Desulfinylation of prop-2-enesulfinic acid: Experimental results and mechanistic theoretical analysis

Article abstract of DOI:10.1021/ja901565s

The potential energy surfaces of the desulfinylation of prop-2-enesulfinic acid (13) in CH₂Cl₂ solution at -15°C have been explored by quantum calculations and analyzed with kinetic data obtained for the reaction in absence or presence of additives. Monomeric 13 adopts a preferred conformation with gauche S=O/σ(C(1)-C(2) bond pairs and the O-H bond pointing toward C(3). It equilibrates with the more stable dimer (13) ₂ (at -15°C) formed by two O-H...O=S hydrogen bonds and in which the S=O/σC(1)-C(2) are gauche also, but the SOH moieties are antiperiplanar with respect to σ(C(1)-C(2)). Dimer (13)₂ undergoes desulfinylation into propene + SO₂ + 13 following a one-step, concerted mechanism. The preferred transition state is a six-membered, chairlike transition structure (C...S elongation and S- O...H...C(3) hydrogen transfer occur in concert) in which the SdO/σ(C(1)-C(2)) bonds are gauche (S=O adopt pseudoaxial positions). There are at least 48 transition states, each one defining a different pathway, all with similar calculated free energies (?G^? = 25.3-28.6 kcal/mol), which makes the bimolecular (autocatalyzed) retro-ene elimination of SO₂ competing (entropy factor) with a monomolecular process for which the transition state (calculated ΔG^? = 24.3 kcal/mol) implies only one molecule of sulfinic acid. This agrees with the experimental rate law of the reaction which is first order in the concentration of dimer (13)₂. SO₂, CF₃COOH, and BF₃· Me₂O do not catalyze the reaction. In the presence of an excess of BF₃·Me₂O the desulfinylation is completely inhibited due to the formation of a stable tetramolecular complex of type (CH₂=CHCH₂SO₂H·BF₃) ₂ (18), for which quantum calculations show that the S=O/σ(C(1)-C(2)) bonds are antiperiplanar whereas the S-OH/σ(C(1)- C(2)) bonds are gauche. Independently of the additive, the retro-ene eliminations of SO₂ are calculated to be concerted and have transition states adopting six-membered cyclic structures in which S=O and σ(C(1)-C(2)) are gauche, the S=O interacting with the additive. Preliminary experiments suggested that the thermodynamically unfavored ene reaction of SO₂ with propene can occur at low temperature using 1 equiv of BF₃.

Full text of DOI:10.1021/ja901565s

Published on Web 06/19/2009
Desulfinylation of Prop-2-enesulfinic Acid: Experimental
Results and Mechanistic Theoretical Analysis
†
‡
,‡
,†
´
´
Adria´n Varela-Alvarez, Dean Markovic´, Pierre Vogel,* and Jose´ Angel Sordo*
Laboratorio de Qu´ımica Computacional, Departamento de Qu´ımica F´ısica y Anal´ıtica,
UniVersidad de OViedo, Principado de Asturias, Spain, and Laboratoire de glycochimie et de
synthe`se asyme´trique (LGSA), Swiss Federal Institute of Technology of Lausanne (EPFL),
CH 1015 Lausanne, Switzerland
´
Received February 28, 2009; E-mail: jasg@uniovi.es (J.A.S.); pierre.vogel@epfl.ch (P.V.)
Abstract: The potential energy surfaces of the desulfinylation of prop-2-enesulfinic acid (13) in CH2Cl2
solution at -15 °C have been explored by quantum calculations and analyzed with kinetic data obtained
for the reaction in absence or presence of additives. Monomeric 13 adopts a preferred conformation with
gauche SdO/σ(C(1)-C(2) bond pairs and the O-H bond pointing toward C(3). It equilibrates with the
more stable dimer (13)2 (at -15 °C) formed by two O-H· · ·OdS hydrogen bonds and in which the SdO/
σC(1)-C(2) are gauche also, but the SOH moieties are antiperiplanar with respect to σ(C(1)-C(2)). Dimer
(13)2 undergoes desulfinylation into propene + SO2 + 13 following a one-step, concerted mechanism. The
preferred transition state is a six-membered, chairlike transition structure (C · · · S elongation and
S-O· · ·H· · ·C(3) hydrogen transfer occur in concert) in which the SdO/σ(C(1)-C(2)) bonds are gauche
(SdO adopt pseudoaxial positions). There are at least 48 transition states, each one defining a different
pathway, all with similar calculated free energies (∆G‡ ) 25.3-28.6 kcal/mol), which makes the bimolecular
(autocatalyzed) retro-ene elimination of SO2 competing (entropy factor) with a monomolecular process for
which the transition state (calculated ∆G‡ ) 24.3 kcal/mol) implies only one molecule of sulfinic acid. This
agrees with the experimental rate law of the reaction which is first order in the concentration of dimer (13)2.
SO2, CF3COOH, and BF3 ·Me2O do not catalyze the reaction. In the presence of an excess of BF3 ·Me2O
the desulfinylation is completely inhibited due to the formation of a stable tetramolecular complex of type
(CH2dCHCH2SO2H·BF3)2 (18), for which quantum calculations show that the SdO/σ(C(1)-C(2)) bonds
are antiperiplanar whereas the S-OH/σ(C(1)-C(2)) bonds are gauche. Independently of the additive, the
retro-ene eliminations of SO2 are calculated to be concerted and have transition states adopting
six-membered cyclic structures in which SdO and σ(C(1)-C(2)) are gauche, the SdO interacting with the
additive. Preliminary experiments suggested that the thermodynamically unfavored ene reaction of SO2
with propene can occur at low temperature using 1 equiv of BF3.
Scheme 1
Introduction
The thermal desulfinylation of ꢀ,γ-unsaturated sulfinic acids
is a useful reaction for the regio- and stereoselective synthesis
of alkenes (Scheme 1).1 In the case of R-substituted sulfinic
acids 1, a concerted retro-ene reaction (formally a retro
hetero-Alder2,3 ene reaction) that leads to the elimination of SO2
is assumed to be responsible for the chirality transfer from the
R-carbon center to the γ-carbon center with formation of alkenes
3. The stereoselectivity of the reaction is explained in terms of
a chairlike transition state 24 that places an optimal number of
substituents in pseudoequatorial positions.5-11 Kinetic investiga-
tions by Young and co-workers4,12 on the desulfinylation of
prop-2-enesulfinic acid in toluene were taken as consistent with
a cyclic transition state (k297K ) 5.5 ( 0.1) × 10-4 s-1, ∆S‡ )
-35 ( 4.5 e.u.; ∆V‡ ) -5.5 ( 1.0 cm3/mol, ∆Vr ) 15 ( 5
cm3/mol).
A relatively small deuterium kinetic isotope effect (kH/kD )
† University of Oviedo.
2.5 ( 0.1) for the H/D transfer also supported the concerted
‡ Swiss Federal Institute of Technology of Lausanne.
(1) Braverman, S. In The chemistry of sulfinic acids, esters and their
deriVatiVes; Patai, S., Ed.; Wiley: Chichester, UK, 1990; pp 298-
303.
(6) Garigipati, R. S.; Morton, J. A.; Weinreb, S. M. Tetrahedron Lett.
1983, 24, 987–990.
(2) Alder, K.; Pascher, F.; Schmitz, A. Chem. Ber. 1943, 76, 27–53.
(3) Hoffmann, H. M. R. Angew. Chem., Int. Ed. Engl. 1969, 8, 556–
557.
(7) Corey, E. J.; Engler, T. A. Tetrahedron Lett. 1984, 25, 149–152.
(8) Baudin, J. B.; Julia, S. A. Tetrahedron Lett. 1988, 29, 3255–3258.
(9) Baudin, J. B.; Julia, S. Bull. Soc. Chim. Fr. 1995, 132, 196–214.
(10) Chochrek, P.; Kurek-Tyrlik, A.; Wicha, J. Pol. J. Chem. 2006, 80,
679–683.
(4) Hiscock, S. D.; Isaacs, N. S.; King, M. D.; Sue, R. E.; White, R. H.;
Young, D. J. J. Org. Chem. 1995, 60, 7166–7169.
(5) Mock, W. L.; Nugent, R. M. J. Org. Chem. 1978, 43, 3433–3434.
(11) Chochrek, P.; Wicha, J. Eur. J. Org. Chem. 2007, 253, 4–2542.
9
10.1021/ja901565s CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 9547–9561 9547

´
Varela-Alvarez et al.
A R T I C L E S
Table 1. Examples of Hetero Ene and Retro-ene Reactions
Scheme 2. Possible Mechanism of Noncatalyzed
(Estimated Standard Heat of Reactions (25 °C, 1 atm, gas phase),
Hetero-Ene-Reaction (Or Retro-Ene Reaction)a
∆H°(in kcal/mol), for Propene (RHdCH3CHdCH2) Reacting with
r
Heteroenophiles)
a (A) Concerted with six-center transition state: H-transfer is concerted
with C-Y bond formation and allylic rearrangement (demonstrated for X
) Y: ArSO2NdSdO, SO2, SeO2, CO2, N2, H2CdO, H2CdCH). (B)
Nonconcerted involves formation of reactive intermediates in which the
H-transfer might not be the rate determining step (observed for X ) Y:
1O2, R′CONdNCOR′, TsNdCHCCl3, ArNdO, SdO, SO3).
ducing corresponding-1-phenylalkane-1-sulfinic acid16 (gain in
aromaticity). At this stage it is interesting to compare the ene-
reaction of SO2 with other related heteroene-reactions (N-
sulfinylbenzenesulfonamides and N-sulfinyl-nonafluorobutane-
sulfonamide under go easy ene-reactions with alkenes, see refs
17-22).
Although the ene-reaction of sulfur monoxide (SdO)23,24 is
estimated to have an exothermicity of ca. -28.7 kcal/mol (based
on ∆H°(SdO, gas) ) 1.2 kcal/mol, ∆H°(MeSOH, gas) ) -45.4
f
f
kcal/mol, ∆H°(CH4) ) -17.9 kcal/mol,14 it has never been
f
reported. Sulfur monoxide prefers to undergo cheletropic [2 +
1]-addition with alkenes25 giving the corresponding thiirane
1-oxides for which no rearrangement into the corresponding
product of ene-reaction has been reported yet. This is surprising
when comparing with the ene-reaction of singlet oxygen26-35
and related ene-reactions of diazene derivatives36-40 and nitroso
a Follows a concerted mechanism with a six-center transition state for
the thermal, non catalyzed reaction (Scheme 2A). b Several step process
that can imply different type of intermediates (Scheme 2B). c Not
observed yet. d Few examples, mechanism unknown. e See decar-
bonylation of 2,2-dimethyl-3-butenal.65 f Using standard heat of
formation (25 °C, 1 atm. gas phase) from ref 14. g See text. h For
electron-poor derivatives.22 i For electron-poor arylNO.22 j See ref 50.
k For 1O2, using ∆H°(1O2)
- )
∆H°(3O2) -22.5 kcal/mol l See
f
TsNdCHCCl3.22
(16) Raasch, M. S.; Smart, B. E. J. Am. Chem. Soc. 1979, 101, 7733–
7734.
