.5... ....|\
.3 . )9; n ,r

.V‘....|l...\.!a? } I \J:
. f. 215:"! .z‘iirtt

7‘.
.y)!(

 

Isl-Ir!
.fvt‘) .I

 

 
 
 
 

THESE»

Date

0-7639

                                                                                                                              

3 1293 00899 758

This is to certify that the

thesis entitled

PERFORMANCE CHARACTERISTICS AND
FAILURE MECHANISM OF STRUCTURAL LIGHT—WEIGHT ‘
VS. NORMAL-WEIGHT CONCRETE MATERIALS 1

presented by '

Mahboob Khan

has been accepted towards fulfillment
of the requirements for

. _j h A ‘ . -
xflflé degree in C 1' r/ /L [79“! Match 17'

\\ Cm \\\_/ M*\u(‘ac_

ajor professor

 

55/9/WS’O

MS U is an Affirmative Action/Equal Opportunity Institution

 

i LIBRARY
iMlchlgan State
1 University

L____.~

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

 

Writ.“ ' J2 ‘5 r.-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution

cthmWnpma-p ‘

PERFORMANCE CHARACTERISTICS AND
FAILURE MECHANISM OF STRUCTURAL LIGHT-WEIGHT

VS. NORMAL-WEIGHT CONCRETE MATERIALS

BY

Mahboob Khan

A THESIS

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

MASTER OF SCIENCE

Department of Civil and Environmental Engineering

1990

 

 

 

Ma- 5754‘

ABSTRACT

PERFORMANCE CHARACTERISTICS AND
FAILURE MECHANISM OF STRUCTURAL LIGHT-WEIGHT
VS. NORMAL-WEIGHT CONCRETE MATERIALS

by
Mahboob Khan

Various aspects of the performance characteristics of
light-weight concrete materials made with a ceramic-based
coarse light-weight aggregate were investigated and compared
with those of normal-weight concrete with similar mix
proportion (by volume) and fresh mix properties.

Two synthetic fibers with different elastic moduli when
used in light—weight and normal-weight concrete produced
comparable effects on flexural strength and toughness, but
the higher—modulus fiber gave higher impact strengths.
Fibers were generally found to be more effective in light-
weight concrete than in normal-weight concrete.

Light-weight concrete materials were, on the average,
40% more permeable than normal-weight concrete materials.
Compressive loading up to 40% of the compressive strength
increased the average permeability of normal—weight and
light—weight concrete by 29% and 8%, respectively,
indicating more severe damage in normal-weight concrete

under compression.

 

The microcracking and failure mechanism of light-weight
concrete under compression and impact loading is
distinguished from that of normal-weight concrete by the
lack of microcrack at the aggregate-paste interfaces and

reduced intensity of microcracks in light-weight concrete.

To Imtiaz Begum

iv

 

 

ACKNOWLEDGEMENTS

The research reported herein was supported by the Research
Excellence Fund of the State of Michigan and the Composite
Materials and Structures Center of Michigan State
University. The materials used in this study were provided
by th 3-M Industrial Mineral Production Division, Allied
Signal Corporation and Fibermesh Company. Their
contributions are gratefully acknowledged. The author would
also like to express his gratitude to Dr. Parviz Soroushian

for his supervision, guidance and encouragement.

TABLE OF CONTENTS

CHAPTER PAQE
List of Tables ix
List of Figures xii
List of Abbreviations xvi
Chapter - Objectives and Scope 1
Chapter 2 — Review of Literature 4
2.1 Introduction ‘ 4
2.2 Light—Weight Aggregate 6
2.3 Mix Design and Manufacturing 7
2.3.1 Light-Weight Concrete Materials 8
2.3.2 Peculiarities of Light-Weight 9
Concrete Mix Proportioning
2.3.3 Mix Proportioning 12
2.3.3.1 Cement Content Method
2.3.3.2 Weight Method
2.3.4 Adjustment.of Mix Proportions
2.3.5 Examples of Mix Proportions
2.3.6 Peculiarities in Manufacturing
Light-Weight Concrete
2.4 Mechanical Properties of Light-Weight 15
Aggregate Concrete.
2.4.1 Compressive Strength 16
2.4.2 Tensile Strength 18
2.4.3 Modulus of Elasticity and 19
Poisson's Ratio
2.4.4 Dynamic Behavior Under Impact 21
Loads
2.4.5 Ductility 21
2.4.6 Creep 22
vi

2.5 Physical Properties 24
2.5.1 Shrinkage 24
2.5.2 Specific Gravity 28
2.5.3 Thermal Properties 31

2.6 Durability and Permeability 32
2.6.1 Permeability 32
2.6.2 Freeze-Thaw Durability 33
2.6.3 Abrasion Resistance 34
2.6.4 Marine Durability and Sulfate 35

Resistance
2.6.5 Fire Resistance 35

2.7 Application _ 36
2.7.1 General 40
2.7.2 Design with Light—Weight Concrete
2.7.3 Economic Aspects of Using Light-Weight

Concrete
Chapter 3 - Effectiveness of Different 43

Synthetic Fibers in Normal-
and Light—Weight

Concrete
3.1 Introduction 43
3.2 Background 44
3.3 Experimental Program 47
3.4 Experimental Results and Discussions 57
3.4.1 Flexural Behavior 58
3.4.2 Impact Strength 69
3.5 Summary and Conclusions 77
Chapter 4 - Permeability of Normal- and 81
Light-Weight Concrete:
Effects of Preloading
4.1 Introduction 81
4.2 Objectives and Scope 82
4.3 Measurement of Permeability 83
4.4 Experimental Program 85
4.5 Test Procedures 88
4.6 Experimental Results and Discussions 91
4.7 Summary and Conclusions 96

 

Chapter 5 - Microcracking and Failure
Mechanism in Light-Weight
versus Normal-Weight
Concrete

1 Introduction

2 Background

3 Experimental Program

4 Image Analysis Procedure

5 Experimental Results and Discussions
6 Summary and Conclusions

Chapter 6 - Summary and Conclusions

6.1 Introduction _

6.2 Effectiveness of Different Synthetic
Fibers in Normal-weight and Light-
Weight Concrete

6.3 Permeability of Normal- and Light-
Weight Concrete Effects of Preloading

6.4 Microcracking and Failure Mechanism
in Light-Weight versus Normal-Weight
Concrete

Appendices:
I. Description of Light-Weight Aggregates

II. Mix Proportioning Procedures

List of References

viii

 

—_—_x~_._...-__-s.... -l._ ,

99

99
102
107
110
120
122
122
123
126

129

131
146

157

Table

Table

Table

Table

Table

Table

Table

Table
Table

Table

Table

Table

Table

Table

Table

3.2

3.3

3.8

3.9
3.10

3.11

3.12

LIST OF TABLES

Gradation Requirements for
Light-Weight Aggregates.

Comparison of Fineness Meduli
by weight and by Vblume for
Typical Light-Weight Aggregate.

Requirements for Structural
Light-Weight Concrete.

The Experimental Program of
Light-Weight and Normal-Weight
Fiber Reinforced Concrete.

Gradation of Normal-Weight
Concrete.

Gradation of Normal-Weight
Fine Aggregate

Gradation of Light-Weight
Aggregate.

Properties of Synthetic Fibers.

Mix Proportion of Normal-Weight
Concrete.

Mix Proportion of Light-Weight
Concrete.

Size and Number of Hardened
Material Test Specimens.

Properties of Fresh Concrete.

Flexural Test Results for
Normal-Weight Concrete.

Flexural Test Results for

Light-Weight Aggregate Concrete.

Flexural Toughness Test Results
for Normal-Weight Aggregate
Concrete.

ix

PAGE

11

19

48

50

50

51

51
52

53

54

57
60

61

62

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table
Table

Table

Table

Table

Table

3.13

3.14

3.15

3.16

3.17

5.3
II.1

II.2

II.3

II.4

II.5

Flexural Toughness Test of

Light-weight Aggregate Concrete.

Impact Test Results (No. of
Blows to First Crack) for
Normal-weight Concrete.

Impact Test Results (No. of
Blows to Failure) for
Normal-Weight Concrete.

Impact Test Results (No. of
Blows to First Crack) for
Light-Weight Concrete.

Impact Test Results (No. of
Blows to Failure) for
Light-Weight Concrete.

AASHTO T277-83 Test Classes of
Chloride Permeability.

Chloride Permeability Test
Results.

Mix Proportion.
Compressive Strength Test

Results of Normal-weight
and Light-weight Concrete.

First Crack Impact Test Results.

Recommended Slump Values.

Cement Content of Light-Weight
and Normal-Weight Concrete.

Approximate Water and Air
Contents Required in Air

Entrained Light-Weight Concrete.

Estimated Water-Cement Ratios.

VOlume of Coarse Aggregate per
Unit VOlume of Concrete.

PAGE

63

70

71

72

73

85

93

103

104

106
147
148

149

151

152

PAGE

Table 11.6 Guidelines for Adjustment of 153
Mixes of Light-Weight Concrete.

xi

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

2.4

2.5

2.6

 

LIST OF FIGURES

Classification of Light-Weight
Concrete.

Effect of Aggregate Shape on
the Compressive Strength of
Light-Weight Concrete.

Typical Compressive Stress-
Strain Curves for Light-Weight
Concrete.

Typical Effects of Confinement
on Stress-Strain Behavior of
Light—Weight and Normal-Weight
Concrete.

Comparison of the Creep of
Light-Weight and Normal-Weight
Concrete.

Relation Between Drying
Shrinkage After 2 Years an
Secant Modulus.

Drying Shrinkage of Prismatic

Light-Weight Concrete Specimens

with Different Dimensions.

Strength VS. Density of Light-
Weight Concretes.

Specific Gravity Factor Vs.
Moisture Content Light-Weight
Aggregates.

Thermal Conductivity Vs. Unit
Weight of Concrete.

Temperature Effects on
Compressive Strength.

Effect of Polypropylene Fiber

Reinforcement on the Workability
of Fresh Light-Weight Concrete.

xii

PAGE

17

20

22

23

25

27

29

30

31

36

45

 

Figure

Figure

Figure
Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

3.
3.
3.
3.

3.

4.
5.
6.
7.

3.9.

3.

4.

4.

10.

.11.

PAGE

Effect of Fiber Reinforcement 46
on the Mechanical Properties
of Light-Weight Aggregate Concrete.

Gradation of Coarse Light-Weight 49
and Normal-Weight Aggregate
ASTM C-330 and ASTM C-33.

Flexure Test Set-Up. 55
Impact Test Set-Up. 56
Flexural Load Deflection Curve. 59

Effect of Synthetic Fiber 64
Reinforcement at 0.25% Volume

Fraction on the Flexural

Strength and Toughness of Normal-
Weight and Light-Weight Concrete.

Relationship Between Flexural 68
Strength Ratio and Modular Ratio.

Relationship Between Flexural 69
Toughness Ratio and Modular Ratio.

Effect of Synthetic Fiber 74
Reinforcement at 0.25% Volume

Fraction on the Impact Strength

at First Crack and Failure of
Normal-Weight and Light-Weight
Concrete.

Relationship Between Impact 77
Strength Ratio and Modular Ratio.

Relationship Between Charge 84
Passed During Test and Total

Integral Chloride Valves in

90 Day Test.

Chloride Permeability Apparatus. 89

xiii

Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

4.3.
4.4.
4.5.

5.8.

5.9.

5.10

 

Specimen for Test.
Electrical Block Diagram.

Effect of 40% Compressive Load
on Permeability of Normal-
Weight and Light-Weight Concrete.

Stress Strain Curve of Concrete
in Compression.

Microcrack Growth in Concrete
Under Compression in Normal-
Weight Concrete.

Location of Transverse and
Longitudinal Slices.

Longitudinal and Transverse
Slices Ready to be Reviewed.

Image Analysis System.

Microcracks in Concrete
Materials.

Microcrack Intensity at
Different Compressive Levels
in Normal-Weight Concrete.

Microcrack Intensity at
Different Impact Loading Levels
in Normal-Weight Concrete
(Transverse Section).

Microcracks in Normal—Weight
Concrete Material (Magnification
5x).

Microcrack Intensity at
Different Compressive Stress
Levels in Light-Weight Concrete.

Microcrack Intensity at
Different Impact Loading Level
in Light-Weight Concrete
(Transverse Section).

xiv

._.-._.-‘~ .

—-.- .....

PAGE

90
91
92

100

101

105

107

108
109,110

111

111

113,114

115

115

PAGE

Figure 5.12. Microcracking of Light-Weight 118,119
Concrete Materials (Magnification
5x) .

 

 

 

NWC
LWC
PPF
PEF

NWPPFRC

LWPPFRC

LWPEFRC

Mod
Elast
Flex

FR

LIST OF ABBREVIATIONS

Normal Weight Concrete
Light-Weight Concrete
Polypropylene Fiber
Polyethylene Fiber

Normal-Weight Polypropylene Fiber Reinforced
Concrete

Light-Weight Polypropylene Fiber Reinforced
Concrete

Light-Weight Polyethylene Fiber Reinforced
Concrete

Modulus
Elasticity
Flexure

Fiber

xvi

 

 

CHAPTER 1

OBJECTIVES AND SCOPE

A comparative study was made on the following
properties of light-weight vs. normal-weight concrete
:materialsx (l) effectiveness of synthetic fibers in
enhancing flexural strength and toughness; (2) permeability
characteristics prior to and after loading; and (3) the
nature of microcracking and failure mechanism under
compression and impact.

In work on fiber reinforcement effects it was checked
if reduced fiber-to—matrix modular ratio in light-weight
concrete (noting that light-weight aggregate reduces the
modulus of elasticity of concrete material) leads to
increased effectiveness of fibers.

Light-weight aggregates are more porous than normal-
weight aggregates; they are, however, more compatible with
the cementitious paste (leading to reduced microcracking at
the paste—aggregate interfaces). These two conditions have
opposite effects on permeability. Increased loading also
causes internal damage in light-weight and normal-weight
concrete materials in distinctly different ways, producing
different effects on permeability. These complex
differences in the permeability characteristics of light-

weight concrete (loaded and unloaded) prompted the work in

this investigation on permeability.

 

 

 

.-_ ____.—_.t..c.._.i... . 1.. NW

The last part of this research dealt with the nature of
microcrack propagation and failure in light-weight vs.
normal-weight concrete materials subjected to compression
and impact loading. Failure in normal-weight concrete
emanates from defects such as microcracks existing at
aggregate-paste interfaces. Under loading, these
microcracks tend to propagate at interfaces and also through
the paste. In light-weight aggregate concrete, the
microcrack intensity at aggregate-paste interfaces is
expected to be less severe than in normal-weight concrete,
and the weakness of light-weight aggregates encourages the
propagation of microcracks through aggregates. These
differences were the subject of investigation in the last
part of this research.

Light-weight concrete materials in this investigation
were made with a ceramic based coarse light-weight aggregate
(and normal-weight fine aggregate). All the normal-weight
and light-weight concrete materials were designed to possess
similar fresh mix proportions and comparable aggregate
gradations; light-weight materials, however, had lower
compressive strength when compared with normal-weight
materials.

The conclusions derived in this investigation are
mainly applicable to the ceramic-based light-weight
aggregates used, and also to the specific ranges of
compressive strength and impact resistance of light-weight

and normal-weight concretes considered.

 

Chapter 2 of this thesis presents a comprehensive
review on the literature of light-weight concrete; this
literature review covers topics beyond the specific subject
matters of this thesis. The work conducted in this research
on fiber reinforcement, permeability characteristics, and
microcracking are presented in Chapter 3, 4, and 5,
respectively. .A summary of the work and the resulting

conclusions are presented in Chapter 6.

CHAPTER 2

REVIEW OF THE LITERATURE

2.1 INTRODUCTION

The unit weight of normal-weight concrete varies from
about 140 to 152 lb/cu.ft. (2258-2435 kg/cu.m.) and it is
generally assumed to be 145 lb/cu.ft. (2324 kg/cu.m.). The
unit weight of concrete can be reduced through the use of
light-weight aggregates. Light-weight concrete may weigh
from 20-115 lb/cu.ft. (320-1843 kg/cu.m.), depending on the
type of light-weight aggregate used and the method of
production. The strength of light-weight aggregate is
roughly proportional to its density. The advantages of
light-weight concrete over normal-weight concrete generally
result from the decrease in foundation size because of
reduced load, the increase in fire reSistance, and the
insulation against heat and sound when light-weight concrete
is used. Light-weight aggregate concrete, however, is
typically 30 to 50% more expensive than normal-weight
concrete (1,6), and it has a greater porosity and more
drying shrinkage than ordinary concrete.

Light-weight concrete can be classified in accordance
to its density, or alternatively based on the purpose for
which it is to be used. Distinctions can be made between
structural and insulating light-weight concretes.
Structural light-weight concrete has a unit weight between

90-115 lb/cu.ft. (1400-1840 kg/cu.m.), with a 28-day

 

 

compressive strength not less than 2500 psi (17 MPa).
Insulating light-weight concrete has a density generally
lower than 50 lb/cu.ft. (800 kg/cu.m.) and a compressive
strength between 100-1000 psi (0.7-7 MPa). Light-weight
concretes with properties in between those of insulating and
structural concretes can be classified as moderate strength
light-weight concrete. (2,7,8)

A variety of light-weight aggregates (natural or man-
made) have been used for the production of light-weight
aggregate concrete. Such aggregates typically weigh less
than 70 lb/cu.ft. (1120 kg/cu.m.). Their light weight
results from the cellular or highly porous nature of their
microstructure. The lightest type of aggregate possible is
air which is used to reduce unit weight in aerated
(cellular, foamed) concrete. Examples of the light-weight
aggregate types commonly used for producing different
classes of light-weight concrete are presented in Figure

2.1.

i Insulating i Moderate-strength: Structural l
'1 concretes i concretes } concretes ‘

. 1
UP” lkghnal 250 500 750 1000 1250 1500 1750 2000
weight

I
Strength range ' (0.7-2.0 MPa) : (7-14 MPa)

I

l

Vermiculite

 

T ‘ I
: (17—41 MPal I
Pumice I Sintered fly ash '

i

I

t

 

r::::i

Perlite ! Scoria I I

Expanded clay or shale

I

Foamed slag 1

Aggregate
type

 

 

I
I
l
I I
l
I

Aerated concrete I
WWI/”WWWA i
I

Other lightweight! No lines (lightweight)

concretes I W
: : : (normal wmqhtl

NO iini‘fi

Figure 2.1 Classification of Light-Weight Concrete (4,7).
(kg/cu.m. x 0.0624 = lb/cu.ft.; MPa x 145 = psi)

 

In Figure 2.1, depending on the unit weight and strength of
concrete, distinctions are made between insulating and
structural light-weight concretes.

While the use of light-weight aggregate in structural
concrete elements has been increasing, (5,10) its primary
use is still as an insulating material or in masonry blocks
and secondary structural members (12). Like other
materials, light-weight aggregate concrete also has certain
limitations and it is only after evaluation of all the
merits and demerits with respect to a particular project
that final decision can be made regarding the suitability of
its application (12).

Major technological advancements have been or are being
made in the area of concrete materials. Examples include
the use of short, randomly distributed fibers for improving
the toughness characteristics, impact resistance and tensile
strength of concrete, or the use of polymers in concrete for
improving the impermeability and adhesiveness of the
material. The field of light-weight concrete needs to take
advantage of these technological developments in order to
retain and increase its rapid growth rate in the years to

come .

2.2 LIGHT—WEIGHT AGGREGATE

Light-weight concrete is a broad term covering
concretes made with a wide variety of aggregates, natural or

artificial, with or without treatment. The ranges of

 

7

strength and density, and in fact all the characteristics of
concrete, vary considerably with different aggregates.

Light-weight aggregates can be generally divided into
natural (crushed rock or minerals) and manufactured,
including industrial by-products and recycled waste products
(2,7,14,49). Light-weight aggregates can also be classified
(based on their density) as ultra light 4-35 lb/cu. ft. (64—
561 kg/cu.m.) and light 35-70 lb/cu.ft. (561-1121 kg/cu.m.).
There are also classifications of light-weight aggregates
based on the potential strength, density or other
characteristics of the resulting light-weight concrete.
Light-weight aggregates having dry unit weights ranging from
4 to 35 lb/cu.ft. (64-561 kg/cu.m.) and thermal
conductivities ranging from 0.45 to 1.5 Btu/ft2.hr.°F (0.65
to 0.22 W/mK) are generally classified as group I and are
used for insulation purposes; whereas light-weight
aggregates having dry unit weights from 35 to 65 lb/cu.ft.
(561 to 1042 kb/cu.m.) are classified as group II and are
used for structural purposes (2,11,38).

A summary of the properties and production techniques
for some commonly used light-weight aggregates is presented

in Appendix I.

2.3 MIX DESIGN AND MANUFACTURING
Mix proportioning of concrete consists of selecting
suitable materials (cement, sand, water, etc.) and

determining the quantities of these ingredients. The

 

economy and performance characteristics of the concrete thus
produced depends on the proportioning of these ingredients.
The procedures adopted for mix proportioning are still
empirical in spite of a considerable amount of work done on
the theoretical aspects of mix proportioning.

2.3.1 Light-Weight Concrete Materials

Portland cement type I is normally used in light-weight
aggregate concrete. Structural elements exposed to marine
environments are made of cements having high tricalcium-
aluminate contents (7%), an amount which can resist sulphate
attack on the elements made with light-weight aggregate
concrete in a similar fashion as in normal weight aggregate
concrete.

Fly ash could be used to improve strength and sulphate
resistance, and to reduce water absorption and improve
curing properties. Light-weight aggregates should satisfy
ASTM C-330 requirements (7). Typical gradation requirements

are given in Table 2.1.

 

Table 2.1. Gradation requirements for light-weight
aggregates (11).

