Construction and Building Materials 52 (2014) 4251

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Inuence of bacterial treated cement kiln dust on the properties of concrete

Kunal a,, Rafat Siddique b, Anita Rajor a
a b

Department of Biotechnology and Environmental Sciences, Thapar University, Patiala, Punjab, India Department of Civil Engineering, Thapar University, Patiala, Punjab, India

h i g h l i g h t s
 Paper presents study on properties of concrete containing bacterial treated CKD.  Bacterial treatment reduces the alkalinity of CKD up to 67% in leachate.  Increase in compressive strength and decrease in water absorption and porosity is observed.  XRD conrms formation of non-expansive ettringite and calcium silicate in CKD-concrete.

a r t i c l e

i n f o

a b s t r a c t
During cement manufacturing, cement kiln dust (CKD) is generated which represents signicant environment concern related to its emission, disposal and reuse due to high alkalinity. This study presents the effect of bacterial (Bacillus halodurans strain KG1) treated cement kiln dust on the compressive strength, water absorption and porosity (at 7, 28 and 91 days) of concrete after reducing the alkalinity. Concrete specimens were prepared with 0%, 5%, 10% and 15% untreated and treated CKD replacing cement. Test results indicated that 7.15% and 26.6% increase in strength of concrete was achieved at 28 and 91 days, respectively, with the addition of bacterial treated 10% CKD whereas reduction in water absorption (20%) and porosity (12.35%) was observed at 91 days. X-ray diffraction (XRD) and scanning electron microscopy (SEM) results suggested that in bacterial treated 10% CKD concrete increased calcium silicate hydrate and formation of non-expansive ettringite in pores dense the concrete structure resulted in increased compressive strength. 2013 Elsevier Ltd. All rights reserved.

Article history: Received 5 July 2013 Received in revised form 1 November 2013 Accepted 12 November 2013

Keywords: Alkalinity Bacteria Cement kiln dust Compressive strength Ettringite

1. Introduction Rapidly growing population generates the need for industrialization and urbanization which in turn increases the demand of building and construction material for infrastructure development. Concrete is one of the most durable and widely used construction material in the world with annual consumption estimated between 21 and 31 billion tones in 2006 [1]. Concrete is made from coarse aggregates (gravel or crushed stone), ne aggregates (sand), water, cement and admixtures. Cement plays an important role in concrete due to its binding properties. Increasing urbanization continuously driving the cement industry to keep growing. According to Oss [2], 3700 million tones of cement (3400 million tones of clinker) was generated in 2012 worldwide, whereas in India 250 million tones of clinker was generated. There is 2.78% and 11.78% increase in cement production was observed in year 2012
Corresponding author. Tel.: +91 9988385367.
E-mail addresses: kunal_pau@yahoo.co.in ( Kunal), siddique_66@yahoo.com (R. Siddique), anitarajor@yahoo.com (A. Rajor). 0950-0618/$ - see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.conbuildmat.2013.11.034

compared to year 2011 and 2010, respectively. This increase in cement production results in towering collection of cement kiln dust from cement plants. Cement kiln dust (CKD) is a ne powdery material generated in large quantities from air pollution control devices (e.g. cyclone, bag house, or electrostatic precipitator) during the production of cement clinker. The generation of CKD has been estimated to be 1520% of clinker production [3], which put world wide CKD generation at an estimated 510680 million tones for the year 2012 and Indian production at 37.550 million tones. Cement kiln dust is a very heterogenous mix both by chemistry and particulate size. The chemical composition of CKD depends upon the raw materials, fuels, kiln type, overall equipment layout, and type of cement being used. The concentration of free lime, sulfates and alkalies in CKD mainly dependent upon the size of particles collected near to the kiln. Coarser particles of CKD contain high content of free lime while the ne particles usually exhibit higher concentration of sulfates and alkalies and lower lime content [4]. Cement kiln dust is generated as a measure to control

