Introduction

There are numerous studies concerning engineering behavior of sand-clay mixtures, which are utilized in many field applications including nuclear waste containment, earth fill dams, and road pavement structures. In the literature, sand-clay mixtures were investigated by analyzing effects of fines content on several geotechical properties such as liquefaction (Karim and Alam 2014), matric suction (Shayea 2001), void ratio (Cubrinovski and Ishihara 2002), undrained shear strength (Thevanayagam 2000), and compressibility (Cabalar and Hasan 2013). In addition, effects of plasticity on several geotechnical properties are well documented in the literature (Vucetic and Dobry 1991; Albrecht and Benson 2001; Jafari et al. 2002; Stark et al. 2005; Bray and Sancio 2006). For instance, Vucetic and Dobry (1991) emphasized that plasticity index caused increase in shear modulus ratio and a corresponding decrease in damping ratio. Albrecht and Benson (2001) studied the volume change during desiccation by applying drying and wetting cycles on soils and results showed that soils of high clay content with large plasticity are likely exposed to volumetric shrinkage strain. Tiwari and Ajmera (2011) experimentally obtained the effects of plasticity, liquid limit and clay content on fully softened and residual shear strength values. The researchers well documented differences between fully softened and residual strength as a function of soil plasticity. Naeini and Jahanfar (2011) investigated the effect of varying soil plasticity on results obtained from direct shear test tests. As the plasticity values were increased up to a value of (16.7%) increase in undrained shear strength is followed with a decrease by increasing plasticity. Sudjianto et al. (2011) performed swelling tests on clays of different plasticities and concluded that one-dimension swelling is increased by increasing soil plasticity. Jafari et al. (2002) performed resonant column tests on clays of different plasticities. The results showed that the medium plastic clay has the highest maximum shear modulus value and the lowest damping ratio. Sridharan and Gurtug (2004) carried out swelling tests on five soils of different plastic values and concluded that volumetric change in montmorillonitic soils is higher than those of other types of clays. They showed that sand-clay mixtures of different fines content and plasticities have different swelling and compaction properties. Bayat et al. (2014) review effects of fines and its plasticity on undrained shear strength and pore water pressure. They concluded that increasing fines content caused a decrease in pore water pressure and rate of this decrease is higher for the mixtures with high plastic bentonite. Stark et al. (2005) indicated that highly plastic clay gains additional shear strength during healing. They concluded that highly plastic clay has greater shear strength in comparison with lowly plastic clay. Bray and Sancio (2006) carried out cyclic triaxial test on four soils of different plasticities showing that cyclic resistance increased regularly when PI > 12. Sawangsuriya et al. (2009) investigated and modeled maximum shear modulus values for various soils based on matric suction changes. The A and C parameters used in this model decreased with the clay content and plasticity values.

In this study, two types (bentonite and kaolinite) of clay were added to the quartz at the contents ranging from 0 to 100%, by dry weight. In this scope, 11 mixtures were used for each type of quartz-clay mixtures. Consistency limits, compaction, and Fall cone tests were performed on different mixtures to evaluate effects of clay content and its plasticity of on maximum dry density (MDD), optimum water content (ωopt), Atterberg limits, and undrained shear strength (su).

Materials and methods

Bentonite and kaolinite were mixed with quartz to obtain mixtures of different clay contents and plasticities. The chemical analysis of these commercially available materials is given in Table 1. Specific gravity of quartz was found to be 2.65 while MDD and ωopt values were found to be 10.27 and 19.50, respectively. Some geotechnical properties of quartz along with its sieve analysis are given Fig. 1.

Table 1 Chemical analyses of materials used during experimental study
Full size table
Fig. 1
figure 1

Particle size distribution of quartz used during experimental study

Full size image

The specific gravity of the kaolinite and bentonite were found to be 2.45 and 2.35, respectively. Standart compaction, Casagrande, plastic limit, and Fall cone experiments were performed on the mixtures. The mixtures were oven dried for 24 h at 105 °C before mixing. The dry mixtures were prepared by mixing clay and quartz in dry state until a homogeneous quartz-clay distribution was observed. ASTM D698 was followed during standart proctor experiment. The mixtures were compacted in three layers and each layer was densified using freely falling rammer from 0.305 m, applying 25 blows per layer. The MDD and corresponding water contents were obtained for each mixture. In Casagrande test, according to ASTM D4318, the samples were completely dried and a sample was taken from all the mixtures and sieved through No.40 sieve (0.425 mm). The liquid limit of the mixtures was determined as the water content corresponding to 25 strokes. Fall cone tests were carried out according to BS 1377 using a British Fall cone apparatus. The Fall cone apparatus includes a specimen cup of 55 mm in diameter and 45 mm in height. Water content corresponding to 20 mm of penetration is recorded as the liquid limit of specimen.

