The structures of phenylphosphonic-acid-mono-(2-hydroxy-phenyl)ester and phenylphosphonic-acid-mono-(3-hydroxy-naphthalen-2-yl)ester - A combined experimental and theoretical study

Article abstract of DOI:10.1016/j.poly.2014.11.017

Two ester derivatives of phenylphosphonic acid were prepared from vicinal aromatic diols and phenylphosphonic dichloride in one-step procedures. The products were characterized by means of multi-nuclear NMR spectroscopy (¹H, ¹³C,

Full text of DOI:10.1016/j.poly.2014.11.017

Polyhedron 87 (2015) 163–169
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
The structures of phenylphosphonic-acid-mono-(2-hydroxy-
phenyl)ester and phenylphosphonic-acid-mono-(3-hydroxy-
naphthalen-2-yl)ester – A combined experimental and theoretical study
⇑
Richard Betz
Nelson Mandela Metropolitan University, Summerstrand Campus (South), University Way, PO Box 77000, Port Elizabeth (Eastern Cape) 6031, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
Two ester derivatives of phenylphosphonic acid were prepared from vicinal aromatic diols and
phenylphosphonic dichloride in one-step procedures. The products were characterized by means of
multi-nuclear NMR spectroscopy (1H, 13C, 31P), mass spectrometry, elemental analysis, UV–Vis spectros-
copy, and melting point as well as vibrational spectroscopy. Their molecular structures were determined
by single crystal X-ray diffraction and found to be open-chain half-esters of the respective diols. Intermo-
lecular contacts and hydrogen bonding were analyzed. Born–Haber-cycles based on DFT calculations
were conducted for the partial hydrolysis of the assumed cyclic phenylphosphonates to assess the
reaction energies for the partial P–O–C cleavage of the chelate rings. The observed partial hydrolysis
was rationalized on grounds of molecular orbital considerations. Experimental data were correlated to
DFT data. The bonding situation of the two compounds was described by means of NBO analyses.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 23 September 2014
Accepted 14 November 2014
Available online 24 November 2014
Keywords:
Phosphorus
NMR spectroscopy
X-ray structure determination
DFT calculations
NBO analysis
1. Introduction
information about cyclic phosphates of nucleosides and some car-
bohydrate derivatives is well-established [10–14]. Several studies
Phosphorus plays a key-role in biochemistry as an important
part of the deoxyribonucleic acid, adenosine triphosphate and many
other compounds in living organisms. The metabolism of nutrients
involves several phosphorylation steps [1]. Founding on the work of
Westheimer and co-workers [2,3], the possible intermediary pen-
tavalency of phosphorus during phosphotransferyl reactions has
been a vivid point of discussion in the literature [4,5]. Recently,
the structure of an intermediate from a phosphoryl transfer reaction
containing a pentacoordinate phosphorus compound was eluci-
dated by means of a single crystal X-ray study [6]. Being ubiquitous
in cellular environments, carbohydrates appear as tempting bond-
ing partners for pentavalent phosphorus. However, sound proof
for the structure of such compounds based on data from single
crystal X-ray studies was lacking completely in the literature up
to several years ago. Important progress was made by Holmes and
co-workers upon the integration of phosphorus into a macrocyclic
framework, and the structures of several carbohydrate-supported
k5-phosphanes were elucidated [7–9].
concerning the hydrolytic stability and cleavage of the cyclic motif
at different pH-values were conducted [15].
The formation of cyclic esters has also been reported for phenyl-
phosphonic acid. Yet, structural information about this class of
molecules is scant and has been focused on the question of the che-
late rings’ conformations [16–21]. Only few work has been done on
the field of hydrolytic stability of these compounds, except for the
ethane-1,2-diol-derived phenylphosphonate [22].
An interesting behaviour has been claimed for the cyclic phen-
ylphosphonates derived from the aromatic diols benzene-1,2-diol
[23] and naphthalene-2,3-diol [24]. These were reported to
undergo P–O–C-cleavage of the chelate ring and the formation of
open-chain half-esters upon contact with air or even prolonged
storage in a desiccator. The confirmation of this heightened sensi-
tivity to moisture and the structures of the respective reaction
products were solemnly based on the acidic behaviour of the
compound and the products of acidic hydrolysis. A thorough char-
acterization of these supposed half-esters was not conducted, the
only physical constant reported being the melting points for both
compounds.
Considerable work was done on the field of the ‘‘precursors’’ of
k5-phosphanes – i.e. the phosphates themselves – where structural
Since this uncertainty about the molecular structures of these
two compounds is an unacceptable starting point for a systematic
study about the hydrolytic stability of cyclic phenylphosphonates,
⇑
Tel.: +27 41 504 4863.
E-mail address: Richard.Betz@nmmu.ac.za
http://dx.doi.org/10.1016/j.poly.2014.11.017
0277-5387/Ó 2014 Elsevier Ltd. All rights reserved.

164
R. Betz / Polyhedron 87 (2015) 163–169
Table 1
the formation of these half-esters was re-investigated and their
molecular structures were determined.
Crystallographic data for the structure determinations of 1 and 2.
