Theoretical and structural studies on mechanism of the Stec reaction

Article abstract of DOI:10.1016/j.tet.2012.04.081

The mechanism of the Stec reaction between phosphoroselenanilidate or phosphonoanilidate and CS₂, activated by strong bases, has been studied computationally, using DFT methods, and experimentally, by low temperature ³¹P NMR spectroscopy. From molecular calculations, the reaction pathway of the reaction has been revealed with several transition states and intermediates, including a low energy spirocyclic pentacoordinate transition state and acyclic tetracoordinate intermediates, which eventually were correlated with short living molecules detected by NMR spectroscopy.

Full text of DOI:10.1016/j.tet.2012.04.081

Tetrahedron 68 (2012) 5554e5563
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Theoretical and structural studies on mechanism of the Stec reaction
,d,
Maria del Carmen Michelini b, Nino Russo b, Stefano Alcaro c, Lucyna A. Wozniak a
*
a Medical University of Lodz, Department of Structural Biology, 90-752-Lodz, 7/9 Zeligowskiego Str., Poland
b Department of Chemistry, University of Calabria, Arcavacata di Rende, Italy
c
ꢀ
ꢀ
ꢁ
Dipartimento di Scienze Farmacobiologiche Universita di Catanzaro “Magna Grccia”, Complesso Ninõ Barbieri, 88021 Roccelletta di Borgia (CZ), Italy
d Centre of Molecular and Macromolecular Studies, Department of Bioorganic Chemistry, Polish Academy of Sciences, Lodz, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
The mechanism of the Stec reaction between phosphoroselenanilidate or phosphonoanilidate and CS2,
activated by strong bases, has been studied computationally, using DFT methods, and experimentally, by
low temperature 31P NMR spectroscopy. From molecular calculations, the reaction pathway of the re-
action has been revealed with several transition states and intermediates, including a low energy spi-
rocyclic pentacoordinate transition state and acyclic tetracoordinate intermediates, which eventually
were correlated with short living molecules detected by NMR spectroscopy.
Received 18 August 2011
Received in revised form 3 April 2012
Accepted 23 April 2012
Available online 30 April 2012
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
rotation TR) within the pentacoordinate intermediate(s) would
result in retention of configuration at the phosphorus atom. The
The Stec reaction, namely the PN/PX (X¼O, S, Se) conversion,
has been established as a convenient method for the synthesis of
numerous P-chiral phosphates, and their thio- or seleno-congeners,
which otherwise could not be simply prepared. Since the reaction is
stereospecific, the P-chiral phosphoramidates are valuable pre-
cursors for the synthesis of a variety of P-chiral derivatives of
phosphorus acids, including various biophosphates and their ana-
logues.1 More recently, the PN/PX reaction was applied in a ster-
eoconvergent synthesis of stereodefined methylphosphonate
analogues of nucleic acids with a predetermined sense of chirality
at the internucleotide methylphosphonates.2
synchronous cleavage of the PeN and CeY bonds also lead to the
corresponding phosphates (or their analogues) with the retained
configuration at phosphorus, and the nitrogen cumulene-like RN]
C]X as a second product.
It should be underlined that although the mechanism presented
in Scheme 1 supports the stereoretentive mode of the PN/PX
conversion, the experimental evidence for the putative participa-
tion of intermediates of either structure III or VI was lacking.
For clarifying the mechanism we merged Density Functional
Theory (DFT) studies of the potential energy profiles, supported
with the structural parameters obtained from the X-ray data and
the low temperature NMR experiments, focused on detection and
identification of any ‘short-time living’ species.
As model compounds, the selenium analogue of pyrocatechol,
i.e., 2-N-phenylamine-2-seleno-1,3,2-benzodioxaphospholane (1)
(Scheme 2) was chosen since there are existing stable PV phos-
phoranes with the 1,2-benzenediol group located in the api-
caleequatorial position, which hinders molecules in predictable
conformations.7
The exhaustive studies3 confirmed a stereoretentive pathway4
for this reaction for acyclic and cyclic (five- or six-membered
ring) compounds alike. It was postulated that the reaction occurred
by the attack of the phosphylamidate anion II at the carbonyl atom
of chalcogenide (Scheme 1),5 with formation of the adduct III. It
was assumed that the adduct III could intramolecularly rearrange,
forming putative pentacoordinate intermediate IV. Such in-
termediate should undergo a polytopal rearrangement into another
pentacoordinate intermediate V. The Westheimer’s rules of apical
entryeapical departure, required placing the nitrogen atom into an
apical position, necessary for the PeN bond cleavage.6 A partici-
pation of hypervalent intermediates of square/rectangular struc-
ture VI stayed questionable. The odd number of polytopal
rearrangements (either Berry pseudorotation Y, or turnstile
2. Results and discussion
2.1. Synthesis and reactivity
2-Aniline-2-seleno-1,3,2-benzodioxaphospholane (1) was pre-
pared from 2-chloro-1,3,2-benzodioxaphospholane (2), treated
with an excess of aniline (2 equiv) at room temperature in toluene
in the presence of elemental selenium. The conversion to the initial
* Corresponding author. Tel.: þ48 42 6393221; e-mail addresses: lucyna.wozniak@
umed.lodz.pl, lawozniak@gmail.com (L.A. Wozniak).
PIII derivative, i.e., 2-aniline-1,3,2-benzodioxaphospholane (3) (31
P
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.04.081

M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
5555
Scheme 1. Plausible mechanism of the PN/PS reaction between 2-N-phenylamine-2-seleno-1,3,2-benzodioxaphospholane 1 (activated with strong bases) and carbon disulfide.1,3
O
O
O
-
P
Se
H
N
H
N
O
O
O
O
O
O
5
Aniline
15 min.
column
Se
Ph
65.3 ppm J = 915 Hz
P
Cl
P-Se
P
Ph
P
chromat
Se
+
Ph
1
2
3
O
O
O
O
77.9 ppm, J = 1018 Hz
N
137.19 ppm
P-Se
173.1 ppm
P
P
Se
Se
6
75.3 ppm J = 1005 Hz
P-Se
1
Scheme 2. Synthesis of 2-aniline-2-seleno-1,3,2-benzodioxaphospholane (1) and 31P NMR chemical shifts and coupling constants JPeSe.
