Charged and neutral particle dynamics in low pressure chlorine and oxygen plasmas: the role of collisional processes

International audienceThe modelling of low pressure industrial plasmas is strongly reliant upon fundamental data describing particle collisions in the gas phase and with surfaces. Such data include electron-heavy particle collision cross sections, surface loss coefficients of reactive neutral species (γ) and thermal energy accommodation coefficients for heavy particle interactions with surfaces (α). The methods required, both theoretically and experimentally, to generate the above data are extremely varied and as such the degree to which these data are known for a given plasma system varies significantly from process to process. Self-consistent plasma simulations provide a strong test for the accuracy of the complete set of fundamental collision data for a given system as the final result often hinges on the accuracy of the data for many different processes in concert. This work seeks to evaluate the state-of-the art of knowledge of a range of fundamental collision data by comparing the results of a self-consistent 2-D plasma simulation (HPEM [1]) and a range of experimental measurements in chlorine and oxygen inductively coupled plasmas. For both gases, it is found that the simulation is sensitive to the electron impact cross section set used as well as the surface coefficients γ and α. In both cases γ determines in large part the density of the reactive atomic species (either atomic chlorine or oxygen), α determines the neutral gas temperature and the electron impact cross sections define the electron density and temperature. Through comparison between the simulated and experimentally measured values of the discussed parameters the most appropriate collision data set for use in self-consistent plasma simulations is inferred. It is hoped that this work with stimulate further direct measurements and calculations of the discussed collision data. [1] M. J. Kushner, J. Phys. D: Appl. Phys., 42, 194013 (2009) Acknowledgement: Funding is acknowledged through the LABEX Plas@Par project, ANR-11-IDEX-0004-02, UK EPSRC Manufacturing Grant (EP/K018388/1) and the York-Paris Collaborative Research Centre