(17) Deleris, G.; Kowalski, J.; Dunogues, J.; Calas, R. Tetrahedron Lett.
1977, 4211–4214.
mechanism with a relatively compact, early transition state. Alk-
2-ene sulfinic acids are generally unstable1 as they undergo
exergonic desulfinylation at room temperature. The difference
in standard heats of formation in the gas phase between sulfinic
acids (RSO2H) and the corresponding hydrocarbons (RH) is
(18) Kresze, G.; Bussas, R. Liebigs Ann. Chem. 1980, 843–857.
(19) Bussas, R.; Kresze, G. Angew. Chem. 1980, 92, 748–749.
(20) Kresze, G.; Bussas, R. Angew. Chem. 1980, 92, 750–751.
(21) Deleris, G.; Dunogues, J.; Gadras, A. Tetrahedron Lett. 1984, 25,
2135–2138.
estimated to be -74.5 kcal/mol.13 Considering ∆H°(SO2, gas)
(22) Starflinger, W.; Kresze, G.; Huss, K. J. Org. Chem. 1986, 51, 37–
40.
f
) -70.9 kcal/mol,14 the ene reaction of SO2 with propene has
an exothermicity (standard heat of SO2 hydrocarbation) of -74.5
+ 70.9 ) -3.6 kcal/mol in the gas phase (Table 1), what is
insufficient to compensate for the cost of entropy of condensa-
tion at temperatures higher than -100 °C. Products of ene-
reaction have been reported for SO2 reacting with allenes giving
the corresponding 1,3-butadiene-2-sulfinic acids15 (exothermicity
of ca. -14 kcal/mol is estimated in this case) and for SO2
reacting with 6-alkylidenecyclohexa-1,3-diene derivatives pro-
(23) Schenk, P. W. Z. Anorg. Allg. Chem. 1937, 233, 385–400.
(24) Schenk, P. W. Z. Anorg. Allg. Chem. 1952, 270, 301–303.
(25) Nakayama, J.; Tajima, Y.; Xue-Hua, P.; Sugihara, Y. J. Am. Chem.
Soc. 2007, 129, 7250–7251.
(26) Gollnick, K.; Kuhn, H. J. Org. Chem. 1979, 40, 287–427.
(27) Stephenson, L. M.; Grdina, M. J.; Orfanopoulos, M. Acc. Chem. Res.
1980, 13, 419–425.
(28) Davies, A. G.; Schiesser, C. H. Tetrahedron 1991, 47, 1707–1726.
(29) Singleton, D. A.; Hang, C.; Szymanski, M. J.; Meyer, M. P.; Leach,
A. G.; Kuwata, K. T.; Chen, J. S.; Greer, A.; Foote, C. S.; Houk,
K. N. J. Am. Chem. Soc. 2003, 125, 1319–1328.
(30) Maranzana, A.; Ghigo, G.; Tonachini, G. Chem.sEur. J. 2003, 9,
2616–2626.
(31) Griesbeck, A. G.; El-Idreesy, T. T.; Adam, W.; rebs, O. CRC
Handbook of Organic Photochemistry and Photobiology; CRC Press
LLC: Boca Raton, FL, 2004; Vol. 8.
(32) Orfanopoulos, M.; Vougioukalakis, G. C.; Stratakis, M. Chemistry
of Peroxides; John Wiley and Sons, Ltd: Chichester, UK, 2006; Vol.
2. (Pt 2).
(12) King, M. D.; Sue, R. E.; White, R. H.; Young, D. J. J. Chem. Soc.
Chem. Commun. 1993, 1797–1798.
(13) Liebman, J. F.; Crawford, K. S. K.; Slayden, S. W. Supplement S:
The Chemistry of Sulphur-Containing Functional Groups; John Wiley
and Sons, Ltd: New York, 1993.
(14) National Institute of Standards and Technology Chemistry WebBook,
NT Standard Reference Data-base Number 69; (http://webbook.nist-
.gov): Gaithersburg, MD, 2000.
(15) Capozzi, G.; Lucchini, V.; Marcuzzi, F.; Melloni, G. Tetrahedron
Lett. 1980, 21, 3289–3292.
(33) Alberti, M. N.; Orfanopoulos, M. Org. Lett. 2008, 10, 2465–2468.
(34) Leach, A. G.; Houk, K. N.; Foote, C. S. J. Org. Chem. 2008, 73,
8511–8519.
(35) Sheppard, A. N.; Acevedo, O. J. Am. Chem. Soc. 2009, 131, 2530–
2540.
9
9548 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
compounds.41-47 Exothermicities of ca. -13.4 and -15.6 kcal/
mol are estimated for the ene-reactions of triplet dioxygen (based
on ∆H°(MeOOH) ) -31.3 kcal/mol, ∆H°(CH4) ) -17.9 kcal/
reaction follows a concerted process62 similar to that proposed
for 3 f 2 f 1. After complex formation between SeO2 and
the alkene, direct transfer of the allylic C-H to the OdSe
moiety occurs in concert with the C-Se bond formation and
the allylic rearrangement. Intermediates such as 1,4-diradical
MeC•H-CH2-Se(O)-O• T zwitterion MeC+HCH2Se(O)O-,
selenirane 1,1-dioxide (analogous to the [1 + 2]-cheletropic
addition of SdO) or 1,2-oxaselenetane 2-oxide (analogous to
ꢀ-sultones) formed by [2 + 2]-cycloaddition of SeO2 to alkenes
have not been detected63,64 (Scheme 2).
f
f
mol14) and diazene (based on ∆H°(HNdNH) ) 50.9 kcal/mol;
f
∆H°(propane) ) -25.0 kcal/mol, ∆H°(CH3CH2CH2NHNH2)
f
f
) +11.4 kcal/mol14). For the estimate of the heat of ene-reaction
of nitroso compounds giving hydroxylamines derivatives (ca.
-28 kcal/mol), we have considered the standard heat of
hydrogenation ∆H°(H-NdO + H2 f H2NOH) ) -35.7
r
kcal/mol48,49 and compared it with the standard heats of
hydrogenation ∆H°(CH2O + H2 f MeOH) ) -21.3 kcal/mol
r
Although the thermal decarboxylation of but-3-enoic acid into
CO2 and propene is exothermic by ca. +5.3 kcal/mol (based
on ∆H°(propane) ) -25.0 kcal/mol, ∆H°(CO2) ) -94.05 kcal/
and ∆H°(CH2dNH + H2 f MeNH2) ) -21.6 kcal/mol which
r
are about -8 kcal/mol more exothermic than the corresponding
f
f
heats of hydrocarbation. Thus, adding to ∆H°(HNO + H2 f
r
mol, ∆H°(butanoic acid) ) -113.7 kcal/mol14), and that it is a
f
H2NOH) this difference of +8 kcal/mol, we estimate the heat
of hydrocarbation of a nitroso compound to amount to ca. -28
kcal/mol (Table 1). In the case of the hydrocarbation of N2 one
estimates a endothermicity of ca. +64.6 kcal/mol (based on
∆H°(MeCH2-CH2-NdNH) ) ∆H°(MeCH2CH2NHNH2)sheat
reaction more exergonic than the desulfinylation of prop-2-ene-
1-sulfinic acid, it requires much higher temperatures to occur
(>300 °C)66 than the latter reaction that occurs below 20 °C
already in solution. Substituent effects on the activation
enthalpy,67-71 kinetic isotopic effects,72 as well as quantum
calculations73-77 support a concerted, six-membered mechanism
in which the carboxylic hydrogen atom is transferred to C(4)
of the but-3-enoic acid in concert with the σ(C(2)-C(1)) bond
breaking (∆H‡ ) 38 ( 1.6 kcal/mol, ∆S‡ ) -10.2 ( 2.5 e.u.
at 377 °C). An analogous reaction is the thermolysis of
homoallylic alcohol which fragmentates into propene and
formaldehyde above 300 °C (∆H‡ = 40 kcal/mol, ∆S‡ ) -8.8
e.u.78,79 see also the retro-ene reaction of propargyl alcohols80)
follow a similar concerted mechanism. Thermolysis (287-375
°C) of allyl ethyl ether giving CH3CHO + propene might also
follow a similar concerted retro-ene mechanism,81,82 as well as
thermolysis of allyl methyl amine giving propene + CH2dNH
(330-420 °C)83 (Scheme 2A). Retro-ene elimination of thio-
aldehydes or thioketones from ꢀ,γ-unsaturated thiols is not
f
f
of hydrogenation of azene of -28.1 kcal/mol), and thus, only
the retro-ene elimination of N2 can be observed.50-52 The ene-
reaction of sulfur trioxide (SO3) is also estimated to be
exothermic by ca. -31.4 kcal/mol based on a heat of hydro-
carbation of SO3 of -126 kcal/mol13 and ∆H°(SO3, gas) )
f
-94 6 kcal/mol. SO3 adds to alkenes with high suprafaciality
at -60 °C giving the corresponding ꢀ-sulfones53,54 that can be
rearranged into the corresponding ꢀ,γ-unsaturated sulfonic
acids.55 Related to the ene-reaction of SO2 is the allylic C-H
56-58
oxidation with SeO2
which occur through an initial ene
reaction followed by a [2,3]-sigmatropic shift.59-61 The ene-
(36) Brimble, M. A.; Heathcock, C. H. J. Org. Chem. 1993, 58, 5261–
5263.
(37) Alberti, M. N.; Vougioukalakis, G. C.; Orfanopoulos, M. Tetrahedron
Lett. 2003, 44, 903–905.
(38) Vougioukalakis, G. C.; Roubelakis, M. M.; Alberti, M. N.; Orfan-
opoulos, M. Chem.sEur. J. 2008, 14, 9697–9705.
(39) Roubelakis, M. M.; Vougioukalakis, G. C.; Angelis, Y. S.; Orfan-
opoulos, M. Org. Lett. 2006, 8, 39–42.
(60) Arigoni, D.; Vasella, A.; Sharpless, K. B.; Jensen, H. P. J. Am. Chem.
Soc. 1973, 95, 7917–7919.
(61) Woggon, W. D.; Ruther, F.; Egli, H. J. Chem. Soc. Chem. Commun.
1980, 706–708.
(40) Lu, X. Org. Lett. 2004, 6, 2813–2815.
(41) Adam, W.; Bottke, N.; Krebs, O. J. Am. Chem. Soc. 2000, 122, 6791–
6792.
(62) Singleton, D. A.; Hang, C. J. Org. Chem. 2000, 65, 7554–7560.
(63) Ra, C. S.; Park, G. Tetrahedron Lett. 2003, 44, 1099–1102.
