Addition of sulfenic acids to monosubstituted acetylenes: A theoretical and experimental study

Article abstract of DOI:10.1002/poc.1557

The reaction of benzenesulfenic acid, generated in situ by thermal decomposition of 3-(phenylsulfinyl) propanenitrile, with monosubstituted acetylenes was experimentally and theoretically investigated at the DFT level using the MPW1B95 density functional.

Full text of DOI:10.1002/poc.1557

Research Article
Received: 13 October 2008,
Revised: 2 March 2009,
Accepted: 6 March 2009,
Published online in Wiley InterScience: 17 April 2009
(www.interscience.wiley.com) DOI 10.1002/poc.1557
Addition of sulfenic acids to monosubstituted
acetylenes: a theoretical and experimental
study
Maria Chiara Aversa a,c, Anna Barattucci a,c, Paola Bonaccorsi a,c
and Alessandro Continib,c
*
The reaction of benzenesulfenic acid, generated in situ by thermal decomposition of 3-(phenylsulfinyl)propanenitrile,
with monosubstituted acetylenes was experimentally and theoretically investigated at the DFT level using the
MPW1B95 density functional. A computational model based on the Hard Soft Acid Base (HSAB) principle was
evaluated for its ability to qualitatively and quantitatively predict the regioselectivity, while kinetics and thermo-
dynamics of the reaction were studied through the analysis of the reaction paths leading to the possible regioisomers.
Copyright ß 2009 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this paper.
Keywords: DFT; HSAB principle; regioselectivity; sulfenic acids; syn-Addition
To our knowledge, the syn-addition reaction of sulfenic acid to
carbon–carbon triple bonds has never been subjected to a
INTRODUCTION
The importance of sulfenic acids RSOH as transient intermediates
in biological processes is widely recognized. Oxidation of thiol
groups in living systems and much of the chemistry of the
penicillin sulfoxides have been considered to involve sulfenic
acids.[1] Unfortunately, most of them are too unstable to be
isolated and, for this reason, much of the knowledge of their
reactions has been derived indirectly through the rationalization
of the final products. In particular, the syn-addition of sulfenic
acids to carbon–carbon triple bonds provides an easy way to
obtain vinyl sulfoxides in mild conditions and with some
regioselectivity, and this reaction has found several applications
in organic synthesis.[2–6] Some theoretical works concerning
the chemistry of sulfenic acids were carried out, most of them
dealing with the mechanism of the sulfoxide thermolysis,
through the combination of theoretical calculations with
instrumental analysis.[7–11] Another computational study com-
pared the syn-elimination from corresponding amine oxides,
sulfoxides, and phosphine oxides, showing that the elimination
from an amine oxide occurs with the lowest activation barrier,
with the sulfoxide being the intermediate case.[12] Other works
were found dealing with the mechanism of sulfoxide reduction
by thiols,[13] with the rearrangement of H2SO to give sulfenic acid
HSOH,[14] or studying the acidity of several inorganic sulfur
oxoacids.[15] Concerning the syn-addition of sulfenic acids to
carbon–carbon multiple bonds (Scheme 1) the generally
accepted reaction mechanism is concerted and, when the
sulfenic acid is trapped by a monosubstituted unsaturated
compound, the product is usually the one in which the partial
positive charge deriving from H and approaching the multiple
bond is better stabilized at the TS level, thus the Markovnikov or
anti-Markovnikov products for electron-donor (ED) or electro-
n-withdrawing (EW) substituents.[16]
systematic computational study. In particular, we were interested
in rationalizing the role of the alkyne substituent in governing the
regiochemical outcome as well as in identifying some reliable
computational models capable of predicting the regiochemistry.
Starting from the excellent work of Jenks and coworkers,[10] we
have optimized a DFT based approach to study the addition
between benzenesulfenic acid and alkynes. Indeed, within the
DFT framework, the development of a simple theoretical model
for the qualitative and semi-quantitative prediction of the
addition regiochemistry has been also possible through the HSAB
principle,[17–19] thus assisting in rationalizing the role of the
alkyne substituent. Moreover, the computational analysis of
the reaction paths leading to the different regioisomers and the
comparison of theoretical and experimental results have
provided new and interesting insights into the mechanism,
kinetics, and thermodynamics of the sulfenic acid syn-addition
reactions.
*
Correspondence to: A. Contini, Istituto di Chimica organica ‘‘A. Marchesini’’,
`
Facolta di Farmacia, Universita degli Studi di Milano, Via Venezian 21, 20133
Milano, Italy.
`
E-mail: alessandro.contini@unimi.it
a
b
c
M. C. Aversa, A. Barattucci, P. Bonaccorsi
Dipartimento di Chimica organica e biologica, Universita degli Studi di
`
Messina, Salita Sperone 31 (vill. S. Agata), 98166 Messina, Italy
A. Contini
Istituto di Chimica organica ‘‘A. Marchesini’’, Facolta di Farmacia, Universita
`
degli Studi di Milano, Via Venezian 21, 20133 Milano, Italy
`
M. C. Aversa, A. Barattucci, P. Bonaccorsi, A. Contini
Centro Interuniversitario di Ricerca sulle Reazioni Pericicliche e Sintesi di
Sistemi Etero e Carbociclici, Italy
J. Phys. Org. Chem. 2009, 22 1048–1057
Copyright ß 2009 John Wiley & Sons, Ltd.

SULFENIC ACIDS WITH MONOSUBSTITUTED ACETYLENES
evident exception of alkyne 3f whose addition to 2 results as the
rate limiting step.
Optimization of the theoretical method
Thermolysis of alkyl sulfoxides was theoretically and experimen-
tally studied by Jenks and coworkers,[10] who did a comparison of
several levels of theory with experimental activation and reaction
enthalpies. DFT calculations, performed with the B3LYP func-
tional,[21,22] led to poor results in terms of activation barriers,
typically several kcal/mol too low in comparison with exper-
imental or high level calculation results. Indeed, despite its wide
use, the B3LYP functional is not recommended for reactions
involving proton transfers.[23–25] Very accurate results, in terms of
reproduction of experimentally derived energy values, were
obtained from MP2/6–311 þ G(3df,2p)//MP2/6–31G(d,p) calcu-
lations, thus this method was our first choice for studying the
addition of sulfenic acids to alkynes. Unfortunately, preliminary
calculations showed that this level of theory was computationally
too demanding for our purposes. Indeed, starting from the
B3LYP/6–31G(d,p) optimized geometry, the optimization of the
TS for the addition of benzenesulfenic acid (2) to phenylacetylene
(3c) at the MP2/6–31G(d,p) level took more than 7 days to
converge on a recent quadcore computer. A less CPU intensive
model was needed, because of high demand of frequency
calculations, essential for thermochemistry evaluations, and the
need to study several reaction paths. DFT appeared to be a
reasonable choice, combining a proper treatment of electron
correlation with a reasonably low computation time.[26] Therefore
we decided to evaluate some of the recent density functionals
developed by the Truhlar group, namely the MPW1B95, MPWB1K,
TPSS1KCIS, MPW3LYP, and MPWKCIS1K,[27–34] for their ability to
reproduce both the experimental activation barriers and the
geometries obtained by high level calculations. All calculations
were performed with the 6–31 þ G(d,p) basis set for all atoms
except sulfur, where additional d functions are recommended for
a proper treatment of its electronic structure.[35,36] Thermolyses
of the sulfoxides a and b (Scheme 3) were then used for the
optimization of the theoretical model.
