Influence of the composition of cement kiln dust on its ...

Cement and Concrete Research 54 (2013) 106–113

Contents lists available at ScienceDirect

Cement and Concrete Research journal homepage: http://ees.elsevier.com/CEMCON/default.asp

Influence of the composition of cement kiln dust on its interaction with fly ash and slag Piyush Chaunsali a, Sulapha Peethamparan b,⁎ a b

Department of Civil and Environmental Engineering, University of Illinois, Urbana-Champaign, IL 61801, USA Department of Civil and Environmental Engineering, Clarkson University, Potsdam, NY 13699, USA

a r t i c l e

i n f o

Article history: Received 26 May 2012 Accepted 2 September 2013 Available online xxxx Keywords: Cement kiln dust (CKD) (D) X-ray diffraction (B) Scanning electron microscopy (B) Transmission electron microscopy (B)

a b s t r a c t Cement kiln dust (CKD), a by-product of the cement industry, contains significant amounts of alkali, free lime, chloride and sulfate. Wide variation reported in the chemical composition of CKDs limits their potential application as a sustainable binder component in concrete. In the current study, the performance of two different CKDs as components in a novel binder is evaluated. Several binders are developed by blending CKDs with fly ash or slag. Binders with 70% CKD were prepared at a water-to-binder ratio of 0.4, and heat-cured at 75 °C to accelerate the strength development. The hydration progress was monitored using X-ray diffraction, and morphological examination was performed using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Ettringite and calcium aluminosilicate hydrate (C-A-S-H) were identified as the main hydration products in the hardened binder system. Strength development of CKD-based binder was found to be significantly influenced by its free lime and sulfate contents. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Cement kiln dusts (CKDs) are byproduct materials generated during the clinker manufacturing process. They contain significant amounts of alkali, sulfate and chlorides. A large portion of CKD with lower alkali, sulfate and chloride contents is recycled as raw material in the cement kiln; the remainder of CKDs is landfilled. Although CKD generation has been reduced significantly in recent years through continuous improvement in the cement manufacturing process, a considerable amount of CKD is still being landfilled. Efforts are underway to utilize the CKD in conjunction with other industrial byproducts in order to develop an alternative binding material for sustainable concrete. The effective utilization of CKD could reduce the environmental issues associated with the disposal of CKDs in landfills. The chemical composition of CKD changes among cement plants and within each cement plant over time, resulting in an uncertainty in the prediction of its performance as a binder component. Wide variations observed in the chemical composition of CKDs limit their potential application as value-added products. The alkali (Na2Oeq), sulfate, and free lime contents of CKD have been reported to be as high as 8, 16, and 30%, respectively [1]. The alkali present in CKD may play an important role in activating aluminosilicate-containing minerals, fly ash and slag, when CKD is used as a major component in such binder systems. In fact, commercial alkalis have been traditionally used in activating aluminosilicate containing materials [2–6]. Additionally, the presence of free lime and calcium

sulfate may also act as activators for aluminosilicates [7–9]. The strength development in lime-containing binder systems is mainly attributed to the formation of calcium silicate hydrate gel (C-S-H) as a result of a pozzolanic reaction, whereas the system containing CaSO4 results in the formation of ettringite (AFt), monosulfate (AFm) and syngenite phases in addition to the C-S-H gel formation [1,10,11]. It was also reported that the presence of alkali makes CaSO4 more effective in activating the slag in the binder system [12,13]. In general, CKDs containing a combination of these activators (CaO, CaSO4, and different forms of alkalis) could potentially result in a binding system with favorable mechanical properties when used in conjunction with fly ash and slag. Recent studies have shown positive initial results supporting this hypothesis [14–19]. In order to understand the influence of the physico-chemical characteristics of CKD on its performance as an effective binder component, a detailed investigation on the interaction of two types of CKDs with fly ash and slag was undertaken. Elevated temperature curing was used in addition to room temperature curing to expedite strength development. The compressive strength development of both binders was monitored under various curing conditions. The hydration progress and microstructural changes were examined using X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). 2. Materials and experimental methods 2.1. Materials