1
2
Chemical formula
Mr/g molÀ1
Crystal system
Spacegroup
a (Å)
C
12H11O4P
C16H13O4P
300.23
2. Experimental
250.18
orthorhombic
Pbca
7.6833(2)
16.5277(3)
17.7383(3)
2252.54(8)
8
orthorhombic
P212121
5.3893(2)
12.4298(4)
20.8061(8)
1393.76(9)
4
2.1. Reagents and instrumentation
b (Å)
Phenylphosphonic dichloride (Merck, synthesis grade), pyrocat-
echol (UNILAB, 98.5%) and 2,3-dihydroxynaphthalene (Aldrich,
98.0%) were used as received without further purification. Toluene
(Merck, analytical grade) and triethylamine (Met-U-Ed Laboratory
Supplies, 99%) were dried over molecular sieves (4 Å) and potas-
sium hydroxide pellets, respectively. All operations were carried
out under an atmosphere of dry nitrogen applying standard
Schlenk techniques. A detailed description of the experimental
procedures is given due to the partial scarce descriptions in the
literature.
c (Å)
V (Å3)
Z
q
l
(g cmÀ3
(mmÀ1
)
)
1.475
0.243
1.431
0.210
Crystal size (mm)
h-Range (°)
0.422 Â 0.119 Â 0.117
2.30–28.36
9207
numerical
0.9169–1.0000
20228
2812
0.0210
0.0134
2461
0.535 Â 0.172 Â 0.071
1.96–28.31
8002
numerical
0.9422–1.0000
13691
3450
0.0180
0.0182
3249
Reflections for metrics
Absorption correction
Transmission factors
Measured reflections
Independent reflections
Rint
1H and 13C NMR spectra were recorded on a Bruker Ultrashield
400 Plus spectrometer at 25 °C at 400 MHz and 101 MHz, respec-
tively, and are referenced to internal TMS or the solvent residual
peak of the deuterated solvent [25]. 31P NMR spectra were
recorded on a Bruker Ultrashield 400 Plus spectrometer at 25 °C
at 109 MHz and are referenced to external H3PO4 (w = 85% in
water). IR spectra were recorded on a Bruker Tensor 27 spectrom-
eter equipped with a Bruker Platinum ATR unit. Mass spectra were
obtained on a Waters Synapt G2 instrument. Elemental analyses
were conducted on a Vario Elementar Microcube ELIII system.
UV–Vis spectra were recorded on a Perkin Elmer Lambda 35 spec-
trometer. Melting points were determined on a Stuart Melting
Point SMP30 apparatus and are uncorrected.
Mean
r
(I)/I
Reflections with I P 2
r(I)
x, y (weighting scheme)
Flack parameter
Parameter
0.0479, 0.7459
–
162
0.0368, 0.2618
À0.03(7)
198
Restraints
0
0
R(Fobs
)
0.0298
0.0863
1.069
0.001
0.0281
0.0705
1.068
0.001
Rw(F2)
S
Shift/errormax
Maximum residual
0.330
0.268
density (e ÅÀ3
)
Minimum residual
À0.357
À0.240
density (e ÅÀ3
)
2.2. DFT calculations and NBO analyses
2.4. Preparation of phenylphosphonic-acid-mono-(2-hydroxy-
phenyl)ester, PhPO(OH)OC6H4OH (1)
The structures of 1 and 2 were optimized with GAUSSIAN 09 [26]
on the B3LYP/6-311++G(2d,p) level of theory with an ultrafine
integration grid and very tight convergence criteria as singlet mol-
ecules under exclusion of symmetry in point group C1 applying a
solvent model for chloroform. The starting geometry was obtained
by means of GaussView [27]. Frequency analyses were performed
on the optimized structures to ensure that these represent minima
on the global electronic potential hypersurface. Optimization of the
cyclic phenylphosphonates as well as the other compounds needed
for calculations in the Born–Haber-cycles were treated in the same
manner. Calculations of NMR shifts were conducted on the
PBE1PBE/6-311++G(2d,p) level of theory using the GIAO method
[28] based on the optimized structures. A solvent model for chloro-
form was used in this step as well. The analysis and visualization of
the molecular orbitals was conducted by the Avogardo [29] as well
as the CHEMCRAFT program suite [30].
In a three-necked round-bottomed flask (500 mL) equipped
with a dropping funnel and a pressure-equalizing valve benzene-
1,2-diol (2.75 g, 25.0 mmol) and triethylamine (6.95 mL, 5.06 g,
50.0 mmol) were dissolved in toluene (100 mL). Under external
cooling with an ice-bath phenylphosphonic acid dichloride
(3.51 mL, 4.87 g, 25.0 mmol) was added with vigorous stirring over
the course of 30 min. In a slightly exothermic reaction the
precipitation of a colourless solid was observed. After completion
of addition, stirring was continued at room temperature overnight.
The reaction mixture was filtered and the colourless filtrate was
evaporated to dryness on a rotary evaporator. A colourless oil
was obtained that solidified upon storage in an open flask at room
temperature overnight, yield 5.33 g, 21.3 mmol, 85.2%. Crystals
suitable for the diffraction study were obtained upon recrystalliza-
tion from boiling toluene.