Kunal et al. / Construction and Building Materials 52 (2014) 4251

43

product quality (low alkali clinker from high alkali raw materials) and to ensure uninterrupted operation of the plant. The major factor preventing return of more dust to the kilns is the high concentration of alkalis in the dust that would cause the alkali content of the clinker to exceed the allowable value. Additionally, the high concentrations of volatiles develop deposits on the walls of the kiln which can result in frequent shut down of the plant. Hence cement plants generate CKD as a means of removing volatile alkalis, chlorides and sulfates from the kiln system. The generation of large quantities of CKD is responsible for a signicant nancial loss to the cement industry in terms of the value of raw materials, processing, energy usage and, above all, disposal and storage. According to US Environmental Protection Agency [5], approximately 15 million tons of CKD is produced annually in United States and major part of it is sent to landlls. Disposal of waste CKD is not only associated with the problem of land use but also with contamination of ground water from leachate. It is estimated that over 200,000 tons a year of landll space could be saved in UK if the surplus CKD could be recycled into the clinker-making process or if alternative uses could be found [6]. Due to the cost associated with the production and disposal of CKD and the strong and strict environmental regulations on the proper management of CKD, the cement industries now showing a keen interest in nding the proper utilization of CKD in a manner that protects human health and the environment. Cement kiln dust has the potential for reuse in many different ways, but the cement making process is the best way to reuse this by-product material and approximately 6067 percent (88.4 million tons) of the total CKD generated in United States is used in this way [7]. The most common benecial uses of CKD are cement replacement, soil stabilization, waste treatment, asphalt pavement and other uses. CKDs cement like properties also makes it a potential replacement for Portland cement in utilization in concrete, owable slurry, etc. Several researchers [821] have reported on some aspects of the utilization of CKD in cement paste, mortar/ concrete. Generally, the cement kiln dust is alkaline in nature (pH 12) and is considered to be caustic. If used in concrete, the alkaline nature needs to be monitored to avoid expansive reaction between alkalis and certain aggregates, which leads to cracking and causes deterioration [22]. Bhatty [1316,23] found that CKD blended cement had reduced workability, setting times and strength. The loss of strength was attributed to the alkalies in the dust. The American Society for Testing and Materials (ASTM) species a limit of 0.6 percent alkali in Portland cement [24,25]. If the amount of alkali in the kiln dust is high, its reusing in the kiln causes the formation of potassium and sodium containing compound and free CaO, leading ultimately to a reduction in the strength of cement [26] and causes crack, deformities and reduce the quality of cement and concrete [27,28]. Juenger and Jennings [22] examined that high alkali content also affects the hydration and microstructure of cement paste and therefore, inuences the concrete properties. Cement kiln dust after reducing the alkalinity can be used in cement-concrete system and may have positive effect on cementconcrete properties. Mohamed and El-Gamal [29] and Gebhardt [30] treated the alkaline CKD with carbon dioxide gas to remove the alkalinity, but these methods are highly expensive, laborious and above all CKD is not reutilized rather landlled. The better alternative to the use of chemical process (involves use of carbon dioxide gas) is the biological treatment of alkaline wastes using bacterial system which grow well at high pH. Studies suggested that alkaliphilic bacteria could degrade pollutants under highly alkaline conditions and had the signicant advantage of not being easily contaminated by neutral microorganisms [3133]. However, reports on the application of alkaliphilic bacteria in treatment of solid alkaline waste are very rare. Thus, in the present study we

have isolated an alkalitolerant bacterial strain (Bacillus sp. KG1) and utilized in reducing the alkalinity of the CKD. The treated CKD then utilized in concrete as partial replacement to cement in different percentages (015%) and investigated the effect on the mechanical properties of concrete.
2. Materials and methods 2.1. Material used 2.1.1. Cement Cement of Indian Standards (IS) mark 43 grade (IS mark 43 grade means that the 28-day compressive strength of cement is 43 MPa) UltraTech brand was used for all mixes. Testing of cement was conducted as per IS: 8112-1989 [34]. The test results conducted on cement are reported in Table 1.

2.1.2. Fine and coarse aggregate Natural sand with 4.75 mm maximum size was used as ne aggregate. The sand was rst sieved through 4.75 mm sieve to remove any particle greater than 4.75 mm and removed the dust. Locally available coarse aggregates having the size of 12.5 mm were used in this work. Testing of ne and coarse aggregates was done as per IS: 383-1970 [35]. Properties of the coarse and ne aggregates used are shown in Table 2.

2.1.3. Cement kiln dust Cement kiln dust (CKD) is ne powdery material of grey-black in color and relatively uniform in size. Table 3 represents the physical properties of CKD compared with literature published. As CKD is derived from same raw materials as cement clinker, despite the fact that it has similar chemical composition to that of ordinary Portland cement, signicant variation in physical and chemical composition of CKDs obtained from different cement plants has been observed. Table 4 shows the chemical composition of CKD and cement used in the study and typical composition of CKD investigated by other researchers.

2.2. Isolation of bacteria Alkaliphilic and/or alkalitolerant bacteria (that tolerate high pH) was isolated from rhizospheric (near to root of plant) soil. The soil samples were suspended in sterile saline solution (0.85% NaCl), diluted properly and plated on enrichment medium containing glucose (10 g/l), peptone (10 g/l), yeast extract (5 g/l), KH2PO4 (1 g/l), agar (15 g/l) and pH was adjusted to 10.5 with 1 N.

Table 1 Properties of cement. Characteristics Fineness (%) Standard consistency Initial setting time (min) Final setting time (min) Specic gravity Values obtained 1% 34% 120 240 3.03 IS: 8112-1989 10 max 30 min 600 max ASTM C150 375 max

Table 2 Properties of coarse and ne aggregates. Characteristics Fineness (%) Specic gravity Water absorption (%) Size (mm) Moisture content (%) Coarse aggregate 6.74 2.59 0.80 12.5 max Nil Fine aggregate 2.72 2.62 1.02 4.75 max 0.16

Table 3 Physical properties of CKD. Physical properties Specic gravity Fineness modulus pH Values 2.39 2.25 >12 Taha et al. [10] 2.4 4.824 Collins and Emery [4] 2.62.8

