Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

G Model

JIEC-1399; No. of Pages 6

Effect of calcination temperature on colorant behavior ofcobalt-aluminate nano-particles synthesized by combustion technique

Shiva Salem *

Chemical Engineering Group, Urmia University of Technology, Urmia, Iran

A R T I C L E I N F O

Article history:

Received 20 January 2013

Accepted 7 June 2013

Available online xxx

Keywords:

Cobalt-aluminate

Spinel

Combustion

Nano-particle

Calcination

A B S T R A C T

In this investigation, the normal nano-crystalline cobalt-aluminate spinel has been successfully

synthesized by the combustion technique. In order to study the colorant behavior of powders after heat

treatment, quantitative and qualitative experiments such as color spectroscopy, X-ray and Raman

spectroscopy were applied. Transmission electron microscopy technique was used to estimate the

particle size and observe the morphology of pigments. The green powder was identified as an inverse

spinel structure whereas a normal spinel corresponding to blue color was produced at higher

temperatures. For obtaining powder with the high colorant efficiency, it is better to carry out calcination

at 1000 8C.

� 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights

reserved.

Contents lists available at SciVerse ScienceDirect

Journal of Industrial and Engineering Chemistry

jou r n al h o mep ag e: w ww .e lsev ier . co m / loc ate / j iec

1. Introduction

Recently, the spinels containing transition metals have been ofgreat importance due to their potential in electrochemical,catalytic and pigment applications [1–4]. Spinels are generallyformulated as A2+B2

3+O4 in which the crystalline structure isclassified as a cubic, isometric, having the anions arranged in acubic close-packed lattice. Also, the A and B cations occupy some orall of the octahedral and tetrahedral sites in the lattice [5]. Thereare two ideal structures: normal spinel structure and inverse spinelstructure. Normal spinel structure is usually identified by cubicclose-packed oxides with one octahedral and two tetrahedral sitesfor each anion. The tetrahedral interstices are smaller than theoctahedrals therefore, the B3+ ions occupy half of the octahedralholes whereas A2+ ions occupy 1/8th of the tetrahedral holesbecause of a charge factor. The lattice energy is maximized whenthe diameter of ions are similar [6,7]. On the other hand, an inversespinel structure is slightly different due to the presence oftransition metals in the crystal field. If the A2+ ions have a strongpreference in the octahedral site, they displace half of the B3+ ionsfrom the octahedral sites to the tetrahedral ones. If the B3+ ionshave a low or zero octahedral site stabilization energy, OSSE, thenthey have no preference and adopt the tetrahedral site [8].

One of the most frequently used spinels is the cobalt-aluminatesystem, CoIICoIII

xAl2�xO4, in which x value is between zero and 2.

* Tel.: +98 4413554352; fax: +98 4413554184.

E-mail addresses: s.salem@che.uut.ac.ir, salem@iust.ac.ir

Please cite this article in press as: S. Salem, J. Ind. Eng. Chem. (2013

1226-086X/$ – see front matter � 2013 The Korean Society of Industrial and Engineer

http://dx.doi.org/10.1016/j.jiec.2013.06.011

The chemical analysis of cobalt-aluminate spinels are more likeCo3O4, Co2AlO4 and CoAl2O4 [3,9]. The coordination of Co2+ is oneof the most important factors, affecting the colorant performanceof cobalt pigment [10,11].

CoAl2O4 is the best known member of cobalt spinels and iswidely used in ceramic, glass, paint industry and color televisiontubes as a contrast-enhancing luminescent pigment to produceThenard’s blue color [12]. CoAl2O4 is opaque and presents goodhiding power in the form of micron-sized pigment whilst the mostattractive feature of the nano-sized CoAl2O4 is transparent [8,13].CoAl2O4 is classified into the category of normal spinels in whichaluminum and cobalt are located in octahedral and tetrahedralsites, respectively [14,15].

The most common method for manufacturing cobalt spinels isthrough a solid-state reaction in which the oxides are mechanicallyground at high calcination temperatures about 1300 8C for a longtime. Though the mentioned process is relatively inexpensive,undesired products can be produced due to a lack of homogeneity,larger and uneven grains due to poor control of stoichiometry[11,16]. Recently, spinels have been synthesized using several wet-chemical techniques such as sol–gel [17–19], emulsion precipita-tion [20], hydrothermal crystallization [16,21] and coprecipitation[22,23]. Among the available solution chemistry routes, thecombustion technique is an inexpensive method that allows thepreparation of highly purified, nano-sized crystalline powders atlower calcination temperatures in a significantly shorter time [24–26].

