Microstructural and Mineralogical Characterization of ... - SAGE Journals

Microstructural and Mineralogical Characterization of Cement Kiln Dust–Activated Fly Ash Binder Piyush Chaunsali and Sulapha Peethamparan new alkali incorporated binding product is formed (10). The potential of calcium hydroxide in activating fly ash has also been evaluated by some researchers (11, 12). As with sodium and potassium hydroxides, calcium hydroxide increases the pH of the systems and causes increased solubility of silicon dioxide (SiO2), which in turn causes the corrosion of densified outer layer of fly ash particles. The present study investigates the use of cement kiln dust (CKD) as an activator for Class F fly ash with the objective of developing alternative binders for making sustainable concrete. CKDs are fine powdery materials collected from the exhaust system of a cement kiln through electrostatic precipitators during the manufacturing process of cement. Beneficial use of CKD has been reported in various applications, such as soil stabilization and solid waste modification, and in controlled low-strength materials (13–19). CKDs have been experimentally demonstrated as an effective stabilizer for different types of expansive and nonexpansive clays by influencing the physical and mechanical properties of the clays (15–19). CKDs contain various percentages of sodium oxide, potassium oxide, and calcium oxide in their chemical compositions; all of these minerals have been independently found to activate fly ash and slag in previous studies (20–23). Being a by-product of cement industry, CKD might be a potentially less expensive alternative for commercial alkalis to activate flay ash. Moreover, CKDs are less caustic than other conventional activators used in alkali activation, making it more suitable in practical applications (18). Some preliminary results have been reported on the use of CKD as an activator for fly ash previously (23). In that study, the effect of curing temperature and the addition of sodium hydroxide on the compressive strength of CKDactivated fly ash (CKD-FA) binder was investigated. In the same study, it was also reported that hydrothermal activation improved the mechanical strength of CKD-FA binder significantly compared with chemical activation. The present paper deals with a detailed mineralogical and morphological study, carried out to understand the hydration of the CKD-FA binder system. At first, a CKD-FA binder material with relatively good compressive strength was developed after determining the optimum curing temperature and optimum binder proportions. Then the compressive strength of the optimum CKD-FA binder was compared with that of pure CKD sample, to establish strength enhancement of the binder through activation. The mineralogical composition and microstructure of developed clinker-free binder were characterized using techniques such as X-ray diffraction (XRD), thermogravimetric analysis (TGA), and scanning electron microscopy (SEM), coupled with an energy dispersive X-ray detector (EDX). Results from the present study show that activation of fly ash by CKD is a promising process to develop an effective binding material and may help lead to beneficial use of CKD waste.

A clinker-free binder for making sustainable concrete was developed using cement kiln dust (CKD) and Class F fly ash. The CKD-activated fly ash binder developed a compressive strength of approximately 30 MPa after 48 h of elevated temperature curing. The mineralogical composition of the developed clinker-free binder was determined with the help of thermogravimetric and X-ray analysis. Microstructure of the hardened binder was examined to identify phases using a scanning electron microscope equipped with an energy dispersive X-ray detector. Various reaction products, such as calcium aluminosilicate hydrate (C-A-S-H), ettringite (Aft), and calcium silicate hydrate, were identified in the hardened clinker-free binder. The major contribution to strength development is attributed to the C-A-S-H gel, which was found extensively as a ground mass in the hardened CKD-activated fly ash system.

