Influence of fines on the resistance to liquefaction of a clayey sand

Transcription

1 Ground Improvement (24) 8, No. 1, Influence of fines on the resistance to liquefaction of a clayey sand R. BOUFERRA and I. SHAHROUR Laboratoire de Mécanique de Lille, University of Sciences and Technologies of Lille, Polytech Lille, France This paper concerns a study of the influence of fines on the resistance to liquefaction of a clayey sand. The study is carried out using a triaxial apparatus on a reconstituted sand clay mixture containing a low fraction of clay. After a description of the soil used in this study, the paper presents results of undrained compression tests together with cyclic liquefaction tests performed on soil samples containing different fractions of clay. Experimental results show that the increase in the fines content up to 15% reduces the resistance to liquefaction of the sand clay mixture. This result is due to the role of fines in reducing the dilatant tendency of the clayey sand used in this study. Keywords: clay; fines content; liquefaction; mixture; resistance; sand; triaxial; undrained Cet exposé concerne une étude de l influence des fines sur la résistance à la liquéfaction d un sable argileux. Cette étude est effectuée en utilisant un appareil triaxial sur un mélange reconstitué sable-argile contenant une faible fraction d argile. Après avoir fait une description du sol utilisé dans cette étude, nous présentons les résultats des essais de compression non drainée ainsi que des essais de liquéfaction cyclique effectués sur des échantillons de sol contenant différentes fractions d argile. Les résultats expérimentaux montrent que l augmentation allant jusqu à 15% de la teneur en fines réduit la résistance à la liquéfaction du mélange sable-argile. Ce résultat est dû au pouvoir des fines à réduire la tendance à la dilatance du sable argileux utilisé dans cette étude. Introduction This paper presents results of an experimental study on the influence of fines content (F C ) on the liquefaction resistance of a clayey sand. Previous works reported conflicting results on this topic. Laboratory and post-earthquake observations show that an increase in the fines content could lead either to an increase or to a decrease in the soil resistance to liquefaction (Shen et al., 1977; Ishihara et al., 198; Tokimatsu and Yoshimi, 1984; Seed et al., 1985; Georgiannou et al., 199; Erten, 1994; Guo and Prakash, 1999). The role of fines is a function of various parameters, but mainly the plasticity index, fines content, cementation, overconsolidation, soil fabric and stress history. Shen et al. (1977) showed that, for a sand allowing sand-to-sand contact, the presence of fines increases the soil resistance to liquefaction. Ishihara et al. (198) noted that the resistance to liquefaction of a soil is constant when the plasticity index is lower than 1; above this value, resistance increases in a significant way with the index of plasticity (I P ). Tokimatsu and Yoshimi (1984) indicated that sands containing more than 1% fines exhibit greater resistance to liquefaction than clean sands with the same SPT N-values. Seed et al. (1985) confirmed this observation. Georgiannou et al. (199) showed that the presence of fines in a soil leads to a more compressible soil fabric, and hence to a significant loss in the soil s resistance to liquefaction. Erten (1994) showed that the resistance to (GI 216) Paper received 31 July 22; last revised 22 August 23; accepted 1 September 23 liquefaction of soils decreases with the addition of fines of low plasticity index. Analysis of experimental tests on silt and a silt clay mixture (Guo and Prakash, 1999) showed that an increase in the plasticity index (PI) decreases the liquefaction resistance of the silt clay mixture in the low range of plasticity, whereas in the high-plasticity range the liquefaction resistance increases with increasing PI. Previous work concerned mainly silty soils; in this paper we present the results of cyclic liquefaction tests on a reconstituted clayey sand with low fines content. Analysis of the results of these tests together with compression undrained tests allows us to draw out the relevant features of the influence of fines content on the resistance to liquefaction of a low-plasticity clay sand mixture. Experimental programme and procedure The experimental study was carried out on soil samples that were reconstituted in laboratory by mixing different amounts of a commercial kaolin clay and Hostun-RF sand. The kaolin clay is composed of 67% silicates and 23% aluminium hydroxide. The Hostun-RF is widely used in France for academic research (Flavigny et al., 199; Al- Mahmoud, 1997); it is a fine sand with a mean particle diameter, D 5 ¼.47 mm. Values of the maximum and minimum voids ratios of this sand were measured according to the AFNOR standard NF P (AFNOR, 2). Tests X # 24 Thomas Telford Ltd

