Chapter 3

Methodology
3.1 General
An experimental program was designed to study the corrosion activities of
reinforcing steel in concrete. Also basic mechanical properties such as compressive
strength were studied. The corrosion activities were investigated by half-cell potential
measurement (HCP) and macro-cell current density measurement (icorr). In addition,
carbonation depth and total chloride content were applied to recheck the corrosion process.
Three different environments were selected for this study. First environment was
carbon dioxide. Concrete specimens were placed in a carbonation chamber having 4%
concentration of carbon dioxide, 40C of temperatures and 555% of relative humidity.
Second environment was chloride attack. Concrete specimens were exposed to 10%
concentration of NaCl solution. For the last environment, specimens were exposed to both
chloride and carbonation alternatively. In this environment, concrete specimens were
exposed to cycles of carbonation and chloride. Each cycle contained 4 days of carbonation
and then 3 days of chloride attack. In addition, examples of application to study corrosion
activity of actual RC structures (concrete walls), which were exposed to marine
environment, were also described. The concrete walls were tested by non-destructive
method and destructive method. All of experimental details were presented in the
following sections.
3.2 Materials
3.2.1 Ordinary Portland cement, fly ash and limestone powder
Ordinary Portland Cement ASTM Type I, manufactured in Thailand by Siam
Cement Group Public Company Limited (Elephant Brand) was used for all mixtures. The
fly ash obtained from Mae-Moh power plant in Thailand was used in all mixtures of the
experiments. This fly ash is a product of lignite after burning process as fuel in the plant. It
is classified as class 2b fly ash according to TIS 2135-2545 (calcium oxide >10%). The
utilized limestone powder was supplied by Surint Omya Chemicals (Thailand) Co., ltd.
The test results of chemical compositions and physical properties of these materials are
shown in Table 3.1.
3.2.2 Aggregates
Natural river sand obtained from natural river, passing a ASTM C33-92a sieve No.
4 (4.75-mm openings), was used as the fine aggregate for all concrete mixes. Its specific
gravity and water absorption were 2.58 g/cm3 and 1.01%, respectively. The crushed
limestone, of which maximum size is 19 mm, was used as the coarse aggregate. It was
18

obtained from Saraburi province with specific gravity of 2.69 g/cm3 and water absorption
of 0.25%. It was retained on the sieve No.4 (4.75-openings) and cleaned before being used.
Table 3.1 Chemical compositions and physical properties of cement,
fly ash and limestone powder
Chemical compositions

Cement

Fly ash

Limestone
powder

Silicon dioxide (%)

20.48
43.88
Aluminum oxide (%)
5.25
24.31
Iron oxide (%)
3.82
12.51
Calcium oxide (%)
65
11.18
Magnesium oxide (%)
0.95
2.73
Sulfur trioxide (%)
1.9
1.43
Insoluble residue (%)
0.13
Sodium oxide (%)
0.01
0.78
Potassium oxide (%)
0.4
2.78
Titanium dioxide (%)
0.25
0.47
Phosphorus pentaoxide (%)
0.05
0.13
Free Lime (%)
0.75
Gypsum Content (%)
5.6
Physical Properties
Specific gravity (g/cm3)
3.15
2.08
Loss on ignition (%)
1.17
0.17
2
Blaine Fineness (cm /g)
3350
3460
Water requirement (%)
100
95.6
Bogues potential compound compositions
C3S (%)
68.13
C2S (%)

7.41

C3A (%)

7.46

C4AF (%)

11.61

0.06
0.09
0.04
54.80
0.57

2.70
44.44
9260
-

3.2.3 Water
Ordinary tap water supplied at concrete laboratory, Sirindhorn International
Institute of Technology, Thammasat University, was used for all concrete.
3.3 Mix proportion
In this study, the cementitious materials used in concrete were ordinary Portland
cement, fly ash and limestone powder. The test parameters were classified by water to
binder ratio, replacement ratio of fly ash and limestone powder. Different fly ash content
(0, 20, 30 and 50%) and limestone powder (0, 5, 10, and 15%) with different w/b (0.4, 0.5
and 0.6) were used in this study. In addition, some mix proportions were designed by using

