Physical and structural properties of calcium iron

Journal of Non-Crystalline Solids 402 (2014) 135–140

Contents lists available at ScienceDirect

Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/ locate/ jnoncrysol

Physical and structural properties of calcium iron phosphate glass doped with rare earth Xiaofeng Liang a,b,⁎, Haijian Li a,b,1, Cuiling Wang a,b, Huijun Yu a,b, Zhen Li a,b, Shiyuan Yang b a b

Analytical and Testing Center, Southwest University of Science and Technology, Mianyang 621010, PR China State Key Laboratory Cultivation Base for Nonmetal Composites and Functional Materials, Southwest University of Science and Technology, Mianyang 621010, PR China

a r t i c l e

i n f o

Article history: Received 23 January 2014 Received in revised form 19 May 2014 Available online 13 June 2014 Keywords: Phosphate glass; Rare earth; Structure; Hardness; Raman spectra

a b s t r a c t The physical and structural properties of calcium iron phosphate glasses with different Gd2O3 contents and various rare earth oxides (Y2O3, La2O3, Nd2O3, Sm2O3, Gd2O3) were systematically studied by investigating their density, Vickers-hardness and Raman spectra. The results show that the compositions which contain up to 10 mol% of rare earth oxides formed homogeneous glasses and no crystalline phases. The properties of density and molar volume were discussed with different Gd2O3 contents and various rare earth elements, and Y-doped glass is an exception in molar volume. Vickers-hardness increases with the increase of cationic field strength of the corresponding rare earth elements. These physical properties have relations with the rare earth glass structure. The shape of the Raman spectra is affected and a very strong depolymerization appears in studied phosphate glasses with the Gd2O3 addition. The relative area of (PO3)2− bonds in Q1 units with different rare earth glasses increase with increasing cationic field strength of corresponding rare earth ions. The P–O distance and rare earth coordination numbers were discussed to understand further the glass structure. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Calcium iron phosphate glasses have recently been researched in glass fiber, biomaterials, semiconductor and optical materials, as well as in more technologically demanding applications related to magnetic materials and high-level radioactive waste immobilization [1–6]. Metcalfe et al. [6] investigated that the dissolution rates of the ternary CaO–Fe2O3–P2O5 glasses were measured gravimetrically on monolithic samples, and the glass compositions had favorable properties on corrosion resistance. Compared to borosilicate glasses for the vitrification of high-level radioactive wastes, an important advantage of the phosphate glasses is given by their ability to incorporate high concentrations of heavy metal oxides and still remain amorphous [7,8]. Rare earth (RE) glasses are known for their superior physical and chemical properties, such as greater hardness, greater thermal characterization and elastic modulus, especially, higher chemical durability, owing to higher field strength than traditional network modifier cations [9–11]. Generally, rare earth elements are assumed to act in the glasses as network modifiers, which are incorporated in the space between [PO4] tetrahedral. Because the ionic radii of the lanthanide elements decrease continuously with their increasing atomic number, their ⁎ Corresponding author at: Analytical and Testing Center, Southwest University of Science and Technology, Mianyang 621010, PR China. Tel.: +86 816 6089507. E-mail address: [email protected] (X. Liang). 1 Co-first authors.

http://dx.doi.org/10.1016/j.jnoncrysol.2014.05.021 0022-3093/© 2014 Elsevier B.V. All rights reserved.

cationic field strength (CFS) also continuously changes with their atomic number. This causes consequent variations in the various properties [12,13]. CFS is defined as CFS = Z/r2, where Z is the valence of the corresponding elements and r is its ionic radius. The ionic radius at the given valence state depends upon the coordination number, which was reported by Shannon [14]. In radioactive waste forms, trivalent lanthanide ions are good surrogates only for the heaviest transuranic elements occurring in HLW such as Cm and Am [15,16]. Indeed, these two actinide elements show the highest contribution during approximately 105 years [17]. They have cation radii r similar to the one of Nd3 + ion (for instance in sixfold coordination: r(Am3+) = 1.01 Å, r(Cm3+) = 0.98 Å and r(Nd3+) = 0.995 Å [14]). On the other hand, gadolinium, a lanthanide with lower non-radiative decay rate and higher thermal neutron capture crosssection, may be added to the final actinide-containing glasses to minimize the likelihood of criticality during the storage period [18–20]. Vibrational spectroscopy has been employed to investigate the structure of glasses and specifically the identification of the main structural units, which are needed to better understand the complex multicomponent glasses used to immobilize high-level liquid waste. It is well known that the network structure of phosphate glasses consists of PO4 tetrahedra which are connected through non-bridging oxygen (NBO) and bridging oxygen (BO) [21]. In this paper, the effect of rare earth oxides on the physical and structural properties of CaO–Fe2O3–P2O5 glasses was systematically studied by investigating its density, Vickers-hardness and Raman spectra.

