Specific heat in strongly hole-doped Iron-based superconductors

In our work we compute the specific heat C(T) in an Fe-based superconductor with only hole pockets. We use a three orbital/three pocket model with two smaller hole pockets made out of d_xz and d_yz orbitals and a larger pocket made out of d_xy orbital. We use the experimental fact that the mass of d_xy fermion is much heavier than that of d_xz/d_yz fermions as an input. We argue that the heavy d_xy band contributes most to the specific heat in the normal state, but the superconducting gap on the d_xy pocket is parametrically smaller than that on d_xz/d_yz pockets. We also argue that in this situation the jump of C(T) at T_c is determined by d_xz/d_yz fermions, and the ratio (C_s - C_n)/C_n is a fraction of that in a one-band BCS theory. Below T_c, C(T) remains relatively flat down to some T^*, below which C(T) rapidly drops. This behavior is consistent with the data for KFe₂As₂ and related materials. We claim that the data on C(T) can be reproduced without assuming that the quasiparticle residue Z on d_xy band is small. We further argue that the very existence of a finite T^* < T_c favors s^+- gap structure over d-wave, because in the latter case T^* would vanish.