Basic requirements for quantitative SIMS analysis using cesium bombardment and detection of MCs + secondary ions
The emission of MCs + secondary ions from targets containing element M was studied as a function of the impact angle of 6-15 keV Cs + primary ions. The MCs + yields measured with sample materials like Cu, Ge and GaAs suggest that the degree of ionization of these dimers is essentially independent of the impact angle θ, at least for θ up to 60° to the surface normal. With Al and Si, on the other hand, the relative MCs + yields at near-normal incidence were anomalously low, i.e., almost two orders of magnitude lower than expected on the basis of the data for Cu or Ge. Comparison with literature data relating to controlled Cs deposition suggests that the efficient neutralization of emitted AlCs + and SiCs + ions at near-normal beam incidence is due to a lowering of the work function by up to 3 eV. This implies the buildup of a high stationary surface concentration of cesium. In order to keep the cesium coverage at a sufficiently low level, bombardment has to be carried out at θ > 60°, in which case the matrix dependent cesium coverage and the associated SIMS matrix effect apparently become negligibly small. The absence of a matrix effect was supported by the observed close agreement between the SiCs + yields for Si and SiO 2. Even the elemental sensitivities, measured for 16 elements, show a relatively small variation. In the extreme case, i.e., comparing the elements Al and Zn (high MCs + yield) with As (low yield) the MCs + yield ratio amounts to a factor of only 30, small compared to variations known from standard SIMS involving the detection of atomic secondary ions. On the basis of the present and previous results it is suggested that the matrix effect under cesium impact originates from material dependent differences in the sputtering yield and the diffusional transport of cesium to the bombarded surface.