Modeling the effect of dislocations on the minority carrier diffusion length of a semiconductor

Abstract

An analysis is given of the minority carrier diffusion and collection in a semiconductor with a surface junction, in the presence of a periodic array of straight dislocations perpendicular to the surface. A dislocation is characterized by its line recombination velocity and is attributed a finite cross section. An approximate solution is obtained for the three-dimensional charge collection probability φ(r) in the unit cell of the array. The integral of φ over the unit cell is used to introduce a new definition of effective diffusion length and to give its analytical expression. It is shown that the relative reduction of diffusion length produced by the dislocation array coincides with the fractional decrease of the total number of junction-injected carriers. This result establishes a connection between the efficiency of charge collection and luminescence of a sample containing dislocations. The analysis is extended to the case of a thin sample, and is compared to previous discussions of the same problem. The consequences of the theory are illustrated by numerical examples and applications to published experimental data.

Grain boundary effect on lifetime in high performance multicrystalline silicon during solar cell processing

“…The lifetime-prediction model, designed in Comsol, is based on Donolato's theory on dislocation recombination strength [15,16]. The model has been designed to fit the predicted lifetime to linescans perpendicular to GBs.…”

Optical Imaging of Diffusive Electrons with Different Kinetic Energies in a Semiconductor

“…Clearly, the three profiles are quite close to each other, which indicates that the diffusion lengths for electrons of different kinetic energies are approximately the same. The correct function for describing the profile should be a modified Bessel function [4,5,6] but corrected for the final laser spot size, which requires solving the diffusion equation numerically. As an approximation, a simple exponential decay function (e −r/L ) was adopted.…”

“…After the PL imaging data is obtained, a computer code, written in Wolfram Mathematica, is used to extract the electron diffusion length. The spatial profile of the PL image is the solution of a diffusion equation including a generation term and a depletion term [4,5,6]. It is usually a decay function of r with a parameter L, where r is the lateral distance measured from the generation site at r = 0; L is the electron diffusion length, assuming that the semiconductor is very thin and diffusion occurs only laterally.…”

Infrared birefringence imaging of residual stress and bulk defects in multicrystalline silicon

“…For example, dislocations can be formed via stress relaxation above the brittle-to-ductile transition temperature, reducing minority carrier lifetime. [115][116][117] Thermal gradients during crystal growth 23,[90][91][92] or cell processing 66,67 are well known to provoke dislocation formation, but similar pathways involving microdefect-related stresses have generally been underappreciated. We observe local stress along GBs ͑Fig.…”