Amide bond formation from selenocarboxylates and aromatic azides

Article abstract of DOI:10.1016/j.tetlet.2005.09.145

A new method of amide bond formation was developed through the reaction of potassium selenocarboxylates with aromatic azides at room temperature. Potassium selenocarboxylates were prepared in situ by the treatment of diacyl selenides with potassium methoxide at 5°C under N₂. After the addition of azide, the reaction was allowed to gradually warm to room temperature and was stirred for 0.5-2 h. Excellent yields were obtained when electron deficient aromatic azides were used.

Full text of DOI:10.1016/j.tetlet.2005.09.145

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 8401–8405
Amide bond formation from selenocarboxylates and aromatic azides
Xinghua Wu and Longqin Hu*
Department of Pharmaceutical Chemistry, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey,
Piscataway, NJ 08854, USA
Received 26 August 2005; revised 22 September 2005; accepted 23September 2005
Available online 10 October 2005
Abstract—A new method of amide bond formation was developed through the reaction of potassium selenocarboxylates with aro-
matic azides at room temperature. Potassium selenocarboxylates were prepared in situ by the treatment of diacyl selenides with
potassium methoxide at 5 ꢀC under N2. After the addition of azide, the reaction was allowed to gradually warm to room temper-
ature and was stirred for 0.5–2 h. Excellent yields were obtained when electron deficient aromatic azides were used.
ꢁ 2005 Elsevier Ltd. All rights reserved.
Formation of an amide bond is usually achieved
through the reaction of an activated carboxylic acid with
an amine. However, there are situations where certain
functional groups are incompatible with the presence
of a free amine. One such situation is the spontaneous
decomposition of 4-aminobenzyl esters through 1,6-
elimination.1 As an alternative to an amine, the use of
non-nucleophilic and non-basic starting material and
intermediates is needed. The azido group has received
attention because it is easily transformed into various
nitrogen-containing functional groups. The reaction of
a thio acid with an organic azide to give the correspond-
ing amide was first reported by Hakimelahi and Just.2
Later, it was proposed that the reaction went through
a rapid in situ reduction of the azide to the correspond-
ing amine prior to amide bond formation.3 Recently,
evidence was presented that excluded the formation of
an amine as the reactive intermediate and revised the
mechanism to proceed via a thiatriazoline intermediate
2 as shown in Scheme 1.4 The reaction was accelerated
by the presence of bases such as 2,6-lutidine. Subse-
quently, other applications and modifications of the thio
acid-azide amidation have appeared.5–7
S
R
R
N
O
+ H +
O
S
S
N
N3 R'
R'
R
N
R
N
H
O
R'
N
2+S
2
3
O
1
Scheme 1. Mechanism proposed for the reaction of thio acids with
azides.
O
LG
LG
LG
direct amidation
R
N
H
N3
4
5
c
i
s
l
y
x
o
b
e
r
a
c
[H]
l
d
o
a
n
l
h
t
o
i
t
n
c
m
e
g
v
i
n
i
n
c
o
p
c
u
o
d
a
polymerization
1,6- elimination
H2N
HN
6
7
LG = a good leaving group (e.g., OCOCH3)
Scheme 2. Direct amidation from aromatic azides avoids 1,6-elimina-
tion of 4-aminobenzyl derivatives.
In our research, quick decomposition of compound 6 in
the form of 4-aminobenzyl-LG (where LG = a good
leaving group) to quinonimine methide 7 and sub-
sequent polymerization, as shown in Scheme 2, made
it impossible to prepare the corresponding amide 5.
The spontaneous 1,6-elimination of 6, therefore, pre-
cluded the use of general amidation procedures that
require a free amine to react with an activated carbox-
ylic acid to form the amide 5. Thus, we needed an alter-
native method to prepare these aryl amides. Ideally, we
would achieve this using our aromatic azide starting
Keywords: Amidation; Selenocarboxylates; Aromatic azides.
*
Corresponding author. Tel.: +1 732 445 5291; fax: +1 732 445
6312; e-mail: LongHu@rutgers.edu
0040-4039/$ - see front matter ꢁ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.09.145