Size lin 3/4in 1/21n 3/ain No.4 No.8 No.16 No.50 No.100
25mm 19mm 12.5mm 9.5mm 4.75mm 2.36am 1.18mm 3006m 1500B

 

Designation

Fine Aggregate
“0.41:00 "' "’ "

Coarse
ate

lin to No. 4 95-100 - 25-60 - 0-10 - - - -
3/4in to No.4 100 90-100 - 10-50 0-15 - — - -
3/810 to No.8 - - 100 80—100 5-40 0-20 0-10 - -

Combined Fine
and Coarse

Mgr-gate

1/21n to 0 -

3/Bin to 0 - - 100 90-100 65-90 35-65 - 10-25 5-15

Water free from salt and organic matters (fit for drinking)
should be used in light-weight aggregate concrete mixtures.
Air entraining admixtures should conform to ASTM C-26.

2.3.2 Peculiarities of Light-Weight Concrete Mix
Proportioning

Light-weight aggregates tend to segregate and float on
the surface of concrete in high-consistency mixtures. This
problem can be controlled by limiting the maximum slump and
by entraining air in fresh concrete mixtures made with
light-weight aggregates (11).

A major factor necessitating adjustments in the
proportioning and control procedures of normal-weight

concrete when applied to light-weight aggregate concrete is

10

the greater water absorption and higher absorption rate of
light-weight aggregates (ll).

Light-weight aggregate mixtures are usually
proportioned by trial on a cement and air content basis at
the required consistency, rather than on a water-to-cement
ratio basis, because water-to-cement ratio cannot be
established with accuracy due to absorption of water by
light-weight aggregates.

Grading of fine and coarse aggregates and the
proportions used have important effects on concrete
properties. A.well-graded aggregate will have a continuous
distribution of particle size, producing a minimum void
content and requiring a minimum amount of cement paste to
fill the voids. This will result in economical use of
cement and will provide maximum strength with minimum volume
change due to shrinkage and temperature effects (11).

In light-weight aggregates, unlike normal-weight
aggregates, the specific gravities of fractions retained on
different sieves are not equal. Hence, it is the volume
occupied by each size fraction, and not the weight of the
materials retained on each sieve, that decides the void
content and cement paste requirements in a mix.

For a typical light-weight aggregate, the percentage
retained on each sieve by volume and weight are given in

Table 2.2.

11

Table 2.2. Comparison of fineness moduli by weight and
by volume for typical light-weight aggregates.(11)

Sieve Opening Percent Cumulative percent
retained by Specific retained by
;p Mp weight VOlume Gravity weight Volume

 

No. 4 .187 4.67 0 0 1.4 O O

No. 8 .0937 2.36 22 26 1.55 22 26

No. 16 .0469 1.18 24 25 1.78 46 51

No. 30 .0234 .600 19 19 1.9 65 70

No. 50 .0117 .300 14 13 2.01 . 79 83

No.100 .0059 150l#”\ 12 10 2.16 91 93
Passing

No. 100 9 7 2.40 100 100

Fineness modulus by weight = 3.03 Fineness modulus by
volume = 3.23

Fineness modulus by volume is greater than fineness
modulus by weight which indicates that light-weight
aggregates require a higher percentage of materials retained
on fine sieves on a weight basis, than do normal-weight
aggregates, to provide an equal size distribution (11).

The particular properties of light-weight aggregates
pose special problems in calculating mix pr0portions for
light-weight concrete. The absolute volume method, which is
the basis of the ACI method for proportioning normal-weight

concrete, can not be used with confidence for light-weight

concrete. This is due to two major factors.

 

 

12

1. Variations in bulk specific gravity of light-weight
aggregates.

2. Changes in the light-weight aggregate moisture content.

The bulk specific gravity of light-weight aggregates
depends upon porosity which varies with particle size.
Hence, grading curves determined on weight basis cannot be
used for proportioning light-weight concrete because they
cannot be converted directly to volumetric basis, as shown
in Table 2.2.

The mix proportioning procedures commonly used for
light-weight aggregate concrete are presented in Appendix

II.

2.3.3. Peculiarities in Manufacturing Light-Weight Concrete
Production of uniform concretes with light-weight
aggregate involves all the procedures and precautions that
are necessary for ordinary concrete.(1) After mix design,
the ingredients should be mixed according to ASTM C-94 as in
the case of normal-weight concrete. The problem is, however,
more difficult where light-weight aggregates are used
because of greater variations in absorption, specific
gravity, moisture content and gradation of aggregates.
‘Uniform results can thus be obtained if the unit weight and
slump tests are performed frequently, and the cement and
water contents of the mix are adjusted as necessary to

compensate for variations in properties (1,2,11,22,35).

13

Dry light-weightaggregates should not be used at the
mixing stage although they will produce a concrete which can
be placed right after being discharged. Continuous water
absorption by dry light-weight aggregates will cause
concrete to segregate and stiffen before placement is
completed.

In order to insure uniformity, light-weight aggregates
should be wetted 24 hours before use. This wetting will
also reduce segregation during stock piling and
transportation, (1). Presaturation of aggregates will help
achieve uniformity but may reduce freeze-thaw durability
(particularly if concrete is not allowed to dry before
exposure to freezing temperature).

It is generally necessary to mix light-weight
aggregates for longer periods than conventional concrete to
assure proper mixing. WOrkability of light-weight concrete
with the same slump as conventional concrete may vary more
widely because of differences in such properties of light-
weight aggregates as porosity and specific gravity.
Workability of freshly made light-weight aggregate concrete
requires special attention because, with high-consistency
mixtures, the aggregate tends to segregate and float on the
surface. To combat this tendency, it is often necessary to
limit the maximum slump and entrain air. Approximately, 5
to 7 percent air entrainment is generally required to lower
the mixing water requirement while maintaining the desired

slump and reduce the tendency for segregation and bleeding.

14

Placing, compaction and finishing of light-weight
aggregate concrete requires less effort than normal-weight
concrete; therefore, 2 to 3 inches (50 to 75 mm) of slump
may be sufficient to obtain workability levels that are
usually shown by 4 to 5 inches (100 to 125 mm) of slump in
normal-weight concretes. A.slump of 2 to 3 inches (50-75
mm) represents a relatively high workability, and a
compaction factor not less 0.8 or a vebe time less than 12
sec correspond approximately to a medium workability. .A
slump in excess of 2-4 inches (50-100mm) may cause
segregation with large light-weight aggregate particles
floating to the top. The tendency towards floating of
larger particles of light-weight aggregate may also be
improved by adjusting the grading of aggregates. This can
be done by crushing the larger particles, adding natural
sand, or adding filler materials. In order to minimize
segregation prior to discharge, the mix should be rotated at
least ten revolutions at maximum speed (3). Concretes made
'with many light-weight aggregates may be difficult to place
and finish because of porosity and angularity of the
aggregates. The placability of concrete can be improved by
adding air-entraining agents.

Quality control during manufacturing is achieved by
keeping the cement content, slump and volume of dry
aggregate constant per cubic yard of concrete regardless of
variations in absorbed or surface water of aggregate. If

the density of aggregate does not vary, constant volume can

15

be maintained by keeping the dry weight of aggregate
constant. Unit weight of fresh concrete should be
determined at frequent intervals. A change in unit weight
indicates a change in air content or change in weight of
damp aggregate. An air content determination will establish
whether or not the correct amount of air is entrained. If
weight of aggregate is changed, it is due to a change in
moisture content, gradation, and unit weight. The aggregate
weight can be easily measured, and this will reveal the
cause of the change.

Different types of light-weight concrete call for
different aggregate gradings; a particular grading which is
suitable for use with one type of light-weight aggregate may
not be suitable with another type (2). Careful selection
and handling of light-weight aggregates to prevent crushing
and contamination, and to ensure uniform.moisture content
and uniform weight, are necessary for achieving quality

light-weight concrete.

2.4 MECHANICAL PROPERTIES OF LIGHT-WEIGHT AGGREGATE
CONCRETE

The following properties of light-weight aggregate
concrete will be discussed in this section: Compressive and
tensile strengths, modulus of elasticity and Poisson's
ratio, impact resistance, ductility and creep

characteristics.

16

2.4.1. Compressive Strength

 

Design compressive strengths of 3000 to 4000 psi (26.7
- 27.5 MPa) at 28 days are common for structural light-
weight concrete. By using high cement contents and good-
quality light-weight aggregates of small size it has been
possible in some precast and prestressed concrete plants to
produce 6000-7000 psi (41.38 - 48.25 MPa) light-weight
concretes (20). Light-weight aggregates with controlled
micr0porosity have been developed to produce 10,000 to
11,000 psi (69-76 MPa) light-weight concretes which
generally weigh 115 to 125 lb/cu.ft. (1843 to 2000
kg/cu.m.) (8,44,45).

With sintered fly ash and expanded clay or shale there
is no difficulty in obtaining concrete strengths up to 6000
lb/in2 (41 MPa) in spite of the high porosity and inherent
weaknesses of aggregates. Pumice, scoria, and some expanded
slags produce concretes of intermediate strengths near 2000
psi (14 MPa). vermiculite and diatomite produce concretes
of very low strength in range of 200-750 psi (1.4-5.2 MPa)
(8).

The following factors associated with light-weight
aggregates may contribute towards the differences in
compressive strength of light-weight and normal-weight

CODC retes 3

l7

1. Aggregate strength.
. Aggregate stiffness.
Aggregate surface texture.

Differences in surface area.

£31.wa

Aggregate shape.

The compressive strength of light-weight concrete
depends partly upon the shape of light-weight aggregate
particles (see Figure 2.2). Concretes made with round
aggregates (volume concentration of sintered fly ash
aggregates = 0.4) have 28-day compressive strengths about
870 to 1160 psi (6 to 8 MPa) higher than those obtained with
elongated aggregates (length/thickness ratio = 4). The '
differences in specific surface area associated with the

differences in shape contribute to strength variations.

- a: 03.13 days
45
L
Q n = 0“. 28 can
10 -
\>\<: " = 0.3. 2 d5”
as _ n- 0-4.2dayt

30 L l I 1 l

 

V J‘
C J!

i
f

 

COMPRESSIVE STRENGTH C: CONCRETE—\ --1

 

 

O
_.
N
w
A
M
O

n = Porosity of Aggregate
Figure 2.2 Effects of aggregate shape on the

compressive strength of light-weight concrete.(29)

18

The shape of light-weight aggregates affects stress
concentration in concrete under load, causing differences in
the compressive strength of light-weight aggregate concretes

made with different aggregate shapes.

2.4.2. Tensile Strength.

Direct tension test results show great scatter because
of misalignment effects, stress concentration at the grips,
and random effects associated with the location of
aggregates. For this reason, direct tension tests are
seldom conducted on concrete materials. The splitting
tensile strength of concrete cylinders (ASTM C-496) is a
convenient relative measure of tensile strength. The data
in Table 2.3 shows that, like normal-weight concrete, the
ratio between the splitting tensile and compressive
strengths decreases significantly with increasing strength
of light-weight concrete. The modulus of rupture of
continuously moist-cured light-weight concrete also behaves
in the same manner; tests on dried specimens show that the
results are extremely sensitive to the moisture state. For
normal-weight concrete the modulus of rupture is usually
between 8vf’c and 12vf’c, whereas for light-weight aggregate
it may range from 6vfc- and 8vfc- (44).

Examination of fractured specimens of light-weight
concrete after splitting tension test clearly reveals that,
unlike normal-weight concrete, the aggregate, and not the

transition zone, is generally the weakest component in the

 

19

system. Holmi, et.al.(4) have presented scanning electron

micrographic evidence of this phenomenon.

 

Table 2.3. Requirements for Structural Light-Weight
Concrete (4).
Air-dried 28-day 28-day splitting 28-day

unit weight, tensile strength, compressive strength,
ng§.r1b/£t3(kq/m§)l Min. [psi (MPa)] Min. [psi (MPa))

 

All light-weight aggregates

110 (1760) 320 (2.2) 4000 (28)
105 (1680) 300 (2.1) 3000 (21)
100 (1600) 290 (2.0) 2500 (17)

Combination of normal sand and light-weight aggregate

115 (1840) 330 (2.3) 4000 (28)
110 (1760) 310 (2.1) 3000 (21)
105 (1680) 300 (2.1) 2500 (17)

The compressive strength and unit weight shall be
obtained based on the average measurements made on three
specimens, and the splitting tensile strength shall be the

average obtained using eight specimens.

2.4.3. Modulus of Elasticity and Poisson’s Ratio
The modulus of elasticity of light-weight concrete Ec,
that is the slope of the initial straight part of the

compressive stress-strain curve, tends to increase with

20

increasing strength of concrete. Typical compressive

stress-strain curves for light-weight concrete are shown in

Figure 2.3.

 

~80

eu L, ksl

unless we

 

 

0

 

. , 0
0 0.00‘ 0.002 0.003 0 00-1
Strann ( .n .m

Figure 2.3. Typical Compressive Stress—Strain Curves for

Light-Weight Concrete (44).

Modulus of elasticity of light-weight concrete is
typically lower than that of normal-weight concrete. The
following equation is recommended for calculating the
modulus of elasticity of light-weight concrete with
compressive strengths ranging from 3,000 to 7,000 psi (21 to

48 MPa) (44).

RC = 40,000 (fc- + 1,000,000)(Wc/l45)1 5

The poisson's ratio for light-weight concrete ranges from

0.17 to 0.21, with an average value of 0.20.

21

2.4.4. Dynamic Behavior Under Impact Loads

Stress wave velocity is about 20% less in light-weight
concrete than in normal-weight concrete, with the length of
stress wave being about the same. Period of vibration is
longer in normal-weight concrete when compared with light-
weight concrete, and vibration damping is greater, probably
due to the interface condition between paste and coarse
aggregates.

Shock and energy absorption of light-weight aggregate
concrete is believed to be substantially greater than those

for normal-weight concrete.

2.4.5. Ductility

Ductility supplies the ability to sustain inelastic
deformations after peak load without a significant drop in
load- resisting capacity prior to collapse. The ductility
of reinforced concrete members depends, among other factors,
on the stress-strain behavior and ductility of concrete
materials.

Light-weight aggregate concrete is increasingly being
used as a structural material in reinforced and prestressed
concrete structures. The strength properties of light-
weight aggregate concretes have been studied extensively,
but the ductility of the material, especially at higher
strengths and for members confined with lateral

reinforcement, has not been understood (17,34).

 

22

very little information is available on complete
compressive stress-strain behavior of light-weight concrete.
Typical comparisons between the ductility of light-weight
and normal-weight concretes (confined and unconfined) are

shown in Figure 2.4.

\ i Luann-dummy".
—-— Comm I081 W052) 4

- - - Unearth-d

 

 

IO"

(:1
\\,\.

2
\ \
\L
Fve‘olkfl _ ‘\~_-\.-K.:

.s 1.0 1.5 2.0 2.5 ' 3.0x
Rointormt Ratio. 9

I

¢utqrv
//
3

 

 

Ductility Flflov,
/
/

 

 

 

 

 

 

 

 

O

 

Figure 2.4. Typical effects of confinement on the stress-
strain behavior of light-weight and normal-weight concretes

(23,34).

2.4.6 Cregp

There are considerable variations in the creep
characteristics of concretes with comparable densities, and
the magnitude of creep depends on the cement content, water-
cement ratio of the paste, modulus of elasticity of
aggregates, and the rate of moisture loss. Creep is usually
greater in light-weight concrete than in normal-weight

concrete at the same strength (see Figure 2.5).

 

23

Light-weight aggregates have relatively low moduli of
elasticity because of their high porosity. The lower
elastic moduli of light-weight aggregates would offer less

restraint to time-dependent deformations such as creep (7).

 

3

.Wflmw

  

8

Crew Imia omain)
an
8

8

 

 

' 1
0 50 100 ISO 200 250 300 350
Tune ldlnl

Figure 2.5. Comparison of the Creep of Light-Weight and

Normal-Weight Concretes.(7)

Present-day research has proved that the effect of
cement paste content on creep in light-weight concrete is
the same as in normal-weight concrete; however, the modulus
of elasticity of the aggregate affects the magnitude of
creep through the medium of exponent a in the expression
(30):

C = cp(1-g- )a

Creep of concrete (made with either light-weight
aggregate or normal-weight aggregate) can be predicted by

this equation for periods up to 7 months under load.

 

24

a = 3(l-flk)
1 + + 2(1 - 2 a) E/Ea

where,
”L, = Poisson’s ratio of surrounding materials.
= Poisson's ratio of aggregate
= Modulus of Elasticity of Surrounding Materials

(Elasticity)

[11
D)
II

Modulus of Elasticity of Aggregate
= velume of Aggregate

Creep of Concrete

0 CID
II

P = Creep of Cement Paste

2.5 PHYSICAL PROPERTIES

This section presents discussions on the shrinkage,
specific gravity and thermal properties of light-weight
concrete materials.

2.5.1 Shrinkage:

The shrinkage of light-weight concrete is generally
greater than that for normal-weight concrete with similar
mix proportions and consistency, but some impervious light-
weight aggregates produce concretes having relatively low
shrinkage (36). Light-weight aggregates usually give higher
shrinkage because they have lower moduli of elasticity and
thus produce smaller restraint against the shrinkage
movements of cement paste. Aggregates with a larger
fraction of fine materials smaller than 75 microns (No 200
sieve) show higher shrinkage movements, as fines lead to

larger void contents.

 

25

The fact that the elastic properties of aggregates
determine the degree of restraint against Shrinkage
movements is demonstrated in the observation that steel
aggregates lead to shrinkage strains that are one-third
less, and expanded shale light-weight aggregates give_one-
third more shrinkage strains, when compared with ordinary
aggregates (46). A typical correlation between shrinkage
and modulus of elasticity of concrete, which depends on the

compressibility of aggregates, is shown in Figure 2.6.

 

1230' I ‘ f
I I
l.
i
i
l

 

 

1000 ‘ .
\ .i

Vol

o 13’

ta»,
O -

l

 

800

6
l0

 

600

 

l
1.1

I l .
c Lightweight Comrete

Shrinkage -

 

 

 

 

 

 

o NOrmol Wright lConcr-ere
203 I { ‘ l I
x I l l l
v 5 IO ‘ 20 2'.) 30 T":
Figure 2.6. Relation Between Drying Shrinkage after 2

years and Secant Modulus (2).

Ref. 4 suggests that light-weight aggregate concrete
exhibits higher moisture movements (i.e. higher rates of
drying shrinkage) and a somewhat higher ultimate shrinkage
(typically 1600 x 10'6) when compared with normal-weight
(aggregate concrete. It seems that the low strength and low
HKNdulus of elasticity of light-weight aggregates have more

pronounced effects on creep than on shrinkage.

26

As mentioned above, the results of many tests (2,48)
show that the shrinkage of light-weight concrete might be 6
to 38% higher than that of normal-weight concrete. Other
test results (47), however, indicate that shrinkage
movements are comparable in light-weight and normal-weight
concrete. It may thus be concluded that shrinkage strains
in light-weight aggregate concrete depend on its specific
types of light-weight aggregate used.

Figure 2.7 shows the drying shrinkage movements of
plain light-weight aggregate (lytag & sand) concrete for
three different sizes of prismatic specimens tested under
constant conditions of 68°F (20°C) and 50% relative
humidity. It can be observed that at the first two days,
unlike normal-weight concrete, there is some expansion in
light-weight concrete and the smallest specimen shows the
largest expansion. This behavior possibly is due to the
water absorption of relatively high amounts of water by
light-weight aggregates (46). The absorbed.water in the
aggregate tends to move to the surface, a phenomenon which
causes the expansion of concrete. The movements of water
depend on the rate of evaporation at the surface and, as the
smallest specimen has the highest surface area-to-volume
ratio, the highest expansion tendencies at early drying
periods are observed with the smallest specimen. As
expected, the smaller specimens also show higher rates of
shrinkage development which again can be attributed to their

higher rate of moisture loss.

27

 

CREE? + SHRINKAGE x 106
8
O
I l
\

200 A 50 x 50 x 320 mm

0 75 x 75 x 3'70 mm

0‘00 X I“) x420mm
l l I I I l I l l 1
1‘0 160 180 200 220

 

 

 

so 100 I o
navmo PERIOD— dlvl

Figure 2.7. Drying shrinkage of prismatic light-weight
concrete specimens with different dimensions (49).

In work with shrinkage-compensating expansive cements
water curing is required; otherwise, expansion potential may
be reduced significantly (10). With light-weight
aggregates, the large quantity of water contained within the
porous aggregate particles produces a much better curing
environment for expansive cement, and the expansion can be
as much as 50% greater than that obtained with comparable

normal-weight aggregate concretes.

28

2.5.2 Specific Gravitv

Since light-weight aggregates generally contain pores,
both permeable and impermeable, the term specific gravity
has to be defined carefully, and there are indeed several
types of specific gravity.

The absolute specific gravity is based on the volume of
solid material excluding all pores, and can therefore be
defined as the ratio of the weight of solid material
excluding the pores (7).

If the volume of the solid includes the impermeable
pores but not the capillary ones, the resulting specific
gravity is called the apparent one. The apparent specific
gravity is then the ratio of the weight of the aggregate
after drying in oven at 212 to 230°F (100 to 110°C) for 24
hours, to the weight of water occupying a volume equal to
that of the solid including impermeable pores.

The specific gravity of natural aggregates varies from
2.6 to 2.7. The values for artificial aggregates extend
from considerably below to very much above this range (7).
The specific gravity is difficult to calculate for light-
weight aggregates because of the water absorption
characteristics of the aggregate.

The specific gravity of light-weight aggregates varies
with their size; the finer particles tend to be heavier than
the larger ones. Hence, on a weight basis, the percentage
of finer materials in light-weight aggregates has to be more

than that in ordinary solid aggregate concrete in order to

29

achieve comparable gradations on a volume basis. As far as
the proportioning of mix is concerned, it is the volume
occupied by each size fraction (and not the weight) that
determines the final cement paste content and workability of
the light-weight concrete mix.

