J. Cent. South Univ. (2012) 19: 2323-2332
DOI: 10.1007/s11771-012-1278-9
Low secondary compressibility and shear strength of Shanghai Clay
LI Qing(李青)1, C. W. W. NG(吴宏伟)1, LIU Guo-bin(刘国彬)2
1. Department of Civil and Environmental Engineering, Hong Kong University of Science and Technology,Hong Kong 999077, China;
2. Department of Geotechnical Engineering, Tongji University, Shanghai 200092, China
? Central South University Press and Springer-Verlag Berlin Heidelberg 2012
Abstract: In order to investigate the compressibility, particularly the secondary compression behaviour, soil structure and undrained shear strength of Shanghai Clay, a series of one-dimensional consolidation tests (some up to 70 d) and undrained triaxial tests on high-quality intact and reconstituted soil specimens were carried out. Shanghai Clay is a lightly overconsolidated soil (OCR=1.2-1.3) with true cohesion or bonding. Due to the influence of soil structures, the secondary compression index Cα varies significantly with consolidation stress and the maximum value of Cα occurs in the vicinity of preconsolidation stress. Measured coefficients of secondary compression generally fall in the range of 0.2%-0.8% based on which Shanghai Clay can be classified as a soil with low to medium secondary compressibility. The effect of soil structures on the compressibility of Shanghai Clay is found to reduce with an increase in depth. Soil structure has an important influence on initial soil stiffness, but does not appear to affect undrained shear strength significantly. Undrained shear strengths of intact Shanghai Clay from compression tests are approximately 20% higher than those from extension tests.
Key words: Shanghai Clay; block sampling; secondary compressibility; soil structure; undrained shear strength
1 Introduction
Shanghai is situated on Yangtze River delta in eastern China. The geology consists mainly of alluvial and marine sediments formed during Quaternary period in the last three million years. Due to alternating warm and cold climate and sea level changes, the uppermost 50 m of the soil deposits are generally categorized into eight layers [1]. In this work, Shanghai Clay is specifically referred to the third and fourth layers which are the major soft clay layers. With the infrastructure development in Shanghai, underground earth structures such as deep excavations and tunnels are increasingly demanded. The thick soft clay layer is the most frequently encountered during construction of excavations and tunnels.
A number of researchers have investigated the properties of Shanghai Clay over the years. Shanghai soft clay is widely considered to have considerable secondary compression (or creep) due to its large void ratio and high natural water content [2]. GAO et al [3] investigated the effects of stress history, loading duration and load incremental ratio on secondary compression of reconstituted Shanghai soft clay. They found that the secondary compression index for reconstituted Shanghai Clay was not significantly changed with increasing consolidation stress up to 1 600 kPa. The relatively constant secondary compression index was also observed by SUN et al [4]. Based on true triaxial test results, YUAN et al [5] concluded that the shear strength of Shanghai soft clay was larger in vertical direction than that in horizontal direction. SUN et al [4] carried out a series of drained triaxial tests on remolded Shanghai soft clay to obtain strength parameters of the soil. However, most aforementioned studies have adopted reconstituted or remolded specimens which exhibit different soil behaviour from those in the field. Research on intact Shanghai Clay is rather limited. Moreover, the effects of soil structure on compressibility and shear strength are rarely investigated in these researches.
The objectives of this work are to investigate the one-dimensional compressibility, especially secondary compression behaviour, soil structure effects and undrained shear strength of Shanghai Clay and to compare the measured results with those of other typical natural soft clays worldwide. A series of long-term one-dimensional consolidation tests (up to 70 d) and undrained triaxial tests on high-quality intact and reconstituted Shanghai Clay have been carried out.
