A Three Step Process for Purification of Fly Ash Zeolites by Hydrothermal Treatment

Applied Clay Science 90 (2014) 122–129

Contents lists available at ScienceDirect

Applied Clay Science

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

Research paper

A three step process for purification of fly ash zeolites byhydrothermal treatment

Bhagwanjee Jha a, D.N. Singh b,⁎a Civil Engineering Department, Indian Institute of Technology Bombay, Powai, Mumbai 400 076, Indiab Civil Engineering Department, Indian Institute of Technology Bombay, Powai, Mumbai 400 076, India

⁎ Corresponding author. Tel.: +91 22 2576 7340; fax:E-mail addresses: [email protected] (B. Jha), dns@civil

0169-1317/$ – see front matter © 2014 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.clay.2013.12.035

a b s t r a c t

Article history:Received 21 March 2013Received in revised form 8 June 2013Accepted 27 December 2013Available online 28 January 2014

Keywords:Hydrothermal treatmentFly ashZeolitesImpuritiesPurification

Over the decades, synthetic zeolites have been synthesized from the fly ash by resorting to its hydrothermal ac-tivationwith NaOH, in one step. However, the activated ash (the residues) has been found to exhibit lower cationexchange capacity. Based on X-ray diffraction spectrometry, main culprits have been identified as Quartz andMullite (i.e., the inactivated or residual fly ash), which are in the form of impurities in the zeolites (i.e., Na-P1,in majority). Also, such impurities have been verified from the field emission gun-scanning electronmicroscopyof the residues. The purification of these zeolites still remains a big challenge and requires a special attention. Insuch a situation, this manuscript demonstrates a “three-step hydrothermal process”; the residues obtained afterStep-2 and Step-3 of this process significantly gain in their characteristics viz. cation exchange capacity by 185and 14% and specific surface area by 478 and 33.42%, respectively. In addition, lowering of the specific gravityand improvement in the SiO2/Al2O3 ratio of the residues are indicative of the formation ofmore porous and silicarich zeolites, in a purified form. Accordingly, an attempt has beenmade to assimilate the outcome of the adoptedprocess and develop a conceptual model to exhibit various complexities involved in overall conversion of the flyash to pure zeolites.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Several researchers have employed conventional hydrothermaltreatment of fly ash (HTF) and its conversion to fly ash zeolites (FAZ),which can be an alternative to some natural as well as man-made com-mercial zeolites, synthesized by using costlier chemicals (Adamczykand Bialecka, 2005; Inada et al., 2005; Kolay and Singh, 2001a,b,c,2002; Kolay et al., 2001; Murayama et al., 2002; Nugteren et al., 2001;Singh and Kolay, 2002). In addition, they have demonstrated thepresence of impurities (IMP), in the FAZ. The IMP mainly consist oftraces of partially activated glass and majority of inactivated Quartzand Mullite crystals, in the alkali activated fly ash (AFA), the residue.These impurities are responsible for lowering the cation exchangecapacity (CEC) of the FAZ and hence their purity, which in turn affectstheir suitability for industrial applications (viz., as an adsorbent whichuptakes heavy metal from contaminated soil and water). In these cir-cumstances researchers' focus has been to develop methodologies,which would facilitate removal of the IMP from the FAZ. However,such studies have been limited to the usage of higher molar solutionsof NaOH (Rayalu et al., 2000), high temperature of the treatment(Adamczyk and Bialecka, 2005; Kim et al., 1997) and/or prolongedperiods of hydrothermal activation of a fly ash (Scott et al., 2001).

+91 22 2576 7302..iitb.ac.in (D.N. Singh).

ghts reserved.

However, metastable FAZs have been reported to dissolve in high alka-line environment during prolonged activation, and hence, these effortsdid not significantly improve the purity of the FAZ (Fansuri et al.,2008; Mortier, 1978; Nugteren et al., 2001). Thus, efforts are necessaryto optimize the molarity of the NaOH solution and duration of activa-tion. In this context, the hydrothermal treatment of the fly ashwas con-ducted by employing different molarities (M = 0.5 and 1.5) of NaOHsolutions, and a duration of 12 h in one step activation, as establishedby Singh and Kolay (2002). However, the CEC of the end products(i.e., residues) has been reported to be less than 150 meq/100 g,which hints at impurities of up to 70% in the AFA, as compared to theCEC value of 500 meq/100 g, which is exhibited by the pure zeolites(e.g. Na-P1) (Mortier, 1978; Scott et al., 2001).

