Surface loss probability of atomic hydrogen for different electrode cover materials investigated in H2-Ar low-pressure plasmas

Abstract

In an inductively-coupled H 2 -Ar plasma at a total pressure of 1.5 Pa the influence of the electrode cover material on selected line intensities of H, H 2 , and Ar are determined by optical emission spectroscopy and actinometry for the electrode cover materials stainless steel, copper, tungsten, Macor , and aluminum. Hydrogen dissociation degrees for the considered conditions are determined experimentally from the measured emission intensity ratios. The surface loss probability β H of atomic hydrogen is correlated with the measured line intensities and β H values are determined for the considered materials. Without the knowledge of the atomic hydrogen temperature, β H cannot be determined exactly. However, ratios of β H values for different surface materials are in first order approximation independent of the atomic hydrogen temperature. Our results show that β H of copper is equal to the value of stainless steel, β H of Macor and tungsten is about 2 times smaller and β H of aluminum about 5 times smaller compared with stainless steel. The latter ratio is in reasonable agreement with literature. The influence of the atomic hydrogen temperature T H on the absolute value is thoroughly discussed. For our assumption of T H = 600 K we determine a β H for stainless steel of 0.39 ± 0.13.

Measurement and modeling of neutral, radical, and ion densities in H2-N2-Ar plasmas

“…In our experiment the geometry of the plasma chamber does not exhibit a simple cylindrical geometry and the plasma-surrounding walls consist of different materials which have different β rad . Furthermore, the radical temperature is a priori not known since radicals are mainly produced by dissociation which is accompanied by a release of potential Franck-Condon energy 44 and T rad can, therefore, exceed the gas temperature considerably. β rad is a function of the radical species 49,83 , the surface material 50,81,83,84 , but most likely also a function of the surface condition 85 (for example, substrate temperature and roughness) and even the plasma parameters 81,86 (for example, ion flux).…”

“…For a detailed discussion of the discrepancy of β rad between our and other groups, see Refs. 43,44 .…”

“…n H /n H 2 was determined by actinometry using the ratio of the H β /Ar 750 lines as described in Ref. 44 . The experimental uncertainty for the dissociation degree is 24 % (see Ref.…”

Wall loss of atomic nitrogen determined by ionization threshold mass spectrometry

“…The measured signal D(λ) (in arbitrary units) results from the emitted photons integrated along the line of sight through the plasma as a function of the wavelength λ in a time interval t int . The signal was relatively quantified by multiplying with the relative sensitivity curve R(λ) obtained for our experimental setting using a calibrated halogen lamp and D 2 arc discharge light source for calibration 8,13 .…”

“…For low-pressure plasmas atomic nitrogen is mainly produced by electron-induced dissociation of N 2 which is predominantly affected by the electron density and electron temperature of the respective plasma. The loss of atomic nitrogen in such plasmas is determined by the flux to the wall [3][4][5][6][7][8] . In our case, as in most other cases, the wall loss is determined by recombination of atomic nitrogen to form molecular nitrogen which desorbs from the surface.…”