T H E U S E ... C O N C R E T E

T H E   U S E  ... C O N C R E T E
University of Pretoria etd – Du Plessis H (2006)
THE USE OF GASIFICATION ASH IN CEMENT AND
CONCRETE
HANLI DU PLESSIS
A dissertation submitted in partial fulfillment for the degree of
MASTER OF ENGINEERING (STRUCTURAL ENGINEERING)
In the
FACULTY OF ENGINEERING
UNIVERSITY OF PRETORIA
November 2005
University of Pretoria etd – Du Plessis H (2006)
SUMMARY
THE USE OF GASIFICATION ASH IN CEMENT AND
CONCRETE
H DU PLESSIS
Supervisor:
Professor E.P. Kearsley
Department:
Civil Engineering
University:
University of Pretoria
Degree:
Master of Engineering (Structural Engineering)
Cement is an essential material in today’s society because, as a major
constituent of concrete, it forms a fundamental element of any housing
or infrastructure development. The chemical process of making cement
clinker produces CO2, a major greenhouse gas contributing to climate
change.
This makes it imperative for us to find ways of using this
resource more efficiently.
Using
waste
from
other
industries,
as
a
raw
material
is
a
huge
opportunity for the cement industry to reduce its environmental impact.
Cement extenders are used as a substitute for some of the Portland
cement in concrete. The reasons for the use of extenders, is a growing
awareness of the engineering, economical and ecological benefits and
the variety of useful enhancements, which they give to the concrete
properties.
University of Pretoria etd – Du Plessis H (2006)
The aim of the research is to determine whether a gasification ash can
be used as a cement extender in concrete. In this study detail of the
manufacturing
of
Portland
cement
(PC),
cement
classes
and
the
hydration of cement will be discussed. Consideration will be given to
properties like the optimisation of sulphate and fineness of cement.
The particle size distribution is discussed with specific reference to the
Rosin-Rammler distribution function. The use of coal combustion byproducts,
specifically
fly
ash,
in
concrete
will
be
discussed.
Consideration will be given to properties like shape, particle size,
mineralogical and chemical composition, durability and the chemical
requirements for using fly ash as a cement extender.
The
physical
properties
of
gasification
ash
indicated
that
the
gasification ash, grinded separate and interground had similar particle
size distributions. The chemical and mineralogical composition of a
gasification
ash
sample
gasification
ash
has
an
was
investigated,
angular
shape
and
and
it
a
was
found
similar
that
chemical
composition as fly ash.
The chemical requirements of the gasification ash meet the majority of
the requirements specified for cement extenders. Where limits are
exceeded it is by a very narrow margin.
The effect of a gasification ash on the short and long term properties
of concrete of both interblending and intergrinding was investigated.
The use of gasification ash as cement extender does not have a
negative impact on the strength development of concrete. There was no
reduction in the tensile strength of concrete. Gasification ash does not
have a detrimental effect on stiffness of concrete, and did not shrink or
creep significantly more than concrete containing fly ash.
The porosity and permeability does not increase when gasification ash
is used as a cement extender.
Gasification ash should therefore not
decrease the durability of concrete.
University of Pretoria etd – Du Plessis H (2006)
SAMEVATTING VAN
DIE BENUTTING VAN GASIFIKASIE-AS IN SEMENT EN
BETON
H DU PLESSIS
Promotor:
Professor E.P. Kearsley
Departement:
Siviele Ingenieurswese
Universiteit:
Universiteit van Pretoria
Graad:
Magister van Ingenieurwese (Struktuur
Ingenieurswese)
Sement is ‘n belangrike materiaal in die hedendaagse gemeenskap
want as ‘n hoof vervangingsmateriaal in beton, vorm dit ‘n belangrike
element
van
enige
behuising
of
infrastruktuur
ontwikkeling.
Die
chemiese proses tydens die vervaardiging van sement stel CO2, ‘n gas
wat lei tot klimaatsverandering, vry. Hierdie aspek maak dit belangrik
vir ons om maniere te kry om hierdie bron meer effektief te gebruik.
Benutting van afval materiale van ander industrieë as a rou materiaal
is a groot kans vir die sement industrie om hul omgewingsimpak te
verminder. Sement-vervangers word gebruik as vervanging vir ‘n deel
van die Portland Sement in beton. The redes vir die gebruik van
vervangers
is
‘n
groeiende
bewuswording
van
die
ingenieurs
University of Pretoria etd – Du Plessis H (2006)
ekonomiese
en
ekologiese
voordeel
asook
die
verskeidenheid
verbeterings wat hulle bydra tot die eienskappe van beton.
Die doel van die navorsing is om te bepaal of gasifikasie-as gebruik
kan word as ‘n sement-vervanger in beton. In hierdie studie word
inligting oor die vervaardiging van Portland sement, sement klasse en
die
hidrasie
eienskappe
van
soos
sement
die
bespreek.
optimisering
Konsiderasie
van
sulfaat
word
gegee
aan
die
fynheid
van
en
sement. Die partikelgrootte verspreiding word bespreek met spesifieke
verwysing na die Rosin-Rammler verspreidingfunksie. Die gebruik van
steenkool as afval produkte, spesifiek Vliegas, in beton word bespreek.
Konsiderasie word gegee aan eienskappe soos vorm, partikelgrootte,
mineralogie en chemiese samestelling, duursaamheid en die chemiese
vereistes vir die gebruik van Vliegas as ‘n sement-vervanger.
Die fisiese eienskappe van gasifikasie-as dui daarop dat gasifikasieas,
apart
gemaal
en
saam
gemaal
dieselfde
partikelgrootte
verspreidings het. Die chemiese en mineralogiese samestelling van
gasifikasie-as is ondersoek en daar is gevind dat gasifikasie-as ‘n
hoekige vorm en dieselfde samestelling as vliegas het.
The chemiese vereistes vir gasifikasie as voldoen aan die meerderheid
van
die
vereistes
gespesifiseer
vir
sement-vervangers.
Waar
die
limiete oorskrei word is dit slegs met ‘n klein margin.
Die effek van gasifikasie as op die kort en lang termyn eienskappe van
beton vir beide saamgemeng en saamgemaal is ondersoek. Die gebruik
van gasifikasie-as as ‘n sement-vervanger het geen negatiewe impak
op die sterkte ontwikkeling van beton nie. Daar was geen vermindering
in die treksterkte van die beton nie. Gasifikasie-as het nie ‘n nadelige
effek op die styfheid van die beton en het nie besonders meer gekrimp
of kruip nie as beton met vliegas nie.
Die porositeit en permeabiliteit het nie vermeerder as gasifikasie-as
gebruik word as ‘n sement-vervanger nie. Gasifikasie-as sal daarvoor
nie die duursaamheid van beton verminder nie.
University of Pretoria etd – Du Plessis H (2006)
ACKNOWLEDGEMENT
I wish to express my appreciation to the following organizations and
persons who made this dissertation possible.
a)
Sasol Technology Research and Development for the donation of
gasification ash tested in this investigation.
b)
Professor E.P Kearsley, my promoter for her guidance and
support.
c)
The following persons for their assistance during the course of
study:
Mr. D. Mostert
Mr. H. Matjie
Mrs. S. Verryn
Mrs. J. Callanan
Personnel of the concrete laboratory of the Civil Engineering
department of the University of Pretoria.
d)
My family and friends for their encouragement and support.
University of Pretoria etd – Du Plessis H (2006)
TABLE OF CONTENTS
PAGE
1.
INTRODUCTION
1-1
1.1
Background
1-1
1.2
Objectives of the study
1-2
1.3
Scope of the study
1-2
1.4
Methodology
1-4
1.5
Organisation of the report
1-4
2.
COMPONENTS AND PROPERTIES OF PORTLAND CEMENT
2-1
2.1
Introduction
2-1
2.2
Cement Manufacture
2-1
2.3
Cement Classes
2-4
2.4
Hydration of Portland Cement
2-5
2.5
2.4.1
The hydration of C3S and C2S
2-5
2.4.2
Hydration of C3A
2-8
2.4.3
Hydration of C4AF
2-9
Heat of Hydration
2-9
2.5.1
2-10
Optimization of Cement Sulphate
2.6
Specific Surface Area
2-12
2.7
Particle size distribution
2-12
2.7.1
2-13
Rosin-Rammler distribution function
2.8
Conclusion
2-15
3.
COMPOSITION AND PROPERTIES OF COAL ASH
3-1
3.1
Introduction
3-1
3.2
Coal Ash
3-2
3.3
Pozzolanic Reaction
3-4
3.4
Fly Ash
3-5
3.4.1
Physical Properties
3-6
3.4.2
Chemical Composition
3-7
3.4.3
Mineralogical Composition
3-8
3.4.4
Chemical Specifications
3-8
University of Pretoria etd – Du Plessis H (2006)
3.5
Influence of fly ash on the properties of concrete
3-9
3.5.1
Fresh Concrete
3-9
3.5.1.1
Water Demand
3-10
3.5.1.2
Workability
3-11
3.5.2
3.6
Hardened Concrete
3-12
3.5.2.1
Compressive Strength Development
3-12
3.5.2.2
Flexural Strength
3-15
3.5.2.3
Modulus of Elasticity
3-16
3.5.2.4
Drying Shrinkage
3-17
3.5.2.5
Creep
3-18
Durability of Concrete
3-19
3.6.1
Porosity
3-19
3.6.2
Permeability
3-19
3.7
Advantages of using Fly ash in Concrete
3-20
3.8
Conclusion
3-21
4.
EXPERIMENTAL PROGRAMME AND TEST PROCEDURES
FOR CEMENT
4-1
4.1
Introduction
4-1
4.2
Preparation of materials
4-1
4.3
Physical and chemical properties of gasification ash
4-4
4.3.1
Particle size distribution
4-4
4.3.1.1
Particle size distribution parameters
4-4
4.3.1.2
Rosin-Rammler particle size
4.3.1.3
4.4
distribution parameters
4-5
Particle size distribution parameters
4-7
4.3.2
Specific surface area
4-8
4.3.3
Scanning electron microscopy (SEM)
4-9
4.3.4
X-ray fluorescence spectroscopy (XRF)
4-9
4.3.5
X-ray diffraction spectroscopy (XRD)
4-10
Standard tests for cementitious materials
4.4.1
Sulphur trioxide content
(SANS 50196-2:1994 / SABS EN 196-2:1994)
4.4.2
4-10
4-11
Loss on Ignition
(SANS 50196-2:1994 / SABS EN 196-2:1994)
4-11
University of Pretoria etd – Du Plessis H (2006)
4.4.3
Free water content
(SANS 6151:1989 / SABS 1151:1989)
4.4.4
Test for fineness of cement and Portland cement
extenders (SANS 6157:2002 / SABS 1157:2002)
4.4.5
4.4.8
4.5
4-15
Soundness
(SANS 50196-3:1994 / SABS EN 196-3:1994)
4-17
Relative density (LSA Method)
4-18
Factors Investigated by Casting Mortar Prisms
4-18
4.5.1
Different Grinding Times
4-18
4.5.2
Different Gypsum Percentages
4-19
4.5.3
Isothermal Conduction Calorimetry
4-20
4.5.4
Different replacements percentages of
Gasification Ash
5.
4-12
Strength factor test
(SANS 50196-1:1994 / SABS EN 196-1:1994)
4.4.7
4-12
Water requirement
(SANS 6156:1989 / SABS SM 1156:1989)
4.4.6
4-12
4-20
EXPERIMENTAL PROGRAMME AND TEST PROCEDURES
FOR CONCRETE
5-1
5.1
Introduction
5-1
5.2
Mix design for concrete mixes
5-1
5.3
Test Conducted on Fresh Concrete Mixes
5-2
5.3.1
5-2
5.4
Slump Test (SANS 586 / SABS SM 82:1994)
Strength Test
5.4.1
Compressive Strength Test
(SANS 5863-1/SABS 863-1994)
5.4.2
5.6
5-3
Splitting Cylinder Test for Tensile Strength
(SANS 625:1994/SABS SM 1253:1994)
5.5
5-3
5-4
Deformation and Volume Change of Concrete
5-4
5.5.1
E-Value Test
5-4
5.5.2
Shrinkage and Creep Test (ASTM C 512-02)
5-6
Durability
5-8
5.6.1
Porosity Test
5-8
5.6.2
Oxygen Permeability Test
5-9
University of Pretoria etd – Du Plessis H (2006)
6.
TEST RESULTS AND DISCUSSION ON CEMENT TESTS
6-1
6.1
Introduction
6-1
6.2
Physical Properties
6-1
6.2.1
Particle Size Distribution Test
6-1
6.2.2
Shape of Particles
6-10
6.3
6.4
6.5
6.6
6.7
6.8
6.9
Chemical Properties
6-12
6.3.1
XRF
6-12
6.3.2
XRD
6-14
6.3.3
Standard Tests for Cementitious Materials
6-15
Effect of Grinding time on the Properties of Interblended
Gasification Ash and Cement
6-16
6.4.1
Mortar Prism Compressive Strength
6-16
6.4.2
Mortar Prisms Flexural Strength
6-17
6.4.3
Particle Size Distribution
6-19
Effect of Grinding time on the Properties of Interground
Gasification Ash and Cement
6-22
6.5.1
Mortar Prism Compressive Strength
6-22
6.5.2
Mortar Prism Flexural Strength
6-23
6.5.3
Particle Size Distribution
6-24
Effect of Gypsum Content on the Properties of Interground
Gasification Ash and Cement
6-27
6.6.1
Mortar Prism Compressive Strength
6-27
6.6.2
Mortar Prism Flexural Strength
6-28
6.6.3
Heat of Hydration
6-29
Effect of Replacement Levels on the Properties of
Interground Gasification Ash and Cement
6-32
6.7.1
Mortar Prism Compressive Strength
6-32
6.7.2
Mortar Prism Flexural Strength
6-34
6.7.3
Particle Size Distribution
6-35
Comparison between Manufactured and Commercial Cement
6-38
6.8.1
Mortar Prism Compressive Strength
6-38
6.8.2
Flexural Mortar Prism Strengths
6-39
6.8.3
Heat of Hydration
6-41
Conclusion
6-41
University of Pretoria etd – Du Plessis H (2006)
7.
TEST RESULTS AND DISCUSSION ON CONCRETE TESTS
7-1
7.1
Introduction
7-1
7.2
Tests Conducted on Fresh Concrete
7-1
7.2.1
7-1
7.3
7.4
7.5
Slump Test
Strength Tests
7-2
7.3.1
Concrete Cube Compression Test Results
7-2
7.3.2
Tensile Strength Results
7-3
Deformation and Volume Change of Concrete
7-4
7.4.1
E-value test results
7-4
7.4.2
Shrinkage and Creep Test
7-6
Durability Tests
7-9
7.5.1
Porosity Test Results
7-9
7.5.2
Oxygen Permeability Test Results
7-11
7.6
Conclusion
7-12
8.
CONCLUSIONS AND RECOMMENDATIONS
8-1
8.1
Conclusions
8-1
8.2
Recommendations
8-3
9.
REFERENCES
9-1
APPENDIX A
Cumulative Particle Size Distribution of Gasification
Ash, Cement and Gasification Ash and Cement
Interground and Interblended
APPENDIX B
Cumulative % Oversize Particle Size Distribution for
Gasification Ash, Cement and Gasification Ash and
Cement Interground and Interblended
APPENDIX C
Rosin-Rammler Distribution Graphs
APPENDIX D
Blaine Surface Area Calculations
APPENDIX E
Mortar Prisms Strength Summary for the Effect of
Grinding Time on the Properties of Interblended
Gasification Ash and Cement
APPENDIX F
Mortar Prisms Strength Summary for the Effect of
Grinding Time on the Properties of Interground
Gasification Ash and Cement
University of Pretoria etd – Du Plessis H (2006)
APPENDIX G
Mortar Prisms Strength Summary for the Effect of
Gypsum Content on the Properties of Interground
Gasification Ash and Cement
APPENDIX H
Mortar Prisms Strength Summary for the Effect of
Replacement Levels on the Properties of Interground
Gasification Ash and Cement
APPENDIX I
Mortar Prisms Strength summary for the Comparison
between Manufactured and Commercial Cement
APPENDIX J
Cube Strength Summary
APPENDIX K
Specific Creep Summary
APPENDIX L
Porosity Summary
APPENDIX M
Permeability Calculations
University of Pretoria etd – Du Plessis H (2006)
LIST OF FIGURES
PAGE
Figure 2.1
Cement manufacture and the influenced aspects
(from www.wbcsdcement.org, 2005)
Figure 2.2
2-3
Changes in Ca2+ in solution curing C3S hydration vs.
time in the absence and presence of gypsum
(Frigione, 1995)
Figure 2.3
Optimization of soluble calcium sulphate (cements)
(Newman, 2003)
Figure 2.4
2-9
Example of normal hydration of Portland cement
(Sandberg, 2004)
Figure 2.5
2-7
2-11
Diagrammatic representation of Rosin-Ramler
distribution function (Wainright and Olorunsogo,
1999)
2-15
Figure 3.1
Coal combustion by-products
3-4
Figure 3.2
A modern pulverised coal-fired thermal power station
(from South African Coal Ash Association, 1999)
3-6
Figure 3.3
Electron microscope photograph of fly ash.
3-7
Figure 3.4
Relationship between water requirement and cement
replacement with fly ash (Naik, 1990)
Figure 3.5
3-11
Development of compressive strengths of Portland
cement and fly ash concretes (from South African
Coal Ash Association, 1999)
Figure 3.6
Effect of coarse fractions of fly ash on compressive
strength development of concretes (Joshi, 1982)
Figure 3.7
3-13
3-14
Compressive strengths of PC and PC/30FA at different
temperatures (South African Coal Ash Association,
1999)
Figure 3.8
3-15
The influence of FA content of the cementitious
material on the flexural/compressive strength ratio of
concrete (from South African Coal Ash Association,
1999)
Figure 3.9
3-16
The influence of FA on the elastic modulus/ compressive
strength of concrete (from South African Coal Ash
Association, 1999)
3-17
University of Pretoria etd – Du Plessis H (2006)
Figure 3.10
Drying shrinkage of concrete incorporating fly ash
(Yuan, 1983)
3-18
Figure 4.1
Laboratory ball mill used in experiment
4-2
Figure 4.2
Steel balls used for grinding
4-2
Figure 4.3
Cumulative particle size of steel balls
4-3
Figure 4.4
Sample of gasification ash clinker
4-4
Figure 4.5
Exponential fit for cumulative % oversize particle
size distribution
4-5
Figure 4.6
Rosin-Rammler distribution graph
4-6
Figure 4.7
Explanation of Rosin-Rammler distribution parameters
4-6
Figure 4.8
Cumulative % oversize distribution parameters
4-7
Figure 4.9
Cumulative % oversize intervals for 3µm and 30µm
4-8
Figure 4.10
Apparatus to determine specific surface area
4-9
Figure 4.11
Photo of sample on flow table after mould is removed
4-14
Figure 4.12
Photo of diameter measurements of flow table test
4-15
Figure 4.13
Photo of mould used
4-15
Figure 4.14
Le Chatelier moulds in humidity cabinet
4-17
Figure 5.1
Measuring the slump (Addis,2001)
5-3
Figure 5.2
Photo of lab set-up for measuring shrinkage and
Creep
5-7
Figure 5.3
Photo of porosity test set up
5-9
Figure 6.1
Graph of cumulative particle size distribution of
cement and gasification ash (grinded separate and
interground)
Figure 6.2
Cumulative particle sizes for gasification ash and
cement interground
Figure 6.3
6-2
6-3
Summary of exponential fitted functions for the
cumulative % oversize particle size distribution
6-3
Figure 6.4
Summary of Rosin-Rammler distributions
6-4
Figure 6.5
Relation between grinding time and position
parameter Xo
Figure 6.6
Figure 6.7
6-7
Relationship between grinding time and slope
parameter (n)
6-8
Scanning electron microscope photo of fly ash
6-10
University of Pretoria etd – Du Plessis H (2006)
Figure 6.8
Scanning electron microscope photo of gasification
ash
6-11
Figure 6.9
Graph indicating Blaine surface area
6-11
Figure 6.10
XRD results for gasification ash
6-14
Figure 6.11
Compressive strengths for interblending gasification
ash and cement
Figure 6.12
Flexural strengths for interblending gasification ash
and cement
Figure 6.13
6-16
6-18
Relation between compressive strengths and
Rosin-Rammler distribution position parameter (Xo)
for interblending gasification ash and cement
Figure 6.14
6-19
Relation between compression strengths and
Rosin-Rammler distribution slope (n) parameter for
interblending gasification ash and cement
Figure 6.15
6-20
Relation between compression strength and particle
size distribution parameters for interblending
gasification ash and cement
Figure 6.16
Compressive Strength for intergrinding gasification
ash and cement
Figure 6.17
6-22
Flexural strengths for intergrinding gasification ash
and cement
Figure 6.18
6-21
6-24
Relation compressive strengths and Rosin-Rammler
distribution position parameter (Xo) for intergrinding
gasification ash and cement
Figure 6.19
Relation between 28-day compression strengths and
Rosin-Rammler distribution slope (n) parameter
Figure 6.20
Figure 6.21
6-25
6-26
Relation between 28-day compression strength and
particle size distribution parameters
6-27
Compressive strengths for gypsum content
6-28
University of Pretoria etd – Du Plessis H (2006)
Figure 6.22
Flexural strengths for gypsum content
Figure 6.23
Rate of heat development for different gypsum
percentages
Figure 6.24
6-29
6-30
The difference in the rate of heat evolution for
different gypsum percentages cement and pure PC
cement
Figure 6.25
Total heat of hydration of different gypsum percentage
cements at 25ºC
Figure 6.26
6-33
Flexural strengths for different replacement levels of
gasification ash
Figure 6.28
6-32
Compressive strength for different replacement
levels of gasification ash
Figure 6.27
6-31
6-34
Relation compressive strengths and Rosin-Rammler
distribution position parameter (Xo) for replacement
levels of gasification ash
Figure 6.29
Relation between 28-day compression strengths and
Rosin-Rammler distribution slope (n) parameter
Figure 6.30
6-36
6-36
Relation between 28-day compression strength and
particle size distribution parameters of replacement
level of gasification ash
Figure 6.31
Compressive strengths for manufactured and
commercially available cement
Figure 6.32
6-38
Flexural strengths for manufactured and
commercially available cement
Figure 6.33
6-37
6-40
Rate of heat development for manufactured and
commercially available cement
6-41
Figure 7.1
Slump test for concrete mixes
7-2
Figure 7.2
Concrete cubes compression strength results
7-3
Figure 7.3
Tensile strength results for concrete mixes
7-4
Figure 7.4
E-value test results for concrete mixes
7-5
Figure 7.5
Shrinkage and creep results for interground
gasification ash mix
Figure 7.6
7-7
Shrinkage and creep results for interblended fly
ash mix
7-8
University of Pretoria etd – Du Plessis H (2006)
Figure 7.7
Shrinkage and creep results for interblended
gasification ash mix
7-8
Figure 7.8
Specific creep results for the three different mixes
7-9
Figure 7.7
Porosity results of concrete mixes
7-10
Figure 7.8
Oxygen permeability test results for concrete mixes
7-11
Figure 7.9
Oxygen permeability index results for the concrete
mixes
7-12
University of Pretoria etd – Du Plessis H (2006)
LIST OF TABLES
PAGE
Table 2.1
Common Portland and Portland fly ash cements
(from SANS 50197-1/SABS EN 197-1:20001)
Table 2.2
2-4
Mechanical and Physical requirements of cement
(from SANS 50197-1/SABS EN 197-1:2000)
2-5
Table 3.1
Typical chemical composition of fly ash (Addis, 2001)
3-8
Table 3.2
Chemical specifications for cement extenders.
(SANS 1491-2:2005 / SABS 1941-2:2005)
3-9
Table 4.1
Mix composition for mortar prisms
4-13
Table 4.8
Mix composition used in test
4-13
Table 4.3
Mortar prism mix composition for interblending cement
and Gasification ash
Table 4.2
4-19
Mortar prism mix composition for intergrinding cement
and Gasification ash
4-19
Table 4.5
Gypsum replacement weights for mortar prisms
4-19
Table 5.1
Mix composition for concrete mixes
5-2
Table 5.2
Concrete mix composition
5-2
Table 6.1
Fitted functions of oversize particle size distribution
6-5
Table 6.2
Rosin-Rammler particle size distribution parameters
6-6
Table 6.3
Oversize particle size distribution parameters
6-9
Table 6.4
XRF results
6-13
Table 6.5
Chemical test results
6-15
Table 6.6
Strength classes of interblending mixes grinding time
6-17
Table 6.7
Percentage of compressive strength achieved for
interblending
6-18
Table 6.8
Strength classes for intergrinding mixes grinding time
6-23
Table 6.9
Percentage of compressive strength achieved for
intergrinding
6-24
Table 6.10 Strength classes for different replacement levels of
gasification ash
6-34
University of Pretoria etd – Du Plessis H (2006)
Table 6.11 Percentage of compressive strength achieved for different
replacement levels of gasification ash
6-35
Table 6.12 Strength classes for different replacement levels of
gasification ash
6-39
Table 6.13 Percentage of compressive strength achieved for different
replacement levels of gasification ash
6-40
Table 7.1
Table comparing tensile and compressive strengths
7-4
Table 7.2
E-value results of the different cylinders for the
different mixes
Table 7.3
7-6
Shrinkage, Creep and Specific Creep Results for the
different mixes
7-6
University of Pretoria etd – Du Plessis H (2006)
1-1
1.
1.1
INTRODUCTION
BACKGROUND
In recent years there has been a significant increase in the use
of waste materials as both cement extenders and fillers in
concrete (Escalante, 2004).
awareness
of
the
The reason for this is a growing
engineering,
economical
and
ecological
benefits, the use of waste materials have in the cement and
concrete industries. Waste materials can however only be used
in concrete if they are not detrimental to the short- or the longterm properties of the concrete.
The main reason for the use of extenders in concrete is their
variety
of
concrete
useful
enhancements
properties.
Cement
of
or
modifications
extenders
have
two
to
the
common
features:
•
their particle size range is similar to or smaller than that
of Portland cement.
•
they become involved in the hydration reactions.
In blended cement production, extenders can be introduced to
the cement by interblending or intergrinding (Erdogdu, 1999).
The feed to Sasol gasifiers principally consists of coarse coal
(>5mm) and extraneous rock fragments (stone). During the
gasification of this coarse coal at elevated temperatures and
pressure a mixture of carbon monoxide and hydrogen (also
referred to as synthesis gas) is produced. The coarse ash is
formed at these elevated temperatures and pressure by the
interaction of inert minerals present in the coal and stone. The
coarse ash is removed from the gasifier and disposed as a byproduct (Van Dyk, 2005).
University of Pretoria etd – Du Plessis H (2006)
1-2
1.2
OBJECTIVES OF THE STUDY
The
main
aim
of
this
investigation
is
to
determine
the
suitability of using gasification ash as cement extender in
concrete. To meet this aim, the objectives of the investigation
are as follows:
•
To investigate the properties of cement manufacturing with
specific reference to cement classes, hydration of cement,
and the optimisation of gypsum, specific surface area and
particle size distribution, to evaluate the performance of
cement produced in the laboratory.
