Density Of Defect Levels Research Articles

Effects of post oxidation of SiO2/Si interfaces in ultrahigh vacuum below 450 °C

Growing SiO2 layer by wet-chemical oxidation of Si surfaces before growth of another insulating film(s) is a used method to passivate Si interfaces in applications (e.g., solar cell, photodiode) at low temperatures (LT) below 450 °C. We report on potential of LT ultrahigh-vacuum (UHV) treatments combined with the wet-chemical oxidation, by investigating effects of LT-UHV oxidation after the wet-chemical growth of SiO2 and before growing Al2O3 film on top of SiO2/Si. This method modifies the SiO2/Si and is found to (i) decrease defect-level density, (ii) increase negative fixed charge density, and (iii) increase carrier lifetime for Al2O3/SiO2/p-Si, as compared to state-of-the-art SiO2/p-Si reference interfaces without LT-UHV. X-ray photoelectron spectroscopy shows that the LT-UHV treatment decreases amount of Si+3 oxidized atoms in chemically grown SiO2 and also amount of carbon contamination. In order to pave the way for further tests of LT-UHV in silicon technology, we present a design of simple UHV instrument. The above-described benefits are reproduced for 4-inch silicon wafers by means of the instrument, which is further utilized to make LT-UHV treatments for complementary SiO2/Si interfaces of the native oxide at silicon diode sidewalls to decrease the reverse bias leakage current of the diodes.

(Invited) Challenges in III-V Surface Passivation and Some Potential Solutions

In many applications, III-V surfaces lie at the buried or embedded interface of an insulator/III-V or metal/III-V junction. The chemical and physical properties of these III-V interfaces are typically different from the uniform III-V bulk crystal structures. Thus, it is difficult to predict the interface behavior (e.g. electronic structure and doping efficiency) on the basis of the established bulk-crystal physics and chemistry. III-V device interfaces often have an amorphous structure and extend over several atomic layers, instead of having a monocrystalline and atomically sharp structure. Reasons for that are discussed in this presentation.Therefore, it is not surprising that the III-V interfaces are a weak part of many practical devices (e.g. laser diode, infrared photodiode, micro-LED, HEMT) causing degradation in the current as well as future III-V applications. III-V interfaces include a high density of point defects, as compared to the uniform bulk crystals areas, which cause the electronic defect levels around the band gap at the interfaces, which further cause the electrical and optical losses. Use of III-V crystals is expected to increase in the areas like LED lighting, infrared optoelectronics, and high-speed transistor circuits, where it is not possible to utilize the Si crystals.Thus it is relevant to investigate and develop the III-V interface properties. The surface or interface passivation means a decrease in the defect-level density or/and in the effects of defect levels. It can be expected that the III-V passivation development requires a multistep procedure combining benefits of the different methods.Some commonly used passivation methods (e.g. wet chemistry, hydrogen forming gas, ALD self cleaning) are summarized in this talk. Our group has studied the issue whether it is useful to include ultrahigh-vacuum (UHV) treatments in a large arsenal of different passivation methods. The traditional surface science has been based on UHV technology. Our group has approached the III-V passivation challenge from a surface-science viewpoint. We have investigated in particular the clean surface properties and the surface oxidation in UHV conditions. Our group has addressed the question whether it is beneficial to pre-oxidize III-V surfaces intentionally because it is very difficult to avoid oxidation of III-V device surfaces. Our results support the previous findings that proper oxidation of III-V interfaces is a potential method to improve the devices performance.We have taken a step toward combining the surface science with the electrical characterization of III-V interfaces to find interconnections between electrical properties and surface-science ones. The high-resolution synchrotron x-ray photoelectron spectroscopy (XPS) has been used. Because XPS is a widely used method, we have addressed specific challenges in XPS analysis and that how the XPS analysis can be developed by including theoretical ab initio calculations. In this talk, it is also attempted to make a comparison to the silicon interface passivation.

