No abstract

No abstract

No abstract

O, T) for an uncompensated metal. " These eA'ects must also be present for a more general Fermi surface. For example in a compensated metal the variation in the relaxation time and effective mass will be reflected in the multiplying constant of (1).In the thermal case, the relaxation times will not be uniformly scaled from the values appropriate to the electrical case so the constant will in general be diferent. Thus the high-"In an earlier discussion (Ref. 8) results of the 2-band model were quoted but the particular assumption made, (cov)I= (cur)2, was not stated. 6eld limits given cannot always be expected to be good approximations and it is possible that the decrease from L(O, T) observed for indium, r ' an uncompensated metal, may be due to this.We should like to thank Professor J. L. Olsen for most illuminating discussions on this subject, One of us (P. W. ) is grateful to the Organization for Economic Co-operation and Development for a fellowship which made possible a very profitable stay in Nottingham.

No abstract

No abstract

The three resonantly enhanced Raman vibrations in trans-iCK)^ and -(CD)^ are described in terms of amplitude modes associated with the Peierls gap. The dispersion in the Raman-line profiles is explained by inhomogeneity. A narrow distribution of the effective electron-phonon coupling constant is required to account for the observed spectra in detail. The resonant Raman vibrations and the ir-active modes induced by doping or photogeneration are all described by the same form of dressed phonon propagator«

No abstract