A new synthetic route to 3-Oxo-4-amino-1, 2, 3-oxadiazole from the diazeniumdiolation of benzyl cyanide: Stable sydnone iminium N-Oxides

Article abstract of DOI:10.1021/jo802343k

Treating benzylcyanide with nitric oxide in basic methanol returns 3-oxo-4-phenyl-5-amino-1, 2, 3- oxadiazole (a 5-iminium-sydnone N-oxide), 5, in addition to the known 2-(hydroxyimino)-2-phenylac-etonitrile, 6, and the bisdiazeniumdiolate imidate salt, 7. Conditions for the separation and purification of the new 1, 2, 3-oxadiazole 5 are described along with its theory, structure, spectroscopy, and reactivity which demonstrate the predominance of the amino tautomer over the N-hydroxide tautomer. New derivatives of 5 include a Schiff base, from the condensation with salicylaldehyde, 8, and a dimethylamino analogue of 5, from its reaction with methyliodide and NaH. As with other related 3-oxo-1, 2, 3-oxadiazoles 5 has pronounced acid/base stability. Theoretical calculations demonstrate that the only other prior sydnone N-oxides prepared were misformulated as the N-hydroxide imines and are better described as 3-oxo-5-amino-derivatives of 1, 2, 3-oxadiazoles.

Full text of DOI:10.1021/jo802343k

A New Synthetic Route to 3-Oxo-4-amino-1,2,3-oxadiazole from the
Diazeniumdiolation of Benzyl Cyanide: Stable Sydnone Iminium
N-Oxides
D. Scott Bohle* and Inna Perepichka
Department of Chemistry, McGill UniVersity, 801 Sherbrooke Street West,
Montreal, H3A 2K6, Quebec, Canada
scott.bohle@mcgill.ca
ReceiVed October 21, 2008
Treating benzylcyanide with nitric oxide in basic methanol returns 3-oxo-4-phenyl-5-amino-1,2,3-
oxadiazole (a 5-iminium-sydnone N-oxide), 5, in addition to the known 2-(hydroxyimino)-2-phenylac-
etonitrile, 6, and the bisdiazeniumdiolate imidate salt, 7. Conditions for the separation and purification of
the new 1,2,3-oxadiazole 5 are described along with its theory, structure, spectroscopy, and reactivity
which demonstrate the predominance of the amino tautomer over the N-hydroxide tautomer. New
derivatives of 5 include a Schiff base, from the condensation with salicylaldehyde, 8, and a dimethylamino
analogue of 5, from its reaction with methyliodide and NaH. As with other related 3-oxo-1,2,3-oxadiazoles
5 has pronounced acid/base stability. Theoretical calculations demonstrate that the only other prior sydnone
N-oxides prepared were misformulated as the N-hydroxide imines and are better described as 3-oxo-5-
amino-derivatives of 1,2,3-oxadiazoles.
Introduction
The singular synthetic access to 1,2,3-oxadiazoles or sydnones
has sharply limited their chemistry and utility. As a consequence,
although these dense highly polarizable mesoionic heterocycles
have many potential applications as new materials, pharmaceu-
ticals, or as synthons for other heterocycles, this potential has
not been realized. The original synthesis of the sydnone core
1, from the cyclodehydration of N-alkyl or aryl N-nitroso-R-
aminoacids with acetic anhydride, eq 1,1 has since been
In this unique and remarkable prior report the product was
formulated as an N-hydroxide B and its unusual acid and base
stability was noted in passing.
elaborated2 but as a rule the chemistry of the N-aryl and alkyl
derivatives is limited by their facile hydrolysis in strong acids
or base. In 1971 Go¨tz and Grozinger3 discovered an alternative
to this synthesis from the nitrosation of the Mannich condensa-
tion of aldehydes with hydroxylamine and cyanide to give 2,
eq 2, with overall yields between 2% and 10%.
Recently two new syntheses of the 1,2,3-oxadiazoles from
the diazeneiumdiolation of acidic carbon precursors were
described, eqs 3 and 4. The reaction of NO with alkynes returns
3, eq 3,4 while with dimethylmalonate in base the product 4
forms in 31% yield, eq 4.5 Unlike the N-aryl/alkyl sydnones 1,
(1) Earl, J. C.; Mackney, A. W. J. Chem. Soc. 1935, 899.
