Perfect switching of photoreactivity by acid: Photochemical decarboxylation versus transesterification of mesityl cyclohexanecarboxylate

Article abstract of DOI:10.1021/ol0065266

(matrix presented) Mesityl cyclohexanecarboxylate has been photodecarboxylated upon excitation at 254 nm to form cyclohexylmesitylene in good yield in neutral acetonitrile solutions. In the presence of a catalytic amount of the acid and ethanol as a nucleophile, however, the same compound undergoes facile transesterification upon irradiation, affording its ethyl ester in almost quantitative yield.

Full text of DOI:10.1021/ol0065266

ORGANIC
LETTERS
2000
Vol. 2, No. 21
3401-3404
Perfect Switching of Photoreactivity by
Acid: Photochemical Decarboxylation
versus Transesterification of Mesityl
Cyclohexanecarboxylate
,†
,†,‡
Tadashi Mori,* Takehiko Wada,† and Yoshihisa Inoue*
Department of Molecular Chemistry, Osaka UniVersity, 2-1 Yamada-oka,
Suita 565-0871, Japan, and Inoue Photochirogenesis Project, ERATO, JST,
4-6-3 Kamishinden, Toyonaka 565-0085, Japan
tmori@ap.chem.eng.osaka-u.ac.jp
Received August 31, 2000
ABSTRACT
Mesityl cyclohexanecarboxylate has been photodecarboxylated upon excitation at 254 nm to form cyclohexylmesitylene in good yield in
neutral acetonitrile solutions. In the presence of a catalytic amount of the acid and ethanol as a nucleophile, however, the same compound
undergoes facile transesterification upon irradiation, affording its ethyl ester in almost quantitative yield.
Photochemical transformations of carboxylic esters have been
studied extensively in terms of cycloaddition, hydrogen
abstraction, isomerization, and rearrangement.1 However, due
to the lack of intense absorption bands in the conventional
ultraviolet region for a simple carboxyl chromophore, the
previous photochemical studies have concentrated mostly on
arenecarboxylates, in which the carboxyl chromophore is
conjugated with the aromatic moiety, providing a readily
accessible π,π* transition.2 Consequently, the photochemistry
of such aromatic esters has attracted much attention from
the mechanistic and synthetic points of view, but unfortu-
nately, a complex mixture of several minor products is
always obtained with a poor material balance.1 For instance,
the photolysis of moderately hindered 2,4-alkylphenyl (aryl)-
alkanoate leads to at least five competing reactions.3 In
contrast to the several competing photochemical reactions
reported for most of the aromatic esters mentioned above,
mesityl benzoate has been reported to undergo photohy-
drolysis in aqueous acetonitrile in the absence of acid, but
with poor product yield (7%).4
Herein, we report an unprecedented photoreactivity switch-
ing of an aryl alkanoate controlled by the acid concentra-
(2) A Study of photodecarboxylation of simple aliphatic carboxylic acids
was carried out in the gas phase, and in general the reaction is more
complicated due to several competing reactions. For reviews: (a) Coyle, J.
D. Chem. ReV. 1978, 78, 97. (b) Budac, D.; Wan, P. J. Photochem.
Photobiol. A: Chem. 1992, 67, 135. (c) Reference 1b, p 384 (Chapter 31).
(3) Bradshaw, J. S.; Loverridge, E. L.; White, L. J. Org. Chem. 1968,
33, 4127.
† Osaka University.
‡ ERATO.
(1) For general reviews on the photochemistry of carboxylic acids and
esters, see: (a) Givens, R. S.; Levi, N. In The Chemistry of Functional
Groups, Supplement B, The Chemistry of Acid DeriVatiVes; Patai, S., Ed.;
John Wiley & Sons, Ltd.: London, 1979; p 641. (b) Pincock, J. A. In CRC
Handbook of Organic Photochemistry and Photobiology; Horspool, W. M.,
Song, P.-S., Eds.; CRC Press: Boca Raton, 1995; p 393 (Chapter 32).
