Measurement 50 (2014) 222228

Contents lists available at ScienceDirect

Measurement
journal homepage: www.elsevier.com/locate/measurement

Evaluation of kaolin slurry properties treated with cement

Mehdi Raftari , Ahmad Safuan A. Rashid , Khairul Anuar Kassim, Hossein Moayedi
Department of Geotechnics & Transportation, Faculty of Civil Engineering, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia

a r t i c l e

i n f o

a b s t r a c t
Kaolin clay has features high compressibility and also very low strength. Stabilization methods are normally used to improve the mechanical and chemical characteristics of cohesive soil. This study has examined the kaolin properties treated with cement using the unconned compression strength (UCS) test, direct shear test, and constant rate of strain (CRS) consolidation test. The strength characteristics of kaolin mixed with cement have been investigated using the UCS test and direct shear test. Then the consolidation behaviour of this treated soil was evaluated by performing the constant rate of strain (CRS) consolidation test. The selected cement content range was 5%, 7.5%, 10%, 12.5% and 15%. Water content was used at twice the liquid limit of kaolin in order to produce a homogeneous and workable sample to be placed inside a curing mould. All the samples were cured for 12 days. Based on the UCS results, it was found that the increment of the cement content led to an increase in unconned shear strength and elasticity modulus of the improved soil and it also caused the water content to decrease after curing. Although the internal friction angle is not considered in saturated clay soils, this experimental result shows that it can be improved by raising the amount of cement. The results of the CRS test indicated a decrease in the slope of the void ratio curve with an increase in cement content. In addition, the variations of void ratio are augmented by the increase of cement content in a constant effective stress. 2014 Elsevier Ltd. All rights reserved.

Article history: Received 19 October 2013 Received in revised form 18 December 2013 Accepted 31 December 2013 Available online 10 January 2014 Keywords: Slurry method Kaolin clay Cement Soil improvement

1. Introduction Kaolin clay has features high compressibility and also very low strength. In order to improve the mechanical characteristics of soil, a stabilization method is usually employed by mixing the cohesive soil with stabilizing agents [14]. According to previous research in the soil stabilizing eld, an increase in the amount of stabilizing agent leads to an increase in the compressive strength, based on the attributes of the soil and binder. In addition, they referred to cement as the strongest binder for stabilizing soft soil [57]. Within Asian countries the employment of cement is more popular than that of lime, not only because it is plentiful and costs less, but also because cement works

Corresponding authors.
E-mail addresses: raftari_m@yahoo.com (M. Raftari), ahmadsafuan@ utm.my (A.S.A. Rashid). 0263-2241/$ - see front matter 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.measurement.2013.12.042

more effectively than lime [8]. A cement composite material is produced due to interaction between clay and a stabilizing agent after the mixing process. These interactions increase the soil strength, which leads to reducing the settlement and improving the bearing capacity [1]. Due to the faster cement hydration process than for lime, the Coastal Development Institute of Technology (2002) and Chew et al. [4] reported that cement creates more strength during the initial period than lime with the same quantities. Cement binders are the most universal stabilizing agents that could be used for different types of soil conditions [6,9,10]. The two main reactions that produce cementitious materials occur during soil cement or lime treatment curing, namely hydration reaction and pozzolanic reaction [4,1113]. For soil cement, the reaction of the mineral with water in the soil produces cement hydration and calcium hydroxide, while for soil-lime treatment, lime hydration and calcium silicate and calcium aluminate are produced

M. Raftari et al. / Measurement 50 (2014) 222228

223

[14,15]. The calcium hydroxide (soil-cement) and calcium silicate and calcium aluminate (soil-lime) then contribute to the pozzolanic reaction. The hydration process and exchange action reduce the soil water content and thus the physical properties will be improved contributing to an increase in soil strength; however, the increase of the shear strength depends mainly on the pozzolanic reaction [16]. From the reaction, new pozzolanic products are developed that bind together the clay particles to produce a stronger soil matrix [13]. The process is completed within several weeks and the majority of the strength increase occurs during this period [15]. It has been documented that the degree of improvement is related to the amount of binder added, binder type and curing time [4,10,15,1719]. The strength increases almost linearly with an increase in the amount of binder [11,12]. For instance, Uddin et al. [20] found that by adding 10% of cement to soft Bangkok clay, the unconned compression strength increased by up to 10 times. The natural moisture content of the soil used ranged from 76% to 84%. It is evident that the strength and stiffness increase at very low cement content. Ductile behaviour was observed at low cement contents below 10%, with no obvious peak stress. At higher cement contents, the treated soil behaved as a brittle material with very high strength and at a strain up to 2% [20]. Terashi et al. [21] observed that the axial strain at failure for treated soil with Portland cement decreased with increasing unconned compressive strength. In this particular research the fundamental characteristics of kaolin clay treated with cement are examined by laboratory investigations. These results can be used for the design of a soil stabilization method and deep soil mixing.