⁎ Corresponding author. Tel.: +1 315 268 4435; fax: +1 315 268 7985. E-mail addresses: [email protected] (P. Chaunsali), [email protected] (S. Peethamparan). 0008-8846/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.cemconres.2013.09.001

Fig. 1 shows the particle size distribution of the CKDs used in this study in comparison with that of the fly ash and slag. CKD(I) was finer

P. Chaunsali, S. Peethamparan / Cement and Concrete Research 54 (2013) 106–113

Fig. 2(c) shows the XRD patterns of both fly ash and slag. The XRD pattern of class ‘F’ fly ash shows a diffuse band (peak at about 22°) corresponding to the glassy phase as well as peaks indicating the presence of crystalline phases of silica and alumina. Similarly, a broader diffuse band was observed in slag, suggesting a higher fraction of amorphous content compared to fly ash.

Percentage Passing (%)

100 CKD (I) CKD (II) Slag Class F Fly Ash

80

107

60

2.2. Experimental methods 40

20

0 0.1

1

10

100

1000

Particle Size (µm) Fig. 1. Particle size distribution.

than CKD(II). CKD(I) was generated in a cement plant that uses a dry processing technology and a long cement kiln for the clinker production, whereas CKD(II) was generated in a cement plant that uses a long cement kiln but a wet processing technology. This difference in the clinker processing technology is reflected in the chemical composition presented in Table 1. The concentration of most of the reactive oxide components was slightly higher in CKD(I), except for CaO and K2O. The main difference in composition of the two CKDs was the alkali (Na2Oeq) content (3.1 versus 6%, respectively), the sulfate (SO3) content (10.62 versus 7.69%, respectively) and the free lime content (5 vs. 1.5%). In addition, CKD(II) had approximately 0.62% chloride which was nearly absent in CKD(I). In reinforced concrete, the permissible water-soluble chloride content of ~0.15% (by weight of cement) has been suggested [20]. The consequence of the high chloride content, in one of the CKDs, on the onset of corrosion of reinforcing bar in CKD-containing concrete is beyond the scope of this paper. Fig. 2(a) and (b) presents the difference in mineralogy of the two CKDs used in the study. Calcium carbonate (CaCO3), quartz (SiO2), anhydrite (CaSO4) and free lime (CaO) were the crystalline phases present in both the CKDs. CKD(II) contained a significant amount of alkali in the form of sylvite (KCl) and syngenite (K2CaSO4·2H2O), as shown in Fig. 2(b).

Table 1 Chemical composition of materials. % weight CKD(I)

CKD(II)

GGBFS

Fly ash (class F)

Chemical composition SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 Cl− Total

14.55 4.46 2.11 61.15 3.84 0.80 3.45 10.62 – 100.98

11.69 2.20 2.93 66.11 0.87 0.87 7.24 7.69 0.62 95.17

36.00 10.53 0.67 39.80 7.93 0.27 0.15 2.11 – 97.46

50.20 28.70 5.72 5.86 1.74 0.96 – 0.51 – 93.69

Additional information Na2Oeq Loss on ignition Mean particle size

3.10 23.40 ~4 μm

6.00 29.06 ~7 μm

0.32 3.00 ~7 μm

0.96 1.85 ~11 μm

Note: The minor components such as TiO2, MnO and P2O5 are excluded from the oxide composition.

