Impurities In InP Research Articles

Annealing behavior of high temperature implanted Fe impurities in n-InP

The lattice location of high temperature implanted Fe impurities in InP and the effect of post-implantation thermal treatments were studied by PIXE-RBS-channeling measurements. The results show that a high fraction (about 85%) of the Fe atoms is incorporated in high-symmetry sites after the high temperature implantation and most of them (more than 60%) occupy substitutional positions. Post-implantation annealing treatments induce the escape of the Fe atoms from their sites and the substitutional fraction decreases by increasing the annealing temperature. Assuming an Arrhenius behavior for this process, the activation energy for the displacement of the Fe atoms from substitutional sites is estimated to be 0.76 eV.

Growth and characterization of Si, S and Zn-doped InP in selective area growth conditions

The incorporation and doping process of Si, S and Zn impurities in InP by metalorganic vapor phase epitaxy have been studied both in standard growth conditions (SGC) and in selective area growth (SAG) conditions using silane (SiH 4), hydrogen sulphide (H 2S) and diethylzinc (DEZn). It was observed that in SGC, the dopant species behaviour, for two different growth rates, is in line with the expectations from the model and some previous reports made on doped GaAs and InP. On the other hand, in SAG conditions, the Si concentration was unexpectedly independent of the growth rate. Furthermore, the Zn and S codoping shows a Zn dopant concentration enhancement in SAG conditions, while it remains constant in SGC. These differences in behaviour are induced by the addition of dopant species transported by lateral vapour phase diffusion from the patterned region to the open region.

Incorporation of Highly Concentrated Iron Impurities in InP by High Temperature Ion Implantation

AbstractIron was introduced in InP by ion implantation with the aim of obtaining a high concentration of substitutional, electrically active, deep level impurities. A substrate temperature higher than 200 °C was maintained during implantation in order to reduce damage accumulation and Fe defect reactions. The lattice position of the implanted Fe atoms and its modification during annealing treatments was studied by means of Proton Induced X-ray Emission (PIXE) in channeling conditions and correlated with the ion induced damage measured by different techniques. The results show that a high fraction of substitutional Fe atoms is present after the implantation. This fraction is progressively reduced during thermal treatments by increasing the annealing temperature, with the formation of inactive Fe aggregates, probably in the form of small Fe-P complexes.

Vacancies and impurities in InP studied using positron lifetime and an improved Doppler-broadening spectrometer

- in Zn-doped p-InP. The line-shape parameter measured by an improved Doppler-broadening spectrometer shows a significant difference between n-type and Fe-doped SI-InP materials: the S-parameter varies from 0.5203 in n-InP to 0.5184 in Fe-doped SI-InP by four times the standard deviation. A possible reason for this difference is discussed.

Impurity Effect on Recombination Processes in InP

The effects of impurities (S, Sn, Fe and Zn) doped in InP on recombination processes were investigated using photoluminescence (PL) and photoacoustic (PA) spectroscopies. The recombination processes were affected by the impurity levels in the bandgap and by the large strain which was induced by the impurities in InP. The processes were radiative for InP with shallow impurity levels such as S-doped and Sn-doped samples and nonradiative for InP with deep impurity levels such as Fe-doped and Zn-doped samples. The results of PL and PA measurements were in a reciprocal correlation. However, both spectra were observed for undoped InP with large strain induced by impurities such as Si. These relationships were explained using three types of configuration coordinate diagram.

Impurities in InP grown by MOVPE using a new precursor Et 2InNMe 2: photoluminescence and Hall study

Epitaxial layers of n-type InP have been grown by metal organic vapor phase epitaxy at different growth temperatures and V/III ratios using a new indium precursor and characterized mainly by photoluminescence and Hall measurements. Their electrical and optical properties are encouraging, although they are not as good as those of the layers grown with conventional precursors. Samples grown in different conditions exhibit quite different properties. In particular, incorporation of Mg and Zn at higher and lower growth temperature, respectively, is shown by photoluminescence spectra recorded at liquid-helium temperature.

Carrier capture on defects in multiband semiconductors.

