The Synthesis and Mechanistic Considerations of a Series of Ammonium Monosubstituted H-Phosphonate Salts

Article abstract of DOI:10.1002/chem.202003090

A series of ammonium monosubstituted H-phosphonate salts were synthesized by combining H-phosphonate diesters with amines in the absence of solvent at 80 °C. Variation of the ester substituent and amine produced a range of ionic liquids with low melting points. The products and by-products were analyzed by spectroscopic and spectrometric techniques in order to get a better mechanistic picture of the dealkylation and formal dearylation observed. For dialkyl H-phosphonate diesters, (RO)₂P(O)H (R=alkyl), the reaction proceeds via direct dealkylation with the reactivity increasing in the order R=iPr?1. For the diphenyl H-phosphonate diesters, (PhO)₂P(O)H, the dearylation was found to proceed via phenol-assisted formation of a 5-coordinate intermediate, (PhO)₃PH(OH), from which P(OPh)₃ and water were eliminated. The presence of an equivalent of water then facilitated the formation of P(OH)₂OPh and the amine, R'NH₂, subsequently abstracted a proton from it to yield [(PhO)PH(O)O]^-[R'NH₃]⁺.

Full text of DOI:10.1002/chem.202003090

Accepted Article
Accepted Article
Title: The synthesis and mechanistic considerations of a series of
ammonium monosubstituted H-phosphonate salts
Authors: Erin Leitao, Keng Lung Lee, Joey Feld, Paul Hume, and Tilo
Söhnel
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202003090
Link to VoR: https://doi.org/10.1002/chem.202003090
01/2020

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
The synthesis and mechanistic considerations of a series of
ammonium monosubstituted H-phosphonate salts
Keng Lung Lee,[a],[c] Joey Feld,[a] Paul Hume,[b],[c] Tilo Söhnel[a],[c] and Erin Leitao *[a],[c]
[a]
Mr KL Lee, Mr J Feld, A/Prof T Söhnel, Dr EM Leitao
School of Chemical Sciences
University of Auckland
Private Bag, 92019, Auckland, 1142, New Zealand
E-mail: erin.leitao@auckland.ac.nz
Dr P Hume
School of Chemical and Physical Sciences
Victoria University of Wellington
Wellington, New Zealand
[b]
[c]
The MacDiarmid Institute for Advanced Materials and Nanotechnology
New Zealand
Supporting information for this article is given via a link at the end of the document.
Abstract: A series of ammonium monosubstituted H-phosphonate
salts were synthesized by combining H-phosphonate diesters with
amines in the absence of solvent at 80°C. Variation of the ester
substituent and amine produced a range of ionic liquids with low
melting points. The products and by-products were analyzed by
spectroscopic and spectrometric techniques in order to get a better
mechanistic picture of the dealkylation and formal dearylation
substituents with strong electron-withdrawing effects (e.g. R = Ph)
help to stabilise the lone pair in the P(III) tautomer, making the H-
phosphonates more reactive. By contrast, electron-donating
substituents (e.g R = Me, Et and i-Pr) destabilise the lone pair,
and the equilibrium lies heavily towards the P(V) tautomer,
making the H-phosphonate diesters less reactive.[10]
Tautomerization is the first step when H-phosphonates are
observed. For dialkyl H-phosphonate diesters, (RO)2P(O)H (R = alkyl), attacked by a nucleophile such as an alcohol,[11] water,[12] or an
the reaction proceeds via direct dealkylation with the reactivity
increasing in the order R = i-Pr < Et < Me corresponding to DFT
calculated activation enthalpies of 22.6, 20.8, and 17.9 kcal/mol. For
the diphenyl H-phosphonate diesters, (PhO)2P(O)H, the dearylation
was found to proceed via phenol-assisted formation of a 5-coordinate
intermediate, (PhO)3PH(OH), from which P(OPh)3 and water were
eliminated. The presence of an equivalent of water then facilitated the
formation of P(OH)2OPh and the amine, R’NH2, subsequently
abstracted a proton from it to yield [(PhO)PH(O)O]-[R’NH3]+.
amine[13] and lead to the formal exchange of the OR substituent
with the nucleophile (e.g. OR’, OH, or NHR’, respectively).[14]
O
P
P H
OH
RO
RO
RO
RO
4
λ5
σ
σ
λ3
3
Figure 1. Tautomerization of H-phosphonate diesters P(V), to disubstituted
phosphite, P(III).
Introduction
H-phosphonate diesters are phosphorus compounds that are
characterised by the presence of multiple functional groups: P-
OR, P-H and P=O.[1] These molecules have attracted a lot of
attention in both organic and inorganic chemistry for the creation
of new phosphorus element bonds, such as P-N,[2], P-O-P and P-
P,[3], P-F,[4], P-S, P-Se and P-Te,[5] and P-C bonds[6] due to their
interesting reactivity as both the phosphorus centre and α-carbon
can be subject to nucleophilic attack. More recently, H-
phosphonate diesters have been used in the formation of
molecules[7] and macromolecules[8] with interesting architectures
and useful potential applications.
The vast majority of the reported reaction chemistry of H-
phosphonate diesters is via nucleophilic attack at the phosphorus
center, however, the α-carbon atom on the OR substituent, is also
susceptible to nucleophilic attack. For example, this reaction can
lead to abstraction of an alkyl group on H-phosphonates to form
an anion when reacted with an amine (e.g. [(RO)PH(O)O]-
Scheme 1). Initially, these anions were postulated to be too
reactive to exist under standard conditions but were theoretically
predicted to be stabilized by hydrogen bonding with the alkylated
amine cation, forming an ionic salt.[1a] Subsequently, these salts
have been postulated as intermediates in the Atherton-Todd
reaction,[15] can be synthesized through direct dealkylation
reactions of H-phosphonate diesters and amines,[16] and have
been observed as by-products when H-phosphonates are used
as precursors in the synthesis of phosphoramidates via copper-
catalysed dehydrogenative coupling.[2d] In the latter, as the
reaction involves the use of an amine as a substrate, the reaction
efficiency suffers as the formation of the ionic salt is competitive
with the dehydrogenative coupling reaction.
One of the most beneficial features of H-phosphonate diesters is
their ability to undergo tautomerization similar to keto-enol
tautomerization. The tautomeric equilibrium exists between a five-
coordinated phosphonate, P(V) (λ5σ4), and three-coordinated
phosphite, P(III) (λ3σ3), favouring the P(V) tautomer over the P(III)
tautomer (Figure 1).[9] The equilibrium position is influenced by
the substituents on the H-phosphonate diesters. For instance,
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
O
O
P
phosphonate diester (1) and amine, in the absence of solvent at
80°C for 20 h (2-4, Scheme 3, Figures S1-S22). By 31P NMR
+
NR'3
P
H
H
RO
RO
RO
spectroscopy, a distinctive doublet can be observed due to the P-
1
H bond in the anion, [(RO)PH(O)O]- (δP = 1.3-11.5 ppm, JPH
=
O
NR'3R
611-664 Hz), with further splitting observed depending on the R
group (Table 1). For example, the characteristic doublet of
pentets of (EtO)2P(O)H, 1b, at 7.3 ppm (1JHP = 680 Hz; 3JPH = 7.9
Hz, coupling to four CH2 protons) was replaced with a doublet of
triplets at 3.9 ppm, corresponding to the anion, [(EtO)PH(O)O]-, in
Scheme 1. Alkylation of an amine by dialkyl H-phosphonate.
Formation of ionic liquids, ionic materials with melting points lower
than 100°C, such as alkylphosphonate imidazolium salts, are also
possible through direct dealkylation of H-phosphonate diesters.
Ionic liquids with a phosphonate anion exhibit high conductivities
(ca. 10-3 S cm-1 at 25°C) which is correlated with low glass-
transition temperatures.[17] Due to the nature of the anion, it is
possible to form multiple complexes with the imidazolium cation.
These solvents have excellent reactivity towards copper, exhibit
substantial corrosion protection of magnesium alloys,[18] are
potential CO2 absorbents[19], can extract polysaccharides from
Japanese cedar[20] and are used as synthons in making new H-
phosphonates.[21]
3
2b (1JPH = 616 Hz; JPH = 8.0 Hz, coupling to two CH2 protons)
over time, as evidenced by 31P NMR spectroscopy (Figure S2).
The 31P NMR peak assignment for 2b is corroborated by a doublet
1
1
signal (δH = 6.63 ppm, JHP = 616 Hz) observed in the H NMR
spectrum (Figure S1). Across the series, the observed coupling
constants and chemical shifts for the various derivatives differ,
thus are influenced by the interaction of the different cations as
well as the attached R groups (e.g. Me, Et, i-Pr, Ph). Despite
compounds 2b and 3b having the same anion, [(EtO)PH(O)O]-,
and the compounds in the series 2d, 3d and 4d, [(PhO)PH(O)O]-,
there are notable differences in the 31P NMR chemical shifts, at
3.9 and 3.6 ppm (2b, 3b respectively) and 1.3, 1.4, 2.1 ppm (2d,
3d, 4d respectively). H-bonding in the form of N-H----O-P
between the cation and anion is prevalent (see Figures 2-3) and
inevitability contributes to the observed changes.