NMR
d
137.22 ppm) was complete within 15 min at room temper-
1JPeSe¼1061 Hz). Reaction of 5 with MeI led to 2-methoxy-2-seleno-
ature. The addition of selenium required phosphoramidite 3 to be
heated in a sealed ampoule for several days in toluene, producing 1
in good yields (over 90% as estimated from 31P NMR spectroscopy).
In contrast, it is worth emphasizing, that addition of elemental
sulfur to 3 providing the corresponding 2-aniline-2-thio-1,3,2-
benzodioxaphospholane (4) was formed at room temperature
within 4 h in almost quantitative yield, and 4 was easily purified by
a silica gel column chromatography.
1,3,2-benzodioxaphospholane (9)
1JPeSe¼1041 Hz).
(
31P NMR
d
65.9 ppm,
The progress of the PN/PS reaction was followed by 31P NMR
spectroscopy. It was found that the activation of 1 with strong bases
resulted in stable anionic forms 11, in the presence of NaH (
d
1
1
56.1 ppm, JPeSe¼885 Hz ), and DBU (
d
57.3 ppm, JPeSe¼910 Hz),
respectively. The reaction of the anionic 2-aniline-2-seleno-1,3,2-
benzodioxaphospholane (11) with an excess of CS2 performed ei-
ther in MeCN, THF, or DMF resulted in formation of the expected 2-
thio-2-seleno-1,3,2-benzodioxaphospholane (12) in almost quan-
titative yields (Scheme 4).
Our efforts to separate the phosphoroselenoanilidate 1 by col-
umn chromatography under analogous conditions failed. Instead,
the 2-oxo-2-seleno-1,3,2-benzodioxaphospholane (5) (31P NMR
1
d
65.2 ppm, JPSe¼915 Hz) was collected from the column as the
major product accompanied by traces of N-phenyl-N,N-bis(2-
2.2. Theoretical studies
seleno-1,3,2-benzodioxaphospholane) (6) (31P NMR
d 73.2 ppm,
1JPSe¼1005 Hz) with no trace of 1 collected after chromatography
(Scheme 3).
The Stec reaction, being
a
variant of the Hornere
WadswortheEmmons (HWE) reaction of dialkyl (aryl) phosphor-
amidates (usually anilidates), activated by sodium hydride, with car-
bonyl electrophiles or carbon disulfide to provide dialkyl phosphates
and the corresponding isocyanates or isothiocyanates, was extended
to phosphoramidothioates (2, X¼S) and phosphoramidoselenoates (2,
X¼Se) with the purpose obtaining several phosphorus acid de-
rivatives, particularly P-chiral compounds in stereospecific man-
ner.1e4 Therefore, the question of the mechanism of stereospecific
The structure of 5 was confirmed by the X-ray analysis.8 Addi-
tionally, it was established that 2-oxo-2-seleno-1,3,2-benzodiox-
aphospholane (5) was easily oxidized under mild anhydrous condi-
tions with t-BOOH, yielding in 1,3,2-benzodioxaphospholane (7) (31P
NMR
trimethylsilyl)acetamide provided exclusively 2-trimethylsilyloxy-2-
seleno-1,3,2-benzodioxaphospholane (8) (31P NMR
63.1 ppm,
d
ꢀ1.8 ppm). Compound
5 after treatment with bis(-
d

5556
M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
H
N
O
O
O
O
O
O
O
-
tBuOOH
silica gel column
P
-
P
1
chromatography
P
7
Se
O
Se
O
5
65.3 ppm J = 915 Hz
77.9 ppm, J = 1018 Hz
P-Se
-1.8 ppm
P-Se
BSA
MeI
O
O
OMe
O
O
OSiMe
3
P
8
P
9
Se
Se
73.2 ppm J = 1005 Hz
63.1 ppm J = 1061 Hz
P-Se
P-Se
1
Scheme 3. Reactivity of 1 (31P NMR chemical shifts and coupling constants JPeSe).
_
H
N
O
O
O
S
_
N
O
O
CS
NaH
2
N=C=S
P
P
+
P
3
Se
Se
O
Se
11 (R1)
56.1 ppm J
P-Se
12 (P1)
120.9 ppm J = 875 Hz
(P2)
= 885 Hz
78.2 ppm J = 1022 Hz
P-Se
P-Se
Scheme 4. Synthesis of 2-thio-2-seleno-1,3,2-benzodioxaphospholane (12).
formation of these products has been of special interest. Here we re-
port a theoretical approach to the reactions of 2-N-phenylamine-2-
seleno-1,3,2-benzodioxaphospholane (3) and 2-N-phenylamine-2-
oxo-1,3,2-benzodioxaphospholane (7) with CS2, with special em-
phasis on the structures of PV-intermediary structures.
reliability of some of the transition states that show low imaginary
frequencies (less than 100 cmꢀ1), we have re-optimized those
structures using the SCF¼very tight option of Gaussian03. No
changes were found in the optimized geometries or the imaginary
frequencies of those structures.
Numerous computational investigations have been reported
using quantum mechanical calculations to elucidate the HWE and
related reactions, mostly focused on the stereochemistry of alkene
formation.9 From these studies it was clear that these reactions can
be rationalized by including solvation effects, since previous cal-
culations on mechanism and stereoselectivity of ylide reactions
carried out as the gas-phase potential energy surface were not
evocative.10 Brandt et al. investigated the HWE reaction using high-
level quantum chemical (QC) methods. They found that the model
reaction of trimethyl phosphonoacetate and formaldehyde was
very sensitive to solvent effect and the rate limiting step was con-
nected to a formation of the oxyanion corresponding to structure III
(Scheme 1).11
For both studied reactions, we report the gas-phase Potential
Energy Profiles (PEPs) calculated as relative energies (including
ZPVE) of the species involved in the reaction pathways with respect
to the ground-state reactant asymptotes at 0 K.