8402
X. Wu, L. Hu / Tetrahedron Letters 46 (2005) 8401–8405
materials (i.e., 4 ! 5). The thio acid/azide reaction is
very rapid if the azide is electron deficient, but the unac-
tivated phenyl azide was shown to require refluxing in
chloroform for several hours for good conversion.4 We
wondered if selenocarboxylates would couple more rap-
idly with azides than thiocarboxylates, since seleno-
derivatives are typically more reactive than their corre-
sponding sulfur analogs.8 The enhanced reactivity
would potentially lower reaction temperature and short-
en reaction time. Although selenocarboxylic acids are
known to be unstable to heat and sensitive to oxygen,
the corresponding alkali metal and piperidinium salts
are relatively stable, especially in the case of aromatic
selenocarboxylates.9 For example, no significant change
occurred in potassium 4-methylbenzeneselenocarboxyl-
ate when it was exposed to air for 5 h. Under oxygen-
free conditions, most aromatic selenocarboxylate salts
can be stored at ꢀ17 ꢀC for at least one month.10 Potas-
sium benzeneselenocarboxylate (10) was thus chosen for
our present studies.
functionalities were selected to explore the scope of the
amidation reaction and to give a better assessment of
the electronic and steric effects on the reaction outcome.
As with thiocarboxylate analogs, it was found that elec-
tron-deficient azides were much more reactive than elec-
tron-rich azides.19 Aromatic azides 12–18 bearing
electron-withdrawing groups such as NO2, Cl, CN,
COCH3, COOH, and COOMe (Table 1, entries 1–9)
gave excellent yields while phenyl azide 19 (Table 1,
entry 10) and aromatic azides 20–22 with electron-
donating groups such as OCH3, CH3, and CH2OH
(Table 1, entries 11–13) gave lower yields of desired
amides with most of the azide starting materials recov-
ered. Electron-withdrawing groups at the para or ortho
position probably stabilize the transition state delocal-
ization of the negative charge on the nitrogen, for exam-
ple, 39 as shown in Scheme 4, and thus facilitate the
formation of the amide bond. Interestingly, the reaction
is slower when the nitro group is at the ortho position as
in 13 than when it is at the para position as in 12. How-
ever, moving the nitro substituent from the para to the
ortho position in the starting azide did not affect the
overall yield of amidation under our current conditions
(Table 1, compare entries 2 and 1). Acetylation of 4-azi-
dobenzyl alcohol 22, as in compound 23, increased the
amidation yield from 44% to 70% (Table 1, compare
entries 14 and 13), and may reflect an inductive elec-
tronic stabilization in the amidation reaction. The suc-
cessful formation of 4-benzoylaminobenzyl acetate (37)
from 4-azidobenzyl acetate (23) (Table 1, entry 14) indi-
cates that the reaction does not involve in situ reduction
of azide 23 to the corresponding amine prior to amide
bond formation, since such reduction would have lead
to the formation of quinonimine methide and subse-
quent polymerization instead of the desired amide, as
discussed earlier and shown in Scheme 2. The reaction
of the aliphatic selenocarboxylate (Table 1, entries 6
and 9) gave comparable yields to aromatic selenocarb-
oxylates. These results are consistent with earlier reports
wherein the reaction yields depend primarily upon
the electronic properties of azides. Based on these
results, a mechanism similar to that proposed for thio-
carboxylates3 is proposed for this amidation reaction
(Scheme 4).
Alkali metal salts of selenocarboxylates can be prepared
by the treatment of trimethylsilyl selenocarboxylic esters
with alkali metal fluorides,11 the reaction of acyl chlo-
rides with alkali metal selenides,12,13 the reaction of
diacyl selenides with alkali metal hydroxide,14 or the
treatment of carboxylic acids with WoollinsÕ reagent in
refluxing toluene.5 We selected the method using diacyl
selenides with potassium methoxide15 as shown in
Scheme 3 since diacyl selenides are relatively stable
and readily prepared. Moreover, aromatic diacyl sele-
nides can be purified by silica gel flash column chroma-
tography.16 We obtained dibenzoyl selenide (9) through
the reaction of 2 equiv of benzoyl chloride (8) with
1 equiv of LiAlHSeH. The latter was freshly prepared
by reacting lithium aluminum hydride with selenium
powder.17,18
For our coupling reaction, potassium benzeneseleno-
carboxylate (10) was generated in situ by mixing di-
benzoyl selenide (9) with 1 equiv of potassium methoxide
in DMSO/EtOAc (1:1) at 5 ꢀC under N2 (Scheme 3).
DMSO was used to increase the solubility of potas-
sium methoxide. After azide addition, the reaction was
allowed to gradually warm to room temperature and
was stirred for 0.5–2 h until TLC and/or LC–MS showed
the disappearance of the starting azide or no further
change of the reaction mixture.
The selenocarboxylate-azide amidation is highly chemo-
selective. All reactions were very clean giving >90%
yields based on recovered starting materials; and no
other side reactions were observed. All unreacted azide
starting materials could be recovered. The lower conver-
sion yields for the less reactive azides could be due to the
limited stability of selenocarboxylates under the current
reaction conditions. To test this hypothesis, we used the
reaction of benzeneselenocarboxylate 10 with 4-nitro-
phenyl azide (12), the fastest amidation reaction in
Table 1, to monitor the stability of selenocarboxylate
under the reaction conditions. When benzeneselenocarb-
oxylate 10 was incubated with 4-nitrophenyl azide (12)
in DMSO/EtOAc (1:1) under our standard conditions,
complete and quantitative amidation was achieved in
less than 5 min as monitored by HPLC. To monitor
the stability of selenocarboxylate under these condi-
tions, dibenzoyl selenide (9) was mixed with 1 equiv
A series of homologous aromatic azides (12–23) bearing
various electron-donating and electron-withdrawing
O
O
O
1 eq
LiAlHSeH
Se
Cl
2
9
8
O
O
R-N3
R
1 eq KOMe
Se-K+
10
Scheme 3. Direct amidation of azides via selenocarboxylates.
N
H
11

