有限二维复式晶格电子结构的尺度效应

以二维复式晶格作为有限系统的集团模型,利用紧束缚近似,在考虑链间耦合及链端效应的情况下,计算了格点数分别为16、32、64、128时π电子在最近邻及次近邻跳跃集团的能谱和态密度.讨论了不同格点数和结构参数对态密度及带宽的影响.     

有限系统的电子结构研究

选取二维和三维亚方晶格作为有限系统的集团模型，在紧束缚近似下，分别计算了不同尺寸集团的态密度。探讨了态密度与集团大小的关系以及受表面效应的影响。     

铁超微粒聚集体的分形结构

由铁超微粒组成的聚集集团的形态结构具有明显的分形体特征和规律性.随着集团聚集密度的增加,呈现出扩展态、补偿态和塌缩态,而且自相似性逐步减弱. 关键词：     

YBa_2Cu_3O_7中Cu、O离子价态的研究

采用不同的原子集团模型研究了 YBa_2Cu_3O_7高 Tc 超导材料中 Cu、O 离子的价态,用局域密度泛函理论自洽求解单电子方程得到系统的电子结构及离子价态.详细讨论了原子集团周围环境离子的电子分布对原子团内 Cu、O 离子价态的影响.结果表明,Cu 离子均为+2价,空穴分布在氧上,这与实验结果一致.在四种氧位中,空穴主要分布在 Ba-O 层内的O(4)上,空穴的分布易受原子团周围离子的电子分布变化的影响.     

哈密顿的维度相关性与声子态密度的卷积律

根据一、二、三维体系晶格振动哈密顿量之间相互关系,给出了不同维度声子态密度之间的卷积规律,分析了态密度的卷积条件及卷积所得态密度所对应的物理图象,阐述了态密度卷积方法在求解晶格振动态密度时的可行性及局限性.最后,给出并讨论了本方法获得的几种不同体系的晶格振动态密度结果.     

Pt基电极材料硫中毒机理的计算机模拟研究

采用基于密度泛函理论的第一性原理方法和超原胞模型,计算并分析了Pt及Cu掺杂的Pt基合金的表面态密度,以及S在Pt及Pt基合金表面的吸附能和态密度情况.考虑了多种掺杂体系及吸附构型,结果表明：Cu掺杂会降低Pt基体系表面费米能级附近态密度,Pt皮肤的存在可以有效地减小Cu掺杂对体系表面态密度的影响;与S在纯Pt表面吸附相比较,S在掺杂体系表面的吸附能较小,且S的吸附对掺杂体系费米能级附近表面态密度影响较小. 以上研究结果有助于为改善Pt基电极材料的催化性能,尤其是其抗S中毒性能提供理论依据.     

不同晶系应变Si状态密度研究

应变Si技术是当前微电子领域研究发展重点,态密度是其材料的重要物理参量.本文基于应力相关KP理论,建立了(001),(101)和(111)晶面施加双轴应力形成的四方、单斜及三角晶系应变Si导带、价带态密度模型.结果表明,除单斜和三角晶系导带底态密度外,应力对其余各态密度均有显著影响.本文所得模型数据量化,可为应变Si材料物理的理解及其他物理参数模型的建立奠定重要理论基础. 关键词： 应变Si KP 态密度     

基于1/f噪声的NPN晶体管辐照感生电荷的定量分离

本文针对NPN双极性晶体管, 在研究辐照感生的氧化层电荷及界面态对晶体管基极电流和1/f噪声的影响的基础上, 建立辐照感生氧化层电荷及界面态与基极电流和1/f噪声的定量物理模型. 根据所建立的模型, 提出一种新的分离方法, 利用1/f噪声和表面电流求出氧化层电荷密度, 利用所求得氧化层电荷密度和表面电流求出界面态密度. 利用本方法初步实现了辐照感生氧化层电荷及界面态的定量计算.     

