Angewandte
Chemie
near the top of the downhill potential along the C-C-S-D
angle for [D]4-methoxythiophenol (Figure 2). Accordingly,
there will be a strong torque exerted on the C-C-S-D dihedral
sample to possibly excite low-frequency modes. In Figure 4,
the translational energy distributions from the photodissoci-
ation of [D]4-fluorothiophenol at 243 nm are compared for
ꢀ
angle as the wavepacket moves out along the repulsive S D
coordinate. Therefore, as the reaction proceeds, the wave-
ꢀ
packet will be spread out not only along the S D bond axis
but also along the C-C-S-D angle. The wide spread of the
wavepacket at the S2/S0 conical intersection is expected to
diminish the probability of the diabatic passage through the
CI significantly. Consistently, the experiment shows a dra-
˜
matic decrease of the X state population, giving a yield of 0.09
˜
for X (0.91 for ffi) in the photodissociation of 4-methoxy-
thiophenol (Figure 2 and Table 1). In other words, the
reaction follows the adiabatic surface with a 91% probability,
whereas the diabatic passage through the CI occurs with a
probability of only 9%.
Figure 4. Total translational energy distribution from the photodissoci-
Table 1: Yields of the XC6H4S(X =H, F, OCH3) radical.
*
ation of [D]4-fluorothiophenol at a backing pressure of 3 atm ( ) and
a(CCSD)
08 (X=H)
318 (X=F)
738 (X=OCH3)
*
0.5 atm ( ). The distributions are normalized for the comparison. A
long tail at the high translational energy region in the distribution at
0.5 atm indicates that the molecule in the jet is vibrationally hot.
57%
74%
91%
43%
26%
9%
two different jet conditions. As the helium backing pressure
goes down from 3 atm to 0.5 atm, the temperature of the jet
increases because of the increase of the seeding ratio. The
Further exploration has been carried out for 4-fluoro-
thiophenol. According to the TDDFT calculations, the most
stable structure of [D]4-fulorothiophenol adopts a geometry
in which the C-C-S-D angle is 318 in the ground state, whereas
it has the minimum energy at the planar geometry in the S2
state (Figure 2). In this case, the initial wavepacket will be
located on the middle of the downhill potential along the C-C-
S-D angle. Therefore, the diabatic passage through the CI will
be less significant than for thiophenol but more probable than
for 4-methoxythiophenol, because the extent of the wave-
packet spread along the dihedral angle in the vicinity of the CI
is likely to be between those of the two other cases.
Experimental findings support this simple model, giving
DFT torsional frequency of 4-fluorothiophenol is 29 cmꢀ1 [23]
,
and it is most likely that a significant portion of the molecules
in the warm jet populate the vibrationally excited state with
several quanta of the C-C-S-D torsional mode. Since the
torsional barrier height is calculated to be very low in the
ground state for 4-fluorothiophenol (Figure 2), the UV
excitation of the molecule in the warm jet would result in
the widening of the initial wavepacket location in the two-
dimensional potential-energy surface along the C-C-S-D
dihedral angle. Therefore, the wavepacket will experience
more significant spread at the CI, giving the lower probability
˜
of diabatic passage to the X state of the fragment. Quite
˜
consistently, the experiment gives a yield of 0.18 for the X
˜
˜
0.26 and 0.74 for yields of the X and ffi states, respectively,
state of FC6H4S in the warm jet, which is much less than the X
for 4-fluorothiophenol (Table 1). It should be noted that the
yield of 0.26 measured for the molecule in the cold jet. This
experimental finding strongly supports the above two-dimen-
sional bifurcation dynamics at the CI.
In conclusion, we have employed the conformational
preference of thiophenol derivatives to manipulate the initial
location of the wavepacket along the C-C-S-D dihedral-angle
˜
Xꢀffi energy gaps of phenylthiyl derivatives (XC6H4SD; X =
H, CH3O, F) are found to be almost identical, thus indicating
that the relative energy of the SOMO localized on sulfur (3p)
ꢀ
is only slightly affected by the chemical substitution. The S D
bond energy is slightly affected by the chemical substitution.
The UV absorption spectra of all thiophenol derivatives
studied herein show very similar features at around 243 nm,
ꢀ
coordinate on two-dimensional PES leading to prompt S D
bond dissociation. The branching ratio between X and ffi
˜
!
where the S2 S0 transition is dominant.
states of the final fragment is controlled by the conforma-
tional preference induced by chemical substitution at the para
position of [D]thiophenol. The experimental trend is success-
fully explained by the simple wavepacket propagation model.
Namely, the wavepacket spread along the C-C-S-D dihedral
angle either through conformational control or vibrational
excitation strongly affects the probability of diabatic passage
through the conical intersection in such a way that the
probability of the diabatic passage at the CI diminishes as the
Another way of manipulating the wavepacket location on
S2 would be the selective excitation of the out-of-plane C-C-
S-D torsional mode in the ground state. However, such an IR–
UV double excitation scheme is experimentally quite
demanding, as the actual torsional mode frequency is too
low (less than 100 cmꢀ1) to be excited by a conventional IR
laser pulse. In this case, the backing pressure of the supersonic
jet is varied to change the vibrational temperature of the
Angew. Chem. Int. Ed. 2008, 47, 1853 –1856
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1855