136

X. Liang et al. / Journal of Non-Crystalline Solids 402 (2014) 135–140

2. Experiment

3. Results

The glass system of xRE2O3–(100 − x) (12CaO–20Fe2O3–68P2O5) (RE = Y, La, Nd, Sm and Gd) was prepared using a standard meltquench technique. Both sides of glass compositions vertical and horizontal were prepared from reagent grade NH4H2PO4, Fe2O3, CaO and RE2O3. In the vertical side, Gd2O3 (x = 0, 2, 4, 6, 8, 10, 12, 14 mol%) was added to CaO–Fe2O3–P2O5 base glass to determine the glassforming range. In the horizontal side, RE2O3, which corresponds to Y2O3, La2O3, Nd2O3, Sm2O3 and Gd2O3, was added to the base glass to study the effect of rare earth oxide on the phosphate glass. After thorough homogeneous mixing the powders (30–40 g) were introduced in corundum crucibles, in order to prevent excess boiling and consequent spillage, water and ammonia in NH4H2PO4 were removed initially by preheating them at 220 °C for about 2 h and then the electric furnace was raised to 1250 °C (heating rate was 10 °C·min−1) for 3 h. The melts were then poured into preheated steel molds, and moved quickly to an annealing furnace, annealed at 475 °C for 2 h and cooled down to room temperature more than 12 h. Samples for property measurements were cut from the ground. XRD analysis was performed on samples employing an X-ray diffractometer (X'Pert PRO; PANalytical, Netherlands). The 2θ scans were made between 5° and 80° with a step width of 0.03° and utilized Cu Kα radiation (λ = 1.5405 Å). Preparation of the samples used a simple top pack loading method for an acquired smooth surface. Scanning Electron Microscope (SEM) and Energy dispersive spectroscopy (EDS) microanalysis were performed by an Ultra 55 scanning electron microscope (Libra 200FE; Carl Zeiss SMT, Germany). The density, D, of each glass was measured at room temperature using the Archimedes method with water as an immersing liquid. The sample weights varied between 3 and 4 g, and the measured densities were reproducible within 0.03 g·cm−3. The molar volume (Vm) was calculated using the relation Vm = ∑(xiMi)/D, where xi is the molar fraction and Mi is the total molecular weight of the component. The composition of the prepared glass was used for the calculation of Vm. The Vickers-hardness was measured by a Hv-1000A hardness tester with a load of 0.3 kgf for 13 s. The absolute hardness values were calculated automatically according to the well known formula and expressed in conventional units HV,

3.1. Glass preparation and physical properties

2

2

HV ¼ 2Fsinðα=2Þ=d ¼ 1:8544F=d ; where d is the average of two indentation diagonals in mm, F is the loading force in kgf and α is the diamond pyramid edge angle in 136°. The hardness was determined from the mean value of 8 indentations. Hence the errors on the measured values correspond to the standard deviation is of 3%. Raman spectra at 400–1600 cm−1 were collected from glass powders using the InVia Raman Microscope (Renishaw, U.K.) at room temperature. The Raman spectra were excited by 514.5 nm light from an argon ion laser. The spectral resolution was about 1–2 cm− 1 and the wavenumber accuracy was 0.2 cm−1. Six multiple measurements per sample were done to check for the potential micron-range heterogeneity and for the effects of sample orientation, both of which have not been found, as it would be expected for a nearly homogeneous glass sample. Because the majority of the bands in the Raman spectra are broad and asymmetric, presenting also some shoulders, a deconvolution of the experimental spectra was necessary. This fact was made with ORIGIN 7.5 program using a Gaussian type function and allowed us a better identification of all the bands which appear in these spectra and their assignments. The proportion of particular structures corresponding to different vibration modes, was calculated from the areas of the fitted Gaussian bands divided to the total area of all bands. The two parameters of each band (peak frequency and relative areas) were allowed to float during the iterations.

