UV-photolysis of HI···CO2 complexes in solid parahydrogen: Formation of CO and H2O

Article abstract of DOI:10.1021/jp9938462

The photochemistry of (HI)_m···(CO₂)_n (m, n = 1, 2, ...) complexes trapped in solid parahydrogen was studied by Fourier transform infrared absorption spectroscopy. Photolysis of the HI in the HI···(CO₂)_n

Full text of DOI:10.1021/jp9938462

J. Phys. Chem. A 2000, 104, 3635-3641
3635
UV-Photolysis of HI‚‚‚CO2 Complexes in Solid Parahydrogen: Formation of CO and H2O†
‡
‡,§
,‡
,|
Mizuho Fushitani, Tadamasa Shida, Takamasa Momose,* and Markku R a1 s a1 nen*
DiVision of Chemistry, Graduate School of Science, Kyoto UniVersity, Kyoto 606-8502, Japan,
Department of Chemistry, Kanagawa UniVersity, Hiratsuka, Kanagawa 259-1293, Japan, and Laboratory of
Physical Chemistry, P.O. Box 55 (A. I. Virtasen Aukio 1), FIN-00014, UniVersity of Helsinki, Finland
ReceiVed: October 29, 1999; In Final Form: January 10, 2000
The photochemistry of (HI)
by Fourier transform infrared absorption spectroscopy. Photolysis of the HI in the HI‚‚‚(CO
m
‚‚‚(CO
2
)
n
(m, n ) 1, 2, ...) complexes trapped in solid parahydrogen was studied
complexes
2 n
)
results in the generation of carbon monoxide and water molecules along with iodine atoms. Analysis of the
observed spectra leads to the conclusion that the energetic hydrogen atom produced by the photolysis of HI
initiates the reaction with one of the CO
followed by the secondary reaction of OH radicals with the matrix hydrogen as OH + H
2
molecules in the complexes as H + CO
2
f OH + CO, which is
2
f H O + H. These
2
four-atom reactions are of practical importance in combustion processes as well as of theoretical importance
in fundamental chemical dynamics. The result of the present study shows the usefulness of the parahydrogen
matrix for fundamental study of photochemistry with the use of weakly bound complexes.
1. Introduction
interaction and the large cage dimension of solid p-H2 allow
the accommodation of sizable complexes with little perturbation.
Photolysis of trapped molecules in a low-temperature solid
In this paper, we report a spectroscopic study of photochemistry
of (HI)m‚‚‚(CO2)n (m, n ) 1, 2, ...) complexes trapped in solid
p-H2. Low-pressure mercury lamp irradiation results in the
disappearance of the infrared absorption of HI and the appear-
provides convenient means for spectroscopic studies of photo-
products and intermediates. As for single molecules isolated in
the solid, photoexcitation may induce dissociation or isomer-
ization of the molecules. If the cage barrier of the solid is low
enough compared with excess kinetic energies of photofrag-
ments, they may be permanently separated. If, on the other hand,
the fragments do not have enough kinetic energies to escape
from the cage, recombination may produce new species in the
cage. The cage-exit energies required for the permanent separa-
tion of the photofragments are on the order of ∼0.5-1.5 eV
for hydrogen atoms and higher for heavier particles in the
familiar rare gas matrixes. Thus, isomerization and cage
8,9
ance of the spin-orbit absorption of iodine atoms. In addition,
the absorptions of CO and H2O are also observed to our surprise.
The findings are successfully accounted for.
2
. Experiments
Hydrogen iodide (HI) was synthesized by the reaction of
10
iodine and tetralin at 207 °C. The produced HI vapor was
purified by passing through a tetralin trap before condensing in
a trap at liquid nitrogen temperature. The gas was further
purified by low-temperature distillation in a vacuum line.
1
recombination are predominant for the most photofragments.
Meanwhile, solid parahydrogen (p-H2) is found to be a
superior matrix for both high-resolution vibration-rotation
The preparation of crystalline p-H2 samples was performed
spectroscopy and the study of photochemical processes.4-6
This feature is due to the extremely weak intermolecular
interaction in solid hydrogen. The cage-exit barrier in solid
hydrogen is very low compared with the rare gas matrixes,
which allows us to study photodissociation of trapped molecules
with ease. Furthermore, being a molecular matrix, a hydrogen
molecule in the solid may participate in secondary chemical
reactions with reactive radicals produced in situ photolysis of
2,3
3,11
in the same way as reported previously.