Extensive literature is available about production and evalua-tion of nano-sized CoAl2O4. Salem et al. concluded that the

), http://dx.doi.org/10.1016/j.jiec.2013.06.011

ing Chemistry. Published by Elsevier B.V. All rights reserved.

S. Salem / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx2

G Model

JIEC-1399; No. of Pages 6

microwave-hydrothermal synthesis technique is not suitabletechnique for production of CoAl2O4 spinel via a reaction in amicrowave digestion system with a limited residence time,controlled temperature and pH, while a very pure nano-crystallinespinel CoAl2O4 is produced by combustion synthesis via a reactionof metal nitrates and glycine [27]. The authors also showed that thepH of the synthesis environment strongly affects the colorantbehavior [28]. The synthesis of a pigment at acidic condition showsa flaming combustion type that is not preferred from the industrialpoint of view. The effect of nano-sized cobalt-aluminate powder onthermal behavior of a typical frit has been investigated by hot stagemicroscopy [29]. The results indicate that the thermal behavior isstrongly influenced by calcination temperature of the cobalt-aluminate powder. However, the inverse variation, expansion, isobserved on shrinkage between 900 and 1030 8C if the powdercalcined at 800 8C is added into the frit. The experimental datashowed that the addition of powders synthesized at a low fuel ratioand calcined at 800 8C leads to an expansion between 940 and1040 8C and beyond this range, the shrinkage decreases, normally[30]. The fuel ratio is efficient in the thermal behavior of modifiedglaze. When a fuel ratio of 0.75 is used the expansion regiondisappears with the exception of the glaze containing the powdersynthesized in an acidic environment.

Few publications exist about evaluation of colorant behavior ofpowder from green to blue as a function of calcination temperatureto achieve an optimum condition in the heat treatment of cobaltspinel. The aim of present investigation is to synthesize nano-sizedCoAl2O4 using a solution-based combustion method to evaluatethe effect of the successive calcination step on powder properties.Laboratory investigation was carried out to determine the colorchanges by evaluating some properties of green and blue powders.This article reports the results of the study on colorant behavior ofcobalt-aluminate spinel by analyzing the carbon content, crystal-lite size, amount of crystalline phase, structure and morphology ofparticles.

2. Materials and methods

2.1. Sample preparation

Analytical grade Co(NO3)2�6H2O, Al(NO3)3�9H2O, both fromSigma–Adrich Company with minimum purity of 99 wt.% asreceived, and glycine, NH2CH2COOH from Sigma Adrich Company,99.5 wt.%, were used as starting materials. A mixed solution ofmetal nitrates was prepared by dissolving above salts in de-ionizedwater with a Co/Al molar ratio of 0.50. The appropriate amount ofglycine was added to adjust the glycine/NO3 molar ratio at 0.56.Subsequently, ammonia was added to control the level of pHaround 7 [28].

In order to obtain a viscous gel, the mixed solution was heatedon a hot plate kept at a steady temperature of 110 8C. The viscousprecursor was then heated rapidly in a preheated furnace kept at500 8C. During the smoldering type combustion, which lasted formore than 7 min, the material underwent foaming followed bydecomposition, generating a large volume of gases. The obtainedvoluminous and foamy combustion ashes were easily crushed toproduce the powders. The crushed powders were further calcinedat 600, 800 and 1000 8C for 1 h to study their thermal behaviors.The powders calcined at 600, 800 and 1000 8C had green, dark andbright blue colors, respectively.

2.2. Sample characterization

In order to obtain a* and b* values by the CIE Lab method, thecolor measurements were conducted using UV–vis spectroscopy(Model Lambda 19, Perkin Elmer, USA) [31]. The diffuse reflectance

Please cite this article in press as: S. Salem, J. Ind. Eng. Chem. (2013

spectra were used to calculate Kubelka–Munk absorption function,F(R), as a follows:

FðRÞ ¼ K

S¼ ð1 � RÞ2

2R(1)

where R is the fractional reflectance, K is the absorption coefficientand S is the scattering coefficient at each wavelength of light in thevisible region, 400–700 nm, [32].

