potentials of cement kiln dust in subgrade improvement … e..pdf · 1 potentials of cement kiln...
TRANSCRIPT
1
POTENTIALS OF CEMENT KILN DUST
IN SUBGRADE IMPROVEMENT
BY
EGBE ENANG ANDREW
PG/M.ENG/09/ 51153
A THESIS SUBMITTED TO THE DEPARTMENT OF CIVIL
ENGINEERING, FACULTY OF ENGINEERING, UNIVERSITY
OF NIGERIA, NSUKKA
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF
MASTER OF ENGINEERING DEGREE (M.ENG) IN CIVIL ENGINEERING
APRIL, 2012
2
CERTIFICATION
Egbe, Enang Andrew, a post graduate Student of the Department of Civil Engineering with
registration number PG/ M.ENG/ 09/51153 has satisfactorily completed the research work for
the Award of Master of Engineering Degree in Civil Engineering. The work embodied in the
dissertation is original and has not been submitted elsewhere for award of a degree.
------------------------------------------------- ---------------------------------------
Egbe Enang Andrew Date
----------------------------------------- ------------------------------------
Engr. Dr F. O. Okafor Date
(Project Supervisor)
------------------------------------------ -------------------------------------
External Examiner Date
------------------------------------------ ----------------------------------
Engr. Dr J.C Ezeokonkwo Date
Head, Civil Engineering Department
----------------------------------------- ------------------------------------
Engr. Professor J.C. Agunwamba Date Dean, Faculty of Engineering
3
DEDICATION
This work is dedicated to God Almighty and to my special wife Mrs Joy Enang Andrew Egbe
and to my beloved daughter Miss Ekema Enang Andrew for their love and encouragement.
4
ACKNOWLEDGEMENT
My profound gratitude goes to my supervisor, Engr. Dr. F. O. Okafor of the Department of Civil
Engineering, University of Nigeria Nsukka, for his guidance and invaluable contribution that
ensured that this research work was properly undertaken. I remain greatly indebted to him. I‟m
also grateful to the Dean of Faculty of Engineering, Engr. Prof. J.C. Agunwamba and other
lecturers of the Department of Civil Engineering, University of Nigeria Nsukka for their support.
I remain very grateful to my wife, Mrs Joy Enang Egbe who has been very supportive and is ever
willing to stand by me.
My sincere appreciation also goes to staff of United Cement Company (UNICEM) and those of
Civil Engineering Department, Cross River University of Technology (CRUTECH) for their
assistance and encouragement toward this research work.
I will forever remain grateful and indebted to my parents Mr/ Mrs Andrew Egbe and the entire
Egbe‟s family for their moral support and encouragement that always inspire me to achieve
heights.
I also acknowledge the support of my Pastor and boss and his wife Pastor/ Mrs Victor Ovat and
that of Pastor/ Dr( Mrs) Obafemi and all other members of Charismatic Renewal Ministries,
Cross River State, especially Mr Caleb Egwom and Mr Godwin Ojisor.
Finally, I wish to appreciate my dear friend, Engr. Desmond Ewa for his contribution and
support and other colleague and friends whose names have not been mentioned here, but have
contributed one way or the other to the successful completion of this research.
5
ABSTRACT
The ever increasing cost of construction materials in Nigeria and other developing countries has
created the need for research into locally and readily available materials and also on how to
convert materials considered to be waste, such as cement kiln dust (CKD) for use in construction
and soil improvement. This study investigated the impact of CKD as additive on sub-base and
base materials. To achieve this, Soil sample collected from Sankwalla– Busiri road, Obanliku
Cross River State classified as an A-2-7 soil on AASHO classification was stabilized with 2-24 %
cement kiln dust (CKD) by weight of the dry soil. The investigation includes evaluation of properties
such as compaction, consistency limits and strength of the soil. The results obtained show that the
increase in CKD content increased the Optimum moisture content (OMC) and a reduced the plasticity.
There was also improvement in the California Bearing Ratio (CBR) and Unconfined Compressive
strength (UCS) with increase in the CKD content. A predictive model was developed and found to
reasonably predict the relationship between properties of soil and the proportion of CKD used. The
coefficients of correlation were high showing a strong relationship between the measured and predicted
values. The study concluded that CKD can be used to improve the properties of soil for construction
purpose and 24% was observed to yield maximum improvement for CBR and UCS.
6
TABLE OF CONTENTS
Page
Title page i
Certification ii
Dedication iii
Acknowledgement iv
Abstract v
Table of content vi
List of Tables ix
List of Figures xi
CHAPTER ONE
1.0 Introduction 1
1.2 Statement of Problem 3
1.3 Aims and Objectives of the research 3
1.4 Significance of the research 4
1.5 Scope of the research 5
1.6 Limitation of the research 5
CHAPTER TWO
2.0 Literature Review 6
2.1 Cement Manufacturing and Production of Cement Kiln Dust 6
2.2 Cement kiln dust definition and formation 8
2.3 Composition of Cement Kiln dust 10
2.4 Properties of Cement Kiln Dust 14
2.4.1 Reactivity of Cement Kiln Dust 14
2.4.2 Loss on Ignition ( LOI) 17
7
2.4.3 pH 18
2.4.4 Particle Size Distribution 18
2.4.5 Specific Surface Area 21
2.4.6 Specific gravity 22
2.4.7 Engineering Properties 22
2.5 Utilization of Cement Kiln Dust 24
2.5.1 Cement Kiln Dust in Soil Stabilization 25
2.5.2 Controlled low Strength Materials 29
2.5.3 Highway base and Subbase 31
2.5.4 Blended Cement and Construction Products 32
2.5.5 Asphalt Pavement 34
2.5.6 Activator for pozzolanas 34
2.5.7 Mining Industry 35
2.5.8 Sludge Stabilization 35
2.6 Design and Analysis of Experiments 36
2.6.1 Strategy of Experiment 36
2.6.2 Mixtures Designs 39
2.6.3 Simplex Designs 40
2.6.4 Derivation of Scheffe‟s 2nd
degree Model 44
2.6.5 Least Square Estimation of model Parameters 46
CHAPTER THREE
3.0 Materials and Method 48
3.1Sample Collection 48
3.2 Materials Preparation 48
3.3 Experimental Tests 48
8
3.3.1 Sieve Analysis 49
3.3.2 Liquid Limit 49
3.3.3 Plastic limit 50
3.3.4 Plasticity index 50
3.3.5 Compaction Test 51
3.3.6 California Bearing Ratio 52
3.3.7 Unconfined Compressive strength (UCS) Test 52
CHAPTER FOUR
4.0 Results and Discussion 54
4.1 Result of Tests on Untreated Soil Sample 54
4.2 Grading Curve Results (Sieve Analysis) 55
4.3 Chemical Composition of the cement kiln dust 55
4.3.1 Atterberg Limit Test 56
4.3.1.1 Effect of cement kiln dust on consistency (Atterberg limit) 58
4.3.2 Compaction Test Result 59
4.3.2.1 Effect of cement kiln dust on the compaction characteristics of the soil 60
4.3.3 California Bearing Ratio Test Result 61
4.3.3.1 Effect of cement kiln dust on California Bearing ratio of soil 62
4.3.4 Unconfined Compressive Strength Test Result 63
4.3.4.1 Effect of cement kiln dust on unconfined compressive strength of soil sample 64
4.4 Response Surface methodology of cement kiln dust stabilized subgrade
and Model verification 65
4.5 Model Verification 76
9
CHAPTER FIVE
5.0 Conclusion and Recommendation 82
5.1 Conclusion 82
5.2 Recommendation 83
References 84
List of Tables
Table Title page
2.1 Statistics on composition of fresh cement kiln dust based on 63 CKD 11
2.2 Average phase composition of cement kiln dust 12
2.3 Mineralogical Composition of four major component of CKD 13
2.4 Summary of CKDs and soils used and strength of treated soil 17
2.5 Effect of particle size on alkali content of CKD (European plants) 21
2.6 Cement kiln dust used for compaction studies 22
2.7 Permeability values, k, of compacted CKD sample 23
4.1 Index Properties of untreated soil sample 54
4.2 Oxide composition of cement kiln dust 55
4.3 Tabulated Atterberg limit test results 56
4.4 Compaction Test Result 59
4.5 Result of CKD on CBR of soil sample 61
4.6 Result of 7,14 and 28 days unconfined Compressive strength with CKD 63
4.7 Test Results after Stabilization with cement kiln dust 66
4.8 Summary of Stabilization Test Result 68
4.9 Selected Controllable Variable for optimum moisture content 69
4.10 Selected Controllable Variable for Maximum Dry Density 70
10
4.11 Selected Controllable Variable for CBR (unsoaked) 72
4.12 Selected Controllable Variable for CBR (soaked) 73
4.13 Selected controllable Variable for unconfined Compressive
Strength ( 7days) 74
4.14 Optimum results of the stabilized soils for discussion 76
4.15 Experimented and Model values for Optimum Moisture Content 77
4.16 Experimented and Model values for Maximum Dry Density 78
4.17 Experimented and Model values for California Bearing Ratio
(Unsoaked) 79
4.18 Experimented and Model values for California Bearing Ratio (soaked) 80
4.19 Experimented and Model values for unconfined Compressive
Strength (7days) 81
List of figures
Figure Title Page
2.1 Cement and Cement Kiln Dust Manufacturing Process 7
2.2 X-Ray Diffraction for Cement Kiln Dust (CKD) 14
2.3 Relationship between Compressive Strength of CKD- fly ash
Blends and TRO of CKD 16
2.4 Examples of CKD particle size distribution 19
2.5 Effect of Kiln dust type and dust collection on particle size distribution 20
2.6 Typical compaction curves for fresh CKD 23
2.7 Lattice design for 3x2 mixes 41
11
2.8 Lattice design for 3x3 mixes 42
4.1 Particle size distribution for untreated soil sample 55
4.2 Variation of liquid limit with proportion of cement kiln dust 57
4.3 Variation of plastic limit with cement kiln dust 57
4.4 Variation of plasticity index with cement kiln dust 58
4.5 Variation of MDD with cement kiln dust 60
4.6 Variation of OMC with cement kiln dust 60
4.7 Variation of California Bearing Ratio (CBR) with cement kiln dust for soaked
and Un-soaked soil sample 62
4.8 Variation of Unconfined Compressive Strength (UCS) with cement kiln dust 64
4.9 Plot of OMC against percentage of cement kiln dust 77
4.10 Plot of MDD against percentage of cement kiln dust 78
4.11 Plot of CBR (unsoaked) against percentage of cement kiln dust 79
4.12 Plot of CBR (soaked) against percentage of cement kiln dust 80
4.13 Plot of Unconfined Compressive Strength (7days) against percentage of
Cement Kiln Dust 81
12
CHAPTER ONE
1.0 INTRODUCTION
In road construction, firm subgrade and base layers are essential components of a pavement
structure. The geotechnical properties like strength and volume stability of the subgrade have
significant influence on the overall performance and durability of the pavement.
Most of the subgrade soils encountered during construction of road are unable to respond to the
imposed stresses, which has become a dominant factor responsible for the failure of pavement
especially in Nigeria (Aigbedion, 2007). Okechukwu et al., (2011) identified inadequacy of
construction materials and poor quality of construction as other factors responsible for road
failure in south eastern Nigeria. Improving the properties of these soils by the addition of
chemicals and cementitious additives becomes not only necessary but very important to make
them suitable for construction. These chemical additives range from waste products to
manufactured materials and include lime, Class C fly ash, Portland cement, proprietary chemical
stabilizers (Parson and Kneebone, 2004) and recent technology like NANO technology.
Soil improvement could be by modification or stabilization, or both. Soil modification is the
addition of modifier to the soil to alter its index properties, while stabilization is the treatment of
soil with modifier to enable their strength and durability to improve such that they become
totally suitable for construction beyond their original classification.
Beside those enumerated above, other materials having cementitous properties like Cement Kiln
Dust (CKD) have proven to have potentials to be useful in the improvement and stabilization of
soils, (Miller and Azad, 2000).
13
Cement Kiln Dust is a waste product from the manufacturing of cement. The quantities, nature
and characteristics of CKD generated depend upon a number of operational factors and
characteristics of inputs to the manufacturing process, including, the raw fed, fuel used, and the
design and operation of the cement kiln. A significant amount of tones of the this industrial by-
product is generated each year by Cement factories all over the world, Nigeria inclusive, and
this by-product have desirable properties that make it suitable for variety of beneficial use and
application including soil improvement. A significant portion of cement kiln dust continues to
be landfill as solid waste. Although disposal as landfill pose little threat to the environment,
disposal at land fill might be viewed as the discarding of a commodity, and which in turn leads to
the consumption of addition (available land) natural resource, NCASI (2003). This practice
flouts the current global cry for sustainable development in terms of economic resource
utilization.
There have been significant research on cement kiln dust (CKD) stabilized subgrade soil and its
utilization for engineering purposes, only a few studies have determined the required inputs of
cement kiln dust stabilized soil that are suitable when designing a new pavement,
(Sreekrishnavilasan and Santagata, 2006). Also, to the researcher‟s knowledge, there are limited
data on the engineering properties of CKD- stabilized soil in Nigeria.
Consequently, the aim of this work is to explore the potentials of utilizing cement kiln dust
(CKD) in improving the properties of soil for construction purpose. The work examined the fill
material used for the construction of Sankwala- Busi Road in Obanliku, Cross River State,
Nigeria, with a view of stabilizing it to obtain a subbase and base material.
14
1.2 STATEMENT OF PROBLEM
The ever increasing cost of construction materials in Nigeria and other developing countries has
created the need for research into locally and readily available materials and also on how to
convert materials considered to be waste for usage to reduce the over dependence on virgin
materials in construction.
On the other hand, cement factories exist in many parts of the world that produces over several
millions tons of cement yearly for the construction industry. Cement kiln dust, the waste product
from cement manufacture is dispose in landfill as waste; this is evident in Nfamoseng in
Akampka in Cross River State where several tons of cement is produced daily. The waste has
been proven to poses cementitous properties. It therefore provides a good alternative to the use of
cement and lime having similar properties for use in improving properties of soils, thereby
reducing cost and reducing or eliminating the problem of waste in the affected area and the
landfill space would be useful for agricultural purposes.
1.3 AIMS AND OBJECTIVES OF THE RESEACH
The primary objective of this research has been drawn from a need for greater understanding
about how Nigeria cement bye products, the cement kiln dust, considered as waste can be used in
the improvement of sandy gravel soils for sub –base / base course in Nigeria.
The objectives of this research can be enumerated more specifically as follows:
To ascertain the suitability of cement Kiln dust in sandy gravel soils improvement by
stabilization
To determine the optimal use of cement kiln dust in improving sandy gravel soils
15
To provide technical data that will encourage the use of cement kiln dust in construction
industry in Nigeria.
1.4 SIGNIFICANCE OF RESEARCH
Large quantities of cement kiln dust (CKD) are generated from cement plant in most cities of the
world. The Vanguard Newspaper,( 2011) reported that Nigeria current local cement production
is put at 10.5 million tonnes and is projected to hit 17 million tonnes in 2011. One metric ton of
cement can produce between 0.6-0.70 tons of CKD. Available literatures showed that cement
kiln dust provides a cheaper alternative to Portland cement and lime in soil improvement by
stabilization, (Miller et al, 2003). Miller (2000) showed that soil improvement by stabilization
using CKD is cheaper and easier alternative as compared to lime, when encountering
problematic soils that are collapsible, expansive and frost-susceptible. Also, cement kiln dust is
very effective in improving the engineering properties of soil, (Solanki etal; 2007). The results of
the cases mentioned above reflect mostly those of other countries. CKD is also available in
Nigeria and if this is realizable, it will find lasting solution to the failure of road due to weak sub-
base and base arising from substandard construction materials used by encouraging the use of
cement kiln dust for soil improvement without incurring excessive cost in the construction
industry, hence justifying the reason for this research. Strength indices will be modelled at the
end of the research for California Bearing Ratio and Unconfined compressive strength so that the
possibility of generating values using dependent and independent variables will be ascertained
from models, that will carter for the laborious and time consuming laboratory work.
16
1.5 SCOPE OF REASEARCH
This research would focus on the potentials of using cement kiln (CKD) in improving the
properties of sandy gravel soils for construction purposes. It examined the fill material (sandy
gravel soil as laterite) used for the construction of Sankwala- Busi Road in Obanliku, Cross
River State, Nigeria, with a view of stabilizing it to obtain a subbase and base material.
The results obtained are modelled for prediction using statistical method.
1.6 LIMITATION
Disturbed soil sample used in this study were obtained from deposits along Sankwalla– Busiri
road, in Obanliku LGA of Cross River State, Latitude 060 33
‟ and Longitude 009
0 14
‟. The
Cement kiln dust was obtained from UNICEM cement plant at Nfamoseng, in Akampka, Cross
River State. The cost of performing the strength indices test and the chemical analysis of the kiln
dust was capital intensive. Also due to the variability in the properties of cement kiln dust, results
obtained from cement kiln dust from other cement factory may be different.
17
CHAPTER TWO
2.0 LITERATURE REVIEW
This chapter contains a discussion of cement kiln dust and its potentials in improving
subgrade soils as reported in several published literatures.
2.1 CEMENT MANUFACTURING AND PRODUCTION OF CEMENT KILN DUST
Cement kiln dust is an industrial by- product of cement manufacturing process. ASTM C 150
defines Portland cement as “hydraulic cement produced by pulverizing clinkers consisting
essentially of hydraulic calcium silicates, usually containing one or more forms of calcium
sulphate as an inter-ground addition”. According to PCA (1992) cement manufacturing is simply
the conversion of calcium and silicon oxides into calcium silicates. Calcium silicates being its
primary constituents, Portland cement is produced by combining materials containing calcium
oxide, silica, aluminina and oxide at high temperature around 1450ºC. The production of
Portland cement is generally a four step process: 1) acquisition of the raw materials 2)
preparation of the raw materials; 3) pyroprocessing of the raw materials to form Portland cement
clinker; and 4) grinding of the clinker to Portland cement see fig 2.1 overlef.
18
Fig 2.1 Cement manufacturing process (Corish ,1995)
Taylor (1997) identify the raw materials for cement manufacturing to include a finely ground
mixture of limestone and clay containing 75% calcium carbonate, 15% silicon dioxide, 3%
aluminium oxide, 2% iron oxide. Minor constituents, generally less than 5% by weight of the
mixture, include magnesium, sulphur, sodium, and potassium. There can be numerous other trace
elements which generally total less than 1% of the mixture. Raw materials must be very
intimately mixed before they are introduced into the kiln. Based on the preparation of the feed
material prior to calcinations, cement kilns are classified as either wet process cement kilns or
dry process cement kilns. In wet process kilns which are generally simpler but less energy
sufficient, feed is prepared in the form of a slurry containing 30% to 40% water, (Mehta 1993).
