THE COMBINED EFFECT OF CEMENT AND LIMESTONE POWDER ON THE STABILIZATION OF SULFATE-BEARING SALINE SOIL IN PAVEMENT CONSTRUCTION

(1)

AND LIMESTONE POWDER ON THE STABILIZATION OF SULFATE-BEARING SALINE SOIL

IN PAVEMENT CONSTRUCTION

Ayazhan Bazarbekova, Bachelor of Engineering

Submitted in fulfilment of the requirements for the degree of Master of Science in Civil & Environmental Engineering

School of Engineering and Digital Sciences Department of Civil & Environmental Engineering

Nazarbayev University

53 Kabanbay Batyr Avenue, Nur-Sultan, Kazakhstan, 010000

Supervisor: Chang-Seon Shon Co-Supervisor: Jong Ryeol Kim

April 2021

(2)

I hereby, declare that this manuscript, entitled “The Combined Effect of Cement and Limestone Powder on the Stabilization of Sulfate-Bearing Saline Soil in Pavement Construction”, is the result of my own work except for quotations and citations which have been duly acknowledged.

Moreover, some components of this manuscript come from the conference paper, entitled

“Potential of Limestone Powder to Improve the Stabilization of Sulfate-contained Saline Soil”, written by me as a part of my graduate studies and published in IOP Conference Series:

Materials Science and Engineering. I also declare that, to the best of my knowledge and belief, it has not been previously or concurrently submitted, in whole or in part, for any other degree or diploma at Nazarbayev University or any other national or international institution.

--- Name: Ayazhan Bazarbekova Date: April 11, 2021

(3)

Abstract

For the modern world with rapidly developing infrastructure, the construction of high- quality roads has always been an issue of primary concern. In this sense, one of the significant factors influencing road pavement's quality and performance is the stability of subgrade material, i.e., soil stability. Therefore, to construct high-quality pavements, the soil must meet specific standards for its mechanical properties and durability parameters. However, the geotechnical properties of soil are determined by soil origin, soil mineralogy, and local environmental conditions, thus, can vary considerably from area to area. For instance, excessive heave occurs in pavements constructed on sulfate-bearing saline soils, the most prevalent soils in Kazakhstan, Central Asia. Salt whiskers in such soils create crystallization pressure that leads to high localized stresses and non-uniform movement of structures in soil. To improve the poor quality of soil and meet the desired end performance criteria in such a pavement construction, stabilization of soil is required, a process that presents the treatment of soil with chemical additives such as cement, lime, fly ash, and calcium chloride, also named as traditional stabilizing agents.

Since soil stabilization is a highly significant issue in constructing both buildings and roads, there has been increasing interest in this topic among researchers. The majority of papers have focused on utilizing the above-mentioned traditional binders and evaluating their effect on soil stabilization. However, less focus has been set on the utilization of recently developed non-traditional stabilizers, such as cement kiln dust, blast furnace slag, and limestone powder.

In this research, therefore, limestone powder, an alternative soil stabilizing material, was used in combination with traditional cement, and its potential performance in the stabilization of sulfate-bearing saline soil was evaluated.

For this purpose, silty sand containing high sulfate and chloride levels was stabilized by 4%, 6%, and 8% pure cement contents and 2%, 4%, and 6% cement contents combined with 2%, 4%, and 6% limestone powder contents. Optimal proportions for mix design were chosen, and series of laboratory tests were conducted to evaluate the improvement in materials characteristics and geotechnical properties of the stabilized soil samples. Material characteristics studied in this research are mineralogy, cation and anion analysis, and pH.

Geotechnical properties include Atterberg limits, optimum moisture content-dry density relationship, unconfined compressive strength, shear strength, friction angle, cohesion, resilient modulus, California bearing ratio, three-dimensional swelling, and dielectric constant.

(4)

Accoroding to experimental results, limestone powder, when added to the cement- treated sulfate-bearing saline soil, improves soil’s mechanical properties and enhances soil durability parameters. Mainly, it decreases soil plasticity, improves soil strength parameters, enhances soil stability, and reduces volumetric swelling and soil moisture susceptibility. Along with the stabilization of soil in terms of mechanical properties and durability parameters, limestone powder, as an industrial waste material, also benefits the environment and economy.

(5)

DEDICATION

I dedicate my graduate thesis work to my parents, who have nursed me with love and affection, supported my decisions, and encouraged me to strive for my goals.

(6)

Acknowledgments

First and foremost, I would like to express my uttermost gratitude to my supervisor, Professor Chang-Seon Shon, and co-supervisor, Professor Jong Ryeol Kim. Their consistent supervision and direction throughout my studies have allowed me to successfully write this graduate thesis and complete the Master of Science program. I am deeply grateful for all the hours of discussion, valuable recommendations, the feedback they have offered me, and the significant financial support they have provided to obtain the necessary laboratory equipment and materials.

Secondly, I would like to thank Construction Materials Laboratory, Soil Mechanics Laboratory, Water Treatment Laboratory, and Core Facilities at Nazarbayev University for providing special equipment used for soil testing.

Moreover, I would also like to express my gratitude towards the research team and special thanks to Aizhan Kissambinova for their great help in the experimental program.

Without their significant involvement, the research could not have been successfully conducted.

Finally, I would like to show my indebtedness to the administration staff at Nazarbayev University, School of Engineering and Digital Sciences, whose efforts a lot of the time go unnoticed. Particularly, I am grateful for the provided safety measures and well-organized campus access during the COVID-19 pandemic.

(7)

List of Abbreviations

C-A-H Calcium aluminate hydrate

CBR California bearing ratio

C-S-H Calcium silicate hydrate

DC Dielectric constant

DD Dry density

DST Direct shear test

EC Electrical conductivity

LL Liquid limit

LSP Limestone powder

MC Moisture content

MDD Maximum dry density

MR Resilient modulus

OMC Optimum moisture content

OPC Ordinary Portland cement

PI Plasticity index

PL Plastic limit

PSD Particle size distribution

SC Sulfate concentration

UCS Unconfined compressive strength

XRD X-ray diffraction

XRF X-ray fluorescence

(10)

List of Tables

Table 3.1: Basic soil characterization ... 21

Table 3.2:Chemical composition of the tested soil, OPC, and LSP ... 23

Table 3.3: Mix design ... 24

Table 4.1: Cation and anion analysis of the stabilized soil ... 33

Table 4.2: Residual strength of the control and the stabilized soil after 3-D swelling test ... 49

(11)

List of Figures

Figure 2.1: Determination of soil stabilization type for varying PI values and sulfate

concentrations ... 15

Figure 2.2: Neutralization, cation exchanged, flocculation and agglomeration in soil stabilization mechanism [8, 25] ... 19

Figure 3.1: Soil gradation ... 20

Figure 3.2: XRD patterns for the tested soil ... 21

Figure 3.3: PSD for OPC and LSP ... 22

Figure 3.4: XRD patterns for OPC ... 23

Figure 3.5: XRD patterns for LSP ... 23

Figure 3.6: Edge-Grim test results for OPC and LSP ... 25

Figure 3.7: The experimental program to evaluate the effect of OPC and LSP on the stabilization sulfate-bearing saline soil ... 26