2.2.1. Sample preparation Previous study by the authors examined the mineralogical and microstructural evolution of CKD(I)-based binders [17,18]. The purpose of this study was to compare the performance of CKDs from two different sources. The strength development, hydration progress and microstructure evolution of CKD(I) and CKD(I)-based binders were examined. All binders contained 70% (by weight) CKD. A constant temperature (75 ± 1 °C), and fixed curing duration (48 h) were used for both CKD(I) and CKD(II)-based binders. Class F fly ash/ground granulated blast furnace slag was homogenized with 70% CKD(I)/(II) (by mass of the total binder) in dry conditions prior to mixing with water. A constant water-to-binder ratio (w/b) of 0.40 was used to prepare the CKD–FA/CKD–Slag pastes following ASTM C 305 procedure. The prepared paste was then poured into 50 mm Plexiglas cube molds for compressive strength testing and 31.75 mm-diameter cylindrical molds for characterization studies. These molded specimens were then compacted on a vibrating table. The compressive strength test specimens were subsequently finished with a flat trowel. All specimens were sealed using a layer of plastic wrap followed by a layer of aluminum foil. The samples underwent 24 h of sealed curing at ambient temperature (23 ± 2 °C), followed by 48 h of heat curing in a precision laboratory oven set to 75 ± 1 °C. During the 48 h of heat curing, the specimens were placed in sealed plastic boxes. After the initial heat curing, the samples were immersed in saturated lime water until the specified day of test. Microstructural characterization was performed at various ages: i) 1 day (after 24 h of sealed curing at ambient temperature), ii) 3 days (after 24 h of sealed curing at ambient temperature and 48 h of heat curing), and iii) 31 days (after 24 h of sealed curing at ambient temperature, 48 h of heat curing and 28 days of saturated lime water curing). Finally, the compressive strengths of cubes were determined after heat curing (3 days) and subsequent saturated lime water curing (31 and 59 days) according to ASTM C 109. 2.2.2. Characterization techniques 2.2.2.1. X-ray diffraction. X-ray diffraction was performed using the Bruker DX-8 diffractometer operating at 40 kV and 30 mA. CuKα radiation of wavelength 1.5405 Å was used. Samples were scanned from 5° to 65° (2θ range) at the rate of 0.02° per second. The samples used for the analysis were ground to produce particles below 75 μm size. 2.2.2.2. Thermogravimetric analysis. A Perkin-Elmer thermogravimetric analyzer was used for thermal analysis. A powdered sample (b75 μm particle size) weighing approximately 25 mg was heated from 50 °C to 1000 °C in a nitrogen environment (flow rate: 40 ml/min) at a 10 °C per minute heating rate. 2.2.2.3. Scanning electron microscopy. The fractured surface of paste samples was examined using a JEOL JSM-7400F electron microscope at accelerating voltage of 15 keV under secondary electron mode. The samples were fractured to expose the fresh surface before mounting them on aluminum stubs using carbon paint. The samples were then sputter-coated with gold–palladium for SEM examination. 2.2.2.4. Transmission electron microscopy. For transmission electron microscopy (TEM) analysis, the powdered sample (passing through

108

P. Chaunsali, S. Peethamparan / Cement and Concrete Research 54 (2013) 106–113 CC CC

L

Q CC

A

CC L CC Arc

CC

CC CC

L

Q

CKD(I)

CC A

CC

S Sg

Q

CC L

CC

CKD(II)

CC CC

S

10 15 20 25 30 35 40 45 50 55 60 65 10 15 20 25 30 35 40 45 50 55 60 65

2-Theta (degrees)

2-Theta (degrees)

(a)

(b) Q

Mu

Q Mu

Mu

Fly Ash

Slag 10

15

20

25

30

35

40

45

50

55

60

65

2-Theta (degrees)

(c) Fig. 2. X-ray diffraction patterns of: (a) CKD(I), (b) CKD(II), and (c) class ‘F’ fly ash and slag (A — Anhydrite, CC — Calcite, L — Free lime, S — Sylvite, Sg — Syngenite, Q — Quartz, Mu — Mullite).

75 μm) was first diluted in isopropanol. Then a drop of solution was dispersed on a copper grid. The grids were kept under vacuum before examination in a JEOL 2010 TEM instrument under an acceleration voltage of 200 kV.

higher fineness of CKD(I) compared to CKD(II) (see particle size distribution in Fig. 1) also might have contributed towards better performance of CKD(I)-based binders. 50

