Structural study of complex tellurite glasses

Structural study of complex tellurite glasses

J O U R N A L OF ELSEVIER Journal of Non-Crystalline Solids 192& 193 (1995) 53-56 Structural study of complex tellurite glasses S. Neov a,*, I. Ger...

247KB Sizes 2 Downloads 31 Views

J O U R N A L OF

ELSEVIER

Journal of Non-Crystalline Solids 192& 193 (1995) 53-56

Structural study of complex tellurite glasses S. Neov a,*, I. Gerasimova a, V. Kozhukharov b, p. Mikula c p. Lukas c a Institute for Nuclear Research and Nuclear Energy, Bulgarian Academy of Sciences, 1784 Sofia, Bulgaria b Higher Institute of Chemical Technology, 1756 Sofia, Bulgaria c Nuclear Physics Institute, 250 68 Rez, Czech Republic

Abstract The short-range atomic order in multicomponent tellurite glasses, containing 80 tool% TeO 2 and 20 tool% CsC1, MnCI2, FeCl z and FeC13, has been studied by the neutron scattering method. Despite the easy deformation of the basic building units of these glasses, the TeO4 polyhedra, the first coordination maximum of the radial distribution functions (RDF) remains well separated. The coordination number of Te is four for all of the compositions studied. For the glasses with either of the 3d metal chlorides as a modifier, the O-O distribution undergoes considerable changes during the transition to the vitreous state. The interpretation of the experimental RDF has been carried out by comparison with model distribution functions composed of the experimental RDF for pure TeO2 glass and crystal-like RDF for the modifier.

1. Introduction Experimental data concerning the glass-formation region (GFR) of binary tellurite systems containing 3d transition metal oxides were published in Ref. [1], while data indicating the wide glass-formation region observed in some tellurium oxide-metal halide and transition metal halide systems was reported later [2]. This work involves a structural investigation of tellurite glasses with composition 80 mol% TeO 2 and 20 mol% TCI, where T represents the cations Mn(II), Fe(II), Fe(III) and Cs(I). The compositions chosen are situated in the middle of the GFR and close to the eutectic point at 18 _ 5 tool% TC1 n. The aim is to establish the influence of the chlorine-containing modifier on the short-range order (SRO) of a highly covalent non-crystalline random network, such

* Corresponding author. Tel: + 359-2 75 8032. Telefax: + 359-2 75 5019. E-mail: [email protected]

as the tellurite matrix, and to clarify the SRO in TeO2-rich mixed oxy-halide tellurite glasses by comparison with diffraction data obtained for 'pure' TeO 2 glass.

2. Experimental and data processing Raw materials (TeO2, MnCI2, FeC12, FeC13 and CsC1) of pro analysis purity were used. 30 g samples were synthesized, using the procedure described in Refs. [2-4], and tested for crystallinity by X-ray diffraction. The samples were not subjected to any thermal treatment following their synthesis. The neutron structure factors, S(Q), where Q is the magnitude of the scattering vector, were obtained using a two-axis diffractometer at the IRT-2000 reactor in Sofia. The incident wavelength, A, was 0.1061 nm, giving a range of Q of 4-100 nm 1 Data were been obtained with the TKSN-400 spectrometer at the Nuclear Physics Institute in Rez-Prague (A =

0022-3093/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved

SSDI 0 0 2 2 - 3 0 9 3 ( 9 5 ) 0 0 3 3 1 - 2

S. Neov et. al. /Journal of Non-Crystalline Solids 192& 193 (1995) 53-56

54

0.083 nm, Q = 5.0-102.5 nm-1). The experimental neutron interference functions were corrected for multiple and inelastic scattering, absorption and the vanadium container scattering as described in Ref. [5].

450

1

356

2

i

8

3

3. Results

L

The measured neutron structure factors, S(Q), for the glasses are presented in Fig. 1. Curves 2 and 3 posses four broad maxima at 20.5, 29.5, 44 and 73 nm-1; a newly formed peak appears at 17 nm -1 in curves 1 and 4. Fig. 2 shows a comparison of the experimental RDFs with the RDF of an almost pure T e e e glass sample (containing 2 mol% P205) [6]. From the experimental RDF curves, it follows that there is no difference in the position of the first peak, P1, at 0.195 nm for the pure T e e 2 glass sample and samples 2, 3 and 4. A shift to a lower value of R (R1 = 0.192 nm) is however seen for sample 1. The second RDF peak, P2, located at 0.285 nm for pure T e e 2 glass and sample 4 exhibits changes which are dependent on the chemical composition; instead of the well-resolved peak Pe, the RDFs of samples 1, 2 and 3 posses a broad double maximum in the region

-5~.

