Diradical Related to the Neocarzinostatin Chromophore
J. Am. Chem. Soc., Vol. 120, No. 33, 1998 8481
Scheme 1
(band 28), 1239 (band 26), 1087 (band 24), 925 (band 20), 809
(band 19), and 744 (band 18) do not appear in the spectrum of
the open form 6 (see Table 2) since all of these bands are typical
of an intact ring structure with an exocycylic methylene group
(CCH2 stretch, HCring bend, ring stretch, etc., Table 1). In turn,
the easily identified asymmetric allene stretching vibration of
-
1
6
(1947 cm , band 36, Table 2) is missing in the spectrum of
compound A, thus confirming that A does not possess an acyclic
structure with an allene group. On the other hand, those
vibrational modes which are normally considered as being
typical of phenol derivatives (phenol: OH stretching at 3637
-
1
-1
-1 23
cm , HOC bending at 1176 cm , OH torsion at 309 cm ;
see also ref 24) appear in the infrared spectrum of both 5 and
-
1
6
1
3
3
: OH stretch, 3635/3670 cm (exptl/calcd); HOC bend, 1173/
-1
-1
169 cm ; OH torsion, 318 cm cal. (5, Table 1); OH stretch,
637/ 3661 cm ; HOC bend, 1175/ 1151 cm ; OH torsion,
66 cm cal. (6, Table 2). Other intensive phenol ring
vibrations (1412 (band 32), 1546 cm (band 37), Table 1),
which according to ROSS-BLYP results could be used to
identify 5, coincide with bands of 4 and, therefore, cannot be
seen in the difference spectrum.
kinetic isotope effect. However, the yield of A formed from
-1
-1
4-d3 via the deuterium shift is too low to clearly assign IR
-1
absorptions other than the OD stretching vibration to this
isotopomer, and therefore, the isotope effect could not be
quantified. The opening of the benzene ring should result in
highly unsaturated molecules with characteristic IR absorptions
such as alkene, allene, or alkyne CC stretching vibrations. Since
no IR absorptions are observed in the 1700-2000 cm range,
it is reasonable to assume that the benzene ring in A is still
intact, and we thus assign A the structure of R,3-didehydro-5-
methyl-6-hydroxybenzene (5, Scheme 1).
-1
The second photoproduct B builds up on prolonged irradiation
6-12 h) of 5 with λ > 360 nm and was identified as 3-hydroxy-
-methylhepta-1,2,4-trien-6-yne (6) (Scheme 1). Hydroxyallene
-
1
(
4
6
3
exhibits several characteristic IR absorptions: OH stretch,
638 (band 45); alkyne CH stretch, 3333 (band 44); asymmetric
We carried out density functional theory (DFT) calculations
for several C8H8O isomers to verify the assignment made for
compound A. While carbene 4 (triplet state), hydroxyallene 6,
and vinyl ketone 7 are well described with standard DFT using
-1
allene stretch, 1947 cm (band 36, see Table 2). On deuteration
(
starting from 4-d6) band 44 is not affected, revealing that the
alkyne group is not deuterated, while band 36 is red-shifted by
-1
2
0 cm . The asymmetrical CCC stretching vibration of allene
the B3LYP energy functional1
4-17
and a 6-31G(d,p) basis,
18
-1
-1
25
at 1957 cm is shifted by 17 cm in H2CCCD2, and thus,
the investigation of diradical 5 cannot be carried out at the
single-determinant DFT level. Therefore, 5 has been treated
with the recently developed restricted-open-shell singlet DFT
-1
the observed shift of 20 cm in B is in accordance with a
doubly deuterated terminal allene group. The hydroxyl function
-1
is also deuterated and νOD of band 45 shifted to 2685 cm .
(ROSS-DFT) approach, which will be described in more detail
The IR spectrum of 6 as well as that of 6-d6 is nicely reproduced
by B3LYP/6-31G(d,p) calculations (Table 2, Figure 2).
in the Experimental Section.
The ROSS-BLYP geometry of 5 (Figure 1a) is similar to the
The isotopic substitution pattern in 6-d6 confirms the mech-
anism outlined in Scheme 1. The primary step is the transfer
of one methyl-hydrogen atom of triplet carbene 4 to the carbonyl
oxygen atom to give diradical 5. Photochemical excitation is
required, however, it is not clear if the hydrogen transfer occurs
on an excited triplet surface or if this is rather a hot (triplet)
ground-state reaction.
