Theory of the thermal magnetocapacitance of multicomponent silicate glasses at low temperature

Abstract

An explanation is put forward for the early findings of Zeller and Pohl ( 1971 Zeller RC Pohl RO 1971 Phys. Rev. B 4 2029 see also R.O. Pohl in Phillips (1981) , Phys. Rev. B, 4, 2029) and, in particular, Stephens ( 1976 Stephens, RB . 1976. Phys. Rev. B, 13: 852 , Phys. Rev. B, 13, 852), and then for the more comprehensive recent findings of Meißner and Strehlow ( 2000 Strehlow, P , Wohlfahrt, M , Jansen, AGM , Haueisen, R , Weiss, G , Enss, C and Hunklinger, S . 2000, 1938. Phys. Rev. Lett., 84 ) for the anomalous magnetic-field-dependent behaviour of the heat capacity of some multicomponent glasses at low temperature (0.1 < T < 4 K) and in weak-to-moderate fields (0 < B < 8T). An essential new ingredient of this theory, together with the standard two-level system approach for cold glasses, is a multi-minima local potential for the quantum motion of metallic ions between near-degenerate wells, separated by relatively shallow barriers. Taking the tunneling entities as independent and assuming a novel probability distribution for the parameters of the multi-welled potential, an expression for the heat capacity C(T,B) is obtained with features that match rather well the reported unusual experimental data for the glasses. The indication coming from this theory for the data available is that the entities involved in tunnelling are nanoclusters of coherent ions.