44 Table 4 Chemical properties of cement and CKD. Constituent (%) Cement

Kunal et al. / Construction and Building Materials 52 (2014) 4251

Cement kiln dust (CKD) Present study Control Bacterial treated 48.19 10.97 2.19 0.58 0.82 0.59 2.41 0.80 0.64 Reviewed literature Maslehuddin et al. [6] 49.3 17.1 4.24 1.14 3.56 2.18 3.84 2.89 15.8 Taha et al. [10] 63.76 15.84 3.57 1.93 1.65 2.99 0.33 2.76 5.38 Udoeyo and Hyee [55] 52.72 2.16 1.09 0.68 0.05 0.11 0.54 42.39

CaO SiO2 Al2O3 MgO SO3 K2O Na2O Fe2O3 CuO ZnO LOI

65.57 23.61 2.16 0.72 1.32 1.03 0.32 2.41 1.39 1.50

55.78 13.17 2.38 0.69 1.13 1.12 2.62 0.89 0.66

Fig. 1. Compressive strength of untreated concrete containing cement kiln dust.

Fig. 2. Compressive strength of concrete containing bacterial treated cement kiln dust. water to cementitious material ratio was kept constant (0.5) to investigate the effect of replacing cement with CKD. The control without CKD was designed as per Indian Standards specications IS: 10262-1982 [37]. 2.5. Preparation and casting of test specimens Concrete cubes of size 150 mm of M20 grade were prepared for compressive strength, water absorption and porosity. The casting of specimens was in accordance with Indian Standard IS: 516-1959 [38]. After casting, the specimens were allowed to remain in iron molds for rst 24 h at room temperature (27 2 C). After that these were demolded and placed in the water tank at room temperature for curing. The specimens were tested after 7, 28 and 91 days of curing period. 2.6. Study of concrete properties Concrete can be made to have high compressive strength. Concrete properties were studied in triplicate. The compressive strength was determined as per IS: 516-1959 [38] specications. Water absorption and porosity of the cubes were determined by using ASTM C 642-97 [39] method at the age of 7, 28 and 91 days. Concrete samples from each mix (after 91 days of curing) was taken from inner core of the matrix (crushed into ne powder by pestle-mortar) and analyzed in powder X-ray diffraction (XRD; PANalytical XPro). The XRD spectrum was taken from 2 h = 5h to 2 h = 60h. The peaks in the new positions of the spectrum were marked, compared and identied from the Joint Committee on Powder Diffraction Standards (JCPDS) data le and from the published literature. Scanning electron microscopic (SEM; JEOL JSM 6510 LV, USA) analysis was performed by mounting small broken concrete specimens (with and without bacterial treated CKD) on brass stubs using carbon tape. The samples were coated with gold and then analyzed at 20 kV.

NaOH. The agar plates were incubated at 37 C for 48 h. Isolated colonies were picked and re-streaked on same agar medium till pure colonies were obtained. The selected colonies were then screened for their tolerance to pH 11 and 12 and performed on minimal (M9) medium containing sucrose (10 g/l), KH2PO4 (2.5 g/l), K2HPO4 (2.5 g/l), (NH4)2HPO4 (1 g/l), MgSO47H2O (2 g/l), FeSO47H2O (0.01 g/l), MnSO44H2O (0.007 g/l) and agar (15 g/l). pH was adjusted to 11 and 12 by KCl NaOH buffer. The cultures were maintained on M9 medium and stored at 4 C for further experimentation. 2.3. Bacterial treatment of CKD On the basis of pH reduction of the alkaline medium, isolate KG1 showed the promising results and utilized for the treatment of CKD. The CKD was mixed with 0.8 OD (optical density; measured by spectrophotometer at kmax 600 nm) value of the selected bacterial strain KG1 (OD 1.0  108 cells) in the ratio CKD to culture (4:1). For bacterial treatment the CKD sample was poured into plastic tubs and bacterial culture KG1 in required proportion was added and mixed manually in a way so that the culture was thoroughly distributed. The treatment mixture was incubated at 35 2 C for 20 days and moisture was maintained by spraying water for the growth of bacterial strain. After 5 days of incubation, sucrose solution (10%) was added only once during the treatment of 20 days to provide carbon source for the bacterial strain KG1. After the completion of the incubation period of 20 days, samples were collected randomly from different places, mixed with water (1:10) in conical asks with shaking (@ 130 rpm for 1 h) to generate leachate and analyzed for alkalinity and chloride along with control treatment [36]. To conrm the decrease in alkalinity, bacterial treated CKD sample was air dried and analyzed with energy dispersive X-ray spectrometry (EDX, JEOL JSM-6510 LV, USA) for change in chemical composition of CKD. 2.4. Concrete mix design The bacterial treated CKD (moisture content 0.02%) was used for the concrete mixtures along with series of control concrete mixtures containing untreated CKD. Seven concrete mix proportions were made. First was the control mix (without CKD) and the other six mixtures contained CKD (three mixes each contained untreated and bacterial treated CKD). Cement was replaced with CKD by weight in proportion of 5%, 10% and 15%. The mix proportion of concrete used was water (0.5): cement (1.0): ne aggregate (1.45): coarse aggregate (2.98). The

3. Results and discussion 3.1. Physical and chemical properties of CKD Specic gravity of CKD was found to be observed as 2.39 whereas typical specic gravity varies between 2.4 and 2.8

Kunal et al. / Construction and Building Materials 52 (2014) 4251

45

Fig. 3. Percent water absorption of (a) untreated and (b) bacterial treated concrete containing cement kiln dust.