19

blended binders i.e. cement, fly ash and limestone powder. The details of mix proportions
of concrete are given in Table 3.2.
The mix proportions of 27 concrete samples are shown in Table 3.2. An example of
description of the mixture designation is as follow; 0.4L10F20 means the sample which
has water to binder ratio of 0.4, limestone powder replacement ratio of 10% and fly ash
replacement ratio of 20%.
Table 3.2 Mix proportions tested in this study
NO.

Designation

w/b

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27

0.4OPC
0.4FA20
0.4FA30
0.4FA50
0.5OPC
0.5FA20
0.5FA30
0.5FA50
0.6OPC
0.6FA20
0.6FA30
0.6FA50
0.4OPC
0.4LP5
0.4LP10
0.4LP15
0.4L10F20
0.5OPC
0.5LP5
0.5LP10
0.5LP15
0.5L10F20
0.6OPC
0.6LP5
0.6LP10
0.6LP15
0.6L10F20

0.4
0.4
0.4
0.4
0.5
0.5
0.5
0.5
0.6
0.6
0.6
0.6
0.4
0.4
0.4
0.4
0.4
0.5
0.5
0.5
0.5
0.5
0.6
0.6
0.6
0.6
0.6

OPC
(kg/m3)
454
348
298
204
399
307
263
181
355
274
236
163
454
430
406
382
306
399
378
357
336
270
355
337
318
299
242

FA
(kg/m3)
0
87
128
204
0
77
113
181
0
69
101
163
0
0
0
0
87
0
0
0
0
77
0
0
0
0
69

LP
(kg/m3)
0
0
0
0
0
0
0
0
0
0
0
0
0
23
45
67
44
0
20
40
59
39
0
18
35
53
35

Water
(kg/m3)
182
174
170
163
199
192
188
181
213
206
202
196
182
181
180
180
171
199
199
198
197
189
213
213
212
211
203

Sand
(kg/m3)
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720
720

Note: w/b is the water to binder ratio; OPC is ordinary Portland cement, FA is fly ash
and LP is of limestone powder

20

Gravel
(kg/m3)
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032
1032

3.4 Sp
pecimen preeparation
R
g steel
3.4.1 Reinforcing
Deformed bars having
g grade SD40 according to TIS 24 ((Thai Industtrial Standard)
with diameter
d
of 12
1 mm and length
l
of 310 mm were used as reinnforcing steeels. They weere
polished by using a wire brushh to remove rust shown on the surfaace and thenn degreased by
b
w
by uusing insulatting tape on both ends so that the 2110
acetonne. All steel bars were wrapped
mm ceenter portion
n of the bar was exposedd. Both tapee-insulated ennds of the stteel bars weere
coatedd again by ep
poxy as show
wn in Figuree 3.1

Figure 3.1
1 Reinforcingg steel used in the specim
mens.
3.4.2 Specimen
S
foor chloride environmen
e
nt
Shape of specimens used
u
for chhloride attacck was moddified from ASTM G1009
which was standarrd test method for determ
mining the effects
e
of chhemical adm
mixtures on thhe
r
nt in concreete exposed tto chloride environment
e
ts.
corrossion of embeedded steel reinforcemen
Specim
men dimensiions are 95 mm
m 200 m
mm 250 mm
m. Three reiinforcing steeel installed in
concreete which diifferencing cover
c
thicknness. One baar put at thee top specim
men. Two baars
put at the bottom specimens which weree placed 80 mm below the top bar.. In additionns,
a 50 mm. The top andd bottom baars
cover thickness off the top barr was variedd as 30 mm and
w
lead w
wires and a 100 resistor. At the top surfacce
were electrically connected with
specim
men was cleeaned for innstalling plaastic dam. The
T plastic dam with dimension of
150mm
m 65mm was placed on the top oof specimenn. The plastic dam and specimen
s
was
attacheed by water proof silicoone adhesivee. The dimennsion details of concrete specimen are
a
given in Figure 3.22.
Pondin
ng by NaCl solutioon
with 10% Cl- by weighht

steeel bar

Ressistor
steeel bar
C
Concrete
95 mm
m
250 mm
m
matic illustrattion of specimen exposed to chloridee.
Figurre 3.2 Schem