2 Geological conditions and soil sampling
Intact samples were taken during excavation of a deep basement located in Puxi area as illustrated in Fig. 1. Figure 2 shows the geological profile and soil properties at the sampling site. The stratigraphy consists of 2 m thick fill underlain by a thin layer of medium silty clay, under which there are soft silt clay and soft clay layers with thicknesses of 4 m and 9 m, respectively. Sandy silt and stiff silty clay layers are followed in succession. Both natural water content (wn) and void ratio (e0) increase with depth till 17 m, after which they decrease gradually. The natural water content of the soft (silty) clay is close to or larger than the corresponding liquid limit, suggesting its soft and compressible features. The main water table is generally 1.0 m below ground. The permeability of the soft clay is in the order of 10-9 m/s.
Block sampling method was adopted in an attempt to obtain high-quality samples. A 300 mm cubic soil block was firstly cut out when the basement was excavated to the designed depth of 8.5 m and 15.5 m (Fig. 2). The soil block was immediately enclosed by a piece of plastic sheet and aluminum foil successively. Soil samples were then carefully sealed with two layers of melt wax and two layers of cheesecloth alternatively. Strong wooden boxes were used to contain the samples and wet cheesecloth was placed between the box and the surface of the sample in order to minimize vibrations during transportation.
Fig. 1 Map of Shanghai showing location of sampling site
Fig. 2 Geological profile and soil properties at Puxi area
3 Soil composition and index property
3.1 Particle size distribution
Figure 3 shows the particle size distribution curves of samples at 8.5 m and 15.5 m depths. There is slight difference in soil compositions at these two depths. The sample at the depth of 8.5 m has 67% silt (0.002- 0.063 mm) and 33% clay (<0.002 mm), while the 15.5 m sample consists of 68% silt and 32% clay.
Fig. 3 Particle size distribution curves
3.2 True cohesion
In order to verify whether there is true cohesion, slaking tests were carried out. Figure 4 shows two 30 mm cubic block specimens of intact Shanghai Clay that have been soaked in distilled water for 80 d. Specimen A was placed with vertical axis parallel to the vertical axis of the site, while specimen B was placed with vertical axis perpendicular to the site vertical axis. Neither of the specimens shows any disintegration during the 80 d soaking period. It is indicated that intact Shanghai Clay indeed has true cohesion or bonding in the field [6].
3.3 Soil plasticity
Liquid limits (wL) of soil samples at 8.5 m and 15.5 m are 51% and 25%, and the plasticity indices (IP) are 45% and 18%, respectively. Figure 5 shows the plasticity chart for Shanghai Clay on the basis of Atterberg limits. Most data fall on or just above the A-line, which is a boundary between clay and silt. According to BS 5930 [7], Shanghai Clay is categorized as a clay with intermediate to high plasticity. Other index properties are summarized in Table 1.
Fig. 4 Slaking tests on block specimens of Shanghai Clay: (a) Specimen A; (b) Specimen B
Fig. 5 Plasticity chart for intact Shanghai Clay
4 Laboratory testing programme
4.1 Specimen preparation
In oedometer tests, a small soil block was trimmed close to the size of cutting ring using a wire saw before the ring was pushed in. The diameter and height of the specimen are 69 mm and 19 mm, respectively. When preparing a triaxial specimen, a soil block with dimensions of approximately 50 mm × 50 mm × 100 mm was firstly cut out from the large block sample. It was then trimmed manually to a cylinder with 38 mm in diameter and 76 mm in height in a soil lathe. In order to minimize moisture loss and hence soil disturbance, specimens were prepared in a temperature- and humidity-controlled room and the whole process lasted for 10 to 15 min.
4.2 One-dimensional consolidation test
Consolidation tests were carried out by using a standard oedometer with a steel ring of 69 mm in internal diameter and 19 mm in height. Silicone grease was lubricated on the inside of the ring in order to reduce side friction between the ring and the soil specimen. All the tests were carried out in a temperature-controlled room ((23±1) ℃) so that the effect of temperature on consolidation tests can be minimized.