In order to better understand the colorant behavior of synthesizedpigments, the residual carbon content of the powders, wasdetermined by elemental analyses (Carlo Erba, Model EA 1110, Italy).

The specific surface area of the powders was measured by theBrunauer, Emmett and Teller, BET, technique (Gemini 2360Apparatus, Micromeritics, Norcross, GA, USA) after degassingunder vacuum at 150 8C.

X-ray diffraction measurements, XRD, were carried out on thecalcined powders using a conventional Bragg-Brentano diffrac-tometer (X’PERT PRO, Philips Research Laboratories) with Ni-filtered CuKa radiation. The patterns were recorded in the 10–8082u range at room temperature, with a scanning rate of0.001 8s�1 and a step size of 0.028. The X-ray diffraction techniquewas used not only to identify the phases in the powders but also toevaluate the average crystal size by using Sherrer equation [33].

The quantitative X-ray analyses of the synthesized powderswere determined by the Rietveld-RIR (Reference Intensity Ratio)method that allows both the crystalline and the amorphousfractions to be determined in a polyphase mixture. In fact, thescattering contribution of any amorphous component eventuallyexisting in the sample is a part of the background and allows itsquantification using an internal standard, suitably chosen, which isadded in a known amount to the investigated mixture and treatedas a mixture component itself. At the end of the Rietveld process,the refined phase fractions are converted into weight fractions andrescaled to the values of the original mixture by the ratio betweenthe refined and the known amount of added standard. Whenever,an amorphous phase exists in the system, the values of the weightfractions are overestimated to satisfy the normalization condition.The percentage of the amorphous phase in the original mixture canbe calculated directly from the weight of the internal standard. Forall the samples, the powder diluted with 10 wt.% corundum (NISTSRM 674a) as internal standard, was side loaded into an aluminumflat holder to minimize preferred orientation. Data were recordedin the 5–14082u range (step size 0.028 and 6 s counting time foreach step). The phase fractions extracted by the Rietveld-RIRrefinements, using GSAS and EXPGUI [34,35], were rescaled on thebasis of the absolute weight of corundum originally added to themixtures as an internal standard, and therefore internallyrenormalized. All the agreement indices (Rwp < 2% andRp < 1.5%) and the additional statistical indicators supplied byGSAS (x2 < 1.2) are indicative of the very good quality of therefinements and testify the accuracy of the estimated weights.

The morphology of the resulted powder was examined with ascanning electron microscope (SEM, Model XL-30, FEI). Finally,transmission electron microscopy (TEM, JEM 2010, JEOL, Tokyo,Japan) equipped with EDS (EDAX PV9900, Philips) was used incharacterizing the particles. For this purpose, the samples wereprepared by dispersing the powders in distilled water and thenplacing a drop of suspension on a copper grid with a transparentpolymer followed by drying in laboratory oven.

Differential thermal analyses (DTA, Model 409, Netzsch,Germany) were performed on the powders calcined at 600, 800and 1000 8C in order to further investigate the effect of subsequentcalcination temperature using ceramic pans as sample holders.Approximately, 10 mg of each powder was subjected to a thermaltreatment from 20 to 1200 8C at a heating rate of 10 8C min�1.

), http://dx.doi.org/10.1016/j.jiec.2013.06.011

Fig. 1. The Kubelka–Munk absorption function, F(R), of powders calcined at 600, 800

and 1000 8C.

S. Salem / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx 3

G Model

JIEC-1399; No. of Pages 6

The structures of the resulting powders after drying wereevaluated by Fourier transform infrared (FT-IR) spectroscopy(Avatar 330, Thermo Nicolet, Germany).

The nature of chemical functional groups of the synthesizedpowders was determined by Raman spectroscopy (Horiba Jobin-Yvon, Longjumeau, France) method. The Raman spectra wererecorded using the exciting wavelength, 632.81 nm, provided byhelium and neon. The laser power was set to 20 mW.

