Orientational dynamics of hydrogen-bonded phenol

Article abstract of DOI:10.1063/1.1809589

We use femtosecond mid-infrared pump-probe spectroscopy to study the effects of hydrogen bonding on the orientational dynamics of the OD-stretch vibration of phenol-d. We study two samples: phenol-d in chloroform and phenol-d in chloroform to which we add

Full text of DOI:10.1063/1.1809589

Orientational dynamics of hydrogen-bonded phenol
Y. L. A. Rezus, D. Madsen, and H. J. Bakker
Citation: The Journal of Chemical Physics 121, 10599 (2004); doi: 10.1063/1.1809589
View online: http://dx.doi.org/10.1063/1.1809589
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/121/21?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Dynamics of hydrogen-bonded OH stretches as revealed by single-mode infrared-ultraviolet laser double
resonance spectroscopy on supersonically cooled clusters of phenol
J. Chem. Phys. 129, 154308 (2008); 10.1063/1.2988494
Relaxation and anharmonic couplings of the O–H stretching vibration of asymmetric strongly hydrogen-bonded
complexes
J. Chem. Phys. 127, 044501 (2007); 10.1063/1.2753840
Hydrogen-bonded OH stretching modes of methanol clusters: A combined IR and Raman isotopomer study
J. Chem. Phys. 126, 194307 (2007); 10.1063/1.2732745
Refinements on solvation continuum models: Hydrogen-bond effects on the OH stretch in liquid water and
methanol
J. Chem. Phys. 112, 5382 (2000); 10.1063/1.481108
Vibrational dynamics of hydrogen-bonded HCl-diethyl ether complexes
J. Chem. Phys. 112, 5127 (2000); 10.1063/1.481069
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56

JOURNAL OF CHEMICAL PHYSICS
VOLUME 121, NUMBER 21
1 DECEMBER 2004
Orientational dynamics of hydrogen-bonded phenol
Y. L. A. Rezus,a) D. Madsen, and H. J. Bakker
FOM-institute for Atomic and Molecular Physics, Kruislaan 407, 1098 SJ Amsterdam, The Netherlands
͑Received 9 July 2004; accepted 2 September 2004͒
We use femtosecond mid-infrared pump-probe spectroscopy to study the effects of hydrogen
bonding on the orientational dynamics of the OD-stretch vibration of phenol-d. We study two
samples: phenol-d in chloroform and phenol-d in chloroform to which we added excess acetone. For
phenol-d in chloroform, we observe rotational diffusion of the OD group around the CO bond, with
a correlation time of 3.7 ps. For phenol-d hydrogen bonded to acetone, the reorientation time is
strongly dependent on the probe frequency, varying from 3 ps on the blue side of the spectrum to
more than 30 ps on the red side. © 2004 American Institute of Physics.
͓DOI: 10.1063/1.1809589͔
I. INTRODUCTION
strength. However, the rapid interconversion of conforma-
tions, which occurs in liquid water, makes it impossible to
directly probe the orientational dynamics of a unique confor-
mation for longer than about 1 ps. Therefore these studies
have relied on models for extracting information about the
frequency dependence of orientational processes. Structur-
ally more complex molecules, such as alcohols, dissolved in
apolar solvents display a much slower rate of interconver-
sion, and it should therefore be possible to probe the reori-
entation of different conformations directly. However, only
few papers have been published that study reorientation in
these systems,13–15 and none have presented a study of the
influence of the hydrogen-bond interaction on reorientational
processes.
Hydrogen bonds play a crucial role in the molecular dy-
namics of many complex systems, such as biomolecules,
solute-solvent complexes, and hydrogen-bonded liquids
͑e.g., water͒. An elegant method for studying the dynamics
of hydrogen-bonded systems is provided by probing the OH-
stretch vibration, as its frequency is highly sensitive to the
local geometry of the hydrogen bond. In general, hydrogen
bonding leads to a strongly broadened OH absorption band
that is redshifted with respect to the free OH absorption. The
broadening of the absorption is a direct consequence of the
large number of conformations present in hydrogen-bonded
systems, differing in hydrogen-bond length and therefore ab-
sorbing at a different frequency; ‘‘the shorter the hydrogen
bond, the lower the OH-stretch frequency,’’ is the rule of
thumb by which this phenomenon is described.1,2
The correlation between hydrogen-bond length and OH-
stretch frequency makes it possible to distinguish the differ-
ent hydrogen-bonded conformations spectroscopically,
which opens the way to study their dynamical properties.
