scispace - formally typeset
Search or ask a question

Showing papers on "Band offset published in 1990"


Journal ArticleDOI
TL;DR: In this article, the conduction and valence-band offsets for GaAs/Ga0.51In0.49P quantum wells were independently estimated by measuring the capacitance transient resulting from thermal emission of carriers from the respective wells.
Abstract: We have independently estimated the conduction‐ and valence‐band offsets ΔEc and ΔEv in GaAs/Ga0.51In0.49P quantum wells by measuring the capacitance transient resulting from thermal emission of carriers from the respective wells. The heterostructure samples were grown by low‐pressure metalorganic chemical vapor deposition. The band offsets are extrapolated from the emission activation energies with appropriate corrections. The estimated values of ΔEc and ΔEv are 0.198 and 0.285 eV, respectively.

131 citations


Journal ArticleDOI
TL;DR: In this paper, the authors analyzed the effect of strain on the valence subband of GaxIn1−xAs/InP single-quantum-well structures using the model solid theory of Van de Walle and Martin.
Abstract: Single‐quantum‐well structures were grown by atmospheric pressure organometallic vapor‐phase epitaxy, with GaxIn1−xAs layers (0≤x≤1) coherently strained to match the lattice parameter of the InP barrier layers in the (100) growth plane. The strain effects on the band lineups were analyzed using the ‘‘model solid’’ theory of Van de Walle and Martin. The hydrostatic strain component for alloys with x≊1 is shown to be sufficient to marginally convert the type‐II lineups for the unstrained case to type I. The band lineups remain type I for x≊0. Considering the effect of strain, the ‖ (3)/(2) , (1)/(2) 〉 valence subband becomes a slowly varying function of x. Band offsets are predicted over the entire alloy composition and compared with the reported data. The photoluminescence (10 K) peak energies for the 100‐A GaxIn1−xAs/InP single quantum wells compare quite favorably with the calculated strained band gap versus x. For nominal monolayer quantum wells, the peak energies are slightly above 1.1 eV over the enti...

103 citations


Journal ArticleDOI
TL;DR: Etude dans le cadre de la fonctionnelle de densite, dans l'approximation de la densite locale, en conjonction avec une geometrie du superreseau pour resoudre l'equation de Schrodinger de l'heterojonction.
Abstract: A theory of heterojunction band offsets is developed within the density-functional framework in the local-density approximation. The linear-muffin-tin-orbital method is used in conjunction with a superlattice geometry for solving Schr\"odinger's equation of the heterojunction. The potential is constructed within the atomic-sphere approximation. Within this context the long-range electrostatics reduce to that of point charges, and the average electrostatic potential of the latter can be used as a local reference level. It is shown that, starting from an arbitrary alignment of the bulk potentials, the correct potential alignment is obtained by minimizing the total energy with respect to a single parameter: the interface dipole. This minimization is equivalent to screening the initially induced dipole by the macroscopic dielectric constant of the interface region of the heterojunction, which can be identified approximately with the harmonic average of the dielectric constants of the two semiconductors. The conditions under which this is valid are discussed. The calculations are performed within a so-called frozen-shape approximation, allowing the potentials to vary only by constant shifts. Almost perfect agreement between calculations using a different shift per atomic layer and calculations using a single shift per semiconductor provide a numerical demonstration that the self-consistent dipole is independent of the details of the dipole profile. They also show that the macroscopic dielectric constant of the heterojunction in the vicinity of the interface can be obtained with reasonable accuracy from the single-parameter variational calculation itself. The calculations also show that linear response is valid over a wide range. The theory is applied to an extensive set of lattice-matched semiconductor (110) interfaces and shown to be in excellent agreement with experimental results and previous more-involved calculations where available. The consequences of the present theory for interface-orientation dependence and metal-semiconductor interfaces are briefly discussed.

