Reynolds - 2013 - Thesis UQ - Eng Correlations For The Characterisation of Reactive Soil Behaviour For Use in Road Design

University of Southern Queensland
Faculty of Health, Engineering & Sciences
Engineering Correlations for the

Characterisation of Reactive Soil Behaviour for
Use in Road Design
A dissertation submitted by
Peter William Reynolds
in fulfillment of the requirements of
Courses ENG4111 and ENG4112 Research Project
towards the degree of
Bachelor of Engineering (Civil)
Submitted: October 2013

Abstract
In the field of road design and construction, expansive or reactive soils are
problematic materials. Clay minerals within reactive soils are subject to large volume
changes when exposed to water, and conversely when they are exposed to prolonged
periods of drying.
The surface movements resulting from the wetting or drying of a reactive soil can
cause distress to structures that are founded on them. This creates safety and
serviceability issues for road users and high maintenance costs to the road authorities
and the community.
Whilst there is a body of data and published information on some of the relationships
between certain material parameters, a definitive engineer’s guide on the correlations
of engineering parameter for expansive soils within Queensland does not appear to
exist.
Currently, the primary reference for site classification in respect to the degree of
reactivity is the Australian Standard AS2870 – Residential Slabs and Footings. Within
this standard, methods are provided to enable an estimation of the range of vertical
movement due to swelling and shrinkage. These estimates are based on the Shrink
Swell Index (Iss), which is determined by as simple soil test method on an undisturbed
soil sample taken from the site of investigation.
In situations where it may be difficult to obtain undisturbed soil samples for shrink
swell testing, using an approximation of the relationship between other test
parameters and the Shrink Swell Index, it may be possible to determine the site
classification and estimate the amount of potential heave or shrinkage, as an
alternative to the Shrink Swell Index.
i
Historical data was gathered from reported site investigations carried out on
Queensland state road projects from 1995 to 2012. From this data, the relationships
between the measures and indices from some of the most commonly used laboratory
methods for characterising reactive soils were examined.
Some useful relationships between various parameters were identified that will assist
engineers in simplifying the identification and classification process.
ii
ENG4111/2 Research Project
Limitations of Use
The Council of the University of Southern Queensland, its Faculty of Health,

Engineering & Sciences, and the staff of the University of Southern Queensland, do
not accept any responsibility for the truth, accuracy or completeness of material
contained within or associated with this dissertation.
Persons using all or any part of this material do so at their own risk, and not at the risk
of the Council of the University of Southern Queensland, its Faculty of Health,
Engineering & Sciences or the staff of the University of Southern Queensland.
This dissertation reports an educational exercise and has no purpose or validity

beyond this exercise. The sole purpose of the course pair entitled “Research Project"
is to contribute to the overall education within the student's chosen degree program.
This document, the associated hardware, software, drawings, and other material set
out in the associated appendices should not be used for any other purpose: if they are
so used, it is entirely at the risk of the user.
Dean
iii
Certification of Dissertation
I certify that the ideas, designs and experimental work, results, analyses and
conclusions set out in this dissertation are entirely my own effort, except where
otherwise indicated and acknowledged.
I further certify that the work is original and has not been previously submitted for
assessment in any other course or institution, except where specifically stated.
iv
Acknowledgments
The author would like to thank the following people for their assistance: to Dr Kazem
Ghabraie (University of Southern Queensland) for his technical guidance, feedback
and support during this project; to Mr Siva Sivakumar (Department of Transport and
Main Roads) for his assistance in the selection of this topic of research and for his
permission in allowing access to the data resources of the Department of Transport
and Main Roads; to my colleague Mr Jeremy Kirjan, for his encouragement and
support; and to my family, for their support and patience in what has been a long wait
for the end of this chapter.
Peter Reynolds
October 2013
v
Table of Contents
Abstract....................................................................................................................................... i
Limitations of Use ....................................................................................................................iii
Certification of Dissertation...................................................................................................... iv
List of Figures.........................................................................................................................viii
List of Tables ............................................................................................................................ ix
1 Introduction....................................................................................................................... 1
1.1 Aims and Objectives of the Project .......................................................................... 2
2 Impacts on Road Performance .......................................................................................... 3
2.1 Distress on Drainage Structures – Case History....................................................... 3

2.2 Distress on Road Embankment – Case History........................................................ 4
3 Background and Literature Review .................................................................................. 6
3.1 Properties of Reactive Soils ..................................................................................... 6

3.2 Mineralogy and Mechanics of Volume Change....................................................... 7
3.3 Soil Suction .............................................................................................................. 9
3.4 Active Depth........................................................................................................... 13
3.5 Applied Stress......................................................................................................... 14
4 Laboratory Methods for Measuring Reactivity............................................................... 14
4.1 Laboratory Measurements...................................................................................... 14
5 Predictions of Surface Movement using AS2870 ........................................................... 26
5.1 General ................................................................................................................... 26

5.2 Instability Index...................................................................................................... 27
5.3 Changes in Soil Suction ......................................................................................... 28
5.4 Previous Assessments of Test Parameters with Shrink Swell Index...................... 28
6 Analysis of Historical Data ............................................................................................. 29
6.1 General ................................................................................................................... 29

6.2 Method of Analysis ................................................................................................ 30
6.3 Sample Population Statistics .................................................................................. 32
6.4 Statistic Relationships ............................................................................................ 33
7 Discussion of Results ...................................................................................................... 33
vi
7.1 General ................................................................................................................... 33
7.2 Correlation of Individual Parameters and Shrink Swell......................................... 34
7.2.1 Correlation of Shrink Swell Index and % Passing 0.075mm............................. 34
7.2.2 Correlation of Shrink Swell Index and % Passing 0.002mm............................. 35
7.2.3 Correlation of Shrink Swell Index and Liquid Limit ......................................... 36
7.2.4 Correlation of Shrink Swell Index and Plasticity Index..................................... 37
7.2.5 Correlation of Shrink Swell Index and Linear Shrinkage .................................. 38
7.2.6 Correlation of Shrink Swell Index and Weighted Plasticity Index .................... 39
7.2.7 Correlation of Shrink Swell Index and Weighted Linear Shrinkage ................. 40
7.2.8 Correlation of Shrink Swell Index and CBR Swell ........................................... 40
7.2.9 Correlation of Shrink Swell Index and Soil Suction.......................................... 42
7.2.10 Correlation of Shrink Swell Index and Swelling Pressure............................. 43
7.2.11 Correlation between shrink swell index and CEC ......................................... 44
7.3 Correlation of Combined Parameter Functions and Shrink Swell Index ............... 44
7.3.1 Correlation of Shrink Swell Index and (LS + CBR Swell) ................................ 45
7.3.2 Correlation of Shrink Swell Index and “Combined Reactivity Index”.............. 46
7.3.3 Correlation of Shrink Swell Index and (LS/CBR Swell) Ratio ......................... 47
7.4 Summary of Correlation Analysis .......................................................................... 48
8 Conclusions..................................................................................................................... 48
8.1 Major Outcomes and Key Findings........................................................................ 48

8.2 Recommendations for Further Work...................................................................... 50
9 List of References ........................................................................................................... 52
Appendix A Project Specification.................................................................................. 60
Appendix B Tabulated Test Data................................................................................... 62
Appendix C Project Aims and Objectives ..................................................................... 75
Appendix D Project Methodology ................................................................................. 77
Appendix E Assessment of Consequential Effects........................................................ 81
Appendix F Risk Assessment........................................................................................ 84
Appendix G Resource Analysis ..................................................................................... 87
Appendix H Project Timeline ........................................................................................ 89
vii
List of Figures
Figure 1 - Distribution of Cracking Clays within Queensland (Main Roads 2000) .................. 2
Figure 2 - Image of damaged abutment wall of culvert at East Warianna Creek ...................... 4
Figure 3 - Location of distressed embankments on Bruce Highway, Yandina.......................... 6
Figure 4 - Diagram of the structure of a montmorillonite layer (Al-Omari 2000) .................... 8
Figure 5 - Comparison of mineralogical structures of kaolinite and montmorillonite (Farris) . 8
Figure 6 - Phase diagram showing soil particles and water in void space................................. 9
Figure 7 - Diagram of soil water potential energy states (Or et al 2005) ................................ 10
Figure 8 - Idealised water content profiles within active depth zone (Nelson et al 2001)....... 13
Figure 9 - CBR test specimens with swell measurement apparatus (Walters 2008) ............... 16
Figure 10 - Hydrometer Testing (Earl 2005) ........................................................................... 17
Figure 11 - Soil consistency with increasing water content (Sivakugan 2000) ....................... 17
Figure 12 - Apparatus for Casagrande liquid test (Earl 2010)................................................. 19
Figure 13 - Apparatus for fall cone liquid limit test (TMR 2013) ........................................... 19
Figure 14 - Plastic limit test specimens adjacent to 3mm diameter guide rod (Earl 2010) ..... 20
Figure 15 - Soil specimen in linear shrinkage test mould (Main Roads 2008)........................ 20
Figure 16 - Core shrinkage specimen ...................................................................................... 23
Figure 17 - Apparatus (Oedometer) for swell tests.................................................................. 24
Figure 18 - Image showing filter paper and soil specimen for suction test (Bulut 2001)........ 25
Figure 19 - Predicted surface movement compared to observed surface movement (Fityus
2005) ........................................................................................................................................ 28
Figure 20 - Site locations referenced by TMR technical reports ............................................. 31
Figure 21 - Linear regression plot of % passing 0.075mm and shrink swell index................. 35
Figure 22 - Linear regression plot of % passing 0.002mm and shrink swell index................. 36
Figure 23 - Linear regression plot of liquid limit and shrink swell index ............................... 37
Figure 24 - Linear regression plot of plasticity index and shrink swell index......................... 38
Figure 25 - Linear regression plot of linear shrinkage and shrink swell index........................ 39
Figure 26 - Linear regression plot of weighted plasticity index and shrink swell index ......... 40
Figure 27 - Linear regression plot of weighted linear shrinkage plot and shrink swell index. 41
Figure 28 - Linear regression plot of CBR swell and shrink swell index................................ 42
Figure 29 - Linear regression plot of soil suction and shrink swell index ............................... 43
Figure 30 - Linear regression plot of swelling pressure and shrink swell index...................... 43
Figure 31 - Linear regression plot of linear shrinkage + CBR Swell and shrink swell index . 45
Figure 32 - Linear regression plot of CRI and shrink swell index........................................... 47
Figure 33 - Linear regression plot of ratio of CBR swell/LS and shrink swell index.............. 47
viii
List of Tables
Table 1 - Suction values corresponding to certain soil states (Lopez et al 1996).................... 12
Table 2 - Guidelines for interpreting strength of relationships (Crewson 2006) ..................... 32
Table 3 - Modified guidelines for interpreting strength of relationships (including R2) ......... 32
Table 4 - Summary of statistical properties of sample data..................................................... 33
Table 5 - Summary of correlation analysis of all parameters with shrink swell index............ 49
ix
1 Introduction
In the field of road design and construction, expansive or reactive soils are
problematic materials. The clay minerals within reactive soils are subject to large
volume changes when exposed to water, and conversely when they are exposed to
prolonged periods of drying. For some clay minerals, such as montmorillonite, the
volume change due to the absorption or removal of water can be as much as 30%.
This can result in resulting surface movements of up to (and sometimes greater
than) 75mm1, causing distress to the structures founded on them resulting in safety
and serviceability issues for road users and high maintenance costs to the road
authorities and the community.
Expansive soils are widely distributed over almost all geographical locations in
the world. In Queensland, expansive or reactive soils are referred to by soil
scientists as “Cracking Clays” or, more commonly, as “Black Soils” (Dept of
Main Roads Qld, 2000). The distribution of these Cracking Clays by land area
covers approximately one third of the state. Figure 1 below illustrates the extent of
these types of soils within Queensland, based on geological soil mapping.
It is important to properly characterise the properties of expansive soils prior to

construction to minimise their impact on the long term performance of
infrastructure.
In the analysis and testing of expansive soils, there are various measures and
indices used by road designers, and pavement/geotechnical engineers to predict
behaviour during the service life of road infrastructure such as pavements,
embankments and culverts. Parameters such as CBR Swell, Shrink Swell Index,
soil suction, weighted plasticity index, swelling pressure, weighted linear
shrinkage, as well as clay content, cation exchange capacity are regularly used.
1
AS2870 - 1996
1
Figure 1 - Distribution of Cracking Clays within Queensland (Main Roads 2000)
Whilst there is a body of data and published information on some of the

relationships between certain material parameters, a definitive engineer’s guide on
the correlations of engineering parameter for Queensland soils and climatic
conditions does not appear to exist. Whilst it is acknowledged that to
comprehensively characterise the expansive behaviour of all Queensland soil
types is beyond the scope of this project, it is hoped that some useful relationships
may be identified during the course of the research.
1.1 Aims and Objectives of the Project

The aim of this project is to identify and quantify the relationships that exist
between key measures of reactive soil behaviour. The study will focus on the most
common methods of laboratory testing used for identifying and characterising
reactive soils within Australia for the purposes of the field of road construction.
Historical data gathered from investigations and testing undertaken since the mid
1990s, by the Department of Transport and Main Roads, Queensland, will be used
to carry out parametric studies. Any linkages or correlations that may exist
2
between the key indicators of reactive soil behaviour will be identified and
quantified for future use as guides in the design of roads using these materials.
The objectives of this project can be summarised as follows:

• To understand the nature of expansive soils and their properties
• To identify by research, the key measures of reactive soil behaviour used by
engineers in road construction.
• To investigate and quantify any relationships that may exist between the key
measures.
• To establish a ranking of reliability for the key measures of reactive soil
behaviour.
• To develop guidelines to assist road designers in the identification and
characterisation of reactive soils during the site investigation phase of road
construction projects.
2 Impacts on Road Performance
The key elements of a road that are at most risk of damage due to reactive soils
are primarily structures on shallow foundations, such as culverts, and pavement
layers. The use of reactive soil as fill materials in embankment construction also
poses a significant risk for long term performance, if adequate controls are not
implemented.
The following section of this report will illustrate by case history, the damaging
effects of reactive soil, if the designs do not adequately account for the properties
of these materials.
2.1 Distress on Drainage Structures – Case History

During a routine inspection of drainage structures in western Queensland in 2011,
the Department of Main Roads Queensland identified major distress to a concrete
culvert on the Kennedy Developmental Road, south west of Hughendon, in north
western Queensland. The structure had undergone severe cracking to the
3
abutment, wing and pier walls2. Figure 2 below shows the extent of the distress
on the culvert.
The distress to the culvert had rendered the structure unsafe, and resulted in its
replacement. Actions required to replace the culvert included the construction of a
temporary side track, demolition of the existing culvert, removal of the reactive
soil to a depth below the zone of influence of the reactive soil, and a replacement
of this soil with non-reactive/stable fill. A new culvert base was constructed
complete with new concrete culvert cells. It was estimated that the final costs of
the replacement works were in the order of $500000.
Based on testing of soil samples taken from the site, it was determined that the
cause of the cracking to the structure was due to the stresses imposed on the
structure as a result of the shrinking and swelling of the foundation soil as a result
of seasonal moisture variations.
Figure 2 - Image of damaged abutment wall of culvert at East Warianna Creek
2.2 Distress on Road Embankment – Case History

Another example of the problems reactive soil can cause to road construction is
when they are used as embankment fill. In this example, the approaches to a
railway overbridge on the Bruce Highway at Yandina in Queensland, were
constructed in 1997-19983. The embankments (maximum height 9m) were
2
MR2469 Department of Transport and Main Roads (2011)
3
MR1923 Qld Dept. of Main Roads (2001)
4
constructed from local soil materials sourced from a site near to the railway
crossing.
Shortly after the completion of construction, longitudinal cracks were observed on

the outside of the edge of the concrete pavement wearing course. These cracks
were sealed with polystyrene bitumen, however, the cracking continued. Due to
concerns about the possible major damage to the concrete pavement, a site
investigation was carried out to determine the cause of the cracking. Samples of
the embankment were tested in the laboratory and found to have high shrink swell
properties. A review of the construction records found that the embankment fill
had not been placed in accordance with the road authority’s technical guidelines
on construction of expansive clay embankments.
The cause of the cracking was found to be as a result of the shrinkage of the
clayey soils within the embankment fill due to a prolonged dry period post
construction. It was proposed that the steep embankment batters (1V:1.5H)
enabled the drying to occur more rapidly than wound normally be expected. In
addition, the flanks of the embankment were not constructed of material suitable
for protecting the embankment core from fluctuating moisture contents, due to
climatic variations.
The adopted remediation action taken to repair the embankment was to flatten the
existing batter slopes using non-expansive soil (1V:2.5H). The estimated total cost
of the final repairs to the embankment was approximately $750000.
As the above two case studies illustrate, the cost to the community of
underperformance of engineering structures is significant. It is important to
properly identify and characterize reactive soil sites and to carefully quantify their
potential for shrink-swell movement.
5
Figure 3 - Location of distressed embankments on Bruce Highway, Yandina
3 Background and Literature Review
3.1 Properties of Reactive Soils

Expansiveness or reactivity is a property of a soil to undergo large volume
changes when subject to the actions of wetting and drying. Consideration of the
mechanism of interaction between water and reactive soils show that the three
most important components are mineralogy, the change in moisture content or
suction due to climatic conditions (atmospheric changes), and the stress applied to
the soil.
Fine grained soils (eg clayey soils) are known to have a greater potential for
reactivity than coarse grained soils (eg soils and gravels). For fine grained soils,
the type and proportion and type of clay mineral present influences the potential
for reactivity. For example, soils with a high proportion of clay mineral
montmorillonite are known to exhibit the very high amount of reactive behaviour.
Another factor which influences expansive behaviour is the potential for change in
the moisture content of a soil, known as soil suction. Moisture conditions beneath
the ground surface are described by the actual moisture content and the water
pressure within the void space; the pore water pressure. Where there is a deficit of
water within the soil, this usually gives rise to a negative pore water pressure,
6
known as soil suction. The drier the soil, the greater the suction, and potential for
absorption of moisture. 4
The depth within a soil at which a stable or equilibrium water content can be
found is known as the Active Depth or Zone of Seasonal Fluctuations. Soil above
the Active Depth is subject to changes in water content and soil suctions, due to
climatic conditions; specifically the seasonal cycles of wetting and drying, which
influence the degree of reactivity of the soil.
In addition to these intrinsic properties, the magnitude of the stresses applied to

the soil by overburden or, for example, embankment loading, also governs the
degree to which the soil can react. Problems of volumetric movement and
swelling pressure on elements of road infrastructure only arise when the
expansiveness is unable to be suppressed.
3.2 Mineralogy and Mechanics of Volume Change

The type of clay mineral is largely responsible for determining the intrinsic
expansiveness of the soil. Kaolinitic clays are relatively non-expansive whilst the
more expansive clays are smectite clays, also known as montmorillonite clays.
The formula for the chemical composition of montmorillonite is
Al4Si8O20(OH)4.nH2O
The structure of the clay particles consists mainly of three (3) layers, octahedral
sheet, usually occupied by aluminium or magnesium (gibbsite) sheets sandwiched
between two sheets of tetrahedral silicon (silica) sheets to give a two (2) to one (1)
lattice structure5. This network typically allows water to enter to the centre of the
clay particle and be retained for long periods of time (Figure 4). The swelling
occurs due to the poor electrostatic bonds between the silica sheets, enabling
4
Nelson et al. (1992)
5
Chen (1988)
7
osmotic pressures to build up and allow for water molecules to be absorbed
between the sheets6.
Figure 4 - Diagram of the structure of a montmorillonite layer (Al-Omari 2000)
By comparison, kaolinite, which exhibits little expansive properties, comprises a

one to one lattice structure Figure 5 and has a much lower affinity to absorb or
retain water molecules within its lattice.
Figure 5 - Comparison of mineralogical structures of kaolinite and montmorillonite (Farris)
6
Al-Omari et al (2000)
8
This retention is responsible for the movement associated with expansive
materials in road construction.
3.3 Soil Suction

A volume of soil is comprised of solid particles and void space. The void space
can be filled with aid and water. When the void space of a soil is completed filled
with water, it is said to be saturated. Conversely, when the void space is only
partly filled with water, it is said to be in an unsaturated state.
The mechanics of the way in which the soil will behave in the unsaturated state, is
governed by the inter-particle forces, the air and water pressures within the void
space, and the surface tension arising from the interactions of the water and air
within the voids. Figure 6 below shows a phase diagram of an unsaturated soil.
Figure 6 - Phase diagram showing soil particles and water in void space
The retention and movement of water in soils and its loss to the atmosphere is best
described in terms of energy. The more strongly water is held by a soil, the greater
is the energy required for it to be released. As such, if water is to be removed from
a moist soil, work has to be done against adsorptive forces; the forces that cause
the adhesion of a thin layer of water to the surface of soil particles. Conversely,
when water is adsorbed by the soil, a negative amount of work is done. The
potential energy required to cause the movement of soil water from a zone where
9
the free energy of water is high (eg standing water table) to one where the free
energy is low (a dry soil), is referred to the total soil water potential.
The key components that comprise the total soil water potential are the Matric
potential, the Osmotic potential, the potential due to Gravitational forces, and the
Pressure potential, due to applied pressures or stresses on a soil.7
The Matric potential is the energy associated with the attraction of water by the
particles within the soil; Osmotic or solute potential describes the energy
associated with the attractive forces within a soil due to water containing salts.
Gravitational potential relates to the energy associate with water at an elevation,
and pressure potential describes the energy due to any applied stresses to the soil.
Figure 7 below shows graphically the soil water energy states.
Figure 7 - Diagram of soil water potential energy states (Or et al 2005)
The total soil water potential can be expressed in the following equation:
7
Or, Tuller and Wraith (2003)
10
(Or, Tuller and Wraith)
Where
= The total soil water potential, in centimetres
= The matric potential, in centimetres
= The osmotic (solute) potential, in centimetres
= The pressure potential, in centimetres
= The gravitational potential, in centimetres
In practical terms, the primary forces acting on soil water held within a rigid soil
matrix are: (i) matric forces resulting from interactions of the solid phase with the
liquid and gaseous phases (matric potential) and (ii) osmotic forces (solute
potential) owing to differences in chemical composition of soil solution.
As stated above, the matric potential is caused by a difference between the air and
water pressures within the pore space within the soil.
This is expressed as the simplified equation:
Where
= pore air pressure, in centimetres
= pore water pressure, in centimetres
Generally, soil water potential is referred to as total suction terms, and is

expressed as:
11
Methods of expressing suctions - there are two units to express differences in
energy levels of soil water:
pF Scale: The soil water potential (total suction) is measured in terms of the height
of a column of water required to produce necessary suction or pressure difference
at a particular soil moisture level. The normal range of the values of suction is
wide. Suction is usually expressed as a logarithm of the height of water column
(cm) to give the necessary suction, in units of picoFarads (pF); ie
Lopes and McManus (1996) published indicative values relating to specific soil
states and are presented in Table 1 below.
Atmospheres or Bars: It is another common mean of expressing suction. One

atmosphere (one bar) is the average air pressure at sea level. If the suction is very
low as occurs in the case of a wet soil containing the maximum amount of water
that it can hold, the pressure difference is of the order of about 0.01 atmospheres
or 1 pF equivalent to a column of water 10 cm in height. Similarly, if the pressure
difference is 0.1 atmosphere the pF will be 20. Soil moisture constants can be
expressed in term of pF values. A soil that is saturated with water has PF 0 while
an oven dry soil has a pF 7.
Suction (pF) Soil State

6.5-7.0 Oven Dry
6.0 Air Dry
5.5 Shrinkage Limit
4.5 Wilting Point (Plants)
2.0-1.5 Swell Limit
1.0 Liquid Limit
Table 1 - Suction values corresponding to certain soil states (Lopez et al 1996)
12
3.4 Active Depth
For the problematic behaviour of reactive soils to manifest itself, the soil must be
subject to prolonged periods of wetting or drying. If the moisture content within a
soil is constant, there is stability. Conversely, where the strata are subject to
seasonal fluctuations of soil moisture, the volumetric changes are observable.
Figure 8 shows a number of idealised water content profiles for a problematic

reactive clay soil. Profile A demonstrates a water content profile in a uniform soil
at an undeveloped site in a dry climate. Below some depth (Depth Zs) an
equilibrium water content exists. This is referred to as the Active Zone.
Figure 8 - Idealised water content profiles within active depth zone (Nelson et al 2001)
Above Depth Zs, the water content decreases due to water losses from the ground
surface, usually evaporation and also transpiration from vegetation. If a cover is
placed on the ground surface that is large enough that edge effects can be
neglected, surface water losses are eliminated, and the water content profile will
come into equilibrium with the environment as shown, for example, by Profile B.
13
If the ground surface is subjected to temperature fluctuations such as due to
summer and winter climates, the water contents in the zone affected by
temperature changes will fluctuate about Profile B. Profile C would be typical of
wet climatic conditions and Profile D would be typical for dry climatic conditions.
The zone in which temperature effects occur, and depths below that in which
climatic effects can change the water content define the zone of seasonal
fluctuation. The depth of this zone would be less than or equal to the depth Zs
(Nelson et al. 2001).
3.5 Applied Stress

The applied stress depends on the structural design and the embankment
geometry. The embankment material will act to suppress vertical movement. The
amount of overburden required to suppress movement depends on the
expansiveness of the soil, and the active depth, as described above.
4 Laboratory Methods for Measuring Reactivity
4.1 Laboratory Measurements

Below is a summary of the most commonly used laboratory methods in the
determination of soil reactivity for roads (Pritchard et al. 2000)
• California Bearing Ratio - % Swell
• Particle Size Distribution using Hydrometer (% Finer that 2 um)
• Atterberg Limits (Liquid Limit, Plasticity Index, Linear Shrinkage)
• Weighted Plasticity Index
• Weighted Linear Shrinkage
• Shrink Swell Index
• Loaded Swell/Swelling Pressure
• Soil Suction – Filter Paper Method
• Cation Exchange Capacity
14
California Bearing Ratio (CBR) - % Swell
CBR tests were developed in 1952 to assess the strength of pavement subgrades
(Lacey, 1998). California Bearing Ratio is defined as the ratio of force required to
cause a circular plunger of 1932mm2 area to penetrate the material for a specified
distance expressed as a percentage of a standard force (Main Roads 2002). Test
specimens are prepared from passing 19.0mm material using a compactive effort
of 596 kJ/m3. They are then tested either in a soaked or unsoaked condition.
The method allows for the determination of CBR Maximum Dry Density (MDD)
and CBR Optimum Moisture Content (OMC) as well as the optional
determination of swell and post penetration moisture content (Main Roads 2002).
This test can be used to simulate climatic conditions and estimate the potential
swell behaviour of expansive clays in its natural state (Fox 2002).
The part of the test that is primarily used as a measure of reactivity is the
percentage of swelling measured at the completion of the 4 day soaking period.
This is performed by using a dial gauge mounted on a frame which, when placed
on top of the CBR mould, aligns the dial gauge centrally over a measuring point.
This measuring point is placed on top of the compacted sample, prior to the
mould and soil being placed in the water bath for soaking.
An initial reading is taking on the dial gauge on the unsoaked compacted soil. At
the end of the soaking period (generally 4 days, but for some tests, 10 days),
another reading is taken at the measuring point on the soaked soil. The
percentage of swelling is calculated as the ratio of the difference in vertical
displacements between the dry and the soaked condition and the compacted
height of the soil sample (gauge length).
15
Figure 9 - CBR test specimens with swell measurement apparatus (Walters 2008)
Particle Size Distribution using Hydrometer

