Applied Clay Science xxx (2011) xxx–xxx

CLAY-02313; No of Pages 9

Contents lists available at SciVerse ScienceDirect

Applied Clay Science

j ourna l homepage: www.e lsev ie r .com/ locate /c lay

Research paper

Strength development in blended cement admixed saline clay

Suksun Horpibulsuk a,⁎, Worawit Phojan a, Apichat Suddeepong a, Avirut Chinkulkijniwat a, Martin D. Liu b,1

a School of Civil Engineering, Suranaree University of Technology, 111 University Avenue, Muang District, Nakhon Ratchasima 30000, Thailandb Faculty of Engineering, University of Wollongong, NSW 2522, Australia

⁎ Corresponding author. Tel.: +66 44 22 4322, +664607.

E-mail addresses: suksun@g.sut.ac.th, suksun@yahoo1 Tel.: +61 2 4221 3035; fax: +61 2 4221 3238.

0169-1317/$ – see front matter © 2011 Elsevier B.V. Alldoi:10.1016/j.clay.2011.10.003

Please cite this article as: Horpibulsuk, S.,doi:10.1016/j.clay.2011.10.003

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 February 2011Received in revised form 26 September 2011Accepted 5 October 2011Available online xxxx

Keywords:Saline clayBlended cementClay-water/cement ratioUnconfined compressive strength

Cement stabilization is extensively used to improve engineering properties of soft saline clays. The effect ofsalinity, which is modified by geological and climate changes, on the strength development in cementadmixed saline clay is investigated in this paper. For a particular curing time and salt content, the strengthdevelopment in saline clay admixed with cement is governed by the clay-water/cement ratio, wc/C. Thestrength increases with the decrease of wc/C. The increase in salt content for a particular water content de-creases the inter-particle attraction of the clay and the cementation bond strength. Hence, for the sameclay-water/cement ratio, the strength of the cement admixed saline clay decreases with increasing salt con-tent. In order to increase strength, and improve the economic and environmental impact, fly ash (FA) andbiomass ash (BA) can be used to substitute Portland cement. The influence of FA and BA on the strength de-velopment of cement admixed saline clay was investigated with unconfined compressive (UC) test and ther-mogravimetric (TG) analysis. FA and BA were dispersing materials, increasing the reactive surface of thecement grains, and hence strength increases as well. The clay-water/cement ratio hypothesis was used suc-cessfully to analyze and assess the strength development of blended cement admixed saline clay at varioussalt contents. An addition of 25% ash can replace up to 15.8% of cement.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Northeast Thailand covers more than one-third of the country of16.9 million ha with 9.25 million ha of agricultural land or 44% of thearable land. There are about 2.8 million ha of saline soil or 17% of thetotal area in northeast Thailand. The soils are classified as severe, mod-erate and slightly saline occupying areas of 240,000, 590,000 and2,020,000 ha, respectively (Yavaniyama et al., 2005). The surface salini-ty originates from the rock salt beads of theMahasarakam Formation ofthe Mesozoic Khorat Group. For wet lands, flood plains and lowlands,the saline soils are generally saturated and consist of soft clay deposits.The salinity fluctuates with time due to climate and geological varia-tions. The soft saline clays are highly compressible and have low shearstrength. One of the extensively used soil improvement techniques isthe in-situ cement stabilization via shallow and deepmixing. This tech-nique is economically viable because cement is readily available at rea-sonable cost in Thailand. Moreover, adequate strength can be achievedin a short time.

89 767 5759; fax: +66 44 22

.com (S. Horpibulsuk).

rights reserved.

et al., Strength developmen

The fundamental mechanical characteristics of cement admixedclays have been experimentally and numerically investigated exten-sively (Horpibulsuk et al., 2004a,b, 2010a; Kamon and Bergado, 1992;Kawasaki et al., 1981; Suebsuk et al., 2010, 2011; Terashi et al., 1979).These investigations mainly focused on the influence of water contentand cement content. The combination effect from the water contentand cement content is expressed by the clay-water/cement ratio(Horpibulsuk and Miura, 2001; Horpibulsuk et al., 2005; Miura et al.,2001). The clay-water/cement ratio, wc/C is defined as the ratio of claywater content to cement content (both in percentage by weight).While the clay water content reflects the microfabric of the soft clay,the cement content influences bonding of that fabric. Based on thisparameter and Abrams' law (Abrams, 1918), a phenomenologicalmodel has been proposed for predicting laboratory strength develop-ment in cement admixed clays at various water contents, cement con-tents and curing times (Horpibulsuk et al., 2003). This model wasrefined to develop a generalized strength equation for cement admixednon-swelling to low swelling clays (Horpibulsuk et al., 2011b).

