A new method for determining absorption cross-sections out of initially
excited vibrational states
Amir Golan, Nitzan Mayorkas, Salman Rosenwaks, and Ilana Bar
Department of Physics, Ben-Gurion University, Beer Sheva 84105, Israel
A first experimental demonstration, combining the methods of vibrationally mediated
photodissociation (VMP) and ionization-loss stimulated Raman spectroscopy (ILSRS) for
measuring cross-sections for dissociation of vibrationally excited levels is reported. The
action spectrum obtained in the VMP of methylamine exhibits enhancement of the H
photofragment yield as a result of initial vibrational excitation and the ILSRS monitors the
fraction of molecules being excited. The partial cross-sections for H production out of the
sampled vibrational states and the extent of mode-selectivity were thus determined.
submitted to J. Chem. Phys.
I.
INTRODUCTION
The detailed investigation and control of molecular dynamics attracts considerable
interest. Studying its important aspects has contributed considerably to our understanding of
reactivity, allowing pursuing and in favorable cases even controlling unimolecular and
bimolecular transformations of polyatomics.1,2,3,4,5,6
Excitation of specific reagent
vibrational states is well suited for exploration of the detailed quantum molecular dynamics
and for driving transformations from particular quantum states of reactants to defined states
of the products. In half-collision processes specific quantum states of reactants, prepared via
one- or two-photon excitation, are subsequently excited to dissociative states and the ensuing
product states probed. If the excitation energy is localized and preserved in the excited bond
or mode until the photodissociation process occurs, bond- or mode-selectivity might occur.
Consequently, selectivity strongly depends on intramolecular energy redistribution (IVR)7,8
and on additional factors including the involved potential energy surfaces (PESs), the
dissociation energy and the possible dissociation channels.
Here we examine this issue for the prototypical seven-atom methylamine (CH3NH2)
molecule, which is the simplest primary amine exhibiting internal-rotation and inversion
motions, found in more complex molecules. Because of its importance, methylamine and its
deuterated isotopologues have been a target for numerous experimental and theoretical
studies
exploring
their
dissociation,9,10,11,12,13,14,15,16,17
and
structural
and
spectroscopic18,19,20,21,22,23,24,25,26 properties. Most of the previous dissociation studies have
been performed without characterizing the initial quantum state of the methylamine on the
~
~
ground, X , or the first electronic state, A . These studies include methylamine photolysis
with broadband radiation (194 – 244 nm),9 where four dissociation pathways were found: a
major channel corresponding to hydrogen atom release through N–H bond fission and three
minor channels for molecular hydrogen elimination from the amine and for C–H and C–N
bond breaking. In addition, the mass-selected translational energies of the fragments10 at 222
nm and the Rydberg-tagged hydrogen (deuterium) translational energies11 at a number of
wavelengths in the 203.0 - 236.2 nm range were studied. Very recently, the predissociation
~
for characterized final quantum states of methylamine on the A state12 and for ~ 243.1 nm
~
excitation of different initial vibrational states on the X state13,14 were investigated.
In
particular, it was interesting to find mode dependent dissociation and photoionization for
methylamine initially prepared in states in the regions of C-H and N-H fundamentals,
implying an incomplete IVR on the ns timescale in a seven atom molecule with a torsional
degree of freedom.
To obtain a more quantitative evaluation of the effect we combined, for the first time,
the methods of ionization-loss stimulated Raman spectroscopy (ILSRS)27,28 and vibrationally
mediated photodissociation (VMP)1,2 and applied them to methylamine in the C-H and N-H
stretching fundamentals region. From the ILSRS measurements we evaluated the stimulated
Raman excitation (SRE) efficacy and from the enhancement obtained for each state, relative
to ~ 243.135 nm dissociation from vibrationless ground state, we could calculate the partial
cross-sections for the dissociation channel leading to H fragments, out of different preexcited vibrational states.
II.