Unfortunately, linear infrared techniques cannot provide dy-
namical information about subensembles, as this requires the
use of at least two light-matter interactions. Nonlinear infra-
red methods do have the ability to provide such information.
In these methods one or more light pulses are used to label a
subensemble, after which the evolution of the system is read
out by a next pulse.
One of these nonlinear techniques is femtosecond mid-
infrared pump-probe spectroscopy. This technique has been
applied extensively to the investigation of the dynamics of
hydrogen-bonded systems.3–9 In addition to providing infor-
mation about vibrational relaxation and spectral diffusion,
the method can also be used to study reorientational pro-
cesses in molecules. In the past reorientational measurements
have thus provided insight into the dynamics of HDO mol-
ecules in liquid heavy water (D2O)10,11 and ionic solvation
shells.12 These studies suggest that the reorientation time of
water molecules depends strongly on the hydrogen-bond
Here we report on the use of polarization and frequency
resolved mid-infrared pump-probe spectroscopy to investi-
gate the influence of the O–D¯O hydrogen-bond length on
the orientational dynamics of the hydrogen-bond donating
OD group.
II. EXPERIMENT
Laser system. The laser system we used for our experi-
ments consists of a commercial Ti:sapphire regenerative am-
plifier that generates pulses with a duration of 100 fs and an
energy of 1 mJ. Its output is used to pump a commercial
two-stage optical parametric amplifier ͑OPA͒ based on
␤-barium borate ͑BBO͒, which we tuned such that the 800
nm pump light is split into signal and idler photons of 1.3
and 2 ␮m, respectively. The idler pulse is frequency-doubled
in a second BBO crystal, and the resulting 1 ␮m light is used
as a seed in a second OPA process in KNbO3 , pumped by
the residual 800 nm light from the first OPA process. This
process yields pulses that are resonant with the OD-stretch
mode ͑4 ␮m͒, with an energy of 4 ␮J and a duration of about
350 fs.
Pump-probe setup. Figure 1 shows a schematic outline
of the pump-probe setup. A small part of the mid-infrared
light is split off by a wedged CaF2 window to obtain probe
and reference pulses. A ␭/2 plate is used to turn the polariza-
tion of the pump pulse by 45° with respect to that of the
probe light. The pump, probe, and reference beams are fo-
a͒
Electronic mail: rezus@amolf.nl
0021-9606/2004/121(21)/10599/6/$22.00
10599
© 2004 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56

10600
J. Chem. Phys., Vol. 121, No. 21, 1 December 2004
Rezus, Madsen, and Bakker
FIG. 1. Schematic outline of the pump-probe setup.
III. RESULTS AND DISCUSSION
cused onto the sample by an off-axis parabolic mirror and
recollimated by an identical mirror. The probe and reference
pulses are focused onto the entrance slit of a spectrometer. In
the spectrometer the pulses are spectrally dispersed onto a
liquid-nitrogen cooled 2ϫ32 MCT ͑mercury-cadnium-
telluride͒ array. Behind the sample a polarizer is placed to
select either the parallel or the perpendicular polarization
component of the probe light with respect to the pump,
which yields the transient absorptions ⌬␣ʈ and ⌬␣Ќ , re-
spectively. From these signals the isotropic signal
Transient spectra and vibrational relaxation. Figure 3
shows the transient spectrum of phenol in chloroform. The
ground state bleach ͑2650 cmϪ1͒ and the excited state ab-
sorption ͑2550 cmϪ1͒ are clearly separated. Apparently the
OD potential has an anharmonicity of 100 cmϪ1. In addition
to the bleach and the excited state absorption, a rather broad
shoulder is observed on the red side of the bleach ͑ϳ2625
cmϪ1͒. This shoulder decays much faster ͑ϳ1 ps͒ than the
bleach and the excited state absorption ͑ϳ10 ps͒, and can be
attributed to clusters of hydrogen-bonded phenol molecules.