Two conventional incremental loading (IL) tests and four long-term consolidation tests (LTC) were performed on intact soil specimens. Another IL test on a reconstituted specimen prepared according to procedures suggested by BURLAND [8] was performed as well. In IL tests, each specimen was loaded in increments with a load incremental ratio of unity and load incremental duration of 24 h. In LTC tests, the specimen was held under each stress level for a period of seven days, except under a certain stress level where the secondary compression was measured for a period of 70 d. Details of the tests carried out in this work are summarized in Table 2.
4.3 Undrained triaxial test
Undrained triaxial compression and extension tests were carried out on both intact and reconstituted specimens. A computer-controlled stress path triaxial apparatus fabricated by GDS Instruments Ltd was used. The apparatus had internal load cell, electrical pore pressure transducer at the bottom, external LVDT transducer and two hydraulic controllers for chamber pressure and back pressure [9].
The specimen was immediately mounted in the triaxial cell after trimming, with porous stones at both ends. Vertical and spiral filter paper cages were used. Each specimen was enclosed by a latex membrane with two o-rings clasped at both ends. Back pressure saturation was conducted subsequently and the specimen was considered to be fully saturated when B value was larger than 0.96. Specimens were isotropically consolidated to designed stress levels, i.e., 100, 200 and 400 kPa in this work. The specimen was then sheared at an axial strain rate of 4.5×10-2 h-1, which was appropriate for pore water pressure equalization under undrained condition. All specimens were sheared up to axial strain of 20%-25% in an attempt to identify the critical state of the soil.
5 Assessment of sample quality
Intact soils would inevitably experience disturbances during the process of sampling, transportation and specimen preparation. In-situ properties may change significantly when subjected to excessive disturbance. Therefore, it is necessary to assess soil disturbance and hence the quality of samples prior to interpretation of oedometer and triaxial test results so that reliable and representative soil parameters can be obtained. The developed volumetric strain, ?εv, when a soil specimen is reconsolidated back to its in-situ stress state, can be considered as an indicator of sample quality. LUNNE et al [10] pointed out that the change in void ratio relative to the initial void ratio, ?e/e0, where ?e is change of the void ratio resulting from reconsolidation and e0 is initial void ratio, was more systematically influenced by soil disturbance and hence a good indicator of sample quality.
Table 1 Summary of index properties
Table 2 Test conditions of one-dimensional consolidation tests
Figure 6 shows the measured values of ?e/e0 in oedometer and triaxial tests on intact soil specimens when they are reconsolidated to the estimated in-situ stresses. It can be seen that measured ?e/e0 values from triaxial tests are less than 0.04, while ?e/e0 values from oedometer tests fall within the range of 0.04-0.06. Based on the criterion suggested by LUNNE et al [10], soil samples used in triaxial tests can be classified as very good to excellent while the samples in oedometer tests are good to fair.
Fig. 6 Sample quality assessment using ?e/e0 method [10] (1- Very good to excellent; 2-Good to fair; 3-Poor; 4-Very poor)
6 One-dimensional consolidation
6.1 Consolidation curve
Figure 7 shows the e -
curves for intact Shanghai Clay. Results of IL test on a reconstituted soil specimen are also included for comparison. The number in bracket denotes the stress level at which the soil was held for a period of 70 d. The preconsolidation stresses 
for samples at 8.5 m and 15.5 m depths are about 85 kPa and 160 kPa, respectively. Given that the estimated in-situ vertical stresses 
are 70 kPa and 120 kPa, the overconsolidation ratios 
of soil samples at these two depths are 1.2 and 1.3. Consolidation curves are almost parallel when vertical stress is lower than the preconsolidation stress, with the 8.5 m curve locating above that of 15.5 m specimen. This is because soils at different depths have gained different structures during deposition history in the form of varied initial void ratios. When applied loading increases beyond the preconsolidation stress, these two curves tend to merge, indicating that the initial soil structures begin to break down. After 200 kPa, these two consolidation curves almost overlap with each other, giving rise to a similar compression index Cc (Table 3).
