Previous ellipsometric studies of the stability of amorphous silicon (a-Si:H) found reversible changes in the pseudo-dielectric functions. These changes were slow to generate and slow to anneal away. These slow changes are associated with a dangling bond related structural change. Since any light-induced change in the dielectric function is useful for photonic engineering, we undertook the present more detailed study of light induced optical effects in a-Si:H. The optical pseudo-dielectric functions of hydrogenated amorphous silicon (a-Si:H) were measured using spectroscopic ellipsometry (SE) and the “through-the-substrate” measurement technique as a function of measurement temperature and bias light illumination. For the first time we report a light-induced change in a-Si:H materials that is fast, bias-light-dependent, reversible, and temperature dependent. This effect, while not completely understood, offers exciting new prospects for photonic engineering.

We examine different approaches to the analysis of noise in amorphous hydrogenated silicon associated with trapping and generation – recombination processes, which appear to predict very different noise spectra. In one approach the broad noise spectrum observed is assumed to be composed of a distribution of Lorentzian noise spectra, each associated with traps at a given energy depth, with appropriate weighting according to the energy distribution of characteristic time constants. This latter weighting is taken to mirror the energy distribution of states in the gap. This represents a linear superposition of the (weighted) contribution from individual trapping levels, each with its own characteristic time constant. This approach thus assumes that each trap level is an independent source of fluctuation in free carrier number, unaffected by the presence of other traps in the material. At first sight this assertion seems plausible, since in the multi-trapping situation envisaged, cross-correlation effects must be very small. However, the presence of several groups of traps, or, in the limit, a continuum, results in a distribution of characteristic time constants, which is not a simple linear superposition of the time constants for each level. Thus the assertion that a flat density of states, or a region which is flat, such as the top of a broadened level, results in a region of 1/f slope in the noise spectrum, may not be valid. We present an alternative model in which the distribution of time constants is appropriately incorporated, and compare the predictions of this model with the ‘superposition’ approach, using computed noise spectra.

Amorphous silicon-based phototransistors are studied as an alternative solution to replace pixel-level amplifiers simplifying large-area imaging systems. We report electrical characterization by means of current-voltage and capacitance measurements. The measured capacitance increases with decreasing frequency of the probe signal and values largely exceeding the geometrical one at low frequencies have been achieved both in the dark and under illumination. In particular, values in excess of 200 μF/cm2 are measured under 220 μW/cm2 illumination at 600 nm at 100 mHz. The capacitance dependence on frequency is interpreted in terms of trapping and release kinetics processes in the base and of the gain of the device.

This contribution presents an approch to use wire-like substrates for thin-film devices. Based on plasma deposition processes for metals, hydrogenated amorphous silicon (a-Si:H) and transparent conductive oxides, solar cells are fabricated onto metal wires coated by a dielectric. Photolithography using a N2-laser and etching steps serve to pattern the thin film layers. Scanning electron microscope pictures and current-voltage characteristics of the device are demonstrated. Simple geometric simulations show the spectrally resolved intensity of the light that is scattered into the a-Si:H layers as a function of the diameter ratio of the substrate material and the deposited thin films. Finally, we discuss light concentrating and light trapping designs of sensors on transparent fibre substrates.

We have investigated the changes, produced by light-soaking, in both the IR and Raman responses of the Si-Hn stretching peaks in the 2000-2100 cm−1 range. Our observations of the IR response are in qualitative agreement with those of Kong and co-workers [1]: that is, short-term light soaking produces an increase in the intensity of the signal and a simultaneous shift to lower frequency. In contrast, short-term light soaking decreases the total intensity of the Raman signal in the 2000-2100 cm−1 range, when normalized to the TO phonon peak at about 480 cm−1. In both cases, these modifications are reversed on annealing at 200° C. We suggest that these changes are attributable to alterations in the environments of the Si-Hn bonds, with the resultant transfer of intensity between IR and Raman matrix elements. Details of the evolution of the components of the Raman spectrum in the 2000-2100 cm−1 range are presented, and compared with IR changes in the same range.