Figure 3.8: Determination of LL ... 27

Figure 3.9: Determination of PL ... 27

Figure 3.10: Soil-stabilizer mixing at the OMC... 27

Figure 3.11: The cylindrical soil specimen cured in the sealed conditions ... 28

Figure 3.12: UCS test of the soil specimen ... 29

Figure 3.13: DST of the soil specimen ... 30

Figure 3.14: 3-D swell test preparation ... 31

Figure 3.15: 3-D swell measurements ... 31

Figure 3.16: DC measurement ... 32

Figure 3.17: Development of salt crystallization during drying in wetting-drying cycles ... 32

Figure 4.1: pH of the control and the stabilized soil ... 34

Figure 4.2: XRD patterns for the stabilized soil ... 35

Figure 4.3: Atterberg limits of the control and the stabilized soil samples ... 36

Figure 4.4: Effect of LSP content on PI ... 37

Figure 4.5: OMC-MDD determination for the tested soil ... 38

Figure 4.6: OMC and MDD of the control and the stabilized soil samples ... 39

Figure 4.7: Effect of LSP on MDD ... 39

Figure 4.8: UCS of the control and the stabilized soil ... 40

Figure 4.9: Effect of LSP on 7-days UCS ... 41

Figure 4.10: Effect of LSP on 28-days UCS... 41

Figure 4.11: Cohesion of the stabilized soil... 42

Figure 4.12: Effect of LSP on 7-days cohesion ... 43

Figure 4.13: Effect of LSP on 58-days cohesion ... 43

Figure 4.14: Friction angle of the control and the stabilized soil... 44

Figure 4.15: MR and CBR of the control and the stabilized soil ... 45

Figure 4.16: 3-D swelling of the control and the stabilized soil ... 47

Figure 4.17: Moisture content of the control and the stabilized soil during 3-D swelling ... 48

Figure 4.18: 28-days and 58-days 3-D swelling of the control and the stabilized soil ... 49

Figure 4.19: Dielectric constant of the control and the stabilized soil ... 50

Figure 4.20: 28-days and 58-days dielectric constant of the control and the stabilized soil .... 51

(12)

Chapter 1. Introduction

1.1. Overview and Problem Statement

Soil's geotechnical properties differ from area to area, depending on the origin of soil, environmental conditions of a region, and soil treatment processes. However, in the construction industry, the soil must satisfy specific standards for its engineering properties, such as plasticity, deformability, strength, and durability parameters. It acts as an engineering medium and a foundation for most structures such as buildings, bridges, roads, etc. Mainly, in pavement construction, the soil is at a subgrade level. Depending on soil properties, road pavement can serve well for a long time, or it may fail after a short period with deformation and cracks developing on the surface of the pavement. In Kazakhstan, for example, heavy textured and saline soils occupy about 41% of the national territory [1]. The salt whiskers in sulfate-rich saline soils grow and create crystallization pressure, which leads to an increase in localized stresses and non-uniform movement of structures within the soil matrix.

Consequently, they result in eventual defects, such as excessive heave and breaking up of pavements constructed on such soils [2, 3].

To improve the poor quality of soils and meet the desired end performance criteria in pavement construction, soil’s geotechnical properties must be strengthened through the stabilization process; particularly chemical stabilization, in which soil is usually treated with chemical additives such as portland cement, fly ash, and calcium chloride [4, 5]. With lime, fly ash, and calcium chloride, cement is suggested as a traditional binder used for soil stabilization.

Cement stabilization of soil is the most common and reliable method for improving soil's mechanical properties such as shear strength and bearing capacity [6]. During the mixing process between soil and stabilizer, strong cations derived from stabilizing agents replace weak cations surrounding soil surface in a cation exchange process. This process leads to the formation of flocculated and agglomerated soil particles, contributing to higher surface tension and increased resistance against compaction. As a result, soil strength is improved [7, 8].

Moreover, in stabilizing soil with traditional calcium-based stabilizing materials, the hydration of cementitious material and pozzolanic reaction produce calcium aluminate hydrate (C-A-H) and calcium silicate hydrate (C-S-H) that continue increasing over time. Eventually, this reaction leads to the long-term improvement of soil's engineering properties [8, 9, 10].

In the past few years, researchers in this area have focused on utilizing industrial solid waste materials, also named by-products, in soil stabilization, such as limestone powder,

(13)

cement kiln dust, and slag. Particularly, limestone powder is one of the main by-products of the aggregate quarrying industry of Kazakhstan. In this sense, as it has been recently studied, limestone powder increases the bearing capacity and reduces weak soil's deformability [11, 12, 13]. Moreover, alternative stabilizers have positive economic and environmental effects, compared to traditional binders, which have limitations such as higher cost and CO₂ emissions to the atmosphere and landfills [9, 12].

Despite many successful studies on soil stabilization, they have most likely focused on evaluating the physical and mechanical properties of stabilized soil, not durability parameters.

Furthermore, existing studies have mainly focused on the effect of a single additive on soil stabilization, particularly traditional agents such as cement, lime, and fly ash. They overlooked the effect of their combination with non-traditional binders such as limestone powder to benefit in economic and environmental aspects. Moreover, little data are available on cement and limestone powder's combined effect on the stabilization of soils containing high salt and sulfate levels. Therefore, the present research aims to study the potential of limestone powder and its combination with cement in the mitigation of salt crystallization, improvement of geotechnical properties, and enhancement of long-term durability of sulfate-bearing saline soils.

1.2. Research Objective and Scopes

First and foremost, this study aims to use limestone powder as a soil stabilizing agent combined with traditional cement. In this sense, limestone powder, which is one of the main by-products of Kazakhstan's aggregate quarrying industry, has positive economic and environmental effects, as it is cheaper industrial waste material and emits less CO₂ to the atmosphere and landfills, comparing to the traditional binders [12]. Moreover, despite many comprehensive studies on soil stabilization and soil stabilizing materials, durability parameters of stabilized soils using limestone powder have not been evaluated as much as physical and mechanical properties. Moreover, little data are available on cement and limestone powder's combined effect on the stabilization of sulfate-bearing saline soils. Therefore, this research's primary objective to evaluate the combined effect of cement-limestone powder blend on the improvement of geotechnical properties and durability parameters of sulfate-bearing saline soil.

For this purpose, the singificant tasks covered in this thesis are summarized as the following:

 Literature review: (i) review the concept of soil stabilization, (ii) identify soil stabilizing materials, and limestone powder as an alternative stabilizer, and (iii) review soil stabilization mechanism.

(14)

 Development of an integrated experimental program: establish experimental programs to evaluate the combined effect of cement and limestone powder on the stabilization of sulfate-bearing saline soil in pavement construction.

 Test result analysis: (i) conduct essential soil and materials characterization, (ii) evaluation of geotechnical properties of stabilized soil, and (iii) evaluate durability of stabilized soil.