Concern over greenhouse gas emission during the manufacturing process of cement has urged the cement and concrete industry to use an increased amount of pozzolanic materials in concrete. Radical efforts are also under way to develop clinker- and cement-free alternative binding materials for producing more environmentally friendly and energy-efficient, sustainable concrete (1). Alkali activation of silica and alumina containing industrial by-products, such as fly ash and slag, can be considered the main precursor in the development of such cement-free binder concretes. Pozzolanic material such as fly ash has been widely studied as a source material for alkali-activated concrete in the last few decades (2–4). Alkali activation of pozzolanic materials is a chemical process that transforms the glassy structures into very well-compacted reaction products. Due to high degree of polymerization exhibited by most of the fly ashes, especially Class F fly ash, chemical and hydrothermal activations are found to be effective methods for accelerating pozzolanic reaction (5). In the case of chemical activation, the type and concentration of the activating solution have significant influence on properties of the resulting binder (6–8). Sodium and potassium hydroxides and sodium silicate have been successfully used in several studies as effective chemical activators for a range of materials (9). The dissolution of alumina and silica from the existing polymeric chain in fly ash takes place at a pH value of ∼13.3 or higher, and a Department of Civil and Environmental Engineering, Clarkson University, 8 Clarkson Avenue, Potsdam, NY 13699. Corresponding author: S. Peethamparan, speetham@ clarkson.edu. Transportation Research Record: Journal of the Transportation Research Board, No. 2164, Transportation Research Board of the National Academies, Washington, D.C., 2010, pp. 36–45. DOI: 10.3141/2164-05

36

Chaunsali and Peethamparan

37

EXPERIMENTAL PROGRAM

100

Materials

CKD Fly Ash

80 Percentage Passing (%)

In this study, Class F fly ash conforming to ASTM C618 and a relatively low free lime content CKD were used for developing the cement-free binder. The Class F fly ash and the CKD are referred as the source material and the activator, respectively. The chemical composition of these materials is presented in Table 1. The total calcium oxide (CaO) content of the CKD was 61.15%, which includes all the CaO present in the calcite and other CaO-bearing minerals, such as quicklime and anhydrite. The loss on ignition of CKD was 23.40%, which suggests that a large portion of the CaO was in the form of calcite (CaCO3). In addition to calcite, the oxide composition of the CKD used in the present study contained large amounts of alkalis [sodium oxide (Na2O) and potassium oxide (K2O)] and sulfates. The particle size distribution of the CKD (as shown in Figure 1a) indicated that this particular CKD was very fine with an average particle size of ∼4 µm. Figure 1b shows the XRD analysis of the CKD that confirmed the presence of calcite, quicklime, anhydrite, and arcanite (K2SO4) minerals. The specific fly ash used for the study had an Si/Al ratio of approximately 1.75. A typical XRD pattern for the fly ash is given in Figure 1b, which shows the presence of both amorphous and crystalline phases. Main crystalline phases identified in the fly ash XRD pattern were quartz (SiO2) and mullite (Al6Si2O13), which are fairly typical for low-calcium (Class F) ashes. The particle size distribution of the fly ash is also included in Figure 1a. The average particle size of the fly ash was ∼20 µm.

60

40

20

0 0.01

0.1

QL CC CC Anh QL CC CC Q Arc CC

% by Mass Fly Ash

CKD

50.20 28.70 5.72 5.86 1.74 0.96 — 0.51 0.96 1.85

14.55 4.46 2.11 61.15 3.84 0.80 3.45 10.62 3.10 23.40

NOTE: Al2O3 = aluminum oxide, Fe2O3 = iron oxide, MgO = magnesium oxide, SO3 = sulfur trioxide, (Na2O)eq = Na2O equivalents, — = not determined.

QL

CKD

Q

Mu

5 TABLE 1 Chemical Compositions of Fly Ash and Cement Kiln Dust

1000

CC

A preliminary set of experiments was conducted to determine the optimum curing temperature conditions and mixture proportion for obtaining maximum compressive strength. The optimum curing temperature and mixture proportions thus obtained were then used for the rest of the study. Source material (fly ash) and activator (CKD) were homogenized in dry conditions before mixing with water. A constant water-to-binder ratio (w/b) of 0.40 was used to prepare the paste following the ASTM C305 procedure.

SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 (Na2O)eq Loss on ignition

100

(a)

Sample Preparation

Chemical Composition

1 10 Particle Size (µm)

10

15

Q

20

Mu

25 30 35 40 45 Angle 2-theta (degrees) (b)

Fly Ash

50

55

60

65

FIGURE 1 Physical properties and mineralogical compositions of fly ash and CKD: (a) particle size distribution and (b) X-ray diffraction pattern (CC ⴝ calcite; Anh ⴝ anhydrite; QL ⴝ quicklime; Arc ⴝ arcanite; Q ⴝ quartz; and Mu ⴝ mullite).

