On self-limitation of UV photolysis in rare-gas solids and some of its consequences for matrix studies

Article abstract of DOI:10.1016/S0009-2614(98)00295-4

UV photolysis of small molecules embedded in rare-gas matrices is examined. We demonstrate that photolysis can be self-limited when products absorb the photolysing radiation. As a result of the rising absorption, in-situ detected luminescence of the photolysis product saturates faster than its concentration. In particular, the present study supports the conclusion that 193 nm photolysis of hydrogen-containing species in Xe matrices produces hydrogen atoms in amounts comparable with the other dissociating part of the precursor. Also, we show that 193 nm radiation activates mobility of hydrogen atoms in annealing, accelerating photochemical processes related to hydrogen mobility.

Full text of DOI:10.1016/S0009-2614(98)00295-4

29 May 1998
Ž
.
Chemical Physics Letters 299 1998 727–733
On self-limitation of UV photolysis in rare-gas solids and some of
its consequences for matrix studies
)
Leonid Khriachtchev , Mika Pettersson, Markku Rasanen
¨ ¨
Laboratory of Physical Chemistry, UniÕersity of Helsinki, P.O. Box 55, FIN-00014 Helsinki, Finland
Received 19 January 1998; in final form 5 March 1998
Abstract
UV photolysis of small molecules embedded in rare-gas matrices is examined. We demonstrate that photolysis can be
self-limited when products absorb the photolysing radiation. As a result of the rising absorption, in-situ detected
luminescence of the photolysis product saturates faster than its concentration. In particular, the present study supports the
conclusion that 193 nm photolysis of hydrogen-containing species in Xe matrices produces hydrogen atoms in amounts
comparable with the other dissociating part of the precursor. Also, we show that 193 nm radiation activates mobility of
hydrogen atoms in annealing, accelerating photochemical processes related to hydrogen mobility. q 1998 Elsevier Science
B.V. All rights reserved.
1. Introduction
As an example, this approach was employed to
generate a novel family of rare-gas containing
molecules, such as HXeH, HXeBr, HXeI, HXeCl
Matrix isolation is considered as a productive
method for spectroscopic studies of atoms and
molecules and their interactions including photo-
w
x
and HKrCl 7,8 . Recently, additional novel
molecules HKrCN, HXeCN, HXeNC and HXeSH
Žw x
chemical interconversion 1 , for a more recent re-
w
x
have been observed 9,10 .
w x.
view see Ref. 2 . Species under study are often
In solid xenon, atomic hydrogen forms a pseudo-
unstable or difficult to deposit into a matrix directly
and in this case they can be stabilized in a rare-gas
host by photolyzing a suitable precursor.
Ž
. w
molecule XenH so-called caged hydrogen 6,11–
x
15 . The caged hydrogen is known to absorb photons
in the 6.2–6.4 eV energy region, which brings it to
the excited charge transfer states. These excited states
decay radiatively and the emission can be observed
at around 255 nm with a lifetime of about 40 ns.
This 255 nm emission of XenH was used to
follow the kinetics of atomic hydrogen concentration
In a number of matrix-isolation studies, UV pho-
w
x
tolysis of hydrogen-containing dopants is used 3–6 .
After photoformation of atomic hydrogen in the par-
ent cage, its cage exit and stabilization in the rare-gas
host is quite probable at low temperature. Annealing
to a certain temperature mobilizes the hydrogen atoms
in the host, and they can reach other species to react.
w
x
in 193 nm photolysis of HBr 16 . It was shown that
the luminescence saturates much faster than the pho-
tolysis does and some loss mechanisms for atomic
hydrogen were discussed. It was concluded that the
)
Corresponding author. E-mail: khriacht@rock.helsinki.fi
0009-2614r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 98 00295-4

(
)
728
L. KhriachtcheÕ et al.rChemical Physics Letters 299 1998 727–733
maximum concentration of hydrogen atoms created
during photolysis is more than an order of magnitude
less than the maximum concentration of Br atoms
In order to describe the luminescence kinetics, we
consider the modifications of optical properties of
the sample during the photolysis. It was shown by
Fajardo and Apkarian that rare-gas matrices can
w
x
17 .
