Acid anhydrides and the unexpected N,N-diethylamides derived from the reaction of carboxylic acids with Ph3P/I2/Et3N

Article abstract of DOI:10.1016/j.tetlet.2015.12.009

The formation of acid anhydrides from the phosphorous-mediated activation of carboxylic acids was investigated. Under various systems, activation of benzoic acid in the presence of base led to the formation of benzoic anhydride at different rates depending on the reactivity of the reagents. Using the Ph₃P-I₂/Et₃N combination, most aryl acids were converted into the corresponding anhydrides in high yields within 5-10 min. However, for nitro-substituted derivatives, unexpectedly, N,N-diethylamides were isolated without anhydride formation. These results indicated the pronounced effect of substituents in governing these potential side reactions which can significantly affect the yields of acylation reactions promoted by phosphonium species.

Full text of DOI:10.1016/j.tetlet.2015.12.009

Accepted Manuscript
Acid anhydrides and the unexpected N,N-diethylamides derived from the reac-
tion of carboxylic acids with Ph3P/I2/Et3N
Wong Phakhodee, Chuthamat Duangkamol, Sirilak Wangngae, Mookda
Pattarawarapan
PII:
S0040-4039(15)30424-X
DOI:
http://dx.doi.org/10.1016/j.tetlet.2015.12.009
TETL 47056
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
29 October 2015
30 November 2015
2 December 2015
Please cite this article as: Phakhodee, W., Duangkamol, C., Wangngae, S., Pattarawarapan, M., Acid anhydrides
and the unexpected N,N-diethylamides derived from the reaction of carboxylic acids with Ph3P/I2/Et3N, Tetrahedron
Letters (2015), doi: http://dx.doi.org/10.1016/j.tetlet.2015.12.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Acid anhydrides and the unexpected N,N-diethylamides derived from the reaction of
carboxylic acids with Ph3P/I2/Et3N
Wong Phakhodee,* Chuthamat Duangkamol, Sirilak Wangngae, and Mookda Pattarawarapan
Department of Chemistry and Center of Excellence for Innovation in Chemistry, Faculty of
Science, Chiang Mai University, Chiang Mai 50200, Thailand
To whom correspondence should be addressed.
Department of Chemistry,
Faculty of Science,
Chiang Mai University,
Chiang Mai
50200
Thailand
Tel. +66 5394 3341
Fax. +66 5389 2277
e-mail : wongp2577@gmail.com

Acid anhydrides and the unexpected N,N-diethylamides derived from the reaction of
carboxylic acids with Ph3P/I2/Et3N
Wong Phakhodee,* Chuthamat Duangkamol, Sirilak Wangngae, and Mookda Pattarawarapan
Department of Chemistry and Center of Excellence for Innovation in Chemistry, Faculty of
Science, Chiang Mai University, Chiang Mai 50200, Thailand
Abstract: The formation of acid anhydrides from the phosphorous-mediated activation of
carboxylic acids was investigated. Under various systems, activation of benzoic acid in the
presence of base led to the formation of benzoic anhydride at different rates depending on the
reactivity of the reagents. Using the Ph3P-I2/Et3N combination, most aryl acids were
converted into the corresponding anhydrides in high yields within 5-10 min. However, for
nitro-substituted derivatives, unexpectedly, N,N-diethylamides were isolated without
anhydride formation. These results indicated the pronounced effect of substituents in
governing these potential side reactions which can significantly affect the yields of acylation
reactions promoted by phosphonium species.
Keywords: substituent effects; triphenylphosphine; iodine; anhydrides; amides
Introduction
Carboxylic acid anhydrides are an important class of reagents and are highly useful as
acylating agents as well as key intermediates in organic synthesis.1 They are usually prepared
by the reaction of carboxylate salts with acylating agents such as acid chlorides,2 coupling of
acid chlorides3 or through a dehydrative coupling reaction of carboxylic acids using reagents
such as thionyl chloride,4 oxalyl chloride,5 triphosgene,6 triazine reagents,7
trichloroisocyanuric acid in combination with triphenylphosphine (Ph3P),8 and carbodiimide.9
While various reagent systems have been developed for the preparation of anhydrides, their
formation are often unwanted side reactions which are encountered during the pre-activation
of a carboxylic acid for an acyl transfer process.10
In the phosphorous-mediated activation of a carboxylic acid, an acyloxyphosphonium
salt is commonly proposed as the key intermediate produced upon treatment of a carboxylic
acid with an in-situ generated phosphonium ions.10e-g This highly reactive species could

