Photochemical [2 + 2] cycloaddition of acetylene to chiral 2(5H)-furanones

Article abstract of DOI:10.1021/jo0264731

The [2 + 2] photocycloaddition of acetylene to chiral 2(5H)-furanones was investigated. The influence on the chemical yield and facial diastereoselectivity of the substituent at the stereogenic center and also the effect of a 4-methyl group were evaluated. A mechanistic proposal based on a simple theoretical conformational analysis is presented. Using a C₂-symmetric bis(lactone) as the substrate, a diastereomeric excess higher than 98% was found.

Full text of DOI:10.1021/jo0264731

P h otoch em ica l [2 + 2] Cycloa d d ition of Acetylen e to Ch ir a l
2(5H)-F u r a n on es
Ramon Alibe´s,† Pedro de March,† Marta Figueredo,† J osep Font,*,† Xiaolin Fu,†
Marta Racamonde,† AÄ ngel AÄ lvarez-Larena,‡ and J uan F. Piniella‡
Departament de Quı´mica and Unitat de Cristal‚lografia, Universitat Auto`noma de Barcelona,
08193 Bellaterra, Spain
josep.font@uab.es
Received September 25, 2002
The [2 + 2] photocycloaddition of acetylene to chiral 2(5H)-furanones was investigated. The influence
on the chemical yield and facial diastereoselectivity of the substituent at the stereogenic center
and also the effect of a 4-methyl group were evaluated. A mechanistic proposal based on a simple
theoretical conformational analysis is presented. Using a C2-symmetric bis(lactone) as the substrate,
a diastereomeric excess higher than 98% was found.
CHART 1
In tr od u ction
The photochemical [2+2] cycloaddition of alkenes to
cyclic enones and R,â-unsaturated lactones is a well-
established approach to preparing cyclobutane com-
pounds.1 This reaction has been widely applied to natural
product synthesis.2 The high levels of regio- and stereo-
selectivity usually achieved and the synthetic versatility
of the adducts render this method synthetically attrac-
tive.3,4 Thus, in the course of our research program on
the stereoselective synthesis of naturally occurring cyclo-
butane pheromones, we have studied extensively the
photocycloaddition of homochiral 2(5H)-furanones to eth-
ylene and other olefins,5 and we have used this reaction
as a key step in the synthesis of both enantiomers of
grandisol.6
of acetylene to 2(5H)-furanones was envisaged as an
appropriate approach to such compounds. So far, this
reaction has received little attention because the number
of reported examples is quite limited9 and, to the best of
our knowledge, a chiral version has not yet been exam-
ined. Therefore, we have prepared the enantiopure
lactones 1a -e and 2a -b (Chart 1) and investigated the
steric course of their photochemical reaction with acety-
lene, which should lead to enantiomerically pure cyclo-
butenes fused to γ-lactones. In this paper, we describe
in detail10 the results of this study designed to evaluate
the factors controlling the facial diastereoselectivity of
the photocycloaddition and also to define the scope of
such a process as a practical synthetic method.
Recently, we have broadened the scope of our research
to the synthesis of chiral polyfunctionalized cyclobutenes.
The utility of such compounds as the precursors of a
variety of products with potential biological activity has
been described.7,8 The stereoselective photocycloaddition
† Departament de Qu´ımica.
‡ Unitat de Cristal‚lografia.
(1) (a) Corey, E. J .; Bass, J . D.; LeMahieu, R.; Mitra, R. B. J . Am.
Chem. Soc. 1964, 86, 5570. (b) Baldwin, S. W. In Organic Photochem-
istry; Padwa, A., Ed.; Marcel Dekker: New York, 1981; p 123. (c)
Crimmins, M. T. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Paquette, L. A., Eds.; Pergamon Press: Oxford, 1991; Vol.
5, p 123.
(2) (a) Demuth, M.; Mikhail, G. Synthesis 1989, 145-162. (b) Bach,
(7) (a) J ung, M. E.; Sledeski, A. W. J . Chem. Soc., Chem. Commun.
1993, 589. (b) Gourdel-Martin, M.-E.; Huet, F. J . Org. Chem. 1997,
62, 2166.
(8) (a) Cadogan, J . I. G.; Cameron, D. K.; Gosney, I.; Tinley, E. J .;
Wyse, S. J .; Amaro, A. J . Chem. Soc., Perkin Trans. 1 1991, 2081. (b)
Binns, F.; Hayes, R.; Hodgetts, K. J .; Saengchantara, S. T.; Wallace,
T. W.; Wallis, C. J . Tetrahedron 1996, 52, 3631.
(9) (a) Kosugi, H.; Sekiguchi, S.; Sekita, R.; Uda, H. Bull. Chem.
Soc. J pn. 1976, 49, 520. (b) Avetisyan, A. A.; Margaryan, A. K.;
Nalbandyan, G. K.; Avetisyan, T. V. Khim. Geterotsikl. Soedin. 1986,
10, 1315.
(10) The preliminary results of this study have been published:
Alibe´s, R.; de March, P.; Figueredo, M.; Font, J .; Racamonde, M.
Tetrahedron Lett. 2001, 42, 6695.
T. Synthesis 1998, 683.
(3) (a) Tanaka, M.; Tomioka, K.; Koga, K. Tetrahedron Lett. 1982,
23, 3401. (b) Tomioka, K.; Tanaka, M.; Koga, K. Chem. Pharm. Bull.
1989, 37, 1201. (c) Tanaka, M.; Tomioka, K.; Koga, K. Tetrahedron
1994, 50, 12829.
(4) (a) Hoffmann, N.; Scharf, H.-D. Tetrahedron Lett. 1989, 30, 2637.
(b) Hoffmann, N.; Scharf, H.-D. Liebigs Ann. Chem. 1991, 1273.
(5) (a) Alibe´s, R.; Bourdelande, J . L.; Font, J . Tetrahedron: Asym-
metry 1991, 2, 1391. (b) Alibe´s, R.; Bourdelande, J . L.; Font, J .; Gregori,
A.; Parella, T. Tetrahedron 1996, 52, 1267. (c) Gregori, A.; Alibe´s, R.;
Bourdelande, J . L.; Font, J . Tetrahedron Lett. 1998, 39, 6961.
(6) Alibe´s, R.; Bourdelande, J . L.; Font, J .; Parella, T. Tetrahedron
1996, 52, 1279.
