1-(N-Acylamino)-1-triphenylphosphoniumalkylphosphonates: General synthesis and prospects for further synthetic applications

Article abstract of DOI:10.1016/j.tetlet.2018.07.040

A general approach for the synthesis of 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonates from readily accessible alkyl imidate hydrochlorides has been developed. The three-step synthesis involves acylation of the imidate hydrochloride with an acyl chloride, the Michaelis–Becker-like addition of diethyl phosphite to the N–acylimidate and subsequent nucleophilic substitution of the ethoxy group of the 1–ethoxyphosphonate derivative with triphenylphosphonium tetrafluoroborate. 1–(N–Acylamino)-1-triphenylphosphoniumalkylphosphonates were demonstrated to be promising intermediates for further synthetic transformations toward α-functionalized derivatives of α–aminophosphonic acids and α,β-dehydro-α-aminophosphonates.

Full text of DOI:10.1016/j.tetlet.2018.07.040

Accepted Manuscript
1-(N-Acylamino)-1-triphenylphosphoniumalkylphosphonates: general synthe-
sis and prospects for further synthetic applications
Anna Kuźnik, Roman Mazurkiewicz, Magdalena Zięba, Karol Erfurt
PII:
S0040-4039(18)30913-4
https://doi.org/10.1016/j.tetlet.2018.07.040
TETL 50149
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
1 June 2018
5 July 2018
16 July 2018
Please cite this article as: Kuźnik, A., Mazurkiewicz, R., Zięba, M., Erfurt, K., 1-(N-Acylamino)-1-
triphenylphosphoniumalkylphosphonates: general synthesis and prospects for further synthetic applications,
Tetrahedron Letters (2018), doi: https://doi.org/10.1016/j.tetlet.2018.07.040
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.

Graphical Abstract

1-(N-Acylamino)-1-triphenylphosphoniumalkylphosphonates: general
synthesis and prospects for further synthetic applications
Anna Kuźnik*,a,b Roman Mazurkiewicz,a,b Magdalena Zięba,a Karol Erfurtc
Dedicated to the memory of Professor Jerzy Suwiński (1939-2017)
aDepartment of Organic and Bioorganic Chemistry and Biotechnology, Silesian University of Technology,
B. Krzywoustego 4, 44-100 Gliwice, Poland
bBiotechnology Center of Silesian University of Technology, B. Krzywoustego 8, 44-100 Gliwice, Poland
cDepartment of Organic Chemical Technology and Petrochemistry, Silesian University of Technology, B. Krzywoustego 4,
44-100 Gliwice, Poland
* Corresponding author. e-mail: anna.kuznik@polsl.pl; Tel: +48 32 237 16 13; Fax: +48 32 237 20 94.
Abstract
A
general
approach
for
the
synthesis
of
1-(N-acylamino)-1-
triphenylphosphoniumalkylphosphonates from readily accessible alkyl imidate hydrochlorides
has been developed. The three-step synthesis involves acylation of the imidate hydrochloride
with an acyl chloride, the Michaelis–Becker-like addition of diethyl phosphite to the
N-acylimidate and subsequent nucleophilic substitution of the ethoxy group of the
1-ethoxyphosphonate
derivative
with
triphenylphosphonium
tetrafluoroborate.
1-(N-Acylamino)-1-triphenylphosphoniumalkylphosphonates were demonstrated to be
promising intermediates for further synthetic transformations toward α-functionalized
derivatives of α-aminophosphonic acids and α,-dehydro-α-aminophosphonates.
Keywords: N-acylimidates, 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonates,
α,-dehydro-α-aminophosphonates, bisphosphonates
Introduction
-Aminophosphonic acids 1, as phosphorus analogues and mimetics of -amino acids,
exhibit a variety of important biological activities.1–4 In the last decade our research group has
developed effective methods for the synthesis of 1-(N-acylamino)alkylphosphonium salts 2
and demonstrated their efficiency as highly reactive -amidoalkylating agents in reactions
with
a
variety
of
nucleophiles.5–7
1-(N-Acylamino)-1-
triphenylphosphoniumalkylphosphonates 4 can be considered both as a subclass of 1-(N-
acylamino)alkylphosphonium salts 2 and phosphonium derivatives of -aminophosphonic
acids 1, as they include both the triphenylphosphonium group of phosphonium salts 2 and the

