Direct, facile synthesis of acyl azides and nitriles from carboxylic acids using bis(2-methoxyethyl)aminosulfur trifluoride

Article abstract of DOI:10.1016/j.tetlet.2007.06.119

A mild, efficient, and practical method for the one-step synthesis of acyl azides from carboxylic acids using bis(2-methoxyethyl)aminosulfur trifluoride is described. The reaction was easily extended to the synthesis of the corresponding nitriles by the inclusion of phosphorous reagents. The method can be applied to the synthesis of optically active nitriles in high yields, and is compatible with fluorous phosphines.

Full text of DOI:10.1016/j.tetlet.2007.06.119

Tetrahedron Letters 48 (2007) 5933–5937
Direct, facile synthesis of acyl azides and nitriles from
carboxylic acids using bis(2-methoxyethyl)aminosulfur trifluoride
a,
b
a
*
Cyrous O. Kangani, Billy W. Day and David E. Kelley
a
Department of Medicine, University of Pittsburgh, Pittsburgh, PA 15213, United States
Departments of Pharmaceutical Sciences and of Chemistry, University of Pittsburgh, Pittsburgh, PA 15213, United States
b
Received 1 June 2007; revised 18 June 2007; accepted 21 June 2007
Available online 24 June 2007
Dedicated to Professor Hoshang E. Master on the occasion of his 60th birthday
Abstract—A mild, efficient, and practical method for the one-step synthesis of acyl azides from carboxylic acids using bis(2-meth-
oxyethyl)aminosulfur trifluoride is described. The reaction was easily extended to the synthesis of the corresponding nitriles by the
inclusion of phosphorous reagents. The method can be applied to the synthesis of optically active nitriles in high yields, and is com-
patible with fluorous phosphines.
Ó 2007 Elsevier Ltd. All rights reserved.
The need for simple and efficient strategies to obtain
complex molecules and their analogues is a driving force
for the development of new methodologies. Acyl azides
and nitriles are important intermediates in organic
chemistry. They are extremely useful in the preparation
In 1999, Lal et al. reported the synthesis and application
of the fluorinating reagent bis(2-methoxyethyl)amino-
sulfur trifluoride (commonly known by the trade name
Deoxo-Fluor) as an alternative to diethylaminosulfur
trifluoride (DAST). While the reactivity profiles of these
two reagents are similar, Deoxo-Fluor is more thermally
1
of amides and heterocyclic compounds. Moreover, nitr-
1
1
iles can be transformed into heterocyclic compounds of
stable than DAST and therefore easier to use.
2
significant biological importance.
We recently reported efficient one-step methods for the
synthesis of amides, benzoxazolines, oxazolines,
oxadiazoles, aldehydes, and ketones from carboxylic
Over the years, a number of methods have been devel-
oped for the synthesis of nitriles, including oxidation
3
4
12
routes from primary amines and aldehydes, and dehy-
acids using the Deoxo-Fluor reagent. These conver-
5
6
dration of amides and aldoximes. Nitriles can also be
synthesized in a one-pot manner from alcohols using a
sions were in response to our pursuit of developing
new analytical methods to locate double bonds in poly-
unsaturated fatty acids and to quantify free fatty acids
in human plasma, inspired by an early work from the
7
variety of reagents. There are, however, few one-pot di-
rect methods for the synthesis of nitriles from carboxylic
8
13
acids.
Georg lab. In continuation of our work, we report
here a highly efficient conversion of carboxylic acids,
using Deoxo-Fluor and sodium azide, to their acyl
azides in a one-pot, direct synthesis. Furthermore, the
inclusion of certain phosphorous reagents allowed con-
version in the same pot to the corresponding nitriles.
The reactions are extremely simple and powerful, and
proceed under very mild conditions.
Many single-step and multi-step methods have been
developed to convert carboxylic acids to acyl azides.
9
,10
Since carboxylic acids are more readily available com-
mercially than are acyl azides and nitriles, we sought a
more robust, mild, selective and highly efficient proce-
dure for the direct conversion of carboxylic acids to acyl
azides as well as nitriles.
