letters K, L, M, etc. The K x-rays are em itted w hen
osphere created by the Professor (Gamow, 1966).
a vacancy in the low est energy level (K; n = 1) is filled.
The electron filling the vacancy in the K -level m ay
originate from the next higher level (L) or from others
associated with electronic orbits having still larg er radii.
5.3 Emission of X-rays
The L -series of characteristic x-rays is generated by
B ohrs m odel of the atom provided a direct expla nation transitions to the L-level (n = 2), and so on. The general
for th e emission of x-rays, which had been discovered in schem e of identifying the characteristic x-rays is m ore
1895 by W ilhelm K onrad R ntgen com plicated than th at shown in Figure 5.1 because
(1845-1923). A lthough B ohrs m odel has been electrons in each of th e K-, L-, and M -energy levels have
replaced by the wave mechanics of W. K. slightly
H eisenberg (1901-1976) and E . Schrdinger
64 THE ELECTRONIC STRUCTURE OF ATOMS
c
LU
L-series X -ray diffraction and x-ray fluorescence have becom
a 7 ' e indispensable tools in geochemical research. The form er
is based on the diffraction of
K-series Electron orbits x-rays by crystals (Klug and A lexander, 1954),
w hereas the latter arises from the emission of
characteristic x-ray by atoms. X -ray fluorescence is now
Figure 5.1 Schematic energy level diagram of atoms used to determ ine the chemical com posi tions of rocks
based on Bohrs model. X-rays of the K-series are including not only the m ajor ele m ents but also many
generated by electronic transitions to the K-energy trace elem ents (N orrish and H utton, 1969).
level (n = 1), whereas the L-series involves transitions
to the L-level (n = 2). Greek letters are used to
indicate the orbit from which the electrons originate.
In reality the characteristic wavelength spectrum is
more complicated than shown here because of small
differences in the energies of the electrons that
5.4 Schrdinger's Model of the
populate the M-, N-, and O-energy levels. Atom
B ohrs m odel w orked for the hydrogen atom and for
ions having only one electron such as H e+, Li2+, and
different energies. These differences are explain able in Be3+. How ever, it did not w ork well for atoms having
term s of wave m echanics and are not p re dicted in B o h two or m ore electrons. A rnold Som m erfeld (1868-1951)
rs m odel of the atom . m ade some im provem ents by intro ducing elliptical
We see in Figure 5.1 th at the K -series of char orbits like those of the planets in the solar system . In
acteristic x-rays is subdivided according to a sim ple addition, he allowed the planes of the orbits to include all
scheme. the free space around
Electronic transition Characteristic x-ray the nucleus, thereby m aking the atom look m ore like a
L to K K-alpha sphere than a flat disk. These refinem ents of B ohrs m
M to K K-beta odel required the introduction of a sec ond quantum num
N to K ber but still did not m ake it applicable to m ultielectron
K-gamma
atoms.
O to K K-delta
The problem with B ohrs m odel is that is treats
electrons as particles whose positions in space can
The L -series is subdivided similarly, but the details in both be determined. The validity of this idea was ques tioned
series get m ore com plicated by Louis-Victor 7th D uke de Broglie
5.4 SCHRDINGER'S M ODEL OF THE ATOM 65
gen erally stated in the o rb its of the B ohr-Som
(L892-1987) in 1923 who form: bers are needed to m erfeld m odel. The x-
dem onstrated that parti describe the m otions of orbital has spherical sym m
cles such as electrons also an electron in the space etry about the nucle us. T
d2ljj d2ip d2lp 87T2m .
have the properties of around the nucleus of an he p -type orbital is sym m
waves. The wavelength A ^? +^7- +jr2+~ ir{E~v),p=0 atom . These quantum etrical about the x-, y-, and
z-axes, and the d-orbitals
of particles of mass m and where ip is a wave num bers are:
velocity v is: are sym m etrical about the
function, m is the mass of n = 1, 2, 3 , . . . , 0 xy, xz, and y z planes.
the electron, E is the total 0
to
X .2
spouaj
k.
m
_2
c
oc
k_
cu
E
re
CD
cu
<U
0
*
tN
<55
59 68
CO
vo a
vo Q
vO Z
VO
1% o
LO U
3
i-
E
E
5
vo I
iO h -
u-> J3
T -O
vO >-
v> (X
iO LU
Pr Er
<
c
[c
u
ai
-0
4-*
cu
cu
E
QJ
WO
^
rx
O JZ
y~
3
^ Q.
Fm N o Lr
G> <Si
LU
co '*-
cn &
A mC m
CQ
t ffl
Md
n
u
Figure 6.1 The periodic table of the elements. The numbering of the groups has been changed to run consecutively from 1 to 18 across the table.
The old method is used here to emphasize the relationship between the electronic structures and highest positive valences of atoms.
76 THE PERIODIC TABLE AND ATOMIC W EIGHTS
attract an electron and
groups. The periods are T he num bering of m ents in th at group. In actually form ions with a
num bered in accordance the vertical colum ns or this version of the periodic charge of - 1 . N
with the first quantum num groups has been confusing table the elem ents in the evertheless, they have
ber of the orbitals that are to say the least. They used s- and p-blocks form ed electronic structures th a t
being filled with increasing to be num bered from left the A groups, and the m ake a valence of + 7 at
atom ic num ber. For to right in accor dance least technically possible.
elem ents in the J-block
with the highest positive The last group of elem
example, the elem ents of com posed the B groups.
valence of the ele- ents are the noble gases H
the first period con tain How ever, the identifi
e, Ne, A r, Kr, Xe, and R
electrons in the Is orbital, cation of A and B groups
n, all of which have filled
the elem ents of the second has not been consistent.
Table 6.1 R e la tio n sh ip BThe
e twvalences
e e n P e ofrio the
d s elem orbitals. Consequently,
period have electrons in
a n d E le c tro n O rb ita ls o f Aents
to mins othe
f th A
e groups, as they neither gain n o r lose
the 2s and 2 p orbitals,
E le m e n ts electrons under n a tu r al
those in third period are used in Figure 6.1, can be
Period Orbitals being conditions and th erefo re
filing 3s and 3p orbitals, deduced directly from the
have a valence of 0 . In
and in the fourth period filled electronic configurations
spite of th at fact, this
electrons enter 4s and 4p 1 \s of the atoms. For example,
group used to be given the
orbitals. H ow ever, the group IA consists of H, Li,
2 2s,2p num ber 8 , although 0 is m
fourth period contains an N a, K, Rb, Cs, and Fr,
3 3s, 3p ore appropriate.
additional group of 1 0 elem each of which has one
ents in which the 3d 4 4s, 3d, 4p electron in an s-orbital
Partly because of
orbitals are filling up. 5 5s, 4 d, 5p outside of a noble gas core
these kinds of inconsisten
Similarly, the fifth and 6 6s, 4f 5d, 6p and thus a valence of +1.
cies, chemists have
sixth periods each contain 7 Is, 5 f 6d, Ip Excluding hydrogen, the
decided to num ber the
elem ents in which d- elem ents in group IA are
Only 17 are known. groups consecutively from
orbitals are being filled. know n as the alkali
left to right. In this new
These elem ents together metals. The elem ents in
scheme the alkali m etals
form the so-called d-block group IIA , called th e
are still in group 1 but the
in the perio d ic table. The alkaline earths, include
noble gases are now in
sixth period contains 14 Be, Mg, Ca, Sr, Ba, and
group 18 (Figure 6.1).
extra ele m ents known as Ra. These elem ents have
Nevertheless, we will
the rare earths with two electrons in an s-
continue the old num
electrons in 4/orbitals. The orbital outside a noble
bering system outlined
rare earths are pulled out
aboveof +because it em
of the table and placed gas core and therefore have valences
separately below it The num bering of the A phasizes the
together with the actinide above continues w ith
elements in which p -block on the right side
5/orbitals are being filled O nce again the num bering
and which include the highest positive valence
transuranium or super consistent with the elec
heavy elements. We see, tronic configuration . Thus
therefore, that the num ber the elem ents in group 7A
of the elem ents in each can have valences of + 7
period is a direct conse because they all have five
quence of the progressive electrons in p -orbitals and
filling of electronic tw o elec trons in y-
orbitals. These orbitals. T he elem ents in
relationships are evident in this group are
Table 6.1. called th e halogens and include F,
and A t. In n atu re th e atom s of these
6.3 BASIC PRINCIPLES OF ATOMIC PHYSICS 77
+2, +3, + 4, + 6 , and even + the sam e place in the periodic table
close relationship betw een 8 in the case of R u and Os.
6.3 Basic they have the sam e (or very sim ilar
electronic structure of the elem ents. properties.
The elem ents w ith The periodic table is Princip The internal
d -orbitals are referred to founded on the recur rence les of composition of the nuclei
as the transition metals. of similar physical and of iso topes is indicated by
The groups in this block of chem ical properties of the
Atomi the shorthand notation
elem ents are num bered in elem ents with increasing c used to identify them .
atom ic num ber. This This notation consists of
accordance with th e
Physic the chemi cal symbol of
valences of the ele m ents statem ent is know n as the
just like the A groups. T periodic law of the elem s the elem ent, preceded by
herefore, group IB consists ents. In addition, elem ents the atomic
We know that atom s have
of the precious m etals Cu, in groups have the same num ber (Z) w ritten as a subscript, and the
nuclei com posed of
A g, and Au, which have valences because they num ber (A ) w ritten as a superscript, where:
protons and neutrons. The
electron configurations of have similar elec tronic A =Z+N
num ber of protons in the
dws 1 and therefore can structures. A s a result,
nucleus determ ines the T herefore, the notation nbC indicates
have a valence of +1. H ow their physical and
positive charge and
ever, cop per can also be chemical properties tend
therefore the num ber of atom or nuclide is an isotope of carbo
+2, and gold can be + 3 . to be similar.
extranuclear electrons nucleus contains 6 protons, and th at the sum
The num bering of the B required to form a neutral
In geochem istry this protons and neutrons is 12.
groups continues w ith IIB atom . The num ber of
m eans that the elem ents From equation 6.1 we find
com posed of Zn, Cd, and electrons and their th at the nucleus also
H g with d ws2 electron of a group tend to occur
together and display distribution in the contains 6 neutrons.
configurations that form +
geochemical coherence. For available orbitals give M ost chemical elem
2 ions. N ext, we move to
m etals, the alkaline earths, atom s their characteristic ents have two or m ore sta
the left to IIIB (Sc, Y, and
noble gases, and the chemical properties. ble isotopes. H owever,
La, d 1s2, valence + 3) and
precious m etals display Therefore, the atom ic num some elements, including
progress from there to
ber (Z) Be, F,Na, Al, P, Mn, Co,
group V IIB (M n, Tc, and well-developed geochem
Re, d5s2, valence + 7). H ical coherence in their and As have only one
indirectly determ ines the
ow ever, th e next three distrb bution in nature. stable
an atom . Since all atom s
m ust have the sam e
groups are anom alous. The iron group (Fe, Ru,
and O s) should have a valence of (or nearly the sam e)
dance w ith their electronic structure chemical properties, it
iron actually has valences of + 2 and +3. Cobalt follows that all atom s
and nickel also have valences of + 2 of a given ele
Therefore, these three m ent m ust have the sam e atom ic num ber.
groups are lum ped togeth The num ber (N
er as group VIIIB . This is atom s affects atom ic m asses
n o t a very elegant solu bearing on chem ical
tion, but it has tradition on atom s of the sam e chem ical
its side. The congeners of tain different num bers
Fe, Co, and Ni form the nuclei. A tom s th at have
platinum group ele m ents ber b u t different n eu tro n num bers
(P G E ) consisting of R u, isotopes. The w ord m
Os, R h, Ir, Pd, and Pt. eans sam e place and
These elem ents can occur indi cates th a t the
in the native state and, in
isotopes of an elem ent
addition, have valences of
all occupy
78 THE PERIODIC TABLE AND ATOMIC W EIGHTS
isotope. Several other elem ents, including K, Rb, Sm, Lu, expressed has changed over the years. Originally, chemists
Re,T h, and U, have long-lived radioactive isotopes that used oxygen as the standard and gave it an atom ic weight
decay so slowly th at they still occur in the solar system of 16.0000. H ow ever, after the discovery of isotopes,
even though they form ed m ore physicists adopted a differ ent scale based on '(JO =
than 6 X 109 years ago by nucleosynthesis reac 16.0000. A ll of the resulting confusion has now been elim
tions in ancestral stars. In addition, m ore than inated by the adoption of l\C as the standard . Therefore,
2 0 0 0
short-lived radioactive isotopes are known the masses of atom s are now expressed in the atom ic
that do not occur in nature because they have decayed mass unit (am u), defined as 1 / 1 2 of the mass of C , which
since the tim e of nucleosynthesis. is the m ost abundant stable isotope of carbon. The masses
A fairly large group of short-lived radioactive of the naturally occurring iso topes have also been determ
isotopes does occur in nature because these iso topes are ined by m ass spec trom etry, and the results are tabulated
continuously produced either by the decay of long-lived in standard
parents (U and Th) or by nuclear reactions involving
cosmic rays. The unstable daughters of uranium and reference books such as the C R C H a ndbook o f
thorium are polonium , astatine, radon, francium , radium , Chemistry and Physics (W east et al., 1986).
actinium , and protactinium . A n o th er large num ber of We are now ready to define the atom ic
radioactive isotopes is produced by nuclear reactions in the weight of an elem ent.
atm osphere: tritium , ,H , ' Be, '^C , 1 7 CI, and m any
others. The atomic weight o f an element is the sum of the
masses of its naturally occurring isotopes weight ed
The abundances of the naturally occurring iso topes in accordance with their abundances.
(stable or long-lived radioactive) are d eter m ined with
mass spectrom eters and are expressed in percent by num A n exam ple will clarify what this means. Silicon has
ber. For exam ple, boron has two stable isotopes whose three naturally occurring stable isotopes whose
abundances a r e 'B = 19.8% abundances and masses are
and " B = 80.2% . This m eans th at out of 1000
Isotope Abundance, % Mass, amu
boron atom s 198 are the isotope 'B and 802 are
?|Si 92.23 27.976927
'jB . Actually, it turns out th at the isotopic com po sitions
of boron and of other elem ents of low atom ic num ber S si 4.67 28.976495
(H , C, N, O, and S) m ay be changed by isotope Si 3.10 29.973 770
fractionation because mass differences am ong the isotopes
of an elem ent affect the kinet ics of chemical reactions and The atom ic weight of silicon is found by m ultiply ing the
the strengths of chem ical bonds. These effects are p art of masses of the isotopes by their abun dances expressed as
the sub ject of isotope geoscience presented in detail in a decim al fractions and adding the resulting products:
textbook by Faure (1986) and will be discussed in C hapter
18 of this book. $Si: 0.9223 X 27.976927 = 25.803 12
fJSi: 0.0467 X 28.976495 = 1.353 20
iSi: 0.0310X 29.973770 = 0.929 18
Sum = 28.085 50
6.4 Atomic Weights
The concept of the atom ic
weight is fundam ental to Therefore, the atomic
all form s of chem istry but weight of silicon, rounded
is n o t always clearly to five significant figures,
understood . O ne reason is 28.086. Chemists regard
for the uncertainty is that the atomic weights of the
the scale by m eans of elem ents as dimension-
which atom ic weights are
6.4 ATOMIC W EIGHTS 79
less num bers because they are expressed relative to the The relationship betw een m oles and the cor
mass of (G reenw ood and Eamshaw, 1984). responding num ber of atom s and m olecules enables us to
The im portance of the atom ic weights of the elem represent chem ical reactions by m eans of equations. For
ents arises from their use in the definition of the gram exam ple, the equation:
-atom ic weight. 2 H , + O , - 2 FLO (6.2)
The gram-atomic weight of an element is equal to the indicates that two m olecules of hydrogen (H 2)
atomic weight in grams.
m ust com bine w ith one m olecule of oxygen (0 2) in
Similarly, the gram -m olecular weight or the gram - form order to form two m olecules of w ater (H 2 0 ) . If
ula w eight of a com pound is defined as we mix 2 g of H 2 with 1 g of 0 2 we would not
achieve the desired proportions of molecules.
The gram-molecular weight (or the gram-formula H ow ever, w hen the am ounts of H 2 and 0 2 are
weight) of a compound is the molecular weight (or expressed in term s of moles, then the H 2 and 0 2
formula weight) in grams. m olecules will be present with the necessary 2 : 1 ratio
required for the reaction.
B oth the gram -atom ic weight and gram -m olecular The atom ic weights of the elem ents listed in Table
weight are referred to as the m ole (mol). This is the basic 6.2 enable us to calculate m olecular weights of com
unit of mass of elem ents and com pounds in chemistry. pounds based on their chem ical form ulas and to convert
am ounts of such com pounds from grams to the
The reason why the m ole is so im portant in
corresponding num ber of moles. For example, the m
chem istry is that one m ole of an elem ent or a
olecular weight o f B a S 0 4 (barite) is calculated from the
com pound always contains a fixed num ber of
atom ic weights of the ele m ents:
atom s or molecules. T hat num ber is:
the m ole stated on the inside back cover of this book. 233.366
W hen barite dissolves in w ater, it dissociates
1 H 1.00794 48 Cd 112.41
2 He 4.00260 49 In 114.82
3 Li 6.941 50 Sn 118.71
4 Be 9.01218 51 Sb 121.75
5 B 10.81 52 Te 127.60
6 C 12.011 53 I 126.905
7 N 14.0067 54 Xe 131.29
8 O 15.9994 55 Cs 132.905
9 F 18.9984 56 Ba 137.33
10 Ne 20.179 57 La 138.906
11 Na 22.9898 58 Ce 140.12
12 Mg 24.305 59 Pr 140.908
13 A1 26.9815 60 Nd 144.24
14 Si 28.0855 61 Pm (145)a
15 P 30.9738 62 Sm 150.36
16 S 32.06 63 Eu 151.96
17 Cl 35.453 64 Gd 157.25
18 Ar 39.948 65 Tb 158.925
19 K 39.0983 66 Dy 162.50
20 Ca 40.08 67 Ho 164.930
21 Sc 44.9559 68 Er 167.26
22 Ti 47.88 69 Tm 168.934
23 V 50.9415 70 Yb 173.04
24 Cr 51.996 71 Lu 174.967
25 Mn 54.9380 72 Hf 178.49
26 Fe 55.847 73 Ta 180.948
27 Co 58.9332 74 W 183.85
28 Ni 58.69 75 Re 186.207
29 Cu 63.546 76 Os 190.2
30 ' Zn 65.39 77 Ir 192.22
31 Ga 69.72 78 Pt 195.08
32 Ge 72.59 79 Au 196.967
33 As 74.9216 80 Hg 200.59
34 Se 78.96 81 Tl 204.383
35 Br 79.904 82 Pb 207.2
36 Kr 83.80 83 Bi 208.980
37 Rb 85.4678 84 Po (209)a
38 Sr 87.62 85 At (210)a
39 Y 88.9059 86 Rn (222)a
40 Zr 91.224 87 Fr (223)a
41 Nb 92.9064 88 Ra 226.025
42 Mo .95.94 89 Ac 227.028
43 Tc (98 )a 90 Th 232.038
44 Ru 101.07 91 Pa 231.036
45 Rh 102.906 92 U 238.029
46 Pd 106.42 93 Np (237)a
47 Ag 107.868 94 Pu (239)a
aMass number of longest-lived isotope in nature.
6.5 SUMMARY 81
both Ba2+ and SO2- is also 9.512 X 10- 6 m ol/L .The of hydrogen ions per m ole of the com pound, and
concentration of Ba2+ in 1 L of saturated barite solution therefore its equivalent weight is equal to its m olecular
can also be expressed as: weight divided by three. In this case, it m akes no
difference that phosphoric acid actual ly gives up its
9.512 X 1CT6 X 137.33 = 1.306 X 10 3 g /L hydrogen ions reluctantly and releases all three of them
A lternatively, we calculate the num ber of B a2+
only u n d er special con ditions.
ions in one liter of a satu rated barite solution by m eans of
A vogadros num ber:
Problems
1. Familiarize yourself with the names, atomic num 5. What is the weight of one atom of 2g2U in grams
bers, and positions in the periodic table of the elements given that the mass of this isotope is 238.050 784 amu?
having atomic numbers from 1 to 36. (Answer: 3.953 x 10~22 g)
2. Calculate the atomic weight of magnesium given the 6. How many atoms of iron are present in 5.00 g of
abundances and masses of its naturally occurring isotopes. hematite (Fe20 3)? (Answer: 3.77 X 1022 atoms)
7. What is the concentration of Cr in the mineral
chromite (FeCr20 4)? (Answer: 46.5%)
Abundance, % Mass, amu
?2Mg 78.99 23.985 042 8. If the concentration of S 0 4~ is 2.5 x 103 mol/L,
how many grams of SO|" are present in one liter?
Mg 10.00 24.985 837 (Answer: 0.24 g)
?^Mg 11.01 25.982 593
9. A solution of A12(S 0 4)3 contains 2 X 10 4 mol/L of
Al3+. H ow many grams of SO2- does it contain per liter?
(Answer: 24.305)
(Answer: 0.029 g/L)
3. Calculate the formula weight of orthoclase feld spar
10. How many gram-equivalent weights of Al3+ and
(KAlSi30 8) using the atomic weights in Table 6.2.
S 0 4' are in 2 mol of A12(S 0 4)3?
(Answer: 278.332)
(Answer: Al3+: 12; SO;". 12)
4. How many moles of Na+ are present in one liter of an
aqueous solution of Na2S 0 4 containing 4.760 g of the
compound? (Answer: 0.0670 mol)
References
A S IM O V , I., 1965. A Short VAN SPRO SE N, J. W., 1969. The
History o f Chemistry. Periodic System o f the
Doubleday, Elements.
Garden City, NY, 263 pp. Elsevier, Amsterdam, 368 pp.
F A U R E , G., 1986. Principles o f W E A S T , R. C., M. J. A S T L E , and W. H.
Isotope Geology. 2nd ed. Wiley, Handbook o f Chemistry and
New York, 589 pp. Raton, FL.
G R E E N W O O D , N. N., and A. E A R W E E K S , M . E ., 1956. Discovery
N S H A W , 1984. Chemistry o f the
Chem. Ed. Pub., 910 pp.
Elements. Pergamon Press,
Oxford, England, 1542 pp.
M A Z U R S , E . G ., 1974. Graphic
Representation o f the Periodic
System During One
Hundred Years. University
of Alabama
Press, Tuscaloosa.
7
Chemical Bonds, Ionic
Radii, and Crystals
The physical and chem ical properties of the ele m ents th e nonm etals. If a com plete transfer of elec
vary with increasing atom ic num ber in accordance with trons is achieved, the electron -donor atom s
the periodic law. Som e of these properties are neatly sum becom e positively charged ions called cations,
m arized in periodic tables available in university book w hereas the electron -acceptor atom s acquire a
stores. It will be helpful to use such a table to study the negative charge and form anions. O ppositely
topics presented in this chapter. charged ions are attracted to each other and
m ay form an electrostatic or ionic bond. M ost
m inerals are chem ical com pounds com posed of
cations (m etals) and anions (nonm etals) held
7.1 Electron Donors Versus together by ionic bonds.
Acceptors W hen two elem ents having sim ilar m etallic
or nonm etallic character interact, electrons are
We can divide the elem ents in the periodic table into two not actually transferred, but are shared by the atoms. The
groups based on their tendency to give up electrons or to sharing of electrons am ong the atom s of two or m ore
attract them . In general, the elem ents on the left side of different elem ents establishes a covalent bond betw een
the periodic table are the electron donors because their them . The elem ents do not necessarily share the
electronic orbitals are largely empty. The elem ents on the electrons equally, which causes covalent bonds to have a
right side of the periodic table have orbitals that are certain am ount of ionic character. Similarly, the transfer
nearly full, which causes them to be electron acceptors. of elec trons am ong atom s of elem ents on opposite sides
The elem ents that are located betw een the alkali m etals of the periodic table is never com plete. As a
on the left and the halogens on the right vary result, ionic bonds have covalent character
progressively in their tendencies to be electron donors or w hose m agnitude depends on the difference in the
acceptors, depending on how close they are to filling their tendency of the atom s to attract electrons to them selves.
orbitals with As atom s becom e m ore equal in their ability to attract
electrons, the covalent character of the bond increases.
electrons. The electron donors are the metals,
The distinction betw een ionic and covalent bonds is th
w hereas the electron acceptors are the non-
erefore an over-sim plification of a very com plex
metals. The m etallic character of the elem ents
interaction am ong atom s of different elem ents (Pauling,
generally decreases from left to right in the peri ods as
1960). The only truly covalent bonds occur when tw o
the orbitals are filling up with electrons.
atom s of the sam e elem ent com bine to form diatom ic
The properties of m etals and nonm etals are com
m olecules such as H 2 , 0 2, or N 2.
plem entary in the sense that the electrons given up by m
etallic elem ents are picked up by
8
3
84 CHEM ICAL BONDS, IONIC RADII, AND CRYSTALS
that, in fact, perm it us to period . The increase
do this. They are coincides with the progres
7.2 M identified and defined as sive filling of the
same effect can be seen in
period 3 in the first ion
e follows. electronic orbitals and ization potentials of M g
a reflects the reluctance of and A1 and of P and S.
atom s to give up electrons The first ionization
s First ionization from orbitals th at are potentials of the transi tion
potential is the
u energy required to
nearly full. For exam ple, m etals of perio d 4, which
we see that in periods 2 have partially filled 3d
r remove one electron
and 3 the first ionization p
from a neutral atom orbitals, vary m uch less
e in a vacuum and to o ten tials increase sharply than those in the first three
w ith increasing atom ic
s place it at rest an
num ber tow ard th e noble
periods. P eriod 4 starts out
infinite dis tance well enough with K and
o away. gas elem ents th at te r m Ca, b u t th en th e first
inate each period . W e can
f Electronegativity is a
measure of the
ioniza tion potentials
even recognize the effect change little as we com
M ability of an atom in on the first ionization plete the
a molecule to attract
e electrons to itself
potential of filling the s- sequence ending w ith Zn . G allium
orbitals or half-filling the the 4p orbitals, which causes it to
t (Pauling, quoted in
p-orbitals. This explains
Cotton and ionization potential than Z
a Wilkinson, 1962, p. why the first ionization n . H ow ever, the first
potential of B is less than
ll 89).
th at of Be and why it
ionization potentials th en
Standard electrode rise rapidly with
i potential is the takes less energy to rem increasing atom ic num
ove an electron from O
c voltage gener ated
than from N. The
ber until period 4 ends w
when one mole of ith Kr. The next period, 5,
C electrons is removed
starting w ith Rb and
h from one mole of an
ending with Xe, likew ise
element or ion in
illustrates the lack of
a the stan dard state.
differentiation am ong the
r elem ents with elec trons
a We deal with the standard in the Ad orbitals. The first
ionization potentials for
c electrode potential in
these and all other elem
Section 14.3 and consider
t here how the first ion ents are listed in Table 7.1.
Ionization potentials
required to remove
additional electrons rise to
high values because it
takes extraordinary am
ounts of energy to rem ove
electrons from atom s that
have acquired a positive
charge because they have
already lost one elec tron
or because the rem aining
electrons com plete ly fill
the orbital they occupy.
Therefore, during chemical
reactions in nature elem
ents generally lose only
their valence electrons,
that is, those that reside
outside of filled orbitals.
2
3 1.0
4 1.5
5 2.0
6 2.5
7 3.0
8 3.5
9 4.0
10
11 0.9
12 1.2
13 1.5
14 1.8
15 2.1
16 2.5
17 3.0
18
19 0.8
20 1.0
21 1.3
22 1.5
23 1.6
24 1.6
25 1.5
26 1.8
27 1.8
28 1.8
29 1.9
30 1.6
31 1.6
32 1.8
33 2.0
34 2.4
35 2.8
36
37 0.8
38 1.0
39 1.3
40 1.4
41 1.6
42 1.8
43 1.9
44 2.2
45 2.2
46 2.2
47 1.9
W
e
a
7.3 BONDING IN MOLECULES 87
The data in Tables 7.1 and 7.2 perm it us to
increasing atom ic num ber in each group.This indi cates
that the binding energy of the first valence electron of predict the ionic character of the bond betw een any two
elem ents in a given group decreases as the volume of the elem ents in the periodic table. In general, we find that
atom s in th at group increases. In other words, the larger bonds betw een neighboring elem ents in the periodic table
an atom , the m ore loosely it holds its valence electrons. are highly covalent. For example, the nonm etals on the
Consequently, the right side of the periodic table form m any fam iliar com
metallic character of the group A elem ents pounds and anion com plexes that are covalently bonded.
increases with increasing atom ic num ber. The H O bond in the H 2 0 , the C O bond in
In the B groups, w here d- and /-orbitals are
being filled, the electronegativities change m uch CO 2 - , the N O bond in N O 3 , the S O bond
less dram atically with increasing atom ic num ber in S 0 4~, and the Si O bond in S i0 4_ are all
than they do in the A groups. A s a result, the strongly covalent.
chemical properties of these elem ents are m ore In addition, we can dem onstrate th at bonds betw een
uniform than those of the elem ents in the elem ents from opposite ends of the peri odic table are
A groups. M oreover, the electronegativities of highly ionic. The best exam ples are the alkali halides (e.g.,
elem ents in B groups do not decrease w ith NaCl, KC1, and CsF). In fact, the Cs F bond is the m
increasing atom ic num ber as they generally do in the A ost ionic bond we know because Cs has the strongest
groups. For exam ple, the electronegativities of the tendency to be an electron donor (lowest electronegativity)
congeners of group V IIIB (Fe, Co, and Ni) actually rise and F is the strongest electron acceptor (highest elec
with increasing atom ic num ber. tronegativity). B onds with w ell-developed ionic character
The concept of electronegativity has m erit because also occur betw een the cations of m et als and com plex
the differences in the electronegativities of elem ents are anions exem plified by C aC O ,, N a N 0 3, M g S 0 4, and
related to the ionic character of the bond form ed by their Fe2 S i0 4, all of which occur naturally as minerals.
atoms. Accordingly, two elem ents with similar
electronegativities form covalent bonds with little ionic
character, whereas elem ents having different A lthough the electronegativities of complex anions
electronegativities form bonds whose ionic character is are not defined, we nevertheless know that the bonds betw
proportional to the m agnitude of the electronegativity een the cation of the m etals and the complex anions are
difference. The relationship betw een the electronegativity highly ionic because these com pounds dissociate into ions
differ ence and the ionic character of chemical bonds is w hen they dis solve in water.
expressed quantitatively by the data in Table 7.2. In Figure
7.1C we see how the bond character
0.2 1 1.8 55
0.3 2 1.9 59
0.4 4 2.0 63
0.5 6 2.1 67
0.6 9 2.2 70
0.7 12 2.3 74
0.8 15 2.4 76
0.9 19 2.5 79
1.0 22 2.6 82
1.1 26 2.7 84
1.2 30 2.8 86
1.3 34 2.9 88
1.4 39 3.0 89
1.5 43 3.1 91
1.6 47 3.2 92
(ID
Px Py
7.4 IONIC CRYSTALS 89
The angle betw een these p -orbitals of oxygen is 90 strongly th at they are in co rp o rated into crystals
because they are sym m etrical about the x-, y-, and e-axes. as w ater of hydration, as, for exam ple, in
T herefore, we expect that the angle betw een the H O C a S 0 4 2H20 (gypsum ).
bonds in the w ater m olecule should also be 90. H ow Therefore, the answ er to the original question
ever, the observed angle is th at w ater is able to dissolve ionic solids
is 104.45 (Fyfe, 1964). because it dissociates them into ions, which are then
T he discrepancy arises because the s- and separated from each other by becom ing
p -o rb itals of O are m erged in such a way th at hydrated. H ow ever, covalent bonds do not disso
th e angles b etw een the h y b rid ized orbitals are ciate in w ater or in other polar solvents.
no longer 90. H ybridization of orbitals is a Therefore, w ater is generally not an effective sol vent for
com m on ph en o m en o n in covalently bonded covalently bonded com pounds.
m olecules and is th e key to understanding the
shapes or ste reo ch em istry o f covalently bonded m
olecules (G reenw ood and E arnshaw , 1984; Pauling, 7.4 Ionic Crystals
1960).