For engineering, economic and environmental impacts, the re-placement of the cement by waste materials such as fly ash (FA)and biomass ash (BA) is attractive and valuable. In Thailand, the pro-duction of FA and BA exceeds by far their utilization. A feasibilitystudy of utilizing these waste materials to partially replace Type IPortland cement is thus considered significant. The role of FA onthe strength development in the blended cement admixed clay

t in blended cement admixed saline clay, Appl. Clay Sci. (2011),

Table 1Pore water chemistry of the saline clay.

Chemical composition Saline clay

1. Electrical conductivity (dS/m) 262. pH 5.33. Na (ppm) 3,7004. Ca (ppm) 2,1945. Mg (ppm) 259.86. K (ppm) 101.17. Chloride (ppm) 11,6008. Sulfate (ppm) 4009. Sulfur (ppm) 2010. Sodium percentage 6011. Sodium absorption ratio (SAR) meq/l 3.312. CEC (NF4OAC) (me/100 g) 1713. Exchangeable sodium percentage (%ESP) 95

Table 2XRF result of PC, FA, and BA.

Chemical composition (%) Silty clay PC FA BA

SiO2 67.04 20.90 44.72 74.12Al2O3 17.30 4.76 23.69 0.57Fe2O3 4.26 3.41 11.03 0.88CaO 0.65 65.41 12.67 1.54MgO 0.63 1.25 2.63 5.91SO3 0.07 2.71 1.28 3.33Na2O 1.57 0.24 0.07 1.71K2O 0.73 0.35 2.87 0.52LOI 7.76 0.96 1.42 11.42

2 S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

was investigated both in macro- and micro-scale experiments(Horpibulsuk and Raksachon, 2010; Horpibulsuk et al., 2009). Unlikeconcrete, the 28-day strength of cement admixed clay with FA is con-siderably lower; FA does not act as pozzolanic material because of thelow abundance of reactive Ca(OH)2. When the soft clay is mixed withcement, the clay and cement particles coalesce into large clay–cementaggregates due to physicochemical interaction (Horpibulsuk et al.,2010b). Fly ash leads to disaggregation of large clay–cement aggre-gates into smaller ones. The dispersion leads to the increase in the re-active surface, and hence strength enhancement. Particle size of FAdoes not affect disaggregation significantly.

By considering that the ash content can be equivalent to cementcontent, the clay-water/cement ratio hypothesis for blended cementadmixed clay (Horpibulsuk et al., 2011a) was developed as follows:“For given set of blended cement admixed clay samples, the strengthdevelopment depends only on the clay-water/cement ratio, wc/C,where the total cement content (C) is the sum of input of cement(Ci) and equivalent cement content (Ce)”. The equivalent cement con-tent (Ce) is equal to ka where k is dispersing factor and a is ash con-tent. Because the pozzolanic reaction is minimal, the Ce mainlydepends upon the dispersing effect, controlled by the ash contentand is not affected by the curing time. Hence, the k-value is practicallyconstant with curing time for a given combination of cementcontent and ash content. The thermogravimetric (TG) analysisshowed that the cementitious products of the blended cementadmixed clay samples are practically the same as long as the wc/Cvalue is the same. Based on Abrams' law and the clay-water/cementratio hypothesis for blended cement admixed clay, the strengthequation was proposed:

qu ¼ Awc

Ci 1þkað Þ� �B ð1Þ

where qu is the compressive strength of blended cement admixedclay at a specific curing time, wc is clay water content, and A and Bare empirical constants. The three parameters A, B and k for differentcuring times can be determined from a multi-regression analysis(MRA). The A-value increases with curing time and is mainly depen-dent upon the clay type. For non- to low-swelling clays, the B-valueis almost constant being equal to 1.27. k is independent of curingtime and ash type and it is about 0.75 for blended cement admixedBangkok clay. For a=0%, Eq. (1) is transformed to the form pro-posed by Horpibulsuk et al. (2011b).

The study on the effect of salinity on the engineering proper-ties is very limited. The change in the pore water chemistry, dueto the salinity change, modifies the clay-water interaction and af-fects the physical and engineering properties of remolded clay.Horpibulsuk et al. (2011c) showed that the generalized stressstate, e/eL (where e is void ratio and eL is liquid limit void ratio)is a useful parameter to interpret the intrinsic engineering proper-ties when their pore water chemistry is changed. As the e/eL de-creases, both the effective stress and shear resistance of theremolded clays increase.

The present paper investigates the influence of salt content onindex properties of saline clay to explain strength development incement admixed saline clay. The possibility of using the clay-water/cement ratio hypothesis to analyze the strength development inblended cement admixed saline clay is examined. The empiricalconstants B and k for different salt contents were also investigated.The B- and k-values obtained from the present work and previousworks (Horpibulsuk et al., 2003, 2011a,b) were analyzed to developa generalized strength equation for different blended cementadmixed clays. Two types of blended cements were used in thisstudy; namely the FA blended cement and the BA blended cement.This study and the generalized strength equation can be fundamental

Please cite this article as: Horpibulsuk, S., et al., Strength developmedoi:10.1016/j.clay.2011.10.003

for analyzing and assessing the strength development in other blend-ed cement admixed clays.

