860
T. Hrn ꢀc ꢁı ꢀr , V. Nehasil / Surface Science 507–510 (2002) 859–864
surface. The intensity of the mixed beam was
2
(QTM). The evaporation rate stability during the
long evaporation time was ensured by the mea-
surement of the evaporated ion current [5].
13
about 1:5 ꢀ 10 molecules/cm at the sample sur-
face. Three ratios of CO and O partial pressures
2
have been used ꢁ1:1,2:1 and 1:2. The total pres-
sure in the vacuum chamber during the scattering
experiments was 1 ꢀ 10ꢁ Pa. The scattered or
desorbed molecules were monitored by the quad-
rupole mass spectrometer (QMS) Leybold Inficon
Immediately after evaporation,samples were
annealed at 630 K for 10 min. There were two main
reasons of the annealing. Firstly,temperature stim-
ulated alloying of Rh and Al was expected [5].
Secondly,samples were stabilised preliminary to
following reactive scattering experiments. Within
these experiments,samples were heated at 570 K
maximally. Annealing of samples at higher tem-
perature than 630 K leads to the significant incor-
porating of Rh layer into Al substrate [5].
Reactive scattering experiments started at room
temperature. The sample was placed into the
mixed beam of CO and O2 and after that the
temperature rose gradually to 570 K and dropped
back to the room temperature again. Mass spectra
of scattered and/or desorbed species were col-
lected during this temperature treatment. This
annealing–cooling cycle was repeated several times
to ensure the stability of investigated effects. The
rate of temperature variation was held at 0.25 K/s.
During the first annealing-cooling cycle,the sam-
ple was oxidised significantly due to impinging
oxygen. With regard to the AES results,we assume
this oxidation applies to the Al component only.
After finishing experiments,samples were taken
from the vacuum chamber and they were analysed
by electron micro probe (EMP).
6
2
000. It was positioned in a differentially pumped
chamber with a small entrance orifice. The sample
was positioned in proximity to this orifice,in order
to examine small surface area where the surface
temperature distribution could be considered as
homogeneous. The temperature was determined
by a chromel–alumel thermocouple fixed on the
sample holder.
A double-pass cylindrical mirror analyser DESA
00 equipped with an electron gun (Steib Instru-
1
ments) was used for Auger electron spectro-
scopy (AES) and electron energy loss spectroscopy
(
EELS) measurements. The primary energy of
an electron beam was 2.5 keV. Measured AES
intensities were normalised to a reference Rh
foil peak intensity. The energy of the Auger Rh
(
45 45
M N N45) peak measured on the reference foil
was equal to 297.3 ꢂ 0.3 eV during all AES ex-
periments.
The Al samples were prepared from Goodfel-
low materials (GFM) 99.999% Al cut into a plate
2
of 10 ꢀ 10 mm area and 1 mm thick. The chem-
ical etching followed the sample polishing. Details
of these procedures are given in [5]. Inside the
vacuum chamber,substrates were sputtered by
The reference Rh sample was GFM 99.9% Rh
þ
10 ꢀ 10 ꢀ 0:125 mm. It was sputtered by Ar ions
(incident energy 1 keV,ion current density at the
þ
ꢁ6
2
Ar ions (incident energy 2 keV,ion current den-
sample 10 A/cm ,15 min). After that,annealing
at 770 K for 20 min in oxygen atmosphere fol-
lowed.
sity at the sample 5 ꢀ 10ꢁ A/cm ,60 min). After
that,annealing at 770 K for 10 min followed
and the ion sputtering with the same parame-
ters has been performed again. After this proce-
dure,the aluminium Auger peaks exhibited only a
metal form,no chemical shift of the peak was
observed. Neither carbon nor oxygen signal was
detected.
7
2
3. Results and discussion
Parameters characterising deposited samples
are given in Table 1. This table contains a com-
parison between the Rh layer thickness estimated
by QTM and EMP techniques. For EMP,it should
be noted the measured thickness was close to the
resolving power of this technique. In the follow-
ing text,the thickness estimated by QTM will be
used instead. The last column of Table 1 contains
Rhodium was evaporated at a room tempera-
ture from the micro electron beam evaporation
source (MEBES) [6]. In this evaporation cell,a
GFM 99.9% Rh wire was evaporated by the high
voltage accelerated electrons. Evaporating rates
were estimated by a quartz thickness monitor