Working at the ALS, a multi-institutional collaboration led by researchers at ALS and Stanford University has identified a pseudogap phase with a nodal-antinodal dichotomy in ferromagnetic manganese oxide materials (manganites). Even though ferromagnetism and superconductivity do not exist together, the pseudogap state found in these manganites is remarkably similar to that found in high-temperature superconducting copper oxide materials (cuprates). This discovery casts new doubts on any direct link between the pseudogap phase and high-temperature superconductivity and adds fire to the debate over one of the great scientific mysteries of our time: What causes high-temperature superconductivity?

Beginning in 1986, the cuprates were discovered to have much higher superconducting transition temperatures (T_c) than known metallic superconductors. Among the possible explanations for high-T_c superconductivity, the pseudogap state has played a prominent role. In metals, electrons occupy a continuous spectrum of quantum states up to a maximum energy known as the Fermi level. When the metal is superchilled to become a superconductor, a forbidden energy gap opens up at the Fermi energy.

Experiments in recent years have shown that the spectrum in the superconducting state of high-T_c cuprates is characterized by a “nodal-antinodal dichotomy": when plotted in reciprocal (momentum) space, the spectrum is shaped like a cloverleaf, exhibiting a gap in the so-called antinodal direction (parallel to the Cu-O bonds), which vanishes along the nodal direction (from the origin running diagonal to the Cu-O chemical bonds). This dichotomous character of the excitation spectrum is generally assumed to be linked to the ”d-wave” symmetry of the superconducting state.

In the high-T_c cuprates, this dichotomous gap structure persists even above T_c and gives rise to the so-called “pseudogap state” with finite “Fermi arcs” replacing the original nodal points of the superconducting state. More intriguingly, the pseudogap appears to have two energy scales, with a low-energy pseudogap comparable to that of the superconducting gap and a much larger high-energy pseudogap. Naturally, the high-energy pseudogap is taken to represent the dominating physics.

The investigators used angle-resolved photoemission (ARPES) at ALS Beamline 10.0.1 to study a two-layer manganite compound consisting of a mixture of lanthanum, strontium, manganese, and oxygen (LSMO). Such manganese oxide materials become ferromagnetic below a certain critical temperature and display colossal magnetoresistance (CMR), meaning their electrical resistance can change by orders of magnitude in the presence of a magnetic field.

ARPES data for three magnifications plots photoelectron intensity variation with energy (relative to the Fermi level) and momentum in the nodal direction. Color represents photoelectron intensity (white=highest to dark blue=lowest). Momentum is determined from the angle of emission. The need for high energy and angular resolution is apparent upon realizing that the data analyzed in this experiment are those contained in the right arm of the light blue “V” in the left low-magnification image.

The investigators were quite surprised when the LSMO manganites displayed the characteristic signatures of the pseudogap state as found in the cuprates, including the nodal and antinodal dichotomy. These findings therefore cast doubt on the assumption that the pseudogap state and the nodal-antinodal dichotomy are hallmarks of the superconductivity state and suggests they are a more general phenomenon characteristic of transition-metal oxides.

Energy distribution curves (EDCs) plot photoelectron intensity variation with energy for fixed momentum. The set of EDCs is obtained by taking vertical line scans through the ARPES data at various momenta. The two features labeled “H” (red) and “Quasi-particle” (blue) are associated respectively with a particular electron band (designated e_g) and a small quasiparticle peak with polaronic features visible both as its small spectral weight (intensity) and its very narrow dispersion (variation of energy with momentum).

Moreover, these ARPES experiments suggest the occurrence of a polaronic metal with anisotropic band structure. A polaron is a "quasiparticle" formed when an electron causes a large distortion around it in the atomic lattice of a crystal. Since the surrounding atoms are much more massive than the bare electron, the polaron behaves as a negatively charged particle with a larger mass and lower mobility than an isolated electron. Earlier research at the ALS had linked the formation of polarons in LMSO to the CMR effect ("Polaron Behavior in CMR Manganites").

The locus of the momenta corresponding to a crossing of the Fermi level (“Fermi momenta”) defines a Fermi surface in momentum space (curved lines in g). G marks the origin of momentum space. Here a series of EDC sets for angular scans crossing the Fermi surface displays the nodal-antinodal dichotomy: the quasiparticle peak in the nodal direction (a) loses its intensity rather quickly while moving towards the antinodal direction (f).

This latest research goes farther and indicates that polaron formation is crucial to CMR. What might be even more interesting is the now emerging picture that polarons are the basic constituents of a poorly understood mysterious metallic phase that is present in both CMR materials and so-called underdoped high-T_c cuprates and likely ubiquitous to all transition-metal oxides. A proper understanding of this new phase will go a long way in capturing the essence of important phenomena such as colossal magnetoresistance and high-temperature superconductivity.

A series of EDC sets taken at temperatures from 15 K to 120 K shows that the nodal quasiparticle feature disappears as the temperature is raised to the ferromagnetic transition temperature (Curie temperature) at which LSMO also becomes insulating. The H feature does not show a dramatic temperature dependence, as predicted by recent theoretical models describing a polaronic scenario.

---

Research conducted by N. Mannella, W. Yang, and X.J. Zhou (Stanford University, Stanford Synchrotron Radiation Laboratory, and ALS); H. Zheng and J.F. Mitchell (Argonne National Laboratory); J. Zaanen (Stanford University, Stanford Synchrotron Radiation Laboratory, and Leiden University, The Netherlands); T.P. Devereaux (University of Waterloo, Canada); N. Nagasosa (University of Tokyo and Correlated Electron Research Center, Tsukuba, Japan); Z. Hussain (ALS); and Z.-X. Shen (Stanford University and Stanford Synchrotron Radiation Laboratory).

Research Funding: U.S. Department of Energy, Office of Basic Energy Sciences (BES); National Science Foundation; and Office of Naval Research. Operation of the ALS is supported by BES.

Publication about this research: N. Mannella, W. Yang, X.J. Zhou, H. Zheng, J.F. Mitchell, J. Zaanen, T.P. Devereaux, N. Nagasosa, Z. Hussain, and Z.-X. Shen, “Nodal quasiparticles in pseudo-gapped colossal magnetoresistive manganites,” Nature 438, 474 (2005).