•
Investigate
the
physical,
chemical
and
mineralogical
composition of a gasification ash sample to determine if
gasification
ash
can
be
used
as
a
cement
extender
in
concrete.
•
Determine
the
compliance
of
gasification
ash
with
the
requirements for use in cement.
•
Determine
an
interblending
optimum
and
grinding
intergrinding
of
time
for
both
the
ash
and
gasification
cement.
•
Determine the effect of replacement level on the properties
of interblended and interground gasification ash and cement.
•
Investigate
the
effect
of
interblended
and
interground
gasification ash and cement on the short and long term
properties of concrete.
•
Investigate
the
effect
of
the
gasification
ash
on
the
durability of concrete.
1.3
SCOPE OF THE STUDY
This study will consist of a literature review on the manufacture
and properties of cement. Detail of different coal ash byproducts is discussed and the use of fly ash as a cement
extender in concrete is considered. Consideration is given to
properties like mineralogical and chemical composition, shape,
University of Pretoria etd – Du Plessis H (2006)
1-3
particle size distributions, chemical requirements of using fly
ash
in
concrete,
strength
development
and
durability
of
concrete.
A gasification ash sample is tested to determine how this waste
material
compares
to
the
extenders
currently
used
in
the
cement industry. Consideration is given to the chemical and
physical properties of gasification ash. All data is analysed to
achieve conclusions on the above-mentioned objectives.
Cement
is
manufactured
by
grinding
clinker.
Particle
size
distributions are investigated and optimum grinding times is
investigated. The optimisation of gypsum is also investigated
with
reference
replacement
to
levels
the
of
hydration
gasification
of
ash
cement.
are
Different
investigated
to
determine restriction on the use of gasification ash in cement.
Gasification
development,
ash
is
used
deformation
in
concrete
and
and
durability
of
the
strength
these
concrete
mixes are examined. This will provide adequate understanding
of the advantages that waste materials have to both the cement
industry and the environment.
To accurately estimate and predict behaviour, it is necessary to
repeat tests on as many independent samples as possible. In
this study the tests were only conducted once. Therefore it is
not statistically correct to draw final conclusions and predict
trends based on the results.
This study does not include the following:
•
The effect of variability of gasification ash as far as
physical and chemical properties is concerned.
•
The response of gasification ash to admixtures.
•
The effect of gasification ash on the long-term durability
of concrete.
•
Durability tests performed excludes the damaging effects
of ultra violet (UV) light and physical scouring.
University of Pretoria etd – Du Plessis H (2006)
1-4
1.4
METHODOLOGY
There were seven principal objectives in this research project.
The methodology followed in each is selected to objectively
evaluate gasification ash for its use in cement and concrete.
•
The physical properties of a gasification ash sample will
be investigated by examining the particle size distribution
and
shape.
This
includes
determining
an
optimum
grinding time and gypsum content.
•
The
chemical
and
mineralogical
composition
of
a
gasification ash sample is investigated and compared to
that
of
extenders
commonly
used
that
are
known
to
enhance the properties of concrete.
•
The compliance of a gasification ash with the chemical
requirements for use in concrete is determined.
•
The
reactivity
of
gasification
ash
is
established
by
considering interblending and intergrinding in mortar, and
testing against fly ash in concrete.
•
The effect of gasification ash on the short and long term
properties
of
concrete
is
established
by
measuring
workability, strength development, elasticity, shrinkage
and creep, porosity and permeability.
•
In conclusion recommendations are made for the use of
gasification ash.
1.5
ORGANISATION OF THE REPORT
The dissertation has been divided into the following sections:
•
Chapter 1 serves as an introduction to the thesis.
•
Portland cement as well as properties of manufacturing of
cement is discussed in chapter 2.
•
The composition and properties of coal ash are discussed in
Chapter 3, with special reference to the properties of fly ash.
University of Pretoria etd – Du Plessis H (2006)
1-5
•
Chapter 4 describes the experimental programme and test
procedures for cement.
•
Experimental programme and test procedures for concrete
are discussed in Chapter 5.
•
Chapter
6
contains
the
results
for
the
properties
and
utilisation of gasification ash as a replacement of cement in
mortar prisms.
•
The results of using gasification ash as a cement extender
in concrete are evaluated in Chapter 7.
•
Chapter 8 contains the conclusions and recommendations of
the study.
•
A list of references follows in Chapter 9.
•
Various
results
appendices
dissertation.
are
are
provided
referred
to
in
in
the
the
appendices.
main
body
These
of
the
University of Pretoria etd – Du Plessis H (2006)
2-1
2.
COMPONENTS AND PROPERTIES OF PORTLAND
CEMENT
2.1
INTRODUCTION
Global population is rising, placing increasing pressure on
essential natural resources such as land and energy. This
makes
it
imperative
for
us
to
find
ways
of
using
these
resources more efficiently. This need for more environmentally
and
socially
agenda
for
sustainable
development
governments,
has
non-governmental
become
a
key
organisations
(NGO) and businesses.
Cement is an essential material in today's society because, as
a major constituent of concrete, it forms a fundamental element
of any housing or infrastructure development. The chemical
process
of
making
cement
clinker
produces
CO2,
a
major
greenhouse gas contributing to climate change.
In this chapter, detail of the manufacturing of Portland cement
(PC), cement classes and the hydration of cement will be
discussed. Consideration will be given to properties like the
optimisation of sulphate and fineness of cement. The particle
size distribution is discussed with specific reference to the
Rosin-Rammler distribution function. This will provide adequate
understanding of the properties of cement and the influence
blended cements have on the cement hydration process.
2.2
CEMENT MANUFACTURE
SANS 50197-1/SABS EN 197-1:2000 states that “cement is a
hydraulic
binder,
a
finely
ground
inorganic
material
which,
when mixed with water, forms a paste which sets and hardens
by means of hydration reactions and processes and which after
hardening, retains its strength and stability even under water.”
University of Pretoria etd – Du Plessis H (2006)
2-2
Cement is manufactured from four raw material oxides: lime,
silica, alumina and ferric oxide. Lime cannot be found in nature
and is produced by heating calcium carbonate (Addis, 2001).
Cement manufacturing consists of quarrying or excavating raw
materials. Cement manufacturers obtain silica, alumina and
ferric oxides from clay or shale. The raw materials are crushed,
blended, milled and preheated in a kiln. In the kiln, calcium
carbonate is converted to calcium oxide at temperatures of
800°C to 1000°C:
CaCO3 → CaO + CO2
(equation 2.1)
The materials flow to a hotter (1450°C) part of the kiln where,
the blend of the four oxides is converted to cement clinker
(Addis, 2001). The clinker is then ground with a small amount
of gypsum into a powder with a specific surface area (Blaine)
of 300 – 350 m²/kg to make ‘Ordinary Portland Cement’, the
most commonly used type of cement (often referred to as
OPC). A schematic presentation of the cement manufacturing
process can be seen in Figure 2.1.
The CO2 emissions from Portland cement manufacturing are
generated
by
two
mechanisms.
These
mechanisms
are
the
calcining of limestone (see equation 2.1) and the combustion of
fuels to generate energy.
Typically, Portland cement contains
the equivalent of about 63.5 % CaO. Consequently, about 1.135
units of CaCO3 are required to produce 1 unit of cement, and
the amount of CO2 released in the calcining process is about
500 kilograms (kg) per metric ton of Portland cement produced
(1,000 pounds [lb] per ton of cement). Total CO2 emissions
depend on energy consumption and generally fall in the range
of 0.85 to 1.35 metric ton of CO2 per metric ton of clinker
(www.wbcsdcement.org, 2005). This means that the cement
industry produces 5% of global man-made CO2 emissions, of
which 50% is from the chemical process, and 40% from burning
fuel. The remainder is split between electricity and transport
uses. Climate protection, and in particular reduction of carbon
University of Pretoria etd – Du Plessis H (2006)
2-3
dioxide
(CO2)
emissions
is
therefore
an
issue
which
very
seriously needs to be addressed by the cement industry.
Figure 2.1 Cement manufacture and the influenced aspects
(from www.wbcsdcement.org, 2005)
Because climate protection has such a high profile in the
industry, effective strategies for managing CO2 emissions are
of
crucial
importance
in
the
marketplace.
options are likely to include:
The
reduction
innovation in improving the
energy efficiency of processes and equipment; switching to
lower carbon fuels; using alternative raw materials to reduce
limestone
use;
developing
CO2
capture
and
sequestration
techniques; and taking advantage of market mechanisms such
as emissions trading and voluntary initiatives.
The use of waste materials decrease the amount of clinker
manufactured
emissions.
and
thus
decreases
the
amount
of
CO2
University of Pretoria etd – Du Plessis H (2006)
2-4
2.3
CEMENT CLASSES
Table 2.1 indicates the most common Portland and Portland fly
ash cement used in South Africa.
Table 2.2 indicates the
mechanical and physical requirements of cement used in South
Africa.
These tables are used to describe the commercially
available cement. The classification of cements in terms of
their strength-giving properties has been practised for many
years.
Table 2.1 Common Portland and Portland fly ash cements (from SANS
50197-1/SABS EN 197-1:2000)
Composition (percentage by mass)
Main constituents
Pozzolana
Main
types
CEM I
CEM II
Portland
fly ash
cement
Minor
additio
nal
constit
uents
clinker
natural
natural
calcined
siliceous
calcer
ous
K
P
Q
V
W
CEM I
95-100
-
-
-
-
0-5
CEM II/A-V
80-94
-
-
6-20
-
0-5
CEM II/B-V
65-79
-
-
21-35
-
0-5
CEM II/A-W
80-94
-
-
-
6-20
0-5
CEM II/B-W
65-79
-
-
-
21-35
0-5
Notation of the
products
Portland
cement
Fly ash
University of Pretoria etd – Du Plessis H (2006)
2-5
Table 2.2 Mechanical and Physical requirements of cement (from SANS
50197-1/SABS EN 197-1:2000)
Compressive strength MPa
Strength
Class
Early Strength
2 Days
7 Days
32.5 N
-
≥16.0
32.5 R
≥ 10.0
-
42.5 N
≥ 10.0
-
42.5 R
≥ 20.0
-
52.5 N
≥ 20.0
-
52.5 R
≥ 30.0
-
2.4
Initial
setting
time
Standard Strength
28 Days
≥ 32.5
≤ 52.5
≥ 75
≥ 42.5
≤ 62.5
≥ 60
≥ 52.5
-
≥ 45
Soundness
(expansion)
mm
≤ 10
HYDRATION OF PORTLAND CEMENT
The setting and hardening of Portland cement occur as a result
of the reaction between the compounds of cement and water.
The
major
produce
dicalcium
compounds
reaction
of
cement
products
silicate
(C2S),
are
that
react
tricalcium
tricalcium
with
water
silicate
aluminate
to
(C3S),
(C3A),
and
tetracalcium aluminoferrite (C4AF) (Neville, 1995).
The presence of gypsum affects the hydration pattern of PC
both kinetically and thermodynamically. Firstly the influence of
gypsum on the principal individual constituents is discussed
and then the influence of the reactions as a whole which occur
with PC hydration.
2.4.1
The hydration of C3S and C2S
The hardening of cement paste is due mainly to the formation
of
Calcium
silicate
hydrate
(CSH),
and
this
formation
is
affected by the presence of CaSO4 in the paste. The gypsum in
cement not only affects the setting time, but also the strength
development. The optimum gypsum content is the value at
which the optimum combination of quantity and quality occurs.
University of Pretoria etd – Du Plessis H (2006)
2-6
Locher et al (1995), have discussed factors governing the
optimum gypsum content and the effects of varying the source
of the sulphate. The situation is complicated by the fact that,
contrary to some early conclusions, the amounts needed to
optimise different properties, such as strength at various ages
and drying shrinkage are not necessarily the same; also, the
amount needed to optimise a given property in a concrete may
not be the same as that required in a paste or mortar (Tang
and Gartner, 1988).
During the early and middle periods of reaction in a cement
paste, gypsum dissolve and react at or close to the surfaces of
the clinker grains, more specifically the aluminate and ferrite
phases. The factor most directly influencing the course of the
early reactions is not so much the relative amounts of calcium
sulphate, aluminate and ferrite phases, as the rate at which the
relevant ionic species are made available at the surface of the
cement grains. Other major factors affecting the supply of
these ions are the particle size distribution of the calcium
sulphate and the distribution in space of the particles (Taylor,
1995).
Frigione
(1995)
has,
by
measuring
the
Ca2+
concentration of the liquid phase shown that the presence of
gypsum increased the C3S hydration rate, Figure 2.2.
The calcium silicate hydrate reactions are the most important
as these are responsible for the majority of strength of the
hardened cement paste (hcp). The hydration products of the
two calcium silicates are similar and differ only in the terms of
the rate at which they occur and in the amount of calcium
hydroxide formed as seen in equation 2.3 and 2.4. The reaction
of the C3S with water is the more rapid of the two.
2C3S + 6H
Alite
→ C3S2H3
+ 3CH
(equation 2.3)
University of Pretoria etd – Du Plessis H (2006)
2-7
Figure 2.2 Changes in Ca2+ in solution curing C3S hydration vs.
time in the absence and presence of gypsum (Frigione, 1995)
The C2S reaction is similar, but it takes place at a slower rate,
contributes less heat than the C3S and is responsible for the
later strength development.
2C2S +
4H
→ C3S2H3
+ CH
(equation 2.4)
Belite
When
water
is
added
to
the
reactions
above,
the
calcium
hydroxide ions are rapidly released into the solution and the
pH of the solution rises rapidly to over 12. During this stage a
considerable amount of heat is evolved.
While the concentration of calcium and hydroxide ions are
building up, a dormant period occurs which coincides with the
ettringite delay from the C3A reaction and helps to explain why
paste
can
remain
workable
for
so
long.
Once
the
critical
concentration of the ions is reached the CSH and CH start to
University of Pretoria etd – Du Plessis H (2006)
2-8
crystallise out from the solution, strength is developed and
workability is lost.
2.4.2
Hydration of C3A
The
reaction
C3A
of
with
water
is
very
fast
and
involves
reactions with sulphate ions supplied by the dissolution of
gypsum (see Equations 2.5 and 2.6).
C3A +
3CSH2 + 26H
→
C3A.3CSH32
(equation 2.5)
Where:
H
- Water
CSH
-
Calcium sulphate (Gypsum)
C3A.3CSH32
- Calcium Sulphate Aluminate Hydrate (Ettringite)
The formation of ettringite slows down the hydration of C3A by
creating a barrier around the cement grains.
portion
of
the
sulphate
has
been
After a certain
consumed,
the
ettringite
becomes unstable and a second reaction begins to take place
(Wainwright, 2004).
Ettringite
C3A
+
CSH + 10H
→
C3A.CSH12
(equation 2.6)
Monosulphate
It is not until this second reaction occurs that the paste begins
to stiffen significantly and workability begins to drop.
The
overall C3A reaction produces a significant quantity of heat but
contribute little to the strength.
An inadequate supply of soluble calcium sulphate can result in
a
rapid
loss
accompanied
of
by
workability
the
release
known
of
as
heat
flash
and
is
set.
This
is
irreversible.
However, if too high a level of gypsum is present; crystals of
gypsum crystallize from the solution and cause a plaster or
false set (cement). If mixing continues or is resumed, the initial
level or workability is restored. The cement manufacturer thus
University of Pretoria etd – Du Plessis H (2006)
2-9
needs to optimize the level of gypsum in the cement and match
this to the reactivity of C3A present. This concept is illustrated
in Figure 2.3.
Figure 2.3 Optimization of soluble calcium sulphate (cements)
(Newman, 2003)
2.4.3
Hydration of C4AF
The ferrite phase (C4AF) is of no great importance compared to
the others. It is a slow reaction (see equation 2.7) with little
heat evolved and contributes little to the strength.
C4AF
+
10H
+
→
2CH
C6AFH12
(equation 2.7)
Sulphoferrite
2.5
HEAT OF HYDRATION
The chemical reactions between unhydrated cement and water
during setting and hardening release heat which results in a
rise
in
temperature
of
the
fresh
concrete.
If
optimum
performance is to be achieved from the cement, it is thought
that the peak hydration rate associated with C3A should be
University of Pretoria etd – Du Plessis H (2006)
2-10
delayed by SO3 additions until the silicate hydration rate has
passed its peak.
2.5.1
Optimization of Cement Sulphate
Neville, (1995) suggested that the optimum gypsum content be
determined
by
observation
hydration.
Optimum
development
and
depletion
soluble
of
of
the
sulphate
dimensional
generation
with
respect
stability
sulphate
used
up
of
to
occurred
by
heat
strength
when
the
of
the
aluminate
hydration occurred at a time later than the maximum heat
evolution from the main silicate hydration peak (Sandberg,
2004). Figure 2.4 shows an example of hydrating PC with
slightly higher than optimum sulphate content monitored at
room temperature by an isothermal conduction calorimeter.
The initial peak in Figure 2.4 occurs as soon as water is added
to the PC, resulting from the ettringite being formed in the C3A
reaction. The calcium hydroxide ions passing into solution in
the cement mixed with water initially display strong exothermic
behaviour by rapid dissolution and initial hydration of mainly
the aluminate phase as seen at A in Figure 2.4.
If sufficient
sulfate is available in the solution, the hydration rate rapidly
decreases as the aluminate reacts with calcium and sulfate to
form ettringite (B on the curve). The formation of ettringite
prevents flash set and allows the concrete to be transported
and placed while it is still fluid. After some time the strength
giving
Alite
hydration
takes
off,
which
results
in
a
broad
exotherm C. Set usually occurs at the initial part of the Alite
exotherm.
The
Alite
and
aluminate
hydration
continue
in
parallel until the mixture runs out of soluble sulphate (at D),
which initiates the formation of aluminates with less sulfate
than ettringite (Sandberg, 2004).
University of Pretoria etd – Du Plessis H (2006)
2-11
Figure 2.4 Example of normal hydration of Portland cement
(Sandberg, 2004)
After the water has been added to the cement the first peak in
the rate of heat evolution is followed by a second peak some 4
to 10 hours later. With the correct amount of gypsum there
should be little C3A available for reaction after all the gypsum
has combined and no further peak in the heat liberation should
occur. Thus optimum gypsum content leads to a desirable rate
of
early
reaction
and
prevents
local
high
concentration
of
products of hydration (Neville, 1995). In consequence the size
of pores in hydrated cement paste is reduced and strength is
increased. The amount of gypsum required increases with the
C3A content and also with the alkali content of the cement.
Increasing the fineness of cement has the effect of increasing
the quantity of C3A available at early stages, and this raises
the gypsum requirement.
The amount of gypsum added to cement clinker is expressed as
the mass of SO3 present; this is limited by the SANS 501971/SABS EN 197-1:2000 to a maximum of 3.5%. There is usually
found to be an optimum SO3 content (2-3% for binders), beyond
which (> 4%), compressive strength begins to decline (Lea,
1998).
Frigione
confirmed
this
statement
using
ISO-RILEM
mortar prisms, for cements with a wide range of particle size
University of Pretoria etd – Du Plessis H (2006)
2-12
distributions, gypsum was added to preground clinker, and
found
that
particle
(1946)
the
size
test
maximum
grading
results
in
strength
(Frigione
indicate
and
that
was
Mara,
2.5%
not
sensitive
1976).
added
SO3
to
Lerch’s,
may
be
sufficient to bring the resulting PC to optimum SO3 level.
2.6
SPECIFIC SURFACE AREA
At most manufacturing plants a ball mill is used to grind the
cement clinker. The principal test carried out at a cement mill
is
the
fineness
determined.
The
test
in
which
specific
the specific
gravity
and
surface area
fineness
of
is
cement
increase with an increase in the grinding time. Bouzoubaâ et al
(1997) found that this increase in fineness was less significant
beyond 2 hours. An optimum of 4 hours grinding time was
established by Bouzoubaâ, beyond which the water requirement
increased and the strength either decreased or did not increase
significantly.
The fineness of cement is a major factor influencing its rate of
hydration, since the hydration reaction occurs at the interface
with water (Lea, 1988). Greater cement fineness increases the
rate at which cement hydrates and thus accelerates strength
development. The effects of greater fineness on paste strength
are manifested principally during the first seven days.
Portland cement is usually ground to a surface area in the
range 300 – 350 m2/kg and rapid hardening Portland cement to
400 – 550 m2/kg.
2.7
PARTICLE SIZE DISTRIBUTION
The particle size distribution curve of a material describes two
properties of the material namely mean particle size as well as
the distribution of other sizes about the mean.
The curve is
usually drawn as the cumulative percentage-values on the yaxis of particles smaller than the corresponding sizes on the x-
University of Pretoria etd – Du Plessis H (2006)
2-13
axis.
The x-axis is drawn to a log scale to accommodate large
ranges of particle sizes.
The shape of the curve gives an
indication of the continuity in size distribution and the slope
describes the wideness or range of the size distribution.
Particle size distributions of interground blended cements are
different than that of separately ground cements (Erdogdu,
1999). The particle size distribution produced when a material
is ground becomes wider as the material becomes easier to
grind.
During
differing
the
intergrinding
grindability,
the
of
particle
cement
size
constituents
distribution
of
of
the
material which is harder to grind becomes narrower, the easier
the other component is to grind. Conversely, the particle size
distribution of the material which is easier to grind becomes
wider the harder the other component is to grind (Lea, 1988).
Approximately 95% of cement particles are smaller than 45µm,
with the average particle around 15µm. Bye, (1999) supported
the generally held view that the 3-30µm fraction makes a major
contribution
to
the
28-day
strength.
The
range
<3µm
is
important for achieving high 1-day strengths.
2.7.1
Rosin-Rammler distribution function
In
searching
for
a
parameter,
which
will
provide
a
more
representative description of the particle size distribution, the
Rosin-Rammler function was investigated.
From
a
probability
point
of
view
Rosin
and
Rammler
investigated the particle size distribution of crushed coal and
developed a function that describes the distribution as (Rosin,
1933):
ƒ(x) = exp(-bxn)
Where:
(equation 2.8)
University of Pretoria etd – Du Plessis H (2006)
2-14
b
-
Fineness
characteristic
measure
of
the
material
being
analysed.
n - A measure of the range of particle sizes.
Rosin and Rammler also found that the function does not only
apply
to
crushed
coal
but
also
to
various
other
powered
materials.
The function was modified as follows:
RR = exp –(x/Xo)n
(equation 2.9)
Where:
The weighting function ƒ(x) is denoted as RR.
Xo
–
the
absolute
size
constant
or
position
parameter
(it
represents the particle size for which 36.8% of the particles
are coarser).
Taking the double logarithm of equation 2.9 we obtain:
lnln(1/RR) = n(lnx-lnXo)
(equation 2.10)
Equation 2.10 describes a straight line plot with a coordinate
system made up of a log scale abscissa for the particle size x,
and an ordinate with a double logarithm of the reciprocal of the
residue RR. The slope of the straight line is n and the line
intercept the horizontal axis at a value describing the particle
size x (Olorunsogo, 1990).
A hypothetical example of the
diagrammatic representation of the particle size distribution by
the Rosin-Rammler distribution function is shown in Figure 2.5.
University of Pretoria etd – Du Plessis H (2006)
2-15
Figure
2.5
Diagrammatic
representation
of
Rosin-Ramler
distribution function (Wainwright and Olorunsogo, 1999)
Figure 2.5 (a) shows the typical particle size distribution of
four different samples having the same n and different Xo.
This illustration shows that the sample represented by the four
plots might have similar ranges of size distribution (denoted by
equal
n)
but
with
possible
varying
degrees
of
fineness
(indicated by the various Xo). The sample represented by plot 1
being the finest and the one by plot 4 the coarsest of the four.
Similarly, Figure 2.5 (b) illustrates a situation whereby the four
samples might be of the same fineness (because of the same
Xo) but possibly with different size ranges (Wainwright and
Olorunsogo, 1999).
2.8
CONCLUSION
•
Test
results
indicate
that
2.5%
added
SO3
may
be
sufficient to bring the resulting PC to optimum SO3 level.
•
The specific gravity and fineness of cement increase with
an increase in the grinding time.
University of Pretoria etd – Du Plessis H (2006)
2-16
•
An optimum of 4 hours grinding time was established
beyond which the water requirement increase and the
strength either decrease or did not increase significantly.
•
Greater
cement
cement
fineness
hydrates
and
increases
thus
the
rate
accelerates
at
which
strength
development.
•
The particle size distribution produced when a material is
ground becomes wider as the material becomes easier to
grind.
•
Approximately 95% of cement particles are smaller than
45µm, with the average particle around 15µm.
•
The 3-30µm fraction makes a major contribution to the
28-day
strength.
The
range
<3µm
is
important
for
achieving high 1-day strengths.
•
The Rosin-Rammler distribution function can be evaluated
as a method providing an easy means of describing the
particle size distribution quantitatively.
University of Pretoria etd – Du Plessis H (2006)
3-1
3.
COMPOSITION AND PROPERTIES OF COAL ASH
3.1
INTRODUCTION
Sustainable development can be defined, as development, which
meets the needs of people living today without compromising the
ability of future generations to meet their own needs. It requires
a
long-term
vision
of
industrial
progress,
preserving
the
foundations upon which human quality of life depends: respect
for basic human need and local and global ecosystems.
Using by-products from other industries as raw material is a
huge
opportunity
for
the
cement
industry
to
reduce
its
environmental impact, because it allows companies to access
materials for use in the kiln and the mill without extracting them
directly from the ground. There are a number of mineral byproducts that contain useful materials that can be extracted for
use
in
cement
production,
or
in
making
concrete.
Typical
additives include slag, coal ash like fly ash and bottom ash, byproducts from blast furnaces and power generation.
Cement, a fine grey powder, sets after a few hours when mixed
with water, and then hardens in a few days into a solid, strong
material. Virtually all the cement produced globally is mixed with
sand, aggregates and water and used to make concrete and
mortars.
Cement extenders are used as a substitute for some of the PC in
concrete. The main reason for the use of extenders is the variety
of
useful
enhancements,
which
they
give
to
the
concrete
properties. Fly ash is one of the cement extenders commonly
used in South Africa. Fly ash consists of finely divided ashes
produced
by
Gasification
burning
ash
is
a
pulverised
produced
as
coal
a
in
power
by-product
stations.
during
the
gasification of coarse coal. The properties of fly ash can be
used as a guideline to investigate other ashes for use as cement
extenders in concrete.
University of Pretoria etd – Du Plessis H (2006)
3-2
In
this
chapter,
the
use
of
coal
combustion
by-products,
specifically fly ash, in concrete will be discussed. Consideration
will
be
given
mineralogical
to
and
properties
chemical
like
shape,
composition,
particle
durability
and
size,
the
chemical requirements for using fly ash as a cement extender.
This will provide adequate understanding of the advantages that
by-products
have
to
both
the
cement
industry
and
the
environment.
3.2
COAL ASH
Currently, close to a billion tons of coal is burned annually in
the world to generate electricity and as a result, nearly 130
million
tons
of
coal
combustion
by-products
(CCBs)
are
produced. One-third of these CCBs are utilized, while the rest
are disposed of mainly in landfills (Schwartz, 2003). Increasing
costs and heightened regulations are making the disposal of
CCBs
an
undesirable
option.
