2
A. Leggio et al. / Tetrahedron Letters xxx (2014) xxx–xxx
formation of side-products in the acylation of primary amines with
carboxylic acid chlorides operating under heterogeneous phase
conditions.31 Nevertheless, in many cases the conversion of start-
ing materials requires long reaction times, limiting the efficiency
of the process.
Benzoylation is a very important task in organic synthesis for
the protection of amino groups. Thus, we studied whether the ami-
nolysis of benzoyl chloride can be performed using different pri-
mary and secondary amines. According to the optimized
conditions, benzoyl chloride (one equivalent) dissolved in diethyl
ether was reacted with an equimolar amount of amine, in the pres-
ence of a mixture of sodium acetate (two equivalents) and silver
acetate (one equivalent) at room temperature (Fig. 1).
We argued that the use of a weak base containing Lewis species
should be striking in assisting the acylation of amines.
Thus, the effectiveness of silver acetate in the reaction of pri-
mary and secondary amines with available or easily affordable acyl
chlorides was investigated. The silver cation enhances the reactiv-
ity of the carbonyl group in acyl chlorides, especially in electro-
philic aromatic substitutions.32,33 Moreover, this metal species
generates silver chloride which precipitates in the most currently
used solvents, allowing the complete and rapid removal of silver
particles at the end of the reaction. Finally, the acetate anion shows
a basic strength similar to those of many tertiary amines used as
bases and it can efficiently be quenched by forming acetic acid in
the reaction environment.
The data summarized in Table 1 indicate that the protocol was
found to be highly effective in the direct preparation of structurally
different amides under very mild conditions. As expected, benzoyl
chloride reacted with different amines to afford the respective
amides 1–5 in good to excellent yields (Table 1, entries a–e). Lower
yields were observed with N-ethyl-N-isopropyl amine and piperi-
dine (Table 1, entries b and e). In these cases the corresponding
amides were obtained in 82% and 81% yields, respectively, most
likely due to the steric requirements imposed by the aliphatic cycle
of piperidine and the ramified chain of the secondary amine. Other
acylating agents different from benzoyl chloride were also investi-
gated. The reaction of N,N-diethylamine with the commercial chlo-
rides of phenylacetic, palmitic, 3-phenylbenzoic, cinnamic and
pyrazine-2-carboxylic acids (Table 1, entries f–j) afforded the
respective amides 6–10 in good yields confirming the synthetic
potential of the protocol. TLC and GC–MS analysis of crude mix-
tures showed the formation of only one product. It is important
to mention that all final products were easily isolated by a simple
hydrolytic work-up of the respective reaction mixtures and
obtained without need for chromatography. 1H and 13C NMR spec-
troscopy furnished data which were consistent with the structures
assigned to the expected compounds.
The feasibility of a silver acetate-assisted aminolysis of acyl
chlorides (Fig. 1) was preliminarily exploited by reacting benzoyl
chloride with N,N-diethylamine. A set of experiments was con-
ducted for modeling the conditions of the treatment. The aminoly-
sis was performed in diethyl ether at room temperature using
almost equimolar amounts of the acyl chloride and secondary
amine. In this first experiment, a diethyl ether solution of benzoyl
chloride (one equivalent) was added to a stoichiometric amount of
N,N-diethylamine dissolved in the same ethereal medium. After
adding silver acetate (three equivalents), reaction went to comple-
tion in 15 min. The formation of a black precipitate was indicative
of the reduction of silver ions caused by exposure to light, whereas
N,N-diethylbenzamide (1) was recovered after hydrolytic work-up
of the reaction mixture in however unsatisfying yield (37%). We
thus repeated the experiment using glassware protected from
exposure to light and maintaining the same reagent stoichiometry
used in the previous test. After 15 min, TLC showed the complete
consumption of benzoyl chloride. The formation of a white precip-
itate indicated that silver cation reduction did not occur (no traces
of black solid were observed) and the expected amide was isolated
in 87% yield. Further optimization studies were carried out using
cheaper sodium acetate (two equivalents) together with silver ace-
tate (one equivalent), instead of three equivalents of the more
expensive silver salt. This modification did not affect the efficiency
of the method, and reaction times and yields remained practically
unaltered. We also experienced that the silver-assisted aminolysis
of benzoyl chloride was strictly depending on the sequence
adopted for reagent adding. In fact, when a mixture of the acylating
agent and silver acetate was stirred for five minutes at room tem-
perature before adding amine, a mixture of N,N-diethylbenzamide
(42%) and N,N-diethylacetamide (53%) was recovered. The interac-
tion between the acetate anion and benzoyl chloride can explain
the reaction outcome. A mixed anhydride was generated and this
species further reacted with the amine affording the two observed
acylated compounds. Oppositely, the benzoyl derivative was exclu-
sively isolated when the amine and silver acetate were added to an
ethereal solution of the chloride, according to the detailed proce-
dure here reported.34
The flexibility of the method and its possible application to
other classes of acylating agents were verified by subjecting a ser-
ies of N-(4-nitrobenzenesulfonyl)-a-amino acid chlorides (Table 2),
prepared as previously reported,35 to the treatment with N,N-
diethyl amine (Fig. 2). All acyl chlorides reacted smoothly with
the secondary amine under the conditions adopted for the cases
reported in Table 1. As desired, the treatment gave the correspond-
ing amides 11–17 which were isolated in good to excellent yields
by the simple hydrolytic work-up of the respective reaction mix-
tures and without need for chromatography. Also in these cases
all products were pure enough for the spectroscopic, TLC, and
GC–MS analysis, and the structures of products were assigned by
1H and 13C NMR.
The reaction of the enantiomeric pair of N-protected amino acid
chlorides prepared starting from L-Phe and D-Phe allowed us to
investigate the effects of the experimental conditions of our
protocol on the chiral configuration of the starting materials. The
two
a-amino acid derivatives reacted with the enantiomerically
pure (S)-1-methylbenzylamine to afford the corresponding diaste-
reomeric amides 16 and 17 which were recovered in good yields
(82% and 86%, respectively) without need for chromatography.
The crude products 16 and 17 were subjected to LC–MS analysis
and NMR spectroscopy. LC–MS runs showed similar retention
times for 16 and 17, whilst 1H NMR spectra demonstrated that
each amide was obtained as a single diastereomer. In fact, the
respective spectroscopic analysis did not show residual signals
attributable to other diastereomers within the sensitivity limits
of the spectroscopic technique, proving that 16 and 17 were
formed without racemization.
N-(4-Nitrobenzenesulfonyl)-a-amino acid chlorides are largely
employed in the synthesis of peptides.35 Therefore, we thought
that the silver acetate-assisted protocol should also be helpful in
preparing short peptides. Exemplificative of the preparation of N-
(4-nitrobenzenesulfonyl)-dipeptides, the coupling of N-(4-nitro-
benzenesulfonyl)-alanine chloride with isoleucine methyl ester
hydrochloride (Fig. 2, R5 = H) was investigated. The corresponding
O
O
R2 R3
N
CH3COOAg
R3
R1
N
R2
+
R1 Cl
H
CH3COONa
Et2O, r.t., 15 min
1-10
Figure 1. The silver acetate-assisted aminolysis of acyl chlorides.