Excitation laser fluence effects on the Ba ... FCH3 + hv τ BaF + CH3 intracluster reaction

Article abstract of DOI:10.1016/S0009-2614(01)01481-6

In this work, results on the Ba...FCH₃ + hv → BaF (Ba^*) + CH₃ (CH₃F) reaction obtained under high laser field conditions are reported. We have found that the two-photodissociation channel yields, i.e. the reactive BaF and non-reactive Ba^* products, exhibited opposite behaviour depending upon the laser fluence. Whereas the BaF yield rises as the laser fluence is increased, the Ba^* yield decreases over the same laser fluence range. In other words, the product (Ba^*)/(BaF) branching ratio changes significantly from low to high laser fluence in such a way that the (photoinitiated) Ba...FCH₃ → BaF + CH₃ reaction may be controlled by changing the intensity of the excitation laser field.

Full text of DOI:10.1016/S0009-2614(01)01481-6

6
February 2002
Chemical Physics Letters 352 (2002) 369–374
www.elsevier.com/locate/cplett
Excitation laser fluence effects on
the Ba ꢀꢀꢀFCH þhm !BaF þCH intracluster reaction
3
3
1
*
A. Gonz ꢀa lez Ure n~ a , K. Gasmi, J. Jim ꢀe nez, R.F. Lobo
a
Unidad de L aꢀ seres y Haces Moleculares, Instituto Pluridisciplinar, Universidad Complutense de Madrid, P Juan XXIII-1°,
8040 Madrid, Spain
2
Received 18 October 2001; in final form 4 December 2001
Abstract
Ã
In this work, results on the Ba ꢀꢀꢀFCH
3
þhm !BaF ðBa ÞþCH
3
ðCH FÞreaction obtained under high laser field
3
conditions are reported. We have found that the two-photodissociation channel yields, i.e. the reactive BaF and non-
Ã
reactive Ba products, exhibited opposite behaviour depending upon the laser fluence. Whereas the BaF yield rises as
Ã
the laser fluence is increased, the Ba yield decreases over the same laser fluence range. In other words, the product
Ã
ðBa Þ/(BaF) branching ratio changes significantly from low to high laser fluence in such a way that the (photoinitiated)
Ba ꢀꢀꢀFCH
3
!BaF þCH
3
reaction may be controlled by changing the intensity of the excitation laser field. Ó 2002
Elsevier Science B.V. All rights reserved.
1
. Introduction
actant molecule is transformed to several prod-
uct molecules, two types of control have been
applied. Namely, the ‘two-beam interference
control’ [4–6] and the ‘two-pulse time delay
control’ [3,7,8]. While in the former case control
is achieved by quantum interference between
different reaction pathways relating the same
initial and final states, the latter method uses a
variable time delay between two extremely short
light pulses.
Laser control of chemical reactions is cur-
rently attracting a considerable attention from
both theoretical and experimental point of view
[
1–3]. Among several schemes to implement the
control of a chemical reaction one of the most
promising is based on quantum interference
control mostly applied to unimolecular reactions.
For this type of processes, in which a single re-
In the present work, we report on the control
of the (photoinitiated) Ba ꢀꢀꢀFCH þhm !BaFþ
3
*
Corresponding author. Fax: +34-1-39-43265.
E-mail addresses: laseres@eucmax.sim.ucm.es, laseres@
3
CH intracluster reaction achieved by changing
the fluence of the excitation laser. Pioneering
work on photofragmentation of FeðH
pluri.ucm.es (A. Gonz a´ lez Ure n˜ a).
1
Present address: Departamento de Fisica, Facultade de
Ciencias e Tecnologia, Universidade Nova de Lisboa, Quinta
da Torre 28025, Monte de Caparica, Lisboa, Portugal.
þ
2
OÞ clus-
n
ters ðn ¼1–9Þas a function of the laser fluence at
different wavelengths was carried out by Mest-
0
009-2614/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved.
PII: S0009-2614(01)01481-6