This assignment is supported by the linear spectrum, which
shows that the tail of the cluster band lies in the region where
the shoulder is observed.
⌬␣ʈϩ2⌬␣Ќ
⌬␣isoϭ
,
͑1͒
3
which is independent of reorientational processes, and the
anisotropy
The decay of the transient absorption ͑Fig. 4͒ is faster on
the red side of the spectrum than on the blue side. We have
used biexponentials to fit these measurements, thereby ac-
counting for the transient bleaching of the tail of the cluster
band, which overlaps with the monomer band and contrib-
utes weakly to the total transient signal ͑ϳ15% at 2650
cmϪ1͒. In Fig. 5 the two derived time constants are plotted as
a function of the probe wavelength. The slow decay time,
which we interpret as the lifetime of the OD-stretch vibration
⌬␣ʈϪ⌬␣Ќ
Rϭ
,
͑2͒
3⌬␣iso
which only reflects reorientation, are constructed.
Sample. In order to study the effect of hydrogen bonding
on the orientational dynamics of the OD group of phenol-d
we compared two samples: a solution of phenol-d in chloro-
form (0.4M) and a solution of phenol-d in chloroform
(0.6M) to which an excess of acetone was added (ϳ3M).
Phenol-d was obtained by dissolving 2 g of phenol-h ͑Ald-
rich͒ in 10 ml of D2O and boiling the solution until all D2O
had evaporated. Chloroform ͑high performance liquid chro-
matography grade͒ was used as received. All measurements
were performed in a static cell with an optical path length of
500 ␮m.
Figure 2 shows the linear IR absorption spectra of these
two samples. Two OD-stretch absorption bands can be ob-
served in the spectrum of phenol in chloroform; a narrow
band at 2650 cmϪ1 ͑FWHMӍ20 cmϪ1͒ and a broad band at
2500 cmϪ1 ͑FWHMӍ200 cmϪ1͒, where FWHM stands for
full width at half maximum. The narrow band is due to mo-
nomeric phenol molecules, whereas the broad band is caused
by clusters of hydrogen-bonded phenol molecules.16 When
an excess of acetone (ϳ3M) is added, the band at 2650
cmϪ1 disappears while the band at 2500 cmϪ1 grows in in-
tensity, which leads us to the conclusion that all phenol mol-
ecules in this sample are hydrogen bonded. As an excess of
acetone is present, we can safely assume that acetone acts as
the hydrogen-bond acceptor, rather than other phenol mol-
ecules.
FIG. 2. Linear IR absorption spectra of phenol-d in chloroform (0.4M) and
phenol-d in chloroform (0.4M) with acetone (ϳ3M). For clarity the latter
spectrum has been shifted upwards by 0.4. The sample thickness was
500 ␮m.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56

J. Chem. Phys., Vol. 121, No. 21, 1 December 2004
Orientational dynamics of phenol
10601
FIG. 5. Time constants obtained from biexponential fits to the transient
spectrum ͑top panel͒ of phenol-d in chloroform as a function of the probe
wavelength (␯pumpϭ2650 cmϪ1). The triangles correspond to the bleaching
of the OD band of monomeric phenol, the squares to its induced absorption
and the circles to the bleaching of the cluster band.
FIG. 3. Transient spectra of phenol in chloroform (0.4M) at delays of 0.3,
5, and 20 ps. The pump pulse was centered at 2650 cmϪ1
.
of monomeric phenol, varies smoothly from about 5 to 11 ps
going from the red to the blue side of the spectrum. This
behavior contrasts with that of the cluster band, which shows
a remarkably constant lifetime of about 1 ps over the entire
probe range.
variation in lifetime can be observed, only at the very blue
edge the lifetime increases.
The role of the hydrogen bond in vibrational relaxation.
The above results show that the hydrogen bond provides,
either directly or indirectly, a very efficient vibrational relax-
ation pathway. For weakly hydrogen-bonded phenol, we ob-
serve that the lifetime decreases from 11 to 5 ps as the
hydrogen-bond strength increases. When the hydrogen-bond
strength is further increased, i.e., by adding the strong
hydrogen-bond acceptor acetone, the lifetime decreases
down to 1 ps. At this point the lifetime seems to reach its
minimum value, and a further increase in hydrogen-bond
strength no longer results in a decrease of the lifetime.