2 R. Bouferra and I. Shahrour gave e min ¼.57 and e max ¼.94. Fig. 1 displays the particlesize distributions of the clay and sand used in this study. The experimental programme included undrained compression and quasi-static cyclic tests for various values of the fines content (ratio of the fines mass to the mass of the soil sample) at a confining pressure, p9 c ¼ 1 kpa. Undrained compression tests were performed at five values of Percentage passing Hostun sand Kaolin clay Particle size: mm Fig. 1. Particle-size distributions of clean Hostun sand and kaolin clay the fines content (F C ¼ %, 5%, 1%, 15% and 2%), and cyclic liquefaction tests were conducted for three values of F C (%, 5% and 15%). For each soil, cyclic tests were performed for three values of the cyclic stress ratio, which allowed for the determination of the liquefaction curve. The plasticity index of the mixtures used in this study was determined using Cassagrande apparatus according to the AFNOR standard NF P (AFNOR, 1993). Tests showed a low plasticity index for these mixtures (I p smaller than 3%). Tests were conducted using a triaxial cell on a medium dense samples, 7 cm in height and diameter, at an initial voids ratio, e ¼.75 (density index, I D ¼.52). Undrained compression tests Figure 2 displays the results of undrained compression tests performed on clean sand and clayey sand at various values of fines content. It shows that the fines content significantly influences the soil response. Clean sand exhibits first a phase of increase in the pore pressure (u), followed by a phase of sharp decrease. The decrease in the pore pressure leads to an increase in the effective confining pressure ( p9 c ) Variation of pore water pressure, u: kpa Axial strain, ε 1 : % Axial strain, ε 1 : % Clean sand F C 5% F C 1% F C 15% F C 2% Fig. 2. Influence of fines content on the response of the sand clay mixture to undrained triaxial compression: stress deviator against axial strain; excess pore water against axial strain; plot of stress paths (mean effective pressure, stress deviator) 2

3 Influence of fines on the resistance to liquefaction of a clayey sand and, consequently, to an improvement of the sand resistance to undrained compression. The increase in the pore pressure results from the contracting tendency of the Hostun sand at low values of the stress ratio q/p9 (p9 and q denote the effective mean stress and the stress deviator), whereas the decrease in the pore pressure is related to the dilatant behaviour of this sand at high values of the stress ratio (q/p9). Results obtained with the clayey sand show that the presence of fines extends the phase of increase in the pore pressure, which indicates that the presence of fines attenuates the soil dilatancy, and consequently reduces its resistance to undrained compression. This result is clearly observed for a small fines content (F C ¼ 5%), and becomes more pronounced when the fines content, F C, increases to 15%. Beyond this value, the clayey sand presents other trends: the increase in fines content causes a recovery of the soil dilatancy and, consequently, an increase in the soil resistance to undrained compression, as illustrated by results of the undrained compression test performed at F C ¼ 2%. The attenuation of the soil dilatancy of the clayey sand can be attributed to the end 17 end end 4 end Fig. 3. Cyclic undrained tests on the sand clay mixture with fines content, F C ¼ 5%. Cyclic stress ratio, R C ¼.1;.2;.3 3

4 R. Bouferra and I. Shahrour role of fines in the creation of a more compressible soil fabric, as noted by Georgiannou et al. (199). Liquefaction tests Figure 3 displays the results of undrained cyclic tests performed on a soil containing 5% of fines for three values Table 1. Influence of fines content on the resistance to liquefaction of the sand clay mixture Clean sand Cyclic stress ratio, R C Number of cycles to cause liquefaction Sand clay Cyclic stress ratio, R C mixture Fines Number of cycles to content ¼ 5% cause liquefaction end 9 end end 4 end end Fig. 4. Cyclic undrained tests on the sand clay mixture with fines content, F C ¼ 15%. Cyclic stress ratio, R C ¼.15;.175;.2 4

