An optimized and scalable synthesis of propylphosphonic anhydride for general use

Article abstract of DOI:10.1016/j.tetlet.2015.02.126

Abstract Propylphosphonic anhydride (T3P) is an effective coupling and dehydrating agent, which has been used for a large number of chemical transformations. An efficient and versatile synthetic method is described to synthesize propylphosphonic anhydride (T3P) in pure form, in an overall yield of 51% in four steps from commercially available diethyl phosphonate.

Full text of DOI:10.1016/j.tetlet.2015.02.126

Tetrahedron Letters 56 (2015) 2014–2017
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An optimized and scalable synthesis of propylphosphonic anhydride
for general use
⇑
Hana Pizova, Pavel Bobal
Department of Chemical Drugs, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Palackeho 1/3, 612 42 Brno, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Propylphosphonic anhydride (T3PÒ) is an effective coupling and dehydrating agent, which has been used
for a large number of chemical transformations. An efficient and versatile synthetic method is described
to synthesize propylphosphonic anhydride (T3PÒ) in pure form, in an overall yield of 51% in four steps
from commercially available diethyl phosphonate.
Article history:
Received 10 December 2014
Revised 15 February 2015
Accepted 26 February 2015
Available online 5 March 2015
Ó 2015 Elsevier Ltd. All rights reserved.
Keywords:
Propylphosphonic anhydride
T3PÒ
Amide coupling
Dehydrating agent
Michaelis–Becker reaction
Michaelis–Arbuzov reaction
Propylphosphonic anhydride (1), (PPAA, T3PÒ) was originally
developed as a coupling reagent for peptide synthesis,1,2 but it is
also used as a dehydrating agent with many advantages. These
include high reaction yields, low impurity profiles due to the water
solubility of the by-products formed from this reagent, and associ-
ated with easy work-ups. The reduced epimerization tendency and
low toxicity make propylphosphonic anhydride a reagent of choice,
even for large-scale synthesis.
Although propylphosphonic anhydride is commercially avail-
able as 50 wt % solution the solvent range is quite limited.
Generally it is sold in ethyl acetate andN,N-dimethylformamide
(DMF). But for many applications, these two solvents are not
appropriate. Ethyl acetate often contains significant amounts of
ethanol and acetic acid, which might interfere with the reacting
substances forming either ethyl esters or derivatives of acetic acid.
This can often lower the yield of the reaction. In addition, at ele-
vated temperatures, ethyl acetate might react with certain compo-
nents of the reaction mixture. The other solvent in which T3PÒ is
commercially available is N,N-dimethylformamide. This solvent
has many advantages over ethyl acetate such as better stability
to nucleophiles and better solubility properties. On the other hand,
liberation of dimethylamine from DMF under certain conditions
may hamper many reactions. Furthermore, removal of DMF from
the reaction mixture is rather difficult and often makes chromato-
graphic purification necessary. In general, one criterion for
selection of a suitable solvent for a reaction is its stability, that
is, the solvent does not react with the components in the reaction
mixture. Therefore, for many reactions, the solvent of choice is an
inert one. If we think about amide bond formation, solvents such as
toluene, methylene chloride, or ether-type solvents are appropri-
ate. This fact encouraged us to investigate the synthesis of T3PÒ
in neat form with further application in a solvent of choice.
However, the synthesis of T3PÒ in an economical fashion from
simple starting materials is difficult. Hence there is a need to
develop improved and convenient methods for the synthesis of
There are more than 200 scientific articles and patent applica-
tions published describing the utilization of T3PÒ for different
types of transformations. From this large number of papers we
must at first mention the use of T3PÒ as a peptide-coupling
promoter.1–4 It is superior to many widely used phosphorus-
containing coupling reagents.4 As a coupling reagent, it has also
found application in the synthesis of Weinreb amides,5,6 esters,7
b-lactams,8 hydroxamic acids,9 and acid azides.10 Several methods
utilize T3PÒ in dehydration processes, for example, the direct
conversion of aldehydes into nitriles,11 formamides into isonitriles
and carboxylic acids or amides into nitriles,12 and alcohols into
alkenes.13 The other applications of this reagent for the formation
of various heterocycles have been compiled in a review article.14
T3PÒ has also been successfully used as
a promoter of
Beckmann,15 Curtius,16 and Lossen17 rearrangements.