CKD(I)-FA CKD(II)-FA CKD(I)-Slag CKD(II)-Slag

3.1. Influence of CKD type on the strength development of CKD-based binders The compressive strength of CKD–FA and CKD–Slag binders with two different CKDs was determined as per ASTM C 109. Fig. 3 shows the compressive strength of CKD-based binders (pastes) at three different curing periods (3, 31, and 59 days) which includes the 48 h of heat curing after initial 24 h of ambient curing and subsequent lime water curing. As shown in Fig. 3, CKD(I)-based binders gained higher strength than the CKD(II)-based binders. The compressive strength of the CKD(I)–Slag binder was in the range of 25–40 MPa, whereas the CKD(II)–Slag binder exhibited only 15–20 MPa strength under various curing conditions. Similar performance of CKD–FA binders can be seen in Fig. 3, where CKD(II)–FA resulted in lowest compressive strength under all curing conditions. The higher strength of the CKD(I)-based binders can be attributed to several factors; the higher free lime, sulfate, and other reactive components such as silica and alumina. Moreover,

Compressive Strength (MPa)

3. Results 40

30

20

10

0 0

10

20

30

40

50

60

70

Age (days) Fig. 3. Effect of CKD type on strength development of CKD–FA and CKD–Slag pastes.


3.2. Hydration progress in CKD-based binders 3.2.1. CKD–FA binder The growth of crystalline phases during the hydration of CKD–FA binders was monitored using XRD. The XRD patterns collected from samples after subjecting them to various curing conditions are presented in Fig. 4(a) and (b) for CKD(I)–FA and CKD(II)–FA mixtures, respectively. The crystalline hydration products identified in 1 day-old CKD(I)–FA paste sample before heat curing were ettringite (E), calcium hydroxide (CH), and gypsum (G). As expected, the pozzolanic reaction between calcium hydroxide and amorphous phases of fly ash was accelerated by heat curing. At the end of 48 h of heat curing, the amount of CH was significantly reduced, potentially due to its consumption during the pozzolanic reaction. Additionally, during the heat curing, the alkalis and sulfates were adsorbed onto the reaction products, mainly to C-S-H [10]. The adsorbed sulfates and the alkalis were presumably released into the pore solution later during the lime water curing

109

period, which contributed to additional precipitation of ettringite and arcanite (Fig. 4(a)). The main difference in the mineralogy of CKD(II)–FA binder compared to the CKD(I)–FA binder was the presence of Friedel's salt and sylvite (Fig. 4(b)) in heat cured samples at 3 days. Although ettringite was detected in one day old CKD(II)–FA sample, the peak almost disappeared after heat curing at 3 days. Friedel's salt and sylvite were found only in 3-day old samples; subsequent lime water curing led to an increase in the amount of ettringite and the disappearance of sylvite. The chlorides seem to have released into the pore solution during the lime water soaking period. Analysis of the pore solution can provide information regarding the fate of chlorides at later ages. Although the pore solution analysis and the fate of the chloride ions are important for predicting the corrosion resistance of reinforced concrete that contain the CKD-based binders, this analysis was beyond the scope of the current study. The use of CKD with high chloride content might cause chloride induced corrosion durability issues in reinforced concrete and

CC CC

CKD(I)-FA CKD(II)-FA

Q

E

Q

Arc E

CC E

CC CC CC CC E

31 d

E E

CC CC CC CC

Q CC

3d

31 d

3d

S Fr

CH

1d

1d

G

5 10 15 20 25 30 35 40 45 50 55 60 65

5 10 15 20 25 30 35 40 45 50 55 60 65

2-Theta (degree)

2-Theta (degree)

(a)

(b)

Fig. 4. X-ray diffraction patterns of CKD(I)–FA and CKD(II)–FA paste (Arc — arcanite, CH — calcium hydroxide, CC — calcite, E — ettringite, Fr — Friedel's salt, G — gypsum, Q — quartz, S — sylvite).