The higher-strength.materials typically have lower void
contents. Foamed concretes, with air acting as light-weight
aggregate, have higher void contents than ordinary concrete.
As observed in Figure 2.8, the strength of light-weight
concrete depends on the density of the material. The
density of concrete is itself a function of specific gravity

of aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

mm’
50 10 20 39 49 50
700
4 - .4...
E O .500
I 3 of O
.C
.1 400 “'
g 00 r i?
5
2 .300
.3 ‘
W I
§ 0
g1 » 200
3 1 /
>100
O O
200 400 600 800

Density - kg [m3

Figure 2.8. Strength vs. Density of Light-Weight Concretes
(2)-

 

30

Another term, specific gravity factor, refers to the
ratio of the weight of the aggregate, as introduced to the
mixer, to the effective volume displaced by aggregate. The
weight of aggregate includes any moisture content, absorbed
or free, at the time of placing the aggregate in mixer.
Specific gravity factor is recommended for calculating
adjustments in mix proportions if any variations in the
properties of fresh light-weight aggregate concrete are
observed. The specific gravity factor is not a true
specific gravity; its value incorporates compensation for
the free water of aggregate, but it is used in exactly the
same way as conventional specific gravity to calculate the
volume relationships. Specific gravity factor depends on
initial moisture content of the aggregate as shown in Figure

2.9.

 

8
I

Specific cavity [actor
‘
‘0
o
I

E
X

 

 

1.20 '
0

 

Aggregate momma content m.

Figure 2.9. Specific Gravity Factor vs. Moisture Content of
Light-Weight Aggregates (7).

 

31

2.5.3 Thermal Properties

The coefficient of thermal expansion of light-weight
concrete is the same as that of normal-weight concrete, but
its thermal conductivity is considerably lower because of
the large amount of void space present in light-weight
aggregates. Thermal conductivity depends upon unit weight
as shown in Figure 2.10. The lower thermal conductivity
makes light-weight concrete more’fire-resistant than normal-
weight concrete; light weight concrete also provides a

better protection against fire for the reinforcement.

Um: wevgh: Hotel.
r75 50 7E ICC 125 I C
. i .

 

    

I

u./n’-n-"H

WE'.
[1012019151079

\

5.:

  
 
 

rniuhn [wily IW/m-K)

IIIIVIIY In...

   

\\

Drv
[1% mOIsture)

 

 

 

i r , . .
400 800 I200 1600 2000 NBC
Umt weight 0‘ Cone/eta 'kg’ml)

Figure 2.10. Thermal Conductivity vs. Unit Weight of
Concrete (7).

The thermal transmittance of light-weight concrete is
about 50% that of normal-weight concrete. This greater

thermal insulation is a key factor in many insulating

32

applications of light-weight concrete. It also has important
effects on prestressing applications because with light-

eight aggregates one gets: (a) greater camber when one side,
is exposed to sun; (b) better response to steam curing; and

(c) better fire resistance.

2.6 DURABILITY AND PERMEABILITY

Durability is the ability of concrete to withstand the
exposure conditions for which it is deSigned over the period
of its life time.

An important factor deciding the durability of concrete
is permeability; water permeation into concrete is a '
necessary condition for the initiation of many forms of
_concrete deterioration.

2.6.1 Permeability

The permeability of light-weight aggregate concrete is
reported to be generally lower than that of the normal-
weight concrete (4). The principal reason for low
permeability of light-weight aggregate concrete is the
general absence of microcracks in the aggregate-cement paste
transition zone. According to Holmet (4), the reduced
intensity of interface microcracks in light-weight concrete
is due to the similarity of elastic moduli between light-
weight aggregates and cement mortar.

Some concretes made with light-weight aggregate may
exhibit higher moisture movements than normal-weight

concrete (2,4). In order to prevent early corrosion of

 

33

reinforcement due to moisture movements in such light-weight
aggregate concretes, the depth of cover should be increased,
generally up to twice that of normal-weight concrete. In
exposed situations the British Code CP 110 requires 0.4" (10
mm) additional cover for light-weight concrete when compared
to dense concretes. Alternatively, the use of a rendered
finish or coating of the reinforcement with rich mortar has
been found useful. In the case of concretes made with
clinker aggregate, there is the additional danger of
corrosion due to the presence of sulfur in the clinker and
coating of steel is necessary (2,4).
2.6.; Fregge-Thaw Durability

The freeze-thaw resistance of light-weight concrete is
reported to be similar to that of ordinary concrete. Air
entrainment should be used when the concrete will be exposed
to repeated freezing and thawing. The moisture content of
aggregates can be critical because when aggregates are close
to saturation, the freezing water in aggregate pores will
force water out of aggregate particles into the surrounding
paste. The resulting hydraulic pressure may cause tensile
failure if sufficient entrained air is not present to
accommodate the excess water. In order to avoid this
situation the aggregates should have as low a moisture
content as practical during mixing, or the concrete should
have ample time to dry out before being exposed to freezing

temperatures.

34

Laboratory freezing and thawing test performed on non-
air-entrained and air-entrained concretes made with light-
weight aggregate and normal-weight natural aggregate
concretes. Each aggregate was used in air-dried and soaked
conditions. The concrete were made at two different
strength levels. On the basis of these tests, the following
observation were made:

(a) The amount of intentionally entrained air required for
adequate durability of concretes made with light-weight
aggregate is similar to that for normal-weight concrete;

(b) Concretes incorporating light-weight aggregate can be
made, through air entrainment, as resistant to the effect of
deicer salts as concretes made with normal-weight aggregate;
and

(c) The spread in durability test results among concretes
made with different light-weight aggregates appears to be
greater than might be encountered with normal-weight
aggregate.

The resistance of light-weight concrete to deicer salt
scaling is reported to be similar to normal-weight aggregate
concrete. Salt sealing resistance is improved by air
entrainment, low water-cement ratio, adequate curing, and a
period of drying before service.

2.6.3 Abrasion Resistance

Light-weight aggregates are porous and therefore more

friable than normal-weight rock and minerals. Consequently,

concretes containing light-weight aggregates generally show

35

poor resistance to heavy abrasion. Replacement of light-
weight fine aggregates with natural sand improves the
abrasion resistance.

2.6.4 Marine Durability & Sulfate Resistance

Marine durability of light-weight aggregate is
generally believed to be comparable to that of normal-weight
aggregate concrete.

Prestressed light-weight concrete piles have been in
service in sea water since 1955 (4). Inspection in 1968
showed no deterioration problems. VOlume change problems
have occurred in some other applications, principally with
prestressed light-weight concrete floor and roof slabs.
These problems are similar to those occurring with
prestressed normal-weight concrete, but are aggravated
because of the low modulus of elasticity of light-weight
concrete and its lower dead load-to-live load ratio, greater
shrinkage, and increased thermal insulation.

The presence of excessive iron sulfide in slag may
cause discoloring and durability problems in concrete
products incorporating slag light-weight aggregates. Under
certain conditions sulfide can be converted to sulfate,
which is undesirable from the stand point of sulfate attack
on concrete. British specification limits the contents of
soluble acid 803 and total sulfide sulfur in slag to 0.7 to

2 percent (4).

 

36

2.6.5 Fire Reeistance

Thermal conductivity of light-weight concrete is about
half as much as that of normal-weight concrete and therefore
the fire endurance is considerably better. The fire
resistance of light-weight aggregate concrete is better by
20 to 50% than normal-weight aggregate concrete.

The type of aggregate affects the strength reduction
under fire, as shown in the Figure 2.11. Light-weight
aggregates gradually lose their strength at temperatures
above 1200°F (684°C). Concretes made with carbonate normal-
weight aggregates (e.g. limestone and dolomite) are also
relatively unaffected by temperature rise up to 1200 to .
1300°F (684 to 740°C) at which time they undergo chemical
change and rapidly loose strength. The quartz in silicon
aggregates, such as quartzite, granite, sandstone and
schists, undergoes a phase change at about 800 to 1000°F
(475 to 590°C), which causes an abrupt change in volume and
spalling of the surface.‘

Temperature ('C)

0 200 400 6th 800
I I T

-.. ‘ ~‘/\Sanct-llghtwelgttt

 

_ 100

      

Carbonate

aggregate \\
60 - \
Slllceous /

aggregate

80'-

40‘-

20 "

Butt

Pink or red Gray in\
I‘

I

Compresssive strength (% of original)

 

 

0 I I L AL I T I l
32 400 800 1200 t600
Temperature ('F)

Figure 2.11 Temperature Effects on Compressive Strength
(40).

 

37

2.7 APPLICATIONS

The use of light-weight concrete leads to reductions in
the overall weight of structures, thereby lowering the
concrete and steel requirements for load-carrying purposes.
This is especially important when dead load is the major
portion of total loads applied on the structure, or when the
weight of the member is a factor to be considered for
transportation and erection purposes.

The development of light-weight pumice concrete with a
28-day compressive strength greater than 3500 psi (24
MPa)(25), has considerably increased the use of light-weight
aggregate concrete in residential construction (31). Pumice
light-weight concrete is strong enough for load-bearing
walls, and it weighs only 80 to 90 lb/ft3 (1280-1440
kg/ma). Reduced weight has also been the determining factor
in the selection of light-weight concrete for prestressed
bridge girders, particularly the suspended span in
cantilever-suspended span bridges (27,33).

Another property of light-weight concrete which is of
great potential use is its low submerged weight. This is
directly attributable to low unit weight (9,10).

Prestressed light-weight concrete weighs 75 to 80% of
prestressed normal-weight concrete in air; when submerged in
water, the ratio is only 50%. This makes prestressed light-
weight aggregate concrete an ideal material for barrages,

ships, and dry docks (39,47).

38

Light-weight aggregate concretes incorporating plastic
aggregates are strong and fire-resistant, and combine good
thermal insulating properties with a high strength-to-weight
ratio. The density of plastic light-weight concrete is in
the low range of 20 to so lb/cu.ft. (320 to 800 Kg/m3).
Typical applications are precast roof and wall panels, floor
and roof fills, core material for sandwich panels, and
light-weight blocks (1).

Some of the most common uses of light-weight concrete
are in composites with normal-weight concrete. Different
combinations have been employed successfully. These
combinations are the following: (a) Prestressed light-weight
concrete joist or girders with a cast-in-place normal-weight
composite slab; (b) Prestressed light-weight concrete joists
or girders with a cast-in-place light-weight concrete slab;
and (c) Prestressed normal-weight concrete joists or girders
with a cast-in-place light-weight concrete slab. .All of
these combinations, especially combination (c), require
careful consideration for the differential moduli of
elasticity and shrinkage. Composite construction has been
employed in USSR where cellular (aerated) concrete is joined
with ribs or slabs.

Lower elastic modulus of light-weight concrete permits
better energy absorption and impact resistance, leading to
applications such as in fender pile and navigation light
stands where the lower elastic modulus permits deflections

under impact loading which delay cracking. Prestressed

39

concrete piles as long as 132 ft (40 m) have been
manufactured and installed with light-weight aggregates.

vermiculite and perlite present two recent developments
in light-weight concrete. vermiculite is made from a
lamellar micaceous material, and perlite from silica lava.
Both raw materials contain combined water, so that, upon a
quick heating to the softening temperature, the material
puffs. For both of these aggregates the fine size is
produced more readily than coarse. Neither of them is
suitable for high-strength structural concrete, but they
have found many other uses, as for concrete partitions and
roofs, inSulating walls, and plaster applications (36). ‘

Concrete masonry represents an important application
field for light-weight aggregates. Concrete blocks are made
in a variety of machines that consolidate a relatively dry
concrete mix into the desired size and shape. The blocks
might be hollow or solid, and load-bearing or non-load-
bearing. Examples of light-weight aggregates typically used
for the production of blocks include: cinder, expanded slag,
granulated slag, expanded shales and clays, pumice, scoria,
and vermiculite.

Examples of important buildings made of light-weight
aggregate concrete include (10):
1 - Hilton Hotel at San Antonio, Texas.
2 - Nation Airline hanger in Miami which is a 111 ft (34 m)

cantilever.

40

3 - A Hyperbolic roof 186 ft (56.7 m) x 192 ft (58.5 m) x 3
in (7.5 mm) thick at Colorado.

4 - John Hancock building 100 stories high, with 5 in (125
mm) thick light-weight aggregate concrete floor slabs acting
compositely with steel beams.

5 - Shell plaza which is 714 ft (217.6 m) tall.

2.7.1 Econopic Aspects of Using Light—Weight Concrete

According to ACT 213R-79, the use of light-weight
aggregate concrete in a structure usually leads to lower
overall cost of the structure. This is in spite of the fact
that light-weight concrete will cost more than normal-weight
concrete per cubic yard, partly because cement contents are
usually higher than for normal-weight concrete for a given
strength. The price of light-weight concrete per unit
volume is typically 30% to 40% higher than that of normal-
weight concrete (10).

The reduced weight associated with the use of light-
weight concrete leads to savings in the structural frame
itself, and particularly to savings in foundation. In good
soils, settlements can be reduced, or alternatively the size
of footing can be reduced. In poor soils, the number of
piles may be reduced. In the case of in-between soils it
may be possible to use spread footing or a mat in lie of
piles. Savings in foundation costs have been cited as a
major economic advantage associated with the use of light-

weight concrete.

41

Wilson cites several examples, including the following,
to demonstrate that use of light-weight concrete can result
in lower costs of foundations, reinforcing steel and
construction.

The construction of the light-weight concrete bridge
deck for the San Francisco - Oakland Bay bridge in 1936
resulted in a $3 million saving in steel (4,23). Since then
numerous light-weight concrete decks have been built
throughout the world.

Strength is not a major consideration in floor slabs;
therefore, a large amount of light-weight aggregate is used
to reduce concrete dead weight in floors of high-rise
buildings. An example of this application is the Lake Point
Tower in Chicago, Illinois, which was built in 1968 and is
71 stories high.(4) The Australian Square in Australia,
completed in 1967, is a circular tower with 50 stories 603.7
ft (184m) high and 1394.5 ft (425m) in diameter (4). .A 13
percent saving in construction cost was achieved through the
use of 333,560 ft3 (31,000 m3) of light-weight concrete in
beams, columns and floors above the seventh-floor level.
Shell plaza, Houston, Texas is an all light-weight concrete
structure of 52 stories containing a 229.6 x 170.6 by 8.2 ft
(70 by 52 by 2.5 m) light-weight concrete pad 59.1 ft (18 m)
below grade. If normal-weight concrete had been used, only
a 35-story structure could have been safely designed due to

limited soil bearing capacity.

42

Light-weight aggregate concrete continues to be used
worldwide in the production of precast concrete elements and
prefabricated panels due to the lower handling,
transportation and construction costs (1,4,15).

It has been suggested (48) that with the rise in labor
costs compared to material costs in the construction
industry, the cost of the light-weight aggregate-concrete
slabs in building structures tends to be further reduced
below that of normal-weight concrete fer large-volume
productions (48). The savings in reinforcement area is
also of significance where live loads are small in

comparison with dead loads.

CHAPTER 3
EFFECTIVENESS OF DIFFERENT SYNTHETIC

FIBERS IN NORMAL- AND LIGHT-WEIGHT
AGGREGATE CONCRETE

3.1 INTRODUCTION

Concrete is weak in tension and fails in a brittle
manner under different stress systems. Any technique that
can help increase the tensile strength and ductility of
concrete would be very valuable. Fiber reinforcement is an
effective technique for this purpose. 'Steel fiber
reinforced concrete has been investigated since 1910, when
Porter (17) attempted to increase the tensile strength of
concrete. During the past decade there has been a growing
Iinterest in the use of different fiber types in cement
paste, mortar and normal-weight concrete, but the field of
light-weight fiber reinforced concrete has been largely
neglected.

In this study synthetic fibers (polypropylene and
polyethylene) were used in normal-weight and light-weight
aggregate concrete materials. The main thrust in this part
of the research was to verify if the increase in fiber-to-
matrix modular ratio in light-weight concrete, which has a
lower elastic modulus when compared with normal-weight
concrete, leads to increased effectiveness of synthetic
fibers in light-weight concrete. The effectiveness of
fibers in different concrete matrices was assessed through

flexure and impact tests.

43

 

44

3.2 BACKGROUND

Different fiber types have been found suitable for use
in normal-weight and light-weight concrete. These fibers
include steel, glass, polymeric, carbon and cellulose; they
vary considerably in both cost and effectiveness as cement
and concrete reinforcement. Steel and glass fibers are
generally used in cement-based materials to improve the
flexural strength, ductility, and impact resistance of the
material. Synthetic fibers generally have lower elastic
moduli than concrete, and can not significantly increase the
flexural strength of concrete; they are, however, effective
in increasing the impact resistance and reducing the
intensity of plastic shrinkage cracking.

Attempts to develop light-weight fiber reinforced
concrete with different types of light-weight aggregate have
been reported in the literature. Successful results have
been obtained through the use of foamed slag with steel
fibers (32).

Like normal-weight concrete, the properties of fresh
light-weight aggregate concrete depends on the size and
shape of fibers (7, 28). Since fibers have relatively large
surface areas, they tend to increase water requirements in
concrete; they also exhibit a tendency to interlock or ball.
As a general rule, concrete workability decreases in the
presence of fibers, and it is difficult to measure the
workability of fiber reinforced concrete; however, the Vebe

test provides a rough measure of workability. The results

45

of compaction, Vebe, and slump tests of steel fiber
reinforced light-weight aggregate concrete are shown in
Figure 3.1 (32), which indicates the damage to workability
of light-weight fiber reinforced concrete with increasing

volume fraction of fibers.

can't-crux. 5 arms

v I ISICOIIOSI

fi
5
I

0.900

0.800 l- I t- - I‘D

manna tonne-nu"

t- L".

 

OJOOl- I *-

OJSO1 oi.

PM

 

 

 

 

a; I “W. ‘

1 L L 1 I l
i 01 WV 0.: on to If u to if ‘ 1.0-
m« rtnunuct or (Ounctto car-curt vow-I

Figure 3.1. Effect of Polymer Fiber Reinforcement on the
Workibility of Fresh Light-Weight Concrete (1 mm = 0.04 in.)

The mechanical properties of light-weight aggregate
concrete generally improved with the addition of fibers up
to a limit. These effects would be different for different

types of light—weight aggregate.

 

 

L8 £4
. E
’ S
I!
r If 260'
E a
II
2 5", 25.0-
9 f:
3 1 24.0-
E 2:10-
5
Z
t- 22.0-
9
2
teq- 5 no-
:uo- E 200-
- i 2%
~ :1 120'- U 19.0-
E s
I groo- 180-
I
3 (
g g. g 8.0- 17.0-
‘ a
\L
0 4. 8 6.0- 18.0-
‘“ <
O a.
3 3. 3 4.0L 15.0

 

TENSION

IMPACT

CRUSHING

“00. 0? RUPTURE

 

 

1 L l 1 j
0 0.3 0.4 0.6 08 1.0
sine PERCENTAGE OF CW'ACTEO CONCRETE VOLUME

(a) Polypropylene Fibers

TENSILE STRENGTH lNlmm'l
. .u .-
9—9—94

 

 

1.8 1
t: I 24.0-
“‘ ‘- 23.0-
22 0(- 210-
20.0” 210-
:5. 13 O- 20.0..
g Q
.t‘ ," IGOI- E 19.0-
E s 5
E 3 _ E
- u 14.0 o no.
ul 2 2
C ( w
2 :3 _ g
a 5 .7. 12.0 a no-
3 u
C I g
‘5 4.0 t, 10.0- i 16.0%-
. ( 3
8 2:. L . L
I 3.0 - 8.0 U 110

 

G

TENSION
// E

CRUSHING

lHPACT

MOO, OF RU'TU RE

1 i L I l ___J
0.1 0.2 0.1 0.4 0.5 06

nus pelican-cs or COMMCTED comet?! VOLUME

Figure 3.2. Effects of Fiber Reinforcement on the

Mechanical Properties of Light-Weight Aggregate Concrete
(in/mm = 144 psi) (32).

 

47

The trends observed in Figure 3.2 regarding the effects
of fiber reinforcement on the flexural and tensile strengths
of light-weight aggregate concrete are basically similar to
those observed for normal-weight concrete. The flexural
strength of light-weight aggregate concrete is shown in
Figure 3.2(b) to increase by about 100 percent with the
addition of steel fibers (18, 32). The increase in flexural
strength with fiber reinforcement is morevpronounced than
the corresponding increase in tensile strength. Both
polypropylene and steel fibers are observed in Figure 3.2 to
be highly effective in improving the impact resistance of
light-weight concrete. Similar trends were observed for-the
effects of fiber reinforcement on the compressive (crushing)
strength and modulus of elasticity (E) of light-weight

concrete .

3.3 EXPERIMENTAL PROGRAM

The effectiveness of different synthetic fibers in
normal-weight and light-weight concrete materials were
investigated experimentally. Six concrete mixes of normal-
weight and light-weight concrete were studied (see Table
3.1); in four mixes synthetic fibers (polypropylene and
high-modulus polyethylene) were used at a 0.25% volume
fraction, and the remaining two mixes were without fibers

(plain).

 

48

Table 3.1. The Experimental Program of Light-Weight and
Normal-weight Fiber Reinforced Concrete.