Samples have been analyzed by XRD to confirm their amorphous or crystalline nature. The XRD patterns of Gd-doped glass samples are given in Fig. 1. As seen from the figures, the patterns are typical of X-ray amorphous when Gd2O3 ≤ 10 mol%. The identified crystalline phase occurs when Gd2O3 ≥ 12 mol%. The major crystalline phase has been identified as GdPO4 (PDF No.01-083-0657). Moreover, the X-ray diffraction analysis of all samples with RE2O3 ≤ 10 mol% indicates the amorphous character, as expected. Fig. 2 shows SEM micrographs of 10 mol% Gd2O3 glass and crystallized 14 mol% Gd2O3 glass. SEM observations of the Gd10 indicate that the surface is smooth and homogeneous in Fig. 2(a). However, the Gd14 has a much rougher surface due to the great crystallization from the glass matrix in Fig. 2(b). Energy dispersive spectroscopy (EDS) microanalysis mounted on the SEM machine was carried out to clarify the compositions of the sample. The EDS microanalyses of the Gd10 block sample show that the atomic ratios of Gd, Ca, Fe, P and O is 3.14, 1.41, 8.25, 19.31 and 67.89%, respectively (Fig. 2(a)). The error of the EDS detector was considered approximately ± 2 at.%, and the actual compositions of these elements basically correspond with the nominal composition of the Gd3.39Ca1.83Fe6.10P20.75O67.93. Thus, the low presence of Ca and Gd was confirmed in the glass system. The average composition of the glassy and crystalline phases of the Gd14 block sample determined by EDS analysis is Gd0.98Ca1.28Fe7.92P16.83O72.92 and Gd2.57Ca1.10Fe6.20P9.22O80.90, respectively (Fig. 2(c)). The results show that the crystalline phase contains small amounts of Ca and Fe elements. The density, mole volume and Vickers-hardness of Gd2O3 doped phosphate glasses were given in Fig. 3. The density of the glasses increased from 2.82 to 3.33 g/cm3 as the Gd2O3 content increased from 0 to 10 mol%. The rate of increase seems to be larger at higher concentrations of Gd2O3. The molar volume decreased slightly with the increase in the concentration of Gd2O3 (except 2 mol% Gd2O3 glass). On the other hand, the hardness of the glass system increased from 470 to 567 kgf/mm2 with the increase of Gd2O3 content as well (except 2 mol% Gd2O3 glass). Fig. 4 shows, respectively, the dependence of density, molar volume and Vickers-hardness on the different rare earth oxides (10 mol%) in the glasses. The density increased and molar volume decreased as the atomic weight of the rare earth elements increase. The hardness increased from 533 kgf/mm2 in La-doped glass to 570 kgf/mm2 in Y-doped glass as the ionic radius of the rare earth elements decreases.

Fig. 1. XRD patterns of phosphate glasses doped with Gd2O3.


137

Fig. 3. The density and mole volume (a) and Vickers-hardness (b) of phosphate glasses doped with Gd2O3. The lines are drawn as guides for eyes.

Fig. 2. SEM micrographs of block samples: (a) 10 mol% Gd2O3; (b and c) 14 mol% Gd2O3.

