Majority carrier transport measurements were made across the potential barriers at (100) twist boundaries in silicon. The bicrystals were prepared by hot-pressing single crystals of lightly doped float-zone material, under ultra-high vacuum conditions. The current - voltage measurements were analyzed using combined drift-diffusion and thermionic emission transport mechanisms, and incorporating some inhomogeneity in the charge distribution at the boundary. Evidence has been found for a small, Nd = 3 × 1010 cm−2, density of mono-energetic defect states near midgap, in bicrystals characterized by a variety of misorientation angles. This density is too small to result from the intrinsic structure of the boundary. In addition, no dependence was found on misorientation angle.

Voids, formed by the condensation of an excess of implantation-induced vacancies, have been recently identified as the defect directly responsible for dopant diffusion and electrical activation anomalies in Si-implanted and annealed GaAs and GaAs/AlGaAs superlattice materials. Depending on the implanted dose, voids can be distributed either throughout the implanted region or in two bands. We have examined the origin of this void distribution difference. In the asimplanted sample associated with the latter case, a buried continuous band of amorphous GaAs has formed. GaAs formed by the recrystallization of amorphous GaAs does not contain excess vacancies, and therefore cannot form voids. However, on either side of the amorphous layer, the excess vacancies can condense to form the observed banded distribution of voids. In the as-implanted sample associated with the former case, a continuous amorphous GaAs layer did not form and therefore, upon annealing, voids are seen throughout the implanted region.

DLTS data taken on MBE grown AlGaAs films show peak heights which increase, maximize, and then decrease with increasing sample temperature. This behavior is difficult to explain within the context of conventional DLTS analysis. We suggest that the data can be accounted for by a trap with a temperature dependent capture cross section in conjunction with a model described by Lee and Borrego in which electron and hole emission rates are comparable. Using this analysis, we obtain an effective trap depth Eeff of 0.35 eV and a capture cross section activation energy Eσ 0.25 eV.

Extensive irradiations were carried on crystalline semiconductors InSb, GaAs and InAs. At low temperatures the frequency dependence of the conductivity is observed to be sublinear (σαωs). where s<1. The dependence of s on both the degree of the induced damage and on temperature is studied and discussed in terms of the quatum mechanical tunnelling and of correlated barrier hopping mechanisms.

At fairly low doses of irradiation activated conductance is observed with s having a weak temperature dependence. As the degree of damage is further increased conduction at helium temperatures takes place by variable range hopping, σαexp(-b/T1/4). The constant b is observed to depend on the effective mass for different semiconductors.

At higher temperatures activated conduction is observed again. The observed activation energies E are Eg 1/4 <E<Eg 1/2 and we believe they could represent hopping of single and double carriers.

Capacitance bridge measurements at low temperature on irradiated specimens of p and n-type Si are reported. Electronic defect levels are seen at Ev+0.10 eV , Ev+0.14 eV and Ec−0.11 eV. The change of the band structure caused by irradiation is also described.

The recombination luminescence involving the A-center, a Cd vacancy paired with a nearest neighbour donor, was investigated in CdTe doped with group VII and III elements. Depending on the type of donor doping, distinct differences in the A-center acceptor binding energies and electron phonon couplings are resolved. Optically detected magnetic resonance shows that the A-center behaves as a shallow effective masstype acceptor consiscent with its small binding energy.

We have analysed the far infrared reflectivity and Raman scattering spectra of phonons in Cd1−xMnxTe, Zn1−xMnxTe, Hg1−xMnxTe, and Hg1−xZnxTe for low composition of x. Unlike CdMnTe and ZnMnTe, weak impurity induced peaks are observed in HgZnTe and HgMnTe, near ˜ 96 and 98 cm−1, well above the LA-phonon band of HgTe. Comprehensive calculation of lattice dynamics based on a semi-empirical Green's function theory has provided convincing arguments to support the above features as gap modes of Zn and Mn in HgTe. The theory also predicts the possibility of resonance modes, above the TA cutoff of HgTe, in HgMnTe and HgZnTe to be observed by IR spectroscopy.

The “bulk” and near-surface regions of N implanted heteroepitaxial ZnSe films were studied using the full width at half maximum (FWHM) of the LO phonon Raman line. The “bulk” FWHM has a minimum below an annealing temperature Ta = 400°C, and increases for higher Ta. This is attributed to the relaxation of residual stress, and to an increased stress from the formation of Zn vacancies. The surface FWHM has a deep minimum near Ta = 500°C which is attributed to the relaxation of the implantation damage at lower Ta, and stress induced by Zn vacancy formation at higher Ta. Another wider peak is found just after implantation and for Ta = 600°C, and results from a sum of two peaks attributed to the heavily damaged region around Rp and to the region with Zn vacancies.

