Oxygen '96
Early Stages of Oxygen Precipitation in Silicon
INTERACTION OF POSITRONS WITH VACANCY-OXYGEN COMPLEXES AND OXYGEN
CLUSTERS IN Si
M. Fujinami
Advanced Technology Research Laboratories, Nippon Steel Corporation
1618 Ida, Nakahara-ku, Kawasaki 211, JAPAN
Keywords: positron, silicon, oxygen clusters, vacancy.
Positron annihilation spectroscopy is one of the most useful
techniques to detect oxygen-related defects as well as vacancy-type
defects in Si. In this study variable-energy positron annihilation
spectroscopy has been applied to study the behaviour of oxygen-related
defects in Si caused by implantation of oxygen ions. It is expected
that excess vacancies and interstitial oxygen atoms can interact
during the annealing stages and that oxygen-related and extended
defects are formed ultimately. The interaction of such defects with
positrons may give vital clues as to their composition. The
samples were prepared by implanting 2x1015 O+
ions/cm2 at 180 keV into p-type CZ-Si(100) wafers at room
temperature. Annealing of the samples from 573 to 10 K for 30 min was
carried out in a nitrogen atmosphere. The Nippon Steel Corporation
slow positron beam was utilised to measure the Doppler broadening of
the annihilation gamma-ray, characterised in terms of the S parameter,
with respect to the incident positron energy (S-E data). The S
parameter is defined as the ratio of the counts in a central region of
the annihilation photopeak to those in the whole of the photopeak.
The value of the S parameter increases when the positrons are trapped
at vacancy-type defects, since the overlap of the positron density
with (low momentum) valence electrons increases. When the positrons
are trapped at vacancy-type defects coupled with oxygen atoms, the
value of S is strongly dependent on the contribution of electrons in
the oxygen atoms.
The S values for the as-implanted sample are
much higher than those of the unirradiated one. The
multivacancy-oxygen complexes such as V3O are the primary
defect found to be formed by room temperature oxygen implantation and
are stable up to 673 K. The first annealing stage (873 K) is strongly
dependent on the oxygen implantation profile and hence, the depth in
the sample. In the implanted region, the S parameter becomes lower
than that for the Si bulk, suggesting the formation of
multivacancy-multioxygen complexes. While multivacancy-based defects
are still evolved in the proximity of the surface, where the oxygen
concentration is low, as the S parameter is large. Upon annealing at
1073 K, high S values are no longer observable, indicating that
vacancy-type defects are eliminated near the surface. But the S-E data
do not return to that of the unirradiated Si, and new positron
trapping sites with very low S are created. The level of S parameter
depends on the dose of oxygen in the range from 2x1014
/cm2 to 2x1015 /cm2. These results
suggest that these defects are due to the oxygen clusters which the
multivacancy-multioxygen complexes are transformed into. From the
ratio of the density of oxygen clusters to the average atomic oxygen
concentration, the mean number of oxygen atoms contained in the oxygen
clusters is estimated to be several tens of atoms. It is clearly
proven that positrons are a sensitive probe to detect
multivacancy-multioxygen complexes and oxygen clusters, whose size is
not observable by transmission electron microscopy.
Full list of abstracts | participants
list | main page.
Last modified: Mon Feb 19 12:11:15 GMT 1996
JG