10.1021/jo0264731 CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/11/2003
J . Org. Chem. 2003, 68, 1283-1289
1283

Alibe´s et al.
other products
TABLE 1. P h otocycloa d d ition of La cton es 1a -e to Acetylen e
entry
substrate
R1
R2
solvent
filter
time
% yielda
3:4 anti:synb
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1a
1b
1c
1d
1e
CO-t-Bu
CO2Mnt
COPh
TBDPS
CO-t-Bu
CO-t-Bu
CO2Mnt
COPh
H
H
H
H
CH3
H
H
H
H
acetone
acetone
acetone
acetone
Pyrex
Pyrex
Pyrex
Pyrex
Pyrex
quartz
quartz
quartz
quartz
quartz
5.5 h
4.7 h
3.3 h
6.6 h
11 h
2.5 h
2.6 h
40 min
3.5 h
6 h
44 (53)
42 (51)
26c
70:30
66:34
68:32d
5c, 6c
acetone
32c (35)c
68 (74)
52 (57)
24c (25)c
54:46c
66:34
59:41
66:34d
7
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
5c, 6c
9
10
TBDPS
CO-t-Bu
CH3
44c (50)c
53:47c
7
a
Isolated yield of the mixture of stereoisomers after column chromatography purification. Numbers in brackets are corrected yields
based on the percent of consumed lactone. Ratio of isolated products. c Yield or isomer ratio from 1H NMR and GC analyses of the
reaction crude. Ratio determined at low conversion of lactone, before the formation of 5c and 6c.
b
d
SCHEME 1
CHART 2
Resu lts a n d Discu ssion
P h otocycloa d d ition of 2(5H)-F u r a n on es 1a -e to
Acetylen e. Compounds 1a ,5a 1c,11 1d ,12 and 1e5b were
synthesized following procedures described in the litera-
ture. The new furanone 1b was easily prepared in 93%
yield by the reaction of (S)-5-hydroxymethyl-2(5H)-fura-
none13 with (-)-menthyl chloroformate and pyridine in
CH2Cl2.
structural and stereochemical assignment was further
confirmed by a single-crystal X-ray diffraction analysis
of 3b.
The photoreaction of 1c with acetylene in acetone
(entry 3) furnished a significantly lower yield of the
expected cycloadducts 3c and 4c, along with variable
amounts of a mixture of two additional products that
lacked double-bond signals in their NMR spectra and
were identified as the cyclobutanes 5c and 6c (Chart 2)
and a large amount of polymeric material. For this
reaction, the diastereomeric excesses of the different runs
were not reproducible, ranging from 44 to 54%. These
misleading results might be attributed, in some part, to
the extension of the photoreduction of the primary
cycloadducts 3c and 4c (vide infra). Monitoring the
reaction by GC showed that, at a low conversion of 1c,
the cyclobutanes 5c and 6c were not present and the ratio
of 3c:4c changed from 68:32 to 77:23 when the formation
of the cyclobutanes became apparent, thus masking the
true stereoselectivity of the cycloaddition. As a conse-
quence, the diastereoselectivity of this reaction has been
determined at low conversions of the starting furanone.
The photocycloaddition of 1c to acetylene in acetonitrile
(entry 8) evolved more rapidly to a similar composition
of the reaction mixture. Finally, all attempts to carry out
the cycloaddition of 1d to acetylene failed (entries 4 and
9). In the foregoing conditions, the starting furanone
underwent decomposition to unidentified products. Other
reaction conditions were also examined (T, filter, con-
centration) without any success.
Substrates 1a -e in acetone or acetonitrile solutions
saturated with acetylene were irradiated through a Pyrex
or quartz vessel with a 125-W medium-pressure mercury
lamp at -20 °C (Scheme 1). The progress of the cyclo-
addition was carefully monitored by GC or 1H NMR
analysis, and the reaction was stopped at an appropriate
time to avoid as much as possible the formation of
byproducts. The results are summarized in Table 1.
Irradiation of an acetone solution of 1a and acetylene
through a Pyrex filter for 5.5 h (entry 1) resulted in the
formation of the two expected cyclobutene diastereomers
3a and 4a in a moderate yield (53%) and with a fair
degree of facial selectivity (40% de). The structure of the
cycloadducts was determined by the detailed analysis of
1
their H and 13C NMR spectra. The value of the coupling
constant between H-4 and H-5 is diagnostic for the anti/
syn stereochemistry of the cycloadducts. A small value
of J 4,5 (around 1.5 Hz) is in agreement with a trans
relationship between these two protons (anti approach),
while larger values (around 6.0 Hz) denote a cis relation-
ship (syn approach).5
When the photochemical reaction of 1a was performed
in acetonitrile through a quartz filter (entry 6), after 2.5
h of irradiation, cycloadducts 3a and 4a were isolated in
a better yield (74%) but with lower de (32%). Similar
results came out for furanone 1b, affording the cyclo-
adducts 3b and 4b (entries 2 and 7). In this case, the
In an attempt to gain some insight into the formation
of the cyclobutane derivatives, the cyclobutene mixture
of adducts 3c and 4c in acetone was submitted to
irradiation under the above conditions except for the
absence of acetylene. The corresponding cyclobutanes 5c
and 6c were cleanly formed. It was then concluded that
these derivatives arise from the photoreduction of the
primary adducts. Similar photoreductions have been
reported in the irradiation of constrained cycloalkenes
(11) Beard, A. R.; Butler, P. I.; Mann, J .; Partlett, N. K. Carbohydr.
Res. 1990, 205, 87.
(12) Ortun˜o, R. M.; Ballesteros, M.; Corbera, J .; Sa´nchez-Ferrando,
F.; Font, J . Tetrahedron 1988, 44, 1711.
(13) Prepared from 1,2:5,6-di-O-isopropylidene-D-mannitol: Mann,
J .; Parlett, N. K.; Thomas, A. J . Chem. Res., Synop. 1987, 369.
1284 J . Org. Chem., Vol. 68, No. 4, 2003

Photochemical [2 + 2] Cycloaddition of Acetylene
in acetone.14 Barlett et al. suggested a mechanism where
the excited triplet acetone transfers its energy to the
cycloalkene, which is converted into the cycloalkane by
two successive hydrogen captures from the solvent. When
mixtures of 3a ,b and 4a ,b in acetone were irradiated,
photoreduction was not detected unless the solutions had
been previously degassed, in which case the cyclobutanes
5a ,b and 6a ,b were slowly formed. The overall experi-
ments suggested to us that an intramolecular photosen-
sitization caused by the benzoyl moiety in 3/4c may
account for the faster formation of the cyclobutane
products 5c and 6c in the course of the irradiation. For
identification purposes, the new compounds 5b, 6b, 5c,
and 6c were independently prepared by the photocyclo-
addition reactions of 1b and 1c with ethylene in acetone.
CHART 3
the 4 position of the 2(5H)-furanone decreases the rate
of the photocycloaddition and also leads to the formation
of the rearrangement product 7, in competition with
cycloadducts 3e and 4e, irrespective of the solvent and
filter used.
The results so far obtained showed that the highest
degree of facial diastereoselectivity was accomplished
with furanone 1a , which bears a pivaloyloxymethyl
group, as was the case in the cycloaddition to ethylene
(56% de).5a At this point, it seemed interesting to evaluate
how the introduction of a methyl group at the 4 position
of the 2(5H)-furanone may influence the steric course of
the cycloaddition process; a remarkable effect was previ-
ously observed on the photocycloadditions to ethylene and
tetramethylethylene.5b,15 To that end, the irradiation of
lactone 1e in acetone in the presence of acetylene was
performed (entry 5), resulting in a mixture of three
products: the cycloadducts 3e and 4e along with the
unexpected 3-vinyl-2(5H)-furanone 7 (Chart 2) in a ratio
of 44:38:18, respectively. In this reaction, a prolonged
irradiation time (11 h) was required to get an acceptable
conversion. The formation of 7 and cyclobutenes 3e and
4e was simultaneous, as was evidenced by GC and NMR
analyses of the reaction mixture at different reaction
times. The structures of cycloadducts 3e and 4e were
All these facts could be explained considering the
mechanism postulated for the photocycloaddition of
enones to alkenes.17,18 It is generally accepted that the
photocycloaddition occurs from the enone triplet excited
state, which reacts with the alkene partner to form a
triplet 1,4-biradical intermediate that collapses to the
coupling adduct after spin inversion. This process com-
petes with fragmentation to regenerate the ground-state
alkene and the enone.19 It has been suggested that the
cycloadduct product distribution is controlled by the
relative amount of the different isomeric 1,4-biradicals
formed, as well as by the extent to which each of them
partitions between closure to the product and reversion
to the ground-state precursors.20,21 The conformation of
the 1,4-biradical intermediates upon spin inversion should
also be an important factor in determining whether the
biradicals close to the products, disproportion to rear-
ranged compounds, or revert to the starting materials.