94 citations


Journal ArticleDOI
TL;DR: Pseudomorphic InGaAs/GaAs multiple quantum well structures with In contents ranging from 18 to 25% were grown by molecular beam epitaxy and investigated by optical absorption, photoluminescence, and electronic Raman scattering as discussed by the authors.
Abstract: Pseudomorphic InGaAs/GaAs multiple quantum well structures with In contents ranging from 18 to 25% were grown by molecular beam epitaxy and investigated by optical absorption, photoluminescence, and electronic Raman scattering Sharp exciton peaks with linewidths of ∼3 meV for the first electron to heavy hole transition are observed in the absorption spectra The electron subband structure was investigated independently by electronic Raman scattering The transition energies are analyzed using a four‐band effective mass Schrodinger equation taking strain into account A conduction‐band offset ratio ΔEc/ΔEg=06 is found for all samples independent of In content

79 citations


Journal ArticleDOI
TL;DR: The (111) double layer with the maximum number of Ge-Ge bonds is found to be energetically the most stable and to have the smallest dipole, but it is still large compared to typical band offsets in semiconductors.
Abstract: We have carried out ab initio self-consistent calculations to investigate the extent to which the band offset can be modified by polar layers. The cases studied are double layers of Ge in bulk GaAs (for which the offsets are totally due to induced dipoles) in (100) and two distinctly different (111) geometries. The (111) double layer with the maximum number of Ge-Ge bonds is found to be energetically the most stable and to have the smallest dipole, but it is still large (0.7 eV) compared to typical band offsets in semiconductors. We argue that this demonstrates the potential possibility of providing a mechanism for band-offset engineering using thin layers at interfaces.

75 citations


Journal ArticleDOI
TL;DR: In this paper, low-temperature photoluminescence excitation spectra of disordered Ga 0.5In0.5P/Al 0.3Ga 0.2In 0.
Abstract: Low‐temperature photoluminescence excitation spectra of disordered Ga0.5In0.5P/Al0.3Ga0.2In0.5P quantum wells lattice matched to (311)B GaAs substrates have been measured for the first time. Transition energies calculated with a k⋅p approach agree with experiment within 3 MeV, over the entire range of quantum well thicknesses (Lz=11–109 A). A conduction‐band discontinuity of 0.65±0.05 is derived.

69 citations


Journal ArticleDOI
TL;DR: In this article, the authors used electrolyte electroreflectance (EER) to characterize ZnSe/GaAs and ZnS/AlAs interfaces and demonstrated that these interfaces are type I.
Abstract: We have used electrolyte electroreflectance (EER) to characterize ZnSe/GaAs and ZnSe/AlAs interfaces. The great sensitivity of EER to interface space‐charge regions enabled us to detect both interface crossover transitions and transitions to triangular‐well interface states. The observation of these transitions provides the first unambiguous proof that the ZnSe/GaAs interface is type I and allowed us to determine the band offsets and band bendings, the diffusion lengths across each interface, and the amount of interdiffusion.

47 citations


Journal ArticleDOI
TL;DR: Observation d'une attenuation exponentielle de l'emission du niveau de cœur 4d de Te en fonction of l'eapisseur de la couche de CdS, ainsi que la persistence du cliche de diffraction des electrons de haute energie de reflexion de type substrat isabelle.
Abstract: Etude par emission photoelectronique. Observation d'une attenuation exponentielle de l'emission du niveau de cœur 4d de Te en fonction de l'eapisseur de la couche de CdS, ainsi que la persistence du cliche de diffraction des electrons de haute energie de reflexion de type substrat. Ceci demontre que les couches sont epitaxiques. Analyse de la structure de bande et de la discontinuite de la bande de valence de CdS

44 citations


Journal ArticleDOI
TL;DR: In this article, the authors used x-ray photoelectron spectroscopy to measure the valence-band offset in situ for strained Si/Ge(100) heterojunctions grown by molecular beam epitaxy.
Abstract: We have used x‐ray photoelectron spectroscopy to measure the valence‐band offset in situ for strained Si/Ge (100) heterojunctions grown by molecular beam epitaxy. Si 2p and Ge 3d core level to valence‐band‐edge binding energies and Si 2p to Ge 3d core level energy separations were measured as functions of strain, and strain configurations in all samples were determined using x‐ray diffraction. Our measurements yield valence‐band offset values of 0.83±0.11 eV and 0.22±0.13 eV for Ge on Si (100) and Si on Ge (100), respectively. If we assume that the offset between the weighted averages of the light hole, heavy hole, and spin‐orbit valence bands in Si and Ge is independent of strain, we obtain a discontinuity in the average valence‐band edge of 0.49±0.13 eV.

