GeₓSe₍₁₋ₓ₎ and AsₓSe₍₁₋ₓ₎ glasses have found use in many technological applications due to their excellent rheological properties and their wide IR transparency window. However, the low glass transition temperatures of these glasses leads to large changes in their properties, due to structural relaxation, over the weeks and months subsequent to their fabrication. Thus, obtaining a more thorough understanding of structural relaxation and its relation to the structure, composition, and processing of these glasses is important in furthering their use. Structural investigations, using NMR and Raman spectroscopies, performed on the GeₓSe₍₁₋ₓ₎ family of glasses show that the structure of these glasses is composed of two distinct microdomains. One corresponds to a rigid GeSe₂-like domain and the other corresponds to a floppy Se domain. These results are compared to other existing structural models for GeₓSe₍₁₋ₓ₎ glasses. Enthalpy measurements on both GeSe₉ and GeSe₄ optical fibers which were aged up to five years demonstrate that both compositions undergo a large amount of enthalpy relaxation in this time period. Raman spectroscopy performed concurrently with enthalpy measurements on the same GeSe₉ and GeSe₄ fibers shows that one of the structural changes taking place within the glass network is the conversion of edgesharing to corner-sharing tetrahedra in the GeSe₂-like phase. Moreover, the rate at which this conversion takes place is shown to be similar to the rate of enthalpy relaxation, suggesting that this structural change is one of the main mechanisms for structural relaxation in GeₓSe₍₁₋ₓ₎ glasses. Implementation of the Tool-Narayanaswamy-Moynihan (TNM) model as a hybrid computer model allowing the prediction of the four relaxation parameters Δh*, log(A), x, and β via optimization of simulated and experimental data was accomplished. It was found that a multi-rate version of the TNM model, which obtains an average set of model parameters via optimization of multiple experimental thermal histories simultaneously, was able to predict relaxation parameters for AsₓSe₍₁₋ₓ₎ glasses within the 2.10 ≤ <r> ≤ 2.50 compositional domain, where <r> is the average bond coordination of the glass network as defined by the Phillips and Thorpe constraints model. Above <r> = 2.50, however, the model fails, due to a bimodal distribution of relaxation times within the glass structure contrary to the TNM model assumption of a unimodal distribution of relaxation times, thus rendering the model inapplicable. Muti-rate modeling of the GeₓSe₍₁₋ₓ₎ family of glasses was also attempted, however the TNM model also fails for this family of glasses due to the inherently bimodal distribution of relaxation times which arises from their bimodal structure.
| Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/202735 |
| Date | January 2011 |
| Creators | King, Ellen Anne |
| Contributors | Lucas, Pierre, Lucas, Pierre, Potter, B. G., Uhlmann, Don, Bureau, Bruno |
| Publisher | The University of Arizona. |
| Source Sets | University of Arizona |
| Language | English |
| Detected Language | English |
| Type | text, Electronic Dissertation |
| Rights | Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. |
Page generated in 0.003 seconds