When seeking an explanation for the frequency indepen-
dence of the lifetime of phenol hydrogen-bonded to acetone,
the first thing that comes to mind is that the strongly
hydrogen-bonded conformations may interconvert rapidly, so
that in fact, an average lifetime is observed. However we will
see below that there is no rapid spectral diffusion, which
The strong variation in vibrational lifetime as a function
of frequency suggests that the ͑monomeric͒ phenol mol-
ecules are hydrogen bonded to the solvent chloroform mol-
ecules. Furthermore, it suggests that the phenol-chloroform
conformations do not interconvert within ϳ10 ps ͑for else a
frequency independent lifetime would have been observed͒.
Figure 6 shows the transient spectra of phenol in chlo-
roform with excess acetone. We observe a broad bleaching
signal and the high-frequency wing of the excited state ab-
sorption on the red side of the probe spectrum. The bleach
decays monoexponentially with a time constant of about 1
ps, which is an order of magnitude faster than for phenol
hydrogen-bonded to chloroform. This time constant is very
similar, however, to that of phenol clusters. Figure 7 shows a
plot of the lifetime of phenol hydrogen-bonded to acetone as
a function of the probe wavelength. Remarkably, hardly any
FIG. 4. Delay scans at three different probe frequencies in the bleaching
FIG. 6. Transient spectra of phenol-d hydrogen-bonded to acetone in chlo-
region of the transient spectrum of phenol-d in chloroform (␯
roform measured at 0.3, 1, and 2 ps. The pump pulse was centered at
pump
ϭ2650 cmϪ1).
2525 cmϪ1
.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56

10602
J. Chem. Phys., Vol. 121, No. 21, 1 December 2004
Rezus, Madsen, and Bakker
FIG. 7. Lifetime of the OD vibration of phenol-d hydrogen-bonded to ac-
etone as a function of probe wavelength (␯pumpϭ2525 cmϪ1). The transient
spectrum is shown in the top panel. The triangles correspond to the bleach-
ing region of the spectrum, the squares to the excited state absorption.
FIG. 8. Anisotropy decay of phenol-d in chloroform at 2650 cmϪ1. The
solid line represents a fit to the data of the form Rϭ(A0ϪAϱ)eϪt/␶
rot
ϩAϱ . The inset illustrates the orientational dynamics that lead to the decay
of the anisotropy.
This reasoning leads us towards attributing the reorien-
tation time of 3.7 ps to the rotational diffusion of the OD-
group around the CO bond. The fact that this time constant
does not depend on frequency shows that the orientational
mobility is negligibly affected by the weak hydrogen bond to
chloroform. In principle, the anisotropy should further decay
to zero if we would measure up to longer delays, at which
the molecular reorientation of phenol becomes important.
Unfortunately, because of the finite relaxation time of the OD
vibration, we were only able to accurately determine the an-
isotropy for delays up to 15 ps.
We compare these results with those obtained for
strongly hydrogen-bonded phenol. Figure 9 shows the an-
isotropy decay of strongly hydrogen-bonded phenol at four
distinct positions in the transient spectrum. Interestingly,
while the lifetime showed no variation over the absorption
means that this explanation cannot be correct. Therefore we
conclude that the relaxation involves a mechanism that
speeds up with increasing strength of the hydrogen bond, but
only up to a certain limiting rate. Of course this leads to the
question what the exact nature of this relaxation channel
could be. Even though it is impossible to provide a conclu-
sive answer based on the experiments described in this paper,
it is unlikely that the hydrogen bond is one of the accepting
modes, since then the relaxation rate is expected to keep
increasing with the hydrogen-bond strength. Instead, the role
of the hydrogen bond may be indirect in matching the energy
of the excited OD-stretch vibration to that of the combination
of ͑other͒ accepting modes.
Effect of hydrogen bonding on reorientation. In order to
obtain the reorientation time of the OD group of phenol in
chloroform, we have measured the anisotropy of the transient
absorption. In Fig. 8 the anisotropy is shown for the peak
position of the bleach. By fitting this anisotropy to a single
exponential we obtained a rotational correlation time of 3.7
ps. Interestingly, whereas the vibrational relaxation showed a
strong frequency dependence for phenol in chloroform, the
reorientation time is fairly frequency independent.
