Return to search

Electrical and spectroscopic studies of new silver gallium thiophosphate glasses

New silver gallium thiophosphate glasses in the Ag<SUB>2</SUB>S-Ga<SUB>2</SUB>S<SUB>3</SUB>-'P<SUB>4</SUB>S<SUB>8</SUB>' system were studied employing a.c. impedance, Raman and IR spectroscopy. 'P<SUB>4</SUB>S<SUB>8</SUB>' does not exist and is used so that the compositions can be conventionally expressed in terms of network modifier and network formers. The glasses were in fact prepared from the elements. The trends in conductivity of other series of glasses were also studied including glasses with added AgI and mixed Ag-Na glasses. These glasses are interesting in that the phosphorus is formally in a IV+ oxidation state whereas other thiophosphate glasses previously reported in the literature contain phosphorus in a V+ oxidation state. Also, Raman spectroscopy has shown that these glasses contain hypothiodiphosphate units (P<SUB>2</SUB>S<SUB>6<SUP>4-</SUB></SUP>) which could be described as octahedral (the P-P occupies the place of the central atom). The phosphate and thiophosphate glasses studied thus far, have had structures based on meta(thio)phosphate chains of corner-sharing tetrahedra. Therefore these hypothiodiphosphate glasses are different in several respects to other thiophosphate glasses. Raman and IR studies of the 0.33Ag<SUB>2</SUB>S-0.33Ga<SUB>2</SUB>S<SUB>3</SUB>-0.33'P<SUB>4</SUB>S<SUB>8</SUB>' glass (or AgGaP<SUB>2</SUB>S<SUB>6</SUB> as it is more usually referred to in this thesis) suggest that the glass is composed of linear chains of -(Ga<SUB>1/2</SUB>P<SUB>2</SUB>S<SUB>6</SUB>Ga<SUB>1/2</SUB>)- units as proposed by Wibbelmann. Studies on the Ag<SUB>(1+x)</SUB>Ga<SUB>(1-x/3)</SUB>P<SUB>2</SUB>S<SUB>6</SUB> series lend weight to this model of the structure. The partial depolymerisation as Ga was replaced by Ag along the series could be followed by Raman spectroscopy. It was possible to relate the increase in conductivity (decrease in activation energy) along the series of changes in local structure. It was found that the compositional trends in activation energy can be explained in terms of the modified Anderson-Stuart model of Martin and Angell and in particular in terms of changes in the jump distance.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:290249
Date January 1991
CreatorsMunro, Brian
PublisherUniversity of Aberdeen
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation

Page generated in 0.0044 seconds