Resonances and surface waves in bounded plasmas

Summary form only given. Surface waves provide a promising means of creating large area plasmas. These waves can uniformly distribute the excitation energy and while presenting a small resistance and zero reactance to the driving source. Experimentally and in our simulations, the electron temperature is low (like 1-3 eV) as is the plasma potential (like 10 Te). The use of surface waves experimentally, and now industrially, to sustain large area plasma sources with device size is comparable to free space wavelength have motivated us to refine the theories of Parker et al. (1964) and Cooperberg, 1998) to be fully electromagnetic. The wave dispersion predicted by the electromagnetic theory differs from the predictions of the prior theories and the results illuminate limitations of the electrostatic model. The use of surface waves have also motivated us to explore the mechanisms by which surface waves heat the plasma. In our 1-D electrostatic simulations (which model a resonant discharge experiment performed by Godyak et al. in 1994), high velocity electron bunches are formed in the sheaths and are alternatively accelerated from each sheath into the bulk plasma each RF cycle. We speculate similar mechanisms provide the ionization in surface wave discharges. We also see in these simulations the plasma makes an abrupt transition from capacitively coupled to resistively coupled and the series resonance "locks" onto the drive frequency; these abrupt transitions resemble "mode-jumping" seen experimentally in large area sources. Furthermore, the density profile of the plasma tracks the drive frequency while in the resonant mode giving a new mechanism by which the plasma parameters can be controlled. We are currently investigating the effect the driving electrode shape has on these resonances and conducting 2-D simulations of a large area surface wave source to explore the ignition of surface wave devices and how the plasma fills in the device.