Utilization
of
CCBs
as
raw
materials results in numerous benefits, including:
•
A decrease in the demand for landfill space
•
Conservation of natural resources
•
A cleaner safer environment
•
Reduced carbon dioxide emissions
•
Significant economic savings for end users
•
A boost in economic development
•
Reduced overall cost of generating electricity
Electricity is the fuel of the “Information age” and power plants
that burn coal account for more than half of the electricity
produced. These power plants also produce coal combustion byproducts like fly ash (which is capture from the exhaust of the
boiler) and bottom ash (which is heavier and falls to the bottom
of the boiler). CCBs are considered to be four distinct and
University of Pretoria etd – Du Plessis H (2006)
3-3
extremely different materials, as seen in Figure 3.1 (American
Coal Ash Association, 1997).
Coal
bottom ash
and
boiler
slag
are
the
coarse,
granular,
incombustible by-products that are collected from the bottom of
furnaces that burn coal for the generation of steam and the
production of electricity.
Bottom ash is a dark gray, granular,
porous, predominantly sand size (-12.7mm) material (Babcock,
1978). Material drops into a water filled hopper at the bottom of
the furnace. The material is removed by means of high-pressure
water jets and conveyed by sluiceways either to a disposal pond
or to a decant basin. From here the material is dewatered,
crushed and stockpiled for disposal or use (Hect, 1975).
Bottom ash applications are snow and ice control, as aggregate
in lightweight concrete masonry units, and raw feed material for
the production of PC. Bottom ash has also been used as a road
base
and
subbase
aggregate,
structural
fill
material
(ASTM
E1861-97), and as fine aggregate in asphalt paving.
Flue gas desulphurisation (FGD) gypsum is also known as
scrubber gypsum. FGD gypsum is the by-product of an air
pollution control system that removes sulphur from the flue gas
in
calcium-based
scrubbing
systems.
It
is
produced
by
employing forced oxidation in the scrubber and is composed
mostly of calcium sulphate.
The
majority
of
FDG-produced
gypsum
used
in
the
United
States (American Coal Ash Association, 1997) is employed for
wallboard, which reduced the need for mining natural gypsum.
As an additive in PC, FDG gypsum is used to retard setting.
This enables wet cement in ready-mix trucks to be transported
greater distances while remaining workable. Gypsum’s high
permeability
(10-3cm/sec)
makes
it
an
excellent
soil
conditioner. Steffan (1991) indicates that FGD gypsum offers
several major benefits as a soil amendment. These include
University of Pretoria etd – Du Plessis H (2006)
3-4
adjustment of soil pH and means of keeping peanuts and other
crops disease-free.
Figure 3.1 Coal combustion by-products
The fourth CCB is fly ash which is most often used in concrete
as a replacement for part of the PC in the mix design. The use
of fly ash as an extender in concrete will be discussed later in
this chapter.
3.3
POZZOLANIC REACTION
The American Society for Testing Materials (ASTM) defines a
pozzolan as ‘a siliceous or siliceous and aluminous material,
which in itself possesses little or no cementitious property but
which
will,
in
finely
divided
form
and
in
the
presence
of
moisture, chemically react with calcium hydroxide at ordinary
temperature
to
form
compounds
possessing
cementing
properties’.
As a pozzolanic material fly ash contains active silica (SiO2),
and is not cementitious in itself but will chemically react with
calcium
hydroxide
cementitious
in
compounds
the
presence
(Illston,
of
2001).
moisture
When
a
to
form
pozzolanic
material is used in conjunction with a PC, the calcium hydroxide
that takes part in the pozzolanic reaction is that produced from
University of Pretoria etd – Du Plessis H (2006)
3-5
the
cement
hydration.
More
quantities
of
calcium
silicate
hydrate are produced (see equation 3.1).
2S + 3CH → C3S2H3
(equation 3.1)
It is clear that the pozzolanic reaction is secondary to the
hydration of PC. The degree of hydration of fly ash is increased
in the presence of gypsum because the surface is activated by
the destruction of the structure of the glass and crystalline
phases caused by the dissociation of Al2O3 reacting with SO4(Uchikawa, 1986). The products of the pozzolanic reaction make
their own contribution to the strength and other properties of the
cement and concrete.
3.4
FLY ASH
Fly ashes (FA) consist of finely divided ashes produced by
burning pulverised coal in power stations as seen in Figure 3.2.
They are removed from the combustion gases and collected by
special
mechanical
Owing
to
the
devices
high
and
electrostatic
temperatures
precipitators.
reached
during
the
instantaneous burning of coal, most of the mineral component
contained in the coal melts and forms small fused drops. The
subsequent sudden cooling transforms them partly or entirely
into spherical glass particles. The recognition that fly ash
exhibits
pozzolanic
constituent
of
concrete.
aluminosilicates
presence
of
properties
that
moisture
will
to
Fly
react
form
has
led
ashes
with
to
its
contain
calcium
calcium
use
as
a
metastable
ions
silicate
in
the
hydrates
(Massazza, 1998). Fly ash produced in South Africa is a fine
powder, the particles of which are round hollow spheres (see
Figure 3.3).
University of Pretoria etd – Du Plessis H (2006)
3-6
Figure 3.2 A modern pulverised coal-fired thermal power station
(Krüger, 1999).
Fly ash is used in concrete either as part of blended cement or
as a separate component added at the stage when the concrete
mix is prepared. The inclusion of fly ash in concrete affects all
aspects of concrete.
As part of the composite concrete mass,
fly ash acts both as a fine aggregate and as a cementitious
component.
It influences the rheological properties of the fresh
concrete and the strength, finish, porosity, and durability of the
hardened mass as well as the cost and energy consumed in
manufacturing the final product (Massazza, 1998).
3.4.1
Physical Properties
Fly ash particles are mostly spherical in shape (see Figure 3.3)
with sizes ranging from approximately 1 to 100μm in diameter,
with an average size of 20μm (Carette, 1986). The surface area
of fly ash particles vary from 2000 cm2/g to 10 000 cm2/g
depending on the proportion of fine particles in the fly ash.
Erdogdu (1998) found that concrete containing a finer fraction of
fly ash gave a better compressive strength than that without fly
ash or concrete containing coarser fly ash. The particle size
distribution, shape and surface characteristics of fly ash have a
University of Pretoria etd – Du Plessis H (2006)
3-7
considerable influence on the water requirement and workability
of
freshly
made
concrete
and
on
the
rate
of
strength
development in hardened concrete.
Figure 3.3 Electron microscope photograph of fly ash.
3.4.2
Chemical Composition
The
chemical
composition
of
fly
ash
depends
on
the
characteristics and composition of the coal burned in power
stations.
The chemical analysis of fly ashes by means of X-ray
fluorescence (XRF) and spectrometry techniques shows that
SiO2, Al2O3, Fe2O3, and CaO are the major constituents of most
fly ashes. Table 3.1 shows the typical chemical composition of
fly ash.
University of Pretoria etd – Du Plessis H (2006)
3-8
Table 3.1 Typical chemical composition of fly ash (Addis, 2001)
Oxide
3.4.3
SiO2
% By mass
45-50
Al2O3
25-30
CaO
4-8
FeO
9-11
MgO
2-4
Na2O + 0.658K2O
1-3
Mineralogical Composition
Both the type and source of fly ash influence its mineralogical
composition. Due to the rapid cooling of burned coal in the
power plant, fly ash consists of noncrystalline particles (≤ 90%),
or glass and a small amount of crystalline material.
In addition to a substantial amount of glassy material, each fly
ash may contain one or more of the four major crystalline
phases:
quartz,
mullite,
magnetite,
and
hematite.
In
subbituminous fly ashes, the crystalline phases may include C3A,
C3A3S, calcium sulphate, and alkali sulphates (Metha, 1989).
The reactivity of fly ashes is related to the noncrystalline phase
or glass. Diamond (1981) pointed out that the composition of
glass in low-calcium fly ashes is different from that in highcalcium fly ashes. X-ray diffraction (XRD) indicates that South
African fly ash consists mineralogically mainly of glass and
some low-quartz (SiO2), mullite (Al6Si2O13) and some quicklime
(CaO).
3.4.4
Chemical Specifications
In South Africa fly ash should comply with the requirements of
the South African standard specification for Portland cement
extenders (SANS 1491-2:2005 / SABS 1941-2:2005). Fly ash
complying with SANS 149-2 may be used as a cement extender
University of Pretoria etd – Du Plessis H (2006)
3-9
with Portland cement for use in concrete when it conforms to
Table 3.2.
Loss
on
ignition,
the
weight
loss
of
fly
ashes
burned
at
temperatures ≤ 1000°C, is related to the presence of carbonates,
combined water in residual clay minerals, and combustion of
free carbon. The water required for workability of mortars and
concretes depends on the carbon content of fly ashes: the
higher the carbon content of a fly ash, the more water is needed
to produce a paste of normal consistency.
Table 3.2 Chemical specifications for cement extenders. (SANS
1491-2:2005 / SABS 1941-2:2005)
Fly ash max
allowed
Test
Sulphur trioxide content, % (m/m)
2.5
Loss on ignition, % (m/m)
5.0
Free water content, % (m/m)
1.0
Fineness, residue retained on a sieve with
square apertures of nominal size 45μm, % (m/m)
Water requirement, % of control
12.5
95
Strength factor, %
6
Soundness, expansion, mm
5
3.5
INFLUENCE OF FLY ASH ON THE PROPERTIES OF CONCRETE
3.5.1
Fresh Concrete
Although concrete is in the fresh state for only a few hours, the
properties
of
fresh
concrete
are
important
because
they
influence the handling of the concrete, the degree to which it
can
be
compacted
and
the
constituents within the concrete.
uniformity
of
distribution
of
University of Pretoria etd – Du Plessis H (2006)
3-10
3.5.1.1
Water Demand
Davis (1937) noted, fly ash differs from other pozzolans, which
increase the water requirement of concrete mixes. The improved
workability allows a reduction in the amount of water used in
concrete. Partial replacement of OPC by fly ash, in concrete
reduces the water requirement to obtain a given consistency
(Malhotra, 1996). This is generally attributed to the spherical
shape and smooth texture of the fly ash particles.
According to Owens (1979) the major factor influencing the
effects of ash on the workability of concrete is the proportion of
coarse material (>45µm) in the ash. Very fine particles of fly ash
get
absorbed
particles
and
on
the
prevent
oppositely
them
charged
from
surface
flocculation.
of
cement
The
cement
particles are thus effectively dispersed and will not trap large
amounts of water, which means that the system will have a
reduced water requirement for flow. In addition, Portland cement
particles are mostly in the size range of 1 to 50μm, which cause
the micro fine particles of fly ash to reduce the void space and
correspondingly the water requirement.
The
relationship
between
the
amount
of
water
and
the
percentage of fly ash replacement for the same workability with
21, 28 and 34 MPa nominal-strength concrete can be seen in
Figure 3.4. As the amount of fly ash increased in the mixture,
the water requirement decreased (Naik, 1990).
University of Pretoria etd – Du Plessis H (2006)
3-11
Figure 3.4 Relationship between water requirement and cement
replacement with fly ash (Naik, 1990)
3.5.1.2
Workability
The
ACI
Committee
116R-00:2000
has
provided
the
most
suitable definition of workability, which reads as follows:
“that
property
of
freshly
mixed
concrete
or
mortar
which
determines the ease and homogeneity with which it can be
mixed, placed, consolidated and finished.”
The workability (fluidity) of a Portland concrete can be improved
when part of the OPC is replaced with fly ash. Brown (1982)
found
that
both
slump
and
V-B
workability
improved
with
increased ash substitutions. The extent of the improvements
depends on the fineness and carbon content of the fly ash.
Rheological properties and fluidity has been found to depend on
the particle size distribution of cement (Gosh, 1983). Lee (2002)
found that the fluidity of a fly ash-cement system increase as
the particle size distribution becomes wider. When workability is
kept constant the water content of a fly ash mix decreases with
an increase in the fineness of the fly ash.
University of Pretoria etd – Du Plessis H (2006)
3-12
3.5.2
Hardened Concrete
The strength of hardened concrete is of fundamental importance
to structural designers. It is also extensively used as an index
of other properties and of concrete quality.
3.5.2.1
Compressive Strength Development
The measured compressive strength of concrete depends on the
intrinsic properties of the concrete. Many variables influence the
strength development of fly ash concrete, these being:
•
properties of fly ash
•
chemical composition
•
particle size
•
reactivity
•
temperature and curing conditions
At early ages the compressive strengths (up to 28 days) of all
concretes
containing
corresponding
fly
concrete
ash
are
containing
lower
normal
than
that
Portland
of
the
cement.
After 28 days, if wet curing is continued, the compressive
strength of fly ash concrete will be higher than the Portland
concrete (Krüger, 1999). This is demonstrated in Figure 3.5.
University of Pretoria etd – Du Plessis H (2006)
3-13
Figure 3.5 Development of compressive strengths of Portland
cement and fly ash concretes (from Krüger, 1999)
Particle size can influence the strength development in two
ways. Particles larger than 45μm influence water requirements
adversely. They counteract to the needs of the methods used to
compensate for the slow rate of reaction of fly ash at early ages.
Cementing activity occurs on the surface of the solid phases,
through
processes
concentrated
pastes.
involving
Figure
the
3.6
diffusion
shows
that
of
materials
finer
imparted greater compressive strengths (Joshi, 1982).
fly
in
ashes
University of Pretoria etd – Du Plessis H (2006)
3-14
Figure 3.6 Effect of coarse fractions of fly ash on compressive
strength development of concretes (Joshi, 1982)
Concrete
containing
fly
ash
shows
strength
gain
as
a
consequence of heating in contrast to the loss of strength that
occurs
with
normal
Portland
cement
(Malhotra,
1996).
This
property of fly ash is of great value in the construction of mass
concrete or in concrete construction at elevated temperatures.
The effect of curing temperature on the strength development of
concrete containing fly ash can be seen in Figure 3.7.
University of Pretoria etd – Du Plessis H (2006)
3-15
Figure
3.7
Compressive
strengths
of
PC
and
PC/30FA
at
different temperatures (Krüger, 1999).
3.5.2.2
Flexural Strength
There is a general trend for the flexural to compressive strength
ratio to increase with an increase in fly ash content. The results
presented in Figure 3.8, points out that the increase is slight for
water
cement
compressive
(w/c)
strength
ratios
ratio
around
of
0.5.
Portland
The
cement
flexural
to
concrete
is
similar to that of PC/FA concrete of similar strength. Comparing
the flexural to compressive strength ratios of the PC and PC/FA
mortar at similar compressive strength, although not reached at
the same age, the results indicate that the ratio for the PC/FA
mortar is somewhat higher than that for the PC mortar.
University of Pretoria etd – Du Plessis H (2006)
3-16
Figure 3.8 The influence of FA content of the cementitious
material on the flexural/compressive strength ratio of concrete
(Krüger, 1999)
3.5.2.3
Modulus of Elasticity
The modulus of elasticity of a material is defined by the stress:
strain curve. The higher the elastic modulus, the more resistant
the material is to deformation.
There
appears
to
be
no
significant
difference
between
the
modulus of elasticity of concrete with or without fly ash at 28
days. However, like compressive strength, concrete with fly ash
has a lower modulus at early age strength and higher modulus
at ultimate strength compared with concrete without fly ash
(Lane, 1982). The effect of fly ash content on the relationship
between compressive strength and elasticity can be seen in
Figure 3.9.
University of Pretoria etd – Du Plessis H (2006)
3-17
Figure
3.9
The
influence
of
FA
on
the
elastic
modulus/
compressive strength of concrete (Krüger, 1999)
3.5.2.4
Drying Shrinkage
When
a
hydraulic
cement-bonded
product
such
as
concrete
looses its free moisture, it shrinks (drying shrinkage) and when
it gains in moisture content, it expands (wetting expansion). The
dimensional
change
concrete
and
is
with
important
variation
property,
in
moisture
because
if
content
of
differential
dimensional movement in the concrete due to such a change is
excessive,
cracking
may
occur.
If
a
dimensionally
stable
aggregate is used, drying shrinkage and wetting expansion can
almost exclusively be attributed to the binder and is affected by
factors such as binder content, water/cement ratio, curing and
the strength of the concrete (Addis, 2001).
Work by Grieve (1991) on concrete made with a South African
fly ash, showed that for similar exposure conditions, the drying
shrinkage of such concretes is very similar to plain PC, over a
range of fly ash contents up to 30%. This is confirmed by
studies of Yuan (1983) which concluded that the replacement of
cement with fly ash has little influence on drying shrinkage (see
figure 3.10). Chindaprasirt (2003) found that the incorporation
of fly ashes reduce the drying shrinkage in comparison with that
University of Pretoria etd – Du Plessis H (2006)
3-18
of PC. The early shrinkage of mortar with finer fly ash was found
to be a little larger than that of the coarser fly ash mortar.
Figure 3.10 Drying shrinkage of concrete incorporating fly ash
(Yuan, 1983)
3.5.2.5
Creep
Creep is defined as the increase in strain (deformation) under a
sustained
stress
(load)
(Holcim,
2005).
Creep
imparts
to
concrete a degree of ductility, which is desirable from the point
of
view
of
structural
behaviour.
However,
creep
also
has
detrimental effects on structures, such as increase deflections
which can result in cracking, loss of pre-stress and buckling of
long columns (Addis, 2001).
Grieve’s (1991) work on concrete incorporating fly ash found
that specific creep was reduced in fly ash mixes relative to plain
mixes with similar 28-day strengths. This confirms the work
done by Pandey (1983) which found that fly ash concrete’s
showed
less
creep
in
the
reference concrete’s showed.
majority
of
specimens
than
the
University of Pretoria etd – Du Plessis H (2006)
3-19
3.6
DURABILITY OF CONCRETE
Durability may be defined as the ability of concrete to remain
fully functional over an extended period under prevailing service
conditions for the purpose for which it was designed. Concrete
will remain durable if movement of aggressive chemicals within
its structure is minimised.
Degradation of concrete can be a result of the environment to
which the concrete is exposed, or from internal causes within
the concrete. The rate of degradation is controlled by the rate at
which moisture, air or other aggressive agents can penetrate the
concrete. Thus considering the various transport mechanisms
like porosity and permeability through concrete will indicate the
influence on the durability of concrete.
3.6.1
Porosity
Hardened cement pastes and concrete contain pores of varying
types and sizes, and therefore the transport of materials through
concrete can be considered as the phenomenon of flow through
a porous medium.
The rate of flow will not only depend on the
porosity, but on the degree of continuity of the pores and their
size.
A
low
porosity
results
in
high
strengths
and
low
permeability in concrete.
The
pozzolanic
reaction
in
fly
ash
produces
more
calcium
silicate hydrate, which tends to fill pore spaces. Fly ash have a
lower water content than PC for the same workability and as a
result of these factors the porosity of fly ash concrete is lower
then PC.
3.6.2
Permeability
Permeability can be defined as the ease with which a liquid or a
gas flows into (through) concrete under a pressure differential
across the concrete. It is measured by the volume of liquid or
University of Pretoria etd – Du Plessis H (2006)
3-20
gas
transmitted
per
unit
area
per
time
per
unit
pressure
difference.
The major effect on the permeability of concrete is the curing
regime.
Air
exposure
disadvantageous.
concrete.
Water
under
curing
dry
conditions
gives
the
is
most
the
most
impermeable
Another effect on the permeability of concrete is the
stronger the concrete and the lower the water/cement ratio; the
more impermeable the concrete.
Fly ash has the effect of reducing the permeability and volume
of large capillary pores in concrete when compared with a plain
CEM I concrete of similar strength (Balim, 1993). This effect
drives form the fineness of the material, the pozzolanic reaction
and the reduced water requirement of the fly ash mixes. The
fluid transport properties of concrete made with fly ash are
therefore considerably reduced, thus enhancing the durability.
3.7
ADVATAGES OF USING FLY ASH IN CONCRETE
The use of fly ash in concrete has the following advantages:
•
Reduction in material cost and the saving of energy by
saving on cement.
•
Reduction in building cost because of improved workability
of concrete
•
Better workability and cohesiveness
•
Reduced water requirement for a given slump
•
Improved impermeability
•
Reduced water penetration of concrete
•
Reduction in shrinkage
•
Improves
the
environments
durability
of
concrete
in
aggressive
University of Pretoria etd – Du Plessis H (2006)
3-21
3.8
CONCLUSION
•
Fly
ash
is
a
waste
material
of
the
combustion
of
pulverised coal in thermal power plants. Fly ash is used
in concrete for economic, environmental and durability
considerations as a cement extender.
•
The products of the pozzolanic reaction make their own
contribution to the strength and other properties of the
cement and concrete.
•
The
pozzolanic,
chemical,
physical
and
durability
properties of fly ash have led to its use as a cement
extender in concrete. The properties of fly ash can be
used to indicate whether a gasification ash is suitable for
use as a cement extender in concrete.
•
Portland cement particles are mostly in the size range of
1 to 50μm, which cause the micro fine particles of fly ash
to reduce the void space and correspondingly the water
requirement.
•
When workability is kept constant the water content of a
fly ash mix decreases with an increase in the fineness of
the fly ash.
•
Concrete containing fly ash shows strength gain as a
consequence of heating in contras to the loss of strength
that occurs with normal Portland cement.
•
There is a general trend for the flexural to compressive
strength ratio to increase with an increase in fly ash
content.
•
There appears to be no significant difference between the
modulus of elasticity of concrete with or without fly ash at
28 days.
•
The
replacement
of
cement
influence on drying shrinkage.
with
fly
ash
has
little
University of Pretoria etd – Du Plessis H (2006)
3-22
•
Fly ash concrete’s showed less creep in the majority of
specimens than the reference concrete’s showed.
•
Fly ash have a lower water content than PC for the same
workability and as a result of these factors the porosity of
fly ash concrete is lower then PC.
•
The fluid transport properties of concrete made with fly
ash are therefore considerably reduced, thus enhancing
the durability.
•
Consideration
of
mineralogical
properties
and
like
chemical
shape,
particle
composition,
size,
strength,
elasticity, shrinkage and the chemical requirements for
using
fly
ash
as
a
cement
extender
should
be
investigated when using gasification ash as a cement
extender.
University of Pretoria etd – Du Plessis H (2006)
4-1
4.
EXPERIMENTAL PROGRAMME AND TEST PROCEDURES
FOR CEMENT
4.1
INTRODUCTION
The aim of this study is to determine whether gasification ash
can be used as a cement extender in concrete.
Currently
pulverised fuel ash (also called fly ash) is widely used as a
cement extender in concrete and the effect of this type of ash
is well established.
gasification
ash
In this study the properties of cement and
is
examined
to
establish
parameters
for
cement blended with gasification ash.
The aim of this chapter is to discuss the testing methods used
in the practical analysis of the reactivity of a gasification ash.
Parameters like grinding time and optimum gypsum content was
established from the physical and chemical properties of the
gasification
ash
by
performing
tests
for
particle
size
distributions; scanning electron microscopy photo’s and x-ray
analysis.
will
be
Standard test performed for cementitious materials
discussed.
Test
methods
include
the
casting
and
testing of mortar prisms and calorimetry testing.
4.2
PREPARATION OF MATERIALS
In an effort to reduce the environmental impact of cement
production, cement manufacturers are increasing the use of
waste materials to replace a fraction of the cement clinker.
In this project cement is manufactured in the laboratory using
cement clinker obtained from a cement factory. The clinker was
ground in a ball mill (see figure 4.1) with 25kg of round steel
balls.
The
steel
balls
(see
figure
4.2)
were
individually
measured and weighed to determine their size distribution as
indicated in Figure 4.3.
University of Pretoria etd – Du Plessis H (2006)
4-2
Figure 4.1 Laboratory ball mill used in experiment
Figure 4.2 Steel balls used for grinding
All the samples ground were sieved through a 1.18µm sieve
and stored in air-tight containers.
University of Pretoria etd – Du Plessis H (2006)
4-3
Cumulative percentage
passing %
100
80
60
40
20
0
1
10
100
Sieve size, mm
Figure 4.3 Cumulative particle size of steel balls
Cement
and
gasification
ash
is
ground
for
different
time
intervals to establish an optimum grinding time. The grinding
time intervals were; 30 minutes, 1hour, 1hour 30 minutes, 2
hours, 2hour 30 minutes and 4hours. The experiment included
interblending cement and gasification ash after grinding each
material
separately,
and
intergrinding
of
cement
and
gasification ash by interblending the two materials in the ball
mill and grinding it together. No gypsum was added to the
cement clinker in the ball mill.
An optimum grinding time was established by considering the
particle size and the flexural and compressive results of mortar
prisms.
The effect of gypsum on cement and gasification ash was
examined. As gypsum has a direct influence on the setting time
and
heat
of
hydration
an
optimum
gypsum
content
was
established. To determine the optimum gypsum content, heat of
hydration curves were established and isothermal calorimetry
tests
were
performed
replacement levels.
on
samples
with
different
gypsum
University of Pretoria etd – Du Plessis H (2006)
4-4
4.3
PHYSICAL AND CHEMICAL PROPERTIES OF GASIFICATION
ASH
The feed to Sasol gasifiers principally consists of coarse coal
(>5mm) and extraneous rock fragments (stone). During the
gasification of this coarse coal at elevated temperatures and
pressure a mixture of carbon monoxide and hydrogen (also
referred to as synthesis gas) is produced. The coarse (see
figure 4.4) ash is formed at these elevated temperatures and
pressure by the interaction of inert minerals present in the coal
and stone. The coarse ash is removed from the gasifier and
disposed as a by-product (Van Dyk, 2005).
Figure 4.4 Sample of gasification ash clinker
4.3.1
Particle size distribution
Laser
technology
is
used
to
investigate
the
particle
size
distribution of materials after grinding.
4.3.1.1
Particle size distribution parameters
By
plotting
the
inverse
of
the
cumulative
percentage
distribution, the cumulative % oversize particle distribution is
obtained. Provision was made for statistical outliers by not
University of Pretoria etd – Du Plessis H (2006)
4-5
taking the five percent smallest diameters and five percent
largest diameters into account.
For each sample a trend line is added to the cumulative %
Cumulative % oversize
oversize graph (see figure 4.5).
-0.0507x
y = 0.9452e
2
R = 0.9978
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
GA,2hr
Expon. (GA,2hr)
0
20
40
60
Particle size (μm)
Figure 4.5 Exponential fit for cumulative % oversize particle
size distribution
4.3.1.2
Rosin-Rammler particle size distribution parameters
The
cumulative
%
oversize
particle
size
distribution
as
discussed in 4.3.1.1 can be represented as a Rosin-Rammler
distribution.
The Rosin-Rammler distribution graphs (Figure 4.6 and Figure
4.7) are examples of how the values for the particle size
distribution parameters were derived.
The modified Rosin-
Rammler distribution graph lnln(1/y) versus lnx is plotted, with
y the fitted functions y = a exp(bx) for the cumulative %
oversize particle size distribution and x the particle size.