X. Wu, L. Hu / Tetrahedron Letters 46 (2005) 8401–8405
Table 1. Amidation reaction of selenocarboxylates with azidesa
8403
Entry
Azide starting material
Amide product
Time (h)
0.5
Yield (%)b
O
O2N
N3
1
98
12
O2N
NH
24
O
N3
NO2
NH
2
3
1.0
0.5
95
96
13
NO2
25
O
Cl
N3
Cl
NH
14
26
O
NC
N3
4
5
0.5
1.0
98
98
NC
NH
15
27
O
O
O
N3
NH
MeO
16
16
MeO
28
O
O
O
N3
6
7
1.0
2.0
91
87
NH
MeO
MeO
29
O
O
O
N3
NH
HO
17
HO
30
O
O
O
O
O
8
9
1.0
1.0
88
N3
N3
NH
18
18
31
O
76 (19)c
NH
32
O
N3
10
11
12
13
14
2.0
2.0
2.0
2.0
2.0
25 (65)c
7 (88)c
NH
19
33
O
MeO
H3C
N3
MeO
H3C
NH
20
34
O
7 (85)c
N3
NH
21
35
O
44 (51)c
70 (24)c
N3
NH
HO
22
HO
36
O
N3
NH
AcO
23
AcO
37
a General conditions: diacylselenide (2 mmol), KOMe (2 mmol), EtOAc/DMSO (1:1), 5 ꢀC, 10 min; then azide (1 mmol), 5 ꢀC– rt.
b Isolated yield (%) of the amide product.
c The yield of the recovered azide starting material is shown in parenthesis.
of potassium methoxide in DMSO/EtOAc (1:1) at 5 ꢀC
to form the selenocarboxylate. The benzeneselenocarb-
oxylate 10 formed was allowed to stand at room temper-
ature and aliquots were combined with 2 equiv of