We have realized color sensors based on vertically integrated thin film structures to overcome color moiré or color aliasing effects. The complete color information of the color aliasing free sensors can be detected at the same position of a sensor array without optical filters. The color separation is realized by the wavelength dependent absorption of amorphous silicon and its alloys. The sensors consist of three amorphous silicon pin diodes. The spectral sensitivity of the sensors can be controlled by the optical and optoelectronic properties of the materials on one hand and the design of the devices on the other hand. In order to investigate the optical wave propagation within the device and to optimize the color separation we have developed an analytical optical model, which takes the optical properties of the materials and the device design into account. The optical model has been combined with a colorimetrical model, which facilitates the benchmarking of the color sensors and the reduction of the color error of the sensors. The analytical colorimetric/optical model and an improved device design will be presented.

We propose a new network rebonding model to describe light-induced metastability in hydrogenated amorphous silicon. This model involves global changes in both the silicon and hydrogen bonding sites. The first step in the process is the breaking of a weak silicon bond generating a dangling bond and floating bond pair. The mobile floating bond diffuses away leaving behind an isolated dangling bond in agreement with ESR measurements. Tight-binding molecular dynamics simulations show clear evidence for each of these processes. Floating bonds are an intermediate transient species that do not remain in the light-soaked steady state. Floating bonds annihilate by reacting with SiH bonds and locally displacing hydrogen atoms. This model provides a new explanation for several major experimental features of the Staebler-Wronski effect, including the t1/3 kinetics for the growth of defects, and the anticorrelation of dangling bonds with H. This new model provides a new platform for understanding the atomistic origins of light-induced degradation.

The hydrogen dilution in the course of production of amorphous silicon (a-Si) influences its structural properties, which affect significantly light-induced degradation. We used FTIR, X-ray reflectivity and GISAXS analysis to monitor the structural changes occurring during the low temperature annealing of undoped a-Si:H films. FTIR results show that upon annealing at very low temperatures, hydrogen is moved from its positions (voids) where it was accumulated unbonded to silicon and is subsequently trapped at dangling bonds, enhancing disorder. X-ray reflectivity and GISAXS measurements confirmed the enhancement of the void size.

Real time spectroscopic ellipsometry (RTSE) has been applied to characterize surface roughening transitions during the growth of undoped hydrogenated silicon (Si:H) thin films by rf plasma-enhanced chemical vapor deposition (PECVD). In particular, the amorphous–to–amorphous surface roughening transition [→] and the amorphous–to–(mixed-phase-microcrystalline) roughening transition [a→(a+µc)] observed during Si:H growth have been studied under different PECVD conditions of hydrogen dilution ratio R=[H2]/[SiH4], rf plasma power P, and substrate temperature T. For Si:H growth on crystalline Si substrates under the different conditions, phase diagrams have been constructed by plotting the bulk film thicknesses at which these transitions occur as a function of R. The effects of the substrate (c-Si wafers versus amorphous Si:H films) on the a→(a+µc) transition of the phase diagram are also explored. The results provide deeper insights into recent attempts to improve the material properties and solar cell performance for a-Si:H i-layers prepared by rf PECVD at higher rates.

We report direct measurement of depth profile of light induced defects (LIDs) in hydrogenated amorphous silicon (a-Si:H) films. These depth profiles were measured by ESR measurements combined with layer-by-layer precise wet etching technique. We discuss those LIDs depth profiles in relation to the spatial depth distribution of photocarriers generation and recombination profiles, as well as the defect creation efficiency. In case of uniformly absorbed light, the light induced defects creation takes place uniformly in depth of the a-Si:H film. In case of nonuniformly absorbed light, the LIDs depth profile is spatially wider than photocarrier generation rate profile G(x), and depends on G(x) sublinearly. Moreover, the LID depth profile agrees well with the photocarrier direct recombination rate profile rather than the profile of total recombination rate or that through defects. The obtained results suggest that the LIDs creation takes place at local site where the photocarriers recombine directly.