(4) Pacifici, J. G.; Zannucci, J. S.; Lappin, G. R.; Ownby, J. C.; Kelly,
C. A. Mol. Photochem. 1971, 3, 175.
10.1021/ol0065266 CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/28/2000

tion: from decarboxylation in neutral acetonitrile to trans-
esterification in acidic acetonitrile containing ethanol, both
of which give their respective products in quantitative yields.
The results, exemplified by mesityl cyclohexanecarboxylate
(1) in Scheme 1, provide not only a convenient, powerful
Table 1. Competitive Photochemical Transesterification and
Decarboxylation of Mesityl Cyclohexanecarboxylate (1) in the
Presence/Absence of Acida
yield, %
additive
none
(mM)
temp, °C
convn, %
2
3
4
(0)b
(0)b
0
0
0
0
0
91
86c
91
92
96
34
93
93
84d
84c
86
89
94
94
0
0
0
0
4
0
91
82
91
48
5
2
19
0
0
45
0
0
0
Scheme 1
MeSO3H
(0.0036)
(0.034)
(0.051)
(0.062)
(0.062)
(0.17)
(0.23)
(0.93)
(1.02)
(4.92)
(5.49)
(0.64)
33
87
27
13
88
78e
38
84
83
94
94
34
91
18
64
93
83
37
86
89
72
90
-40
25
0
0
0
0
0
0
0
CF3CO2H
BF3/OEt2
0
tool for controlling the photoreactivity of aromatic esters but
also a versatile photolabile protectiving group5 which is
readily converted to other ester groups upon irradiation in
the presence of an appropriate acidic alcohol. Alternatively,
decarboxylation to the corresponding alkylmesitylene may
be performed in neutral solvents.
a Irradiated for 2 h in quartz tubes under Ar with a low-pressure mercury
lamp (254 nm) in acetonitrile containing 1 (1.8-2.3 mM) and ethanol (0.34
M), unless otherwise indicated. b No ethanol added. c Irradiation was
performed in pentane. d In 2-propanol (0.34 M) instead of ethanol. e Yield
of isopropyl ester.
First of all, it should be emphasized that thermally the
mesityl ester 1 was entirely stable under the acidic conditions
employed in the present study, i.e., acetonitrile solutions
containing 0.34 M ethanol and e 5 mM methanesulfonic
acid, and was recovered in >99% yield after 72 h of standing
in the dark at 25 °C. Even in the presence of 20 mM of the
acid, no transesterification product was detected on GC after
1 week at 25 °C. In photochemical runs, acetonitrile solutions
of 1 (2 mM) containing ethanol (0.34 M) were irradiated at
254 nm under an argon atmosphere, mostly at 0 °C, in the
presence and absence of acid (0-5 mM). The irradiations
led to the efficient formation of the transesterification
products, i.e., ethyl cyclohexanecarboxylate (2) and mesitol
(3), and/or the decarboxylation products (4), depending on
the conditions employed. The conversions and product yields
obtained after 2 h irradiation under a representative variety
of conditions are listed in Table 1.
In the absence of acid, the photolysis of 1 exclusively
afforded the decarboxylation products, i.e., 2,4,6-trimethyl-
1-cyclohexylbenzene (4a) and its positional isomers (4b-
d), in a ratio of 87:3:2:8; 4a was identified by comparison
with an authentic specimen synthesized independently,6 while
the latter three were tentatively assigned to the positional
isomers of 4a, produced in the secondary photorearrangement
upon prolonged irradiations7 on the basis of the GC retention
times and mass spectra.
of photolysis from decarboxylation to transesterification,
maintaining an equally excellent material balance (Scheme
1). As shown in Table 1, even in the presence of methane-
sulfonic acid concentrations as low as 0.034 mM, the
photolysis afforded the transesterification (2 and 3) and
decarboxylation (4) products in comparable yields. Upon
increasing acid concentrations of up to 0.10 mM, the yields
of 2 and 3 rapidly dominate, at the expense of 4, reaching a
plateau thereafter. Other typical Brønsted and Lewis acids,
such as trifluoroacetic acid and trifluoroborane etherate, also
switched the photoreaction mode to the transesterification,
affording 2 and 3 in excellent yields. The use of 2-propanol
instead of ethanol also led to the smooth formation of the
corresponding ester, though at slightly slower rates but in
correspondingly excellent yields under similar conditions.
