Jtam-A4.dvi


JOURNAL OF THEORETICAL

AND APPLIED MECHANICS

55, 2, pp. 523-533, Warsaw 2017
DOI: 10.15632/jtam-pl.55.2.523

EXPERIMENTAL STUDY OF THE SHEAR RESISTANCE OF GRANULAR

MATERIAL: INFLUENCE OF INITIAL STATE

Abdelhamid Flitti, Noureddine Della

Laboratory of Material Sciences and Environment (LMSE), University Hassiba Benbouali of Chlef (Algeria)

e-mail: n.della@univ-chlef.dz

Ramiro D. Verástegui Flores

iMMC, Catholic University of Louvain (Belgium)

The shear strength of sand and itsmechanical properties can be affected by numerous para-
meters. This work presents an experimental investigationwhich aims to study the influence
of the fines content, the depositional method and the grain size on the shear strength of
Chlef sand. Tests were conducted with the shear box on two types of soil, the natural sand
and the clean sand-silt mixture. Dense samples (Dr=88%)were reconstituted through dry
deposition for each type of the material. An additional series of tests was carried out on
a medium dense natural sand (Dr = 52%) prepared by dry and wet (w = 3%) deposition
methods. All specimens were subjected to normal stresses of 100kPa, 200kPa and 300kPa
and therewasno immersion ofwater.The tests results showthat thebehavior of sandcanbe
affected by three parameters, the fines content, the depositionmethod and the particle size.
Themaximum shear stress and the friction angle decrease as the fines content increases, the
initial water content increases, the effective grain size diameter decreases and the uniformity
coefficient increases. The cohesion intercept increases with the increasing fines content and
decreasing initial water content. Overall, the samples prepared by the dry deposition me-
thod showmore resistance than those prepared by the wet deposition method. The results
obtained are generally in agreement with the previous research on drained and undrained
saturated sand in the literature.

Keywords: dry sand, fine content, water content, grain size, depositional method

1. Introduction

Soil of the Chlef region is vulnerable to earthquakes and its mechanical effects as the north of
Algeria is a part of the African tectonic plate. On October 10th 1980, the Chlef region was hit
by an earthquake of magnitude 7.3 [12], considered the strongest in its history. Much damage
occurred due to landslides, pavement deformation and liquefaction [3]. Due to all these facts,
the characterization of the mechanical behavior of soil of this region and especially the sand of
Chlef river is of relevance to mitigate and prevent similar disaster in the future.

The shear strength of soil was studied through the direct shear test. The experiment showed
that very similar shear strength results could be obtained on saturated sand and dry sand,
provided that the sand remained saturated and that drainage took place freely during shear,
and in both cases the effective stresses were equal to the total stresses (Head and Epps, 2011).

The behavior of saturated Chlef sand is the topic of study of many researchers, but studies
conducted on the dry chlef sand are rare in the literature.

The objective of this research was to study the effect of the fines content, the deposition
method and the particle size on the shear strength of the dryChlef sand (without saturation by
immersion in water) using the direct shear apparatus.



524 A. Flitti et al.

2. Literature review

The effect of the fines content and the sample preparationmethod on the liquefaction resistance
of saturated soil was the subject of some controversial research, because no consensus could be
found in the literature.

Researchers who studied the effect of the fines content on the resistance of saturated soils are
divided into three groups; some say that an increase in thefines content increases the liquefaction
resistance (Chang et al., 1982; Amini and Qi, 2000), others say that it reduces the liquefaction
resistance (Shen et al., 1977;Troncosco andVerdugo, 1985; Finn et al., 1994;Vaid, 1994; Zlatovic
and Ishihara, 1997; Arab, 2009; Belkhatir et al., 2013, 2014), while the third group of researchers
conclude that the liquefaction resistance decreaseswith an increasing fines content to aminimum
value, then it rises (Law and Ling, 1992; Koester, 1994; Bouferra and Shahrour, 2004).