Fine grained soils such as silts and clays and have particles smaller than
0.060mm. To determine the grain size distribution of a material <0.075mm sieve
the Hydrometer method is commonly used. This test is important in determining
the amount of clay present in a soil structure, therefore detailing the amount of
plasticity and potential swell the material possesses.
Soil is mixed with water and a dispersing agent, stirred vigorously, and allowed
to settle to the bottom of a measuring cylinder. As the soil particles settle out of
suspension the specific gravity of the mixture reduces.
An hydrometer is then used to record the variation of specific gravity with time.
By relating Stoke’s Law, velocity of a free falling sphere to its diameter the test
data is reduced to provide particle diameters and the percent weight of the sample
finer than a particular particle size.
The hydrometer test is usually discontinued when the percentage of clay sized
particles has been determined.
16
Figure 10 - Hydrometer Testing (Earl 2005)
Atterberg Limits
In the year 1911 Atterberg proposed the limits (liquid limit LL , plastic limit PL
and shrinkage limit SL ) of consistency in an effort to classify the soils and
understand the correlation between the limits and engineering properties like
compressibility, shear strength and permeability (Casagrande, 1932).
The limits represent the water holding capacity at different states of consistency.
The limits are the more prominent procedures for gathering information on the
expansive nature and mechanical behavior of clay soils (Williams, A, 1958).
Figure 6 shows a diagram of the different states of soil consistency with

increasing water content, including the Atterberg Limits.
Figure 11 - Soil consistency with increasing water content (Sivakugan 2000)
17
A useful set of classification data for identifying the swell potential of expansive
subgrades are the liquid limit (LL) and plasticity index (PI). The liquid limit is
the water content at which a soil changes from the liquid state to a plastic state
while the plastic limit is the water content at which a soil changes from the
plastic state to a semisolid.
The Plasticity Index is derived from the plastic limit and liquid limit and is
represented by the equation below:
PI = LL − PL ( %) (Head)
Where:
PI = Plasticity index (%)
LL = Liquid Limit (%)
PL = Plastic Limit (%)
Liquid Limit (LL)

There are two methods to describe the liquid limit (LL) namely percussion cup
method and fall cone method. In the percussion cup method, liquid limit is
defined as the moisture content corresponding to a specified number of blows
required to close a specified width of groove for a specified length (Casagrande,
1932 and 1958). In Queensland, this test is generally carried in accordance with
Australian Standard AS1289.3.1.1
The principle of the determination of the liquid limit from the fall cone method is
that the liquid limit corresponds to a value of the depth of soil penetration due to
a steel cone of specified mass and dimensions.
It is important to note that this depth of penetration value will vary, depending on
the test method used for the determination. For example, the Department of
Transport and Main Roads Queensland method Q104A, specifies that the cone
penetration at liquid limit is a value of 15.5mm. Alternatively, the penetration at
liquid limit for the Australian Standards AS1289.3.9.1 is 20mm.
18
The liquid limit data considered in this study has been determined by both
Casagande and fall cone methods, using Australian Standard and Transport and
Main Roads methods.
Figure 12 - Apparatus for Casagrande liquid test (Earl 2010)
Figure 13 - Apparatus for fall cone liquid limit test (TMR 2013)
Plastic Limit (PL)

Plastic limit is the water content at which the soil begins to crumble when rolled
into 3 mm threads. This is the minimum water content at which the soil will
deform plastically. This gives an accurate representation of the minimum
moisture content a fine grained soil can contain before deformation occurs. This
test can be determined by Australian Standard and Transport and Main Roads
methods (AS1289.3.2.1 and Q105).
19
Figure 14 - Plastic limit test specimens adjacent to 3mm diameter guide rod (Earl 2010)
Shrinkage limit (SL) – Linear Shrinkage Test

The shrinkage limit is the water content dividing the semi-solid and the solid
state of the soil. It is the water content at which further reduction in moisture
content does not result into a decrease in volume of the soil mass. This test can be
determined by Australian Standard and Transport and Main Roads methods
(AS1289.3.4.1 and Q106). A sample of fine-grained soil, at approximately the
liquid limit of the soil, is placed in a shallow trough-shaped mould of 150mm in
length, 25mm wide and 15mm deep. The soil and mould are placed in a low
drying temperature oven for 2 hours and then placed in a high temperature drying
oven overnight.
The length of dried specimen is measured and the linear shrinkage is calculated
as the ratio of the change in length due to drying and the original length.
Figure 15 - Soil specimen in linear shrinkage test mould (Main Roads 2008)
20
Loaded Swell/ Swelling Pressure Test
The loaded swell test provides an indication of the moisture content, soil strength
and surface pressure required to suppress swell of a fully saturated soil. This test
is performed on an undisturbed sample of soil in its natural (insitu) state.
The soil is trimmed in a steel ring so as to be confined laterally. It is placed in an

oedometer apparatus and inundated with water. It is allowed to swell vertically
until it achieves a constant value. This completes the Free Swell test, and from
this, the percentage of swelling strain observed from the insitu moisture content
to the saturated soil condition can be calculated.
Once swell is complete, a series of vertical pressures are applied to the specimen,
consolidating it until it reaches the original height of the specimen. A graph of
Voids Ratio (%) and. Pressure (kPa) can be plotted and from this, an estimation
of the loading required to suppress expansion can be found. This loading value
(in kPa) can then be related to a measurement of soil overburden. All loaded
swell data used in this study has been determined using ASTM test method
D4546 - 03.
Shrink Swell Index (Iss)

The shrink swell test is composed of companion core shrinkage and swelling
tests, carried out on undisturbed soil samples from their initial field moisture
contents. The vertically oriented sample is usually obtained from the ground,
using a 50-mm-diameter thin-walled tube. The sample is extruded from the tube
as a soil core and a suitable portion of the sample is selected for the preparation
of a shrinkage core (70–100 mm long) and a swell core (20–25 mm long). Test
samples must be from adjacent portions of the core to ensure that water content
and both compositional and structural differences are minimized. The test is
carried out in accordance with Australian Standard AS1289.7.1.1.
The shrinkage and swell tests are then conducted simultaneously, as follows.
21
Shrinkage Test—This component of the test is identical in procedure to the core
shrinkage test, although fewer measurements are required as the shrink-swell
index is based on the oven-dried state. A shrinkage core, 45–50 mm in diameter
and a length of 1.5–2 diameters, is trimmed from the soil sample. Where
possible, it is selected and trimmed to be free of major structural defects and
loose material. Initial dimensions and mass are recorded. Small pins are added to
each end as reference points to facilitate consistent measurements of sample
length as drying proceeds.
The shrinkage core is firstly air-dried. Regular measurements of length and mass
are taken until shrinkage ceases. The core is then oven-dried to a constant mass at
105–110

Throughout the drying process, the core is kept in a shallow tray, so that any
crumbs that become loose during the test are not lost, as this would affect
moisture content calculations.
Swelling Test—this involves a simplified oedometer test in which the sample (of
measured mass) is installed in a steel ring, (of measured volume; usually around
20 mm high and 50 mm in diameter) and placed in a consolidation apparatus. A
gauge to monitor the sample height is then zeroed under a nominal seating
pressure of 5 kPa. A load of 25 kPa or the estimated in situ overburden pressure
(whichever is greater) is then applied for 30 minutes to record any initial
settlement or seating adjustment. This displacement is used to correct the initial
sample height for determination of swelling strain.
After re-zeroing the displacement gage, the sample is inundated with distilled
water and allowed to swell until the swelling increment, in a period of not less
than 3 h duration, is not more than 5 % of the total recorded swell.
22
The initial water content is determined from the sample trimmings, and the final
water content is measured from the extracted sample at the end of the test. In the
sample preparation process, particular care is taken to ensure that the sample
neatly fills the sample ring, as voids and recompacted or remolded portions will
accommodate internal adjustments in the volume of the sample and hence, affect
the realized vertical swell. Shrinkage strains and swell strains, measured in the
respective tests, are then combined to give a shrink-swell index (Iss). 8 This is
given by the following equation, (as per AS1289.7.1.1):
∈sw
∈sh +
I ss = 2 in %/pF
1.8
Where
∈sh = shrinkage strain in %
∈sw = swelling strain in %
The factor of 1.8 is a constant that is adopted for all Australian soils. This relates
to the linearly varying part of soil suction changes corresponding to the changes
in the axial strains during shrinking and swelling of the soil.
Figure 16 - Core shrinkage specimen
8
Fitkus et al (2009)
23
Figure 17 - Apparatus (Oedometer) for swell tests
Soil Suction – Filter Paper Method

This method uses laboratory grade filter paper to measure the matrix suction
properties of an undisturbed soil. The filter paper is initially calibrated for suction
at different moisture contents.
Once the calibration curve is completed, the test specimen of soil is carefully
dissected into a minimum of three layers. Three discs of filter paper (pre-weighed
using a balance with an accuracy to 0.0001g) are sandwiched between the soil
slices. The sample is tightly wrapped in plastic, ensuring the paper discs are fully
in contact with soil only. The soil and filter paper is stored in a sealed container
in a controlled temperature environment for 7 days. At the completion of the test
duration, the filter papers are individually removed from between the soil slices
and their wet mass is measured. The papers are then dried to a constant mass in a
drying oven.
The moisture content for each filter paper is then determined, and the average of
all three results is calculated. Using the calibration curve for the specified grade
of filter paper, the soil suction value is determined from the average filter paper
moisture content. The soil suction data used for this study has been determined
using BRE Information Paper Method IP 4-93.9
9
Crilly and Chandler (1998)
24
Figure 18 - Image showing filter paper and soil specimen for suction test (Bulut 2001)
Cation Exchange Capacity

This test method describes the procedures for measuring the soluble and bound
cations as well as the cation exchange capacity (CEC) of fine-grained inorganic
soils.
Clay minerals in fine-grained soils carry a negative surface charge that is
balanced by bound cations near the mineral surface. These bound cations can be
exchanged by other cations in the pore water, which are referred to as soluble
cations. The cation exchange capacity is a measure of the negative surface charge
on the mineral surface. The CEC generally is satisfied by calcium (Ca), sodium
(Na), magnesium (Mg), and potassium (K), although other cations may be
present depending on the environment in which the soil exists. This test method
was developed from concepts described previously in Lavkulich (1981) and
Rhoades (1982) In soils with appreciable gypsum or calcite, dissolution of these
minerals will release Ca in solution that may affect the measurement.
In this test method, the soluble salts from the mineral surface are washed off with
de-ionized water and then the concentration of soluble salts within the extract is
measured. The bound cations of the clay are measured by using a solution
containing an index ion that forces the existing cations in the bound layer into
solution. The total concentrations of bound and soluble cations in this solution
are measured.
The CEC is measured by displacing the index ion with another salt solution and
measuring the amount of the displaced index ion.
25
It was found during the collection of the Cation Exchange Capacity data that the
suppliers of the testing did not use methodology from well recognized suppliers
of standards. Instead it was found that the suppliers used in-house methods based
on conventional chemical analysis techniques.
5 Predictions of Surface Movement using AS2870
5.1 General
The Australian Standard AS2870 uses the following equation to predict the
characteristic surface movement that a particular layer of soil may produce under
seasonal moisture variations:
Where
= characteristic surface movement, in millimeters
= instability index in % picoFarads (pF)
= soil suction change averaged over the thickness of the layer under
consideration, in picoFarads (pF)
= thickness of layer under consideration, in millimeters
= number of soil layers within the design depth of suction change10
The key parameters within this equation are:

• the instability index ( )
• the soil suction change ( )
• the depth of the clay layer ( )
A point worth noting in reference to the principle behind the determination of the
reactivity for a site within then Residential Slabs and Footings Construction
standard AS2870, was made by Brown et al. (2002). Specifically, it was observed
that AS2870 does not refer to standards tests such as Atterberg limits or linear
10
AS2870 Section 2.3(2011)
26
shrinkage to determine reactivity. Instead surface movement (ys) is the primary
characteristic used to classify reactive soils. As detailed above, this incorporates
the use of soil suction data and the instability index derived from shrink-swell
tests into the calculation of surface movement.11
Caunce (2010) observed that the AASHTO guidelines for assessing expansive
soils utilise LL, PI and soil suction, whereas Brown (2002) quotes the Australian
Standard, AS2870, stating that there are no clear tests to determine clay reactivity.
In addition, Brown also postulates that movement is simply a function of
mineralogy, proportion of clay, moisture change, loading and lateral restraint.
As such, from these statements and observations above, it can be concluded that
the instability index accounts for a number of different properties relating to the
reactivity of a soil, within the one single parameter.
5.2 Instability Index

As discussed above, the instability index ( ), in basic terms is a measure of the
soil reactivity. This index value is a function of a number of factors including
matric and osmotic suction and the stress state of the soil, and the moisture
history12. This property can be best approximated by the Shrink Swell Index .
Fityus (2005) reported on the results of a field trial by Cameron in 1989, from a
site in South Australia – Gilles Plains. The mean observations of the ground
surface movements taken over one wet and one dry cycle were compared against
the predicted surface movements based on laboratory values of the shrink swell
index of soil taken from the site. The study found that the predicted movements
calculated using the shrink swell index were a good estimated of actual ground
surface movement. Figure 19 below illustrates these findings.
11
Caunce (2010)
12
Cameron (1989)
27
Figure 19 - Predicted surface movement compared to observed surface movement (Fityus 2005)
5.3 Changes in Soil Suction

