Several recent measurements have shown that annealing of metastable defects in a- Si:H can be accelerated by the presence of light. This is the opposite of the usual light-induced defect generation, and no existing rate equation explains it while maintaining the necessary symmetry of generation and recovery processes, and consistency with the stretched-exponential transients that best describe observed generation and anneal behavior. This paper shows that this light-enhanced annealing (LEA) can be explained readily by the usual rate equation leading to stretched exponentials with no other terms by allowing a variation of coefficients with temperature or light intensity. This equation then leads to good simulations of observed LEA. Interpretation of these results in terms of distributional changes is presented, and an experimental test is proposed.

We have found that a broad distribution of the annealing activation energies is necessary to interpret the annealing behavior for the density of metastable defects created by 77K light soaking, while a narrow distribution is sufficient for describing the annealing behavior for the density of those created by RT light soaking. We have also found that the recovery of the degraded photoconductivity is easier than the annealing of the photocreated defects for both 77K and RT light soaking and that, at a given defect density, 77K light soaking degrades the photoconductivity more than RT light soaking. Our results suggest that the photocreated defects with a smaller annealing activation energy are more effective recombination centers than those with a larger annealing activation energy.

Many photoelectronic properties of a-Si,N:H alloys prepared by remote PECVD (RPECVD) from two N-atom source gases - N2 and NH3 - are the same; however, the photo-induced changes in the electrical properties in alloys with -2.1 eV bandgaps are ∼3 to 5 times greater in alloys deposited from NH3, which display Si-NH, as well as SiH bonding. Based on this result, we show that bonding groups important in the Staebler-Wronski effect include (i) ≡SiH, and nearest-neighbor (ii) ≡Si-NH-Sis and/or ≡Si-O-Sis in which the respective N and O-atoms make H-bonds with the sSiH group. The model, based on ab-initio calculations, includes a H-exchange reaction in which trapping of photo-generated holes promotes a transfer of the H-atom from the ^SiH group to a nearest-neighbor ≡Si-NH-Si≡ creating (i) a Si-dangling bond (Si*) and (ii) a Metastable (≡Si-NH2-Si≡) + group. Calculations indicate that neutral (≡Si-NH2≡Sis) ° is unstable, so that relaxation of (≡Si-NH2-Si≡) + groups can occur by trapping of a thermally-released (trapped) electron during a post-light-soaking thermal-anneal. The same type of model is developed for hole/electron trapping-induced H-atom transfer between ≡SiH and ≡Si-0-Si≡ groups in other a-Si:H Materials.

We report the results of a study of the metastability under illumination by high intensity red light of device quality a-Si:H thin films deposited using a wide range of deposition conditions. The process variables included substrate temperature, pressure, rf power, and dilution of silane by He or H2. In-situ Monitoring of the sample conductivity and defect density during light-soaking provides the kinetics of the degradation of the electronic properties of the films. We observe equilibration of the photoconductivity and of the defect density. The characteristic time of equilibration τse of the defect density varies by more than an order of magnitude, dividing the samples into two groups: one group with a τse on the order of 103 seconds, the other with a τse on the order of 104 seconds. Low steady state defect densities combined with high ημτ products are observed for “standard” a-Si:H deposited between 100°C and 250°C and He-diluted films deposited above 250°C.

The dependence of the light-induced degradation on the band gap and hydrogen content in Amorphous silicon and related alloys is discussed. A simple model is presented which predicts that the light-induced degradation should be strongly correlated with the optical gap. In particular, for materials with Egap smaller than approximately 1.2eV no degradation is expected. The implications and consequences of the model for the task of reducing metastability in Amorphous silicon alloys are discussed.

The interaction mechanisms of keV-electrons with the hydrogenated Amorphous semiconductor are briefly discussed and the differences to the metastable defect creation by photons are set out. Based on the knowlegde of the energy dissipation mechanisms of keV-electrons in the hydrogenated Amorphous semiconductor, a model for the creation of metastable defects by keV-electron irradiation is developed and its quantitative agreement with the experimental results is shown.

We have annealed PECVD a-Si:H films at 250, 300, and 350°C and measured the evolution of the infrared absorption spectrum. We observe that, during the initial stage of such a heat treatment, atomic hydrogen migrates from the isolated state to the clustered state. Thus diffusion of atomic hydrogen must occur around 300°C. Microvoids with internal surfaces covered with SiH bonds appear to be more stable than voids lined with SiH2 bonds and (SiH2)n polymers.

