could be effected quite well with standard column chroma-
tography rather than the laborious series of fractional
crystallizations previously reported.8,10
According to the literature,8 five recrystallizations of a 50/
50 mixture of diastereomers afforded the meso-isomer in 95%
purity, while a single column afforded the meso-isomer in
99% purity (by GC). A second column purification on the
enriched d,l-isomer afforded the diastereomer in 94% purity
while the literature reports it in only 65% purity. Zinc metal
reduction of the stereochemically enriched dinitro compounds
to bishydroxyl amines 5 followed by oxidation to the
corresponding diazetine dioxides proceeded according to
literature precedent.7c
Figure 2. Solvent-dependent pathways for the decomposition of
1a
We recently reported the kinetics of decomposition of
compound 1b as determined by 1H NMR spectroscopy.1 We
desired to compare the rate of decomposition of compounds
bond in the diazetine dioxide to form a transient diradical
followed by bond rotation and cleavage of the remaining
C-N bond. We wondered whether simple alkyl-substituted
diazetine dioxides might exhibit the same sort of scrambling
of stereochemistry upon decomposition.
The stereochemically defined diazetine dioxides 1c and
1d have been briefly reported in the literature as synthetic
intermediates, but their decomposition was apparently not
investigated.6 We synthesized compounds 1c and 1d accord-
ing to literature procedures that generally mirrored the
synthesis of the well-known tetramethyl derivative 1b
(Scheme 1).7 Attempted synthesis of the dinitro intermediates
1
1c,d with that of 1b, but H NMR spectroscopy was not
nearly as appealing a method because of the greater
complexity of the NMR spectra of 1c,d (and, hence, expected
overlap of signals). We found that the decomposition of 1b
could be readily followed by UV-vis spectroscopy at 100
°C in high-boiling n-butanol as solvent by following loss of
the absorption at 257 nm (corresponding to the azo dioxide
functional group). Clean, first-order kinetics were observed.
The rate determined (k ) [1.2 ( 0.2] × 10-4 s-1) was
consistent with the rate determined by NMR spectroscopy
in DMSO-d6 (k ) [8.9 ( 0.1] × 10-5 s-1).1 Compounds 1c
(k ) [2.2 ( 0.1] × 10-4 s-1) and 1d (k ) [1.7 ( 0.2] ×
10-4 s-1) exhibited marginally higher rates of decomposition,
possibly attributable to the slightly greater steric interactions
of the larger ethyl groups on these compounds relative to
the four methyl groups of 1b (the cis-oriented ethyl groups
of 1c affording the greatest steric interactions and therefore
the highest reaction rate). Despite very subtle differences,
all of the compounds (as would be expected) decomposed
at similar rates.
Scheme 1. Synthesis of Diazetine Dioxides 1c and 1d
For product analysis, decomposition of the diazetine
dioxides was effected in the heated injector port (set at 200
°C) of a GCMS. In both cases, only NO and the correspond-
ing alkenes were observed. Thermolysis of the enriched d,l-
a Separation of the diastereomers was effected via standard
column chromatography to afford enriched meso- and d,l-isomers.
bReaction of the meso-enriched dinitro compound led to 1c.
Table 1. Thermal Decomposition of Diazetine Dioxides 1c and
1d
4 in aqueous solution according to methods reported in the
literature were not successful in our hands.7c,8 We were,
however, able to synthesize 4 via coupling of the preformed
lithium salt of nitrobutane with independently prepared
2-bromo-2-nitrobutane in DMSO.9 Furthermore, we found
that separation of the meso- and d,l-dinitro stereoisomers
(6) White, D. K.; Greene, F. D. J. Am. Chem. Soc. 1978, 100, 6760-
6761.
(7) (a) Singh, P.; Boocock, D. G. B. Tetrahedron Lett. 1971, 42, 3935-
3938. (b) Greene, F. D.; Gilbert, K. E. J. Org. Chem. 1975, 40, 1409-
1415. (c) Ullman E. F. Diazacyclobutanes. U.S. Patent 4,032,519, June 28,
1977.
(8) Shustov, G. V.; Tavakalyan, N. B.; Shustova, L. L.; Pleshkova, A.
P.; Kostyanovskii, R. G. IzV. Akad. Nauk SSSR, Ser. Khim. 1982, 364-
375.
(9) Kornblum, N.; Boyd, S. D.; Pinnick, H. W.; Smith, R. G. J. Am.
Chem. Soc. 1971, 93, 4316-4318.
compd
temp (°C)
ratio Z:E
% stereoretention
1ca
1ca
1db
1db
200
350
200
350
94:6
95
83
95
89
83:17
10:90
15:85
a Enriched sample (99% meso: 1% d,l). b Enriched sample (94% d,l:
6% meso).
3006
Org. Lett., Vol. 9, No. 16, 2007