10.1021/jo802343k CCC: $40.75
Published on Web 01/14/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 1621–1626 1621

Bohle and Perepichka
TABLE 1. Reaction Conditions for the Diazeniumdiolation of
Benzyl Cyanide in the Presence of Basea
conditions
yield, %
T, °C
BnCN
+
5
6
7
BnCN/THF
(0-3) °C to rt
(0-3) °C to rt
no reaction occurred
no reaction occurred
BnCN/CH3OH
+
BnCN/CH3ONa/CH3OHb rt
BnCN/CH3OK/CH3OH
trace 10.5
3
51
1:1
1:1
1:0.1
1:1.5
1:2
1:2
-20 °C
(0-3) °C to rt
(0-3) °C to rt
(0-3) °C (48 h) trace 22
(0-3) °C to rt
60 °C, 24 h
+
trace
+
trace <5
10
30
trace
32
32
6
11
+
5
trace
-
trace
4
9
-
70
-
BnCN/KOH/CH3OH
the N-oxides 2A, 3, and 4 are remarkably stable toward acids
and bases and have an emerging unique chemistry that includes
the facile electrophilic decarboxylation6 of 4 and the kinetic or
thermodynamically controlled O- or N-methylation.7 In the
course of extending our studies of the diazeniumdiolation of
carbanions with nitric oxide8 we reexamined the reaction9 of
nitric oxide with benzyl cyanide described by Arnold et al. and
have found that in addition to the nonheterocyclic products
described by these authors a new 1,2,3-oxadiazole, 5, can be
isolated from these reaction mixtures.9 In this paper we describe
(1) a new synthesis of a new 3-oxo-1,2,3-oxadiazole, 5, (2) the
structure and spectroscopy of the species derived from benzyl-
cyanide, (3) theoretical results relating to its ground state
tautomeric structure, and (4) its reactions with salicylaldehyde
and methyliodide to give a Schiff base and dimethylamino
derivatives, respectively. Together these results demonstrate that
5 is an amine derivative A of the sydnone N-oxide family and
not the N-hydroxy tautomer 2B.
1:1
1:2.5
1:5
1:1
rt
rt
rt
+
10
11.1
5
+
+
46
+
trace
20
25
+
+
0-3 °C
8
30
BnCN/K2CO3/CH3OH
3:1
rt
+
trace trace
-
a Yields for the dipotassium salt 7 were determined from the dried
precipitate isolated from the reaction mixture. The yields for 5 and 6
were determined after chromatographic separation of the reaction
mixture. b Conditions were the same as used in ref 9 with 7 in this case
being isolated as a disodium salt.
that although methanol reacts with nitric oxide to give formate,10
this is a slow reaction under these conditions and no potassium
formate is observed in the precipitate.
The identity of the new product in this reaction, 3-oxo-4-
amino-5-phenyl-1,2,3-oxadiazole, 5, was clearly established by
X-ray diffraction, spectroscopy, and elemental analysis. The
X-ray crystallography for 5, Figure 1, was particularly informa-
tive due to the high-quality crystals grown for 5 from acetone/
ether and which allow for data collection in the Ewald sphere
out to 56° in 2θ. As a consequence, all protons were located in
the penultimate electron density maps and could be refined
independently. At this stage two protons were clearly located
on the amino nitrogen in 5, which also refined as a nonplanar
geometry with bond angles at the amino nitrogen summing to
347.2°. For a disordered structure with an imine tautomer B
the geometry at the nitrogen is anticipated to be coplanar with
the ring. In the final refinement, the amino hydrogen atoms were
refined isotropically while for the phenyl ring a riding model
was used for the five carbon bound hydrogens. The resulting
metric parameters are listed in Table 2 with the ORTEP
depiction shown in Figure 1. Attempts to refine this structure
as an N-hydroxyimino tautomer B with riding models for a
proton on O(1) and N(3) consistently gave poorer models for
the electron density as evidenced by the higher R values
(>0.5%), larger Sgof values, and larger metric parameter esd
values. The phenyl and sydnone rings are not coplanar and form
a dihedral angle of 32.3(1)°. Crystal packing in 5 juxtaposes
the amine hydrogens and the N-oxide at the 3 position to form
