Surface Science 167 (1986) L175-LI80
North-Holland, Amsterdam
L175
SURFACE SCIENCE LETTERS
O B S E R V A T I O N OF A NEW ATOMIC-LIKE PEAK IN THE IONI N D U C E D AUGER S P E C T R U M OF Si
L. D E F E R R A R I I S a,d) O. G R I Z Z I b), G.E. Z A M P I E R I , E.V. A L O N S O d)
a n d R.A. B A R A G I O L A c,d)
Centro Atomico Bariloche, Instituto Balseiro, 8400 Bariloche, Argentina
Received 30 August 1985; accepted for publication 21 November 1985
We have observed Si L-shell Auger electrons from a silicon surface bombarded with 20 keV
Ar + ions at different angles of incidence. The measurements of angle-resolved electron energy
distributions allowed us to separate the contributions to the Auger electron spectra coming from
de-excitations inside and outside the solid. At grazing angles of incidence and observation, the
Auger electron distributions resemble those from gas-phase atomic collisions. In these conditions
we observed an atomic-like Auger peak with an energy close to that of the high energy edge of the
Si L2-tVVtransition.
The L-Auger spectra of Mg, A1, a n d Si excited by heavy-ion b o m b a r d m e n t
have some interesting features which are the subject of current studies. Three
or more n a r r o w peaks are clearly observed s u p e r i m p o s e d o n a b r o a d structure.
The narrow peaks, resembling atomic-like transitions, have received different
i n t e r p r e t a t i o n s [1-4]; the m a i n p o i n t of discussion b e i n g whether these Auger
electrons are emitted inside or outside the solid [1,5-7]. The b r o a d structure is
very similar to the Auger spectrum excited by electron b o m b a r d m e n t , a n d has
been u n a n i m o u s l y assigned to L2,3VV transitions [1-4] involving two valence
b a n d 0¢) electrons. A close c o m p a r i s o n reveals that in the i o n - i m p a c t case the
b r o a d structure extends to higher energies t h a n in e l e c t r o n - i n d u c e d spectra.
Some authors observed this discrepancy in the high energy edge in A1 a n d Si
a n d ascribed it to possible changes in the electron energy d i s t r i b u t i o n in the
solid d u r i n g ion b o m b a r d m e n t of Al[3], a n d charging of the sample in the case
of Si[2].
Recently, Whaley a n d T h o m a s [8] (WT) c o n f i r m e d the existence of this
discrepancy in the A1 a n d Si spectra, a n d its absence in the Mg spectrum.
a) Fellowof the Consejo Nacional de InvestigacionesCientificas y Tecnicas.
bl Universidad Nacional del Comahue, Centro Regional Bariloche. 8400 Bariloche, Argentina.
c) ALTES S.E., C.C. 112, 8400 Bariloche, Argentina.
d) Instituto Balseiro, Universidad Nacional de Cuyo.
L176
L. de Ferrariis et al. / Auger spectrum of Si
These authors also proposed a detailed modelling of the spectra which accounts
for all the significant features. Their model assumes that each spectrum is the
superposition of a band-like component, taken equal to the electron-induced
L2.3VV spectrum, and atomic-like peaks corresponding to excited sputtered
neutral atoms and ions decaying in free space. To obtain an accurate fit of the
A1 and Si spectra the authors assumed the existence of a new atomic-like peak,
located at almost the same energy as the high-energy edge of the L 2 N V
feature. Since this peak is superimposed on the sharp edge of the L2.~VV
feature, it appears in normal spectra only as a small shoulder, and could
account for the mentioned shift of the high-energy edge in the ion-induced
spectra.
Saiki and Tanaka [6] (ST) showed that when Auger electron spectra are
observed at emission angles close to the surface, the atomic-like peaks predominate over the broad band-like component. They found also a good
agreement with the assumption of a cosine distribution for the bulk structure
and an isotropic distribution for the atomic-like peaks, coming from the decay
of excited sputtered atoms in vacuum. In contradistinction with WT they
attributed the small shoulder in the high-energy edge to Auger electrons
generated inside the solid.
Our work was intented to clarify this controversy. We studied the Auger
electron spectra of Si bombarded with 20 keV Ar + ions at different angles of
incidence of the projectile and varying the emission angle at which the
electrons are observed.
Experiments were performed under ultra-high vacuum conditions ( < 10 ~
Tort) with an equipment which will be described in detail elsewhere.
