IR multiple-photon dissociation of Br(CF2)4COI molecules by excitation via C=O (1794 cm-1) stretching and skeleton (961 cm-1) vibrations

Article abstract of DOI:10.1016/S0009-2614(98)00100-6

Dissociation of Br(CF₂)₄COI molecules resulting from infrared multiple-photon excitation (IR-MPE) by the second (via C=O stretch) and the first (via skeleton vibrations) harmonics of TEA CO₂ laser was studied. It was found that in the two cases of IR-MPE, dissociation occurs via two distinct channels with the formation of C₂F₄ and Br(CF₂)₄I molecules as final products. At the same time, the relative ratio of these final products was found to depend on the channel of molecular excitation.

Full text of DOI:10.1016/S0009-2614(98)00100-6

10 April 1998
Ž
.
Chemical Physics Letters 286 1998 277–283
ž
/
IR multiple-photon dissociation of Br CF2 4COI molecules
y1
ž
/
by excitation via C-O 1794 cm
stretching and skeleton
/
vibrations
y1
ž
961 cm
A.V. Dem’yanenko, E.A. Ryabov, V.S. Letokhov
Institute of Spectroscopy, Russian Academy of Sciences, 142092 Troitzk, Moscow region, Russia
Received 15 October 1997; in final form 5 January 1998
Abstract
Dissociation of Br CF2 4COI molecules resulting from infrared multiple-photon excitation IR-MPE by the second via
Ž
.
Ž
.
Ž
.
Ž
.
C-O stretch and the first via skeleton vibrations harmonics of TEA CO2 laser was studied. It was found that in the two
Ž
.
cases of IR-MPE, dissociation occurs via two distinct channels with the formation of C2F4 and Br CF2 4I molecules as final
products. At the same time, the relative ratio of these final products was found to depend on the channel of molecular
excitation. q 1998 Elsevier Science B.V.
1. Introduction
tivity of IR-MPD has not yet been realized. It is
assumed that the main problem in achieving mode-
selectivity is caused by intramolecular vibrational
Ž
.
Infrared multiple-photon excitation IR-MPE has
been used for quite a long time as a method of
excitation of ground-state molecules to high lying
vibrational states, which leads subsequently to uni-
Ž
.
Ž
relaxation IVR which results in fast pico- and
.
sub-picosecond time scale redistribution of the vi-
brational energy from a resonant mode over all
Ž
w
x.
Ž
w x.
molecular dissociation see e.g. 1–3 . One of the
advantages of IR-MPE is its collisionless nature.
This property is a base for the selectivity of pro-
cesses induced by IR laser excitation. Two types of
selectivity might be considered. The first type is
intermolecular selectivity. Laser isotope separation
molecular modes see recent review 9 . Therefore,
a possible way to overcome the problem of IVR and
to realize intramolecular selectivity is to use ultra-
w x
short laser pulses 6 .
As for site-selectivity there is one more way to
realize it. The idea is based on the use of prolate,
w
x
w
x
Ž
1–4 by IR-MPE is an example of realization of this
chain-like molecules 10 , where a transient non-
.
type of selectivity. The second type is intramolecular
selectivity which includes processes selective with
statistical distribution of vibrational energy might be
expected. In this case, unimolecular decay might
Ž
.
Ž
respect to either some specific bond mode or to
happen in some part of the molecule probably near
Ž
.
.
some group of bonds part of the molecule . In the
the mode under excitation before redistribution of
first case one can speak about ‘mode-selectivity’
the absorbed energy all over the molecule. This
raises the natural question of how fast IR-MPE must
be to realize the above situation. Vibrational energy
w
x
5,6 and, in the second case, about site- or regio-
w
x
selectivity 7,8 . As far as we know, the mode-selec-
0009-2614r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S0009-2614 98 00100-6

(
)
278
A.V. Dem’yanenko et al.rChemical Physics Letters 286 1998 277–283
transfer in a one-dimensional chain of diatomic
bution in the course of regio-selective excitation of
CO bonds in the neighborhood of these atoms. If the
Br atom close to the excited CO bond site detaches
sooner than the more remote and having a lower
bond energy I atom, this would point to a non-statis-
tical vibrational distribution in the case of suffi-
ciently fast IR-MPE. These molecules also have
delocalized skeleton vibrations in the 10.4 mm re-
gion which provides a comparison of the results of
IR-MPE via two different channels. Theoretical esti-
w
x
molecules was analyzed in Ref. 11 . At certain
vibrational energy levels a non-statistical vibrational
distribution could persist even on a nanosecond time
scale. Local vibrational equilibrium is established in
this model in a time scale of some tens of picosec-
onds.