42 citations


Journal ArticleDOI
Mark S. Hybertsen1
TL;DR: First-principles total-energy-minimization calculations show that the interface bonds are significantly strained in nominally lattice-matched As/InP(001) heterostructures, in agreement with recent x-ray measurements.
Abstract: First-principles total-energy-minimization calculations show that the interface bonds are significantly strained in nominally lattice-matched ${\mathrm{In}}_{0.53}$${\mathrm{Ga}}_{0.47}$As/InP(001) heterostructures, in agreement with recent x-ray measurements. Anion intermixing at the interface reduces this strain. The calculated valence-band offset is sensitive to interface bond lengths so an energetically (meta)stable interface structure must be used. Then the calculated valence-band offset is independent of intermixing (hence measured strain) and in quantitative agreement with experiment.

41 citations


Journal ArticleDOI
TL;DR: In this paper, the optical properties of ZnSe-ZnSxSe1-x strained-layer superlattices (SLSs) are described and compared with the theoretical predictions of lattice dynamic models for multilayer structures.

Journal ArticleDOI
TL;DR: In this article, the authors used x-ray photoelectron spectroscopy to measure the valence band offset in situ for strained Si/Ge (100) heterojunctions and for AlSb/ZnTe (100), grown by molecular-beam epitaxy.
Abstract: We have used x-ray photoelectron spectroscopy to measure the valence band offset in situ for strained Si/Ge (100) heterojunctions and for AlSb/ZnTe (100) heterojunctions grown by molecular-beam epitaxy. For the Si/Ge system, Si 2p and Ge 3d core level to valence band edge binding energies and Si 2p to Ge 3d core level energy separations were measured as functions of strain, and strain configurations in all samples were determined using x-ray diffraction. Our measurements yield valence band offset values of 0.83±0.11 eV and 0.22±0.13 eV for Ge on Si (100) and Si on Ge (100), respectively. If we assume that the offset between the weighted averages of the light-hole, heavy-hole, and spin-orbit valence bands in Si and Ge is independent of strain, we obtain a discontinuity in the average valence band edge of 0.49±0.13 eV. For the AlSb/ZnTe (100) heterojunction system, we obtain a value of –0.42±0.07 eV for the valence band offset. Our data also suggest that an intermediate compound, containing Al and Te, is formed at the AlSb/ZnTe (100) interface.

Journal ArticleDOI
TL;DR: In this paper, the conduction-band offset in GaAs•Ga1−xAlxAs multiquantum wells was measured and the authors showed that the usual approximation in calculating the Fermi level at 0 K can lead to significant error in the determination of the band offset.
Abstract: We report measurements of the conduction‐band offset in GaAs‐Ga1−xAlxAs multiquantum wells. Admittance spectroscopy is used to measure the activation energy for thermionic emission of electrons over GaAlAs barriers and we discuss the validity of the technique in the presence of a low electric field. We show that the usual approximation in calculating the position of the Fermi level at 0 K can lead to significant error in the determination of the band offset. For an alloy composition x=0.39, we found a conduction‐band offset of 303 meV.

Journal ArticleDOI
TL;DR: In this article, strong luminescence efficiency in the blue was observed with spectra showing only weak alloy disorder effects, and a measurement for the band offsets in a model which properly included exciton effects was obtained.
Abstract: Optical characteristics of a new, weakly strained wide‐gap II‐VI quantum well have been studied with emphasis on the confined exciton states Strong luminescence efficiency in the blue is observed with spectra showing only weak alloy disorder effects Magneto‐optical studies yield a measurement for the band offsets in a model which properly includes exciton effects

Journal ArticleDOI
TL;DR: On considere les trois principales orientations cristallographiques cristal periodique moyenne de ces heterojonctions de l'interface dans laquelle l' interface est consideree comme une perturbation dans un cristAL periodiquemoyenne.
Abstract: On considere les trois principales orientations cristallographiques ((001), (110) et (111)) de ces heterojonctions. Analyse des resultats selon une approche de reponse lineaire dans laquelle l'interface est consideree comme une perturbation dans un cristal periodique moyenne. Description des effets des contraintes interfaciales et des autoenergies induites par le desordre et la correlation electronique