However, our recent data on the formation and
w
x
often not be regarded as optically thin 20,21 . This
makes it necessary to account for the depletion of the
excitation field during its penetration into the matrix
layer, and, in quantitative kinetic studies, the rate
equations should be integrated over both time and
depth. If the matrix contains a sufficient amount of a
precursor, hydrogen atoms formed in the photolysis
can provide strong absorption of the photolysis radia-
tion and this decreases the photolysis efficiency for
the deeper parts of the sample. Also, the excitation
of hydrogen atoms becomes less efficient in the
inner parts of the sample, which reduces the detected
emission from that area. In other words, the lumines-
cence method detects essentially the surface layer
when the matrix contains a large number of hydro-
gen atoms. The estimations show that a 100 mm
thick matrix with precursorrXes1:2000 and an
area of 1 cm2 contains ;2=1017 precursor
molecules. One 10 mJ pulse of 193 nm radiation
consists of ;1016 photons. This means that such a
matrix can absorb totally the radiation after longer
photolysis.
IR-induced decomposition of HXeI molecules em-
ploying HI photolysis proposes a much higher rela-
w
x
tive concentration of hydrogen atoms 18 . Indeed,
up to 50% of the iodine atoms formed during photol-
ysis were shown to participate in the formation of
HXeI molecules in annealing. By taking into account
the simultaneous diffusion-controlled formation of
HXeH molecules, one can expect a large part of the
hydrogen atoms to stabilize in solid xenon. This
estimate was also quantitatively confirmed in the
studies on formation of HXeNCrHXeCN molecules
w x
9 .
In this Letter, we explain the remarkable disagree-
ment between the luminescence and infrared-absorp-
tion estimates on the concentration of hydrogen atoms
in Xe matrices by employing the approach of opti-
cally thick matrices. This research was motivated by
the importance of hydrogen atoms in low-tempera-
ture photochemistry and by the wide practical use of
UV photolysis of hydrogen-containing species and
luminescence probing. We have chosen HCN as a
precursor for this study because of reliable IR detec-
tion of the precursor and CN radicals formed in the
photolysis. Also, it is useful for our study that HCN
The qualitative picture can be put onto a quantita-
tive level, similar to considerations by Fajardo and
w
x
Apkarian 20,21 . We consider a matrix layer being
photolyzed from one side. The photolysis efficiency
w
x
is a quite weak absorber of 193 nm radiation 19 .
Ž
.
is assumed to be proportional to k1 I x,t where
Ž
.
I x,t is the laser intensity in relative units at a given
2. Qualitative considerations and computational
results
coordinate and time and k1 is the photolysis time
Ž
.
constant for I x,t s1. The relative concentration of
Ž
.
hydrogen atoms nH x,t , establishes the absorption
Ž
.
The known behavior of XenH luminescence re-
coefficient k2 nH x,t , the factor k2 depending on
the transition oscillator strength. The luminescence
signal from hydrogen atoms can be written as
w
x
ported by LaBrake and Weitz 16 is the starting
point for our consideration. In fact, under in-situ 193
nm photolysis, the emission reaches the saturation
level when only a small portion of a hydrogen-con-
taining precursor is photolyzed. Further photolysis
results only in slow depletion of the steady-state
emission. This experimental picture could be ex-
plained by employing extensive loss processes for
hydrogen atoms. However, the studies of 193 nm
photolysis of HI and HCN in Xe matrices propose a
S t sAk LI x,t n x,t d x
1
Ž .
Ž .
Ž
.
H Ž
.
2H
0
where A describes the efficiency of luminescence
detection and L is the thickness of the matrix layer.
In this simple model, the precursor absorption is
neglected.
