Electronic Structure ofNiS1−xSexacross the Phase Transition

Abstract

We report very highly resolved photoemission spectra of NiS1−xSex across the so-called metalinsulator transition as a function of temperature as well as composition. The present results convincingly demonstrate that the low temperature, antiferromagnetic phase is metallic, with a reduced density of states at EF . This decrease is possibly due to the opening of gaps along specific directions in the Brillouin zone caused by the antiferromagnetic ordering.While metal-insulator transition has been one of the outstanding problems in condensed matter physics [1], the specific case of a transition in hexagonal NiS [2] as a function of temperature has been possibly the most controversial, spanning a period of three decades. The controversy relates to the very basic issue of the nature of the transition, with various groups describing it as a metal-insulator transition [2][3][4][5][6][7], while nearly equal number of groups claiming it to be a metal-metal transition [8][9][10][11][12]. The transport and magnetic properties are that NiS is a highly conducting (≈10 −5 ohm-cm) Pauli paramagnetic metal at room temperature with a characteristic metallic dependence of resistivity on temperature [10,11]. Near 260 K, the system undergoes a first order phase transition, with a nearly two orders of magnitude increase in resistivity (≈10 −3 ohm-cm) [10,11]. The system becomes antiferromagnetic below the transition [3], showing a volume expansion of about ≈1.9% without any change in crystal symmetry. While it is generally agreed that the high temperature phase represents an example of a highly conducting metallic compound, the controversy continues to exist concerning the nature of the low temperature phase. Normally one would expect that transport measurements would readily resolve the question concerning the metallic or insulating phase of the ground state. A very unusual nearly temperature independent resistivity down to about 4 K [10,11], however, does not provide any unambiguous clue. While in a metallic system the resistivity is expected to decrease with temperature, in contrast to the observed behavior, an insulating ground state should have very manifest exponentially increasing resistivity, when the temperature scale is lower than the band gap. Thus, if the low temperature phase is insulating, the band gap should be much smaller than 4 K (< 0.5 meV). On the other hand, if it is metallic, it has to be an unusual state where the resistivity is nearly independent of temperature over a wide range of temperature (4 K ≤ T ≤ 260 K). Unfortunately, theoretical studies have not been of much help in settling these discussions in favor of either of the views. The earliest non-self-consistent band structure calculations [13] correctly predicted the high temperature Pauli paramagnetic phase to be metallic, while the low temperature anti-ferromagnetic phase yielded a small gap. However, subsequent self-consistent band structure calculations [14] based on local spin density approximation (LSDA) failed to yield any gap. Similarly, the LDA+...

Electronic structure of millerite NiS

“…For the preparation of the millerite samples, polycrystalline samples of hexagonal NiS was prepared first by solid state reaction [1,7]. Hexagonal NiS sample was then sealed in quartz tube in high vacuum and was maintained at 573 K for about 2-3 weeks and was cooled slowly to room temperature over a period of 8 hours, to obtain the millerite phase.…”

Optical study of the metal–nonmetal transition in NiS

“…Effects of substituting other elements for Ni or S have been also studied in detail. 7,8 Recently, two high-resolution photoemission studies 9,10 have revealed a finite density of states (DOS) around the Fermi energy (E F ) in the LT phase, but they have given contrasting interpretations: Nakamura et al 9 have concluded that there is a small correlation-induced band gap with an unusually sharp band edge, and that the observed finite DOS at E F is due to thermal and instrumental broadenings of the edge. On the other hand, Sarma et al 10 have concluded that the LT phase is an "anomalous metal".…”

Defect evolution during the phase transition of hexagonal nickel sulfide studied by positron annihilation spectroscopy

“…Many other basic physical properties of the hexagonal NiS can be found in the related Refs. [11][12][13]. As Hall coefficients could not be observed in the HT phase [8], the defects in that phase are inconclusive.…”