Theoretical Prediction and Experimental Tests of Conformational Switches in Transition States of Diels-Alder and 1,3-Dipolar Cycloadditions to Enol Ethers

Article abstract of DOI:10.1021/jo971408q

Transition structures for the cycloadditions of butadiene, acrolein, nitrosoethylene, and methyl-enenitrone to 1-butene, silyl vinyl ether, and methyl vinyl ether have been located using ab initio RHF theory with the 3-21G basis set and with density funct

Full text of DOI:10.1021/jo971408q

1064
J . Org. Chem. 1998, 63, 1064-1073
Th eor etica l P r ed iction a n d Exp er im en ta l Tests of Con for m a tion a l
Sw itch es in Tr a n sition Sta tes of Diels-Ald er a n d 1,3-Dip ola r
Cycloa d d ition s to En ol Eth er s
J ian Liu, Satomi Niwayama, Ying You, and K. N. Houk*
Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095
Received J uly 30, 1997
Transition structures for the cycloadditions of butadiene, acrolein, nitrosoethylene, and methyl-
enenitrone to 1-butene, silyl vinyl ether, and methyl vinyl ether have been located using ab initio
RHF theory with the 3-21G basis set and with density functional theory using the Becke3LYP
functional and the 6-31G* basis set. The computational results show that there is a switch in the
conformation of the enol ether from syn (COCC ) 0°), which is favored by 2.3 kcal/mol in the
reactant, to anti (COCC ) 180°), which is favored by 1.2-6.6 kcal/mol in the various transition
structures studied here. The results are consistent with the experimental stereoselectivities in
reactions of chiral enol ethers observed by Denmark and Reissig. The preference of the anti
conformation in the transition structures is due primarily to electrostatic effects and, to a lesser
extent, steric effects. The preference is predicted to be influenced significantly by polar solvents.
The magnitude of this preference was calculated theoretically and measured experimentally: the
rates of cycloadditions of conformationally fixed enol ethers, 2,3-dihydrofuran and 2-methylene-
tetrahydrofuran, with 1-nitroso-1-phenylethylene and with C-benzoyl-N-phenylnitrone were studied.
Observed relative rates are in good agreement with prediction and confirm that enol ethers adopt
the s-trans conformation in transition states of cycloadditions.
Sch em e 1
In tr od u ction
Enol ethers have been used in inverse electron-demand
Diels-Alder and 1,3-dipolar cycloadditions for the syn-
thesis of five- and six-membered ring systems in natural
products.1 Previous theoretical studies on the stereose-
lectivity focused mainly on the endo/exo selectivity of
these cycloaddition reactions.2 The goal of our investiga-
tion was to determine the effect of dienophile conforma-
tions on the rates and stereoselectivities of these types
of cycloadditions. We have discovered a general phe-
nomenon for the cycloaddition reactions of enol ethers,
in which a clear-cut “conformational switch” occurs
during the course of reaction.3 Thus, the ground-state
conformation of the reactant is changed to a dramatically
different conformation in the transition structure (Scheme
1). This conformational switch reverses the sense of
π-facial stereoselectivity of reactions of chiral enol ethers
from that which would be expected on the basis of facial
accessibility in the ground state. The results provide an
explanation for the stereoselectivities of cycloadditions
of chiral enol ethers reported by Denmark et al., Reissig
et al., and others.4
The stereoselectivities of addition reactions are some-
times rationalized in terms of steric or torsional effects
involving the reactant conformation.5 This explanation
is satisfactory for reactions when the conformation of the
transition structures mimics that of the reactants; how-
ever, the Curtin-Hammet principle shows that this will
not always be the case.6 There are many examples where
consideration of the ground state leads to successful
rationalization of the reaction stereoselectivity, but there
(1) Diels-Alder reactions: (a) Denmark, S. E.; Thorarensen, A.
Chem. Rev. (Washington, D.C.) 1996, 96, 137 and references therein.
(b) Boger, D. L. Chemtracts 1996, 9, 149 and references therein. (c)
Denmark, S. E.; Dappen, M. S.; Sternberg, J . A. J . Org. Chem. 1984,
49, 4741. (d) Tietze, L.-F.; Gluesenkamp, K.-H.; Harms, K.; Remberg,
G. Tetrahedron Lett. 1982, 23, 1147. (e) Arnold, T.; Orschel, B.; Reissig,
H.-U. Angew. Chem., Int. Ed. Engl. 1992, 31, 1033. (f) Hippeli, C.;
Reissig, H.-U. Liebigs Ann. Chem. 1990, 217. (g) Zimmer, R.; Reissig,
H.-U. Liebigs Ann. Chem. 1991, 553. (h) Zimmer, R.; Reissig, H.-U. J .
Org. Chem. 1992, 57, 339. (i) Fisera, T.; Dandarava, M.; Kovac, J .;
Gaplovsky, A.; et al. Collect. Czech. Chem. Commun. 1982, 47, 523.
1,3-Dipolar cycloadditions: (j) Nitrones, Nitronates and Nitroxides;
Wiley: New York, 1990; pp 1-104. (k) Annunziata, R.; Cinquini, M.;
Cozzi, F.; Raimondi, L. Gazz. Chim. Ital. 1989, 119, 253.
(2) (a) Desimoni, G.; Colombo, G.; Righetti, P. P.; Tacconi, G.
Tetrahedron 1973, 29, 2635. (b) Reissig, H.-U.; Hippeli, C.; Arnold, T.
Chem. Ber. 1990, 123, 2403. (c) McCarrick, M. A.; Wu, Y.-D.; Houk,
K. N. J . Org. Chem. 1993, 58, 3330. (d) Gouverneur, V. E.; Houk, K.
N.; Pascual-Teresa, B.; Beno, B.; J anda, K. D.; Lerner, R. A. Science
1993, 262, 204. (e) Rastelli, Q.; Bagatti, M.: Gandolfi, R.; Burdisso,
M. J . Chem. Soc., Faraday Trans. 1994, 90, 1077. (f) Sisti, N. J .;
Motorina, I. A.; Dau, M. T. H.; Riche, C.; Fowler, F. W.; Grierson, D.
S. J . Org. Chem. 1996, 61, 3715. (g) Dharale, D. D.; Trombini, C. J .
Chem. Soc., Chem. Commun. 1992, 1268.
(4) (a) Arnold, T.; Reissig, H.-U. Synlett 1990, 514. (b) Denmark, S.
E.; Thorarensen, A. J . Org. Chem. 1994, 59, 5672. (c) Denmark, S. E.;
Schnute, M. E.; Marcin, L. R.; Thorarensen, A. J . Org. Chem. 1995,
60, 3205. (d) Denmark, S. E.; Thorarensen, A.; Middle, D. S. J . Org.
Chem. 1995, 60, 3574.
(5) (a) Toromanoff, E. Tetrahedron 1981, 37, 3141. (b) Bucourt, R.
Top. Stereochem. 1974, 8, 159.
(3) Other conformational switches can be found in nucleophilic
additions to carbonyl compounds (Coxon, J . M.; Houk, K. N.; Luibrand,
R. T. J . Org. Chem. 1995, 60, 418 and references therein) and in nitrile
oxide cycloadditions to chiral alkenes (Houk, K. N.; Duh, H.-Y.; Wu,
Y.-D.; Moses, S. R. J . Am. Chem. Soc. 1986, 108, 2754).