Unlike monodealkylation with dialkyl H-phosphonates,
monodearylation from diphenyl H-phosphonate is not possible
due the absence of an electrophilic α-carbon available for
nucleophilic attack. Instead, diphenyl H-phosphonate has been
postulated to undergo a disproportionation reaction to form a
monophenyl H-phosphonate anion and triphenylphosphite in the
presence of base (Scheme 2).[22]
O
P
O
P
80°C
20 h
'
+
H
H
NH2R
R
O
R
O
R
O
O
NH2R''R'
=
R' Ph, Bn,
2-4
1
O
P
O
P
n
base
or
Bu
2
:
3
= n-
:
2
R' Bn
+
R'
Bu
H
:
:
=
=
=
=
=
1a
1b
H
R
R
R
R
Me
Et
i-Pr
P(OPh)3
-
H
PhO
:
:
=
=
=
or
3b
3d
R
R
Et
Ph
PhO
R''
H
R
:
=
=
=
=
2a
R
R
R
R
Me
:
:
:
PhO
1c
1d
2b
Et
O
:
:
2c
2d
Ph
i-Pr
Ph
:
4
=
R' Ph
:
=
4d
R
Ph
Scheme 2. Base catalysed disproportionation of diphenyl H-phosphonate.
Scheme 3. Formation of a series of monosubsituted phosphonate salts (2-4)
from the corresponding H-phosphonate diesters (1) and amines.
To improve the selectivity and efficiency during the synthesis of
ammonium monosubstituted H-phosphonate salts as well as in
oxidative dehydrogenative coupling reactions between H-
phosphonates and amines, it is crucial to understand the
Table 1. Selected data for compounds 2-4.
formation of these ionic salts. In this work,
a series of
Compound
R/R’/R’’
δ
P P-H /ppm
Conversion[a] or
Yield[b] /%
monosubstituted H-phosphonate ammonium salts were
synthesized and characterized, including by thermogravimetric
analysis and mass spectrometry. To gain insight into the
mechanisms, kinetic studies were performed by monitoring the
phosphorus-containing compounds using 31P NMR spectroscopy.
Spectroscopic and spectrometric techniques were combined in
order to observe the presence of any intermediates during the
formation of the ion-pair and the experimental results were
supported by DFT calculations.
(1JPH /Hz)
11.5 (617)
3.9 (616)
6.4 (614)
1.3 (645)
3.6 (611)
1.4 (644)
2.1 (664)
2a
2b
2c
2d
3b
3d
4d
Me/Bn/mix
Et/Bn/mix
i-Pr/Bn/mix
Ph/Bn/H
80[a]
83[a]
16[a]
45[b],[c]
78[a]
Et/Bu/mix
Ph/Bu/H
48[a],[c]
40[b],[c]
Results and Discussion
Ph/Ph/H
Synthesis
monosubstituted H-phosphonate salts
and
characterization
of
ammonium
[a] conversion as determined by 31P NMR spectroscopy after 20 hours of
reaction time at 80°C (2a was at 20°C). [b] isolated yield after recrystallization.
[c] maximum yield/conversion = 50% due to the formation of P(OPh)3 during the
synthesis.
A series of ammonium monosubstituted H-phosphonate salts
were synthesized by combining the two components, H-
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
The ammonium cations, [R’R’’NH2]+, in the monosubstituted H-
phosphonate salts can be characterized by using a mixture of 1H
NMR spectroscopy and ESI mass spectrometry. The ammonium
cation of compound 2b was observed as a mixture of [BnNH3]+
P(OPh)3 and PhOH during the synthesis (which is also present in
trace amounts in the starting material, 1d, observed by 1H NMR
spectroscopy), the maximum conversion (or yield) possible when
starting with compound 1d is 50% (Table 1).
3
and [BnEtNH2]+. Two quartets (δH = 2.7-3.8 ppm, JHH = 7.1 Hz)
3
Single crystals were grown for the ammonium phenyl H-
phosphonate salts, 2d and 4d, by slow diffusion of hexanes into
a saturated dichloromethane solution; the crystal structures were
solved by single crystal X-ray diffraction (Figures 2-3, S23-S24
Tables S1-S3). From the molecular structures, both the P and N
centres exhibit slightly distorted tetrahedral geometry as the
smallest bond angle is 102.48(8)° and the largest is 120.21(9)°
(Figures 2a and 3a).
and two triplets (δH = 1.1-1.4 ppm, JHH = 7.1 Hz), one
corresponding to the Et group on the cation, [BnEtNH2]+, and the
other corresponding to the Et group on the anion, [(EtO)PH(O)O]-
were observed by 1H NMR spectroscopy (Figure S1). This
assignment was also supported by ESI mass spectrometry
([C9H16NO3P
+
H]+ 218.0946 m/z (cal’d 218.0946 m/z),
[C11H20NO3P + H]+ 246.1254 m/z (cal’d 246.1254 m/z), Figure
S3). However, the presence of a mixture of cations precludes
purification for compounds 2a, 2b, 3b. To account for the
observation that either a proton or an ethyl group was formally
transferred to the amine, either a one-step or a two-step route to
the cations is possible (Scheme 4).
1
Step
a)
direct
dealkylation
O
P
O
P
R'NH2
R
H
H
RO
O
RO
O
R'RNH2
=
R
i-Pr, Et, Me
=
R' Bn, Bu
Hoffmann
elimination
2
Step
-
CH2CH2
O
O
=
R
Et
P
H
P
H
RO
RO
+
-
R'NH2
O
O
R'NH3
R'RNH2
R'RNH
deprotonation
b)
Scheme 4. One-step and two-step routes to forming ammonium cations.
The one-step route involves the direct transfer of the alkyl group,
R, from (RO)2P(O)H via nucleophilic attack from the amine (Step
1, Scheme 4 producing [R’RNH2]+). There are two options for the
second step to produce the protonated ammonium cation,
[R’NH3]+ i) Hoffmann elimination of the alkene directly from
[R’RNH2]+ (e.g. for R = Et, loss of CH2=CH2)[23] or ii) deprotonation
of the ammonium salt (e.g. [R’RNH2]+) via a second equivalent of
base (R’NH2) to yield R’RNH and [R’NH3]+ (Step 2, Scheme
4).[16a] The presence of mixed cations due to alkyl transfer in the
series was limited to starting with either 1a ((MeO)2P(O)H), 1b
((EtO)2P(O)H) or 1c ((i-PrO)2P(O)H) and the Ph groups from 1d,
was not observed to be transferred to the ammonium cation (see
Figures S10 and S18). This suggests that a different mechanism
is in operation for the diphenyl H-phosphonate, 1d, resulting in
exclusively the ammonium cation, [R’NH3]+, in compound 2d.
Unlike the dialkyl H-phosphonates, the α-carbon atom of diphenyl
H-phosphonate is not vulnerable to nucleophilic attack so direct
dearylation of the amine is not possible. Instead, during the
synthesis of the series 2d, 3d, 4d, another significant resonance
was observed by 31P NMR spectroscopy at 128.0 ppm,
corresponding to P(OPh)3 agreeing with the previously reported
studies on the base catalyzed disproportionation of 1d (Scheme
2, and Mechanistic considerations).[22] Due to the generation of
c)
Figure 2. Molecular structure of 2d showing a) the separated ion pair b)
hydrogen bonding from the perspective of the ammonium cation c) hydrogen
bonding from the prospective of the phosphate anion, with thermal ellipsoids
shown at the 50% probability level.
a)
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
temperature. Moreover, all the TGA spectra showed a steep
increase in mass loss at 100°C which is characteristic for water
loss. Considering the reactions were performed without using any
solvents, this suggests that the ionic liquids are quite hygroscopic,
capturing moisture from the atmosphere. As an alternative
explanation, protic ionic liquids sometimes exist in equilibrium
with their respective neutral compounds (Figure 4) and this
equilibrium causes the loss of electrostatic forces of attraction
making the liquid more volatile. Lastly, full decomposition of all of
the compounds was not observed even at 600°C, which is
unusual compared to other protic ionic liquids reported on
literature.[25]
b)
O
P
O
P
+
H
NHRR'
H
RO
RO
HO
O
NH2R'R
liquid
ionic
neutral
compounds
Figure 4. Protic ionic liquids, such as ammonium monoalkyl H-phosphonate, in
equilibrium with the starting acid and base.
c)
Figure 3. Molecular structure of 4d showing a) the separated ion pair b)
hydrogen bonding from the perspective of the ammonium cation c) hydrogen
bonding from the prospective of the phosphate anion, with thermal ellipsoids
shown at the 50% probability level.
