Competition between hetero-Diels-Alder and cheletropic addition of sulfur dioxide. Theoretical and experimental substituent effects on the relative stability of 3,6-dihydro-1,2-oxathiin-2-oxides (sultines) and 2,5- dihydrothiophene-1,1-dioxides (sulfolenes). Anomeric effects in sultine and 6-substituted derivatives

Article abstract of DOI:10.1021/jo981679g

At low temperature and in the presence of CF₃COOH, SO₂ undergoes Diels-Alder additions with (E)-1-acetoxybutadiene (8d) giving a 1:10 mixture of diastereomeric 6-acetoxysultines (9d + 10d). The Van't Hoff plot for equilibria 8d + SO₂ ? 9d + 10d led to ΔH(r) = -7.0 ± 0.3 kcal/mol, ΔS(r) = -42 ± 3 cal·mol^-1·K ^-1. At 20 °C, 8d underwent a slow cheletropic addition with SO₂ giving 2-acetoxysulfolene (11d, ΔH(r) ? -11.5 kcal/mol), the structure of which was established by single-crystal X-ray diffraction studies. (E)-Chloro (8e) and (E)-bromobutadiene (8f) did not undergo Diels- Alder additions with SO₂, even in the presence of protic or Lewis acid promoters. Low yields of 2-chloro- (11e) and 2-bromosulfolene (11f) were obtained at 20 °C. The structure of 11e was confirmed by single-crystal X- ray diffraction. The potential energy hypersurfaces of the Diels-Alder and cheletropic additions of SO₂ to butadiene (8a), (E)-piperilene (8b), (E)-1- methoxy- (8c), (E)-1-acetoxy- (8d), and (E)-1-chlorobutadiene (8e) were studied by ab initio quantum calculations at the MP2/631G* level. In agreement with the experiment, 6-substituted sultines 9X and 10X were less stable than the corresponding 2-substituted sulfolenes 11X for X = Me, OAc, Cl. With X = OMe, the two diastereomeric 6-methoxysultines (9c, 10c) and 2- methoxysulfolene (11c) were calculated to have similar stabilities. This is attributed to a stabilizing thermodynamic anomeric effect or gem- sulfinate/methoxy disubstitution effect in 9c, 10c. Such effects were not detected for sulfinate/acetoxy (9d, 10d) and sulfinate/chloro (9e, 10e) disubstitutions. The relative instability of 2-acetoxy- (11d) and 2- chlorosulfolene (11e) compared with their cycloaddents is attributed to repulsive interactions between the SO₂ moieties and the 2-substituents. The Alder endo mode of [4 + 2] cycloaddition of SO₂ is predicted to be faster than the 'anti-Alder mode' of additions for dienes 8X, X = Me, OMe, OAc, Cl. The resulting diastereomeric sultines 9X and 10X, respectively, exist as equilibria (energy barrier: ca. 5-6 kcal/mol) of two conformers 9X ? 9'X, 10X ? 10'X. In general, the conformers 9X, 10X with pseudoaxial S=O group are preferred (conformational anomeric effect of the sulfinate moiety). Repulsive interactions between pseudoaxial S=O and polar cis-6-substituents (e.g.: X = OMe, OAc) in 9X may render conformers 9'X (with the S=O and 6-X groups in pseudoequatorial positions) as stable as conformers 9X. The calculations predict the existence of conformational anomeric effects of 2-3 kcal/mol for the gem-sulfinate/methoxy (9c, 10'c) and gem-sulfinate/acetoxy disubstitution (9d, 10'd).

Full text of DOI:10.1021/jo981679g

9490
J . Org. Chem. 1998, 63, 9490-9499
Com p etition betw een Heter o-Diels-Ald er a n d Ch eletr op ic
Ad d ition of Su lfu r Dioxid e. Th eor etica l a n d Exp er im en ta l
Su bstitu en t Effects on th e Rela tive Sta bility of
3,6-Dih yd r o-1,2-oxa th iin -2-oxid es (Su ltin es) a n d
2,5-Dih yd r oth iop h en e-1,1-d ioxid es (Su lfolen es). An om er ic Effects
in Su ltin e a n d 6-Su bstitu ted Der iva tives
Tino Ferna´ndez,† Dimas Sua´rez,† J ose´ A. Sordo,*,† Fre´de´ric Monnat,‡ Elena Roversi,‡
Antonio Estrella de Castro,‡ Kurt Schenk,‡ and Pierre Vogel*,‡
Departamento de Quı´mica Fı´sica y Analı´tica, Universidad de Oviedo, Oviedo, Principado de Asturias,
Spain, and Section de chimie de l’Universite´ de Lausanne, BCH-Dorigny,
CH-1015 Lausanne, Switzerland
Received August 18, 1998
At low temperature and in the presence of CF3COOH, SO2 undergoes Diels-Alder additions with
(E)-1-acetoxybutadiene (8d ) giving a 1:10 mixture of diastereomeric 6-acetoxysultines (9d + 10d ).
The Van’t Hoff plot for equilibria 8d + SO2 a 9d + 10d led to ∆Hr ) -7.0 ( 0.3 kcal/mol, ∆Sr )
-42 ( 3 cal‚mol-1‚K-1. At 20 °C, 8d underwent a slow cheletropic addition with SO2 giving
2-acetoxysulfolene (11d , ∆Hr = -11.5 kcal/mol), the structure of which was established by single-
crystal X-ray diffraction studies. (E)-Chloro (8e) and (E)-bromobutadiene (8f) did not undergo Diels-
Alder additions with SO2, even in the presence of protic or Lewis acid promoters. Low yields of
2-chloro- (11e) and 2-bromosulfolene (11f) were obtained at 20 °C. The structure of 11e was
confirmed by single-crystal X-ray diffraction. The potential energy hypersurfaces of the Diels-
Alder and cheletropic additions of SO2 to butadiene (8a ), (E)-piperilene (8b), (E)-1-methoxy- (8c),
(E)-1-acetoxy- (8d ), and (E)-1-chlorobutadiene (8e) were studied by ab initio quantum calculations
at the MP2/6-31G* level. In agreement with the experiment, 6-substituted sultines 9X and 10X
were less stable than the corresponding 2-substituted sulfolenes 11X for X ) Me, OAc, Cl. With X
) OMe, the two diastereomeric 6-methoxysultines (9c, 10c) and 2-methoxysulfolene (11c) were
calculated to have similar stabilities. This is attributed to a stabilizing thermodynamic anomeric
effect or gem-sulfinate/methoxy disubstitution effect in 9c, 10c. Such effects were not detected for
sulfinate/acetoxy (9d , 10d ) and sulfinate/chloro (9e, 10e) disubstitutions. The relative instability
of 2-acetoxy- (11d ) and 2-chlorosulfolene (11e) compared with their cycloaddents is attributed to
repulsive interactions between the SO2 moieties and the 2-substituents. The Alder endo mode of [4
+ 2] cycloaddition of SO2 is predicted to be faster than the “anti-Alder mode” of additions for dienes
8X, X ) Me, OMe, OAc, Cl. The resulting diastereomeric sultines 9X and 10X, respectively, exist
as equilibria (energy barrier: ca. 5-6 kcal/mol) of two conformers 9X a 9′X, 10X a 10′X. In general,
the conformers 9X, 10X with pseudoaxial SdO group are preferred (conformational anomeric effect
of the sulfinate moiety). Repulsive interactions between pseudoaxial SdO and polar cis-6-
substituents (e.g.: X ) OMe, OAc) in 9X may render conformers 9′X (with the SdO and 6-X groups
in pseudoequatorial positions) as stable as conformers 9X. The calculations predict the existence
of conformational anomeric effects of 2-3 kcal/mol for the gem-sulfinate/methoxy (9c, 10′c) and
gem-sulfinate/acetoxy disubstitution (9d , 10′d ).