1.3. Thesis Structure

This thesis consists of five chapters. Background information on soil stabilization and its application in pavement construction is presented in Chapter 1. Moreover, it explains the research's novelty, the motivation behind the thesis, and the paper's main objectives and tasks.

Further, Chapter 2 presents a literature review, particularly, it collects, reviews, and integrates all relevant information on soil stabilization, its mechanism, and stabilizing agents, including a deeper description of limestone powder. Chapter 3 describes the materials, mix design, and methodology used in this research. This section contains the detailed characterization of materials used in the experiment, notably sulfate-bearing saline soil, cement, and limestone powder. Moreover, it presents the mixtures designed for the investigation and the experimental program itself. Following this, Chapter 4 presents the results of the tests and discusses these findings. Particularly, the effect of cement and limestone powder on the stabilization of sulfate- bearing saline soil in terms of the geotechnical properties and durability parameters of stabilized mixes are analyzed and evaluated. Conclusions regarding the findings described in the previous section are drawn in Chapter 5. Moreover, some recommendations for further studies are given in the last chapter.

(15)

Chapter 2. Literature Review

A literature search, as the first step of the research, is completed under this task and presented in this chapter. The focus of this task is to thoroughly collect, review, and integrate all relevant information on the topic of this paper. The information comes from a variety of sources including books, published journals, reports and memos, unpublished reports and memos, a database of industries and agencies, and etc. The compilation of such a comprehensive listing provides a strong foundation for the research on which to formulate and to implement the results of the work.

The following paragraphs present a comprehensive literature review on the following topics: (a) soil stabilization; (b) soil stabilizing materials, including (c) a deeper description of limestone powder as an alternative stabilizer; and (d) soil stabilization mechanism.

2.1. Soil Stabilization

Weak and soft subgrade soils often result in poor performance and a short lifetime of road pavements constructed on these soils. To improve the poor quality of soils and meet the desired end performance criteria in pavement construction, it is necessary to stabilize the soil.

The process of soil stabilization aims to enhance soil's engineering properties physically, mechanically, and chemically [14]. Physical stabilization of soil refers to the modification of soil in terms of its particle size distribution or plasticity by adding or subtracting different soil fractions, also named blending to obtain the material meeting the specified soil gradation or soil plasticity, respectively [15]. Mechanical type of soil stabilization includes various techniques, such as compaction, wetting-drying cycles, and fiber reinforcement, applied to achieve the modification of soil porosity, mitigation of free swelling of the soil, and improvement of mechanical properties of soil and soil stability, respectively [8, 10, 15]. Though physical and mechanical types of soil stabilization are essential techniques involved in the material selection and preparation stages, the term “stabilization” in pavement construction mainly refers to soil's chemical treatment [9, 15]. Hence, this paper focuses on chemical stabilization of soil, also named additive or binder stabilization of soil, a process of improving the engineering properties of soil achieved by the addition of chemical stabilizers such as lime, Portland cement, and calcium chloride [4, 5]. In roadway construction, chemical stabilization with calcium-based stabilizing materials (CBSMs) such as lime and cement can enhance many of the subgrade soil's engineering properties. These include compressive strength, bearing capacity, resilient

(16)

modulus, shear strength, soil stability, plasticity, and long-term durability expressed by mitigation of volumetric swelling and reduction of moisture susceptibility.

The selection criteria for stabilizing sulfate-rich soils have been described by the Texas Department of Transportation [16]. The procedure involves two main steps: (a) risk assessment and (b) soil exploration. Risk assessment addresses the major question about the potential risk for sulfate-induced heave on pavements constructed on the selected area and is performed by identifying soil formation, soil mineralogy, basic soil properties, local climatic characteristics, and drainage features. Soil exploration is a determination of the sulfate concentration of soil.

Sulfate concentration is the main criterion for the selection of soil treatment types classified as (a) traditional, (b) modified, and (c) alternative. The precise determination of soil stabilization type is summarized in Figure 2.1.

Figure 2.1: Determination of soil stabilization type for varying PI values and sulfate concentrations

Plasticity Index (PI)

PI>15

Moderately to high expansive soils

PI≤15

Minimally expansive soils

Sulfate concentration (SC) in ppm Sulfate concentration (SC) in ppm

SC≤3000 3000<SC≤8000 SC>8000

Traditional treatment

Alternative treatment Modified

treatment

SC≤3000 SC≥3000

Low potential for sulfate-induced

heave;

therefore, adequate mixing and moisture are

sufficient

Stabilization with calcium- based additives Regular

mix design with a minimum 24 hours of

mellowing

Single lime application;

mellowing;

additional moisture treatment

Removing and replacing

sulfate soil;

blending in non-plastic

soils

(17)

2.2. Soil Stabilizing Materials

Chemical stabilization of soil is achieved through the addition of stabilizing materials, which include a wide array of binders such as Portland cement, lime, industrial solid waste materials, polymers, fibers, reagents, bitumen, and etc. Generally, stabilizing agents are classified as traditional and non-traditional additives. As a conventional binder, cement is the most commonly used stabilizing agent for soil treatment, as it allows to achieve the most effective soil strength improvement [6]. Along with cement, examples of traditional binders include calcium-based materials such as lime, fly ash, and calcium chloride. When mixed in an aqueous phase, the soil-additive mixture undergoes immediate chemical reactions, such as cation exchange, flocculation, and agglomeration. These processes contribute to an instant improvement of soil properties and prolonged time-dependent chemical reactions, such as hydration and pozzolanic reaction, which provide a progressive increase of soil strength [8, 9, 17].

The adverse environmental effects of high CO2 emissions to the atmosphere and landfills, and the high cost of traditional binders, have motivated researchers to propose new economically friendly and cost-effective stabilizing materials. Non-traditional stabilizers, mainly quarry by-product materials, such as cement kiln dust, blast furnace slag, and limestone powder, are the most recently developed materials to stabilize soil stabilization [9]. Like the conventional calcium-based stabilizing materials, the cation exchange, flocculation, agglomeration, hydration, and formation of cementitious materials (C-S-H and C-A-H) are the main mechanisms contributing to the enhancement of the geotechnical properties of stabilized soils.

Non-calcium-based stabilizing materials, such as polymers and fibers, are also used in the stabilization of soil. Polymers, as it has been reported, mitigate soil liquefication, enhance resistance against moisture susceptibility, and improve soil reaction to weathering actions [18, 19]. Fiber reinforcement, comparing to stabilization using calcium-based binders, provides more ductile behavior of stabilized soil [20]. In addition, the combination of calcium-based and non-calcium-based additives is also a widespread practice implemented in the stabilization of soil.

2.2.1. Limestone Powder as an Alternative Stabilizer

Environmental and economic limitations associated with the traditional stabilizing agents have become a motivation for the researchers to propose the use of non-traditional binders. Particularly, limestone powder, one of Kazakhstan's main by-products from the

(18)

aggregate quarrying industry, is an environmentally friendly and cost-effective alternative to the traditional stabilizing additives. Generally, as for the calcium-based stabilizing materials, the major mechanisms behind soil properties' improvement are cation exchange, flocculation, agglomeration, hydration, and C-A-H and C-S-H formation. It has been reported that limestone powder, used for stabilizing fine-grained soils, increases soil strength and bearing capacity and reduces deformability of weak soil [11, 12, 13]. Improvement of bearing capacity and reduction of deformability of weak soil leads to a reduction in thickness of pavement layers, which means significant savings in construction materials and, hence, significant savings in construction cost.