undergo nucleophilic attack with a carboxylate anion to provide the anhydride by-product
before reacting with the desired nucleophile. Such side reactions could lead to low yield, long
reaction time, harsh reaction conditions or even inexplicable results.
Under the well-known Mitsunobu conditions (Ph3P/DEAD), it has been suggested that
one of the reasons for the lower yields of esters from starting acids with higher pKa is due to
competitive acid anhydride formation via acyloxyphosphonium salts.10d, 11 Indeed, in a study
by Jenkins and co-workers,12 it was found that benzoic acid gave a high yield of the
corresponding anhydride, whereas the more acidic 4-nitrobenzoic acid did not lead to any of
the anhydride side product.
To explore whether a similar outcome persists with other related phosphine reagent
systems and to better explain the outcome of the related acylation reactions, herein, anhydride
formation under various phosphorous-mediated carboxylic acid activation conditions was
thoroughly investigated using a variety of aromatic acids with a range of pKa values.
In our initial investigation, benzoic acid was used as a model substrate and various
factors affecting anhydride formation including the type of base and halogenated additives
were examined. The effect of tertiary amine base was first compared using the combination
of Ph3P-iodine as the acid activation system. As shown in Table 1, benzoic anhydride was
i
quantitatively obtained within ca. 5 min using either Et3N or Pr2NEt (entries 1 and 2).
Replacing these bases with imidazole gave a lower yield of the corresponding anhydride with
a longer reaction time (entry 3).
Various halogenated additives were evaluated in the presence of Et3N. Under similar
conditions, N-bromosuccinimide (NBS) gave a lower yield of the anhydride relative to that
using I2 (entry 4 vs entry 1). When Ph3P-N-chlorosuccinimide (NCS) or Ph3P-
carbontetrachloride (CCl4) combinations were used, the reaction was sluggish, requiring
longer reaction times or increased temperature (entries 5 and 6). It was noted that using the
reported Ph3P-trichloroacetonitrile (CCl3CN) system (entry 7),13 a high yield of benzoic
anhydride (92%) was obtained after 1 h at room temperature although, in this case, an acyl
chloride was proposed as the key intermediate.

Table 1. Formation of benzoic anhydride using different reagent systemsa
time
entry
Reagent
base
yield (%)
(min)
1
2
3
4
5
6
7
Ph3P/I2
Et3N
5
99
Ph3P/I2
iPr2NEt
imidazole
Et3N
5
97
Ph3P/I2
30
5
85
Ph3P/NBS
Ph3P/NCS
Ph3P/CCl4
Ph3P/CCl3CN
92
Et3N
30
60
60
88
Et3N
87b
9013
Et3N
aReaction conditions: benzoic acid (0.41 mmol), Ph3P (0.62 mmol),
halogenated reagent (0.62 mmol), base (1.23 mmol), CH2Cl2 (2
mL), 25 °C. bThe reaction was carried out at 60 °C.
To study the substituent effect on this transformation, we systematically investigated
anhydride formation using the Ph3P-I2/Et3N conditions using various aromatic acids with a
range of pKa values. According to Table 2, weaker acids (larger pKa value) such as benzoic
acid and those derivatives containing electron-donating substituents (entries 1-8) reacted
extremely well to provide the corresponding anhydrides in high yields within 5-10 min. Steric
factors had little influence on the reaction and the presence of a substituent at the ortho
position did not significantly lower the yields (see entries 2, 6). Stronger acids (smaller pKa
value) such as chloro-substituted derivatives (entries 9-11) were also smoothly converted to
the corresponding anhydrides. However, using substrates bearing the acetyloxy or hydroxy
groups (entries 12-13) led to slightly lower conversion. Apart from benzoic acid derivatives,
other aryl acids such as naphthyl derivatives as well as a conjugated system (entries 14-16)
also reacted rapidly.