1H NMR (DMSO-d6): d = 11.43 (s, 1 H, OH), 9.06 (s, 1 H, OH),
7.85–7.76 (m, 2 H, Har), 7.59–7.39 (m, 3 H, Har), 7.13–7.09 (m,
1 H, Har), 6.94–6.83 (m, 2 H, Har), 6.71–6.64 (m, 1 H, Har) ppm.
13C NMR (DMSO-d6): d = 148.5 (d, J = 4.8 Hz), 139.0 (d, J = 7.0 Hz),
131.9 (d, J = 3.2 Hz), 131.4 (d, J = 10.0 Hz), 128.3 (d, J = 14.7 Hz),
124.9 (d, J = 1.2 Hz), 121.5 (d, J = 3.2 Hz), 119.1 (d, J = 1.2 Hz),
117.3, 113.0 (d, J = 10.3 Hz) ppm. 31P NMR (DMSO-d6):
d = 14 ppm. Elemental analysis found (calculated for C12H11O4P):
C, 57.81% (57.61%); H, 4.52% (4.43%). ICP-AES, found (calculated
for C12H11O4P): P, 12.11% (12.38%). HRMS (ESI+): m/z = found
251.0473 (calculated for C12H12O4P+1: 251.0473). HRMS (ESIÀ): m/
2.3. Crystallography
The diffraction studies on single crystals were performed at
200(2) K on a Bruker Kappa APEX-II CCD diffractometer (Mo Ka,
k = 0.71073 Å). Structure solutions and refinement procedures
were conducted by means of the SHELX program suite [31]. Crystal-
lographic details of the structures are summarized in Table 1.
Carbon-bound hydrogen atoms were calculated using the riding-
model approximation and were included in the refinement with
their UH values set to 1.2Ueq(C). The hydrogen atoms of the hydro-
xyl groups were located on a DFM and refined freely. Metrical
parameters discussed for the crystal structures were obtained
using PLATON [32], graphical presentations were prepared using
Mercury [33] and ORTEP-III [34].
z = found 249.0316 (calculated for
(neat): = 3367 (m, OH), 3069 (w, CH), 3055 (w, CH), 3029 (w,
C
12H10O4PÀ1 : 249.0317). IR
m
CH), 1602 (m), 1591 (m), 1494 (s), 1467 (m), 1438 (m), 1395 (w),
1351 (w), 1316 (w), 1290 (w), 1265 (m), 1207 (s), 1182 (m),

R. Betz / Polyhedron 87 (2015) 163–169
165
1144 (s), 1098 (s), 1072 (m), 1031 (m), 1005 (m), 995 (m), 987 (s),
942 (m), 924 (s), 877 (m), 828 (m), 761 (m), 750 (s), 721 (m), 700
(m), 690 (m), 617 (w), 595 (m), 568 (m), 554 (m), 526 (s), 484 (m),
459 (m), 426 (m) cmÀ1. UV–Vis (acetonitrile): kmax = 211, 213, 215,
267, 272 nm; UV–Vis (cyclohexane): no distinct absorption max-
ima in the range between 195 and 1100 nm. Melting point: 119–
123 °C (lit.: 123–124 °C [23]).
hydroxyl group on the aromatic diol as well as the hydroxyl group
on phosphorus remain unaffected.
The coordination polyhedron around the central atom can be
described as tetrahedral. P–O bond lengths for all oxygen atoms
are found in the range that was reported for two comparable
monoesters derived from phenol [35] and ethanol [36] and substi-
tuted phenylphosphonic acids. The values are also in good agree-
ment with the most common values reported for other
derivatives of phenylphosphonic acids whose metrical parameters
have been deposited with the Cambridge Structural Database [37].
The least-squares planes through the aromatic moieties enclose
an angle of approximately 57°. Hydrogen bonding is a prominent
feature in the crystal structure of 1. The hydroxyl groups furnish
the formation of infinite strands along [100]. In terms of graph-set
analysis [38,39], two different patterns were observed. On the unary
level, two antidromic chains intercross at the – formally – double-
bonded oxygen atom on phosphorus. The pattern exclusively sup-
ported by the hydroxyl group of the aromatic diol necessitates a
C11(7) descriptor while the chain involving the phosphorus-bonded
hydroxyl groups can be described by means of a C11(4) descriptor.
On the binary level, these chains form rings with a R22(11) descriptor.
2.5. Preparation of phenylphosphonic-acid-mono-(3-hydroxy-
naphthalene-2-yl)ester, PhPO(OH)OC10H6OH (2)
In a three-necked round-bottomed flask (500 mL) equipped
with a dropping funnel and a pressure-equalizing valve naphtha-
lene-2,3-diol (4.00 g, 25.0 mmol) and triethylamine (6.95 mL,
5.06 g, 50.0 mmol) were dissolved in toluene (100 mL). Under
external cooling with an ice-bath phenylphosphonic acid dichlo-
ride (3.51 mL, 4.87 g, 25.0 mmol) was added with vigorous stirring
over the course of 30 min. In a slightly exothermic reaction the pre-
cipitation of a colourless solid was observed. After completion of
addition, stirring was continued at room temperature overnight.
The reaction mixture was filtered and stored in an open flask at
ambient conditions over night. A colourless crystalline solid pre-
cipitated from the filtrate, yield 5.28 g, 17.6 mmol, 70.4%.
