A new type of amide formation from thiocarboxylic acid and alkyl azide

Article abstract of DOI:10.1016/S0040-4039(02)01397-7

We studied the coupling of thiocarboxylic acid and alkyl azide using various triaryl phosphines. Amide formation greater than 95% was achieved when the free-formation of Staudinger intermediate with electron deficient triaryl phosphines was employed.

Full text of DOI:10.1016/S0040-4039(02)01397-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 6309–6311
A new type of amide formation from thiocarboxylic acid and
alkyl azide
Sang-Don Park, Jung-Hee Oh and Dongyeol Lim*
Department of Applied Chemistry, Sejong University, Seoul 143-747, Republic of Korea
Received 9 July 2002; accepted 11 July 2002
Abstract—We studied the coupling of thiocarboxylic acid and alkyl azide using various triaryl phosphines. Amide formation
greater than 95% was achieved when the free-formation of Staudinger intermediate with electron deficient triaryl phosphines was
employed. © 2002 Elsevier Science Ltd. All rights reserved.
Amide formation from carboxylic acids and amines has
been studied comprehensively and numerous proce-
dures have been developed and implemented. Most of
the amide formation involves the generation of free
amino groups and activation of carboxylic acid as the
method of choice, but an efficient chemoselective amide
formation is still in demand to synthesize various
chemicals.
First we tried the previously reported condition2a to
obtain an amide product from thiobenzoic acid and
benzyl azide. After heating a mixture of all the reagents
in various solvents, we could not produce an amide
with more than 40% yield. We also found various
equivalents of benzyl azide, triphenyl phosphine and
organic bases resulted in lower yields. So we prepared a
phosphazene intermediate by first heating a mixture of
benzyl azide and triphenyl phosphine. We observed the
gas (N2) evolved from the reaction mixture and moni-
tored the phosphazene formation by NMR.4 After 10–
48 hours of heating we observed no more phosphazene
formation from various triaryl phosphine and benzyl
azides in CH3CN. The electron rich tri-4-
methoxyphenyl phosphine required a longer time than
tri-4-fluorophenyl phosphine. When thiocarboxylic acid
was added to the phosphazene solution, fast transfor-
mation to an amide was observed.5
Formation of Staudinger intermediate1 (phosphazenes
from azides and tertiary phosphines) is a convenient
way of generating nucleophilic amine from azide
(Scheme 1). Studies by Vilarrassa et al. have shown that
this intermediate can be coupled to acids, acid halides
or acid anhydrides to yield an amide in an inter- or
intramolecular manner.2 Recent studies by Saxon et al.
and Nilson et al. presented further improvement of the
reaction providing an efficient tool for successful liga-
tion of peptides or proteins.3
With variations of phosphines and solvents, we
attempted a coupling reaction of benzyl azide and
thiobenzoic acid (Scheme 2).6
In an attempt to prepare thioamides, we have found
that amide can be obtained from monothiocar-
boxylic acids and phosphazenes. In this paper we
describe the first successful amide formation from
monothiocarboxylic acid and azide with various triaryl
phosphines.
Among the phosphines we tried, triphenyl phosphine
was the best, yielding up to 91% coupling product in
CH3CN (entry 1). Other solvents, such as methylene
R2
R2
R2
R2
-N2
N
N
N
R23P
+
R1
N
P
R2
R1
R1
P
N3
R2
solvents
Scheme 1. Staudinger intermediate from azide and phosphine.
Keywords: Staudinger intermediate; amide; thiocarboxylic acid; azide; phosphine.
* Corresponding author. Tel.: +822-3408-3218; fax: +822-462-9954; e-mail: dylim@sejong.ac.kr
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01397-7