In tr od u ction
1,3-diene (e.g., 1),7 SO2 gives at -78 °C adducts that
decompose above -20 °C into polymeric material. In the
presence of a Lewis acid, they (e.g., 2) engender zwitter-
ionic species (e.g., 3) that react with enoxysilanes (e.g.,
4), giving silyl sulfinates (e.g., 5), the hydrolyses of which
generate the corresponding â,γ-unsaturated sulfinic acids
At low temperature and in the presence of a suitable
protic or Lewis acid catalyst, simple 1,3-dienes add
reversibly to SO2 via a hetero-Diels-Alder reaction giving
the corresponding 3,6-dihydro-1,2-oxathiin-2-oxides (sul-
tines).1,2 Above -50 °C the sultines undergo fast cyclo-
reversion liberating the starting diene and SO2 that can
undergo cheletropic additions3 above -40 °C giving the
corresponding 2,5-dihydrothiophene-1,1-dioxide (sul-
folenes).4 With (E)-1-methoxybutadiene,5,6 (E)-2-methyl-
1-silyloxybutadiene, and (E,E)-2-methyl-1-silyloxypenta-
(3) Woodward, R. B.; Hoffmann, R. The Conservation of Orbital
Symmetry; Academic Press: New York, 1970; Turk, S. D.; Coob, R. L.
In “1,4-Cycloaddition Reactions; Hamer, J ., Ed.; Academic Press: New
York, 1967; p 13; Dewar, M. J . S. J . Am. Chem. Soc. 1984, 106, 209-
219 and preceding papers.
(4) De Bruin, G. Proc. K. Ned. Akad. Wet. 1914, 17, 585; Backer, H.
J .; Strating, J . Recl. Trav. Chim. Pays-Bas 1934, 53, 525; 1943, 62,
815.
(5) Deguin, B.; Vogel, P. Helv. Chim. Acta 1993, 76, 2250.
(6) Deguin, B.; Roulet, J .-M.; Vogel, P. Tetrahedron Lett. 1997, 38,
6197.
(7) Roulet, J .-M.; Puhr, G.; Vogel, P. Tetrahedron Lett. 1997, 38,
6201.
† Universidad de Oviedo.
‡ Universite´ de Lausanne.
(1) Deguin, B.; Vogel, P. J . Am. Chem. Soc. 1992, 114, 9210.
(2) Deguin, B.; Vogel, P. Tetrahedron Lett. 1993, 34, 6269; see also:
Heldeweg, R. F.; Hogeveen, H. J . Am. Chem. Soc. 1976, 98, 2341;
Durst, T.; Te´treault-Ryan, L. Tetrahedron Lett. 1978, 2353.
10.1021/jo981679g CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/13/1998

Hetero-Diels-Alder and Cheletropic Addition of SO2
J . Org. Chem., Vol. 63, No. 25, 1998 9491
Sch em e 1
also were carried out for the conformational analysis of
the diastereomeric 6-substituted sultines 9X a 9′X and
10X a 10′X with X ) Me, MeO, and AcO.
In what follows, the axial or equatorial nature of the
SdO bond and of the 6-substituent X in sultines is
designated by means of A or B, respectively, in paren-
theses (the first letter refers to the SdO bond, the second
to the substituent X).
Th eor etica l Meth od s
Reactants, products, and transition structures corre-
sponding to reactions I and II (Scheme 2) were fully
optimized at the MP2/6-31G* level of theory. All the
structures located on the potential energy hypersurface
were characterized (Hessian matrix) at the HF/6-31G*
level by means of vibrational frequency calculations. The
G2(MP2, SVP) theory introduced recently by Curtiss and
co-workers10 was used to improve the energetic results.
In this method, a basis set extension correction, ∆MP2,SV
,
is defined as
∆
MP2,SV ) E[MP2/6-311 + G(3df, 2p)] -
(e.g., 6) undergoing fast retro-ene elimination of SO2 at
20 °C. This led us to propose a new carbon-carbon-bond
forming procedure that can generate three contiguous
stereogenic centers in a one-pot procedure (Scheme 1) and
one (E)-alkene unit as shown with 1 + 4 f 7.7
To make progress with the chemistry of sultines, we
need to understand the substituent effects on activation
energies and on the thermodynamical parameters of the
hetero-Diels-Alder additions (reaction I) and of the
cheletropic additions (reaction II) of sulfur dioxide to (Z)-
1-substituted butadienes 8X.
This report presents experimental studies for the
reactions of SO2 with (E)-1-acetoxy- (8d ), (E)-1-chloro-
(8e), and (E)-1-bromobutadiene (8f), as well as ab initio
quantum calculations for systems with X ) H, Me, MeO,
OAc, Cl (Scheme 2). In a preliminary work,5 we had
assumed that 6-methoxy substitution should stabilize
sultine 9c more than sulfolene 11c because of a geminal
dioxy substitution effect (thermodynamic anomeric ef-
fect8). Our calculations supported this hypothesis. Nev-
ertheless, and contrary to expectation, no stabilizing gem-
sulfinate/acetoxy and gem-sulfinate/chloro substituent
effects have been found.
E(MP2/6-31G*) (1)
and the total G2(MP2, SVP) energy, E0, is given by
E0 ) E[QCISD(T)/6-31G*] + ∆MP2,SV
+
E(ZPE) + HCL (2)
where E(ZPE) is the zero-point vibrational energy, and
HCL (higher-level correction) is an empirical parameter
to account for remaining basis set deficiencies that have
no effect on the energy differences ∆E0 computed in the
present work.
Because the use of the quadratic configuration interac-
tion to compute the first term in eq 2 exceeded our
computer availabilities for X ) OMe and OAc (155 and
185 basis functions involved, respectively), the G2(MP2,
SVP) values were estimated in such cases by assuming
averaged values of the difference {∆E[QCISD(T)/6-31G*]
- ∆E(MP2/6-31G*)}obtained from calculations on X )
H, CH3, and Cl. Calculations show that, because of the
rather small variances involved, the approach should
work quite well (see Table 12 in the Supporting Informa-
tion).
The electrostatic solvent effects were estimated by
performing single-point calculations using a general self-
consistent reaction field (SCRF) model proposed for
quantum chemical computations on solvated molecules,11
where the solvent is represented by a dielectric con-
tinuum characterized by its relative static dielectric
permittivity ꢀ0.
All calculations were carried out using the Gaussian
94 packages of programs.12 ∆H° values were computed
within the ideal gas, rigid-rotor, and harmonic oscillator
The structures of the 6-alkylsultines have been inferred
1
from their H and 13C NMR spectra (dilute solutions in
SO2/CD2Cl2, low temperature).1,5 The data allowed rec-
ognition of the diastereomeric sultines but were not
sufficient to decide whether the SfO bond prefers the
pseudoequatorial (steric effect) or the pseudoaxial posi-
tions (conformational anomeric effect9). This question will
be examined by quantum calculations for the parent
sultine 9a a 9′a for which no experimental data are
available. If butadiene undergoes the hetero-Diels-Alder
addition with SO2 under protic acid- or Lewis acid-
catalyzed conditions, it does so very slowly. Calculations
(10) Curtiss, L. A.; Redfern, P. C.; Smith, B. J .; Radom, L. J . Chem.