21

3.4.3 Specimen
S
foor carbon diioxide envirronment
Dimension
ns of prism specimens
s
e
exposed
to carbon dioxidde are 200 mm
m 200 mm
m
250 mm.
m Concreete cover of steel
s
was varried at 20 mm
m, 30 mm, 440 mm and 50
5 mm. At thhe
centerr of each speecimen, a refference bar w
was placed. Each reinfoorcing bar was electricallly
conneccted with the reference bar
b by usingg lead wires and a 100 resistor as shown in Fig
3.3.
In additionn, some conccrete specim
mens were deesigned with 10 mm of concrete
c
covver
bonation wass expected to be slow. Size
S of concrete specimeen
thicknness because rate of carb
is sim
milar with specimen exxposed to teesting chlorride environnment. Figurre 3.4 show
ws
schem
matic illustrattion of specimen exposed to carbonaation for 10 mm
m of coverr thickness.

Figuure 3.3 Schem

matic illustraation of speccimen with 20,
2 30, 40, annd 50 mm stteel coveringg
thhickness expposed to carbbonation

Fiigure 3.4 Schhematic illusstration of sppecimen with 10 mm steeel covering thickness

exposedd to carbonattion
3.4.4 Specimen
S
foor combined
d CO2 and Cl
C - environm
ment
Concrete specimen
s
forr combinnedd CO2 and Cl
C - environm
ment is show
wn in Fig 3.5.
Concrrete specimeen exposing to both caarbon dioxid
de and chloride was deesigned to be
b
similar to the speecimen for carbonationn. But the only
o
dissimillarity is thee existence of
m 65 mm
m that was placed on the
t top of thhe
plasticc dam with the dimensiion of 150 mm
specim
men.

22

Figure 3.5 Schematic illustration of specimen with 10 mm concrete cover for combined
carbonation and chloride exposure.

3.5 Exposure conditions

All specimens for testing HCP and icorr in all exposure condition were demolded 24
hours after casting and were cured by covering with wet clothes for another 27 days in the
laboratory atmosphere. After removing the specimens from the curing room, they were
immediately exposed to the test conditions. Three major categories of exposure conditions
are classified as individually ponding with solution, individually exposed to carbonation
and combined cyclic exposure between ponding with chloride solution and exposure to
carbonation.
(1) In case of carbon dioxide, specimens were exposed in a carbonation chamber in which
the conditions were controlled at 4 % carbon dioxide concentration, 40 C, and 555%
RH. Figure 3.6(a) shows arrangement of specimens exposed to accelerated carbonation
condition in the carbonation chamber.
(2) In case of chloride attacks, these groups of specimens were continuously ponded with
solution that has concentrations of chloride of 10% NaCl by weight. This solution was
used as the continuous ponding solution above the top surface of every specimen.
Normal air condition was controlled at 30C and 555% RH. Please be noted that
chloride concentration of normal sea water in Thailand is approximately 3% NaCl by
weight. Figure 3.4(b) shows condition of specimens exposed to chloride solution
(3) In case of combined cyclic exposure between ponding with Cl- solution and exposing
to carbonation. Specimens were firstly placed in carbonation chamber for 4 days, and
then ponded with chloride solution for 3 days to complete one cycle.