In a typical reaction to form acyl azides, the carboxylic
acid (0.43 mmol, 1 equiv) and diisopropylethylamine
(
DIPEA) (0.86 mmol, 2 equiv) were dissolved in CH Cl
2 2
*
0
Corresponding author. Tel.: +1 412 647 6796; fax: +1 412 692
165; e-mail: kanganic@msx.dept-med.pitt.edu
(2 mL) in an open test tube. NaN (1.3 mmol, 3 equiv)
was then added as a 0.5 M DMSO solution (2.7 mL).
3
2
040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.06.119

5934
C. O. Kangani et al. / Tetrahedron Letters 48 (2007) 5933–5937
Table 1. One-pot synthesis of acyl azides from carboxylic acids
The mixture was cooled to 0 °C. Because of the presence
of DMSO, the reaction mixture solidifies at this temper-
ature. Deoxo-Fluor (0.65 mmol, 1.5 equiv) was added
dropwise over a 15–30 min period, converting the solid
to a solution. For aliphatic carboxylic acids, the solution
mixture was kept at 0 °C for the course of the reaction.
For aromatic carboxylic acids, the reaction mixture was
kept at 0 °C for 15 min after Deoxo-Fluor addition, and
was then allowed to warm to room temperature. The
reaction mixture was occasionally shaken on a vortex
mixer for the times listed in Table 1. The solvent was
removed in a SpeedVac (centrifugal lyophilizer) with
no heating of the sample. (We should note that the use
of sodium azide in CH Cl could possibly lead to the
O
O
iPr2NEt, CH2Cl2,
NaN3
0.5M, DMSO)
+ (
R
OH
o
Deoxo-Fluor, 0 C - rt
R
N3
a
Entry Carboxylic acid
Product (Yield, %) Reaction
time, min
O
1
1
2
3
(90) 75
(92) 60
(87) 60
OH
O
2
3
OH
2
2
Br
formation of CH (N ) and/or HN , which are explo-
2
3 2
3
O
1
4
sive; however, in the carefully designed reaction and
solvent evaporation conditions used here, no problems
were encountered. The order of addition of the reagents
is critical with sodium azide being added to carboxylic
acid only after the amine base is present to avoid the
OH
O N
2
O
formation of HN ). The residue was taken up in diethyl
4
5
4
5
(83) 75
(90) 50
3
OH
OH
ether and washed with water. After drying over MgSO4
and filtration, the solvent was evaporated either under a
stream of dry nitrogen or in a chemical fume hood
(
depending upon the volatility of the product) to give
the crude product. The product was purified by flash
silica gel column chromatography using a heptane–ether
eluent.
OH
O
O
Both aromatic (Table 1, entries 1–5) and aliphatic (en-
tries 6–12) carboxylic acids converted cleanly to azides.
All reactions resulted in high isolated yields, and conver-
sion of carboxylic acid was complete as determined by
GC–MS. 4-Bromobenzoic acid (entry 2) was efficiently
converted to the corresponding azide, while 4-nitro-
benzoic acid (entry 3) gave the corresponding acyl azide
along with 10% of the primary amine (derived from Cur-
tius rearrangement of the acyl azide). The efficacy of this
method was explored by converting ortho-substituted
carboxylic acids, such as salicylic acid to give 2-hydroxy-
benzoyl azide (entry 4). This method was found
applicable to a-substituted carboxylic acids such as
diphenylacetic acid (entry 5). As expected,12 no geomet-
ric isomerization of disubstituted alkenes (entries 9 and
6
7
6
7
(94) 30
(92) 30
OH
OH
OH
1
4
O
1
5
O
8
9
8
9
(95) 30
(85) 30
1
6
O
OH
OH
7
7
O
1
0
10
11
12
(90) 30
(96) 30
(94) 30
1
0) was observed.
7
7
Our experiences with the Deoxo-Fluor reagent have
shown that it can be exceptionally selective, depending
O
11
1
2
OH
OH
upon the reaction conditions. As triphenylphosphine
PPh ) is known to facilitate conversion of acyl azides
17
O
(
3
1
5
to nitriles, we examined whether the addition of a
phosphorous reagent to the single pot would allow
access to nitriles directly from carboxylic acids via
in situ generated acyl azides.