Almost pure p-H2
gas (with the residual orthohydrogen less than 0.01%) was
obtained by passing normal hydrogen gas through a hydrous
ferric oxide catalyst at 13.8 K. The p-H2 gas thus converted
was mixed with the vapor of CO2 and HI. The mixing ratio of
5
6
CO2/HI/p-H2 was 1:16:5×10 and 15:1:2×10 . The mixed gas
was then introduced slowly into an optical cell which was kept
at 8.0 K during the crystallization.
6
,7
precursors.
In such reactions at cryogenic temperatures
All the infrared measurement and irradiation were carried out
at 4.9 K. Fourier transform infrared (FTIR) spectra were
recorded by a Bruker IFS 120 HR spectrometer with a resolution
7
quantum tunneling may be important, which is interesting in
its own right.
Weakly bound complexes with a fixed structure in solid p-H2
are even more interesting than single molecules because such
complexes may be regarded as a prearranged reactant system
in photoinduced bimolecular reactions. The weak intermolecular
-1
of 0.1 and 0.01 cm . A globar and a tungsten lamp were used
as the light source with a KBr beam splitter and a liquid N2
cooled InSb and MCT detector. A 20 W low-pressure mercury
lamp which mainly emits light at 253.7 nm with a small fraction
of light at 184.9 nm was used without filters for the photolysis.
†
Part of the special issue “Marilyn Jacox Festschrift”.
*
To whom correspondence should be addressed. E-mail: momose@
3
. Results
kuchen.kyoto-u.ac.jp; morasane@pcu.helsinki.fi.
‡
Kyoto University.
Kanagawa University.
University of Helsinki.
5
§
3.1. System of CO2/HI/p-H2 ) 1:16:5×10 . FTIR spectra
|
5
of the sample with a composition of CO2/HI/p-H2 ) 1:16:5×10
1
0.1021/jp9938462 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/15/2000

3636 J. Phys. Chem. A, Vol. 104, No. 16, 2000
Fushitani et al.
5
Figure 1. FTIR spectra in the HI stretching region of a sample in solid parahydrogen with a composition of CO
2
/HI/p-H
2
) 1:16:5×10 : (a) the
sample immediately after the deposition; (b) same as (a) after a continual 1020 min UV irradiation. The horizontal arrows indicate the spectral
2 3 2
region is divided into three groups of HI, (HI) , and (HI) clusters which are little affected by CO .
Figure 2. Collection of the spectral information on the HI stretching vibration for monomer, dimers, and trimers. The references for Xe, Kr, and
Ar matrixes are refs 12, 13, and 14, respectively. The stick spectrum for p-H is constructed by reproducing the major observed peaks in Figure 1a.
The frequencies of gaseous HI were measured at room temperatures with the spectrometer described in the Experimental Section.
2
-
1
in the HI absorption region are shown in Figure 1. The
composition was chosen to have HI dominantly over CO2.
Traces a and b represent the spectra before and after a 1020
absorptions below ∼2160 cm amplified by a factor of 10 are
attributed to trimeric HI. The infrared absorption of HI monomer
in matrixes is known to be notoriously weak, and there is a
statement in the literature that “the absorbance of monomer HI
appears to be significantly weaker than the absorbance of HI
that forms a hydrogen bond to another molecule. It is therefore
difficult to observe monomer HI in a matrix without significant
min UV irradiation, respectively. Referring to the literature on
HI in the gas and in various rare gas matrixes,1
2-14
which is
schematically summarized in Figure 2, the spectrum in Figure
a is divided into the three regions as shown by the horizontal
1
-
1
14
arrows. In the region above ∼2210 cm , two faint peaks are
dimer absorption”. The faint and intense absorptions in the
-
1
seen at 2229.1 and 2225.7 cm which are associated with
first two spectral regions in Figure 1a are consistent with the
cited statement. The very weak absorption in the third region
assigned to the trimeric system implies that the absolute
-
1
monomeric HI, whereas the rich absorptions at ∼2210 cm to
-
1
∼
2160 cm are assigned to dimeric HI, and the very feeble

UV-Photolysis of HI‚‚‚CO2 Complexes
J. Phys. Chem. A, Vol. 104, No. 16, 2000 3637
6
Figure 3. FTIR spectra in the HI stretching region of sample in solid parahydrogen with a composition of CO
2
/HI/p-H
2
) 15:1:2×10 : (a) the
sample immediately after the deposition; (b) same as (a) after an initial 180 min UV irradiation; (c) same as (b) after an additional 240 min
irradiation followed by a dark period of 300 min and a further irradiation for 240 min. Thus, the total irradiation time amounts to 180 + 240 × 2
)
660 min. The peaks shown with frequencies appear newly contrary to the case of Figure 1a. The dashed-line arrow in (c) shows the CO stretching
vibration. The multiple peaks may reflect the subtle difference of the environment of CO.
