Novel photolactonisation from xanthenoic esters

Article abstract of DOI:10.1016/S0040-4039(01)01296-5

In the context of our work on photocleavable fragrance precursors, we have discovered a new photo-fragmentation of xanthenoic esters into xanthene- and formyl radicals. This homolytic cleavage has not been reported previously. Thus, unsaturated formyl radicals cyclise to lactones of various ring size.

Full text of DOI:10.1016/S0040-4039(01)01296-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6519–6522
Novel photolactonisation from xanthenoic esters
Caroline Plessis† and Sam Derrer*
8
Fragrance Research Chemistry, Givaudan Du¨bendorf AG, Uberlandstrasse 138, CH-8600 Du¨bendorf, Switzerland
Received 4 July 2001; accepted 19 July 2001
Abstract—In the context of our work on photocleavable fragrance precursors, we have discovered a new photo-fragmentation of
xanthenoic esters into xanthene- and formyl radicals. This homolytic cleavage has not been reported previously. Thus, unsaturated
formyl radicals cyclise to lactones of various ring size. © 2001 Elsevier Science Ltd. All rights reserved.
The free radical cyclisation to lactones is well known.1
In this way, lactones have been made from allyl a-halo-
carboxylates,1a,b haloalkyl acrylates,1c vinyl 2-haloben-
zoates,1d xanthates1f,g and phenylseleno carbonates.1h
The latter two classes require tributyltin hydride to
generate a formyl radical via homolytic cleavage of either
the SeꢀCO or the SꢀCO bond, followed by the
intramolecular addition of this radical to a car-
bonꢀcarbon double bond (Scheme 1). We have discov-
ered the photochemical formation of such formyl radicals
from xanthenoic esters and their cyclisation to lactones.
As part of our work on photocleavable fragrance precur-
sors,2 we envisaged that xanthenoic esters may release
alcohols, based on a disclosure of Barton and co-work-
ers, who observed the photolytic cleavage of xanthenoic
aryl esters into xanthene and the corresponding phenols.3
They rationalised their findings by postulating a
homolytic cleavage of the COꢀOAr bond (according to
the well precedented photochemical Fries rearrange-
ment),4 followed by decarbonylation. However, we have
found that only a small amount of alcohol is formed from
alkenyl xanthenoic esters 1a–f.5 The major products are
lactones of various ring size, depending on the location
of the unsaturation. These findings are a clear indication
that the photolytic cleavage of xanthenoic esters derived
from alcohols other than phenols occurs a to the car-
bonyl group to yield the corresponding formyl and
xanthene radicals. To the best of our knowledge this
fragmentation of xanthenoates under UV-irradiation has
not yet been described.
O
O
O
( )n
X
O
X = SePh, SC(S)OR
free radical cyclisation
( )n
initiation
O
O
O
( )n
( )n
O
We report the transformation of such compounds to
lactones. Scheme 2 illustrates the general reaction and
Scheme 1.
R2
R1
O
( )
R3
R1
n
R1
R1
( )n
R1
( )n
3a-f
O
O
( )n
R3
hν (>300 nm)
R3
( )n
R3
O
O
and/or
+
+
H
O
HO
EtOH
R3
R2
O
R2
R2
O
R2
4a-f
2a-f
O
1a-f
Scheme 2.
Keywords: xanthenoic ester; xanthenoate; photo-fragmentation; xanthene radical; formyl radical; lactonisation; lactone.
* Corresponding author. Tel.: +41 1 824 23 79; fax: +41 1 824 29 26; e-mail: samuel.derrer@givaudan.com
† Current address: MANE, 620 Route de Grasse, 06620 Le Bar-Sur-Loup, France.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01296-5