a centrosymmetric eight-member hydrogen-bonded ring, Figure
2. The heavy atoms in this ring are coplanar and minor
deviations of the hydrogens H1A and H1B are 0.12 and 0.19 Å
Results
The reactions of nitric oxide with organic substrates are often
strongly dependent upon the base, pressure, temperature, and
salts formed. These factors influence the equilibrium between
the NO monomer and the NO dimer, both of which are
competent electrophiles for these carbanions.7 Although the
products from this reaction, eq 5 and Table 1, are usually
dominated by the bisdiazeniumdiolate imidate salt 7, conditions
for the isolation of 6 in 70% yield and for the 3-oxo-1,2,3-
oxadiazole 5 in 22% yield are described. This modest yield of
5 is offset by the ease of isolation that simply requires a filtration
from the dipotassium salt 7 and a chromatographic separation
from any unreacted starting material and the imidate 6. The
products in eq 5 are all air and thermally stable derivatives which
do not require any additional care in handling or storage. Note
(2) Gilchrist, T. L.; O’Neill, P. M. In ComprehensiVe Heterocyclic Chemistry
II; Katrizky, A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon: London,
UK, 1996; Vol. 5, pp 165-178.
(3) Gotz, M.; Grozinger, K. Tetrahedon. 1971, 27, 4449–4456.
(4) Sugihara, T.; Kuwahara, K.; Wakabayashi, A.; Takao, H.; Imagawa, H.;
Nishizawa, M. Chem. Commun. 2004, 216–217.
(7) Bohle, D. S.; McQuade, L. E.; Perepichka, I.; Zhang, L. J. Org. Chem.
2007, 72, 3625–3631.
(8) Arulsamy, N.; Bohle, D. S. J. Org. Chem. 2006, 71, 572–581.
(9) Arnold, E. V.; Keefer, L. K.; Hrabie, J. A. Tetrahedron Lett. 2000, 41,
8421–8424.
(5) Arulsamy, N.; Bohle, D. S. Angew. Chem., Int. Ed. 2002, 41, 2089–
2091.
(6) Bohle, D. S.; Ishihara, Y.; Perepichka, I.; Zhang, L. Tetrahedron Lett.
2008, 48, 4550–4552.
(10) DeRosa, F.; Keefer, L. K.; Hrabie, J. A. J. Org. Chem. 2008, 73, 1139–
1142.
1622 J. Org. Chem. Vol. 74, No. 4, 2009

Stable Sydnone Iminium N-Oxides
FIGURE 1. ORTEP representation of 5 with 50% thermal ellipsoids
shown for non-hydrogen atoms and the amine hydrogen atoms shown
as isotropic spheres.
TABLE 2. Experimental and Theoretical Dimensions for
Aminosydnones
FIGURE 3. Infrared spectra (KBr) for 5 shown in the bottom panel
(A) with N-H, O-H, and C-H stretching bands at 3447, 3298, 3249,
and 3113 cm-1 and for the deuterium exchanged sample shown in the
DFT,
DFT,
,
metric parameter
exptla b
amino Ac hydroxoimide Bc
top panel (B) with new peaks at 2578, 2443, and 2304 cm-1
.
C(1)-N(3), Å
1.3440(15) 1.36173
1.3739(16) 1.36689
1.3817(19) 1.41389
1.2893(13) 1.24665
1.3100(14) 1.31470
1.4014(13) 1.40844
1.3458(15) 1.33927
1.4601(16) 1.46133
132.11(11) 133.04745
130.78(11) 131.60755
126.32(10) 125.09716
1.26885
1.45302
1.34163
1.36263
1.30710
1.38721
1.40599
1.45302
124.01
C(1)-C(2), Å
short or long exocyclic N-C and N-O bond lengths. The
contrast of the solid state dimensions of C(1)-N(1) and
N(1)-O(1) with the calculated values for A and B shows them
to be much closer to the -NH2 tautomer A. A likely conse-
quence of the solid state H-bonds shown in Figure 2 is a slight
decrease in the C(1)-N(1) bond length and an increase in
N(1)-O(1) from those predicted in the gas phase. To aid in
the assignment of A and B we contrast the IR of the normal
NH2 and deuterated ND2 derivatives in Figure 3. Here three
N-H· · ·O bands are observed in addition to the characteristic
fingerprint bands for the 3-oxo-1,2,3-oxadiazole ring. The
number and relative intensities of the bands correspond well to
those predicted for a gas phase tetramer calculated at the
B3LYP/3-21+G* level, as well as those found for simple
aromatic amines such as aniline. Significantly, the band at 1660
cm-1 shifts to 1639 cm-1 on deuteration, and corresponds to
an antisymmetric combination of a ring stretch with exocyclic
CdN and NsO modes.
