2
G. W. Breton, M. Turlington / Tetrahedron Letters xxx (2014) xxx–xxx
Scheme 1. Cookson’s method for the synthesis of urazole 5.
reagent and its unique properties relative to other TADs, we sought
a gram-scale route for the synthesis of its immediate precursor,
urazole 5. Herein, we report two convenient pathways for the syn-
thesis of 5. One of these methods proved to be amenable to the
synthesis of other urazole-precursor derivatives as well.
Figure 1. Compound 7 and its possible precursors, 8 and 9.
Results and discussion
Synthesis of 5 via in situ production of methyl isocyanate from
the decomposition of 1,3-dimethyl-1-nitrosourea
Synthesis of 5 via ethyl phenyl hydrazine-1,2-dicarboxylate
As semicarbazides 10 can be readily cyclized to the correspond-
ing urazoles using Cookson’s method (Scheme 3), we were also
interested in developing a general method for their gram scale
synthesis that provides an alternative to the use of isocyanates
and does not depend on the commercial availability of 1,3-dial-
kylureas. To this end, we envisioned that ethyl phenyl hydrazine-
1,2-dicarboxylate 11 could serve as a common synthetic precursor
of semicarbazides 10 via reaction with various amines (see
Scheme 4). This strategy was supported by several reports demon-
strating the efficiency of the phenoxide leaving group in the addi-
tion of nitrogen nucleophiles to aryl carbamates under mild
conditions.12b,16
Dicarboxylate 11 was readily prepared via reaction of carbazate
2 with phenyl chloroformate (1 equiv) in the presence of DIPEA
(Hunig’s base) in CH2Cl2 (Scheme 4).17 Acidic extraction success-
fully removed the amine and treatment of crude 11 in CH3CN with
2.2 equiv of a commercially available solution of methylamine in
THF resulted in clean displacement of the phenol leaving group
and formation of semicarbazide 4. The phenol byproduct could
be conveniently separated from 4 via selective extraction of phenol
from an aqueous solution of the crude reaction mixture (thereby
avoiding the necessity of column chromatographic purification)
to afford pure 4 in 84% yield over two steps (Scheme 4, R = CH3).
Cyclization of the semicarbazide using NaOEt/EtOH as before
provided urazole 5.
In order to examine the generality of this method the reaction
of 11 with other saturated amines in CH3CN was explored. Dicar-
boxylate 11 can be prepared on a large scale (10 g) and is a bench
stable solid. Treatment of 11 with 2.2 equiv of alkyl amines
allowed for the synthesis of N-substituted semicarbazides 10a–d
(Table 1) in good yields, providing a method for semicarbazide for-
mation from a common precursor while avoiding the use of isocy-
anates that have traditionally been used. Unfortunately, aniline
proved to be unreactive with 11 at room temperature or even at
reflux temperatures, presumably because of its lower nucleophilic-
ity relative to the alkyl amines. Addition of DIPEA as a base catalyst
resulted in consumption of 11 but the desired semicarbazide 10e
was not observed. The synthesized semicarbazides 10a–d can be
Our first approach towards the synthesis of 5 mimicked the
general synthetic route established by Cookson (Scheme 1).10 Since
methyl isocyanate is no longer commercially available, we sought a
suitable substitute. 1,3-dimethyl-1-nitrosourea (6), readily synthe-
sized in high yield by aqueous-phase nitrosylation of 1,3-dimethyl-
urea, is known to decompose in aqueous solution at elevated
temperatures to form methyl isocyanate (3) and N-nitrosomethyl
amine (Scheme 2).13a,b The nitrosoamine byproduct spontaneously
decomposes further to generate diazomethane which rapidly
reacts with water to be converted into methanol. The generated
methyl isocyanate (3), however, has been shown to be trappable
in situ by added amines to form 1-methyl-3-alkyl urea, as well
as by hydrazine to form 4-methylsemicarbazide (Scheme 2),
thereby conveniently avoiding the need to isolate the toxic methyl
isocyanate.13a,c We suspected that the methyl isocyanate might
also be trappable by commercially-available ethyl carbazate (2).
Heating an aqueous solution of 1,3-dimethyl-1-nitrosourea and
one equivalent of ethyl carbazate indeed resulted in efficient trap-
ping of the liberated methyl isocyanate by the carbazate to form
semicarbazide 4. A small amount (ꢀ6%) of an impurity was
detected in the crude reaction mixture that proved impossible to
separate from 4. However, when we subjected the reaction mixture
to Cookson’s cyclization conditions (NaOEt/EtOH) to form the
desired urazole 5,14 the impurity apparently similarly cyclized to
afford 1,4-dimethylurazole 7 (identified by its 1H NMR spectrum)15
suggesting that the initial impurity had the structure of either 8 or
9 (Fig. 1). It is conceivable that a small amount of 4 is methylated
by the diazomethane byproduct to ultimately give rise to the
observed 8 or 9. Fortunately, however, 7 could easily be separated
from 5 via a simple filtration process by taking advantage of its
greater solubility in CH2Cl2 relative to the nearly insoluble 5,
providing pure urazole 5 in 73% yield. This method, therefore, pro-
vides for the synthesis of 5 in multigram quantities with excellent
overall yield while avoiding the necessity of purification of the
product or any intermediates by column chromatography.
Scheme 3. Cookson’s method for urazole synthesis via cyclization of semicarbaz-
ides (10).
Scheme 2. Decomposition of 1,3-dimethyl-1-nitrosourea in aqueous solution.