A study on the kabachnik - Fields reaction of benzaldehyde, cyclohexylamine, and dialkyl phosphites

Article abstract of DOI:10.1002/hc.20767

The Kabachnik-Fields reaction of benzaldehyde, cyclohexylamine, and dimethyl phosphite carried out at 80°C in acetonitrile takes place via an imine (PhCi￡Ni-^cHex) intermediate, as the monitoring by in situ Fourier transform IR spectroscopy suggested. The corresponding α-hydroxyphosphonate was also formed in a quantity of 13% that was not converted to α-aminophosphonate under the conditions applied. The outcome was similar to the Kabachnik-Fields reaction with diethyl phosphite as the P-component. Molecular modeling and subsequent DFT calculations carried out under solventless conditions supported the experimental results and indicated the formation of a high number of ideally positioned H bonds as the key determinant for the conformation of the starting, intermediate, and product states. The relative energies of the possible intermediates were in accord with the observation that the formation of the α-hydroxyphosphonate is a "dead-end" route.

Full text of DOI:10.1002/hc.20767

Heteroatom Chemistry
Volume 23, Number 2, 2012
A Study on the Kabachnik–Fields Reaction of
Benzaldehyde, Cyclohexylamine, and Dialkyl
Phosphites
Gyo¨rgy Keglevich,1 No´ra Zsuzsa Kiss,1 Do´ra K. Menyha´rd,2
Andra´s Fehe´rva´ri,1 and Istva´n Csontos1
1Department of Organic Chemistry and Technology, Budapest University of Technology and
Economics, H-1521 Budapest, Hungary
2Laboratory of Structural Chemistry and Biology, Institute of Chemistry, Eo¨tvo¨s Lo´ra´nd
University, H-1117 Budapest, Hungary
Received 4 July 2011; revised 14 September 2011
INTRODUCTION
ABSTRACT: The Kabachnik–Fields reaction of ben-
zaldehyde, cyclohexylamine, and dimethyl phosphite
carried out at 80◦C in acetonitrile takes place via
an imine (PhC N cHex) intermediate, as the moni-
toring by in situ Fourier transform IR spectroscopy
suggested. The corresponding α-hydroxyphosphonate
was also formed in a quantity of 13% that was
not converted to α-aminophosphonate under the con-
ditions applied. The outcome was similar to the
Kabachnik–Fields reaction with diethyl phosphite as
the P-component. Molecular modeling and subsequent
DFT calculations carried out under solventless con-
ditions supported the experimental results and indi-
cated the formation of a high number of ideally po-
sitioned H bonds as the key determinant for the con-
formation of the starting, intermediate, and product
states. The relative energies of the possible intermedi-
ates were in accord with the observation that the for-
mation of the α-hydroxyphosphonate is a “dead-end”
α-Aminophosphonic acids may be considered
as phosphorus analogues of α-amino acids [1].
They can act as antimetabolites, for which they
are important targets for drug research because
of their low mammalian toxicity [2]. Diaryl α-
aminophosphonate derivatives are, for example,
selective and highly potent inhibitors of serine
proteases and thus can mediate the pathophys-
ical processes of cancer growth, metastasis, os-
teoarthritis, or heart failure [3]. Dialkylglycine de-
carboxylase [4] and leucine aminopeptidase [5] are
also effectively inhibited by α-aminophosphonates.
Cyanoacrylate [6] and amide derivatives [7] of
α-aminophosphonates are antivirally active com-
pounds and inactivators of tobacco mosaic virus.
Furthermore, certain α-aminophosphonates proved
to be suitable for the design of continuous drug-
release devices due to their ability to increase
the membrane permeability of a hydrophilic probe
molecule [8].
ꢀ
C
route.
2011 Wiley Periodicals, Inc. Heteroatom Chem
23:171–178, 2012; View this article online at wileyonlineli-
brary.com. DOI 10.1002/hc.20767
The solvent-free, catalyst-free, Kabachnik–
Fields reaction provides a green procedure for the
synthesis of such compounds [9]. Although the reac-
tion itself has been known for more than 50 years
[10,11], an up-to-date and comprehensive knowl-
edge of its mechanism is still missing.