concentration of the trimers is very small, which is reasonable
if the dimerization is the dominant process in the impinging
premixed gas before the crystallization in the optical cell (see
the Experimental Section).
that new absorptions appear in Figure 3 at around 2225-2220,
-
1
2195-2175, and 2150-2130 cm , the wavenumbers of the
major peaks being given in the figure. Since the difference
between the two samples should be attributed to the difference
of the relative concentration of CO2 over HI, the much stronger
absorptions under the arrow “monomer” are assigned to the
monomeric HI affected by CO2. Similarly, the newly appearing
peaks under the arrows “dimer” and “trimer” are attributed to
(HI)2 and (HI)3 moieties interacting with CO2. Here, the notation
of CO2 does not necessarily mean a single CO2 molecule but
small clusters of (CO2)n (n ) 1, 2, ...), because a supplementary
experiment of CO2/p-H2 systems under the same experimental
condition indicates the formation of the clusters (see Figure 5
below).
Since most of the HI molecules in the premixed gas have
little chance to encounter CO2 because of the low ratio of CO2/
HI ) 1:16, most of the peaks are regarded as due to the
monomers, dimers, and trimers of HI uncomplexed with CO2.
-
1
Since the two peaks at 2229.1 and 2225.7 cm barely appear
in the sample of a high ratio of CO2/HI ) 15:1, which is
discussed in the next subsection, we attribute the two peaks to
two different HI monomers uncomplexed with CO2, although
we have no definite interpretation of the appearance of the two
peaks at the moment. In passing, we note that the structure
reported for monomeric HI in Ar matrix14 is absent in solid
p-H2. Furthermore, we note from Figure 2 that the frequencies
Comparison of traces a and b in Figure 3 shows that after an
initial 180 min irradiation the absorptions in the dimer and trimer
regions vanish almost completely, whereas the decrease of the
absorption in the monomer region is a little reluctant. The
additional irradiation for 240 × 2 min further changes the
spectrum from trace b to c in Figure 3. The decrease is
accompanied with the appearance of new absorptions at around
-
1
of monomeric HI in p-H2 are red-shifted by ∼10-20 cm
relative to those in Ar while those of dimeric and trimeric HI
in p-H2 are comparable to those in Ar.
As shown in Figure 1b almost all the peaks disappeared after
a continual 1020 min irradiation. No significant new absorption
-
1
-1
appeared in the spectral region from 4000 to 1000 cm except
2145-2140 cm , as shown under the dashed-line arrow in
-
1
those at 7639.2, 7638.5, and 7637.8 cm . These are assigned
Figure 3c. The new peaks are assigned to carbon monoxide in
reference to a reported peak of CO in Ar matrix at 2148.8
8
to the spin-orbit absorption of the iodine atom, the details of
9
-1 15
which will be published in a separate paper. Therefore, in the
cm . Recently, Fajardo et al. observed the absorption of CO
sample of CO2/HI ) 1:16 the major event of photolysis must
in the similar spectral region by their rapid deposition tech-
nique.
1
6
be the photodissociation of the HI moiety.
6
3
.2. System of CO2/HI/p-H2 ) 15:1:2×10 . The spectra of
Panels a and b in Figure 4 demonstrate the temporal change
of the integrated intensities of the absorption in the HI stretching
region for HI‚‚‚(CO2)n and (HI)2‚‚‚(CO2)n, respectively, as a
function of UV irradiation time and standing period under dark.
Panel c shows the accompanying change of the intensity of the
CO absorption. The increase in the CO absorption is correlated
with the decrease in the absorption of HI‚‚‚(CO2)n in the first
irradiation and reirradiation periods. However, the (HI)2‚‚‚(CO2)n
rapidly decreases to zero in the initial irradiation period, and
6
the sample with a composition of CO2/HI/p-H2 ) 15:1:2×10
in the HI absorption region are shown in Figure 3. The
composition was chosen to have CO2 overwhelmingly relative
to HI. Traces a-c represent spectra before UV irradiation (a),
after a 180 min irradiation (b), and after an additional irradiation
of 240 min plus a dark period of 300 min plus a further 240
min irradiation to make the total irradiation time equal to 660
min (c). Comparison of trace a of Figures 1 and 3 indicates

3638 J. Phys. Chem. A, Vol. 104, No. 16, 2000
Fushitani et al.
