Activation of carboxylic acids by Burgess reagent: An efficient route to acyl ureas and amides

Article abstract of DOI:10.1016/j.tetlet.2006.01.015

Carboxylic acids upon treatment with Burgess reagent are converted to novel mixed sulfocarboxy anhydrides. Subsequent treatment of such mixed anhydrides with amines at elevated temperatures yields acyl ureas and amides. The ratio of the two products appears to be temperature controlled. The method provides a simple and convenient route to diverse acyl ureas starting from carboxylic acids and amines.

Full text of DOI:10.1016/j.tetlet.2006.01.015

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 1825–1828
Activation of carboxylic acids by Burgess reagent: an efficient
route to acyl ureas and amides
Derek Wodka, Michael Robbins, Ping Lan, Rogelio L. Martinez,
John Athanasopoulos and Gergely M. Makara*
Merck & Co., Merck Research Laboratories, Department of Target Validation, RY80Y-325, 126 E. Lincoln Ave,
Rahway, NJ 07065, USA
Received 21 November 2005; revised 28 December 2005; accepted 5 January 2006
Available online 25 January 2006
Abstract—Carboxylic acids upon treatment with Burgess reagent are converted to novel mixed sulfocarboxy anhydrides. Subse-
quent treatment of such mixed anhydrides with amines at elevated temperatures yields acyl ureas and amides. The ratio of the
two products appears to be temperature controlled. The method provides a simple and convenient route to diverse acyl ureas start-
ing from carboxylic acids and amines.
Ó 2006 Elsevier Ltd. All rights reserved.
Burgess reagent has been a versatile agent in organic
synthesis for decades. Over the years it has been
employed in diverse transformations such as dehydra-
tions of alcohols to alkenes1 and amides to nitriles2 or
synthesis of oxazolines3 and thiazolines.4 It has also
been noted that primary alcohols can be converted to
the corresponding methyl carbamates5 and that the reac-
tion of Burgess reagent with epoxides can lead to cyclic
sulfamidates.6 Sulfamidates have also been prepared
from 1,2-diols or epoxyalcohols, while sulfamides are
the products starting from aminoalcohols using the
agent.7 An immobilized version of the Burgess reagent
has been found to work with similar outcome as the
solution phase reagent in the synthesis of oxadiazoles.8
The acyl urea moiety has been an important functional
group in medicinal chemistry and has been incorporated
in marketed drugs as well as investigational candidates.
The synthesis of acyl ureas has typically been achieved
via condensation of either acylisocyanates with amines
or isocyanates with amides.9 Alternative routes starting
from ureas such as boronic acid-catalyzed condensation
with acids10 or acylation using alkenyl esters11 have also
been disclosed. Our results represent the first synthesis
of acyl ureas directly from carboxylic acids and amines.
The initial attempts to mix the Burgess reagent with
acids were carried out in tetrahydrofuran but the reac-
tion turned instantly heterogeneous. In addition, many
acids had rather poor solubility. A more polar solvent,
dimethylacetamide, was used successfully as an alterna-
tive. Unfortunately, the large solvent signal observed by
LC–MS made the careful study of the various interme-
diate species difficult. Ultimately, we decided to run
the reactions in acetonitrile for reasons as follows. In
acetonitrile there is no solvent peak in the UV, ELSD,
or mass spectrum, the solvent has a low boiling point
and good microwave absorption, and it does not gener-
ate dimethylamine, a competing nucleophile during
urea formation. Diisopropylethylamine (DIPEA) was
employed as an additive to avoid solubility issues with
the reactant acids. The Burgess reagent is quite stable
but we noticed that, after prolonged exposure to ambi-
ent conditions, an increasing excess was required for
complete acid consumption. The excess active reagent,
Surprisingly, no report has been published on reacting
Burgess-type reagents with carboxylic acids. We specu-
lated whether acids could be activated by the reagent
to give a mixed anhydride derivative, which upon treat-
ment with amines could lead to amides. Indeed, our tests
indicated that an active species is formed almost instan-
taneously, however, treatment with amines at 150 °C
yields predominantly acyl ureas while amide formation
is favored at lower temperatures.
Keywords: Burgess; Acyl urea; Amide.
*
Corresponding author. Tel.: +1 732 594 3053; fax: +1 732 594
2130; e-mail: gergely_makara@merck.com
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.01.015

