Long-term behavior of lime-stabilized kaolinite clay

ORIGINAL ARTICLE

Long-term behavior of lime-stabilized kaolinite clay

Aydın Kavak • Gokhan Baykal

Received: 22 August 2010 / Accepted: 8 October 2011 / Published online: 28 October 2011

� Springer-Verlag 2011

Abstract Clay soils create many problems for highway

construction and they have to be replaced or improved by

stabilization for satisfactory performance. Lime stabiliza-

tion is a well-established technique to improve the per-

formance of clays. Cementitious minerals form upon

mixing of clay with lime causing an improvement in

strength and durability. In the study, the changes in the

microfabric of long-term cured lime-stabilized kaolinite

clay using X-ray diffraction pattern, scanning electron

microscope and unconfined compressive strength (UCS) is

presented. Unconfined compression test samples at two

different lime contents (4 and 12% by weight) were pre-

pared and cured in a humidity room for long time curing.

The UCS of pure kaolinite was originally 125 kPa, which

increased to 1,015 kPa after 1 month and to 2,640 kPa (21

times the initial value) after 10 years for cured lime-sta-

bilized kaolinite samples. Similar long-term strength

increases were also observed for stabilized kaolinite with

12% lime. Calcium aluminate silicate hydrate minerals

were detected in the structure of the kaolinite. This sug-

gests pozzolanic reactions with lime stabilization may

continue in the long-term for up to 10 years.

Keywords Lime stabilization � Lime � Soil �Stabilization � Clay � Kaolinite

Introduction

Environmental impact of road construction on the ecology

is detrimental due to need for large volume of fill material.

When the performance in the site is not suitable for road

construction, borrow material is needed which is dredged

from nearby river or mined from surrounding pits. Instead

of using borrow material, it may be more economical to use

in situ soil provided that the performance criteria are met.

Lime stabilization of clay soil is widely done to improve

the performance of in situ soil to the level that, the required

performance criteria are satisfied. Improvement of the

existing soil reduces the need for cut and fills works,

minimizing the environmental impacts.

Soil stabilization by lime refers to the admixture of this

material in the form of calcium hydroxide to the soil and

the compaction of the mixture at the optimum water con-

tent. These chemical processes modify the soil structure

whereby larger grain aggregates are formed (Broderick and

Daniel 1990), leading to several advantages in the suit-

ability of the soil for road construction. Lime stabilization

is widely used, both to make clay-bearing soils suitable as

base, sub-bases and sub-grades to increase the strength, and

to produce other properties of useful infrastructure mate-

rials that contain clay. The stabilization process occurs

because calcium hydroxide (lime CaOH2) reacts favorably

with some but not all of the clay. The reaction causes the

clay’s properties to change substantially (Clare and

Crunchley 1957). Plasticity and the tendency to expand

with added moisture are immediately reduced, as the

troublesome cations of sodium and potassium are replaced

by those of calcium or magnesium. Also, lime causes

flocculation, a lumping together of clay particles, which

increase particle size (Sides and Barden 1971). The addi-

tion of lime to soil causes an immediate increase in soil pH

A. Kavak (&)

Department of Civil Engineering, Kocaeli University,

Umuttepe Camp., Kocaeli, Turkey

e-mail: [email protected]

G. Baykal

Department of Civil Engineering,

Bogazici University, Bebek/Istanbul, Turkey

e-mail: [email protected]

123

Environ Earth Sci (2012) 66:1943–1955

DOI 10.1007/s12665-011-1419-8

due to partial disassociation of the calcium hydroxide

(Ladd et al. 1960). The calcium ions in turn combine with

the reactive hydrous silica and/or aluminates, which harden

with time to affect stabilization of the soil. The relatively

high pH of the pore water facilitates the formation of this

cementitious gel by increasing the reactivity of the surface

silica. Over a longer period, there can be a substantial

strength gain through a pozzolanic reaction with any

available silica or alumina.

When stabilizing soil with lime, the amount of calcium

required depends on the eventual use of the mix, and the

objectives of stabilization. As indicated, lime stabilization

of natural materials for bases and sub-bases is intended both

to reduce plasticity and the accompanying volume changes

with moisture content, and to increase strength. The com-

mon test for reduction in plasticity and volume change is the

plasticity index (PI). Results with one group of clays

showed that values reduced by one-third to two-thirds with

3% lime and to a non-plastic condition with 5%. Strength

increases are often measured by standard tests such as

unconfined or triaxial compression, California bearing ratio

(CBR), or flexure test. The strength of lime-stabilized soils

usually increases not only in the short time but also con-

tinues over time since pozzolanic action takes place (Kavak

1996). Solanki et al. (2009) showed the variation of mod-

ulus of elasticity (ME) and unconfined compressive strength

(UCS) values with lime content. Similar to resilient mod-

ulus (Mr), the modulus of elasticity and UCS essentially

showed the same trend. It is evident that there is a signifi-

cant increase in ME and UCS with increasing lime content.

Eades and Grim (1963) studied kaolinite and lime

reactions. They concluded that the reaction of lime and

kaolinite causes the formation of new crystalline phases,

which are tentatively identified as calcium silicate

hydrates. This reaction seems to take place by the lime

reacting with kaolinite particles around the edges, with a

new phase forming around a core of kaolinite.

