3.. n3? . ..
$3132.12...

.5. .
t5ﬁ??:
x3 .x,
a: I
Z

mama.- w‘I.

v

.0
‘
.x
o

.I-
I35,

.x...t
. q. .vvnvns
3...,

 

Z. 2.) A
:1. . $3.1: 7:. _
. 1 is . 211.425.-

 

 

 

 

 

MOCHIGAN TATE UNIV

l'L“llll’ll'lllllllllllll‘l'

THESlS 1 r'
7

 

 

 

 

 

 

This is to certify that the

dissertation entitled

VALUE-ADDED USE OF CLEAN COAL TECHNOLOGY
BY-PRODUCT IN CONCRETE

presented by

Jer-Wen Hsu

has been accepted towards fulﬁllment
of the requirements for

 

 

Ph.D. degreein Civil and Environmental
Englneerlng
‘@ SNvA

 

Major professor

Date 8/3/1995

 

MS U i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

LIERARY
Michigan $tate
1 University

 

 

 

PLACE N RE‘I’URN BOX to roman thlo chockout from your rocord.
TO AVOID FINES rotum on or before doto duo.

DATE DUE DATE DUE DATE DUE

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l I

MSU IoAn Afﬁrmative Action/Emu Opportunity IWIOH
W

 

 

 

 

l

 

 

VALUE-ADDED USE OF CLEAN COAL TECHNOLOGY BY—PRODUCT
IN CONCRETE

By

J er-Wen Hsu

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Civil and Environmental Engineering

1995

ABSTRACT

VALUE-ADDED USE OF CLEAN COAL TECHNOLOGY BY-PRODUCT
' IN CONCRETE

By

J er-Wen Hsu

Coal ﬂy ash and clean coal technology by-products represent major volumes of solid
wastes disposed of in landﬁlls. Environmental concerns have led to increased adoption of
the clean coal combustion technologies in power plants, substantially increasing the
volumes of clean coal technology by-products. This research evaluates the by-products
of the calcium spray dryer clean coal technology as a partial replacement for Portland
cement in concrete. This by-product possesses higher calcium and sulfur oxide contents,
and lower silicon, aluminum and iron oxide contents when compared with the

conventional Class F coal ﬂy ash, which is commonly used in concrete.

The clean coal technology by-product, when replacing 15 ~ 30% by weight of cement in
concrete, produced concrete materials with performance characteristics comparable to
those obtained with conventional coal ﬂy ash. The clean coal by-product concrete
possessed better scaling resistance than ﬂy ash conventional concrete. When compared
with plain concrete, the clean coal by-product concrete had comparable (or slightly

reduced) 28-day compressive strength, improved chloride permeability, and comparable

scaling resistance and freeze-thaw durability. The clean coal technology by-product
concrete, like ﬂy ash concrete, responded better than plain concrete to extended moist

curing periods.

Capillary pore system characteristics were found to be a fundamental property explaining
the effects of the clean coal technology by-product in concrete. There was a general
reduction in capillary porosity, particularly after longer moist curing periods, in the
presence of the clean coal technology by-product. The pozzolanic reaction of the by-
product seemed to block the pores and thus reduce their continuity, which explains the
drop in permeability in the presence of the by-product. There was a marked effect of
prolonged moist curing on the total capillary porosity and particularly on the volume of
larger pores. This explains the highly beneﬁcial effects of extended moist curing on the
engineering prOperties‘ of the clean coal technology by-produet concrete. It is my general
conclusion that the by-product of calcium spray dryer offers a strong potential for use as a

mineral admixture with positive effects on permeability and thus durability of concrete.

ACKNOWLEDGMENTS

I am deeply indebted to Dr. Parviz Soroushian (the chairperson of guidance committee)
for his constant direction, instruction, supervision and encouragement during the course
of this work, and to the member of guidance committee, Dr. Mark B. Snyder, Dr. Robert

Wen and Dr. Richard Hill for their invaluable advice, guidance and inspiration.

I also thank my parents, Mr. Chi-Hsiao Hsu and Ms. Hsueh-Ru Liou, for their belief in
education and encouragement. This gratitude is extended to Chi-Fan Hsu, Shang-Shian

Hwang, and Siavosh Ravanbakhsh.

Finally, I wish to express my greatest appreciation to my wife, Chi-Pei Chow, for her

sacriﬁce, encouragement and support during my work.

TABLE OF CONTENTS

Page
LIST OF TABLES .................................................................................................. vii
LIST OF FIGURES ................................................................................................ ix
LIST OF ABBREVIATIONS ................................................................................ xii
CHAPTER 1 INTRODUCTION ........................................................................... 1
CHAPTER 2 FLY ASH EFFECTS ON THE SCALING RESISTANCE OF
CONCRETE: A LITERATURE REVIEW ................................. 3
2.1 Introduction ................................................................................ 3
2.2 Scaling Mechanisms ................................................................... 4
2.3 Fly Ash Effects ........................................................................... 6
2.4 Conclusions ................................................................................ 16

CHAPTER 3 FACTORS EFFECT THE AIR VOID SYSTEM ON THE
POZZOLANIC MATERIALS: A LITERATURE REVIEW 18

3.1 Introduction ................................................................................ 18
3.2 Background ................................................................................. 20
3.3 Literature Test Procedure and Results ........................................ 22

CHAPTER 4 CLEAN COAL TECHNOLOGY BY-PRODUCT S: A

LITERATURE REVIEW .............................................................. 40
4.1 Introduction ................................................................................ 40
4.2 Clean Coal Technologies ............................................................ 41
4.3 Clean Coal Technology By-Products ......................................... 52
4.4 Potential Applications ................................................................. 56
4.5 Utilization in Concrete ................................................................ 57

CHAPTER 5 CONCRETE MICROSTRUCTURE: A LITERATURE

REVIEW .......................................................................................... 66
5.1 Introduction ................................................................................ 66
5.2 Pore Structure ............................................................................. 66
5.3 Capillary Pore System Characteristics ........................................ 68
CHAPTER 6 EXPERIMENTAL PROGRAM .................................................... 81

V

6.1 Introduction ................................................................................ 81

6.2 Materials and Mix Proportions ................................................... 83
6.3 Test Procedures ........................................................................... 86
CHAPTER 7 EXPERIMENTAL RESULTS AND ANALYSES:
ENGINEERING PROPERTIES ................................................... 95
7.1 Introduction ................................................................................ 95
7.2 Performance of The Clean Coal Technology By-Product
Versus Conventional Coal Fly Ash ............................................ 95
7.3 Effects of 15% Class F Fly Ash ................................................... 105
7.4 Effects of 30% Class F Fly Ash ................................................... 112
7.5 Effects of 15% Clean Coal Technology By—Product ................... 118
7.6 Effects of 30% Clean Coal Technology By-Product ................... 123
CHAPTER 8 STRUCTURE-PROPERTY RELATIONSHIPS ......................... 129
8.1 Introduction ................................................................................ 129
8.2 Experimental Work ..................................................................... 130
8.3 Pore Size Distribution ................................................................. 130
8.4 Structure-Property Relationship ................................................. 136
CHAPTER 9 SUMMARY AND CONCLUSIONS ............................................. 139
9.1 Introduction ................................................................................ 139
9.2 Experimental Program ................................................................ 139
9.3 Experimental Results on Engineering Properties ....................... 140
9.4 Structure-Property Relationships ................................................ 143
BIBLIOGRAPHY ................................................................................................... 144

vi

Table
Table 2.1

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

Table 4.7

Table 5.1

Table 6.1

Table 6.2

Table 6.3

Table 6.4

Table 7.1

LIST OF TABLES

Average Deicer Scaling Test Results for Various Curing Conditions
(Gebeler et al., 1989).3 .........................................................................

Chemical Characteristics (%) of Clean Coal Technology By-
Products and Conventional Coal Fly Ash (ICF Northwest, 1988 &
Malhorta) ............................................................................................

Physical Characteristic of Clean Coal Technology By-Products and
Conventional Coal Fly Ash (ICF Northwest, 1988) ..........................

By-Product Samples ...........................................................................
Chemical Characteristic .....................................................................
Physical Characteristic .......................................................................

Properties of Clean Coal Technology By-Products Vs. ASTM C 618
Requirements ......................................................................................

Cement Replacement Testing Result .................................................

Relatipnship Between Pore Size and Freezing Temperature (Philleo,
et a1.) ..................................................................................................

25 Factorial Design of Experiments ...................................................

Chemical and Physical Properties of Cement Conventional Coal Fly
Ash, and Clean Coal Technology By-Product ...................................

Mix Pr0portions ..................................................................................
Chloride Ion Permeability Based on Charge Passed ..........................

Subcomponent of the Experimental Program .....................................

vii

Page

9

57

57
58
59

61

64

65

69

82

84
84
91

96

Table 7.2

Table 7.3

Table 7.4

Table 7 .5

Table 7 .6

Results of the Factorial Analysis of Variance ..................................... 101

Effects of 15% Class F Fly Ash .......................................................... 105
Effects of 30% Class F Fly Ash .......................................................... 112
Effects of 15% Clean Coal Technology By-Product .......................... 118
Results of the Factorial Analysis of Variance ..................................... 123

viii

Figure
Figure 2.1

Figure 2.2

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4
Figure 4.5
Figure 5.1

Figure 5.2

LIST OF FIGURES

A Summary of the Multiple Comparison Analysis of Scaling Test
Results ..............................................................................................

Deicer Scaling Performance Versus Compressive Strength of
Low-Temperature-Cured Concretes with or without Fly Ash.3 ......

Variation of the Slump with Time for the Mixes Retempered with
Water only at 45 Min. .......................................................................

Variation of the Spacing Factor with Time for the Mixes
Retempered with Water only at 45 Min. ..........................................

Variation of the Spacing Factor with Time for the Nonair-
entrained Mixes Retempered with Water and a Normal Dosage of
Air-entraining Agent at 45 Min. ......................................................
Variation of the Spacing Factor with Time for the Air-entrained
Mixes Retempered with Water and a Normal Dosage of Air-
entraining Agent at 45 Min. .............................................................
AF BC Process Located at Michigan State University .....................
Spray Dry FGD System Schematic ..................................................
Typical Spray Dryer/Particulate Collection Flow Diagram .............
Typical Dry Sodium Injection System Schematic ...........................

Structure of Cement Paste Versus Fly Ash-Cement Paste ...............

Relationship of Strength with Gel/Space Ratio (Powers, 195 8)‘.

Pore Size Distribution Versus Age in Ordinary Portland Cement
Mortar.l ............................................................................................

ix

Page

12

14

33

34

36

37
44
46
47
50
55

71

76

Figure 5.3
Figure 5.4
Figure 6.1

Figure 6.2

Figure 7.1

Figure 7.2

Figure 7.3

Figure 7.4

Figure 7.5
Figure 7.6
Figure 7.7
Figure 7.8
Figure 7.9
Figure 7.10
Figure 7.11
Figure 7.12
Figure 7.13

Figure 7.14

Pore Size Distribution Vs. Age in a Mortar Containing Fly Ash ..... 77

Fly Ash Effects on the Pore Size Distribution of Pastes.l ................ 79
Gradation Curve of Coarse Aggregates ............................................ 85
Gradation Curve of Fine Aggregate ................................................. 86
Compressive Strength Test Results (Means & 95% Conﬁdence
Intervals) .......................................................................................... 97
Rapid Chloride Permeability Test Results (Means & 95%
Conﬁdence Intervals) ....................................................................... 98
Deicer Salt Scaling Resistance Test Results (Means & 95%
Conﬁdence Intervals) ........................................................................ 99
Freeze-Thaw Resistance Test Results (Means & 95% Conﬁdence
Intervals) .......................................................................................... 100
Ash Content Effects the Scaling Resistance .................................... 103
Prolonged Curing Effects on the Scaling Resistance ....................... 104
Test Results (Means & 95% Conﬁdence Interval) ........................... 107
Water to Cement Ratio Effects on the Permeability ........................ 109
Air Content Effects on the Permeability .......................................... 110
Test Results (Means & 95% Conﬁdence Intervals) ........................ 114
Fly Ash Effects on Concrete Permeability ...................................... 116
Fly Ash Interacts with Prolong Curing Period ................................ 117
Test Results (Means & 95% Conﬁdence Intervals) ........................ 120
Effects of Ash Content Versus Prolonged Curing .......................... 122

Figure 7.15

Figure 7.16

Figure 7.17

Figure 8.1
Figure 8.2

Figure 8.3

Figure 8.4
Figure 8.5

Figure 8.6

Figure 8.7

Test Results (Means & 95% Conﬁdence Interval) ..........................

Effects of the Clean Coal Technology By-Product Versus Moist
Curing Period ..................................................................................

Effects of the Clean Coal Technology By-Product Versus Moist
Curing Period ..................................................................................

Pore Size Distribution of Portland Cement Concrete .......................
Pore Size Distribution of Fly Ash Concrete .....................................

Pore Size Distribution of the Clean Coal Technology By-Product
Concrete ............................................................................................

Pore Size Distribution at the Age of 56 Days ..................................
Total Porosity Versus Compressive Strength ..................................
Regression Analysis Between Larger Pore Volume and
Compressive Strength (Regression Line and 95% Conﬁdence
Region) .............................................................................................
Regression Analysis Between Smaller Pore Volume and

Compressive Strength (Regression Line and 95% Confidence
Region) .............................................................................................

xi

125

126

127

131

132

134

135

137

137

138

Abbreviation

ASTM
C-S-H
mm
MPa
nm
W/C

A

um

LIST OF ABBREVIATIONS

Meaning

American Standards for Testing Materials
calcium silicate hydrates

millimeter

million pascal

nanometer

water-cement ratio

angstrom

micron

xii

CHAPTER 1

INTRODUCTION

Environmental concerns have led to increased use of clean coal combustion technologies
in coal-burning power plants. This has led to the generation of tremendous volumes of
clean coal technology by-products which are largely disposed of in landﬁlls. These by-
products generally comprise conventional coal ﬂy ash, the unreacted sorbent used to react
with the sulfur oxide present in emissions, and the sorbent reaction products. Since
conventional coal ﬂy ash has found a large-volume market in concrete construction, it
seems prudent to evaluate the potential of clean coal technology by-products for use in

this market.

This research evaluates the by-product of the calcium spray dryer (clean coal) technology
for use as mineral admixture (to partially replace Portland cement) in concrete. The
research compares the effects of the clean coal by-product and conventional Class F coal
ﬂy ash on the structure and properties of concrete materials with diverse mix proportions
which are subjected to different moist curing durations. Both short-term mechanical

properties and long-term durability characteristics are considered.

Chapters 2, 3, 4 and 5 present literature surveys on the scaling resistance of ﬂy ash
concrete, air void system characteristics of ﬂy ash concrete, clean coal technology by-

products, and microstructural characteristics of concrete. The experimental program

2
statistical analyses of results are presented in Chapter 7. The results of rnicrostructural

studies and the structure-property relationships are given in Chapter 8. Chapter 9

presents a summary of the research project and its conclusions.

CHAPTER 2
FLY ASH EFFECTS ON THE SCALING RESISTANCE OF CONCRETE:
A LITERATURE REVIEW

2.1 INTRODUCTION

The use of salt on concrete pavements to remove ice may cause an accelerating surface
disintegration of pavements in the form of scaling. Concrete that is adequately air-
entrained for ﬁost resistance may nevertheless be damaged by repeated application of
deicing salts; freezing and thawing of concrete in combination with salt action present

one of the most aggressive inﬂuences on concrete.

Concrete pavements commonly use ﬂy ash as a partial substitute for cement. Concerns
have, however, been raised regarding the scaling resistance of ﬂy ash concrete. The
properties of the surface layer of concrete play an important role in the scaling
phenomenon.l Scaling is most likely to occur on surfaces that have been overvibrated,
trowelled too early and too long, subjected to plastic shrinkage, or where excessive
bleeding has occurred. In these conditions a weak layer of paste or mortar tends to form
either at the surface or just below it, and there may exist microcracks or bleeding
channels through which surface solutions can be transported to lower levels. Careful mix
design, placing and ﬁnishing should eliminate many potential problems. If adequate
moist curing is followed by a period of drying before deicing chemicals are applied,

tendencies towards scaling should be reduced substantially.l

2.2 SCALING MECHANISMS

A number of physical and chemical mechanisms have been proposed to cause scaling, as

discussed below.

2.2.1 Physical Mechanisms

The consumption of heat required to melt ice when deicer is applied causes a rapid drop
in the temperature of concrete just below the surface, which may cause differential
thermal strains. Deicing chemicals may accumulate in concrete just below the surface to
form relatively concentrated solutions. When rain water accumulates on the surface, the
phenomenon of osmosis occurs, whereby water ﬂows to equalize concentration
differences. Considerable osmotic pressure can be created by this effect, causing rupture
of cement paste. The additional free moisture present at concrete surfaces may also

encourage the growth of microscopic or macroscopic ice lenses near the surface.

Scaling often occurs in the form of small ﬂat ﬂakes breaking away from the pavement
mass. The ﬂakes appear to be sound; this is also true for the mass of concrete. Quite
likely, the reason for this phenomenon lies in deicer gradient and the degree of saturation.
Above the failure plane of the scale, the solution of deicer in concrete pores is too

concentrated to produce sufﬁcient freezing to cause failure (antifreeze effect). Saturation

5

at the surface is determined by the vapor pressure of the solution present. Strong calcium
chloride solutions have a much lower vapor pressure than pure water. These solutions
draw moisture from concrete, lowering the surface saturation. The background saturation
is determined by the original mix proportions, curing conditions, and the environment.
The higher this background level, the more vulnerable the concrete is to deicer scaling.
The concentration gradient and the degree of saturation together produce scaling in the

form of ﬂakes.

A variable which has to be considered with respect to its possible contribution to the
scaling process is temperature. Every freeze-thaw cycling process inherently involves a
temperature cycling process. The inﬂuence of temperature cycling on the absorption of

deicer solutions plays a role in deciding the scaling of concrete materials.

2.2.2 Chemical Mechanisms

While salt scaling is generally believed to be a physical action, there are some chemical
phenomena which may pronounce the effect. The pH of pure calcium chloride solutions
can vary from 6.0 in dilute solutions to 4.0 in saturated solutions. It is likely that these
acid solutions react directly with the cement paste in concrete to yield readily soluble

salts.

6
It has been shown that saturated solutions of calcium chloride, without freeze-thaw

cycling, can be deleterious to concrete. The mode of attack depends on the water-cement
ratio of concrete. It has been hypothesized that at low water-cement ratios, leaching is the
key deterioration mechanism under exposure to saturated calcium chloride solutions,
while at higher water-cement ratios (0.7 by weight), degradation results mainly from

deposition and crystallization.

An interesting property of the attack on concrete by saturated calcium chloride is its
dependence on temperature. The specimens cycled from 45 to 75°F (6.2 to 240°C)
deteriorate much faster than those cycled from 15 to 45°F (-9.4 to 62°C). This increase

in reactivity with increasing temperature is characteristic of a chemical reaction.

The effect of strong solutions of sodium chloride, the other common deicing chemical, in

the absence of freeze-thaw cycling has received little attention.

2.3 FLY ASH EFFECTS

A combination of different factors inﬂuence the scaling resistance of concrete."2 These
factors include mix proportions (cement content, water-cementitious ratio, level of
cement substitution with ﬂy ash, etc.), cement type, ﬁnishing and curing procedures, and

exposure conditions after the placement of concrete.

Fly ash may change the scaling resistance of concrete through its effects on: (a) reduction
of water requirements for fresh mix workability; (b) reduction of permeability and ﬁlling
of large capillary pores with relatively insoluble calcium silicate hydrates; 3’4’5‘6’7 (c)
causing slower rates of strength gain at early ages; (d) increase in the inﬂuence of
temperature on the rate of reactions in concrete; (e) destabilizing the entrained air void
system in fresh concrete; and (f) changing the chemical composition of concrete and pore
water. These factors are expected to have opposing effects on the scaling resistance of

concrete. The speciﬁc effects of ﬂy ash on surface characteristics of concrete have not

been fully investigated.

While the mechanisms through which ﬂy ash inﬂuences the scaling resistance of concrete
are not well understood, a number of investigators have performed experimental studies
on the scaling resistance of ﬂy ash concrete. The reported test data dealing with the
effects of curing procedures, ﬂy ash type, ﬂy ash content and water-cement ratio on the

scaling resistance of ﬂy ash concrete are summarized below.

Gebeler et al.(1989)3 conducted scaling tests on ﬁve sets of air-entrained concrete slabs
cured in the following conditions: moist-cured at 73°F (23°C); air-cured at 73°F (230 F);
membrane-cured at 73°F (23°C); membrane-cured at 40°F (44°C); and moist-cured at
40°F (44°C). Concrete mixtures with Class C or Class F ﬂy ash as well as control

mixtures without ﬂy ash were investigated. Five replications were made in scaling tests

8
with each class of ﬂy ash under each curing condition. Mixtures with ﬂy ash had total

binder contents of 517 lb/cu.yd. (307 Kg/cu.m.) and 474 lb/cu.yd. (282 Kg/cu.m.), and
ﬂy ash-binder ratios between 0.4 and 0.45; those without ﬂy ash had cement contents of
517 lb/cu.yd. (308 Kg/cu.m.) and 474 lb/cu.yd. (283 Kg/cu.m.). The fresh mix air

contents were almost constant in all mixtures, ranging from 6 to 6.5 percent.