2. Laboratory investigation

2.1. Soil sample

Soil sample was collected from Phimai district, Nakhon Ratchasima,Thailand at a 2 meter depth. The Phimai clay is saline alluvium that isclassified as Typic Natraqualfs subgroup according to soil taxonomy.The soil sample is composed of 17% sand, 45% silt and 38% clay. It hasspecific gravity 2.65 and liquid and plastic limits 43% and 24%, respec-tively. Groundwater was ca. 1 m below the ground surface. The watercontent was 39% i.e. close to liquid limit. Based on the Unified Soil Clas-sification System (USCS), it is classified as low plasticity clay (CL). Thefree swell test (Prakash and Sridharan, 2004) showed that the clay isclassified as low-swelling type with free swell ratio (FSR) of 1.3. TheFSR is defined as the ratio of equilibrium sediment volume of 10 g ofoven-dried soil passing through a 425 mm sieve in distilled water (Vd)to that in kerosene (Vk). Thismethodwas employed because it is simpleand predicts the dominant clay mineralogy of soils satisfactorily(Horpibulsuk et al., 2007). The X-ray diffraction (XRD) results showthat the saline clay consists mainly of quartz and subordinate kaolinite.This is in agreement with the free swell test result, indicating that theclay has low swelling potential. Pore water chemistry was determinedby atomic absorption spectrometry (AAS) (Table 1). The chemical com-position of the clay was determined by X-ray fluorescence (XRF) analy-sis (Table 2). The grain size distribution curve of the saline clay is shownin Fig. 1. This clay has electrical conductivity, ECe 26 dS/m and sodiumadsorption ratio, SAR 3.34. According to the high ECe (higher than 15),this soil is classified as very strongly saline and dispersive (US SalinityLaboratory Staff, 1995).

2.2. Cement, fly ash, and biomass ash

Type I Portland cement (PC), fly ash (FA) from Mae Moh powerplant in the north of Thailand and biomass ash (BA) from Thai

nt in blended cement admixed saline clay, Appl. Clay Sci. (2011),

Fig. 1. Grain size distribution of saline clay, fly ash, biomass ash and Portland cement.

3S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

Power Supply Company Limited in Chachoengsao province were usedin this study. FA was a by-product from burning of brown coal and BAwas a by-product of rice husk, bark, board wood and seed husk.Chemical composition obtained from the XRF analysis of PC, FA andBA is given in Table 2. Both FA and BA were passed through a 45 μmsieve to remove larger particles. Total amount of the major compo-nents SiO2, Al2O3 and Fe2O3 in FA and BA are 79.44% and 75.57%, re-spectively. FA is classified as class F fly ash in accordance with ASTMC 618 whereas BA is referred to as “off-specification” because itmeets neither the class C nor class F criteria. The LOI of the BA ismuch higher than that of the FA because the latter was obtainedfrom high temperature (>950 °C) whereas the BA was obtained atca. 700 °C. Grain size distribution curves of PC, FA, and BA obtainedfrom the laser diffraction are also shown in Fig. 1. Specific gravitiesof PC, FA and BA are 3.15, 2.54, and 1.95, respectively. The PC andBA particles are irregular in shape whereas the FA particles arespherical.

2.3. Methodology

The saline clay was passed through a 2-mm sieve to removecoarser particles. The clay was submerged for a week to dissolve thesoluble salts and subsequently was mixed with salt to attain salt con-tents of 0.075, 1.3, 3, 5, 10 and 15% dry weight of clay. The salt wasobtained from a salt field in Phimai area. Its composition is shownin Table 3. NaCl is the main constituent of the salt (95.5%). Theindex tests were performed on the saline clay samples at differentsalt contents according to the American Society for Testing andMaterials (ASTM) standard (ASTM D 4318). The test was performedto illustrate the influence of salt content on index properties of salineclay to explain the strength development in cement admixed salineclay.

The role of salt content and ash content on the strength develop-ment was investigated based on the unconfined compression test re-sults. The water contents of the samples with different salt contentswere adjusted to the range of liquidity indices (IL) i.e., 1.0, 1.5, and2.0. The liquidity index was used in this investigation as an indicatorto refer the initial water content of the clay in relation to the plasticity

Table 3Chemical composition of the rock salt.