Croft (1964) examined kaolinite samples by X-ray dif-

fraction method. The preliminary examinations revealed a

rapid attack uniformly distributed over all surfaces of the

clay minerals. The intensity and sharpness of all kaolinite

lines, as well as those of calcium hydroxide, had dimin-

ished considerably after reaction for 1 week. Croft (1964)

also observed the mixture by electron microscopy; the

particles revealed a tendency towards pseudohexagonal

shape; most were, however, irregular and ragged. Glenn

and Handy (1963) cured kaolinite and lime mixtures for

2 years. X-ray diffraction shows a reduction in kaolinite

peak intensities after reaction with lime. However, much

unreacted Ca(OH)2 remains in the calcitic lime mix after

2 years. Small amounts of Ca(OH)2 and much Mg(OH)2

remain in the dolomitic lime mixes. There is an evidence of

12.6-A material from reaction with dolomitic limes, and a

7.6-A product, which appears strongest from reaction with

calcitic lime but is also present in the other mixes.

The in situ performance of lime-stabilized soils

depending on time is also a very important issue for sta-

bilization. As an example, lime stabilization, soil replace-

ment and soil reinforcement by geosynthetics methods

were used to improve the soil when widening old railway

lines to double track in the Czech Republic. After 5 years

in service, the slab section was causing permanent prob-

lems for the track maintenance and had to be reconstructed,

while the lime-improved section behaved very well. Herle

(2005) investigated the subsoil anomalies in this research

on a radar record and also conducted unconfined com-

pression tests on samples taken from test pits. Following

this study, the use of small quantities of lime such as 1, to

2% for rail track subgrade improvement has been widely

accepted as a very efficient and economic method for

reconstructing old rail corridors. In comparisons with other

methods (e.g. soil replacement, soil reinforcement), it is

also environmentally friendly, as it is not necessary to shift

large quantities of soil to and from the site. Another

important example was the construction, in 1969, of a test

section with base and sub-base layers of lateritic soil sta-

bilized with calcitic lime near the town of Cruz Alta, in

Brazil. Despite the lack of major rehabilitation works,

35 years later the road retained an acceptable serviceability

level. Nunez et al. (2005) studied cored samples and new

samples prepared in the laboratory. Cored specimens

showed higher strength and modulus than those compacted

in the laboratory, due to the longer curing time of the cured

samples and to some that of carbonation that took place in

the laboratory samples.

While long-term stabilization data exists in the litera-

ture, virtually all of this is from field scale projects and not

from a controlled laboratory environment. Although it is

known that pozzolanic reactions continue for many years,

there is limited available laboratory data in the literature

for the long time curing and in many studies, curing time

for this process is less than 1 year. In the following table, a

list of similar studies with the maximum curing time in the

laboratory environment are listed.

As it can be seen from the Table 1, there are many

researches related to investigate the effect of curing time

on lime stabilization but, the curing time under controlled

conditions in these researches is less than 1 year. Locat

et al. (1990) studied shear strength development of sensi-

tive clays on lime stabilization under controlled environ-

ment as much as 300 days curing time. Uddin et al. (1997)

(cross reference Little 1999), cured a plastic clay for

180 days at lime contents between 2.5 and 15%. Very

substantial strength gains occurred between 60 and

180 days. Strength gains were maximal at 10% lime and

decreased by 15%. Evans (1998) stabilized highly plastic

1944 Environ Earth Sci (2012) 66:1943–1955

123

Queensland Black clays with 8% lime, which were cured

for 28 days and 26 weeks. As in previous reports, the

strength gain between 28 days and 26 weeks was very

substantial, demonstrating the long-term pozzolanic

process. Baykal (1989) studied the effect of micro mor-

phological development on the elastic moduli of lime-

stabilized bentonite samples cured for 180 days. The

improvements in the lime-stabilized bentonite were shown

by unconfined compression tests, chemical analysis, and

micro morphological investigation using X-ray diffrac-

tometer (XRD) and scanning electron microscope (SEM)

techniques.

The objective of this study was to investigate the long-

term behavior of lime-stabilized kaolinite clay. The study

examined the changes in the microstructure of lime-stabi-

lized kaolinite using X-ray diffraction pattern, SEM elec-

tron microscope images and the changes in strength for

long time cured kaolinite samples under controlled condi-

tions. In the present study, two groups of work were mat-

ched and a full dataset was obtained for a period of

10 years. The first group tests were conducted on lime-

stabilized kaolinite. During this study, many samples were

also left for long-time curing in a temperature and humidity

controlled curing room. Unconfined compression tests

were conducted at various curing times, and the micro

fabric of lime stabilized kaolinite was investigated using

XRD, SEM and EDS methods for 10 years cured

specimens.

Table 1 Literature studies related to UCS tests of lime stabilized soils

References Titles Tests Curing time

Locat et al. (1990) Laboratory investigations on the lime stabilization of

sensitive clays: shear strength development (Clays)

Proctor test, shear strength, SEM 0, 1, 3, 10, 30, 100 and

300 day(s)

Ladd et al. (1960) Recent soil-lime research at the Massachusetts Institute

of Technology (Kaolinite/Montmorillanite)

Dynamic compaction test,

classification tests, unconfined

compression test, CBR tests

0, 7 and 28 day(s)

Hilt and Davidson

(1960)

Lime fixation in clayey soils (Kaolinite/

Montmorillonite)

Classification tests, chemical

analysis, pH test, unconfined

compression test

28 days

Croft (1964) The processes involved in the lime-stabilization of clay

soil (Kaolinite/Montmorillonite)


analysis, pH test, X-ray

analysis, SEM

60 days for X-ray

Hilt and Davidson

(1961)