Fig. 7 Consolidation curves for intact and reconstituted specimens
Table 3 Summary of consolidation results
In LTC tests, two extended sustained loading periods up to 70 d (Table 2) were applied. One is close to the determined preconsolidation stress (i.e. 200 and 400 kPa) and the other is 800 kPa. It seems that the consolidation curve is not significantly changed by the 70 d sustained loading near preconsolidation stress. However, the other larger sustained loading (800 kPa in this work) makes the change of void ratio for the following loading increment smaller. This is because a part of natural structures may be recovered in a relatively short time. It should be noted that the initial void ratio of specimen LTC3 is smaller than that of LTC2, indicating different initial soil structures. The difference in soil structures is the reason for the discrepancy in the locations of consolidation curves between LTC3 and other two tests (LTC1 and LTC2).
The reconstituted soil specimen was prepared from a slurry, which was thoroughly mixed at a water content of 1.4 times the liquid limit wL in a consolidometer under a vertical stress of 100 kPa. Due to the side friction between the inner face of consolidometer and the sample, final vertical effective stress is about 65 kPa. The reconstituted soil sample has a similar consolidation curve to intact samples but with smaller Cc and Cs values. A summary of compression index Cc and swelling index Cs of soil specimens from different depths is given in Table 3.
6.2 Secondary compressibility
Secondary compressibility of Shanghai Clay is generally described in terms of secondary compression index which is defined as Cα=de/dlg t. It is the slope of e-lg t curve after the completion of primary consolidation.
Figure 8 shows variations of measured secondary compression index Cα with vertical effective stress. Each specimen shows a well defined maximum secondary compression index in the vicinity of preconsolidation stress. The maximum value of secondary compression index may correspond to a significant destructuration process which is discussed later on. The soil specimen at 8.5 m exhibits a more significant reduction in Cα than 15.5 m specimen because of different soil structures developed during deposition history. It is of interest to note that Cα value of the reconstituted specimen is 1/3 to 1/2 that of intact specimens at high stresses. This implies that the structures of intact specimens are not completely destroyed even at a stress up to 1 200 kPa. It is the remained the structures that give rise to the difference in secondary compression index between intact and reconstituted specimens [11].
Fig. 8 Measured secondary compression index Cα at varied vertical effective stress 
Figure 9(a) shows variations of Cα with time under varied vertical effective stress (test IL1). It can be seen that when the vertical effective stress is below preconsolidation stress, Cα is almost constant. It may be because soil structures have rarely been destroyed as a result of true cohesion (bonding) at low stresses. As consolidation stress increases, soil structures destroy gradually and hence secondary compression index increases. After soil structures break down significantly near preconsolidation stress, destructuration process slows down, giving rise to an reduction in Cα value at higher stresses. Figure 9(b) shows a typical long-term consolidation curve for Shanghai Clay. After the completion of primary consolidation, the secondary compression takes effect at a varied secondary compression rate. This observation differs from the results of GAO et al [3] that the secondary compression index of Shanghai soft clay was almost constant with time. The discrepancy may lie in the fact that soil structures retained by the intact soil in this work are different from those of reconstitued specimens adopted by GAO et al.
Fig. 9 Variations of secondary compression index with time: (a) Test IL1; (b) Test LTC1 and LTC2
6.3 Cα/Cc concept for Shanghai Clay
MESRI and GODLEWSKI [12] proposed Cα/Cc concept initially and pointed out for any soil, the ratio of secondary compression index (Cα) to the corresponding compression index (Cc) was constant for any time, effective stress and void ratio. The secondary compression can be considered as a continuation of the volume change initiated during primary compression [13]. It is implied that soils with high compressibility (i.e., large Cc value) would exhibit high rate of secondary compression as well based on the concept.