In such circumstances, the main challenges are to investigate thezeolitization potential of the residues, and to establish a suitable processfor the treatment of their impurity content, resulting in pure form ofpolycrystalline zeolites (Fansuri et al., 2008; Hollman et al., 1999;Querol et al., 2002, 2007). In this context, a process of recycling theresidues, up to three steps, by employing a hydrothermal system at100 °C, is being proposed in this manuscript. In short, this processaims at using the supernatant (i.e., spent NaOH solution as a by-product) for further activation of the residues and improving theircharacteristics (viz., physico-chemical, mineralogical and morpho-logical), conforming to increased yield of zeolites. Also, this study isfocused to ascertain a suitable step of the treatment, which yieldsbetter zeolites. Apart from these efforts, this manuscript finally

2.00

2.25

2.50

2.75

3.00

0

20

40

60

80

100

120

-7.22%(+4.54%)

-1.15%(+9.6%)

+0.76%(+11.76%)

-2.25%(+10.75%)

(+11%)G

Step of treatment

0 1 2 3 0 1 2 3

(a)(+13%)

2.35

OFA

+52.95%(+545.95%)

+33.42%(+595.42%)

+453%(+493%)

+478%(+562%)

(+84%)

(+40%)

SS

A (

m2 /

g)

M 0.5, 1.5

(b)

6.68OFA

Fig. 1.Effects of recycling and step of the treatment on (a)G and (b) SSA of thefly ash and residues,where percentage increase (plus sign) and decrease (minus sign) in the parameter after,Steps-2 and 3, are marked. The data shown within parenthesis represent variations w.r.t. virgin fly ash, OFA.

123B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

attempts to clarify some complexities in the treatment-cum-purificationprocess, as a conceptual model.

2. Experimental investigation

2.1. Materials and process

The virgin or original fly ash (OFA) was procured from a hopper ofelectrostatic precipitator of a thermal power plant in Maharashtra,India. The NaOH used for preparing its stock solution (i.e., 0.5 M to1.5 M) in deionized water, was supplied by Thomas Baker, Mumbai,India. In order to activatemost of the constituents of the ash and synthe-size pure fly ash zeolites, a solution to the fly ash ratio (L/S = 10)(Watek et al., 2008) wasmaintained for preparing a fly ash–NaOH slur-ry (Jha and Singh, 2012). The hydrothermal treatment of this slurry wascarried out in three successive steps, each for 12 h. Step-1 of the treat-ment was followed by a Step-2, as recycling of the residue and thesupernatant, obtained from Step-1. Similarly, the Step-3 was also arecycling of the residue and the supernatant, obtained from Step-2(Jha and Singh, 2011). The many intermediate processes involved ineach step of the treatment are: (i) separation of the residues from thesolution (by a centrifuge working at 1000 rpm for 30 min), (ii) doublewashing (up to pH=10) of the residueswith distilledwater, (iii) drying(by setting oven temperature at 100 ˚C for 12 h) of the residues and final-ly, (iv) a grinding (by hand, using a mortar and pastel). Especially, thegrinding of the residues was intentionally done to peel off all fly ashzeolites, deposited as thin film on the surface of the impurities. In thisway, the inactivated underlying portion of the ash was brought incontactwith the reagent solution, the supernatant. To examine the effectof recycling andmolarity, residues were characterized for their physico–chemico–mineralogical, morphological characteristics and FTIR analysiswas conducted to identify the functional groups present in them. Forthe sake of brevity, the residues have been designated by combiningtwo parameters; (i) the molarity (M) of the NaOH used in Step-1 and

Table 1Chemical composition (% by weight) of the OFA and AFA samples.

Sample Al2O3 SiO2 Fe2O3 Na2O K2O Ba

OFA 26.0 63.8 5.1 0.05 0.66 0.10.5-S1 27.9 57.7 5.4 4.67 0.55 0.10.5-S2 25.5 58.3 5.0 5.53 0.63 0.00.5-S3 33.8 52.5 3.7 5.79 0.61 0.01.5-S1 39.8 41.2 4.63 9.29 0.33 0.01.5-S2 26.4 53.1 5.89 10.2 0.29 0.01.5-S3 25.1 55.4 5.78 9.42 0.30 0.1

(ii) a designation for the step of treatment. For example, the designationssuch as 0.5-S1, 0.5-S2 and 0.5-S3 correspond to the residues obtainedafter the treatments by using 0.5 M NaOH in Step-1, Step-2 and Step-3,respectively.

2.2. Physical characterization

Specific gravity (G) and specific surface area (SSA) of the residueswere determined by employing a Helium gas Ultra Pycnometer(Quantachrome, USA; ASTM D 5550-06) and following the standardethylene glycol monoethyl ether (EGME) method (Arnepalli et al.,2010; Cerato and Lutenegger, 2002; Jha and Singh, 2011), respectively.The results (i.e., average of the three trials) obtained from these investi-gations are depicted in Fig. 1(a, b).

2.3. Chemical characterization

2.3.1. X-ray fluorescence (XRF) studiesAnXRF set up (Phillips 1410, Holland)was used for determination of

the chemical composition (in term of major oxides) of the residues, byfollowing the procedure available in the literature (Jha and Singh,2011, 2012; Singh and Kolay, 2002) and the results are presented inTables 1 and 2.

