Polar Semiconductor Surfaces Research Articles

Effective approach for calculating individual energy of step edges on polar AlN(0001) and GaN(0001) surfaces

We propose a direct approach for calculating the individual energy of step edges on polar semiconductor surfaces using ab initio calculations. The approach is applied to the single-bilayer step edges on AlN(0001) and GaN(0001) surfaces and clarifies characteristic features of the stability of step edges depending on the atomic configurations. It is found that the stable steps have either threefold-coordinated atoms or hydrogen-terminated edge atoms on the upper terrace. The calculated results suggest that the phenomena related to the stability of step edges on various polar group-III nitride surfaces would benefit from the evaluations of step formation energies.

Tuning hydrogen adsorption on pure and doped ZnO (0001¯) surfaces by a simple electron counting model

Hydrogen (H) adsorption strengths on oxygen-terminated (0001¯) surfaces of pure and doped wurtzite ZnO are investigated under varying H surface coverage conditions. Consistent with the prediction of the classical electron counting rules, a 12 monolayer (ML) of adsorbed H changes the electronic structure of pure ZnO (0001¯) surface from metallic to semiconductor state by saturating unpaired electrons of surface oxygen atoms. This closed-shell electron configuration of the ZnO (0001¯) surface significantly reduces the adsorption strengths of subsequent H atoms, making the dissociative adsorption of a H2 molecule endothermic. We apply a simple electron counting model to predict and tune the coverage-dependent H adsorption strengths on general polar semiconductor surfaces. This model is confirmed by our investigations of H adsorption on (0001¯) surfaces of ZnO with a series of dopant elements (Na, Mg, Al, Ti, Fe, Sn, etc.). It can also be applied to H adsorption on other similar polar semiconductors, such as ZnO (0001¯) containing O vacancies, wurtzite GaN (0001¯), and zincblende ZnS (1¯1¯1¯) surfaces.

Growth mechanism and surface atomic structure of AgInSe2

The growth of (112)A-oriented AgInSe2 on GaAs (111)A and its surface reconstruction were studied by scanning tunneling microscopy, atomic force microscopy, and other techniques. Films were grown by a sputtering and evaporation method. Topographic STM images reveal that the film grew by atomic incorporation into surface steps resulting from screw dislocations on the surface. The screw dislocation density was ∼1010 cm2. Atomically resolved images also show that the surface atomic arrangement appears to be similar to that of the bulk, with a spacing of 0.35–0.41 nm. There is no observable reconstruction, which is unexpected for a polar semiconductor surface.

Surface energy and the common dangling bond rule for semiconductors.

Equilibrium shape and surface energies are among the most basic properties of finite crystals. Yet, an effective approach for accurately calculating individual energy for polar semiconductor surfaces is still lacking, and there is not a general rule regarding surface energies of different orientations. Here, we suggest a wedge-shaped geometry for calculating individual surface energies by direct, first-principles methods. Applications to prototypical semiconductors, Ge, GaAs, and ZnSe, establish a surprisingly simple common dangling bond rule relating surface energies to local chemical similarities.

Schottky barrier heights at polar metal/semiconductor interfaces

Using a first-principle pseudopotential approach, we have investigated the Schottky barrier heights of abrupt Al/Ge, Al/GaAs, Al/AlAs, and Al/ZnSe (100) junctions, and their dependence on the semiconductor chemical composition and surface termination. A model based on linear-response theory is developed, which provides a simple, yet accurate description of the barrier-height variations with the chemical composition of the semiconductor. The larger barrier values found for the anion- than for the cation-terminated surfaces are explained in terms of the screened charge of the polar semiconductor surface and its image charge at the metal surface. Atomic scale computations show how the classical image charge concept, valid for charges placed at large distances from the metal, extends to distances shorter than the decay length of the metal-induced-gap states.

Evolution of elementary excitations at a doped polar semiconductor surface in a depletion-layer formation process

We investigate the evolution of elementary excitations at a doped polar semiconductor surface in a depletion-layer formation process. The elementary excitations analyzed are two coupled plasmon-phonon modes and a surface optical-phonon mode involving screening charges. Surface excitations of free carriers are calculated by taking account of the Coulomb interaction between carriers, the dynamical exchange-correlation effect, and the coupling with surface polar phonons, on the basis of thermal-equilibrium states calculated self-consistently by the local-density approximation. We focus our attention on the coupling character and the spatial structure of each surface-excitation mode, as well as the energy dispersion and the energy-loss intensity involved in the excitation. The induced charge-density distribution in the excitation mode is composed of a carrier component due to carrier density fluctuation, a phonon component arising from longitudinal polarphonon polarization, and two on-surface components originating from the termination of the polar phonon and background polarization at the surface. The coupling character is elucidated by the phase relation and the amplitude ratio among these induced charge components. The spatial structure is visualized by the contour map of the induced charge-density distribution. We follow the variation in the coupling character and the spatial structure in the depletion-layer formation process. Our analysis gives a clear explanation of the variation in each of the three loss peaks in the electron energy-loss spectrum.

