We investigated the electrical properties of Mg-doped GaN grown by alternative pulse supplies of source and dopant materials in metalorganic vapor phase epitaxy. We obtained the hole concentration of 6×1018cm-3 for p-type GaN grown on a sapphire substrate by repetition of supply and purging of Ga and Mg sources in the constant NH3 flow, while that of p-type GaN grown by the constant feeding of Ga and Mg sources was 2×1018cm-3. By using alternative feedings of Ga source and NH3 with Mg-Si co-doping, we obtained a highly hole concentration of 2×1019cm-3 for p-type GaN which was grown directly on a low temperature AlN buffer layer. We also obtained the hole concentration of 6×1018cm-3 for p-type GaN which was grown on an AlGaN layer on a SiC substrate by alternative co-doping technique. The activation energies for Mg-doped GaN grown by the pulse feedings of source materials were lower than that for GaN grown by continuous supplies of source materials as used in the conventional technique.

Deep intrinsic energy levels near the middle of the band gap in CdTe have been reported in a number of experiments. Based on earlier defect-supercell electronic structure calculations, at least some of these features have been attributed to the second ionization level of the Cd vacancy, while the TeCd antisite, possibly complexed with a Cd vacancy, has also been suggested to account for some midgap levels. Using high-accuracy LDA calculations with full lattice relaxation out to third neighbors, we find that (i) both acceptor states of the Cd vacancy are shallow, (ii) the donor states of an isolated TeCd are both more than 1 eV above the valence band maximum, (iii) the TeCd-VCd complex does indeed have acceptor states near midgap in CdTe and probably accounts for the native defect states in that energy range.

We present an investigation of the diffusion and melt doping of Cu in CdGeAs2. Cu was found to be an acceptor, its introduction in the melt allowed the controlled introduction of holes from ∼8·1016 cm-3 to ∼1·1018 cm -3. Introduction of Cu from the melt was also found to improve the homogeneity of the crystal. The saturation solubility of Cu introduced via diffusion doping was found to have an exponential dependence on the diffusion temperature, from 723 K to 873 K, obeying the equation N = N0 exp (- δH / kT) with the parameters N0 = 6.10+19 cm-3, δH = 0.3 eV, and k equal to Boltzman's constant. From 873 K to 923 K, the saturation solubility was found to decrease exponentially with temperature.

Results of the application of a combination of synchrotron radiation based analytical techniques, X-ray Beam Induced Current (XBIC) and microprobe X-ray Fluorescence (ν-XRF) to the analysis of shunts and lifetime limiting defects in solar cells are reported. XBIC, a new lifetime measurement technique similar to the Laser Beam Induced Current (LBIC) technique, uses a focused X-ray beam to generate minority charge carriers, which are then collected by the p-n junction of the solar cell. The X-ray beam is focused down to a spot size varying from approximately 1×1 νm to 5×5 νm, depending on the settings of focusing mirrors and slits. The sample stage is moved by computer-controlled step motors with sub-micron accuracy. Since the X-ray Beam Induced Current, which characterizes the minority carrier diffusion length in the spot where the X-ray beam hits the sample, and the X-ray Fluorescence signal, which characterizes the chemical nature of the precipitates under the beam, are measured at the same time, the chemical nature of the defects and impurities and their recombination activity can be studied simultaneously, in situ, and with a micron-scale resolution.

We present the results of the applications of these techniques to low lifetime regions in fully processed solar cells. The solar cells were pre-characterized by LBIC and thermography, and regions of interest (containing shunts) were selected. An ν-XRF scan in this area of low lifetime revealed the presence of silver and titanium far from the contact strip, suggesting a process-induced defect.

4H-SiC samples were implanted at room temperature with 30 keV D+ ions at a dose of 5×1016 D+/cm2. Nuclear reaction analysis (NRA) and secondary ion mass spectroscopy (SIMS) measurements were performed to study the deuterium profiles after subsequent annealing at 1000-1250°C for 10min.Also, analytical techniques: RBS/C and thermal desorption spectroscopy (TDS) were carried out to characterize the evolution of implantation induced defects upon annealing. According to the NRA measurements, no deuterium release was found in the sample annealed at 1000°C. However, increasing the temperature to 1150°C led to a 40% decrease of deuterium content. Similar results about the evolution of D profiles upon annealing have also been obtained by SIMS measurements. In addition, SIMS measurements show that the maximum of the deuterium concentration shifts to the surface. Deuterium desorption at annealing temperatures higher than 1000°C was further confirmed by TDS experiments. Results from RBS/C indicated that during the desorption of deuterium, the implantation induced damage was annealed. These results are discussed.