Following curing, the slabs were ponded with a 1/4-in. (6-mm) layer of water on the test
surface for 3 days, after which the freeze-thaw deicer test cycles commenced. Freezing
took place in a room maintained at 0 j; 3°F (~18 j; 1.7°C); thawing was accomplished in a
room maintained at approximately 70°F (21°C). To start the thawing, ﬂake calcium
chloride was applied to the l/4-in. (6-mm) layer of ice on the top surface at the rate of 2.4
lb/yd2 (1.40 Kg/Mz). After thawing, the slabs were washed and 1/4-in. (6-mm) layer of
water was replaced. Slabs were subjected to one freeze-thaw cycle per day. The amount

of scaling was determined visually at selected intervals of 100, 200 and 300 cycles.

A summary of the scaling test data at 300 cycles reported by Gebeler et al.(l989)3 is

presented in the Table 2.1.

9

 

 

 

 

 

 

 

 

Table 2.1 Average Deicer Scaling Test Results for Various Curing Conditions
(Gebeler et al., 1989).3
Scale rating at 300 cycles“)
Class of ﬂy ash 73°F (23°C) 40°F (44°C)
Moist cure Air Cure Membrane Moist Cure Membrane
Cure cure
C 3- 3 2 2- 2-
F 3 3- 2+ 2 2+
Control
Mixtures 2- 2 2- 1+ 2
517 poy‘b’
474 pcym 1+ 2+ 2 1 2-

 

 

 

 

 

 

(a) Where: 0 = No scaling
1 = Slight scaling
2 = Slight to moderate scaling

3 = Moderate scaling

4 = Moderate to severe scaling
5 = Severe scaling
(b) To convert pounds per cubic yard to kilograms per cubic meter, multiply by 0.594.

Analysis of the data presented in Table 2.1, disregarding the observed experimental

errors, have led Gebeler et al.(1989)3 to make the following conclusions:

Under moist curing at 73°F (23°C), air-entrained concretes with ﬂy ash

showed somewhat less resistance to deicer scaling than the control concretes, and

there was essentially no signiﬁcant difference in deicer scaling performance between

either class of ﬂy ash.

Air-entrained concretes with ﬂy ash, when air-cured at 73°F (23°C), showed

somewhat less resistance to deicer scaling than either of the control concretes. The

lack of moisture during curing did not have signiﬁcant effects on the deicer scaling

resistance of concretes with ﬂy ash. However, the control concretes did show slightly

 

10

less deicer scaling resistance after air curing as compared to moist curing. This was
particularly noticeable for the control mix with lower cement content. Overall, the
lack of moisture during curing had relatively small effects on the deicer scaling
resistance of concretes with or without ﬂy ash. Some test data reported in the
literature, however, suggest that air-drying may have beneﬁcial effects on deicer
scaling resistance.8

Membrane curing at 73°F (23°C) improved the deicer scaling resistance of

ﬂy ash concrete as compared to moist curing or air curing at 73°F (23°C).

For the control air-entrained concretes (without ﬂy ash), the effect of membrane
curing on deicer scaling resistance was minor.

Air-entrained concretes without ﬂy ash which were moist cured at 40°F (44°C)
showed slightly better resistance to deicer chemicals than air-entrained concretes with
ﬂy ash.

For specimens that were membrane cured at 40°F (44°C), concretes with Class C ﬂy
ash had essentially the same deicer scaling resistance as control concretes. However,
concretes with Class F ﬂy ash showed slightly less resistance to deicer chemicals than
concretes with Class C ﬂy ash or those without ﬂy ash.

Examination of the overall performance of air-entrained concretes cured at moderate
and low temperatures, and subjected to deicer chemicals during freezing and thawing,
indicates that low-temperature moist curing of concretes with or without ﬂy ash

resulted in somewhat better resistance to deicer chemicals. However, concretes with

ll

ﬂy ash that were membrane cured showed essentially no difference in resistance to

deicer chemicals for the two temperatures.

A statistical analysis of the data summarized in Table 2.1 was conducted in order to
account for random experimental errors in the analysis of the observed trends. The
results of statistical analysis (multiple comparison”) indicated that random experimental
errors are relatively large and, given the limited number of replications in scaling tests
reported by Gebeler et al.(1989)3 , the stated conclusions can not be veriﬁed statistically.
The results of the multiple comparison study performed on the scaling test results of
Gebeler et al.(1989)3 are summarized in Figure 2.1. This ﬁgure presents the mean values
of the scaling test results as well as the range of the least signiﬁcance difference for each
condition. The least signiﬁcant difference is the smallest difference for which
signiﬁcance can be declared (in this case with 5% level signiﬁcance, that is with 5%
chance of error in declaring a signiﬁcance difference). An analysis of Figure 2.1
indicates that, considering the random experimental errors for the given test data at 5%
level of signiﬁcance, the addition of ﬂy ash to concrete does not have signiﬁcant adverse
effects on scaling resistance under similar curing conditions. The only exception to this
rule was for moist curing at 73°F ( 23°C), where ﬂy ash (particularly Class F) caused
negative effects on scaling resistance. Considering the trends observed, however, one
may suggest that more replications of tests are needed in order to have a more powerful

conclusions regarding ﬂy ash effects on scaling resistance.

12

300 Cocoa “om homo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aOI- ’ «p '-
Human“!
__ t Gooocﬂym
32 ,_ r : emolFlonl
_ "' -- 3 Control
25- ‘ - “Malia“!
. I o croucnym
2‘_ _ o .cloooFAlonh
1" J n r . 6.0"“
J j_ ' ~ l Intranet-““78“!
Zl' l. 1, 7 CloooCFlonu
- I CloooFFlonh
4 l. 0 Control
”if . .. Houuoooc «occur
" to Cloooc 11m
12" - l - i. n ClouFFlonn
- 12 Control.
08" « HololCuroootoooooJ
' - ﬁCIoooC'lonn
i4 Mfllym
0.4? J is Gourd

 

 

 

123456789l0ul2l3l4l5

OIuooooo can. on. My no

Figure 2.1 A Summary of the Multiple Comparison Analysis of Scaling Test Results.

Klieger et a1. (1987)9 have also reported the results of scaling resistance tests on ﬂy ash
and control concretes subjected to different curing conditions. The ﬂy ash concrete used
in this study contained 388 lb/cb.yd. (230 Kg/cu.m.) of cement and 129 lb/cu.yd. (77

Kg/cu.m.) of ﬂy ash (a 25% cement replacement with ﬂy ash). The control concretes had

13

a cement content of 517 lb/cu.yd. (307 Kg/cu.m.), with no ﬂy ash. Curing in this study

was of ﬁve different types: moist curing at 73°F (23°C); air curing at 73°F (23°C);

membrane curing at 73°F (23°C); membrane curing at 40°F (44°C); and moist curing at

40°F (44°C). The severity of scaling was determined visually.

The scaling test data reported by Klieger et al(1987)9 indicated that :

Air-entrained concretes with Class C ﬂy ash cured at low temperature generally
showed good performance when subjected to freezing and thawing in water or salt
solution.

The lack of moisture during curing had only a minor effect on the deicer scaling
performance of air-entrained concretes with or without ﬂy ash.

Air-entrained concretes with either class of ﬂy ash and moist-cured at 73°F (23°C)
showed approximately equal performance when subjected to deicer chemicals during
freezing and thawing.

Air-entrained concrete with Class F ﬂy ash cured at low temperature had slightly less
resistance to deicer scaling than air-entrained concretes with Class C ﬂy ash or
without ﬂy ash.

Air-entrained concrete with Class C ﬂy ash, membrane cured and stored at low
temperature, showed essentially the same deicer scaling performance as air-entrained

control concretes without ﬂy ash.

l4
Gebeler et al. (1989)3 suggest that strength level is an important factor deciding the

scaling resistance of concrete, and ﬂy ash effects on deicer scaling performance can be
partly illustrated by the ﬂy ash effects on compressive strength. Figure 2.2 illustrates
that, for concretes with or without ﬂy ash cured at 40°F (44°C), deicer scaling resistance
generally improves with increasing compressive strength. This ﬁnding is consistent with
Kliger’s9 conclusion that “The development of a certain level of strength has merit as an
index to the amount of curing required for air-entrained concrete prior to permitting the

use of deicers.”

300 Cycles Dolcor Scollng Rotlng

 

 
 
 
  
  
  

 

 

 

5
0 Class F Fly Ash
+ Class C Fly Ash
4 e .
9K 517 pcy Conlr0l
474 pcy Contr0l
3 — 95% Cont. Level
O —— Regression Line
. 95% Cont. Level
2. .-
1 .—
0 l l l I . I
2500 3000 3500 4000 4500 5000

28-Day Compressive Strength ‘

Figure 2.2 Deicer Scaling Performance Versus Compressive Strength of Low-
Temperature-Cured Concretes With or Without F 1y Ash.3

15
Johnston (1987)10 suggests that a safe upper limit of water-to-cement plus ﬂy ash ratio in

ﬂy ash concrete with respect to scaling is between 0.40 and 0.53, which is comparable to
the limit of 0.45 on water-to-cement ratio recommended for normal concrete exposed to
deicing agents. The laboratory and ﬁeld test data analyzed by Johnston (1987)10 indicate
that ﬂy ash concrete with 550 lb/cu.yd. (326 Kg/cu.m.) of cement and 275 lb/cu.yd. (163
Kg/cum.) of ﬂy ash, can be satisfactorily used for applications involving deicer scaling

exposures, provided that the water-to-cement plus ﬂy ash ratio is about 0.4.

Malharta et al. (1987)11 report test results on concretes with (Class F) ﬂy ash content of
158 lb/cu.yd. (94 Kg/cu.m.) and cement content of 473 lb/cu.yd. (281 Kg/cu.m.), with
25% of cement substituted by ﬂy ash, indicating that increased water-to-cement plus ﬂy
ash ratio has negative effects on surface scaling. At water-to-cement plus ﬂy ash ratios of
0.6 and 0.4, slight scaling could be observed after 300 and 500 cycles of freezing and

thawing, respectively.

Tyson (1987)12 and Johnston (1987)10 suggest that there are upper limits on cement
replacement with ﬂy ash beyond which scaling can become a problem. Test results
presented by Johnston (1987)IO on air-entrained concretes with cement plus ﬂy ash
contents of 503 lb/cu.yd. (299 Kg/cu.m.) and 488 lb/cu.yd. (290 Kg/cu.m.), at water-to-
cement plus ﬂy ash ratio of 0.53 indicate that scaling for the concrete without ﬂy ash is
negligible; it is slight for the concrete with 16% ﬂy ash, and severe for the concrete with

12.13

more than 27% ﬂy ash. Tyson suggests that scaling test results performed on

l6
concretes with 423, 517 and 564 lb/cu.yd. (250, 305 and 335 Kg/cu.m.) of cement plus

ﬂy ash at ﬂy ash replacement levels of 25 and 50% indicate that optimum levels (for the
speciﬁc ﬂy ash types used in this investigation) appear to be closer to 25 percent than 50

percent.

2.4 CONCLUSIONS

Relatively large variations (random experimental errors) are observed in the limited test
data reported on the scaling resistance of ﬂy ash concrete. Statistically sound conclusions
regarding ﬂy ash effects on the scaling resistance of concrete can not be derived from the
available experimental results. The following general trends may, however, be
distinguished in the reported data:

(a) Fly ash effects on scaling resistance seem to be dependent on the curing conditions.
For membrane-cured concretes, the scaling resistance of ﬂy ash concrete (at 25%
cement replacement level, with either Class C or Class F ﬂy ash) was similar to that
of conventional concrete.

(b) The scaling resistance of ﬂy ash concrete seems to be dependent on the ﬂy ash type.
While under same curing conditions Class C and Class F ﬂy ashes produced similar
results, under other curing conditions Class C had a tendency to produce better
scaling resistance than Class F ﬂy ash.

(c) Scaling resistance of concrete, with or without ﬂy ash, seems to improve with

l7

increasing compressive strength of concrete. The relationships between scaling
resistance and strength have been observed to be similar in conventional and ﬂy
ash concretes (with 25% cement substitution with ﬂy ash).

(d) There seem to exist limits on cement substitution with ﬂy ash and also on water-

binder ratio, beyond which ﬂy ash concrete may exhibit pronounced deicer scaling.

CHAPTER 3
FACTORS EFFECT THE AIR VOID SYSTEM ON THE

POZZOLANIC MATERIALS:A LITERATURE REVIEW

3.1 INTRODUCTION

The use of internal vibration to consolidate fresh concrete has been one of the most
beneﬁcial technological advances in concrete industry. It has greatly facilitated
placement and consolidation of concrete of relatively stiff consistency and as a results,
has permitted use of stronger and less massive reinforced concrete members. In virtually
all applications, internal vibration has permitted wider use of denser, higher quality

concretes.

Although considerable research has been undertaken to evaluate the effectiveness of
vibration for consideration of concrete, its effects on the entrained air-void system and

freeze-thaw durability of concrete are less well deﬁned.1

Past researchs’6 has shown that ﬂy ash can adversely affect the production of an adequate
air content, but explanation of this effect has not been thoroughly establish. Other
researchersm’9 have found higher air-entraining admixture requirements for concretes
containing ﬂy ash than for concrete not containing ﬂy ash. Larson1° concluded that the

higher air entraining admixture cement was due to the carbon content of the ﬂy ash.

l8

19
However, the air-entraining admixture requirement reported for concrete containing these

ﬂy ashes is still higher than for concrete without ﬂy ash. Therefore, another main task is
to obtain a better understanding of factors that inﬂuence the air entraining admixture
requirement.4 Nowadays, adding water to a plastic mix to increase slmnp is an extremely
common practice, even though it is not recommended because it increases the porosity of
concrete. Concrete often arrives on site more than half an hour after initial mixing.
Placement operations can take anywhere ﬁom 10 to 60 min, depending on the ﬁeld
conditions and the size of load. When the slump decreases to an unacceptable level
during the operations, water is added to the mix and, very often, experienced ﬁeld
inspectors will tolerate what can be termed "reasonable" retempering, i.e., enough to
increase slump by 50 or 60 m. If the air content of the fresh concrete is below the
required minimum percentage when it arrives on site, it is also common to add an amount
of air-entraining agent. Usually, just enough is added to reach the minimum acceptable

value of air content because of the strength reduction due to the air-content increase.

Does retempering by adding water 30 or 60 min. after initial mixing have an effect on the
air-void system, and particularly on the value of spacing factor? How much air-
entraining agent must be used at about the same amount, not only increase air content but
also decrease the spacing factor below the usual 200 um limit? These are very important
questions concerning freeze-thaw durability and there are few data on the subject in the
technical literature. Langan and Ward17 have reported that using 2 to 5 percent of the

original dosage of air-entraining agent is sufﬁcient to increase the air content to the

20
original level and will have little effect on the spacing factor. Results of ﬁeld test by

Burgl° have shown that adding water to increase slump by an average of 25 mm had little

inﬂuence on the spacing factor (L).

3.2 BACKGROUND

Thermodynamically, air void in fresh concrete is intrinsically unstable. Because of air
pressure differentials within entrained air voids of different sizes, and at least temporary
under saturation of air in mix water in fresh concrete, there is a natural tendency for air in
smaller than average voids to dissolve into solution, and for air to be transferred to larger
than average voids. The net result of these tendencies is an overall decrease in the
ﬁneness of the air-void system, growth of larger than average voids, and an increase in air
content. There also is a tendency for air voids at exposed surfaces in to burst because

of higher pressures in the air voids compared with ambient atmospheric pressure. Both
tendencies are facilitated by high water-cement ratios. It is possible, however, that most
spontaneous changes in air void systems are prevented from occurring while concrete is
in the mechanically undisturbed state because of the cohesive strength of the concrete and

the comparatively long diffusion paths between bubbles.

21
Work by Backstrom, Burrows, Mielenz, and Wolkodoff showed that compacction,

including internal vibration, resulted in decreased air contents and smaller average size of
air voids. Their work also indicated that vibration had little effect on void spacing

factor (L). These authors also showed that longer periods of internal vibration resulted in
greater loosed of air and further reduction in average size of air voids, where L remained
virtually unaffected.2 Work by Crawley3 Produced results similar to those summarized
above and further showed that frequency of vibration affected the stability of entrained air
void system. Using frequencies of 6800 and 13,000 vibrations per minute (vpm), air
losses were found to be greater with the higher frequency and to increased slump of fresh
concrete. Both frequencies were found to remove nearly all entrained air when duration
of vibration was exceed 1.5 and 4.5 minutes for concrete with initial contents up to 9%.
These periods are considerably longer than commonly encountered in practice. Both
ﬁequencies also were found, after vibration for 90 second, to cause pronounced
segregation of coarse aggregate and uneven distribution of entrained air voids. It is thus
evident that internal vibration signiﬁcantly alters the character of the entrained air-void

system. Most signiﬁcantly, it can be expected to reduced air contents.

The newly developed Dodson11 "Foam Index" test was investigated. He suggested that
this test could be used to estimate the air-entraining admixture requirement of Portland-

cement concrete mixtures containing ﬂy ash.4

22
3.3 LITERATURE TEST PROCEDURE AND RESULTS

David C. Starkl in his test, four parameters were used to determine the characteristics and
performance of the concrete prisms. Air void characteristics of the hardened concrete
were determined by linear traverse methods described in ASTM Designation: C457-82
Standard Practice for Microscopical Determination of Air Void Content and Parameters
of the Air Void System in Hardened Concrete This was done on one of the four prisms
for each variable combination. Numbers of / voids measured with chord intercepts in
successive 12 micron intervals, up to 240 microns, also were recorded. Performance
during freezing and thawing was evaluated by changing in weight and fundamental
transverse frequency of vibration. Weights were measured on a balance sensitive to 0.5 g.
All prisms were weight at 73+ 1°F in the saturated surface-dry condition. Fundamental
transverse frequency of vibration was determined in accordance with ASTM Designation:
C215-60 (Reapproved 1976) Standard Test Method for fundamental Transverse,
Longitudinal, and Torsional Frequencies of Concrete Specimens. Relative dynamic
modulus of elasticity, P, expressed in percent, was calculated from the following
equation:

P=(N1/N)2xWx100 (1)
Where:

N = flmdamental transverse frequency at zero cycles

N1 = fundamental transverse frequency at indicated number of cycles

W = ratio of weight of concrete prism at indicated number of cycles to weight at zero

23

cycles.

As stated in ASTM C666-80, relative dynamic modulus of elasticity can be calculated
ﬁom Equation 1, disregarding weights are relatively constant. However, signiﬁcant
weight losses developed for most of the concrete prisms; thus the weight change
parameter, W, was included in the calculation of P. In this study, concretes were made
with normal water-cement ratios of 0.4, 0.5, 0.6, and 0.7 and were subjected to internal
vibration at frequencies of 0; 8000; 11,000; and 14,000 vpm prior to freeze-thaw testing
in 4% NaCl solution. Air void characteristics of the various concretes were determined
from linear traverse measurements while ﬁ'eeze-thaw durability were measurements of

weight loss and changes in relative dynamic modulus of elasticity.

Test data revealed that, for all four water-cement ratio concretes, internal vibration, as
carried out in this study, generally adversely affected the quality of the entrained air-void
systems, compared with those in concretes that were not vibrated. Vibration led to
reduced air contents and increased average void sizes and spacing factors. Change in air
void parameters described above generally were reﬂected in the overall trends in relative
freeze-thaw durability. Based on weight loss and change in relative dynamic modulus of
elasticity, the greatest resistance to freeze-thaw deterioration occurred, in general, in the

lowest water-cement ratio concretes subjected either to no vibration, or to vibration

24

frequencies of 8000 vpm. Conversely, the lowest durability levels were found in the
higher water-cement ratio vibrated at 14,000 vpm. None of the vibrated 0.40, 0.50, and
0.60 water-cement ratio concretes had durability levels as low as the 0.70 water-cement
ratio concrete that was not vibrated. Results of this study thus support previous ﬁnding
that internal vibration can signiﬁcantly alter entrained air void system in concrete.2
However, the manner in which vibration is carried out evidently determines whether it
will adversely affect the ﬁnal characters of air void systems and consequent durability of
concrete in ﬁ'eeze-thaw exposures. In the present study, both were adversely affected by
internal vibration. Since internal vibration has been used successfully in many
applications for many years, the ﬁndings of this study illustrate the fact that it also can be
used incorrectly with resultant adverse effects on the concrete durability. This possibility
should bee kept in mind, particularly when high water-cement ratio concretes are being
used.1 Gebler et al.4 in their investigation were to evaluate the effect of ﬂy ash on the air
void stability, and air entraining admixture requirement in concrete, and to investigate the
applicability of "Foam Index" test. Fly ashes were selected to represent a wide range of
chemical and physical compositions, as well as geographical location. Tests were

performed on both freshly mixed and hardened concretes containing ﬂy ash.