3. Results and discussion

The CIE Lab colorimetric coordinates allowed the characteriza-tion of the pigment color. Table 1 reports the values of a* and b* forthe synthesized powders. As reported, the powder calcined at600 8C presents green color whereas the powders calcined athigher temperatures show dark and light blue colors, respectively.It can be noticed that the increment in the calcination temperaturestrongly decreases b* value and slightly increases the amount of a*.Moreover, the increment in temperature causes a lighting of thecolor. This variation mainly consists a decrease in b* as reported inTable 1, reaching a minimum value when the calcinationtemperature increases up to 1000 8C. The powder calcined atlow temperature showed a different colorant behavior aspresented in data of Table 1. The colorant behavior of the powdercalcined at 600 8C is remarkably different compared to that in othercases. The powder calcined at 1000 8C has higher a* and loweramount of b*. The whiteness of powders i.e. L* approximatelyremains constant and does not change with increment incalcination temperature. On the other hand, the amounts ofcarbon and nitrogen after calcination at different temperatures arevery low, indicating that the colorant differences between thepowders is not related to the carbon or nitrogen content.

Fig. 1 shows Kubelka–Munk absorption function of the powderscalcined at 600, 800 and 1000 8C. The F(R) curves of the powderscalcined at 800 and 1000 8C indicate three absorption peaks in thevisible region around 537, 580 and 633 nm due to spin allowedtransitions of Co2+ ions in tetrahedral coordination which gives riseto blue color, whereas the mentioned peaks disappeared in thepattern of the green powder calcined at 600 8C [13,36].

The results of the specific surface area and measurements of thecrystal size are presented in Table 2. As can be seen, the crystals ofthe powder calcined at 1000 8C present the lowest specific surfacearea among the three samples. During this stage the smallestcrystals disappear in favor of the larger ones, which grow further.This phenomenon is due to crystal growth and the variation incrystallite size will be discussed by TEM micrographs. As aconsequence the total number of crystals is reduced. Thisphenomenon results in a decrease of the specific surface area.The improved crystallinity of the powder is reflected in the lowerspecific surface area. The calcination temperature has a majoreffect on the specific surface area. Temperature is an importantfactor which affects the rate of crystal growth. Calcination of thepowders at 600 8C can not promote the crystal growth. Corre-sponding to data reported in Table 2, green powder calcined at600 8C presents the highest specific surface area whereas bluepowder shows the lowest specific surface area.

Table 1The colorant behavior of powders as a function of calcination temperature, carbon

and nitrogen percentage.

Temperature (8C) L* a* b* C (wt.%) N (wt.%)

600 39.4 �3.45 �1.5 0.53

800 41.0 �1.64 �7.3 0.40 0.14

1000 41.4 �1.09 �10.1 0.30

Please cite this article in press as: S. Salem, J. Ind. Eng. Chem. (2013

The XRD patterns of studied samples are presented in Fig. 2. TheX-ray diffraction analysis shows the crystal structure of allsamples. However, the degree of crystallinity is different. Thereis a clear difference between the samples calcined at 600 8C andthose calcined at 800 and 1000 8C. This figure shows that there isno clear difference in the peak position between the XRD patternsof green and blue powders. It also indicates that higher calcinationtemperatures promote crystallinity, considerably. In the powdercalcined at 1000 8C, the sharp peaks were attributed to the increasein the crystallinity of powder. Moreover, the shift of the diffractionpeaks was not observed in the samples when the powders werecalcined at high temperatures.

Crystallite size, D, was estimated from X-ray line broadeningmeasurements according to the Scherrer formula.

D ¼ klbcosu

(2)

where b is the breadth of the observed diffraction line at its half-intensity maximum, k is the so-called shape factor, which usuallytakes a value of about 0.9, u is the diffraction angle, in radian, and lis the wave length of the X-ray source used in XRD [33]. From XRDdata, the crystallite size was estimated, Table 2. Based on Scherer’sformula crystallite size can be calculated from the XRD peaks [37].The calcination temperature affects predominantly the crystallitesize and, as a consequence, specific surface area. From thecrystallization point of view, the calcined temperature is theinfluential parameter. With increasing temperature, the crystallitesize increases while the specific surface area decreases, signifi-cantly. Therefore, the results obtained by average crystallite sizeare in agreement with the results obtained by specific surface areameasurements.

In order to investigate the effect of heat treatment oncrystallization and phase transformation in the powder preparedby combustion technique, the Rietveld-RIR experiment was used.

Table 2The specific surface area and average crystal size of calcined powders as function of

temperature.

Temperature (8C) Specific surface

area (m2 g�1)

Average crystal

size (nm)

Color

600 80 18 Green

800 43 23 Dark blue

1000 13 62 Bright blue

), http://dx.doi.org/10.1016/j.jiec.2013.06.011

Fig. 2. The X-ray diffraction patterns of calcined powders.

