Chemical potential shift in La 1 − x Sr x MnO 3 : Photoemission test of the phase separation scenario

– We have studied the chemical potential shift in La1−xSrxMnO3 as a function of doped hole concentration by core-level x-ray photoemission. The shift is monotonous, which means that there is no electronic phase separation on a macroscopic scale, whereas it is consistent with the nano-meter scale cluster formation induced by chemical disorder. Comparison of the observed shift with the shift deduced from the electronic specific heat indicates that hole doping in La1−xSrxMnO3 is well described by the rigid-band picture. In particular no mass enhancement toward the metal-insulator boundary was implied by the chemical potential shift, consistent with the electronic specific heat data. It is generally recognized that the manganites R1−xAxMnO3 with the perovskite structure, where R is a rare-earth (R = La, Pr, Nd) and A an alkaline-earth metal (A = Sr, Ba, Ca), exhibit a very rich phase diagram because of competition between the ordering and fluctuations of the spin, lattice and charge degrees of freedom [1]. They are antiferromagnetic (AFM) insulators for undoped materials (x = 0) while hole doping produces a ferromagnetic (FM) metallic phase, which shows colossal magnetoresistance (CMR). Historically, the ferromagnetism in this phase has been explained by double exchange mechanism [2]. Materials with x close to 0.5 have been the focus of recent studies because most of the manganites show so-called CE-type antiferromagnetic chargeand orbital-ordered insulating state at x = 0.5. When this spin-charge-orbital ordering completes with metallic behavior, they also exhibit a remarkable CMR effect, which becomes the subject of intensive studies. The origin of the CMR behaviors has been debated for many years since it was pointed out that double exchange interaction alone is insufficient to explain the large resistivity change [3]. Recently, the possibility of phase separation as the origin of the CMR is discussed by theoretical studies. Moreo et al. [4] has pointed out the possibility of the coexistence of the antiferromagnetic or ferromagnetic insulating phase and the ferromagnetic metallic phase