Amidations Using N,N′-Carbonyldiimidazole: Remarkable Rate Enhancement by Carbon Dioxide

Article abstract of DOI:10.1021/jo049949k

Carbon dioxide catalyzes the reaction of imidazolides with amines to form amides. A substantial rate enhancement is observed in the presence of CO ₂ compared to the CO₂-free case. The scope and limitations of this reaction are discussed.

Full text of DOI:10.1021/jo049949k

Am id a tion s Usin g
SCHEME 1
N,N ′-Ca r bon yld iim id a zole: Rem a r k a ble
Ra te En h a n cem en t by Ca r bon Dioxid e
Rajappa Vaidyanathan,* Vikram G. Kalthod,
Duc P. Ngo, J erad M. Manley, and Sean P. Lapekas
Chemical Research and Development, Pfizer Inc.,
0
200-91-201, 7000 Portage Road,
Kalamazoo, Michigan 49001
rajappa.vaidyanathan@pfizer.com
Received J anuary 7, 2004
SCHEME 2
Abstr a ct: Carbon dioxide catalyzes the reaction of imida-
zolides with amines to form amides. A substantial rate
enhancement is observed in the presence of CO2 compared
to the CO2-free case. The scope and limitations of this
reaction are discussed.
N,N ′-Carbonyldiimidazole (CDI, 1) is a widely used
reagent to form amides from carboxylic acids and amines
(
Scheme 1).1 Typically, this is a one-pot procedure
wherein a mixture of the acid and CDI is allowed to stir
until the activation reaction to form the intermediate
imidazolide 2 is complete. The amine is then added to
the reaction mixture to lead to the desired amide. The
method has been applied to a variety of aliphatic,
aromatic, and heterocyclic carboxylic acids. The inter-
mediate imidazolides 2 obtained upon reaction of car-
boxylic acids with CDI possess reactivity comparable to
acid chlorides; however, they are more easily handled,
and may be isolated if necessary. Moreover, imidazole,
the byproduct obtained when the imidazolide reacts with
the amine, is easily removed from the reaction mixture
by an acidic wash. These advantages, coupled with the
relatively low cost of CDI, render this method an attrac-
tive alternative to the carbodiimide-based reagents such
as DCC and EDC.
sion in ca. 4 h and was typically complete in less than
1
2 h. However, when attempts were made to run this
chemistry in the pilot plant on a multi-kilogram scale,
the amidation reaction was substantially slower, requir-
ing almost 50 h to reach completion.
One potential explanation was that the CO released
2
in the imidazolide formation step was not removed
completely from the reactor in the pilot plant prior to
2
addition of diamine 6 and that this lingering CO reacted
with the amine to form the carbamate salt, thus slowing
down the amidation reaction. To verify this hypothesis,
two experiments were set up in the laboratory, starting
with isolated imidazolide 5. In one case, CO
into a slurry of imidazolide 5 and imidazole in THF for
5 min prior to the addition of the amine, and the other
was run under CO -free conditions. The reaction with
CO reached 88% conversion in 4 h, while the CO -free
2
was bubbled
The synthesis of 3, an exciting split kinase inhibitor
2
under clinical development, is depicted in Scheme 2. The
1
key acid aldehyde 4 is activated with CDI to form the
intermediate imidazolide 5 in situ. Addition of N,N-
diethylethylenediamine (6) to the reaction mixture leads
to the imine-amide intermediate, 7. Imine formation
competes with amide formation, necessitating the use of
excess amine; however, the imine itself does not pose any
problems in the subsequent coupling step. Coupling of 7
with 5-fluorooxindole 8 leads to the final product, 3.3
During our exploratory studies on this route in the
laboratory, the amidation reaction reached 90% conver-
2
2
2
reaction was substantially slower and was only 37%
4
complete in the same time. These experiments clearly
(
3) (a) J in, Q.; Mauragis, M. A.; May, P. D. PCT Int. Appl. WO
2
003070725, 2003. (b) Guan, H.; Liang, C.; Sun, L.; Tang, P. C.; Wei,
C. C.; Mauragis, M. A.; Vojkovsky, T.; J in, Q.; Herrinton, P. M. PCT
Int. Appl. WO 2002066463, 2002. (c) Mauragis, M. A.; J in, A.; Fleck,
T. J . Manuscript in preparation. (d) Manley, J . M.; Kalman, M. J .;
Conway, B. G.; Ball, C. C.; Havens, J . L.; Vaidyanathan, R. J . Org.
Chem. 2003, 68, 6447.
(
4) The reactions were monitored by HPLC. Imines 7 and 9 were
(
1) Armstrong, A. In Encyclopedia of Reagents for Organic Synthesis;
Paquette, L. A., Ed.; J ohn Wiley & Sons: Chichester, 1995; Vol. 2, pp
006 and references therein.
2) (a) Tang, P. C.; Miller, T.; Li, X.; Sun, L.; Wei, C. C.; Shirazian,
quantitatively hydrolyzed to the corresponding aldehydes, 26 and 27
respectively under the HPLC conditions.
1
(
S.; Liang, C.; Vojkovsky, T.; Nematalla, A. S. WO 01/60814. (b) Mendel,
D. B.; Laird, A. D.; Xin, X.; Louie, S. G.; Christensen, J . G.; Li, G.;
Schreck, R. E.; Abrams, T. J .; Ngai, T. J .; Lee, L. B.; Murray, L. J .;
Carver, J .; Chan, E.; Moss, K. G.; Haznedar, J . O.; Sukbuntherng, J .;
Blake, R. A.; Sun, L.; Tang, C.; Miller, T.; Shirazian, S.; McMahon,
G.; Cherrington, J . M. Clin. Cancer Res. 2003, 9, 327.
1
0.1021/jo049949k CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/03/2004
J . Org. Chem. 2004, 69, 2565-2568
2565