Table 3Proportion of crystalline and amorphous phase of samples measured by the

Rietveld-RIR method.

Temperature (8C) Crystalline

phase (wt.%)

Amorphous

phase (wt.%)

x2 Rwp Rp

Without calcination 74.6 25.4 1.183 0.0188 0.0149

600 70.8 29.2 1.102 0.0186 0.0148

800 88.0 12.0 1.097 0.0182 0.0145

1000 98.2 1.8 1.162 0.0181 0.0144

Fig. 3. The SEM images of powders calcined at (a) 600, (b) 800 and (c) 1000 8C.

S. Salem / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx4

G Model

JIEC-1399; No. of Pages 6

In Table 3, the refined weights of the crystalline phases and relativeestimated amorphous fraction were reported for each powder usedfor the colorant tests as a function of temperature. Also thecrystalline phase of the powder prepared without calcination isincluded in Table 3. Noteworthy, the crystalline phase increasedwhen the samples were heated from 600, 800, up to 1000 8C,indicating further growth of the nanocrystals inside the powderparticles. At room temperature and 600 8C, it contains minimumcrystalline phase. The calcination of powder results in a highdegree of crystallization. Even, the heat treatment at 800 8Cproduces nearly 15 wt.% more crystallite phase than the startingpowder prepared without calcination. The amorphous phasefraction detected in the powders nonlinearly decreases withtemperature. This point tends to change the colorant behavior ofsynthesized powder with calcination temperature. The fraction ofcrystallite phase for the powder heated at 1000 8C is about 98 wt.%,the highest value found. There is a relation between the colorantbehavior of calcined powder and crystallite phase content, whichchanges the color of the pigment. The relation depends oncrystallite phase content and crystal size, consequently. Thefraction of crystallite phase in starting sample and the powdercalcined at 600 8C is about 75 and 71 wt.%, respectively. These arelower than the content of crystallite phase for the samples calcinedat 1000 8C. For this powder the crystallization was fully accom-plished.

The microstructure of the powders prepared by the presentprocess, after calcined at 600, 800 and 1000 8C for 1 h, wasexamined by SEM, as shown in Fig. 3. As it can be seen from themorphology of particles, there is a distribution of small particlesand large agglomerates. These agglomerates consist of very fineparticles that are cold welded together. TEM micrographs of thepowders after calcination at 800 and 1000 8C are represented inFig. 4. The microstructure of the particles after heating at 800 8C isobserved to be nearly spherical, with particle size in the range 10–30 nm. When temperature increases to 1000 8C, the particle size ofthe powder is found to be 50–70 nm. The morphology of theparticles was observed to be more regular in shape when thesamples were heated from 800 to 1000 8C, indicating furthergrowth of the nanocrystals inside the powder. The growth could bevia an aggregation of primary particles by increasing temperature.The TEM observations on the particles are in agreement with thedata calculated by Scherrer’s equation, Table 2.

Please cite this article in press as: S. Salem, J. Ind. Eng. Chem. (2013

The DTA curves for the powders calcined at 600, 800 and1000 8C are illustrated in Fig. 5. It is interesting to note that the DTAcurve of powder calcined at 600 8C shows a exothermic peakaround 800 8C and the curve of powder calcined at 800 8C indicatesa weaker exothermic peak at the same temperature but there is nopeak in the curve of powder calcined at 1000 8C. This phenomenonmay be due to spinel phase transformation started at 800 8C. It canbe concluded that further calcination at 1000 8C is necessary forcomplete crystallization reaction.

), http://dx.doi.org/10.1016/j.jiec.2013.06.011

Fig. 4. The TEM micrographs of powders calcined at (a) 800 8C and (b) 1000 8C.

Fig. 5. The DTA of powders calcined at 600, 800 and 1000 8C.

Fig. 6. The FT-IR spectra of green and blue powders calcined at 600, 800 and 1000 8C.

S. Salem / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx 5

G Model

JIEC-1399; No. of Pages 6

Please cite this article in press as: S. Salem, J. Ind. Eng. Chem. (2013

The structures of the powders were analyzed using FT-IRspectroscopy after calcination at 600, 800 and 1000 8C, as shown inFig. 6. It is interesting to note that the green and blue powders havedifferent FT-IR spectra. By comparing the spectra, it can beconcluded that the normal spinel is formed after calcination at800 8C. Mainly the peaks for Al–O vibrations in the normal spinelcan be identified around wave number of 550 and 640 cm�1 with ashoulder at around 590 cm�1, which is attributed to the vibrationalband of normal spinel CoAl2O4 [9,17,38] whereas the band around661 cm�1 observed in the spectrum of green powder indicates thatnormal spinel is not formed at 600 8C. Based on FT-IR spectra thedifference between green and blue powders is related to a differentcrystal structure.