2.3.2. Determination of the CECThe cation exchange capacity (CEC) of the residues was deter-

mined by following the ammonium acetate method (Adamczykand Bialecka, 2005; ISIRC, 1992; Querol et al., 2007). Accordingly,the sample was initially saturated with Na+ and freed from organicimpurities after being washed thrice with sodium acetate solutionand subsequently with 99% isopropyl alcohol, respectively. Finally,the Na+ was exchanged with NH4

+ after washing the sample thriceby ammonium acetate solution. The results, average of three tests,are presented in Fig. 2.

O CaO MgO MnO2 P2O5 SrO TiO2

4 1.88 0.39 0.16 0.16 0.25 1.522 1.62 0.38 0.05 0.05 0.04 1.489 1.63 0.32 0.04 0.06 0.04 1.518 1.64 0.36 0.04 0.05 0.04 1.289 1.88 0.36 0.05 0.06 0.04 1.589 1.84 0.41 0.05 0.05 0.04 1.560 1.80 0.41 0.05 0.04 0.04 1.49

Table 2Important parameters of the OFA and the residues, AFA samples.

Ratio OFA AFA

0.5-S1 0.5-S2 0.5-S3 1.5-S1 1.5-S2 1.5-S3

SiO2/Al2O3 2.45 2.07 (−16) 2.28 (+10) 1.55 (−32) 1.03 (−58) 2.01 (+95) 2.20 (+9)Na2O/Al2O3 0.002 0.16 (+7900) 0.21 (+31) 0.17 (−19) 0.23 (+11400) 0.38 (+65) 0.37 (−3)

Note: a number under parenthesis represents percentage increase or decrease, as designated by+or− sign, respectively due to its stage in comparison to immediate previous stage of thetreatment.

124 B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

2.4. Mineralogical characterization

2.4.1. X-ray diffraction analysisThe mineralogical composition of the residues was evaluated by

resorting to X-ray diffraction (XRD) studies and by employing a spec-trometer (Phillips 2404, Holland). The sample was scanned for twicethe diffraction angle (2θ) ranging from 0 to 60°, and the diffractogramsare presented in Fig. 3. Theminerals, present in the sample were identi-fied by matching actual d-spacing of an unknown mineral with the dif-fraction data file (ICDD, 1994). Accordingly, designations; Q, ML, H, P, S,F and C, used in Fig. 3, represent the minerals Quartz, Mullite, Hema-tite, zeolite Na-P1, Hydroxysodalite, Faujasite (Na-X) and Chabazite,respectively.

2.4.2. FEG-SEM studiesA field emission gun-scanning electron microscopy (FEG-SEM) set

up (JEOL JSM-7600 F, Japan) was employed to determine the morphol-ogy of the residues and the micrographs are presented in Fig. 4(a to g).

2.4.3. Determination of FTIR absorption bandsTo evaluate transitions in mineral phases of the fly ash and structure

of the bonds of Si4+, Al3+ and OH− formed in the residues, a Fouriertransform infrared (FTIR) set up (Perkin Elmer Paragon 1000 PC) wasemployed. The sample preparation was carried out by resorting toKBr pellet method (Li, 2005). To obtain a spectrum of the infrared ab-sorption bands, the pellet was scanned in the frequency range of4000 to 400 cm−1. The resulting variation in the percentage trans-mittance of the residues with frequency (in terms of wave number)is presented in Fig. 5.

2.4.4. Thermal analysisThermogravimetric analysis (TGA) and differential thermal anal-

ysis (DTA) set up (Diamond TG/DTA, PERKIN ELMER, USA) wasemployed for thermal analysis of the fly ash and the residues (VanReeuwijk, 1972). Around 8 to 10 mg of the sample, contained in anα-Al2O3 crucible, was heated continuously up to 1000°°C, at a rate

0

100

200

300

400

500

+16%(+1282.6%)

+14%(+1536%)

CE

C (

meq

/100

g)

Step of treatment

M 0.5, 1.5

0 1 2 3

(+1337%)

+79.6%(+1266.6%)

+185%(+1522%)

(+1187%)8OFA

Fig. 2. Variation in the CEC of the samples, where percentage increase in the CEC, aftersteps-2 and 3 has been marked, by a plus sign and vice-versa. The data shown withinparenthesis represent the variations w.r.t. virgin fly ash, OFA.

of 10 °C/min (Majchrzak-Kuceba and Nowak, 2004; Nandre et al.,2012; Van Reeuwijk, 1972). During these analyses, a controlled envi-ronment (i.e. an inert gas, N2, flow rate between 0 to 1000 ml/min,under vacuum 10−2 Torr) was maintained for monitoring actualtransitions (viz., mass change, dehydration, oxidation, mineralphase, crystallization and all endothermic and exothermic activities)undergone by the sample. The results of these analyses are presentedin Fig. 6.