5
MCSCF3-21G geometry of 1 and suggests that the CC bonds
adjacent to the radical centers are slightly shortened, which is
in line with the spin polarization pattern shown in Figure 1b.
In Table 1, calculated ROSS-BLYP/6-31G(d,p) vibrational
frequencies, relative intensities, and calculated isotope shifts are
compared with the available experimental data. Each vibrational
mode was characterized on the basis of an adiabatic analysis
see Table 1).22 The adiabatic stretching frequencies are
Recently we described the photochemistry of p-benzoquinone
diazide carboxylic acids as a route to 2,4-didehydrophenols.26
Thus, irradiation of 4-diazocyclohexa-2,5-diene-1-one-2-car-
boxylic acid with λ ) 435 nm cleanly yields 2,5-cyclohexadien-
(
included into Figure 1a and confirm the variation in the CC
bond as indicated in Figure 1b.
Comparison of calculated and experimental spectra provides
further proof for the existence of 5. The weak IR bands at 1249
1
-one-2-carboxylic acid-4-ylidene (8), which on 600-700-nm
irradiation eliminates CO2 and simultaneously transfers the
hydrogen atom of the carboxylic acid group to the carbonyl
oxygen atom. In contrast, the H-transfer in 4 to give diradical
5 requires UV irradiation. This difference in reactivity results
from the short nonbonding O-H distance of only 1.73 Å in 8
(
13) Gebicki, J.; Krantz, A. J. Chem. Soc., Perkin Trans. 2 1984, 1617-
1
621.
(
(
14) Becke, A. J. Chem. Phys. 1993, 98, 5648-5652.
15) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1993, 98, 11623.
(
16) Becke, A. D. Phys. ReV. A: Gen. Phys. 1988, 38, 3098-3100.
(
calculated at the B3LYP/6-31G(d) level of theory) compared
to 2.79 Å in the lowest energy conformation (C2V symmetry)
of 4 (Figure 3).
Photochemical ring cleavage of diradical 5 (Myers cyclo-
reversion) results in the formation of 6. Since the methylene
(
17) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B: Condens. Matter 1988,
3
7, 785-789.
(18) Hariharan, P. C.; Pople, J. A. Chem. Phys. Lett. 1972, 66, 217.
(19) Miehlich, B.; Stoll, H.; Savin, A. Mol. Phys. 1997, 91, 527-536.
(20) Andrews, J. S.; Murray, C. W.; Handy, N. C. Chem. Phys. Lett.
1
993, 201, 458-464.
(
21) Gr a¨ fenstein, J.; Kraka, E.; Cremer, D. Chem. Phys. Lett., in press.
22) (a) Konkoli, Z.; Cremer, D. Int. J. Quantum Chem. 1998, 67, 1-10.
(
(23) Michalska, D.; Bienko, D. C.; Abkowicz-Bienko, A. J.; Latajka, Z.
J. Phys. Chem. 1996, 100, 17786-17790.
(b) Konkoli, Z.; Larsson, J. A.; Cremer, D. Int. J. Quantum Chem. 1998,
6
2
1
7, 11-28. (c) Konkoli, Z.; Cremer, D. Int. J. Quantum Chem. 1998, 67,
9-40. (d) Konkoli, Z.; Larsson, J. A.; Cremer, D. Int. J. Quantum Chem.
998, 67, 41-55. (e) Cremer, D.; Larsson, J. A.; Kraka, E. In Theoretical
(24) Kraka, E.; Cremer, D.; Bucher, G.; Wandel, H.; Sander, W. Chem.
Phys. Lett. 1997, 268, 313-320.
(25) Eaton, D. R.; Thompson, H. Proc. R. Soc. A 1959, 250, 39.
(26) Bucher, G.; Sander, W.; Kraka, E.; Cremer, D. Angew. Chem. 1992,
104, 1225-1228; Angew. Chem., Int. Ed. Engl. 1992, 31, 1230-1233.
and Computational Chemistry, Vol. 5, Theoretical Organic Chemistry;
Parkanyi, C.; Ed.; Elsevier: Amsterdam, 1998; pp 259-327.