To evaluate the air void stability, air contents of plastic concrete were measured over a
period of 90 min. After completion of initial mixing. A l-min. remix, including
retempering, was used at 30 min. intervals after the initial mixing. Concrete specimens

were cast at 30 min. intervals for subsequent linear traverse measurements.

25

The "Foam Index" test for estimating the air entraining admixtures requirement was
evaluated. Results of this test were compared to air entraining admixture requirement for
concrete. This investigation was also directed toward obtaining a better understanding
admixture requirements of concrete. The test concrete were proportion for a nominal
cementitious material content of 517 pounds per cubic yard (307 Kg/m3) with 75% Type

I Portland cement and 25% ﬂy ash by weight of Cementitious material.

Two control concrete mixtures without ﬂy ash were also prepared. One control concrete
had a nominal cement content of 517 pcy (307 Kg/m3); the other control concrete had a
nominal cement content of 474 pcy (281 Kg/m3). The latter was used to provided early
age strength development more nearly equal to concrete containing ﬂy ash as well as for

other durability studies.

The ﬂy ash concretes used in this study contain 388 pcy (230 Kg/m3) cement and 129 pcy
(77 Kg/m3) of ﬂy ash. Relative to the control mixes, this ﬂy ash concrete can be
considered as a 25% cement replacement concrete to be compared with the 517 pcy (3 07
Kg/m3) control and as a partial replacement plus admixture for comparison with the 474
pcy (281 Kg/m3) control. In the latter case, 86 pcy (51 Kg/m3) of cement (17%) is

replaced by ﬂy ash and 43 pcy (26 Kg/m3) of ﬂy ash (8%) is used as an admixture.

26’

All concrete mixtures were proportioned to have a slump of 3 in. i 1 in. (75 + 25 mm),

with sufﬁcient neutralized Vinsol resin to obtain a nominal air content of 6 i 1 % in the

plastic concrete.4 The following ﬁndings and conclusions are based in large part on an

arbitrary separation of the ﬂy ashes used in this study into ASTM C 618 Class C and F

designations based on 10% CaO content as a dividing line. Therefore, there may be

exceptions to these general classiﬁcations. The ﬁndings and conclusion are presented

below:

1.

Generally, concretes containing Class C ﬂy ash require less air entraining admixture
than those concrete with Class F ﬂy ash. All concretes with ﬂy ash required more air

entraining admixture than the Portland-cement concrete without ﬂy ash.

. Plastic concrete containing Class C ﬂy ash tend to lose less air than concretes with

Class F ﬂy ash.

Spacing factors (L) on specimens cast over a period of 90 min. were essentially
constant for the majority of concretes containing ﬂy ash. In addition, the initial
spread of results of speciﬁc surface and voids per inch (n) was essentially similar

for concretes containing Class F or Class C ﬂy ash. However, when measured on
specimens cast at 90 min., concretes with Class F ﬂy ash exhibited greater

variability of results for these air void parameters than concretes with Class C ﬂy ash.
As the air entraining admixture requirement increases for a concrete containing ﬂy

ash, the air loss increases.

. Air contents in plastic concrete containing Class F ﬂy ash at 90 minutes after

completion of mixing were reduced as much as 59%. Similar results were obtained

27

on hardened concrete specimens cast at various time intervals after mixing.

6. The "Foam Index" test appears to be a good rapid test to predict relative air entraining
admixture dosage requirements for concrete.

7. As the organic matter content, carbon content, and loss on ignition of ﬂy ash
increase, the air entraining admixture requirement increases, as does the loss of air in
plastic concrete.

8. Generally, as total alkalis in ﬂy ash increase, the air entraining admixture requirement
decrease.

9. As the speciﬁc gravity of a ﬂy ash increase, the retention of air in concrete increases.
Concrete containing a ﬂy ash that has a high lime content (Class C ﬂy ash) and less
organic matter tends to be less vulnerable to loss of air.

10. Generally as the 80;, content of ﬂy ash increases, the retained air in concrete

increase.

3.3.1 Effect of Water/Cement Ratio

Water/cement ratio inﬂuences the size distribution of air voids because the cement paste
air content and water phase viscosity are greatly altered as water/cement ratio changes. '2

As the water/cement ratio decreases, the viscosity of the water phase increases and the air

content of cement paste decreases.

28

The effects of varying water/cement ratio on several air void system parameters were
summarized in the following.

1. Increasing w/c while holding the ratio of air-entraining admixture to concrete volume
constant greatly increases the total air content of the paste and produces larger air
bubbles, thereby decreasing the speciﬁc surface of the voids. Note that spacing
factor and freeze-thaw expansion decrease initially before developing an overall
trend of increase. This indicates that the inclusion of some larger air bubbles may
produce a better size distribution of air voids for freeze-thaw durability.

2. Increasing w/c while maintaining approximately constant air content results in the

use of less air-entraining admixture at higher w/c ratios. Thus larger bubbles are
again produced, speciﬁc surface decreases, spacing factor increases, and freeze-thaw

resistance decreases accordingly.

3.3.2 Air-Entraining Admixtures

Air-entraining admixture are usually water-insoluble materials such as natural wood
resins, fats, and oils that are processed and combined with the concrete mix water to
produce microscopic voids in the cement paste. These entrained voids combine with
voids produced by excess mix water and entrapped air to provide an air void size

distribution that enhances freeze-thaw durability.

29
Air-entraining admixture can signiﬁcantly alter the properties of both ﬂesh and harden

concrete”. Fresh air-entrained concrete is considerably more plastic and workable than
nonair-entrained concrete. It can be placed and hardened with less segregation and has a

lower tendency toward bleeding. Hardened air-entrained concrete. ‘3 ’14’15

This improved
durability can be traced to improved mix uniformity, permeability, and absorption

characteristics that are brought about by the increased number of small, closely-spaced

voids in the paste.

Commonly-accepted critical values for air void spacing factor range from 0.006 to 0.01
in. (0.15 to 0.25 mm).16 Analysis of hardened concrete air void systems reveals that the
spacing factor for nonair-entrained concrete is generally higher than critical, while
properly air-entrained concretes are generally lower than critical15 and suggest that lack
of adequate air-entrainment can seriously damage concrete subject to moderate or severe

exposure.

While the use of entrained air improve the durability characteristics of Portland cement
concrete, it is also known that reducing the w/c ratio can produce similar results (as
described previously). However, tests by the US. Bureau of Reclamation found that air-
entrained pastes with high w/c ratios were more durable than nonair-entrained pastes with
low w/c ratios.l3 These ﬁndings lead to the conclusions that the relative improvement of
durability due to air-entrainment is greater than increased durability that results when w/c

is decreased.

30

The importance of air-entrainment in producing durable concrete is widely accepted.
However, most known air-entraining admixtures (especially wood resin-based
admixtures) have not been entirely satisfactory with respect to the stability of the number
and/or volume of voids they have produced.14 They have shown tendencies toward

increased or decreased air content as mixing time is extended.

Another disadvantage of air-entrainment is the reduction of concrete compressive
strength. Losses of up to 10% per 1% increase in entrained air have been reported. These
are believed due in part to irregular void size and the coalescence of small voids during
mixing or consolidation to produce much larger voids. These tendencies vary widely
with the type of air-entraining admixture. For example, microscopic examination of a
concrete produced using a methyl ester-derived cocamide diethanolamine (also known as
cocamide DEA) found bubbles that are spherical in shape and have thick walls that resist
coalescence. The wood resin-based admixtures produced bubbles with relatively thin
walls that tend to be less spherical and prone to coalescence.l4 These difference were
believed to account for differences in air content stability in the mixed and hardened
concrete, with the cocamide DEA admixture performing more favorably. These more
favorable air void sizes and distribution were found to result in increases in freeze-thaw
durability of 400% and compressive strength increases of 25% at early ages and 10% at

later ages.

31

From the above, it is clear that air—entrained concretes can produce air void systems that
substantially improve concrete workability and freeze-thaw durability with minimal
losses of compressive strength. However, there are substantial differences in the

effectiveness and deleterious effects of various air-entraining admixtures.

Pigeon et al.l° in their research was to studying the inﬂuence of retempering on the
characteristics of the air-void system. It consists of preparing a mix with predetermined
parameters and taking four sets of samples to determined the air-void characteristics of
the harden concrete: at 10, 25, 55,and 85 min. after the initial water-cement contact.
Periods of mixing followed by rest periods are used in laboratory to simulate agitation in

mixer trucks.

Water retempering was investigated by adding enough waterto the mix at 45 min,
between the second and third sample preparation, to increase the slump from about 50 to
100 mm. This represents approximately 10 Kg of water per m3 of concrete and, thus , a
water-cement ratio increase of the order of 0.03 (0.45 to 0.48). Taking into account usual
ﬁeld practice and the effect on strength and freeze-thaw durability of the water-cement
ratio, this amount of water was considered to be the upper limit of reasonable
retempering. Four laboratory mixes were prepared using two different air entraining
agents, each at two different dosages: one normal and one low. This was done to study

the effect or retempering on two mixes with a normally stable air-void system, as well as

32

on two with a probably unstable one. The two mixes with a normal amount of air-

entraining agent were prepared a second time under ﬁeld conditions.

Four laboratory and six ﬁeld mixes were made, again systematically using two
admixtures, to see if air-entraining admixtures can produce a correct air-void system, or
signiﬁcantly alter the existing one, when they are incorporated into the fresh concrete

much later than the initial mixing period.

In the four laboratory concretes and in the ﬁrst four ﬁeld mixes, two different procedures
were tested: two mixes were prepared with no air-entraining agent with the full dosage
added at 45 min, two mixes were made with a low dosage of the admixture added in
large quantity at 45 min. To serve as a reference point, two additional ﬁeld mixes with a
low air-entraining agent dosage were fabricated and not retempered. Water was always
added with the admixture when the mixes were retempered to insure proper dispersion.
Standard concretes with normal dosage of air-entraining agent (and no retempering) were

fabricated for the ﬁrst series of tests on air-void stability and, thus, were not repeated.

Water and air-entraining agent retempering was always carried out at 45 min. This
allows air-void stability to be analyzed with two sets of samples before and after
retempering, and also very approximately represents the time at which usual ﬁeld mixes

are retempered. '7

33
The test results described indicate clearly that the air void spacing factor does not vary

signiﬁcantly when a mix is retempered by adding a sufﬁcient quantity of water to
increase the slump from approximately 50 to just about 100 mm. This is illustrated in F ig,
3.1 and 3.2, which show, respectively, the variation of the slump and variation of the

spacing factor with time for ﬁve of the six mixes retempered with water only.

 

 

 

 

 

 

 

 

 

150 'l
o-wm
OW
;\ OW
I” k L : II —
E ‘
5
E \_
. 50

 

 

 

 

 

 

 

 

o 1'—
0 15 30 . 45 oo 75 so
MdW(M)

Figure 3.] Variation of the Slump with Time for the Mixes Retempered with Water
only at 45 Min.

34

 

 

 

?

 

 

 

 

 

 

Spacing Ioctor (pm)

§
IWM
\T
l
. . i\/

1111..

 

 

 

 

 

 

 

 

 

 

MUMlM)

Figure 3.2 Variation of the Spacing Factor with Time for the Mixes Retempered with
Water only at 45 Min.

The results shown in Fig. 3.1 and 3.2, although contrary to popular belief, are
nevertheless extremely clear: when the slump increases sharply, the values of spacing
factor stay constant. This is true for the three ﬁeld and laboratory mixes with reduced

dosages of the admixture and thus higher values of the spacing factor.

In some cases, retempering seems to cause a small increase in air content with a
corresponding decrease of the speciﬁc surface, although the spacing factor stays

relatively constant. Probably because less energy is required to entrain air in a less

35

viscous paste, remixing a more ﬂuid mix, as already pointed out, tends to cause the
entrainment of additional air voids. These are large air voids, which have little effect on
the spacing factor but increase the speciﬁc surface. As the results for air content and

speciﬁc surface indicate, these air voids are not stable and tend to disappear with time.

The variation of the spacing factor with time for the mixes retempered with water and air-
entraining agent at 45 min. is shown in Fig. 3.3 and 3.4. For the ﬁeld and laboratory
mixes that were nonair-entrained at the start (Figure 3.3), the spacing factor after addition
of the air entraining agent is below 200 um and stable in every case. For the mixes made
with a reduced dosage the admixture (Figure 3.4), the value of the spacing factor
immediately after retempering is correct, but it tends to increase with time for concretes
containing the SH admixture. It is thus possible to entrain a correct air void system 45
min. after the initial water-cement contact, which means that the entraining and
stabilizing effects of the air-entraining admixtures are still effective when various
chemical reactions of hydration have been going on for that period of time. It is also
possible to correct an unsatisfactory air-void system by adding more air-entraining agent
after 45 min. The amount added, however, must be substantial, i.e. of this case, and the

air-void spacing factor does not seem to become perfectly stable.

36

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1000 J
o M
can
A o m
E
3 a m
3 30° 0 m ——
8
O
C
To 400
a
m
..
0 , .
0 15 30 45 60 75 90 ‘

mammal

Figure 3.3 Variation of the Spacing Factor with Time for the Nonair-entrained mixes
Retempered with Water and a Normal Dosage of Air-entraining Agent at

45 minute.

37

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

500
I I
. M
....... o
E I W
a 000 ﬁ (~—
3 a
3 dig,
a 5.. a ,,
g m ooooooooooooooooo g
“ 'E"’Oo.....hoo' w m
0
0 15 ’0 “ .0 7‘ .0

Figure 3.4 Variation of the Spacing Factor with Time for the air-entrained mixes
Retempered with Water and a Normal Dosage of Air-entraining Agent at

‘45 minute.
It is also evident from the test results described that small air-content variations (less than
about 2 percent) cannot be used to predict spacing factor variations. The spacing factor
can increase signiﬁcantly even if the aif content stay relatively stable ﬁ'om 10 to 25 min.,
and the spacing factor can be stable while the air content decreases very substantially
from 10 to 25 min. Of course, air losses can also correspond to spacing factor increases.
These last examples illustrate another very important aspect of the problem of air

entrainment: the difference between air-void production and stability. The air-void

38

characteristics in samples prepared after 25 min. can be quite different from those in

samples prepared after 10 min., even in mixes with normal initial air contents or in mixes

with normal dosages of the air-entraining agent. The spacing factor can often be

inadequate aﬁer 25 min. even if the value after 10 min. was lower than 200 um (Figure

3.2 to 3.4). This phenomenon seems to be more common in laboratory mixes, probably

in part due to the high energy developed by the mixer.

17

3.3.3 Summary

1.

Retempering by adding enough water to increase the slump from approximately 50 to
about 100 mm has no signiﬁcant inﬂuence on the value of the air-void spacing factor,

although it often causes a small increase in air content.

. In a previously nonair-entrained mix, it is possible to entrained an adequate air-void

system with a spacing factor of less than 200 pm if a normal dosage of air-entraining
agent, with a little water, is incorporated into the mix 45 min. after the initial water-

cement contact.

. It is possible to correct an unsatisfactory air-void system by adding more of the air-

entraining agent after 45 min., but to lower signiﬁcantly the value of the spacing factor,
the quantity of admixture that is added must be signiﬁcant (more than about 30 to 50

percent of the normal dosage) to cause a marked increase of the air content.

. Air-content variations of the order of l or 2 percent should not be used to predict

spacing factor variations.

39
5. The characteristics of the air-void system tend to be more stable in ﬁeld mixes than in

laboratory mixes.
6. The air-entraining agents used in the series of tests can be both give satisfactory
results, but at low dosages only the synthetic detergent admixture can produce an

adequate and stable spacing factor.

CHAPTER 4
CLEAN COAL TECHNOLOGY BY-PRODUCTS:

A LITERATURE REVIEW

4.1 INTRODUCTION

The abundance of coal makes it one of the nation's most important strategic resources in
building a more secure energy future. Coal can be one of the most useful energy sources
of the country well into the let century and beyond. With current prices and technology,
U.S. recoverable resources of coal could supply the nation's coal consumption at current
rates for nearly 300 years (U .S. Department of Energy, 1989).1 However, if coal is to
reach its full potential and be both environmentally acceptable and economically
competitive, an expanded slate of advanced clean coal technologies must be fully
developed and implemented to provide substantially improved options that are superior to

today's choices.

Almost 50% of the cmrent inventory of electrical generating capacity in the United States
will be over 30 years old by 1997 (US. Department of Energy, 1989).1 The need to
replace or refurbish this capacity, plus adding new capacity to keep pace with the rising
demand for electricity, means that a major investment in electrical generation capacity
should begin by the mid 1990's. Better technologies must be available for use on a

commercial basis prior to the year 2,000 to avoid the economic and environmental

4O

4 1
penalties associated with continued investments in only the currently available

state-of-the-art commercial technologies.

Through the 1980's concerns over acid rain and related air pollution issues have had the
strongest impact on coal use. Many countries have taken steps to reduce sulfur dioxide
($02) and nitrogen oxides (N Ox), precursors of acid rain. In the US, the House and
Senate bills amending the Clean Air Act include the provisions aimed at controlling acid
rain through retroﬁt to existing plants. About 40% of the US. electricity generating

units, representing 60% of capacity, could exceed the proposed criteria.

Recent concerns about global warming have also led to growing pressure for national and
international action to curb emissions of " greenhouse gases," mainly carbon dioxide.
Although scientists continue to debate the potential severity and even the existence of
global warming, it makes sense to ﬁnd ways to minimize coal's contribution to C02
emissions. Coal-ﬁred generation and the industrial use of coal account for approximately

one-third of C02 emissions in the United States.

4.2 CLEAN COAL TECHNOLOGIES

In 1983 a coalition of coal producers, utilities and equipment manufacturers proposed that

the federal government join in a collaborative effort to accelerate the development of

42

clean coal technologies designed to provide for effective and economical control of
polluting emissions. This led to the Clean Coal Demonstration Program in 1986,

administrated by the Department of Energy.

Recent work on clean coal has focused on reducing emissions of 802 and NOX, which
contribute to acid rain, through intervention before, during and after combustion.
Technologies include:
1. Coal Cleaning and Upgrading
2. 802 Control
-Lime/Limestone Flue Gas Desulfurization (F GD)
-Fumace Sorbent Injection (F SI)
-Duct Sorbent Injection
3. NOx Control
4. Fludized-Bed Combustion

5. Integrated Gasiﬁcation/Combined Cycle (IGCC)

4.2.1 Atmospheric Fluidized Bed Combustion

In ﬂuidized-bed combustion, a granulated bed material (limestone or dolomite)
encompasses the pulverized burning coal. This mixture of crushed coal and limestone is

held in suspension by an upﬂow of combustion air. The coal burns in the bed and in the

4 3
freeboard above the bed while the limestone is calcined to form lime which absorbs sulfur

oxides, reducing their emission into the environment. This process generates large
quantities of spent dolomite or limestone materials, magnesium or calcium sulfate, ash,
and some unburned coal. During combustion, a portion of this material is continuously
withdrawn to maintain optimum bed conditions and is carried by the combustion gases
through the furnace to cyclones where a substantial fraction of the particulate matter is
collected and recycled to the bed. The remaining particulate matter is carried by the ﬂue
gas to a fabric ﬁlter or electrostatic precipitator, where it is collected for disposal or use.
Figure 4.1 presents a ﬂow diagram of AF BC process located at Michigan State

University, East Lansing, Michigan.

The ﬂuidized-bed process generates by weight about three times as much spent bed
residue as ﬂy ash, and the total quantity of ﬂuidized bed residues represents the total of
the original dolomite or limestone sorbent plus the solid ashes ﬁom the burned coal and

the chemical combination of sulfur and other reactive materials.2

The start—up of an atmospheric ﬂuidized-bed boiler requires heating a portion of the bed
to 315° C (600° F) to ignite the injected coal. After ignition, the temperature of the bed
rises until the energy released in the bed equals the energy absorbed by the boiler tubes
plus that in the hot gases leaving the bed, and the system achieves normal equilibrium.
Excellent heat transfer and high heat capacity is maintained in a ﬂuidized-bed boiler at a

uniform temperature, typically in the range of 800 to 900° C (1472 to 1652 °F).2

4 4
High heat-transfer coefficients between the ﬂuidized-bed and immersed steam generation

surfaces reduce the steam tubing requirements and also permit operation at lower and
more uniform bed temperatures. The lower temperatures reduce NOx emissions and slag,
decrease equipment corrosion, and produce ashes with chemical and physical properties

which differ somewhat from those of conventional ash.

 

 

 

—-— SUPERHEATEN

—- NOT CYCLONE!

—- E COMOMIZEH

HEL FEEDERS—

  

COWUSTDR—

-—-- TUBUL All AIHHEATEI

1'

LIME STONE —

BED ASH COOLEﬁs—

[0. FM

 

 

 

 

 

Figure 4.1 AFBC Process Located at Michigan State University.

