Journal of Non-Crystalline Solids 356 (2010) 1447–1451

Contents lists available at ScienceDirect

Journal of Non-Crystalline Solids

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

Sol–gel synthesis of bioglass-ceramics using rice husk ash as a source for silica andits characterization

J.P. Nayak, S. Kumar, J. Bera ⁎Department of Ceramic Engineering, National Institute of Technology, Rourkela, Orissa-769008, India

⁎ Corresponding author. Tel.: +91 9437246159; fax:E-mail address: jbera@rediffmail.com (J. Bera).

0022-3093/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.jnoncrysol.2010.04.041

a b s t r a c t

a r t i c l e i n f o

Article history:Received 7 February 2010Received in revised form 22 April 2010Available online 9 June 2010

Keywords:Sol–gel;Bioglass-ceramics;Silica;Rice husk ash;Bioactivity

SiO2–Na2O–CaO based bioglass-ceramics was synthesized through sol–gel route using rice husk ash as silicasource. The decomposition behavior of gel was evaluated. Sodium–calcium–silicate phases were crystallizedabove 700 °C. Pellet made of glass-ceramics powder was sintered at 900 °C. In vitro bioactivity andbiodegradability of glass-ceramics were investigated by incubation in simulated body fluid and Tris buffersolution respectively. Scanning electron microscopy, energy dispersive spectroscopy and X-ray diffractionwere used to monitor the surface deposition on glass-ceramics during incubation. The material showed agood bioactivity with a formation of carbonated hydroxy apatite phase in 3 days of incubation. A quickdegradation of the material was observed in Tris solution associated with an increase of pH due to thedissolution of Ca2+ and Na+ ionic species. All these results suggest the glass-ceramics, prepared utilizing ricehusk ash, would be a low cost biomaterial for potential biomedical applications.

+91 6612462201.

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Since the pioneering work by Hench et al. in 1972 [1], bioactiveglass and glass-ceramics have attracted much attention towardsclinical applications due to their good bioactivity, osteoconductivityand biodegradability characteristics [2–5]. When bioactive materialsare implanted in human body, a biologically active carbonatedhydroxyl apatite layer is formed on their surfaces. Subsequently, theimplant bonds to and integrates with living bone [6,7]. In vivoimplanting study revealed that bioactive glasses show no local orsystemic toxicity, no inflammation, no foreign-body response, andbond to both soft and hard tissues without an intervening fibrouslayer [8]. Glass-ceramics has added advantage of better mechanicalproperties than glass [9]. Since, glass-ceramics is composed ofdifferent phases, some are more soluble than others in body fluid;the bioactivity process starts at more soluble areas [10]. Therefore, it isproposed that bioactive glass-ceramics with mechanical propertiesmuch closer to that of natural bone can be prepared [11]. For example,A-W glass-ceramics shows a high bioactivity and high mechanicalstrength compared to other glasses [12,13].

Generally, glass-ceramics are prepared by the traditional meltingand subsequent crystallization of glass [14–16]. This method hasdisadvantage of evaporation of volatile component during high-temperature processing. The sol–gel technique, requiring low proces-sing temperature, is an alternative approach to prepare glass-ceramics

without melting operation [17–19]. This route enables a wide range ofcompositions with high purity, homogeneity and production ofvarious shapes; such as monoliths, powders, fibers or coatings [20].Additionally, glass-ceramics obtained by the route exhibit highersurface area and porosity which are the critical factors for theirbioactivity [21].

Conventional sol–gel process uses alkoxysilane precursors, liketetraethylorthosilicate, tetramethylorthosilicate etc., as silica precur-sors. However, such precursors are expensive. Silica source of ricehusk ash (RHA) can be a cheap raw material for sol–gel synthesis ofglass-ceramics. RHA mainly contains amorphous silica with minorother metal impurities. Amorphous silica can be extracted easily assodium silicate [22,23] which further may be used as a precursor forthe sol–gel synthesis.

The aim of this paper is to report the preparation of bioglass-ceramics in the system of SiO2–CaO–Na2O through sol–gel route usingRHA as silica source. In vitro bioactivity and biodegradability of theglass-ceramics were investigated to analyze its feasibility as abiomaterial.