Phys. 1996, 104, 5148. Curtiss, L. A.; Raghavachari, K.; Redfern, P.
C.; Pople, J . A. J . Chem. Phys. 1997, 106, 1063. See also: Curtiss, L.
A.; J ones, C.; Trucks, G. W.; Raghavachari, K.; Pople, J . A. J . Chem.
Phys. 1990, 93, 2537; Curtiss, L. A.; Raghavachari, K.; Pople, J . A. J .
Chem. Phys. 1993, 98, 1293.
(11) Rivail, J . L.; Rinalidi, D.; Ruiz-Lo´pez, M. F. In Theoretical and
Computational Models for Organic Chemistry, NATO ASI Series C,
Vol. 339, Formosinho, S. J .; Csizmadia, I. G., Eds., Kluwer Academic
Publishers: Dordrecht, 1991; Dillet, V.; Rinaldi, D.; Angya´n, J . G.;
Rivail, J . L. Chem. Phys. Lett. 1993, 202, 18; Dillet, V.; Rinaldi, J . L.
J . Phys. Chem. 1994, 98, 5034.
(8) Davidson, E. R.; Chakravorty, S.; Gajewski, J . J . New J . Chem.
1997, 21, 533 and references cited therein; Cramer, C. J .; Truhlar, D.
G.; French, A. D. Carbohydr. Res. 1997, 298, 1 and references cited
therein; Rakus, K.; Verevkin, S. P.; Peng, W.-H.; Beckhaus, H.-D.;
Ru¨chardt, C. Liebigs Ann. 1995, 2059; Beckhaus, H.-D.; Dogon, B.;
Verevkin, S. P.; Ha¨drich, H.; Ru¨chardt, C. Angew. Chem., Int. Ed. Engl.
1990, 29, 320; Hati, S.; Datta, D. J . Org. Chem. 1992, 57, 6056.
(9) J uaristi, E.; Cuevas, G. Tetrahedron 1992, 48, 5019 and refer-
ences cited therein; Tvaroska, I.; Bleha, T. Adv. Carbohydr. Chem.
Biochem. 1989, 47, 45.

9492 J . Org. Chem., Vol. 63, No. 25, 1998
Ferna´ndez et al.
Sch em e 2
approximations,13 as implemented in Gaussian 94. A
temperature of 298.15 K and a pressure of 1 atm were
assumed in the calculations.
estimate of ∆Hr (10b a 11b) = -8.3 kcal/mol. No
equilibrium constant could be measured for reaction 8c
+ SO2 a 9c (or 10c), because 8c was polymerized
concurrently with the formation of the adduct when
equimolar amounts of SO2 were used instead of large
excesses.
Resu lts
The reactions of SO2 with (E)-1-acetoxybutadiene (8d )
were better behaved. In the presence of CF3COOH, 8d
added to SO2 gave a 1:10 mixture (thermodynamic ratio)
of two diastereomeric sultines 9d /10d , the structures of
Experimental thermochemical data gave ∆H°r (8a +
SO2 a 11a ) ) -16.4 ( 1.0 kcal/mol,14 ∆H° (8b + SO2 a
r
11b) ) -14.9 kcal/mol,15 and ∆H° (isoprene + SO2
a
r
3-methyl-2,5-dihydrothiophen-1,1-oxide) ) -16.9 ( 1.1
kcal/mol in the gas phase. For the latter equilibrium in
benzene, Isaacs and Laila16 reported ∆Hr ) -15.6 kcal/
mol. For the hetero-Diels-Alder addition isoprene + SO2
a 4-methyl-3,6-dihydro-1,2-oxathiin-2-oxide, we mea-
sured an equilibrium constant K(I) ) 0.03 mol-1 dm3 at
-60 °C. Assuming ∆Sr(I) ) -42 e.u. (see below),17 one
calculates ∆Hr(I) ) -7.5 kcal/mol for this equilibrium.
Similar values of ∆Hr(I) and ∆Sr(I) can be retained for
8a + SO2 a 9a . One thus can propose a first estimate of
∆Hr (9a a 11a ) = -9 kcal/mol for the difference in
stability between the parent sulfolene 11a and sultine
9a .18 For equilibrium 8b + SO2 a 10b (most stable
diastereomer), K(I) ) 0.004 mol-1dm3 was measured at
-60 °C, allowing one to estimate ∆Hr (8b + SO2 a 10b)
) -6.6 kcal/mol with ∆Sr(I) ) -42 e.u. This leads to an
1
which were assigned from their H NMR characteristics.
Irradiation of the H-6 signal (δH ) 6.69 ppm) of the cis-
6-acetoxysultine 9d led to a positive NOE on H-3 (δH
)
3.36 ppm) and not on H′-3 (δH ) 3.84 ppm). The fact that
H′-3 is more deshielded than H-3 by 0.48 ppm is
consistent with an axial SdO bond syn with respect to
H′-3.19 This suggests therefore that conformer 9d is
favored over 9′d , as predicted by our calculations. Ir-
radiation of the H-6 signal (δH ) 6.48 ppm) of the major
trans-6-acetoxysultine 10d led to a positive NOE on the
signal assigned to H′-3 (δH ) 3.61 ppm) and not on H-3
(δH ) 3.56 ppm). In this case, the small chemical shift
difference observed for H-3 and H′-3 (∆δ ) 0.05 ppm) is
consistent with an equilibrium of conformers 10d a 10′d
in which both conformers are populated similarly, in
agreement with our calculations. Van’t Hoff plots (-70
to -30 °C) afforded ∆Hr(I) ) -7.0 ( 0.3 kcal/mol and
∆Sr(I) ) -42 ( 3 e.u. for equilibrium 8d + SO2 a 9d +
10d . The structure of 11d was established unambigu-
ously by single-crystal X-ray diffraction studies. At 25
°C, an equilibrium constant K (8d + SO2 a 11d ) ) 0.19
mol-1 dm3 was measured. Assuming ∆Sr(II) ) -42 e.u.
for this equilibrium, we estimate that ∆Hr (8d + SO2 a
11d ) = -11.5 kcal/mol. This gives an estimate of 4.5 kcal/
mol for the stability difference between sultines 9d + 10d
and sulfolene 11d , a value significantly smaller than that
found for the unsubstituted and alkyl-substituted ana-
logues (Table 2).
(12) Gaussian 94, Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill,
P. M. W.; J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T. A.;
Petersson, G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M.
A.; Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian,
Inc., Pittsburgh, PA, 1995.
(13) McQuarrie, D. A. Statistical Thermodynamics; University Sci-
ence Books: Mill Valley, CA, 1973.
(14) Cox, J . D.; Pilcher, G. Thermochemistry of Organic and Orga-
nometallic Compounds; Academic Press: New York, 1970.
(15) Drake, L. R.; Stowe, S. C.; Partanski, A. M. J . Am. Chem. Soc.
1946, 68, 2521.
(16) Isaacs, N. S.; Laila, A. A. R. J . Chem. Res., Synop. 1977, 10; J .
Chem. Res., Miniprint 1977, 188.
(17) Similar value measured for the hetero-Diels-Alder addition of
1,2-dimethylidenecyclohexane to SO2: Monnat, F.; Vogel, P. in prepa-
ration.