Correspondence to: Gyo¨rgy Keglevich; e-mail: gkeglevich@mail
.bme.hu.
Contract grant sponsor: Hungarian Scientific and Research
Fund.
Cherkasov et al. studied the mechanism of the
Kabachnik–Fields reaction in detail. One possibility
Contract grant number: OTKA K83118.
2011 Wiley Periodicals, Inc.
c
ꢀ
171

172 Keglevich et al.
H
H
Route A
δ− δ+
(RO)2PH
O
H
PhCHO
+
(RO)2PH
O
H
N
H
H2N
(RO)2P
O
H
H2N
PhCH
N
H
2
+
O
H
(RO)2P CH OH
Ph
H
(RO)2P CH NH
Ph
H
H2N
1
− H2O
PhC
OH
O
O
O
O
(RO)2P
(RO)2P CH
H
Ph
SCHEME 1 From Refs. [12,15].
H
H2N
3
Route B
R = Me (a), Et (b)
is that an imine is formed from the carbonyl com-
pound and the (primary) amine, and then the di-
alkyl phosphite is added onto the C N moiety of
the imine. The other route that they considered in-
volves the formation of an α-hydroxyphosphonate by
the addition of the dialkyl phosphite to the carbonyl
group of the oxo component. Then the hydroxyphos-
phonate undergoes substitution by the amine to af-
ford α-aminophosphonate. Mainly on the basis of ki-
netic studies, it was thought that the mechanism is
dependent on the nature of the reactants. For exam-
ple, the condensation of aniline, benzaldehyde, and
a dialkyl phosphite was shown to follow the “imine”
mechanism [12,13]. A similar route was experienced
for the reaction of benzaldehyde, propylamine, and
diethyl phosphite [14].
In another example, Cherkasov et al. suggested
that the reaction of the more nucleophilic cyclohexyl-
amine, benzaldehyde, and a dialkyl phosphite takes
place via the “hydroxyphosphonate” route. Here, an
interaction was shown to precede the addition of
the dialkyl phosphite on the C O group of the oxo
compound. According to this, an H bond is formed
between the P(O)H moiety of the phosphite and the
nitrogen atom of the amine (Scheme 1) [12,15].
Later, however, Matveeva and Zefirov proved
that the condensation of cyclohexylamine, benzalde-
hyde, and dialkyl phosphite follows the “imine route”
and concluded that there is no experimental evi-
dence for the hydroxyphosphonate route being in-
volved [16].
SCHEME 3
dependent on the components of the reaction, but
the “imine” route is more general than the pathway
involving an “α-hydroxyphosphonate” intermediate
[1]. Gancarz and Gancarz suggested that a reversible
formation of the α-hydroxyphosphonate may also
occur; at the same time, if it is rearranged to the
corresponding phosphate, this is a “dead-end” route
[18]. It can be said that in the Kabachnik–Fields reac-
tion, a soft nucleophile, that is the dialkyl phosphite,
and a hard nucleophile, that is the amine, compete
for the electrophilic carbonyl compound. The softer
the carbonyl compound is, the faster it reacts with
the softer P nucleophile and the slower it reacts with
the harder amine nucleophile [19].
On the basis of the contradicting literature data,
it was a challenge for us to study the Kabachnik–
Fields reaction of cyclohexylamine, benzaldehyde,
and dialkyl phosphites in detail. We wished to clarify
which mechanism is effective.
RESULTS AND DISCUSSION
The model reaction studied was the condensa-
tion of cyclohexylamine, benzaldehyde, and dialkyl
phosphites affording dialkyl α-cyclohexylamino-α-
phenyl-methylphosphonate (1) (Scheme 2).
In principle, the Kabachnik–Fields reaction
under discussion may take place either via
imine (Schiff base) 2 or α-hydroxyphosphonate 3
(Scheme 3).
