Analysis of strength development in cement-stabilized silty clay from microstructural considerations

https://doi.org/10.1016/j.conbuildmat.2010.03.011Get rights and content

Abstract

This paper analyzes the strength development in cement-stabilized silty clay based on microstructural considerations. A qualitative and quantitative study on the microstructure is carried out using a scanning electron microscope, mercury intrusion pore size distribution measurements, and thermal gravity analysis. Three influential factors in this investigation are water content, curing time, and cement content. Cement stabilization improves the soil structure by increasing inter-cluster cementation bonding and reducing the pore space. As the cement content increases for a given water content, three zones of improvement are observed: active, inert and deterioration zones. The active zone is the most effective for stabilization where the cementitious products increase with cement content and fill the pore space. In the active zone, the effective mixing state is achieved when the water content is 1.2 times the optimum water content. In this state, the strength is the greatest because of the highest quantity of cementitious products. In the short stabilization period, the volume of large pores (larger than 0.1 μm) increases because of the input of coarser particles (unhydrated cement particles) while the volume of small pores (smaller than 0.1 μm) decreases because of the solidification of the cement gel (hydrated cement). With time, the large pores are filled with the cementitious products; thus, the small pore volume increases, and the total pore volume decreases. This causes the strength development over time.

Introduction

Soil in northeast Thailand generally consists of two layers. The upper layer (varying from 0 to 3 m thickness) is wind-blown and has been deposited over several decades. It is clayey sand or silty clay with low to moderate strength (12 < N < 20, where N is the standard penetration number). This upper soil is problematic because it is sensitive to changes in water content [1]. Its collapse behavior as a result of wetting is illustrated by Kohgo et al. [2]; and Kohgo and Horpibulsuk [3]. The lower layer is residual soil that is weathered from claystone and consists of clay, silt, and sand [4]. It possesses very high strength (generally N > 30) and very low compressibility. One of the most common soil improvement techniques for upper soil is to compact the in situ soil (in relatively a dry state) mixed with cement slurry. This technique is economical because cement is readily available at a reasonable cost in Thailand. Moreover, adequate strength can be achieved in a short time.

Stabilization begins by mixing the soil in a relatively dry state with cement and water specified for compaction. The soil, in the presence of moisture and a cementing agent becomes a modified soil, i.e., particles group together because of physical–chemical interactions among soil, cement and water. Because this occurs at the particle level, it is not possible to get a homogeneous mass with the desired strength. Compaction is needed to make soil particles slip over each other and move into a densely packed state. In this state, the soil particles can be welded by chemical (cementation) bonds and become an engineering material.

The effects of some influential factors, i.e., water content, cement content, curing time, and compaction energy on the engineering characteristics of cement-stabilized soils have been extensively researched [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20]. However, these previous investigations have mainly focused on the mechanical behavior: the microstructural study is limited. It is vital to understand the changes in engineering properties that result from changes in the influential factors.

Models of the microstructure of fine-grained soils have been developed and modified since 1953 by geotechnical engineers to help understand soil behavior. Lambe’s model is the first conceptual model, which considers clay particles to be single platelets. Since Lambe developed his theory, there have been significant improvements in microstructure observation techniques, leading towards complete description of the microstructure of fine-grained soils in relation to their engineering behavior such as the works reported by Gillott [21] and Collins and McGown [22]. Aylmore and Quirk [23]; Olsen [24]; and Nagaraj et al. [25] have revealed that the basic element of the microstructure of natural clay is not the single platelet but domains composed of various aggregated platelets.

Because cement and clay interacts with water, when clay is mixed with cement and water, clay and cement particles group together into large clay–cement clusters [26]. The cement gel is stable in the intra-aggregate and inter-aggregate pores because of the attractive forces (caused by physicochemical forces), and the capillary forces between the clay–cement clusters and the cement gel, respectively.

Abduljauwad [27] observed the microstructure changes of stabilized soils using scanning electron microscopes (SEMs). Keshawarz and Dutta [28] reported that the particles of uncemented soil appear as a blocky arrangement of loosely packed particles while the cemented soil has an abundance of tobermorite crystals. Previous works focusing on clay mineralogy [27], [29], [30], [31], [32] used X-ray diffraction techniques to investigate the mineralogical changes and to identify the reaction products formed when lime is added to clay soils.

Even though available researches exists on microstructure of cement-stabilized clay, they mainly focus on particular water content and curing time and do not cover all microstructural tests. This paper attempts to investigate the microstructural changes in cement-stabilized silty clay to explain the different strength development according to the influential factors, i.e., cement content, clay water content and curing time. Two sets of cemented samples were prepared for this study: samples with cement content C = 0–10% (practical range) and C > 10%. In the first set, the investigation illustrates the role of the influential factors on the strength and microstructure development and determines the effective mixing state in the practical range. The second set is used to further understand the strength and microstructure development with cement to facilitate the determination of proper quantity of cement to be stabilized. The unconfined compressive strength was used as a practical indicator to investigate the strength development. The microstructural analyses were performed in this paper using a scanning electron microscope, mercury intrusion porosimetry, and thermal gravity tests.