Chemical composition Content (%)

Sodium chlorine (NaCl) 95.5Moisture 3.0Insoluble substances 1.0Sulfate 0.35Calcium 0.15

Please cite this article as: Horpibulsuk, S., et al., Strength developmendoi:10.1016/j.clay.2011.10.003

characteristics before cement is admixed, similar to Horpibulsuk et al.(2003, 2011a,b) and Miura et al. (2001). This intentional increase inwater content simulates the water content increase, taking place inwet method of dispensing cement admixture in deep mixing. Theclay samples with their water content, corresponding to the abovelevels of IL were thoroughly mixed with the blended cements(PC+FA and PC+BA). The cement content, Ci, varied from 0 to30%, which is commonly used for the improvement of clay withhigh water content (Horpibulsuk et al., 2003, 2011a,b). FA and BAcontents, a, varied from 0 to 60% by weight of cement. The mixingtime was arbitrarily fixed at 10 min in accordance with Miura et al.(2001) and Horpibulsuk et al. (2003). The uniform pastes were trans-ferred to cylindrical containers (50 mm diameter and 100 mmheight), taking care to prevent any air entrapment. After 24 h, the cy-lindrical samples were dismantled, and wrapped in vinyl bags andstored in a humidity room of constant temperature (20±2 °C) for dif-ferent curing times. Unconfined compression (UC) tests were run onsamples after 7, 14, 28, 60 and 90 days of curing. The rate of verticaldisplacement in UC tests was 1 mm/min.

To examine the possibility of the clay-water/cement ratio hypoth-esis, the capability of the wc/C as a prime parameter in analyzing thestress–strain response and strength development in saline clay mustbe proved. Unconfined compression (UC) tests on the samples havingthe same wc/C for different water contents were carried out. The wc/Cvalues of 3, 5 and 10 at 28 days of curing were considered. The role ofash as dispersing material on the growth of cementitious productswas illustrated by the thermogravimetric (TG) analysis. The blendedcement admixed samples were broken from the center into smallfragments. The samples were frozen at −195 °C by immersion in liq-uid nitrogen for 5 min and were evacuated at a pressure of 0.5 Paat −40 °C for 5 days (Horpibulsuk et al., 2009, 2010b). Prior to TGtesting, the dried samples were ground in a ball mill and were sievedthrough a 150 μm sieve. Approximately 10–20 mg of the samples wastaken for the analysis. The heat rate was maintained at 10 °C/min andthe samples were heated up to 1000 °C. When heating the samples attemperature between 450 and 580 °C, Ca(OH)2 is decomposed intocalcium oxide (CaO) and water (El-Jazairi and Illston, 1977, 1980;Midgley, 1979; Wang et al., 2004) as in Eq. (2).

CaðOHÞ2→ CaO þ H2O ð2Þ

The variation in the abundance of the cementitious products(hydrated calcium silicates, hydrated calcium aluminates and hydrat-ed calcium aluminum silicates) can be expressed by the variation ofCa(OH)2 because they are the hydration products. Horpibulsuk et al.(2009, 2010b, 2011a,b) have successfully used this technique toapproximate the Ca(OH)2 of the cement admixed clay for explainingthe growth of cementitious products and strength development.

Fig. 2. Index properties of saline clay for various salt contents.

t in blended cement admixed saline clay, Appl. Clay Sci. (2011),

Fig. 3. Relationship between Ip and wL of the saline clay for various salt contents.

Fig. 5. Role of wc/C on the strength development in the saline clay for different watercontents and cement contents.

4 S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

Finally, a generalized strength prediction equation for the blendedcement admixed saline clay for different salt contents is suggestedbased on the clay-water/cement ratio hypothesis. The proposed equa-tion was verified by the separate tests of the BA blended cementadmixed saline clay for different water contents (wc=43 to 62%), ce-ment contents (C=10 to 30%) and ash contents (a=10, 15 and 25%).

3. Results

3.1. Strength development in cement admixed saline clay

Fig. 2 shows the influence of salt content on the index propertiesof the saline clay samples. As the salt content increases, the liquidlimit, the plastic limit and the plasticity index decrease. The decreasein liquid limit is due to the compression of diffuse double layer inkaolinite particles. Nevertheless, the relationship between plasticityindex, Ip and liquid limit, wL of all saline clay samples still lies abovethe A-line (Fig. 3). Since the relationship between Ip and wL is uniquefor all the saline clay samples, either Ip or wL can be used to explainthe variation in their intrinsic engineering properties with salt con-tent. In this study, the liquid limit, which is widely considered in cor-relating index properties to intrinsic properties (Burland, 1990;Horpibulsuk et al., 2007, 2011c; Nagaraj et al., 1998), was used.

The strength development in saline clay admixed with cement atthe same clay-water/cement ratio (same water content and cementcontent, wc=53% and C=20%) but different salt contents is shownin Fig. 4. As the salt content increases, the strength decreases. For aparticular water content, the generalized stress state e/eL decreases

Fig. 4. Strength development with time of the saline clay for various salt contents.

Please cite this article as: Horpibulsuk, S., et al., Strength developmedoi:10.1016/j.clay.2011.10.003

with the salt content due to the decrease of the liquid limit; therefore,the inter-particle attraction of the saline clay decreases. The presenceof the salt reduces both the inter-particle attraction of the saline clayand the cementation bond strength. It is well known that the salt isdetrimental for hydration. As such, the presence of salt decreasesthe effective stress (cementation bond strength and inter-particle at-traction) and the strength of the cement admixed saline clay.