(6) Seeman, J . I. Chem. Rev. (Washington, D.C.) 1983, 83, 84.
S0022-3263(97)01408-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/30/1998

Experimental Tests of Conformational Switches
J . Org. Chem., Vol. 63, No. 4, 1998 1065
Ta ble 1. Ca lcu la ted En er gies (a u ) a n d Rela tive En er gies
(k ca l/m ol, in P a r en th eses) of s-Cis a n d s-Tr a n s Alk en es
by RHF /3-21G a n d Beck e3LYP /6-31G* Meth od s
are others where there is a switch in conformation in the
transition structures as compared to that of the reac-
tants.7 We have predicted theoretically, and have veri-
fied experimentally through studies of rates of cycload-
ditions of some conformationally fixed enol ethers, that
such a conformational switch occurs.
olefins
RHF/3-21G
Becke3LYP/6-31G*
s-cis methyl vinyl ether
s-trans methyl vinyl ether -190.850 10 (3.3) -193.110 42 (2.3)
-190.855 30 (0.0) -193.114 06 (0.0)
s-cis silyl vinyl ether
s-trans silyl vinyl ether
s-cis 1-butene
-440.634 47 (0.0) -444.534 41 (0.0)
-440.631 09 (1.5) -444.533 25 (0.7)
-155.241 94 (0.8) -157.220 38 (0.4)
-155.243 26 (0.0) -157.221 07 (0.0)
Ca lcu la tion Meth od s
s-trans 1-butene
Geometry optimizations were carried out at the restricted
Hartree-Fock (RHF) level and with density functional theory
(DFT) and the Becke3LYP8 functional using the GAUSSIAN94
program.9 For the reactions with ethylene, 1-butene, silyl
vinyl ether, and methyl vinyl ether, the reactants and transi-
tion structures were fully optimized with the 3-21G basis set
for RHF and with the 6-31G* basis set for Becke3LYP
calculations. For the reactions with conformationally fixed
dienophiles and dipolarophiles, the reactants and transition
structures were fully optimized at the RHF level with the
3-21G basis set. Each of the transition structures gave only
one imaginary harmonic vibrational frequency, corresponding
to the formation of the C-C and C-X bonds. The self-
consistent reaction field (SCRF)10 cavity model with the
SCIPCM option11 was used to calculate the solvation energies
of the gas-phase-optimized structures in order to establish the
effect of solvent on the reaction rates and stereoselectivities.
The reported charges are Mulliken charges.
double bond in the s-cis conformation.12 Consequently,
the dipole moment of the molecule is small. In 1-butene,
the s-cis conformation is a local minimum but lies 0.8
kcal/mol above the gauche minimum; here there are no
significant electrostatic effects and steric hindrance
destabilizes the s-cis conformer relative to the s-trans
conformer.
Tr a n sition Str u ctu r es for th e Cycloa d d ition Re-
a ction s of Nitr osoeth ylen e, Acr olein , Bu ta d ien e,
a n d Nitr on e w ith Eth ylen e. Calculations for each of
these systems have been reported using a variety of
different methods.13 The cycloadditions involving nitrone
and acrolein reacting with ethylene have been studied
using low-level ab intio methods or semiempirical meth-
ods, while those with butadiene13c and nitrosoethylene13j,k
have been studied using higher levels of theory such as
Becke3LYP/6-31G* and MP2/6-31G*.
Resu lts a n d Discu ssion
Concerted asynchronous transition structures were
found for both Diels-Alder and 1,3-dipolar cycloadditions
by RHF and DFT methods in our calculations. Previous
studies on related cycloadditions have predicted concerted
transition structures with geometries similar to those
found here.13 The transition structures for the reactions
of nitrosoethylene, acrolein, butadiene, and nitrone with
ethylene have been located using RHF/3-21G and
Becke3LYP/6-31G* methods in order to allow comparison
with the cycloadditions of substituted dienophiles or
dipolarophiles. Figure 1 shows the geometries of the four
transition structures and the activation energies pre-
dicted by these two methods. The activation energy
predicted for the cycloaddition of butadiene with ethylene
is 24.8 kcal/mol (Becke3LYP/6-31G* with ZPE included),13c
which agrees remarkably well with the experimental
value of 27.5 ( 2 kcal/mol.14 The acrolein activation
energy is roughly the same. Nitrosoethylene is predicted
to be a much more reactive diene than acrolein in the
inverse electron-demand Diels-Alder reactions, also in
agreement with experiment.15 The reactivities are nor-
mally controlled by LUMOdiene-HOMOdienophile interac-
tions for these two inverse electron-demand Diels-Alder
reactions. There is 0.14 electron transferred from eth-
ylene to both acrolein and nitrosoethylene in the respec-
tive Diels-Alder reactions. Nitrone should undergo 1,3-
1. Th eor etica l P r ed iction of th e Con for m a tion a l
Sw itch in Cycloa d d ition s of En ol Eth er s. Con for -
m a tion s of Vin yl Eth er s a n d 1-Bu ten e in Gr ou n d
Sta tes. The lowest energy conformation of methyl vinyl
ether is the s-cis planar conformation, as demonstrated
by both experiment and theory.12 The s-trans conforma-
tion is nearly 2.0 kcal/mol higher in energy. This
preference is further supported by RHF/3-21G and
Becke3LYP/6-31G* calculations. Table 1 lists the cal-
culated and relative energies of s-cis and s-trans methyl
vinyl ether, silyl vinyl ether, and 1-butene. Although the
bulky silyl group decreases the preference for the s-cis
conformation because of steric hindrance, the s-trans
conformation is still 1.5 kcal/mol higher in energy by RHF
and 0.7 kcal/mol by DFT. The s-cis preference can be
explained in terms of electrostatic effects: the lone pair
electrons on the oxygen avoid the π electrons of the
(7) For example: Martinelli, M. J .; Peterson, B. C.; Khan, V. V.;
Hutchison, D. R.; Houk, K. N. J . Org. Chem. 1994, 59, 2204.
(8) (a) Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864. (b)
Kohn, W.; Sham, L. J . Phys. Rev. 1965, 140, A1133. (c) Lee, C.; Yang,
W.; Parr, R. G. Phys. Rev. 1988, B37, 785. (d) Becke, A. D. Phys. Rev.
1988, A38, 3098. (e) Michlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem.
Phys. Lett. 1989, 157, 200.
(9) Gaussian94 (Revision A.1), Frisch, M. J .; Trucks, G. W.; Schlegel,
H. B.; Gill, P. M. W.; J ohnson, B. G.; Robb, M. A.; Cheeseman, J . A.;
Keith, T. A.; Peterson, G. A.; Montgomery, J . A.; Raghavachari, K.;
Al-Laham, A. A.; Zakrzewski, V. G.; Oriz, J . V.; Foresman, J . B.;
Cioslowski, J .; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.;
Peng, C. Y.; Ayla, 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.
(10) (a) Wong, M. W.; Frisch, M. J .; Wiberg, K. B. J . Am. Chem.
Soc. 1991, 113, 4776. (b) Wong, M. W.; Wiberg, K. B.; Frisch, M. J . J .
Am. Chem. Soc. 1992, 114, 523. (c) Wong, M. W.; Wiberg, K. B.; Frisch,
M. J . J . Chem. Phys. 1991, 95, 8991. (d) Wong, M. W.; Wiberg, K. B.;
Frisch, M. J . J . Am. Chem. Soc. 1992, 114, 1645.