Current-voltage characteristics of iron-implanted silicon based Schottky diodes

Current-voltage (I–V) measurements were carried out on undoped and iron (Fe) doped n-silicon (n-Si) to establish and study a change in electrical properties of the material-based diodes with Fe doping concentration. Fe doping was achieved by implantation at the energy of 160 keV to fluences of 1015, 1016 and 1017 ion/cm2. The obtained results indicated that the diodes were well fabricated and Fe doping resulted in a diode behaviour changing from normal exponential to ohmic I–V behaviour. This ohmic behaviour was explained in terms of Fe-induced defect levels that were positioned at the centre of the energy gap. An I–V ohmic region increased with fluence indicating that the density of defect levels has increased with Fe implantation fluence. A change in diode conduction mechanism domination and parameters with fluence were investigated. The obtained I–V properties of Fe doped Si-based diodes were similar to those of the diodes that were fabricated on radiation-hard materials indicating that Fe, too, is a promising dopant in a quest to improve radiation-hardness of Si to be used in high energy physics experiments.

Correlating facet orientation, defect-level density and dipole layer formation at the surface of polycrystalline CuInSe2 thin films

Individual grains of chalcopyrite solar cell absorbers can facet in different crystallographic directions at their surfaces. To gain a deeper understanding of the junction formation in these devices, we correlate variations in the surface facet orientation with the defect electronic properties. We use a combined analytical approach based on scanning tunneling spectroscopy (STS), scanning electron microscopy, and electron back scatter diffraction (EBSD), where we perform these experiments on identical surface areas as small as 2 × 2 µm2 with a lateral resolution well below 50 nm. The topography of the absorber surfaces indicates two main morphological features: micro-faceted, long basalt-like columns and their short nano-faceted terminations. Our STS results reveal that the long columns exhibit spectral signatures typical for the presence of pronounced oxidation-induced surface dipoles in conjunction with an increased density of electronic defect levels. In contrast, the nano-faceted terminations of the basalt-like columns are largely passivated in terms of electronic defect levels within the band gap region. Corresponding crystallographic data based on EBSD experiments show that the surface of the basalt-like columns can be assigned to intrinsically polar facet orientations, while the passivated terminations are assigned to non-polar planes. Ab-initio calculations suggest that the polar surfaces are more prone to oxidation and resulting O-induced defects, in comparison to non-polar planes. Our results emphasize the correlation between morphology, surface facet orientations and surface electronic properties. Furthermore, this work aids in gaining a fundamental understanding of oxidation induced lateral inhomogeneities in view of the p-n junction formation in chalcopyrite thin-film solar cells.

Defect tolerant zero-bias topological photocurrent in a ferroelectric semiconductor

Lattice defect is a major cause of energy dissipation in conventional electric current due to the drift and diffusion motions of electrons. Different nature of current emerges when noncentrosymmetric materials are excited by light. This current, called the shift current, originates from the change in the Berry connection of electrons' wave functions during the interband optical transition. Here, we demonstrate the defect tolerance of shift current using single crystals of ferroelectric semiconductor antimony sulfoiodide (SbSI). Although the dark conductance spreads over several orders of magnitude in each crystal due to the difference in the density of defect levels, the observed shift current converges to an identical value. We also reveal that the shift current is scarcely disturbed by the surface defects while they drastically suppress the conventional photocurrent. The defect tolerance is a manifestation of the topological nature of shift current, which will be a crucial advantage in optoelectronic applications.

Oxidation/reduction cycles and their reversible effect on the dipole formation at CuInSe2 surfaces

The defect-electronic properties of {112} microfaceted surfaces of epitaxially grown ${\mathrm{CuInSe}}_{2}$ thin films are investigated by scanning tunneling spectroscopy and photoelectron spectroscopy techniques after various surface treatments. The intrinsic ${\mathrm{CuInSe}}_{2}$ surface is found to be largely passivated in terms of electronic defect levels in the band-gap region. However, surface oxidation leads to an overall high density of defect levels in conjunction with a considerable net surface dipole, which persists even after oxide removal. Yet, a subsequent annealing under vacuum restores the initial condition. Such oxidation/reduction cycles are reversible for many times providing robust control of the surface and interface properties in these materials. Based on ab initio simulations, a mechanism where oxygen dissociatively adsorbs and subsequently diffuses to a subsurface site is proposed as the initial step of the observed dipole formation. Our results emphasize the relevance of oxidation-induced dipole effects at the thin film surface and provide a comprehensive understanding toward passivation strategies of these surfaces.

