remains nearly unchanged for 7c (350 nm). The large
solvation change in 7z further implies its zwitterionic
property. According to the calculation, the electronic excita-
tion between S0 and S1 of 7c is dominated by the HOMO f
LUMO excitation (Figure 3), in which the electron density
of HOMO is mainly located on the lone pair of O(8) atoms
and the p-methoxybenzamine moiety. Conversely, the elec-
tron density is transferred to the NO2 group in the LUMO.
The net change of the electron density during the HOMO
f LUMO transition is expected to cause the C-O bond
weakening and consequently cleavage. We also performed
a similar calculation on 7t, considered to be another possible
neutral product (see Scheme 3). The calculated S0f S1
transition of 28 570 cm-1 (350 nm) is much higher in energy
than that of the product (7z) absorption peak experimentally
resolved to be 500 nm. The result justifies the assignment
of 7z to a zwitterionic species, Theoretically, the 7c f 7z
formation involves only the cleavage of C(sp3)-O bonding
at the junction of the heterobicycle, while for 7t, if it were
the photoproduct, in addition to the bond cleavage, another
hydrogen transfer from N to O would be necessary, leading
to a dynamically unfavorable pathway.
“trapping” the 7z species. Note that similar hydrogen bonding
formation is not evident in other synthesized analogues
presented in this study.
We have also made attempts to analyze the product yield
in a more quantitative manner. Using a 30 mW/cm2 355 nm
Nd:YAG pulse laser to illuminate the entire volume of 3
mL of 7c solution (7.8 × 10-5 M in acetonitrile) for 100
min, we then observed that the absorbance at 500 nm was
increased from 0.003 to 0.172, corresponding to an increase
of 4.9 × 10-5 M of the 7z production. Taking the ratio of
the number of 7z molecules being produced versus the
number of photons being absorbed, the yield of 7z production
was then estimated to be 0.12 ( 0.02%.
Since the above photochromic phenomena, the associated
dynamics, and efficiency are independent of O2, the reaction
less likely takes place in the triplet manifold. To draw the
relative thermodynamics in the excited singlet state, we
simply added the observed S0 f S1 absorption peak in terms
of energy to the calculated energy for 7c and 7z. The result,
depicted in Scheme 4 of the Supporting Information, indi-
cates that 7c f 7z is thermodynamically favorable only by
∼1.0 kcal/mol along the S1 state. Alternatively, it is more
plausible that the photochromism proceeds via a nonadiabatic
(i.e., surface crossing) process from S1 (7c) to S0 (7z). Never-
theless, at this stage, whether the reaction takes place strictly
in the excited state or via crossing to the ground state poten-
tial energy surface is pending detailed dynamics approaches.
The oxazabicycle 7c reported here, to the best of our
knowledge, represents the first example of a heterobicyclic
molecule exhibiting satisfactory photochromic properties.
Although the overall photochromism cycle is relatively slow
and the maximum absorption at 500 nm (ꢀ ∼500 M-1 cm-1,
see Figure 1) is low, we believe that 7c or its analogues may
still merit future applications. For example, the 7z f 7c
reverse reaction is of importance in that the ∼34 kcal/mol
is chemically stored in 7z and may be re-exploited in the
presence of suitable catalysts. Moreover, in the current
system, both 7c and 7z are nonluminescent despite their
anchoring coumarin moiety. Thus, work focusing on the
improvement of the reaction reversibility as well as the
utilization of fluorescence based on coumarin analogues to
enhance the detection should be of great interest. We thus
believe that this work may open an exciting new avenue for
future development of the photochromic dyes with novel
molecular structures.
Among 7a-e only 7c exhibits the aforementioned pho-
tochromic properties. The fact that only the oxazabicycle with
a nitro substituent at the ortho position of the bridgehead
benzene ring exhibited the expected photochromic behavior
implied that the ortho substituent, but not meta and para, on
the bridgehead benzene ring plays an important role in the
photochemistry. We initially speculated that the increasing
steric hindrance between the ortho substituent and the nearby
methyl groups on the heterobicyclic skeleton made 7c more
susceptible to a C-O bond rupture than the corresponding
p-nitro-substituted 7b. However, neither the o-chloro-func-
tionalized (7d) nor the o-methoxy-substituted (7e) analogue
exhibits any photochromic behavior, suggesting that factors
other than the steric effect govern the photochromism. An
additional experiment was also performed and found no
photochromic property when the amino group (N(2)) on 7c
was methylated. The result thus drew our attention on the
possible role of hydrogen bonding formation between the
iminium hydrogen (N(2)-H) and nitrooxygen (N(10)-O).
The result of molecular simulation for 7c produced an
estimated distance between the oxygen atom on the o-nitro
group and the amino hydrogen atom of 2.39 Å (see Figure
S2 in the Supporting Information), while this distance for
7z even decreases to as short as 1.98 Å. It is thus reasonable
to propose that 7z can be further stabilized via enhancing
the intramolecular hydrogen bonding strength between the
iminium hydrogen and the nitrooxygen. This hypothesis is
supported by the result that 7z is calculated to be 34.2 kcal/
mol (adding CH3CN as the solvent) higher in energy than
7c, while based on a similar method, we calculated the energy
of zwitterionic product of 7b, if there were any, to be higher
in energy than 7b by 39.3 kcal/mol. In comparison, the
difference of ∼5 kcal/mol manifests the stabilization energy
given by the hydrogen bonding formation in 7z. We thus
tentatively propose that the strong N(2)-H-O-N(10) in-
tramolecular hydrogen bonding plays a certain role in
Acknowledgment. We thank the National Science Coun-
cil, Taiwan, for financial support.
Supporting Information Available: Additional experi-
mental details. This material is available free of charge via
OL702482C
(9) Kone´, M.; Illien, B.; Graton, J.; Laurence, C. J. Phys. Chem. A 2005,
109, 11907-11913.
(10) Dutoi, A. D.; Head-Gordon, M. Chem. Phys. Lett. 2006, 422, 230-
233.
(11) Lynch, B. J.; Fast, P. L.; Harris, M.; Truhlar, D. G. J. Phys. Chem.
A 2000, 104, 4811-4815.
(12) Cance`s, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107,
3032-3041.
5290
Org. Lett., Vol. 9, No. 25, 2007