Numerical simulations of the current-voltage characteristics of PECVD-microcrystalline silicon based p-i-n diodes were performed to study the affect of defect density and mobility on solar cell performance. Depending on the combination of both parameters the ideality factor increases or decreases with applied forward bias. The reason is the variable contribution of volume recombination to the total diode current and space charge stored in defect states. The decrease in dark current with reduced hydrogen dilution can partly be attributed to a decrease in recombination centers by the same factor as predicted for midgap defect states by the analytic diode theory. Microcrystalline silicon solar cells deposited in the highly crystalline regime (high H-dilution) are limited by recombination of photogenerated carriers and high dark current. Both can be attributed to a large number of recombination centers. The fill factor of our state-of-theart solar cell is limited by the dark current for small illumination intensities, by series resistance for high illumination levels and by both at its maximum under AM1.5 illumination. Short-circuit current and open-circuit voltage pairs measured under intensities from 10-6 to 30 suns reveal a diode characteristic corresponding to an ideality factor of one at large forward bias.

The piezoresistive behavior of optimized n-type and p-type microcrystalline silicon films deposited on polyethylene terephthalate plastic substrate by hot-wire and radio-frequency plasma-enhanced chemical vapor deposition, at a substrate temperature of 100 °C, is studied. A 4-point bending jig allowed the application of positive and negative strains in the films. Repeated measurements of the relative changes in the resistance of the samples during the strained condition showed reversible behavior, with p-type microcrystalline films having positive gauge factor in the range from 25 to 30 and n-type [.proportional]c-Si:H films having negative values of gauge factor from -40 to -10. The induced strain in the films was in the range between 0 and ±0.3%. A sensor utilizing the piezoresistive property of doped [.proportional]c-Si:H was used to map a contour with the shape of an Archimedes' spiral.

The vibrational density of states for the SiH(D) and SiH2(D2) groups in amorphous silicon nitride are calculated. The Si-H(D) bond-stretching modes of the SiH2(D2) groups are found to split, with the Si-D bond-stretching mode having larger splitting. The difference between the Si-D bond-stretching frequency of the SiD group and the Si-D asymmetrical stretching frequency of the SiD2 group has about the same value as that of the Si-H bonds. The differences in the effective masses of the Si-H and Si-D bonds and the degrees of the bond-stretching splitting of the SiH2 and SiD2 groups have compensating effects.

We numerically simulate performance data of hydrogenated amorphous silicon (a-Si:H) and copper indium gallium diselenide (CIGS) based solar cells for various illumination conditions. For ease of comparison, we model typical single junctions with the very same software. The study allows us to evaluate the cell feasibility in different hybrid electronic systems like smart cards, wrist watches, transponder systems, and mobile sensors. At an illumination intensity of 1 sun, the optical bandgap of the absorber material and the series resistances determine the spectral sensitivity of the solar cell to particular illumination spectra. For intensities of 10-2 suns and so-called D65 spectrum, which represents daylight under cloudy skies, the efficiency of a-Si:H solar cells nearly equates the CIGS cell performance although the AM 1.5 efficiency of the CIGS diode exceeds the one of our a-Si:H cell by more than a factorof two. Infrared-weighted black body radiation leads to superior performance of the CIGS type, whereas for ultraviolet-weighted illumination the a-Si:H cell shows better performance. For intensities below 10-4 suns theexternal shunt resistance dominates the current-voltage characteristics of both cell types, resulting in poor performance independent of the incident spectrum. We complete our study by simulating the solar-powered charging process of a gold capacitor, which serves us as a model for the energy storage within a hybrid electronic system. The charging behavior under various realistic illumination conditions shows particular cellcharacteristics: high open circuit voltages qualify a-Si:H solar cells for electronic systems that require increased voltages and CIGS cells are suited for applications with higher current need.