Figure 4. The upper two panels show the temporal change of the
integrated intensities of the HI stretching vibration for monomeric HI
and (HI) complexed with small CO clusters. The plots are obtained
2 2
by the integration of peaks under the arrows of “monomer” and “dimer”
spectral regions in Figure 3. The lower two panels are obtained by
plotting the integrated intensity of the CO stretching vibration in Figure
Figure 5. The upper two panels show the spectra in the antisymmetric
stretching region of CO
2 2 2 2
for CO /p-H samples. The ratio of CO of
(a) over (b) is about 1:400. Both spectra are unaffected by prolonged
UV irradiations. The lower two panels are for the same sample as for
Figure 3. Traces c and d correspond to traces a and c in Figure 3,
respectively. Trace e is the difference between traces d and c. The
downward dips and a few overshoots with the asterisks are explained
in the text.
3
2
c and the intensity of the antisymmetric stretching vibration of H O
shown in the right-hand-side of Figure 6, respectively. The horizontal
portions in the eye-guide curves denoted by “Dark” corresponds to a
3
00 min dark period at 4.9 K. Except for this period, the abscissa is
for the UV irradiation time.
A similar trend of the overall decrease is observed also in
13
there is no correlation between (HI)‚‚‚(CO2)n and CO. As for
the last panel d in Figure 4 arguments will be postponed until
the end of this section.
the ν3 region of CO2 which is present in the natural abundance
in the sample, as shown in the left-hand-side panel of Figure 6.
Traces a and b in the left-hand-side panel of Figure 6 are for
the sample giving the spectra in traces a and b of Figure 3 as
well as traces c and d of Figure 5. The difference between traces
b and a is shown in trace c in the left-hand-side panel of Figure
6. As for a few upward shoots designated with the asterisks
in both trace e in Figure 5 and trace c in the left-hand-side
panel of Figure 6, an explanation will be given in the next
section.
Traces a and b in Figure 5 demonstrate the absorption in the
spectral region of the ν3 antisymmetric stretching vibration of
CO2 for CO2/p-H2 systems without HI. The concentration ratio
of CO2 is 5a/5b = 1:400. The difference of the spectra between
the two systems indicates the formation of CO2 clusters of
different sizes. The dimer formation of CO2 in Ar matrix is
reported earlier.17 Also, CO2 clusters up to tetramers are reported
for a three-component system of CO2/HCl/Ar, which resembles
the present system.18 Both spectra in traces a and b in Figure 5
did not change upon irradiation with full UV light from the
lamp for 300 min, and no appearance of CO was observed, in
contrast to the case for Figure 3.
Trace c in Figure 5 demonstrates the ν3 mode of CO2 for the
sample giving the spectrum of the HI stretching in Figure 3a.
After a total of 660 min irradiation, the spectrum in trace c
changed to trace d in Figure 5, in contrast to the photoinsensitive
systems giving the spectra in traces a and b. The difference of
between traces d and c is shown in trace e in Figure 5, which
indicates that the downward dips predominate over the upward
shoots at lower frequencies. In consistence with the downward
dips, the integrated intensity of the spectrum in Figure 5d is
about 20% less than that in Figure 5c. Such a decrease must be
ascribed to CO2 in HI‚‚‚(CO2)n and (HI)2‚‚‚(CO2)n complexes
because in the CO2/p-H2 system there is no such decrease in
the CO2 absorption.