A
linear trend line and equation is also added to these graphs
(see figure 4.6). The slope and interception with the horizontal
axis of the line is taken as the n value and ln Xo value of the
modified Rosin-Rammler function respectively.
University of Pretoria etd – Du Plessis H (2006)
4-6
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9665x - 2.8296
2
R =1
Slope (n) = 0.9665
Ln Xo = 2.952
Position parameter Xo
= 19.14 μm
1
GA,2hr
Linear (GA,2hr)
0.5
0
0
1
2
3
4
5
ln (Particle size(μm))
Figure 4.6 Rosin-Rammler distribution graph
With this analysis 36.8% of the particles (per mass) are greater
than the Xo value (position parameter in μm). This parameter is
an indicator of the particle size.
The n-value represents the
range of the particle size distribution of the particle sizes
greater than Xo (see figure 4.7).
Cumulative % oversize
100%
90%
80%
(n) Slope indicates the range of
the particle sizes greater than Xo
70%
60%
50%
40%
36.8%
30%
20%
10%
0%
0
10
20
Xo
30
40
50
60
Particle size (μm)
Figure
4.7
parameters
Explanation
of
Rosin-Rammler
distribution
University of Pretoria etd – Du Plessis H (2006)
4-7
Particle size distribution parameters
The corresponding 50% oversize particle size distribution in
μm can be read from the exponential fit graph (See figure 4.8).
This particle size gives an indication of the average particle
size of each sample. In the same way the 10% oversize particle
size (D10), which shows the difference in number of larger
particles in each sample.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Cumulative % oversize
4.3.1.3
y = 0.9452e
-0.0507x
2
R = 0.9978
GA,2hr
Expon. (GA,2hr)
0
10 D50
20
30
40 D10
50
60
Particle size (μm)
Figure 4.8 Cumulative % oversize distribution parameters
Figure 4.9 indicates the 3 µm and the 30µm oversize particle size
distributions. This particle size can give an indication of the % of the
particle which lies in the <3µm and 3-30µm intervals, which can
correspond to the values cement manufacturer’s currently use to limit
the fineness of cement.
University of Pretoria etd – Du Plessis H (2006)
4-8
Cumulative % oversize
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
y = 0.9452e
-0.0507x
2
R = 0.9978
GA,2hr
Expon. (GA,2hr)
03 µm
10
20
30
40
50
60
Particle size (μm)
Figure 4.9 Cumulative % oversize intervals for 3µm and 30µm
4.3.2
Specific surface area
A standard method of determining the specific surface area of
cement which is based on the resistance to air flow through a
compact of cement and was developed by Blaine. Figure 4.10
shows the apparatus used to determine specific surface area.
A constant volume method where the time, t, required to pass a
fixed volume of air through a compact bed of cement (12.7 mm
diameter and 15 mm in depth), of standard porosity is related
to the specific surface of cement by the following relationship:
Sw = K√t
Where K= 281.3915 (constant for apparatus)
(equation 4.1)
University of Pretoria etd – Du Plessis H (2006)
4-9
Figure 4.10 Apparatus to determine specific surface area
4.3.3
Scanning electron microscopy (SEM)
A Hitachi X-650 Scanning Electron Microanalyzer was used to
take the micrographs of the samples. Samples were mounted
on
aluminium
stubs
using
conductive
glue
and
were
then
investigate
the
coated with a thin layer of gold.
4.3.4
X-ray fluorescence spectroscopy (XRF)
An
XRF
Spectrometer
was
used
to
characteristic spectra of elements present in the solid sample.
For quantification analysis, the intensity of characteristic line
of the element analysed was measured.
The coarse solid sample was initially ground to a particle size
of 100% <200µm. The powdered sample was calcined at 850ºC
for 4 hours in order to remove all organic compounds and water
contained in the sample. The calcined sample was converted
University of Pretoria etd – Du Plessis H (2006)
4-10
into
a
solid
(Li2B4O7).
placed
solution
The
in
of
characteristic
fusion
prepared
sample
compartment
by
holders
the
line
solid
of
XRF
with
lithium
solution
and
and
placed
spectrometer.
the
element
to
borate
standard
in
The
be
tetra
the
were
sample
intensity
determined
of
a
was
measured. The concentration of the element in the sample was
calculated from the intensity measured.
4.3.5
X-ray diffraction spectroscopy (XRD)
All the samples were received as dry, fine powder. A mass of
approximately 4g of each sample, were further ground and
homogenised by hand in an agate mortar.
The additional grinding, as required for a quantitative XRD
analysis, was done using the agate segments in a McCrone
micronising mill over 10 minutes. This fine grinding followed by
spiking the samples with 10% (by mass) CaF2 as an internal
standard,
was
required
to
conduct
the
quantitative
XRD
analyses.
Approximately 0.5g of the ground sample was placed in a
stainless steel sample holder and exposed to the X-ray beam
to generate the sample’s diffraction pattern.
4.4
STANDARD TESTS FOR CEMENTITIOUS MATERIALS
Fly ash complying with the requirements of the South African
standard specification for Portland cement extenders (SANS
1491-2:2005 / SABS 1491-2:2005) conforms to the following
tests:
¾
Sulphur trioxide content
¾
Loss on ignition
¾
Free water content
¾
Fineness
¾
Water requirement
University of Pretoria etd – Du Plessis H (2006)
4-11
4.4.1
¾
Strength Factor
¾
Soundness
Sulphur trioxide content (SANS 50196-2:1994 / SABS EN
196-2:1994)
1 g of the sample is weighed; 90mℓ of water and 10 mℓ of HCI
is added while stirring vigorously.
The solution is heated and
left to digest just below boiling point for 15 minutes. The
solution is filtered and washed where after the solution is
heated while BaCl2.2H2O is added to the solution. The solution
is digested for 12 hours and then filtered and washed.
Paper
is placed in crucible and precipitate is ignited to constant mass
at 300°C.
The sulphate content is now determined for the
sample.
(equation 4.2)
SO3 content = (a - b) x 34.3
c
where:
a = mass of barium sulphate found (g)
b = mass of barium sulphate found in blank determination (g)
c = mass of sample taken (g)
4.4.2
Loss on ignition (SANS 50196-2:1994 / SABS EN 196-2:1994)
Weigh about 1g of sample into crucible and ignite at 850°C for
15
minutes.
Allow
cooling
in
desiccator
and
ascertaining
whether constant mass is obtained by weighing the sample.
Determine the LOI.
LOI % = (a - b) x 100
a
where:
a = mass of sample taken (g)
b = mass of residue (g)
(equation 4.3)
University of Pretoria etd – Du Plessis H (2006)
4-12
4.4.3
Free water content (SANS 6151:1989 / SABS 1151:1989)
Weigh about 1g of sample into crucible and dry the sample for
1 hour in oven at 100°C to a constant mass. Allow cooling in
desiccator and determining the mass of the dry sample.
Free water content % = (a - b) x 100
(equation 4.4)
b
where:
a = mass of sample (g)
b = mass of dried sample (g)
4.4.4
Test for fineness of cement and Portland cement extenders
(SANS 6157:2002 / SABS 1157:2002)
A mass of 1 g of the sample is taken after being dried in an
oven and placed on a 45μm sieve. The sample is washed with
the use of a nozzle spraying water under a pressure of 7080kPa.
Remove the sieve from under the nozzle and dry the
sieve with the residue in an oven. Determine the mass of the
sieve and residue.
Fineness, % (m/m) = c – b x 100
(equation 4.5)
a
where:
a = dried mass of specimen (g)
b = mass of sieve (g)
c = mass of sieve plus residue (g)
4.4.5
Water requirement (SANS 6156:1989 / SABS SM 1156:1989)
The mortar was mixed according to SABS Method 866 and the
flow test was performed on the sample. The mix composition
for mortar prisms is seen in Table 4.1.
University of Pretoria etd – Du Plessis H (2006)
4-13
Table 4.1 Mix composition for mortar prisms
Material
Quantity
Water
225 g
Cement
450 g
Standard reference sand
In
this
project
35%
1350 g
of
the
cement
was
replaced
with
gasification ash as can be seen in Table 4.2.
Table 4.2 Mix composition used in test
Material
Quantity
Water
225 g
Gasification Ash
140 g
Cement
310 g
Standard reference sand
1350 g
The flow table is dried and the mould is placed in the centre.
The mould is half-filled with a layer of mortar (25 mm) and
tamped 20 times with a tamper. The mould is filled and again
tamped 20 times. Excess mortar is cut off and mould is lift
vertically. In a period of 25 seconds, the table is raised and
dropped 25 times through a height of 12.7 mm. A photo of the
flow table test can be seen in Figure 4.11.
University of Pretoria etd – Du Plessis H (2006)
4-14
Figure 4.11 Photo of sample on flow table after mould is
removed
The diameter of the mortar is measured at four approximately
equal-spaced intervals. The average diameter is calculated and
taken as the flow. The increase in diameter is expressed as a
percentage of the original nominal diameter of the lower part of
the mortar specimen (100 mm). The diameter measurements on
the flow table can be seen in the photo in Figure 4.12.
Calculate to the nearest 1% the water requirement of the
material as follows:
Water requirement = Wt x 100
(equation 4.6)
Wr
where:
Wt = mass of water to produce flow value of the mortar to
within 5% points of the reference mix (g)
Wr = actual mass of water used in reference mix (g)
University of Pretoria etd – Du Plessis H (2006)
4-15
Figure 4.12 Photo of diameter measurements of flow table test
4.4.6
Strength factor test (SANS 50196-1:1994 / SABS EN 1961:1994)
The strength of cement can be determined by casting mortar
prisms.
The
specification
prescribes
a
standard
mix
composition as indicated in Table 4.1. The mortar composition
for the strength test was as indicated in Table 4.2. Mortar was
mixed according to SABS Method 866. The test specimens are
40 mm x 40 mm x 160 mm prisms as seen in Figure 4.13.
Figure 4.13 Photo of mould used
University of Pretoria etd – Du Plessis H (2006)
4-16
The specimens are moulded after mixing of the mortar. Moulds
are filled with the first of two layers of mortar, spread equally
in
each
mould
compartment
and
then
compacted
with
the
vibrating table for 20 seconds. After compacting the second
layer, the surface of the specimen is levelled and the specimen
are covered with plastic and placed in a constant temperature
room at 22°C and 55% relative humidity (RH) overnight.
Mortar prisms are cast as for the strength factor test (see 4.4.6
strength factor test) and the flow table test (see 4.4.5) is
conducted on each different mortar mix.
Each batch of mortar
is mixed mechanically using a mixer. Materials are added into
the mixing bowl and mixed for a constant time of 30 seconds
where after the water is added and mixed for a further 90
seconds. Samples are cast in sets of three prisms to obtain
strengths on 2, 7, 28 days to study the strength development
over a long period.
Specimens are demoulded after 24 hours and placed in water
at 25˚C for curing.
Firstly
the
flexural
strength
is
determined.
Prisms
are
supported on supports 100 mm apart and loaded with a point
load at midspan.
The flexural strength is calculated as follows:
Rf = 1.5 x Ff x l
b3
(equation 4.7)
where:
Rf = flexural strength (MPa)
Ff = load applied to the middle of the prisms (kN)
l = distance between the supports (mm)
b = side of the square section of the prism (mm)
The same prisms tested in the flexural test are tested for
compressive strength. The two halves of each prism are tested
University of Pretoria etd – Du Plessis H (2006)
4-17
in the press and the average of the six results obtained is
taken as the compressive strength. The compressive strength
of a prism is calculated as follows:
Rc =
(equation 4.8)
Fc
1600
where:
Rc = compressive strength (MPa)
Fc = maximum load at fracture (N)
1600 = 40 mm x 40 mm, area of the platens (mm)
4.4.7
Soundness (SANS 50196-3:1994 / SABS EN 196-3:1994)
The purpose of this test is to determine the risk of expansion
due
to
hydration.
Prepare
a
cement
paste
of
standard
consistency and fill a lightly oiled Le Chatelier mould, using
only the hands and not vibration. Place the apparatus in a
humidity (98%) cabinet for 24 hours at 20°C. Figure 4.14
shows the Le Chatelier apparatus in a water bath in a humidity
cabinet. Measure the distance between the indicator points.
Figure 4.14 Le Chatelier moulds in humidity cabinet
University of Pretoria etd – Du Plessis H (2006)
4-18
Heat mould for 30 minutes and maintain the water-bath at
boiling temperature for three hours.
Measure the distance
between the indicator points and allow mould to cool to 20°C.
Measure distance between indicator points.
(equation 4.9)
Soundness = c – a
where:
a = measurement after 24 hours in humidity cabinet (mm)
c = measurement after cooling specimen to 20°C (mm)
4.4.8
Relative density (LSA Method)
The relative density of the sample is determined by:
Relative density =
c
(equation 4.10)
(c + e) - d
where:
c = mass of sample (g)
d = mass of flask and sample and water (g)
e = mass of pyknometer and water (g)
4.5
FACTORS INVESTIGATED BY CASTING MORTAR PRISMS
4.5.1
Different Grinding Times
Mortar prisms were cast for the different grinding time intervals
with a constant replacement of cement with gasification ash of
35%. Mortar prisms were water cured and compressive and
flexural strengths were determined after 2, 7 and 28 days. The
mix composition for both interblending cement and gasification
ash after grinding each material separately, and intergrinding
University of Pretoria etd – Du Plessis H (2006)
4-19
of
cement
and
gasification
ash
by
interblending
the
two
materials in the ball mill and grinding it together is seen in
Table 4.3 and Table 4.4. The gypsum content was constant at
2.5% replacement of the cement content.
Table
4.3
Mortar
prism
mix
composition
for
interblending
for
intergrinding
cement and Gasification ash
Material
Mass (g)
Cement
292.5
Gasification ash
157.5
Gypsum
7.31
Water
225
Standard reference sand
1350
Table
4.4
Mortar
prism
mix
composition
cement and Gasification ash
Material
4.5.2
Mass (g)
Cement and Gasification ash interground
450
Gypsum
7.31
Water
225
Standard reference sand
1350
Different Gypsum Percentages
Mortar prisms were cast for in the ball mill to test the effect of
different gypsum percentages and to establish the difference
between the laboratory interblended and interground cements
with a commercially available CEM I 42.5. Table 4.5 shows the
different weights for gypsum as a percentage of the cement
content.
University of Pretoria etd – Du Plessis H (2006)
4-20
Table 4.5 Gypsum replacement weights for mortar prisms
Gypsum
replacement %
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
4.5.3
Gypsum
content (g)
0.00
1.46
2.93
4.39
5.85
7.31
8.78
Isothermal Conduction Calorimetry
In isothermal conduction calorimetry, the heat of hydration of
cement is directly measured by monitoring the heat flow from
the specimen when both the specimen and the surrounding
environment
are
maintained
at
approximately
isothermal
conditions. Approximately 3g of water were inoculated into an
equivalent mass of reactant powder that had been placed in a
copper sample cup and placed within the calorimetry cavity.
The cups were sealed with plastic film to minimize evaporation
of water. Each reactant was allowed to equilibrate separately
to 25ºC, prior to mixing. The water was equilibrated in a
syringe and when equilibrium had been achieved the plastic
film was penetrated and the water injected over the solids. The
rates of heat evolution, dQ/dt in mW/g were measured and
recorded using a computer data acquisition system.
The heat of hydration of a mortar mix was also investigated.
Mortar
mixes
Thermocouples
were
were
mixed
and
inserted
cast
into
into
the
a
steel
pocket
of
cylinder.
the
steel
cylinders’ lid. The cylinder and thermocouple were placed into
a temperature isolation flask and left in a constant temperature
room at 25ºC.
The thermocouples recorded the temperature of the mortar over
112 hours. Afterwards the thermocouples were removed and
the data were downloaded with a computer system.
University of Pretoria etd – Du Plessis H (2006)
4-21
4.5.3
Different replacements percentages of Gasification Ash
Mortar prisms were cast for different replacement percentages
of Gasification ash. Gasification ash were replaced in the ball
mill and grounded with cement. These replacement percentages
were:
0%,
10%,
20%,
35%
and
55%.
The
replacement
percentages were selected as the highest interval currently
used in commercially available CEM I (0%), CEM II (6-10% and
21-35%) and CEM IV (36-55%).
University of Pretoria etd – Du Plessis H (2006)
5-1
5.
EXPERIMENTAL PROGRAMME AND TEST PROCEDURES
FOR CONCRETE
5.1
INTRODUCTION
The aim of this study is to determine whether gasification ash
can be used as a cement extender in concrete.
Currently
pulverised fuel ash (also called fly ash) is widely used as a
cement extender in concrete and the effect of this type of ash
is well established.
In this study the properties of cement and
concrete containing gasification ash will be compared to the
properties of cement and concrete containing fly ash.
The aim of this chapter is to discuss the testing methods used
in the practical analysis of the reactivity of a gasification ash.
The
mix
discussed.
design
The
for
different
testing
of
mixes
concrete
of
concrete
cubes,
will
cylinders
be
and
shrinkage beams will be discussed with reference to the testing
apparatus, as well as the standardised testing methods used.
5.2
MIX DESIGN FOR CONCRETE MIXES
Concrete was batched for three different mixes to investigate
cube strengths, tensile strengths, E-values, shrinkage, creep,
porosity and permeability.
Each mix (see Table 5.1) had a water/cement ratio of 0.6. The
aggregate content was made up of dolomite sand and 1/3 of
9.5mm and 2/3 19mm granite stone. A 35% substitution of
cementitious materials was considered where applicable for the
mixes. The size of the mixes was 43 ℓ.
University of Pretoria etd – Du Plessis H (2006)
5-2
Table 5.1 Mix composition for concrete mixes
Materials
Quantity 43ℓ
Water
9
Per kg/m³
ℓ
210
ℓ
Cement
9.8
kg
227
kg
Cementitious material
5.3
kg
123
kg
Dolomite Sand
37.7
kg
875
kg
9.5mm Granite Stone
14.2
kg
330
kg
19mm Granite Stone
28.4
kg
660
kg
The mix composition remained constant and only the type of
cementitious material used was changed. The cement used in
the mixes can be seen in Table 5.2.
Table 5.2 Concrete mix composition
Mix
Description
Abbreviation
Mix 1
Intergrinding of cement and
gasification ash
IG
Mix 2
Interblending of cement and
gasification ash
IB GA
Mix 3
Interblending of cement and fly
ash
IB FA
5.3
TEST CONDUCTED ON FRESH CONCRETE MIXES
5.3.1
Slump Test (SANS 586 / SABS SM 82:1994)
The slump test is a method to measure the consistency of the
concrete.
It
does
however
not
test
all
the
consistency
requirements and has a limited application.
In the slump test the mould is filled in three equal layers,
subjecting each layer to 25 blows from the tamping rod while
the mould is firmly held down by standing on the foot pieces.
The surface is smoothed, the cone is removed and the slump is
measured to the nearest 5 mm. The slump is the distance
University of Pretoria etd – Du Plessis H (2006)
5-3
between the top of the inverted mould and the highest point of
the concrete as indicated in Figure 5.1.
Figure 5.1 Measuring the slump (Addis, 2001)
5.4
STRENGTH TESTS
5.4.1
Compressive Strength Test (SANS 5863-1/SABS 863-1994)
Test specimens are crushed between two platens in a hydraulic
press. The rate of load application influences the compressive
strength results and is specified at a uniform rate of 0.3 MPa/s
± 0.1 MPa/s.
All the cubes were water cured. Three cubes (100x100x100)
from each mix were crushed on 2, 7 and 28 days and the
average
of
the
three
cube
strengths
was
defined
as
the
strength. The compressive strength is recorded to the nearest
0.5MPa.
University of Pretoria etd – Du Plessis H (2006)
5-4
5.4.2
Splitting Cylinder Test for Tensile Strength (SANS
625:1994/SABS SM 1253:1994)
In South Africa the tensile strength of concrete is determined
indirectly by breaking beam specimens in flexure or by splitting
cylinders by applying line loads.
The cylinder is placed with its axis horizontal between the
platens of the testing machine, and the load is increased until
failure occurs by indirect tension in the form of splitting along
the vertical plane takes place. For each mix, two cylindrical
samples were split.
The following equation was applied to determine the tensile
strength of the concrete according to the elastic theory:
f = 2 x P
(equation 5.1)
πdℓ
where:
f = tensile strength (MPa)
P = compression load at failure (N)
d = diameter of cylinder (mm)
ℓ = length of cylinder (mm)
All the cylinders were water cured and two cylinders from each
mix were used to determine the tensile strength.
5.5
DEFORMATION AND VOLUME CHANGE OF CONCRETE
5.5.1
E-Value Test
Deformation takes place when a load is applied to a structural
material. If the ratio of the applied compressive strength to the
longitudinal strain produced is constant, the constant is called
the modulus of elasticity (Young’s modulus).
University of Pretoria etd – Du Plessis H (2006)
5-5
The non-linearity of the stress-strain relationship for concrete
is mainly due to the non-linear stress-strain response of the
paste.
A portion of the curve may be regarded as being
effectively linear, and at stresses within this range the elastic
modulus may be taken as the slope of the linear portion. For
this portion Hooke’s law may be used to determine the modulus
of elasticity.
There is no SABS test method to determine the static elastic
modulus of concrete and therefore ASTM C 469-02 is used. The
tests determine initial tangent modulus (Young’s) as well as
the
secant
modulus
corresponding
to
one
third
of
the
compressive failure stress. The test involves loading a cylinder
at
a
constant
rate
and
recording
the
load
(stress)
and
deformation (strain) of the specimen. A stress-strain curve is
determined from which the modulus of elasticity is determined.
The modulus of elasticity was determined as follows:
∆L = GL – GF
(equation 5.2)
2
ε = ∆L
(equation 5.3)
L
P = (PL – PF) x 1000
(equation 5.4)
A = πd2
(equation 5.5)
4
σ = P
(equation 5.6)
A
E = σ
ε
(equation 5.7)
University of Pretoria etd – Du Plessis H (2006)
5-6
where:
GL = Last reading of gauge (µm)
Gf = First reading of gauge (µm)
L = distance between measuring points (mm)
PL = last strength interval (kN)
PF = first strength interval (kN)
d = diameter of cylinder (mm)
E = static modulus of elasticity (GPa)
Cylinders were water cured for 28 days. For the determination
of the modulus of elasticity, two cylinders of each mix were
tested after the 28-day compressive results of the cubes were
determined.
5.5.2
Shrinkage and Creep Test (ASTM C 512-02)
Volume
change
occurs
in
concrete
in
both
the
fresh
and
hardened state. Structural performance is most concerned by
the volume change associated with an interchange of moisture
between hardened concrete and the environment.
Swelling
occurs
when
the
net
flow
of
moisture
from
the
environment to the concrete cause a volume increase and
shrinkage occurs when a net outflow from the concrete to the
environment results in a decrease in volume. Conventional
concrete
usually
contains
more
water
than
what
can
be
chemically combined with the cement and there is a tendency
for
moisture
to
be
lost
from
the
concrete,
resulting
in
shrinkage.
Creep is defined as the increase in strain (deformation) under
a sustained stress (load). When loaded, concrete experiences
an
instantaneous
elastic
strain,
which
is
recoverable.
In
addition, an inelastic creep strain takes place that is only
partially recoverable.
University of Pretoria etd – Du Plessis H (2006)
5-7
Shrinkage is tested in conjunction with creep. Three cylinders
were cast and cured in water for 28 days. For the test, one
cylinder is used for shrinkage and two cylinders for creep
testing. The shrinkage test is a natural drying method where
the cylinder is dried in a controlled environment of 65±5%
relative
humidity
and
22±2ºC.
The
shrinkage
movement
is
measured over a period of time.
The test method for creep measures the load-induced time
dependent compressive strain at selected ages for concrete.
The cube compressive strengths of the different mixes are
determined on 28 days. The cylindrical samples are placed into
a loading frame and loaded with 40% of the compressive load.
Strain readings are taken immediately before and after loading,
and thereafter at regular time intervals.
University of Pretoria etd – Du Plessis H (2006)
5-8
Figure 5.2 Photo of lab set-up for measuring shrinkage and
creep
5.6
DURABILITY
Durability can be defined as, “the capability of maintaining the
serviceability of a product, component assembly or construction
over a specified time”.
The quality of concrete cannot be determined by only using the
strength test as this approach does not give an adequate
indication of the quality of the cover concrete.
The cover
concrete acts as a barrier between the reinforcing steel and
University of Pretoria etd – Du Plessis H (2006)
5-9
the external aggressive environment and its quality is of great
importance in durability considerations.
The durability of concrete is a function of porosity. Strength
increases with decreasing porosity and porosity is affected by
capillary
porosity
and
air-void
porosity.
Porosity
defined as the percentage voids in a sample.
may
be
Voids, which are
spaces filled with water, are a result of the failure to expel all
the air from the wet concrete.
The method that has been adopted is to measure a fluid
transport parameter of the material such as permeability to
liquids or gasses.
Fluid transport properties are influenced by
capillary porosity and the degree of interconnection capillaries.
Permeability may be defined as the ease with which a liquid or
gas can pass through a specific material.
5.6.1
Porosity Test
For each of the three mixes, two water-cured samples were
tested after 28 days.
A core was drilled from 150x150 cubes
and oven-dried at a temperature of 100˚C for 24 hours to
ensure that all moisture was removed. After determining the
weight
of
the
oven-dried
samples,
they
were
placed
in
a
vacuum for 24 hours. The samples were then submerged in deaired distilled water and once again placed in a vacuum for 3
hours.
concrete
This was done to ensure that all the voids in the
samples
were
filled
with
water.
Finally
the
submerged weight of the samples was determined. The porosity
test set up can be seen in Figure 5.3.
University of Pretoria etd – Du Plessis H (2006)
5-10
Figure 5.3 Photo of porosity test set up
The following equation was applied to determine the porosity of
the concrete:
Porosity = masssat – massdry
(equation 5.8)
masssat – masswat
where:
masssat = weight of saturated sample (g)
massdry = weight of oven-dried sample (g)
masswat = weight of sample in water (g)
5.6.2
Oxygen Permeability Test
The oxygen permeability of the concrete was determined for
two samples of each mix after 28 days. A core was drilled from
the 150x150 cube and oven-dried at 100˚C for 24 hours to
prevent
expansion
permeability readings.
cracks,
which
would
influence
the
The apparatus consists of a series of
valves, which delivered oxygen at a known pressure through
the samples, and calibrated glass tubes through which the rate
University of Pretoria etd – Du Plessis H (2006)
5-11
at which the oxygen bubbles moved could be measured with a
stopwatch.
The
following
equation
was
applied
to
determine
the
permeability coefficient, k, of the concrete.
K = 2 x t x Q x e x P
(equation 5.9)
A (P22 – P12)
where:
t = sample thickness (m)
e = oxygen viscosity (2.02-5N.s/m2)
K = oxygen permeability (l/m2)
Q = volume per second passing through sample (m/s³)
A = cross sectional area of sample (m2)
P1 = atmospheric pressure (Pa)
P2 = applied pressure (Pa)
The oxygen permeability index (OPI) can be calculated as:
OPI = -Log10K
(equation 5.10)
University of Pretoria etd – Du Plessis H (2006)
6-1
6.