Junction capacitance methods were employed to investigate the metastable states in a-Si:H with hydrogen modulated Multilayer sample as well as companion homogeneous samples. Each layer in the multilayer sample was grown from the same gas mixture (SiH4+Ar or SiH4+H2) as that of homogeneous samples. We observed that in state A and metastable states produced by a short time light soaking the defect density of each layer in the multilayer sample track well the defect density of the corresponding homogeneous sample. This implies that “local” aspects of metastable defect creation dominate in the earlier light soaked states, and also the stable defects in state A are due to local properties of each layer as introduced during growth. However, for longer time light soaked states and succeeding partial annealed states the defect level of the multilayer sample is higher than either of homogeneous samples. However, the spatial variations of the stable defects are finally recovered when the sample is annealed at the same temperature as the initial state A.

Hevolution spectra were measured for hydrogenated Amorphous silicon (a-Si:H) films of varying thickness. A low temperature evolution peak at about 380 °C was determined to arise from a surface monolayer. The bulk limited main peak of the evolution spectrum arises from a very narrow range of binding energies (<0.2 eV). The evolution spectra of a-Si:H, hydrogenated polycrystalline Si, and c-Si containing platelet structures are virtually identical in both size and temperature of the various peaks. Less than 5% of the H is released in an evolution peak occurring at about 800 °C, a peak possibly due to H bound at isolated dangling bonds.

A deuterium secondary ion mass spectrometry (SIMS) study of hydrogen diffusion in undoped and boron-doped a-Si0.86C0.4:H deposited by an electron cyclotron resonance plasma is described. The undoped films deposited at 250°C clearly indicated deuterium-hydrogen interdiffusion at T ≥ 350°C. The dispersion parameter a of the power-law time dependent diffusion constant D = D00(ωt)−α decreased from ∼0.3 at T = 350 and 400°C to ∼0.1 at 450°C, and the activation energy for a diffusion length of 1000 Å was ∼1.0 eV. These results are discussed in relation to previous studies of a-Si:H. The diffusion in ∼0.2 and ∼0.6 at.% B-doped a-Si0.86C0.14:H sharply differs from that in B-doped a-Si:H, where an enhancement of up to ∼103 was previously observed. In doped a-Si0.86C0.14:H, the diffusion of most of the H atoms is strongly suppressed, but a small fraction undergoes fast diffusion. IR Measurements indicate that the B-doping reduces the bulk-like Si-H stretch vibration at ∼2000 cm1. Upon annealing, the Si-CHn and C-H wag modes at ∼780 and ∼1000 cm−1, resp., increase, while the 640 and ∼2000 cm1 Si-H wag and stretch modes, resp., weaken, indicating transfer of hydrogen from Si- to C-bonds, in which the H atoms are apparently deeply trapped. As in a-Si:H, the fast diffusion component is apparently due to carrier recombination-enhanced weak Si-Si bond breaking. The results suggest that B-doping also induces microvoids and enhances the rate of breaking of weak Si-C bonds, leading to enhanced trapping of H.

For plasma-grown a-Si:H films deposited at substrate temperatures between 100 and 580°C the diffusion of hydrogen was studied by SIMS profiling of H/D interdiffusion and by hydrogen evolution. The solubility for deuterium penetration into a-Si:H is found to be limited by the hydrogen concentration. Diffusion prefactors and activation energies evaluated for a constant diffusion length follow a Meyer-Neldel rule. The diffusion free energies obtained by interdiffusion and evolution experiments agree closely up to about 400°C but disagree at higher substrate temperatures.

Previous measurements of local hydrogen motion in intrinsic, doped, and compensated hydrogenated Amorphous silicon (a-Si:H) using the 1H nuclear magnetic resonance (NMR) dipolar echo method have shown that the local hydrogen motion is much faster than the macroscopic diffusion would indicate but that the local motion follows the same trends with doping and defect density as the macroscopic diffusion. We report the effect of light soaking on the local motion of hydrogen in hydrogenated Amorphous silicon. Measurements are presented on 10−3 P-doped a-Si:H at 297 K. After light soaking with infrared-filtered, white light of intensity -400 MW/cm2 for 75 hours, the electron spin resonance (ESR) spin density increases to -101 spins/cm After light soaking 1H NMR dipolar echo measurements on this sample show that the dipolar spin-lattice relaxation time, T1D, is ∼4 Ms. After thermal annealing at 190 C for two hours the value of T1Dreturns to its pre-irradiation value of ∼ 11 Ms. The local rate of motion, which scales with TID-1 thus increases with the paramagnetic defect density. The general implications of this result for descriptions of both microscopic and macroscopic Motion of hydrogen in a-Si:H are discussed.