The polarity of the w ater m olecule therefore arises W hen ions com e together to form a crystal they assem ble
not only because of th e unequal sharing of the electrons in regular patterns dictated by the requirem ent of closest
betw een H and O but also because one of the hybridized p packing. This require m ent arises because the strength of
-orbitals of O containing the two paired electrons points the electro static bond (F) betw een cations and anions is
away from the other two, which are involved in the inversely proportional to the square of the dis tance (r)
bonding. In addition, the H zO m olecule is inherently betw een them:
polar because the O atom has m ore electrons than the H
atoms.
Coordination
f - -'55
radius ratio exceeds a structure when the radius
critical value, ratio is
cA o
dim ensional patterns of fitting anions around cations. For
planar
example, four fold coordination occurs when the anions
occupy the corners of a tetrahedron with the cation in its
center. Similarly, sixfold coordination is realized in an
octahedron w here six anions occupy the cor 0.225-0.414 4 O A \ Tetrahedral
y
now been superseded by 12
1 H +1 extremely small
3 Li +1 0.68 0.82
4 Be +2 0.25 0.35
5 B +3 0.10 0.20
6 C -4 2.60 (P)
7 N +4 0.16 (A)
+3 0.16 (A)
8 +5 0.13 (A) 1.34
O -2 1.27 1.28 1.30 1.32
9 F -1 1.21 1.22 1.23 1.25 1.21 1.24 1.40
11 Na +1 1.07 1.08 1.10
12 Mg +2 0.66 0.75 0.80 0.97
13 Al +3 0.47 0.56 0.61
14 Si +4 0.34 0.48
15 P +3 0.44 (A)
16 +5 0.25 0.35 (A) 1.78
S -2 1.56 1.72
17 +6 0.20 0.30 (A) 1.65
Cl -1 1.67 1.72
+5 0.20 0.34 (A)
19 K +7 0.28 0.27 (A) 1.54 1.59 1.63 1.67 1.68
+1 1.46
20 Ca +2 1.08 1.15 1.20 1.26 1.36 1.43
21 Sc +3 0.83 0.95
22 Ti +2 0.94
+3 0.75
23 V -1-4 0.61 0.69
+2 0.87
+3 0.72
+4 0.67
24 Cr +5 0.44 0.54 0.62
+2 0.81 (L)
0.90 (H)
+3 0.70
+4 0.52 0.63
+5 0.43 0.52 (A)
25 +6 0.38 1.01
Mn +2 0.75 (L)
0.92 (H)
+3 0.66 0.66 (L)
0.73 (H)
+4 0.62
+6 0.35
+7 0.34 0.46 (A)
Table 7 .3 (continued)
Coordination numbers
z Symbol Valence // III IVs V
vr-h VII VIII IX
X
26 Fe +2 0.71 (H) 0.69 (L)
+3 0.57 (H) 0.86 (H)
0.63 (L)
27 Co +2 0.65 (H) 0.73 (H)
0.73 (L)
+3 0.83 (H)
0.61 (L)
28 Ni +2 0.69 (H)
0.77
+3 0.64 (L)
29 Cu +1 0.54 0.68 (H)
Coordination numbers
<
~o
nJ
Q
20 25
Figure 7.5 Variation of ionic radii with atomic number and valence. Note that ions having negative charges are larger
than those having positive charges. Note also that the radii of ions forming isoelectronic series decrease with increasing
atomic number (e.g., Li+ to F7+, Na+ to Cl7+, etc). The radii of ions in the same group that have the same charge increase
with atomic number (e.g., Li+ to Rb+, F to Br~, etc) [based on the ionic radii reported by Ahrens (1952) for sixfold
coordination].
alizations about the way 3. The radii of ions of
they vary within the p eri the same elem ent
odic table. We therefore decrease with
derive a num ber of con increasing positive
clusions about the ionic charge and
radii of the elem ents by increase with
exam ining Figure 7.5. increasing negative
charge.
1. The ionic radii of
isoelectronic
4. The radii of ions with charges of
series decrease
+3 am ong the transition m etals
with increasing
fourth period decrease
atom ic num ber
atom ic num ber, thereby
for both
traction of the
positively and
electron cloud as
negatively
the 3d orbitals are
charged ions.
filled.
2. The radii of ions 5. Ions of different
with the sam e elem ents may
charge in a group
have simi lar
increase with
increasing atom ionic radii.
ic num ber, th at
is, dow nw ard in
the periodic
table.
PROBLEMS 97
Exam ples to illustrate the last observation can be 7.6 Summary
found in Figure 7.5 or by consulting Table 7.3.
The chemical elem ents can be divided into electron
Na", Ca2+, Cd2+, Y3+, ~ 1.00 donors and electron acceptors, depending on
Tl3+, R EE(+3),Th4+ (sixfold coordination) w hether their valence orbitals are nearly em pty or
H f4+,Zr4+ ~ 0.80 A nearly full. The tendencies of elem ents to be elec
(sixfold coordination) tron donors, and therefore to act like metals, is
reflected by their first ionization potentials and
Fe3+, Co3+, Ni3+ ~ 0.65 A
electronegativities. Elem ents with low ionization
(sixfold coordination)
potentials and low electronegativities have strongly
developed m etallic character. In general, the m etal
The sim ilarity of the ionic radii of different ele lic character of the elem ents decreases from left to
m ents is an im portant phenom enon because it right across the periodic table and increases dow n
m ay perm it the ion of one elem ent to substitute ward in those groups in which the elem ents have
for an ion of another elem ent in ionic crystals valence electrons in s- and p-orbitals.
(G oldschm idt et al., 1926). Chemical bonds can be classified into ionic or
The radii of ions in aqueous solution may be covalent bonds, depending on whether the valence
significantly changed by hydration. In this connec electrons are effectively transferred or shared,
tion it is helpful to define the ionic potential as the respectively. In reality, all ionic bonds have varying
ratio of the charge of an ion in electronic charges degrees of covalent character as indicated by the
to its radius m easured in angstrom units. The ionic magnitude of their electronegativity difference.
potential is a m easure of the density of the electri Polar solvents such as w ater are able to break ionic
cal charge on the surface of an ion and reflects the bonds because of electrostatic interactions betw een
strength of the electrostatic bond the ion can form ions in the solid phase and charges on the w ater
with another ion of the opposite charge or with a molecules. The hydration of cations and anions fur
polar molecule like water. Ions with high ionic ther enhances the solubility of ionic solids in water.
potential attract m ore w ater molecules and form The radius ratio of a cation with respect to
hydrated ions that are larger than those of ions the surrounding anions determ ines its coordina
having lower ionic potentials. For example, the tion num ber and hence affects the three -dim en
radii of the hydrated ions of the alkali m etals sional patterns the ions form in ionic crystals. The
decrease from Li+ to Cs+ even though their ionic ionic radii of the elem ents in the periodic table
radii in crystals increase with atomic num ber vary predictably, depending on the atom ic num
(Table 7.3). This can be explained by considering bers of the elem ents, the electrical charges of the
the ionic potentials of the alkali metals: Li+ = 1.2. ions, and on the coordination num ber.
N a+ = 0.91. K + = 0.68. R b + = 0.64. and Cs+ = T he m etallic ch aracter of the elem ents, the
0.56. Therefore, even though L i+ (radius = 0.82 ) kinds of bonds they form with each other, and
is much smaller than Cs+ (radius = 1.78 ), it the radii of ions are all related to their position
forms a larger hydrated ion than Cs+ because of its in th e periodic table and thus to their electronic
higher ionic potential. configurations.
Problems
1. Determine the percent 2. Look up the ionization
ionic character of the fol- potentials of silicon in a
lowing bonds, based on the recent edition of the CRC
electronegativity differ- Handbook o f Chemistry
ences: (a) NaCl; (b) FeCl2; and Physics (Weast et al.,
(c) CuCl; (d) A1C13; (e) 1986) and plot them in
CC14. coordi-
98 CHEM ICAL BONDS, IONIC RADII, AND CRYSTALS
nates of electron volts (eV) versus the number of elec trons 6. Calculate the volume of an ion of Na+ in cubic
removed. meters (assume sixfold coordination).
3. Explain why the third and fifth ionization poten tials 7. Calculate the radius ratio for a planar structure
of silicon are significantly higher than those that with fourfold coordination.
immediately precede them. 8. Determine the ionic potentials of the ions in the
isoelectronic series K+ to Mn7+ and derive a conclu sion
4. Explain why the first ionization potential of sulfur
about the size of the hydrated ions in the iso electronic
(Z = 16) is less than that of phosphorus (Z = 15).
series (use sixfold and fourfold coordination depending on
5. Write an equation to represent the dissociation of the availability of data in Table 7.3).
(NH4)2S 0 4 into ions in an aqueous solution.
References
A H R E N S , L . H., 1952. The use of P A U L IN G , L., 1927. The sizes of
ionization potentials Part 1. ions and the structure o f
Ionic radii of the elements. ionic crystals. J. Amer.
Geochim. Cosmochim.
Chem. Soc., 49:763.
Acta,
PA U L IN G , L., 1960. Nature o f
2:155.
the Chemical Bond, 3rd ed.
C O T T O N , F. A., and G. W IL K IN S O
Cornell University Press,
N , 1962. Advanced Inorganic
Ithaca, NY.
Chemistry. Wiley, New S A R G E N T -W E L C H SC IE N T IF IC C
York, 959 pp. O ., 1980. Periodic Table o f the
F y f e , W. S., 1964. Elements. 7300 Linder
Geochemistry o f Solids. Ave., Skokie, IL 60076.
McGraw-Hill, New
SH A N N O N , R. D., and C.T. P R E W
York, 199 pp.
IT T , 1969. Effective crystal
G O L D S C H M ID T , V. M., T.
B AR T H , G. L U N D E , radii in oxides and
and W. fluorides. A eta
Z A C H A R I A S E N , 1926. Crystallogr., Sect. B,
Geochemische 25:925-946.
Verteilungsgesetze der SH A N N O N , R. D., 1976. Revised
Elemente. VII. Skrifter effective ionic radii and
Norske Videnskaps.-Akad. sys tematic studies of
Oslo, I. Mat.-Naturv. interatomic distances in
Klasse, No. 2. halides and chalcogenides.
G R E E N W O O D , N. N., and A. E A R Acta Crystallogr. Sect. A,
N S H A W , 1984. Chemistry o f the
32:751-767.
Elements. Pergamon Press,
W E A S T , R. C., M. J. A S T L E ,
Oxford, England, 1542 pp.
and W. H. B E Y E R , 1986. CRC
H E Y D E M A N N , A., 1969. Tables.
Handbook o f Chemistry
H andbook o f Geochemistry,
and Physics. CRC Press,
Verlag, Berlin, 442 pp.
Boca Raton, FL.
W H IT T A C K E R , E. J. W., and R.
M U N T U S , 1970. Ionic radii
for use in geochemistry.
Geochim. Cosmochim.
Acta, 34:945-956.
8
Ionic Substitutions in
Crystals
O ne of the principal objectives of geochemistry has been structure of crystals could be understood in term s of the
to explain the distribution of chemical elem ents in the E size and charge of the ions and that an ion of one elem ent
arth (Chapter 1). Before this objective could be achieved, can replace the ion of another if it is similar in size and
it was first necessary to determ ine how the elements are charge. These insights are sum m arized by a set of
actually dis tributed by analyzing large numbers of rocks generalizations know n as
and minerals from all over the Earth. F. W. Clarke took a G oldschm idts rules o f substitution.
giant step in that direction by his own efforts as an 1. The ions of one elem ent can extensively replace
analytical chemist and by presenting the data of those of another in ionic crystals if their radii
geochemistry in his book, first published in 1908. In
differ by less than about 15% .
subsequent years, geochemists continued to ana lyze
2. Ions w hose charges differ by one unit sub stitute
geological materials in order to determ ine the chemical
readily for one another provided electrical
composition of the E arth and to under stand the reasons
neutrality of the crystal is m ain tained. If the
for the observed variation in chemical composition of
charges of the ions differ by m ore than one unit,
different kinds of rocks.
substitution is gener ally slight.
This was by no m eans an idle exercise
because the inform ation contributed directly to 3. W hen two different ions can occupy a p ar ticular
the understanding of the origin of m etallic and position in a crystal lattice, the ion with the
nonm etallic ore deposits. The search for ore higher ionic potential form s a
deposits of all kinds continues to be one of the principal stronger bond with the anions surround ing the
tasks of geologists in society. In a very real sense, site.
geologists are still expected to know where to find the 4. Substitution m ay be limited, even w hen the size
good rocks that are the foun dation of our m odern and charge criteria are satisfied, w hen the com
industrial society. peting ions have different electronegatives and
form bonds of dif ferent ionic character.
such as albite (NaAlSi3 0 8) or halite (NaCl). We know, of 8.2 Camouflage, Capture, and
course, that this substitution does not occur. The reason is
Admission
that copper forms m ore cova lent bonds than sodium, as
indicated by their elec tronegatives. A ccording to G oldschm idts first rule, ions that have sim
ilar radii and equal charges replace each o th er extensively
The occurrence of elem ents in minerals and rocks in ionic crystals. The extent of substitution that actually
either as major or m inor constituents depends on their takes place depends on
abundances and chemical properties. In general, the most th e concentration of the ions in the m edium in
abundant elem ents form the m in eral com pounds within which th e crystals are form ing, on the tem pera
which the m inor elem ents may be accom m odated by ture, and on the com patibility of their bonds and
ionic substitution, in interstitial lattice positions, in fluid coordination num bers. Crystals form ing at high
inclusions, or as exsolved m ineral phases. Some m inor tem p eratu re are m ore tolerant of foreign ions
elem ents also occur in accessory minerals such as zircon than crystals form ing at a low tem perature.
or apatite. T herefore, the concentration of trace elem ents in crystals
The tendencies of elem ents to form m inerals are m can be used to estim ate the tem perature of form ation of
ost clearly displayed during the crystallization of a cooling certain m inerals. For example, th e concentration of Fe2+
magma. Therefore, G oldschm idts rules prim arily apply in sphalerite (Fe, ZnS) increases with the tem perature
during this process, which was thought to so rt the ions provided that enough iron was available to saturate the
according to their size and charge. Those ions th at do not spha lerite (K ullerud, 1959).
fit into the m ajor rock-form ing m inerals are said to be in
c o m pa tib le and therefore accum ulate in the residual m The second rule applies to substitution of ions of sim
agm a. Consequently, the incom patible elem ents are ilar size but having different charges. W hen th e charge
concentrated in late-stage differentiates of magmas, difference is greater than one, substitu
including aplite dikes, pegm atites, and hydrotherm al tion is limited because of the difficulty in m ain
veins. The elem ents in this category include K +, R b +, C taining electrical neutrality. C harge deficiencies
s+, Sr2+, B a2+, the rare earth elem ents (R E E ), Z r4+, H th at result from substitution of ions of unequal
f 4 +,N b 5 +,Ta5 +,T h4+, and U 4+. A ll of these elem ents charge m ust be com pensated by a second substitu tion
ultim ately associate them selves with rocks com posed of involving an ion having a different charge. This process of
silicate m iner als and, for that reason, are also referred to coupled substitution contributes to the diversification of
as the large ion lithophile (LIL) group. chemical com positions of m any minerals. A n alternative
to coupled substitu tion displayed by the clay m inerals is
adsorption of ions on the charged surfaces of small
G oldschm idts rules w ere critically review ed by m crystals.
any geochem ists, including Shaw (1953) and Burns and The third rule describes the effect of the ionic
Fyfe (1967). The rules are at best a first potential on the relative strengths of ionic bonds of ions
approxim ation to which there are m any excep com peting for the sam e site. Ions having a higher ionic
tions th at arise because the replacem ent of one potential (charge/radius ratio) form a stronger bond than
ion by another is generally site-specific, especially in com their com petitors and are therefore preferentially
plex com pounds. The problem is m agnified because the incorporated into the crystal. Similarly, com peting ions
ionic radii, used as a criterion in G oldschm idts rules, with lower ionic potentials are discrim inated against and
depend on the site w here the replacem ent takes place. In are ini tially excluded from crystals forming from a m agm
spite of these lim ita tions, G oldschm idts rules still serve a. Consequently, ions with high ionic poten tials are
a useful p u r pose by identifying som e of the m ajor concentrated in early-form ed crystals in a cooling magma,
factors that m ust be considered in the possible replacem w hereas those with low ionic potentials are concentrated
ent of ions in crystals form ing in a cooling m agm a or in the residual magma and enter late-form ing crystals.
from an aqueous solution.
8.2 CAM O UFLAGE, CAPTURE, AND ADMISSION 101
the relatio n sh ip betw een
A ccording to the In this case, the m inor in o rd er to preserve the th e ionic potentials of
fourth rule, a m inor ion elem ent does not form its electrical neutrality of the cations and the stren g th
can replace a m ajor ion own m inerals but is hidden in the
crystal crystals
lattice. of the bonds they form w
only if their another elem ent. Cam ouflage is displayed ith anions. This
electronegativi ties are Z r4+ (0.80 ) and H f4 Admission involves relationship m anifests
similar. W hen two m inor +(0.79 A ) because the entry of a itself in som e cases in th e
ions th at are sim ilar in hafnium rarely form s its foreign ion that has m elting tem p eratu res o f
size and charge com pete own minerals and is a lower ionic the com pounds they
always present in the m potential than the form . For exam ple, the M
for the sam e site, the one
ineral zircon (Z rS i0 4). major ion because it g sili
w hose electronegativity is
We could say that zircon has either a lower cate fo rsterite (M g2 S i0 4) has a h
m ore sim ilar to that of
crystals do not distinguish charge or a larger tem p eratu re (1910 C) th an th e F
the m ajor ion is preferred
between Z r4+ and H f4+ radius, or both. alite (Fe2 S i0 4), w hich m elts at 1503
because the bonds are m
m inerals form a solid solution k
ore com patible. ions and accept both with
m ineral olivine ((M g, F e ) 2 S i0 4)
equal ease.
The occurrence of R b + in lizes from cooling m agm
We see th at G
K -feldspar and other as of basaltic com posi
oldschm idts rules of Capture takes place tion . E arly form ed olivine is enrich
substitu tion provide when a minor potassium m inerals is an
fo rsterite (M g) end m em ber, w her
criteria for predicting the element enters a exam ple of admission
form ing at low er tem p
extent to which m inor crystal because R b + (1.57 )
eratu res is enriched in the
elem ents can replace m preferentially has a sm aller ionic
fayalite (Fe) end m em
ajor ele m ents in the m because it has a potential than K + (1.46
ber. A possible
inerals they form . A s a higher ionic ). O ther exam ples are
explanation is that early
result, we can predict how potential than the the replace m ent of Ca2+
form ed olivine captures
m inor elem ents distribute ions of the major (1.08 ) by Sr2+ (1.21 A )
Mg2+ (0.67 A)
th em selves w hen m element. in calcite, and the
in favor of Fe2+ (0.74 A ) because
inerals crystallize from a substitution of Cl- (1.72
ionic p o ten tial (A hrens, 1952).
cooling m agm a or from a E xam ples of capture A ) by B r (1.88 ) in
feldspar an o rth ite (C aA l2 Si2 O g)
supersaturated aqueous occur in the form ation of chlorides. T he extent to
m elting tem p eratu re than albite (
solu tion. Som e elem feldspar crystals, which m which ions are adm itted
presum ably because C
ents are selectively ay capture B a2+ (1.44 into a particular lattice site
a2+ (1.08 A ) has a higher
concentrated into the solid ) or Sr2+ (1.21 ) in decreases as the differ
ionic p o ten tial th an N a + (1.10 A
phase, whereas others rem place of K + (1.46 A ). A s ence in the radii of com
1952). A s a result, early -form ed p
ain in the liquid phase. a result, the peting ions increases. For
The different ways in concentrations of these example, I (2.13 )
which the ions elem ents in the residual replaces C1 (1.72 A )
m agm a decrease during much less than B r (1.88
of m inor elem ents are p a rtitioned
th e crys tallization of K A ) and B a2+ (1.44 A ) is
solid and liquid phases are described by
-feldspar. H ow ever, the less abundant in calcite
term s th at form the heading of rep
this section.
lace m ent of th e than Sr2+ (1.21 A).
univalent K + ion by a Evidently, adm ission of
foreign ions into the
Camouflage occurs divalent ion requires a
crystal of a m ajor elem
when the minor coupled substitution of A
ent is ultim ately
element has the l3+ for Si4+
controlled by the size
same charge and a
criteri on expressed in G
similar ionic radius
oldschm idts first rule.
as the major
element it is
H ie third ru le m
replacing.
akes a statem en t about
102 IONIC SUBSTITUTIONS IN CRYSTALS
anorthite celsian
betw een crystals and the liquid from which they E quation 8.3 indicates that the ratio of the con
form ed. centrations of the trace elem ent y in m inerals A
We therefore define the distribution coeffi and B is equal to a constant K, which is equal to
cient (D ): the ratio of the distribution coefficients of y in
Cx those m inerals at the tem perature of form ation.
(8.1) In principle, the tem perature dependence of
D =c'
the distribution coefficients of a trace elem ent in
w here C* is the concentration of a m inor elem ent different m inerals can be determ ined experim en
in the crystal (x) of a m ineral and C 1 is the con tally. The tem perature of form ation of rock sam
centration of that elem ent in the liquid (1 ) from ples can then be determ ined from the ratio of the
which the crystal form ed under equilibrium con m easured concentrations of the selected trace
ditions. The distribution coefficient of a particular elem ent in these two m inerals (M cIntyre, 1963).
elem ent in a specific m ineral m ay be > 1 , < 1 , or Consider, for exam ple, the replacem ent of
= 1. T he m agnitude of D is related to the verbal Z n2+ (0.68 , fourfold coordination) in spha
descriptors as follows: lerite (ZnS) and Pb2+ (1.26 , sixfold coordina
D> 1 capture tion) in galena by Cd2+ (0.88 , fourfold
coordination, 1.03 A, sixfold coordination). The
1
D< adm ission electronegativities of the three elem ents are simi
D= 1 cam ouflage lar and the bonds they form with sulfur are about
85 % covalent. A n experim ental study by B ethke
The num erical values of D m ust be determ ined and B arton (1971) indicated th at Cd is strongly
experim entally in the laboratory or by analysis of partitioned into sphalerite (sp) coexisting with
natural systems in which both crystals and liquid galena (gn) betw een 600 and 800 C and th at the
can be sampled. tem perature dependence of the distribution coef
T he observed distribution of trace elem ents ficients is given by:
in crystals can be used to estim ate th e tem p era (2080 - 0.0264P)
ture of form ation of coexisting m inerals. R ecall lo g tf(C d ) = - 1.08 (8.4)
th a t the extent of substitution is tem p eratu re
d ep en d en t because crystals becom e m ore to le r where:
ant of foreign ions as the tem p eratu re increases. r vn
T herefore, distribution coefficients are, in g en er K (C d) =
al, tem p eratu re d ep en d en t and are also affected - f 2 for Cd
by the com positions of th e liquid and th e crys C= concentration of CdS in weight percent
tals, as well as by pressure. If two m inerals A and
T=
B coprecipitate or crystallize from the sam e absolute tem perature in kelvins
solution or m agm a, a m inor elem en t (y) can P = pressure in atm ospheres
e n te r b oth m inerals A and B at a p articu lar te m
The relationship betw een P'(C d) and the tem
peratu re . The distribution coefficients of ele
m ent y are: p eratu re is found to be highly nonlinear w hen
cx equation 8.4 is plotted in linear coordinates in
and B= ( ^) ( 8 .2 ) Figure 8.2. H ow ever, in coordinates of log K (y-
C1/ coordinate) and 1 /T (x -coordinate) equation 8.4
W/A is a straight line in the slope-intercept form:
T he ratio of the distribution coefficients for ele
m ent y is: y=mx+b (8.5)
and Jupiter. Fragments resulting from collisions B, Al, O, Si, and som e of the transition m etals such
am ong them continue to be deflected into E arth - as Sc, Ti, V, Cr, Mn, and som e of their congeners.
crossing orbits and impact on the E arth as m ete The noble gases, H, and N m ake up the a tm o p h ile
orites. Goldschmidt, as well as Ida and W alter group.
N oddack in Berlin, analyzed the m etallic, sulfide, T here are a few surprises in this classification.
and silicate phases of many m eteorites and d eter For example, O is lithophile rather than atmophile,
m ined from the results how the elem ents had been and C, as well as P, dissolves in m etallic Fe in the
partitioned into the three immiscible liquids dur absence of O. Thallium is a chalcophile elem ent
ing the geochemical differentiation of the parent although it com m only substitutes for K + in silicate
bodies of m eteorites (N oddack and N oddack, minerals. We note also that Au, Sn, and M o are
1930). The inform ation derived from the study of siderophile, and presum ably have been carried
m eteorites was consistent with the chem ical com away by metallic Fe to form the core of the E arth.
positions of the silicate slag, F e -C u sulfide m atte, The same is true for Co and Ni, which occur as sul
and m etallic Fe, all of which form during the fides in ore deposits but prefer liquid Fe when
sm elting of ore. given a choice. O n the other hand, Ti, Cr, and Mn,
The resulting classification of the elem ents in which are com m only associated with Fe in igneous
Table 8.2 contains four groups th at G oldschm idt rocks, are not siderophile but lithophile elements.
nam ed siderophile (iron liquid), chalcophile (sul
fide liquid), lithophile (silicate liquid), and 8.6 .Summary
a tm o p h ile (gas phase). N ote th at several elem ents
occur in m ore than one group and that such sec The ions of different elem ents can substitute for
ondary affinity is indicated by parentheses. In gen each other provided their radii, charges, and elec
eral, we see from Figure 8.3 th at the elem ents in tronegativities are similar. In addition, they must
group V IIIB are siderophile together with C, Mo, be com patible with the lattice site in term s of
Au, Ge, Sn, and P. The congeners of groups IB and their radius ratios and coordination numbers.
IIB are joined by the elem ents G a, In, TI, Pb, As, G oldschm idts rules provide a rational basis for
Sb, and Bi, as well as by S, Se, and Te, to form the understanding the distribution of trace elem ents
chalcophile group. The lithophile elem ents include but do n o t give enough em phasis to the character
the alkali m etals (group IA ), alkaline earths of the site of replacem ent in the crystal lattice.
Ru, Rh, Pd Ag, Cd, In Na, Mg, Al, Si, P, , Cl (C), N, (O), Ne
Os, Ir, Pt Hg, TI K, Ca, (Ga), (Ge), , , Br Ar, Kr, Rn
plus plus Rb, Sr, , I
Mo, Re, Au (Ge), As, S Cs, Ba, (TI), plus
(W), and (Sn), Sb, Se Sc, Ti, V, Cr, Mn, (Fe)
C, P, Ge Pb, Bi, Te Y, Zr, Nb
Sn, (As), (Pb) and (Cr), (Fe) La, Hf, Ta, W
(Mo) REE, Th, , U, and
(H)
"The elements in parentheses occur primarily in another class.
8.6 SUMMARY 107
Figure 8.3 Geochemical classification of the elements in the periodic table.The classification is based on the way the
elements distribute themselves between an iron liquid (siderophile), a sulfide liquid (chalcophile), a silicate liquid
(lithophile), and a gas phase (atmophile).
The extent to which tion of the feldspars and of
108 IONIC
SUBSTITUTIONS IN
replacem ent of ions takes other aluminosilicate m
CRYSTALS from the chem ical com
place is described by the ineral groups.
terms camouflage, capture, The distribution positions of m eteorites
and admission. Trace elem coefficient is a quantitative m etallic Fe, transition m and from m etallurgical
ents whose ions closely m easure of ionic etal sulfides, and silicates. studies in th e smelting of
resem ble those of m ajor substitution. The num The geochem ical sulfide and oxide ores.
elements in terms of size, erical val ues of classification of the elem
charge, and distribution coefficients ents, which is based on
electronegativity are cam must be d ete r m ined this phenom enon, was
ouflaged in the crystals experim entally as a derived
formed by the major function of tem perature,
elements. Trace elem ents chem ical compositions of
that are captured during liquids and crystals, and of
the crystalliza tion of a pressure. The distribution
magma are concentrated of trace elem ents betw
into early-formed crystals een two coexisting m
and are depleted in the inerals th at form ed under
residual liquid. Ions that equilibrium conditions
are adm itted into the from the sam e liquid can
crystals of a m ajor ele m be used to determ ine the
ent are initially enriched in tem p eratu re of form
the residual magma and ation . Such geotherm om
subsequently enter late- eters are of great interest
forming crystals. in geochem istry but their
application is restricted by
Substitution of A l3+ the conditions that m ust
for Si4+ in silicate struc be im posed.
tures causes a charge im
balance in the lattice that
is neutralized by the com The geochem ical
plem entary introduction properties of the elem ents
of certain cations that are are reflected by th eir
com patible with the par distribution betw een a gas
ticular lattice sites. This phase and am ong natural
phenom enon of coupled liquids com posed of
substitution accounts for
the chemical composi
BEL
SON
Problems
1. Determine the coordination numbers of Ca2+ and 8. The melting temperatures of the fluorides of the
SF+ relative to O2-, and use the result to predict the alkali metals and alkaline earths are listed below. Consider
substitution of Ca2+ by Sr2' in calcite (coordination : 6) the observed variation in terms of both the ionic character
and aragonite (coordinate : 8). of the bonds and the ionic potentials. How should the
melting temperatures vary according to Goldschmidts
2. Predict whether Hg2+ can replace Sr2+ in the min rules? Which compounds conform to this prediction and
eral strontianite (SrC03). which do not?
3. Explain the formula of leucite (feldspathoid,
Melting
KAlSi20 6) in terms of coupled substitution.
Com pound temperature C
4. Examine the formula of the feldspathoid sodalite LiF 845
(Na8A l6Si60 24Cl2) and explain it on the basis of cou pled
substitution. NaF 993
KF 858
5. Silver (A g+) is a common trace element in galena
where it replaces Pb2+. Identify the ion of a chalcophile RbF 795
element that is best suited to enter galena with A g+ in a CsF 682
coupled substitution. BeF2 800 (sublimes)
6. Lithium (Li+) andMg2+ have similar radii and elec MgF2 1261
tronegativities, yet Li+ does not replace Mg2+ in olivine. CaF2 1423
Explain the reason for this occurrence and suggest anoth SrF2 1573
er host mineral for Li+ in which it does replace Mg2+. BaF2 1355
7. Zircon crystals (ZrSi04) commonly admit urani um
(U4+) but strongly exclude Pb. Deduce the valence of Pb in
magma from this observation.