23

(a) Speecimens in carbonation

c
c
chamber

(b) Pondingg with chloride solution

Figure 3.6 E
Exposure connditions
3.6 Methods of teesting
3.6.1 Compressiv
C
ve Strength
In this stud
dy, the cylinnder specimeens with thee size of 1000x200 mm were used for
f
testingg compressiv
ve strength. This test coonforms to ASTM
A
C39//C39M-99 (Standard Teest
Methood for Com
mpressive Strength
S
of Cylindricaal Concrete Specimenss). After thhe
specim
mens were cured for a period of 7, 14, 28 and 91 dayys, they weere tested for
f
comprressive strenngth. Three specimens
s
w
were used forr each mixtuure and the result
r
was thhe
averagge of the two
o measured values.
v
3.6.2 Carbonation
C
n Tests
The rate of
o carbonatiion in real environmen
nt is usuallyy slow due to low CO
O2
concenntration in the
t environm
ment. Acceleerated carboonation test was conduccted. Concreete
cube specimens with the size
s
of 1000100100 mm were used in the accelerateed
carbon
nation test. The
T concretee specimens for testing carbonation depth were demolded 24
2
hours after castinng and weree water-cureed for 27 days
d
before being carboonated in thhe
acceleerated carbonnation chambber inventedd in this studyy for a periood of 4, 8 and
d 12 weeks.
The teemperature and
a relative humidity
h
in the carbonattion chambeer were contrrolled at 40C
and 555 5 %, resspectively. The
T carbon ddioxide conccentration waas 4% (40,0
000 ppm). Thhe
carbon
nation chambber used in this
t study is shown in Fiig 3.7.
At the agge of measuurement as shown in Fig
F 3.8, specimens weere tested for
f
carbon
nation depthh in the laborratory. The specimens were
w
split annd cleaned. The
T depths of
carbon
nation weree determined
d by spraying on a freshly
f
brokken surface with 1% of
phenolphthalein in
n the solution of 70% ethhyl alcohol (RILEM, 19988). The ph
henolphthaleein
on is colorleess and usedd as an acid-bbase indicattor. The coloor of the solu
ution changes
solutio
into pu
urple when pH is higheer than the raange of apprroximately nnine. Thereffore, when thhe
solutio
on is sprayeed on a brooken concrete surface, the carbonaated portion is uncoloreed
(concrrete color) annd non-carbonated portiion is purplee. The depth of carbonattion is defineed
as the thickness of the uncoloored carbonaated portion. The averagge depths froom 8 points of
nation were measured
m
annd reported.
carbon

24

Fiigure 3.7 Caarbonation chhamber for carbonation

c
ddepth test

Figure 3.8 Broken surffaces of speccimens after spraying witth phenolphtthalein

3.6.3 Determinati
D
ion of chlorride content
A
C11552
Total chloride contentts in concreete were dettermined acccording to ASTM
after the
t specimenns has been ponded. Fivve grams off powdered concrete
c
at the
t surface of
reinforrcing bar was
w collected
d. Then, the sample of powdered concrete
c
wass dissolved in
HNO3 (nitric acid) solution, and
a automatiically titrated
d against N//20 AgNO3 (silver
(
nitratte)
of
solutio
on as shown
n in Fig 3.9. Total chhloride conteent was calculated as percentage
p
chloridde ion by weeight of the powder
p
sampple.

25

Figure 3.9 Chloride meeasurementss (Potensiom

metric titration)

