Showing papers on "Band offset published in 1987"
PARC1
TL;DR: A first-principles approach to deriving the relative energies of valence and conduction bands at semiconductor interfaces is presented, along with a model which permits a simple interpretation of these band offsets, and results are reported for a large number of lattice-matched interfaces.
Abstract: We present a first-principles approach to deriving the relative energies of valence and conduction bands at semiconductor interfaces, along with a model which permits a simple interpretation of these band offsets. Self-consistent density-functional calculations, using ab initio nonlocal pseudo-potentials, allow us to derive the minimum-energy structure and band offsets for specific interfaces. Here we report results for a large number of lattice-matched interfaces, which are in reasonable agreement with reported experimental values. In addition, our systematic analysis leads to the important conclusions that, for the cases considered, the offsets are independent of interface orientation and obey the transitivity rule, to within the accuracy of our calculations. These are necessary conditions for the offsets to be expressible as differences between quantities which are intrinsic to each of the materials. Based on the information obtained from the full interface calculations, we have developed a new and simple approach to derive such intrinsic band offsets. We define a reference energy for each material as the average (pseudo)potential in a “model solid,” in which the charge density is constructed as a superposition of neutral (pseudo)atomic densities. This reference depends on the density of each type of atom and the detailed form of the atomic charge density, which must be chosen consistently for the different materials. The bulk band structures of the two semiconductors are then aligned according to these average potential positions. For many cases, these model lineups yield results close to those obtained from full self-consistent interface calculations. We discuss the comparison with experiments and with other model theories.
508 citations
TL;DR: In this paper, it is argued that the absolute hydrostatic deformation potentials for tetrahedral semiconductors with the linear muffin-tin-orbital method must be screened by the dielectric response of the material before using them to calculate electron-phonon interaction.
Abstract: It is argued that the absolute hydrostatic deformation potentials recently calculated for tetrahedral semiconductors with the linear muffin-tin-orbital method must be screened by the dielectric response of the material before using them to calculate electron-phonon interaction. This screening can be estimated by using the midpoint of an average dielectric gap evaluated at special (Baldereschi) points of the band structure. This dielectric midgap energy (DME) is related to the charge-neutrality point introduced by Tejedor and Flores, and also by Tersoff, to evaluate band offsets in heterojunctions and Schottky-barrier heights. We tabulate band offsets obtained with this method for several heterojunctions and compare them with other experimental and theoretical results. The DME’s are tabulated and compared with those of Tersoff’s charge-neutrality points.
370 citations
TL;DR: In this paper, the valence and conduction band discontinuities for the lattice matched (Ga,In)P/GaAs heterojunction have been determined by capacitance-voltage (C‐V) profiling.
Abstract: The valence and conduction band discontinuities for the lattice matched (Ga,In)P/GaAs heterojunction have been determined by capacitance‐voltage (C‐V) profiling. Both p‐p and n‐n heterojunctions were profiled, in order to obtain separate and independent values for both the valence‐band‐edge discontinuity (ΔEv) and the conduction‐band discontinuity (ΔEc). The band lineup is found to be of the straddling type with the valence‐ and conduction‐band discontinuities 0.24 and 0.22 eV, respectively, with an estimated accuracy of ±10 meV. Computer reconstruction of the C‐V profiles was used to check the consistency of the data. The band offset data indicate that the (Ga,In)P/(Al,Ga)As system should be staggered for a certain range of Al compositions.
181 citations
TL;DR: In this paper, it was shown that if the GaAs layer thickness is less than ∼30 A, the lowest conduction-band state is confined to the AlAs barrier region instead of the GAs quantum well region resulting in a spatial separation of electrons and holes.