Where there is a deficit of water within the soil, this usually gives rise to a
negative pore water pressure, known as soil suction. Essentially, the drier the soil,
the greater the suction, and potential for absorption of moisture.
Soil suction is a property of soil which is a function of the degree of saturation of

the soil and the size of the void spaces within the soil structure. For fine grained
soils such as clays, the void space within the soil structure is relatively small
compared to that of a granular soil. As a result, when a clayey soil is partially
saturated, it is difficult for water to escape from the void space.
5.4 Previous Assessments of Test Parameters with Shrink Swell

Index
Earl (2005) reported on research into the relationships between the shrink swell
index, Atterberg limits and linear shrinkage. Earl found a poor correlation existed
between the shrink-swell index and linear shrinkage, from analysis of test data on
clays from the Shepparton Formation in northern Victoria. However, Earl reported
reasonably good correlations between the shrink-swell index and the plasticity
28
index weighted by the percentage clay fraction, and similarly with the shrink-
swell index and the plasticity index weighted by the percentage clay fraction.
It should be noted that Earl used the a value of the coefficient of determination
(R2) > 0.8 as his criterion for determining if the strength of the relationships
examined in his work would be useful for the purposes of estimation. Whilst this
is considered a reasonable approach, this study has adopted the combined criteria
of a coefficient of correlation (r) > 0.7 and a coefficient of determination of (R2) >
0.5 for determining whether a relationship between parameters would be valid in
the estimation of the shrink swell index/instability index.
Earl also referenced work by Cameron (1989) which identified a correlation

between shrink swell index and linear shrinkage, with a coefficient of correlation
of r= 0.76, for a broad range of soils.
In a similar study, Wan et al. (2002) reported a strong correlation between the
shrink swell of a soil and liquid limit was reported for volcanic soils from
Honolulu.
In research conducted by Earl (2005) a poor correlation was between the shrink-
swell index and LS, however, reasonably good correlations were shown between
the shrink-swell index and the PI factored by the clay fraction and similarly with
the shrink-swell and the PL factored by the clay fraction.
6 Analysis of Historical Data
6.1 General
Based on a search of public records from Department of Transport and Main
Roads Queensland, laboratory test data were extracted from geotechnical reports
on site investigations carried out between 1992 and 2012. The geotechnical
reports were filtered for their relevance to sites that were identified as having
reactive soil. Data from a total of 108 technical reports was reviewed. From these
reports a total of 33 reports were found to have relevant laboratory test
29
information. The data from these specific reports were extracted and compiled for
the purposes of this research.
The reports detailed investigations throughout Queensland’s state road network,

ranging from locations from South East Queensland, to the Darling Downs,
Central Queensland and North Western Queensland. A map of the sites and the
relevant reports to which they relate is shown in Figure 20.
6.2 Method of Analysis

The objective of the analysis of the historical laboratory data is to determine if
there are any relationships between the various parameters obtained from different
laboratory techniques. As previously discussed, the basis for the classification of
reactive sites by characteristic movement, as per AS2870, is the instability index
( ). This is best approximated by the Shrink Swell Index, . Using the as
the independent variable of the relationship, each test parameter was plotted in a
scatter diagram. Using a linear regression analysis of the data, the coefficient of
correlation , and the coefficient of determination were determined for
each set of data. These statistical parameters were used to determine (a) the
strength of the relationship between and the other variables, and (b) measure
of how well the calculated regression line represents the data. Table 2 below gives
some guidance on the interpretation of strength of the relationships.
30
Figure 20 - Site locations referenced by TMR technical reports
Using the criteria for strength of relationship from Crewson above, the
corresponding values of the coefficient of determination (R2) are presented in
Table 3.
This analysis was performed using the shrink swell index as the dependent
variable and the other parameters were treated as independent variables.
31
Range of Coefficient
Strength of Relationship
of Correlation (r) Values
0.9 to 1.0
Very High
(or -0.9 to -1.0)
0.7 to 0.9
High
(or -0.7 to -0.9)
0.5 to 0.7
Moderate
(or -0.5 to -0.7)
0.3 to 0.5
Low
(or -0.3 to -0.5)
0 to 0.3
None
(0 to -0.3)
Table 2 - Guidelines for interpreting strength of relationships (Crewson 2006)
Range of Coefficient Strength of Range of Coefficient of

of Correlation (r) Values Relationship Determination (R2) Values
0.9 to 1.0 0.81 to 1.0
Very High
(or -0.9 to -1.0)
0.7 to 0.9 0.49 to 0.81
High
(or -0.7 to -0.9)
0.5 to 0.7 0.25 to 0.49
Moderate
(or -0.5 to -0.7)
0.3 to 0.5 0.09 to 0.25
Low
(or -0.3 to -0.5)
0 to 0.3 0 to 0.09
None
(0 to -0.3)
Table 3 - Modified guidelines for interpreting strength of relationships (including R2)
6.3 Sample Population Statistics

Table 4 below summarizes the size and distribution of the sample populations of
data for each test parameter used in this study.
32
<0.075m <0.002m LL (%) PI (%) LS (%) WPI WLS Suction - Swelling Iss CBR Swell - CBR Swell - CEC
m m Filter Paper Pressure %/pF 4 Day Soak 10 Day Soak (mEq/100g)
(kPa) (kPa) (%) (%)
N 183 55 209 208 209 209 188 29 21 100 56 13 10
MEAN 77.8 56.6 67.7 40.7 17.5 3710.7 1604.6 1776.7 321.1 3.0 3.8 4.5 39.2
MEDIAN 82.0 57.0 65.8 38.4 17.4 3443.2 1601.0 2008.0 140.0 2.7 3.3 4.0 28.5
SD 17.3 14.8 23.4 17.5 5.4 1808.5 600.9 937.5 371.2 2.0 3.0 3.0 23.3
CV (%) 22 26 35 43 31 49 37 53 116 68 79 66 59
MIN 16 16.6 19.6 4 2.6 346 225 4 49 0.1 -1.9 0.5 19
MAX 100 88 143 100 29.2 9900 2800 3354 1400 9 11.5 7.9 91
Table 4 - Summary of statistical properties of sample data
Despite the high number of test results extracted from the records, there were a
high number of occurrences in which not all laboratory tests were performed on
each soil sample. This had the consequence of reducing the number of sets of
paired data able to be analyzed through linear regression. For example, in
correlating shrink swell index with the soil suction parameter, only 8 pairs of data
out of a possible 29 were considered valid, as being obtained from the same
representative soil sample. As a result, it was not possible to analyse some
relationships, as the sample population was considered too small to be
meaningful.
6.4 Statistic Relationships

Table 5 below summarizes the results of the analysis of the relationships between
the shrink swell index and the other key parameters. Figures 21 to 33 show the
linear regression analysis plots for each pair of parameters compared.
7 Discussion of Results
7.1 General
The intention of this study was to apply a statistical approach to analyzing
historical data from laboratory testing from site investigation activity,
accumulated over a reasonable period of time, to include as many reactive soil
areas within Queensland. Data was limited to that contained in the records of the
Department of Transport and Main Roads – this was, however, considered a
reasonable source of data, as Departmental records of laboratory test results from
site investigations date back to the late 1960s.
33
In gathering the data, the main objective was to collect as many sets of data which
included the Shrink Swell Index in addition to the other parameters of interest.
The purpose of this was ultimately to enable a direct substitution of one of a
number of test parameters used to characterize reactive soil properties for the
Shrink Swell Index in the AS2870 equation for the determination of the
characteristic surface movement,. For this reason, it was important to ensure that
reasonable sample populations of data could be gathered for data sets/pairs

including the Shrink Swell Index as the primary parameter.
The limitation of this approach, however, was that the Australian standard test
method for the Shrink Swell Index AS1289.7.1.1 was first published in 1992.
Indeed, the inclusion of the characteristic surface movement equation in the
Residential Slabs and Footings Standard 2870 was only first introduced in 1996.
This means that prior to year of publication, the Shrink Swell test did not formally
exist and consequently, there is no data older than 1992 of relevance to this study
for Queensland or Australia-wide, for that matter. As a result, the period of time
considered for data gathering for this study was from 1992 to 2013.
7.2 Correlation of Individual Parameters and Shrink Swell

A discussion of the correlation of each individual parameter with the shrink swell
index is presented in following sections of the report. The criteria adopted for this
study for assessing whether a particular relationship would be sufficiently reliable
to enable the substitution of the shrink swell index with a paired parameter, are
based on guidelines given by Crewson (2006), as discussed in Section 5.4.
Specifically, where the coefficient of correlation (r) > 0.7 and the coefficient of
determination of (R2) > 0.5 for any particular correlation, the relationship is rated
of sufficient reliability to be used in the estimation of the characteristic surface
movement (ys).
7.2.1 Correlation of Shrink Swell Index and % Passing 0.075mm

The test data used in for the regression analysis are presented in Appendix B. The
data is shown plotted in Figure 21 with a regression curve fitted to the data sets.
34
The size of the sample population for the data is N=67, which appears reasonable
to enable the regression to have validity.
The resulting trendline is a linear function with r = 0.543 and R2 = 0.2557. Using
the criteria for correlation in Table 3, it can be concluded that the data set shows
the relationship between the shrink swell index and the percentage finer than
0.075mm is of low to moderate strength. As a result, the percentage finer than
0.075mm would not be a reliable parameter to use for the estimation of the
characteristic surface movement (ys).
Figure 21 - Linear regression plot of % passing 0.075mm and shrink swell index
7.2.2 Correlation of Shrink Swell Index and % Passing 0.002mm

The size of the sample population for the data is N=14, which appears too small to
enable the regression to have validity. The following analysis must be viewed in
the context of the small sample population.
35
the relationship between the shrink swell index and the percentage finer than
0.002mm is of low strength. As a result, based on this data, the percentage finer
than 0.002mm would not be a reliable parameter to use for the estimation of the
characteristic surface movement (ys). However, lack of data available for this
correlation makes the outcome inconclusive.
Figure 22 - Linear regression plot of % passing 0.002mm and shrink swell index
7.2.3 Correlation of Shrink Swell Index and Liquid Limit

The size of the sample population for the data is N=69, which appears sufficient to
enable the regression to have validity.
the relationship between the shrink swell index and the plasticity index is of high
strength. As a result, based on this data, the liquid limit would be a reliable
parameter to use for the estimation of the characteristic surface movement (ys).
36
By using the linear relationship of the line of best fit below, the shrink swell
index, can be estimated, and this estimated value can be used for the calculation of
the characteristic surface movement, using the AS2870 equation as presented in
Section 5.1.
Iss=0.0745(LL)-1.611
Figure 23 - Linear regression plot of liquid limit and shrink swell index
7.2.4 Correlation of Shrink Swell Index and Plasticity Index

strength. As a result, based on this data, the plasticity index would be a reliable
parameter to use for the estimation of the characteristic surface movement (ys).
37
Section 5.1.
I ss = 0.0882( PI ) − 2.637
Figure 24 - Linear regression plot of plasticity index and shrink swell index
7.2.5 Correlation of Shrink Swell Index and Linear Shrinkage

the relationship between the shrink swell index and the plasticity index is of
moderate strength. As a result, based on this data, the plasticity index would not
be a reliable parameter to use for the estimation of the characteristic surface
movement (ys).
38
Figure 25 - Linear regression plot of linear shrinkage and shrink swell index
7.2.6 Correlation of Shrink Swell Index and Weighted Plasticity Index

strength. As a result, based on this data, the weighted plasticity index would be a
reliable parameter to use for the estimation of the characteristic surface movement
(ys). By using the linear relationship of the line of best fit below, the shrink swell
Section 5.1.
I ss = 0.0009(WPI ) − 1.274
39
Figure 26 - Linear regression plot of weighted plasticity index and shrink swell index
7.2.7 Correlation of Shrink Swell Index and Weighted Linear Shrinkage

The test data used in the regression analysis is presented in Appendix B. The data
is shown plotted in Figure 27 with a regression curve fitted to the data sets. The
size of the sample population for the data is N=67, which appears sufficient to
the relationship between the shrink swell index and the plasticity index is of
moderate strength. As a result, based on this data, the weighted plasticity index
would not be a reliable parameter to use for the estimation of the characteristic
surface movement (ys).
7.2.8 Correlation of Shrink Swell Index and CBR Swell

identify any trend but may not be sufficient to be conclusive in defining the
relationship. The following analysis must be viewed in the context of the
relatively small sample population.
40
Figure 27 - Linear regression plot of weighted linear shrinkage plot and shrink swell index
In addition, the swelling measurements from tests where the soaked period was 10
days were included in the analysis with the data from tests where the soaking
period was 4 days.
The resulting trendline is a linear function with r = 03821 and R2 = 0.607. Using
the relationship between the shrink swell index and the CBR Swell is of high
strength. As a result, based on this data, the CBR Swell would be a reliable
parameter to use for the estimation of the characteristic surface movement (ys). By
using the linear relationship of the line of best fit below, the shrink swell index,
can be estimated, and this estimated value can be used for the calculation of the
characteristic surface movement, using the AS2870 equation as presented in
Section 5.1.
I ss = 0.3321(CBRSwell ) − 2.382
However, it is recommended that a larger sample population be used to confirm

this correlation.
41
^
Z
&
^

/
^
^
Z^
Figure 28 - Linear regression plot of CBR swell and shrink swell index
7.2.9 Correlation of Shrink Swell Index and Soil Suction

data is shown plotted in Figure 29. The size of the sample population for the data
is N=8, which is too small to enable the regression to have validity. As a result
regression analysis was performed on this set of data. In order to assess the
relationship between this data set, a larger sample population is required. The
outcome for this data set is inconclusive.
42
Figure 29 - Linear regression plot of soil suction and shrink swell index
7.2.10 Correlation of Shrink Swell Index and Swelling Pressure