3. Results and discussion

From Fig. 1(a, b) it is apparent that Step-1 causes an increase in the Gand SSA of the residues by 13% and 84%, respectively, corresponding to1.5 M NaOH, which can be opined to be a superior reagent than 0.5 MNaOH. These findings are in agreement with those reported by Kolayand Singh (2002) and can be attributed to various effects of ash–alkaliinteraction in Step-1 such as; (i) etching at fewer areas on interactingsurface of the ash and (ii) creation of surface pores, visible in the micro-graphs (refer Fig. 4). These pores facilitate release of entrapped gasesfrom the inner regions of the ash particles, which are responsible forthe higher value of G of the residues (Jha and Singh, 2012; Kolay andSingh, 2002). Conversely, Step-2 causes a minor change in the specificgravity, while a significant increase (by 478%) in the SSA is noticeable.This could be due to significant dissolution of the impurities and precip-itation of the resulting soluble Si, Al andNa as smaller particles in size upto 50 nm, which could get incorporated in the residues of Step-2.Further, Step-3 results in reduction of G and continuous increase ofthe SSA. Incidentally, a decrease in G of the residues may be attributedto: (i) enhanced formation of porous compounds and (ii) polymeriza-tion of Si andAl tetrahedra,which also corresponds to increased volumeof a zeolitic framework structure, present in the residues. Based on var-iations of G and SSA it can be inferred that Steps-2 and 3, are responsiblefor better zeolitic characteristics of the residues (i.e., purified form ofresidues of Step-1).

Also from Table 1 reduction in most of the oxides (viz., SiO2, Fe2O3,BaO, P2O5 and TiO2) after Step-1 can be observed. Especially, reducedquantity of the heavy metal oxides is also responsible for reduced G

0 10 20 30 40 50 60

Rel

ativ

e in

tens

ity

2

OFA

0.5-S10.5-S20.5-S31.5-S1

1.5-S2

1.5-S3

Q

PQ PPQML

SC

QML

H

C

ML ML QQ

F

Fig. 3. Variation in XRD patterns of the samples.

125B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

values of the residues after Steps-2 and 3, corresponding to 1.5MNaOH.In addition, fromTable 2 it can be noticed that residues of Step-1 and 1.5M NaOH exhibit a SiO2/Al2O3 ratio (=58%) lower than that of the orig-inal fly ash, which indicates enhanced dissolution of SiO2 (Quartz). Inci-dentally, a significant increase in Na2O/Al2O3, from 0.002 to 0.23(higher than 0.5 M NaOH) indicates bonding of more Na+, in the res-idues and hence their improved CEC value by 1337% (refer Fig. 2), ascompared to the fly ash. This may be due to significant reduction inXRD peak of Quartz and minor reduction in peak of Mullite, presentas impurities in the zeolite Na-P1 (refer Fig. 3). Based on Fig. 3, a

(b)

(c)

(d)

(a)

25µm

450

80 nm

800 nm

10050

75-100nm

P

P

S

P

P

Scale bar length=10

Scale bar length=100nm

Scale bar length=100nm

Scale bar length=100nm

Fig. 4. FEG-SEM micrographs of all the samples. (a) OFA, (b) 0.5-S

summary of the mineralo-morphological transitions occurring tothe fly ash when a three-step process is adopted has been listed inTable 3.

From this table it can be noted that a highly increased content of thefly ash zeolites (viz., Na-P1 and Faujasite) and absence of Quartz andMullite in the residues 1.5-S2, establish their superiority over otherresidues. The formation of these zeolites also gets verified from Fig. 4,which exhibits several morphologies such as: (i) radiating fibers androd shaped morphology resembling zeolites Na-P1 length rangingfrom 450 to 800 nm (refer Fig. 4b, c) and (ii) ball shaped morphology

(e)

(f)

(g)

2.5µm

30µm

P

Pore

P

Honeycombs

S

830

200

300 nm

C

F

S

PF

C Pore

µm

Scale bar length=100nm

Scale bar length=1µm

Scale bar length=100nm

1, (c) 0.5-S2, (d) 0.5-S3, (e) 1.5-S1, (f) 1.5-S2 and (g) 1.5-S3.

4000 3200 2400 1600 800

1595

.3

3785

.6

3499

.7

1657

.6

451.

4

824.

8

1297

.5

998.

110

97.4

0.5-S3

0.5-S2

Tra

nsm

ittan

ce (

%)

Wave number (cm-1)

OFA

0.5-S1

1.5-S3

1.5-S2

1.5-S1

285.999

62.3

Fig. 5. FTIR spectra of the OFA and the residues in the frequency range 4000–400 cm−1.

126 B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

in diameter ranging from 75 to 100 nm, which resemblesHydroxysodalite (refer Fig. 4d). In addition, Fig. 4(f) exhibits formationof honeycombs (i.e., agglomerates of premature zeolites i.e., under pro-cess of precipitation transforming into sheet shaped Chabazite). These al-terations in the fly ash are commensurate with increasing inter particlepore volume, as noticed in the micrographs (refer to Fig. 4b to g). Basedon these observations, it can be inferred that three-step activation ismore effective than the conventional treatment (i.e., Step-1) of thefly ash.