Table 2. Ph3P-I2 mediated formation of symmetric anhydrides[a]
entry
1
R
pKa of 114
4.19
4.09
4.47
4.36
3.97
3.98
4.27
4.37
2.89
2.81
3.98
3.49
2.97
3.69
4.17
4.44
3.44
yield of 2 (%)
C6H5
98
96
92
72
90
83
89
90
90
89
99
72
57
87
89
83
0
2
2-CH3OC6H4
4-CH3OC6H4
3,4-(CH3O)2C6H3
3,5-(CH3O)2C6H3
2-CH3C6H4
3-CH3C6H4
4-CH3C6H4
2-ClC6H4
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
3-ClC6H4
4-ClC6H4
2-AcOC6H4
2-HOC6H4
1-naphthyl
2-naphthyl
PhCH=CH
4-NO2C6H4
[a]Reaction conditions: carboxylic acid (0.41 mmol), Ph3P (0.62
mmol), I2 (0.62 mmol), Et3N (1.23 mmol), CH2Cl2 (2 mL), 0 °C -
RT, 5-10 min.
In contrast to the results obtained with most aryl acids, aromatic acids bearing nitro
group(s) were found to behave differently. When 4-nitrobenzoic acid was subjected to the
above described conditions (Table 2, entry 17), no trace of the respective anhydride was
detected (TLC) which was in accordance with those reported under Mitsunobu conditions.12
Adjusting the reaction conditions by increasing the amount of the reagents, changing the base
or solvents, or increasing the reaction temperature did not lead to any detectable anhydride.
We attempted to isolate the product from the reaction of 4-nitrobenzoic acid with polymer-
bound Ph3P-I2 and Et3N; however we were also unable to obtain any anhydride product

following the work-up procedures described in a reported literature.15 Thus, it is plausible
that the corresponding acyloxyphosphonium salt necessary for anhydride formation was not
formed or rapidly decomposed to another less reactive intermediate.
To our surprise, it was found that when the standard reaction conditions were applied for
a longer period of time, N,N-diethyl-4-nitrobenzamide was observed after a few hours and
significant amounts of the amide could be isolated after 16 h (Table 3, entry 1). Similar
results were also obtained when the experiment was repeated with other nitro-substituted aryl
acids. As shown in Table 3, all of the test substrates could be converted into the
corresponding amide products to different extents at various rates (the yields were un-
optimized). Notably, the respective amide from 2-chloro-5-nitrobenzoic acid was produced in
high yield within 3 h. Since there was no other source of diethylamine except from
triethylamine, it is possible that the N,N-diethylamides could be derived from amidation of
the acids with Et3N via a highly reactive intermediate(s) where both coupling partners are
activated toward C-N bond formation.
Table 3. Unexpected formation of N,N-diethylamides 3 from nitro-substituted aryl acidsa
entry
R
pKa of 114
3.44
yield of 3 (%)
1
2
3
4
5
4-NO2C6H4
63
45
52
50
75
3-NO2C6H4
3.46
3,5-(NO2)2C6H3
2-NO2C6H4
2.82
2.17
2-Cl-5-NO2C6H3
2.17
aReaction conditions: carboxylic acid (0.41 mmol), Ph3P (0.62
mmol), I2 (0.62 mmol), Et3N (1.23 mmol), CH2Cl2 (2 mL), 0 °C-
RT, 3-16 h.
To determine whether amide formation was correlated to the pKa of the acids, control
experiments were carried out using 3-chlorobenzoic acid and benzoic acid as substrates. It
was found that, after 16 h reaction, the reaction of 3-chlorobenzoic acid produced the
respective amide in a trace amount, whereas the product from the reaction of benzoic acid

was not observed. This data indicated that amide formation is possible when the formed
intermediate is sufficiently reactive due to possessing an electron withdrawing group.
Although the direct amidation of carboxylic acids with tertiary amines has been
previously reported, the reaction required metal catalysis and proceeded under more vigorous
conditions.16 Thus, our system is the first reported formation of N,N-substituted amides from
carboxylic acids using tertiary amine nucleophile under metal-free and mild conditions.
It should be noted that since pKa measures equilibrium in aqueous media, it can only be
used for a rough estimation regarding the acidity of the carboxylic acids and may not ideally
represent their reactivity. Some observed deviation from the trend could result from solvation
effects which significantly affects nucleophilicity and the leaving group ability of the formed
carboxylate ions. Nevertheless, based on the results observed in Tables 2 and 3, some
correlation could be made between the pKa values of the acids and their ability to form
anhydrides or amides. Most of the aryl acids weaker than 4-nitrobenzoic acid generally
produce the corresponding anhydrides rapidly, whereas stronger acids could give rise to the
N,N-diethylamides. Both reactions are expected to take place in competition with the
acylation reaction involving weak nucleophiles when Et3N is applied as base. However,
competing amidation with the weakly nucleophilic base would be less likely in the presence
of other commonly used nucleophiles. Thus, the acylation reaction of acidic substrates is
most likely to give significantly improved yields in comparison with those using other less
acidic derivatives where the competitive anhydride side reaction would be predominant.
Based on our previous experience and the presented results, a mechanism for the Ph3P-I2
mediated reaction toward anhydrides and formation of the unexpected amides was proposed
(Scheme 1). Addition of iodine to Ph3P results in formation of a pentavalent
triphenylphosphinediiodide I and triphenylphosphonium iodide II. Subsequent treatment with
the carboxylic acid and Et3N, produces a pentacoordinate intermediate III which is in
equilibrium with acyloxyphosphonium IV.17 With weak acid substrates, the formed
carboxylate anions are more nucleophilic, and thus could undergo rapid acyl substitution with
IV to yield the anhydride product. However, with strong aryl acids containing an electron
withdrawing group, their corresponding carboxylates are weak nucleophiles. It is highly
likely that, once IV is produced, the highly electrophilic carbonyl groups of IV could undergo
nucleophilic substitution with triethylamine with the thermodynamically favourable release
of triphenylphosphine oxide leading to the respective amide upon removal of ethyl iodide.