Abstract: Realistic tight‐binding calculations for the GaAs‐AlAs superlattice clearly demonstrate that if the GaAs layer thickness is less than ∼30 A, the lowest conduction‐band state is confined to the AlAs barrier region instead of the GaAs quantum well region resulting in a spatial separation of electrons and holes. This observation exemplifies a failure of the widely used Kronig–Penney model and emphasizes the importance of the k vector selection rule. This property can be used in photoluminescence spectroscopy to determine the band offset of thin layer superlattices for an essentially arbitrary composition of AlxGa1−xAs. A complete two‐dimensional map of the electronic character as a function of the GaAs and the AlAs layer thickness is obtained. The possibility that Xxy states may lie below Xz states for narrow wells is proposed, and implications of the present calculation on interpreting spectroscopic data are discussed.
76 citations
TL;DR: The Hg-Te-CdTe system is used as an example in which it is found that the valence-band maximum of HgTe lies 0.35 eV above that of CdTe, which is in good agreement with the x-ray photoemission spectroscopy measurement of the heterojunction offset.
Abstract: A method to determine a natural valence-band offset (NVBO), i.e., the change in the valence-band maximum energy which is intrinsic to the bulk band structures of semiconductors is proposed. The HgTe-CdTe system is used as an example in which it is found that the valence-band maximum of HgTe lies 0.35 + or - 0.06 eV above that of CdTe. The NVBO of 0.35 eV is in good agreement with the X-ray photoemission spectroscopy measurement of the heterojunction offset. The procedure to determine the NVBO between semiconductors, and its implication on the heterojunction band lineup and the electronic structures of semiconductor alloys, are discussed.
65 citations
TL;DR: In this article, a band-aligned superlattice laser is proposed which uses the intersubband optical transition in a bandaligned super lattice, where the miniband discontinuity within the conduction of valence band functions as a band offset in the heterojunction structure.
Abstract: A novel infrared laser is proposed which uses the intersubband optical transition in a band‐aligned superlattice. In this band‐aligned superlattice laser, the miniband discontinuity within the conduction of valence band functions as a band offset in the heterojunction structure, and the population inversion is achieved by current injection as in the conventional heterojunction laser. It is more flexible than a heterojunction laser of a quantum well laser since one may tailor the bandwidth and band structures as well as the band gap of the minibands. Also indirect band‐gap materials like Si and Ge can be used for lasing in the intersubband transitions. The intersubband optical transition is similar to an atomic two‐level system which exhibits low threshold current, and a gain coefficient with weak temperature dependence and a narrow spectrum which is determined only by the line‐shape function. These special features make the band‐aligned superlattice laser competitive with and perhaps superior to the quantum well dot laser which is not presently feasible.
61 citations
TL;DR: It is found that the valence-band offset ΔE v at the lattice-matched GaAs/AlAs(001) interface is in very good agreement with recent experimental results, which contribute to understanding the underlying mechanism of the band discontinuity.
Abstract: The valence-band offset ΔE v at the lattice-matched GaAs/AlAs(001) interface is derived from highly precise self-consistent all-electron local-density band-structure calculations of the (GaAs) n (AlAs) n (001) superlattices (with n ≤ 3). We calculate ΔE v , by using the core levels — available uniquely from an all-electron approach—as reference energies. Since these are experimentally accessible quantities, a direct comparison with experiment is, in principle, possible. We find that ΔE v = 0.5 ± 0.05 eV, in very good agreement with recent experimental results (ΔE v = 0.45−0.55 eV). Calculated core-level shifts are also compared to experiment. These results, which are closely related to changes in the charge-density distribution at the interface, contribute to understanding the underlying mechanism of the band discontinuity.
59 citations
TL;DR: In this article, an effective dipole theory is presented to estimate the band lineups at the interface of a lattice matched or nearly matched semiconductor heterojunction, which causes additional shift ΔEv in the difference of the band edges.
Abstract: An effective dipole theory is presented to estimate the band lineups at the interface of a lattice‐matched or nearly matched semiconductor heterojunction. The theory is based on the formation of an effective dipole at the interface which causes additional shift ΔEv in the difference of the band edges. A set of equations are derived from which δEv can be solved iteratively. The calculation requires the values of the top of the valence band and several bulk band‐structure parameters of the constituent semiconductors as input. The dipole effect is evaluated by considering the charge transfer induced by the penetration of the effective mass electrons representing the bulk band states into the quantum barrier of the neighboring semiconductor. The theory is applied to predict the band offset values of more than 100 heterojunctions involving group IV, III‐V, and II‐VI semiconductors. Of the 30 heterojunctions for which the experimental data have been reported, the predicted values differ from the data by only ab...