Prior to running TGA, the melting points of 2d and 4d were
determined to be 112.3°C and 116.4°C, respectively (Table 2).
For the viscous liquids 2b, 3b, and 3d it was crudely determined
that melting points are below 0°C. Compound 2d had the highest
Tpeak which was 222°C. The TGA curve for 2d also showed
consistent loss of mass starting from 50°C and the compound
The three P-O bonds in the anions have bond lengths of 1.616(2)
Å and 1.615(2) Å (P1-O1(Ph), single bond), 1.476(2) Å and
1.484(2) Å (P1-O2 with more double bond character), and
1.494(2) Å and 1.498(2) Å (P1-O3 with more single bond
character) for compounds 2d and 4d, respectively. Kee et al.
reported similar ionic compounds with comparable P-O
distances.[16a] For example, the anion [(BnO)PH(O)H]- has a P-
O(Bn) distance of 1.590(1) Å and the other P-O bond lengths are
1.479(1) Å and 1.504(1) Å. However, when the anion is
protonated, as in (ArO)PH(O)OH (Ar = 2,6-dimethylphenyl)
synthesized by Murugavel et al., the three bond lengths are
1.579(2) Å (P-O(Ar)), 1.527(2) Å (P-O(H)) and 1.469(2) Å (P=O),
suggesting that both of the P-O bonds without substituents in 2d
and 4d fall somewhere between double bonds and single
bonds.[24] The ion pairs are held together with a substantial
amount of H-bonding. From the perspective of the anion,
[PhOPH(O)O]-, the oxygens, O2, with more double bond
character are bonded to one cation and the anionic oxygens, O3,
are H-bonded to two cations (Figures 2b and 3b). The cation,
[R’NH3]+ (R’ = Bn, Ph), containing three hydrogens is H-bonded
to three anions with O-H distances of 1.77(3) Å, 1.88(3) Å and
1.89(3) Å for 2d (Figure 2c) and distances of 1.80(4) Å, 1.83(4)
Å and 1.84(4) Å for 4d (Figure 3c).
started to decompose at Tonset
= 131°C. By contrast, 4d
experienced mass loss starting from 56°C, which is in agreement
with the instability observed both in solution and in the solid state
(Figure S22). In addition, compound 4d had the lowest Tpeak at
161°C and the lowest Tonset at 113°C. For compound 3d, the TGA
curve depicted multistage decomposition, with Tonset at 125°C and
the maximum rate of weight loss at Tpeak = 181°C (Figure S19).
Compounds 2b and 3b showed similar decomposition curves with
identical Tonset at 162°C and Tpeak at 181°C (Figures S4 and S15).
Table 2. Decomposition temperatures for selected ionic salts
Compound
Tonset /°C
Tpeak /°C
Melting point /°C
2b
2d
3b
3d
4d
162
131
162
125
113
181
222
181
181
161
< 0
11.3
< 0
< 0
116.4
Compounds 2a-c and 3b are viscous liquids and therefore could
have utility as protic ionic liquids.[18-19, 25] To compare with similar
H-phosphonate ionic liquids, thermogravimetric analysis (TGA)
was performed on the products (Figures S4, S11, S15, S19, S22).
However, the ionic salts suffered some weight loss at relatively
low temperatures (38-100°C) which is not commonly observed for
these types of ionic salts. It is postulated, in this case, that the
trace amounts of amines and impurities present in the compounds
(see Figures S1-S22) resulted in evaporation at a lower
According to a review paper published by Cao and Mu,[26] five
levels of thermal stability can be categorized based on the value
of Tonset. When Tonset is lower than 250°C, the thermal stability of
ionic liquids is classified as least stable while the most stable ionic
liquids are those that possessed Tonset greater than 400°C. This
suggests that the ionic liquids formed from H-phosphonate and
amines are among the least thermally stable and this will limit their
applications. Furthermore, the authors compared the relationship
between thermal stability and chain length. Although the effect of
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
chain length was not significant, it was clear that the shorter chain
length tends to have higher thermal stability. The was rationalized
as, even though a longer chain length can increase Van der
Waals forces, the bulky structure minimizes the electrostatic
interaction between the cation and anion.[26]
intermediate aminolysis product, 5b ((EtO)(BnNH)P(O)H, δP
=
17.7 ppm, 1JPH = 637 Hz, 3JPH = 9.4 Hz Scheme 5) was detected
as a doublet of pentets in the 31P NMR spectrum, reached its
highest concentration at 60 minutes when it could be detected by
ESI-MS ([C9H14NO3P + Na]+ 222.0662 m/z (cal’d 222.0660 m/z),
Figure S31) and then slowly disappeared over 7 hours. The
presence of a persistent by-product, 6b, ((EtO)(BnNH)P(O)OH, δP
= 28.6 ppm) was also observed during the reaction of 1b with
benzylamine. The initial rate for formation of 2b was determined
to be 0.131 min-1 at 80°C (Figure S32).
Mechanistic considerations
For the series of compounds produced by monodealkylation of
1a-c using benzylamine, 2a-c, the reactions were monitored over
time by 31P NMR spectroscopy. Dimethyl H-phosphonate, 1a,
reacted with benzylamine within minutes at 80°C, so for this
Table 3. Intermediates and products formed during the kinetic monitoring of the
conversion of 1 to 2
derivative, the temperature was reduced to 20°C. Consistent with
what was reported by Troev et al.[23a] a doublet of quartets for 2a
(δP = 11.5 ppm, 1JPH = 619 Hz, 3JPH = 12 Hz) was observed by 31
P
1
Starting Material
Intermediate
or Product
δP P-H /ppm, splitting, JHP /Hz,
3JHP /Hz,
NMR spectroscopy (Table 2, Figure S25). An intermediate
aminolysis product,[14] 5a ((MeO)(BnNH)P(O)H, δP = 14.2 ppm,
1JPH = 696 Hz,3JPH = 12 Hz Scheme 5) was also observed at early
reaction times as a doublet of sextets in the 31P NMR spectra
(Table 3) and then disappeared after 7 hours. ESI-MS analysis
1a
1a
1b
1b
1b
1c
1c
1c
1c
1d
1d
1d
1d
1d
2a
11.5, dq, 619, 12
14.2, ds, 696, 12
9.0, dt, 615, 8.2
17.7, dp, 637, 9.4
28.6, tt
5a
1
and H NMR spectroscopy were performed on the sample after
2b
5b
6b
2c
the timed experiment, revealing that the cation in 1a was a mixture
of [BnMeNH2]+, [BnMe3N]+ and [BnNH3]+(Figures S26-S27) with
the [(MeO)PH(O)O]- anion visible in the negative infusion
spectrum (Figure S28). Presence of [BnMe3N]+ is presumably
possible by first demethylating 1a to form [BnMeNH2]+ which can
be deprotonated by BnNH2 to form [BnNH3]+ along with the
7.0, dd, 613, 9.5
15.5, br d, 635
26.9. dt
stronger nucleophile, BnMeNH.
Then the 2° amine can
5c
demethylate another molecule of 1a to generate [BnMe2NH]+ and
the deprotonation/demethylation steps can occur once more to
yield the quaternary ammonium cation, [BnMe3N]+. After
integrating the 31P{1H} NMR spectra over time and plotting the
relative integrations, the initial rate of formation of 2a was found
to be 0.006 min-1 at 20°C (Figure S29).
6c
7c
3.7, br s
2d
5d
8
6.6, d, 645
9.2, dt, 716, 9.3
17.7, ds, 647, 10.5
15.7, d, 667
O
P
O
P
H
H
BnHN
BnHN
RO
HO
9d
P(OPh)3
9a-d
8
128, s
H
2O
ROH
NH2Bn
ROH
[a] conversion as determined by 31P NMR spectroscopy after 20 hours of
reaction time at 80°C. [b] maximum conversion = 50% due to the formation of
P(OPh)3 during the synthesis.
O
P
O
P
O
P
NH2Bn
ROH
H
H
OH
BnHN
RO
RO
RO
RO
[O]
RO
1a-d
7a-d
Moving from diethyl to diisopropyl H-phosphonate resulted to an
increase in the number of side reactions possible (Table 3,
Figures S33-S36). As with the synthesis of 2a and 2b, an
aminolysis intermediate, 5c ((i-PrO)(BnNH)P(O)H, δP = 15.5 ppm,
1JPH = 635 Hz Scheme 5) was observed as a broad doublet by
31P NMR spectroscopy at early reaction times reaching its
maximum concentration in solution at 180 minutes before
disappearing. The aminolysis reaction was more competitive in
the reaction with 1c relative to 1a and 1b, due to the presence of
the bulky isopropyl groups that inhibits nucleophilic attack at the
α-carbon. However, based on the observations of previous
reactions, the aminolysed products are unstable and exist in
equilibrium with the reactants, eventually favouring the formation
of more thermodynamic product, 2. By-products 6c ((i-
PrO)(BnNH)P(O)OH, δP = 26.9 ppm) and a minor amount of 7c
((i-PrO)2P(O)OH, δP = 3.7 ppm) were also spectroscopically
observed (Figure S36). After the course of the reaction, the main
species present was 2d (Figure S37) containing both the
protonated and alkylated ammonium cations ([BnNH3]+ and [Bn(i-
5a-d
[O]
a
b
:
=
R
R
R
R
Me
Et
i-Pr
Ph
NH2Bn
O
:
:
=
=
c
d
O
P
:
=
P
H
OH
BnHN
RO
RO
NH2R''Bn
O
6a-d
=
or
R
R''
H
2a-d
Scheme 5. Aminoloysis, hydrolysis, alcoholysis, and oxidation of
postulated routes to the intermediates observed during the synthesis ionic salts
2.