Figure 10 compares measured secondary compression indices of Shanghai soft clay together with other four different natural soft soils published in the literature. Experimental data on intact as well as reconstituted specimens are adopted in regression analysis. The regression line did not pass through the origin initially, with the intercept value of 7×10-4 on the Cα axis. AL-SHAMRANI [14] suggested that the regression line passes through the origin for the majority of inorganic clays. Therefore, the small intercept value is neglected and the regression line is made to pass through the origin. It can be seen that the value of Cα/Cc for Shanghai Clay is 0.034 with a correction coefficient of 0.94. By comparing the Cα/Cc values ranging from 0.01 to 0.07 for a variety of natural geomaterials [15] and the comparisons shown in the figure, it is clear that Shanghai Clay does not exhibit distinctively significant secondary compression characteristics as commonly assumed by many researchers and engineers in the mainland of China. On the contrary, the secondary compressibility of Shanghai soft clay is low.
Fig. 10 Comparison of Cα/Cc of Shanghai Clay with other four typical soft clays
6.4 Evaluation of secondary compressibility
The coefficient of secondary compression defined as εα=Cα/(1+e)=?ε/?lg t is commonly used to quantify secondary compressibility of a soil. According to the value of εα, geotechnical materials can be classified into six categories, ranging from very low secondary compressibility to very high secondary compressibility [13]. Figure 11 shows the relationship of coefficient of secondary compression εα with natural water content wn of Shanghai Clay. For comparison, other 13 different natural soil deposits are included as well. The coefficients of secondary compression of intact Shanghai Clay generally fall within the range of 0.2%-0.8%, based on which Shanghai Clay is classified as a soil with low to medium secondary compressibility. The low secondary compressibility is consistent with measured ground settlements induced by deep excavation and tunneling in Shanghai [19-20].
6.5 Evaluation of soil structure
In order to investigate the influence of structure of Shanghai Clay, consolidation test results are interpreted in terms of void index Iv proposed by BURLAND [8]:
(1)
where 
and 
are void ratios on the intrinsic compression line (ICL) corresponding to vertical effective stress of 100 and 1 000 kPa, respectively.
Fig. 11 Comparison of secondary compression index of intact Shanghai Clay with other clays worldwide [18]
Figure 12 illustrates the non-linear compressibility and the effect of progressive destructuration of Shanghai Clay. Compression curves measured from intact and reconstituted soil specimens are shown against the intrinsic compression line (ICL) and sedimentary compression line (SCL) proposed by BURLAND [8]. The consolidation curve from conventional IL test on reconstituted soil specimen agrees well with that predicted by BURLAND’s equation [8]. The difference between the two curves at stresses below 100 kPa may be caused by the structures developed during specimen preparation. For simplicity, the ICL predicted by BURLAND’s equation is adopted for analysis. It is clear that all intact soil specimens cross the ICL at relatively low stress and continue to keep above the ICL and touch the SCL when the vertical effective stress approaches the preconsolidation stress. The normalized consolidation curves gradually bend towards the ICL beyond the preconsolidation stresses without converging with the ICL. This means that soil structures are not destroyed completely even at 1 200 kPa. Convergence might have been achieved if higher vertical effective stress is applied. The difference in structures between intact and reconstituted specimens may result in the divergence of the normalized consolidation curves from ICL [21]. The distance between normalized consolidation curve and ICL is normally considered as an indicator of soil structure. It is, therefore, indicated that the effect of soil structures on compressibility is less significant for samples at shallow depth (8.5 m) than that for samples from greater depth (15.5 m). The difference becomes negligible when vertical stress exceeds preconsolidation stress. The decrease of structure effects with an increase in depth was also observed in stiff London Clay [21-22].
Fig. 12 Effect of progressive destructuration on compressibility
7 Undrained shear strength
Figures 13(a) and (b) show the deviatoric stress- axial strain relationship and stress paths from CIUC tests on intact Shanghai Clay. Deviatoric stress increases rapidly upon shearing and peak shear stress occurs at axial strain of 5%-10% for each specimen. At any particular strain, the deviatoric stress is found to increase with an increase in consolidation stress before shearing. All specimens exhibit small but noticeable drop in deviatoric stress, indicating a strain-softening behaviour. This is consistent with the observation that a failure plane is developed at the end of shearing. Stress paths from CIUC tests reflect a typical contractive response to shearing. Although shearing is terminated at an axial strain of 20%, slight change in deviatoric stress can still be observed. In this work, the specimen is assumed to reach critical state at 20% axial strain for simplicity. The critical stress ratio is about 1.21, which corresponds to a critical frictional angle of 30°.