The obtained result by the FT-IR spectroscopy could beexplained more by the results of the Raman spectroscopytechnique. Fig. 7 shows the Raman spectra of the green and bluepowders. The normal spinel characteristics can be distinguishedin the Raman spectrum of blue powder and inverse spinel peakswere observed in green powder spectrum [38–40]. The Ramanspectra clearly showed that the calcined sample at 1000 8C havethe highest peak intensity at 380 cm�1. The intensity directlydepends upon the crystallinity of sample therefore, the highestdegree of crystallinity appears to be achieved after calcination at1000 8C. Comparing the Raman spectra of the heated powders at600 and 1000 8C lead to the conclusion that there is no indicationof normal spinel peaks in green powder. Therefore, the powderprepared by combustion technique has favorite colorant behaviorat 1000 8C.

Fig. 7. The Raman spectra of green and blue powders calcined at 600 and 1000 8C.

), http://dx.doi.org/10.1016/j.jiec.2013.06.011

S. Salem / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx6

G Model

JIEC-1399; No. of Pages 6

4. Conclusions

The nano-particles of CoAl2O4 spinel were successfully man-ufactured by solution-based combustion technique from Co(N-O3)2�6H2O, Al(NO3)3�9H2O as starting materials. According to theexperimental results, the color of cobalt-aluminate spinel isstrongly related to the successive calcination temperature. Thesynthesized powder has favorite colorant behavior when it iscalcined at 1000 8C. The amorphous phase remains in loweramount, less than 2 wt.%, and the particle size of the powder wasfound to be 50–70 nm after heat treatment at this temperature.Regardless crystallinity and crystallite size, the colorant behavior isalso related to spinal phase transformation which is observable at800 8C. In particular, DTA technique is more accurate for thedetection phase transformation than the XRD. The Raman testresults show that the green powder is characterized by an inversespinel structure whereas a normal spinel, CoAl2O4, correspondingto blue color is produced at higher temperatures.

Acknowledgments

The author would like to acknowledge Prof. F. Bondioli and Dr. L.Pasquali for their contribution during the investigation.

References

[1] A.L. Fernandez, L. Pablo, Pigment and Resin Technology 31 (2002) 350–356.[2] N. Ouahdi, S. Guillemet, B. Durand, R.E. Ouatib, L.E. Rakhob, R. Moussab, A. Samdi,

Journal of the European Ceramic Society 28 (2008) 1987–1994.[3] A. Nakatsuka, Y. Ikeda, Y. Yamasaki, N. Nakayama, T. Mizota, Journal of Solid State

Communication 128 (2003) 85–90.[4] S. Meseguer, M.A. Tena, C. Gargori, R. Galindo, J.A. Badenes, M. Llusar, G. Monros,

Ceramics International 34 (2008) 1431–1438.[5] D.M.A. Melo, J.D. Cunha, J.D.G. Fernandes, M.I. Bernardi, M.A.F. Melo, A.E. Marti-

nelli, Journal of Materials Research Bulletin 38 (2003) 1559–1564.[6] L.K.C. Souza, J.R. Zamian, G.N.R. Filho, L.E.B. Soledade, I.M.G. Santos, A.G. Souza, T.

Scheller, R.S. Angelica, C.E.F. Costa, Dyes and Pigments 81 (2009) 187–192.[7] K.E. Sickafus, J.M. Wills, Journal of the American Chemical Society 82 (1999)

3279–3292.[8] J.K. Burdett, G.L. Price, S.L. Price, Journal of the American Chemical Society 104

(1982) 92–95.[9] C. Wang, S. Liu, L. Liu, X. Bai, Journal of Materials Chemistry and Physics 96 (2006)

361–370.

Please cite this article in press as: S. Salem, J. Ind. Eng. Chem. (2013

[10] I.S. Ahmed, H.A. Dessouki, A.A. Ali, Spectrochimica Acta Part A. 71 (2008) 616–620.