(64) Ra, C. S.; Park, E. M.; Park, G. Bull. Korean Chem. Soc. 2008, 29,
2513–2516.
(42) Adam, W.; Bottke, N.; Engels, B.; Krebs, O. J. Am. Chem. Soc. 2001,
123, 5542–5548.
(43) Leach, A. G.; Houk, K. N. Chem. Commun. 2002, 1243–1255.
(44) Adam, W.; Bottke, N.; Krebs, O.; Lykakis, I.; Orfanopoulos, M.;
Stratakis, M. J. Am. Chem. Soc. 2002, 124, 14403–14409.
(45) Quadrelli, P.; Romano, S.; Piccanello, A.; Caramella, P. J. Org. Chem.
2009, 74, 2301–2310.
(65) Crawford, R. J.; Lutener, S.; Tokunaga, H. Can. J. Chem. 1977, 55,
3951–3954.
(66) Smith, G. G.; Blau, S. E. J. Phys. Chem. 1964, 68, 1231–1234.
(67) Bigley, D. B. J. Chem. Soc. 1964, 3897–3899.
(68) Noyce, D. S.; Heller, R. A. J. Am. Chem. Soc. 1965, 87, 4325–4328.
(69) Bigley, D. B.; Thurman, J. C. J. Chem. Soc. 1965, 6202–6205.
(70) Bigley, D. B.; Thurman, J. C. J. Chem. Soc. B: Phys. Org. 1966,
1076–1077.
(46) Adam, W.; Krebs, O.; Orfanopoulos, M.; Stratakis, M.; Vou-
gioukalakis, G. C. J. Org. Chem. 2003, 68, 2420–2425.
(47) Fokin, A. A.; Yurchenko, A. G.; Rodionov, V. N.; Gunchenko, P. A.;
Yurchenko, R. I.; Reichenberg, A.; Wiesner, J.; Hintz, M.; Jomaa,
H.; Schreiner, P. R. Org. Lett. 2007, 9, 4379–4382.
(48) Anderson, W. R. Combust. Flame 1999, 117, 394–403.
(49) Feller, D.; Dixon, D. A. J. Phys. Chem. A 2003, 107, 10419–10427.
(50) Jabbari, A.; Sorensen, E. J.; Houk, K. N. Org. Lett. 2006, 8, 3105–
3107.
(71) Bigley, D. B.; May, R. W. J. Chem. Soc. B: Phys. Org. 1967, 557–
560.
(72) Bigley, D. B.; Thurman, J. C. J. Chem. Soc. B: Phys. Org. 1967,
941–943.
(73) Dewar, M. J. S.; Ford, G. P. J. Am. Chem. Soc. 1977, 99, 8343–
8344.
(51) Ripoll, J. L.; Vallee, Y. Synthesis 1993, 659–677.
(52) Mikami, K.; Shimizu, M. Chem. ReV. 1992, 92, 1021–1050.
(53) Nagayama, M.; Okumura, O.; Noda, S.; Mori, A. J. Chem. Soc. Chem.
Commun. 1973, 841.
(74) Lee, I.; Cho, J. K. Bull. Korean Chem. Soc. 1984, 5, 51–52.
(75) Lee, I.; Cho, J. K.; Lee, B. S. J. Comput. Chem. 1984, 5, 217–224.
(76) Wang, Y. L.; Poirier, R. A. Can. J. Chem. 1994, 72, 1338–1346.
(77) Dunn, M. E.; Shields, G. C.; Takahashi, K.; Skodje, R. T.; Vaida,
V. J. Phys. Chem. A 2008, 112, 10226–10235.
(54) Morley, J. O.; Roberts, D. W.; Watson, S. P. J. Chem. Soc. Perkin
Trans. 2 1999, 1819–1826.
(78) Smith, G. G.; Yates, B. L. J. Chem. Soc. 1965, 7242–7246.
(79) Inaba, T.; Sakamoto, M.; Watanabe, S.; Takahashi, I.; Fujita, T.
J. Chem. Technol. Biotechnol. 1990, 48, 483–492.
(80) Hopf, H.; Kirsch, R. Angew. Chem. 1985, 24, 796–797.
(81) Egger, K. W.; Vitins, P. Int. J. Chem. Kinet. 1974, 6, 429–435.
(82) Vitins, P.; Egger, K. W. J. Chem. Soc. Perkin Trans. 2 1974, 1292–
1293.
(55) Cerfontain, H.; Kramer, J. B.; Schonk, R. M.; Bakker, B. H. J. R.
Netherlands Chem. Soc. 1995, 114, 410–420.
(56) Dupont, G.; Zacharewicz, W. Bull. Soc. Chim. Fr. 1935, 2, 533–
539.
(57) Guillemonat, A. Ann. Chem. Fr. 1939, 11, 143–211.
(58) Wiberg, K. B.; Nielsen, S. D. J. Org. Chem. 1964, 29, 3353–3361.
(59) Sharples, K. B.; Lauer, R. F. J. Am. Chem. Soc. 1972, 94, 7154–
7155.
(83) Vitins, P.; Egger, K. W. J. Chem. Soc. Perkin Trans. 2 1974, 1289–
1291.
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9549

´
Varela-Alvarez et al.
A R T I C L E S
Scheme 3. One-Pot Synthesis of Stereotriads via SO2-Umpolung
described, but examples of ene-reaction of thiocarbonyl com-
pounds are known.84,85
lation of ꢀ,γ-unsaturated sulfinic acid in more detail with the
goal to find suitable reaction conditions under which undesired
side-reactions such as R*OH elimination and retro-aldol cleav-
age would be suppressed. An alternative route is the Pd-
catalyzed desulfinylation of silyl ꢀ,γ-unsaturated sulfinates.101
As stated above, ꢀ,γ-unsaturated sulfinic acids are unstable
at room temperature. They are formed as intermediates in the
hydrolysis of sulfinamides,5,6,8-11 the reduction of ꢀ-ketosul-
fones,86 and the oxidation of allylic thiols.87 They have been
proposed as intermediates in SO2-mediated alkene isomer-
ization.1,16,88-90 As we have shown, this might not always be
true.91 In the case of 1,1-dialkylethenes and trialkyl substituted
alkenes, SO2 generates first polysulfone polymers with the
alkenes. The latter are responsible for hydrogen atom abstraction
from the alkenes and generate allyl radical intermediates that
finally provide the most stable alkene isomers.92
As we had found that SO2 catalyzes its hetero-Diels-Alder
and cheletropic additions,102-104 we hoped that it would also
catalyze the desulfinylation of ꢀ,γ-unsaturated sulfinic acid under
less acidic conditions than protic or Lewis acid. We thus have
investigated the kinetics of the desulfinylation of prop-2-
enesulfinic acid (13) under various conditions. As the reaction
in toluene is quick at room temperature,4,12 we decided to work
1
at -15 °C in CD2Cl2, using H NMR to follow the reactions.
In 1997, our group uncovered a new C-C bond-forming
reaction cascade that condenses butadien-1-yl ethers 4, enox-
ysilanes 5 and SO2 to generate ꢀ,γ-unsaturated silyl sulfinates
6. After desilylation, the corresponding sulfinic acid 7 undergo
SO2 elimination with formation of (E)-alkenes 8 with high
stereoselectivity.93 The reaction cascade has been applied to
several diene(4)/enoxysilane(5) pairs,94-97 as well as to diene(9)/
allylsilane(10) pairs98 (Scheme 3), thus giving us the possibility
to construct, in one-pot operations, stereotriads 8 and 12 that
can be used directly in the synthesis of long-chain polyketides
and polypropionates.99,100 In some cases, the desulfinylations,
which require acidic conditions, are accompanied by decom-
position of the desired polyfunctional products. To avoid this,
we have been forced to study the mechanism of the desulfiny-
As we shall see, and in contrast to the studies reported by Young
and co-workers,4 the rate laws are more complicated than first
order in sulfinic acid. We also find that CF3COOH does not
catalyze the reactions much better than sulfinic 13 acid itself.
To our surprise SO2 has no significant effect on the rate of the
desulfinylation of 13. At low concentration, BF3 ·Et2O has no
significant effect, whereas at high concentration, it inhibits the
reaction, in contrast with the efficient catalytical effect of
BF3 ·Et2O,105-108 on related carbonyl-ene reactions.3,109-117
In parallel with these kinetics, we have carried out extensive
quantum calculations to approach a better understanding of our
observations and to delimit the possible mechanisms of the retro-
ene reaction (Schemes 1 and 2). This led us to find various
transition structures and paths for the desulfinylations. The
favored transition structures for uncatalyzed and “catalyzed”
reactions involve chairlike structures H atom transfer from C-H
to OdS occurs in concert with the C-S bond formation, in
which the SdO bond of the sulfinyl group not involved in the
(84) Motoki, S.; Watanabe, T.; Saito, T. Tetrahedron Lett. 1989, 30, 189–
192.
(85) El-Sayed, I.; Franek, W.; Abdel-Megeed, M. F.; Yassin, S. M.;
Gylling, A.; Senning, A. Sulfur Lett. 1995, 19, 59–66.
(86) Baldwin, J. E.; Adlington, R. M.; Ichikawa, Y.; Kneale, C. J. J. Chem.
Soc. Chem. Commun. 1988, 702–704.
(87) Engler, T. A.; Ali, M. H.; Vander Velde, D. Tetrahedron Lett. 1989,
30, 1761–1764.
(101) Huang, X. G.; Craita, C.; Vogel, P. J. Org. Chem. 2004, 69, 4272–
4275.
(88) Rogic, M. M.; Masilamani, D. J. Am. Chem. Soc. 1977, 99, 5219–
5220.
(102) Fernandez, T.; Sordo, J. A.; Monnat, F.; Deguin, B.; Vogel, P. J. Am.
Chem. Soc. 1998, 120, 13276–13277.
(89) Masilamani, D.; Rogic, M. M. J. Am. Chem. Soc. 1978, 100, 4634–
4635.
(103) Monnat, F.; Vogel, P.; Rayon, V. M.; Sordo, J. A. J. Org. Chem.
2002, 67, 1882–1889.
(90) Masilamani, D.; Reuman, M. E.; Rogic, M. M. J. Org. Chem. 1980,
45, 4602–4605.
(104) Monnat, F.; Vogel, P.; Meana, R.; Sordo, J. A. Angew. Chem., Int.
Ed. 2003, 42, 3924–3927.
(91) Markovic, D.; Vogel, P. Angew. Chem., Int. Ed. 2004, 43, 2928–
2930.
(105) Maruoka, K.; Ooi, T.; Yamamoto, H. J. Am. Chem. Soc. 1990, 112,
9011–9012.
(92) Markovic, D.; Varela-Alvarez, A.; Sordo, J. A.; Vogel, P. J. Am.
Chem. Soc. 2006, 128, 7782–7795.
(106) Achmatowicz, O.; Szechner, B. J. Org. Chem. 1972, 37, 964–967.