The irradiation temperature moderately affected both the
reaction rate and the product distribution. At -40 °C the
reaction rate (consumption of the starting material) was
significantly reduced, although the material balance remained
good-to-excellent. At 25 °C the reaction rate increased;
however, the material balance (particularly that of yield of
2) was lowered and the decarboxylation product 4 was
appreciably favored. Both the decreased yield of 2 and the
increased yield of 4 at 25 °C may be rationalized by an
increased escape of the radical pairs from the solvent cage,
resulting in the enhanced decarbonylation8 and decarboxy-
In contrast, the addition of a very small amount of acid to
the system (in the µM order) dramatically switched the mode
(7) The three positional isomers 4b-d are known to be formed through
the 1,2-shift of the alkyl group via a benzvalene intermediate. Indeed, control
experiments with pure 4a under similar photochemical conditions gave the
same isomers. See: (a) Kaplan, L.; Wilzbach, K. E.; Brown, W. G.; Yang,
S. S. J. Am. Chem. Soc. 1965, 87, 675. (b) Burgstahler, A. W.; Chien, P.-
L. J. Am. Chem. Soc. 1964, 86, 2940. (c) Wilzbach K. E.; Kaplan, L. J.
Am. Chem. Soc. 1964, 86, 2307. Thus, the isomers 4b-d are tentatively
assigned to 2,3,5-, 2,3,6-, and 2,4,5-trimethyl-1-cyclohexylbenzenes.
(5) For a general review on photoprotecting groups: Pillai, V. N. R.
Org. Photochem. 1987, 9, 225.
(6) (a) Kim, J. N.; Chung, K. H.; Ryu, E. K. Tetrahedron Lett. 1994,
35, 903. (b) Kotsuki, H.; Ohishi, T.; Inoue, M.; Kojima, T. Synthesis 1999,
603. (c) Kotsuki, H.; Ohishi, T.; Inoue, M. Synlett 1998, 255. (d)
Hickinbottom, W. J.; Rogers, N. W. J. Chem. Soc. 1957, 4124.
3402
Org. Lett., Vol. 2, No. 21, 2000

Scheme 2
lation of the relevant precursors 5 and 8 at the elevated
temperature.
triplet sensitizations with acetophenone and fluorene13 have
failed to give any product under similar conditions, ester 1
being completely recovered. In search of transient species
formed upon irradiation, we performed spectroscopic moni-
toring of the sample solution during irradiation to find a
moderate absorption band centered at ca. 320 nm, which is
assigned as the n,π* transition of a cyclohexadienone
derivative.14 On the basis of the smooth reactivity switching
with increasing acid concentration and the spectroscopic
evidence, we tentatively propose the mechanism shown in
Scheme 2 for this photoreaction.15
The solvent effect provides us with valuable information
about the nature of the intermediates. As shown in Table 1,
the photolysis of 1 in acetonitrile ([MeSO3H] ) 1.02 mM)
exclusively afforded the transesterification products 2 and
3, whereas the formation of decarboxylation product 4 was
favored upon irradiation in pentane ([MeSO3H] ) 0.93 mM),
yet with comparable conversion for both solvents. This can
be accounted for in terms of a charge-transfer character of
intermediate 5, i.e., [acylδ+ ‚‚‚ phenoxyδ-].9
To elucidate the origin and mechanism of this unique pH-
controlled photoreactivity switching, spectroscopic examina-
tions of 1 in the ground and excited states were carried out
in the presence/absence of acid in acetonitrile containing
ethanol (0.34 M). In the UV spectral examinations of 1 with
added acid, no noticeable ground-state interactions were