For compressive strength determination, the paste was poured into a cubical mold of 50 × 50 × 50 mm and compacted adequately using a table vibrator. The specimens were then kept in ambient temperature for 24 h before demolding. After 24 h, the specimens were demolded, wrapped in aluminum foil, and subjected to heat curing in a laboratory oven for 48 h. Some of the specimens were subjected to saturated lime water curing for an additional period of 28 days, followed by the initial heat curing. The saturated lime water was used to limit the leaching of ions from the pore solution due to pH gradient, for CKD has large amount of alkalis. The saturated lime water curing was not carried out in the preliminary set of experiments, in which the objective was to determine optimum curing temperature and mixture proportions. All the samples, including those used for

38

1. 24 h of ambient temperature curing [designated “24 h (A)” series], 2. 24 h in ambient temperature followed by 48 h of heat curing [designated “24 h (A) + 48 h (H)” series], and 3. 24 h in ambient temperature followed by 48 h heat curing and subsequent extended curing in lime water for 28 days [designated “24 h (A) + 48 h (H) + 28 d (L)” series].

40

Compressive Strength (MPa)

preliminary analysis, were allowed to cool down to ambient temperature for 30 min immediately after heat curing. Compressive strengths were determined according to ASTM C109. For mineralogical and microstructural analysis of CKD and CKDFA binders, paste specimens were prepared in small polypropylene cylinders. The same sample preparation procedure described previously was used, but the samples were prepared using the optimum CKD dosage (70%) and temperature of curing (75°C). Samples were collected for the analysis at the end of three curing regimes. They are

Transportation Research Record 2164

Determination of Optimum Binder Proportion The compressive strength was expected to have dependence on the proportions of CKD and fly ash in the binder. To determine the optimum mixture proportion that attains maximum compressive strength, samples of CKD-FA binder with varying CKD dosages were tested after heat curing at 75°C for 48 h. As shown in Figure 2b, the samples with 70% of CKD and 30% of fly ash by weight of total binder achieved maximum compressive strength. This optimum mixture proportion was used for all the further experiments. Setting Time and Flow Test To study the fresh paste properties, the workability (flow test) and the setting time of the CKD-FA paste prepared at the optimum proportion were determined. The setting time of cement paste is usually determined using Vicat test described in ASTM C191. In the standard test method, the paste used for determining the setting time

10

50

60

70 80 Temperature (°C) (a)

90

100

90

100

40


A previous study has shown that the hydrothermal activation enhanced the compressive strength of CKD-FA binder significantly (23). Hence, as mentioned earlier, a preliminary study was carried out to determine the optimum temperature for heat curing. To determine the effect of temperature on the compressive strength, three curing temperatures were used: 60°C, 75°C, and 90°C. The cubes were heat cured for 48 h at the respective temperatures, and compressive strength was determined after the specimens were cooled to room temperature. For this part of the study, the binder was prepared with 50% fly ash and 50% CKD. The w/b used was 0.40. Figure 2a shows the variation of the compressive strength with curing temperature. It can be seen that the maximum strength was achieved when temperature was 75°C. Hence, the optimum temperature of 75°C (for maximum compressive strength) was used for the rest of the study.

20

0

The collected samples were broken into small chunks and soaked in acetone for at least 2 days to arrest further hydration. The acetone soaked samples were dried for 1 day in vacuum desiccators before the analysis and stored in airtight glass vials. A small portion of the sample was ground and sieved through a 75-µm size sieve, and this powder sample was used for the TGA and the XRD analysis. Remaining samples were used for the SEM examination. Determination of Optimum Temperature for Heat Curing

50% CKD - 50% FA

30

30

20

10

0 50

60

70

80

CKD dosage in CKD-FA binder (%) (b) FIGURE 2 Effect of (a) curing temperature and (b) CKD dosage on compressive strength of CKD-FA binder (w/b ⴝ 0.40; curing temperature ⴝ 75ⴗC; and curing duration ⴝ 48 h).