portland cement (OPC) and kaolin. OPC powder with a specic gravity of 3.083.18 was purchased from Lafarge Malaysia Berhad. The mixture proportions depended on the amount of cement and water added to the kaolin needed to gain a certain strength of mixture, which was ascertained using several trial mix designs. The typical mineralogical composition of OPC used in this study is given in Table 2. All the specimens were prepared by mixing the kaolin and the OPC powder with de-aired water. Oven-dried kaolin powder and OPC were poured into the mixing bowl of a food mixer at a controlled room temperature of 23 C. The food mixer initially ran at a low speed of 50 revolutions per minute to avoid spillage for 1 min. The mixer was stopped after the colour of the powder changed uniformly. This technique was employed to ensure that the powder was well mixed. The water was then added, with the amount being sufcient to make sure that the sample could be homogeneously mixed and would have a high enough degree of workability to be placed inside a mould. This amount was twice the liquid limit (LL) of kaolin suggested by Rashid et al. [7] and Rashid [22]. The mixer was then started for 1 min at the same low speed to avoid spillage. It was then stopped and the material was scraped off the paddle and sides of the bowl before mixing was resumed for another 10 min. This procedure was conducted to achieve a homogeneous mix as suggested by the Japanese Geotechnical Society Standard [23]. The samples inside the mould were cured for 12 days at a controlled room temperature of 23 C and relative humidity of 95%. 3. Results and discussion The inuence of kaolin treated with different percentages of cement has been investigated. In order to obtain the stiffness parameters of cement-stabilized kaolin clay, the UCS test and direct shear test were used with British standard BS 1377: Part7:1990. Also, the CRS test was employed to examine the settlement behaviour. 3.1. Unconned compressive strength As was mentioned above, each specimen for the UCS test was made and cured according to stabilized soil specimens without compaction (JGS 0821). Then the mixture was poured in 3 layers into PVC tubes with a 38 mm internal diameter by 78 mm height. Air bubbles were removed by lightly tapping the mould against a rigid plate after pouring each layer of slurry. The above operations were terminated once air bubbles were no longer observed on the soil surface. A palette knife was used to trim the top of the soil before a cap was put on the top of the tube and sealed with tape. The results of UCS tests of cured samples at 12 days are presented in Figs. 16 and Table 3. In the unconned compression strength, qu improved with increasing cement content, which agrees with the results of earlier investigations. The increase in strength with cement content is attributed mainly to the cement hydration that leads to the dissociation of calcium ions that eventually react with

2. Material and methods The tests were run on kaolin clays using samples obtained from commercial powdered kaolin (L2B20) supplied by Kaolin (Malaysia) Sdn. Bhd. Their index characteristic values are summarized in Table 1. As kaolin powder has more than 80% of particles classied as clay particles (under 2 lm size) of specic gravity, Gs, 2.60 was used. Selection of the stabilized material and determination of the exact amount to be used as an additive were critical as they directly affect the structural performances of the ground improvement method. A soilcement mixture was prepared using a mix of ordinary

Table 1 Characteristics of Kaolin Clay. Properties Liquid limited Plastic limited Plasticity index Particle density Particle density Particle density Particle density Particle density Particle density Characteristics values 54 30 24 2.60 2.61 2.63 2.64 2.66 2.67

(g/cm3) (g/cm3) (g/cm3) (g/cm3) (g/cm3) (g/cm3)

kaolin Kaolin Kaolin Kaolin Kaolin Kaolin

+5% cement +7.5% cement +10% cement +12.5% cement +15% cement

224

M. Raftari et al. / Measurement 50 (2014) 222228

Unconfined Shear Strength (kPa)

