Effect of annealing on the specific heat of optimally doped Ba(Fe0.92Co0.08)2As2

Specific heat versus field for LiFe(1-x)Cu(x)As

LiFeAs is one of the new class of iron superconductors with a bulk [Formula: see text] in the 15-17 K range. We report on the specific heat characterization of single crystal material prepared by self-flux growth techniques with significantly improved properties, including a much decreased residual gamma, γ(r) (≡C/T as T → 0), in the superconducting state. Thus, in contrast to previous explanations of a finite γ(r) in LiFeAs being due to intrinsic states in the superconducting gap, the present work shows that such a finite residual γ in LiFeAs is instead a function of sample quality. Further, since LiFeAs has been characterized as nodeless with multiple superconducting gaps, we report here on its specific heat properties in zero and applied magnetic fields, to compare to similar results on nodal iron superconductors. For comparison, we also investigate LiFe(0.98)Cu(0.02)As, which has the reduced T(c) of ≈9 K and an H(c2) of 15 T. Interestingly, although presumably both LiFeAs and LiFe(0.98)Cu(0.02)As are nodeless, they clearly show a non-linear dependence of the electronic density of states (is proportional to specific heat γ) at the Fermi energy in the mixed state with the applied field, similar to the Volovik effect for nodal superconductors. However, rather than indicating nodal behavior, the satisfactory comparison with a recent theory for γ(H) for a superconductor with two isotropic gaps in the presence of impurities argues for nodeless behavior. Thus, in terms of specific heat in a magnetic field, LiFeAs can serve as the prototypical multiband, nodeless iron superconductor.

Specific heat discontinuity, ΔC, at Tc in BaFe2(As0.7P0.3)2-consistent with unconventional superconductivity

We report the specific heat discontinuity, ΔC/T(c), at T(c) = 28.2 K of a collage of single crystals of BaFe(2)(As(0.7)P(0.3))(2) and compare the measured value of 38.5 mJ mol(-1) K(-2) with other iron pnictide and iron chalcogenide (FePn/Ch) superconductors. This value agrees well with the trend established by Bud'ko, Ni and Canfield, who found that ΔC/T(c) is proportional to aT(c)(2) for 14 examples of doped Ba(1 - x)K(x)Fe(2)As(2) and BaFe(2 - x)TM(x)As(2), where the transition metal TM = Co and Ni. We extend their analysis to include all the FePn/Ch superconductors for which ΔC/T(c) is currently known and find ΔC/T(c) is proportional to aT(c)(1.9) and a = 0.083 mJ mol(-1) K(-4). A comparison with the elemental superconductors with T(c) > 1 K and with A-15 superconductors shows that, contrary to the FePn/Ch superconductors, electron-phonon-coupled conventional superconductors exhibit a significantly different dependence of ΔC on T(c), namely ΔC/T(c) is proportional to aT(c)(1.9). However ΔC/γT(c) appears to be comparable in all three classes (FePn/Ch, elemental and A-15) of superconductors with, for example, ΔC/γT(c) = 2.4 for BaFe(2)(As(0.7)P(0.3))(2). A discussion of the possible implications of these phenomenological comparisons for the unconventional superconductivity believed to exist in the FePn/Ch is given.