The possibility for acyl substitution of the reactive anhydride with Et3N could also be
envisaged.
It is also possible that III could undergo iodide substitution with triethylamine leading to
intermediate V. Upon de-ethylation, the pentacoordinate phosphorane VI is generated. A
Berry pseudorotation of the groups around the phosphorus center of VI then bring the
diethylamino and acyl groups in close proximity.18 Subsequent intramolecular substitution at
the acyl functional group could produce the tertiary amide with the release of
31
triphenylphosphine oxide as a strong driving force. To confirm the proposed mechanism, P
13
and C NMR studies of the formed intermediates need to be carried out which will be the
subject of future work.
Scheme 1. Proposed mechanism for the Ph3P-I2 mediated anhydride and amide bond
formation.
In summary, we report a systematic study toward anhydride formation using
phosphorous-mediated carboxylic acid activation. It was found that the reaction was highly
dependent on the electronic nature of the carboxylic acid substrates as reflected by their pKa
values. With the Ph3P-I2/Et3N system, most aromatic acids, except for the more acidic nitro-
substituted derivatives, could give rise to anhydrides in high yields within short reaction
times, indicating that competitive anhydride formation could be problematic in acylation
reactions involving weak nucleophiles. In such cases, anhydride formation could potentially
be avoided or minimized if the base is added after introducing the nucleophile.17a The

unprecedented finding regarding with the unexpected formation of amides may also lead to
new interesting organic syntheses using inexpensive, stable, and easy to handle tertiary
amines as nucleophiles. Further studies on this amidation reaction are being pursued and will
be reported in due course.
Acknowledgements
Financial support from the Thailand Research Fund through the Royal Golden Jubilee Ph.D.
Program to S.W. (Grant No. PHD/0206/2556) and C.D. (Grant No. PHD/0086/2557) as well
as the Center of Excellence for Innovation in Chemistry (PERCH-CIC), Chiang Mai
University, Thailand are gratefully acknowledged.
Supplementary Material
1
General experimental method, spectroscopic data for all the compounds, and H and 13C
NMR spectra of representative products.
References
1.
(a) El-Faham, A.; Albericio, F. Chem. Rev. 2011, 111, 6557-6602.(b) Shaaya, O.;
Magora, A.; Sheskin, T.; Kumar, N.; Domb, A. J. Pharm. Res. 2003, 20, 205-211. (c)
Tachibana, Y.; Kawasaki, H.; Kihara, N.; Takata, T. J. Org. Chem. 2006, 71, 5093-
5104. (d) Funasaka, S.; Mukaiyama, T. Bull. Chem. Soc. Jpn. 2008, 81, 148-159. (e)
Luo, F.; Pan, C.; Qian, P.; Cheng, J. J. Org. Chem. 2010, 75, 5379-5381. (f) Cirriez,
V.; Rasson, C.; Riant, O. Adv. Synth. Catal. 2013, 355, 3137-3140.
(a) Rambacher, P.; Maeke, S. Angew. Chem. Int. Ed. Engl. 1968, 7, 465. (b) Hajipour,
A. R.; Mazloumi, G. Synth. Commun. 2002, 32, 23-30. (c) Fife, W. K.; Zhang, Z. D.
Tetrahedron Lett. 1986, 27, 4933-6. (d) Leadbeater, N. E.; Scott, K. A. J. Org. Chem.
2000, 65, 4770-4772.
(a) Kim, J.-G.; Jang, D. O. Bull. Korean Chem. Soc. 2009, 30, 27-28. (b) Plusquellec,
D.; Roulleau, F.; Lefeuvre, M.; Brown, E. Tetrahedron 1988, 44, 2471-6. (c) Wang,
J.-X.; Hu, Y.; Cui, W. Synth. Commun. 1994, 24, 3261-8. (d) Roulleau, F.;
Plusquellec, D.; Brown, E. Tetrahedron Lett. 1983, 24, 4195-6. (e) Serieys, A.;
Botuha, C.; Chemla, F.; Ferreira, F.; Perez-Luna, A. Tetrahedron Lett. 2008, 49,
5322-5323.
2.
3.
4.
(a) Adduci, J. M.; Ramirez, R. S. Org. Prep. Proced. 1970, 2, 321, 323-5. n(b)
Kazemi, F.; Kiasat, A. R. Phosphorus, Sulfur Silicon Relat. Elem. 2003, 178, 2287-
2291.(c) Kazemi, F.; Kiasat, A. R.; Mombaini, B. Synth. Commun. 2007, 37, 3219-
3223.
5.
6.
7.
Rinderknecht, H.; Ma, V. Helv. Chim. Acta 1964, 47, 162-5.
Kocz, R.; Roestamadji, J.; Mobashery, S. J. Org. Chem. 1994, 59, 2913-14.
Kaminski, Z. J.; Kolesinska, B.; Malgorzata, M. Synth. Commun. 2004, 34, 3349-
3358.
8.
9.
Rouhi-Saadabad, H.; Akhlaghinia, B. Chem. Pap. 2015, 69, 479-485.
Hata, T.; Tajima, K.; Mukaiyama, T. Bull. Chem. Soc. Jap. 1968, 41, 2746-7.