 

 

 

 

 

 

 

Fiber Reinf. Condition
Concrete
Plain 0.25% Poly. 0.25% Polyeth
Light -Weight
Normal-Weight

 

 

 

The basic mix ingredients were Type I Portland cement,
light-weight coarse aggregate, normal-weight aggregate
(coarse and fine), water, superplasticizer (naphtalene
formaldehyde sulfonate-based) only in fibrous mixtures, and
an air-entraining agent. Gradations of the normal-weight
coarse and fine aggregates are shown in Tables 3.2 and 3.3
respectively. The gradation of the light-weight coarse
aggregate is shown in Table 3.4. The light-weight aggregate
used in this investigation (43) was ceramic-based with
specific gravity of about 0.6 (ranging from 0.58 to 0.62).
Macrolite ceramic spheres are low-density spheres (36 lb/cu.
ft.) consisting of multiple minute independent air cell
encircled by an out shell which is very tough. The sphere
has less absorption capacity as compared to other light-
weight aggregate (less than 0.5% by weight) and can function
at a higher temperature. The properties of synthetic fibers
are shown in Table 3.5. The higher modulus polyethylene
fiber used in this study is observed to possess higher
tensile strength and subtantially higher elastic modulus

when compared with polypropylene fibers.

49

Percentage Passing
1°C

L

 

100 ”' "‘“‘““"

 

 

 

0 U l 1 l
O 0.2 0.4 0.6 0.8 1 1.2

-6- ASTM 3:30 -*- ASTM 330 + Grad. or Material

(a) Light-Weight Aggregate

Percentage passing
120

 

 

 

 

 

O 0.2 0.4 0.6 0.8 1 1.2

—“ ASTM 33 -+" ASTM 33 + Gradof Material

(b) Normal-Weight Aggregate

Figure 3.3.--Gradation of Coarse Light-Weight and Normal-
Weight Aggregate ASTM C-330 and C-33.

The fine normal-weight aggregates satisfy gradation
requirements of ASTM C-33; the normal- and light-weight
coarse aggregate gradations both satisfy the ASTM C-33 and
C-330 requirements (maximum aggregate size was 3/4" (19 mm)
both in light-weight and normal-weight coarse aggregate.

Figure 3.3 compares the gradations of the two coarse

aggregates with the ASTM C-33 requirements.

 

 

50

Table 3.2.--Gradation of Normal-Weight Coarse Aggregate
(Specific Gravity = 2.52)

 

 

 

Sieve No. % Passing
3/4" (19 mm) 100
5/8" (15.5 mm) 99
1/2" (12.5 mm) 94
3/8" (9.5 mm) 50
0.265" (6.625 mm) 30
#4 (4.75 Mm) 9
#8 (2.36 [inn 5

 

Table 3.3.--Gradation of Normal-Weight Fine Aggregate
(Specific Gravity 2.54)

 

 

 

Sieve No.‘ % Passing
3/8" (9.5 mm) 100
#4 (4.75 mm) 99
#8 (2.36 mm) 88
#16 (148 .u m) 74
#30 (450 a,m) 48
#50 (300 ‘11n) 22

#100 (lSOum) 9

 

 

 

51

Table 3.4.--Gradation of Light-weight Aggregate

 

 

U.S. SIEVE 3/4" 1/2" 3/8" #4 #8
(19 mm) (12.5 mm) (9.5 mm) (4.75 mm) (2.36 mm)

PASSING 100 98.5 48.5 12.12 2

 

Table 3.5.--Properties of Synthetic Fibers (145 Psi = l Mpa)

 

 

 

 

NAME OF LENGTH SPECIFIC MODULES 0F TENSILE

FIBER (IN) GRAVITY ELASTICITY STRENGTH
K81 K81

POLYETHYLENE 0.75 0.96 I 17 375

 

The variables in the experimental design were matrix
type (normal- and light-weight) fiber type (polypropylene or
polyethylene) and fiber volume fraction (0% and 0.25%). The
coarse aggregate/cement and fine aggregate/coarse aggregate
ratios were kept constant at 2.0 and 1.0 by volume,
respectively, in all mixes. A slump of 3".: .5"

(75 mm i 12.5 mm) was maintained in plain and fiber
reinforced concrete mixtures. The superplasticizer dosage
was adjusted for achieving the required slump. Fibers tend
to damage the fresh mix workability and increase the
required dosage of superplasticizer. An air entraining

agent was used to maintain the air content of all fresh

52

mixtures in the range from 7% to 9% in order to ensure the
frost resistance of the resulting concretes. For this
purpose the dosage rate of the air entraining agent was
adjusted in different mixes.

Table 3.6 represents mix proportions for normal-weight
plain and fibrous concretes.

Table 3.6.--Mix Proportion of Normal weight Concrete

 

 

 

. Air Super
Aggregate Entraining Plasti-
Mix Cement* Water* Coarse* Fine* Agent Cizer
% of cement (Solid)*
by weight
Plain 704(418) 311(185.4) 1454(866) 1469(871) 0.1 ---
Concrete
Polypro- 738(438) 315(187) 1440(869) 1460(855) 0.05 4.38(2.6)
pylene Fiber
Reinforced
Concrete
Polye- 728(432) 323(192) 1466(870) 1478(877) 0.05 3.89(2.3)
thylene
Fiber Rein—
forced
Concrete

 

* lb/yd3 (kg/m3)

Mix proportions of the light-weight plain and fibrous
concrete mixtures are shown in Table 3.7. Normal-weight
fine aggregate (sand) was used with light-weight coarse

aggregates in the light-weight concretes considered in this

investigation.

Table 3.7.--Mix

53

Proportion of Light-Weight Aggregate

 

 

 

Concrete
Air Super
Aggregate Entraining Plasti-
Mix Cement* Water* Coarse* Fine* Agent Cizer
% of cement (Solid)*
by weight
Plain 800(475) 315(187) 404(240) 1617(960) 0.07 ----
Concrete
Poly- 800(475) 320(190) 404(240) 1621(962) 0.045 6.08(4.0)
propylene
Fiber Rein-
forced
Concrete
Polye- 800(475) 326(191) 404(240) 1626(965) 0.049 6.4(4.95)
thylene
Fiber Rein-
forced
Concrete

 

* 1b/yd3(kg/m3)

A conventional concrete mixer was used for

manufacturing all the plain and fibrous concrete mixtures.

The following mixing procedure was adopted:

1. Add the coarse and fine aggregates with half the
into the mixer.

Ivatxar

2. Start the mixer and add cement into it while the
mixer is running.

3. Add the remainder of water with air entraining

agent.

4. Add superplasticizer.

5. Add fibers gradually in order to avoid balling.

6. Continue mixing for three minutes.

54

7. Stop the mixer for two minutes.

8. Start the mixer and run for two minutes.

In the case of plain concrete all the above steps were
taken except that the addition of fibers and super-
plasticizer was eliminated from the process.

The fresh mix was tested for slump (ASTMC 143), Vebe
Time British Standard (BS 1881), air content (ASTMC 231),
and unit weight (ASTMC 138). Samples were made for the
hardened material flexure and impact tests. The hardened
specimens were cast in molds and compacted through external
vibrator. The specimens were demolded after 24 hours,
during which they were covered under wet burlap and plastic
sheet and stored at 74°F (22°C). Thereafter the specimens
were moist-cured at 74°F (22°C) and 100% relative humidity
for five days, and then air dried for 22 days. The size and
number of samples are shown in Table 3.8. Each hardened
material test was replicated ten times (on ten different
specimens) in order to provide data for statistical

analysis.

Table 3.8. Size and Number of Hardened Material Test Specimens.

 

 

 

FLEXURE IMPACT
SIZE 4" X 4" X 14" 6" did 2.5" LONG
(100 mm x 100 mm x 350 mm) (150 x 52mm)

NUMBER PER MIX 10 12

 

 

55

The flexure tests on hardened material were conducted
by four point loading on the 4" x 4“ x 14" (100 x 100 x 350
mm) specimens. Deflections were measured at the center of
the specimen with respect to the loading point (Figure 3.4).
This method of displacement measurement eliminated any error
associated with the rigid body movements of the specimen or

penetration at support or load points into the specimen.

 

 

 

 

 

J?
ff
~11

 

Anglo pace:
(as unm- u pouibh in uudtll)

Figure 3.4. Flexure Test Set-Up

The flexural loading was displacement-controlled with a
quasi-static deflection rate of about 1/1000 times the span
length per minute. These flexure tests will produce a
flexure load deflection curve which can be characterized
through strength (modulus of rupture) and toughness. The

Japanese Concrete Institute specifications (51) were

56

followed for calculating flexural toughness. Toughness was
defined as the area underneath flexural load-deflection
curve up to a deflection equal to the span length divided by
150.

The impact tests were carried out following the
procedure recommended by ACI committee 544 (54). This test
measures the amount of impact energy necessary to start a
visible crack in concrete and then to continue to open that
crack with failure.

The equipment for the impact test (Fig. 3.5) consists
of a standard lO-pound (448N) compaction hammer with 18 in.
(457 mm) drop, a 2.5 in. (635 mm) diameter steel ball, and
positioning fixtures. The impact test is performed by
dropping the hammer repeatedly and recording the number of
blows required to cause the first visible crack at top and

then ultimate failure.

  
  
   

lit-IAI- Hanks—-
lulu—It‘l-
rum-mun.

illllllnk!

Figure 3.5. Impact Test Set-Up

57

3.4 EXPERIMENTAL RESULTS AND DISCUSSIONS

The fresh mix test results are presented in Table 3.9.
The slump and vebe.time test results are indicative of a
reasonable success in this study in achieving comparable
fresh mix workability conditions in all the mixtures. The
air content test results also indicate that all the fresh
mixtures had similar air contents. Fiber reinforcement is
observed in Table 3.9 to have negligible effects on unit
weight. On the average the light-weight concrete mixtures
(with and without fibers) had a unit weight of 66% that of
normal-weight concrete materials.

Table 3.9. Test Results of Fresh Concrete

 

 

UNIT WEIGHT SLUMP VEBE TIME AIR CONTENT
1b(kg) in (mm) (sec.) (%)

 

Normal Weight 146(2296) 3.25" 5—6 sec. 7.5
Polyethylene

Fiber Reinforced

Concrete

Light Weight 95.96(1537) 3.5" 5 sec. 8.25
Polyethylene

Fiber Reinforced

Concrete

Light—Weight 95(2331) 2.75" 4 sec. 7.75
Polypropylene

Fiber Reinforced

Concrete

Light Weight 96(1546) 2.5" 6.5 sec. 8.00
Plain Concrete

Normal Weight 145(2295) 2.75” 4.5 sec. 7.25
Polypropylene

Fiber Reinforced

Concrete

 

58

Table 3.9.--Continued

 

 

UNIT WEIGHT SLUMP VEBE TIME AIR CONTENT

 

lb(kg) in (mm) (sec.) (%)
Normal weight 143(1517) 2.75" 5.5 sec. 7.1
Polypropylene
Fiber Reinforced
Concrete

3.4.1 FLEXQRAL BEHAVIOR

Typical load deflection curve for normal-weight and
light-weight concretes incorporating no fibers and 0.25
percent polypropylene or polyethylene fibers are shown in
Figures 3.6a and 3.6b. The flexural strength test results
are given in Tables 3.10 and 3.11 for normal-weight and
light-weight concrete materials, respectively. Tables 3.12
and 3.13 present the flexural toughness for normal-weight
and light-weight cements, respectively.

Figures 3.6a and 3.6b are indicative of the positive
effects of fibers on the post-peak ductility and toughness
of concrete materials in flexure. The initial stiffness
does not seem to be influenced significantly by the fiber
reinforcement, but flexural strength tends to increase in
the presence of synthetic fibers. Polyethylene fibers seem
to be more effective than polypropylene fibers in enhancing
the flexural behavior of normal-weight and light—weight

concrete .

F lexural load
4

S9

 

 

 

pptrc
" ' potrc

\ —--"'_ --_--— —————————————
..............

 

 

 

 

 

 

O l l g
0 0.02 0.04 0.00 0.08
Deflchon
(a) Normal-Weight Concrete
Flexural Load
4
3..
__ Dc
2- ppfrc
"‘ petc
é
1-{1' ________
O l J l l l J l
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
Denecfion

(b) Light-Weight Concrete

Figure 3.6.--Flexura1 Load Deflection Curve

 

60

Table 3.10 Flexural Strength Test Result for Normal-Weight
Aggregate Concrete in Psi (145 Psi = 1 Mpa).

 

 

Polypropylene Polyethylene
Plain Fiber Reinforced Fiber Reinforced
Concrete Concrete Concrete
606.7 688 768
571 771 593
532.4 660 693
444.4 646 661
454 688 782
481 660 675
544.2 644 676
604.9 748 . 700
666.8 732 782
545.8
Sample Mean - 545 Psi 695 Psi 696 Psi
Standard
Deviation - 95 Psi 44 Psi 56 Psi
Co-efficient of
Variation - 17.16% 6.95% 8.18%

95% Confidence Interval

Lower Limit 456 662 652
Upper Limit 622 727 739

 

 

61

Table 3.11.--F1exura1 Strength Test Results for Light-Weight
Aggregate Concrete in Psi (145 Psi = 1 Mpa)

 

 

Polypropylene Polyethylene
Plain Fiber Reinforced Fiber Reinforced
Concrete Concrete Concrete
142.00 303.00 248.00
169.00 194.00 332.00
130.00 170.00 278.00
278.00 204.00 442.00
280.00 330.00 447.00
305.60 351.80 262.60
254.40 303.80 317.00
254.60 207.50 261.90
172.10 380.50 417.10
215.70 318.80 419.20
Sample Mean - 227 Psi 282 Psi 342 Psi
Standard
Deviation - 62 Psi 71.24 Psi 84 Psi
Co-efficient of
Variation - 27.6% 25% 24.37%

95% Confidence Interval

Lower Limit 175 231 284
Upper Limit 285 333 400

 

 

62

Table 3.12.--Flexural Toughness Test Results for Normal-Weight
Aggregate Concrete in K-in. (1 K-in - 113 N.M.)

 

 

 

Polypropylene Polyethylene

Plain Fiber Reinforced Fiber Reinforced
Concrete Concrete Concrete
0.0047 0.024 0.100
0.0040 0.021 0.059
0.0039 0.041 0.047
0.0051 0.067 0.056
0.0071 0.065 0.071
0.0076 0.051 0.073
0.0049 0.076 0.095
0.0035 0.067 0.076
0.0048 0.053 0.067
Standard

Deviation - 0.0014 0.019 0.017
95% Confidence Interval

Lower Limit 0.004 0.038 0.059
Upper Limit 0.006 0.636 0.083

 

Table 3.13.--Flexura1 Toughness Test of Light-Weight Concrete in

 

 

 

K—in (1 K-in 113 N.M.).
Polypropylene Polyethylene

Plain Fiber Reinforced Fiber Reinforced
Concrete Concrete Concrete
0.0026 0.029 0.041
0.0014 0.033 0.063
0.0051 0.07 0.050
0.0040 0.06 0.049
0.003 0.03 0.045
0.0035 0.02 0.055
0.004 0.039 0.051
0.001 0.034 0.048
0.0033 0.031 0.062
Mean - 0.0031 0.038 0.051
Standard

Deviation - 0.0019 0.018 0.0189
95% Confidence Interval
Lower Limit 0.0019 0.036 0.038
Upper Limit 0.0043 0.06 0.063

 

Chi-Square goodness-of-fit test confirmed the normality
of sample distributions for flexural strength test results
at 5% level of significance, but showed poor fit of the
flexural toughness test data to normal distribution. The
relatively small sample size and the relatively large
variations in toughness test data might be responsible for
the deviation of replicated toughness test results from
normal distribution.

The flexural strength and toughness test results are

shown in Figures 3.7a and 3.7b, respectively.

64

FLEXURAL STRENGTH (PSI)

  
    
     

mean . Sid. Dev

Mean - Sld Dev,

  

 

............. N\

NORMAL-WEIGHT OONC LIGHT—WEIGHT CONC

PC PPFRC (:1 PEFRC

(a) Flexural Strength

Teug mess
1
Mean 08m. Dav.

Mean

Man —scd. Dav

 

   

 

Normal—weight Conc. Light-Weight 0000.

pc PPFRC {:3 PEFRC

(b) Toughness

Figure 3.7 Effect of Synthetic Fiber Reinforcement at 0.25%
Volume Fraction on the Flexural Strength and Toughness of

Normal—Weight and Light-Weight Concrete Materials (PC —

Plain Concrete; PPFRC = Polypropylene Fiber Reinforced
Concrete; and PEFRC =

Polyethylene Fiber Reinforced
Concrete).

 

65

On the average, the flexural strength of normal-weight
concrete increased by 26% and 27% in the presence of 0.25%
volume fraction of polypropylene and polyethylene fibers,
respectively, and the corresponding percentage increases in
the case of light-weight concrete were 24% and 52%,
respectively. The average increases in flexural toughness
of normal-weight concrete in the presence of 0.25% volume
fraction of polypropylene and polyethylene fibers were
1,000% and 1,300%, respectively, and the corresponding
increases in the case of light-weight concrete.were 1,300%
and l,500%, respectively.

Simple observation of the flexure test data together
with analyses of variance of the results at 5% level
significance led to the following conclusion regarding the
effectiveness of polypropylene and polyethylene fibers in
enhancing the flexural properties of normal—weight and
light-weight concrete materials:

1. Both the synthetic fibers increase the flexural

strength and toughness of normal-weight and light-

weight ccncrete materials;

2. Polyethylene fibers are more effective than
polypropylene fibers in enhancing the flexural
strength and toughness of light-weight concrete;

and

 

66

3. Considering the variations in flexural strength
and toughness test results, in the flexural
behavior of normal-weight concrete with
polyethylene fiber reinforcement are not
significantly superior to those obtained through
polypropylene fiber reinforcement; this is in
spite of the fact that polyethylene fibers produce
higher average values of flexural toughness than

polypropylene fibers;

4. There is a tendency in the variations in fluxural
strength test results to be reduced with fiber

reinforcement; and

5. The percentage increase in flexural strength with
polypropylene fiber reinforcement is comparable in
light-weight and normal-weight concrete.
Polyethylene fibers, however, tend to be more
effective in light-weight concrete than in normal-
weight concrete (as far as the percentage increase
in flexural strength is concerned). Both these
conclusions are valid from the generated test

data, at 5% level of significance.

The ratio of fiber modulus of elasticity to that of

matrix is suggested to be one of the factors deciding the

 

67

effectiveness of fibers in concrete materials. The higher
this modular ratio the more effective it would be. With
this hypothesis one may conclude that the same fibers can be
more effective in light-weight concrete than in normal-
weight concrete because of the lower modulus of elasticity
of light-weight concrete. In order to verify the validity
of this hypothesis the relationship between the flexural
strength ratio (of fibrous to plain concrete materials) were
investigated (see Figure 3.8) for the four fiber reinforced
concrete materials considered in this project (polypropylene
and polyethylene fiber reinforced concrete, with normal-
weight and light-weight aggregates). The relationship be-
tween flexural strength ratio and modular ratio in Figure
3.8 is not consistent; this figure does not provide evidence
of the dependence of flexural strength ratio on modular
ratio, suggesting that the effectiveness of fibers in
increasing flexural strength is not dependent on modular
ratio for the specific conditions of this investigation.

The correlation coefficient of flexural strength ratio
versus modular ratio is 0.334, confirming the weak

dependency of flexural strength ratio on modular ratio.

 

 

68

3 FLEX. STR. FRO/PC

 

omoooo

wasn‘t

 

 

o L J L l l

4 6 8 1o 12
MOD. OF ELAS. FRO/PC

‘ NWPPFRC + LWPPFRC *‘ NWPEFRC '3 WPEFRC

 

Figure 3.8 Relationship Between Flexural Strength Ratio and
Modular Ratio.

The relationship between toughness ratio (of fibrous to
plain concrete) and modular ratio in Figure 3.9 also does
not provide evidence for the dependence of flexural
toughness ratio on modular ratio, but the effectiveness of
fibers in increasing flexural toughness is not dependent on
modular ratio for the specific condition of this
investigation. The correlation coefficient of flexural
toughness ratio versus modular ratio is 0.358, conforming

the weak dependency of flexural toughness ratio on modular

ratio.

 

69

O TOUGHNESS FRO/PC

 

40"

307

 

 

 

20-.* o
.1 g E

10di+ at a
2+ l

O . T r

0 5 I10

Mod. of Elastcity of FR/PC
' NWPPFRG 4' LWPPFRC 4" NWPEFRC ‘3 LWPEFRC

Figure 3.9.-~Re1ationship Between Flexural Toughness Ratio
and Modular Ratio

3.4.2 Impact Strength

Tables 3.14 and 3.15 present the impact strength test
results at first crack and failure, respectively, for
normal-weight concrete, and Tables 3.16 and 3.17 present the
impact test results at first crack and failure,

respectively, for light-weight concrete.

70

Table 3.14 Impact Test Results (No. of Blows to First
Crack) for Normal-Weight Concrete.

 

 

Polypropylene Polyethylene
Plain Fiber Reinforced Fiber Reinforced
Concrete Concrete Concrete
11 23 42
9 - 26 21
7 11 24
13 14 42
24 20 20
10 13 24
20 11 50
33 32 18
17 12 135
23 22 25
Mean 16 18 40
Std Dev 8 8.5 33
Coeffi- 50% 46% 46%
cient of
Variation

95% Confidence Interval
Lower Limit - 1]. 7 20

Upper Limit - 20 29 60

 

 

 

71

Table 3.15.--Impact Test Results (No. of Blows to Failure)
for Normal-weight Concrete.

 

 

Polypropylene Polyethylene
Plain Fiber Reinforced Fiber Reinforced

Concrete Concrete Concrete

13 58 81

11 51 51

8 30 74

14 38 81

25 46 64

12 40 50

23 45 38

35 47 155

18 55 45

24 38 75
Mean 17 46 71
Standard -
Deviation 8 9 33
Coeff— 45% 20% 46%
icient of
Variation
95% Confidence Interval
Lower Limit - 12 39 51

Upper Limit - 21 46 91

 

 

72

Table 3.16.--Impact Test Results (No. of Blows to First

Crack) for Light-Weight Concrete.