Fig. 5 shows the typical stress–strain relationships in unconfinedcompression tests of cement admixed saline clay samples for differ-ent initial water contents and cement contents, but the same wc/Cvalues and salt content 1.3%. It is noted that the wc/C ratio is theprime parameter controlling the strength development at the con-stant salt content. As the wc/C value decreases, the strength increasesdue to the stronger cementation bond. Similar stress–strain behaviorof all the admixed samples, having the same wc/C values, is observed.

Fig. 6. Relationship between strength and unit weight versus fly ash content.

nt in blended cement admixed saline clay, Appl. Clay Sci. (2011),

Fig. 7. TG results of the blended cement admixed clay at different as contents.

Fig. 9. Relationship between strength and clay-water/cement ratio for BA blended ce-ment admixed saline clay.

5S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

It is concluded that the strength of cement admixed saline clay is con-trolled both by the clay-water/cement ratio and the salt content.

3.2. Role of fly ash on strength development

The strength development for the FA blended cement admixed sa-line clay is shown in Fig. 6, for 53% water, 20% cement and 28 days ofcuring. Both the strength and unit weight increase with fly ash con-tent and being maximum at a=25%. They decrease gradually withthe ash content when a exceeds 25%. Fly ash content in excess of25% possibly surrounds the cement grains and hinders the interactionbetween water and cement grains. Consequently, the abundance of

Fig. 8. Relationship between strength and clay-water/cement ratio for FA blended ce-ment admixed saline clay.

Please cite this article as: Horpibulsuk, S., et al., Strength developmendoi:10.1016/j.clay.2011.10.003

cementitious products decrease and so does strength when a>25%.Similar behavior was reported for Bangkok clay admixed with theFA and BA blended cement (Horpibulsuk et al., 2011a). The optimalash content is 25% and is independent of the cement content becauseit is determined in proportion to the cement content.

Fig. 7 shows the weight loss of the FA blended cement admixed sa-line clay at 53% water for different cement contents and fly ash con-tents. All the samples contain the same amount of cement. For therange of ash content tested (a≤25%), the weight loss for all theblended cement admixed samples is higher than that of the cement

Fig. 10. Relationship between strength and clay-water/cement ratio of FA blended ce-ment admixed saline clay for various salt contents.

t in blended cement admixed saline clay, Appl. Clay Sci. (2011),

Fig. 11. Strength development with time for the FA blended cement admixed saline clay.

6 S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

admixed sample. Besides the Ca(OH)2, kaolinite, which contains ap-proximately 14% crystalline water, dehydrolyzes at the temperaturebetween 450 and 580 °C. The increase in weight loss with ash contentindicates that the input ash enhances the amount of Ca(OH)2 becausethe amount of soil in the mix slightly decreases with the addition of

Fig. 12. Strength development with time for th

Please cite this article as: Horpibulsuk, S., et al., Strength developmedoi:10.1016/j.clay.2011.10.003

the FA. This higher amount of Ca(OH)2 is associated with the higherstrength for the blended cement admixed clay. In contrast, for flyash concrete, the strength increase due to pozzolanic reaction is asso-ciated with decrease in Ca(OH)2 content (Chindaprasirt et al., 2004,2006; Sinsiri et al., 2006). This result confirms that the pozzolanic

e BA blended cement admixed saline clay.

nt in blended cement admixed saline clay, Appl. Clay Sci. (2011),

Fig. 13. Generalized strength development for the FA and BA blended cement admixedsaline clay.

7S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

reaction is not important for the blended cement admixed clay. Thecontribution of the ash to the strength development is mainly dis-persing effect.

Based on Eq. (1), the analysis of strength development for the ef-fective dispersing range (a≤25%) is performed (Figs. 8 and 9). Thegeneralized two-dimensional qu-wc/C plots for the FA and BA blendedcement admixed saline clay with 1.3% salt at different curing timesare shown. Similarly, the role of salt content on the strength develop-ment for 7 and 28 days of curing is illustrated in Fig. 10. It is evidentthat the clay-water/cement ratio hypothesis can be applied to cementadmixed saline clay with a wide range of salt content. The variation ofparameter A is marked and depends on curing time and salt content.

Table 4Strength prediction of the BA blended cement admixed saline clay.