Journal ArticleDOI
TL;DR: In this paper, the authors used heterostructure bipolar transistors to experimentally determine band offsets in lattice-matched In0.53Ga0.47As and Al0.48In0.52As/Al0.5As devices.
Abstract: Heterostructure bipolar transistors are used to experimentally determine band offsets in lattice‐matched In0.53Ga0.47As devices. Valence‐band offsets of ΔEV=0.24 eV for Al0.48In0.52As/In0.53Ga0.47As and ΔEV=0.34 eV for InP/In0.53Ga0.47As are measured. Because of band filling in the base, these values place important constraints on p‐type doping levels and emitter injection efficiency in practical devices.

Journal ArticleDOI
TL;DR: The results show that the Katnani-Margaritondo method is unreliable in determining offsets for heterojunctions where significant Fermi-level pinning may occur and where the local structure and chemistry of the interface depends strongly on the specific heterojunction.
Abstract: Synchrotron-radiation soft-x-ray photoemission spectroscopy was used to investigate the development of the electronic structure at the CdS/${\mathrm{CuInSe}}_{2}$ heterojunction interface. CdS overlayers were deposited in steps on single-crystal p- and n-type ${\mathrm{CuInSe}}_{2}$ at 250 \ifmmode^\circ\else\textdegree\fi{}C. Results indicate that the CdS grows in registry with the substrate, initially in a two-dimensional growth mode followed by three-dimensional island growth as is corroborated by reflection high-energy electron-diffraction analysis. Photoemission measurements were acquired after each growth in order to observe changes in the valence-band electronic structure as well as changes in the In 4d, Se 3d, Cd 4d, and S 2p core lines. The results were used to correlate the interface chemistry with the electronic structure at these interfaces and to directly determine the CdS/${\mathrm{CuInSe}}_{2}$ heterojunction valence-band discontinuity and the consequent heterojunction band diagram. These results show that the Katnani-Margaritondo method is unreliable in determining offsets for heterojunctions where significant Fermi-level pinning may occur and where the local structure and chemistry of the interface depends strongly on the specific heterojunction.

Journal ArticleDOI
TL;DR: Pour un superreseau (m,n)=(15,15) on observe une structure de bande de type I qui devient de type II a une pression d'environ 0,22 GPa.
Abstract: We have measured low-temperature photoluminescence and photoluminescence excitation spectra of (GaAs${)}_{\mathrm{m}}$/(AlAs${)}_{\mathrm{n}}$ symmetric superlattices (m=n) near the type-I-to-type-II crossover under high hydrostatic pressures. For an (m,n)=(15,15) superlattice we observe a type-I band structure which becomes type II at a pressure of \ensuremath{\sim}0.22 GPa. Based on pressure coefficients, the zero-pressure splitting is extrapolated as 27(4) meV. When (m,n)=(12,12), the superlattice is type II at zero pressure, showing two luminescence bands separated by 46(4) meV which linearly diverge in energy with increasing pressure. This implies the crossover occurs when 12lm,nl15. From intensity measurements on the (m,n)=(15,15) sample, the type-I recombination rate is estimated to be 1000 times slower than scattering into the lowest-energy X-like conduction-band state. We also find the valence-band offset to be dependent on pressure. These measurements then permit us to determine the pressure coefficients of all the lowest-energy confined states relative to the AlAs-like valence band. Near crossover we see no evidence of interaction between the type-I and type-II conduction-band states.

Journal ArticleDOI
TL;DR: In this paper, the authors measured the band discontinuity of the β-SiC/Si heterojunction by photoemission to determine its application to Si-heterojunction bipolar transistors.