(107) Okachi, T.; Fujimoto, K.; Onaka, M. Org. Lett. 2002, 4, 1667–1669.
(108) Okachi, T.; Onaka, M. Stud. Surf. Sci. Catal. 2003, 145, 141–144.
(109) Oppolzer, W.; Snieckus, V. Angew. Chem., Int. Ed. Engl. 1978, 17,
476–486.
(93) Deguin, B.; Roulet, J. M.; Vogel, P. Tetrahedron Lett. 1997, 38,
6197–6200.
(94) Roulet, J. M.; Puhr, G.; Vogel, P. Tetrahedron Lett. 1997, 38, 6201–
6204.
(110) Snider, B. B. Acc. Chem. Res. 1980, 13, 426–432.
(111) Song, Z. G.; Beak, P. J. Am. Chem. Soc. 1990, 112, 8126–8134.
(112) Achmatowicz, O.; Bialecka-Florjanczyk, E. Tetrahedron 1996, 52,
8827–8834.
(95) Turks, M.; Murcia, M. C.; Scopelliti, R.; Vogel, P. Org. Lett. 2004,
6, 3031–3034.
(96) Turks, M.; Huang, X. G.; Vogel, P. Chem.sEur. J. 2005, 11, 465–
476.
(113) Braddock, D. C.; Hii, K. K. M.; Brown, J. M. Angew. Chem., Int.
Ed. 1998, 37, 1720–1723.
(97) Craita, C.; Didier, C.; Vogel, P. Chem. Commun. 2007, 2411–2413.
(98) Turks, M.; Fonquerne, F.; Vogel, P. Org. Lett. 2004, 6, 1053–1056.
(99) Vogel, P.; Turks, M.; Bouchez, L. C.; Markovic, D.; Varela-Alvarez,
A.; Sordo, J. A. Acc. Chem. Res. 2007, 40, 931–942.
(100) Vogel, P.; Turks, M.; Bouchez, L.; Craita, C.; Huang, X. G.; Murcia,
M. C.; Fonquerne, F.; Didier, C.; Flowers, C. Pure Appl. Chem. 2008,
80, 791–805.
(114) Yamanaka, M.; Mikami, K. HelV. Chim. Acta 2002, 85, 4264–4271.
(115) Morao, M.; McNamara, J. P.; Hillier, A. H. J. Am. Chem. Soc. 2003,
125, 628–629.
(116) Choomwattana, S.; Maihom, T.; Khongpracha, P.; Probst, M.;
Limtrakul, J. J. Phys. Chem. C 2008, 112, 10855–10861.
(117) Clarke, M. L.; France, M. B. Tetrahedron 2008, 64, 9003–9031.
9
9550 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
Figure 1. B3LYP/6-31+G(d,p) structures of the species located on the PES of the retro-ene reaction of monomeric prop-2-enesulfinic acid (13). In parentheses,
CCSD(T)/6-311+G(d,p)//B3LYP//6-31+G(d,p) Gibbs free energy differences, including solvent effects, with respect to 0.5 ·(13)2 (Table 2).
Theoretical Methods
pericyclic reaction occupies pseudoaxial position which prefers
to be gauche with respect to the C(1)-C(2) bond. This
stabilizing gauche effect is also found in prop-2-ene sulfinic
acid itself and in its dimer that forms in CH2Cl2 at -15 °C.
Thus, our calculations have uncovered a case of stabilizing
gauche effect that involves a polar SdO bond and a much less
polar C-C bond. Other possible mechanisms involving the
intermediacy of ꢀ-sultane, thirane-1,1-dioxide or 1,4-diradical
T zwitterionic intermediates require higher activation free
energies than the concerted mechanism.
Geometry optimizations were carried out using density
functional theory (DFT)118 with the Becke three-parameter
Lee-Yang-Parr119,120 (B3LYP) hybrid functional and Pople’s
6-31+G(d,p) basis set.121 All the geometrical parameters were
fully optimized, and all the structures located on the potential
energy surfaces (PESs) were characterized as minima or
transition structures by computing the corresponding Hessian
matrices and examining the number of imaginary frequencies
from them. Graphical analysis of the imaginary frequencies of
the transition structures, as well as intrinsic reaction coordinate
(118) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density
Functional Theory, 2nd ed.; Wiley-VCH: Weinheim, 2000.
(119) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785–
789.
(120) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(121) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab initio
Molecular Orbital Theory; Wiley: New York, 1986.
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9551

´
Varela-Alvarez et al.
A R T I C L E S
(IRC) calculations, allowed us to interconnect the different
structures located on the PESs and then construct the corre-
sponding energy profiles.
The energy predictions were improved by performing single-
point CCSD(T)/6-311+G(d,p)//B3LYP/6-31+G(d,p) calcula-
tions.121,122 Solvent effects were estimated through quantum
mechanical free energy calculations based on the self-consistent
reaction field (SCRF) method augmented by atomic surface
tensions.123 The so-called solvation model 5.43 (SM5.43R)124,125
was employed. This model is specially suitable in the present
case as it was parametrized to predict the free energy of solvation
of solutions with species containing H, C, O, and S atoms in
organic solvents (including CH2Cl2) using B3LYP/6-31G(d)
method to describe the electronic structure of the solute.
Calculations were carried out for CH2Cl2 as a solvent, thus
simulating experimental conditions. The ion convention standard
state (1M)122 was used to compute the changes in Gibbs free
energies.
The thermodynamic functions (∆H, ∆S, and ∆G) were
estimated within the ideal gas, rigid rotor, and harmonic
oscillator approximations.126 A temperature of 258.15 K was
assumed. Kinetics calculations were performed within the
context of the conventional transition state theory (CTST),127
estimating tunnel effects by means of the small curvature
tunnelling (SCT) approximation.128 The Gaussian03 package
of programs129 was used to carry out the electronic structure
calculations.
Figure 2. B3LYP/6-31G+G(d,p) structure for the (13)2 dimer.
Table 2. CCSD(T)/6-311+G(d,p)//B3LYP//6-31+ G(d,p) Relative
Energy Values for the Intermediates and Transition Structures of
the Monomolecular and Bimolecular Mechanisms of the Retro-Ene
Elimination of SO2 from Prop-2-enesulfinyl Acid (13)a
structure
∆U
∆H
∆S
∆G
∆Gsolv,SMX
dimer
monomer
0.5 ·(13)2
13a
T1
13b
T2
T3
T4
0.0
0.0
0.0
0.0
0.0
3.5
8.2
11.7 11.1 19.2
16.8 15.3 17.9
13.0 12.4 20.5
35.6 31.6 15.1
37.2 33.1 15.4
6.9
11.4
7.9
28.5
29.9
28.4
30.4
5.0
monomolecular
bimolecular
24.3
25.9
25.8
28.4
-5.6
31.7 28.6
33.8 30.5
0.9
0.4
T5
Propene + SO2
11.6
9.0 57.0 -3.4
a ∆U, ∆H, and ∆S are calculated at 258.15 K and 1 atm. ∆G and
∆Gsolv,SMX calculated using standard state (1 M, 258.15 K).122 All
energy values are given in kcal/mol, entropies (∆S) in cal/K ·mol.
Scheme 4
Results and Discussion
Quantum Calculations. In a first series of calculations, we
examined the PES of prop-2-enesulfinic acid (13) without any
additive. As expected for sulfinic acids in solution,130-133 dimers
exist in equilibrium with the monomeric acids due to favorable
SdO· · ·H-OS hydrogen bonding. In the case of 13 + 13 a
(13)2 (Figure 2 and Table 2), we calculate for CH2Cl2 solution
at -15 °C ∆G ) -3.5 kcal/mol, or K(13 + 13 a (13)2) )
919. Interestingly, both molecules of 13 in dimer (13)2 adopt
the same privileged conformation shown in Scheme 3 in which
the SdO and σ(C(1)-C(2)) bond are gauche, the C-H bond
at C(2) pointing toward the oxygen center of SdO. The S-OH
moiety is antiperiplanar with respect to σ(C(1)-C(2)).
1). The greater stability of 13a with respect to 13b (1.5 kcal/
mol) might be attributed to hydrogen bridging between the SOH
group and the π-electrons of the alkene unit that is more
favorable in 13a than in 13b (Table 2).
As found experimentally (sealed tube experiments), the
conversion of (13)2 into 13 + propene + SO2 is exergonic at
-15 °C. Monomer 13a undergoes retro-ene elimination of SO2
following a one-step, concerted mechanism for which transition
structure T2 is located (Figures 1 and 3, Table 2) and which
shows a C-S bond significantly elongated (from 1.833 Å in
13a to 2.257 Å in T2) and a high degree of C(3)-H bond
formation (from 2.748 Å in 13a to 1.382 Å in T2). In T2, SdO
and σ(C(1)-C(2)) are gauche, as in 13a. This transition structure
realizes a six-membered ring with H-C(1) bonds both staggered
with S-O, and SdO bonds (Scheme 5). Furthermore, if one
considers the plane realized by S, C(1), and C(3), the oxygen
center of S-OH is below this plane whereas C(2) is above this
plane, making it to resemble a chair type of structure with an
axial SdO group. Our calculations predict a second transition
state for the retro-ene elimination of SO2 from monomeric 13
which is 1.6 kcal/mol higher in free energy than T2. Such a
transition structure, T3, adopts a conformation resembling that
of a boat six-membered ring in which S-O and σ(C(1)-C(2))
bonds, as well as one of the two C(1)-H bonds and SdO double
bond, are nearly eclipsed. These eclipsing interactions are
avoided in T2, thus explaining the lower free energy of T2
compared with T3 (Table 2). Scheme 5 shows the two possible
reaction pathways for the monomolecular retro-ene mechanism.
Two conformers are found for monomeric 13. In both of
them, 13a and 13b, SdO/σ(C(1)-C(2)) and S-OH/σ(C(1)-C(2))
are gauche. Monomers 13a and 13b are equilibrated by rotation
about the σ(C(1)-C(2)) bond through transition state T1 (Figure
(122) Cramer, C. J. Essentials of Computational Chemistry: Theories and
Models; Wiley: Chichester, 2002.
(123) Chamberlin, A. C.; Kelly, C. P.; Thompson, J. D.; Lynch, B. J.; Xidos,
J. D.; Hawkins, G. D.; Li, J.; Zhu, T. H.; Volobuev, Y.; Rinaldi, D.;
Liotard, D. A.; Cramer, C. J.; Truhlar, D. G. SMXGAUS, Version
3.4.2; University of Minnesota: Minneapolis, 2007.
(124) Thompson, J. D.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. A
2004, 108, 6532–6542.
(125) Thompson, J. D.; Cramer, C. J.; Truhlar, D. G. Theor. Chem. Acc.
2005, 113, 107–131.
(126) McQuarrie, D. A. Statistical Thermodynamics; University Science
Books: Mill Valley, CA, 1973.
(127) Laidler, K. J. Chemical Kinetics; Harper and Row: New York, 1987.