Fig. 1 presents the precursor concentration and
the photolysis laser intensity as a function of the
penetration depth for two different photolysis times.
w
x
rather large proportion of hydrogen atoms 9,18 and
here on we try to reconcile these two experiments.

(
)
L. KhriachtcheÕ et al.rChemical Physics Letters 299 1998 727–733
729
more quantitative consideration difficult. The first
factor is the influence of Rayleigh scattering. This
non-coherent effect changes the direction of the light
wave propagation due to optical inhomogenity of the
medium. Essentially, the intensity distribution after
scattering is written as
I ; 1qcos2u rl4
2
Ž .
Ž
.
u
where u is the angle of scattering and l is the
Ž
w x.
radiation wavelength see for instance Ref. 22 . It is
important to note that the cross-section of Rayleigh
scattering increases with radiation frequency. While
a sample does not scatter much in IR absorption
measurements, it can be significantly scattering at
the photolysis wavelength. From the viewpoint of the
effect under discussion, Rayleigh scattering can en-
hance the effective absorption of the sample because
it increases the pathlength of the radiation in the
sample bulk. This is illustrated by Fig. 3 where we
compare the spatial distributions of photolysis effi-
ciency with and without scattering. The influence of
scattering was estimated by a simple model of one-
Fig. 1. The precursor concentration and the laser intensity as a
function of penetration depth into matrix bulk calculated for two
photolysis times. The arrows show the axes, to which the curves
refer.
In the beginning of the photolysis, both the precursor
concentration and the laser intensity are set equal to
1. During the photolysis, the medium becomes more
absorbing. This absorption leads to a decrease of the
laser intensity with depth. As a result, the photolysis
is more efficient near the surface, compared with
deeper in the bulk, where it slows down with time.
Thus, the progress of the photolysis reduces its
efficiency, i.e. self-limitation of the process takes
place.
Ž
.
dimensional scattering forward and back with ab-
sorption happening after twelve scatterings of a pho-
ton. It is evident from Fig. 3 that Rayleigh scattering
limits strongly the penetration of radiation into the
bulk and enhances the radiation intensity in the
surface layer. In reality, three-dimensional scattering
in inhomogeneous matrix morphology can produce
Fig. 2 compares the kinetics of the product con-
centration and the corresponding luminescence for
two thicknesses of the sample with the same precur-
sor concentration. For the thinner sample, the differ-
ence between the two curves is not dramatic, about
20%. For the thicker sample, the curves differ greatly,
in particular, the luminescence starts saturating when
only about 30% of the precursor is photolyzed.
Similarly, the concentration and luminescence ki-
netic curves become closer to each other for more
diluted matrices and the difference increases with the
initial precursor concentration.
Our calculations explain qualitatively the differ-
ence between the kinetics of atomic hydrogen con-
centration and the corresponding luminescence.
However, it is clear that the results give only qualita-
tive trends without accurate estimates and we can
mention at least two important factors, which make
Fig. 2. The product concentration and the product luminescence as
a function of photolysis time calculated for two matrix thick-
nesses. The arrows show the axes to which the curves refer.

(
)
730
L. KhriachtcheÕ et al.rChemical Physics Letters 299 1998 727–733
The HCN samples were photolyzed by an excimer
Ž
.
laser ELI 76, Estonian Academy of Science at 193
nm, with pulse energies up to 20 mJ and a pulse
duration of ;10 ns. The spectra were recorded with
Ž
.
an FT-IR spectrometer Nicolet 60SX and with a
Ž
.
single UV–VIS monochromator Spex 270M
equipped with a gated ICCD camera Princeton In-
struments . The concentrations of HCN and CN were
Ž
.
measured as proportional to the IR absorptions at
y1
w
x
w
x
718 24 and 2040 cm
9,25 , respectively.
A typical UV emission generated by 193 nm
photolysis of the samples is shown in Fig. 4a and it
agrees well with the literature data on XenH emis-
sion. It should be noted that this 255 nm emission
undergoes some shape changes during photolysis.