observed between 1 and methanesulfonic acid concentrations
of up to 0.7 M. However, the fluorescence spectrum did not
appear to be influenced by the addition of acid up to a
concentration of 0.03 M. Further addition of the acid (0.1-
2.2 M) caused a bathochromic shift of the fluorescence peak
from 290 to 296 nm, as well as gradual significant decreases
in fluorescence intensity. The fluorescence quenching be-
havior of 1 with higher concentrations of methanesulfonic
acid was analyzed quantitatively to give a linear Stern-
Volmer plot with a slope of 1.10 M-1. From the fluorescence
lifetime of 1 (τ ) 1.7 ns), measured independently by the
single-photon-counting technique,10 we could determine the
quenching rate constant as kq ) 6.5 × 108 M-1 s-1, which
is lower by more than 1 order of magnitude than the
diffusion-controlled rate constant in acetonitrile (1.9 × 1010
M-1 s-1).11
Since the reactivity switching from decarboxylation to
transesterification is complete at acid concentrations as low
as 0.1 mM, and no appreciable fluorescence quenching is
observed at such a low acid concentration, the excited singlet
state of 1 cannot be the immediate precursor to 2/3 or 4. A
mechanism postulating the formation of a ketene intermediate
and the subsequent alcoholysis has been proposed,12 but this
possibility is clearly ruled out in the present case since
deuterium incorporation was not found in the product 2 when
the photolysis was carried out in ethanol-O-d. Attempted
In neutral solutions, the singlet excited state of 1 undergoes
three possible processes, i.e., deactivation to 1 and two types
of bond dissociations, giving the radical pair intermediates
5 and 8. The radical pair 5 may afford the cyclohexadienone
derivative 6, which is similar to the intermediate postulated
in the photo-Fries rearrangement. Both 5 and 6 can easily
revert to the starting material 1 in the absence of any trapping
agent such as a proton.16 On the other hand, the radical pair
(8) (a) Finnegan, C. C.; Kuntson, D. Chem. Commun. 1966, 172. (b) J.
Am. Chem. Soc. 1967, 89, 1970. (c) Barton, D. H. R.; Chow, Y. L.; Cox,
A.; Kirby, G. W. Tetrahedron Lett. 1962, 1055. (d) Barton, D. H. R.; Chow,
Y. L.; Cox, A.; Kirby, G. W. J. Chem. Soc. 1965, 3571. (e) Horspool, W.
M.; Pauson, P. L. J. Chem. Soc. 1965, 5162.
(9) Coppinger and Bell claimed that in photo-Fries rearrangement the
radical pair should exist on the experimental basis but it is very hard to
explain why the two components in the pair always remain associated only
in terms of solvent cage. Hence, they envisaged the charge-transfer
interaction developed in the radical pair on the basis of the substituent effect
on the quantum yield. Coppinger, G. M.; Bell, E. R. J. Phys. Chem. 1966,
70, 3479.
(10) Mori, T., unpublished results: τ ) 1.7 ns (25 °C) at 0.28 mM in
MeCN in the presence of EtOH (0.34 M) and τ ) 1.6 ns in the presence of
MeSO3H (57 mM) at 25 °C.
(11) Murov, S. L.; Carmichael, I.; Hug, G. L. In Handbook of
Photochemistry, 2nd ed.; Mercel Dekker: New York, 1993; p 210.
(12) Gutsche, C. D.; Oude-Alink, A. M. J. Am. Chem. Soc. 1968, 90,
5855.
(13) Triplet state energies (ET) of sensitizers are 311 and 284 kJ/mol,
respectively. ET of ester 1 is assumed to be ∼320 kJ/mol. See, ref 11, Section
1.
(14) (a) Grabner, G.; Ko¨hler, G.; Marconi, G.; Monti, S.; Venuti, E. J.