Isolation and investigation of a lime-montmorillonite

crystalline reaction product (Bentonite/Kaolinite)


analysis, pH test, X-ray

analysis, SEM

30 days

Taylor and Arman

(1960)

Lime stabilization using preconditioned soils

(Kaolinite/Montmorillanite)

Classification tests, Proctor test,

Tri-axial test

40 days

Mitchell and El Jack

(1964)

The fabric of soil-cement and its formation (Kaolinite/

Montmorillonite)


analysis, Proctor test,

unconfined compression tests

240 days

TRB State of the Art

Report-5 (1987)

(Little 1999)

Lime stabilization of Illinois soil, Field evidences

indicates some soil-lime mixtures can continue to gain

strength for excess of 10 years

Unconfined compression tests,

field evidences

75 days for laboratory

works, field

evidences for

10 years

Baykal (1989) The Effect of Micromorphological Development on the

elastic moduli of Fly-ash, Lime Stabilized Bentonite


analysis, unconfined

compression tests, XRD, SEM

1, 28, 90 and 180 days

Shackelford and Daniel

(1991)

Diffusion in saturated soil II : results for compacted

clay (Kaolinite)

Classification tests, Proctor tests,

diffusion test

30 and 109 days

Doty and Alexander

(1968)

The time-dependent behavior of pozzolanic strength

development of 12 different soils cured at 23 C and 38

C. Significant strength increases are observed between

180 and 360 days

Classification tests, PI

comparisons

From 7 days to

360 days

Uddin et al. (1997)

(Little 1999)

The time-dependent behavior of pozzolanic strength

development of a plastic clay

Unconfined compression tests Between 60 days and

180 days

Evans (1998) Soils highly plastic Queensland Black clays, the

strength gain between 28 days and 26 weeks was very

substantial demonstrating the long-term pozzolanic

process

Classification tests, PI

comparisons, unconfined

compressive strength

Between 28 days and

26 weeks

Environ Earth Sci (2012) 66:1943–1955 1945

123

Materials and methods

Methodology

In the study, the geotechnical properties of natural kaolinite

were determined at the beginning. The optimum lime

quantity for kaolinite clay was selected as the minimum

lime percentage in weight that produced the maximum

UCS. Atterberg limit tests and pH tests were also con-

ducted at various lime contents between 2 and 10% to

determine the optimum lime content of the kaolinite clay.

In order to determine the short- and long-term effects of

excessive lime amount in the clay soil, studies have been

made also with 12% of lime percentage. Kaolinite with 4

and 12% lime by weight are denoted as K4 and K12,

respectively.

Kaolinite was air-dried and stored in plastic containers

in the laboratory. Kaolinite was taken from the container

and the required weight was mixed with a predetermined

amount of lime by using a hand mixer with speed control.

Thorough mixing was essential to evenly distribute the

moisture. During the mixing period, the rate of the mixer

was held constant. After 5 min of mixing, water was slowly

added to the mixture. The amount of water added corre-

sponds to the optimum moisture content of the soil and

lime mixture. The Harvard miniature compaction apparatus

was used to compact the samples for unconfined com-

pression tests with a specially designed hammer. The

standard Proctor energy level was used for unconfined

compressive strength tests conducted in the study. In order

to obtain comparable stress–strain curves between samples,

equal compaction energy per blow per unit compacted

volume was used and the hammer-to-mold diameter ratio

was maintained. After compaction, the specimens were

carefully trimmed. The specimens were first sealed in

plastic containers and then the same samples were sealed in

aluminum foil and have been kept in a humidity room at a

temperature of 21�C and humidity of more than 95%.

Keeping the moisture content constant is essential for the

development of cementitious products. The weights of

specimens with and without the cover were recorded and

the specimens were checked to determine whether they had

lost water content during curing.

To determine the mineralogical characteristics of the

kaolinite chemical analysis, XRD and SEM analyses and

environmental scanning electron microscope (ESEM)

analyses were made of natural samples and samples to

which lime was added. Chemical analyses were conducted

with added 4% lime by weight and the ratios of the

structure that forms the components were determined.

XRD analyses were conducted on natural and long-term

cured lime-stabilized kaolinite samples. Small sub-samples

were taken from the unconfined compression test samples

and were powdered and dried in a sterilizer. Copper K-aradiation with wavelength B1.54 A was used. Diffraction

patterns showing two theta angles with respect to intensity

values were plotted. The samples used for SEM images

were kept in a low temperature sterilizer for 24 h. Before

being placed under the microscope, the samples, which

have a non-conductive surface, were coated with a very

thin conductive carbon layer. The samples analyzed by

ESEM scanning electron microscope were not coated.

The Kaolinite clay used in the present study is mainly

composed of silica (61.6%) and alumina (13.31%). Iron

oxide, magnesium oxide and Calcium oxide also are

present (4–5%) (Table 2).

To determine the geotechnical properties of the Kao-

linite sieve analysis, Atterberg limit tests, Specific gravity

and standard and modified Proctor tests were conducted.

Tests were conducted in accordance with ASTM

D421&422, ASTM D4318, ASTM D 854, ASTM D

698-78 and ASTM D 1557-78 (Bowles 1992). The geo-

technical characteristics of the natural kaolinite are shown

below, in Table 3.

Physical and chemical properties of the lime used in the

study are given in Table 4 (Nuh Cement and Lime Fac-

tory). The lime is stored in the laboratory within a plastic

cover, which is inside a sealed glass container to prevent

the carbonation effect of the CO2 present in the air.

The hydrated calcium lime used in the studies was

manufactured by Nuh Company. Calcium hydroxide con-

tent varies between 80 and 86%.