We discuss the growth of a-Si:H Materials and devices using a low pressure remote ECR plasma. We show that by using this plasma in an etching mode with a high H flux, we can grow high quality a-Si:H films at high temperatures (325–375 C). These films have significantly improved stability compared to standard a-Si:H films deposited using glow discharge. We can further improve the stability of these films by incorporating minute (sub ppm levels) of boron during growth. We also report on the fabrication of devices at these temperatures using this very reactive plasma. We discuss the precautions taken to obtain good devices, and discuss how the devices can be further improved.

We report on the free carrier Mobilities Measured by polarized electroabsorption Method. The result shows that the electron mobility reaches about 11.2cm2/Vs for device quality undoped a-Si:H, while the hole mobility is about 20% of the electron Mobility. Alloying with carbon leads to a continuous reduction of Mobility, with the largest drop (15%) for a carbon concentration of about 10 at.%, this being in sharp contrast to a less-pronounced effect by germanium alloying.

In this paper we describe a new method for the determination of the carrier lifetime (r) and the carrier Mobilities (μn, μp) in semiconductors. This technique utilizes a moving intensity grating that is generated by superposition of frequency shifted laser beams for the illumination of the sample. The Material parameters are extracted from the short circuit current in the sample induced by the moving grating as a function of it's velocity (vgr). We solve the continuity equations and Poisson's equation in the small signal approach for the modulated electron and hole densities and show how these densities result in an electric field that in turn acts on the electrons and holes in order to yield a short circuit current density jsc. From a fit of this expression for jsc to experimental data we determine μn, μp and τ.

An overview is given on the time and carrier density range covered by steady-state and transient photocarrier grating experiments which are commonly used to determine ambipolar transport parameters. In addition to the optically-detected transient grating method we discuss some details of a new version of the transient grating Method, the electrically-detected transient grating technique (EDTG) applied to hydrogenated Amorphous and microcrystalline silicon films. Numerical simulation shows that the diffusion coefficient is time and intensity dependent as expected for dispersive transport.

We have measured electroabsorption (EA) in a-Si:H p-i-n solar cells in order to estimate their built-in potentials Vbi. This Method was pioneered by the Osaka University group; in simple cases Vbi is identified with a parameter V0 obtained by a simple linear fit to the dependence of EA on external bias voltage. Using signal averaging techniques we are able to Measure V0 with a reproducibility of better than 2%. In one sample from the Institute of Energy Conversion with an a-SiC:H p+ layer we estimate Vbj = 0.97 ± 0.06 V, which is in good agreement with previous JVT Measurement. The reason that the error estimate is larger than the reproducibility error is the presence of several small effects not accounted for by the original EA theory: in particular V0 depends slightly on the laser wavelength. We attribute these effects to the small portion of Vbi dropped in the a-SiC:Hp+ layer, which has a different electroabsorption than a-Si:H. We discuss models for the wavelength dependence.

The constant photocurrent Method (CPM) is often used to measure the sub-bandgap absorption for the determination of the defect density. However, the absolute value of the derived defect density depends on the method of data analysis and the calibration factor. Normally the calibration factor is obtained from electron spin resonance (ESR) but the defect pool model gives rise to doubt whether ESR detects the same defects as CPm. Therefore, we propose combined total-yield photoelectron spectroscopy (TYPES) and CPM Measurements on n-type a-Si:H to determine the calibration factor. Furthermore, we calculate CPM spectra by extending an approach to simulate photoconductivity, taking into account the full set of optical transitions, and compare the results with standard evaluation Methods.

We have deposited n-i-p diodes with microcrystalline intrinsic layers for radiation detection applications. The diodes show interesting electrical characteristics which have not been reported before. From TOF Measurement for our best samples we obtained μe values which are about 3 times larger than our standard a-Si:H. for μτ values approximately a factor of 2 improvement was observed. The N*D values derived from hole-onset measurements show lower ionized dangling bond density than normal a-Si:H Material. We have proposed a simple model which can very well explain the experimental results.

The effective temperature definition which combines electric field and lattice temperature is critically discussed based on dark conductivity and drift mobility data measured over a wide range of temperature (10 K ≤ T ≤ 300 K) and electric field (102 V/cm ≤ F ≤ 6×105 V/cm). The reasonable values for the localisation lengths of electrons (αe ≡ 7–8 A) and holes (αh, ≡ 4 A) as well as the overall good agreement between predicted and deduced features support the effective temperature concept. A statistical transit time equation for transport at low temperature and high field is given and discussed based on time-of-flight data measured at T = 40 K.