Surface Charge Ordering Transition:αPhase of Sn/Ge(111)

Abstract

We have identified and characterized a charge-density wave transition ͑T c ϳ 260 ± C͒ in the lowcoverage a phase of the Sn͞Ge(111) interface both experimentally and theoretically. Charge ordering is accompanied by a structural distortion from ͑ p 3 3p 3 ͒R30 ± to ͑3 3 3͒ symmetry. Density-functional theory calculations are unable to correctly reproduce the observed ground state and, more importantly, indicate that Fermi surface nesting does not play a role in this transition. Both signal the importance of many-body effects in this system. Experiment and theory indicate that the Sn͞Ge(111) overlayer is fundamentally different from the Pb͞Ge (111) overlayer previously reported. [S0031-9007(97)04249-X] PACS numbers: 73.20.At, 61.16.Ch, 68.35.Md, 71.45.LrA charge-ordered state, or charge-density wave (CDW), incorporates a symmetry-lowering periodic redistribution of valence charge driven by a reduction in the system's total electronic energy, resulting in a small periodic lattice distortion [1]. This phenomenon is most likely to occur in reduced dimensions [2,3], for example, in the layered perovskites [4]. The loss of coordination at a crystal surface might also be expected to invite CDW formation, but few genuine instances of surface charge-density waves have been reported to date. One clear example was recently discovered at the Pb͞Ge(111)-a vacuum interface [5]. This low-density overlayer transforms from the room-temperature (RT) ͑ p 3 3p 3 ͒R30 ± structure to the low-temperature (LT) ͑3 3 3͒ ground state. Charge ordering (ϳ0.5 Å corrugation) and accompanying lattice distortion occur gradually and reversibly with T c ϳ 220 ± C. Density function theory calculations indicate the chargeordered ͑3 3 3͒ structure is the ground state of this system [5]. The opening of a E g ϳ 65 meV band gap below T c (later confirmed by photoemission measurements [6]) must be the consequence of correlation effects [5]. We therefore conjectured that although enhanced electron-phonon coupling (enabled by Fermi surface nesting) drives this transition, many-body interactions stabilize the ground state.Our motivation to study the a phase of Sn͞Ge(111) was threefold. First, seeking verification that Pb͞Ge(111) charge ordering is not an isolated quirk of nature resulting from Pb's exotic properties, we considered the other isostructural adsorbate/substrate combinations that exist. Al, Ga, In, Sn, and Pb all form the same low-density overlayer atop both Si(111) and Ge(111) [7,8]. However, overlayers composed of the trivalent species (Al, Ga, In) will probably not undergo a CDW distortion, as they are semiconducting [7] with an even number of electrons per unit cell at RT.Second, it is imperative to test several theoretically derived concepts, including Fermi surface nesting, and the idea that the CDW ground state is a Mott-Hubbard insulator [5]. Intuitively, the Sn overlayer bandwidth w should be less than that of the Pb overlayer, because the metallic r Sn is ϳ10% smaller than r Pb . For the same reason one expects the on-site Coulomb repuls...

First principles calculations of charge and spin density waves of -adsorbates on semiconductors

“…1. This is in agreement with recent calculations reported for the Sn/Ge(111) [4] surface, and disagrees with earlier claims for Pb/Ge(111) [1]. Since a low temperature transition to a 3 × 3 phase has been observed on both Sn/Ge (111) and Pb/Ge(111), the role of a Fermi surface nesting as the driving force of the transition remains to be clarified.…”

Surface Peierls transition on Cu(001) covered with heavier p-block metals

“…It was suggested that the second layer is composed of an In 1/4 Cu 3/4 alloy with (2 × 2) periodicity [16]. 2 The In coverage for (9 √ 2 × 2 √ 2)R45 • was first reported as 1.0, but was later determined to be 0.5 [70]. The above three surfaces undergo phase transitions upon heating.…”

“…As solid surfaces and interfaces provide quasi-two-dimensional electron systems, efforts have long been paid to find Peierls transitions, which fruited in the last decade in the discovery of intriguing surface phase transitions which were suggested to be the Peierls, or CDW, phase transition [1][2][3][4].…”

“…The transition on the Ge(111) surface covered with 1/3 monolayers of heavier p-block metallic elements 1 such as Pb and Sn [1,2] has been most intensively studied. (For reviews on this surface from various aspects, see Refs.…”

The next 25 years of surface physics

“…This room temperature (RT) structure is referred to as the a-phase, whose family of adsorbate-substrates has been studied by the surface community for $20 years [18], but it was not until 1996 that someone cooled the sample and discovered that there was a phase transition to a (3 Â 3) structure [19,20]. Carpinelli discovered that for both Sn or Pb on Ge(1 1 1) the low temperature phase was a CDW [19,20]. On the left of Fig.…”