(128) Fernandez-Ramos, A.; Miller, J. A.; Klippenstein, S. J.; Truhlar, D. G.
Chem. ReV. 2006, 106, 4518–4584.
(129) Frisch, M. J.; Gaussian03, ReVision C.01; Gaussian, Inc.: Wallington,
CT, 2004.
(130) Wright, W. G. J. Chem. Soc. 1949, 683–692.
(131) Wudl, F.; Lightner, D. A.; Cram, D. J. J. Am. Chem. Soc. 1967, 89,
4099–4101.
(132) Minkwitz, R.; Kornath, A.; Lohmann, U. Z. Naturforsch. Teil B 1996,
51, 739–743.
9
9552 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
intermediate mechanism for some particular ene reactions as a
direct consequence of the presence of valley-ridge inflections
on their PESs. Indeed, the existence of the so-called branching
points (those having all vibrational frequencies positive but one
being zero)145,146 has been shown to be crucial in order to
rationalize fundamental aspects of chemical reactivity29,144,147
and weak interactions.148,149 We have explored the possibility
of existence of some thiirane 1-dioxide 14, ꢀ-sultane 15, or
diradical intermediates 16a, 16b (see Figure 1 and Scheme 7)
on the PES for the retro-ene reaction of 13. Figure 1 shows the
four structures we located as intermediates at the B3LYP/6-
31+G(d,p) level. Their corresponding CCSD(T)/6-311+G(d,p)//
B3LYP/6-31+G(d,p) Gibbs free energies, including solvent
effects, become 30.2, 1.4, 34.2, and 34.9 kcal/mol, respectively.
Therefore, participation of the 1-dioxide 14 and diradical
intermediates 16a, 16b in the retro-ene elimination mechanism
is unlikely. The ꢀ-sultane 15, resulting from a formal [2 +
2]-cycloaddition, is slightly less stable than sulfinic acid 13.
As expected,150-152 a concerted mechanism involves a rather
high energy transition state (37.5 kcal/mol). A two-step process
would imply diradical structures 16a, 16b; both also are too
high in free energy. As a conclusion, our calculations suggest
the monomolecular retro-ene elimination of SO2 from prop-2-
enesulfinic acid to proceed through a concerted, pericyclic
pathway.
Figure 3. B3LYP/6-31+G(d,p) structures of the species located on the
PES of the retro-ene reaction of prop-2-enesulfinic acid (13) in the presence
of a second molecule of 13. Some selected bond distances are given in Å.
In parentheses, CCSD(T)/6-311+G(d,p)//B3LYP//6-31+G(d,p) Gibbs free
energy differences, including solvent effects, with respect to 0.5 ·(13)2 (Table
3).
In a second series of calculations we examined whether dimer
(13)2 (Figure 2) would be able to undergo the retro-ene
elimination of SO2 without monomerization, thus involving
“autocatalysis” of the ene reaction. Transition structures T4
(∆G‡ ) 25.4 kcal/mol) and T5 (∆G‡ ) 28.4 kcal/mol) (Figure
3, Table 2) correspond to the lowest free energy transition states
calculated for bimolecular mechanisms. The data suggest that
a bimolecular mechanism, at least through transition state T4,
can compete at -15 °C with the monomolecular retro-ene
reactions involving transition states T2 and T3, as observed
experimentally (see below). Transition structures T4 and T5
can be formally viewed as transition structures T2 and T3,
respectively, stabilized by a hydrogen bonding between their
SdO moieties and the corresponding SOH groups of the
spectrator molecule 13. There are also weaker electrostatic
interactions involving the SdO moiety of spectrator molecule
13 and the C-H bond at C(2) in T2 or the C-H bond at C(1)
and C(3) in T3. These interactions represent favorable enthalpic
effects as can be seen in Table 1: ∆H‡ (T2) ) 31.6 kcal/mol vs
∆H‡ (T4) ) 28.6 kcal/mol. However, entropic contributions
Scheme 5
It should be mentioned that a weak complex between SO2 and
propene was located on the PES. However, it resulted higher
in Gibbs free energy than products (SO2 + propene) and,
consequently, it has been not included in the ∆G profile shown
in Scheme 6.
For the ene-reactions of alkenes with singlet dioxygen,29,33,43,134
with nitrosoarenes,42,43,46,135,136 triazolinediones,39,46,137-143 and
aldehydes3,111-114,116 mechanisms involving intermediates have
been invoked in some cases. Interestingly, Houk, Singleton, and
co-workers29 and Singleton and co-workers144 have shown
theoretical evidence supporting the possibility of a two-step no-
‡
clearly favor T2: ∆Gsolv,SMX‡ (T2) ) 24.3 kcal/mol vs ∆Gsolv,SMX
(T4) ) 25.8 kcal/mol.
It is well known that consideration of transition structures
involving additional molecules of one of the reactants acting
(133) Beckmann, J.; Finke, P.; Hesse, M.; Wettig, B. Angew. Chem., Int.
Ed. 2008, 47, 9982–9984.
(144) Bekele, T.; Christian, C. F.; Lipton, M. A.; Singleton, D. A. J. Am.
Chem. Soc. 2005, 127, 9216–9223.
(134) Orfanopoulos, M. Mol. Supramol. Photochem. 2001, 8, 243–285.
(135) Adam, W.; Krebs, O.; Orfanopoulos, M.; Stratakis, M. J. Org. Chem.
2002, 67, 8395–8399.
(145) Bosch, E.; Moreno, M.; Lluch, J. M.; Bertran, J. Chem. Phys. Lett.
1985, 160, 543–548.
(146) Quapp, W.; Heidrich, D. J. Mol. Struct. (THEOCHEM) 2002, 585,
105–117.
(136) Leach, A. G.; Houk, K. N. Org. Biomol. Chem. 2003, 1, 1389–1403.
(137) Ohashi, S.; Butler, G. B. J. Org. Chem. 1980, 45, 3472–3476.
(138) Seymour, C. A.; Greene, F. D. J. Am. Chem. Soc. 1980, 102, 6384–
6385.
(147) Brana, P.; Menendez, B.; Fernandez, T.; Sordo, J. A. J. Phys. Chem.
A 2000, 104, 10842–10854.
(148) Valdes, H.; Rayon, V. M.; Sordo, J. A. Chem. Phys. Lett. 1999, 309,
265–273.
(139) Cheng, C. C.; Seymour, C. A.; Petti, M. A.; Greene, F. D.; Blount,
J. F. J. Org. Chem. 1984, 49, 2910–2916.
(149) Valdes, H.; Sordo, J. A. Chem. Phys. Lett. 2001, 333, 169–180.
(150) Sordo, J. A.; Gonza´lez, J.; Sordo, T. L. J. Am. Chem. Soc. 1992,
114, 6249–6251.
(140) Elemes, Y.; Foote, C. S. J. Am. Chem. Soc. 1992, 114, 6044–6050.
(141) Chen, J. S.; Houk, K. N.; Foote, C. S. J. Am. Chem. Soc. 1997, 119,
9852–9855.
(151) Assfeld, X.; Ruiz-Lo´pez, M. F.; Gonza´lez, J.; Lo´pez, R.; Sordo, J. A.;
Sordo, T. L. J. Comput. Chem. 1994, 15, 479–487.
(152) Bran˜a, P.; Gimeno, J.; Sordo, J. A. J. Org. Chem. 2004, 69, 2544–
2550.
(142) Singleton, D. A.; Hang, C. J. Am. Chem. Soc. 1999, 121, 11885–
11893.
(143) Acevedo, O.; Squillacote, M. E. J. Org. Chem. 2008, 73, 912–922.
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9553

´
Varela-Alvarez et al.
A R T I C L E S
Scheme 6. CCSD(T)/6-31+G(d,p)//B3LYP/6-31+G(d,p) + SM5.43R ∆G Profile
Table 3. CCSD(T)/6-311+G(d,p)//B3LYP/6-31+G(d,p) Relative
Energy Values for the Intermediates and Transition Structures of
the Retro-Ene Reaction of 13 in the Presence of Additive A (SO2,
BF3, CF3COOH)a
T2 through its S center. There is also a weak electrostatic
interaction between one SdO moiety of SO2 and C-H at C(2).
In the case of SO2 approaching T3, as in T7, again the SdO
moiety of T3 interacts with the S center of SO2 and a weak
electrostatic interaction can be recognized between one SdO
moiety of SO2 with C-Hs at C(1) and C(3) of T3. Comparison
of the differences of enthalpies (∆H, Tables 2 and 3) calculated
for T6 (23.2 kcal/mol)/T2 (31.6 kcal/mol) and T7 (24.8 kcal/
mol)/T3 (33.1 kcal/mol) one finds that SO2 stabilizes both
transition structures T2 and T3. Nevertheless, because of the
negative entropy this requires, the calculated free energy of
activation are higher at -15 °C for the SO2-assisted reactions
(25.2 and 27.5 kcal/mol) than for the monomolecular retro-ene
reactions (24.3 and 25.9 kcal/mol). This prediction is confirmed
experimentally (see Table 4).
Lewis acids promote the hetero-Diels-Alder and cheletropic
additions of SO2.155,156 We thus examined the effect of BF3 on
the desulfinylation of 13. Experiments were carried out using
both BF3 and BF3 ·Me2O moieties. We focused our theoretical
analysis on the reaction in the presence of BF3 to keep the
computational cost within reasonable limits. Some comments
on the stability of 19 when employing BF3 ·Me2O can be found
in Section D. Calculations show that both conformers of 13
form stable complexes 17 (BF3 + 13a) and 18 (BF3 + 13b).
Interestingly, dimer (13)2 combines with 2 equiv of BF3 forming
tetramolecular complex 19 (Figure 5). In CH2Cl2 at -15 °C
complexes 17 (-11.2 kcal/mol), 18 (-9.8 kcal/mol), and 19
(-15.5 kcal/mol) are definitively more stable than BF3 + 0.5·
a ∆U, ∆H, and ∆S calculated at 258.15 K and 1 atm. ∆G and
∆Gsolv,SMX calculated using the standard state 1 M at 258.15 K.122 All
energy values are given in kcal/mol, entropies (∆S) in cal/K ·mol.
Scheme 7
as a catalyst can sometimes be relevant.153 As previous studies
on the hetero-Diels-Alder and cheletropic addition of sulfur
dioxide had shown that SO2 promotes these reactions,102,103,154
we examined whether this could also be the case for the ene-
reaction of SO2 with propene. Our results are summarized in
Table 3 for the lowest possible transition states T6 and T7
(Figure 4). Formally, on approaching transition structure T2 as
in T6, one molecule of SO2 interacts with the SdO moiety of
Figure 4. B3LYP/6-31+G(d,p) structures of the transition states located
on the PES of the retro-ene reaction of 13 in the presence one molecule of
SO2. Some selected bond distances are given in Å. In parentheses, CCSD(T)/
6-311+G(d,p)//B3LYP//6-31+G(d,p) Gibbs free energy differences, includ-
ing solvent effects, with respect to 0.5 ·(13)2 (Table 3).
