Orbital Order and Insulator-Metal Transition with a d-wave Pseudogap in a Two-Dimensional Model for Manganites.

Mukul S. Laad

MPI for the Physics of Complex Systems, Dresden, Germany

The multitude of orbital (OO) and magnetic ordered (MO) ground states in perovskite manganites, R1-xAxMnO3 with R=La,Pr,.. and A=Ca,Sr, as a function of hole doping, x, are an unambiguous fingerprint of the crucial role of multi-orbital electronic correlations in these Mott-Hubbard systems. Understanding these ordered states lies at the heart of a consistent resolution of the colossal magneto-resi stance (CMR) observed in manganites, where melting of the OO induces a low-$T$ insulator-metal transition for 0.25 x 0.45. Recently, a d-wave pseudogap in the "bad metallic" phase of bilayer manganites has been extracted from angle-resolved photoemission spectroscopy by Dessau and co-workers. This has prompted inevitable comparison with similar features observed in underdoped cuprates. However, given the very different physics of these two systems, its origin in manganites is a enigmatic, and promises to shed invaluable light on our microscopic understanding of these materials.

We have used a doubly-degenerate Hubbard model with:

(i) realistic hopping structure characteristic of the eg-based perovskites,

(ii) strong, multi-orbital coulomb interactions and Jahn-Teller distortions,

to study these aspects of manganite physics. Using (a) multi-orbital DMFT, and (b) cluster diagonalisation studies, we focus on the orbital ordered states as a function of hole doping, x, as well as on the AFOO-FI to PI-FM transition. For x=0 (pure LaMnO3), we find the correct AFOO (of 3x2-r2/3y2-r2 orbitals) Mott insulating state. Moreover, the zig-zag (diagonal) as well as the stripe OO phases observed for x=1/2 and for x=2/3,3/4 are exactly obtained. These various OO states are EXACTLY representable as an ``alloy'' ordering of the 3x2-r2/3y2-r2 orbitals on each Mn3+ site. Remarkably, the "bad metallic" state in the DMFT turns out to have a d-wave pseudogap, in full agreement with ARPES. Implications for understanding CMR will be touched upon.

Back