370
A. Gonz ꢀa lez Ure ~n a et al. / Chemical Physics Letters 352 (2002) 369–374
dagh and co-workers [9]. Linear dependence of
ꢁ
2
fluences up to 100 mJ=cm . Here the two photo-
2
Ba ꢀꢀꢀFCH
3
þhmpump
cross-sections ð0:05 A Þwas always found even at
%BaF þCH3
!
ðk ¼617:7 nmÞ
fragmentation channels i.e., reactive BaF and
Ã
Ã
non-reactive Ba , yields exhibited opposite be-
haviour depending upon the laser fluence. Indeed,
whereas the BaF yield increases with increasing
&Ba þCH F
3
BaF þhmprobe ðk ¼266 nmÞ!BaFþ
Ba þhmprobe ðk ¼266 nmÞ!Baþ
Ã
laser fluence, the Ba yield decreases over the
same laser fluence range. In other words, the
Ã
Ã
product [Ba ]/[BaF] branching ratio changes sig-
þ
The BaF signal is produced by direct one
nificantly from low to high laser fluence condi-
tions. It is emphasised that we used nanosecond
laser pulses being the excitation laser fluence the
only adjustable variable. The simplicity of this
scheme could also be of interest for practical
applications.
photon ionisation of BaF product absorption of
the 266 nm photon, as demonstrated in previous
work (see for example Fig. 2 of [12]). With respect
Ã
to the Ba product, direct one photon ionisation of
2
. Experimental
A detailed description of the molecular beam
apparatus is given elsewhere [10,11] and only a
brief description will be presented here. The
weakly bound complex Ba ꢀꢀꢀFCH
3
is produced in
a laser vapourisation source followed by super-
sonic expansion, using a gas mixture of He with a
1
3
0% of CH F. The molecular beam is then probed
in a different chamber using a laser ionisation
coupled with a linear time-of-flight mass spec-
trometer.
Two lasers are used for the probing of the
molecular beam. A tunable laser induces the re-
action in the Ba ꢀꢀꢀFCH
while a second laser ionises the complex and the
photofragmentation products. An
3
weakly bound complex
Ba ꢀꢀꢀFCH
3
infinity-XPO laser with a bandwidth of typical
À1
5
cm and a pulse duration of 3.5 ns is used for
excitation of the complex, while the fourth har-
monic of a Continuum NY80 Nd:YAG laser (266
nm, pulse duration 7 ns) is used for ionisation.
The excitation laser arrives about 7–10 ns earlier
to the beam-laser interaction region than the
ionisation laser. This is achieved using an elec-
tronic delay generator provided by the infinity-
XPO laser.
Fig. 1. Energy level diagram of the Ba þCH
3
F system. The left
column shows the Ba electronic energy levels and the right one
the BaF energy levels. The centre column displays an approx-
imate location of the Ba ꢀꢀꢀFCH
3
electronic states including the
ÃÃ
The pump and probe scheme used in the present
experiment is the following:
hypothetical ðBa ꢀꢀꢀFCH Þ Rydberg state. See text for com-
3
ments.

A. Gonz ꢀa lez Ure ~n a et al. / Chemical Physics Letters 352 (2002) 369–374
371
Ba ground state is not possible at 266 nm, so only
Ã
3. Results and discussion
metastable Ba product can be ionised. See Fig. 1
for an energy diagram of the system here investi-
gated. Under the present experimental conditions
the 266 nm probing laser intensity was always kept
at a very low intensity to guarantee that two-
photon absorption process was, if any, a very
minor effect. It should be recalled that the pump-
Fig. 2 (solid line) shows a time-of-flight mass
spectrum of the Ba þFCH system when the
3
fundamental output of the Nd:YAG is used for
vapourisation and the species in the beam are
ionised with the 266-nm radiation. A significant
depletion of the monomer signal is observed when
the excitation OPO laser is tuned to 617.7 nm and
allowed to enter into the detection chamber (da-
shed line). It should be indicated that no depletion
is observed for other species contained in the
beam, indicating that these should probably ab-
sorb at different wavelengths. Furthermore the
enhancement of both Ba and BaF signals is clearly
noticed due to complex photofragmentation (see
[12,13] for further details).
ing laser wavelength excites the A
X electronic
complex only, i.e., it
transition of the Ba ꢀꢀꢀFCH
3
does not affect the free Ba atoms present in the
beam expansion at all (see [11] for details).
The experiment consisted of measuring simul-
þ
þ
taneously the BaF and Ba signals at two dif-
ferent conditions: (a) using UV photons
ðk ¼266 nmÞand (b) using first a pump XPO laser
ðk ¼617:7 nmÞ and then, 10 ns later, the UV
Ã
Ã
ðk ¼266 nmÞlaser. The BaF and Ba action sig-
As for the Ba signal, it should be emphasised
Ã
nals were estimated from their, (b) minus (a) sig-
nals, respectively. This procedure was then
repeated at different XPO laser fluences. The latter
was modified by using a high energy variable at-
tenuator (Newport mod. 935-5-OPT). This proce-
dure ensured that the focus and optical features of
the laser beam remained unchanged.
that no Ba ion signal was observed with the pump
laser only, even though higher fluences were
employed. Thus, taking into consideration the
energetic diagram, one can conclude that the non-
reactive photofragmentation channel here investi-
gated, with the k ¼266 nm probing, corresponds
1
3
to metastable BaðD ; PÞstates.
1
Fig. 2. Solid line: time-of-flight mass spectra of the Ba þCH
3
F system. Strong depletion of the Ba ꢀꢀꢀFCH
3
complex is observed when
the OPO laser is tuned to 617.7 nm. The enhancement of both Ba and BaF signals is also clearly noticed as a result of the complex
photofragmentation.