1 as
The reaction of diethyl H-phosphonate, 1b, with benzylamine took
4 days to reach completion at 80°C. Similar to 2a, 2b was
observed as a doublet of triplets (δP = 9.0 ppm, 1JPH = 615 Hz, 3JPH
= 8.2 Hz) by 31P NMR spectroscopy (Table 3, Figure S30). The
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
Pr)NH2]+) as observed by ESI mass spectrometry (Figures S38).
The reaction between diisopropyl H-phosphonate, 1c, with
benzylamine to form 2c was more than two orders of magnitude
slower than the formation of 2b at 80°C with an initial rate of 0.001
min-1 (Figure S39). Furthermore, unlike with 2a and 2b, the
reaction between 1c and benzylamine did not reach completion
after 7 days at 80°C and approximately 22% of 1c remained
unreacted with only 51% of 2c formed along with 16% of 6c.
O
O
P
NR'3R
80°C
20 h
+
P
H
H
NR'3
RO
:
RO
RO
and/or
NR'3H
O
10
2, 3,
1
=
=
=
:
:
=
=
=
2 NR'3 BnNH2
NR'3
NR'3 Et3N
1b
1d
R
R
Et
Ph
NR'3 BnNH2
,
n
-BuNH2
n
:
3
-BuNH2
Et N
,
3
:
10
Diphenyl H-phosphonate, 1d, reacted with benzylamine instantly
when the two reactants were mixed together at 20°C. After 30 s
of stirring, about 94% of 1d was consumed to produce various
Scheme 6. Reaction comparison of 1b and 1d with different amines.
phosphorus compounds according to numerous peaks in the 31
P
NMR spectra (Table 3, Scheme 5, Figure S40-S42), including
P(OPh)3 (δP =128.0 ppm) as a major product. Among all of these
phosphorus species, 2d (δP = 6.6 ppm, 1JPH = 645 Hz) comprised
Table 4. The effect of the amine on the formation of the ammonium
monosubstituted H-phosphonate salts
Starting
Material
Amine
δ
P P-H /ppm
Conversion[a] /%
37% of the total mixture. Interestingly, three signals
(1JHP /Hz)
7.3 (680)
3.6 (611)
3.9 (616)
4.4 (616)
4.5 (669)
1.4 (644)
1.3 (645)
1.9 (663)
1
corresponding to 5d ((PhO)(BnNH)P(O)H, δP = 9.2 ppm, JPH
=
716 Hz, 3JPH = 9.3 Hz), 9d ((HO)(BnNH)P(O)H δP =15.8 ppm, 1JPH
= 667 Hz,) and 8 ((BnNH)2P(O)H, δP = 17.7 ppm, 1JPH = 647 Hz,
3JPH = 10.5 Hz) were observed by 31P NMR spectroscopy during
the initial stages of the reaction, but slowly diminished over time
(Scheme 5). To assist with the aforementioned assignments, the
ESI mass spectrum at 30 min revealed two major phosphorus
species [(BnNH2)(PhO)P(O)H + Na]+ (5d) and [(BnNH2)2P(O)H +
H]+ (8) (Figure S42) which were not observed in the end of the
reaction. Both of these intermediates were the result of aminolysis
substitution reactions. As the PhO substituent has a stronger
electron withdrawing effect on the H-phosphonate when
compared with alkyoxy substituent, tautomerization, and hence
aminolysis, is favoured.
1b
1b
1b
1b
1d
1d
1d
1d
none
n/a
78
n-BuNH2
BnNH2
Et3N
83
19
none
n/a
48[b]
45[b]
14[b]
n-BuNH2
BnNH2
Et3N
[a] conversion as determined by 31P NMR spectroscopy after 20 hours of
reaction time at 80°C. [b] maximum conversion = 50% due to the formation of
P(OPh)3 during the synthesis.
Overall, the order of reactivity for the various H-phosphonate
diesters with benzylamine follows the trend: 1d > 1a > 1b > 1c. In
addition to comparing the difference in the reactivity by modifying
the R group, it is also possible to modify the nucleophilicity of the
amine (Scheme 6). Reacting diethyl H-phosphonate, 1b, with
amines BnNH2, n-BuNH2 and Et3N produced the corresponding
ammonium ethyl H-phosphonate salts in 78% (2b), 83% (3b), and
19% conversion (10b) after 20 h at 80°C, respectively (Table 4,
Figures S43-S44). These results indicate that the Et3N is too
sterically bulky to effectively monodealkylate 1b and the two
primary amines have similar nucleophilicity in this reaction. By 1H
NMR spectroscopy it was clear that 10b contained a mixture of
both [Et4N]+ and [Et3NH]+ cations. Reacting diphenyl H-
phosphonate, 1d, with the same three amines, BnNH2, n-BuNH2
and Et3N, produced the corresponding ammonium phenyl H-
phosphonate salts in 48% (2d), 45% (3d), and 14% conversion
(10d) after 20 h at 80°C, respectively (Table 4, Figures S44-S45).
In addition, starting from 1d, P(OPh)3 was produced, along with
PhOH indicating that both of these species are involved in the
mechanism. Additionally, all of the cations formed from 1d are
protonated amines (e.g. [BnNH3]+, [n-BuNH2]+, or [Et3NH]+).
DFT Calculations
DFT calculations were performed in order to gain a deeper
understanding of the underlying mechanism of salt formation.
Calculations employed the B3LYP exchange-correlation
functional,[27] with the 6-311+(d,p) basis set and Grimme’s
empirical dispersion correction (D3).[28] Solvent effects were
included by the use of a polarizable continuum model (PCM).
Dibutylamine was selected for the PCM because of its chemical
similarity to benzylamine, and comparable dielectric constant.
For the alkyl (R = Me, Et, i-Pr) series 1a-c, it was possible to locate
transition states for direct nucleophilic dealkylation by
benzylamine (Scheme 7). The activation enthalpies for these
processes follow the trend expected from steric considerations
(17.9, 20.8, and 22.6 kcal/mol for Me, Et, and i-Pr, respectively).
The possibility that compounds 5a-c were intermediates in the
formation of 2a-c was considered. However, this interpretation is
inconsistent with the observed reaction kinetics (see previous
Section, and Supporting Information). The transient accumulation
of 5a-c, together with the fact that no other phosphorus-containing
species were observed in significant quantities, means that
consumption of 5a-c would need to be rate-limiting under this
hypothesis. This would imply that the rate of formation of 2a-2c
would increase with concentration of 5a-5c, however this is not
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
Scheme 8. Previously proposed mechanism for the formation of 2d and
the case. Therefore, the interpretation given in Scheme 4, is
favored wherein reversible formation of 5a-5c leads to an initial
accumulation of these species, which are subsequently
consumed as the irreversible formation of 2a-2c proceeds. The
observation of analogous species 5d in the reaction of 2d with
BnNH2, which follows a different mechanism, further supports this
conclusion.
P(OPh)3.[22]
Instead, on the basis of previous work,[14] we propose that
diphenyl H-phosphonate, 1d, reacts with trace phenol (present in
2d), which results in the formation of 5-membered intermediate,
Int-1. Following pseudorotation, Int-1 undergoes phenol-assisted
dehydration to form P(OPh)3 and water (Scheme 9). The water
produced in this way can then react with a second equivalent of
1d to form the corresponding five-membered intermediate Int-2
and hydrolyse 1d to 9d (see also Scheme 5). DFT calculations
indicate that the formation of five-membered intermediates Int-1
and Int-2 are associated with activation barriers of +15.8 and
+27.4 kcal/mol respectively. Phenol-assisted loss of the apical
substituent of five-membered intermediates Int-1 and Int-2 is
associated with activation barriers of +10.7 and +7.6 kcal/mol,
respectively. The high activation barrier associated with the
formation of Int-2 can be explained by activation of H2O through
H-bonding or protonation from the primary ammonium salt
present in the mixture.