Thus, the biradical conformers that are not spatially
oriented for bond closure would revert to the ground-state
precursors, and an interradical distance (IRD) minor of
3 Å has been considered appropriate for ring closing.22
According to this general hypothesis, the isomeric 1,4-
biradical intermediates Ia -IVa and Ie-IVe (Chart 3)
have to be considered in the photoreaction of 1a and 1e
with acetylene, respectively. We have calculated the
relative stability and geometry of the four possible
intermediate biradicals I-IV in both cases.23 The equi-
librium geometry of the lowest-energy conformer and
some of the other significant and energetically close
conformers was optimized by applying the semiempirical
AM1 model. The results are summarized in Table 2.
According to their lower calculated energies, the triplet
1,4-biradicals that are likely to play an important role
as cycloaddition intermediates are I and II. For lactone
1a , the minimum-energy conformation of both biradical
intermediates Ia and IIa gave suitable IRDs for fast ring
1
determined by the analysis of the their H and 13C NMR
spectra and were confirmed by a single-crystal X-ray
diffraction of 3e. The noteworthy diminution of the
stereoselectivity observed (compare entries 1 and 5)
evidences the importance of the vinylic methyl group in
directing the final outcome of the photochemical process.
The structure of diene 7 was straightforwardly assigned
1
by H and 13C NMR analyses of enriched samples with
the help of a DEPT experiment.
White and co-workers reported that the photochemical
[2 + 2] cycloaddition of 4-methyl-5,6-dihydro-2-pyranone
to acetylene in acetonitrile gave the corresponding cy-
cloadduct in 73% yield.16 According to these authors, the
use of acetonitrile is crucial for the success of the reaction
because other solvents favored the formation of dispro-
portion products. Contrary to this result, the photochemi-
cal reaction of 1e with acetylene in acetonitrile (entry
10) furnished the cycloadducts 3e and 4e and also the
diene 7 in a 43:38:19 ratio, respectively (61% total yield).
Using Vycor as a filter, the photocycloaddition became
very slow, affording essentially the same proportion of
products. Therefore, the presence of a methyl group at
(17) Loufty, R. O.; de Mayo, P. J . Am. Chem. Soc. 1977, 99, 3559.
(18) Schuster, D. I.; Lem, G.; Kaprinidis, N. A. Chem. Rev. 1993,
93, 3.
(19) Schuster, D. I.; Heibel, G. E.; Brown, P. B.; Turro, N. J .; Kumar,
C. V. J . Am. Chem. Soc. 1988, 110, 8261.
(20) Bauslaugh, P. G. Synthesis 1970, 287.
(14) Barlett, P. D.; Roof, A. A. M.; Winter, W. J . J . Am. Chem. Soc.
1981, 103, 6520.
(15) Hoffmann, N.; Buschmann, H.; Raabe, G.; Scharf, H.-D. Tet-
rahedron 1994, 50, 11167.
(16) (a) White, J . D.; Matsui, T.; Thomas, J . A. J . Org. Chem. 1981,
46, 3376. (b) White, J . D.; Avery, M. A.; Carter, J . P. J . Am. Chem.
Soc. 1982, 104, 5486.
(21) Andrew, D.; Weedon, A. C. J . Am. Chem. Soc. 1995, 117, 5647.
(22) (a) Audley, M.; Geraghty, N. W. A. Tetrahedron Lett. 1996, 37,
1641. (b) Froese, R. D.; Lange, G. L.; Goddard, J . D. J . Org. Chem.
1996, 61, 952. (c) Busque´, F.; de March, P.; Figueredo, M.; Font, J .;
Margaretha, P.; Raya, J . Synthesis 2001, 1143.
(23) The calculations were performed using the PC SPARTAN plus
program of Wavefunction, Inc.
J . Org. Chem, Vol. 68, No. 4, 2003 1285

Alibe´s et al.
SCHEME 2
TABLE 2. Rela tive En er gies a n d Geom etr ica l Da ta for
th e Mor e Sta ble Con for m er s of Bir a d ica l In ter m ed ia tes
I-IVa
SCHEME 3
E
IRDc VR-MeH
E
IRDc VR-MeH
(kcal/mol)b (Å)
Dd (Å)
(kcal/mol)b (Å)
Dd (Å)
Ia
-93.58
-93.77
-92.47
-90.94
-103.44
2.9
2.9
2.9
2.8
3.6
I′e
IIe
IIIe
IVe
-103.41
-103.73
-92.76
-92.96
2.9
2.9
2.8
2.8
2.9
2.5
IIa
IIIa
IVa
Ie
3.6
a
The calculations were performed using the PC SPARTAN plus
b
program of Wavefunction, Inc. Energy calculated by the AM1
model. c Interradical distance. Distance from the vinyl radical to
d
the closest methyl proton (VR-MeH D).
cyclic double bond. Both intermediates I′e and IIe, with
calculated IRDs appropriate for biradical closure, present
suitable distances for a 1,5-hydrogen shift between the
vinyl radical and at least one of the methyl protons.
Therefore, the product distribution resulting for the above
reaction would be dominated by the extent that inter-
mediates I′e and IIe partition between reversion to the
starting products, disproportion, or closure to cyclobutenes
(Scheme 2).
P h otocycloa d d ition of Bis[2(5H)-fu r a n on es] 2a ,b
to Acetylen e. Although the chemical yields of the former
reactions were quite satisfactory, the moderate diaster-
eomeric excesses accomplished prompted us to consider
the use of C2-symmetric analogues of 1, that is, 2a ,b, as
substrates in the photoreaction with acetylene. In recent
papers, we have described the preparation of several bis-
[2(5H)-furanones] and their photocycloadditions to eth-
ylene. We studied the influence of the protecting groups
of the central diol unit on the facial diastereoselectivity
and found an overall anti selectivity higher than 98% for
the bis(trimethylsilyl) derivative.25 Encouraged by these
previous results, we have carried out the cycloaddition
of the bis(lactones) 2a and 2b, which are synthetically
equivalent to 1, to acetylene (Scheme 3).
closing and the preferential formation of the anti adduct
is likely to be determined by the higher steric demand of
the syn face of the lactone. For lactone 1e, it was expected
that the methyl group would sterically hinder the ap-
proach of acetylene to the â position and, moreover, it
would stabilize the radical center formed as a result of
the attack of the alkyne at the R position of the lactone.