Fig. 4. Percent porosity of (a) untreated and (b) bacterial treated concrete containing cement kiln dust.

Fig. 5. X-ray diffraction shown by concrete without CKD.

[4,10]. Fineness modulus and pH was found to be observed as 2.25 and 12, respectively. CKDs on the average are typically characterized by higher alkali and sulfur content which is one of the main reason from removing the dust from kilns. As CKD is derived from the same raw materials as ordinary Portland cement, despite that it has signicant variation in chemical composition obtained from different cement

plants. The chemical composition of typical CKD, CKD and cement used in this study is shown in Table 4. Compounds of lime, silica, alumina and iron constitute the major composition of CKD followed by alkali (K2O) and sulfur (SO3). Analysis of CKD showed the absence of Na and Cl with small amounts of Cu, Mg and Zn. Lime was found to be observed as maximum value of 55.78% whereas silica was found to be observed as 13.17%. Alkalinity of

46

Kunal et al. / Construction and Building Materials 52 (2014) 4251

Fig. 6. X-ray diffraction shown by concrete containing 5% CKD; (a) untreated CKD, and (b) bacterial treated CKD.

the CKD was mainly due to the presence of K (K2O; 1.12%) and S (SO3; 1.13%) components. 3.2. Bacterial treatment of CKD Bacterial treated CKD sample after 20 days were analyzed for alkalinity and chloride as per method of APHA [36]. The alkalinity and chloride of control CKD leachate was 1467 mg/l and 460 mg/l, respectively, whereas after treated with bacterial strain KG1, the alkalinity and chloride content of CKD reduced signicantly (p < 0.05; t-test at 95% condence limits) to 480 mg/l (32.72%) and 73.33 mg/l (15.94%), respectively. All the treatments were performed in three replications and the results are the average of three readings. The EDX analysis of bacterial treated CKD also revealed the signicant (p < 0.0001; two way ANOVA) reduction of K2O and SO3 content by 47.32% and 24.43%, respectively, compared to control CKD (Table 4). The fact behind the reduction of alkalinity is due to the production of organic acid (acetic and formic acid) by bacterial enzymes in fermentation and respiration process of bacterial metabolism which eventually reduces the alkalinity [33]. 3.3. Effect of bacterial treated CKD on concrete properties During fresh concrete mixing (before casting), the samples were collected from different places and the leachate generated was analyzed for alkalinity of the fresh concrete mix leachate. The

alkalinity of the control concrete (0% CKD) was 1666.67 mg/l which keep on decreasing as the cement was replaced with untreated CKD in different percentages of 5 (1553.33 mg/l), 10 (1526.67 mg/l) and 15 (1440 mg/l). This is supported by the fact that the alkalinity of cement is 1893.33 mg/l which is higher than CKD alone and 0% CKD control concrete. When cement is replaced with different percentages of untreated CKD, the alkalinity of concrete (5%, 10% and 15%) goes on decreasing instead of increasing. After addition of bacterial treated CKD in concrete, 64%, 62.8% and 60% alkalinity was observed in 5%, 10% and 15% treated CKD concrete compared to control concrete mix. After casting at the age of 7 days, the control mix (0% CKD) showed compressive strength of 23.23 N/mm2, whereas 5%, 10% and 15% CKD (untreated) concrete showed compressive strength of 23.78, 24.31 and 23.03 N/mm2, respectively (Fig. 1). The compressive strength of the CKD control (untreated) concrete increased with increase in curing period. At the age of 28 days, the compressive strength of the control CKD concrete (0% CKD) was 34.82 N/mm2 whereas of 5%, 10% and 15% control CKD concrete was 35.78, 36.29 and 34.53 N/mm2, respectively. Similarly at 91 days of curing period, there was 40.44, 41.89, 44.12 and 40.03 N/mm2 of compressive strength in 0%, 5%, 10% and 15% CKD control concrete (Fig. 1). Maslehuddin et al. [8] reported decrease in compressive strength (>5%) of concrete mixes (10% and 15% replacement) at all ages (3, 7, 14, 28, 56 and 91 days). The authors concluded that up to 5% CKD could be used without compromising the compressive strength of concrete. El-Aleem et al. [40] concluded that up

Kunal et al. / Construction and Building Materials 52 (2014) 4251

47

Fig. 7. X-ray diffraction shown by concrete containing 10% CKD; (a) untreated CKD, and (b) bacterial treated CKD.