=
+22.6, TS1c R i-Pr
,
=
TS1b R Et
+20.8,
,
=
+17.9, TS1a R Me
,
H
HO
P
OPh
OPh
O
P
PhO
PhO
pseudorotation
PhO
HO
+
PhOH
H
H
P
H
PhO
PhO
OPh
O
P
OPh
Int-1
PhOH
+
1d
BnNH2
O
P
(RO)2
σ
4
λ5
H
RO
σ
5
λ5
H
O
+
+
PhOH
=
=
=
H
1a
2O
R
R
R
Me
Et
i-Pr
,
P(OPh)3
σ
3
λ3
BnNH2R
1b
,
1c
,
H
=
=
=
2c
-20.0,
-22.6,
-24.5,
R
R
R
i-Pr
Et
,
PhO
HO
OH
P
OPh
O
P
pseudorotation
PhO
HO
2b
2a
,
+
H
H
2O
P
H
H
PhO
PhO
Me
,
HO
OPh
OPh
Int-2
PhOH
1d
Scheme 7. Direct nucleophilic dealkylation of dialkyl H-phosphonates by BnNH2
and calculated activation/reaction enthalphies. All values given in kcal/mol.
Transition state diagram shown for 1a, R = Me.
σ
4
λ5
σ
λ5
5
BnNH3
BnNH2
O
P
H
PhO
+
OPh
2 PhOH
P(OH)2
9d
O
2d
For compound 1d (R = Ph) nucleophilic displacement is precluded
by the nature of the substituent. In attempting to develop a
mechanistic hypothesis for this reaction it was important to
account for the formation of P(OPh)3, which is rapidly generated.
σ
3
λ3
σ
λ5
4
Scheme 9. Proposed mechanism for the formation of 2d and P(OPh)3 from 1d
in the presence of trace PhOH.
A
proposal involving phosphonate disproportionation has
previously been reported[22] (Scheme 8). However, this
mechanism posits nucleophilic attack on a phosphite centre
bonded to a more electrophilic phosphonate. All attempts to locate
transition states corresponding to this step were unsuccessful.
Conclusion
In summary,
a series of ammonium monosubstituted H-
phosphonate salts were prepared and thoroughly investigated by
ESI-MS, multi-nuclear NMR spectroscopy, TGA and single crystal
BnNH3 BnNH2
O
P
OPh
P
O
P
OPh
P
XRD,
where
appropriate.
Evidence
for
the
direct
monodealkylation of (RO)2P(O)H (1) using both primary and
tertiary amines, with the generation of a low concentration of the
monoaminoloysis species, (RO)(NHR’)P(O)H was observed for
the series R = Me, Et, i-Pr. The rate of dealkylation experimentally
followed the order Me > Et > i-Pr, corroborated by DFT
calculations. A mixture of the protonated [R’NH3]+, and alkylated
2°, 3° and even 4° ammonium cations (e.g. [R’RNH2]+, [R’R2NH]+,
[R’R3N]+, respectively) were observed for the dialkyl H-
phosphonates.
PhO
H
H
PhO
PhO
1d
O
OPh
O
OPh
H
BnNH2
PhOH
BnNH3
P(OPh)3
O
+
P
PhO
O
2d
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
MS: [C9H16NO3P + H]+ 218.0946 m/z (cal’d 218.0946 m/z), [C11H20NO3P +
H]+ 246.1254 m/z (cal’d 246.1254 m/z), [C11H20NO3P + BnNH2(Et)]+
381.2297 m/z (cal’d 381.2297 m/z). Tonset = 162°C; Tpeak = 188°C.
Addition of amine to diphenyl H-phosphonate (1d), resulted to a
relatively more complicated mixture on route to producing the
ionic salt [(PhO)PH(O)O]-[BnNH3]+ (2d).
DFT calculations
suggested that the presence of trace amounts of PhOH acts as
an autocatalyst for the reaction eliminating H2O from a 5-
coordinate intermediate (Int-1). Hydrolysis of 1d (via Int-2) then
generated compound 9d (observed experimentally) along with
PhOH, followed by proton abstraction from the amine to afford
product 2d.
Synthesis of phenyl H-phosphonate benzylammonium salt, 2d.
1d (3.0 mmol, 0.57 mL) and benzylamine (3.0 mmol, 0.33 mL) were added
to a round bottom flask equipped with a stir bar. The mixture was stirred at
80°C for 20 hours. After the reaction, solids precipitated and
dichloromethane (DCM) was added to redissolve the solid. Hexane was
added dropwise to form a layer on DCM. The mixture was allowed to
recrystallize and the contents were filtered using a sintered filter funnel.
The white crystalline solids were collected and dried under vacuum. Yield
= 0.36 g (45%), mp = 112.3°C (see Figures S5-S11, S23). 1H NMR (400
MHz, CDCl3): δ 7.12-7.28 (m, 10H), 6.70 (d, 1JHP = 644 Hz, 1H), 3.70 (s,
2H). 1H NMR (400 MHz, d6-DMSO): δ 8.24 (s, 3H), 7.36-7.47 (m, 5H), 7.23
Understanding the ease of synthesis, the mechanistic details and
selectivity in making ammonium monosubstituted H-phosphonate
salts is key to using them as synthetic precursors for the creation
of new H-phosphonates, such as heterodialkyl H-
phosphonates,[21] or as other synthons, such as in the
phosphorylation of chitosan.[29]
3
(t, 3JHH = 7.9 Hz, 2H), 7.05 (d, JHH = 8.3 Hz, 2H), 6.97 (t, 3JHH = 7.4 Hz,
1H), 6.83 (d, 1JHP = 593 Hz, 1H), 3.98 (s, 2H). 31P NMR (162 MHz, CDCl3):
δ 1.3 (d, 1JPH = 645 Hz). 31P NMR (162 MHz, d6-DMSO): δ -2.5 (d, 1JPH
=
593 Hz). 13C{1H} NMR (100.6 MHz, d6-DMSO): δ 42.35, 120.36, 121.88,
128.30, 128.55, 128.75, 128.99, 134.5. ESI-MS: [C13H16NO3P + H]+ obs’d
266.0944 m/z (cal’d 266.0942 m/z), [C13H16NO3P + BnNH3]+ 373.1668 m/z
(cal’d 373.1676 m/z), [2(C13H16NO3P) + H]+ 531.1808 m/z (cal’d 531.1808
Experimental Section
m/z), [3(C13H16NO3P) + H]+ 796.2649 m/z (cal’d 796.2676 m/z). Tonset
=
General considerations and instrumentation.
131°C; Tpeak = 222°C.
All reagents and solvents were used as received from either Sigma-Aldrich
or pH Scientific, unless otherwise stated. AR grade solvents were used.
1H, 13C, 31P/31P{1H} NMR experiments were recorded using a Bruker
Avance 400 MHz spectrometer. 1H chemical shifts are recorded in parts
per million (ppm) downfield tetramethylsilane and are referenced to
residual protium in the NMR solvent (e.g. CHCl3 δ = 7.26 ppm). Chemical
Synthesis of ethyl H-phosphonate n-butylammonium salt, 3b.
1b (5.0 mmol, 0.67 mL) and n-butylamine (5.0 mmol, 0.50 mL) were added
to a round bottom flask equipped with a stir bar. The mixture was stirred at
80°C for 20 hours. After the reaction, a DCM and hexane mixture was
added to the mixture. The mixture was shaken and sonicated and the
mixture was allowed to stand so the hexane and DCM mixture formed two
layers. The product was dissolved in the DCM layer. The hexane layer was
decanted and the extraction process was repeated three times. The DCM
and hexane residue was remove under vacuum overnight. A colourless,
shifts for carbon are reported in ppm relative to CDCl3 (δ = 77.3 ppm). 31
P
chemical shifts are reported in ppm relative to an external standard of 85%
phosphoric acid (δ = 0 ppm). Data are represented as follows: chemical
shift, multiplicity (app = apparent, br = broad, s = singlet, d = doublet, t =
triplet, q = quartet, p = quintet, m = multiplet), coupling constants in Hertz
(Hz), and integration. High resolution mass spectrometry measurements
were made on a Bruker microTOF-QII mass spectrometer, equipped with
a KD Scientific syringe pump, in positive ion ESI mode. Thermogravimetric
analysis measurements were run with TA Instrument’s thermogravimetric
analyzer Model Q600 IR. The data was processed using TRIOS computer
software connecting to the analyzer. TGA experiments were performed
with the ramp rate of 10°C/min up to 600°C. 50 µL platinum pans (for liquid
samples) and 100 µL ceramic pans (for solid samples) were used to
contain the samples. X-ray diffraction measurements of single crystals (2d
and 4d) were performed on a Rigaku Oxford Diffraction XtaLAB-Synergy-
S single-crystal diffractometer with a PILATUS 200K hybrid pixel array
detector using Cu Kα radiation (λ = 1.54184 Å; Table S1). The data was
processed with the SHELX2018/3[30] and Olex2 1.3[31] software packages.