References
A H R E N S , L. H ., 1952. The use M a s o n , B., 1962. Meteorites.
of ionization potentials Part 1. Wiley, New York, 274 pp.
Ionic M C I N T Y R E , W. L., 1963. Trace
radii of the elements. element partition coefficients
Geochim. Cosmochim.
Acta, 2:155. a review of theory and
B E T H K E , P. M., and P. B. B A R T applications to geology.
O N , J R ., 1971. Distribution
Geochim.
of some minor elements Cosmochim. Acta, 27:1209-
1264.
between coexisting sulfide
N O D D A C K , I., and W. N O D D A C K ,
minerals.
1930. Die Hufigkeit der
Econ. Geol., 66:140-163.
chemischen Elemente.
B U R N S , R. G., and W. S. F Y F E , Naturwissenschaften,
1967.Trace element 18:757-764.
distribution rules and their P A U L IN G , L., 1927. The sizes of
significance. Chem. Geol., ions and the structure of
2:89-104. ionic crystals, J. Amer.
G O L D S C H M ID T , V. M., 1937. The Chem. Soc., 49:765-790.
principles of distribution R IN G W O O D , A . E ., 1955. T h e p
of chemical elements in rin c ip le s g o v e rn in g tra c e
minerals and rocks. J. e le m e n t d is trib u tio n d u
Chem. Soc. London, rin g m a g m a tic cry stalliza tio
1937:655-673. n.
K U L L E R U D , G., 1959. Sulfide Geochim. Cosmochim.
systems as geological Acta, 7:189-202.
thermome S H A W , D .M ., 1953 .T h e c am o u
ters. In P. H. A
fla g e p rin c ip le a n d tra c e e
(Ed.), Researches in le m e n t d is trib u tio n in m a g
Geochemistry,
m a tic m in erals . J. Geol.,
vol. 1,301-335. Wiley, New 61:142 -151 .
York, 511 pp.
Ill
AQUEOUS GEOCHEMISTRY
AND THE STABILITY OF
MINERALS
Im portant chemical reactions occur at the surface of the E arth when m inerals are exposed to w
ater, carbon dioxide, and oxygen. The principles th at govern such reac tions can be used to explain the
interactions of ions in aqueous solutions and to con
strain the stabilities of solid com pounds with which the ions are in equilibrium . The
inform ation derivable from such studies is conveniently presented in graphical form
because E arth Scientists are accustom ed to using m aps and diagram s in their work.
9
Acids and Bases
Chemical reactions in nature commonly take place in the H owever, we deliberately added a larger quanti ty of salt
presence of water, which acts as the m edium within to the beaker than will dissolve in the w ater it contains.
which ions and molecules can interact. W hen two ions or As soon as the solid NaCl comes in contact
m olecules approach each other closely, they have an with the w ater in the beaker, it begins to dissolve
opportunity to establish a bond betw een them . W hen
and the concentration of N a+ in the w ater
they do so, we say that a chemical reaction has taken
place and a product has form ed. Such chemical reac
increases with time. A fter a few m inutes the N a +
tions are represented by m eans of algebraic equations concentration becom es constant regardless of
that express the balance of masses and charges of the how long we continue the experim ent. A t this
reactants and products. We use such equations to tim e the beaker still contains som e solid NaCl
understand chemical reactions in nature, and therefore that did not dissolve in the water. In order to interpret the
need to develop the skill of writing and interpreting results of this experim ent we turn to Figure 9.1, which is
them . For this reason, we begin the third part of this the recorder trace of our experim ent.
book with a very simple exercise.
We see that the N a+ concentration increased rapidly
at first but eventually leveled out and becam e constant.
W hen no m ore NaCl would dis solve in the w ater at the
tem perature and pressure of the experim ent, the solution
9.1 Chemical Reactions and becam e saturated with respect to NaCl. W hen we
Equilibria reached that con dition, the reaction achieved a state of
equilibrium in which the rate of solution of NaCl is equal
In order to understand how chemical reactions to the rate at which it precipitate s. Therefore, at equi
work we first consider what happens when sodi librium the am ount of excess solid NaCl and the
um chloride is dissolved in pure w ater at room concentrations of N a+ and Cl- ions in the saturat ed
tem perature and atm ospheric pressure. The solution are invariant with respect to time. Once the
apparatus consists of a beaker containing contin uously reaction reaches equilibrium, it stops being productive
stirred water, into which is placed an ion-sensitive because the forward and back ward reactions are
electrode capable of m easuring the concentration of N a+ occurring simultaneously and at the same rates. We can
ions. The electrode is con nected to a strip-chart recorder indicate this condition by means of equations. A t
operating at a known chart speed. We ignore all practical equilibrium the forward reaction:
limi tations and com plications and concentrate our
attention entirely on the events that occur when
N aCl(s) ^ N a+ + Cl (9.1)
we place a quantity of solid NaCl into the and the backw ard reaction:
w ater in the beaker. The results of this experi
m ent are entirely predictable based on past N a + + CI N aCl(s) (9.2)
experience: the salt will dissolve in the water.
110
9.1 CHEMICAL REACTIONS AND EQUILIBRIA 111
expressed by an equation The reaction we used
is in a state of equilibrium changes in the physical in our experim ent
. and chemical conditions, it involves the dissociation
O ur experim ent responds by reestablishing of a solid into ions.
illustrates the im portant a new state of equilib Therefore, in this reaction
point that chemical rium . This property of no chemical bonds are
reactions have a natural chemical reactions was actually form ed by close
ten dency to progress tow expressed by H enry Le encounters betw een ions
ard equilibrium . In order Chtelier in 1888 in these or molecules. We chose
to reach this condition th e words: A ny change in one that reaction mainly
reactants are consum ed of the variables that because it is familiar to all
and products are form ed determ ine the state of a of us. A nother reaction
such th at the am ounts of system in equilibrium that is familiar to
both change with time. causes a shift in the geologists is the decom
We can think of this dise position of equilibrium in position of calcium
quilibrium condition as a direction that tends to carbonate by hydrochloric
Figure 9.1 Increase of the
concentration of Na+ with the active m ode of a reac counteract the change in acid. This reac tion can be
time after the addition of tion. A fter a reaction has the variable under represented by the
solid NaCl to pure water reached equilibrium it consideration (M oore, following equation:
at 25 C. This experiment becom es inactive in the 1955, pp. 79-80). In other
illustrates the point that sense th at the am ounts of
+ 2 HC1 - words,
Ca2+ +when
2 C1a+reaction
H zO + Cat 0 2 (9.4)
chemi cal reactions reactants and products per equilibrium is disturbed, it
progress toward a state of unit w eight or unit reestablishes equilibri um
equilibrium in which the volume of w ater becom e The carbon dioxide that is
by counteracting the
concentrations of produced escapes in the
constant. In fact, w hen a disturbance. This is the
reactants and products are
chemical reaction at essence of Le Chteliers form of bubbles when the
invariant with time. In this
equilibrium is disturbed principle. This principle reaction is carried out in
illustration the reaction
reaches equilibrium when by will be very useful in the contact with the atm
the solution becomes interpretation of equa osphere. U nder these
saturat ed with respect to tions representing conditions the reaction
NaCl. While reactions are chemical reactions. cannot reach equilibrium
at disequi librium they because the C 0 2 does not
effectively convert O ur experim ent accum ulate. Therefore,
reactants into products leads us also to understand the reaction will continue
When reactions reach
the m eaning of the in the direction indicat ed
equilibrium, the rate at
concept of solubility, by the arrow in equation
which products are
converted back to which is defined as the am 9.4 until either the HC1 or
reactants equals the for ount of a com pound th at the C a C 0 3 is used up. In
ward reaction rate and, as dis solves to form a this case, the reaction is
a result, the net change in saturated solution. prevented from reaching
the concentrations of Therefore, in order to m equilibrium and instead
reactants and products is easure or calculate the runs to completion.
zero. solubility of a com pound
the reaction by which it
occur at the same rates. dissolves m ust be at
equilibrium . This is an im
We represent this state by
portant point to which we
m eans of double arrows:
will return when we
N a C l( s ) ^ N a + + Cl discuss solubility
That is, double arrows calculations.
mean that the reaction
112 ACIDS AND BASES
Here we encounter the first of many differ ences The rate of the forw ard reaction is:
between geochemistry and chemistry. In the chemistry v f = /cf(A )(B ) (9.6)
laboratory the dissolution of C a C 0 3 by HCl can be
and the rate of th e backw ard reaction is:
carried out in such a way that the COz cannot escape. In
this case, the reaction will achieve equilibrium and excess = k b(C )(D ) (9.7)
C a C 0 3 and H C l will then coexist without appearing to
w here (A ), (B), (C), and (D ) are the m olar
react with each other. However, when this reaction occurs
under natural conditions, it does not achieve equilibrium am ounts or concentrations of the com pounds or elem
because the CO z escapes into the atm osphere. Similarly, ents sym bolized by A, B, C, and D and k f and
chemical reactions on the surface of the E arth may not
k b are proportionality constants. A t equilibrium
achieve equilibrium because some of the prod ucts v f = v b. Therefore:
continually escape either because they are gases or
kf(A )(B ) = k b(C )(D ) (9.8)
because ions and molecules are carried away by the
movement of groundwater. which can be rew ritten as:
kf (9.9)
A nother reason why some reactions in nature fail to (C )(D ) = K
achieve equilibrium is th at their reaction rates are slow.
kh (A )(B )
This is especially true of reactions involving the transform
ation of a solid com pound into another solid. Such w here K is the equilibrium constant. In the gener al case
reactions require a com plex chain of events and their of a reaction at equilibrium represented by:
rates are deter m ined by the slowest or the rate-determ aA+bB^cC+dD (9.10)
ining step in the process. However, we will see th at the the Law of M ass Action, first form ulated by
fail ure of some reactions in nature to achieve equi librium
does not diminish the im portance of this condition as an G uldberg and W aage in 1863, takes the form:
aid to understanding the geo chem istry of the surface of K (9.11)
the E arth . (CT(D)d =
(A )fl(B )
w here a, b, c, and d are the m olar coefficients
taken from a balanced equation representing the reaction,
9.2 The Law of Mass Action and (A ), (B), (C), and (D ) are the con centrations of
reactants and products of the reac tion at equilibrium . The
The alchem ists originally believed that chemical reactions left side of equation 9.11 is the reaction quotient, which
occur becuse certain elem ents and com pounds love each changes continu ously as a reaction proceeds tow ard
other. R o b ert Boyle expressed reservations about this equilibrium .
idea in 1661 and Claude Louis de B erthollet pointed out W hen a reaction has achieved equilibrium, the
about 150 years later that the direction of chem ical concentrations of reactants and products do not change
reactions can be reversed by adding excess am ounts of and the reaction quotient becom es a con stant know n as
one of
the equilibrium constant K (M oore, 1955).
the products (M oore, 1955). The apparent
reversibility of chemical reactions ultim ately led A lthough it is true that at equilibrium the rate
to the conclusion that reactions achieve a state of of the forw ard reaction is equal to that of the
equilibrium w hen the rate of forw ard reaction is equal to backw ard reaction, the Law of Mass A ction
the rate of the backw ard reaction . For a simple reaction expressed in equation 9.11 cannot be derived from
such as: kinetics. The reason is th at chem ical reactions con sist of
sequences of discrete steps occurring at their own rates.
(9.5)
Therefore, the overall rate of a
9.2 THE LAW OF M ASSACTION 113
reaction is not necessarily a linear function of the m olar tant in very dilute solutions and vanishes at infinite
concentrations of reactants and products, but m ay vary as dilution when y = 1.0. We will treat the activity coefficient
the square, the cube, or even the square root of the as a dimensionless num ber, which enables us to express
concentrations. T herefore, the Law of M ass A ction was activities in the same units as concentration.
initially stated only as a generalization of experim ental
results and did not have the force of a scientific law. It The units of concentration of ions and m ole cules in
turns out, how ever, th at the Law of Mass A ction can be aqueous solutions are based on am ounts expressed in m
derived from the first and second laws of therm odynam ics oles but differ in term s of weight or volum e of solvent or
for reactions am ong ideal gases at equilibrium . In order to solution.
apply it to reactions am ong ions and m ol ecules in
Molality (m) is the number of moles of solute per kg
aqueous solutions we m ust replace their m olar
of water.
concentrations by their activities. The activ ity a of an ion
Formality (F) is the number of moles of solute per kg
in solution is related to its m olar concentration c by the
of solution.
activity coefficient y such that:
Molarity (M) is the number of moles of solute per
liter of solution.
Normality (N) is the number of equivalent weights of
a = yc (9.12) solute per liter of solution.
The activity coefficients correct the m olar con
The m olarity and norm ality are expressed in terms of
centration of ions for the interference by other ions in real volumes of solution, whereas m olality and for mality are
solutions. In m ost cases, the values of activity coefficients based on weight of the solvent or solu tion, respectively.
are less than one, which indi cates th at the effective The latter are to be preferred because they are independent
concentration (or activity) of the tem perature, whereas m olarity and normality vary
of ions is less than their actual concentration . with tem per ature because of the expansion of w ater with
N um erical values of activity coefficients can be increasing tem perature. The data in Table 9.1 indi
calculated from the D ebye -H ckel theory and
its extensions, which we will discuss later. cate that the volum e of 1 g of w ater increases from
T he Law of M ass A ctio n sta te d in e q u a tion 1.000 13 m L at 0C to 1.043 42 m L at 100C or
9.11 m ust now be restated in term s of activ ities: If a about 4.3% . M oreover, the specific volume of
reactio n rep resen ted by equ atio n 9.10 is at equilibrium , water has a m inim um value of 1.000 00 m L at
then: 3.98C. Therefore, concentrations based on vol umes m
ust specify the tem perature at which the volume was m
(9.13)
[C]c[D ]' = K easured. On the other hand, it is m ore convenient to m
[A)[B]b easure volumes of liquid than it is to weigh them .
where the brackets [ ] symbolize activities, wheras Therefore, both kinds of concentration units are in use.
parentheses ( ) symbolize concentrations of ions and W hen a chem ical reaction betw een a solid and its
molecules. U nfortunately, there is no uniform convention ions and molecules in solution has achieved equilibrium ,
to represent activities. We use brackets for activities and the Law of Mass A ction applies with the following
parentheses for concentrations following the exam ple of conventions:
G arrels and Christ (1965) and K rauskopf (1979), w
hereas Stumm and M organ (1970) used the opposite 1. The activities of ions and molecules must be
convention and D rever (1988) symbolized activities by a expressed in terms of moles but may be referred
low er case a with appropriate subscripts. The difference to a unit weight of solvent (molal ity), a unit of
between activity and concentration is unim por weight of solution (formality), or a unit of volume
of solution (molarity).
114 ACIDS AND BASES
Table 9.1 D e n sity a n d S pecific V o lu m e o f solutions in which the ratio of w ater to solute is sig
P u re W a te r F re e o f A ir nificantly reduced. The convention to express the partial
pressures of gases in units of atm ospheres is arbitrary but
T , C D e n sity , g / m L V o lu m e , m L / g
affects the num erical values of equi librium constants.
0 0.99987 1.00013
The tem parature and pressure at equilibrium m ust be
3.98 1.00000 1.00000 specified because equilibri-ium constants vary with tem
5 0.99999 1.00001 perature. The effect of pressure on equilibrium constants is
10 0.99973 1.00027
negligible under Earth-surface conditions.
15 0.99913 1.00087
20 0.99823 1.00177
25 0.99707 1.00293
30 0.99567 1.00434
9.3 Dissociation of Weak Acids
35 0.99406 1.005 97
and Bases
40 0.99224 1.00782
45 0.99025 1.00984 Following the tradition of the Swedish chemist Svante
50 0.98807 1.01207 Arrhenius, we define an acid as a com pound that releases
55 0.98573 1.01447 hydrogen ions when it is dissolved in water. Similarly, a
base is a com pound that releases hydroxyl ions in aqueous
60 0.98324 1.01704
solution. These defini tions are no longer accepted by
65 0.98059 1.01979
chemists who now prefer B rpnsteds definition that an
70 0.97781 1.02269
acid is any sub stance that can donate a proton to another
75 0.97489 1.02575
sub stance and a base is a substance that can accept a
80 0.97183 1.02898
proton . Strictly speaking, protons or hydrogen ions (H +)
85 0.96865 1.03236 cannot exist unattached in w ater and interact with w ater
90 0.96534 1.035 90 molecules to form hydronium ions (H 3 0 +). We adopt the
95 0.96192 1.03958 A rrhenius concept of acids and bases because in
100 0.95838 1.04342 geochemistry we deal only with aqueous solutions of
eletrolytes and we will represent protons as H + regardless
SO U R C E : Weast et al. (1986), p. F-10.
of whether they actually occur in this form.
Hydrochloric acid (HC1) hydroxides of alkali dissociated into ions. T herefore, at equilibrium :
[CH 3 C O O H ] = 0.1 - x
metals
Nitric acid (H N 03) hydroxides of alkaline [CH 3 C O O -] = [H +] = x
earths (except Be)
Sulfuric acid (H2S 0 4) CH3COOH Substituting into equation equation 9.16, we note that
lanthanum hydroxide -> C H 3 C O O - 9.15 we obtain a qua K is a very small num ber,
(La(OH)3) + FL dratic equation: which im plies th at only a
(9.14) sm all fraction of the
x2
D issociation reactions = 1.76
The com m on w eak 0.1 x
acids include acetic acid like this one reach equilib
(C H 3 C O O H ), carbonic rium very quickly and are This equation can be
acid (H 2 C 0 3), phosphor then subject to the Law of solved by putting it into
ic acid (H 3 P O 4 ), and Mass Action: the standard form:
silicic acid (H 4 S i0 4). A [H +][CH 3 COQ~ ax2 + bx
m ong the weak bases we which then yields the quadratic formula:
m ention am m onium [CH 3 C O O H ]
hydrox ide (N H 4 O H ), w here K = 1.76 X 10~ 5 at 25 C and -b (b2 -
nickel hydroxide (N i(O small value of K indicates that at x=
H )2), cop p er hydroxide m ost of the acetic acid will rem ain in the undis
(C u (O H )2) and the sociated m olecular form . In ordInerour
to case, we find th at a = 1,
hydroxides of the R E E s tion 9.15 we now specify thatand c = - 1 . 7 6 x l 0 -6 and that
except La. A clear com pound was dissolved in 1 Lyields x l = +1.32 X IO- 3 and
distinction m ust be m ade We discard the negative
betw een the strength of an root of the quadratic
acid or base and its co n equation and conclude
centration. For exam ple, m olar solution of acetic acid contains:
hydrochloric [H +] = 1.32 x
acid rem ains a strong acid even in dilute solu
[CH 3 C O O '] = 1.32 X
tion because it is com pletely dissociated . Sim i
larly, acetic acid rem ains [CH 3 C O O H ] = 0.1 - 1.32
a w eak acid even in a co n The degree of dissociation
cen trated solution because
solution of acetic acid is:
it is only partially 3 2
D- 1 X /
dissociated.
The activity of hydrogen ions is a useful p ara The dissociation constant of water, like all other
equilibrium constants, varies with tem pera ture, as shown
m eter because it reflects directly or indirectly the in Table 9.2. Therefore, th e neutral
9.3 DISSOCIATION OF W EAK ACIDS AND BASES 117
1979). The reactions [H 2
Table 9.2 D isso c ia tio nrapidly
C o n sachieve
ta n t have the sam e S(aq)] =
o f W a te r a t D iffe re n t T equilibrium
e m p e ra tuand
re sthen concentration in both of 0.1 - X
T, C - lo g K w
satisfy the Law of Mass the equi libria in which [H
Action: they occur. However, the +]
0 14.9435 =x
am ount of H + produced
[H +][H S ~] by the first equilibrium
5 14.7338 [HS
10 14.5346 (9.30) is m uch greater ~] =
[ h +][s2 ]
15 14.3463
than th at contributed by x
14.1669
the second equilibrium
20 [HS] (9.31) because K x is From the Law of Mass A
24 14.0000 ction we obtain:
H ere we encounter for the about 1 million tim es
25 13.9965
first tim e two sim ulta larger than K v
neous equilibria th at have (9.34)
S O U R C E : Weast et al. 0 .1 -
(1986), p.D-164. (H + and H S - ). Obviously, The problem is to x
calculate the activities of
Since K x is a very small
all ionic and m olecular
num ber, we predict that
point of the pH scale is at pH species in a solution con
only a small fraction of H
Actually, the data in Table taining 0.1 mol of H 2S
2 S(aq) will be dissociat
K w = IO- 1 4 0 at 24C and at 25 C it is slightly dissolved in 1 L of pure w
er than this value. ater at 25 C. We will dem ed at equilibrium ,
Several weak acids onstrate two strategies for which justifies the
are capable of yielding solving problem s assum ption
two or m ore hydrogen involving two or m ore th at x 0.1 and that 0.1 - x 0
ions or protons per m simul taneous equilibria: equation 9.34 becomes:
olecule of acid. W hen 7.0
A dds
Acetic CH3COOH 4.75 1
Arsenious H3A s 0 3(aq) 9.2 2
Arsenic H3A s 0 4(aq) 2.2 7.0 11.5 2
Boric H3B 0 3(aq) 9.2 2
Carbonic H2C 0 3(aq) 6.35 10.3 2
Hydrofluoric HF(aq) 3.2 2
Phosphoric h 3p o 4 2.1 7.2 12.4 2
Hydrosulfuric H 2S(aq) 7.0 12.9? 2
Sulfuric H2S 0 4(aq) 2.0 2
Hydroselenic H 2Se(aq) 3.9 15.0 2
Selenic H 2S e 0 4(aq) 1.9 2
Silicic H4S i0 4 9.71 13.28 9.86 13.1 3
(pK 4)
Bases (Hydroxides)
Ammonium NH4OH(aq) 4.7 ; 2
Aluminum, amorph. Al(OH)3 12.3 10.3 9.0 2
Aluminum, gibbsite a i ( o h )3 14.8 10.3 9.0 1.4 2
Beryllium, amorph. Be(OH)2 2.2 2
Cadmium Cd(OH)2 10.5 3.9 4.1? 6.7 2
Cobalt Co(OH)2 10.6 4.3 5.2 6.5 2
Copper Cu(OH)2 13.0 6.3 2.9 2
Ferrous Fe(OH)2 10.6 4.5 5.1 8.4? 2
Ferric, amorph. Fe(OH)3 16.5 10.5 11.8 4.4 2
Lead, red PbO + H20 9.0 6.3 1.4 4.4 2
Magnesium Mg(OH)2 8.6 2.6 2
Manganese Mn(OH)2 9.4 3.4 5.1 2
Mercurous, red HgO + H20 14.8 10.6 4.5 3.6 2
Nickel Ni(OH)2 11.1 4.1 4 7 2
Silver 2 Ag20 + 5 H20 3.4 5.7 2
Thorium, amorph. Th(OH)4 10.3 5.8 10.8 2
uo2 u o 2( o h )2 14.2 8.2 3.6 (P*4) 2
Vanadium v ( o h )3 11.7 2
VO v o ( o h )2 15.2 8.3 2
Zinc, amorph. Zn(OH), 10.5 5.0 1.9 4.4? 2
Explanations: pK = - log K
Zn(OH)2(s) Zn(OH) + + OH- K,
Zn(OH)+ ^ Zn2+ + OH K2
Zn(OH)2(s)^Zn(OH)2(aq) Kaq
Zn(OH)2(s) + OH- =^Zn(OH)3 KA (amphoteric)
means does not apply or unknown.
b(l)Weast et al. (1986); (2) Krauskopf (1979); (3) calculated from thermodynamic data of Lindsay (1979).
9.4 SO LUBILITY OF SPARINGLY SOLUBLE BASES 121
combining equation 9.54 and 9.52. In ord er to accomplish ous equilibria existing in a given environm ent. Therefore,
this, we first invert equation 9.54: the degree of dissociation of w eak acids and bases and the
resulting solubility of o th erwise insoluble weak bases are
C d(O H )2 ( a q ) ^ C d ( O H ) 2 (s) 1 / K aq (9.57) controlled by the activity of H + ions in the environm ent.
[Mg2+] =
[Mg2+] [ O H l i o -726
= 11.48 X 10- 5 m o l/L (9.71)
[M g (O H *)] - 10~ <9-62>
[H +][O H - ] = IO 1 4 0 (9.63) T he solubility of M g(O H ) 2 in pure w ater at
PH
Figure 9.2 Solubility of brucite (Mg(OH)2) in pure water at 25C as a function of pH.
Mg(OH)2 dissociates to form Mg(OH)+ and Mg2+ ions, whose activities in a saturated
solution are strongly pH dependent, as given by equations 9.73 and 9.74. The resulting
solubility of brucite is the sum of the concentrations of the Mg-bearing ions in the
solution and is indicated by the dashed line. The coordinates of point P on the diagram
represent conditions that will cause brucite to precipitate. The straight lines representing
activities of Mg2+ and M g(OH)+ have been arbitrarily terminated at unit activities
because in nature other Mg compounds will precipi tate at elevated concentrations of
Mg ions.
enough to m aintain the equilibrium line. We
equilibrium . T herefore, conclude, therefore, that
the solubility of brucite in brucite will precipitate in
w ater can be used to all environm ents that lie
define its stability as a to the right of the
solid phase in a geological solubility limits in Figure
environm ent. 9.2. In
other words, the activities of the
C onsider point P in M g(O H ) 2 constrain its stability
Figure 9.2 w hose coordi w ater containing M g2+
nates are [Mg2+] = 10 2 m and M g(O H ) + .
o l/L and pH = 10. The
coordinates of P do not
satisfy equation 9.74, 9.5 pH
which implies that brucite Control
cannot be in equilibri um
with a solution of that com of
position because, at Dissoci
this value of [Mg2+],
high. The application
ation
to equations 9.59 and 9.60 Equilibr
indicates th at the pres ia
ence of excess O H will
result in the precipitation The dissociation of weak
of solid M g(O H ) 2 until acids and bases is con
both [Mg2+] and [O H - ] trolled by the pH of the
have been reduced geochem ical environ m
sufficiently to place point ent. In order to exam ine
P on this concept we return to
the dissociation of H 2
S(aq) represented by
124 ACIDS AND BASES
equations 9.30 and 9.31. H ow ever, here we rew rite the rro m m e mass oaianee:
m ass-action equations (9.32 and 9.33) as follows: (H 2 S) + (H S - ) + (S2-) = 1 X 10- 2 m o l/L
PH
Figure 9.3 The pH dependence of the abundances of S-bearing ions and molecules in a solution of H2S(aq)
containing a fixed amount of S in solution (equation 9.79). Note that H2S is dominant at pH < 7.0, HS" domi nates
between pH = 7.0 and 12.9, and that S2_ is dominant only at pH > 12.9. However, all of the ions are pre sent at all pH
values even though their abundances may be small compared to that of the dominant ion.
In a very similar m anner, settles out slowly as a
silicon dioxide reacts with gelatinous precipitate (Iler.
w ater to form silicic acid, 1979). Given sufficient
such that at equilibrium: time, the precipitate expels
w ater and begins to
S i0 2 (am orph.) + 2 H 20
crystallize, forming opal A
and opal C T as interm
ediate phases (K astner et
S i0 2 is a solid com pound al., 1977).The process
that can either be am or ends with the
phous or assum e several crystallization of a
different polym orphic cryptocrystalline variety
crystalline forms, of quartz called chal
including quartz, cedony, which forms chert
cristobalite, and tridymite. or flint and occurs in geo-
Silicic acid actually forms des as agate. Q uartz is
as a result of chemical w highly insoluble, and
eathering of the com m on dissolves in w ater only
rock-forming sparingly, even during
aluminosilicate minerals long periods of geologic
such as the feldspars and time. Am orphous silica,
micas rather than by howev er, is much m ore
solution of crys talline or reactive and generally
am orphous S i0 2. maintains equilibrium
However, when its solu with silicic acid.
bility is exceeded, it forms
am orphous silica, which We are now in a
position to discuss the
solu bility of am
orphous silica based on
equation 9.84 and on
the dissociation T the
silicic acid:
126 ACIDS AND BASES
H4Si04 ^ H 3 S i0 4 +H + (9.85)
K x = 10 -9.71 of the concentrations of the Si-bearing ions in solu
H 3 S i0 4" ^ H 2 S iO ^ + H + K 2 = 10
1128
(9.86) tion. The first dissociation step (equation 9.85)
m akes the largest contribution to the increase in
9S6
H* 2,S^ i0 24- ^ HSiO] + H + ,v3K -=. u10 (9.87) the solubility of S i0 2 (am orphous) with increasing pH .T
he other ions come into play only in extrem e ly basic
H SiO 3- ^ SiO4 + H + K 4 = IO 1 3 1 0 (9.88) solution.
The solubility of am orphous silica as a func
W hen am orphous silica and p u re w ater have reached tion of pH is shown graphically in Figure 9.4.
equilibrium at 25 C (equation 9.84), the activity of silicic N ote that the concentration is expressed in m il
acid is given by the Law of Mass A ction: ligrams per liter (m g/L ) of SiOz because it is
com m only stated this way in chemical analyses of w ater.
The S i0 2 is not actually present in m olecu lar form but
[H 4 S i0 4] = IO 2 7 4 m o l/L (9.89) represents H 4 S i0 4 to which it can be related by equation
9.84. Therefore, the concen tration of S i0 2 stated in
The equilibrium (equation 9.84) is independent of the pH;
milligrams per liter can be recalculated as H 4 S i0 4 in
therefore, the activity of [H 4 Si 4] in a sat urated
moles per liter by the
solution of silicic acid in contact w ith solid am orphous S
following consideration. If S i0 2 = 25.0 m g/L , its
i0 2 is constant and varies only with tem perature. The
m olar concentration is 25.0/60.08 m m ol/L or
dissociation constants of silicic acid are all very small,
0.416 X IO- 3 m ol/L , w here 60.08 is the m olecular
making this a very weak acid. We can calculate the degree
weight of S i0 2. E quation 9.84 indicates that for
of dissociation
every m ole of am orphous S i0 2 that dissolves
of silicic acid at pH = 7.0 by proceeding stepwise
1 m ol of H 4 S i0 4 is produced. Therefore, since
from equation 9.85: 3
0.416 X 10 mol of S i0 2 dissolved to give a con
[H +][H 3 SiQ4 ] IO 9 7 1 (9.90) centration of 25.0 m g/L , the concentration of
=
H 4 S i0 4 in that solution is also 0.416 X 10- 3 or
[H4 S i0 4]
4.16 X 10~ 4 m ol/L .
[H 3 S i0 4] = ic r 971 x n r 2-74
The concentration of SiO z in Figure 9.4 is
io- p lotted on a linear rath er than a logarithm ic scale
= IO" 5 4 5 m o l/L (9.91) in o rd er to em phasize th at the solubility of am or
phous S i0 2 at 25 C increases steeply at pH > 8 because of
1(T5-45 X 102
the dissociation of H 4 S i0 4 to H 3 S i0 4 , which introduces
2 -7 4
IO additional Si-bearing ions into the w ater. T herefore, even
= 0.19% (9.92) a small decrease in the pH of basic solutions containing H
4 S i0 4 may reduce the solubility of S i0 2 enough to
Therefore, the activity of H 3 S i0 4 at pH = 7.0 is
saturate the solution and to cause am orphous S i0 2 to be
about 500 tim es less than that of H 4 Si 4 and does deposited . For exam ple, a decrease of 0.1 pH units from
not contribute appreciably to the solubility of 8.5 to 8.4 of a saturated solution can result in the
am orphous S i0 2. deposition of 1.37 mg of am orphous SiO z p er liter of
A t pH = 8.0 the activity of H 3 S i 0 4 increas solution. Such seem ingly sm all am ounts quickly increase
es to 10 4 4 5 m o l/L (equation 9.91) and D x rises w hen we apply them to th e large volum e of groundw ater
to 1.95% . A t pH = 9.0 [H 3 S i 0 4] increases to that m ay flow through an aquifer in the course of geologic
10- 3 4 5 m o l/L and D x reaches 19.5%. We see that the time. Subtle changes in the geochem ical environm ent can
degree of dissociation of silicic acid is very low at pH < be m agnified by tim e into large-scale tran s form ations.
