STRUCTURE AND ELECTRICAL PROPERTIES OF SrO-BOROVANADATE (V2O5)z(SrO)0.2(B2O3)0.8-z GLASSES
SrO -borovanadate glasses with the nominal composition (V 2 O 5 ) z( SrO )0.2 ( B2 O 3 ) 0.8-z, 0.4≤z ≤0.8 were studied by direct current (DC) electrical conductivity, inductively coupled plasma (ICP) spectroscopy, Fourier transform infrared (FT-IR) spectroscopy and X-ray-powder-diffraction (XRD). These glasses were prepared by a normal quench technique and the actual compositions of the glasses were determined by ICP spectroscopy. XRD patterns confirm the amorphous nature of the present glasses. The temperature dependence of DC electrical conductivity of these glasses has been studied in terms of different hopping models. The IR results agree with previous investigations on similar glasses and it has been concluded that similar to SrO -vanadate glasses, metavandate chainlike structures of SrV 2 O 6 and individual VO 4 units also occur in these SrO -borovanadate glasses. The SrV 2 O 6 and VO n polyhedra predominate in the low B 2 O 3 containing SrO -borovanadate glasses as the B substitutes into the V sites of the various VO n polyhedra and only when the B 2 O 3 concentration exceeds the SrO content do BO n structures appear. This qualitative picture of three distinct structural groupings for the Sr -vanadate and Sr -borovanadate glasses is consistent with the proposed glass structure on previous IR and extended X-ray absorption fine structure (EXAFS) studies on these types of glasses. The conductivity results were analyzed with reference to theoretical models existing in the literature and the analysis shows that the conductivity data are consistent with Mott's nearest neighbor hopping model. However, both Mott VRH and Greaves models are suitable to explain the data. Schnakenberg's generalized polaron hopping model is also consistent with the temperature dependence of the activation energy, but the various model parameters such as density of states, hopping energy obtained from the best fits were found to be not in accordance with the prediction of the Mott model.