(11) Wiberg, K. B.; Keith, T. A.; Frisch, M. J .; Murcko, M. J . Phys.
Chem. 1995, 99, 9072.
(13) (a) Li, Y.; Houk, K. N. J . Am. Chem. Soc. 1993, 115, 7478. (b)
Houk, K. N.; Gonzalez, J .; Li, Y. Acc. Chem. Res 1995, 28, 81. (c)
Goldstein, E.; Beno, B.; Houk, K. N. J . Am. Chem. Soc. 1996, 118, 6036.
(d) Brown, F. K.; Singh, U. C.; Kollman, P. A.; Raimondi, L.; Houk, K.
N.; Bock, C. W. J . Org. Chem. 1992, 57, 4868. (e) Houk, K. N.; Sims,
J .: Watts, C.; Luskus, I. J . J . Am. Chem. Soc. 1973, 95, 7301. (f) Houk,
K. N.; Sims, J .; Duke, R. E., J r.; Strozier, R. W.; George, J . K. J . Am.
Chem. Soc. 1973, 95, 7287. (g) Pascal, Y. L.; Chanet-Ray, J .; Vessiers,
R.; Zeroual, A. Tetrahedron 1992, 48, 7197. (h) Lee, I.; Han, E. S.; Choi,
J . Y. J . Comput. Chem. 1984, 5, 606. (i) Tietze, L. F.; Femen, J .; Anders,
E. Angew. Chem., Int. Ed. Engl. 1989, 28, 1371. (j) J ursic, B. S.;
Zdravkovski, Z. J . Org. Chem. 1995, 60, 3163. (k) Sperling, D.;
Mehlhorn, A.; Reissig, H.-U.; Fabian, J . Liebigs Ann. 1996, 1615.
(14) Rowley, D.; Steiner, H. Discuss. Faraday Soc. 1951, 10, 198.
(15) Boger, D. L. Chem. Rev. (Washington, D.C.) 1988, 86, 781 and
references therein.
(12) For example: (a) Owen, N. L.; Seip, H. M. Chem. Phys. Lett.
1968, 48, 162. (b) Owen, N. L.; Sheppard, N. Trans. Faraday Soc. 1964,
60, 634. (c) Cahill, P.; Gold, L. P.; Owen, N. L. J . Chem. Phys. 1968,
48, 1620. (d) Gallinella, E.; Cadioli, B. J . Mol. Struct. 1991, 249, 343.

1066 J . Org. Chem., Vol. 63, No. 4, 1998
Liu et al.
F igu r e 2. Transition structures and activation energies for
the cycloaddition of methyl vinyl ether with butadiene, ac-
rolein, nitrosoethylene, and nitrone with Becke3LYP/6-31G*
and RHF/3-21G (in parentheses).
F igu r e 1. Transition structures and activation energies for
the cycloadditions of ethylene with butadiene, acrolein, nitro-
soethylene, and nitrone with Becke3LYP/6-31G* and RHF/3-
21G (in parentheses).
s-trans endo transition structures with methyl vinyl ether
are shown in Figure 2. The forming bond lengths in
s-trans endo transition structures are similar to those
found in the s-cis endo, s-trans-exo, and s-cis exo transi-
tion structures. For the inverse electron-demand Diels-
Alder cycloadditions with nitrosoethylene and acrolein,
the controlling frontier orbital interactions involve the
LUMO(diene) and HOMO(dienophile). This is reflected
in the charge transfer (B3LYP/6-31G*) from methyl vinyl
ether of 0.23 and 0.20 to nitrosoethylene and acrolein,
respectively. For the cycloadditions with nitrone and
butadiene, there is less charge transfer: 0.07 and 0.08
electron, respectively.
The methoxy group increases the asynchronicity mea-
sured as the difference between the bond lengths of the
two forming bonds in the transition structures. The
forming bonds R to the methoxy group are 0.15-0.35 Å
longer than those â to the methoxy group. The â carbons
are more nucleophilic than the R carbons. Consequently
there is more bond formation at the â carbons in the
transition structures.
The calculated energies, relative energies, and dipole
moments of the transition structures for the cycloaddi-
tions of methyl vinyl ether are listed in Table 2. For the
reactions with nitrosoethylene, nitrone, and acrolein, the
endo transition structures are favored by 0.9 kcal/mol or
less in the gas phase and by 1.1 kcal/mol or less in
solution using the DFT method. Boger explained this
endo preference in terms of hyperconjugative anomeric-
type interactions in which the lone pair of electrons (nO)
on the diene or dipole oxygen interact with the σ*CO
orbital of the ether, therefore stabilizing the endo transi-
tion structure when arranged antiperiplanar.16 In accord
Sch em e 2
dipolar cycloadditions with alkenes very easily, as
predicted by the very low activation energy of 11.4 kcal/
mol, obtained at the Becke3LYP/6-31G* level for the
parent system. The charge transfer from ethylene to
nitrone is only 0.02 electron in the transition state. The
C-O bond lengths in the nitrosoethylene and nitrone
transition structures are about only 0.02 Å shorter than
the C-C bonds, while the forming C-C bond in the
acrolein reaction is shorter than the forming C-O bond.
However, since a normal C-O bond length (1.48 Å) is
shorter than a single C-C bond length (1.54 Å), there
is, nevertheless, more C-C than C-O bond formation
in all of these cycloaddition transition structures.
Tr a n sition Str u ctu r es for th e Cycloa d d ition s of
Meth yl Vin yl Eth er . For each of the cycloadditions of
methyl vinyl ether with butadiene, acrolein, nitrosoeth-
ylene, and nitrone, four transition structures could be
located, referred to here as the s-cis endo, s-trans endo,
s-cis exo, and s-trans exo (Scheme 2). There are only
small differences in the geometries of the transition
structures obtained by RHF/3-21G and B3LYP/6-31G*
methods. The relative energies of the transition struc-
tures obtained by both methods are similar, with RHF/
3-21G predicting a slightly larger preference for the
s-trans conformation in the transition structures. The
(16) Boger, D. L.; Corbett, W. L.; Curran, T. T.; Kasper, A. M. J .
Am. Chem. Soc. 1991, 113, 1713.