The results obtained on the effect of the depositional method are not all in agreement, some
authors foundthat the samplespreparedbythesedimentationmethodpresentahigher resistance
to liquefaction than the samples preparedbyothermethods suchas thedry funnelpluviationand
the wet deposition (Zlatovic and Ishihara, 1997); others found that the liquefaction resistance
of the samples prepared by the wet deposition is larger than that of samples prepared by the
dry funnel pluviation (Mulilis et al., 1977; Yamamuro andWood, 2004). Canou (1989), Ishihara
(1993), Benahmed et al. (2004) found that the resistance of the samples prepared by the dry
funnel pluviation is more elevated than by the wet deposition method. The tests performed by
Della et al. (2009) on saturated Chlef sand confirmed this result, showing that the dry funnel
pluviation method gives stable samples (dilating) while the wet deposition method encourages
contractance.

It is known in the literature that the particle size significantly affects the resistance to
soil liquefaction. Whether from studies conducted in the laboratory (Lee and Fitton, 1968) or
in situ observations (Tsuchida, 1970; Seed and Idriss, 1971), many boundaries of particle size
distribution curves have been proposed to identify liquefiable soils. To study the effect of grain
size, a series of undrained tests were performed by Belkhatir et al. (2011). They found that the
undrained shear strength at the peak and the undrained residual shear strength decreased as
the coefficient of uniformity increased while the average diameter decreased and fines content
increased up to 50%.

Due to the lack of studies conducted on the unsaturated sand of Chlef region (Northern
Algeria), it was suggested to study the effect of the fines content, the deposition method and
the particle size on the shear strength of the dry Chlef sand (without saturation by immersion
in water) using the direct shear apparatus.

3. Material tested

The tests were realized on sand from the Chlef river (which crosses the city of Chlef to the west
ofAlgiers). However, two types of theChlef sandwere used; the natural sand and the clean sand
mixed with different fractions of silt (from 0% to 40%). TheChlef river silt shows low plasticity
with a plasticity index equal to 5.81%. Figure 1 showsmicrophotographs of the natural and the
clean sand. The properties of the natural sand, the clean sand-silt mixture and the silt used in
this study are illustrated in Table 1. The grain size distribution curves of the tested soils are
shown in Fig. 2.

Figure 3 shows the variation of the maximum and the minimum void ratio with the fines
content. It is clear that both decrease with the increasing fines content (Fc) up to a value
Fc = 30% beyond which they begin to increase following the same trend (similar observation
were reported by Belkhatir et al. (2013)).



Experimental study of the shear resistance of granular material... 525

Fig. 1. Microphotographs: (a) Chlef natural and (b) Chlef clean sand (zoom 50x)

Table 1. Index properties of materials used

Material
Fc GS

emin emax
D10 D30 D50 D60

CU CC
[%] [g/cm3] [mm] [mm] [mm] [mm]

Natural sand 1 2.700 0.586 0.946 0.171 0.311 0.463 0.538 3.153 1.055

Clean sand 0 2.652 0.632 0.795 0.266 0.431 0.596 0.700 2.634 0.999

Silty sand
(clean sand
+ fines
content)

10 2.654 0.536 0.703 0.077 0.369 0.549 0.643 8.304 2.733
20 2.655 0.458 0.697 0.029 0.298 0.510 0.616 21.622 5.058
30 2.657 0.449 0.687 0.017 0.087 0.420 0.535 30.630 0.811
40 2.658 0.504 0.759 0.011 0.057 0.307 0.437 38.305 0.662

Silt 100 2.667 0.991 1.563 – 0.015 0.029 0.036 – –

Fig. 2. Grain size distributed curves of the tested materials



526 A. Flitti et al.

Fig. 3. Maximum andminimum void ratios versus fines content

4. Experimental program

In this study, a series of testswere performedon the natural sand and the clean sand-siltmixture
for a relative density Dr = 88% (also Dr = 52% for some tests) under three normal stresses
σN =100kPa, 200kPa and 300kPa. To study the effect of the depositional method, dry or wet,
the initial water contents were set tow=0% and 3%, respectively.