All non-hydrogen atoms were refined anisotropically. Mercury 2020.1[32]
was used to visualize the molecular structure. All DFT calculations were
done using Gaussian 09.[33] Note: due to the presence of by-products and
multiple ammonium salts, compounds 2a and 2c were not isolated cleanly,
however, were generated under the same conditions as 2b and 2d below.
viscous ionic liquid was formed. Conversion
= 78% (some mixed
ammonium salt remained preventing isolation and purification; see
Figures S12-S15). 1H NMR (400 MHz, CDCl3): δ 8.61 (s, 3H), 6.78 (d, 1JHP
= 611 Hz, 1H), 3.88 (p, 3JHH = 6.9 Hz, 2H), 2.90 (q, 3JHH = 7.0 Hz, 2H), 2.81
3
(m, 2H), 1.67 (m. 2H), 1.41 (m, 4H), 1.27 (t, JHH = 7.0 Hz, 3H), 0.94 (t,
3JHH = 7.5 Hz, 3H). 31P NMR (162 MHz, CDCl3): δ 3.6 (dt, 1JPH = 611 Hz,
3JPH = 8.1 Hz). ESI-MS: [C6H18NO3P + H]+ 184.1098 m/z (cal’d m/z
184.1097), [C6H18NO3P
[C6H18NO3P
n-BuEtNH2]+ 313.2614 m/z (cal’d 313.2615 m/z),
[2(C6H18NO3P) + H]+ 423.2735 m/z (cal’d 423.2747 m/z). Tonset = 162°C;
peak = 181°C.
+
H]+ 212.1414 m/z (cal’d m/z 212.1410),
+
T
Synthesis of phenyl H-phosphonate n-butylammonium salt, 3d.
1d (5.0 mmol, 0.95 mL) and n-butylamine (5.0 mmol, 0.50 mL) were added
to a round bottom flask equipped with a stir bar. The mixture was stirred at
80°C for 20 hours. After the reaction, a DCM and hexane mixture was
added. The mixture was shaken and sonicated. The mixture was allowed
to stand so that hexane and DCM mixture formed two layers. The product
was dissolved in the DCM layer. The hexane layer was decanted and the
extraction process was repeated three times. The DCM and hexane
residue was remove under vacuum overnight. The peach coloured ionic
liquid was formed. Conversion = 48% (presence of phenol and P(OPh)3
remained preventing isolation and purification; see Figures S16-S19). 1H
Synthesis of ethyl H-phosphonate benzylammonium salt, 2b.
1b (5.0 mmol, 0.67 mL) and benzylamine (5.0 mmol, 0.55 mL) were added
to a round bottom flask equipped with a stir bar. The mixture was stirred at
80°C for 20 hours. After the reaction, a DCM and hexane mixture was
added to the mixture. The mixture was shaken and sonicated. The mixture
was allowed to stand so the hexane and DCM mixture formed two layers.
The product dissolved in the DCM layer. The hexane was decanted and
the extraction process was repeated three times. The DCM and hexane
residue was remove under vacuum overnight. A pale-yellow ionic liquid
was formed. Conversion = 95% (some mixed ammonium salt remained
preventing isolation and purification; see Figures S1-S4). 1H NMR (400
MHz, CDCl3): δ 7.28 – 7.50 (m, 5H), 6.63 (d, 1JHP = 616 Hz, 1H), 3.99 (s,
2H), 3.75 (m, 2 x 2H), 1.32 (t, 3JHH = 7.1 Hz, 3H) 1.20 (t, 3JHH = 7.1 Hz, 3H).
31P NMR (162 MHz, CDCl3): δ 3.9 (dt, 1JPH = 616 Hz, 3JPH = 8.0 Hz). ESI-
NMR (400 MHz, CDCl3): δ 7.40 (s, 3H), 7.00-7.30 (m, 5H), 7.10 (d, 1JHP
=
639 Hz, 1H), 2.70 (t, 3JHH = 7.6 Hz, 2H), 1.53 (m, 2H), 1.25 (m, 2H) 0.79 (t,
3JHH = 7.02 Hz, 3H). 31P NMR (162 MHz, CDCl3): δ 1.4 (d, 1JPH = 644 Hz).
ESI-MS: [C10H18NO3P
[C10H18NO3P
BuNH3]+ 305.2004 m/z (cal’d m/z 305.1989 m/z),
[2(C10H18NO3P) + H]+ 463.2145 m/z (cal’d 463.2121 m/z), [2(C10H18NO3P)
BuNH3]+ 536.3038 m/z (cal’d 536.3013 m/z), [2(C10H18NO3P)
(O-)(PhO)P(O)H 2H+] 621.2275 m/z (cal’d 624.2275 m/z),
+
H]+ 232.110 m/z (cal’d 232.1097 m/z),
+
+
+
+
[3(C10H18NO3P) + H+] 694.3158 m/z (cal’d 694.3073 m/z), [3(C10H18NO3P)
+ BuNH3]+ 767.4016 m/z (cal’d 767.4037 m/z). Tonset = 125°C; Tpeak
=
181°C.
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
Synthesis of phenyl H-phosphonate phenylammonium salt, 4d.
1d (3.0 mmol, 0.57 mL) and aniline (3.0 mmol, 0.27 mL) were added to a
round bottom flask equipped with a stir bar. The mixture was stirred at
Keywords: H-phosphonate diesters • ionic salt • amine •
mechanism • dealkylation
80°C for 20 hours.
After the reaction, solids precipitated and
dichloromethane (DCM) was added to dissolve the solid. Hexane was
added dropwise to form a layer on DCM. The mixture was allowed to
recrystallize and the contents were filtered. The white-brownish crystals
were collected, and dried under vacuum. Yield = 0.24 g (40%), mp =
116.4°C. Note this product degrades quite quickly in solution to an
insoluble solid making characterisation a challenge (see Figures S20-S22,
[1] a) K. Troev, Chemistry and Application of H-Phosphonates,
Elsevier Science, 2006, p vii-viii, 11-105,; b) M. K. Sobkowski, A.;
Stawinski, J., Top. Curr. Chem. 2015, 361, 137-177.
[2] a) Y. B. Zhou, J. Yang, T. Q. Chen, S. F. Yin, D. Q. Han and L. B.
Han, Bull. Chem. Soc. Japan 2014, 87, 400-402; b) M. Meazza, A.
Kowalczuk, L. Shirley, J. W. Yang, H. Guo and R. Rios, Adv. Syn. Cat.
2016, 358, 719-723; c) J. Fraser, L. J. Wilson, R. K. Blundell and C.
J. Hayes, Chem. Commun. 2013, 49, 8919-8921; d) X. J. Jin, K.
Yamaguchi and N. Mizuno, Org. Lett. 2013, 15, 418-421; e) J. Q.
Zhang, Y. S. Xiong, A. S. C. Chan and G. Lu, RSC Adv. 2016, 6,
84587-84591; f) G. Wang, Q. Y. Yu, S. Y. Chen and X. Q. Yu,
Tetrahedron Lett. 2013, 54, 6230-6232; g) Y. J. Li and F. S. Liang,
Tetrahedron Lett. 2016, 57, 2931-2934; h) G. Hong, X. Y. Zhu, C. Hu,
A. N. Arum, S. Y. Wu and L. M. Wang, J. Org. Chem. 2016, 81, 6867-
6874.
[3] Y. B. Zhou, S. F. Yin, Y. X. Gao, Y. F. Zhao, M. Goto and L. B. Han,
Angew. Chem. Int. Ed. 2010, 49, 6852-6855.
[4] A. K. Purohit, D. Pardasani, A. Kumar, D. R. Goud, R. Jain and D.
K. Dubey, Tetrahedron Lett. 2015, 56, 4593-4595.
[5] a) J. X. Wang, X. L. Wang, H. J. Li and J. Yan, J. Organomet.
Chem. 2018, 859, 75-79; b) X. H. Zhang, D. G. Wang, D. An, B. S.
Han, X. Song, L. Li, G. Q. Zhang and L. X. Wang, J. Org. Chem. 2018,
83, 1532-1537; c) H. Huang, J. Ash and J. Y. Kang, Org. Biomol.
Chem. 2018, 16, 4236-4242.
[6] a) A. Kondoh and M. Terada, Chem. Eur. J. 2018, 24, 3998-4001;
b) E. O. Yurko, T. V. Gryaznova, K. V. Kholin, V. V. Khrizanforova and
Y. H. Budnikova, Dalton Trans. 2018, 47, 190-196; c) E. O. Yurko, T.
V. Gryaznova, V. V. Khrizanforova, M. N. Khrizanforov, A. V.
Toropchina, Y. H. Budnikova and O. G. Sinyashin, Russ. Chem. Bull.
2018, 67, 102-107; d) M. R. Patil, N. P. Dedhia, A. R. Kapdi and A. V.