 

 

Polypropylene Polyethylene
Plain Fiber Reinforced Fiber Reinforced

Concrete Concrete Concrete

1 6 8

2 6 13

3 10 10

2 6 12

7 7 6

7 6 12

5 5 9

4 9 11

2 4 6

4 4 14
Mean 4 6 10
Standard
Deviation 2 2.1 2.8
Coeff— 57% 34% 29%
icient of
Variation
95% Confidence Interval
Lower Limit 2.2 4.6 7.7
Upper Limit 5.2 7.6 11.7

 

73

Table 3.17 Impact Test Results (No. of Blows to Failure) of
Light-weight Concrete.

 

 

 

Polypropylene Polyethylene
Plain Fiber Reinforced Fiber Reinforced

Concrete Concrete Concrete
3 25 43
5 20 64
6 31 45
6 34 61
9 25 43
10 31 27
9 46 31
6 38 42
6 40 44
5 35 64

Mean 7 32 47

Standard 2 . 2 7 . 7 13

Deviation

Coeff- 33% I 24% 28%

icient of

Variation

95% Confidence Interval

Lower Limit - 4 27 37

Upper Limit - 8 38 55

 

Chi-square goodness-of-fit tests at 5% level of
significance showed poor fit of the impact strength test
data to normal distribution. The relatively small sample
size and the relatively large variations in impact strength
test results might be responsible for the deviation of

replicated impact test data from normal distribution.

74

The first crack and ultimate impact strength test
results for normal-weight and light-weight concrete

materials are shown in Figures 3.10a and 3.10b,

 

  
  

   

 

 

respectively.
1151 Crack Impact Resistance(# Of Blows)

100 —- ~~

60 ~ 1 e— — ~ -----

60 1. - 1 I- - ,,,,M9.a_0 1,,.Std-...Dev.-.,,.._._.,., . -
40 ~

2° ".1...—31576;.'""“" ‘ ‘ ' '

' ' ’ 7 ‘7/
0 mm%

Normal Weight conc nght-Welgm conc

PC 1:] PPFRC PEFRC

(3)

Impact Strength at First Crack

Impact Reslstance (# 01 Blows) to Fallur
Mean r 510 Dev

 

 

Normal-Weight conc Light—Weight Cone

 

2"‘Itii‘Z'IPC EPPRFC -PEFRC

(b) Impact Strength at Failure

Figure 3.10.—-Effect of Synthetic Fiber Reinforcement at
0.25% Volume Fraction on the Impact Strength at First Crack

and Failure of Normal-Weight and Light-Weight Concrete
Materials.

75

The average first-crack impact resistance of normal-
weight concrete is improved by 12% and 150% through
polypropylene and polyethylene fiber reinforcement,
respectively. The corresponding improvements in the case of
light-weight concrete are 50% and 180%, respectively, noting
that plain light-weight concrete has a relatively low impact
resistance. The ultimate impact strength of normal-weight
concrete is improved by 170% and 310% through polypropylene
and polyethylene fiber reinforcement, respectively. The
corresponding improvements in ultimate strength of light-
weight concrete are 357% and 570%, respectively.
Reinforcement with synthetic fibers leads to more
significant improvements in the ultimate impact resistance
of concrete materials than in the first crack impact
resistance.

It should be noted that on the average, the first-crack
impact resistance of light-weight concrete materials (plain
or fibrous) is about 25% to 35% of the first-crack impact
resistance of corresponding normal-weight concrete
materials. In the case of ultimate impact resistance, the
value for light-weight concrete is about 40% to 70% that of
normal-weight concrete.

Simple observations of impact test results together
with factorial analyses of variance of the data at 5% level

of significance led to the following conclusions regarding

 

7-6

the effectiveness of polypropylene and polyethylene fibers

in enhancing the impact properties of normal-weight and

light-weight concrete materials.

1.

Both synthetic fibers increases the first-crack
and ultimate impact strength of normal-weight and
light-weight concrete materials.

Polyethylene fibers are more effective than
polypropylene fibers in enhancing the impact
strength (1st crack and ultimate) of lightaweight
concrete and normal-weight concrete.

Both polypropylene and polyethylene fibers are
more effective, at 5% level of significance, in
increasing the ultimate impact assistance of
light-weight concrete when compared with normal-
weight concrete.

The lightaweight concrete materials considered in
this investigation (plain or fiber reinforced)
possess lower impact resistance when compared with
the corresponding normal-weight concrete materials

(at 5% significance level).

The relationship between the ratios of fibrous concrete

to plain concrete ultimate impact strengths and the modular

ratio (of fibers to matrix) is presented in Figure 3.11.

The modular ratio by itself does not seem in this figure to

truly represent the degree of effectiveness of specific

fiber types in a concrete matrix for the conditions of this

investigation. The correlation coefficient for impact

 

 

77

strength ratio versus modular ratio was 0.475,
the weak dependence of impact strength ratio 0

ratio.

GIMPACT STRENGTH OF FRO/PC

 

5L+ ' 0

++ 9’1” H
I. . ’. I .
m

 

O—b

MODULUS OF ELAST.FIBER/PC
' NWPPRFC + LWPPRFC ‘ NWPEFRC ° LWPEFRC

Figure 3.11.--Relationship Between Impact Stre
Modular Ratio.

3.5 SUMMARY AND CONCLUSIONS

The effectiveness of different synthetic

normalsweight and lightaweight concrete materi
investigated experimentally. Six concrete mix
weight and lightaweight concrete were studied;
synthetic fibers (polypropylene and polyethyle
at 0.25% volume fraction, and the remaining tw
without fibers (plain). The lightdweight conc
incorporated a ceramic-based light-weight coar
with normal-weight fine aggregate. The normal
lightaweight coarse aggregates had similar gra
satisfying the ASTM.C-33 and C-330 requirement

proportions of coarse aggregate, fine aggregat

 

 

 

78

were kept constant in all mixtures. The dosage of
superplasticizer was adjusted to give a constant slump in
different mixes. The air content of different mixtures was
maintained at a desirable level for frost resistance through
the use of an air entraining agent at required dosages. The
high-modulus polyethylene fiber used in this investigation
had a higher tensile strength and a substantially higher
modulus of elasticity when compared with the polypropylene
fiber.

The fresh concrete mixtures were tested for slump, vebe
time, air content, and unit weight. The hardened concrete
materials were tested for flexural behavior and impact
resistance. The hardened material test data were replicated
in order to allow for powerful analyses of variance of the
test data. The following conclusions, derived based on the

generated test data, are valid at 5% level of significance:

1. Fiber reinforcement has negligible effects on the
unit weight of light-weight and normal-weight
concrete materials. In the specific mixtures of
this study, the light—weight concrete materials
had unit weight of about 66% those of normal
weight concrete (with and without fibers).

2. Both synthetic fibers increased the flexural

strength and toughness of normal-weight and light-

weight concrete materials.

 

79

Considering the variations in flexural strength
and toughness test results, the improvement in
flexural behavior of normal-weight concrete with
polyethylene fiber reinforcement were not
significantly superior to those obtained through
polypropylene fiber reinforcement; this is in
spite of the fact that polyethylene fibers
produced higher average values of flexural
toughness than polypropylene fibers.

There is a tendency in variations in the flexural
strength test results to be reduced with fiber
reinforcement.

.As far as the percentage increase in flexural
strength is concerned, polypropylene fibers
produced comparable results in light-weight and
normal-weight concretes at 5% level of
significance. Polyethylene fibers are more
effective in light-weight concrete than in normal-
weight concrete.

No consistent relationship was observed in this
study between the improvements in flexural
properties with fiber reinforcement and the ratio
of fiber-to-matrix modular elasticity.

Both synthetic fibers increased the first crack
and ultimate impact strength of normal-weight and

light-weight concrete materials.

 

10.

11.

80

Polyethylene fibers were more effective than
polypropylene fibers in enhancing the impact
resistance of light-weight and normal-weight
concrete materials.

Both polypropylene and polyethylene fibers were
more effective at 5% level of significance in
increasing the impact resistance of light-weight
concrete when compared with normal-weight
concrete.

No consistent relationship was observed between
the ultimate impact strength ratio (of fibrous to
plain concrete) and the modular ratio of fibers to
matrix.

The light-weight concrete materials considered in
this investigation (plain or fibrous) possessed
lower flexural strength, toughness, and impact
resistance when compared with the corresponding
normal-weight concrete materials (at 5% level of
significance); an exception to this rule was
observed for the flexural toughness of light-
weight polypropylene fiber reinforced concrete
‘which, at 5% level of significance, was comparable
to that of normal-weight polypropylene fiber

reinforced concrete.

 

CHAPTER 4

PERMEABILITY OF NORMAL- AND LIGHT-
WEIGHT CONCRETE: EFFECT OF PRELOADING

4.1 INTRODUCTION

Permeability is one of the important factors effecting
the durability of concrete. It controls the rate of entry
of moisture that may contain aggressive chemicals
responsible for the disintegration of concrete.
Permeability of concrete depends on the volume and size of
the interconnected capillary pores in cement paste, the
:microcracks that are present at the aggregate paste
interfaces, and possibly the pore system characteristics of
aggregate.

The interface microcracks seem to play an important
role in establishing the interconnection between capillary
cavities and increasing the permeability of concrete. They
appear due to the differential shrinkage and thermal strains
between the cement paste and aggregates, and can grow under
externally applied loads. The interface microcracks are
larger in width than most capillary pores present in cement
paste .

Light-weight aggregates are typically more porous than
IIormal-weight aggregates; they, however, may produce more
<iesirable interface conditions with reduced microcracking.
flihese two effects of light-weight aggregates tend to

lirlfluence permeability in Opposite directions.

81

IIIIIIII-i___

82

4.2. OBQECTIVE AND SCOPE

The nature of the interface microcrack system is
expected to be different in normal-weight and light-weight
concrete materials. Light-weight aggregates with lower
elastic modulus are expected to be more compatible with
cement paste than normal-weight aggregates. Less interface
cracks are expected to occur in light-weight concrete (4,9).
Light-weight aggregates are, on the other hand, generally
:more porous and thus more susceptible to cracking under
stress when compared with normal-weight aggregates. These
differences are expected to lead to different permeability
characteristics in light-weight concrete when compared with
normal-weight concrete. One objective of this investigation
is to compare the permeability of light-weight and normal-
‘weight concrete materials.

Considering the significance of interface microcracks
in deciding the permeability of concrete materials, the
increase in microcrack intensity under increasing load is
expected to influence the permeability of concrete
:materials. The nature of microcrack propagation is
different in light—weight concrete when compared with
Imormal-weight concrete. Another objective of this
:investigation is to assess and compare the effects of
lLoading on the permeability of normalaweight and light-

Vweight concrete materials.

83

This study has been conducted with specific types of
light-weight and normal-weight aggregates, and with certain
concrete mix proportions. The results are strictly
applicable to these conditions.

The results are strictly applicable to these
conditions. Within these conditions, however, replicated
tests were conducted and the results were analyzed
statistically in order to produce reliable conclusions with

specified levels of confidence.

 

4.3 MEASUREMENT OF PERMEABILITY _

Concrete permeability can be measured by different »
methods, including hydraulic head and rapid chloride
permeability test techniques. In the hydraulic head method
the concrete specimen is placed between two chambers having
different pressure of fluid flow from the high-pressure
chamber through the specimen to the lower chamber.
Permeability in this method is usually expressed in ft/sec
(m/sec) of fluid moving through the specimen (60). In the
rapid chloride permeability test method a positive
correlation has been shown between chloride ion penetration
into concrete after 90 days of ponding and the electrical
charge passed through the sample of concrete. An example of

this correlation is shown in Figure 4.1.(58,59,60)

 

84

Luther
2.0 "' /

    
    

/,/ 952 CONFIDENCE LIMITS

 

Tolol Integral Chlorlde ,90'd0y pondlnq

1—

0 IO 20 .30 4O 50 60
Charge passed ' CoulombsnhousandS' Rapid lest 'Slobs

Figure 4.1.--Relationship Between Charges Passed During
Rapid Chloride Test and Total Chloride Values in 90-Day
Ponding Test.(59)

Basically, lower passed charge measurements were
associated with lower permeability. Ultimately, a version
of this passed electrical charge method was standardized as
AASHTO T277-83 (Rapid Chloride Permeability Test). .Although
the AASHTO T277-83 method measures a specific passed charge
number, the result is reported as lying within one of the
following ranges or "classes": High, Moderate, Low, Very

Low and Negligible. Table 4.1 represents the rapid chloride

permeability passed charge levels (60).

 

85

Figure 4.1.--AASHTO T277 Test Classes of Chloride
Permeability from Reference 60.

 

 

CHARGE PASSED CHLORIDE TYPICAL
COULOMBS PERMEABILITY OF
>4000 HIGH High Water Cement

Ratio (“0.6)
Portland Cement
Concrete

2000 - 4000 MODERATE Moderate Water
Cement Ratio (0.4-
0.5) Portland Cement
Concrete.

1000 - 2000 LOW Low‘Water Cement
Ratio (0.4) Portland
Cement Concrete

100 - 1000 VERY LOW Latex Modified
Concrete
<100 NEGLIGIBLE Polymer Impregnated
Concrete; Polymer
Cement

4.4 EXPERIMENTAL PROGRAM

The effects of two variables on the permeability of
concrete were studied experimentally. The variables were
aggregate type (normal-weight versus light-weight), and the
level of preloading (unloaded verses loaded in compression
to 40% of compressive strength).

The basic concrete mix ingredients were Type I Portland
Cement, light-weight and normal-weight fine sand, water, and
an air entraining agent.

Gradation of light-weight and normal-weight aggregates
are shown in Tables 3.2 and 3.4. The normal-weight and

light-weight coarse aggregates both had a maximum particle

86

size 1-3/4" (19 mm) and satisfied the ASTM C-330 and ASTM C-
33 requirements. The normal-weight fine aggregates also
satisfied ASTM C-33 gradation requirement. It is worth
emphasizing that the aggregate gradations (by volume) in
normal-weight and light-weight concrete materials were
comparable.

.A slump of 3" i .5" (75 mm i 12.5mm) was maintained in
both light-weight and normal-weight concrete materials,
through adjustment of water content. An air entraining
agent was used to maintain the air content of fresh mixture
in the range of 7 to 9% in order to ensure the frost
resistance of concrete materials.

The mix proportions and fresh.mix properties for

normal-weight aggregate concrete is shown below:

1bzxd3 (kgzm3)
Cement = 704 (433.75)

Normal Weight
Coarse Aggregate

1461 (867.5)

Normal Weight
Fine Aggregate

1469 (871.5)

Water = (314) (185)

Air Entraining = 1% by weight of cement.
Agent

Slump = 3 inches

Air Content = 8%

Unit Weight 146 (2340)

 

87

The Mix Proportion for Light-Weight Concrete was as

follows:

lblyd3 (kgzm3
Cement = 808 (475)
Light Weight
Coarse Aggregate = 404 (240)
Normal-Weight = 1600 (950)
Fine Aggregate
Water = 318 (187)
Air Entraining = 1% by weight of cement.
Agent .
Slump = 2.75 inch
Air Content = 8.9%
Unit Weight = 95 (1522)

A conventional rotary drum concrete mixer was used for
the production of normal-weight and light-weight concrete
materials. The procedure adapted for producing the

materials is described below:

1. Add the coarse and the fine aggregate with half
the water into the mixer;

2. Start the mixer and add cement as the mixer is
running.

3. Add remaining water with air entraining agent.

4. Continue mixing of concrete for three minutes.

5. Stop the mixer for two minutes

6. Start the mixer and run for two minutes.

From.each mix (normal-weight and light-weight) twenty
specimens (10 for unloaded and 10 for loaded permeability
tests) were made. The specimens were prepared by placing

concrete inside molds and compacting it through external

88

vibration. The sizes of cylindrical samples for loaded and
unloaded permeability tests were 4 in. (100 mm) diameter by
8 in. (200 mm) long. The specimens were moist-cured for
five days (one day inside the moulds, and four days after
demoulding), and were then air-dried until the test usage of

28 days.

4.5 TEST PRQCEQURE§

The fresh mix workability was assessed by the slump
test (ASTM C-143). The unit weight and air content tests on
fresh mix were conducted according to ASTM C-138. The
specimens used for the preloaded permeability tests were
subjected to a compressive stress equal to 40% of the
compressive strength of concrete. The compressive strength
of normal-weight concrete materials, obtained through the
performance of 10 replicated compression tests on 4 in. (100
mm) diameter by‘8 in. (200 mm) high cylindrical specimens
were 4850 psi (33 Mpa) and 2280 psi (15.5 Mpa) for normal-
weight and light-weight concrete materials respectively.

The corresponding co-efficient of variations were 7% and 18%
respectively.

The test procedures outlined in AASHTO 277-83(59) were
used to conduct chloride permeability tests on normal-weight
and light-weight concrete materials. The cylindrical
specimens to be used for this standard test are 4 in. (100
mm) in diameter and 2 in. (50 mm) high; they were cut from

the mid-height of the cylindrical samples (unloaded or

89

loaded) prepared from the mixtures considered in this

investigation.

 

 

Figure 4.2.--Vacuum Saturation Apparatus

The equipment for the chloride permeability test
consists of a vacuum saturation apparatus (Fig. 4.2),
separatory funnel (500 m1 capacity), beaker, vacuum
desiccator, vacuum pump, voltage application apparatus and
cell. Test procedure is as follows: (1) Brush the side of
the specimen (4 in. diameter, 2 in. long) with rapid setting
epoxy; (2) Place specimen in a beaker (1000 ml), and
transfer the beaker to the vacuum desiccator; (3) Seal
desiccator and attach it to a vacuum pump; (4) Start vacuum
pump and maintain vacuum in the desiccator for three hours;
(5) Drain water from the separatory funnel (500 ml) through
the stopcock into the beaker after three hours of running

the pump, submerge the specimen inside the beaker in water;

90

(6) Allow the pump to run one hour after submerging the
specimen in water; (7) Stop the pump and soak specimen under
water for (18:1) hours.

The specimen is then removed from water and sealed
temporarily with silicon rubber sealant to the cell (Figure

4.3).

 

Figure 4.3.--Specimen for Testing.

The left side of the cell containing the top surface of
the specimen is filled with a 3% solution (by weight) of
NaCl, and the right side of the cell is filled with a 0.3N
NaCl solution. The cell is then attached to post as shown

in Figure 4.4.

91

3f own ovu

 

IOOV ES. —-- "mum"
—- Hookup WI"

 

?

 

rum, ai 03°" ova
dc

 

200 «M £5.
'00 IM Shunt

D I?

J
1 ‘4

T0 3.0". NOCI To 0.3N NOON

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4.4. Electrical Block Diagram

A 60 DC voltage is applied across the specimen for 6
hours. The total cumulative charge passed through the

specimen during this 6 hours is noted.

4.6 EXPERIMENTAL RESULTS AND DISCUSSIONS

The raw test data for the chloride permeability tests
performed on normal-weight and light-weight aggregate
concrete materials are shown in Table 4.2.

Chi-Square goodness-of-fit tests showed poor fit of the
test data to normal distribution. This deviation may be due
to the relatively small sample size and relatively large
variations in test results. The permeability test results
for light-weight and normal-weight concretes are shown in
Figure 4.5. Simple observations together with one-way and
two-way analyses of variance of test results (at 5% level of

significance) indicate that:

92

     

  

        

 

77777777

Figure 4.5. Permeability Test Results for Light-Weight and

Normal—Weight Concrete.

93

Table 4.2.--Permeabi1ity Test Results

 

Normal weight Light-weight

 

 

Concrete Concrete
Unloaded Loaded Unloaded Loaded
Coulomb Coulomb Coulomb Coulomb
4,600 4,977 6,442 6,378
2 4,600 5,799 6,595 6,615
3 1,203 2,349 4,135 5,380
4 2,080 2,349 1,625 2,122
5 2,840 2,173 2,778 3,335
6 3,450 7,195 5,798 6,627
7 1,880 2,780 4,136 4,326
1,767 2,211 3,388 3,510
9 4,193 4,649 6,442 6,328
10 3,880 4,856 3,380 3,910
Mean 3,040 3,933 4,471 4,853
Std Dev. 1,251 1,794 1,751 1,624
Co-Efficient 41% 45% 39% 31%
of Variance
95% Confidence
level
Lower Limit 2,144 2,549 3,218 3,690
Upper Limit 3,935 5,217 5,724 5,015

 

weight concrete is 47% more than that of the

The average chloride permeability of the unload light-

corresponding normal-weight concrete; the variations in

chloride permeability test results are, however, so

94

large (about 40% Co-efficient of variation) that a one
way analysis of variance of test results, given the
small sample size (10 specimens for each condition),
can hardly confirm the difference in permeability of
light-weight and normal-weight concrete materials at 5%

level of significance.

Loading up to 40% of compressive strength increases the
average permeability of normal-weight and light-weight
concrete by 29% and 8%, respectively. Pre-loading
seems to increase the permeability of normal-weight
concrete more than light-weight concrete, possibly‘
because the aggregate-paste interfaces in normal-weight
concrete, when compared with light-weight concrete, are
more susceptible to microcracking. Considering the
large variations in test results, one way analysis of
variance showed no major difference in the permeability
of loaded and unloaded concrete materials for the
relatively small sample size of 10 for each condition

at 10% level of significance.

Two-way analysis of variance also showed no effects (at
10% level of significance) of preloading on the
chloride permeability of normal-weight and light-weight

concrete materials.