Results

Determination of the optimum lime content

Optimum lime content can be defined as the lowest lime

content giving the maximum unconfined compressive

Table 2 Chemical analysis of

kaolinite (%)SiO2 61.60

Al2O3 13.31

Fe2O3 5.27

TiO2 0.75

CaO 4.31

MgO 4.52

Na2O 0.57

K2O 1.76

Cr2O3 0.026

BaO 0.03

SO3 0.08

P2O5 0.08

MnO 0.138

L.O.I. (loss of ign.) 6.27


123

strength of lime-stabilized soils, by also taking into con-

sideration PI and pH values. In order to find the optimum

lime percentage for the kaolinite clay, unconfined com-

pression tests at various lime contents and pH tests using

the Eades and Grim (1963) procedure and Atterberg limit

were conducted.

Atterberg limits

After adding lime, structural transformation and floccula-

tion begins immediately. Kaolinite was initially mixed with

lime and was then kept under curing conditions for one

hour (Kavak and Akyarli 2007) before Atterberg limit tests

were performed. The tests were conducted in accordance

with ASTM 4318. The experiments were done by adding

lime at weight percentages of 0, 2, 4, 5, 6, 8, 10%

accordingly.

The liquid limit value for pure kaolinite was found to be

37.5, which increased to 42.5 by the addition of 2% lime.

The liquid limit is around 41–42 for higher lime percent-

ages. The plastic limit also increased from 29.8 to 34.2

percent with the addition of 2% of lime and remained

relatively constant at higher percentages of lime. The

results of liquid and plastic limit and PI values for kaolinite

with different lime percentages are presented in Table 5.

At 4% lime content, the PI decreased from 7.7 to 6.5,

compared with the original specimen. This percent was

almost the lowest value and this lime content was chosen as

the design lime content for the kaolinite, Atterberg limit

tests give similar design lime content, at 4%. The PI value

drops to 7.5 and does not change much for higher lime

percentages.

UCS tests

The unconfined compressive strength was conducted from

0 to 12% of lime at 2% increments using 0–28 days cured

specimens. A summary of the tests is shown in the Fig. 1,

below. The UCS tests showed that the lowest lime content

that produced maximum soil strength was 4% lime.

In the Fig. 1, the effect of curing and of lime content can

be seen clearly. The UCS of natural kaolinite is 125 kPa,

reaching to 1,000 kPa at 4% lime content after 28 days

curing. For almost all cured specimens, lime content of

more than 4% had little effect on the UCS. Based on Fig. 1,

the optimum lime content can be taken as 4%, taking into

economic factors of the stabilization work.

Table 3 Geotechnical properties of kaolinite clay

Property

Soil classification

Classification-USCS ML

Classification-AASHTO A-4

Atterberg limits

LL (%) 38

PL (%) 30

PI (%) 8

Sieve analysis USCS

Sand (75–2,000 lm) (%) 0

Silt (2–75 lm) (%) 42

Clay (\2 lm) (%) 58

Specific gravity 2.62

Organic material (%) 0

Compac. param. (standard proctor)

Dry unit weight (kN/m3) 15.3

Optimum moist. cont. (%) 21.5

Comp. param. (modified proctor)

Dry unit weight (kN/m3) 15.6

Optimum moist. cont. (%) 18.6

Natural water content (%) 6–8

Activity 0.15

Table 4 Physical and chemical properties of lime

Chemical name Calcium hydroxide

Physical appearance Dry white powder

Boiling temperature (�C) 100

Heat of fusion (�C) 580

Bulk density ( kg/m3) Max. 500

Specific gravity 2.21

Over 90 l (%) 3–6

Over 63 l (%) 7–10

pH (25�C) 12.4

Ca(OH)2 (%) 80–86

Active CaO (%) 60.6–65.15

CaO ? MgO (%) 83–93

MgO 1–2

Insoluble material (%) max. 1

Bounded H2O (%) 19.4–20.85

S max. 0.5

R2O3 (Al2O3 ? Fe2O3) (%) max. 1

Table 5 Atterberg limits of kaolinite at different lime contents

Lime cont. (%) LL (%) PL (%) PI (%)

0 37.5 29.8 7.7

2 42.5 34.2 8.3

3 41.2 33.8 7.4

4 41.1 34.3 6.5

5 40.8 33.9 6.9

6 40.9 34.2 6.7

8 41.1 32.8 7.3

10 41.7 33.6 8.1


123

pH tests

The pH level is an indicator of the reaction between lime

and soil. The pH of the soil increases to 12.4, with optimum

amount of lime (Eades and Grim 1963). At this pH, the soil

alumina and soil silica (pozzolans) become soluble and

combine with free calcium and water to form cementitious

products: calcium aluminate silicate hydrates (CASH) and

calcium-aluminate-hydrates (CAH) (Little 1999).

The pH tests were also conducted according to the

method of Eades and Grims (1963) using lime contents

from 1 to 10 at 1% increments. In Fig. 2, the variation of

pH with lime for kaolinite samples is shown. The pH value

increases to a maximum of almost 12.4 at 4% lime.

Based on the Atterberg limit, UCS and pH test results, the

optimum lime for the research content was chosen as 4%.