44 citations
TL;DR: In this paper, Li et al. measured low temperature optical transmission spectra (TS) and room temperature photoreflectance (PR) spectra to investigate several In x Ga 1− x AS GaAs strained multiple quantum wells (MQWs) grown by molecular beam epitaxy (MBE).
35 citations
TL;DR: In this paper, the authors report a very careful, self-consistent, relativistic pseudopotential calculation of the interfacial dipole double-layer potential, valence-band offset, and formation enthalpy.
Abstract: We report a very careful, self-consistent, relativistic pseudopotential calculation of the interfacial dipole double-layer potential, valence-band offset, and formation enthalpy of (GaAs${)}_{3}$(AlAs${)}_{3}$(110). A comparison is made with identical calculations for the (001) superlattice with the following results [(001) in parentheses]: The interfacial dipole layer is 315 (154) meV. The formation enthalpy per twelve-atom unit cell is -21.9 (+1.7) meV. The valence-band offset is 447 (446) meV. This lends credence to the idea that the band offset is a difference of bulk quantities and that vastly different interfaces set up whatever double layer is necessary to maintain that difference.
32 citations
TL;DR: It is shown that the Schottky-barrier height and the band offset at semiconductor heterojunctions can coherently be related through the energy of dangling bonds through a charge-dependent tight-binding scheme.
Abstract: It is shown that the Schottky-barrier height and the band offset at semiconductor heterojunctions can coherently be related through the energy of dangling bonds. At Schottky barriers the dangling bonds become resonant states, in which case the Friedel sum rule imposes that the dangling-bond energy aligns with the metal Fermi energy. For heterojunctions the condition of negligible charge transfer corresponds to an alignment of the dangling-bond energies. The proposed theory naturally allows an identification of Tersoff's midgap level with the dangling-bond energy. Detailed calculations are worked out in a charge-dependent tight-binding scheme and show good correlation with experimental data, confirming the validity of the description.
TL;DR: In this article, the conduction-band discontinuity of In0.53Ga0.47As−In0.52(Ga1-xAlx)0.48As heterostructures was determinedto be ΔEc(x)=(0.73±0.03)ΔEg(x) (0.
Abstract: The conduction-band discontinuity, ΔEc(x), of In0.53Ga0.47As–In0.52(Ga1-xAlx)0.48As heterostructures was determinedto be ΔEc(x)=(0.73±0.03)ΔEg(x) (0\leqslantx\leqslant1), by a comparison of the electron (and hole) energy levels in quasi-parabolic (multi-stepped) In0.53Ga0.47As–In0.52(Ga1-xAlx)0.48 As quantum welts observed by photoluminescence excitation spectroscopy, with numerical solutions of Schrodinger's equation, where non-parabolicity of the conduction band was considered.
Patent•
IBM1
TL;DR: In this article, double heterojunction bipolar transistors with an energy band offset in the valence or conduction band and the other of the bands being substantially aligned at the heterjunction are presented.
Abstract: Heterojunction bipolar transistors are disclosed having an energy band offset in the valence or conduction band and the other of the bands being substantially aligned at the heterojunction. For an npn transistor the conduction band is substantially aligned and the bandgap difference is in the valence band. A pnp type transistor is also disclosed wherein all the bandgap difference is in the conduction band and the valence band is substantially aligned. The npn type transistor provides improved hole confinement in the base as well as enhanced electron injection and collection. In one embodiment of a double heterojunction bipolar transistor, materials are selected that utilize Ga compounds in the base and Al and/or In compounds in the emitter and collector, which have a valence band offset of approximately 400 meV or greater and an aligned conduction band at both of the heterojunctions.