Scheme 1.
RESULTS AND DISCUSSION
Experimental test set
Some representative examples for evaluating the regiochemistry
of the addition were planned in order to cover the main electronic
and steric features of the alkyne substituent (Scheme 2).
Thus, acetylenes bearing an aromatic strong or weak ED group
(3a and 3b respectively), a phenyl substituent (3c), an aromatic
strong EW group (3d), an aliphatic EW group (3e), and a sterically
encumbered weak ED group (3f) were reacted with benzene-
sulfenic acid, generated in situ through the thermal decompo-
sition of 3-(phenylsulfinyl)propanenitrile (1) (alkyne/1 molar ratio
6:1). Sulfenic acid precursor 1 has been obtained in two simple
steps from benzenethiol.[20] The reactions were conducted both
in toluene and in acetonitrile at their reflux temperature in order
to evaluate solvent influences in governing the regiochemical
outcome. Yields have been almost quantitative in all the cases.
Results reported in Scheme 2 suggest that the solvent plays no
relevant action in ruling the regiochemistry, because an apolar
solvent such as toluene and a polar aprotic solvent such as
acetonitrile lead essentially to the same regioisomeric ratio. The
main observed change concerns the reaction time, much shorter
when the addition is conducted in toluene (from 25 to 70 min for
3a and 3e,f, respectively). Bearing in mind that the reaction is
conducted at reflux, the observed differences in reaction kinetics
are probably due to the higher boiling point of toluene, and not
to specific solvent effects. It should also be noted that, as
suggested by calculation results extensively discussed later on, in
most cases the reaction rate limiting step seems to be the
decomposition of 1 to give benzenesulfenic acid 2, with the
Activation enthalpies resulting from the different method-
ologies were compared with those experimentally obtained,
while the most important geometrical parameters of reactants a
and b and the corresponding TSs were compared with the MP2/
6–311 þ G(3df,2p) geometries through a regression analysis. As
shown in Table 1, the overall best performance was obtained with
the MPW1B95 and TPSS1KCIS density functionals, with slightly
better results for the first method. MPW1B95/6–31 þ
G(d,p),S(3df) geometries are shown in Fig. 1, where MP2/
6–311 þ G(3df,2p) values are also reported for comparison.
Scheme 2. (Reaction time and products obtained by the addition of
sulfenic acid 2 to alkynes 3)
Scheme 3. Reaction models used for density functional evaluation
J. Phys. Org. Chem. 2009, 22 1048–1057
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

M. C. AVERSA ET AL.
HSAB model
Table 1. Performance of different density functionals in
respect to experiments or MP2 calculations
DFT, by adopting the electron density r as the central quantity,
provides an excellent tool for the theoretical study of chemical
systems at a reasonable computational cost. Indeed, several
chemical concepts can be precisely defined in the framework of
the conceptual DFT,[37–39] and quantities such as the chemical
potential, electronegativity, hardness, and softness actually
correspond to the linear responses of r with respect to changes
in external potential (n) and number of electrons (N). The relevant
reactivity indexes are the electron chemical potential m ¼ (dE/
dN)n(r), which measures the escaping tendency of electrons
(E ¼ molecular energy), and the softness S ¼ (dN/dm)n(r), which
describes the propensity of the molecule to gain or lose electrons
as a response to a change in m.[37] The above indexes are tied to
the chemical reactivity through the HSAB principle, which is in
turn deeply rooted in DFT. As our objective was to study the
regioselectivity, an atomic reactivity index was needed and the
best suited for this purpose was the local softness s(r) ¼ (dr(r)/
dm)n(r), which describes the sensitivity of r at the point r to a
variation of m.[37] The local softness actually corresponds to the
Fukui function f(r), defined by Parr and Yang,[40] multiplied by the
global softness S and contains the same information as Fukui
functions plus additional information about the total molecular
softness. As r(r) is a discontinuous function of N, within the finite
difference approximation and expressing f(r) in the condensed
form, three local softness indexes can be obtained for each atom
in the molecule: sþ for its reactivity toward nucleophiles, sꢀ for
electrophiles, and s0 for radicals,[41] and the reaction is favored
between atoms having similar s values. Thus, in accordance with
the local HSAB principle,[42] a regioisomer is favored when the
new bonds are formed between atoms with equal softness.
Local softness values, calculated for the atoms involved in the
addition of benzenesulfenic acid (2) to alkynes 3a–f and reported
in Table 2, show that alkynes 3a–d are more reactive toward
electrophiles at the unsubstituted carbon, where the sulfenic acid
proton will be driven leading to a Markovnikov-like addition
product, as generally observed for the syn-addition reactions of
sulfenic acids.[16] On the other hand, in methyl propiolate (3e) the
C—R carbon bears an sꢀ value higher than C—H, which is also
characterized by a particularly high sþ, suggesting that the
addition of benzenesulfenic acid will lead to an anti-Markovnikov
product as the major regioisomer, in accordance with the
experimental findings. The highest sꢀ value computed for the C—H
in alkyne 3f could suggest the Markovnikov product as the
major regioisomer when benzenesulfenic acid is reacted with
trimethylsilylacetylene, in net discordance with the experimental
findings. However, as the general accepted mechanism of the
syn-addition of sulfenic acids to multiple bonds is concerted, the
simultaneous fulfillment of local HSAB principle at all the reaction
centers should be considered. To this end, an expression which
measures such fulfillment in a least square sense has been
proposed for cycloaddition reactions,[43,44] and we applied it to
the addition reaction of 2 to 3a–f (see Table 2; D4 and D5 are
referred to the Markovnikov and anti-Markovnikov regioisomers,
respectively). Considering that the smaller the value of D the
greater is the extent that the HSAB principle is satisfied, our
results reported in Table 2 are in perfect concordance with the
experiments and this suggests that the HSAB principle is a
valuable tool also to get preliminary insights into the reaction
mechanism. Moreover, it is interesting to note that the HSAB
model not only provided a correct regiochemistry prediction for
Methoda
DHz a DHz b
r2b
Standard errorb
B3LYP
29.3
32.1
35.5
29.0
29.3
36.1
33.0
41.0
43.7
46.6
41.6
41.0
47.2
42.0
0.99997
0.99999
0.99998
0.99998
0.99998
0.99996
0.25
0.17
0.22
0.19
0.23
0.29
MPW1B95
MPWB1K
TPSS1KCIS
MPW3LYP
MPWKCIS1K
Exp.[10]
a DFT calculations performed with the 6–31 þ G(d,p),S(3df)
basis set.
b Regression analysis between DFT and MP2/6–311 þ (3df,2p)
geometrical parameters reported in Reference 10. A total of 18
bond distances and 2 bond angles were compared.
