66 Ying Luo et al.
with the next two excited states S
2
and S
3
. Here it is important to note that the trans-2
geometry (i.e., = 180
) is the natural-state and the cis-geometry (i.e., = 0
)isthe
metastable-state. Without external excitations, the large PE barrier 1.98 eV in the ground
state prevents the transition from trans-2-butene to cis-2-butene. It is also clear from
Figure 6(B) that there is no barrier in the first excited singlet state S
1
, which is ideal for
an ultra-fast switching between the trans- and cis- isomers.
Potential energy curves can be used to illustrate this optically induced isomerization
process by following the superimposed arrows. The process begins with a required
optical excitation (∼846 eV, 146 nm) of the “electronic-state” of the trans-geometry
from S
0
to S
1
. This is followed by a non-radiative decay to the S
1
PE valley minimum,
which corresponds to a 90
rotation about the reaction coordinate. At this point, the
molecule undergoes an “electronic” radiative decay from S
1
to S
0
. This is followed by
a second non-radiative decay to the cis-geometry, which corresponds to a second 90
rotation about the reactive coordinate. At this point, butene will remain in the metastable
cis-2 geometry until thermal relaxation of the system back to the ground state.
To estimate the probability that the excited trans-2-butene will follow the isomer-
ization process described above, instead of relaxing back to its own ground state, we
investigate the molecular dynamics of butene in the first singlet excited state S
1
by
combining the theory of Newton’s Dynamics and standard quantum chemistry software
Gaussian 98. We began the investigation from cis-geometry ( =0
and calculated the
time for the excited cis-geometry to achieve the energy minimum ( = 90
and the
excited trans-geometry ( =180
of S
1
. The calculation procedure is as follows: (i) we
calculated the potential energy of the ground state and the excitation energy of S
1
of the
optimized cis-butene using Gaussian 98 package and by adding the two energies we got
the potential energy of the excited cis-butene. Note that we took this potential energy as
the total energy of butene ( i.e. we assume the kinetic energy of excited cis-butene is 0).
To initiate the rotation, we actually started from a very small with a very small initial
velocity. (ii) For every 10
, we calculated 20 points (i.e., we performed Gaussian 98
calculation every 05
). For every point, we calculated the ground-state energy without
optimization and the corresponding excitation energy; thus we got the potential energy
of the excited state. By subtracting the potential energy from the total energy, we got
the kinetic energy. Then we used the energies of 20 points to calculate the time for
butene to rotate 10
by integrating Newton’s equations.
The relation between the potential energies of excited states and time is given in
Figure 7, and for demonstrative purposes, the PE curves of the ground state S
0
is
presented on the same graph. We can see that the time for excited trans-2-butene to
relax to the valley minimum is about 100 fs and symmetrically it takes almost the same
time to go from the minimum to cis-2-butene. On the other hand, according to the
spontaneous emission theory the time for excited trans-2-butene to relax back to ground
state is about 59 ×10
3
s. So the isomerization process will be significantly probable.
In Figure 8, we demonstrate how the dihedral angle of C1–C2
==
C3–C4 double bond, or
the geometric structure of butene, will change with time.
The lowest vibrational frequencies and the associated IR intensities that were cal-
culated for trans-2-butene and cis-2-butene are compared in Figure 9. These results
indicate a significant difference in spectral signatures of the two molecular conforma-
tions and one that has the general quantitative characteristics needed by the previously
discussed bio-molecular architecture.