The right-hand-side panel of Figure 6 shows the FTIR
spectrum of the rotationally resolved antisymmetric stretching
-1
vibration of H2O. The peaks at 3765.7, 3763.6, and 3762.4 cm
in trace a of the right-hand-side panel of Figure 6 are due to
the residual water vapor in the optical path. These unwanted
peaks are absent in traces b and c in the right-hand-side panel
of Figure 6 by fortuitous cancellation of the background signal
of moisture. Instead, there appears a newly growing peak at
-
1
3765.5 cm which is slightly red-shifted against the peak at
-
1
3765.7 cm in trace a. This new peak is assigned to H2O
produced by UV irradiation in the solid p-H2 in reference to
-
1
19
the absorption at 3734.3 cm in Ar matrix. The peak begins
to appear in the initial irradiation for 180 min and continues to
grow in the subsequent irradiation for 240 min. It remains
unchanged in the next dark period for 300 min until it restarts
to grow in the reirradiation period for 240 min. The temporal
behavior of this H2O peak is shown in Figure 4d, which is in
parallel with the behavior of the absorption of CO as shown in

UV-Photolysis of HI‚‚‚CO2 Complexes
J. Phys. Chem. A, Vol. 104, No. 16, 2000 3639
Figure 6. (left) Spectra in traces a and b represent the 13CO
antisymmetric stretching region for the sample giving the spectra in Figure 3. They
2
correspond to the sample before and after irradiation, respectively, and trace c is the difference between traces b and a. The three correspond to
1
2
spectra in traces c-e in Figure 5 for the CO
2
2
absorption. (right) A portion of the rotation-vibration spectrum of H O in the antisymmetric
stretching region for the sample giving the spectra in Figure 3. Trace a is for the sample before UV irradiation. The three peaks at 3765.7, 3763.6,
and 3762.4 cm are the signals due to residual moisture in the optical path that failed to be canceled. Trace b is after 180 min UV irradiation, and
trace c is after further irradiation for 480 min. The peak at 3765.5 cm , slightly shifted from the peak at 3765.7 cm for gaseous H
O produced in solid p-H by the reaction discussed in text.
-
1
-
1
-1
2
O, is due to
H
2
2
Figure 4c. The parallel behavior of CO and H2O definitely
indicates that their production is synchronized.
4
. Discussion
From Figures 3-6 it is seen that photolysis of the system of
6
CO2/HI/p-H2 ) 15:1:2×10 in section 3.2 induces the decrease
in the absorptions of the spectral regions of both the HI
stretching and the antisymmetric stretching vibration of CO2
accompanied by the increase in the absorption of CO and H2O.
Thus, the generation of the CO and H2O must be related to a
photochemical decomposition of HI‚‚‚(CO2)n complexes. We
propose the following reactions in the complexes:
HI + hν f H* + I
(1)
(2)
(3)
#
H* + CO f HOCO f OH + CO
2
Figure 7. Energy diagram relevant to analyze the photochemical
processes of HI‚‚‚CO complexes. Numerical values are collected from
various sources in the literature.2
OH + H f H O + H
2
2
2
0-22
To rationalize the scheme, we have collected relevant
energetic information from various sources in the literature
20-22
2
2
different excited states, the branching ratio of I( P3/2) and I( P1/2)
is wavelength dependent. It is reported that at 266 nm excitation
and collectively illustrated in Figure 7. Since the numerical
values in the figure are for the gas-phase reactions or for those
obtained by computation, the small stabilization energy from
the formation of the van der Waals complexes in the present
work in the p-H2 matrix is not taken into account. The lower
left-hand-side corner in the figure represents the initial state,
while the right-most end corresponds to the final state. Since
the dissociation energy of HI is 3.05 eV,2 the excess energies
upon the UV excitation at 253.7 nm ()4.88 eV) are 1.83 and
2
24
about 64% of the I atom is in the ground P3/2 state and that
at a wavelength of 253.7 nm used in the present work the
percentage of the excited iodine atoms is about 66%. As shown
in the left-hand side of Figure 7, the kinetic energy of 1.83 eV
25
2
of the hydrogen atoms produced with I( P3/2) is sufficient to
surmount the 1.06 eV activation barrier for the reaction between
H and CO2 with an approximately 0.8 eV surplus energy at the
barrier top (see refs 22, 26, 27, and references therein). On the
3
2
2
2
0
.88 eV for the production of I( P3/2) and I( P1/2), respectively.
other hand, the H atom formed with I( P1/2) does not have
The excess energy is mainly transferred to the translational
energy of the split H atom because of the heavy mass of I atom.
Since the broad ultraviolet absorption of HI is composed of three
enough energy to overcome the reaction barrier of 1.06 eV.
The reaction of H atoms with carbon dioxide is known to
2
8-30
proceed via the intermediate HOCO.
The intermediate

3640 J. Phys. Chem. A, Vol. 104, No. 16, 2000
Fushitani et al.
species in solid Ar is characterized by Milligan and Jacox, the
respectively. Previously, Donaldson et al. also reported the
reaction, (HI)2 + hν f H2 + I2, in the molecular beam.
-1 31,32
44
strongest absorptions being at 1843.6 and 1211.3 cm .