The calculated energies are in agreement with this
expectation. The least-hindered anti approach of the
acetylene would lead to the predominant formation of Ie
over IIe, but the calculated IRD for the minimum-energy
conformer of Ie gave a value too high to favor collapse
over reversion. On the contrary, for intermediate IIe
derived from the syn approach, the relatively short
distance between the two radical sites allows fast closure
to products. Nevertheless, a very close minimum-energy
conformation, I′e, also derived from anti bonding and
with an appropriate IRD for ring closing, has been found.
Therefore, this could explain the poor facial diastereo-
selectivity observed for 1e.
The occurrence of such intermediates could also ac-
count for the formation of diene 7. A plausible mechanism
for the formation of this compound would be an intra-
molecular disproportion via a 1,5-hydrogen transfer5b,24
from the methyl group to the vinyl radical leading to an
ene-type product, followed by isomerization of the exo-
Thus, irradiation of 2a in a solution of acetonitrile
saturated with acetylene through a quartz filter resulted
(25) (a) de March, P.; Figueredo, M.; Font, J .; Raya, J . Tetrahedron
Lett. 1999, 40, 2205. (b) de March, P.; Figueredo, M.; Font, J .; Raya,
J . Org. Lett. 2000, 2, 163.
(26) See for example: (a) Blomfield, J . J .; Owsley, D. C. Org.
Photochem. Synth. 1976, 2, 36. (b) Binns, F.; Hayes, R.; Ingham, S.;
Saengchantara, S. T.; Turner, R. W.; Wallace, T. W. Tetrahedron 1992,
48, 515.
(24) (a) Lange, G. L.; Organ, M. G.; Lee, M. Tetrahedron Lett. 1990,
31, 4689. (b) Lewis, F. D.; Reddy, G. D.; Elbert, J . E.; Tillberg, B. E.;
Meltzer, J . A.; Kojima, M. J . Org. Chem. 1991, 56, 5311. (c) Curran,
D. P.; Shen, W. J . Am. Chem. Soc. 1993, 115, 6051. (d) Capella, L.;
Montevecchi, P. C.; Navacchia, M. L. J . Org. Chem. 1996, 61, 6783.
1286 J . Org. Chem., Vol. 68, No. 4, 2003

Photochemical [2 + 2] Cycloaddition of Acetylene
in the formation of a crude mixture whose 1H and 13C
NMR spectra showed a main set of signals consistent
with a highly symmetric bis(cyclobutene) adduct. The
relative configuration of this major product 8a was
assigned as before by means of the value of the vicinal
coupling constant J 4,5, which is 1.5 Hz, indicative of an
anti-anti stereochemistry. Neither of the other cyclo-
adducts, 9a or 10a , was detected in the mixture (de >
98%). Purification by column chromatography allowed
the isolation of 8a in 56% yield. However, the product
showed some instability during this process, presumably
due to the sensitive silyl protecting groups, and conse-
quently the crude mixture of the photochemical reaction
was directly desilylated by treatment with TBAF, afford-
ing the corresponding diol 11a in 43% overall yield for
the two steps. Consequently, the use of the C2-symmetric
bis[2(5H)-furanone] 2a in the photocycloaddition to acet-
ylene has remarkably improved the facial selectivity.
Next, we explored the photochemical behavior of the
bis[2(5H)-furanone] 2b that bears a methyl group at the
4 position. Irradiation of this bis(lactone) in acetonitrile
gave a very complex mixture for which NMR spectra
displayed only traces of the signals attributable to the
expected bis(cyclobutene). Attempts to obtain any defined
compound by column chromatography failed. A similar
result was found when the reaction was performed in
acetone. Several trials in other conditions were also
unsuccessful. This negative result contrasts with that
previously observed in the photocycloaddition of ethylene
to the same substrate25b and can be rationalized on the
basis of the influence that the vinylic methyl group exerts
in the course of the photochemical reaction.
electron-impact mass spectra were recorded at 70 eV. The
HRMS spectrum was performed at the SIDI in the Universidad
Auto´noma de Madrid. Microanalyses were performed at the
Servei d’Ana`lisi Qu´ımica de la Universitat Auto`noma de
Barcelona.
(1R,2S,5R)-Men th yl [(5S)-2-Oxo-2,5-d ih yd r ofu r a n -5-yl]-
m eth yl Ca r bon a te (1b). (-)-Menthyl chloroformate (2.8 mL,
13.1 mmol) was added dropwise to an ice-cooled solution of
(S)-5-hydroxymethyl-2(5H)-furanone (1.00 g, 8.7 mmol) and
dry pyridine (1.5 mL) in dry CH2Cl2 (25 mL). The mixture was
stirred overnight as it came to room temperature. After the
addition of water (15 mL), the organic layer was successively
washed with 5% HCl (3 × 20 mL), saturated aqueous NaHCO3
(3 × 20 mL), and brine (2 × 20 mL) and dried (Na2SO4).
Evaporation of the solvent and chromatography of the residue
(3:1 hexanes-EtOAc) afforded 1b (2.38 g, 93% yield): mp 94-
95 °C (white solid from EtOAc-pentane); [R]D -154.3° (c 1.05,
CHCl3); IR (KBr) 2959, 2917, 2868, 1743, 1264; 1H NMR (400
MHz) δ 7.44 (dd, J ) 5.9, 1.8 Hz, 1H), 6.17 (dd, J ) 5.9, 2.2
Hz, 1H), 5.21 (m, 1H), 4.47 (td, J ) 10.9, 4.5 Hz, 1H), 4.35 (d,
J ) 4.5 Hz, 2H), 2.00 (m, 1H), 1.86 (m, 1H), 1.64 (m, 2H), 1.43
(m, 1H), 1.36 (m, 1H), 1.02 (m, 2H), 0.88 (d, J ) 6.5 Hz, 3H),
0.85 (d, J ) 7.0 Hz, 3H), 0.84 (m, 1H), 0.73 (d, J ) 7.0 Hz,
3H); 13C NMR δ 172.0, 154.4, 152.1, 123.4, 80.5, 79.2, 65.4,
46.8, 40.6, 34.0, 31.4, 26.0, 23.3, 21.9, 20.6, 16.3; MS (CI, NH3)
m/z 296 (M+ + 18, 100). Anal. Calcd for C16H24O5: C, 64.85;
H, 8.16. Found: C, 65.03; H, 8.33.
Gen er a l P r oced u r e for th e P h otocycloa d d ition s of
2(5H)-F u r a n on es to Acetylen e a n d Eth ylen e. Irradiations
were performed in a small conventional photochemical reactor
(two-necked vessel fitted with a Vycor, Pyrex, or quartz
immersion-type cooling jacket) using a medium-pressure,
125-W mercury lamp. Methanol at -15 °C was used for the
refrigeration of the immersion well jacket. The vessel was
externally cooled at -20 °C with a dry ice-CCl4 bath. The
progress of the reaction was monitored by GC or 1H NMR
analysis of aliquot samples. Acetylene (acetone free) or eth-
ylene gas was bubbled through the solution for 15 min. Once
the lamp was turned on, a slow flow of gas was maintained
throughout the irradiation. Caution: acetylene is hazardous
and subject to strict safety regulations.26
Con clu sion s
In summary, we have investigated thoroughly the
photochemical [2 + 2] cycloaddition of chiral 2(5H)-
furanones to acetylene. Among the studied derivatives,
furanone 1a , which bears the pivaloyl group, gave the
best overall yield and diastereofacial selectivity. The
presence of a methyl group at the â position of the lactone
slowed the rate of the photocycloaddition and led to the
formation of a disproportion adduct in competition with
cyclobutene formation. A mechanistic proposal based on
a simple theoretical conformational analysis accounts for
the experimental observations. When a C2-symmetric bis-
(lactone) as the photoactive substrate is used, the bis-
(cyclobutene) adduct is formed with a fair yield and a
diastereomeric excess higher than 98%. The cyclobutene
compounds synthesized are small, densely functionalized
molecules that could serve as useful precursors in asym-
metric synthesis. Active investigation in this field is being
carried out in our laboratory.