The size of the sample population for the data is N=19, which appears too small to
enable the regression to have validity. The following analysis must be viewed in
the context of the small sample population.
the relationship between the shrink swell index and the swelling pressure is of low
to moderate strength. As a result, based on this data, the swelling pressure would
not be a reliable parameter to use for the estimation of the characteristic surface
movement (ys). However, lack of data available for this correlation makes the
outcome inconclusive.
Figure 30 - Linear regression plot of swelling pressure and shrink swell index
43
7.2.11 Correlation between Shrink Swell Index and CEC
This relationship was unable to be investigated due to the very low data sets
available from the historical Main Roads records. Only four valid data sets were
able to be considered for analysis – clearly this sample population is insufficient.
As a result the regression analysis between shrink swell index and cation
exchange capacity (CEC) was not carried out, and is not presented in this report.
7.3 Correlation of Combined Parameter Functions and Shrink

Swell Index
In the analysis above, only relationships between individual pairs of parameters
were examined for potential correlation. However, it is also acknowledged that
statistically valid relationships might also exist between combinations of
parameters, as related by simple functions.
To explore this proposal in more detail, consideration was given to the method in
which shrink swell index is calculated in AS1289.7.1.1.
As detailed in Section 4.1, the shrink swell index parameter is a summation

between the strain measured in shrinkage and half of the swelling of the soil
specimen, divided by a constant. Fundamentally, the shrink swell index is a
function of the shrinkage strain and the swelling strain of the soil.
Linear Shrinkage and CBR Swell and Shrink Swell Index

Of the other parameters considered in this study, the linear shrinkage and the CBR
Swell are direct measures of shrinkage strain and swelling strain respectively,
despite the obvious differences in the methodology used to obtain these
parameters.
On this basis, it was considered useful to investigate whether any relationship

existed between the shrink swell index and some mathematical function between
the linear shrinkage and the CBR Swell parameters.
44
7.3.1 Correlation of Shrink Swell Index and (LS + CBR Swell)
In an attempt to model the shrink swell index equation, it was decided to at first
consider a correlation between the simple summation of the linear shrinkage and
the CBR Swell values, and the shrink swell index. The results of this are shown in
Figure 31 below, and the values are tabulated in Appendix B.

&
Z

/
^
^
>^ Z^
Figure 31 - Linear regression plot of linear shrinkage + CBR Swell and shrink swell index
It should be noted that the correlation is based on a modest sample population (ie
N = 27). However, taking this into account, the correlation between the sum of
linear shrinkage and CBR swell was assessed to be of high strength (r = 0.798, R2
= 0.637).
As a result, based on this data, the combination of the linear shrinkage added to
the CBR Swell would be a reliable parameter to use for the estimation of the
characteristic surface movement (ys). By using the linear relationship of the line of
best fit below, the shrink swell index, can be estimated, and this estimated value
can be used for the calculation of the characteristic surface movement, using the
AS2870 equation as presented in Section 5.1.
I ss = 0.216( LS + CBRSwell ) − 1.202
However, it is recommended that a larger sample population be used to confirm

this correlation.
45
7.3.2 Correlation of Shrink Swell Index and “Combined Reactivity Index”
Another combination of parameters was investigated for correlation with shrink
swell index. The parameters of linear shrinkage and CBR Swell, measuring
shrinkage and swelling strain were used in an arrangement similar to the equation
used for the calculation of the shrink swell index. This new quantity postulated by
the author, is defined by the equation below and is provisionally described by the
author as the Combined Reactivity Index (CRI):
LS + (CBRSwellx500)
CRI =
1.8
The size of the sample population for the data is N=25. Despite the modest size of
the data set, it appears sufficient to identify trend. However, the population size
may not be sufficient to conclusively define the relationship. Hence following
analysis must be viewed in this context.
The correlation between the CRI and the shrink swell index was assessed to be of
high strength (r = 0.887, R2 = 0.787). This relationship was observed to show the
best correlation of all the parameters examined in this study.
As a result, based on this data, the CRI would be a reliable parameter to use for
the estimation of the characteristic surface movement (ys). By using the linear
relationship of the line of best fit below, the shrink swell index, can be estimated,
and this estimated value can be used for the calculation of the characteristic
surface movement, using the AS2870 equation as presented in Section 5.1.
I ss = 0.0017(CRI ) − 0.3081
The values used in the analysis are tabulated in Appendix B.
46

&
Z

^
^
Z /
Figure 32 - Linear regression plot of CRI and shrink swell index
7.3.3 Correlation of Shrink Swell Index and (LS/CBR Swell) Ratio

A third combination of parameters was investigated for correlation with shrink
swell index. The CBR Swell and the linear shrinkage were combined in a ratio;
specifically CBR Swell/linear shrinkage.
The ratio of the CBR Swell to the Linear Shrinkage was calculated for each set of
data in which there was a corresponding shrink swell index test result. The linear
regression of the CBR Swell/LS ratio and the shrink swell index is shown in
Figure 33, and the values are tabulated in Appendix B.
Figure 33 - Linear regression plot of ratio of CBR swell/LS and shrink swell index
47
Using Table 3, it can be seen that despite a relatively modest sample population
(N=27), the parameters exhibit a high strength relationship (r=0.777, R2=0.603).
Thus as the ratio of CBR Swell and linear shrinkage correlates well with shrink
swell index for soils over a range of locations, it can be can be cautiously
classified as a useful parameter for the purposes of estimation.
Section 5.1.
I ss = 10.675(CBRSwell / LS ) + 0.684
7.4 Summary of Correlation Analysis

Table 5 below summarises the results of the analysis including the factor and the
constants of the regression lines. For the purposes of using the regression equations to
calculate an estimation of the shrink swell index, only the parameters in which a high
strength relationship was determined should be considered.
8 Conclusions
8.1 Major Outcomes and Key Findings

This report has introduced the main properties of expansive soils and highlighted
some of the risks these materials pose to the serviceability and safety of road
assets. It has provided an overview of the main approach for the design of
structures on reactive soil sites (AS2870) within Queensland (and Australia) and
the primary method for determining the magnitude of the potential surface
movements likely to be encountered for such sites.
48
No. of
Data
Coefficient Description
Pairs Coefficient of
of of
Parameter (N) Determination Factor Constant
Correlation 2 Relationship
Corrected (R )
(r) Strength
for
Outliers)
Moderate-
<0.075mm 67 0.543 0.256 0.0430 0
Low
<0.002mm 14 0.417 0.169 Low 0.0729 0
LL 69 0.791 0.626 High 0.0745 -1.611
PI 71 0.719 0.517 High 0.0882 -2.637
WPI 67 0.705 0.523 High 0.0009 -1.274
LS 71 0.662 0.438 Moderate 0.2791 -1.496
WLS 67 0.679 0.520 Moderate 0.0031 0.151
CBR Swell 26 0.821 0.607 High 0.3321 2.382
Suction 8 NA NA NA NA NA
Loaded Swell/Swell Moderate-
19 0.583 0.340 0.0053 1.926
Pressure Low
CEC 4 NA NA NA NA NA
LS + CBR Swell 27 0.798 0.637 High 0.216 1.202
LS + (CBRSwellx500)
25 0.887 0.787 High 0.0017 -0.3081
1.8
CBR Swell/LS 27 0.777 0.603 High 10.675 0.684
Table 5 - Summary of correlation analysis of all parameters with shrink swell index
This study has also identified some of the laboratory test parameters used to
identify and characterise soils with shrink/swell potential. Relationships between
these parameters and the shrink swell index have been investigated to determine if
alternatives to the shrink swell could potentially be used in the calculation of
characteristic surface movement (ys).
Test data gathered from site investigations throughout Queensland have been
statistically analyzed to determine the relationships that exist between the various
parameters and shrink swell index.
49
Of the relationships identified, it was found that the following parameters,
determined by common laboratory tests, showed a high strength of correlation
with the Shrink Swell Index:
• Liquid limit
• Plasticity Index
• Weighted Plasticity Index
• CBR Swell
It was also found that unique combinations of these standard test parameters
exhibited high strength of correlation. Specifically:
• Linear Shrinkage + CBR Swell
• “Combined Reactivity Index”
• CBR Swell/Linear Shrinkage
It is the conclusion of this report that, based on the strength of the correlations,
it would seem a reasonable approach for a designer or engineer to substitute
these parameters for the Shrink Swell Index (and by association the Instability
Index), in the calculation of the characteristic surface movement (ys), as per
AS2870 – Residential Slabs and Footings.
Some caution would be required in using these parameters, however, as the

sample populations on which the correlations are based contain modest to small
numbers of data.
8.2 Recommendations for Further Work

The source of data for this investigation was limited to the historical records of
the Department of Transport and Main Roads, from 1992 to 2013. Despite the
number of published reports available, and the large numbers of reactive soil sites
investigated throughout Queensland over this period, the amount of laboratory
test data available from this activity was found to be limited.
50
This was particularly the case for the less common laboratory tests considered,
such as soil suction and cation exchange capacity.
To investigate these relationships in more detail, other sources of data need to be

considered, such as local government authorities, commercial construction
materials laboratories, and consultant engineering firms within Queensland.
51
9 List of References
Al-Omari, R.R and Oraibi, W.K. (2000), “Cyclic Behavior of Reinforced Expansive
Clay”, Soils and Foundations Vol. 40, No. 2, Japanese Geotechnical Society.
ASTM International (2003), Standard Test Method for One Dimensional Swell or
Settlement Potential of Cohesive Soils, D4546 – 03, 100 Barr Harbor Drive, PO Box
C700, West Conshohocken, PA 19428-2959, United States.
Bulut R., Lytton R.L. and Wray W.K. (2001), Expansive clay soils and vegetative
influence on shallow foundations, Proceedings of Geo-Institute Shallow Foundation
and Soil Properties, Committee Sessions at the ASCE 2001 Civil Engineering
Conference
Casagrande, A. (1958), Notes of the design of the liquid limit device, Geotechnique,
vol. 8, No. 2, pp 84-91.
Cameron D.A. (1989), Tests for Reactivity and Prediction of Ground Movement,
Australian Civil Engineering Transactions, pp121-131
Caunce C.M. (2010) – Effective Road Pavement Design for Expansive Soils in
Ipswich, University of Southern Queensland
Chen, F. H. (1988), Foundations on Expansive Soils, Elsevier Science Publishers
Clarke G.M. and Cooke D. (1998), A basic course in statistics. Arnold London, 4th
edition
Crewson P. (2006), Applied Statistics Handbook, AcaStat Software
Crilly, M. and Chandler, R., (1993); A method of determining the state of desiccation
in clay soils, BRE Information Paper, February 1993
52
Department of Transport and Main Roads, Queensland, (2011), Technical Report
MR2469, East Warianna Creek Culvert Culvert Foundation Expansive Soil
Investigation
Department of Transport and Main Roads, Queensland, (2011), Technical Report

MR2470, Auchmar Culvert Culvert Foundation Expansive Soil Investigation
Department of Transport and Main Roads, Queensland, (2012), Materials Testing

Manual: List of Test Methods, Queensland Volume 1
Earl D. (2005), To determine if there is a correlation between the shrink swell index
and Atterberg limits for soils within the Shepparton Formation, University of
Southern Queensland
Fox E., 2000, A climate based design depth of moisture change map of Queensland
and the use of such maps to classify sites under AS 2870-1996, Australian
Geomechanics, December 2000
Fredlund D.G., Rahardjo H., Fredlund M.D., (2012), Unsaturated Soil Mechanics in
Engineering Practice, Wiley & Sons
Fityus, S.G., Cameron, D.A., and Walsh, P.F., (2009), The Shrink Swell Test,
Geotechnical Testing Journal, Vol. 28, No. 1.
Gallagher A.G., (1991), Developments in the Measurement of Soil Suction,

Queensland University of Technology.
Gourley GS, Newill D and Schreiner H.D, (1993), Expansive soils: Research
Strategy, First International Symposium on Engineering Characteristics of Arid Soils,
City University, London
53
Head, K. H., (1980), Manual of Soil Laboratory Testing, Volume 1: Soil
Classification and Compaction Tests, Pentech Press, London.
Head, K. H., (1982), Manual of Soil Laboratory Testing, Volume 2: Permeability,

Shear Strength and Compressibility Tests, Pentech Press, London.
Lopez, D and McManus, KJ. (1996), Soil Index Methods to Predict Ground
Movements, 7th Australia New Zealand Conference on Geomechanics:
Geomechanics in a Changing World: Conference Proceedings. Barton, ACT:
Institution of Engineers, Australia, 1996: 921-924. National conference publication
(Institution of Engineers, Australia) ; no. 96/07.
Lucian, C. (2012), Soil-Suction Measurements Using the Filter Paper Method to

Evaluate Swelling Potential, International Journal of Scientific Engineering and
Technology, Volume No.1, Issue No.4, pg : 49-54
Nelson, J. and Miller, D., (1992), Expansive Soils: Problems and Practice in
Foundation and Pavement Engineering, 1992, Wiley & Sons
Nelson, J.D, Overton, D.D. and Durkee, D.B., (2001), Depth of Wetting and the
Active Zone, pp95-109, Proceedings from ASCE Conference (2001)
Or, D., M. Tuller and J.M. Wraith, (2005), Soil Water Potential, In: Encyclopedia of
Soils in the Environment, (ed. Hillel D.), Elsevier Science, Oxford, 270-277
Pritchard, R. and Wijeyakulasurily C. V.(2000), Drainage Structures on Expansive