Modern cement plants favour the dry process, which is more energy efficient than the wet
process because the water used in the slurry must subsequently be evaporated before the
19
clinkering operation. An efficient dry- kiln will consume only about 60% of energy required to
produce a ton of cement in a typical wet- process kiln (PCA, 1992). A further advancement in
cement kiln technology is the preheater/ precalciner kiln, where fuels are combusted in the
preheater system just upstream of the rotary kiln. Here the moving raw material powder is
dispersed in a stream of hot gas coming from the kiln. Initial heating to about 800ºC is carried
out in preheater using the carbon dioxide (CO2) evolved from the limestone and the hot
combustion gases from the fuel. Lower grade fuel can be used in a precalciner kiln because the
temperatures required for calcinations are much lower than the temperatures needed to fuse the
minerals into clinker.
The well- homogenized raw materials are fed into the upper end of cylindrical rotary kilns, huge
ovens that can range in size from 3.7 to 5.5m in diameter, 46 to 183m in length and rotate at 1-4
revolutions per minute. Typically the length to diameter ratio of kilns ranges from 30: 1 to 40:1.
The kiln is slightly inclined (3-4%) and rotates about its longitudinal axis. As the raw mix travels
down the kiln, it is gradually heated to 1450ºC in the burning or clinkering zone. The raw
materials are fed into the upper end of the kiln while fuel is burned in the lower end. Pulverized
coal ash is the most commonly used fuel, though oil, natural gas and lignite are also used.
Successive reactions in different regions of the kiln produce hard pellets called clinker by partial
fusion of raw materials, typically 3-20 mm in diameter, which when grounded with gypsum and
other additives produce the fine powder called Portland cement.
2.2 CEMENT KILN DUST DEFINITION AND FORMATION
Cement kiln dust (CKD) is defined by Adeln et al (1993) as a “fine particulate matter that
consists of entrained particles of clinker, raw materials, and partially calcined raw materials”.
20
Cement industries generate millions of metric tons of cement kiln dust, as a measure to control
product quality ( low alkali clinker from high alkali raw materials) and to ensure uninterrupted
operation of the plant( Kessler, 1995). It is also define by Garth et al. (2003) as a secondary
product made during the manufacture of another product (cement). As the kiln rotates the raw
materials move slowly from the upper end to the lower end at a rate controlled by the slope and
rotational speed of the kiln. In the hotter part of the kiln, potassium, sodium, chlorine and some
other elements present in the raw materials or fuel are partially or wholly volatilized. These
volatiles are not allowed to pass into the clinker. The oxygen for combustion of the fuel is
provided by the rapid flow of air, which moves against the flow of raw material. The swift gas
and continuous raw feed agitation are turbulent in nature and result in large quantities of
particulate matter being entrained in the combustion gases. These combustion gases released
from the fuel move up the kiln counter to the downward flow of raw materials. The gas flow
picks up partially burned raw materials and the volatilized materials, carrying them up the kiln.
The entrained particulate matter (as well as various precipitates) constitutes cement kiln dust
(CKD) and is subsequently removed from the kiln exhaust gases by air pollution control
equipment. The nature and quantity of CKD produced strongly depends on the raw feed, fuel
used, as well as on the design and operation of the cement kiln.
Modern cement plants are equipped with provisions like suspension heaters which largely
capture the dust (which is not removed by the dust collection system) before the gas escapes
through the chimney. This dust depending on its composition (content of alkali, sulphates or
chlorides) is recycled back into the kiln either by mixing it with the raw meal or with the fuel
(insuffation). The major factors preventing return of more dust to the kiln is the high
concentration of alkalis in the dust that would cause the alkali content of the clinker to exceed
21
the recommended allowable value (0.6%). Also the high concentrations of volatiles develop
deposits on walls of the kiln which can result in frequent shut down of the plant. Therefore plants
generate CKD as a means of removing volatile alkalis, chlorides and sulphate from the kiln
system (Kessler, 1995)
The nature and amount of CKD can be significantly affected by the design, operation and
materials used in a cement kiln. For example, a cement plant using raw materials, fuels low in
alkalis (potassium and sodium), low chlorine and sulphur will generate lesser amounts of CKD.
The CKD generated by intermitted bypass operation will typically have much higher levels of
alkalis, volatile metal salts, and oxides than CKD generated from a continuous bypass.
The production of CKD strongly depends upon the type of process and design of gas velocities in
the kiln, (PCA, 1992). Other factors such as kiln performance and dust collection system also
play vital roles. The quantity of CKD generated varies according to the amount of dust contact
and gas velocities of each kiln system. Steuch (1992 ) reported that the largest amount of dust is
generated from long dry kilns in which the dust is stirred up by chains and the gas velocities are
high. On the other hand, in preheated kilns, feed loading is high and the resulting dust contact
with kiln gases is short. Some of the wet kilns produce the lowest amount of dust, mainly
because these kilns contain pebble-size dust agglomerates that are difficult to sweep away by the
moving kiln gases. Thus the CKD generated is fairly low.
2.3 COMPOSITION OF CEMENT KILN DUST (CKD)
Cement kiln dust (CKD) is a particulate matter that is collected from kiln exhaust gases and
consist of entrained particulate of clinker, raw materials, particularly calcined raw materials, and
fuel ash enriched with alkali sulphate halides and other volatiles (Adeln et al,1993).
22
While the chemical composition of ordinary Portland cement from most of the plants in the
world is found to be remarkably consistent and despite the fact that CKD is derived from the
same raw materials as Portland cement clinker, a noticeable variation in chemical composition
and physical characteristics has been observed for CKDs obtained from different plant. This
variability can be ascribed to differences in the type of kiln operation, the dust collection facility
(e.g dust collection from the alkali bypass of precalciner kilns have been observed to be typically
coarser, more calcined, and concentrated with alkali volatiles, the location within the system
where the dust is collected, and the fuel used (Klemm, 1980).
Klemm (1980) reported that dust from gas oil-fired kilns contained higher proportions of soluble
alkalis as compared to those from coal fired kilns.
Collins and Emery (1983) also reported that there are often significant difference between total
and separated dust collected, with the finer dust particles usually having a higher concentration
of sulphates and alkalis and a lower free lime content. An assessment of the variability in the
chemical composition of fresh CKD for 63 different samples documented in literatures, shows
that CaO and SiO2 are the major constituents for all CKDs, (Sreekrishnavilasam and Santagata,
2006).
Table 2.1: Statistics on composition of fresh CKD based on 63 CKD Samples
23
The data shown in Table 2.1 indicates a large range in variation for all the oxides, as well as for
LOI and the free lime content. Compared to Portland cement, CKDs on the average are typically
characterised by higher alkali content, in particular potassium (the high alkali content, is one of
the reasons for removing the dust), and by higher sulphur content.
Haynes and Kramer (1982) conducted an extensive study on the mineralogical composition of
CKD using X-ray Diffraction patterns. They analysed 113 CKD samples from 102 plants in US
and observed that the major constituents of CKD are Calcite, and to a lesser degree, lime,
anhydrite, quartz and dolomite. They reported the average phase composition for CKD shown
below.
Table 2.2: Average phase composition of CKD(Haynes and Kramer)
Constituent % by weight
CaCO3 55.5
SiO3 13.6
CaO 8.1
K2SO4 5.9
Fe2O3 2.1
KCl 1.4
MgO 1.3
Na2SO4 1.3
CaSO4 5.2
Al2O3 4.5
KF 0.4
Others 0.7
24
Muller (1977) investigated nearly 100 European CKD and discovered four major component of
the dust. The summarized mineralogical composition of the four components are contain in
Table 2.3.
Table 2.3 Mineralogical Composition of four major component of CKD (Bhatty, 1995)
Additional data on the mineralogical composition of CKD have been reported by Baghdadi
(1990) and Konsta-Gdoutos and Shah (2003). The X-ray diffraction pattern for a CKD produced
by Arabian Cement Co. reported by Baghdadi (1990) showed calcium carbonate (calcite) as the
predominant constituent. Konsta- Gdoutos and Shah (2003) conducted X-ray diffraction studies
on several CKDs to study the hydration and properties of novel blended cements based on
cement kiln dust and blast furnace slag. The CKD samples identified as E, P, A, and X, were
representative of various plant operating conditions affecting dust composition and reactivity,
such as the feed raw materials, kiln type, and dust collection systems. Results from the XRD
analysis of the four CKD samples are as shown in figure 2.2, Overleaf. Calcite was identified as
the prevailing phase in all four samples.
25
Figure 2.2 X-Ray diffraction for four CKDs (Konsta-Gdoutos and Shah, 2003)
2.4 PROPERTIES OF CEMENT KILN DUST
2.4.1 REACTIVITY OF CKD
The reactivity of any binding material, including CKD, has been accepted as being a function of
its overall physical and chemical make up. While the overall reactions and products associated
with the interaction of CKD with water are not completely understood, and are likely to vary
depending on the particular CKD under consideration, there have been efforts to identify some
simple parameters that can be used to predict the reactivity of CKD, based on its oxide
composition. Two empirical parameters that have been proposed are the Hydration Modulus,
HM (Kamon and Nontananandh, 1991) and the Total Reactive Oxide content, TRO (Collins and
Emery, 1983).
26
Kamon and Nontananandh (1991) defined the hydration modulus as:
HM = CaO/(SiO2+Al2O3+Fe2O3) where CaO, SiO2, etc, are expressed as
percentage values. Using alite (HM~1.7) and belite (HM~2.4), the most common cementitious
phases present in Portland cement, as references they suggested that for a CKD to be
characterized as reactive, the HM should lie between 1.7 and 2.4. Based on tests conducted with
two fresh CKDs with hydration modulus 2.05 and 2.3 (free lime content is not provided), Miller
et al. (2003) reported a good correlation between the HM and the effectiveness of the CKD in
stabilizing three soils (Soils employed are CH, CL and ML as per USCS). Since cementitious
phases in the form of alite and belite are rarely found in CKD, and given that free lime is likely
the most significant source of reactivity in CKD, the literature data on CKD samples were
analyzed to establish the correlation, if any, between free lime content and HM. The result
showed that there is no direct correlation between free lime content and HM. CKDs with
practically no free lime have values of HM falling within the limits set by Kamon and
Nontananandh (1991). Conversely, CKDs with free lime content values as high as 27% are
observed to have HM outside the recommended range. Overall, these observations suggest that
the validity of the HM as an indicator of the reactivity of the CKD is debatable.
Collins and Emery (1983) performed a study to determine the effectiveness of replacing kiln
dusts (both CKD and lime kiln dust) in lime-fly ash aggregate road base system, and introduced
a second parameter, the total reactive oxide content (TRO), as an indicator of the reactivity of
CKD (note that the basis for the definition for TRO is not reported).
27
The TRO is defined as:
TRO = (CaO+MgO-LOI)-(K2O+Na2O) where CaO, MgO, K2O, Na2O and LOI are expressed as
percentage values (note that the use of the equation is not recommended by the authors for a total
alkali content greater than 6%).
Figure 2.3 Relationship between compressive strength of CKD-fly ash blends and TRO of CKDs
(Collins and Emery, 1983).
Based on this study they proposed that the higher the TRO of a CKD, the greater the potential
contribution of the CKD in terms of strength development. Figure 2.3 shows the relationship
between the 7-day compressive strength obtained and the TRO of the CKDs used. Bhatty et al.
(1996) performed a comparison between the TRO and the strength development in CKD
stabilized soils using data from the study performed by McCoy and Kriner (1971) on the use of
CKD in soil stabilization. These data show no correlation between strength developed and the
TRO. In fact the CKD (C) with the highest TRO yields the lowest strength as shown in table 2.4.
It should be however noted that the relationship between TRO and compressive strength for
28
CKD stabilized soil and CKD-fly ash blends need not be expected to be the same as both involve
different mechanisms.
Table 2.4: Summary of CKDs and soils used and strength of treated soils (McCoy and Kriner,
1971)
As reflected by the correlation coefficient, there is a clear correlation between TRO and free
lime content. These data support the hypothesis by Collins and Emery (1983) that a higher total
reactive oxide signifies a CKD with more reactive phases.
2.4.2 Loss on Ignition (LOI)
Loss on ignition is a very important property of CKD as it relate to it performance in terms of
strength. In CKD the loss on ignition (LOI), i.e., the loss in mass associated with heating to
~950°C, is contributed by chemically bound water, CO2 and noncarbonated carbon (Haynes et
al.,1995). The sources of noncarbonated carbon in CKD are the unburned clinker or the fuel oil.
Thermo gravimetric analysis (TGA) can be used to identify the exact contributions of each of
these compounds. Haynes et al. (1995) analyzed 113 fresh CKD samples from 102 plants by
TGA. The result showed that the chemically bound water is typically low in CKD, ranging from
0.4 to 3.8%. The CO2 content ranged from 4.4 to 34.4% and free noncarbonated carbon varied
29
from 0.01 to 1.83%. Hence the LOI of CKD is primarily contributed by the CaCO3 that
undergoes de-carbonation when ignited.
Bhatty et al. (1996) suggests that high LOIs typically is an indication of high carbonate and low
free lime. McCoy and Kriner (1971) reported that soil treated with low LOI and high free lime
CKD provided promising results in terms of strength.
Most CKDs have a LOI in the 20-35% range. The LOI shows significant variation, depending on
the plant operation. For example Todres et al. (1992) compared CKDs from three different kiln
types and observed that the CKDs from long wet and long dry kilns had LOI significantly greater
(25% and 30%, respectively) than that measured on CKD obtained from an alkali bypass system
(LOI~4.0%). LOI values of landfilled CKDs are generally higher compared to the parent fresh
CKD. When the CKD is exposed to the atmosphere the free lime and any other cementitious
phases present, commonly undergo hydration and carbonation. Also evaporation of moisture
from an open stock pile may result in the formation of gypsum and alkali hydroxides (Bhatty,
1996). Each of these compounds will contribute to higher values of the LOI.
2.4.3 pH
pH is a measure of the acidic or alkaline content of a substance. The chemical composition of
CKD shows that it contains significant amounts of alkalis, which are considered to be caustic. As
a result the pH of a CKD-water mixture is typically about 12.0 or greater. Miller et al. (2003)
reported pH values of 12.48 and 12.65 for two fresh CKD.
2.4.4 Particle Size Distribution
An important physical characteristic of CKD is its particle size distribution. As CKD is a waste
product that is collected from the exhaust gas stream, it is a very fine, powdery material of
30
relatively uniform particle size. PCA (1992) reported that CKD typically has a mass median
particle diameter of 10 μm even though the raw materials are much larger in mean diameter.
Data from the literature showed that the particle size distribution of CKD depends on the process
technology, method of dust collection, chemical composition of CKD, alkali content. The
particle size analysis of CKD is typically carried out using some form of sedimentation test
based on Stokes‟ law. Selections of results collected from the literatures that illustrate the range
in particle size of cement kiln dust are shown in fig 2.4 below. Typical curves for ordinary
portland cement and two microcements (Santagata and Collepardi, 1998) are also plotted for
comparison purposes. As seen in the figure, the various fresh CKDs show significant variation in
the mean particle size (D50 = 2.8 μm to 55 μm), as well as in the gradation (Cu =5-25). For the
most part, these values straddle the data for ordinary portland cement, and are greater than those
reported for microcements.
Figure 2.4 Examples of CKD particle size distributions from the literature
31
Figure 2.5 Effect of kiln type and dust collection system on particle size distribution of CKD
(Gdoutos and Shah, 2003)
The particle size of CKD depends greatly on the type of kiln system used. This is illustrated in
Figure 2.5 which shows the results of particle size analyses performed at the Construction
Technology Laboratories Inc. (CTL) (Todres et.al., 1992) on CKDs representing different plant
operations. The figures indicate that the dusts collected from dry kilns are finer than those from
wet and semi-wet/semi-dry kilns.
Corish and Coleman (1995) reported that there are typically significant variations in the alkali
concentration depending on the particle size fraction. Table 2.5 indicates the tendency for the
percentage of potassium to increase significantly in the finer fraction of CKD, suggesting that the
alkali concentration may be higher in finer CKDs. Collins and Emery (1983) have also reported
that there are often significant differences between total and separated dust collected, with the
finer dust particles usually having a higher concentration of sulphates and alkalis, and a lower
free lime content.
32
Table 2.5 Effect of particle size on alkali content of CKD (European plants) (Corish
andColeman, 1995)
2.4.5 Specific Surface Area
The surface area per unit weight of the material is termed as specific surface expressed as cm2/g
or m2/kg (Blaine). This is an indirect measure of the CKD grading. Blaine fineness values for
CKD reported in the literature vary between 2300 and 14000 cm2/g (McCoy and Kriner, 1971;
Collins and Emery, 1983; Baghdadi, 1990; Konsta-Gdoutos and Shah, 2003) and are consistently
above typical values for ordinary Portland cement (~3000-5000 cm2/g) suggesting that in the
case of CKD, particle texture and morphology may play a significant role. In an extensive study
on the stabilization effects of CKD, McCoy and Kriner (1971) found that calcined CKD has
lower fineness (2290 cm2/g) than undercalcined CKD (9370 cm
2/g). Malhotra and
Ramezanianpour (1994) reported that the specific surface areas of fly ash are a function of the
raw material and method of dust collection. For example, fly ash collected from modern
electrostatic precipitators generally has a higher specific surface area compared to dust collected
by cyclones. Considering the variability in the raw materials, process technology, fuel used, dust
collection methods associated with the generation of CKD, this observation is likely to apply
also to CKD.
33
2.4.6 Specific Gravity
The measured specific gravity of a particulate material is affected by the chemistry and structure
of the individual particles. For similar chemical composition, the particles with solid structure
tend to have greater specific gravity than particles with hollow and porous structure (Huang,
1990). Baghdadi (1990) reported that the specific gravity of CKD is typically in the range of 2.6-
2.8, less than that of Portland cement (Gs~3.15).
2.4.7 Engineering Properties
Not much literature to the researcher‟s knowledge is available on the mechanical and
engineering properties of CKD. One of the published work by Todres et al ( 1992) focused on
the compaction behaviour and permeability of CKD from three sources. Table 2.6 summarizes
some of the CKD physio- chemical characteristics of the CKD used. Todres et al (1992)
performed the standard proctor test on the three samples ( G,H,S) and discovered that
compaction of CKD respond to compaction in a manner similar to that of fine grained soils
where the dry density increased with water content up to maximum and then decreased with
further increase in water content see figure 2.11
Table 2.6 CKDs used for compaction studies performed by Todres et al. (1992)
34
Todres et al. (1992) also conducted permeability tests on the three CKD specimen compacted to
different densities directly inside the permeability moulds. The measured values of the hydraulic
conductivity, k, obtained from these tests are presented in table 2.7, and the study revealed that
for compacted CKD, low permeability values could be achieved by increasing the compaction
effort.
