Synthesis and Decomposition of a 1,2-Diazetine Dioxide
TABLE 1. Computed Energies of Compounds Relevant to the
Decomposition of Diazetine Dioxide 1aa
compd
Ea (hartrees)
1a
3a
3b
NO
4
-800.71679
-540.83905
-540.84665
-129.93143
-800.75874
-800.70326b
generally has a higher activation barrier than C-N bond
cleavage leading to loss of NO.16a,17 The picture that emerges
from these various studies suggests that formation of forma-
doxime occurs exclusively from the cis-azo dioxide form (by
an as yet unknown mechanism), while monomerization (and
ensuing C-N bond scission) occurs at higher temperatures.
In one regard, compound 1a is essentially an azo dioxide
constrained to a cis conformation.9 Its thermal behavior appears
to parallel that of the cis dimer of nitrosomethane. Low-
temperature decomposition (ca. room temperature) leads to
diphenyl glyoxime as the major product in the highly polar
solvent DMSO-d6 and as a competing reaction product in CDCl3.
Isomerization probably results from direct conversion of the
diazetine dioxide rather than via initial ring opening to the
bisnitroso compound 2 since, as mentioned earlier, there is no
evidence for even transient formation of 2. The direct Et3N-
catalyzed conversion of 1a to diphenyl glyoxime in CHCl3 is
consistent with previous studies of base-catalyzed isomerizations
of azo dioxide compounds to the corresponding oximes.18
Thermal decomposition of diazetine dioxide 1a apparently
takes place via two closely competing routes: (i) isomerization
to diphenyl glyoxime and (ii) C-N bond scission to form a
transient diradical intermediate followed by loss of N2O2 to form
trans-stilbene 3b. We computed the energies of the compounds
involved in both of these transformations at the B3LYP/6-
311+G(d,p)//B3LYP/6-31G(d) levels and the results are sum-
marized in Table 1. The isomerization pathway to form
glyoxime 4 is predicted to be exothermic by 26 kcal/mol, while
loss of NO to form 3b is predicted to be endothermic by 5 kcal/
mol. However, the formation of 3b will almost certainly be
favored entropically. Thus at room temperature (and especially
in the base-catalyzed reaction) the isomerization process is able
to compete, while at higher temperatures, where entropic effects
are more strongly exerted, NO elimination is favored. Notice
that formation of the bisnitroso compound 2 (via mechanism
A) is predicted to be endothermic by 8 kcal/mol and that
formation of the stereochemically retained cis-stilbene 3a (via
mechanism B) is predicted to be endothermic by 10 kcal/mol,
which further suggests that these pathways are less likely than
the alternative modes of decomposition. We are currently
extending our computational studies on model compounds to
more closely examine the transition states expected from the
various possible modes of decomposition.
2
a B3LYP/6-311+G(d,p)// B3LYP/6-32G(d). b Computed at the most
stable conformation as determined by a Monte Carlo distribution generated
by a MMFF calculation.
CDCl3) δ 7.24 (s, 10H), 6.43 (s, 2H); 13C NMR (15 MHz, CDCl3)
δ 129.9, 129.5, 129.3, 128.6, 78.6; IR (solid) cm-1 3065, 2993,
1541, 1452, 692; UV (CH3OH) λmax ) 264 (ꢀ ) 19 000).
Thermolysis of 1a. Diazetine dioxide 1a (16 mg, 0.07 mmol)
was dissolved in 1 mL of CHCl3 (passed through a column of Al2O3
to remove acidic impurities) and the solution was quickly heated to
reflux in a 5 mL conical vial fitted with a reflux condenser and dry-
ing tube. A reddish-brown gas evolved quickly from the heated solu-
tion and dissipated within 30 min. The solution was cooled and
the CHCl3 removed under reduced pressure. 1H NMR analysis of the
resulting white solid showed only trans-stilbene (identified by com-
parison to a commercially available sample). Analysis by GCMS
revealed small amounts of benzaldehyde (2%) and (Z)-1,2-diphenyl-
1-nitroethylene (<1%) in addition to trans-stilbene. The reaction
products were identified (GC retention times and mass spectral data)
by comparison to commercially available samples. A control run
established that trans-stilbene is not isomerized to cis-stilbene even
at the high temperatures of the GC injection port (350 °C).