Time(days)

Water content, wc

(%)cement content, Ci(%)

BA content, a(%)

7 43 10 157 43 10 207 43 30 157 43 30 207 53 10 157 53 10 207 53 30 157 53 30 207 62 10 157 62 10 207 62 30 157 62 30 2028 43 10 1528 43 10 2028 43 30 1528 43 30 2028 53 10 1528 53 10 2028 53 30 1528 53 30 2028 62 10 1528 62 10 2028 62 30 1528 62 30 2090 43 10 1590 43 10 2090 43 30 1590 43 30 2090 53 10 1590 53 10 2090 53 30 1590 53 30 2090 62 10 1590 62 10 2090 62 30 1590 62 30 20

Please cite this article as: Horpibulsuk, S., et al., Strength developmendoi:10.1016/j.clay.2011.10.003

The strength of cement admixed saline clay decreases with salt con-tent; therefore, the A-value decreases as the salt content increases.The values of B and k are practically identical for both the FA and BAblended cement admixed saline clay and irrespective of water con-tent, cement content, ash content and curing time. The variations ofparameters B and k are very small (1.25–1.27 and 0.74–0.78, respec-tively) for both the FA and BA blended cement admixed saline clay.On the contrary, the k-value of the fly ash concrete increases signifi-cantly with curing time (Papadakis and Tsimas, 2002). It is concludedfrom the present work and the previous works (Horpibulsuk et al.,2009, 2011a,b) that the dispersing effect is independent of salt con-tent and grain size and shape of the ashes used. The B- and k-valuescan be taken as 1.27 and 0.75 for the non- to low-swelling clays.

From Eq. (1), at a particular curing time, the strength ratio at dif-ferent clay-water/cement ratio is proposed as follows:

q wc=Cð Þ1q wc=Cð Þ2

¼ wc=Cð Þ2wc=Cð Þ1

� �1:27ð3Þ

where q(wc/C)1 is the strength to be estimated at clay-water/cementratio of (wc/C)1 and q(wc/C)2 is the strength value at clay-water/cementratio of (wc/C)2. From the above equation, it is possible to assess thestrength at any other clay-water/cement ratio (clay-water content,cement content and fly ash content).

3.3. Strength development with curing time

The typical strength development with curing time for the FA andBA blended cement admixed saline clay with 1.3% salt is shown inFigs. 11 and 12, respectively. The strength development with curing

wc/C Laboratory strength, qul(kPa)

Predicted strength, qup(kPa)

3.87 1057 10333.74 1031 10771.29 3908 40801.25 4000 42524.72 580 8054.57 638 8391.57 3266 31791.52 3063 33135.57 357 6545.39 425 6821.86 2308 25821.80 2440 26913.87 1600 Reference3.74 1635 17271.29 5641 65401.25 5769 68174.72 1120 12914.57 1192 13451.57 4874 50691.52 4898 53125.57 676 10485.39 803 10931.86 3412 41401.80 3916 43153.87 2120 21823.74 2181 22741.29 8029 86131.25 8204 89784.72 1340 17004.57 1631 17721.57 6652 67111.52 6641 69955.57 1336 13815.39 1357 14391.86 4933 54511.80 5407 5682

t in blended cement admixed saline clay, Appl. Clay Sci. (2011),

8 S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

time (days) in natural logarithmic scale can be expressed as linearvariation. It is evident that at a particular wc/C value, the strength de-velopment with time is controlled by the A-value only because the B-and k-values for all practical purposes can be regarded as constant.Even though the A-values are different for different salt contents,the rate of strength development with time is identical because thecementitious products influence the rate predominantly. The general-ized strength development for the FA and BA blended cementadmixed saline clay is presented in the form (vide Fig. 13):

qDq28

¼ 0:099þ 0:281 lnD ð4Þ

where qD is the strength after D days of curing, q28 is the 28 day-strength and D is the curing time (days). This generalized strengthdevelopment is comparable to that proposed for cement admixedclays (Horpibulsuk et al., 2003, 2011a,b), and accounts for the effectsof clay water content, cement content and ash content.

4. Interrelationship among strength, clay-water/cement ratio andcuring time

Because the clay-water/cement ratio hypothesis is valid for ce-ment admixed saline clay, the generalized strength equation can bedeveloped in the same way as for the original one (Horpibulsuk etal., 2011a). The generalized interrelationship among strength, curingtime and wc/C for assessing strength development of the blended ce-ment admixed saline clay in which the wc/C ranges from 1 to 6 isexpressed by combination of Eqs. (3) and (4).

q wc=Cð ÞDq wc=Cð Þ28

( )¼ wc=Cð Þ28

wc=Cð ÞD

� �1:270:099þ 0:281 lnDð Þ ð5Þ

where q(wc/C)D is the strength of the blended cement admixed salineclay to be estimated at clay-water/cement ratio of (wc/C)D after Ddays of curing and q(wc/C)28 is the strength of the blended cementadmixed saline clay at clay-water/cement ratio of (wc/C) after28 days of curing and

C ¼ Ci 1þ 0:75að Þ: ð6Þ

The k-value of 0.75 is considered for a≤25%. Using Eq. (5), the as-sessment of the strength development in the BA blended cementadmixed saline clay with 1.3% salt is presented in Table 4. The 28-day strength of the sample made up at wc=43%, C=20% anda=10% was taken as a reference. The error from the prediction is ac-ceptable for the common engineering practice with the mean abso-lute error being less than 8.2%. The expression proposed is simplewithin the framework of Abrams' law and requires the strengthdata of a trial mix. Based on the previous researches (Horpibulsuket al., 2011a,b) and the present work, it is concluded that Eq. (5) isvalid for low-swelling clays with a wide range of salt content. TheB- and k-values can be taken as 1.27 and 0.75, respectively. The equa-tion requires only a laboratory strength value of the cement admixedclay (with or without fly ash and biomass ash) for a particular curingtime and mixing condition (water content and cement content).