⇑
Corresponding author. Tel.: +420 54156 2927.
E-mail address: bobalp@vfu.cz (P. Bobal).
http://dx.doi.org/10.1016/j.tetlet.2015.02.126
0040-4039/Ó 2015 Elsevier Ltd. All rights reserved.

H. Pizova, P. Bobal / Tetrahedron Letters 56 (2015) 2014–2017
2015
propylphosphonic anhydride (T3PÒ) from readily available dialkyl
phosphonates.
O
O
O
Br
O
O
O
Besides Wissmann’s procedure1,2 the synthesis of T3PÒ 1 has
been described vaguely in the patent literature18–20 without any
optimized procedure (Scheme 1). Many patent procedures have
been tested without satisfactory results. In the original procedure1
for the synthesis of T3PÒ, propylphosphonic dichloride (2) was
used as the starting material. Its partial hydrolysis followed by
heating led to formation of T3PÒ 1, which was used without further
purification by distillation. The drawback of this procedure was the
liberation of a large amount of HCl and the fact that propylphos-
phonic dichloride (2) is not readily available nowadays. In addition,
the synthesis of propylphosphonic dichloride (2) using AlCl3 and
P
P
P
P
reflux
O
O
O
O
O
O
O
5
6
7
8
Scheme 2. Michaelis–Arbuzov reaction of 1-bromopropane with triethyl
phosphite.
O
R
O R
H
P
1. NaH, THF, 30-40 °C
2. n-PrBr, THF, reflux, 4 d
P
O
O
O
O
R
R
21
22
9
R = Me
10 R = Et
11 R = Me, 46%
PCl3 or PCl5 is rather unpleasant. The other method consists
of two steps.18 In the first step, propylphosphonic acid (3) is trea-
ted with acetic anhydride at elevated temperature to form an oli-
6
R = Et, 81%
Scheme 3. Michaelis–Becker reaction of dialkyl phosphonates.
gomeric phosphonic acid anhydride intermediate
which is then distilled under reduced pressure.
4 (Fig. 1),
Of the two methods, the latter would appear to be the more
convenient. Therefore our attention was focused on the
optimization of this procedure. The starting propylphosphonic acid
is available in a limited quantity as a laboratory reagent. So we
decided to start the synthesis from trialkyl phosphites or dialkyl
phosphonates.
O
R
OH
conc. HCl
reflux, 8 h
Method A
P
P
O
OH
O
O
R
3
11 R = Me
Method A: R = Me, 82%, mixture of solid and oil
R = Et, 88%, colorless solid
Method B: R = Me, 33%, mixture of solid and oil
R = Et, 82%, dark solid
6
R = Et
Traditionally, alkyl phosphonates are prepared via the
Michaelis–Arbuzov23,24 or the Michaelis–Becker reactions.25 Thus,
we initially tested the Michaelis–Arbuzov reaction23,24 of 1-bromo-
propane with triethyl phosphite (5). The reaction was carried out
from 70 °C to 90 °C. According to GC–MS analysis, formation of a
mixture of products occurred. Besides the desired diethyl propy-
lphosphonate (6) and residual phosphite 5, formation of diethyl
ethylphosphonate (7) and triethyl phosphate (8) as side products
was observed (Scheme 2). The first side product (7) was formed
from liberated bromoethane, and triethyl phosphate (8) was the
product of oxidation of triethyl phosphite (5). In order to reduce
the formation of side products the reaction was carried out with
a large excess of bromopropane, but the effect on the impurity pro-
file was negligible. Attempts to separate the products formed were
unsuccessful. We also tested the Michaelis–Becker reaction25
according to the patent procedure.22 Unfortunately, the reaction
of 1-bromopropane with dimethyl phosphonate (9) in the presence
of NaH as the base in anhydrous diethyl ether did not give the
desired intermediate. According to GC–MS analysis, formation of
dimethyl propylphosphonate (11) was not achieved, probably
due to the low reaction temperature or the poor solubility of the
O
TMS
H2O
TMSCl, NaBr
CH3CN, 60 °C, 3 h
P
O
O
TMS
12
Method B
Scheme 4. Scope of the hydrolysis of dialkyl phosphonates.
sodium salt of dimethyl phosphonate in diethyl ether. To overcome
this problem, the solvent was exchanged for anhydrous tetrahy-
drofuran. In this case, dimethyl propylphosphonate (11) containing
a small amount of methyl propyl propylphosphonate was isolated
in 46% yield (Scheme 3).