Single point calculations were performed on the gas-phase
optimized structures in order to include solvent effects. Those ef-
fects were computed within the framework of the Self-Consistent
Reaction Field Polarized Continuum Model (SCRF-PCM)21e23 as
implemented in Gaussian03. The default UAHF33 set of solvation
radii was used in the calculations to build the cavity according to
the molecular topology, hybridization, and formal charge. In this
method, the solvent is modeled as a continuum of dielectric con-
stant . In order to determine the solvent effects on the reaction
3
The initial step was to model the gas-phase reactions. With this
aim, theoretical calculations of all the reactants, intermediates,
transition states, and products were performed within the frame-
work of DFT. In particular, the Perdew and Wang density gradient
correction to the exchange and correlation functionals (denoted as
PW91PW91) was used,12e14 together with the aug-cc-pVTZ-PP
basis set for selenium,15 and 6-311þþG(2d,2p) extended basis
set16e18 for the rest of the atoms. All the calculations were carried
out with Gaussian03 code.19 We refer to this level of calculation as
PW91/G03, hereafter.
energetic, each of the species along the reaction pathways is placed
in a cavity within the solvent. The polarized continuum model
defines the cavity as the union of a series of interacting spheres
centered on the atoms, and uses a numerical representation of the
polarization of the solvent. This type of model, however, does not
consider specific interactions between the solute and the solvent.
After inclusion of solvent effects on each of the species involved
in the reaction mechanism, the PEPs are reconstructed as relative
energies (including the gas-phase ZPVE correction) with respect to
the corresponding reactant asymptotes.
All geometrical structures were optimized without symmetry
restrictions. For each optimized system a vibrational analysis was
performed to determine its character (minima or saddle point) and
to evaluate the zero-point vibrational energy corrections (ZPVE),
which were included in all the reported energies. We have ensured
that every transition state has only one imaginary frequency, and
that the vibrational mode associated to that frequency corresponds
to the correct movement of the involved atoms. All the minima
connected by a given transition state were confirmed performing
IRC (Intrinsic Reaction Coordinate) computations, as implemented
in Gaussian03 program.20 The default SCF¼tight keyword was used
in all geometry optimizations. However, in order to check for the
We first analyzed the gas-phase PEPs of reactions PN(Se)/P(S)
Se and PN(O)/P(S)O (Fig. 1) The geometrical structures of the all
the involved stationary points are reported in Figs. 2 and 3,
respectively.
The reaction of the phosphoroselenoanilidate anion 11 (R1) and
CS2 starts with the surpassing of an activation barrier of about
11 kcal/mol (TS1). It is important to indicate that in the first transi-
tion state (TS1 i252 cmꢀ1, Fig. 2) the CS2 molecule is already acti-
vated, as evidenced by the S(1)eC(3)eS(2) angle of 142.8ꢁ. After the
system surpasses the first transition state, the formation of the first
insertion intermediate (Int1), takes place. In this structure the CeN
ꢂ
bond can be considered as already formed (1.478 A). Further

M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
5557
geometries, using the SCRF-PCM approach. We have considered two
different solvents, namely MeCN and THF. The computed potential
energy profiles for the PN(Se)/P(S)Se reaction including solvent
effects are reported in Fig. 4 and compared with gas-phase PEP. The
inclusion of the solvent effects induces important changes in the
shape of the PES. Indeed, the initial barrier increases by around
3 kcal/mol with respect to the gas-phase value. In contrast, the in-
sertion intermediate (Int1) and the second transition state (TS2)
show higher stabilities. The second insertion intermediate (Int2)
and the last transition state (TS3) are destabilized by 3 and 4 kcal/
mol, respectively. The exothermic effect of the whole process is in-
creased with respect to the gas-phase value by around 5 kcal/mol.
The gas-phase PEPs for the PN(O)/P(S)O reaction are changed
by the inclusion of solvent effects in a similar way. In particular,
there is an increase in the first barrier of around 5 kcal/mol,
whereas the first insertion intermediate (Int10) and the second
transition state (TS20) are more stabilized than in the gas-phase
reaction by 4 and 2 kcal/mol, respectively. The second insertion
intermediate (Int20) is slightly less stabilized, with respect to the
asymptotes than in the gas-phase reaction, being at 7 kcal/mol
under the reactant asymptote. The same effect is found for the last
transition state, TS30, by about 4 kcal/mol. From reactants to
products, the process is exothermic by around 16 kcal/mol (Fig. 5).
TS2
12
TS1
11
9
10
0
Int1
8
TS1’
8
TS2’
3
R1+ CS2
R1’+ CS2
Int1’
TS3
-2
TS3’
Int2
-9
P1’+P2
-11
P1 + P2
-10
-12
Int2’
Reaction Coordinate
Fig. 1. Gas-phase potential energy profiles for reaction PN(Se)/P(S)Se (full line) and
PN(O)/P(S)O (dashed line). The relative energies (RE, in kcal/mol) of the transition
states, intermediates and products are calculated with respect to the ground-state
reactant asymptotes (R1þCS2 and R10þCS2, respectively).
diminishing of the S(1)eC(3)eS(2) angle (127.1ꢁ) and a slight
lengthening of the CeS bonds are also evidenced in this structure
(Fig. 2). The formation of this intermediate is endothermic by 8 kcal/
mol. The system then overcomes the second transition state (TS2,
i126 cmꢀ1), which involves a reaction barrier of 4 kcal/mol. TS2 is
12 kcal/mol above the reactant asymptotes and is structurally very
close to TS1, with the major shortening of the PeS(1) distance, from
3.245 A in the first intermediate to 2.855 A in the TS2. The reaction
then proceeds to the formation of a second insertion intermediate
(Int2) with a four-member ring hypervalent pentacoordinate spi-
rocyclic structure. Int2 is more stable than Int1 by around 17 kcal/
mol and is 9 kcal/mol below the reactants asymptote. The last step of
the reaction takes place after the system surpasses the last transition
state (TS3, i78 cmꢀ1), which involves a reaction barrier of 7 kcal/mol.