It should be noted that the reaction of cyclo-
hexylamine, benzaldehyde, and dibutylphosphine
oxide, which can be regarded to be as an extended
Kabachnik–Fields condensation, was shown to pro-
ceed according to the “imine” mechanism [12,17]. It
seems to be probable that the actual mechanism is
At first, the condensation with dimethyl phos-
phite was studied. The possible components of the
reaction, Schiff base 2, α-hydroxyphosphonate 3a,
O
O
H
O
H
H2N
+
+
(RO)2P
PhC
(RO)2P CH NH
H
− H2O
H
Ph
R = Me (a), Et (b)
1
SCHEME 2
Heteroatom Chemistry DOI 10.1002/hc

Study on the Kabachnik–Fields Reaction of Benzaldehyde, Cyclohexylamine, and Dialkyl Phosphites 173
(MeO)2P(O)CH(Ph)NHC6H11
(MeO)2P(O)CH(Ph)OH
PhCH=NC6H11
(MeO)2P(O)H
C6H5CHO
CH3CN
Wavenumber (cm −1
)
FIGURE 1 IR spectra of the reaction components measured in acetonitrile solution.
and α-aminophosphonate 1a, were prepared in sepa-
rate experiments (see the Experimental section), and
the infrared (IR) spectra of benzaldehyde, dimethyl
phosphite, imine 2, hydroxyphosphonate 3a, and
aminophosphonate 1a were recorded in acetonitrile
solution. The spectra are shown in Fig. 1; the char-
acteristic absorptions are summarized in Table 1.
Because of its lack of IR absorption in the region
of 800–1700 cm−1, the spectrum of cyclohexylamine
was not included.
and aminophosphonate 1a can be recognized from
the series of signals at 1038, 1060, and 1251 or 1031,
1056, and 1247 cm−1. The absorptions at around
1035 and 1058 cm−1 are due to the νP
C stretching
O
vibration, whereas those at around 1249 cm−1 are
the consequence of the νP vibration.
O
The Kabachnik–Fields reaction monitored by in
situ Fourier Transform (FT) IR spectroscopy was
carried out by heating an acetonitrile solution of
dimethyl phosphite and cyclohexylamine to 80◦C,
and benzaldehyde was then added. Collection of
the spectra was started when all the three com-
ponents were in the flask. The benzaldehyde was
converted instantly to Schiff base 2, which was
transformed to aminophosphonate 1a by a reaction
with dimethyl phosphite. The intermediate (2) ap-
peared at 1644 cm−1 and could be seen for ca. 2 h
with the maximum concentration at 10 min. With
the disappearance of the imine (2), the aminophos-
phonate (1a) appeared, as it was suggested by the
The reagents benzaldehyde and dimethyl phos-
phite revealed intense absorptions at 1702 or 980,
1042 and 1266 cm−1, respectively. The peaks at
1702, 980/1042, and 1266 cm−1 are due to νC
,
O
νP
C, and νP
stretching vibrations, respectively.
O
O
The imine (2) showed intense signals at 969 and
1644 cm−1. As the first one overlaps with the peak
of (MeO)2P(O)H at 980 cm−1, the absorption at
1644 cm−1 that is due to the νC N vibration may help
in the identification. The hydroxyphosphonate 3a
TABLE 1 Characteristic IR Absorptions of the Reaction Components Measured in Acetonitrile Solutiona
CH3CN
C6H5CHO
1702
(MeO)2P(O)H
PhCH NC6H11
1644
(MeO)2P(O)CH(Ph)OH
(MeO)2P(O)CH(Ph)NHC6H11
1440
1379
1204
1266
1042
980
1251
1060
1038
1247
1056
1031
919
969
aThe molar ratio of the corresponding component and acetonitrile is 1:10.
Heteroatom Chemistry DOI 10.1002/hc

174 Keglevich et al.
(MeO)2P(O)CH(Ph)NHC6H11
(MeO)2P(O)H
PhCH=NC6H11
Wavenumber (cm−1)
FIGURE 2 A segment of the time-dependent IR spectrum for the Kabachnik–Fields reaction of dimethyl phosphite, cyclohexyl-
amine, and benzaldehyde at 80◦C in acetonitrile.