In addition, three C–H. . .p contacts stemming from two hydrogen
atoms on the pyrocatechol’s aromatic system and one hydrogen
atom of the phosphorus-bonded phenyl group as donors and both
1H NMR (DMSO-d6): d = 9.07 (s, 1 H, OH), 8.25 (s, 1 H, OH), 7.90–
7.81 (m, 2 H, Har), 7.67–7.61 (m, 3 H, Har), 7.58–7.43 (m, 3 H, Har),
7.35–7.21 (m, 3 H, Har) ppm. 13C NMR (DMSO-d6): d = 148.3 (d,
J = 4.9 Hz), 140.5 (d, J = 7.3 Hz), 132.1, 131.9 (d, J = 2.9 Hz), 131.4
(d, J = 10.1 Hz), 131.2, 129.4, 128.3 (d, J = 14.8 Hz), 127.5 (d,
J = 0.8 Hz), 126.8, 125.4 (d, J = 27.0 Hz), 123.3, 117.8 (d, J =
3.6 Hz), 111.0 ppm. 31P NMR (DMSO-d6): d = 14 ppm. Elemental
analysis found (calculated for C16H13O4P): C 64.24% (64.00%), H
aromatic systems as acceptors can be observed.
p-Stacking does
not contribute to the stabilization of the crystal structure as the
shortest intercentroid distance between two centers of gravity
was measured at 4.6299(8) Å in between the two different aromatic
systems in neighbouring molecules.
An X-ray analysis of the reaction product 2 obtained from phe-
nylphosphonic acid dichloride and naphthalene-2,3-diol yielded a
similar picture for the coordination environment around phospho-
rus (Fig. 2). P–O bond lengths are in good agreement with the val-
ues apparent in 1. The phosphorus atom is present in the center of
a tetrahedral coordination polyhedron.
4.28% (4.36%). ICP-AES, found (calculated for
10.41% (10.32%). HRMS (ESIÀ): m/z = found 299.0480 (calculated
for C16H12O4PÀ: 299.0473). IR (neat):
= 3387 (m, OH), 3062 (w,
C16H13O4P): P
m
CH), 2962 (w, CH), 1638 (w), 1594 (w), 1510 (s), 1471 (s), 1458
(m), 1438 (m), 1400 (w), 1366 (w), 1348 (w), 1320 (w), 1273
(m), 1260 (m), 1205 (m), 1193 (m), 1168 (m), 1135 (s), 1097 (s),
1071 (m), 1029 (m), 1001 (s), 986 (s), 949 (m), 922 (m), 892 (s),
880 (m), 867 (s), 808 (m), 756 (m), 744 (s), 718 (m), 691 (m),
655 (m), 619 (m), 590 (m), 550 (m), 525 (s), 496 (s), 473 (s), 445
The least-squares planes through the aromatic moieties enclose
an angle slightly below 45°. The hydrogen bonding system in the
crystal structure of 2 slightly differs from the one observed in the
crystal structure of 1. In the naphthalene-2,3-diol-supported phen-
yphosphonate, two homodromic chains were observed with a C11(4)
and
a
C11(7) descriptor on the unitary level of graph-set
(m), 423 (m), 413 (s), 357 (w) cmÀ1
. UV–Vis (acetonitrile):
analysis[38,39]. Both chains intercross at the – formally – dou-
ble-bonded oxygen atom on phosphorus and furnish the formation
of cyclic patterns with a R23(13) descriptor on the binary level. In
total, strands along [100] are formed. In addition, two C–H. . .O
contacts whose range falls by more than 0.1 Å below the sum of
van-der-Waals radii of the atoms participating in them can be
observed in the crystal structure of 2. The latter are exclusively
supported by one of the hydrogen atoms in ortho and in meta posi-
tion on the phosphorus-bonded phenyl group and exclusively
apply the oxygen atom of the free hydroxyl group of the aromatic
kmax = 266, 272, 313, 327 nm; UV–Vis (cyclohexane): kmax = 213
nm. Melting point: 139–142 °C (lit: 145–147 °C [24]).
3. Results and discussion
3.1. Crystal structure analysis
Phenylphosphonic-acid-mono-(2-hydroxy-phenyl)ester (1) was
first prepared by Berlin and Nagabhushanam upon the attempted
synthesis of the cyclic ester of phenylphosphonic acid and
benzene-1,2-diol. According to an infrared analysis conducted
immediately after the workup the desired cyclic ester was claimed
to have been isolated by the authors. The compound was then
reported to undergo hydrolytic cleavage of the chelate ring even
upon storage in a desiccator [23]. A similar behaviour was
observed by these authors for the naphthalene-2,3-diol-derived
phenylphosphonate 2 (cf. Scheme 1) [24].
The results of a single crystal X-ray analysis on the solidified
compound, 1, obtained after reacting phenylphosphonic dichloride
and benzene-1,2-diol confirmed the findings by Berlin and
Nagabhushanam [23]. In the molecule, a benzene-1,2-diol moiety
acts as a monodentate ligand and is esterified with phenylphos-
phonic acid by only one of its hydroxyl groups (Fig. 1). The second
diol as acceptor. No C–H. . .
p contacts are observed, however, the
shortest intercentroid distance between two centers of gravity –
found in between the two different ring systems of the aromatic
diol in neighbouring molecules – is measured at a slightly shorter
value than in 1, at 4.5491(12) Å.
3.2. Physical and spectroscopic properties
13C NMR spectra of crystals of 1 and 2 dissolved in DMSO-d6
show the expected number of resonances in the aromatic region.
