Ion bombardment sputtering of gallium arsenide, gold and tantalum.

Sputtering yields have been measured for neon, argon and krypton sputtering of gold and for argon and tellurium sputtering of gallium arsenide in the energy range from 50 to 400 keV. Values of yields, estimated by activation analysis of sputtered deposits, are compared with other experimental results and with theory and are shown to be too high as a result of errors in ion dose measurement. Yields for gold and gallium arsenide are shown to increase by about 10% over the dose range from 1017 to 1017 argon ions/cm2 as a result of changes in surface topography during sputtering. Polythene hemispheres were used to collect sputtered atoms and are shown to be very efficient (> 90%) for gallium and gold atoms. Measurements of surface stoichiometry, sputtering of cooled and heated targets and the use of nickel coated collectors were carried out for gallium arsenide and led to the conclusion that the sticking efficiency of sputtered arsenic atoms was very low giving rise to lower estimates of arsenic yield compared with those of gallium. The Rutherford backscattering technique has been used to assess the quality of tantalum thin films deposited on glass and on vitreous carbon substrates by R.F. Sputtering and by electron beam evaporation. Sputter deposited films were shown to be superior with respect to purity and reproducibility. Effects of implanting to very high doses (> 1018 ions/cm2) with argon, nitrogen or oxygen ions were examined by backscattering which has been shown to be a very useful tool for studying thin films. Estimates of ion ranges were compared with L.S.S. theory and sputtering yields were also calculated for the different ions used. Finally the limitations of the backscattering method are discussed with respect to very high impurity levels