Furthermore, it has been studied that, when mixed with expansive soil, limestone powder enhances strength parameters, reduces plasticity index and liquid limit, increases liquid limit, and mitigates volumetric swelling of expansive soils [21, 22]. Moreover, along with improving soil properties and saving the construction cost, limestone powder, an industrial waste product, is an environmentally sustainable stabilizing agent due to lower CO2 emissions to the atmosphere and landfills.

2.3. Soil Stabilization Mechanism

Generally, the primary mechanism of soil stabilization using calcium-based stabilizing agents involves (a) hydration, (b) cation exchange, (c) flocculation and agglomeration, and (d) pozzolanic reaction [8, 9, 17]. Cation exchange, followed by flocculation and agglomeration, occurs immediately after soil and stabilizer are mixed in the aqueous environment. It, thus, contributes to the instant improvement of soil properties such as plasticity and short-term strength. Hydration process and pozzolanic reaction, which result in the formation of C-S-H and C-A-H, lasts longer, months or even years after soil-stabilizer mixing, and provide the long- term enhancement of the geotechnical properties of soil such as continuous improvement of strength parameters, increase in resilient modulus and modulus of elasticity, and reduction of free swell. Moreover, (e) potential carbonation resulting from chemical reactions involved in cement- and lime-treatment of soil has been identified in hot, dry climates with difficult curing control conditions [23]. Strength improvement and plasticity reduction, achieved through soil stabilization, can reverse when stabilized soil is exposed to carbonation. Since carbonation entirely depends on the environmental conditions, stabilized soil specimens are recommended to be cured in sealed containers [24].

The first two processes involved in the stabilization mechanism are cation exchange and formation of flocculated and agglomerated particles, which take place immediately after soil- stabilizer mixing. As a result, they provide instant improvement of soil properties such as

(19)

plasticity and short-term strength. The surface of sulfate-containing saline soil is negatively charged, and it attracts positively charged ions from the stabilizing agent when soil and stabilizer are mixed in the presence of water. This process provides neutralization of the negative charges surrounding the soil surface. Higher valence cations (e.g., Ca²⁺) from the stabilizer, particularly cement and limestone powder, replace lower valence cations (e.g., H⁺ and Na⁺) surrounding soil surface through a phenomenon called cation exchange. This phenomenon leads to the flocculation and agglomeration processes that are associated with the aggregation of soil particles. Moreover, the increased electrolyte concentration in the system due to the cation exchange reduces the thickness of the electrical diffuse double layer (DDL) at the particle-liquid interface. As the thickness of the DDL reduces, the spacing between individual soil particles also reduces, further enhancing the formation of flocculated soil particles. Flocculation and agglomeration of soil particles result in higher surface tension, increased cohesion between soil particles, reduced plasticity, and improved soil strength, particularly at an early age, as the cation exchange occurs instantaneously [8]. The neutralization and cation exchange processes are visualized in Figure 2.2.

The hydration process and pozzolanic reaction last longer, thus, provide progressive improvement of soil strength parameters. Hydration, particularly OH^-, increases soil pH, promoting the dissolution of aluminates and silicates from the soil matrix. Calcium, available from the stabilizing agent, reacts with alumina and/or silica, and free water in pozzolanic reaction, and produces calcium aluminate hydrate (C-A-H) and calcium silicate hydrate (C-S- H). The process is also referred to as solidification. It improves soil strength and stiffness due to the binding nature of the cementation gels produced in the pozzolanic reaction [25].

Moreover, the pozzolanic reaction takes place months and years, as long as sufficiently high pH is obtained for the dissolution of aluminates and silicates from the soil matrix [26]. The amount of these compounds increases over time; therefore, they continuously contribute to the long-term improvement of soil strength [8].

(20)

Figure 2.2: Neutralization, cation exchanged, flocculation and agglomeration in soil stabilization mechanism [8, 25]

When sulfate-containing soil is treated with cement and/or lime, its pH increases to above 12.0, promoting the dissolution of soil particles and the release of aluminum and sulfate into the system. Calcium is released from the calcium-based stabilizing material, and water is supplied as a source of soil stabilization and soil mixing. As a result, the aluminum-sulfate- calcium-water reaction produces the ettringite minerals, which can hold a large amount of water within the material, resulting in its expansion [27]. The stabilization with low-calcium-based additives, such as cement, fly ash, ground granulated blast furnace slag, is recommended to reduce the selling potential in sulfate-bearing saline soils over the use of high-calcium-based benders, such as lime [28, 29]. In addition, the use of a non-calcium-based stabilizer, particularly metakaolin-based geopolymer, is proposed for the mitigation of free swelling in sulfate-rich soils [30].

(21)

Chapter 3. Materials, Mixtures, and Methods

Based on the comprehensive literature review, an adequate selection of materials, design of optimum mixtures, and development of the experimental program has been performed and described in this chapter.

3.1. Materials 3.1.1. Soil

In the current research, a sulfate-bearing saline soil collected from West Kazakhstan was studied. According to the AASHTO classification system, the tested soil is classified as silty or clayey gravel and sand [31]. The tested soil's gradation is obtained according to the TxDOT specification and is shown in Figure 3.1 [32]. Basic soil properties, including AASHTO soil classification, Atterberg limits, and optimum moisture content-maximum dry density relationship, were determined as summarized in Table 3.1. Moreover, the tested soil's chemical properties, particularly cation and anion analysis, performed by the Dionex ICS-600 Ion Chromatography System, and pH of the soil, measured by the Tex-128-E test method, are also presented in the Table 3.1 [33]. Atterberg limits test results (liquid limit, plastic limit, plasticity index) show low plasticity value for the soil, resulting in lower volumetric swelling compared to that of clayey soil. However, high sulfate and chloride concentrations in the tested soil are expected to promote salt crystallization in the soil matrix, which results in pavement failure due to sulfate-induced heave.

Figure 3.1: Soil gradation

(22)

Table 3.1: Basic soil characterization Geotechnical properties

Property Value Property Value

AASHTO classification A-2-4(0) Liquid limit (%) 19.16

Optimum moisture content (%) 10.80 Plastic limit (%) 16.67 Maximum dry density (kg/m³) 1941.00 Plasticity index (%) 2.49

Chemical properties

Cations (ppm) Calcium Sodium Potassium Magnesium

6983.81 6682.37 801.37 664.87

Anions (ppm) Sulfate Chloride

16931.00 10681.98

pH 6.32

The mineralogical analysis performed using XRD is shown in Figure 3.2. Particularly, the soil mainly consists of quartz (SiO2), gypsum (CaSO4H2O), and calcite (CaCO3). The presence of gypsum provides high sulfate concentration in the soil, which is expected to result in salt crystallization, causing poor soil performance and damage due to sulfate-induced heave.