TEST RESULTS AND DISCUSSION ON CEMENT TESTS
6.1
INTRODUCTION
In chapter 6, the results of the experimental tests discussed in
chapter 4 are reviewed and analysed to examine the reactivity
of a gasification ash used as a cement extender.
Firstly the physical properties of the gasification ash will be
discussed with reference to the particle size distribution and
the shape of the particles.
The chemical properties of the
gasification ash will be analysed by considering the results
from XRF, XRD and standard tests for extenders.
Thereafter the results of the mortar prism results for intervals
of grinding times and different gypsum percentages will be
illustrated and discussed. The discussion of the results takes
into consideration that tests were conducted on a single set of
samples.
Limitations to the testing method will be discussed
and possible improvements will be recommended.
6.2
PHYSICAL PROPERTIES
6.2.1
Particle Size Distribution Test
The
water
demand
and
workability
of
controlled by the particle size distribution.
cement
paste
is
As nothing can be
done to alter the mineralogical characteristics, the control of
particle size distribution is the only practical method by which
the cementitious activity can be enhanced.
The particle size distribution of the cement and gasification
ash ground for a two hour time interval can be seen in Figure
6.1. It is observed that the gasification ash, grinded separate
and
interground
with
cement
had
similar
particle
size
distributions. There is a considerable difference in particle size
University of Pretoria etd – Du Plessis H (2006)
6-2
between the gasification ash and cement grinded separately for
the same time interval. The gasification ash had a particle size
range between 0.08 μm and 50μm, where the cement particles
ranged between 2μm and 110 μm. The results for all the
grinding time intervals can be seen in Appendix A. These
results indicate that the cement clinker is harder than the
gasification ash clinker. The particle size of intergrinding 65%
cement and 35% gasification ash shows that the particle size
distribution for time intervals 1,5 hours and 2 hours were
identical. Further grinding did not reduce the fineness of the
interground sample significantly (see Figure 6.2). Thus from
the
particle
size
distribution
it
is
clear
that
an
optimum
grinding time can be established.
Higher fineness provides a greater surface area to be wetted,
resulting in an acceleration of the hydration reaction. It was
thus expected that the smaller particle size of the gasification
ash should require the same or more water than fly ash.
Cumulative particle size %
100
90
80
70
IG,2hr
60
GA,2hr
50
CEM,2hr
40
Fly ash
30
20
10
0
0.01
0.1
1
10
100
1000
Particle size (μm)
Figure 6.1 Graph of cumulative particle size distribution of
cement and gasification ash (grinded separate and interground)
University of Pretoria etd – Du Plessis H (2006)
6-3
Cumulative particle size %
100
90
80
IG, 30min
70
60
IG,1hr
50
IG,2hr
40
30
IG,2.5hr
IG,1.5hr
IG,4hr
20
10
0
0.1
1
10
100
1000
Particle size (μm)
Figure 6.2 Cumulative particle sizes for gasification ash and
cement interground
The relevance of the Rosin-Rammler particle size distribution
parameters was evaluated by comparing the fitted functions for
the
cumulative
%
oversize
different grinding times.
particle
size
distributions
for
Both the constants in the equation
vary with grinding time and therefore the constants as such
cannot be used to compare the samples (See table 6.1). The
functions for gasification ash are plotted in Figure 6.3. The rest
Cumulative % oversize
of the particle size distributions can be viewed in Appendix B.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
GA,30min
GA,1hr
GA,1.5hr
GA,2hr
GA,2.5hr
GA,4hr
0
50
100
150
200
250
Particle size (μm)
Figure 6.3 Summary of exponential fitted functions for the
cumulative % oversize particle size distribution
University of Pretoria etd – Du Plessis H (2006)
6-4
According
to
Figure
6.3
the
position
parameter
(Xo)
and
therefore the particle size distribution decrease as the grinding
time increases for gasification ash.
It is clear that the grinding
time has a significant influence on the range of the particle
size distributions greater than the position parameter (Xo). The
shorter grinding times seem to have more of the larger (coarse)
particles. This trend can also be seen when comparing the
Rosin-Rammler slope (n) parameters (see Figure 6.4). See
Appendix C for rest of graphs.
The graph clearly indicates that the particle size of gasification
ash decreases as the grinding time increase. The fitted trend
line that best describes the measured values is an exponential
function. The statistical R-square value for all the samples of
the exponential fit is between 0.995 and 0.999. The exponential
equation is therefore a true representative of the cumulative %
oversize particle size distribution.
ln ln (1/cumulative %
oversize)
1.5
GA,30min
GA,1hr
1
GA,1.5hr
GA,2hr
0.5
GA,2.5hr
GA,4hr
0
0
1
2
3
4
5
6
ln (Particle size (μm))
Figure 6.4 Summary of Rosin-Rammler distributions
University of Pretoria etd – Du Plessis H (2006)
6-5
Table 6.1 Fitted functions of oversize particle size distribution
Fitted function
Sample
y = a e(bx)
R²
Xo
IG,30 min
a
0.955
b
-0.0198
0.9996
47.90
IG,1hr
0.975
-0.0354
0.9997
28.01
IG, 1.5hr
0.9929
-0.0514
0.998
19.88
IG, 2hr
0.9835
-0.05
0.9983
20.00
IG, 2.5hr
0.9693
-0.0532
0.9986
18.20
IG, 4hr
0.9822
-0.0608
0.9963
16.69
GA,30min
0.9429
-0.0122
0.9991
78.35
GA, 1hr
0.9329
-0.0284
0.9989
33.05
GA, 1.5hr
0.9331
-0.0382
0.9986
24.51
GA, 2hr
0.9452
-0.0507
0.9978
19.14
GA, 2.5hr
0.9254
-0.0554
0.9981
16.64
GA, 4hr
0.8942
-0.0781
0.9952
10.93
Cem,30min
0.9675
-0.024
0.9936
37.71
Cem, 1hr
0.7121
-0.0144
0.8842
33.11
Cem, 1.5hr
0.9411
-0.0249
0.9858
37.29
Cem, 2hr
0.8794
-0.0237
0.9681
29.43
Cem, 2.5hr
0.9077
-0.0247
0.9791
32.04
Cem, 4hr
0.8179
-0.0179
0.9454
25.71
IB, 30 min
0.9031
-0.0165
0.9874
45.51
IB, 1hr
0.8898
-0.023
0.9792
31.66
IB, 1.5hr
0.9607
-0.03
0.9973
30.55
IB, 2hr
0.945
-0.033
0.9946
25.93
IB, 2.5hr
0.9142
-0.0321
0.9897
25.59
IB, 4hr
0.8079
-0.0256
0.9545
23.27
The values of the slope n and the position parameter Xo for all
the samples can be seen in Table 6.2. The equations for the
linear trend line fitted to the Rosin-Rammler distribution graph
and the statistical R-square value of the linear fit are also
listed in Table 6.2.
University of Pretoria etd – Du Plessis H (2006)
6-6
Table 6.2 Rosin-Rammler particle size distribution parameters
Fitted function
Sample
y = cx +d
R²
Xo
n
c
d
IG,30 min
0.9743
-3.7795
1
47.90
0.9743
IG,1hr
0.9848
-3.2664
1
28.01
0.9848
IG, 1.5hr
0.9959
-2.9494
1
19.88
0.9959
IG, 2hr
0.9902
-2.951
1
20.00
0.9902
IG, 2.5hr
0.9813
-2.8498
1
18.20
0.9813
IG, 4hr
0.9896
-2.7544
1
16.69
0.9896
GA,30min
0.9653
-4.0996
1
78.35
0.9653
GA, 1hr
0.958
-3.347
1
33.05
0.958
GA, 1.5hr
0.9589
-3.0666
1
24.51
0.9589
GA, 2hr
0.9665
-2.8296
1
19.14
0.9665
GA, 2.5hr
0.9543
-2.6897
1
16.64
0.9543
GA, 4hr
0.934
-2.2793
0.9999
10.93
0.934
Cem,30min
0.9816
-3.6309
1
37.71
0.9816
Cem, 1hr
0.816
-3.1629
0.9984
33.11
0.816
Cem, 1.5hr
0.9662
-3.5129
1
37.29
0.9662
Cem, 2hr
0.9307
-3.3684
0.9999
29.43
0.9307
Cem, 2.5hr
0.9469
-3.4172
0.9999
32.04
0.9469
Cem, 4hr
0.8908
-3.4053
0.9996
25.71
0.8908
IB, 30 min
0.9424
-3.7753
0.9999
45.51
0.9424
IB, 1hr
0.9348
-3.4207
0.9999
31.66
0.9348
IB, 1.5hr
0.9777
-3.3921
1
30.55
0.9777
IB, 2hr
0.9686
-3.2527
1
25.93
0.9686
IB, 2.5hr
0.9499
-2.0479
0.9999
25.59
0.9499
IB, 4hr
0.8839
-3.0507
0.9995
23.27
0.8839
The effect of grinding time on particle size distribution can be
seen in Figure 6.5. From the graph it is observed that as the
grinding time is extended the particles become finer, resulting
in the position parameter decreasing. This trend is prominent
for the gasification and as well as the blended cement. The
cement clinker alone has no clear trend, since the position
parameter of the cement after 30 minutes is less than that of
intergrinding or gasification ash. After 1 hour the position
University of Pretoria etd – Du Plessis H (2006)
6-7
parameters for all three are approximately the same. Hereafter
gasification
ash
and
intergrinding
continues
to
reduce
as
Position Parameter X o (µm)
grinding time is extended, while the cement increases slightly.
90
80
70
60
IG
50
GA
40
CEM
30
20
10
0
0.5
1
1.5
2
2.5
4
grinding
time
Grinding time (hours)
Figure
6.5
Relation
between
and
position
parameter Xo
The position parameters of cement range between 25μm and
38μm and there is no clear trend for the position parameter of
cement to reduce with increased grinding time. The position
parameter of gasification ash and intergrinding is significantly
smaller than that of cement, even after 4 hours of grinding.
After 2 hours of grinding the reduction in position parameter for
gasification
ash
and
intergrinding
is
not
as
noticeable
as
before. The graph indicates that an exponential relation exists
between particle size and grinding time. It seems as if an
optimum
grinding
time
can
be
established.
For
both
the
intergrinding and gasification ash this optimum seems to be in
the region of 2 hours. Grinding for longer than this optimum
time would not decrease the particle size considerably more
and would only add to the cost of grinding.
In Figure 6.6 the slope is plotted as a function of the grinding
time.
From this graph it can be seen that for intergrinding and
University of Pretoria etd – Du Plessis H (2006)
6-8
gasification ash the grinding time does not have an influence
on the slope (n) value, which represents the range of the
particle size distribution of the particle sizes greater than the
position parameter. It is observed that for the gasification ash
and
intergrinding,
the
slope
after
2
hours
grinding
is
a
maximum and thereafter the slope decreases. Thus after 2
hours the size range is narrow and any increase or decrease in
grinding time make the size range wider.
For cement the range
varies for each grinding time interval.
Intergrinding had the
larger slope (n) and would have a narrower distribution than
the gasification ash and the cement. This difference in slope is
however very small and observed trends are not deemed to be
significant.
1.2
Slope (n )
1
0.8
IG
0.6
GA
CEM
0.4
0.2
0
0.5
1
1.5
2
2.5
4
Grinding time (hours)
Figure
6.6
Relationship
between
grinding
time
and
slope
parameter (n)
The values of the oversize particle size distribution parameters
(D50) and (D10) can be seen in Table 6.3. These parameters
can be compared for different samples.
University of Pretoria etd – Du Plessis H (2006)
6-9
Table 6.3 Oversize particle size distribution parameters
Oversize particle size
3 - 30µm
distribution parameters
Sample
(%)
<3 µm (%)
D50 (µm)
D10 (µm)
IG,30 min
32.1
113.6
41.73
9.18
IG,1hr
19.4
64.2
55.20
13.76
IG, 1.5hr
14.0
44.4
62.05
18.01
IG, 2hr
13.9
45.4
62.18
18.48
IG, 2.5hr
12.7
43.1
62.74
19.65
IG, 4hr
11.6
37.3
64.44
22.07
GA,30 min
50.9
183.5
31.78
7.34
GA, 1hr
21.2
78.7
48.51
14.82
GA, 1.5hr
15.8
58.9
56.30
16.55
GA, 2hr
12.4
44.1
59.57
22.01
GA, 2.5hr
10.5
40.4
59.79
25
GA, 4hr
7.3
31.3
58.34
32.48
Cem,30 min
26.2
92.8
52.82
5.66
Cem, 1hr
22.2
100.9
57.08
6.32
Cem, 1.5hr
23.9
86.5
55.41
6.83
Cem, 2hr
21.4
83.4
57.94
7.13
Cem, 2.5hr
22.1
85.1
56.99
7.42
Cem, 4hr
22.7
106.7
53.25
7.97
The
D50
and
D10
results
in
the
table
indicate
that
for
intergrinding of 35% gasification ash and 65% cement in the
ball
mil
and
for
separately
grounded
gasification
ash
and
cement the oversize particle size distribution parameters, D50
and D10, decrease for increased grinding time. This trend is
more observable for the gasification ash. The gasification ash
has a softer clinker than the cement and this is seen since the
gasification ash is easier to grind. For the cement there is no
clear trend. This is due to the harder cement clinker. It seems
that increased grinding did not affect the particle size. The
table also indicate the 30-3µm and <3µm percentage particles.
University of Pretoria etd – Du Plessis H (2006)
6-10
These results indicate that with increased grinding time the %
of
particles
<3µm
increase.
The
cement
remained
almost
constant between 52 and 58µm. For both the intergrinding and
gasification ash the % particles become more as grinding time
increase. All of the results in the table indicate that with
increased grinding time the particles become smaller.
6.2.2
Shape of Particles
Particle shape has a strong influence on the water requirement
of concrete. The unique spherical shape of fly ash makes its
water requirement lower and thus improves the workability (see
Figure 6.7). The gasification ash has no unique structure and
could be described as angular as can be seen in Figure 6.8.
Angular
increase
shapes
in
have
porosity
higher
and
a
water
requirements,
decrease
in
strength
and
an
may
be
expected.
Figure 6.7 Scanning electron microscope photo of fly ash
University of Pretoria etd – Du Plessis H (2006)
6-11
Figure 6.8 Scanning electron microscope photo of gasification
ash
The Blaine surface area (Appendix D) for cement, gasification
ash and interground cement and gasification ash is seen in
Figure 6.9.
Cement
Gasification ash
Intergrinding
2
Blaine Surface Area m /g
800
700
600
500
400
300
200
100
0
30 min
1 hour
1.5 hours
2 hours
2.5 hours
Grinding time (hours)
Figure 6.9 Graph indicating Blaine surface area
4 hours
University of Pretoria etd – Du Plessis H (2006)
6-12
From the graph it can be seen that cement has a surface area
ranging between 300m2/g and 450m2/g. This range of surface
area corresponds to commercial cement as reported by Bhatty
et al (2004). The surface area of cement increase for the
grinding time intervals up to 2 hours, after which the surface
area remains constant for 2.5 hours and slightly decreases for
4 hours grinding. This indicates that longer grinding does not
keep
increasing
optimum
surface
the
surface
area
for
area.
the
There
cement.
seems
Initially
to
be
an
after
30
minutes, gasification ash has the lowest surface area while the
interground sample is close to the cement sample. From 1 hour
grinding, both the gasification ash and interground gasification
ash and cement sample has a higher surface area than the
cement. The gasification ash and interground gasification ash
and cement is approximately the same up to 2 hours of grinding,
Thereafter the gasification ash’s surface area increase more
than the interground. This indicates that the surface area of
gasification
ash
increase,
while
the
cement
reached
an
optimum surface area. After 2 hours, when cement reached an
optimum surface area, the interground gasification ash and
cement sample increased less in surface area.
6.3
CHEMICAL PROPERTIES
6.3.1
XRF
The XRF results of the gasification ash provided insight into
the
behaviour
in
reactivity
in
concrete.
The
chemical
composition of both gasification and fly ash can be seen in
Table 6.4.
These results indicate that gasification ash has a chemical
composition similar to that of fly ash and it should thus be
acceptable to use as a cement extender in concrete. There are
University of Pretoria etd – Du Plessis H (2006)
6-13
three elements in the form of oxides present in the gasification
ash (MnO, Cr2O3
and V2O5) which are not in fly ash.
These
The P2O5 -
elements should have no effect on the reactivity.
content is lower in the gasification ash but was not considered
to be significant. The varying amount of Al2O3 within the ashes
is not so important in determining the influence on properties
in concrete, thus the lower content of Al2O3 in gasification ash
would not have detrimental effects on the concrete. These
results indicate that gasification ash should be acceptable to
use as a cement extender in concrete.
Table 6.4 XRF results
Elements
Gasification Ash
%
Typical values of
South African Fly
Ash* %
Fe2O3
6.8
3.7-4.7
MnO
0.13
0
Cr2O3
0.63
0
V2O5
0.02
0
TiO2
1.43
1.4-1.9
CaO
8.17
7.1-10.5
K2O
0.83
0.5-1.2
P2O5
0.7
1.1-1.4
SiO2
48.5
45-49
Al2O3
23.5
29-31
MgO
2.3
1.8-2.8
Na2O
0.5
0.1-0.8
Cl
0
0
S
0.4
0
SO3
0.49
0.5-1.0
Loss on ignition
5.18
5.0
*(SANS 1491-2:2005 / SABS 1491-2:2005)
University of Pretoria etd – Du Plessis H (2006)
6-14
6.3.2
XRD
The
XRD
results
showed
that
gasification
ash
has
high
percentages of quartz (SiO2) and mullite (Al6SiO13) which is
similar to fly ash.
The XRD analysis of gasification ash shows that anorthite,
sodian intermediate ((Ca, Na)(Si, Al)4O8), mullite (Al6SiO13),
alpha-quartz
(SiO2),
present
ash,
in
were
found
diopside
to
be
the
major
minerals
(Ca(Mg,Al)(Si,Al)2O6),
indialite
(Mg2Al4Si5O18) and gehlenite (Ca2Al2SiO7) were identified in
small concentrations.
These results indicate that gasification
ash should be acceptable for use as a cement extender in
concrete.
GASIFICATION
1500
1400
1300
1200
1100
900
800
700
600
500
400
300
200
100
2 Theta (Cu K-alpha)
GASIFICATION - File: HANLI04-5.raw - Type: 2Th/Th locked - Start: 5.000 ° - End: 70.000 ° - Step: 0.040 ° - Step time: 1.5 s - Temp.: 25 °C (Room) - Time St arted: 0 s - 2-Theta: 5.000 ° - Theta: 2.500 ° - Chi: 0.00 ° 00-046-1045 (* ) - Q uartz, syn - S iO2 - Hexagonal - I/Ic PDF 3.4 00-018-1202 (I) - Anorthit e, sodian, intermediate - (Ca,Na)(S i,Al)4O8 - Triclinic 00-015-0776 (I) - Mullite, syn - Al6S i2O13 - Orthorhombic -
Figure 6.10 XRD results for gasification ash
70
60
50
40
30
20
10
0
6
Lin (Counts)
1000
University of Pretoria etd – Du Plessis H (2006)
6-15
6.3.3
Standard Tests for Cementitious Materials
The results of the standard tests conducted are summarized in
Table 6.5. The chemical requirements of the gasification ash
comply with the specification as set for fly ash in all but two
cases.
The gasification ash has a higher loss on ignition
(LOI), due to the higher carbon content, which gives the ash a
dark grey colour. The water requirement of gasification ash is
more than 100% and this can be explained when considering
the angular shape of the gasification ash. The strength factor
of gasification ash is 2% higher than fly ash.
The factors that
do not meet the limits as set for fly ash do not cause a great
concern but it should be investigated.
Table 6.5 Chemical test results
Test
Sulphur trioxide content, %
(m/m)
Loss on ignition, % (m/m)
Free water content, %
(m/m)
Gasification
Limits for
ash
Fly ash
SANS 50196-2
0.49
<2.5
SANS 50196-3
5.18
<5.0
SANS 6151
0.19
1
SANS 6157
6.5
12.5
SANS 6156
> 100
95
SANS 50196-1
8
>6
SANS 50196-3
0
<1
LSA Method
2.59
2.3
LSA Method
7802
Method
Fineness, residue retained
on a sieve with square
apertures of nominal size
45um, % (m/m)
Water requirement, % of
control, max
Strength factor, %
Soundness, expansion,
mm, max
Relative density
Specific surface area,
cm2/g
Not
quantified
University of Pretoria etd – Du Plessis H (2006)
6-16
6.4
EFFECT OF GRINDING TIME ON THE PROPERTIES OF
INTERBLENDED GASIFICATION ASH AND CEMENT
Mortar Prism Compressive Strength
Figure 6.11 shows the results for interblending gasification ash
and cement manufactured in the laboratory. See Appendix E for
the mortar prism strength summary. Both the gasification ash
and cement clinker were ground for different time intervals.
A
mix of cement grinded for 2 hours and not interblended can
also be seen in the graph. From the results it is observed that
continually,
throughout
the
strength
development,
the
mix
containing cement with 2 hours grinding time had the highest
compressive strength. After 28 days the 2.5 hour mix was
almost equal to the 2 hour mix. The mix made with cement
containing
only
clinker
achieved
the
same
compressive
strength after 28 days. The difference in compressive strength
of mixes grinded for less than 2 hours was approximately 10
MPa.
This
difference
clearly
indicates
than
grinding
time
should not be shorter than 2 hours.
Compressive strength MPa
6.4.1
50
45
40
35
30
25
20
15
10
5
0
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
Cement (2 hrs)
2 days
7 days
28 days
Days
Figure
6.11
Compressive
strengths
for
interblending
gasification ash and cement for different grinding times
University of Pretoria etd – Du Plessis H (2006)
6-17
The compressive strength results of the mortar prisms are
defined as a strength class in Table 6.6 from SANS 501971/SABS EN 197-1:2000.
Table 6.6 Strength classes of interblending mixes grinding time
30 min
1 hour
1.5
hours
2
hours
2.5
hours
4
hours
cement
(2hrs)
Strength
class
N/A to
32.5 N
32.5R
32.5R
42.5 R
42.5 N
42.5 N
42.5 N
Percentage
strength
development
after 7 days
63%
75%
71%
76%
65%
73%
70%
From the strength class requirements the difference in the
compressive mortar strength for the mixes are highlighted.
Grinding
times
shorter
than
30
minutes
does
not
achieve
strengths for an acceptable strength class. Grinding times of 1
and 1.5 hours produce a cement strength class of 32.5R. A
cement strength class of 42.5 R was produced for 2 hours
grinding time and for the grinding times longer than 2 hours
cement with a strength class of 42.5 N was achieved. When
consideration is given to the cement grinded for 2 hours and
not
interblended,
it
is
clear
that
gasification
ash
had
a
contribution to the early strength development of the mortar
prisms. Table 6.6 also indicates the percentage of 28 day
strength gained after 7 days, which indicates that mostly 70%
of the strength is achieved after 7 days. Grinding for 2 hours
increased the 7 day strength to 76% of the 28 day strength.
6.4.2
Mortar Prisms Flexural Strength
The flexural strengths of the interblended mixes (Figure 6.12)
indicate the same trend as did the compressive strength for the
mix
grinded
for
2
hours
(Appendix
E).
After
28
days
the
interblended mixes grinded for longer than 2 hours exceeded
the flexural strength of the cement only mix. The difference in
flexural strengths for the mixes was approximately 1 MPa,
University of Pretoria etd – Du Plessis H (2006)
6-18
which is not a considerable difference. The flexural strength
results shows that grinding times greater than 2 hours achieved
Flexural strength MPa
better flexural strengths.
10
9
8
7
6
5
4
3
2
1
0
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
Cement (2 hrs)
2 days
7 days
28 days
Days
Figure 6.12 Flexural strengths for interblending gasification
ash and cement for different grinding times
Table
6.7
gives
a
percentage
of
the
compressive
strength
achieved by the flexural strength after 28 days. It is expected
that the flexural strength should be 10% of the compressive
strength. All of the mixes had a flexural strength of at least
19% of the compressive strength. The results indicate that
interblending
gasification
ash
and
cement
achieved
better
flexural strengths than a cement only mix.
Table 6.7 Percentage of compressive strength achieved for
interblending
Percentage
of
compressive
strength
achieved
30 min
1 hour
1.5
hours
2
hours
2.5
hours
4
hours
cement
(2hrs)
24%
22%
21%
19%
19%
20%
18%
University of Pretoria etd – Du Plessis H (2006)
6-19
Particle Size Distribution
The
effect
of
the
particle
size
distribution
on
the
28-day
compression strengths can be seen in Figure 6.13 and Figure
6.14. The Rosin-Rammler particle size distribution parameters
(as discussed in 4.3.1.2) are plotted as a function of the 28day compression strengths.
The graph indicates that the compressive strength increased
with decreasing position parameters (Xo) for interblending of
cement
clinker
and
gasification
ash.
Thus
finer
particle
achieved greater strengths. The interblended mix of 4 hours
grinding time had the smallest position parameter (Xo), but
achieved slightly lower strengths than the 2.5 hour and 2 hour
mixes. It can be seen in Figure 6.13 that for grinding times
longer than 2 hours the position parameter (Xo) decreases
slightly while no significant increase in compressive strength is
observed.
Compressive Strength
MPa
6.4.3
50
IB,30 min
40
IB, 1hr
30
IB, 1.5hr
IB, 2hr
20
IB, 2.5hr
10
IB, 4hr
0
0
10
20
30
40
50
Position parameter X o (μm)
Figure
6.13
Relation
Rosin-Rammler
between
distribution
compressive
position
interblending gasification ash and cement
strengths
and
(Xo)
for
parameter
University of Pretoria etd – Du Plessis H (2006)
6-20
Figure
6.13
indicates
that
position
parameters
30μm achieved lower compressive strengths.
larger
than
There was a
considerable increase in compressive strength from 30μm to
25μm. From the graph it seems as if a peak is reached at 25μm,
where the highest strengths were reached. It is thus concluded
that 25μm, and thus 2 hours is an optimum when position
parameters is compared to the compressive strength. However,
the difference in compressive strength between the 2, 2.5 and
4 hours are small and thus a 2-hour grinding time seems the
most effective when considering strength, particle sizes and
cost of grinding.
In Figure 6.14 the slope (n) is plotted as a function of the 28day compression strength. The smallest slope, after 4 hours
grinding had high compressive strength.
and
2.5
hours
strengths
but
had
approximately
higher
slopes.
There
Grinding times of 2
the
is
same
no
compressive
clear
trend
to
establish that smaller slopes give higher compressive strengths.
However, the difference in compression strength between the 2
and 4 hour grinding time is small and optimally the 2 hour
grinding time is most effective when strength, particle size and
Compressive strength
(MPa)
cost of grinding is considered.
50
IB,30 min
40
IB, 1hr
30
IB, 1.5hr
IB, 2hr
20
IB, 2.5hr
10
IB, 4hr
0
0.85
0.9
0.95
1
Slope (n )
Figure
6.14
Relation
Rosin-Rammler
between
distribution
compression
slope
interblending gasification ash and cement
(n)
strengths
parameter
and
for
University of Pretoria etd – Du Plessis H (2006)
6-21
In
Figure
6.15,
the
oversize
particle
size
distribution
parameters (D50) and (D10) are plotted as a function of the 28day compressive strength in order to find the influence of
Compressive Strength
(MPa)
compressive strength on the particles size distribution.