110


hence should be evaluated thoroughly prior to using it in the field. Only traces of CH were present in the CKD(II)–FA mixture, mainly because of the low free lime content of CKD(II). 3.2.2. CKD–Slag binder Fig. 5 compares the hydration progress of CKD(I)–Slag and CKD(II)– Slag binder. Ettringite was present in both binders at all ages, although it was found in smaller quantities in CKD(II)–Slag binder. Interestingly, the precipitation of Friedel's salt (C3A·CaCl2·10H2O) was observed in CKD(II)–Slag binder at early ages (i.e., before and after heat curing) (Fig. 5(b)). Konsta-Gdoutos and Shah also observed the Friedel's salt at early-age in CKD–Slag binder prepared with CKD having high chloride content [14]. In their study, Friedel's salt was observed in the CKDs with 1.4% and 1.8% chloride content. After saturated lime water curing, Friedel's salt was not detected in CKD(II)–Slag binder. It is known that heat curing results in adsorption of sulfate ions into C-S-H gel [10], and after subsequent saturated lime water curing, the concentration of sulfate ions in the pore solution increases, presumably due to the desorption of the sulfate ions from the C-S-H. This increase of sulfate concentration in the pore

solution favors the precipitation of ettringite at later ages along with the depletion of Friedel's salt as shown in Fig. 5(b). As discussed before, it appears that the chloride ions were released into pore solution during the saturated lime water curing. Contrary to what was observed in CKD(I)– Slag binder, only traces of CH were detected in CKD(II)–Slag binder. This can be attributed to the relatively low free lime present in CKD(II). 3.3. Microstructural changes in CKD-based binders 3.3.1. CKD–FA binder 3.3.1.1. SEM. Fig. 6(a) and (b) shows the morphological difference in the reaction products of CKD(I)–FA and CKD(II)–FA mixtures after 48 h of heat curing, respectively. It appears that, as shown in Fig. 6(a), almost all of the fly ash particles were covered with reaction products in CKD(I)–FA mixture, whereas in CKD(II)–FA mixtures the reaction was not that extensive. Even fine fly ash particles were present without showing any signs of reaction (Fig. 6(b)). Fig. 7(a) and (b) shows the SEM micrographs of a fly ash grain and the surrounding area in

CC

CC

CKD(I)-Slag

E

CKD(II)-Slag

Q CC E

E

Q CC CC CC CC E

31 d

E

CC CC CC CC

CC E Q

3d

CH G

1d

31 d

3d

Fr

Sg

1d

5 10 15 20 25 30 35 40 45 50 55 60 65

5 10 15 20 25 30 35 40 45 50 55 60 65

2-Theta (degree)

2-Theta (degree)

(a)

(b)

Fig. 5. X-ray diffraction patterns of CKD(I)–Slag and CKD(II)–Slag paste (CH — calcium hydroxide, CC — calcite, E — ettringite, Fr — Friedel's salt, G — gypsum, Q — quartz, Sg — syngenite).


111

Fig. 6. SEM micrographs of: (a) CKD(I)–FA, and (b) CKD(II)–FA after heat curing (3 days).

CKD(I)–FA and CKD(II)–FA binders, respectively. Higher reactivity of fly ash in CKD(I)–FA binder also resulted in higher strength in comparison to CKD(II)–FA binder.

of reaction gel can be observed in Fig. 9(a) and (b). The difference in the morphology of reaction gel may also have an influence on strength development in CKD–Slag binders.

3.3.1.2. TEM. To further examine the morphology of C-A-S-H gel present in CKD-based fly ash binders, the microstructure was examined under a TEM. A fibrillar morphology for C-A-S-H is clearly evident in this TEM micrograph. In TEM examination, some of the regions (including the location around the fly ash particle) had a relatively different morphology, as seen in Fig. 8(b).

3.3.2.2. TEM. TEM examination revealed the fibrillar morphology of C-S-H gel formed in CKD(I)–Slag and CKD(II)–Slag binders (Fig. 10).

3.3.2. CKD–Slag binder

It is evident that CKD(I)-based binders attained higher strength than CKD(II)-based binders. The fly ash and slag were activated by calcium hydroxide and alkalis released during CKD hydration. Furthermore, calcium sulfate acted as an activator in the presence of alkalis [10,12].

3.3.2.1. SEM. Fig. 9 shows the microstructure of CKD(I)–Slag and CKD(II)–Slag paste after heat curing. A relatively different morphology

4. Discussion 4.1. Factors influencing the development of compressive strength

Fig. 7. SEM micrographs of: (a) CKD(I)–FA, and (b) CKD(II)–FA after heat curing (3 days).