Correlating the Local Defect-Level Density with the Macroscopic Composition and Energetics of Chalcopyrite Thin-Film Surfaces.

The unusual defect chemistry of polycrystalline Cu(In,Ga)Se2 (CIGSe) thin films is a main issue for a profound understanding of recombination losses in chalcopyrite thin-film solar cells. Especially, impurity-driven passivation of electronic levels due to point defects segregating at the surface and at grain boundaries is extensively debated. By combining current imaging tunneling spectroscopy with photoelectron spectroscopy, the local defect-level density and unusual optoelectronic grain-boundary properties of this material are correlated with the macroscopic energy levels and surface composition. Vacuum annealing of different CIGSe materials provides evidence that Na diffusion from the glass substrate does not affect the surface defect passivation or grain-boundary properties of standard Cu-poor materials. Furthermore, we find no major impact on the observed thermally activated dipole compensation or the accompanying change in surface band bending (up to 0.6 eV) due to Na. In contrast, Cu-rich CIGSe shows an opposing surface defect chemistry with only minor heat-induced band bending. Our results lead to a comprehensive picture, where the highly desirable type inversion at the p/n interface in standard chalcopyrite thin-film solar cells is dominated by band bending within the CIGSe absorber rather than the result of Na impurities or an n-type defect phase segregating at the interface. This is in accordance with recent studies suggesting a surface reconstruction as the origin for Cu depletion and band-gap widening at the surface of chalcopyrite thin films.

Multiparameter Electronic Tools Quality Level Linear Transformation Models

Continuous control main probability characteristic modeling methods for multiparameter electronics tools has been made, when separate independent device parameters defect level probabilistic distributions are “a priori” known. Defected devices flow in control operation is targeted to localized repair operation in this stage, then devices with second type errors goes again into control and “rotates” until all devices are accepted as good. Second type classification errors probabilities (defect device is accepted as good) in control and repair operations are described by one generalized error model, which is used in linear defect level transformation by different parameters. For all device defect level transformation, defect level transformation by different parameters is used. For all defect level transformation, defect levels densities by different parameters combination, is used referencing by transformation model. It is offered to use approximated models instead of exact whole device defect level probabilistic density by different parameters, described by beta law density, because it is the whole model and the exact defect level density is expressed by many different models in every partial interval of integration. This method simplifies modeling procedure, without decreasing engineering analysis accuracy.

Direct Evidence for a Reduced Density of Deep Level Defects at Grain Boundaries ofCu(In,Ga)Se2Thin Films

The unusual optoelectronic properties of chalcopyrite grain boundaries (GBs) have become the subject of an intense debate in recent years. In this work we investigate the defect density at GBs of Cu(In,Ga)Se2 by scanning tunneling spectroscopy. Contrary to our expectation, our results give evidence for a reduced density of deep level defects and point to an increased density of defect levels in resonance with the lower conduction band at GBs. Our findings imply low recombination activity at GBs, and thus can explain the low impact of GBs on the efficiency of chalcopyrite based solar cells.

Two-Parameter Electronic Devices Quality Models

Continuous control main probability characteristic modeling methods for multiparameter electronics devices has been made, when separate independent device parameters defect level probabilistic distributions are “a priori” known. Defected devices flow in control operation is targeted to localized repair operation in this stage, then devices with second type errors goes again into control and “rotates” until all devices are accepted as good. Second type classification errors probabilities (defect device is accepted as good) in control and repair operations are described by one generalized error model, which is used in linear defect level transformation by different parameters. For all device defect level transformation, defect level transformation by different parameters is used. For all defect level transformation, defect levels densities by different parameters combination, is used referencing by transformation model. It is offered to use approximated models instead of exact whole device defect level probabilistic density by different parameters, described by beta law density, because it is the whole model and the exact defect level density is expressed by many different models in every partial interval of integration. This method simplifies modeling procedure, without decreasing engineering analysis accuracy.