Double crystal x-ray rocking curves from single crystal CdTe and CdZnTe of various degrees of crystalline perfection were recorded under diffraction conditions in which the extinction distance varied from under 10 microns to more than 100 microns. In crystals of intermediate quality the rocking curves recorded under conditions of long extinction length showed structure not evident in those recorded under conditions of short extinction length. Integrated reflectivity measurements were performed at both long and short extinction lengths. The results have been interpreted within the framework of a transition from kinematic to dynamical diffraction.

The Perturbed Angular Correlation (PAC) method has been used to study In -vacancy complexes in AgCl and II - VI compounds. Although these materials are quite different in structure and properties, they are related by photoconductive properties which make them interesting intrinsically and for application to photography (silver halides) and to the design of photodetectors and photovoltaic cells (II - VI compounds). We have obtained the first PAC results for silver halides and ternary II-VI compounds.

For AgCl, measurement at 77 K clearly reveals two different 111In - vacancy complexes characterized by quadrupole interaction frequencies Vq1 = 21.5 MHz and Vq2=43 MHz, both having axially symmetric electric field gradient (EFG) tensors with principle axes along <100> directions. Temperature dependence of the PAC signal allows a determination of the activation energy and attempt frequency for vacancy motion by measurement of the damping arising from dynamic effects.

For 4% Zn - doped bulk CdTe samples, vacuum annealed in the range 400°C to 550°C, PAC shows that a large fraction of the 111 In probe atoms occupy non-cubic lattice sites having an EFG characterized by vQ = 60 MHz, η = 0.2. The concentration of this complex was decreased by annealing between 400°C and 600°C in a Cd overpressure, suggesting tha the observed EFG is caused by a 111In - cadmium vacancy complex. Similar measurements of Cd0.8Mn0.2Te showed an identical EFG which, however, appeared at a lower annealing temperature.

X-ray section topography has been used to study the distribution and size of precipitates resulting from heat treatment of MCZ silicon. A low density of precipitates was found, enabling individual precipitate images to be studied. Images have been simulated by numerical solution of Takagi's equations and the magnitude of the strain field deduced by comparison with experiment. Excellent agreement has been found in the details of simulated and experimental images. The effective defect volume increased monotonically with annealing temperature. The effect of surface relaxation and long range curvature on the accuracy of determining the microscopic strain field by matching simulated and experimental images has been investigated.

End-of-range damage produced by n-type dopants is frozen-in if that damage is produced in germanium-rich regions in crystalline silicon. This is not the case with p-type dopants. High germanium doses were implanted into silicon; the wafers were annealed at 1000 C for 30 minutes under nitrogen. Phosphorus or boron then were implanted into the crystalline siliconfollowed by a 900 C, 30 minute neutral ambient anneal.

Explanation of the freeze-in phenomenon relies on results of semiempirical quantum chemical calculations that showed that group V dopants, whether interstitial or substitutional, as well as interstitial silicon, would be attracted to the vicinity of substitutional germanium sites in germaniumrich regions ofthe crystalline silicon. This general preferential diffusion behavior is not calculated to occur with the p-type dopants from group III. When substitutional, in fact, the group III impurities are predicted to be repelled from the germanium-rich regions.

TEM studies show that any residual damage with boron as the implant and germanium present is low in density and within 20 nm of the silicon interface. This is contrary to studies involving phosphorus where damage is exclusively associated with the end-of-range of the phosphorus implant. Both sets of results are in accord with the theoretical considerations.

The degradation mechanism of a-SiN:H films under current injection is investigated. It is shown that the degradation of a-SiN:H films is closely related to the hole trapping into Si-Si, Si-H, and Si0 defects. It is presumably concluded that the hole trapping centers are the Si-Si defects in the valence band tail. We estimate that the hole trapping cross-section is nearly equal to 10−20cm2.

Samples of SIMOX have been prepared by implantation in a high-current implanter (density ≍ 1 mA/cm2) and by annealing at 1300°C for 6 hours. Transmission electron microscopy reveals unusual structure in these samples. Spectroscopic ellipsometry has been used to analyze these structures. Ellipsometric measurements were collected at an angle of incidence of 75. deg, with photon energies from 1.5 to 5.0 eV, and using a rotating polarizer configuration. The measurements were analyzed with three models: a three-layer model, a four-layer model, and a five-layer model. The five-layer model provided the best fit of the three. This model identified a layer of crystalline Si inclusions (“islands”) within the SiO2 layer. A method is presented that provides initial estimates for the thicknesses of the top three layers to help start the regression analysis.