4 5
4.2.2 Calcium Spray Drying

Spray drying is a process in which a slurry or solution is atomized into a hot gas stream to
evaporate essentially all of the moisture. A dry product can then be separated from the
gas. The spray dry F GD process consists of hot ﬂue gas leaving the boiler and entering
the spray dry chamber. Figure 4.2 is a schematic diagram of this process. In this reaction
vessel, the gas is contacted with an atomized alkali solution, such as a lime slurry.
Atomization can be achieved with either a rotary or nozzle-type atomizer and is necessary
to maximize the SOz/reagent contact. The water in the slurry is evaporated by the ﬂue
gas as the sulfur dioxide in the ﬂue gas reacts with the alkali, forming dry sulﬁte and
sulfate particles.3 A variety of calcium-based and sodium—based alkalis have been

evaluated, but slaked lime has the greatest commercial potential.

Some spray dryers act as a cyclone and collect some of the dry material in hoppers at the
bottom of the contact chamber, while others rely on fabric ﬁlters or electrostatic
precipitators to remove all the dry waste products from the gas stream. A ﬂow diagram
for such a system is shown in Figure 4.3. Although both systems work well, baghouses
have an advantage over ESPs because some 802 removal is achieved on the bags. In
either case, the ﬂue gas exiting the spray dryer is directed to the particulate control
equipment. The ﬂue gas exiting the spray dryer is never allowed to reach its saturation

temperature and is normally maintained at a temperature at least 11°C (20°F) above the

4 6
saturation temperature. Normal approach to saturation temperatures may range ﬁom -6 to

10°C (20 to 50°F)“.

In some installations, it is desirable to reheat the ﬂue gas before it enters the particulate
control equipment. Reheating prevents condensation in downstream equipment when a
very low approach to saturation temperature is used. The need for reheat and the type of

reheat is dependent on speciﬁc site conditions.

In a lime-based system, dry waste material is sometimes recycled to improve the system's
lime utilization. The waste material recycled contains ﬂy ash, unreacted lime, and waste
reaction products. This also is the material collected by the particulate control equipment
for later disposal or utilization. The solid waste is a dry ﬂowable material with handling
characteristics similar to ﬂy ash. Normally the mixture is pneumatically conveyed to an

ash silo for storage prior to disposal.

ll

.31

 

 

 

 

 

 

WNW

 

 

 

 

 

 

 

 

‘l

 

 

 

“Q" I.
u".

:ilq—Z‘eﬁwg’“

Figure 4.2 Spray Dry FGD System Schematic

 

 

 

 

47

 

 

 

 

 

 

   

 

cum OAS TO
AM“
”01' OI m on In“
--: -ﬁ
no. I
. I Ct to
—
‘ I . l 81'“!
g ‘ I“
All
was:
on I.
”I”?
“I“.

 

J mun IICVCtl 0‘ noun
n

l

 

.0. .7 met oouooo
uu'u‘ov on m cutout
tour

Figure 4.3 Typical Spray Dryer/Particulate Collection Flow Diagram.

4.2.3 Limestone Furnace Injection

Combustion-zone injection involves the addition of sorbent or reagent directly to the
boiler along with pulverized coal. By injecting the sorbent directly into the combustion
zone, the eﬂective residence time is prolonged, resulting in greater S02 conversions
Studies performed to date indicate that limestone injection is capable of removing from
50 to 80% $02, but that greater removal (up to 90%) can be achieved using nahcolite as
the sorbent.° Furnace addition of bicarbonate also yielded good results; however, sorbent

injection into the combustion zone caused several problems related to slagging, fouling,

4 8
and corrosion.5 The amount of S02 that can be captured is dependent on the type and
amount of sorbent, its mixing with combustion gases and ﬂy ash in the fumace, and its
thermal history.° The highly reactive limestone and dolomite, when used as the injected
sorbents, can create chemical scaling problems in certain absorber designs under
particular operating conditions. In addition, it is desirable to avoid contact between the
ash collected from a limestone injected boiler and water until the material is conveyed to
its ultimate disposal. This is good practice because the dry/wet interfaces in the removal

system may be a source of plugging or deposit buildup, which would reduce the capacity

of the conveying system.7

A promising 802 removal technology utilizes a combined system of furnace limestone
injection with dry scrubbing. In addition to greater possible reduction of 802 than can be
achieved by either system alone, the combined technologies can represent signiﬁcant
savings in reagent costs by using limestone rather than lime. The low—cost limestone is
injected into a furnace and is calcined into lime, which undergoes reaction with sulfur
species in the ﬂue gases. The collected lime and ash materials are then recycled and used
as the principal reagent in the dry scrubbing technology for high-sulfur coals, as well as

for low-sulfur coals.

4 9
4.2.4 Sodium Sorbent Injection

This "dry scrubbing" process removes 802 by injecting a dry sodium-based sorbent into
the ﬂue gas stream ahead of particulate control devices to produce a dry end product.
This dry F GD technology offers unique advantages in terms of low cost and compactness

of required process equipment.

In this process, the ﬂue gas exits the combustor, ﬂows through a duct, then through a heat
exchanger, and ﬁnally through a fabric ﬁlter or electrostatic precipitator. Use of a
baghouse generally produces better results because unreacted alkali on the bag surface
continues to react with the S02. The sorbent injection point has been varied from the
furnace burners to injection directly into the particulate collector,8 with the majority of
studies concentrating on injection into the heat exchanger section or just ahead of the
baghouse, since the bulk of S02 removal occurs as the gas passes through the sorbent

cake which has built up on the bags.9

Initial investigations used nahcolite, a naturally occurring sodium bicarbonate, as the
alkali sorbent, but the lack of large supplies of this material prompted studies using other
alkalis. Only sodium carbonate and sodium bicarbonate were found to be consistently
effective,10 with bicarbonate being the more reactive.8 When dry powdered sodium
carbonates and bicarbonates are pneumatically injected into the inlet duct of a fabric

ﬁlter, a considerable degree of reaction occurs between the sulfur oxides in the gas and

 
 

5 0
the ﬁnely divided sodium compounds. The dry particulate product is then collected with

the ﬂy ash in the baghouse.

Figure 4.4 presents a schematic of a typical dry injection system. Regardless of whether
the sorbents used in commercial applications will be man-made sodium carbonate or
bicarbonates, or their naturally occuning forms, reﬁned trona and nahcolite, waste
products from the dry injection process will be a mixture of ﬂy ash, reaction products,

and unreacted sodium compounds.

 

 

 

 

 

   
  
 

 

 

cum GAS
OUTLET
nus GAS r 41
onv soaps m
pnecoar
onv sonaeur h
INJECTION
HOPPER “mouse
onv .__l [ml

sonaeur

sronacs ”M I

 

 

 

wasrs sronace ___.. WASTE
REMOVAL

 

 

 

Figure 4.4 Typical Dry Sodium Injection System Schematic.

5 1
The waste products from this process will be generated in very large quantities and will

require a signiﬁcant area for land disposal. The waste will contain high levels of soluble
sodium salts and could require pretreatment prior to disposal. Pelletizing and/or
chemical-physical stabilization may be necessary, although regeneration of sodium

compounds from the dry waste using technology from other ﬁelds is being tested.8

The above clean coal technologies, besides controlling the 802 and NOX emissions, also
inﬂuence the emission of the greenhouse gases. With respect to S02, some of the clean
coal technologies improve the efﬁciency of the conversion of coal to useful energy.
Technologies such as ﬂuidized-bed combustion and integrated gasiﬁcation/ combined
cycle will consume less coal per unit of useful energy produced and thus lower the
amount of C02 emitted per unit of useful energy produced. Other clean coal
technologies, however, result in lower net amount of thermal conversion efﬁciency and

hence, slightly increase the rates of C02 emission.

Many clean coal technologies trade off the reduction in polluting emissions with an
increase in the production of solid wastes, particularly ﬂy ash. According to the US.
Department of Energy (1989),1 the use of clean coal technologies may lead to more than
100% increase in the solid waste generation of power plants by the year 2010 (to
approximately 220 million tons per year). Solid waste disposal will thus continue to be a

growing problem for many utilities. Large land areas will be needed for disposal of the

52

increasingly large amount of ﬂy ash produced by the power plants adopting clean coal

technologies.

Stricter regulations on landﬁll design and monitoring add to overall waste disposal costs.
The development of solid waste utilization programs can help many utilities to decrease
their overall waste disposal costs while adopting the emerging clean coal technologies.
Many conventional power plant "wastes" have proven to be marketable by-products.
Today, conventional pulverized coal combustion ﬂy ash is considered an essential raw
material in ordinary concrete, controlled low-strength cementitious materials (e.g.,
ﬂowable ﬁll), and roller compacted concrete. Similarity of the clean coal technology
by-products (ﬂy ashes) to conventional coal combustion ﬂy ash is indicative of the strong

potentials for beneﬁcial use of the clean coal technology by-products in construction.

4.3 CLEAN COAL TECHNOLOGY BY-PRODUCTS

Many of the advanced clean coal technologies produce a "waste" that is comparable to
conventional coal combustion ﬂy ash. The sorbent, which is generally added to control
polluting emissions, is intimately associated with the clean coal technology ﬂy ashes
(Radian Corporation, 1988).“ This "modiﬁed ﬂy ash" includes conventional ﬂy ash
particles that are coated with sorbent and sorbent reaction products, as well as smaller

(non-ﬂy ash) particles consisting of reacted and/or unreacted sorbent. Depending upon

53
the particular clean coal combustion system, the volume of "modiﬁed ﬂy ash" is

approximately twice that of conventional coal ﬂy ash due to the addition of sorbent
material. As the advanced clean coal technologies are employed by full scale commercial
systems, they will result in larger volumes of waste which will have to be managed by the
electric utilities involved. Given the current marketability of conventional coal
combustion ﬂy ash for various applications and similarity of the so-called "modiﬁed ﬂy
ashes" to conventional ﬂy ash, the utilization potentials of these new by-products warrant

investigation.

Conventional coal ﬂy ash is a by-product of the combustion of pulverized coal in thermal
power plants. It is removed by the dust collection system as a ﬁne particulate residue
from the combustion gases before they are discharged into the atmosphere. Conventional
coal ﬂy ash is a siliceous and aluminous material which in itself possesses little or no
cementitious value but which will, in ﬁnely divided form and in the presence of moisture,
chemically react with calcium hydroxide at ordinary temperature to form compounds

possessing cementitious properties.

Fly ash contains alumina-silicates that react with the calcium ions released by Portland
cement in the presence of moisture to form Calcium Silicate Hydrates. Figure 4.5a
schematically shows the structure of hydrated cement paste (no ﬂy ash used). The
structure (see Figure 4.5a) consists of "A" which represents aggregations of Calcium

Silicate Hydrate particles (a stable cement hydration product with desirable strength

54
development and durability characteristics); "H" representing Calcium Hydroxide in the

form of large crystals which are unstable in the presence of water especially when some
aggressive chemicals are also present; and "C" representing capillary void system which
is the route for entry of water and aggressive chemicals responsible for the aging of

concrete.

The structure of ﬂy ash-cement paste is presented in Figure 4.5b. The major change is
that the Calcium Hydroxide ("H") crystal products, as a result of pozzolanic reaction,
change to Calcium Silicate Hydrate ("A") particles of low density which ﬁll the large
capillary voids and reduce the continuity of the capillary void system. This blocking of
capillary voids reduces the permeability of concrete. Reduced permeability and
substitution of Calcium Hydroxide with Calcium Silicate Hydrate in the presence of ﬂy
ash, which is used in concrete as a partial substitute for Portland cement, provide
improvements in resistance against sulfate attack, alkali-aggregate reaction, and diffusion
of chloride ions which are responsible for the corrosion of embedded steel reinforcement.
Fly ash is also capable of enhancing the workability of fresh concrete mixtures and
reducing the heat of hydration of concrete. There are, however, concerns regarding the
adverse effects of ﬂy ash on the deicer salt scaling resistance and air entrainment for

freeze-thaw durability of concrete.

55

 

(a) Cement Paste

 

(b) Fly Ash-Cement Paste

Figure 4.5 Structure of Cement Paste Versus Fly Ash-Cement Paste.

5 6
4.4 POTENTIAL APPLICATIONS

Technical and economical advantages of utilizing conventional coal ﬂy ash in concrete
have led to the development of a large-volume market for ﬂy ash accounting for half of
all the ﬂy ash utilized in the United States (approximately 10 million tons per year). In
the recent years, with the increased production of ﬂy ashes failing to satisfy speciﬁc (e.g.,
Loss On Ignition) ASTM requirements for pozzolanic concrete admixtures, many new

techniques have evolved to utilize conventional coal ﬂy ashes in cementitious materials.

By-products of three major clean coal technologies (Calcium Injection, Furnace Injection,
and Spray Dryer) possess chemical and physical characteristics which closely simulate
those of conventional coal ﬂy ashes. Tables 4.1 and 4.2 compare the chemical and
physical characteristics, respectively, of these clean coal technology by-products with
those of conventional coal ash. These tables are indicative of the potentials of clean coal
technology by-products to repeat the success of conventional coal ﬂy ash in ﬁnding high-
value and large-volume markets in concrete. The relatively high CaO content of the clean
coal technology by-products, when compared with conventional coal ﬂy ash, suggests
that these by-products present not only pozzolanic qualities but also cementitious
potentials in the sense that, in the presence of moisture, they themselves may be capable
of releasing sufﬁcient calcium ions for strength development through pozzolanic
reaction. This phenomenon Can reduce the need for Portland cement as a source of

calcium ions in concrete.

57

 

 

 

 

Table 4.1 Chemical Characteristics (%) of Clean Coal Technology By-Products and
Conventional Coal Fly Ash (ICF Northwest, 1988 & Malhorta).
By-Product A1203 CaO F303 MgO K20 SiOL E20 803
Calcium Inj 31.4 15.4 8.9 1.1 3.4 30.0 1.2 NA
Furnace Inj 30.0 16.8 16.9 0.7 2.1 27.9 1.0 3.5
Spray Dryer 24.9 20.0 6.5 2.6 0.8 21.3 1.8 10.25
Coal Ash 12-23 1-13 3-26 1-3 1-3 35-60 .1-5 .2-4

 

 

 

 

 

 

 

 

 

 

 

 

NA - Not Available

Table 4.2 Physical Characteristic of Clean Coal Technology By-Products and

Conventional Coal Fly Ash (ICF Northwest, 1988).

 

 

 

 

 

 

By-Product % Retained on No. 325 Sieve
Calcium Injection 6
Furnace Injection 20
Spray Dryer 3
Conventional Coal Ash 14—35

 

 

 

4.5 UTILIZATION IN CONCRETE
4.5.1 Chemical and Physical Requirements

Six advanced 802 control by-products samples were used in a laboratory testing program

reported by Perri et a1. (1987):12 an AF BC baghouse ash, an AF BC spent bed material,

58

two calcium spray dryer ashes, a limestone ﬁrmace injection ash, and a dry sodium

injection ash. Table 4.3 gives the by-product sample numbers (which are referred to in

the subsequent table) and shows where the by-product samples were obtained from. The

chemical and physical analyses of the by-products are presented in Tables 4.4 and 4.5

 

 

 

 

 

 

respectively.
Table 4.3 By-Product Samples.
Sample No. Plant Utility Location
AF BC:
TV03 (bed) Shawneea Tennessee Valley Paducah, KT
TV05 (ash) Shawneea Tennessee Valley Paducah, KT
Spray Dryer:
LR07 Laramie River Basin Electric Power Wheatland, WV
HSOS Holcomb Sunﬂower Electric Holcomb, KS
Limestone Injection:
OL03 Lakeview TGS Ontario Hydro Mississauga, Ontario
Sodium Injection:
NX04 R.D. Nixon City of Colorado Spring Colorado Springs, CO

 

 

 

 

 

59

 

 

 

 

 

 

Table 4.4 Chemical Characteristic.

Sample No. A1203 CaO Fe203 MgO K20 SiOz NaZO SO3
AFBC:

TV03 2.72 45.1 4.77 0.62 0.31 3.17 0.27 6.50
TV05 15.0 22.6 18.9 0.51 1.93 15.3 0.34 17.3
Spray Dryer:

LR07 21.1 26.9 6.11 2.33 0.74 17.7 2.08 12.3
HSOS 24.9 30.0 6.51 2.62 0.75 21.3 1.81 10.3
Limestone
Injection: 17.8 36.1 13.2 0.63 1.11 15.8 0.48 6.25
OL03

Sodium
Injection: 28.9 4.54 2.50 1.16 0.77 25.2 24.8 12.0
NX04

 

 

 

 

 

 

 

 

 

As shown on Table 4.4 all the by-products, except for sodium injection ash NX04,

primarily consist of calcium, silica, alumina, sulfur, and iron oxides. Calcium oxide

concentrations ranged from 20-45% by weight for the calcium-based advanced S02

control by-products. The sodium-based process by-product had 25% by weight

concentration of sodium oxide and a relatively low CaO content of 4.5%. The by-product

silicon oxide concentrations (3 -25% by weight) were much lower than would be

encountered in conventional pulverized coal (PC) ﬂy ash which tends to have an Si02

content in the 30-60% range.13 The sulfur trioxide concentration of the by-product

samples ranged ﬁ'om 6-17%; these concentrations are much higher than normally found

in conventional pulverized coal ﬂy ash which are generally under 3%.13 All other

 

6 0
parameters (A1203, Fe203, MgO, and K20) are present in concentrations which are

similar to those found in conventional pulverized coal ﬂy ash.

Table 4.5 provides the by-product physical analyses test results. All of the by-products,
except for the AF BC spent bed material, were ﬁne ashes with the majority of particles in
the silt size range (2-75 um). The majority of AF BC spent bed material particles fell in
the coarse sand size range (425 um-Z um). The physical property with the most
variability among the six by-product samples analyzed was permeability; the by-product
permeability values varied from 3.4 x 10‘4 cm/sec (1.34 x 10'4 in/sec) for the AFBC

spent bed to 7.7 x 10'9 cm/sec (3 x 10'9 in/sec) for the AF BC baghouse ash.

61

33.. .E: at can 02 a :3 ....s .2. 3.5.3 . 32......
0.3.. .53 ms. 02 02 o :0 .3532 on ten 33; on 02 a :3 ....s .o... 3.....3 . aco- 2...
33; .E: a... 02 oz o :0 3:33. on uco 3.3: 9 oz o :3 ....s .2. 3.2.33 . E3 333
.0)... .EE 3 o. 02 o co ©0530. on :5: .2... 3.2.35 . .23.. #030 5.54 c. “.2233. no Caz-$30 o
p.324 82 . «z
.9353 2.3.3.302 . 32
.9322. 2.3.3.3.. .o 3333 .E... v.3... oz . ..z

 

 

 

 

 

 

 

 

 

 

 

.352
9...... o . ... o. 2 3 a. a 8. a o . 3 .99.
o 69.3.... :5.va
«.9... ~ 32 on 3 oo oo 2 on co 3 . o 3.0
39.3.... 0.2.5..
5.9.5 . m v~ o~ mm no 3 2. mo 0 . a new:
50.... o A. on 2 .o 3 «m 3 3 . ~ (2 3...
203.0 30:;
0.0.: 3 m an an no on so nm 8. o o o 233 mo>p
«.9... n 32 .2 an o~ no 0. oo. o o. ~3 o. 309. 2.)»
9.34
Corn... ‘0 or «o 3.0 :3. w... 9...: be: .233
3553633 .2... 2.3.3 .2... .23.... 23.20.). 636...... 3.33 . o5. 3.3.. .02 0.9.3.3.
53.530 .
.3.
.33.. 3.33 . 533.55 2.» 23......

 

 

 

 

 

 

 

 

mu.hm.¢uhu<¢<xu .(U_m>z3 FUDOO¢3$O

2.3.8.0820 18.92.. m... 2%....

62

The technical criteria for the utilization of conventional coal ﬂy ash in cementitious
materials are provided in ASTM C 618. These criteria cover: (1) Silica, Alumina, and
Iron Oxides (SiOz+A1203+Fe203); (2) 803 content; and (3) Particles passing No. 325
sieve. The clean coal technology by-products are observed in Table 4.6 to satisfy these
speciﬁc criteria There is, however, a need to fully evaluate these by-products against all
the criteria speciﬁed in ASTM C618; some speciﬁc concerns relate to: (l) potentially
high sulfate content of some clean coal technology by-products in the form of the reaction
products of sorbent with the $03 in ﬂue gas; and (2) potential increase in the Loss On
Ignition (L01) of clean coal technology by-products due to insufficient ignition of the
sorbent and its reaction products. Any potential problem in these areas will encourage the
adoption of a performance-based approach to the use of clean coal technology by-
products in concrete, and will increase the emphasis on lower-grade (e. g. ﬂowable ﬁll)
applications of the by-product where the ASTM standards for pozzolanic concrete

admixtures should not be necessarily satisﬁed.

ASTM C618 provides speciﬁcations for use of ﬂy ash as a mineral admixture in Portland
cement concrete. Table 4.6 presents by-product physical and chemical analyses along
with the corresponding ASTM C618 physical and chemical requirements reported by
Perri et al. (1987).12 None of the by-products meets the ASTM C618 requirements for

combined SiOz + A1203 + Fe203 or for 803 concentration. Only the spray dryer ash

6 3
HS05 had an LOI concentration below 6.0%. All by-products, except the AF BC spent

bed material, meet the limits of percent retained on a No. 325 Sieve and for autoclave
expansion. Furnace injection ash OL03, AF BC baghouse ash TV05, and spray dryer ash

H805 achieved pozzolanic activity indexes above the required 75%.

