| dc.description.abstract | Main Conclusions from the Present Studies
Alternate Methods for Pair Correlation Function
The search for alternate methods of obtaining the pair correlation function g(r)g(r)g(r) from the structure factor S(Q)S(Q)S(Q), especially avoiding the use of a modification function in the Fourier transformation, was shown to be essential. The one-dimensional Monte Carlo (MC) method appears to be one such alternative. The extrapolated region of S(Q)S(Q)S(Q) for GeSe2_22? was found to agree with the experimental data of Susman et al. regarding the positions of peaks and valleys. Moreover, the excellent agreement between the derived number density and the experimentally measured number density (within 0.5%) clearly demonstrates the superiority of this method over conventional techniques.
The observation of a distinct feature around 3.0 Å in g(r)g(r)g(r) obtained by the MC method, compared to the conventional method, also highlights its advantage. However, the algorithm used in the present investigation cannot account for the statistical noise present in the data.
Intermediate Range Order (IRO) in Ge–Se Glasses
The intermediate range order in Gex_xx?Se1?x_{1-x}1?x? glasses increases with increasing Ge content. The intensity of the First Sharp Diffraction Peak (FSDP) is maximum for x=0.33x = 0.33x=0.33, and the position of the peak shifts progressively to smaller QQQ values with increasing xxx. The coherence length was found to be maximum for x=0.33x = 0.33x=0.33 and x=0.4x = 0.4x=0.4. The increase in IRO with xxx suggests that it is associated with Ge–Ge correlations.
Basic Structural Units in Ge–Se Glasses
For x=0.1x = 0.1x=0.1 and x=0.2x = 0.2x=0.2, the basic building blocks are Se chain segments and Ge(Se1/2_{1/2}1/2?)4_44? tetrahedra.
At x=0.1x = 0.1x=0.1, these tetrahedra predominantly cross-link Se chains and are not interconnected.
At x=0.2x = 0.2x=0.2, some tetrahedra are interconnected, although most are still bridged by single Se atoms.
For x=0.33x = 0.33x=0.33, tetrahedral connectivity is complete, with Ge(Se1/2_{1/2}1/2?)4_44? being the basic units.
For x=0.4x = 0.4x=0.4, Ge2_22?(Se1/2_{1/2}1/2?)4_44? ethane-like molecules are the basic units. At this composition, Ge–Ge bonds are present in the glass, averaging one Ge–Ge bond per Ge site.
Edge-Sharing (ES) Sites
The fraction of edge-shared (ES) sites in GeSe2_22? glass appears to differ from that in the crystal. The fraction of ES sites increases with xxx. For x=0.4x = 0.4x=0.4, some Ge sites have two ES tetrahedra, meaning the Ge tetrahedron is completely interconnected in an ES configuration.
Reverse Monte Carlo (RMC) Simulation Studies
To confirm these results and understand the behavior of higher peaks, especially the peak around 4.7 Å, Reverse Monte Carlo (RMC) studies were carried out. The main features of the present simulation study are:
The RMC method, initially proposed by McGreevy and Pusztai, is not robust enough to generate the correct short-range order (SRO) of GeSe2_22? glass.
Molecular Dynamics (MD) simulation, even without a three-body term, is able to reproduce the correct SRO.
Incorporation of proper constraints in the RMC algorithm results in a match of the partial pair correlations with those obtained in MD simulation using a two-body potential.
The bond-angle distributions from the RMC study are in qualitative agreement with MD results using a two-body potential.
The RMC results produce a glass structure in which Ge(Se1/2_{1/2}1/2?)4_44? tetrahedra are more distorted. Further, the structure mimics a two-body effect rather than the non-central forces known to be present in covalent glasses.
The number of two-fold rings in the present study is higher compared to other reports in the literature.
About 9% of edge-shared Ge sites have two edge-shared neighbors. | |