Figure 2.6 Typical compaction curves for fresh CKDs (Todres et al., 1992)
Table 2.7 Permeability values, k of compacted CKD sample (Todres et al., 1992)
35
2.5 UTILIZATION OF CEMENT KILN DUST
It is practically impossible for cement plant (e.g. UNICEM) to achieve 100% recycling of CKD.
Land filling continued to be the typically adopted mode of storage of CKD to accommodate the
increased production of CKD. There are several efforts by cement industries to find practical
applications of CKD due to high cost associated with the disposal and the strict environmental
regulations for management and disposal.
The re-utilization of CKD provides the opportunity to save higher quality natural
materials, reduces the cost of disposal, freeing precious landfill space and great progress toward
achieving 100% recycling scenarios being advocated by social environment policies.
The first commercial use of CKD was reported in 1912, when it was used as potassium rich
fertilizer for citrus and other crops, ( Sreekrishnavilasam and Santagata, 2006). Klemm (1982)
reported that CKD was used to produce potassium nitrate for explosives during World War 1.
Besides the cases mentioned above, several researchers have investigated the reuse of CKD in a
number of fields. However , as with other industrial waste materials, due to the large volumes of
materials involved in highway construction, there has, and continues to be great interest in
exploring the viability of using CKD in partial substitution of traditional construction materials (
Sreekrishnavilasam and Santagata, 2006).
A number of researchers have investigated the use of CKD for subgrade sludge and as
partial additive to produce blended cement for concrete construction. Also ASTM D 5050 list a
few potential applications for CKD ( e.g. soil stabilization, solidification and stabilization of
coasts materials and agricultural lime. It may be worthwhile to note that while much work has
been done in trying to identify large scale applications for CKD there are clear “barriers” to the
36
widespread utilization of CKD as a construction material. Some of these barriers have been
encountered in the past with other industrial by-products, including concerns over the
environmental impact and regulatory issues, the perception of the new material as
“experimented‟‟, and economic related concerns associated with special design and
constructions, additional training required for the design and construction team (
Sreekrishnavilasam and Santagata, 2006).
. There are also specific additional concerns for CKDs that are associated with the variability in
proportion of these materials. It has been proven that there exist no “average‟‟ CKD, and that
physical and chemical properties of CKD vary remarkably from plant to plant, depending upon
the feed raw materials, type of kiln operation, dust collection facility, and fuel used. As such, the
re-utilization of CKD is likely to be best considered on a plant-by-plant basis. Other noticeable
barrier to the utilization of CKD is its fineness which brings a concern in many geotechnical
applications due to interior properties and construction difficulties such as compaction and dust
control that typically characterize similarly graded geomaterials. Also as like other industrial
bye-products, issues like environmental stability, field verification and long term performance
have to be clearly addressed before CKD can be considered as a practical alternative to other
construction materials.
2.5.1 Cement Kiln Dust in Soil Stabilization
Lime and cement, have been extensively used as chemical admixtures in both shallow and deep
stabilization of soils, in order to improve properties such as strength and stiffness. Despite
several studies undertaken to investigate the use of CKD as an alternative to these traditional
materials for treating both clays and sands, there remain many fundamental questions regarding
37
its effectiveness as a soil stabilizer. This is partly due to the fact that CKD have chemical and
physical properties varying over a wide range.
The first investigation of CKD for soil stabilization was reported by McCoy and Kriner (1971)
who employed CKDs with different free lime contents and soils of different plasticity. The
results were compared with those obtained from hydrated lime and Portland cement, it was
concluded that the use of CKDs with appropriate composition at adequate addition levels was
promising for stabilizing soil (Sreekrishnavilasam and Santagata, 2006).The report of McCoy
and Kriner (1971) also reveals that CKDs with high free lime and low LOI provide a 7-day
compressive strength comparable to that measured on soil-cement mixtures and significantly
higher than that of hydrated lime-soil mixture. It was also noted that the presence of high alkali
content in the CKD adversely affected the compressive strength.
Studies by other researchers , Baghdadi (1990), Zaman et, al.(1992), Baghdadi (1995), Miller
and Azad (2000), Miller et, al.,(2003) testify the potential of using CKD for soil stabilization,
though at dosages (8-30%) substantially greater than those used for other admixtures.
Baghdadi (1990), reported that CKD with 5.3% free lime and LOI=26% bring about a
considerable improvement in the strength of kalolinite and reduction in the plasticity index PI of
the bentonite. Zaman et, al., (1992) working on expansive clay and CKD with LOI=28%
reported an improvement in strength and reduction in the plasticity index (PI). Miller and Azad
(2000) performed a laboratory study to evaluate the effectiveness of cement kiln dust (CKD) as
soil stabilizer, the study revealed that increase in the unconfined compressive strength(UCS) of
soil occurred with the addition of CKD. The increase in UCS was inversely proportional to the
plasticity index (PI) of the untreated soil. A reduction in plasticity index with increased CKD
38
content was observed. Collins and Emery (1983) demonstrated the effectiveness of substituting
CKD for lime in a number of lime-ash-sandy aggregate systems for subbase construction. The
result shows that the majority of the CKD-treated fly ash and aggregate mixtures resulted in
materials which are comparable in strength, durability, dimensional stability, and other
engineering properties, to those of the conventional lime-fly ash aggregate.
The possibility of using CKD in stabilizing sandy soils for pavement subgrade application was
examined by Napeirala (1983). The results from the study showed that an addition of 15% CKD
having 5.9% free CaO and MgO and 0.97% total alkalis (K2O + Na2O) ensured a compressive
strength of 0.25Mpa, which is standard practice in Poland for the subgrade within 14days of the
treatment. Bhatty et al., (1996) conducted a study on the use of cement kiln dust in stabilizing
soil. it was reported that CKD with high free lime (2.66%) and moderate alkalies ( ≈4.6%
expressed as Na2O equivalent) produced mixtures with compressive strengths comparable to
those obtained with cements and lime. CKD having low free lime (0.5%) and low alkalies (2.2%
Na2O equivalent) gave lower strengths. In general, CKDs with high free lime and moderate
alkalies gave enhanced stabilization in terms of improved compressive strengths and reduced
plasticity. It is worthy of note that higher alkalies in CKDs can counter the stabilization reactions
because of ionic interference, (Rahman et al., 2011).
An experimental study to investigate the effectiveness of CKD for stabilizing kaolinite clay was
performed by Peethamparan et al., (2006). Two CKDs having free lime content of 13.85 and
5.32% and LOI of 14.22 and 29.65% respectively were employed. The percentages of CKD
varied from 8 to 25% by dry weight of clay. They reported that the strength of CKD treatment
kaolinite clay is proportional to the CKD content and also the free lime content. A considerable
39
improvement in strength of kaolinite was observed. Also the CKD with higher free lime content
(13.85) gave a compressive strength twice that of a CKD with lower free lime content (5.32%).
Miller et al, 2003 reported the field evaluation of CKD treated subgrade performed in 2000 by
the Oklahoma DOF. The results of sandy lean clay (PI≈15-30%) treated with three different
CKDs confirmed laboratory observations(significant improvement in the properties that can be
obtained using CKD, but also the great variability in stabilization results depending on the type
of CKD used), and highlighted a number of issues relevant to construction (e.g problem posed by
wind blown CKD).
Documented results in literature indicate that, due to wide variation in the physical and chemical
properties of CKD, general conclusions on its validity as a soil stabilizer cannot be drawn.
Moreover, the investigations conducted so far have been limited to comparative experimental
investigation and mechanisms responsible for improved behaviour remains unclear. These
factors have essentially to date prevented a more extensive use of CKD in soil stabilization, (
Sreekrishnavilasam and Santagata, 2006).
One of the most noticeable phenomena that are mentioned in almost every study on lime, fly ash
or CKD stabilization is the ability of the binder to change the plasticity characteristics of the soil.
Study by researchers (Macoy and Kriner, 1971; Baghdadi, 1990; Zaman et al, 1993; Miller and
Azad, 2000) has shown that the addition of CKD to moderately plastic soil generally causes an
immediate increase in plastic limit and reduction in plasticity index. Trend of both increasing and
decreasing liquid limit with CKD percentage depending on the soil used are reported by
researchers (Zaman et al.,1992; Miller and Azad, 2000; Santagata and Bonet, (2002). Al-Refeai
and Al-Karni (1999) conducted an experimental study on the utilization of cement kiln dust for
40
ground modification, by adding different percentages of CKD on the engineering properties of
three different problematic soils in central region of Saudi Arabia. The result shows significant
improvement in their engineering properties. The plasticity index and liquid limit of clay
decreases significantly as the percentage of added cement kiln dust increases the optimum
moisture content and maximum dry density of dune sand increases as the cement kiln dust
increases. The unconfined compressive strength was found to increase substantially during the
first seven days.
Solanki and Zaman (2009) conducted a laboratory performance evaluation of subgrade soils
stabilized with sulphate-bearing cementitious additives namely lime, class c fly ash and cement
kiln dust. The study shows that 15% CKD gives the highest unconfined compressive strength
and resilient modulus for 28days curing. There was however decrease in durability for soils
stabilized with CKD after 60days curing. In general, the plasticity characteristics of soil were
dramatically modified by the addition of CKD. The increases or decreases in limits are not only
dependent on the type of soil treated but also on the chemical composition (majorly the free lime
cement) of CKD. Also the addition of CKD brings about improvement in the engineering
properties of soil, (mainly California bearing ratio and unconfined compressive strength).
2.5.2 CONTROLLED LOW STRNGTH MATERIALS
Controlled low-strength materials (CLSM) are defined as self-compacting, cementitious
materials. Used primarily as backfill in lieu of compacted fill (ACI, 1994). CLSM is defined by
ASTM as a “mixture of soil, cementitious materials, water and sometimes admixtures that
harden into materials with higher strength than the soil but less than 8270kpa (1200psi)‟‟. In
general, CLSM describes a fill technology that is used in place of compacted backfill and that
41
contains cement, fly ash, fine aggregate, water and sometimes chemical admixtures mixed in
varying proportions to meet the requirements of strength and flexibility, ( Sreekrishnavilasam
and Santagata, 2006). The primary property of CLSMs in all applications according to
Sreekrishnavilasam and Santagata, (2006), is their flexibility/self levelling mixture. CLSMs
requires no compaction and are therefore ideal for use in tight and restricted-access areas where
placing and compacting soil or granular fill is difficult or even impossible. The consistency of
flowable fills used in geotechnical applications is similar to that of a lean grout or slurry, yet
several hours after placement the material hardens enough to support traffic loads without
settlement.
CLSM is generally used in non-structural applications below grade for which low strengths are
desired. In these cases, the ultimate strength of the the CLSM is intended to be no greater than
that of the surrounding soil, ASTM standards, various researchers and ready mixed concrete
companies have reported various strength for CLSM. The recommendations for a minimum
strength are intended to ensure that CLSM has adequate bearing capacity and does not deform
excessively under load. Maximum strength recommendations (which typically vary anywhere
between 345 and 690kpa at 28days) ensure that CLSM can be removed with conventional
excavating equipment, ( Sreekrishnavilasam and Santagata, 2006).
There have been significant efforts in utilising various industrial by-products in CLSM. One of
the most commonly used by-products in CLSM is fly-ash that is found to improve flowability
and reduces segregation and bleeding, ( Sreekrishnavilasam and Santagata, 2006).
Current researches now focus on the feasibility of using CKD as one of the constituent in CLSM.
In a research to investigate the utilization of CKD in CLSM, Al-jabri et al.,(2002) conducted a
42
preliminary study on the use of one CKD (with LOI=9.67%) in CLSM in replacement of cement.
The two mixes tested by these authors (one with 249kg/m3
of CKD and 47kg/m3 of cement; the
other with only 296kg/m3 of CKD) both showed satisfactory strength at 28days (2735kpa and
1045kPa respectively). It was concluded that CKD with good cementry properties could be used
as a partial or full substitute for cement in flowable fill. Another research to investigate the
combined use of CKD (LOI=25.1%) and fly ash at different ratio in CLSM was conducted by
Pierce et al., (2003). This study involved evaluation of both the fluid and hardened state
properties of the CLSM, and comparison with the results obtained with fly ash. The results
showed that high flowability and relatively rapid setting time could be achieved with CKD-
CLSM mixture. Also the 28-day compressive strength was observed to increase with CKD-fly
ash ratio (for CKD: FA= 1:12, 1:6 and 1:1 the average 28 day compressive strength were 45, 71
and 356 kpa). It was concluded that by varying the amounts of CKD, fly ash and water, it was
possible to create a self-consolidating material with a wide range of hardened and fluid state
properties for field application. Katz and Kovler (2004) also compared the properties of CLSM
mixes using different waste materials like fly ash, CKD, dust from asphalt quarry, bottom ash
and quarry waste. The results of this study indicated that CKD- CLSM were characterized by
higher water demands, lower bleeding, high shrinkage and higher setting time compared to the
fly ash- CLSM. In terms of strength the CKD and fly-ash based CLSMs showed comparable
results.
2.5.3 HIGHWAY BASE AND SUBBASE
No much study has been done on the utilization of CKD in highway bases and subbases. Collins
(1983) carried out an extensive study in order to determine the effectiveness of substituting CKD
for hydrated lime in fly ash-aggregate road base mixture. 33 CKD samples were selected for the
43
study. Optimum ratios for CKD-fly ash combinations were determined on the basis of the
strength developed in test cylinders. CKD-fly ash control mixes were prepared with a typical
class F bituminous coal fly ash. These were compared with a control mix consisting of one part
by weight of commercial high calcium hydrated lime and four parts by weight of same fly ash.
Only11 of the 45 kiln dust exhibited lower 7-day strengths than the lime-fly ash control mix. In
terms of long term strength development similar results were reported for both the kiln dust-fly
ash mixes and the lime-fly ash aggregate. Most CKD-fly ash aggregate mixture were found to be
dimensionally stable over extended period regardless of whether samples were submerged or
cured in moist room. Volume changes were found to be generally negligible or comparable in
magnitude to those measured on conventional lime- fly ash mixtures. Considerable volumetric
expansion was documented for mixtures manufactured with CKD with high sulphate content
(10% or greater). CKD-fly ash aggregate test mixes showed excellent freezer than durability,
except for two mixes in which high sulphate, high alkali cement kiln samples were used. In
general, the study concluded that CKDs having a sulphate content greater than or equal to 10%
should not be considered or should be used with utmost caution in a pozzolanic base system.
2.5.4 BLENDED CEMENTS AND CONSTRUCTION PRODUCTS
Researchers have carried out a lot of study on the use of CKD in blended cements (e.g Bhatty,
1983,1986; Sanduo, 1986; Daughterty and Funnel, 1983;Klemm 1980).
Bhatty (1983,1986) conducted a series of studies on the use of CKD in blended cements, and
published variable results. Cements blended reportedly had reduced strength, setting time and
workability. The strength loss was attributed to the presence of high alkalis in the dust. The
addition of fly ash with CKD lowered the alkalis contents and resulted in improved strength.
44
Additions of slag generally reduced workability but improved blends strengths because of
activation from the high lime content of CKD. The results also demonstrated that particular
ratios of alkalis, chlorides, and sulphates are important for better performance of cement blends.
A high alkali content results into increased susceptibility to alkali-aggregate reaction, which may
be blocked by the slag or fly ash constituents. Bhatty (1986) suggested that the use of finely
ground limestone and cement kiln dust should be allowed in standard ASTM specifications for
Portland cement.
Daugherty and Funnel (1983) investigated the incorporation of low levels of By-products
in Portland cement and the effects of cement quality, cement, concrete, and aggregate. This was
done by grinding and blending CKD with Portland cement. Their data showed no evidence of
adverse effects on blends setting time, soundness, or shrinkage with up to 10% CKD addition.
These blends strength however, varied probably because of variations in composition of dust
used in these blends.
Strung et al., (1985) investigated the role that volatile alkalis (Na2O, K2O) present in CKD plays
on blends properties. They reported that when salts other than sulphate are present, both of these
alkalis are preferentially incorporated into C3A to give orthorhombic form. N2O orthorhombic
C3A is retarded in hydration while that of K2O is accelerated. They also reported that when these
alkalis are present as sulphates, cement setting as well as compressive strength development is
accelerated.
Klemms (1980) reported that blended cement containing CKD can work out satisfactory
only if the alkali content is not excessive, otherwise fly ash or blast furnace slag should be added
to reduce potential alkali-aggregate reaction problem. El-sayed et al.,(1991) investigated the
45
effect of CKD on the compressive strength of cement paste and on the corrosion behaviour of
embedded reinforcement. They reported that substitution of up to 5% by weight of cement by
CKD produced no adverse effect on the cement paste strength or on the reinforcement passivity.
Daous (2004) investigated the utilization of CKD and fly ash in cement blends at various ratios.
These blends were tested for their water requirements for normal consistency, initial setting
times, and compression and tensile strength were compared to those of Portland cement. Test
results showed that blends containing as low as70% Portland cement can still exhibit adequate
mechanical strength if only CKD is used as the blending waste material.
2.5.5 ASPHALT PAVEMENT
When it is used as mineral filler in asphalt pavements, cement kiln dust can reduce asphalt
cement volume requirements by 15-25%. The CKD provides stripping resistance for the
pavement (Klemm, 1993). CKD and asphalt binder can also be combined to create low ductile
mastic asphalt that has been successfully used in Europe for bridge deck waterproofing and
protection (Davis and Hooks, 1975).
2.5.6 ACTIVATOR FOR POZZOLANS
Researchers, (Sprouse, 1984, Akin 1995, Gdouts et al.; 2003 etc) have carried out studies on the
use of CKD as an activator for latent pozzolanic material. Latent hydraulic materials develop
pozzolanic activity and act as hydraulic cement once their glass network disintegrates when
attacked by OH- ions. The solubility of Si, Ca, Al and Mg are functions of pH. At a pH lower
than 11.5, the equilibrium solubility of silica is low and slag does not dissolve. As a result, more
Ca2+
and Mg2+
enter into solution and impermeable aluminosilicate coating covers the surfaces
of the slag grains inhibiting further hydration. Hence, a chemical activator is required for further
46
hydration of slag. Activators generally include all alkali hydroxides and salts, with the least
soluble salts being the most effective. The high alkali and sulphate content make CKD a
potential candidate as an activator for pozzolanic materials; ( Sreekrishnavilasam and Santagata,
2006).
Series of experimental studies carried out by Gdoutos et al (2003) on the interaction of CKD
with slag in order to explore the feasibility and approach for developing a new generation of
more durable CKD-activated slag blends records that CKD can be successfully utilized to
activate blast furnace slag. CKD concrete, with good performance in terms of setting time and
mechanical properties, can be produced with CKD covering a wide range of chemical
composition and finess and can be regarded as a potential future material to ordinary Portland
cement.
