Author : Caleb Timothy Nelson
Publisher :
Page : 286 pages
File Size : 47,87 MB
Release : 2010
Category : Fluorocarbons
ISBN :
Book Description
Understanding dominant reaction channels for important gas-phase species in fluorocarbon plasmas is crucial to the ability to control surface evolution and morphology. In order to accomplish this goal a modified GEC reference ICP reactor is used in tandem with Fourier transform infrared spectroscopy (FTIR) to measure the densities of stable species. Integrated absorption cross-sections are presented for all fundamental bands in the 650 cm -1 to 2000 cm -1 region for C 3 F 6 , C 4 F 8 , C 3 F 8 , C 2 F 6 , C 2 F 4 , and CF 4 . The results show that although the absorption profile changes significantly, the integrated absorption cross-sections, with the exception of CF 4 , do not change significantly as gas temperature increases from 25°C to 200°C. However, the internal temperature of the absorbing species can be estimated from the rotational band maximum in most cases. Species densities obtained with the aforementioned cross-sections are used with a novel analysis technique to quantify gain and loss rates as functions of residence time, pressure, and deposited power. CF 4 , C 2 F 6 , C 3 F 8 , and C 4 F 10 , share related production channels, which increase in magnitude as the plasma pressure, deposited power, or surface temperature are raised. CF 2 is primarily produced through a combination of surface production (the magnitude also increases with temperature) and electron impact dissociation of C 2 F 4 , while it is predominantly lost in the large reactor to gas-phase addition to form C 2 F 4 . Time-resolved FTIR results are used to measure a cross-section of 1.8x10 -14 cm 3 /s for the reaction between CF 2 radicals creating C 2 F 4 . Finally, C 2 F 4 originates through the electron impact dissociation of c- C 4 F 8 . The loss process for C 2 F 4 is undetermined, but the results indicate that it could occur on reactor surfaces. Neither the density of fluorine nor the ion flux to the chuck surface changes substantially with wall temperature. We show that increases in the deposition rate in a heated chamber are due to an increase in the fluxes of depositing neutral species. Furthermore, the sticking coefficient for these species does not change significantly as a function of surface temperature. Instead, surface temperature elevates the yield of etchant species, which terminate broken bonds to increase the desorption rates of stable species.