(153) Mene´ndez, M. I.; Suarez, D.; Sordo, J. A.; Sordo, T. L. J. Comput.
Chem. 1995, 16, 659–666.
(154) Vogel, P.; Sordo, J. A. Curr. Org. Chem. 2006, 10, 2007–2036.
(155) Deguin, B.; Vogel, P. J. Am. Chem. Soc. 1992, 114, 9210–9211.
9
9554 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
Table 4. Experimental Kinetic Data Obtained for the
Desulfinylation of 13a in CD2Cl2 at -15 °Cb
in T10 and C(3) in T11. Calculated activation free energies are
23.2 and 24.0 kcal/mol for T10 and T11, respectively, what is
somewhat lower than for the monomolecular reaction implying
transition structures T2 (24.3 kcal/mol) and T3 (25.9 kcal/mol),
and the bimolecular mechanism involving transition structure T4
(25.8 kcal/mol) and T5 (28.4 kcal/mol).
n(TFA)/
n(tin sulf)
k′′/(mol1/2
/
entry
additive (other conditions)
k′/min-1
kg1/2 ·min)
1
2
3
4
5
6
7
8
0.47
0.86
1.19
1.88
5.60
0.89
0.82
0.73
0.83
0.92
0.80
-
0.00575
0.00565
0.00578
0.00564
0.0055
-
TFA (0.19 equiv)
TFA (0.88 equiv)
TFA (4.6 equiv)
BF3 OMe2, (0.19 equiv)
BF3 OMe2, (10 equiv)
TBAF (0.26 equiv)
SO2 (12.99 equiv)
C6D12, 25 °C
0.00231
0.00335
0.00928
0.00065
In the particular cases where the retro-ene reactions occur in
the presence either of an activating second molecule of sulfinic
acid or a molecule of CF3CO2H, an alternative pathway was
identified on their respective PESs. According to this second
mechanism, the molecule accompanying the initial sulfinic acid
reactant is not a mere spectator like in the regular mechanism
discussed above (see T4, T5 (Table 2, Figure 4), T10 and T11
(Table 3, Figure 6) but it plays an active role forming part directly
of the reaction coordinate. Indeed, in the case of the reaction in
the presence of a CF3CO2H molecule, the SO2H hydrogen in the
sulfinic acid species is transferred to the carboxylic group of
CF3CO2H at the same time that the hydroxylic hydrogen in this
latter species is being transferred to the CH2 terminal group of the
prop-2-enesulfinic acid (see T10′ and T11′ structures in the
Supporting Information). For the retro-ene reaction in the presence
of a second molecule of 13, this alternative mechanism involves
the transfer of the SO2H hydrogen from the reacting 13 to the
accompanying molecule 13 and the simultaneous transfer of the
SO2H hydrogen from the accompanying molecule 13 to the
propene-like fragment in the reactant 13 (see T4′ and T5′ structures
in the Supporting Information). According to our calculations
(Tables S2 and S4 in the Supporting Information), the above-
described alternative mechanism involve much higher energy
barriers [∆Gsolv,SMX ) 38.1 (T4′), 37.8 (T5′), 33.1 (T10′) and 33.0
(T11′) kcal/mol; to be compared with the corresponding values
25.8 (T4), 28.4 (T5), 23.2 (T10) and 24.0 (T11)] and can be safely
discarded. Obviously, disfavoring entropy contributions must result
much larger than the enthalpic stabilizations associated to the
O-H· · ·O interactions present in the T4′, T5′, T10′, and T11′
structures. Consequently, such alternative pathways were not
included in Tables 2-3. (Structures T4′, T5′, T10′, and T11′ were
initially located at lower levels of theory by Mr. Ruben Meana
during a preliminary search on the PES at a very earlier stage of
his research.).
0.0061
no reaction
0.00581
0.0063
0.02927
too fast to
be measured
by NMR
0.00004
0.00099
9
10
11
CD2Cl2, 25 °C
a Prepared by addition of CF3COOH (TFA) solution in CD2Cl2 to a
CD2Cl2 solution of tributyltin prop-2-ene-sulfinate (tin sulf).4 b See text
for definitions (eqs 1-7).
(13)2 (Table 3). Scheme 8 shows the energy profile for the retro-
ene reaction of 13 in the presence of BF3. Interestingly, the SdO
bonds of 19 which coordinates BF3 are not gauche with respect
to σ(C(1)-C(2)), as it is the case for complex 17. The two
S-O-H groups are gauche with respects to σ(C(1)-C(2)) and
relatively important electrostatic interaction between S-OH and
F-B contribute to the high stability of this tetramolecular
complex (see Scheme 9). Weak complexes SO2 · · ·BF3,157
SO2 · · ·propene, and BF3 · · ·propene, located on the PES, resulted
higher in Gibbs free energy than products (BF3 + SO2 +
propene) and, consequently, they were not included in the ∆G
profile shown in Scheme 8.
For the two lowest lying transition states T8 and T9 which
can be formally viewed as transition structures T2 and T3,
respectively, coordinated to BF3 by interaction of the SdO
moiety with the boron center of BF3, we calculated activation
free enthalpies of 8.3 and 12.3 kcal/mol with respect to 0.5 ·(13)2
+ BF3. This would correspond to a substantial catalytic effect
of BF3 on the desulfinylation of both monomeric and dimeric
forms of 13 if they would not be highly stabilized through
coordination with BF3. In fact, our calculations give ∆G‡(0.5·19
f T8) ) 23.8 kcal/mol, thus suggesting a insignificant catalytic
effect of BF3 in CH2Cl2 at -15 °C. These calculations suggest
that the ene reaction of propene with sulfur dioxide, which is
endergonic at -15 °C in CH2Cl2/SO2 mixtures, might become
exergonic in the presence of 1 equiv of BF3. Indeed, when
propene was added to a premixed mixture of SO2 and BF3 in
Kinetics Analyses of the Desulfinylation of Prop-
2-enesulfinic Acid. 1. Experimental Results. Table 4 collects the
experimental results of a series of kinetics measurements of the
disulfitative retro-ene reaction of prop-2-ene-1sulfinic acid in
the presence of different additives (A ) CF3CO2H, BF3OMe2,
TBAF, SO2, C6D12, and CD2Cl2). Experimental details can be
found in the Supporting Information.
1
CD2Cl2, a quick reaction occurred already at -78 °C. The H
In the absence of additive A (entries 1, 2; Table 4), the rate
of appearance of propene (P) fitted the following rate law:
NMR spectrum of the crude reaction mixture showed signals
of 13 (minor) and those of polymeric material (major). Thus, it
should be possible to realize ene reactions of SO2 with alkenes
if a suitable analogue of BF3 is willing to catalyze them and to
form stable complexes with the sulfinic acids formed, without
inducing polymerization.
d[P]/dt ) k′[(13)2]
(1)
where k′ is the global rate constant corresponding to the process:
As we had shown that CF3COOH is a good catalyst for the
hetero-Diels-Alder and cheletropic additions of sulfur di-
oxide,155,158 we examined whether this protic acid would also
catalyze the retro-ene elimination of SO2 from prop-2-enesulfinic
acid. Our calculations are summarized in Table 3 and Figure 6 for
the lowest free energy transition states T10 and T11 that can be
formally seen as T2 and T3, respectively, interacting with
CF3COOH through hydrogen bonding between SdO moiety and
OH of CF3COOH. Weak electrostatic interaction is probably
present between CdO of CF3COOH and one C-H bond of C(1)
Integration of eq 1 gives
[P] ) 2Co(1 - e-k′t
)
(2)
with k′ ) k′2K1 (K1 ) k1/k-1), and Co ) initial concentration of
dimer (13)2.
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9555

´
Varela-Alvarez et al.
A R T I C L E S
In the presence of additives A ) CF3CO2H, TBFA, or SO2
(entries 3, 4, 5, 8, 9; Table 4), the rate of appearance of P fitted
the following rate law:
with k′′ ) K11/2k′′2[A]
Integration of eq 3 gives
d[P]/dt ) k′[(13)2] + k′′[(13)2]1/2
(3)
2
where k′ and k′′ are global rate constants corresponding to the
(k′′ + k′ C ) exp(-k′t/2) - k′′
√
0
[P] ) 2 C -
(4)
{
}
[
]
0
following processes:
k′
Experimental [P] vs t curves can be found in the Supporting
Information. Fittings to eqs 2 and 4 provide the values for k′
and k′′ rate constants collected in Table 4.
In Figure 7 are presented four kinetics for the decay of prop-
2-enesulfinic acid in the retroene reaction. In absence of any
with k′ ) K1k′2, and
Figure 5. B3LYP/6-31+G(d,p) structures of the species located on the PES of the retro-ene reaction of 13 in the presence of BF3. Some selected bond
distances are given in Å. CCSD(T)/6-311+G(d,p)//B3LYP//6-31+G(d,p) Gibbs free energy differences, including solvent effects, with respect to 0.5 ·(13)2
(Table 3).
9
9556 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
mer). Our calculations suggest that a pre-equilibrium between
13 and its dimer (13)2 is established very fast. Indeed, no energy
barrier for the dimer formation was detected on the PES. A
double hydrogen-bonding-like interaction (Figure 2) with
S-OH· · ·H bond distances of 1.631 Å is responsible for the
dimer stabilization. The energy profile arising from our calcula-
tions (see Scheme 6) suggests a half-order in the sulfinic acid
dimer (first-order rate law in the sulfinic acid monomer) defined
by Scheme 10.
According to our calculations (see Table 2), (13)2 is about
3-5 kcal/mol more stable than the corresponding monomer 13
in any of the two conformations 13a or 13b at -15° in CH2Cl2.
Conformers 13a and 13b interconvert in each other through a
transition structure (see T1 in Figure 1 and Scheme 3) involving
a small barrier of less than 5 kcal/mol (Table 2).
B. First-Order Mechanism for the Desulfinylation of the
Dimer (Second-Order Mechanism in the Monomer). Let us now
consider the possibility that two molecules of 13 take part in
the process according to a first-order mechanism in the sulfinic
acid dimer (second-order mechanism in the monomer) defined
by Scheme 11.
Two molecules of 13 as conformers 13a or 13b interact with
each other giving rise to the products of the retro-ene process
(SO2 + propene + 1/2 dimer) through transition structures T4
and T5 (Table 2, Figure 4), respectively. One of the molecules
plays the role of activator and the second molecule undergoes
itself the rearrangements required to render the retro-ene
products in a concerted way as described above.
Figure 6. B3LYP/6-31+G(d,p) transition structures located on the PES
of the retro-ene reaction of 13 in the presence of one molecule of CF3COOH.
Some selected bond distances are given in Å. CCSD(T)/6-311+G(d,p)//
B3LYP//6-31+G(d,p) Gibbs free energy differences, including solvent
effects, with respect to 0.5 ·(13)2 (Table 3).