Table 4.7 provides the results of the cement replacement testing for three of the by-
product samples. The concrete mixes were prepared with 30% by weight replacement of
the Portland cement with advanced 802 control ash. Even though these ashes did not
meet the ASTM C618 chemical requirements, they did produce acceptable quality
concretes when compared with both the no ﬂy ash and conventional ﬂy ash control
mixes. Although the spray dryer ash mixes had exceptionally long setting times (17-21
hours to ﬁnal set), their compressive strength (even at early ages) and durability exceeded
the 100 percent cement control. These mixes also performed very well in freeze/thaw
testing; the spray dryer ash concrete mixes showed no surface deterioration after 300

freeze/thaw cycles.

64

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4.6 Properties of Clean Coal Technology By-Products Vs. ASTM C 618
Requirements.
SiOz
+
A1203 Retained Pozzolanic
+ on No.3 25 Activity Autoclave
Sample No. Fe203 S03 LOI Sieve Index Expansion
Reference
Speciﬁcation:
ASTM C618 50-70(min) 5.0 (max) 6.0 (max) 34 (max) 75 (min) 0.8 (max)
ASTM 595* 20 (max) 75 (min)
AFBC: .
TV03 (bed) 10.66 6.50 100 NA NA
TV05 (ash) 49.18 17.25 12.64 0 88 NA
Spray Dryer:
LR07 45.03 12.25 10 69 +0.05
H805 52.71 10.25 3 107 +0.03
Furnace
Injection: 46.72 6.25 10.97 20 108 +0.08
OL03
Sodium
Injection: 56.58 12.00 0 NA NA
NX04
Key: NA - Not Analyzed
* ASTM C595 requirements for pozzolan for use in blended cements.
By-Product SiOz+Ale3+FezO3 % SO3 % % > #325
Calcium Injection 70 NA 6
Furnace Injection 75 3.5 20
Spray Dryer 53 NA 3
ASTM C 618 Min. 50 Max. 5 Max. 34

 

 

 

 

 

 

 

6S

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4.7 Cement Replacement Testing Result
Spray Spray Furnace Control Control
Dryer Dryer Injection #1 #2
LR07 HSOS OL03 (no ash) (conv.ash)a
Mixture Pr0portions (1 yd’)
Cementitious Material
1. Cement (lbs)b 410 410 410 590 410
2. Ash (lbs) 180 180 180 0 180
Fine Aggregate (lbs) 1380 1380 1380 1380 1380
Coarse Aggregate (lbs) 1875 1875 1875 1875 1875
Admixtures
1. Air Entrainment (ozs)c 7.73 7.73 25.1 7.92 4.42
2. Water Reducer (ozs)d 12.3 12.3 12.3 17.2 12.3
Water (lbs) 265.5 265.5 165.5 247.5 211.5
General Data
Slump (inches) 4.0 4.0 3.5 4.5 4.5
Air Content (%) 5.8 5.6 4.5 5.6 6.0
Unit Weight (lb/113) 147.1 146.3 133.3 145.7 147.1
Concrete Temperature (°F) 83 79 84 NA NA
Time of Set (hrzmin)
Initial (500 psi) 13:55 18:30 3:50 6:00 6:15
Final (4000 psi) 17:00 21 :15 5:35 8:00 8:00
Compressive Strengthe
3-day (psi) 3523 3607 2063 3280 3140
7-day (psi) 4557 4645 2581 4280 4100
28—day (psi) 5682 5093 3784 4940 5310
90-day (psi) 6331 6343 6083 5626 6740
Modulus of Rupture (psi)I 597 630 543 763 702
(at 28 days)
Freeze/Thaw Durabilityg
Durability Factor 99.3h 94.2h 89.6 92.5 98.25
Length Change (%) 0.04 0.04 0.07 -0.02 -0.11
Notes:

a Conventional ﬂy ash from a pulverized coal-ﬁred boiler collected by ESP from the United Power

Association's Coal Creek Station in Underwood, North Dakota. A lignite ash with approximately 40%

SiOz, 12% A1203, 7% Fe203 and 24% CaO.

b ASTM C150 Type I cement.

c Air entrainment agent - Master Builders MB-VR.

d Water reducing agent - Gillford Hill PSI

e ASTM C39, average of three samples.
f ASTM C78, average of three samples.

g ASTM C666, procedure A @ 28 days, average of three samples.
h No surface deterioration after 300 freeze/thaw cycles.
i Slight surface deterioration after 300 freeze/thaw cycles.

 

CHAPTER 5
CONCRETE MICROSTRUCTURE:

A LITERATURE REVIEW

5.1 INTRODUCTION

Concrete is a porous structure, usually containing 20 to 40 percent by volume of void
space. This void space strongly inﬂuences the concrete performance, characteristics.
Diﬁ‘erent pore sizes affect different characteristic like strength, shrinkage, permeability,
creep, freeze-thaw durability, etc. This chapter reviews the past research studies

concerning concrete pore system.

5.2 PORE STRUCTURE

In general, concrete microstructure can be categorized into entrapped air void system,
entrained air void system, capillary pores, and gel. The entrapped air void system has
irregular voids which are greater than 1 mm in size. This system is the typical symptom
of improper handling of fresh concrete, and may cause signiﬁcant drop in the strength of
hardened concrete. It is usually caused by a too stiff fresh concrete or an improper

consolidation practice such as vibrating too short, using a vibrator head which is too

66

67

The entrained air void system is between few microns and 1 millimeter in void size. The
entrained air void system is purposely put into the concrete structure. These void system
has a spherical shape and a uniformly distributed void size. A well entrained air void

system can help hardened concrete overcome the freezing and thawing problem, and can

enhance the concrete durability. Past, researchers have intensively studied this area.

Capillary void system represents the unﬁlled space in the hydrated cement paste. They
range in size from 0.01 micron to few micron. Concrete consists roughly one part by
volume of cement with three parts of water. During hydration, cement reacts with water
and produces hydration products. There exists a gel gap in between hydration products
because they can not fully occupy the space previously occupied by cement and water.
Excess of mixing water thus causes higher capillary porosity. The capillary pore system
is somewhat continuous; this is partly caused by the evaporation of excess water through
microchannels towards the surface. Continuity of the capillary pore system increases the

permeability of concrete and damages its durability.

68

5.3 CAPILLARY PORE SYSTEM CHARACTERISTICS

The microstructure of Portland cement paste was ﬁrst studied in details by Power et a1.
(1946)6. This study concluded that: (1) when a unit volume of cement hydrates, it
produces 2.2 volumes of hydrated material; (2) the hydrated material, commonly called
gel because of its colloidal nature, has an inherent porosity of about 26 percent; (3)
because of the small size of the gel pores, they attract any water that is available, and
when the water enters the pores, since much of it is adsorbed water rather than free water,

it has a speciﬁc volume of 0.9.

Based on the above conclusions, a unit mass of Portland cement would react with a
certain (~ 0.38) unit mass of water in the hydration process. When in a cement paste the
water-to-cement ratio exceeds this level, after the completion of hydration there will still
be unﬁlled space in the original cement and water ﬁlled space. As hydration process
continues, this space would be subdivided into isolated pockets. Power calls these
isolated pockets “capillary pores” in order to distinguish them from the gel pores. Under
the condition of water to cement ratio less than 0.38, there would be insufﬁcient water to
hydrate the cement, and thus portion of cement remains unreacted. Unless the hydration

process continues over time, there will exist unreacted particles in the cement paste.

Furthermore, Power observed that the water in the gel will not freeze at temperatures as

low as -20°C when water in capillary pore did freeze. More recently, Kashi and Weyer

69
(1989)14 conﬁrmed these ﬁndings. The relationship between pore size and freezing

temperature is presented in Table 5.1.

Table 5.1 Rel3ationship Between Pore Size and Freezing Temperature (Philleo, et
al.)

Freezing Temperature Size of pore required for ice formation at
selected
nm
-2 ' 20.8

-6 8.3
-10 5.9
~15 4.4

-20 3.5

 

5.3.1 Relationship between Strength and Porosity

The most popular property to evaluate concrete is compressive strength. Three types of
semi-empirical models for predicting concrete strength in terms of its porosity have been
widely used12 :

Balshin model: 6 = 00 (1-P)A
31>

Ryshkewitch model: 0' = 60 e'

Schiller model: 0 = C 1n (PCR / P)

70

In the Balshin and Ryshkewitchu" models, 60 represents the compressive strength at zero
porosity and P is the porosity. In the Rayshkewitch’s model, the PCR is zero strength
porosity. A, B and C are material constants. In general, the Ryshkewitch model is

accurate at low porosity level, but the Schillier’s model is good at high porosities.

Brownde and Power present their model of compressive strength of Portland cement
paste in terms of the gel to space ratio6:
Strength = constant 0 x3

where x = 2.06 Vc a / (Vc + Wolc)

Where c is the weight of cement, V6 is the speciﬁc volume of cement, W0 is the volume
of mixing water, and on is the degree of hydration of cement. This equation is very
sensitive to small changes when the gel to space ratio is above 0.2. Furthermore, in
Power’s later research, he simpliﬁed the relationship of porosity and strength. Power
found that the 28-day compressive strength fc was related to the gel/space ratio, or to the
ratio between the solid hydration products in the system and the total space:

Fc = a x3
where a is the intrinsic strength of the material at zero porosity (p) and (x) the solid/space
ratio or the amount of solid fraction in the system, which is therefore equal to (1-p).

Power’s data are shown in Figure 5.1. The value of (a) was found to be 34,000 psi (234

MPa).

71

 

 

”9 16— MORTAR

~" oMilA ,

.E ‘4'- lMixB

3 AMixC

". 12 1-

.0

to

.8

a 8 '-

0

.5 6 1-

N

~

0 4 r- A

f

a

g 2 - 1.554.000):3
a l 1 l 1

 

 

O
O 0.2 0.4 0.6 0.8 I.O
Gel/Space Ratio (11)

Figure 5.1 ‘Relationship of Strength with Gel/Space Ratio ( Powers, 1958).'

Jambor (1973)9 did the study on the pore structure by measuring the pore size
distribution, pore volume, and compressive strength. He concluded that the higher

compressive strength, the smaller the average pore size at a given porosity. Furthermore,

he expresses the low-porosity compressive strength as:

s=KJPo—P /(w/c+P)

72
where the P is the total porosity, w/c is the water to cement ratio, P0 is the porosity at the

onset of the hydration, and K is a coefﬁcient dependent on the type of cement, curing

conditions, and volume concentration of cement in the paste.

Odler (1972)19 developed a relationship between strength and porosity where porosity
was measured, using the mercury intrusion technique. The following relationship was
derived under the conditions cement hydration at 25°C and with the water-to-cement ratio
ranging from 0.25 to 0.31:

S = So ' a P<10nm ' b P10-100nm ' C P>100nm

where a, b and c are constants, P is porosity, and SO is the strength at zero porosity.

Beaudoin and Feldman(1994)4 concluded in their research on and modulus of electricity
that the relationship between elastic modulus and porosity can be given as:

E = E0 CCBP.

5.3.2 Relationship between Permeability and Porosity

There are ntunerous ways to measure the permeability of hardened concrete and paste.

All of the methods measure how the liquid or electron can pass through the capillary

pores. This depends on how big the thread of the capillary pores is and how well they are

73
connected. The fundamental relationship of permeability and porosity can be expressed

from the Hagen-Poisevill law as follows”:

Q/A = K AP/(nh)
Where Q/A is the volume ﬂux passing through the specimen area A, AP is the pressure
difference between the two sides of the specimen, 1] is the liquid viscosity, h is the
thickness of the specimen, and K is the coefﬁcient of permeability. K can be related to
the average pore radius yt by the equation of:

K = w 37,2

where w is a form factor (for example 1/ 8 for cylindrical pores) and s is the porosity.

A more exact expression which includes the pore size distribution is presented in the
following:
K = C I r2 f(r) dR

where C is constant.

Setzer et al, (1973)10 discussed the main factors correlating the permeability with the pore
size distribution. They concluded that an average pore radius aﬂects permeability, and
derive a simpliﬁed equation based on the Hagen-Poiseville law”, using an effective
porosity of s/ 3. The value rm determined from mercury intrusion data was taken from the
maximum of differential curve. The ratio rm/rt increased greatly with water-to-cement

ratio, and rt was always smaller than rm.

74

Nyame et al.13’20

conducted a similar research on the relationship between pore structure
and permeability. They used the technique of mercury intrusion to obtain the pore size I
distribution, and concluded that a linear relationship exists between log (water
permeability) and mean pore radius. In this study, they correlated the pore structure and

permeability using the radius obtained from the differential mercury intrusion curve:

K = 1.684rm3 x 10'22

The value of the correlation coefficient was 0.96, but for the values of rm less than 100

nm the scatter of data was found to be large.

Bier et al,(l 988)20 measured permeability of water through oven-dried and non-oven-
dried specimens. They found that oven drying creates visible cracks and increases
permeability by about an order of magnitude. These high values, however, decreased
during further exposure to water and reached the same values as observed for samples not
dried in a short period. This result occurred despite the fact that the oven dried samples
had coarser pore structures as determined by mercury intrusion after the experiment. It
was concluded that the deposition of particles into the cracks rapidly reduces
permeability. These results emphasize the difficult in making precise conclusions with

regard to pore size distribution-penneability relationships.

75
5.3.3 Effects of Pozzolans on the Pore Structure of Hardened Concrete

Cebrera has concluded that ﬂy ash additions to concrete leads to a more reﬁned pore
system. The pore structure in concrete is altered in at least to ways: (1) ﬂy ash will
reduce the amount of water needed to maintain good workability, thus producing a tighter
matrix; and (2) ﬂy ash will react with calcium hydroxide produced dtuing hydration to
form products which will have a tendency to ﬁll empty spaces in the matrix, thus creating
a more discontinuous pore system. Figure 5.2 shows the pore size distribution of an
ordinary Portland cement mortar designed for a compressive strength of 40 MPa and a 50
mm slump. It can be seen that the total porosity decreases with time and that the volume
of large pores (>01 mm) decreases sharply with age. On the other hand, Figure 5.3
depicts the pore size distribution of a mortar obtained from a ﬂy ash concrete containing
30 percent ﬂy ash by weight of cementitious component. It was designed to meet the
conditions stated above for the plain matrix, but the water required to maintain constant
slump was reduced by 9 percent. Examination of Figure 5.3 shows that the total early
porosity (at 3 days of age) was reduced, perhaps due to the lower water content. It is also
evident from the ﬁgure that the volume of large pores (>0.01 mm) has been reduced in

the ﬂy ash mortar, particularly at later ages.

76

28 Days

F
0.5 - °
0 0.3 We . °

A 0.4 .
D 0.5 .

O 0.5

' ‘ 0.7

0.3 ._ I 0.8 .

0 0.9

0.2 - ,

0.1 - .

 

Penetration Volume (cc/g)

 

 

 

Pore Diameter 3

0.5 F

0.4 .—

07 VIC

0.3 *-

0.2 ..

0.l {-

Penetration .Volume (cc/cl

 

 

 

Figure 5.2 Pore Size Distribution Versus Age in Ordinary Portland Cement Mortar].

77

instrusion volume (2)

 

 

 

 

 

 

5° 1
tale-merit
0 Ida,
0 Joe
40 :7“: “0
"do
A Y.
A
0
so 2

  
    

 

 

 

 

 

 

20

 

 

 

10

 

 

 

 

 

 

 

 

i

 

 

09059.0.0

 

 

 

 

  

 

 

 

 

20

 

10

 

 

 

 

 

 

 

20 2 0.2 0.02 0.002
equivalent pore diameter (um)

Figure 5.3 Pore Size Distribution Vs. Age in a Mortar Containing Fly Ash.

78
Mehta and Manmohan (1981)20 have studied the effect of pozzolans on the pore size

distribution of hardened cement paste. Class F ﬂy ash was added at 10, 20 and 30 percent
by weight of cementitious material, and type 11 Portland cement was used throughout the
investigation. The pastes were prepared using distilled water at a water to solid ratio of
0.5. In order to reduce bleeding, the pastes were mixed for 30 minutes. Mercury
porosimetry was used to evaluate the pore size distribution of the samples (the contact
angle used was 130°). The results obtained are shown in Figure 5.4 a and b, where it can
be seen that the pozzolanic lime-silica and lime-alumina reactions were slow. For
example, at 28 days of age, all ﬂy ash pastes exhibited a coarser pore system than the
corresponding control pastes. After 1 year of curing, all pastes exhibited ﬁner porosities,
but the ﬂy ash pastes had more reﬁned porosities than the controls. These results indicate
that the pore reﬁning effect of a pozzolan is a ﬁrnction of its pozzolanic reactivity.
Moreover, it was concluded that type F ﬂy ash was not effective in reducing the volume

of larger pores at early ages.

79

penetration volume (cc/g)
0.30

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 ‘O 10% FA >
0.25.4. A 20: FA -

0 I 30: FA 28 days
o.2o«-—n..'_'i°""°'_

db }l ”a?
0.15
0.10
0.05

 

 

 

 

 

 

 

Q t 3 e : ,
150 100 80 60 50 4O 30 20 15 10 7 4,5
pore diameter (nm)

Figure 5.4 Fly Ash Effects on the Pore Size Distribution of Pastesl.

Hooton studied the effcts of ﬂy ash at 25 and 35 percent replacement levels on the pore
size distribution of sulfate-resistant Portland cement (Type V). Two different water-to-
cement ratios of 0.25 and 0.36 were used in this study. The pore size distributions were
measured by mercury intrusion (a contact angle of 140° was used). The results showed
that the pore structure of the pastes containing ﬂy ash (water-to-solids ratio of 0.25) was
not ﬁner that of controls. However, this was found to be inconsistent with the strength
and permeability data obtained for equivalent samples, which showed evidence of
continued pozzolanic reaction with time. For pastes having a water-to-cementitious

material ratio of 0.36, at 25 percent ﬂy ash the results showed coarser porosities at 7 days

80
at curing, but at 28 days the porosities of both control and ﬂy ash pastes were found to be

similar. At this same age for the 35 percent replacement level, the volume of pores larger
than 0.025 mm was found to be smaller than that of controls. After 28 days, higher
porosities were noticed for pores ﬁner than 0.005 mm, indicating the presence of larger

quantities of gel pores.

CHAPTER 6

EXPERIMENTAL PROGRAM

6.1 INTRODUCTION

The experimental program presented in this Chapter has been designed to provide
comprehensive and statistically reliable information on the impacts of ﬂy ash and calcium
spray dryer ﬂy ash (a clean coal technology by-product) on the freeze-thaw durability,
deicer salt scaling resistance and relevant engineering and microstructural properties of
concrete. The results can help distinguish between the effects of conventional ﬂy ash and
a speciﬁc clean coal ﬂy ash on concrete properties, and also provide a basis to develop
structure-property relationships and illustrate the mechanisms of action of conventional
and clean coal ﬂy ash in concrete materials; (this research emphasizes the freeze-thaw

durability and deicer salt scaling resistance of concrete).

The effects of ﬁve variables on the structure and engineering properties of concrete were
investigated in a 25 factorial design of experiments (see Table 6.1):

0 ﬂy ash type (conventional Class F and Calcium Spray Dryer)

0 ﬂy ash content (15% and 30% replacement by total weight of cement)

e water-cementitious material ratio (0.40 and 0.45 by weight)

- moist curing period (28 and 56 days)

0 air content (2.5% and 7.5%)

81

82

Control concrete specimens (with 0% ash content) were also considered at all the levels

of water-cementitious ratio, moist curing period and air content.

Table 6.1 25 Factorial Design of Experiments

 

w/c Ratio 0.40 0.45

 

Ash Content 15% 30% 15% 30%

 

Air Content 2.5% 7.5% 2.5% 7.5% 2.5% 7.5% 2.5% 7.5%

 

 

 

 

 

 

 

 

 

 

 

 

 

Curing Period 28 56 28 56 28 56 28 56 28 56 28 56 28 56 28 56
Class F :1: at: a1: :1: at: :1: a1: :1: :1: :1: e at: :1: at: at: :1:
Calcium * * 1! it It * 3F It: * * * III * * 1‘ *
Spray

Control * * * * at: :1: at: *

 

 

 

 

 

 

 

 

 

The concrete specimens produced based on the above experimental design were subjected

to the following tests:

0 ASTM C 666, Resistance of Concrete to Rapid Freezing and Than

0 ASTM C 672, Scaling Resistance of Concrete Surface Exposed to Deicing Chemical

0 ASTM C 1202, Electrical Indication of Concrete’s Ability to Resist Chloride Ion
Penetration

0 ASTM C 39, Compressive Strength of Cylindrical Concrete Specimen

o ASTM C 457, Microscopical Determination of Parameters of the Air-Void System in
Hardened Concrete

0 Mercury Intrusion

 

6.2

83
MATERIALS AND MIX PROPORTIONS

The Materials used in this research are listed below:

Cement: regular Type I Portland cement (Table 6.2)

Coarse Aggregate: crushed limestone with a maximum aggregate size of 1/2 inch
(12.5mm) (Figure 6.1)

Fine Aggregate: regular concrete sand (Figure 6.2)

Conventional Coal Combustion By-Product: Class F Fly Ash, from Lansing Board of
Water and Light (Table 6.2)

Clean Coal Technology By-Product: Calcium Spray Dryer Fly Ash, from American
Power Association, North Dakota Plant (Table 6.2)

Air Entraining Agent: Darair, a type of Vinsol resin (AEAl), produced by W. R.