The (E)-1-chloro (8e) and (E)-1-bromobutadiene (8f)
had been reported not to add SO2.20 In our hands, no
(18) Acyclic sulfones RS(O)2R′ have been evaluated to be -7 to -15
kcal/mol more stable than their isomeric sulfinates RS(O)OR′: Lieb-
man, J . F.; Crawford, K. S. K. In Supplements: The Chemistry of
Sulphur-Containing Functional Groups, Patai, S.; Rapoport, Z., Eds.,
J . Wiley & Sons Ltd: New York, 1993; p 197.
(19) Bassindale, A. R.; Iley, J . N. In The Chemistry of Sulphinic
Acids Esters and their Derivatives; Patai, S., Ed.; J . Wiley & Sons
Ltd: New York, 1990; p 128.
(20) Klebanskii, A. L.; Sorokina, R. M.; Khavin, Y. Z. J . Gen. Chem.
USSR (Engl. Transl.) 1947, 17, 235; Chem. Abstr. 1948, 42, 514.

Hetero-Diels-Alder and Cheletropic Addition of SO2
J . Org. Chem., Vol. 63, No. 25, 1998 9493
F igu r e 1. Structures located on the MP2/6-31G* potential energy hypersurface for the cycloaddition reactions of 1,3-butadiene
and SO2 (see text and Scheme 2 for notation).
products resulting from hetero-Diels-Alder additions
were observed with these dienes between -100 and +40
°C, in the absence or in the presence of acid promoters.
Together with their polymerization, mediocre yield of
sulfolenes 11e (13%) and 11f (12%) were obtained after
long reaction times at 50 °C. This suggests that 1-chloro
and 1-bromo substitutions of butadiene render the het-
ero-Diels-Alder additions too slow or that they destabi-
lize the corresponding sultines, making them invisible
under our conditions. The structure of 11e was confirmed
by chemical correlation with 2-chlorosulfolane21 and by
X-ray diffraction studies (see Supporting Information).
The structure of 11f was confirmed by catalytical hydro-
genation (H2, 30 atm) that gave 2-bromosulfolane, a
compound identical with that obtained by bromination
(Br2/Ph-H) of sulfolane treated with 1.2 equiv of EtMgBr.
electron-deficient dienophiles, (E)-1-methoxybutadiene
(8c) is calculated to react significantly faster than all
other dienes 9X toward SO2 via TSIc (see Scheme 2). We
also find that the cheletropic addition 8c + SO2 a 11c is
easier than for the other dienes. Consistently with the
experiment, dienes 8a and 8e are very slow in both
reactions I and II. In (E)-1-acetoxybutadiene, the com-
puted energy barriers adopt intermediate values. It must
be pointed out that direct interconversions between
sultine 9X and corresponding sulfolenes 11X (reaction
III) can be discarded from the ab initio calculations.23 As
discussed elsewhere,24 the symmetries of diene and SO2
molecular orbitals allow for both HOMO(diene) f
LUMO(SO2) and HOMO(SO2) f LUMO(diene) interac-
tions during formation of the two C-S bonds of sul-
folenes, but only the HOMO(diene) f LUMO(SO2) in-
teraction contributes predominantly in the formation of
the CsS and CsO bonds of sultines. In agreement with
other experiments1,2,6 previous calculations predicted25
that Lewis acids would catalyze the hetero-Diels-Alder
additions better than the corresponding cheletropic ad-
dition of SO2 (reaction II).24 The notable increase in net
charge transfer from (HOMO) diene to (LUMO) dieno-
phile causes an important reduction in the activation
energy, thus favoring the formation of sultines at low
temperatures. On the other hand, as expected,26 our
calculations show somewhat lower energy barriers for
reactions in a solution than in a vacuum.
Ca lcu la tion s An d Discu ssion
Figures 1-5 show the structures located (transition
structures and products) computed at the MP2/6-31G*
level for reactions I and II in Scheme 2, including the
most relevant geometrical parameters (in X ) AcO,
Figure 4 includes, besides the four sultines depicted in
Scheme 2, four additional conformers arising as a con-
sequence of the greater complexity of this substituent).
Comparison between the calculated and experimental (X-
ray diffraction) geometries of sulfolenes 11d and 11e
generally showed good agreement, except for dihedral
angles that, as expected,22 were slightly less reliable (both
theoretical and experimental geometrical parameters are
provided as Supporting Information). Table 1 collects the
energetic results for all the transition structures, sul-
folenes, and the most stable sultine conformers obtained
by pathways I and II in Scheme 2.
From the thermodynamic viewpoint, we learn from
Table 1 that for 8c and 8d the exothermicities of
reactions I and II are rather similar, whereas for 8a , 8b,
and 8e the cheletropic additions of SO2 are more exo-
(23) Sua´rez, D.; Iglesias, E.; Sordo, T. L.; Sordo, J . A. J . Phys. Org.
Chem. 1996, 9, 17.
(24) Sua´rez, D.; Sordo, T. L.; Sordo, J . A. J . Am. Chem. Soc. 1994,
116, 763. Sua´rez, D.; Gonza´lez, J .; Sordo, T. L.; Sordo, J . A. J . Org.
Chem. 1994, 59, 8058.
From the kinetic viewpoint, data in Table 1 show that,
according to expectations for Diels-Alder additions with
(25) Sua´rez, D.; Sordo, T. L.; Sordo, J . A. J . Org. Chem. 1995, 60,
2848.
(26) Sua´rez, D.; Assfeld, X.; Gonza´lez, J .; Ruiz-Lo´pez, M. F.; Sordo,
T. L.; Sordo, J . A. J . Chem. Soc. Chem. Commun. 1994, 1683.
(21) Kattenberg, J . Tetrahedron 1973, 29, 4149.
(22) Hehre, W. J .; Radom, L.; Schleyer, P. v. R.; Pople, J . A. Ab Initio
Molecular Orbital Theory; J ohn Wiley & Sons: New York, 1986.

9494 J . Org. Chem., Vol. 63, No. 25, 1998
Ferna´ndez et al.
F igu r e 2. Structures located on the MP2/6-31G* potential energy hypersurface for the cycloaddition reactions of (E)-1-
methylbutadiene and SO2 (see text and Scheme 2 for notation).
thermic than the corresponding hetero-Diels-Alder ad-
ditions. Experimental data (Table 1) confirm that the
stability of the sulfolene relative to the sultine for diene
8d is considerably smaller than in dienes 8a , 8b. The
theoretical predictions reproduce the experimental data
available for the reactions studied (I, II, and III). The
accuracy of the computed reaction energies correspond
to what should be expected when using the G2(MP2,
SPV) theory in calculations involving hypervalent mol-
ecules such as SO2.27 Especially satisfactory are experi-
mental and theoretical data collected for equilibrium III.
For the parent sultine 9a (see Figure 1), the SdO bond
is predicted to occupy a pseudoaxial position; its con-
former 9′a with a pseudoequatorial SdO bond is calcu-
lated to be ca. 2.6 kcal/mol less stable at the G2(MP2,
SVP) level (the MP2/6-31G* value including zero-point
energy is 3.4 kcal/mol). We shall use this latter theoreti-
cal level in the conformational analysis of sultines 9X a
9′X and 10X a 10′X for X ) Me, MeO, AcO (Scheme 3
and Table 2). The axial preference for the SdO bond in
9a is consistent with the conformational anomeric effect
(27) Raghavachari, K.; Curtiss, L. A. In Modern Electronic Structure
Theory, Part II; Yarkony, D. R., Ed.; World Scientific: Singapore, 1995;
Chapter 14.