2. Experimental procedure

2.1. Preparation of glass-ceramics

The glass-ceramics with composition (mol%) 50 SiO2, 25 Na2O and25 CaO [24], was synthesized through sol–gel process. Sodium silicatesolution (containing required stoichiometric ratio of SiO2:Na2O) andCa(NO3)2·4H2O (Merck, India) were used as precursors. Sodiumsilicate solution was prepared by boiling RHA (SiO2≥99.87%, ±0.01)

Fig. 1. TGA/DSC plot of soda-lime-silicate gel powder.

Fig. 2. XRDpatterns of gel powder; (a) as prepared, and after calcination at (b) 300 (c) 500(d) 700 and (e) 900 °C.

1448 J.P. Nayak et al. / Journal of Non-Crystalline Solids 356 (2010) 1447–1451

in 1 M NaOH solution. RHA was prepared by burning cleaned husk at700 °C in sufficient flowing air for 6 h. Calcium nitrate was dissolvedin concentrated nitric acid to make the solution highly acidic. Silicatesolution was then added drop wise into that highly acidic Ca(NO3)2solution under constant stirring condition to avoid gel precipitation,as the gel was precipitated immediately when Ca (NO3)2 solution wasadded to sodium silicate solution. Final pH of the sol was at 3. The solturned into gel within 30 min at room temperature. The gel was agedfor 3 days at 70 °C followed by 150 °C drying for 48 h, as per the dryingschedule followed by Rainer et al. [25]. Dried gel was crushed andground into fine powder.

The decomposition behavior of gel was evaluated by thermogravimetric analysis and differential scanning calorimetry (TGA/DSC,Netzsch STA 449 C Jupiter) with 10 °C/minute heating rate. Gelpowder was calcined at different temperatures up to 700 °C for 2 h toevaluate crystallization behavior. Calcined powders were analyzed fortheir crystalline phase content using X-ray diffractometer (XRD,Philips PW 1830, Holland). Final powder (calcined at 700 °C for 2 h)was then ground and made into pellets by pressing. Pellets weresintered at 900 °C for 2 h. Sintered pellets showed an average bulkdensity of about 1.5 g/cc (±0.2) and about 20% (±2) apparentporosity. Crystalline phases of sintered specimen were evaluated byXRD.

2.2. In vitro bioactivity

The bioactivity of glass-ceramics was assessed following CuneytTas protocol [26] in simulated body fluid (SBF). The SBF was preparedby dissolving reagent-grade NaCl, NaHCO3, KCl, Na2HPO4·2H2O,MgCl2·6H2O, CaCl2·2H2O, and Na2SO4 (Merck, India) in deionizedwater. The solution was buffered to pH 7.4 with Tris buffer andhydrochloric acid. The sintered pellets were immersed in SBF andincubated for 3, 14 and 21 days at 37 °C. The glass-ceramics specimenwithout soaking in SBF is termed as zero days' specimen. After beingsoaked, the pellets were rinsed with deionized water and dried indesiccator at room temperature. The surfaces of specimen wereinvestigated by XRD for phase analysis, scanning electron microscopy(SEM, Jeol JSM-6480LV) for morphology and energy dispersivespectroscopy (EDS, Oxford Instrument, INCA) for elemental analysis.

2.3. In vitro degradation

In order to study the dissolution features of glass-ceramics, Trisbuffer solutions was chosen because, Tris is the plain buffering agentused in most SBF preparations. Tris solutions do not contain ions andthus represent maximum solubility and minimum reprecipitationactivity for a bioactive material [27]. Pure Tris was dissolved indeionized water with a concentration of 6.1 g/l. The solution pH waslowered to 8 using 1 M HCl solution. Sintered pellets were immersedin Tris solution and incubated at 37 °C for different time periods up to7 days. Tests were carried out in triplicate. pH of Tris solution (±0.01accuracy) and soaked weight of specimen were measured after eachsoaking experiment.