Experimental Tests of Conformational Switches
J . Org. Chem., Vol. 63, No. 4, 1998 1067
Ta ble 2. Ca lcu la ted En er gies (a u ), Rela tive En er gies (k ca l/m ol, in P a r en th eses), a n d Dip ole Mom en ts (Debye) for
Tr a n sition Str u ctu r es of Cycloa d d ition s of Meth yl Vin yl Eth er in Ga s P h a se a n d Solu tion (RHF /3-21G,
Beck e3LYP /6-31G*)
RHF/3-21G
Becke3LYP/6-31G*
dipole
moment
dipole
moment
structures
gas phase
SCRFa
gas phase
SCRF
nitrosoethylene + methyl vinyl ether
s-cis endo
-396.316 79 (5.8) -396.331 71 (3.5)
-396.326 10 (0.0) -396.337 36 (0.0)
-396.316 29 (6.2) -396.331 31 (3.8)
-396.324 57 (1.0) -396.334 89 (1.5)
5.47
2.80
5.45
2.62
-400.971 44 (5.2) -400.982 05 (3.4)
-400.979 80 (0.0) -400.987 54 (0.0)
-400.972 53 (4.6) -400.982 80 (3.0)
-400.978 44 (0.9) -400.985 77 (1.1)
5.15
3.17
5.24
3.00
s-trans endo
s-cis exo
s-trans exo
nitrone + methyl vinyl ether
s-cis endo
s-trans endo
s-cis exo
s-trans exo
-358.683 09 (6.9) -358.693 84 (5.4)
-358.694 12 (0.0) -358.702 48 (0.0)
-358.685 71 (5.3) -358.695 89 (4.1)
-358.691 25 (1.8) -358.699 65 (1.8)
4.86
1.90
4.56
2.38
-362.882 20 (5.7) -362.887 99 (4.8)
-362.891 28 (0.0) -362.895 60 (0.0)
-362.884 64 (4.1) -362.890 12 (3.4)
-362.890 03 (0.8) -362.894 39 (0.8)
3.81
1.52
3.23
1.72
acrolein + methyl vinyl ether
s-cis endo
s-trans endo
s-cis exo
s-trans exo
-380.491 91 (5.3) -380.502 79 (3.8)
-380.499 63 (0.5) -380.508 88 (0.0)
-380.491 53 (5.5) -380.501 48 (4.6)
-380.500 37 (0.0) -380.507 57 (0.8)
4.29
2.01
2.84
1.82
-384.990 22 (5.8) -384.998 33 (5.3)
-384.999 42 (0.0) -385.005 59 (0.0)
-384.991 67 (4.9) -384.999 32 (3.9)
-384.999 20 (0.1) -385.004 29 (0.8)
4.36
2.76
3.86
2.33
butadiene + methyl vinyl ether
s-cis endo
s-trans endo
s-cis exo
s-trans exo
-344.845 11 (2.2) -344.850 94 (1.8)
-344.847 09 (1.0) -344.853 31 (0.3)
-344.846 76 (1.2) -344.852 43 (0.9)
-344.848 61 (0.0) -344.853 80 (0.0)
1.58
1.80
1.87
1.94
-349.058 70 (3.3) -349.061 69 (3.4)
-349.063 65 (0.2) -349.066 86 (0.1)
-349.060 20 (2.4) -349.063 17 (2.5)
-349.064 01 (0.0) -349.067 08 (0.0)
1.34
1.68
1.75
1.82
a
The SCRF model uses ꢀ ) 9.18 D for Diels-Alder reactions and ꢀ ) 2.14 D for 1,3-dipolar cycloadditions. It is the same for Tables
3-5.
with this, we find that for the cycloadditions of vinyl
The SCRF model was used in this work to calculate
the relative energy of each transition structure by single-
point calculations of the transition states placed in a
solvent cavity of dielectric constant 9.18 D, which is that
of the common solvent methylene chloride of Diels-Alder
reactions of enol ethers, or 2.14 D of benzene which is
often used for 1,3-dipolar cycloadditions. The preference
for the s-trans over the s-cis transition structures de-
creases in polar solvents. For example, for the reaction
of nitrosoethylene and methyl vinyl ether, the s-trans
endo transition state is 5.8 kcal/mol lower in energy than
s-cis in gas phase. The difference obtained by RHF and
DFT, drops to 3.5 or 3.4 kcal/mol in an SCRF cavity with
ꢀ ) 9.18 D. The same trends are found for the other
cycloadditions. The smallest effect is with the butadiene
cycloadditions, where the dipole moments are rather
similar for different transition states.
ethers with butadiene, where there can be no such orbital
interaction, the exo transition structures are favored
because of the steric hindrance in the endo transition
structure.
A large preference, 2.4-4.9 kcal/mol (Becke3LYP/6-
31G*), was found for the s-trans conformation of methyl
vinyl ether in both the endo and exo transition structures.
This preference is due to electrostatic and steric effects
discussed in detail later in this paper. For the Diels-
Alder reactions of methyl vinyl ether with 1-aza-1,3-
butadiene, the s-trans conformation of methyl vinyl ether
in the transition state is the only one reported from AM1
calculation by Fowler.2f
The Diels-Alder reactions, which have relatively non-
polar transition structures, normally exhibit small sol-
vent effects. However, for hetero-Diels-Alder cycload-
ditions, the transition structures have significant polarity.
Solvent can effect the selectivity of Diels-Alder reaction
if the different transition structures have different polar-
ity. Table 2 lists the gas-phase dipole moments of the
s-trans and s-cis transition structures. The significant
difference between the dipole moments indicates that
polar solvents are likely to change the relative energies
of the s-trans and s-cis transition structures.
Three theoretical models including both semiempirical
and ab initio methods have been reported for the model-
ing of solvent effects on Diels-Alder reactions.17 The first
model involves Monte Carlo simulations, which represent
the solvent as a set of discrete molecules interacting
through simple pairwise additive potentials. The super-
molecule model includes a few discrete solvent molecules
coordinated to the solute. Continuum cavity models such
as the self-consistent reaction field (SCRF) model repre-
sents the solvent as an infinite polarizable continuum,
characterized by its dielectric constant, ꢀ, surrounding a
variable-shaped cavity, in which the solute is placed.
Tr a n sition Str u ctu r es for th e Cycloa d d ition s of
Silyl Vin yl Eth er . For cycloadditions with silyl vinyl
ether, the transition structures with the s-cis conforma-
tion could not be located. Calculations starting from an
s-cis conformer did not converge or terminated at the
s-trans conformation. When single-point calculations
were carried out by rotating the OSiH3 substituent by
180°, the energies of the s-cis conformers were found to
be approximately 8 kcal/mol higher than the s-trans
conformers. Figure 3 shows all of the s-trans exo
transition structures for the cycloadditions of silyl vinyl
ether.
The silyl vinyl ether reactions exhibit a slight prefer-
ence for exo-Diels-Alder cycloadditions, in contrast to
the reactions of methyl vinyl ether. This exo selectivity
is consistent with Reissig’s experimental results.2b The
steric hindrance of the siloxy group in the endo transition
structures has a greater effect than the nO-σ*CO interac-
tion; the exo transition structures are more stable in
sterically hindered cases. Solvent effects (Table 3) do not
significantly influence the exo selectivities, since there
are no large differences between dipole moments of exo
and endo transition structures. The SiH3 group is
(17) For review of modeling methods of solvent effects on the Diels-
Alder reactions, see: Cativiela, C.; Garcia, J . I.; Mayoral, J . A.;
Salvatella, L. Chem. Soc. Rev. 1996, 209.

1068 J . Org. Chem., Vol. 63, No. 4, 1998
Liu et al.