372
A. Gonz ꢀa lez Ure ~n a et al. / Chemical Physics Letters 352 (2002) 369–374
The individual photofragmentation channel
yields are displayed on Figs. 3 and 4 as a function
Ã
of the laser fluence. Both DBa and DBaF signals
Ã
were obtained using the same procedure. The DBa
was calculated by subtracting the Ba (UV) signal
Ã
Ã
from the Ba (UV + VIS). Here UV and VIS stand
for the signal measured when the UV
ðk ¼266 nmÞand XPO ðk ¼617:7 nmÞlaser were
employed, respectively. The most interesting fea-
ture is the laser fluence dependence of both reac-
Ã
tive, DBaF, and non-reactive, DBa , yields. While
Ã
DBa increases reaching a maximum around
2
1
:9 mJ=cm with a subsequent decline beyond this
fluence value, the DBaF always increases over the
Ã
studied laser fluence range. Clearly, the ½Ba ꢁ/[BaF]
branching ratio as depicted in Fig. 5 is not con-
Fig. 4. Same presentation as in Fig. 3, but now for the BaF
channel. Notice the opposite behaviour of the present BaF
Ã
channel when compared with previous Ba data, in particular
2
for laser fluences beyond ꢂ2 mJ=cm .
Ã
Fig. 3. Ba photofragmentation yield ratios as a function of the
Ã
Ã
laser fluence. (a) Ba signal with the UV laser, (b) Ba signal
when VIS pump laser operated 10 ns before the probe UV laser
Ã
Ã
arrives to the ionisation region. DBa signal obtained by sub-
tracting (a) from (b) signal. In all cases the same relative units
are used.
Fig. 5. Laser fluence dependence of the ½Ba ꢁ/[BaF] branching
ratio. Points: present experimental results; dotted line is just a
guide to the eye.