(1R,4S,5S)-4-P ivaloyloxym eth yl-3-oxabicyclo[3.2.0]h ept-
6-en -2-on e (3a ) a n d (1S,4S,5R)-4-P iva loyloxym eth yl-3-
oxa bicyclo[3.2.0]h ep t-6-en -2-on e (4a ). A solution of 1a (94
mg, 0.47 mmol) in freshly distilled acetone (65 mL) saturated
with acetylene was irradiated through a Pyrex filter for 5.5 h.
Evaporation of the solvent and chromatography of the residue
(5:1 hexanes-EtOAc) afforded a 70:30 mixture of cycloadducts
3a and 4a (40 mg, 0.18 mmol, 44% yield) and some unreacted
1a (17 mg, 0.08 mmol, 18%). Repeated chromatographies (0
f 15% EtOAc in hexane) allowed the separation of pure 3a
and 4a . 3a : oil; [R]D -211.9 (c 1.1, CHCl3); IR (film) 2975, 2874,
1
1768, 1734, 1484, 1171; H NMR (400 MHz) δ 6.35 (dt, J )
2.7, 0.5 Hz, 1H), 6.30 (dd, J ) 2.7, 0.8 Hz, 1H), 4.60 (dddd, J
) 3.0, 3.0, 1.5, 0.5 Hz, 1H), 4.26 (dd, J ) 12.0, 3.0 Hz, 1H),
4.12 (dd, J ) 12.0, 3.0 Hz, 1H), 3.70 (dd, J ) 3.5, 0.8 Hz, 1H),
3.45 (ddd, J ) 3.5, 1.5, 0.5 Hz, 1H), 1.20 (s, 9H); 13C NMR δ
178.0, 174.6, 140.7, 139.2, 76.2, 65.7, 47.6, 44.2, 38.6, 27.1; MS
(CI, NH3) m/z 242 (M+ + 18, 100), 225 (M+ + 1, 35). Anal.
Calcd for C12H16O4: C, 64.27; H, 7.19. Found: C, 64.14; H,
7.13. 4a : oil; [R]D +128 (c 0.25, CHCl3); IR (film) 2960, 2926,
2873, 1770, 1732, 1481, 1285; 1H NMR (400 MHz) δ 6.34 (dd,
J ) 2.7, 0.9 Hz, 1H), 6.28 (dd, J ) 2.7, 0.9 Hz, 1H), 4.61 (td,
J ) 7.2, 4.8 Hz, 1H), 4.30 (dd, J ) 11.8, 4.8 Hz, 1H), 4.22 (dd,
J ) 11.8, 7.2 Hz, 1H), 3.70 (dd, J ) 3.6, 0.9 Hz, 1H), 3.65 (ddd,
J ) 7.2, 3.6, 0.9 Hz, 1H), 1.20 (s, 9H); 13C NMR δ 178.1, 174.6,
139.7, 138.7, 75.3, 63.2, 47.6, 43.4, 38.8, 27.1; MS (CI, NH3)
m/z 242 (M+ + 18, 100), 225 (M+ + 1, 15). Anal. Calcd for
Exp er im en ta l Section
Gen er a l Meth od s. The solutions were concentrated using
an evaporator at 15-20 Torr. Flash column chromatographies
were carried out on silica gel (230-400 mesh). Melting points
were determined at the hot stage and are uncorrected. The
signals in the IR spectra are reported in cm-1. The 1H NMR
and 13C NMR spectra were recorded at the Servei de Resso-
na`ncia Magne`tica Nuclear de la Universitat Auto`noma de
Barcelona at 250 and 62.5 MHz or 400 and 100 MHz (when
specified) in CDCl3 solutions, unless otherwise indicated. The
C
12H16O4: C, 64.27; H, 7.19. Found: C, 64.32; H, 7.09.
When the irradiation was performed through a quartz filter
in freshly distilled acetonitrile (70 mL) for 2.5 h, from lactone
1a (100 mg, 0.50 mmol) after the purification of the crude
J . Org. Chem, Vol. 68, No. 4, 2003 1287

Alibe´s et al.
material by column chromatography, the following fractions
were obtained: a mixture (66:34) of 3a and 4a (77 mg, 0.34
mmol, 68% yield) and lactone 1a (8 mg, 0.04 mmol, 8%).
128.4, 75.3, 63.7, 47.6, 43.3; MS (CI, NH3) m/z 262 (M+ + 18,
100), 245 (M+ + 1, 23).
When the irradiation was performed through a quartz filter
in freshly distilled acetonitrile (70 mL) for 40 min, from lactone
1c (100 mg, 0.46 mmol) after the purification of the crude
material by column chromatography, the following fractions
were obtained: a mixture (56:19:17:8) of 3c, 4c, 5c, and 6c
(36 mg) and some unreacted lactone 1c (4 mg, 0.02 mmol, 4%).
(1R,2S,5R)-Men th yl (1R,4S,5S)-2-Oxo-3-oxabicyclo[3.2.0]-
h ep t-6-en -4-ylm eth yl Ca r bon a te (3b) a n d (1R,2S,5R)-
Men th yl (1S,4S,5R)-2-Oxo-3-oxa bicyclo[3.2.0]h ep t-6-en -
4-ylm eth yl Ca r bon a te (4b). A solution of 1b (81 mg, 0.27
mmol) in freshly distilled acetone (55 mL) saturated with
acetylene was irradiated through a Pyrex filter for 4.7 h.