Oxygen Plasma Damage in GaAs Directly Exposed to Surface-Wave Plasma

We have developed a down-flow ashing system for microprocessing of large substrates. The system is equipped with a surface-wave-plasma source of which the “slot-type” antenna was specially designed to generate a uniform and high-density plasma. In this work, the oxygen-plasma-induced damage to the GaAs surface was investigated as a function of the plasma-substrate distance D and exposure time T. These samples were evaluated by X-ray photoelectron spectroscopy (XPS), ellipsometry, photoluminescence (PL), deep level transient spectroscopy (DLTS) and current–voltage (I–V) measurements. XPS results showed that the part of the oxide layer near the surface consists mainly of Ga2O3 and As2O3 while deeper in the oxide layer, Ga2O3 predominantly exists. Ellipsometry results showed that the thickness of the oxide layers increases up to 100 Å with decreasing D and increasing T. PL intensity of free exciton (FE) and (e,A0) observed in GaAs decreased with decreasing D and increasing T. The origin of this decrease in PL intensity was confirmed by DLTS measurements to be due to an increase in EL2 (0.73 eV) defect level density. I–V measurement showed that the effective barrier height decreases with decreasing D. This is probably due to the creation of surface states and to an increase in EL2 density, which lead to an increase in the recombination current at the interface or depletion layers. These results demonstrate that a low amount of damage can be ensured at D≧157 mm. The developed ashing system and condition is currently being used for the mass production of actual semiconductor devices.

Two kinds of enhancement in photoluminescence by chemical etching of porous silicon

We investigated the effect of chemical etching of porous silicon in a solution of HF acid and ethanol. The first etching for 20 s enhanced photoluminescence and shifted to the higher energy. The successive etching enhanced photoluminescence further but peak position was not changed. Photoacoustic spectra evidenced that the first etching induced a blue shift while the successive etching diminished the density of defect levels distributed under 1.3 eV.

Deep-level characterization in semi-insulating GaAs by photo-induced current and Hall effect transient spectroscopy

Deep levels in semi-insulating (SI) GaAs single crystals were studied by photo-induced current transient spectroscopy (PICTS) and photo-induced Hall effect transient spectroscopy (PHETS). A method of measuring the defect level density using PHETS was developed and used to measure the EL6 density in SI-GaAs. Investigations were made into EL2 and EL6, the main defect levels in SI-GaAs, to find the density distribution in the as-grown wafer and the annealing behaviour in the temperature range from 600 to 1100°C. The density distribution of EL2 across the wafer showed a so-called W-shaped pattern similar to the dislocation density, but the EL6 distribution showed an M-shaped pattern. Annealing at temperatures from 800 to 1000°C increased the density of each level, whereas annealing at 1100°C diminished it. These experimental results were well explained by tentative defect models for EL2 (AsGa VGa) and EL6 (AsGa Asi).

Growth Kinetics and Annealing Studies of the “Cathodic” Plasma Oxidation of Silicon

A study of the growth kinetics for plasma‐enhanced oxidation in an electrodeless RF system based on an earlier design is presented. Oxidation growth rates and thickness uniformity variations across 75 mm diam wafers for various plasma parameters are given. The time‐dependence of the grown oxides were found to fit a power law model. Other parameters investigated include RF frequency in the range 6–21 MHz, RF power from 100–500 W, chamber pressure from 85 to 500 mtorr, and oxygen in argon concentration in the plasma from 1 to 100% oxygen. The etch rate of the plasma oxide in 10:1 buffered oxide etch was found to be comparable to that of thermally grown oxides. Defect levels (net coulombic charge) and breakdown values for postoxidation‐annealed oxides were measured initially with poly‐gated capacitors, prior and subsequent to post‐poly anneal. Defect levels and breakdown values for postoxidation annealed oxides were also obtained with aluminum‐gated capacitors, with thermal oxides used as “controls” and subjected to the same annealing treatments. It was found that those oxides which had received a postoxidation anneal in oxygen at 1000°C for 15 min were comparable to the thermal “controls” in defect level density and electric breakdown integrity.