Ž
In our search for non-statistical effects including
.
site-selectivity in unimolecular decay of IR-MP ex-
cited molecules, we have chosen two isomers
Ž
.
Ž
.
w x
I CF2 4COBr and Br CF2 4COI as examples of pro-
late molecules. These molecules have strong isolated
mations, made in Ref. 12 , have shown that a non-
Ž
.
statistical transient vibrational distribution may in-
deed occur in molecules of considered prolate-type
in real experimental conditions.
Ž
.
absorption bands in the 5.5 mm region see Fig. 1a ,
related to the C-O stretch bond for which specific
site excitation is possible. This bond is placed re-
spectively in the vicinity of Br or I atoms. Insofar as
the bond energies of these atoms are different, one
can try to observe a non-statistical vibrational distri-
This Letter is devoted to the first results of studies
of IR-MP dissociation of Br CF2 4COI molecules in
static cell conditions. The molecules were excited
both via CO stretching mode 1794 cm
Ž
.
y1
Ž
.
and via
Ž .
Ž
.
Ž
. Ž .
Ž
.
Ž .
Ž .
Fig. 1. IR absorption spectra of a Br CF2 4COI gas ps0.8 Torr, Ls12 cm . b Br CF2 4COI before 1 and after 2 irradiation with
Ž
.
Ž . Ž .
Ž
.
second-harmonic radiation of TEA CO2 laser ps0.8 Torr .
c
IR absorption spectrum of C2 F4 ps0.6 Torr .
d IR absorption
Ž
.
Ž
.
spectrum attributed to Br CF2 4 I ps0.3 Torr .

(
)
A.V. Dem’yanenko et al.rChemical Physics Letters 286 1998 277–283
279
y1
Ž
.
delocalized skeleton vibration 961 cm . The re-
sults obtained are presented below.
than 70% of the pulse energy. The AgGaSe2 crystal
was used as a non-linear element to generate the
second harmonic.
The CO2 laser radiation was collimated with a
system of mirrors and NaCl lenses and directed to
the AgGaSe2 crystal. At the output of the crystal the
radiation with the 2vlas frequency was separated
2. Experimental set-up and procedure of mea-
surements
Ž
.
Ž
.
Ž .
Br CF2 4COI was synthesized purity 98% , by
Deev et al. from the Perm branch of the State
Institute of Applied Chemistry. Before the experi-
ments the substance was purified by trap-to-trap
low-temperature vacuum distillation. The saturation
pressure vapor of the compound was 2.1 Torr at
ts21"28C. Fig. 1a presents the IR spectrum of
from the CO2 laser radiation vlas with a LiF plate,
Ž
.
focused by NaCl lens fs14 cm and directed into
Ž
.
an aluminum gas cell Ls12 cm, B1.2 cm with
NaCl windows. Because of comparatively low en-
Ž
.
ergy of the second harmonic radiation 15–20 mJ
we used focused irradiation geometry in order to
obtain a high enough dissociation yield. For simplifi-
cation of the comparison of the results the same
geometry was used for IR-MPD with the first har-
monic of the 12 CO2 laser.
Ž
.
Br CF2 4COI in the spectral region of 1900–400
cmy1. The most intensive bands are 1794, 1213.5,
1181.4, 1105.1, 961, and 656.7 cmy1. As far as we
know, the IR spectrum of this compound and assign-
ment of the vibrations are absent in the literature.
The irradiation procedure and product analysis
were made as follows. The gas cell was filled with
Ž
.
