ISSN:
1432-0630
Keywords:
07.75
;
61.70
;
79.20
Source:
Springer Online Journal Archives 1860-2000
Topics:
Mechanical Engineering, Materials Science, Production Engineering, Mining and Metallurgy, Traffic Engineering, Precision Mechanics
,
Physics
Notes:
Abstract The minimum-detection limits achievable in SIMS analyses are often determined by transport of material from surrounding surfaces to the bombarded sample. This cross-contamination (or memory) effect was studied in great detail, both experimentally and theoretically. The measurements were performed using a quadrupole-based ion microprobe operated at a secondary-ion extraction voltage of less than 200 V (primary ions mostly 8keV O 2 + ). It was found that the flux of particles liberated from surrounding surfaces consists of neutrals as well as positive and negative ions. Contaminant species condensing on the bombarded sample could be discriminated from other backsputtered species through differences in their apparent energy spectra and by other means. The apparent concentration due to material deposited on the sample surface was directly proportional to the bombarded area. For an area of 1 mm2 the maximum apparent concentration of Si in GaAs amounted to ∼5 × 1016atoms/cm3. The rate of contamination decreased strongly with increasing spacing between the bombarded sample and the collector. The intensities of backsputtered ions and neutrals increased strongly with increasing mass of the target atoms (factor of 10 to 50 due to a change from carbon to gold). The effect of the primary ion mass (O 2 + , Ne+, and Xe+) and energy (5–10keV) was comparatively small. During prolonged bombardment of one particular target material, the rate of contamination due to species not contained in the sample decreased exponentially with increasing fluence. In order to explain the experimental results a model is presented in which the backsputtering effect is attributed to bombardment of surrounding walls by high-energy particles reflected or sputtered from the analysed sample. The level of sample contamination is described by a formula which contains only measurable quantities. Cross-contamination efficiencies are worked out in detail using calculated energy spectra of sputtered and reflected particles in combination with the energy dependence of the sputtering yield of the assumed wall material. The experimental findings are shown to be good agreement with the essential predictions of the model.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF00616059
Permalink