3.2. Raman spectroscopy Glass structure was investigated using Raman spectra. It is well known that the Raman spectra of phosphate glasses are usually characterized in the frequency region of 400–1600 cm−1 [22]. Fig. 5 shows the experimental Raman spectra of the studied calcium iron phosphate glass system doped with Gd2O3. In the Gd-doped glasses, the lowerfrequency band shifts markedly with increasing Gd2O3 content from 708 cm−1 up to 733 cm−1. It is assigned to the symmetric stretching

mode of P\O\P bonds in Q1 units [23]. The higher frequency region (900–1300 cm− 1) of the Raman spectra of the base glass (RE0) has two dominant vibrational bands at 1072 cm−1 and 1205 cm−1, which can correspond to the symmetric stretching mode of (PO3)2− bonds in Q1 units and symmetric stretching mode of ‘strained’ (PO2)− (strained structural units, possibly three- or four-membered rings) in Q2 units, respectively [24–26]. With increasing Gd2O3 content in the composition, the two bands merge into one band at 1102 cm−1, it is assigned to the symmetric stretching mode of (PO3)2− bonds in Q1 units [26]. Fig. 6 shows the Raman spectra of the studied calcium iron phosphate glass system doped with different rare earth. After the base glass doped 10 mol% rare earth oxides, the band at 708 cm−1 shifts basically to the high frequency band at 733 cm−1, and the most intense peak in the spectra is observed within the range of 1092–1102 cm−1 (except Nd-doped glass). The band at 1143 cm− 1 implies a bond order intermediate between those of Q1 and Q2 tetrahedra, while a similar band in zinc polyphosphate glass was observed by Brow et al. [27]. As the atomic weight of the rare earth elements increases, the position of the most intense peaks has an obvious red shift. 4. Discussion The crystallization tendency of the rare earth calcium iron phosphate glasses can be related to the high amount of rare earth ions that act as a nucleating agent in this base glass. SEM micrographs and structural properties are focused on those samples which have been

138


Fig. 5. Raman spectra of phosphate glasses doped with Gd2O3.

The intensities and the shifts of the Raman bands depending on the kind and concentration of network modifiers and the correlation of these data with the structural groups which appear in the studied

Fig. 4. The density and mole volume (a) and Vickers-hardness (b) of phosphate glasses doped with different rare earth.

confirmed to be XRD amorphous. Hence, the glass formation has a wider range to modify its network by containing 10 mol% RE2O3. According to XRD patterns only one, GdPO4, phase crystallizes in the sample, the microstructure of the Gd14 sample observed by EDS demonstrates that all the Gd, Ca, Fe, P and O elements exist in the glass matrix. The primary phase for samples in which crystallization occurs during cooling is rich in various elements. The density and molar volume measurements have been done to understand the structural changes occurring in these glasses. It is observed in Fig. 3. that density increases with the increase of Gd2O3 content. This is an expected result as high relative molecular mass Gd2O3 filled the interstice of glass network [28]. This decreases in Vm is mainly attributed to the higher molecular weight of Gd2O3 and decrease in the total number of oxygen atoms [29]. The increase of the hardness is possible only due to structure “tightening” [12]. The variation of Vm with different rare earth glasses is shown in Fig. 4. The densities increase with the increase of the molecular mass of the rare earth. The Vm of the Y-doped glass may be attributed to create a large number of nonbridging oxygens (NBOs) [29,30]. Hardness is a physical property, which indirectly reflects the strength of the bonds in the material. It is often related to the CFS, which was reported by Lofaj et al. [12]. Hardness values increase with the increase of CFS (or reduction of ionic radius) of the corresponding rare earth elements. Similar results were observed by Lofaj et al. [12]. Ramesh et al. [31] thought that the attractive forces between RE3 + and surrounding structural units increase as the field strength of the RE cations increases. If this is the case then the hardness increases.

Fig. 6. Raman spectra of phosphate glasses doped with different rare earth.


Fig. 7. Deconvoluted Raman spectra of phosphate glasses using a Gaussian-type function for the base glass (a) and the doped 10 mol% Gd2O3 (b).

glasses are investigated. These results allowed us to understand the modifier role of the rare earth oxides. The shift of the band near 708 cm−1 may be attributed to a change in the in-chain P\O\P bond angle as an effect of the Gd2O3 network modifier on the glass structure [32,33]. The larger frequency of the symmetric band is a result of the smaller P\O\P bond angle, characteristic for shorter phosphate chain length [34]. The result of peak deconvolution indicates a number of 5 peaks in the region from 800 cm-1 to 1400 cm−1 in Fig. 7. In the base glass, the new band at ~968 cm−1 is assigned to the symmetric stretching mode of non-bridging (PO4)3− oxygen ions in Q0 units [27], while that at ~1139 cm-1 is similar to the band at 1143 cm−1 in the Nd-doped glass. The band is assigned to phosphorus–oxygen stretching modes in Q1 phosphate chain terminator units formed by the scission of phosphate chains [27]. The band at ~1292 cm−1 is assigned to the P = O symmetric stretching [21,35].