8.0 and rises rapidly as the pH increases above th at value.
A s a result, the solubility of am or phous silica also rises
because it is based on the sum
9.7 SUMMARY 127
per liter by converting first to micromoles of Si and then
to m icrograms of SiO,. Fof average river water:
form ulate the Law of Mass A ction, which was later bined effect of all chemical reactions that are tak ing place
derived from the principles of therm odynam ics. simultaneously.
Chem ical reactions in the natural environ Som e bases are insoluble and actually occur as m
m ent may fail to reach equilibrium because the inerals in nature. For exam ple, M g(O H ) 2 form s the m
products escape from the site of the reaction or because ineral brucite. The solubility of such insoluble bases can
reactions involving the transform ation of one solid into be calculated by solving a set of sim ultaneous equations
another are very slow. In addition, som e natural reactions derived from the Law of Mass A ction and from the
are irreversible. requirem ent of con servation of m ass and electrical
Acids and bases are defined as com pounds th at charge.
release H + or O H - ions in aqueous solutions, W eak acids play an im portant role in geo
respectively. The bases of som e elem ents are am photeric chem istry because the abundances of the ions
and act like bases in acidic environ m ents and like acids they form by dissociation are controlled by the pH
in basic environm ents. W eak acids and bases dissociate of the environm ent. This phenom enon affects the
only partially depend ing on the p H of the environm ent. solubility of am orphous silica, which increases
In general, the p H of natural environm ents results from rapidly at p H > 8 because of the dissociation of
the com silicic acid.
I LE R
Problems
1. If the concentration of an ion in a solution is 5.0 X 10-2 bottle. Assay: 98.0%; net weight: 4.08 kg; molecular
m ol/L at 25 C, what is its concentration in the same weight: 98.08; specific gravity: 1.84. (Answer 36.8)
solution at 45 C? Use the data in Table 9.1.
5. What volume of 35 N (normal) H2S 0 4 is required
(Answer: 4.96 x 10-2 mol/L)
to make 250 mL of 1.5 N H2S 0 4?
2. Derive a relationship between the molarity (M), the
6. Calculate the pH of hydrofluoric acid containing
formality (F), and the density (d) of a solution.
0.1 mol of HF per liter of solution. Find the dissociation
3. Calculate the normality of hydrochloric acid (HC1) constant in Table 9.3.
from the following information provided by
7. Calculate the activities of all ions and the pH of a
the label on the bottle. Assay: 37%; net weight: 2.72 kg;
solution containing 0.1 mol of phosphoric acid per liter of
molecular weight: 36.46; specific gravity: 1.18.
(Answer: 12.0) solution. Find the equilibrium constants in Table 9.3.
8. Calculate the solubility of gibbsite (Al(OH)3) at pH
4. Calculate the normality of sulfuric acid (H2S 0 4)
= 5.0. Find the dissociation constants in Table 9.3.
based on the information provided by the label on the
References
C R E S S M A N , E. R., 1962. , R. K., 1979. The
Nondetrital siliceous Chemistry o f Silica.
sediments. Wiley, New York,
Chapter T, 1-23. In M. 866 pp.
Fleischer (Ed.), Data o f K S T N E R , M.. J. B. K EEN
Geochemistry, U.S. Geol. E , and J.
M. G IESKES ,
Surv. Paper 440-T.
1977.
D R E V E R , J. I., 1988. The
Diagenesis of siliceous
Geochemistry o f Natural
oozes I. Chemical
Waters, 2nd ed. Prentice-
controls on the rate of opal
Hall, Upper Saddle River,
A and opal CT
NJ, 435 pp.
transformation: an experi
G A R R E L S , R. M ., and C. L . C H R
mental study. Geochim.
IS T , 1965. Minerals,
Cosmochim. Acta, 41:1041-
Solutions and Equilibria.
1060.
Harper & Row, New York,
450 pp.
REFERENCES 129
K R A U S K O P F , K . B., 1979. Introduction to Geochemistry, 2nd ed. W E A S T , R. C., M. J. A S T L E , and W. H. B E Y E R (Eds.), 1986. CRC
McGraw-Hill, New York, 617 pp. Handbook o f Chemistry and Physics, 66th ed. CRC Press, Boca
L IN D S A Y , W. L ., 1979. Chemical Equilibria in Soils. Wiley, New Raton, FL.
York, 449 pp. W EDE POHL , K. H. (Ed,), 1972. Silicon. Handbook of
M O O R E , W. J., 1955. Physical Chemistry, 2nd ed. Prentice-Hall, Geochemistry, vol. II-2. Springer-Verlag, Berlin.
Upper Saddle River, NJ, 633 pp.
ST U M M , W., a n d J. J. M O R G A N , 1970. Aquatic Chemistry. An
Introduction Emphasizing Chemical Equilibria in Natural
Waters. Wiley, New York, 583 pp.
10
Salts and Their Ions
W hen an acid and a base are m ixed, the hydro gen ion of sulfuric acid and alum inum hydroxide, is quite soluble in
the acid com bines with the hydroxyl ion of the base to w ater and dissociates readily into ions:
form w ater. The rem aining anion of the acid and the
A l2 (S 0 4 )3 ( s ) ^ 2 A l3+ + 3 SO 2- (10.1)
cation of the base form a salt, which may precipitate or
rem ain in solu tion depending on its solubility. Salts are We assum e for the tim e being that these ions do
nam ed n o t interact with w ater, that this reaction is the
after the acids th at provided the anions. For
only source of A l3+ and S 0 4_ ions, that no other
exam ple, hydrochloric acid form s chlorides, sul
furic acid form s sulfates, carbonic acid form s car
ions are present, and that the w ater tem perature
bonates, and silicic acid form s silicates. The naturally is 25 C. We see from equation 10.1 that each
occurring salts of these and other acids are m inerals th a t m ole of alum inum sulfate that dissolves p ro
form the rocks of the crust of the E arth . The only m duces two moles of A l3+ and three m oles of S 0 4_ in
inerals th at are not salts are the oxides, hydroxides, and the solution. Therefore, if x is the num ber of m oles of
native elements. salt that dissolve in a saturated solution, the
Evidently, the study of salts and their properties concentrations of the resulting ions will be
is an im portant subject in geochem istry. In this (A l3+) = 2x and ( S 0 4~) = 3x. A ccording to the
chapter we will study the solubility of salts in Law of Mass A ction (equation 9.13):
w ater and the interactions of th e resulting ions [A13 +]2 [S 0 4 - ] 3 = K sp (10.2)
w ith m olecules of water.
w here K sp is the solubility product constant, and
the activity of excess solid alum inum sulfate is equal to
10.1 Solubility of Salts one. If the concentrations of the ions
W hen salts dissolve in w ater, they dissociate into the are equal to the activities in the solution,
anions and cations of the acid and the base from which [Al3+] = 2x and [SO2 -] = 3x. Substituting into
the equation 1 0 . 2 we obtain:
they w ere derived. N eutral m ole cules of salts m ay also
(2x)2(3 x f = K sp (10.3)
form , but their abundance is low in m ost cases. W hen a
salt has form ed a sat urated solution, a state of w here K sp = 69.19 (calculated from its solubility
equilibrium exists betw een the ions in the solution and
given by W east et al., 1986).The solubility product
any excess salt rem aining in the solid state. Therefore, the
constant is a large num ber, indicating th at alu
Law of Mass A ction applies and can be used to calculate
m inum sulfate is, in fact, a very soluble salt. For
the activities of the ions in a saturated solution and hence
this reason, it does not occur naturally as a m iner al.
the solubility of the salt.
Equation 10.3 reduces to 108x5 = 69.19, which yields x =
We choose alum inum sulfate [A12 (S 0 4)3] to 0.9147. Therefore, a saturated solution of alum inum
illustrate how the Law of M ass A ction can be used for sulfate contains 2x = 1.829 m o l/L of A l3+ and 3x =
2.744 m ol/L of SO;;- . The A l3+ reacts with w ater to
this purpose. A lum inum sulfate, a salt of
form insoluble A l hydroxide.
130
10.1 SO LU BILITY OF SALTS 131
The am ount of alum inum sulfate that dissolves in a constant, the solution is undersaturated and th ere fore can
saturated solution is equal to x = 0.9147 m ol/L . Since the dissolve the m ineral. If IA P = K , the solution is saturated
m olecular weight of aluminum sulfate is and a state of equilibrium exists betw een the solid and its
342.1478, its solubility is 0.9147 X 342.1478 = ions.
3.13 x 102 g /L of water. In the case we are considering, it turns out
M ost m inerals are m uch less soluble in w ater than that C a S 0 4 does not actually precipitate from a
alum inum sulfate, and their solubility p ro d uct constants supersaturated solution of its ions. Instead, g yp
have correspondingly sm aller val ues. Table 10.1 contains sum (C a S 0 4 2H 2 0 ) form s and subsequently
the solubility product constants of som e com m on m crystallizes to anhydrite. This is another exam ple
inerals taken p ri of the irreversibility of som e geochem ical reac
m arily from a com pilation by K rauskopf (1979) tions we first m entioned in Section 9.6. In this
but augm ented with data from Lindsay (1979). case, C a S 0 4 dissolves in w ater but does no t p re
These constants can be used to calculate the activ ities of cipitate at 25 C. Similarly, m ost silicate m inerals
the ions in saturated solutions of m inerals by m eans of dissolve in w ater b u t do n o t precipitate from
the Law of Mass Action. aqueous solutions o f their ions at low tem p era
We are also interested in finding out w hether a ture. Som e silifcate m inerals precipitate from
solution is actually saturated with respect to a specific m aqueous solutions only at elevated tem peratures or require
ineral. For example, if a sample of w ater contains 5.00 X a specific starting m aterial like volcanic glass, and the
10- 2 m o l/L of Ca2+ and 7.00 x 10 - 3 m o l/L of S 0 4~, we silicate m inerals of igneous rocks crystallize only from
may w ant to know w hether this solution is saturated with silicate m elts at tem p era tures close to 1000 C.
respect to calcium sulfate (anhydrite). In order to find out,
we first write the equation to represent the disso ciation of In the case of the calcium sulfate solution we are
C a S 0 4 into its ions: considering, gypsum precipitates until the activities of the
ions are reduced sufficiently so that the IA P is equal to K
sp and solid gypsum is in equilibrium with its ions:
C a S 0 4 ^ C a 2+ + SO2- (10.4)
If the solution is saturated with respect to anhy drite, the its ions:
product of activities of the ions in the solution m ust equal [Ca2 +][S 0 2_] = IO- 4 6 (10.8)
10-4,5. We assume for the sake of argum ent th at the m
easured concentrations If x m oles of gypsum precipitate, the activities of the ions
are equal to the activities of the ions and th ere at equilibrium will be:
fore calculate the ion activity product (IA P):
5
[Ca2+] = 5.00 X IO- 2 - x (10.9)
(5.00 X 10- 2 )(7.00 X 10~3) = 35.00 X 10 3
p K*
Sulfides
a- e S a ip o 4 A1P04
Ag S 2H20
2S y H6K3A15(P 0 4)S -18H20
-A F - C aH P04 2HzO C aH
g e P04 a~Ca3(P 0 4)2 /3-
N
2S S Ca3(P 0 4)2
i
Ag2 2
S
S
B F
e P
i
S b
2
2 S
S
3
F S
e b
C
2 2
a S
S S
3 3
C a
d - S
S H n
g S
C S
o -K gS a
S H -
g Z
( 2 n
a S S
) -ZnS
M
C g P
o S h
S o
( M s
/ n p
3 S h
) a
M t
C n e
u S s
2
A
S M g
n
C 3
S
u P
2
S
a_^ 0
M
e0.9 o
5^ 4
S
a
2
-
F
e a
S -
N
F i
galena
stibnite
bismuthinitesphalerite
oldhamite wurtzite
greenockite
chalcocite berlinite
covellite variscite
pyrrhotite K-taranakite
troilite brushite
troilite monetite
pyrite
marcasite whitlockite
cinnabar (red)
metacinnabar (black)
[Hg2+][S2-]
(green)
(pink) alabandite
haurite
molybdenite
134 SALTS AND THEIR IONS
Table 10.1 (continued)
p tfab
p iC-b
Fluorides
AgF 100.0 (L)
BaF, 5.8 (K)
CaF2 fluorite 10.4 (K)
MgFz sellaite 8.2 (K)c
PbF, 7.5 (K)
SrF2 8.5 (K)
N atural w ater samples are m uch m ore com plex w here y 2 is the activity coefficient of Ca2+ and Ba2+
than the solution we have been discussing because they defined by equation 9.12. By substituting into equation
contain a m ixture of cations and anions and therefore may 10.18, setting y 2 = 1 .0 , and drop ping (H +) and (O H
precipitate several dif ferent salts. For example, a solution ), we obtain:
m ay be satu rated with respect to both gypsum and
barium sulfate (barite). In this case we have two sim ulta 2X10- + 2>O0^ = 2(son ( ia 2 1 )
[SO2 ]
neous equilibria with a com m on ion: [so2-]
If we stipulate that [S 0 4 ] = (SO 2 ), equation
C a S 0 4 2H 2 O ^ C a 2+ + SO 2 + 2H zO (10.13) 10 .21
reduces to:
B a S 0 4 ^ B a 2+ + SO2 (10.14) 10 -4.6 - 10.0 _ (10.22)
1 0
2. Strong acid + weak base. The bicarbonate ion ( H C 0 3) is itself an acid because it
3. W eak acid + strong base. can release H + into solution. H ow ever, it also
4. W eak acid + weak base. hydrolyzes:
138 SALTS AND THEIR IONS
HCOJ+H20^H 2C03 +OH Km (10.25) K2C03 in one liter of pure w ater at 25 C. In addi
2
tion, we specify th at C 0 gas does not escape from
By applying the Law of M ass A ction to reaction 10.24
the solution. Since i^ A 1 = 10- 6 , 3 5 and K A2 = 10- 1 0 3
we have: (Table 9.3) and K w = 10-14, we find that
[H C O j][O H - ] /CH] = 10- 3 7 and K m = 10 - 7 65. Since 7CH 1 is n ear
(10.26)
= Ku ly 9000 tim es larger than K m , the first reaction is m uch
[C O M m ore productive than the second. We th ere fore attem pt
and from equation 10.25: a stepwise solution. If x m oles of the carbonate ion
[h2co3][oh-] (10.27) hydrolyze in the first step, we have at equilibrium :
= K,H2
[H C O j] [CO3 -] = 0.1 - x
To find the value of the first hydrolysis constant (7fH1) we [H C O j] = [O H - ] =
replace [O H - ] in equation 10.26 by: Substituting into equation 10.26 yields:
[O H - ] = (10.28)
[H +]
= io -3-7 (10.35)
w here K w is the dissociation constant of water. 0.1
Therefore, from equation 10.26: If x 0 .1 , 0 . 1 x ~ 0 . 1 and, from equation
10.35, x = 10-235, which is only 4.5% of 0.1.
[H C O 3 K (10.29)
K, Therefore, the approxim ation that x 0.1 is jus
[ c o r ] [ n +] tified. The second hydrolysis step produces a neg ligible
Since the dissociation of carbonic acid proceeds stepwise: am ount of additional hydroxyl ion equal to [O H - ] = 10 - 7 6
5 m ol/L . The pH of a solution
Table 10 .3 List
of Values for Constants Used
Table 10.2 Chemical Composition of Water in the D ebye-H ckel Theory Expressed in
of the Mississippi River at New Orleans Equation 10.46
Concentration Temperature, C A B
0 0.4883 0.3241
Ion ppm m o l/k g
Ions a,
K+,C L ,B r ,r ,N O ; 3
OH , F,H S - , B rO J,IO ;,M n 04 3.5
N aLH C O j, H2POL HSOj, Hg2+, SO2 -, SeOf- , CrOj , HPO2 -, PO3- 4.0-4.5
Sr2+, Ba2+, Ra2+, Cd2+, Hg2+, S2 -, WO2- 5.0
Li+, Ca2+, Cu2+, Zn2+, Sn2+, Mn2+, Fe2+, Ni2+, Co2+ 6
Mg2+, Be2+ 8
H+, A l3+, Cr3+, REE3+ 9
Th4+,Zr4+,C e4+,Sn4+ 11
SOURCE: Garrels and Christ (1965).
coefficients for ions of different charges and electrolyte solutions can be attributed to crow d
hydrated radii shown in Figure 10.2. A third equa ing of the ions, which counteracts the interfer
tion developed by Davies yields reliable results up ence th at occurs at low er ionic strengths.
to / = 0.5: The interference of other ions in the solution
fl/2 with the ions of a salt causes th e solubility of the
- lo g y = A t 0. 2/ (10.47) salt to increase. For exam ple, the solubility of
1 + I 1' 2 C a S 0 4 in pure w ater at 25 C ( 7 = 1.0, K sp =
25
F urther extensions of these m ethods by Pitzer 10-4,5) is IO- 2 or 5.6 X 10- 3 m ol/L . H ow ever, in
(1973) enabled H arvie and W eare (1980) to calcu w ater of the Mississippi R iver ( / = 4.4 X 10- 3
late the solubilities of m inerals in brines having y 2 = 0.759) the concentration of Ca2+ in equilib
ionic strengths greater than 2 0 . rium with anhydrite is:
Activity coefficients for ions of different (C a2+) = [Ca2 10-2.25 7.4 X 10- 3 m o l/L
charges in solutions of varying ionic strengths, as 0.759 =
?2
determ ined from the Davies equation (10.47), are (10.48)
listed in Table 10.5. N ote that the values of the
which implies an increase of the solubility of
activity coefficients in Table 10.5 decrease with
increasing ionic strength of the solution and with anhydrite of 32.1% .
increasing charge of the ion. For example, we find This result illustrates the point that the Law
by interpolating in Table 10.5 that the activity of of M ass A ction yields inaccurate results unless the
N a+ in w ater of the Mississippi River concentrations of reactants and products of a
(/ = 4.4 X 10-3) is about 0.936, w hereas that of reaction at equilibrium are converted to activities
Ca2+ is 0.759, and that of L a3+ is only 0.543. In by m eans of activity coefficients in order to cor
other words, the presence of other ions reduces the rect them for the nonideality of electrolyte solu
effective concentration of N a+ in the Mississippi tions. Conversely, the results of calculations based
River by 6.4%, that of Ca2+ by 24.1%, and that of on the Law of Mass A ction are expressed in term s
La3+ by 45.7% . In general, activity coefficients of activities and m ust be converted to concentra
decrease with increasing ionic strength, but tions before they are interpreted . The solubility
increase again at high values of I and may becom e calculation of anhydrite in M ississippi R iver
larger than one (G arrels and Christ, 1965). The w ater indicates that the presence of other ions in
increase of the activity coefficient in highly saline a solution m ust not be ignored.
142 SALTS AND THEIR IONS
C a C 0 3 ^ C a 2+ + C O 2'
CO2 + H20^HC03 +
HC03 + H20^H2C0
10.4 SOLUBILITY OF CALCIUM CARBONATE 143
[H +][O H _] = [O H " ] 2
If the solution is open to We notice th at according
The sixth equation arises from the
to equation 10.56: Finally, from equation 10.57:
bonic acid is in equilibrium w ith
ide, which has a p artial for electrical neutrality: = [C P 2] = 3 [H +] = 10 140
atm osphere: [ h 2 c o 3] 32.2
2(C a2+) + (H +) = 2(C O ]-) + ( H C 0 3) + O H ") (10.58) [O H ]
H 2 C 0 3 ^ C 0 2 (g) + H 20
= 9.3
We now convert the
W hen calcite crystals are activities of the ions to
Evidently, the activity of
in equilibrium with their their corresponding
carbonic acid is fixed by
ions in pure w ater open to concentrations for
the partial pressure of C 0
the atm osphere at 25 C, substitution into equation
2. We can use this result in
seven ions and m olecules 10.58. In order to do so we
equation 10.55 to express [
are present: must know the ionic
H C 0 3] as a func tion of strength of the solution,
Ca2+, C O ]", H C O J, [O H - ]: which depends on the
H 2 C 0 3, C 0 2, H +, O
concentrations of the ions
H
[ it con tains. Strictly
H ow ever, the partial H speaking, we cannot
pressure of C O z is fixed 2 determ ine the ionic
by strength w ithout first com
c
the com position of pleting the calcu lation. H
o
soil gases may contain up to ten ow ever, we cannot com
3
C02 as the open atm osphere. plete the calcula tion w
]
need only six independent ithout knowing the ionic
[
equations in order to strength. We deal w ith this
O
specify the concentrations Catch-22 situation by
H
of all ions and m ole cules iteration. T hat is, we
-]
in a saturated solution of assum e that all activity
[H C O 3 ] =
calcite. We obtain
(9.3 coefficients are equal
five equations by applying
A ction to equations 10.49 through (10.60)
the dissociation equilibrium io
5
equation 10.49:
=
[Ca2 +][C O ]1 = 102 6 1
8[O
From equations 10.50 and 10.51: H ]
[hco3][oh-] From equation 10.54 we
[C O ]-] ~~ derive a relationship betw
een [C O ]- ] and [O H - ]:
[ h 2 c o 3 ][o h ~ ] [C O ]] = [ h c o 3 ] [ o h ]
= 1 0 - 37
[H C O 3]
= 106 3 1 8 [O H " ] 2
to one, m ake the calculation, and then use the results to We can now calculate the activities of the
determ ine the ionic strength and hence the activity other ions in the solution by substituting
66
coefficents from the D avies equation (10.47) or from [O H - ] = 10- 5 into equations 10.60-10.63. The
Table 10.5. We then return to equation 10.58, convert the results are:
activities of all ions to the appropriate concentrations, and [H C O - ] = 102 6 1 8 [O H - ] = IO2 6 1 8
solve the problem a second time. If necessary, the process X 10 -5.66
can be repeated several tim es to achieve a desired level of = io -3042 = 9.08 X 10- 4 m o l/L
accuracy. [C O |- ] = io 6 -3 1 8 [ o h - ] 2 = IO6 3 1 8 x io -11-32
T herefore, we now proceed w ith the solu tion of the = IO- 5 0 0 2 = 9.95 X 10- 6 m o l/L
problem by substituting equations 10.60, 10.61, 10.62, and IQ -14 .668 14.668
10.63 into equation 10.58, assum ing th at all activity
[Ca2 [O H - ] 2 10- 1 L 3 2 = 10-3-348
coefficients are equal to one:
4.48 X 10- 4 m o l/L
2
x i c r 14-668 io -140 , , , 1R. r ,, [JJ+] =
+ r Ti = (2 X 10 )[O H _] 10-14- IO- 1 4 0 - 8 .3 4
[O H - ] 10
+ 102 6 1 8 [O H - ] + [O H ]
We simplify equation 10.64 by elim inating the We check the correctness of these results by cal culating
the IA P of calcite in this solution:
term for (H +) because the solution will be basic
and by com bining 102 6 IS[O H _] + [ O H ] as IA P = (9.95 X 10-6) X (4.48 X 10-4) = 10- 8 3 5
102 6 1 9 [ o h - ]. N ext, we clear the fractions by m ul ( 10.6 8 )
tiplying all rem aining term s by [O H - ] 2 and collect
term s: which is identical to for calcite and therefore
106 6 1 9 [O H - ] 4 + 102 6 1 9 [O H - ] 3 = 10- 1 4 3 6 5(10.65) confirms the correctness of the calculation. The ionic
619
strength of the solution is I = 1.4 X 10-3, and the activity
D ividing each term by 106 yields: coefficients of the ions from
]3 Table 10.5 are y 1 = 0.966 and y 2 = 0.857.
[O H - ] 4 + 10- 4 0 0 0 [O H - = IO- 2 0 9 8 5 (10.66)
R eturning to equations 10.60-10.63, we now
convert the activities to concentrations:
W e solve this equation by trial and erro r starting w ith [O 1 0 2 6 1 8 [ o h - ] _ 102 6 1 8 [O H - ]
N ote that equation 10.52 is the sum of 10.71 and 10.72 to form a stalactite hanging from the roof of the cavern. W
and th at m olecular C 0 2 dissolved in w ater is thereby hen the rate of flow of groundw ater is so rapid that the
elim inated from consideration. We can use these solution does not reach equilib rium on the roof of the
equations to predict qualitatively how the solubility of cavern, additional calcite m ay precipitate after the w ater
calcite is affected by changes in the partial pressure of C 0 has dripped onto the floor of the cavern, thereby forming a
2. A n increase in the partial pressure of C O z at constant stalag mite. Ultim ately, stalactites and stalagm ites may
tem perature increases th e concentration of carbonic acid join to becom e pillars or curtains of calcite.
in the solution. Consequently, m ore calcite dissolves by
equation 10.70 in accordance with Le C hteliers The effect of the partial pressure of C O , on the
principle. Similarly, a decrease in the partial pres sure of solubility of calcite is shown quantitatively in Figure 10.3.
CO z causes a saturated solution of calcite to becom e H ere we see that the concentration of Ca2+ in a saturated
supersaturated and results in the p re cipitation of calcite solution of calcite at 25 C increases from about 20 m g /L
until equilibrium is restored. to about 45 m g /L w hen the partial pressure of C 0 2
increases from 3 X 10~ 4 atm in the air to 3 X IO- 3 atm in
The partial pressure of C 0 2 at a site on the surface of soil gases. Consequently, each liter of w ater appearing on
the E arth m ay decrease as a result of photosynthesis of the roof of a cavern could, under ideal condi tions, deposit
aquatic plants, which has the effect of com bining C 0 2 and 25 mg of calcite. The am ount of cal cite th at is actually
H20 to produce glu cose (C 6 H 1 2 0 6) and 0 2 with the help deposited depends on the partial pressure of C 0 2 in the
of ultraviolet radiation . Consequently, com m unities of soil above a given cavern, on the tem perature, and on the
aquatic plants can low er the concentration of C 0 2 (aq) in establish m ent of equilibrium betw een calcite and the
the w ater during daylight hours. This may cause calcite to groundw ater before the w ater reaches the cavern. N
precipitate if the w ater becom es super saturated with evertheless, stalactites and stalagmites can grow
respect to calcite. C ertain colonial algae are capable of
precipitating calcite by this m echanism , thereby form ing
thinly lam inated cal cite m ounds called strom atolites (W
alter, 1977).
We now examine the equation we have con structed to tially valuable m ineral deposit and, in that sense, is not
learn by the application of Le C hteliers principle how desirable from our point of view.
the reaction at equilibri um responds to changes in certain The probability that a chemical reaction in nature,
environm ental parameters. We recall that reactions in such as equation 10.78, can achieve equi librium depends,
nature may not achieve equilibrium but run to completion am ong other things, on the w ater/rock ratio. If there is an
if one of the products escapes from the system. In the case abundance of w ater
we are considering, this could occur if K + and H 4 S i0 4 com pared to the volum e of rock, reactions are
are rem oved by the m ovem ent of the water in which the likely to run to com pletion because the environ
reaction is taking place. Therefore, the conversion of K m ent controls the reaction. If, on the other hand, the w
-feldspar to kaolinite is favored by the m ovem ent of ater/rock ratio is small, the reactions control the environm
groundw ater or surface water in response to a hydraulic or ent and equilibrium is possible, p ro vided reaction rates
topographic gradient. We also note that nine moles of w are adequate. Therefore, reac tions taking place in large
ater are consumed for each mole of kaolinite that is bodies of water and in aquifers with high percolation rates
produced. Therefore, the form ation of kaolinite is favored having large w ater/rock ratios are m ore likely to run to
by an abun dance of water, which, in addition, must be com pletion than to achieve equilibrium , whereas reac
acidic to provide the H + needed by the reaction. We con tions in small volumes of pore w ater (small w ater/rock
clude, therefore, that K -feldspar is converted into kaolinite ratios) may achieve a localized state of equilibrium betw
in places where an abundance of acidified w ater is een ions and their solids. Such restricted environm ents
available and where the w ater flows at a sufficient rate to occur in the pore spaces of rocks in which m ineral cem
rem ove the soluble products of the reaction. Also, since ents m ay be deposited by reactions that m aintain a state
reaction rates generally double for every 10 C increase in of equilibrium .
tem perature, the form a tion of kaolinite is likely to be
favored by tropical climatic conditions near the equator of The Law of Mass A ction enables us to deter mine the
the Earth. activities of ions and molecules required for equilibrium in
a chemical reaction represented by a balanced equation.
Therefore, we return to equation 10.78 and place that
The reaction betw een K -feldspar and water involves reaction at equilibri um. By applying the Law of Mass A
the hydrolysis of a salt derived from a weak acid and ction we obtain:
strong base. We predict that such a reaction should rem ove
[H 4 SiQ 4 ]4 [K + ] 2
H + from the solution and increase its pH .This is indeed K (10.79)
the case here because H + is consum ed by the reaction. [H + ] 2
However, the pre dicted increase of the pH can only be We linearize this equation by taking logarithms:
dem onstrated in the laboratory because natural waters are
4 log[H 4 S i0 4] + 2 log [K+] - 2 log [H +]
buffered (Wollast, 1967).Therefore, when this reac tion (10.80)
occurs in nature, the H + originates from the environm ent = log K
by the dissociation of carbonic or hydrosulfuric acid, or In order to reduce the num ber of variables to two, we
some other reaction. If the supply of H + is large and if the
com bine K + and H + as the [K +]/[H +] ratio:
soluble products are rem oved, the reaction can run to
completion. If, however, the w ater is basic and the soluble 4 log [H4S iO j + 2 log | | J | = log K (10.81)
products rem ain at the site, the reaction can achieve We rewrite this equation as:
equilib rium . In that case, the conversion of K -feldspar to
kaolinite stops and the reaction becom es unpro ductive. [K 5 log K -2 log[H 4 S i0 4] (10.82)
log =
Therefore, the establishm ent of chemical equilibrium may
p reven t the form ation of a poten [H +]
which is the equation of a straight line in coordi nates of
log [K+]/[H +] and log [H 4 S i0 4] having a
10.6 TRANSFORMATION OF POTASSIUM FELDSPAR TO KAOLINITE 151
and vice versa, in the environ m ents containing
slope m = - 2 and an presence of water at will proceed in the less H 4 S i0 4 than is
25C.The line labeled direction th at moves it required for equilibrium of
intercept of b = ) log K . It
10.82 represents
is the locus of all points w closer to a state of reaction 10.78.
conditions at which K-
hose coordinates are the equilibrium . If the w
feldspar and kaolin ite
logarithm s of the [K+]/[H ater/ro ck ratio is small, The activity of silicic
coexist in equilibrium.