In this work a new structure is proposed for position sensitive detectors consisting of glass/Cr/a-Si:H(n+)/a-Si:H(i)/ZnO, where the ZnO forms an heterojunction with the a-Si:H(i). The results show that this structure works with success in the fabrication of linear position sensitive detectors. The devices present a good nonlinearity of ͌ 2% and a good sensitivity tothe light intensity. The main advantages of this structure over the classical p-i-n are an easier to built topology and a higher yield due to a better immunity to the a-Si:H pinholes, since the ZnO does not diffuse so easily into a-Si:H as the metal does, which are the cause of frequent failure in the p-i-n devices due to short-circuits caused by the deposition of the metal over the a-Si:H. In this structure the illumination is made directly on the ZnO, so a transparent substrate is not needed and a larger range of substrates can be used.

The feasibility of the new ‘Expanding Thermal Plasma’ technique for the deposition of a-SiNx:H at high deposition rates (typically ∼20Å/s) has been investigated. The structural film properties of the a-SiNx:H are reported for various process conditions and the application of the material as antireflection coating on (multi)crystalline silicon solar cells is studied. Furthermore, the performance of the material for surface and bulk passivation is investigated.

The structural properties of nip-µc-Si:H solar cells are investigated by transmission electron microscopy, X-ray diffraction and Raman spectroscopy. Different structural compositions are obtained by variation of the gas mixture during preparation by plasma enhanced chemical vapour deposition. Nucleation and growth of the n-layer onto textured TCO substrate was found to be similar to the growth on glass substrates. The growth of the i-layer follows a local epitaxy. This implies that the structure of the n-layer is of special importance regarding the control of the microstructure in microcrystalline Si nip solar cells.

This paper reviews recent progress in large-area a-Si/a-SiGe tandem solar cells in Sanyo. Much effort has been devoted to increasing both the stabilized efficiency and the process throughput. A key issue in increasing the stabilized efficiency is thinner i-layer structure with an improved optical confinement effect. High-rate deposition of the i-layers has been investigated using rf (13.56MHz) plasma-CVD method while keeping the substrate temperature below 200 °C. A high photosensitivity of 106 of a-Si:H films maintain up to the deposition rate (Rd) of 15 Å/s by optimizing hydrogen dilution and other deposition conditions. It is of great importance to utilize the effect of hydrogen dilution which can reduce the incorporation of excess hydrogen in the films. The world's highest conversion efficiency of 11.2% has been achieved for a large-area (8252cm2) a-Si/a-SiGe tandem by combining the optimized hydrogen dilution and other solar cell related technologies.

A soft ionization laser-based technique using 10.5 eV photon energy has been used to probe radical growth precursors in the hot wire chemical vapor deposition (HW-CVD) of a-Si:H. Using a Re filament, it is shown that Si, SiH3, and Si2H6are the major silicon-containing species formed from the hot wire dissociation of silane, and SiH2is at most a very minor product. However, chamber history is found to influence the radical species produced; i.e. SiH3 and Si2H6are largely related to the chamber wall and filament conditions. The gas species produced by W and Re filaments at wire temperatures between 1000oC and 2000oC have been studied and compared. Heating the filament to higher temperatures increases the flux of Si, SiH3 and Si2H6 in a similar fashion for both filament materials. Above 1800oC, the Si intensity saturates, while SiH3 and Si2H6show monotonic increase without saturation up to 2000oC.

We have observed the existence of medium range order via fluctuation microscopy in amorphous silicon grown at 230°C. We hypothesize that this structure develops during the highly non-equilibrium growth process; high densities of ordered surface nuclei are produced which are subsequently buried and forced into an unfavorable energy state. These nm sized regions are distorted in the bulk due to strain, but remain topologically crystalline. In this work, we alter the growth energetics both at the surface and sub-surface during magnetron sputter film deposition with two kinds of particle bombardment, respectively: a controllable flux of low-energy (20eV) Ar+ ions, and higher energy (100eV) D vs. H neutrals. With this method, we demonstrate for the first time control over the intensity of this medium-range structural order at a constant substrate temperature as seen primarily with fluctuation electron microscopy, but also Raman scattering, spectroscopic ellipsometry, and SAXS. We suggest that these bombardments can increase adspecie surface mobility or drive local sub-surface restructuring (“kinetic annealing”), thus increasing or decreasing the size, density and/or strength of the ordered regions.