10.
11.
(a) McNulty, J.; Capretta, A.; Laritchev, V.; Dyck, J.; Robertson Al, J. J Org Chem
2003, 68, 1597-600. (b) Morcillo, S. P.; Alvarez de Cienfuegos, L.; Mota, A. J.;
Justicia, J.; Robles, R. J. Org. Chem. 2011, 76, 2277-2281. (c) Bae, S.; Lakshman, M.
K. J. Org. Chem. 2008, 73, 1311-1319. (d) Ahn, C.; Correia, R.; DeShong, P. J. Org.
Chem. 2002, 67, 1751-1753. (e) Xu, S.; He, Z. RSC Adv. 2013, 3, 16885-16904. (f)
Pedrosa, L. F. Synlett 2008, 1581-1582. (g) But, T. Y. S.; Toy, P. H. Chem. - Asian J.
2007, 2, 1340-1355.
(a) Hughes, D. L.; Reamer, R. A. J. Org. Chem. 1996, 61, 2967-71. (b) Dodge, J. A.;
Trujillo, J. I.; Presnell, M. J. Org. Chem. 1994, 59, 234-6. (c) Grochowski, E.;
Stepowska, H.; Michejda, C. J. Bull. Pol. Acad. Sci., Chem. 1984, 32, 129-34.
Harvey, P. J.; von Itzstein, M.; Jenkins, I. D. Tetrahedron 1997, 53, 3933-3942.
Kim, J.; Jang, D. O. Synth. Commun. 2001, 31, 395-399.
12.
13.
14.
Harding, A. P.; Wedge, D. C.; Popelier, P. L. A. J. Chem. Inf. Model. 2009, 49, 1914-
1924.
15.
16.
Metay, E.; Leonel, E.; Condon, S.; Nedelec, J.-Y. Tetrahedron 2006, 62, 8515 -8524.
(a) Xiong, B.; Zhu, L.; Feng, X.; Lei, J.; Chen, T.; Zhou, Y.; Han, L.-B.; Au, C.-T.;
Yin, S.-F. Eur. J. Org. Chem. 2014, 2014, 4244-4247. (b) Zhang, X.; Yang, W.;
Wang, L. Org. Biomol. Chem. 2013, 11, 3649-3654. (c) Li, Y.; Ma, L.; Jia, F.; Li, Z.
J. Org. Chem. 2013, 78, 5638-5646.
17.
18.
(a) Wangngae, S.; Duangkamol, C.; Pattarawarapan, M.; Phakhodee, W. RSC Adv.
2015, 5, 25789-25793. (b) Duangkamol, C.; Wangngae, S.; Pattarawarapan, M.;
Phakhodee, W. Eur. J. Org. Chem. 2014, 7109-7112.
Schenk, S.; Weston, J.; Anders, E. J. Am. Chem. Soc. 2005, 127, 12566-12576.

Graphical Abstract
Acid anhydrides and the unexpected N,N-diethylamides Leave this area blank for abstract info.
derived from the reaction of carboxylic acids with
Ph3P/I2/Et3N
Wong Phakhodee,* Chuthamat Duangkamol, Sirilak Wangngae, and Mookda Pattarawarapan