TL;DR: In this paper, the authors used admittance spectroscopy to measure the band offsets of semiconductor heterojunctions and found the conduction band offset to be ΔEc=250±10 meV and the valence band offset was ΔEv=346±10 V. The sum of these offsets agrees very well with the known band gap difference between InP and Ga0.47In0.53As.
Abstract: We discuss the use of admittance spectroscopy to measure the band offsets of semiconductor heterojunctions. By using this method to analyze the dynamic response of p–n junctions containing lattice‐matched InP/Ga0.47In0.53As superlattices we can independently determine both the conduction and valence band offsets for this materials system. We find the conduction band offset to be ΔEc=250±10 meV and the valence band offset to be ΔEv=346±10 meV. The sum of these offsets agrees very well with the known band‐gap difference between InP and Ga0.47In0.53As. The ratio of the conduction band offset to the valence band offset is thus 42:58.
TL;DR: In this article, the valence band offsets of the common anion CdTe-HgTe, HgTe-ZnTe, ZnTe-CdTe, and GaAs-AlAs semiconductor pairs are calculated from the core level energies.
Abstract: The valence band offsets of the common‐anion CdTe–HgTe, CdTe–ZnTe, ZnTe–HgTe, and GaAs–AlAs semiconductor pairs are calculated from the core level energies. The good agreement obtained with experiment for lattice‐matched systems and a simple electrostatic model analysis suggest interface dipoles to have only a small effect. Furthermore, the microscopic origin of the failure of the common‐anion rule in lattice‐matched systems is identified: it is found that participation of cation d orbitals (neglected by tight‐binding and pseudopotential approaches alike) in the valence band maxima is responsible for much of the band offset in these systems.
TL;DR: For quantum wells composed of narrow-band-gap materials the subbands exhibit nonparabolic dispersion, and energy-band calculations based on the two-band model indicate that the gap centers of the well and barrier are offset such that the valence-band barrier is increased by about 100 meV.
Abstract: For quantum wells composed of narrow-band-gap materials the subbands exhibit nonparabolic dispersion. We exploit this property in analyzing far-infrared cyclotron resonance in a PbTe quantum well confined by a lattice-matched ${\mathrm{Pb}}_{0.894}$${\mathrm{Eu}}_{0.106}$${\mathrm{Se}}_{0.100}$${\mathrm{Te}}_{0.00}$ barrier. Observation of cyclotron resonance permits simultaneous measurement of carrier density and cyclotron mass, whose energy dependence leads to a determination of the band offset at the interfaces. Energy-band calculations based on the two-band model indicate that the gap centers of the well and barrier are offset such that the valence-band barrier is increased by about 100 meV.
TL;DR: In this article, the natural valence band offset between semiconductors in a common anion alloy system can be determined through photoemission core level measurements, and the result is in approximate agreement with the experimental value of 0.17±0.05 eV.
Abstract: The natural valence‐band offset (NVBO) between semiconductors in a common anion alloy system can be determined through photoemission core level measurements. In this work, we tested this method in the InxGa1−xAs system. The NVBO between GaAs and InAs is measured to be 0.11±0.05 eV. This result is in approximate agreement with the experimental value of 0.17±0.07 eV determined by x‐ray photoemission spectroscopy measurements.
01 Oct 1987
TL;DR: In this article, two frequency selective subreflectors with Jerusalem cross or double square elements are used in cascade to give a dual/triple band feed system at 28, 14 and 6 GHz.
Abstract: Two frequency selective subreflectors with Jerusalem cross or double square elements are used in cascade to give a dual/triple band feed system at 28, 14 and 6 GHz. Their performance in this configuration is described.
TL;DR: In this article, the expected band alignments for pseudomorphic InP/InxGa1−xAs heterostructures for growth on (001) InP are presented; wherein 0≤x≤1.0 relative to the lattice matched value is assumed.