Ta ble 3. Ca lcu la ted En er gies (a u ), Rela tive En er gies (k ca l/m ol, in P a r en th eses), a n d Dip ole Mom en ts (Debye) for
Tr a n sition Str u ctu r es of Cycloa d d ition s of Silyl Vin yl Eth er in Ga s P h a se a n d Solu tion (RHF /3-21G, Beck e3LYP /6-31G*)
RHF/3-21G
Becke3LYP/6-31G*
dipole
dipole
structures
gas phase
SCRF
moment
gas phase
SCRF
moment
nitrosoethylene + silyl vinyl ether
s-trans endo
-646.106 65 (1.8) -646.116 55 (1.5)
-646.109 47 (0.0) -646.119 01 (0.0)
2.88
2.83
-652.398 69 (0.9) -652.405 92 (0.2)
652.400 05 (0.0) -652.406 24 (0.0)
2.97
2.72
s-trans exo
nitrone + silyl vinyl ether
s-trans endo
s-trans exo
-608.480 99 (0.0) -608.489 02 (0.0)
-698.479 74 (0.8) -608.487 94 (0.7)
3.44
4.48
-614.314 37 (0.0) -614.318 33 (0.0)
-614.314 01 (0.2) -614.318 20 (0.1)
2.38
3.20
acrolein + silyl vinyl ether
s-trans endo
s-trans exo
-630.282 19 (3.7) -630.290 18 (2.3)
-630.288 01 (0.0) -630.293 86 (0.0)
1.53
1.80
-636.418 99 (1.5) -636.424 20 (0.4)
-636.421 37 (0.0) -636.424 91 (0.0)
1.92
1.68
butadiene+silyl vinyl ether
s-trans endo
s-trans exo
-594.628 58 (0.2) -594.632 86 (0.1)
-594.628 94 (0.0) -594.632 98 (0.0)
1.13
1.75
-600.482 46 (1.8) -600.484 72 (1.9)
-600.485 26 (0.0) -600.487 75 (0.0)
0.90
1.11
effects. Solvents exert very small effects on the relative
energies of the transition structures, because the differ-
ences between the dipole moments of each transition
structure are quite small. The ethyl substituent de-
creases the activation energies of the cycloadditions with
nitrosoethylene and acrolein relative to the reactions with
ethylene. The transition structures are less asynchro-
nous than those for the vinyl ethers. The differences
between the two forming bond lengths are approximately
0.1 Å (Figure 4).
s-Tr an s Con for m ation s Ar e P r efer r ed in th e Tr an -
sition Sta tes. The most dramatic conclusion drawn
from our calculations is that the enol ether adopts the
s-trans conformation in the transition state. This is
especially striking in the case of methyl vinyl ether,
which is known from experiment and theory to have an
s-cis conformation. Therefore, for enol ethers there is a
switch of conformation going from the ground state to
the transition state. In the transition states of reactions
with the electron-deficient nitrosoethylene, acrolein, and
nitrone, the s-trans preferences are 4-6 kcal/mol in gas
phase. For butadiene, the s-trans preference drops to
1-2 kcal/mol.
Vinyl ethers exist in the s-cis conformation to minimize
the dipole interactions between the oxygen lone pairs and
the π-system of the carbon-carbon double bond. The
s-trans preference observed in the transition structures
of these cycloadditions can be explained on the basis of
electrostatic interactions between the lone pairs of the
vinyl ether oxygens and the partial positive charge of the
alkene in the transition state. This is shown qualita-
tively in Figure 5. The interactions with the terminal
oxygens of the dienes or the nitrone might also be
important. There is also a steric component to these
effects which is reflected in the fact that s-cis transition
structures cannot be located for silyl enol ether reactions.
The 1-butene reactions have 0-3 kcal/mol preferences
for the s-trans transition structures, as compared to the
0.4 kcal/mol preference for the anti conformation in the
ground state.
F igu r e 3. Transition structures and activation energies for
the cycloadditions of silyl vinyl ether with butadiene, acrolein,
nitrosoethylene, and nitrone with Becke3LYP/6-31G* and
RHF/3-21G (in parentheses).
predicted to be directed somewhat toward the diene or
dipole, except in the case of butadiene. The dihedral
angles between the SiH3 moiety and the vinyl group,
CdC-O-Si, are in the range of 30-45°. This may be
attributed to the electrostatic attraction between the
silicon atom and the oxygen of the diene or dipole.18
Tr a n sition Str u ctu r es for th e Cycloa d d ition s of
1-Bu ten e. The reactions of 1-butene were also studied
in order to enable comparison to a dienophile lacking an
oxygen to influence the conformation. Table 4 gives the
results. The cycloadditions of 1-butene show very small
endo/exo preferences, since there are no significant
orbital interactions to stabilize either transition struc-
ture. Although the s-trans preference is still observed,
the differences between the s-trans and s-cis transition
structures are much smaller than those for the vinyl
ethers. The s-trans preferences are caused by steric
Since the s-trans preference is due primarily to elec-
trostatic effects, as reflected in the dipole moments of the
s-cis transition structures, solvent effects are expected
to be significant. Polar solvent effects reduce the s-trans
preference dramatically for enol ethers, because the polar
solvent stabilizes the transition structures with s-cis
conformations relative to those with s-trans conforma-
tions. The solvent effects on the relatively nonpolar
butene reaction are predicted to be very small.
(18) Seebach, D.; Maetzke, T.; Petter, W.; Klotzer, B.; Plattner, D.
A. J . Am. Chem. Soc. 1991, 113, 1781.

Experimental Tests of Conformational Switches
J . Org. Chem., Vol. 63, No. 4, 1998 1069
Ta ble 4. Ca lcu la ted En er gies (a u ), Rela tive En er gies (k ca l/m ol, in P a r en th eses), a n d Dip ole Mom en ts (Debye) for
Tr a n sition Str u ctu r es of Cycloa d d ition s of 1-Bu ten e in Ga s P h a se a n d Solu tion (RHF /3-21G, Beck e3LYP /6-31G*)
RHF/3-21G
Becke3LYP/6-31G*
dipole
dipole
structures
gas phase
SCRF
moment
gas phase
SCRF
moment
nitrosoethylene + 1-butene
s-cis endo
-360.704 83 (2.7)
-360.709 25 (0.0)
-360.706 61 (1.0)
-360.708 48 (0.5)
-360.712 78 (2.8)
-360.717 23 (0.0)
-360.714 25 (1.9)
-360.716 22 (0.6)
3.51
3.72
3.50
2.62
-365.077 39 (1.8)
-365.080 29 (0.0)
-365.079 02 (0.8)
-365.079 77 (0.3)
-365.083 23 (1.8)
-365.086 10 (0.0)
-365.085 16 (0.6)
-365.085 59 (0.3)
3.37
3.59
3.52
3.53
s-trans endo
s-cis exo
s-trans exo
nitrone + 1-butene
s-cis endo
323.073 20 (2.3)
-323.076 82 (0.0)
-323.075 78 (0.7)
-323.076 61 (0.1)
-323.080 30 (2.1)
-323.083 74 (0.0)
-323.082 57 (0.7)
-323.083 35 (0.2)
3.25
3.52
3.38
3.52
-326.992 62 (1.8)
-326.995 13 (0.2)
-326.994 59 (0.6)
-326.995 48 (0.0)
-326.996 34 (1.7)
-326.998 79 (0.2)
-326.998 28 (0.5)
-326.999 12 (0.0)
2.48
2.69
2.48
2.67
s-trans endo
s-cis exo
s-trans exo
acrolein + 1-butene
s-cis endo
-344.876 56 (3.6)
-344.881 70 (0.4)
-344.880 16 (1.3)
-344.882 30 (0.0)
-344.882 23 (3.2)
-344.887 30 (0.0)
-344.884 87 (1.5)
-344.886 73 (0.5)
2.15
2.32
1.94
2.04
-349.093 86 (2.8)
-349.098 23 (0.1)
-349.096 57 (1.1)
-349.098 40 (0.0)
-349.097 70 (2.8)
-349.102 19 (0.0)
-349.100 16 (1.3)
-349.101 58 (0.4)
2.33
2.55
2.17
2.26
s-trans endo
s-cis exo
s-trans exo
butadiene + 1-butene
s-cis endo
-309.236 01 (1.8)
-309.238 94 (0.0)
-309.237 69 (0.8)
-309.237 74 (0.7)
-309.238 22 (1.9)
-309.241 18 (0.0)
-309.239 93 (0.8)
-309.240 03 (0.7)
0.70
0.72
0.91
0.84
-313.169 17 (1.5)
-313.171 56 (0.0)
-313.170 25 (0.8)
-313.170 73 (0.5)
-313.170 55 (1.5)
-313.172 91 (0.0)
-313.171 69 (0.8)
-313.172 22 (0.4)
0.57
0.59
0.77
0.72
s-trans endo
s-cis exo
s-trans exo
F igu r e 5. Electrostatic effects in the transition states of
Diels-Alder with enol ethers.