1826
D. Wodka et al. / Tetrahedron Letters 47 (2006) 1825–1828
however, readily reacts with amines at room tempera-
ture to result in sulfamidates, which is in accord with re-
cent reports.7 As such, it was critical to destroy the
excess reagent prior to amine addition. This was easily
accomplished by heating the reaction mixture to 80 °C
for 15 min. No sulfamidates could be detected once
the amines were added subsequent to heating, while com-
plex 2 appeared unscathed. The mass spectrum in the
LCMS of novel intermediates 2 derived from all acids
indicated a strong signal for the parent ion in negative
Table 1. Products (5/4) obtained via synthetic route (b) in Scheme 1
Entry
Acid
Amine
Method
1
Yielda
35/15
O
NH2
NH2
a
OH
OH
b
c
d
1
1
1
10/25
50/0
30/0
O
O
O
N
H
O
O
OH
O
O
OH
NH2
O
NH2
e
f
1
1
1
1
1
1
1
1
40/5
45/35
25/15
50/0
45/15
5/20
40/5
0/0
OH
OH
H3CO
N
N
O
NH
O
g
hb
i
NH2
OH
O
NH2
OH
F
O
HO
OH
NH2
NH2
O
j
OH
O
NH2
k
l
OH
N
F3C
O
OH
NH2
NH
O
NH2
NH2
m
n
1
2
15/25
50/10
OH
NHBoc
O
OH
a Isolated yields for acyl ureas/amides (%).
b Purified by filtration of crude mixtures due to poor solubility under HPLC conditions.

D. Wodka et al. / Tetrahedron Letters 47 (2006) 1825–1828
1827
mode electrospray and a weaker di-sodiated m/z of the
parent ion in positive mode electrospray. Intermediate
2a (R1 = phenyl, Table 1) was purified by HPLC and
further characterized by 1H NMR, 13C NMR, and
HRMS. All data were consistent with the proposed
structure 2 in Scheme 1. Frequently, we also observed
another smaller UV and mass signal in the LCMS cor-
responding to the sodiated ion of the acyl urethanes 3.
The acyl urethanes of entries 2a,b were also purified
A limited set of experiments was done to optimize the
experimental conditions for acyl urea formation. Longer
reactions led to product degradation, higher and lower
temperatures resulted in decreased purity. Conventional
heating gave comparable ratios of amides and corre-
sponding acyl ureas for aryl acids but higher amide
ratios for alkyl acids (data not shown). In general micro-
wave heating appeared to deliver somewhat higher
yields. The acyl urea products of primary amines for
most (especially aryl) entries precipitate in acetonitrile
upon cooling. To achieve a generally applicable parallel
process the solvent was stripped off and the residue was
dissolved in DMSO for preparative HPLC purification.
The ratio for the isolated products was verified by analy-
sis of the LCMS UV spectra of the crude mixtures
using the purified samples as controls for normalization.
We made no attempt to further optimize the reaction
conditions or workup procedures for the individual en-
tries, although, some sensitive functional groups (such
as hydroxyl, amino, or Boc) may clearly benefit from
custom conditions.
1
and the structures were confirmed by matching the H
NMR spectrum to that in the literature.
The ratio of the acyl urea and amide products in our
experience is governed by the relative rates of conversion
of intermediate 2 to either the corresponding amide 4 or
urethane 3. Amide formation for the more reactive alkyl
carboxylic acids is much faster and cleaner than that for
aryl acids. The same effect in route (b) in Scheme 1
results in a significant amount of amide by-product for
alkyl acids as the temperature ramp-up time is sufficient
to convert a large portion of 2 to the amides. To confirm
this hypothesis, purified 3a was heated under conditions
identical to route (b). Clean conversion to acyl urea 5a
without any detectable amide formation was observed
by LCMS. In addition, the rearrangement required for
conversion of 2 to 3 was easily blocked by two ortho
methyl groups in entry 5j and thus only traces of the acyl
urea could be isolated along with the amide product.
This rearrangement is proposed to be analogous to the
one required for the formation of methyl carbamates
from primary alcohols.5 Little amide formation is
observed for unhindered aryl acids under the standard
conditions toward acyl ureas. No amide could be
detected, however with the hindered tert-butylamine
(5d), which further substantiates the mechanism put
forth herein. Amides cannot typically be synthesized in
reactions of similar hindered nucleophiles and active
esters such as p-nitrophenyl or pentafluorophenyl esters.
It was also important to verify that the route to acyl
ureas 5 does not go through amides 4 due to the amine
reacting with a decomposition product of Burgess
reagent. To that extent, we heated crude amide mixtures
obtained at 80–150 °C but no acyl ureas were detected.
The method is rather insensitive to the electronic and
steric nature of the nucleophile in preparation of the
acyl ureas (Table 1). Both electron rich and electron
deficient aryl amines and alkyl amines work equally
well. In fact, less reactive amines lead to higher acyl urea
ratios as route (c) toward amides is shut down. On the
other hand, significant steric hindrance on the acid
blocks the conversion of 2 to acyl urethanes as exempli-
fied by entry 5j. No significant conversion to amides
takes place with amines of low nucleophilicity. Both pri-
mary and secondary amines react equally well and no
significant electronic effect on the acid side was observed
in either routes (b) and (c). The failure of pyrrole-2-car-
boxylic acid (5l) signals that the known high reactivity of
the Burgess reagent with amines and alcohols does
require these functional groups present in the carboxylic
acid be protected. Boc protection was tolerated in the
microwave under both routes (b) and (c) conditions.
Route (c) leads to no acyl urea products albeit amides
for aryl acids were found to be contaminated by variable
O
2
O
O
O
O
S
O
O
b
a
R1
O
N
O
R1
OH
O
R1
N
H
O
O O
S
NH
N
N
O
1
3
6
R2R3NH
O
R2R3NH
R2R3NH
O
c
b
b
O
4
R1
N
NR2R3
R1
NR2R3
H
4
5
Scheme 1. Reaction of carboxylic acids with Burgess reagent and amines. Reagents and conditions: (a) acetonitrile (1 mL/0.1 mmol acid), DIPEA
(1.5 equiv), 5 min at rt; Burgess reagent (6, 1.4 equiv), 5 min at rt; 15 min at 80 °C (lw); (b) R2R3NH (1.5 equiv), 8 min at 150 °C (lw); (c) R2R3NH
(3 equiv), 1 h at 80 °C (lw).

