Electron paramagnetic resonance and microhardness of binary vanadium tellurite glasses

Electron paramagnetic resonance and microhardness of binary vanadium tellurite glasses

J O U R N A L OF Journal of Non-Crystalline Solids 146 (1992) 261-266 North-Holland NON-CRYS LSESOLIDS Electron paramagnetic resonance and microhar...

394KB Sizes 0 Downloads 13 Views


Journal of Non-Crystalline Solids 146 (1992) 261-266 North-Holland


Electron paramagnetic resonance and microhardness of binary vanadium tellurite glasses N e e n a Chopra and Abhai Mansingh Department of Physics and Astrophysics, University of Delhi, Delhi- 110007, India

Pawan M a t h u r Department of Chemistry, University of Delhi, Delhi - 110007, India Received 21 May 1991 Revised manuscript received 19 May 1992

Electron paramagnetic resonance (EPR) spectra and microhardness measurements of binary vanadium tellurite (V205TeO 2) glasses have been taken. The spectra are typical of V 4+ ions present in vanadyl (VO 2+) form in the glass. Sufficiently resolved spectra are obtained only for glasses with low concentrations of V205 ( < 30 mol% V205). Hyperfine lines begin to overlap with the increase in V205 and finally collapse for higher concentrations of V205. EPR parameters have therefore been evaluated only for low concentrations of V205. Microhardness of V2Os-TeO 2 glasses decrease with increasing V205 content. The microhardness of 60VzOs-40TeO 2 glass is less than that of 60MoO3-40P205 glass reported by Selvaraj and Rao. Both microhardness and EPR studies indicate that the covalency of V - O bonds decreases with in'creasing V205 content.

1. Introduction

Electron paramagnetic resonance (EPR) is a powerful spectroscopic technique which can yield detailed information about the properties of solids on an atomic scale. EPR spectra of binary transition metal oxide (TMO) glasses have been reported by several workers [1-12]. However, very few authors [1,2] have attempted to analyze the chemical bonding parameters from the spectra of these glasses. Muncaster and Parke [1] have made a detailed investigation of the EPR spectra of TMO glasses as a function of composition. They reported an increase in covalency of these glasses with increasing TMO content. However, the microhardness data of TMO glasses reported by Selvaraj

Correspondence to: Dr N. Chopra, Department of Physics and Astrophysics, University of Delhi, Delhi-ll0007, India.

and Rao [13] indicate a decrease in covalency with increasing TMO concentration. Muncaster and Parke [1] studied in V 2 O s - T e O 2 glasses while Selvaraj and Rao [13] studied M o O 3 - P 2 0 5 glasses. The difference in the results of the EPR and microhardness may be due to either a difference in TMO and glass-former or to different preparation techniques. Data of EPR spectra and microhardness on a set of samples prepared in the same batch are expected to provide correlation of spectra with microhardness data. In the present paper, we report EPR spectra and microhardness of V2Os-TeO 2 glasses.

2. Glass preparation and experimental details

Binary vanadium tellurite glasses were prepared by melt quenching. Appropriate amounts of powders of V205 (Fluka) and TeO 2 (Aldrich

0022-3093/92/$05.00 © 1992 - Elsevier Science Publishers B.V. All rights reserved

N. Chopra et al. / EPR and microhardness of V2Os-TeO 2 glasses


Chemical Company) were accurately weighed and mixed in a quartz tube. The tube was suspended in a furnace to heat the mixture to 930°C for 1 h in air. The glasses were prepared by pouring the melt onto a precooled metal plate. The amorphous state was confirmed by X-ray analysis. Xray diffraction of the polished samples was taken with Philips (Model p w - l l 3 0 / 0 0 ) X-ray diffractometer in which Cu K s (h = 1.5405 A) radiation was used as a source. E P R spectra of powdered glass samples were recorded at 120 K and at room temperature, on a J E O L JES-FE 3 × G ESR spectrometer operating in the X-band. Microhardness numbers of T M O glass with finely polished surfaces were determined using Zwick 3212 hardness tester. The measurements were made at room temperature. A uniform load of 0.3 kg was applied for 15 s. A number of indentations were made on each glass sample to obtain meaningful averaged results. The Vickers hardness number, H v , was calculated using the formula H v = 1 . 8 5 4 F / d 2 where F is the test load and d is diagonal length of indentation.

