5768
H. Koroniak et al. / Tetrahedron Letters 45 (2004) 5767–5769
O
F
O
F
CH3
O
F
CH3
O
CF3
N
N
h , λ>300 nm
ν
CF3
F
H O
2
N
N
CH3
CH3
6
7
O
F
CH3
N
CF3
O
N
F
CH3
8
was expected to undergo further transformations.
Moreover the presence of fluorine atoms in the cyclo-
butene moiety should stabilize its structure.4a Surpris-
well-known reaction, we believe that in this case tor-
quoselectivity10 seems to be an important and domi-
nating factor. It seems that in the ring opening of the
intermediate cyclobutene 8, among the two substituents
–CF3 and –F responsible for the isomerization, the
fluorine being an efficient p-electron donor will prefer-
entially rotate outward, forcing the CF3 group to rotate
inward. As a result formal trans–cis isomerization is
observed. It is just another experimental example sup-
porting the torquoselectivity prediction based on com-
putational calculations.10
ingly, during irradiation5 of
1
with UV light
(k > 300 nm) in water, we isolated 1,3-dimethyl-(5,6-di-
hydrourac-6-yl)-difluoroacetic acid 2, as the only stable
product.6 Irradiation of the same substrate 1 with higher
energy light (k ¼ 254 nm) led to cleavage of the C(5)–
C(trifluorovinyl) bond and gave 1,3-dimethyluracil as
the sole product.
The mechanism of these transformations seems to in-
volve a Michael type addition–elimination reaction after
the photochemical electrocyclization. The more stable
keto 5 versus enol 4 tautomer then reacts with a nucleo-
phile (water) leading to the most thermodynamically
stable product––a derivative of difluoroacetic acid 2. In
the case of irradiation with higher energy light (low-
pressure mercury lamp, k ¼ 254 nm) simple cleavage of
a C–C bond occurs.
References and notes
1. (a) De Clerq, E.; Descamps, J.; De Somer, P.; Barr, P. J.;
Jones, A. S.; Walker, R. T. Proc. Natl. Sci. U.S.A. 1979, 76,
2947–2951; (b) Jones, A. S.; Rahim, S. G.; Walker, R. T.
J. Med. Chem. 1981, 24, 759–760; (c) Coe, P. L.; Harnden,
M. R.; Jones, A. S.; Noble, S. A.; Walker, R. T. J. Med.
Chem. 1982, 25, 1329–1334; (d) De Clerq, E.; Desgranges,
C.; Herdewijn, P.; Sim, I. S.; Jones, A. S.; McLean, M. J.;
Walker, R. T. J. Med. Chem. 1986, 29, 213–217.
2. (a) Koroniak, H.; Fiedorow, P.; Pluskota, D.; Karwatka,
P.; Abboud, K. A. J. Mol. Struct. 1994, 323, 215–221; (b)
Koroniak, H.; Karwatka, P.; Pluskota, D.; Fiedorow, P.;
Jankowski, J. J. Fluorine Chem. 1995, 71, 135–137.
It would appear that the final product 2 is formed via the
labile intermediate 3, the structure of which we were not
able to determine experimentally. We suggest that the
keto 5–enol 4 equilibrium should favour the keto tau-
tomer 5, which can sometimes be reversed.4 On the basis
of DFT calculations, however, keto form 5 is 0.4 kcal/
mol (1.7 kJ/mol) more stable than the enol form 4.7
3. Welch, J.; Eswarakrishan, S. Fluorine in Bioorganic
Chemistry; John Wiley and Sons, 1991.
4. (a) Lemal, D. M. Acc. Chem. Res. 2001, 34, 662–671; (b)
Lindner, P. E.; Correa, R. A.; Gino, J.; Lemal, D. M.
J. Am. Chem. Soc. 1996, 118, 2556–2563; (c) Lindner,
P. E.; Lemal, D. M. J. Org. Chem. 1996, 61, 5109–5115.
5. Photoirradiations were carried out in a water cooled
reactor with an immersed medium pressure mercury lamp
using a 1 mm Pyrex filter which provided UV light with
k > 300 nm. The phototransformations were followed by
HPLC analysis. The concentration of the irradiated water
solutions of 1, 6 and 7 was in the range of 10ꢀ3 M (usually
1 mM), due to the solubility of the starting material.
6. Pure 2 was isolated from the irradiated solution of 1. After
1.5 h irradiation of 1, the solvent (water) was removed in
vacuo. The solid precipitate was collected and showed to
consist of 2 and unreacted traces of 1. After column
chromatography (to remove unreacted 1) and recrystalli-
zation, samples of pure 2 (purity checked by HPLC,
reverse phase C-18 column) were submitted for analysis.
These findings are in contrast to the photochemical
transformations
of
1,3-dimethyl-5-(E-pentafluoro-
propenyl)uracil. It is surprising, that in this case only
E–Z isomerization was observed. No traces of inter-
mediates analogous to 4 or 5 were found in the reaction
mixture. The irradiation of a water solution of 6 led to a
photostationary state involving equilibration with 1,3-
dimethyl-5-(Z-pentafluoropropenyl)uracil 7, where the
more sterically congested 7 is the major product (HPLC
ratio 6/7 ¼ 1:1.54).9 Irradiation of 7 as the starting
material also led to the identical photostationary state.
This experimental observation suggests that the uni-
molecular transformation is much faster than the
bimolecular nucleophilic reaction with molecules of
solvent (water), as observed in the case of 1. Although
trans–cis photochemically induced isomerization is a
1
Yield (isolated) 55%. H NMR, (DMSO-d6, in ppm in d
scale downfield from TMS), 2.60 (br m, 2H, C5–H2), 2.96