Nevertheless, taking into account similar molecules
0.8 Torr of Br CF2 4COI gas. The absorption spec-
trum of this compound was measured before and
y1
w
x
13 , the band with 1794 cm
biguously assigned to the stretching vibration of the
frequency is unam-
Ž
.
after the irradiation. IR spectra of Br CF2 4COI and
dissociation products in the region of 600–1900
cmy1 were recorded by SPECORD-M82 spectro-
photometer. The spectra were registered in a digital
form which made their computer processing possi-
ble. The spectrum of irradiated compound along with
the spectrum of parent gas normalized to CO band
intensity are shown in Fig. 1b.
Preliminary studies of the dependence of the dis-
sociation yield on the number of laser pulses show
that after consumption of about 40% of the parent
molecules the dissociation slows down due to satura-
tion effects. Thus, in the measurements, the maxi-
mum attainable consumption did not exceed 30–35%.
This was achieved by appropriate choice of the
number of laser pulses that was varied, depending on
laser intensity, in the range of 15–900 shots.
Ž
.
bond –C-O nC - O . The frequency nC - falls
within a tuning range of the double-frequency radia-
tion of the TEA 13CO2 laser. Two absorption bands,
both related to skeleton vibrations, fall into opera-
tional region of TEA 12 CO2 laser — the red wing of
band ns1105 cmy1 and band ns961 cmy1. For
reasons presented below, we chose a comparatively
weak band in the 961 cmy1 region for IR-MPE with
the first harmonics of CO2 laser. The assignment of
w
x
fluorocarbons bands in this region 14–16 makes it
possible to suggest that this band contains a certain
percentage of C5 stretching skeleton vibration
Ž
.
Ž
.
nC — C . Thus, Br CF2 4COI molecules can be ex-
cited in two ways: through the vibration nC - O by
radiation with 2vlas and through nC — C vibration by
radiation with vlas
.
Ž
.
Ž
.
Br CF2 4COI molecules were dissociated using
The consumption of Br CF2 4COI was deter-
mined from the IR spectra measured before and after
irradiation. Then the parameter bG , the fraction of
molecules in the gas cell dissociated in a pulse, was
Ž
the second-harmonic radiation of 10P18 line vlas
s
898 cmy1 of the TEA CO2 laser that operated on
.
the mixture 13CO2rN2rHe 2:1:9 , and the first
Ž
.
harmonic of 10P8 vlas s954 cmy1 line of the
calculated the notations and transformed formula
Ž
.
Ž
are taken from Ref. 1 : bGs1y I0rI1 1r N, where
b is the dissociation yield, N is the number of
irradiation pulses, I0 and I1 are the integrated ab-
sorption coefficients of band nC - O before and after
irradiation, G is the ratio of the exposed gas volume
w x.
Ž
.
CO2 laser that operated on the mixture
12
Ž
.
CO2rN2rHe 2:1:9 . In both cases the radiation
pulse was standard in shape, it had a peak with its
halfwidth of about 80 ns and a ‘tail’ with total
duration of about 1 ms. The peak contained more

(
)
280
A.V. Dem’yanenko et al.rChemical Physics Letters 286 1998 277–283
Ž .
–CF3 group, and for cyclic compounds C4 F8, C3F6
13–15 are also absent. Comparison of the spectrum
to the total volume of the cell. The yields of the
dissociation products Yi were calculated in the same
manner.
In the course of this experiment we did not calcu-
late the dissociation yield b but just estimated it
because, for focused irradiation, the fluence F at the
2vlas and vlas frequencies varied by several times
along the whole cell.
w
x
Ž
.
with that of the I CF2 4 I molecule reveals similar
y1
Ž .
structure of strong bands in the 1200 cm region
w
x
related to skeleton vibrations 18 . In addition, the
spectrum of the IR-MPD products of the molecule
Ž
.
Ž
.
I CF2 4COBr after subtracting the C2 F4 spectrum
Ž
.
coincides with that of Br CF2 4COI.
On the basis of the above arguments we assume
Ž
.
that IR-MPD of the Br CF2 4COI molecule with
both 2vlas and vlas frequencies of CO2 laser gives
rise to only two final products detectable in the IR
3. Results and discussion
Ž
.
Ž .
After irradiation of Br CF2 4COI with either 2vlas
or vlas frequencies of CO2 laser radiation a pro-
nounced change of IR absorption spectrum was ob-
absorption spectrum: C2 F4 and Br CF2 4 I. Registra-
Ž
.
tion of other possible products CO, I2 Br2 was not
made due to the difficulty of this procedure. The
Ž
.