The shape can be stabilized by annealing after pho-
Fig. 3. The photolysis efficiency as a function of penetration depth
into matrix bulk calculated for two cases, without scattering
Ž
.
Ž
.
Ž
.
Ž
.
tolysis to 40 K , with Fig. 4a showing emission of
such an annealed sample. The stabilization upon
annealing suggests a site origin of the effect and
similar trends were noticed, for instance, by Zoval
and Apkarian for sulfur atoms in argon and krypton
kS s0 and with scattering kS rkA s12 . The absorption kA
was assumed to be constant along the coordinate and one dimen-
sional scattering was treated.
complicated modification of the laser radiation distri-
bution in the sample bulk. Secondly, some additional
absorbers to some extent can influence the result. For
example, the absorption of the precursor was ne-
glected in our computations because of its weakness
w x
matrices 5 . This interesting phenomenon of the
emission modification is under study in our labora-
tory.
Under 193 nm photolysis of HCN in Xe matrices,
decomposition of HCN correlates reasonably well
with the formation of CN radicals although some
formation of HNC isomers takes place as indicated
w
x
19 . Also, some modification of luminescence kinet-
ics can arise from minor absorption of luminescence
Ž
photons by matrix impurities molecular oxygen, as
an example and water impurities can contribute to
atomic hydrogen concentration.
by the absorptions at 3577, 2021 and 477 cmy1
.
2
w
x
9,26 . The 255 nm emission saturates faster than
decomposition of the precursor happens, in agree-
ment with the observation by LaBrake and Weitz
w
x
16 . This saturated value decreases slowly in longer
3. Experimental apparatus and results
photolysis and sometimes some persistent proportion
of the precursor is observed.
The experiments were performed with solid-state
xenon matrices doped with HCN in various propor-
In order to test the hypothesis described above,
we followed the time-dependence of the 255 nm
emission. Just after saturation of the emission, we
measured a level of HCN decomposition. The exper-
iment was run for three samples with the same
Ž
.
tions typically 1:2000 prepared by a standard
1
manometric technique with synthesized HCN and
Ž
.
xenon )99.99%, Aga . The gas mixture was de-
posited onto a CsI substrate at 7.5 K in a closed
Ž
.
HCNrXe ratio 1:2000 and different thicknesses
Ž
.
cycle helium cryostat HC-2, APD . The temperature
Ž
.
was measured with a silicon diode accuracy 0.5 K
and a resistive heater provided higher temperatures.
2 In Ref. 26 , the absorptions of HNC were measured in Ar
matrixes and the results correlate well with our measurements in
solid xenon. To our knowledge, there is no literature data about
HNC IR absorptions in Xe matrixes.
w
x
1 HCN was synthesized from H2 SO4 and KCN 23 , dried with
phosphorus pentoxide and purified by vacuum distillation.
w
x

(
)
L. KhriachtcheÕ et al.rChemical Physics Letters 299 1998 727–733
731
Ž
saturation level of atomic hydrogen emission ;90%
.
of the maximum . This result is in perfect agreement
with the previous discussions.
Also, we carried out the same measurement using
Ž
a ten times more diluted sample HCNrXes
.
1:20000 with matrix thickness of 300 mm. The
dilution was found to slow down saturation of the
255 nm emission relative to the decomposition of
HCN, again in agreement with our calculations. For
Ž
this last sample, the saturation of the emission ;
.
90% of the maximum corresponds to about 30%
decomposed HCN, which is three times more than
that proportion for the ten times more concentrated
sample of the same thickness.
In a final experiment, we studied thermal bleach-
ing of the 255 nm luminescence and the influence of
Ž
irradiation on this process. A sample HCNrXes
.
1:20000 was first properly photolyzed at 7.5 K.
Then, we increased the temperature to 50 K and
detected the luminescence excited by weak probing
radiation. The luminescence was measured every 10
min for 10 s with the pulse energy density -0.1
mJrcm2 at a 2 Hz repetition rate and such probing
irradiation was found not to influence the bleaching
process much. This essentially temperature-induced
Ž
bleaching is shown in the first stage of Fig. 5 0–80
Ž .