should be prepared using the normal consistency water content. In the present study, setting time was performed on a paste prepared at specific w/b rather than using the normal consistency water content. The purpose of the test was to evaluate the setting time of the CKDactivated binder with a specific w/b. The setting time of the developed binder was compared with that of the CKD alone paste. Unlike fly ash, CKD reacts with water and forms new reaction products, resulting in the setting and hardening of CKD pastes. Flow test was performed according to ASTM C 437 to determine the flow of CKD and CKD-FA binders at a w/b of 0.40. Thermogravimetric Analysis Thermogravimetric analysis was performed using a Perkin-Elmer thermogravimetric analyzer. In this test, a powdered sample (pass-


X-Ray Diffraction Analysis XRD analysis was carried out using Bruker DX-8 diffractometer. CuKα radiation with a wavelength of 1.504 Å was used, and samples were scanned in 2θ range from 5° to 65° at a scanning rate of 0.02° per s. The hand-ground powdered sample passing through the 75-µm (No. 200) sieve was also used for the XRD examination.

40 100% CKD 70% CKD-30% FA Penetration Depth (mm)

ing through the 75-µm sieve) weighing 25 ± 2 mg was heated in a nitrogen environment from 50°C to 1,000°C at the rate of 10°C per min. The nitrogen gas flow rate was kept as 40 mL/min. TGA and derivative thermogravimetry (DTG) curves were obtained to identify hydration products in the CKD and CKD-FA binder systems.

39

Scanning Electron Microscopy

30 Initial Setting (25 mm)

20

10

0 0

SEM was performed using a JEOL JSM-7400F electron microscope, coupled with the EDX, operating in secondary mode. The sample was fractured to get a smooth surface with few undulations and was stuck on an aluminum stub using carbon paint. Gold coating was applied on the fractured surface of the sample. A lower electron image detector was used for taking images in the present study.

200

400

600

800

1000

1200

Elapsed Time (minutes) (a) 100

80

DISCUSSION OF RESULTS

The early-age properties such as the workability and setting time of the developed binder were determined and are presented in Figure 3. The fresh properties of a CKD paste at the same w/b were also evaluated to compare the difference in behavior of the CKD and CKD-FA pastes. Class F fly ash used in the present study did not have any self-cementing properties. Thus, it is assumed that the fly ash paste does not set during the 24-h period. Results show that the CKD-FA binder exhibits longer setting time than the CKD paste does. As can be seen in Figure 3a, the initial setting time was observed approximately 6 h for the CKD-FA binder system, while the CKD alone paste samples were set approximately 90 min after mixing. From the field application viewpoint, such a long setting time for CKD-FA binder might not be attractive. Methods to address this issue are currently being investigated. The flow values of CKD-FA binder and CKD samples are shown in Figure 3b. Compared with CKD, CKD-FA binder had a larger flow value. As expected, the addition of fly ash increased the workability of CKD because of the spherical shape of fly ash particles and the dilution effect resulting from the partial replacement of CKD. Compressive Strength The compressive strength of CKD alone and CKD-FA binder after 48 h of heat curing is shown in Figure 4. The addition of 30% of fly ash enhanced the compressive strength of CKD by 50%, from ∼13 MPa to 28 MPa. This significant enhancement in compressive strength of CKD by adding fly ash, which does not have any binding capability, indicates the important role played by the underlying activation reactions. The 28MPa compressive strength obtained for the CKDFA binder developed in this study is comparable with other conventional binders, demonstrating its potential to be used as a successful binder material in the field. It was also found that most of the acti-

Flow (%)

Setting Time and Flow Test

60

40

20

0 100% CKD

70% CKD - 30% FA (b)

FIGURE 3 Early-age properties of CKD and CKD-FA binder: (a) setting time test and (b) flow test (w/b ⴝ 0.40).

vation reactions occurred during the 48 h of heat curing. Compressive strength of CKD-FA binder after 28 days of lime water curing did not show any significant improvement as compared with compressive strength after 48 h of heat curing. That indicates no significant development of any additional strength-giving reaction during the extended curing period in the saturated lime water. Mineralogical Characterization