+] ratio and of [H 4 S i0 4] Point P on the line was the environm ental param acid is lim ited by its sol
required for equilibrium cho sen arbitrarily to help eters can change until they ubility, as shown in Figure
of reaction 10.78. Thus, determine the direction of satisfy equation 10.82 and 9.4. T herefore, am or
any point P on the line in reaction 10.78 when [H4S equilibrium is established. phous silica precipitates w
Figure 10.5 (equation i0 4] is increased or If the w ater/ro ck ratio is hen its solubility limit is
decreased. Line 10.83 reached and the activity of
10.82) represents an large, the environm ental
represents the solubility of [H 4 S i0 4] in Figure 10.5
environm ent in which K- param eters m ay not be
amorphous silica, which cannot rise above the value
feldspar and kaolinite limits the activity of [H4S affected appreciably by
coexist in equilibrium required for equilibri
i0 4] to values less than the reaction we are now
um in the reaction:
with the ions and m IO2 74 m ol/L, as shown consider ing. The reaction
S i0 2 (am orph .) + 2 H 2 0 ^ H 4 S i0 4
olecules in equation in equation 9.89. (Line therefore runs to com
10.78. Further, 10.82 was plotted from pletion A t equilibrium , [H 4 S i0 4] = 10 2 7 4 m
any point th at is n o t on equation 10.16 of Garrels
and Christ, 1965.) tion 9.89). Therefore, a line has b
environm ent in which K -feldspar and does n o t achieve equilibrium .
Figure 10.5 at log [H 4 S i0 4] = -2 .7
are n o t at equilibrium . In The straight line we have draw n in
im penetrable boundary
10.5 (equation 10.82) is a boundary th a t separates
th e because [H4 S i0 4]
environm ents in which K -feldspar
duced by reaction 10.78 cannot rise to higher
from those in which values. The two lines in
kaolinite is produced . We Figure 10.5, therefore,
assum e h ere th at the
divide the com
reaction represented by
equation 10.78 is, in fact, positional plane into
reversible. A lthough this three fields th at
m ay n o t be tru e in all represent environm
cases, a K -feldspar know ental condi tions
n as adularia does form
needed to stabilize K
from kaolinite at low tem
p eratu re (M ensing and -feldspar, kaolinite, and
Faure, 1983; K astner and am orphous silica.
Siever, 1979). In any
case, if we m ove from
point P on the equilibrium
cr
1 .5 17
References
B U T L E R , J. N., 1982. Carbon HARDlE,L.A.,andH.REuGST
Dioxide Equilibria and Their ER,1970.Theevolutionofclosed
Applications. Addison- -
Wesley, Reading, MA, 259 basin brines. Mineral. Soc.
pp. Amer. Spec. Pub. 3,273-290.
G A R R E L S , R. M., and C . L. C H H A R V IE , C. E., and J. H . W E A R E ,
R IS T , 1965. Solutions, 1980. The prediction of
Minerals and Equilibria. min-eral solubilities in
Harper & Row, New York, natural waters: the Na-K-
450 pp. Mg-Ca-Cl-S04-
154 SALTS AND THEIR IONS
H20 system from zero to high concentration at 25 C. basement core from Scioto County, Ohio, USA. Contrib.
Geochim. Cosmochim. Acta, 4 4 :9 8 1 -9 9 7 . Mineral. Petrol., 82:327-333.
K s t n e r , M., and R. S i e v e r , 1979. Low temperature feldspars in P I T Z E R , K. S., 1973. Thermodynamics of electrolytes. I.
sedimentary rocks. Amer. J. Sei., 2 7 9 :4 3 5 -4 7 9 . Theoretical basis and general equations. J. Phys. Chem.,
K r a u s k o p f , K. B., 1979. Introduction to Geochemistry, 2nd ed. 77:268-277.
McGraw-Hill, New York, 617 pp. W A L T E R , M. R., 1977. Stromatolites. Elsevier, Amsterdam, 790 pp.
L IN D S A Y , W. L ., 1979. Chemical Equilibria in Soils. Wiley, New
Y o rk , 449 pp . W E A S T , R. C., M. J. A S T L E , and W. H. B E Y E R (Eds.), 1986. CRC
L IV IN S T O N E , D. A., 1963. Chemical composition of rivers and lakes. Handbook o f Chemistry and Physics, 66th ed. CRC Press, Boca
In M . Fleischer (Ed.), Data of Geochemistry, 6th ed. Raton, FL.
Ch. G., U.S. Geol. Surv. Prof. Paper 440-G, 64 pp. W O L L A S T , R., 1967. Kinetics of the alteration of K-feldspar in
M E N S I N G ,T. M ., and G. F A U R E , 1983. Identification and age of buffered solutions at low temperature. Geochim. Cosmochim.
neoformed Paleozoic feldspar (adularia) in a Precambrian Acta, 31:635-648.
11
Thermodynamics
The science of therm odynam ics originally evolved but cannot be proved or disproved. They m ust be
from a variety of observational evidence accepted as articles of faith. The science of th e r
indicating that m echanical energy is trans m odynam ics has evolved from these so-called
form able into heat. For example, Benjamin laws by careful deductive reasoning aided by
Thom pson (1753-1814) of W oburn, M assachu m athem atics.
setts, b etter known as C ount R um ford of the
H oly R om an Em pire, was im pressed by the
am ount of heat generated during the boring of
11.1 Definitions
cannon at the arsenal in Munich. In 1798 he pro posed th Therm odynam ics describes the world the way it appears
at the heat was form ed by conversion of the m echanical to an observer in term s of certain m ea surable properties
energy expended by the boring and actually calculated the of m atter, such as volume, pres sure, tem perature, and
am ount of heat gener ated by one horse turning the borer chemical com position. M atter is regarded as continuous
for one hour. C ount R um fords ideas about the and indivisible. In this regard therm odynam ics is the
equivalence of w ork and heat were supported by Sir H opposite of
um phry D avy (1778-1829), who dem onstrated in 1799 statistical m echanics in which m atter is treated as
that two pieces of ice could be m ade to melt by rubbing a collection of particles in m otion. In order to
them together by a clockwork m echa nism in a vacuum . en ter the world of therm odynam ics we m ust first becom
Nevertheless, the scientific com m unity did not accept e fam iliar with certain definitions and con ventions th at
this idea for nearly 50 years until James R Joule (1818- may differ from com m on usage.
1889), working in his fath ers brewery in M anchester, The basic entity for consideration in therm o
England, carefully m easured the rise in tem perature dynamics is the system, which is that p art of the universe
caused by stirring w ater with a mechanical paddle wheel. whose properties are under considera tion. The system is
H is research betw een 1840 and 1849 clear ly dem separated from the rest of the universe, know n as the
onstrated that the expenditure of a certain am ount of surroundings, by a b o u n d ary whose properties can be
work always produced the same am ount of heat (M oore, defined. A system is said to be open w hen m atter can
1955). pass across the boundary and closed w hen m atter is
prevented from entering or leaving the system. The bound
ary may prevent not only m atter but also heat and any
The observation that work is transform able into heat other form of energy from passing through it. Systems that
is an extension of the principle of con servation of energy. have such im pervious boundaries are said to be isolated.
It can be restated by saying th at w henever w ork is done, The investiga tor m ust identify the system to be studied
heat energy is con sum ed, implying that one m ust and specify the properties of its boundary.
always pay a price by the expenditure of energy if w ork
is to be accom plished.
The properties of a system are classified as being
Therm odynam ics is founded on three basic either intensive or extensive. Intensive properties, such as
principles, which arise from com m on experience pressure, tem perature, and
155
156 T H ER M O D Y N A M IC S
density, are independent of the am ount of m atter that is strained by specifying the value of one of the variables. W
present, w hereas the extensive properties, such as volum e hen the num ber of variables exceeds the num ber of
and mass, depend on the mass and are therefore additive. equations by two, the system is divariant, and so on.
W hen the properties of a system are changed, it
experiences a change in its state. For exam ple, w hen a W hen a chem ical system is at equilibrium , the num
quantity of gas in a cylinder is com pressed by m oving a ber of phases (p ), the num ber of com po nents (c), and
frictionless piston, the system undergoes a change in state. the num ber of degrees of freedom
(/) m ust satisfy the Gibbs phase rule (N ordstrom and M
The chem ical com position of therm odynam ic unoz, 1986):
system s is expressed in term s of com ponents and p+f=c +2 (1 1 . 1 )
phases. N ordstrom and M unoz (1986, p. 67)
defined a phase as a uniform , hom ogeneous, Therefore, systems in which the num ber of phases is
physically distinct, and m echanically separable equal to the num ber of com ponents have two degrees of
portion of a system . W hen the system under con freedom . In other words, pressure and tem perature m ust
sideration is a rock, the m inerals of which the rock is com be specified in this case in order to com pletely define the
posed are the phases. Phases need not be solids, but can state of the system .
also be liquids or gases. For exam ple, a system containing All systems contain a certain am ount of heat, which
ice, w ater, and w ater vapor has three phases (solid, liquid, is defined as that which flows across the boundary of a
and gas).The aqueous phase in such a system m ay also closed system during a change in its state by virtue of a
contain a num ber of ionic or m olecular species in difference in tem perature between the system and its
solution. surroundings. M ore over, heat always flows from a point
The com ponents of a system are the chemical of higher to a point of lower tem perature.
constituents by m eans of which the chemical Thermodynamics defines heat as though it w ere an
com positions of the phases of a system can be invisible fluid whose direction of flow is indicated by its
com pletely described. The choice of com ponents is algebraic sign. H eat is positive w hen it flows from the
arbitrary. For exam ple, the chem ical com posi tion of a sur roundings to the system and negative when it flows
rock specim en can be expressed in term s of oxides or in from the system to the surroundings. We can rem em ber
term s of chem ical elements. Flowever, the num ber of this im portant convention by recogniz ing that systems
com ponents required to describe the com positions of all are selfish. W hen a system gains heat, the added heat
of the phases p re sent in a system at equilibrium is has a positive sign, and when it loses heat the sign is
constrained by the G ibbs phase rule. negative.
The num ber of variables th at m ust be speci Work is defined in mechanics as the product of force
fied in order to define th e state of a system is times distance, whereas in thermodynamics work is
equal to the num ber of com ponents plus the defined as a quantity that flows across the boundary of a
num ber of intensive properties, such as pressure and tem system during a change in its state and is completely
perature . T he state of the system is com - pletely defined convertible into the lifting of a weight in the surroundings.
w hen the num ber of variables is W ork, like heat, is an algebraic quantity, but the sign
equal to the num ber of in dependent relations convention of work is opposite to that of heat. W hen
betw een them . If th e num ber of independent work flows to the surroundings it has a positive sign, and
equations is less than th e n um ber of variables, the system when it flows to the system it has a negative sign. A
has variance or degrees o f freedom . In this case, th e state nother way to visualize the sign convention for work is to
of th e system cannot be com plete ly constrained . W hen say that when the system does work, the work is positive,
th e num ber of variables exceeds the num ber of in but when the surroundings do it, the work is negative.
dependent equations by one, th e system is univariant and W ith these definitions and conventions we have now
has one degree of freedom . U nivariant system s can be set the stage to introduce the first law of
con
11.3 ENTHALPY 157
ut
int/ ayait/H i .
therm odynam ics. The presentation will be selec tive and
based on plausibility rather than m athe m atical rigor. M
ore com plete and rigorous presentations can be found in AE=qw (11.4)
college level text books on physical chemistry. Therm
odynam ics for The first law of therm odynam ics implies that
use by geologists is presented in books by K ern
and W eisbrod (1967), W ood and Fraser (1976), work and heat are related and th at some of the
Fraser (1977), N ordstrom and M unoz (1986), and A heat that flows into the system during a change in state is
nderson (1996). converted into w ork and therefore does not augm ent the
internal energy.
We can im agine th at a reversible change in state can
11.2 The First Law be carried out in small increm ental steps such th at the
internal energy changes by small am ounts designed by dE
Every system whose state has been defined by specifying . Now we can restate the first law in differential form:
the num erical values of its properties contains a certain
am ount of energy. This energy arises from the heat
d E = dq d w (11.5)
content of the system as well as from other internal
sources such as gravitation al, electrical, magnetic, m w here dq and dw are very small increm ents of heat and w
olecular, atomic, and even nuclear phenom ena. The am ork, respectively. The m ost com m on form of w ork a
ount of internal energy (E ) of a particular system is system can do during a change in its state is to expand
difficult to determ ine. However, its m agnitude depends against the constant pressure of the surroundings. Thus we
on the state of the system . W hen a system undergoes a can specify that:
change in its state in a reversible m anner, its internal
dw =PdV (11.6)
energy also changes. The m agnitude of the change in the
w here P is the external pressure and d V is an
internal energy AE of a system that undergoes a reversible
change in state from state 1 to state 2 is given by: increm ental change in the volum e of the system . A
simple dem onstration indicates that P d V does have the
dim ensions of w ork (force X distance):
AE=E2Ex (11.2) PdV= force X (distance) 3
= force X distance
w here E 2 and E x are the internal energies of the system (distance ) 2
in states 2 and 1, respectively. It makes no difference how (11.7)
the change in state is carried out, provided that it is Therefore, we can express the first law by the equation:
reversible, which requires that the system is never far
from being in equilibrium . If the system is returned to its
d E = dq - P d V (11.8)
original state, the net change in its internal energy is zero.
We con clude therefore th at the change in the internal
energy of a system undergoing a change in state along a
circular path is zero: 11.3 Enthalpy
The statem ent of the first law represented by equation 1 1 . 8
AE = (E 2 - E x) + (E x - E z) = 0 (11.3) enables us to predict the change in the internal energy of a
system during a change in its state. The equation indicates
The first law o f thermodynamics states that the th at the increase in the internal energy of the system is
increase in the internal energy of a system during a equal to the heat absorbed from the surroundings m inus
reversible change in its state is equal to the heat that the work done by the expansion of the system against the
flows across its boundary from the surround ings (q) pressure exerted by th e surroundings. We can
minus the work done by the system (w).
158 TH ER M O D Y N A M IC S
This is a very im Standards, and in com
carry out such a change in portant and concrete their compounds. The pilations by Kelley (1962),
result because qP is a m reference state is defined in Robie et al. (1978), B arin
state by integrating
easurable quantity. For terms of certain standard et al. (1989), and Woods
equation 1 1 . 8 from state 1 to
example, w hen a chem properties that define the so- and G arrels (1987).
state 2 : However, it is im portant
ical reaction is carried called standard state. The
out in a well-insulated to rem em ber that
tem perature of the standard
d E = J dq vessel called a calorim standard enthalpies of
The integration yields: state is equal to 25C
eter, the heat evolved or form ation and other therm
(298.15 K ),the pres sure is 1
consum ed is indicated odynam ic constants are
atm (recently redefined as 1
E 2 - E , = (q2 - q j - by the change in tem measure m ents and are
bar, which is equal to 10 6
perature inside the subject to experim ental
W e simplify this result by replacing dyne/cm 2, or 0.987 atm,
calorim eter. Therefore, errors. Consequently, the
which is an am ount of h eat added at 29.53 in. Hg), and the
equation 11.16 enables values reported by
activity of ions and
pressure. In addition, we m us to m easure the different investigators for
ultiply the P V terms: molecules in aqueous
change in the enthalpy of a given com pound
solution is equal to one. In
E 2 - Ej = q p - a system as a result of a commonly dif fer from
addition, we specify that the
P V 2 + P chem i cal reaction, b u t each other by am ounts
V 1 (11.11) enthalpies of the pure elem
we still do not know the that may or may not be
ents in their stable state of
By rearranging the term s of am ount of enthalpy the within the error of the m
aggregation in the standard
this equation we obtain: system contains. easurem ents. Such
state are equal to zero. Thus
discrepancies in therm
(E 2 + P V 2) - we have the definition: odynam ic constants affect
In order to establish
a scale for enthalpy we the results of calculations
(E, + EV ,) = qP
The enthalpy o f in which they are used and
(11.12) now define a reference
formation of cause internal
It is advantageous at this state for chemical elem
compounds and their inconsistencies. Some of
point to define a new kind ents and ions and molecules in the
of energy called the aqueous solution is
enthalpy (H ): the heat absorbed or
given off by chemical standard enthalpies
H =E +PV reactions in which the of form ation
The enthalpy is a function of the state of the sys compounds, ions, and listed in
tem , as isthe internal energy.Therefore, the molecules form from A ppendix B are average
the elements in the values of two or m ore
enthalpy of the system in states
standard state. determ inations that
tfj = E, + P V X
appear to be com patible
We represent this im portant
H2=E2+PV2 with in the errors of m
param eter by the sym bol
easurem ent. However, no
Substituting equations 11.14 and 11.15 into equa script identifies the
assur ance can be given th
standard state and the sub
tion 1 1
.1 2
yields: script/sym bolizes form ation at . these values are
The standard enthalpies of accurate to within the
H2- Hx=qP=
other therm odynam ic constants stated num ber of
In other words, the change pounds, ions, and molecules significant figures or that
in the enthalpy of a sys tem have been m easured the listed values are
(AH ) during a reversible experim entally and are internally consistent.
change in its state at tabulated in the CRC H
constant pressure is equal andbook o f Chemistry and
to the heat absorbed by the Physics (W east et al.,
system during th at change 1986), in Technical N otes
in state. of the National B ureau of
11.4 HEATS OF REACTION 159
4.1840 joules (J) so that 1 kcal/m ol is equal to
11.4 Heats of Reaction
4.1840 kJ/m ol. A t the present tim e both units are in use.
W hen two com pounds or elem ents, A and B, react
to form a product A 2 B3 by the reaction: W hen AHR of a reaction is positive, the reac tion is
2 A + 3 B A 2 B3 (11.17) endothermic, which m eans that h eat flows from the
surroundings to the system . Conversely, w hen AHR is
a certain am ount of heat is either used up or given off. We negative, the reaction is exotherm ic and heat flows from
can now calculate the am ount of this heat (.q P) from the the system to the surround
enthalpy change of the system when the reaction takes ings. E ndotherm ic reactions consum e heat,
place in the standard state. In other words, we treat the w hereas exotherm ic reactions produce heat.
reaction as a change in state and calculate the difference in In order to illustrate how heats of reaction
the enthalpy (AHr ) betw een the final state (A 2 B3) and are calculated, we choose th e reaction betw een
the ini tial state (2 A + 3 B). Therefore, the change in hydrogen and oxygen to form w ater in the
enthalpy of the reaction in the standard state is: gaseous state, sym bolized by (g):
2 H 2 (g) + 0 2 (g) - 2 H zO(g) (11.20)
AHR = Hf (A 2 B3) - [2Hf (A ) + 3 /T /B )] (11.18)
The am ount of h eat absorbed or given off w hen this
reaction occurs at 1 atm and 25 C is given by:
where AHR is the heat of the reaction in the stan dard AHR = 2 //} (H 2 0 (g ))
state and Hf is the standard enthalpy of for m ation of A 2
- [2H/ (H 2 (g)) + Hf ( 0 2(g))}
B 3, A, and B, respectively. Note that the heat is evolved
or absorbed as a result of, or during, the reaction and not The standard enthalpies of H 2 and 0 2 are both
when the reaction is at equilibrium . A fter a reaction has equal to zero by definition, so th at in this case:
reached equilibri um, the heat of the reaction (AH R) is
AHR = 2 //} (H 2 0 (g )) (11.22)
zero because the rates of the forward and backward
reactions are exactly equal. Thus, we can state that at The standard enthalpy of form ation of w ater vapor is
equilibrium AH R = 0. However, the reference state is a 57.80 kcal/m ol (A ppendix B ).Therefore:
defined condition in which equilibrium does not occur. AHR = 2 ( -5 7 .8 0 ) = -1 1 5 .6 kcal (11.23)
Therefore AHR is never zero. In general, AHR is
We conclude that this reaction is strongly exother mic. In
calculated by summing the standard enthalpies of the
fact, this reaction is com m only used to dem onstrate the
products and by subtracting the enthalpies of the reactants:
explosive nature of som e chem i cal reactions. N ote that
the reaction is endotherm ic w hen equation 1 1 . 2 0 is
reversed.
H eat effects are also associated with the dis sociation
AHR = 2 niHfi ( p r o d u c t s ) w , / / ) , (reactants)
of salts into ions, w ith phase transform a tions, and with
(11.19) mixing or dilution of solutions. For exam ple, heat is
absorbed from the surroundings w hen w ater evaporates to
w here n is the m olar coefficient of each reactant form w ater vapor:
and product taken from a balanced equation rep H 2 0 (l) -> H zO (g) (11.24)
resenting the reaction and i identifies the com
pounds or ions th at participate in the reaction. W hen the evaporation occurs in the standard state:
Standard enthalpies are expressed in units of AHR = H/ (H 2 0 (g )) - /T )(H 2 0 (l)) (11.25)
kilocalorie per m ole (kcal/m ol), where the nor
From A ppendix B:
m al calorie is defined as the am ount of heat
required to raise the tem perature of 1 g of w ater from AHR = (-5 7 .8 0 ) - ( - 6 8 .3 2 ) = +10.52 kcal (11.26)
14.5 to 15.5 C. The calorie is equivalent to
160 TH ER M O D YN A M IC S
Evidently, the evaporation of w ater at 25 C and 1 atm all of the heat that enters a system during a reversible
pressure is an endotherm ic process that rem oves heat from change in its state is available to do work. Similarly, the
the surroundings and thereby causes the tem perature to direction in which a reaction proceeds spontaneously is
decrease. The heat required to evaporate one mole of a not determ ined solely by the enthalpy change because
liquid is called the latent heat o f evaporation. W hen the systems have an additional property called entropy. We
vapor con denses to form the liquid, the heat given off is return to this phenom enon in Section 11.6.
called the latent heat o f condensation. In other words,
evaporation is endotherm ic, whereas con densation of w
ater vapor is exothermic, and the latent heat of evaporation
is equal in m agnitude to the heat of condensation. 11.5 Heat Capacity
Similarly, fusion or m elt ing of solids is endotherm ic,
W hen heat is added to a solid, a liquid, or a gas, the tem
whereas crystalliza tion of liquids is exotherm ic, and the
perature of the substance increases. This fam iliar
latent heat o f fusion is equal to the latent heat o f
observation can be stated as an equation:
crystallization.
dq = C d T (11.27)
N ote th at we are unable to calculate the
w here dq is an increm ent of heat added, C is a constant
latent heat of fusion or crystallization of w ater
known as the heat capacity, and d T is the corresponding
because it takes place at 0C rather than at 25 C.
increm ental increase in the tem perature expressed in
Therefore, in order to calculate the latent heat of fusion of
kelvins (K). In therm ody namics, tem peratures are
w ater, we m ust first extrapolate the enthalpies of form
always expressed on the absolute or Kelvin scale, which
ation from 25 to 0C. In gener al, chemical reactions on the
has its zero point at 273.15C, w hen all m olecular m
surface of the E arth occur at tem peratures betw een 0 and
otion stops and the heat content of pure crystalline m
100 C, and their heats of reaction therefore com m only
aterials becomes equal to zero.
differ from those calculated at 25 C.
The heat capacity is a characteristic property
The response of a reaction at equilibrium to a
of chemical elem ents and their com pounds that
change in tem perature can be predicted on the
m ust be m easured experim entally. It turns out
basis of Le C hteliers principle. If the forw ard
that heat capacities actually vary with tem pera
reaction is exotherm ic, an increase in the tem pera ture
ture and that it m akes a difference w hether the
favors the backw ard reaction because it con sum es heat
heat is added at constant pressure or at constant volume.
and therefore counteracts the increase in the tem perature.
On the surface of the E arth pressure is constant, and we
Consequently, the equilibrium will shift to the left m aking
therefore w ork with the heat capacity at constant
the equilibrium con stant smaller. Similarly, a decrease in
pressure symbolized by CP. Consequently, we restate
tem perature applied to a reaction at equilibrium in which
equation 11.27 as:
the forw ard reaction is exotherm ic causes the equilib rium
to shift to the right and increases the value of the dqP = CPd T (11.28)
equilibrium constant. Evidently, the enthalpy change of a
We now recall that the heat added to a closed system
chem ical reaction is related to the way in which the num
during a reversible change in its state at constant pressure
erical value of the equilibrium constant varies with tem
is equal to the increase in the enthalpy of the system . This
perature. This relation ship is form ally expressed by the
conclusion was expressed by equation 11.16, which we
vant H off equa tion, which will be derived in Section
now restate in terms of differentials:
11.10.
, , b X 10 3r , , s,n
d H = a I d T + b X 10 3 TdT
T JT T
162 TH ER M O D YN A M IC S
th at all systems possess a property called entropy, nam ed of the system . However, unlike internal energy and
by the G erm an physicist R udolf Clausius (1822-1888). O enthalpy, the absolute value of the entropy of a system can
ne way to explain entropy is to say that a certain fraction be determ ined by virtue of the third law of therm odynam
of the enthalpy of a system is not convertible into w ork ics.
because it is con sum ed by an increase in the entropy. A M easurem ents indicate th at the heat capaci ties of
nother way to explain the m eaning of entropy is to com pure crystalline m aterials decrease with decreasing tem
pare it to the property of random ness recognized in perature and approach zero at absolute zero. This evidence
has been elevated into the third law o f thermodynamics,
statistical mechanics. W hen one tries to predict which states:
the behavior of a collection of particles, one com es to the
conclusion th at they will m ost p ro b ably assum e a The heat capacities of pure crystalline substances
condition of m axim um random ness. This conclusion become zero at absolute zero.
leads to the generalization th at every system that is left to
itself will, on the average, change tow ard a condition of Since dq = C d T (equation 11.28) and
maximum random ness. The im plication of this statem ent dS = d q /T (equation 11.37), it follows that:
is th at the entropy of a system increases sponta neously dS= c(y) (11.39)
and that energy m ust be spent to reverse this tendency.
1"-/.Mr)
(11'40)
ASR= 2 -^(p ro d u cts) 2 ,5(reactants) w here H is the enthalpy and S is the entropy. The change
in the Gibbs free energy of a chem ical reaction in the
(11.41) standard state is:
W hen ASR is positive, the entropy of the system AGR = AHR - TASR (11-47)
increases as a result of the change in state, w here as w
hen AS^ is negative, it decreases. For exam ple, when The Gibbs free energy is a function of the state of the
liquid w ater evaporates in the standard state to form w system like the internal energy and the enthalpy. It is
ater vapor: therefore subject to the sam e con ventions regarding the
H 2 0 ( l) -> H 2 0 (g ) (11.42) standard state as enthalpy and is m easured in kilocalories
per mole:
the change in entropy is (K rauskopf, 1979):
The standard Gibbs free energy o f formation of a
ASR = 45.10 - 16.71 = +28 .39 cal/d eg (11.43)
compound (Gf) is the change in the free energy of
the reaction by which it forms from the ele ments in
The increase in the entropy is consistent with our the standard state.
expectation that w ater m olecules in the gas phase are m
ore random ly distributed than m olecules of liquid water. N um erical values of Gf for com pounds and ions in
Similarly, w hen solid N aCl dissoci ates to form ions: aqueous solution are listed in A ppendix B together with
values of Hf . The value of AGR for a reaction is
calculated by summing the values of Gf of th e products,
N aCl(s) -> N a+ + Cl" (11.44) each m ultiplied by its m olar coefficient, and by
the change in entropy is (K rauskopf, 1979): subtracting from it the sum of the Gf values of the
reactants:
ASR = 14.0 + 13.5 - 17.2 = +10.3 cal/d eg (11.45)
AGR = 2 n , Gfi(products) (reactants)
Evidently, the entropy increases when N aCl dis solves to (11.48)
form ions in aqueous solution because the ions occupy The algebraic sign and m agnitude of AGR
fixed lattice positions in the crys tal w hereas the ions in
the solution are mobile. H ow ever, for other reactions depend on the sign and m agnitude of AH R and ASR in
ASR may be nega tive, which implies a reduction in the equation 11.47. If the forw ard reaction is exotherm ic
entropy of the system. (AHR is negative) and th e entropy increases (ASR is
positive), then the tw o term s of equation 11.47 com bine
to m ake AGR negative. If, how ever, the entropy
decreases (ASR is nega
tive), then TASR in equation 11.47 is positive. If
11.7 Gibbs Free Energy AHR is negative, A GR will be less negative than
AHR and may even becom e positive if TA S > AH .