The
lifetime of the intermediate in the gas phase is reported to be a
The possibility that CO and H2O originate from the photolysis
at 184.9 nm of CO2 is ruled out because no spectral change
occurred in the CO2/p-H2 systems as mentioned in connection
with traces a and b in Figure 5. We examined the other
possibility of the formation of H2O via the photolysis at 184.9
nm of the OH radical in reaction 3 because the dissociation
2
9,30
few picoseconds.
Consistently, the formation of the OH
radical in the gas-phase photolysis of HI and CO2 at about 253.7
3
3
nm is reported to occur on a subpicosecond time scale.
These pieces of information lead immediately to reactions 1
and 2. As for reaction 3, the diagram in Figure 7 indicates that
the excess energy of 0.72 eV may yield initially hot OH radicals
which may react easily with the surrounding hydrogen mol-
ecules. Since the activation energy for this reaction is about
4
5
energy of the radical of 4.35 eV is lower than the photon
energy of 184.9 nm. However, this possibility is slim unless
the OH radical absorbs the second photon with an extreme
efficiency to yield the reactive O atom. The decrease of a part
of the ν3 absorption of CO2 shown in trace e of Figure 5 and in
trace c of the left-hand-side panel of Figure 6 along with the
synchronous formation of CO and H2O are now firmly attributed
to the consecutive reactions 2 and 3 which are initiated by the
absorption of light by the HI moiety in the HI‚‚‚(CO2)n
complexes.
0.24 eV, one may expect the detection of the OH radical if it is
relaxed quickly to the ground state. However, the absorption
of the OH radical reported at 3554.1 cm-1 for an Ar matrix34
was not detected in the present work, which implies either the
radical is hot enough to surmount the barrier of 0.24 eV in
Figure 7 or the relaxed radical may tunnel to the system in the
time scale of the present experiment. The latter possibility is
intriguing because it is proposed that tunneling reactions of
We now consider a few upward shoots in trace e of Figure
5
and trace c in the left-hand-side panel of Figure 6 designated
neutral radicals with the hydrogen molecule may proceed fast
if the reactions are exothermic enough.35
with the asterisks. As mentioned in connection with Figure 7,
the photolysis of HI at 253.7 nm is reported to yield a pair of
In any event, the indication of a rapid reaction of the OH
radical with the hydrogen molecule ensures the correlation of
the formation of CO and H2O. According to the literature,
excitation of the OH radical to either the first or the second
excited vibrational state enhances the rate of reaction 3 only
by less than 50%, while the excitation of the H2 molecule to
2
2
I( P3/2) + H* (∼1.83 eV) and a pair of I( P1/2) + H* (∼0.88
eV) in an approximate ratio of 1:2. So far, we have focused on
the processes initiated by the former only. As for the latter, to
produce the less energetic H atoms, the consecutive reactions
2
2
and 3 are prohibitive so that the counterpart I( P1/2) may
2
3
6,37
quickly relax to I( P3/2) and form a loose complex with (CO2)n.
Such a complex between (CO2)n and the very polarizable I atom
is expected to give red-shifted (CO2)n absorptions.
the V ) 1 state increases the rate by a factor of 150.
How
relevant this information to the present study in the solid
hydrogen is yet to be pursued.
In conclusion, we have successfully observed the consecutive
reactions of one of the CO2 molecules in HI‚‚‚(CO2)n complexes
with an H atom produced by the photolysis of HI to give the
products OH + CO, which is followed by OH + H2 f H2O +
H. The salient feature of the present work is that we employed
the van der Waals complex as the dopant in the p-H2 matrix.
Since the geometrical structure of the complex in the matrix is
fixed, the system is reminiscent of the “precursor-geometry-
limited” method in the gas-phase study which was introduced
Weakly bound complexes in the gas phase often possess well-
defined structures. For example, the HF and HCl complexes
38
with the CO2 molecule are linear while the HBr complex has
a slanted T-shape for the Br‚‚‚CO2 skeleton with the molecular
axes being nearly perpendicular and the distance between the
3
9
centers of mass of the two components being about 3.6 Å.
This structure has been confirmed in pulsed nozzle FT-MW
40,41
measurements also.