phosphonate group of compounds 1 (Fig.1). It could be assumed that nucleophilic substitution
of the triphenylphosphonium group at the -position of compounds 4 would enable the simple
synthesis of variously -functionalized -aminophosphonic acid derivatives. It could also be
expected that the Wittig reaction of phosphonium salts 4 with a proton at the α-position (R2 =
H) would enable the direct synthesis of important -dehydro--aminophosphonic acid
derivatives. The elimination of triphenylphosphonium tetrafluoroborate from phosphonium
salts 4 with a proton at the -position (R2 = R6R7CH) represents another pathway towards
-dehydro--aminophosphonic acid derivatives.
Figure 1. General structures of -aminophosphonic acids 1 and their -functionalized derivatives 3, 5
as
well
as
1-(N-acylamino)triphenylphosphonium salts
2
and 1-(N-acylamino)-1-
triphenylphosphoniumalkylphosphonates 4.
Recently, we reported a method for the synthesis of hitherto unknown diethyl 1-(N-
acylamino)-1-triphenylphosphoniumalkylphosphonates 4 by electrochemical oxidative -
methoxylation of diethyl 1-(N-acetylamino)alkylphosphonates 6 to 1-(N-acetylamino)-1-
methoxyalkylphosphonates 7, followed by substitution of the methoxy group with
triphenylphosphonium tetrafluoroborate (Scheme 1).8 We also showed the usefulness of these
compounds in the synthesis of α-aminobisphosphonates 5 as well as their phosphonyl-
phosphinyl and phosphonyl-phosphinoyl unsymmetrical analogues.8 However, the reaction
scope for the synthesis of compounds 4 was proven to be limited. Electrochemical -
methoxylation of 1-(N-acetylamino)alkylphosphonates was possible only in the case of
compounds 6 with small substituents at the -position (R2 = H or Me). Attempts to extend

this method to 1-(N-acetylamino)alkylphosphonates with more bulky substituents at the -
position (R2 = Ph or iPr) failed, as electrochemical oxidation in methanol gave intractable
mixtures of many compounds in these cases.
Scheme 1. Synthesis of 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonates
4
via
electrochemical α-methoxylation of 1-(N-acylamino)alkylphosphonates 6 followed by substitution of
the methoxy group with triphenylphosphonium tetrafluoroborate.8
Herein, we report a more general strategy for the synthesis of 1-(N-acylamino)-1-
triphenylphosphoniumphosphonates 4 based on N-acylimidates 9 as starting compounds
(Scheme 2). The oxidation state of the imine carbon in N-acylimidates 9 and in the -carbon
in 1-(N-acylamino)-1-triphenylphosphoniumphosphonates 4 are the same, thus the difficult
oxidation step in the discussed syntheses has been omitted. The second part of this paper
demonstrates
the
potential
applications
of
1-(N-acylamino)-1-
triphenylphosphoniumalkylphosphonates as new reagents for the synthesis of variously
-functionalized -aminophosphonic acid derivatives.
Results and Discussion
Imidate salts 8 are a well-known class of compounds most commonly synthesized
from nitriles via the Pinner synthesis, but are also available from amides or lactams via
O-alkylation. Imidoesters, their less stable free bases, can be obtained from imidoyl halides by
reaction with alkoxide or phenolate anions.9–11 The acylation of N-unsubstituted imidate salts
with acetyl chloride in the presence of triethylamine to N-acetylimidates 9 was carried out by
modification of the procedures described by Alves and co-workers and Inoue and co-workers
(Scheme 2).12,13 In the case of N-acylformimidates, the acylation products were too unstable
to be isolated and were therefore reacted in situ.