? b-



O <2

X n


3. Results

The microhardness of binary vanadium tellurite glasses is given in table 1. Electron paramagnetic resonance spectra of binary vanadium tellurite (V2Os-TeO2) glasses, recorded at room temperature, are shown in fig. 1. The figure shows that the spectra are typical of V 4 + ions present in vanadyl (VO 2÷) form in the glass. It also shows that resolved hyperfine spectra were obtained only for glasses with lower concentrations of V205

Table 1 Microhardness numbers of vanadium tellurite glasses Sample


Microhardness (kg mm -2)


10V205-90TeO 2 20V205 -80TeO 2 30V205 -70TeO 2 40V205-60TEO 2 50V205 -50TeO z 60V205-40TEO 2

250.2 240.6 236.8 225.2 213.7 207.0

+ 1.4 + 3.6 + 3.3 +_3.7 + 3.9 + 2.9



3500 GAUSS



Fig. 1. EPR spectra of vanadium tellurite glasses recorded at room temperature (gain = 1 × 10).

(up to 30 mol% V205). Hyperfine lines begin to overlap with an increase in V205 and finally collapse for higher V205 concentrations. No resolved structure is found for glasses containing 40 and 50 mol% V205. This lack of structure is attributed to the increase in V 4+ ion concentration with composition. It may however be noted that Muncaster and Parke [1] obtained resolved hyperfine spectra of V2Os-TeO 2 glasses even for higher concentration of V205. This resolution was achieved by melting the glasses under more

N. Chopra et al. / EPR and microhardness of V2Os-TeO 2 glasses


















Fig. 2. EPR spectrum of 50V2Os-50TeO 2 glass recorded at 120 K (gain = 6.3 × 1).

oxidizing conditions to reduce the concentrations of V 4+ ions. Figure 2 shows the EPR spectrum of 50V20 550TeO 2 glass recorded at 120 K. It is observed that hfs lines are unresolved even at such a low temperature and the spectrum is similar to that recorded at room temperature for this composition.

4. Discussion Figures 1 and 2 indicate no change in the peak to peak line width of the spectrum with temperature. This is expected for line broadening due to spin-spin interaction since the corresponding term in the spin Hamiltonian is not temperaturedependent. Friebele et al. [11] also observed that the line width of vanadium phosphate glasses remain constant throughout this temperature range. Spin-spin interactions can be either magnetic dipolar or exchange. Since the effects due to these two interactions cannot be resolved, it is difficult to determine the predominant cause of


the disappearance of hfs lines for higher concentrations of V2Os in our glasses. Probably both mechanisms contribute to it. It may be noted that Muncaster and Parke [1] attributed the disappearance of the hfs lines to the dipolar broadening mechanism. Lynch and Sayer [14], however, interpreted their results on the basis of exchange narrowing of the resonance line in the disordered system. Horvfith et al. [6] studied vanadium phosphate glasses and suggested that the dominating interaction must be long range superexchange at low V 4+ concentrations ( < 2.3 × 1021 spin cm -3) and magnetic dipolar at high V 4+ concentration. Following this discussion, we attribute our results primarily to exchange interactions as the V 4+ ion concentration of x V 2 O s - ( 1 0 0 - x ) T e O 2 (x = 10, 20, 30, 40, 50) glasses is ~ 10 20 spins cm -3. Since V 4+ ions are at some distance from each other, the interaction must be superexchange. The intensity (peak to peak height) of the unresolved spectrum recorded at 120 K is observed to be greater than that recorded at room temperature, This difference suggests that the coupling is between ions in different valence states. It is therefore concluded that superexchange interaction of the form V 4 + - O - V s+ may be the cause of disappearance of hfs lines in the spectra.

4.1. Variation of EPR parameters with composition Electron paramagnetic resonance spectra obtained for binary vanadium tellurite glasses can be described by an anisotropic axial spin Hamiltonian corresponding to a 3d electron localized on a single vanadium ion Slv (I = 7/2), given by

x . = gj#HzSz + g ~#(H~S~ + H , s , )

+AiiIzS ~ + A ~ (SxI x + SyIy).


The solution of the above Hamiltonian is given by Hil(mi) = Hql(0) - A l i ( m i )

- [ A 2 /2HH(O)][I( I + 1) - m 2]


N. Chopraet al. / EPR and microhardnessof V2Os-TeO2 glasses

264 and

H A(mI) = H A(0) - - A A ( m I ) - [ ( A ~ + A Z ) / 4 H A (0)] ×[I(I+

1 ) - m 2]


for the parallel and perpendicular hyperfine lines, respectively. Here, Hll(0 ) = hu/(gll/3);