C(2)-N(1), Å
N(1)-O(1), Å
N(1)-N(2), Å
N(2)-O(2), Å
O(2)-C(1), Å
C(2)-C(3), Å
C(2)-C(1)-N(3), deg
C(1)-C(2)-C(3), deg
N(1)-C(2)-C(3), deg
129.356
128.38627
175.565
N(1)-C(2)-C(3)-C(4), deg 147.39(12) 147.35407
a Value from X-ray diffraction. b Metrics for the phenyl ring carbons
are not included here. c Calculated by density functional theory with a
triple-ꢀ basis set: B3LYP/6-311++G**.
The vibrational spectroscopy is consistent with the presence
of the N-oxo amino tautomer A versus the N-hydroxoimino form
B and the bond lengths, particularly the N-O and C-N
exocyclic bonds, correspond to those expected theoretically for
the N-oxo amino tautomer.
FIGURE 2. Hydrogen bonding between the amine and N-oxide groups
in 5. The center of inversion located at the midpoint relates two halves
of a ring.
out of this plane. ORTEP plots for this ring are depicted in
Figures S1 of the Supporting Information.
To assist with the structural assignment of 5, 2, and 9
(discussed below), ab initio density functional calculations were
used to determine the relative energies and vibrational frequen-
cies for the possible ground states. For 5 the two key tautomers
A and B both correspond to theoretical stationary points which
are local ground states with the geometries shown in the figures
in the Supporting Information. Their Cartesian coordinates and
normal vibrational modes are listed in Tables S2-S4 of the
Supporting Information. In general density function theory with
triple-ꢀ basis sets, B3LYP/6-311++G** gives very good
representations for the geometries, energies, and vibrations of
these types of heterocycles.7
In its reactions 5 also behaves like a nucleophilic amine as
well as a weak Bronsted acid. For example, with salicylaldehyde
5 adds slowly in toluene at reflux to give the yellow Schiff base
adduct 8, eq 6. This condensation is slower than that of typical
aromatic amines and reflects the weak nucleophilicity and
basicity of the amine in 5. This amine can also be dimethylated
with methyl iodide and sodium hydride, eq 7. The dimethy-
lamino product, 9, has similar UV and fingerprint-IR properties
In the gas phase the isomer 5A is 21.11 kcal·mol-1 more
stable than 5B. The gas phase optimized ab initio ground state
structures (B3LYP/6-311++G**) for A and B predict relatively
J. Org. Chem. Vol. 74, No. 4, 2009 1623

Bohle and Perepichka
formulated as a bis(N-methyl) derivatives of 2 or as O-acetylated
mesoionic analogues.3 Although the spectroscopic data presented
for the bis-N-methylation of 2 is unequivocal, their remains
considerable uncertainty in the acetylation site. We have found
that theory and X-ray crystallography is often required to predict
and to rigorously determine the site of electrophilic addition7
and the proposed structures of these acetylated products must
be considered tentative. There are potentially three acetylation
sites in 2 and to evaluate which is most likely to be the
thermodynamically preferred we have calculated the ground-
state geometries and energies with DFT for all three acetylated
isomers for acetylated 1,2,3-oxadiazine, Figure 4. Acetylation
of the amine nitrogen to give C gives a more stable isomer by
over 21.6 kcal ·mol-1 over the exocyclic O addition product D
and the ring acetylated product E. While substitution at the 4
position to give the derivatives prepared by Go¨tz and Grozinger
might affect the stabilities of C-E slightly, the most likely
product in his preparation is that related to C and this fits well
with the cursory IR data given in his paper, 1730 and 1660,
which most closely matches the strong bands associated with
this isomer. Go¨tz formulated his product as that shown in D.