Table 2 Typical mineralogical composition of ordinary portland cement (OPC). Oxides Calcium oxide (CaO) Magnesium oxide (MgO) Aluminum oxide (Al2O3) Ferric oxide (Fe2O3) Silicon dioxide (SiO2) Sulfur trioxide (SO3) Amount % 65 3 6 3 20 2.5

25 20 15

110 105 100 95

10 90 5 0 85 80 17.5

Table 3 Results of unconned compression test. Test number 1 2 3 4 5 Cement content % 5 7.5 10 12.5 15 Elasticity modulus 234 281 326.2 416.09 579.09 UC strength 8.4 13.16 16.96 27.68 43.5 Final water content% 102 98.1 95 92 90

2.5

7.5

10

12.5

15

Cement Content (%)

Unconfined shear strength (kPa) water content %

Fig. 2. The variation of unconned shear strength and water content at different amount cement contents.

soil silica and soil alumina leading to the formation of pozzolanic products. These pozzolanic products bind together the clay particles or clusters of clay particles (or clay minerals), and create a new bonded, stronger soil matrix. The cement hydration process and the resultant pozzolanic reaction can last for months, or even years. Therefore, with enough water after mixing, the strength of the treated soil is expected to increase over time. Fig. 1 shows the unit weight after curing of the three specimens (AC) with similar test conditions. Accordingly, variations of the unit weight for all specimens versus the amount of cement content at constant water content and curing time are plotted. It can be seen that an increase in the cement content will increase the unit weight of the specimen. This is due to the change in the shrinkage of the stabilized kaolin clayey soils with cement over the hydration process. As this process increases, by increasing both the cement content and given

time for the curing, the unit weight will be increased [13]. For a specied water content and curing time, the water content after curing decreased with increasing cement content. The offered reduction in Fig. 2 veries that some of the existing water content participates in the cement hydration process. Fig. 3 presents the linear equation of normalized after-curing water content (xi/xf) and different cement contents. Fig. 4 also shows the stressstrain relationship of treated soil for various admixed cement contents. Youngs soil modulus are obtained from the unconned compression strengths of 5%, 7.5%, 10%, 12.5% and 15% cement content, with a curing time of 12 days and remoulding water content twice the LL (Fig. 5). Stabilization with cement changes the material behaviour into a more brittle state. The higher stress was achieved at a lower strain of 0.05% for cement-treated samples. On the vertical axis of Fig. 6, the modulus of elasticity is plotted, which is dened by the secant modulus of elasticity in

Specimen Category

13

13.5

14

14.5

15

15.5

16

Unit Wegth (KN/m3)

15% cement sample A, B and C 10% cement sample A, B and C 5% cement sample A, B and C 12.5% cement sample A, B and C 7.5% cement sample A, B and C

Fig. 1. The UCS results for 12 days curing of samples.

Fig. 3. The linear equation of normalized after curing water content (xi/xf) and different amount cement contents.

Water content (%)

M. Raftari et al. / Measurement 50 (2014) 222228

225

50

40

35 30 25 20 15 10 5 0 0 0.05 0.1 0.15 0.2 0.25 0.3

Modulus of elasticity (kPa)

45

Axial Stress (kPa)

Unconfined compressive strength q u (Kpa

Axial Strain (%)

5% cement 12.5% cement 7.5% cement 15% cement 10% cement

Fig. 6. Linear equation of elasticity modulus and unconned compressive strength with increase cement contents.

Fig. 4. The stressstrain relationship of treated soil.

