Equilibrium and kinetic deuterium isotope effects on the hetero-Diels-Alder addition of sulfur dioxide

Article abstract of DOI:10.1002/anie.200351732

The C-S bond is formed to a greater extent than the C-O bond in the transition state of the [4+2] cycloaddition of S02 to dimethylidenecyclohexane (see scheme). Both deuterium isotope effects and quantum calculations lead to this conclusion.

Full text of DOI:10.1002/anie.200351732

Communications
Hetero-Diels–Alder Reaction
Equilibrium and Kinetic Deuterium Isotope
Effects on the Hetero-Diels–Alder Addition of
Sulfur Dioxide**
FrØdØric Monnat, Pierre Vogel,* RubØn Meana, and
JosØ A. Sordo*
At room temperature butadiene and alkyl-substituted 1,3-
dienes that can adopt the s-cis conformation add to sulfur
dioxide giving the corresponding sulfones (2,5-dihydrothio-
phene-1,1-dioxides),[1] which are about 10 kcalmolꢀ1 more
stable than the isomeric sultines (3,6-dihydro-1,2-oxathiin-2-
oxides).[2][3] In the presence of acid catalysts and at temper-
atures below ꢀ408C the latter equilibrate with the 1,3-dienes,
which results in hetero-Diels–Alder additions that are much
faster than the corresponding cheletropic additions.[4] The
hetero-Diels–Alder addition of 1,2-dimethylidenecyclohex-
ane (1) to SO2 does not require promotion by an acid
[Eq. (1)]. However, we have demonstrated that the reaction is
[*] Prof. Dr. P. Vogel, Dr. F. Monnat
Institute of Molecular and Biological Chemistry
Swiss Institute of Technology, BCH
1015 Lausanne-Dorigny (Switzerland)
Fax: (+41)21-693-93-75
E-mail: pierre.vogel@epfl.ch.
Prof. Dr. J. A. Sordo, R. Meana
Laboratorio de Química Computacional
Departamento de Química Física y Analítica
Universidad de Oviedo
Oviedo, Principado de Asturias (Spain)
Fax: (+34)98-523-78-50
E-mail: jasg@correo.uniovi.es
[**] This work was supported by the Swiss National Science Foundation,
the Centro Svizzero di Calcolo Scientifica (Manno), the SOCRATES
(Oviedo/EPFL) program, the SociØtØ Vaudoise des Sciences
Naturelles (Lausanne), and DGI (BQU2001-3600-C02-01, Madrid,
Spain).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3924
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200351732
Angew. Chem. Int. Ed. 2003, 42, 3924 –3927