' [~.11

k

.4 R (rim)

0 . I3

, 0,12

OI

400

..

A :" ~ " " \

-

7

,"i ".,,

P4

co

/

c

z

-/

\

, "

\

~/

-

z

x

200

:".

1

0 . ,7

see

E

2

i

-

Ps "

0.1B - - 0.16

of 0.35 > R > 0.25 nm. The peaks P3 and P4 at 0.38 and 0.46 nm, respectively, are typical for all of the glasses investigated. Fig. 3 illustrates the correlation between the experimental RDF curves and those for an additive model. The model RDFs are calculated on the basis of the experimental RDF of pure T e e 2 glass [6] and 500 -

2.0

i

Fig. 2. Radial distribution functions for: (1) 80Tee 2 +20MnCI2; (2) 80Tee 2 +20FeCI2; (3) 80Tee 2 +20FeCI 3 and (4) 80Tee 2 + 20CsC1 glasses compared with that for pure T e e z glass [6].

3.0

ky\

,4

~

A

~-'~

V

3 84 l e e

-

,

e'eo

i

le

,

~k

2

i

se

,

i

4~

,

~

s

i

6e

,

I

7e

,

i

ee

,

i

,

L

mode exp.

,

9e i~01

e

O ( nm - 1 ) Fig. 1. Neutron structure factors for glasses from the systems T e O 2 - T C I , , with TCI, = (1) MnC12, (2) FeC12, (3) FeC13 and

(4) CsC1.

-



e.1

e2

0.4

o.s

e.6

e.7

R (nm) Fig. 3• Experimental and model RDFs for the glasses studied: (1) 8 0 T e e 2 +20MnCI2; (2) 8 0 T e e 2 +20FeC12; (3) 80Tee 2 + 20FeC13 and (4) 8 0 T e e 2 + CsC1.

s. Neov et. al. /Journal of Non-Crystalline Solids 192&193 (1995) 53-56

crystal-like distribution of the corresponding modifier taken according to their molar concentration [3,5]. A comparison with the second peak, P2, of the model curve shows deviations for the glass sample containing 3d metal halide (Fig. 3(a),-(c)). A strictly additive SRO model can be accepted only for sample 4 (CsCl) (Fig. 3(d)).

4. Basic stereochemicai characteristics of the polyhedra The coordination state of tellurium atoms in glassy networks has been studied in detail [7,8]. The basic structural unit of tellurite glasses, the TeO 4 polyhedron, is very variable and can easily be attacked by the modifier. This assumption is in accordance with Gilespy's hypothesis [9] for the mechanism of TeO 4 polyhedron deformation and the transition from 4 to 3 + 1 to three-fold Te coordination. The structural chemistry of MnCI 2 is similar to that of FeC12. Each cation is sixfold coordinated in relation to the chlorine atoms and the average distance is RMn_CI = 0.221 nm. The arrangement of the MnCI 6 octahedra is the same as in CdC12 [10]. The structure of FeC13 is similar to the layer structures of BiI3, YC13 and ScC13 [10] with an average distance RFe_Cl = 0.214 nm. According to Wells [10], each cation in CsC1 is eightfold-coordinated with respect to the CI ions, which are situated on the vertices of a cube. These structural units have an average interatomic distance, Rcs_cl, of 0.356 nm.