In addition, Fig. 4(b to d) exhibits remarkable growth inwidth of thezeolitic shapes, which increase from 80 to 100 nm. This could be a resultof an enhanced nucleation of soluble Si4+, Al3+ andNa+ in the superna-tant and agglomeration of their products (i.e., the zeolites) in the resi-dues of Step-2 (refer Fig. 3) from 1.5 M NaOH solution, which causesincrease in Na2O/Al2O3 and SiO2/Al2O ratios by 65% and 95%, respec-tively (refer Table 2).

Most interestingly, it can be noticed from Fig. 2, that, CEC of thesesamples increases (by 1187 to 1337%) tremendously, due to Step-1.Step-2 causes further improvement (by 185%) in the CEC, for 1.5 MNaOH. However, Step-3 results in a minor increase in the CEC. Thus,Step-2 and 1.5 M NaOH can be thought to be a better combination,which results in a larger gain in the CEC of the residues. This can be at-tributed to an augmented formation of a new zeolite (ball shapedHydroxysodalite) along with Na-P1 and hence, improvement in theirpurity. In agreement to Grutzeck and Siemer (1997), this can be attrib-uted to increased linkage of cations (Na+), and hence more CEC by105.4% (refer Fig. 2). Conversely, Step-3 causes significant decrease inNa2O/Al2O3 and SiO2/Al2O in case of residues from 0.5 M NaOH. Thiscan be ascribed to conversion of Na-P1 to Hydroxysodalite fromStep-2 to Step-3 (refer Figs. 3 and 4), and this conforms to less in-crease in the CEC (refer Fig. 2).

Furthermore, Fig. 4 exhibits significant transitions in shape and sizeof the crystalline constituents from the OFA to the residues of Steps-1, 2and 3 of the treatments. In addition, the particle size (=2.5 to 30 μm) ofthe minerals in the fly ash reduces to 50 to 800 nm in the residues, de-pending on the type of alkali solution and the step of the treatment.From the micrographs it can be inferred that the shape of the newlyformed crystals in the residues vary from rod shaped (marked as P), inStep-1, to spheroidal balls (marked as S) to honeycombs (C) in therecycling and Step-3 of the treatment.

The mineralogical phase transition in the residues of the three steps(refer Table 3) gets verified from the FTIR spectra and TG–DTA curvespresented in Figs. 5 and 6, respectively. The disappearance of a weakband (i.e. Si–OH, silanol) at high frequency (3785.6 cm−1) from theFTIR spectrum of the fly ash (refer Fig. 5), and simultaneously appear-ance of broader and deeper bands (i.e. Si\OH\Al) at low frequency(3499 cm−1) in the residues confirms significant conversion of Si andAl tetrahedra from Quartz and Mullite to a new product, which com-prises of new structural bonds (Li and Wu, 2003; Sun, 1993, 1994).Such development favors mutual coordination between Si and Al tetra-hedra by a bridging hydroxyl functional group in the residues. This find-ing is in agreement with that proposed by the previous researchers(Criado et al., 2007; Rayalu et al., 2005). However, a sharper band at1657.6 cm−1 confirms relatively more adsorbed water in the residues1.5-S3 (as compared to its predecessors 1.5-S2 and 1.5-S1, and the res-idues resulting from 0.5 M NaOH treatments), which could be possiblebecause of higher surface negativity and increased presence of AlO4

5−

tetrahedralwhich results in higher zeolitic contents. This again gets sub-stantiated from another sharper and deeper band at 998.1 cm−1, whichcorresponds to higher crystallinity of the zeolitic framework (Si\O\Albonds) in these residues 1.5-S3. In general, the intensity of these bandsincreases at the shifted low frequency in the spectra of Steps-2 and 3.These findings are indicative of increased stretching vibration ofSi\O\Al bonds, which reveal higher yield of fly ash zeolites in the res-idues of subsequent steps of treatment (Coates, 2000; Criado et al.,2007; Jacob and Mortier, 1982; Li, 2005; Li and Wu, 2003; Rayaluet al., 2005; Sun, 1993, 1994). The subsequent steps are responsible

for major structural alterations in the residues and growth of moredistinct morphology of the crystals, which resemble with the commonfly ash zeolites, as marked in the micrographs (refer Fig. 4).

