Absolute yields of the exciton induced desorption at the surface of solid rare gases
Absolute yields of the photo-induced desorption at the surface of solid rare gases were studied in the excitonic excitation region. Both metastable and total desorption yields depend strongly on excitation energy and film thickness of rare gas solids. The absolute desorption yields and their depende...
Gespeichert in:
| Datum: | 2003 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
2003
|
| Schriftenreihe: | Физика низких температур |
| Schlagworte: | |
| Online Zugang: | http://dspace.nbuv.gov.ua/handle/123456789/128821 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Zitieren: | Absolute yields of the exciton induced desorption at the surface of solid rare gases / I. Arakawa, T. Adachi, T. Hirayama, M. Sakurai // Физика низких температур. — 2003. — Т. 29, № 3. — С. 342-350. — Бібліогр.: 32 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-128821 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1288212018-01-15T03:03:59Z Absolute yields of the exciton induced desorption at the surface of solid rare gases Arakawa, I. Adachi, T. Hirayama, T. Sakurai, M. Electronically Induced Phenomena: Low Temperature Aspects Absolute yields of the photo-induced desorption at the surface of solid rare gases were studied in the excitonic excitation region. Both metastable and total desorption yields depend strongly on excitation energy and film thickness of rare gas solids. The absolute desorption yields and their dependence on film thickness were quantitatively reproduced by a simulation based on the diffusion of excitons in the bulk and the kinetic energy release by a cavity ejection mechanism and an excimer dissociation one followed by internal sputtering. 2003 Article Absolute yields of the exciton induced desorption at the surface of solid rare gases / I. Arakawa, T. Adachi, T. Hirayama, M. Sakurai // Физика низких температур. — 2003. — Т. 29, № 3. — С. 342-350. — Бібліогр.: 32 назв. — англ. 0132-6414 PACS: 71.35.-y, 79.20.La http://dspace.nbuv.gov.ua/handle/123456789/128821 en Физика низких температур Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
Electronically Induced Phenomena: Low Temperature Aspects Electronically Induced Phenomena: Low Temperature Aspects |
| spellingShingle |
Electronically Induced Phenomena: Low Temperature Aspects Electronically Induced Phenomena: Low Temperature Aspects Arakawa, I. Adachi, T. Hirayama, T. Sakurai, M. Absolute yields of the exciton induced desorption at the surface of solid rare gases Физика низких температур |
| description |
Absolute yields of the photo-induced desorption at the surface of solid rare gases were studied in the excitonic excitation region. Both metastable and total desorption yields depend strongly on excitation energy and film thickness of rare gas solids. The absolute desorption yields and their dependence on film thickness were quantitatively reproduced by a simulation based on the diffusion of excitons in the bulk and the kinetic energy release by a cavity ejection mechanism and an excimer dissociation one followed by internal sputtering. |
| format |
Article |
| author |
Arakawa, I. Adachi, T. Hirayama, T. Sakurai, M. |
| author_facet |
Arakawa, I. Adachi, T. Hirayama, T. Sakurai, M. |
| author_sort |
Arakawa, I. |
| title |
Absolute yields of the exciton induced desorption at the surface of solid rare gases |
| title_short |
Absolute yields of the exciton induced desorption at the surface of solid rare gases |
| title_full |
Absolute yields of the exciton induced desorption at the surface of solid rare gases |
| title_fullStr |
Absolute yields of the exciton induced desorption at the surface of solid rare gases |
| title_full_unstemmed |
Absolute yields of the exciton induced desorption at the surface of solid rare gases |
| title_sort |
absolute yields of the exciton induced desorption at the surface of solid rare gases |
| publisher |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| publishDate |
2003 |
| topic_facet |
Electronically Induced Phenomena: Low Temperature Aspects |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/128821 |
| citation_txt |
Absolute yields of the exciton induced desorption at the surface of solid rare gases / I. Arakawa, T. Adachi, T. Hirayama, M. Sakurai // Физика низких температур. — 2003. — Т. 29, № 3. — С. 342-350. — Бібліогр.: 32 назв. — англ. |
| series |
Физика низких температур |
| work_keys_str_mv |
AT arakawai absoluteyieldsoftheexcitoninduceddesorptionatthesurfaceofsolidraregases AT adachit absoluteyieldsoftheexcitoninduceddesorptionatthesurfaceofsolidraregases AT hirayamat absoluteyieldsoftheexcitoninduceddesorptionatthesurfaceofsolidraregases AT sakuraim absoluteyieldsoftheexcitoninduceddesorptionatthesurfaceofsolidraregases |
| first_indexed |
2025-07-09T09:58:34Z |
| last_indexed |
2025-07-09T09:58:34Z |
| _version_ |
1837162948195254272 |
| fulltext |
Fizika Nizkikh Temperatur, 2003, v. 29, No. 3, p. 342–350
Absolute yields of the exciton induced desorption at the
surface of solid rare gases
I. Arakawa and T. Adachi
Department of Physics, Gakushuin University, Mejiro, Tokyo 171-8588, Japan
E-mail: ich.arakawa@gakushuin.ac.jp
T. Hirayama
Department of Physics, Rikkyo University, Ikebukuro, Tokyo 171-8501, Japan
M. Sakurai
Department of Physics, Kobe University, Nada, Kobe 657-8501, Japan
Received December 10, 2002
Absolute yields of the photo-induced desorption at the surface of solid rare gases were studied
in the excitonic excitation region. Both metastable and total desorption yields depend strongly on
excitation energy and film thickness of rare gas solids. The absolute desorption yields and their de-
pendence on film thickness were quantitatively reproduced by a simulation based on the diffusion
of excitons in the bulk and the kinetic energy release by a cavity ejection mechanism and an
excimer dissociation one followed by internal sputtering.
