Evelyne Sauvain

Effects of dangling bonds on the recombination function in amorphous semiconductors

Abstract A closed-form expression for the recombination function in steady-state illuminated hydrogenated amorphous silicon (a-Si: H) is given for the case that recombination occurs mainly via the dangling-bond states. Based on the three charge conditions (positive, neutral or negative) for the dangling bonds, the three corresponding occupation functions are derived; an expression for the recombination function R DB is thereby obtained. The latter differs considerably from the commonly used Shockley-Read and Hall function RSRH. As an illustration, the limiting carrier in an a-Si: H p-i-n solar cell under reverse voltage is shown to be either one with a longer drift length, using KSRH, or one with a shorter drift length, using R DB. Therefore one can conclude that the use of the proper recombination function is critical for a discussion of the relevant physical parameters involved in the description of p-i-n devices. Interpretation for ambipolar diffusion length and photoconductivity are also mentioned as ...

Effect of the dangling-bond charge on the ambipolar diffusion length in a-Si: H

Abstract The light intensity dependence of the photoconductivity (σph), of the ambipolar diffusion length (L amb) and of the ratio of the effective mobilities (b = μn∗/μp∗) has been measured on undoped a-Si: H. L amb and b have been measured using the steady-state photocarrier grating (SSPG) method. The kinetics of the light-induced degradation (Staebler-Wronski effect) of a typical undoped hydrogenated amorphous silicon (a-Si: H) film is presented: it illustrates that the measured value of L amb varies in a quite different manner with deep defect density that σph. All these results can be explained by incorporating the effect of the dangling-bond charge into the model (used hitherto), that included only the effect of the charge trapped in the bandtails.

Subbandgap absorption spectra of slightly doped a-Si:H measured with constant photocurrent method (CPM) and photothermal deflection spectroscopy (PDS)

Abstract The authors report on systematic measurements of the absorption spectra in the visible near-infrared range performed on a series of slightly phosphorus- and boron-doped amorphous hydrogenated silicon (a-Si:H) samples. Optical transmission spectra, Constant Photocurrent Method (CPM) and Photothermal Deflection Spectroscopy (PDS) have been used to measure the absorption coefficient (α) in differently doped samples. Variation of the gas phase doping levels in the ppm range is shown to lead to relevant differences between CPM and PDS spectra. The latter demonstrate that additional care has to be taken when evaluating deep defect density from CPM when the Fermi level is not constant, and especially so, if the Fermi level is below midgap.

Characteristic lengths for steady-state transport in illuminated, intrinsic a-Si:H

The steady-state transport equations are solved for a general small-signal case. This solution leads to the spatial distribution of the carriers in the bulk of undoped hydrogenated amorphous silicon (a-Si:H) under illumination. The general small-signal differential equations for electrons and holes, including a local and/or external electric field, are first established. Subsequently, the analysis is restricted to the case with a negligible external electric field. Two characteristic lengths appear in the analytical solution. They are interpreted for the lifetime regime as the ambipolar diffusion length and the dielectric relaxation length. Depending on the material parameters, one or the other of these lengths dominates the transport. Illustrative numerical examples for typical solar-cell-quality a-Si:H are given. >

Characteristic lengths for transport in illuminated intrinsic a-Si:H

Abstract The five coupled differential equations that describe one-dimensional transport in semiconductor are linearised, using a small-signal approach. The general homogeneous solution of this system of differential equations has four characteristic lengths. Considering diffusion and drift separetely, they reduce to two characteristic lengths L+ and L. Furthermore, a material parameter allows one to distinguish between two different behaviors: the “relaxation” and the “lifetime”-type material. In “lifetime”-type material, it is the ambipolar drift or diffusion length that dominates transport. The dielectric relaxation lengths governs space-charge, if any. These theoretical results are discussed in the experimental framework of hole/ambipolar μτ measurements such as Steady State Photcarrier Grating (SSPG), Surface Photo Voltage (SPV), and Steady-State Hecht Plot (SSHP). Of these, SSPG is considered the only reliable method to measure small-signal μτ.

Consistency between experimental data for ambipolar diffusion length and for photoconductivity when incorporated into the ‘‘standard’’ defect model for a‐Si:H

Reasonable consistency between experimental data for the ambipolar diffusion length and experimental data for the photoconductivity is demonstrated for steady‐state measurements performed on a‐Si:H layers. This consistency is obtained based on the ‘‘standard’’ defect model for a‐Si:H. In this model the dangling bonds are taken into account, considering their amphoteric behavior and treating then as recombination centers, whereas the band tails are taken into account as simple two‐valued defects acting as traps. Consistency is obtained based on (1) a particular form of the recombination function such as is considered appropriate for the dangling bonds, as well as, additionally, (2) the local charge neutrality condition. The experimental data analyzed are power laws of the ambipolar diffusion length and of the photoconductivity (versus light intensity); they are obtained for a series of slightly p and n‐doped samples including the undoped case.

Effects of dangling bond charge on ambipolar transport measurements in a-Si:H

The dependence of the ambipolar diffusion length (L amb ) and of the photoconductivity (σ ph ) on the dangling bond density have been investigated by observing the increase of the dangling bond density during light soaking (Steabler Wronski effect). Two regimes are observed: in the first one, L amb is independent of the dangling bond density, whereas σ ph decreases. In the second regime, L amb 2 decreases in the same way as σ ph . The effect of dangling bond charge on ambipolar transport can explain these two behaviors.

Assessment of Free Electron to Free Hole Ratios in a-Si:H By Ambipolar Diffusion and Photoconductivity Measurements

The authors report on systematic photoconductivity σ ph and ambipolar diffusion length L amb measurements carried out on a series of lightly p- and n-doped samples and monitored throughout the entire light-induced degradation process. The results are interpreted based on a transport model involving free carriers (nf and pf), localized charge on bandtails, as well as dangling bond and dopant charges.

The µτ Problem: New Results on Micro-Doped a-Si:H Films

Comparative µτ-product measurements performed on a series of slightly doped (micro- doped) a-Si:H films by steady-state photocarrier grating (SSPG), steady-state photoconductivity (SSPC) and time of flight (TOF) are presented. The observed discrepancy between transient and steady-state majority carriers µτ-products is discussed within the framework of a dangling bond recombination model. Furthermore, results on the variation of µτ with doping and light-soaking are reported and ratio of capture cross-sections of charged to neutral defects are deduced.

Measurement of ambipolar mobility-lifetime product and its significance for amorphous silicon cells

In order to evaluate correctly the ambipolar diffusion length (L/sub amb/) or the ambipolar drift length (L/sub e/) from a steady-state photocarrier grating (SSPG) diffusion or drift measurement, the condition of charge quasi-neutrality has to be maintained everywhere in the material (ambipolarity condition). This is shown theoretically by calculating the experimentally accessible parameter ( beta ) without assuming that the ambipolarity condition holds. The effect of nonambipolar behavior on the experimental plots, both for diffusion and for drift, is derived. Measured SSPG plots for undoped a-Si:H are given, illustrating both ambipolar and nonambipolar cases.<<ETX>>