6520
C. Plessis, S. Derrer / Tetrahedron Letters 42 (2001) 6519–6522
Table 1 summarises the results.6 cis-3-Hexenyl xan-
thenoate (1a) represents a standard case where the
above mentioned reaction conditions led to the g-lac-
tone 2a, that was isolated in good yield beside some
cis-3-hexenol (3a, 10%). The formate 4a was only
observed in trace amounts. Xanthenoate 1b was studied
under various conditions: the standard procedure gave
a mixture of diastereoisomeric d-lactones 2b in 42%
yield. In this experiment, the alcohol 3b and the for-
mate 4b were isolated in 12 and 5%, respectively. Due
to the relatively long reaction times (the half-life of 1b
upon irradiation with a 150 W lamp was approximately
11 h), several alterations to the conditions were made.
age) of g-lactones at 254 nm to yield formates.7 Con-
ceivably then, irradiation of 1b at shorter wave lengths
(<300 nm, quartz cell) does either not lead to lactones,
or leads to their rapid cleavage. Thus, the initial reac-
tion conditions in ethanol turned out to be the best for
our purpose. The ( )-isopulegol-derived xanthenoate 1c
gave 50% of a 75:25 mixture of diastereoisomeric lac-
tones rac-2c. NOESY-spectroscopy confirmed the trans
ring fusion. Only little alcohol 3c and formate 4c was
formed. Irradiation of the 5-hexenol derivative 1d
yielded 34% of the eight-membered-ring lactone 2d
beside an inseparable 3:4 mixture of six- and seven-
membered lactones 2%d. The formation of the d-lactone
can be rationalised by isomerisation of the terminal
carbonꢀcarbon double bond under the reaction condi-
tions. Again, only small amounts of alcohol 3d and
formate 4d were formed.
No reaction took place in dioxane and toluene. In
acetonitrile and methanol, the disappearance of the
xanthenoate was much faster, however, neither lactone
2b nor the corresponding formate 4b could be isolated.
The only isolable and characterised product was xan-
thone (5). No lactone either was formed by performing
the reaction in a quartz vessel. Simonaitis and Pitts
described the opening (through homolytic bond cleav-
Linalyl xanthenoate 1e gave a complex mixture of
products. However, the four diastereoisomers of
bicyclic lactone 2%e were identified by GLC–MS analysis
of an enriched fraction. The combined GLC-yield
Table 1. Results of the photolysis of alkenyl xanthenoates 1a–f

