Three-dimensional Fermi surface determination by angle-resolved photoelectron spectroscopy

Abstract

Angle-resolved photoelectron spectroscopy ͑ARPES͒ is commonly applied to map the shape of Fermi surfaces. Here we quantify the errors of simple criteria for extracting Fermi vectors by ARPES that are induced by strongly varying matrix elements. Sophisticated methods for determining the three-dimensional Fermi vector based on temperature and photon energy dependent photoemission are discussed with reference to data of the quasi-two-dimensional system 1T-TiTe 2 .

High-resolution angle-resolved photoemission investigation of the electronic structure of Cr-intercalated1T‐TiTe2

“…26 The structures in the bonding energies region beyond 2 eV are mostly formed by the Te5p states, as was already known from earlier calculations. 15,16 Whereas the composition of bands remains the same in pristine and intercalated compounds, the contrast is better in TiTe 2 . Their shape, incidentally symmetric around ⌫ in the undoped compound, gets substantially distorted on intercalation with Cr ͑the band d͒.…”

“…Such dependence of this peak's intensity, and also of its width, on k ʈ are consistent with earlier reported data. [14][15][16] The pinned energy position of the peak reveals a dispersionless band, which is not revealed in DFT calculation. Apart from this, the dispersion of other local maxima in angle-resolved spectra of TiTe 2 falls astonishingly well onto the calculated DFT band structure, as will be discussed in a minute.…”

“…15 However, this assumption has been revised in some works. 15, 16 Rossnagel et al 16 have shown that the k Ќ component is not negligibly small. It should be noted that a decrease of the lattice parameter on intercalation makes the crystal more dense and less "quasi-two-dimensional," than it could have been implied for pure TiTe 2 .…”

“…We use at present an allelectron method, that however does not lead to any noticeable differences in the energy dispersion as compared to that of Ref. 16.…”

“…15 Following these studies, which were limited by the atomic sphere approximation and the neglect of the spin-orbit coupling, a pseudopotential calculation with Gaussian-type orbitals and spin-orbit interaction included has been reported in Ref. 16. We use at present an allelectron method, that however does not lead to any noticeable differences in the energy dispersion as compared to that of Ref.…”