Grace & Company

84

Table 6.2 Chemical and Physical Properties of Cement Conventional Coal Fly Ash,
and Clean Coal Technology By-Product.

 

 

 

 

 

 

 

 

 

 

 

Class F Clean Coal
Cement Fly Ash Technology By-
Product

Silicon Dioxide (SiOZ) 20.58 49.3 28.73
Aluminum Oxide (A1203) 5.35 26.7 16.09
Ferric Oxide (Fe203) 2.56 8.95 4.32
Calcium Oxide (CaO) 62.44 1.83 29.63
Magnesium Oxide (MgO) 3.46 1.5 5.24
Sulfur Oxide (S03) 3.46 0.52 4.48
Alkalies as (NaZO) 0.77 3.8 2.5
Loss on Ignition 2.0 5.5 3.17
Fineness (% retained on 10.7 19.6 24.03
#325 sieve)

 

 

 

 

 

The mix designs were developed based on the absolute volume method.1 The total
cementitious material content was kept constant, and the target amount of air content was
achieved by adjusting the dosage of air-entraining agent. Many trial mixes were

investigated in order to develop the ﬁnal mix proportions.

The base mix considered in this project is presented in the Table 6.3.

Table 6.3 Mix Proportions.

 

 

 

 

 

 

Gradation Quantity Source
Coarse Aggregate 6AA 1800 lb crushed limestone
Fine Aggregate 2NS 1400 lb concrete sand
Cement Type I 560 lb Portland cement
Water 224 lb

 

 

 

 

 

85

Percent Passing (%)

 

 

 

eo /
40 I" “‘— /

 

 

20 /

 

 

o l
o- 0.2 0.4 0.6 0.8.-
Sieve Opening Size (in)

Figure 6.1 Gradation Curve of Coarse Aggregates.

86

0 Percent Passing (‘15)

100 ‘

 

 

 

 

 

:2..- /
j

o . . .
o 0.2 0.4 0.6 0.8

Sieve Opening Size (in)

 

 

 

Figure 6.2 Gradation Curve of Fine Aggregate.

6.3 TEST PROCEDURES

Most test procedures conducted in this research project follow the ASTM guidelines.

This section brieﬂy reviews these test procedures.

6.3.1 Resistance of Concrete to Rapid Freezing and Thawing (ASTM C666)

This test determines the concrete’s freezing and thawing resistance. The freeze-thaw test

specimens are 3 inch x 4 inch x 16 inch (7.62 x 10.16 x 40.6 cm). Concrete is kept inside

87
mold for 24 hours; after demolding, the specimens are stored in saturated lime water until

the time to start the test. Right before starting the test, the specimens are temporarily
stored inside a tempering tank in order to cool down the specimen temperature to prevent
initial reading errors. The tempering tank has a temperature within the range of : 2°F

(: 1.1°C) of the target thaw temperature for specimens in the actual freezing-and-thawing
cycles. After starting the freeze-thaw cycles measurements were done on specimens,

every 25 cycles in thaw condition.

The key measurements conducted here are:

1. Length Change (ASTM C 215), using a comparator capable of measuring to the
nearest 0.0001 in (0.0025 mm).

2. Weight Change, measured using a scale accurate to 0.0022 lb (1 g).

3. Fundamental Transverse Frequency (ASTM C 490).

Based on these measurement, the dynamic modulus of elasticity can be calculated as
follows:
P, =(n12/n2) x 100
where:
Pc = relative dynamic modulus of elasticity (percent), after c cycles of freezing and
thawing,
n = ﬁlndamental transverse frequency at 0 cycles of freezing and thawing

n, = ﬁrndamental transverse frequency after c cycles of freezing and thawing

88
All the specimens are subjected to 300 cycles for freezing and thawing unless they fail

earlier. Each cycle is approximately 6 hours.

6.3.2 Scaling Resistance of Concrete Surface Exposed to Deicing Chemicals
(ASTM C 672)

This test determines the resistance to scaling of a horizontal concrete surface exposed to
freezing-and-thawing cycles in the presence of deicing chemical. The surface resistance
is evaluated qualitatively by visual examination. The specimen size used in this test was

8 inch x 10 inch x 3 inch (20.3 x 25.4 x 7.6 cm).

All specimens are stored in curing room until the testing age; they are then placed in a
freezing environment for 16 hours. At the end of this time the specimens are removed
from freezer and placed in laboratory air at 73°F (23°C) and a relative humidity of 45 to
55% for 8 hours. These cycles are repeated typically 50 times, ﬂushing off the surface

every 5 cycles. Visual observation of the surface is made after completion of all cycles.

Visual rating of the surface in accordance with the following scale:

R |° C l' I' [S E
0 no scaling
1 very slight scaling (max. 1/8 in., 3.2 m, depth, no coarse

aggregate visible)

89

2 slight to moderate scaling

3 moderate scaling (some coarse aggregate visible)

4 moderate to severe scaling

5 severe scaling (coarse aggregate visible over entire surface)

6.3.3 Microscopical Determination of Parameters of the Air-Void System in
Hardened Concrete (ASTM C 457)

The test conducted here is ASTM procedure A, the linear-traverse method. This test
method uses a microsc0pical device to determinate the air content of hardened concrete,
and the speciﬁc surface, void frequency, spacing factor, and paste-air ratio of the air-void
system in hardened concrete. The 1” x 3” x 4” (2.5 x 7.6 x 10.1 mm) specimens used in
this test A were obtained by saw cutting one of the freeze-thaw specimen. The rough
surface of the saw cut specimen was subjected to a series of lapping procedures in order
to obtain satisfactory visual microscopical observations. The lapping procedure involved
the use of a commercially available lapping machine. All the specimens were subjected
to the following sequence: (1) apply No. 1560 grit (with the maximum size of 75 run) for
12 minutes; (2) wash the polishing surface; (3) apply No. 2320 grit (with the maximum
size of 35 pm) for 12 minutes; (4) wash the polishing surface; (5) apply No. 2800 grit

(with the maximum size of 14.5 um ) for 12 minutes; and (6) wash and dry the surface.

90

Because the maximum size of coarse aggregate is 3/4 inch (19 mm), the total traveling

distance is 90 inches (2286 mm).

The microscope used here had a magniﬁcation of 75x. The testing procedure, under the
microscopic, involved counting all the aggregates and voids passing through the reference
point located in one of the eyepieces. The testing apparatus will automatically store all

the observed data and calculate the desirable results.

6.3.4 Electrical Indication of Concrete’s Ability to Resist Chloride Ion Penetration
(ASTM C 1202)

This test monitors the amount of electric current passed through a 4 in. (10.1 mm)
diameter cylinder with a thickness of 2 inches (50.8 mm) for a period of 6 hours. A
potential difference of 60 V dc is maintained across the ends of the specimen. One end of
the specimen is immersed in a sodium chloride solution, and the other in a sodium
hydroxide solution. The total charge passed, in Coulombs, has been found to be related

to the resistance of the specimen to chloride ion penetration.

All the casting specimens will be allowed to surface dried in air for at least 1 hour. A
surface coating epoxy is then brushed onto the side surface of the specimen. After the
epoxy coating cures, the specimen is placed into a vacuum chamber; the pressure is

maintained at less than 0.04 in (1 mm) Hg for 3 hours. The vacuum chamber is then

91

ﬁlled with tap water and vacuum is continued for another hour. The Specimen is
subsequently soaked without vacuum for 18 hours, after which the specimen is placed in
the testing cell. One side of the cell is ﬁlled with 3% NaCl solution, and the other side
with 0.3 N NaOH solution. The coulombs electricity passing through concrete cell are
measured every 30 minutes for a total period of 6 hours. The ﬁnal result will be
compared with the following ASTM C 1202 table in order to assess the chloride

permeability level.

Table 6.4 Chloride Ion Permeability Based on Charge Passed

 

 

 

 

 

 

Charge Passed (coulombs) Chloride Ion Penetrability
> 4,000 High
2,000 - 4,000 Moderate
1,000 - 2,000 Low
100 - 1,000 Very low
< 100 Negligible

 

 

 

 

6.3.5 Compressive Strength of Cylindrical Concrete Specimen (ASTM C79)

All the specimens used in this test were 6 inches (152 mm) in diameter and 12 inches
(305 mm) high. After completion of the moist curing period, the specimen are air dried
for approximately one hour before capping. After capping, the specimens are placed in

the hydraulic compressive machine and subjected to a compressive force at a rate around

92
35 psi/sec (0.24 MPa/sec). The maximum load divided by the area becomes the ﬁnal

result.

6.3.6 Bleeding of Concrete (ASTM C 232)

This test determines the relative quantity of mixing water that will bleed from a freshly
mixed concrete sample. The bleeding test conducted here is the ASTM C 232 Method A
- Sample Consolidated by Tamping. In this test a half cubic foot volume container is
ﬁlled with ﬁ'eshly mixed concrete. The cover is kept in place through out the test, except
when drawing the water. The water that has accumulated on the surface is drawn off with
pipet at 10-minute intervals during the ﬁrst 40 min. and at 30-minute intervals thereafter
until cessation of bleeding. To facilitate the collection of bleeding water, the specimen is
tilted careﬁrlly by placing a 2-in (SO-mm) block under the container 2 minutes prior to
each time the water is withdrawn. All the collected water is accumulated to a 100-ml
gradual. The volume of bleeding water per unit area of surface, V, is then calculated as
follows:

V = Vl/A
where:
Vl = volumer of bleeding water measured during the selected time interval, ml, and

2
A = area of exposed concrete, cm

93
6.3.7 X-Ray Diffraction

The X-Ray diffraction is a commonly used method to detemline the mineralogical
composition of material. In this study, a powder sample (ground from the hardened
concrete sample) was struck by an X-ray beam. The regular crystalline nature of many of
the phases in the sample causes diffraction of the incoming X-rays at particular angles.
The intensity and angles of these diffraction effects produce a pattern of peaks (the
diffraction pattern or diffractogram) representing what phases are present in the residual.
Each crystalline compound has its own peak positions and intensities, and these
characteristics (X-ray powder patterns) can be used to assess the mineralogical

characteristics of material.

6.3.8 Concrete Surface Finishing

In order to understand the effects of ﬁnishing techniques on the surface scaling of
concrete, additional specimens were prepared and subjected to different ﬁnishing
procedures. The ﬁnishing procedures consisted of: (1) placing the fresh mix into
specimen mold; (2) screeding or striking off the top surface; (3) bullﬂoating or darbying
to eliminate high and low spots and to embed large aggregate particles. Aﬁer bleeding,
the extra specimens were ﬁnished with a steel bullﬂoating. The purpose of the ﬁnal

ﬁnishing is to seal the surface and provide a smooth ﬁnish.

94
6.3.9 Mercury Intrusion

In the mercury intrusion porosimetry test technique, mercury is forced under pressure into
dried concrete specimens. The pressure required to force mercury into the voids is related
to the size of the entranceway of the pores. Using this relationship, by recording the
pressure and the volume of mercury intruded, one may obtain an estimation of the pore
size distribution. The tests conducted here used a contact angle of 117 degree, and the

pressure could reach as high as 30,000 psi (206 MPa).

CHAPTER 7

EXPERIMENTAL RESULTS AND ANALYSES:
ENGINEERING PROPERTIES

7.1. INTRODUCTION

The extensive nature of the experimental data produced in this research makes it
necessary to analyze subgroups of test results in order to reach practical conclusions. The
sequential analysis we have conducted concerns: (1) the performance of clean coal
technology by-product versus conventional coal ﬂy ash; (2) effects of the clean coal by-
product; and (3) effects of conventional coal ﬂy ash. This chapter focuses only on the
engineering properties of concrete. Microstructural analyses and structure-property

relationships will be discussed in the next chapter.

7.2. PERFORMANCE OF THE CLEAN COAL TECHNOLOGY BY-
PRODUCT VERSUS CONVENTIONAL COAL FLY ASH

The subcomponent of experimental program analyzed here in order to compare the
performance of the clean coal technology by-product and the conventional ﬂy ash is

shown in Table 7.1.

95

96
Table 7.1 Subcomponent of the Experimental Program.

 

 

 

 

 

 

 

 

Variables Experimental Program

W/C 0.4 0.45

Ash Content 15% 30% 15% 30%

Air Content 2.5% 7.5% 2.5% 7.5% 2.5% 7.5% 2.5% 7.5%
Curing Period 28 56 28 56 28 56 28 56 28 56 28 56 28 56 28 56
Clean C031 e e t e e e e e e e e e e e e 1:
Conventional # t t t t t It i Q t t i t i * *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The engineering properties considered here are compressive strength, rapid chloride
permeability, deicing salt scaling resistance and freeze-thaw durability. Test results are

presented in Figures 7.1, 7.2, 7.3 and 7.4.

 

97

Compressive Strength (MPa)

 

 

 

 

 

 

 

[3 clean coal - 28 “7’7 clean coal - 56
fly ash - 2e :1 fly ash - 56

(a) Water-Cement Ratio = 0.4

V Compressive Strength (MPa)

 

 

 

 

 

 

 

 

 

 

80
195% Conﬁdence Intervals
60 -
40 r
20 '-
('1
80%-2.5%
Replacement - Air
[3 clean coal - 28 clean coal - 56
222529 tly ash - 28 [:1 fly ash - 56

(b) Water-Cement Ratio = 0.45.

Figure 7.1 Compressive Strength Test Results (Means & 95% conﬁdence Intervals).

98

6 Permeability (Thousand Coulombsl
[95% Conﬁdence Intervals

I2 ........ I'

15%-7.5% 30%-2.5%

 

Replacement — Air
[:1 clean coal - 28 clean coal - 56
fly ash - 28 I: 11y ash - 56

(a) Water-Cement Ratio = 0.4

.4 Permeability (Thousand Coulombs)
I 7 + [95% Conﬁdence Intervals

 

 

 

 

7
6 .-
5
4
2
1
o ' '
15%-2 % 15%-7.5% 30%-2.5% 30%-7.5%
Replacement — Air
C: clean coal - 28 "“7""? clean coal - 56
My ash-28 1:] 11y ash-56

(b) Water-Cement Ratio = 0.45

Figure 7.2 Rapid Chloride Permeability Test Results (Means & 95% conﬁdence
Intervals).

Scaling Rate
1':

 

1 95% Conﬁdence Intervals

 

15%-7.5% 30757596
Replaceme”t ’ Air
/ clean coaI ' 56
I: 11y at;h ' 56

I: clean coal — 28
Hy ash - 28

(a) Water-Cement Ratio = 0.4

Scaling Rate

y
r
z
/

 

 

30%-7.5%

 

1 95% Conﬁdence Intervals

 

 

23:22:. /; "is 2 A5 3
15952595 15%—7.5% 30%-2.5
Replacement — Air

   

7‘77 clean coal - 56
CI fly ash - 56

I: clean coal — 28
ﬂy ash - 28

(b) Water-Cement Ratio = 0.45

Figure 7.3
Intervals).

   

30%-7.5%

  

 

Deicer Salt Scaling Resistance Test Results (Means & 95% conﬁdence

100

Durability Factor (%)

 

          

 

 

 

 

 

 

 

 

 

 

120
t 95% Conﬁdence Intervals
4O _ ., . ..
37 252255:
O‘ '///4 l .7 5355355
15%-2.5% 15%-7.5%
Replacement - Air
1:] clean coal - 28 77 clean coal - 56
11y ash - 28 i: ﬂy ash - 56 ,

(a) Water-Cement Ratio = 0.4

Durability Factor (%)

 

1 20
195% Confidence Intervals
100 - ~ A V l A r “I V M 72-3323:-

eeeeeee
eeeeeee

60 *- ' '7 /5553255

eeeeeee

eeeeeee
eeeeeee
eeeeeee
eeeeeee
eeeeeee
eeeeeee
.......
eeeeeee

eeeeeee

eeeeeee

eeeeeee

eeeeeee
eeeeeee
eeeeeee
eeeeeee
eeeeeee

  

  

 

 

 

 

 

 

/ e

 

 

 

 

15%-7.5% 30%—2.5% 30%-7.595
Replacement - Air

D clean coal - 28 7/7/ clean coal - 56
fly ash - 28 Cl tly ash - 56

(b) Water-Cement Ratio = 0.45

Figure 7.4 F reeze-Thaw Resistance Test Results (Means & 95% Conﬁdence
Intervals).

The results of factorial analysis of variance are summarized in Table 7.2.

101

 

 

 

 

 

 

 

 

 

 

 

 

Table 7.2 Results of the Factorial Analysis of Variance.
Variables Compressive Permeability Scaling Freeze-Thaw
Strength Resistance Resistance
Ash Type * * - I -
W/C *** I I -
Ash Content * * * I I -
AirContent *** I *** ***
Curing Period *** I I -
I : Significant Interaction
* : Inﬂuential at 90% confidence level
** - : Inﬂuential at 95% conﬁdence level
"* : Inﬂuential at 99% conﬁdence level

7.2.1 Compressive Strength

The factorial analysis of variance of the compressive strength test results indicates that

there is a signiﬁcant difference between the clean coal technology by-product and the

conventional Class F ﬂy ash at 95 percent conﬁdence level. On the average, concrete

mixtures with conventional coal ﬂy ash provided 5% more compressive strength than

mixtures with the clean coal by-product. This slight drop could partly result from the

improved workability of fresh mix in the presence of coal ﬂy ash. Increasing the ﬂy ash

level ﬁ'om 15% to 30% caused an average drop of 7% in compressive strength while the

same increase in the content of the clean coal by-product led to a drop of 15% in

 

102

compressive strength. Overall, the performance of clean coal by-product as a mineral

additive in concrete was satisfactory from the strength point of view.

7.2.2 Rapid Chloride Permeability

On the average, specimens with the clean coal by-product showed a 9% more chloride
permeability than those with conventional ﬂy ash. This increase, however, was not
statistically signiﬁcant. In the case with lower (15%) ash contented high (7.5%) air
content, chloride permeability actually dropped by 11% in mixtures with the clean coal
by-product when compared those with conventional coal ﬂy ash. Increasing the ﬂy ash
and clean coal by-product contents from 15% to 30% led to drops of 50% and 33% in
permeability, respectively. Increasing the moist curing period from 28 to 56 days caused
drops of 65% and 54% in the permeability of concrete mixtures with conventional coal

ash and the clean coal by-product, respectively.

7.2.3 Scaling Resistance

The analysis of variance of scaling test results indicates that there are interactions

between most variables; only air content was independent and had statistically signiﬁcant

effects on scaling resistance. Overall, the scaling rate of concrete with conventional ﬂy

103
ash was 5% worse than that with the clean coal by-product. Increased ash content

actually improved the scaling resistance with the clean coal by-product while it had
adverse effects with conventional Class F ﬂy ash (Figure 7.5). Prolonged moist curing

improved the scaling resistance of concrete specimens with both conventional ﬂy ash and

the clean coal by-product (Figure 7.6).

Scaling Rate

 

 

 

 

 

 

 

 

 

   

Ash Content

9,9531. Fly ASh

 

1:] Clean Coal

Figure 7.5 Ash Content Effects the Scaling Resistance.

104

Scaling Rate

 

 

 

 

 

 

 

 

 

 

1: Clean Coal Fly Ash

Figure 7.6 Prolonged Curing Effects on the Scaling Resistance.

7.2.4 Freeze-Thaw Resistance

The analysis of results show that the hardened concrete air content is the only factor
effects the hardened concrete freeze-thaw resistance. Irrespective of different mix

proportioning speciﬁcs, sufﬁcient air content guaranteed that the concrete sustains

repeated freezing and thawing.

105
7.3 Effects of 15% Class F Fly Ash

The overall analysis in this section focuses on the performance of plain concrete vs. that

with 15% replacement of cement with Class F ﬂy ash. Figure 7.7 shows the relevant test

results, and Table 7.3 presents the results of the analysis of variance of the test results.

Table 7.3 Results of the Factorial Analysis of Variance.

 

 

 

 

 

 

 

 

 

 

 

Variables Compressive Permeability Scaling F reeze-Thaw
Strength Resistance Resistance
W /C 4! at: It! It: * at: _
Ash Content - - - -
Air Content *** - - ***
Curing PeriOd *IIHII #1101: *4"? _
I : Signiﬁcant Interaction
" : Inﬂuential at 90% conﬁdence level
** : Inﬂuential at 95% conﬁdence level

*" : Inﬂuential at 99% conﬁdence level

 

 

80 Compressive Strength (MPa)

195% Conﬁdence Intervals

 

 

 

0.4-2.5% 0.4-7.5% 0.45-2.5% 0.45-7.5%
Binder Ratio - Air
[erry Ash - 28 “Fly Ash - 56 -Plain - 28 1:1P1aln - 56

(a) Compressive Strength

Permeability (Th0usand Coulombs)
195% Conﬁdence Intervals

 

MQAGGVO)

 

 

 

 

0.4-2.5% 0.4-7.5% 0.45-2.5% 0.45-7.5%
Binder Ratio — Air
13H): Ash - 2a Wm, Ash — 56 3555553 Plain - 28 DPiain - 56

(b) Rapid Chloride Permeability

107

 

Scaling Rate
5

1 95% Conﬁdence Intervals

 

7/ .......

