An unusual and efficient reaction of 2-methylene-3-phenyloxetane in the presence of lithium and 4,4'-di-tert-butylbiphenyl in THF

Article abstract of DOI:10.1016/S0040-4039(00)00060-5

When 2-methylene-3-phenyloxetane (4) was treated with excess lithium and DTBB (cat.), lactone 6 was isolated as an unexpected product in high yield. It is postulated that 6 arises from a coupling of a radical enolate derived from 4 and the enolate of acetaldehyde, a product of THF decomposition under the reaction conditions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00060-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1859–1862
An unusual and efficient reaction of
2-methylene-3-phenyloxetane in the presence of lithium and
4,40-di-tert-butylbiphenyl in THF
Mehrnoosh Hashemzadeh and Amy R. Howell ∗
Department of Chemistry, University of Connecticut, Storrs, Connecticut 06269-3060, USA
Received 2 November 1999; accepted 6 January 2000
Abstract
When 2-methylene-3-phenyloxetane (4) was treated with excess lithium and DTBB (cat.), lactone 6 was isolated
as an unexpected product in high yield. It is postulated that 6 arises from a coupling of a radical enolate derived
from 4 and the enolate of acetaldehyde, a product of THF decomposition under the reaction conditions. © 2000
Elsevier Science Ltd. All rights reserved.
In the preceding paper,1 the formation of an enolate dianion 2 via the reductive ring-opening of 3,3-
dimethyl-2-methylene-4-phenyloxetane (1) was described (Scheme 1). Oxetane 1 was used because prior
work on reductive ring-opening of heterocycles had demonstrated that more efficient and regioselective
reactions occurred when a phenyl group was located on the carbon alpha to the heteroatom.2–5 In terms
of potential utility in organic synthesis, we felt it was important to ascertain if a phenyl substituent at
C-4 was required for reductive ring-openings of 2-methyleneoxetanes. Because of the volatility of 3, we
decided to examine the reactivity of 2-methylene-3-phenyloxetane (4).6 The unexpected and interesting
result of this study is disclosed in this communication.
Scheme 1.
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00060-5
tetl 16340

1860
Fig. 1. HETCOR of 6
2-Methylene-3-phenyloxetane (4) was added to a stirred suspension of lithium (15 equiv.) and 4,40-
di-tert-butylbiphenyl (DTBB) (0.1 equiv.) in THF at −78°C. After 20 min the reaction was placed in
an ice bath. Contrary to the cases where dianion formation was successful, no deep green color was
1
observed. A H NMR spectrum of the reaction mixture showed the formation of a single product. If
the dianion of 4 had formed, the expected product from an aqueous quench would have been methyl
ketone 5 (C10H12O). This was clearly not the case (see Fig. 1). Although the proton count (12) was
the same, the chemical shifts were not consistent with ketone 5. Further, the 13C NMR (Fig. 1) of the
purified compound demonstrated that it contained 12 carbons, rather than 10, and suggested that it was
an ester (peak at 163.0 ppm), rather than a ketone. Microanalytical and HRMS data on the purified,