Evaporation of the solvent and chromatography of the residue
(4:1 hexanes-EtOAc) afforded a 66:34 mixture of cycloadducts
3b and 4b (37 mg, 0.11 mmol, 42% yield) and recovered lactone
1b (14 mg, 0.05 mmol, 17%). Repeated chromatographies (5:1
hexanes-EtOAc) of the first fraction furnished analytical
samples of pure 3b and 4b. 3b: mp 110-112 °C (colorless
needles from EtOAc-pentane); [R]D -171.8 (c 0.71, CHCl3);
(1R,4S,5S)-5-Meth yl-4-pivaloyloxym eth yl-3-oxabicyclo-
[3.2.0]h ep t -6-en -2-on e (3e), (1S,4S,5R)-5-Met h yl-4-p iv-
a loyloxym eth yl-3-oxa bicyclo[3.2.0]h ep t-6-en -2-on e (4e),
a n d (5S)-4-Met h yl-5-p iva loyloxym et h yl-3-vin yl-2(5H )-
fu r a n on e (7). A solution of 1e (80 mg, 0.38 mmol) in freshly
distilled acetone (55 mL) saturated with acetylene was irradi-
ated through a Pyrex filter for 11 h. Evaporation of the solvent
and chromatography of the residue (4:1 hexanes-EtOAc)
afforded a 44:38:18 mixture of 3e, 4e, and 7 (35 mg, 39% yield)
and some unreacted lactone 1e (6 mg, 0.03 mmol, 7%). A
second column chromatography furnished pure 4e and a
mixture of 3e and 7 as an oil. The crystallization of this oil
from EtOAc-pentane gave 3e as colorless needles. The mother
liquor contained 3e and 7, all attempts to separate these
compounds were unsuccessful, and the enriched fractions of
diene 7 were analyzed. 3e: mp 53-54 °C (colorless needles
from EtOAc-pentane); [R]D -134.3 (c 1.2, CHCl3); IR (KBr)
2978, 1879, 1770, 1727, 1285, 1179; 1H NMR (400 MHz) δ 6.38
(dd, J ) 2.6, 1.2 Hz, 1H), 6.30 (dd, J ) 2.6, 0.8 Hz, 1H), 4.54
(dd, J ) 3.0, 2.3 Hz, 1H), 4.38 (dd, J ) 12.2, 3.0 Hz, 1H), 4.09
(dd, J ) 12.2, 2.3 Hz, 1H), 3.25 (s, 1H), 1.42 (s, 3H), 1.18 (s,
9H); 13C NMR δ 177.7, 175.3, 146.1, 137.2, 78.3, 63.5, 52.8,
50.4, 38.7, 27.0, 16.4; MS (CI, NH3) m/z 256 (M+ + 18, 46),
239 (15), 228 (13), 227 (100). Anal. Calcd for C13H18O4: C,
65.53; H, 7.61. Found: C, 65.53; H, 7.70. 4e: oil; [R]D +178.9
(c 0.20, CHCl3); IR (film) 2973, 2875, 1778, 1729, 1152; 1H
NMR δ 6.33 (dd, J ) 2.8, 0.9 Hz, 1H), 6.30 (dd, J ) 2.8, 0.7
Hz, 1H), 4.33 (dd, J ) 7.4, 4.7 Hz, 1H), 4.25 (dd, J ) 11.7, 4.7
Hz, 1H), 4.19 (dd, J ) 11.7, 7.4 Hz, 1H), 3.24 (s, 1H), 1.49 (s,
3H), 1.21 (s, 9H); 13C NMR δ 178.1, 174.1, 142.9, 137.3, 80.4,
62.8, 53.0, 50.7, 38.6, 27.0, 20.1; MS m/z 239 (M+ + 1, 15),
238 (M+, 100), 137 (32), 119 (22), 93 (23). Anal. Calcd for
1
IR (KBr) 2956, 2929, 2868, 1765, 1742, 1287; H NMR δ 6.32
(dd, J ) 2.7, 0.6 Hz, 1H), 6.28 (dd, J ) 2.7, 1.1 Hz, 1H), 4.56
(td, J ) 3.9, 1.6 Hz, 1H), 4.48 (td, J ) 10.9, 4.5 Hz, 1H), 4.27
(dd, J ) 11.6, 3.9 Hz, 1H), 4.16 (dd, J ) 11.6, 3.9 Hz, 1H),
3.67 (dd, J ) 3.4, 1.1 Hz, 1H), 3.43 (ddd, J ) 3.4, 1.6, 0.6 Hz,
1H), 2.01 (m, 1H), 1.89 (m, 1H), 1.64 (m, 2H), 1.44 (m, 1H),
1.38 (m, 1H), 1.01 (m, 2H), 0.88 (d, J ) 6.6 Hz, 3H), 0.87 (d, J
) 7.2 Hz, 3H), 0.84 (m, 1H), 0.75 (d, J ) 7.0 Hz, 3H); 13C NMR
δ 174.3, 154.4, 140.7, 139.2, 79.2, 76.1, 67.7, 47.2, 46.8, 44.0,
40.6, 34.0, 31.3, 26.1, 23.3, 21.9, 20.6, 16.3; MS (CI, NH3) m/z
340 (M+ + 18, 100). Anal. Calcd for C18H26O5: C, 67.06; H,
8.13. Found: C, 67.07; H, 8.08. 4b: oil, [R]D +80.0 (c 0.65,
CHCl3); IR (film) 2956, 2871, 1766, 1743, 1456, 1260; 1H NMR
δ 6.34 (dd, J ) 2.7, 1.0 Hz, 1H), 6.28 (dd, J ) 2.7, 0.8 Hz, 1H),
4.63 (ddd, J ) 6.9, 6.6, 5.4 Hz, 1H), 4.50 (td, J ) 10.9, 4.5 Hz,
1H), 4.33 (dd, J ) 11.6, 6.9 Hz, 1H), 4.26 (dd, J ) 11.6, 5.4
Hz, 1H), 3.69 (m, 1H), 3.65 (ddd, J ) 6.6, 3.6, 1.0 Hz, 1H),
2.03 (m, 1H), 1.92 (m, 1H), 1.65 (m, 2H), 1.44 (m, 1H), 1.39
(m, 1H), 1.02 (m, 2H), 0.89 (d, J ) 6.5 Hz, 3H), 0.87 (d, J )
7.0 Hz, 3H), 0.84 (m, 1H), 0.76 (d, J ) 7.0 Hz, 3H); 13C NMR
δ 173.6, 154.4, 139.8, 138.5, 79.1, 74.8, 66.1, 47.5, 46.9, 43.2,
40.6, 34.0, 31.3, 26.0, 23.2, 21.9, 20.6, 16.2; MS (CI, NH3) m/z
340 (M+ + 18, 100). Anal. Calcd for C18H26O5: C, 67.06; H,
8.13. Found: C, 66.60; H, 8.30.
C
13H18O4: C, 65.53; H, 7.61. Found: C, 65.41; H, 7.70. 7: 1H
When the irradiation was performed through a quartz filter
in freshly distilled acetonitrile (70 mL) for 2.6 h, from lactone
1b (98 mg, 0.34 mmol) after the purification of the crude
material by column chromatography, the following fractions
were obtained: a mixture (59:41) of 3b and 4b (56 mg, 0.17
mmol, 52% yield) and lactone 1b (8 mg, 0.03 mmol, 8%).
NMR δ 6.37 (dd, J ) 17.3, 10.4 Hz, 1H), 6.26 (dd, J ) 17.3,
2.8 Hz, 1H), 5.48 (dd, J ) 10.4, 2.8 Hz, 1H), 4.90 (dd, J ) 4.0,
3.0 Hz, 1H), 4.40 (dd, J ) 12.2, 3.0 Hz, 1H), 4.33 (dd, J ) 12.2,
4.0 Hz, 1H), 2.06 (s, 3H), 1.16 (s, 9H); 13C NMR δ 177.9, 171.2,
155.9, 125.1, 123.7, 121.2, 80.2, 61.4, 38.8, 27.0, 12.0.
When the irradiation was performed through a quartz filter
in freshly distilled acetonitrile (70 mL) for 6 h, from lactone
1e (100 mg, 0.47 mmol) after the purification of the crude
material by column chromatography, the following fractions
were obtained: a mixture (43:38:19) of 3e, 4e, and 7 (61 mg,
54% yield) and lactone 1e (12 mg, 0.04 mmol, 12%).