95

While one-way analysis of variance at 10% level of
significance showed higher permeability in unloaded
light-weight concrete when compared with unloaded
normal-weight concrete, since loading increased the
average permeability of normal-weight concrete more
than that of light-weight concrete, no difference could
be detected between the permeability of preloaded
normal-weight and light-weight concrete materials
through one-way analysis of variance at 10% level of
significance. This observation was made (considering
the relatively large variation in test results) in
spite of the fact that average permeability of
preloaded light-weight concrete was 23% higher than

that of preloaded normal-weight concrete.

Considering the relatively large variations in
permeability test results (when compared with the
effects of preloading in light-weight versus normal-
weight concrete), the sample size chosen in this
investigation (ten permeability test specimens for each
condition) did not seem to be sufficient for powerful
analyses of variance of test results. The power of
one-way analysis of variance conducted in this
investigation typically was less than 0.5, indicating
that there is about a 50% chance of accepting a false
hypothesis for the sample sizes used in this

investigation.

96

4.7 SUMMARY AND CONCLUSION

The differences between permeability characteristics of
normal-weight and light-weight concrete materials were
investigated and the effects of compressive loading (up to
40% of compressive strength) on the permeability of normal-
weight and light-weight concrete materials were studied.
The light-weight and normal-weight concrete had comparable
mix proportions (by volume) and performed similarly in the
fresh state as far as slump and air content were concerned
(the dosage of air entraining agent was adjusted for
achieving comparable air contents). The light-weight coarse
aggregate used in this investigation was ceramic-based, with
specific gravity 0.58 and relatively low water absorption.
The normal-weight coarse aggregates (crushed limestone) had
a gradation similar to that of light-weight coarse
aggregate, both satisfying ASTM requirements. The fine
aggregate in both normal-weight and light-weight concrete
was natural sand satisfying the ASTM gradation requirements.

The normal-weight concrete considered in this
investigation had a specific gravity of 2.32 and a
compressive strength 4880 psi (33 Mpa). The light-weight
concrete had a specific gravity of 1.52 and a compressive
strength 2280 psi (15.5 Mpa). The permeability test was
conducted following the AASHTO T227-83 rapid chloride
permeability test procedure which measures permeability
based on the rate of diffusion of chloride ion into

concrete .

97

Ten replicated permeability tests were conducted for

each condition (unloaded and preloaded concrete materials of

normal and light weight) and the results were analyzed

statistically in order to compare the permeability in

different conditions in light of the variations in test

results. The following conclusions were derived based on

the test data generated in this investigation.

1.

Light-weight concrete materials were on the average 47%
more permeable than normal-weight concrete materials.
The variation in test results were, however, high (40%
coefficient of variation), and a one-way analysis of
variance at 5% level significance could hardly confirm
the difference between the permeability of light-weight
and normal-weight concrete. This difference was more

easily distinguishable at 10% level of significance.

Compressive loading up to 40% of compressive strength
increased the average permeability of normalaweight and
light-weight concrete by 29% and 8%, respectively,
indicating that interface microcracks under compression
are more severe in normal-weight concrete than in
light-weight concrete. The variation in test results
were so high that one-way and twoaway analyses of
variance at 10% level of significance could not confirm
the effect of loading on the permeability of normal-

weight and light-weight concrete.

98

While unloaded light—weight concrete was more permeable
than unloaded normal-weight concrete (as confirmed by
oneaway analysis of variance at 10% level of
significance), smaller effect of loading on the
permeability of light-weight concrete led to a
condition where preloaded light-weight and normal-
weight concretes showed comparable permeability
characteristics as indicated by the results of one-way
analysis of variance of the test data. This was partly
a result of large variations in test results, noting
that preloaded light-weight concrete was still 23% more

permeable than preloaded normal-weight concrete.

Large variations in permeability test data require
larger sample sizes to be used for more powerful

statistical analyses of the permeability test data.

 

 

 

CHAPTER 5
MICROCRACKING AND FAILURE MECHANISM

IN LIGHT-WEIGHT VERSUS NORMAL-WEIGHT
CONCRETE

5.1 INTRODUCTION

Failure in concrete materials results from the
propagation and interconnection of microcracks in the
material under different load and environmental effects.
Microcracks exist in concrete materials, particularly at
aggregate-cement paste interfaces, prior to any loading due
to differential movements between aggregate and cement paste
caused by shrinkage and thermal movements or settlement of
aggregates in the paste.

Past investigations on the microcracking and failure
mechanism.of concrete materials have mainly dealt with the
effect of quasi-static loading on normal-weight concrete.
This investigation is concerned with the effects of impact
as well as quasi-static loads on microcracking and failure

in normal-weight and light-weight concrete materials.

5.2 BACKGROUND

When concrete is subjected to increasing compression,
at a certain critical stress the volume of concrete begins
to increase rather than to decrease; the significance of
this phenomenon was first recognized by Brandtzaseg (57),
who suspected that the beginning of volume dilation

indicates the beginning of internal disruption. Since then,

99

100

many investigators have realized the importance of critical
stress. HSU, et a1. (57), observed that in the vicinity of
critical stress (70 to 90 percent of ultimate stress),
internal microcracking through mortar starts to increase
significantly (57).

Research in the last 20 years has proved that failure
of concrete is subject to the process of microcracking,
particularly at the interfacial region between cement paste
and aggregate particles (7,2), which is the weakest link in
concrete composites. These cracks were attributed to
bleeding, segregation and volume change of cement paste
during hydration of cement. .

Like in other brittle materials, the failure process in
concrete consists of three stages: crack initiation; slow
crack growth; and rapid crack growth (4). The three stages
determine the stress-strain behavior and failure of

concrete, as shown in Figure 5.1 for normal-weight concrete

i
Fast growth
of matrix cracks
75 :
Slow growth of matrix
crack + bond cracks
5. 1
Slow growth of bond
cracks
30 Y

Negligible bond cracks

(4). (56,57)
mo

 

Stress (%)

 

O

 

Strain

Figure 5.l.--Stress strain curve of concrete in Compression

(7)-

101

Below about 30 percent of the ultimate compression
load, the stress-strain curve is almost linear (4) and at
this stage, for normal-weight concrete, bond cracks exist in
the material (Figure 5.2a); but these are quite stable and
have little tendency to propagate. At 75-80% level of the
ultimate compression load, propagation of cracks start in
normal-weight concrete, as typically shown in Figures 5.2b
and 5.2c for the pre-peak and post-peak conditions
respectively. Severe growth and interconnection of
microcracks in the post-peak region cause the fracture of

concrete (See Figure 5.2d.

 

(c) Post-Peak (d) Failure

Figure 5.2.--Microcrack Growth in Normal-Weight Concrete
Under Compression (56).

 

 

 

102

5.3 EXPERIMENTAL PROGRAM

Microcracking and failure mechanism of light-weight
versus normal-weight concrete under quasi—static and impact
loads were investigated experimentally.

The basic mix ingredients were Type I Portland cement,
light-weight coarse aggregate (for the light-weight
mixtures), normal-weight coarse aggregate (for the normal—
weight mixtures, normal-weight fine aggregate (for all
mixtures), water, and an air entraining agent.

Gradation of normal-weight coarse and fine aggregates,
and properties of the light-weight coarse aggregate, were
presented in Chapter 3. The normal-weight and light-weight
concrete materials had comparable gradations. A slump of 3"
i 0.5" (75 i 12.5 mm) was maintained in normal-weight and
light-weight concretes through adjustments in water content.
An air entraining agent was used to maintain the air content
of fresh mixtures in the range from 7% to 9% in order to
ensure the frost resistance of the resulting concrete. For
this purpose, the dosage rate of the air entraining agent
was adjusted in the normal-weight and light-weight mixtures.
The mix proportion for normal-weight and light-weight
concrete are shown in Table 5.1. Both normal-weight and
light-weight concrete materials have comparable proportions

by volume of aggregates and cement.

 

103

Table 5.1 Mix Proportion

Coarse Fine
Cemen wate Aggregage Aggregate
lb/yd lb/yd lb/yd 11mm3
NAME (kg/m3) (kg/m3) (kg/m3) (kg/m )

 

Normal-Weight 708(431) 311(185) 1452(862) 1458(865)
Concrete

Light-Weight 800(476) 328(192) 404(240) 1592(949)
Concrete

A conventional concrete mixer was used for the
production of normal-weight and light-weight concrete. The
procedure adapted for producing concrete is as follows:

1. .Add the coarse and the fine aggregate with half of
the water into the mixer;

2. Start the mixer and add all the cement while the
mixer is running.

3. Add the remainder of water with air entraining
agent.

4. Continue mixing of concrete for three minutes.

5. Stop the mixer for two minutes

6. Start the mixer and continue mixing for two
minutes.

The specimens for microcracking studies through image
analysis were cylinders 3 in. (75 mm) in diameter and 6 in.
(150 mm) in length for quasi-static compression tests and 6
in. (150 mm) in diameter and 2.5 in. (62.5 mm) in length for
impact tests. They were cast in molds and externally
vibrated for compaction. The specimens were kept under a
wet burlap with plastic sheet for 24 hours and then demolded
and further moist-cured for another 4 days before being air-

dried until the test age of 28 days. A total of thirty six

 

 

 

 

104

compression and sixteen impact specimens were manufactured
and tested for each of the normal-weight and light-weight
concrete mixtures.

The workability of fresh mixtures was assessed by slump
tests (ASTM-C-143); the unit weight and air content tests on
fresh mix were conducted according to ASTM C-l38.

The compression tests were displacement-controlled with
a quasi-static strain rate of 10'5/sec. The compressive
strength values were measured for ten specimens of each mix
and are shown in Table 5.2.

Table 5.2. Compressive Strength Test Results of Normal-
Weight and Light-Weight Concrete.

 

Normal-Weight Concrete, psi (Mpa) Light-Weight Concrete, psi (Pma)

 

 

4542 (31) 2075 (14)
4712 (32.5) 2381 (16)
5185 (35) 2648 (18)
4934 (34) 2867 (19)
4735 (32) 2507 (17)
5610 (38) 2262 (15)
4682 (32) 1485 (10)
4592 (31) 2110 (14)
4721 (32.5) 1870 (15)
4312 (30) 1970 (13.5)

Mean 4733 (32) 2215 (15)

Standard 345 (2.37) 400 (2.75)

Deviation

Co-efficient 7.2% 18%

of Variation

95% Confidence

Interval

Lower Limit 4496 (31) 1928 (13)

Upper Limit 4990 (34) 2502 (15)

105

Based on the average compressive strength (f’c)
obtained from ten tests for each mix eight stress levels
were selected and two specimens of the same mix were loaded
to eight stress levels. The following stress levels were

cousidered in this study:

0.00 fc 0.93 fc Post-Peak
0.30 fc Pre-Peak 0.89 fc Post-Peak
0.83 fc Pre—Peak 0.81 fc Post-Peak
1.00 fc Peak 0.71 fc Post-Peak

Each specimen was then encased in a polyethylene fiber
reinforced mortar in order to maintain the integrity of
specimens during loading and preparation for image analysis
(especially those loaded to large strains in the post-peak
region). The specimens were then cut transversely and
longitudinally; 1/2" (12 mm) slices were cut from the center

of specimens as shown in Figure 5.3.

7

_. Umpruduul
6 0 In. She:
Tnnwuu

S be:

   

 

 

 

 

V
(a) Transverse Slice (b) Longitudinal Slice

Figure 5.3.--Location of Transverse and Longitudinal Slices

106

Impact tests were performed following the procedure
recommended by ACI committee 544 (54). This test measures
the amount of impact energy necessary to start a visible
crack in concrete and then continue the opening of cracks
until failure. The impact test is performed by dropping a
standard 18 lb. (8.2 kg) hammer repeatedly and recording the
number of blows required to cause the first crack on the top
of the specimens.

The first-crack impact strength test results for 10
specimens of normal-weight and light-weight concrete are
shown in Table 5.3.

Table 5.3. First Crack Impact Test Results

 

 

 

Normal-Weight Concrete Light-Weight Concrete
11 1
9 2
23 7
7 4
13 3
24 5
10 7
10 4
20 2
33 2
17
Mean 16 4
Standard 8.26 2
Deviation
Co-efficient 54% 50.0%

of Variance

95% Confidence
Interval

Upper Limit 20 7

Lower Limit 11 4

107

Microcracking studies were conducted on impact
specimens subjected to 0%, 35%, and 70% of the average
first-crack impact loads. Image analyses were conducted
only on transverse slices (Figure 5.3a) of impact test

specimens.

5.4 IMAGE ANALYSIS PROCEDURE

Concrete slices were washed in a jet of water and
allowed to dry in the lab environment for 24 hours. They
were then stained with black indian ink, ground with silicon
carbide on rotating laps over a sequence of 5 grit sizes:
#180, 240, 320, 400, and 600. These specimen preparation
steps help distinguish the microcracks under microscope.
Figure 5.4 shows a view of longitudinal and transverse

slices after being prepared for microstructural studies.

T
JI—OJ in, E M m

   

(a) Longitudinal Slice (b) Transverse Slice
Figure 5.4. Longitudinal and Transverse Slices Ready to be
Viewed.
The slices were then examined for microcracking

characteristics using an image analysis system (typical

magnification 25 x). The cracks at the prepared faces of
the slices were visible as black lines.

Image analyses were conducted after dividing the
surface area of the slice into about thirty 13 by 13 mm (0.5
by 0.5 in.) squares, each to be viewed as a separate field

of measurement (see Figure 5.5).

 

Figure 5.5. Image Analysis System.

Once subdivided, the slices were looked at through a
microscope connected to the image analysis system. For each
field of view of microscope (which covers a 4 mm by 3 mm
area within each 13 mm by 13 mm square of the mesh), the

following three measurements were taken:

109

1. The intensity of bond microcracks at aggregate—
cement interfaces shown in Figure 5.6 (with intensity
defined as the total microcrack length per unit area);

2. The intensity of matrix microcracks (Figure 5.6b);
and

3. Microcrack orientations defined as the average
inclination of microcracks with respect to the direction of
loading (performed on the longitudinal slices only).

The measurements on microcrack intensity will reveal
information on the process of failure-in concrete materials
under increasing stress levels and impact loads. The
inclination of microcracks will also provide indications of

the nature of failure mechanism under compression.

 

(a) Aggregate—Cement Interface Microcrack

110

 

(b) Matrix Microcrack

Figure 5.6. Microcracks in Concrete Materials.

5.5 EXPERIMENTAL RESULTS AND DISCUSSIONS

 

The results of microstructural studies on the process
of microcrack propagation and failure under compression and
impact loads in normal-weight and light—weight concrete
materials are discussed in this section.

The results of microcrack intensity (microcrack length
per unit area of the cross section) measurements at
different levels of compressive stress and impact load on
normal—weight concrete materials are presented in Figures

5.7 and 5.8.

111

 

 

   

 

”nun/m
l2’—— —- _____
10 —~ —*~—— ,
e
6
, ,.__._A
2 A_l _

coon: .30 lo 0.701: 0811: '831c .esic ,snc Lolc " :3: (133“, cm; 1,, 0,3”; Q55“, (15.918 0.911.; 1.06 n:
Compresslve Stress Level Compresle sless level
-Bond and Guam crack I Daona crock -anx such
(a) Transverse Slice (b) Longitudinal Slice

Figure 5.7.--Microcrack Intensity at Different Compressive
Stress Levels in Normal-Weight Concrete.

mm/ mm2
6

 

     

 

 

 

l I l
0% Impact level 35% Impact level 70% Impact level

DBond crack Matrix crack

Figure 5.8. Microcrack Intensity at Different Impact
Loading Levels in Normal—Weight Concrete (Transverse
Slices).

112

Figures 5.7 and 5.8 indicate that normal-weight
concrete incorporates microcracks (mainly at aggregate-paste
interfaces) before any loading. These microcracks result
from differential shrinkage and thermal movements between
the paste and aggregates. Under initial compression or
impact loading (up to about 30% of compressive or impact
strengths) in normal-weight concrete, the intensity of
microcracks at aggregate-paste interfaces, and within the
paste increase rather rapidly. Thereafter, the increase in
microcrack intensity with increasing compression or impact
loading takes place mainly in the cement paste phase of
normal-weight concrete materials. Typical micrographs of
normal-weight concrete materials subjected to compression
loading up to 70% of compressive strength, and impact
loading up to 70% of impact resistance, are presented in
Figures 5.9a and 5.9b, respectively. The aggregate-paste
interface and paste microcracks can be observed in these

micrographs.

113

 

(a) Under Compression Loading (70% Compressive Strength)

114

 

(b) Under Impact Loading (70% of Impact Loading)

Figure 5.9. Microcrack of Normal-Weight Concrete Material
(Magnification 5x).

115

The microcrack intensities in light-weight concrete
under increasing levels of compressive stress are shown in

Figures 5.10 and 5.11.

turn/m0

         

°'°°'° ”0” 07°": °‘°“° 0'33 ‘° 039 ‘6 091 Ic 1.00m o.ooic 0.80m more cane O83Ic came am to 1.00“:
lv tre Level
DMatrlx ORAGQCRACK Compress B S 38
.mu: Or m crack.

(a) Transverse Slice (b) Longitudinal Slice

Figure 5.10.--Microcrack Intensity at Different Compressive
Stress Levels in Light-Weight Concrete.

mm/ mm2

 

 

 

 

 

 

I l l l l . l V. J

O ‘1. IMPACT LEVEL 35 %Impact Level 70% Impact Level

 

EEK/lam Or Agg.Cracks

Figure 5.11. Microcracking Intensity at Different Impact

Loading Levels in Light-Weight Concrete (Transverse
Section).

116

The process of microcracking and failure in light-
weight concrete is distinguished from that in normal-weight
concrete by the lack of interface cracks between cement
paste and light-weight aggregates. Before any loading as
shown in Figures 5.10 and 5.11, light-weight concrete
incorporates only matrix cracks. This could be illustrated
by the lower stiffness of light-weight aggregates which
provide less restraint against shrinkage and thermal
movements in paste, preventing the development of major
interfacial stresses. Reduced settlement of light-weight
aggregates inside the paste and thus reduced bleeding in
light-weight concrete also leads to improved interface zone
characteristics, further reducing the chance of microcrack
development at aggregate-paste interfaces in light-weight
concrete.

Comparison of Figures 5.7 and 5.10 and Figures 5.8 with
5.11 indicate that the microcrack system is more stable in
light-weight concrete. Microcrack propagation in light-
weight concrete takes place at a relatively slow rate
initially (prior to about 50% of compressive or impact
strength), and it does not reach the relatively high
microcrack intensities observed in normal-weight concrete
near ultimate compressive strength or impact resistance.
The propagation of microcracks in light-weight concrete
takes place either within the paste or through the
aggregates. This could result from the relative weakness of

light-weight aggregates when compared with the interface

117

zone, and also from the fact that light-weight aggregates,
due to their reduced stiffness, are more compatible with
paste, causing a reduction in the level of interface cracks
under load.

Typical micrographs of light-weight concrete materials
subjected to compressive loading up to 70% of compressive
strength, and impact loading of up to 70% impact strength,

are presented in Figures 5.12(a) and 5.12(b),.respectively.

118

 

(a) Under Compressive Loading (70% of Compressive Load)

119

 

(b) Under Impact Loading (70% Impact of Impact Strength).

Figure 5.12. Microcracking of Light-Weight Concrete
Materials (Magnification 5x).

120

The fact that microcracks develop only within the paste
or through the light-weight aggregate, but not at the
interface between the paste and light-weight aggregates, can
be observed in these figures.

Measurements of microcrack inclinations (with respect
to the longitudinal axis of compressive specimens along
which loading was applied) on the longitudinal slice
indicated that microcracks are inclined on the average by
20° and 300 in nonmal-weight and lightsweight concrete

materials, respectively.

5.6 SUMMARY AND CONCLUSION

The process of microcrack propagation and failure in
normal-weight and light-weight concrete materials under
compression and impact loads were investigated
experimentally using image analysis techniques. The results
indicated that:

(1) Microcracks exist in concrete prior to any
loading. In the case of normaleweight concrete, these
microcracks appear both at the aggregate paste interfaces
and also within the paste; for light-weight concrete,
however, microcracks appear initially only within the paste.

(2) Microcrack intensities tend to increase under
increasing compression and impact loads. This increase
takes place in normal-weight concrete first mainly at the
interfaces and then, at higher load levels, dominantly

within the paste. In the case of light-weight concrete,

121

microcrack propagation takes place within the paste and
through the aggregates, but not at the interfaces. The rate
of microcrack propagation in light-weight concrete under
compression and impact loads is lower than.normal-weight
concretes, and the microcrack intensity does not reach the
relatively high levels observed in normal-weight concrete
under ultimate compression and impact loadings.

(3) Matrix microcracks tend to be more inclined (with
respect to the compressive stress direction) in light-weight
concrete than in normal-weight concrete.

The lack of interface microcracks and the relatively
stable nature of the microcrack system in light-weight
concrete, when compared with normal-weight concrete, may be
attributed to a number of factors including: (a) better
compatibility of lightaweight aggregates with cement paste
(as far as the similarity of their elastic moduli are
concerned) which reduce the interfacial stresses caused by
differential shrinkage and thermal movements between the
paste and aggregates; (b) reduced settlement of light-weight
aggregates within the paste which leads to reduced bleeding
and improved interface zone characteristics; (c) relative
weakness of light-weight aggregates when compared with the
interface zone; and (d) capability of flexible light-weight
aggregates to absorb the energy transferred to the material

by impact loading without damage to the interface zone.

CHAPTER 6

SUMMARY AND CONCLUSIONS

6.1 INTRQDUCTION

Structural light-weight concrete materials possess
higher strength-to-unit weight rations when compared with
normal-weight concrete. This provides opportunities for
reduction of dead load in concrete-based high rise buildings
and long-span bridges. Production of now lower-cost light-
weight aggregates using industrial by-products and recycled
waste presents a recent development which can lead to major
increases in the use of light-weight concrete. various
aspects of the performance characteristics of light-weight
concrete materials made with a ceramic-based coarse light-
weight aggregate were investigated and compared with those
of normal-weight concretes with similar mix proportions (by
volume) and fresh mix properties. The research was
performed in three phases: (1) Assessment of the
effectiveness of synthetic fibers with different elastic
moduli; (2) measurement of permeability characteristics as
influenced by loading; and (3) investigation of the
microcracking process and failure mechanism under static and
dynamic loads. .A summary of the work conducted under each
of these categories and the relevant conclusions are

presented below.

