Trimeron ordering, bandgap and polaron hopping in magnetite

Nikita Fominykh, Vladimir Stegailov · 2024

Read Trimeron ordering, bandgap and polaron hopping in magnetite by Nikita Fominykh, Vladimir Stegailov (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-20865.txt.

Description

Abstract In this work, we apply the DFT+U approach for a detailed ab initio study of the refined structure of the low-temperature phase of magnetite [M. S. Senn et al., Nature 481 , 173 (2012)]. We compare the electronic properties of this structure and several alternatives with respect to the presence of trimeron ordering and the bandgap properties.

Catalog record

CREATORNikita Fominykh, Vladimir Stegailov
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-20865

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Trimeron ordering, bandgap and polaron hopping in magnetite Nikita Fominykh, Vladimir Stegailov Abstract In this work, we apply the DFT+U approach for a detailed ab initio study of the refined structure of the low-temperature phase of magnetite [M. S. Senn et al., Nature 481 , 173 (2012)]. We compare the electronic properties of this structure and several alternatives with respect to the presence of trimeron ordering and the bandgap properties. The connection of the trimeron arrangement with site-selective doping of magnetite is discussed. Calculations of the polaron hopping energy allow us to make one step forward toward understanding the complex interplay of polaronic and bandgap contributions to electronic properties of the magnetite. Preprint: APS/123-QED I Introduction The physics of magnetite has been an enigmatic puzzle for nearly a century [ 1 , 2 ] with its pinnacle represented by the Verwey transition [ 3 , 4 ] which is one of the fundamental problems of condensed matter physics and involves a complex interplay of crystal structure and lattice vibrations with electronic structure and strong electron correlations [ 5 , 6 ] . The physics of the Verwey transition illustrates the modern level of consistency among experimental methods, theoretical approaches, and computational tools available for understanding the properties of materials. The significant confusion associated with magnetite physics stems from the fact that magnetite was the first substance where the metal-insulator transition was detected: in fact, as a generalization of the Verwey transition. At the same time, modern experimental studies describe this transition undoubtedly as a semiconductor-semiconductor transition (see, e.g. [ 7 , 8 ] ). The development of ab initio electronic structure methods follows several complementary avenues [ 9 , 10 ] and materials with strong electronic correlations represent one of the most complicated classes that has progressed during the last decades through many stages of rethinking (see, e.g., [ 11 , 12 , 13 ] ). Subtle nuances of electronic structure methods were shown to be crucial for understanding magnetite properties [ 14 , 15 , 16 , 17 ] . The question of the crystal structure of magnetite imposes fundamental limitations on the theoretical understanding of its electronic structure. On the one hand, high-temperature (HT) phase has a well established cubic Fd\bar{3}m lattice. However, this lattice is not stable at low temperatures and therefore can not be applied for static zero-temperature ab initio electronic structure calculations. On the other hand, the crystal structure of the low-temperature (LT) phase has been the subject of debate for a long time. Presumably, the first attempt to apply the ab initio theory taking into account strong correlations was made in 1996 by Anisimov et al. [ 18 ] using the DFT+U method. Szoteck et al. [ 19 ] applied the SIC-DFT approach for strong correlations and considered cubic, Pmc2_{1} and P2/c candidate structures. After the next major refinement of the LT phase structure in 2001-2002 by Wright, Attfield, and Radaelli [ 20 , 21 ] , Leonov et al. [ 22 ] and Jeng et al. [ 23 ] presented the DFT+U calculations for the P2/c structure. It was shown [ 22 , 24 ] that the resulting orbital ordering obeys the Kugel-Khomskii model [ 25 , 26 , 27 ] . Subsequent refinement of the LT phase structure was published in 2012 by Senn, Wright, and Attfield [ 28 ] showing that the Cc structure was the best variant. Electronic structure calculations for the Cc structure [ 29 , 30 , 31 , 32 ] provided systematically higher bandgap values than predicted by previous studies. To our knowledge, these systematic changes in bandgap values have not received any appropriate attention yet that would be very desirable in the context of the significant complexity of interpretations proposed for high-precision experimental data on magnetite electronic properties [ 33 , 34 , 16 , 17 , 7 , 35 , 36 , 8 ] . The key feature of the Verwey transition is a sharp decrease in conductivity about two orders of magnitude below T_{V}\sim 125~\text{K} . It was this transition that stimulated the concept of metal-insulator transition [ 37 ] frequently used to describe the differences between the LT and HT phases of magnetite. The closure of the bandgap at temperatures above T_{V} is usually implied. Infrared optical conductivity data [ 33 , 34 ] , the photoemission spectroscopy (PES) results [ 38 , 39 , 40 , 41 , 42 , 43 ] and scanning tunneling spectroscopy (STS) results [ 44 , 35 ] could be interpreted as the manifestation of a bad metallic state above T_{V} . However, these methods can be quite sensitive to surface states, and at the same time no clear Fermi edge is observed. Moreover, the conductivity value itself is much lower than the minimum Ioffe-Regel conductivity for metals [ 35 ] . For the HT phase, the dc conductivity follows the Arrhenius law with an activation energy E_{a}^{HT}\sim{50}~\text{meV} . In the LT phase, the conductivity follows the Arrhenius law also with a slightly higher activation energy E_{a}^{LT}\sim{100}~\text{meV} [ 3 , 45 , 46 , 8 ] , similar values are also observed for terahertz conductivity [ 47 ] . There are several models that describe charge transport in magnetite. The bandlike model with delocalized electron transport proposed by Cullen and Callen [ 48 ] . In contrast, other models suggest localized charge transport via polarons formed because of strong electron-phonon interactions, such as small polarons [ 49 ] , bipolarons [ 50 ] and molecular polarons [ 51 ] . Later, Ihle and Lorenz assumed the superposition of the small polaron band and the hopping conductivity [ 52 ] . The conductivity behavior shows a clear semiconductor-semiconductor type of Verwey transition [ 8 ] . The concept of a semiconductor-semiconductor transition is further supported by a series of papers, in which the bandgap E_{g} is observed both above and below T_{V} . Using temperature-dependent high-resolution PES data, Park et al. [ 53 ] showed that the bandgap does not vanish above the Verwey transition but undergoes a jump decrease from E_{g}^{LT}\sim 150 meV to E_{g}^{HT}\sim 100 meV. Later, similar results were obtained using two less surface-sensitive methods: the soft X-ray PES [ 7 ] and the hard X-ray PES [ 54 ] gave slightly lower values E_{g}^{LT}\sim 100 meV and E_{g}^{HT}\sim 50 meV, which are in a good agreement with the conductivity activation energies. In all of these works, the possible role of polarons on electronic properties is emphasized. In addition, a STS study by Jordan et al. is reported in which a constant E_{g}^{LT}\approx E_{g}^{HT}\sim 200 meV surface gap is observed below and above the Verwey transition [ 55 ] . Later, with a single magnetite nanocrystal, Hevroni et al. observed a decrease in the bandgap just above the transition to E_{g}^{HT}\sim 75 meV [ 36 ] and in another study of single magnetite nanocrystal a higher value for low temperature phase E_{g}^{LT}\sim 320 meV and E_{g}^{HT}\sim 70 meV were observed [ 56 ] . Table 1: The review of the selected previous ab initio studies of magnetite that provided the bandgap E_{g} values. Selected key properties of previous ab initio magnetite studies. The Fe B

Read the complete text