Abstract: Estimates of the anticipated band alignments for pseudomorphic InP/InxGa1−xAs heterostructures for growth on (001) InP are presented; wherein 0≤x≤1.0. Linearity and transitivity of the valence‐band offset, ΔEv, are assumed for a given in‐plane lattice parameter, a∥≡a0(InP). Valence‐band offsets for ternary heterojunctions are obtained via linear interpolation of the self‐consistent interface calculations of Van deWalle and Martin (unpublished), whereas ΔEv for the lattice matched (In, Ga) As/InP heterojunction is taken as an input parameter. It is found that ΔEc shows a rather gradual increase for 0≤x 0.53. The present estimates imply an increase in ΔEc by more than 0.2 eV as x→1.0 relative to the lattice matched value, and that the partitioning of ΔEg between conduction and valence band in strained heterostructures maintains its lattice matched value. Potential applications are reviewed in brief.
TL;DR: In this paper, the current-voltage behavior of a Hg0.78Cd0.22Te-CdTe-Hg 0.78cd0., HgTe, CdTe barrier layer was studied experimentally in the presence of molecular beam epitaxy.
Abstract: Current-voltage behavior is studied experimentally in a Hg0.78Cd0.22Te-CdTe-Hg0.78Cd0.22Te heterostructure grown by molecular beam epitaxy. At temperatures above 160 K, energy-band diagrams suggest that the dominant low-bias current is thermionic hole emission across the CdTe barrier layer. This interpretation yields a direct determination of 390±75 meV for the HgTe-CdTe valence-band discontinuity at 300 K. Similar analyses of current-voltage data taken at 190–300 K suggest that the valence-band offset decreases at low temperatures in this heterojunction.
TL;DR: In this article, a discussion of the theoretical models trying to explain the semiconductor interfaces formation is presented; the results obtained with different methods used to calculate heterojunctions band offsets: the induced density of interface states model (IDISM), the selfconsistent tight-binding model (SCTBM), and the self-consistent local density method (SCLDM) are considered; all of them yield results in good agreement with the experimental evidence.
Abstract: A discussion of the theoretical models trying to explain the semiconductor interfaces formation is presented. Metal-semiconductor and semiconductor-semiconductor junctions are discussed. While the metal-semiconductor interface is worse known due to the uncertainties in the chemisorption sites of the metal on the semiconductor surface, heterojunctions are better understood. We discuss the results obtained with different methods used to calculate heterojunctions band offsets: the induced density of interface states model (IDISM), the selfconsistent tight-binding model (SCTBM) and the selfconsistent local density method (SCLDM) are considered; all of them yield results in good agreement with the experimental evidence. The band offset accuracy for low ionicity semiconductors is higher in the SCLDM (better than 0.05 eV) and lower in the IDISM (better than 0.20 eV), while the SCTBM yields results with an intermediate accuracy, better than 0.1 eV.
TL;DR: In this article, the authors used local pseudopotentials with spin-orbit coupling to calculate electronic and optical properties of (GaAs) N (AlAs) 2 (001) superlattices with the number of atomic monolayers ranging from 1 to 28.
TL;DR: In this article, the authors performed capacitance-voltage measurements on a periodic 20-20 A GaAs-GaAlAs (30%Al), n -type (3 x 10 16 cm -3 doped, 1.7 μm thick structure) and showed that the material is homogeneous in the growth direction.
TL;DR: In this paper, the energy differences of sublevels in rectangular single-quantum-well structures are calculated with adjustable well width and band offset parameter Qc (= Delta Ec/Delta Eg).
Abstract: The authors use the usual effective mass envelope function approach to study the sub-band levels of GaAs-AlxGa1-xAs quantum well structures by considering the complex band structures and the non-parabolicity of bulk component materials. The energy differences of sublevels in rectangular single-quantum-well structures are calculated with adjustable well width and band offset parameter Qc (= Delta Ec/ Delta Eg). Comparison of the results with experimental data suggests that Qc is around 0.65. Also, they present a feasible method for determining Qc independently of other input parameters (such as the effective masses, the well width and the Al concentration) by the field-induced Stark effect on sublevels of electrons and holes or exciton peaks in quasi-parabolic multiple-quantum-well structures.
TL;DR: In this paper, the authors report on current-voltage characteristics obtained from vertical carrier transport measurements at Si/SiGe superlattices, where the period length of the super-lattice was varied from 40 A up to 1000 A. The n-type layers were grown on (100) Si n+ substrates by molecular beam epitaxy.