50
IB,30 min
45
IB, 1hr
40
D50
D10
IB, 1.5hr
35
IB, 2hr
30
IB, 2.5hr
IB, 4hr
25
0
20
40
60
80
100
120
140
Oversize Particle size distribution
parameters (μm)
Figure
particle
6.15
Relation
size
between
distribution
compression
parameters
strength
for
and
interblending
gasification ash and cement
As
the
average
particle
size
becomes
smaller
the
28-day
compression strength increases. As the 10% largest particle
size for different grinding times become smaller compressive
strengths also increases. After 2 hours grinding the maximum
compressive
strength
is
obtained.
The
difference
in
compressive strength between 2, 2.5 and 4 hours is small and
thus an optimum limit is reached after 2 hours grinding for both
the 50% and 10% largest particle size distribution parameters.
University of Pretoria etd – Du Plessis H (2006)
6-22
6.5
EFFECT OF GRINDING TIME ON THE PROPERTIES OF
INTERGROUND GASIFICATION ASH AND CEMENT
Mortar Prism Compressive Strength
Figure 6.16 shows the results for gasification ash and cement
clinker interground in the ball mill for different time intervals.
See summary of strength results in Appendix F. The results
indicate that intergrinding for 2 hours achieved the highest
compressive strength during the strength development. Same
as for the interblending, grinding for shorter than 1.5 hours
achieved
strengths
of
approximately
10
MPa
less.
The
compressive strength of the cement only mix where slightly
higher than the mixes interground for longer than 2 hours. This
difference of 1 MPa is omissible. From the results it can be
seen that intergrinding time should not be shorter than 1.5
hours.
Compressive strength MPa
6.5.1
50
45
40
35
30
25
20
15
10
5
0
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
Cement (2 hrs)
2 days
7 days
28 days
Days
Figure 6.16 Compressive Strength for intergrinding gasification
ash and cement for different grinding times
From
the
strength
class
requirements
in
Table
6.7
the
difference in interblending and intergrinding can be observed in
terms of strength. Grinding times shorter than 30 minutes for
intergrinding
did
not
achieve
strengths
for
an
acceptable
University of Pretoria etd – Du Plessis H (2006)
6-23
strength class. The same strength class were achieved for
grinding 1, 1.5, 2 and 4 hours. For intergrinding the 2.5 hour
grinding
achieved
a
lower
strength
class
than
with
the
interblending at 32.5 R. This is similar to the 1 and 1.5 hour
grinding time strength class. A cement strength class of 42.5 R
was produced for 2 hours grinding time and for grinding 4
hours cement with a strength class of 42.5 N was achieved.
When consideration is given to the cement grinded for 2 hours
and not interblended, it is clear that gasification ash had a
contribution to the early strength development of the mortar
prisms.
Table 6.8 also indicates the percentage of 28 day strength
gained after 7 days, which indicates that at least 70% of the
strength
is
achieved
after
7
days
for
all
the
mixes.
Intergrinding achieved a faster rate of strength development
after 7 days than the interblending mixes (Table 6.6).
Table 6.8 Strength classes for intergrinding mixes grinding time
6.5.2
30 min
1
hour
1.5
hours
2
hours
2.5
hours
4
hours
cement
(2hrs)
Strength
class
N/A to
32.5 N
32.5R
32.5R
42.5 R
32.5 R
42.5 N
42.5 N
Percentage
strength
development
after 7 days
71%
75%
73%
75%
76%
73%
70%
Mortar Prism Flexural Strength
The flexural strengths of intergrinding in Figure 6.17 indicate
that
grinding
flexural
times
strengths
longer
than
(Appendix
F).
1.5
The
hours
achieved
difference
in
higher
flexural
strengths is less than 1 MPa. The graph indicates that although
the difference in flexural strength is omissible, grinding times
longer than 1.5 hours achieved greater flexural strengths and
thus grinding for less than 1.5 hours is not advisable.
University of Pretoria etd – Du Plessis H (2006)
6-24
Flexural strength MPa
10
9
8
7
6
5
4
3
2
1
0
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
Cement (2 hrs)
2 days
7 days
28 days
Days
Figure 6.17 Flexural strengths for intergrinding gasification ash
and cement
Table
6.9
gives
a
percentage
of
the
compressive
strength
achieved by the flexural strength after 28 days for intergrinding.
All of the mixes had a flexural strength of at least 20% of the
compressive strength, which is higher than the expected 10%.
The lowest percentage is achieved by the cement only mix. The
results indicate that intergrinding gasification ash and cement
achieved better flexural strengths than a cement only mix.
Table 6.9 Percentage of compressive strength achieved for
intergrinding
Percentage
of
compressive
strength
achieved
6.5.3
30 min
1
hour
1.5
hours
2
hours
2.5
hours
4
hours
cement
(2hrs)
26%
24%
22%
20%
22%
20%
18%
Particle Size Distribution
The
effect
of
the
particle
size
distribution
on
the
28-day
compression strengths can be seen in Figure 6.18 and Figure
University of Pretoria etd – Du Plessis H (2006)
6-25
6.19. The Rosin-Rammler particle size distribution parameters
(as discussed in 4.3.1.2) are plotted as a function of the 28day compression strengths.
The graph indicates that compressive strength increase with
decreasing position parameters (Xo) for intergrinding of cement
clinker
and
gasification
greater
strengths.
It
ash.
can
be
Thus
seen
finer
in
particle
Figure
achieved
6.18
that
for
grinding times longer than 2 hours the position parameter (Xo)
decreases slightly while no significant increase in compressive
Compressive Strength
MPa
strength is observed.
50
IG,30 min
40
IG,1hr
30
IG, 1.5hr
20
IG, 2hr
10
IG, 2.5hr
IG, 4hr
0
0
10
20
30
40
50
60
Position parameter X o (μm)
Figure 6.18 Relation compressive strengths and Rosin-Rammler
distribution
position
parameter
(Xo)
for
intergrinding
gasification ash and cement
Figure
20μm
6.19
indicates
achieved
that
position
considerably
lower
parameters
compressive
larger
than
strengths.
There was a considerable increase in compressive strength
from 30μm to 20μm. The highest strengths were reached at
20μm and 16μm. The 30-3µm used by cement manufacturers as
a limit on fineness, effectively show that after 1.5 hours the %
particle
almost
remained
constant
(Table
6.3)
as
did
the
compressive strength. From the graph it seems as if a peak is
reached at 20μm. It is thus concluded that 20μm, and thus 2
University of Pretoria etd – Du Plessis H (2006)
6-26
hours is an optimum when position parameters is compared to
Compressive strength
(MPa)
the compressive strength.
50
IG,30 min
40
IG,1hr
30
IG, 1.5hr
20
IG, 2hr
IG, 2.5hr
10
0
0.97
IG, 4hr
0.975
0.98
0.985
0.99
0.995
1
Slope (n )
Figure 6.19 Relation between 28-day compression strengths
and Rosin-Rammler distribution slope (n) parameter
In Figure 6.19 the slope (n) is plotted as a function of the 28day compression strength. From this graph it seems that with
increased
grinding
time
the
slope
(n)
decrease
and
the
compression strengths increase. The smallest slope, after 4
hours grinding had the greatest compression strength.
Again
the difference in compression strength between the 2 and 4
hour grinding time is small and optimally the 2 hour grinding
time is most effective when strength, particle size and cost of
grinding is considered.
In
Figure
6.20,
the
oversize
particle
size
distribution
parameters (D50) and (D10) are plotted as a function of the 28day compressive strength in order to find the influence of
compressive strength on the particles size distribution.
University of Pretoria etd – Du Plessis H (2006)
6-27
28-Day Compressive Strength
(MPa)
50
IG,30 min
45
IG,1hr
40
IG, 1.5hr
D50
D10
IG, 2hr
35
IG, 2.5hr
30
IG, 4hr
25
0
20
40
60
80
100
120
Oversize Particle size distribution parameters
(μm)
Figure 6.20 Relation between 28-day compression strength and
particle size distribution parameters
As
the
average
particle
size
becomes
smaller
the
28-day
compression strength increases. As the 10% largest particle
size for different grinding times become smaller compressive
strengths also increases.
It is interesting to note that the
distribution curves converge, indicating that the particle sizes
become smaller and more uniform in size at higher strengths.
After 4 hours grinding the maximum compressive strength is
obtained.
However
the
compressive
strength
difference
between 4 hours and 2 hours is small and thus an optimum
limit is reached after 2 hours grinding for both the 50% and
10% largest particle size distribution parameters.
6.6
EFFECT OF GYPSUM CONTENT ON THE PROPERTIES OF
INTERGROUND GASIFICATION ASH AND CEMENT
6.6.1
Mortar Prism Compressive Strength
Figure 6.21 shows the compressive results for different gypsum
contents. See Appendix G for a summary of the strengths. The
highest
compressive
strength
is
achieved
by
3%
gypsum
content but the difference in compressive strength for the
different gypsum contents is not significant. All of the mixes
University of Pretoria etd – Du Plessis H (2006)
6-28
can be classified as 32.5R in strength class. From the results
there is no clear indication that there is an optimum gypsum
content.
Compressive Strength MPa
45
0%
40
0.50%
35
1%
30
1.50%
25
2%
20
2.50%
15
3%
10
5
0
2 Days
7 Days
28 Days
Days
Figure 6.21 Compressive strengths for gypsum content
6.6.2
Mortar Prism Flexural Strength
The
flexural
strengths
(Appendix
G)
for
different
gypsum
contents can be seen in Figure 6.22. All of the mixes were a
blend of 35% gasification ash, intergrinded with cement. After
28 days the highest compressive strength is achieved by 2%
gypsum content. The difference in compressive strength for the
different mixes is small. From the flexural strength results
there is no clear indication that there is an optimum gypsum
content.
University of Pretoria etd – Du Plessis H (2006)
6-29
10
Flexural Strength MPa
9
8
0%
7
0.50%
6
1%
5
1.50%
4
2%
3
2.50%
2
3%
1
0
2 Days
7 Days
28 Days
Days
Figure 6.22 Flexural strengths for gypsum content
6.6.3
Heat of Hydration
Figure 6.23 shows the rate of heat development for different
gypsum percentages. From the graph it is observed that the
first peak (initial reaction) is delayed for low percentages of
gypsum. All the first peaks for the gypsum percentages have a
lower heat evolution than the reference 100% cement mix. A
3% replacement of gypsum peaks at the same time as the 100%
cement, however at a lower heat evolution. For all the curves it
is observed that the curves are smooth and no shoulders is
observed after the first heat peak which would indicate flash or
false set. The duration for testing was 36 hours and restricts
the complete curve of hydration.
University of Pretoria etd – Du Plessis H (2006)
6-30
1 cem 0% gypsum
1 cem 0.5% gypsum
1 cem 1% gypsum
1 cem 1.5% gypsum
1 cem 2% gypsum
1 cem 2.5% gypsum
1 cem 3% gypsum
Cement 100%
dQ/dt (mwatts/g.sec)
2.5
2
1.5
1
0.5
0
0.02
3
6
8
11
14
17
19
22
25
28
30
33
Time (hrs)
Figure 6.23 Rate of heat development for different gypsum
percentages
Figure 6.24 represents the evolution of the hydration heat of
mortars made with the different percent of gypsum substitution
relative to 100% PC mortar, with a reference point of zero
being assigned to the hydration heat developed by the 100%
PC mortar. This representation of the data shows clearly the
effect
of
different
percentages
of
gypsum
on
the
heat
development. Low percentages of gypsum decrease the heat
output after the induction period. The heat difference with
respect to the pure cement paste is positive for these mixes
after the induction period. For higher gypsum percentages the
initial heat output is less than for low gypsum percentages.
After the induction period the heat difference for the higher
gypsum percentages as negative. Higher gypsum percentages
seem to have a constant heat output after the induction period.
36
University of Pretoria etd – Du Plessis H (2006)
6-31
1 cem 0% gypsum
1 cem 0.5% gypsum
1 cem 1% gypsum
1 cem 1.5% gypsum
1 cem 2% gypsum
1 cem 2.5% gypsum
1 cem 3% gypsum
36
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
0.02
Δ (dQ/dt)
1.8
-1.8
Time (hours)
Figure 6.24 The difference in the rate of heat evolution for
different gypsum percentages cement and pure PC cement
This
was
further
confirmed
by
the
plotting
the
total
heat
evolved with time in Figure 6.25. The total heat evolved at the
first
10
hours
increased
with
decreasing
gypsum
contents.
After 10 hours the pure cement has a considerable increase in
heat
where
low
gypsum
percentages
decrease
the
heat
evolution. Higher gypsum percentages have a constant heat
evolution which is lower than the pure cement. From the results
it is clearly observed that gypsum decrease the heat evolution
of cement pastes. This is true for the smallest percentages of
gypsum. There is no clear indication that gypsum percentages
considerably
enhance
the
hydration
of
cement.
It
can
be
deduced that a 2.5% gypsum addition is a good average to use
in further testing when heat of hydration and compressive
strength is considered.
University of Pretoria etd – Du Plessis H (2006)
6-32
240
Q (mWatss/g)
Cum 0%
200
Cum 0.5%
160
Cum 1%
Cum 1.5%
120
Cum 2%
80
Cum 2.5%
Cum 3%
40
Cum cement
0
0
10
20
30
40
Time (hours)
Figure
6.25
Total
heat
of
hydration
of
different
gypsum
percentage cements at 25ºC
6.7
EFFECT OF REPLACEMENT LEVEL ON THE PROPERTIES OF
INTERGROUND GASIFICATION ASH AND CEMENT
6.7.1
Mortar Prism Compressive Strength
Figure 6.26 shows the results for gasification ash and cement
clinker
interground
Appendix
H
for
for
different
summary
of
replacement
strength
results.
levels.
The
See
results
indicate that a replacement level of 10% achieved the highest
compressive strength. Replacement levels more than 10% were
gradually lower in compressive strength. Replacing 10% of
cement with gasification ash achieved a higher compressive
strength
than
the
mix
with
0%
replacement.
This
clearly
indicates that gasification ash replacement can increase the
compressive
strength
of
mortar
pastes.
The
difference
in
compressive strength between 35% and 55% is approximately
15MPa, which is considerable. From the results it can be seen
that a 10% replacement of gasification ash had the highest
compressive
lower
strength.
compressive
The
20%
strengths
but
and
35%
was
replacement
still
considered
had
as
sufficient for use in cement. A replacement of 55% had low
University of Pretoria etd – Du Plessis H (2006)
6-33
compressive
strengths
and
is
not
recommended
as
a
Compressive Strength
MPa
replacement for cement.
70
60
50
40
30
20
10
0
0%
10%
20%
35%
55%
2 Days
7 Days
28 Days
Days
Figure 6.26 Compressive strength for different replacement
levels of gasification ash
From
the
strength
class
requirements
in
Table
6.10
the
difference in replacement level can be observed in terms of
strength. Replacement levels of 0% to 20% achieved the same
strength
class,
52.5N.
The
35%
replacement
achieved
strengths for a lower strength class, 42.5N. The 55% did not
achieve a compressive strength to be classified. The results
indicate that gasification ash replacement has no detrimental
effect on the compressive strength development of cement
paste. Low replacement levels achieved higher compressive
strengths than high replacement levels.
Table 6.10 also indicates the percentage of 28 day strength
gained after 7 days, which indicates that at least 75% of the
strength is achieved after 7 days for mixes with a replacement
level between 0% and 35%. The 10% and 20% replacement
level achieved a faster rate of strength development after 7
days than the 0% replacement level. The 55% replacement
level achieved lower strength development than the expected
70%.
University of Pretoria etd – Du Plessis H (2006)
6-34
Table 6.10 Strength classes for different replacement levels of
gasification ash
10%
20%
35%
55%
Strength class
52.5 N
52.5 N
52.5 N
42.5 N
N/A to
32.5N
Percentage
strength
development
after 7 days
83%
87%
85%
75%
58%
Mortar Prism Flexural Strength
The flexural strengths (Appendix H) of different replacement
levels
of
gasification
ash
in
Figure
6.27
indicate
that
replacement levels lower than 35% achieved higher flexural
strengths. The difference in flexural strengths is approximately
1 MPa. The graph indicates that although the difference in
flexural strength is small, replacement levels less than 35%
achieved
greater
flexural
strengths
and
thus
a
55%
replacement level is not advisable.
Flexural Strength MPa
6.7.2
0%
12.0
10.0
0%
8.0
10%
6.0
20%
4.0
35%
2.0
55%
0.0
Tension
2 Days
7 Days
Days
Figure 6.27 Flexural strengths for different replacement levels
of gasification ash
University of Pretoria etd – Du Plessis H (2006)
6-35
Table 6.11 gives a percentage of the compressive strength
achieved by the flexural strength after 28 days for intergrinding.
All of the mixes had a flexural strength of at least 20% of the
compressive strength, which is higher than the expected 10%.
The
cement
only
mix
seen
in
Table
6.7
and
6.9
had
a
percentage of 18% while the 0% replacement was 17%. The
ratio varied with repeating and further testing should be done
to establish limits on the strengths. The lowest percentage is
achieved by the cement only mix. The results indicate that
intergrinding
gasification
ash
and
cement
achieved
better
flexural strengths than a cement only mix.
Table 6.11 Percentage of compressive strength achieved for
different replacement levels of gasification ash
Percentage of
compressive
strength
achieved
6.7.3
0%
10%
20%
35%
55%
17%
17%
18%
22%
26%
Particle Size Distribution
The
effect
of
the
particle
size
distribution
on
the
28-day
compression strengths can be seen in Figure 6.28 and Figure
6.29. The Rosin-Rammler particle size distribution parameters
(as discussed in 4.3.1.2) are plotted as a function of the 28day compression strengths.
The graph indicates that compressive strength decrease with
decreasing position parameters (Xo) for replacement levels of
gasification ash. It can be seen in Figure 6.28 that for a 10%
replacement level of gasification ash the position parameter
(Xo) is smaller than the 0% replacement. Replacement levels of
20% and 55% had the same position parameter (Xo) as the 10%
replacement level, but with lower compressive strengths. The
35% replacement level had a smaller position parameter (Xo)
University of Pretoria etd – Du Plessis H (2006)
6-36
than the other replacement levels. There is no indication that
Compressive Strength
MPa
finer particle achieved greater strengths.
70
60
50
40
30
20
10
0
0%
10%
20%
35%
55%
0
10
20
30
40
Position parameter X o (μm)
Figure 6.28 Relation compressive strengths and Rosin-Rammler
distribution position parameter (Xo) for replacement levels of
gasification ash
Figure 6.28 indicates that there is no significant indication that
position parameters had an effect on the compressive strength
Compressive Strength
MPa
development of the mortar mixes.
70
0%
60
50
40
10%
30
20
10
0
35%
20%
55%
0.1
0.3
0.5
0.7
0.9
1.1
Slope n
Figure 6.29 Relation between 28-day compression strengths
and Rosin-Rammler distribution slope (n) parameter
In Figure 6.29 the slope (n) is plotted as a function of the 28day compression strength. From this graph it seems that the
University of Pretoria etd – Du Plessis H (2006)
6-37
slope (n) is approximately 1 for all the replacement levels of
gasification ash but 0%. The compressive strength decreased
for different replacement level while the slope (n) remained the
same. There is no significant indication that the slope (n) had
an effect on the compressive strength of the mortar mixes.
In
Figure
6.30,
the
oversize
particle
size
distribution
parameters (D50) and (D10) are plotted as a function of the 28day compressive strength in order to find the influence of
compressive strength on the particles size distribution.
Compressive Strength
MPa
70
60
0%
50
10%
D50
40
D10
20%
30
35%
20
55%
10
0
0
20
40
60
80
100
Oversize Particle size distribution parameters
(μm)
Figure 6.30 Relation between 28-day compression strength and
particle size distribution parameters of replacement level of
gasification ash
As
the
average
particle
size
becomes
smaller
the
28-day
compression strength decreases. As the 10% largest particle
size
for
different
compressive
replacement
strengths
also
levels
decreases.
become
It
is
smaller
however
interesting to note that there is no clear indication that the
average particle size or the 10% largest particle size becomes
smaller for different replacement levels. The average particle
size for the different replacement levels range between 15µm
and 20µm, while the 10% largest particle size range between
40µm and 95µm. From the graph there is no optimum limit for
University of Pretoria etd – Du Plessis H (2006)
6-38
both
the
50%
and
10%
largest
particle
size
distribution
parameters.
6.8
COMPARISON BETWEEN MANUFACTURED AND
COMMERCIAL CEMENT
Mortar Prism Compressive Strength
The
cement
manufactured
in
the
lab
was
compared
to
commercially available CEM I 42.5R. See Appendix I for a
summary
of
laboratory
strengths.
was
either
The
cement
interblended
manufactured
(interblend
in
in
mixer)
the
or
interground (interblend in ball mill and ground together) for 2
hours with a 2.5% gypsum content. Figure 6.31 indicates the
compressive strengths for interblended and interground cement
manufactured in the lab and a commercially available CEM I
42.5R. From the graph it is observed that intergrinding cement
and gypsum achieved compressive strengths higher than the
CEM I 42.5R, while interblending achieved lower compressive
strengths than the CEM I 42.5R.
IB CEM+GYPSUM
Compressive Strength
MPa
6.8.1
IG CEM +GYPSUM
CEM I 42.5 R
70
60
50
40
30
20
10
0
2 Days
7 Days
28 Days
Days
Figure
6.31
Compressive
strengths
for
manufactured
and
commercially available cement
From
the
strength
class
requirements
in
Table
6.12
the
difference can be observed in terms of strength. Interblending
University of Pretoria etd – Du Plessis H (2006)
6-39
can be classified as 42.5N while intergrinding is classified as
52.5N.
Table 6.12 also indicates the percentage of 28 day strength
gained after 7 days, which indicates that interblending and
intergrinding achieved at least 80% of the strength after 7 days.
The
CEM
I
42.5
R
achieved
only
68%
of
its
strength
development after 7 days. From the results it is clear that the
interground cement manufactured in the lab performed better
than a commercially available 42.5R. The interblended cement
manufactured cement performed like a 42.5N but had a faster
strength development after 7 days than the 42.5R.
Table 6.12 Strength classes for different replacement levels of
gasification ash
6.8.2
IB CEM +
Gypsum
IG CEM +
Gypsum
CEM I 42.5R
Strength class
42.5 N
52.5 N
42.5 R
Percentage
strength
development
after 7 days
82%
80%
68%
Flexural Mortar Prism Strengths
Figure
6.32
shows
the
flexural
strengths
(Appendix
I)
for
interblended and interground cement manufactured in the lab
and a commercially available CEM I 42.5N. From the graph it is
observed
that
interground
cement
strengths than the CEM I 42.5N.
achieved
similar
flexural
University of Pretoria etd – Du Plessis H (2006)
6-40
Flexural Strength MPa
IB CEM+GYPSUM
IG CEM +GYPSUM
CEM I 42.5 R
12
10
8
6
4
2
0
2 Days
7 Days
28 Days
Days
Figure
6.32
Flexural
strengths
for
manufactured
and
commercially available cement
Table 6.13 gives a percentage of the compressive strength
achieved by the flexural strength after 28 days. Interblending
achieved the highest percentage of flexural strength compared
to compressive strength. The interground and CEM I 42.5R
mixes achieved smaller percentages but all three of the mixes
achieved a percentage higher than 10% which is expected. The
results indicate that the both the interblended and interground
cement performed similar to the commercially available CEM I
42.5R.
Table 6.13 Percentage of compressive strength achieved for
different replacement levels of gasification ash
Percentage of
compressive
strength
achieved
IB CEM +
Gypsum
IG CEM +
Gypsum
CEM I
42.5R
20%
16%
19%
University of Pretoria etd – Du Plessis H (2006)
6-41
6.8.3
Heat of Hydration
Figure
6.33
shows
the
rate
of
heat
development
for
the
manufactured and commercially available cement. From the
graph it is observed that the cement manufactured in the lab
peak at the same time as the CEMI 42.5R but at a lower heat
evolution. After the first peak both the mixes decrease without
any shoulders which indicate flash or false set. The results
indicate that the cement manufactured in the lab performed
similar to the commercially available CEM I 42.5R but lower
heat of hydration temperatures is observed for the cement
manufactured in the lab.
dQ/dt (mWatts/g.sec)
2
1.75
1.5
1.25
CEM I 42.5
1
LAB CEM
0.75
0.5
0.25
0
0 5 9 14 18 23 27 32 36 41 45 50 54 59 63 68
Time (hours)
Figure 6.33 Rate of heat development for manufactured and
commercially available cement
6.9
CONCLUSION
•
It is observed that the gasification ash, grinded separate
and interground had similar particle size distributions.
The particle size of gasification ash ground separately is
considerably finer than that of cement ground separately
for the same time interval. This indicates that the cement
clinker is harder than the gasification ash clinker.
University of Pretoria etd – Du Plessis H (2006)
6-42
•
The position parameter (Xo) and therefore the particle
size distribution decrease as the grinding time increases
for gasification ash.
•
It
seems
as
if
an
optimum
grinding
time
can
be
established. For both the intergrinding and gasification
ash this optimum seems to be in the region of 2 hours.
•
The gasification ash has no unique structure and could be
described as angular.
•
The XRF, XRD and chemical specification results indicate
that gasification ash should be acceptable to use as a
cement extender in concrete.
•
The compressive strength, flexural strength, particle size
and
Rosin-Rammler
distribution
parameters
clearly
indicate that grinding time should not be shorter than 2
hours for interblending and intergrinding of gasification
ash and cement.
•
It can be deduced that a 2.5% gypsum addition is a good
average to use in further testing when heat of hydration
and compressive strength are considered.
•
The compressive strength, flexural strength, particle size
and Rosin-Rammler distribution parameters confirm that
replacement
levels
of
gasification
ash
should
range
between 10% and 35%.
•
The cement manufactured in the laboratory performed
similar
in
available
strength
CEM
I
development
42.5R
but
to
lower
the
heat
commercially
of
hydration
temperatures are observed for the cement manufactured
in the laboratory.
University of Pretoria etd – Du Plessis H (2006)
7-1
7.
TEST RESULTS AND DISCUSSION ON CONCRETE TESTS
7.1
INTRODUCTION
In chapter 7, the results of the experimental tests discussed in
chapter 5 on concrete are reviewed and analysed to examine
the reactivity of a gasification ash when used in concrete.
Thereafter the results of the concrete cast are illustrated and
discussed.
tensile
The results include slump, compression strength,
strength,
E-value,
shrinkage,
oxygen permeability test results.
creep,
porosity
and
The discussion of the results
takes into consideration that tests were conducted on a single
set of samples.
Limitations to the testing method will be
discussed and possible improvements will be recommended.