Patent
06 Sep 1990
TL;DR: In this paper, an indium arsenide well monolayer is formed in an InGaAs channel region and functions to move a first quantized energy level E 0 closer to the bottom of the channel region quantum well thereby increasing electron concentration by increasing effective band offset potential.
Abstract: A heterojunction field effect transistor (HFET) having a source, drain, and channel, wherein the channel comprises a quantum well and at least one mono-atomic well or barrier layer is provided. The mono-atomic well or barrier layer has a different bandgap than the channel region and serves to modify electron wave function and conduction band energy in the channel region. Preferably, an indium arsenide well monolayer is formed in an InGaAs channel region and functions to move a first quantized energy level E 0 closer to the bottom of the channel region quantum well thereby increasing electron concentration by increasing effective band offset potential. Another embodiment uses an aluminum arsenide monolayer as a barrier monolayer in the InGaAs channel. By varying location of the monolayers, confinement of electrons in the channel can be improved.

Journal ArticleDOI
TL;DR: La theorie reproduit avec precision les spectres d'une large gamme d'echantillons pour des discontinuites des bandes de valence allant de 0,3 a 0,6.
Abstract: The complete k\ensuremath{\cdot}p Hamiltonian with strain is solved numerically to obtain the energies and wave functions of ${\mathrm{In}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{\ensuremath{-}}\mathit{x}}$As-GaAs superlattices. The electron, heavy-hole, light-hole, and split-off bands are treated in a unified description in which the only adjustable parameters are the respective zone-center effective masses in each material. It is shown that the theory accurately reproduces the spectra of a wide range of published samples for valence-band offsets ranging from 0.3 to 0.6. It is also found that the transition energies are relatively insensitive to the valence-band offset over a wide range of offsets. The electron\char21{}light-hole exciton energy is fitted more closely at the lower offset values, and suggests a valence-band offset close to 0.4. At this offset, the light holes exhibit borderline type-II behavior, and are only slightly localized in the GaAs layers.

Journal ArticleDOI
TL;DR: In this paper, experimental data of energy gaps, band offsets, and energy levels in AlxGa1−xAs/GaAs quantum wells were utilized for the determination of Γ electron and light hole effective masses in GaAs compounds on the basis of the author's relations between these quantities.
Abstract: Experimental data of energy gaps, band offsets, and energy levels in AlxGa1−xAs/GaAs quantum wells are utilized for the determination of Γ electron and light hole effective masses in AlxGa1−xAs compounds on the basis of the author’s relations between these quantities The temperature dependences of electron and light hole effective masses in GaAs are also obtained

Journal ArticleDOI
TL;DR: In this paper, the orientation and the growth sequence dependence of the valence-band offset ΔEv at the lattice matched common anion GaAs•AlAs interfaces were systematically studied.
Abstract: We systematically studied the orientation and the growth sequence dependence of the valence‐band offset ΔEv at the lattice‐matched common anion GaAs‐AlAs interfaces. High quality GaAs‐AlAs heterojunctions were carefully grown on GaAs substrates with three major orientations, namely, (100), (110), and (111)B. The core level energy distance ΔECL between Ga 3d and Al 2p levels was measured by in situ x‐ray photoemission spectroscopy. ΔECL is found to be independent of the substrate orientation and the growth sequence, which clearly indicates the face independence of ΔEv. This result suggests that the band lineup at lattice‐matched isovalent semiconductor heterojunctions is determined by the bulk properties of the constituent materials. ΔEv is determined to be 0.44 ± 0.05 eV.

Journal ArticleDOI
TL;DR: In this article, the optical properties of a novel GaAs/GaP strained-layer superlattice were studied and compared with a theoretical model, and the first observation of direct (type−I) and indirect (type‐II) transitions in GaAs-GaP heterojunction was made.
Abstract: The optical properties of a novel system, the GaAs/GaP strained‐layer superlattice, are studied and compared with a theoretical model. Photoluminescence and photoreflectance measurements revealed that among the set of superlattices under study type‐I and type‐II behaviors (similar to those found in the lattice‐matched GaAs/AlAs system) are present. The evolution of the photoluminescence peaks as a function of temperature and excitation density supported the assignment of the transitions involved. This is to our knowledge the first observation of direct (type‐I) and indirect (type‐II) transitions in strained‐layer superlattices. A comparison with a theoretical model has led to an estimation of the conduction‐band offset as 0.4 eV, which is the first value obtained from experiment in a GaAs/GaP heterojunction.