Figure 3.2: XRD patterns for the tested soil

3.1.2. Stabilizers

Ordinary Portland cement and its combination with limestone powder were used as stabilizing agents for the experimental work. Particularly, limestone powder was obtained by crushing locally collected limestones using a jaw crusher and then ground the material using a ball mill. The particle size distribution of the stabilizers is shown in Figure 3.3. According to

(23)

it, LSP is relatively well-graded, thus, has coarser particles with 73% fines content, passing sieve No. 325 (45 μm), comparing to relatively uniformly graded OPC with 95% fines content.

Moreover, due to its poor uniform gradation, OPC has a lower content of very fine particles than well-graded LSP, which has approximately the same content of coarser and finer particles.

Grain size distribution of stabilizing agents, particularly, predominant fine particles, comparing to sand, allow these stabilizers to act as a filling material and contribute to binding of particles in the soil-stabilizer matrix, which is expected to result in increased cohesion, reduced plasticity, and improved strength at an early age.

Figure 3.3: PSD for OPC and LSP

The mineralogical analysis of cement and limestone powder obtained by XRD and chemical composition of the tested soil and stabilizers obtained by XRF are presented in Figures 3.4 and 3.5 and Table 3.2, respectively. The presence of calcium silicate oxide and calcium silicate in OPC and calcite in LSP is expected to promote the pozzolanic reaction in soil- stabilizer mixing, thus contributing to prolonged strength improvement.

(24)

Figure 3.4: XRD patterns for OPC

Figure 3.5: XRD patterns for LSP

Table 3.2:Chemical composition of the tested soil, OPC, and LSP

Compound Soil (%) OPC (%) LSP (%)

SiO2 20.22 21.05 11.16

Al2O3 4.55 3.79 3.62

Fe2O3 11.68 4.47 9.44

Na2O 0.84

0.43 0.60

K2O 4.16 2.58

MgO 1.66 1.77 1.00

CaO 39.78 64.48 68.54

TiO2 1.61 - 1.23

SO3 10.60 2.88 0.70

MnO 0.66 - 0.62

Cl 3.70 0.01 0.59

(25)

3.2. Mixtures

A total of 10 mixtures, including a control sample (soil only), were set up to achieve the research objectives. Stabilizers contents were determined using the Edges-Grim test method described in TxDOT specification [34]. The Edges-Grim test calculates the recommended percentage of stabilizing agents. The minimum percentage of additive required to achieve the pH of soil-stabilizer mixture equal to 12.4 indicates a sufficient pH value for pozzolanic reaction to take place. According to test results, shown in Figure 3.6, the minimum recommended cement and limestone powder contents are 4% and 2%, respectively. Four different cement contents (2, 4, 6, and 8%) and three different limestone powder contents (2, 4, and 6%) were selected. Different combinations of cement and limestone powder contents were designed in order to evaluate the effect of each additive on the stabilization of sulfate-bearing saline soil. Based on the evaluation, select the optimum stabilizer content and mixture. Mixtures were designed as follows:

Table 3.3: Mix design

No. OPC LSP Notes

1 - - Control (soil only)

2 2% 2%

3 2% 4%

4 2% 6%

5 4% -

6 4% 2%

7 4% 4%

8 6% -

9 6% 2%

10 8% -

(26)

Figure 3.6: Edge-Grim test results for OPC and LSP

3.3. Methods

The study's experimental program can be categorized into basic material characterization, mix design and sample preparation, determination of the material characterization of the stabilized soil samples, evaluation of the geotechnical properties, and the durability of soil-cement-limestone powder mixtures. The detailed procedure with specified tests and test methods is presented in Figure 3.7.

(27)

Figure 3.7: The experimental program to evaluate the effect of OPC and LSP on the stabilization sulfate-bearing saline soil

3.3.1. Determination of Basic Material Characterization

Soil gradation was performed in accordance with the Tex-110-E specification. Based on it, the following grain sizes are included in the analysis: 1 in, 3/8 in, sieve No. 4, sieve No. 10, sieve No. 60, and sieve No. 200. The Atterberg limits (LL, PL, and PI) of the tested soil were determined as described in the Tex-104-E, Tex-105-E, and Tex-106-E specifications and shown in Figures 3.8 and 3.9 [35, 36, 37]. The optimum moisture content-maximum dry density relationship for the soil was obtained, according to the procedure provided in the ASTM D698- 12e2 specification [38]. The chemical analysis of the tested soil, particularly cation and anion analysis, was conducted by the Dionex ICS-6000 Ion Chromatography System. Moreover, the pH of the natural soil was measured based on the Tex-128-E test method [33].

The mineralogical compositions of both the tested soil and the additives were obtained by means of XRD patterns, which were determined using Rigaku SmartLab and analyzed using MDI Jade 6. The chemical compositions of the soil, OPC, and LSP were determined using AxiosmAX XRF spectrometer by PANalytical.

Soil Stabilizer

Gradation: Tex-110-E Minerology: XRD

Chemical composition: XRF Ion chromatography pH: Tex-128-E

Atterberg limits: Tex-105-E, 104-E, and 106-E

OMC & MDD: ASTM D698-12e2 USC: ASTM D1633-17

DST: ASTM D3080

PSD: Mastersizer 3000 Minerology: XRD

Chemical composition: XRF Determination of basic soil and

stabilizer characterization

Mix design and sample preparation

Geotechnical properties Durability

Atterberg limits: Tex-106-E OMC & MDD: ASTM D698-12e2 USC: ASTM D1633-17

DST: ASTM D3080

MR and CBR: empirical relation

3-D swelling: TTI 3-D swell test Tube suction: Tex-144-E Evaluation

Material characterization of the stabilized soil

(28)

Figure 3.8: Determination of LL Figure 3.9: Determination of PL

3.3.2. Mix Design and Sample Preparation

A total of 10 mixtures, including a plain soil sample and nine soil mixtures containing different amounts and types of stabilizer, were designed as described in the previous sections.

As presented in Figure 3.10, the soil samples were mixed thoroughly with the different cement and limestone powder percentages at their corresponding optimum moisture contents.

Determination of OMC for varying stabilizer contents is described in the following sections.

After thoroughly mixing soil-stabilizer mixtures at their corresponding OMC, the cylindrical specimens (4 in. × 4.5 in. and 50 mm × 100 mm) were compacted with the standard Proctor compaction energy and cured in the sealed conditions for further testing, as shown in the Figure 3.11.

Figure 3.10: Soil-stabilizer mixing at the OMC

(29)

Figure 3.11: The cylindrical soil specimen cured in the sealed conditions

3.3.3. Determination of the Material Characteristics of the Stabilized Soil

The stabilized soil's material characteristics include cation and anion analysis, performed by the Dionex ICS-6000 Ion Chromatography System, pH, measured in accordance with the Tex-128-E test specification, mineralogical composition of the cement- and limestone powder-treated soil mixtures associated with XRD patterns of the samples [33]. The stabilized samples were cured for 7-days, and the samples exposed to 58-days volumetric swelling under capillary soak conditions were tested for the material characteristics. The 7-days cured samples represent the instant effect of the soil stabilization that indicates cation exchange and formation of flocculated and agglomerated particles, while the samples after 58-days volumetric swelling provide the realistic long-term durability assessment.