H± (0) = h u / ( g A/3);

m I is the nuclear magnetic quantum number of the vanadium nucleus taking the values ___7 / 2 , +_5/2, +_3/2, +_1/2;/3 is the Bohr magneton, 1, is the spectrometer frequency; gll and g A are the parallel and perpendicular components of the anisotropic g tensor; and All and A A are the parallel and perpendicular components of the hyperfine tensor in magnetic field units. An iteration procedure similar to that followed by Muncaster and Parke [1] and Bandyopadhyay [15] has been used for calculating the E P R parameters from the spectra shown in fig. 1. These parameters could be calculated only for glasses with low V205 (_< 30 mol%) concentrations since the spectra for the glasses with high V205 content are not sufficiently resolved. The values of the spin Hamiltonian parameters (gll, g A, All and A A) thus obtained are given in table 2. The 60% confidence limits determined by normal statistical methods are within _ 0.0005 for the g-values and +_0.5 × 10 -4 cm-1 for the A-values. The values of gll and g A depend critically on the type of distortion around the V 4+ ions. The axial distortion can either be threefold (trigonal) or fourfold (tetragonal). In the presence of trigonal symmetry, Gladney and Swalen [16] determined that either a singlet or a doublet can be the ground state depending on the type of distor-

tion. According to their results, a singlet ground state requires gll >> g A and the doublet ground state determines that the value of gll and g A are near zero. An octahedral site of V 4+ ion with a trigonal distortion can thus be excluded, since the values of gll and g A given in table 2 do not satisfy the above conditions. Since grl < g A < ge ( = 2.0023) and All > A j_ for an octahedral site with tetragonal compression [17], the V 4+ ions in the present glasses appear in tetragonally distorted octahedral sites. This assignment is in agreement with the results obtained by others [1-4]. Table 2 shows that both gll and gA decrease with increasing V20 5 content. Since one of the effect of covalency is to make the g-value more nearly equal to ge ( = 2.0023), the above results suggest that the covalent bonding between the vanadium ion and the surrounding ligands decreases with increasing VzO 5. Parameter Agll/Ag A (Agll=ge-gll; A g A = g e - - g A; g~ = 2.0023) measures the tetragonality of V 4÷ sites and is given in table 2. A decrease in ( A g J A g A) with increasing V205 is observed. This decrease indicates an improvement in the octahedral symmetry of the V 4+ site with increasing V205 content. The spin Hamilton±an parameters are related to the chemical bonding parameters by the following equations [1,15]: glr-- gel 1 - ( 4 a 2/32A/E 2)],


g A=ge[l-(y2/32A/gl)],






~ A g A]"

Here k is proportional to the isotropic Fermi

Table 2 ESR parameters of vanadium tellurite glasses Sample



10V205-90TeO 2 20VeOs-80TeO z 30V2Os-70TeO z







(×10 -4 cm-~)

(×10 -4 cm -1)


(XI0 -4 cm-~)

(>(10-4 em-I)

62.28 57.66 54.72

4.4 4.34 2.9

77.6 80.6 81.9

29.98 29.68 29.45

1.8939 1.9778 169.9 1.8864 1.9676 167.9 1.8853 1.9615 166.1

1-a 2

1-3 '2

0.1031 0.0411 0.032

0.4103 0.1648 0.018

N. Chopra et aL / EPR and microhardness of ~Os-Te02 glasses

contact interaction, p = 2/30/3Ny0(dxylr -3 I dxy), /3N is the nuclear magneton, 70 is the gyromagnetic ratio, E 1 and E 2 are the energies of transitions b 2 ~ e* and b 2 ~ b~, respectively./32 is the measure of the in-plane w-bonding with the equilateral ligands and is assumed to be equal to 1 for many oxide glasses containing VO 2÷ ion [15]. An estimate of covalency of the w-bonding between the V 4+ ion and the vanadyl oxygen is given by the expression ( 1 - 72) and that of o- (Sigma) bonding with equilateral ligands is given by (1 a 2) [1,15]. The values of (1 - a 2) and (1 - 72) are given in table 2. It may be noted that the magnitudes of ( 1 - a 2) and ( 1 - y2) depend on the assumed values of the energy of transitions E1 (12000 cm -1) and E 2 (16500 cm -1) and the spin orbit coupling constant A (249 cm-t). Therefore, only the trends in the variation of magnitude of bonding parameters with composition are discussed. We observe a decrease in ( 1 - 3,2) and ( 1 - a 2) with increase in V205, indicating a decrease in the covalency of respective bonds with increasing VzO 5. Equations (6) and (7) can be split into their component parts: All= _p[(g)f12 + Agll + (3) Ag ±] -p/32k

=A;j -




=A'. -p132k."