As pointed out in a previous section, the activation free energy
for the bimolecular mechanism, ∆Gsolv,SMX‡(T4) ) 25.8 kcal/
mol, is slightly larger than the corresponding value for the
monomolecular mechanism, ∆Gsolv,SMX‡(T2) ) 24.3 kcal/mol.
This theoretical prediction apparently contradicts the experi-
mental evidence (see Table 4) which shows that the retro-ene
reaction of 13 is a first-order reaction in the dimer with a reaction
rate constant of kglobal ) 5.7 × 10-3 min-1. However, one must
take into consideration the fact that the structure of the transition
structures defining the first-order mechanism, T4 and T5, are
conformationally very complex. In fact, Figure 8 collects
the 24 different conformations a· · ·x, all of them energetically
close to each other (25.3-26.3 kcal/mol), adopted by T4. There
are another 24 equivalent structures for T5 (Table 5) (see Figures
additive, Figure 7a and b, the kinetics follows first order decay
with rate law values fitted as k′a ) 0.00575 min-1 (R2
)
a
0.98313) and k′b ) 0.00565 min-1 (R2 ) 0.99482). Figure 7c
b
and d represents decay of prop-2-enesulfinic acid with 0.19 or
0.88 equiv of TFA as additive. If these kinetics are fitted with
equations combining mixed half and first order rate laws the R2
value are R2 ) 0.99904 and R2 ) 0.9984, respectively, and
c
d
the rate constant values are k′c ) 0.00578 min-1 and k′′c )
0.00055 mol1/2/kg1/2 ·min and k′d ) 0.00564 min-1 and k′′d )
0.00335 mol1/2/kg1/2 ·min, respectively. When the decay is fitted
as first rate law the R2 value are smaller and found to amount
to R2 ) 0.99791 and R2 ) 0.99659, respectively.
c
d
Theoretical Analysis. A. Half-order mechanism for the des-
ulfinylation of the dimer (1st-order mechanism in the mono-
Scheme 8. CCSD(T)/6-31+G(d,p)//B3LYP/6-31+G(d,p) + SM5.43R ∆G Profile
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9557

´
Varela-Alvarez et al.
A R T I C L E S
Figure 7. Appearance of prop-1-ene at -15 °C arising from a mixtures made of 0.47 equiv of TFA to allyl tributyltin sulfinate (S1). (a) n(TFA)/n(prop-
2-enesulfinic acid) ) 0.47 red line represent first order fitting. (b) n(TFA)/n(prop-2-enesulfinic acid) ) 0.86 red line represents first-order fitting. (c) n(TFA)/
n(prop-2-enesulfinic acid) ) 1.19 red line represent mixed order fitting (dotted line first-order fitting). (d) n(TFA)/n(prop-2-enesulfinic acid) ) 1.88 red line
represent mixed order fitting (dotted line first-order fitting).
Scheme 9
Scheme 11. Bimolecular Mechanism
Scheme 10. Monomolecular Mechanism
the CTST estimate of the rate constant for a half-order
mechanism in the dimer which, after tunnelling corrections (κ
) 18.611), becomes 1.45 × 10-5 mol1/2/kg1/2 ·min. Thus, while
our calculations suggest a mixed half-order + first-order kinetics
in the dimer, experimental fittings (see Table 4) shows that
desulfinylation of 13 proceeds through a first-order mechanism
in the dimer (13)2. It is clear that our theoretical predictions
cannot reach the very high level of accuracy required to discern
the subtle factors controlling the change from a mixed half-
order + first-order mechanism to a full first-order mechanism.
In this context, let us just mention a few well-known drawbacks
associated with the theoretical treatments: (a) for quite different
reasons,159,160 theoretical kinetics predictions are far from being
a trivial task, even for relatively simple small-sized gas phase
S1-S2 of the Supporting Information) on the PES, all of them
contributing to the rate constant. We have employed the CTST
theory127 to get a rough estimate of the theoretically predicted
reaction rate constants both in the gas phase and in solution.
As can be seen in Table 5, the global rate constant for the first-
order rate law in the dimer becomes 1.65 × 10-5 min-1
(transmission coefficient, κ ) 11.065, was computed using the
SCT approximation).128 This value should be compared with
(156) Suarez, D.; Sordo, T. L.; Sordo, J. A. J. Am. Chem. Soc. 1994, 116,
763–764.
9
9558 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
Figure 8. Twenty-four possible conformers (a· · ·x) of similar relative free energies generated by rotation about the C(1)-S bond (red arrows) and about
the C(1)-C(2) bond (blue arrows) for transition structure T4 (“autocatalysis”). In parentheses, CCSD(T)/6-311+G(d,p)//B3LYP//6-31+G(d,p) Gibbs free
energy differences.
systems. One of the most critical factors is that kinetics
predictions do involve the calculation of the exp(-∆G‡/RT)
factor in the CTST mathematical expression for the rate
constant.127 Thus, relatively small errors in the theoretical
estimate of ∆G‡ do produce considerable deviations in the
predicted rate constants. A trivial calculation tells us that an
error in 1 kcal/mol in ∆G‡ (the goal, not always achieved at
all, in today’s theoretical thermodynamic predictions for small-
sized systems)161-165 gives rise to a discrepancy in 1 order of
magnitude in the rate constant. Thus, the discrepancies between
theoretically predicted (Table 5) and experimentally determined
(Table 4) rate constants, represent about 2 kcal/mol error in the
calculation of ∆G‡ which, according to our own experience,161-163
is the best level of accuracy attainable with today’s software
and hardware utilities affordable for the kind of systems (number
of electrons) dealt with in the present study. These unavoidable
discrepancies between experimentally measured and theoreti-
cally predicted rate constants, that have been emphasized in
recent works (see, for example, refs 166-168), should be kept
in mind when assessing and judging the appropriateness of a
given theoretical treatment. (b) As can be inferred from
inspection of Table 5, solvent effects exert a remarkable
influence on the ∆G‡ and k values. Despite the fact we employed
a theoretical model recently developed to properly account for
solvent effects,124,125 limitations inherent to the solvent theoretical
models are still considerable,122 and (c) although impressive
advances in the accuracy of DFT methodologies have been
accomplished during the past few years,169 it is clear that the
simultaneous accurate representation of kinetics and thermodynam-
ics for a given process like the one studied here, is not a trivial
task.170 We decided to carry out the geometry optimizations by
using the most popular and widely employed DFT functional:
(157) Peebles, S. A.; Sun, L.; Kuczkowski, R. L.; Nxumalo, L. M.; Ford,
T. A. J. Mol. Struct. 1998, 471, 235–242.
(158) Roversi, E.; Scopelliti, R.; Solari, E.; Estoppey, R.; Vogel, P.; Bran˜a,
P.; Menendez, B.; Sordo, J. A. Chem.sEur. J. 2002, 8, 1336–1355.
(159) Cimas, A.; Rayon, V. M.; Aschi, M.; Barrientos, C.; Sordo, J. A.;
Largo, A. J. Chem. Phys. 2005, 123, 114312-1–114312/11.
(160) Gonzalez-Lafont, A.; Lluch, J. M.; Varela-Alvarez, A.; Sordo, J. A.
J. Phys. Chem. B 2008, 112, 328–335.
(166) Golden, D. M.; Barker, J. R.; Lohr, L. L. J. Phys. Chem. A 2003,
107, 11057–11071.
(167) Maranzana, A.; Barker, J. R.; Tonachini, G. Phys. Chem. Chem. Phys.
2007, 9, 4129–4141.
(161) Feller, D.; Sordo, J. A. J. Chem. Phys. 2000, 112, 5604–5610.
(162) Feller, D.; Sordo, J. A. J. Chem. Phys. 2000, 113, 485–493.
(163) Sordo, J. A. J. Chem. Phys. 2001, 114, 1974–1980.
(164) Feller, D.; Peterson, K. A.; de Jong, W. A.; Dixon, D. A. J. Chem.
Phys. 2003, 118, 3510–3522.
(168) Maranzana, A.; Barker, J. R.; Tonachini, G. J. Phys. Chem. A 2008,
112, 3666–3675.
(169) Sousa, S. F.; Fernandes, P. A.; Ramos, M. J. J. Phys. Chem. A 2007,
111, 10439–10452.
(165) Feller, D.; Peterson, K. A.; Crawford, T. D. J. Chem. Phys. 2006,
124, 054107-1–054107/17.
(170) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput.
2006, 2, 364–382.
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9559

´
Varela-Alvarez et al.
A R T I C L E S
Table 5. Gas-Phase and Solution Gibbs Free Energies (kcal/mol)
for the 48 Transition Structures Located on the PES for the
First-Order Mechanism in the Sulfinic Acid Dimer As Computed at
the CCSD(T)/6-311+G(d,p)//B3LYP/6-31+G(d,p) Level, Using the
Standard State 1 M at 258.15 Ka
Scheme 12. Mechanism of the Desulfinylation in the Presence of
Additive SO2 or CF3COOH
structure
∆G(gas)
k(gas)
∆Gsolv,SMX
k(solv)
T4-a
T4-b
T4-c
T4-d
T4-e
T4-f
T4-g
T4-h
T4-i
28.4
28.7
29.0
28.2
28.4
29.0
28.8
28.0
28.1
28.6
29.0
28.2
28.5
27.9
27.9
28.3
29.3
28.6
27.9
28.1
27.6
28.1
28.7
28.2
30.4
29.8
29.7
30.6
30.5
29.8
29.0
29.6
29.6
29.3
29.9
30.1
29.3
29.1
29.1
29.4
29.7
30.0
29.9
30.6
30.6
30.0
28.7
28.0
3.24 × 10-09
1.80 × 10-09
1.00 × 10-09
4.78 × 10-09
3.24 × 10-09
1.00 × 10-09
1.48 × 10-09
7.06 × 10-09
5.81 × 10-09
2.19 × 10-09
1.00 × 10-09
4.78 × 10-09
2.66 × 10-09
8.58 × 10-09
8.58 × 10-09
3.93 × 10-09
5.60 × 10-10
2.19 × 10-09
8.58 × 10-09
5.81 × 10-09
1.54 × 10-08
5.81 × 10-09
1.80 × 10-09
4.78 × 10-09
6.56 × 10-11
2.11 × 10-10
2.57 × 10-10
4.44 × 10-11
5.40 × 10-11
2.11 × 10-10
1.00 × 10-09
3.12 × 10-10
3.12 × 10-10
5.60 × 10-10
1.74 × 10-10
1.18 × 10-10
5.60 × 10-10
8.27 × 10-10
8.27 × 10-10
4.61 × 10-10
2.57 × 10-10
1.43 × 10-10
1.74 × 10-10
4.44 × 10-11
4.44 × 10-11
1.43 × 10-10
2.80 × 10-09
7.06 × 10-09
1.22 × 10-07
25.8
25.8
26.2
25.6
25.3
25.8
26.0
25.4
25.7
25.9
26.1
25.4
25.8
25.6
25.5
25.8
26.3
25.8
25.4
25.9
25.4
25.9
26.0
25.7
28.4
28.1
27.9
28.6
28.2
27.7
27.6
27.8
27.9
27.8
27.9
28.0
27.7
27.5
27.5
27.7
27.8
28.1
27.8
28.2
28.1
27.8
26.3
25.6
5.14 × 10-07
5.14 × 10-07
2.36 × 10-07
7.59 × 10-07
1.36 × 10-06
5.14 × 10-07
3.48 × 10-07
1.12 × 10-06
6.25 × 10-07
4.23 × 10-07
2.87 × 10-07
1.12 × 10-06
5.14 × 10-07
7.59 × 10-07
9.23 × 10-07
5.14 × 10-07
1.94 × 10-07
5.14 × 10-07
1.12 × 10-06
4.23 × 10-07
1.12 × 10-06
4.23 × 10-07
3.48 × 10-07
6.25 × 10-07
3.24 × 10-09
5.81 × 10-09
8.58 × 10-09
2.19 × 10-09
4.78 × 10-09
1.27 × 10-08
1.54 × 10-08
1.04 × 10-08
8.58 × 10-09
1.04 × 10-08
8.58 × 10-09
7.06 × 10-09
1.27 × 10-08
1.87 × 10-08
1.87 × 10-08
1.27 × 10-08
1.04 × 10-08
5.81 × 10-09
1.04 × 10-08
4.78 × 10-09
5.81 × 10-09
1.04 × 10-08
1.94 × 10-07
7.59 × 10-07
1.65 × 10-05
T4-j
T4-k
T4-l
Transition structures T6 (25.2 kcal/mol) and T7 (25.4 kcal/
mol) for the retro-ene reaction in the presence of SO2 (see Table
3, Figure 4) are slightly higher than the energy barriers of the
desulfinylation reaction in the absence of SO2 (Table 2).