PACS: 71.35.–y, 79.20.La
1. Introduction
When the surface of a solid rare gas is irradiated
with vacuum ultraviolet light, it can eject atoms and
clusters in ground, electronically and vibrationally
excited, or ionized states. A variety of desorption
mechanisms have been proposed for each desorbed spe-
cies and for the excitation energy region. Two types of
mechanisms are known to play an essential role in the
desorption induced by the excitonic excitation; a cav-
ity ejection (CE) mechanism is due to the repulsive
interaction between the excited atom with an inflated
electron cloud and the matrix with a negative electron
affinity and an excimer dissociation (ED) one is the
dissociation of an excimer in the vicinity of the sur-
face. The investigation of the desorption phenomena
initiated by the excitonic excitation is one of the most
powerful tools to reveal the dynamic characters of the
exciton in rare gas solids [1]. The desorption of ex-
cited neutral particles of Ne and Ar has extensively
been investigated for two decades [2]; the close rela-
tionship between exciton formation [3] and the detail
of the desorption mechanisms [4] has been almost
fully elucidated. On the other hand, the desorption
mechanism of the ground state neutral, as well as its
desorption yield, has not been clarified yet though it
can safely be said that they are the main component in
the desorbed species.
We report the measured results on absolute yields
of the metastable desorption and the total one which
are induced by the excitonic excitation from the sur-
faces of solid Ne, Ar, and Kr. We also show that the
experimental results for Ne and Kr are satisfactorily
reproduced by the simple model simulation of the
desorption process.
2. Desorption mechanism
The negative electron affinity of the matrix is
known to be essential for the CE process to have a re-
pulsive interaction between the excited atom and the
surrounding ground state atoms. The metastables
desorbed through the CE process were observed for
the solid whose electron affinity EA is negative,
namely, solid Ne (EA = –1.4 eV) and Ar (–0.4 eV),
but not for solid Kr (0.3 eV) and Xe (0.5 eV) [5]. The
kinetic energy Ek of the metastable desorbed through
© I. Arakawa, T. Adachi, T. Hirayama, and M. Sakurai, 2003
the CE process is originated from the lattice distortion
energy around the exciton, which can roughly be esti-
mated from the difference between the excitation en-
ergy Ex of the exciton and the excitation energy Eg of
the corresponding state of an isolated atom in the gas
phase [1]. The energy difference, Ex – Eg, is divided
among three terms, Ek: the kinetic energy of a de-
sorbing particle, Ecoh: the cohesive energy of an atom
on the surface, and Elatt: the energy absorbed in the
lattice [6]. It has been revealed that 30–70 % of the
lattice distortion energy is transferred into the kinetic
energy of the desorbing metastable from the surface
exciton S1 of Ne and Ar [4]. It has been also experi-
mentally studied by the molecular dynamic calcula-
tion [7]. In the case of Ne, for example, Ek of the
metastable which is desorbed by the first order surface
exciton is 0.18 eV at the peak of the distribution,
while the excitation energy Ex is (17.17±0.03) eV
and Eg of the corresponding 2p53s states are
16.619–16.848 eV [8]. The excitation energy Ex of the
bulk exciton B1 of Ne is (17.57±0.03) eV. Con-
sidering the low cohesion energy, 0.019 eV, of Ne, the
lattice distortion energy, Ex – Eg, of 0.9–0.7 eV is
large enough to squeeze the excited atom out of a
second or third underlying layer to vacuum. The
squeezed excited atom may push overlying atoms for-
ward. This internal sputtering process can result in a
large total desorption yield of Ne. In solid Ar, though
the lattice distortion energy is estimated at about
0.7–0.4 eV, the internal sputtering process must be
less efficient because of its larger cohesion energy of
0.068 eV.
Dissociative relaxation of a rare gas excimer in the
gas phase yields the kinetic energy of about 1 eV [9].