112


Fig. 8. (a) TEM micrograph of CKD(I)–FA after heat curing (3 days), and (b) the light morphology of C-S-H around the shell of fly ash particle.

Higher strength of CKD(I)-based binders can be attributed to higher amount of ettringite and increased activation of fly ash and slag particles. Ettringite was present in larger amounts in CKD(I)-based binders than in CKD(II)-based binders, possibly due to higher SO3 and alumina contents. In addition, the higher lime content of CKD(I) resulted in a larger amount of C-A-S-H gel formed during the pozzolanic reaction in CKD(I)-based binders than in CKD(II)-based binders. In spite of the higher alkali content, CKD(II) did not exhibit better performance compared to CKD(I) with respect to the strength development. The calcium hydroxide formed during CKD hydration triggers the pozzolanic reaction in fly ash-containing binders and results in the formation of C-AS-H. Since CKD(I) had higher free lime content, larger amounts of reaction gel are expected to form in CKD(I)–FA based binders. In addition, higher sulfate content in CKD (I) caused more ettringite formation in CKD(I)-based binders. It seems that the small difference in free lime and sulfate contents of the CKD significantly alters the strength of the resulting binder. The results of this study agree with the findings of a previous study where the sulfate and lime contents of CKD were found to be the key factors influencing the binder system's strength development [14].

4.2. Implication of CKD chemistry on the composition and morphology of microstructure The morphology of reaction gel in CKD-based binders was observed to be fibrillar, as illustrated by TEM examination. Furthermore, SEM

analysis indicated a difference in the morphology of the binder resulting from two different CKDs. The dissolution of fly ash particles, as shown by SEM examination of the CKD–FA binders, indicated the fly ash's higher reactivity in the presence of CKD(I), which resulted due to CKD(I)'s higher lime and sulfate contents. It is noted that the elevated temperature curing might alter the morphology of the reaction product (C-A-S-H) in CKD-based binders. Therefore, the results presented here are solely for the samples heat cured at 75 °C.

4.3. Role of alkalis and sulfates on stability of ettringite In this study, ettringite was found to be stable in CKD(I)-based binders even at high temperature (75 °C), as indicated by X-ray diffraction. On the other hand, the amount of ettringite was reduced after heat treatment in CKD(II)-based binders. The stability of ettringite at a particular temperature depends on the pH of the pore solution and the availability of sulfate ions. Damidot and Glasser [21] reported the pH ranges for ettringite stability as 10.87–12.25 and 10.43–12.52 for 85 °C and 25 °C respectively. Their study suggests that the upper threshold for pH is reduced with an increase in temperature. Similarly, the sulfate concentration dictates which form would be dominant between AFt and AFm phases. Therefore, higher alkali content and lower sulfate content of CKD(II) appear to have decreased the stability of ettringite at 75 ° C. Experimental findings in this work clearly show the beneficial effect of free lime and sulfate contents in strength development of CKD-based binders. However, the efficacy of alkalis in CKD needs to be investigated

Fig. 9. SEM micrographs of: (a) CKD (I)–Slag and (b) CKD(II)–Slag paste after 48 h of heat curing (3 days).


113

Fig. 10. TEM micrographs of: (a) CKD(I)–Slag paste, and (b) CKD(II)–Slag paste after heat curing (3 days), showing the fibrillar morphology.