EXPERIMENT
The experiments were performed in a home built time-of-flight mass spectrometer
(TOFMS). A mixture of ~ 5 % methylamine seeded in argon was expanded through a pulsed
nozzle to produce a molecular beam.
beams (pump,
p,
and Stokes,
S)
This molecular beam was crossed by two visible laser
for SRE and a counterpropagating ultraviolet (UV) beam
for inducing the ILSRS and VMP processes, shown schematically in Fig. 1. A doubled Nd :
yttrium–aluminum–garnet (YAG) laser delivered the second harmonic (532 nm) with energy
of ~ 450 mJ in ~ 5 ns pulses at a frequency of 10 Hz and bandwidth of about 0.2 cm−1. The
532 nm output was split by a dichroic mirror to pump a dye laser and to act as the pump
beam in the SRE. 80 % of this energy pumped the dye laser, operating with DCM diluted in
methanol, to produce the tunable Stokes beam and the rest of the green beam was used as
p.
Excitation of the vibrational modes in the 2800 - 3400 cm-1 range was achieved by scanning
the dye laser at wavelengths in the 625 to 650 nm region. The vertically polarized SRE
beams (energies of ~ 30 mJ) were spatially and temporally overlapped and focused (75 cm
focal length plano convex lens) onto the TOFMS interaction region. The counterpropagating
UV beam was provided by the doubled output of a tunable dye laser (Coumarine 480)
pumped by the third harmonic of another Nd:YAG.
The ~ 120 J UV beam was focused
with a 40 cm focal length plano convex lens.
In the ILSRS experiments, Fig. 1(a), the UV laser was tuned to 236.295 nm,
corresponding to excitation of the vibrationless ground state to the
9'
~
(NH2 wag) on the A
state,12 and inducing resonant two photon ionization (R2PI). When the ILSRS transferred
population out of the vibrationless ground state, it led to a reduction in the detected molecular
ion signal, allowing measurement of the Raman signature of methylamine.
In the VMP experiment, Fig. 1(b), a fixed wavelength of 243.135 nm was used to
~
promote vibrationally excited molecules to the first excited electronic state, A , and to probe
the ensuing H photofragments. This wavelength was chosen for probing the H photofragment
via (2 + 1) resonantly enhanced multiphoton ionization (REMPI) through the 2s 2S
two-photon transition.
1s 2S
For monitoring the Raman signatures of methylamine and for
detecting the H photofragments, the SRE beams preceded the UV beam by ~ 15 ns. This
delay and the above mentioned energies of the SRE beams were chosen to minimize the
photoionization of the excited vibrational states. In addition, special care was taken to assure
that the size of the SRE beams in the interaction region is larger than that of the UV beam.
Ions formed via the ionization processes were extracted, accelerated and passed
through two pairs of orthogonal deflectors and an Einzel lens prior to entering the field free
drift region and eventual detection by a microsphere plate. The detector output was fed into a
digital oscilloscope, a boxcar integrator and passed to a personal computer for further
analysis. Raman signatures of methylamine and SRE dependent action spectra, monitoring
the yield of the H photofragments released in the VMP, were obtained by sampling the
molecular ions or the H ion yield, respectively, while scanning the
S
laser wavelength and
fixing the UV laser wavelength at the above mentioned wavelengths.
III.
RESULTS AND DISCUSSION
Figure 2 shows the H action (a) and the ILSR (b) spectra in the N-H and C-H stretch
fundamentals region of methylamine, where in the former more bands appear due to the
higher signal to noise ratio. Moreover, it is interesting to note that the bands appearing in
both types of spectra show a different intensity pattern. In particular, features corresponding
to 2 5, the overtone of the CH3 degenerate deformation, and
CH3 symmetric deformation with
5,
5
+
6,
the combination of the
are of low intensity in the ILSR spectrum, and become
more pronounced and of the order of stretches [N-H symmetric ( 1), degenerate CH3 ( 2),
and CH3 symmetric ( 3)] in the action spectrum. This can be due to the different origin of
these two types of spectra, where the ILSR spectrum reflects the extent of depletion of
ground state methylamine molecules as a result of vibrational excitation, while the action
spectrum the yield of the departing H photofragments.
Thus the ILSR spectrum represents
the Raman vibrational spectrum at low temperature and is very similar to our previously
measured room temperature photoacoustic Raman spectrum (PARS).13 Nevertheless,
comparing the action spectrum to the ILSRS spectrum is advantageous since it exhibits
narrowed rotational state distribution and reduced inhomogeneous structure.
The action
spectrum arises from the dissociation of vibrationally pre-excited and vibrationless ground
state [baseline in Fig. 2(a)] methylamine molecules.
enhancement of the 2
5
and the
~
the UV excitation, through the A
5
+
6
As previously suggested,13 the large
bands in the action spectrum seems to be related to
~
X transition (see below).