Ž
.
served see Fig. 1b , which points unambiguously to
formation of new products in the course of IR-MPD
of parent molecules. In this process new absorption
bands occurred. In addition, the intensity of some
amount of dissociated Br CF2 4COI is balanced with
that of resultant products. By taking into considera-
tion the results presented below, we can assume that
there are two channels of final product formation:
Ž
parent absorption bands 1213.5, 1181.4, 1105.1, and
2Br CF 4COI
Ž
.
2
656.7 cmy1 did not decrease as could be expected
.
Ž
.
due to consumption of parent compound but in
some cases increased substantially.
Absorption spectrum measured after irradiation
comprises both bands of the products of IR-MPD
and bands of the reduced amount of parent com-
pound. By ‘subtracting’ from this spectrum the one
measured before irradiation after a proper normal-
ization to CO band intensity we readily obtain a
spectrum of the products of IR-MPD. It was found
n"v™
Br CF 4 IqCO
1
Ž
.
Ž .
2
q
½
m"v™ 2C2 F4 qCOqBrqI
2
Ž .
We have referred absolute amounts of dissociation
products to their absorption spectra. For C2 F4 this
w
x
Ž
.
was made by using data from Ref. 17 , for Br CF2 4 I
by comparing its spectral intensity with the amount
of consumed parent gas for the spectra which have
no remarkable bands of C2 F4. The resultant accuracy
of determining the product yields was 10–20%.
Since the irradiation geometry with the second
and first harmonics was the same, we can directly
compare measured results. Fig. 2 shows the depen-
dencies of the bG parameter on laser fluence F for
irradiation with either 2vlas and vlas. It is evident
Ž
.
Ž
that some of the originated bands 1181, 1186, 1195,
1332, and 1342 cmy1 may be attributed to a C2 F4
.
w
x
molecule, whose spectrum is well known 17 . For
some irradiation conditions low fluencies of 2vlas
mainly C2 F4, without substantial impurities of other
products, was formed. This makes it possible to
Ž
.
Ž
.
extract pure C2 F4 spectrum see Fig. 1c and subtract
it from the spectrum of the products. Subsequent
from Fig. 2 that, for excitation with 2vlas, the
0.91"15%
Ž
.
dependence is nearly linear bGAF
which
Ž
w x
analysis shows that the resulting spectrum see Fig.
usually points 1 to high values of dissociation yield
close to saturation. It should be noted that the effi-
.
Ž
.
1d may be assigned to the molecule Br CF2 4 I. We
did not find this spectrum in the literature and our
conclusion is based on indirect evidence; however, a
number of considerations presented below allows us
to draw the above conclusion. First, we can exclude
molecules of the type X CF2 Y X,YsI,Br with
shorter chains n-4; spectra of these compounds are
well known 14,15 . The absorption bands which are
characteristic for double and triple C-C bonds, for
Ž
.
ciency of Br CF2 4COI MPD by second-harmonic
radiation is rather high. Our estimations show that
the dissociation yield can reach 100% when the
fluence is Fs3 Jrcm2. For the excitation with
vlas, the bG dependence is steeper: bGAF1.6"15%
.
Ž
.
Ž
.
n
Fig. 3 shows the dependencies of the yields of the
products for the two cases of IR-MPE. Even though
these dependencies are different, the C2 F4 yield
w
x
Ž
.

(
)
A.V. Dem’yanenko et al.rChemical Physics Letters 286 1998 277–283
281
formation of different ratios of final dissociation
products.
First, the observed difference can be explained by
the formation of different vibrational distributions of
molecules under excitation with vlas and 2vlas laser
frequencies — which is quite possible because exci-
Ž
tation conditions doubled photon energy, different
.
bands, etc. differ substantially. Then for the equal
total energy absorbed, in the first case skeleton
vibrations , the well-known bimodal distribution 1
may be formed, a small part of the molecules will be
excited very high above dissociation limits for both
Ž
.
w x
Ž
.
channels see Fig. 5a and will decay very rapidly in
comparison to a collisional deactivation time. In the
Fig. 2. The dependence of the fraction of dissociated molecules
bG on the fluence of second and first harmonic of CO2 laser
radiation.