2.5.7 MINING INDUSTRY
Studies by the U.S. Bureau of Mines (Haynes and Kramer, 1982) have documented the use of
CKD as partial hydraulic filler for backfilling coal mine shafts and tunnels. This application can
potentially use large quantities of CKD.
2.5.8 SLUDGE STABILIZATION
Another of the most common uses of cement kiln dust (CKD) is for the stabilization and
solidification of waste. Stabilization / solidification of waste( solid or liquid) or contaminated
soil generally refers to the process of reducing the mobility of hazardous substances and
contaminants in waste through both physical and chemical means by the use of additives so as to
yield a product or material suitable for land disposal or for other beneficial uses. CKD has
reportedly been used in conjunction with other cementitious products such as slag cement, lime
47
kiln dust, fly ash, hydrated lime and Portland cement in the stabilization of contaminated soils
and sludges, (Mackay and Emery, 1992). Nicholson (1982), developed a CKD-fly ash process
that is applicable to sewage sludge solidification which is also appropriate for other wastes. A
method of solidifying and immobilizing hazardous waste in an organic matrix using CKD as one
of the component has been suggested by Funderburk (1990). Kigel et al. (1994) developed a
chromium ore waste treatment utilizing CKD.
2.6 DESIGN AND ANALYSIS OF EXPERIMENTS
2.6.1 Strategy of experimentation
Literally, an experiment is a test. Investigators perform experiments in virtually all fields
of inquiry, usually to discover something about a particular process or system. More formally, an
experiment is a test or series of tests in which purposeful changes are made to the input variables
of a process or system so that we may observe and identify the reasons for changes that may be
observed in the output response (Montgomery2005).
Experimentation plays an important role in product realization activities, which consist of
new product design and formulation, manufacturing process development, and process
improvement. The objective in many cases may be to develop a robust process, that is, a process
affected minimally by external sources of variability.
As an example of an experiment, suppose that a metallurgical engineer is interested in
studying the effect of two different hardening processes, oil quenching and saltwater quenching,
on an aluminium alloy. Here the objective o f the experimenter is to determine which
48
quenching solution produces the maximum hardness for this particular alloy. The engineer
decides to subject a number of alloy specimens or test coupons to each quenching medium and
measures the hardness of the specimens after quenching. The average hardness of the specimen
treated in each quenching solution will be used to determine which solution is best. As we
consider this simple experiment, a number of important questions come to mind:
Are these two solutions the only quenching media of potential interest?
Are there any other factors that might affect hardness that should be investigated or controlled in
this experiment (such as, for example, the temperature of the quenching media)?
How many coupons of alloy should be tested in each quenching solution?
How should the test coupons be assigned to the quenching solutions, and in what order should
the data be collected?
What method of data analysis should be used?
What difference in average observed hardness between the two quenching media will be
considered important?]
All of these questions, and perhaps many others, will have to be answered satisfactorily before
the experiment is performed.
Experimentation is a vital part of the scientific (or engineering) method. Now there are
certainly situations where the scientific phenomena are so well understood that useful results
including mathematical models can be developed directly by applying these well – understood
principles (Montgomery 2005). Continuing, he reported that the models of such phenomena that
49
follow directly from the physical mechanism are usually called mechanistic models. A simple
example is the familiar equation for current flow in an electrical circuit, Ohm‟s law, E = IR.
However, most problems in science and engineering require observation of the system at
work and experimentation to elucidate information about why and how it works. Well – designed
experiments can often lead to a model of system performance; such experimentally determined
models are called empirical models, he posited.
In general, experiments are used to study the performance of processes and systems. The process
under consideration can be as a combination of operations, machines, methods, people, materials
and other resources that transforms some input (often a material) into an output that has one or
more observable response variables. Some of the process variables and material properties x1, x2,
. . . xp are controllable, whereas other variables y1, y2, . . . yp are uncontrollable (although they
may be controllable for purposes of a test).
The objectives of the experiment may include the following:
Determining which variables are most influential on the response y.
Determining where to set the influential x‟s so that y is almost always near the desired nominal
value.
Determining where to set the influential x‟s so that variability in y is small.
Determining where to set the influential x‟s so that the effects of the uncontrollable variables y1,
y2, . . . yp are minimized.
The process of achieving the above objectives in the experimentation of a system or process is
termed Characterizing of a process. That is, in characterization experiment, the experimenter is
50
usually interested in determining which process variables affect the response (Montgomery
2005). The next logical step is to optimize, that is, to determine the region in the important
factors that leads to the best possible response. For example if the response is yield, the
experimenter would look for a region of maximum yield, whereas if the response is variability in
a critical product dimension, the experiment would seek a region of minimum variability.
Once we have found the region of the optimum, a second experiment would typically be
performed. The objective of this second experiment is to develop an empirical model of the
process and to obtain a more precise estimate of the optimum operating conditions for the
controllable and uncontrollable variables. This approach to process optimization is called
response surface methodology, (Montgomery, 2005).
2.6.2 Mixture Designs
Many products are mixtures of several components (ingredient). Characteristics of the products
such as the strength of steel, the efficacy of a chemical pesticide, or the viscosity of a liquid
detergent, depend only on the relative proportions of the components in the mixture. Studying
changes in a product's properties caused by varying the ingredient proportions is the objective
of performing mixture experiments (Cornell, 1991).
The inherent restriction that the sum of the component proportions equal unity creates
different design strategies than are usually employed with independent factors where factorial
arrangements are quite common. Experimental designs for exploring the entire mixture simplex
region as well as for exploring only a subregion of the simplex are presented. In those cases
where four or more components are considered and a subregion is to be investigated, computer-
aided designs are the rule rather than the exception. Design criterion base on the properties
51
(variance and bias) of the prediction equation are mentioned briefly and some suggestions are
made for future research in mixture experiments.
In mixture experiments, the levels of individual components of the mixture are not independent
[22, 24, 109]. For example, if x1, x2. . . xp denote the proportions of p components of a mixture,
then;
0 ≤ 𝑥𝑖 ≤ 1 𝑖 = 1, 2, ⋯ , 𝑝; and 2.1
𝑥1 + 𝑥2 + ⋯ 𝑥𝑝= 1 (i.e 100%) 2.2
With three components, pi (i = 1, 2, 3), the mixtures space is a triangle with vertices
corresponding to formulations that are pure blends (mixtures that are 100% of a single
component.) The mixture design dependent composition, as given in Equation (2.2) is usually
illustrated using simplex designs. They are also used to study the effects of mixture components
on the response variable (Montgomery, 2005).
2.6.3 Simplex Designs
Simplex is the structural representation (shape) of the line or plane joining the assumed
positions of the constituent materials (atoms) of the mixtures (Jackson, 1983, Obam, 2006). They
are used to study the effects of mixture components on the response variable (Montgomery,
2005).
A (p, m) simplex lattice design for p components consist of points defined by the following
coordinate settings; the proportions assumed by each component take the m + 1 equally spaced
values from 0 to 1.
52
That is,
𝑥𝑖 = 0,1
𝑚,
2
𝑚, ⋯ , 1 𝑖 = 1, 2, ⋯ , 𝑝 2.3
Where,
p = Number of components of the mixture
m = Maximum possible number of the components which the mixture can be composed of
(Mixture level). For example, when p = 3 and m = 2; then, the possible number of runs is:
(x1, x2, x3) = (1, 0, 0), (0, 1, 0), (0, 0, 1), (½, ½, 0), (½, 0, ½), (0, ½, ½) = 6 runs
These can be shown in a simple lattice as
Figure 2.7: (3, 2) Lattice Design
For a (3, 3) lattice design, the number of runs is:
(x1, x2, x3) = ( 1,0,0); (0,1,0); (0,0,1); (1/3, 1/3, 1/3); (1/3, 0, 2/3); (2/3, 0, 1/3); (0, 1/3, 2/3); (0,
2/3, 1/3); (1/3, 2/3, 0); (2/3, 1/3, 0) = 10 runs
The simple design lattice for (3, 3) lattice is shown in Figure 2.8
(1/2,0,1/2)
x3 = 1
(0,0,1)
(0,1/2,1/2)
(1/2,1/2,0)
(1/2,1/2,0)
x2 = 1
(0,1,0)
x1 =1
(1,0,0)
53
Figure 2.8: (3,3) Lattice Design
Simple lattice designs combinations (p, m) such as (4, 2), (4, 3), (4, 4), (5, 2), (5, 3), . . . (i, j) for
i = 1 – n and j = 1 – n can be obtained. The combinations depend on the number of components
of the mixture, p and the level of combination (mixture level or order of the lattice), m. In
general, the number of points in a (p, m) simplex lattice design is given as:
𝑁 = 𝑝 + 𝑚 − 1 !
𝑚! 𝑝 − 1 ! 2.4
Where,
N = Number of points in a simplex lattice,
m = Number of levels individual components.
P = Number of components/variables in the lattice.
x2 = 1
(0, 1, 0)
(2/3, 1/3, 0)
(1/3, 2/3, 0)
(0, 2/3, 1/3) (0, 1/3, 2/3)
x3 = 1
(0, 0, 1)
(2/3, 0, 1/3)
(1/3, 0, 2/3)
x1 =1
(1, 0, 0)
(1/3, 1/3,
1/3)
54
An alternative to simplex lattice design is simplex centroid design (Scheffe, 1963, Kiefer
1959). Mixture models differ from the usual polynomials employed in response surface work
because of the constraint (Montgomery 2005, Myers et al 2004, Cornell 1991, Obam, 2006).
𝑥𝑖 = 1 2.5
The standard forms of the mixture models (Scheffe‟s Models) that are in widespread use
are (Montgomery 2005):
𝐿𝑖𝑛𝑒𝑎𝑟: 𝐸 𝑦 = 𝛽𝑖𝑥𝑖
𝑝
𝑖=1
2.6
𝑄𝑢𝑎𝑑𝑟𝑎𝑡𝑖𝑐 𝐸 𝑦 = 𝛽𝑖𝑥𝑖
𝑝
𝑖=1
+ 𝛽𝑖𝑗 𝑥𝑖
𝑝
𝑖<𝑗
𝑥𝑗 2.7
Full cubic:
𝐸 𝑦 = 𝛽𝑖𝑥𝑖
𝑝
𝑖=1
+ 𝛽𝑖𝑗 𝑥𝑖
𝑝
𝑖<𝑗
𝑥𝑗 + 𝛽𝑖𝑗 𝑥𝑖
𝑝
𝑖<𝑗
𝑥𝑗 𝑥𝑖 − 𝑥𝑗
+ 𝛽𝑖𝑗𝑘 𝑥𝑖
𝑝
𝑖<𝑗 <𝑘
𝑥𝑗 𝑥𝑘 2.8
Special cubic:
𝐸 𝑦 = 𝛽𝑖𝑥𝑖
𝑝
𝑖=1
+ 𝛽𝑖𝑗 𝑥𝑖
𝑝
𝑖<𝑗
𝑥𝑗 + 𝛽𝑖𝑗𝑘 𝑥𝑖
𝑝
𝑖<𝑗 <𝑘
𝑥𝑗 𝑥𝑘 2.9
55
Where,
𝛽𝑖 = Expected response to the pure blend 𝑥𝑖 = 1 and 𝑥𝑗 = 0 when 𝑖 ≠ 𝑗
𝐸 𝑦 = Expected response
The portion 𝛽𝑖𝑥𝑖𝑝𝑖=1 is called the linear blending portion. When curvature arises from
nonlinear blending between component pairs, the parameters 𝛽𝑖𝑗 represent either synergistic or
antagonistic blending. Higher order terms are frequently necessary in mixture models because
(1) the phenomena studied may be complex and (2) the experimental region is frequently the
entire operability region and are therefore large, requiring an elaborate model (Montgomery
2005). Mixture models which are known as Scheffe‟s models from mere observation are distinct
from ordinary polynomials by the absence of independent constant variables in the models.
2.6.4 Derivation of Scheffe’s 2nd
degree Models
The mixture models are most times referred to as Scheffe‟s Models. The Scheffe‟s Model for 2nd
degree polynomial is as follows: Assuming our mixture models are possesses curvature in the
system then a polynomial of higher degree such as given below must be used.
E y = 𝛽0 + 𝛽𝑖𝑥𝑖2
𝑝
𝑖=1
+ 𝛽𝑖𝑗 𝑥𝑖
𝑝
𝑖<𝑗
𝑥𝑗 + ⋯ + ∈ 2.10
i.e. Equation of independent variable for 2nd
degree polynomial.
Where,
E(y) = Ff = f( CKD, Soil),
Let CKD variable be x1
Soil variable be x2,
56
Number of components of the mixture, p = 2
𝐸(𝑦) ⇒ 𝐹𝑓 = 𝑓 𝑥1, 𝑥2
Expanding Equation 2.10, we have:
⇒ 𝐹𝑓 = 𝛽0 + 𝛽1𝑥1 + 𝛽2𝑥2 + 𝛽12𝑥1𝑥2 + 𝛽11𝑥12 + 𝛽22𝑥22
2 2.11
The constrain of Scheffe‟s equation for mixture is that 𝑥1 + 𝑥2 + ⋯ 𝑥𝑝= 1
⇒ 𝑥1 + 𝑥2 = 1 2.12
Thus, let
𝛽0 = 𝛽0 ∙ 1 = 𝛽0 𝑥1 + 𝑥2
Also, it can be seen that
𝑥12 = 𝑥1𝑥1
𝑏𝑢𝑡, 𝑥1 = 1 − 𝑥2
⇒ 𝑥12 = 𝑥1 ∙ 1 − 𝑥2
⇒ 𝑥12 = 𝑥1 − 𝑥1𝑥2
Similarly,
𝑥22 = 𝑥2 − 𝑥1𝑥2
Putting these expressions into the general response expression,
⇒ 𝐹𝑓 = 𝛽0 𝑥1 + 𝑥2 + 𝛽1𝑥1 + 𝛽2𝑥2 + +𝛽12𝑥1𝑥2 + 𝛽11 𝑥1 − 𝑥1𝑥2 + 𝛽22 𝑥2 − 𝑥1𝑥2
57
= 𝛽0 + 𝛽1 + 𝛽11 𝑥1 + 𝛽0 + 𝛽2 + 𝛽22 𝑥2 + 𝛽12 − 𝛽11 − 𝛽22 𝑥1𝑥2
Let
𝛽0 + 𝛽1 + 𝛽11 = 𝜇1; 𝛽0 + 𝛽2 + 𝛽22 = 𝜇2; 𝛽12 − 𝛽11 − 𝛽22 = 𝜇12;
⇒ 𝐹𝑓 = 𝜇1𝑥1 + 𝜇2𝑥2 + 𝜇12𝑥1𝑥2 2.13
This can be put in compact form as:
𝐹𝑓 = 𝜇𝑖𝑥𝑖
𝑝
𝐼<𝑖<𝑝
+ 𝜇𝑖𝑗 𝑥𝑖
𝑝
1<𝑖<𝑗 <𝑞
𝑥𝑗 2.14
Equation 2.14 is the same as Equation 2.10. They are called 2nd
Scheffe‟s models for two
variables – response.
2.6.5 Least Squares Estimation of the Model Parameters
This procedure is referred to as model fitting (Montgomery, 2005; Myers and Montgomery,
1997). The method of least square chooses the coefficients 𝜇𝑖 and 𝜇𝑖𝑗 in Equation 2.14 so that the
sum of the squares of the errors between the experimental expected response (𝐹𝑓𝑖 and 𝐹𝑓𝑖𝑗 ) and
predicted expected response (𝐹 𝑓𝑖 and 𝐹 𝑓𝑖𝑗 ) is minimized.
The least square function is expressed as:
𝐿 = 𝐹𝑓 − 𝜇𝑖𝑥𝑖
𝑝
𝐼<𝑖<𝑝
+ 𝜇𝑖𝑗 𝑥𝑖
𝑝
1<𝑖<𝑗 <𝑞
𝑥𝑗
2
2.15
The function L is to be minimized with respect to 𝜇𝑖(𝑖 = 1 − 2) and 𝜇𝑖𝑗 (1 ≤ 𝑖 ≤ 𝑗)
58
The least square estimator must satisfy
𝑑𝐿
𝑑𝜇𝑖 𝑖=1−2
= −2 𝐹𝑓𝑖 − 𝜇𝑖𝑥𝑖
𝑝
𝐼<𝑖<𝑝
+ 𝜇𝑖𝑗 𝑥𝑖
𝑝
1<𝑖<𝑗 <𝑞
𝑥𝑗
𝑛
𝑖=1
= 0 2.16
And
𝑑𝐿
𝑑𝜇𝑖𝑗 1<𝑖<𝑗 <𝑞
= −2 𝐹𝑓𝑖 − 𝜇𝑖𝑥𝑖
𝑝
𝐼<𝑖<𝑝
+ 𝜇𝑖𝑗 𝑥𝑖
𝑝
1<𝑖<𝑗 <𝑞
𝑥𝑗
𝑛
𝑖=1
𝑥𝑖𝑥𝑗 = 0 2.17
Equations 2.16 and 2.17 would result to a system of homogenous equations, which when solved
would give the values of the unknowns: 𝜇𝑖 𝑎𝑛𝑑 𝜇𝑖𝑗 of Equation 14.0. This procedure is very
adaptable to a degree of polynomials.
59
CHAPTER THREE
3.0 MATERIALS AND METHODS
3.1 SAMPLE COLLECTION
The soil used for this study was collected by method of disturbed sampling at average depths of
1 – 1.2m, from deposits along Sankwalla– Busiri road, Obanliku, Cross River State on lie
Latitude 060 33
‟ and Longitude 009
0 14
‟. The cement kiln dust used was obtained from United
Cement Company of Nigeria (Unicem), Calabar.
3.2 MATERIALS PREPARATION
The collected soil was air – dried and stored in jute bags in workshop 3 of the Department of
Civil Engineering, Cross River University of Technology (CRUTECH), Calabar. The Kiln dust
collected was stored in a waterproofing bag.
3.3 EXPERIMENTAL TESTS
In preparation of all specimens, the required amounts of kiln dust by dry weight of soil and soil
were measured and mixed in the dry state before addition of water and conduction of any test.
The various tests performed on the soil are as follows:
Sieve Analysis
Atterberg Limits
Compaction
California Bearing Ratio (CBR)
Unconfined Compressive Strength (UCS)
60
All tests were performed in accordance with the BS1377 (1990), BS1924 (1990) and modified
by the Nigerian General Specification (1997) for Road & Bridges.
Specimens for the Unconfined Compressive Strengths (UCS) and the California Bearing Ratio
(CBR) were prepared at maximum dry density (MDD) and optimum moisture content (OMC) at
the Proctor compaction. The chemical composition of the kiln dust was assessed through the X-
ray diffraction method (XDM).