Standard and modified proctor tests

Moisture and density relations of the kaolinite were deter-

mined using standard and modified Proctor compaction

tests. The tests were conducted according to the ASTM D

698-78 and D 1557-78 (Bowles 1992). Following these

tests, comparisons were made of the behavior of the lime-

stabilized kaolinite between standard and modified Proctor

tests. The soil and lime mixtures were prepared for com-

paction test as described above in the section on sample

preparation. First, Proctor tests were conducted using the

natural kaolinite and then the kaolinite was mixed with the

chosen optimum lime content and the tests were repeated

for this case. A summary of the maximum dry unit weight

and optimum water content of kaolinite with and without

lime is shown in the Table 6. The Proctor curves are shown

in the Fig. 3.

Optimum water content and dry unit weight for kaolinite

samples was found to be 21.5% and 15.3 kN/m3, respec-

tively, for standard Proctor tests. In the light of the standard

Proctor test results, the optimum water content increases to

25% and maximum dry unit weight decreases to 14.6 kN/m3

with 4% lime. The trend is also similar for modified

Proctor tests, while optimum water content increases to

24.5%, the maximum dry unit weight drops to 14.7 kN/m3

with 4% of lime. Optimum water content of kaolinite

0

200

400

600

800

1000

1200

0 2 4 6 8 10 12

Unc

. Com

p. S

tr. (

kPa)

Lime Content %

0 Day0 Day0 Day avg.1 Day1 Day1 Day avg.7 Days7 Days7 Days avg.28 Days28 Days28 days avg.

Fig. 1 UCS tests of kaolinite at

different lime contents

Fig. 2 Kaolinite pH values with respect to lime content %

Table 6 Maximum dry unit weight and optimum water content

values of kaolinite with and without lime

Maximum dry unit

weight cd (kN/m3)

Optimum moisture

content wt (%)

Kaolinite (standard) 15.3 21.5

Kaolinite (modified) 15.6 18.6

Kaolinite ? 4% by

weight lime (standard)

14.6 25.0

Kaolinite ? 4% by

weight lime

(modified)

14.7 24.5

Kaolinite ? 12% by

weight lime (standard)

14.5 26.5


123

sample and dry unit weight for kaolinite samples were

found as 26.5% and 14.5 kN/m3, respectively, for K12

samples. It is clear that, optimum water content increases

with the increases of lime content and maximum dry unit

weight decreases with increasing lime.

According to the compaction test results, it is observed

that for lime-stabilized kaolinite samples, the results are

similar for both of standard and modified tests. The vari-

ation in the optimum water content and maximum dry unit

weight is less than 3 and 1%, respectively, between stan-

dard and modified Proctor tests. As a result of compaction

tests, it can be said that optimum water content values

increase and maximum dry unit weight values decrease

with lime stabilization. The Proctor curves become flatter

for lime-stabilized samples and smoother curves are

obtained. The results are similar to those reported in pre-

vious studies by Neubauer and Thompson (1972) and

Solanki et al. (2009).

According to the results of standard and modified

Proctor tests for natural and lime-stabilized kaolinite clay,

it can be said that although the compaction energy values

are very different for standard and modified Proctor tests,

the moisture versus dry unit weight graphs of kaolinite

clays with lime are very similar. Kaolinite behaves similar

to that of a granular material in micro fabric after lime

stabilization.

Unconfined compression tests

Unconfined compression tests were conducted on kaolinite

samples prepared with standard energy at the 4% optimum

lime content and cured at different periods ranging from

0 days to 123 months. The samples were prepared for UCS

tests using a Harvard miniature compaction set according

to the defined methodology described above in the section

on sample preparation. The specimens were cured at 21�C

in a room with greater than 95% humidity. The maximum

curing time used in the study is more than 10 years. The

unconfined compressive strength of natural kaolinite was

125 kPa, reaching 1,015 kPa for lime-stabilized kaolinite

specimens with 4% lime that were cured for 28 days. In the

long-term, for a period of 123 months curing UCS reached

2,640 kPa. The increase of strength was 21 times higher

than the original strength. Three typical unconfined com-

pression tests for natural kaolinite K4, cured for 28 days

and K4 cured for 123 months are shown in Fig. 4.

The behavior of the lime-stabilized kaolinite was very

different from that of natural kaolinite. The modulus of

elasticity, which is the slope of the stress–strain curve

changes sharply for lime-stabilized kaolinite, and the soil

shows brittle behavior with lime. The unit strains at failure

decreases to around 1% for lime-stabilized and cured

specimens, compared to 2.3% for natural kaolinite samples.

UCS test results for K4 samples at different curing periods

up to 123 months are shown in the Fig. 5 below.

In Fig. 5, it is clearly seen that, the UCS of lime-stabi-

lized kaolinite increases with time. Each UCS value in the

bar graph is the average value of three UCS tests and the

maximum variation in these three tests was found to be

10% at maximum variation. The UCS increased from the

reference value of 125 kPa to 1,000 kPa in 1 month, to

2,064 kPa after 96 months and finally reached to 2,640 kPa

at 123 months curing time.

In order to see the changes in strength and behavior in

the short- and long-term, a series of UCS control tests were

conducted with 12% content, which was the maximum

value used for the lime determination studies. The stress–

strain curve for unconfined compression tests of kaolinite

with 12% lime are shown in Fig. 6 below. The unconfined

compression strength increases with curing time and

reaches to 2,507 kPa in 123 months. The strength increase

was very high like K4 samples which is approximately 20

times higher than the initial kaolinite’s unconfined com-

pression strength.