It is known that the ED process in solid produces crys-
tal defects [10]. If ED occurs in the surface vicinity of
a rare gas solid, a large number of atoms in the surface
layer will be released into vacuum by the internal
sputtering process.
3. Experimental
The experimental apparatus for the photostimu-
lated desorption (PSD) study was settled at BL5B
beam line in the UVSOR facilities of the Institute for
Molecular Science, Okazaki, Japan [11–13]. A solid
rare gas film was prepared on a platinum substrate at a
temperature of about 6 K which was attached to a liq-
uid helium cryostat in an ultra-high vacuum chamber
with a base pressure below 10–8 Pa. The amount of the
gas condensed on the substrate, the film thickness,
was calculated from the exposure, which was the
product of the pressure and the duration of deposition,
assuming the condensation coefficient to be unity.
The wavelength resolution, � ��/ , of the vacuum
ultraviolet light source was about 500 ± 200 in the
range of � between 20 and 100 nm. The light intensity
was monitored by the photoelectron current from a
gold plated mesh, through which the light beam was
introduced onto the sample surface. We adopted the
value of 0.07 electrons/photon as the photoelectron
yield at gold surface, which is the average value of the
published data by Samson [14] between 52 and
100 nm of wavelength. The intensity of photon flux
ranged between 1011 and 1012 photons/s in continu-
ous mode and between 105 and 106 photons/pulse in
pulse mode within a diameter of 3 mm at the sample.
3.1. Metastable desorption measurement
The monochromatized photon beam is pulsed by a
mechanical chopper in order to measure the
time-of-flight (TOF) spectra of the desorbed
metastable atoms. The width and the interval of the
pulsed beam are 15 µs and 2.5 ms, respectively. The
photon beam is incident at 20 deg of the normal direc-
tion of the sample surface.
The desorbed metastable atoms are detected by an
open electron multiplier tube (EMT, Hamamatsu,
R595) with a CuBe dynode as a first electrode. The
EMT is fixed at a distance of 360 mm from the sample
in the normal direction of the sample. The diameter of
the entrance of EMT is 8 mm, which corresponds to
the detection solid angle of 3.1·10–5 sr. The lifetimes
of the metastable (3P0,2) rare gas atoms are known to
be much longer than the flight time of the detected
atoms which is less than 1 ms in our experimental sys-
tem. The charged particles are rejected by applying
suitable voltages to the grid mesh in front of the de-
tector and the 1st dynode of EMT. In order to cal-
culate the overall metastable yields, which were in-
tegrated for all direction in half space, we utilized
the angular distribution of the desorbed metastables
which have already been reported by our group
[15–17].
3.2. Total desorption measurement
The total desorption rate was calculated from the
pumping speed and the rise of the partial pressure in
the vacuum chamber during irradiation of the sample.
The pumping speed of the turbo molecular pump and
cold surfaces for desorbed rare gas was determined
from the pressure measured by an extractor gauge in-
stalled in the chamber and from the flow rate cali-
brated volumetrically using a reference volume and a
Baratron pressure gauge as a reference. A little rise of
the partial pressure during irradiation was detected by
a quadrupole mass spectrometer which was calibrated
against the extractor gauge. It should be noted that
Absolute yields of the exciton induced desorption at the surface of solid rare gases
Fizika Nizkikh Temperatur, 2003, v. 29, No. 3 343
the uncertainty of the relative sensitivity of the ex-
tractor gauge was cancelled in the present method of
determining the desorption rate.
3.3. Accuracy of measurements
In the metastable yield determination, the uncer-
tainty of the results was estimated to be ± 30% by a
quadrature sum of all sources of error; light intensity
measurement, detection efficiency of the photomul-
tiplier, geometrical configuration, etc. The main part
of the error came from the estimation of the amount of
higher order light from the monochromator [12,18].
The uncertainty in determining the absolute yield
of the total desorption was estimated as large as ±1/3
of the order of magnitude; an error bar ranging from
–50% to +100% of the obtained value should be added
to the obtained values. In addition to the uncertainty
in the intensity measurement of the incident photon, it
was caused by difficulties in determining the sensiti-
vity of the mass spectrometer and the pumping speed
for the desorbed species.
Noticeable changes in the desorption yields were
observed after a few hours exposure of the sample
in the chamber at a pressure of lower 10–8 Pa range,
where the main component was a rare gas itself. It was
probably due to the adsorption of the common resi-
dual gases, H2, H2O, CO, in the UHV system [19,20].
The uncertainty caused by this impurity effect is
within the above mentioned error bar in the present
experiment.
4. Desorption of Ne
4.1. Experimental results
The desorption yields of Ne metastable in the
excitonic excitation region is shown in Fig. 1 as a
function of incident wavelength. Most of the observed
peaks in the figure can be assigned to a series of the
bulk (B) and the surface (S) excitons reported by
Saile and Koch [21], where S‘ peak is caused by the
exciton in the 2p53p state which is allowed at the sur-
face because of the reduced symmetry [22]. The four
TOF spectra of the Ne metastable in Fig. 2 were ob-
tained by the excitations at S1, B1, S’, and B2
excitons.