(MeO)2P(O)CH(Ph)NHC6H11
(MeO)2P(O)H
PhCH=NC6H11
Time (h)
FIGURE 3 Concentration profile for the Kabachnik–Fields reaction studied at 80◦C in acetonitrile.
signals at 1031 and 1056 cm−1. The three-
dimensional (3D) diagram for the Kabachnik–Fields
reaction under discussion can be seen in Fig. 2. The
reaction mixture was analyzed by means of 31P and
13C NMR, as well as mass spectrometry of the oil
obtained after evaporation and flash column chro-
matography. The major component (87%) of the oil
was α-aminophosphonate (1a) as expected, but α-
hydroxyphosphonate (3a) was also present as a mi-
nor (13%) product. For the details, see the Experi-
mental section.
hydroxyphosphonate (3a) is also formed to a smaller
extent, and that this species (3a) does not undergo
further reaction with the cyclohexylamine present in
the mixture under the conditions of the reaction (at
80◦C). Hence, the formation of hydroxyphosphonate
3a is a “dead-end” route from the point of view of
formation of the α-aminophosphonate 1a.
The relative concentration–time diagram ob-
tained after deconvolution is shown in Fig. 3. The
intermediacy of imine 2 can be well seen, and the
reaction time is approximately 3.5 h.
It was also possible to reproduce the IR spec-
tra of the reaction components such as dimethyl
phosphite, Schiff base 2, and aminophosphonate 1a
(Fig. 4). It can be seen that the real IR spectra are
The above experiment serves as direct evidence
for the pathway via Schiff base 2 (“route A” in
Scheme 3) in the case of the three-component con-
densation under discussion. It is also seen that the α-
Heteroatom Chemistry DOI 10.1002/hc

Study on the Kabachnik–Fields Reaction of Benzaldehyde, Cyclohexylamine, and Dialkyl Phosphites 175
(MeO)2P(O)CH(Ph)NHC6H11
PhCH=NC6H11
(MeO)2P(O)H
Wavenumber (cm–1)
FIGURE 4 IR spectra for the reaction components obtained from the 3D diagram after deconvolution.
almost identical to those obtained by deconvolution
(Fig. 1 vs. Fig. 4).
The Kabachnik–Fields reaction of benzaldehyde,
diethyl phosphite, and cyclohexylamine was also
carried out in acetonitrile solution at 80◦C. After
a 4-h reaction time, 31P NMR analysis of the con-
centrated reaction mixture suggested the presence
of 85% α-aminophosphonate 1b and 15% of α-
hydroxyphosphonate 3b.
Reactants (A)
Imine intermediate (B)
Theoretical Calculations of the Solventless
Kabachnik–Fields Reaction
The reactant, intermediate, and product state mod-
els were studied by molecular modeling methods and
density functional calculations. For the details, see
the Experimental section. Both possible routes, the
one through an imine and the other through an α-
hydroxyphosphonate intermediate, were considered
under one-pot, solvent-free, and catalyst-free con-
ditions. The solvent- and catalyst-free accomplish-
ments were found to be especially efficient under
microwave irradiation [9,20–22]. Stereostructures of
the assemblies in the different states and relative en-
ergies are presented in Figs. 5 and 6 and Table 2,
respectively.
Products (D)
α-Hydroxyphosphonate
intermediate (C)
FIGURE
5 Stereostructure of the reactant (benzalde-
hyde, cyclohexylamine, and dimethyl phosphite (A)), imine
(B), hydroxyphosphonate (C), and product (D) assemblies
calculated.
In the reactant state (A), the dimethyl phosphite
and the amine molecules are bound by an H bond,
with an H bond distance of 3.11 A (measured be-
TABLE 2 Relative Energies for the Four States Calculated
Relative Energy
˚
(kJ mol−1
)
tween donor and acceptor), whereas the aldehyde
carbonyl oxygen is positioned halfway between the
methoxy groups of the phosphite, with an average
Reactant (benzaldehyde, cyclohexylamine,
and dimethyl phosphite (A))
Imine intermediate (B)
Hydroxyphosphonate intermediate (C)
Product (D)
0.0
−18.6
−40.5
−42.9
˚
O· · ·Cmethyl distance of 3.49 A. It should be noted
that the P O segment proved to be a more powerful
H bond attractant than the C O moiety of aldehyde.