   

  

  

O {/77 n ....... A , ,1 ........ /,,...-/ ....
0.4-2.5% 0.4-7.5% 0.45-2.5% 0.45-7.5%
Binder Ratio — Air
EFly Ash - 2a 4/ Fly Ash - 56 -Plain - 28 [:1Plalh - 56

(c) Deicer Salt Scaling Resistance

Durability Factor (96)
120

 

195% Conﬁdence Intervals
100 r' . '

  

530-»
so—

40—

 

 

 

0.4-2.5%

    

0.45-7.5%

0.4-7.5% 0.45-2.5%
Binder Ratio - Air
[:lFly Ash - 28 Fly Ash - 56 -Plain - 28 1:1Piain - 56

(d) Freeze-Thaw Resistance

Figure 7.7 Test Results (Mean & 95% conﬁdence Interval).

 

 

108
7.3.1 Compressive Strength

The overall analysis shows there is no statistically signiﬁcant difference between the class
F ﬂy ash concrete and plain concrete in compressive strength. Both plain and class F ﬂy
ash concrete have the same trends of dropping 12% compressive strength in response to
the increase in water to cement ratio. With increased dosage of entrained air from 2.5%
to 7.5%, plain concrete has a tendency towards a strength loss of about 15%. Under the
same condition, the 15% class F ﬂy ash concrete would drop 22% of the strength. This
indicates that the class F ﬂy ash is more sensitive to water content. Under the prolonged
curing period, both class F ﬂy ash and plain concrete show an increase in strength of 14%

and 20%, respectively.

7.3.2 Rapid Chloride Permeability

Overall, there is a signiﬁcant difference between the class F ﬂy ash and plain concrete in
the performance of rapid chloride permeability. The plain concrete shows an average of
24% higher permeability value than the class F ﬂy ash concrete. Under the condition of
increased water content, ﬂy ash concrete has a 32% increase in permeability value and the
plain concrete shows a 55% increase. Figure 7.8 presents the results of changing water to

binder ratio effects on the permeability.

109

Permeability (Coulombs)

 

6000

l

5000

4000b

1

8000

I

2000

1

1000

 

 

 

 

 

 

 

Figure 7.8 Water to Cement Ratio Effects on the Permeability.

Air content is another important factor which affects permeability. Overall, there is only
6% increase in plain concrete permeability when increasing the air content from 2.5% to
7.5%, but 28% increase in the case of ﬂy ash concrete. Figure 7.9 presents the air content
effects on permeability. It can be noted that the ﬂy ash is very effective in reducing
permeability. Irrespective of the mix pr0portions, ﬂy ash signiﬁcantly reduces the
chloride permeability of concrete. Class F Fly ash, a round and ﬁne particle, can ﬁt well

between all ingredients during the mixing and thus block the continuity of capillary pores

110
structure. The pozzolanic action of ﬂy ash further ﬁlls and block capillary pores with the

pozzolanic reaction products.

Permeability (Coulombs)

 

    
     
   

 

 

 

 

 

 

 

 

5000
4000 — WM // .
%/ g
7 ,
3000 - 7 ¢ :
2000 ’- 7 g:
1000 *-
0
Air Content
[:1 Fly Ash Plain

 

Figure 7.9 Air Content Effects on the Permeability.

One of the most important characteristics of ﬂy ash concrete is the proper curing to

ensure the pozzolanic reaction which densiﬁes the microstructure of paste. Usually, this

lll

reaction takes place over time in the presence of moisture. In order to take advantage of
this phenomenon, a prolonged moist curing period is essential. 0f with Class F ﬂy ash,
permeability dropped by 70%, and by 487. For plain concrete when the curing period

was increased from 28 days to 56 days.

7.3.3 Deicer Salt Scaling Resistance

The Class F ﬂy ash concrete with 15% ﬂy ash performed better than the plain concrete
from the scaling resistance point of view. This result may be illustrated by reasons such
as: (1) 15% replacement of cement by ﬂy ash would not signiﬁcantly affect the early age
strength of the paste; (2) the use of Class F ﬂy ash signiﬁcantly reduces the permeability
of hardened concrete; and (3) the pozzolanic reaction ﬁlrther densiﬁes the microstructure
and helps concrete relieve the adverse effects of deicer salt under repeated freezing and

thawing.

Both ﬂy ash and plain concrete were sensitive to changes in water to binder ratio.
Increased air content did not change the scaling resistance of plain concrete, but improved
that of ﬂy ash content from 2.9 to 2.2. Under prolonged moist curing ﬂy ash concrete
showed a dramatic improvement in scaling resistance from 3.2 to 1.9, while plain

concrete showed a smaller improvement from 3.1 to 2.8.

7.3.4 Freeze-Thaw Resistance

112

The analysis of results show that the hardened concrete air content is the only factor

affecting freeze-thaw resistance. Irrespective of different mix proportions, sufﬁcient air

content will guarantee that the concrete sustains repeated freezing and thawing.

7.4 EFFECTS OF 30% CLASS F FLY ASH

The overall analysis in this section focuses on the performance of plain concrete versus

that with 30% replacement of cement with Class F ﬂy ash. Table 7.4 presents the results

of the analysis of variance of the test results, and Figure 7.10 shows the relevant test

 

 

 

 

 

 

 

 

 

 

 

results.
Table 7.4 Results of the Factorial Analysis of Variance.
Variables Compressive Permeability Scaling Freeze-Thaw
Strength Resistance Resistance
W /C 31‘ II! '1‘ It! _
Ash Content *** *** 1 -
Air Content *** - ** ***
Curing Period ** * * * * 1 -
l : Signiﬁcant Interaction
* : Inﬂuential at 90% conﬁdence level
" : Inﬂuential at 95% conﬁdence level

*** : Inﬂuential at 99% conﬁdence level

 

010)

A

NCO

113

Compressive Strength (MPa)
80

 

195% Conﬁdence Intervals
60 r

40”

20—

 

 

0.4—2.5% 0.4—7.5% 0.45-2.5% 0.45-7.5%
Binder Ratio - Air
L—JFly Ash '- 28 Fly Ash - 56 -Plain — 28 1:1Plain - 56

(a) Compressive Strength

Permeability (Thousand Coulombs)
195% Conﬁdence Intervals

 

    

 

 

0.4-2.5% 0.4—7.5% 0.45-2.5% 0.45-7.5
Binder Ratio — Air
:lFly Ash - 28 Fly Ash - 56 -Piain - 28 [:lPlaln — 56

(b) Rapid Chloride Permeability

 

ll4

Scaling Rate
5

 

l 95% Conﬁdence Intervals

   
   

 

 

04-25% 0.4-7.5% 0.45—2.5% 0.45—7.5%
Binder Ratio — Air

2 Fly Ash - 55 -Plaln — 28 Clplaln - 55

 

i:lFly Ash - 28

(c) Deicing Salt Scaling Resistance

0 Durability FactOr (%)

 

i 95% Conﬁdence Intervals

80*

60-

20"

 

 

0 .
0.4—2.5%

 

Binder Ratio - Air
:Fly Ash — 28 -Fly Ash - 56 -Plaln - 28 :lPlaln — 56

(d) Freeze-Thaw Resistance

Figure 7 .10 Test Results (Mean & 95% conﬁdence Intervals).

115
7.4.1 Compressive Strength

Overall, the plain concrete provide a higher compressive strength than class F ﬂy ash.
Under the condition of increased air content, Class F ﬂy ash showed a decrease of 10% in
strength while the corresponding dr0p in the case of plain concrete was 15%. With
increasing water to binder ratio, ﬂy ash and plain concretes showed drops of 5 and 10%,
respectively. Prolonged curing signiﬁcantly increased the ﬂy ash and plain concrete

strengths by about 20%.

7.4.2 Rapid Chloride Permeability

Overall, the analysis shows that there is a signiﬁcant difference between the permeability
of ﬂy ash and plain concrete. Figure 7.11 compares the rapid chloride permeability of ﬂy
ash and plain concrete. It can be seen that the plain concrete pOssess permeability values
as high as 140% more than the ﬂy ash concrete. Under the condition of prolonged curing,
both ﬂy ash and plain concrete show a drop of about 50% in the initial permeability

value.

ll6

Permeability (Coulombs)

 

5000

 

4000 '

3000 '

2000 —

 

1000 ‘

 

 

 

 

 

 

 

1
Fly Ash Plain

Figure 7.11 Fly Ash Effects on Concrete Permeability.

7.4.3 Deicer Salt Scaling Resistance

In general, there is a signiﬁcant interaction between the use of ﬂy ash and its curing
period as far as the deicer salt scaling resistance of concrete is concerned. The test results
are shown in Figure 7.12. Under prolonged curing, ﬂy ash concrete tends to improve its
scaling resistance by 26%, while plain concrete shows an improvement of only 7%. It
should be noticed that without a proper curing period of 56 days, ﬂy ash would perform

16% worse than the plain concrete from the scaling resistance point of view.

ll7

Scaling Rate

 

 

 

 

 

 

 

 

EFIy Ash

 

. 757,? f//f -
" --, Plain

Figure 7.12 F ly Ash Interacts with Prolong Curing Period.

7.4.4 Freeze-Thaw Resistance

The analysis of results shows that the hardened concrete air content is the dominant factor
in regard to freeze-thaw durability. Other factors, including ﬂy ash content, are not

critical in this regard as far as sufﬁcient air content is entrained into concrete.

118

7.5 EFFECTS OF 15% CLEAN COAL TECHNOLOGY BY-PRODUCT

The overall analysis in this section focuses on the performance of plain concrete versus

that with 30% replacement of cement with the clean coal technology by-product. Table

7.5 presents the results of the analysis of variance of the test data, and Figure 7.13 shows

the relevant test results.

 

 

 

 

 

 

 

 

 

 

 

Table 7.5 Results of the Factorial Analysis of Variance.
Variables Compressive Permeability Scaling Freeze-Thaw
Strength Resistance Resistance
W/C * III III _ _
Ash Content *** - -
Air Content * - - ***
Curing Period ** *** ** -
I : Signiﬁcant Interaction
* : Inﬂuential at 90% conﬁdence level
" : Inﬂuential at 95% conﬁdence level

*** : Inﬂuential at 99% conﬁdence level

 

119

Compressive Strength (MPa)

 

I 95% Conﬁdence Intervals

 

  

 

l l/r/ I

C‘ 4-2.5% 0.4-7.5% 0.45-2.5% 0.45—7.5%
Binder Ratio - Air
|:lclean coal - 28 -clean coal - 56 -Plain - 28 [:1Plein - 56

 

(a) Compressive Strength

Permeability (Thousand Coulombs)
Ci

:, 195% Conﬁdence Intervals .

 

 

  

1‘.) 1’1 J—~ (1R (31

.1

 

 

1‘1 4-2.5% 0.4—7.5% 0.45-2.5% 0.45-7.5%
Binder Ratio — Air
:jclean coal - 28 -clean coal — 56 -Plain - 28 :lPlaln - 56

(b) Rapid Chloride Permeability

120

Scaling Rate

 

I 95% Conﬁdence Intervals

4 F'-

    

 

 

 

 

 

O.45'2.5% 0.45-7.5%
Binder Ratio - Air
[:lclean coal - 28 -c|ean coal - 56 -P|ain - 28 BPlain - 56

(c) Deicer Salt Scaling Resistance

Durability Factor (%)

 

 

 

 

l 95% Conﬁdence Intervals
‘lOO '
80
6O
4O
20
O /
0.45—2.5%
Binder Ratio 5 Air
1: clean coal - 28 /7/ clean coal - 56

 

Plain — 28 C] Plain - 55

(d) F reeze-Thaw Resistance

Figure 7 .13 Test Results (Mean & 95% conﬁdence Interval).

121
7.5- l Compressive Strength

Overall, there was no signiﬁcant effect of the clean coal technology by-product on
compressive strength. Prolonged moist curing period would increase the compressive
Strength of concretes with clean coal technology by-product and plain concrete by 14%

atld 1 9%, respectively.

7-5 -2 Rapid Chloride Permeability

Sta'atistical analysis indicates that there is a signiﬁcant effect of the clean coal technology

bY-product on permeability. There is an average drop of 20% in permeability with the
CJean coal technology by-product. Prolonged moist curing beneﬁted both clean coal
technology by-product and plain concretes by reducing their permeability by 56% and

<18%, respectively (Figure 7.14).

122

De r meability (Thousand Coulombs)

 

 

 

 

 

 

 

 

I:igure 7.14 Effects of Ash Content Versus Prolonged Curing.

7 ~ 5.3 Deicer Salt Scaling Resistance

There is no statistically signiﬁcant effects of the clean coal technology by-product on the
Scaling resistance of concrete. Prolonged moist curing would increase the scaling
resistance of the clean coal technology by-product concrete by 17% and that of the plain

concrete by 11%.

123
7.5-4 Freeze-Thaw Resistance

The analysis of results shows that the hardened concrete air content is the only key factor

in determining the freeze-thaw durability of concrete.

7-6 EFFECTS OF 30% CLEAN COAL TECHNOLOGY BY-PRODUCT

The overall analysis in this section focuses on the performance of plain concrete versus
that with 30% replacement of cement with Class F ﬂy ash. Table 7.6 presents the results

of the analysis of variance of the test data, and Figure 7.10 shows the relevant test results.

 

 

 

 

 

 

 

 

 

 

 

Table 7.6 Results of the Factorial Analysis of Variance.

Variables Compressive Permeability Scaling Freeze-Thaw

Strength Resistance Resistance

W/C =1: :1: a1: _
Ash Content **"' ** * -
Air Content - - * ***
Curing Period *** *** ** -
I : Signiﬁcant Interaction
"‘ : Inﬂuential at 90% conﬁdence level
"‘ * : Inﬂuential at 95% conﬁdence level
* t *

: Inﬂuential at 99% conﬁdence level

124

Compressive Strength (MP8)

I“.

 

‘4

0

l 95% Conﬁdence Intervals

 

 

 

 

 

 

  

 

0.4-7.5% 0.45-2.5% 0.45-7.5%
Binder Ratio - Air
:1 clean coal - 28 -clean coal - 56 -Plain - 28 EPlain - 56
(a) Compressive Strength
Permeability (Thousand 00u|ombsi
F 195% Conﬁdence Intervals
r ............
i
i
F
1 H], "m
l' 4-2.5% 0.4—7.5% 0.45-2.5% 0 115-7.5%
Binder Ratio - Air
Eclean coal - 28 clean coal - 56 -Plain — 28 l:lPlain - 56

 

(b) Rapid Chloride Permeability

 

SC
5

125

aling Rate

 

 

   
   

1 95% Conﬁdence Intervals

0.4-2.5% 0.45—2.5% 0.45-7.5%
Binder Ratio - Air

l:iclean coal — 28 -ciean coal - 55 -Plain - 28 DPlain - 56

(c) Deicer Salt Scaling Resistance

Durability Factor (%)

 

100
80
60
40

20

 

l:l clean coal - 28

 

195% Conﬁdence Intervals

 

0.45-2.5% 0.45-7.5%
Binder Ratio - Air

 

I clean coal - 56

Plain - 28 :3 Plain — 56

Figure 7.15

(d) Freeze-Thaw Resistance

Test Results (Mean & 95% conﬁdence Interval).

 

126
7.6.1 Compressive Strength

Overall, the replacement of 30% of cement by the clean coal technology by-product has a
negative effect on compressive strength; there is an average drop of 15% in compressive
strength. Although prolonged moist curing would substantially increase the strength of
both the clean coal technology by-product concrete, it would still be 16% lower than that

of plain concrete after 56 days of moist curing (Figure 7.16)

OCompressive Strength (MPa)

 

40*

 

10‘-

 

 

 

 

 

 

 

Figure 7.16 Effects of the Clean Coal Technology By-Product Versus Moist Curing
Period.

127
7.6.2 Rapid Chloride Permeability

There is a statistically signiﬁcant difference between the permeability of clean coal
technology and plain concretes. The results indicates an average 50% drop in chloride
permeability when using the clean coal technology by-product in concrete. Extended
moist curing period reduces permeability by as much as 50% (Figure 7.17). In Figure 17,
after prolonged curing, the clean coal technology by-product concrete possessed 50% less

permeability when compared with plain concrete.

Permeability (Coulombs)

 

5000

4000 F ~

I

3000

I

2000

1000

 

 

 

 

 

 

 

 

272 Plain

1:: Clean Coal

Figure 7.17 Effects of the Clean Coal Technology By-Product Versus Moist Curing
Period.

128
7.6.3 Deicer Salt Scaling Resistance

In general, there is no statistically signiﬁcant difference between the scaling resistance of
the clean coal technology by-product and plain concretes. Prolonged moist curing
enhanced the scaling resistance of the clean coal technology by-product and plain

concretes by 18% and 8%, respectively.

7.6.4 Freeze-Thaw Resistance

The analysis of results shows that the hardened concrete air content is the only factor

affecting its freeze-thaw durability.

CHAPTER 8

STRUCTURE-PROPERTY RELATIONSHIPS

8.1 INTRODUCTION

The properties of hardened concrete are mostly determined by its pore structure -
typically the cement paste pore structure. There are many methods that can be used to
measure the pore structure of hardened concrete. The mercury intrusion porosimetry has

been the most popular and useful one.

The technique for mercury intrusion process involves forcing the mercury to penetrate
into the specimen from the surface toward the inner area. During the process of intrusion,
with increasing intrusion pressure, entry into the smaller pore systems can be obtained.
By observing the pressure and volume of the intruded mercury, the pore entry size and
the volume of different pore systems can be evaluated. Hence, the typical pore size
distribution obtained by intrusion can be seen as a distribution of pore volume vs. pore

entry sizes.

This pore size distribution and pore volume can be also correlated with the compressive
strength and permeability. The strength of hardened concrete has a direct relationship
with the pore structure. Over time, inside the hardened concrete, the unhydrated

cementitious material hydrates continuously (as far as sufﬁcient moisture is present).

129

130

This continuous reaction densiﬁes the concrete structure, blocks the continuous capillary

pores and improves the strength of concrete.

Rapid chloride permeability of hardened concrete measures the diffusion of electrical
irons through the pore structure. Hence, the entry size of pores controls the quantity of

ions passing through the capillary pores during the permeability test.

8.2 EXPERIMENTAL WORK

This experimental program considered plain, ﬂy ash, and clean coal technology by-
product concretes. Initial analyses are based on the general characteristics of pore
structure in order to compare different cementitious materials. Subsequently, regression
analyses are conducted in order to develop structure-property relationships in the form of
the relationships between pore size characteristics and strength of concrete materials

(irrespective of the binder composition).

8.3 PORE SIZE DISTRIBUTION

All mixes discussed in this section had a water-binder ratio of 0.4, aggregate-binder ratio

of 5, and air content of 2.5%. In the case of ﬂy ash and clean coal technology by-product

131

concretes, the replacement level was 30%. Figure 8.1 presents the pore size distribution
of Portland cement concrete at different ages. The pore system size is rather uniformly
distributed at the two ages considered. Over time, pores of all size were ahnost equally
reduced in intrusion volume as a result of continued reactions within the binder. At both

ages, a peak volume of pores was observed at a pore size of approximately 0.1 micron.

 

 

 

Intrusion Volume (1: 0.001 cclg)

 

 

MNOQPGIOOia-NIONNVN
thuomthi‘Ogva
"ommmv- vole-Pogo;

0 0897

0 0549

0.0336 ..

0 0205

0 0125
0.00767 I

 

Pore Diameter (micro) l
‘ +28-days +56-day

 

 

Figure 8.1 Pore Size Distribution of Portland Cement Concrete.

Figure 8.2 presents the pore size distribution of ﬂy ash concrete. The Class F coal ﬂy ash
particles used in this study are in the range of 5 to 45 micron in size. The presence of ﬂy

ash reduced the uniformity of pore size distribution. While an increase was observed in

Intrusion Volume (11 0.001 0019)

132

the volume of smaller pores (<0.1 micron), the volume of larger pores was reduced. At
later ages, one could see a drop in pore volume resulting from continued pozzolanic
reaction. Fly ash seems to be effective in blocking the continuos capillary pore system,
which should reduce the permeability of concrete. It seems that ﬂy ash concrete beneﬁts

more than conventional concrete ﬁom prolonged curing.

 

30..

88383

 

 

 

 

 

Pore Diameter l l

Figure 8.2 Pore Size Distribution of Fly Ash Concrete.

