Abstract
Although it is well established that small polarons are formed in transition metal oxide (TMO) glasses, no satisfactory quantitative analysis of d.c. and a.c. transport exists. The temperature dependences of the d.c. and a.c. conductivities for TMO glasses are re-examined. The charge transport cannot be explained quantitatively by the small-polaron model (strong electron-lattice coupling). An alternative possibility—a multiphonon tunnelling process with weak coupling—is also discussed. The a.c. conductivity, which is strongly correlated to the d.c. conductivity and which has never been explained by current theories based on the pair approximation, is interpreted in terms of the continuous-time random-walk approximation.