Heteroatom Chemistry DOI 10.1002/hc

176 Keglevich et al.
Reactants
(A)
0
−5
−10
−15
−20
−25
−30
−35
−40
−45
Imine
intermediate
(B)
Hydroxy-
phosphonate
intermediate
(C)
Product
(D)
Product
(D)
FIGURE 6 Relative energies for the reactant (benzaldehyde, cyclohexylamine, and dimethyl phosphite (A)), imine (B), hy-
droxyphosphonate (C), and product (D) states calculated.
The “imine intermediate” and the phosphite are
held together by two H bonds with the participation
of the water molecule released in the condensation of
the amine and the aldehyde. In this associate (B), the
substitution of α-hydroxyphosphonates, which are
quite hindered model compounds, by amines is stud-
ied separately, and our results will be published in
due course. On the other hand, the reaction proceeds
quite easily through the imine intermediate. The re-
sults of the calculations are in full agreement with
our preparative experiments and with our observa-
tions by in situ FT IR measurements. It can also be
seen that the Kabachnik–Fields reaction under dis-
cussion proceeds similarly under solventless condi-
tions and in acetonitrile solution. The energetic or-
dering of the conformations was found highly depen-
dent on the number of ideally positioned H bonds of
the arrangements, which was, therefore, concluded
to be the driving force of the assembly of the studied
states.
In conclusion, our study combining monitor-
ing by in situ FT IR spectroscopy and preparation
suggests that the Kabachnik–Fields reaction of cy-
clohexylamine, benzaldehyde, and dialkyl phosphite
takes place via the imine intermediate at 80◦C in
acetonitrile used as the solvent; however, some α-
hydroxyphosphonate is also formed that does not
take part in a subsequent substitution under the
conditions of the reaction applied. The theoretical
results obtained by calculations indicate that the
situation may be similar under solvent-free condi-
tions. This means that in our case, the course sug-
gested by Cherkasov and coworkers [12,15] does not
work and the α-hydroxyphosphonate may not be
a real intermediate. At the same time, the conclu-
sion of Matveeva and Zefirov that the Kabachnik–
Fields condensation under discussion takes place
via the imine route [16] is confirmed. However, the
course of the Kabachnik–Fields reactions may, in
˚
N· · ·O and the O· · ·O distances are 2.93 and 2.84 A,
respectively.
In the structure of the complex formed between
the α-hydroxyphosphonate intermediate and the un-
reacted amine (C), the amine participates in two H
bond interactions, one with the double-bonded oxy-
gen of the P O group and one with the α-hydroxyl
substituent of the hydroxyphosphonate.
In the product state (D), the water molecule
formed as the side product is fixed to the α-
aminophosphonate by two H bonds. The first one
involves an interaction between the oxygen of the
P O group and one of the hydrogen atoms of wa-
ter, whereas the other one is based between the
oxygen atom of water and the proton of the amino
group. The former is a considerably stronger associ-
˚
ation with a donor–acceptor O O distance of 2.73 A,
whereas the length of the latter H bond measured at
˚
the O N atoms is 3.19 A.
The relative stability of the aforementioned four
states (A–D) can be regarded as a good indicator to
show how the reaction may take place (Fig. 6 and
Table 2). It can be seen that the formation of the
α-hydroxyphosphonate associate C (or intermediate
(3a)) is not unfavorable at all. However, it represents
a “dead-end” route, as species 3a does not undergo
substitution reaction with cyclohexylamine under
the conditions of the reaction applied to afford the
corresponding α-aminophosphonate (1a). This may
be due to the marginal energy gain (2.4 kJ mol−1)
of the final substitution step. The possibility of the
Heteroatom Chemistry DOI 10.1002/hc

Study on the Kabachnik–Fields Reaction of Benzaldehyde, Cyclohexylamine, and Dialkyl Phosphites 177
general, depend on the nature of the starting mate-
rials [12,13,15,16] and the conditions applied.
(KBr) 700, 776, 1027, 1054, 1153, 1238, 1452, 2957,
3266 cm−1.