Sch em e 3
F igu r e 4. Transition structures and activation energies for
the cycloadditions of 1-butene with butadiene, acrolein, nitro-
soethylene, and nitrone with Becke3LYP/6-31G* and RHF/3-
21G (in parentheses).
2. Exp er im en ta l Tests of th e In flu en ce of Con -
for m a tion on th e Ra tes of Cycloa d d ition s of En ol
E t h er s. There is no direct way to observe which
conformation is present in the transition state of the
reaction. However, these results lead to the prediction
that enol ethers fixed in different conformations will react
at different rates.
To test these predictions, the cycloaddition reactions
of several conformationally fixed enol ethers with 1-ni-
troso-1-phenylethylene, 1, and C-benzoyl-N-phenylni-
trone, 2, were carried out experimentally (Scheme 3).
Conformations of the enol ethers were fixed by incorpora-
tion into five-membered rings. In 2,3-dihydrofuran, 3b,
the conformation is s-cis, while in 2-methylene tetrahy-
drofuran, 4b, the conformation is s-trans. Theoretically,
the s-trans enol ether should be more reactive than the
s-cis. However, the substitution patterns in 2,3-dihy-
drofuran (1,2-disubstituted) and 2-methylenetetrahydro-
furan (1,1-disubstituted) are different, and this will
influence the rates also. Therefore, controls were per-
formed to determine the contribution of structural dis-
similarity to the rate difference. Cyclopentene (3a ) and
methylenecyclopentane (4a ) provide the appropriate
models with 1,2- and 1,1-disubstitution, respectively.
Ab initio calculations were performed on these mol-
ecules to locate the Diels-Alder and 1,3-dipolar cycload-
dition transition structures. RHF/3-21G calculations

1070 J . Org. Chem., Vol. 63, No. 4, 1998
Liu et al.
Ta ble 5. Ca lcu la ted En er gies (a u ) a n d Activa tion En er gies(k ca l/m ol) for Tr a n sition Str u ctu r es of Cycloa d d ition in Ga s
P h a se a n d Solu tion a n d th e Dip ole Mom en ts (Debye) a t RHF /3-21G
compounds
E (gas phase)
E(SCRF)
dipole moment
∆Ea(gas phase)
∆Ea(SCRF)
cyclopentene + nitrosoethylene
endo
-398.365 03
-398.366 82
-437.190 70
-398.372 93
-398.375 12
-437.198 13
2.72
3.78
3.54
21.0
19.9
18.3
20.5
19.1
18.1
exo
methylenecyclopentane + nitrosoethylene
dihydrofuran + nitrosoethylene
endo
exo
-433.978 30
-433.977 02
-433.993 50
-433.992 46
5.83
5.80
19.7
20.5
17.2
17.8
2-methylenetetrahydrofuran+nitrosoethylene
endo
exo
-472.816 22
-472.814 68
-472.827 02
-472.825 69
3.25
3.03
14.0
15.0
14.8
15.9
cyclopentene + nitrone
endo
exo
-360.732 33
-360.734 65
-399.557 09
-360.739 46
-360.741 67
-399.563 43
3.22
3.34
3.40
14.3
12.8
12.1
18.0
16.7
16.4
methylenecyclopentane + nitrone
dihydrofuran + nitro
endo
exo
-396.344 71
-396.345 11
-396.355 59
-396.356 50
4.85
5.11
13.5
13.2
16.5
16.0
2-methylenetetrahydrofuran + nitrone
endo
exo
-435.183 99
-435.179 42
-435.190 59
-435.187 40
1.42
2.54
7.6
9.8
13.0
15.0
were performed for both gas and solution phase. The
activation energies for each reaction and the dipole
moments of these transition structures are listed in Table
5. The Diels-Alder reaction of nitrosoethylene with
2-methylenetetrahydrofuran has an activation energy
which is 5.7 kcal/mol lower in the gas phase than the
reaction with 2,3-dihydrofuran. In a cavity with a
dielectric constant of 9.18 D, which is the ꢀ for the
reaction solvent methylene chloride, this difference drops
to 2.4 kcal/mol because of the higher dipole moments and
solvation energies of the s-cis transition states. In a
reaction with methylene nitrone, the reaction of 2-meth-
ylenetetrahydrofuran has an activation energy 5.6 kcal/
mol lower in the gas phase and 3.0 kcal/mol lower in a
cavity (with dielectric constants of 2.14 D which is the ꢀ
for reaction solvent benzene) than the 2,3-dihydrofuran.
The activation energies for the Diels-Alder reaction of
nitrosoethylene with methylenecyclopentane are lower
by 1.6 kcal/mol in the gas phase and 1.0 kcal/mol in
solution relative to those of nitrosoethylene with cyclo-
pentene. Finally, the activation energies for the 1,3-
dipolar reaction of methylene nitrone with methylenecy-
clopentane are lower by 0.7 kcal/mol in gas phase and
0.3 kcal/mol in solution phase than those of methylene
nitrone with cyclopentene.
We have carried out the Diels-Alder reactions of
1-phenyl-1-nitrosoethylene, 1, with 2-methylenetetrahy-
drofuran, 4b, 2,3-dihydrofuran, 3b, methylenecyclopen-
tane, 4a , and cyclopentene, 3a , as well as the 1,3-dipolar
cycloadditions of C-benzoyl-N-phenylnitrone, 2, with each
of the four alkenes (Scheme 3). 1-Nitroso-1-phenyleth-
ylene, 1, was generated in situ by base treatment of
ω-chloroacetylbenzene oxime.19 The nitrone 220 and
olefin 4b21 were prepared according to methods described
in the literature. The Diels-Alder reactions were carried
out in freshly distilled methylene chloride at room
temperature, while the 1,3-dipolar cycloadditions were
carried out in benzene. Each Diels-Alder reaction yields
only one pair of enantiomers. Endo and exo products
from the 1,3-dipolar reactions could be separated by flash
chromatography. All structures for pure products were
1
verified using H NMR, 13C NMR, IR, MS, or elemental
analysis. The structures of the endo products, 7c and
8b, were determined by X-ray crystallography.22 The
structures of 7a and 7b were determined from the
coupling constants in the 1H NMR. The exo/endo product
ratios for the 1,3-dipolar cycloadditions were determined
from isolation of the individual adducts. The endo/exo
ratio was 3.2:1 for 2-methylenetetrahydrofuran, 1.1:1 for
2,3-dihydrofuran, and 1.2:1 for cyclopentene.