The electrical properties of S+ implanted in SI GaAs have been studied. The rapid diffusion and redistribution of S+ implanted in GaAs after conventional thermal annealing (CTA) depends not on conventional diffusion of S+ or VAs, but on the enhanced diffusion by ion implantation. By employing rapid thermal annealing (RTA) techniques enhanced diffusion can be restrained, redistribution of S+ implantation can be decreased greatly and a thin active layer suitable for fabricating GaAs MESFET devices can be obtained.

Laser annealed arsenic implanted silicon specimens with doses ranging from 6×1015 to 7×1016 As/cm 2 have been investigated by transmission electron microscopy (TEM) and double crystal xray diffractometry (DCD). For the highest implant dose, laser powers ranging from 1.1 to 2.2 J/cm 2 have been used. Experimental observations show two new features for this kind of specimen. First, in some cases, TEM micrographs evidence small (˜50Å diameter) precipitate-like defects and/or dislocation loops confined within the heavily doped region. Second, in some cases, DCD shows a positive strain in addition to the negative strain attributed to 90% As in substitutional sites. X-ray rocking-curve simulations reveal that the negative strain drops to zero around 1000Å before the end of the As distribution. This might be related to the presence of Si interstitials in the deepest region of the As distribution.

Synchrotron white beam X-ray diffraction topography in transmission geometry has been used to non-destructively investigate defect structures in silicon single crystal wafers, both prior and subsequent to a 60 second rapid thermal processing (RTP) treatment at 1050°C. Prior to RTP dislocations, precipitates and swirl defects were observed and characterized. Following RTP the following effects were observed: glide of individual dislocations and dislocation multiplication; and the enhancement of the strain field associated with the swirl defects. Precipitates appeared unaffected by RTP. This work shows that synchrotron topography is capable of non-destructively revealing significant dislocation motion induced by RTP under conditions were such motion is not thought to occur. This dislocation motion is likely to be detrimental to device performance. The technique enables determination of the conditions required to avoid such dislocation motion.

The effects of Co and Ti silicide film formation on diffusion of buried Sb-doped layers in Si have been investigated. Sb profile analysis by secondary ion mass spectrometry shows that greatly enhanced, non-uniform Sb diffusion occurs during reactions of various thicknesses (30- 300 nm) of Co and Ti by rapid thermal annealing. A simple non-equilibrium intrinsic diffusion model is invoked to estimate time-averaged excess vacancy concentrations. Vacancy concentrations of about 107 times equilibrium values are shown to exist during CoSi2 formation by reaction of a 30 nm Co film at 700°C for 5 min in Ar.Diffusion enhancements at large distances from silicide stripe edges are observed by bevel and etch techniques. These effects tend to decrease with increasing annealing time, indicating that film stresses may play an important role in the interfacial point defect injection process.

Contactless deep level transient spectroscopy using microwave reflection at 35 GHz is presented and it is proved to be a powerful technique for characterizing trapping levels in semiconductors without the necessity of special sample preparation. The technique consists of measuring the transient decay in photoconductivity after a monochromatic light pulse has been applied to the semiconductor. The photoconductivity after the light pulse is caused by emission of carriers from trapping levels filled during the light pulse and by scanning the sample temperature it is possible to determine their activation energy. On Si-implanted layers on LEC grown SI-GaAs substrate we have detected three trapping levels, located at 0.15, 0.18 and 0.27 eV below the conduction band, by using a 1060 nm YAG laser, and one level, 0.13 eV below the conduction band, by using a 633 nm HeNe laser. The technique has been applied to LEC SI-GaAs and no trapping levels have been observed above the EL2 level.

We have investigated the dissociation of donor-hydrogen pairs in Schottky barrier capacitors fabricated from phosphorus and arsenic doped silicon, at temperatures in the range 24 to 110°C. At zero bias the release of donor trapped hydrogen does not follow first order kinetics. Introduction of minority carriers by illumination or forward bias injection enhances the rate of hydrogen release and redistribution, but otherwise leaves the functional form relatively unchanged. Numerical modeling shows that the debonding rate must be proportional to the local hole density. If the dissociation process is carried out in reverse diode bias, some of the released hydrogen rebonds deeper in the silicon, outside the depletion region. In addition, a portion of the charge density in the depleted region is seen to be metastable and disappears after several hours at room temperature. This observation suggests the existence of a positive charge state of the donor-hydrogen pair which may be the precursor to dissociation. While several groups have recently pointed out that reverse-bias annealing data suggest the existence of a negative charge state of interstitial hydrogen, our observation of an electronically controlled debonding rate. casts considerable doubt on the firmness of that conclusion.