As shown in Figs. 13(c) and (d), the stress-strain behaviour from CIUE tests is similar to that from CIUC tests except that deviatoric stresses decrease significantly at higher consolidation stresses (200 and 400 kPa). It is worth noting that the specimen tends to form a neck (having a much smaller area) appearing at about one third of the specimen height to the bottom during extension test. Necking generally results in non-uniform strains developed in the specimen and impedes the achievement of the critical state. Therefore, no well- defined critical state can be identified based on stress paths from CIUE tests. The maximum deviator stress failure criterion is used and hence the peak friction angle under extension is 
20°. It is also found that the initial slope of deviator stress and axial strain curve for intact specimen is steeper than that of reconstituted one. Small initial stiffness of reconstituted soil specimen may be induced by the destroy of soil structures during sample preparation.
Fig. 13 Stress-strain curves and stress paths: (a), (b) CIUC test; (c), (d) CIUE test
Figure 14 compares the undrained shear strengths (corresponding to peak shear strength) for CIUC and CIUE tests. Undrained shear strength appears to increase linearly with vertical consolidation stress prior to shearing for both cases. However, the undrained shear strength from compression test is consistently higher than that from extension test. This is because interparticle bonds and soil structures contribute to peak strength mobilized under compression, whereas, the interparticle bonds and interlocks are destroyed as soon as extension begins. It should be noted that there is no significant difference in undrained shear strengths for intact and reconstituted soil specimens at 200 kPa. One possible reason is that both specimens have similar water contents before shearing.
Fig. 14 Comparison of undrained shear strength between compression and extension tests
8 Conclusions
1) High-quality samples of Shanghai Clay can be obtained by using block sampling technique. The soil tested is lightly-overconsolidated with overconsolidation ratio of 1.2 and 1.3 at 8.5 m and 15.5 m, respectively. Slaking test shows that natural Shanghai Clay possesses true cohesion or bonding.
2) The secondary compression index Cα?of intact Shanghai Clay is almost constant when consolidation stress is smaller than or in the vicinity of preconsolidation stress, while Cα decreases significantly with time under consolidation stress larger than preconsolidation stress. This is likely to be attributed to the progressive destructuration process of the soil.
3) The value of Cα/Cc ratio of Shanghai Clay is approximately 0.034, which falls within the lower range of ratios for inorganic clay and silt. Based on the classification in terms of coefficient of secondary compression, Shanghai Clay can be classified as a soil with low to medium secondary compressibility.
4) Soil structures of intact Shanghai Clay break down gradually with the increase of vertical effective stress, whereas some structures still take effect at a stress up to 1 200 kPa. The influence of soil structure of intact Shanghai Clay on the compressibility is found to reduce with the increase in depth.
5) Undrained shear strengths of intact Shanghai Clay from compression tests are approximately 20% higher than those obtained from extension tests. Soil structure has an important influence on initial soil stiffness, but does not appear to affect undrained shear strength significantly.
References
[1] GAO Da-zhao, WEI Dao-duo, HU Zhong-xiong. Geotechnical properties of Shanghai soils and engineering applications [J]. ASTM Special Technical Publication, 1986: 161-177.