The wavelength dependence of the absolute yield of
the total photo-desorption in the range between 52
and 77 nm is shown in Fig. 3 for a film thickness of 73
atomic layers, where the peaks caused by the exciton
excitation are also clearly seen. The absolute value of
the desorption yield in the figure represents the aver-
age number of the Ne atoms desorbed by one incident
photon. The large and continuous background was due
to the desorption induced by ionization by the second
344 Fizika Nizkikh Temperatur, 2003, v. 29, No. 3
I. Arakawa, T. Adachi, T. Hirayama, and M. Sakurai
7570656055
Wavelength, nm
S1
B1
S'
B2
Eg
N
e
In
te
n
si
ty
,
ar
b
. u
n
its
*
Fig. 1. The desorption yields of the Ne metastable in the
excitonic excitation region as a function of incident wave-
length.
8006004002000
Flight time, µs
B1
S'
B2
Photon
B
A
C
D
S1
In
te
n
si
ty
,
ar
b
.u
n
its
Fig. 2. The time-of-flight spectra of the metastable Ne
desorbed by the excitation of the four types of excitons.
2.0
1.5
1.0
0.5
0.0D
es
o
rp
t i
o
n
y
ie
ld
,a
to
m
s/
ph
ot
on
75706560
Wavelength, nm
S1
B1
S'
B2
B3
73 atomic layers
Fig. 3. The absolute yields of the total desorption of the Ne
in the excitonic excitation region as a function of incident
wavelength.
or higher order light from the monochromator. The
peaks caused by the bulk excitons, B1 and B2, develop
at the film thickness of a few tens atomic layers and
the absolute yields seem to be saturated at around 1.6
and 1 atoms/photon, respectively, for the film
thicker than 100 atomic layers. The peak heights due
to the surface excitons seem to keep constant values of
about 0.3 and 0.1 atoms/photon for S1 and S’, respec-
tively, over the entire thickness range as expected for
the surface excitations. The absolute yields of the
metastable and the total desorption by the four differ-
ent excitonic excitations are summarized in Table.
4.2. Schematic model of the desorption process
The mechanism which works in each desorption
process and the branching ratio of the relaxation cas-
cade which leads to the desorption can be deduced
from a careful examination of the TOF spectra of the
metastables and the absolute yields of the metastable
and the total desorption. The mechanisms which work
in the desorption process are also listed in Table.
The higher kinetic energy peak A (Ek = (1.4±
±0.1) eV) [23] in Fig. 2, which is only observed by
the excitation of the higher order excitons (S‘ and
B2), is caused by the excimer dissociation of the
highly excited dimer: Ne2** � Ne + Ne* + Ek. This is
also the case for Ar [24]. The lower kinetic energy
peak, that is CE one, is obviously composed of two
components B (Ek = (0.21±0.02) eV) and C (Ek =
= (0.18±0.02) eV) [12]. The metastables which com-
pose the peaks B and C are ejected directly from the
surface excitons S‘ and S1, respectively [15]. The dif-
ference in the kinetic energy is caused by the differ-
ence in the magnitude of the repulsive interaction.
The low energy tail D is only observed by the bulk ex-
citations (B1 and B2) [15] and is likely caused by the
bulk exciton trapped in the underlying layer just be-
low the surface. They are ejected into the vacuum,
loosing their kinetic energy by collision with atoms in
the overlayer. These TOF spectra can be used as a fin-
gerprint to identify each exciton. These spectra also
show clearly the relaxation channel of the excitons
Absolute yields of the exciton induced desorption at the surface of solid rare gases
Fizika Nizkikh Temperatur, 2003, v. 29, No. 3 345
Table
The absolute yields of the metastable and the total desorption from solid Ne, Ar, and Kr
Rare gases
Initial
excitation
Desorbed species Yield, atoms/photon Desorption mechanisms
Ne
S1
metastable (2.3 ± 0.7)·10–3 CE
total 0.3 CE + S/CE(?%) + S/ED(?%)
S’
metastable (7.8 ± 2.3)·10–4 CE + ED(� 1%)
total 0.1 CE + S/CE(?%) + S/ED(?%)
B1
metastable (1.4 ± 0.4)·10–3 CE + S/CE(a few %)
total 1.6 ± 0.3 S/CE + CE(� 10%) + S/ED(a few %)
B2
metastable � 1·10–3 CE + S/CE(a few %) + ED(� 1%)
total 1 ± 0.2 S/CE + CE(� 10%) + S/ED(a few %)
Ar
S1
metastable � 1·10–5 CE
total 0.1 CE
S2
metastable not determined CE + ED
total 0.1 CE
B1
metastable not determined CE
total 0.23 CE + S/ED
B2
metastable not determined CE + ED
total 0.16 CE + S/ED
Kr
S1(3/2) total 0.015 S/ED
S1(1/2) total 0.01 S/ED
S2(3/2)
metastable not determined ED
total � 10–3 S/ED
B1(3/2) total 0.03 S/ED
B1(1/2) total 0.02 S/ED
B2(3/2)
metastable not determined ED
total � 10–3 S/ED
from the higher energy state to the lower one; from B1
to S1, from S‘ to S1 and excimer, and from B2 to S’,
S1, and excimer.