1861
unknown compound were consistent with a molecular formula of C12H12O2. The isolated yield based on
that molecular formula was 84%.
Working with the molecular formula of C12H12O2 and with the 1H and 13C spectra, several structural
features could be deduced. The presence of an alkene was suggested by the 1H (peaks at 5.9 and 5.3 ppm)
and 13C (seven peaks in the sp2 region — four from the phenyl ring, and one from the ester, leaving two
peaks) NMR spectra. Further, a HETCOR analysis (Fig. 1) demonstrated that both alkene protons were
on the same carbon (114.9 ppm), indicating a disubstituted terminal alkene. The HETCOR spectrum also
showed that the proton at 4.96 ppm correlated with the carbon at 78.3 ppm. Presumably, this carbon
was directly attached to an oxygen. The molecular formula indicated seven elements of unsaturation.
The phenyl ring, carbonyl and alkene accounted for six of those, suggesting the presence of a ring. The
C_O stretch in the IR spectrum appeared at 1800 cm−1 (thin film), and the λmax (CHCl3) was 260 nm.
These observations, combined with the chemical shifts, were consistent with what would be expected for
γ-lactone 6.7
In trying to account for the reproducible formation of this product we recognized that the extra
carbons and the extra oxygen could not come from the partial decomposition of 2-methyleneoxetane
4; the isolated yield made this impossible. Thus, the only logical source for the additional carbons and
the oxygen would be THF. Indeed, the reaction did not proceed in either DME or ether. It is known
that in the presence of butyllithium THF decomposes somewhat to give ethylene and the enolate of
acetaldehyde.8 Thus, we presumed that our additional carbons and oxygen were furnished by the enolate
of acetaldehyde. If that is accepted, the formation of 6 is still a challenge to rationalize. A partial potential
pathway is shown in Scheme 2.
Scheme 2.
It is reasonable that the hydrogen atom at C-3 of 4 could be abstracted, since it is both allylic
and benzylic. The importance of this feature is confirmed by the fact that 3-methyl-2-methylene-3-
phenyloxetane (8) is completely inert under the reaction conditions, even if the reaction mixture is
elevated to room temperature. Elimination via a radical pathway seems more plausible than reduction
of benzylic radical 7 to an anion, since the enolate resulting from a reduction/elimination process would
not react with the enolate of acetaldehyde. Further, there is precedent for the coupling of an enolate
and an enolate radical.9,10 It is harder to predict the sequence of reactions that follows, including the
order of bond formation. Nevertheless, the types of reactions shown are not unreasonable under the

1862
reaction conditions. What is both remarkable and noteworthy is the efficiency of the whole process. We
are currently trying to ascertain the tolerance of this reaction to functionality at C-4.
Acknowledgements
Helpful discussions with Theodore Cohen, David Van Vranken and Dennis Curran, as well as helpful
suggestions by a referee, are gratefully acknowledged. We would like to thank Xiangqun Xie for
performing a COSY of 6. A.R.H. thanks the NSF for a CAREER Award.
References
1. Hashemzadeh, M.; Howell, A. R. Tetrahedron Lett. 2000, 41, 1855–1858.
2. Bartmann, E. Angew. Chem., Int. Ed. Engl. 1986, 25, 653–654.
3. Mudryk, B.; Cohen, T. J. Org. Chem. 1989, 54, 5657–5659.
4. Choudhury, P. K.; Almena, J.; Foubelo, F.; Yus, M. Tetrahedron 1997, 53, 17 373–17 382.
5. Azzena, U.; Pilo, L. Synthesis 1999, 664–668.
6. Dollinger, L. M.; Ndakala, A. J.; Hashemzadeh, M.; Wang, G.; Wang, Y.; Martinez, I.; Arcari, J. T.; Galluzzo, D. J.; Howell,
A. R.; Rheingold, A. L.; Figuero, J. S. J. Org. Chem. 1999, 64, 7074–7080.
7. Our 1H NMR spectrum did not correspond with the 1H NMR data of the two previously reported syntheses of 6: (a) Larock,
R. C.; Leuck, D. J.; Harrison, L. W. Tetrahedron Lett. 1988, 29, 6399–6402. (b) Kang, S.-K.; Bai, T.-G.; Kulak, A. N. Synlett
1999, 324–326. However, the structure of 6 is fully consistent with our HETCOR and COSY (not shown) data.
8. Bates, R. B.; Kroposki, L. M.; Potter, D. E. J. Org. Chem. 1972, 37, 560–562.
9. Cohen, T.; McNamara, K.; Kuzemko, M. A.; Ramig, K.; Landi Jr., J. J.; Dong, Y. Tetrahedron 1993, 49, 7931–7942.
10. Russell, G. A.; Khanna, R. K. J. Am. Chem. Soc. 1985, 107, 1450–1452.