1828
D. Wodka et al. / Tetrahedron Letters 47 (2006) 1825–1828
Table 2. Amides obtained via synthetic route (c) in Scheme 1
In addition, we describe a mild method for the synthesis
of amides. In the latter route, the Burgess reagent acts as
a weak coupling reagent, which is of practical use espe-
cially for highly reactive or unstable carboxylic acids.
Entry
Acid
Amine
Yielda
40
O
NH2
NH2
4a
OH
Supplementary data
OH
4b
4c
65
20
0
O
The following supplementary data is available: experi-
mental details, representative LCMS chromatograms,
NMR spectra for all new compounds, and characteriza-
tion of 2a. Supplementary data associated with this
article can be found, in the online version, at doi:10.
1016/j.tetlet.2006.01.015.
O
N
H
O
O
OH
O
NH2
4e
OH
O
H3CO
N
References and notes
NH2
4m
35
1. Lamberth, C. J. Pract. Chem. 2000, 342, 518–521.
2. Claremon, D. A.; Phillips, B. T. Tetrahedron Lett. 1988,
29, 2155–2158.
3. Miller, C. P.; Wipf, P. Tetrahedron Lett. 1992, 33, 907–
910.
OH
NHBoc
a Isolated yields for amides (%).
4. Miller, C. P.; Wipf, P. Tetrahedron Lett. 1992, 33, 6267–
6270.
5. Burgess, E. M.; Penton, H. R.; Taylor, E. A. J. Org.
Chem. 1973, 38, 26–31.
6. Rinner, U.; Adams, D. R.; dos Santos, M. L.; Abboud, K.
A.; Hudlicky, T. Synlett 2003, 9, 1247–1252.
7. Nicolaou, K. C.; Snyder, S. A.; Longbottom, D. A.;
Nalbandian, A. Z.; Huang, X. Chem. Eur. J. 2004, 10,
5581–5606.
amounts of acyl urethanes 3. While the isolated yields
for amides were modest (Table 2) purification was found
to be simple as the LCMS chromatograms were clean.
Based on our results, active species 2 appears to be less
reactive than some of the often used active esters (i.e.,
pentafluorophenyl or p-nitrophenyl) of carboxylic acids.
This low reactivity could potentially become useful
when reactive sites of multi-functionalized reagents need
to be differentiated or when extremely reactive carb-
oxylic acid substrates are used. Complex 2 derived from
alkyl acids is more reactive (reactions slowly progress at
rt) while aryl acids need prolonged heating for good
conversion.
8. Brain, C. T.; Paul, J. M.; Loong, Y.; Oakley, P. J.
Tetrahedron Lett. 1999, 40, 3275–3278.
9. Klabunde, T.; Wendt, K. U.; Kadereit, D.; Brachvogel,
V.; Burger, H.-J.; Herling, A. W.; Oikonomakos, N. G.;
Kosmopoulou, M. N.; Schmoll, D.; Sarubbi, E.; von
Roedern, E.; Schonafinger, K.; Defossa, E. J. Med. Chem.
2005, 48, 6178–6193.
In summary, a versatile method for the synthesis of acyl
ureas from acids and amines has been developed. The
protocol enables the synthesis of diverse sets of acyl
ureas from a vast commercially available reagent pool.
10. Yamamoto, H.; Ishihara, K.; Maki, T. Synlett 2004, 1355–
1358.
11. Seiller, B.; Heins, D.; Bruneau, C.; Dixneuf, P. H.
Tetrahedron 1995, 51, 10901–10912.