Scheme 2. Synthesis of 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonates 4 from imidate
hydrochlorides 8 via 1-(N-acylamino)-1-ethoxyalkylphosphonates 3. Reagents and conditions: (i) acyl
chloride, Et3N, Ar atmosphere, 0-20 ºC, 2-18 h, 76-87%; (ii) HP(O)(OEt)2, K2CO3, 18-crown-6, 20 ºC,
4-120 h, 38-78%; (iii) Ph3P.HBF4, 20-85 ºC, 0.1-6 h, 82-96%.
The Michaelis-Becker-like addition of diethyl phosphite to N-acylimidates 9 under
PTC conditions in hexane in the presence of K2CO3 and the appropriate crown ether (18-
crown-6) at room temperature resulted in the expected 1-(N-acylamino)-1-
1
ethoxyphosphonates 3a-d (Scheme 2, Table 1). The reaction progress was monitored by H
and 31P NMR spectroscopy, thus determining the optimal reaction time. The obtained 1-
alkoxy derivatives of 1-aminophosphonic acid are a rare class of compounds. Pudovik and
co-workers described the synthesis of dimethyl and diethyl 1-(N-acetylamino)-1-ethoxy-1-
phenylmethylphosphonates (3, R1 = Me, R2 = Ph) via the addition of diethyl phosphite to the
corresponding N-acetylimidate in the presence of sodium methoxide or sodium ethoxide.14 In
contrast to the well-known electrochemical -alkoxylation of 1-(N-acylamino)acid esters,15
the analogous alkoxylation of 1-(N-acylamino)phosphonic acid esters 6 to 1-(N-acylamino)-1-
alkoxyphosphonic acid esters was reported in the literature only twice and, as was already
mentioned, the scope of application of the electrochemical alkoxylation of these compounds
proved to be very limited.8,16
Table 1. Synthesis of 1-(N-acylamino)-1-ethoxyalkylphosphonates 3 via the Michaelis-Becker-like
reaction.
Entry
R1
R2
Procedurea
Time [h]
Yield [%]
Me
CH2Ph
Me
H
H
Bb
Bb
A
18
18
4
49c
38c
78
3a
3b
3c
Me

Me
Ph
A
120
57
3d
a PTC reaction conditions: ethyl N-acylimidate (2 mmol), diethyl phosphite (2.4 mmol), K2CO3 (2.72 mmol),
b
18-crown-6 (0.27 mmol), hexane (6.4 mL), 20 °C. Procedure B differs from procedure A in that ethyl N-
acylformimidates were in situ subjected to the Michaelis-Becker-like reaction. c The yield of product was
calculated relative to the starting formimidate hydrochloride due to the instability of the intermediate N-acyl
derivative.
As recently demonstrated, the electrophilic reactivity of the -carbon in
1-(N-acylamino)-1-alkoxyalkylphosphonates 7 was relatively low.8 Our attempts to transform
these compounds to the corresponding -(N-acylamino)bisphosphonates 5 in the Michaelis–
Arbuzov-like reaction with triethyl phosphite failed.8 To enhance the electrophilic activity of
the 1-ethoxy derivatives of 1-aminophosphonic acid, the 1-ethoxy group was successfully
displaced by the triphenylphosphonium group in the presence of catalytic
methyltriphenylphosphonium iodide, thus obtaining the hitherto unknown 1-(N-acylamino)-1-
triphenylphosphoniumalkylphosphonates 4a-d in good yields (Scheme 2, Table 2). The
optimal reaction time was determined based on NMR experiments. The synthesized
phosphonium salts were purified by dissolution in CH2Cl2 and precipitation of the crystalline
salt with Et2O.
Table 2. Synthesis of 1-(N-acylamino)-1-triphenylphosphoniumphosphonates 4 from the
corresponding 1-ethoxyphosphonate derivatives 3.
Entry
R1
R2
Procedure
temp. [°C]
Time [h] Yield [%]
Me
CH2Ph
Me
H
H
Ca
Ca
Db
Db
85
85
20
20
6
4
86
93
96
82
4a
4b
4c
4d
Me
Ph
0.1
0.75
Me
Reaction conditions: diethyl 1-(N-acylamino)-1-ethoxyalkylphosphonate (0.8 mmol), Ph3P.HBF4 (0.8 mmol),
CH2Cl2 (5 mL). a 85 °C, reduced pressure. b 20 °C, atmospheric pressure.
As
expected,
preliminary
experiments
proved
that
the
obtained
1-triphenylphosphonium derivatives of 1-aminophosphonic acid 4a-d were compounds of
high and diversified reactivity (Scheme 3).