Fig. 4. 193 nm photolysis of HCN in solid xenon at 7.5 K. a The
Ž .
in-situ excited XenH emission spectrum.
b Comparison of data
for HCN concentration measured by IR absorption with lumines-
Ž
.
cence kinetics measured for three matrix thicknesses 300 crosses ,
Ž
.
Ž
.
50 circles and 15 mm triangles . The radiation with ;3
mJrpulsercm2 was used for photolysis. The luminescence data
were scaled to approximately coincide to each other. The lines are
used for guiding the eye.
300, 50 and 15 mm. The thickness of the thinner
matrix was measured from interference patterns in
the IR absorption spectra and the ratios were con-
firmed by comparing the HCN IR absorptions, which
were found to correlate with the deposition times.
The comparison of decomposition efficiencies for
the three samples is presented in Fig. 4b, the time-axis
being scaled to overlap the luminescence dependen-
cies. As the most essential result, a strong difference
in the precursor decomposition efficiency was ob-
served for the three samples. For the 300, 50 and 15
mm thick samples, respectively, decomposition of
about 10%, 30% and 65% corresponds to the same
Fig. 5. Kinetics of XenH emission signal in different experimental
conditions. The strong 193 nm radiation has a pulse energy
density of about 10 mJrcm2 at 2 Hz. The probing radiation used
at the first stage is weaker by a factor of 100 and the period of a
measurement was limited to 10 s. The luminescence signals under
strong and probing radiation are normalized to each other at the
point of switching the strong radiation on. The cooling period
Ž
.
from 50 to 7.5 K is excluded.

(
)
732
L. KhriachtcheÕ et al.rChemical Physics Letters 299 1998 727–733
.
min . Then, we applied strong 193 nm radiation to
further photolysis. From this point of view, the limi-
tation of photolysis efficiency is essentially self-limi-
tation. The self-limitation of the photolysis explains
the persistent part of the precursor, which is often
observed in practice. In the framework of the present
study, this limitation is most probable in matrices of
poor optical quality providing efficient scattering.
The effect under discussion takes place when hydro-
gen atoms in Xe matrices are generated by 193 nm
photolysis of HCl, HBr, HCN, H2 S, etc. As a further
example, we can mention photolysis of hydrogen
peroxide at wavelengths resonant with OH radicals.
Also, the absorption of photolysis radiation by the
products influences the in-situ detected luminescence
of the products. In particular, the luminescence does
not follow the concentration of the products but
saturates faster than the concentration does. This
applies especially to scattering samples. In photolysis
experiments with Xe matrices, we noticed often some
increase of scattering, which was visible in the IR
spectra. According to the present discussion, the
possible increase of scattering during photolysis re-
duces the luminescence signal when the concentra-
tion is constant. This process may be responsible for
the experimentally observed decrease of the 255 nm
luminescence during long photolysis.
2
Ž
.
the sample ;10 mJrpulsercm , 2 Hz . A fast
decrease of the 255 nm emission was observed as
shown in the second stage of Fig. 5 80–110 min ,
the luminescence signal being normalized to that
excited by the weak probing radiation. The emission
signal saturates at some level ;20% of a starting
level and does not change more under longer irradi-
ation. When the temperature of the sample was
lowered to 7.5 K the cooling period is not shown in
the picture , the strong irradiation returned the lumi-
nescence signal to rather close to the starting level.
Such cycles under the strong irradiation photo-in-
duced bleaching at 50 K and photolysis at 7.5 K can
be repeated without an essential decrease of the
steady-state emission.
Ž
.
Ž
.
Ž
.
Ž
.