Cement Kiln Dust Paste TGA and X-ray analyses of the CKD paste are presented in Figure 5a and 5b, respectively. Both the analyses showed very similar characteristics. The main hydration products found after 24 h of curing at room temperature were ettringite (∼80°C), gypsum (∼120°C), calcium hydroxide (∼420°C), and calcite (∼700°C) (Figure 5a). The formation of gypsum and calcium hydroxide can be attributed to the

40


24 h (A) 24 h (A) + 48 h (H) 24 h (A) + 48 h (H) + 28 d (L)

100

30

CC E

TGA

90 Weight (%)


40

20

CC E CC

80

CH

70

10

E G CH CH

60

DTG

0

0 100% CKD

200

40 Tem 0 600 pera 80 ture (°C) 0

70% CKD - 30% FA

FIGURE 4 Compressive strength of CKD and CKD-FA binder (w/b ⴝ 0.40; curing temperature ⴝ 75ⴗC; and curing duration ⴝ 48 h).

1000

(a) CC

presence of anhydrite and quicklime present in the CKD. Some of the calcium sulfate might have been consumed during the formation of ettringite [C3A(CaSO4) 䡠 32H2O], while the rest of the anhydrites seem to be converted to gypsum at the end of 24 h of curing period. After 48 h of heat curing, a large increase in the ettringite peak height and complete absence of gypsum were observed, indicating that the gypsum formed was eventually consumed for the formation of additional ettringite. The considerable strength gain observed in the CKD alone samples could be attributed to the formation of ettringite (19). Only a slight change occurred in the calcium hydroxide peak (DTG) between the 24-h ambient temperature cured and 48-h heatcured CKD paste samples. The CaO was fully converted to calcium hydroxide in the first 24 h of hydration, and hence, any significant formation of additional calcium hydroxide after the 24-h period can be discounted. Therefore, the relatively stable peak of calcium hydroxide after 24 h of hydration indicates that it was not consumed for the formation of any reaction products during the heat curing period (Figure 5a). Similarly, as evident in X-ray analysis (Figure 5b), calcium hydroxide was present in all the samples irrespective of the method of curing. In addition, the 28 days of curing in saturated lime did not have any effect on the ettringite and calcium hydroxide formation, which can be verified through the unchanged DTG peaks for these hydration products in CKD paste. The XRD pattern of samples after 28 days saturated lime curing did not show any difference from that of 48 h of heat curing, supporting the TGA observation. The presence of the calcium hydroxide even after heat curing and saturated lime curing can be attributed to the absence of any possible pozzolanic reaction in the CKD paste, which would otherwise consume the calcium hydroxide generated in the system. The unaltered presence of calcite peak at all ages strongly confirms its inertness in the system.

Cement Kiln Dust–Fly Ash Paste The TGA and X-ray diffraction curves of CKD-FA binder are given in Figure 6a and 6b, respectively. Both the TGA and XRD curves

E CC Q Anh

E CH E

CH

CC CC CC

CC 24 h (A) + 48 h (H) + 28 d (L)

Arc

G

5

10

24 h (A) + 48 h (H)

G

15

20

24 h (A)

25 30 35 40 45 50 Angle 2-theta (degrees) (b)

55

60

65

FIGURE 5 Mineralogical characterization of CKD paste: (a) TGA with DTG plots and (b) XRD patterns (E ⴝ ettringite; CH ⴝ calcium hydroxide; CC ⴝ calcite; Anh ⴝ anhydrite; Arc ⴝ arcanite; Q ⴝ quartz; and G ⴝ gypsum).

of CKD-FA paste were very similar to that of CKD paste after 24 h of ambient temperature curing. As in the case of CKD samples, ettringite, gypsum, and calcium hydroxide were observed as the main reaction products after 24 h of ambient curing in CKD-FA binder. However, significant difference in TGA and XRD analysis can be noticed between CKD-FA binder and CKD samples after 48 h of heat curing. While the calcium hydroxide peaks were observed at all ages at the same level in pure CKD samples, calcium hydroxide was completely consumed during the 48 h heat curing of CKD-FA binder (Figure 6a and 6b). This phenomenon indicates that the 48-h heat curing activated the pozzolanic reaction between calcium hydroxide formed in the hydrated CKD and the reactive silica and alumina present in the fly ash, forming cal-