3. Results

3.1. Thermal decomposition and phase formation behavior of gel powder

The thermal decomposition behavior of gel precursor wasevaluated using TGA/DSC. Fig. 1 shows TGA/DSC curves of gelprecursor. The decomposition may be divided into four differentzones as shown in the figure. Zone I, extending up to ∼150 °C, shows∼12% weight loss with a broad endothermic peak at 120 °C. Zone II(150–520 °C) shows an endothermic peak at 227 °C and a smallweight loss (5%). Zone III (520–730 °C) shows a major weight loss of∼35% and four endothermic peaks at 580, 650, 690 and 720 °C

respectively. There is very small or no weight loss in zone IV (730–1000 °C).

The phase formation behavior during thermal decomposition andheat treatment of glass-ceramics precursors were investigated byXRD. Fig. 2 shows the XRD patterns of precursors after calcining atdifferent temperatures. As prepared gel is completely amorphous(Fig. 2(a)). However, NaNO3 phase crystallizes upon heating the gel.Fig. 2(b) shows well crystallized NaNO3 phase on the specimen afterheat treatment at 300 °C. The intensity of the NaNO3 peaks decreasesupon heat treatment above 500 °C (Fig. 2(c)). Sodium–calcium–

silicate phases were crystallized at and above 700 °C. Two phases;combeite-I, Na6Ca3Si6O18 (JCPDS 77-2189) and Na2Ca2Si2O7 (JCPDS10-0016) are crystallized initially (Fig. 3(d)). Finally, the major phaseis combeite-I. Some amount of combeite-II, Na4Ca4Si6O18 is alsodetected in the specimen that was heat treated at 900 °C (Fig. 3(e)).

3.2. Assessment of bioactivity in vitro

The bioactivity of glass-ceramics was assessed in vitro using SBFsolution. Fig. 3 shows the XRD pattern of glass-ceramics surface thoseincubated for different period of times in SBF along with untreatedsample. The specimen without SBF treatment (zero days) showstypical sodium–calcium–silicate phases as described above. Thesecrystalline phases are absent in 3-day SBF incubated specimen. 3-day

Fig. 3. XRD patterns of glass-ceramics surfaces after incubation in SBF for different timeperiod.

1449J.P. Nayak et al. / Journal of Non-Crystalline Solids 356 (2010) 1447–1451

specimen shows the presence of amorphous glassy phase and someamount of crystalline calcium phosphate-based phases. Carbonatedhydroxy apatite and hydrated calcium phosphate phases areidentified in the specimen. Amount of these two phases increases in14 and 21 day specimen. The amorphous glassy phase seems to beincreased in 14 day specimen.

Formation of bone like apatite layer during bioactivity experimentswas also evaluated by investigating the change of surface morphology.Fig. 4 shows the surface morphology of glass-ceramics specimen; thoseincubated in SBF with different time periods along with EDS elementalanalysis. Untreated glass-ceramics surface has acicular structuredmorphology (Fig. 4(a)) with typical EDS spectra showing Si, Ca andNa. The surface morphology changes with incubation periods. 3-daysurface shows the formationof individual apatitegrainswith size∼1 µm(Fig. 4(b)).With increasing incubation periods, apatite grains grows likespherical balls with an average diameter of ∼1.5 µm size in 14 days(Fig. 4(c)). 21-day surface (Fig. 4(d)) shows ∼5 µm size clusters withcauliflower like morphology covering the entire surface. The EDSspectra of these grain/clusters show the presence of P and C in additionto Na, Ca and Si; which indicates the formation of carbonated calciumphosphate hydrated phases.

Fig. 4. SEM image and EDS spectra of glass-ceramics; (a) before SBF i

3.3. In vitro dissolution

In vitro degradability test was carried out in Tris solution. Theprocess was monitored by water uptake of specimen and the changein pH of Tris-solution. Fig. 5 shows the percentage of water absorbedby the specimen and pH variation of the media with differentincubation time periods. The rate of water absorption, that is the rateof dissolution was rapid up to about 1 day, moderate up to about2 days and slow afterwards. The specimen shown in Fig. 5 was havingan initial apparent porosity of ∼18 % in Tris solution. The porosityincreases to about 21% in 7 days due to the dissolution of glass-ceramics. The similar trendwas obtained for pH of themedium, whichchanges from 8 to 10 within 7 days.