In order to estimate the basis set effects on the energies and
geometries, all the stationary points were re-optimized at the
MPW1B95/6–311 þ G(3df,2p) level. The activation enthalpies of
reaction models A and B decreased by 0.5 and 1.6 kcal/mol,
respectively, and no relevant changes were observed in
geometries. Almost the same energetic results (DHz ¼ 31.6 and
42.0 kcal/mol for models A and B, respectively) were obtained by
performing single point calculations at the MPW1B95/
6–311 þ G(3df,2p) level, thus justifying the use of the lighter
basis set for geometry optimizations.
Figure 1. MPW1B95/6–31 þ G(d,p),S(3df) optimized reactants and TSs
for the thermolysys of sulfoxides a and b, with MP2/6–311 þ G(3df,2p)
values in parenthesis.[10] Distances are reported in angstroms, angles in
degrees
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1048–1057

SULFENIC ACIDS WITH MONOSUBSTITUTED ACETYLENES
Table 2. Local softness computed for atoms involved in the addition of 2 to 3a–f,a HSAB fulfilment degrees D4 and D5, chemical
potential differences Dm2–3 (eV), and differences in grand potential variation DDVR1-R2 (kJ/mol)b
c
c
sþ
sþ
sꢀ
sꢀ
C-R
D4
D5
Dm2–3
DDV4–5
—
RC CH
—
R
C-H
C-R
C-H
3a
3b
3c
3d
3e
3f
p-CH3OC6H4
p-CH3C6H4
Ph
p-NO2C6H4
COOCH3
TMS
0.83
0.73
0.60
0.55
1.19
0.34
ꢀ0.08
ꢀ0.02
0.21
ꢀ0.11
ꢀ0.40
ꢀ0.21
0.83
0.89
0.79
0.99
0.53
0.75
ꢀ0.15
ꢀ0.03
0.23
0.18
0.62
3.43
3.13
2.71
3.22
5.23
4.02
4.44
4.11
3.40
3.65
1.61
2.94
ꢀ0.05
0.26
0.45
1.71
1.40
0.90
ꢀ3.9 ꢂ 10ꢀ2
ꢀ4.8 ꢂ 10ꢀ4
ꢀ1.6 ꢂ 10ꢀ2
2.0 ꢂ 10ꢀ1
4.2 ꢂ 100
0.56
5.9 ꢂ 10ꢀ1
a The following values were computed for benzenesulfenic acid (2): sþ(S) ¼ 1.94; sꢀ(S) ¼ 1.46; sþ(H) ¼ 1.86; sꢀ(H) ¼ 0.06.
b For the calculation of the HSAB fulfillment degrees D and grand potential variation DDV the acidic hydrogen was considered the
electrophile and sulfur the nucleophile.
c D4 ¼ (sþH–sꢀC–H)2 þ (sꢀS–sþC–R)2; D5 ¼ (sþH–sꢀC–R)2 þ (sꢀS–sþC–H)2.
entries 1–4, where the same result could be obtained by simply
applying the Markovnikov’s rule, but also for the more trivial entry
5 where, the alkylsylyl group being a weak ED, the Markovnikov’s
rule would have had predicted the wrong regioisomer.
isolated reactants (see the Computational analysis of the reaction
paths section). However, our impression is that some attention
should be paid when quantitative predictions are searched
through DDV calculations, since a quantitative prediction can
only be made through a linear regression analysis which provides
the actual relationship between DDV and DDEz. Indeed, by
analyzing some of the examples reported in the literature,[46,48,49]
it can be observed that the weighted linear regression equations
are rather different from case to case. This means that a proper
experimental ‘‘training set’’ should be prepared for each specific
study, thus limiting the applicability of this valuable method to
those problems well covered by experimental examples.
It was recently reported that, within the HSAB theory,
quantitative regioselectivity predictions could also be made by
calculating the variation DV in the grand potential.[45] Indeed, it is
assumed that when two reactants approach each other the
interaction occurs between pairs of atoms located in different
molecules, and charge is transferred within such pairs in the very
first step of the bond-forming interaction between the specific
atoms. Such a transfer equalizes the electron chemical potential
and induces a variation DV of the grand potential (that is the
natural thermodynamic quantity describing the behavior of the
reactant atoms, which are open subsystems freely exchanging
energy and electrons) of the system. For concerted reactions, the
difference DDV between two regioisomers is expected to be
proportional to the energy difference between their TSs thus
enabling a quantitative prediction without the need to locate the
TSs, as reported by several successful examples.[46–50] We then
decided to apply this strategy by calculating DV for each
regioisomer, according to Eqns (1) and (2)
Computational analysis of the reaction paths
The main power of the local HSAB model is that predictions can
be made by performing simple calculations on the isolated
reactants only, without the need to localize all the TSs. However,
this step is absolutely necessary for a thorough analysis of the
reaction mechanism. For example, the exceptionally high value of
DDV obtained for the reaction of methyl propiolate (3e) remains
unclear. Indeed, on the basis of the local softness computed for
3e, the Markovnikov product 4e should not be obtained.
Moreover, theoretical considerations on the reaction kinetics and
thermodynamics can only be made by localizing all the stationary
points along the potential energy surface (PES). Finally, the
geometrical analysis of TSs can elucidate important aspects of the
reaction mechanism such as the synchronicity degree. For this
reasons, all the TSs and products for the two regioisomeric paths
of the reaction of 2 with 3a–f were located and optimized at the
same level of theory discussed above. Activation free energies are
collected in Table 3, where differences between the two
regioisomeric paths 4 and 5 are also reported for clarity.
The mechanism of decomposition of sulfoxide 1 was also
investigated in order to qualitatively evaluate the possibility for
the reaction of benzenesulfenic acid 2 with alkynes 3a–e to
compete with the reverse reaction of 2 with acrylonitrile. Results
are graphically depicted in Scheme 4, where activation and
reaction free energy ranges are also reported.
ꢀ
ꢀ
ꢁ
ꢁ
1
SHSCꢀH SSSCꢀR
SHSCꢀH SSSCꢀR
2
DV4 ¼ ꢀ ðm2 ꢀ m3Þ ꢁ
þ
(1)
(2)
2
1
SHSCꢀR SSSCꢀH
2
DV5 ¼ ꢀ ðm2 ꢀ m3Þ ꢁ
þ
2
SHSCꢀR SSSCꢀH
The obtained results, collected in Table 2, are in linear
relationship with the activation energies DDEz obtained from the
experimental ratio 4:5 through the Arrhenius equation, with the
only exception of methyl propiolate (3e) which behaves as an
outlier. The difference in activation energy for the two
regioisomers 4 and 5 was obtained as DDEz ¼ !RT ꢁ ln(Y), where
T is the reaction temperature (383.6 K for toluene) and Y is the
experimental ratio of 4:5. Analogously, the ratio 4:5 can be
predicted from computed activation energies as Y ¼e(Ea5-Ea4)/RT
.