The tests were performed using a square direct shear box 60×60mm2. The initial sample
height was 25mm. The test consisted in placing a sample in the shear box and subjecting it to
a vertical load N that represented the normal stress applied (100kPa, 200kPa, 300kPa) and a
horizontal load T which was gradually increased. The direct shear test allowed measuring the
peak and residual shear strength corresponding to every normal stress.

Two methods were used to set-up the sample, the dry deposition and the wet deposition
methods. In thewetdepositionmethod, thedrysandwasmixedthoroughlywitha small quantity
of water (3%) until a homogeneous soil sample was obtained. In the dry depositionmethod, the
sandwasdeposited in thedry state. To achieve the two relative densities, the samplewas divided
into three layers. Each layer was compacted to achieve the dense state (Dr = 88%), however,
no compaction was necessary to achieve the medium dense state (Dr = 52%), so only the
sample surface was leveled off. After the set-up, the samples were sheared at a constant speed
(1mm/min).

5. Test results

5.1. Effect of fines content

To study the effect of fines content on the shear strength of dry sand, five samples of sand-silt
mixture (Fc=0% to 40%) were tested under three normal stresses. Figure 4a shows the effect
of the fines content on the shear stress of dry sand (σN =300kPa,Dr=88%). For sampleswith
Fc=0%to20%, themobilized shear stress increasedwith the increasinghorizontal displacement
to reach the maximum value (located between 2 and 3mm) and then it gradually decreased;
whereas the mobilized shear stress of samples with Fc=30% and Fc=40% did not show the
peak value (Fig. 4a).



Experimental study of the shear resistance of granular material... 527

Figure 4b shows theMohr-Coulomb failure envelope that represents the relationship between
the maximum shear stress τmax and the normal stress σN according to the following formula

τmax =C+σN tanϕ (5.1)

whereC andϕ are the cohesion intercept and the friction angle, respectively.

Figure 4b shows clearly that the slope of the failure envelope decreases with the increasing
fines content. The reliability of these results is high considering the limited spread obtained for
each series of tests (R2 =0.97∼ 0.99).

Fig. 4. Effect of the fines content on the strength: (a) shear stress versus horizontal displacement
σN =300kPa,Dr=88%, (b) maximum shear stress versus normal stress

Fig. 5. Maximum shear stress (τmax) versus fines content (Fc), σN =100, 200, 300kPa,Dr=88%

The development of themaximum shear stress with the fines content is shown in Fig. 5. It is
clear from this figure that themaximum shear stress decreases with the increasing fines content



528 A. Flitti et al.

for the three normal stresses (σN =100, 200, 300kPa), but the decrease ismore pronounced for
σN =300kPa.

Figure 6 shows the effect of the fines content on themecanical properties (cohesion intercept
and friction angle) of the sand-silt mixture for the dense state (Dr=88%). It can be seen that
the cohesion increases with the increasing fines content (Fig. 6a). For the friction angle, it is
clear that it decreases lineary with the increasing fines content (Fig. 6b). This decrease in the
resistance is probably due to the presence of fine particles between grains of sand that promote
reduction in the contact between sand particles. The same effect was found by Arab (2009) on
the saturated sand.

Fig. 6. Variation of mechanical properties with the fines content (Fc),Dr=88%: (a) cohesion versus
fines content, (b) friction angle versus fines content

5.2. Effect of the mode of deposition

In order to study the effect of the depositional method (dry or wet), a series of tests were
performed on the natural sand. Two values of initial water content were used; w = 0% for
the dry case, and w = 3% for the wet case. Specimens at two relative densities were tested
(Dr = 52%, Dr = 88%) subjected to three levels of normal stresses (σN = 100, 200 and
300kPa).