A point worth mentioning is that the anisotropy does not
decay to zero as one would expect in the case of free rota-
tional diffusion, but to a value of ϳ0.04. This nonzero end
level can be explained by considering the two motions that
play a role in the reorientation of the OD group; the first is
the rotation of the OD group around the CO-bond axis, and
the second is the rotation of the phenol molecule as a whole,
which of course also contributes to the reorientation of the
OD group. We assume that the first process occurs much
faster than the second, so that on the time scale of our ex-
periment, we essentially only observe the rotation of the OD
group around the CO-bond axis ͑see inset to Fig. 8͒. As this
rotation restricts the reorientation of the OD group to a lim-
ited portion of the unit sphere, a nonzero anisotropy will be
measured at long delays. The end level of the anisotropy is
determined by the solid angle that can be covered by the OD
group and therefore by the C–O–D angle.
FIG. 9. Anisotropy decay at different frequencies in the transient spectrum
of phenol-d hydrogen-bonded to acetone. The solid curves are monoexpo-
nential fits with an end level of 0.04.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56

J. Chem. Phys., Vol. 121, No. 21, 1 December 2004
Orientational dynamics of phenol
10603
bedded in a network of hydrogen bonds, contrary to phenol
that forms a complex with only a single acetone molecule.
The paradox is resolved, when one realizes that the reorien-
tation of water molecules is assisted by the rapid breaking
and reformation of hydrogen bonds that occurs in liquid wa-
ter. It is more difficult to understand however that the reori-
entation of the OD group of phenol, when it is weakly hy-
drogen bonded to chloroform, proceeds slower ͑3.7 ps͒ than
that of hydrogen-bonded HDO molecules ͑2.6 ps͒. A tenta-
tive explanation may be that the rotation of the OD group is
considerably hindered by the phenyl ring, as gas phase stud-
ies and quantum chemical calculations17 indicate that the OD
group lies preferentially in the plane of the phenyl ring, lead-
ing to a significant barrier for internal rotation around the
CO-bond axis.
FIG. 10. Reorientation time of the OD group of phenol-d hydrogen-bonded
to acetone as a function of the probe wavelength (␯pumpϭ2525 cmϪ1). The
arrows indicate reorientation times that greatly exceed 30 ps, but that could
not be fitted accurately. The top panel shows the transient spectrum. The star
corresponds to the reorientation time measurement for phenol-d in chloro-
form.
IV. CONCLUSION
We studied the influence of hydrogen bonding on the
orientational dynamics of the OD group of phenol-d. These
dynamics are measured by probing the anisotropy of the ex-
citation of the OD-stretch vibration. For weakly hydrogen-
bonded phenol molecules dissolved in chloroform, the an-
isotropy decays with a time constant of 3.7 ps to a nonzero
value. This partial decay of the anisotropy results from the
rotational motion of the OD group around the CO-bond axis.
The molecular reorientation of phenol that would lead to a
complete decay of the anisotropy takes place on a much
slower time scale. The rotational diffusion of the OD group
is observed to slow down with increasing hydrogen-bond
strength, resulting in an increase of its time constant to val-
ues Ͼ30 ps. Hydrogen bonding also affects the vibrational
band, we now see that the reorientation proceeds much faster
on the blue side of the spectrum than on the red side. As the
anisotropy decays more or less linearly over the range that
we can measure, it is impossible to unambiguously fit these
curves with monoexponentials including a nonzero end level.
In order to be able to assign a decay time to these curves, we
have fitted them to monoexponentials with an end level that
is the same as the one we found for phenol hydrogen-bonded
to chloroform ͑0.04͒. The reasoning underlying this choice is
that hydrogen bonding may affect the reorientation time of
the OD group, but not the solid angle over which it can
reorient, so that the final anisotropy should be the same.