[2] XIE Ning, SUN Jun. Rheological behaviour of Shanghai clays [J]. Journal of Tongji University, 1996, 24(3): 233-237. (in Chinese)
[3] GAO Yan-bin, ZHU He-hua, YE Guan-bao, XU Chao. The investigation of the coefficient of secondary compression Cα in oedometer tests [J]. Chinese Journal of Geotechnical Engineering 2004, 26(4): 459-463. (in Chinese)
[4] SUN De-an, CHEN Bo, ZHOU Ke. Experimental study of compression and shear deformation characteristics of remolded Shanghai soft clay [J]. Rock and Soil Mechanics, 2010, 31(5): 1389-1394. (in Chinese)
[5] YUAN Ju-yun, YANG Xi-zhang, ZHAO Xi-hong, DONG Jian-guo. Experimental study on anisotropic characteristic of Shanghai soft clay [J]. Dam Observation and Geotechnical Tests, 1996, 20(2): 10-14. (in Chinese)
[6] ATKINSON J H, CHARLES J A, MHACH H K. Examination of erosion resistance of clays in embankment dams [J]. Quarterly Journal of Engineering Geology, 1990, 23: 103-108.
[7] BSI. BS 5930: Code of Practice for Site Investigations [S]. London: British Standard Institution, 1999.
[8] BURLAND J B. On the compressibility and shear strength of natural clays [J]. Géotechnique, 1990, 40(3): 329-378.
[9] CHEN R. Expetimental study and constitutive modelling of stress-dependet coupled hydraulic hysteresis and mechanical behaviour of an unsaturated soil [D]. Department of Civl and Enviromental Engineering, The Hong Kong University of Science and Technology, 2007.
[10] LUNNE T, BERRE T, STRANDVIK S. Sample disturbance effects in soft low plastic Norwegian clay [C]// Proceedings of Symposium on Recent Developments in Soil and Pavement Mechanics. Rio de Janeiro, Brazil, 1997: 81-102.
[11] MESRI G, ROKHSAR A, BOHOR B F. Composition and compressibility of typical samples of Mexico city clay [J]. Géotechnique, 1975, 25(3): 527-554.
[12] MESRI G, GODLEWSKI P. Time-and stress-compressibility interrelationship [J]. ASCE J Geotech Eng Div, 1977, 103(5): 417-430.
[13] MESRI G. Coefficient of secondary compression [J]. ASCE J Soil Mech Found Div-v 99, 1973(SM1): 123-137.
[14] AL-SHAMRANI M A. Application of the Cα/Cc concept to secondary compression of sabkha soils [J]. Canadian Geotechnical Journal, 1998, 35(1): 15-26.
[15] TERZAGHI K, PECK R B, MESRI G. Soil mechanics in engineering practice [M]. New York:John Wiley and Sons, Inc., 1996: 43-44.
[16] MESRI G. Primary compression and secondary compression [J]. Geotechnical Special Publication, 2003, 119: 122-166.
[17] NASH D F T, SILLS G C, DAVISON L R. One-dimensional consolidation testing of soft clay from Bothkennar [J]. Géotechnique, 1992, 42(2): 241-256.
[18] NG C W W, LI Qing, LIU Guo-bin. Characteristics of one-dimensional compressibility of Shanghai Clay [J]. Chinese Journal of Geotechnical Engineering, 2011, 33(4): 630-636. (in Chinese)
[19] LIU G B, NG C W W, WANG Z W. Observed performance of a deep multistrutted excavation in Shanghai soft clays [J]. Journal of Geotechnical and Geoenvironmental Engineering, 2005, 131(8): 1004-1013.
[20] LEE K M, JI H W, SHEN C K, LIU J H, BAI T H. Ground response to the construction of Shanghai Metro Tunnel-Line 2 [J]. Soils and Foundations, 1999, 39(3): 113-134.
[21] GASPARRE A, NISHIMURA S, COOP M R, JARDINE R J. The influence of structure on the behaviour of London Clay [J]. Géotechnique, 2007, 57(1): 19-31.
[22] GASPARRE A, COOP M R. Quantification of the effects of structure on the compression of a stiff clay [J]. Canadian Geotechnical Journal, 2008, 45(9): 1324-1334.
(Edited by HE Yun-bin)
Foundation item: Project(GRF618006) supported by the Research Grants Council of the Hong Kong Special Administrative Region, China
Received date: 2011-10-12; Accepted date: 2012-03-28
Corresponding author: LI Qing, PhD; Tel: +852-64815737; E-mail: lq_0501230@126.com
