Abstract
The idea of gaplessness, already familiar in several areas of superconductivity, is considered in connection with high-current superconductivity. A calculation is made of the critical temperature Tc at which a normal metal with zero current becomes thermodynamically unstable against decay into a high-current super-conducting phase. It is pointed out that at Tc all the various suggested forms of high-current electron-electron interaction, both retarded and nonretarded, are equivalent, so that a calculation of Tc does not suffer from ambiguity of choice of interaction. It is found that a transition temperature exists when the Bardeen, Cooper, and Schrieffer (BCS) parameter N(0)V0>~0.43 [assuming N(0)Vc0=0.1]. In fact, for 0.43<N(0)V0<0.63, two values of Tc exist, since at sufficiently low temperatures, the normal phase is thermodynamically stable against high-current superconductivity. When N(0)V0=0.43, Tc=0.259ωkB. At T=0, a separate calculation is made of the N(0)V0 required to obtain high-current superconductivity with a finite thermal energy gap 2(ε0pFv0). This calculation, using a nonretarded interaction, includes the correction of an algebraic error in the 1959 paper by Parmenter. The result, N(0)V0>~0.70, is in close agreement with the recent work of Hone, who made use of a retarded interaction in calculating that one must have N(0)V0>~0.67 in order to get finite-thermal-gap high-current superconductivity. This suggests that retardation effects are not crucial in determining the occurrence or non-occurrence of high-current superconductivity. It is pointed out that experimental measurements of Morin and Maita have suggested that some transition-metal superconductors have large enough values of N(0)V0 for both the gapless form and the finite-thermal-gap form of high-current superconductivity to occur. Either form, if it exists, will exhibit the same form of electrical instability that is known to occur under exceptional conditions in conventional low-current superconductivity.