However, there seems no structural study
on the HI‚‚‚CO2 complex in the gas phase. Therefore, we have
no definite idea on the structure of the assumed complexes
between HI and CO2 in the p-H2 matrix. The appearance of a
multitude of peaks even in HI‚‚‚(CO2)n system may suggest the
coexistence of a number of local minima on the energy surface
of the complexes, each minimum corresponding to a structure
and to a particular spectral peak. The preferential disappearance
of some peaks in the HI stretching region (Figure 3) as well as
in the CO2 ν3 region (traces c vs d in Figure 5 and traces a vs
b in the left-hand-side panel of Figure 6) may indicate that there
is a distribution of the relative orientation and/or the distance
between HI and the counterpart interacting (CO2)n (n ) 1, 2,
4
6,47
48-50
by Jouvet and Soep
and by the Wittig group,
and is
extensively developed notably by Zewail and Wittig.20
Lastly, we should like to point out that the present system is
closely related to the intensively studied four-atom reactions
of OH + CO f CO2 + H and OH + H2 f H2O + H. The
former is regarded as an important reaction in combustion. The
latter is responsible for the chain propagation in the hydrogen
combustion. These reactions are the subject of the highly
sophisticated theoretical study also. Comprehensive references
30
2
2,26,27
are in the recently published papers.
We hope that the
present approach provides complementary information to the
gas-phase experimental as well as the theoretical studies.
.
..) and that in some complexes the coupled reactions 1 and 2
proceed preferentially. It is desirable to disclose the structure
of each complex and correlate the structure with the chemical
reactivity.
Acknowledgment. The study was partially supported by a
Grant-in-Aid for Scientific Research of the Ministry of Educa-
tion, Science, Culture, and Sports of Japan. M. R a¨ s a¨ nen
acknowledges support from the Academy of Finland.
As shown in Figure 4, the rapid decrease of the (HI)2‚‚‚(CO2)n
in the initial stage of UV irradiation is not correlated with growth
of CO and H2O. Although not shown in the figure, the very
weak (HI)3‚‚‚(CO2)n signal in Figure 3a also behaved similarly
to the (HI)2‚‚‚(CO2)n complexes. The result is reasonable
because the dominant photochemical process in the dimeric HI
system is considered to be one-photonic, that is, (HI)2 + hν f
References and Notes
(
1) See, for example: Brus, L. E.; Bondybey, V. E. J. Chem. Phys.
1
976, 65, 71.
(2) Oka, T. Annu. ReV. Phys. Chem. 1993, 44, 299.
(3) Momose, T.; Shida, T. Bull. Chem. Soc. Jpn. 1998, 71, 1.
4
2
(4) Fajardo, M. E.; Tam, S.; Thompson, T. L.; Cordonnier, M. E. Chem.
Phys. 1994, 189, 351.
5) Momose, T.; Uchida, M.; Sogoshi, N.; Miki, M.; Masuda, S.; Shida,
T. Chem. Phys. Lett. 1995, 246, 583.
H2 + I2 in view of our study of photolysis of (CH3I)2 and
C2H5I)243 in solid p-H2, where we observed an efficient
production of C2 H6 and disproportionation of the ethyl radical,
(
(

UV-Photolysis of HI‚‚‚CO2 Complexes
J. Phys. Chem. A, Vol. 104, No. 16, 2000 3641
(
6) Fushitani, M.; Sogoshi, N.; Wakabayashi, T.; Momose, T.; Shida,
T. J. Chem. Phys. 1998, 109, 6346.
7) Momose, T.; Hoshina, H.; Sogoshi, N.; Katsuki, H.; Wakabayashi,
T.; Shida, T. J. Chem. Phys. 1998, 108, 7334.
(29) Scherer, N. F.; Khundkar, L. R.; Bernstein, R. B.; Zewail, A. H. J.
Chem. Phys. 1987, 87, 1451.
(30) Scherer, N. F.; Sipes, C.; Bernstein, R. B.; Zewail, A. H. J. Chem.
Phys. 1990, 92, 5239.
(
8) Pettersson, M.; Nieminen, J. Chem. Phys. Lett. 1998, 283, 1.
(31) Milligan, D. E.; Jacox, M. E. J. Chem. Phys. 1971, 54, 927.
(32) Jacox, M. E. J. Chem. Phys. 1988, 88, 4598.
(33) Ionov, S. I.; Brucker, G. A.; Jaques, C.; Valachovic, L.; Wittig, C.
J. Chem. Phys. 1992, 97, 9486.
(
(
11) Tam, S.; Fajardo, M. E.; Katsuki, H.; Hoshina, H.; Wakabayashi,
(34) Pehkonen, S.; Pettersson, M.; Lundell, J.; Khriachtchev, L.;
R a¨ s a¨ nen, M. J. Phys. Chem. A 1998, 102, 7643.