Hydrogen implantation has been used to anneal defects produced in p-type silicon by boron implantation. Boron implantation is performed with an energy of 300 keV to a dose of 1×109 cm-2. Deep level transient spectroscopy measurements show the production of four hole traps (Ev + 0.21, 0.35, 0.50, 0.55 eV) by boron implantation. Subsequent hydrogen implantation is performed with energies of 60, 90, 120 and 150 keV to a dose of 2×1010 cm-2. Among four traps produced by boron implantation, the most significant effect of hydrogen implantation is observed on one trap (Ev + 0.50 eV). A 62% decrease in concentration is caused for this trap by hydrogen implantation with energies of 120 and 150 keV. This partial annealing is ascribed to the reaction of boron-implantation-induced defects with point defects produced by hydrogen implantation.

Structure and properties of amorphous hydrogented silon nitride (a- SiNx:H) are studied using density functional methods. Models of a-SiNx:H were generated using a random network algorithm. In addition we investigarted the N-H terminated surfaces of β-Si3N4.

A comparison between the vibrational spectra of the surface and of the bulk models shows that within the bulk the frequency of N-H bond stretching (≈ 3150 cm-1) is lower than at the surface (≈3300 cm-1). Both spectra show an asymmetric form of the principal peak tailing towards space of hydrogen with nearby atoms. Car-Parrinello moleculat dynamic simulations at elevated temperatures show hopping of bulk hydrogen between different two bonding sites. Moreover, hydrogen acts as network transformer releasing internal stresses.

Silicon samples were implanted with He ions at 1.6 MeV using doses ranging from 1×1016 cm-2 to 1×1017cm-2 with different fluxes (0.4νA/cm2 - 2.0νA/cm2) and annealed at high (1000°C) and low temperatures (800°C). The implantation induced-defect structure and their distribution in the depth of the sample were studied by cross section electron microscopy (XTEM). An unexpected consequence of the flux on the defect population and density was found solely for 2×1016 cm-2, which is the upper threshold to get nano-bubbles at such large implantation depth. Nuclear Reaction Analysis (NRA) were performed to measure the ratio of He remaining in the bubbles as a function of time and temperature anneal. Some samples were gold or nickel diffused at temperatures ranging from 870°C to 1050°C prior to He implantation. The gettering efficiency of the implantation-induced defects was measured by secondary ion mass spectroscopy (SIMS), after a high temperature getter annealing. SIMS profiles exhibit a shape and a width closely related to the presence of the defects (observed by XTEM) which are very efficient sinks for all kinds of metal impurities. The bubbles were found to be more efficient traps than the dislocation loops.

Defect engineering principles are integral to the design and manufacture of high-quality CCD image sensors. As examples, we describe the use of epitaxial silicon for defect control, hydrogen passivation of interface defects, and several forms of impurity gettering. The high sensitivity of image sensors to contaminants reveals that boron segregation gettering of iron dominates gettering by oxygen precipitates for both fast and slow cooling cycles. We estimate that the gettering efficiency for iron is 99.95%.

Epitaxial layers of 4H-SiC are grown on (0001) substrates inclined toward <1120> and <1100> directions. Defects in these films are characterized by deep level transient spectroscopy (DLTS) in order to clarify the dependence of concentrations and activation energies on substrate inclination. DLTS results show no such dependence on substrate inclination but show thickness dependence of the concentration.

We investigated the roughness of the electronic interfaces of GaAs p-n multilayers using cross-sectional scanning tunneling microscopy. We demonstrate that these interfaces exhibit a much larger roughness than the underlying essentially perfect ‘metallurgical’ interface, due to the individual long range electrostatic screening fields around each dopant atom near the interface and due to a clustering of dopant atoms. The clustering and the inherently connected local lack of dopant atoms gives rise to charge carrier depletion zones extending locally through entire nominally homogeneously doped layers once the layer thickness is close to the cluster dimensions. Thus, local variations in the dopant atom distribution limit the precision of the spatial and energetic positioning of the Fermi energy in nanoscale semiconductor structures.

Among many doping issues of diamond semiconductors, establishment of a shallow n-type impurity is a very important and yet challenging subject. Although both nitrogen (N) and hydrogen (H) are omnipresent impurities in diamond, they create only deep half-filled states in the energy gap. In this study, we present a theoretical proposal of a mechanism that makes N donors in diamond as shallow as possible. A complex of two adjacent substitutional N atoms with H sitting between the N atoms, an N-H-N defect, has a donor level approximately 1 eV shallower than of an isolated substitutional N defect.

A number of experimental techniques were used to characterize structural quality, ultrafast carrier dynamics and deep center properties of low-temperature-grown GaAs doped with Be. GaAs layers grown at 280 °C, doped with the Be concentration from 5×1017 cm-3 to 2×1019 cm-3 and annealed at temperatures between 500 and 800 °C were studied. Electron trapping times in these samples varied from hundreds of femtoseconds to several picoseconds. A non-monotonous electron trapping time dependence on Be doping level is explained by the influence of triple-charged gallium vacancies and single-charged Be-acceptors on the number of ionized As antisite defects.