TABLE 1. Ra tes of Am id a tion of Im id a zolid es w ith a n d w ith ou t CO2
a
b
t1/2 is the time required for the amidation reaction to reach 50% conversion by HPLC. The product imine-amides were hydrolyzed
c
d
to the corresponding aldehyde-amides under the HPLC conditions. The reaction was 48% complete in 330 min. The reaction was 11%
complete in 510 min. The reaction was 43% complete in 275 min. f The reaction was 100% complete in 30 min. g The reaction was 97%
e
h
complete in 1 min. The reaction was 34% complete in 1 min and 92% complete in 10 min.
demonstrate that CO
deleterious effect on the reaction rate! Based on these
results, a conscious attempt was made to retain the CO
2
has a beneficial rather than
2
campaign led to removal of CO from the reactor and thus
slowed the amidation reaction.
In separate experiments, CO was bubbled through
2
2
liberated in the first step in solution during the next pilot
plant campaign. As expected, the amidation reaction
reached ca. 90% conversion in 4 h and was complete in
less than 12 h without any complications. In retrospect,
better agitation and venting in the pilot plant in our first
solutions of the amine, imidazolide, and the solvent prior
to the reaction. Another reaction was carried out at 50
psig CO . Interestingly, all these reactions proceeded at
2
the same rate as the normal reaction (one-pot conversion
of the acid to the amide via the imidazolide without
2
566 J . Org. Chem., Vol. 69, No. 7, 2004