Indeed,
a
least-square linear regression results in
DDV ¼ 0.02 ꢁ DDEz þ 0.30 with a correlation coefficient r2 ¼ 0.99
only if the value computed for 3e is excluded. This could be due
to secondary interactions at the TS level, which alter the value of
DV in a way that cannot be predicted by calculations on the
The decomposition of 1 to give 2 and acrylonitrile occurs
through an activation barrier DGz ¼ 25.3 and appears to be
slightly endothermic (DG ¼ 2.4). Keeping the sum of the energies
J. Phys. Org. Chem. 2009, 22 1048–1057
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

M. C. AVERSA ET AL.
Table 3. Activation and reaction free energies (kcal/mol) for the addition of 2 to 3a–f, and activation and reaction free energy
differences (kcal/mol) between regioisomers 4 and 5a
DGz
DGz
DG4
DG5
DDGz
4–5
DDG4–5
—
RC CH
—
R
4
5
3a
3b
3c
3d
3e
3f
p-CH3OC6H4
p-CH3C6H4
Ph
p-NO2C6H4
COOCH3
TMS
24.6
25.0
25.0
25.2
22.9
29.1
27.1
27.9
27.9
26.4
22.0
26.6
ꢀ10.0
ꢀ10.0
ꢀ9.8
ꢀ11.8
ꢀ12.5
ꢀ12.5
ꢀ12.2
ꢀ18.2
ꢀ5.5
2.5
2.9
2.9
ꢀ1.8
ꢀ2.5
ꢀ2.7
ꢀ2.4
ꢀ2.3
ꢀ1.3
ꢀ9.8
1.4
ꢀ0.9
ꢀ15.9
ꢀ4.3
ꢀ2.5
a Calculated as the sum of MPW1B95/6–311 þ G(3df,2p) energy and the thermal correction to Gibbs free energy obtained from
thermochemical calculations (383.6 K, 1 atm, in order to simulate the reaction experimental conditions for toluene) at the MPW1B95/
6–31 þ G(d,p),S(3df) level. Activation barriers are calculated as the energy difference between TSs and the sum of the energies of
isolated reactants 1 and 3a–f, while reaction energies correspond to the energy differences between products and the sum of the
isolated reactants energies.
of 1 and 3 as the reference, the activation barriers for the step
leading to the isolable products 4 and 5 ranges from 22.0 kcal/
mol (computed for the addition of 2 to 3e to give 4e) to 26.6 kcal/
mol (obtained for the addition of 2 to 3f to provide 4f). With the
exception of 3e, where a quite low DGz was obtained due to the
strong EW group COOCH3 which lowers the activation barriers,[10]
the competition of the addition of benzensulfenic acid 2 to
alkynes 3 with the reverse reaction bringing back to 1 appears to
be possible due to the very low DDGz between the two processes.
Indeed, the addition of 2 to 3a to give 4a is favored over the
reaction of 2 with acrylonitrile by only 0.6 kcal/mol. An even lower
difference was computed for 3b,c (0.3 kcal/mol) and for 3d
(0.1 kcal/mol), accordingly to the different electronic effects
exerted by the aromatic substituent. For alkyne 3f a considerably
higher activation barrier was observed and competition with the
reverse reaction was considerable. Indeed, the step leading to
product 5f is kinetically unfavored with respect to the reverse
reaction by 1.2 kcal/mol, evidently due to the TMS substituent
steric encumbrance which considerably raises the activation
barrier. The steric hindrance of the TMS group is even more
evident in the Markovnikov TS-4f (see Fig. 2) which in fact was
never isolated in the adopted experimental conditions.
barriers, but, once again, methyl propiolate (3e) behaves as an
outlier having the lowest activation barrier (DGz ¼ 22.9 and
22.0 kcal/mol for TS-4 and TS-5, respectively) and the longest
reaction time (70 min when reacted in toluene). This suggests
that some other chemical equilibrium may play a role in the
addition of sulfenic acids to methyl propiolate. Indeed, the local
HSAB analysis for compounds 3a-d,f showed that the highest sꢀ
value, which distinguishes the most nucleophilic atom, is
obtained for the alkyne C—H. The early stage of the addition
reaction will be thus characterized by the formation of a complex
between the phenylsulfenic acid 2 and alkynes 3a–d,f which will
be closely related to the corresponding TSs. Otherwise, the
highest sꢀꢀvalue for compound 3e is observed on the carbonyl
oxygen (sO¼C ¼ 1:08) meaning that, in this case, a reactant
complex (RC) structurally unrelated with the addition TS can be
formed by the non-covalent interaction of the carbonyl oxygen
with the sulfenic acid hydrogen. To support this hypothesis, the
early stage RC for the reaction of phenylsulfenic acid 2 with either
alkyne 3a or alkine 3e was obtained through a reverse intrinsic
reaction coordinates (IRC) analysis starting from TS-4a and TS-5e,
respectively, followed by full optimization. To be sure that the
lowest energy complexes were obtained, different conformations
were also optimized and compared. The most favored
geometries are represented in Fig. 3.
It can also be observed that the reaction times (see Table 1) are
in a general linear dependence with the computed activation
A visual inspection of the obtained complexes shows that the
geometry of RCa closely resembles the corresponding TS-4a
Scheme 4. Representation of the potential energy surface for the ther-
mal decomposition of 1 in the presence of alkynes 3. Activation and
reaction free energy differences are reported in kcal/mol
Figure 2. Optimized TS-4f and TS-5f
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1048–1057

SULFENIC ACIDS WITH MONOSUBSTITUTED ACETYLENES
Figure 4. Predicted versus experimental percentage of regioisomers 4
kinetic control even if the reactions are performed under reflux
conditions. Indeed, the anti-Markovnikov regioisomers 5 are the
thermodynamic products, being 1.3–2.7 kcal/mol more stable
than the corresponding products 4, but a kinetic-thermodynamic
competition cannot be expected in the adopted conditions due
to the high activation barriers of the reverse reactions. The
geometrical and vibrational analysis of the optimized stationary
points and IRC analyses, reported in Fig. 5, confirm that the
reactions studied herein are concerted, in concordance to the
generally accepted mechanism.[16]
Figure 3. Reactant complexes obtained between 2 and 3a (RCa) or 3e
(RCe) and free energy barriers between RCs and corresponding TSs
structure, while RCe converged to a structure which is far away
from the TS geometry (similar results were obtained starting from
either TS-4e or TS-5e) and is characterized by a hydrogen bond
between the acidic sulfenic acid hydrogen and the carbonyl
oxygen, as predicted by the HSAB analysis. The free energies
calculated for both complexes were found to be higher than
those associated to the sum of the free energies of the isolated
reactants 2 and 3a,e, but a DG of 6.4 kcal/mol was obtained for
RCa, while a DG of 2.4 kcal/mol was computed for RCe. This
means that, by considering the DG of the RC as the reference, a
DGz of 15.8 kcal/mol would be necessary for the conversion of
RCa in the corresponding TS-4a, while a DGz of 17.2 kcal/mol
would be required to go from RCe to TS-5e, thus justifying the
striking differences in reaction time. Different hypotheses of
competing reactions were also thoroughly considered, but both
computational results and chemical logic agreed in suggesting
the above explanation as the most convincing.
Coming back to the main question of the reaction regiochem-
istry, useful information can be obtained through the activation
and reaction free energy differences DDGz
and DDG4–5,
4–5
respectively. Indeed, for the first three terms 3a–c the energy
barrier leading to the Markovnikov products 4 is about 3 kcal/mol
lower than the one calculated for products 5, which in fact were
never experimentally obtained. The DDGz between TS-4d and
TS-5d drops to 1.4 kcal/mol, and as a consequence traces of the
product (E)-1-(4-nitrophenyl)-2-(phenylsulfinyl)-ethene (5d) can
be expected. The opposite is observed for 3e, where the
anti-Markovnikov product 5e is kinetically favored over 4e by
0.9 kcal/mol, and for 3f where the DDGz raises to 2.5 kcal/mol in
favor of the anti-Markovnikov TS-5f. The ratios between 4 and 5
were calculated from the above-mentioned energy differences
through the Arrhenius equation, and compared with experimen-
tal results through a regression analysis.