Synthesis of Dimethyl α-Cyclohexylamino-
benzylphosphonate (1a)
EXPERIMENTAL
Equipment
A mixture of 0.16 mL (1.7 mmol) of dimethyl phos-
phite, 0.17 mL (1.7 mmol) of benzaldehyde, and
0.2 mL (1.7 mmol) of cyclohexylamine was mea-
sured in a sealed tube and irradiated (20–30 W) in
a CEM microwave reactor equipped with a pres-
sure controller at 100◦C for 30 min. Chromatogra-
phy (silica gel as the absorbent and 3% methanol
in dichloromethane as the eluant) of the crude mix-
ture furnished 0.22 g (45%) of aminophosphonate
1a. 31P NMR (CDCl3) δ 26.5; 13C NMR (CDCl3) δ 24.2
and 24.7 (C3), 25.9 (C4), 31.7 and 34.2 (C2), 53.1
(J = 6.5) and 53.3 (J = 2.2) (OCH3), 53.8 (J =
7.1, C1), 57.1 (J = 154.3, CHN), 127.6 (d, J = 3.1,
The 31P and 13C NMR spectra were taken on
a Bruker DRX-500 spectrometer (Bruker BioSpin
GmbH, Karlsruhe, Germany) operating at 202.4 and
125.7 MHz, respectively. The couplings are given in
Hertz.
The synthesis of aminophosphonate was car-
ried out in a CEM Discover microwave reactor
(CEM Microwave Technology Ltd., Buckingham,
UK) equipped with a pressure controller using 20–
30 W irradiation.
In situ FT IR measurements were conducted
using a ReactIR 1000 equipment (Mettler-Toledo
Inc., Columbus, OH). An attenuated total reflectance
(ATR) diamond measuring head was placed in a 100-
mL four-necked flask equipped with a dropping fun-
nel, a condenser, a thermometer, and a magnetic
stirrer. The temperature was maintained by using
an appropriately adjusted oil bath.
∗
∗
ꢁ
2
ꢁ
C ), 128.2 (J = 13.3, C ), 128.3 (J = 9.5, C ), 136.2
ꢁ
4
3
ꢁ
(J = 2.7, C1 ) *may be reversed. (The numbers 1–4
refer to the carbon atoms of the cyclohexyl group,
whereas numbers 1ꢁ–4ꢁ refer to the carbon atoms of
the phenyl ring.) δC [24] (CDCl3) 24.8, 25.3, 26.4, 32.2,
34.7, 53.6, 53.8, 54.4, 57.5 (J = 153.6), 128.3, 128.7,
128.9, 136.6; IR (neat) 700, 830, 1032, 1185, 1245,
1452, 2853, 2928 cm−1.
Synthesis of Imine 2
Monitoring the Kabachnik–Fields Reaction
Carried out at 80◦C
A mixture of 3.3 mL (32.5 mmol) of benzaldehyde
and 3.7 mL (32.5 mmol) of cyclohexylamine in 20 mL
of acetonitrile was stirred at 26◦C for 3 h. The sol-
vent and volatile components were then removed in
vacuum to afford 5.9 g (97%) of Schiff base 1. IR
(neat) 694, 753, 965, 1449, 1644, 2856, 2916 cm−1.
13C NMR (CDCl3) δ 24.6 (C3), 25.6 (C4), 34.3 (C2),
A solution of 5.5 mL (60.0 mmol) of dimethyl phos-
phite and 7.6 mL (66.0 mmol) cyclohexylamine in
30 mL of acetonitrile was heated to 80◦C, and 6.1 mL
(60.0 mmol) of benzaldehyde was added dropwise in
5 min. The mixture was stirred for 4 h. The resulting
3D diagram is shown in Fig. 2.
After evaporating the acetonitrile, according to
31P NMR, the crude mixture consisted of 87% of
aminophosphonate 1a (31P NMR (CDCl3) δ 26.5) and
13% of hydroxyphosphonate 3a (31P NMR (CDCl3)
δ 24.1). Chromatography of the crude product fur-
nished 13.1 g (76%) of aminophosphonate 1a.
An analogous reaction, but applying 7.7 mL
(60.0 mmol) of diethyl phosphite instead of dimethyl
phosphite, was also carried out. A similar workup
afforded a mixture consisting of 85% of aminophos-
phonate 1b (31P NMR (CDCl3) δ 24.2, δP [16] 24.3)
and 15% hydroxyphosphonate 3b (31P NMR (CDCl3)
δ 21.5, δP [25] 21.5).