Hydrolysis of the phosphonate prepared was performed by two
methods (Scheme 4). The first tested (method A) was hydrolysis
with concentrated hydrochloric acid.22 During heating, the reac-
tion mixture became dark in color and the resulting acid was diffi-
cult to purify. All attempts to optimize the recrystallization failed
and the crystalline product 3 contained a large amount of a yellow
oily residue. Purification via the disodium salt together with
extraction with an organic solvent and subsequent conversion back
into the acid was successful, but lengthy and difficult, because the
acid was highly soluble in water. However, we managed to isolate
the acid 3 in 29% yield.
We next attempted to progress through transesterification to
silyl ester 12 and its subsequent hydrolysis (method B) to the acid
3.26–28 The reaction proceeded without problems, but we again
encountered the same problem with an oily residue present in
the crystalline propylphosphonic acid after isolation, and the yield
of this reaction was only 33%. Furthermore, the intermediate was
contaminated with iodine formed from the Me3SiCl/KI system used
in this transformation.
O
O
Cl
OH
P
P
O
P
P
O
O
Cl
OH
P
O
O
O
2
1
3
Scheme 1. Synthesis of propylphosphonic anhydride (T3PÒ).
O
P
O
P
O
P
A significant improvement was achieved when instead of
dimethyl phosphonate (11), the diethyl homologue 6 was used.
The Michaelis-Becker reaction of diethyl phosphonate (10) with
1-bromopropane proceeded without the formation of a mixed
ester and the product was isolated in 81% yield after distillation
(Scheme 3). The transesterification into silyl ester 12, followed by
hydrolysis in aqueous medium at room temperature led to an acid
in 71% yield (Scheme 4). The formation of iodine was eliminated by
HO
O
O
OH
n
4
Figure 1. The general structure of oligomeric phosphonic acid anhydride
intermediate 4.

2016
H. Pizova, P. Bobal / Tetrahedron Letters 56 (2015) 2014–2017
phosphonate, represents a superior alternative method to produce
this important green reagent in comparison with literature reports.
The prepared propylphosphonic anhydride either neat, or as a
solution in the solvent of a choice, shows activity comparable with
or better than commercial solutions. In addition, a few examples of
amide bond formation have been highlighted with very good
results. Investigations on the use of this reagent in amide coupling
for the synthesis of new biologically active compounds are under-
way in our laboratory.
O
P
O
P
O
P
OH
Ac2O
reflux, 24 h
P
HO
O
O
OH
OH
O
3
n
300 °C
4
0.3 mbar
distillation
O
O
P
P
P
O
O
O
O
Representative procedures are described
T3P 1 (72%)
Diethyl propylphosphonate (6).22 To
a mixture of 17.4 g
Scheme 5. Synthesis of propylphosphonic anhydride (1).
(0.73 mol) of NaH (free of mineral oil) in 100 ml of anhydrous
THF was added a solution of 50.0 g (0.36 mol) of dry diethyl
phosphonate (10) in 50 ml of anhydrous THF at 30–40 °C, and
the mixture was stirred overnight. Next, a solution of 49.0 g
(0.4 mol) of 1-bromopropane in 50 ml of anhydrous THF was
added dropwise and the mixture was heated under reflux for 4 d
under an argon atmosphere. To the mixture was slowly added
130 ml of H2O followed by extraction with Et2O (3 Â 150 ml). The
organic layer was dried over anhydrous MgSO4 and the solvents
were removed under reduced pressure. Distillation of the residue
gave 52.7 g (81%) of diethyl propylphosphonate (6).
H
OH
N
H2N
N
N
O
O
T3P, PhMe
80 °C, 24 h
Ph
O
O
Ph
O
O
16
13 86%
X
H2N
H
N
X
Propylphosphonic acid (3).22 Method A. A mixture of 20.00 g
(0.11 mol) of diethyl propylphosphonate (6) and 100 ml of concen-
trated HCl was heated under reflux for 8 h. Next, one half of
the volume of acid was removed by atmospheric distillation and
the remainder was evaporated to the dryness under reduced
pressure on a rotary evaporator. The crude solid was triturated
with hexane and filtered to yield 12.1 g (88%) of propylphosphonic
acid (3) as off-white crystals.