3.3.1 SIEVE ANALYSIS
Sieve analysis test was carried out to determine the particle size distribution of the soil. The test
displayed the percentage of different soil samples that were retained on each sieve. About 300g
of the oven- dried soil sample was washed through sieve No 200. The remnants retained on the
sieve No 200 were later oven – dried, and thereafter sieved mechanically. The retained samples
were weighed and the percentage retained was calculated in relation to original weight as
follows:
Percentage Passing = weight retained *100/original weight.
3.3.2 LIQUID LIMIT
Liquid limit signifies the percentage of moisture at which the sample changes, with decrease in
moisture content from a viscous state to plastic one. It is also referred to as a minimum moisture
content at which the soil will flow under its own weight.
About 50g of soil passing through sieve 425µmm were used for each liquid limit test. The soil –
kiln dust mixture was mixed with portable water to form a uniform paste. A portion of the
resulting mortar was placed in the Casagrande liquid limit apparatus, grooved as per
61
specification and the lever lifted up and drooped continually until the groove closed to the
standard „V‟ shaped while the number of blows to achieve this was recorded. This was repeated
for increased moisture content and portion for moisture content were taken. On the plot of
moisture content versus number of blows, the moisture content corresponding to 25 blows gave
the liquid limit. The results of the liquid limits at different percentages of kiln dust are presented
and discussed in chapter Four.
3.3.3 PLASTIC LIMIT
Plastic limit signifies the percentage of moisture at which the sample changes, with decreasing
weight, from plastic to solid to semi – solid state. In this condition, the soil mortar begins to
crumble when rolled into thread of 3mm diameter. At plastic limit, the moisture will not separate
the soil particles but will produce enough surface tension to give contact pressure between the
solid grains and thus cause the soil to act as a semi- solid.
To about 50g of soil passing sieve 425µmm was added CKD at various percentage requirement.
The soil – kiln dust mix was with sufficient quantity of water till it became plastic and can be
easily moulded with fingers. A portion of it was rolled on a glass plate until a thread of about
3mm diameter was formed. The moisture content at which such threads started showing signs of
crumbling was determined as the plastic limit. The results of plastic limit at different percentages
of kiln dust are presented and discussed in chapter Four.
3.3.4 PLASTICITY INDEX
The Plasticity Index is simply the numerical difference between the liquid limit and the plastic
limit for a particular material and indicates the magnitude of the range of moisture content over
which the soil remains plastic. It is a measure of the cohesive qualities of the binder resulting
62
from the clay content. Also, it gives some indication of the amount of swelling and shrinkage
that will result in the wetting and drying of that fraction tested. If some soils do not have
sufficient mechanical interlock, they require amounts of cohesive materials to give a satisfactory
performance. A deficiency of clay binder may cause ravelling of gravel wearing courses during
dry weather and excessive permeability.
PI=LL-PL
3.3.5 COMPCTION TEST
Compaction is the pressing of soil particles close to each other by mechanical methods. Air is
expelled from the void spaces in the soil mass during compaction and, therefore, the mass
density is increased. Compaction of a soil mass is done to improve its engineering properties. It
generally increases the shear strength of the soil, and hence the stability and bearing capacity,
Arora (2009). The density achieved by compressing the soil is called maximum dry density
(MDD). The optimum moisture content (OMC) is the moisture content at which the maximum
dry density is achieved under a compactive effort.
For all the soil – kiln dust mixtures, standard Proctor compaction test was carried out. This is
because the required energy is easily achieved in the field, Musa (2008). Compaction was done
in three layers, each receiving 25 blows of the standard 2.5Kg rammer falling over a height of
300m. The collar of the mould was then removed and the compacted sample weighed while
specimen was taken around the middle to determine the corresponding moisture content. This
procedure was repeated with varying moisture contents until compacted sample started falling.
Results of compaction tests at different percentages of kiln dust are also presented and discussed
in chapter Four.
63
3.3.6 CALIFORNIA BEARING RATIO
The California Bearing Ratio (CBR) Test is a measure of the strength of a compacted soil sample
against that of crushed rock. It is measured by an empirical test devised by the California State
Highway Association. The CBR is obtained by measuring the relationship between the force
when a plunger of cross sectional area of 1935mm2 was made to penetrate the soil at a given rate.
It is the resistance to a penetration of 2.5mm of a standard cylindrical plunger with a diameter of
49.6mm, which is expressed as a percentage of the known resistance of the plunger to various
penetration s in crushed aggregate of 13.41KN at 2.5mm and 20KN at 5.0mm.
The maximum dry density and optimum moisture content obtained from the Proctor test above
was then used to compact the CBR samples. The CBR test was conducted after 48hours soaking
in water. The test involved allowing a standard plunger (as described above) to penetrate the
compacted sample at controlled rate while the load required for various levels of penetration was
recorded. The results of the CBR tests at different kiln dust percentages are presented and
discussed in chapter Four.
3.3.7 UNCONFINED COMPRESSION STRENGHT (UCS) TEST
This test was carried out in accordance with BS1377; part7: 1990; section 7.2 and section
4.1 of BS 1924 part2: 1990 with some modifications. Compaction was done at the maximum dry
density of the optimum moisture content of the respective mixtures. Samples were then extruded
with extruder. The dimensions of the specimen were 76mm long and 38mm in diameter. The
specimens were sealed in transparent cellophane bags after which they were kept in desiccators
to prevent loss of water by evaporation. The samples were cured for respective periods of 7, 14
and 28 days.
64
The unconfined compressive strength was determined by testing the cylindrical
specimens using standard load frame. The cylindrical specimens were subjected to a steadily
increasing axial load until failure occurs. The results of the unconfined compressive test (UCS)
at different percentages of kiln dust are presented and discussed in Chapter Four.
65
CHAPTER FOUR
4.0 RESULTS AND DISCUSSION
4.1 RESULTS OF TESTS ON UNTREATED SOIL SAMPLE
Results of the index properties of the untreated soil are presented in Table4.1 below, while the
effect of cement kiln dust are presented in the Figures 4.1 to 4.8.
Table4.1 Index Properties of untreated soil sample
DESCRIPTION RESULTS
Liquid Limit (%) 42
Plastic Limit (%) 27.3
Plasticity Index (%) 14.7
Percentage Passing BS Sieve 200 (%) 2.10
AASHTO Classification A-2-7(1)
USCS Classification SC
Maximum Dry Density (kg/m3) 1830
Optimum Moisture Content (%) 12.18
Unconfined Compressive Strength (KN/m2) 395
California Bearing Ratio (after 24 hrs soaking) 2.2
Specific Gravity 2.54
Colour Reddish Brown
66
4.2 GRADING CURVE RESULTS (SIEVE ANALYSIS)
The sieve analysis result of the untreated soil is shown in the figure 4.1 below
Fig 4.1 Particle Size distribution
4.3 CHEMICAL COMPOSITION OF THE CEMENT KILN DUST
The oxide composition of the cement kiln dust is shown in Table 4.2. The composition of
calcium oxide (CaO) alone is 55.06%. This is responsible for the ion exchange between soil and
the CKD resulting in the formation of more granular material and strength development. The lost
in ignition (LOI) for the cement kiln dust was 4.38 while the specific gravity was 2.60. The lost
in ignition is the loss in mass associated with heating to ~950oC, usually contributed by
chemically bound water and noncarbonated carbon. Soil treated with low LOI and high free lime
CKD provided promising results in terms of strength McCoy and Kriner (1971)
Table 4.2: Oxide composition of Cement kiln dust
Oxides SiO2 Al2O3 Fe2O3 CaO MgO SO3 K2O Na2O F.CaO LOI
Percentage (%)
14.79 4.51 2.64 55.06 2.66 1.48 1.26 0.18 4.04 4.38
0
10
20
30
40
50
60
70
80
90
0.001 0.01 0.1 1
pe
rce
nta
ge p
assi
ng
(%)
Particle size(mm)
67
4.3 STABILIZATION TEST RESULT
4.3.1 Atterberg Limit Test
Table 4.3 Tabulated Atterberg limit test result
CKD (%) LIQUID PLASTIC PLASTICITY
LIMIT (%) LIMIT (%) INDEX(%)
0 42 27.3 14.7
2 40 26 14
4 38.2 24.67 13.5
6 37.4 24 13.4
8 37.82 24.5 13.32
10 36.7 23.7 13
12 35.7 22.9 12.8
14 33 20.6 12.4
16 32 20 12
18 34 22 12
20 36 24.33 11.67
22 36.7 25.3 11.2
24 36.2 25.3 10.9
28 37 26 11
32 30.3 20.2 10.1
36 29 19 9.2
68
Fig 4.2 Variation of liquid Limit with Cement Kiln Dust
Fig 4.3 Variation of Plastic Limit with Cement Kiln Dust
20
25
30
35
40
45
0 5 10 15 20 25 30
liqu
id li
mit
(%
)
CKD(%)
10
15
20
25
30
0 5 10 15 20 25 30
pla
stic
lim
it (
%)
CKD (%)
69
Fig 4.4 Variation of plasticity index with Cement Kiln Dust
4.3.1.1 Effect of cement kiln dust (CKD) on Consistency ( Atterberg Limit)
The variation of liquid limit with CKD is shown in fig 4.2. The liquid limit decreased with the
addition of CKD up to 12% and then increased slightly with further increase in the addition of
CKD. This initial reduction in the liquid limit according to (Talal and Awad 1998) is attributed
to cementitious properties of CKD due to high content of calcium oxide (CaO) which aid
flocculation and aggregation of soil particles. The increased in liquid limit beyond 12% CKD
content is attributed to the extra water required to turn the soil-CDK mix to fluid.
Figure 4.3 shows the relationship between plastic limits and CKD content. The trend is similar to
that of the liquid limit. The reasons for the variation of the liquid limit with CKD content are also
similar to that of the variation of plastic limit with CKD content.
Plasticity Index and CKD content relationship is shown in figure 4.4. The plasticity index
decreased from 14.7% to 11.2% with increase in CKD content from 0% to 24%. This trend may
6
8
10
12
14
16
0 5 10 15 20 25 30
Pla
stic
ity
ind
ex
(%)
CKD(%)
70
be attributed to the replacement of the finer soil particles by CKD with consequent reduction in
plasticity index (Okafor, et al 2009). This trend is also observed by several researchers, ( Zaman
1992, Miller et al;1997).
4.3.2 COMPACTION TEST RESULT
The results of effect of various proportion of cement kiln dust on compaction characteristic of the soil are
shown in Table 4.4
Table 4.4 Compaction Test Result
S/N CKD (%) Maximum Dry
Density (MDD)Kg/
M3
Optimum
Moisture Content
(OMC) [ %]
1 0 1.88 12.18
2 2 1.86 12.34
3 4 1.79 12.49
4 5 1.80 12.52
5 6 1.82 12.61
6 8 1.75 12.72
7 9 1.72 12.87
8 10 1.70 13.00
9 12 1.68 13.27
10 14 1.60 13.61
11 15 1.58 13.84
12 16 1.49 13.95
13 18 1.44 14.30
14 20 1.42 14.64
15 22 1.40 15.27
16 24 1.40 15.89
71
Fig 4.5 Variation MDD with cement kiln dust
Fig 4.6 Variation OMC with cement kiln dust
4.3.2.1 Effect of cement kiln dust on the compaction characteristics of the soil
The relationship between MDD and CKD content is shown in figure 4.5. The results indicate that
between 0% and 12% CKD content, the MDD increased from 1.83Mg/m3 to 2.60Mg/m
3
respectively. This is as a result of CKD occupying the void within the soil matrix. It may also be
due to the relatively higher specific gravity of CKD (2.60) to that of the soil (2.54).( Ola , 1975;
500
1000
1500
2000
2500
3000
3500
0 5 10 15 20 25 30
MD
D K
g/m
3
CKD (%)
4
6
8
10
12
14
16
18
20
0 5 10 15 20 25 30
OM
C
CKD (%)
72
and Oriola and Moses, 2010). While the decreased in MDD after 12% optimum CKD content
could be attributed to the brittle nature of soil with higher CKD and this is one of the drawbacks
identified by Maine (2006) in the use of CKD for soil stabilization.
Figure 4.6 shows the variation of OMC with CKD content. The result shows that the OMC
increased with increase in CKD content. This result agrees with previous research by Zaman et al
(1992) and Miller and Zaman (2000). The increase in OMC is due to the additional water needed
to coat the surface area and to lubricate the entire matrix for hydration process (Eze-Uzomaka et
al, 2010).
4.3.3 CALIFORNIA BEARING RATIO (CBR) TEST RESULT
The California bearing ratio result of the soil treated with different percentages of cement kiln dust is
shown in table 4.5 below.
Table 4.5 Result of CKD on CBR of treated soil sample
CKD (%) CBR soaked CBR unsoaked
0 2.2 22
2 3.3 30
4 4.3 44
6 6 47
8 7.1 52
10 9 55
12 10.5 57
14 13 59
16 16.6 60
18 20 62.5
20 22.5 64
22 25 78
24 28 80
73
Fig 4.7 variation of CBR of soil with cement kiln dust for both soak and unsoak sample
4.3.3.1 Effect of cement kiln dust on California Bearing Ratio of soil
Figure 4.7 shows the relationship between the CBR and CKD content at optimum moisture
content (OMC). The result shows that the CBR for unsoaked sample increase from 22% to
80% with increase in CKD content from 0% to 24 % respectively. The reason for the
increment in CBR may be because of the gradual formation of cementitious compound in the
soil by reaction between the CKD and some amount of CaOH present in soil.
The trend of the soaked CBR was similar to the Unsoaked though with lower values.
The CBR is a measure of the strength of the subgrade
0
10
20
30
40
50
60
70
80
90
0 5 10 15 20 25 30
C B
R V
alu
es
(%)
CKD (%)
CBR soak
CBR unsoak
74
4.3.4 UNCONFINED COMPRESSIVE STRENGTH TEST RESULT
The unconfined compressive strength result of soil sample treated with cement kiln dust is
shown in table 4.6 below.
Table 4.6 Results of 7, 14 and 28 days unconfined compressive strength with CKD
CKD(%) 7 days UCS (MN/m2 ) 14 days UCS (MN/m
2 ) 28 days UCS (MN/m
2 )
0 395 395 395
2 460 495 498
4 525 594 600
6 568 658 674
8 611 721 747
10 724 830 860
12 838 938 973
14 973 1064 1088
16 1107 1189 1204
18 1240 1320 1346
20 1373 1452 1487
22 1366 1454 1489
24 1360 1454 1490
75
Fig 4.8 variation of unconfined compressive strength of soil with cement kiln dust
4.3.4.1 Effect of cement kiln dust on Unconfined Compressive Strength soil sample
Unconfined compressive strength (UCS) is the most common and adaptive method of
evaluating the strength of stabilized soil. It is the main test recommended for the
determination of the amount of additive required to be used in stabilization of soil (Singh and
Singh 1991). Results of variation of UCS with increase in CKD for 7days, 14days and 28days
are shown in fig 4.8 above. The USC values increased as the percentage of CKD increases
from 0% to 24%. Beyond this point no significant increased in UCS is observed with increase
in CKD content. The increase in the UCS is attributed to the formation of cementitious
compound between CaOH present in the soil and the CKD.
0
200
400
600
800
1000
1200
1400
1600
0 5 10 15 20 25 30
UC
S (
MN
/m2 )
CKD ( %)
7 days UCS
14 days UCS
28 days UCS
76
4.4 RESPONSE SURFACE METHODOLOGY OF CKD STABILIZED SUBBASE
AND MODEL VERIFICATION
The Response Surface methodology is the development of an empirical model of the
process, the mixture matrix and to obtain a more precise estimate of the optimum operating
conditions for the controllable and uncontrollable variables. The controllable variables are
prepared soil samples and CKD, while the uncontrollable variables are Optimum Moisture
Content OMC, Maximum Dry Density MDD, California Bearing Ratio CBR and Unconfined
Compressive Strength (UCS)
As discussed in Section 2.6.4, the method of least square chooses the coefficients 𝜇𝑖
and 𝜇𝑖𝑗 in Equation 2.15 so that the sum of the squares of the errors between the experimental
expected response (𝐹𝑓𝑖 and 𝐹𝑓𝑖𝑗 ) and predicted expected response (𝐹 𝑓𝑖 and 𝐹 𝑓𝑖𝑗 ) are
minimized. The function L is to be minimized with respect to 𝜇𝑖(𝑖 = 1 − 2) and 𝜇𝑖𝑗 (1 ≤ 𝑖 ≤
𝑗).
Satisfying this condition by using equations 2.16 and 2.17, the least square method
results to a system of linear equations with three unknown constants 𝜇1, 𝜇2 𝑎𝑛𝑑 𝜇12 to be
determined for each of the CKD-soil matrix for the various uncontrolled variables.
77
Table 4. 7a TEST RESULTS AFTER STABILIZATION WITH CKD
Table 4.7a Optimum Moisture Content (OMC)
S/N % CKD Sample (%) OMC (%)
1 0 100 12.18
2 2 98 12.34
3 4 96 12.49
4 5 95 12.55
5 6 94 12.61
6 8 92 12.72
7 9 91 12.86
8 10 90 13.00
9 12 88 13.27
10 14 86 13.61
11 15 85 13.78
12 16 84 13.95
13 18 82 14.30
14 20 80 14.64
15 22 78 15.27
16 24 76 15.89
Table 4.7b Maximum dry density (MDD)
S/N % CKD Sample (%) MDD (kg/m3)
1 0 100 1.83
2 2 98 1.96
3 4 96 1.79
4 5 95 1.80
5 6 94 1.82
6 8 92 1.75
7 9 91 1.72
8 10 90 1.70
9 12 88 1.68
10 14 86 1.60
11 15 85 1.58
12 16 84 1.49
13 18 82 1.44
14 20 80 1.42
15 22 78 1.40
16 24 76 1.40
78
Table 4.7c CALIFORNIA BEARING RATIO
Table 4.7d UNCONFINED COMPRESSIVE STRENGTH
S/N
CKD (%)
Sample
(%) CKD (%) CBR soaked CBR un-soaked
1 0 100 0 2.2 22
2 2 98 2 3.3 30
3 4 96 4 4.3 44
4 6 94 6 6 47
5 8 92 8 7.1 52
6 10 90 10 9 55
7 12 88 12 10.5 57
8 14 86 14 13 59
9 16 84 16 16.6 60
10 18 82 18 20 62.5
11 20 80 20 22.5 64
12 22 78 22 25 78
13 24 76 24 28 80
S/N CKD( %) Sample (%) Unconfined compressive
Strength ( MN/m2)
7 days 14 days 28 days
1 0 100 395 395 395
2 2 98 460 495 498
3 4 96 525 594 600
4 6 94 568 658 674
5 8 92 611 721 747
6 10 90 724 830 860
7 12
88 838 938 973
8 14
86 973 1064 1088
9 16
84 1107 1189 1204
10 18
82 1240 1320 1346
11 20
80 1373 1452 1487
12 22
78 1366 1454 1489
13 24
76 1360 1454 1490
79
Table 4.8 SUMMARY OF STABILIZATION TEST RESULT
S/N CKD (%)
SOIL
OMC (%)
MDD MG/M3
CALIFORNIA BEARING RATIO
UNCONFINED COMPRESSIVE STRENGTH(MN/m3)
Soaked
(%)
Unsoaked
( %) 7 days 14 days 28 days
1 0 100 12.18 1.83 2.2 22 395 395 395
2 2 98 12.34 1.96 3.3 30 460 495 498
3 4 96 12.49 2.10 4.3 44 525 594 600
4 5 95 12.20 2.13 5.2 46 547 626 637
5 6 94 12.61 2.17 6 47 568 658 674
6 8 92 12.72 2.24 7.1 52 611 721 747
7 9 91 12.86 2.42 8.1 54 668 776 804
8 10 90 13.00 2.58 9 55 724 830 860
9 12 88 13.27 2.60 10.5 57 838 938 973
10 14 86 13.61 2.54 13 59 973 1064 1088
11 15 85 13.78 2.33 14.8 59.5 1040 1127 1146
12 16 84 13.95 2.13 16.6 60 1107 1189 1204
13 18 82 14.30 2.04 20 62.5 1240 1320 1346
14 20 80 14.64 1.96 22.5 64 1373 1452 1487
15 22 78 15.27 1.95 25 78 1366 1454 1489
16 24 76 15.89 1.94 28 80 1360 1454 1490
Thus, the least square method results to a system of linear equations with three unknowns
constants 𝜇1, 𝜇2 𝑎𝑛𝑑 𝜇12 to be determined for each of the response.