We analyze electron capture on a defect that exhibits no electron-phonon interaction. In this case, capture occurs by a cascade mechanism via the defect excited states. In direct-band-gap materials, when the temperature is too high to allow the filling of the \ensuremath{\Gamma} excited states at the bottom of conduction band, capture takes place via the excited states of a higher band, provided that they are deep enough. Experimental evidence for capture via the excited states of the L and X bands is provided for several defects or impurities in InP, GaAs, and related alloys.

Electronic structure of S and Sn impurities in InP.

Sulfur is an unavoidable contaminant in InP. Recent experiments indicate that the ionization energy of InP:S is very close to that of InP:Sn. We have performed a fully self-consistent discrete-variational linear-combination-of-atomic-orbitals method to study the electronic structure of S and Sn substituted at the In site in InP. Calculated results indicate S to be an acceptor in contrast to donorlike behavior observed in Sn. It is concluded that S occupies the P site and/or forms defect complexes in InP. Comparison with experimental results is made wherever available.

Theoretical study of native defects and impurities in InP.

A study is made of the trends in the deep levels, formation energies, and abundances of native defects and impurities in InP. We predict deep-level positions and equilibrium abundances for the eight native tetrahedral point defects. These consist of two vacancies, two antisites, and the two native atoms at both nonequivalent tetrahedral interstitial sites. The effects on native defects of electron irradiation and the introduction of common dopants Zn and S in InP are also investigated. The theoretical studies are done within the local-density approximation using a recently developed parameter-free self-consistent pseudopotential and pseudoatomic orbital scheme.

Theoretical study of substitutional group IV impurities in InP

The electronic structure of group IV impurities in InP has been analysed using a discrete variational Xalpha self-consistent field method in the embedded scheme within the local-density framework. The pure semiconductor gap is calculated to be 1.44 eV in very good agreement with the experimental value of 1.41 eV. The defected clusters are studied by substituting C, Si, Ge and Sn at the central In site. All the impurities but C show a donor-like behaviour with states introduced deep in the gap region of the host semiconductor. The charge distributions and the density of states are calculated. Comparison with experimental results are made wherever available.

Iron redistribution and compensation mechanisms in semi-insulating Si-implanted InP

The Si and Fe depth distributions have been measured in Si-implanted semi-insulating (SI) InP as a function of implant temperature and post-implant annealing technique (either furnace annealing or rapid thermal annealing). Depth profiles obtained by secondary ion mass spectrometry and damage measurements obtained by Rutherford backscattering demonstrate that Fe redistributes into regions of residual damage during thermal processing. The active electrical carrier profiles measured by electrochemical profiling show differences between atomic and electrical carrier profiles which depend on whether the substrate is semi-insulating or undoped, and on the implant temperature. These differences are interpreted in terms of three different compensating mechanisms: (a) the amphoteric nature of Si impurities in InP, so that Si can be self-compensating, (b) carrier compensation caused by the redistribution of Fe, and (c) implantation-related damage effects and/or stoichiometry imbalance induced by the Si implant.

Electrical properties of ytterbium-doped InP grown by metalorganic chemical vapor deposition

A study has been carried out on the electrical properties of ytterbium-doped InP grown by metalorganic chemical vapor deposition for the purpose of clarifying the electrical behavior of ytterbium impurities in InP. The epilayers were characterized by Hall effect measurements and by deep level transient spectroscopy. The results strongly suggest that ytterbium introduces an acceptor-like level 0.03±0.01 eV below the conduction band which is related to isolated ytterbium atoms or single ytterbium atoms complexed with native defects. No signs of precipitation or clustering of ytterbium are observed for concentrations below 1018 cm−3 . Contrary to previous hypotheses, no evidence has been found for the existence of an ytterbium-related acceptor level in the lower half of the band gap.

Electronic structure due to hydrogen and vanadium as substitutional impurities in InP.

The self-consistent local-density theory is used in a cluster model to calculate the charge distribution, one-electron energy spectra, and the density of states for pure InP and for H and V substituted at the In site. The pure semiconductor gap is found to be 0.8 eV, consistent with experiments. The hydrogen impurity introduces a trapping center at E/sub v/+0.37 eV in the gap region: a semi-insulating-like behavior. The transition metals are generally found to introduce deep levels in the gap region, but for vanadium we find the gap swept clean of any impurity level, with the last partially occupied state close to the bottom of the conduction band. This is consistent with deep-level transient spectroscopy experiments where no vanadium-related deep levels in the band gap are found down to 4 K.