Apart from these alterations, no shift in bands and their increasedintensity from 824.8 to 451.4 cm−1, can be attributed to bending vibra-tions, which could be due to improved porosity of the residues, (Jacoband Mortier, 1982). This gets confirmed by increased weight loss(refer Fig. 6c) and a distinct endothermic peak (refer Fig. 6d), whichhint at (i) an increased presence of surface and structural water andhence (ii) sharper band at 1657.6 cm−1 (refer Fig. 5) in the zeolitic res-idues 1.5-S3, as compared to their counterpart 0.5-S3, which undergoessudden drop in weight. This could be attributed to the presence ofcarbonaceous impurities in the residues and their conversion to gaseslike CO2, which is not the case with the residues corresponding to 1.5M NaOH. In addition, the lowering of DTA curves of residues 1.5-S3 isalso an indication of their less heat flow capacity and less mineralphase transition. Incidentally, the residues 1.5-S3 exhibit many TGApeaks beyond 1000 °C (refer Fig. 6c), which can be attributed to thethermal conversion of some minerals to gaseous products entrappedin the zeolitic pores. However a flatter DTA curve reveals (i) negligiblevariation in mineralogy of the residues, (ii) increased thermal stabilityof the residues obtained from Step-3 as compared to Steps-2 and 3and hence (iii) the creation of the most purified form of the residuesobtained from the Step-3 treatment with 1.5 M NaOH.

3.1. A conceptual model for step-wise purification

Furthermore, improvement in various characteristics of the flyash, as discussed above has been employed to develop a model forpredicting mineral transition and physical transformation occurringto it, due to the three-step process. In line with the findings of previ-ous researchers (Brouwers and Vaneijk, 2002; Inada et al., 2005), amodel for the fly ash particle and its three-steps of activation has beenconceptualized and depicted in Figs. 7 and 8, respectively. Inada et al.(2005) have reported that the outer region, of the OFA particle (referFig. 7) contains glass, which interacts first with the alkali duringStep-1 of the treatment and it gets etched slowly (Jha and Singh,2012). Subsequently, the underlying Quartz interacts with the superna-tant during Step-2. As reported by Brouwers and Vaneijk (2002), the in-nermost Mullite gets exposed and activated completely due to theStep-3 treatment. Hence, it can be opined that such activation of thefly ash particle, results in a remarkable increase in the CEC (referFig. 2). This highlights tremendous reduction in the impurities (IMP)in the residues (refer Figs. 5 and 6), and hence high yield of the fly ashzeolites (refer to Figs. 3 and 4). Fig. 8 intends to simplify the complexeffects of successive activations of the fly ash, in three-steps. In such

85

90

95

100

Wei

ght (

%)

0.5-S1

0.5-S3

0.5-S2

(a)

85

90

95

100

Wei

ght (

%)

1.5-S2

1.5-S1

1.5-S3

(c)

0 200 400 600 800 1000

0

10

20

30

40

0.5-S1

0.5-S2

0.5-S3

(b)

0 200 400 600 800 1000

0

10

20

30

40

1.5-S3

1.5-S21.5-S1

Endothermic

(d)

Exothermic

0 200 400 600 800 1000

96

98

100

-30

-20

-10

End

o do

wn

( V

)

Wei

ght (

%)

Temp. (°C)

TGA

DTA

(e)

Temp. (°C) Temp. (°C)

0 200 400 600 800 1000 0 200 400 600 800 1000

Temp. (°C) Temp. (°C)

End

o do

wn

( V

)

End

o do

wn

( V

)

Fig. 6. TG–DTA analysis of residues of three steps of hydrothermal treatmentwith (a, b) 0.5MNaOH, (c, d) 1.5MNaOH and (e) Fly ash, where S1, S2 and S3 are designations for Steps-1, 2and 3, respectively.

127B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

situation, grinding of the residues, may cause peeling off and crumblingof the deposited crystals of the FAZ (refer Fig. 4), which thus exposesthe roughened boundary of the IMP from Step-1 and Step-2. In fact,the activated boundary contains many nano-sized pores, as observedin Fig. 4(b to g), which could facilitate increased surface contact (referto Fig. 1b) and also an improved interaction through a pore, up to theinner region of the OFA particle, with the supernatant. Accordingly, res-idues of Steps-2 and 3 can be assumed to contain a number of thermallystable shapes (shown by white color), which are commensuratewith the conversion of primarily synthesized rod shaped zeolite

Table 3Minerals present in the OFA and AFA samples.

Mineral OFA 0.5-S1 0.5-S2 0.5-S3 1.5-S1 1.5-S2 1.5-S3

Quartz +++ ++ – – ++ – –

Mullite +++ ++ + + ++ + –

Na-P1 – + +++ ++ ++ +++ ++Na-X (Faujasite) – – + ++ + +++ ++Chabazite – – – – – + +Hydroxysodalite – – + + + +

Note:+++,++,+,− represent very high, high and low content, respectively, whereas,dash represents not identifiable.

Na-P1 to polycrystalline shapes (viz., balls in 75–100 nm diameter ofHydroxysodalite, thickened rods and honeycombs of Na-P1) of FAZ(refer Fig. 4). Fig. 8 demonstrates that polycrystalline fly ash zeolitesget crystallized in a supernatant due to the solubility of unstable zeolitesin an alkaline environment, which have low SiO2/Al2O3 ratio, after Step-1(refer Table 2); this is in agreement with the observations reported byWilkin and Barnes (1998). A significant increase in this ratio, due to

Q

Outer core

GS Surface layer

Inner core

ML

Smooth boundary

Fig. 7. 3D model of the OFA particle.