具有态守恒赝隙的光子晶体中两能级原子自发辐射的增强与抑制

论证了在赝带隙光子晶体中存在一个全频率域态总数守恒规则，在完全带隙光子晶体中还存在一个局域态总数守恒规则.态总数守恒规则指出，如果一个光子晶体的态密度在某些频率范围存在相对于等效介质态密度的谷，则一定由其他频率范围内相对于等效介质态密度的峰来补偿.使用符合态总数守恒规则的态密度模型，解释了态密度调制导致的自发辐射谱增强、抑制、变窄、红移、蓝移以及谱分裂等光子晶体中的量子光学现象.该理论比较适合研究在具有赝带隙的光子晶体中大量随机分布的发光原子或分子的自发辐射行为. 关键词： 光子晶体 自发辐射 态密度 光子赝带隙     

Ⅲ族元素掺杂对SnO2电子结构及电学性能的影响

采用密度泛函理论研究了Ⅲ族元素掺杂对SnO2电子结构及电学性能的影响.态密度分析结果表明,以替代位存在的Ⅲ族杂质均使SnO2的费米能级明显向低能态方向移动,使得价带顶不完全填满,因此在SnO2中均充当受主作用.部分态密度分析结果表明,相对于掺Al的SnO2,Ⅲ族元素中的Ga及In对费米能级附近态密度贡献较大,其主要贡献来自Ga3d态或In4d态,这预示着在SnO2中掺Ga或In能实现更好的p型掺杂效果.电离能计算结果进一步表明,在Al,Ga及In三种元素中,替位In有最小的电离能(0.06 eV),这说明其在SnO2中能形成最浅的受主能级,因而在同等掺杂情况下,可引入最高浓度的空穴,从而实现最佳的p型掺杂效果.     

A tight binding analysis of the step-dependent density of 〈111〉 surface states for covalent semiconductors

The influence of steps on the electronic dangling bonds surface states is studied in a tightbinding approximation. It is first shown how surface states can be studied separately as a two dimensional problem, using the concept of effective interactions between dangling bonds. This technique is applied to perfect 〈111〉 surfaces where a nearest neighbour approximation reproduces fairly well the existing dispersion curves. This simplified model is then used for a systematic analysis of the existence of localized states associated with steps. The results are found to provide a theoretical interpretation of the observed features appearing when step densities are important.     

Equilibrium states of an Ising ferromagnet in the low temperature region

We investigate low temperature properties of an Ising ferromagnet when the nearest neighbour coupling constant is dominant. We show that all the translationally invariant equilibrium states are a superposition of only two extremal states.     

On the influence of size and roughness on the electronic structure of transition metal surfaces

The dependence of surface electronic structure on local and farther environment for transition metals is investigated in the tight binding approximation associated with the moment method. Local densities of states (LDS) on fcc cubooctahedral and on icosahedral clusters of increasing size, from 13 up to 2057 atoms, are studied for corner sites and central sites on differently oriented faces. The general LDS features are determined by the nearest neighbour shell but outstanding details, such as the occurrence of a strong central peak, are dependent on each particular cluster site and dimension. Only for particles larger than one thousand atoms the LDS at the center of (100) and (111) faces look like those of the corresponding infinite surfaces. The importance of symmetry is pointed out by the comparison of LDS at equivalent sites on cubo-octahedra (cubic symmetry) and icosahedra (five-fold symmetry). These results might explain the variation of activity with the catalyser dispersion, as the adsorption energy depends on the relative position of the adatom atomic level and the substrate LDS peaks and Fermi level.     

Optimum ground states of generalized Hubbard models with next-nearest neighbour interaction

We investigate the stability domains of ground states of generalized Hubbard models with next-nearest neighbour interaction using the optimum groundstate approach. We focus on the -pairing state with momentum P=0 and the fully polarized ferromagnetic state at half-filling. For these states exact lower bounds for the regions of stability are obtained in the form of inequalities between the interaction parameters. For the model with only nearest neighbour interaction we show that the bounds for the stability regions can be improved by considering larger clusters. Additional next-nearest neighbour interactions can lead to larger or smaller stability regions depending on the parameter values. Received 30 March 1999 and Received in final form 3 May 1999     

On the existence conditions of localized states in linear chains

The existence conditions of localized states in linear chains with one or two impurities, are investigated in the tight-binding model. The influence of the coupling between impurity and its neighbour is stressed, in the cases of infinite and semi-infinite chains.     