122

123

6.2 EFFECTIVENESS OF DIFFERENT SYNTHETIC FIBERS IN NORMAL-
WEIGHT AND LIGHT-WEIGHT AGGREGATE CONCRETE

The effectiveness of different synthetic fibers in
normal-weight and light-weight concrete materials was
investigated experimentally. Six concrete mixes of normal-
weight and light-weight concrete were studied; in four mixes
synthetic fibers (polypropylene and polyethylene) were used
at 0.25% volume fraction, and the remaining two mixes were
without fibers (plain). The light-weight concrete mixtures
incorporated a ceramic-based light-weight coarse aggregate
with normal-weight fine aggregate. The normal-weight and
light-weight coarse aggregates had similar gradations
satisfying the ASTM C-33 and C-330 requirements. The volume
proportions of coarse aggregate, fine aggregate and cement
were kept constant in all mixtures. The dosage of
superplasticizer was adjusted to give a constant slump in
different mixes. The air content of different mixtures was
maintained at a desirable level for frost resistance through
the use of an air entraining agent at required dosages. The
high-modulus polyethylene fiber used in this investigation
had a higher tensile strength and a substantially higher
modulus of elasticity when compared with the polypropylene
fiber.

The fresh concrete mixtures were tested for slump, Vebe
time, air content, and unit weight. The hardened concrete
materials were tested for flexural behavior and impact
resistance. The hardened material test data were replicated

in order to allow for powerful analyses of variance of the

test data.
generated

1.

124

The following conclusions, derived based on the
test data, are valid at 5% level of significance:
Fiber reinforcement has negligible effects on the
unit weight of light-weight and normal-weight
concrete materials. In the specific mixtures of
this study, the lightdweight concrete materials
had unit weight of about 52% those of normal
weight concrete (with and without fibers).

Both synthetic fibers increase the flexural
strength and toughness of normal-weight and light-
weight concrete materials.

Considering the variations in flexural strength
and toughness test results, the improvement in
flexural behavior of normal-weight concrete with
polyethylene fiber reinforcement are not
significantly superior to those obtained through
polypropylene fiber reinforcement; this is in
spite of the fact that polyethylene fibers produce
higher average values of flexural toughness than
polypropylene fibers.

There is a tendency in variations in flexural
strength test results to be reduced with fiber
reinforcement.

.As far as the percentage increase in flexural
strength is concerned, polypropylene fibers
produce comparable results in light-weight and

normal-weight concretes at 5% level of

10.

11.

125

significance. Polyethylene fibers are more
effective in light-weight concrete than in normal-
weight concrete.

No consistent relationship was observed in this
study in the improvements in flexural properties
‘with fiber reinforcement and the ratio of
fiber-to-matrix modular ratio of elasticity.

Both synthetic fibers increased the first crack
and ultimate impact strengths of normal-weight and
light-weight concrete materials.

Polyethylene fibers were more effective than
polypropylene fibers in enhancing the impact ‘
resistance of light-weight and normal-weight
concrete materials.

Both polypropylene and polyethylene fibers were
more effective at 5% level of significance, in
increasing the impact resistance of light-weight
concrete when compared with normal-weight
concrete.

No consistent relationship was observed between
the ultimate impact strength ratio (of fibrous to
plain concrete) and the modular ratio of fibers to
matrix.

The light-weight concrete materials considered in
this investigation (plain or fibrous) possessed
lower flexural strength, toughness, and impact

resistance when compared with the corresponding

126

normal-weight concrete materials (at 5% level of
significance); an exception to this rule was
observed for the flexural toughness of light-
weight polypropylene fiber reinforced concrete
which, at 5% level of significance, was comparable
to that of normal-weight polypropylene fiber
reinforced concrete.
6.3 PERMEABILITY CHARACTERISTICS OF LIGHT-WEIGHT AND
NORMAL-WEIGHT CONCRETE
The differences between permeability characteristics of
normal-weight and light-weight concrete materials were
investigated and the effects of compressive loading (up to
40% of compressive strength) on the permeability of normal-
weight and light-weight concrete materials were studied.
The light-weight and normal-weight concretes had comparable
mix proportions (by volume) and performed similarly in the
fresh state as far as slump and air content were concerned
(the dosage of air entraining agent was adjusted for
achieving comparable air contents). The light-weight coarse
aggregate used in this investigation was ceramic-based, with
specific gravity 0.58 and relatively low water absorption.
The normal-weight coarse aggregates (crushed limestone) had
a gradation similar to that of light-weight coarse
aggregate, both satisfying ASTM requirements. The fine
aggregate in both normal-weight and light—weight concrete

was natural sand satisfying the ASTM gradation requirements.

127

The normal-weight concrete considered in this
investigation had a specific gravity of 2.32 and a
compressive strength 4880 psi (33 Mpa). The light-weight
concrete had a specific gravity of 1.52 and a compressive
strength 2280 psi (15.5 Mpa). The permeability test was
conducted following the AASHTO T227-83 rapid chloride
permeability test procedure which measures permeability
based on the rate of diffusion of chloride ion into
concrete.

Ten replicated permeability tests were conducted for
each condition (unloaded and preloaded concrete materials of
normal and light weight) and the results were analyzed
statistically in order to compare the permeability in
different conditions in light of the variations in test
results. The following conclusions were derived based on
the test data generated in this investigation.

1. Light-weight concrete materials were on the
average 47% more permeable than normal-weight
concrete materials. The variation in test results
were, however, high (40% coefficient of
variation), and a one-way analysis of variance at
5% level significance could hardly confirm the
difference between the permeability of light-
weight and normal-weight concrete. This
difference was more easily distinguishable at 10%

level of significance.

128‘

Compressive loading up to 40% of compressive
strength increased the average permeability of
normal-weight and light-weight concretes by 29%
and 8%, respectively, indicating that interface
microcracks under compression are more severe in
normal-weight concrete than in light-weight
concrete. The variations in test results were so
high that one-way and two-way analyses of variance
at 10% level of significance could not confirm the
effect of loading on the permeability of normal-
weight and light-weight concrete.

While unloaded light-weight concrete was more
permeable than unloaded normal-weight concrete (as
confirmed by one-way analysis of variance at 10%
level of significance), smaller effects of loading
on the permeability of light-weight concrete led
to a condition where preloaded light-weight and
normal-weight concretes showed comparable
permeability characteristics as indicated by the
results of one-way analysis of variance of the
test data. This was partly a result of large
variations in test results, noting that preloaded
light-weight concrete was still 23% more permeable

than preloaded normal-weight concrete.

129

4. Large variations in permeability test data require
larger sample sizes to be used for more powerful
statistical analyses of the permeability test

data.

6.4 MICROCRACKINQ AND FAILURE MECHANISM IN LIGHT-WEIGHT
VERSUS NORMAL-WEIGHT CONCRETE

The process of microcrack propagation and failure in
normal-weight and light-weight concrete materials under
compression and impact loads were investigated
experimentally using image analysis techniques. The results
indicated that:

(1) Microcracks exist in concrete prior to any
loading. In the case of normal-weight concrete, these
microcracks appear both at the aggregate paste interfaces
and also within the paste; for light-weight concrete,
however, microcracks appear initially only within the paste.

(2) Microcrack intensities tend to increase under
increasing compression and impact loads. This increase
takes place in normal-weight concrete first mainly at the
interfaces and then, at higher load levels, dominantly
within the paste. In the case of light-weight concrete,
microcrack propagation takes place within the paste and
through the aggregates, but not at the interfaces. The rate

of microcrack propagation in light-weight concrete under

 

130

compression and impact loads is lower than normal-weight
concretes, and the microcrack intensity does not reach the
relatively high levels observed in normal-weight concrete
under ultimate compression and impact loads.

(3) Matrix microcracks tend to be more inclined (with
respect to the compressive stress direction) in light—weight
concrete than in normal-weight concrete.

The lack of interface microcracks and the relatively
stable nature of the microcrack system in light-weight
concrete, when compared with normal-weight concrete, may be
attributed to a number of factors including: (a) better
compatibility of light-weight aggregates with cement paste
(as far as the similarity of their elastic moduli are-
concerned) which reduces the interfacial stresses caused by
differential shrinkage and thermal movements between the
paste and aggregates; (b) reduced settlement of light-weight
aggregates within the paste which leads to reduced bleeding
and improved interface zone characteristics; (c) relative
weakness of light-weight aggregates when compared with the
interface zone; and (d) capability of flexible light-weight
aggregates to absorb the energy transferred to the material

by impact loading without damage to the interface zone.

APPENDIX I

DESCRIPTION OF LIGHT-WEIGHT AGGREGATES

APPENDIX I
DESCRIPTION OF LIGHT-WEIGHT AGGREGATES

This appendix presents the properties and production
techniques of some commonly used light-weight aggregates.

Aggregate: Leca (Expanded Clay) (12,14,42):
Dry unit weight: 35-65 lb/cu.ft. (561-1041 kg/cu.m.)

Particle shape: Spherical

Concrete unit weight: 80-100 lb/cu.ft. (1282-1602.5
kg/cu.m4)

Concrete 28-day compressive strength: 2400-2500 psi (16.5-
17.25 MPa)

Typical concrete applications: structural elements and'

insulating components.

Aggregate: Foamed Slag (Expanded Slag) (2,12,14,42):
Dry unit weight: 43.4 .b/cu.ft. (694 kg/cu.m.)

 

Water absorption (% by weight): 5-25

Production technique: This is an artificial light-weight
aggregate manufactured from.molten blast furnace slag which
is heated up to 2192°F (1200°C); water and air are used for
bloating the slag.

Typical practice size: 0.2-0.48 in. (5-12 mm)

Particle shape: Irregular.

Concrete unit weight: 86.8-117.8 lb/cu.ft. (1400-1900
kg/cu.m.)

Concrete 28-day compressive strength: 1015-4060 psi (7-28

MPa)

131

132
Typical concrete applications: structural reinforced

concrete, and roof insulation.

Aggregate: Aglite (Sintered Shale)
Dry unit weight: 44 lb/cu.ft. (700 kg/cu.m.)

 

Production technique: Raw materials are diffused in a
rotary kiln at temperature of 1832 to 2732°F (1000 to
1500°C); expansion occurs due to generation of gases. The
fused materials are cooled and reduced to the required
particle size.

Typical particle size: 0.4-0.6 in (10-15mm)

Particle shape: Irregular

Concrete unit weight: 100-115 lb/cu.ft. (1613-1855
kg/cu.m4)

Concrete 28-day compressive strength: 6000 psi (41.4 MPa)
Typical concrete applications: Structural light-weight

concrete, roof insulation, and non-load-bearing walls.

Aggregate: Lygag (2,12,41,42):
Dry unit weight: 37-63 lb/cu.ft. (597-1016 kg/cu.m.)

Production technique: The raw material for the production
of this aggregate is fly ash and coal dust. the fly ash and
coal slurry are mixed together in a mixer. Mixing should be
done carefully in order to keep the water and carbon

contents controlled to produce strong pellets; these pellets

133

are then heated up to a temperature of 2192 to 2372°F (1200

to 1300°C). This sintered fly ash aggregate is then crushed
and graded.

Typical particle size: 0.2-0.5 in (5-12 mm)

Particle shape: Irregular

Concrete unit weight: 85-120 lb/cu.ft. (1371-1935 kg/cu.m.)
Concrete 28-day compressive strength: 2000-6000 psi (13.8-

41.38 MPa)

Typical concrete applications: Structural reinforced and

prestressed concrete.

Aggregate: Solite (expanded solite) (2,12,42):
Dry unit weight: 40-55 lb/cu.ft. (640-880 kg/cu.m.)

 

Production technique: Expanded solite is manufactured by
heating solite in a rotary kiln at 1832 to 2192°F (1000 to
1200°C). The expanded mass thus produced due to the
generation of gases in the kiln is cooled gradually to
reduce the specific gravity of the expanded structure.
Typical particle size: 0.2-0.4 in (5-10 mm)

Particle shape: Irregular

Concrete unit weight: 50-118 lb/cu.ft. (806-1903 kg/cu.m.)
Typical concrete applications: Structural reinforced and
prestressed concrete elements in high-rise buildings (to

reduce load on foundations).

134

Aggregate: Cinder (2,4):
Dry unit weight: 45-65 lb/cu.ft. (720-1040 kb.cu.m4)

Production techniques: Cinders are residues from combustion
of coal or coke in industrial furnaces.

Concrete unit weight: 45-95 lb/cu.ft. (720-1000 kg/cu.m.)'
Concrete 28-day compressive strength: 305-1000 psi (2.1-6.9
MPa).

Typical concrete applications: Concrete block, and plain
concrete.

Comments: Use of Cinders in reinforced concrete is not
recommended because of sulphate contents which affect

durability of concrete.

Aggregate: Expanded slag (2,4,14,42):

Unit weight: 20-70 lb/cu.ft. (320-ll2l.8 kg/cu.m.)

Water absorption (% by weight): 5-25

Production techniques: There are two methods of producing

expanded slag:

i) When molten slag is coming out of the-furnace, water in
the form of spray comes in contact with it. The steam
thus generated makes the slag surface porous, leading
to weight reductions. This process is called water jet
process.

ii) A controlled amount of water is agitated with molten
slag. Steam is entrapped in molten slag and, together
with gas, reduces the slag weight. This process is

called machine process.

135
Typical particle size: 0.2-0.4 in (5-10 mm)
Particle shape:t Irregular
Concrete unit weight: 70-115 1b/cu.ft. (1121.8-1843
kg/cu.m.)
Compressive strength: 2000-6000 psi (13.8-42.4 MPa)
Concrete thermal conductivity: 0.98-1.9 Btu/ft2.hr.°F (1.7-
3.3 W/mK)
Cost (s/yd3): 5-12 in 1972
Typical concrete applications: Insulating and structural

concrete components .

Expanded shale and clay (2,4,14):
Unit weight: 35-65 lb/cu.ft. (561-1042 kg/cu.m.)

Water absorption (% by weight): 5-15

Production technique: The aggregates are produced by
heating raw materials (clay, shale); at 1832 - 2192°F (1000-
1200°C) expansion of these materials occurs due to
generation of gases. These gases are entrapped within the
diffused mass which is cooled to produce light-weight
materials. Raw materials are processed to the desired size
before heating.

Concrete unit weight: 70-115 lb/cu.ft. (1122-1829.5
kg/cu.m.)

Concrete 28-day compressive strength: 2000-6000 psi (13.8-
41.4 MPa)

Concrete thermal conductivity: 1.1-2.6 Btu/ft2.hr.°F (1.9-

4 . 5W/mK)

 

136

Particle shape: Spherical

Cost ($/yd3): 5-12 in 1972

Typical concrete applications: Prestressed and reinforced
concrete elements.

Comments: If pelletized materials are used for the
manufacture of aggregates, the particles will be nearly
spherical in shape and they will have less water absorption

rate .

Pumice and scoria, lava, tuft, volcanic cinder (2,14):
Dry unit weight: 33-55 lb/cu.ft. (529-881 kg/cu.m4)

Production technique: Natural light-weight aggregates.-
Concrete unit weight: 50-80 lb/cu.ft. (801-1282 kg/cu.m.)
Concrete 28-day compressive strength: 580-725 psi (4-5 MPa)
Concrete thermal conductivity: 0.087-0.17 Btu/ft2.hr.°F
(0.15-0.30 W/mK)

Typical applications: Roof insulation

Comments: Pumice causes high shrinkage.

Ceramic spheres (43)
Dry unit weight: 35-65 lb/cu.ft. (560-1041 kg/cu.m.)

water absorption (% by weight): less than 0.5

Typical particle size: 0.12 - 0.40 in. (0.3 - 10 mm)
Particle shape: Spherical

Concrete thermal conductivity: 0.67-0.86 Btu/ft2.hr.°F

(1.08-1.48 W/mK)

137

Typical concrete applications: Insulation, light-weight
structural concrete, and light-weight precast concrete
units.

Comments: Macrolite ceramic spheres, typical ceramic
spheres, are low-density spheres consisting of multiple
minute independent air cells encircled by an outer shell
which is very tough. The sphere has less absorption
capacity as compared to other light-weight aggregates and

can function at high temperatures.

Lycrete (42,CP110):
Aggregate unit weight: 87 lb/cu.ft. (1394 kg/cu.m,)

Production technique: The raw material is remnant of steel
production; these remnants are heated up to 1832-2732°F
(1000-1500°C) in kiln. Expansion occurs due to generation
of gases during calcination.

Typical particle size: 0.2-0.56 in (5-14 mm)

Particle shape: Spherical

Unit weight of concrete: 45-125 lb/cu.ft. (1400-2000
kg/cu.m,)

Concrete 28-day compressive strength: 980-7000 psi

Typical concrete applications: Structural reinforced

concrete 0

Expanded pplystyxene (14):
Unit weight: 2-10 lb/cu.ft. (32-192 kg/cu.m.)

Water absorption (% by weight): 5-10

 

138

Production technique: Produced by heating and treatment of
polystyrene. Raw material is saturated hydrocarbon (low
molecular weight).

Unit weight of concrete: 20-55 lb/cu.ft. (322-877 kg/cu.m,)
Concrete 28-day compressive strength: 100-1800 psi (0.69-
12.4 MPa)

Typical concrete applications: Insulation concrete
Comments: More research is needed to make plastic-

originated aggregates suitable for structural use.

Carbonized cereal (14):
Unit weight: 4-12 lb/cu.ft. (64-192 kg/cu.m.)

Production technique: Raw materials are wheat, corn and
rice. By the process ofpuffing and distilling, the raw
materials are converted into aggregates.

Unit weight of concrete: 40 lb/cu.ft. (641 kg/cu.m.)
Concrete 28-day compressive strength: 800 psi (5.5 MPa)
Concrete thermal conductivity: <0.58 Btu/ft2.hr.°F (<1.0
‘W/mK)

Cost (S/yd3): 2.7 in 1972

Typical concrete application: roof decking

Comments: The aggregates have large pores and are highly
porous. If the large pores are either reduced or

eliminated, the strength of aggregate could be improved.

139

Exfoliated vermiculite (2,7,14):
Unit weight: 4-12 lb/cu.ft. (64.5-192 kg/cu.m.)

water absorption (% by weight): 20-35

Production technique: vermiculite is magnesium-aluminum-
iron hydrous silicate. Expanded vermiculite is manufactured
by heating vermiculite to 1202-1832°F (650-1000°C). Due to
high temperature, vermiculite expands and becomes light as
compared to its original weight.

Unit weight of concrete: 25-60 lb/cu.ft. (403-968 kg/cu.m.)
28-day compressive strength: 95-420 psi (0.65-3.9 MPa)
Concrete thermal conductivity: 0.29-0.64 Btu/ft2.hr.°F
(0.5-1.1 W/mK)

Cost ($/yd3): 4.3-13 in 1972

Typical concrete applications: Roof insulation

Comments: More expensive than regular light-weight
aggregates; highly porous; greater water absorption

capacity.

Expanded glass (7,12,14:

Unit weight: 15-30 lb/cu.ft. (250-500 kg/cu.m.)

Water absorption (% by weight): 5-10

Production technique: Origin is glass. Production of
expanded glass is possible by heating pelletized mixtures of
glass and expanding the material in rotary kiln at 1292-
1472°F (700-800°C).

Concrete unit weight: 17 lb/cu.ft. (1209 kg/cu.m.)

7-day compressive strength: 1300 psi (8.9 MPa)

140

Concrete thermal conductivity: 0.162 Btu/ft2.hr.°F
(0.28W/mK)
Typical applications: roof insulation, blocks, and panels.

Comments: New material, yet to be improved.

Expanded perlite (2,14):
Unit weight: 29-113 lb/cu.ft. (465-1810 kg/cu.m.)

Water absorption (% by weight): 10-50

Production technique: It has a volcanic origin. The raw
material is heated up rapidly to 1832°F (1000°C) for the
production of expanded perlite aggregate. During the
heating process the moisture which is present in the raw
material converts into steam and causes expansion of the
original material and reduction of weight.

Concrete unit weight: 35-60 lb/cu.ft. (564-960 kg/cu.m.)
28-day compressive strength: 80-500 psi (.55-3.4.MPa)
Concrete thermal conductivity (k factor): 0.29-0.58
Btu/ft2.hr.°F (0.5-1.0 W/mK)

Typical applications: Roof insulation

Comments: This aggregate has a relatively high water
absorption capacity; concrete made with expanded perlite is

fast-drying and can be finished rapidly.

Brick rubble (14):
Unit weight: 47 lb/cu.ft. (758 kg/cu.m.)

Water absorption (% by weight): 19-36

Production technique: Origin is clay. Sand and clay are

 

141

blended in water; bricks are made from this mixture and
dried in the sun. The dried bricks are heated in kiln and,
after cooling, cut to size.

Particle shape: Irregular

Unit weight of concrete: 110-120 lb/cu.ft. (1779-1935
kg/cu.mt)

Concrete 28-day compressive strength: 3000 psi 20.7 MPa).
Concrete thermal conductivity (k factor): 2.2-2.9
Btu/ft2.hr.°F (3.8-5.0 wlmx)

Typical applications: Concrete blocks; Floors.

Fused clinker (4):
Unit weight: 65-100 1b/cu.ft. (1048-1613 kb/cu.m4)

Concrete thermal conductivity (k factor): 1.8-3.8
Btu/ft2.hr.°F

Typical concrete applications: Interior concrete; precast
clinker blocks.