As depicted in Fig. 4, the linear correlation between computed
and experimental ratios is excellent (r2 ¼ 0.995) and this fact,
while strengthening the adopted computational method,
suggests that the regiochemical outcome is completely under
Figure 5. IRC calculations for the reactions of 2 with 3a and 3e at
MPW1B95/6–31 þ G(d,p),S(3df). Relative energies are referred to the
energy of the isolated reactants. This figure is available in colour online
at www.interscience.wiley.com/journal/poc
J. Phys. Org. Chem. 2009, 22 1048–1057
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

M. C. AVERSA ET AL.
Table 4. Selected Wiberg bond order indexes.a BOI differ-
ences between the forming bonds in TSs are reported in
parenthesis.
Species
C–H
C–S
C–C
S–O
O–H
TS-4a
TS-5a
4a
0.37
0.27
0.90
0.88
0.35
0.27
0.90
0.88
0.35
0.26
0.89
0.88
0.34
0.18
0.89
0.87
0.15
0.15
0.89
0.87
0.36
0.30
0.89
0.87
0.29 (0.08)
0.41 (0.14)
0.88
2.35
2.31
1.89
1.82
2.36
2.32
1.89
1.82
2.36
2.33
1.89
1.83
2.36
2.33
1.89
1.83
2.47
2.40
1.86
1.85
2.39
2.38
1.93
1.94
1.04
1.02
1.24
1.25
1.03
1.01
1.24
1.25
1.03
1.01
1.25
1.25
1.04
0.98
1.25
1.25
0.95
0.97
1.24
1.26
1.04
1.01
1.25
1.25
0.39
0.46
5a
0.93
TS-4b
TS-5b
4b
0.31 (0.04)
0.41 (0.14)
0.88
0.41
0.45
5b
0.93
TS-4c
TS-5c
4c
0.30 (0.05)
0.42 (0.16)
0.88
0.40
0.46
Figure 6. Optimized TSs for the addition of 2 to 3a and 3e. Selected
˚
distances are reported in A
5c
0.93
TS-4d
TS-5d
4d
0.31 (0.03)
0.48 (0.30)
0.87
0.41
0.52
5d
0.93
(R ¼ p-CH3OC6H4) and an even more evident dependence seems
to exist for path 5 with a DBOI of 0.14, obtained for TS-5a,b
(R ¼ p-CH3OC6H4 and p-CH3C6H4, respectively), which raises to
0.28 and 0.30 for TS-5e (R ¼ COOCH3) and TS-5d
(R ¼ p-NO2C6H4), respectively. Generally, it can be concluded
that ED substituents, while favoring the Markovnikov like TS-4,
increase the reactions’ asynchronicity by anticipating the
protonation step, as the C—H BOI is slightly closer to its final
state than the C—S. Contrarily, EW substituents favor the
anti-Markovnikov like TS-5 and increase the asynchronicity by
anticipating the nucleophilic attack of the S atom, as confirmed
by the C—S BOI higher than C—H. The peculiar behavior of
methyl propiolate (also observed in the HSAB analysis where
product 4e was found to be decidedly unfavored, in net
disagreement with the experimental evidence) is still unclear.
However, a singularity consisting in an attractive weak hydrogen
interaction possible for TS-4e but not for TS-5e was revealed by
the analysis of the TS geometries, depicted in Fig. 6.
TS-4e
TS-5e
4e
0.33 (0.18)
0.43 (0.28)
0.88
0.57
0.55
5e
0.93
TS-4f
TS-5f
4f
0.38 (0.02)
0.45 (0.15)
0.91
0.40
0.43
5f
0.91
a Carbon–carbon BOIs for reactants 3a–f are 2.83, 2.84, 2.84,
2.83, 2.86, and 2.93, respectively. S–O and O–H BOIs for 2 are
0.89 and 0.71, respectively.
Table 4 reports the selected Bond Order Indexes (BOIs)
obtained from the Natural Bond Orbital (NBO) analysis of the
optimized stationary points,[51–54] which can be compared in
order to gain information about the ‘‘earliness’’ or ‘‘lateness’’ of
TSs, as well as to evaluate the general synchronicity of the
examined reactions. The computed values for the S—O bond in
products confirms the ylide character of sulfoxides, as previously
observed by Jenks[10] and by Sundberg and Molina[55] through an
Atoms-in-Molecules analysis.
Indeed, the three parameters d (the H ꢁ ꢁ ꢁ O distance), u (the
C—H ꢁ ꢁ ꢁ O angle) and D (the distance between the C—H carbon
˚
and the carbonyl oxygen) measured for TS-4e (d ¼ 2.64 A,
˚
u ¼ 106.68, D ¼ 3.13 A) perfectly fall within the ranges reported
in the literature for similar interactions.[56] The energy contri-
bution of the hydrogen interaction was estimated by performing
—
The inspection of BOIs calculated for the C—H and C—S bonds
in TSs reveals that all reactions follow a quite synchronous
mechanism, although syn-additions leading to products 4 result
a PES scan over the rotation of the C—C O s bond, followed by
—
a full optimization of the rotamer TS, which resulted in 1.3 kcal/
mol less stable than TS-4e at the MPW1B95/6–31 þ G(d,p),S(3df)
level of theory. It is reasonable that such a hydrogen interaction
stabilizes TS-4e with respect to TS-5e and allows to obtain the
product 4e together with 5e, in contrast to the corresponding
result from the HSAB analysis which, being conducted on the
isolated reactants, cannot consider any secondary interaction at
the TS level.
from
a slightly more synchronous mechanism than the
corresponding path 5, where the sulfur addition on the
unsubstituted alkyne carbon foregoes the protonation step.
The difference between the BOIs (DBOI) of the two forming bonds,
which can be taken as a measure of synchronicity, ranges from
0.02 to 0.18 (TS-4f and TS-4e, respectively) for path 4, while
values between 0.14 and 0.28 (TS-5a,b and TS-5e, respectively)
are obtained for path 5, with the addition of methyl propiolate
being the most asynchronous in both cases. This is quite
surprising, as opposite relationships between the synchronicity
and the electronic character of the R substituent can be observed
for paths 4 and 5. Indeed DBOI increases from 0.03 to 0.08
by switching from TS-4d (R ¼ p-NO2C6H4) to TS-4a
CONCLUSIONS
DFT calculations with the MPW1B95 functional proved to be
adequate for studying the syn-addition of sulfenic acids onto
monosubstituted acetylenes. A model based on the local HSAB
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1048–1057

SULFENIC ACIDS WITH MONOSUBSTITUTED ACETYLENES
principle was evaluated for its ability to predict the regiochemical
outcome of the studied reactions and qualitative predictions
were perfectly in line with experimental results. Quantitative
predictions based on the grand potential variation DDV were
found to be correct for 3a–d,f, but failed for methyl propiolate
(3e) where the anti-Markovnikov regioisomer 4e was predicted
as the only product. On the other hand, the computational
analysis of all the stationary points along the reaction PES allowed
to accurately compute the differences in activation energies for
all the examples herein reported and provided an explanation of
the peculiar behavior observed for methyl propiolate. Moreover,
the role of the alkyne substituent in affecting the activation
barriers, ruling the regiochemical outcome and influencing the
reaction mechanism was rationalized and discussed, showing
how the ED substituents favor the addition of sulfur on the most
substituted carbon while increasing the corresponding TSs
asynchronicity by anticipating the protonation step. Contrarily,
EW substituents favor the addition of sulfur on the least
substituted carbon and increase the corresponding TSs asyn-
chronicity by anticipating the nucleophilic attack of the S atom.