Hetero-Diels-Alder and Cheletropic Addition of SO2
J . Org. Chem., Vol. 63, No. 25, 1998 9495
F igu r e 3. Structures located on the MP2/6-31G* potential energy hypersurface for the cycloaddition reactions of (E)-1-
methoxybutadiene and SO2 (see text and Scheme 2 for notation).
(3.5 kcal/mol preference for axial SdO vs equatorial Sd
O) reported for 1,2-oxathian-2-oxide, the dihydro ana-
logue of 9a .28
most stable diastereomeric 6-methoxysultine (Table 1).
Because conformers 10c and 10′c are calculated to have
the same stability, we can admit that the stabilizing
anomeric effect of the SdO and 6-MeO groups are
similar. Their cumulative effect (as realized in 9c) is
larger than the SdO...OMe repulsion (in 9c) by 2.8 kcal/
mol, because 9′c in which the conformational anomeric
effects are absent is calculated to be 2.8 kcal/mol less
stable than conformer 9c.
Experiments show that 6-methylsultine 10b is the
most stable diastereomeric adduct (pseudoequatorial
methyl group).1 Our calculations provide the mechanistic
details (see Scheme 3). Conformer 9b is formed through
the lower energy transition structure TSIb. An alterna-
tive path via TS′Ib leads to the formation of the most
stable conformer 10b. Therefore, 9b is the kinetically
favored conformer, whereas 10b is formed when the
thermodynamic control is established, in full agreement
with experimental data.1 The formation of 10b through
the interconversions 9b a 10′b a 10b or 9b a 9′b a
10b implies barriers of ca. 70 kcal/mol [see Scheme 3;
transition structures TS(9b f 9′b), TS(10b f 10′b),
TS(9b f 10′b), and TS(9′b f 10b) connecting all
conformers are depicted in Figure 2]; they confirm the
hypothesis that the isomerization 9b a 10b occurs via
[4 + 2] cycloreversion and cycloadditions.
The experimental stability difference between ano-
meric 2-methoxytetrahydropyran amounts to 0.78 kcal/
mol in favor of the conformer with an axial OMe group.9
Our calculations for the 6-methoxysultines suggest that
a similar stabilizing conformational anomeric effect exists
for the gem-sulfinate/methoxy disubstitution but it must
compete with a destabilizing electrostatic interaction
when both the SdO and MeO moieties occupy cis-
pseudoaxial positions (Figure 3). Indeed, the MP2/6-31G*
calculations (Table 2) predict the diastereomer 9c is more
stable than 10c by 0.2 kcal/mol only, whereas the more
reliable G2 (MP2, SVP) method estimates that 10c is the
In the 6-acetoxysultines (Figure 4; more conformers
than pairs 9d a 9′d and 10d a 10′d are found on the
energy hypersurface of this system due to rotamers about
the O-COMe bond of the acetoxy substituent; only the
most stable rotamers are shown) our calculations (Tables
1 and 2) predict that sultine 10d is more stable than its
diastereomer 9d in agreement with our experiments.
Thus, in the 6-methyl-(9b, 10b) and 6-ethylsultines,5 the
trans-6-substituted sultine 10d is more stable than the
cis-isomer 9d . As in the case of the 6-methoxysultines
10c a 10′c one calculates that conformers 10d and 10′d
1
have the same stability,29 in agreement with our H NMR
data for this system (see above). In this case one also can
invoke a conformational anomeric effect of the SdO bond
of the sulfinate moiety to be similar to that of the gem-
sulfinate/methoxy system. 9d is calculated to be 3 kcal/
mol less stable than its diastereomer 10d . Conformer 9′d
is estimated to be only 0.5 kcal/mol more stable than 9d ,
consistent with our 1H NMR data. One can thus admit
that the repulsive interaction between the two cis-
pseudoaxial SdO and AcO groups in 9d overwhelms the
cumulative conformational anomeric effects of these
(29) Attempts to slow the interconversions of conformer pairs 9d a
9′d and 10d a 10′d failed because no significant line broadening could
be detected at -110 °C, the temperature at which two phases start to
form.
(28) Harp, D. N.; Gleason, J . G. J . Org. Chem. 1971, 36, 1314 and
references cited therein.

9496 J . Org. Chem., Vol. 63, No. 25, 1998
Ferna´ndez et al.
F igu r e 4. Structures located on the MP2/6-31G* potential energy hypersurface for the cycloaddition reactions of (E)-1-
acetoxybutadiene and SO2 (see text and Scheme 2 for notation).
groups by 3 kcal/mol (9′d does not benefit from any
conformational anomeric effect). Our calculations seem
to exaggerate the stability difference between 9d and
10d , because our experimental data give ∆G ) 0.9 kcal/
mol at 198 K for equilibrium 10d a 9d in SO2/CD2Cl2/
CF3COOH.
In the 6-chlorosultines, the MP2/6-31G* calculations
(Table 2) favor 9e over its diastereomer 10e by 0.2 kcal/
mol, whereas the G2(MP2, SVP) methods predict 10e to
be more stable than 9e by 1.2 kcal/mol (Table 1). We thus
must admit that repulsive SdO...Cl interactions in 9e
compete with the hypothetical conformational anomeric
effect of the gem-sulfinate/chloro system, or that the
latter does not exist.
The most striking outcome of our studies is the
observation that the gem-sulfinate/methoxy substitution
is accompanied by a thermodynamic anomeric effect8 that
is significantly larger than those of the gem-sulfinate/
acetoxy and gem-sulfinate/chloro systems. To confirm this
phenomenon further we have calculated the energies of

Hetero-Diels-Alder and Cheletropic Addition of SO2
J . Org. Chem., Vol. 63, No. 25, 1998 9497
F igu r e 5. Structures located on the MP2/6-31G* potential energy hypersurface for the cycloaddition reactions of (E)-1-
chlorobutadiene and SO2 (see text and Scheme 2 for notation).