The relationship between themobilized shear stress and thehorizontal displacement is shown
in Fig. 7. For the medium dense state (Dr = 52%), it is clear from Fig. 7a that the shear
stress of the dry specimens rapidly increases until a horizontal displacement of about 3mm
occurs. Beyond that, the increase is less significant. The results on the specimens prepared by
wet deposition show a mobilized shear stress that increases continuously with the increasing
horizontal displacement. In the dense state (Dr=88%), Fig. 7b shows that the shear stress of
the dry method rises to reach a peak value (between 3mm and 6mm), while the shear stress
of the wet method keeps increasing with the horizontal displacement as in the medium dense
specimens.

Figure 8 shows the variation of the maximum shear stress τmax (deduced from Fig. 7) with
the initial water content in the two states, the dense and themedium dense. It is clear that the
maximum shear stress of the dry specimens is higher than that of thewet specimens. This result
is in agreement with that obtained by Della et al. (2009) on the saturated sand. The difference
in the resistance between the dry and wet deposition method is more pronounced in the dense



Experimental study of the shear resistance of granular material... 529

Fig. 7. Shear stress versus horizontal displacement, σN =100, 200 and 300kPa: (a) medium dense state
(Dr=52%), (b) dense state (Dr=88%)

Fig. 8. Maximum shear stress (τmax) versus water content (w), σN =100, 200 and 300kPa,
Dr=52% and 88%

state. Furthermore, the difference in the resistance between the dense state and the medium
dense state is more apparent in the dry deposited samples than in the wet deposited samples.

The effect of the depositionalmethod in terms of the initial water content on themechanical
properties of sand is shown in Fig. 9. It is clear that the cohesion intercept decreases with the
increasing initial water content for both the dense and the medium dense state (Fig. 9a), but



530 A. Flitti et al.

the trend of the decrease is more pronounced in the dense state. Concerning the friction angle,
it can be seen from Fig. 9b that it also decreases with the increasing initial water content with
the same trend for the two relative densities. This decrease in the friction angle confirms the
previous result and allows saying that the dry samples aremore resistant than the wet samples.

Fig. 9. Variation of the mechanical characteristics with the initial water content (w), Dr=52%
and 88%: (a) cohesion versus water content, (b) friction angle versus water content

5.3. Effect of the particle size

Only the effective grain size diameter D10 and the uniformity coefficient Cu were chosen
as parameters to study the particle size effect on the shear stress and the friction angle of the
sand-silt mixture. For each normal stress level (σN =100, 200 and 300kPa), it was found that
themaximum shear stress increases according to a logarithmic trendwith an increasing effective
grain size diameter D10 (Fig. 10a). About the friction angle, Fig. 10b shows that it increases
with the increasing effective grain size diameter and decreasing fines content.
As it is known, theuniformity coefficient (Cu) is a crude shapeparameter (Holtz andKovacs,

1981) and it represents the ratio of the 60% particle size (D60) to the 10% particle size (D10)
(Head, 2006). Figure 11 shows the effect of the uniformity coefficient on themechanical behavior
of the sand-silt mixture. By increasing the fines content from 0% to 40%, the uniformity coeffi-
cient values range between 2.6 and 38.3, this is whyFigs. 11a and 11b present similar tendencies
and results with those of Figs. 5 and 6b, respectively showing that the maximum shear stress
and the friction angle decrease linearly with the increasing uniformity coefficient. These findings
are in agreement with those found by Belkhatir et al. (2011) on the saturated sand.

6. Conclusion

Results of experimental research on the influence of the fines content, the depositionalmethod in
terms of the initial water content and the grain size on the shear stress as well as themechanical
properties of Chlef river sand (natural and clean sand-silt mixture) have been presented.
Shear tests were performed in a direct shear apparatus. Two relative densities were evalu-

ated Dr = 52% and 88% for the natural sand using two sample preparation methods, the dry
deposition and the wet deposition, while only dense (Dr = 88%) specimens of clean sand-silt
mixture were tested using the dry deposition. The specimens were sheared dry without water
immersion and theywere subjected to three levels of normal stressesσN =100, 200 and300kPa.