∗
∗
ꢁ
2
ꢁ
69.7 (C ), 127.9 (C ), 128.3 (C ), 130.1 (C ), 136.6
ꢁ
1
4
3
ꢁ
(C1 ), 158.1 (C N) *may be reversed. (The numbers
1–4 refer to the cyclohexyl group, whereas numbers
1ꢁ–4ꢁ refer to the phenyl ring.)
Synthesis of Dimethyl α-Hydroxy-
benzylphosphonate 3a
5.7 mL (56.6 mmol) of benzaldehyde was added to
a suspension of 21.0 g (0.21 mol) of Brockmann II
Al2O3 in 40 mL of CH2Cl2. Then the mixture was
stirred for 2 h. The dichloromethane was evaporated,
and the mixture was kept at 26◦C for 1 week. Af-
ter completion, the mixture was extracted with 3 ×
50 mL of CH2Cl2. The solvent of the combined ex-
tracts was evaporated, and the solid residue so ob-
tained was washed with 2 × 5 mL of diethyl ether
to give 9.8 g (80%) of α-hydroxyphosphonate 3a as
white crystals in a pure form. Mp 101–102◦C, mp
[23] 102◦C; 31P NMR (CDCl3) δ 24.3, δP [23] 24.3; IR
Molecular Modeling
Molecular modeling and subsequent Density Func-
tional Theory (DFT) calculations were carried
out. Monte Carlo multiple minimum (MCMM)
Heteroatom Chemistry DOI 10.1002/hc

178 Keglevich et al.
[3] Sienczyk, M.; Oleksyszyn, J. Curr Med Chem 2009,
16, 1673.
[4] Liu, W.; Rogers, C. J.; Fisher, A. J.; Toney, M. D. Bio-
chemistry 2002, 41, 12320.
[5] Grembecka, J.; Mucha, A.; Cierpicki, T.; Kafarski, P.
J Med Chem 2003, 46, 2641.
[6] Long, N.; Cai, X. J.; Song, B. A.; Yang, S.; Chen, Z.;
Bhadury, P. S.; Hu, D.Y.; Jin, L. H.; Xue, W. J Agric
Food Chem 2008, 56, 5242.
[7] Hu, D. Y.; Wan, Q. Q.; Yang, S.; Song, B. A.; Bhadury,
P. S.; Jin, L. H.; Yan, K.; Liu, F.; Chen, Z.; Xue, W. J
Agric Food Chem 2008, 56, 998.
[8] Danila, D. C.; Wang, X. Y.; Hubble, H.; Antipin, I. S.;
Pinkhassik, E. Bioorg Med Chem Lett 2008, 18, 2320.
[9] Keglevich, G.; Szekre´nyi, A. Lett Org Chem 2008, 5,
616.
search [26]—as implemented in MacroModel [27]—
involved the random variation (within the range of
0–180◦) of a randomly selected subset of all tor-
sional angles (a minimum of 2◦ and a maximum of
12◦), combined with the variable molecules selection
(MOLS) method for translations and rotations of the
complexing partners with respect to each other. The
combined MCMM/MOLS procedure allowed ran-
◦
˚
dom translation (0–5 A) and rotation (0–180 ) during
a Monte Carlo step. Calculations consisted of 50,000
steps. The perturbed structures were minimized us-
ing a truncated Newton algorithm [28]. Calcula-
tions were carried out using the optimized potential
for liquid simulations 2005 force field [29]. Atomic
charges were calculated by using the electrostatic
potential method [30] from B3LYP/6-31G∗∗ wave
functions of the geometry-optimized noninteracting
partners of each discussed state, using Jaguar [31].
The applied MCMM procedure results in a col-
lection of conformers that span the energy range
specified in the calculation (50 kJ mol−1, in our case).
The low-energy conformers of this set describe an
ensemble that can be expected to shape the macro-
scopic descriptors of the given state. Therefore, in
each case, conclusions were drawn based on trends
that were general within the low-energy conforma-
tions of the given state, which were considered as
such if their conformational energy did not exceed
the global minimum (lowest) energy by more than
8.2 kJ mol−1 (approximately 2 kcal mol−1).