[11] T. Mimani, S. Ghosh, Current Science 78 (2000) 892–896.[12] P.M.T. Cavalcante, M. Dondi, G. Guarini, M. Raimondo, G. Baldi, Dyes and Pigments

80 (2009) 226–232.[13] W. Li, J. Li, J. Guo, Journal of the European Ceramic Society 23 (2003) 2289–2295.[14] N. Ouahdi, S. Guillemet, J.J. Demai, B. Durand, L.E. Rakho, R. Moussa, A. Samdi,

Journal of Materials Letters 59 (2005) 334–340.[15] Z.Z. Chen, E.W. Shi, W.J. Li, Y.Q. Zheng, J.Y. Zhuang, B. Xiao, L.A. Tang, Materials

Science and Engineering: B 107 (2004) 217–223.[16] Z. Chen, E. Shi, W. Li, Y. Zheng, W. Zhong, Materials Letters 55 (2002) 281–284.[17] M.S. Niasari, M.F. Khouzani, F. Davar, Journal of Sol–Gel Science and Technology

52 (2009) 321–327.[18] D. Segal, Journal of Materials Chemistry 7 (8) (1997) 1297–1305.[19] F. Yu, J. Yang, J. Ma, J. Du, Y. Zhou, Journal of Alloys and Compounds 468 (2009)

443–446.[20] M. Lade, H. Mays, J. Schmidt, R. Willumeit, R. Schomacker, Journal of Colloids and

Surfaces 163 (2000) 3–15.[21] W.L. Suchanek, R.E. Riman, Journal of Advanced Science and Technology 45 (2006)

184–193.[22] S. Chokkaram, R. Srinivasan, D.R. Milbrun, B.H. Davis, Journal of Molecular

Catalysis A: Chemical 121 (2-3) (1997) 157–169.[23] S. Britto, A.V. Radha, N. Ravishankar, K.P. Vishnu, Journal of Solid State Science 9

(2007) 279–286.[24] R.D. Purohit, S. Saha, A.K. Tyagi, Ceramics International 32 (2006) 143–146.[25] T. Peng, X. Liu, K. Dai, J. Xiao, H. Song, Materials Research Bulletin 41 (2006)

1638–1645.[26] M.D. Lima, R. Bonadimann, M.J. Andrade, J.C. Toniolo, C.P. Bergmann, Journal of

the European Ceramic Society 26 (2006) 1213–1220.[27] Sh. Salem, S.H. Jazayeri, F. Bondioli, A. Allahverdi, M. Shirvani, Journal of Ceramics

Processing Research 12 (6) (2011) 677–681.[28] Sh. Salem, S.H. Jazayeri, F. Bondioli, A. Allahverdi, M. Shirvani, A.M. Ferrari,

International Journal of Applied Ceramic Technology 29 (2011) 1–11.[29] Sh. Salem, S.H. Jazayeri, F. Bondioli, A. Allahverdi, M. Shirvani, Thermochimica

Acta 521 (2011) 191–196.[30] Sh. Salem, S.H. Jazayeri, F. Bondioli, A. Allahverdi, M. Shirvani, Ceramic Science

and Technology 2 (2011) 169–178.[31] R.S. Hunter, The Journal of the Optical Society of America 48 (1958) 985–993.[32] R.A. Eppler, Journal of American Ceramic Society Bulletin 69 (2) (1990) 228–230.[33] H.P. Klug, L.E. Alexander, Chemical Education 32 (1955) 228–235.[34] A.C. Larson, R.B. Von Dreele, General Structure Analysis System (GSAS). Los

Alamos National Laboratory Report LAUR, 2000, 86–748.[35] B.H. Toby, Journal of Applied Crystallography 34 (2001) 210–213.[36] G.Y. Lee, K.H. Ryu, H.G. Kim, Y.Y. Kim, Bulletin of the Korean Chemical Society 30

(2) (2009) 373–377.[37] B.D. Cullity, Elements of X-Ray Diffraction, Second ed., Addison-Wesley, Mass,

1978.[38] E. Mwenesongole, A Raman and XRD Study of the Crystal Chemistry of Cobalt

Blue, University of Pretoria, 2008 Master of Science Thesis.[39] D. Waal, Journal of Raman Spectroscopy 35 (2004) 646–649.[40] L.D. Kock, D. Waal, Journal of Raman Spectroscopy 38 (2007) 1480–1487.

), http://dx.doi.org/10.1016/j.jiec.2013.06.011