Fig. 8. A conceptual model of the process to explain stepwise purification of the FAZ, where black dots, hexagons, oval, inner most circle andwhite shapes represent glass, Quartz, Mullite,hollow core of the fly ash particle and fly ash zeolites, respectively.

128 B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

Step-2, confirms the increased stability and high CEC of the residues, aswell (refer Fig. 2 and Table 2). From Figs. 2, 5 and 6 the residues ofStep-3 can be assumed to be enriched in more thermally stable form ofzeolites FAZ at a temperature lower than 600 °C.Mineral phase transitionbeyond this temperature in the TG–DTA curves can be attributed to con-version of Quartz and Mullite impurities to metastable zeolites.

4. Conclusions

The findings of the study are remarkably encouraging and it hasbeen observed that an initial increase in specific gravity of the residuesis followed by its decrease due to recycling steps in the adopted process,which are indicative of a significant yield of more porous and puremorphologies of fly ash zeolites. In addition, a continuous increase inboth the SSA andCEC of the residueswith the recycling steps of this pro-cess confirms the transition of the zeolite Na-P1 to pure residues, whichcomprises of a number of polycrystalline zeolites and inter particlepores. The recycling Steps-2 and 3 result in disappearance of the XRDpeaks, corresponding to major constituents of the fly ash and growingup of the peaks of zeolites (viz., Na-P1, Hydroxysodalite, Chabaziteand Faujasite). The second recycling treatmentwith 1.5MNaOH resultsin highly improved characteristics of the residues, as compared to 0.5MNaOH. Hence, superiority of the adopted process over a one steptreatment of the fly ash for synthesizing pure fly ash zeolites getsestablished.

References

Adamczyk, Z., Bialecka, B., 2005. Hydrothermal synthesis of zeolites from polish coal flyash. Pol. J. Environ. Stud. 14 (6), 713–719.

Arnepalli, D.N., Hanumantharao, B., Shanthakumar, S., Singh, D.N., 2010. Determination ofdistribution coefficient of geomaterials and immobilizing agents. Can. Geotech. J. 47(10), 1139–1148.

ASTMD-5550–06, Standard test: method for specific gravity of soil solids by helium gas pyc-nometer. 100 Barr Harbor Drive, P.O. Box C700, West Conshohocken, PA 19428–2959,USA.

Brouwers, H.J.H., Vaneijk, R.J., 2002. Fly ash reactivity: extension and application of a shrink-ing core model and thermodynamic approach. J. Mater. Sci. 37, 2129–2131.

Cerato, A.B., Lutenegger, A.J., 2002. Determination of surface area of fine-grained soils bythe ethylene glycol monoethyl ether (EGME) method. Geotech. Test. J. 25 (3),314–320.

Coates, J., 2000. Interpretation of infrared spectra, a practical approach. In: Meyers, R.A.(Ed.), Encyclopedia of Analytical Chemistry. John Wiley & Sons Ltd, Chichester,pp. 10815–10837.

Criado, M., Fernandez-Jimenez, A., Palomo, A., 2007. Alkali activation of fly ash: effectof the SiO2/Na2O ratio: part I: FTIR study. Microporous Mesoporous Mater. 10,180–191.

Fansuri, H., Pritchar, D., Zhang, D., 2008. Manufacture of Low-Grade Zeolites from Fly Ashfor Fertilizer Applications. QCAT Technology Transfer Centre, Technology Court,Pullenvale Qld 4069, Australia.

Grutzeck, M.W., Siemer, D.D., 1997. Zeolites synthesized from class F fly ash and sodiumaluminate slurry. J. Am. Ceram. Soc. 89 (9), 2449–2453.

Hollman, G.G., Steenbruggen, G., Janssen-Jurkovicov, M., 1999. A two-step process for thesynthesis of zeolites from coal fly ash. Fuel 78, 1225–1230.

ICDD, International Centre for Diffraction Data, 12 Campus Boulevard, Newtown Square,PA 19073–3273, 1994, U.S.A.

Inada, M., Eguchi, Y., Enomoto, N., Hojo, J., 2005. Synthesis of zeolite from coal fly asheswith different silica–alumina composition. Fuel 84, 299–304.

ISIRC, 1992. World Soil Information Data Base. Wageningen UR.Jacob, P.A., Mortier, W.J., 1982. An attempt to rationalize stretching frequencies of lattice

hydroxyl groups in hydrogen zeolites. Zeolites 2, 226–230.Jha, B., Singh, D.N., 2011. A review on synthesis, characterization and industrial applica-

tion of fly ash zeolites. J. Mater. Educ. 33 (1–2), 65–132.Jha, B., Singh, D.N., 2012. Zeolitization characteristics of fly ashes from wet- and dry-

disposal systems. Acta Geotech. Slov. 9 (2), 63–71.Kim, W., Jung, S., Ahn, B.J., 1997. Synthesis of Na-P1 zeolite from coal fly ash. J. Ind. Eng.