This step is associated with the cleavage of NeP and C(3)eS(1)
bonds. Our calculations indicate that the overall process, from re-
actants to products, is exothermic by 11 kcal/mol.
2.3. 31P NMR studies
It was found that the activation of phosphoramidates can be
efficient not only in the case of sodium or potassium hydrides, but
also can be achieved in the reactions of strong organic bases.
Among them DBU was found to be superior because of solubility
and homogeneity of the reaction system. From previous experi-
ments it is known, however, that DBU activation results in slower
conversion, therefore low temperature reactions were performed
in DMF and THF, with NaH as activator.
ꢂ
ꢂ
It was found that at ꢀ60 ꢁC the activation in DMF was almost si-
multaneous, and two closely resonating compounds were formed
1
1
(
31P NMR
d
57.75 ppm, JPSe¼890 Hz; 58.66 ppm, JPSe¼906 Hz) in
a 2:1 ratio. The signal of the higher field species disappeared first (at
ꢀ40 ꢁC), whereas the second one disappeared slowly in the course of
the reaction and corresponded to formation of 2-thio-2-seleno-1,3,2-
A similar reaction mechanism can be traced for PN(O)/P(S)O.
The potential energy profile for the gas-phase reaction is sketched
in Fig. 1 (dashed line), with the optimized geometries of all the
involved species reported in Fig. 3.
benzodioxaphospholane (12) (31P NMR
d
120.66 ppm, 1JPSe¼876 Hz).
The initial 2:1 ratio of the preliminary intermediate species is in-
variable in different solvents like THF (at ꢀ40 ꢁC) or at MeCN (at 0 ꢁC),
and product 12 formation is always accompanied by the disappear-
The initial barrier height is only slightly lower than that for
PN(Se)/P(S)Se, namely 9 kcal/mol over the asymptote limit. The
first insertion intermediate (Int10) is more stable than the corre-
sponding minima for the previous reaction by around 5 kcal/mol.
The next step of the reaction involves the overcoming of the second
transition state (TS20), which involves an energetic barrier of
5 kcal/mol. TS20 is 8 kcal/mol above the reactant asymptotes, and is
characterized by an imaginary frequency (i124 cmꢀ1) that involves
the stretching of the PeS(2) bond (Fig. 3).12 As compared to the
previously described reaction, the geometrical structure of TS20 is
very close to that of (I0), albeit the major difference between the
structures is reflected in the PeS(2) distance. After the system
surpasses the second transition state, the formation of a second
insertion intermediate (Int20) takes place. The implied exothermic
effect is of about 15 kcal/mol. As in the previous reaction, Int20 is
more stable than the previous reaction intermediate, by around
9 kcal/mol. In contrast to the reaction PN(Se)/P(S)Se, the reaction
PN(O)/P(S)O products are slightly higher in energy with respect to
the last intermediate. The whole reaction was found to be exo-
thermic by around 11 kcal/mol.
ance of the corresponding intermediate at
d
57.75 ppm,1JPSe¼890 Hz.
At ꢀ25 ppm and ꢀ40 ppm, we observed traces of relatively long
living species (below 10%). Their chemical shifts imply penta-
coordinate phosphorus compounds.7b,24 However, based on the
chemical shifts it is unfeasible to assess their structure, since there
are limited data on pirocatechol phosphoranes,25 particularly those
with spiro structures involved.26 We have not detected any JPeSe
1
coupling constants, therefore the presence of selenium in high field
compounds cannot be confirmed. There are literature reports,
however, of several selenium containing PV compounds lacking
a 1JPeSe, although a reason for this phenomenon is not clear.27
31P NMR chemical shifts were calculated using the Gauge In-
variant Atomic Orbital Method (GIAO)28 with both Gaussian03 and
ADF2004.01 packages.29 In the case of Gaussian03 calculations, we
employed the mPW1PW9130 functional on the PW91PW91/cc-aug-
pVTZ-PP(Se), 6-311Gþþ(2d,2p) optimized geometries, using the
same basis sets (mPW1PW91/G03, hereafter).
Previous theoretical studies have also shown the difficulty in
suggesting a reliable choice of a functional, due to the significant
variance in performance of the different methods,31 with the
mPW1PW91 functional found to have a good performance in pre-
vious NMR chemical shift theoretical studies.31a We also performed
As previously mentioned, the solvent effects were evaluated by
performing single point calculations on the gas-phase optimized

5558
M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
Fig. 2. Most relevant geometrical parameters of the species involved in the PN(Se)/P(S)Se reaction. Bond distances are in angstroms and angles in degrees. Imaginary frequencies
are reported between parentheses.
some test calculations using the B3LYP functional,32 which has also
been reported to be efficient for the calculation of NMR proper-
ties.33 However, for the systems under study it was found that the
mPW1PW91 functional gave a better agreement with the observed
values. We therefore report the results obtained at the
mPW1PW91/cc-aug-pVTZ-PP(Se), 6-311Gþþ(2d,2p) level of the-
ory (mPW1PW91/G03).
The NMR spinespin coupling constants were calculated using
the ADF package. The zero-order regular approximation (ZORA)
was employed in all ADF calculations. For these calculations we
chose the PW91 functionals (exchange and correlation) together
with the TZ2P basis sets, as implemented in ADF code (PW91/ADF,
hereafter). All ADF calculations were performed after the structures
were re-optimized at the same level of theory used for NMR cal-
culations. The geometrical parameters obtained at the PW91/ADF
level of theory with the ADF program for substrate 1, and its anion
11, Int1, Int2 and the product P1 are presented at Fig. 6, together
with the corresponding values obtained at the PW91/G03 level. A
comparison between the structural parameters obtained at the
different levels of theory shows that there is a fairly good agree-
ꢂ
ment, in particular the bond distances differ by less than 0.02 A and
angles by less than 6ꢁ. The difference between calculated dihedral
angles was in some cases a little larger, namely up to 15ꢁ (substrate
1). The computed 31P chemical shifts are reported in Table 1 to-
gether with the measured values, whereas the different contribu-
tions to the spinespin coupling constants are given in Table 2.

