Superconductivity in the complex metallic alloy beta-Al3Mg2

Based on low temperature electrical resistivity and specific heat measurements, we have shown that beta-Al3Mg2 undergoes a phase transition into a superconducting ground state at Tc=0.87 K. Microscopically, superconductivity can be understood in terms of the phonon-mediated BCS model. An exponential behavior of the specific heat well below Tc implies a nodeless superconducting gap in the electronic density of states, of the order of 1.6 K. The initial slope of the upper critical field is deduced to be about −0.2 T/K, while an extrapolation T→0 yields mu0Hc2 about 0.14 T. The limiting pair breaking mechanism seems to be orbital pair breaking, as concluded from the model of Werthamer et al. Superconductivity in beta-Al3Mg2 occurs in a crystal environment without inversion symmetry. Broken inversion symmetry has a distinct influence on the superconducting phase, which usually relies on the formation of pairs of electrons in degenerate states with opposite momentum. The availability of such states is normally guaranteed by time reversal and inversion symmetries. The absence of inversion symmetry would favor a strong antisymmetric spin-orbit coupling and, as a consequence, a mixture of spin-singlet and spintriplet pairs in the superconducting condensate can be expected. The small values of the upper critical field, however, seem to exclude a substantial portion of spin-triplet pairs in the condensate. Moreover, the lightweight elements Al and Mg may be responsible for only a minimal spin-orbit coupling in beta-Al3Mg2; hence, the spin-singlet condensate dominates. Additionally, the very complex crystal structure is supposed to smooth the effect of the missing inversion symmetry. A rather conventional superconductivity seems to appear, which also follows from the agreement of the upper critical field with Werthamer’s model. Presently, only a small number of superconductors without inversion symmetry have been found. Although the crystal structure of beta-Al3Mg2 appears to be rather complicated, the various physical quantities derived in both the superconducting and the normal state region turn out to be simple. In the first approximation, some of these quantities even look like a balanced superposition of pure Al and Mg. The latter follows from macroscopic measurements such as the specific heat and microscopic data like those derived from NMR as well.