Comments: Highly expansive, could not be used in reinforced

concrete .

Air cooled slag (14):
Unit weight: 70-90 lb/cu.ft. (1129-1451 kg/cu.m.)

Water absorption (% by weight): 1-5
Production technique: Raw material is iron ore. Molten
slag is let to cool under atmospheric pressure.

Concrete thermal conductivity (k factor): 3.48-4.64

142

Btu/ft2.hr.°F (6-8 W/mK)

Unit weight of concrete: 115-130 lb/cu.ft. (1885-2096
kg/cu.m.)

Concrete 28-day compressive strength: 3000-6000 psi (2-.68-
41.34 MPa)

Particle shape: Irregular

Typical particle size: 0.08 in. (2 mm)

Typical concrete applications: Structural concrete; precast

blocks.

Granulate slag (14):
Unit weight: 55-65 lb/cu.ft. (887-1048 kg/cu.m.)

Water absorption (% by weight): 4-15

Production technique: Raw material is iron ore. Produced
by rapid quench of the molten slag.

Concrete 28-day compressive strength: 3000 psi (21 MPa).
Cost (slyd3): 0.6-l.2 in 1972

Typical applications: Structural concrete; prestressed
concrete.

Comments: It is a glassy material.

Sypopal (14):
Unit weight: 65-80 lb/cu.ft. (1048-1290 kg/cu.m.)

Water absorption (% by weight): 0.7-2.0
Production technique: The origin of syn0pal is calcium
silicate. The manufacturing of synopal consists of heating

of raw material in a double kiln process in which the

 

143
materials react in the first kiln and the product is then
annealed in the second kiln.
Concrete unit weight: 115-125 lb/cu.ft. (1855-2016
kg/cu.m.)
Concrete 28-day compressive strength: 3000-6000 psi (20.7-
41.9 MPa)
Typical concrete applications: Structural concrete; highway
pavements.
Comments: Concrete has high light reflection and skid

resistance.

144

APPENDIX II

MIX PROPORTIONING PROCEDURES

 

 

APPENDIX II

MIX PROPORTIONING PROCEDURES

Common methods of light-weight aggregate concrete mix
porportioning are presented in this Appendix. Procedures
for the adjustment of mix proportions and examples of light-
weight concrete mix design are also given.

As mentioned in Chapter 2, there are differences in mix
proportioning of normal-weight and light-weight concretes.
This is partly because light-weight aggregates have a high
absorption capacity and thus the water-cement ratio of
light-weight aggregate concrete cannot be fixed (13).

Hence, mix proportioning of light-weight concrete is
generally accomplished by the cement content-strength method
(11), instead of the method based on water—cement ratio. In
the "cement content-strength method“ of mix proportioning,
the cement content depends on the strength of light-weight
aggregate concrete (which is normally specified prior to mix
design). The quantity of water is decided based on the
desired slump value. With the cement and water contents
known, the aggregate content in the mix design is then
decided. An alternative method for the mix proportioning of

light-weight concrete is the "weight method".

II.1 "Cement Content" Method
The following steps should be taken in the "cement
content-strength method" for the proportioning of available

materials into a mix suitable for a specific job.

146

147

Step 1. Selection of slump: If slump is not specified, a

value suitable for a job can be taken from Table II.1.

Table II.1. Recommended slump values (11).

Type of Construction Slump in (mm)
Max. Min.
Beam & reinforced wall 4 (100) 1 (25)
Building column 4 (100) 1 (25)
Floor slab 3 (75) 1 (25)

The slump values suggested in Table II.1 can be used where

vibrators are employed for the consolidation of concrete.

Step 2. Selection of Maximum Aggregate Size: The maximum

aggregate size should not exceed 1/5 of the narrowest

dimension between sides of form, 1/3 the depth of the slab,
or 3/4 of the minimum spacing between reinforcing bars. If
concrete can be placed without voids and honeycombing these

limitations can be waived.

Step 3. Calculation of Cement Congent: As the cement
content varies for different types of aggregate, the
aggregate supplier should be consulted regarding the cement
content to be used in a mix. In case information from
producer/supplier are not available, the cement content can

be taken from Table II.2.

148

At the same compressive strength, light-weight
concretes typically have higher cement contents than normal-

weight concretes (Table 11.2).

Table 11.2 Cement content of light-weight and normal-

weight concretes (7).

Compressive Cement content lb/yd3 (kg/cu.m,)
Strength Light-Weight Normal-Weight

 

 

lb/in2 (MPa)

2500 (17) 425-700 (255-420) 350-550 (210-330)
3000 (21) 475-750 (285-450) >350-600 (210-360)
4000 (28) 550-850 (330-510) 400-700 (240-420)
5000 (35) 650-950 (370-560) 500-750 (300—450)

Step 4. Water and Air Contents: .Air contents in the range

 

of 5 to 9% should be incorporated into the mix in order to
avoid frost damage. Factors influencing slump are the water
and air contents and the maximum aggregate size. Table II.3
presents approximate values for water requirements in air-
entrained light-weight concrete, depending on the required
values of slump and aggregate size. Suggestions for the
required air content as a function of the exposure condition

and maximum aggregate size are also presented.

Table II.3 Approximate water and air contents required

in air entrained light-weight concrete (ll).

149

Slump in inch Water in lb/yd3 of concrete for
indicated max. size of aggregate.
3/8" 1/2" 3/4"

Air Entrained Concrete.

 

1 to 2 _ 305 295 280
3 to 4 340 325 305
5 to 6 355 335 315
Exposure of Recommended average total
concrete air content percent.

Mild 4.5 4.0 4.0
Moderate 6.0 5.5 5.0
Extreme 7.5 7.0 6.0

Step 5. Estimation of Light-Weight Aggregate VOlume: Total

volume of light-weight aggregate required, measured as the
sum of uncombined volumes on a dry-loose basis, is
typically from 28 to 34 cu.ft. per cu.yd’ of concrete (1.0
to 1.2 cu.m. per cu.m. of concrete). Of this quantity, the
fine aggregate volume may be from 40 to 60 percent. The
aggregate producer may be consulted before making a final
decision on aggregate volume. Aggregate supplier should
also be consulted for first-trial mix proportions; such mix
proportions may be the result of previous experience or
previously established mix proportions for concrete. This
first-trial mix proportion may be adjusted as necessary to
change the properties of concrete.

The “weight method" for mix proportioning of light-
weight aggregate concrete, which is an alternative to the

"cement content-strength" method is described below.

 

150

11.2 "Weightr Method

 

 

This procedure for selection of mix proportions can be
applied to both light-weight and semi light-weight aggregate
concrete. In semi light-weight concrete, normal-weight fine
aggregate is used in order to prevent floating of coarse
light-weight aggregate. The specific gravity of aggregates,
which is required for batch weight estimation, is determined
by pycnometer method. Effective mixing water could be
calculated by dry spin method.

Estimates of water-cement ratio, minimum cement
content, air content, slump, maximum size of aggregate,
strength and unit weight of aggregate, may be made before
designing a mix for a specific job.

If the slump and maximum size of aggregate are not
specified for a mix, the value of slump and aggregate size
may be chosen as follows. Slump could be selected from
Table 2.3. Maximum aggregate particle size should be less
than 1/5 of the narrowest dimension between the sides of
form (2,11,16); the supplier/producer of aggregate may have

more accurate suggestions for the size of aggregates.

Estimation of mixing water and air contents: Generally, the

strength of concrete determines the water-cement ratio of a
mix. Durability and finishing properties of concrete are
factors which can not be ignored while deciding the water-

cement ratio of concrete. Relationships between water-cement

151

ratio and strength may be developed for different types of
aggregate and cement. If such relationships are not
available, Table 11.4 can be used for the selection of
water/cement ratio..

Depending on aggregate texture and shape, mixing water
requirements may be somewhat above or below the tabulated
values, but the values given in the table are sufficiently

accurate for the first estimates.

Table 11.4 Estimated water-Cement Ratios (11).

 

Compressive strength Approximate water-cement ratio.

 

of concrete. NOn-air entrained Air entrained
psi MPa concrete concrete.
6000 41 .41 -
5000 34 .48 .40
4000 28 .57 .48
3000 21 .68 .50
2000 14 .52 .74

Cement content: Light-weight aggregates absorb water and
the water-cement ratio depends on the rate of absorption of
light-weight aggregates. Hence, it is difficult to use
water-cement ratio for the calculation of cement content,
because water-cement ratio will vary during mixing.

Proportioning of light-weight concrete mixtures on the basis

 

152

of cement content is thus advisable. The cement content can

be selected from Table II.2.

Light-weight coarse aggregate content: The amount of light-

weight coarse aggregate can be found out from Table II.5 for
different types of fine aggregate having different fineness
moduli. The fine aggregate content will be the difference
between total weight and the sum total of the weight of

water, cement and light-weight coarse aggregate.

 

Table II.5 VOlume of coarse aggregate per unit of volume of

concrete (11).

Maximum size VOlume of dry-rodded coarse aggregate per unit
Aggregate volume of concrete for different fineness

moduli of sand.

 

inch mm 2x40 2x60 2x89 3499,
3/8 10 0.50 0.45 0.46 0.44
1/2 13 0.59 0.57 0.55 0.53
3/4 19 0.66 0.64 0.62 0.60.PA

Light-weight aggregate has a high absorption capacity.
The absorption of water leads to an increase in the weight
of light-weight aggregate. The absorbed water by aggregate
will be deducted from the mixing water added to the batch of

materials to be used in concrete mixer for producing

153

concrete .

II.3 Adjustment of mix proportions

After completion of mix proportioning for a light-
weight concrete mix, some adjustments in the mix proportions
are necessary either due to absorbed water in light-weight
aggregate or variations in materials at field or laboratory.
If these adjustments are small, rule of thumb, using
guidelines of Table II.6, may be applied for adjusting the
mix proportions.

Table II.6 Guidelines for Adjustment of Mixes of Light-
Weight Concrete (11).

 

Change regpired Amount of change lb (kg) changes in
"otheragpaptities
per yd (m ) of
concrete.

Fineness content +1% +4.41 (2) water

Air content +1% -6.61 (3) water

Slump +1" (+25mm) +132 (6) water

+ 3% cement.

A procedure based on absolute volume method can be used for
making adjustments in mix proportions. This procedure
consists of calculating the quantity of light-weight
aggregate in concrete which depends on the moisture content
and specific gravity of aggregate. The specific gravity can

be determined by pycnomenter method (11).

 

 

154

11.4 Examples of Mix Proportions
Typical mix proportions for light-weight aggregate

concrete reported in the literature are presented below.

(a) Ref. 28

(lb) (kg)
Type I cement 611 277
Fly ash 97 44
Livlite coarse 350 159'
Livlite medium. 416 142
Livlite fine 314 142
Water 346 157
WRDA 56 oz 1.6
Vinsol Resin 44 oz 1.3
(b) Ref. 2

(1b) (kg)
Cement 550 34.3
Fine aggregate 895 55.5
Coarse aggregate 705 44.0

Water 485 30.3

 

155

(c) Ref. 11

(1b) (kg)
Cement 529 33.0
Coarse light-weight Aggregate 696 43.5
Fine aggregate 365 23.0

Water 443 27.5

 

 

REFERENCES

10.

12.

13.

REFERENCE

Concrete Manual, 7th Edition. U.S. Water Resource
Technical Publication, Bureau of Reclamation, Denver,
1966.

Neville, AtM., Properties of Concrete, 2nd Edition,
Wiley, New York, 1963.

White, George R., Concrete Technology, 3rd Edition,
Albany, New York, Delamar Publisher, 1977.

Mehta, P. Kumar, Concrete Structure Properties and
Materials, Prentice-Hall Inc., Englewood Cliffs, New
Jersey 07632.

Balaguru, P. and Krushnan, V. Rama, "Properties of
Light-Weight Fiber Reinforce Concrete SP105-17," Fiber
Reinforced Concrete Properties and Application 1987,
pp. 305-322. '

Nilson, AIM. and Winter, G., Design of Concrete
Structure, 10th Edition, McGraw Hill, New York, 1986.

Sidney Mindness and J. Francis Young, Concrete,
Prentice-Hall, Englewood Cliffs, New Jersey 07632.

Paul, Krieger and Hansan, J.A., "Freeze-Thaw Test of
Light-weight Concrete," American Concrete Institute
Journal, January 1961, pp. 779-796.

Litivin, A. and Fioralo, A.E., "A Light-Weight
Aggregate Concrete for OTEC Cold Water Pipe," Concrete
International, VOlume 3, No. 3, March 1981, pp. 48-55.

Pfeirer, D.W., "The Structural Use of Light-weight
Concrete In U.S.A.," Concrete, VOlume 3, No. 2, Feb
1969, pp. 60.11. ACI Committee 211, "(ACI 211.2-
81)," American Concrete Institute, 1981, pp. 18. Also
Manual of Concrete Practice 1984, Part 1.

Bobrowski, J. and Bardhan-Roy, B.K., "Structural
Assessment of Light-Weight Aggregate," Concrete, Volume
5, NO. 7, JUlY 1971' pp. 229-232.

ACI Committee 213, Report 213-79, "Guide for Structural

Light-Weight Concrete," Concrete International, VOlume
1, No. 2, 1979, pp. 33-62.

157

 

 

 

14.

15.

16.

17.

18.

19.

21.

22.

23.

24.

25.

26.

158

Berger, R.L., "Synthetic Aggregate," Proceeding of the
Conference, UNIVERSITY of ILLINOIS at CHICAGO CIRCLE,
Chicago, Illinois, Dec. 1972, pp. I-VIII-21VIII.

Shah, S. P. and Skarendahl, A., "Steel Fiber Concrete,"
Elsevier Science Publishers, 1986.

ACI Committee 335, ”State-of-the.Art Offshore Concrete
Structure for the Artic," Concrete International,
Vblume 7, No. 8, Aug. 1985, pp. 23-33.

Shah, S. P., Naaman,.A. E., and Moreno, "Effect of
Confinement on the Ductility of Light-Weight Concrete,"
International Journal of Cement Composite and Light-
Weight Concrete (Harlow), VOlume 1, February 1983, pp.
15-25 0

Craig, R. J., "Light-Weight Reinforced Fiber Concrete
Behavior and Uses," in Print.

Nichols, G. W. and Better, L., "Bond and Tensile
Capacity of Light-Weight Aggregate," ACI Journal, Title
no. 67-65, December 1970, pp. 959-962.20. Leonnard,
F., Prestressed Concrete Design and Construction, 2nd
Edition, Wilhelm Ernst and Son, Berlin, 1964.

George, Michele St., "Concrete Aggregate from Waste
Water Sludge," Concrete International VOlume 8 No. 11,
November 1986, pp. 27-30.

Concrete Manual, 8th Edition, U.S.A, A.Water Resources
Technical Publication, Bureau of Reclaimation, Denver.

Phileo, R. E., "Light-weight Concrete Bridges,"
Concrete International, VOlume 8, No. 11, November
1986, pp. 19-220

Takada, H., Uchida, I., and Sakurada, T., "Development
of Light-weight Durable Fiber Glass-Reinforced Concrete
(FRC)," SP105-10 Reinforced Concrete Properties and
Applications, 1987, pp. 179-188.

John Carmichae, "Pumice Concrete Panels," Concrete
International Velume 8, No. 11, November 1986, pp. 31-
32.

Nanni A. ,Corbitt G., and M. Phang, "Compaction and
Light SFRC Mine Cribs." SP105-l9, Fiber Reinforced
Concrete Properties and Applications, ACI Publication
1987, pp. 351-359.

 

27.

28.

29.

300

31.

32.

33.

34.

35.

36.

38.

159

Muller-Rochholz, J.F.W. and Weber, J.W., "Traffic
Vibratic of Bridge Deck and Hardening of Light-Weight
Concrete," Concrete International Vblume 8, No. 11,
November 1980, pp. 23-26.

Swamy, R.N. and Jojagha, AJH., "Workability of Steel
Fiber Reinforced Light-Weight Aggregate Concrete," the
International Journal of Cement Composite, and Light-
weight Aggregate Concrete, V61. 4, No. 2 May, 1982,
pp. 103-109.

Andrzej quryszkiewiez, "The Effect of Aggregate Shape
Upon the Strength of Structural Light-weight Concrete M
E," Magazine of Concrete Research, volume 25, No 83,
June 1973, pp. 81-86.

Rutledge, S.E. and Neville, A.M., "The Influence of
Cement Paste Content on the Creep of Light-Weight
Aggregate Concrete," Magazine of Concrete Research,
June 1966, pp. 69-74.

Fullet, P.M., "Structural Light-Weight Aggregate
Concrete, Why Should be Used for Multi-Story Structures
and Car Park?," Concrete, Aug. 1985, pp. 31.

Ritchie, and Al-Kayyali, 0.0., "The Effects of Fibre
Reinforcement on Light-Weight Aggregate Concrete,"
Fiber-Reinforced Cement and Concrete, RILEM Symposium,
1975, Construction Press Ltd., pp. 247-256.

Nedh Burns, T.Y., Design of Prestressed Concrete
Structure, 3rd Edition, John Wiley and Son.

Jaime Moreno, "Light-Weight Concrete Ductility,"
Concrete International, VOlume 8 No. 11, November 1986,
pp. 15-18.

ACI Committee 613, "Recommended Practice for Selecting
Proportion for Structural Light-weight Concrete,"
JOurnal of the American Concrete Institute, Title 55-
18, September 1958, pp. 305-313.

Troxell, G.G., Davis, H.E., Kelly, J.W., Composition
and Properties of Concrete, McGraw-Hill Book Co.,
1968.37. Shah, S.P. and Skarnedahl,.A., Steel Fibers,
Applied Science, Publisher London, New York.

"Light-Weight Aggregates for Concrete Masonry Units,"
ASTM-C-33l-87.

 

 

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

160

Brieef Bender, P.E., "Economics and Uses of Light-
weight Concrete in Prestressed Structure,” Prestressed
Concrete Institute, VOlume 25, No. 6, Nov./Dec. 1980,
PP. 66’670

MacGregor, J. G., Reinforced Concrete Mechanical
Design, Prentice Hall, Englewood Cliffes, New Jersey
07632.

Swamy, R.N. and Lambert, G.H., ”The Microstructure of
Lytag Aggregate" International Journal of Cement
Composites and Light-Weight Concrete, V61. 3, No.4
1981, pp. 273-282.

Concrete Society Data Sheet 1980, “Structural Light-
weight Aggregate Concrete."

Macrolite Ceramic Spheres Data Sheet, Effective: Jan.
1, 1987.

Nilson, A. H., Design of Prestressed Concrete, 2nd
Edition, John Wiley and Sons, 1987.

Martinez, 8., Nilson, A.H. and Slate, "Short Term
Mechanical Properties of High-Strength Light-Weight
Concrete," Research Report no. 82-9, Department of

Structural Engineering, Cornell University, August
1982.

Wang, S.T., Shah, S.P., and Naaman, A.F., "Stress
Strain Curves of Normal and Light-weight Concrete in
Compression," ACI, VOlume 75, No. 11, November 1978.

Concrete Society, "Structural Light.Aggregate Concrete
for Marine and Off Shore Applications," Technical
Report no. 16, May 1979.

Concrete Society, WA Comparative Study of the Economics
of Light-Weight Aggregate Concrete," Report of the
Cost/Price Comparison Working Party..May 1972

Aranti. C, and Sanag Khara, "Crip and Shrinkage in
Light-Weight Aggregate Concrete," Magazine of Concrete
Research, V01. 36, No. 128, Sept. 1984, pp. 165-171.

Soroshian, P., and Mahboob Khan, "Light-Weight
Concrete: STATE-OF-ART," MSU ENGR-89-018. December
1989 College of Engineering, Michigan State University

Japanese Concrete Institute, "JEI Standard for Test
Methods of Fiber Reinforced Concrete," Report No. JCI-
SF-1984-6800

 

52.

53.

54.

55.

56.

57.

58.

59.

60.

161

.Allied Fiber, "Spectra High Performance Fiber," Allied
Signal, Inc., Petersburg, Virginia, 1985. pp. 10

Soroushian, P., and Khan, A., "Allied Signal
Polyethylene Fiber Application to Coarse Aggregate
Concrete," Report No. MSU-ENGR 89-010, College of
Engineering, Michigan State Univeristy, June 1989, 3688

ACI Committee - 544, "Measurement of Fiber Reinforced
Concrete Properties," ACI September - October 1982, pp.
1-200

Shah, S.P., “Application of Fracture Mechanics to
Cemtitious Composite," Advance Science Institute Series
E, Martinus Nijhoff Publishers 1985.

Shah, S.P. and Sankar, "Internal Cracking and Strain-
Solfint Response of Concrete Uner Uniaxial
Compression," ACI Material Journal Title No. 84--M22.

Hsu, Thomas T.C.; State, Floydoo; Sturmen, Gerald, M.;
and Winter, George, "Microcracking of Plain Concrete
and Shape of the Stress-Strain Curve," ACI Journal,
proceeding V. Go. Mo. 2 Feb 1963, pp. 209-224

AASHTO T-277-83, "Standard Method of Listing for Rapid
Determination of Chloride Permeability of Concrete,"
Standard Specification for Transportation Materials and
Methods of Sampling and Listing Part II, Washington,
D.C., 1987. ‘pp. 1229-1234.

Whitting, D., "Rapid Determination of Chloride
Permeability ofConcrete," Report No. FHWA/RD-81/119,
Federal Highway Administration, Material Division,
Washington, D.C., August 1981, pp. 166

Luther, D., "Silica Fume (Microsilica) Effect on
Concrete Permeability and Steel Corrosion," MSU
Technology Seminar - 4 Feb 1990, pp. 5-9 to 5-13

 

 

av— V _ llllllllllllllllllllllllllllllllllllll