FIGURE 4. Optimized ground states and relative B3LYP/6-311++G**
energies (kcal mol-1) for 3-oxo-4-methyl-5-amino-1,2,3-oxadiazole
acetylation isomers. Corresponding isomers for 3-oxo-5-amino-1,2,3-
oxadiazole, F, are listed but not shown. Strong vibrational acetyl and
ring modes are listed below each isomer. DFT also predicts a strong
amide like stretch at 1281 cm-1 found for C.
as 5 with its 1H and 13C NMR spectra indicating that the methyl
resonances are equivalent on the NMR time scale.
There are three prominent absorption bands in the ultraviolet
spectrum of 5 and these shift to lower energies when treated
with base and are then restored when treated with acid, Figure
S4 in the Supporting Information. There is no change in this
spectrum upon treatment with acid or dissolution in 1 M HCl.
The relative acidity of these protons is demonstrated by their
downfield shift in the 1H NMR spectrum to 8.05 ppm in contrast
with many aromatic amines which have a typical shift range of
3-5ppm.Aspectrophotometrictitrationin50:50water-methanol
with 1 M sodium hydroxide solution gives an isosbestic point
at 314 nm, and a pKa ) 11.4 for these protons.
The X-ray diffraction structure and DFT results also suggest
that A best describes the tautomeric structure. For example, the
exocyclic nitrogen-carbon bond length in 5, 1.3440(15) Å, is
well within the range of aromatic amine C-N bond lengths,
over 2100 entries, found in the Cambridge Crystallographic
database, 1.38(3) Å. Although there is some pyramidalization
of the geometry of the nitrogen in 5, where the three bond angles
at the nitrogen sum to 347°, this value should not be overin-
terpreted given the general problems of refining hydrogen atom
positions from X-ray data. However, DFT predicts N-pyrami-
dalization and in the gas phase these angles sum to 346.51°, in
remarkable agreement with experiment. Consistent with the
formation of the amine tautomer A for 5 is that it readily forms
a Schiff base 8 with salicylaldehyde, eq 6, and a dimethylamino
derivative, 9, eq 6.
Discussion
The diazeniumdiolation of carbanions is a general reaction
that requires careful control of the reaction conditions to achieve
product yield and specificity. The utility of the reaction is
exemplified here in that a stable heterocyclic N-oxide derivative
is readily prepared in a single step from the condensation of
readily available precursors. Thus while at first glance there is
an array of complex products formed during the reaction in eq
5, by varying the condition yields can be maximized to select
from the different products. The ease of product separation and
their inherent stability are important practical features in these
preparations as well.
The role of the N-oxide in stabilizing these heterocycles was
perhaps not appreciated by Go¨tz and Gronzinger in their initial
ground-breaking studies of 2. In their paper they are designated
variously as tautomer A seven times and as tautomer B seven
times and designated as imines in the title. Given that at the
point of time of their discoveries only 3-alkyl, aryl, and aminated
sydnones were known it is perhaps logical to formulate 2 as
derivatives with N-O single bonds rather than as N-oxides. We
now know the N-oxide is an important stabilizing factor in these
heterocycles and that it is responsible for their unusual stability.
These distinctions in tautomers are important if we are to
understand the reactivity of 2 and 5. Thus the Mannich
condensation products 2, which readily add electrophiles such
as methyliodide or acetic anhydride to give products, originally
Finally, although N-oxides are often written with the designa-
tion N+-O-, for many Lewis representations the valence bond
formulation does not really express the remarkable strength,
shortness, and inertness of these bonds and this functionality.
These bond lengths are typically much shorter than NsO single
bonds and even those found in many formally NdO double
bonding nitroso species. In terms of reactivity, heterocyclic
N-oxides are often poor ligands, and their weak Lewis basicity
correlates with poor O-atom nucleophilicity. One way to think
of the N-oxide substituent in these and many heterocycles is as
a lone pair place holder that blocks any reaction in which the
lone pair contributes substantially. In the case of the 1,2,3-
oxadiazoles the reaction which is blocked is the ring open diazo-
1624 J. Org. Chem. Vol. 74, No. 4, 2009

Stable Sydnone Iminium N-Oxides
tautomer G in eq 8.11 With this localization and ring opening
reaction blocked these compounds are more stable to acids,
bases, and heat.
were evaporated. IR (KBr, cm-1) 3109 w, 3068 w, 2578 m, 2443 m,
2303 m, 1639 s, 1509 s, 1465 w, 1449 w, 1371 s, 1327 s, 1309 m,
1284 m, 1237 w, 1192 w, 1181 m, 1060 w, 1013 m, 982 s, 910 m,
859 m, 799 w, 764 s, 691 s, 637 w, 620 w, 594 s, 511 m, 484 w.
