Investigation of electron density and temperature in an argon inductively coupled plasma and feasibility of plasma monitoring

Abstract

In semiconductor etching processes, plasma parameters such as plasma density and electron temperature, are important factors which affect the processing outcomes. Monitoring such parameters is crucial as the etching processes are becoming more complex. This work conducts a feasibility study on non-intrusive plasma monitoring in a two-chamber argon inductively coupled plasma. As a non-intrusive diagnostic tool, a VI probe was used to measure the voltage and current waveforms induced on the antenna of the ICP. To measure the plasma parameters in both the source and the expanding region, cut-off probes were used to measure plasma density whereas Langmuir probes were used to measure electron temperature. To investigate the effect of capacitive coupling on plasma parameters, measurements with and without a Faraday shield were compared. For plasma monitoring, a circuit model using Bayesian probability theory was investigated, and its limitations are discussed. As a better alternative, a neural network was constructed to infer plasma parameters and the results are presented.