Chemistry Reference
In-Depth Information
Thus, atomic steps on silicon (111) surface at elevated temperature during gold
deposition shows nontrivial behavior complicated by the impact of the DC heating.
The physical reason for changing of the effective charge of the silicon adatoms
still is not clear. It was reported that gold atoms on the silicon surface may act
as an activator during the surfactant-mediated growth. During growth of silicon on
the Si(111)-(5
2) Au surface gold atoms segregated to the topmost surface. The
diffusion length of silicon atoms on gold-deposited silicon surface was reported to
be larger than that on the (7
×
7) surface [
65
]. Aoki et al. [
66
] reported decreasing of
adatom density on Si(111) surface, which was measured from atomic step shifting
toward the step-down direction at 780
◦
C after deposition of 0.3ML of gold. The
advance of the atomic steps means that silicon atom density on the gold-adsorbed
(1
×
1) surface decreases with an increase of gold coverage. Taking into account
these facts one can consider modification of the detachment/attachment rates of the
atomic steps induced by gold atoms in a way similar to surfactant-mediated epitaxy.
Our observations of the atomic step motion with increased rates in both directions
(step-up and step-down) confirm this consideration. So we can conclude that gold
atoms on the silicon surface produced drastic modifications of the diffusion proper-
ties on the surface at this temperature.
×
11.2.4 The Kinetics of the Gold-Induced Step Bunching
The kinetics of step bunching induced by gold adsorption was investigated in real
time and compared with one of current-induced step bunching at the same tem-
perature. Figure
11.6
demonstrates series of REM images illustrating the process
of gold-induced step bunching at 900
◦
C under the step-down DC direction. We
found that step bunching at this condition occurs faster than one on clean sili-
con surface. The temporal measurement of average distance between neighboring
bunches revealed that the bunching rate was at least two order times higher than one
without gold deposition at the same temperature. It should be pointed out that the
bunch formation in this case occurs mainly by means of atomic step moving in the
direction of sublimation. However, some step bunches were observed to be moving
in the growth direction producing coagulation of several step bunches in one.
While all presented results were obtained for the same sample temperature of
900
◦
C, we also investigated the temperature dependence of the silicon morphology
during gold deposition. In the first temperature range (830-1050
◦
C) we have no
observed significant differences in this effect and the same transitions were qual-
itatively registered. Only slow increasing of the critical gold concentration with
temperature enlargement was found probably due to increased departing of gold
atoms from the surface at higher temperatures. The most essential fact was observed
in the second temperature range (1050-1250
◦
C) where, in contradiction to the
first temperature one, two critical gold concentrations were observed and the three
gold coverage ranges were found only. At higher temperatures the investigations
of gold-induced transformations are embarrassed due to high probability of gold
atoms to depart from the surface and restricted power of gold evaporator. However,