4. Discussion
The experiments described above distinguish two
possible origins of the different time dependencies
for XenH luminescence and HCN decomposition:
absorption of the photolysis radiation and possible
chemical losses. In the model of self-limited photoly-
sis, decomposition of the precursor and product lu-
minescence depend on the sample thickness in dif-
ferent ways. Namely, decomposition of the precursor
takes longer and the luminescence saturates faster for
thicker samples, i.e. their kinetics have an opposite
thickness dependence. Of course, the loss reactions
The analysis of the present results together with
w
x
our recent data on IR decomposition of HXeI 18
supports the conclusion that the major part of the
dissociated hydrogen atoms is stabilized in the ma-
trix. In annealing, most of the hydrogen atoms partic-
ipate in the formation of such xenon-containing
species as HXeH and HXeI, and only a minor part of
them restores the precursor and forms H2 , for in-
stance, in the reaction
w
x
discussed in Refs. 16,17 are independent of film
thickness. Thus, the difference in kinetics of photoly-
sis efficiency and luminescence of photolysis prod-
ucts arises essentially from the increasing absorption
3
of the medium at the photolysis wavelength.
HqHI™H2 qI .
3
Ž .
The effect under discussion occurs when the pho-
tolysis products absorb the photolyzing radiation and
the photolyzed front layer covers the bulk preventing
More extensive discussions of this subject can be
w
x
found elsewhere 18 . It should be noticed here that
the hydrogen atoms arising from dissociation of wa-
ter impurities are neglected. This approximation is
justified because the impurity water concentration
was estimated to be less than 1:20000. The strong
dependence of the hydrogen luminescence on the
HCN concentration supports this approximation at
least for the matrices with HCNrXes1:2000. In
the framework of the present study, dissociation of
water impurities slows down the photolysis and ac-
3 We obtained more experimental evidence of this conclusion
while studying 193 nm photolysis of HCN in Kr matrices. CN
radicals in solid krypton were found to absorb the 193 nm
radiation and the CN emission at ;790 nm was detected. Thus,
we could measure both luminescence and IR absorption of the
same product. Indeed, the CN luminescence saturated faster than
CN concentration measured by IR absorption.

(
)
L. KhriachtcheÕ et al.rChemical Physics Letters 299 1998 727–733
733
celerates the saturation of the luminescence and this
process can influence particularly the kinetics mea-
recombine mainly into xenon-containing neutral
species. Additionally, 193 nm radiation activates the
mobility of the hydrogen atoms during annealing.
This radiation-enhanced mobility accelerates photo-
chemical processes involving hydrogen mobility.
Ž
.
sured in the more diluted HCNrXes1:20000
sample.
In luminescence kinetics studies, it is also quite
important to account for the radiation-induced en-
hancement of the hydrogen mobility because the
probing radiation can change the concentration influ-
encing the estimates of the rate constants. Fig. 5
demonstrates the increase of atomic hydrogen mobil-
ity while irradiating the matrix at 50 K. This temper-
References
w x
1
Ž .
L.E. Andrews, M. Moskovits Eds. , Chemistry and Physics
Ž
.
of Matrix Isolated Species Elsevier, Amsterdam, 1989 .
w x
2
V.E. Bondybey, A.M. Smith, J. Agreiter, Chem. Rev. 96
Ž
.
1996 2113.
Ž
.
ature 50 K is known to mobilize hydrogen atoms in
w x
Ž
.
3
w x
4
H.M. Kunttu, J.A. Seetula, Chem. Phys. 189 1994 273.
E.I. Tarasova, A.M. Ratner, V.M. Stepanenko, I.Ya. Fugol,
M. Chergui, R. Schriever, N. Schwentner, J. Chem. Phys. 98
w x
Xe matrices 6 and their thermal diffusion causes
the slow decay of the luminescence without irradia-
tion. The 193 nm irradiation gives some additional
kinetic energy to individual hydrogen atoms in the
process of XenH decay, and the mobility of hydro-
gen atoms enhances dramatically, about by a factor
of 102 in our experiments, thus dominating the
thermal mobility at 50 K. On the other hand, Fig. 5
also proves the essential participation of the lattice in
atomic hydrogen mobility. Indeed, when decreasing
the temperature, we obtained stabilization of hydro-
gen atoms again in the network. Thus, it is not
enough for an atom to have kinetic energy to travel
from cage to cage but also the lattice should be
activated. It seems to us that the observed behavior
can be interpreted by using the known model of local
Ž
.