Kumar, J. Org. Chem. 2018, 83, 4477-4490; e) T. Q. Chen, C. Q. Zhao
and L. B. Han, J. Am. Chem. Soc. 2018, 140, 3139-3155; f) W. B. Wu,
Y. C. Wong, Z. K. Tan and J. Wu, Catal. Sci.Technol. 2018, 8, 4257-
4263; g) B. Z. Lin, S. S. Shi, R. C. Lin, Y. Q. Cui, M. J. Fang, G. Tang
and Y. F. Zhao, J. Org. Chem. 2018, 83, 6754-6761; h) L. Chen, X. Y.
Fang and Y. X. Zou, Org. Biomol. Chem. 2018, 16, 951-956; i) W. C.
Yang, K. Wei, X. Sun, J. Zhu and L. Wu, Org. Lett. 2018, 20, 3144-
3147; j) A. Matsude, K. Hirano and M. Miura, Org. Lett. 2018, 20,
3553-3556; k) P. H. Li, G. W. Wang, X. J. Zhu and L. Wang,
Tetrahedron 2019, 75, 3448-3455; l) C. Y. Li, T. Q. Chen and L. B.
Han, Dalton Trans. 2016, 45, 14893-14897; m) H. Fu, T. Yang, J. Q.
Shang, J. L. Zhou, M. Sun and Y. M. Li, Org. Chem. Front. 2017, 4,
1777-1780; n) Y. K. Yang, C. Qu, X. L. Chen, K. Sun, L. B. Qu, W. Z.
Bi, H. Hu, R. Li, C. F. Jing, D. H. Wei, S. K. Wei, Y. Q. Sun, H. Liu and
Y. F. Zhao, Org. Lett. 2017, 19, 5864-5867; o) J. Xu, P. B. Zhang, X.
Q. Li, Y. Z. Gao, J. Wu, G. Tang and Y. F. Zhao, Adv. Syn. Cat. 2014,
356, 3331-3335.
[7] a) R. Karpowicz, J. Lewkowski, M. Stasiak, A. Czopor, P. Tokarz,
A. Krol, D. Rogacz and P. Rychter, Phosphorus Sulfur Silicon Relat.
Elem. 2018, 193, 423-436; b) D. T. Keough, D. Rejman, R. Pohl, E.
Zbornikoa, D. Hockova, T. Croll, M. D. Edstein, G. W. Birrell, M.
Chavchich, L. M. J. Naesens, G. K. Pierens, I. M. Brereton and L. W.
Guddat, ACS Chem. Bio. 2018, 13, 82-90; c) A. Kraszewski and J.
Stawinski, Pure Appl. Chem. 2007, 79, 2217-2227; d) M. Froneman
and T. A. Modro, Tetrahedron Lett. 1988, 29, 3327-3330; e) A. B.
Mirgorodskaya, L. Y. Zakharova, R. A. Shagidullina, L. A.
Kudryavtseva, I. M. Shermergorn, V. E. Belskii and B. E. Ivanov, Bull.
Russ. Aca. Sci. Ussr Div. Chem. Sci. 1990, 39, 1013-1017.
S24). 1H NMR (400 MHz, CDCl3): δ 7.00-7.38 (m, 10H), 7.04 (d, 1JHP = 668
1
Hz, 1H). 31P NMR (162 MHz, CDCl3): δ 2.1 (d, JPH = 664 Hz). Tonset
=
113°C; Tpeak = 161°C.
Kinetic monitoring of 1a-1d with benzyl amine.
H-phosphonate diester, 1 (1a: 0.02 mol, 1.80 mL; 1b: 0.05 mol, 8.33 mL;
1c: 0.03 mol, 3.90 mL; 1d: 0.02 mol, 3.80 mL;) was added to a vial or 2
necked round bottom flask containing magnetic stir bar. 1 was brought to
the appropriate temperature (20°C for 1a and 1d or 80°C for 1b and 1c).
Benzylamine (1a: 0.02 mol, 2.20 mL; 1b: 0.05 mol, 5.50 mL; 1c: 0.03 mol,
4.20 mL; 1d: 0.02 mol, 2.20 mL) was added and a stopwatch was started.
0.05 mL of the mixture was extracted using a pipette and transferred to an
amber vial containing 0.45 mL CHCl3. The amber vial was immediately
stored in the fridge to stop the reaction. The samples were analyzed using
31P NMR spectroscopy.
1a: The time points where 0.05 mL sample was extracted were: 4.0, 6.0,
15, 20, 30, 60, 80, 375, 1476 and 1735 min (see Figures S25-S29). 31P
NMR (162 MHz, CDCl3): δ 11.5 (2a, dq, 1JPH = 619 Hz, 3JPH = 12 Hz), 14.2
(5a, ds, 1JPH = 696 Hz, 3JPH = 12 Hz).
1b: The time points where 0.05 mL sample was extracted were: 0.5, 2.0,
4.0, 6.0, 8.0, 10, 60, 300, 1440, 2880 and 7080 min (see Figures S30-
S32). 31P NMR (162 MHz, CDCl3): δ 9.0 (2b, dt, 1JPH = 615 Hz, 3JPH = 8.2
Hz), 17.7 (5b, dp, 1JPH = 637 Hz, 3JPH = 9.6 Hz).
1c: The time points where 0.05 mL sample was extracted were: 10, 20, 60,
180, 300, 1440, 2760, 4200, 5640 and 10020 min (see Figures S33-S39).
31P NMR (162 MHz, CDCl3): δ 3.7 ((i-PrO)2P(O)OH), 7.0 (2c, dd, 1JPH
613 Hz, 1JPH = 9.5 Hz), 15.5 (5c, br d, 1JPH = 635 Hz), 26.9 (6c, dt).
=
1d: The reaction occurred much too fast to monitor over time as the
starting material 1d was already nearly used up when the first spectrum
was taken after 4 min (see Figures S40-S42). 31P NMR (162 MHz, CDCl3):
δ 6.6 (2d, d, 1JPH = 647 Hz), 9.2 (5d, dt, 1JPH = 716 Hz, 3JPH = 9.3 Hz), 15.7
(9d, d, 1JPH = 667 Hz), 17.7 (8d, ds, 1JPH = 647 Hz, 3JPH = 10.5 Hz), 128.0
(P(OPh)3, s).
The effect of amine on the reaction.
For both 1b and 1d, two amines (BnNH2 and n-BuNH2) have been used to
make the ionic salts 2b, 2d, 3b, 3d. Two further reactions with the tertiary
trimethylamine were completed for comparison.
10b: 1b (5.0 mmol, 0.67 mL) and triethylamine (5.0 mmol, 0.70 mL) were
added to a round bottom flask equipped with a stir bar. The mixture was
stirred at 80°C for 20 hours. Conversion = 19% (see Figures S43-S44)
10d: 1d (3.0 mmol, 0.57 mL) and triethylamine (3.0 mmol, 0.42 mL) were
added to a round bottom flask equipped with a stir bar. The mixture was
stirred at 80°C for 20 hours. Conversion = 14% (see Figures S45-S46).
Acknowledgements
K. L. L., J. F., T. S. and E. M. L. would like to acknowledge the
School of Chemical Sciences and the MacDiarmid Institute for
financial support. P.A.H. acknowledges support from the
MacDiarmid Institute for a Personal Post-Doctoral Fellowship. All
calculations were performed using the VUW High Performance
Computing Cluster “Rāpoi”. We thank Tatiana Groutso for
collecting the single crystal X-ray diffraction data.
[8] a) K. Troev, S. Shenkov and U. Delimarinova, Phosphorus Sulfur
Silicon Relat. Elem. 1992, 68, 107-114; b) S. Hossain, S. K. Gupta
and R. Murugavel, CrystEngComm 2015, 17, 4355-4366; c) A.
Lemeune, A. Y. Mitrofanov, Y. Rousselin, C. Stern, R. Guilard, Y. Y.
Enakieva, Y. G. Gorbunova and S. E. Nefedov, Phosphorus Sulfur
Silicon Relat. Elem. 2015, 190, 831-836; d) M. Blaha, M. Trchova, P.
Bober, Z. Moravkova, Z. D. Zujovic, S. K. Filippov, J. Prokes, J. Pilar
and J. Stejskal, Syn. Met. 2017, 232, 79-86; e) Z. Moravkova, M.
Trchova, J. Dybal, M. Blaha and J. Stejskal, J. Appl. Polym. Sci. 2018,
9
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
135; f) H. Busch, E. Schiebel, A. Sickinger and S. Mecking,
Macromolecules 2017, 50, 7901-7910.
[30] G. M. Sheldrick, Acta Crystallogr. C 2015, 71, 3-8.
[31] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and
H. Puschmann, J. Appl. Crystallogr. 2009, 42, 339-341.
[32] C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P.
Shields, R. Taylor, M. Towler and J. van De Streek, J. Appl.
Crystallogr. 2006, 39, 453-457.