41

Morphological Characterization 24 h (A) 24 h (A) + 48 h (H) 24 h (A) + 48 h (H) + 28 d (L)

100 E

TGA

Weight (%)

90

CC E

80

CC CC

70 E G

60

DTG

0

CH

200

400 600 Tem pera 800 ture (°C)

1000

CC

Q E

E

CC

Arc

CC CC CC CC E

24 h (A) + 48 h (H) + 28 d (L)

24 h (A) + 48 h (H)

CH G

5

CH

Four SEM micrographs taken from various locations of a 48-h heatcured CKD paste sample are presented in Figure 7. A number of mineralogically and morphologically different minerals were identified with the help of EDX spectrum. Locations where the four EDXs were taken are marked as 1, 2, 3, and 4 in the micrographs. From the EDX spectrum, the compositions at these locations were identified as sodium potassium sulfate, 1; calcium carbonate crystals, 2; ettringite crystals, 3; and calcium silicate hydrate (C-S-H)-like gel, 4. Specimens were sputter coated with gold (Au), and hence, the Au peak appeared in all the EDX spectra. Generally, the microstructure of the 48-h heat-cured (75°C) CKD paste seemed to be porous, as seen in Figure 7d.

Cement Kiln Dust–Fly Ash Paste

(a)

E

Cement Kiln Dust Paste

24 h (A)

10 15 20 25 30 35 40 45 50 55 60 65 Angle 2-theta (degrees) (b)

FIGURE 6 Mineralogical characterization of CKD-FA paste at different curing periods: (a) TGA with DTG plots and (b) XRD patterns (E ⴝ ettringite; CH ⴝ calcium hydroxide; CC ⴝ calcite; Arc ⴝ arcanite; Q ⴝ quartz; and G ⴝ gypsum).

cium silicate hydrate or calcium aluminosilicate (C-A-S-H) gel. Another interesting observation was the formation of significantly large amount of potassium sulfate (arcanite) in the 28-day-old saturated lime-cured specimen (Figure 6b). The appearance of the K2SO4 peak in the XRD analysis of the prolonged saturated lime-cured sample seems to question the common notion that the alkalis (K2O and Na2O) present in the CKD may be used up in the formation of new strength-giving compound sodium or potassium aluminosilicate hydrate, as in the case of alkali-activated fly ash or slag. Though the alkalis present in the CKD powder increased the initial pH of the solution and helped the dissolution of fly ash, based on our results the alkalis present in the CKD did not seem to be permanently incorporated in the C-A-S-H structure.

Figure 8 shows the microstructure of a 48-h heat-cured CKD-FA paste. As seen in the CKD paste sample earlier, ettringite was present in CKD-FA at many locations. In the CKD-FA sample, the ettringite (Figure 8a) was found mostly around the fly ash particles. Most of the fly ash particles were covered with a reaction shell, as seen in Figure 8b. The approximate chemical composition of the outer shell was determined at location 5 by the EDX analysis, and a typical composition is presented in the pattern marked as point 5. The composition of the shell was different from that of the unreacted inner fly ash surface, shown in spectrum point 6. The higher calcium and oxygen peaks in spectrum point 5 compared with those of spectrum point 6 suggest the initiation of reaction products formation on the surface of fly ash particle. The exact quantitative chemical composition cannot be obtained using the EDX analysis of the fractured surface; only a qualitative comparison can be made. Generally, the microstructure of CKD-FA sample was much denser than that of CKD. The gel-like structure (Figure 8c) found as a ground mass in the CKD paste had a composition of C-A-S-H. A typical spectrum collected from such a location point 7 shows the composition of gel as C-A-S-H. The presence of this reaction gel is the major difference between the CKD-FA binder and 48-h heatcured CKD paste sample. Hence, improved strength exhibited by the CKD-FA binder system after heat curing could be attributed to these reaction products. Also of interest is that some fly ash particles are still unreacted in the system (Figure 8d), while others were dissolved or partially reacted (Figure 8b). This indicates further room for dissolution of fly ash particles forming more reaction gel and potential strength enhancement.