to 6% CKD replacement there is no signicant reduction in compressive strength of hardened mortar whereas above this percentage the compressive strength decreased sharply. This reduction in strength may be due to decrease in cement content, increase in free lime in cement dust, increased porosity or the formation of chloro and sulpho-aluminate phases which leads to softening and expansion of the hydration products. In this study, up to 10% CKD replacement (control CKD concrete) showed increased strength whereas in 15% CKD control concrete compressive strength was decreased. The increase in compressive strength up to 10% CKD replacement levels in control concrete mix may be due to an appropriate alkalinity that increases the dissolution of silicate and formation of calcium silicate hydrate which is responsible for increased compressive strength [41]. The decreased compressive strength in 15% CKD control concrete may be due to increased alkalinity, porosity and lower cement content at 91 days of curing. These results were in concomitant with the ndings of Bhatty [1316,23] and Rehsi and Garg [26]. Studies reported that high alkalinity in dust causes cracks, reduces the quality of cement and concrete, affects the hydration and microstructure of cement paste and ultimately reduced the strength of concrete [22,27,28]. Fig. 2 shows the compressive strength of bacterial treated CKD containing concrete cubes (5%, 10% and 15%) at the age of 7, 28 and 91 days. Similar trend of increased compressive strength was also found in bacterial treated CKD concrete at all curing ages. The early strength (compressive strength at 7 days of curing) of bacterial

treated CKD concrete showed decrease in strength values compared to control CKD concrete cubes. This may be due to decrease in the alkalinity of concrete mix containing bacterial treated CKD. The CKD was treated with bacterial strain KG1 to reduce the alkalinity which was then used as a replacement to cement in different percentages (5%, 10% and 15%). The heat of hydration is due to the alkalinity present in the cement-concrete mix which results in the pozzolanic reaction and developed the early strength. The reduced alkalinity slows the pozzolanic reaction of CKD-cement mixture and lowers the early strength of concrete compared to control concrete mixes. According to Pu [42], the strength of the cement-concrete mix containing active mineral additives such as CKD can be considered as composed of two parts: the rst part of strength is contributed by the hydrates that are formed by the hydration of clinker in cement, and the second part is contributed by the additional hydrates obtained from the secondary reaction between active silica and alumina oxides in mineral additives with free calcium hydroxides obtained from the hydration of clinker. This second part is responsible for the strength developed in bacterial treated CKD concrete in later ages (28 and 91 days). In 10% bacterial treated CKD concrete 7.15% and 2.81% increase was observed in compressive strength compared to 0% and 10% CKD control concrete, respectively at 28 days of curing, whereas at the age of 91 days, this increase was 26.6% and 16.09%, respectively. Above 10% CKD addition (bacterial treated) the strength decreased and this may

48

Kunal et al. / Construction and Building Materials 52 (2014) 4251

Fig. 8. X-ray diffraction shown by concrete containing 15% CKD; (a) untreated CKD, and (b) bacterial treated CKD.

Fig. 9. SEM image of control concrete without CKD.

be attributed to decrease in cement content and hydration reaction. Several studies have been reported the improvement of compressive strength of the cement mortar by inclusion of microorganisms [4345]. Researchers proposed a new phenomenon known as biocalcication or microbially induced calcite precipitation to improve the overall strength and performance of cement mortar by repairing the cracks and pores of the structure. In this study,

bacterium was used to reduce the alkalinity of CKD and then utilized in partial replacement to cement in concrete. The late strength developed after 28 days of curing and is probably due to reduced hydration reaction, increased CS/CSH gel formation (evidenced from XRD results discussed later in the text) and deposition of bacterial cells or spores within the pores of CKD-cement-sand matrix which plugs the pores with in the concrete. Konsta-Gdoutos and Shah [19] and Salem and Ragai [46] reported that presence of alkali concentration plays an important role in initial hydration of cement pastes. Cement paste containing alkaline CKD showed increase hydration resulted in early strength. Water absorption and porosity are directly related to compressive strength of concrete. Increased water absorption and pore size decreased the compressive strength of concrete. Results of water absorption and porosity shows decrease of 20% and 12.35%, respectively, in 10% bacterial treated CKD concrete compared to untreated CKD control concrete at 91 days of curing (Figs. 3 and 4). Water absorption and porosity decreases with increase in CKD concentration but above 10% the water absorption and porosity increases in both control and bacterial treated CKD concrete (at the ages of 7, 28 and 91 days). This is due to decrease in cement content and increase in CKD concentration which decreases the binding of CKD-cement-aggregate in concrete and develops pores. X-ray diffraction (XRD) analysis of concrete samples with or without bacterial treated cement kiln dust shows peaks of quartz (Q), calcium silicate hydrate (CSH), calcite (C), larnite (L) and ettringite (E) phases on comparing the values of 2h/d/I/I of the

Kunal et al. / Construction and Building Materials 52 (2014) 4251

49

Fig. 10. SEM image shows 5% cement kiln dust containing concrete; (a) untreated, and (b) bacterial treated.

Fig. 11. SEM image shows 10% cement kiln dust containing concrete; (a) untreated, and (b) bacterial treated.

