Selenium adsorption on Cs-covered Si(100) 2 × 1 surfaces

Article abstract of DOI:10.1016/j.ssc.2003.12.009

This report involves the study of Se adsorption on caesiated Si(100) 2 × 1 surfaces in ultra high vacuum (UHV) using low energy electron diffraction, Auger electron spectroscopy, thermal desorption spectroscopy and work function measurements. Selenium atoms on Cs/Si(100) 2 × 1 surface adsorb initially on uncaesiated portions of Si and subsequently on the Cs overlayer. The presence of Se increases the binding energy of Cs on Si(100). For Cs and Se coverages above 0.5 ml CsSe and Cs_xSe _ySi_z, compound formation was observed. The coadsorption of Se and Cs induces a high degree of surface disorder, while desorption most probably causes surface etching. The presence of Cs on Si(100) 2 × 1 surfaces prevents the diffusion of Se into the Si substrate and greatly suppresses the formation of SiSe₂ and SiSe₃, detected when Se is adsorbed on clean Si(100) 2 × 1 surfaces.

Full text of DOI:10.1016/j.ssc.2003.12.009

Solid State Communications 129 (2004) 637–641
www.elsevier.com/locate/ssc
Selenium adsorption on Cs-covered Si(100) 2 £ 1 surfaces
*
A.K. Sotiropoulos, M. Kamaratos
Department of Physics, University of Ioannina, P.O. Box 1186, 451 10 Ioannina, Greece
Received 28 November 2003; accepted 7 December 2003 by E.L. Ivchenko
Abstract
This report involves the study of Se adsorption on caesiated Si(100) 2 £ 1 surfaces in ultra high vacuum (UHV) using low
energy electron diffraction, Auger electron spectroscopy, thermal desorption spectroscopy and work function measurements.
Selenium atoms on Cs/Si(100) 2 £ 1 surface adsorb initially on uncaesiated portions of Si and subsequently on the Cs overlayer.
The presence of Se increases the binding energy of Cs on Si(100). For Cs and Se coverages above 0.5 ml CsSe and Cs Se Si ,
x
y
z
compound formation was observed. The coadsorption of Se and Cs induces a high degree of surface disorder, while desorption
most probably causes surface etching. The presence of Cs on Si(100) 2 £ 1 surfaces prevents the diffusion of Se into the Si
substrate and greatly suppresses the formation of SiSe
q 2003 Elsevier Ltd. All rights reserved.
2
and SiSe
3
, detected when Se is adsorbed on clean Si(100) 2 £ 1 surfaces.
PACS: 71.20Dg; 73.30 þ y
Keywords: A. Alkali; A. Selenium; A. Silicon; D. Adsorption; E. Auger electron spectroscopy (AES); E. Low energy diffraction (LEED);
E. Thermal desorption spectroscopy (TDS); E. Work function
1
. Introduction
The coadsorption of Cs with S on Si(100) 2 £ 1 and Cs with
Se on Si(111) 7 £ 7 surfaces have been performed very
recently [10,14]. It is obvious that more scientific effort is
necessary to understand the mutual effect of Se and alkali
coadsorbates on Si surfaces.
The properties of alkali metals on metallic and
semiconducting surfaces and their interaction with electro-
negative elements such as oxygen have been the subject of
numerous investigations [1–7]. In contrast to the case of
oxygen, the coadsorption of alkali metals with the other
elements in group VI on solid surfaces is very scarce
In this investigation we study the adsorption of Se on
caesiated Si(100) 2 £ 1 surfaces at room temperature. Data
were acquired using the surface analysis techniques of low
energy electron diffraction (LEED), Auger electron spec-
troscopy (AES), Thermal desorption spectroscopy (TDS)
and work function (WF) measurements.
[
8–14]. The presence of alkali metals on metallic and
semiconducting surfaces increases drastically the sticking
coefficient and the maximum amount of oxygen on the
surface and enhances its oxidation [2,4–7]. On the other
hand, the presence of alkali metals does not change the
sticking coefficient of S on Ni(100) [8,9,12].
2
. Experimental
Semiconductor compounds of Se have been receiving
increased attention lately, due to their unique optical,
structural and electrical properties [15,16]. Deposition
studies of Se and compounds such as ZnSe, InSe and
GaSe have been performed on surfaces of Si [15,17–19].
The experiments were carried out in an ultra high
10
2
vacuum (UHV) champer with base pressure ,10
The champer is equipped with a four grid LEED system, a
mbar.
hemicylindrical mirror analyzer (CMA) with a resolution of
0
.3 eV for AES, a quadrupole mass spectrometer (UTI) for
*
Corresponding author. Tel.: þ30-265-109-8453; fax: þ30-265-
09-8694.
E-mail address: mkamarat@cc.uoi.gr (M. Kamaratos).
1
TDS an electron gun for WF measurements using the diode
method and facilities for Ar ion bombardment.
0038-1098/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ssc.2003.12.009