Method B. To a solution of 3.00 g (0.02 mol) of diethyl propy-
lphosphonate (6) in 10 ml of dry MeCN, 5.43 g (0.05 mol) of
TMSCl and 5.14 g (0.05 mol) of dry NaBr were added. The mixture
was heated at 60 °C for 3 h. Precipitated NaCl was filtered off and
the solvent and low boiling materials were evaporated under
reduced pressure. The residue was treated with 10 ml of H2O and
stirred at room temperature overnight. The mixture was washed
with hexane (2 Â 10 ml) and evaporated to dryness. The residue
was triturated with hexane and the remaining solid material was
filtered to give 1.46 g (71%) of propylphosphonic acid (3).
Propylphosphonic anhydride (1).18 Propylphosphonic acid (3)
10.00 g (0.08 mol) was dissolved in 53 ml (0.56 mol) of Ac2O in a
round-bottomed flask. The solution was heated under reflux for
24 h under an argon atmosphere. Residual Ac2O and AcOH were
evaporated under reduced pressure on a rotary evaporator and
the oligomeric phosphonic acid anhydride 4 was distilled using a
Kugelrohr apparatus at 300 °C/0.3 mbar to yield 6.17 g (72%) of
cyclic propylphosphonic anhydride (1) as a colorless viscous oil,
which could be dissolved in inert organic solvents with the aid of
sonication.
OH
T3P, PhMe
80 °C, 24 h
N
N
O
O
17
14 X = 4-Br, 68%
15 X = 3-F, 87%
Scheme 6. Examples of tested amide couplings.
using Me3SiCl/NaBr and the yield reached 82%, however, the purity
of the product was not satisfactory. On the other hand, simple
hydrolysis with concentrated hydrochloric acid was superior and
pure off-white propylphosphonic acid (3) was isolated in 88% yield.
The transformation into T3PÒ was carried out in refluxing acetic
anhydride for 24 h with subsequent distillation of oligomeric phos-
phonic acid anhydride intermediate4 (Scheme 5). We found that
reflux in acetic anhydride for 8 hours as described in a patent18
was not sufficient, so we extended the time to 24 h. Residual acetic
anhydride and acetic acid were removed under reduced pressure
on a rotary evaporator. Distillation was carried out in a simple
short-path bulb-to-bulb distillation apparatus (Kugelrohr) at a
pressure of 0.3 mbar. In order to achieve a sufficient degree of
cyclization to propylphosphonic anhydride (1), the temperature
during the distillation had to reach 300 °C.
Propylphosphonic anhydride (1) was isolated as a colorless,
highly viscous oil and could be stored in pure form in a closed con-
tainer for several months without decomposition. After warming it
could be removed from the container and transferred to a reaction
flask. Compatible solvents such as toluene, tetrahydrofuran,
methylene chloride, etc., can be added and the reagent applied
for further use as a solution. Standard 50 wt % solutions were
tested in amide formation. We have demonstrated the efficacy of
the T3PÒ prepared in this manner for the synthesis of heterocyclic
anilides13–15 and have compared this method with traditional
procedures.29–31 The results were compared with those cited in
the literature and it was found that the anilides (Scheme 6) pre-
pared using this protocol were obtained in comparable or
improved yields and purities.
Acknowledgements
Support for this work was provided by the project IGA VFU Brno
108/2013/FaF.
Supplementary data
In conclusion, we have developed an optimized and versatile
method for the synthesis of cyclic propylphosphonic anhydride
(T3PÒ), which can be conveniently applied to a wide variety of
chemical transformations. This simple process, which features four
simple high-yielding steps from commercially available diethyl
Supplementary data (Further experimental details, GC data,
NMR and MS spectra for compounds 1, 3, 6 and 13–15.) associated
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.tetlet.2015.02.126.

H. Pizova, P. Bobal / Tetrahedron Letters 56 (2015) 2014–2017
2017
15. Augustine, J. K.; Kumar, R.; Bombrun, A.; Mandal, A. B. Tetrahedron Lett. 2011,
52, 1074.