Scheme 3. Potential applications of 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonates 4 for
the -functionalization of -aminophosphonic acid derivatives. Reagents and conditions: (i)
R3R4POR5, (iPr)2EtN, Ph3P+Me I, 20-60 ºC, 0.3-6 h, 43-78%; (ii) KCN, 18-crown-6, 20 ºC, 24 h,
50%; (iii) MeC(O)CF3, K2CO3, 18-crown-6, 50 ºC, 4 h, 43%; (iv) (iPr)2EtN, 20 ºC, 5 h, 98%; (v) TBD,
20 ºC, 24 h, 62%.
It was demonstrated that the phosphonium group of phosphonium salts 4 can easily be
displaced by triethyl phosphite or methyl (diphenyl)phosphonite in a Michaelis–Arbuzov-like
reaction to give N-acylamino)alkylenebisphosphonates 5 or their phosphonyl-phosphinoyl
asymmetrically substituted analogue, typically in good yields (Scheme 3, Table 3). As before,
the optimal reaction time was determined by 1H and 31P NMR spectroscopy.
It was also demonstrated that the phosphonium group of phosphonium salts 4c can
easily be displaced by a carbon nucleophile, e.g. the cyanide anion with formation of a new
C-C bond (cf compound 10).
We also proved that phosphonium salt 4a with a proton at the α-position underwent
the Wittig reaction with trifluoroacetone under PTC conditions to give the corresponding α,-
dehydro-α-aminophosphonic acid ester 11a. Moreover, -elimination of phosphonium
tetrafluoroborate from 1-(N-acylamino)-1-triphenylphosphoniumphosphonate with a proton at
the -position (4c) in the presence of Hünig’s base offers another route to the important
-dehydro--aminophosphonic acid derivative 11b. It is noteworthy that we were also able
to transform 1-(N-acylamino)-1-ethoxyethylphosphonate 3c into the same -dehydro--
aminophosphonic acid derivative 11b by the -elimination of ethanol, although this reaction
required a much stronger base [1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD)] and a longer
reaction time, and resulted in a much lower yield. -Dehydro--aminophosphonic acid

derivatives are widely used for asymmetric synthesis of -aminophosphonic acids by
stereoselective hydrogenation in the presence of chiral catalysts.17
Table 3. Michaelis–Arbuzov-like reaction of diethyl 1-(N-acylamino)-1-triphenylphosphoniumalkyl-
phosphonate tetrafluoroborates 4 with phosphorus nucleophiles
Substrate
R1
Nucleophile
Product
temp. [ºC] Time [h]
Entry
R2
R3
R4
R5
Entry Yield [%]
Me
CH2Ph
Me
H
H
Ph
Ph
Me
Et
60
60
20
0
4
4
78
73
70
43
4a
4b
4c
4d
5a
5b
5c
5d
EtO
EtO
EtO
EtO
EtO
EtO
Me
Ph
Et
6
Me
Et
0.3
Reaction conditions: diethyl 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonate tetrafluoroborate (0.5
mmol), phosphorus nucleophile (0.75 mmol), Ph3P+Me I– (0.12 mmol), (iPr)2EtN (0.05 mmol), CH2Cl2 (1.5 mL).
Conclusion
In conclusion, the Michaelis–Becker-like addition of diethyl phosphite to readily
accessible N-acylimidates followed by displacement of the ethoxy group by
triphenylphosphonium tetrafluoroborate was proven to be a novel and general method for the
synthesis of hitherto unknown 1-(N-acylamino)-1-triphenylphosphoniumalkylphosphonates
4a-d. The charged triphenylphosphonium group at the -position of compounds 4 can be
displaced by diversified nucleophiles to give variously -functionalized -aminophosphonic
acid derivatives. Both the Wittig reaction of phosphonium salts 4 which are able to generate
phosphonium ylides (R2 = H) and the -elimination of triphenylphosphonium
tetrafluoroborate from phosphonium salts with hydrogen at the -position (R2 = R6R7CH)
enable easy access to important -dehydro--aminophosphonic acid derivatives.
Supplementary data (for online publication)
1H, 13C and 31P NMR spectra of all synthesized compounds 3, all phosphonium salts 4 and
new bisphosphonates 5b,d as well as derivatives 10 and 11a,b are presented in the
Supplementary data file.