The quantum efficiency of the exciton creation can
be estimated from the photo-absorption coefficient of
solid Ne at the wavelength corresponding to the bulk
exciton (B1) excitation. We assume here that the ex-
citation probability of S1 is the same as that of B1 in
each layer at their own excitation energy. The estima-
tion based on the data by Pudewill et al. [25] leads to
the value of 0.1 excitons/photon in each layer.
The relaxation cascade of the first order surface
exciton S1 of solid Ne is schematically shown in
Fig. 4. Incident light of the wavelength of 72.2 nm
will excite surface excitons with the probability of 0.1
excitons/photon in the first layer. 90% of the incident
photons pass through the solid Ne film and will be
scattered or absorbed at the substrate. The surface
exciton may decay radiatively or may be desorbed by
the CE mechanism as a metastable in 3P2,0 state or as
an excited atom in 3P1 or 1P1 state which decay into
the ground state with a short lifetime of 1–10 ns. In
our study, we have determined the absolute yields of
the metastable desorption, 0.0023 Ne*/photon, and
of the total desorption, 0.3 Ne/photon. Considering
the absolute total desorption yield at S1 excitation of
0.3 atoms/photon, the value, 0.1 excitons/photon,
of the initial excitation probability must be underesti-
mated since it is likely that a surface exciton can be
desorbed only a single excited atom in the CE process.
The larger desorption yield may be explained by other
desorption processes: the desorption of a dimer by
the CE process and the ED process which yields two
or more desorbing atoms. However, the branching ra-
tio to these two processes must be far smaller than
that to the single atom desorption by the CE process.
Another possibility is the excitation by the light re-
flected at the substrate surface. If this is the case, the
initial excitation probability doubles at most. In any
case, it can be concluded that almost all the surface
excitons yield the desorption of one Ne atom or, in
other words, that the desorption probability of the
surface excitons S1 is almost unity. The comparison
between the metastable desorption yield and the total
one shows that about 1% of the desorbed Ne atoms are
in the metastable state.
The above-mentioned scheme cannot be applied to
the analysis of the desorption yield caused by S‘ exci-
tation because the quantum efficiency for S’ creation
cannot be estimated in the same manner; the excita-
tion of 2p53p state is only allowed at the surface of the
solid. It can be said from the comparison between the
total desorption yields by S‘ and by S1, if we assume
that the desorption probability of S‘ exciton is also
unity, that the excitation cross section of S‘ is one
third of that of S1. The metastable TOF spectrum by
S‘ excitation shows that almost two thirds of S‘
exciton decays into S1 state at the surface before the
desorption and that a few % of S‘ forms excimers in a
higher excited state. It is also worth noting that the
ratio of the number of the metastable species to the to-
346 Fizika Nizkikh Temperatur, 2003, v. 29, No. 3
I. Arakawa, T. Adachi, T. Hirayama, and M. Sakurai
photon 72.2 nm
~ 0.1
~ 0.9
S1 exciton
radiative decay
3P2,0
3 P1 , 1 P1
Y META
�
�
0.0023 Ne*/photon
Y TOTAL
0.3 Ne/photon
Fig. 4. Schematic chart of the relaxation cascade of the first order surface exciton S1 in solid Ne.
tal number of desorption events is about 1% also in the
case of S‘ excitation.
The relaxation cascade of the first order bulk
exciton B1 of solid Ne are schematically shown in
Fig. 5. When solid Ne is irradiated by 70.5 nm pho-
tons, the probability for the bulk exciton creation
in the second underlying layer is estimated at 0.1 exci-
tons/photon, 0.09 in the third layer, and 0.1(1–0.1)(n–2) in
the nth layer. Radiative decay may occur immediately
or on the way of migration in the bulk. Some of them
may reach the surface and convert into the surface
exciton S1, which will follow the scenario in Fig. 4.