Elementary excitations at doped polar semiconductor surfaces with carrier-depletion layers

We make a comparative analysis of coupled carrier plasmon–polar phonon modes at doped polar semiconductor surfaces in the absence and the presence of a carrier-depletion layer. Our analysis shows how the effect of the depletion layer appears in the spatial structure of each excitation mode visualized in a contour map of the induced charge-density, the coupling character elucidated by the phase relation and the amplitude ratio of the carrier component and the polar-phonon component in the induced charge-density distribution, and the energy-loss intensity due to the surface excitation.

Absence of a step-edge barrier on a polar semiconductor surface with reconstruction

National Research Institute for Metals, 1-2-1, Sengen, Tsukuba, Ibaraki 305-0047, Japan~Received 9 December 1999!Based on purely geometrical considerations, we point out that a polar semiconductor surface with recon-struction does not satisfy the necessary condition for the existence of an Ehrlich-Schwoebel~ES! step-edgebarrier. We further show by the explicit Monte Carlo calculations that the kinetic surface roughening observedon a GaAs~001! surface can be accounted for by the stability and complexity of the (234) reconstruction, andhence an ES barrier is unnecessary to account for it. Finally, we point out that it is incorrect to use thesolid-on-solid~SOS! model with an ES barrier for surface growth studies or for atomistic growth simulations.I. INTRODUCTION

Chemical reactions of ammonia with polar and non-polar nitride semiconductor surfaces

We present results from ab-initio local-orbital calculations performed to study the impact of ammonia on GaN. We focus on chemical reactions on the non-polar (101̄0) and (112̄0) surfaces as well as the dissociative adsorption of NH 3 on GaN(0001̄), for which an energy barrier is obtained by computing the potential energy surface. For the adsorption of NH 3 on GaN(0001̄), finite-temperature molecular-dynamics simulations show that the molecule splits into NH 2 and H. The NH 2 group binds to two of the three second-layer Ga atoms neighboring the vacancy, while the dissociating H binds to a first-layer nitrogen atom. This reaction lowers the total energy of the system by 2.05 eV. The energy barrier estimated for dissociative adsorption of NH 2 and H on the vacancy p(2×2) reconstruction of the GaN(0001̄) surface is 0.5 eV.

Accounting for stoichiometry changes on compound semiconductor surfaces

We discuss how to self-consistently account for stoichiometry changes on surfaces with complex reconstructions, e.g. polar compound semiconductor surfaces. The key aspect of the methodology is that surface steps are allowed to act as a reservoir where atoms may be added or removed. The method is specifically applied to molecular beam epitaxy (MBE) and atomic layer epitaxy (ALE) results for GaAs(100) surfaces. The method easily demonstrates why only ∼ 12 monolayer (ML) of As is needed to convert from the Ga-rich surface to the (2 × 4) As-rich surface (as observed experimentally), even though the latter surface has an As coverage of 34 ML in the top layer. We also demonstrate how to convert from a more complex reconstruction, having two incomplete layers, to a simpler reconstruction having only one incomplete layer. The methodology also shows that ideal ALE of GaAs(100) cannot occur by cycling between the known adsorbate-free Ga-rich and As-rich surface reconstructions, because any such transition would not yield the observed 1 ML per cycle growth rate. We briefly discuss how adsorbates may stabilize ideally terminated (i.e. vacancy free) III–V surfaces. For example, methyl groups adsorbed on GaAs(100) exhibit a (1 × 2) LEED pattern, which is not seen for the clean GaAs(100) surface reconstructions. By using the electron counting model we interpret this structure as 12 ML CH3 adsorbed on a complete layer (1 ML) of dimerized Ga atoms. The ideal termination of the Ga-rich GaAs(100)-(1 × 2)-CH3 surface now allows for a plausible ALE mechanism which yields 1 ML deposition per cycle.