Therefore, calculations suggest that SO2 does not catalyze the
reaction 13 f propene + SO2, in agreement with experimental
data (Table 4).
In the presence of CF3CO2H, the free energy of activation
(see Table 3) becomes slightly lower: 23.2 (T10) and 24.0 (T11)
kcal/mol, thus suggesting a moderate catalytic effect which is
also consistent with experimental data collected in Table 4. The
estimated global rate constant for the kinetic law equation in
eq 3 becomes 8.85 × 10-5 mol1/2/kg1/2 ·min, which, bearing in
mind the theoretical limitations to make quantitative kinetics
predictions mentioned in the preceding section, agrees reason-
ably well with the experimentally estimated rate constant of
9.38 × 10-3 mol1/2/kg1/2 ·min.
T4-m
T4-n
T4-o
T4-p
T4-q
T4-r
T4-s
T4-t
T4-u
T4-v
T4-w
T4-x
T5-a
T5-b
T5-c
T5-d
T5-e
T5-f
T5-g
T5-h
T5-i
D. Desulfinylation in the Presence of BF3 ·Me2O. The free
energy diagram for the desulfinylation of 13 in the presence of
BF3 is somewhat more complicated (see Scheme 8) than those
for the other reactions presented above. This is due to the fact
that BF3 generates with 13 two relatively stable complexes 17
and 18 and, more importantly, the tetramolecular complex 19
(Table 3, Figure 5). No data have been reported yet for the
enthalpies of complex formation of BF3 with sulfinic acids. The
enthalpy change for the reaction of gaseous BF3 and Et2O in
CH2Cl2 has been measured to amount to -18.8 kcal/mol. With
MeSOMe in CH2Cl2, the reaction is more exothermic and gives
a heat of complex formation of -25.7 kcal/mol.171 Thus, if
sulfoxides can be considered to mimick sulfinic acids for their
Lewis basicity, it is expected that BF3 ·Me2O, the additive
employed in our experiments instead of BF3, will form stable
complex with dimeric (13)2, producing 19. By manometric
measurements, McLaughlin and Tamres,172 determined for
liquid BF3 ·Me2O a heat of complex formation of -13.65 (
0.2 kcal/mol (25 °C) and an entropy of complex formation of
-33.1 ( 0.6 e.u. (25 °C). -15 °C (our reaction conditions)
∆G ) -13650 - 258.15(-33.1) = -5.1 kcal/mol.
T5-j
T5-k
T5-l
T5-m
T5-n
T5-o
T5-p
T5-q
T5-r
T5-s
T5-t
T5-u
T5-v
T5-w
T5-x
total
a The corresponding CTST rate constants (min-1) are also included
(tunnel contributions were estimated with the SCT approximation).
B3LYP.118-120 Although some of its shortcomings in predicting
stabilities or kinetic barriers170 are expected to be overcome through
the CCSD(T)//DFT(B3LYP) single-point calculations we carried
out, unavoidable inaccuracies in the geometrical parameters must
necessarily accompany the energy predictions. Bearing in mind
the above limitations, we consider that the degree of agreement
between theoretical and experimental rate constants reached in the
present work is rather acceptable.
Then,
BF3 + Me2O a BF3 · Me2O ∆G ) -5.1kcal/mol
On the other hand, according to our calculations (see Table
2 and Scheme 5),
BF3 + 0.5(13)2 a 0.5(19) ∆G ) -15.5kcal/mol
C. Desulfinylation in the Presence of SO2 or CF3CO2H. The
possibility of catalysis for the retro-ene reaction of 13 acid was
explored by analyzing the effect of the presence of additive A
such as SO2 or CF3CO2H. The structures located on the PES
(Table 3) suggest the mechanism in Scheme 12.
and consequently,
0.5(13)2 + BF3 · Me2O a 0.5(19) + Me2O ∆G )
-10.4kcal/mol
9
9560 J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009

Desulfinylation of Prop-2-enesulfinic Acid
A R T I C L E S
Therefore, the real stability of 19, with respect to products, when
employing BF3 ·Me2O instead of BF3, will be about 5 kcal/mol.
This result is a very relevant theoretical prediction which allows
us to interpret the experimental observation (see Table 4) that
the retro-ene reaction in the presence of BF3 ·Me2O is inhibited
as the amount of the latter increases. Indeed, the formation of
19 is a competitive process with respect to the retro-ene reaction
(see Scheme 8). The greater exergonicity for the formation of
19 will inhibit the formation of the retro-ene products (SO2 +
propene), in agreement with the experimental observation.
thiirane oxides, and that of SO3 which give products of ene-reaction
via ꢀ-sultone intermediates, the ene-reaction of SO2 with alkene
does not follow mechanisms in which either thiirane 1,1-dioxide
of ꢀ-sultane would be intermediates. In fact the retro-ene elimina-
tion of SO2 from prop-2-enesulfinic acid follows a converted
mechanisms in which the hydrogen atom transfer is rate-determin-
ing as for the thermal decarboxylation of but-3-enoic acid into
propene + CO2, and the thermal fragmentations of ꢀ,γ-unsaturated
carbonyl imine and diazene compounds into alkenes- + RR′CdO,
RR′CdNH, and N2, respectively. The exergonic ene-reaction of
SO2 with alkenes follows the same mechanism as the ene-reaction
of SeO2 with alkenes.
Conclusion
13 in CH2Cl2 solution at -15 °C equilibrates with a dimeric
form (13)2 which undergoes the retro-ene elimination of SO2
with production of propene and 1 equiv of 13. In the transition
state of this one-step, concerted reaction, one molecule of
sulfinic acid makes a hydrogen bond with the SdO moiety of
the molecule undergoing the pericyclic reaction. Although protic
acids such as CF3COOH, or Lewis acids such as SO2 are known
to catalyze other pericyclic reactions of SO2 such as the hetero-
Diels-Alder and cheletropic additions to 1,3-dienes, these acids
are not efficient catalysts of the retro-ene elimination of SO2.
In the presence of catalytic amount of BF3 ·Me2O the reaction
is almost not affected. When using a large amount of BF3 ·Me2O,
the reaction is completely inhibited because of the formation
of a stable tetramolecular complex of type (prop-2-enesulfinic
acid)2, (BF3)2 (19). Quantum calculations suggest concerted
retro-ene elimination of SO2 from prop-2-enesulfinic acid with
transition states in which the C(1) · · ·S bond elongation and the
S-O· · ·H· · ·C(3) transfer occurs in concert, and this indepen-
dently of the nature of the additive that interacts with the SdO
moiety. The favored transition states adopt six-membered,
chairlike structures in which the SdO moieties is gauche with
σ(C(1)-C(2)) bond of the propenyl unit (SdO occupies an
pseudoaxial position). This stabilizing gauche effect involving
a polar SdO and less polar C-C bond is also found in 13 in
CH2Cl2 solution, in its dimeric form (13)2. This is not the case
anymore in complex 19 in which the SOH/σ(C(1)-C(2)) are
gauche. Contrary to the reaction of SdO with alkenes that generate
Acknowledgment. We thank the Swiss National Science
Foundation, the Roche Research Foundation, the ”Secre´tariat d’Etat
a` l’Education et a` la Recherche (SER), the CSCS (Manno,
Switzerland), MEC (Madrid, Spain, Project No. CTQ 2007-67234-
C02-01/BQU), and FICYT (Gobierno del Principado de Asturias,
Project No. IB08-023) for financial support. A long time ago, Mr.
Florentino Fernandez performed a preliminary exploration of the
PESs corresponding to the retro-ene reaction of prop-2-enesulfinic
acid in the presence of SO2, BF3, HF, and H3O+ by employing
lower levels of theory. Mr Ruben Meana located some structures,
included in the Supporting Information section, corresponding to
an alternative mechanism for the case of the presence of TFA or a
second molecule of sulfinic acid. We thank all such contributions
at the very early stages of this research.
Supporting Information Available: Calculated CCSD(T)/6-
311 + G(d,p)//B3LYP/6-31 + G(d,g) absolute energy values
for prop-2-enesulfinic acid and its dimer, intermediates pre- and
post-transition state complexes and transition structures, in the
absence or presence of additives (SO2, BF3, CF3COOH) (Tables
S1-S5). Representations of all possible conformers of bimo-
lecular transition structures T4a-x, T5a-x (Figures S1-S2). 1H
NMR spectra of 13 + BF3 ·Me2O in CD2Cl2/toluene internal
reference. Experimental details on the preparation of 13 and
on the kinetic measurements and sample preparations (Figures
S3-S11). Complete ref 129. This material is available free of
charge via the Internet at http://pubs.acs.org.
(171) Maria, P. C.; Gal, J. F. J. Phys. Chem. 1985, 89, 1296–1304.
(172) McLaughlin, D. E.; Tamres, M. J. Am. Chem. Soc. 1960, 82, 5618–
5621.
JA901565S
9
J. AM. CHEM. SOC. VOL. 131, NO. 27, 2009 9561