13C–31P coupling is observed for most of the resonances. 1H NMR
spectra of both compounds exhibit signals indicative for the pres-
ence of acidic hydroxyl groups. For both compounds, a 31P reso-
nance at +14 ppm is detected. IR spectra confirm the presence of
free hydroxyl groups.

166
R. Betz / Polyhedron 87 (2015) 163–169
Scheme 1. Synthetic pathways for compound 1 and 2.
acid dichloride and benzene-1,2-diol was repeated as described
in the experimental section, however, under inert conditions, i.e.
a blanket of dry nitrogen in flame-dried glassware and the use of
scrupulously dried solvents (toluene) and reagents (triethylamine).
31P NMR spectra were recorded on samples taken directly from the
reaction mixture as well as an aliquot of the reaction mixture
spiked with several drops of water in the NMR sample tube. While
the anhydrous sample showed a single resonance at 48 ppm, the
sample that was mixed with several drops of water lacked this sig-
nal entirely. Instead, a resonance at 30 ppm was recorded in this
case. The latter is in agreement with 31P NMR spectra of compound
1 recorded in the same solvent. The findings for the reaction mix-
ture remain unchanged over the course of several weeks of storage
at room temperature if inert conditions are maintained. A sample
of the reaction mixture stored in a beaker with contact to the
atmosphere, however, shows the disappearance of the resonance
at 48 ppm and the emergence of the resonance at 30 ppm within
24 h. Therefore, it can be concluded that, as an intermediate of
the reaction described in the experimental section, a phosphorus
compound sensitive to hydrolysis is formed that undergoes
hydrolytic cleavage to yield compound 1. Although no parallel
experimental sequence was conducted for the reaction between
naphthalin-2,3-diol and phenylphosphonic dichloride under inert
conditions as just described for benzene-1,2-diol, a corresponding
behaviour ought to be expected on grounds of the similar open-
chain reaction product 2.
Fig. 1. Structure of PhPO(OH)O-C6H4OH (1) (50% probability ellipsoids). Selected
individual bond lengths (in Å), angles and torsional angles (in °): P1–O1 1.5398(9),
P1–O2 1.4884(9), P1–O21 1.5890(9), P1–C11 1.7791(12), O2–P1–O1 115.35(5),
O2–P1–O21 106.69(5), O1–P1–O21 104.34(5), O2–P1–C11 112.04(6), O1–P1–C11
108.39(6), O21–P1–C11 109.65(5), C11–P1–O21–C21 À44.71(10).
3.4. Quantum chemical calculations
3.4.1. Geometric parameters and NMR shifts
Starting from a geometry whose atomic coordinates resemble
the structures obtained by X-ray diffraction, the optimized struc-
tures of 1 and 2 were found to closely maintain the found confor-
mations in the solid state. The geometric parameters calculated for
1 and 2 are in good agreement with the experimentally-found val-
ues (cf. Table 2). The slight deviations apparent can be rationalized
by taking into account packing effects in the crystal that are not
accounted for by the DFT calculations applying a solvent model.
The optimized structures were analyzed by computational
methods with regards to their isotropic NMR shielding tensors at
the PBE1PBE/6-311++G(2d,p) level of theory applying a solvent
model for chloroform. The calculated NMR shifts for phosphorus
were found at +16 ppm for compound 1 and at +17 ppm for com-
pound 2. Taking into account the marked solvent effect apparent
in 31P NMR spectroscopy, these values are in good agreement with
the ones found experimentally for both compounds dissolved in
DMSO-d6 (+14 ppm).
Fig. 2. Structure of PhPO(OH)O-C10H6OH (2) (50% probability ellipsoids). Selected
individual bond lengths (in Å), angles and torsional angles (in °): P1–O1 1.5314(11),
P1–O2 1.4935(10), P1–O21 1.6041(10), P1–C11 1.7744(13), O2–P1–O1 116.29(6),
O2–P1–O21 110.41(6), O1–P1–O21 102.56(6), O2–P1–C11 111.94(6), O1–P1–C11
106.63(6), O21–P1–C11 108.38(6), C11–P1–O21–C21 69.81(11).
Both compounds were found to be strong acids in water. The pH
value for a saturated aqueous solution of 1 was measured at
approximately 1.0, the corresponding value for 2 was found at
around 1.5 (both determinations were observed by means of a
colour-changing indicator stick). The solubility of 1 – where a con-
centration slightly below 0.1 m can be obtained – in water is mark-
edly higher than that of 2.
3.3. 31P NMR study
3.4.2. Bonding situation
To assess the course of the reaction and the role the absence and
presence of water plays, the reaction between phenylphosphonic
To assess the bonding situation in 1 and 2 – especially with
regard to the P@O bond – NBO analyses were conducted for both

R. Betz / Polyhedron 87 (2015) 163–169
167
Table 2
For 1 as well as for 2, the break-up of the chelate ring upon the
addition of one molecule of water was found to be exothermic as
was the full hydrolysis to phenylphosphonic acid and the respec-
tive aromatic diol (cf. Table 4).
Comparison of selected averaged experimental bond lengths and angles (XRD) in 1
and 2 with the corresponding parameters calculated at the B3LYP/6-311++G(2d,p)
level of theory applying a solvent model for chloroform (DFT).