Section snippets

Laboratory investigation

The soil sample is silty clay collected from the Suranaree University of Technology campus in Nakhon Ratchasima, Thailand at a depth of 3 m. The soil is composed of 2% sand, 45% silt and 53% clay. Its specific gravity is 2.74. The liquid and plastic limits are approximately 74% and 27%, respectively. Based on the Unified Soil Classification System (USCS), the clay is classified as high plasticity (CH). During sampling, the groundwater had disappeared. The natural water content was 10 percent.

Compaction and unconfined compression test results

Compaction and strength test results for samples with C = 0 to 10% are shown in Fig. 3, Fig. 4, Fig. 5, Fig. 6. Fig. 3 shows the plots of dry unit weight versus molding water content of the uncemented and the cemented samples compacted under the standard and modified Proctor energies. It is noted that the maximum dry unit weight of the cemented samples is higher than that of the uncemented samples whereas their optimum water content is practically the same. This characteristic is the same as that

Uncemented samples

For compacted fine-grained soils, the soil structure mainly controls the strength and resistance to deformation, which is governed by compaction energy and water content. Compaction breaks down the large clay clusters into smaller clusters and reduces the pore space (vide Fig. 2, Fig. 8). Fig. 9 shows SEM photos of the uncemented samples compacted under the modified Proctor energy at water contents in the range of 0.8–1.2OWC. On the wet side of optimum (vide Fig. 9c), a dispersed structure is

Discussion

Based on the cluster theory [25], [32], the pores are classified into two categories: inter-aggregate pores (larger than 0.01 μm) and intra-aggregate pores (smaller than 0.01 μm). After mixing clay with cement, the formation of clay–cement clusters due to physicochemical interaction reduces the small inter-aggregate pore (0.01–0.1 μm) volume and slightly increases the large inter-aggregate pore (0.1–10 μm) volume, resulting in the increase in the dry unit weight. Because of the growth of

Conclusions

This paper analyzes strength development considering soil microstructure using a scanning electron microscope, mercury intrusion pore size distribution measurements, and thermal gravity analysis. The following conclusions can be advanced from this study.

  • 1.

    The strength development with cement content for a specific water content is classified into three zones: active, inert and deterioration. In the active zone, the volume of pores smaller than 0.1 μm significantly decreases with the addition of

Acknowledgements

The authors would like to acknowledge the financial support provided by theCommission on Higher Education (CHE) and the Thailand Research Fund (TRF) under the contract MRG5080127. Facilities, equipments and financial support provided by the Suranaree University of Technology are appreciated. The technical comments by Prof. T.S. Nagaraj (now deceased), Indian Institute of Science, as well as those by Dr. Amnat Apichatvullop and Dr. Theerawat Sinsiri, Suranaree University of Technology, are

References (45)

  • B. El-Jazairi et al.

    The hydration of cement plate using the semi-isothermal method of thermogravimetry

    Cem Concr Res

    (1980)
  • K.S. Wang et al.

    The hydration characteristics when C2S is present in MSWI fly ash slag

    Cem Concr Res

    (2004)
  • S. Horpibulsuk et al.

    A case history on underpinning for a distressed building on hard residual soil underneath non-uniform loose sand

    Soils and Found

    (2008)
  • Kohgo Y, Tamrakar SB, Tang HG. Investigations on the mechanical properties of typical soils distributed in northeast...
  • Kohgo Y, Horpibulsuk S. Simulation of volume change behavior of yellow soil sampled from Khon Kaen City in Northeast...
  • Udomchore V. Origin and engineering characteristics of the problem soils in the Khorat Basin. Northeastern Thailand,...
  • Terashi M, Tanaka H, Okumura T. Engineering properties of lime treated marine soils and DMM. In: Proceedings of the 6th...
  • Kawasaki T, Niina A, Saitoh S, Suzuki Y, Honjo Y. Deep mixing method using cement hardening agent. In: Proceedings of...
  • G.W. Clough et al.

    Cemented sands under static loading

    J Geotechn Eng Div, ASCE

    (1981)
  • Kamon M, Bergado DT. Ground improvement techniques. In: Proceedings of the 9th Asian regional conference on soil...
  • Uddin K. Strength and deformation behaviour of cement treated Bangkok clay. Doctoral Thesis, Asian Institute of...
  • J.H. Yin et al.

    Strength and stiffness of Hong Kong marine deposit mixed with cement

    Geotechn Eng J

    (1998)
  • Cited by (485)

    View all citing articles on Scopus
    View full text