638
A.K. Sotiropoulos, M. Kamaratos / Solid State Communications 129 (2004) 637–641
The Si(100) 2 £ 1 sample was cleaned by heating at
300 K. The temperature of the sample was measured by a
sition on Cs-covered Si(100) and Si(111) surface is that the
1
Si (91 eV) Ap-pH decreases and almost disappears at high
Se doses on Si(111) surface, whereas on Si(100) surface it
levels off at a value, which is 25% of its initial value (clean
surface) independent of Cs precoverage. The stability of Cs
(563 eV) Ap-pH at low Se coverages indicates that the Se
adsorption sites are different than that of Cs and there would
be no masking of the Cs adatoms. At high Se doses (.10)
Se begins to cover the Cs adatoms, for the Cs (1 ml)/Si(100)
surface system and the Cs (563 eV) Ap-pH decreases.
Fig. 2 shows the change of the Cs (47 eV) Ap-pH versus
Se deposition on Si(100) 2 £ 1 covered with 0.5 and 1.0 ml
of Cs at room temperature. On the same figure we also show
the Se (45 eV) Ap-pH versus Se deposition on clean Si(100)
surface. Unfortunately, the Se (45 eV) Auger peak does not
resolve from the Cs (47 eV) Auger peak. The Se (45 eV)
Ap-pH during Se deposition on clean Si(100) 2 £ 1 surface
increases with a continuous lower rate as the Se doses
increase in agreement with previous results [17]. Selenium
deposition on Cs-covered Si(100) surfaces causes a slight
increase of the (45–47 eV) Ap-pH up to 0.5 ml of Se
deposition. Higher Selenium deposition leads to a continu-
ous decrease of the (45–47 eV) Auger peak. The initial
increase of the (45–47 eV) Auger peak height as the Se
adsorption increases indicates that the Se adsorbs on
different adsorption sites and the intensity of the two
peaks is added. This is in agreement with previous data of Se
on Cs-covered Si(111) and Cs on S-covered Si(100)
surfaces [10,14]. The decrease of the Cs (47 eV)–Se
(45 eV) Auger peak height caused by Se deposition above
0.5 ml is attributed to a strong Cs–Se interaction, which
results to a change of the chemical bond of Cs and Se and/or
to a reorganization of the surface. As it was observed
previously, the intensity of the Auger peaks which involves
valence electrons as it is the Cs (47 eV) are substantially
affected by the chemical bond [20].
Cr–Al thermocouple calibrated by an infrared pyrometer in
the range 900–1300 K. The surface was considered clean
when the main impurity of carbon was below the detection
limit of the Auger spectroscopy. The clean surface shows a
very sharp 2 £ 1 reconstructed LEED pattern. Caesium was
evaporated from a commercial dispenser source (SAES
getters). Elemental Se was evaporated by thermal dis-
sociation of WSe
2
single crystal flakes enclosed in
envelopes of thin Ta plates which were heated by passing
the required electrical current. The atomic Se produced by
the source was estimated to be 30% of the total yield [14].
The respective dosages of Cs and Se were calibrated by
analyzing the LEED and Auger spectroscopy data separ-
ately deposited on clean Si(100) 2 £ 1 surface at room
temperature. The surface atomic density of a saturation layer
(
1 ml) of Cs and Se was considered equal to that of the
1
4
outermost layer of Si(100) 2 £ 1 surface, 6.8 £ 10
2
atoms/cm .
3
. Results
Fig. 1 shows the Cs (563 eV) Auger peak-to-peak height
Ap-pH) as a function of Se doses after 0.5 and 1.0 ml of Cs
(
deposition on Si(100) 2 £ 1 surfaces at room temperature.
On the same figure we show the corresponding Si (91 eV)
Ap-pH versus Se deposition. The Si (91 eV) Ap-pH
decreases drastically at low Se doses and levels off at high
Se coverages. On the other hand, the Cs (563 eV) Ap-pH is
constant at low Se deposition and continuous to be constant
at high Se doses for the Cs (0.5 ml)/Si(100) surface, while it
decreases for Cs (1 ml)/Si(100) system. Analogous behavior
has been observed for Se deposition on Cs-covered Si(111)
surfaces [14]. A substantial difference between Se depo-
The 2 £ 1 LEED pattern observed after Cs deposition
disappears for Se coverages, higher than 0.5 ml. This result
Fig. 1. The Cs (563 eV) Ap-pH as a function of Se deposition on
clean and Cs-covered Si(100) surfaces.
Fig. 2. The Cs (47 eV) and Se (45 eV) Ap-pH as a function of Se
doses on clean and Cs-covered Si(100) surfaces.