4. Discussion

The decomposition behavior of gel precursor is required for theoptimal synthesis of glass-ceramics. The gel is associatedwith two typesof water namely physically and chemically bonded water. About 12%weight loss and associated endotherm at 120 °C (zone I, Fig. 1) areattributed to the loss of physisorbedwater from gel. A small weight lossin the zone II (150 to 520 °C), might be due to the loss of remainingportion of physically absorbed water. This zone shows an endothermicpeak at 227 °C under which there is no weight loss. This may be due tothe eutectic melting of NaNO3–Ca(NO3)2 of gel. Similar eutectic meltingendotherm has been reported [28] for NaNO3–Ca(NO3)2 at 232 °C. Themajor weight loss of 35% in zone III is due to the decomposition ofsodium and calcium nitrate as well as by the removal of chemicallybonded water. The endothermic peaks at 580, 650, 690 and 720 °C maybe attributed to the decomposition of Ca(NO3)2 [29] and NaNO3 [30],which releases O2, NO and N2 gases during their decomposition.

Decomposition of gel should be correlatedwith phase formation tooptimize the processing parameters of glass-ceramics. By nature, thegel is amorphous (Fig. 2 (a)). Gel contains a continuous network ofsolid and water. Upon heat treatment, crystallization of NaNO3 occursin the gel (Fig. 2(b)). Similar crystallization has been reported in sol–gel glass upon heating above room temperature [31]. Then amount ofNaNO3 decreases (Fig. 2(c)) due to its partial decomposition at 500 °C.Dynamic decomposition of nitrates starts around at 530 °C (Fig. 1).Since the calcination was carried out in static condition at 500 °C for2 h, some decomposition takes place. Nitrate decomposition com-pletes upon calcinations at 700 °C for 2 h (Fig. 2(d)). Simultaneously,

ncubation, and after incubation for (b) 3 (c) 14, and (d) 21 days.

Fig. 5. Water uptake of glass-ceramics pellet and change in pH of medium withincubation time in Tris solution. (The lines joining data points are drawn as guides tothe eyes.).

1450 J.P. Nayak et al. / Journal of Non-Crystalline Solids 356 (2010) 1447–1451

calcium and sodium reacts with silica to form different sodium–

calcium–silicate phases. The major phase is Na6Ca3Si6O18, because,NaO:CaO:SiO2 in the precursor is in the ratio of 3 (Na2O):3 (CaO):6(SiO2). Crystallinity of these ceramic phases increases when fired at900 °C (Fig. 2(e)) due to growth phenomena. It is evident that thecrystallization of ceramic phases occurs mainly in zone IV of Fig. 1; inwhich the DSC curve shifts towards positive, that is, exothermicdirection due to this crystallization.

The bioactivity in vitro is evaluated by checking the ability of bonelike apatite phase formation on the surface of biomaterial in simulatedbody fluid. The glass-ceramics shows a good bioactivity in vitro.Hydrated calcium phosphate phase was form within 3 days ofincubation (Fig. 3). Sodium–calcium–silicate phases of glass-ceramicswere completely dissolved in SBF within 3 days, due to the release ofNa+ and Ca2+ ions into the solution. Dissolution of surface ions leavesa glassy matrix on the surface. Glassy matrix increased in 14 daysspecimen; this might be due to some structural changes in the matrix.

The apatite formation upon bioglass-ceramics is speculated toproceed by the following mechanism [32]. The glass-ceramics releasesCa2+ and Na+ ions from its network via an exchange with the H+ andH3O+ ions in SBF to formSi–OHgroupson the surface. The Si–OHgroupsformed induce apatite nucleation through the formation of amorphouscalcium silicate and calcium phosphate. Once nucleated, the apatite cangrow spontaneously as the body fluid is highly supersaturated withrespect to apatite under normal condition. Released Ca2+ and Na+ ionsof glass-ceramics also accelerate apatite nucleation by increasing theionic activity product of apatite in the fluid. Itmay be concluded that theglass-ceramics prepared in present investigation, show a good bioac-tivity in 3 days and comparable to that of A-W glass-ceramics [12,13].Good bioactivity of the glass-ceramics is due to its sol–gel origin ofsynthesiswhichprovide abundant amount of Si–OHfunctional groupontheir surfaces [33] for apatite nucleation.