Fig. 12. SEM image shows 15% cement kiln dust containing concrete; (a) untreated, and (b) bacterial treated.

peaks by JCPDS data le (Figs. 58). Peaks of different phases in treatments shows the intensity corresponding to the strength of concrete. Alite (C3S) is the major mineral component (>50%) found in cement and upon hydration forms calcium silicate hydrate or calcium silicate which hardens the cement slurry and is responsible for initial (13 days) and nal strengths [47]. The second major component found in cement is C2S or Belite or Larnite (Ca2SiO4). Larnite reacts with water to form calcium silicate hydrate or calcium silicate and portlandite, and responsible for the development of late strength. Neville [48], Molnar et al. [49] and Jumate and Manea [47] studied that hydration and hydrolysis reaction of C3S and C2S mineral components produce calcium silicate hydrate (also known as Tobermorite) gels and later the solid phase develops crystals during curing period leading to strengthening of the cement-concrete mixes.

In 10% bacterial treated CKD concrete (Fig. 7b) the increased formation of CSH resulted in increased strength compared to 10% CKD control concrete (Fig. 7a). Ettringite formation in 10% bacterial treated CKD concrete, due to less alkali content, was nonexpansive and lled the pore structure in concrete resulted in dense structure and increased the compressive strength. Min and Mingshu [50] stated that the high concentration of hydroxyl ions (i.e. high pH values) due to higher alkali content of the solution results in the expansive type of ettringite. According to Heinz and Ludwig [51], several factors affect the formation of ettringite such as sulfate content, pH of the solution and availability of the calcium hydroxide. Higher alkali content increased the solubility of sulfate ions in solution which being absorbed by CSH resulted in formation of expansive type of ettringite. The XRD results (Figs. 6 and 7) shows increased intensity of CS (21, 26, 29 degree 2h) and

50

Kunal et al. / Construction and Building Materials 52 (2014) 4251

nonexpansive ettringite (7, 17 and 34 degree 2h) in bacterial treated CKD concrete (5% and 10%) responsible for the strength development in concrete where as in 15% bacterial treated CKD concrete, reduction in cement content reduced the required alkalinity (needed for hydration reaction) which in turn decreased the CSH and thus reduced the strength compared to control concrete. Figs. 912 show the SEM analysis of control (0% CKD) and, untreated and bacterial treated CKD (5%, 10% and 15%) containing concrete. In 0% CKD concrete the SEM image shows the formation of calcium silicate hydrate and the hydration reaction formed dense structure resulted in increased compressive strength. Similar type dense structures was also observed in 5% and 10% CKD concrete but are less porous than 0% control at 28 days of curing period. In 15% CKD control concrete (Fig. 12a) voids were shown resulted in highly porous structure and exhibited decrease in strength compared to other control treatments. Samoui et al. [52] observed more reticular and porous structure in high alkali cement paste compared to low alkali paste. In 10% bacterial treated CKD concrete less porous and highly dense structure was formed at 28 days of curing (Fig. 11b) due to reduced hydration and alkalinity, compactness of the materials in concrete and this could possibly explain the increase in strength compared to untreated CKD concrete. In 15% bacterial treated CKD concrete (Fig. 12b) highly porous nature and voids due to less cement content and CSH formation that do not bind actively the materials, thus resulted in reduced strength in concrete. At the age of 28 days, in 15% bacterial treated CKD concrete needle shaped ettringites were seen on the interface of the cement paste-aggregate (Fig. 12b) which generates localized pressure and causes expansion of the crystallized structure and reduces the strength of concrete whereas in 10% bacterial treated CKD concrete, at 28 days of curing ettingite formation was observed in voids (Fig. 11b) which increases the density, reduces porosity and strengthens the concrete. These results were in accordance with the ndings of Divet and Pavoine [53] and Famy et al. [54] indicating the expansive nature of ettringites on the outer surface of calcium silicate hydrate surface at the interfaces of cement paste-aggregate and non-expansive nature of ettringite in the voids of the concrete i.e. microporous zones in the cement paste, pores or bubbles. The expansive ettringite exerts pressure to the aggregates forming gaps at the interfaces resulting in increased porosity and reduced strength. These results suggest that bacterial treatment of CKD reduced its alkalinity and improved the strength of concrete (up to 10% CKD) in later ages due to increased calcium silicate hydrate gel formation and formation of non expansive ettringite.

4. XRD and SEM analysis reveals the increased formation of CSH gel and nonexpansive ettringite formation which supports the increased compressive strength in 10% bacterial treated CKD concrete. 5. Further investigation is necessary to identify the effect of decreased alkalinity on durability properties of bacterial treated CKD concrete.