1
2
2
0
a
1
Yield of pure, isolated products (characterized by GC–MS, and
and C NMR).
H
1
3
Preliminary experiments to this end showed that optimal
reactions required a base that is less hindered than
DIPEA, such as triethylamine (TEA) or pyridine, to
promote the conversion to nitriles. After a detailed study
of reaction conditions with palmitic acid as the example,
we found that treatment of 1 equiv of carboxylic acid,
0 °C led to the formation of hexadecanenitrile in 90%
yield after 30 min. It should be noted that less triphenyl-
phosphine (1.2 equiv) can be used, but only if added
after conversion to the azide is complete.
3
equiv of NaN (in DMSO), and 3 equiv each of TEA
To explore the generality and scope of the one-pot
method, various carboxylic acids were examined (Table
3
and PPh in CH Cl with 1.5 equiv of Deoxo-Fluor at
3
2
2

C. O. Kangani et al. / Tetrahedron Letters 48 (2007) 5933–5937
5935
2
). Both aromatic and aliphatic carboxylic acids were
smoothly and efficiently converted to nitriles within
0–60 min under these reaction conditions. Note that
active compounds,16 we applied this method for the syn-
thesis of nitriles with chiral centers a and b to the origi-
nal carbonyl system (Table 2, entries 8 and 9). No
racemization was observed by chiral GC analysis. This
is the first report of chiral nitriles being synthesized di-
rectly from their corresponding chiral carboxylic acids.
3
the reaction was accelerated as compared to acyl azide
formation, perhaps due to TEA and/or the presence of
PPh . Benzoic acids with electron-withdrawing and
3
-
releasing groups (entries 1 and 2), and hindered carbox-
ylic acids (entries 3–5 and 9) provided high yields of nitr-
iles. Since optically active nitriles are important
synthetic intermediates for a number of biologically
The pathway for conversion of the acid to the nitrile is
envisioned to proceed as follows (Fig. 1). Deoxo-Fluor
is known to convert carboxylic acid to acyl fluoride,
1
3
whose fluoride is displaced by azide to give acyl azide.
Table 2. One-pot synthesis of nitriles from carboxylic acids
Reaction between acyl azide and PPh results in the for-
3
mation of acyl triazaphosphadiene. This cyclizes to oxa-
O
NaN3
(0.5M, DMSO)
TEA, PPh , CH Cl ,
3
2
2
phosphatriazine, which loses N in a concerted fashion
2
+
R CN
17
R
OH
o
to yield nitrile and Ph PO. Amide (and the associated
Deoxo-Fluor, 0 C - rt
3
acyl iminophosphorane) is an unlikely intermediate, be-
cause when we replaced the acid with amide and used
the same reaction conditions, the conversion to the ni-
trile was incomplete (650%), even with longer reaction
times and in the presence of excess Deoxo-Fluor (up
to 10 equiv). The pathway as given in Figure 1 does
not, however, provide an explanation for the previously
mentioned accelerated nitrile formation, the need for
excess TEA and PPh , nor the fact that Ph PS is
a
Entry Carboxylic acid
Product
(Yield, %)
Reaction
time, min
O
CN
1
13 (90) 60
OH
O
3
3
CN
also an abundant by-product of the reaction, so the
mechanism will require further examination.
2
OH
14 (92) 50
Br
Br
Other phosphorous reagents were also examined for
suitability in the conversion of palmitic and 4-bromo-
benzoic acids to their corresponding nitriles (Table 3).
OH
3
4
15 (87) 50
16 (88) 50
Ph
Ph
CN
CN
O
Although the reaction with PPh was efficient, removal
3
Ph
of its oxide and sulfide (the major non-volatile by-prod-
ucts of the reaction) was difficult and required two or
three column chromatographic purifications.