The experimentally determined value of the meta-
stable yield, 0.0014 Ne*/photon, via S1 state sug-
gests that the total desorption yields via S1 excitons
are about 0.2 atoms/photon considering the ratio of
the total yield to the metastable one in the case of S1
excitation. This means that the conversion rate from
B1 exciton to S1 exciton is 0.2. The excited atom in
the second or third layers can be squeezed out because
of the large lattice distortion and can blow a number
of atoms in the surface layer away. The desorption of
metastable Ne from the underlying layers, which
losses the kinetic energy by collisions with neutral at-
oms in the overlayers, was experimentally identified
in the TOF spectra of Fig. 2 as a tail D. The large de-
sorption yields of the order of unity brought by the
bulk exciton can be attributed to this internal sputter-
ing mechanism by excited atoms in the second and
deeper layers beneath the surface [7].
The TOF spectrum of desorbed metastables by the
excitation of B2 exciton is quite similar to that by S‘
except the signal caused by a long life time (several
hundreds �s) luminescence. The relaxation of B2 into
S1, S’, and the excimer is clearly distinguished from
the spectrum. Though the total number of excitons ex-
cited by B2 excitation is essentially the same to that
by B1 excitation, there is an obvious difference be-
tween the total desorption yields by B1 and B2 excita-
tions. This is probably because of the difference in the
distribution of the initial excitation in the Ne film and
because of the relaxation channel of B2, which yields
the luminescence stated above.
5. Desorption of Kr
5.1. Metastable desorption
The desorption of metastable species from solid Kr,
which has been observed by electron and high energy
Absolute yields of the exciton induced desorption at the surface of solid rare gases
Fizika Nizkikh Temperatur, 2003, v. 29, No. 3 347
photon 70.5 nm
� 0.1 � 0.1 0.9 � 0.1 0.9 (n –2)
radiative decay
B1 exciton
S1
3P 2,0
3 P2,0
3 P 2,1,0 , 1 P 1
- - - -
migration in solid
��
Fig. 5. Schematic chart of the relaxation cascade of the first order bulk exciton B1 in solid Ne.
ion bombardment, is attributed to the ED mechanism.
However, it has been reported that adsorption of a
small amount of hydrogen makes it possible to desorb
metastables by the CE mechanism by changing the
electron affinity into negative [26]. Because of this
phenomenon, it is difficult to determine the absolute
desorption yields of a metastable via the process in-
trinsic of pure solid Kr. This is also the case for solid
Xe.
5.2. Total desorption
The wavelength dependence of the photo-de-
sorption intensities from solid Kr for three different
film thicknesses is shown in Fig. 6, where the curves
are the direct output from the mass spectrometer with
no compensation applied for a higher order light. The
arrows show the wavelength corresponding to the ex-
citation of the series of excitons in solid Kr. The coin-
cidence between the exciton excitation energies and
the peaks or the shoulders in the figure clearly shows
that the excitons induce the desorption. The de-
sorption yields were estimated by the peak height
above the continuous background which was due to
the bulk ionization caused by the higher order light
from the monochromator. The desorption yield at
S1(3/2) excitation has apparently no thickness de-
pendence and is about 0.015 atoms/photon. The de-
sorption by the bulk exciton creation becomes detec-
table at a few tens atomic layers in our experimental
condition. The desorption yields at B1(3/2) and
B1(1/2) excitation increase with the film thickness
and reach saturated values, 0.03 and 0.02 atoms/pho-
ton, respectively, at a thickness of about 100 atomic
layers. The absolute yields of the total desorption by
the five different excitonic excitations are summarized
in Table.
5.3. Desorption model and simulation of total
desorption yields
The desorption process caused by excimer dissocia-
tion and internal sputtering after the bulk exciton ex-
citation is considered as sequential processes as sche-
matically shown in Fig. 7: i) creation of a free exciton,
ii) migration of the exciton, iii) formation of an
excimer, iv) dissociation of the excimer, and v) colli-
sion cascade followed by internal sputtering. In order
to calculate the desorption yields, each step should
quantitatively be evaluated. The detail of the calcula-
tion will be published elsewhere [27]. Though the be-
havior of the exciton at a vacuum interface and metal
substrate is not well understood, the photoemission
data of Kr were well described by assuming that the
metal surface is a perfect sink and that the vacuum in-
terface is a perfect reflector for the exciton [28]. The
final process v), which is one of the most important
factors to give the theoretical desorption yields, has
extensively been studied by a classical molecular dy-
namic simulation [7]. A distribution of the initial ex-
citation in the process i) and a diffusion length in the
migration ii) affect the depth profile of the distribu-
tion of the trapped excimer and, therefore, determine
the film thickness dependence of the total desorption
yield. The initial distribution of the excited bulk
excitons is estimated from the photo-absorption coeffi-
cient [29]. The only unknown parameter is the diffu-
sion length L of the free exciton, which is expected to
be highly affected by crystal condition and tempera-
ture. As is shown in Fig. 8, both the absolute value
and the thickness dependence were satisfactorily re-
produced by choosing the parameter L between 5 and
10 nm. This value shows good agreement with the esti-
mation between 1 and 10 nm by Schwentner et al.