venting the CO
amount of CO
2
). This suggests that only a critical
in solution is needed to catalyze the
2
reaction; more does not necessarily help.
Several substrates were examined in an attempt to
explore the scope of this CO catalysis, and the results
2
are summarized in Table 1. We chose to use t1/2 (time
required for the amidation reaction to reach 50% comple-
tion) as a convenient point of comparison of the rates of
the catalyzed and uncatalyzed reactions (see the Experi-
mental Section for a typical procedure).
As seen in Table 1, the CO
substantially faster than the corresponding CO
reactions. Significantly, in the case of the reaction of the
imidazolide of acid 10 with benzylamine, the CO -free
2
-catalyzed reactions were
2
-free
2
reaction was only 11% complete after 8.5 h, while the
catalyzed reaction reached 50% completion in less than
3
h (entry 4). Interestingly, the CO
in entries 5-7 reached completion within 30-60 min,
while the CO -free reactions took several hours to reach
2
-catalyzed reactions
2
F IGURE 1. Rates of amidation with and without CO .
2
even 50% conversion. In the case of the reaction of
benzoyl imidazole with benzylamine, the catalyzed reac-
tion was virtually complete within 1 min, while the
uncatalyzed reached completion in ca. 10 min (entry 8).
When either the amine or the acid was highly sterically
encumbered, the reaction did not proceed at all (entries
SCHEME 3
9
and 10). The inference drawn from these experiments
is that CO catalyzes amidation of imidazolides; this
2
effect is perceptible and pronounced when either the
imidazolide or the amine is less reactive due to steric or
electronic reasons (entries 1-7). In unhindered sub-
strates, the effect still exists; but since the uncatalyzed
course of their investigation on the reaction of amines
with dimethyl carbonate (DMC) to form methyl carbam-
ates, Aresta and co-workers established (through labeling
2
reaction is inherently fast, the catalytic effect of CO is
barely noticeable. In cases where the sterics are over-
bearing (2,4,6-trimethylbenzoic acid or tert-butylamine),
the reaction does not proceed at all.
studies) that the carbonyl carbon in the product was from
8
DMC and not CO
2
. Their mechanistic studies suggested
the intermediacy of carbamic-carbonic anhydride 24
(Scheme 3).
The next question was: “can one ‘jump-start’ a slow,
CO -free reaction by bubbling in CO mid-way through
2 2
Based on our observations and literature precedent,
the following mechanism may be postulated to explain
the catalytic effect of CO on amidation reactions. Reac-
tion of the amine with CO would lead to the alkylam-
monium N-alkyl carbamate, 23. Nucleophilic attack of
the oxygen center of carbamate 23 on the imidazolide
would give intermediate 25 (analogous to 24), which upon
the reaction?” The reaction of imidazolide 22 (derived
from acid 10) with benzylamine was chosen as the test
reaction since the rate difference between the catalyzed
and uncatalyzed reactions was most pronounced in this
case (Table 1, entry 4). In the experiment, a mixture of
2
2
2
2, imidazole and benzylamine was allowed to stir at 45
2
extrusion of CO would lead to the amide (Scheme 4,
°
C. After 1 h (ca. 1% conversion to the amide), CO was
2
mechanism a). Alternately, the nitrogen center of tau-
tomer 28 may attack the carbonyl of the imidazolide to
directly give the amide (Scheme 4, mechanism b).
In summary, we have demonstrated the catalytic effect
sparged into the reaction mixture for 15 min, and the
mixture was stirred at 45 °C. As expected, the reaction
rate increased dramatically, and was nearly identical to
that of the CO
conclusively establishes the fact that CO
2
catalyzed reaction (Figure 1). This
catalyzes the
of CO
2
in amidation reactions using CDI. This work
released
2
reaction and may be used to “jump-start” slow amidation
reactions.
clearly establishes the need to retain the CO
2
in the imidazolide formation step in solution during the
subsequent reaction of the imidazolide with the amine.
While the mechanistic aspects of this reaction may
require further elucidation, its utility in organic synthesis
is clear.
In the case of the amidation reactions listed in Table
1
2
, a precipitate was observed when CO was bubbled
through a solution of the amine in THF, suggesting the
formation of the alkylammonium N-alkyl carbamate, 23.5
The ambident nucleophilic nature of these carbamates
is well-documented. They have been shown to react with
(7) (a) Aresta, M.; Quaranta, E. Tetrahedron 1992, 48, 1515. (b)
Aresta, M.; Quaranta, E. J . Org. Chem. 1988, 53, 4154. (c) Aresta, M.;
Quaranta, E. J . Chem. Soc., Dalton Trans. 1992, 1893. (d) McGhee,
W. D.; Pan, Y.; Riley, D. P. J . Chem. Soc., Chem. Commun. 1994, 699.
6
alkyl halides to give both N-alkylation and O-alkylation
7
products, depending on the reaction conditions. In the
(e) McGhee, W.; Riley, D.; Christ, K.; Pan, Y.; Parnas, B. J . Org. Chem.
1
995, 60, 2820. (f) Salvatore, R. N.; Chu, F.; Nagle, A. S.; Kapxhiu, E.;
(
(
5) Hampe, E. M.; Rudkevich, D. M. Chem. Commun. 2002, 1450.
6) Yoshida, Y.; Ishii, S.; Watanabe, M.; Yamashita, T. Bull. Chem.
Cross, R. M.; J ung, K. W. Tetrahedron 2002, 58, 3329.
(8) (a) Aresta, M.; Quaranta, E. Tetrahedron 1991, 47, 9489. (b)
Aresta, M.; Dibenedetto, A. Chem. Eur. J . 2002, 8, 685.
Soc. J pn. 1989, 62, 1534.
J . Org. Chem, Vol. 69, No. 7, 2004 2567