Acicular morphology of glass-ceramics is due to the precipitationof ceramics in glassy matrix (Fig. 4(a)). XRD analysis reveals that,ceramics dissolves in SBF within 3 days of incubation. The surfacemorphology also supports this as there are no acicular grains in 3-daysurface (Fig. 4(b)). Calcium hydrated phosphate phases are detectedin 3, 14 and 21 days specimen. So, the grains and clusters shown bythese surfaces (Fig. 4(b–d)) are attributed to the formation ofphosphate phases. Similar changes in apatite morphologies havebeen reported for sol–gel derived glasses and glass-ceramics [34].

The dissolution of glass-ceramics is dependent on Ca2+ and Na+

within it [27,35]. As discussed earlier, the ions were leached out from

the glass-ceramics, exchanged with H+ ions from the solution. It isapparent that the pH of the medium will increase by the dissolution.To compensate the leached out material, equilibrium water moleculediffuse into the specimen and fills the void space created by theleached material; hence water absorption increases [36]. The ratedissolution as well as pH increment decreases after about 2 to 4 daysdue to the decrease in surface Ca2+ and Na+ ionic concentration.

Based on the study, a physical model similar to that proposed byKokubo [12], may be used to explain these results. As like the model,ceramic combeite phases are considered to be embedded in glassymatrix and Ca, Na and silicate ions are dissolved fromboth the combeiteand theglassymatrix. pHof themediumincreases due tohighbasicityofNa+ ion. The rate of dissolution aswell aspH incrementdecreases after 3to 4 days due to the decrease in the dissolving ions at the surface. Themodel also proposed that the dissolution of Ca2+ ion increases thedegree of super saturation of surrounding fluid with respect to apatiteand thedissolutionof silicate ionsprovides favorable sites for nucleationof the apatite on the surface of glass-ceramics.

In summary, the present sol–gel glass-ceramic material shows goodbioactivity and bio-dissolution within 3 days. It is well established thatsol–gel bioglasses have higher bioactivity than melt driven bioglasses[1]. This is because the sol–gel processing has the advantage withrespect to the control of material's surface chemistry which directlyrelates to the bioactivity [37]. Themajor differences in surface chemistrybetween gel- andmelt driven-glasses are [37]: (i) gel-glass has a largervolume fraction of nanometreporosity on the surface; (ii) gel-glass has alarger concentration of silanols on the surface; and (iii) gel-glass hasmetastable three-membered and four-membered siloxane rings on thesurface. All these surface chemistry render the gel driven glass or glass-ceramics as a better bioactive material than conventional melt drivenglass-ceramics. Results also show that, rice husk ash can successfully beused to synthesize a bioactive sol–gel glass-ceramics.

5. Conclusions

Soda-lime-silica based bioglass-ceramicswas synthesized via sol–gelroute utilizing sodium silicate, derived from rice husk ash. Gel powderwas calcined at 700 °C for 2 h to get a reactive glass-ceramics powder.The calcined powder mainly contains combeite-I (Na6Ca3Si6O18)crystalline phase dispersed in amorphous glass matrix. Sintered pelletwith about 20% apparent porosity was tested for bioactivity andbiodegradability. Crystalline combeite phase of the glass-ceramics wasfound to dissolve easily in SBF and Tris buffer solution. Carbonatedhydroxyapatite was formed on the surface of the glass-ceramics within3 days of incubation in SBF. Finally, it may be concluded that rice huskash may be low costing raw material for the preparation of bioglass-ceramic materials through simple sol–gel route.

Acknowledgements

The authors are thankful to the Ministry of Environment andForests, Government of India, New Delhi for providing the researchgrant vide sanction no. 19/50/2004 RE.

References

[1] L.L. Hench, R.J. Splinter, W.C. Allen, T.K. Greenlee, J. Biomed. Mater. Res. 5 (1972)117.