Ph
Ph
OH
O
Ph
Ph
Ph
O
Ph
We therefore tested phosphorous reagents that, along
with their oxides and sulfides, could be more easily
removed. Triethylphosphine and triethylphosphite,
which are, along with their oxides and sulfides, suffi-
ciently volatile to be removed on a SpeedVac, served this
purpose well, providing good yields of nitriles.
5
6
7
8
9
17 (90) 50
18 (85) 60
19 (90) 60
20 (90) 60
21 (85) 30
CN
OH
Ph
Ph
O
CN
Ph
Ph
OH
OH
Phosphine reagents more sterically hindered than PPh3
such as tri-tert-butylphosphine or tri-o-tolylphosphine
did not give nitriles, yet did not interfere with acyl azide
formation. Use of a phosphorus ylide, (carbethoxymeth-
ylene)triphenylphosphorane, in the reaction gave nitr-
iles, but in low yield and the reactions required a long
Ph
CN
O
Ph
O
Ph
CN
CN
Ph
CN
OH
OH
O
O
–
+
Ph
O
N Na
3
Deoxo-Fluor
O
R
OH
R
F
1
1
1
0
1
2
22 (90) 30
23 (91) 30
24 (90) 30
14
OH
OH
1
4
O
P
:
PPh3
O
N
O
R
N N N
CN
N
R
N
16
16
P
O
O
N
N
CN
7
R-CN + N2 + Ph PO
3
OH
7
R
N
7
7
a
1
Yield of pure, isolated products (characterized by GC–MS, and
and C NMR).
H
Figure 1. Proposed mechanism for the conversion of carboxylic acids
to nitriles.
13

5936
C. O. Kangani et al. / Tetrahedron Letters 48 (2007) 5933–5937
Table 3. Reaction compatibility with various phosphorus reagents
NMR) for all products) associated with this article can
be found, in the online version, at doi:10.1016/
j.tetlet.2007.06.119.
Phosphorous reagent
Carboxylic acid
O
O
OH
OH
1
4
Br
References and notes
a
a
Et
3
P
(100%), 30 min
(100%), 90 min
No product
(100%), 30 min
a
a
1. (a) Br a¨ se, S.; Zimmermann, V.; Gil, C.; Knepper, K.
Angew. Chem., Int. Ed. 2005, 44, 5188; (b) Patai, S. In
Chemistry of the Azido Group; Interscience: New York,
(
Bu
EtO)
3
P
(100%), 90 min
t
3
P
No product
No product
1
971; p 397; (c) Lwowski, W. Azides and Nitrenes:
Reactivity and Utility; Academic Press: New York, 1984,
p 205; (d) Scriven, E. F. V.; Turnbull, K. Chem. Rev. 1988,
P
No product
8
8, 297; (e) Moore, H. W.; Goldish, D. M. In Chemistry of
Halides Pseudo-Halides and Azides; Patai, S., Rappoport,
Z., Eds.; Wiley: Chichester, UK, 1983; Vol. 1, p 321.
. (a) Wipf, P. Chem. Rev. 1995, 95, 2115; (b) Khanna, I. K.;
Weier, R. M.; Yu, Y.; Xu, X. D.; Koszyk, F. J.; Collins, P.
W.; Koboldt, C. M.; Veenhuizen, A. W.; Perkins, W. E.;
Casler, J. J.; Masferrer, J. L.; Zhang, Y. Y.; Gregory, S.
A.; Seibert, K.; Isakson, P. C. J. Med. Chem. 1997, 40,
1634; (c) Gu, X.-H.; Wan, X.-Z.; Jiang, B. Bioorg. Med.
Chem. Lett. 1999, 9, 569; (d) Ducept, P. C.; Marsden, S. P.
Synlett 2000, 692; (e) Shie, J.-J.; Fang, J.-M. J. Org. Chem.
2003, 68, 1158.
2
O
a
a
OEt
(35%), 24 h
(50%), 24 h
Ph P
3
Ph P
2
a
a
(
100%), 50 min
(100%), 30 min
3
. (a) Capdevielle, P.; Lavigne, A.; Sparfel, D.; Baranne-
Lafont, J.; Cuong, N. K.; Maumy, M. Tetrahedron Lett.