133
Figure 8.3 shows the pore size distribution of the clean coal technology by-product

concrete. It can be seen that the clean coal technology by-product concrete has a
relatively high volume of larger pores (~ 100 micron diameter). But this group of pores
was almost diminished after prolonged curing. The overall pore size distribution was not
uniform but somewhat concentrated around three areas. Although this non-uniformity
may reﬂect the blocking effect which yields reduced permeability, the concentration of
pore groups may damage the strength of concrete. Over time, the pore groups between 4
to 0.4 micron were not changed at all. This is an indication that the clean coal technology
by-product is lacking a certain group of hydration systems when compared with the
conventional Class F ﬂy ash. This may be explained by the factor that the clean coal
technology by-product contains, in addition to ﬂy ash, unreacted sorbent and some
reacted sorbent products. Some of these reacted products would no longer react with
moisture to form hydration products. Another observation in Figure 8.3 is that prolonged
curing is essential to remove the potentially problematic large-size capillary pores in the

presence of the clean coal technology by-products.

134

 

 

 

 

 

 

3
B
o
F
o
°.
0
x
v
o
E
2
O
>
c
.2
§
I
E “1 rrrll_r r I'tf A
an ”DOV-NQNNQNNO’ much
gtgmgasmehqggmvggﬁggg
*eamnolu- ‘N""°°°Soqo,oo.g
oooooo

 

Pore Diameter (micro)

 

Figure 8.3 Pore Size Distribution of the Clean Coal Technology By-Product Concrete.

Figure 8.4 compares the capillary pore size distributions of Portland cement, Class F coal
ﬂy ash and clean coal technology by-product concretes after 56 days of moist curing.

The clean coal technology by-product and particularly coal ﬂy ash were effective in
reducing the volume content of large capillary pores. The clean coal technology by-
product produces a performance somewhat between coal ﬂy ash and pure Portland
cement. The reduction of large capillary pore volume is a key positive effect of these

pozzolanic materials in concrete.

135

 

llT
w w a m w

.28 so... 5 oEa_o> 533....

l illili - 2.5.8
‘ i .. 88.8

10

.e 88.8
H 38.8
mm 58.8
.. 83

h as
.m «and
H. ~88
w 8..

1N5?

 

Pore Diameter (Micron)

 

 

l —e— Portland Cement + Fly Ash - - le- - -Clean Coal Technology By-Productl

Figure 8.4 Pore Size Distribution at the Age of 56 Days.

l 3 6
8.4 STRUCTURE-PROPERTY RELATIONSHIP

In this section we intend to develop a relationship between the pore system characteristics
and the compressive strength of concrete. Since larger pores tend to be more damaging,
we have divided the capillary pore system to two groups - those larger than 1 micron, and
those smaller than 1 micron. Since we intend to derive generic structure-property
relationships, we have considered all mixes, irrespective of mix proportions and binder

composition.

Figure 8.5 presents the relationship between compressive strength and total capillary
porosity. There is a trend in compressive strength to drop with increasing capillary

porosity.

The division of capillary pores to larger and smaller pores did not signiﬁcantly reduce the

scatter in this relationship (see Figures 8.6 and 8.7).

137

60 T I T I

 

Capillary Porosity (%)

 

0”

 

 

20 l l I; l

3000 4000 5000 6000 7000 8000

 

Compressive Strength (Psi)

Figure 8.5 Total Porosity Versus Compressive Strength.

 

004 i 7 T l ]

     

i.

0.02

cor ’

Large Pore Size (>000‘irrlm)

 

 

 

0 00 l
3000 4000 5000 8000 7000 8000

Compresswe Strength (Psi)

Figure 8.6 Regression Analysis Between Larger Pore Volume and Compressive Strength
(Regression Line and 95% conﬁdence region).

006

003

Small Pore Size (<0 001mm)

0.02

0.04 1

3000 4000 5000 6000 7000 8000

138

 

T T i I l

Compressive Strength = 12259 - 180505 * Small Pore Size

0 09
N

   
      
  
  

$00 °

 

 

l .

l

Compresswe Strength (Psi)

Figure 8.7 Regression Analysis Between Smaller Pore Volume and Compressive
Strength (Regression Line and 95% conﬁdence region).

CHAPTER 9

SUMMARY AND CONCLUSIONS

9.1 INTRODUCTION

Coal ﬂy ash and clean coal technology by-products represent major volumes of solid
waste. Broad implementation of clean coal combustion technologies have led to the
generation of increasing volumes of clean coal technology by-products. This research
determines the potential of a clean coal technology (calcium spray dryer) by-product to
partially replace Portland cement in concrete. The research compares the performance of
conventional coal ﬂy ash with this clean coal technology by-product, and also addresses
some key durability characteristics of concrete materials incorporating these industrial

by-products.

9.2 EXPERIMENTAL PROGRAM

The experimental study investigated the effects of ﬁve variables on the engineering
properties and structure of concrete in a 25 factorial design of experiments (with three
replications):

- ash type (conventional Class F ﬂy ash versus the by-product of the calcium spray

dryer clean coal technology)

139

140

o ash content (15% and 30% replacement by weight of cement)
e water-binder ratio (0.4 and 0.45 by weight)
0 moist curing period (28 and 56 days)

0 air content (2.5% and 7.5%)

The concrete specimens produced based on the above experimental design were subjected
to the following tests:

0 F reeze-thaw durability

o Deicer salt scaling resistance

0 Air void system characteristics

0 Chloride permeability

o Compressive strength

a Mercury intrusion

9.3 EXPERIMENTAL RESULTS ON ENGINEERING PROPERTIES

9.3.1 Clean Coal Technology By-Product Vs. Conventional Coal Fly Ash

Compared to conventional coal ﬂy ash, the clean coal technology by-product had lower

SiOz, A1203 and F6203 contents but higher CaO, MgO and 803 contents. Its Loss On

l4l
Ignition was lower than that of coal ﬂy ash, and it also had a somewhat higher ﬁneness

(i.e. it was slightly coarser).

Conventional coal ﬂy ash produced concrete materials with ~ 5% higher compressive
strength than the clean coal technology by-product. This could be attributed to the
somewhat higher workability and compatibility of ﬂy ash concrete. The overall
performance of the clean coal technology by-product was satisfactory ﬁ'om the strength
point of view. Compared to coal ﬂy ash, the clean coal technology by-product produced
concretes with ~9% more chloride permeability. The difference was, however,
statistically insigniﬁcant. Both conventional coal ﬂy ash and clean coal technology by-
products were highly effective in reducing the chloride permeability of concrete. Their

increased dosages led to ﬁrrther reduction of concrete permeability.

The clean coal technology by-product produced concretes with ~ 5% improvement in
deicer salt scaling resistance when compared with conventional coal ﬂy ash. Increased
dosages of the clean coal technology by-product enhanced the scaling resistance of

concrete while the reverse was true for conventional coal ﬂy ash.

Both conventional coal ﬂy ash and the clean coal technology by-product produced

concretes of sufﬁcient freeze-thaw durability as far as sufﬁcient air content was entrained.

142
9.3.2 EFFECTS OF 15% AND 30% CLASS F FLY ASH

Replacement of 15% of cement with Class F ﬂy ash generated concrete materials with
comparable strength, reduced chloride permeability, improved scaling resistance, and
comparable freeze-thaw durability. At 30% ﬂy ash content, concrete materials (when
compared with plain concrete) possessed reduced compressive strength, improved
chloride permeability, reduced deicer salt scaling resistance, and comparable freeze-thaw
durability. At both 15% and 30% replacement levels, ﬂy ash concrete responded better to

prolonged moist curing than plain concrete.

9.3.3. EFFECTS OF 15% AND 30% CLEAN COAL TECHNOLOGY BY-
PRODUCT

Replacement of 15% of cement with the clean coal technology by product produced a
comparable compressive strength, improved chloride permeability, and comparable
scaling resistance and freeze-thaw durability. At 30% dosage of the clean coal
technology by-product, concrete showed reduced compressive strength, reduced chloride
permeability, and comparable scaling resistance and freeze-thaw durability. In the
presence of the clean coal technology by-product, concrete responded better to prolonged

periods of moist curing.

143
9.4. STRUCTURE - PROPERTY RELATIONSHIPS

The capillary pore size distribution of concrete was affected by the presence of ﬂy ash
and the clean coal technology by-product Fly ash and, to some extent, the clean coal
technology by-product were effective in reducing the total pore volume of concrete. This
effect was more pronounced after longer moist curing periods. Fly ash concrete had ﬁner
capillary pores which could result from the ﬁlling of larger pore with the products of
pozzolanic reaction. Prolonged moist curing, which promotes continued pozzolanic
reaction, led to further reﬁnement of the capillary pore system in ﬂy ash concrete. The
clean coal technology by-product reduced the uniformity of pore size distribution which
may reﬂect on blocking of some pores. There was an increase in the volume of larger
pores, which diminished with continued moist curing. The blocking of pores has positive
effects from the permeability point of view, but the increase in the volume of larger pores

damages strength.

Capillary porosity was found to be a dominant and fundamental property determining the
compressive strength of concrete. Irrespective of the mix composition and the presence
of ﬂy ash or the clean coal technology by-product, compressive strength showed a clear

drop with increasing capillary porosity.

BIBLOGRAPHY

Chapter 2

1

. Sidney Mindess and J. Francis Young “Concrete” Published by Prentice-Hall, Inc.

pg. 569-570.

. Frederick P. Browne and Philip D. Cady “Deicer Scaling Mechanisms In Concrete”,

ACI SP 47-6,pg.101-ll9.

. S.H. Gebler and P. Klieger “Effect of Fly Ash on the Durability of Air-Entrained

Concrete”, ACI SP 91-23, pg.483-519.

. Higginson, E., “Manual de Control del Concrete.” Publication Numero 68, Del

Instituto de Ingenieria, Universidad Nacional Autouma de Mexico, 1962.

. Abdun-Nur, E., “Fly Ash in Concrete, an Evaluation,” Bulletin 284, Highway Research

Board, 135 pages, 1961.

. Davis, R., “A Review of Pozzolanic Materials and Their Use in Concretes,”

Symposium on Use of Pozzolanic Materials in Mortars and Concretes, STP-99,
American Society for Testing and Materials, Philadelphia, PA, pages 3-15, 1949.

. Davis, R., “Pozzolanic Materials - With Special Reference to Their Use in Concrete

Pipe,” Technical Memorandum, American Concrete Pipe Association, Arlington, VA.
30 pages. 1954.

. Verbeck, G. and Klieger, P. “Studies of ‘Salt’ Scaling of Concrete,” Highway Research

Board Bulletin 150, Jan. 1956, 13 pages. Also PCA Research Department Bulletin 83.

9. P. Klieger and S. Gebler “Fly Ash and Concrete Durability” ACI SP loo-56,1987,

pp.1043-1069.

10. C. Johnston “Effects of Microsilica and Class C Fly Ash on Resistance of Concrete to

Rapid Freezing and Thawing and Scaling in the Presence of Deicing Agents”, ACI SP
100-61, pg.l 183-1204.

1 1. V. Malhotra, G. Carette and T. Bremner “Durability of Concrete Containing

Supplementary Cementing Materials in Marine Environment”, ACI SP 100-63,
pg. 1227-1259.

12. Samuel S. Tyson “Freeze-Thaw Durability of Coal Fly Ash Concrete”, Materials

Symposium on Concrete for the 1990’s, Hyatt Regency Hotel, Columbus, Ohio,
November 1, 1989.

144

145

13. “Deicer-Scaling Resistance of Lean Concrete Containing Fly Ash”, Portland Cement
Association, Concrete Technology Today, July 1989, Spp.

14. Roger G. Pertersen “Design and Analysis of Experiments” Published by Marcel
Dekker, Inc. pp. 74-77.

Chapter 3

1. David C. Stark "Effect of Vibration on the Air -Void System and Freeze-Thaw
Durability of Concrete," Portland Cement Association, Research and Development
Bulletin RD092.01T, 1986.

2. Backstrom, James E.; Burrows, Richard W.; Mielenz, Richard C.; and Wolkodoff,
Vladimir E., "Origin, Evolution, and Effects of the Air Void System in Concrete, Part
3-Inﬂuence of Water-Cement Ratio and Compaction," Journal of the American
Concrete Institute, No. #, Volume 330, September 1958, pp 359-375.

3. Crawley, Walter 0., "Effects of Vibration on Air Content of Mass Concrete," Journal
of American Concrete Institute, No. 10, Vol. 24, June 1953, pp. 909-920.

4. S. H. Gebler and P. Klieger, "Effect of Fly Ash on the Air-Void Stability of Concrete,"
American Concrete Institute publication SP-79 July-August 1983.

5. Larson, 0., "Effect of Substitutions of Fly Ash for Portions of Cement in Air-
Entrained Concrete, Highway Research Board, Proceedings of the Thirty-Second
Annual Meeting, 1953, pp. 328-335.

6. Pasko, T.,Larson, T., "Some Statistical Analyses of the Strength and Durability of Fly
Ash Concrete," ASTM Proceedings Vol. 62, 1962, pp.1054-1067.

7. Perenchio, W. and Klieger, P., "Further Laboratory Studies of Portland-Pozzolan
Cements," (RD041T) Portland Cement Association, Skokie, IL, 1976.

8. Isenberger, K, "Fly Ash Concrete-Compressive Strength and Freeze-Thaw
Durability," Iowa Department of Transpiration, Highway Division, Ofﬁce of
Materials, 1981.

9. Pavlovich, R., "Fly Ash as a Partial Replacement for Portland Cement for-Concrete
and Soil Cement - Mixture Physical Properties," Wyoming Highway Department,
1 979.

10. Larson, T., "Air-Entrainment and Durability of Aspects of Fly Ash Concrete," ASTM
Proceedings, Vol. 64, 1964, pp. 866-886.

146

11. Dodson, V., "Foam Index Test," Presented at the Transportation Research Board,
Washington, DC, January 1980.

12. Backstrom, J .E.; Burrows, R.W.; Mielenz, R.C.; Wolkodoff, V.E. Origin, "Evolution,
and Effects of the Air-Void System in Concrete - Part 3-Inﬂuence of W/C Ratio and
Compaction." American Concrete Institute Proceedings; July, 1958 - June, 1959;
Volume 55.

13. Cordon, W.A., "Freezing and thawing of concrete - Mechanism and Control."
American Concrete Institute Monograph No. 3. Iowa State: American Concrete
Institute.

14. Gutrnann, P.F., "Bubble Characteristics as Strength and Freeze/Thaw Durability."
AC1 Material Journals; September- October, 1988.

15. Fagerlund, G., "Critical Moisture Contents at Freezing of Porous Materials." 2nd
International CBI/RILEM Symposium on Moisture Problems in Buildings,
Rotterdam, September 10 to 12, 1974.

16. Signiﬁcance of Test and Properties of Concrete and Concrete- Making Materials.
STB 169B. American Society for testing and materials; 1978.

17. Michel Pegeon, Francois Saucier, and Patrick Plante, "Air-Void Stability, Part IV:
Retempering," ACI Materials Journal, May-June, 1990, pp. 252-259.

Chapter 4

1. US. Department of Energy, "Clean Coal Technology Demostratioin Program - Final
Programmatic Environmental Impact Statement, "Publication DOE/EIS-0146,
November, 1989.

2. Stone, Ralph and Richard, L. Kahle. Environmental Assessment of Solid Residues
from Fluidized-Bed Fuel Processing: Final Report, EPA-600/7-78-107. June, 1978.
Prepared for US. Environmental Protecting Agency.

3. The National Lime Association. Effecive Use of Lime for Flue Gas Desulfurization.
Conference Proceedings, Denver, Colorado, September 27-28, 1983.

4. Eriksen, Robert L. “The Development of Dry Flue Gas Desulfurization at Basin
Electric Power Cooperative,” presented at the Second Conference on Air Quality
Management in the Electric Power Industry, Austin, Texas, January 22-25, 1980.

I47

5. SCS Engineers. Recovery, Utilization, and Disposal of Solid By-Products
Generated by Dry Flue Gas Desulfurization Systems: State of the Art and Research
Needs, RP 1260-16, Final Report, March 1981, prepared for Electric Power Research
Institute.

6. Martin, G. Blair and James H. Abbott. “EPA’s LIMB R&D Prograrn-Evolution, Status
and Plans.”

7. Combustion Engineering, Inc. Feasibility of Furnace Limestone Injection for 802

Control, RP 1836-3, Final Report, September 1983, prepared for Electric Power
Research Institute.

8. Radian Corporation, “Dry Reagent-Dry System.” In The Evaluation and Status of
Flue Gas Desulfurization Systems, RP 982-28, Final Report, January 1984, prepared
for Electric Power Research Institute.

9. KVB, Inc. Bench-Scale Study of the Dry Removal of S02 with Nahcolite and Trona,
RP 982-8, Final Report, March 1981, prepared for Electric Power Research
Institute.

10. Grutle, ROM. and DC. Gehri. “Perspectives on the Development of Dry Scrubbing-
The Coyote Story,” Proceedings: Symposium on Flue Gas Desulfurization - Houston,
October 1980, Volume 2, EPA-600/9-81-019b, April 1981, prepared for the US.
Environmental Protection Agency.

11. Radian Corporation, "Sampling and Analytical Protocal for Advanced Sulfur Dioxide
Control Wastes." Publication EPRI CS-5625, Electric Power Research Institute,
February 1988.

12. Perri, J .S., Collings, B.A., Manz, OE, and Golden, D.M., “Utilization of Advanced
802 Control By-Products: Laboratory Test Results,” EPRI Report CS-5362,
Volume 1 1987, pp. 8-1 to 8-19.

13. Baker/T SA, Inc. Classiﬁcation of Fly Ash for Use in Cement and Concrete. EPRI CS-
5116. April 1987.

Chapter 5

1. Mehta, P. Kumar and Monteiro, Paulo, “Concrete,” Second Edition, Prentice Hall,
1993, pp. 28.

2. Mehta, P. Kumar and Monteiro, Paulo, “Concrete,” Second Edition, Prentice Hall,
1993, 548 pp.

148

3. Philleo, Robert E., and Clifton, James R., “Pore Structure of Concrete and Freezing
Vulnerability of High-Performance Concrete,” NTST Report 89-4186, November
1989, 18 pp.

4. Beaudoin, J. J ., F eldman, R. F. and Tumidajski, P. J ., “Pore Structure of Hardened
Portland Cement Pastes and Its Inﬂuence on Properties,” Advanced Cement Based
Material, January 1994, pp 224 - 236.

5. Attiogbe, Emmanuel K., Nmai, Charles K. and Gay, Frank T., “Air-Void System
Parameters and Freeze-Thaw Durability of Concrete Containing Superplasticizers,”
Concrete International, July, 1992, pp 57-61.

6. Powers, TC. and Brownyard, “Studies of the Physical Properties of Hardened
Portland Cement Paste,” Journal of the American Concrete Institute, Oct.-Dec., 1946;
Jan.- April, 1947, pp. 101-132, 249-336, 469-504, 549-602, 669-712, 845-880, 933-
992.

7. Ryshkewitch, E. J ., American Cer. Soc. 1953, Volume 36, pp. 65 - 68.

8. Schiller, K. K., Cement and Concrete Research, 1971, Volume 1, pp. 419-422.

9. Jambor, J ., “Pore Structure and Properties of Mateirals”, Conference Proceeding,
Prague 1973, Volume 11, 159-161.

10. Graf, H. and Stzer, M., “Pore Structure and Permeability of Cementitious Materials”,
Material Research Society, 1988, pp. 337-347.

11. Power, T. C., J. American Ceram. Society, Volume 41, No. 1, pp. 1-6, 1958

12. Brown, P.W. and Shi, D., “Porosity/Permeability Relationships,” Materials Science
of Concrete II, 1994, pp. 83-109.

13. “Fly Ash in Concrete,” Properties and Performance, Edited by K. Wesche, RILEM
Report 7, E & FN SPON, 1991, pp. 56-57.

14. F eldman, R.F., Beaudoin, J .J ., Proceeding, 6th International Congr. Chem. of Cem.
Moscow, 1974,11, pp.288-293.

15. Feldman, R. F., Beaudoin, J. J. Cem. Con. Res. 1976, pp. 389-400
16. Van Brakel, J. Powder Technology, 1975, Volume 11, pp. 205-236

17. Braf, J ., Setzer, “Pore Structure and Permeability of Cementitious Materials 1988, M.
Matls, Res. Soc., 1988, pp. 337-347.

149

18. Bier, T.A.; Ludirdja, D.; Young, J. F., Berger, RL., “Pore Structure and Pemeability
of Cementitious Mateirals,” Mat. Res. Soc., Proc., 1988, pp. 235-242.

19. Odler, I., Yudenfreund, M., Skalny, J., Brunauer, S., Cem. Con. Res., 1972, pp.463-
480.

20. Manmohan, D.; Mehta, P., “Enﬂuence of Pozzolanic, Slag, and Chemical Admixtures
on the Pore Size Distribution and Perrneabilities of Hardened Cement Pastes,”
Cement, Concrete and Aggregates - CCAG DP, 1981, Volume 3, pp.63-67.

Chapter 6

l. Kosmatka, S. H. and Panarese, W. C., 1993, “Design and Control of Concrete
Mixtures,” Portland Cement Association, Thieteenth Edition.

 

nrcHran STATE UNIV. LIBRARIES
lllllllllllllllllllllllllllllllllllllllllllllllllll
31293014116002