6310
S.-D. Park et al. / Tetrahedron Letters 43 (2002) 6309–6311
Table 2. Reaction of benzyl azide in acetonitrile
chloride, dioxane and DMF, gave lower yields (entries
2–5). Benzene and hexane gave very low yields (entries
6–7). Vilarrasa et al. reported better yields of amide
formation from propionic acid, benzyl azide, and Ph3P
in refluxing benzene (95%) and refluxing hexane (78%)
when compared with the same reaction in refluxing
CH3CN (61%).2a When BOP or PyBOP was used, no
product was obtained, implying that reduced benzyl
amine is not an intermediate for the amide formation
(Table 1).
Entry
Thioacid
R3P
Yield (%)
1
2
3
4
5
6
7
8
9
Thiobenzoic acid
Ph3P
91
91
97
62
(4-F-Ph)3P
(3-F-Ph)3P
(4-CH3O-Ph)3P
(2-Furyl)3P
Ph3P
(4-F-Ph)3P
(3-F-Ph)3P
(4-CH3O-Ph)3P
(2-Furyl)3P
91
Thioacetic aid
72a
54a
81a
50a
89a
Based on the result of Table 1 in which triphenylphos-
phine is the best reagent, we evaluated benzyl azide
coupling with two thioacids using substituted phenyl
and furyl phosphines in acetonitrile (Scheme 3). Inter-
estingly, with both thiobenzoic acid and thioacetic acid,
tri-4-fluorophenyl phosphine gave better yields than
triphenyl phosphine, while tri-4-methoxyphenyl phos-
phine gave lower yields. When tri-2-furyl phosphine
was employed, the same or slightly better yields were
obtained (entries 5, 10). These substituent effects indi-
cate that this reaction is sensitive to the electron density
at phosphine (Table 2).
10
a Yields were determined by NMR integration of the methylene peaks
against benzyl acetate as an internal standard.
O
O
R23P
O
OMe
+
R1
N
H
N3
R1
SH
MeO
CH3CN
O
Scheme 4. Amide formation from azido methylacetate and
triaryl phosphine.
In order to expand this type of reaction, we changed
benzyl azide to azido methylacetate (Scheme 4). We
consider this azide a glycine analogue that can be
applied to peptide chemistry.
The results of the coupling reactions of azido methylac-
etate and two thioacids are listed in Table 3. Similar to
benzyl azide reactions, electron deficient phosphine
gave the highest yield, up to 96%. Phenyl and
methoxyphenyl phosphines gave lower yields with both
thioacids. Again 2-furyl phosphine gave a similar or
better yield compared to triphenyl phosphine.
O
O
N3
Ph
SH
+
R3P
Ph
Ph
N
H
Ph
Considering the reaction mechanism, two intermediates
could possibly be formed (Scheme 5). If the reaction
solvent
Scheme 2. Amide formation from benzyl azide and thioben-
zoic acid.
Table 3. Reaction of azido methylacetate and thiocar-
boxylic acid in CH3CN
Table 1. Reaction of benzyl azide and thiobenzoic acid
Entry
Thioacid
R3P
Yield (%)
Entry
R3P
Solvent
Yield (%)
1
2
3
4
5
6
7
8
Thiobenzoic acid
Ph3P
49
96
62
94
(4-F-Ph)3P
(4-CH3O-Ph)3P
(2-Furyl)3P
Ph3P
(4-F-Ph)3P
(4-CH3O-Ph)3P
(2-Furyl)3P
1
2
3
4
5
6
7
8
9
Ph3P
Ph3P
Ph3P
Ph3P
Ph3P
Ph3P
Ph3P
Bu3P
BOPa
PyBOPb
CH3CN
CH2Cl2
1,4-Dioxane
DMF
91
35
62
85
45
23
14
26
0
Thioacetic aid
68a
96a
49a
69a
CHCl3
Benzene
Hexane
CH3CN
CH3CN
CH3CN
a Yields were determined by NMR integration of the methylene peaks
against benzyl acetate as an internal standard.
10
0
a Benzotriazolyl-oxy-tris(dimethylamino)-phosphonium hexafluoro-
phosphate.
b Benzotriazolyl-oxy-trispyrroldino-phosphonium
phate
hexafluorophos-
O
O
R23P
+
R1
N
H
Ph
Ph
N3
R1
SH
CH3CN
Scheme 3. Amide formation from thiocarboxylic acid and
benzyl azide with triaryl phosphine.
Scheme 5. Possible intermediate for amide formation.

