W. Ensinger et al., Plasma immersion ion implantation of complex-shaped objects: an experimental study on the treatment homogeneity, SURF COAT, 128, 2000, pp. 265-269
Plasma immersion ion implantation is a technique which is often quoted to t
reat complex-shaped objects homogeneously. The present study shows that thi
s is not necessarily the case, if the plasma conditions are not chosen in a
very narrow predetermined window. It is not only important to know the dos
e distribution of the implanted ions across a three-dimensional object, but
also the intensity of sputter etching depending on the position. Therefore
, two different techniques, sensitive to these issues have been applied. Th
e objects investigated are macro-trenches with different aspect ratios. U-s
haped stainless steel sample holders with different trench widths have been
lined with silicon wafer segments. Additionally a Ta metal strip, coated w
ith a Ta2O5 layer, has been folded into the sample holders. These trenches
have been treated in an argon plasma. The implanted Ar ion dose has been de
termined by Rutherford backscattering spectrometry of the Si pieces. The am
ount of sputtered material could be evaluated by determining the colour cha
nge of the Ta2O5/Ta strips. The colour of the oxide on the metal is very se
nsitive to the oxide thickness. It is shown that under the conditions appli
ed, the treatment is not homogeneous. A larger amount of Ar is implanted at
the trench top side compared to the trench bottom. The trench side walls a
re hardly affected by ion implantation. The sputter results are different f
rom the implantation ones. It is shown that the trench bottom is most affec
ted by sputtering, considerably more than the top side. The inner side wall
s of the trenches are for all aspect ratios also hardly affected by sputter
ing. The results are explained in terms of the sheath dimensions and the ex
pansion of the sheath as well as by the three-dimensional geometry of the s
ample holders. (C) 2000 Elsevier Science S.A. All rights reserved.