Swelling Treatment By Using Sand for Tamia Swelling Soil G. E. Abdelrahman 1, M. M. Shahien 2 1 Department of Civil Engineering, Cairo University-Fayoum Branch, Fayoum, Egypt 2 Department of Civil Engineering, Tanta University, Egypt Abstract- The aim of this research is to study the effect of both sand content percent and gradation on the Tamia swelling clay. It describes an experimental investigation of the physical and mechanical properties of compacted sandy clay. Two sands with different gradation were used. Both sands are fine to medium and well graded sand. Three different sand percentages by weight were used. The results from Atterberg Limits, Compaction, Oedometer, and Direct Shear box tests were compared to find out the effect of sand gradation on swelling soil stabilization. The results show the variations of the maximum dry unit weight and undrained shear strength with the molding moisture content for material with different sand contents and gradation. Sand gradation showed no significant effect on Atterberg limits. It is concluded that the addition of 40% well graded sand reduced the free swell and swelling pressure by 70% and 90%, respectively. While the addition of the same percent of uniform sand reduced the free swell and swelling pressure by 50% and 94%, respectively. Keywords Swelling clay, Atterberg limits, Oedometer, Compaction, Shear strength, Stabilization. I. INTRODUCTION The most common techniques used for treating expansive soil deposits are lime, cement, and salt stabilization. The last two types reduce liquid and plastic limits and increase both shrinkage limit and shear strength. Many investigators studied swelling and engineering properties of compacted sandy clay. Faure [1] studied the evolution of compaction curves versus water content with variation of clay content, clay mineralogy, sand content, and compactive effort by using kaolin and montmorillonite mixed separately with sand in different proportions. Daniel and Wu [2] investigated compacted clayey sand liners and covers for arid sites. They used soil samples from west Texas, samples were compacted and tested for hydraulic conductivity, strength, and volume changes. Faure and Mata [3] investigated the penetration resistance along compaction curves and concluded that the maximum dry density of compaction curves had no effect on penetration resistance values. Alawaji [4] investigated the effect of clay/sand contents, initial water content, and dry density on the swell potential, strength, and compressibility of compacted sandy clay soils. He used Thomamah sand from surface sand dune in north- eastern Riyadh with Al-Ghatt expansive clay from the same region. Leelani et al. [5] investigated the effect of adding fine and coarse uniform sands on stabilizing active clay. The parameters investigated in that study includes Atterberg limits, compaction characteristics and shear strength. Swelling tests were not included in their study. The presence of expansive clay and shale in many places in Fayoum city such as Tamia, Elazab, Demmo, Khorshed, Komosheem, and Elberka with the presence of sand in adjacent parts of these regions suggested the potential use of mixed sandy clay soil in road subgrade, shallow foundation, or for other engineering application. In this study, sand was used as stabilizing material for the Tamia swelling soil in Fayoum city. Two types of sand with different gradation were used. Both types were added to Tamia swelling clay with 10%, 25%, and 40% by weight. Tests performed were Atterberg limits, oedometer tests, compaction tests, and direct simple shear box test on compacted sample at optimum moisture content. The objective of this study is to compare the different test results after using two different sizes of sand treatment. II. MATERIAL AND METHODOLOGY A Tamia Clay Tamia is a small village, located at the north of the Fayoum city, the expansive clay has 0.36% fine sand and 99.64% passing U.S. sieve No.200. The clay was classified as high plasticity clay (CH), according to USCS. The samples were brought form a strata in a soil profile that consists of three meter of cemented white silty clay followed by five meter of light yellowish silty clay, which was encountered till the end of the borehole at 12m with some lime for the last two meters. Ground water table is not encountered in the soil profile available data.
2 TABLE 1 Tamia Clay Properties Clay Properties Tamia Wc % 17.5 Dry Unit Weight, γ d t/m 3 1.78 Liquid Limit, LL% 72 Plastic Limit, PL % 33 Plasticity Index, IP % 39 Shrinkage Limit, SL % 9 Shrinkage Index. IR % 63 Montmorillonite % 31.98% Axial Swelling % 5.1% Swelling Pressure t/m 2 87 Swelling Strain (mm) 0.97 % Free Swell 100 Clay Fraction (% finer <0.002 mm) 52% Silt Content / Sand Content 47.64% / 0.36% Activity 0.81 B Stabilizing Sand Two sands were used in stabilizing the clay, fine to medium uniform sand and well graded sand. The grain size and gradation, classification, and compaction characteristics are summarized in Table (2).The main variable between the two sands is the size gradation of the sand. Properties TABLE 2 Sand Properties Uniform Fine Sand Well Graded Sand D 10, mm 0.23 0.08 D 30, mm 0.375 0.43 D 60, mm 0.51 0.57 Coefficient of Curvature 1.22 2.85 Coefficient of Uniformity 2.22 7.10 USCS SP SW Max. Dry Unit Weight, t/m 3 1.80 1.83 Optimum Moisture Content, % 7.5 10.0 C Sample preparation Both sands were mixed to Tamia clay. Three sand percentages were used by weight 10%, 25%, and 40%. Samples were prepared at three selected water contents; at optimum moisture content, dry and wet sides of optimum of the compaction curves. Samples for Oedometer, and Direct shear box tests were taken at optimum moisture content after the compaction test D Oedometer Tests Oedometer tests were used to investigate the influence of both sand size gradation and sand content on the swelling potential of Tamia clay. The swelling potential was measured for the Natural non-treated silty clay samples. Treated specimens were prepared by compact the Tamia silty clay in proctor compaction test at optimum moisture content with desired mixed sand percentage for both uniform and well graded sands, undisturbed samples were taken from the compaction mold to be used in oedometer. In the Oedometer, a constant pressure of about 1 t/m 2 was applied and then water was added while the sample allowed to swelling in the vertical direction under the constant pressure until it reached equilibrium. The swelling pressure was then measured by applying additional load to reduce the volume of the sample to its pre-inundation volume. E Compaction Test Maximum dry density, γ d, and optimum moisture content, OMC, were determined from compaction tests which were carried out on the Tamia clay soil mixed by 10%, 25%, and 40% with uniform and well graded sands. For each sand percentage, four different water contents were added to the mixed soil to reach the dry and the wet sides on compaction tests. F Direct Simple Shear Box Test Treated samples with 40% sand content were taken from compacted soil after compaction test at optimum moisture content and then tested in the direct simple shear both. The samples were sheared at fast enough rate (0.5mm/minute) to obtain undrained shear strength of the treated samples. III. RESULTS A Atterberg Limits Liquid limit, LL, plastic limit, PL, and shrinkage limit, SL, were tested before and after adding the stabilizing uniform and well graded sands. Typical results are shown in Figure 1. The results indicate that LL and PL
3 values decrease with the increase in sand content for both uniform and well graded sands. Shrinkage limit slightly increases with the increase of sand content. Figure 1 shows that added sand gradation does not have major influence on Atterberg limits results. LL & PL & SL % 80 60 40 20 0 Percentage of Sand % LL Fine Sand LL Graded Sand PL Fine Sand PL Graded Sand SL Fine Sand SL Graded Sand Fig. 1 Variation of consistency limits with different percentage of uniform and well graded sands According to Snethen [11] if shrinkage index (IR) is 30 to 60, swelling potential is high, and for IR >60, swelling potential is very high. Tamia swelling clay is classified as high to very high swelling according to the fore mentioned empirical methods.. The addition of 40% of uniform sand reduced the measured free swell from 100% for the natural soil to 50% for the admixture. On the other hand, the addition of the same amount of well-graded sand reduced the free swell % fro 100 to 30%. Figure 2 shows the Oedometer test results. The swelling strains of the natural clay, clay plus 40% well graded sand, and clay plus 40% uniform sand are 5.1%, 0.13%, and 0.08%, respectively. The swelling pressure of the natural clay was 87 t/m 2. The addition of 40% well graded sand and uniform sand reduced the swelling pressure to 8 t/m 2 and 5.3 t/m 2, respectively. For graded sand: LL= 0.77 % of Sand + 74.33 (R 2 =0.98) (1) PL = 0.49 % of Sand + 30.49 (R 2 =0.92) (2) SL = 0.1 % of Sand + 8.6 (R 2 =0.95) (3) For uniform sand Swelling Percentage % 6 5 4 3 2 1 0 0.1 1 10 100 Applied Stress kg/cm2 Clay before treatment Clay with 40% Graded Sand Clay with 40% Uniform Sand LL= 0.68 % of Sand + 77.55 (R 2 =0.95) (4) PL = 0.397 % of Sand + 27.96 (R 2 =0.68) (5) SL = 0.06 % of Sand + 8.89 (R 2 =0.75) (6) Where, the R 2 indicates the corresponding coefficient of determination. B Swelling Potential According to the Egyptian Code, part 5, [6], Tamia clay is considered as high swelling clay. Seed et al, [7], classified the swelling clay from low to very high according to the percentage of clay content and the activity with the swelling potential. Dakshanamurthy and Raman, [8], classified the swelling clay from low to very high according to the liquid limit and the plasticity index. Fig. 2 Effect of sand treatment on swelling strain and swelling pressure. C Compaction Tests Maximum dry unit weight and optimum moisture content were determined using the proctor compaction test. Figures 3 and 4 show typical results of adding three different percentage of uniform and graded sand. The results reflect the effect of sand content on the unit weight and optimum moisture content. Samples with larger amount of sand result in higher dry unit weight and lower optimum moisture content values which are consistent with the general trend shown by Henderson and Lisle [12], Faure and Mata [3], Alawaji [4], and Leelani et al. [5]. Unit weight of clay treated with well graded sand is higher than that of the clay treated with uniform sand. Van der Merwe [9] and similarly Williams and Donaldson [10] a version of which is in the Egyptian Code), swelling potential according to clay fraction and plasticity index.
4 Dry Density t /m3 2.00 1.75 1.50 1.25 1.00 Compaction Tests For Treated Clay with Uniform Sand Water Content Wc% Fig. 3 Variation of dry density with molding water content with uniform sand Dry Density t /m3 2.00 1.75 1.50 1.25 1.00 Compaction Tests For Treated Clay with Graded Sand Water Content Wc% 10 % Sand 25% Sand 40% Sand Fig.4 Variation of dry density with molding water content with graded sand 10 % Sand 25% Sand 40% Sand The maximum dry density γ d kg/m 3 and the optimum water content OMC% were regressed with uniform sand percentage or graded sand percentage, and liquid limit percentage and plastic limit percentage as follows: γ d = 82.4 uniform sand %+ 66.5 LL+119.7 PL 6442.3 (R 2 =0.998) γ d = 29.8 graded sand %+ 52.5 LL 44.4 PL 1045.3 (R 2 =0.98) OMC%= 0.12 uniform sand% + 1.39 LL 4.92 PL+19.75 (R 2 =0.97) OMC%= 0.82graded sand% 1.65 LL+2.04 PL+ 91.83 (R 2 =0.98) D Shear Box Test Result Direct shear box test was used to measure the undrained shear strength for undisturbed clay sample. The undrained shear parameter c u was 7.1t/m 2. As the sand content increases the shear strength increase, Alawaji [4] and Leelani et al. [5]. Figure 5 shows the results of the direct shear box test results for natural and treated clay. In general, the addition of 40% of sand (uniform or well graded) and compacting to maximum dry unit weight and optimum water content results in an increase in the undrained shear strength of the clay. The addition of well graded sand resulted in more pronounced increase in undrained shear strength with the increase in effective normal stress as compared to slight increase in the case of adding uniform sand (Figure 5- Table 3). Shear Stress ( kg/cm2) 2.5 2.0 1.5 1.0 0.5 0.0 Shear Box Test Results 0 0.5 1 1.5 2 Normal Stress ( kg/cm2) Clay before treatment 40% Graded Sand 40% Uniform Sand Fig. 5 Shear strength before and after sand treatment TABLE 3 Tamia Clay Properties before and after sand treatment Clay Properties Tamia before Tamia after treatment treatment 40% Uniform 40% Well sand Graded sand Wc % 17.5 10 11 γ dry t/m 3 1.78 1.92 1.9 LL% 72 49 45 PL % 33 15 13 IP % 39 34 32 SL % 9 11 13 IR % 63 38 32 C u t/m 2 6 15.6 14 ϕ 6.28 9.1 23.27 Axial Swelling (mm) 0.97 0.015 0.025 Swelling Pressure t/m 2 87 5.34 8 Swelling Potential % 5.1 0.008 0.013 % Free Swell 100% 50% 30%
5 IV. CONCLUSIONS 1. By correlating activity with swelling pressure, and by plotting these results on the Egyptian Codes charts it is found that Tamia samples are considered swelling soils. This must be taken into consideration in foundation design. The addition of sand to the expansive soil as stabilizing material is efficient, beneficial from economical and environmental views. 2. Treating the clay by adding sand results in reduction in liquid and plastic limits of swelling clay and slight increase in shrinkage limit... There was no major influence of gradation on the results. 3. The addition of 40% well graded sand reduced the free swell and swelling pressure by 70% and 90%, respectively. While the addition of the same percent of uniform sand reduced the free swell and swelling pressure by 50% and 94%, respectively. 4. The addition sand resulted in an increase in maximum dry unit weight and a decrease in the optimum water content. The trend is pronounced with the increase in sand content.. The addition of well graded sand gave higher increase in dry unit weight than the increase by adding uniform sand. [7] H.B. Seed, R.J. Woodward, and R. Lundgren, Prediction of swelling potential for compacted clay, Soil Mechanics and Foundation Division, ASCE, Vol. 88, No. SM3, June (1962), pp. 53-87. [8] V. Dakshanamurthy, and V. Raman, A simple method of identifying an expansive soil Soils and Foundations, Japanese Society, Vol.13, No.1, March (1973), pp. 97-104. [9] Van der Merwe, D. H. (1964). The prediction of heave from the plasticity index and percentage clay fraction of soils, The Civil Engineer in South Africa, Vol. 6, pp. 103-107. [10] Williams, A. A. B., and G. Donaldson (1980). Building on expansive soils in South Africa 1973-1980, Proceedings of 4 th Int. Conf. On Expansive Soils, Denver, Vol. 2, p. 834. [11] Snethen, D. R. (1980). Characterization of expansive soils using soil suction data Proceedings of 4 th Int. Conf. On Expansive Soils, Denver, Vol. 1, pp. 54-75. [12] C. Henderson and L. Lisle The effects of soil water content and bulk density on the compactibility and soil penetration resistance of some western Australian sandy soils. Australian Journal Soil Research 26, 1988, pp. 391-400. 5. The addition of sand increased the undrained shear strength of the compacted clay. The increase in undrained shear strength with the increase in effective normal stress is higher in the case of adding well grade sand as compared to that in the case of adding uniform sand. REFERENCES [1] A. Faure, Des caracteristiques de la fraction argileue dans le mecanisme de tasssements des sols. C.R. Acad. Sc., D (278), pp. 1175-1178, 1974. Quted from Ref. [4]. [2] D. E. Daniel and Y.K. Wu Compacted clay liners and covers for arid sites. Journal of Geotechnical Engineering, ASCE, vol. 119, No. 2, 1993, pp. 223-237. [3] A. Faure and J.D. Mata, Penetration resistance value along compaction curves. Journal of Geotechnical Engineering, ASCE, vol. 120, No. 1, 1994, pp. 46-59. [4] H. A. Alawaji, Swell and strength characteristics of compacted sandy clay soils. Third International Geotechnical Engineering Conference Cairo University-Egypt, 1997, pp. 303-314. [5] Leelani, P.T., P.V. Compton, and S. Abdul-Baki (1992). Active clay stabilized by admixtures of fine and coarse sands, Proceedings of the 7 th Int. Conf. On Expansive Soils, Dallas, Texas, Vol. 1, pp. 370-374. [6] Egyptian Code, part 5, (2001).