An addition of 25% ash is recommended for an economic mix de-sign. By substituting a=25% in Eq. (6), the economic input of cementfor the blended cement admixed clay in terms of the total equivalentcement, Ci is 0.842C. This recommended mix design can reduce theuse of cement up to 15.8% 1−0:842

1

� �� 100%�

.

5. Conclusions

The presence of salt content affects the index properties of salineclay. The liquid limit decreases as salt content increases due to the

Please cite this article as: Horpibulsuk, S., et al., Strength developmedoi:10.1016/j.clay.2011.10.003

compression of diffuse double layer of kaolinite particles. For a partic-ular water content, the generalized stress state e/eL increases withincreasing the salt content. This reduces the effective stress (inter-particle attraction) and shear strength of the saline clay. The clay-water/cement ratio hypothesis is applicable for analyzing and asses-sing the strength development. For the cement admixed saline claysamples with different salt contents but the same clay-water/cementratio, the samples with lower salt content exhibit higher strength dueto the higher effective stress (cementation bond strength and inter-particle attraction of the saline clay). Due to the dispersing effect, flyash and biomass ash can be used to reduce the amount of cementadded. The 25% ash content is the most effective amount and can re-duce addition of cement up to 15.8%. This ash content yields highestcementitious products and strength. This dispersion increases the re-active surface for hydration, which results in the increase in cementi-tious products and strength. Based on the present work, a generalizedstrength equation for blended cement admixed non- to low swellingclays is suggested. The B- and k-values can be taken as 1.27 and 0.75for all curing times. The equation facilitates the determination ofproper quantity of blended cement to be admixed for different in-situ and field mixing water contents.

Acknowledgments

This work was supported by the Higher Education ResearchPromotion and National Research University Project of Thailand,Office of Higher Education Commission. The financial support and fa-cilities that were provided by the Suranaree University of Technologywere appreciated.

References

Abrams D.A., 1918. Design of Concrete Mixtures. In: Structural Materials Research Lab-oratory, Lewis Institute, Chicago, Bulletin 1, 20pp.

Burland, J.B., 1990. On the compressibility and shear strength of natural clays. Geotech-nique 40 (3), 329–373.

Chindaprasirt, P., Homwuttiwong, S., Sirivivatnanon, V., 2004. Influence of fly ash fine-ness on strength, drying shrinkage and sulfate resistance of blended cement mor-tar. Cement and Concrete Research 34, 1087–1092.

Chindaprasirt, P., Jaturapitakkul, C., Sinsiri, T., 2006. Effect of fly ash fineness on micro-structure development of blended cement plates. Construction and BuildingMaterials 21, 1534–1541.

El-Jazairi, B., Illston, J.M., 1977. A simultaneous semi-isothermal method of thermogra-vimetry and derivative thermogravimetry, and its application to cement plates.Cement and Concrete Research 7, 247–258.

El-Jazairi, B., Illston, J.M., 1980. The hydration of cement plate using the semi-isothermal method of thermogravimetry. Cement and Concrete Research 10,361–366.

Horpibulsuk, S., Miura, N., 2001. A new approach for studying behavior of cement sta-bilized clays. Proceedings of 15th International Conference on Soil Mechanics andGeotechnical Engineering (ISSMGE). Istanbul, Turkey, vol. 3, pp. 1759–1762.

Horpibulsuk, S., Raksachon, Y., 2010. Fly ash as a dispersing material in cement stabili-zation. Proceedings of Geo-Shanghai 2010. China, pp. 137–142.

Horpibulsuk, S., Miura, N., Nagaraj, T.S., 2003. Assessment of strength development incement-admixed highwater content clays with Abrams' law as a basis. Geotechnique53 (4), 439–444.

Horpibulsuk, S., Bergado, D.T., Lorenzo, G.A., 2004a. Compressibility of cement admixedclays at high water content. Geotechnique 54 (2), 151–154.

Horpibulsuk, S., Miura, N., Bergado, D.T., 2004b. Undrained shear behavior of cementadmixed clay at high water content. Journal of Geotechnical and Geoenvironmen-tal Engineering, ASCE 130 (10), 1096–1105.

Horpibulsuk, S., Miura, N., Nagaraj, T.S., 2005. Clay-water/cement ratio identity ofcement admixed soft clay. Journal of Geotechnical and GeoenvironmentalEngineering, ASCE 131 (2), 187–192.