C. Plessis, S. Derrer / Tetrahedron Letters 42 (2001) 6519–6522
6521
solvent
H
References
OH
O
1. (a) Halo-acetates: Nagashima, H.; Seki, K.; Ozaki, N.;
Wakamatsu, H.; Itoh, K.; Tomo, Y.; Tsuji, J. J. Org.
Chem. 1990, 55, 985–990; Belletire, J. L.; Mahmoodi, N.
O. J. Nat. Prod. 1992, 55, 194–206; Hanessian, S.; Di
Fabio, R.; Marcoux, J.-F.; Prud’homme, M. J. Org.
Chem. 1990, 55, 3436–3439; Barth, F. O.; Yang, C. Tet-
rahedron Lett. 1990, 31, 1121–1124; Lee, E.; Yoon, C.
H.; Lee, T. H. J. Am. Chem. Soc. 1992, 114, 10981–
10933; Lee, E.; Yoon, C. H.; Lee, T. H.; Kim, S. Y.;
Ha, T. J.; Sung, Y.-S.; Park, S.-H.; Lee, S. J. Am.
Chem. Soc. 1998, 120, 7469–7478; Terent’ev, A. B.;
Vasil’eva T. T.; Kuz’mima, N. A.; Ikonnikov, N. S.;
Orlova, S. A.; Mysov, E. I.; Belokon’, Y. U. Russ.
Chem. Bull. 1997, 46, 2096–2098.(b) Malonates: Back, T.
G.; Gladstone, P. L. Synlett 1993, 699–700; Back, T. G.;
Gladstone, P. L.; Parvez, M. J. Org. Chem. 1996, 61,
3806–3814; Baldovini, N.; Bertrand, M.-P.; Carrie`re, A.;
Nouguier, R.; Plancher, J.-M. J. Org. Chem. 1996, 61,
3205–3208; Snider, B. B.; McCarthy, B. A. Tetrahedron
1993, 49, 9447–9452; Snider, B. B.; Kiselgof, J. Y. Tet-
rahedron 1998, 54, 10641–10648.(c) Halo-alkyl ‘acrylates’:
Haaima, G.; Routledge, A.; Weavers, R. T. Tetrahedron
Lett. 1989, 30, 5159–5162; Haaima, G.; Weavers, R. T.
Tetrahedron Lett. 1988, 29, 1085–1088; Lynch, M.-J.;
Simpson, J.; Weavers, R. T. Aust. J. Chem. 1993, 46,
203–212; Haaima, G.; Hanton, L. R.; Lynch, M.-J.;
Mawson, S.; Routledge, A.; Weavers, R. T. Tetrahedron
1994, 50, 2161–2174; Clive, D. L. J.; Zhang, J. J. Chem.
Soc., Chem. Commun. 1997, 549–550; Porter, N. A.;
Chang, V. H.-T. J. Am. Chem. Soc. 1987, 109, 4976–
4981; Ihara, M.; Setsu, F.; Shohda, M.; Taniguchi, N.;
Fukumoto, K. Heterocycles 1994, 37, 289–292.(d) o-
Bromo-benzoate: Zhang, W.; Pugh, G. Tetrahedron Lett.
1999, 40, 7595–7598.(e) Xanthyl-alkyl acrylates:
Vela´squez, S.; Huss, S.; Camarasa, M.-J. J. Chem. Soc.,
Chem. Commun. 1991, 1263–1265.(f) Xanthates: Saicic,
R.; Zard, S. Z. J. Chem. Soc., Chem. Commun. 1996, 14,
1631–1632.(g) Thionolactones: Iwasa, S.; Yamamoto, M.;
Kohmoto, S.; Yamada, K. J. Org. Chem. 1991, 56,
2849–2853.(h) Seleno derivatives: Bachi, M. D.; Bosch E.
J. Org. Chem. 1992, 57, 4696–4705; Bachi, M. D.;
Bosch, E. Heterocycles 1989, 28, 579–582.
'tandem' radical
cyclisation
O
O
O
2'e
formyl radicals
OH
O
O
OH
6
Scheme 3.
amounts to about 9%. Linalool (3e) was obtained in 4%
beside alcohol 6 (7%). The latter resulted from inter-
molecular radical reaction with ethanol (Scheme 3).
None of the formate 4e was observed. A mechanism for
the formation of these bicyclic lactones is proposed in
Scheme 3, involving a ‘tandem’ radical addition first to
the vinylic then to the prenylic carbonꢀcarbon double
bond. Such a behaviour of formyl radicals derived from
xanthates was also observed by Iwasa and co-work-
ers.1g Finally, 10-undecenol-derived xanthenoate 1f
gave only 6% of the expected 13-membered ring lactone
2f, showing the limitations of this method. The forma-
tion of formate 3f (12%) as well as intermolecular side
reactions became predominant. Among the by-products
we identified in particular 7 (12%) and 8 (5%). 10-
Undecenol (3f) was obtained in 13% yield.
O
OH
O
O
7
O
5
OH
8
2. See for example: Anderson, D.; Fra´ter, G. US Patent
6,096,918, 1999.
3. Barton, D. H. R.; Chow, Y. L.; Cox, A.; Kirby, G. W.
J. Chem. Soc. 1965, 3571–3578.
In conclusion, we have shown for the first time that
alkenyl xanthenoates fragment under UV-irradiation
(l>300 nm) to generate xanthene and unsaturated for-
myl radicals that may cyclise to give the corresponding
lactones in acceptable yields. The optimisation of the
system with regard to reaction conditions and additives,
as well as the alteration of the chromophore are subject
of continued efforts.
4. Kobsa, H. J. Org. Chem. 1962, 27, 2293.
5. The xanthenoates, prepared in good yields via the DCC-
mediated coupling of the xanthenoic acid with the alco-
hols, were fully characterised by 1H and 13C NMR, IR
and mass spectroscopy.
6. Approximately three-millimolar solutions of the starting
materials 1a–f in ethanol were irradiated with a 150 W
medium pressure Hg-lamp in a water cooled Pyrex®
photoreactor (>300 nm). Larger scale experiments were
conducted with a 700 W lamp. A typical experiment
consists of irradiating an ethanolic solution of the xan-
thenoate 1a–f (3 g in 3 l) with a 700 W lamp. The
conversion was monitored by GLC-analysis. Reaction
times varied from 48 to 96 h for a conversion of >80%.
Acknowledgements
We thank Professor Dr. A. Vasella, ETH Zu¨rich
(Switzerland) for his support. The help rendered by Dr.
J. Schmid, Fragrance Research Analytics, for GLC–MS
support was greatly appreciated.

6522
C. Plessis, S. Derrer / Tetrahedron Letters 42 (2001) 6519–6522
The solvent was evaporated and the residue taken up in
1H and 13C NMR, IR and MS, the others were identified
by GLC or GLC–MS from enriched fractions (all for-
mates 4 were prepared independently from their corre-
sponding alcohols for GLC-identification purposes). Solid
xanthone was obtained by further Kugelrohr distillation
and the structure of other xanthene-derivatives was tenta-
tively assigned on the basis of a GLC–MS analysis.
7. Simonaitis, R.; Pitts, J. N. J. Am. Chem. Soc. 1969, 91,
108–112.
MTBE (200 ml). This was washed with saturated aq.
NaHCO3 (200 ml) and brine (100 ml), dried (MgSO4)
and concentrated in vacuo. The crude, a yellow oil, was
diluted with poly(ethylene glycol) (Mn ꢀ400, 5 g) and
the volatiles collected by Kugelrohr distillation under
reduced pressure. The products were separated by flash
chromatography on silica. Compounds 2a–d and 2f, 2%d,
3a–d, 4b and 5 were isolated and fully characterised by