1993 7786.
w x
Ž .
5
w x
6
J. Zoval, V.A. Apkarian, J. Phys. Chem. 98 1994 7945.
J. Eberlein, M. Creuzburg, J. Chem. Phys. 106 1997 2188,
and references therein.
Ž
.
w x
7
M. Pettersson, J. Lundell, M. Rasanen, J. Chem. Phys. 102
¨ ¨
Ž
.
1995 6423.
w x
8
M. Pettersson, J. Lundell, M. Rasanen, J. Chem. Phys. 103
¨ ¨
Ž
.
1995 205.
w x
9
M. Pettersson, J. Lundell, L. Khriachtchev, M. Rasanen, J.
¨ ¨
Ž
.
Chem. Phys. 1998 in press.
w
w
w
x
10 L. Khriachtchev, M. Pettersson, E. Isoniemi, M. Rasanen, J.
¨ ¨
Ž
.
Chem. Phys. 108 1998 5747.
x
11 M. Creuzburg, F. Koch, F. Wittl, Chem. Phys. Lett. 156
Ž
.
1989 387.
x
12 F. Wittle, M. Creuzburg, R. Schriever, J. Lumin. 48r49
Ž
.
1991 611.
w
w
w
w
w
x
Ž
.
.
.
13 M. Kraas, P. Gurtler, Chem. Phys. Lett. 174 1990 396.
¨
w x
lattice melting developed by Tarasova et al. 4 .
x
Ž
14 M. Kraas, P. Gurtler, Chem. Phys. Lett. 183 1991 264.
¨
x
Ž
15 M. Creuzburg, F. Wittle, J. Mol. Struct. 222 1990 127.
x
Ž
.
16 D. LaBrake, E. Weitz, Chem. Phys. Lett. 211 1993 430.
x
Ž
.
17 D. LaBrake, E.T. Ryan, E. Weitz, J. Chem. Phys. 102 1995
5. Conclusions
4112.
w
w
w
w
x
18 M. Pettersson, J. Nieminen, L. Khriachtchev, M. Rasanen, J.
¨ ¨
In the present study, we have demonstrated that
UV photolysis in solid-state matrices can be self-
limited when the photolysis products absorb the pho-
tolysis radiation. Practically, this leads to slowing
down the photolysis efficiency with time. The growth
of the absorbers can lead to fast saturation of the in
situ detected luminescence. As a result, the lumines-
cence signal does not follow the product concentra-
tion accurately. The 193 nm photolysis of hydrogen-
containing species in Xe matrices produces hydrogen
atoms in amounts comparable with the other dissoci-
ating part of the precursor. In annealing, these parts
Ž
.
Chem. Phys. 107 1997 8423.
x
19 H. Okabe, Photochemistry of Small Molecules, 1st edn.
Ž
.
Wiley, New York, 1978 .
x
Ž
.
.
20 M.E. Fajardo, V.A. Apkarian, J. Chem. Phys. 85 1986
5660.
x
Ž
21 M.E. Fajardo, V.A. Apkarian, J. Chem. Phys. 89 1988
4102.
w
w
x
x
Ž
.
22 P.W. Milloni, J.H. Eberly, Lasers Wiley, New York, 1988 .
.
23 E. Knozinger, R. Wittenbeck, J. Chem. Phys. 80 1984
¨
Ž
5979.
w
w
x
x
Ž
.
24 A.D. Abbate, C.B. Moore, J. Chem. Phys. 82 1985 1255.
25 G. Schallmoser, A. Thoma, B.E. Wurfel, V.E. Bondybey,
Ž
.
Chem. Phys. Lett. 219 1994 101.
w
x
Ž
.
26 D.E. Milligan, M.E. Jacox, J. Chem. Phys. 47 1967 278.