1
2
On grounds of these values, the full hydrolysis of the cyclic pre-
cursors of 1 and 2 should, therefore, be even more favourable than
the cleavage of only one P–O–C bond which is in contradiction to
the experimental findings as the two compounds can be stored
at ambient conditions without further modification. To explain this
behaviour, it is instructive to consider the mechanism of the
hydrolysis reactions observed in the light of the respective molec-
ular orbitals found in the individual compounds. For both – the
partial as well as the full hydrolysis via a stepwise process – one
necessary step will be the addition of a molecule of water to the
phosphorus atom. As can be seen from Fig. 3 depicting the highest
occupied molecular orbital (HOMO) and the lowest unoccupied
molecular orbital (LUMO) of the cyclic precursor of 1, the LUMO
is centered on the phosphorus atom. This allows for the attack of
a molecule of water on the phosphorus atom, therefore facilitating
the hydrolysis.
A different picture is apparent for the HOMO and the LUMO of 1
shown in Fig. 4. While the HOMO is centered on the aromatic sys-
tem of the catechol moiety and neither engulfs the phosphorus
atom nor the phenyl group bonded to it, the LUMO exclusively
spans the phosphorus-bonded phenyl group, excluding the phos-
phorus atom. The addition of another molecule of water to the lat-
ter – that would be a crucial step in the second stage of hydrolysis
to effect the split into phenylphosphonic acid and catechol – is
hampered as a consequence.
A similar situation is at hand for the HOMO and the LUMO in
the cyclic precursor of 2 (Fig. 5) as well as 2 (Fig. 6) itself, therefore
explaining the identical behaviour of partial hydrolysis with the
cleavage of only one P–O–C bond. While the HOMO in 2 is centered
on the – formally – double-bonded oxygen atom, the LUMO in 2 is
centered on the hydroxyl group bonded to the phosphorus atom.
Both molecular orbitals do not extend to the phosphorus atom.
The energetic gap between the respective HOMO and the LUMO
is calculated to 492 kJ/mol in 1 and 437 kJ/mol in 2. Within a
margin of 3% these values are in good agreement with the values
calculated for the corresponding energetic gaps for the respective
cyclic precursors of 1 and 2 therefore ruling out the observed
partial hydrolysis to be caused purely on grounds of differing
activation barriers.
XRD
DFT
XRD
DFT
d(P@O) (Å)
1.488
1.540
1.473
1.607
1.619
1.797
116.01
109.84
100.45
1.494
1.531
1.604
1.774
116.29
110.41
102.56
1.474
1.598
1.634
1.791
116.65
113.84
98.64
d(P–OH) (Å)
d(P–OC) (Å)
d(P–C) (Å)
\(O–P–OH) (°)
\(O–P–OC) (°)
\(HO–P–OC) (°)
1.589
1.779
115.35
106.69
104.34
compounds. In both molecules, the d orbital contribution to the
bonds established in between the phosphorus atom and the atoms
it is bonded to, respectively, is small. The hybridization of the
phosphorus atom is found at sp2.04d0.04 (1) and sp1.96d0.04 (2) for
the P@O bond. The latter finding is also corroborated by the respec-
tive natural electronic configuration of phosphorus – s0.78p1.71d0.09
(1) and s0.79p1.71d0.09 (2) – that shows the negligible participation
of diffuse orbitals in the bonds established by it. Invariably, the
P@O bond shows around 26% phosphorus character with
a
sp1.77d0.02 hybridization on the oxygen atom. While the natural
charge for the phosphorus atom is positive and nearly identical
in value in 1 and 2, all other atoms bonded to the phosphorus atom
bear a negative natural charge (cf. Table 3).
If Rydberg orbitals as well as core-centered electrons are
neglected, second order perturbation theory calculations invari-
ably show the interaction of one lone pair stemming from the
phosphorus-bonded oxygen atom of the respective aromatic diol
in 1 and 2 with the central atom to be the most stabilizing elec-
tronic factor within each compound. For 1, the stabilization energy
is calculated at 1420 kJ/mol while for 2 a lower value of 1330 kJ/
mol is predicted.
3.4.3. Thermochemistry
To estimate the ease of hydrolytic cleavage of the P–O–C chelate
ring, Born–Haber-cycles for both compounds were calculated for
the partial hydrolysis of the cyclic precursor compounds on
grounds of the vibrational analyses calculated at the B3LYP/6-
311++G(2d,p) level of theory for all molecular entities involved.
The general exothermicity for the partial and full hydrolysis
reactions is in accordance with the observations made for the
base-catalyzed hydrolysis of the cyclic phenylphosphonate derived
from ethane-1,2-diol. In this case, this behaviour was attributed to
ring tension among the five-membered chelate ring [22]. Taking
into account the preparative procedure for the synthesis of this
Table 3
Natural charges of the phosphorus atom and the atoms it is bonded to in 1 and 2 with
the corresponding parameters calculated at the B3LYP/6-311++G(2d,p) level of theory
applying a solvent model for chloroform (DFT).