7.2
TESTS CONDUCTED ON FRESH CONCRETE
7.2.1
Slump Test
The slump test results (as indicated in figure 7.1) illustrate that
the concrete mix interblended with fly ash had the highest
slump and thus a high workability. It is expected that due to
the filler effects characteristic of fly ash it will exhibit a slightly
lowered paste water demand. This increases the cohesiveness
of the mix which improves its workability (Holcim, 2005). The
mixes
with
indicate
gasification
that
the
water
ash
were
demand
less
of
workable.
concrete
Results
containing
gasification ash is higher than that of the mix interblended with
fly
ash.
The
slump
test
results
indicate
that
the
use
of
gasification ash as cement extender will results in a reduction
in the workability of concrete.
University of Pretoria etd – Du Plessis H (2006)
7-2
Slump mm
200
160
120
80
40
0
IG
IB FA
IB GA
Mixes
Figure 7.1 Slump test for concrete mixes
7.3
STRENGTH TESTS
7.3.1
Concrete Cube Compression Test Results
Figure 7.2 shows the compressive strength of the different
concrete mixes. The highest strength was achieved by the
intergrinded gasification ash mix, but the strength difference is
not greater than 5 MPa for all the mixes after 28 days (see
Appendix
J).
The
lowest
strengths
were
achieved
by
the
interblending with fly ash. These results clearly indicate that
the use of gasification ash as cement extender does not have a
negative impact on the strength development of concrete.
After 7 days the gasification ash intergrinded achieves a high
compressive
strength.
This
value
could
statistically
be
considered as an outlier but no trends can be concluded from
the results as only one set of samples were tested. This value
visibly indicates that more testing should be done so that
statistical conclusions could be drawn about the behaviour of
compressive strength of gasification.
The compressive strength test is however not a good indicator
of concrete durability as no direct relationship exists between
University of Pretoria etd – Du Plessis H (2006)
7-3
the two characteristics.
durability,
should
The quality of concrete, in terms of
therefore
not
be
deduced
from
the
compressive strength, as is often the case in the construction
Compressive strength
MPa
industry.
35
30
25
20
IG
IB FA
15
10
IB GA
5
0
24 hour
7 days
28 days
Time
Figure 7.2 Concrete cubes compression strength results
7.3.2
Tensile Strength Results
Theoretically the tensile strength is expected to be 10% of the
compressive strength. Results in Table 7.1, shows that the
tensile strength of the mixes was all approximately 3 MPa. All
of the mixes had a flexural strength to compressive strength
comparison
of
10%
to
12%.
Higher
than
10%
for
the
interblending mixes was due to a lower compressive strength
than the intergrinding mix. The flexural strength results of
gasification ash, intergrinded or interblended was marginally
higher than the mix interblended with fly ash. The tensile
results are however within 0.3 MPa of each other and therefore
the
difference
is
not
considerable.
Interblending
with
gasification ash has a slightly higher tensile strength than the
fly ash.
University of Pretoria etd – Du Plessis H (2006)
7-4
Table 7.1 Table comparing tensile and compressive strengths
Mixes
Tensile strength
MPa
Compression
Strength MPa
Comparison %
IG
3.3
31.5
10%
IB FA
3.02
26.7
11%
IB GA
3.35
28.0
12%
From
Figure
7.3,
it
can
be
seen
that
interblending
and
intergrinding with gasification ash achieved the highest tensile
strengths. These results indicate that the use of gasification
ash as cement extender does not results in a reduction in the
Tensile strength MPa
tensile strength of concrete.
4
3
IG
2
IB FA
IB GA
1
0
Mixes
Figure 7.3 Tensile strength results for concrete mixes
7.4
DEFORMATION AND VOLUME CHANGE OF CONCRETE
7.4.1
E-value test results
The elastic modulus represents the material stiffness of the
concrete to an imposed stress. Figure 7.4 shows the average of
two results for each mix. The results for the E-value show that
intergrinding
with
gasification
ash
achieved
the
highest
University of Pretoria etd – Du Plessis H (2006)
7-5
modulus
of
elasticity.
Interblending
with
gasification
ash
achieved the lowest modulus of elasticity.
Modulus of Elasticity GPa
50
40
30
IG
IB FA
IB GA
20
10
0
Mixes
Figure 7.4 E-value test results for concrete mixes
Gasification ash intergrinded achieved a higher modulus of
elasticity than the interblending mixes (see table 7.2). This
could be due to the better interface of the particles already
mixing
when
compressive
grinded
strength
together
of
and
intergrinding.
due
There
to
is
the
a
higher
constant
difference of 2 GPa between each of the cylinders for all three
mixes. No trends can be concluded from the results as only one
set of samples were tested. The high stiffness of intergrinding
gasification ash indicates that more testing should be done so
that statistical conclusions could be drawn about the stiffness
of concrete when gasification ash is intergrinded with cement.
These results indicate that the use of gasification ash as
cement extender does not have a detrimental effect on the
stiffness of concrete.
University of Pretoria etd – Du Plessis H (2006)
7-6
Table 7.2 E-value results of the different cylinders for the
different mixes
7.4.2
Mix
E-value
Cylinder 1
E-value
Cylinder 2
IG
38.8 GPa
40.3 GPa
IB FA
37.8 GPa
35.9 GPa
IB GA
35.4 GPa
33 GPa
Shrinkage and Creep Test
Shrinkage is caused by drying therefore factors that contribute
to the drying of concrete such as relative humidity, size and
shape of the concrete member as well as the concrete mix
proportions and materials will influence shrinkage.
Table 7.3 indicates the shrinkage, creep and specific creep for
each
of
the
mixes
after
309
days.
From
the
results
it
is
observed that intergrinding and interblending of gasification
ash
achieved
higher
shrinkage,
creep
and
specific
creep
results than the interblended fly ash mix. These differences are
only marginal.
Table 7.3 Shrinkage, Creep and Specific Creep Results for the
different mixes
MIX
Shrinkage
(microstrain)
Creep
(microstrain)
Specific Creep
(microstrain/Mpa)
IG GA
413.4
830.7
201.1
IB FA
326.8
643.7
169.8
IB GA
378.0
783.5
184.6
Previous studies conducted by Badenhorst (2003) showed that
for a 70/30 FA blended cement it is expected that for different
aggregate the shrinkage differs. A dolomite sand and granite
stone was used in the mix design of the concrete for all three
University of Pretoria etd – Du Plessis H (2006)
7-7
mixes. Granite has an expected shrinkage of 500 microstrain
while dolomite has an expected shrinkage of 300 microstrain.
The results of the three mixes fall in this range and thus the
difference
in
shrinkage
for
the
mixes
is
not
considerable.
Figures 7.5, 7.6 and 7.7 indicate the shrinkage and creep for
each of the three mixes. The results for shrinkage indicate the
gasification ash did not shrink significantly more than fly ash.
Deformation
(microstrain)
IG GA
o
73 kN Load; 55% to 60% RH; 25 C
1000
900
800
700
600
500
400
300
200
100
0
Creep
Creep
Shrinkage
0
50
100
150
200
250
300
350
Time (days)
Figure
7.5
Shrinkage
gasification ash mix
and
creep
results
for
interground
University of Pretoria etd – Du Plessis H (2006)
7-8
Deformation
(microstrain)
IB FA
o
67 kN Load; 55% to 60% RH; 25 C
1000
900
800
700
600
500
400
300
200
100
0
Creep
Creep
Shrinkage
0
50
100
150
200
250
300
350
Time (days)
Figure 7.6 Shrinkage and creep results for interblended fly ash
mix
Deformation
(microstrain)
IB GA
o
75 kN Load; 55% to 60% RH; 25 C
1000
900
800
700
600
500
400
300
200
100
0
Creep
Creep
Shrinkage
0
50
100
150
200
250
300
350
Time (days)
Figure
7.7
Shrinkage
and
creep
results
for
interblended
gasification ash mix
Creep of concrete is load induced, and is influenced by factors
associated with the application of load and the ability of the
concrete to withstand the load. The potential of the concrete to
creep is determined by mix materials and proportions of the
concrete.
University of Pretoria etd – Du Plessis H (2006)
7-9
Table
7.3
indicate
gasification
ash
that
mixes
had
interground
higher
and
creep
interblended
values
than
the
interblended fly ash mix. The difference in the creep values are
however not considerable (see figure 7.5, 7.6 and 7.7). The
results for creep indicate the gasification ash did not creep
significantly more than fly ash.
Figure 7.8 indicate the specific creep (See Appendix K) for the
three mixes. It is observed that over time the specific creep for
the three mixes is approximately the same.
results
indicate
that
gasification
ash
The specific creep
has
a
specific
creep
similar to fly ash.
55% to 60% RH; 25oC
Specific Creep
(microstrain / MPa)
300
250
200
IG GA
150
IB FA
IB GA
100
50
0
0
50
100
150
200
250
300
350
Time (days)
Figure 7.8 Specific creep results for the three different mixes
7.5
DURABILITY TESTS
7.5.1
Porosity Test Results
It can be observed from Figure 7.7 that for interblending the
porosity of the gasification ash was lower than fly ash. This is
due to a finer particle size of the gasification ash than the fly
ash, which results in a more even distribution of solid particles
in the concrete.
University of Pretoria etd – Du Plessis H (2006)
7-10
The concrete mix interblended with fly ash has the highest
porosity,
which
disperse
shows
uniformly
that
the
through
cement
the
particles
water
did
not
resulting
in
agglomerations and leaving large spaces which do not contain
cement.
These spaces form capillary pores and entrain air.
The result of this is a lower strength of the concrete. The
intergrinding with fly ash had a finer particle size and results in
a lower porosity. Interblending with gasification ash had a
slightly higher porosity than the intergrinding with gasification
ash;
this
could
be
due
to
better
interlocking
of
particles
grinded together. See Appendix L for Porosity summary.
The difference in porosity between all three mixes is between
15% and 16%. Porosity values (British Concrete Society, 2000)
for a highly porous concrete is greater than 15%. Al three of
the mixes fall into this category and thus the difference in
results is omissible.
The results indicate that the use of
gasification ash as cement extender does not results in an
increase
in
porosity.
It
is
therefore
anticipated
that
the
concrete containing gasification ash will be no less durable
than concrete currently manufactured.
Porosity %
18%
12%
IG
IB FA
IB GA
6%
0%
Mixes
University of Pretoria etd – Du Plessis H (2006)
7-11
Figure 7.7 Porosity results of concrete mixes
Oxygen Permeability Test Results
The results of the permeability test (as indicated in figure 7.8)
show that the type of ash used did not affect the permeability
of mixes. The mix interblended with fly ash achieved the lowest
permeability. High permeability is due to a poor quality of
cement paste-aggregate interface.
Fly ash has the effect of
reducing the permeability, which is observed from the results.
Intergrinding had the highest permeability and would absorb
the most water. See Appendix M for Permeability calculations.
1.2E-20
Permeability (m/s)
7.5.2
1E-20
8E-21
IG
IB FA
6E-21
IB GA
4E-21
2E-21
0
Mixes
Figure 7.8 Oxygen permeability test results for concrete mixes
Permeability values (British Concrete Society, 2000) for a good
quality concrete with a low permeability is smaller than 2 x 1018
and all of the results fall into this category.
OPI values as
seen in Figure 7.9 indicate an excellent class of durability for
concrete with a greater than 10 OPI (Alexander, 1999). These
results indicate that the use of gasification ash in concrete
achieves a low permeability and a durable concrete can be
expected.
University of Pretoria etd – Du Plessis H (2006)
7-12
Oxygen permeability index
24
18
IG
12
IB FA
IB GA
6
0
Mixes
Figure 7.9 Oxygen permeability index results for the concrete
mixes.
7.6
CONCLUSION
•
The use of gasification ash as a cement extender in
concrete would have ecological and economical benefits
in the cement industry.
•
The
slump
test
results
indicate
that
the
use
of
gasification ash as cement extender will results in a
reduction in the workability of concrete.
•
The use of gasification ash as cement extender does not
have a negative impact on the strength development of
concrete. There was no reduction in the tensile strength
of concrete.
•
The use of gasification ash as cement extender does not
have a detrimental effect on the stiffness of concrete.
•
The results for shrinkage indicate the gasification ash did
not shrink significantly more than fly ash. Creep results
indicate that gasification ash did not creep significantly
more than fly ash. The specific creep results indicate that
gasification ash has a specific creep similar to fly ash.
University of Pretoria etd – Du Plessis H (2006)
7-13
•
The results indicate that the use of gasification ash as
cement
porosity.
extender
does
not
results
in
an
increase
in
It is therefore anticipated that the concrete
containing gasification ash will be no less durable than
concrete currently manufactured.
•
The use of gasification ash in concrete achieves a low
permeability and a durable concrete can be expected.
University of Pretoria etd – Du Plessis H (2006)
8-1
8.
CONCLUSIONS AND RECOMMENDATIONS
8.1
CONCLUSIONS
The aim of the research is fulfilled and explained by giving
consideration to each of the following conclusions.
Investigating
the
physical,
chemical
and
mineralogical
composition of a gasification ash sample had the following
results:
•
The gasification ash has no unique structure and could be
described as angular.
•
The
chemical
gasification
and
ash
mineralogical
indicates
that
the
composition
ash
have
of
similar
elements than fly ash and is within the allowable range
for use as a cement extender in cement and concrete.
•
The physical properties of gasification ash indicated that
the gasification ash, ground separate and interground
with cement in the ball mill had similar particle size
distributions.
There
is
a
considerable
difference
in
particle size between the gasification ash and cement
ground
separately
indicates
that
the
for
the
cement
same
clinker
time
is
interval.
harder
than
This
the
gasification ash clinker.
•
The position parameter (Xo) and therefore the particle
size distribution decrease as the grinding time increases
for gasification ash.
Investigating the physical properties of cement manufacturing
with specific reference to grinding time, the optimisation of
gypsum, specific surface area and particle size distribution had
the following results:
University of Pretoria etd – Du Plessis H (2006)
8-2
•
It
seems
as
if
an
optimum
grinding
time
can
be
established. For both the intergrinding and gasification
ash this optimum seems to be in the region of 2 hours.
•
The compressive strength, flexural strength, particle size
and
Rosin-Rammler
distribution
parameters
clearly
indicate than grinding time should not be shorter than 2
hours for interblending and intergrinding of gasification
ash and cement.
•
It can be deduced that a 2.5% gypsum addition is a good
average to use in further testing when heat of hydration
and compressive strength is considered.
The
effect
interblended
of
replacement
and
level
interground
on
the
gasification
properties
ash
and
of
cement
shows that:
•
The compressive strength, flexural strength, particle size
and Rosin-Rammler distribution parameters confirm that
replacement
levels
of
gasification
ash
should
range
between 10% and 35%.
•
The cement manufactured in the lab performed similar in
strength development to the commercially available CEM
I
42.5R
but
lower
heat
of
hydration
temperatures
is
observed for the cement manufactured in the lab.
•
Intergrinding of gasification ash and cement in the ball
mill is in my opinion better than interblending gasification
ash and cement in the mixer.
The effect of interblended and interground gasification ash and
cement on the short and long term properties of concrete had
the following results:
•
The use of gasification ash as a cement extender in
concrete would have ecological and economical benefits
in the cement industry.
University of Pretoria etd – Du Plessis H (2006)
8-3
•
The
slump
test
results
indicate
that
the
use
of
gasification ash as cement extender will results in a
reduction in the workability of concrete.
•
The use of gasification ash as cement extender does not
have a negative impact on the strength development of
concrete. There was no reduction in the tensile strength
of concrete.
•
The use of gasification ash as cement extender does not
have a detrimental effect on the stiffness of concrete.
•
The results for shrinkage indicate the gasification ash did
not shrink significantly more than fly ash. Creep results
indicate that gasification ash did not creep significantly
more than fly ash. The specific creep results indicate that
gasification ash has a specific creep similar to fly ash.
The effect of the gasification ash on the durability of concrete
showed:
•
The use of gasification ash as cement extender does not
results
in
an
increase
in
porosity.
It
is
therefore
anticipated that the concrete containing gasification ash
will
be
no
less
durable
than
concrete
currently
manufactured.
•
The use of gasification ash in concrete achieves a low
permeability and a durable concrete can be expected.
•
Gasification ash can be used as a cement extended in
concrete.
8.2
RECOMMENDATIONS
•
The
environmental
manufacturing
of
impact
cement
of
decreasing
can
benefit
gasification ash as a cement extender.
CO2
from
in
the
using
University of Pretoria etd – Du Plessis H (2006)
8-4
•
The use of gasification ash as a cement extender in
concrete is highly recommended due to characteristics
like
increased
strength
development
and
reduced
in
properties
permeability.
•
The
effect
of
variability
chemical
of
gasification ash should be further investigated.
•
The
effect
influence
of
admixtures
the
water
on
demand
gasification
of
concrete
ash
could
and
cause
concrete to be expensive.
•
Long term testing should be done on gasification ash to
determine the effect on concrete durability.
•
Testing
should
be
repeated
on
gasification
ash
to
determine a trend for the behavior of gasification ash
when used as a cement extender in concrete.
•
The grinding times and results are not necessarily the
only
answer
since
influence on this.
the
laboratory
equipment
has
a
University of Pretoria etd – Du Plessis H (2006)
9-1
9.
REFERENCES
ACI Committee 232.1R-94. Use of Natural Pozzolans in
Concrete. ACI Journal, February 1995,
Addis, B.J. and Owens, G. 2001. Fulton’s Concrete
Technology.
Eight
Edition.
Cement
and
Concrete
Institute. Midrand, South Africa.
Alexander, M.G Mackechnie, J.R and Ballim, Y. Guide
to the use of durability indexes for achieving durability
in concrete structures. University of Cape Town and
University
of
Witwatersrand.
Research
Monograph
No.2. Cape Town. 1999.
American Coal Ash Association. 1997. Coal combustion
Product-Production and Use. Alexandria. Virginia.
ASTM
C
469-02.
Standard
test
method
for
static
modulus of elasticity and poisson’s ratio of concrete in
compression.
West
Conshohocken.
American
Society
for Testing and Materials. 2002.
ASTM C 512-02. Standard test method for Creep of
Concrete in Compression. American Society for Testing
Materials. West Conshohocken. Pennsylvania. 2002.
ASTM
C
1861-97.
Standard
Guide
for
Use
of
Coal
Combustion By-Products in Structural Fills. American
Society
for
Testing
Materials.
West
Conshohocken.
Pennsylvania. 2002.
Babcock and Wilcox. 1978. Steam. Its Generation and
Use. New York, NY.
University of Pretoria etd – Du Plessis H (2006)
9-2
Badenhorst,
S.
2003.
A
Critical
review
of
concrete
drying shrinkage, test methods, factors influencing and
shrinkage prediction. Johannesbug. University of the
Witwatersrand. July 2003. (MSc Project Report).
Ballim,
Y.
1999.
Localising
international
concrete
models – the case of creep and shrinkage prediction.
Proceedings
concrete
of
5th
the
technology
International
for
conference
developing
countries.
on
New
Delhi. Volume 1. Pt 7. November 1999. pp 66-78.
Bhatty,
J.I.
Innovations
Miller,
in
F.M.
and
Portland
Kosmatka,
Cement
S.H.
2004.
Manufacturing.
Portland Cement Association. Illinois.
Bouzoubaâ, N. Zhang, M.H. Bilodeau, A. and Malhotra,
V.M.
1997.
The
effect
of
grinding
on
the
physical
properties of fly ashes and a Portland cement clinker.
Cement and Concrete Research. Volume 27, Number.
12. pp 1862 – 1874.
British
Concrete
Society.
Society
technical
report
2000.
number
British
54.
Concrete
Diagnosis
of
deterioration in concrete structures. Identification of
defects,
evaluation
and
development
of
remedial
action. Concrete Society. London.
Brown,
J.H.
concrete
1982.
with
International
The
PFA
strength
and
substitution.
Symposium
on
the
In
Use
workability
of
Proceedings,
of
PFA
in
Concrete. University of Leeds, UK. Apr 14-16. Edited
by J.G. Cabrera and A.R. Cusens. Department of Civil
Engineering. University of Leeds, Leeds, UK. pp 151161.
Bye,
G.C.
Portland
cement.
production and properties. 2
nd
1999.
Composition,
edition. Thomas Telford.
University of Pretoria etd – Du Plessis H (2006)
9-3
Carette, G.G and Malhotra, V.M. 1986. Characterization
of Canadian fly ashes and their relative performance in
concrete.
Energy,
Mines
and
Resources
Canada.
CENMET Report 86-6E. Ottawa. ON.
Chindaprasirt, P. Homwuttiwong, S and Sirivivatnanon,
V. 2004. Influence of fly ash fineness on strength,
drying shrinkage and sulphate resistance of blended
cement
mortar.
Cement
and
Concrete
Research.
Volume 34. pp 1087 – 1092.
Davis, R.E Carlson, R.W Kelly, J.W and Davis, H.E.
1937. Properties of cements and concretes containing
fly ash. Journal of the American Concrete Institute,
Volume 33, pp 577-612.
Diamond,
S.
distribution
characteristics
and
Lopez-Flores,
between
between
F.
physical
lignitic
1981.
and
and
On
the
chemical
bituminous
fly
ashes. Proceedings, Symposium on Effects of Fly Ash
Incorporated in Cement and Concrete. Edited by S.
Diamond. Materials Research Society. Boston, MA. pp
34-44.
Erdogdu, K. Tokyay M. and Türker, P. 1999.Comparison
of
intergrinding
and
separate
grinding
for
the
production of natural pozzolan and GBFS-incorporated
blended
cements.
Cement
and
concrete
research,
Volume 29, pp 743-746.
Escalante-Garcia, and Sharp, J.H. 2004. The chemical
composition and microstructure of hydration products in
blended
cements.
Cement
Volume 26, pp 967-976.
and
concrete
composites,
University of Pretoria etd – Du Plessis H (2006)
9-4
Frigione, G. 1983. Gypsum in Cement. In: Ghosh SN
ed. Cement Technology. Pergamon Press.New York. pp
485-535.
Frigione, G. and Mara, S. 1976. Relationship between
particle size distribution and compressive strength in
Protland
Cement.
Cement
and
Concrete
Research.
Volume 6. pp 113-128.
Ghosh, R.S. and Timusk, J. 1981. Creep of fly ash
concrete. Journal of the American Concrete Institute.
Volume 87. pp 351-357.
Grieve,
G.R.H.
African
fly
1991.
ashes
The
on
the
influence
engineering
of
two
South
properties
of
concrete. PhD thesis. Johannesburg. Univeristy of the
Witwatersrand.
Hect, N.L. and Duvall, D.S. 1975. Characterisation and
Utilization of Municipal and Utility Sludges and Ashes:
Volume
III–Utility
Research
Coal
Center.
Ash.
U.S.
National
Environmental
Environmental
Protection
Agency.
Holcim.
2005.
Materials
Handbook.
Holcim
South
Africa.
Illston, J.M. and Domone, P.L.J. 2001. Construction
Materials. Their nature and behaviour. Third Edition.
Spon Press, London.
Joshi, R.C. 1982. Effect of coarse fraction (+ #325) of
fly
ash
on
concrete
properties.
In
Proceedings,
6th
International Symposium on Fly Ash Utilization. Reno,
NV. March 1982. U.S. Department of the Environment.
Washington, DC. DOE/METC/82-52. pp 77-85.
University of Pretoria etd – Du Plessis H (2006)
9-5
Krüger,
J.E.
1999.
Guide
2:
Origin,
History
and
Properties of South African Fly Ash and Portland Fly
Ash Cements. Guides on the use of South African fly
ash as a cement extender. The South African Coal Ash
Association, South Africa.
Lane, R.O. and Best, J.F. 1982.Properties and use of
fly
ash
on
Portland
cement
concrete.
Concrete
International. Volume 4, Number 7. pp 81-92.
Lea FM. 1970. Chemistry of cement and concrete. 4th
ed. Edward Arnold Publishers Ltd. Great Britain.
Lee, S.H. Kim, H.J. Sakai, E and Daimon, M. 2003.
Effect of particle size distribution of fly ash-cement
system on the fluidity of cement pastes. Cement and
concrete research, Volume 33, pp 763-768.
Lerch,
W.
1946.
The
influence
of
gypsum
on
the
hydration and properties of Portland cement pastes.
PCA Bulletin. Number 12.
Locher, F.W. Sprung, S. and Korf, P. 1973. The effect
of particle size distribution on the strength of Portland
cement. Zement-Kalk-Gips. Volume 26. pp 349-355.
Massazza,
F.
1998.
Chemical
analysis
of
Portland
cements. Chapter 4: Lea’s Chemistry of Cement and
Concrete,
Fourth
Edition,
Edited
by
Hewlett,
P.C.
Arnold, London.
Malhotra, V.M. and Mehta, P.J. 1996. Pozzolanic and
Cementitious
Materials.
First
Breach Publishers, Amsterdam.
Edition.
Gordon
and
University of Pretoria etd – Du Plessis H (2006)
9-6
Metha,
P.K.
1989.
Pozzolanic
and
cementitious
by-
products in concrete – Another look. Proceedings, 3rd
CANMET/ACI International Conference on the use of
Fly
Ash,
Silica
Fume,
Slag
and
Other
Mineral
By-
Products in Concrete, Trondheim, Norway. June 18-23.
Edited by V.M. Malhorta. American Concrete Institute.
Detroit, MI. Special Publication SP-114, Volume 1, pp
1-43.
Naik, T.R. and Ramme, B.W. 1990. Effect of high-lime
fly
ash
content
compressive
on
water
strength
of
demand,
time
concrete.
of
ACI
set
and
Materials
Journal. Volume 87. pp 619-626.
Neville,
A.M.
1995.
Properties
of
Concrete.
Fourth
Edition. Longman Group Ltd. London.
Newman,
J.
2003.
Advanced
concrete
technology.
Elsevier Ltd.
Olorunsogo,
F.T.
1990.
Effect
of
particle
size
distribution of ground granulated blast furnace slag on
some properties of slag cement mortar. PhD thesis,
University of Leeds.
Owens, P.L. 1979. Fly ash and its usage in concrete.
Concrete: The Journal of the Concrete Society. Volume
13. pp 21-26.
Pandey,
S.P.
Analysis:
Technology.
1983.
Science
In:
The
and
Ghosh
SN
Particle
Size
Applications
ed.
Cement
Pergamon Press.New York. pp 735-775.
Distribution
in
Cement
Technology.
University of Pretoria etd – Du Plessis H (2006)
9-7
Rosin, P. and Rammler, E. 1933. The laws governing
the fineness of powdered coal. Journal of the Institute
of Fuel. Volume 7. pp 29-33.
Sandberg, P. 2005. The use of Isothermal calorimetry
to optimise cement sulphate Part 1 – Cement without
admixtures.
Internal
Document.
Grace
Construction
Products.
SANS 1491/SABS 1491:1989. Portland cement extenders,
Part 1: Ground granulated blastfurnace slag, Part 2: Silica
fume, Part 3: Fly ash. South African Bureau of Standards.
Pretoria.
SANS 50196-1/SABS 196-1:1994. Methods of testing
cement,
Part
1:
Determination
of
strength.
South
African Bureau of Standards. Pretoria.
SANS 50196-2/SABS 196-1:1994. Methods of testing
cement, Part 1: Sulphur trioxide content and loss on
ignition. South African Bureau of Standards. Pretoria.