Experimental study of the shear resistance of granular material... 531

Fig. 10. Effect of the effective grain size diameterD10: (a) maximum shear stress (τmax) versus the
effective grain size diameter, (b) friction angle versus the effective grain size diameter

Fig. 11. Effect of the uniformity coefficientCu: (a) maximum shear stress (τmax) versus the uniformity
coefficient, (b) friction angle versus the uniformity coefficient

Test results showed that the fines content, the depositional method and the grain size affect
the strength of thedry sand. Itwas found thatby increasing thefines content, the strength of the
sand-silt mixture decreases, the friction angle decreases and the cohesion increases. Concerning
the effect of the depositional method, it was shown that the maximum shear stress of the dry
deposited specimens was higher than that of the wet deposited specimens. Also, the friction
angle and the cohesion decreased with the increasing initial water content at deposition. About
the effect of the grain size, the tests illustrated that the maximum shear stress and the friction
angle increased with the increasing grain size diameter D10 and the decreasing fines content.
Furthermore, the results showed that themaximum shear stress and the friction angle decreased
linearly with the increasing uniformity coefficient and the fines content.



532 A. Flitti et al.

The results obtained from this study performed on the dryChlef river sand are in agreement
with those carried out on the saturated sand reported in the literature.

Acknowledgments

The testingwasperformed at the Laboratory ofMaterial Sciences andEnvironment (LMSE) atChlef

University and The Institute of Mechanics, Materials, and Civil Engineering (iMMC) of the Université

Catholique de Louvain (UCL). The authors express their gratitude to all who were assisting in the

preparation of this paper.

References

1. Amini F., Qi G.Z., 2000, Liquefaction testing of stratified silty sands, Journal of Geotechnical
and Geoenvironmental Engineering, 126, 3, 208-217

2. Arab A., 2009, Comportement monotone et cyclique dun sable limoneux, C.R. Mecanique, 337,
621-631

3. Assistance pour la réalisationd’une étudedemicrozonation sismiquedans la régiondeChlef (Phase
II), Résultats et recommandations du projet, Unesco, PNUD, Paris, 1985

4. Belkhatir M., Arab A., Della N., Schanz T., 2014, Laboratory study on the hydraulic
conductivity and pore pressure of sand-silt mixtures,Marine Georesources and Geotechnology, 32,
2, 106-122, DOI: 10.1080/1064119X.2012.710712

5. Belkhatir M., Arab A., Schanz T., Missoum H., Della N., 2011, Laboratory study on the
liquefaction resistance of sand-silt mixtures: effect of grading characteristics,Granular Matter, 13,
599-609

6. BelkhatirM., SchanzT.,ArabA., 2013,Effect of fines content andvoid ratio on the saturated
hydraulic conductivity and undrained shear strength of sand-silt mixtures, Environmental Earth
Sciences, DOI 10.1007/s12665-013-2289-z, 2469-2479

7. Benahmed N., Canou J., Dupla J.C., 2004, Structure initiale et propriétés de liquéfaction
statique d’un sable,C.R. Mecanique, 332, 887-894

8. Bouferra R., Shahrour I., 2004, Influence of fines on the resistance to liquefaction of a clayey
sand,Ground Improvement, 8, 1, 1-5

9. Canou J., 1989, Contribution à l’étude et à l’évaluation des propriétés de liquéfaction d’un sable,
Thèse de Doctorat de l’Ecole Nationale Des Ponts et Chaussées, Paris

10. Chang N.Y., Yeh S.I., Kaufman L.P., 1982, Liquefaction potential of clean and silty sands,
Proceedings 3rd International Earthquake Microzonation Conference, Seattle, Washington, 2,
1017-1032