8404
X. Wu, L. Hu / Tetrahedron Letters 46 (2005) 8401–8405
without going through the nucleophilic and basic free
N3
N
R
Se
N
amine intermediates. It should be useful when amide
bonds need to be formed between molecules with func-
tional groups incompatible with free amines and basic
conditions.
O
N
O
R
R
R
(12-23)
Se
38
39
N
R
Se
R
N
H+
NH
Acknowledgments
N
R
O
O
R
We gratefully acknowledge the financial support of a re-
search grant from the State of New Jersey Commission
on Cancer Research and a research scholar grant from
the American Cancer Society (to L.H.). We thank Pro-
fessors Lawrence Williams and Spencer Knapp for help-
ful discussions.
N2 +Se
40
24-37
Scheme 4. Mechanism proposed for the reaction of selenocarboxylates
with aromatic azides.
References and notes
1.0
0.8
0.6
0.4
0.2
0.0
1. Guibe-Jampel, E.; Wakselman, M. Synth. Commun. 1982,
12, 219–223.
2. Hakimelahi, G. H.; Just, G. Tetrahedron Lett. 1980, 21,
2119–2122.
3. Rosen, T.; Lico, I. M.; Chu, D. T. W. J. Org. Chem. 1988,
53, 1580–1582.
4. Shangguan, N.; Katukojvala, S.; Greenberg, R.; Williams,
L. J. J. Am. Chem. Soc. 2003, 125, 7754–7755.
5. Fazio, F.; Wong, C.-H. Tetrahedron Lett. 2003, 44, 9083–
9085.
6. Knapp, S.; Darout, E. Org. Lett. 2005, 7, 203–206.
7. Zhu, X. M.; Pachamuthu, K.; Schmidt, R. R. Org. Lett.
2004, 6, 1083–1085.
0
20
40
60
80
100
120
8. Pearson, R. G.; Sobel, H. R.; Songstad, J. J. Am. Chem.
Soc. 1968, 90, 319–326.
Time (min)
9. Ishilhara, H.; Muto, S.; Kato, S. Synthesis 1986, 128–
130.
10. Niyomura, O.; Tani, K.; Kato, S. Heteroatom Chem. 1999,
10, 373–379.
Figure 1. Stability of phenylselenocarboxylate 10 in DMSO/EtOAc
(1:1) at room temperature as monitored by conversion to N-(4-
nitrophenyl) benzamide (24) in an HPLC assay.
11. Niyomura, O.; Kato, S.; Kanda, T. Inorg. Chem. 1999, 38,
507–518.
12. Kawahara, Y.; Kato, S.; Kanda, T.; Murai, T. Bull. Chem.
Soc. Jpn. 1994, 67, 1881–1885.
13. Kojima, Y.; Ibi, K.; Kanda, T.; Murai, T.; Kato, S. Bull.
Chem. Soc. Jpn. 1993, 66, 990–992.
14. Ishihara, H.; Hirabayashi, Y. Chem. Lett. 1976, 5, 203–
204.
15. Kageyama, H.; Takagi, K.; Murai, T.; Kato, S. Z.
Naturforsch. 1989, 44b, 1519–1523.
16. Zhao, H. R.; Zhao, X. J.; Huang, X. Synth. Commun.
2002, 32, 3383–3388.
17. Ishihara, H.; Koketsu, M.; Fukuta, Y.; Nada, F. J. Am.
Chem. Soc. 2001, 123, 8408–8409.
18. Koketsu, M.; Nada, F.; Hiramatsu, S.; Ishihara, H. J.
Chem. Soc., Perkin Trans. 1 2002, 737–740.
19. General procedure: To a solution of diacyl selenide
(2.0 mmol) in EtOAc (20 mL) was added a suspension of
KOMe (2.0 mmol) in DMSO (20 mL) at 5 ꢀC under
nitrogen atmosphere. The resulting mixture was stirred at
5 ꢀC for 10 min. Then, the starting azide (1.0 mmol) was
added into the above mixture. The reaction was allowed to
gradually warm to room temperature. The reaction mix-
ture was poured into ice water when TLC showed the
disappearance of the starting azide or no further change of
the reaction mixture. The black selenium powder was
removed by filtration. The filtrate was extracted with
EtOAc. After removal of solvents, the crude product was
purified by flash column chromatography.
4-nitrophenyl azide (12). The amide product 24 was ana-
lyzed to determine the amount of benzeneselenocarb-
oxylate that remained in solution. Figure 1 shows the
plot of normalized amide product 24 formed, i.e., rel-
ative selenocarboxylate 10, against the time when excess
azide was added. Indeed, selenocarboxylate 10 was not
stable and has an apparent half-life of about 25 min
under these reaction conditions.20 This is consistent with
the results shown in Table 1, wherein less reactive azides
gave lower yields. Increasing the reaction time and tem-
perature did not improve the yields of amide formation
for the less reactive azides. Efforts are underway to find
reaction conditions underwhich selenocarboxylates are
stable and thus can be used to facilitate the amidation
reaction with electron-rich azides.
In summary, we developed a method for the formation
of amide bonds under mild conditions through the reac-
tion of selenocarboxylates with azides. Excellent yields
were obtained when electron-deficient aromatic azides
were used. The reaction is highly chemoselective and
very clean with no detectable side products. It is com-
patible with a variety of functional groups including
hydroxy, ketone, carboxylic acid, ester, and nitrile. This
method offers the advantage of mild reaction conditions

X. Wu, L. Hu / Tetrahedron Letters 46 (2005) 8401–8405
8405
20. Stability measurement: The stability of benzeneseleno-
carboxylate 10 in DMSO/EtOAc (1:1) at room temper-
ature was measured by monitoring the product
formation between benzeneselenocarboxylate 10 and 4-
nitrophenyl azide 12 under the above general reaction
conditions. The reaction between benzeneselenocarboxyl-
ate 10 and 4-nitrophenyl azide 12 to form the product
amide 24 was complete in less than 5 min (data not
shown). Such fast product formation allows for the
accurate measurement of the amount of benzeneseleno-
carboxylate 10 remained in solution during our stability
study. Briefly, aliquots of freshly prepared benzenesele-
nocarboxylate 10 in DMSO/EtOAc (1:1) were incubated
at room temperature and the amount of benzeneseleno-
carboxylate 10 was measured at different time intervals
by the reaction with 2 equiv of 4-nitrophenyl azide (12)
for at least 1 h. The reaction mixtures were separated on
a Chromolith SpeedROD RP-18e column (50 · 4.6 mm)
at 1 mL/min with a 10-min gradient of 10–90% aceto-
nitrile containing 0.1% formic acid on a Shimadzu 2010
LCMS system. The N-(4-nitrophenyl) benzamide product
24 was quantitated based on UV absorption at 220 nm,
and its structure was confirmed by the mass spectrometry
detector. The peak area of N-(4-nitrophenyl) benzamide
product 24 represents the amount of benzeneselenocarb-
oxylate 10 that remained at a given time point and was
plotted against the time of incubation. The data was
fitted to a single, two parameter exponential decay
equation in SigmaPlot to obtain the first-order rate
constant.