10

11

12

13

14

15

16

17

18

14 16 18 20 22 24 26 28 30 32 34 36

Max

imum

Dry

Uni

t W

eigh

t , k

N/m

3

Water Content , %

Kaolinite (Standard)

Kaolinite + 4% Lime (Standard)

Kaolinite (Modified)

Kaolinite + 4% Lime (Modified)

Fig. 3 Standard and modified Proctor test results for kaolinite Fig. 4 UCS tests for K4 samples


123

Elastic modulus values are determined from UCS tests

by using the slopes of the stress–strain curve between 20

and 50% of the failure stress. The increase in UCS values is

in the range of 20 times the original strength but the

increase in elastic moduli is approximately 50 times

compared to original one for K4 samples. The elastic

modulus values are calculated and presented in the Figs. 4

and 6.

Similar to the results of the K4 samples, the behavior of

the K12 lime-stabilized kaolinite was very different to that

of natural kaolinite. The modulus of elasticity which is the

slope of the stress–strain curve changes sharply for lime-

stabilized kaolinite, the soil shows brittle behavior with

lime. The unit strain at failure decreases to 1.3% for lime

stabilized and cured specimens compared to 2.3% for

Fig. 5 UCS tests conducted in

123 months for K4 at various

curing periods

0

500

1000

1500

2000

2500

3000

0 0.5 1 1.5 2 2.5 3

UC

S(k

Pa)

Strain (%)

Natural

1 day

7 days

28 days

6 months

123 months

Fig. 6 UCS tests, stress–strain curves for K12 samples

Fig. 7 Unconfined

Compressive Strength Values

for K12


123

natural kaolinite samples. In Fig. 7, it is clearly seen that,

similar to K4 samples, the UCS of K12 lime-stabilized

kaolinite increases with time.

Each UCS value in the bar graph is the average value of

three UCS tests and the maximum variation between these

three tests was found to be 10%. UCS increased from the

reference value of 125 kPa to 1,000 kPa in 1 month,

1,550 kPa at 115 months and finally reached 2,507 at

123 months curing time.

The strength gain of the lime-stabilized samples is very

high in the long-term. Many design methods for lime-sta-

bilized soils use the unconfined compressive strength of

28-day or 3-month cured samples as design strength. In

Fig. 7, the long-term increase in strength at 123 months is

2.9 times for K4 samples and 3.5 times for K12 samples

compared to their respective 28-day cured values. The very

significant strength increases further illustrate the impor-

tance of maintaining enough lime and high enough pH to

continue pozzolanic reactivity in a reactive clay soil. As

long as this maintained strength gain in the long-term is

possible (Little 1999). Laboratory data indicate that, under

suitable conditions, strength gain can continue for in excess

of 10 years.

Typical basic characteristics of lime-stabilized kaolinite

samples were calculated and final values are shown in

Table 7. The water content and drying effect might be very

important in the long-term.

The drying is less than 1% compared to original values.

The void ratio is 0.83 for K4 samples and 0.92 for K12

samples. Although, voids in K12 are more than K4 sam-

ples, saturation is higher in K4 samples compared to K12

samples.

CBR experiments

California bearing ratio tests are conducted to lime stabi-

lized kaolinite samples for both of standard and soaked

cases. The tests are conducted according to the ASTM D

1883-87. Experiments have been done on both of samples

K4 and kaolinite. The samples have been prepared first by

mixing at optimum water content and leaving an hour in

the laboratory for initial reactions then compacted in CBR

moulds. The samples are compacted and left in the

laboratory in a humidity room at 21�C. The tests are con-

ducted on samples cured for 1, 7 and 28 days. The varia-

tion in CBR values are shown in Fig. 8. CBR values for

many cured lime-soil mixtures are very high and indicate

the extensive development of pozzolanic cementing

products.

California bearing ratio values increased sharply with

lime for both of standard and soaked CBR tests. For

standard CBR, the CBR of natural kaolinite was found as

15, this value reached to 89 with 4% of lime in 28 days,

which one is 5.9 times higher than the original value. The

similar sharp increases are also observed for the soaked

CBR tests. The soaked CBR value of kaolinite was found

as 2, the value increased with 4% lime to 58 in 28 days.

The value is 29 times higher than the initial value. The

swelling for lime stabilized kaolinite is found less than 1%.

Micro analysis

XRD analysis

X-ray diffractometer analyses were conducted of the K4

and K12 samples, which were cured for long-term

(10 years). The tests were conducted at Bogazici Univer-

sity Advance Technologies R&D center in Istanbul Turkey.

40 kV and 40 mA Copper K-a radiation was used in XRD

analysis. The samples were supplied to the laboratory in

powdered form and were dried in a sterilizer prior to

analysis. XRD analyses were made for kaolinite, K4 and

K12 samples.

D-spacings obtained by XRD patterns belong to the

hydrated calcium aluminum silicate (CASH) mineral

named laumontite. Chemical formula of Laumontite is

CaAl2Si4O12–4H2O. Since this mineral belongs to the

silicate group, this is probably one of the reasons for the

long-term strength gain observed in the samples.

Table 7 Basic characteristics of the lime stabilized UCS samples

K4, 10 years K12, 10 years

Water cont. (%) 25.2 24.4

Unit wt (kN/m3) 17.51 16.6

Dry unit wt (kN/m3) 13.99 13.3

Weight (g) 112.5 106.6

Void ratio 0.83 0.92

Saturation (%) 79.0 69.3

Fig. 8 Standard and soaked CBR values


123

Energy dispersive X-ray spectroscopy

Energy dispersive X-ray spectroscopy (EDS) was used for

the elemental analysis and chemical characterization of K4

samples cured for 10 years. As a type of spectroscopy,

EDS relies on the investigation of a sample through

interactions between electromagnetic radiation and matter,

analyzing X-rays emitted by the matter in response to

impacts from charged particles. The EDS analysis of the

samples was conducted by the micro analysis group at the

Sisecam company research laboratories, using a Thermo

Noran system six X-ray microanalysis system.