Ž
.
grows sharper with F than that for Br CF2 4 I in
either case. For correct comparison of the yield ratio
of IR-MPD products, it is essential to relate the
results for the 2vlas irradiation with that for the vlas
one. The parameter we used to tie together the
results of both experiments was the fraction of
molecules dissociated per pulse — bG . Thus we
have compared the IR-MPD yields ratio for equal
Ž
.
consumption of parent molecules see Fig. 4 . From
this figure it can be seen that the yields ratio differs
by about an order of magnitude for IR-MPE via
Ž
different channels C-O stretch and skeleton vibra-
.
tions for the same consumption of the parent
molecules.
Thus, we have measured the dependencies of the
consumption bG and the yields of primary products
Y1 and Y2 on laser fluence in the course of IR-MPE
Ž .
via two channels: a stretching vibration nC - O
s
1794 cmy1
;
b
skeleton vibrations nC — C s961
Ž .
cmy1. The ratio of the yields of products of IR-MPD
Ž
.
of Br CF2 4COI molecule in a static cell was found
to depend substantially on the channel of excitation.
w
x
Similar results were obtained earlier 19 where it
Ž
.
was shown that when CF3 CF2 3COI molecule is
Ž
excited by the second harmonic of CO2 laser via
C-O bond , the primary MPD products are C4 F9 I
.
and CO, whereas in the case of excitation by CO2
Ž
.
Fig. 3. The dependence of yields of final products Br CF2 4 I and
Ž
.
laser pulse via skeleton vibration the products are
C4 F9 I, C2 F4 and CO.
Ž .
Ž . Ž
C2 F4 on the fluence of the 2vlas a and vlas radiation b . The
units of measurement of yields are arbitrary, but the same for both
.
There are a few possible reasons for the observed
products within each figure.

(
)
282
A.V. Dem’yanenko et al.rChemical Physics Letters 286 1998 277–283
Ž
.
the primary product Br CF2 4 I to C2 F4 at the vlas
frequency. This, however, may be possible only for
Ž
.
vibrationally excited Br CF2 4 I molecules. The
choice of the comparatively weak band nC — C s961
cmy1 for IR-MPD via skeleton vibrations was due to
necessity to avoid IR-MPD of the primary product
Ž
.
Br CF2 4 I. Special measurements of dependence of
the products yield ratio on the number of laser shots
vlas s958 cmy1 show that this ratio is constant
with accuracy of about 5%. Thus, the second mecha-
Ž
.
nism seems to be highly improbable.
Third, it is quite possible that the formation of
various MPD products after IR-MPE through the
modes nC - O and nC — C results from the non-statis-
Fig. 4. The dependence of final products yields ratio on the
Ž
.
tical decay of Br CF2 4COI. This might be the most
intriguing result since, in this case, the non-equi-
dissociation yield b of parent molecules for IR-MPE via C-O
bond nC - O s1794 cmy1 and skeleton vibrations nC — C s961
Ž
.
.
Ž
cmy1
.
Ž
.
librium intramolecular not intermolecular distribu-
tion is responsible for the observed results which, in
principle, is a main goal of our research. We cannot
assert it definitely because, among other things, pri-
mary products of IR-MPD are unknown. This is a
subject for further investigation.
Ž
.
other cases C-O stretch , a major portion of
molecules may be excited rather low as compared
Ž
.
with the dissociation limits see Fig. 5b and due to
low dissociation rate intermolecular collisions, after
the laser pulse is over, will influence substantially
the process of IR-MPD decreasing remarkably the
yield via higher dissociation channel. Although the
dissociation yield in the case of excitation at vlas is
4. Conclusion
The experiments performed have shown that com-
lower than that at 2vlas 2% at Fv, min s3 Jrcm2 as
Ž
Ž
.
Ž
.
pounds like X CF2 4COY X,YsI,Br can dissoci-
ate effectively when modes of carbonyl group as
well as skeleton vibrations are excited, respectively,
by the second- and first-harmonic radiation of TEA
2
.
against near 100% at F2 v, max s3 Jrcm , the ex-
cited molecules in the first case may have a higher
average vibrational energy. Thus, these molecules
will probably decompose via a high-energy channel
D2.