(1R,2S,5R)-Men t h yl [(1R,4S,5S)-2-Oxo-3-oxa b icyclo-
[3.2.0]h ep t-4-yl]m eth yl Ca r bon a te (5b) a n d (1R,2S,5R)-
Men th yl [(1S,4S,5R)-2-Oxo-3-oxa bicyclo[3.2.0]h ep t-4-yl]-
m eth yl Ca r bon a te (6b). A solution of 1b (103 mg, 0.35 mmol)
in freshly distilled acetone (70 mL) saturated with ethylene
was irradiated through a Pyrex filter for 6 h. Evaporation of
the solvent and chromatography of the residue (5:1 hexanes-
EtOAc) afforded a 74:26 mixture of cycloadducts 5b and 6b
(75 mg, 0.23 mmol, 66% yield) as a white solid and some
unreacted lactone 1b (6 mg, 0.02 mmol, 6%). Repeated chro-
matographies with the same eluent allowed the isolation of
the analytical samples of 5b and 6b. 5b: mp 104-105 °C
(colorless needles from EtOAc-pentane); [R]D -81.2 (c 1.01,
CHCl3); IR (KBr) 2962, 2921, 2868, 1775, 1734, 1387, 1269;
1H NMR (400 MHz) δ 4.53 (td, J ) 3.8, 0.8 Hz, 1H), 4.47 (td,
J ) 10.9, 4.5 Hz, 1H), 4.21 (dd, J ) 11.7, 3.8 Hz, 1H), 4.12
(dd, J ) 11.7, 3.8 Hz, 1H), 3.13 (dddd, J ) 10.0, 6.4, 2.0, 2.0
Hz, 1H), 3.02 (m, 1H), 2.53 (m, 1H), 2.39 (m, 1H), 2.14 (m,
2H), 2.02 (m, 1H), 1.89 (m, 1H), 1.65 (m, 2H), 1.43 (m, 1H),
1.37 (m, 1H), 1.02 (m, 2H), 0.89 (d, J ) 6.5 Hz, 3H), 0.88 (d, J
(1R,4S,5S)-4-Ben zoyloxym eth yl-3-oxabicyclo[3.2.0]h ept-
6-en -2-on e (3c) a n d (1S,4S,5R)-4-Ben zoyloxym eth yl-3-
oxa bicyclo[3.2.0]h ep t-6-en -2-on e (4c). A solution of 1c (100
mg, 0.46 mmol) in freshly distilled acetone (55 mL) saturated
with acetylene was irradiated through a Pyrex filter for 3.3 h.
Evaporation of the solvent and chromatography of the residue
(6:1 hexanes-EtOAc) afforded a mixture (52:16:23:9) of 3c,
4c, 5c, and 6c (40 mg) as a white solid. A second chromatog-
raphy (from 12:1 to 4:1 hexanes-EtOAc) provided the follow-
ing fractions: (i) a mixture of 3c and 5c as a white solid and
(ii) a mixture of 4c and 6c as a white solid. All attempts to
separate 3c and 4c from 5c and 6c, respectively, were
unsuccessful, and the enriched fractions were analyzed. 3c:
1H NMR (400 MHz) δ 7.98 (m, 2H), 7.55 (m, 1H), 7.42 (m, 2H),
6.37 (dd, J ) 2.6, 0.9 Hz, 1H), 6.32 (dd, J ) 2.6, 0.9 Hz, 1H),
4.72 (td, J ) 3.2, 1.5 Hz, 1H), 4.50 (dd, J ) 12.0, 3.2 Hz, 1H),
4.44 (dd, J ) 12.0, 3.2 Hz, 1H), 3.72 (dd, J ) 3.5, 0.9 Hz, 1H),
3.50 (ddd, J ) 3.5, 1.5, 0.9 Hz, 1H); 13C NMR δ 174.6, 166.1,
140.7, 139.2, 133.5, 129.6, 129.2, 128.6, 76.3, 66.0, 47.7, 44.2;
MS (CI, NH3) m/z 262 (M+ + 18, 100), 245 (M+ + 1, 23). 4c:
1H NMR (400 MHz) δ 8.03 (m, 2H), 7.56 (m, 1H), 7.44 (m, 2H),
6.37 (d, J ) 2.6 Hz, 1H), 6.34 (d, J ) 2.6 Hz, 1H), 4.75 (ddd,
J ) 7.6, 5.8, 4.4 Hz, 1H), 4.57 (dd, J ) 12.0, 4.4 Hz, 1H), 4.47
(dd, J ) 12.0, 7.6 Hz, 1H), 3.74 (m, 1H), 3.73 (d, J ) 3.5 Hz,
1H); 13C NMR δ 173.8, 166.1, 139.7, 138.7, 133.4, 129.7, 129.3,
1288 J . Org. Chem., Vol. 68, No. 4, 2003

Photochemical [2 + 2] Cycloaddition of Acetylene
) 7.0 Hz, 3H), 0.88 (m, 1H), 0.76 (d, J ) 7.0 Hz, 3H); 13C NMR
δ 179.7, 154.5, 82.2, 79.1, 67.9, 46.9, 40.5, 38.7, 36.4, 34.0, 31.4,
26.1, 24.8, 23.6, 23.3, 21.9, 20.6, 16.3; MS (CI, NH3) m/z 342
(M+ + 18, 100). Anal. Calcd for C18H28O5: C, 66.64; H, 8.70.
Found: C, 66.45; H, 8.77. 6b: mp 70-74 °C (colorless needles
from EtOAc-pentane); [R]D -115.2 (c 0.30, CHCl3); IR (KBr)
2948, 2963, 2926, 1769, 1744, 1274, 1251, 1178; 1H NMR (400
MHz) δ 4.63 (ddd, J ) 7.3, 5.9, 4.7 Hz, 1H), 4.52 (td, J ) 10.9,
4.4 Hz, 1H), 4.40 (dd, J ) 11.7, 7.3 Hz, 1H), 4.23 (dd, J ) 11.7,
4.7 Hz, 1H), 3.24 (m, 1H), 3.14 (m, 1H), 2.51 (m, 1H), 2.26 (m,
1H), 2.11 (m, 2H), 2.03 (m, 1H), 1.93 (m, 1H), 1.66 (m, 2H),
1.45 (m, 1H), 1.39 (m, 1H), 1.03 (m, 2H), 0.90 (d, J ) 6.5 Hz,
3H), 0.88 (d, J ) 7.0 Hz, 3H), 0.88 (m, 1H), 0.77 (d, J ) 7.0
Hz, 3H); 13C NMR δ 179.3, 154.5, 79.1, 78.2, 65.8, 47.0, 40.7,
39.7, 36.5, 34.0, 31.4, 26.0, 23.3, 23.2, 21.9, 20.7, 19.4, 16.2;
MS (CI, NH3) m/z 342 (M+ + 18, 100). Anal. Calcd for
(65 mL) of 3/4a , 3/4b, or 3/4c (2.5 mM) were previously
degassed and irradiated under an argon atmosphere until the
disappearance of the starting material (3/4a , 5 h; 3/4b, 5 h;
3/4c, 2 h). 1H NMR and GC analyses of the crude reactions
showed only signals corresponding to the cyclobutanes 5/6a ,
5/6b, or 5/6c.