Acknowledgements The authors wish to express their gratitude to SERB, Department of Science and Technology, Govt. of India for the support in this research work. Authors also acknowledge the support of Department of Biotechnology & Environmental Sciences and Department of Civil Engineering, Thapar University, Patiala (India) for infrastructure and making this research possible. References
[1] European Concrete Platform ASBL, Sustainable benets of concrete structures, <http://www.bibm.eu/Documenten/Book%20nal%20version.pdf>; 2009. [2] Oss HG. Minerals Information Cement. Retrieved January 2013, from United States Geological Survey, <http://minerals.usgs.gov/minerals/pubs/commodity/ cement/mcs-2013-cemen.pdf>; 2013. [3] USEPA. Cement Kiln Dust. Retrieved January 15, 2009, from USEPA Wastes. <http://www.epa.gov/epawaste/nonhaz/industrial/special/ckd/cement2.htm>; 1993. [4] Collins RJ, Emery JJ. Kiln dusty ash systems for highways bases and subbases. Federal Highway Administration, Report No. FHWA/RD-82/167, Washington DC; 1983. [5] EPA. US, Environmental Protection Agency. Report to congress on cement kiln dust, EPA-530-R-94-001; 1993. [6] Maslehuddin M, Al-Amoudi OSB, Shameem M, Ibrahim M, Rehman MK. Usage of cement kiln dust in concrete products research review and preliminary investigations. Constr Build Mater 2008;22:236975. [7] Adaska WS, Taubert DH. Benecial uses of cement kiln dust. In: IEEE/PCA 50th Cement Industry Technical Conference, Miami, Florida; 2008. p. 19. [8] Maslehuddin M, Al-Amoudi OSB, Rehman MK, Ali MR, Barry MS. Properties of cement kiln dust concrete. Constr Build Mater 2009;23:235761. [9] Al-Jabri K, Taha R, Al-Harthy AS, Al-Oraimi S, Al-Nuaimi A. Use of cement bypass dust in owable ll mixtures. Cem Concr Agg 2002;24:537. [10] Taha R, Al-Rawas A, Al-Jabri K, Al-Harthy AS, Hassan H, Al-Oraimi S. An overview of waste materials recycling in the Sultanate of Oman. Res Conserv Recycl 2004;41:293306. [11] Al-Harthy AS, Taha R, Al-Maamary F. Effect of cement kiln dust (CKD) on mortar and concrete mixtures. Constr Build Mater 2003;17:35360. [12] Batis G, Rakanta E, Sideri E, Chaniotakis E, Papageorgiou A. Advantages of simultaneous use of cement kiln dust and blast furnace slag. In: Proceedings of the International Congress Challenges of Concrete Construction, Dundee, UK; 2002. p. 6772. [13] Bhatty MSY. Use of Cement Kiln Dust in Blended Cements, vol. 15. World Cement Technology, London, UK; 1984a. p. 12634. [14] Bhatty MSY. Kiln Dust Cement Blends, Evaluated, Rock Products, vol. 88. Chicago, Illinois, USA; 1984b. p. 4765. [15] Bhatty MSY. Use of cement kiln dust in blended cements: alkali-aggregate reaction expansion, vol. 16. World Cement Technology, London, UK; 1984c. p. 38692. [16] Bhatty MSY. Properties of blended cements made with Portland cement, cement kiln dust, y ash, and slag, vol. 1. In: Proceedings of the International Congress on the Chemistry of Cement, Communications Theme-3; 1986. p. 11827. [17] El-Sayed HA, Gabr NA, Hana S, Mohran MA. Reutilization of by-pass kiln dust in cement manufacture. In: Proceedings of the International Conference on Blended Cement in Construction; 1991. [18] Heikal M, Aiad I, Helmy MI. Portland cement clinker, granulated slag and bypass cement dust composites. Cem Concr Res 2002;32:180512. [19] Konsta-Gdoutos MS, Shah SP. Hydration and properties of novel blended cements based on cement kiln dust and blast furnace slag. Cem Concr Res 2003;33:126976. [20] Shoaib MM, Balaha MM, Abdel-Rahman AG. Inuence of cement kiln dust substitution on the mechanical properties of concrete. Cem Concr Res 2000;30:3717. [21] Wang K, Shah SP, Mishulovich A. Effects of curing temperature and NaOH addition on hydration and strength development of clinker-free y ash binders. Cem Concr Res 2004;34:299309. [22] Juenger MCG, Jennings HM. Effect of high alkalinity on cement pastes. ACI Mater J 2001;98:2515. [23] Bhatty MSY. Use of Kiln Dust in Blended Cements. PCA R&D Serial No. 1717, Portland Cement Association, Skokie, Illinois, USA; 1983.

4. Conclusions 1. Treatment of cement kiln dust with Bacillus halodurans strain KG1 has positive effects on the properties of CKD concrete. 2. Increase in 7.15% and 26.6% compressive strength of concrete having 10% bacterial treated CKD after 28 and 91 days, respectively, and decrease in water absorption (20%) and porosity (12.35%) at 91 days was achieved whereas above 10%, decrease in strength was observed due to reduced hydration reaction and lower cement content. 3. The late strength development in bacterial treated CKD concrete is probably due to reduced hydration reaction, increased calcium silicate hydrate gel formation and deposition of bacterial cells or spores within the pores of concrete.