m¼!(IP þ EA)/2 and S ¼ 1/(IP!EA), where IP and EA are the vertical
ionization potential and the electron affinity, respectively. The
cþondensed form of the local softness was calculated as
sk ¼ [qk(N þ 1)!qk(N)]S and skꢀ ¼ [qk(N)!qk(N!1)]S for the reactivity
towards nucleophiles and electrophiles, respectively, were qk(N),
qk(N þ 1) and qk(N–1) represent the gross electron population of
the atom k in the neutral, anionic, and cationic systems,
respectively. Basis sets using Cartesian d and f functions were
always required and all calculations were performed with the
Gaussian03 package.[59]
Synthetic part: general
Solvents were purified according to standard procedures. Light
petroleum (petrol) used refers to the fraction boiling at 30–50 8C.
All reactions were monitored by TLC on commercially available
precoated plates (Aldrich silica gel 60, F254) and the products
were visualized with vanillin [1 g dissolved in CH3OH (60 ml) and
conc. H2SO4 (0.6 ml)]. Silica gel Aldrich 60 was used for column
1
chromatography. H and 13C NMR spectra were recorded on a
Varian Mercury 300 spectrometer at 300 and 75 MHz respectively
in CDCl3 solutions with Si(CH3)4 as the internal standard: the
attributions are supported by Attached Proton Test (APT) and
homodecoupling experiments; the indicated proton and carbon
nuclei by pertain to the phenylsulfinyl group. Mass spectra were
measured by Electron Impact (EI, 70 eV) with a Finnigan MAT 90
instrument. Compounds 4b,c,e,[63–65] and 5d,e[66,67] have been
already described. Additional spectral data are reported for
compounds 4b,c,e and 5e.
EXPERIMENTAL
Theoretical calculations
Reactants 2, 3a–f, products 4a–f and 5a–f and the corresponding
TSs-4a-f and TSs-5a-f were fully optimized in the gas phase at
the MPW1B95/6–31 þ G(d,p),S(3df) level of theory.[35,36]
Vibrational frequencies were computed at the same level of
theory to define optimized geometries as minima (no imaginary
frequencies) or TSs (a unique imaginary frequency corresponding
to the vibrational stretching of the forming/breaking bonds) and
to calculate ZPVE and thermochemical corrections to electronic
energies (1 atm, 355.0 and 383.6 K, corresponding to the
experimental conditions for acetonitrile or toluene at reflux).
Single point calculations were performed at the MPW1B95/
6–311 þ G(3df,2p) in the gas phase and in solution (toluene and
acetonitrile) using both the PCM and CPCM solvent models.[57,58]
Different topological models and set of atomic radii were
adopted for the construction of the molecular cavity (the default
united atom model UA0 as implemented in Gaussian03,[59] the
Pauling model, also referred as the Merz–Kollman, model,[60] or
the Bondi model[61]), but none of the above strategies
outperformed gas-phase calculations in terms of reproduction
of the experimental outcome. For this reason, all the results
herein reported are referred to gas-phase calculations. Compet-
ing biradical mechanisms for the sulfoxide syn-eliminations were
excluded by Jenks and coworkers through a CASSCF analysis of
the TSs;[10] however the RHF-UHF stability of the wave function
was checked for TSs-4a,e and TSs-5a,e, but both RMPW1B95 and
UMPW1B95 methods provided the same energy, and no
instabilities were observed. IRC analyses were performed at
the MPW1B95/6–31 þ G(d,p),S(3df) level starting from TSs-4a,e
and TSs-5a,e requesting in each case a step size of 0.1 or
0.05 amu1/2 Bohr. HSAB calculations were performed on reactants
2, 3a–f optimized at the MPW1B95/6–31 þ G(d,p),S(3df) level.
The anion and cation of 2, 3a–f were treated at the UMPW1B95/
6–31 þ G(d,p),S(3df) level using the geometry of the neutral
system. Atomic electron populations were evaluated with the
Merz–Kollman charge scheme.[62] Reactivity indexes were
computed within the finite difference approximation as
General procedure for the thermolysis of sulfoxide 1 in the
presence of alkynes 3
To a 0.2 M solution of 1 in toluene or acetonitrile, the commercial
acceptor 3 (6 equiv) was added, and the mixture maintained at
reflux temperature. When the reaction appeared complete by
TLC (disappearance of 1, see reaction time in Scheme 1), the
crude reaction product was purified by flash column chroma-
tography on silica gel (EtOAc/petrol from 9:1 to 4:1).
1-(4-Methoxyphenyl)-1-(phenylsulfinyl)-ethene (4a). Pale yel-
low oil. TLC Rf (EtOAc/petrol 1:1) 0.55; 1H NMR: d 7.5–7.3 (m, 5H,
3
H-200-600), 7.14 and 6.80 (AA0BB0 system, Jortho ¼ 9.2 Hz, 4H,
H-20,30,50,60), 6.18 and 5.84 (two d, 2Jgem ¼ 0.5 Hz, 2H, H2-2), 3.77 (s,
13
3H, CH3); C NMR: d 160.2 (C-40), 153.8 (C-1), 142.9 (C-100), 131.0
(C-400), 128.9 and 125.1 (C-20,60,200,300,500,600), 125.9 (C-10)þ, 115.2
(C-2), 114.0 (C-30,50), 55.2 (CH3); MS: m/z (%) 258 (15) [M ], 133
(10), 132 (100), 76 (7). Element. anal. Calcd. (%) for C15H14O2S
(258.34): C, 69.74; H, 5.46. Found: C, 69,68; H, 5.47.
1-(4-Methylphenyl)-1-(phenylsulfinyl)-ethene (4b).[63] TLC
Rf (EtOAc/petrol 1:1) 0.67; 1H NMR: d 7.4–7.3 (m, 5H, H-200-600), 7.1
(m, 4H, H-20,30,50,60), 6.21 and 5.87 (two s, 2H, H2-2), 2.30 (s, 3H,
CH3); 13C NMR: d 154.2 (C-1), 142.9 (C-100), 139.1 (C-40), 131.1 (C-400),
130.7 (C-10), 129.3, 128.9, 127.4, and 125.2 (C-20,30,50,60,200,30,500,600),
115.6 (C-2), 21.2 (CH3); MS: m/z (%) 242 (2) [Mþ], 126 (6), 116 (100).