139

Fig. 7(a) and (b) shows the deconvoluted spectroscopy of the base glass and doped 10 mol% Gd2O3 glass, respectively. With the increase of Gd2O3 content, the relative area of (PO3)2 − bonds in Q1 units increases and the relative area of the P_O decreases (Table 1). This behavior is attributed to structural rearrangements in the main phosphate network due to the replacing of P_O by P\O\Gd bonds [36,37]. Moreover, the relative area of the P\O bond in Q1 chain terminator units decreases from 26.84% in the base glass to 9.34% in 10 mol% Gd-doped glass with the increase of Gd2O3 content, and the frequency of the band decreases gradually. Brow et al. [27] showed that the P\O\Zn bonds with ZnO addition can participate in the pyrophosphate glass structure by replacing the Q1 terminal oxygen, which is due to pyrophosphate disproportionation reaction in glass melts. It is then deduced that P\O\Gd bonds participate in the pyrophosphate glass structure by replacing the Q1 terminal oxygen, until pyrophosphate units dominate the structure. Accordingly, the glass network becomes more crosslinked, with consequent increases in hardness with the addition of Gd2O3. The most notable change in doped 10 mol% rare earth glasses is the decrease of the relative area of the P_O bonds compared to the base glass in Table 2. The change is attributed to the replacing of P_O by P\O\RE bonds. A similar behavior in rare earth phosphate glasses was observed by Cole et al. [37]. The relative area of (PO3)2 − bonds in Q1 units is seen to increase in the order: Nd10 b La10 b Sm10 b Gd10 b Y10. The order is mainly attributed to the effect of the cation field strength (except Nd-doped glass). Cations with smaller sizes or higher field strength will stabilize NBO more effectively and contain a higher fraction of NBO than cations with larger sizes or lower field strength, which in turn help to stabilize the modifier cation in the melt network [38]. The structure analysis of Raman spectra shows that higher cation field strength provides more non-bridging oxygen to depolymerize the glass network. Neutron diffraction studies indicate a decrease in the mean P\O bond length with increasing modifier field strength [39]. Rufflé et al. [40] thought that larger field strength cations increased structural rigidity, and so does increasing glass transition temperature. For example, Y-doped phosphate glass could have smaller the P–O distance than La-doped glass, which could lead to the change of hardness in the rare earth glasses. In addition, the extended X-ray absorption fine structure (EXAFS) results showed that the R3+ ions are surrounded exclusively by between six and eight oxygen ions [37]. Hoppe et al. [41] showed that rare earth coordination numbers (NREO) were clearly different for meta- and ultraphosphate glasses with values of ~6.5 and ~8.0, respectively. With the increase of Gd2O3 content, oxygen to phosphorus molar ratio increases from 3.02 to 3.27, which results in the increase of NREO. Moreover, NREO increases as the rare earth ions increase in size [21]. Cations with higher coordination numbers can strengthen the glass structure. 5. Conclusions The impact of different Gd2O3 contents and various rare earth oxides (Y2O3, La2O3, Nd2O3, Sm2O3, Gd2O3) on the physical and structural properties of CaO–Fe2O3–P2O5 glasses has been systematically studied. X-ray diffraction analysis and SEM micrographs show that the compositions

Table 1 Deconvolution parameters (the Raman bands and the relative area) and the bands assignments of phosphate glasses doped with Gd2O3. Range frequency (cm−1)

938–969 1056–1069 1111–1139 1175–1210 1278–1298

Assignments

(PO4)3−sym (PO3)2−sym

Relative areas dependence of vibrational bands (%)

0

stretch (Q ) stretch (Q1) P–O stretch (Q1 chain terminator) ‘strained’ (PO2)−sym (Q2) (P_O)sym stretch