M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
5559
Fig. 3. Most relevant geometrical parameters of the species involved in the PN(O)/P(S)O reaction. Bond distances are in angstroms and angles in degrees. Imaginary frequencies
are reported between parentheses.
A comparison between the 31P chemical shifts computed for
substrate 1 and the anionic structures R1 and 12 (P1); and the
corresponding measured values shows that even when the trend is
well reproduced, the differences between the measured chemical
shifts and the calculated ones are significant. As previously dem-
onstrated, theoretical modeling of phosphorus NMR chemical
shielding is very challenging.34 The chemical shifts are known to be
very sensitive to the molecular geometry.34b,35 Accordingly, one of
the reasons for the quantitative disagreement observed between
the calculated and experimental values could be the approach used
for the inclusion of solvent effects. Due to the excessive computa-
tional cost of performing geometry optimizations including solvent
effects, single point calculations were done on gas-phase optimized
geometries. A comparison between the computed values shows
that, in all cases the calculations overestimate the experimental
values, and that the inclusion of the solvent effects does not notably
improve the agreement. With the only exception of substrate 1, all
of the species studied here are negatively charged and both media
(MeCN and THF) are relatively polar, therefore solvent effects are
expected to be strong. Between the theoretical approaches used
here, the one that gives the better agreement with the experiments
is PW91/ADF, which could be due to the better performance of the
TZVP Slater-type basis sets. However, we have not performed
a complete comparison of different functional/basis sets as to
conclude that the better performance is mainly associated to the
functional or basis sets employed in each approach.

5560
M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
MeCN
coordinate intermediates with a four-membered ring have been
20
10
0
THF
Gas-phase
postulated and their intermolecular rearrangements according to the
Westheimer rules have been assumed to be responsible for the ster-
eoretentive mechanism of this reaction.1,3 However, so far, experi-
mental evidence of these hypervalent species has not been available.
In practice, the definite pentacoordinate structures could vary
between the TBP (like structures IV and V in Scheme 1) and the
square/rectangular pyramidal (SP/RP) (structure VI) geometry. In
general, the TBP geometry, with a different level of distortion, is
favored, and the SP/RP geometry is mostly limited to spirobicyclic
derivatives or ones containing strained three-/four-membered
rings.36 It is experiential that a less apicophilic thio or seleno
group in the TBP geometry is more apicophilic in the SP/RP
geometry.37
Intramolecular exchange processes in acyclic pentacoordinate
phosphoranes appear to occur mostly through a Berry pseudor-
otation.38 Even for a monocyclic and spirocyclic phosphorane, this
is a good option, although in principle a turnstile mechanism could
also be operating. Barriers of pseudorotation are influenced by
charge state, apicophilicity of ligands, intramolecular hydrogen
bonding, cyclic structure, and solvation.39
TS1
14
TS2
12
10
9
13
11
8
5
TS3
2
5
2
Int1
R1 + CS2
-2
-6
-6
-10
-20
-9
Int2
-11
P1 + P2
-16
-16
Reaction Coordinate
Fig. 4. Potential energy profile for PN(Se)/P(S)Se reaction including solvent effects for
MeCN (full line) and THF (dashed line). The gas-phase PEP is included as a reference
(dotted line).
In the case of the reaction of anion 11 with CS2, the initial ad-
dition resulted in transition state TS1 (the energy barrier was cal-
culated as 11 kcal/mol in the gas phase, and corrected to 14 kcal/
mol in MeCN). The first minimum in the potential energy surface
(Int1) corresponding to the intermediate of tetracoordinate struc-
ture III was identified with 8 kcal/mol in the gas phase dropping to
5 kcal/mol in MeCN. One can take into consideration the relatively
small differences in structure between II and III (R1 and IntI in
Fig. 3), and compare it with variable temperature NMR data, where
the signal of anion 11 is accompanied at low temperatures by
20
MeCN
THF
TS1’
14
Gas-phase
TS2’
13
10
8
9
7
TS3’
2
3
6
0
1
0
-10
-20
-1
R1’ + CS2
-2
Int1’
-7
-7
a
d
close but distinguishable higher field signal
(
31P NMR
57.75 ppm, JPeSe¼890 Hz and 58.66 ppm JPeSe¼906 Hz), origi-
1
1
-11
-12
P1’ + P2
nally in a 2:1 ratio. If so, one could assume this signal of being
Int2’
-16
-17
a tetracoordinate ionic intermediate III (Int1, 31P NMR
1JPeSe¼890 Hz).
d 57.75 ppm,
The second minimum of the potential energy profile corre-
Reaction Coordinate
sponds to a spirocyclic pentacoordinate intermediate of the pos-
Fig. 5. Potential energy profile for the PN(O)/P(S)O reaction including solvent effects
for MeCN (full line) and THF (dashed line). The gas-phase PEP is included as a reference
(dotted line).
tulated structure
V (Int2 in Fig. 2). The formation of this
intermediate is preceded by an overcoming of the second energy
barrier lowered in solvent to 10 kcal/mol and transition state TS2.
The second intermediate, which is a priori more stable (struc-
ture V), and should be a result of pseudorotation40 with the PeS
bond placed in an apical position, has a distorted structure of TBP
with SePO(1) and SePO(2) angles of 99.1ꢁ and 121.1ꢁ, respectively.
The structure of the second intermediate of the PN(O)/P(S)O re-
action shows very similar characteristics, with the representative
angles O(3)PO(1)¼118.2ꢁ and O(3)PO(2)¼100.3ꢁ, respectively. In
both intermediates, the PO(1) and PO(2) distances are not equal,11
additionally confirming the distorted structures of these species.
On the potential surface there is no identified structure of pre-
liminary pentacoordinate species IV, which would evolve via
pseudorotation to V. Such a result, however, is not completely
unexpected. Recently, the ab initio studies at RHF and B3LYP levels
of theory of the Staudinger reaction showed that the mechanism
would produce a pentacoordinate intermediate with a four-mem-
bered ring with the sp3 nitrogen atom in an apical position.41
However, the cyclization transition state (analogous to in-
termediate III of the Stec reaction) was found to convert directly
into more a stable isomer without intermediacy of the initial
pentacoordinate.