80
4.4.1 OPTIMUM MOISTURE CONTENT
Table 4.9: Selected Controllable Variables for OPTIMUM MOISTURE CONTENT
S/NO X1 X2 X1 X2 X12
X22
X12 X2 X1 X2
2 X1
2X2
2
1. 0.00 100.00 0.00 0.00 10000.00 0.00 0.00 0.00 2. 2.00 98.00 196.00 4.00 9604.00 392.00 19208.00 38416.00 3. 4.00 96.00 384.00 16.00 9216.00 1536.00 36864.00 147456.00 4. 6.00 94.00 564.00 36.00 8836.00 3384.00 53016.00 318096.00 5. 10.00 90.00 900.00 100.00 8100.00 9000.00 81000.00 810000.00 6. 12.00 88.00 1056.00 144.00 7744.00 12672.00 92928.00 1115136.00 7. 14.00 86.00 1204.00 196.00 7396.00 16856.00 103544.00 1449616.00 8. 18.00 82.00 1476.00 324.00 6724.00 26568.00 121032.00 2178576.00 9. 20.00 80.00 1600.00 400.00 6400.00 32000.00 128000.00 2560000.00 10. 24.00 76.00 1824.00 576.00 5776.00 43776.00 138624.00 3326976.00 ∑ 110.00 890.00 9204.00 1796.00 79796.00 146184.00 774216.00 11944272.00
The homogenous set of equations resulting from Equations 2.16 and 2.17 are as follows:
𝜇1 𝑥12 + 𝜇2 𝑥1 𝑥2 + 𝜇12 𝑥1
2 𝑥2 = 𝐹𝑖 𝑥1 4.1
𝜇1 𝑥1 𝑥2 + 𝜇2 𝑥22 + 𝜇12 𝑥1𝑥2
2 = 𝐹𝑖 𝑥2 4.2
𝜇1 𝑥12 𝑥2 + 𝜇2 𝑥1𝑥2
2 + 𝜇12 𝑥12 𝑥2
2 = 𝐹𝑖 𝑥1𝑥2 4.3
S/NO ∑Fi X1 ∑FiX2 ∑FiX1X2
1. 0.00 1218.00 0.00 2. 24.68 1209.32 2418.64 3. 49.96 1199.04 4796.16 4. 75.66 1185.34 7112.04 5. 130.00 1170.00 11700.00 6. 159.24 1167.76 14013.12 7. 190.54 1170.46 16386.44 8. 257.40 1172.60 21106.80 9. 292.80 1171.20 23424.00 10. 381.36 1207.64 28983.36
∑ 1561.64 11871.36 129940.56
81
Substituting the numerical values of the terms as given in Table 4.9 in homogenous
Equations (4.1-4.3), we have:
1796.00𝜇1 + 9204.00𝜇2 + 146184.00𝜇12 = 1561.64
⇒ 9204.0𝜇1 + 79796.00𝜇2 + 774216.00𝜇12 = 45188.00
146184.00𝜇1 + 774216.00𝜇2 + 11944272.00𝜇12 = 571320.00
⇒
𝜇1
𝜇2
𝜇12
= 1796.00 9204.00 146184.00 9204.0 79796.00 774216.00
146184.00 774216.00 11944272.00
−𝟏
1561.64
45188.00571320.00
= 0.67000.12260.0053
⇒ 𝐹𝑓 = 𝜇1𝑥1 + 𝜇2𝑥2 + 𝜇12𝑥1𝑥2
Where Ff = OMC,
x1 = Percentage proportion of CKD and x2 = percentagae proportion of sample
⇒ OMC = 0.670 x1 + 0.1226 x2 + 0.00527 x1x2 4.4
4.4.2 MAXIMUM DRY DENSITY
Table 4.10 Selected Controllable Variables for MAXIMUM DRY DENSITY
S/NO X1 X2 X1 X2 X12
X22
X12 X2 X1 X2
2 X1
2X2
2
1. 0.00 100.00 0.00 0.00 10000.00 0.00 0.00 0.00 2. 2.00 98.00 196.00 4.00 9604.00 392.00 19208.00 38416.00 3. 4.00 96.00 384.00 16.00 9216.00 1536.00 36864.00 147456.00 4. 6.00 94.00 564.00 36.00 8836.00 3384.00 53016.00 318096.00 5. 10.00 90.00 900.00 100.00 8100.00 9000.00 81000.00 810000.00 6. 12.00 88.00 1056.00 144.00 7744.00 12672.00 92928.00 1115136.00 7. 14.00 86.00 1204.00 196.00 7396.00 16856.00 103544.00 1449616.00 8. 18.00 82.00 1476.00 324.00 6724.00 26568.00 121032.00 2178576.00 9. 20.00 80.00 1600.00 400.00 6400.00 32000.00 128000.00 2560000.00 10. 24.00 76.00 1824.00 576.00 5776.00 43776.00 138624.00 3326976.00 ∑ 110.00 890.00 9204.00 1796.00 79796.00 146184.00 774216.00 11944272.00
82
Substituting the numerical values of the terms as given in Table 4.10 in homogenous
Equations (4.1- 4.3), we have:
1796.00𝜇1 + 9204.00𝜇2 + 146184.00𝜇12 = 240.14
⇒ 9204.0𝜇1 + 79796.00𝜇2 + 774216.00𝜇12 = 1930.86
146184.00𝜇1 + 774216.00𝜇2 + 11944272.00𝜇12 = 20207.56
⇒
𝜇1
𝜇2
𝜇12
= 1796.00 9204.00 146184.00 9204.0 79796.00 774216.00
146184.00 774216.00 11944272.00
−𝟏
240.14
1930.8620207.56
= −0.33790.01790
4.669 × 10−3
⇒ 𝐹𝑓 = 𝜇1𝑥1 + 𝜇2𝑥2 + 𝜇12𝑥1𝑥2
Where Ff = MDD,
x1 = Percentage proportion of CKD and x2 = percentagae proportion of soil sample
⇒ MDD = −0.3379 x1 + 0.0179 x2 + 4.669 × 10−4 x1x2 4.5
S/NO ∑Fi X1 ∑FiX2 ∑FiX1X2
1. 0.00 183.00 0.00 2. 3.92 192.08 384.16 3. 8.40 201.60 806.40 4. 13.02 203.98 1223.88 5. 25.80 232.20 2322.00 6. 31.20 228.80 2745.60 7. 35.56 218.44 3058.16 8. 36.72 167.28 3011.04 9. 39.20 156.80 3136.00 10. 46.32 146.68 3520.32
∑ 240.14 1930.86 20207.56
83
4.4.1 CALIFORNIA BEARING RATIO ( Un-soaked)
Table 4.11: Selected Controllable Variables for CALIFORNIA BEARING RATIO
(Unsoaked)
S/NO X1 X2 X1 X2 X12
X22
X12 X2 X1 X2
2 X1
2X2
2
1 0.00 100.00 0.00 0.00 10000.00 0.00 0.00 0.00 2 2.00 98.00 196.00 4.00 9604.00 392.00 19208.00 38416.00 3 4.00 96.00 384.00 16.00 9216.00 1536.00 36864.00 147456.00 4 6.00 94.00 564.00 36.00 8836.00 3384.00 53016.00 318096.00 5 10.00 90.00 900.00 100.00 8100.00 9000.00 81000.00 810000.00 6 12.00 88.00 1056.00 144.00 7744.00 12672.00 92928.00 1115136.00 7 14.00 86.00 1204.00 196.00 7396.00 16856.00 103544.00 1449616.00 8 18.00 82.00 1476.00 324.00 6724.00 26568.00 121032.00 2178576.00 9 20.00 80.00 1600.00 400.00 6400.00 32000.00 128000.00 2560000.00 10 24.00 76.00 1824.00 576.00 5776.00 43776.00 138624.00 3326976.00 ∑ 110.00 890.00 9204.00 1796.00 79796.00 146184.00 774216.00 11944272.00
Substituting the numerical values of the terms as given in Table 4.11 in homogenous
Equations (4.1 – 4.3), we have:
1796.00𝜇1 + 9204.00𝜇2 + 146184.00𝜇12 = 6912.00
⇒ 9204.0𝜇1 + 79796.00𝜇2 + 774216.00𝜇12 = 45188.00
146184.00𝜇1 + 774216.00𝜇2 + 11944272.00𝜇12 = 571320.00
⇒
𝜇1
𝜇2
𝜇12
= 1796.00 9204.00 146184.00 9204.0 79796.00 774216.00
146184.00 774216.00 11944272.00
−𝟏
6912.00
45188.00571320.00
= −1.25700.2639
4.60 × 10−2
S/NO ∑Fi X1 ∑FiX2 ∑FiX1X2
1 0.00 2200.00 0.00 2 60.00 2940.00 5880.00 3 176.00 4224.00 16896.00 4 282.00 4418.00 26508.00 5 550.00 4950.00 49500.00 6 684.00 5016.00 60192.00 7 826.00 5074.00 71036.00 8 1134.00 5166.50 92988.00 9 1280.00 5120.00 102400.00 10 1920.00 6080.00 145920.00
∑ 6912.00 45188.00 571320.00
84
⇒ 𝐹𝑓 = 𝜇1𝑥1 + 𝜇2𝑥2 + 𝜇12𝑥1𝑥2
Where Ff = CBR unsoaked,
x1 = Percentage proportion of CKD and x2 = percentagae proportion of soil sample
⇒ CBR = −1.257𝑥1 + 0.2639 x2 + 4.60 × 10−2 x1x2 4.6
4.4.1 CALIFORNIA BEARING RATIO (soaked)
Table 4.12: Selected Controllable Variables for CALIFORNIA BEARING RATIO
(soaked)
S/NO X1 X2 X1 X2 X12
X22
X12 X2 X1 X2
2 X1
2X2
2
1 0.00 100.00 0.00 0.00 10000.00 0.00 0.00 0.00 2 2.00 98.00 196.00 4.00 9604.00 392.00 19208.00 38416.00 3 4.00 96.00 384.00 16.00 9216.00 1536.00 36864.00 147456.00 4 6.00 94.00 564.00 36.00 8836.00 3384.00 53016.00 318096.00 5 10.00 90.00 900.00 100.00 8100.00 9000.00 81000.00 810000.00 6 12.00 88.00 1056.00 144.00 7744.00 12672.00 92928.00 1115136.00 7 14.00 86.00 1204.00 196.00 7396.00 16856.00 103544.00 1449616.00 8 18.00 82.00 1476.00 324.00 6724.00 26568.00 121032.00 2178576.00 9 20.00 80.00 1600.00 400.00 6400.00 32000.00 128000.00 2560000.00 10 24.00 76.00 1824.00 576.00 5776.00 43776.00 138624.00 3326976.00 ∑ 110.00 890.00 9204.00 1796.00 79796.00 146184.00 774216.00 11944272.00
S/NO ∑Fi X1 ∑FiX2 ∑FiX1X2
1 0.00 2200.00 0.00 2 6.60 323.40 646.80 3 17.20 412.80 1651.20 4 36.00 564.00 3384.00 5 90.00 810.00 8100.00 6 126.00 924.00 11088.00 7 182.00 1118.00 15652.00 8 360.00 1640.00 29520.00 9 450.00 1800.00 36000.00 10 672.00 2128.00 51072.00
∑ 1939.80 9940.20 157114.00
85
Substituting the numerical values of the terms as given in Table 4.12 in homogenous
Equations(4.1 -4.3), we have:
1796.00𝜇1 + 9204.00𝜇2 + 146184.00𝜇12 = 1939.80
⇒ 9204.0𝜇1 + 79796.00𝜇2 + 774216.00𝜇12 = 9940.20
146184.00𝜇1 + 774216.00𝜇2 + 11944272.00𝜇12 = 157114.00
⇒
𝜇1
𝜇2
𝜇12
= 1796.00 9204.00 146184.00 9204.0 79796.00 774216.00
146184.00 774216.00 11944272.00
−𝟏
1939.80
9940.20157114.00
= 3.340
0.022242.90 × 10−2
⇒ 𝐹𝑓 = 𝜇1𝑥1 + 𝜇2𝑥2 + 𝜇12𝑥1𝑥2
Where Ff = CBR soaked,
x1 = Percentage proportion of CKD and x2 = percentagae proportion of soil sample
⇒ CBR = 3.34𝑥1 + 0.0224 x2 + 2.9 × 10−2 x1x2 4.7
4.4.1 UNCONFINED COMPRESSIVE STRENGTH (7DAYS)
Table 4.13: Selected Controllable Variables for UNCONFINED COMPRESSIVE
STRENGTH (7DAYS)
S/NO X1 X2 X1 X2 X12
X22
X12 X2 X1 X2
2 X1
2X2
2
1 0.00 100.00 0.00 0.00 10000.00 0.00 0.00 0.00 2 2.00 98.00 196.00 4.00 9604.00 392.00 19208.00 38416.00 3 4.00 96.00 384.00 16.00 9216.00 1536.00 36864.00 147456.00 4 6.00 94.00 564.00 36.00 8836.00 3384.00 53016.00 318096.00 5 10.00 90.00 900.00 100.00 8100.00 9000.00 81000.00 810000.00 6 12.00 88.00 1056.00 144.00 7744.00 12672.00 92928.00 1115136.00 7 14.00 86.00 1204.00 196.00 7396.00 16856.00 103544.00 1449616.00 8 18.00 82.00 1476.00 324.00 6724.00 26568.00 121032.00 2178576.00 9 20.00 80.00 1600.00 400.00 6400.00 32000.00 128000.00 2560000.00 10 24.00 76.00 1824.00 576.00 5776.00 43776.00 138624.00 3326976.00 ∑ 110.00 890.00 9204.00 1796.00 79796.00 146184.00 774216.00 11944272.00
86
Substituting the numerical values of the terms as given in Table 4.13 in homogenous
Equations (4.1- 4.3), we have:
1796.00𝜇1 + 9204.00𝜇2 + 146184.00𝜇12 = 119766.00
⇒ 9204.0𝜇1 + 79796.00𝜇2 + 774216.00𝜇12 = 725834.00
146184.00𝜇1 + 774216.00𝜇2 + 11944272.00𝜇12 = 9827812.00
⇒
𝜇1
𝜇2
𝜇12
= 1796.00 9204.00 146184.00 9204.0 79796.00 774216.00
146184.00 774216.00 11944272.00
−𝟏
119766.00
725834.009827812.00
= 68.1903.6200
−0.2467
⇒ 𝐹𝑓 = 𝜇1𝑥1 + 𝜇2𝑥2 + 𝜇12𝑥1𝑥2
Where Ff =𝑈𝐶𝑆,
x1 = Percentage proportion of CKD and x2 = percentage proportion of sample
⇒ UCS(7days) = 68.19 x1 + 3.62 x2 + 0.2467 x1x2 4.8
S/NO ∑Fi X1 ∑FiX2 ∑FiX1X2
1 0.00 39500.00 0.00 2 920.00 45080.00 90160.00 3 2100.00 50400.00 201600.00 4 3408.00 53392.00 320352.00 5 7240.00 65160.00 651600.00 6 10056.00 73744.00 884928.00 7 13622.00 83678.00 1171492.00 8 22320.00 101680.00 1830240.00 9 27460.00 109840.00 2196800.00 10 32640.00 103360.00 2480640.00
∑ 119766.00 725834.00 9827812.00
87
4.5 MODEL VERIFICATION
The models in Equation 4.4 through Equation 4.8 will be verified using the selected
optimum results of the stabilized soils given in Table 4.8 This test results are reproduced as in
Table 4.14
Table 4.14: OPTIMUM RESULTS OF THE STABILIZED SOILS FOR
DISCUSSIONS
S/NO
CKD
(%)
SAMPLE
(%)
OMC
(%)
MDD
Mg/m3
CBR
unsoaked
(%)
CBR
soaked
(%)
7 days
UCS
(MN/m2)
1. 5.00 95.00 12.55 2.10 47 5.2 547
2. 8.00 92.00 12.72 2.24 52 7.1 611
3. 9.00 91.00 12.86 2.42 54 8.1 668
4. 15.00 85.00 13.78 2.33 59.5 14.8 1049
5. 16.00 84.00 13.95 2.13 60 16.6 1070
6. 22.00 78.00 15.27 1.93 78 25 1366
88
OPTIMUM MOISTURE CONTENT
Using equation 4.4, the following values were obtained:
Table 4.15: Experimented and Model values for Optimum Moisture Content
S/NO
CKD
(%)
SAMPLE (%)
OMC (%)
Experiment Model
1 5.00 95.00 12.55 12.75
2 8.00 92.00 12.72 12.76
3 9.00 91.00 12.86 12.87
4 15.00 85.00 13.78 13.76
5 16.00 84.00 13.95 13.95
6 22.00 78.00 15.27 15.27
The correlation coefficient, r = 0.9977
Figure 4.9 Plot of Optimum moisture content against percentage CKD
8
9
10
11
12
13
14
15
16
5 10 15 20 25
OM
C
% CKD
Experiment
Model
89
MAXIMUM DRY DENSITY
Using equation 4.5, the following values were obtained:
Table 4.16: Experimented and Model values for Maximum Dry Density
S/NO
CKD
(%)
SAMPLE (%)
MDD Mg/m3
Experiment Model
1 5.00 95.00 2.13 2.23
2 8.00 92.00 2.24 2.38
3 9.00 91.00 2.42 2.41
4 15.00 85.00 2.33 2.40
5 16.00 84.00 2.13 2.37
6 22.00 78.00 1.93 1.97
The correlation coefficient r =0.8728
Figure 4.10 Plot of Maximum dry density against percentage CKD
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
5 10 15 20 25
MD
D M
g/m
3
CKD (%)
Experiment
Model
90
CALIFORNIA BEARING RATIO (Un-soaked)
Using equation 4.6, the following values were obtained:
Table 4.17: Experimented and Model values for California Bearing Ratio (Un-soaked)
S/NO
CKD
(%)
SAMPLE (%)
CBR (%)
Experiment Model
1 5.00 95.00 47 40.64
2 8.00 92.00 52 48.09
3 9.00 91.00 54 50.38
4 15.00 85.00 59.5 62.23
5 16.00 84.00 60 63.88
6 22.00 78.00 78 71.87
The correlation coefficient r =0.9247
Figure 4.11 Plot of California bearing ratio (un-soak) against percentage CKD
5
15
25
35
45
55
65
75
5 7 9 11 13 15 17
CB
R (
%)
CKD (%)
Experiment
Model
91
CALIFORNIA BEARING RATIO (soaked)
Using equation 4.7, the following values were obtained:
Table 4.18: Experimented and Model values for California bearing Ratio (Soaked)
S/NO
CKD
(%)
SAMPLE (%)
CBR (%)
soaked
Experiment Model
1 5.00 95.00 5.2 5.05
2 8.00 92.00 7.1 7.40
3 9.00 91.00 8.1 8.35
4 15.00 85.00 14.8 15.03
5 16.00 84.00 16.6 16.35
6 22.00 78.00 25 25.46
The correlation coefficient r =0.9994
Figure 4.12 Plot of California bearing ratio (soaked) against percentage CKD
0
5
10
15
20
25
30
5 10 15 20 25
CB
R(%
)
CKD (%)
Experiment
Model
92
UNCONFINED COMPRESSIVE STRENGTH (7days)
Using equation 4.8, the following values are obtained:
Table 4.19: Experimented and Model values for Unconfined Compressive Strength
S/NO
CKD
(%)
SAMPLE (%)
UCS KN/ m2
7days
Experiment Model
1 5.00 95.00 547 567
2 8.00 92.00 611 697
3 9.00 91.00 668 741
4 15.00 85.00 1049 1016
5 16.00 84.00 1070 1064
6 22.00 78.00 1366 1359
The correlation coefficient r =0.9937
Figure 4.13 Plot of Unconfined Compressive Strength (7 days) against percentage CKD
0
200
400
600
800
1000
1200
1400
1600
5 10 15 20 25
UC
S (M
N/m
2 )
CKD ( %)
Experiment
Model
93
5.0 CONCLUSIONS AND RECOMMENDATION
5.1 CONCLUSIONS
From the results and analysis carried out from this study, the following conclusion can be
drawn
1. The soil used for the study was obtained from Sankwala- Busi road in Obanliku,
Cross River State and classified as A-2-7 (1) base on AASHT0 (1986) classification
system, ( sandy gravel). In its natural state the soil is suitable as filled material.