348 Fizika Nizkikh Temperatur, 2003, v. 29, No. 3
I. Arakawa, T. Adachi, T. Hirayama, and M. Sakurai
125120115110105
Wavelength, nm
B2(1/2)
B2(3/2)
S2(3/2) S1(3/2)
1200 atomic layers
10 atomic layers S1(1/2)
B1(1/2) B1(3/2)
150 atomic layers
D
e
so
rp
tio
n
yi
e
ld
s,
ar
b
. u
n
its
Fig. 6. The total desorption yields as a function of inci-
dent wavelength for three different thicknesses of Kr
films.
iii)
i)
iv)
v)
ii)
Fig. 7. The model of sequential steps of the excimer disso-
ciation process in solid Kr.
from the photoemission study [28]. We have also ap-
plied this method of simulation to the ESD case of
solid Kr and obtained satisfactory results [27].
Though we assume that the CE mechanism does not
work at the surface of solid Kr, the surface exciton ex-
citation resulted in significant total desorption. This
may be caused by the excimer dissociation on the sur-
face. However, we cannot exclude the effect of hydro-
gen adsorption which makes the CE process effective.
6. Desorption of Ar
The desorption of metastables from solid Ar by
both CE and ED mechanisms has been reported by
many researchers [24,30,31]. There are some experi-
mental difficulties, however, in determining the abso-
lute yields of Ar metastable; unknown angular distri-
bution of desorbed Ar, low signal intensity, and
unstable detection efficiency of a secondary electron
multipliers. Nevertheless we have obtained a rough es-
timation of the absolute metastable yield of 1·10–5
atoms/photon by S1 excitation [12]. Though the de-
termination of the absolute yields of total desorption
has already been made in the same way as for Kr (see
Table) the comparison between the experimental re-
sults and the simulation is not so straightforward as in
the Kr case. It is expected for solid Ar that both CE
and ED mechanisms contribute to the total desorption
in comparable measure. The model calculation of the
total yield at B1 excitation, in which only the ED
mechanism is assumed under conditions similar to the
Kr case, gives a theoretical desorption yield of almost
one tenth of the experimental value. This large dis-
crepancy is probably because the bulk excitons are ef-
ficiently converted to the surface ones at the vacuum
interface and they are desorbed by the CE mechanism.
The model simulation which takes both CE and ED
mechanisms into account is under progress.
7. Summary
We have measured the absolute yields of
metastable and total desorption at the surface of solid
Ne, Ar, and Kr by the photostimulated desorption
technique. The experimental results are well ex-
plained by the model in which the excitons play a cru-
cial role. However the mechanism which leads to the
desorption for each rare gas is different. For solid Ne,
the predominant mechanism which works for both
metastable and total desorptions is the CE one. The
large lattice distortion energy makes the internal sput-
tering process possible and results in a large total
desorption yield of the order of 1 atoms/photon by
bulk exciton excitation. For solid Kr, the desorption
process can be well described solely with the ED
mechanism in which the migration of a free exciton be-
fore trapping into the excimer state determines the ab-
solute yields and its film thickness dependence. In
solid Ar, both CE and ED mechanisms work for the
desorption, which make the model calculation of the
desorption yield difficult. The mechanism which
works in the desorption process from solid Xe must be
solely the ED one as is the case for solid Kr. Unfortu-
nately, we have not obtained the data on absolute to-
tal desorption yields which can be compared with the
simulation result. This is because of the experimental
difficulty due to the low desorption signal intensity
and the limitation of our experimental apparatus.
It was demonstrated that the desorption study
could reveal the dynamics of excitons in rare gas solids.
The promising step in the future study may be a sys-
tematic investigation of the temperature dependence
of the desorption phenomena on a well defined sample
crystal. The desorption study at the surface of a rare
gas alloy [32] will also yield interesting information
on the dynamics of the exciton in a heterogeneous
system.
Acknowledgment
This work was supported by the Joint Studies Pro-
gram of the Institute for Molecular Science and partly
by a Grant-in-Aid for Scientific Research from the
Ministry of Education, Science, Sports, and Culture,
Japan.
Absolute yields of the exciton induced desorption at the surface of solid rare gases
Fizika Nizkikh Temperatur, 2003, v. 29, No. 3 349
0.08
0.06
0.04
0.02
0.00
10 100 1000
Thickness of Kr film, atomic layers
1nm
5nm
10nm
50nm
Diffusion length L
A
b
so
lu
te
d
e
so
rp
tio
n
yi
e
ld
s,
at
o
m
s/
p
h
o
to
n
Fig. 8. Thickness dependence of the total desorption yield
of solid Kr; the experimental results (solid square) and
the simulated ones (lines) for the diffusion length as a pa-
rameter.
1. I. Arakawa, Mol. Cryst. Liq. Cryst. 314, 47 (1998).
2. F. Colletti, J.M. Debever, and G. Zimmerer, J. Phys.
Lett. 45, 467 (1984).
