Interestingly, the nitrogen atom of 7a is not trigonal planar
in marked contrast to that of 1, which is isoelectronic to 7a,
though other structural features of 7a are almost the same
as those of 1 and 3a.4,6a Pyramidal configuration for the
nitrogen atom of 7a would emerge from the steric repulsion
between the Martin ligand and the phenyl group on the
nitrogen atom.
phosphorus analogue 1. This is the first finding for aziridine
formation pathways from heteracyclobutanes with highly
coordinate main-group elements. Compounds 10 and 11 are
most likely formed by the ring-opening reaction of aziridine
8, giving azomethine ylide 12, followed by the hydrolysis
of 12 with water in the reaction mixture, indicating that 10
and 11 are the secondary products of 8 (Scheme 3).
Consequently, the four-membered ring significantly devi-
ates from planarity as indicated by the torsion angle [Se1-
N1-C2-C1] of 7a [21.9(1)°]. The Se1-C1 bond length
[1.991(2) Å] is slightly longer than that of previously reported
oxaselenetanes 3a [1.923(7) Å].6a Taking into account the
standard deviations, other common bond lengths of the four-
membered ring are almost same as those of 1 and 3a.4,6a
The Se1-O1 and Se1-C3 bond lengths are also almost same
as those of 3a. The bond angle between two apical bonds
deviates by 21.30(7)° from 180°, probably due to its ring
strain. Such deviation was also observed for 1 and 3a.4,6a
Since the spectral features of both 7a and 7b are very similar,
7b should be another diastereomer of 1,2-selenazetidine in
regard to the configuration at the selenium, that is, the phenyl
group at 4-position in 7b is trans to that of the lone pair but
cis in 7a.
Scheme 3
The above experimental result shows that a 1,2-selen-
azetidine has a reactivity similar to that of 1,2-oxaselenetanes
3b at the point of giving the corresponding three-membered
ring compound on the thermolysis. Such ligand coupling
reactions on selenium would be favorable compared with
the olefin formation, because the energy gain obtained by
the formation of the double-bond between selenium and
oxygen or nitrogen is too small to be the driving force of
the latter reaction. We previously reported the possibility that
the 1,2-oxachalcogenetanes are intermediates of the Corey-
Chaykovsky reaction of chalcogenonium ylides with carbonyl
compounds.5,6b In addition, such an aziridine formation of a
sulfonium ylide with a cyclic imine was previously reported
as a nitrogen version of the Corey-Chaykovsky reaction.14
The result of thermolysis of 7a giving 8 indicates that a 1,2-
selenazetidine might be the intermediate of the selenium
version of this reaction.
Thermolysis of 7a at 210 °C in xylene-d10 was monitored
1
by H and 19F NMR spectroscopy (Scheme 2). The signals
Scheme 2
Similarly, thermolysis of 7b (xylene-d10, at 210 °C) also
gave 8 (79%), 9 (100%), 10 (15%), and 11 (16%) as final
products, but 7a was also observed during the reaction. The
thermal reaction of 7b at 150 °C gave predominantly 7a
(78%) with a small amount of 8 (13%) and 9 (13%) in 91%
conversion of the reaction (10 h), resulting in the formation
of 8, 9, 10, and 11 as the final products (225 h). These results
indicate that 7b isomerized to 7a before its thermolysis.
In summary, we have revealed an interesting difference
in reactivity between azetidines with group 15 elements and
those with group 16 elements. Hence, the above experimental
due to the corresponding aziridine 8 and cyclic selenenate 9
increased with the decrease in those of 7a. Finally, 8 (78%),
9 (100%), 10 (16%), and 11 (16%) were obtained after
complete consumption of 7a.13 No formation of the corre-
sponding olefin was observed, in sharp contrast to the
(11) A crystal data of 7a. C25H15F12NOSe, FW ) 652.34, colorless
crystals, monoclinic, space group P21/n, a ) 9.7010(2), b ) 15.3350(5),
and c ) 16.8410(5) Å, â ) 101.446(2)°, V ) 2455.5(1) Å3, Z ) 4, Fcalcd
) 1.764 g cm-3, T ) 150 K. Of the 4474 reflections which were collected,
4340 were unique. The structure was solved by direct methods and expanded
by using Fourier techniques. The final cycle of full-matrix least-squares
refinement on F2 was based on all 4340 observed reflections and 361
variable parameters and converged with unweighted and weighted agreement
factors of R1 ) 0.039 and wR2 ) 0.118.
(8) For reviews on the 77Se NMR, see: (a) Luthra, N. P.; Odom, J. D.
In The Chemistry of Organic Selenium and Tellurium Compounds; Patai,
S., Rappoport, Z., Eds.; John Wiley & Sons: New York, 1986; Vol. 1;
Chapter 6, pp 189-241. (b) Paulmier, C. Selenium Reagents and Intermedi-
ates in Organic Synthesis; Pergamon: Oxford, 1986; pp 17-21.
(9) For selenuranes with Se-N bond(s), see: (a) Roesky, H. W.;
Ambrosius, K. Z. Naturforsch. 1978, 33B, 759-762. (b) Fujihara, H.; Mima,
H.; Erata, T.; Furukawa, N. J. Chem. Soc., Chem. Commun. 1991, 98-99.
(c) Fujihara, H.; Mima, H.; Ikemori, M.; Furukawa, N. J. Am. Chem. Soc.
1991, 113, 6337-6338. (d) Fujihara, H.; Ueno, Y.; Chiu, J.-J.; Furukawa,
N. J. Chem. Soc., Perkin Trans. 1 1992, 2247-2248. (e) Fujihara, H.; Mima,
H.; Erata, T.; Furukawa, N. J. Am. Chem. Soc. 1993, 115, 9826-9827. (f)
Fujihara, H.; Mima, H.; Furukawa, N. Tetrahedron 1996, 52, 13951-13960.
(g) Mima, H.; Fujihara, H.; Furukawa, N. Tetrahedron 1998, 54, 743-
752.
(12) Crystallographic data for the structure reported in this paper have
been deposited with the Cambridge Crystallographic Data Centre as
supplementary publication no. CCDC-149505. Copies of the data can be
obtained free of charge on application to CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax (+44)1223-336-033; e-mail deposit@ccdc.cam.ac.uk).
(13) The yields of these products were determined on the basis of the
1
integral of H and 19F NMR spectra of the reaction solution.
(10) Granoth, I.; Martin, J. C. J. Am. Chem. Soc. 1979, 101, 4618-
4621.
(14) Hortmann, A. G.; Robertson, D. A. J. Am. Chem. Soc. 1967, 89,
5974-5975.
Org. Lett., Vol. 3, No. 5, 2001
693