A.K. Sotiropoulos, M. Kamaratos / Solid State Communications 129 (2004) 637–641
639
indicates that the presence of Se leads to a disorder of the
surface.
Fig. 3 shows the WF change of the Cs-covered Si(100)
observed after oxygen exposure on Cs-covered Si surfaces
[6,21]. This is attributed to different interaction of oxygen
and selenium with the Cs/Si(100) surface system.
2
£ 1 surfaces as a function of Se deposition. On clean
Fig. 4 shows a series of the Cs (133 amu) thermal
desorption spectra after 1 ml of Cs deposition and
subsequent Se adsorption. The presence of Se increases
the binding energy of Cs in agreement with previous results
Si(100), Se deposition causes an increase of the WF, up to a
maximum value near 1 ml, with a subsequent small decrease
in agreement with previous results [17]. Selenium on Cs
covered Si(100) surface causes a drastic increase of the WF
for all Cs coverages. This behavior is the same as that
observed for Se deposition on Cs-covered Si (111) 7 £ 7
surfaces [14], but it differs slightly to that of oxygen on Cs
covered Si surfaces. At high Cs coverages (1 ml), on Si(100)
or GaAs(100) the WF of the surface decreases with oxygen
deposition at low oxygen exposures [6,21]. The difference
of the WF change curve of Se on Cs-covered Si(100) surface
in correlation to that of oxygen on Cs-covered Si(100)
surface is attributed to the different interaction of oxygen
and Se with the Cs-covered Si system. Oxygen atoms
initially submerge under the Cs layer and interacts with Cs
overlayer, whereas the Se on Cs covered Si(100) surface
occupies different adsorption sites, and interacts initially
with Si atoms to form SiSe. Another substantial difference
on the WF curves between oxygen and Se deposition on Cs-
covered Si surfaces is that after Se adsorption the increase of
the WF is so high that the WF turns back, close to its initial
value of the clean Si surface. The same large increase of the
WF has been measured after Se deposition on Cs-covered
Si(111) surface [14]. Such large increase of the WF was not
[
14]. The main desorption peak of Cs after 4 ml of Se
deposition appears at around 1050 K.
Fig. 5 shows a series of SiSe
2
compound thermal
desorption spectra after 2 ml of Se deposition on Si(100)
surface covered with different amounts of Cs. The peak of
SiSe
presence of a very small amount of Cs (0.06 ml) shifts
slightly the SiSe TD peak to lower temperature. Higher Cs
precoverages causes a drastic decrease of intensity of the
SiSe peak which disappears for Cs coverages higher than
.5 ml. This is attributed to the strong Se–[Cs/Si(100)]
2
from Se/Si(100) system appears at ,980 K. The
2
2
0
interaction for Cs coverages higher than 0.5 ml, which leads
to a reorganization of the surface in agreement with the
results of Auger spectroscopy (Fig. 2) and LEED observ-
3
ations. The SiSe thermal desorption peak indicates almost
the same behavior and we do not show it here. Analogous
results have been observed previously for the system Se/Cs/
Si(111) [14]. On the other hand, the SiSe desorption peak is
not affected substantially from Cs precoverage. The above
results indicate that initially, the Se occupies different sites
Fig. 4. The Cs (133 amu) thermal desorption spectra after 1 ml Cs
deposition from Cs/Si(100) 2 £ 1 and Se/Cs/Si(100) surface
systems, at a constant heating rate.
Fig. 3. Work function changes of Se on clean and Cs covered
Si(100) 2 £ 1 surfaces as a function of Se doses.

