Heavy-particle induced secondary electrons in capacitive radio frequency discharges driven by tailored voltage waveforms

Abstract

Particle-in-Cell/Monte Carlo Collision simulations are performed to investigate the effects of heavy-particle induced secondary electrons (SEs) on the ionization dynamics and on the control of ion properties at the electrodes in geometrically symmetric capacitively coupled argon discharges driven by tailored voltage waveforms. The driving voltage waveform is composed of a maximum of four (1≤N≤4) consecutive harmonics of the fundamental frequency of 13.56 MHz and is tailored by adjusting the identical phases of the even harmonics, θ. The simulations are carried out at neutral gas pressures of 3 Pa (nearly collisionless low-pressure regime) and 100 Pa (collisional high-pressure regime). Different approaches are used in the simulations to describe the secondary electron emission (SEE) at the electrodes: we adopt (i) constant ion-induced secondary electron emission coefficients (SEECs), γ, and (ii) realistic, energy-dependent SE yields for ions and fast neutrals. The mean ion energy at the electrodes, ⟨Ei⟩, can be controlled by θ at both pressures, for both approaches adopted to describe the SEE in the simulations. At a low pressure of 3 Pa, we obtain largely different dependencies of the ion flux at the electrodes, Γi, on θ, depending on the value of the γ-coefficient. For γ=0.2, Γi remains nearly constant as a function of θ, independently of the choice of N, i.e., the mean ion energy can be controlled separately from the ion flux by adjusting θ. However, for values of γ different from 0.2, the quality of the separate control of the ion properties changes significantly. At a high pressure of 100 Pa, independently of the choice of γ, for a given N≥2, the ion flux varies as a function of θ. At both pressures, the surface conditions affect the plasma parameters and the quality of the separate control of ion properties at the electrodes. Adopting realistic, energy-dependent SE yields for heavy particles in the simulations can lead to significantly different results compared to those obtained by assuming constant SEECs.