Spectroscopic evidence that oxygen suppresses Si incorporation into vapor phase epitaxial InP

Photothermal ionization spectroscopy is employed to monitor the residual donor impurities in InP grown by vapor phase epitaxy, and study the effect of adding O2 to the carrier gas. The dominant donor species are found to be Si and S, irrespective of the O2 addition, and their concentrations are obtained with the aid of Hall measurements. The result shows that the O2 addition reduces the incorporation of Si into InP but does not affect the incorporation of S.

Electronic structure of alkali-metal impurities in InP

The electronic structure of the substitutional alkali metal impurities in InP has been studied by MSXαSCF method. The shift of the tDBH2-level for different impurities is determined. In none of the cases the impurity level is shifted out of the band gap. Li and Na impurities with their shallow acceptor levels behave as p-type semiconductors whereas K impurity with its deep level behave somewhat like a semi-insulator.

Theoretical study of the electronic states of substitutional transition-metal impurities in InP

The electronic structure of the transition-metal impurities Cr, Mn, Fe, Co, Ni and Cu and of a vacancy in InP have been studied theoretically using the multiple-scattering X alpha -SCF method. It is found that the dangling bond hybrid t2 level moves into the band gap. For Fe this is in the middle of the gap providing a clue to the semi-insulating effect of Fe in InP. The crystal-field resonance level e goes from the gap into the valence band as one goes from Cr to Cu. The corresponding t2 level goes successively deeper into the valence band from Cr to Cu. The population density study shows that the general trend is an increase of transfer of charge as one scans the latter part of the first transition-metal series.

Determination of residual impurities in InP grown by the Czochralski method, using charged particle activation analysis

InP samples were analyzed by charged particle activation analysis. Proton activation at 12 MeV permits the nondestructive survey of 30 elements. The main impurities are Ti, Fe, Ga and Zn (1015 to 2·1016 at/cm3). Oxygen was analyzed separately by triton activation at 3 MeV; the oxygen concentration is normally less than 1016 at/cm3.

Carrier concentration effects on radiation damage in InP

Minority carrier diffusion length and carrier concentration studies have been made on room-temperature 1-MeV electron irradiated liquid-encapsulated Czochralski grown Zn-doped p-InP. The damage rate for the diffusion length and carrier removal rate due to irradiation have been found to strongly decrease with an increase in the carrier concentration in InP. These phenomena suggest that the induced defects interact with impurities in InP. A preliminary study on the annealing behavior has also been performed.

Extremely rapid outdiffusion of n-type impurities in InP

During sealed ampoule annealing at 550 °C, an unexpected, deep diffuson front was observed in n-type InP crystals using defect etching and cathodoluminescence (CL) imaging. The diffusivities of the fronts were determined to be 1–5×10−8 cm2 s−1 which is approximately two orders of magnitude greater than that of the rapidly diffusing Zn impurity. By measuring the changes in the CL intensity and the photoluminescence linewidth to determine the changes in carrier concentration, the diffusion fronts were shown to be caused by the outdiffusion of the dopants. Although outdiffusion of Cr and Fe, affecting an ∼1 μm surface region during thermal annealing at 550 °C for 30 min, has previously been reported for InP, we believe this is the first reported observation of rapid outdiffusion of n-type impurities that can affect InP crystals to a substantial depth (∼200 μm) for similar annealing conditions.

Low pressure-MOCVD growth of Ga0.47In0.53As–InP heterojunction and superlattices

We report the latest results of a continuing study of low pressure-metal–organic chemical vapor deposition (LP-MOCVD) growth of InP and Ga0.47In0.53As–InP heterojunctions and superlattices, which describe the general growth of InP and GaInAs with specific application to quantum well heterostructure. A study of the sources and control of residual impurities in InP and Ga0.47In0.53As, the effects of source purity upon residual impurities, mobility, and 2 K PL intensity are detailed. Total impurity concentrations as low as 6×1014 cm−3 for InP and 3×1014 cm−3 for Ga0.47In0.53As have been obtained. Mobility as high as=μ (300)=5350, μ (77)=60 000 cm2 V−1 s−1 for InP, and μ (300)=12 000, μ (77)=60 000, μ (2)=100 000 cm2 V−1 s−1 for GaInAs layers have been measured.