The second law of therm odynam ics implies that the
N ote that TA S is in fact a form of enthalpy
increase in the enthalpy of a system during a reversible
because AS = A H /T at constant pressure and
change in its state at constant tem pera ture is dim inished
tem perature (equations 11.16 and 11.37).
because a certain am ount of the enthalpy is consum ed by
Therefore, TAS has the dim ensions of A H and the two
an increase in the entropy of the system . This insight was
term s can be added or subtracted from each other.
used by J. Willard Gibbs (1839-1903) to define a new
form of energy, now known as the Gibbs free energy (G): W hen AGR of a reaction is negative, the for ward
G=H-TS (11.46) reaction has excess energy w hen it occurs in the standard
state. W hen AGR is positive, there is
164 TH ERM O D YN AM ICS
a deficiency of energy for the forw ard reaction in the We can also calculate AHR and ASR for reac
standard state and the reaction will therefore run in the tion 11.51 using values of Hf from A ppendix B
backw ard direction. In general, AGR is a m easure of the and of Sf from K rauskopf (1979):
driving force of a chem ical reac tion such as equation AHR =[(-129.7) + (-161.84)] - [(-288.6)]
11.17:
= -2.94 kcal (11.53)
2 A + 3 B > A 2 B 3
ASR =[(-12.7) + (-13.6)] - [(+22.2)]
If AGR of this reaction is negative, the reaction pro ceeds
as w ritten from left to right in the standard state. W hen = 48.5 c al/d eg (11.54)
the reaction ultimately reaches equilib rium , the product
is m ore abundant than the reac tants so that the N ote that ASR for the dissociation of calcite into ions is
equilibrium constant K is large: negative, implying a decrease in the entropy. N ote also
that the entropy change is expressed in calories per degree
[ a 2 b 3] (11.49)
=K (cal/deg) rather than in kilocalories (kcal). We can now
[A ] 2
[B]: com bine AHR and ASR for reaction 11.51 and calculate
If, however, A G R for reaction 11.17 in the standard AGr , thereby testing the internal consistency of the
therm odynam ic data:
state is positive, the reaction runs the other way:
2 A + 3 BA 2 B 3 (11.50)
AGR = - 2 .9 4 - 298.15(48.5)
(11.55)
Therefore, th e equilibrium constant is small because the
1000
reactants are m ore abundant than the product. Evidently,
the sign and m agnitude of AG^ of a reaction can be used w here 298.15 is the tem perature of the standard state and
to predict both the direc tion in which a reaction runs in we divide the TAS term by 1000 in order to convert
the standard state and the m agnitude of the equilibrium calories to kilocalories.
constant. The result is:
In conclusion, we calculate AGR for the fam il iar AGR = +11 .5 kcal (11.56)
reaction: which is in satisfactory agreem ent with AGR =
C a C 0 3 - Ca2+ + C O f (11.51) + 11.43 kcal calculated from equation 11.52. N ote that
calcite AGR for the dissociation of calcite into ions is positive
using the data in Appendix B: even though the reaction is exotherm ic, that is, AHR is
AGR = [(-132 .3) + (-126 .17)] - [(-269 .9)] negative. This occurs because m ore heat energy is
consum ed to decrease the entropy than is produced by the
= +11.43 kcal (11.52) dissociation. Therefore, in this case AHR is not a reliable
predictor of the
Since AGR is positive, the reaction runs from right to left direction in which the reaction runs in the stan
in the standard state and calcite precipi tates. The dard state.
equilibrium constant for reaction 11.51 should therefore The sign of AHR does, however, enable us to
be a small num ber, as we know it to be. It is not use Le Chateliers principle to predict how the equilibrium
surprising that the reaction (equa tion 11.51) runs to the constant of reaction 11.51 varies with tem perature. Since
left in the standard state because the activities of the ions the reaction is exotherm ic as written, it produces heat
in the standard state are each equal to one and the ion which increases the tem perature. Therefore, if reaction
activity product is therefore much larger than the equilib 11.51 is at equilibri um and the tem perature is decreased,
rium constant. the reaction moves to the right because that tends to
increase
11.8 DERIVATION OF THE LAW OF MASS ACTION 165
dG = dq - P d V + P d V
the tem perature. W hen Substituting into 2 " A - ' 2 n iGi = R T '2 , n i \n P i
equilibrium is equation 11.60
Next, we assum e that the
reestablished at a lower Since dS = d q /T , dq = yields:
m ixture of ideal gases is
tem perature, the TdS and equation 11.59 the result of a chemical
equilibrium constant is therefore reduces to: dG = R T ^ p j
reaction and specify th at
larger than it was before. Integrating from the standard pressure
dG = V dP the m olar coefficients (n)
Therefore, the solubility-
som e other pressure (P): of the products are p o s
product constant of calcite We can use equation 11.60
dG = itive and those of the
J
increases with decreas ing to calculate the change in
reactants are negative.
tem perature and calcite the Gibbs free energy of
The provision is not new
becomes m ore soluble. A one m ole of an ideal gas
or arbitrary but goes back
s we discussed in Section as a function of pressure at yields: to the convention
10.4, thermodynamics constant tem pera G P - G = R T (lnregarding
P the algebraic
provides a rational ture. Since the volume
of gases varieswith pres sign of heat. If we choose
explanation for the tem Since P = 1 atm , In P
sure, wem ust express to represent the chem ical
perature dependence of the
the volum e in = 0. Therefore, equation reac tion by the equation:
solubility of calcite and
equation 11.64 reduces to:
allows us to predict the
11.60 by an equation that
course of other chemical GP- G=RT\nP aA +bB^cC +dD
relates volume to pres
reactions provided the
sure. A ccording to the w here GP is the m olar
necessary data are
ideal gas law, for one mole free energy of the ideal
available.
of gas: gas at pressure P, G is the
m olar free energy in the
11.8 Deri standard state, and T is the
vatio V = tem perature of the
standard state in
n of where R is the gas kelvins.The m olar free
the constant. energy is also called the
Law chemical potential (/x),
and eq u a tion 11.65 can
of be w ritten as:
Mass
Actio ix = fjL + R T \ n P
Problems
1. Calculate the 5. Calculate K r for the
enthalpy change when (NaAlSijOg), water, and
fluorite (CaF2) dissolves in (Al2Si20 5(OH)4) and silicic acid.
water in the standard state. 6. Apply the Law of
2. Based on the result in Mass Action to the
Problem 1, predict how the reaction in Problem 5,
solubility of fluorite varies take logarithms to the
with temperature. base 10, and express the
3. Calculate the resulting equation in
standard free energy terms of log [Na+]/[H +]
change of the reaction by and log [H4S i0 4].
which fluorite dissociates 7. The water of the
into ions. Mississippi River at
4. Calculate the Baton Rouge, Louisiana,
solubility of fluorite in has the following
water at 10, 20, and 30 C chemical composi tion at
and express each in terms pH = 7.2.
of the concen tration of
Ca2+ in mg/L. (Assume y =
1 for all species.)
REFERENCES 171
ppm ppm Mississippi River at Baton Rouge; that is, determine
G ib b site, a n d A m o rp h o u s Silica in th e P re se n c e o f W a te r a t 25 C
a n d 1 a tm p re s s u re 3
Microcline-kaolinite
Kaolinite-gibbsite
Muscovite-kaolinite
Muscovite-gibbsite
K A l3 Si3 O 1 0 (O H ) 2 + 2 K + +
K A l3 Si3 O 1 0 (O H ) 2 + 2 H
If sufficient microcline is available to drive reaction diagram s for many different systems at different tem
12.18 until the chemical com position of the w ater peratures and pressures was published by Bowers et al.
reaches point E, all three m inerals (microcline, kaolinite, (1984).
and muscovite) coexist in equilibrium w ith the ions in the The stability limits of albite (N aA lSi 3 0 8) and
solution . Consequently, the chemical w eathering anorthite (C aA l2 Si2 0 8) are shown in Figure 12.2, and
reactions stop at point E provided that the tem perature the relevant equations are listed in Table 12.2. These
rem ains constant and the system rem ains closed. diagram s can be interpreted in the same way as the
microcline diagram . They are, in fact,
The reaction path shown in Figure 12.1 is one of an com plem entary to Figure 12.1 because they not
infinite num ber of possible paths whose course depends only introduce N a+ and Ca2+, respectively, but
on the initial chemical com position and p H of the water. also involve H + and H 4 S i0 4. In m ost natural envi ronm
The work of Steinm ann et al. (1994) indicated th at the ents all three m inerals are likely to be p re sent.
shape of reaction paths in Figure 12.1 is especially sen Therefore, the activities of H + and H 4 S i0 4 of natural
sitive to the initial pH of the water. If pH > 5, the reaction solutions reflect the progress of all three chemical
paths proceed from point A to the gibb-site-m uscovite reactions occurring simultaneously. In addition, other
boundary w ithout entering the kaolinite field. reactions may be taking place involving other silicate
minerals, as well as car bonates, phosphates, sulfides, or
sulfates. As a result, predictions about the progress of one
The purpose of this discourse is to em phasize that par ticular reaction occurring in such com plex natural
the reactions betw een microcline (or anoth er m ineral) environm ents may be affected by the other reac tions that
and acidified w ater not only cause transform ations are occurring simultaneously.
among the solid phases but also affect the chemical com
position of the water. These reactions therefore contribute The stability diagram of albite in Figure 12.2A
significantly to the chemical evolution of w ater both on includes the Na-m ica paragonite and therefore is sim ilar
the surface and below the surface of the E arth . Therefore, to that of microcline. The solubility limit of m agadiite
activity diagrams like Figure 12.1 p ro vide useful inform (Surdam and Eugster, 1976) is listed in Table 12.2 but is
ation about several geochem i cal questions: not shown in Figure 12.2A.
The stability of analcim e is discussed in Sec
2.2
tion 1 with other zeolite minerals, w hereas
reactions am ong the clay m inerals are discussed
1. W hat environmental conditions are required to in C hapter 13.
allow a particular mineral to form? The stability field of the Ca-m ica m argarite is
2. W hat m inerals are stable in a given geo chemical not shown in Figure 12.2B because it is a rare
environm ent? m ineral and its standard free energy of form a
3. W hat ions or molecules are consum ed or tion is not well known. In addition, the activity of
produced when an unstable m ineral reacts in a Ca2+ in aqueous solutions reacting with an o r
given geochemical environm ent? thite is lim ited by the solubility of calcite, which
4. How does the w ater evolve chemically depends on the fugacity of C 0 2 of the environ
w hen reactions occur in a closed system m ent and on the H + activity. The equation that governs
with a small w ater/rock ratio? the solubility lim it of calcite can be con structed as
follows. We know that calcite dis solves to form Ca2+ and
The answers to these kinds of questions provide inform H C O j at p H values betw een 6.4 and 10.3 (Section 9.5).
ation that is useful in hydrogeology, sedi m entary H ow ever, the bicarbonate ion is also in equilibrium w ith
petrology, soil science, geochem ical prospecting, m ole cular H 2 CO j, which equilibrates w ith C 0 2 (g) + H
economic geology, clay mineralogy, and other im portant 2 0 . Therefore, we express the solubility of cal cite in term
[Na+] (1)
loS Tu+T = 2 lQg [H4SiQ4] - 0.19
[H+]
Kaolinite-gibbsite
logWT = +7-63 ^
Paragonite-gibbsite
Albite-pyrophyllite
Pyrophyllite-Kaolinite
Table 1 2 .2 (c o n tin u e d )
CaCO, + 2 H V Ca + CO, + H ,0
[ca2+l
lQg VfTTIT = 13-38 if [c 2] = 3 x 10-4 atm (5)
LH J
Based on the thermodynamic data in Appendix B.
182 MINERAL STABILITY DIAGRAMS
thereby avoid having to introduce the bicarbon ate ion A nother consequence of the calcite solubility limit is
into the equations:. that neither gibbsite nor kaolinite can be converted into
anorthite because log [Ca2 +]/[H + ] 2 of the solution cannot
C a C 0 3 (s) + 2 H 3 Ca2+ + CO ,(g) + H 2 0 (l)
rise above 13.38. Therefore, the reactions betw een
(12.19)
acidified w ater and anor thite that produce either kaolinite
In the standard state, AGR = -13.441 kcal and
or gibbsite are
not reversible at 25 C. If the fugacity of C O z is
hence K = 109'85. If [C 0 2] = 3 X IO" 4 atm:
increased by a factor of 10 to 3 X 10- 3 atm ,
(12.20)
1 0 9
[Ca2+]
log [Ca2 +]/[H + ] 2 = 12.38 (equation 12.20), which
= 10 13.38
log [Na+]/[H +]
log [Ca2+]/[H +]2
all of which have identical m olar S i/A l ratios of 25 C only in those environm ents th at lie within the
small triangular field labeled L . In addi tion, the
1
2 :
but differ in the num ber of w ater molecules
they contain. Therefore, we choose laum ontite to conversion of anorthite to laum ontite is irreversible
represent these minerals on the stability diagram in Figure because the calcite solubility b ound ary at atm ospheric
12.3B because it forms in near-surface envi ronm ents and values of the fugacity of C 0 2 prevents the [Ca2 +]/[H + ] 2
may subsequently convert to leon- ratio of natural solu tions from increasing to values within
hardite or wairakite. L eonhardite form s by the anor thite field.
dehydration of laum ontite upon exposure to the
atm osphere, whereas wairakite becom es stable We can com bine the stability diagram s of analcim e
betw een 150 and 300 C at 1 atm pressure (Bowers et al., and laum ontite (Figure 12.3) by specify ing that the
1984).The transform ation of anorthite to lau m ontite solutions are saturated with respect to am orphous silica.
proceeds as follows (line 1, Figure 12.3B): This presents no problem for laum ontite, which can be in
equilibrium with such a solution. How ever, analcime,
C aA l2 Si2 O g + 2 H 4 S i0 4^ C aA l2 Si4 0 1 2 4H 20 paragonite, and gibbsite are not stable in contact with
(12.27) silica-satu rated solutions. Therefore, we anticipate that
A GR = -1 1 .3 2 kcal K = IO8 3 0 kaolinite is the stable phase in silica-saturated solutions
having low activities of N a+ and Ca2+ and that kaolinite
log [H4S iO j = - 4 .1 5 (12.28)
can coexist in equilibrium with albite and laum ontite as
the activities of N a+ and Ca2+ in the solution increase.
The decom position of laum ontite to kaolinite is Actually, pyrophyl-lite forms from kaolinite just below
represented by the equation: the silica-sat-
C aA l2 Si4 0 ]2' 4H 20 + H 20 + 2 H +^
A l2 Si2 0 5 ( 0 H ) 4 + Ca2+ + 2 H 4 S i0 4 (12.29) uration limit (Figure 12.2). H owever, we prefer
kaolinite because it is far m ore abundant as a
AGR = -10 .73 kcal K = 107 ' 8 7 w eathering product of alum inosilicate m inerals
[Ca2+1 (1 2 3 ) than pyrophyllite. Accordingly, we begin the con struction
7
of this diagram with the conversion of kaolinite to albite
l0 g W F = ~ 2 1 o g [H 4Si4] + 7 .8
in a silica-saturated solution:
E quation 12.30 has been plotted as line 2 in
A l2 Si2 O s(O H ) 4 + 2 N a+ + 4 S i0 2 (a m o rp h .)^
Figure 12.3B. Finally, the laum ontite-pyrophyllite
equilibrium is represented by: 2 NaAlSijOg + H 20 +2H+ (12.33)
AGR = +13.899 kcal K = 1(T1 0 1 9
C aA l2 Si4 0 1 2 4H 20 + 2 H V
A l2 Si4 O 1 0 (O H ) 2 + Ca2+ + 4 H 20 (12.31) log [N a+] 12.4) (12,34)
= +5 .09 (line 1, Figure
A GR = -1 8 .0 0 kcal K = io + 1 3 -2 0 [H +]
= +13 .20 (12.32)
[Ca2 + 1 The reaction of kaolinite with Ca2+ to form lau
log + 12
m ontite is represented by:
[H
A l2 Si2 0 5 (O H ) 4 + Ca2+ + 2 S i0 2 (am orph.)
E quation 12.32 forms line 3 in Figure 12.3B. All other
boundaries are the same as in Figure 12.2B. + 3 H 2 0 ^ CaA l2 Si4 0 1 2 4H 20 + 2 H + (12.35)
We see th at laum ontite has a large stability 15
A GR = +17.941 kcal K = IO- 1 3
field in Figure 12.3B, but only a very small p art
of it lies w ithin the com positional range of n a t log ICa2+l
ural solutions. T herefore, anorthite, or any other A l- r +12 = +13.15 (line 2, Figure 12.4) (12.36)
bearing com pound, can alter to laum ontite at LH J
186 MINERAL STABILITY DIAGRAMS
, [Na+] o s ! lo
tCa2+ - 1.48
log [Ca2 +]/[H + ] 2 log T7JTT = '5
[H 4] B [H 4 ] 2
Figure 12.4 Stability diagram for albite (Ab), kaolin ite (line 3, Figure 12.4) (12 .41)
(K), laumontite (L), and calcite (C) in contact with
solutions saturated with respect to amorphous S i02 (Si) at The three lines derived from reactions 12.33, 12.35,
25 C. Analcime, paragonite, anorthite, and gibb site were and 12.37 have been p lo tted in Figure 12.4
excluded from this diagram because they are unstable in to g eth er with the solubility lim it of calcite
silica-saturated solutions as shown in Figure 12.3. (Table 12.2B, equation 5) for [C 0 2] = 3 X 10 4
However, pyrophyllite is probably stable, but was excluded
because kaolinite is more abundant as a weathering
atm . The resulting diagram delineates the con
product of Al-silicates than pyrophyl lite. The calcite- ditions under w hich albite decom poses to kaoli
saturation limit is based on nite + silica or laum ontite + silica. Its virtue
[C 02] = 3 x IO-4 atm (Table 12.2B, equation 5). The lies in the fact th at the environm ental condi
numbered lines (1,2, and 3) were derived from equa tions are described in term s of both N a 4 and
tions 12.33,12.35, and 12.37, respectively. C a24, but it is lim ited by the instability of cer
tain solids in the presence of silica-saturated solutions.
E quilibrium betw een albite and laum ontite in sil ica-
saturated solutions is based on:
Natural
forsterite, enstatite, talc, and
anthophyllite by magnesite, serpen tine,
+ 12 - solutions sepiolite, or amorphous silica. If
amorphous silica precipitates first, and if
one of the more soluble Mg silicates
continues to dissolve, the [Mg2+]/[H +]2
of the solution rises until sepiolite
+ 10 -
N\ \ Y coprecipitates with silica. Similarly, the
Mg2*, H*, H4S i0 4 -Silica
\
\ \ initial precipitation of magnesite may
1------- 1------- r t '-------- 1------- -------- r eventually result in the formation of
8 -6 -4 -2 0 serpentine, provided the activity of H4S
log [H4S iO J i0 4 increases sufficiently.
12.3 MAGNESIUM SILICATES 189
Kaolinite-chlorite
Mg(OH)2 ^ M g 2+ + 2 O H '
B. Solubility of Mg Silicates
Forsterite
Mg2S i0 4 + 4 H V 2 M g 2+ + H4S i0 4
AGr = -38.76 kcal K = 10+ 2 8 -4 2
fMe2+l
log = ~0'5 log [H4Si 4] + 142 (1)
Enstatite
Table 12 .3 (continued)
2 .6 kcal K = IO1 7 7 8
surface of the E arth m ust be 1 atm or less. If the pressure
of gases rises to higher values, they expand against the atm It follows th at at equilibrium the fugacity of 0 2
osphere, which m eans th at bubbles form in w ater m ust be io - 8 8 -9 2 atm , which is also less than the
exposed to the atm osphere lowest fugacity of 0 2 in equilibrium with liquid
12.5 FU GACITY DIAGRAMS195
w ater. Therefore, m etallic Fe cannot be stable on the surface of the E arth, w here th e fugacity of 0 2 Taking logarithm s to the base 10:
m ust be > 10 8 3 1 atm . H ow ever, m agnetite does
5 log [ 0 2] - 3 log [C 0 2] = -32 .61 (12.59)
occur in surface environm ents and, therefore, is
log [ 0 2] = 6 log [C 0 2] - 65.22 (12.60)
stable at 0 2 fugacities that are com patible with
liquid water. Similarly, h em atite reacts with C02 to form
We consider next the reaction betw een mag netite
and hem atite in the hope that this equilibri um will siderite according to:
provide a stability limit for m agnetite that is w ithin the Fe2 0 3 + 2 C 0 2^ 2 F e C 0 3 + j 0 2 (12.61)
range of perm issible 0 2 fugacities:
AGR = + 45 .52 kcal K = IO' 3 3 3 7
2 FeO Fe2 0 3 + 2 02 = 3Fe203
(12.56) and:
AG0 = - 4 7 .6 kcal K ,34.89
= 10; log [ O j = 4 log [C O J - 66.74 (12.62)
Thus, m agnetite and hem atite are in equilibrium
at [ 0 2] = IO- 6 9 8 atm, which is within the perm is The equations representing solid -gas equilibria derived
sible range of 0 2 fugacities. H em atite is the most above have been plotted in Figure 12.8 in coordinates of
O -rich oxide of Fe and can coexist with w ater at 0 2 log [ 0 2] and log [C 0 2], The result is a fugacity diagram
fugacities from IO- 6 9 8 to 1 . 0 atm . for the oxides and
Next, we add siderite (F e C 0 3) to the system by carbonates of Fe th at indicates the fugacities of
reacting m agnetite and hem atite with CO z gas. Since Fe the tw o gases th at are required to stabilize each
in siderite is divalent, we start with m ag netite because it of the m inerals we have considered . T he d ia gram also
also contains divalent Fe. We construct the desired predicts the direction of solid -gas reactions th a t occur w
equation with m agnetite as the reactant and siderite as the hen a particular m ineral is placed in an environm ent in
product. Then we which a n o th er m in eral is stable.
add C 0 2 to the reactant side to balance the C in
siderite. Finally, we com plete the equation by For exam ple, so-called ironstone concretions com
adding 0 2 as needed for balance: posed of siderite may be exposed to the atm osphere as a
result of erosion of the enclosing sedim entary rocks.
FeO Fe2 0 3 + 3 C 0 2 = F e C 0 3 + 5 0 2 (12.57)
Figure 12.8 indicates th at hem atite is stable in contact
with the atm osphere.
N ote that this is an oxidation -reduction reaction and that
Therefore, siderite is transform ed into hem atite
the equations cannot be balanced in term s of mass only,
in accordance w ith equation 12.61. A t equilibri
as we just did. In this case, two of the three divalent Fe
um betw een siderite and hem atite:
ions in siderite give up one electron each for a total of
[o2]
-33.37 (12.63)
two. These elec trons are accepted by one atom of 0 2 gas, 1 /2
= (12.58)
K
L10-3337 J
[co2]3 10
16.51 (12.64)
= 10 atm
196 MINERAL STABILITY DIAGRAMS
hem atite, and siderite, provided the fugacity of O z is low T herefore, certain m inerals cannot be in equilib
enough. W hen native Cu is exposed to the atm osphere, it rium in the presence of w ater and also cannot
tarnishes by form ing a layer of tenorite or m alachite or form by transform ation from other m inerals.
both, depending on the fugacity of C 0 2. This phenom enon is exem plified by anorthite,
which does not form as an authigenic m ineral at
E arth -surface tem peratures, w hereas albite and
K -feldspar do occur in places w here alkali-rich
brines can react w ith volcanic glass, clay m iner als, or
12.6 Summary
other kinds of solids.
W hen the silicate m inerals of igneous and meta-m orphic The conditions required for stability of m in erals can
rocks are exposed to acidified water, they may dissolve in also be expressed by their congruent solution or by their
co n g ru en tly to produce clay m inerals and zeolites as interaction with gases such as 0 2, C 0 2, and S2. For
well as insoluble oxides and hydroxides, depending on th e example, the non-alum inous Mg silicates dissolve
environm ental conditions. The chem ical reactions th at congruently to form Mg2+ and H 4 S i0 4, which may
take place precipitate as a different Mg silicate whose solubility is
during these transform ations can be used to exceeded. The com bination of congruent solution of a
define the conditions required for equilibrium prim ary Mg silicate and the precipitation of a
w hen both reactants and products are stable. secondary m ineral results in the replacem ent of one m
T herefore, equations can be derived from the iner al by another. The apparent conversion of ensta-tite to
Law of Mass A ction, which specify the conditions serpentine is an exam ple of such paired reactions.
required for equilibrium betw een the solids.
T he resulting m ineral stability diagram s
define the environm ental conditions in which The equilibrium betw een w ater and its con stituent
each of the solids in a given com positional sys gases, 0 2 and H 2, defines the stability range of w ater w
tem is stable. In addition, the diagram s predict ithin which all reactions on the surface of the E arth take
place. Solid oxides, car
w hat happens to a m ineral w hen it is exposed to
conditions outside of its own stability field. The reactions bonates, and sulfides m ust sim ilarly rem ain in
0
equilibrium with their constituent gases 2,
th at occur in such cases n o t only cause transform ations
am ong the solids but also affect C 0 2, and S2, respectively. These solid -gas reac
the chem ical com position, or quality, of w ater tions can be used to construct fugacity diagram s th at
b o th above and below the surface of the E arth . contain the stability fields of the solids appropriate to a
W hen the reactions occur in closed system s with sm all w particular system . The fugacity
ater/ro ck ratios, the chem ical com posi diagram s of different m etals can be superim
tion of th e w ater m ay evolve along curved reac posed on each other to form com bination dia
tion paths. Such conditions exist in the pore grams. These diagram s effectively subdivide the
spaces or fractures of rocks in subsurface w here stability field of w ater into environm ents in
reactions betw een m inerals and w ater not only which suites of m inerals of different elem ents
co n trib u te to the chem ical evolution of the can coexist in equilibrium .
w ater but also cause the deposition of m ineral The virtue of activity and fugacity diagrams is that
cem ents. they depict the relationships betw een m iner als and
The solubility lim its of certain com pounds, aqueous solutions or gases in a readily usable form . A s a
including calcite, m agnesite, am orphous silica, result, idle speculation is replaced by understanding and
and others, which can p recip itate from natural progress is made in understanding the chemical reactions
solutions, significantly restrict th e chem ical com p o sitio that con tribute to geological processes.
n s of aq u eo u s so lu tio n s in nature .
REFERENCES 199
Problems
1. Construct the equations necessary to introduce 6. What mineral precipitates first when forsterite
pyrophyllite into the activity diagram for microcline in dissolves congruently in water having the composition of
Figure 12.1. point P in Figure 12.5?
2. Construct the equations necessary to intro duce Na- 7. Construct a fugacity diagram for the sulfides and
smectite (Na0.33Al2.33S>3.670lo(OEI)2 GJ - 1277.76 carbonates of Pb.
kcal/mol) into the activity diagram for albite in Figure
8. Construct a fugacity diagram for the sulfides and
12.2.
carbonates of Zn.
3. Construct a solubility diagram for goethite (FeOOH)
9. Combine the two diagrams for Zn and Pb.
as a function of pH.
4. What are the environmental conditions in terms of 10. What are the fugacities of 0 2 and COz at which
native Cu, cuprite, and azurite coexist in equilibrium?
[Mg2+]/[H +]2 and [H4S i0 4] at which enstatite and
forsterite coexist in equilibrium? 11. Can hematite and siderite both join that assem
5. Is equilibrium possible between enstatite and blage in equilibrium with each other?
forsterite in the presence of aqueous solutions in nature?
References
at low temperature. Amer.
Mineral., 69:692-700.
B A R R E R , R. M., 1978. Zeolites K S T N E R , M., 1971. Authigenic
J O H N S O N , G . K., H. E. F L O T O W ,
and Clay Minerals as P. A. G. O 'H A R E , and W. S. feldspars in carbonate rocks.
Sorbents and Molecular W IS E , 1982.
Amer. Mineral., 56:1413-
1442.
Sieves. Academic Press, Thermodynamic studies of K S T N E R , M., and R. S I E V E R ,
London, 497 pp.
zeolites: analcime and 1979. Low temperature
B o w e r s , T. S., K . J. J a c k s o
n , and H . dehydrated analcime. feldspars in sedimentary
C. H E L G E S O N , 1984. Amer. Mineral., 67: 736- rocks. Amer. J. Sei., 279:
Equilibrium Activity 748. 435-479.
Diagrams. Springer- L IS IT Z IN A , N. A., and G. YU. B U
Verlag, New York,
T U Z O V A , 1982. Authigenic
397 pp.
zeo lites in the sedimentary
C O O M B S , D. S., A. J. E L L IS , W.
S. F Y F E , and A. M. T A Y L O R , mantle of the world ocean.
1959. Sed. Geol., 31: 33^2 .
The zeolite facies, with M E N S IN G .T . M., and G. F A U R E ,
comments on the 1983. Identification and
interpretation of age of neoformed Paleozoic
hydrothermal syntheses. feldspar (adularia) in a
Geochim. Cosmochim. Precambrian basement
Acta, 17: 52-107. core from Scioto County,
G A R R E L S , R. M., and C. L. C H R Ohio, U.S.A. Contrib.
IS T , 1965. Solutions, Mineral Petrol., 82: 327-
Minerals and Equilibria. 333.
Harper & Row, New York, P E T Z IN G , J., and R. C H E S T E R ,
450 pp. 1979. Authigenic marine
G O T T A R D I , G ., and E . G A L L I, zeolites and their
1985. Natural Zeolites, relationship to global
Springer-Verlag, New volcanism. Marine Geol.,
York, 390 pp. 29: 253-272.
H A Y , R. L., 1966. Zeolites and ST E IN M A N N , P., P. C. L I C H T N E R ,
zeolitic reactions in and W. SH O T Y K , 1994.
sedimenta ry rocks. Geol. Reaction path approach to
Soc. Amer., Spec. Paper No. mineral weathering
85,130 pp. reactions. Clays and Clay
H E L G E S O N , H . C., R. M . G A R R E Minerals, 42:197-206.
L S , and F. T. M A C K E N Z IE , 1969. S U R D A M , R. C., and H. P. E U G S
Evaluation of irreversible T E R , 1976. Mineral
reactions in geochemical reactions in the
processes involving sedimentary deposits of the
minerals and aqueous Lake Magadi region,
solutions. II. Application. Kenya. Geol. Soc. Amer.
Geochim. Cosmochim. Bull., 87:1739-1752.
Acta, 3 3 :4 5 5 -4 8 1 . S U T H E R L A N D , F. L., 1977.
H E M IN G W A Y , B. S., and R. A. R Zeolite minerals in the
O B I E , 1984. Jurassic dolerites of
Thermodynamic Tasmania: Their use as
properties of zeolites: low- possible indicators of
temperature heat burial depth. Geol. Soc.
capacities and Aust. J., 24:171-178.
thermodynamic functions W A L K E R , G. P. L., 1960.
for phillipsite and Zeolite zones and dike
clinoptilolite. Estimates of distribution in relation to
the thermochemical the structure of the basalts
properties of zeolitic water of eastern Iceland.
J. Geol., 68: 515-528.
13
Clay Minerals
Clay m inerals form one of the largest and m ost difference to the geochem ical processes that
highly diversified groups of m inerals known. The m ay be occurring. Since we encountered the
m ajority of clay m inerals are alum inosilicates clay m inerals kaolinite, pyrophyllite, and sm ec tite in C
and they crystallize as phyllosilicates. How ever, hapter 12, it is now necessary for us to exam ine these m
som e clay m inerals are M g-silicates or Fe-sili- inerals in m ore detail.
cates, others are n o t crystalline, or they are not
phyllosilicates. The property com m on to all clay 13.1 Crystal Structure
m inerals is the small size of their crystals or
grains with diam eters less th an 2 m icrom eters Clay m inerals are divided into three groups based on
(iom). A s a result, sedim ent in th at size range is their crystallographic properties: (l) platy clay m inerals
som etim es referred to as the clay-size fraction or even as (phyllosilicates), (2 ) fibrous clay minerals, and (3) am
clay. This usage is incorrect in geochem istry because orphous clay. The platy clay m inerals are much m ore
clay m inerals are defined on the basis of their crystal abundant than the other two groups and include m ost of
structure and chem ical com position and not by the size the m inerals that traditionally m ake up the clay
of their crystals or minerals. The fibrous clays sepiolite and palygorskite
grains. M oreover, the clay-size fraction of sedi (also know n as attapulgite), though not as com m on as
m ent frequently includes sm all grains of other the platy clays, do occur in many different kinds of
m inerals (quartz, feldspar, calcite, etc.), as well as grains depositional environm ents, including m arine and
of insoluble oxides or hydroxides of Fe, Al, M n, and so lacustrine sedim ent, hydrotherm al deposits,
on, which may be partly or com plete ly am orphous. N
and soils of arid regions (Singer and G alan,
evertheless, clay m inerals are typically fine grained and
1984). So-called am orphous clay, referred to as
therefore cannot be identified solely by their physical or
allophane, is an alum inosilicate of variable chem ical
optical p ro p erties. Instead, the classification of clay m
com position that is am orphous to x-rays in m ost cases
inerals relies heavily on their crystallographic and cer tain
but does diffract electrons. It may be an interm ediate
physical properties th at are revealed by x-ray diffraction.
product in the form ation of phyllosilicate clays from
various kinds of m ateri als, but its abundance is not well
Clay m ineralogy has evolved into an im por ta n t docum ented because it is not directly detectable by x-ray
field w ith applications in geology, agrono dif fraction (Carroll, 1970).
my, ceram ic engineering, soil m echanics, and
Clay m inerals are described in m any text books on
geochem istry. We will be concerned prim arily
mineralogy and in num erous specialized books,
w ith the classification of clay m inerals, with
including those by Grim (1968), Millot (1970), W eaver
th eir stability in d ifferent geochem ical environ m ents,
and Pollard (1973), Velde (1977), Sudo and Shim oda
w ith ion exchange phenom ena, and w ith th e use of clay
(1978), Sudo et al. (1981), and Caims-Smith and H artm
m inerals for dating sedim entary rocks. The presence of
an (1987). In addition, several scientific journals are
clay m inerals in n atural environm ents frequently m akes
devoted to clay min
a significant
200
13.1 CRYSTAL STRUCTURE 201
eralogy, including Clays and Clay Minerals, Clay sharing hydroxyl ions and have an A l/O H ratio of 1: 3
Minerals Bulletin, Clay Science, and Clays and instead of 1 : 6 .