1
2
P
P@O
P–O(Catechol)
P–O(H)
P–C
+2.39
À1.10
À0.84
À1.00
À0.46
+2.38
À1.10
À0.84
À0.99
À0.46
´
compound given by Holy [40] where the organic phase containing
the cyclic ester derived from this aliphatic diol was washed with
water, the cyclic phenylphosphonate derived from ethane-1,2-diol
seems to be inert against water at least for a short time. DFT calcu-
lations on the cyclic phenylphosphate of ethane-1,2-diol as well as
its half-ester corresponding to the two title compounds allowed an
assessment of the reaction energies for the partial as well as the
full hydrolysis of this ester derived from the simplest vicinal ali-
phatic diol (cf. Table 5). An inspection of the pertaining HOMO
and LUMO in the cyclic phenylphosphate of ethane-1,2-diol sup-
ports its observed persistence in water as both molecular orbitals
of this compound do not span its phosphorus atom.
Table 4
Calculated reaction energies for the hydrolytic cleavage of the respective cyclic
precursors to 1 and 2 upon addition of one molecule of water (‘‘Partial’’) as well as the
full hydrolysis to phenylphosphonic acid and the respective aromatic diol (‘‘Full’’). All
values are given in kJ/mol.
Esterified diol
Catechol
Naphthalin-2,3-diol
The reason for the heightened sensitivity of the respective cyclic
precursors of 1 and 2 towards atmospheric moisture in comparison
to the cyclic phenylphosphonic acid ester of ethane-1,2-diol could
also be explained on grounds of the higher acidity of the hydroxyl
groups on the aromatic systems in aqueous solution (pKa1
Partial (to 1)
Full
Partial (to 2)
Full
D
D
D
Etot
E0
H298
À23.7
À13.2
À15.8
À48.4
À32.6
À39.3
À27.0
À16.5
À19.0
À48.2
À32.1
À39.1

168
R. Betz / Polyhedron 87 (2015) 163–169
Fig. 3. The highest occupied (HOMO) as well as the lowest unoccupied (LUMO) molecular orbital in the cyclic precursor of 1. The isocontour surfaces were drawn at the 0.005
cut-off level.
Fig. 4. The highest occupied (HOMO) as well as the lowest unoccupied (LUMO) molecular orbital in 1. The isocontour surfaces were drawn at the 0.002 and the 0.07 cut-off
level.
Fig. 5. The highest occupied (HOMO) as well as the lowest unoccupied (LUMO) molecular orbital in the cyclic precursor of 2. The isocontour surfaces were drawn at the 0.002
and 0.003 cut-off level.
Fig. 6. The highest occupied (HOMO) as well as the lowest unoccupied (LUMO) molecular orbital in 2. The isocontour surfaces were drawn at the 0.02 cut-off level.

R. Betz / Polyhedron 87 (2015) 163–169
[5] R.R. Holmes, Acc. Chem. Res. 37 (2004) 746.
169
Table 5
[6] S.D. Lahiri, G. Zhang, D. Dunaway-Mariano, K.N. Allen, Science 299 (2003)
2067.
[7] N.V. Timosheva, A. Chandrasekaran, R.R. Holmes, J. Am. Chem. Soc. 127 (2005)
12474.
[8] N.V. Timosheva, A. Chandrasekaran, R.R. Holmes, Inorg. Chem. 45 (2006) 3113.
[9] N.V. Timosheva, A. Chandrasekaran, R.R. Holmes, Inorg. Chem. 45 (2006)
10836.
Calculated (B3LYP/6-311++G(2d,p) level of theory applying solvent models for
chloroform) reaction energies for the hydrolytic cleavage of the cyclic diester of
phenylphosphonic acid and ethane-1,2-diol to the respective monodentate ester upon
addition of one molecule of water to PhPO(OH)–OCH2CH2OH (‘‘Partial’’) as well as the
full hydrolysis to phenylphosphonic acid and ethane-1,2-diol (‘‘Full’’). All values are
given in kJ/mol.
[10] K. Watenpaugh, J. Dow, L.H. Jensen, S. Furberg, Science 159 (1968) 206.
[11] Q. Zheng, S. Dou, Y. Gu, Wuli Xuebao 30 (1981) 1369.
[12] M.E. Druyan, M. Sparagana, S.W. Peterson, J. Cycl. Nucl. Res. 2 (1976) 373.
[13] B.S. Reddy, W. Saenger, Acta Crystallogr., Sect. B: Struct. Sci. 34 (1978) 1520.
[14] F. Sartilli-Piscil, S. Cruz, M. Sanchez, H. Hopfl, C. Anaya de Parodi, L. Quinterno,
Tetrahedron 59 (2003) 4077.
Partial
Full
D
D
D
Etot
E0
H298
À17.7
À10.4
À10.9
À15.1
À3.1
À7.4
[15] H.G. Khorana, G.M. Tener, R.S. Wright, J.G. Moffatt, J. Am. Chem. Soc. 79 (1957)
430.
[16] R.C.G. Killean, J.L. Lawrence, I.M. Magennis, Acta Crystallogr., Sect. B: Struct.
Sci. 27 (1971) 189.
[17] D.C. Craig, R.F. Toia, V. Wacher, Aust. J. Chem. 42 (1989) 977.
[18] W.N. Setzer, M. Brown, A. Arif, D.G. Vanderveer, Phosphorus, Sulfur Silicon
Relat. Elem. 54 (1990) 187.
[19] W.N. Setzer, M.L. Brown, C.-K. Wu, E.J. Meehan, Heteroatom. Chem. 2 (1991)
533.
benzene-1,2-diol: 10.87 [41] or 9.85 [42], pKa1 naphthalene-2,3-
diol 8.6 [43], pKa1 ethane-1,2-diol: 14.22 [42]). This denotes the
precursors of 1 and 2 to more closely resemble mixed acid anhy-
drides instead of cyclic esters of phenylphosphonic acid, therefore
explaining a heightened degree of hydrolytic sensitivity.