Soils in Western Queensland, Technical Note WQ37, Queensland Department of
Main Roads(2000)
Punthutaecha, K and Puppala, J; Vanapalli,S and Inyang, H (2006), Volume Change

Behaviors of Expansive Soils with Recycled Ashes and Fibers, Journal of Materials in
Civil Engineering
54
Queensland Department of Main Roads, (1993), Technical Note 7 – Controlling
Moisture in Pavements, Transport Technology
Queensland Department of Main Roads, (2002), Technical Report CR667,

Gordonstone Creek Crossings Foundation Investigations
Queensland Department of Main Roads, (1999), Technical Report R3177, Bruce

Highway Upgrade – Yandina to Cooroy (Cut J) – Geotechnical Investigation
Queensland Department of Main Roads, (1999), Technical Report MR1780,

Thompson River Project, Culvert Site Classification
Queensland Department of Main Roads, (1999), Technical Report MR1795, Deep

Creek and D’Aigular Highway Expansive Clay Testing

Hughendon – Winton Road Culvert Bases – Expansive Property Results
Queensland Department of Main Roads, (2000), Technical Report MR1846, Wondalli

Creek Base Slab – 122/17D/30
Queensland Department of Main Roads, (2000), Technical Report MR1892, Culvert

Base Slab on Expansive Clay – Yaralla Deviation

Foundation Investigation – Doctor’s Creek

Foundation Investigation – Lagoon Creek

Foundation Investigation – Barkly Highway
55
Macalister Culvert – Job Number 124/18C/16

Glenorchy Straight – Additional Geotechnical Investigation

Chippendale Creek Culvert Investigation
Queensland Department of Main Roads, (2004), Technical Report MR2140, Warrego

Highway (Bowenville to Dalby) Culvert Investigation
Queensland Department of Main Roads, (2008), Technical Report MR2326, Site

Investigation of Cracking at Cockatoo Creek Culverts
Queensland Department of Main Roads, (2008), Technical Report MR2334, Deep

Creek Channel – Culvert Investigation
Queensland Department of Main Roads, (2009), Technical Report MR2366, Bribie

Island Road – Expansive Subgrade Investigation
Queensland Department of Main Roads, (2009), Technical Report MR2367, Exit 54

Pacific (M1) Motorway at Coomera Supplementary Investigation
Queensland Department of Main Roads, (1999), Technical Report R3162, Bruce

Highway Upgrade (Brisbane to Gympie) Additional Geotechnical Investigation
Queensland Department of Main Roads, (2002), Technical Report R3263, Gatton

Bypass Package 2 (CH30000 – 39400) Job No. 114/18A/54.2
Queensland Department of Main Roads, (2006), Technical Report R3379, Mackay –

Bucasia Road between Holts Road and Golflinks/Habana Roads (Ch.2.6km to
Ch.3.3km) - Job No. 120/856/12
56
Queensland Department of Main Roads, (2006), Technical Report R3380, Final
Report: Geotechnical Investigation for Linkfield Rd Overpass Duplication, Volume 1:
Current Investigation
Queensland Department of Main Roads, (2007), Technical Report R3403,

Geotechnical Investigation at Tamborine-Oxenford Deviation Link Road
Queensland Department of Main Roads, (2008), Technical Report R3419, Warrego

Highway (Macalister-Warra) Culverts: Expansive Subgrade Investigation

Cunningham Highway/Amberley Interchange Preliminary Geotechnical Report

Cunningham Highway/Amberley Interchange Interpretive Preliminary Geotechnical
Report
Ramanujam J (1999), Rehabilitation of Pavement on Expansive Soils, Workshop on

low volume roads, Roma Bungil Cultural Community Centre, Roma Reeves, I, 2001,
Risk Management – Materials [presented at Road System and Engineering
Technology Forum 2001],
Snethen Dr D. R (1980), Expansive Soils in Highway Subgrades, U.S Department of

transportation, United States of America Smith R.L (2004), Achieving the Goal of
Management of Reactive Clays, Australian Institute of Building Surveyors
Conference, Australia
Standards Australia, (1995), AS 1289.3.2.1-1995 Methods of testing soils for

engineering purposes Method 3.2.1: Soil classification tests- Determination of the
plastic limit of a soil – Standard method, Standards Australia, Homebush, NSW.
57
engineering purposes Method 3.3.1: Soil classification tests- Calculation of the
plasticity index of a soil, Standards Australia, Homebush, NSW.

engineering purposes Method 3.4.1: Soil classification tests - Determination of the
linear shrinkage index of a soil – Standard method,
Standards Australia, Homebush, NSW.

engineering purposes Method 7.1.1: Soil reactivity tests- Determination of the
shrinkage index of a soil – Shrink-swell index, Standards Australia, Homebush, NSW.
Standards Australia, (1996), AS 2870-1996 Residential slabs and footings -

Construction, Standards Australia, Homebush, NSW.
Vanderstaay, A. G. And Reeves, I. N. (2000), Soils of Western Queensland,

Technical Note WQ32 , Queensland Department of Main Roads (2000)
Walters K. A., (2008), Investigation of Construction Practices and Test Procedures for
Road Pavements on Expansive Subgrades Volume I, University of Southern
Queensland
Waters T (2002), Moisture in Pavements, Road System and Engineering Technology,

Department of Main Roads
Watford Day, R. W. (1999), Geotechnical and foundation engineering design and

construction, McGraw-Hill Companies, New York.
Yanagisawa E, Moroto, N and Mitachi, T (1998), Problematic Soils, Vol 1,

