Measurements of the specific heat of Mg11B2 from 1 to 50 K, in magnetic fields to 9 T, give the Debye temperature, Theta = 1050 K, the coefficient of the normal-state electron contribution, gamma(n) = 2.6 mJ mol(-1) K-2, and a discontinuity in the zero-field specific heat of 133 mJ mol(-1) K-1 at T(c) = 38.7 K. The estimated value of the electron-phonon coupling parameter, lambda = 0.62, could account for the observed T(c) only if the important phonon frequencies are unusually high relative to Theta. At low T, there is a strongly field-dependent feature that suggests the existence of a second energy gap, about 4 times smaller than the major gap.

Abstract. -We show that the specific heat of the superconductor MgB2 in zero field, for which significant non-BCS features have been reported, can be fitted, essentially within experimental error, over the entire range of temperature to Tc by a phenomenological two-gap model. The resulting gap parameters agree with previous determinations from band-structure calculations, and from various spectroscopic experiments. The determination from specific heat, a bulk property, shows that the presence of two superconducting gaps in MgB2 is a volume effect.The discovery of superconductivity in MgB 2 [1] raised the questions of its nature and the origin of its relatively high transition temperature T c ∼ 40 K. Specific heat (C) is a powerful tool to aid in answering these questions and, more generally, to provide information on the thermodynamics of the transition. Several groups have reported such measurements on MgB 2 [2-10]. It is now established that C significantly deviates from the standard BCS behaviour. First, a large excess in C is observed in the vicinity of T c /4 [2-6]. Second, an exponential fit of C(T ) in the region T ≪ T c indicates a gap ratio 2∆ 0 /k B T c only onequarter to one-third of the isotropic BCS value [3,4,6]. This excess was interpreted as a possible sign of a second superconducting gap, whose existence is predicted by band-structure calculations [11][12][13]. The specific heat near T c is puzzling also with the jump ∆C at T c consistently smaller than the BCS weak-coupling lower bound. In this Letter, we present an empirical two-gap model that fits the experimental data over the whole range of temperature to T c . This model resolves the apparent contradiction between different analyses of the specific heat, and relevant parameters show good agreement with determinations based on independent experiments.We focus on two sets of specific-heat data obtained independently in two different laboratories. Experimental methods and results have been described elsewhere [2,3,5,6]. The unusual excess specific heat at ∼ T c /4, which denotes the presence of excitations within the c EDP Sciences

No abstract

No abstract

The specific heats of two samples of UPts have been measured in the vicinity of the transition to the superconducting state. In both cases the specific-heat anomalies are sharper than any previously observed, and two maxima are clearly resolved. The results are interpreted as evidence of a splitting of the transition and unconventional pairing. A model that is consistent with the known sample dependence of the superconducting-state specific heat is used to derive "intrinsic" values of the related parameters.PACS numbers: 74.70.Tx, 65.40.Em, 74.30.Ek, 74.60.Mj From the initial discoveries of superconductivity in CeCu2Si2/ UBei3,^ and UPts, ^ it has been clear that these heavy-fermion superconductors (HFS) are unusual, and it was soon recognized"^ that the coupling mechanism and the nature of the superconducting state might be unconventional. A number of differences between the properties of the superconducting state in HFS and in conventional BCS superconductors have been observed, for example, in the temperature dependences of both transport and thermodynamic properties. However, the interpretation of these results has been clouded by questions associated with sample quality, and particularly by the inhomogeneity implied by the broad superconducting transitions generally observed. In the case of UPts, the temperature dependences of the upper critical field ^ and the rf susceptibility,^ as well as the field dependence of the ultrasonic attenuation,^ suggest the existence of two distinct superconducting states that occur at different fields. It has since been pointed out that this is to be expected for rf-wave pairing, and that even in zero field there should be two transitions occurring at different temperatures.^ It seems possible that these transitions would appear as two separate anomalies in the specific heat, C, as is observed in the case of liquid ^He. The discontinuity in C at the critical temperature Tc, AC(Tc), can also be expected to give information about the nature of the pairing in the superconducting state (see Ref. 9 and others cited there). However, most measurements on UPt3 have been made on samples that showed broad transitions with no sign of structure in ^3,10-12 Y\iQ exceptions are measurements on a series of three samples that were prepared at Grenoble: Measurements there, ^^~^^ and also at Berkeley on one of the samples, ^^ have shown "shoulders" on the high-temperature sides of the anomalies at Tc. The recurrence of that feature from sample to sample was highly suggestive, but, in view of the known dependence of Tc on sample quality, ^' the results were interpreted cautiously, and until now it has not been claimed that this structure was an intrinsic property of UPts.Specific-heat measurements on two new samples of UPt3, each of which shows two distinct maxima near Tc that correspond to two transitions separated by approximately 60 mK, are presented in this Letter. The new samples were prepared in diflferent laboratories by different techniques, and in both cases their properties reflec...