[9] a) B. G. Janesko, H. C. Fisher, M. J. Bridle and J. L. Montchamp,
J. Org. Chem. 2015, 80, 10025-10032; b) W. J. Bailey and R. B. Fox,
J. Org. Chem. 1963, 28, 531-534; c) J. Stawinski and A. Kraszewski,
Acc. Chem. Res. 2002, 35, 952-960; d) E. E. Nifant'ev, Russ. Chem.
Rev. 1978, 47, 835-858.
[33] M. J. Frisch; G. W. Trucks; H. B. Schlegel; G. E. Scuseria; M. A.
Robb; J. R. Cheeseman; G. Scalmani; V. Barone; B. Mennucci, G. A.
Petersson; H. Nakatsuji; M. Caricato; X. Li; H. P. Hratchian; A. F.
Izmaylov; J. Bloino; G. Zheng; J. L. Sonnenberg; M. Hada; M. Ehara;
K. Toyota; R. Fukuda; J. Hasegawa; M. Ishida; T. Nakajima; Y.
Honda; O. Kitao; H. Nakai; T. Vreven; J. A. Montgomery Jr. ; J. E.
Peralta; F. Ogliaro; M. Bearpark; J. J. Heyd; E. Brothers; K. N. Kudin;
V. N. Staroverov; R. Kobayashi; J. Normand; K. Raghavachari; A.
Rendell; J. C. Burant; S. S. Iyengar; J. Tomasi; M. Cossi; N. Rega; J.
M. Millam; M. Klene; J. E. Knox; J. B. Cross; V. Bakken; C. Adamo; J.
Jaramillo; R. Gomperts; E. Stratmann; O. Yazyev; A. J. Austin; R.
Cammi; C. Pomelli; J. W. Ochterski; R. L. Martin; K. Morokuma; V. G.
Zakrzewskil; G. A. Voth; P. Salvador; J. J. Dannenberg; S. Dapprich;
A.D. Daniels; Ö. Farkas; J. B. Foresman; J. V. Ortiz; J. Cioslowski; D.
J. Fox; Gaussian, Inc.: Wallingford CT, 2013.
[10] a) B. Kurscheid, W. Wiebe, B. Neumann, H. G. Stammler and B.
Hoge, Eur. J. Inorg. Chem. 2011, 5523-5529; b) D. B. Vincze, P.;
Abranyi-Balogh, P., Phosphorus Sulfur Silicon Relat. Elem. 2019, 194,
359-360; c) J. E. Griffiths and A. B. Burg, J. Am. Chem. Soc. 1960,
82, 1507-1508.
[11] a) K. A. G. t. Petrov, R. G., Russ. Chem. Rev. 1966, 35, 622-631;
b) M. J. Gallagher and H. Honegger, J. Chem. Soc. Chem. Commun.
1978, 54-55.
[12] a) F. H. Westheimer, Acc. Chem. Res. 1968, 1, 70-78; b) Z. Luz
and B. Silver, J. Am. Chem. Soc. 1961, 83, 4518-4521; c) J. P. Guthrie,
J. Am. Chem. Soc. 1977, 99, 3991-4001; d) M. C. Mitchell, R. J. Taylor
and T. P. Kee, Polyhedron 1998, 17, 433-442; e) J. R. Peyser and J.
P. Ferris, Origins Life Evol. B. 2001, 31, 363-380.
[13] A. Sobkowska, M. Sobkowski, J. Cieslak, A. Kraszewski, I. Kers
and J. Stawinski, J. Org. Chem. 1997, 62, 4791-4794.
[14] K. L. Lee, J. Feld, Y. Ben-Tal, Z. Y. Guo, P. Hume and E. M.
Leitao, Asian J. Org. Chem. 2019, 8, 1415-1425.
[15] a) V. Mitova, N. Koseva and K. Troev, RSC Adv. 2014, 4, 64733-
64736; b) E. M. Georgiev, Kaneti, J., Troev, K., Roundhill, D. M., J.
Am. Chem. Soc. 1993, 115, 10964-10973.
[16] a) D. E. Bryant, C. Kilner and T. P. Kee, Inorg. Chim. Acta 2009,
362, 614-616; b) M. D. M. Gray and D. J. H. Smith, Tetrahedron Lett.
1980, 21, 859-860; c) H. Krawczyk, Synth. Commun. 1997, 27, 3151-
3161; d) N. S. Kumar, S. Kumaraswamy, M. A. Said and K. C. K.
Swamy, Org. Process Res. Dev. 2003, 7, 925-928; e) K. Troev and R.
Tsevi, Phosphorus Sulfur Silicon Relat. Elem. 1998, 133, 61-68; f) K.
Troev, Heteroat. Chem. 2000, 11, 205-208.
[17] J. Pitawala, J. Scheers, P. Jacobsson and A. Matic, J. Phys.
Chem. B 2013, 117, 8172-8179.
[18] a) A. E. Jimenez, A. Rossi, M. Fantauzzi, T. Espinosa, J. Arias-
Pardilla, G. Martinez-Nicolas and M. D. Bermudez, ACS Appl. Mater.
Interfaces 2015, 7, 10337-10347; c) T. Espinosa, J. Sanes and M. D.
Bermudez, Coatings 2015, 5, 39-53; d) Y. Y. Han, D. Qiao, S. W.
Zhang and D. P. Feng, Tribol. Int. 2017, 114, 469-477.
[19] a) J. Palgunadi, J. E. Kang, M. Cheong, H. Kim, H. Lee and H. S.
Kim, B. Korean Chem. Soc. 2009, 30, 1749-1754; b) J. Palgunadi, J.
E. Kang, D. Q. Nguyen, J. H. Kim, B. K. Min, S. D. Lee, H. Kim and H.
S. Kim, Thermochimica Acta 2009, 494, 94-98.
[20] Y. Fukaya, R. Asai, S. Kadotani, T. Nokami and T. Itoh, Bull.
Chem. Soc. Jpn. 2016, 89, 879-886.
[21] a) A. D. Dal-Maso, Legendre, F., Blonski, C., Hoffmann, P., Synth.
Commun. 2008, 38, 1688-1693; b) B. Boutevin, Y. Hervaud, T.
Jeanmaire, A. Boulahna and M. Elasri, Phosphorus Sulfur Silicon
Relat. Elem. 2001, 174, 1-14; c) Y. Jin, M. Sun, H. Fu and Y. F. Zhao,
Chem. Lett. 2004, 33, 116-117; d) A. Zwierzak and M. Kluba,
Tetrahedron 1973, 29, 1089-1094.
[22] A. Kers, I. Kers, J. Stawinski, M. Sobkowski and A. Kraszewski,
Tetrahedron 1996, 52, 9931-9944.
[23] a) K. Troev and D. M. Roundhill, Phosphorus Sulfur Silicon Relat.
Elem. 1988, 37, 243-245; b) E. M. Georgiev, R. Tsevi, V. Vassileva,
K. Troev and D. M. Roundhill, Phosphorus Sulfur Silicon Relat. Elem.
1994, 88, 139-145; c) V. Vassileva, E. M. Georgiev, K. Troev and D.
M. Roundhill, Phosphorus Sulfur Silicon Relat. Elem. 1994, 92, 101-
107.
[24] S. Hossain, Gupta, S. K., Murugavel, R., CrystEngComm 2015,
17, 4355-4366.
[25] M. L. Williams, S. P. Holahan, M. E. McCorkill, J. S. Dickmann
and E. Kiran, Thermochim. Acta 2018, 669, 126-139.
[26] Y. Y. Cao and T. C. Mu, Ind. Eng. Chem. Res. 2014, 53, 8651-
8664.
[27] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098-3100; b) A. D. Becke,
J. Chem. Phys. 1993, 98, 5648-5652; c) C. Lee, Yang, W., Parr, R.
G., Phys. Rev. B. 1988, 37, 785-789.
[28] S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys.
2010, 132.
[29] C. Coquery, F. Carosio, C. Negrell, N. Causse, N. Pebere and G.
David, Surf. Interfaces 2019, 16, 59-66.
10
This article is protected by copyright. All rights reserved.

10.1002/chem.202003090
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
=
R
direct
alkyl
O
P
O
P
dealkylation
H
+
H
RO
RO
neat
RO
20-80°C,
N
Bn H2
O
=
R
=
phenyl
R
phenyl
N
BnR' H2
H
via
(PhO)2PH(OH)2
=
R
or
R'
I on it. Dialkyl and diphenyl H-phosphonates undergo dealkylation and dearylation, respectively, upon heating with amines to afford
the corresponding ionic salts. The mechanisms for these processes, which form ammonium monosubstituted H-phosphonate salts
held together by H-bonds, were interrogated both experimentally and theoretically, and the results show that they are very different.
Institute and/or researcher Twitter usernames: @MacDiarmidInsti
11
This article is protected by copyright. All rights reserved.