Proceedings of the international symposium on problematic soils
58
Yang H P, Zheng J and Zhang R (2007), Addressing Expansive Soils, Institute of
Special Soils, pg 64-69
59
Appendix A Project Specification
60
For Peter Reynolds
Topic Engineering Correlations for the Characterisation of Reactive Soil
Behaviour For Use in Road Design
Supervisors Dr Kazem Ghabraie
Enrolment ENG 4111 – S1, X, 2013
ENG 4112 – S2, X, 2013
Project Aim To identify and quantify the relationships that exist between key
measures of reactive soil behaviour.
PROGRAMME Issue A -13th March 2013
1. Research the background information relating to the
properties, identification and classification of reactive soils
as they relate to road design and construction
2. Identify sources of data containing published site
classification information for road infrastructure projects
within Queensland
3. Obtain and extract relevant data from published sources
and collate based on locality and geological formation,
where possible.
4. Investigate and analyse soil property parameters for any
significant trends or correlations.
5. Determine whether there is any specific parameters which
are more reliable indicators of reactive soil behaviour and
identify any limitations to the use of these parameters
6. Rank soil property parameters in order of reliability and
how economical they are to determine
7. Develop a guideline for the investigation and identification
of reactive soil, including a table of correlations of soil
parameters
61
Appendix B Tabulated Test Data
62
Tabulated Data for Correlation of % passing 0.075mm sieve and Shrink Swell Index
^ E / W&
GS00-447 0-1.3 82 4.2
GS00-444 1.3-2.1 92 5
BS00-1291/GS00-504 0-0.5 67 0.2
BS00-1292/GS00-505 0.5-1.0 62 0.3
BS00-1295/GS00-511 0.5-1.0 71 3
Performed by consultants 0.25-.7 62 1.7
Performed by consultants 0.7-2.1 16 0.8
Performed by consultants 0-1.2 77 0.4
BS99-0757 0.7-2.1 84 4.5
GS00-456, 00/1044 0.2-0.5 58 2.9
GS00-457, 00/1045 0.5-0.8 54 3.7
GS00-458, 00/1046 0.8-1.3 62 2.7
GS00-454, 00/1042 0.3-0.6 63 1.6
GS00-455, 00/1043 1.7-2.1 77 1.9
GS00-442, 00/964 0-0.4 90 6.1
GS00-443, 00/965 0.4-1.4 88 5.8
GS00-495,00/1075 0.3-0.6 73 1
GS00-496,00/1076 1.7-2.0 89 1
GS00-497,00/1077 0.25-0.65 87 3.7
GS00-498,00/1078 0.65-1.1 86 4.5
GS00-499,00/1079 1.7-2.0 86 4.7
GS00-500,00/1080 0.25-0.55 61 0.1
GS00-501,00/1081 0.7-1.0 89 3.8
GS00-502,00/1082 1.4-1.6 93 1.9
GS99/207.F/99-207 NA 86 1.7
GS99/208.F NA 90 1.4
GS01/395 0-1.7 43 1
GS01/396 3.0-3.3 44 2.2
GS01/397 1.0-1.45 41 1.4
GSO4/418 0.5 - 0.8 63 3.9
GSO4/420 1.7 - 2.0 52 2.6
Performed by consultants 2.0 -2.4 85 5.8
63
Performed by consultants 4.0 - 4.4 51 4.7
Performed by consultants 0.0 - 2.5 98 7.4
Performed by consultants NA 98 7.3
GS04-418 0.5-0.8 63 3.9
GS08-434 1.5 94 5.1
GS08-438 0.5 82 2.1
GS11-1353/CLON110772 0-1.3 59 1.4
GS11-1354/CLON110773 1.3-2.5 51 2
GS11-1471/11110313 2.5-2.95 58 2.7
GS99-207 NA 90 1.7
GS99-208 NA 90 1.4
GS06-061/062 0.5-0.9 98 7.4
603224/GS06-089/GS06-100 0-0.5 83 3.7
603228/GS06-093/GS06-102 1 99 5.6
603552/GS06-0110 1.0-1.4 91 5.1
603554/GS06-0113 1.5-1.9 91 7.4
GS07-525/526 1.5-2.9 84 1.2
GS07-528/529 3.0-3.9 73 4.5
GS07-536 0.5-0.9 78 3.5
GS10-072/129 0.3-3.1 98 4.9
GS10-078/079 0.2-0.9 83 4.8
GS01-212RA NA 82 1
GS01-212RB NA 82 1.4
GS01-211RA NA 80 1.7
GS01-211RB NA 80 2.4
GS06-809 3-3.45 100 6
GS06-789 1.0-1.45 88 5.5
GS06-793/809 2.5-2.9 100 9
GS06-798/836/837 1.0-1.45 99 2.7
GS10-129/072 0.3-3.1 98 4.9
GS10-078/79 0.2-0.9 83 4.8
64
Tabulated Data Used for Correlation between
% Finer than 0.002mm and Shrink Swell Index
^ E / W&
GS99/207.F/99-207 NA 49 1.7
GS99/208.F NA 50 1.4
GS04-418 0.5-0.8 45 3.9
GS04-420 1.7-2.0 43 2.6
603224/GS06-089/GS06-100 0-0.5 51 3.7
603228/GS06-093/GS06-102 1.0 77 5.6
603554/GS06-0113 1.5-1.9 76 7.4
GS07-525/526 1.5-2.9 64 1.2
GS07-528/529 3.0-3.9 63 4.5
GS07-536 0.5-0.9 60 3.5
GS06-809 3-3.45 61 6
GS06-789 1.0-1.45 46 5.5
GS06-793/809 2.5-2.9 61 9
GS06-798/836/837 1.0-1.45 65 2.7
65
Tabulated Data for Correlation between Liquid Limit and Shrink Swell Index
^ E >> / W&
GS00-097.B 1.5-1.9 27 1.6
Performed by consultants 0-1.2 28.2 0.4
BS00-1292/GS00-505 0.5-1.0 34.4 0.3
GS00-096.B 1.0-1.4 35.4 1
GS00-094.B 0-0.4 35.4 3
GS00-457, 00/1045 0.5-0.8 35.8 3.7
GS00-500,00/1080 0.25-0.55 36.6 0.1
GS00-454, 00/1042 0.3-0.6 42 1.6
BS00-1291/GS00-504 0-0.5 43 0.2
GS01/395 0-1.7 43.2 1
GS08-438 0.5 44 2.1
BS00-1295/GS00-511 0.5-1.0 44 3
Performed by consultants 0.7-2.1 45 0.8
Performed by consultants 0.25-.7 45.6 1.7
GS00-456, 00/1044 0.2-0.5 46.4 2.9
GS01/397 1.0-1.45 48 1.4
GSO4/418 0.5 - 0.8 48 3.9
GS04-418 0.5-0.8 48 3.9
GS00-455, 00/1043 1.7-2.1 48.6 1.9
GS00-502,00/1082 1.4-1.6 49 1.9
GS11-1471/11110313 2.5-2.95 49.6 2.7
GS99/208.F NA 50.8 1.4
GS99-208 NA 50.8 1.4
GS00-458, 00/1046 0.8-1.3 52 2.7
GS00-495,00/1075 0.3-0.6 53.8 1
GS07-630 0.9 58 4.8
GS00-496,00/1076 1.7-2.0 58.8 1
GS99/207.F/99-207 NA 59 1.7
GS99-207 NA 59 1.7
GS01-212RA NA 60.8 1
GS01-212RB NA 60.8 1.4
GS04-420 1.7-2.0 61.6 2.6
GS00-501,00/1081 0.7-1.0 61.8 3.8
GSO4/420 1.7 - 2.0 62 2.6
GS11-1353/CLON110772 0-1.3 63.4 1.4
GS11-1354/CLON110773 1.3-2.5 64.6 2
GS07-536 0.5-0.9 67 3.5
GS01/396 3.0-3.3 70.2 2.2
603224/GS06-089/GS06-
100 0-0.5 71 3.7
BS99-0757 0.7-2.1 71.4 4.5
Performed by consultants 4.0 - 4.4 71.8 4.7
GS01-211RA NA 72.2 1.7
GS01-211RB NA 72.2 2.4
66
GS00-499,00/1079 1.7-2.0 74.6 4.7
GS00-447 0-1.3 74.8 4.2
GS08-434 1.5 76 5.1
GS06-789 1.0-1.45 77 5.5
GS00-498,00/1078 0.65-1.1 77.4 4.5
GS00-444 1.3-2.1 78.8 5
GS00-442, 00/964 0-0.4 79.6 6.1
GS10-078/079 0.2-0.9 80 4.8
GS10-078/79 0.2-0.9 80 4.8
Performed by consultants 2.0 -2.4 80.2 5.8
GS00-443, 00/965 0.4-1.4 81 5.8
GS00-497,00/1077 0.25-0.65 81.6 3.7
GS07-528/529 3.0-3.9 83 4.5
GS07-538/539 1.5-1.9 84 2.6
603552/GS06-0110 1.0-1.4 88 5.1
GS06-798/836/837 1.0-1.45 90 2.7
GS10-072/129 0.3-3.1 95 4.9
GS10-129/072 0.3-3.1 95 4.9
603554/GS06-0113 1.5-1.9 100 7.4
GS06-809 3-3.45 105 6
GS06-793/809 2.5-2.9 105 9
Performed by consultants NA 107 5.6
67
Tabulated Data for Correlation between Plasticity Index, Linear Shrinkage
and Shrink Swell Index
Sample/Article Number Depth (m) PI (%) LS (%) Iss %/PF
GS00-447 0-1.3 45.8 18 4.2
GS00-444 1.3-2.1 43.8 19.2 5
GS00-094.B 0-0.4 19.8 12.8 3
GS00-096.B 1.0-1.4 19.8 12.8 1
BS00-1291/GS00-504 0-0.5 29.4 14.2 0.2
BS00-1292/GS00-505 0.5-1.0 21.2 9.8 0.3
BS00-1295/GS00-511 0.5-1.0 31.1 12.6 3
Performed by consultants 0.25-.7 26.6 15.8 1.7
Performed by consultants 0.7-2.1 18 10 0.8
Performed by consultants 0-1.2 14.2 8.6 0.4
Performed by consultants 0-2.0 15.4 9 0.6
BS99-0757 0.7-2.1 49.2 20.4 4.5
GS00-456, 00/1044 0.2-0.5 26.6 15 2.9
GS00-457, 00/1045 0.5-0.8 19.2 10.6 3.7
GS00-458, 00/1046 0.8-1.3 32.8 14.8 2.7
GS00-454, 00/1042 0.3-0.6 23 13.2 1.6
GS00-455, 00/1043 1.7-2.1 28.6 15.4 1.9
GS00-442, 00/964 0-0.4 44.8 20 6.1
GS00-443, 00/965 0.4-1.4 49.4 18.8 5.8
GS00-495,00/1075 0.3-0.6 24 13.2 1
GS00-496,00/1076 1.7-2.0 28 13.4 1
GS00-497,00/1077 0.25-0.65 43 16.2 3.7
GS00-498,00/1078 0.65-1.1 48 19 4.5
GS00-499,00/1079 1.7-2.0 41.2 19 4.7
GS00-500,00/1080 0.25-0.55 18.4 10.4 0.1
GS00-501,00/1081 0.7-1.0 33 13.2 3.8
GS00-502,00/1082 1.4-1.6 19.2 9.6 1.9
GS99/207.F/99-207 NA 30.6 15 1.7
GS99/208.F NA 23.8 14.2 1.4
Performed by consultants NA 64 25.2 5.6
GS01/395 0-1.7 25.2 12.2 1
GS01/396 3.0-3.3 46.4 15.8 2.2
68
GS01/397 1.0-1.45 31 13.2 1.4
GSO4/418 0.5 - 0.8 28.8 16.8 3.9
GSO4/420 1.7 - 2.0 37.6 19.4 2.6
Performed by consultants 2.0 -2.4 45.0 20.2 5.8
Performed by consultants 4.0 - 4.4 36.0 16.2 4.7
GS04-418 0.5-0.8 28 16.8 3.9
GS04-420 1.7-2.0 37.6 19.4 2.6
GS08-434 1.5 46 16 5.1
GS08-438 0.5 24 11.5 2.1
GS11-1353/CLON110772 0-1.3 31.4 16.2 1.4
GS11-1354/CLON110773 1.3-2.5 34.6 16.4 2
GS11-1471/11110313 2.5-2.95 23.6 6.6 2.7
GS99-207 NA 30.6 15 1.7
GS99-208 NA 23.8 14.2 1.4
GS06-061/062 0.5-0.9 40.4 17.6 7.4
603224/GS06-089/GS06-100 0-0.5 36 21 3.7
603228/GS06-093/GS06-102 1 89 28 5.6
603552/GS06-0110 1.0-1.4 52 21.5 5.1
603554/GS06-0113 1.5-1.9 70 27 7.4
GS07-525/526 1.5-2.9 65 25 1.2
GS07-528/529 3.0-3.9 58 23.5 4.5
GS07-536 0.5-0.9 39 19 3.5
GS07-538/539 1.5-1.9 52 21.5 2.6
GS10-072/129 0.3-3.1 70 23 4.9
GS10-078/079 0.2-0.9 56 22 4.8
GS01-212RA NA 32.4 16.4 1
GS01-212RB NA 32.4 16.4 1.4
GS01-211RA NA 42.8 17.2 1.7
GS01-211RB NA 42.8 17.2 2.4
GS07-630 0.9 36.8 15 4.8
GS06-809 3-3.45 69 20 6
GS06-789 1.0-1.45 53 18.5 5.5
GS06-793/809 2.5-2.9 69 20 9
GS06-798/836/837 1.0-1.45 57 23 2.7
GS10-129/072 0.3-3.1 70 23 4.9
GS10-078/79 0.2-0.9 56 22 4.8
69
Tabulated Data for the Correlation of Weighted Plasticity Index, Weighted Linear
Shrinkage and Shrink Swell Index
^ E t>^ tW/ / W&
GS00-447 0-1.3 680 1196 0.1
GS00-444 1.3-2.1 1416 2931 0.2
BS00-1291/GS00-504 0-0.5 979 2118 0.3
BS00-1292/GS00-505 0.5-1.0 851 1406 0.4
BS00-1295/GS00-511 0.5-1.0 792 1355 0.6
Performed by consultants 0.25-.7 765 1388 0.7
Performed by consultants 0.7-2.1 300 540 0.8
Performed by consultants 0-1.2 664 1118 0.9
Performed by consultants 0-1.2 1022 1858 1
BS99-0757 0.7-2.1 1443.2 2851.2 1
GS00-457, 00/1045 0.5-0.8 1323 2218 1.4
GS00-458, 00/1046 0.8-1.3 1102 2588 1.4
GS00-454, 00/1042 0.3-0.6 993 1925 1.4
GS00-455, 00/1043 1.7-2.1 1323 2218 1.4
GS00-442, 00/964 0-0.4 1443.2 2851.2 1.4
GS00-443, 00/965 0.4-1.4 1292 2252 1.6
GS00-495,00/1075 0.3-0.6 1169 1969 1.7
GS00-496,00/1076 1.7-2.0 1338 2730 1.7
GS00-497,00/1077 0.25-0.65 1338 2730 1.7
GS00-498,00/1078 0.65-1.1 1462 3638 1.7
GS00-499,00/1079 1.7-2.0 1525 2831 1.9
GS00-500,00/1080 0.25-0.55 902 1805 1.9
GS00-501,00/1081 0.7-1.0 850 1792 2
GS00-502,00/1082 1.4-1.6 966 2016 2.1
GS99/207.F/99-207 NA 1128 3313 2.2
GS99/208.F NA 1462 3638 2.4
GS01/395 0-1.7 1707 3309 2.6
GS01/396 3.0-3.3 1707.2 3308.8 2.6
GS01/397 1.0-1.45 1372 3041 2.7
GSO4/418 0.5 - 0.8 614 2195 2.7
GSO4/420 1.7 - 2.0 2300 5700 2.7
Performed by consultants 2.0 -2.4 1256 2226 2.9
Performed by consultants 4.0 - 4.4 1259 3117 3
70
Performed by consultants 0.0 - 2.5 1672 3432 3.5
Performed by consultants NA 957 1734 3.7
GS04-418 0.5-0.8 1503 3999 3.7
GS04-420 1.7-2.0 1869 3204 3.7
GS08-434 1.5 1224 3059 3.8
GS08-438 0.5 1596 2736 3.9
GS11-1353/CLON110772 0-1.3 1562.4 2604 3.9
GS11-1354/CLON110773 1.3-2.5 1669 4246 4.2
GS11-1471/11110313 2.5-2.95 1975 4782 4.5
GS99-207 NA 1737 4387 4.5
GS99-208 NA 2350 5800 4.5
GS06-061/062 0.5-0.9 1765 3827 4.7
603224/GS06-089/GS06-100 0-0.5 1569 3486 4.7
603228/GS06-093/GS06-102 1 1980 5040 4.8
603552/GS06-0110 1.0-1.4 1980 5040 4.8
603554/GS06-0113 1.5-1.9 2277 6930 4.9
GS07-525/526 1.5-2.9 2277 6930 4.9
GS07-528/529 3.0-3.9 1910 4358 5
GS07-536 0.5-0.9 1584 4550 5.1
GS10-072/129 0.3-3.1 2043 4940 5.1
GS10-078/079 0.2-0.9 1813 5194 5.5
GS01-212RA NA 2800 8900 5.6
GS01-212RB NA 1747 4589 5.8
GS01-211RA NA 1946 4335 5.8
GS01-211RB NA 2000 6900 6
GS06-809 3-3.45 1928 4319 6.1
GS06-789 1.0-1.45 1600 3500 7.3
GS06-793/809 2.5-2.9 1600 3500 7.4
GS06-798/836/837 1.0-1.45 1760 4040 7.4
GS10-129/072 0.3-3.1 2565 6650 7.4
GS10-078/79 0.2-0.9 2000 6900 9
71
Tabulated Data for Correlation of CBR Swell and Shrink Swell Index
Z^ KD Z^ KD
^ E / W&
^ ^
GS00-447 0-1.3 6.7 4.2
GS00-444 1.3-2.1 8.8 5
BS99-0757 0.7-2.1 5.4 4.5
GS00-456, 00/1044 0.2-0.5 4.3 2.9
GS00-457, 00/1045 0.5-0.8 3.7 3.7
GS00-454, 00/1042 0.3-0.6 1.8 1.6
GS00-455, 00/1043 1.7-2.1 3.1 1.9
GS00-442, 00/964 0-0.4 7.8 6.1
GS00-443, 00/965 0.4-1.4 6.5 5.8
GS00-495,00/1075 0.3-0.6 1 1
GS00-496,00/1076 1.7-2.0 1 1
GS00-497,00/1077 0.25-0.65 6.75 3.7
GS00-498,00/1078 0.65-1.1 4.55 4.5
GS00-501,00/1081 0.7-1.0 1.2 3.8
GS00-502,00/1082 1.4-1.6 0.5 1.9
GS99/207.F/99-207 NA 3.4 1.7
GS99/208.F NA 1.6 1.4
GS08-434 1.5 4 4 5.1
GS08-438 0.5 3 3 2.1
GS99-207 NA 3.4 1.7
GS99-208 NA 1.6 1.4
GS10-072/129 0.3-3.1 7.9 7.9 4.9
GS10-078/079 0.2-0.9 7.3 7.3 4.8
GS06-798/836/837 1.0-1.45 0.6 0.6 2.7
GS10-129/072 0.3-3.1 7.9 7.9 4.9
GS10-078/79 0.2-0.9 7.3 7.3 4.8
72
Tabulated Data for the Correlation of Soil Suction and Shrink Swell Index
^ ^
^ E D & / W&
W W
GS04-418 0.5-0.8 2668 3.9
GS04-420 1.7-2.0 3354 2.6
GS07-525/526 1.5-2.9 1235 1.2
GS07-528/529 3.0-3.9 1101 4.5
GS06-809 3-3.45 77.4 6.0
GS06-789 1.0-1.45 4.0 5.5
GS06-793/809 2.5-2.9 195.9 9.0
GS06-798/836/837 1.0-1.45 195.9 2.7
Tabulated Data for the Correlation of Swelling Pressure and Shrink Swell Index
^
^ E W / W&
W
GS99/207.F/99-207 NA 109.2 1.7
GS99/208.F NA 113.9 1.4
GS99-421 NA 615 3.2
GS04-418 0.5-0.8 250 3.9
GS04-420 1.7-2.0 320 2.6
GS08-434 1.5 140 5.1
GS11-1353/CLON110772 0-1.3 52 1.4
GS11-1354/CLON110773 1.3-2.5 49 2
GS11-1471/11110313 2.5-2.95 105 2.7
GS99-207 NA 109 1.7
GS99-208 NA 114 1.4
GS07-525/526 1.5-2.9 120 1.2
GS07-528/529 3.0-3.9 230 4.5
GS07-536 0.5-0.9 120 3.5
GS07-538/539 1.5-1.9 110 2.6
GS10-072/129 0.3-3.1 450 4.9
GS10-078/079 0.2-0.9 300 4.8
GS10-129/072 0.3-3.1 450 4.9
GS10-078/79 0.2-0.9 300 4.8
73
Tabulated Data for Correlation of CBR Swell/LS Ratio and Shrink Swell Index
Z^
KD
^ E >^ Z^ >^ / W&