Results and discussion

The results of compaction, consistency limits experiments on quartz-clay mixtures were given in Table 1. Initially, compaction properties of the mixtures were determined. It should be noted that, repeatability of test results were ensured by carrying out three consecutive tests (Table 2). Figure 2 shows the results of the quartz-kaolinite mixtures. As kaolinite content was increased, MDD was increased up to 20% clay content then decreased. On the other hand, optimum water content decreased up to 20% clay content and then increased. Such behavior could be related to existence of a threshold fines content (FCt). Numerous studies emphasize that there is a specific fines content that control the mechanical behavior of sand-clay mixtures (Thevanayagam 2000; Xenaki and Athanasopoulos 2003). The researchers explained that below a certain FCt, coarse particles govern mechanical behavior of coarse-fine mixtures. Beyond FCt, fines control the mechanical behavior. Such behavior in MDD and ωopt of kaolinite quartz mixtures revealed that, FCt is around 20% fines content. The reason of such behavior could also be attributed to fill voids between the quartz sand particles by kaolinite up to a content of 20%. Similar results were reported by De Magistris et al. 1998 Chiu and Shackelford (1998), Komine (2004), and Purvana et al. (2012). De Magistris et al. (1998) mixed different sand types with bentonite and the compaction tests were carried out. The addition of bentonite increased the MDD up to certain FCt and then decreased. Optimum water content, on the other hand, increased continuously or decreased up to FCt (depending on sand type) then increased. Figure 3 shows the compaction values of the bentonite with quartz. The maximum dry density increased with the 10% bentonite content and then decreased. Optimum water content showed a steady increase with bentonite content. Compaction curves of different clay contents in Fig. 4 show (20%, 50%, 80%) the effects of plasticity on compaction results. The results of MDD and ωopt of different mixtures were compared in Figs. 5 and 6, respectively. The MDD results showed an increase up to the addition of 10% bentonite while it was found to be 20% for kaolinite, after a decrease was observed. The main difference between the two clay types is thought to be plasticity, compressibility, and matric suction. It was previously stated that the increases up to a certain FCt means clay particles have filled the voids of quartz particles. At this point, it was considered that bentonite filled voids between quartz particles with 10% bentonite content while in quartz-kaolinite mixtures this ratio was found to be 20%. In addition, compressibility of materials could be reason of such behavior. Simpson and Evans (2015) stressed that compressibility of sand-clay mixtures decrease up to 10% then increased. The authors attributed changes in MDD behavior to compressibility of mixtures. Compressibility of mixtures could decrease up to 20% therefore MDD could also be increased. As for bentonite, compressibility of mixtures could decrease up to 10% then increased. Furthermore, Park and Santamarina (2017) stated that plasticity of fines affect the control behavior of course-fines in compression index so different plasticity between kaolinite and bentonite might be reason that different FCt. In Fig. 5, mixtures prepared with kaolinite have greater MDD than those with bentonite at a given quartz content. Similar results were found by Kolay and Ramesh (2016). They mixed kaolinite and bentonite with sand and fly ash. Their results showed that at a given sand content or fly ash MDD of mixtures with kaolinite was higher than those with bentonite. The reason also could be higher void ratio of bentonite clay compared to kaolinite clay. Maio et al. (2004) and Jacobs et al. (2007) found similar results denoting that bentonite has a greater void ratio compared to kaolinite. In addition, the reason could be difference of compressibility properties between bentonite and kaolinite. Bentonite compressibility might be higher than kaolin; therefore, MDD of mixtures with bentonite is higher than that of kaolinite. Kolay and Ramesh (2016) concluded that bentonite has higher compressibility than kaolinite. They found that mixtures with kaolinite have higher MDD those of bentonite. The reason for this behavior might be that the bentonite has higher suction due to its higher surface area. Such an approach is also pointed out studies carried out by Murray (2000) and Shayea (2001).