35 30
y = 579.09x

Axial Stress (kPa)

25 20 15 10
y = 281.53x y = 416.09x y = 326.28x

5 0 0

y = 234x

0.05

Axial Strain (%)

5% cement 12.5% cement 7.5 % cement 15 % cement 10 % cement

Fig. 5. Elasticity modulus in stressstrain curve at a half of unconned compressive strength.

was unconsolidated, the normal load was applied instantly. Rate of direct shear was controlled at 0.6 mm/ min. For the tests, normal stress of the 56.4, 110.9 and 219.9 kPa were chosen. Six specimens were prepared from each cement percentage. The specimens were kept in the laboratory at a temperature of 23 C and relative humidity of 95% until the tests were performed. As regards unstable slope analyses, where the maximum shear strength is the most important value, the shear parameters were computed using the maximum shear stresses. The effect of OPC on maximum shear stress for different cement content is shown in Fig. 7. Increasing the cement percentage leads to raising the shear yield point of kaolin slurry treated with cement. As expected, the addition of cement has a signicant effect on the cohesion of all specimens. Fig. 8 shows the cohesion variations with an increase in the cement amount. Fig. 8 illustrates that the increase in cement content had a negligible incremental effect on the internal friction angle of kaolin. This low growth is due to the cement reaction and formation of pozzolanic products. Eventually, these pozzolanic products will cause the aggregation of particles to be able to occupy larger areas and hence modify the effective grading of the soils. 3.3. Rapid consolidation test A constant rate of strain consolidation apparatus was used for investigation of the consolidation behaviour of kaolin treated with cement. The standard triaxial load frame was used for applying a constant rate of displacement. Thus, specimen size is often limited to 100 mm diameter with varying lengths using a laboratory apparatus [24]. In preparation, the CRS samples were cured into the moulds, which were larger than CRS rings, based on achieving soil disturbance in a minimum cutting and tting sample technique. The pore pressure in the tests with a high rate of strain increased quickly, which does not follow Darcys law [25]. For the majority of the researchers, determination of the strain rate in the rst steps of CRS is difcult for controlling the pore pressure.

the stressstrain curve at half the unconned compressive strength for various cement contents respectively. In this graph, an elasticity modulus (Es) was introduced and dened as Eq. (1).

Es K c qu

where Kc is the correlation factor of all the specimens, and qu is the unconned compressive strength for all stabilized soils respectively. 3.2. Direct shear strength Direct shear tests were performed following BS 1377: Part 7:1990 and were carried out on slurry kaolin treated with different amount cement samples based on JGS 0821. Samples were cured at 60 mm square mould and 25 mm height at 12 days. Given that the direct shear test

226

M. Raftari et al. / Measurement 50 (2014) 222228

140 120 100 80 60 40 20 0 0 50 100 150 200 250

Shear stress (kPa)

Normal Stress (kPa)

Untreated 10% cement Linear (Untreated) Linear (10% cement) 5% cement 12.5% cement Linear (5% cement) Linear (12.5% cement) 7.5% cement 15% cement Linear (7.5% cement) Linear (15% cement)

Fig. 7. Maximum shear stress versus normal stress.

Before starting the test, it is crucial to saturate the CRS soil sample to ensure the measurement accuracy of the pressures. Normalized strain rate (b) plays an important role in the strain rate estimation. Eq. (2) is applied to determine the strain rate that on it, cv is the consolidation coefcient, r is the strain rate, and Ho is the initial thickness of the specimen.

r H2 o cv

Computing the strain rate relating to b requires an evaluation of cv before beginning the CRS test. In 1993, Lee proposed a maximum input value of 0.1 for b but he had no idea about the minimum value. It has been observed that for small values of b (less than 0.1), the nite strain solution is closely approximated by the innitesimal strain solution. Based on this nding, a test analysis procedure

25

20 15

20 15 10 10 5 0 2.5 5 7.5 10 12.5 15 5 0 17.5

Cement content (%)

Cohession (kPa) Internal friction angle (degree)

Fig. 8. The variation of the cohesion and internal friction angle versus cement content.

Internal friction angle (degree)