SANS 50196-3/SABS 196-1:1994. Methods of testing
cement, Part 1: Soundness. South African Bureau of
Standards. Pretoria.
SANS
50197-1/SABS
197-1:2000.
Cement
Part
1:
Composition, specification and conformity criteria for
common cements. South African Bureau of Standards.
Pretoria.
SANS
5862/SABS
SM
862:1994.
Concrete
tests
–
Consistence of freshly mixed concrete, Part 1: Slump
tests, Part 2: Flow test, Part 3: Vebe test, Part 4:
Compacting factor and compaction index. South African
Bureau of Standards. Pretoria.
University of Pretoria etd – Du Plessis H (2006)
9-8
SANS 5863/SABS Method 863-1994. Concrete tests –
Compressive
strength
of
hardened
concrete.
1st
Revision. South African Bureau of Standards. Pretoria.
SANS 6151/SABS Method 863-1994. Freewater content
of
Portland
cement
extenders.
1st
Revision.
South
African Bureau of Standards. Pretoria.
SANS 6156/SABS Method 863-1994. Water requirement
of
Portland
cement
extenders.
1st
Revision.
South
African Bureau of Standards. Pretoria.
SANS
6157/SABS
Method
863-1994.
Fineness
cement and Portland cement extenders. 1
st
of
Revision.
South African Bureau of Standards. Pretoria.
SANS 6253/SABS Method 863-1994. Concrete tests –
Compressive
strength
of
hardened
concrete.
1st
Revision. South African Bureau of Standards. Pretoria.
Schwartz, K.P. 2003. The outlook for CCPs. Electric
Perspectives. July/August 2003.
Steffan,
P.
and
Dean,
G.
1991.
FGD
Gypsum
Utilization: Survey of current practices and Assessment
of
Market
Potential.
Proceedings,
Air
Waste
Management Conference. p. 1.
Tang,
F.J.
Sulphate
and
Source
Gartner,
on
E.M.
Portland
1988.
Influences
Cement
of
Hydration.
Advances in Cement Research. Volume 1. Number 2.
April 1988.
Taylor,
H.F.W.
1982.
Telford. Reprinted 1982.
Cement
Chemistry,
2nd
ed.
University of Pretoria etd – Du Plessis H (2006)
9-9
Uchikawa, H. 1986. Effect of blending components on
hydration
and
structure
formation.
Proceedings,
8th
International Congress on the Chemistry of Cements.
Rio de Janeiro, Brazil. Sept, 22-27. Specials Report
Volume 1.
Van Dyk, J.C. Keyser, M.J. and Coertzen, M. 2005.
World
of
Coal
Ash
Conference.
SASOL’s
unique
position in SYNGAS Production in South Africa. Coal
Sources
using
SASOL
Lurgi
Fixed
Bed
Dry
Bottom
Gasifiers. Paper Presented.
Wainwright, P.J. and Olorunsogo, F.T. 1999. Effects of
PSD of GGBS on some durability properties of slag
cement mortars. Journal of the South African Institution
of
Civil
Engineering.
Volume
41,
Number
1
First
Quarter 1999. pp 9-17.
Wainwright,
P.J.
2004.
Portland
Cement.
School
of
Civil Engineering. University of Leeds. Lecturer Notes.
www.wbcsdcement.org,
agenda
for
sustainable
development. 2005.
Yuan, R.L and Cook, J.E. 1983. Study of Class C fly
ash concrete. Fly Ash, Silica Fume, Slag and Other
Mineral
By-products
in
Concrete,
first
international
conference. Montebello, Canada. SP-79. pp 307-319.
University of Pretoria etd – Du Plessis H (2006)
APPENDIX A
CUMULATIVE PARTICLE SIZE DISTRIBUTION OF GASIFICATION
ASH, CEMENT AND GASIFICATION ASH AND CEMENT
INTERGROUND AND INTERBLENDED
University of Pretoria etd – Du Plessis H (2006)
A-1
Cumulative Particle size distribution of Gasification ash
Cumulative particle size
distribution %
120
100
GA,30min
80
GA,1hr
GA,1.5hr
60
GA,2hr
GA,2.5hr
40
GA,4hr
20
0
0.01
0.1
1
10
100
1000
Particle size (µm)
Cumulative Particle size
distribution %
Cumulative Particle size distribution of Cement
100
90
CEM,30min
80
70
60
CEM,1hr
CEM,1.5hr
50
40
30
20
10
0
0.01
CEM,2hr
CEM,2.5hr
CEM,4hr
0.1
1
10
Particle size (µm)
100
1000
University of Pretoria etd – Du Plessis H (2006)
A-2
Cumulative particle size
distribution %
Cumulative particle size distribution of gasification ash and
cement interground
100
90
80
70
60
50
40
30
20
10
0
IG, 30min
IG,1hr
IG,1.5hr
IG,2hr
IG,2.5hr
IG,4hr
0.1
1
10
100
1000
Particle size (μm)
Cumulative particle size
distribution %
Cumulative particle size distribution of gasification ash and
cement interblended
100
90
80
70
60
50
40
30
20
10
0
IB 30 min
IB 1hr
IB 1.5 hr
IB 2 hr
IB 2.5hr
IB 4 hr
0.1
1
10
Particle size (μm)
100
1000
University of Pretoria etd – Du Plessis H (2006)
APPENDIX B
CUMULATIVE % OVERSIZE PARTICLE SIZE
DISTRIBUTIONS FOR GASIFICATION ASH, CEMENT AND
GASIFICATION ASH AND CEMENT INTERGROUND AND
INTERBLENDED
University of Pretoria etd – Du Plessis H (2006)
B-1
Cumulative % oversize
Cumulative % oversize particle size distribution
y = 0.9429e-0.0122x
R2 = 0.9991
100%
80%
60%
GA,30min
Expon. (GA,30min)
40%
20%
0%
0
50
100
150
200
250
Particle size (μm)
Cumulative % oversize
Cumulative % oversize particle size distribution
-0.0284x
y = 0.9329e
2
R = 0.9989
100%
80%
GA,1hr
60%
Expon. (GA,1hr)
40%
20%
0%
0
20
40
60
80
Particle size (μm)
100
120
University of Pretoria etd – Du Plessis H (2006)
B-2
Cumulative % oversize
Cumulative % oversize particle size distribution
y = 0.9331e-0.0382x
R2 = 0.9986
100%
80%
60%
GA,1.5hr
Expon. (GA,1.5hr)
40%
20%
0%
0
20
40
60
80
Particle size (μm)
Cumulative % oversize
Cumulative % oversize particle size distribution
y = 0.9452e-0.0507x
R2 = 0.9978
100%
80%
60%
GA,2hr
40%
Expon. (GA,2hr)
20%
0%
0
10
20
30
40
Particle size (μm)
50
60
University of Pretoria etd – Du Plessis H (2006)
B-3
Cumulative % oversize particle size distribution
y = 0.9254e-0.0554x
R2 = 0.9981
Cumulative % oversize
100%
80%
60%
GA,2.5hr
Expon. (GA,2.5hr)
40%
20%
0%
0
10
20
30
40
50
60
Particle size (μm)
Cumulative % oversize
Cumulative % oversize particle size distribution
y = 0.8942e-0.0781x
R2 = 0.9952
100%
80%
60%
GA,4hr
40%
Expon. (GA,4hr)
20%
0%
0
10
20
Particle size (μm)
30
40
University of Pretoria etd – Du Plessis H (2006)
B-4
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
-0.024x
y = 0.9675e
2
R = 0.9936
80%
60%
CEM,30min
40%
Expon. (CEM,30min)
20%
0%
0
50
100
150
Particle size (µm)
Cumulative % oversize particle size distribution
-0.0144x
Cumulative % oversize
100%
y = 0.7121e
2
R = 0.8842
80%
60%
CEM,1hr
Expon. (CEM,1hr)
40%
20%
0%
0
50
100
150
200
Particle size (µm)
250
300
University of Pretoria etd – Du Plessis H (2006)
B-5
Cumulative % oversize particle size distribution
-0.0249x
Cumulative % oversize
100%
y = 0.9411e
2
R = 0.9858
80%
60%
CEM,1.5hr
Expon. (CEM,1.5hr)
40%
20%
0%
0
50
100
150
Particle size (µm)
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
-0.0237x
y = 0.8794e
2
R = 0.9681
80%
60%
CEM,2hr
Expon. (CEM,2hr)
40%
20%
0%
0
50
100
Particle size (µm)
150
200
University of Pretoria etd – Du Plessis H (2006)
B-6
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
-0.0247x
y = 0.9077e
2
R = 0.9791
80%
60%
CEM,2.5hr
Expon. (CEM,2.5hr)
40%
20%
0%
0
50
100
150
Particle size (µm)
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
-0.0179x
y = 0.8179e
2
R = 0.9454
80%
60%
CEM,4hr
Expon. (CEM,4hr)
40%
20%
0%
0
50
100
150
Particle size (µm)
200
250
University of Pretoria etd – Du Plessis H (2006)
B-7
Cumulative % oversize particle size distribution
-0.0198x
y = 0.955e
Cumulative % oversize
100%
2
R = 0.9996
80%
60%
IG, 30min
Expon. (IG, 30min)
40%
20%
0%
0
50
100
150
200
Particle size (μm)
Cumulative % oversize particle size distribution
-0.0354x
y = 0.975e
100%
Cumalative % oversize
2
R = 0.9997
80%
60%
IG,1hr
Expon. (IG,1hr)
40%
20%
0%
0
20
40
60
Particle size (μm)
80
100
University of Pretoria etd – Du Plessis H (2006)
B-8
Cumulative % oversize particle size distribution
-0.0514x
y = 0.9929e
2
R = 0.998
Cumalative % oversize
100%
80%
60%
IG,1.5hr
Expon. (IG,1.5hr)
40%
20%
0%
0
20
40
60
80
100
Particle size (μm)
Cumulative % oversize particle size distribution
Cumalative % oversize
100%
y = 0.9835e
-0.05x
2
R = 0.9983
80%
60%
IG,2hr
Expon. (IG,2hr)
40%
20%
0%
0
20
40
60
Particle size (μm)
80
100
University of Pretoria etd – Du Plessis H (2006)
B-9
Cumulative % oversize particle size distribution
100%
Cumalative % oversize
-0.0532x
y = 0.9693e
2
80%
R = 0.9986
60%
IG,2.5hr
Expon. (IG,2.5hr)
40%
20%
0%
0
20
40
60
80
100
Particle size (μm)
Cumulative % oversize particle size distribution
-0.0608x
y = 0.9822e
Cumalative % oversize
100%
2
R = 0.9963
80%
60%
IG,4hr
Expon. (IG,4hr)
40%
20%
0%
0
20
40
60
Particle size (μm)
80
100
University of Pretoria etd – Du Plessis H (2006)
B-10
Cumulative % oversize particle size distribution
-0.0165x
y = 0.9031e
2
R = 0.9874
Cumulative % oversize
100%
80%
60%
IB 30 min
Expon. (IB 30 min)
40%
20%
0%
0
50
100
150
200
Particle size (μm)
Cumulative % oversize particle size distribution
-0.023x
y = 0.8898e
2
R = 0.9792
Cumulative % oversize
100%
80%
60%
IB 1hr
Expon. (IB 1hr)
40%
20%
0%
0
50
100
Particle size (μm)
150
200
University of Pretoria etd – Du Plessis H (2006)
B-11
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
y = 0.9607e
80%
-0.03x
2
R = 0.9973
60%
IB 1.5 hr
40%
Expon. (IB 1.5 hr)
20%
0%
0
50
100
150
200
Particle size (μm)
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
-0.0334x
y = 0.945e
2
R = 0.9946
80%
60%
IB 2 hr
Expon. (IB 2 hr)
40%
20%
0%
0
50
100
Particle size (μm)
150
200
University of Pretoria etd – Du Plessis H (2006)
B-12
Cumulative % oversize particle size distribution
Cumulative % oversize
100%
-0.0321x
y = 0.9142e
2
R = 0.9897
80%
60%
IB 2.5hr
40%
Expon. (IB 2.5hr)
20%
0%
0
50
100
150
200
Particle size (μm)
Cumulative % oversize particle size distribution
-0.0256x
y = 0.8079e
2
R = 0.9545
Cumulative % oversize
100%
80%
60%
IB 4 hr
Expon. (IB 4 hr)
40%
20%
0%
0
50
100
Particle size (μm)
150
200
University of Pretoria etd – Du Plessis H (2006)
APPENDIX C
ROSIN-RAMMLER DISTRIBUTION GRAPHS
University of Pretoria etd – Du Plessis H (2006)
C-1
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9653x - 4.1996
2
R =1
1
GA,30min
Linear (GA,30min)
0.5
0
0
2
4
6
ln (Particle size(µm))
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.958x - 3.347
2
R =1
1
GA,1hr
Linear (GA,1hr)
0.5
0
0
1
2
3
ln (Particle size(µm))
4
5
University of Pretoria etd – Du Plessis H (2006)
C-2
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9589x - 3.0666
2
R =1
1
GA,1.5hr
Linear (GA,1.5hr)
0.5
0
0
1
2
3
4
5
ln (Particle size(µm))
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9665x - 2.8296
2
R =1
1
GA,2hr
Linear (GA,2hr)
0.5
0
0
1
2
3
ln (Particle size(µm))
4
5
University of Pretoria etd – Du Plessis H (2006)
C-3
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9543x - 2.6897
2
R =1
1
GA,2.5hr
Linear (GA,2.5hr)
0.5
0
0
1
2
3
4
5
ln (Particle size(µm))
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.934x - 2.2793
2
R = 0.9999
1
GA,4hr
Linear (GA,4hr)
0.5
0
0
1
2
ln (Particle size(µm))
3
4
University of Pretoria etd – Du Plessis H (2006)
C-4
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9816x - 3.6309
R2 = 1
1
CEM,30min
Linear (CEM,30min)
0.5
0
0
2
4
6
8
ln (Particle size(µm))
Rosin-Rammler distribution graph
y = 0.816x - 3.1629
2
R = 0.9984
ln ln (1/Cumulative %
oversize)
1.5
1
CEM,1hr
Linear (CEM,1hr)
0.5
0
0
1
2
3
4
ln (Particle size(µm))
5
6
University of Pretoria etd – Du Plessis H (2006)
C-5
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9662x - 3.5129
2
R =1
1
CEM,1.5hr
Linear (CEM,1.5hr)
0.5
0
0
1
2
3
4
5
6
ln (Particle size(µm))
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9307x - 3.3684
2
R = 0.9999
1
CEM,2hr
Linear (CEM,2hr)
0.5
0
0
2
4
ln (Particle size(µm))
6
University of Pretoria etd – Du Plessis H (2006)
C-6
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.9469x - 3.4172
2
R = 0.9999
1
CEM,2.5hr
Linear (CEM,2.5hr)
0.5
0
0
2
4
6
ln (Particle size(µm))
Rosin-Rammler distribution graph
ln ln (1/Cumulative %
oversize)
1.5
y = 0.8908x - 3.4053
2
R = 0.9996
1
CEM,4hr
Linear (CEM,4hr)
0.5
0
0
1
2
3
4
ln (Particle size(µm))
5
6
University of Pretoria etd – Du Plessis H (2006)
C-7
Rosin-Rammler distribution graph
ln ln (1/Cumulative particle
size distribution)
1.5
y = 0.9743x - 3.7795
2
R =1
1
IG, 30min
Linear (IG, 30min)
0.5
0
0
1
2
3
4
5
6
ln (Particle size(µm))
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler distribution graph
1.5
y = 0.9848x - 3.2664
2
R =1
1
IG,1hr
Linear (IG,1hr)
0.5
0
0
1
2
3
ln (Particle size(µm))
4
5
University of Pretoria etd – Du Plessis H (2006)
C-8
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler distribution graph
1.5
y = 0.9959x - 2.9494
2
R =1
1
IG,1.5hr
Linear (IG,1.5hr)
0.5
0
0
1
2
3
4
5
ln (Particle size(µm))
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler distribution graph
y = 0.9902x - 2.951
2
R =1
1.5
1
IG,2hr
Linear (IG,2hr)
0.5
0
0
1
2
3
ln (Particle size(µm))
4
5
University of Pretoria etd – Du Plessis H (2006)
C-9
Rosin-Rammler distribution graph
ln ln (1/Cumulative particle
size distribution)
1.5
y = 0.9813x - 2.8498
2
R =1
1
IG,2.5hr
Linear (IG,2.5hr)
0.5
0
0
1
2
3
4
5
ln (Particle size(µm))
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler distribution graph
1.5
y = 0.9896x - 2.7544
2
R =1
1
IG,4hr
Linear (IG,4hr)
0.5
0
0
1
2
3
ln (Particle size(µm))
4
5
University of Pretoria etd – Du Plessis H (2006)
C-10
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler Distribution graph
1.5
y = 0.9424x - 3.7753
2
R = 0.9999
1
IB 30 min
Linear (IB 30 min)
0.5
0
0
2
4
6
ln (Particle size(µm))
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler Distribution graph
1.5
y = 0.9348x - 3.4207
2
R = 0.9999
1
IB 1hr
Linear (IB 1hr)
0.5
0
0
2
4
ln (Particle size(µm))
6
University of Pretoria etd – Du Plessis H (2006)
C-11
Rosin-Rammler Distribution graph
ln ln (1/Cumulative particle
size distribution)
1.5
y = 0.9777x - 3.3921
2
R =1
1
IB 1.5 hr
Linear (IB 1.5 hr)
0.5
0
0
2
4
6
ln (Particle size(µm))
Rosin-Rammler Distribution graph
y = 0.9686x - 3.2527
ln ln (1/Cumulative particle
size distribution)
1.5
2
R =1
1
IB 2 hr
Linear (IB 2 hr)
0.5
0
0
2
4
ln (Particle size(µm))
6
University of Pretoria etd – Du Plessis H (2006)
C-12
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler Distribution graph
1.5
y = 0.9499x - 2.0479
2
R = 0.9999
1
IB 2.5hr
Linear (IB 2.5hr)
0.5
0
0
2
4
ln (Particle size(µm))
ln ln (1/Cumulative particle
size distribution)
Rosin-Rammler Distribution graph
1.5
y = 0.8839x - 3.0507
2
R = 0.9995
1
IB 4 hr
Linear (IB 4 hr)
0.5
0
0
2
4
ln (Particle size(µm))
6
University of Pretoria etd – Du Plessis H (2006)
APPENDIX D
BLAINE SURFACE AREA CALCULATIONS
University of Pretoria etd – Du Plessis H (2006)
D-1
Cement
Sample
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
7.83gr
time (sec)
125.1
138.9
175.3
217.1
228.5
227.6
Gasification ash
Sample
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
6.65gr
time (sec)
68
173.8
253.8
355
437.7
668.5
Intergrinding
Sample
30 min
1 hour
1.5 hours
2 hours
2.5 hours
4 hours
7.28gr
time (sec)
131.4
186.2
255.4
333
348
558.1
2
cm /g
3147
3316
3726
4146
4254
4245
cm2/g
2320
3710
4483
5302
5887
7275
cm2/g
3226
3840
4497
5135
5249
6648
RD = 3.2
m2/g
315
332
373
415
425
425
RD =2.715
m2/g
232
371
448
530
589
728
RD = 2.976
m2/g
323
384
450
513
525
665
University of Pretoria etd – Du Plessis H (2006)
APPENDIX E
MORTAR PRISMS STRENGTH SUMMARY FOR THE EFFECT OF
GRINDING TIME ON THE PROPERTIES OF INTERBLENDED
GASIFICATION ASH AND CEMENT
University of Pretoria etd – Du Plessis H (2006)
E-1
Summary Table
Interblending Gasification ash and Cement
Compression
(MPa)
30
min
1
hour
1.5
hours
2 hours
2.5
hours
4
hours
Cement
(2 hrs)
2 days
10.1
11.9
11.2
20.1
18.4
17.6
17.6
7 days
19.5
24.1
23.5
34.3
29.7
31.3
31.5
28 days
30.8
32.1
33.1
45.1
45.4
43
44.9
Tension (MPa)
30
min
1
hour
1.5
hours
2 hours
2.5
hours
4
hours
Cement
(2 hours)
2 days
3
3.5
3.3
4.9
4.5
4.8
3.8
7 days
5.2
5.8
5.5
7.3
6.7
7.4
7.3
28 days
7.3
7.7
8.2
8.9
8.8
8.9
8.2
University of Pretoria etd – Du Plessis H (2006)
APPENDIX F
MORTAR PRISMS STRENGTH SUMMARY FOR THE EFFECT OF
GRINDING TIME ON THE PROPERTIES OF INTERGROUND
GASIFICATION ASH AND CEMENT
University of Pretoria etd – Du Plessis H (2006)
F-1
Summary Table
Intergrinding Gasification Ash and Cement
Compression
(MPa)
30
min
1 hour
1.5
hours
2
hours
2.5
hours
4 hours
Cement
(2 hours)
2 days
11.8
13.3
17
21.5
19.5
19.2
17.6
7 days
19.4
24.6
29.3
32.4
31.6
31.7
31.5
28 days
27.3
32.9
40.1
43.2
41.6
43.6
44.9
Tension (MPa)
30
min
1 hour
1.5
hours
2
hours
2.5
hours
4 hours
Cement
(2 hours)
2 days
3
3.4
4.2
4.8
4.5
5.1
3.8
7 days
4.6
6
6.5
7.3
6.7
7.2
7.3
28 days
7
8
8.8
8.6
9
8.8
8.2
University of Pretoria etd – Du Plessis H (2006)
APPENDIX G
MORTAR PRISMS STRENGTH SUMMARY FOR THE EFFECT OF GYPSUM
CONTENT ON THE PROPERTIES OF INTERGROUND
GASIFICATION ASH AND CEMENT
University of Pretoria etd – Du Plessis H (2006)
G-1
Summary Table
Gypsum Content
Compression
(MPa)
0%
0.50%
1%
1.50%
2%
2.50%
3%
2 days
10.7
11.1
11.7
11.6
12
12.1
15.2
7 days
25.1
24.5
24.9
25.7
27.6
25.7
28.6
28 days
33.7
36.1
36.9
32.9
36.2
37.1
38.9
Tension (MPa)
0%
0.50%
1%
1.50%
2%
2.50%
3%
2 days
3.2
3.3
3.2
3.4
3.3
3.3
4.5
7 days
6
6
5.5
6.2
5.9
5.7
6.2
28 days
7.0
7.8
7.7
7.5
8.0
7.7
7.6
University of Pretoria etd – Du Plessis H (2006)
APPENDIX H
MORTAR PRISMS STRENGTH SUMARY FOR THE EFFECT OF
REPLACEMENT LEVEL ON THE PROPERTIES OF
INTERGROUND GASIFICATION ASH AND CEMENT
University of Pretoria etd – Du Plessis H (2006)
H-1
Summary Table
Replacement Level
Compression (MPa)
0%
10%
20%
35%
55%
2 Days
32.1
34.4
26.1
18.3
6.5
7 Days
47.0
50.3
45.4
32.6
16.9
28 Days
56.9
57.9
53.2
43.3
28.9
Tension (MPa)
0%
10%
20%
35%
55%
2 Days
6.8
7.2
4.7
4.3
1.5
7 Days
8.8
9.2
9.1
7.2
4.4
28 Days
9.7
9.8
9.6
9.5
7.5
University of Pretoria etd – Du Plessis H (2006)
APPENDIX I
MORTAR PRISMS STRENGTH SUMMARY FOR THE COMPARISON
BETWEEN MANUFACTURED AND COMMERCIAL CEMENT
University of Pretoria etd – Du Plessis H (2006)
I-1
Summary Table
Compression
(MPa)
IB CEM+GYPSUM
IG CEM +GYPSUM
CEM I 42.5 R
2 Days
18.8
29.8
22.3
7 Days
34.5
49.4
36.2
28 Days
42.0
61.5
53.0
Tension (MPa)
IB CEM+GYPSUM
IG CEM +GYPSUM
CEM I 42.5
2 Days
4.2
5.9
5.0
7 Days
6.4
8.3
7.8
28 Days
8.3
10.0
9.9
University of Pretoria etd – Du Plessis H (2006)
APPENDIX J
CUBE STRENGTH SUMMARY
University of Pretoria etd – Du Plessis H (2006)
J-1
Cube Strength Summary Table
Summary
24 hour
7 days
28 days
IG (MPa)
6.33
22.64
31.5
IB FA (MPa)
6.31
15.93
26.7
IB GA (MPa)
7.59
19.29
28.0
University of Pretoria etd – Du Plessis H (2006)
APPENDIX K
SPECIFIC CREEP SUMMARY
University of Pretoria etd – Du Plessis H (2006)
K-1
Specific Creep Summary Table
Time (days)
IG GA
(microstrain)
IB FA
(microstrain)
IB GA
(microstrain)
0
0.00
0.00
0.00
1
10.48
4.15
11.13
2
14.30
20.25
21.80
3
24.30
29.59
30.61
7
38.12
44.65
42.67
8
43.84
51.92
48.70
10
49.08
57.63
53.80
13
58.61
68.02
64.47
15
67.19
77.88
72.82
17
71.48
83.59
75.60
41
97.69
97.09
89.98
48
116.27
113.70
115.49
53
120.08
120.97
116.42
64
129.61
124.09
119.20
120
148.20
137.07
136.83
127
155.82
142.78
141.00
132
159.16
145.89
143.78
146
163.92
153.68
150.28
167
171.07
160.95
158.16
269
188.23
175.49
173.00
309
201.09
169.78
184.60
University of Pretoria etd – Du Plessis H (2006)
APPENDIX L
POROSITY SUMMARY
University of Pretoria etd – Du Plessis H (2006)
L-1
Summary Table
Mix
Porosity
IG
15.1%
IB FA
15.9%
IB GA
15.4%
University of Pretoria etd – Du Plessis H (2006)
APPENDIX M
PERMEABILITY CALCULATIONS
University of Pretoria etd – Du Plessis H (2006)
M-1
Summary Table
Mix
IG
IB FA
Pressure bar
2
2
2
2
2
Diameter (m)
0.06935
0.0694
0.06935
0.0693
0.0694
Area (m2)
0.0037773
12
0.003782
76
0.00377
73
0.0037718
67
Time average
(s)
4.85
10.93
8.29
Reading( m3/s)
1.88262E05
8.3735E06
Thickness of
disc (m)
0.02
0.02
e
2.02E-16
2.02E-16
1.621E-20
7.213E21
Permeability
(m/s)
Permeabilty
average (m/s)
OPI
IB GA
2
2
2
2
0.0693
5
0.0037
77
0.0037
83
0.0693
5
0.0037
77
0.003782
76
0.0694
5
0.0037
88
12.83
10.81
11.15
11.53
8.29
9.79
1.1E-05
7.10137E06
8.42E06
7.78E06
7.85095E
-06
1.08E05
9.38E06
0.02
0.02
0.02
0.02
0.02
0.02
0.02
2.02E16
7.21E21
2.02E16
6.37E21
2.02E16
9.07E21
2.02E16
8.14E21
2.02E16
9.461E21
2.02E-16
6.125E-21
0.0694
2.02E-16
6.688E21
1.096E-20
6.569E-21
7.967E-21
19.96
20.18
20.098
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement