- Email: [email protected]
Shear strength-suction relationship of compacted Ankara clay
Applied Clay Science 49 (2010) 400–404
Contents lists available at ScienceDirect
Applied Clay Science j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l a y
Shear strength-suction relationship of compacted Ankara clay Erdal Çokça ⁎, Hüseyin P. Tilgen Department of Civil Engineering, Middle East Technical University, 06531 Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 28 August 2008 Received in revised form 26 August 2009 Accepted 27 August 2009 Available online 4 September 2009 Keywords: Ankara clay Compaction Moisture content Shear strength Suction Unsaturated soil
a b s t r a c t The relationship between shear strength and soil suction of compacted Ankara clay was investigated at different moisture contents. Soil samples were tested at optimum moisture content (w = 20.8%), drier than optimum (w = 14.8%, 16.8%, 18.8%) and wetter than optimum (w = 22.8%, 24.8%, 26.8%). Direct shear tests were performed. Soil suctions were measured by the filter paper method after direct shear tests. Direct shear tests and soil suction measurements were also performed on soaked samples. The soil suction versus moisture content curve was developed. The shear strength variation with respect to soil suction was found for all normal stresses. The soil suction versus moisture content curves bear a close relationship to the unsaturated shear strength of the soil. An increase in soil suction increases the shear strength. © 2009 Elsevier B.V. All rights reserved.
1. Introduction
2. The strength of compacted clays
Compacted soils are part of many earth structures. Therefore their shear strength is very important in geotechnical problems such as bearing capacity, slope stability, lateral earth pressures. Usually, to determine shear strength, laboratory samples are tested which have the same moisture content and dry density as the fill, but the fill can be wetted in the future. For example, the bearing capacity of shallow footings on compacted fills is commonly based on the compressive strength of the unsaturated soil. The strength measurements are often performed on soil specimens which have negative pore water pressures (suction). The assumption is then made for design purposes that conditions in the future will remain similar, but this may not be a realistic assumption. The fill can be wetted; the source of wetting primarily is rainfall. Rising groundwater may be a source of wetting. In addition, broken utility lines, utility trenches, street subgrades, permeable layers, gravel packed subdrains, all act as subsurface conduits that lead water to fill (Croney et al., 1958). The Ankara clay is an overconsolidated and fissured clay at most places, and may have swelling properties. In this study, soil suction and shear strength of unsaturated compacted Ankara clay were measured. The soil suction was measured by the filter paper method. Direct shear testing was used to measure shear strength. Additional direct shear tests were performed where the compacted samples were soaked and consolidated prior to shearing. The relationships between moisture content, soil suction and shear strength were found.
Previous studies on the shear strength of unsaturated compacted soils (Diamond, 1970; Brackley, 1973, 1975; Zein, 1985; Delage et al., 1996; Kong and Tan, 2000; Toll, 2000) show the development of an aggregated fabric in materials compacted drier than the optimum moisture content. This type of aggregation was not found to exist in the materials compacted on the wet side of the optimum moisture content. Vanapalli et al. (1996) measured the shear strength of statically compacted glacial till specimens using a modified direct shear apparatus. Specimens were prepared at three different water contents and densities (i.e. corresponding to drier than optimum, at optimum, and wetter than optimum conditions). Various net normal stresses and matric suctions were applied to the specimens. Soil-water characteristic curves were developed using a pressure plate apparatus on specimens. According to Vanapalli et al. (1996), “soils compacted at various “initial” water contents and to various densities should be considered as “different” soils from a soil mechanics behavioral standpoint even though their mineralogy, plasticity, and texture are the same. The engineering behavioral change from one specimen to another will vary due to differences in soil structure or aggregation.” Lambe and Whitman (1979) states that “for a given compactive effort and dry density, the soil tends to be more flocculated for compaction on the dry side as compared to compaction on the wet side (i.e. on the wet side the soil is more dispersive). In general, an element of flocculated soil has a higher strength than the same element of soil at the same void ratio but in a dispersed state.” Brackley (1973, 1975) proposed a model of unsaturated clay fabric in which the clay particles grouped together in ‘packets’. The soil
⁎ Corresponding author. Tel.: + 90 312 2102435; fax: + 90 312 2105401. E-mail address: [email protected] (E. Çokça). 0169-1317/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.clay.2009.08.028
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
within the aggregation or ‘packets’ is held together by suction, and the packets act as individual particles. The larger effective particle size produces more frictional behavior and hence a higher angle of friction. According to Toll (2000) “fabric plays a vital role in determining the engineering behavior of compacted soils. Clayey materials compacted drier than optimum moisture content develop an aggregated or ‘packet’ fabric. The presence of aggregations causes the soil to behave in a coarser fashion than would be justified by the grading.” Fredlund and Rahardjo (1993) used modified direct shear apparatus to measure the shear strength (consolidated-drained tests), and matric suctions were measured by pressure plate tests. Fredlund and Rahardjo (1993) state that “the suction influences the shear strength of unsaturated compacted specimens.”
3. Soil suction The total suction, ψ of a soil is made up of two components, namely, the matric suction, (ua – uw), and the osmotic suction, π. The matric suction component is commonly associated with the capillary phenomenon arising from the surface tension of the water. The pore-water in a soil generally contains dissolved salts; the decrease in relative humidity due to the presence of dissolved salts in pore-water is referred to as the osmotic suction π. The free energy of the soil water (total suction, ψ) can be determined by measuring the vapor pressure of the soil water or the relative humidity (RH) in the soil. The direct measurement of RH in a soil can be conducted using a device called a psychrometer (Fredlund and Rahardjo, 1993). The RH in a soil can be indirectly measured using a filter paper as a measuring sensor. The filter paper is equilibrated with the suction in the soil.
3.1. Soil suction measurements with filter paper method The working principle behind the filter paper method is that the filter paper will come to equilibrium with the soil either through vapor flow or liquid flow, and at equilibrium, the suction value of the filter paper and the soil will be the same (Fig. 1). When the psychrometer method is compared to the filter paper method, the filter paper method gives more consistent results (Bulut et al., 2000). If the filter paper is allowed to absorb water through vapor flow (non-contact method), then total suction is measured (Fig. 1). After equilibrium is established between the filter paper and soil in a constant temperature environment, the water content of the filter paper disc is measured. Then, by using a filter paper calibration curve of water content versus suction, the corresponding suction value is found from the curve, so the filter paper method is an indirect method of measuring soil suction. Therefore, a calibration curve should be constructed or be adopted, curves were presented for different filter papers in ASTM D 5298-92, Standard Test Method for Measurement of Soil Potential (Suction) Using Filter Paper in soil suction measurements.
Fig. 1. Non-contact filter paper method for measuring total suction.
401
The filter paper method can reliably be used with suctions from about 80 kPa to in excess of 6000 kPa, a much larger range than any other single technique (Chandler and Gutierrez, 1986). 4. Experimental study 4.1. Introduction The soil sample (Ankara clay) used in this study was taken from the METU (Middle East Technical University, Ankara Turkey) campus area and classified according to Unified Soil Classification System by using the test results of the sieve analysis, hydrometer analysis and Atterberg limits. Also specific gravity, maximum dry density and optimum moisture content (o.m.c.) (Fig. 2) were determined. The dry density versus moisture content curve of the sample was plotted by using the compaction test results in which standard Proctor compaction mould and hammer was used. Soil index and compaction properties are given in Table 1. Free swell and suction properties are given in Table 2. After the Proctor compaction test, soil samples were tested at optimum moisture content (w= 20.8%), drier than optimum (w= 14.8%, 16.8%, 18.8%) and wetter than optimum (w= 22.8%, 24.8%, 26.8%). Shear strengths were determined by using direct shear apparatus. All samples were sheared under 75 kPa,150 kPa, and 225 kPa normal stresses. After that the same samples were used in filter paper suction measurements as described in Section 3.1. 4.2. Procedure for direct shear test In this test program 14 sets of direct shear tests (ASTM D3080, 2003) were made and each set contains 3 individual direct shear tests. To prepare the samples, oven dried samples were mixed with an appropriate mass of water and left 24 hours in a plastic bag in a humidity room to form a homogeneous mixture. After that the samples were compacted dynamically by using a Proctor compaction mould. Then samples were taken out from the compaction mould by using direct shear mould and hydraulic jack. It is very important to prevent the soil sample from moisture loss since the sheared samples were used for suction measurements after the direct shear test. Therefore, the direct shear box was wrapped with nylon stretch film and covered with moisturized cloth after placing the sample in the direct shear machine. The emplaced samples were left one day for consolidation under normal stress of σn = 75 kPa, σn = 150 kPa, and σn = 225 kPa. Each set contains three stages and each stage finalized in two days. After consolidation, samples were sheared. In order to prepare and test a soaked sample, the same procedure was used but before the direct shear machine was switched on, the samples were soaked and left for 24 hours under weights giving normal stresses of 75 kPa, 150 kPa, and 225 kPa.
Fig. 2. Dry density versus moisture content.
402
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
Table 1 Index and compaction properties of the tested clay. Property
Value
Specific Gravity, (Gs) Liquid Limit, (LL) Plastic Limit, (PL) Plasticity Index, (PI) Clay Fraction, (% finer than 2 µm) Activity, (PI / % finer than 2 µm) Optimum Moisture Content, (wopt) Maximum Dry Density, (ρd) Classification (According to Unified Soil Classification System)
2.73 48% 21% 27% 67.9% 0.40 20.8% 1.67 Mg/m3 CL
Since the Consolidated Drained test procedure was followed, horizontal displacement rate was very important. Therefore after each consolidation day t50 (time to reach 50% consolidation) and t90 (time to reach 90% consolidation) values were determined by using the log time and root time method (ASTM D2435, 2003). Using the following equations the failure times were calculated for each sample: tf = 11:7 t90
ð1Þ
tf = 50 t50
ð2Þ
After a few trials, it was seen that time to failure (tf) was between 6 to 10 hours and lateral displacement δ to reach the peak soil strength was between 3 mm to 5 mm. So, by using the following formula the displacement rate was chosen as 1.0 × 10- 4 mm/sec, which was slower than calculated, to be on the safe side: V = δ = tf
ð3Þ
4.3. Soil total suction measurements with filter paper method After the direct shear test, the tested sample was taken out from the direct shear box. After taking samples for moisture content check, the remaining sample was used for filter paper suction measurements. In this study ash free Whatman no. 42 type filter papers were used. At least 75 percent volume of a glass jar was filled with the soil (Fig. 1). A ring type support was put on top of the soil to provide a noncontact system between the filter paper and the soil. Two filter papers were placed on the ring. Then, the glass jar lid was sealed. After that, the glass jar was put into a box and in a controlled temperature room for equilibrium. The suggested equilibrium period is at least one week. After the equilibrium period, the filter paper moisture content was measured to nearest 0.0001 g accuracy. The calibration curve obtained in this study for Whatman no.42 type filter paper, by following ASTM D5298-92 (1992) procedure, is given in Fig. 3. After obtaining the filter paper moisture content (w) value, the calibration curve (Eq. (4)) was used to get the suction (ψ) value of the soil sample. Log ψ ðkPaÞ = 5:1887 – 0:0741w
Fig. 3. Calibration curve of Whatman no. 42 type filter paper used in this study.
5. Test results 5.1. Shear strength test results The shear stress – shear displacement and shear strength – vertical stress relationships for undisturbed Ankara clay (at natural moisture content) are given in Figs. 4 and 5 as examples. Similar behaviour was observed for all moisture contents. 5.2. Total suction-moisture content relationships The moisture content versus log total suction relationship is given in Fig. 6 for as compacted samples (there is no normal stress acting on the sample). Fig. 6 indicates that the soil suction increases as the water content decreases at the dry side of optimum. In this range of water content the moisture content versus log suction behaviour is linear (regression equation is given on the Fig. 6). Above the o.m.c. it has an almost constant value. Similar behaviour was observed for σn = 75 kPa, 150 kPa and 225 kPa samples. At the wet side of the optimum the soil suction attains steady and is not a function of the moisture content. 5.3. Shear strength and total suction relationships The shear strength of the compacted samples which is defined as the maximum shear stress measured in the direct shear tests is plotted against total soil suction on the logarithmic scale as shown in Fig. 7 for the three normal stress ranges studied and for the soaked samples are given in Fig. 8. The moisture content of the data points (from left to right) are 26.8%, 24.8%, 22.8%, 20.8%, 18.8%, 16.8% and 14.8%. Fig. 7 shows that, for all normal pressures the change in total suction with respect to shear strength shows similar behavior (curves are drawn to show the general trend of data points). Towards the dry side of o.m.c., the shear strength increases. The shear strength versus log soil suction behavior depends on the moisture content and the normal stress level.
ð4Þ
Table 2 Swell and suction properties of the undisturbed sample. Test
Results
Free Swell Natural Water Content (w) Applied Normal Pressures (σn) Total Suction (Ψ)
2.4% 21% 75 kPa 3136 kPa
150 kPa 4110 kPa
225 kPa 4414 kPa
Fig. 4. Shear stress versus shear displacement graphs for undisturbed samples.
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
403
Fig. 7. Shear strength-total suction graphs for σn = 75, 150, 225 kPa.
gated. Experiments were done on samples compacted at optimum moisture content (w= 20.8%), drier than optimum (w = 14.8%, 16.8%, 18.8%) and wetter than optimum (w= 22.8%, 24.8%, 26.8%). The following conclusions were drawn:
Fig. 5. Shear strength versus vertical stress graph for undisturbed samples.
The shear strength versus soil suction behavior can be explained as: the clayey material tends to be more flocculated for compaction on the dry side of o.m.c. as compared to compaction on the wet side of o.m.c. Suction seems to generate a resistance to slippage at the contacts between the particles (or aggregates) as the moisture content decreases on the dry of o.m.c. It is noted that the role of clay aggregates on shear strength of the clay is substantially reduced at about o.m.c. The decrease in shear strength with increasing moisture content (at the wet side of o.m.c.), is attributed to decreasing soil suction. From Fig. 8, due to soaking, the initial moisture content loses its importance for the shear strength and total suction relationship. As normal pressure increases, shear strength increases, too. When the shear strength-total soil suction graphs of soaked samples are compared with as compacted samples, the trend for compacted samples disappears due to soaking.
The total soil suction increases as the moisture content decreases at the dry side of optimum, and in this range of moisture content the moisture content versus log suction behaviour is linear and above the o.m.c. it has an almost constant value. For all normal pressures change in total suction with respect to shear strength shows similar behavior. Towards the dry side of o.m.c., shear strength increases with decrease in moisture content. Due to soaking, the initial moisture content loses its importance for the shear strength and total suction relationship. As normal pressure increase, shear strength increases, too. An increase in soil suction increases the shear strength. This study shows the effect of the compaction moisture content and soaking on the shear strength of the compacted soil. Therefore, if there is a possibility of wetting of the compacted fill in the future, to properly study the shear strength behaviour of the unsaturated compacted soil, tests should be performed on the soil in its ‘as compacted’ condition and after soaking. Acknowledgments
6. Conclusions Effects of compaction moisture content and soaking on the unsaturated shear strength and suction of Ankara clay were investi-
Fig. 6. Total suction-moisture content graph for as compacted samples.
The help provided by the technician Mr. Ali Bal under the supervision of the writers is gratefully acknowledged. The writers are grateful for the help provided by Ms. Nilgün N. Çokça during the review stage of the manuscript. The writers also would like to thank the reviewers of the paper for their many helpful suggestions.
Fig. 8. Shear strength-total suction graphs for σn = 75, 150, 225 kPa for soaked samples.
404
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
References ASTM D2435, 2003. Standard test method for one-dimensional consolidation properties of soils using incremental loading. American Society of Testing and Materials. Annual Book of ASTM Standards, Philadelphia 04.08, 1–10. ASTM D3080, 2003. Standard test method for direct shear test of soils under consolidated drained conditions. Annual Book of ASTM Standards, Philadelphia 04.08, 1–7. ASTM D5298-92, 1992. Standard test method for measurement of soil potential (suction) using filter paper. American Society of Testing and Materials. Annual Book of ASTM Standards, Philadelphia 15.09, 156–161. Brackley, I.J.A., 1973. Swell pressure and free swell in a compacted clay. Proceedings 3rd International Conference Expansive Soils, Haifa, Vol. 1. Academic Press, Jerusalem, pp. 169–176. Brackley, I.J.A., 1975. A model of unsaturated clay structure and its application to swell behavior. Proceedings 6th African Conference Soil Mechanics and Foundation Engineering, Durban. Balkema, Rotterdam, pp. 65–70. Bulut, R., Park, S.W., Lytton, R.L., 2000. Comparison of total suction values from psychrometer and filter paper. In: Rahardjo, H., Toll, D.G., Leong, E.C. (Eds.), Unsaturated Soils for Asia. Springer, Singapore, pp. 269–273. Chandler, R.J., Gutierrez, C.I., 1986. The filter paper method of suction measurement. Geotechnique 36, 265–268.
Croney, D., Coleman, J.D., Black, W.P.M., 1958. Studies of the movement and distribution of water in soil in relation to highway design and performance. HRB spec. Washington D.C. Report 40, pp. 226–250. Delage, P., Audiguier, M., Cui, Y.J., Howat, M.D., 1996. Microstructure of a compacted silt. Canadian Geotechnical Journal 33, 150–158. Diamond, S., 1970. Pore size distribution in clays. Clays and Clay Minerals 18, 7–23. Fredlund, D.G., Rahardjo, H., 1993. Soil Mechanics for Unsaturated Soils. John Wiley & Sons, Inc., pp.1-77, pp.247-258. Kong, L.W., Tan, L.R., 2000. Study on shear strength and swelling – shrinkage characteristic of compacted expansive soil. In: Rahardjo, H., Toll, D.G., Leong, E.C. (Eds.), Unsaturated Soils for Asia. Springer, Singapore, pp. 515–519. Lambe, T.W., Whitman, R.V., 1979. Soil Mechanics. Wiley, New York. Toll, D.G., 2000. The influence of fabric on the shear behavior of unsaturated compacted soils. Advances in Unsaturated Geotechnics. : Geotechnical Special Publication, vol. 99. American Society of Civil Engineers, pp. 222–234. Vanapalli, S.K., Fredlund, D.G., Pufahi, D.E., 1996. The relationship between the soilwater characteristic curve and the unsaturated shear strength of a compacted glacial till. Geotechnical Testing Journal ASTM 19, 19–28. Zein, A.K.M., 1985. Swelling characteristics and micro fabric of compacted black cotton soil. PhD. thesis, University of Strathclyde, Glasgow.
Contents lists available at ScienceDirect
Applied Clay Science j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l a y
Shear strength-suction relationship of compacted Ankara clay Erdal Çokça ⁎, Hüseyin P. Tilgen Department of Civil Engineering, Middle East Technical University, 06531 Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 28 August 2008 Received in revised form 26 August 2009 Accepted 27 August 2009 Available online 4 September 2009 Keywords: Ankara clay Compaction Moisture content Shear strength Suction Unsaturated soil
a b s t r a c t The relationship between shear strength and soil suction of compacted Ankara clay was investigated at different moisture contents. Soil samples were tested at optimum moisture content (w = 20.8%), drier than optimum (w = 14.8%, 16.8%, 18.8%) and wetter than optimum (w = 22.8%, 24.8%, 26.8%). Direct shear tests were performed. Soil suctions were measured by the filter paper method after direct shear tests. Direct shear tests and soil suction measurements were also performed on soaked samples. The soil suction versus moisture content curve was developed. The shear strength variation with respect to soil suction was found for all normal stresses. The soil suction versus moisture content curves bear a close relationship to the unsaturated shear strength of the soil. An increase in soil suction increases the shear strength. © 2009 Elsevier B.V. All rights reserved.
1. Introduction
2. The strength of compacted clays
Compacted soils are part of many earth structures. Therefore their shear strength is very important in geotechnical problems such as bearing capacity, slope stability, lateral earth pressures. Usually, to determine shear strength, laboratory samples are tested which have the same moisture content and dry density as the fill, but the fill can be wetted in the future. For example, the bearing capacity of shallow footings on compacted fills is commonly based on the compressive strength of the unsaturated soil. The strength measurements are often performed on soil specimens which have negative pore water pressures (suction). The assumption is then made for design purposes that conditions in the future will remain similar, but this may not be a realistic assumption. The fill can be wetted; the source of wetting primarily is rainfall. Rising groundwater may be a source of wetting. In addition, broken utility lines, utility trenches, street subgrades, permeable layers, gravel packed subdrains, all act as subsurface conduits that lead water to fill (Croney et al., 1958). The Ankara clay is an overconsolidated and fissured clay at most places, and may have swelling properties. In this study, soil suction and shear strength of unsaturated compacted Ankara clay were measured. The soil suction was measured by the filter paper method. Direct shear testing was used to measure shear strength. Additional direct shear tests were performed where the compacted samples were soaked and consolidated prior to shearing. The relationships between moisture content, soil suction and shear strength were found.
Previous studies on the shear strength of unsaturated compacted soils (Diamond, 1970; Brackley, 1973, 1975; Zein, 1985; Delage et al., 1996; Kong and Tan, 2000; Toll, 2000) show the development of an aggregated fabric in materials compacted drier than the optimum moisture content. This type of aggregation was not found to exist in the materials compacted on the wet side of the optimum moisture content. Vanapalli et al. (1996) measured the shear strength of statically compacted glacial till specimens using a modified direct shear apparatus. Specimens were prepared at three different water contents and densities (i.e. corresponding to drier than optimum, at optimum, and wetter than optimum conditions). Various net normal stresses and matric suctions were applied to the specimens. Soil-water characteristic curves were developed using a pressure plate apparatus on specimens. According to Vanapalli et al. (1996), “soils compacted at various “initial” water contents and to various densities should be considered as “different” soils from a soil mechanics behavioral standpoint even though their mineralogy, plasticity, and texture are the same. The engineering behavioral change from one specimen to another will vary due to differences in soil structure or aggregation.” Lambe and Whitman (1979) states that “for a given compactive effort and dry density, the soil tends to be more flocculated for compaction on the dry side as compared to compaction on the wet side (i.e. on the wet side the soil is more dispersive). In general, an element of flocculated soil has a higher strength than the same element of soil at the same void ratio but in a dispersed state.” Brackley (1973, 1975) proposed a model of unsaturated clay fabric in which the clay particles grouped together in ‘packets’. The soil
⁎ Corresponding author. Tel.: + 90 312 2102435; fax: + 90 312 2105401. E-mail address: [email protected] (E. Çokça). 0169-1317/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.clay.2009.08.028
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
within the aggregation or ‘packets’ is held together by suction, and the packets act as individual particles. The larger effective particle size produces more frictional behavior and hence a higher angle of friction. According to Toll (2000) “fabric plays a vital role in determining the engineering behavior of compacted soils. Clayey materials compacted drier than optimum moisture content develop an aggregated or ‘packet’ fabric. The presence of aggregations causes the soil to behave in a coarser fashion than would be justified by the grading.” Fredlund and Rahardjo (1993) used modified direct shear apparatus to measure the shear strength (consolidated-drained tests), and matric suctions were measured by pressure plate tests. Fredlund and Rahardjo (1993) state that “the suction influences the shear strength of unsaturated compacted specimens.”
3. Soil suction The total suction, ψ of a soil is made up of two components, namely, the matric suction, (ua – uw), and the osmotic suction, π. The matric suction component is commonly associated with the capillary phenomenon arising from the surface tension of the water. The pore-water in a soil generally contains dissolved salts; the decrease in relative humidity due to the presence of dissolved salts in pore-water is referred to as the osmotic suction π. The free energy of the soil water (total suction, ψ) can be determined by measuring the vapor pressure of the soil water or the relative humidity (RH) in the soil. The direct measurement of RH in a soil can be conducted using a device called a psychrometer (Fredlund and Rahardjo, 1993). The RH in a soil can be indirectly measured using a filter paper as a measuring sensor. The filter paper is equilibrated with the suction in the soil.
3.1. Soil suction measurements with filter paper method The working principle behind the filter paper method is that the filter paper will come to equilibrium with the soil either through vapor flow or liquid flow, and at equilibrium, the suction value of the filter paper and the soil will be the same (Fig. 1). When the psychrometer method is compared to the filter paper method, the filter paper method gives more consistent results (Bulut et al., 2000). If the filter paper is allowed to absorb water through vapor flow (non-contact method), then total suction is measured (Fig. 1). After equilibrium is established between the filter paper and soil in a constant temperature environment, the water content of the filter paper disc is measured. Then, by using a filter paper calibration curve of water content versus suction, the corresponding suction value is found from the curve, so the filter paper method is an indirect method of measuring soil suction. Therefore, a calibration curve should be constructed or be adopted, curves were presented for different filter papers in ASTM D 5298-92, Standard Test Method for Measurement of Soil Potential (Suction) Using Filter Paper in soil suction measurements.
Fig. 1. Non-contact filter paper method for measuring total suction.
401
The filter paper method can reliably be used with suctions from about 80 kPa to in excess of 6000 kPa, a much larger range than any other single technique (Chandler and Gutierrez, 1986). 4. Experimental study 4.1. Introduction The soil sample (Ankara clay) used in this study was taken from the METU (Middle East Technical University, Ankara Turkey) campus area and classified according to Unified Soil Classification System by using the test results of the sieve analysis, hydrometer analysis and Atterberg limits. Also specific gravity, maximum dry density and optimum moisture content (o.m.c.) (Fig. 2) were determined. The dry density versus moisture content curve of the sample was plotted by using the compaction test results in which standard Proctor compaction mould and hammer was used. Soil index and compaction properties are given in Table 1. Free swell and suction properties are given in Table 2. After the Proctor compaction test, soil samples were tested at optimum moisture content (w= 20.8%), drier than optimum (w= 14.8%, 16.8%, 18.8%) and wetter than optimum (w= 22.8%, 24.8%, 26.8%). Shear strengths were determined by using direct shear apparatus. All samples were sheared under 75 kPa,150 kPa, and 225 kPa normal stresses. After that the same samples were used in filter paper suction measurements as described in Section 3.1. 4.2. Procedure for direct shear test In this test program 14 sets of direct shear tests (ASTM D3080, 2003) were made and each set contains 3 individual direct shear tests. To prepare the samples, oven dried samples were mixed with an appropriate mass of water and left 24 hours in a plastic bag in a humidity room to form a homogeneous mixture. After that the samples were compacted dynamically by using a Proctor compaction mould. Then samples were taken out from the compaction mould by using direct shear mould and hydraulic jack. It is very important to prevent the soil sample from moisture loss since the sheared samples were used for suction measurements after the direct shear test. Therefore, the direct shear box was wrapped with nylon stretch film and covered with moisturized cloth after placing the sample in the direct shear machine. The emplaced samples were left one day for consolidation under normal stress of σn = 75 kPa, σn = 150 kPa, and σn = 225 kPa. Each set contains three stages and each stage finalized in two days. After consolidation, samples were sheared. In order to prepare and test a soaked sample, the same procedure was used but before the direct shear machine was switched on, the samples were soaked and left for 24 hours under weights giving normal stresses of 75 kPa, 150 kPa, and 225 kPa.
Fig. 2. Dry density versus moisture content.
402
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
Table 1 Index and compaction properties of the tested clay. Property
Value
Specific Gravity, (Gs) Liquid Limit, (LL) Plastic Limit, (PL) Plasticity Index, (PI) Clay Fraction, (% finer than 2 µm) Activity, (PI / % finer than 2 µm) Optimum Moisture Content, (wopt) Maximum Dry Density, (ρd) Classification (According to Unified Soil Classification System)
2.73 48% 21% 27% 67.9% 0.40 20.8% 1.67 Mg/m3 CL
Since the Consolidated Drained test procedure was followed, horizontal displacement rate was very important. Therefore after each consolidation day t50 (time to reach 50% consolidation) and t90 (time to reach 90% consolidation) values were determined by using the log time and root time method (ASTM D2435, 2003). Using the following equations the failure times were calculated for each sample: tf = 11:7 t90
ð1Þ
tf = 50 t50
ð2Þ
After a few trials, it was seen that time to failure (tf) was between 6 to 10 hours and lateral displacement δ to reach the peak soil strength was between 3 mm to 5 mm. So, by using the following formula the displacement rate was chosen as 1.0 × 10- 4 mm/sec, which was slower than calculated, to be on the safe side: V = δ = tf
ð3Þ
4.3. Soil total suction measurements with filter paper method After the direct shear test, the tested sample was taken out from the direct shear box. After taking samples for moisture content check, the remaining sample was used for filter paper suction measurements. In this study ash free Whatman no. 42 type filter papers were used. At least 75 percent volume of a glass jar was filled with the soil (Fig. 1). A ring type support was put on top of the soil to provide a noncontact system between the filter paper and the soil. Two filter papers were placed on the ring. Then, the glass jar lid was sealed. After that, the glass jar was put into a box and in a controlled temperature room for equilibrium. The suggested equilibrium period is at least one week. After the equilibrium period, the filter paper moisture content was measured to nearest 0.0001 g accuracy. The calibration curve obtained in this study for Whatman no.42 type filter paper, by following ASTM D5298-92 (1992) procedure, is given in Fig. 3. After obtaining the filter paper moisture content (w) value, the calibration curve (Eq. (4)) was used to get the suction (ψ) value of the soil sample. Log ψ ðkPaÞ = 5:1887 – 0:0741w
Fig. 3. Calibration curve of Whatman no. 42 type filter paper used in this study.
5. Test results 5.1. Shear strength test results The shear stress – shear displacement and shear strength – vertical stress relationships for undisturbed Ankara clay (at natural moisture content) are given in Figs. 4 and 5 as examples. Similar behaviour was observed for all moisture contents. 5.2. Total suction-moisture content relationships The moisture content versus log total suction relationship is given in Fig. 6 for as compacted samples (there is no normal stress acting on the sample). Fig. 6 indicates that the soil suction increases as the water content decreases at the dry side of optimum. In this range of water content the moisture content versus log suction behaviour is linear (regression equation is given on the Fig. 6). Above the o.m.c. it has an almost constant value. Similar behaviour was observed for σn = 75 kPa, 150 kPa and 225 kPa samples. At the wet side of the optimum the soil suction attains steady and is not a function of the moisture content. 5.3. Shear strength and total suction relationships The shear strength of the compacted samples which is defined as the maximum shear stress measured in the direct shear tests is plotted against total soil suction on the logarithmic scale as shown in Fig. 7 for the three normal stress ranges studied and for the soaked samples are given in Fig. 8. The moisture content of the data points (from left to right) are 26.8%, 24.8%, 22.8%, 20.8%, 18.8%, 16.8% and 14.8%. Fig. 7 shows that, for all normal pressures the change in total suction with respect to shear strength shows similar behavior (curves are drawn to show the general trend of data points). Towards the dry side of o.m.c., the shear strength increases. The shear strength versus log soil suction behavior depends on the moisture content and the normal stress level.
ð4Þ
Table 2 Swell and suction properties of the undisturbed sample. Test
Results
Free Swell Natural Water Content (w) Applied Normal Pressures (σn) Total Suction (Ψ)
2.4% 21% 75 kPa 3136 kPa
150 kPa 4110 kPa
225 kPa 4414 kPa
Fig. 4. Shear stress versus shear displacement graphs for undisturbed samples.
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
403
Fig. 7. Shear strength-total suction graphs for σn = 75, 150, 225 kPa.
gated. Experiments were done on samples compacted at optimum moisture content (w= 20.8%), drier than optimum (w = 14.8%, 16.8%, 18.8%) and wetter than optimum (w= 22.8%, 24.8%, 26.8%). The following conclusions were drawn:
Fig. 5. Shear strength versus vertical stress graph for undisturbed samples.
The shear strength versus soil suction behavior can be explained as: the clayey material tends to be more flocculated for compaction on the dry side of o.m.c. as compared to compaction on the wet side of o.m.c. Suction seems to generate a resistance to slippage at the contacts between the particles (or aggregates) as the moisture content decreases on the dry of o.m.c. It is noted that the role of clay aggregates on shear strength of the clay is substantially reduced at about o.m.c. The decrease in shear strength with increasing moisture content (at the wet side of o.m.c.), is attributed to decreasing soil suction. From Fig. 8, due to soaking, the initial moisture content loses its importance for the shear strength and total suction relationship. As normal pressure increases, shear strength increases, too. When the shear strength-total soil suction graphs of soaked samples are compared with as compacted samples, the trend for compacted samples disappears due to soaking.
The total soil suction increases as the moisture content decreases at the dry side of optimum, and in this range of moisture content the moisture content versus log suction behaviour is linear and above the o.m.c. it has an almost constant value. For all normal pressures change in total suction with respect to shear strength shows similar behavior. Towards the dry side of o.m.c., shear strength increases with decrease in moisture content. Due to soaking, the initial moisture content loses its importance for the shear strength and total suction relationship. As normal pressure increase, shear strength increases, too. An increase in soil suction increases the shear strength. This study shows the effect of the compaction moisture content and soaking on the shear strength of the compacted soil. Therefore, if there is a possibility of wetting of the compacted fill in the future, to properly study the shear strength behaviour of the unsaturated compacted soil, tests should be performed on the soil in its ‘as compacted’ condition and after soaking. Acknowledgments
6. Conclusions Effects of compaction moisture content and soaking on the unsaturated shear strength and suction of Ankara clay were investi-
Fig. 6. Total suction-moisture content graph for as compacted samples.
The help provided by the technician Mr. Ali Bal under the supervision of the writers is gratefully acknowledged. The writers are grateful for the help provided by Ms. Nilgün N. Çokça during the review stage of the manuscript. The writers also would like to thank the reviewers of the paper for their many helpful suggestions.
Fig. 8. Shear strength-total suction graphs for σn = 75, 150, 225 kPa for soaked samples.
404
E. Çokça, H.P. Tilgen / Applied Clay Science 49 (2010) 400–404
References ASTM D2435, 2003. Standard test method for one-dimensional consolidation properties of soils using incremental loading. American Society of Testing and Materials. Annual Book of ASTM Standards, Philadelphia 04.08, 1–10. ASTM D3080, 2003. Standard test method for direct shear test of soils under consolidated drained conditions. Annual Book of ASTM Standards, Philadelphia 04.08, 1–7. ASTM D5298-92, 1992. Standard test method for measurement of soil potential (suction) using filter paper. American Society of Testing and Materials. Annual Book of ASTM Standards, Philadelphia 15.09, 156–161. Brackley, I.J.A., 1973. Swell pressure and free swell in a compacted clay. Proceedings 3rd International Conference Expansive Soils, Haifa, Vol. 1. Academic Press, Jerusalem, pp. 169–176. Brackley, I.J.A., 1975. A model of unsaturated clay structure and its application to swell behavior. Proceedings 6th African Conference Soil Mechanics and Foundation Engineering, Durban. Balkema, Rotterdam, pp. 65–70. Bulut, R., Park, S.W., Lytton, R.L., 2000. Comparison of total suction values from psychrometer and filter paper. In: Rahardjo, H., Toll, D.G., Leong, E.C. (Eds.), Unsaturated Soils for Asia. Springer, Singapore, pp. 269–273. Chandler, R.J., Gutierrez, C.I., 1986. The filter paper method of suction measurement. Geotechnique 36, 265–268.
Croney, D., Coleman, J.D., Black, W.P.M., 1958. Studies of the movement and distribution of water in soil in relation to highway design and performance. HRB spec. Washington D.C. Report 40, pp. 226–250. Delage, P., Audiguier, M., Cui, Y.J., Howat, M.D., 1996. Microstructure of a compacted silt. Canadian Geotechnical Journal 33, 150–158. Diamond, S., 1970. Pore size distribution in clays. Clays and Clay Minerals 18, 7–23. Fredlund, D.G., Rahardjo, H., 1993. Soil Mechanics for Unsaturated Soils. John Wiley & Sons, Inc., pp.1-77, pp.247-258. Kong, L.W., Tan, L.R., 2000. Study on shear strength and swelling – shrinkage characteristic of compacted expansive soil. In: Rahardjo, H., Toll, D.G., Leong, E.C. (Eds.), Unsaturated Soils for Asia. Springer, Singapore, pp. 515–519. Lambe, T.W., Whitman, R.V., 1979. Soil Mechanics. Wiley, New York. Toll, D.G., 2000. The influence of fabric on the shear behavior of unsaturated compacted soils. Advances in Unsaturated Geotechnics. : Geotechnical Special Publication, vol. 99. American Society of Civil Engineers, pp. 222–234. Vanapalli, S.K., Fredlund, D.G., Pufahi, D.E., 1996. The relationship between the soilwater characteristic curve and the unsaturated shear strength of a compacted glacial till. Geotechnical Testing Journal ASTM 19, 19–28. Zein, A.K.M., 1985. Swelling characteristics and micro fabric of compacted black cotton soil. PhD. thesis, University of Strathclyde, Glasgow.