Phys. Chem. 1990, 94, 3609. (b) Weiner, S. A.; Mahoney, L. R. J. Am.
Chem. Soc. 1972, 94, 5029. See also: (c) Mahoney, L. R.; DaRooge, M.
A. J. Am. Chem. Soc. 1970, 92, 890. (d) Becker, H.-D. J. Org. Chem. 1965,
30, 982.
(15) We thank reviewers of ACS for suggestions and comments on the
reaction mechanism.
(16) (a) Jackson, L. B.; Waring, A. J. J. Chem. Soc., Perkin Trans. 1
1988, 1791. (b) Waring, A. J.; Zaidi, J. H. J. Chem. Soc., Perkin Trans. 1
1985, 631.
Org. Lett., Vol. 2, No. 21, 2000
3403

8 either regenerates 1 or decarboxylates to give radical pair
9, which in turn recombine in the solvent cage to afford 4.
In the presence of acid, the labile intermediate 6 is trapped
by acid and the resulting intermediate 7 undergoes solvolysis
to give the products 2 and 3.17
In the present study, we have demonstrated that the
presence of acid completely switches the photoreactivity of
mesityl alkanecarboxylate from decarboxylation to transes-
terification. In this context, recent reports18 by Hoffmann
and Pete of the acid-induced intramolecular photocycload-
dition of alkenyloxybenzene derivatives are quite intriguing
from the viewpoint of controlling photobehavior by acids.
This research shows that acids can play a more vital role in
photochemistry, with the active control of photobehavior by
acids applied as a more general idea to a much wider range
of photoreactions. Work on the scope, limitations, and
detailed mechanism of the present acid-controlled photo-
behavior is currently in progress.
Acknowledgment. We thank Professor Qing-Xiang Guo
and Mr. Tingwei Mu at the University of Science and
Technology of China for ab initio calculations and Dr. Guy
A. Hembury and Dr. Michael Oelgemo¨ller at the Inoue
Photochirogenesis Project for valuable discussion and as-
sistance in the preparation of this manuscript. Financial
support (to T.M.) of this work from a Grant-in-Aid for
Scientific Research of the Ministry of Education, Science,
Sports and Culture of Japan (No. 12740346) and from the
Shorai Foundation for Science and Technology are gratefully
acknowledged.
(17) Bicyclic biradical/zwitterionic intermediates bearing a fused-oxetane
ring, such as 10 in the present case, have been suggested previously in the
photo-Fries rearrangement of phenyl esters. The radical pair intermediate
5 is accepted in general in the mechanism of photo-Fries rearrangement,
and indeed, ab initio calculations clearly demonstrate that the fused-oxetane
intermediate 10 (∆Hf ) -91.4 kJ mol-1, relative to the energy of excited
singlet state of 1) is less stable than the radical pair 5 (∆Hf ) -201.6 kJ
mol-1). However, 10 is much more stable than the radical pair 8 (∆Hf )
-6.0 kJ mol-1), suggesting that intermediate 10 could not rigorously be
ruled out from the energetic point of view. See: (a) Reese, C. B.; Anderson,
J. C. J. Chem. Soc. 1963, 1781. (b) Sander, M. R.; Hedaya, E.; Trecker, D.
J. J. Am. Chem. Soc. 1968, 90, 7249. (c) Sander, M. R.; Trecker, D. J. J.
Am. Chem. Soc. 1967, 89, 5725.
Supporting Information Available: Figures of the
spectral changes during the irradiation of mesityl cyclohex-
anacarboxylate 1. This material is available free of charge
via the Internet at http://pubs.acs.org.
(18) (a) Hoffman, N.; Pete, J.-P. J. Org. Chem. 1997, 62, 6952. (b)
Hoffman, N.; Pete, J.-P. Tetrahedron Lett. 1998, 39, 5027.
OL0065266
3404
Org. Lett., Vol. 2, No. 21, 2000