Figure 10 shows the reorientation time as a function of
probe wavelength for the entire probe range. On the blue side
of the band the anisotropy decays with a time constant that is
comparable to that of phenol hydrogen-bonded to chloroform
͑3 ps͒, suggesting that also for these molecules the rotation
of the OD group is not hindered by the hydrogen bond. On
the red side of the band however, the reorientation time in-
creases dramatically to over 30 ps.
These results show that hydrogen bonds hinder the rota-
tion of the donating hydroxyl group, and that the strongest
hydrogen bonds are most effective at doing so. The fact that
we were actually able to probe the orientational dynamics of
unique conformations, implies that these conformations live
for a relatively long time. The rate of interconversion must
be slower than about 5 ps, for else we would not have ob-
served different values of the anisotropy at this delay.
The time scales that have been identified in this paper
can be placed in a broader perspective by making a compari-
son with the study by Nienhuys, Van Santen, and Bakker on
the reorientation of HDO molecules in liquid heavy water.10
In this study HDO molecules were found to reorient with a
time constant of 2.6 ps, which seems extremely rapid com-
pared to the time constant of Ͼ30 ps observed for the
strongly hydrogen-bonded OD group of phenol-d. This is
particularly true considering that an HDO molecule is em-
relaxation. The vibrational lifetime ␶ decreases with in-
creasing hydrogen-bond strength but only down to a limiting
value of ϳ1 ps.
life
ACKNOWLEDGMENTS
This work is part of the research program of the ‘‘Stich-
ting voor Fundamenteel Onderzoek der Materie ͑FOM͒,’’
which is financially supported by the ‘‘Nederlandse organi-
satie voor Wetenschappelijk Onderzoek ͑NWO͒.’’ We thank
M. Bonn for carefully reading the manuscript.
1 A. Novak, Struct. Bonding ͑Berlin͒ 18, 177 ͑1974͒.
2 W. Mikenda, J. Mol. Struct. 147, 1 ͑1986͒.
3 M. Lim and R. M. Hochstrasser, J. Chem. Phys. 115, 7629 ͑2001͒.
4 N. Huse, K. Heyne, J. Dreyer, E. T. J. Nibbering, and T. Elsaesser, Phys.
Rev. Lett. 91, 197401 ͑2003͒.
5 J. C. Deak, S. T. Rhea, L. K. Iwaki, and D. D. Dlott, J. Phys. Chem. A 104,
4866 ͑2000͒.
6 A. Pakoulev, Z. Wang, Y. Pang, and D. D. Dlott, Chem. Phys. Lett. 380,
404 ͑2003͒.
7 C. J. Fecko, J. D. Eaves, J. J. Loparo, A. Tokmakoff, and P. L. Geissler,
Science 301, 1698 ͑2003͒.
8 G. M. Gale, G. Gallot, F. Hache, N. Lascoux, S. Bratos, and J.-C. Leick-
nam, Phys. Rev. Lett. 82, 1068 ͑1999͒.
9 R. Laenen, C. Rauscher, and A. Lauberau, Phys. Rev. Lett. 80, 2622
͑1998͒.
10 H.-K. Nienhuys, R. A. Van Santen, and H. J. Bakker, J. Chem. Phys. 112,
8487 ͑2000͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56

10604
J. Chem. Phys., Vol. 121, No. 21, 1 December 2004
Rezus, Madsen, and Bakker
11 S. Woutersen, U. Emmerichs, and H. J. Bakker, Science 278, 658 ͑1997͒.
12 M. F. Kropman, H.-K. Nienhuys, and H. J. Bakker, Phys. Rev. Lett. 88,
077601 ͑2002͒.
14 R. Laenen and C. Rauscher, Chem. Phys. Lett. 274, 63 ͑1997͒.
15 R. Laenen and K. Simeonidis, Chem. Phys. Lett. 299, 589 ͑1999͒.
16 U. Liddel and E. D. Becker, Spectrochim. Acta 10, 70 ͑1957͒.
17
13 K. J. Gaffney, I. R. Piletic, and M. D. Fayer, J. Chem. Phys. 118, 2270
͑2003͒.
B. J. C. Cabral, R. G. B. Fonseca, and J. A. M. Simoes, Chem. Phys. Lett.
ˆ
258, 436 ͑1996͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.138.73.68 On: Sat, 20 Dec 2014 18:20:56