RE0

Gd2

Gd4

Gd6

Gd8

Gd10

8.92 24.58 26.84 22.45 17.19

13.85 27.98 23.05 26.12 8.99

8.52 30.19 20.27 30.96 10.05

10.29 32.70 24.03 25.71 7.26

13.87 32.67 14.07 34.15 5.23

12.10 39.41 9.34 31.99 7.14

140


Table 2 Deconvolution parameters (the Raman bands and the relative area) and the band assignments of phosphate glasses doped with rare earth. Range frequency (cm−1)

938–968 1056–1066 1111–1139 1175–1210 1269–1298

Assignments

(PO4)3−sym (PO3)2−sym

Relative areas dependence of vibrational bands (%)

0

stretch (Q ) stretch (Q1) P–O stretch (Q1 chain terminator) ‘strained’ (PO2)−sym (Q2) (P_O)sym stretch

RE0

Y10

La10

Nd10

Sm10

Gd10

8.92 24.58 26.84 22.45 17.19

11.42 39.60 14.09 27.06 7.83

9.89 34.07 18.68 31.68 5.67

6.60 32.58 26.52 25.31 8.99

6.40 38.59 13.72 33.84 7.45

12.10 39.41 9.34 31.99 7.14

contain up to 10 mol% of rare earth oxides formed homogeneous glasses and no crystalline phases. The properties of density and molar volume were discussed with different Gd2O3 contents and various rare earth elements, and Y-doped glass is an exception in molar volume. Vickershardness values increase with the increase of CFS of the corresponding rare earth elements. In the Raman spectra, the characteristic bands of these glasses due to the stretching vibrations were identified and analyzed. A very strong depolymerization appears in studied phosphate glasses with the Gd2O3 addition. The relative area of (PO3)2− bonds in Q1 units with different rare earth glasses increases with increasing cationic field strength of corresponding rare earth ions. The P–O distance and rare earth coordination numbers were discussed to understand further the glass structure. Acknowledgments This work was supported by the Science Foundation of Southwest University of Science and Technology (11zx7157) and the Scientific Research Fund of Sichuan Provincial Education Department (14ZA0105). References [1] M.S. Mohammadi, I. Ahmed, N. Muja, S. Almeida, C.D. Rudd, M.N. Bureau, S.N. Nazhat, Acta Biomater. 8 (2012) 1616–1626. [2] I. Zebger, F. Pfeifer, N. Nowack, J. Non-Cryst. Solids 351 (2005) 3443–3457. [3] M.M. El-Desoky, A. Al-Shahrani, Physica B 371 (2006) 95–99. [4] I. Ardelean, C. Andronache, P. Păşcuţă, Mod. Phys. Lett. B 17 (2003) 1271–1275. [5] B. Kumar, S. Lin, J. Am. Ceram. Soc. 74 (1991) 226–228. [6] B.L. Metcalfe, S.K. Fong, I.W. Donald, Glass Technol. 46 (2005) 130–133. [7] D.E. Day, Z. Wu, C.S. Ray, P. Hrma, J. Non-Cryst. Solids 241 (1998) 1–12. [8] W. Huang, D.E. Day, C.S. Ray, C.W. Kim, J. Nucl. Mater. 346 (2005) 298–305. [9] J. Marchi, D.S. Morais, J. Schneider, J.C. Bressiani, A.H.A. Bressiani, J. Non-Cryst. Solids 351 (2005) 863–868. [10] A. Aronne, S. Esposito, P. Pernice, J. Non-Cryst. Solids 51 (1997) 163–168. [11] S.L. Lin, C.S. Hwang, J. Non-Cryst. Solids 202 (1996) 61–67. [12] F. Lofaj, R. Satet, M.J. Hoffmanna, A.R. de Arellano López, J. Eur. Ceram. Soc. 24 (2004) 3377–3385. [13] S. Hampshire, M.J. Pomeroy, J. Non-Cryst. Solids 344 (2004) 1–7. [14] R.D. Shannon, Acta Crystallogr. A Found. Crystallogr. 32 (1976) 751–767.