Table 8 shows the constituents of K4 kaolinite clay

specimens cured for 10 years determined by EDS.

The stabilized K4 samples consist mainly of Silicon

dioxide (63%), aluminum oxide (24%), sulfur trioxide

(9%) and lime (4%).

SEM analysis

Scanning electron microscopy studies were carried out at

two institutions. The samples examined with Jeol JSM

6360 LV scanning electron microscope at Sisecam

research laboratory, were coated with carbon. The envi-

ronmental scanning electron microscope at Bogazici Uni-

versity does not need coating for nonconductive specimens

(The XL 30 Philips ESEM scanning electron microscope).

The specimens studied at Bogazici University were not

coated (Fig. 9).

In the studies, images of kaolinite clay at 2,500, 5,000

and 10,000 magnifications were compared with K4 sam-

ples cured for 10 years. The images are shown in Figs. 10,

11, 12, 13, 14 and 15.

In the figures above, images from untreated soil indicate

crumbled and flaked arrangement of soil particles. Higher

magnified views (5,0009 and 10,0009) of the treated K4

samples indicate the presence of cubic cementitious crys-

tals. As lime reacts with the clay particles, an agglomera-

tion effect takes place which forms bigger particles, with a

consequent increase in pore size area, resulting in well-

aggregated crumbs.

In the following images in Figs. 16 and 17, K4 kaolinite

clay at 350 magnifications were compared with K12 sam-

ples cured 10 years.

The images show that increased lime in K12 samples

creates porous fabric compared to that of K4 samples. The

higher void ratios calculated and presented in Table 7 are

in competence with the visual observations.

The SEM and XRD images of the long-term K4 samples

show that the original platy structure of the kaolinite

mineral is still preserved.

Furthermore, SEM images of the K12 samples show

quartz, calcite and kaolinite minerals within the structure.

Microfabric evaluation of stabilized clays is a challenging

process. It is very difficult to determine the reaction prod-

ucts due to their low concentration. However, microfabric

Table 8 EDS analysis of K4 kaolinite clay cured for 10 years

Element Net counts Weight (%) Formula Compound (%)

O 18,067 51.34 –

Al 5,327 12.50 Al2O3 23.62

Si 11,641 29.33 SiO2 62.75

S 1,182 3.64 SO3 9.1

Ca 602 2.62 CaO 3.66

Ti 36 0.23 TiO2 0.38

Cr 38 0.33 Cr2O3 0.48

Total 100.00 100.00

Fig. 9 XRD analysis of natural

kaolinite and samples with 4

and 12% lime (Bogazici Univ.,

2004)


123

Fig. 10 Kaolinite natural clay 92,500

Fig. 11 K4, 92,500 (10 years cured)


Fig. 13 K4, 95,000 (10 years cured)


Fig. 15 K4, 910,000 (10 years cured)


123

development of K4 and K12 samples was evident at 10,000

and 2,000 magnification after 123 months.

To determine the relationships between mineralogy and

macroscopic geotechnical parameters, Chryssochoou

(2010) completed a detailed mineralogical analyses of

dredged material (DM) stabilized with lime, cement kiln

dust (CKD), class F fly ash and two cements that was cured

for 6 months. A direct correlation between mineralogy and

engineering parameters was not found, largely due to the

complexity of the system involved. Rajasekaran and Nar-

asimha (1996) showed crumbs of floccules associated with

soil–lime reaction products of the images of the samples

treated with lime.

Although, it is very hard to correlate microfabric to

engineering behavior quantitatively, some qualitative

comments can be made. Addition of lime results in an open

microfabric due to flocculation. The pozzolanic reactions

cause formation of cementitious crystals which in term

densify the fabric causing an increase in strength and a

decrease in strain at failure values. The densification of

microfabric is supported by the results of UCS tests with

nearly 50 times increase in elastic modulus values while

the UCS increase is in the range of 20 times.

X-ray diffractometer and ESEM analysis offered min-

eralogical proof of the changes which occur in kaolinite

clays when treated with lime. This mineralogical proof was

supported by changes in physical properties such as

Atterberg limits, compressive strength measured on the

same soil.

Conclusions

The present study used three methods, UCS tests, pH

method and Atterberg limits to determine the optimum

lime content for the kaolinite samples. All the methods

used in the study gave expected and consistent results. The

performance and behavior of the lime-stabilized kaolinite

with 4% lime is similar to that of with 12% lime in both

short- and long-term tests. The long-term behavior of the

samples also supports the validity of the methods over

longer periods.

In the long-term, the unconfined compressive strength of

lime-stabilized kaolinite increases continuously although

the rate of increase is less in later years. The increase in

strength of lime-stabilized samples is very high compared

to that of the natural kaolinite clay samples. The uncon-

fined compressive strength of natural kaolinite was

125 kPa; this increased to 2,640 kPa for lime-stabilized

kaolinite samples cured for 123 months, which is 21 times

the original value. Since ion exchange, flocculation and

carbonation reactions between lime and clay samples occur

in a short time, the reason for the strength increase lime-

stabilized kaolinite even at 123 months is probably due to

pozzolanic reactions. Tests on long-term cured kaolinite

samples with 12% of lime showed similar results, with a

UCS value around 20 times higher than the initial strength

of natural kaolinite. Increasing lime content to a value of

three times of optimum did not improve the long-term

strength behavior. These increases can also be explained by

pozzolanic reactions. XRD and ESEM studies also showed

that laumontite minerals are present in both K4 and K12

samples. A typical CASH named Laumontite has been

detected in K4 and K12 samples after 10 years of curing.