640
A.K. Sotiropoulos, M. Kamaratos / Solid State Communications 129 (2004) 637–641
same as that of the main peak of Cs (Fig. 4). This means
possibly that the CsSe are fragments of a Si Cs Se
x
y
z
compounds. This is supported by previous results of Se on
Cs-covered Si(111) surfaces [14].
4
. Conclusion
In this investigation we study the structural and chemical
interaction between Se, Cs and Si during Se deposition on
Cs/Si(100) 2 £ 1 surface systems. The investigation took
place in UHV utilizing the surface analysis techniques
LEED, AES, TDS and WF measurements. Adsorption of Se
on Cs/Si(100) surface system results in Se being initially
adsorbed on sites different than that occupied by Cs up to
0
.5 ml of Se deposition and forms SiSe compound. Higher
Se deposition leads to a strong Cs–Se and Si–Cs–Se
interaction, which leads to a disorder of the surface. The
CsSe and Si–Se–Cs compounds are removed from the
surface at about 1050 K. The presence of Se increases
the binding energy of Cs on Si(100) 2 £ 1 surface, whereas
the binding energy of Se remains approximately equal to that
on the clean Si(100) surface. The presence of Cs prevents
the diffusion of Se into the Si substrate and suppresses
2
Fig. 5. This figure shows the SiSe thermal desorption spectra
desorbed from Se/Si(100) and Se/Cs/Si(100) surfaces, at a constant
heating rate, from 300 to 1300 K.
2 3
substantially the formation of SiSe and SiSe .
from those of Cs and forms SiSe up to a Se coverage of
0
.5 ml. Higher Se deposition leads to a strong Se–Cs–Si
interaction which, prevents the diffusion of Se into the Si
substrate and the formation of SiSe or SiSe compounds.
Fig. 6 shows the CsSe and Si CsSe compounds thermal
References
2
3
[1] C.A. Papageorgopoulos, J.M. Chen, J. Vac. Sci. Technol. 9
1971) 570.
2
(
desorption peaks after Se deposition on Cs covered Si(100)
surfaces. These compounds appear at about 1050 K for Se
and Cs coverages higher than 0.5 ml. This temperature is the
[
[
[
2] C.A. Papageorgopoulos, J.M. Chen, Surf. Sci. 52 (1975) 40
see also pp. 52.
3] M. Kiskinova, L. Surnev, G. Blisnakov, Surf. Sci. 104 (1981)
2
40.
4] K. M u¨ ller, G. Besold, K. Heinz, in: H.P. Bonzel, A.M.
Bradshaw, G. Ertl (Eds.), In: Alkali Metal Adsorption on
Metals and Semiconductors, Elsevier, Amsterdam, 1980.
5] H.I. Starnberg, P. Soukiassian, Z. Hurych, Phys. Rev. B39
[
[
[
[
[
(
1989) 12775.
6] G. Boishin, M. Tikhov, M. Kiskinova, L. Surnev, Surf. Sci.
61 (1992) 224.
2
7] J. Higasa, F.H. Geuzebroek, Y. Yamashita, A. Namiki, Surf.
Sci. 295 (1993) L55.
8] C.A. Papageorgopoulos, M. Kamaratos, A.C. Papageorgopou-
los, Surf. Sci. 433 (1999) 806.
9] A.C. Papageorgopoulos, M. Kamaratos, J. Phys.: Condens.
Matter 12 (2000) 9281.
[
[
[
[
10] A.C. Papageorgopoulos, D. Mosby, C.A. Papageorgopoulos,
Surf. Rev. Lett. 5 (1998) 85.
11] C.A. Papageorgopoulos, M. Kamaratos, A.C. Papageorgopou-
los, Surf. Sci. 481 (2001) 143.
12] C.A. Papageorgopoulos, M. Kamaratos, A.C. Papageorgopou-
los, Surf. Sci. 402 (1998) 120.
13] S. Anderson, J.P. Pendry, J. Phys. C. Solid State Phys. 9 (1976)
2
721.
Fig. 6. This figure depicts the TDS of Si
2
CsSe and CsSe compounds,
[14] A.C. Papageorgopoulos, M. Kamaratos, J. Phys.: Condens.
Matter 14 (2002) 5255.
desorbed from Se/Cs/Si(100) surfaces, at a constant heating rate.

A.K. Sotiropoulos, M. Kamaratos / Solid State Communications 129 (2004) 637–641
641
[
[
15] F. Proix, V. Panella, S. Monkad, A. Glebor, J.P. Lacharme, M.
Eddriel, K. Aminer, C.A. Sebenne, J.P. Toennies, J. Eur. Phys.
B5 (1998) 919.
[17] A.C.Papageorgopoulos,M.Kamaratos,Surf.Sci.466(2000)173.
[18] R.D. Bringans, M.A. Olmstead, Phys. Rev. B39 (1989) 12785.
[19] M. Shuang, B.R. Schroeder, M.A. Olmstead, Phys. Rev. B61
(2000) 7215.
16] R. Tuttle, J.S. Ward, T. Duda, T.A. Berens, M.A. Contreras,
K.R. Ranganathan, A.L. Tennant, J. Keane, K. Emery, R.
Nouf, MRS Proc., vol. 426, Materials Research Society,
Pittsburgh, PA, 1996, p. 143.
[20] R. Weissman, K. M u¨ ller, Surf. Sci. Rep. 105 (1981) 251.
[21] M. Kamaratos, S. Kennou, S. Ladas, C.A. Papageorgopoulos,
J. Phys.: Condens. Matter 1 (1989) 6071.