(
Rf = C8F17)
1
2
990, 31, 3305; (b) Gao, S.; Herzig, D.; Wang, B. Synthesis
001, 544; (c) Chen, F.-E.; Kuang, Y.-Y.; Dai, H.-F.; Lu,
Rf
a
Percent conversion as determined by GC–MS. All reactions com-
pletely consumed the carboxylic acids. Where the values are <100%,
the remaining materials were the respective acyl azides.
L.; Huo, M. Synthesis 2003, 2629; (d) De Luca, L.;
Giacomelli, G. Synlett 2004, 2180.
. (a) Bose, D. S.; Narsaiah, A. V. Tetrahedron Lett. 1998,
4
3
9, 6533; (b) Chen, F.-E.; Fu, H.; Meng, G.; Cheng, Y.;
time (24 h) and excess phosphorus reagent (10 equiv). A
L u¨ , Y.-X. Synthesis 2000, 1519; (c) Erman, M. B.; Snow, J.
W.; Williams, M. J. Tetrahedron Lett. 2000, 41, 6749; (d)
Lai, G.; Bhamare, N. K.; Anderson, W. K. Synlett 2001,
230; (e) Sharghi, H.; Sarvari, M. H. Tetrahedron 2002, 58,
10323; (f) Sharghi, H.; Sarvari, M. H. Synthesis 2003, 243.
. (a) Nakajima, N.; Ubukata, M. Tetrahedron Lett. 1997,
fluorous analogue of PPh , whose unreacted form and
3
by-products can be removed by selective solvent or solid
1
8
phase partitioning, was found to be on par with PPh3
for the conversion to nitriles.
5
6
3
7
8, 2099; (b) Bose, D. S.; Narsaiah, A. V. Synthesis 2001,
, 5237.
In summary, an efficient new one-pot direct method for
the synthesis of various acyl azides from carboxylic
acids was found. The method provides excellent yields
in short reaction times. Inclusion of certain phospho-
rous reagents in the reaction provides a simple and effi-
cient method for the direct conversion of carboxylic
acids to nitriles, in high yield and without racemization.
The compatibility with volatile phosphorous reagents as
well as a fluorous phosphine provides means to simplify
purification of nitrile from the reaction mixture.
. (a) Wang, E.-C.; Lin, G.-J. Tetrahedron Lett. 1998, 39,
047; (b) De Luca, L.; Giacomelli, G.; Porcheddu, A. J.
4
Org. Chem. 2002, 67, 6272; (c) Chandrasekhar, S.;
Gopalaiah, K. Tetrahedron Lett. 2003, 44, 755; (d)
Czekelius, C.; Carreira, E. M. Angew. Chem. 2005, 117,
618.
7. (a) Chen, F.-E.; Li, Y.-Y.; Xu, M.; Jia, H.-Q. Synthesis
2002, 1804; (b) Iranpoor, N.; Firouzabadi, H.; Akhlaghi-
nia, B.; Nowrouzi, N. J. Org. Chem. 2004, 69, 2562; (c)
Mori, N.; Togo, H. Synlett 2005, 1456.
8
. (a) Huber, V. J.; Bartsch, R. A. Tetrahedron 1998, 54,
281; (b) Mlinari c´ -Majerski, K.; Margeta, R.; Veljkovi c´ , J.
9
Acknowledgments
Synlett 2005, 2089.
9
. (a) Ito, M.; Koyakumaru, K. I.; Ohta, T.; Takaya, H.
Synthesis 1995, 376; (b) Laszlo, P.; Polla, E. Tetrahedron
Lett. 1984, 25, 3701; (c) Kaiser, C.; Weinstock, J. Org.
Synth. 1971, 51, 48; (d) Bolm, C.; Schiffers, I.; Dinter, C.
L.; Defr e` re, L.; Gerlach, A.; Raabe, G. Synthesis 2001,
This investigation was supported by funding from
the National Institutes of Health of the University of
Pittsburgh Obesity and Nutrition Research Center
(
DK46204). We thank Professor Paul Floreancig for
1
719; (e) Bolm, C.; Schiffers, I.; Atodiresei, I.; Hacken-
helpful discussions regarding the mechanism of the
reaction.
berger, C. P. R. Tetrahedron: Asymmetry 2003, 14, 3455.
