1072
Can. J. Chem. Vol. 85, 2007
1
action sequence, 4,6-dichloro-[1,3,5]triazin-2-yl group has to
be the leaving group, because the acidity of 4,6-dichloro-
[1,3,5]triazin-2-ol 3b is higher (pK = 3.2 (1d)) than that of
aliphatic carbonic acids or benzoic acid. As a result, target
amides are formed, and compound 3b is washed off by basic
solutions. Cyanuric acid (3a), with pK value about 4.7, is
acidic enough to be the leaving group. We would prefer to
consider the formation of amides promoted by cyanuric
chloride as a case of activated ester method (see Scheme 2),
which has a wide application in organic (mostly peptide)
synthesis (5). The method involves the conversion of acids
to esters with hydroxy compounds whose anions are good
leaving groups (4-nitrophenol, 1-hydroxybenzitriazole, and
many others). The esters then react with amines or amino
acids to furnish amides.
To embark on experiments with acetic acid as solvent, we
first combined aniline, cyanuric chloride, and Et3N in the
molar ratio 1:1:1 (1 equiv. each). Triethylamine, a strong
base, was neutralized with acetic acid, and the mixture was
cooled to ambient temperature before being added to the so-
lution of cyanuric chloride. Aniline was also neutralized
with acetic acid beforehand (see experimental details later).
We tried different sequences of the reagent addition. When
aniline (1 equiv.) was added immediately or even 3 h after
Et3N (1 equiv.), aryl amination proved to be the predominant
reaction. Only trace amounts of acetanilide 4 were detected.
Further experiments showed that the course of the reaction
could be changed when no less than 3 eqiuv. of Et3N was
added to the reaction mixture. To completely eliminate the
aryl amination, the acetic acid solution of cyanuric chloride
and Et3N must be left for a long time (10–12 h) before the
addition of aniline. Aniline was taken in excess (~4 equiv.)
to provide maximum yield. Extraction of the reaction mix-
ture with benzene and subsequent treatment with dilute hy-
drochloric acid and alkali gave a chromatographically pure
sample of 4 (the purity and identity was checked by means
of 1H NMR and HPLC). The isolated yield was 77%, assum-
ing all three chlorine atoms of 1 reacted. The yield of 4 is
slightly higher (81%) when Et3N is replaced by sodium ace-
tate and is considerably lower (56%) when pyridine is used
instead.
etry and H NMR. No trace of acetyl chloride was detected.
We may assume that the formation of insoluble cyanuric
acid 3a could be an additional driving force of the
solvolysis, which leads to the formation of acetic anhydride.
Finally, we found it possible to perform acylation of aniline
with acetic acid as a “green reaction” simply by heating ani-
line in a large volume (~ 80 mol excess) of acetic acid under
reflux (~1 h) in the absence of catalysts or dehydrating
agents, as recommended in (3a).
Meanwhile, no trace of 4 was detected when 3 equiv. of
all three reagents were mixed simultaneously. Thus, when
aniline (3 equiv.) was added to the reaction mixture immedi-
ately after Et3N (3 equiv.), aryl amination prevailed and a
precipitate was formed (see Experimental section). The
major product (92%, according to HPLC) was 6-chloro-
N,N′-diphenyl-[1,3,5]triazine-2,4-diamine (Scheme 1, di-
substituted product 2a). Up to 66% of this compound was
isolated by simple filtration of the reaction mixture after 6 h
exposure at ambient temperature. Et3N can be replaced by
sodium acetate or pyridine with the same result. An excel-
lent yield (96%, isolated) of di-substituted product was
achieved when water was added to the reaction mixture (to-
gether with the reagents), which precipitated the reaction
product. To obtain pure mono-substituted product 2b in the
reaction with 1 equiv. of aniline, the temperature should be
maintained at about 15 °C and water should also be added.
Otherwise, substantial amounts of di-substituted product 2a
are formed.
All these experiments demonstrate that acetate ion, as a
weaker nucleophile, fails to compete with aniline in the re-
action with cyanuric chloride in acetic acid solutions. It was
established in the late 1940s that salts and acids are dissoci-
ated to a very little extent (6) in acetic acid medium, in con-
trast to water solutions. Thus, acetates are barely dissociated
in anhydrous acetic acid, and consequently the effective con-
centration (activity) of the nucleophile is very low. Note that
neat acetic acid exists in the form of dimers with strong
intramolecular hydrogen bonds and, as acid, dissociates
weakly (6).
Conclusions
The analysis of an acetic acid solution containing cyanuric
chloride and Et3N (in the form of its acetate) revealed some
interesting facts. First, a precipitate of cyanuric acid 3a was
formed. It was identified by mass spectroscopy and elemen-
tal analysis; the isolated yield was about 72%. Further,
triethylamine hydrochloride was isolated with 92% yield
from the liquid phase upon diluting with hexane. The forma-
tion of this salt confirms the proposed reaction scenario
(Scheme 2). Regarding the reactivity of cyanuric chloride in
general, our results are consistent with some of the previ-
ously reported data, showing that selective substitution of
one chlorine atom in cyanuric chloride is often problematic
(1d). To explain the formation of cyanuric acid (3a), one
should keep in mind that acetic acid is a protonic solvent.
Such solvents must cause solvolysis of the “active ester” 5.
We observed a rapid formation of 3a when water was added
to the solutions of 1 and Et3N in acetone or acetic acid. In
dry acetic acid medium, the only possible solvolysis product
was acetic anhydride (Scheme 2, 6), whose presence in the
reaction mixture was confirmed by means of mass spectrom-
Two reactions of aniline in acetic acid solutions contain-
ing cyanuric chloride and hydrogen chloride acceptors have
been established, not considering the acid–base interaction
between aniline and acetic acid. Either transformation, aryl
amination or acylation, can be performed smoothly and with
high yields under proper reaction conditions. The results
show that acetic acid can be used as solvent for aryl
amination of cyaniric chloride 1 and most probably other
heterocyclic halides with no risk of acylation. This is due to
the lower nucleophilicity of the acetate ion and a weak dis-
sociation of acetates in anhydrous acetic acid medium. The
presence of HCl acceptors (triethylamine, pyridine, sodium
acetate) does not influence the aryl amination, which pre-
dominates in acetic acid solution. Mono- or di-phenylamino-
1,3,5-triazine derivatives (2a and 2b) were obtained in good
and excellent yields when acetic acid solutions of cyanuric
chloride, aromatic amines (1 or 2 equiv.), and an acetate (1
or 2 equiv., respectively) were combined simultaneously at
ambient temperature. On the other hand, a clean procedure
© 2007 NRC Canada