Quantitative analysis of optical emission spectroscopy for plasma process monitoring

Abstract

In the field of plasma materials processing, various plasma parameters should be evaluated quantitatively and precisely to control the plasma process adequately, particularly with non-invasive methods, one of which is optical emission spectroscopy (OES) measurement. It has sufficient scientific feasibility to derive the electron density N e, electron temperature T e, and the electron energy distribution function (EEDF) even for various processing plasmas in a state of non-equilibrium. In this review, previous studies are reviewed to measure the N e, T e, and EEDF values of argon plasma with low-electron temperature (T e ≃ 1–10 eV) under not only low-pressure conditions but also atmospheric-pressure discharge using the OES measurement. First, to diagnose low-pressure discharge argon plasmas, we explain the basics and applications of the “collisional radiative model”, which models the population kinetics of the excited states in plasma at the elementary process level in non-equilibrium plasma. Methods for analyzing the plasma parameters are shown from the actual measurement results of emission spectra, including machine learning analysis of the excited-state populations. Next, the research results of the method to measure N e, T e, and EEDF are introduced for the measurement of atmospheric-pressure non-equilibrium plasmas using OES measurement of continuum emission, which also includes methods based on machine learning and data-scientific methods for the analysis of the OES data observed as bremsstrahlung of free electrons scattered against neutral molecules.