The sample was a commercial single crystal Si (1000 ~2 cm), 4 mm wide,
mirror polished. Ar ÷ ions were produced in a radio frequency source and mass
analyzed by a switching magnet. The beam was 0.5 mm in diameter and its
angular dispersion was 0.1 °. Electrons were energy analyzed with a cylindrical
mirror analyzer (CMA) designed for angular measurements~ particularly near
the incident beam direction. The energy resolution was 1.5%. Fig. 1 shows the
experimental arrangement. The ion beam forms an angle of 42.3 ° with respect
to the spectrometer axis and the angle between this axis and the acceptance
cone of the analyzer is also 42.3 ° . The analyzer could be rotated 180 ° around
its symmetry axis, under this condition the axis of the acceptance cone
describes a cone with a semi-aperture of 42.3 °. This disposition allows the
observation angle to be varied from 0 ° to 84.6 ° with respect to the incident
beam direction. The angular acceptance is 3.2 ° .
The sample could be rotated around the analyzer axis changing the beam
incidence angle from 90 ° to 47.7 ° with respect to the sample normal.
Fig. 2a shows the Si Auger spectrum taken at a projectile incidence angle of
47.7 ° and an electron emission angle of 55°; the angles are measured with
respect to the sample normal. This spectrum is quite similar to spectra already
L177
L. de Ferrariis et al. / Auger spectrum of Si
SAMPLE
6o~
/ ~2.3"C°NE/
,
INCIDENT
/
/
Fig. 1. Schematic diagram of the experimental set-up.
published [1,6,8]. Three atomic-like peaks, A, B and D, are clearly observed
superimposed on the broad structure associated with L2,3W transitions. Two
small shoulders, C and E, are detected on the high and low energy edge of the
main atomic-like peak, respectively. The structure F at 108.8 eV is normally
assigned to Auger transitions in Si atoms with a double L2, 3 vacancy.
We have searched for the existence of an atomic-like peak in the region of
the shoulder E, by measuring Auger spectra at an incidence angle of 47.7 ° and
varying the observed electron ejection angle in the angular region 470-90 ° .
Figs. 2a and 2b show two limiting cases of Si Auger spectra taken under
different emission angles, 55 ° and 86 ° respectively.
We can observe in these figures a strong variation of the relative intensities
of band-like and atomic-like components, but the absolute intensities of peaks
B and D and their width at half maximum ( F W H M ~ 4.4 eV) remain independent of the emission angle (withing experimental error) in full agreement with
ST.
The band-like contribution decreases with increasing emission angles due to
the small attenuation lengths of Auger electrons in solids. Electrons which
escape from the solid at angles close to the surface should suffer, on the
average, more inelastic collisions than those which escape near the surface
normal. Additionally they are reflected at the surface when their velocity in the
direction of the surface normal is not enough to traverse the surface barrier.
L178
50
L. de Ferrariis et al. / Auger spectrum of Si
--
I
Oi:
I
D
47.7 deg
0o= 86
deg
25
I
I
I
.~"
I
--
b)
.
E
G)
i--
~1oo
0i: &77 deg
0o: 55 deg
B
A
75 2 : _
""
ol
•
I L,.
I
,~,:,,~,.,~.
C"
-,:"
,.-'N>~-: '~
..
.E
LLI
z
'.I
50
25
0
•
I
I
50
70
"J'x..
F
--
I
90
110
ELECTRON ENERGY (eV)
Fig. 2. Energy distributions of electrons ejected from Si under 20 keV Ar + bombardment. The
incidence angle was 47.7 ° and the observed ejection angle was 55 ° (a), and 86 ° (b), with respect to
the sample normal. The energy resolution was 1.5%. All other experimental conditions were the
same for both spectra. Notice that in (b) the structure E appears as a well-defined shoulder and
that the intensity of the bulk structure diminishes considerably while atomic-like peaks remain
almost unchanged. (Notice that both spectra have the same vertical scale.)
The isotropic behaviour of the narrow peaks confirms previous assumptions
that they correspond to Auger electrons emitted in the vacuum by the
sputtered target atoms [2-4,6-8]. We can also see in fig. 2b that feature E
becomes clearer than in the previous spectrum, thus behaving like an atomic-like
peak, in contrast with the behaviour of the broad structure.
With this encouraging result we measured similar spectra but bombarding
the sample at grazing incidence. In this condition, the atomic collision cascade
produced by the projectile should develop closer to the surface. Thus we expect
that the atomic-like peaks will predominate even more over the band-like
component.