Previously,13 we determined the enhancements factors for each of the features by
dividing the peak areas of a particular feature in the action spectrum to that in the PARS and
normalizing them to the intensity of the
2
band. Here an alternative and more quantitative
approach is used, where the partial absorption cross sections for H photofragments
production, out of the vibrationally excited states, is determined by accounting for the
pumping efficiency or the fraction of excited molecules, fexc, and the enhancement factors, Ef,
for each of the vibrational states.
The fexc in the VMP experiment can be directly obtained by measuring the depletion
of ground state methylamine molecules in the ILSRS experiment. This can be done since fexc
in both types of experiments should be similar, provided that the fraction of the jet cooled
molecular beam illuminated by the SRE laser beams is the same and the ILSR and VMP
experiments are performed under identical conditions. Nevertheless, since the ILSRS
involves a R2PI process and the VMP a single-photon vibronic resonance followed by a (2 +
1) REMPI for H detection, some mismatch between the effective focal volume in the two
processes could be possible. To compensate for this source of error in evaluating fexc from the
ILSRS measurements and using it in the VMP measurements larger spot size SRE beams
were used than that of the UV beam.
Therefore, by measuring the decrease in the parent
molecular ion signal (the base line of the ILSRS signal) as a result of vibrational excitation,
the fexc was obtained.
By averaging over six experiments the values for the different
vibrational states of methylamine were obtained (with standard deviation of ~ 20 %) and are
given in Table 1.
The Ef were determined via the VMP experiment in which the methylamine was
excited with combined energies (SRE + UV) of ~ 43 900 – 44 500 cm-1 to accesses the
2 9' and
7'
+
9'
( 7' - CH3 rock and
7'
+
~
– NH2 wag)12 combination bands on the A state and
9'
predissociated thereafter. The action spectrum of Fig. 2(a) was used to determine Ef by taking
the ratio between the maximum signal for a specific band and the off-resonant signal (the
base line) in its vicinity, and they are given in Table I.
Considering that fexc for the different vibrational states leads to the corresponding Ef
and accounting for the cross-section for excitation of vibrationless ground state methylamine
molecules at 243.135 nm,
243
, the values for the partial cross-sections, for H fragments
production out of the vibrational states,
taking into account the measured value of
cross-sections,
vib
vib
, at this wavelength can be calculated. Thus,
243
, 0.05 Mb,29 (1 Mb = 10-18 cm2), the partial
, were calculated via the relation
vib
= Ef
I. Also given, for comparison, are the cross-sections,
243
/fexc and are given in Table
, for excitation of vibrationless
ground state at the combined (vibrational + UV) energies as retrieved from Ref. [29].
As can be seen from Table I, the
and
vib
values are in the 1.2 - 1.8 Mb and 5.3 -
24.0 Mb range, showing that the latter are a sensitive function of the excited vibrational state
and the one accessed on the upper electronic state. This behavior, where
considerably larger than
vib
are
, enables eventual significant production of H photoproducts and
monitoring of action spectra and also exploring vibrational effects. The states possessing
CH3 deformation excitation, 2
5
and
5
+
6,
present the highest cross-sections for 243.135
nm excitation, about 480 and 330 times greater, respectively, than
243
. This means that
the UV excitation is induced most effectively, through favorable Franck Condon (FC)
factors, i.e., larger overlap integrals between the vibrational wavefunctions of the ground and
first electronic states, or photodissociation efficiencies. Nevertheless, we have previously
shown13 that the REMPI spectrum of the parent, monitored by temporally overlapping SRE
and UV beams, and the H action spectrum are characterized by similar patterns. Since the
REMPI process probes directly the electronic transition, we came to the conclusion that the
UV excitation efficiency is only due to improved FC factors.
The C-H vibrational states on the ground electronic state are the result of strong
resonances, via the exchange of two quanta of the CH3 deformation mode with one quantum
of the C-H stretching mode;21 presumably there is some difference in their vibrational
content even on the ns timescale of our experiment.
The observed mode-dependent
photodissociation and, as shown previously,13 photoionization, indicates that these modes
survive IVR in a seven atom molecule with a torsional degree of freedom on a relatively long
time scale.