Clay Technology. Articles on clay mineralogy are also The tetrah ed ral and octahedral layers are joined to
published in the Am erican Mineralogist, the form tw o-layer clays (T -O ), three-layer clays (T -O -T ),
Journal o f the Soil Science Society o f America, and the or m ixed-layer clays that are mix tures of different tw o-
Journal o f Sedimentary Petrology. layer and three-layer clays. The linking of the octahedral
The phyllosilicate clay minerals consist of two kinds and tetrahedral sheets in T - O clay m inerals takes place
of layers of different chemical composition and by sharing of O atom s and hydroxyl radicals, which
coordination. One of these is the tetrahedral layer (T), reduces the num ber of hydroxyls by one each time such a
which is composed of silica tetrahedra. The second layer link is established. T herefore, tw o-layer clays, such as
consists of A l hydroxide with A l3+ ions located in the kaolinite, are com posed of Si, Al, O, and O H in the
centers of octahedra and is there fore known as the proportions S i0 2 5 A l(0 H )2. We m ultiply the form ula by
octahedral layer (O). In some clay minerals, such as tw o in ord er to clear the decim al frac
serpentine and talc, the octahedral layer is composed of
Mg hydroxide, which forms the m ineral brucite. tion and obtain the form ula of kaolinite and its
Therefore, the octahedral layer is also referred to as the polymorphs:
brucite layer or the gibbsite layer, depending on its A l2 Si2 0 5 ( 0 H ) 4 (13.1)
composition.
W hen A l3+ occurs in the octahedral layer, only two Similarly, the form ula of three-layer clays
of every three available sites are filled. T herefore, such
clays are said to be dioctahedral. W hen A l3+ is replaced (T -O -T ), such as pyrophyllite, is 2 (S i0 2 s)A l(O H ) or
by Fe3+ or by other triva lent ions, the m inerals rem ain Si2 0 5 A l(0 H ) . A gain, we m ultiply by two and obtain:
dioctahedral. H ow ever, when the two A l3+ ions are
replaced by three divalent ions of Fe2+, M g2+, Z n2+, A l2 Si4 O 1 0 (O H ) 2 (13.2)
etc., then all positions in the octahedral layer are filled and
the clay is trioctahedral. B ased on these distinc tions we The structures of two- and three-layer clays are illustrated
conclude that kaolinite is dioctahedral, w hereas in Figure 13.1.
serpentine is trioctahedral. The phyllosilicate clay m inerals have widely varying
chem ical com positions because the A l3+ ion in the
The tetrahedral layer consists of silica tetrah e dra octahedral layers can be wholly or p art ly replaced by
that are linked together by sharing O atom s at three of the Fe3+, C r3+, Fe2+, M g2+, Z n2+, L i+, and m any other
four corners. The linkages occur only at the bases of the ions. In addition, some of the Si4+ ions in the tetrah ed ral
tetrahedra, thus forming a sheet with the unshared O atom layer are replaced by A l3+. This results in an excess of
s all pointing in the same direction. If all four O atom s of negative charge, which is neutralized by the adsorption of
silica tetrahedra are shared (as in feldspar), then each Si cations to the outer surfaces of the tetrahedral layers of
atom con adjoin ing clay units. The substitutions within the layers
trols only one half of its O atoms. Therefore, the are constrained by the requirem ent for electrical
S i/O ratio in feldspar and other tektosilicates is
1: 2 (Section 8.3). Since in phyllosilicates only the three neutrality, which is m et w hen th e cations in the
basal O atoms are shared, the S i/O ratio is 1:1 .5 . W hen octahedral layer have a com bined charge of
+ 6
the unshared O atom is counted with the basal ones, the S and the tetrahedral layers contain four m oles of
i/O ratio becom es 1 : 2.5. Si + A l per form ula weight. T he tw o-layer clays
The octahedral layer has six-hydroxyl ions located at
perm it very little substitution of either A l or Si, w
the corners of an octahedron. Therefore, the ratio of Al to
hereas the three -layer clays, w ith the exception of
hydroxyl ions is basi cally 1 : 6 . However, the octahedra
that m ake up the gibbsite sheet are linked at the corners pyrophyllite, are characterized by extensive
by substitutions in either the octahedral layer or the
tetrahedral layer, o r in both.
202 CLAY MINERALS
(a) Two-layer clay: A lS i0 2.5(0 H ) 2
13.2 Classification and Chemical
\ ^ ^ ^ 7 Composition
\ 1 \ / \ / . /Tetrahedral layer
Tetrahedral
a. Two-Layer Clays
Octahedral The phyllosilicate clays are subdivided into the two-layer,
three-layer, and m ixed-layer clays. The two-layer clays
Tetrahedral consist of two subgroups, each of which contains a num
ber of m ineral species, as listed in Table 13.1.
2. SMECTITES The
smectite group contains a
large num ber of m ineral
species that differ in chem
ical com position because
of the com plete or partial
replacem ent of A l34' in
204 CLAY MINERALS
m ineral species that differ from each other in chem ical The additional cations, w hich neutralize th e negative
composition. For example, m ontm orillonite is a prom charges caused by im perfect su b stitu tions w ithin th e
inent m em ber of the dioctahedral subgroup of smectites layers, are adsorbed on the su r
in which Mg2+ occurs in the octahedral layer, but the faces of the m ineral grains and on the
replacem ent of Si4+ by A l3 4 in the tetrahedral layer is tetrah ed ral layers of the clay units. T herefore,
quite limited. For example, a sample of m ontm orillonite they are partly or com pletely exchangeable and can be
from G erm any has the form ula (Carroll, 1970,Table 6 ): replaced by o th er cations or by the polar
m olecules of w ater in th e environm ent. The
adsorbed ions located on the o u ter surfaces of
[(Al[.7 7 FeQ0 3 M go2 o)(Si3 7 4 AlQ2 6 )O lo(O H )2]
the tetrah ed ral layers bond th e structural units
(octahedral) (tetrahedral) together. W hen the bonding involves the cations of m
(13.3) etals, the interlayer distances betw een ad ja cent
C a0 .i6 ' N ao.o7 Mgo.o4 structural units are short. H ow ever, w hen certain clay m
(exchangeable) inerals are im m ersed in w ater, the interlayer cations are
This form ula can be readily interp reted in term s of the replaced by m olecules of w ater, which have m uch sm
three-layer m odel if we determ ine the elec trical charges aller n et positive charges than m etal ions. C onsequently,
in each of the layers. The octahedral layer contains 1.77 A a larger num ber of w ater m olecules is required to n e u
l3+ + 0.03 F e3+ + 0.20 M g2+, which adds up to + 5 .8 tralize excess negative charges g enerated w ithin the
electronic charges. Since th e octahedral layer is supposed layers. As a result, the interlayer distances increase as w
to contain six positive charges, th ere is a deficiency of 0 . 2 ater m olecules crow d in and the clay expands.
posi tive charge, which is equivalent to an excess of 0.2
negative charge. In the tetrahedral layer replacem ent of
som e Si4+ by A l3+ causes a defi ciency of positive The swelling of clay m inerals in the sm ectite group
charge equivalent to a charge of -0 .2 6 . Evidently, the is a characteristic property that is used to identify them .
substitutions in th e two kinds of layers have caused a total For example, w hen natural m ont
charge of 0.2 + ( - 0 .2 6 ) = 0.46. In the example m orillonite is heated at 300 C for an hour or
before us, the negative charges are neutralized by more, the interlayer distance decreases from
adsorption of 0.16 C a2+ + 0.07 N a+ + 0.04 M g2+, which about 15 angstrom s (A ) to about 9 A because adsorbed w
together provide 0.47 positive charge. The agreem ent ater m olecules are expelled. W hen the clay is
betw een the negative charges in the layers ( 0.46) and subsequently exposed to the vapor of eth ylene glycol (H
the positive charges of the addi tional ions (+0.47) is well O C H 2 C H 2 O H ) for several hours, the interlayer
distance expands to 17 or 18 .
within the uncertainty of the chemical analysis of this m
M ontm orillonite com m only forms by alter ation of
ineral. The overall
volcanic ash and therefore is the principal m ineral of
bentonite beds, which form by alter ation of volcanic ash
charge distribution for this m ontm orillonite is deposits. The well-known swelling of bentonite, when it is
given by adding up the ionic charges of all ele placed in water, is actually caused by the expansion of the
m ents that m ake up this mineral: interlayer distances of m ontm orillonite. The m ineral also
occurs in soils of arid regions w here leaching is lim ited
1.77(A13+) + 0.03(Fe3+) + 0.20(M g2+) by lack of w ater and the pH is above 7.
+ 3.74(Si4+) + 0.26(A13+)
Beidellite is a dioctahedral sm ectite contain ing little
+ 0.16(Ca2+) + 0 .07(N a+) if any Fe2+ or Mg2+ and in which nega tive charges
originate prim arily by replacem ent of Si4+ by A l3+ in
+ 0.04(M g2+) - 10(O 2~) - 2(O H )
= +0 .01 (13.4) the tetrahedral layer. A ccording to G rim (1968), its form
ula can be w ritten as:
13.2 CLASSIFICATION AND CHEM ICAL COM POSITION 205
am ounts of Fe3+ and charge of -1 .0 8 is neutralized by ad
[(A l2 , 7 )(S i3 .1 7 A l0 .8 3 ) O Fe2+. Negative charges The sm ectite m and Ca2 4 in the interlayer position
arise in the tetrahedral ineral sauconite is characteristically contains both M g2 4
N ote that the replacem layer by replacem ent of character ized by the molecules in the interlayer
ent of Si4+ by A l3+ in Si4 4 by A l3+ and may be presence of Z n 2+ in the sites. The w ater can be
the tetrahedral layer has partly com pensated by octahedral layer and by expelled by heating at
caused a net charge of -0 . addition al cations in the lim ited replacem ent of 500C, which causes the
83, which is partly com octahedral layer. The form Si4+ by A l3 4 in th e interlayer distance to
pensated by the positive ula of saponite according decrease from about 14 A
tetrahedral layer. C arroll
charge (+0.51) generated to G rim (1968) is: to about 9 . A fter
(1970) recalculat ed a
by an excess of 0.17 Al3+ heating at 500 C or less,
chem ical analysis of
in the [(M g ^ S i^ A lg j^ O n /O H ^ ] N a verm i culite rehydrates
sauconite to obtain the
octahedral layer. The rem aining Hectorite is a Li-bearing spontaneously; however,
following form ula:
charge ( 0.83 + 0.51 = 0.32) heating at 700C causes
adsorption of 0.33 moles of M g2 4 and L i+ occur together irreversible dehydration. In
weight of beidellite. layer, but no substitutions
[(Z n 2 4 0 Mg0 1 8 A10 2 2 F'eQ.n)addi tion, when verm
N ontronite, the third occur in the tetrahedral (Si3 .4 7 A l0 ,5 3 )O iculite is treated with a
dioctahedral sm ectite in layer. Thus the form ula solution of KC1, the
Table 13.1, contains Fe3+ for hectorite is: The charge im balance in interlayer Mg2 4 ions are
in the octahedral layer in the octahedral layer of this replaced by K 4,
[(Mg2 .6 7 Li0 .3 3 )(Siclay m ineral is +0.33, that
place of A l3+ with m inor and it becom es nonexpandable. Vermicul
am ount of Fe2+ and M The sm ectite m ineral of the tetrahedral layer is - fers from sm ectites by having a higher charg
g2+. The net charge im 0 .5 3 for a net charge of the tetrahedral layer, by commonly con
stevensite is similar to
balance arises from sub -0 .2 0 , which is
hec torite because it is M
stitutions of Si4 4 by Al3+ neutralized by the
g-rich and lacks Al in the
in the tetrahedral layer. adsorption of
tetrahedral layer. How exchangeable
Therefore, the form ula of
ever, stevensite also lacks cations in interlayer sites.
nontronite is:
Li in the octahedral layer. The w ord verm iculite
[(Fe2 +)(Si3 6 7 Alo3 3 ) 0 1 0 (O H ) L atin verb vermiculari, w hich
worm s. The clay m ineral
A s in beidellite, the b o th in dio ctah ed ral and
negative charge generated w ith o th er clay m inerals
in the tetrahedral layer p ro d u ct of micas. A ccording to
may be partly com verm iculite from B are Hills, M aryland, has the
pensated by the presence form ula:
of additional Fe34, Fe24,
[(Feo.24 ^ 6 2 . 7 0 Fegg4 Ni001)
or M g2 4 in the octahedral
layer. W hen A l3+ in the (Si2.73Al1 .26 ) 0 1 o(O H )
octahe dral layer is
replaced by Cr34, the Evidently, Mg2 4 dom
resulting dioc tahedral sm inates in the octahedral
ectite is know n as layer, and the Si4 4 in the
volkhonskoite. tetrahedral layer is
The trioctahedral extensively replaced by A
smectites include the m l34. The octahedral layer
iner al species saponite, has an excess positive
hectorite, and sauconite. charge of + 0 .2 2 , which
partly com pensates for
In
the strong negative charge
saponite, the octahedral layer
of - 1 .3 0 in
m arily of Mg2+ in place of
the tetrahedral layer. The
206 CLAY MINERALS
M g2+ in the exchangeable interlayer sites, and by being 2M polym orphs and are derived by m echanical
less expandable. weathering of igneous and m etam orphic rocks. Therefore,
the 2M illite in sedim entary rocks is detrital in origin, w
3 . 1LUTE T he pow er of K + ions to stabilize clay
hereas the lM d or 1M poly m orphs may consist of
m inerals is well illustrated by illite, which is
authigenic illite. The x-ray diffraction spectra of the
listed in Table 13.1 w ith the m ica m inerals.
different muscovite poly m orphs in the clay fraction of
A ctually, illite could also be regarded as a special
sedim entary m ateri als were w orked out by Yoder and
m em ber of the sm ectite group. H ow ever, it
Eugster (1955).
differs from typical sm ectites by having a
nonexpandable lattice because of th e presence of K + ions 4. GLAUCONITE OR GLAUCONY Glauconite is a
in interlayer sites. Illite was nam ed by G rim et al. (1937) dioctahedral clay m ineral in which Fe3+, Fe2+, and Mg2+
after the state of Illinois in the U n ited States. T hey replace A l3+ in the octahedral layer. Negative charges in
intended it to represent m ica-like clay m inerals as a the octahedral and tetrahedral layers are neutralized by
group, rath er than a specific clay m ineral, and expressed interlayer K +, Ca2+, and N a+. The concentration of K +
the form ula of illite as: in glauconite is variable and increases with time depending
on the availability of K + in the environment. Johnston and
[(A l2, F e 2, M g2, M g 3 )(S i4 _I A l,)O 1 0 (O H )2] K , Cardile (1987) reported that glauconite from Point Jackson,
Francosia, Wisconsin, has the formula:
(13.11)
The form ula indicates that the octahedral layer m ay
contain Al, Fe, or M g and th at illites may be dioctahedral [(Fe2 097 A l0 8 4 9 M g 0 442Ti 0.003 M noom)
(2 A l3+ or 2 Fe3+) o r trioctahedral (3 M g2+). The value (Si3.61lAIo.38g)010(O H )2]K 0 7 2 5 C a(,n% (13.13)
of x is less than one and usu ally varies betw een 0.5 and
In this sam ple of glauconite, Fe2+ dom inates in
0.75. The negative charge generated in the tetrahedral layer
the octahedral layer, and the deficiency of posi
is com pensated by K + ions, which fit into hexagonal
tive charges in both the octahedral and tetrah e
holes form ed by the silica tetrahedra of the tetra
dral layers is largely com pensated by interlayer
hedral layers. A s a result, the K + ions are not K + and Ca2+ ions.
exchangeable, the spacing of the interlayers is G lauconite is an authigenic clay m ineral that forms
fixed, and the m ineral is not expandable. in the m arine environm ent by transform a tion from other
H ow ever, illites m ay be interlayered with m ont- kinds of m aterial (M cRae, 1972).
m orillonite or other clay m inerals th at are It commonly occurs as greenish pellets about
1 -2
The only
dioctahedral chlorite is an
Al-rich variety known as
cookeite described by
Brown
208 CLAY MINERALS
recrystallize during low -grade regional m etam or phism to of hydrotherm al alteration of pyroxene and amphi-bole.
the Fe-chlorite thuringite. One of the largest accumulations is in the Hawthorn Form
ation of M iocene age in G eorgia and Florida, which was
c. Mixed-Layer Clays deposited in fresh to brack ish water lagoons.
The clay m inerals we have described occur as
interstratified layers in the so-called mixed-layer clays, Sepiolite is also com posed of chains of silica
which are very com m on in soils and in about tetrahedra but it differs from palygorskite in spe cific
70% of clay-rich sedim entary rocks (W eaver, details. G rim (1968) reported two different structural form
1956). To some extent the principal types of ulas for this m ineral based on a review of the literature
m ixed-layer clays depend on climatic factors that affect available at the time. The
the environm ent within which chemical w eathering form ulas currently in use (see A ppendix B) are
occurs. U nder m oist and tropical con Mg2 Si3 0 6 ( 0 H ) 4 and Mg4 Si6 O t5 (O H ) 2 6H 2 0 .
ditions the stratification in m ixed-layer clays is Carroll (1970) reported that a chemical analysis of
m ontm orillonite-halloysite-kaolinite, w hereas in sepiolite from L ittle C ottonw ood, U tah, yields the
hum id tem perature regions the m ixed-layer clays consist formula:
of m ontm orillonite-chlorite-m ica or m ica-interm ediate
[(Mg7 4 2 MnQ5 3 Feon2) (Si| 1 6 7 A l024Feo J9 ) 0 3 2]Cu() 4 5
products-illite (Carroll, 1970). The crystallographic
properties of mixed-layers (13.18)
clays have been discussed by Z en (1967), Evidently, M g2+ is partly replaced by M n2+ and Fe3+, w
R eynolds (1967, 1980), and C orbat and hereas Si4+ is replaced by A l3+ and Fe3+. The structure
Tettenhorst (1987). has a net charge of -0 .3 7 , which is largely neutralized by
exchangeable Cu2+.
d. Fibrous Clay Minerals
The fibrous clay m inerals palygorskite and sepio-lite are
com posed of double chains of silica tetra- 13.3 Gibbs Free Energies of
hedra with an S i/O ratio of 4 : 11. The channel
betw een the double chains is occupied by w ater
Formation
m olecules that are lost stepwise during heating up The therm odynam ic properties of the three-layer and m
to 850C, at which tem perature the structure is ixed-layer clay m inerals are difficult to establish because
destroyed. N either palygorskite nor sepiolite of their highly variable chem i cal compositions. As a
expand w hen treated w ith organic liquids. result, standard free ener gies or enthalpies of form ation
P alygorskite is a M g-silicate with some are available only in certain specific cases or apply only to
replacem ent of M g2+ and Si4+ by A l3+. Its struc ture idealized com positions th at do not reflect the true
was w orked out by B radley (1940), who reported the form chemical diversity of real clay minerals.
ula:
5 In order to overcom e this problem, Tardy and
M g Si8 O 2 0 (O H )2 (O H 2 ) 4 4 H zO (13.17)
Garrels (1974), Chen (1975), and Nriagu (1975) pro posed
w here O H 2 is bound w ater and H 20 is w ater in
empirical m ethods for estimating standard free energies of
channels like th at of zeolites. The structure p re sented form ation of clay minerals. Tardy and Garrels
above is electrically neutral but m ay con tain (1976,1977),Tardy and G artner (1977), Tardy and
exchangeable C a2+ to com pensate charge im balances Vieillard (1977), and Tardy (1979) subse quently
caused by lim ited substitutions in this structure. generalized this approach to estimate the standard free
Palygorskite is called attapulgite in the U nited States after energies of form ation of other kinds of compounds and
an occurrence of this clay in A ttapulgus, G eorgia. their ions in aqueous solution.
The m ethod of Tardy and G arrels (1974) is based on
Palygorskite is an authigenic mineral that occurs in the prem ise th at the standard free ener gy of a
alkaline, unleached soils and as a product phyllosilicate can be treated as the sum of
13.3 GIBBS FREE ENERGIES OF FORMATION 209
th e free energies of its oxide and hydroxide com ponents. E qu atio n s 13.22-13.24 can be solved by sub
The num erical values of the free energies o f the com stitution and yield the standard free energies
ponents m ust reflect the fact that they of fo rm atio n in th e silicate lattice of th e th ree
exist in a silicate lattice and not in free form . com ponents listed in Table 13.2. O th er com po
T herefore, they m ust first be determ ined from n en ts can be ad d ed by considering the A l sili
phyllosilicate m inerals whose standard free ener cates kaolinite, pyrophyllite, and m uscovite,
gies of form ation are know n from experim ental which yield values for G^(A12 0 3), G ^(H 2 O s),
determ inations. The first step tow ard th at end is and G ^(K 2 O s).Tardy and G arrels (1974) also
to restate the structural form ulas of phyllosili- determ in ed th e stan d ard free energies of fo rm a tion in
cates in term s of oxide com ponents, except that the silicate lattice of exchangeable cations and w ater. T
M g is converted to the hydroxide until all of the hydroxyl hese values are listed in Table 13.2.
ion in the form ula has been assigned to M g(O H )2. For
example, serpentine is rew ritten: In order to illustrate their m ethod of estim at ing
standard free energies of form ation of a real
M g3 Si2 O s(O H ) 4 = 2M g(O H ) 2 M gO 2 S i0 2
(13.19)
clay m ineral, Tardy and G arrels (1974) chose a typical m Table 13 .4 Standard Free Energies of Formation
ontm orillonite: of Clay Minerals Based on Experimental
Determinations and Estimates by the Method of
[(A lisF eQ ^ M ggjX S ijjA lt^ O u /O H ^ ] M g04
Tardy and Garrels (1974)
(13.25)
G} G
The form ula is electrically neutral, which is im portant Mineral or (measured), (estimated),
because the exchangeable cations do contribute to th e free locality kcal/m ol kcal/mol
energy of the m ineral and therefore cannot be om itted . In Dlite
Table 13.3 the for m ula for this m ontm orillonite has been
recast in term s of m olar am ounts of the oxide com Fithian -1277.23 -1278.8
ponents. These are then m ultiplied by their respective G^ Beavers Bend 1274.7a -1276.2
(silicated) values from Table 13.2 and the products are Goose Lake -1272.1s -1273.4
summed. The result is 1282.5 kcal/m ol, which is a Montmorillonite
reasonable value for the standard free
Aberdeen -1230.6C -1228.3
Belle Fourche 1240.6b -1236.4
energy of form ation of m ontm orillonite. For
exam ple, A ppendix B contains a value of Colony, Wyoming 1241.5d -1240.6
1258.84 kcal/m ol listed by Lindsay (1979) for a Other Phyllosilicates
M g-m ontm orillonite. Annite (1) -1149.3 -1150.2
M ore specific com parisons betw een free Annite (2) -1151.7 -1149.1
energies of form ation, estim ated by the m ethod
Clinochlore -1961.8 -1958.7
of Tardy and G arrels (1974), and experim entally
Phlogopite -1410.9 -1400.7
determ ined values are presented in Table 13.4.
Greenalite -720.0
T he experim ental determ inations w ere m ade by
Minnesotaite -1055.0
Dlite formulas
Table 13 .3 Estimate
of the Standard Free Fithian,
Energy of Formation of a Typical [(A l1, 4Fea29Mg0.l9)(Si3, 1Al[..49)O
M ontmorillonite with the Formula Stated in 10(OH)2]Ka64 Beavers Bend,
Equation 13.25 [(AI] 6ftFeQ 2()Mgo i3)(Si3 ^2A1q 3g)OjQ(OH)2]K(j 53
1 2 3 Goose Lake,
Component, G (silicated), 1X2, [(AIj 5^FeQ24Mgo i5)(Si3 AIq 3 5 )0 ]q(0 H)2 ]Kq j y
moles kca l/m o l kcal
Montmorillonite formulas
1.0 a i 2o 3 -382.4 -382.4
Aberdeen,
0.1 Fe20 3 -177.7 -17.77 [(Ali.29Feo335Mg0 445)(Si3 82A10 i9)O
0.3 Mg(OH)2 -203.3 -60.99 10(OH)2]Ka415 Belle Fourche,
3.5 S i0 2 -204 .6 -716.1 [( Al i .5 lsFejj 225Mg0 29) (Si3 935Al0
0.7 H2Oa -59 .2 -41.44 065) O Colony,
0.4 MgO(exch.) -159 .5 -63 .8 [(A li,2 Fe3 l 2 Mgo.2 9 )(Si3 .8 1 Alo.i9 )Oio(OH)2 ]K0.4()
Sum -1282.5 kcal
"Routson and Kittrick (1971).
a0.3 Mg(OH)2 uses up 0.6 hydrogen atom, leaving 1.4 hydro gen bKittrick (1971a).
atoms to make 0.7 mole of H20 . Kittrick (1971b).
S O U R C E : Tardy and Garrels (1974). dWeaver et al. (1971).
13.4 STABILITY DIAGRAMS 211
dissolving the m inerals and by analyzing the other direct com parisons in Table 13.4 are also
resulting saturated solutions. For exam ple, impressive.
W eaver et al. (1971) dissolved a sam ple of m ont-m
orillonite from Colony, Wyoming:
13.4 Stability Diagrams
[(A li.5 2 F e 0 .2 2 ^ 0 .2 9 )(^ *3 .8 lA lo .i9 )0 10( O H ) 2] M g 0 2o
The three -layer clays are m ore sensitive to the
+ 6.10 H + + 3.57 H 2 0 ^ geochem ical environm ent in which they exist
1.71 A l3+ + 3.81 H 4 S i0 4 (13'26) than m ost o th er m inerals because of the p res
+ 0.11 F e 2 0 3 (s) + 0.49 Mg2+ ence of exchangeable ions and m olecules. T heir
approach to equilibrium takes place by exchange of the
adsorbed ions with ions or m olecules from th e environm
A t equilibrium in a saturated solution the Law of Mass A
ent, by transform ations involving the chem ical com
ction applies:
position of the layers and their
[Mg2 +]0 4 9 [H 4 SiO 4 ]3 8 1 [Al3 + ] 1 -7 1 stacking, and perhaps even by th e form ation of
[H + ] 6 1 0 m ixed-layer clays. A lthough ion exchange is
rapid, reactions th at require adjustm ents in the
(13.27)
chem ical com positions of the octahedral and
where Fe2 0 3 (s) and H 2 0(1) were given activities of one. tetrah ed ral layers, and in the bonding betw een
The value of the equilibrium constant, cal culated from individual clay units, are slow. Consequently, n a t ural m
equation 13.27 using the m easured activities of the ions aterials com m only contain m ixtures of several clay m
and molecules, was found to be K = IO0" 1. Since log K = inerals th at cannot be in equilibri um with each o th er u n
-AG*/1.364, it fol lows that AGR = 1.352 kcal. The d er one set of environ m ental conditions. For this reason,
standard free energy change that takes place during the we now turn
disso ciation of the m ontm orillonite from Colony, to the construction and use of activity diagram s
Wyoming, in accordance with equation 13.26 can be for clay m inerals in o rd er to identify the stable
expressed as: phases and to indicate the direction of reactions am ong the
clay m inerals in a given geochem ical environm ent.
[(All.8oMgo.25)(Sb.50JMo.5o)Olo(OIi)2]lC().33 -1300.98
(1307.485)
If we multiply kaolinite by
7 and K -montmorillonite to the left side of the
by 6, we can balance AI to equation to balance Si
a good approximation. As and 2 moles of K + to
a result, we must add 8
balance K. Finally, we
moles of S i0 2
need 7 moles
(amorphous)
of H 20 on the right side to balance O and 2 moles of
H + to balance H . Equation 13.29 yields
13.4 STABILITY DIAGRAMS 213
AGR = +35.511 kcal and K = l ( r 2 6 0 3 . From the
Law of Mass Action:
[ I2 = | Q 26.03 (13.30)
[K + ] 2
and therefore (line 1, Figure 13.2):
rK +l
log FTTZi- = 13.02 (13.31)
[H-
Equation 13.29 i n d i c a t e s that K-montmorillonite
is f a v o r e d o v e r k a o l i n i t e i n e n v i r o n m e n t s w h e r e [ H +] is
l o w a n d [ K + ] is h i g h . S u c h c o n d i t i o n s o c c u r i n p o o r l y d r
a i n e d s o i l s u n d e r a r id c l i m a t i c c o n d i t i o n s o r in t h e o c e
log [M g2+]/ [H +]2
answhereseawaterprovidesa
source ofK+atapH o f a b o u t 8.2. T h e f o r m a t i o n Figure 13.3 Stability of selected minerals in the sys tem
ofNa-montmorillonite from kaolinite takesplace K20 - M g 0 -A l20 3- S i0 2-H 20 -H C l at 25 C at
b y a r e a c t i o n t h a t is a n a l o g o u s t o e q u a t i o n 13.29 1 atm pressure in the presence of amorphous silica. The
(line 2, F i g u r e 13.2, l o g [ N a + ] / [ H + ] = 13.32). positions of the lines were adjusted as indicated in the
The direct conversion of K-montmorillonite footnotes to Table 13.6 to overcome internal inconsis
toNa-montmorilloniteinvolvesonlyanexchangeoft
tencies in the thermodynamic data. Nevertheless, the
diagram shows that the clay minerals kaolinite (K), illite
hetwoelements:
(I), montmorillonite (Mg-mont.), and chlorite (Ct) can
3 [ ( A l 2 . 00)( S i3 . 67A l0 .3 3 )O 10( O H ) 2] K 0.33 + N a + ^ form as weathering products of microcline (Mi),
3 [ ( A 1 2.oo) ( S i 3 .67 A l 0 33) O 10(O H ) 2] N a 0 33 + K + muscovite (Mu), and phlogopite (Ph), which crystallize
from cooling magma or grow by metamorphic recrys
(13.32) tallization of clay minerals at elevated temperatures and
AG = +0.413 k c a l K = IO 0 -3 0 2 pressures.