Ag±]-p/32k (9)

In eqs. (8) and (9), the first term is the contribution of the 3dxy electron to the hyperfine structure and the second term, p/32k, is the anomalous contribution of the S electrons. The values of IAIII and I A'. I were calculated using the above equations and are also given in table 2. It is noted that I A I and I A± [ decrease, ] All l increase while A' l remains almost constant with increasing V205. The variation of the EPR parameters and of the degree of covalency with composition are not consistent with those reported by Muncaster and Parke [1]. It may be pointed out that covalency is only a small fraction of the bonding and the major contribution is from ionic bonding. The changes in the covalency with composition may be influenced by impurities present in the sam-


25°I ~'..,~

2~o~ - \ J











Mo~e °/0 V205

Fig. 3. Variation of microhardness with composition for vanadium teUurite glasses.

ples. Muncaster and Parke [1] have not specified the purity of their samples. The preparation condition of the glass may also be responsible for the differences. Muncaster and Parke's data are based on the glasses prepared in oxygen ambient, while the presently investigated glasses have been prepared in air atmosphere. It is, therefore, difficult to ascertain the exact cause of the variance in the reported results of Muncaster and Parke and present studies.

4.2. Microhardness The variation of microhardness with composition is shown in fig. 3. The microhardness decreases with increasing V205 content. Selvaraj and Rao [13] suggested that covalently bonded structures are more difficult to deform than the ionically bonded structure. Hence the decrease in microhardness may be related to a decrease in covalency. Therefore, covalency of V2Os-TeO 2 glasses decreases with increasing V205 content. This is in accordance with the EPR results discussed above. It may be mentioned that Selvaraj and Rao [13] also obtained similar results for MoO3-P205 glasses. The microhardness value of the 60VaOs-40TeO 2 glass is however lower than that of 60MoO3-40P205 glass (403 kg mm -2) [13].


N. Chopra et al. / EPR and microhardness of V2Os-TeO 2 glasses

5. Conclusions

(1) Superexchange interaction of the form V4+-O-V5+ results in the disappearance of hfs lines. (2) V 4+ ions exist in tetragonally distorted octahedral sites in vanadium tellurite glasses. (3) Tetragonal distortion decreases with increasing V205 content. (4) Both microhardness and EPR studies indicate a decrease in covalency of the V - O bonds with increase in V205. The authors wish to thank Drs Ajay Dhar and Rajeev Chopra National Physical Laboratory (Delhi) for microhardness measurements of vanadium tellurite glasses. Financial assistance for this work from the Council of Scientific and Industrial Research (India) to one of the authors (N.C.) is gratefully acknowledged.

References [1] R. Muncaster and S. Parke, J. Non-Cryst. Solids 24 (1977) 399.

[2] AI. Nicula and E. Culea, Phys. Status Solidi (b)142 (1987) 265. [3] G,F. Lynch, M. Sayer and S.L. Segel, J. Appl. Phys. 42 (1971) 2587. [4] V.M. Nagiev, Sov. Phys. Solid State 7 (1966) 2204. [5] AI. Nicula, E. Culea and I. Lupsa, J. Non-Cryst. Solids 79 (1986) 325. [6] L.I. Horv~th, I. Geresdi and T. Sz6r6nyi, J. Non-Cryst. Solids 70 (1985) 429. [7] A.A. Hosseini and C.A. Hogarth, J. Mater. Sci. Lett. 7 (1988) 593. [8] E.E. Khawaja, M. Sakhawat Hussain, M.A. Khan and J.S. Hwang, J. Mater. Sci. 21 (1986) 2812. [9] E.E. Khawaja, F. Tegally, J.S. Hwang, A.S.W. Li and A.A. Kutub, J. Mater. Sci. 20 (1985) 3074. [10] M. Elahi, M.H. Hekmat-Shoar, C.A. Hogarth and K.A.K. Lott, J. Mater. Sci. 14 (1979) 1997. [11] E.J. Friebele, L.K. Wilson and D.L. Kinser, J. Am. Ceram. Soc. 55 (1972) 164. [12] V.K. Dhawan, A. Mansingh and M. Sayer, J. Non-Cryst. Solids 51 (1982) 87. [13] U. Selvaraj and K.J. Rao, J. Non-Cryst. Solids 72 (1985) 315. [14] G.F. Lynch and M. Sayer, J. Phys. C: Solid State. Phys. 6 (1973) 3661. [15] A.K. Bandyopadhyay, J. Mater. Sci. 16 (1981) 198. [16] H.M. Gladney and J.D. Swalen, J. Chem. Phys. 42 (1965) 1999. [17] H.G. Hecht and T.S. Johnston, J. Chem. Phys. 46 (1967) 23.