Ta ble 1. Ca lcu la ted En er gies (k ca l/m ol)a of th e MP 2/6-31G* Op tim ized Tr a n sition Str u ctu r es (TS) a n d P r od u cts Rela tive
to Rea cta n ts for th e Diels-Ald er (I) a n d Ch eletr op ic (II) Rea ction s of 1-X-bu ta d ien e (X ) H, CH3, OMe, OAc, Cl) w ith
Su lfu r Dioxid e
b
c
c,d
reactions
TS
∆Eb
products
∆Hel
∆Hint
∆Hr
∆Hr(III)d,f
butadiene + SO2 (Ia)
TSIa
19.2 (-0.6)
19.0 (-1.1)
15.7 (-0.4)
21.3 (-0.2)
17.3 (-0.9)
11.5 (0.2)
15.2 (-0.3)
16.0 (-0.4)
14.5 (-0.1)
18.9 (-1.1)
18.5 (-0.9)
18.8 (-0.4)
23.6 (0.1)
9a
11a
9b
10b
11b
9c
10c
11c
9d
10d
11d
9e
-5.2 (-1.2)
-13.5 (-3.1)
-5.9 (-0.8)
-7.2 (-0.9)
-14.3 (-2.6)
-7.5 (-1.0)
-8.4 (0.2)
2.6
2.5
2.4
2.4
2.5
2.4
2.5
2.6
2.2
2.3
2.3
2.2
2.2
2.3
-2.6 [-7.6]e
-10.9 [-16.4]
-3.5
butadiene + SO2 (IIa)
TSIIa
TSIb
TS′Ib
TSIIb
TSIc
TS′Ic
TSIIc
TSId
TS′Id
TSIId
TSIe
-8.3 [-9.0]
(E)-piperilene + SO2 (Ib)
(E)-piperilene + SO2 (I′b)
-4.8 [-6.6]
-11.8 [-14.9]
-5.1
(E)-piperilene + SO2 (IIb)
-7.0 [-8.3]
-0.5
(E)-1-MeO-butadiene + SO2 (Ic)
(E)-1-MeO-butadiene + SO2 (I′c)
(E)-1-MeO-butadiene + SO2 (IIc)
(E)-1-AcO-butadiene + SO2 (Id)
(E)-1-AcO-butadiene + SO2 (I′d)
(E)-1-AcO-butadiene + SO2 (IId)
(E)-1-Cl-butadiene + SO2 (Ie)
(E)-1-Cl-butadiene + SO2 (I′e)
(E)-1-Cl-butadiene + SO2 (IIe)
-5.9
-9.0 (-2.5)
-3.0 (0.0)
-6.4
-0.8
-6.6 (-0.8)
-9.1 (-1.8)
-3.6 (-2.0)
-4.4 (-0.8)
-10.0 (-2.6)
-4.3 [-7.0]
-6.8 [-11.5]
-1.4
-2.5 [-4.5]
TS′Ie
TSIIe
10e
11e
-2.2
-7.7
19.6 (-0.5)
-5.5
a
b
G2(MP2,SVP) energies (see the text for more details). MP2/6-31G* SCRF estimate of the electrostatic solvent effect using a relative
permittivity of 13.3 to simulate the experimental conditions in parentheses. c Computed using HF/6-31G* ∆Hvib contributions scaled by
d
0.8905 and including ∆Hrot and ∆Htrans
.
Values obtained from experimental constants or from thermochemical data in parentheses.
e Experimental ∆Hr(I) value for isoprene.1,5 f Theoretical data correspond to the most stable conformer of the substituted sultine (see
text) in all cases.
Ta ble 2. Rela tive En er gies (MP 2/6-31G* Va lu es
In clu d in g Zer o-P oin t En er gy Cor r ection ; k ca l/m ol) of th e
Differ en t Con for m er s of 6-Meth oxysu ltin es,
6-Acetoxysu ltin es, a n d 6-Ch lor osu ltin es (See Text for
Nota tion )
disubstitution effect) of the 6-methoxy group in both
diastereomeric forms (compare ca.. -13 kcal/mol for OMe
with -6 to -7 kcal/mol for Me). Such an effect is not
detected for 6-acetoxy substitution of the sultines, whereas
a destabilizing gem-sulfinate/chloro effect might exist. No
stabilizing gem-MeO/sulfone effect can be seen with
reaction 11a + CH3OMe a 11c + CH4 (compare -5.5
kcal/mol for OMe with -6.1 kcal/mol for Me) confirming
that the relative stability of 2-methoxysulfolene (11c) is
similar to that of its isomeric sultines 9c and 10c (∆Hr-
(III) ) -0.5 kcal/mol; see Table 1). The extra stabilization
caused by the thermodynamic anomeric effect that
stabilizes the 6-methoxysultines (9c, 10c) does not
overwhelm the intrinsic greater relative stability of
diastereomers
cis a cis
trans a trans
9X (AA) 9′X (BB) 10X (AB) 10′X (BA)
6-methoxysultines (X ) c)
6-acetoxysultines (X ) d)
6-chlorosultines (X ) e)
(0.0)
3.0
(0.0)
2.8
2.5
3.1
0.2
(0.0)
0.2
0.4
0.0
0.3
the isodesmic reactions IV and V (Scheme 4). At the MP2/
6.31G* level (see Table 3) one confirms that the sultines
are stabilized by a thermodynamic anomeric effect (gem-

9498 J . Org. Chem., Vol. 63, No. 25, 1998
Ferna´ndez et al.
Sch em e 3. (MP 2/6-31G*) [∆Hr , ∆Eq in k ca l/m ol]
9X and 10X resulting from the hetero-Diels-Alder ad-
ditions of SO2 to (E)-piperilene (8b), (E)-1-acetoxybuta-
diene (8d ), and (E)-1-chlorobutadiene (8e) are less stable
than the corresponding 2-substituted sulfolenes 11X (X
) Me, OAc, Cl) resulting from the cheletropic additions
8X + SO2. With (E)-1-methoxybutadiene (8c), SO2 is
predicted to generate two diastereomeric 6-methoxysul-
tines 9c and 10c and 2-methoxysulfolene (11c) with
similar stabilities. This is attributed to a stabilizing
thermodynamic anomeric effect or gem-sulfinate/methoxy
disubstitution effect in 9c, 10c. There is no such effect
for the sulfone/methoxy disubstitution in 11c. In agree-
ment with our experiments, the calculations predict
destabilizing interactions in 2-acetoxysulfolene (11d ) and
2-chlorosulfolene (11e) caused by electrostatic repulsions
between the SO2 group and the 2-acetoxy and 2-chloro
substituents, respectively. Except for parent butadiene
(8a ), the uncatalyzed hetero-Diels-Alder additions of SO2
to 1,3-dienes (8b-8e) are predicted to be faster than the
corresponding cheletropic additions. The Alder mode of
addition which leads to cis-6-substituted sultines (9X a
9′X) is always predicted to be faster than the “anti-Alder”
mode of addition giving the diastereomeric trans-6-
substituted sultines (10X a 10′X). Sultines can adopt two
pseudo-chair conformations.
Sch em e 4
In the parent sultine, conformer 9a with a pseudoaxial
SdO bond is preferred over conformer 9′a with a pseu-
doequatorial SdO bond by ca. 3 kcal/mol, thus indicating
the existence of a conformational anomeric effect for the
sultinate moiety of the sultines. In the cis-6-methylsul-
tine (9b) with pseudoaxial SdO and methyl groups, the
barrier energy for the interconversion into its conformer
9′b (with pseudoequatorial SdO and methyl groups) is
estimated to be 5-6 kcal/mol. The steric factor makes
the cis-6-methylsultine (9b) about 1 kcal/mol less stable
than the trans-6-methylsultine (10b). The calculations
predict the existence of conformational anomeric effects
in 6-methoxy- (9c, 10c) and 6-acetoxysultines (9d , 10d ),
which amounts to ca. 3 kcal/mol. In the cis-diastereomers
9c and 9d electrostatic repulsions between the pseudo-
axial SdO, OMe, and OAc groups, respectively, are
competing with the conformational anomeric effects and
render the trans-diastereomeric sultine 10c as stable as
9c. For the 6-acetoxy-substituted analogue, the trans-
isomer 10d is estimated to be more stable than the cis-
sultine 9d in agreement with our experiments. The
pseudoaxial SdO/pseudoaxial OAc repulsion is evaluated
to be somewhat more important than the pseudoaxial Sd
O/pseudoaxial OMe repulsion. Consequently, conformers
10c and 10′c have similar stabilities. Conformer pairs
9d a 9′d and 10d a 10′d are calculated to have similar
Ta ble 3. MP 2/6-31G* Rea ction En er gies (k ca l/m ol) for
Isod esm ic P r ocesses IV a n d V (See Text) In volvin g
1-X-Bu ta d ien e Su ltin es a n d Su lfolen es (X ) CH3, OMe,
OAc, Cl)
CH3 OMe OAc
(b) (c) (d)
Cl
(e)
reactions
X
9a + CH3X a 9X + CH4
9a + CH3X a 10X + CH4
∆E(IV) -6.2 -13.4 -5.6 -4.7
∆E(IV) -7.2 -13.2 -8.6 -4.5
11a + CH3X a 11X + CH4 ∆E(V) -6.1
-5.5 -3.0 -1.2
2-methoxysulfolene (11c). Why only one adduct of (E)-
1-methoxybutadiene (8c) with SO2 has been observed so
far is still unexplained. Is it a sultine or a sulfolene?