Chem. 3 (3), 185–190.Kolay, P.K., Singh, D.N., 2001a. Effect of zeolitization on physico–chemico–mineralogical

and geotechnical properties of the lagoon ash. Can. Geotech. J. 38 (5), 1105–1112.Kolay, P.K., Singh, D.N., 2001b. Effect of zeolitization on compaction, consolidation and

permeation characteristics of a lagoon ash. J. Test. Eval. 28 (6), 425–430.Kolay, P.K., Singh, D.N., 2001c. Physical, chemical, mineralogical and thermal properties of

cenospheres from an ash lagoon. Cem. Concr. Res. 31 (4), 539–542.Kolay, P.K., Singh, D.N., 2002. Characterization of alkali activated lagoon ash and its appli-

cation for heavy metal retention. Fuel 8, 483–489.Kolay, P.K., Singh, D.N., Murti, M.V.R., 2001. Synthesis of zeolites from lagoon ash. Fuel 80

(5), 739–745.Li, G., 2005. FTIR Studies of Zeolitic Material: Characterization and Environmental Appli-

cations. Thesis and Dissertation University of Iowa.Li, C., Wu, Z., 2003. Microporous materials characterized by vibrational spectroscopies.

Handbook of Zeolite Science and Technology. Marcel Dekker, Inc.Majchrzak-Kuceba, I., Nowak, W., 2004. Application of model-free kinetics to the study of

dehydration of fly ash-based zeolite. Thermochim. Acta 413, 23–29.Mortier, W.J., 1978. Zeolite electro-negativity related to physicochemical properties.

J. Catal. 55, 138–145.

129B. Jha, D.N. Singh / Applied Clay Science 90 (2014) 122–129

Murayama, N., Yamamoto, H., Shibata, J., 2002. Mechanism of zeolite synthesisfrom coal fly ash by alkali hydrothermal reaction. Int. J. Miner. Process. 64,1–17.

Nandre, S.J., Shitole, S.J., Ahire, R.R., 2012. Thermal and optical studies of gel grown cobaltiodate crystal. Arch. Phy. Res., 3(1), pp. 70–77.

Nugteren, H.W., Moreno, N., Sebastia, E., Querol, X., 2001. Determination of the availableSi and Al from coal fly ashes under alkaline conditions with the aim of synthesizingzeolites products. International Ash Utilization Symposium, Centre for Applied Ener-gy Research, University of Kentuchy, Paper No. 71.

Querol, X., Moreno, N., Uman, Juan, R., Hernandez, S., Fernandez, P.C., Ayora, C., Janssen,M., Garcia, J.M.J, Linares, S.A., Cazorla, A.D., 2002. Application of zeolitic material syn-thesized from fly ash to the decontamination of waste water and flue gas. J. Chem.Technol. Biotechnol. 77, 292–298.

Querol, X., Moreno, N., Alastuey, Juan, R., Andres, J.M., Lopez-Soler, A., Ayora, C., Medinaceli,A., Volero, A., 2007. Synthesis of high ion exchange zeolites from coal fly ash. Geol. Acta5 (1), 49–57.

Rayalu, S.S., Meshram, S.U., Hasan, M.Z., 2000. Highly crystalline faujasite zeolites from flyash. J. Hazard. Mater. B77, 123–131.

Rayalu, S.S., Udhoji, J.S., Meshram, S.U., Naidu, R.R., Devodatta, S., 2005. Estimation of crys-tallinity in fly ash-based zeolite-A using XRD and IR spectroscopy. Res. Commun.,Current Sci., 89(12), pp. 2147–2151.

Scott, J., Guang, D., Naeramitmarnsuk, Thabuot, M., Amal, R., 2001. Zeolite synthesis fromcoal fly ash for the removal of lead ions from aqueous solution. J. Chem. Technol.Biotechnol. 77, 63–69.

Singh, D.N., Kolay, P.K., 2002. Simulation of ash water interaction and its influence on ashcharacteristics. Prog. Energy Combust. Sci. 28, 267–299.

Sun, D.M., 1993. An analysis of the structure of 13× molecular sieve in ion exchange.Vacuum 44 (2), 75–78.

Sun, D.M., 1994. Dependence of X zeolite adsorption properties on electro-negativity andvibration frequency. Vacuum 45 (1–2), 1175–1179.

Van Reeuwijk, L.P., 1972. High-temperature phases of zeolites of the natrolite group. Am.Mineral. 57, 499–510.

Watek, T.T., Saito, F., Zhang, O., 2008. The effect of low solid/liquid ratio on hydrothermalsynthesis of zeolites from fly ash. Fuel 87, 3194–3199.

Wilkin, R.T., Barnes, H.L., 1998. Solubility and stability of zeolites in aqueous solutions: I.analcime, Na-, and K-clinoptilolite. Am. Mineral. 83, 746–761.