55

According to the model used, the double RDF peak at 0.25-0.35 nm, P2, is formed by six O-O, two T e - O and six Fe(Mn)-chlorine distances. There is a good correlation in the position and amplitude of P2 between the experimental and model RDFs for sample 4. A 'smearing' effect is observed for the tellurite glasses containing MnC12, FeCI 2 and FeC13. The effect of the added crystal-like metal-chlorine distribution is insufficient to explain the observed changes. At this composition, the oxygen network is strongly influenced by the chemical interaction with the modifier. The O - O and Te-(second coordination sphere) O distances are dispersed over the region between 0.25 and 0.35 nm. From previous studies [7,11-13], it is known that the T e - T e interatomic distances are very variable and very sensitive to the vitreous transition from the melt. The first T e - T e coordination maximum is located at 0.385 nm and its amplitude is always very small. The experimental RDFs for the mixed oxyhalide tellurite glasses possess a peak, P3, at this distance (see Figs. 2 and 3). In the context of previous investigations, it can be concluded that this peak is mainly due to the interatomic distances for the halide modifier. The peak, P4, at 0.48 nm is formed mainly by the O - O distances. It is found in all of the model and experimental RDF curves. This peak is strongly broadened only in the model RDF for sample 4. In the region of peak P5 there are many distances (e.g., O-O, Mn-C1, C1-C1, Cs-Cs, Fe-Fe) which are superimposed and any interpretation will be quite arbitrary.

5. Discussion 6. Conclusion The RDFs presented in Figs. 2 and 3 and the structural characteristics of the TeO: and metal halides indicate that the peak, P1, at 0.195 nm is exclusively the result of the distribution of four T e - O distances in the T e O 4 polyhedra. P1 is stable and only a small shift to shorter distance can be seen for sample 1. The well-expressed minimum between P1 and P2 shows that there is no major TeO 4 deformation in mixed oxy-halide binary tellurite glasses. The T e - O (axial) bonds in the TeO 4 bipyramids are not so variable as for some tellurite glasses with oxide modifiers [11-13].

The analysis of the RDFs obtained by neutron scattering experiments leads to the following conclusions concerning the SRO in the glasses studied. (1) The fourfold coordination of the Te atoms is conserved for all of the compositions studied. (2) In the T e O 2 - M n C I 2, TeO2-FeC12 and TeO2-FeC13 glasses, the oxygen network is strongly influenced by the chlorine-containing modifiers. (3) A strictly additive model of the modifier incorporation in the TeO 2 matrix is only applicable for the TeO2-CsC1 glass short-range order.

56

S. Neov et. al. /Journal of Non-Crystalline Solids 192&193 (1995) 53-56

(4) In the case of the isostructural modifiers MnC12 and FeCI 2, the SRO in the corresponding TeO 2 glasses is similar.

This work was sponsored by the National Science Fund of Bulgaria, Research Contract F-98.

References [1] V. Kozhukharov, M. Marinov and I. Gugov, J.NonCryst.Solids 28 (1978) 429. [2] V. Kozhukharov, M. Marinov and I. Gugov, Abstracts of 1st Int. Symp. on Halide and Other Non-Oxide Glasses, March 1982, Cambridge, UK, Paper 4 (1982). [3] S. Neov, V. Kozhukharov, I. Gerasimova and B. Sidzhimov, J.Non-Cryst. Solids 126 (1990) 255.

[4] H. Burger, W. Vogel and V. Kozhukharov, Infrared Phys. 25 (1985) 395. [5] S. Neov, V. Kozhukharov, Ch. Trapalis and P. Chieux, Phys. Chem. Glasses 36 (2) (1995) 89. [6] S. Neov, S. lshmaev and V. Kozhukharov, these Proceedings, p. 61. [7] S. Neov, I. Gerasimova, K. Krezhov, B. Sidzhimov and V. Kozhukharov, Phys. Status Solidi (a)47 (1978) 743. [8] V. Kozhukharov, S. Neov, I. Gerasimova and P. Mikula, J. Mater. Sci. 21 (1986) 1707. [9] R. Gilespy, Geometry of Molecules (Mir, Moscow, 1975) p. 98. [10] A. Wells, Structural Inorganic Chemistry, 4th Ed. (Pergamon, Oxford, 1975) pp. 142, 199, 348-375, 542. [11] S. Neov, I. Gerasimova, V. Kozhukharov and M. Marinov, J.Mater. Sci. 15 (1980) 1153. [12] S. Neov, I. Gerasimova, B. Sidzhimov, V. Kozhukharov and P. Mikula, J.Mater. Sci. 23 (1989) 749. [13] V. Kozhukharov, H. Burger, S. Neov and B. Sidzhimov, Polyhedron 5 (1986) 771.