An endo preference is observed for the 1,3-dipolar
cycloaddition of the nitrone to 2-methylenetetrahydro-
furan; the 3.2:1 endo/exo ratio indicates that the endo
transition structure is favored by 0.7 kcal/mol. The endo
and exo products of the 1,3-dipolar cycloaddition of
nitrone with 2,3-dihydrofuran and cyclopentene are
formed in comparable ratios, in agreement with calcula-
tions. The energy differences predicted by RHF/3-21G
between the endo and exo products in the cycloadditions
of the nitrone with 2,3-dihydrofuran and cyclopentene are
1.3 and 0.5 kcal/mol, whereas the selectivity observed
corresponds to a smaller endo preference.
To determine the relative rates of reactions of the enol
ethers constrained to be s-cis and s-trans, the reactions
of mixtures of 2,3-dihydrofuran and 2-methylenetetrahy-
drofuran with 1-nitroso-1-phenylethylene and with C-
benzoyl-N-phenylnitrone were investigated. 1-Nitroso-
1-phenylethylene was reacted with a 2.5-fold excess of a
mixture of the enol ethers. C-Benzoyl-N-phenylnitrone
was reacted with a 20-fold excess of enol ethers for the
1,3-dipolar cycloaddition competitive reactions. HPLC
was used to determine the product ratios. The ratios of
the relative reaction rates for 2-methylenetetrahydrofu-
ran/2,3-dihydrofuran were 33.5 for the Diels-Alder reac-
tion and 12.2 for the 1,3-dipolar cycloaddition. For
reference, the reactions with mixtures of cyclopentene
and methylenecyclopentane were also investigated. A
2.5-fold excess of a mixture of alkenes was employed for
reactions with 1-nitroso-1-phenylethylene and 20-fold
excess of a mixture of alkenes for reaction with C-benzoyl-
N-phenylnitrone. The ratios of reaction rates for meth-
(19) For the preparation of oxime refer to: Korten, H.; Scholl, R.
Ber. 1901, 34, 1901.
(20) Averas, M. C.; Cum, G.; Stragno d’Alcontres, G.; Uccella, N. J .
Chem. Soc., Perkin Trans. 1 1972, 222.
(21) Ireland, R. E.; Haebich, D. Chem. Ber. 1981, 114, 1418.
(22) Niwayama, S.; You, Y.; Houk, K. N.; Khan, S. Acta Crystallogr.
C 1996, 52, 486.

Experimental Tests of Conformational Switches
J . Org. Chem., Vol. 63, No. 4, 1998 1071
Sch em e 4
ylenecyclopentane/cyclopentene were 13.5 for the Diels-
Alder reaction and 10.4 for the 1,3-dipolar cycloaddition.
The differences in the activation energies for the
2-methylenetetrahydrofuran (s-trans conformation) and
2,3-dihydrofuran (s-cis conformation) reactions are 2.4
kcal/mol for the Diels-Alder reaction and 3.0 kcal/mol
for the 1,3-dipolar cycloaddition according to RHF/3-21G
calculations in the dielectric cavity. The experimental
results show that the s-trans enol ether, 2-methylene-
tetrahydrofuran, has a lower activation energies than
s-cis, dihydrofuran, by 2.1 and 1.5 kcal/mol for the Diels-
Alder and 1,3-dipolar cycloadditions, respectively. The
values are comparable to the calculations but show lower
selectivity than predicted.
The calculated differences in activation energies for the
reactions of methylenecyclopentane and cyclopentene are
1.0 kcal/mol for the Diels-Alder reaction and 0.3 kcal/
mol for the 1,3-dipolar cycloaddition in solution; the
differences in activation energies experimentally are 1.5
and 1.4 kcal/mol, respectively. Here the calculated values
predict comparable but lower selectivity for alkene than
found experimentally.
F igu r e 6. Relative energies and s-cis and s-trans conforma-
tions for the chiral enol ethers 11 and 12 optimized by the
MM2* force field.
s-cis conformer, the diene can access the enol ether from
both π-faces in 9, which should lead to no π-facial
selectivity. When s-cis, the diene can access the re face
of 10, which would lead to different selectivity from that
observed experimentally. Thus the stereochemical result
of experiment requires the s-trans conformer to be
present in the transition state.1c Denmark et al. ratio-
nalize this conformation as resulting from steric effects
in the transition state.
Conformation searches were done on chiral enol ethers
11 and 12 using the MM2* force field in Macromodel 5.024
in order to investigate whether the s-cis or s-trans
conformations of these enol ethers are involved in the
transition states of Diels-Alder cycloadditions. The s-cis
and s-trans conformations are the ground-state confor-
mations for both these enol ethers as shown in Figure 6.
The s-cis conformation is slightly more stable than the
s-trans conformation in 12, and both are the same energy
for 11. If the transition states adopt the s-cis conforma-
tion of the enol ether, the diene would be able to access
the enol ether from both the re and si faces, and there
would be no selectivity. However, if the reactions take
the s-trans conformation in the transition state, the re
face is blocked by the other parts of the enol ether.
Consequently, the transition state with the s-trans
conformation of enol ether leads to the stereoselectivity
observed experimentally.1 The stereoselectivities of Di-
els-Alder reaction of chiral enol ethers proved that the
The experimental results verify the theoretical predic-
tion of a preference for the transition states of enol ethers
to be an s-trans arrangement, although when compared
in solution, the preference with enol ethers is only slightly
larger than the preference for alkenes. The investigation
into the polarities of different solvents on the preference
of the enol ether cycloaddition should be able to confirm
that polar solvent can stabilize the s-cis transition state
more relative to the s-trans transtion state. For the
Diels-Alder reaction of nitrosoethylene, there is a 1 kcal/
mol greater increase in preference for reaction of the
s-trans-constrained enol ether than for the alkyl model,
while in the nitrone cycloaddition, the experiment prefer-
ence for the s-trans transition state with the ether and
alkyl cases are essentially the same. The results indicate
that any s-cis preference that may be present in acyclic
enol ethers is eliminated in cycladdition transition states.
3. Im p lica tion s for Ster eoselectivity. Chiral enol
ethers, such as 9, 10, 11, and 12 (Scheme 4), have been
used in Diels-Alder cycloadditions to control π-facial
selectivities.1 Denmark et al. reported molecular me-
chanics calculations on the conformations of the chiral
enol ethers, 9 and 10, that they have used in Diels-Alder
reactions.1c,23 The s-cis conformer was predicted to be the
preferred ground-state conformation. However, for the
(24) Macromodel V5.0: Mohamadi, F.; Richards, N. G. J .; Guida,
W. C.; Liskamp. R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson,
T.; Still W. C. J . Comput. Chem. 1990, 11, 440. MM2* is the
MacroModel implementation of MM2.
(23) Denmark, S. E.; Senanyake, C. B. W.; Ho, C.-D. Tetrahedron
1990, 46, 4857.

1072 J . Org. Chem., Vol. 63, No. 4, 1998
Liu et al.
Hz, 1H), 2.78 (dd, J ) 17.9, 8.7 Hz, 1H), 2.29 (dd, J ) 17.9,
enol ethers adopt the s-trans conformation in the cy-
cloaddition transition states.