CKD-FA Paste After Prolonged Curing Period As discussed earlier, the compressive strength of the CKD-FA paste samples cured in saturated lime water for 28 days did not show any variation from those of 48-h heat-cured samples. This indicates the stability of the reaction product of CKD-FA system when expose to moisture. Figure 9 shows the scanning electron micrograph taken at two locations of samples after 28 days of saturated lime water curing. The main difference in the microstructure of the 48-h heat-cured and the 28-day moist-cured CKD-FA system was the presence of arcanite (K2SO4) crystals in the moist-cured samples. EDX and XRD analyses also confirmed the presence of arcanite in the 28-day moist-cured CKD-FA system. The presence of arcanite could be due

42


2

1

(a)

(b)

3

4

(c)

1

(d)

2

(e)

(f)

4 3

(g) FIGURE 7

SEM micrographs of 48-h heat-cured CKD paste, with EDX patterns.

(h)


43

Gel

E

5

6

Gel

(a)

(b)

Unreacted Fly Ash

CASH Gel

7

(c)

(d)

5

(e) FIGURE 8

6

(f)

SEM micrographs of 48-h heat-cured CKD-FA paste, with EDX patterns.

7

(g)

44


Gel

8

9

(a)

(b)

9

8

(c) FIGURE 9

(d)

SEM micrographs of CKD-FA paste after 28 days of lime curing, with EDX patterns.

to its recrystallization from the pore solution during prolonged curing period. However, further research is needed for verifying the role of the alkalis in the CKD-activated fly ash binder system.

CONCLUSION The CKD used in the present study was found to be effective in activating Class F fly ash. Within the study scope, the optimum binding proportion that gave the maximum compressive strength after 48 h of heat curing was 70% CKD and 30% fly ash. The developed binder achieved a compressive strength of approximately 28 MPa after 48 h. The setting time of the developed binder was approximately 6 h, and the flow value was relatively lower. It was found that most of the strength development was completed by the end of 48 h of heat curing, and further moisture curing in the lime water did not cause significant strength gain. The high pH environment created by the dissolution of the alkalis and the transformation of the quicklime present in the CKD to calcium hydroxide create a favorable environment for extensive dissolution of the fly ash in the CKD-FA system. Once the reactive silica and alumina

are available, the calcium hydroxide reacts with silica and alumina forming C-A-S-H. The thermal activation (heat curing) accelerated the pozzolanic reaction, which helped in formation of the C-A-S-H gel. The thermal activation also favored the formation of more ettringite in the system, and that might have also contributed towards the strength development. Although the alkali oxides caused increased dissolution of fly ash, it seems that the alkalis were not permanently incorporated in C-A-S-H gel structure. Further research is needed to quantify the presence of alkalis and its stability in the C-A-S-H gel. REFERENCES 1. Mehta, P. K. Greening of the Concrete Industry for Sustainable Development. Concrete International, Vol. 24, No. 7, 2002, pp. 23–28. 2. Davidovits, J. Geopolymers: Inorganic Polymeric New Materials. Journal of Thermal Analysis, Vol. 37, 1991, pp. 1633–1656. 3. Palomo, A., M. W. Grutzeck, and M. T. Blanco. Alkali-Activated Fly Ashes: A Cement for the Future. Cement and Concrete Research, Vol. 29, No. 8, 1999, pp. 1323–1329. 4. Xie, Z., and Y. Xi. Hardening Mechanisms of an Alkaline-Activated Class F Fly Ash. Cement and Concrete Research, Vol. 31, No. 9, 2001, pp. 1245–1249.