We have obtained better results for the calculated values of cou-
pling constants at the PW91/ADF level of theory, for 2-aniline-2-
seleno-1,3,2-benzodioxaphospholane (1) (JPeSe¼1040 Hz) where
they almost merge with the experimental ones (JPeSe¼1018
e1028 Hz), and, with a slightly less accurate value for the anionic
species 11.
3. Conclusions
Over the past few decades quantum mechanical calculations
have provided considerable insight into the mechanisms of
chemical reactions. Careful examination of the potential energy
profiles calculated by QM/MM methods, support or question
a mechanism due to possibility of evaluation of structures proposed
for presumed transition states and intermediates.10,11
It is accepted that the Stec reaction, being a particular variation of
the WordswortheEmmons reaction (Scheme 1), involves a two-step
process: an initial addition of an electrophile to the activated phos-
phoramidates (or their thio- or seleno-analogues), followed by for-
mation of intermediates (and their rearrangements), with
simultaneous break of the PeN and CeS bonds and elimination of
cumulene, resulting in formation of the corresponding phosphate,
phosphorothioate, or phosphoroselenoate, respectively. Penta-
The question arises about the rate limiting step, and relevance of
assignment of the 31P NMR signal close to the anion 11 as an in-
termediate. Since the inclusion of the solvent effect increases the
initial barrier by around 3 kcal/mol with respect to the gas-phase,
the intermediate and the second transition state (TS2) show

M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
5561
Fig. 6. The geometrical parameters of 1, 11, Int1, Int2, and 12 at the PW91/G03 (PW91/ADF) levels of theory.
Table 1
Experimental and theoretical 31P chemical shifts,
d
(in ppm) and JPeSe values (in Hz)a
Anionic substrate 11 (R1)
Substrate 1
Int1
Int2
12 (P1)
d
JPeSe
1018
d
JPeSe
880
d
JPeSe
d
JPeSe
d
JPeSe
875
Expt
PW91/ADF
mPW1PW91/G03 (gas-phase)
mPW1PW91/G03 (CH3CN)
mPW1PW91/G03 (THF)
77.9
105
118
120
119
59.8
92
120.9
179
212
211
211
ꢀ1040
ꢀ925
118
133
133
133
ꢀ987
d
124
59
ꢀ871
d
ꢀ891
d
100
94
d
d
d
d
d
57
57
d
d
d
95
d
d
d
d
a
Chemical shifts are calculated with respect to the computed H3PO4 absolute values¼295 ppm (PW91/ADF); 303 ppm (mPW1PW91/G03 in gas-phase); 300 ppm
(mPW1PW91/G03 in MeCN); 300 ppm (mPW1PW91/G03 in THF). Experimental value¼328 ppm.
a higher stability, with TS2 around 9 kcal/mol over the reactant
asymptotes. It turns out to be quite apparent that the first stage of
the reaction is the rate-determining step. However, the differences
of energy for the TS1 and TS2 in solvent are rather moderate
potential energies of TS1 and TS2 (
D
E wꢀ1 kcal/mol in a gas phase
versus 5 kcal/mol in MeCN), corroborating a strong but multifac-
eted stabilization of ionic TS1/TS2. Experimental data support this
interpretation, because in the case of the PN(O)/P(S)O reaction we
have not observed stable anions accompanied by much less stable
species resonating at higher field.
(D
E¼0.58 kcal/mol), casting doubt over such a postulate. The sol-
vent effects (solvation of anionic species) increase differences in the

5562
M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
benzene, and the X-ray analysis was performed.8 The structural
parameters obtained for the structure of (details in
Supplementary data) were used in molecular calculations.
Table 2
Calculated (PW91/ADF) contributions to JPeSe values (in Hz)
5
DOa
PSOb
FCþSDc
Total JPeSe
Substrate 1
Anionic substrate 11
Int1
Int2
12 (P1)
0.17
0.17
0.20
0.18
0.15
ꢀ134.9
ꢀ126.6
ꢀ146.2
ꢀ135.0
ꢀ135.5
ꢀ905.2
ꢀ798.8
ꢀ914.2
ꢀ736.2
ꢀ755.6
ꢀ1040
ꢀ925
ꢀ987
ꢀ871
ꢀ891
4.4. 2-Aniline-2-thio-1,3,2-benzodioxaphospholane (4)
A solution of 2-chloro-1,3,2-benzodioxaphospholane (0.88 g,
0.05 mol) was added dropwise to a suspension of sulfur (1.6 g,
0.05 mol) and aniline (10 mL, 0.11 mol) in dry toluene under argon
and was sealed via Teflon gauge. The reaction mixture was stirred at
room temperature overnight. After extraction with a 5% solution of
hydrochloric acid, drying with MgSO4, and concentration, a crude
oily product was purified by means of silica gel chromatography
(CHCl3) to give pure 4 as white amorphous solid after vacuum
concentration (2.23 g, 85%). Rf (CHCl3/EtOH, 95:5 v/v) 0.63; MS CI
m/z 264.1 (M), 232.1 (MꢀS), 173.1, 306.2 (requires C12H10O2N1P1S1
263.0168); dP (C6D6) 77.03; dH (C6D6) 7.36 (2H, d, J 7.93) 7.35 (2H, J
7.81), 7.12 (2H, J 7.81), 6.87, 6.92 (m, 4H).
a
Diamagnetic orbital contribution.
Paramagnetic orbital contribution.
Fermi-contactþspin-dipolar contribution (FC and SD contributions cannot be
b
c
separated in this formalism).
The resulting major difference in mechanism proposed before,
and the one appearing from the reported theoretical calculations at
this level of approximations, is manifested in a consecutive rather
than simultaneous break of the PeN and PeS bonds. The de-
composition of the TS2 proceeds in the way of the advanced PeN
bond breaking followed by the PeS breaking. The structure of the
TS3 for the PN(Se)/P(S)O reaction already has a tetracoordinate
structure with the PeN bond split.