Kunal et al. / Construction and Building Materials 52 (2014) 4251 [24] Wilson RD, Anable WE. Removal of alkalies from Portland CKD. US department of Interior, Bureau of mines report of investigations number 9032, 2; 1986. [25] Kosmatka SH, Panarese WC. Design and control of concrete mixtures. 13th ed. Skokie, Illinois: Portland Cement Association; 1990. p. 156. [26] Rehsi SS, Garg SK. Production of cement from cement kiln dust. Indian Conc J 1986;60:37. [27] Ghanem H, Zollinger D, Lytton R. Determination of the main parameters of alkali silica reaction using system identication method. J Mater Civil Eng 2010;22:86573. [28] Davis TA, Hooks DB. Disposal and Utilization of Waste Kiln Dust from Cement Industry, National Environmental Research Center, Ofce of Research and Development, United States Environmental Protection Agency, Cincinnati, Ohio, USA, EPA 6702-75-043; 1975. [29] Mohamed AMO, El-Gamal MM. Method for treating cement kiln dust. US Patent 8,043,426 B2; 2011. [30] Gebhardt RF. Method of treating cement kiln dust for recovery and recycle. US Patent 6331207; 2001. [31] Tanabe H, Yoshihara K, Tamura K, Kobayashi Y, Akamatsu I, Niyomwan N, et al. Pretreatment of pectic wastewater from orange canning process by an alkalophilic Bacillus sp. J Ferment Technol 1987;65:2436. [32] Ntougias S, Zervakis GI, Ehaliotis C, Kavroulakis N, Papadopoulou KK. Ecophysiology and molecular phylogeny of bacteria isolated from alkaline two phase olive mill wastes. Res Microbiol 2006;157:37685. [33] Kumar R, Sharma A, Dhall P, Kulshreshtha NM, Kumar A. Neutralization of alkaline waste waters using a blend of microorganisms. World Acad Sci Eng Technol 2011;76:5036. [34] BIS 81121989. Specications for 43-grade Portland cement. New Delhi, India: Bureau of Indian Standards; 1989. [35] IS 3831970. Specications for coarse and ne aggregates from natural sources for concrete. New Delhi, India: Bureau of Indian Standards; 1970. [36] APHA, AWWA, WEF. Standard methods for examination of water and waste water. 21st ed. Washington (DC): American Public Health Association; 2005. [37] IS 102621982. Recommended guidelines for concrete mix design. New Delhi, India: Bureau of Indian standards; 1982. [38] IS 5161959. Indian standard code of practice methods of test for strength of concrete. New Delhi, India: Bureau of Indian Standards; 1959. [39] ASTM. Standard test method for density, absorption and voids in hardened concrete. ASTM C642, West Conshohocken, PA; 1997. [40] El-Aleem SA, Abd-El-Aziz MA, Heikal M, El-Didamony H. Effect of cement kiln dust substitution on chemical and physical properties and compressive strength of portland and slag cements. Arabian J Sci Eng 2005;30:26373.

51

[41] Wang K, Konsta-Gdoutos MS, Shah SP. Hydration, rheology, and strength of ordinary portland (OPC)-cement kiln dust (CKD)-slag binders. ACI Mater J 2002;99:1739. [42] Pu X. Investigation on pozzolanic effect of mineral additives in cement and concrete by specic strength index. Cem Concr Res 1999;29:9515. [43] Ghosh P, Mandal S, Chattopadhyay BD, Pal S. Use of microorganism to improve the strength of cement mortar. Cem Concr Res 2005;35:19803. [44] Vempada SR, Reddy SSP, Rao MVS, Sasikala Ch. Strength enhancement of cement mortar using microorganisms an experimental study. Int J Earth Sci Eng 2011;4:9336. [45] Park SJ, Chun WY, Kim WJ, Ghim SY. Application of alkaliphilic biolm forming bacteria to improve compressive strength of cement-sand mortar. J Microbiol Biotechnol 2012;22:3859. [46] Salem Th, Ragai MSh. Electrical conductivity of granulated slag cement kiln dust-silica fume paste at different porosities. Cem Concr Res 2001;31: 7817. [47] Jumate E, Manea DL. X-Ray difraction study of hydration processes in the Portland cement. J Appl Eng Sci 2011;1:7986. ; 2003. [48] Neville AM. Proprieta t ile betonului. Bucures ti: Editura tehnica [49] Molnar L, Manea D, Aciu C. The study of hydration processes of cement based on latest generation methods, vol. 1. In: Proceedings of the Internaional Scientic Conference, CIBv 2010, ISSN 1843-6617, Transilvania University Press; 2010. p. 21521. [50] Min D, Mingshu T. Formation and expansion of ettringite crystals. Cem Concr Res 1994;24:11926. [51] Heinz D, Ludwig U. Mechanism of Secondary Ettringite Formation in Mortars and Concretes. ACI SP100-105; 1987. p. 205971. [52] Samoui N, Berube MA, Fournier B, Bissonnette B, Durand B. Effects of alkali addition on the mechanical properties and durability of concrete. Cem Concr Res 2005;35:20312. [53] Divet L, Pavoine A. Delayed ettringite formation in massive concrete structures: An account of some studies of degraded bridges. in: International RILEM TC 186-ISA Workshop on internal sulfate attack and delayed ettringite formation, Villars, Switzerland; 2002. p. 98126. [54] Famy C, Scrivener KL, Atkinson A, Lachowski E, Brough AR. Characterization of calcium-silicate hydrate products in expansive and non-expansive heat cured mortars-electron microscopy study. ACI Mater J SP 2000;192 24(192):385402. [55] Udoeyo FF, Hyee A. Strengths of cement kiln dust concrete. J Mater Civil Eng 2002;14:5246.