Therefore,atequilibrium:
Table 13.6 Equations for Mineral Stability Diagram (Figure 13.3) in the System K20 - M
g 0 - A l20 3- S i0 2-H C l-H 20 at 25 C, 1 atm Pressure, and in Equilibrium with
Amorphous Silica
Kaolinite-muscovite
3 Al2Si20 5(0 H )4 + 2 K+^ 2 KAl2Si3AlO 10(OH)2 + 3 H20 +2H+
Illite-muscovite
log^|=1.481og^?-5.52 (2)
Kaolinite-Mg-montmorillonite
[(Al2 oo)(Si3.f,7Al 33)O |o(OH)2]Mgu 167 + 0.608 K* + 0.086 Mg2+ + 0.405 H20
A GR = -3 .2 8 ab kcal K = IO241
Mg-montmorillonite-chlorite
Chlorite-phlogopite
[K+] n ^ w t Mg2+
, [H + ]- 0 .5 1 o g [H+]
Microline-illite
[K+l TMe2+l
log W \ = -147 log + 633 (8)
Muscovite-microcline
Microcline-phlogopite
KAlSi30 8 + 3 Mg2+ + 4 H2O ^K M g3Si3A lO 10(OH)2 + 6 H +
tem atic variations in the chem ical com position of the Table 13 .7 SolubilityEquations of Pure Fe-
fluid in the im m ediate vicinity of a m ineral grain m ay Silicate Minerals in Pure Water at 25 Ca
cause a sequence of reactions to occur such th at phlogopite
Fayalite
first alters to chlorite, which changes to M g -m ontm
Fe2S i0 4 + 4 H +^ 2 F e 2+ + H4S i0 4
orillonite, which ultim ately
decom poses to kaolinite or even gibbsite. A GR = -20 .76 kcal K = 101522
Similarly, illite m ay form either m icrocline or
m uscovite by a reduction of [K +] o r decrease in pH . D log I f f = ~0'5 108 [l*SiC)J + 7-61 (1)
epending on the existence of the appropri ate chem ical Ferrosilite
gradients, m uscovite flakes m ay alter to kaolinite or illite, FeSiOj + HzO + 2 H +^ F e 2+ + H4S i0 4
which m ay transform into AGR = -7.663 kcal K = IO5 62
illite -m ontm orillonite or illite -kaolinite mixed-
layer clays. H ow ever, predictions of reaction
products based on therm odynam ics alone are not always log I f f = los [Ii4Si 4] + 5-62 (2)
reliable because unfavorable kinetics may p rev en t the Greenalite
reactions from reaching equilibrium or from going to com Fe3Si20 5(0 H )4 + 6 H+^
pletion . 3 Fe2+ + 2 H4S i0 4 + H20
AGfi = -18.577 kcal K = 1013 M
b. Solubility of Ferrous Silicates
The ferrous silicates include fayalite (Fe2 S i0 4), ferrosilite
(F e S i0 3), and the tw o-layer phyllosili cate greenalite. In log I F F =~0-67108[H4Si 4] +0,53 (3)
addition, the carbonate m ineral siderite (F e C 0 3) and the Siderite
hydroxide [Fe(OH)2] m ay form . We assum e th at Fe is F eC 03 + 2 H +^ F e 2+ + C 0 2 + H20
predom inantly in the divalent state and th a t m inerals AGR = +0.509 kcal K = IO037
containing Fe3+ do not occur in the system . loS t S J = 3 1 5 at [C J = 3 x 1 C r 4 atm (4)
(5)
equation in Table 13.7, which have been plotted in Figure A GR = -15.924 kcal K = IO1167
13.4 in coordinates of log [Fe2 +]/[H + ] 2 and log [H4 S i0 4]. lFe2+l
The lines are the boundaries betw een solutions that are iogW =1L67
undersatu rated or supersaturated with respect to each Amorphous silica
miner al. For example, fayalite dissolves in solutions S i02(amorph.) + 2 H20 ^ H 4S i0 4
represented by points lying to the left of the equilib rium A G R = +4.044 kcal K = 10~2%
line (equation 1, Table 13.7). Evidently, greenalite log [H4S i0 4] = -2 .96 (6)
(equation 3, Table 13.7) is less soluble, and therefore m ore
stable, than either fayalite or ferrosilite in contact with w Based on thermodynamic constants in Appendix B.
ater at 25 C. However, the com position of natural
solutions in contact with the atm osphere is restricted by
the solubility
The solubility data in Figure 13.4 indicate th at
,boundaries of siderite (equation 4, Table 13.7) and am greenalite could precipitate in place of siderite if log [Fe 2
orphous silica (equation 6 , Table 13.7). Therefore, even +]/[H + ] 2 is greater than about 7. The fugacity of C O z th at
greenalite dissolves in natural solu tions and precipitates w ould perm it this can be cal culated from the solubility
only from solutions that have low bicarbonate ion equation for siderite in Table 13.7:
activities.
13.4 CO LLO IDAL SUSPENSIONS AN D ION EXCHANGE 217
T r~ m ost com m on iron silicate m ineral in unaltered
\ i parts of the G unflint F orm ation in w estern
\
IFe (O H )2
\ Ontario, C anada. It occurs as light, green to olive green
\
10 granules up to 3 m m in diam eter, except in the banded ch
- "\
\ \ ert-sid erite and tuffaceous shale facies. The absence of
X\ X\j)' greenalite from siderite - bearing rocks is explained by the
\ N - f - '- ? difference in their solubilities w hen th e solutions are in
+ \ I N% . '-v
contact with C 0 2 of the atm osphere. H ydrotherm al solu
X [C 0 2] = 10 7 4 atm \ SideriteX \ %
>sl \ \
tions discharged into th e G u nflint basin could precipitate
greenalite and am orphous silica if they had sufficiently
CN Greenalite | \
6 precipitates low carbonate ion concen trations to prevent siderite from
00
\ forming.
o
low carbonate ion immiscible in an o th er
4- M innesotaite is com
concentrations and are liquid are know n as e m u
[C 0 2] = 10 3 5 atm m on in the Fe-silicate
2- Natural l sions, and solid particles o
close to being saturated facies of the Biwabik Iron
r liquid droplets dis persed
with respect to am Form ation of northern
in a gaseous m edium form
solutions Minnesota, but is absent
orphous silica. Such aerosols. Such colloidal
from the G unflint
solutions m ay be dis suspensions m ay appear to
-8 -6 Formation. Jam es (1966)
charged by volcanic be h om oge neous, and the
lo pointed out that this dif
particles do n o t settle o u t
g hotsprings in shallow m ference betw een the two
under the influence of
[ arine iron form ations m ay be
gravity. Som e kinds of
H basins w here cherty iron form ations
related to the fact th a t th
particles in sols attract w
4 deposited in Precam brian e Biwabik Form ation was
ater m olecules to them
S G oodw in (1956) rep o rted intruded by the E m barras
selves and are therefore
i0 G ranite over m uch of its
said to b e hyd ro p h ilic, w
4] length, w hereas the G
hereas others lack this
unflint Form ation was not
tendency and are classified
Figure 13.4 Solubility of intruded by granite.
as h y d r o p h o bic. A sol
ferrous silicates in pure Consequently, he agreed
water at 25 C. com posed of hydrophilic
with Tyler (1949) that m
Greenalite, which is a particles m ay set to form a
innesotaite is a product of
two-layer clay analogous translucent o r tran sp aren t
m eta morphism, which m
to kaolinite and gel. Som e
serpentine, is soluble in ade the Biwabik Form
natural solutions open to ation m ore susceptible to
the atmosphere. It can chem ical w eathering than
pre cipitate at 25 C, only the Gunflint Form ation
when [C 02] = IO7 4 and resulted in the form
atm or less, from ation of hem atite ore
solutions that are nearly bodies in M innesota.
saturated with respect to
amorphous silica.
Greenalite is the most
abundant primary Fe-
silicate in the Gunflint 13.5 Colloid
Iron Formation of
al
Proterozoic age in
western Ontario, Suspen
Canada, and its absence
from the siderite facies sions
implies that the siderite and
facies was deposited
where the fugacity of Ion
C 0 2 was greater than IO
7'4 atm.
Exchan
ge
W hen clay m inerals and o
[Fe.2+1 10
th er sm all particles with
[H + ] 2
[C O J diam eters of < 1 g m are
which yields [COz] = dispersed in pu re w ater,
they form a colloidal
greenalite can suspension called a sol.
precipitate only from Suspensions of sm all
solutions that have very droplets of a liquid th at is
218 CLAY MINERALS
gels can be dispersed again into a sol, w hereas in others Similar reactions m ay take place with atom s of Si, Fe,
the process of gel form ation is irreversible. For example, Mn, etc., on the surface of colloidal particles th at
concentrated solutions of silic acid can polym erize to form contribute to their total electrical charges.
am orphous silica parti cles, and the resulting colloidal The presence of electrical charges on the sur face of
suspension can set to form a gel. H ow ever, the silica gel colloidal particles overcom es their te n dency to settle out
does not redisperse spontaneously, but slowly loses w ater and keeps them in suspension indefinitely unless the
as it crystallizes to form quartz (variety chal cedony) with charges are neutralized by the addition of an electrolyte or
opal as an interm ediate product (see Section 9.6). the passage of an electrical current. In general, colloidal
suspen sions are stabilized in dilute electrolyte solutions
because ions from the solution are needed to build up the
The properties of colloidal suspensions arise from surface charge. H ow ever, at high electrolyte
electrical charges on the surface of the p ar ticles. The concentrations the surface charges of the particles are
presence of electrical charges on col loidal particles is neutralized and the suspension coagulates, th at is, the
related to th eir small size and resultant large surface area. particles settle out.
The atom s on the sur faces of solid particles of all kinds of
m aterials are incom pletely bonded because they are not The polarity of the surface charge of colloidal
sur rounded by ions of the opposite charge. Therefore, such particles depends on the activity of H + in the water.
particles attract ions from the solution in which they are Consequently, we define the isoelectric p oint as the pH of
suspended and thereby acquire a characteristic charge. W the aqueous m edium at which the surface charge of
hen clay m iner als are suspended in w ater, som e of the particles of a specific m ater ial is equal to zero. If the pH
adsorbed interlayer ions m ay be lost, leaving the is less than the iso electric point, the charge is positive by
individual particles with a negatively charged surface. In virtue of equation 13.36; w hereas if the pH is greater than
addition, clay particles also have surface ions with the isoelectric points the charge is negative according to
unsatisfied charges and hence becom e charged not only equation 13.37. The isoelectric points of different kinds of
because of internal crystallographic rea sons but also m aterials in Table 13.8 range from pH = 1.0 (am orphous
because of the surface effects that silica) to pH = 9.1 (A12 0 3, corundum ). The isoelectric
points can be used to deduce the polarity of the surface
charge of colloidal particles of each m aterial in acid and
occur in all small particles. basic media. The data in Table 13.8 indicate that colloidal
The surfaces of silicate, oxide, or hydroxide particles com posed of am orphous silica, quartz, M n-
particles expose atom s of Si, Al, Fe, M n, etc., dioxide, m ontm orillonite, and kaolin ite have a negative
which are bonded to O 2 or O H ' radicals within surface charge, except in high ly acidic environm ents
the solid. H ow ever, O H radicals on the surface caused, for example, by local discharge of acid m ine
exert an attractive force on positively charged ions in the waters. The oxides of Fe form particles whose surface
solution, including H +. Therefore, such particles acquire a charges may change from positive to negative betw een pH
positive charge in strongly acidic solutions by virtue of the val ues 5 and 7. H ow ever, Fe-oxide or hydroxide and
reaction: TiOz particles in the oceans (pH = 8.2) have neg ative
charges. The oxides and hydroxides of Al stand out from
A l O H (_) + H + -> A l O H (2+) (13.36) the others in Table 13.8 because they form positively
w here the parentheses indicate n e t positive or negative charged particles in all but highly basic environm ents.
charges th at are not necessarily equal to one electronic
unit. In strongly basic solutions the O H - radicals m ay
release H +, thereby causing a n et negative charge at the
surface:
It turns out th at the presence of electrolytes in a
A l O H (" } - A l 0 (_) + H + (13.37) colloidal suspension affects the developm ent
13.5 CO LLO IDAL SUSPENSIONS AND ION EXCHANGE 219
Table 13 .8 Isoelectric
Points of Various Natural Materials When
Suspended in Water of Varying Acidity but Lacking Dissolved
Electrolytes
Material Polarity in Polarity in
pH acid medium basic medium
S i0 2(amorph.) 1.0-2.5 _a -
Quartz 2.0 - -
M n02 (different forms) 2.0 -4.5 - -
Albite 2.0 - -
Montmorillonite 2.5 - -
Kaolinite 4.6 - -
Hematite 5 -9 + -t-
commonly 6-7
Magnetite 6.5 + -
Goethite 6 -7 + +
Limonite (Fe20 3 H20 ) 6 -9
Anatase (T i02) 7.2 + -
Gibbsite ~9 + +
Corundum 9.1 + +
Periclase (MgO) 12.4 + +
The polarity changes from negative to positive at pH values less than the isoelectric point. Thus,
the charge is negative if the pH is greater than the isoelectric point. A cid and basic are defined
relative to pH = 7.0 at 25 C.
After Drever (1982),Table 4.2;Stumm and Morgan (1970).
of surface charges of the size or larger, adsorb
particles. Therefore, the p cations (m etals), w
H at which surface charges
hereas positively
of colloidal particles
becom e equal to zero in
charged particles
the presence of elec adsorb anions (non-
trolytes differs from the m etals). Particles of all
isoelectic point in ways th oceans scavenge m etal
at depend on the com transfer them to the sedim
position of the particles ent accum ulating at the
and the charges of the ions bottom of the oceans.
of the electrolyte. The pH Consequently, deep-sea
at which the surface sedim ent is enriched in m
charge of colloidal parti any m etals com pared to
cles in the presence of norm al m arine sedim ent,
electrolytes is equal to which is dom inated by
zero is know n as the zero detrital sedim ent derived
from nearby continen
po in t o f charge (Z P C ).
tal areas (Section 4.5).
The num erical values of
Colloidal particles in
the Z P C depend not only
soils act as tem
on the com position of the porary
colloidal particles but also repositories of ions
on the com position of the released by chemical w
electrolytes in the sus eather ing or by the
pension.
addition of fertilizers. The
resulting capacity of soils
to store plant nutrients
The charges of
enhances their fertility. In
colloidal particles of
addition, the fixation of
different com positions
ions in soils and sedim ent
affect geochem ical
enhances the solubility of
processes in the oceans, in
m inerals and results in the
soils, and in the
purification of groundw
deposition of sedim ent ater.
because negatively
charged particles, of
colloidal
220 CLAY MINERALS
W hen clay m inerals or other particles of col loidal or The capacity of clay m inerals to act as ion
larger size are dispersed in w ater, their surfaces becom e exchangers varies widely, depending at least in part on the
charged and attract ions of the opposite charge, which are charge im balances th at originate within the crystal lattice.
adsorbed by m eans of electrostatic bonds. W ith a few For example, kaolinite commonly has a low cation
exceptions, m iner al surfaces are not particularly selective exchange capacity (C E C ), w hereas m ontm orillonite
and attract ions of different elem ents w ithout m uch and verm iculite have large CECs. The C E C of a sam ple
discrim ination. This m eans that the adsorbed ions are of clay is determ ined by m eans of a standard procedure: 1 0
exchangeable for others in the solution depending prim 0 g of dry clay is repeatedly placed into NaCl solutions at
arily on their activity. The result ing phenom enon of ion pH = 7.0 until all exchangeable cations are replaced by N
exchange can be rep re sented by a simple exchange a+. The N a -saturated clay is then added to concen trated
reaction: KC1 solutions and the am ount of N a + released by the
clay at pH = 7.0 is determ ined by analysis and is
expressed in units of milliequiva-lents per 100 g of dry
AX+ B+ ^ B X +A+ (13.38)
clay. The C E C values of clay m inerals are related to their
which indicated that ion A + on the adsorber is exchanged chem ical com posi tions and therefore vary widely. H ow
for ion B + in the solution. Such exchange reactions reach ever, the val ues in Table 13.9 are fairly typical of the
equilibrium very rapidly and are then subject to the Law of respective clay minerals.
Mass Action; that is:
1BX1[A 0 =K (13 39) The C E C of a clay m ineral can be estim ated from
[A X ][B + ] " 1 the net negative charge of its structural for mula. For
where K cx is the exchange constant. If the exchang er example, the m ontm orillonite rep re sented by form ula
accepts A + and B + with equal ease, then K ex = 1.0. 13.3 has a charge of -0 .4 6 electronic units per form ula
However, in general this is not the case because A + and B weight which is neu tralized by exchangeable Ca2+,
+ do not have the same charge/radius ratio and their Mg2+, and N a+. The C E C of this m ineral caused only
hydrated ions have different radii. M oreover, divalent or by the charge in the layers is:
trivalent ions form stronger electrostatic bonds than
univalent ions and may therefore be adsorbed
preferentially.
E quation 13.39 can be restated in the form:
Table 13 .9 Cation Exchange Capacity
lA ll = ur f ( AX A n (13.40) (CEC) of Various Clay Minerals
[B +] ex\(B X )/
CEC, m eq/100 g
w here (A X ) and (BX ) are the m olar concentra tions of Mineral at pH = 7.0
A and B on the exchanger and n is an Kaolinite 3-15
exponent w ith values betw een 0 . 8 and 2 . 0
Chlorite 10-40?
(B erner, 1971; G arrels and Christ, 1965) . These
Illite 10-40
equations explain why the activities of ions in the aqueous
phase of colloidal suspension are con trolled by ion Glauconite 11- 20+
exchange. For example, if A + is added to a colloidal Palygorskite 20-30
solution containing A + and B + in ion exchange Allophane -70
equilibrium with clay m iner als in suspension, A + will Smectite (montmorillonite) 70-100
displace B + from exchange sites and thereby reestablish Vermiculite 100-150
equilibri um in accordance with Le C hteliers principle.
SOURCE: Garrels and Christ (1965).
13.6 DATING OF CLAY MINERALS 221
0.46 X 100 X 1000 this circum stance by dating detrital m inerals such as
C EC = 36025---------- = m eq/100 g muscovite, feldspar, or zircon to determ ine their p ro ven
(13.41) a n ce by identifying their source by its age (Faure and
Taylor, 1981). However, the age of sedim entary rocks can
This value is probably an underestim ate because it does
be determ ined only by d at ing m inerals th at form ed at
not include charges caused by broken bonds. Nevertheless,
the tim e of deposition (C lauer and C haudhuri, 1995).
the result indicates that 100 g of this clay can adsorb 64.6
mmol of Ca2+, which am ounts to 40.08 X 64.6/1000 = U nfortunately, m ost of the authigenic m iner als in
2.59 g of Ca or 2.97 g of Na. These am ounts are not only sedim entary rocks are not suitable for d at ing by isotopic
substantial, but the presence of N a+, Ca2+, or other m ethods because they do not contain radioactive p aren t
cations has a signifi-gant effect on the mechanical
elem ents in sufficient concentration or because they do
properties of the clay.
not retain the radiogenic daughter isotope quantitatively.
The presence of N a+ in exchangeable positions
These shortcom ings effectively elim inate several authin-
increases the plasticity of clay and can m ake soil
genic m inerals th at com m only occur in sedim en tary
containing such clay unsuitable for agriculture. The best
rocks, including calcite, dolom ite, quartz, gypsum,
soils contain clay m inerals having a mixture of Ca2+, H
anhydrite, hem atite, m agnetite, pyrite, pyrolusite, halite,
+, M g2+, and K + as exchangeable cations. However, in
sylvite, and other evaporite m in erals. However, certain
the ceramic industry, where plasticity is a desirable
clay m inerals are authi genic and do contain K and Rb,
property, clay may be converted to the Na-form by suitable
which m ake them potentially datable by the K - A r and R
treatm ent prior to its use.
b -S r m eth ods described in C hapter 16.
Sm ectite clays expand w hen w ater m olecules
replace exchangeable ions, and slurries of sm ec tite may
set to form a solid that can revert to a liq uid w hen it is G lauconite and illite are best suited for d at ing of
disturbed. This property, called thixotropy, characterizes sedim entary rocks, but only under narrow ly defined
bentonite, which is com posed largely of sm ectite clay circum stances. For exam ple, glaucony
minerals, and m akes it useful as a sealant of building in the form of greenish pellets and encrustations
foundations and as a drilling m ud in the petroleum in m arine sedim entary rocks contains varying
industry. am ounts of K, which stabilizes the lattice by
strengthening the bonds betw een clay units. A s a result,
the retentivity of glaucony with respect to radiogenic 40A
r im proves with increasing K con
13.6 Dating of Clay Minerals ten t and the m ineral becom es closed to 40A r
O ne of the prim ary tasks in the E arth Sciences is to w hen KzO > 6.0% (O din and M atter, 1981).
establish an accurate chronology of the history of Therefore, only glauconitic micas w ith the
required K concentration can yield reliable dates by the K
the E arth, which is recorded in the sedim enta ry
-A r m ethod. The selection criteria for dating glaucony by
rocks th at have survived to the Present. This has the R b -S r m ethod are not as well defined, partly because
turned out to be a very difficult task because radiogenic 87Sr is less
sedim entary rocks are generally not datable by m obile than 4 0 A r. H ow ever, both m ethods are
m ethods based on radioactivity. One of several also sensitive to gain or loss of the radioactive
reasons for this difficulty is that sedim entary p arent elem ent so that w ell-crystallized glau
rocks are com posed prim arily of detrital particles conitic m ica is also preferable for dating by the
of m inerals th at crystallized to form igneous or R b -S r m ethod. H ow ever, even well-crystallized
m etam orphic rocks long before they w ere incor glauconies may be appreciably altered by th e r
porated into clastic sedim entary rocks, such as m al m etam orphism , tectonic deform ation, and
sandstones or shales. O ne can m ake a virtue of
222 CLAY MINERALS
ion exchange w ith brines in the subsurface.
in the tetrahedral layer Si4+ is partially replaced by A l3+.
T herefore, glauconies th at rem em ber their age The charge im balance th at results from this substitution in
are valuable exceptions rath er than the rule the tetrahedral layer may be partly com pensated by
(O din, 1982). additional cations in the octahedral layer. How ever, the net
The application of the R b -S r m ethod to negative charge arising from either the tetrahedral or th e
dating clay-rich sedim entary rocks has been octahedral layer, or both layers, is neutralized by cations
sum m arized by C lauer (1982). The w hole-rock that are adsorbed to the outer surfaces of the terahedral
isochron m ethod som etim es yields useful results (C layer and thereby bond individual clay units to each other.
ordani et al., 1985), especially in unfossilifer-ous shales of Potassium ions are espe cially effective in this regard
Precam brian age containing large am ounts of accum because they fit into the hexagonal rings of silica
ulated radiogenic 8 7 Sr. In gen tetrahedra th at m ake up the tetrahedral layer.
eral, how ever, it is b e tte r to use m ethods that Consequently, K -bearing clays like illite have short
rely on the properties of a specific authigenic interlayer spacings and do not expand in water. Clays that
m ineral, such as illite. M orton (1985) dem on have adsorbed N a+, Ca2+, or M g2+ ions have wider
strated th at coarse-grained illite (1 - 2 //.m) in interlayer spacings than K-clays and expand in water
U p p er D evonian black shales of Texas is com because the inter layer cations are displaced by water
posed of the 2M polym orph of detrital origin molecules.
and yielded a provenance d a te by th e R b -S r
m ethod th at p red ates the tim e of deposition . The chemical diversity of clay m inerals m akes it
H ow ever, lM d illite in the <0.2-/u,m fraction in difficult to ascertain their therm odynam ic prop erties.
this case dated the tim e of diagenetic alteration Several em pirical schemes have been p ro posed to
of the illite by m igratingbrines. The fit of data overcom e this difficulty, including one by
points to a R b -S r isochron was dram atically Tardy and G arrels (1974). Their m ethod for esti
im proved by treating fine-grained lM d illites m ating standard free energies is based on the
w ith am m onium acetate, which rem oved R b + oxides and hydroxides that m ake up the layers of two- or
and S r2+ from exchangeable sites. three-layer clays and on the adsorbed ions, including H +.
However, in order to construct m in eral stability diagram s
of general validity, stan dardized smectite-type clay m
inerals are used. The results indicate that prim ary mica m
13.7 Summary inerals (m us covite, biotite, and phlogopite) alter to a
variety of clay m inerals depending on environm ental
Clay m inerals are classified according to their crystal condi tions. M ineral stability diagram s can be used to
structure into platy clays, fibrous clays, and x-ray am understand natural geochemical processes such as the form
orphous clays. The platy clays are com posed of sheets of ation of greenalite in cherty iron form a tions and the
silica tetrahedra joined at their bases (tetrahedral layer, T) incompatibility of greenalite and siderite.
and Al hydroxide octahedra joined by sharing hydroxyl
ions at the corners (octahedral layer, O ). The phyllosilicate
clays are m ade up of tw o-layer (T -O ) and three-layer (T Clay m inerals and other small particles can form
- O - T ) packages. The chemical diversity of clay m colloidal suspensions in dilute electrolyte solutions
inerals results from com plete or partial replacem ent of A because of electrical charges on their surfaces. Such
l3+ in the octahedral layer and of Si4+ in the tetrah ed ral suspensions may be stable for long periods of tim e but are
layer. coagulated by an increase in the concentration of
electrolytes. The polarity of the electrical charges of
In the octahedral layer, A l3+ can be replaced by colloidal particles of different com positions depends on
Fe3+, Fe2+, M g2+, Z n2+, Li+, and C r3+, w hereas the pH of the
PROBLEMS 223
solution. Clay m inerals, quartz, feldspar, and M n oxides capacity (C E C ), which can be m easured by m eans of a
are negatively charged except in strongly acidic environm standardized procedure . T he physical properties of clays
ents with pH < 2.0. The various oxides and hydroxides of are strongly affected by the kinds of adsorbed cations.
Fe can have positive or negative surface charges in the pH
range from 5 to 9, w hereas A l oxides and hydroxide C ertain authigenic K -bearing clay m inerals are
particles have positive charges except at pH > 9. dateable by the K -A r and R b -S r m ethods and therefore
provide rare opportunities to determ ine the ages of sedim
C ations th at are adsorbed on the charged surfaces of entary rocks. How ever, only clay samples whose chemical
small particles in contact with dilute electrolyte solutions com position, crystallinity, or grain size has been evaluated
are exchangeable for ions in the solution. These cation should be used, and the resulting dates may be affected by
exchange reactions can achieve a state of equilibrium th a t therm al m etam orphism , structural deform ation, or ion
causes w ater-particle m ixtures to respond to changes in exchange with circulating brines. D ating of sedi m entary
rocks by m eans of clay m inerals requires a thorough
w ater com positions in accordance w ith Le understanding of clay m ineralogy as well as of
C h teliers principle. The ability of sm all p a rti geochemistry.
cles of different m inerals to adsorb exchange
able ions is expressed by its cation exchange
Problem
s
1. Determine the
Component %
electrical charges of the
S i0 2 45.20
octahedral and tetrahedral
layers and check whether a i2o 3 37.02
the exchange able ions Fe20 3 0.27
balance the charge in the
FeO 0.06
following clay mineral.
MgO 0.47
[ (Fe j 3 4 5 Mg0 595Mn CaO 0.52
0.004)
aH20 is water removed
(Si3.g36^0.112^e0.05l) at 105 C, whereas HzO +
lo(H)2]Ko- is lost only at high
779Ca0.076 temperature.
a i 2o 3 17.64 H20+ 10.30 8. Calculate the solubility product constant for hec-
Fe20 3 8.70 Sum 99.57 torite at 25 C, assuming congruent solution based on
FeO 37.96 the formula
6. Estimate the standard free energy of formation of [(Mg2.6? Li 0.33) (Si 4) 0 10(OH)2]Na0 33
[(^1.50^go.34Fe0.17pe0.04)
(S*3.83AIO.I7)OIO(OH)2]K(M0
References
G R I M , R. E., 1968. Clay
Minerology, 2nd ed. McGraw-
B E R N E R , R. A., 1971. Principles Numerical Dating in Hill,
o f Chemical Sedimentology. Statigraphy, vol 1, 245-276. New York, 596 pp.
McGraw-Hill, New York, Wiley,
240 pp. Chichester, England, 630 pp.
B R A D L E Y , W. F., 1940. The C L A U E R , N., and S. C H A U D H U R I
structural scheme of Environments; Isotopic Dating
attapulgite.
Verlag, Heidelberg, 359 pp.
Amer. Mineral., 25:405^110.
C O R B A T O . C. E and R. T. T E T T
B R IN D L E Y , G. S., 1961. Kaolin, E N H O R S T , 1987. Analysis of
H ey, M. H., 1954. New review of the chlorites. Mineral. Mag., Gibbs energies of formation of layer silicates. Geochim.
30:277-292. Cosmochim. Acta, 38:1101-1116.
H O W E R , J., 1961. Some factors concerning the nature and ori T A R D Y , Y., and R. M. G A R R E L S , 1976. Prediction of Gibbs energies
gin of glauconite. Amer. Mineral., 46:313-334. of formation I. Relationships among Gibbs energies of
James, H. L., 1966. Chemistry of the iron-rich sedimentary rocks. formation of hydroxides, oxides and aqueous ions. Geochim.
In M. L. Fleischer (Ed.), Data of geochemistry. U. S. Cosmochim. Acta, 40:1051-1056.
Geol. Surv. Prof. Paper, 440-W; 60 pp. T A R D Y , Y., and R. M. G A R R E L S , 1977. Prediction of Gibbs energies
J O H N S T O N , J. H ., and C. M. C A R D I L E , 1987. Iron substitution in of formation of compounds from the elements
montmorillonite, illite, and glauconite by 57Fe Mssbauer II. Monovalent and divalent metal silicates. Geochim.
spectroscopy. Clays and Clay Minerals, 35:170-176. Cosmochim. Acta, 41:87-92.
K I T T R I C K , J. A., 1971a. Stability of montmorillonites: I. Belle T A R D Y , Y., and L . G A R T N E R , 1977. Relationships among Gibbs
Fourche and Clay Spur montmorillonites. Soil Sei. Soc. energies of formation of sulfates, nitrates, carbonates, oxide,
Amer. Proc., 35:140-145. and aqueous ions. Contrib. Mineral. Petrol, 63:89-102.
K I T T R I C K , J. A., 1971b. Stability of montmorillonites: II. T A R D Y , Y., and P. V I E I L L A R D , 1977. Relationships among Gibbs
Aberdeen montmorillonite. Soil Sei. Soc. Amer. Proc., free energies and enthalpies of formation of phosphates,
35:820 -823 . oxides and aqueous ions. Contrib. Mineral. Petrol.,
L IN D S A Y , W. L ., 1979. Chemical Equilibira in Soils. Wiley, New 63:75-88.
Y o rk , 449 pp . T A R D Y , Y., 1979. Relationships among Gibbs energies of for
M C R A E , S. G., 1972. Glauconite. Earth Sei. Rev., 8(4):397^140. mation of compounds. Amer. J. Sei., 279:217-224.
M ILLOT , G., 1970. Geology o f Clays. Springer-Verlag, Berlin, 429 T Y L E R , S. A., 1949. Development of Lake Superior soft iron ores
pp. from metamorphosed iron formation. Geol. Soc.
M O R T O N , J. P., 1985. Rb-Sr dating of diagenesis and source age of Amer. Bull, 60:1101-1124.
clays in Upper Devonian black shales of Texas. Geol. Soc. Amer. V E L D E , B., 1977. Clays and Clay Minerals in Natural and
Bull., 96:1043-1049. Synthetic Systems. Elsevier, Amsterdam, 218 pp.
N R I A G U , J. O., 1975. Thermochemical approximations for clay W E A V E R , C. E., 1956. The distribution and identification of
minerals. Amer. Mineral., 60:834-39. mixed-layer clays in sedimentary rocks. Amer. Mineral.,
O D I N , G . S. (E d .), 1982. Numerical Dating in Stratigraphy, vols. 41:202-221.
1 and 2. Wiley, Chichester, England, 1040 pp. W EAVE R , R . M ., M . L . JA C K S O N , a n d J. K . SY E R S , 1971.
O D T N , G . S., a n d A . M A T T E R , 1981. D e g la u c o n ia ru m o rig in e . M ag n e siu m a n d silico n activ itie s in m a trix s o lu tio n s o f
Brindley and G. Brown (Eds.), Crystal Structures o f Clay muscovites. Geochim. Cosmochim. Acta, 8:225-280.
Minerals and Their X-ray Identification. Mineral. Soc. London, Zen, E -A n, 1967. Mixed-layer minerals as one-dimensional
Monograph 5,495 pp. crystals. Amer. Mineral., 52:635-660.
R O U T S O N , R. C., and J. A. K I T T R I C K , 1971. Illite solubility. Soil Zen, E -A n, 1972. Gibbs free energy, enthalpy and entropy of ten
Sei. Soc. Amer. Proc., 35:714718. rock-forming minerals: calculations, discrepancies,
S IN G E R , A., and E. G A L A N (Eds.), 1984. Palygorskite-Sepiolite. implications. Amer. Mineral., 57:524-553.