Journal ArticleDOI
TL;DR: In this paper, the crystal structures and the band structures of ultrathin layered ZnSe/ZnS superlattices are calculated by the Valence Force Field model and the self-consistent pseudopotential method.
Abstract: The crystal structures and the band structures of ultrathin layered ZnSe/ZnS superlattices are calculated by the Valence Force Field model and the self-consistent pseudopotential method. The resulting crystal structures show that the macroscopic elastic theory simulates strained structures of the present superlattices well. It is shown that some energy levels oscillate due to the symmetry with changing the layer thickness while strain causes another level splitting. The possibility that the ZnSe/ZnS superlattice on the GaAs substrate becomes the strain-induced light-hole gap material with decreasing the layer thickness is suggested. The structural stability and the band offset are also discussed.

Journal ArticleDOI
TL;DR: In this article, the authors reported on the calculations of energy band gaps based on the semi-empirical tight-binding model for short-period (ZnTe)m(ZnSe)n and (znS)m (ZnsSe)N strained-layer superlattices (SLSs) and obtained an unstrained valence-band offset of 1.136±0.1 eV for this super-layer.
Abstract: We report on the calculations of energy band gaps based on the semiempirical tight‐binding model for short‐period (ZnTe)m(ZnSe)n and (ZnS)m(ZnSe)n strained‐layer superlattices (SLSs). During the calculation, much attention has been paid to the modeling of strain effect. It is found that (ZnTe)m(ZnSe)n superlattices grown on InAs, InP, and GaAs substrates show very different electronic properties from each other, which is consistent with experimental results now available. Assuming that the emission observed for (ZnTe)m(ZnSe)n SLS originates from intrinsic luminescence, we obtain an unstrained valence‐band offset of 1.136±0.1 eV for this superlattice. On the other hand, the band gap of (ZnS)m(ZnSe)n superlattice grown coherently on GaP is found to exhibit a much stronger structure dependence than that grown coherently on GaAs. The difference of energy gap between superlattice with equal monolayers (m=n) and the corresponding alloy with equal chalcogenide composition is also discussed.

Journal ArticleDOI
TL;DR: In this article, the electronic band structures of the lattice mismatched InAs/GaAs system have been calculated for three different lattice constants using ab initio pseudopotentials and the effect of strain on the band discontinuities has been investigated with respect to the interface interlayer separation as well as the strain.
Abstract: The electronic band structures of the lattice mismatched InAs/GaAs system has been calculated for three different lattice constants using ab initio pseudopotentials. The effect of strain on the band discontinuities has been investigated with respect to the interface interlayer separation as well as the strain. The superlattice electronic structure is examined within the context of interface states and charge localisation.

Journal ArticleDOI
TL;DR: In this article, a first principles total energy self-consistent pseudopotential calculation is used to predict the band offset in the lattice matched superlattice InAs/Al 0.8Ga0.2As0.14Sb0.86.
Abstract: A first principles total energy self‐consistent pseudopotential calculation is used to predict the band offset in the lattice‐matched superlattice InAs/Al0.8Ga0.2As0.14Sb0.86. We find that inclusion of interface strain changes the character of the band offset from nominally type II to strongly type II. The predicted band offset at the minimum energy configuration is in excellent agreement with the value determined from infrared photoluminescence measurements.

Journal ArticleDOI
TL;DR: In this article, Gaussian local-orbital and linearized augmented plane wave basis sets were used to calculate the electronic structure and bonding characteristics of an ideal unrelaxed diamond/BN (110) interface and a series of ideal diamond/Ni (100) and (111) interfaces.

Journal ArticleDOI
TL;DR: In this paper, a review of the present understanding of band edge and free carrier properties in Hg-based superlattices such as HgTe-CdTe is presented.
Abstract: The present understanding of band edge and free carrier properties in Hg-based superlattices such as HgTe-CdTe is reviewed. The authors emphasise the relation between distinctive aspects of the superlattice band structures obtained theoretically and corresponding features in the magneto-transport, magneto-optical and optical data. Theory predicts a high sensitivity of the free carrier properties to the magnitude of the valence band offset. One finds that nearly all of the main experimental results (excepting low-temperature negative differential resistance) are qualitatively consistent with a large offset, but are difficult to explain if the offset is small. The valence band offset controversy thus appears to have been largely resolved.