2. The geotechnical properties of soil such as CBR, UCS were greatly improved by the
addition of CKD to the soil. 24% of CKD addition was observed to yield maximum
improvement. The soil after improvement could be used as sub base and base course
material.
3. The hydration of CKD after addition of water releases at the primary stabilization
stage, calcium silicate hydrate (C-S-H), monosulphoaluminates (C4ASH12) and
calcium hydroxides (CaOH).
4. The calcium hydroxide released during the primary stabilization also reacts with the
silicate in the soil to form calcium silicate hydrate ( C-S-H) and calcium aluminate
hydrates (C-A-H). this pozzolanic reaction with gelatinous hydrate also contributes to
improvement in strength of the soil.
5. The addition of CKD brings about alteration in plasticity properties of the soil. CKD
was very effective in reducing the plasticity of the soil.
6. At specified CKD content (0-12%) treatment of the soil it showed an increased in
MDD from 1.083Mg/m3 to 2.6Mg/m
3 after which there was a slight reduction in
MDD. The OMC generally increased with increase in CKD content.
7. The CKD used has LOI and free lime content of 4.3 and 4.04% respectively. Both
content were low. The LOI of the CKD relates to its performance in terms of strength.
94
The low content of LOI and free lime content also contributed to the increased in
strength of the soil after stabilization.
8. Models developed correspond well with experimental results. The model can be
successfully be used to predict the soil-CKD properties in the absence of experimental
data for soil.
5.2 RECOMMENDATIONS
From this research work the following recommendation are made:
1. Cement kiln dust (CKD) can be used to improve the California bearing ratio (CBR) and
unconfined compressive strength (UCS) of soils. The improvement would make it suitable for
used as sub-base and base course material in highway construction, thereby reducing
construction cost.
2. Any widespread use of a particular CKD should be accompanied by careful evaluation of its
properties (e.g. its free lime content, LOI) and possibly also of its effectiveness in
combination with the specific soil(s) to be treated.
3. Treatment of other soil type in Nigeria for improvement with CKD should be investigated.
4. The addition of other additives like lime to CKD for soil improvement should be
investigated. This may reduce the percentage of CKD required.
95
LIST OF REFERENCE
Abeln, D.L., Hastings, R.J., Scxhreiber, R.J., and Yonley, C. (1993): “Detailed
Illustration of Contingent Management Practices for Cement Kiln Dust,” Research
and Development Bulletin, SP115T, Portland cement Association, Skokie, IL.
Aigbedion, I., (2007): “Geophysical Investigation of Road failure using Electromagnetic
Profile along Opoji-Uwenlenbo and Illeh in Ekpoma- Nigeria” Middle –East Journal
of Scientific Research 2(3-4) pp 111-1115.
Alabadan, B.A., M.A. Olutoye, M.S. Abolarin and M. Zakariya (2005): “Partial Replacement
of Ordinary Portland Cement (OPC) with Bambara Groundnut Shell Ash (BGSA) in
Concrete”. Leonardo Electronic Journal of Practices and Technologies. Issue 6, pp.43-
48, January-June 2005. Vol. 04.08, American Society for Testing and Material,
Philadelphia.
Al-Harthy, A. S., Taha, R., and Al-Maamary, F., (2003) “Effect of Cement Kiln. Dust (CKD)
on Mortar and Concrete Mixtures,” Construction and Building Materials, Volume 17,
Number 5, July 2003 , pp. 353-360(8)
Al-Jabri, K., Taha,R., Al-Harthy, A., Al-Oraimi,S., and Al-Nuaimi,A.A. (2002). “Use of
Cement by-pass Dust in Flowable Fill Mixtures,” Cement, Concrete and Aggregates,
24 (2), pp. 53-57.
Al-Refeai, T.O and Al- Karni, A.A (1999) “Experimental Study on the Utilization of cement
kiln Dust for Ground Modification” J. king Saud University, Engineering Science
Vol. 2 No(2) pp 217-232
Arora, K. R. (2008) “Soil Mechanics and Foundation (Geotechnical) Engineering” Standard
Publisher Distributions, Delhi, India.
Baghdadi, Z.A., Fatani, N., and Sabban, N.A. (1995). “Soil Modification by Cement Kiln
Dust,” Journal of Materials in Civil Engineering, ASCE 7 (4), pp. 218-222.
Baghdadi, Z.A., (1990): “Engineering Study of Kiln Dust- Kaolinite Mixture” Proceedings of
the tenth Southeast Assian Geotechnical Conference, Taipei, Taiwan pp 17-21
Bhatty, J. I. (1995): “Alternative Uses of Cement Kiln Dust,” Research and Development
Bulletin, RP327, Portland cement Association, Skokie, IL, USA.
Bhatty, J.I., Bhattacharja, S., and Todres, H.A. (1996): “Use of Cement Kiln Dust in
Stabilizing Clay Soils”, Research & Development Bull, RP343, Portland Cement
Assoc.,Skokie, IL
BS 1377 (1990): “Methods of Testing Soils for Civil Engineering Purpose” British Standard
Institute, London
96
Bye, G.C. (1983): “Portland Cement-Composition, Production and Properties,” The
Institute of Ceramics, Pergamon Press, USA.
Collins, R, J., and Emery, J. J. (1983). “Kiln Dust-Fly Ash System for Highway Bases and
Subbases,” Federal Highway Administration Report FHWA/RD-82/167, U.S
Department of Transportation, Washington D.C.
Corish A. and Coleman T. (1995): Cement kiln dust. Concrete (London), 29(5), pp. 40-42.
Cornell, J. A., (1991): “Mixture Designs for Product Improvement Studies,” Communications
in Statistics - Theory and Methods, Volume 20, Issue 2 , pp 391 – 416.
Daugherty, E.D., and Funnell, J.E. (1983): The incorporation of low levels of by-product
in Portland cement and the effects of cement quality, cement, concrete and
aggregates. Vol. 5, No. 1. American Society for Testing and Materials,
Philadelphia, Penn. USA. pp. 14–20.
Davis, T.A., and Hooks, D.B., (1975): “ Disposal and utilization of Waste Kiln Dust
from Cement Industry” National Environmental Research Centre, Office Of
Research and Development, US Environmental Protection agency, Cincinnati,
Ohio, USA, EPA 670-2-75-043.
Diamond, S. and Kinter, EB. (1965). “Mechanism of Soil-Lime Stabilization- An
Interpretive Review,” Highway Research Record No.92.
El-Didamony H, Helmy IM, Amer AA. 1992. Utilization of cement dust in blended
cement. Pakistan Journal of Scientific and Industrial Research 35(7–8): 304-308.
El-Sayed, H.A., Gabr, N.A., Hanafi, S., and Mohran, M.A. 1991. Reutilization of by-pass
kiln dust in cement manufacture. In Proceedings of the international conference on
blended cement in construction. Sheffield, UK. September 1991.
Emery, J., Mackay, H., Umar, A., Vanderveer, G., and Pichette, J. 1992. Use of Wastes
and Byproducts as Pavement Construction Material. In Proceedings of the 45th
Canadian Geotechnical Conference, Toronto.
Eze-Uzomaka O.J and Agbo (2010) “ Suitability of quarry dust as improvement to cement
stabilized- laterite for road bases. Electronic Journal of Geotechnical Engineering
15(8) pp 1053-1066
Funderburk, R. F. (1990): “ Method and Mixture for Solidifying and immobilizing various
Waste Contaminants in an Organic Matrix” Patent and Trademark Office, USA
Department of Commerce, Washinton, DC, USA, Patent No4, 909,849.
Greer.L.W., and Matz., T.L. (1996). “Cement Kiln Dust Enforceable Agreement-Yes or
No?”., IEE-IAS.
Haynes, W.B., and Kramer, G.W. (1982): “Characterization of U.S.Cement Kiln Dust,”
Information Circular #8885, U.S Bureau of Mines, U.S. Department of the Interior,
Washington D.C.
97
Hilt, G.H., and Davidson, D.T. (1960): “Lime Fixation in Clayey Soils,” Highway
Research Bull. 262, pp.20-32.
Ingles, O. G. and Metcalf, J. B. (1972): Soil Stabilization Principles and Practice,
Butterworths, Sydney.
Kadiyali, L. R (2008) Traffic Engineering and Transportation Planning. Khanna Publishers,
New Delhi.
Kamon, M., and Nontananandh, S. (1991): “Combining Industrial Waste with lime for Soil
Stabilization,” Journal of Geotechnical Engineering, Vol.117, No.1, January, pp.1-17.
Katz,A.M., and Kovler,K. (2004): “Utilization of Industrial by-Products for the
Production of Controlled Low Strength Materials (CLSM),” Waste Managament
Vol.24, Issue 5, pp. 501-512.
Kessler, G.R. (1995): “Cement Kiln Dust (CKD) Methods for Reduction and Control,”IEEE
Transaction on Industry Applications 31(2), pp.407-412.
Kiefer, J. C., (1959): “Optimum Experimental Designs.” Journal of Royal Statistical Society,
Series B Vol. 21, pp. 272-304.
Kigel, M.Y., Shultis, J.F., Goldman, E. S., and Demytrk, M. K., ( 1994): “ Method of
Detoxification and Stabilization of Soils Containated with Chromium Ore Waste”
Patent and Trademark Office, US Department of Commerce, Washington DC, USA,
Patent No. 5,304,710.
Klemm, W.A. (1980): “Kiln Dust Utilization,” Martin Marietta Laboratories Report.
MML 80-12, Baltimore, MD.
Klemm, W.A., (1993): Cement kiln dust: a look at its uses and characteristics. In
Proceedings of the 29th International Cement Seminar, Rock Product, San
Francisco, California, USA.
Konsta-Gdoutos M.S., and Shah, S.P. (2003): “Hydration and Properties of Novel
Blended Cements Based on Cement Kiln Dust and Blast Furnace Slag,” Cement and
Concrete Research, 33 (8), pp.1269-1276.
Kumar,R., Kanaujia, V.K., and Ranjan, A. (2002): “An Experimental Study on Potential
Cement Kiln Dust in Stabilization of Fy Ash,” Cement, Concrete and Aggregates (24)
1, pp.25-27.
MacKay, M.and Emery, J. (1992): “Stabilization and Solidification of Contaminated
Soils and Sludges Using Cementitious System-Selected Case Histories,”
Transportation Research Record No. 1458, pp. 67-72.
Malhotra, V.M., and Ramezaniapour, A.A. (1994): “Fly Ash in Concrete,” MSL 94-
45(IR), Canada Center for Mineral and Energy Technology.
98
McCoy, W.J., and Kriner, R.W., (1971): “Use of Waste Kiln Dust for Soil
Consolidation,” Mill Session Papers, Portland Cement Association, Skokie, IL,
U.S.A.
Medubi, A. (1998): Stabilization of Black Cotton Soil using Superphosphate Fertilizer
Processing Residue as Admixture Unpublished M.Sc. Thesis Department of Civil
Engineering, Ahmadu Bello University, Zaria.
Mehta, P.K., and Monterio., (1993): “Concrete- Microstructure, Properties and
Materials” Prentice-Hall, Inc.
Miller, G.A., Zaman, M., Rahman, J., and Tan, K.N. (2003): “Laboratory and Field
Evaluation of Soil Stabilization Using Cement Kiln Dust,” Final Report, No. ORA
125-5693, Planning and Research Division, Oklahoma Department of Transportation.
Miller,G.A., and Azad, S. (2000): “Influence of Soil Type on Stabilization With Cement Kiln
Dust,” Journal of Construction and Building Materials 14 (2), 89-97.
Mitchell, J.K. (1993): “Fundamentals of Soil Behavior” 2nd Edition, John Wiley and Sons
Inc.
Montgomery, D. C; Designs and Analysis of Experiments: 6th
ed., John Wiley and Sons,
New York, 2005.
Moses, G. (2008): “stabilization of black cotton soil with ordinary portland cement Using
Bagasse ash as admixture” IRJI Journal of Research in Engrg. Vol.5 No.3 , pp. 107-
115
Muller, H.P. (1977): “What is Dust?: Characterization and Classification of Kiln Dust,” 24th
Technical Meeting, Report No. MA 77/2505/E, Holderbank Management and
Consulting Ltd., Technical Center, Material Division, Aargau, Switzerland.
Myers, R. H., D. C. Montgomery, G. C. Vining, C. M. Borror, and S. M. Kowalski; (2004):
“Response Surface Methodology: A Retrospective and Literature Survey,” Journal of
Quality Technology, Vol. 36, pp. 53477.
Myers, R. H., and D. C. Montgomery(1997): “A Tutorial on Generalized Linear Models,”
Journal of Quality Technology, Vol. 29, pp 274 – 291.
Nataraya, M.C and Nalanda, Y (2007): “ Performance of Industrial by- Product in controlled
low- strength material ( CLSM)” Waste Management, in Press, Corrected, 791
Nevilli,A.M., (1997): “Properties of Concrete,” Fourth Edition, John Wiley & Sons, Inc, New
York, USA.
Nicholson, J.P. (1983): “Cement Kiln Dust: Where it is Going?,” Pit and Quarry, July,pp. 98-
104.
99
Nigerian General Specification (1997): Bridges and Road Works. Federal Ministry of Works,
Lagos, Nigeria
Nisbet., M. (1997): “The 3Rs and Cement Kiln Dust: Opportunities for Reduction, Reuse and
Recycling,” For Presentation at the Air & Waste Management Association‟s 90th
Annual meeting & Exhibition, Toronto, Canada.
Obam, S. O., (2006): “The Accuracy of Scheffe‟s Third Degree Over Second Degree,
Optimization Regression Polynomials,” Nigerian Journal of Technology, Vol. 25, No
2, September, pp. 5 – 15.
Okafor, F.O and Okonkwo U. N., (2009): “Effects of Rice Husk Ash on Some Geotechnical
Properties of Lateritic Soil Leonardo Electronic Journal of Practices and Technologies
Issue 15, p. 67-74
Okechukwu, P. A., and Celestine O. O., (2011): “ Geotechnical Assessment of Road Failure
in Abakalike Area, Southeastern Nigeria”. International Journal of Civil and
Environmental Engineering, Vol. II No 2 pp 12-24
Ola, S.A. (1978): “The geology and geotechnical properties of the black cotton soils of north-
eastern Nigeria. In Tropical Soils of Nigeria in Engineering Practice. A. A. Balkema/
Rotterdam pg 131-144.
Olugbenga, O. A; Adetuberu A. A (2010): Characteristics of Bamboo leaf Ash stabilization
on lateritic soil in Highway construction. International Journal of Engineering and
Technology Vol 2(4) pp 212-219
Oriola, F and Moses, G (2010): “Groundnut shell Ash stabilization of Black cotton Soil”.
Electronic Journal of Geotechnical Engineering pg 415-428
Parson, R. L., and Kneebone E., (2004) “ Use of cement kiln dust for subgrade stabilization”,
Kansas Department of Transportation. Final Report No
Peethamparan, S., Olek, J., and Helfrich, K.E. (2006): “Evaluation of the Engineering
Properties of Cement Kiln Dust (CKD) Modified Kaolinite Clay,” the 21st
International Conference on Solid Waste Technology Management, March 26-29,
Philadelphia, USA.
Pierce, C.E., Tripathi, H., and Brown, W.B. (2003): "Cement Kiln Dust in Controlled
Low-Strength Materials," ACI materials journal, vol.100, No.6 pp 455-462.
Portland Cement Association, (1992). “An Analysis of Selected Trace Metals in Cement and
Kiln Dust,” SP109, Skokie, IL.