3.3.4. Determination of the Geotechnical Properties of the Stabilized Soil

Geotechnical properties determined in this study include Atterberg limits (LL, PL, PI), optimum moisture content, maximum dry density, 7-days and 28 days unconfined compressive strength, shear strength parameters such as friction angle and cohesion. Moreover, resilient modulus and California bearing ratio of stabilized mixtures were obtained through the empirical relation.

The Atterberg limits (LL, PL, and PI) were measured for both the plain and the stabilized soil samples. The optimum moisture content-maximum dry density relationship (M-D curve) of the natural soil was obtained as described in the previous section, and then OMC for the stabilized mixtures was calculated according to the soil-cement testing proposed by TxDOT in the Tex-120-E specification[39]. Equation 3.1 allows calculating nearer OMC for stabilized soil samples without running a new OMC-MDD relationship determination for each stabilizer content. % cement increase is considered as the total stabilizer content (%OPC + %LSP).

(30)

% molding water% OMC from MD curve0.25 (%  cement increase) (3.1) Maximum dry densities of soil-cement-limestone powder mixtures were calculated as shown in Equation 3.2. The moist density of specimen is a density of soil stabilized with cement or its combination with limestone powder, mixed at the corresponding optimum moisture content, and compacted with standard Proctor compaction energy.

1 100 Moist density Dry density

 OMC

 (3.2)

Wet mass Moist density

Volume

 (3.3)

As presented in Figure 3.12, the 7- and 28-day UCS of the natural soil and the stabilized mixtures were measured as described in the ASTM D1633-17 specification [40]. The 7- and 28-days cured cylindrical soil specimens of 50 mm diameter and 100 mm height were compressed at a loading rate equal to 1 mm/min under unconfined conditions, and the maximum load that the tested samples could withstand was determined as UCS. Shear strength, friction angle, and cohesion of the plain and the stabilized soil samples were determined through the Direct Shear Test according to the ASTM D3080 specification [41]. The 7-days cured samples and the samples exposed to 58-days volumetric swelling, including 28-days persistent swelling and 30-days wetting-drying cycles, were tested in the DST as shown in Figure 3.13.

Figure 3.12: UCS test of the soil specimen

(31)

Figure 3.13: DST of the soil specimen

Resilient modulus and California bearing ratio of the stabilized mixtures were obtained using the empirical relations provided in Equations 3.4, 3.5, and 3.6. CBR is a function of % particles passing No. 200 sieve (P200) and PI.

( ) 2555 0.64

MR psi  CBR (3.4)

( ) 1500 , 20

MR psi  CBR where CBR (3.5)

(%) 75

1 0.728 200

CBR  P PI

   (3.6)

3.3.5. Determination of the Durability of the Stabilized Soil

Durability assessment was performed by measuring the three-dimensional (3-D) swelling and dielectric constant of the plain and stabilized soil samples. The 3-D swelling test was conducted as described by Texas Transportation Institute in order to assess the volumetric expansion of the sample associated with the formation of ettringite minerals when cement- and limestone powder-treated sulfate-bearing saline soil is exposed to prolonged capillary suction [42]. The 3-D swell test was performed, as shown in Figures 3.14 and 3.15, by covering the 4- inch diameter and 4.5-inch height specimen with a wet towel and rubber membrane, placing filter paper and porous stone on the bottom and filter paper, plastic sheet, porous stones on the top of the sample. The specimen was placed in a container with deionized water, letting the water to soak through the sample for a particular period. The volumetric expansion of the sample was measured periodically. After the 28-days of continuous capillary soak, the samples

(32)

were exposed to wetting-drying conditions, performed by drying the samples for two days and placing them back in the container with water for three days, total in a 30-days period.

The residual unconfined compressive strength of the stabilized soil samples after the 3- D swell test was also determined. The residual UCS of the samples exposed to the volumetric swelling under capillary suction assesses the moisture susceptibility of the soil. The obtained strength value is considered as a more realistic approach since it imitates in-situ conditions and provides an assessment of long-term durability of the soil.

Figure 3.14: 3-D swell test preparation

Figure 3.15: 3-D swell measurements

The dielectric constant quantifies the moisture susceptibility of material and categorizes the material as “good, marginal, and poor”. DC value is determined in the Tube Suction test, described in the Tex-144-E test method [43]. The 4-inch diameter and 4.5-inch height specimen was covered with a rubber membrane, placed filter paper and porous stone on the bottom and filter paper, plastic sheet, and porous stone on the top, and exposed to continuous capillary suction conditions in the container. DC values for each sample were measured using a

(33)

percometer, as shown in Figure 3.16. Similar to the 3-D swelling test, the samples were exposed to wetting-drying cycles, total 30-day, after 28-days of continuous capillary suction. The effect of wetting-drying cycles is presented in Figure 3.17.

Figure 3.16: DC measurement

Figure 3.17: Development of salt crystallization during drying in wetting-drying cycles

(34)

Chapter 4. Test Results and Discussion

The results of the tests described in the previous sections are presented in this Chapter.

Mainly, Chapter 4 evaluates OPC and LSP's effect on the material characteristics of the stabilized soil, improvement of the geotechnical properties, the enhancement of its long-term durability of the soil-stabilizer mixtures.

4.1. Material Characterization of the Stabilized Soil

Material characterization of the stabilized soil samples includes cation and anion analysis, pH, and mineralogical composition of the soil-stabilizer mixtures. The 7-day cured samples and the samples exposed to 58-days volumetric swelling were tested in order to determine the short-term and the long-term effect of the soil stabilization, respectively. Cation and anion concentrations in the natural and the stabilized samples are shown in Table 4.1.

Table 4.1: Cation and anion analysis of the stabilized soil

Cation (ppm) Anion (ppm)

Ca²⁺ Na⁺ K⁺ Mg²⁺ SO42- Cl^-

Control 6983.81 6682.37 801.37 664.87 16931.00 10681.98 7-days

2%OPC+2%LSP 5740.32 5001.21 260.40 9.05 11712.34 7538.38 2%OPC+4%LSP 4841.45 4460.50 215.54 7.12 10098.22 6696.80 2%OPC+6%LSP 4936.73 4227.02 236.06 6.24 10221.67 6469.27 4%OPC 6710.86 6765.66 394.59 12.59 12951.56 10276.09 4%OPC+2%LSP 5030.45 4643.48 419.31 7.17 10404.88 7047.44 4%OPC+4%LSP 5259.05 4763.29 356.42 5.60 10519.02 7208.32 6%OPC 5684.40 4760.15 398.79 5.63 10143.65 7334.04 6%OPC+2%LSP 5893.82 4384.49 431.67 6.70 11064.09 6705.38 8%OPC 6264.56 4952.64 474.76 22.28 11631.59 7526.07