Element. anal. Calcd. (%) for C15H14OS (242.34): C, 74.34; H, 5.82.
Found: C, 74.37; H, 5.80.
1-Phenyl-1-(phenylsulfinyl)-ethene (4c).[64] TLC Rf (EtOAc/
petrol 1:1) 0.48; 1H NMR: d 7.5–7.3 (m, 10H, H-20-60,200-600), 6.25 and
5.91 (two d, 2Jgem ¼ 0.5 Hz, 2H, H2-2); 13C NMR: d 154.3 (C-1), 142.6
(C-100), 133.6 (C-10), 131.1 (C-400), 129.0, 128.9, 128.5, 127.5, and
J. Phys. Org. Chem. 2009, 22 1048–1057
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc

M. C. AVERSA ET AL.
125.1 (C-20-60,200,300,500,600), 116.2 (C-2); Element. anal. Calcd. (%) for
C14H12OS (228.31): C, 73.65; H, 5.30. Found: C, 73.60; H, 5.25.
1-(4-Nitrophenyl)-1-(phenylsulfinyl)-ethene (4d). Light yel-
[8] G. Winnewisser, F. Lewen, S. Thorwirth, M. Behnke, J. Hahn, J. Gauss, E.
Herbst, Chem. Eur. J. 2003, 9, 5501-L 5510.
[9] S. Nakamura, S. Kusuda, K. Kawamura, T. Toru, J. Org. Chem. 2002, 67,
640–647.
[10] J. W. Cubbage, Y. Guo, R. D. McCulla, W. S. Jenks, J. Org. Chem. 2001,
66, 8722–8736.
[11] S. Lacombe, M. Loudet, E. Banchereau, M. Simon, G. Pfister-Guillouzo,
J. Am. Chem. Soc. 1996, 118, 1131–1138.
[12] B. S. Jursic, THEOCHEM 1997, 389, 257–263.
´
[13] B. Balta, G. Monard, M. F. Ruiz-Lopez, M. Antoine, A. Gand, S.
Boschi-Muller, G. Branlant, J. Phys. Chem. A 2006, 110, 7628–7636.
`
[14] J. Poater, M. Sola, M. Duran, X. Fradera, J. Phys. Chem. A 2001, 105,
2052–2063.
[15] A. H. Otto, R. Steudel, Eur. J. Inorg. Chem. 2000, 617–624.
[16] D. R. Hogg, In The Chemistry of Sulphenic Acids and their Derivatives,
(Ed.: S. Patai), John Wiley & Sons, Chichester, 1990. pp 388–395.
[17] R. G. Pearson, J. Am. Chem. Soc. 1963, 85, 3533–3539.
[18] P. K. Chattaraj, H. Lee, R. G. Parr, J. Am. Chem. Soc. 1991, 113,
1855–1856.
1
low crystals, m.p. 122–124 8C. TLC Rf (EtOAc/petrol 1:1) 0.50; H
3
NMR: d 8.12 and 7.37 (AA’BB’ system, Jortho ¼ 8.8 Hz, 4H,
H-20,30,50,60), 7.4–7.3 (m, 5H, H-200-600), 6.40 and 6.05 (two d,
2Jgem ¼ 0.8 Hz, 2H, H2-2); 13C NMR: d 153.0 (C-1), 148.0 (C-40), 141.8
(C-100), 140.0 (C-10), 131.7 (C-400), 129.2, 128.5, 125.1, andþ123.7
(C-20,30,50,60,200,300,500,600), 119.8 (C-2); MS: m/z (%) 273 (4) [M ], 148
(26), 126 (100). Element. anal. Calcd. (%) for C14H11NO3S (273.31):
C, 61.52; H, 4.06. Found: C, 61.50; H, 4.10.
Methyl 2-(phenylsulfinyl)-2-propenoate (4e).[65] TLC Rf
1
(EtOAc/petrol 1:1) 0.55; H NMR: d 7.7–7.5 (m, 5H, H-200-600),
6.87 and 6.71 (two d, 2Jgem ¼ 0.5 Hz, 2H, H2-3), 3.72 (s, 3H, CH3); 13
C
NMR: d 162.2 (C-1), 147.3 (C-2), 143.0 (C-100), 131.7 (C-400), 129.2
and 125.9 (C-200,300,500,600), 128.5 (C-3), 52.3 (CH3); MS: m/z (%) 210
(100) [Mþ], 179 (14), 126 (31), 125 (99), 77(62). Element. anal.
Calcd. (%) for C10H10O3S (210.25): C, 57.13; H, 4.79. Found: C, 57.10;
H, 4.77.
[19] A. Cedillo, P. K. Chattaraj, R. G. Parr, Int. J. Quantum Chem. 2000, 77,
403–407.
[20] H. Adams, J. C. Anderson, R. Bell, D. N. Jones, M. R. Peel, N. C. O.
Tomkinson, J. Chem. Soc., Perkin Trans. 1998, 3967–3974.
[21] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[22] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[23] V. Barone, L. Orlandini, C. Adamo, Chem. Phys. Lett. 1994, 231,
295–300.
(E)-1-(4-Nitrophenyl)-2-(phenylsulfinyl)-ethene (5d).[66] It
was identified by comparison with published characterization
data.
Methyl (E)-3-(phenylsulfinyl)-2-propenoate (5e).[67] TLC Rf
[24] V. Barone, L. Orlandini, C. Adamo, Int. J. Quantum Chem. 1995, 56,
697–705.
1
(EtOAc/petrol 1:1) 0.52; H NMR: d 7.6–7.5 (m, 5H, H-200-600), 7.50
[25] B. G. Johnson, C. A. Gonzales, P. M. W. Gill, J. A. Pople, Chem. Phys. Lett.
1994, 221, 100–108.
[26] R. G. Parr, W. Yang, In Density Functional Theory of Atoms and
Molecules, Oxford University Press, Oxford, 1989.
[27] A. Dybala-Defratyka, P. Paneth, J. Pu, D. G. Truhlar, J. Phys. Chem. A
2004, 108, 2475–2486.
[28] Y. Zhao, D. G. Truhlar, J. Phys. Chem. A 2004, 108, 6908–6918.
[29] A. D. Boese, J. M. L. Martin, J. Chem. Phys. 2004, 121, 3405–3416.
[30] Y. Zhao, J. Pu, B. J. Lynch, D. G. Truhlar, Phys. Chem. Chem. Phys. 2004,
6, 673–676.
[31] Y. Zhao, D. G. Truhlar, Phys. Chem. Chem. Phys. 2005, 7, 2701–2705.
[32] J. Pu, D. G. Truhlar, J. Phys. Chem. A 2005, 109, 773–778.
[33] B. J. Lynch, Y. Zhao, D. G. Truhlar, J. Phys. Chem. A 2005, 109,
1643–1649.
(AB d, J2,3 ¼ 14.8 Hz, 1H, H-3), 6.75 (AB d, 1H, H-2), 3.78 (s, 3H, CH3);
13C NMR: d 164.3 (C-1), 151.3 (C-3), 141.4 (C-100), 131.8 (C-400), 129.7
and 124.7 (C-200,300,500,600), 123.6 (C-2), 52.3 (CH3); MS: m/z (%) 210
(20) [Mþ], 162 (100), 131 (64), 109 (46), 77 (31). Element. anal.