PL studies of oxygen precipitation related defects, stress relaxation related defects and doping striations in various silicon materials are presented. The sample spectrum includes a variety of dopant species, and the dopant concentration range covers several 1014 cm-3 to several 1019 cm-3. Lightly doped, precipitation-annealed polished wafers were intentionally contaminated with Fe, Ni and Cu. Several types of epi wafers based on heavily doped substrates have been investigated after full device processing. PL intensity in the investigated doping concentration range is controlled by three basic recombination mechanisms: radiative recombination competing with multi phonon Shockley-Read-Hall (SRH) and Auger recombination. SRH recombination is the major competing mechanism at low dopant concentration, and Auger recombination becomes important at increasing doping levels. Even though not yet fully understood, the PL technique applied in this study has generated practically useful results.

The work reported here deals with studying the defects induced by light ion (Al, N) implantation in 4H and 6H silicon carbide, both p-type (Al-doped) and n-type (N-doped). Confocal Micro-Raman (MR) was used for monitoring the (480 – 540) cm-1 spectral region of amorphous silicon. A broad peak forms in this region because of silicon atoms relocated as interstitials, translating into a locally stressed crystalline lattice. The locally relaxed lattice at these atoms' locations of origin also gives a broadening of the characteristic Raman peaks of each type of material. UV/Vis Optical Absorption (OA) Spectroscopy has also been employed as a good tool for dopant and carrier trapping levels embedded in the band gap of the silicon carbide material. MR and OA data collected from virgin samples, as implanted, and after annealing at two different temperatures (1100 and 1600 °C) are discussed in this paper.

Low temperature microsensors are designed for cryogenic applications. As a material for the sensors we use heavily doped compensated Ge films deposited on the semi-insulating GaAs substrates. We present the results of experimental and theoretical study of the low temperature resistance as a function of temperature and magnetic field for some models of temperature sensors. The computer simulations show a good agreement with experimental data.

The GaN/AlGaN multi-quantum-wells (MQWs) have been grown via metalorganic chemical vapor deposition (MOCVD). Micro-Photoluminescence (PL) measurement has been performed on non-, In- and Si- doped GaN/AlGaN multi-quantum-well samples in the temperature ranges of 90-300 K. In the non-doped GaN/AlGaN multi-quantum-wells we observed the free exciton peak at 3.4587 eV at 90 K. Other exciton related peaks are located at 3.4346, 3.4177, 3.394, 3.3129eV, which are probably associated with the strongly localized excitons involving the defects. In In-doped GaN/AlGaN MQWs, the free exciton peaks have a slight red-shift from 3.4712 to 3.4629 eV, but the PL intensities become stronger with increasing trimethylindium (TMIn) flow from 10.6 to 42.6-νmol/min. With Si-doping in the well layers, photoluminescence exhibits an envelope of exciton bands ranged from 3.4796eV (free exciton) to 3.43915eV. The excitonic peaks in the bands vary in intensity and position with sample temperature. In addition, we have also observed the LO phonon replica of AlGaN interacted by the laser line due to the resonance effect.

We present experimental results of dopant- and self-diffusion in extrinsic silicon doped with As. Multilayers of isotopically controlled 28Si and natural silicon enable simultaneous analysis of 30Si diffusion into the 28Si enriched layers and dopant diffusion throughout the multilayer structure. In order to suppress transient enhanced self- and dopant diffusion caused by ion implantation, we adopted a special approach to dopant introduction. First, an amorphous 250-nm thick Si layer was deposited on top of the Si isotope heterostructure. Then the dopant ions were implanted to a depth such that all the radiation damage resided inside this amorphous cap layer. These samples were annealed for various times and temperatures to study the impact of As diffusion and doping on Si self-diffusion. The Si self-diffusion coefficient and the dopant diffusivity for various extrinsic n-type conditions were determined over a wide temperature range. We observed increased diffusivities that we attribute to the increase in the concentration of the native defect promoting the diffusion.

The strain relaxation and the misfit dislocation density in micron size windows with different mask materials were studied. Experiments showed that the misfit dislocation density and the strain in the SiGe films would be affected not only by the size of the windows but the mask material. Experiments also showed that after annealing the SiGe films grown in the micron size windows would be much more stable than films grown in the large area on the same wafer.

The formation of the Schottky barrier height at metal-semiconductor interfaces has often been discussed in terms of interface states. This paper examines theoretical models that specifically relate defects and other interface states (such as MIGS) to the interface dipole. These models usually rely on the assumption, which is also invoked in most experimental determination of interface state distribution, that the interface dipole can be broken up into the product of an interface charge and an interface distance/width. Various inconsistencies associated with this assumption are discussed, suggesting that the formation of the interface dipole is an integral process which cannot be divided. It is shown that the inclusion of the bond polarization at semiconductor interfaces, proposed in recent work, gives quantitative account of experimental Fermi level pinning effect. In addition, the bond polarization concept can be extended to the theoretical description of band offsets at epitaxial semiconductor heterojunctions. These results underscore the importance of correctly handling the chemistry at semiconductor interfaces in order to understand their electronic properties.