These results imply that the prepared states live considerably longer than
expected from the dependence of IVR rates on the internal rotation barrier height. 13,30 This
vibrational mode-dependent dissociation could be assisted by the ultrafast radiationless
transition taking place during the N-H bond fission. This transition is enabled by the conical
~
~
intersection between the ground, X , and first electronic state, A of methylamine.15,16,17
IV.
CONCLUSIONS
The present study combines, for the first time, VMP with ILSRS to provide a
quantitative measurement of partial absorption cross-sections out of initially excited
vibrational states. By determining the fexc from the ILSR spectrum and the Ef from the action
spectrum, the
vib
for a specific UV wavelength could be determined, implying a mode
dependent behavior, which is quite surprising for such a large and flexible molecule.30
Beyond the first demonstration, the importance of the ILSRS method is that it can be applied
to predissociating molecules or molecules with a UV chromophore, for attaining the resulting
selectivity.
Acknowledgments
The support of this research by the Israel Science Foundation (ISF) under grant No.
839/05 and by the James Franck Binational German-Israeli Program in Laser-Matter
Interaction is gratefully acknowledged.
1
F. F. Crim, J. Phys. Chem. 100, 12725 (1996).
2
I. Bar and S. Rosenwaks, Int. Rev. Phys. Chem. 20, 711 (2001).
3
H. A. Bechtel, J. P. Camden, D. J. A. Brown, M. R. Martin, R. N. Zare, and K.
Vodopyanov, Angew. Chem. Int. Ed. 44, 2382 (2005).
4
S. Yoon, R. J. Holiday, and F. F. Crim, J. Phys. Chem. B 109, 8388 (2005).
5
A. Teslja and J.J. Valentini, J. Chem. Phys. 125, 132304 (2006).
6
S. Yan, Y.-T. Wu, B. Zhang, X.-F. Yue, and K. Liu, Science 316, 1723 (2007).
7
K. K. Lehmann, G. Scoles, and B. H. Pate, A. Rev. Phys. Chem. 45, 241 (1994).
8
D. J. Nesbitt and R. W. Field, J. Phys. Chem. 100, 12735 (1996).
9
J. V. Michael and W. A. Noyes, J. Am. Chem. Soc. 85, 1228 (1963).
10
G. C. G. Waschewsky, D. C. Kitchen, P. W. Browning, and L. J. Butler, J. Phys.
Chem. 99, 2635 (1995).
11
M. N. R. Ashfold, R. N. Dixon, M. Kono, D. H. Mordaunt, and C. L. Reed, Philos.
Trans. R. Soc. London, Ser. A 355, 1659 (1997); C. L. Reed, M. Kono, and M. N. R.
Ashfold, J. Chem. Soc., Faraday Trans. 92, 4897 (1996).
12
D.-S. Ahn, J. Lee, J.-M. Choi, K.-S. Lee, S.J. Baek, K. Lee, K.-K. Baeck and S. K.
Kim, J. Chem. Phys. 128, 224305 (2008).
13
A. Golan, S. Rosenwaks, and I. Bar, J. Chem. Phys.125, 151103 (2006); ibid. Isr. J.
Chem. 47, 11 (2007).
14
R. Marom, U. Zecharia, S. Rosenwaks, and I. Bar, Chem. Phys. Lett. 440, 194
(2007); ibid. Mol. Phys. 106, 213 (2008); ibid. J.Chem. Phys. 128, 154319 (2008).
15
E. Kassab, J. T. Gleghorn, and E. M. Evleth, J. Am. Chem. Soc. 105, 1746 (1983).
16
K. M. Dunn and K. Morokuma, J. Phys. Chem. 100, 123 (1996).
17
C. Levi, G. J. Halász, Á. Vibók, I. Bar, Y. Zeiri, R. Kosloff, and M. Baer, J. Chem.
Phys. 128, 244302 (2008).
18
M. Kreglewski, J. Molec. Spectrosc. 72, 1 (1978); M. Kreglewski, J. Molec.
Spectrosc. 133, 10 (1989).
19
Y. Hamada, N. Tanaka, Y. Sugawara, A. Y. Hirakawa, M. Tsuboi, S. Kato, and K.
Morokuma, J. Mol. Spectrosc. 96, 313 (1982).