30

25

that closely follows Wissa et al. (1971) is suggested. In this laboratory investigation, the strain rate was calculated 0.025 mm/min with the result of the ordinary consolidation test according to Eq. (2) by Smith and Wahls (1969). Fig. 9 is a schematic arrangement of the rapid consolidation cell equipment that was available in the geotechnical laboratory of the University of Technology, Malaysia. The size of the steel ring was 100 mm internal diameter and 30 mm height. The maximum capacity of the cell was 10 kN, which was applied by load frame and led to a contact pressure of 1250 kPa. This pressure allowed comparisons to be made with results from conventional consolidation tests on specimens consolidated to 1250 kPa. Furthermore, it caused the design parameters to be measured at the correct levels of stress [24]. After curing, the void ratio of specimens was between 1.25 and 1.4 with the initial water contents of twice the LL. Figs. 10 and 11 present the void ratio changing to the logarithm of effective vertical stress by running the stages of CRS test. The actual consolidation pressure was measured in the CRS test process. Increasing the amount of cement caused more deection and shifted the void ratio curve towards the more effective stress region. Furthermore, this deection and shift caused the consolidation yield stress to increase in the CRS test results. Also, Fig. 10 demonstrates that with the raising of the cement content, the pre-consolidation pressure increases slowly. This may be due to the higher initial water content. Based on Fig. 11, the variations in void ratio are augmented by the increase in cement content in a constant effective stress. The drainage behaviour of samples was presented for controlling the results of the CRS consolidation test in Fig. 12. It shows the amount of pore water pressure during the test time. The pore water pressure was changed according to the amount of the admixed cement content. Samples with higher cement content showed higher pore water pressure (Fig. 12). In addition,

Cohesion (kPa)

M. Raftari et al. / Measurement 50 (2014) 222228

227

10

Pore water pressure (kPa)

50

100

150

200

250

300

Time (min)
7.5% cement 15% cement 5% cement

Fig. 9. Equipment schematic diagram of the constant rate of strain test (Rapid Consolidation Cell Equipment, RACE).

Fig. 12. The pore water pressure during the time.

1.6 1.4
5% cement 7.5% cement 10% cement 12.5% cement 15% cement

Void Ratio (e)

1.2 1 0.8 0.6 0.4 0.2 0 1 10 100

1000

Fig. 10. Void ratio (e) log r0 Curve.

0 0.2 0.4

10

100

1000
5% cement 7.5% cement 15% cement

10000

Fig. 13. Variations of ub/r0 and effective stress.

Void Ratio (e)

0.6 0.8 1 1.2 1.4 1.6 1.8

it continues smoothly. In other words, the hydraulic gradient trends toward close to zero. Additionally, the curve shifts out with an increase in the amount of cement, which means that the hydraulic gradient is increased with more cement at constant effective stress. 4. Conclusion In this paper, kaolin properties treated with ordinary Portland cement were studied using laboratory tests. These results are fundamental information for the design of deep soil mixing. The following conclusions were drawn:  The increase in the cement content will increase the unit weight, and unconned compression strength. This process increases with both the cement content and given curing time. The increase in strength with the cement content is attributed mainly to the cement hydration that leads to the dissociation of calcium ions.

Fig. 11. Variation of void ratio (e) versus log r0 .

higher pore water pressure is measured in the bottom of the samples during the CRS experiments. As a way to verify the outcomes, variations of effective stress and ub/r0 are plotted as shown in Fig. 13. This illustrates that the trend of the curve goes downward signicantly at the beginning of the tests, and after ub/r0 = 0.05

228

M. Raftari et al. / Measurement 50 (2014) 222228

  

The dissolved silica and alumina eventually react with soil silica and soil alumina leading to the formation of pozzolanic products. For a specied water content and curing time, the water content after curing decreased with increasing cement content. After 12 days curing time, it decreased from 108% to 90% after mixing with 15% cement content. This means that some of the existing water content participates in the cement hydration process. The elasticity modulus linearly increased with increased cement content. Comparing the 5% and 15% cement content, the elasticity modulus increased by 147%. The addition of cement had a signicant effect on cohesion and the increase in cement content had a negligible incremental effect on the internal friction angle of kaolin. After curing, the void ratio decreased with the increase in the amount of cement. Increasing the amount of cement caused the consolidation yield stress to increase in the CRS test results. With the raising of the cement content, the pre-consolidation pressure increased slowly.