(35) Herbst, E. Annu. ReV. Phys. Chem. 1995, 46, 27.
(36) Zellner, R.; Steinert, W. Chem. Phys. Lett. 1981, 81, 568.
(37) Glass, G. P.; Chaturvedi, B. K. J. Chem. Phys. 1981, 75, 2749.
(38) Baiocchi, F. A.; Dixon, T. A.; Joyner, C. H.; Klemperer, W. J.
Chem. Phys. 1981, 74, 6544.
T.; Momose, T. J. Chem. Phys. 1999, 111, 4191.
(12) Kunttu, H. M.; Seetula, J. A. Chem. Phys. 1994, 189, 273.
(13) Bowers, M. T.; Flygare, W. H. J. Chem. Phys. 1966, 44, 1389.
(14) Engdahl, A.; Nelander, B. J. Phys. Chem. 1986, 90, 6118.
(15) Dubost, H.; Abouaf-Marguin, L. Chem. Phys. Lett. 1972, 17, 269.
(16) Fajardo, M. E.; Tam, S. 53rd Ohio State University International
Symposium on Molecular Spectroscopy, FC05, Columbus, OH, 1998.
(39) Zeng, Y. P.; Sharpe, S. W.; Shin, S. K.; Wittig, C.; Beaudet, R. A.
J. Chem. Phys. 1992, 97, 5392.
(40) Rice, J. K.; Lovas, F. J.; Fraser, G. T.; Suenram, R. D. J. Chem.
Phys. 1995, 103, 1263.
(41) Sharpe, S. W.; Zeng, Y. P.; Wittig, C.; Beaudet, R. A. J. Chem.
Phys. 1990, 92, 943.
(42) Momose, T.; Miki, M.; Uchida, M.; Shimizu, T.; Yoshizawa, I.;
Shida, T. J. Chem. Phys. 1995, 103, 1400.
(43) Sogoshi, N.; Wakabayashi, T.; Momose, T.; Shida, T. J. Phys.
Chem. A 1997, 101, 522.
(44) Fan, Y. B.; Randall, K. L.; Donaldson, D. J. J. Chem. Phys. 1993,
98, 4700.
(17) Guasti, R.; Schettino, V.; Brigot, N. Chem. Phys. 1978, 34, 391.
(18) Fourati, N.; Silvi, B.; Perchard, J. P. J. Chem. Phys. 1984, 81, 4737.
(19) Ayers, G. P.; Pullin, A. D. E. Spectrochim. Acta A 1976, 32, 1629.
(20) Jaques, C.; Valachovic, L.; Ionov, S.; B o¨ mer, E.; Wen, Y.; Segall,
J.; Wittig, C. J. Chem. Soc., Faraday Trans. 1993, 89, 1419.
21) Alagia, M.; Balucani, N.; Casavecchia, P.; Stranges, D.; Volpi, G.
G. J. Chem. Phys. 1993, 98, 8341.
22) Loomis, R. A.; Lester, M. I. Annu. ReV. Phys. Chem. 1997, 48,
43.
23) Huber, K. P.; Herzberg, G. Molecular Spectra and Molecular
Structure. IV. Constants of Diatomic Molecules; Van Nostrand: New York,
(
(
6
(
1
(45) Herzberg, G. Molecular Spectra and Molecular Structure. I. Spectra
of Diatomic Molecules, 2nd ed.; Van Nostrand: New York, 1950.
(46) Jouvet, C.; Soep, B. Chem. Phys. Lett. 1983, 96, 426.
(47) Jouvet, C.; Soep, B. J. Chem. Phys. 1984, 80, 2229.
(48) Buelow, S.; Radhakrishnan, G.; Catanzarite, J.; Wittig, C. J. Chem.
Phys. 1985, 83, 444.
1
340.
R. Chem. Phys. 1998, 231, 245.
(
27) Alagia, M.; Balucani, N.; Casavecchia, P.; Stranges, D.; Volpi, G.
(49) Radhakrishnan, G.; Buelow, S.; Wittig, C. J. Chem. Phys. 1986,
84, 727.
(50) Buelow, S.; Noble, M.; Radhakrishnan, G.; Reiseler, H.; Wittig,
C.; Hancock, G. J. Phys. Chem. 1986, 90, 1015.
G. J. Chem. Soc., Faraday Trans. 1995, 91, 575.
28) Chen, Y.; Hoffmann, G.; Oh, D.; Wittig, C. Chem. Phys. Lett. 1989,
59, 426.
(
1