Horpibulsuk, S., Shibuya, S., Fuenkajorn, K., Katkan, W., 2007. Assessment of engineer-ing properties of Bangkok clay. Canadian Geotechnical Journal 44 (2), 173–187.

Horpibulsuk, S., Rachan, R., Raksachon, Y., 2009. Role of fly ash on strength and micro-structure development in blended cement stabilized silty clay. Soils and Foundations49 (1), 85–98.

Horpibulsuk, S., Liu, M.D., Liyanapathirana, D.S., Suebsuk, J., 2010a. Behavior of cemen-ted clay simulated via the theoretical framework of the Structured Cam Claymodel. Computers and Geotechnics 37, 1–9.

Horpibulsuk, S., Rachan, R., Raksachon, Y., Suddepong, A., Chinkulkijniwat, A., 2010b.Analysis of strength development in cement-stabilized silty clay based on micro-structural considerations. Construction and Building Materials 24, 2011–2021.

nt in blended cement admixed saline clay, Appl. Clay Sci. (2011),

9S. Horpibulsuk et al. / Applied Clay Science xxx (2011) xxx–xxx

Horpibulsuk, S., Rachan, R., Suddeepong, A., 2011a. Assessment of strength develop-ment in blended cement admixed Bangkok clay. Construction and BuildingMaterials 25 (4), 1521–1531.

Horpibulsuk, S., Rachan, R., Suddeepong, A., Chinkulkijniwat, A., 2011b. Strength devel-opment in cement admixed Bangkok clay: laboratory and field investigations. Soilsand Foundations 51 (2), 239–251.

Horpibulsuk, S., Yangsukaseam, N., Chinkulkijniwat, A., Du, Y.J., 2011c. Physical, com-pressibility and permeability characteristics of Bangkok clay compared withthose of kaolinitic and montmorillonitic soils. Applied Clay Science 52, 150–159.

Kamon, M., Bergado, D.T., 1992. Ground improvement techniques. Proceedings of9th Asian Regional Conference on Soil Mechanics and Foundation Engineering,pp. 526–546.

Kawasaki, T., Niina, A., Saitoh, S., Suzuki, Y., Honjo, Y., 1981. Deep mixing method usingcement hardening agent. Proceedings of 10th International Conference on SoilMechanics and Foundation Engineering. Stockholm, pp. 721–724.

Midgley, H.G., 1979. The determination of calcium hydroxide in set Portland cements.Cement and Concrete Research 9, 77–82.

Miura, N., Horpibulsuk, S., Nagaraj, T.S., 2001. Engineering behavior of cement stabi-lized clay at high water content. Soils and Foundations 41 (5), 33–45.

Nagaraj, T.S., Pandian, N.S., Narasimha Raju, P.S.R., 1998. Compressibility behavior ofsoft cemented soils. Geotechnique 48 (2), 281–287.

Papadakis, V.G., Tsimas, S., 2002. Supplementary cementing materials in concrete partI: efficiency and design. Cement and Concrete Research 32, 1525–1532.

Please cite this article as: Horpibulsuk, S., et al., Strength developmendoi:10.1016/j.clay.2011.10.003

Prakash, K., Sridharan, A., 2004. Free swell ratio and clay mineralogy of fine-grainedsoils. Geotechnical Testing Journal, ASTM 27 (2), 220–225.

Sinsiri, T., Jaturapitakkul, C., Chindaprasirt, P., 2006. Influence of fly ash fineness on cal-cium hydroxide in blended cement paste. Proceedings of Technology and Innova-tion for Sustainable Development Conference (TISD2006). Khon Kaen, Thailand.

Suebsuk, J., Horpibulsuk, S., Liu, M.D., 2010. Modified Structured Cam Clay: a constitu-tive model for destructured, naturally structured and artificially structured clays.Computers and Geotechnics 37, 956–968.

Suebsuk, J., Horpibulsuk, S., Liu, M.D., 2011. A critical state model for overconsolidatedstructured clays. Computers and Geotechnics 38, 648–658.

Terashi, M., Tanaka, H., Okumura, T., 1979. Engineering properties of lime treatedmarine soils and DMM. Proceedings of 6th Asian Regional Conference on SoilMechanics and Foundation Engineering, pp. 191–194.

US Salinity Laboratory Staff, 1995. Diagnosis and improvement of saline and alkalinesoils. USDA Handbook No.60, Washington DC.

Wang, K.S., Lin, K.L., Lee, T.Y., Tzeng, B.Y., 2004. The hydration characteristics when C2Sis present in MSWI fly ash slag. Cement and Concrete Research 26, 323–330.

Yavaniyama, A., Lertsirivorakul, R., Sriboonruang, V., 2005. Effect of land use manage-ment on groundwater and soil salinization in northeast Thailand. Workshop onManagement of Tropical Sandy Soils for Sustainable Agriculture, Khon Kaen,Thailand.

t in blended cement admixed saline clay, Appl. Clay Sci. (2011),