further to determine their role on mechanical strength development of the binder. 5. Conclusions The influence of CKD composition on the microstructure and strength development of CKD-based binders is discussed in this study. The essential findings are summarized as follows: • Free lime and sulfate contents of CKD influenced the strength development of CKD-based binders significantly. The CKD with higher free lime and sulfate contents attained higher compressive strength. • Ettringite and C-S-H are the main hydration products found in CKDbased binders. In CKD with higher alkali content and lower sulfate content, the formation of ettringite was decreased during the heat curing. • TEM investigation revealed the fibrillar morphology of C-S-H gel in CKD-based binders. References [1] S. Peethamparan, J. Olek, J. Lovell, Influence of chemical and physical characteristics of cement kiln dusts (CKDs) on their hydration behavior and potential suitability for soil stabilization, Cem. Concr. Res. 38 (6) (2008) 803–815. [2] J. Davidovits, Geopolymers: inorganic polymeric new materials, J. Therm. Anal. 37 (8) (1991) 1633–1656. [3] S.D. Wang, K.L. Scrivener, P.L. Pratt, Factors affecting the strength of alkali activated slag, Cem. Concr. Res. 24 (6) (1994) 1033–1043. [4] A. Palomo, M.W. Grutzeck, M.T. Blanco-Varela, Alkali activated fly ashes: a cement for the future, Cem. Concr. Res. 29 (8) (1999) 1323–1329.

[5] C. Shi, R.L. Day, Early strength development and hydration of alkali-activated blast furnace slag/fly ash blends, Adv. Cem. Res. 11 (4) (1999) 189–196. [6] A. Fernández-Jiménez, A. Palomo, Composition and microstructure of alkali activated fly ash binder: effect of the activator, Cem. Concr. Res. 35 (10) (2005) 1984–1992. [7] C. Shi, Studies on several factors affecting hydration and properties of lime–pozzolan cements, J. Mater. Civ. Eng. 13 (6) (2001) 441–445. [8] P. Mira, V.G. Papadakis, S. Tsimas, Effect of lime putty addition on structural and durability properties of concrete, Cem. Concr. Res. 32 (5) (2002) 683–689. [9] C.S. Poon, S.C. Kou, L. Lam, Z.S. Lin, Activation of fly ash/cement systems using calcium sulfate anhydrite (CaSO4), Cem. Concr. Res. 31 (2001) 873–881. [10] H.F.W. Taylor, Cement Chemistry, 2nd edition Thomas Telford, London, 1997. [11] A. Gruskovnjak, B. Lothenbach, F. Winnefeld, R. Figi, S.-C. Ko, M. Adler, U. Mäder, Hydration mechanisms of super sulphated slag cement, Cem. Concr. Res. 38 (2008) 983–992. [12] M. Regourd, Structure and Behavior of Slag Cement Hydrates, Proceedings of the 7th International Congress on the Chemistry of Cement, Paris, vol. I, 1980, pp. 9–26. [13] A. Gruskovnjak, B. Lothenbach, L. Holzer, R. Figi, F. Winnefeld, Hydration of alkali-activated slag: comparison with ordinary portland cement, Adv. Cem. Res. 18 (3) (2006) 119–128. [14] M.S. Konsta-Gdoutos, S.P. Shah, Hydration and properties of novel blended cements based on cement kiln dust and blast furnace slag, Cem. Concr. Res. 33 (8) (2003) 1269–1276. [15] C.E. Pierce, H. Tripathi, T.W. Brown, Cement kiln dust in controlled low-strength materials, ACI Mater. J. 100 (6) (2003) 455–462. [16] K. Wang, S.P. Shah, A. Mishulovich, Effects of curing temperature and NaOH addition on hydration and strength development of the clinker-free CKD–fly ash binders, Cem. Concr. Res. 34 (2) (2004) 299–309. [17] P. Chaunsali, S. Peethamparan, Microstructural and mineralogical characterization of cement kiln dust-activated fly ash binder, Transp. Res. Rec. 2164 (2010) 36–45. [18] P. Chaunsali, S. Peethamparan, Evolution of strength, microstructure and mineralogical composition of a CKD–GGBFS binder, Cem. Concr. Res. 41 (2) (2011) 197–208. [19] P. Chaunsali, S. Peethamparan, Novel cementitious binder incorporating cement kiln dust: strength and durability, ACI Mater. J. 110 (3) (2013) 297–304. [20] M. Thomas, Chloride thresholds in marine concrete, Cem. Concr. Res. 26 (4) (1996) 513–519. [21] D. Damidot, F.P. Glasser, Thermodynamic investigation of the CaO–Al2O3–CaSO4– H2O System at 50 °C and 85 °C, Cem. Concr. Res. 22 (6) (1992) 1179–1191.