58-days

2%OPC+2%LSP 4859.41 1509.97 223.90 20.49 5628.18 1623.66 2%OPC+4%LSP 5451.18 3474.75 189.51 14.44 8262.55 3598.50 2%OPC+6%LSP 6795.18 3135.18 201.87 25.63 11833.37 3635.28 4%OPC 5900.93 4070.65 293.37 20.30 11286.48 4868.86 4%OPC+2%LSP 6152.33 3630.45 342.48 49.26 11038.46 4610.68 4%OPC+4%LSP 6031.01 3611.80 326.64 48.00 11145.30 4700.59 6%OPC 6569.17 3755.78 377.06 19.35 11596.63 5166.02 6%OPC+2%LSP 6552.37 3761.40 439.17 42.09 12062.76 5152.18 8%OPC 5666.99 3457.90 469.04 46.74 10647.34 4670.22

(35)

According to cation and anion analysis, OPC- and LSP-stabilization of sulfate-bearing saline soil reduce sulfate and chloride concentrations in the tested soil. As a result, the originally acidic soil with a pH value of 6.32 transforms into the alkaline material with a pH value almost twice the original one when stabilized with calcium-based OPC and LSP. pH values above 12.0 achieved in the 7-days cured samples stabilized with 4% OPC + 2% LSP, 4% OPC + 4% LSP, 6% OPC, 6% OPC + 2% LSP, and 8% OPC are sufficient to promote pozzolanic reaction in these mixtures, which contributes to the long-term strength improvement. However, pH values of the samples exposed to the 58-days volumetric swelling under capillary suction are slightly lower, comparing to those of the 7-days cured samples. The slight drop in pH value can be attributed to the potential carbonation of the OPC- and LSP-treated soil samples when the samples reacted with CO2 in the room during drying periods in the wetting-drying cycles. The results of the pH measurements performed for the 7-days cured samples and the 58-days swelling samples are illustrated in Figure 4.1.

Figure 4.1: pH of the control and the stabilized soil

The mineralogical composition of the stabilized soils is presented in Figure 4.2.

According to the XRD patterns of the stabilized soils, ettringite was formed in all samples due to the reaction of calcium available from the calcium-based stabilizing material, aluminum and sulfate available from the sulfate-bearing saline soil, and water provided as the source for soil- stabilizer mixing. Ettringite is capable of storing large amounts of water within the material, thus promoting the volumetric swelling of the OPC- and LSP-stabilized soil. As the tested soil is a low plasticity soil, and plasticity is expected to reduce after the stabilization, the expansion is expected to be not as high as for clayey soil.

(36)

Figure 4.2: XRD patterns for the stabilized soil

4.2. Evaluation of the Geotechnical Properties of the Stabilized Soil

The geotechnical properties of the cement- and limestone powder-stabilized sulfate- bearing saline soil, including Atterberg limits, optimum moisture content, maximum dry density, unconfined compressive strength, shear strength, particularly, cohesion and friction angle, resilient modulus, and California bearing ratio are discussed. Changes in these properties associated with the soil stabilization mechanism and soil chemistry are analyzed and explained in this section.

4.2.1. Atterberg Limits

The soil stabilization mechanism that induces cation exchange and formation of flocculated and agglomerated particles at an early age aims to reduce soil plasticity. Indeed, as Atterberg limits test results show, stabilization of sulfate-bearinf saline soil with pure cement and its combination with limestone powder provides a reduction of soil plasticity except for one

(37)

case. As presented in Figure 4.3, except for the sample stabilized with 2% OPC + 2% LSP, all other mixtures show a decrease in soil PI, with the maximum 91% reduction achieved in the 8%

OPC treated soil sample.

Figure 4.3: Atterberg limits of the control and the stabilized soil samples

The maximum PI reduction of 91% in 8% OPC treated soil can be compared to the 86%

decrease in PI of the soil mixture containing 6% OPC content combined with 2% LSP content.

It shows that cement and limestone powder's combined effect is almost as strong as the effect of cement used in a higher percentage. Moreover, it was recorded that 2% OPC content in combination with 2% and 4% LSP contents was not efficient in reducing soil plasticity, with the former combination resulting in the plasticity increase of about 84.5%. This behavior may be attributed to the introduction of the finer particles with OPC and LSP, which are sufficient to change the gradation of the soil-stabilizer mixture and not sufficient to promote the stabilization mechanism, particularly, cation exchange in the soil-stabilizer matrix and flocculation and agglomeration of the mixture particles. However, the soil sample treated with the combination of 2% OPC and 6% LSP has lower PI, comparing to the samples stabilized with 4% OPC and its combination with 2% and 4% LSP. This again shows that higher cement content can be replaced by lower cement content with the addition of limestone powder, as the mixtures stabilized with the combination of lower OPC content and LSP result in the same or even lower soil plasticity as the ones treated with higher OPC content. Moreover, an increase in LSP content provides a more significant reduction in soil PI, except for the 4% OPC + 2%

LSP treated soil sample. Nevertheless, the 30% PI reduction associated with 4% OPC content

(38)

is almost similar to the plasticity decrease of 29% associated 4% OPC + 2% LSP content. Thus, the trend, shown in Figure 4.4, may be generalized as stated above.

Figure 4.4: Effect of LSP content on PI

4.2.2. Optimum Moisture Content-Dry Density Relationship

The moisture content-dry density relation (M-D curve) of the natural soil was obtained, and the optimum moisture content at which the tested soil exhibits the highest dry density was determined as shown Figure 4.5. Further, the values of OMC and MDD for the OPC- and LSP- treated samples were obtained using the empirical relationship (Equations 3.1, 3.2, and 3.3) described in the previous sections. The effect of OPC and LSP addition on the stabilization of sulfate-bearing saline soil, in terms of OMC and MDD, is illustrated in Figure 4.6.

(39)

Figure 4.5: OMC-MDD determination for the tested soil

Generally, the stabilization of soil is required to increase MDD and decrease OMC of soil. The soil samples treated with 2% OPC + 2% LSP and 4% OPC lead to the significant MDD increase, while the samples treated with 2% OPC + 4% LSP, 4% OPC+ 2% LSP, and 6% OPC have almost the same MDD as the control sample. The other mixtures show a significant drop in MDD values. The combination of 2% OPC and 2% LSP contents act as a filling material due to the small particle sizes of the binders, thus, contribute to the MDD increase. This behavior can also be linked to the behavior of the same additive combination in changing Atterberg limits described in the previous section, particularly increasing soil plasticity (instead of reducing it). The reduced MDD, corresponding to the majority of the stabilized soil samples, may be explained by the increased soil resistance against compaction due to the formation of flocculated and agglomerated soil-stabilizer particles during cation exchange. Based on the mathematical equation (Equation 3.1) provided in the previous sections, OMC is a linear function of the stabilizer content. Thus, the value increases when the total stabilizer content rises. The increased OMC may also be attributed to the flocculation and agglomeration of soil-stabilizer particles during cation exchange: the flocculated and agglomerated particles occupy larger spaces, which, in turn, contributes to an increase in void ratio in soil-stabilizer matrix

(40)

Figure 4.6: OMC and MDD of the control and the stabilized soil samples

As stated in Equation 3.1 provided by the TxDOT, the higher stabilizer content corresponds to the higher OMC value. Accordingly, the increased additive content leads to the reduced MDD value. Thus, an increase in LSP content causes a decrease in MDD value, as shown in Figure 4.7.