20
M. Oda and N. Ohashi, J. Molec. Spctrosc. 142, 57 (1990).
21
C. Levi, J. M. L. Martin, and I. Bar, J. Comput. Chem. 29, 1268 (2008).
22
H. L. Fang, R. L. Swofford, and D. A. C. Compton, Chem. Phys. Lett. 108, 534 (1984).
23
H. Wolff and E. Wolff, Ber. Bunsen-Ges. Phys. Chem. 69, 467 (1965).
24
U. Merker, H. K. Srvastava, A. Callegary, K. K. Lehmann, and G. Scoles, Phys.
Chem. Chem. Phys. 1, 2427 (1999).
25
D. P. Taylor and E. R. Bernstein, J. Chem. Phys. 103, 10453 (1995).
26
S. J. Baek, K.-W. Choi, Y. S. Choi, and S. K. Kim, J. Chem. Phys. 118, 11026
(2003); M. H. Park, K.-W. Choi, S. Choi, S. K. Kim, and Y. S. Choi, J. Chem. Phys.
125, 084311 (2006).
27
G.V. Hartland, B.F. Henson, V.A. Venturo, R.A. Hertz, and P.M. Felker, J. Opt. Soc.
Am. B 7, 1950 (1990); B.F. Henson, G.V. Hartland, V.A. Venturo, R.A. Hertz, and
P.M. Felker, Chem. Phys. Lett. 176, 91 (1991); G.V. Hartland, B.F. Henson, V.A.
Venturo, and P.M. Felker, J. Phys. Chem. 96, 1164 (1992); S. Lee, J.S. Chung, P.M.
Felker, J. L. Cacheiro, B. Fernandez, T.B. Pedersen, and H. Koch. J. Chem. Phys.
119, 12956 (2004).
28
P. M. Felker, P.M. Maxton, and M.W. Scaeffer, Chem. Rev. 94, 1787 (1994).
29
M.-J. Hubin-Franskin, J. Delwiche, A. Giuliani, M.-P.Ska, F. Motte-Tollet, I.C.
Walker, N.J. Mason, J.M. Gingell, and N.C. Jones, J. Chem. Phys. 116, 9261 (2002).
30
D. S. Perry, G. A. Bethardy, and X. L. Wang, Ber. Bunsenges. Phys. Chem. 99, 530
(1995).
TABLE I. The excited vibrational modes of methylamine, fraction of excited molecules,
enhancement factors, partial cross-sections for the 243.135 nm H dissociation out of different
initially prepared vibrational states and for excitation of vibrationless ground state molecules
at the corresponding combined energies.
Vibrational
mode
3
+
2 5
5
2
6
2
12
1
a
fexc (%)
Ef
11.2
3.5
3.5
7.2
9.8
12.4
15.6
11.5
16.8
14.0
10.4
16.6
vib
a
[Mb]
7.0
16.4
24.0
9.7
5.3
6.7
b
[Mb]
1.2
1.7
1.9
1.5
1.5
1.2
Partial cross-sections in Mb (1 Mb = 10-18 cm2) for 243.135 nm excitation out of
different vibrational states, calculated while accounting for the fraction of
vibrationally excited molecules, the enhancement factor for H fragments production
and the cross section for 243.135 nm excitation of vibrationless ground state
molecules.
b
Cross-sections retrieved from Ref. [29] for excitation of vibrationless ground state
molecules by energies corresponding to the respective combined excitation energies
(vibrational + UV) for each of the vibrations.
Figure Captions
Fig. 1: Schematics of the concepts for (a) ionization-loss stimulated Raman spectroscopy,
where resonant two photon ionization of the molecular parent, follows the depletion of
ground state reactant species as a result of stimulated Raman excitation and (b) vibrationally
mediated photodissociation, where pre-excited molecules are dissociated and the ensuing H
photofragments are probed by (2 + 1) resonantly enhanced multiphoton ionization.
Fig. 2: Methylamine vibrational spectra: (a) H action spectrum obtained via 243.135 nm
dissociation of vibrationally excited molecules and (b) ionization-loss stimulated Raman
spectrum exhibiting the depletion of parent molecule ionization signal due to vibrational preexcitation.
H+
UV
2
2s S
UV
UV
UV
CH3NH2+
~
CH 3 NH 2 A
H
UV
p
(a)
Fig. 1.
S
v
v =0
~
CH 3 NH 2 X
(b)
2
1s S
Fig. 2.