Figure 3.10 Half-cell potentiaal instrumentts

H
potential measurement
3.6.4 Half-cell
CP
The corrosion potentiial of reinfforcing steell was meassured by meeans of HC
measu
urement. Thhe HCP vallue indicatees the probaability of ccorrosion acctivity of thhe
reinforrcing steel located
l
beneeath the halff cell, as deescribed by ASTM
A
C 87
76. The setuup
basicaally consists of an exterrnal copper//copper sulfaate (Cu/CuS
SO4) electrodde (half celll),
conneccting wires and
a a high im
mpedance vooltmeter. Thhe half-cell ppotential meaasurement has
been widely
w
used in the field due to its ssimplicity. General
G
agreement amonng researcheers
have been
b
reachedd that it effectively indiccates the proobability of aactive corrossion along thhe
steel reinforcemen
r
nt in concrette. In case off chloride ex
xposure, speecimens weree air-dried for
f
1 hourr before the HCP measuurement. In case of carrbonation, sppecimens weere wetted by
b
coveriing with weet clothes for
fo 12 hourss to increasse the moistture in concrete because
enviro
onment in caarbonation chhamber was very dry. Measurement
M
was conduccted on the toop
surface at every 20
2 mm alon
ng the direction of top reinforcing bar. HCP was
w measureed
o chloride exposure
e
andd every weekk for carbonn dioxide expposure.
every day in case of

26

3.6.5 Corrosion
C
c
current
denssity measurrement
Actually, corrosion
c
wiill be firstlyy initiated ass micro-cell corrosion between anodde
n the top reiinforcing barr. However, measurement of micro--cell corrosioon
and caathode within
curren
nt is very diifficult. In thhis study, oonly macro-ccell corrosioon current was
w measureed.
Macroocell corrosioon current was
w a good iindicator of corrosion innitiation of th
he reinforcinng
steel. Corrosion current
c
(I) was
w measured indirectly as a voltagge drop acrooss the 100
b
reinfforcing barss as shown in
resistoor connectedd externally between thhe top and bottom
Figuree 3.11. The corrosion
c
currrent is calcuulated basedd on Ohms law as shownn in Eq (1).

V = IR

(3.1)

where V is the volltage drop accross the ressistor (V), I is the curreent due to coorrosion (A
A),
R is a resistor (1000 in this stu
udy). Voltm
meter is a high
h impedancee type with a resolution of
0.01 mV
m as recom
mmended by ASTM G
G109. Then, corrosion ccurrent denssity (icorr) was
calculaated as show
wn in Eq (2)..
I
(3.2)
icorr =
A
where icorr is the corrosion
c
cuurrent densityy (A/cm2), and A is thee surface areea of exposeed
reinforrcing steel. In this study
y, exposed ssteel has thee surface areea of 79.1688 cm2 (surfacce
area of the center portion
p
with
hout epoxy coat).

Figgure 3.11 Scchematic illuustrations weere testing thhe voltage.

n site half-ccell potential measurem
ment of reinfforced concrete structu
ures
3.7 On
mental invesstigation on a reinforceed concrete to
This sectioon describess an experim
m
nt carried out
o from weetting condiition to those
compaare half-celll potential measuremen
obtainned from dryying conditio
on. Three meeasured build
ding locationns, all reinfoorced concreete

27

wall members,
m
weere investigaated. First loocation (SD22-R) is a parrt of concrete wall whicch
was reepaired by cooncrete. Seco
ond locationn (SD2-C) is the existingg unrepaired wall (same as
that of
o the first location butt without reepair). Finall location (SP3-L)
(
is a newly built
concreete wall. Thhree walls are
a located 112 km from
m sea. The tthree measuured walls are
a
shownn in Figs 4.122 and 4.13
Half-cell potential
p
vaalues, concreete cover thhickness annd chloride content weere
mined to indiicate states of
o corrosionn. Covering meter
m
was used to determ
mine concreete
determ
cover thickness annd locationss of steel baars inside thee concrete. H
Half-cell po
otential testinng
t determinee the corrosiion activity of steel barrs. However, HCP testinng
(HCP)) was used to
was conducted affter wetting for 12 houurs and 1 hour
h
(duratioon of wettinng was a teest
meter). In add
dition, reinfo
forced concrete wall meembers were drilled by a drill gimleet.
param
The deepth of drilliing to take concrete pow
wder sampless for chloridee content anaalysis is everry
10 mm
m up to 50 mm
m from the exposed surrface.

Figuree 3.12 Locatiions of SD2--C and SD2--R

Figure 3.13 Locations of SP3-L

28