[10] Kabachnik, M. I.; Medved, T. Ya. Dokl Akad Nauk
SSSR 1952, 83, 689; 84, 717.
[11] Fields, E. K. J Am Chem Soc 1952, 74, 1528.
[12] Cherkasov, R. A.; Galkin, V. I. Russ Chem Rev 1998,
67, 857 (Usp Khim 1998, 67, 940; in Russian).
[13] Galkin, V. I.; Zvereva, E. R.; Sobanov, A. A.; Galkina,
I. V.; Cherkasov, R. A. Zh Obshch Khim 1993, 63,
2225.
[14] Keglevich, G.; Fehe´rva´ri, A.; Csontos, I. Heteroat
Chem 2011, 22, 599.
[15] Galkina, I. V.; Sobanov, A. A.; Galkin, V. I.;
Cherkasov, R. A. Russ J Gen Chem. 1998, 68, 1398
(Zh Obshch Khim 1998, 68, 1465; in Russian).
[16] Matveeva, E. D.; Zefirov, N. S. Dokl Chem 2008,
420, 137 (Dokl Akad Nauk 2008, 420, 492; in
Russian).
[17] Galkina, I. V.; Galkin, V. I.; Cherkasov, R. A. Zh
Obshch Khim 1998, 68, 1469.
[18] Gancarz, R.; Gancarz, I. Tetrahedron Lett 1993, 34,
145.
[19] Gancarz, R. Tetrahedron 1995, 51, 10627.
[20] Kabachnik, M. M.; Zobnina, E. V.; Beletskaya, I. P.
Synlett 2005, 1393.
[21] Xu, Y. S.; Yan, K.; Song, B.; Xu, G. F.; Yang, S.; Xue,
W.; Hu, D. Y.; Lu, P. P.; Ouyang, G. P.; Jin, L. H.;
Chen, Z. Molecules 2006, 11, 666.
[22] Xia, M.; Lu, Y. D. Ultrason Sonochem 2007, 14,
235.
[23] Rowe, B. J.; Spiling, C. D. Tetrahedron: Asymmetry
2001, 12, 1701.
Final conformation and energy ordering of the
different states were obtained by DFT gas-phase op-
timization (B3LYP/6-31G**) of characteristic, low-
energy conformers of the force-field calculations
(typically 3–6 low-energy conformers), followed by
further B3LYP/6-311G∗∗++ optimization of the two
best conformers.
[24] Mu, X-J.; Lei, M-Y.; Zou, J-P.; Zhang, W. Tetrahedron
Lett 2006, 47, 1125.
ACKNOWLEDGMENTS
[25] Keglevich, G.; To´th, V. R.; Drahos, L. Heteroat Chem
2011, 22, 15.
[26] Chang, G.; Guida, W. C.; Still, W. C. J Am Chem Soc
1989, 111, 4379.
[27] Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.;
Hendrikson, T.; Still, W. C. J Comput Chem 1990, 11,
440.
[28] Ponder, J. W.; Richards, F. M. J Comput Chem 1987,
8, 1016.
This work is connected to the scientific program
of the “Development of quality-oriented and har-
monized R + D + I strategy and functional model
at BME” project. This project is supported by the
´
New Hungary Sze´chenyi Plan (Project ID: TAMOP-
4.2.1/B-09/1/KMR-2010-0002). DKM acknowledges
the support of the Bolyai Foundation of the Hun-
garian Academy of Sciences.
[29] Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.;
Jorgensen, W. J. J Phys Chem B 2001, 105, 6474.
[30] Woods, R. J.; Khalil, M.; Pell, W.; Moffat, S.
H.; Smith, V. H., Jr. J Comput Chem 1990, 11,
297.
REFERENCES
[1] Zefirov, N. S.; Matveeva, E. D. Arkivoc 2008, 1, (i).
[2] Kafarski, P.; Lejczak, B. Curr Med Chem Anti-Cancer
Agents 2001, 1, 301.
[31] Jaguar, version 4.1; Schro¨dinger, Inc.: Portland, OR.
Heteroatom Chemistry DOI 10.1002/hc