Surface stoichiometry and the role of adsorbates during GaAs atomic layer epitaxy

Several questions regarding the stoichiometry of atomic layer epitaxy (ALE) arise because most polar compound semiconductor surfaces reconstruct into structures terminated with less than monolayer coverage at the outermost layer. In this paper we briefly discuss how to self-consistently account for stoichiometry changes on surfaces that are not ideally terminated. The key aspect of the methodology is that surface steps are allowed to act as a reservoir where atoms may be added or removed. The methodology shows that ideal ALE of GaAs(100) cannot occur by cycling between the known adsorbate-free Ga-rich and As-rich surface reconstructions, because no such transition would yield the observed 1 monolayer (ML) per cycle growth rate. In fact, ideal ALE (1 ML/cycle) must involve at least one adsorbate induced surface reconstruction. Adsorbates may stabilize ideally terminated (i.e. vacancy free) III–V surfaces because of their ability to passivate dangling bond states. For example, methyl groups adsorbed on GaAs(100) exhibit a (1 x 2) LEED pattern, which is not seen for the clean GaAs(100) surface reconstructions. By using the electron counting model we interpret this structure as 1 2 ML CH 3 adsorbed on a complete layer (1 ML) of dimerized Ga atoms. The GaAs(100)-(1 x 2)-Ch 3 surface was also examined using surface infrared spectroscopy (SIRS) using a multiple internal reflection geometry. This surface exhibits relatively sharp infrared linewidths suggestive of a well ordered structure. The polarization dependence of the symmetric stretching and bending CH 3 modes also supports our proposed structure of the GaAs(100)-(1 x 2)-Ch 3 surface. The ideal termination of the Ga-rich GaAs(100)-(1 x 2)-CH 3 surface for a plausible ALE mechanism which yields 1 ML deposition per cycle.

Structure of GaAs(001) surfaces: The role of electrostatic interactions.

We report first-principles total-energy calculations for the GaAs(001) surface. Our results indicate that the 2[times]4 reconstruction corresponds to the [beta]2(2[times]4) structure, which exhibits two As dimers in the top layer and a third As dimer in the third layer. This structure has a lower surface energy than the [beta](2[times]4) model, which has three As dimers in the top layer. We also find that a model recently proposed by Skala [ital et] [ital al]. [Phys. Rev. B [bold 48], 9138 (1993)] for the structure of the Ga-rich 4[times]2 phase is energetically unfavorable. From our results we conclude that electrostatic interactions between the charged building blocks of polar semiconductor surfaces play an important role in determining the equilibrium structure. We introduce a simple model for estimating these interactions.

LCAO‐Calculations of the Valence Band Structure of Modified Polar Semiconductor Surfaces: Hydrogen‐, Copper‐, and Ruthenium‐Covered Gallium Arsenide (001)

AbstractResults of local density first principles LCAO calculations for a 13 layer slab of free and modified As‐terminated (1 × 1) GaAs(001) surfaces are presented. The surfaces are covered by a monolayer of hydrogen, copper, or ruthenium. The surface calculation for the hydrogen covered surface gives hydrogen‐induced and GaAs surface states in agreement with angle‐resolved photoemission results within the energy range where they are found. For the systems GaAs(001):Cu and GaAs(001): Ru metal induced gap states are predicted with a density of 2 × 1015 cm−2. The calculated surface states indicate a higher Schottky barrier for the ruthenium than for the copper covered surface.

A New Slab Model Approach for Electronic Structure Calculation of Polar Semiconductor Surface

A new slab model approach is examined for the electronic structure calculation of a polar semiconductor surface. Our proposed slab model contains fractionally charged H atoms which completely terminate a noninteresting surface of the slab. We have studied in detail the electronic structure of the polar GaAs(001) surface in order to examine the reliability of our slab model. Calculation is performed by the ab initio pseudopotential method. We have found that our slab model can accurately describe a polar semiconductor surface. In addition, this model can greatly reduce the computational time.

Origin of two distinct surface optical phonon modes on an n-type polar semiconductor surface

We investigate surface optical phonons on a doped polar semiconductor surface for the case of heavy doping where carrier electrons are well degenerate and where plasmon energies are higher than phonon energies. Our theoretical analysis asserts that there exist two distinct surface optical phonon modes, which form two energy dispersion branches, and our analysis clarifies the origin of these two distinct surface optical phonon modes. On the basis of our analysis the energy dispersion of the surface phonon on an n-type GaAs surface which was recently observed by electron energy loss spectroscopy can be interpreted as switching from one dispersion branch to the other.

Dielectric response of a degenerate polar semiconductor surface

We investigate the surface dielectric response of a degenerate polar semiconductor, on the basis of a model which involves electron gas response and phonon polarization. Carrier electrons are described by a degenerate electron gas in the infinite-barrier jellium model and the random phase approximation. Phonon polarization is described by a Lorentzian-type oscillator model. Our formulation is very powerful for calculating spatial structure of surface excitation modes. We apply our model to an n-type InSb surface, in conjunction with an EELS experiment, in particular, with variation of plasmon and phonon loss features with exciting beam energy.

Schottky Barriers without Midgap States

It is found that modification of the surface reconstruction on clean silicon surfaces could automatically bring the Fermi energy at the surface to midgap. The mechanism does not depend upon the existence of surface states in the gap. The corresponding effect is not expected on polar semiconductor surfaces.