Acknowledgments
This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
Declarations of interest: none.
References:
1. Kafarski P, Lejczak B. Curr Med Chem-Anti-Cancer Agents. 2001; 1: 301–312.
2. Orsini F, Sello G, Sisti M. Curr Med Chem. 2010; 17: 264–289.
3. Kukhar V. P.; Hudson H. R. Aminophosphonic and Aminophosphinic Acids: Chemistry and
Biological Activity; John Wiley & Sons Ltd: Chichester, 2000.
4. Huang J, Chen R. Heteroatom Chem. 2000; 11: 480–492.
5. Październiok-Holewa A, Adamek J, Mazurkiewicz R, Zielińska K. Phosphorus, Sulfur, and
Silicon Relat Elem. 2013; 188: 205–212.
6. Adamek J, Mazurkiewicz R, Październiok-Holewa A, rymel M, Kuźnik A, Zielińska K. J Org
Chem. 2014; 79: 2765–2770.
7. Adamek J, Mazurkiewicz R, Październiok-Holewa A, Kuźnik A, rymel M, Zielińska K, Simka
W. Tetrahedron. 2014; 70: 5725–5729.
8. Kuźnik A, Mazurkiewicz R, rymel M, Zielińska K, Adamek J, Chmielewska E, Bochno M,
Kubica S. Beilstein J Org Chem. 2015; 11: 1418.
9. Roger R, Neilson DG. Chem Rev. 1961; 61: 179–211.
10. Neilson D. G. In The chemistry of amidines and imidates; Patai, S., Ed.; John Wiley & Sons,
1975; pp 385-489.
11. Pilotti A, Reuterhall A, Torssell K, Lindblad C-G. Acta Chem Scand. 1969; 23: 818–824.
12. Alves MJ, Durães MM, Fortes AG. Tetrahedron. 2004; 60: 6541–6553.
13. Inoue H, Nakagawa T, Seo S, Ohsawa N. U.S Patent 0305896 A1, 2012.
14. Pudovik AN, Zimin MG, Konovalova IV, Pozhidaev VM, Vinogradov LI. Zh Obshch Khim.
1975; 45: 30–37.
15. Steckhan, E. In Organic Electrochemistry; Lund, H., Hammerich, O., Eds.; CRC Press: New
York, 2001.
16. Ginzel K-D, Brungs P, Steckhan E. Tetrahedron. 1989; 45: 1691–1701.
17. Kuźnik A, Mazurkiewicz R, Kuźnik N. Curr Org Synth. 2013; 10: 411–424.
18. Ohme R, Schmitz E. Angew Chem Int Ed. 1967; 6: 566–566.

Highlights
1. Hitherto unknown 1-amino-1-phosphoniumalkylphosphonate salts were described.
2. A general approach for α-functionalization of α-aminophosphonates was developed.
3. Two effective pathways toward α,-dehydro-α-aminophosphonates were proposed.
4. A new route for the synthesis of bisphosphonates was reported.
5. Transformation of N-acylimidates into aminophosphonate derivatives was described.

Graphical Abstract