The route to high temperature superconductivity in transition metal oxides
The discovery of high temperature superconductivity in cuprates was possible only through an intimate knowledge of perovskite oxides which have been synthesized and characterized for decades at the IBM in the Zürich laboratoty. Especially SrTiO3 and LaAlO3 have been in the focus at IBM as was presented in Volume 1 of a series [1]. Probably for the first time detailed microscopic investigations of the local properties of these compounds have been obtained by studying by means of EPR the surroundings of transition metal impurities in these materials. These experiments enabled the identification of the order parameter of the structural instability observed in these oxides. However, the idea to search for superconductivity came later motivated by theoretical considerations that metallic hydrogen could become superconducting at high temperatures. Since SrTiO3 is an insulator it was thought that the implantation of hydrogen would render it metallic and eventually also superconducting. This approach failed since the carrier density remained always too small. In sequence it was then tried to achieve a metallic state in oxide perovskites by varying their composition which was in so far promising as reduced SrTiO3 exhibits superconductivity at 0.3K [2]. Furthermore it was subsequently shown [3] that Tc can be enhanced to 1.2K by doping SrTiO3 with Nb. In spite of the fact that Tc was far below values achieved in A15 compounds, a remarkable observation was connected with the Nb doped perovskite, namely for the first time long before predicted two-gap superconductivity was realized here.