Section 17. Chalcogenides. I. Processing, composition, structure and defects
Self-organization and anisotropy in amorphous chalcogenides

https://doi.org/10.1016/j.jnoncrysol.2005.11.140Get rights and content

Abstract

The intermediate, self-organized phase, in the As–Se system has been modeled on the basis of clusters with one or more selenium–selenium bridges. The X-ray diffraction patterns of the glass models at the limits of the ‘floppy phase-stressed rigid phase’ window have been calculated. The structural and optical anisotropy in a matrix of clusters containing Se–Se bonds is supposed to be a consequence of breaking and re-linking the broken clusters in the direction of the polarized light beam that acts upon the chalcogenide material.

Introduction

The concept of global connectedness has played a useful role for understanding the network glasses and amorphous films. The idea emerged at the middle of the last century when the chalcogenides were examined as a function of composition [1].

Later, it was demonstrated that the nearest-neighbour valence interactions can serve as Lagrangian constraints [2]. This led to the prediction of a floppy-to-rigid phase transition, when the number of constants per atom increases to three. Recently, Boolchand [3] revealed not just one transition but two rigidity transitions in network glasses. The two transitions define the limits of an intermediate phase, called the Boolchand phase. The first transition is between the floppy phase (the elastic under constrained phase characterized by the mean atomic coordination r¯<2.40) and the rigid but stress-free Boolchand phase. The second transition is between the Boolchand phase and the stressed rigid phase. In strictly random networks the two transitions are coincident giving rise to only one elastic phase transition as predicted by Phillips [2] and Thorpe [4].

The window opened between the two elastic transitions was experimentally demonstrated and it is characteristic to chalcogenides. Since the discovery of the Boolchand phase many papers were published [5], [6], [7], [8], [9], [10], but, nevertheless, the structure of this phase remained a challenging problem for the physics of glasses. Lucovsky [11] has shown that self-organization in oxide and chalcogenide thin films based on the intermediate phase provides the basis for device applications. Strand [12] pointed out the remarkable successes in creating the OUM (Ovonic Universal Memory) with amorphous chalcogenide films.

In this paper, we report our results on the modeling of the intermediate (Boolchand phase) in the As–Se glass and we show that specific cluster units could be the most important ingredients that govern the structure.

Section snippets

Modeling

Twenty five years ago Phillips [13] demonstrated that molecular models for AsxS1−x and AsxSe1−x glasses near x = 0.4 could explain the anomalies in the temperature dependence of diffraction data.

The first observation of a microscopic anomaly was the first sharp diffraction peak in glassy As2Se3. This peak in As–S films was explained as the signature of spheroid clusters (e.g. As4S4), which are stable as free molecules; annealing of the films reduces the peak intensity, and suggests that the

Discussion

The intermediary self-organized phase in As–Se glass seems to be based on special clusters.

The Se–Se bridge content extends from one Se–Se bridge (minimum possible is zero bridge, that corresponds to As2Se3 limit) to six Se–Se bridges (maximum possible is nine Se–Se bridges, that corresponds to the composition As2Se6).

It is interesting to note that in the case of the phosphorus chalcogenide PxSe1−x the extent of the Boolchand phase is from xc(1) = 0.28 to xc(2) = 0.40. If we consider the same

Conclusions

The intermediate phase in As–Se amorphous chalcogenides is suggested to be based on specific molecular clusters, more or less polymerized. The anisotropy in a matrix of Se–Se bridged clusters can find a simple explanation on the basis of breaking and directionally linking of clusters. The network glass made by polymerization of the clusters with multiple Se–Se bridges gives rise to a rigid not stressed network, specific to the Boolchand phase.

Acknowledgement

The partial financing through the CERES Program (Project 3/117) is kindly acknowledged.

References (16)

  • J.C. Phillips

    J. Non-Cryst. Solids

    (1979)
  • M.F. Thorpe

    J. Non-Cryst. Solids

    (1983)
  • M.F. Thorpe et al.

    J. Non-Cryst. Solids

    (2000)
  • J.P. de Neufville et al.
  • P. Boolchand et al.

    J. Optoelectron. Adv. Mater.

    (2001)
  • M.V. Chubinsky et al.

    Curr. Opin. Solid State Mater. Sci.

    (2001)
  • D.G. Georgiev et al.

    Europhys. Lett.

    (2003)
  • S. Mamedov et al.

    J. Phys.: Condens. Matter

    (2003)
There are more references available in the full text version of this article.

Cited by (0)

View full text