Table 2 Summary of test results
Full size table
Fig. 2
figure 2

Compaction curves of quartz-kaolinite mixtures

Full size image
Fig. 3
figure 3

Compaction curves of quartz-bentonite mixtures

Full size image
Fig. 4
figure 4

Compaction curves for quartz with different clay types

Full size image
Fig. 5
figure 5

Variation of MDD of quartz with different clay types

Full size image
Fig. 6
figure 6

Variation of OMC of quartz with different clay types

Full size image

Fall cone test used for estimating several geotechnical properties such as liquid limit, plastic limit, sensitivity, compressibility, and undrained shear strength of fine grained soil (Kumar and Wood 1999; Feng 2001). Figure 7 presents the cone penetration-water content relationship of the quartz-kaolinite mixtures. Figure 7 clearly shows that, as the water content increases, the penetration increases. In addition, the penetration values increased up to 30% with the kaolinite at a given water content. Beyond 30%, the penetration values decreased. Penetration-water content relationship of bentonite and quartz mixtures are given in Fig. 8. As the bentonite content increases, the penetration value decreases at a given water content. Contrary to quartz-kaolinite mixtures, the penetration value was stable up to a FCt of 20%; however, it was continuously decreased for rest of the samples tested. This behavior indicated that the mixtures gain strength against decreasing penetration all addition of bentonite. This could be attributed to the differences in the plasticity of the different clays. The penetration-water content relations could be interpreted by using clay matrix water content penetration relations. Kumar and Wood (1999), Cabalar and Mustafa (2015) indicated that clay matrix water content governed Fall cone penetration. The authors used Eq. 1;

$$ {w}_c=\frac{m_w}{m_c} $$
(1)

where wc is water content of clay matrix, mw, and mc is masses of water and clay, respectively. Clay matrix water content-su relationships are present in Figs. 9 and 10. It was clearly seen that above 20% clay content penetration values were governed by clay matrix.

Fig. 7
figure 7

Variation of penetration with water content for the quartz-kaolinite mixtures

Full size image
Fig. 8
figure 8

Variation of penetration with water content for the quartz-bentonite mixtures

Full size image
Fig. 9
figure 9

Variation of penetration with clay matrix water content for the quartz-kaolinite mixtures

Full size image
Fig. 10
figure 10

Variation of penetration with clay matrix water content for the quartz-bentonite mixtures

Full size image

It has been indicated that the Fall cone testing is also a strength measuring device. Hansbo (1957) suggested that undrained su can be found by using Eq. 2.

$$ {s}_u=k\times \frac{mg}{d^2} $$
(2)

