Dynamic electrical response and discharge mode transitions in CF4/Ar capacitively coupled plasmas with automatic matching networks

Abstract

This study investigates the dynamic stabilization of capacitively coupled plasma (CCP) discharges in a CF4/Ar environment through an automatic external L-type matching network using one-dimensional particle-in-cell/Monte Carlo collision (PIC/MCC) simulations. Driven at a source voltage of 100 V and a frequency of 27.12 MHz, the plasma system exhibits complex impedance evolution and non-linear kinetic transitions as the CF4 mixing ratio increases. Our results demonstrate that the automatic matching algorithm, by iteratively optimizing the matching capacitances (Cm1 and Cm2), achieves a series-like resonance that magnifies the electrode voltage to 3.5–4.5 times the source value. Specifically, at high electronegativity (80% CF4), the discharge encounters a “high-resistance barrier” characterized by severe electron depletion and initial impedance mismatch. The matching network facilitates a synchronized “staircase” evolution of the electrode voltage, current, and charged species densities, effectively preventing discharge collapse and ensuring global stability. Spatial analysis reveals a fundamental heating mode transition: While low CF4 fractions support sheath-edge localized heating (α and Drift-Ambipolar modes), the 80% CF4 case shifts to a bulk-dominated Ohmic heating regime with electron temperatures exceeding 4 eV. This transition is further confirmed by the temporal evolution of the electron energy probability function, which shows a robust recovery of the high-energy tail during the matching iterations. This work underscores that automatic external circuit matching is not merely a tool for power efficiency but acts as a vital kinetic regulator that sustains discharges in attachment-heavy, high-pressure environments, providing critical insights for industrial plasma etching processes.