References
[1] D.T. Bergado, L.R. Anderson, N. Miura, A.S. Balasubramaniam, Soft ground improvement in lowland and other environments, ASCE (1996). [2] A. Porbaha, H. Tanaka, M. Kobayashi, State of the art in deep mixing technology: part II. Applications, Proc. ICE-Ground Improvement 2 (1998) 125139. [3] M. Topolnicki, In situ soil mixing, Ground Improvement (2004) 331 428. [4] S.H. Chew, A.H.M. Kamruzzaman, F.H. Lee, Physicochemical and engineering behavior of cement treated clays, J. Geotech. Geoenviron. Eng. 130 (2004) 696706. [5] H. Ahnberg, G. Holm, L. Holmqvist, C. Ljungcrantz, The use of different additives in deep stabilization of soft soils, in: Proceedings of the international conference on soil mechanics and foundation engineering, AA BALKEMA, New Dehli, 1994, pp. 11911194. [6] E.S. Stab, Development of design and construction methods to stabilize soft organic soils: design guide soft soil stabilization, industrial and materials technologies programme (Brite-EuRam III), European Commission, CT97-0351, Project No. BE, 2002, pp. 96 3177.

[7] A.S.A. Rashid, M.N. Norhazilan, Estimation of wall friction of chamber box using consolidation characteristic, Appl. Mech. Mater. 174 (2012) 21372141. [8] B.B. Broms, Stabilization of soft clay with lime columns, in: Proc. Seminar on Soil Improvement and Construction Techniques in Soft Ground, NTU, Singapore, 1984. [9] M. Kivelo, B.B. Broms, Mechanical behavior and shear resistance of lime/cement columns, in: Proceeding of the International Conference on Dry Mix Methods for Deep Soil Stabilization, 1999, pp. 193200. [10] M. Kitazume, State of practice report-Field and laboratory investigations, properties of binders and stabilized soil, in: Proc. of the International Conference on Deep Mixing Best Practice and Recent Advance, 2005, pp. 660684. [11] A. Porbaha, S. Shibuya, T. Kishida, State of the art in deep mixing technology. Part III: geomaterial characterization, Proc. ICE-Ground Improvement 4 (2000) 91110. [12] E. Cokca, Use of Class C y ashes for the stabilization of an expansive soil, J. Geotech. Geoenviron. Eng. 127 (2001) 568573. [13] G.A. Lorenzo, D.T. Bergado, Fundamental characteristics of cementadmixed clay in deep mixing, J. Mater. Civ. Eng. 18 (2006) 161174. [14] S. Saitoh, Y. Suzuki, K. Shirai, Hardening of soil improved by the deep mixing method, in: Proc. of the 11th International Conference on Soil Mechanics and Foundation Engineering, 1985, pp. 17451748. [15] K. Engan, K.S. Gijutsu, The Deep Mixing Method: Principle, Design, and Construction, CRC Press, 2002. [16] B. Broms, Lime and Lime Cement Columns, Spon Press, 2004. [17] R. Babasaki, M. Terashi, T. Suzuki, A. Maekawa, M. Kawamura, E. Fukazawa, JGS TC Report: factors inuencing the strength of improved soil, Grouting Deep Mixing (1997) 913918. [18] M. Terashi, Theme lecture: deep mixing brief state of the art, in: Proceeding of the 14th International Conference on Soil Mechanics and Foundation Engineering, Hamburg, Germany, 1997, pp. 2475 2478. [19] S. Horpibulsuk, N. Miura, D.T. Bergado, Undrained shear behavior of cement admixed clay at high water content, J. Geotech. Geoenviron. Eng. 130 (2004) 10961105. [20] K. Uddin, A.S. Balasubramaniam, D.T. Bergado, Engineering behavior of cement-treated Bangkok soft clay, Geotech. Eng. 28 (1997) 89 119. [21] M. Terashi, H. Tanaka, T. Mitsumoto, Y. Niidome, S. Honma, Fundamental properties of lime and cement treated soils (2nd report), Report of the Port and Harbour Research Institute, vol. 19, 1980, pp. 3362. [22] A.S.A. Rashid, Behaviour of Weak Soils Reinforced with Soil Columns Formed by the Deep Mixing Method, University of Shefeld, United Kingdom, 2011. [23] Japanese Geotechnical Society Standard, Practice for Making and Curing Stabilized Soil Specimens without Compaction, 2000. [24] K.A. Kassim, B.G. Clarke, Constant rate of strain consolidation equipment and procedure for stabilized soils, ASTM Geotech. Test. J. 22 (1999) 1321. [25] H. Ahmadi, H. Rahimi, A. Soroush, Investigation on the characteristics of pore water ow during CRS consolidation test, Geotech. Geol. Eng. 29 (2011) 989997.