Self-consistent calculations for the electronic structure of a vacancy in copper. A solution of the embedding problem

The charge density and the local density of states for a vacancy in Cu and for the first shell of Cu neighbours are calculated by the KKR-Green's function technique. The muffin-tin potentials for the vacancy and the neighbour shell atoms are determined self-consistently in the local density approximation of density functional theory. By the use of the proper host Green's function the embedding of this cluster of 13 perturbed muffin-tins into the infinite array of bulk Cu muffin-tin potentials is described rigorously, thus representing a solution of the embedding problem. The calculations demonstrate a rather large charge transfer of 1.1 electrons from the first neighbour shell to the vacancy.     

Ab-initio calculation of the electronic structure and energetics of the unreconstructed Au(001) surface

The electronic structure, surface and relaxation energies, and the electric field gradient for the unreconstructed Au(001) surface were calculated by means of the ab-initio all-electron full-potential linearized augmented plane wave slab method. The valence states were calculated within the standard semi-relativistic approach whereas the core states are treated in a fully relativistic way. The Au(001) surface was modelled by free slabs of 5, 7, and 9 layers. From the 9-layer calculation a work function of 5.39 eV was obtained. For the surface energy a value of 1.30 J/m² for the unrelaxed geometry was derived from the total energies of the 7- and the 9-layer slabs. From total energy minimization of the 7-layer slab, a negative, inward relaxation of −2.6% and a relaxation energy of 14.3 × 10⁻³ J/m² were derived. To discuss a mechanism of reconstruction, particular surface states were analyzed in detail in terms of the band structure, layer-dependent density of states and the charge density distribution. Differences of surface and central-layer charge densities show a gain of charge in z-direction localised below and also, to a smaller extent, above the surface atoms. We find a very small gain of delocalised charge in the surface plane between the nearest neighbour positions at the expense of more localised s-d hybridised states. The electric field gradient component Φ_zz was obtained in a two energy window calculation for which the Au5p states were also treated as band states. The resulting Φ_zz values are −16.50 × 10¹⁷ V/cm² surface layer, and −3.3 × 10¹⁷ V/cm² for the subsurface layer.     

The density dependence of the diffusion constant in interacting lattice gases: Application to surface diffusion of 0 onW(110)

The influence of interactions between adsorbed particles on their diffusion constant is investigated by kinetic Ising models with independent nearest neighbour hops. This leads to expressions for the ratio of the diffusion constant at arbitrary coverage to its value at =0 as a function of the interaction energies relative to temperature. It is shown that under certain conditions this quantity obeys a particle-hole symmetry. Exact results in the whole range of densities are given in one dimension for nearest neighbour interaction. They already yield a qualitative agreement with experimental results and are also compared to corresponding numerical simulations. The introduction of a next nearest neighbour interaction is shown to produce drastic changes in the density dependence of the diffusion constant in some of the cases. A generalized quasichemical approximation and a virial expansion are made in two dimensions, leading to a better agreement with the measurements.     

Electronic vacancy states in silicon by the chemical pseudopotential method

The electronic states of a neutral vacancy in Si are studied through the chemical pseudopotential method by creating a vacancy in a large crystal unit cell containing up to 54 atoms. A localized vacancy state is found in the forbidden gap and its energy is shown to be convergent with respect to the size of the cell. The density of states of the valence band is modified by the presence of the vacancy with additional peaks which give charge localization on the vacany nearest neighbour atoms.     

Competing interactions in double-chains: Ising and spherical models

We investigate the thermodynamic properties of double chains of Ising and spherical spins with different first and a crossed second neighbour interaction in zero field. The interest is focussed on the region where different ground states are nearly degenerate due to competing interaction constants. The Ising system shows quasi-singular behaviour of the susceptibility for certain ratios of parameters. Moreover the nearest neighbour correlation function exhibits a sharp crossover from high-temperature “compensation-point” to low temperature ferro- or antiferromagnetic behaviour. An analogy is found between compensation points and tricritical points of higher dimensional systems.