Santagata, M.C., and Bobet, A. (2002): “ The Use of Cement Kiln Dust (CKD) for
Subgrade Stabilization/Modification,”. Joint Transportation Research Program
Report, SPR 2575. West Lafayette, IN.
100
Santagata, M.C. and Collepardi, M. (1998). “Selection of Cement-Based Grouts for Soil
Improvement,” Grouts and Grouting, GSP No. 80, ASCE, pp. 177-195.
Sayah, A.I. (1993): “Stabilization of Expansive Clay Using Cement Kiln Dust,” M.S.
Thesis, University of Oklahoma, Norman, OK, U.S.A.
Scheffe, H., (1963) “The Simplex-centroid Design for Experiments with Mixtures,” Journal
of Royal Statistical Society, Series B Vol. 25, pp. 235-251.
Scheffe, H., (1958) “Experiments with Mixtures,” Journal of Royal Statistical Society, Series
B Vol. 20, pp. 344-360.
Schreiber, R.J, and Riney, S.M. (1995): “Cement Kiln Dust: An Overview,” Report No.0
7803-2456-0/95, IEEE-IAS.
Scott, M. M., and Ferguson E. G., (2005): “ Stabilization of soil with self-cementing coal
Ashes” World of Coal Ash Conference. April 11-15, 2005, Lexington, Kentucky,
USA. Retrieve from http:// www.flyash.info on 22/3/2012
Shoaib, M.M., Balaha, M.M., and Abdel-Rehman A.G. (2000). “Influence of Cement
Kiln Dust Substitution on the Mechanical Properties of Concrete,” Cement and
Concrete Research 30 (3), pp. 371-377.
Singh G., Singh J., (1991): “Highway Engineering” Standard Publisher Distributions, Nai
Sarak, India, pp 608-610.
Sreekrishnavilasam, A., King, S., and Santagata, M.C. (2006). “Characterization of Fresh and
Landfilled Cement Kiln Dust for Reuse-in-Construction Applications,” Journal of
Engineering Geology (Article in press).
Sreekrishnavilasam, A., and Santagata, M.C. (2006). “The Development of Criteria for
utilization of cement kiln dust (CKD) in highway infrasructures ” Joint Transport
Research Program. Paper 272 retrieved from http//docs.lib.purdue.edu/jtrp/272.
Sreekrishnavilasam, A., Rahardja, S., Kmetz, R., and Santagata, M.C (2006): “Soil
Treatment Using Fresh and Landfilled CKD”, Journal of Building Construction and
Materials(Article in press)
Steuch, H.E. (1992). “Review of Dust Return System,” Emerging Technologies for Kiln Dust
Management, SP 113, Portland Cement Association, Stokie, Illinois, U.S.A.
Taylor, H.F.W. (1997). “Cement Chemistry,” 2nd Edition, Thomas Telford Publishing,
London, UK.
Todres, H.A., Mishulovich, J., and Ahmed, J. (1992): “Cement kiln Dust
Management:Permeability,” Research & Development Bulletin RD103T, Portland
Cement Assoc., Skokie, IL.
Udoeyo, F.F., and Hyee, A. (2002): “Strength of cement kiln dust concrete,” Journal of
Materials in Civil Engineering, ASCE 14 (6), pp.524-526.
101
USEPA, (1993). “Report to Congress on Cement Kiln Dust,” Office of Solid Wastes.
U.S. Environmental Protection Agency.
Wang, K., Gdoutos, M.S.K., and Shah, S.P. (2002): “Hydration, Rheology, and Strength of
Ordinary Portland Cement (OPC) - Cement Kiln Dust (CKD) - Slag Binders,” ACI
Materials Journal, 99 (2), 173-179.
Zaman,M., Laguros,J.G., Sayah.A. (1992): “Soil Stabilization Using Cement Kiln
Dust,”Proceedings 7th International Conference on Expansive Soils. Dallas, Texas,
pp. 347-351.
Zaman, M. and Solanki, P. (2010): “Laboratory Performance Evaluation of Subgrade Soils
Stabilized with Sulfate Bearing Cementitious Additives,” ASTM J. of Testing and
Evaluation, Vol. 38, No. 1, pp. 1-12.
102
Appendix A
CKD
(%) PARAMETER
TEST
NUMBER
LIQUID PLASTIC PLASTICITY
1 2 3
LIMIT
(%)
LIMIT
(%) INDEX(%)
0 NOB 12 23 30 42 27.3 14.7
MC 45 41.2 39
2 NOB 16 24 40 40 26 14
MC 43 40 36
4 NOB 20 25 50 38.2 24.67 13.5
MC 40 36 34
6 NOB 17 25 42 37.4 24 13.4
MC 41 38 33
8 NOB 13 25 30 37.82 24.5 13.32
MC 42 39 31
10 NOB 14 27 45 36.7 23.7 13
MC 40 31 33
12 NOB 15 29 51 35.7 22.9 12.8
MC 38 34 34
14 NOB 12 27 46 33 20.6 12.4
MC 38.2 33.8 28.3
16 NOB 10 20 40 32 20 12
MC 41 28 24
18 NOB 12 18 34 34 22 12
MC 38 36.5 31
20 NOB 12 30 53 36 24.33 11.67
MC 40 38 28
22 NOB 15 23 35 36.7 25.3 11.2
MC 39 37 38
24 NOB 15 23 35 36.2 25.3 10.9
MC 39 37 38
28 NOB 13 25 30 37 26 11
MC 38 36 35
32 NOB 15 24 37 30.3 20.2 10.1
MC 37 31 23
36 NOB 15 30 44 29 19 9.2
MC 37 24 15
Note: Mc= Moisture Content; NOB= Number of blows
103
Appendix B
TABULATED COMPACTION TEST RESULT
CKD
(%) PARAMETER TEST NUMBERS MDD
(Mg/m3)
OMC
(%) 1 2 3 4 5
0 𝛾( Mg/m
3) 1.79 1.84 1.88 1.85 1.80
1.88 12.18 M (%) 6.40 7.8 12.18 15.50 17.40
2 𝛾( Mg/m
3) 1.75 1.81 1.86 1.82 1.79
1.86 12.34 M (%) 8.70 9.40 12.34 13.80 15.10
4 𝛾( Mg/m
3) 1.70 1.73 1.75 1.79 1.72
1.79 12.49 M (%) 5.40 8.1 11.45 12.49 18.40
5 𝛾( Mg/m3) 1.73 1.75 1.80 1.74 1.70
1.80 12.52 M (%) 5.57 10.50 12.52 15.70 18.90
6 𝛾( Mg/m3) 1.56 1.52 1.66 1.82 1.56
1.82 12.61 M (%) 6.07 8.88 11.46 12.61 15.60
8 𝛾( Mg/m3) 1.63 1.67 1.75 1.73 1.70
1.75 12.72 M (%) 5.36 9.03 12.72 15.83 18.96
9 𝛾( Mg/m3) 1.63 1.67 1.72 1.67 1.64
1.72 12.87 M (%) 5.95 9.70 12.87 16.90 18.00
10 𝛾( Mg/m3) 1.60 1.64 1.70 1.67 1.66
1.70 13.00 M (%) 5.20 8.24 13.00 15.97 19.20
12 𝛾( Mg/m3) 1.45 1.53 1.68 1.62 1.57
1.68 13.27 M (%) 4.65 9.20 13.27 16.45 18.70
14 𝛾( Mg/m3) 1.48 1.52 1.60 1.57 1.54
1.60 13.61 M (%) 6.02 9.92 13.61 13.82 15.90
15 𝛾( Mg/m3) 1.45 1.51 1.58 1.53 1.46
1.58 13.84 M (%) 6.60 7.51 13.84 16.62 18.20
16 𝛾( Mg/m3) 1.42 1.44 1.45 1.49 1.44
1.49 13.95 M (%) 4.76 8.84 10.84 13.95 17.98
18 𝛾( Mg/m3) 1.40 1.41 1.44 1.42 1.41
1.44 14.30 M (%) 6.89 11.75 14.30 17.45 19.20
20 𝛾( Mg/m3) 1.40 1.41 1.42 1.41 1.40
1.42 14.64 M (%) 7.20 10.00 14.64 15.83 18.70
22 𝛾( Mg/m3) 1.35 1.37 1.40 1.38 1.36
1.40 15.27 M (%) 6.90 11.02 15.27 16.32 18.70
24 𝛾( Mg/m
3) 1.34 1.36 1.40 1.34 1.33
1.40 15.89 M (%) 7.72 10.59 15.89 16.04 18.74
M= Moisture Content (%) 𝛾= dry density in Mg/m3
104
Appendix C
TABULATED UNCONFINED COMPRESSIVE STRENGTH TEST
RESULT
CKD(%) PARAMETER 7 DAYS CURING
RESULT
14 DAYS CURING RESULT
28 DAYS CURING
RESULT
1 2 AV 1 2 AV 1 2 AV
0
δc (MN/m2) 398 392
395 392 398
395 388 402
395
A1X10 -4
(m2) 23.38 23.38 23.83 23.83 23.85 23.85
2
δc (MN/m2) 458 462
460 490 499
495 500 495
498
A1X10 -4
(m2) 24.13 24.13 24.13 24.13 24.13 24.13
4 δc (MN/m2) 519 531 525 588 600 594 612 588 600
A1X10 -4
(m2) 23.38 23.38 24.13 24.13 24.13 24.3
6 δc (MN/m2) 568 568 568 652 664 658 683 664 674
A1X10 -4
(m2) 23.38 23.38 23.38 23.38 23.38 23.38
8 δc (MN/m2) 617 604 611 715 727 721 755 739 747
A1X10 -4
(m2) 23.58 23.58 23.58 23.58 23.58 23.58
10 δc (MN/m2) 730 718 724 829 831 830 867 854 860
A1X10 -4
(m2) 24.48 24.48 23.58 23.58 23.58 23.58
12 δc (MN/m2) 844 832 838 942 934 938 978 968 973
A1X10 -4
(m2) 24.48 24.48 23.58 23.58 23.58 23.58
14 δc (MN/m2) 978 968 973 1067 1069 1064 1093 1084 1088
A1X10 -4
(m2) 23.38 23.38 24.48 24.48 23.83 23.83
16 δc (MN/m2) 1111 1103 1107 1192 1186 1189 1208 1200 1204
A1X10 -4
(m2) 24.88 24.88 24.48 24.48 24.48 24.48
18 δc (MN/m2) 1240 1240 1240 1324 1318 1320 1345 1346 1346
A1X10 -4
(m2) 23.83 23.83 23.83 23.83 23.83 23.83
20 δc (MN/m2) 1370 1375 1373 1456 1449 1452 1482 1492 1487
A1X10 -4
(m2) 23.83 23.83 23.58 23.58 23.58 23.58
22 δc (MN/m2) 1378 1354 1366 1458 1450 1454 1483 1494 1489
A1X10 -4
(m2) 23.13 23.13 24.13 24.13 23.58 23.58
24 δc (MN/m2) 1387 1333 1360 1456 1451 1454 1484 1496 1490
A1X10 -4
(m2) 23.83 23.83 24.13 24.13 24.13 24.13
Note:δc= Maximum Stress Sustain by UCS Sample in MN/m2 (UCS
Values)
A1 Corrected Area at Maximum Stress in m
2
AV= Average
105
Appendix D
SPECIFIC GRAVITY TEST RESULT OF THE CKD SAMPLE
BS 1377: 1975
S/N CKD
SAMPLE NO A B
BOREHOLE NO
DEPTH OF SAMPLE
LOCATION
MASS OF BOTTLE (M1) g 457.00 457.00
MASS OF BOTTLE + SAMPLE (M2) g 859.00 858.00
MASS OF BOTTLE +WATER+SAMPLE (M3) g 1480.00 1482.00
MASS OF BOTTLE FULL OF WATER ( M4) g 1234.00 1237.00
MASS OF WATER USED (M3-M2) g 621.00 630.00
MASS OF SAMPLE USED (M2-M1) g 402.00 395.00
Gs= M2-M1/[(M4-M1)-(M3-M2)] 2.56 2.63
MEAN SPECIFIC GRAVITY, Gs 2.60
106
APPENDIX E
SPECIFIC GRAVITY TEST RESULT OF THE SOIL SAMPLE
BS 1377: 1975
S/N SOIL SAMPLE
SAMPLE NO
BOREHOLE NO A B
DEPTH OF SAMPLE
LOCATION
MASS OF BOTTLE (M1) g 457.00 457.00
MASS OF BOTTLE + SAMPLE (M2) g 862.00 865.00
MASS OF BOTTLE +WATER+SAMPLE (M3) g 1485.00 1481.00
MASS OF BOTTLE FULL OF WATER ( M4) g 1239.00 1235.00
MASS OF WATER USED (M3-M2) g 623.00 616.00
MASS OF SAMPLE USED (M2-M1) g 405.00 408.00
Gs= M2-M1/[(M4-M1)-(M3-M2)] 2.55 2.52
MEAN SPECIFIC GRAVITY, Gs 2.54
107
Appendix F
CORRELATION BETWEEN THE EXPERIMENTAL AND MODEL OPTIMUM MOISTURE
CONTENTS
Substituting the values obtained into the equation below,
r = 0.9977
The tabulated, r value at 4 degrees of freedom for p= 0.005 from table of correlation coefficient is
0.811. Our calculated r, 0.9977 value exceeds the tabulated at p=0.05, so the experimented and
modelled results are positively correlated. The implication of this is that the model can be use to
predict other results.
S/N X (Experimental) Y (Model) X2 Y
2 XY
1 12.55 12.75 157.50 162.56 160.01
2 12.72 12.76 161.80 162.82 162.37
3 12.86 12.87 165.38 165.64 165.51
4 13.78 13.76 189.89 189.34 189.61
5 13.95 13.95 194.60 194.60 194.60
6 15.27 15.27 233.17 233.17 233.17
∑ 81.13 81.36 1102.34 1108.13 1105.27
108
Appendix G
CORRELATION BETWEEN THE EXPERIMENTAL AND MODEL MAXIMUM DRY DENSITY
Substituting the values obtained into the equation below,
r = 0.8728
The tabulated, r value at 4 degrees of freedom for p= 0.005 from table of correlation coefficient is
0.811. Our calculated r, 0.8728 value exceeds the tabulated at p=0.05, so the experimented and
modelled results are positively correlated. The implication of this is that the model can be use to
predict other results.
S/N X (Experimental) Y (Model) X2 Y
2 XY
1 2.13 2.23 4.54 4.97 4.75
2 2.24 2.38 5.02 5.66 5.33
3 2.42 2.41 5.86 5.81 5.83
4 2.33 2.40 5.43 5.76 5.59
5 2.13 2.37 4.54 5.62 5.05
6 1.93 1.97 3.72 3.88 3.80
∑ 13.18 13.76 29.11 31.70 30.35
109
Appendix H
CORRELATION BETWEEN THE EXPERIMENTAL AND MODEL CALIFORNIA BEARING
RATIO (UN-SOAKED)
Substituting the values obtained into the equation below,
r = 0.9247
The tabulated, r value at 4 degrees of freedom for p= 0.005 from table of correlation coefficient is
0.811. Our calculated r, 0.9247 value exceeds the tabulated at p=0.05, so the experimented and
modelled results are positively correlated. The implication of this is that the model can be use to
predict other results.
S/N X (Experimental) Y (Model) X2 Y
2 XY
1 47 40.64 2,209.00 1,651.61 1,910.08
2 52 48.09 2,704.00 2,312.65 2,500.68
3 54 50.38 2,916.00 2,538.14 2,720.52
4 59.5 62.23 3,540.25 3,872.57 3,702.69
5 60 63.88 3,600.00 4,080.65 3,832.80
6 78 71.87 6,084.00 5,165.30 5,605.86
∑ 350.50 337.09 21,053.25 19,620.92 20,272.63
110
Appendix I
CORRELATION BETWEEN THE EXPERIMENTAL AND MODEL CALIFORNIA BEARING
RATIO (SOAKED)
Substituting the values obtained into the equation below,
r = 0.9994
The tabulated, r value at 4 degrees of freedom for p= 0.005 from table of correlation coefficient is
0.811. Our calculated r, 0.9994 value exceeds the tabulated at p=0.05, so the experimented and
modelled results are positively correlated. The implication of this is that the model can be use to
predict other results.
S/N X (Experimental) Y (Model) X2 Y
2 XY
1 5.2 5.05 27.04 25.50 26.26
2 7.1 7.40 50.41 54.76 52.54
3 8.1 8.35 65.61 69.72 67.64
4 14.8 15.03 219.04 225.90 222.44
5 16.6 16.35 275.56 267.32 271.41
6 25 25.46 625.00 648.21 636.50
∑ 76.80 77.64 1,262.66 1,291.41 1,276.79
111
Appendix J
CORRELATION BETWEEN THE EXPERIMENTAL AND MODEL 7 DAYS COMPRESSIVE
STRENGTH
Substituting the values obtained into the equation below,
r = 0.9937
The tabulated, r value at 4 degrees of freedom for p= 0.005 from table of correlation coefficient is
0.811. Our calculated r, 0.9937 value exceeds the tabulated at p=0.05, so the experimented and
modelled results are positively correlated. The implication of this is that the model can be use to
predict other results.
S/N X (Experimental) Y (Model) X2 Y
2 XY
1 547 567 299,209 321,489 310,149
2 611 697 373,321 485,809 425,867
3 668 741 446,224 549,081 494,988
4 1049 1016 1,100,401 1,032,256 1,065,784
5 1070 1064 1,144,900 1,132,096 1,138,480
6 1366 1359 1,865,956 1,865,956 1,856,394
∑ 5,311 5,444 5,230,011 5,386,687 5,291,662
112
Appendix K
Table Correlation Coefficients
Degrees of Freedom Probability, p
0.05 0.01 0.001
1 0.997 1.000 1.000
2 0.950 0.990 0.999
3 0.878 0.959 0.991
4 0.811 0.917 0.974
5 0.755 0.875 0.951
6 0.707 0.834 0.925
7 0.666 0.798 0.898
8 0.632 0.765 0.872
9 0.602 0.735 0.847
10 0.576 0.708 0.823
11 0.553 0.684 0.801
12 0.532 0.661 0.780
13 0.514 0.641 0.760
14 0.497 0.623 0.742
15 0.482 0.606 0.725
16 0.468 0.590 0.708
17 0.456 0.575 0.693
18 0.444 0.561 0.679
19 0.433 0.549 0.665
20 0.423 0.457 0.652
25 0.381 0.487 0.597
30 0.349 0.449 0.554
35 0.325 0.418 0.519
40 0.304 0.393 0.490
45 0.288 0.372 0.465
50 0.273 0.354 0.443
60 0.250 0.325 0.408
70 0.232 0.302 0.380
80 0.217 0.283 0.357
90 0.205 0.267 0.338
100 0.195 0.254 0.321
Retrieved from www.biology.ed.ac.uk/research/groups/jdeacon/statistics/tress.11.html