Angewandte
Chemie
promoted by SO2 itself (rate law of second order in [SO2]).[5]
The unstable sultine 2 resulting from that cycloaddition has
been crystallized and analyzed by X-ray diffraction studies at
ꢀ1008C.[6] High-level quantum calculations[6,7] have sug-
gested that the hetero-Diels–Alder addition of SO2 to 1,2-
dimethylidenecyclohexane is concerted and asynchronous,
The rate laws d[4]/dt = k4[3][SO2]2, d[5]/dt = k5[3][SO2]2
(formation of regioisomeric sultines 4, 5), and d[6]/dt = k6[3]
[SO2]2 (formation of sulfolene 6) were followed between ꢀ75
and ꢀ548C and with a 10- to 40-fold excess of SO2 in CD2Cl2.
The third-order rate constants k4, k5, and k6 are reported in
Table 1. One finds a kinetic regioselectivity k4/k5 = 1.11 Æ 0.01
at ꢀ758C and 1.0798 Æ 0.0003 at ꢀ548C. The thermodynamic
ꢀ
ꢀ
with the C S bond formed to a greater extent than the C O
bond in the transition state. We confirm this hypothesis by
measuring the kinetic deuterium isotope effect[8] of the
reactions of 1 and its dideuterated derivative 3 with SO2.
Diene 3 equilibrates with the two regioisomeric adducts 4 and
5 (Scheme 1). The experimental kinetic isotope effects are
Table 1: Third-order rate constants [m2 sꢀ1] determined by 1H NMR
spectroscopy with toluene as an internal reference.
T [K]
k4 ”106
k5 ”106
k6 ”109
k4/k5
198
2.13Æ0.15
4.03Æ0.16
1.91Æ0.13
3.73Æ0.15
76.9Æ0.5
338Æ3
1.11Æ0.001
219.1
1.0798Æ0.003
regioselectivity given by [4]/[5] could not be measured at
ꢀ758C because equilibrium was reached too slowly. At
ꢀ548C, however, it amounts to [4]/[5] = 0.73 Æ 0.04 (no
change in this product ratio after more than 300 h at ꢀ548C).
We have also measured the relative rate constants for the
disappearance of dienes 1 and 3. The ratio of rate constants
k2/2k4 and k2/2k5 give the kinetic deuterium isotope effect for
the formation of sultines 4 and 5, respectively, where k2 is the
rate constant for the appearance of non-deuterated sultine 2.
This was done by following the reactions of a 1:1 mixture of
dienes 1 and 3 ([1 + 3] < 0.2 m, 2.5–3 m SO2 in CD2Cl2) at
ꢀ758C by 13C NMR spectroscopy (toluene as internal refer-
ence; because of the vicinal deuteration, the chemical shifts of
C1 in 1 and 2 are different). The ratio k4/k5 = 1.11 Æ 0.01
(Table 1) led to k2/2k4 = 0.89 Æ 0.04 (kH/kD for the formation
of 4) and k2/2k5 = 0.99 Æ 0.05 (kH/kD for the formation of 5) at
ꢀ758C.
Scheme 1. The reaction of SO2 with 3.
compared with those estimated by quantum calculations. We
have been able also to measure the deuterium isotope effect
on the equilibrium 4Ð5 and have found that it is opposite to
the kinetic deuterium isotope effect on the regioselectivity of
the addition. This is the first example of Diels–Alder
additions for which kinetic and equilibrium isotope effects
are compared.
Diene 3 (98% D2) was prepared as outlined in Scheme 2,
starting with the reduction of cis-cyclohexane-1,2-dicarbox-
ylic anhydride (7) with one equivalent of NaBD4 in THF at
08C. The resulting lactone 8 was reduced with LiAlH4 to give
diol 9. Esterification with 4-toluenesulfonyl chloride (TsCl)
gave 10, which underwent double elimination of toluenesul-
fonic acid to provide 3 (26% overall yield based on 7).
Equilibrium deuterium isotope effects have shown that C–
D bonds at sp3(C) centers are lower in energy than those at
sp2(C) centers .[9] It was thus expected that the transition
ꢀ
structure 11 (in which the C S bond is formed to a greater
ꢀ
extent than the C O bond) should be more stable than 12 and
the corresponding transition structure arising from the non-
ꢀ
deuterated dienes. On the contrary, if the C O bond should
be formed first, transition structure 13 is expected to be more
stable than 14 and the transition structures of the SO2 addition
to 1 (Scheme 3). Our experimental data (k4/k5 = 1.1, k2/2k4 =
0.89, and k2/2k5 = 0.99 at ꢀ758C) are consistent only with
transition structure 11, as predicted by quantum calcula-
tions.[2] Although the calculations did not locate any inter-
mediates along the reaction hypersurface of the hetero-
Diels–Alder addition of butadiene + 2SO2,[5] we cannot
exclude formally that 11 and 12 equilibrate with diradical or
zwitterionic intermediates before the formation of sultines 4
and 5, respectively.
Isotope effects were computed by means of the QUIVER
program,[10] which employs the Bigeleisen–Mayer formula-
tion[11] within the transition-state-theory approximation.[12]
Scaling factors of 0.943 [MP2/6-31G(d)] and 0.9614
[B3LYP6-31G(d)] were used in the calculations.[13] The
kinetic deuterium isotope effects calculated are reported in
Table 2 for the hetero-Diels–Alder addition of 1/3 + 2SO2
and of butadiene (15)/1,1-dideuterobutadiene (16) + 2SO2
following concerted mechanisms.
Scheme 2. Synthesis of diene 3. DMAP=dimethylaminopyridine,
DMSO=dimethyl sulfoxide, pyr.=pyridine.
Angew. Chem. Int. Ed. 2003, 42, 3924 –3927
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3925

