Revisiting argon glow discharges: where do the current models fail and why

In this work we develop a 0D kinetic model for argon, which is validated against various experimental data from the literature for a wide range of pressures and reactor geometries. We discuss the measurements of the different data sources and perform an overview of the available rate coefficients for the various processes. The validation of the model includes a comparison against the reduced electric field, the electron density and temperature, and the metastable density when available. A straightforward sensitivity analysis on the rate coefficients and cross-section datasets used for the electron impact excitation and ionisation is performed, defining the systematic error inherent to the model. The model provides a qualitatively good picture of the main trends and underlying physics, but some quantitative discrepancies remain when comparing the predictions with experiment. A machine-learning framework will be developed to enhance the agreement by identifying key sources of improvement/error on the kinetic scheme.