The fabric and structure of cementation minerals did not

change after 10 years curing under the controlled labora-

tory conditions.

The strength improvement of nearly 2.5 times with

respect to 28 days curing is attributed to maturity of

cementitious minerals. The strength increases after

1 month and 3 months are very important because many

lime stabilization design methods uses the 28-day strength’

Fig. 16 K4, 9350 (10 years cured)

Fig. 17 K12, 9350 (10 years cured)


123

as a design value. The strength in the long-term is found to

be around 3 times higher compared to that of 28-day

strength for both K4 and K12 samples. This suggests that

classical design parameters are very conservative in terms

of long-term lime stabilization behavior.

Chemical stabilization of clay soils with lime is an

environmentally friendly, economical and technically fea-

sible method for road construction. The present study has

demonstrated the fact that the cementitious minerals, which

contributed to strength, formed in the short-term are still

available and maintain their original fabric and structure

without deterioration.

References

Baykal GI (1989) The effect of micro morphological development on

the elastic moduli of fly-ash, lime stabilized bentonite. PhD

dissertation, Louisiana state university

Bowles JE (1992) Engineering properties of soils and their measure-

ments, 4th edn. McGraw-Hill, NY

Broderick GP, Daniel D (1990) Stabilizing compacted clay against

chemical attack. J Geotech Eng Div 116(10):1549–1567

Chryssochoou M, Grubb DG, Drengler KL, Malasavage NE (2009)

Stabilized dredged material III: a mineralogical perspective.

J Geotech Geo-environ Eng 136:1037

Clare KE, Crunchley AE (1957) Laboratory experiments in the

stabilization of clays with hydrated lime. Geotechnique

VII:97–111

Croft JB (1964) The process involved in the lime stabilization of clay

soils. Australian research board, vol 2, part 2, pp 1169–1203

Doty R, Alexander ML (1968) Determination of strength equivalency

for design of lime-stabilized roadways, Report No. FHWA-CA-

TL-78-37

Eades JE, Grim RE (1963) A quick test to determine lime

requirements for lime stabilization. Highway research board,

Record No. 139, pp 61–72

Evans P (1998) Lime stabilization of black clay soils in Queensland,

Australia. Presentation to the National Lime Association Con-

vention, San Diego, California

Glenn G, Handy RL (1963) Lime–clay mineral reaction products.

Highway research record, No. 29, pp 70–83

Herle V (2005) Performance of lime improved railway sub-grade after

5 years in service. Tremti, International conference. Paris C026,

pp 1–6

Hilt GH, Davidson DT (1960) Lime fixation in clayey soils, Highway

Research Board, Bul. No. 262, pp 20–32

Hilt GH, Davidson DT (1961) Isolation and investigation of a lime–

montmorillonite crystalline reaction product. Highway Research

Board, Bul. No. 304, pp 52–64

Kavak A (1996) The behavior of lime stabilized clays under cyclic

loading. PhD dissertation, Bogazici University

Kavak A, Akyarlı A (2007) A field application for lime stabilization.

Eng Geol 51-6:987–997

Ladd CC, Moh ZC, Lambe TW (1960) Recent soil–lime research at

the Massachusetts I.T. Highway Research Board, Bul. No. 262,

pp 64–85

Little DN (1999) Evaluation of structural properties of lime stabilized

soils and aggregates, vol 1: summary of findings prepared for the

national lime association

Locat H, Berube M, Choquette M (1990) Laboratory investigations on

the lime stabilization of sensitive clays shear strength develop-

ment. Can Geotech J 27:294–303

Mitchell JK, El Jack SA (1964) The fabric of soil cement and it’s

formation, 14. In: National Conference on clays and clay

minerals, proceedings of the 14th national conference on clays

and clay minerals, pp 364–366

Neubauer CH, Thompson MR (1972) Stability properties of uncured

lime-treated fine grained soils. In: Highway research record 281,

HRB, National Research Council, Washington, DC

Nunez WP, Lovato RS, Malysz R, Ceratti AP (2005) Revisiting

Brazilian State Road 377: a well-succeeded case of lime-

stabilized road base. Tremti International conference, Paris

C036, pp 1–11

Rajasekaran G, Narasimha R (1996) Lime migration studies in marine

clays. Ocean Eng 23(4):325–355

Shackelford CD, Daniel D (1991) Diffusion in saturated soils. ASCE

J Geotech Eng 117:467–503

Sides G, Barden L (1971) The microstructure of dispersed and

flocculated samples of kaolinite, illite and montmorillonite. Can

Geotech J 8(3):391–399

Solanki P, Hhoury N, Zaman MM (2009) Engineering properties and

moisture susceptibility of silty clay stabilized with lime, class C

fly ash, and cement kiln dust. J Mater Civil Eng 21:749–757

Taylor WH, Arman A (1960) Lime-stabilization using preconditioned

soils. Transportation research board, Bul. No. 262, pp 1–13

Transportation Research Board (1987) Transpotation research board

state-of-the-art report No.5, Washington, DC

Uddin et al (1997) Cross-reference Little DN (1999)


123

Date post:	03-Dec-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Long-term behavior of lime-stabilized kaolinite clay

Documents