2-(Hydroxyimino)-2-phenylacetonitrile, 6: mp 102-104 °C
The 1,2,3-oxadiazole heterocycle is readily stabilized by the
presence of an N-oxide substitution at the 3-position. Thus with
this substituent in place it is possible to isolate 2-N-alkylated
heterocycles, to decarboxylate the ring for 4-carboxylate-
substituted derivatives, and it is possible to isolate 5-amino
derivatives such as 5, 8, and 9. Clearly there is now considerable
scope for chemistry of the 3-oxides of the 1,2,3-oxadiazoles,
and there are likely to be a wide range of new heterocycles that
can be prepared from these simple rings.
(lit.12 mp 100 °C); H NMR (DMSO, 300 MHz, ppm) δ 7.49 -
1
7.55 (m), 7.69-7.72 (m), 7.92-7.95 (m), 13.8 (s); 13C NMR
(CDCl3, 300 MHz, ppm) δ 109.3, 126.5, 129.1, 129.3, 131.7, 149.9;
IR (KBr, cm-1) 3360 s, 2975 w, 2815 w, 2236 m, 1569 w, 1498 m,
1450 m, 1417 m, 1318 w, 1282 s, 1060 s, 1030 m, 1001 m, 967 s,
920 w, 765 s, 689 s, 675 s, 661 m, 607 m, 480 m; UV/vis (CH3OH,
λ
max, nm) 260. Anal. Calcd for C8H6N2O: C, 65.74; H, 4.13; N
19.16. Found: C, 65.47; H 3.81; N, 19.07.
Preparative Conditions for the Selective Synthesis of 6. Under
nitrogen in a glass pressure reaction bottle 0.50 g (4.27mmol) of
benzyl cyanide was added to a solution of potassium hydroxide in
absolute methanol (30 mL). The reaction flask was repeatedly
flushed with nitrogen, pressurized with nitric oxide (250 kPa), and
warmed to 60 °C (oil bath) for 24 h. After completion of the
reaction, the yellow reaction mixture was flushed with nitrogen and
the solvent was evaporated. Recrystallization of the crude product
from acetone/hexane gave pure compound 6, yield 0.42 g (70%).
Synthesis of Schiff Base 8. Under nitrogen, a mixture of 5 (0.022
g, 0.18mmol) and salicylaldehyde (0.031 g, 1.75 mmol) in 10 mL
of toluene was refluxed for 24 h. After evaporation, the crude yellow
product was purified by column chromatography on silica gel with
hexane/ethyl acetate (1:2, v/v) as an eluent, Rf 0.5, to give 0.02 g
(42.5%) of compound 8 as a bright yellow solid: mp 162 °C (DSC);
1H NMR (CDCl3, 400 MHz, ppm) δ 7.06 (t, J ) 8 Hz, 2H), 7.54
(m, 5H), 7.94 (d, J ) 8 Hz, 2H), 9.14 (s, CH), 11.63 (s, OH); 13C
NMR (CDCl3, 300 MHz, ppm) δ 118.4, 120.8, 122.4, 127.6, 128.3,
129.3, 129.5, 130.8, 134.9, 137.6, 158.9, 162.4, 168.2; IR (KBr,
cm-1) 3048 w, 2923 w, 2851 w, 1658 m, 1618 s, 1594 s, 1561 s,
1502 w, 1485 w, 1451 m, 1430 s, 1379 s, 1350 m, 1327 m, 1281 m,
1249 m, 1233 m, 1213 m, 1233 s, 1117 w, 1007 w, 997 w, 977 w,
904 m, 810 w, 757 s, 747 s, 684 s, 648 m, 585 s, 505 m, 478 w,
462 m; Raman 1662 s, 1622 s, 1607 m, 1599 m, 1566 s, 1506 w,
1465 m, 1450 m, 1434 w, 1395 m, 1354 w, 1330 m, 1292 w,
1251 m, 1234 s, 1217 m, 1178 s, 1155 m, 1031 m, 999 s, 979 w,
813 m, 793 s, 761 w, 712 w, 646 w, 616 w, 558 w, 475 w, 442 w,
396 w, 350 w, 291 w, 246 w, 225 w; UV/vis (CH3OH, λmax, nm)
339, 394; MS (ESI), m/z calcd for C15H11N3O3 [M] 281, found [M
+ H]+ 282 (100%), [M + Na]+ 304 (10%), exact mass calcd for
C15H11N3O2 [M - H]- 280.07280, found 280.07277.