11. Della N., Arab A., Belkhatir M., Missoum H., 2009, Identification of the behavior of the
Chlef sand to static liquefaction,C.R. Mecanique, 337, 282-290

12. El Asnam Algeria Earthquake of October 10, 1980, A Reconnaissance and Engineering Report,
Earthquake Engineering Research Institute, Berkeley, California, January 1983

13. Finn W.L., Ledbetter R.H., Wu G., 1994, Liquefaction in silty soils: design and analysis,
Ground Failures Under Seismic Conditions, Geotechnical Special Publication, ASCE, 44, 51-76

14. Head K.H., 2006,Manual of Soil Laboratory Testing, Volume 1: Soil Classification and Compac-
tion Tests, Third Edition,Whittles Publishing, Scotland, UK, p. 416

15. Head K.H., Epps R.J., 2011,Manual of Soil Laboratory Testing, Volume 2: Permeability, Shear
Strength and Compressibility Tests, Third Edition,Whittles Publishing, Scotland, UK, p. 499

16. Holtz R.D., Kovacs W.D., 1981, An Introduction to Geotechnical Engineering, Prentice Hall,
EnglewoodCliffs, New Jersey, 733 pages



Experimental study of the shear resistance of granular material... 533

17. IshiharaK., 1993, Liquefaction and flow failure during earthquakes,Géotechnique, 43, 3, 351-415

18. Koester J.P., 1994, The influence of fines type and content on cyclic strength,Ground Failures
Under Seismic Conditions, Geotechnical Special Publication, ASCE, 44, 17-33

19. Law K.T., Ling Y.H., 1992, Liquefaction of granular soils with non-cohesive and cohesive fines,
Proceedings of the Tenth World Conference on Earthquake Engineering, Rotterdam, 1491-1496

20. Lee K.L., Fitton J.A., 1968, Factors affecting the cyclic loading strength of soil, Symposium on
Vibration Effects of Earthquakes on Soils and Foundations, ASTM STP, 450, 71-95

21. Mulilis J.P., Seed H.B., Chan C.K., Mitchell J.K., Arulanadan K., 1977, Effects of
sample preparation on sand liquefaction, Journal of Geotechnical Engineering Division, ASCE,
103 (GT2), 91-108

22. Seed H.B., Idriss I.M., 1971, Simplified procedure for evaluating soil liquefaction potential,
Journal of the Soil Mechanics and Foundations Division, ASCE, 97, SM9, 1249-1273

23. Shen C.K., Vrymoed J.L., Uyeno C.K., 1977, The effects of fines on liquefaction of sands,
Proceedings of the 9th International Conference on Soil Mechanics and Foundation Engineering,
Tokyo, 2, 381-385

24. Troncosco J.H., Verdugo R., 1985, Silt content and dynamic behavior of tailing sands, Pro-
ceedings of the 12th International Conference on Soil Mechanics and Foundation Engineering, San
Francisco, USA, 1311-1314

25. TsuchidaH., 1970,Prediction and countermeasure against liquefaction in sanddeposits (in Japa-
nese),Abstract of the Seminar of the Port and Harbour Research Institute, Ministry of Transport,
Yokosuka, Japan, 3.1-3.33

26. Vaid V.P., 1994, Liquefaction of silty soils,Ground Failures Under Seismic Conditions, Geotech-
nical Special Publication, ASCE, 44, 1-16

27. Yamamuro J.A., Wood F.M., 2004, Effect of depositional method on the undrained behavior
andmicrostructure of sand with silt, Soil Dynamics and Earthquake Engineering, 24, 751-760

28. Zlatovic S., Ishihara K., 1997, Normalized behavior of very loose non-plastic soils: effects of
fabric, Soils and Foundations, 37, 4, 47-56

Manuscript received May 28, 2016; accepted for print October 14, 2016