A fast method for the calculation of electron number density and temperature in laser-induced breakdown spectroscopy plasmas using artificial neural networks

Abstract

A fast and precise method for the determination of electron temperature and electron number density in laser-induced plasmas is presented. The method is based on the use of a simple artificial neural network (ANN), trained on a suitable set of laser-induced breakdown spectroscopy spectra. The training procedure is quite fast; once the ANN is set, the determination of plasma temperature and electron number density is almost instantaneous, allowing the possibility of measuring these parameters, with good precision, in real time. A direct application of this new method could be the characterization of plasmas generated during pulsed laser deposition process of thin films and nanoparticles generation. The plasma electronic parameters will help to tune the energies involved in the stoichiometry and crystallization control of those nanostructured materials. As an example, the characteristics of the plasma induced by a Nd:YAG laser on a pure titanium target are determined, at different laser fluences.