5 Influence of fines on the resistance to liquefaction of a clayey sand of the cyclic stress ratio R C ¼.1,.2 and.4 (R c ¼ q max / (2 p9 c ), where q max denotes the amplitude of the stress deviator). It can be observed that cyclic loading induces an increase in the pore pressure to a value that is equal to the effective confining pressure ( p9 c ), indicating the occurrence of a full liquefaction. The increase in the pore pressure depends strongly on the cyclic stress ratio (R C ): consequently the number of cycles required to cause liquefaction depends on (R C ). It is equal to 2 cycles for a stress ratio R C ¼. 1, and decreases to 6 and 1 when the stress ratio (R C ) increases to.2 and.3 respectively. Table 1 summarises the variation of the number of cycles required to cause liquefaction with the cyclic stress ratio (R C ) for both the clean sand and the sand clay mixture with fines content ¼ 5% (Bouferra, 2). It shows clearly that the presence of clay dramatically reduces the soil resistance to liquefaction. Indeed, for the stress ratio R C ¼. 3, the sand clay mixture exhibits liquefaction after one cycle, whereas 14 cycles are required to cause the liquefaction of the clean sand. This result is consistent with the results of undrained compression tests, which showed that the presence of clay attenuates the dilatant behaviour of the soil, and consequently reduces its resistance to liquefaction. They also well agree with results reported by Georgiannou et al. (199) and Erten (1994), which showed that resistance to liquefaction of the soil decreases with the addition of fines. Figure 4 displays the results of cyclic tests performed on the sand clay mixture with fines content F C ¼ 15%. Tests were performed for three values of the cyclic stress ratio R C ¼. 15,. 175 and. 2. It can be observed that the three values of the cyclic stress ratio lead to a full liquefaction. For the cyclic stress ratio R C ¼. 15 liquefaction occurs after 1 cycles, while 5 cycles and 1 cycle are required to cause liquefaction for the cyclic stress ratio R C ¼.175 and.2 respectively. A comparison of these results with those obtained with the mixture containing 5% of fines (Fig. 3) shows that the increase in the fines content from 5% to 15% induces an important reduction in the soil resistance to liquefaction for stress ratio values (R C ) exceeding. 17. This finding is consistent with results of undrained compression tests, which showed that the increase in the fines content intensifies the contracting behaviour of the soil mixture. For the stress ratio R C ¼. 15, soil mixtures with F C ¼ 5% and 15% exhibit the same resistance to liquefaction. This result indicates that, for low values of the stress ratio (R C ), the fines content does not affect the soil response to undrained loading paths, as illustrated in the beginning of the undrained compression test (Fig. 2). Figure 5 illustrates the influence of the fines content on the liquefaction potential curve. It clearly shows that the presence of a small fraction of fines (smaller than 15%s) greatly reduces the soil s resistance to liquefaction. This reduction is augmented by the increases in the fines content, in particular for high values of the cyclic stress ratio (R C ), those that producing soil liquefaction rapidly (number of cycles requires to cause liquefaction less than 1). Conclusion This paper has presented a study of the influence of the presence of fines on the resistance to liquefaction of a reconstituted sand clay. Analysis was performed on medium dense mixtures with low values of fines content and Cyclic stress ratio, R c Number of cycles plastic index. Undrained compression tests showed that the increase in the fines content up to 15% reduces the dilatant tendency of the sand clay mixture, and consequently reduces its resistance to triaxial undrained compression. This result was confirmed with cyclic tests, which clearly show that the increase in the fines content in the range 15% leads to a significant reduction of the soil resistance to liquefaction. References Clean Hostun sand Sand clay mixture (5% clay) Sand clay mixture (15% clay) Fig. 5. Influence of fines content on resistance to liquefaction of sand clay mixture AFNOR (2) Norme NF P 94-59: Reconnaissance et Essais- Determination des masses volumiques minimale et maximale des sols non conérents. AFNOR. AFNOR (1993) Norme NF P 94-51: Sols: Reconnaissance et Essais- Détermination des linite s d Atterberg Limite de liquidité à la compelle Limite de plasticité au ronleau. AFNOR. Al Mahmoud M. (1997) Etude en laboratoire du comportement des sables sous faibles contraintes. Thèse de doctorat à l USTL, Lille. Bouferra R. (2) Etude en laboratoire de la liquéfaction des sols. PhD thesis, Université des Sciences et Technologie de Lille, France. Erten D. (1994) Effect of Fines Content on Liquefaction Potential of Sands. PhD thesis, Rutgers, State University of New Jersey-New Brunswick, UMI. Flavigny E., Desrues J. and Palayer B. (199) Le sable d Hostun RF. Revue Française de Géotechnique, No. 53, Georgiannou V. N., Burland J. B. and Hight D. W. (199) The undrained behaviour of clayey sands in triaxial compression and extension. Géotechnique, 4, No. 3, Guo T. and Prakash S. (1999) Liquefaction of silts and silt clay mixtures. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 125, No. 8, Ishihara K., Troncoso J., Kawase Y. and Takahashi Y. (198) Cyclic strength characteristics of tailing materials. Soils and Foundations, 23, No. 4, Seed H.B., Tokimatsu K., Harder L. F. Jr and Chung R. M. (1985) The influence of SPT procedures in soil liquefaction resistance evaluations. Journal of Geotechnical Engineering, ASCE, 111, No. 12, Shen C. K., Vrymoed J. L. and Uyeno C. K. (1977) The effect of fines on liquefaction of sands. Proceedings of the 9th International Conference on Soil Mechanics and Foundation Engineering, Tokyo, 2, Tokimatsu K. and Yoshimi Y. (1984) Criteria of soil liquefaction with SPT and fines content. Proceedings of the 8th World Conference on Earthquake Engineering, San Francisco, 3, Discussion contributions on this paper should reach the editor by 1 June 24 5