Communications
Table 3: Calculated kinetic deuterium isotope effects for nonconcerted
hetero-Diels–Alder addition of SO2 + butadiene![COSOCH2CHCH-
CH2C][a].
Reaction
Ratio
T [8C]
ꢀ54
1.075
0.610
0.56
ꢀ75
1.086
0.568
0.52
+25
1.046[c]
0.727[c]
0.69
[b]
15, 16 + SO2!4’
kH/kD
k0H/k0D
[d]
15, 16 + SO2!5’
D
Kinetic regioselectivity[e]:
[a] Calculated (B3LYP6-31G(d)) energy barrier: 25.3 kcalmolꢀ1, relative
stability of the diradical intermediate: 18.5 kcalmolꢀ1, second energy
barrier: 19.1 kcalmolꢀ1. [b] kH/kD =k(15
+
SO2)/k(16
+
SO2!4’).
[c] B3LYP6-31G(d). [d] kH0 /kD0 D =k(15 + SO2)/k(16 + SO2!5’). [e] Ratio
(k0H/kD0 D)/(kH/kD).
Scheme 3. Transition structures leading to products 4 and 5.
Calculations predict smaller kH/kD values than those
found experimentally, but the calculated kinetic regioselec-
tivities k4/k5 are similar to the experimental values. This
ꢀ
confirms that the C S bond is formed to a greater extent than
ꢀ
the C O bond in the transition state of the hetero-Diels–
Alder additions of sulfur dioxide (see Figure 1). Calculations
ꢀ
made for the butadiene +SO2 reaction predict the C O bond
to be formed to a greater extent in the transition state of the
concerted mechanism. For the reaction involving a diradical
Figure 1. Calculated transition structure for the hetero-Diels–Alder
addition 1 + 2SO2. a) Top view, b) side view. Large pale balls=sulfur,
small black balls=oxygen, medium gray balls=carbon, small pale
balls=hydrogen. Distances in ꢀ.
¼
intermediate COSOCH2CH CH2CH2C calculations lead to k4/
k5 values much smaller than those observed (Table 3).
Furthermore, the “energy of concert”[14] (the difference
between the activation energies of the concerted and diradical
transition states) is ca. 13 kcalmolꢀ1 in favor of the concerted
mechanism.
respectively) agrees rather well with the experimental meas-
urements (0.73 Æ 0.04).
The theoretical prediction for the thermodynamic regio-
selectivity at the B3LYP6-31G(d) level of theory (0.758 and
0.718 for equilibria involving one and two molecules of SO2,
The equilibrium deuterium isotope effect that renders 5
more stable than 4 can be interpreted in terms of the
preference for dueterium to substitute at C–H moieties with
the highest stretching vibration energy. In sultines 4 and 5
¼
the acidifying effect of the S O moiety makes the C4–H
Table 2: Calculated kinetic deuterium isotope effects for concerted hetero-Diels–Alder
additions.
bond weaker than the C1–H bond. Thus deuterium prefers
C1, rendering 5 more stable than 4.The kinetic isotope
effects are not dominated by the thermodynamic isotope
Reaction
Ratio
T [8C]
ꢀ54
0.725
(0.830)[c]
0.869[b]
(0.841)[c]
1.10
effects (Dimroth principle: DH° = aDH
reinforcing the validity of the experimental test used
here to define the asynchronous character of the concerted
hetero-Diels–Alder additions of SO2.[14]
+ b), thus
ꢀ75
0.692
(0.746)
0.801
(0.763)
1.12
+25
[a]
1/3 + 2SO2![11]!4 + SO2
1/3 + 2SO2![12]!5 + SO2 k0H/kD0
Kinetic regioselectivity[i]:
kH/kD
(0.714)
0.776
0.816[b]
[d]
(0.734)
1.06
Received: April 23, 2003 [Z51732]
[e]
15/16 + 2SO2![11’]!4’ + SO2
kH/kD
0.707
(0.709)
0.695
(0.785)
0.693
(0.754)
0.780
(0.685)
0.741
0.829[b,j]
Keywords: ab initio calculations · cycloaddition ·
(0.673)
0.658
(0.806)[c]
0.796[f]
.
heterocycles · isotope effects · sulfur dioxide
(0.755)
(0.865)[g]
0.729
[h]
k0H/k0D
15/16 + 2SO2![12’]!5’ + SO2
0.824[b]
(0.721)
0.748
(0.647)
(0.842)[c,k]
0.861[f]
[1] G. de Bruin, Proc. K. Ned. Akad. Wet. 1914, 17, 585; H. J.
Backer, J. Strating, Recl. Trav. Chim. Pays-Bas 1943, 62, 815.
[2] B. Deguin, P. Vogel, J. Am. Chem. Soc. 1992, 114, 9210; T.
Fernµndez, D. Suµrez, J. A. Sordo, F. Monnat, E. Roversi, A.
Estrella de Castro, K. Schenk, P. Vogel, J. Org. Chem. 1998,
63, 9490; F. Monnat, P. Vogel, J. A. Sordo, Helv. Chim. Acta
2002, 85, 712.
[3] For early examples of hetero-Diels–Alder additions of SO2,
see: R. F. Heldeweg, H. Hogeveen, J. Am. Chem. Soc. 1976,
98, 2341; T. Durst, L. TrØteault-Ryan, Tetrahedron Lett. 1978,
(0.789[g]
[a] kH/kD =^ k1/2k4. [b] B3LYP6-31G(d). [c] B3LYP6-31G(d) for the reaction with only one
equivalent of SO2. [d] k0H/kD0 =^ k1/2k5. [e] kH/kD =^ butadiene
2SO2 vs. 1,1-D2-
+
butadiene + 2SO2 giving a sultine analogous to 4. [f] MP2/6-31G(d). [g] MP2/6-
31G(d) for reaction with only one equivalent of SO2. [h] k0H/kD0 =^ butadiene + 2SO2 vs.
1,1-D2-butadiene + 2SO2 giving sultine analogous to 5. [i] k0H/kD0 (5)/kH/kD(4) and kH0 /
k0DD(5’)/kH/kD(4’); calculated (B3LYP6-31G(d) including zero-point energy). [j] Energy
barrier: 4.8 kcalmolꢀ1. [k] Energy barrier: 12.2 kcalmolꢀ1
.
3926
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2003, 42, 3924 –3927