S.-D. Park et al. / Tetrahedron Letters 43 (2002) 6309–6311
6311
follows path A, thioamide will be obtained, but we
failed to detect thioamide formation. On the other
hand, a stable amide product was obtained, confirming
that the reaction went through path B. The equilibrium
between thiono- and thiolo-carboxylic acid and differ-
ences in bond length and nucleophilicity may play a
crucial role in yielding amide products.
A. E.; Kosower, E. M. J. Org. Chem. 1996, 61, 1689–1701;
(h) Bosch, I.; Gonzalez, A.; Urpi, F.; Vilarrasa, J. J. Org.
Chem. 1996, 61, 5638–5643; (i) Tang, Z.; Pelletier, J. C.
Tetrahedron Lett. 1998, 39, 4773–4776; (j) Ariza, X.; Urpi,
F.; Viladomat, C.; Vilarrasa, J. Tetrahedron Lett. 1998, 39,
9101–9102; (k) Mizuno, M.; Muramoto, I.; Kobayashi, K.;
Yaginuma, H.; Inazu, T. Synthesis-Stuttgart 1999, 162–
165; (l) Mizuno, M.; Haneda, K.; Iguchi, R.; Muramoto,
I.; Kawakami, T.; Aimoto, S.; Yamamoto, K.; Inazu, T. J.
Am. Chem. Soc. 1999, 121, 284–290; (m) Boullanger, P.;
Maunier, V.; Lafont, D. Carbohydr. Res. 2000, 324, 97–
106; (n) Malkinson, J. P.; Falconer, R. A.; Toth, I. J. Org.
Chem. 2000, 65, 5249–5252.
In conclusion, we were able to obtain amide products
from monothiocarboxylic acid and alkyl azide for the
first time in satisfactory yields. From the optimization
of the reaction, we found that electron deficient aryl
phosphine is the best reagent in CH3CN. This reaction
would be generally applicable to chemoselective amide
formation from thiocarboxylic acid and azide. Further
applications in the syntheses of glycopeptides and pep-
tidomimetics are in progress and the results will be
published in due course.
3. (a) Saxon, E.; Bertozzi, C. R. Science 2000, 287, 2007–
2010; (b) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T.
Org. Lett. 2000, 2, 1939–1941; (c) Saxon, E.; Armstrong, J.
I.; Bertozzi, C. R. Org. Lett. 2000, 2, 2141–2143; (d)
Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett.
2001, 3, 9–12.
4. Chemical shift of benzylic proton in CDCl3 was moved,
upon mixing with triphenylphosphine, from 4.32(s) to
4.41(s), then slowly to 3.87(s), corresponding to benzyl
azide, phosphazide and phosphazene, respectively.
5. Upon addition of thiobenzoic acid to the phosphazene
solution, NMR shift of 4.66 (d, J=5.7 Hz) appeared
quickly.
Acknowledgements
This study was supported by a grant from the Korea
Health 21 R&D Project, Ministry of Health and Wel-
fare, Republic of Korea (HMP-00-B-21500-0113).
6. General procedure: Triphenyl phosphine (342 mg, 1.3
mmol) was added to a solution of benzyl azide (145 mg,
1.09 mmol) in acetonitrile (10 mL) and the mixture was
stirred for 1 h at 0°C and 40 h at 65°C. To this mixture
was added thiobenzoic acid (100 mg, 0.72 mmol) and the
solution was stirred for 24 h at room temperature. After
the reaction mixture was concentrated, water (10 mL) was
added and the solution was extracted with EtOAc (10
mL×3). The combined organic layer was washed with
saturated NaHCO3 and brine, dried over MgSO4 and
concentrated. After purification via SiO2 chromatography,
139 mg of N-benzyl benzamide was obtained (91% yield).
1H NMR (CDCl3, 200 MHz) l 7.9–7.75 (2H, m), 7.6–7.4
(8H, m), 6.7–6.4 (1H, bs), 4.66 (2H, d, J=5.7 Hz); mass
Calcd for C14H15NO: 211, Found: 211.
References
1. (a) Staudinger, H.; Meyer, J. Helv. Chim. Acta 1919, 2,
635–646; (b) Gololobov, Yu. G.; Kasukhin, L. F. Tetra-
hedron 1992, 48, 1353–1406.
2. For selected examples, see: (a) Garcia, J.; Urpi, F.; Vilar-
rasa, J. Tetrahedron Lett. 1984, 25, 4841–4844; (b) Garcia,
J.; Vilarrasa, J. Tetrahedron Lett. 1984, 27, 639–640; (c)
Urpi, F.; Vilarrasa, J. Tetrahedron Lett. 1986, 27, 4623–
4624; (d) Bosch, I.; Romea, P.; Urpi, F.; Vilarrasa, J.
Tetrahedron Lett. 1993, 34, 4671–4674; (e) Inazu, T.;
Kobayashi, K. Synlett 1993, 869–870; (f) Bosch, I.; Urpi,
F.; Vilarrasa, J. J. Chem. Soc., Chem. Commun. 1995,
91–92; (g) Shalev, D. E.; Chiacchiera, S. M.; Radkowsky,