[2] L.L. Hench, Curr. Orthop. 14 (2000) 7.[3] J. Wilson, S.B. Low, J. Appl. Biomater. 3 (1992) 123.[4] M. Vallet-Regi, C.V. Ragel, A.J. Salinas, Eur. J. Inorg. Chem. (2003) 1029.[5] L.L. Hench, J. Am. Ceram. Soc. 74 (1991) 1487.[6] P.J. Li, I. Kangasniemi, K. Degroot, T. Kokubo, A.U. Yliurpo, J. Non-Cryst. Solids 168

(1994) 281.[7] M.M. Pereira, A.E. Clark, L.L. Hench, J. Biomed. Mater. Res. 28 (1994) 693.[8] L.L. Hench, J. Biomed. Mater. Res. 41 (1998) 511.[9] T. Kokubo, S. Ito, S. Sakka, T. Yamamuro, J. Mater. Sci. 21 (1986) 536.

[10] S. Padilla, J. Roman, A. Carenas, M. Vallet-Regi, Biomater 26 (2005) 475.[11] A. Salinas, M. Vallet-Regi, Z. Anorg, Allg. Chem. 663 (2007) 1762.

1451J.P. Nayak et al. / Journal of Non-Crystalline Solids 356 (2010) 1447–1451

[12] T. Kokubo, Biomater 12 (1991) 155.[13] T. Kokubo, S. Ito, T. Huang, T. Hayashi, S. Sakka, T. Kitsugi, J. Biomed. Mater. Res. 24

(1990) 331.[14] T. Kasuga, Y. Abe, J. Non-Cryst. Solids 243 (1999) 70.[15] Y. Zhang, J.D. Santos, J. Non-Cryst. Solids 272 (2000) 14.[16] S. Kumar, P. Vinatier, A. Levasseur, K.J. Rao, J. Solid State Chem. 177 (2004) 1723.[17] Z. Hong, A. Liu, L. Chen, X. Chen, X. Jing, J. Non-Cryst. Solids 355 (2009) 368.[18] D.B. Joroch, D.C. Clupper, J. Biomed. Mater. Res. 82A (2007) 575.[19] J. Roman, S. Padilla, M. Vallet-Regi, Chem. Mater. 15 (2003) 798.[20] J. Lao, J.M. Nedelec, Ph. Moretto, E. Jallot, Nucl. Instrum. Meth. Phys. Res. B 245

(2006) 511.[21] P. Sepulveda, J.R. Jones, L.L. Hench, J. Biomed. Mater. Res. 59 (2005) 340.[22] U. Kalapathy, A. Proctor, J. Shultz, Bioresour. Technol. 73 (2000) 257.[23] J.P. Nayak, J. Bera, Trans. Ind. Ceram. Soc. 68 (2009) 91.[24] S. Fujibayashi, M. Neo, H.-M. Kim, T. Kokubo, T. Nakamura, Biomater 24 (2003)

1349.

[25] A. Rainer, S.M. Giannitelli, F. Abbruzzese, E. Traversa, S. Licoccia, M. Trombetta,Acta Biomater. 4 (2008) 362.

[26] A. Cuneyt Tas, Biomater 21 (2000) 1429.[27] M.G. Cerruti, D. Greenspan, K. Powers, Biomater 26 (2005) 4903.[28] E.A. Dancy, P. Nguyen-Duy, Thermo. Acta 42 (1980) 59.[29] C. Ettarh, A.K. Galwey, Thermochimica Acta 288 (1996) 203–219.[30] Y. Hoshino, T. Utsunomiya, O. Abe, Bull. Chem. Soc. Jpn 54 (1981) 1385.[31] J. Phalippou, M. Prassas, J. Zarzycki, J. Non-Cryst. Solids 48 (1982) 17.[32] O. Peital, E.D. Zanotto, L.L. Hench, J. Non-Cryst. Solids 292 (2001) 115.[33] T. Kokubo, H. Takadama, Biomater 27 (2006) 2907.[34] M. Mami, A. Lucas-Girot, H. Oudadesse, R. Dorbez-Sridi, F. Mezahi, E. Dietrich,

Appl. Surf. Sci. 254 (2008) 7386.[35] D.A. Cortes, A. Medina, J.C. Escobedo, S. Escobedo, M.A. Lopez, J. Biomed. Mater.

Res. A 70 (2004) 341.[36] B.C. Bunker, J. Non-Cryst. Solids 179 (1994) 300.[37] W. Cao, L.L. Hench, Ceram. Int. 22 (1996) 493.