1
Determination of the Rate of Decomposition of 1a by H
NMR Spectroscopy. Diazetine dioxide 1a (20 mg, 0.08 mmol)
was dissolved in either DMSO-d6 or CDCl3 and 5 µL of 1,2-
dichloroethane was added as an internal standard. The resulting
solution was transferred to an NMR tube. For the high-temperature
kinetics runs, the sample was frozen under vacuum and the tube
flame sealed. The samples were submerged in a silica gel bath
heated to the desired temperature and removed at regular intervals
for analysis by NMR spectroscopy. For the 29 °C runs (i.e.,
temperature in the cavity of the NMR instrument), the NMR tube
was tightly capped and sealed with paraffin. In all cases, the rate
of decrease of the proton signal corresponding to the hydrogens at
the 3- and 4-positions of the diazetine dioxide ring was determined
relative to the internal standard by integration. An average of at
least three runs were taken in each case. The reaction mixtures
were analyzed directly by GCMS. The reaction products were
identified (GC retention times and mass spectral data) by compari-
son to commercially available samples.
Decomposition of 1a in the Presence of Triethylamine.
Diazetine dioxide 1a (20 mg, 0.08 mmol) was dissolved in a mixture
of 1 mL of CHCl3 and 0.5 mL of Et3N. The solution was allowed
to sit for 3 h after which it was directly analyzed by GCMS.
Experimental Section
3,4-Diphenyl-1,2-diazetine 1,2-dioxide (1a). To 122 mL of a
freshly prepared solution of dimethyldioxirane in acetone19 at 0 °C
(∼0.07 M, 5 equiv) was added 0.364 g (1.72 mmol) of meso-1,2-
diphenylethylenediamine (Aldrich) as a solid at one time. The
solution quickly became blue-green in color, which then dissipated
within 30 s. The solution was stirred for 1 h and the acetone
removed under reduced pressure. Column chromatography (SiO2,
1:1 hexane/EtOAc) afforded 0.11 g of a pale yellow solid. This
solid was chromatographed a second time (SiO2, CH2Cl2) to afford
Computational Details
All of the computations reported in this paper were obtained
with the Becke3LYP density functional as implemented within the
Spartan ’02 program.20 Geometries and frequencies were computed
(20) Spartan ’02; Wavefunction, Inc.: Irvine, CA. Kong, J.; White, C.
A.; Krylov, A. I.; Sherrill, C. D.; Adamson, R. D.; Furlani, T. R.; Lee, M.
S.; Lee, A. M.; Gwaltney, S. R.; Adams, T. R.; Ochsenfeld, C.; Gilbert, A.
T. B.; Kedziora, G. S.; Rassolov, V. A.; Maurice, D. R.; Nair, N.; Shao,
Y.; Besley, N. A.; Maslen, P. E.; Dombroski, J. P.; Daschel, H.; Zhang,
W.; Korambath, P. P.; Baker, J.; Byrd, E. F. C.; Van Voorhis, T.; Oumi,
M.; Hirata, S.; Hsu, C.-P.; Ishikawa, N.; Florian, J.; Warshel, A.; Johnson,
B. G.; Gill, P. M. W.; Head-Gordon, M.; Pople, J. A. J. Comput. Chem.
2000, 21, 1532.
1
26.2 mg of 1a (7% yield) as a white solid. H NMR (60 MHz,
(18) Di Giacomo, A. J. Org. Chem. 1965, 30, 2614-2617.
(19) Crandall, J. K.; Batal, D. J.; Sebasta, D. P.; Lin, F. J. Org. Chem.
1991, 56, 1153-1166.
J. Org. Chem, Vol. 72, No. 4, 2007 1415