References and notes
16. Augustine, J. K.; Bombrun, A.; Mandal, A. B.; Alagarsamy, P.; Atta, R. N.; Selvam,
P. Synthesis 2011, 1477.
1. Wissmann, H.; Kleiner, H. J. Angew. Chem., Int. Ed. Engl. 1980, 19, 133.
2. Wissmann, H. Phosphorus Sulfur 1987, 30, 645.
3. Escher, R.; Bünning, P. Angew. Chem., Int. Ed. Engl. 1986, 25, 277.
4. Rzepecki, P.; Gallmeier, H.; Geib, N.; Cernovska, K.; Koenig, B.; Schrader, T. J.
Org. Chem. 2004, 69, 5168.
5. Burkhart, F.; Hoffmann, M.; Kessler, H. Angew. Chem., Int. Ed. Engl. 1997, 36,
1191.
6. Sharnabai, K. M.; Nagendra, G.; Vishwanatha, T. M.; Sureshbabu, V. V.
Tetrahedron Lett. 2013, 54, 478.
7. Wedel, M.; Walter, A.; Montforts, F. P. Eur. J. Org. Chem. 2001, 1681.
8. Crichfield, K. S.; Hart, J. E.; Lampert, J. T.; Vaid, R. K. Synth. Commun. 2000, 30,
3737.
9. Ech-Chahad, A.; Minassi, A.; Berton, L.; Appendino, G. Tetrahedron Lett. 2005,
46, 5113.
10. Basavaprabhu; Narendra, N.; Lamani, R. S.; Sureshbabu, V. V. Tetrahedron Lett.
2010, 51, 3002.
11. Augustine, J. K.; Atta, R. N.; Ramappa, B. K.; Boodappa, C. Synlett 2009, 3378.
12. Meudt, A.; Scherer, S.; Boehm, C. WO2005123661, 2005; Chem. Abstr. 2005,
144, 69403.
17. Vasantha, B.; Hemantha, H. P.; Sureshbabu, V. V. Synthesis 2010, 2990.
18. Wehner, M.; Kirschbaum, B.; Deutscher, L.; Wagner, H. J.; Hoessl, H.
WO2005014604, 2005; Chem. Abstr. 2005, 142, 198208.
19. Cumming, G. R.; Fuller, G. WO2010001085, 2010; Chem. Abstr. 2010, 152,
119820.
20. Mou, Y.; Zhang, Z.; Li, X.; Lu, Y.; Zhang, G. CN 103483386, 2014; Chem. Abstr.
2014, 160, 190243.
21. Crofts, P. C.; Kosolapoff, G. M. J. Am. Chem. Soc. 1953, 75, 3379.
22. Hill, K. L.; McCarthy, J. F. US 3837834, 1974; Chem. Abstr. 1974, 82, 12288.
23. Michaelis, A.; Kaehne, R. Ber. Dtsch. Chem. Ges. 1898, 31, 1048.
24. Arbuzov, A. E. J. Russ. Phys. Chem. Soc. 1906, 38, 687.
25. Michaelis, A.; Becker, T. Ber. Dtsch. Chem. Ges. 1897, 30, 1003.
26. Katritzky, A. R.; Pilarski, B.; Johnson, J. W. Org. Prep. Proced. Int. 1990, 22, 209.
27. Rabinowitz, R. J. Org. Chem. 1963, 28, 2975.
28. Morita, T.; Okamoto, Y.; Sakurai, H. Bull. Chem. Soc. Jpn. 1981, 54, 267.
29. Bobal, P.; Sujan, J.; Otevrel, J.; Imramovsky, A.; Padelkova, Z.; Jampilek, J.
Molecules 2012, 17, 1292.
30. Gonec, T.; Bobal, P.; Sujan, J.; Pesko, M.; Guo, J.; Kralova, K.; Pavlacka, L.; Vesely,
L.; Kreckova, E.; Kos, J.; Coffey, A.; Kollar, P.; Imramovsky, A.; Placek, L.;
Jampilek, J. Molecules 2012, 17, 613.
13. Meudt, A.; Scherer, S.; Boehm, C. WO2005123632, 2005; Chem. Abstr. 2005,
144, 69544.
14. Basavaprabhu; Vishwanatha, T. M.; Panguluri, N. R.; Sureshbabu, V. V. Synthesis
2013, 45, 1569.
31. Mukaiyama, T. Tetrahedron 1981, 37, 4111.