What do we learn from impurities and disorder in high-Tc cuprates?

A series of experimental studies established that the differing morphologies of the phase diagrams versus hole doping nh of the various cuprate families are mostly controlled by defects and disorder. In the minimally disordered cuprate Yttrium Baryum Copper Oxide (YBCO) we introduced controlled detfects that allowed us to probe the metallic and superconducting states. We demonstrate that the extent of the spin glass phase and the superconducting dome can be controlled by the concentration of spinless (Zn, Li) impurities substituted on the planar Cu sites. NMR frequency shift measurements establish that these defects induce, in their vicinity, a cloud with a Kondo-like paramagnetic behavior. Its “Kondo” temperature and spatial extent differ markedly between the pseudogap and strange metal regimes. We have performed transport measurements on single crystals with a controlled content of in-plane vacancies introduced by electron irradiation. At high T, the inelastic scattering of the carriers has been found independent of disorder and completely governed by the excitations of the correlated electronic state. The low T upturns in the resistivity associated with single-site Kondo-like scattering are qualitatively in agreement with local magnetism induced by spinless impurities. The apparent metal insulator crossover is only detected for a very large defect content, and part of the large resistivity upturn remains connected with Kondo-like paramagnetism. In the superconducting state, the defect-induced reduction of Tc scales linearly with the increase in residual resistivity induced by disorder. High-field magnetoresistance experiments permit us to determine the paraconductivity due to superconducting fluctuations. The latter vanishes beyond a temperature T’c and a field H’c that both decrease with increasing in-plane defect content. In the pseudogap regime, the weaker decrease of T’c with respect to that of Tc reveals a large loss of superconducting phase coherence in the presence of disorder. In light of our experimental results, we initiate a discussion of its interplay with pair breaking. Our data also permit us to confirm that the differing phase diagrams are due to competing orders or disorders that are family-specific. In the ideal phase diagram of a disorder-free cuprate, 2D superconductivity should persist at low doping. This ensemble of experimental results provides serious challenges for the theoretical understanding of superconductivity in these correlated electron systems.