4.5. Low temperature studies by 31P NMR
In conclusion, the reported studies provide additional in-
formation on the mechanism of the PN/PX (O, S, Se) conversion,
giving for the first time a comparison of the experimental results
involving various temperature NMR studies and theoretical DFT
studies. The potential surfaces allow for identification of the ma-
jority of the previously postulated intermediates and mark several
well-defined transition states.
Low temperature experiments were performed in DMF with
toluene-d8 as internal standard added (5:1 v/v). The reagents were
frozen gradually in layers in NMR tubes, sealed with septa under
argon, and frozen in liquid nitrogen. The tubes were warmed up
slowly to ꢀ60 ꢁC in NMR machine and spectra were registered
every 20 ꢁC.
Acknowledgements
4. Experimental section
4.1. General
Authors kindly acknowledge the support and stimulating in-
spiration of Prof. Wojciech J. Stec.
Reactions were carried out under a positive pressure of dry ar-
gon. Solvents and reagents were purified according to standard
laboratory techniques and distilled directly into reaction vessels.
Column chromatography and TLC analyses were performed on
silica gel, and silica gel HP TLC precoated F254 plates, respectively.
NMR spectra were recorded on a Bruker Avance DRX 500 spec-
trometer, operating at 500.13 MHz (1H), and 202.46 MHz (31P).
Supplementary data
An ORTEP drawing with selected bond parameters for 5, variable
temperature 31P NMR spectra for 1, and details on theoretical cal-
culations. Supplementary data associated with this article can be
found in the online version, at doi:10.1016/j.tet.2012.04.081.
Chemical shifts (d) are reported relative to TMS (1H) and 80% H3PO4
(
31P) as external standards. Mass spectra were recorded on a Fin-
References and notes
nigan Mat 95 (NBA, Csþ gun operating at 13 keV).
1. (a) Stec, W. J. Acc. Chem. Res. 1983, 16, 411 and references cited therein; (b)
Comprehensive Organic Name Reactions, 3V Set; John Wiley & Sons Ltd: New
York, NY, 2009, pp 2647e2650.
4.2. 2-Aniline-2-seleno-1,3,2-benzodioxaphospholane (1)
2. (a) Wozniak, L. A.; Pyzowski, J.; Stec, W. J. J. Org. Chem. 1996, 61, 879; (b) Stec,
W. J.; Wozniak, L. A.; Pyzowski, J.; Niewiarowski, W. Antisense Res. Dev. 1997, 7,
381; (c) Wozniak, L. A.; Wieczorek, M.; Pyzowski, J.; Majzner, W.; Stec;, W. J. J.
Org. Chem. 1998, 63, 5395; (d) Pyzowski, J.; Wozniak, L. A.; Stec, W. J. Org. Lett.
2000, 2, 771.
3. Wozniak, L. W.; Okruszek, A. Chem. Soc. Rev. 2003, 32, 158.
4. The single example of partial stereoselectivity of the PN/P18O conversion in
synthesis of isotopomeric adenosine 50-O-[16O,17O,18O]-phosphates was re-
ported by Gerlt, J. A.; Coderre, J. A.; Mehdi, S. In Advances in Enzymology;
Meister, A., Ed.; Wiley & Sons: New York, NY, 1983; Vol. 55, p 291.
5. An alternative concerted four-center mechanism could also be taken into
consideration, with a transition state analogous to intermediate V.
6. Westheimer, F. H. The Polytopal Rearrangement at Phosphorus In. Rearrange-
ments in Ground and Excited States; Mayo, E., Ed.; Academic: New York, NY,
1980; Vol. 2, p 229.
A solution of 2-chloro-1,3,2-benzodioxaphospholane (0.88 g,
0.05 mol) was added dropwise to a suspension of selenium (3.95 g,
0.05 mol) and aniline (10 mL, 0.11 mol) in dry toluene (40 mL)
under argon and was sealed via a Teflon gauge. The reaction mix-
ture was stirred at 130 ꢁC for 3 days (31P NMR, 96%). After ex-
haustive extraction with 0.05 M citric acid followed by water,
drying with MgSO4 and concentration, a viscous liquid was dis-
solved in CH2Cl2, precipitated from toluene, and 1 obtained as
white amorphous solid (13.94 g, 90%). Rf (CHCl3/EtOH, 95:5 v/v)
0.60; MS CI m/z 312.1 (Mþ2), 232.1 (100) (MꢀSe), 248.1 (MꢀSeþO),
450 (Mþpyrocatechol) (requires C12H10O2N1P1Se1 310.962); dP
(C6D6) 77.8 (1JPeSe 1018 Hz).
7. (a) Ramirez, F.; Marecek, J. F.; Okazaki, H. J. Am. Chem. Soc. 1976, 98, 5310; (b)
Kumara Swamy, K. C.; Satish Kumar, N. Curr. Sci. 2003, 85, 1256; (c) Sau, A. C.;
Holmes, J. M.; Day, R. O.; Holmes, R. R. Inorg. Chem. 1983, 22, 1771; (d) Timo-
sheva, N. V.; Prakasha, T. K.; Chandrasekaran, A.; Day, R. O.; Holmes, R. R. Inorg.
Chem. 1995, 34, 4525.
4.3. 2-Hydroxy-2-seleno-1,3,2-benzodioxaphospholane (5)
8. Szyrej, M.; Wieczorek, W.; Wozniak, L. A. Arkivoc 2011, vi, 286.
9. (a) Robiette, R.; Richardson, J.; Aggarwal, V. K.; Harvey, J. N. J. Am. Chem. Soc.
2005, 127, 13468; (b) Robiette, R.; Richardson, J.; Aggarwal, V. K.; Harvey, J. N. J.
Am. Chem. Soc. 2006, 128, 2394; (c) Appel, R.; Loos, R.; Mayr, H. J. Am. Chem. Soc.
2009, 131, 704.