Angewandte
Chemie
2353; see also: J.-H. Liu, A.-T. Wu, M.-H. Huang, C.-W. Wu, W.-
S. Chong, J. Org. Chem. 2000, 65, 3395.
[4] E. Roversi, F. Monnat, P. Vogel, K. Schenk, P. Roversi, Helv.
Chim. Acta 2002, 85, 733; E. Roversi, R. Scoppelliti, E. Solari, R.
Estoppey, P. Vogel, P. Braæa, B. MenØndez, J. A. Sordo, Chem.
Eur. J. 2002, 8, 1336.
[5] T. Fernµndez, J. A. Sordo, F. Monnat, B. Deguin, P. Vogel, J. Am.
Chem. Soc. 1998, 120, 13276.
[6] D. Markovic, E. Roversi, R. Scoppelliti, P. Vogel, R. Meana, J. A.
Sordo, Chem. Eur. J., 2003, in press..
[7] D. Suµrez, T. L. Sordo, J. A. Sordo, J. Am. Chem. Soc. 1994, 116,
763; D. Suµrez, T. L. Sordo, J. A. Sordo, J. Org. Chem. 1995, 60,
2848.
[8] For examples of kinetic deuterium isotope effects used to
establish the synchronicity or asynchronicity of concerted Diels–
Alder reactions: M. Taagepera, E. R. Thornton, J. Am. Chem.
Soc. 1972, 94, 1168; J. J. Gajewski, K. B. Petersen, J. R. Kagel,
Y. C. J. Huang, J. Am. Chem. Soc. 1989, 111, 9078; D. A.
Singleton, S. R. Merrigan, B. R. Beno, K. N. Houk, Tetrahedron
Lett. 1999, 40, 5817; D. A. Singleton, S. R. Merrigan, A. A.
Thomas, Tetrahedron Lett. 1999, 40, 639; K. N. Houk, J.
Gonzµlez, Y. Li, Acc. Chem. Res. 1995, 28, 81; N. Chronakis,
M. Orfanopoulos, Org. Lett. 2001, 3, 545.
[9] J. J. Gajewski, N. D. Conrad, J. Am. Chem. Soc. 1979, 101, 6693.
[10] W. Saunders, K. E. Laidig, M. Wolfsberg, J. Am. Chem. Soc.
1989, 111, 8989.
[11] J. Bigeleisen, M. G. Mayer, J. Chem. Phys. 1947, 15, 261.
[12] K. J. Laidler, Chemical Kinetics, 3rd ed., Harper & Row, New
York, 1987.
[13] A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502.
[14] For other studies on the hetero-Diels–Alder addition mecha-
nisms, see e.g.: L. Kupczyk-Subotkowska, H. J. Shine, J. Am.
Chem. Soc. 1993, 115, 5296; G. Orlova, J. D. Goddard, J. Org.
Chem. 2001, 66, 4026; A. G. Leach, K. N. Houk, J. Org. Chem.
2001, 66, 5192; A. G. Leach, K. N. Houk, Chem. Commun. 2002,
1243.
Angew. Chem. Int. Ed. 2003, 42, 3924 –3927
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3927