Ž
.
CO2 laser. The IR-MPD of Br CF2 4COI molecule
is possible through two different channels. The final
Second, the difference in MPD products may be,
in principle, caused by the subsequent dissociation of
Ž
.
observed MPD products are Br CF2 4 I, C2 F4, with
the channel of C2 F4 formation being realized at
higher F.
It was found that, although formation of these
products was observed for IR-MPD both with 2vlas
Ž
.
Ž
.
C-O stretch and vlas skeleton vibrations of CO2
laser radiation, the products yield ratio in the two
cases is quite different; that is to say the composition
of final products depends on molecular excitation.
Acknowledgements
Fig. 5. Two possible distributions of vibrationally excited
Special thanks are due to Prof. C.B. Moore for his
work to support this research program. This work
was partly supported by the Russian Foundation for
Ž .
molecules: a small part of molecules is excited very high above
Ž
. Ž .
b major portion of
dissociation limit bimodal distribution ,
molecules is excited rather low above dissociation limit.

(
)
A.V. Dem’yanenko et al.rChemical Physics Letters 286 1998 277–283
283
w x
8
V. Lopez, V. Fairen, S.M. Lederman, R.A. Marcus, J. Chem.
Basic Research and the Lawrence Berkeley Labora-
tory.
Ž
.
Phys. 84 1986 5459.
w x
Ž .
9
D.Y. Nesbitt, R. Field, J. Phys. Chem. 100 1996 12735.
w
x
10 V.S. Letokhov, Non-Linear Laser Chemistry with Multiple-
Ž
.
Photon Excitation Springer, Berlin, 1983 .
References
w
w
x
Ž
.
11 R. Bolton, S. Nordholm, Chem. Phys. 182 1994 263.
x
12 A. Demianenko, V. Letokhov, A. Makarov, E. Ryabov,
w x
1
Ž
.
V.N. Bagratashvili, V.S. Letokhov, A.A. Makarov, E.A.
Ryabov, Multiple Photon Infrared Photophysics and Photo-
Faraday Discuss. 102 1995 301.
w
w
x
Ž
.
13 D. Welti, Infrared Vapor Spectra Heyden–Sadtler, 1970 .
Ž
.
x
Ž
.
chemistry Harwood Academic Publ., Chur, 1985 .
14 D.C. Weiblen, in: J.H. Simons Ed. Fluorine Chemistry
w x
2
Ž .
Ž
.
C.D. Cantrell Ed. Multiple-Photon Excitation and Dissocia-
tion of Polyatomic Molecules Springer, Berlin, 1986 .
Academic Press, 1954 .
Ž
.
w
w
x
15 L.M. Sverdlov, M.A. Kovner, E.P. Kraynov, Vibrational
w x
Ž
.
Ž
.
3
w x
4
D.V. Lupo, M. Quack, Chem. Rev. 87 1987 181.
Spectra of Polyatomic Molecules Moscow, Nauka, 1970 .
x
Ž
.
J.L. Lyman, in: L.J. Radziemski, R.W. Solarz, J.A. Paisner
Eds. , Laser Spectroscopy and its Applications Dekker,
16 D.A.C. Compton, D.M. Rayner, J. Phys. Chem. 86 1982
Ž
.
Ž
1628.
.
w
w
x
Ž
.
New York, 1987 p. 417.
17 J.R. Nielsen, H.H. Claassen, J. Chem. Phys. 18 1950 812.
w x
Ž
.
x
5
w x
6
I. Oref, B.S. Rabinovitch, Acc. Chem. Res. 12 1979 166.
N. Bloembergen, A.H. Zewail, J. Phys. Chem. 88 1984
18 M. Hauptstein, C.S. Stoces, A.V. Grosse, J. Am. Chem. Soc.
Ž
.
Ž
.
74 1952 1974.
w
x
5459.
19 I.K. Bildinov, V.B. Laptev, E.A. Ryabov, L.M. Tumanova,
w x
7
Ž .
Ž
.
V. Lopez, R.A. Marcus, Chem. Phys. Lett. 93 1982 232.
Laser Chem. 12 1992 173.