A. Gonz ꢀa lez Ure ~n a et al. / Chemical Physics Letters 352 (2002) 369–374
373
stant but depends of the laser field employed to
induce the reaction.
For an understanding of the present results it
should be remarked that as the complex is irradi-
ated with more and more photons a coherent su-
perposition of both ground and excited states is
probably suitable to achieve the control of the
intracluster reaction by changing the laser fluence.
One of the crucial questions concerning the in-
terpretation of present results is whether or not a
multiphoton process is controlling the fluence de-
pendence of the different reaction channels, be-
cause if such a process was present it could open
new routes to turn on the reaction and produce the
BaF. In this view, the enhancement of the BaF
yield, as the laser fluence increases, would reflect
the predominance of this new route rather than the
(favoured) faster photofragmentation channel due
to the shortening of the (excited complex) lifetime
as the Rabi frequency increases. This alternative is
further discussed below.
created. In addition, the Ba ꢀꢀꢀFCH decomposes
3
via two channels, e.g.
2
Ã
1
Ã
Ba þFCH
and consequently the lifetime of the excited
3
ðBa ꢀꢀꢀFCH
3
Þ!BaF þCH
3
Ã
ðBa ꢀꢀꢀFCH
3
Þ may be influenced not only by the
two channel photofragmentation dynamics but
also by the intense laser field conditions. In this
Ã
view, the actual ðBa ꢀꢀꢀFCH Þ concentration is
3
not longer constant but follows a periodical time
dependent functionality controlled by the Rabi
frequency and so by the laser fluence F. Indeed
Let us consider for example the absorption of
two-photon by the Ba ꢀꢀꢀFCH complex. A two-
3
À1
photon mechanism would put 36:283 cm into
the system. As can be seen in the energetic diagram
Ã
2
½
Ba ꢀꢀꢀFCH ꢁwill vary as ꢂsin ðX=2Þt, in which
3
1
=2
þ
the Rabi frequency, X, is proportional to ꢂF
of Fig. 1 this would form a Ba ꢀꢀꢀCH F ion
3
1
=2
(
hereafter we will write X ¼aF , a being a laser
surrounded by the Rydberg electron in a different
orbital. In principle this state might evolve to form
fluence independent parameter). The coherent su-
perposition of complex excited and ground states
leads to a shortening of its lifetime which may
favour the faster photodissociation channels as F,
and so X, increases. As a result, photofragmenta-
tion products could be selectively formed as long
as the reaction time is comparable with the Rabi
period 1=X. It is out of the scope of the present
Letter to describe the calculation of the transition
moment and the Rabi frequency for the studied
transition as a function of the laser fluence.
However, as it will be published elsewhere [14], the
Rabi period lies about the picosecond time domain
over the low laser fluence range used in the present
experiment, reaching the sub-picosecond regime at
high fluences [14]. To the best of our knowledge
there is no experimental information about the
reaction time associated to the non-reactive
þ
BaF ꢀꢀꢀCH underneath the Rydberg state and
3
Ã
Ã
subsequently BaF þCH . Thus, this BaF could
3
then be easily excited by the probe ðk ¼266 nmÞ
photon.
The Ba ꢀꢀꢀFCH absorption at 618 nm was
3
studied in [16] (see, for example, its Section 3.1),
þ
monitoring the strong peak of the Ba ꢀꢀꢀFCH
3
signal caused by the simultaneous absorption of
the one pump photon (k ¼618 nm) and one probe
photon ðk ¼400 mÞ. Typical values of the (120 fs)
pump and probe laser fluences were 0.5 and
2
1 mJ=cm , respectively. For the pump laser the
used fluence of 0:5 mJ=cm and 120 fs pulse du-
2
9
2
ration gave a laser intensity of ca. 4 ꢃ10 W=cm .
As reported in [16] a linear dependence was always
found for each photon over a range of 4 in laser
fluence.
Ã
channel yielding Ba products. However, the re-
Obviously, the comparison between the possible
two-photon regime induced in the present nano-
second experiment and previous [16] femtosecond
experiment must integrate the pulse duration.
Taking the latter into account one can show that
the two-photon absorption process would there-
fore be much less probable in the present experi-
ment than in the previously reported femtosecond
study [16].
action time for the BaF channel has been mea-
sured to be about 270 fs by the femtosecond pump
and probe technique [15,16]. Thus one can assume
that the reaction time for the non-reactive channel
could well be of the order of picoseconds as it is
the (usual) case in photodissociation dynamics of
van der Waals molecules. Bearing all this in mind,
it seems that the employed laser conditions are

