Surface-State Bipolaron Formation on a Triangular Lattice in thesp-TypeAlkali-Metal/Si(111)Mott Insulator

Abstract

We report on new low-energy electron diffraction, scanning tunneling microscopy, and angle-resolved photoemission spectroscopy studies of alkali-metal/Si(111) previously established as having a Mott-insulating ground state at surface. The observation of a strong temperature dependent Franck-Condon broadening of the surface band together with the novel sqrt[3] x sqrt[3] --> 2(sqrt[3] x sqrt[3]) charge and lattice ordering below 270 K evidence a surface charge density wave in the strong electron-phonon coupling limit (g approximately 8). Both the adiabatic ratio variant Planck's over 2piomega_{0}/t approximately 0.8 and the effective pairing energy V_{eff} = U - 2gvariant Planck's over 2piomega_{0} approximately -800 meV are consistent with the possible formation of a bipolaronic insulating phase consisting of alternating doubly occupied and unoccupied dangling bonds as expected in the Holstein-Hubbard model.

Giant Alkali-Metal-Induced Lattice Relaxation as the Driving Force of the Insulating Phase of Alkali-Metal/Si(111):B

“…We found 0.1 and 0.19 for Si up and B down atoms and 0.008 and 0.025 for Si down and B up atoms. Hence, a moderate charge modulation q % 0:3e is obtained from DFT calculations being far from the value q ¼ 2e expected in the bipolaronic limit previously supposed to be reached [18,19].…”

“…Angle-resolved photoemission measurements reveal the 2 ffiffiffi 3 p symmetry and an alkalimetal-dependent gap [19]. A novel interpretation in terms of a lattice-driven bipolaronic insulating ground state, instead of a Mott-Hubbard one, has been proposed showing a quantitative agreement with the angle-resolved photoemission spectroscopy (ARPES) spectra [18]. Contrary to former assumption, saturation coverage has been proved to be 0.5 monolayer leading to the possible alternation of empty and doubly occupied Si dangling bonds in agreement with the bipolaronic model on the hexagonal lattice [19,20].…”

“…This evidences a complex interplay between lattice and electronic degrees of freedom at surface. In this context, a novel 2 ffiffiffi 3 p  2 ffiffiffi 3 p R30 surface reconstruction (2 ffiffiffi 3 p in the following) has been evidenced close to saturation in alkali-metal/Si(111):B [18]. Angle-resolved photoemission measurements reveal the 2 ffiffiffi 3 p symmetry and an alkalimetal-dependent gap [19].…”

Absolute coverage determination in the K/Si(111):B-23×23R30∘surface

“…[25][26][27] Weitering et al proposed that it could also be the signature of a Mott insulator, 27 since they assumed a coverage of one K atom per √ 3 unit cell (1/3 ML), which should lead to one half filled, metallic surface state in a one-electron picture, 28,29 Very recently, the 2 √ 3 × 2 √ 3R30 • reconstruction ("2 √ 3" hereafter) together with a similar surface electronic structure as in Ref. 27 have been observed 30 and generalized to all alkali adsorbates, 31 Still assuming a 1/3 ML coverage would now give an even number of electrons (four) per 2 √ 3 unit cell and, therefore, produce a mere band insulator in contradiction with the former scenario. It is then very necessary to determine experimentally the absolute coverage of the 2 √ 3 surface to solve the controversy, even it is not so easy from a practical point of view.…”

Spectroscopic indications of polaronic behavior of the strong spin-orbit insulator Sr3Ir2O7

“…However, their intensity distribution, which reflects the overlap of the initial state with different excited final states, follows a Gaussian envelope. This provides a hallmark of a PHYSICAL REVIEW B 87, 241106(R) (2013) polaronic system which allows its experimental observation by photoemission, 27,[36][37][38][39] without recourse to detailed theoretical treatments or necessitating a microscopic identification of the dominant bosons. With increasing temperature, the linewidth of the Gaussian envelope broadens further, reflecting the presence of thermally excited bosons, 40 entirely consistent with our measured temperature-dependent linewidths [Figs.…”