3. P. Feulner, T. Muller, A. Puschmann, and D. Menzel,
Phys. Rev. Lett. 59, 791 (1987).
4. T. Kloiber and G. Zimmerer, Radiat. Eff. Def. Sol.
109, 219 (1989).
5. N. Scwentner, F.-J. Himpsel, V. Saile, M. Skibowski,
W. Steinmann, and E.E. Koch, Phys. Rev. Lett. 34,
528 (1975).
6. D.E. Weibel, A. Hoshino, T. Hirayama, M. Sakurai,
and I. Arakawa, in: Desorption Induced by Electronic
Transitions DIET V, E.B. Stechel, A. Burns, and
D.R. Jennison (eds.), Springer, Berlin (1993), p. 333.
7. S. Cui, R.E. Johnson, and P.T. Cummings, Phys.
Rev. B39, 9580 (1989).
8. N.E. Small-Warren and L.-Y. Chow Chiu, Phys. Rev.
A11, 1777 (1975).
9. T.R. Connor and M.A. Biondi, Phys. Rev. A140, 778
(1965).
10. I.Ya. Fugol’, O.N. Grigorashchenko, A.N. Ogurtsov,
and E.V. Savchenko, J. Luminescence 53, 517 (1992).
11. M. Sakurai, T. Hirayama, and I. Arakawa, Vacuum
41, 217 (1990).
12. T. Hirayama, A. Hayama, T. Koike, T. Kuninobu,
I. Arakawa, K. Mitsuke, M. Sakurai, and E.V. Sav-
chenko, Surf. Sci. 390, 266 (1997).
13. I. Arakawa, T. Adachi, T. Hirayama, and M. Sakurai,
Surf. Sci. 451, 136 (2000).
14. J.A.R. Samson, J. Opt. Soc. Am. 54, 6 (1964).
15. I. Arakawa, D.E. Weibel, T. Nagai, M. Abo, T. Hi-
rayama, M. Kanno, K. Mitsuke, and M. Sakurai,
Nucl. Instrum. Meth. Phys. Res. B101, 195 (1995).
16. M. Sakurai, T. Nagai, M. Abo, T. Hirayama, and
I. Arakawa, J. Vac. Soc. Jpn. 38, 298 (1995).
17. D.E. Weibel, T. Hirayama, and I. Arakawa, Surf. Sci.
283, 204 (1993).
18. M. Sakurai, T. Adachi, T. Hirayama, and I. Arakawa,
UVSOR Activity Report 2001, UVSOR-29, 58
(2002).
19. T. Kuninobu, A. Hayama, T. Hirayama, and I. Ara-
kawa, Surf. Sci. 390, 272 (1997).
20. A. Hayama, T. Kuninobu, T. Hirayama, and I. Ara-
kawa, J. Vac. Sci. Technol. A16, 979 (1998).
21. V. Saile and E. E. Koch, Phys. Rev. B20, 784 (1979).
22. K. Inoue, H. Sakamoto, and H. Kanzaki, Solid State
Commun. 49, 191 (1984).
23. T. Hirayama, T. Nagai, M. Abo, I. Arakawa, K. Mit-
suke, and M. Sakurai, J. Electr. Spectr. Rel. Phen.
80, 101 (1996).
24. I. Arakawa and M. Sakurai, in: Desorption Induced
by Electronic Transitions DIET IV, G. Betz and
P. Varga (eds.), Springer, Berlin (1990), p. 246.
25. D. Pudewill, F. Himpsel, V. Saile, N. Schwentner,
M. Skibowski, and E.E. Koch, Phys. Status Solidi
B74, 485 (1976).
26. Hayama, T. Kuninobu, T. Hirayama, and I. Arakawa,
J. Vac. Sci. Technol. A16, 979 (1998).
27. T. Adachi, T. Hirayama, T. Miura, I. Arakawa, and
M. Sakurai, Surf. Sci. (to be published).
28. N. Schwentner, G. Martens, and H.W. Rudolf, Phys.
Status Solidi B106, 183 (1981).
29. B. Sonntag, in: Rare Gas Solids, vol. II, M.L. Klein
and J.A. Venables (eds.), Academic, New York (1977).
30. D.J. O’Shaughnessy, J.W. Boring, S. Cui, and R.E.
Johnson, Phys. Rev. Lett. 61, 1635 (1988).
31. I. Arakawa, M. Takahashi, and K. Takeuchi, J. Vac.
Sci. Technol. A7, 2090 (1989).
32. D.E. Weibel, T. Nagai, T. Hirayama, I. Arakawa, and
M. Sakurai, Langmuir 12, 193 (1996).
350 Fizika Nizkikh Temperatur, 2003, v. 29, No. 3
I. Arakawa, T. Adachi, T. Hirayama, and M. Sakurai
|