(1R,1′R,4S,4′S,5S,5′S)-4,4′-[(1R,2R)-1,2-Bis(tr im eth ylsi-
lyloxy)eth a n e-1,2-d iyl]bis(3-oxa bicyclo[3.2.0]h ep t-6-en -
2-on e) (8a ). A solution of 2a (100 mg, 0.27 mmol) in freshly
distilled acetonitrile (70 mL) saturated with acetylene was
irradiated through quartz for 2 h. Evaporation of the solvent
and chromatography of the residue (6:1 hexanes-EtOAc)
through alumina afforded 8a (64 mg, 0.15 mmol, 56% yield)
as an oil: 1H NMR δ 6.43 (dt, J ) 2.7, 0.8 Hz, 2H), 6.21 (dd,
J ) 2.7, 0.7 Hz, 2H), 4.58 (ddd, J ) 3.2, 1.5, 0.7 Hz, 2H), 3.54
(dd, J ) 3.2, 1.0 Hz, 2H), 3.50 (m, 4H), 0.15 (s, 18H); 13C NMR
δ 174.9, 141.8, 137.9, 78.6, 76.4, 46.9, 44.2, 0.13.
C
18H28O5: C, 66.64; H, 8.70. Found: C, 66.88; H, 8.71.
(1R,4S,5S)-4-Ben zoyloxym et h yl-3-oxa b icyclo[3.2.0]-
(1R,1′R,4S,4′S,5S,5′S)-4,4′-[(1R,2R)-1,2-Dih idr oxyeth an e-
1,2-d iyl]bis(3-oxa bicyclo[3.2.0]h ep t-6-en -2-on e) (11a ). A
solution of 2a (94 mg, 0.25 mmol) in freshly distilled aceto-
nitrile (70 mL) saturated with acetylene was irradiated
through quartz for 2 h. After the evaporation of the solvent,
the residue was dissolved in dry THF (1 mL) and a solution of
1 M TBAF in THF (1.0 mL, 1.00 mmol) was added at room
temperature under an argon atmosphere. The mixture was
stirred for 2 h. Evaporation of the solvent and chromatography
of the residue (4:1 hexanes-EtOAc) afforded 11a (30 mg, 0.10
mmol, 43% yield from 2a ) as a white solid: mp 178-180 °C
(EtOAc-pentane); [R]D -125.5 (c 0.26, acetone); IR (KBr) 3398,
h ep ta n -2-on e (5c) a n d (1S,4S,5R)-4-Ben zoyloxym eth yl-
3-oxa bicyclo[3.2.0]h ep ta n -2-on e (6c). A solution of 1c (98
mg, 0.45 mmol) in freshly distilled acetone (70 mL) saturated
with ethylene was irradiated through a Pyrex filter for 6 h.
Evaporation of the solvent and chromatography of the residue
(3:1 hexanes-EtOAc) gave cycloadducts 5c and 6c (78 mg, 0.32
mmol, 71% yield) as a 76:24 mixture and some unreacted
lactone 1c (13 mg, 0.06 mmol, 13%). Repeated chromatogra-
phies (from 8:1 to 5:1 hexanes-EtOAc) allowed the isolation
of the analytical samples of 5c and 6c. 5c: mp 78-80 °C (white
solid from EtOAc-pentane); [R]D -55.0 (c 0.8, CHCl3); IR (KBr)
2988, 2944, 1761, 1723, 1286, 1159; 1H NMR (400 MHz) δ 7.95
(m, 2H), 7.55 (m, 1H), 7.42 (m, 2H), 4.67 (ddd, J ) 3.8, 3.2,
1.0 Hz, 1H), 4.44 (dd, J ) 12.2, 3.8 Hz, 1H), 4.38 (dd, J ) 12.2,
3.2 Hz, 1H), 3.17 (m, 1H), 3.08 (m, 1H), 2.56 (m, 1H), 2.42 (m,
1H), 2.18 (m, 2H); 13C NMR δ 179.9, 166.1, 133.4, 129.6, 129.1,
128.6, 82.7, 65.7, 38.9, 36.6, 24.8, 23.7; MS (CI, NH3) m/z 264
(M+ + 18, 100), 247 (M+ + 1, 51). Anal. Calcd for C14H14O4:
C, 68.28; H, 5.73. Found: C, 68.36; H, 5.75. 6c: mp 78-81 °C
(white solid from EtOAc-pentane); [R]D +91.6 (c 0.41, CHCl3);
IR (KBr) 2971, 2952, 1770, 1719, 1270, 1166; 1H NMR (400
MHz) δ 8.02 (m, 2H), 7.60 (m, 1H), 7.44 (m, 2H), 4.76 (ddd, J
) 7.0, 5.9, 4.7 Hz, 1H), 4.54 (dd, J ) 12.0, 7.0 Hz, 1H), 4.51
(dd, J ) 12.0, 4.7 Hz, 1H), 3.31 (m, 1H), 3.17 (m, 1H), 2.54 (m,
1H), 2.34 (m, 1H), 2.15 (m, 2H); 13C NMR δ 179.5, 166.2, 133.4,
129.7, 129.3, 128.5, 78.7, 63.5, 39.8, 36.6, 23.3, 19.5; MS (CI,
NH3) m/z 264 (M+ + 18, 100), 247 (M+ + 1, 12). Anal. Calcd
for C14H14O4: C, 68.28; H, 5.73. Found: C, 68.31; H, 5.77.
Gen er a l P r oced u r e for th e P h otor ed u ction of Cy-
clobu ten es 3a -c a n d 4a -c. Irradiations were performed
using the same procedure as above but in the absence of
acetylene. The progress of the reaction was monitored by GC
1
3359, 2951, 1753, 1172; H NMR (acetone-d6) δ 6.49 (dt, J )
2.7, 0.8 Hz, 2H), 6.29 (dd, J ) 2.7, 0.8 Hz, 2H), 4.62 (dd, J )
1.4, 0.7 Hz, 2H), 3.73 (s, 2H), 3.66 (ddd, J ) 3.4, 1.4, 0.8 Hz,
2H), 3.57 (dd, J ) 3.4, 0.8 Hz, 2H), 2.81 (m, 2H); 13C NMR
(acetone-d6) δ 174.3, 142.3, 138.7, 77.6, 74.1, 47.8, 45.7. HRMS
(FAB): (M+ + 1) calcd for C14H15O6, 279.0869; found, 279.0868.
Ack n ow led gm en t. We gratefully acknowledge the
financial support of DGI (BQU2001-2600), CIRIT (Project
2001SGR00178), and the Generalitat de Catalunya and
the Agencia Espan˜ola de Cooperacio´n Internacional for
grants (to M.R. and X.F., respectively).
Su p p or tin g In for m a tion Ava ila ble: Crystallographic
information files and ORTEP drawings for compounds 3b and
3e and the full assignment of 1H and 13C NMR spectra of
compounds 1b; 3a ,b,d ,e; 4a ,b,d ,e; 5b,c; 6b,c; 7; 8a ; and 11a .
This material is available free of charge via the Internet at
http://pubs.acs.org.
1
or H NMR analysis of aliquot samples. The acetone solutions
J O0264731
J . Org. Chem, Vol. 68, No. 4, 2003 1289