Figure 4.7: Effect of LSP on MDD

4.2.3. Unconfined Compressive Strength

The stabilization of soil aims to achieve both the short-term and the long-term strength improvements through the stabilization mechanism, particularly cation exchange, flocculation

(41)

and agglomeration soil-stabilizer particles, and pozzolanic reaction, resulted in the formation of C-A-H and C-S-H. As shown in Figure 4.7, the stabilized mixtures have considerably higher UCS than the untreated soil, which reflects the main purpose of the stabilization process. The highest increase in UCS of about 171% corresponds to the 8% OPC-treated soil sample.

Figure 4.8: UCS of the control and the stabilized soil

In general, the cement content increase correlates with the instant and prolonged strength improvement, whereas the limestone powder content increase from 2% to 4 reduces soil strength, regardless of curing age, as illustrated in Figures 4.9 and 4.10. The LSP content increase from 4% to 6% at fixed 2% OPC content slightly enhances both short- and long-term strength. Interestingly, while the LSP content increase from 0% to 2% for the 6% OPC-treated soil increases the 7-days and the 28-days UCS significantly, the similar increase in LSP content for the 4% OPC-treated soil shows the opposite trend. Overall, the addition of LSP in the 2%

OPC-treated and the 4% OPC-treated soil samples is not efficient in promoting bot instant and prolonged strength improvement. This behavior may be attributed to the gradation of the binders, particularly, the fact that LSP has well-graded grain size distribution, comparing to uniformly-graded OPC. The introduction of coarse particles together with fine particles, when LSP is added to the system, especially at higher contents than OPC, results in the poor binding of the soil-stabilizer particles and insufficient surface tension, that together lead to the reduction of the soil compressive strength, as it can be seen from the UCS test results.

(42)

Figure 4.9: Effect of LSP on 7-days UCS

Figure 4.10: Effect of LSP on 28-days UCS

4.2.4. Shear Strength

The stabilization of soil aims to increase shear strength of soil. Namely, the flocculation and agglomeration of soil-stabilizer particles during cation exchange and the formation of cementation gel, a product of the pozzolanic reaction, result in the increased surface tension of soil-stabilizer matrix, thus, higher cohesion and friction angle of stabilized soil. Along with the effect of stabilizers, the shear strength parameters also rise under the effect of curing age.

Accordingly, the DST results, illustrated in Figure 4.11, show a significant increase in soil cohesion due to the addition of stabilizers and curing age. Initially, the tested soil is cohesionless sand, and after mixing it with cement and limestone powder, the cohesion of the soil-stabilizer

(43)

mixture rises, with the maximum increase up to 62 kPa and 63 kPa associated with the 7-days and 58-days 8% OPC-treated soil sample, respectively.

Figure 4.11: Cohesion of the stabilized soil

Generally, the increased stabilizer content, both OPC and LSP, results in increased soil cohesion. However, the effect of LSP is more moderate, comparing to that of OPC, as provided in Figures 4.12 and 4.13. Particularly, for the 7-days cured samples, the LSP content increase from 0% to 2% has a more significant effect in increasing soil cohesion for the 4% OPC- stabilized sample rather than 6% OPC-stabilized one. Similarly, increasing LSP content from 2% to 4% is more effective in promoting soil cohesion for the samples treated with 2% OPC rather than the ones treated with 4% OPC, for both 7-days and 58-days samples. This trend may be explained by the particle size distribution of the stabilizers: while OPC is uniformly graded and mainly consists of particles between 10 μm and 45 μm, LSP has more particles coarser than 45 μm and more particles finer than 10 μm. The introduction of both coarse and fine particles, when LSP is added to the soil-cement mixture, changes the overall gradation of the system, and coarse particles weaken the effect of the stabilizer in promoting cohesion of the stabilized mixture. Interestingly, however, for the 6% OPC-stabilized soil samples exposed to the 58-days volumetric swelling, the increase in LSP content from 0% to 2% results in the reduction of soil cohesion, though not significant. In this sense, the sample treated with 4% OPC + 4% LSP can be compared to the samples treated with 6% OPC + 2% LSP. It can be concluded that, in the long-term perspective, the combination of lower cement and higher limestone powder contents can be more effective in enhancing shear strength parameters, namely cohesion, rather than the

(44)

combination of higher cement and lower limestone powder contents. Decreasing OPC and increasing LSP contents is not only efficient technique in terms of shear strength improvement, but also an environmentally-friendly and cost-effective solution for soil stabilization.

Figure 4.12: Effect of LSP on 7-days cohesion

Figure 4.13: Effect of LSP on 58-days cohesion

The internal friction angle of the soil was not subjected to significant changes under the effect of the stabilization. As presented in Figure 4.14, friction angle values for the treated samples remain almost constant. This behavior may be attributed to the fact that cement and limestone powder do not affect the interlocking stress between soil-stabilizer particles.

(45)

Figure 4.14: Friction angle of the control and the stabilized soil

4.2.5. Resilient Modulus and California Bearing Ratio

Resilient modulus, MR, and California bearing ratio, CBR, are the key parameters used for the assessment of pavement performance. In this paper, MR and CBR were obtained using the empirical relation provided in the previous chapter, and the calculated values are presented in Figure 4.15. Usually, the unit for MR is pound per square inch (psi), however, for convenience, the values were converted to MPa. In general, the soil stabilization mechanism promotes an increase in both MR and CBR, thus, contribute to improved pavement performance and the prolonged road lifetime. The beneficial effect of the soil stabilization is evidential for all soil-stabilizer mixtures, except for the 2% OPC + 2% LSP treated sample. Based on Equation 3.6, the CBR of the treated soil depends on its plasticity, therefore, as PI of the 2% OPC + 2%

LSP treated sample increased significantly, as described in the previous sections, CBR for the mixture decreased. The CBR reduction led to the MR reduction of the same sample, according to Equation 3.4. The other stabilized samples show an increase in both MR and CBR due to the positive effect of OPC and LSP on soil plasticity. The most significant increase in MR and CBR is experienced in 6% OPC and 8% OPC treated samples. The addition of LSP to the 2% OPC + 2% LSP stabilized soil samples increases both MR and CBR values, the same trend is applicable for the 4% OPC + 2% LSP treated samples, though the MR and CBR increase in relatively low. Moreover, the stabilization of soil with 2% OPC + 6% LSP leads to the more effective MR and CBR improvement, comparing to the stabilization with 4% OPC + 2% LSP.

In this sense, the use of lower cement content in combination with higher limestone powder content is more beneficial in terms of enhancing MR and CBR values rather than the use of