5. Taylor, H. F. W. Cement Chemistry, 2nd ed. Thomas Telford, London, 1997. 6. Fernández-Jiménez, A., and A. Palomo. Composition and Microstructure of Alkali Activated Fly Ash Binder: Effect of the Activator. Cement and Concrete Research, Vol. 35, No. 10, 2005, pp. 1984–1992. 7. Criado, M., A. Fernández-Jiménez, A. G. de la Torre, M. A. G. Aranda, and A. Palomo. An XRD Study of the Effect of SiO2/Na2O Ratio on the Alkali Activation of Fly Ash. Cement and Concrete Research, Vol. 37, No. 5, 2007, pp. 671–679. 8. Li, D., Y. Chen, J. Shen, J. Su, and X. Wu. The Influence of Alkalinity on Activation and Microstructure of Fly Ash. Cement and Concrete Research, Vol. 30, No. 6, 2000, pp. 881–886. 9. Hardjito, D., and B. V. Rangan. Development and Properties of LowCalcium Fly Ash-Based Geopolymer Concrete. Research Report. Curtin University of Technology, Perth, Australia, 2006. 10. Fraay, A. L. A., J. M. Bijen, and Y. M. deHaan. The Reaction of Fly Ash in Concrete: A Critical Examination. Cement and Concrete Research, Vol. 19, No. 2, 1989, pp. 235–246. 11. Mira, P., V. G. Papadakis, and S. Tsimas. Effect of Lime Putty Addition on Structural and Durability Properties of Concrete. Cement and Concrete Research, Vol. 32, No. 5, 2002, pp. 683–689. 12. Shi, C. Studies on Several Factors Affecting Hydration and Properties of Lime–Pozzolan Cements. Journal of Materials in Civil Engineering, Vol. 13, No. 6, 2001, pp. 441–445. 13. Baghdadi, Z. A., M. N. Fatani, and N. A. Sabban. Soil Modification by Cement Kiln Dust. Journal of Materials in Civil Engineering, Vol. 7, No. 4, 1995, pp. 218–222. 14. Pierce, C. E., H. Tripathi, and W. Brown Travis. Cement Kiln Dust in Controlled Low-Strength Materials. ACI Materials Journal, Vol. 100, No. 6, Nov.–Dec. 2003, pp. 455–462. 15. Bhatty, J. I. Alternative Uses of Cement Kiln Dust. R&D Bulletin 327. Portland Cement Association, Skokie, Ill., 1995.

45

16. Peethamparan, S., J. Olek, and S. Diamond. Physicochemical Behavior of Cement Kiln Dust–Treated Kaolinite Clay. In Transportation Research Record: Journal of the Transportation Research Board, No. 2059, Transportation Research Board of the National Academies, Washington, D.C., 2008, pp. 80–88. 17. Peethamparan, S., J. Olek, and S. Diamond. Mechanism of Stabilization of Na-Montmorillonite Clay with Cement Kiln Dust. Cement and Concrete Research, Vol. 36, No. 7, 2009, pp. 580–589. 18. Peethamparan, S., and J. Olek. A Study on the Effectiveness of Cement Kiln Dusts (CKDs) in Stabilizing Na-Montmorillonite Clay. Journal of Materials in Civil Engineering, ASCE, Vol. 20, No. 2, 2008, pp. 137–146. 19. Peethamparan, S., J. Olek, and J. Lovell. Influence of Chemical and Physical Characteristics of Cement Kiln Dusts (CKDs) on Their Hydration Behavior and Potential Application in Soil Stabilization. Cement and Concrete Research, Vol. 38, 2008, pp. 803–815. 20. White, D. J., and K. L. Bergeson. Long-Term Strength and Durability of Hydrated Fly-Ash Road Bases. In Transportation Research Record: Journal of the Transportation Research Board, No. 1755, Transportation Research Board of the National Academies, Washington, D.C., 2001, pp. 151–159. 21. Amin, A. M., E. Ebied, and H. EI-Didamony. Activation of Granulated Slag with Calcined Cement Dust. Silicates Industriels, Vol. 60, No. 3–4, 1995, pp. 109–115. 22. Konsta-Gdoutos, M. S., and S. P. Shah. Hydration and Properties of Novel Blended Cements Based on Cement Kiln Dust and Blast Furnace Slag. Cement and Concrete Research, Vol. 33, No. 8, 2003, pp. 1269–1276. 23. Wang, K., S. P. Shah, and A. Mishulovich. Effects of Curing Temperature and NaOH Addition on Hydration and Strength Development of the Clinker-Free CKD-Fly Ash Binders. Cement and Concrete Research, Vol. 34, No. 2, 2004, pp. 299–309. The Basic Research and Emerging Technologies Related to Concrete Committee peer-reviewed this paper.