3-Oxo-4-phenyl-5-dimethylamino-1,2,3-oxadiazole, 9. To a
solution of 5 (0.1 g, 0.565 mmol) in anhydrous DMF (1 mL) under
nitrogen at room temperature was portion-wise added 0.06 g (2.5
mmol) NaH, 60% dispersion in mineral oil in DMF (2-3 mL) via
a syringe. In addition, the mixture was stirred 30 min then cooled
to 5 °C. Methyl iodide (0.21 g, 1.5 mmol) in DMF was slowly
introduced and the reaction mixture was left overnight at ambient
temperature. The solution was poured into cold water and repeatedly
extracted with chloroform, washed with brine, and dried over
magnesium sulfate. After removal of the solvent the crude product
was purified by column chromatography on silica gel with hexane/
ethyl acetate (1:1, v/v, Rf 0.6) as an eluent; yield 0.076 g (66%,
from DCM/hexane) of 9; mp 105-107 °C; 1H NMR (CDCl3, 400
Conclusion
In basic methanol benzyl cyanide reacts rapidly with nitric
oxide to give product distributions which are dependent upon
the base, stoichiometry, temperature, and pressure. Among the
products is the unusual heterocycle 3-oxo-4-phenyl-5-amino-
1,2,3-oxadiazole, 5, which correponds to an N-oxide resulting
from the cyclic addition of the NO dimer to the benzyl cyanide.
As with the other 3-oxo-1,2,3-oxadiazoles 5 is stable toward
acids and bases and is best represented as an aromatic amine.
Experimental Section
General. All reagents and solvents were used as supplied
commercially, except for methanol, which was distilled from
sodium. Nitric oxide was passed over potassium hydroxide pellets
and was used in an evacuated and purged Schlenck line before
introduction into the medium pressure reaction vessel. The apparatus
for the diazeniumdiolation reactions has been described in detail
in prior publications8 with a key detail being the use of a thick-
walled glass pressure bottle for the reactions.
Reaction of Benzyl Cyanide with Nitric Oxide. In a typical
procedure, benzyl cyanide was added to a solution of potassium
methoxide in methanol in a pressure bottle. After 4 cycles of
nitrogen:vacuum degassing, the solution was exposed to 250 kPa
nitric oxide for 4 h with stirring at the listed temperature. During
this period, as the nitric oxide is consumed, the pressure is topped
up with fresh gas until no further gas is consumed. The reaction
mixture is allowed to warm to room temperature and kept under
an atmosphere of nitric oxide overnight. During this reaction copious
amounts of a light cream colored precipitate form. After venting
the excess nitric oxide, the precipitate is filtered and washed with
methanol/dichloromethane (1:1). The resulting snow white material
is bis-diazeniumdiolated imidate, 7. The filtrate is reduced in volume
and the crude product was purified by column chromatography on
silica gel with hexane/ethyl acetate (1:1, v/v) as an eluent to give
unreacted starting benzyl cyanide, compound 5 (Rf 0.4) and 6
(Rf 0.8).
