Porosity and roughness determination of porous silicon thin films by genetic algorithms

TL;DR

This paper presents a genetic algorithm-based method to accurately determine the porosity, surface roughness, and optical constants of porous silicon thin films from reflectance data, validated against SEM measurements.

Contribution

It introduces an evolutionary algorithm approach for extracting multiple properties of porous silicon films solely from optical reflectance measurements.

Findings

Genetic algorithms accurately fit reflectance data.

Porosity and surface quality correlate well with SEM results.

Effective medium approximation effectively models optical constants.

Abstract

The problem of determining the porous silicon (PSi) optical constants, thickness, porosity, and surface quality using just reflectance data is board employing evolutionary algorithms. The reflectance measurements were carried out of PSi films over crystalline silicon (c-Si) substrate, and the fitting procedure was done by using a genetic algorithm. The PSi is treated as a mixture of c-Si and air. Therefore, its effective optical constants can be correlated with the porosity trough effective medium approximation (EMA). The results show that genetic fitting has a good match with the experimental measurements (Near UV-Vis reflectance) and the thickness obtained by scanning electron microscopy (SEM).