Further experimental studies are now required to
confirm our preliminary work.5 It is possible that sul-
folene 11c has not been observed together with sultine
9c because the rate of formation of 11c is significantly
smaller than that of 9c as predicted (Table 2). Because
8c + SO2 a 9c decomposed quickly above -20 °C, the
conditions were not realized that would allow for 11c to
be formed concurrently with the polymerization of diene
8c. The calculated energy for equilibrium Vd (-3.0 kcal/
mol; Table 3) suggests a destabilizing effect of the acetoxy
substituent in the 2-acetoxysulfolene 11d , as found
experimentally (∆Hr(IIId) ) -11.5 kcal/mol compared
with ∆Hr(IIIb) ) -14.9 kcal/mol; Table 2). A somewhat
larger destabilization effect is calculated for the chloro
substituent, in agreement with our experiments. This can
be interpreted in terms of repulsive electrostatic interac-
tions between the SO2 moiety and the AcO, Cl substit-
uents of the sulfolenes (Figures 4 and 5).
1
relative stabilities, in agreement with H NMR data.
Exp er im en ta l Section
NMR Sa m p le p r ep a r a tion a n d Th er m och em ica l P a -
r a m eter s. CD2Cl2 and toluene were distilled over anhyd CaH2,
CF3COOH over P2O5. SO2 was filtered through a column of
alcaline aluminum oxide 90 (activity I, Merck) before use. In
a weighed, dry 5-mm NMR Pyrex tube, dienes 8X (0.05-0.3
mmol), toluene (internal reference, 5-15 mg), CD2Cl2 (0.2-
0.3 g), and CF3COOH (0.01-0.2 mmol; catalyst, if any) were
mixed under an Ar atmosphere at 20 °C. The solution was
degassed by several freeze-thaw cycles at 10-2 Torr. Degassed
SO2 (0.1-0.4 mL) was transferred to the above mixture on the
vacuum line. The NMR tube was sealed under vacuum. The
NMR tube, frozen in liquid N2, was warmed to -100 °C in
Our calculations suggest the existence of hydrogen
bondings of the type Me...OSO (see 11b, Figure 2) and
OMe...OSO (see 11c, Figure 3) for 2-methylsulfolene and
2-methoxysulfolene, respectively.
Con clu sion
Experiments and high-level ab initio quantum calcula-
tions agree that the diastereomeric 6-substituted sultines

Hetero-Diels-Alder and Cheletropic Addition of SO2
J . Org. Chem., Vol. 63, No. 25, 1998 9499
4
liquid EtOH/liquid N2 and transferred into the Bruker ARX-
400 spectrometer probe cooled to -90 °C. Other tubes were
left in thermostated EtOH baths at various temperatures until
equilibria were reached, then cooled to -100 °C, and trans-
ferred into the spectrometer probe cooled to -90 °C. When
viscous solutions were obtained at temperatures lower than
-70 °C, mixtures of CD2Cl2/CFCl3 were used instead of pure
CD2Cl2. When the equilibrium constant measurements were
terminated, the NMR tube was allowed to reach room tem-
perature and was weighed together with the piece of tube left
over after sealing: this allowed verification of the exact amount
of SO2 introduced in the diene solution. Spectra were recorded
for 90° pulses with a delay of 10-15 s between pulses, same
spectral window (16-32 scans). The equilibrium constants at
a given temperature were evaluated for at least three different
NMR samples and had reproducible values within experimen-
tal error, varying the initial cycloaddent ratio. Enthalpies and
entropies of reactions were obtained through Van't Hoff plots,
the equilibrium constants being determined by using the
concentrations of the products and reactants.
) 2.6, J (H-C(3),H-C(5b)) ) 2.4, H-C(3)), 6.43 (dddd, 1H,
3J (H-C(3),H-C(4)) ) 8.8, 3J (H-C(4),H-C(5b)) ) 3.0, 3J (H-
C(4),H-C(5a)) ) 2.9, J (H-C(4),H-C(2)) ) 1.2, H-C(4)).
4
2-Ch lor o-2,5-d ih yd r oth iop h en e-1,1-d ioxid e (11e). The
procedureis the same as for the preparation of 11d , starting
from (E)-1-chlorobutadiene30 (8e, 5 g, 0.057 mol) and heating
to 50 °C for 12 h. Yield: 1.14 g (13%), yellowish oil that
crystallizes from 1:1 Et2O/pentane: colorless crystals used for
the X-ray analysis, mp 40-42 °C. 1H NMR (400 MHz,
3
CDCl3)δH 3.85 (m, H-C(5)), 5.28 (m, H-C(2)), 6.26 (dddd, J
(H-3,H-4) ) 8.6, 3J (H-3,H-2) ) 3J (H-3,H-5) ) 4J (H-3,H-5) )
2.0, H-C(3)), 6.29 (dddd, 3J (H-3,H-4) ) 8.6, 3J (H-4,H-5) )
3.0, 3J (H-4,H-5) ) 2.5, 4J (H-2,H-4) ) 1.1, H-C(4)); 13C NMR
(100.6 MHz, CDCl3)δC 53.3 (t, 145, C(5)), 69.7 (d, 170, C(2)),
127.4 (d, 176, C(4)), 128.7 (d, 179, C(3)).
2-Br om o-2,5-d ih yd r oth iop h en e-1,1-d ioxid e (11f). The
procedure is the same as for the preparation of 11e, starting
from (E)-1-bromobutadiene30 (8f, 1 g, 0.0075 mol). Yield: 0.16
1
2
g (11%), yellow oil. H NMR (400 MHz, CDCl3)δH 3.83 (m, J
3
4
) 16, J (H-C(5), H-C(4)) ) 2.5, J (H-C(5), H-C(2)) ) 2.5,
4J (H-C(5), H-C(3)) ) 1.5, H-C(5)); 3.9 (m, J ) 16, J (H′-
2
3
Da ta for 6-Acetoxy-3,6-d ih yd r o-1,2-oxa th iin -2-oxid es
(9d a 9′d and 10d a 10′d ). In the presence of one equiv, of
CF3COOH, (E)-1-acetoxybutadiene (8d ) reacted with SO2 at
-75 °C to give a 1:10 mixture of two diastereomeric sultines.
This ratio was the same at the beginning and at the end of
the reaction and at higher temperatures, when cycloreversion
4
5
C(5), H-C(4)) ) 3, J (H′-C(5), H-C(3)) ) 2.1, J (H′-C(5),
5
H-C(2)) ) 0.9, H′-C(5)); 5.4 (m, J (H-C(2), H-C(5)) ) 1.6,
5J (H-C(2), H′-C(5)) ) 0.9, H-C(2)); 6.23 (m, 3J (H-C(3),
H-C(4)) ) 8.6, 4J (H-C(3), H′-C(5)) ) 0.9, 4J (H-C(3),
3
H-C(5)) ) 2.5, H-C(3)); 6.32 (m, J (H-C(4), H-C(3)) ) 8.6,
1
3J (H-C(4), H′-C(5)) ) 3, 3J (H-C(4), H-C(5)) ) 2.5, H-C(4)).