Fig. 3 shows the Si Auger spectrum taken with a projectile incidence angle
of 86 ° and an electron emission angle of 78 ° (16 ° with respect to the incident
L. de Ferrariis et aL / Auger spectrum of Si
I
I
L179
I
D
10 O i = 86 deg.
eo= 78 deg.
8
-
i-.
C
Z
"
I
I
>r~
r~
B
A
6
E
-
o¢"
<
uJ
Z
•
.y?.,.,
I
50
..~
.~., ,..:,,.
:'.~ ..,.
I
I
70
90
ELECTRON ENERGY (eV)
I
110
Fig. 3. Energy distribution of electrons ejected from Si under 20 keV Ar + bombardment with an
incidence angle of 86 ° and an observation angle of 78 ° . The bulk structure almost disappeared and
atomic lines became narrower than in fig. 1. Structures C and E are now well-defined atomic-like
peaks.
b e a m direction). In this c o n d i t i o n of a l m o s t grazing o b s e r v a t i o n the a p p a r e n t
size of the b e a m spot is 1.5 mm, which does not c h a n g e the a n a l y z e r
p e r f o r m a n c e . It is observed that the b r o a d feature has a l m o s t d i s a p p e a r e d ,
though n o t completely, while features C a n d E a p p e a r as well-defined a t o m i c like peaks. W e can also notice that a t o m i c - l i k e p e a k s b e c a m e n a r r o w e r than in
previous spectra ( F W H M 2.7 eV).
A t grazing incidence a l m o s t all the projectiles are reflected at the surface [9],
d i m i n i s h i n g the total A u g e r electron yield. In the ideal c o n d i t i o n s of a t o m i c a l l y
flat surfaces all the projectiles are reflected, a n d excitation collisions, which
a c c o r d i n g to the electron p r o m o t i o n m o d e l require small distances of closest
a p p r o a c h , are not possible. T h e ejection of s p u t t e r e d recoils, excited in a
p r o j e c t i l e - t a r g e t collision, m a y be possible due to the presence of a t o m i c steps
in the surface. T o excite a Si 2p electron, the d i s t a n c e of closest a p p r o a c h m u s t
be less than 0.65 ,~ a n d the c o r r e s p o n d i n g m o s t p r o b a b l e scattering angle of
L180
L. de Ferrariis et a L / Auger ,spectrum of Si
the recoils [10] is less than 13 ° with respect to the surface normal. Therefore
the D o p p l e r b r o a d e n i n g [3] for observation angles close to the surface should
be small, as observed.
Finally we must p o i n t out that peak A has a behaviour different from that
of the other atomic-like peaks, it is broader than B, C, D a n d shifts to lower
energies when the emission angle is close to the surface (see figs. 2b a n d 3). We
have no e x p l a n a t i o n for this effect.
In conclusion, using angle-resolved electron energy d i s t r i b u t i o n measurem e n t s c o m b i n e d with b o m b a r d i n g the sample at grazing incidence, we observed
Auger spectra of Si which resemble those from gas-phase atomic collisions a n d
present conclusive evidence of the existence of an atomic-like peak as postulated by W T with almost the same energy as the high energy edge of the
L2.NV transition.
References
[11 L. Viel, C. Benazeth and N. Benazeth~Surface Sci. 54 (1976) 635.
[2] K. Wittmaack, Surface Sci. 85 (1979) 69.
[3] R.A. Baragiola, Springer Ser. Chem. Phys. 17 (1981) 38:
R.A. Baragiola, E.V. Alonso and H.J.L. Raiti, Phys. Rev. A25 (1982) 1969.
[4] W.A. Metz, K.O. Legg and E.W. Thomas, J. Appl. Phys. 51 (1980) 2888.
[5] J. Mischler, N. Benazeth, M, Negre and C. Benazeth, Surface Sci. 136 (1984) 532.
[6] K. Saiki and S. Tanaka, Nucl. Instr. Methods B2 (1984) 512.
[7] G. Zampieri and R.A. Baragiola, Phys. Rev. B29 (1984) 1480.
[8] R. Whaley and E.W. Thomas, J. Appl. Phys. 56 (1984) 1505.
[9] J.P. Biersack and W. Eckstein, Appl. Phys. A34 (1984) 73.
[10] M.T. Robinson, Rept. ORNL-4556 (Oak Ridge National Laboratory, 1970).