[15] P. Loiseau, D. Caurant, J. Nucl. Mater. 402 (2010) 38–54. [16] G.K. Liu, V.V. Zhorin, M.R. Antonio, S.T. Li, C.W. Williams, L. Soderholm, J. Chem. Phys. 112 (2000) 1489. [17] P. Loiseau, D. Caurant, N. Baffier, L. Mazerolles, C. Fillet, J. Nucl. Mater. 335 (2004) 14–32. [18] X. Deschanels, S. Peuget, J.N. Cachia, T. Charpentier, Prog. Nucl. Energy 49 (2007) 623–634. [19] Y.H. Zhang, A. Navrosky, H. Li, L.Y. Li, L.L. Davis, D.M. Strachan, J. Non-Cryst. Solids 296 (2001) 93–101. [20] G.M. Tao, H.T. Guo, L. Feng, M. Lu, W. Wei, B. Peng, J. Am. Ceram. Soc. 92 (2009) 2226–2229. [21] R.K. Brow, J. Non-Cryst. Solids 263&264 (2000) 1–28. [22] C. Ivascu, A. Timar Gabor, O. Cozar, L. Daraban, I. Ardelean, J. Mol. Struct. 993 (2011) 249–253. [23] A.M.B. Silva, R.N. Correia, J.M.M. Oliveira, M.H.V. Fernandes, J. Eur. Ceram. Soc. 30 (2010) 1253–1258. [24] A. Moguš-Milanković, A. Šantić, S.T. Reis, K. Furić, D.E. Day, J. Non-Cryst. Solids 351 (2005) 3246–3258. [25] G.S. Henderson, R.T. Amos, J. Non-Cryst. Solids 328 (2003) 1–19. [26] O. Cozar, D.A. Magdas, L. Nasdala, I. Ardelean, G. Damian, J. Non-Cryst. Solids 352 (2006) 3121–3125. [27] R.K. Brow, D.R. Tallant, S.T. Myers, C.C. Phifer, J. Non-Cryst. Solids 191 (1995) 45–55. [28] G. Srinivasa Rao, B.K. Sudhakar, H.N.L. Prasanna, V. Devasahayam, N.R.K. Chand, Mater. Lett. 65 (2011) 378–380. [29] S. Rani, S. Sanghi, Anshu, A. Agarwal, N. Kishore, V.P. Seth, J. Phys. Chem. Solids 69 (2008) 1855–1860. [30] S. Rani, S. Sanghi, A. Agarwal, V.P. Seth, Spectrochim. Acta A 74 (2009) 673–677. [31] R. Ramesh, E. Nestor, M.J. Pomeroy, S. Hampshire, J. Eur. Ceram. Soc. 17 (1997) 1933–1939. [32] A. Bertoluzza, M.A. Battaglia, R. Simoni, D.A. Long, J. Raman Spectrosc. 14 (1983) 178–183. [33] G.B. Rouse Jr., P.J. Miller, W.M. Risen, J. Non-Cryst. Solids 28 (1978) 193–207. [34] J. Koo, B.-S. Bae, H.-K. Na, J. Non-Cryst. Solids 212 (1997) 173–179. [35] K. Meyer, J. Non-Cryst. Solids 209 (1997) 227–239. [36] Y.M. Lai, X.F. Liang, G.F. Yin, S.Y. Yang, J.X. Wang, H.X. Zhu, H.T. Yu, J. Mol. Struct. 1004 (2011) 188–192. [37] J.M. Cole, E.R.H. van Eck, G. Mountjoy, R. Anderson, T. Brennan, G. Bushnell-Wye, R.J. Newport, G.A. Saunders, J. Phys. Condens. Matter 13 (2001) 4105–4122. [38] J.S. Wu, J.F. Stebbins, J. Non-Cryst. Solids 362 (2013) 73–81. [39] U. Hoppe, G. Walter, D. Stachel, A.C. Hannon, Z. Naturforsch, A 50 (1995) 684–692. [40] B. Rufflé, S. Beaufils, Y. Délugeard, G. Coddens, J.P. Ambroise, E. Guéguen, R. Marchand, Mater. Res. Soc. Symp. 407 (1996) 155. [41] U. Hoppe, R.K. Brow, D. Ilieva, P. Jóvári, A.C. Hannon, J. Non-Cryst. Solids 351 (2005) 3179–3190.