374
A. Gonz ꢀa lez Ure ~n a et al. / Chemical Physics Letters 352 (2002) 369–374
On the other hand, the possibility that such
ðBa ꢀꢀꢀFCH
laser fluence as the only adjustable parameter to
achieve the control of the photoinitiated reaction.
Furthermore the application to a bimolecular re-
action by clustering the reactants in a van der
Waals should also be remarked. The simplicity of
this scheme could be of interest when one consid-
ers its potentiality for practical applications.
ÃÃ
3
Þ
Rydberg state leads into
þ
þ
BaF þCH
3
through Ba ꢀꢀꢀFCH
3
!BaF þCH
3
reaction underneath of the Rydberg electron seems
very unlikely. The ðBa ꢀꢀꢀFCH
þ
3
Þ
seems to be a
ionisation
potential (see Fig. 4 in [11]) was measured, the
very stable ion. When the Ba ꢀꢀꢀFCH
3
þ
Ba ꢀꢀꢀFCH signal increased smoothly with the
3
laser ionisation energy even though the
BaF þCH
þ
3
channel was energetically opened
see energetic diagram), which can only be inter-
Acknowledgements
(
þ
preted as that the Ba ꢀꢀꢀFCH fragmentation into
The authors acknowledge helpful discussions
with J.M. Mestdagh on the results presented in
this work. This work received financial support
from DGES of Spain Grant PB97-0272 and the
Ram oꢀ n Areces Foundation.
3
þ
BaF þCH
3
should be a very minor channel, at
least at present excitation energies.
In summary, taking all the above considerations
into account we conclude that multiphoton pro-
cesses can be ruled out under the present experi-
mental conditions.
References
[
[
[
1] R.N. Zare, Science 279 (1998) 1875.
2] S.A. Rice, Nature 409 (2000) 422.
4
. Concluding remarks
3] A. Assion, T. Baumert, M. Bergt, T. Brixner, B. Kiefer,
V. Seyfried, M. Strehle, G. Gerber, Science 282 (1998)
The most interesting finding of the present work
is that the ðBa ꢀꢀꢀFCHÞ two-photodissociation
Ã
9
[4] M. Shapiro, P.J. Brumer, J. Chem. Phys. 84 (1986) 4103.
19.
channel yields, i.e. the reactive BaF and non-re-
Ã
[
[
[
5] M. Shapiro, P.J. Brumer, J. Chem. Phys. 97 (1992)
259.
active Ba products, exhibited opposite behaviour
depending upon the laser fluence. The BaF yield
rises as the laser fluence is increased. However, the
6
6] L.-C. Zhu, V. Kleiman, X.-N. Li, S.P. Lu, K. Trentelman,
R.J. Gordon, Science 270 (1995) 77.
7] D. Tannor, S.A. Rice, J. Chem. Phys. 83 (1985) 5013.
Ã
Ba yield reached a maximum at low fluences with
a subsequent decline over the high laser fluence
range. This opposite behaviour was attributed to a
different reaction times associated to each indi-
vidual photofragmentation channel of the
[8] T. Baumert, M. Grosser, R. Thalweiser, G. Gerber, Phys.
Rev. Lett. 67 (1991) 3753.
[9] L. Dukan, L. del Fabbro, P. Pradel, O. Sublemontier, J.M.
Mestdagh, J.P. Visticot, Eur. Phys. J. D 3 (1998) 257.
[
10] S. Skowronek, R. Pereira, A. Gonz ꢀa lez Ure n~ a, J. Chem.
Ba ꢀꢀꢀFCH
3
. In this picture, the faster channel, the
Phys. 107 (1997) 1668.
[11] S. Skowronek, R. Pereira, A. Gonz ꢀa lez Ure n~ a, J. Phys.
Chem. 101 (1997) 7468 (Special issue on Stereodynamics).
BaF product, would be favoured with respect to
Ã
the slower one, the Ba product, as the excited
complex lifetime becomes shorter as the Rabi fre-
quency increases. This would justify the observa-
[
12] S. Skowronek, A. Gonz ꢀa lez Ure n~ a, in: R. Campargue
Ed.), Atomic and Molecular Beams: The State of the Art,
(
Springer, Berlin, 2000, p. 353.
Ã
tion that the product ðBa Þ/(BaF) branching ratio
[
13] S. Skowronek, A. Gonz ꢀa lez Ure n~ a, Prog. React. Kinet.
Mech. 24 (1999) 101.
changes significantly from low to high laser fluence
way that the (photoinitiated)
[
14] A. Gonz ꢀa lez Ure n~ a et al., in preparation.
15] V. Stert, P. Farmanara, W. Radloff, F. Noack, S. Skow-
ronek, J. Jimenez, A. Gonz ꢀa lez Ure n~ a, Phys. Rev. A 59 (3)
in such
Ba ꢀꢀꢀFCH
a
[
3
þhm !BaF þCH reaction could be
3
controlled.
A key feature of the present investigation is the
use of nanosecond laser pulses with the excitation
(
16] P. Farmanara, V. Stert, W. Radloff, S. Skowronek, A.
1999) 1727.
[
Gonz ꢀa lez Ure n~ a, Chem. Phys. Lett. 304 (1999) 127.