where, m is the mass of cone, g is the gravitational acceleration, k is a constant, which changes based on the angle of the cone and it is found to be 0.85 for the 30° British cone (Wood 1985), d is the penetration of cone. The su values found using Eq. 1 are shown in Fig. 11 for quartz-kaolinite mixtures. Figure 11 clearly shows that the undrained shear strength has decreased by up to 30% kaolinite then increased. This behavior is consistent with intergranular void ratio approach (Thevanayagam 2000). The researcher used the concept of intergranular void ratio (eg) and interfines void ratio (ef) instead of global void ratio (e) to clarify behavior of sand-clay mixtures. According to this approach, the mixture is considered to consisting of coarse grains, fine grains, and voids. Fines content less than FCt (%), the fine grains floats in the voids of coarse graines and decrease number of contact points between coarse grains and the mechanical behavior is mainly governed by the contact points of coarse grains. The mechanical behavior of the soil is reduced with addition of fines in this part due to contact points. Then fine grains fill voids between coarse grains, so the fine grains govern the main behavior of soil the case that fines content exceeds FCt. In this part, the concept of ef is used as a indicator of sand-clay mixtures behavior. Therefore penetration decrease up to 30% kaolinite content could be explained by theory of Thevanayagam (2000). In addition, Vallejo and Mawby (2000) concluded that porosity controls the shear strength of mixtures. It is reasonable that the values of the undrained shear strength values and the MDD values with kaolinite content behavior interchange same FCt. Besides, various formulas between water content and su were suggested. According to the critical state concept, slope of these curves could be related to compression behavior (Wood 1990). In this context, it is evident that as the kaolinite ratio increases, the slope of line is increased. This behavior shows that compressibility generally increases with kaolinite addition, which is similar to outcomes reported in literature (De Magistris et al., 1998; Kolay and Ramesh 2016). Figure 12 shows su values of quartz-bentonite mixtures. It was evident that undrained shear strength is controlled by water content of soil matrix. However, as bentonite content is increased, the su values were increased unlike kaolinite, there is no decrease in up to 30% (FCt) was observed. As a result of these different behaviors of bentonite and kaolinite clay, the undrained shear strength behavior of bentonite was controlled by viscous shear stress and viscous double layer water as Sridharan and Prakash (1999a) concluded, su decrease up to FCt was not observed. Since the particle arrangement forces control the undrained shear behavior, there might be decrease in contact points of quartz particles with the kaolinite additions therefore su values be decrease up to FCt. Water content—undrained su are proposed for each mixtures in Fig. 12. The compressibility, which is the slope of the curves, is clearly increases as the bentonite content increases. Since different clays of plasticity added plasticity values are different, it is desired to investigate the effect of these values on su values. Figure 13 includes variation of strength of quartz including different amounts of bentonite and kaolinite. It is evident that quartz with bentonite mixtures has greater su values than those including kaolinite for a constant water content. The plasticity and liquid limit difference between kaolinite and bentonite could be given as the reason behind this. Sridharan and Prakash (1999b) stated that liquid limit and the undrained shear strength parameter are consistent with each other in a consistent manner. Greater the liquidity of bentonite mixtures, higher the su values. Furthermore, additional shear stresses due to the suction forces (Kochmanová and Tanaka 2011; Vanapalli 2009; Likos and Jaafar 2014) could be evaluated as the reason of different behaviors. Bentonite has greater effective stress since it has more suction stress than kaolinite. This behavior could also be attributed to different plasticity of bentonite and kaolinite (Fig. 16). A similar interpretation was performed by Blatz et al. (2002) who made a review of strength and stiffness behavior of bentonite quartz mixtures. They found that as plasticity of mixtures increased su values also increased. Besides, at a given water content, it was clearly shown that the slope of the mixtures with bentonite is much higher than that of kaoline. This indicates that bentonite has a higher compressibility values than kaolinite. Similar results found by Kolay and Ramesh (2016). Liquid limit from fall cone, plastic limit values of quartz with clay additions are shown in Figs. 14 and 16, respectively. These figures indicated that the liquid limit values were increased with clay additives and it is obvious that bentonite has greater liquid limit value than kaolinite at a given clay content (Fig. 15). Similar behaviors were observed by Sridharan and Prakash (1999b), Tiwari and Marui (2003), and Shayea (2001). In addition, Fig. 16 clearly indicated mixtures with bentonite have higher plasticity than those with kaolinite. In addition, apparent liquid limit values are shown in Fig.15. Apparent liquid limit defines liquid limit properties for only clay matrix (Cabalar and Mustafa 2015). Comparision of liquid limit values (Fig. 14) and apparent liquid limit values (Fig. 15) clearly indicated that difference between maximum and minimum values of mixtures are clearly decrease in Fig. 15. It proves that liquid limit of mixtures are governed by clay particles as Cabalar and Mustafa (2015) concluded (Fig. 17). Comparison of the liquid limit values found in the Fall cone test with the liquid limit values found in the Casagrande test show in Fig. 18. Different formulas with R2 correlation between Fall cone and Casagrande liquid limit values have been proposed for quartz-kaolinite and quartz-bentonite mixtures.

Fig. 11
figure 11

Variation of undrained shear strength with water content for the quartz-kaolinite mixtures

Full size image
Fig. 12
figure 12

Variation of undrained shear strength with water content for the quartz-bentonite mixtures

Full size image
Fig. 13
figure 13

Variation of undrained shear strength with water content for the quartz with different clay types

Full size image
Fig. 14
figure 14

Liquid limit values as a function of clay content

Full size image
Fig. 15
figure 15

Apparent liquid limit values as a function of clay content

Full size image
Fig. 16
figure 16

Plastic limit values as a function of clay content

Full size image
Fig. 17
figure 17

Plasticity index values as a function of clay content

Full size image
Fig. 18
figure 18

Comparision of Casagrande and Fall cone liquid limit values

Full size image

Conclusions

An experimental study is carried out to evaluate the liquid limit, plastic limit, plasticity index, and undrained shear strength properties of two types of clay (kaolinite and bentonite)-quartz mixtures regarding to clay content, plasticity effect of clay contents. Experimental results indicated that liquid limit, plastic limit, and plasticity index values are increased with both kaolinite and bentonite content. Consistency limits of bentonite-quartz mixtures were greater than those including kaolinite. Compaction tests revealed that MDD of mixtures increase with clay content up to FCt then decreased. This ratio is determined to be 20% for mixtures with kaolinite while 10% for bentonite. Optimum water content on the other hand, increased with 20% kaolinite content then decreased while it was increased continuously for bentonite content ranging among 0% to 100%. Fall cone results indicated that clay inclusion increased the resistance aganist penetration. Undrained shear strength values obtained from the Fall cone test governed by water content of clay matrix. Undrained shear strength results indicated that su values decreased up to 20% kaolinite content then increased whereas bentonite additions increased su values continuously. Furthermore, quartz-kaolinite mixtures has greater maximum dry density that those with bentonite whereas quartz-bentonite mixtures has greater undrained shear strength than those with kaolinite.