SCHEME 4
vacuo to remove all CO
and imidazole was diluted with 10 mL of THF. In a separate
flask, CO was bubbled through a solution of the amine (7.8
2
. This mixture containing the imidazolide
2
mmol, 1.3 equiv) in THF (10 mL) for 15 min. This solution was
added to the solution of the imidazolide and imidazole and
stirred at 45 °C. The progress of the reaction was monitored by
HPLC. For the examples in entries 1 and 2 (Table 1), 3 equiv of
amine was added for both the catalyzed and uncatalyzed
reactions.
CO
mmol) and CDI (7.2 mmol) in THF (20 mL) was stirred at 45
C. When HPLC indicated complete conversion to the imida-
zolide, the mixture was concentrated to dryness in vacuo to
remove all CO . This mixture containing the imidazolide and
2
-F r ee Rea ction s. A mixture of the carboxylic acid (6
°
2
imidazole was diluted with 10 mL of THF. A solution of the
amine (7.8 mmol, 1.3 equiv) in 10 mL of THF was added to the
solution of the imidazolide and imidazole, and stirred at 45 °C.
The progress of the reaction was monitored by HPLC. For the
examples in entries 1 and 2 (Table 1), 3 equiv of amine was
added for both the catalyzed and uncatalyzed reactions.
Ack n ow led gm en t. We thank Drs. Peter G. M.
Wuts, Thomas A. Runge, Bruce A. Pearlman, Peter
Giannousis, Thomas J . Beauchamp, Gerald A. Weisen-
berger, J ames R. Gage, and Professor Scott Denmark
for helpful discussions. We also thank William C.
Snyder and Chadd R. Gromaski for analytical support.
Exp er im en ta l Section
Typ ica l Exp er im en ta l P r oced u r e. The following experi-
mental procedures were used for the reactions depicted in Table
1
. Authentic standards of all amides were synthesized by
treatment of the corresponding acids and amines with EDC and
HOBt, and the products were characterized by H and C NMR
and compared to literature values.
1
13
Su p p or tin g In for m a tion Ava ila ble: Characterization
data for compounds 12 and 27. This material is available free
of charge via the Internet at http://pubs.acs.org.
2
Rea ction s Ru n in th e P r esen ce of CO . A mixture of the
carboxylic acid (6 mmol) and CDI (7.2 mmol) in THF (20 mL)
was stirred at 45 °C. When HPLC indicated complete conversion
to the imidazolide, the mixture was concentrated to dryness in
J O049949K
2
568 J . Org. Chem., Vol. 69, No. 7, 2004