1
0. (a) Shao, H.; Colucci, M.; Tong, S.; Zhang, H.; Castel-
hano, A. L. Tetrahedron Lett. 1998, 39, 7235; (b) Froeyen,
P. Phosphorus, Sulfur Silicon Relat. Elem. 1994, 89, 57; (c)
Gumaste, V. K.; Bhawal, B. M.; Deshmukh, A. R. A. S.
Tetrahedron Lett. 2002, 43, 1345; (d) Bandgar, B. P.;
Pandit, S. S. Tetrahedron Lett. 2002, 43, 3413; (e)
Kobayashi, S.; Kamiyama, K.; Iimori, T.; Ohno, M.
Supplementary data
Supplementary data (a description of the general meth-
1
13
ods, GC–MS, and spectroscopic data ( H and
C

C. O. Kangani et al. / Tetrahedron Letters 48 (2007) 5933–5937
5937
Tetrahedron Lett. 1984, 25, 2557; (f) Canonne, P.; Akssira,
M.; Dahouh, A.; Kasmi, H.; Boumzebra, M. Heterocycles
was very good, the reaction took a much longer time, 12–
15 h, to proceed to completion; (b) Bretherick, L. Chem.
Eng. News 1986, 64, 2; (c) Peet, N. P.; Weintraub, P. M.
Chem. Eng. News 1993, 71, 4; (d) Hruby, V. J.; Boteju, L.;
Li, G. Chem. Eng. News 1993, 71, 4.
1
993, 36, 1305; (g) Arrieta, A.; Aizpurua, J. M.; Palomo,
C. Tetrahedron Lett. 1984, 25, 3365; (h) Sridhar, R.;
Perumal, P. T. Synth. Commun. 2003, 33, 607.
1
1
1. (a) Lal, G. S.; Pez, G. P.; Pesaresi, R. J.; Prozonic, F. M.;
Cheng, H. J. Org. Chem. 1999, 64, 7048; (b) Lal, G. S.;
Pez, G. P.; Pesaresi, R. J.; Prozonic, F. M. Chem.
Commun. 1999, 215.
15. Shilov, W.; Lippmann, E. Z. Chem. 1986, 26, 101.
16. (a) Banfi, L.; Basso, A.; Gandolfo, V.; Guanti, G.; Riva,
R. Tetrahedron Lett. 2004, 45, 4221; (b) Harrowven, D. C.;
Pascoe, D. D.; Demurtas, D.; Bourne, H. O. Angew.
Chem. 2005, 117, 1247; Angew. Chem., Int. Ed. 2005, 44,
1221; (c) Lee, D.; Kim, D.; Yun, J. Angew. Chem., Int. Ed.
2006, 45, 2785.
17. Shalev, D. E.; Chiacchiera, S. M.; Radkowsky, A. E.;
Kosower, E. M. J. Org. Chem. 1996, 61, 1689.
18. (a) Dandapani, S.; Curran, D. P. Tetrahedron 2002, 58,
3855; (b) Lindsley, C. W.; Zhao, Z.; Newton, R. C.;
Leister, W. H.; Strauss, K. A. Tetrahedron Lett. 2002, 43,
4467.
2. (a) Kangani, C. O.; Kelley, D. E. Tetrahedron Lett. 2005,
4
6, 8917; (b) Kangani, C. O.; Kelley, D. E.; Day, B. W.
Tetrahedron Lett. 2006, 47, 6289; (c) Kangani, C. O.;
Kelley, D. E.; Day, B. W. Tetrahedron Lett. 2006, 47,
6
497.
1
3. White, J. M.; Tunoori, A. R.; Turunen, B. J.; Georg, G. I.
J. Org. Chem. 2004, 69, 2573.
1
4. (a) Originally, we performed the reaction by adding NaN
as a solid (i.e., not predissolved in DMSO). While the yield
3