^
GS00-495,00/1075 13.2 1 0.08 1
GS00-496,00/1076 13.4 1 0.07 1
GS99/208.F 14.2 1.6 0.11 1.4
GS99-208 14.2 1.6 0.11 1.4
GS01-212RB 16.4 1.9 0.12 1.4
GS00-454, 00/1042 13.2 1.8 0.14 1.6
GS99/207.F/99-207 15 3.4 0.23 1.7
GS99-207 15 3.4 0.23 1.7
GS00-455, 00/1043 15.4 3.1 0.20 1.9
GS00-502,00/1082 9.6 0.5 0.05 1.9
GS08-438 11.5 3 0.26 2.1
GS01-211RB 17.2 2.1 0.12 2.4
GS00-456, 00/1044 15 4.3 0.29 2.9
GS00-457, 00/1045 10.6 3.7 0.35 3.7
GS00-497,00/1077 16.2 6.75 0.42 3.7
GS00-501,00/1081 13.2 1.2 0.09 3.8
GS00-447 18 6.7 0.37 4.2
BS99-0757 20.4 5.4 0.26 4.5
GS00-498,00/1078 19 4.55 0.24 4.5
GS10-078/079 22 7.3 0.33 4.8
GS10-078/79 22 7.3 0.33 4.8
GS10-072/129 23 7.9 0.34 4.9
GS10-129/072 23 7.9 0.34 4.9
GS00-444 19.2 8.8 0.46 5
GS08-434 16 4 0.25 5.1
GS00-443, 00/965 18.8 6.5 0.35 5.8
GS00-442, 00/964 20 7.8 0.39 6.1
74
Appendix C Project Aims and Objectives
75
The aim of this project is to identify and quantify the relationships that exist
between key measures of reactive soil behaviour. A study of the existing methods
used for identifying and characterising reactive soils within the field of road
construction, will be conducted. Historical data gathered from investigations and
testing undertaken since 2000, by the Department of Transport and Main Roads,
Queensland, and other local road authorities, will be used to carry out parametric
studies. Any linkages or correlations that may exist between the key indicators of
reactive soil behaviour will be identified and quantified for future use as guides in
the design of roads using these materials.
The objectives of this project can be summarised as follows:
• To understand the nature of expansive soils and their properties
• To identify by research, the key measures of reactive soil behaviour used by
engineers in road construction.
• To investigate and quantify any relationships that may exist between the key
measures.
• To establish a ranking of reliability for the key measures of reactive soil
behaviour.
• To develop guidelines to assist road designers in the identification and
characterisation of reactive soils during the site investigation phase of road
construction projects.
• To develop a simple model to estimate the loads and pressures required to
suppress volume change within an expansive soil foundation or embankment
76
Appendix D Project Methodology
77
Project Methodology The specific objectives as listed earlier in Appendix 2 will now be
considered in terms of how each task will be undertaken, the tasks involved in each
project and why the decision was made to carry out the work this way.
To Understand the Nature and Properties of Expansive Soils. This will be achieved by
conducting a literature review of both print and web-based sources, from Australia and
Overseas. This will ensure that a wide variety of information is obtained.
Within Australia and internationally, the subject of reactive soils, their effect on the built
environment is well documented. Along with Australia, countries such as South Africa
and North America have large zones of arid and semi arid land, where generally, these
soils are at their most troublesome.
Researching the background information on reactive soils will help to give the author an
understanding of their behaviour.
To Identify By Research The Key Measures Of Reactive Soil Behaviour. This will also
be achieved by research well by reviewing the methods that have been adopted, by road
engineers within Queensland and overseas, to identify and characterise expansive soils.
This information should provide guidance on the effectiveness of each method which will
allow the author to rank the different methods and determine which method will be the
most effective tool for use in the investigation and design processes for the road engineer.
Local shires and engineering bodies will also be investigated to determine what
investigation tools have been adopted for their site investigation activities. By identifying
78
parameters that provide the most optimum characterisation of these soils, more cost
effective testing and investigation programs can be implemented. This is a particularly
worthwhile goal in the current state and national economic environments.
Investigate And Quantify Any Relationships That May Exist Between The Key
Measures. This will be done by gathering data from test results obtained from site
investigations carried out for state and local shire road projects. Test data will be
sourced from Transport and Main Roads soil laboratories or Main Roads databases, as
well as other publically available information from local shire authorities. The data will
be analysed to determine whether any significant relationships exist between the various
key parameters.
Establish A Ranking Of Reliability For The Key Measures Of Reactive Soil Behaviour.
By analysing the outcomes of the data analysis, a review of each method will be
conducted and a ranking will be given to each test method in terms of the meaningfulness
of the reported values, the time required for the test, the ability of industry to perform the
test and the skill of the operator required to perform the test.
Develop Guidelines To Assist Engineers In Conducting Site Investigations The intent of
this objective is to provide a list of the minimum testing requirements to conduct an
investigation to characterise the expansive properties of a site. In reference to the
Australian Standard for the Residential Slabs and Footings AS2870, complementary
information can be developed based on any identified relationships between parameters
as well as the ranking score, which will assist the engineer in customising a testing
program for site investigations to manage risk and provide economy.
79
Present project results in the required oral and written formats
This project as part of the course ENG 4112 will be submitted as a written dissertation
and also presented at the University of Southern Queensland’s 2013 Professional Practice
Seminar.
80
Appendix E Assessment of Consequential Effects
81
Environmental, Sustainability and Consequential Safety Effects This project is being
undertaken with the support of The Department of Transport Main Roads and it is the
intent to develop guidelines for the improved identification and characterisation of
reactive soil sites.
In this way, this project aims to minimise the effects of reactive soils, and therefore
reduce the level of failures of roads constructed on these problematic materials. This will
result in more efficient use of financial and natural resources, by reducing the need for
maintenance works and rehabilitation. As a consequence, it is intended that any decrease
in maintenance and rework will increase the sustainability of infrastructure in the future.
The main activities for this project involve information and data gathering, which are
considered to represent no environmental risk.
Immediate safety concerns have been addressed in Appendix C. Identified risks are and
will be managed and control methods implemented.
Ethical Effects The ethical responsibility for the work is fully accepted by the author and
at no time will the author compromise the tenets of the Engineers Australia Code of
Ethics, as detailed below:
Source: http://www.engineersaustralia.org.au. Code of Ethics 2010.
“As engineering practitioners, we use our knowledge and skills for the benefit of the
community to create engineering solutions for a sustainable future. In doing so, we
strive to serve the community ahead of other personal or sectional interests.”
82
Improved understanding of expansive soils and their properties will be to the benefit of
the community. It will enable to road planners and designers to make more informed
decisions on the best solutions for the construction of roads on problematic materials.
“Demonstrate integrity”
Publication of the project will promote the honour, integrity and dignity of the
engineering profession.
“Practise competently”
The author’s area of education and current vocational specialisation is in the area civil
engineering including high focus on construction materials and geotechnical testing.
Careful consideration and diligence will be sought throughout the duration of the project.
Professional advice from members with greater competence will be sought in situations
when specialised skills or knowledge are required.
“Exercise leadership”
The author will ensure all actions and results throughout the project are fair, honest and in
the best interest of the community, client, employers and colleagues.
“Promote sustainability”
Development and delivery of the project is supported by Department of Transport and
Main Roads, which has protocol and guidelines which ensure that all work completed is
in the best interest of the state of Queensland. As an employee of this organisation, the
author has undertaken appropriate workplace training and has the awareness of the
Department’s focus on provide the best service for the community, now and for future
generations.
83
Appendix F Risk Assessment
84
Risk Assessment In all projects it is essential that all risks are identified, assessed and
controlled. This is done by safe workplace procedures and a controlled work
environment.
The Project has the major tasks involved:
1. Office Work Research
2. Office Work Reporting
A summary of the Hazard Identification, Potential Risks and Control Methods is
presented in Table A6-1.
85
Specific Activity Hazard and Risks C L Risk Controls/treatments C L Risk
Level Level
Research and Sitting at a desk for Mod P High Observe correct Mod U Medium
Reporting prolonged periods posture
– potential back Use lumbar support on
and neck injuries chair
Take regular breaks

incorporating stretches
and standing
Research and Viewing a Mod P High Limit computer use to Mod U Medium
Reporting computer screen no more than 1 hour at
for prolonged a time.
periods – potential
eye strain and
Ensure adequate
headaches
breaks are taken away
from the computer (at
least 20 mins)
During periods of use,

focus to infinity at
regular intervals
C = Consequence. L = Likelihood
Risk Matrix
Likelihood Consequence
Insignificant Minor Moderate Major Severe
Almost certain MEDIUM HIGH HIGH EXTREME EXTREME
Likely MEDIUM MEDIUM HIGH HIGH EXTREME
Possible LOW MEDIUM HIGH HIGH HIGH
Unlikely LOW LOW MEDIUM MEDIUM HIGH
Rare LOW LOW MEDIUM MEDIUM HIGH
Table F-1 Risk Assessment Matrix.
86
Appendix G Resource Analysis
87
Resource Analysis
To efficiently complete a project it is essential that prior investigations into the
availability of resources, costs involved in using these resources and feasibility of these
resources are completed before the project can be commenced. There are very few
resources needed to complete this project, however the ones that are required are critical
items and without them the project would not be able to be completed. All items can be
sourced readily by the author, or from Department of Transport and Main Roads. The
timing of external resource dependant tasks can be managed to negate any impact on the
delivery time of the project.
The resources likely to be required are shown in Table A6 below:
Resource Possible Problems Control Measures
Transport and Main Roads Documents unavailable or Ensure adequate lead times
Library and technical delays in delivery in making library
reports archive borrowing requests
Reports and test data from Unavailability of key staff Alternative sources of data
Local Authorities contacts and not able to may need to be found.
obtain permission to
access data
Computer Printing and Some restrictions on Some minor printing costs
Internet access at TMR. may be required. Most
internet access will be
from private account – not
likely to cause delays
Table G-1 Resources Required
88
Appendix H Project Timeline
89
Task Description Predecessors Resources Start End Mar April May Jun Jul Aug Sep Oct
1 Specification - PR Mar Mar

2 Background 1 PR, TMR Apr May
3 Literature 1 PR, TMR, Apr May
Review Other
4 Appreciation 3 PR, TRM, May May
Other
5 Data Collection 3 PR Jun Jul
6 Data Analysis 5 PR Jul Aug
7 Review of 5,6 PR Aug Aug
Results
8 Develop 3,5,6 PR Aug Aug
Guidelines
9 Conclusions 3,7,8 PR Sep Sep
10 Reporting 6,7,8,9 PR Sep Oct
PR – Peter Reynolds, TMR – Transport and Main Roads, Other – Local Road Authorities.
Table H-1 – Project Timeline
90

Reynolds - 2013 - Thesis UQ - Eng Correlations For The Characterisation of Reactive Soil Behaviour For Use in Road Design

Uploaded by

Copyright:

Available Formats

Reynolds - 2013 - Thesis UQ - Eng Correlations For The Characterisation of Reactive Soil Behaviour For Use in Road Design

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Reynolds - 2013 - Thesis UQ - Eng Correlations For The Characterisation of Reactive Soil Behaviour For Use in Road Design

Uploaded by

Copyright:

Available Formats

University of Southern Queensland

Faculty of Health, Engineering & Sciences

Engineering Correlations for the

Peter William Reynolds

in fulfillment of the requirements of

Courses ENG4111 and ENG4112 Research Project

towards the degree of

Bachelor of Engineering (Civil)

Submitted: October 2013

Faculty of Health, Engineering & Sciences

ENG4111/2 Research Project

The Council of the University of Southern Queensland, its Faculty of Health,

This dissertation reports an educational exercise and has no purpose or validity

Limitations of Use ....................................................................................................................iii

List of Tables ............................................................................................................................ ix

1.1 Aims and Objectives of the Project .......................................................................... 2

2 Impacts on Road Performance .......................................................................................... 3

2.1 Distress on Drainage Structures – Case History....................................................... 3

3 Background and Literature Review .................................................................................. 6

3.1 Properties of Reactive Soils ..................................................................................... 6

4 Laboratory Methods for Measuring Reactivity............................................................... 14

4.1 Laboratory Measurements...................................................................................... 14

5 Predictions of Surface Movement using AS2870 ........................................................... 26

5.1 General ................................................................................................................... 26

6 Analysis of Historical Data ............................................................................................. 29

6.1 General ................................................................................................................... 29

7 Discussion of Results ...................................................................................................... 33

8.1 Major Outcomes and Key Findings........................................................................ 48

9 List of References ........................................................................................................... 52

Appendix A Project Specification.................................................................................. 60

Appendix B Tabulated Test Data................................................................................... 62

Appendix C Project Aims and Objectives ..................................................................... 75

Appendix D Project Methodology ................................................................................. 77

Appendix E Assessment of Consequential Effects........................................................ 81

Appendix F Risk Assessment........................................................................................ 84

Appendix G Resource Analysis ..................................................................................... 87

Appendix H Project Timeline ........................................................................................ 89

It is important to properly characterise the properties of expansive soils prior to

Whilst there is a body of data and published information on some of the

1.1 Aims and Objectives of the Project

The objectives of this project can be summarised as follows:

2 Impacts on Road Performance

2.1 Distress on Drainage Structures – Case History

Figure 2 - Image of damaged abutment wall of culvert at East Warianna Creek

2.2 Distress on Road Embankment – Case History

Shortly after the completion of construction, longitudinal cracks were observed on

3 Background and Literature Review

3.1 Properties of Reactive Soils

In addition to these intrinsic properties, the magnitude of the stresses applied to

3.2 Mineralogy and Mechanics of Volume Change

Figure 4 - Diagram of the structure of a montmorillonite layer (Al-Omari 2000)

By comparison, kaolinite, which exhibits little expansive properties, comprises a

Figure 5 - Comparison of mineralogical structures of kaolinite and montmorillonite (Farris)

3.3 Soil Suction

Figure 7 below shows graphically the soil water energy states.

Figure 7 - Diagram of soil water potential energy states (Or et al 2005)

This is expressed as the simplified equation:

(Or, Tuller and Wraith)

Generally, soil water potential is referred to as total suction terms, and is