4. Conclusion
[20] I.A. Litvinov, O.N. Kataeva, V.A. Naumov, R.P. Arshinova, S.G. Gnevashev, Russ.
Chem. Bull. 40 (1991) 196 (original: Izv. Nats. Akad. Nauk. Resp. Kaz., Ser.
Khim. 1 (1991) 214).
The X-ray structure analyses on 1 and 2 confirm the correct
formulation of Berlin’s and Nagabhushanam’s reaction sequences.
For both compounds, the supposed hydrolytic cleavage of the
respective cyclic precursors was found to have occurred. Upon
Born–Haber-cycles based on DFT calculations this partial hydroly-
sis was found to be exothermic. In contrast to a cyclic ester of phe-
nylphosphonic acid derived from the simplest aliphatic vicinal diol,
the P–O–C cleavage among the chelate ring of the corresponding
compounds applying aromatic vicinal diols is already induced by
atmospheric moisture while no further hydrolysis is observed.
The latter experimental findings can be rationalized on grounds
of molecular orbital considerations that explain the (non-)feasibil-
ity of the addition of a molecule of water to the central phosphorus
atom in the various compounds. With this knowledge at hand a
concise study about the formation and hydrolytic stability of cyclic
esters of phenylphosphonic acid can be conducted.
[21] N.S. Magomedova, A.N. Sobolev, M.P. Koroteev, N.M. Pugashova, V.K. Bel’skii,
É.E. Nifant’ev, J. Struct. Chem. 32 (1991) 368 (original: Zh. Strukt. Khim. 32
(1991) 67).
[22] J.-C. Yang, D.G. Gorenstein, Tetrahedron 43 (1987) 479.
[23] K.D. Berlin, M. Nagabhushanam, J. Org. Chem. 29 (1964) 2056.
[24] K.D. Berlin, M. Nagabhushanam, Tetrahedron 20 (1964) 2709.
[25] H.E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 62 (1997) 7512.
[26] 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, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, 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, 2009.
[27] R. Dennington, T. Keith, J. Millam, Semichem Inc., Shawnee Mission KS, 2009.
[28] K. Wolinski, J.F. Hinton, P. Pulay, J. Am. Chem. Soc. 112 (1990) 8251.
[29] M.D. Hanwell, D.E. Curtis, D.C. Lonie, T. Vandermeersch, E. Zurek, G.R.
Hutchison, J. Cheminf. 4 (2012) 1.
Acknowledgements
[30] G.A.
Zhurko,
Chemcraft
(Version
1.7,
build
132).
<http://
www.chemcraftprog.com>.
The author acknowledges financial support by Nelson Mandela
Metropolitan University – South Africa. The Center for High Perfor-
mance Computing in Cape Town is thanked for generous allocation
of calculation time. Mr. Felix Odame is thanked for helpful
discussions.
[31] G.M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr. 64 (2008) 112.
[32] A.L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr. 65 (2009) 148.
[33] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock, L.
Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, J. Appl. Crystallogr. 41
(2008) 466.
[34] L.J. Farrugia, J. Appl. Crystallogr. 45 (2012) 849.
[35] T. Miyagawa, A. Furuko, K. Maeda, H. Katagiri, Y. Furusho, E. Yashima, J. Am.
Chem. Soc. 127 (2005) 5018.
[36] C. Nolde, M. Schürmann, M. Mehring, Z. Anorg. Allg. Chem. 633 (2007) 142.
[37] F.H. Allen, Acta Crystallogr., Sect. B: Struct. Sci. 58 (2002) 380.
[38] J. Bernstein, R.E. Davis, L. Shimoni, N.-L. Chang, Angew. Chem., Int. Ed. Engl. 34
(1995) 1555.
[39] M.C. Etter, J.C. MacDonald, J. Bernstein, Acta Crystallogr., Sect. B: Struct. Sci. 46
(1990) 256.
Appendix A. Supplementary data
CCDC 1024667 and CCDC 1024668 contain the supplementary
crystallographic data for compound 1 and compound 2. These data
can be obtained free of charge via http://www.ccdc.cam.ac.uk/con-
ts/retrieving.html, or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-
336-033; or e-mail: deposit@ccdc.cam.ac.uk.
[40] A. Holy´, Synthesis 4 (1998) 381.
[41] J. Lacour, C. Ginglinger, C. Grivet, G. Bernardinelli, Angew. Chem., Int. Ed. Engl.
36 (1997) 608.
[42] D.R. Lide, Handbook of Chemistry and Physics, 74th ed., CRC Press, 1993.
[43] J.T. van Germert, Aust. J. Chem. 21 (1968) 2203.
References
[1] L. Stryer (Ed.), Biochemistry, third ed., W. H. Freeman and co., New York, 1988.
[2] F.H. Westheimer, Acc. Chem. Res. 1 (1968) 70.
[3] J.A. Gerlt, F.H. Westheimer, J.M. Sturtevant, J. Biol. Chem. 250 (1975) 5059.
[4] R.R. Holmes, Acc. Chem. Res. 31 (1998) 535.