TL;DR: In this article, the conduction band offset of Ga0.94Al 0.06As:Ga0.57Al0.43As isotype n−n heterojunctions was determined to be 0.54−0.64.
Abstract: We have undertaken electrochemical C‐V measurements of Ga0.94Al0.06As:Ga0.57Al0.43As isotype n‐n heterojunctions in order to determine the conduction band offset ΔEc. Samples were grown by liquid‐phase epitaxy; carrier concentrations were determined to be, respectively, 1016 cm−3 for the small band‐gap layer and 1016 cm−3 or 1017 cm−3 for the large band‐gap layer. From results obtained with various different samples, we determine ΔEc/ΔEg≊0.54–0.64, quite close to the new accepted value of the GaAs:GaAlAs interface. Moreover, the density of fixed interface charges are estimated and found to be related to the monolayer doping density of the large band‐gap layer.
TL;DR: In this article, the electronic band structures of (GaAs)n/(AIAs)1 and (GAAs) 1/(AAs) n(n=1~10) superlattices were investigated by means of an improved tight-binding method in which the overlap integrals up to the second nearest-neighbor atoms, including new parameters, were explicitly taken into account.
Abstract: The electronic band structures of (GaAs)n/(AIAs)1 and (GaAs)1/(AIAs)n(n=1~10) superlattices were investigated by means of an improved tight-binding method in which the overlap integrals up to the second nearest-neighbor atoms, including new parameters, were explicitly taken into account. Kroemer's rule for the band offset value was employed. The resulting band structure of superlattices in the extended zone scheme were compared with the bulk band structure while paying attention to the band-folding effect due to a periodic insertion of one atomic monolayer. A possibility of transforming an indirect-gap material into a direct-gap material by such artificial insertion of another material is discussed and the optical oscillator strength at the Γ-point is estimated for each n.
TL;DR: In this paper, the electronic band structure of superlattices is investigated by means of an improved tight binding method, in which the overlap integrals up to the second nearest neighbor atoms, including new parameters, are explicitly taken into account in order to improve the fitting of the lowest conduction band in the bulk materials to the results of the pseudopotential method.
Abstract: The electronic band structure of (GaAs)n/(AIAs)n(n=1~4) superlattices is investigated by means of an improved tight binding method, in which the overlap integrals up to the second nearest neighbor atoms, including new parameters, are explicitly taken into account in order to improve the fitting of the lowest conduction band in the bulk materials to the results of the pseudopotential method. The two cases of band offset values based on Dingle's rule and Kroemer's rule are employed and the resulting band structure is compared. The oscillator strength between the valence band top and the three of the lower conduction band minima at Γ-point is calculated and compared with a photoluminescence experiment.
01 Jan 1987
TL;DR: In this paper, selectively-doped double-heterojunction FETs have been fabricated using highly conductive epitaxial layers grown by MBE, and the maximum drain current of 600 mA/mm and high transconductance values of 350 - 470 mS/mm were obtained for 0.5 µm gate FET.
Abstract: High-current driving strained N-Al 0.27 Ga 0.73 As /In 0.15 Ga 0.85 As/N-Al 0.15 Ga 0.85 As selectively-doped double-heterojunction FETs have been fabricated using highly conductive epitaxial layers grown by MBE. The maximum drain current of 600 mA/mm and high transconductance values of 350 - 470 mS/mm at drain currents of 330 - 400 mA/mm were obtained for 0.5 µm gate FETs. The strained double-heterojunction structures showed high sheet electron concentrations well exceeding 3×1012/cm2and sheet resistance of 260 - 320 ohm, which is one third that for conventional GaAs/N-AlGaAs single-hetero-junction structures. Because of large band offset at the bottom heterointerface, short channel effects were greatly reduced. Consequently, the decrease of threshold voltage due to the reduction of gate length from 5 µm to 0.5 µm was only 0.2 V and the drain conductance for 0.5 µm-gate FET was less than 10 mS/mm.