13C NMR (100.6 MHz, CDCl3)δC 57.7 (t, 145), 58.3 (t, 155),
126.6 & 129.5 (2d, 176).
was observed. Equilibrium was reached after 10 h. H NMR
(400 MHz, CD2Cl2/SO2, 198 K) of the major sultine 10d : δH
2.23 (s, Ac), 3.56 and 3.61 (2m, H2C(3)), 6.14 (m, H-C(5)), 6.26
(m, H-C(4)), 6.48 (m, H-C(6)), 3J (H-5,H-6) = 4J (H-4,H-6) )
2.4 ( 0.2 Hz, 3J (H-4,H-5) ) 11.6 ( 0.2 Hz (through homo-
nuclear decoupling); 13C NMR (100.6 MHz, CD2Cl2/SO2, 198
2-Ch lor osu lfola n e. Sulfolene 11e (88 mg, 0.6 mmol) in
CH2Cl2 (3 mL) was pressured with H2 (1 atm) and stirred at
20 °C for 1 h in the presence of 10% Pd/C (20 mg). After
filtration (Celite) and solvent evaporation, flash chromatog-
raphy on silica gel (CH2Cl2) gave 53 mg (60%) of a colorless
oil identical (spectral data) with a sample of 2-chlorosulfolane
prepared from sulfolane according to Kattenberg.21
2-Br om osu lfola n e. A mixture of sulfolene 11f (150 mg,
0.09 mmol), MeOH (3 mL), and 10% Pd/C (20 mg) was
degassed on the vacuum line and pressurized with H2 (30 bar).
After shaking at 25 °C for 12 h, the catalyst was filtered off,
the solvent evaporated, and the residue purified by flash
chromatography on silica gel (CH2Cl2) giving 40 mg (25%) of
yellowish oil identical (spectral data) with the 2-bromosul-
folane prepared according to Faith et al.31
1
K)δC: 20.6 ppm (q, J (C,H) ) 131 Hz, Ac), 45.2 (t, 142, C(3)),
84.0 (d, 180, C(6)), 119.5 (d, 176, C(4)), 122.8 (d, 171, C(5)),
172.5 (s, COO); assignments confirmed by 1H/13C-correlated
two-dimensional spectra HETCOR and HSQC. Equilibrium
constant K (298 K) ) 0.022 ( 0.006 dm3mol-1. Data for 9d :
1H NMR (400 MHz, CD2Cl2/SO2, 198 K)δH 2.36 (s, Me), 3.36
(ddt, H-C(3)), 3.84 (ddt, H′-C(3)), 6.00 (dq, H-C(5)), 6.69 (m,
2
3
H-C(6)), 6.17 (m, H-C(4), J (H-3,H′-3) ) 17.3, J (H-3,H-4)
) 4.8, 3J (H′-3,H-4) ) 4.0, 3J (H-4,H-5) ) 11.2, 3J (H-5,H-6) )
1.9, 4J (H-3,H-5) ) J (H′-3,H-5) ) 2, 4J (H-4,H-6) ) 1.3, 5J
4
(H-3,H-6) ) J (H′-3,H-6) ) 2.1; 13C NMR (100.6 MHz, CD2-
5
Cl2/SO2, 198 K)δC 20.6 (q, 131, Ac), 46.7 (d, 142, C(3)), 87.6 (d,
180, C(6)), 122.0 (d, 171, C(4)), 123.7 (d, 171, C(5)), 173.0 (s,
COO), assignments confirmed by HETCOR and HSQC. Equi-
librium constants for reaction 8d + SO2 a 9d + 10d : K )
0.103 ( 0.005 dm3mol-1 (at 183 K), 0.0394 ( 0.004 (193 K),
0.0244 ( 0.004 (198 K), 0.0158 ( 0.004 (203 K), 0.0110 (
0.0034 (208 K), 0.0080 ( 0.0031 (213 K), 0.0037 ( 0.0006 (223
K), 0.0020 ( 0.00012 (233 K), 0.0009 ( 0.00006 (243 K).
2-Acetoxy-2,5-d ih yd r oth iop h en e-1,1-d ioxid e (11d ). A
mixture of (E)-1-acetoxybutadiene (1 g, 8.9 mmol) and SO2
(3.97 g, 62 mmol, 7 equiv) was placed in a Pyrex tube, degassed
on the vacuum line. After sealing the tube, the mixture was
left at 20 °C for 6 days. After cooling in liquid N2, the tube
was opened and the SO2 evaporated. The residue was purified
by flash chromatography on silica gel (EtOAc/light petroleum
1:1) yielding 671 mg (3.8 mmol, 43%) of a colorless solid that
recrystallized from CH2Cl2/pentane, yielding colorless needles
used for the X-ray diffraction studies. In solution, 11d is slowly
decomposed into (E)-1-acetoxybutadiene and SO2 at 20 °C. It
Ack n ow led gm en t . We thank the Swiss National
Science Foundation, the “Centro Svizzero di Calcolo
Scientifico” (Manno), the “Centre Informatique” of the
University of Lausanne, and FICYT (Principado de
Asturias) for financial support.
Su p p or tin g In for m a tion Ava ila ble: Data for 11d , 11e,
11f. Tables of bond lengths, bond angles, and dihedral angles
for crystalline (X-ray) and MP2/6-31G* structures 11d , 11e.
Total energies for all the structures located as computed at
the different levels of theory, and ∆E[QCISD(T)/6-31G*] -
∆E[MP2/6-31G*] values corresponding to the MP2/6-31G*
optimized transition structures and products for the hetero-
Diels-Alder addition and the cheletropic addition of SO2 to
(E)-1-X-butadiene (X ) H, Me, and Cl) (19 pages). This
material is contained in libraries on microfiche, immediately
follows this article in the microfilm version of the journal, and
can be ordered from the ACS; see any current masthead page
for ordering information.
1
is stable in the crystalline form. mp 65-68 °C. H NMR (400
MHz, CDCl3)δH 2.23 (s, 3H, Ac), 3.77 (dddd, 1H, 2J (H-
C(5a),H-C(5b)) ) 17.2, 3J (H-C(4),H-C(5b)) ) 3.0, 4J (H-
5
C(3),H-C(5b)) ) 2.4, J (H-C(2),H-C(5b)) ) 1.1, H-C(5b)),
3.80 (dddd, 1H, 2J (H-C(5a),H-C(5b)) ) 17.2, 3J (H-C(4),H-
C(5a)) ) 2.9, 4J (H-C(3),H-C(5a)) ) 2.6, 5J (H-C(2),H-C(5a))
J O981679G
3
) 2.4, H-C(5a)), 5.93 (dddd, 1H, J (H-C(3),H-C(2)) ) 2.9,
(30) Keegstra, M. A.; Verkruijsse, H. D.; Andringa, H.; Brandsma,
L. Synth. Commun. 1991, 721.
(31) Faith, H. E.; Kautsky, M. P.; Abreu, B. E. J . Org. Chem. 1962,
27, 2889.
5J (H-C(2),H-C(5a)) ) 2.4, 4J (H-C(4),H-C(2)) ) 1.2, 5J (H-
C(2),H-C(5b)) ) 1.1, H-C(2)), 6.16 (dddd, 1H, 3J (H-C(3),H-
3
4
C(4)) ) 8.8, J (H-C(3),H-C(2)) ) 2.9, J (H-C(3),H-C(5a))