Product 1 product was dissolved in chloroform and poured on
a silica column under argon. Column chromatography was per-
formed using chloroform as an eluent. Compound 5 was collected
from the column major product (yield >80%, 31P NMR
d 65.2 ppm,
10. (a) Aggarwal, V. K.; Harvey, J. N.; Robiette, R. Angew. Chem., Int. Ed. 2005, 44,
5468; (b) Silva, M. A.; Bellenie, B. R.; Goodman, J. M. Org. Lett. 2004, 6, 2559.
1JPSe¼915 Hz) as anilinium salt. Compound 5 was crystallized from

M.del.C. Michelini et al. / Tetrahedron 68 (2012) 5554e5563
5563
11. (a) Brandt, P.; Norrby, P.-O.; Martin, I.; Rein, T. J. Org. Chem. 1998, 63, 1280; (b)
Ando, K. J. Org. Chem. 1999, 64, 6815.
25. Krawczyk, E.; Skowronska, A.; Michalski, J. J. Chem. Soc., Dalton Trans. 2002,
4471.
12. Burke, K.; Perdew, J. P.; Wang, Y. In Electronic Density Functional Theory: Recent
Progress and New Directions; Dobson, J. F., Vignale, G., Das, M. P., Eds.; Plenum: New
York, NY, 1998.
26. (a) Puke, C.; Erker, G.; Aust, N. C.; Wurthwien, E. U.; Frohlich, R. J. Am. Chem. Soc.
1998, 120, 4863; (b) Wilker, S.; Laurent, Ch.; Sarter, Ch.; Puke, C.; Erker, G. J. Am.
Chem. Soc. 1995, 117, 7293.
13. Perdew, J. P. In Electronic Structure of Solid ’91; Ziesche, P., Eschrig, H., Eds.;
Akademie: Berlin, 1991; p 11.
27. Sase, S.; Kano, N.; Kawashima, T. J. Am. Chem. Soc. 2002, 124, 9706.
28. (a) Gauss, J. Chem. Phys. Lett. 1992, 191, 614; (b) Gauss, J. Chem. Phys. Lett. 1996,
255, 327; For a description of the GIAO approach, see Ditchfield, R. Mol. Phys.
1974, 27, 789; Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112, 851.
29. (a) te Velde, G.; Bickelhaupt, F. M.; van Gisbergen, S. J. A.; Fonseca Guerra, C.;
Baerends, E. J.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931; (b)
Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor. Chem. Acc.
1998, 99, 391; (c) ADF2004.01, SCM, Theoretical Chemistry. Vrije Universiteit:
Amsterdam, Netherlands. http://www.scm.com.
14. Perdew, J. P.; Burke, K.; Wang, Y. Phys. Rev. B 1996, 54, 16533.
15. Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. J. Chem. Phys. 2003, 119, 11113.
16. Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650.
17. Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. V. R. J. Comput. Chem.
1983, 4, 294.
18. Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265.
19. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Fores-
man, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision C.02; Gaussian: Wallingford CT, 2004.
30. Adamo, C.; Barone, V. J. Chem. Phys. 1998, 108, 664.
31. (a) Magyarfalvi, G.; Pulay, P. J. Chem. Phys. 2003, 119, 1350; (b) Bayse, C. A. J.
Chem. Theory Comput. 2005, 1, 1119; (c) Shaghaghi, H.; Ebrahimi, H.; Tafazzoli,
M.; Jalali-Heravi, M. J. Fluorine Chem. 2010, 131, 47.
32. (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648; (b) Lee, C.; Yang, W.; Parr, R. G.
Phys. Rev. B 1988, 37, 785.
33. Robert, V.; Petit, S.; Borshch, A.; Bigot, B. J. Phys. Chem. A 2000, 104, 4586.
34. (a) Chesnut, D. B. J. Phys. Chem. A 2005, 109, 11962; (b) Mazurek, A. P.; Do-
browolski, J. Cz.; Sadlej, J. J. Mol. Struct. 1997, 436e437, 435.
35. Chruszcz, K.; Barnska, M.; Czarniecki, K.; Proniewicz, L. M. J. Mol. Struct. 2003,
651e653, 729.
36. Holmes, R. R. Chem. Rev. 1996, 96, 927.
37. Lattman, M. Hypervalent Compounds In. Encyclopedia of Inorganic Chemistry;
King, R. B., Ed.; JohnWiley
& Sons: Chichester, England, 1994; Vol. 3,
20. (a) Gonzales, C.; Shlegel, H. B. J. Chem. Phys. 1989, 90, 2154; (b) Gonzales, C.;
Shlegel, H. B. J. Phys. Chem 1990, 94, 5523.
pp 1496e1511; 96, 927e950 and references therein.
38. Sheldrick, W. S. Top. Curr. Chem. 1978, 73, 1.
21. Miertos, S.; Tomasi. J. Chem. Phys. 1982, 65, 239.
39. Lopez, C. S.; Faza, O. N.; de Lera, A. R.; York, D. M. Chem.dEur. J 2005, 11, 2081.
40. (a) Gilheany, D. G. In The Chemistry of Organophosphorus Compounds; Patai, S.,
Ed.; The Chemistry of Functional Groups Series; Wiley & Sons Ltd: Chichester,
UK, 1992; Vol. 3, Chapter 1, doi:10.1002/0470034424; (b) Gillespie, P.; Ramirez,
F.; Ugi, I.; Marquarding, D. Angew. Chem., Int. Ed. Engl. 1973, 12, 91.
41. Alajarin, M.; Conesaa, C.; Rzepa, H. S. Perkin Trans. 2 1999, 1811.
22. Cossi, M.; Barone, V.; Commi, R.; Tomasi, J. Chem. Phys. Lett. 1996, 255, 327.
23. Barone, V.; Cossi, M.; Menucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107, 3210.
24. (a) Kumara Swamy, K. C.; Kumara Swamy, S.; Said, M.; Krishna Kishore, R. S.;
Herbst-Irmer, R.; Pulm, M. Curr. Sci. 2000, 78, 473; (b) Kojima, S.; Takagi, R.;
Akiba, K. J. Am. Chem. Soc. 1997, 119, 5970; (c) Fu, H. J. Chem. Soc., Perkin Trans. 1
1997, 2021.