3-Oxo-4phenyl-5-amino-1,2,3-oxadiazole, 5: mp 145-147 °C;
1H NMR (DMSO, 400 MHz, ppm) δ 7.37 (t, J ) 7 Hz, 1H), 7.46
(t, J ) 8 Hz, 2H), 7.64 (d, J ) 8 Hz, 2H), 8.05 (s, NH2); 13C NMR
(CDCl3, 300 MHz, ppm) δ 105.8, 124.1, 127.5, 128.4, 128.5, 163.6;
IR (KBr, cm-1):3447 m, 3298 m, 3249 m, 3113 m, 1660 s, 1507 s,
1468 w, 1450 w, 1375 s, 1322 m, 1310 m, 1282 m, 1247 s, 1160 w,
1064 w, 1013 m, 957 s, 911 w, 844 w, 801 w, 765 s, 709 w, 693 s,
65 m, 643 w, 625 w, 613 w, 506 m; UV/vis (CH3OH) λmax, nm (ε
(mol/L cm)-1) 233 (12750), 256 (10100), 290 (9646); (NaOH 2M)
(12) Wieland, H. Ann. Chem. 1903, 328, 154–156.
(13) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson,
J. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Kamaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez,
C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.;
Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian98;
Gaussian Inc.: Pittsburgh, PA, 1998.
λ
max, nm (ε (mol/L cm)-1) 279 (8300), 334 (11400). Anal. Calcd
for C8H7N3O2: C, 54.23; H, 3.98; N, 23.71. Found: C, 54.14, H,
3.74, N 23.50. Deuterium derivative of 5: Protio 5 dissolved in a
1:1 mixture of d6-acetone/D2O stirred for an hour and the solvents
(11) Blocher, A.; Zeller, K. P. Angew. Chem. 1991, 103, 1489–1490. See
also Angew. Chem., Int. Ed. Engl. 1991, 1430, 1476-1487.
J. Org. Chem. Vol. 74, No. 4, 2009 1625

Bohle and Perepichka
MHz, ppm) δ 2.94 (s, 6H), 7.44 (s, br, 5H); 13C NMR (CDCl3,
300 MHz, ppm) δ 38.6, 107.9, 124.4, 128.8, 129.8, 131.5, 161.9;
IR (KBr, cm-1) 3072 w, 3032 w, 2932 w, 2821 w, 1665 s, 1604 m,
1524 m, 1480 w, 1446 w, 1379 s, 1289 m, 1270 m, 1156 w, 1070 m,
1015 m, 944 m, 912 m, 789 w, 760 m, 729 m, 701 m, 692 m,
649 m, 580 w, 496 w, 467 w. UV/vis (DCM) λmax, nm (ε (mol/L
cm)-1) 240 (9800), 287 (11100); MS (electron impact), m/z exact
mass calcd for C10H11N3O2 [M] 205.0851, found 205.0855.
Computational Details. All of the calculations described above
were performed with Gaussian 98.13 We have studied the neutral
amine 5 as both A and B tautomers as well the possible acetylation
isomers of Go¨tz C-E. Computations were carried out at the
restricted Hartree-Fock (RHF),14 and Density Functional Theory
(DFT). DFT calculations used the hybrid B3LYP functional and
triple-ꢀ 6-311++G** basis sets.14-16 Vibrational frequencies were
calculated for all stationary points to define them as either minima
or transition states.
full-matrix least-squares on F2 of all data, using SHELXTL
software.17 Key crystallographic data is summarized in Table S1in
the Supporting Information.
Acknowledgment. We gratefully acknowledge the CRC, CFI,
and NSERC for support in the forms of a CRC, CFI, and
discovery grant to DSB, and the CIHR for a Chemical Biology
Fellowship to Dr. Inna Perepichka.
Supporting Information Available: General experimental
instrumentation, crystallographic data (CIF files, Tables S1-S5),
and density functional results for 5 and its tautomers. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO802343K
X-ray Crystallography: Single-crystal X-ray diffraction experi-
ments were carried out with a BRUKER SMART CCD diffracto-
meter using graphite-monochromated Mo-KR radiation (λ ) 0.71073
Å). The structures were solved by direct methods and refined by
(14) Roothaan, C. C. J. ReV. Mod. Phys. 1951, 23, 69–89.
(15) Lee, C. Y. W.; Parr, R. G. Phys. ReV. B 1988, 37, 785–789.
(16) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(17) SHELXTL, SHELXTL 5.10 ed.; Bruker AXS, Inc.: Madison, WI., 1997.
1626 J. Org. Chem. Vol. 74, No. 4, 2009