No abstract

cision of about 1%. The HQ values are estimated from a parabolic extrapolation which, because of the limited range of T, is probably only accurate to about 10%. From the value of idHc/dT)^^, the specific heat discontinuity at T^ is computed to be 2.20 mJ/mole deg for Ru. Using the calorimetrically determined value^ of y = 3.35 mJ/mole deg^ (where y is the normal electronic specific-heat coefficient), one obtains (AC/yT^)j:= j =1.37, in reasonable agreement with the Bardeen-Cooper-Schrieffer (BCS) value of 1.43.®The T^ value obtained here is roughly intermediate between the values reported in previous measurements. ^»^°'" (The scatter of our points suggests that our absolute accuracy in T is no better than ±0.007°K.) Our values of (dHc/dT)^^ are significantly larger than the results of Hulm and Goodman^^ but in reasonable agreement with the value of Carruthers and Connolly.^^ However, it seems noteworthy that, aside from the apparent absence of an isotope effect, these Ru specimens show no other discernable anomalies. Phys. Rev. 112, 1522 (1958.^Unfortunately the relatively large transition width characteristic of a spherical specimen makes precise determination of H^ more difficult and contributes to the experimental scatter evident in Fig. 2. The polarization of the conduction electrons that is induced by a magnetic field lowers the free energy of the normal state relative to the superconducting state. Indeed, it has recently been shown^^^ that, in the limit of complete field penetration,^ this mechanism imposes an upper limit on the critical field obtainable in ordinary superconductors. In a rare earth ferromagnetic metal the effective exchange field ^^ff impressed on the conduction electrons, via the exchange interaction with the rare earth spin S, is in general so large as to inhibit the occurrence of superconductivity in zero external field H, We would like to point out that in certain ferromagnetic metals ^eff opposes H and allows for the conduction electron polarization to be cancelled so that, if in addition, these metals possess a predominantly attractive electron-electron interaction, the possibility arises that superconductivity will occur in the compensation region. Since the cancellation will presumably occur for H--^^eff, critical fields may be obtained which exceed, by an order of magnitude, the limit derived for ordinary superconductors.^'^ The free energy of the normal state of a metal in zero field F^Q is lowered by an amount IX/,^ in a magnetic field. Characterizing the free energy of the superconducting state requires that one specify the field configuration and spin magnetization. The assumptions underlying the Clogston limits were (1) complete field penetration (no Meissner effect) and (2) that the magnetic 290

No abstract

No abstract

No abstract