Calcd. (%) for C10H10O3S (210.25): C, 57.13; H, 4.79. Found: C, 57.15;
H, 4.80.
(E)-1-(Trimethylsilyl)-2-(phenylsulfinyl)-ethene (5f). TLC Rf
(EtOAc/petrol 1:1) 0.70; 1H NMR: d 7.6–7.5 (m, 5H, H-2000;-6000;), 6.96
(AB d, J1,2 ¼ 17.5 Hz, 1H, H-2), 6.63 (AB d, 1H, H-1), 0.13 (s, 9H, CH3);
13C NMR: d 146.7 (C-2), 143.4 (C-100), 136.3 (C-1), 131.0 (C-400), 129.4
and 124.7 (C-200,300,500,600), - 1.7 (CH3); Element. anal. Calcd. (%) for
C11H16OSSi (224.39): C, 58.88; H, 7.19. Found: C, 58.91; H, 7.17.
´
[34] Y. Zhao, N. Gonzalez-Garc´ıa, D. G. Truhlar, J. Phys. Chem. A 2005, 109,
2012–2018.
[35] E. Juaristi, R. Notario, M. V. Roux, Chem. Soc. Rev. 2005, 34, 347–354.
´
[36] M. V. Roux, M. Temprado, P. Jimenez, J. Z. Davalos, R. Notario, G.
Mart´ın-Valcarel, L. Garrido, R. Guzman-Mej´ıa, E. Juaristi, J. Org. Chem.
´
´
´
Acknowledgements
2004, 69, 5454–5459.
[37] H. Chermette, J. Comput. Chem. 1999, 20, 129–154.
[38] P. Geerlings, F. De Proft, W. Langenaeker, Chem. Rev. 2003, 103,
1793–1874.
[39] P. W. Ayers, J. S. M. Anderson, L. J. Bartolotti, Int. J. Quantum Chem.
2005, 101, 520–534.
`
The authors thank the ‘‘Ministero dell’Istruzione, Universita e
Ricerca’’ (MIUR) for financial support and the ‘‘Centro Interuni-
versitario Lombardo per l’Elaborazione Automatica’’ (CILEA) for
computational facilities.
[40] R. G. Parr, W. Yang, J. Am. Chem. Soc. 1984, 106, 4049–4050.
[41] W. Yang, W. J. Mortier, J. Am. Chem. Soc. 1986, 108, 5708–5711.
´
´
[42] J. L. Gazquez, F. Mendez, J. Phys. Chem. 1994, 98, 4591–4593.
[43] D. Sengupta, A. K. Chandra, M. T. Nguyen, J. Org. Chem. 1997, 62,
6404–6406.
REFERENCES
´
[44] S. Damoun, G. Van de Woude, F. Mendez, P. Geerlings, J. Phys. Chem. A
1997, 101, 886–893.
[1] L. A. G. M. Van Den Broek, L. P. C. Delbressine, H. C. J. Ottenheijm, In:
The Chemistry of Sulphenic Acids and their Derivatives, (Ed.: S. Patai),
John Wiley & Sons, Chichester, 1990. pp 701–721.
[2] M. C. Aversa, A. Barattucci, P. Bonaccorsi, P. Giannetto, D. N. Jones, J.
Org. Chem. 1997, 62, 4376–4384.
[3] M. C. Aversa, A. Barattucci, P. Bonaccorsi, P. Giannetto, M. Policicchio, J.
Org. Chem. 2001, 66, 4845–4851.
[4] V. Aucagne, M. C. Aversa, A. Barattucci, P. Bonaccorsi, P. Giannetto, P.
[45] A. Ponti, J. Phys. Chem. A 2000, 104, 8843–8846.
[46] A. Ponti, G. Molteni, J. Org. Chem. 2001, 66, 5252–5255.
[47] G. Molteni, A. Ponti, Tetrahedron 2003, 59, 5225–5229.
[48] G. Molteni, A. Ponti, Chem. Eur. J. 2003, 9, 2770–2774.
[49] A. Ponti, G. Molteni, Chem. Eur. J. 2006, 12, 1156–1161.
[50] G. Molteni, A. Ponti, ARKIVOC (Gainesville, FL) 2006, 49–56.
[51] J. P. Foster, F. Weinhold, J. Am. Chem. Soc. 1980, 102, 7211–7228.
[52] A. E. Reed, F. Weinhold, J. Chem. Phys. 1983, 78, 4066–4073.
[53] A. E. Reed, F. Weinhold, J. Chem. Phys. 1985, 83, 1736–1740.
[54] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899–
926.
¨
Rollin, A. Tatibouet, J. Org. Chem. 2002, 67, 6925–6930.
[5] M. C. Aversa, A. Barattucci, P. Bonaccorsi, P. Giannetto, J. Org. Chem.
2005, 70, 1986–1992.
[6] M. C. Aversa, A. Barattucci, M. C. Bilardo, P. Bonaccorsi, P. Giannetto, P.
¨
Rollin, A. Tatibouet, J. Org. Chem. 2005, 70, 7389–7396.
[7] H. Beckers, S. Esser, T. Metzroth, M. Behnke, H. Willner, J. Gauss, J.
Hahn, Chem. Eur. J. 2006, 12, 832–844.
[55] J. A. Dobado, H. Mart´ınez-Garc´ıa, J. Molina Molina, M. R. Sundberg, J.
Am. Chem. Soc. 1999, 121, 3156–3164.
[56] G. R. Desiraju, Chem. Commun. (Cambridge, UK) 2005, 2995–3001.
www.interscience.wiley.com/journal/poc
Copyright ß 2009 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2009, 22 1048–1057

SULFENIC ACIDS WITH MONOSUBSTITUTED ACETYLENES
[57] M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem. Phys. 2002, 117,
43–54.
[58] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24,
669–681.
Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision
C.02, Gaussian, Inc., Wallingford CT, 2004.
[60] Ed.: R. C. Weast, Handbook of Chemistry and Physics, CRC Press,
Cleveland, OH, 1981.
[61] A. Bondi, J. Phys. Chem. 1964, 68, 441–451.
[62] B. H. Besler, K. M. Merz, Jr., P. A. Kollman, J. Comput. Chem. 1990, 11,
431–439.
[59] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant,
J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, 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, P. Y. Ayala, K.
Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G.
Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C.
[63] X. Creary, M. E. Mehrsheikh-Mohammadi, M. D. Eggers, J. Am. Chem.
Soc. 1987, 109, 2435–2442.
[64] M. Kawakita, K. Yokota, H. Akamatsu, S. Irisawa, O. Morikawa, H.
Konishi, K. Kobayashi, J. Org. Chem. 1997, 62, 8015–8017.
[65] F. Leyendecker, M. T. Comte, Tetrahedron 1986, 42, 1413–1421.
[66] J. H. van Steenis, J. J. G. S. van Es, A. van der Gen, Eur J. Org. Chem.
2000, 2787–2793.
[67] F. A. Davis, A. J. Friedman, U. K. Nadir, J. Am. Chem. Soc. 1978, 100,
2844–2852.
J. Phys. Org. Chem. 2009, 22 1048–1057
Copyright ß 2009 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/poc