4.4 Hz, 2H), 2.23 (m, 1H), 1.88 (m, 3H), 1.56 (m, 2H) ppm. 13
C
NMR (CDCl3, 200 MHz) 160.6 (s, 1C), 135.8 (s, 1C), 129.8 (d,
1C), 128.6 (d, 2C), 125.7 (d, 2C), 78.8 (d, 1C), 37.2 (t, 1C), 32.0
(d, 1C), 31.3 (t, 1C), 24.7 (t, 1C), 22.8 (t, 1C) ppm. MW for
Con clu sion
C
13H15NO: calcd 201.1154, found 201.1152.
3-P h en yl-2-a za -1,7-d ioxa bicyclo[4.3.0]n on -2-en e (5b):
A general type of stereoselectivity has been found for
enol ethers in inverse electron-demand Diels-Alder
reactions and 1,3-dipolar cycloadditions. The s-trans
preference of enol ethers in transition structures is an
electrostatic effect which decreases in polar solvents.
Measurements of rates of reactions with conformationally
fixed enol ethers have verified this preference. The
s-trans transition states can explain the π-facial stereo-
selectivities in cycloadditions of chiral enol ethers ob-
served by Denmark, Reissig, and others.
yield ) 90.0%, mp 64-65 °C. IR (KBr): 3063.8, 2982.5, 2941.8,
2905.2, 1592.8, 1446.8, 1080.3, 979.9, 897.1, 762.0, 700.2 cm-1
.
1H NMR (DCCl3, 200 MHz) 7.73 (m, 2H), 7.41 (m, 3H), 5.66
(d, J ) 6.1 Hz, 1H), 4.04 (ddd, J ) 15.3, 7.2, 2.1 Hz, 1H), 3.89
(ddd, J ) 15.3, 7.4, 1.8 Hz, 1H), 2.93 (m, 1H), 2.74 (d, J ) 4.9
Hz, 2H), 2.18 (m, 1H), 1.77 (m, 1H) ppm. 13C NMR (CDCl3,
200 MHz) 164.9 (s, 1C), 135.0 (s, 1C), 130.4 (d, 1C), 128.8 (d,
2C), 125.9 (d, 2C), 102.1 (t, 1C), 68.4 (d, 1C), 37.0 (t, 1C), 30.3
(d, 1C), 25.1 (t, 1C) ppm. MW for C12H13NO2: calcd 203.0946,
found 203.0950.
3-P h en yl-2-a za -1-oxa sp ir o[5.4]d ec-2-en e (6a ): yield )
83.8%, mp 62-63 °C. IR (KBr): 3063.8, 2961.1, 2870.1,
Exp er im en ta l Section
1592.5, 1446.8, 1012.8, 910.5, 754.3, 688.7 cm-1
.
1H NMR
Ma ter ia ls a n d Meth od s. Commercial grade reagents and
solvents were used without further purification except as
indicated below. Methylene chloride and benzene were dis-
tilled from calcium hydride. ω-Chloroacetylbenzene oxime,19
C-benzoyl-N-phenylnitrone,20 and tetrahydro-2-methylene-2H-
furan21 were prepared by standard methods. All the reactions
were stirred magnetically in oven-dried glassware under an
inert atmosphere.
(CDCl3, 200 MHz) 7.62 (m, 2H), 7.29 (m, 3H), 2.57 (t, J ) 6.9
Hz, 2H), 1.89 (t, J ) 6.9 Hz, 2H), 1.81 (m, 4H), 1.61 (m, 2H),
1.19 (m, 2H) ppm. 13C NMR (CDCl3): 153.7 (s, 1C), 136.2 (s,
1C), 129.3 (d, 1C), 128.4 (d, 2C), 125.2 (d, 2C), 85.2 (s, 1C),
37.0 (t, 2C), 28.1 (t, 1C), 24.4 (t, 2C), 20.9 (t, 1C) ppm. MW
for C14H17NO2: calcd 215.1310, found 215.1311.
3-P h en yl-2-a za -1,7-d ioxa sp ir o[5.4]d ec-2-en e (6b): yield
) 95.6%, mp 47-48 °C. IR (KBr): 3063.8, 2988.7, 2951.1,
2926.1, 2876.2, 1593.8, 1444.9, 1080.3, 983.8, 773.6, 702.2
Gen er a l P r oced u r e for Diels-Ald er R ea ct ion s of
1-P h en yl-1-n itr osoeth ylen e w ith Dien op h iles. 1-Nitroso-
1-phenylethylene was formed from ω-chloroacetylbenzene
oxime in basic solution and reacts immediately with dieno-
philes to form Diels-Alder products.25 Solutions of ω-chloro-
acetylbenzene oxime (0.002 mol) and the dienophiles (cyclo-
pentene, methylenecyclopentane, 2,3-dihydrofuran, or 2-meth-
ylenetetrahydrofuran) (0.01 mol) in methylene chloride (40
mL) were stirred with freshly ground sodium carbonate (2.12
g, 0.02 mol) at room temperature for 48 h. The reaction were
stopped by removing the sodium carbonate with a Celite
funnel. The solvent methylene chloride was evaporated, and
the resulting oil was purified using flash chromatography.
Gen er a l P r oced u r e for th e Com p etitive Diels-Ald er
Rea ction s. A solution of ω-chloroacetylbenzene oxime (0.085
g, 0.5 mmol) and 2,3-dihydrofuran (1.25 mmol), 2-methylene-
tetrahydrofuran (1.25 mmol) or cyclopentene (1.25 mmol), and
methylenecyclopentane (1.25 mmol) in 10 mL of methylene
chloride was stirred with freshly ground sodium carbonate (5
mmol) at room temperature for 48 h. The reactions were
stopped by filtering the base out of the solution and chilling
the reaction mixture at 0 °C. The ratios of products formed
for the competitive reactions were checked using HPLC with
Nova/Pak C18 column and H2O/CH3CN as solvent; pure product
solutions were used as standard.
Gen er a l P r oced u r e for th e 1,3-Dip ola r Cycloa d d ition s
of C-Ben zoyl-N-p h en yln itr on e. A solution of C-benzoyl-N-
phenylnitrone (180 mg, 0.80 mmol) and dipolarophiles (2-
methylenetetrahydrofuran, 2,3-dihydrofuran, methylenecyclo-
pentane, or cyclopentene) (20.4 mmol) in dry benzene (3 mL)
was stirred for 48 h at room temperature. The solvent was
removed on rotary evaporator, and the residue was chromato-
graphed to yield cycloadduct.
Gen er a l P r oced u r e for th e Com p etitive 1,3-Dip ola r
Cycloa d d ition of C-Ben zoyl-N-p h en yln itr on e. The solu-
tion of C-benzoyl-N-phenylnitrone (90 mg, 0.40 mmol) and
2-methylenetetrahydrofuran (8.0 mmol), 2,3-dihydrofuran (8.0
mmol) or methylene cyclopentane (8.0 mmol), and cyclopentene
(8.0 mmol) in dry benzene (4 mL) was stirred for 48 h at room
temperature. The ratios of cycloadducts formed for the
competitive reactions were checked using HPLC with Nova/
Pak C18 column and H2O/CH3CN as solvent; pure product
solutions were used as standard.
cm-1 1H NMR (CDCl3, 200 MHz) 7.72 (m, 2H), 7.35 (m, 3H),
.
4.03 (m, 2H), 2.76 (m, 2H), 2.25 (m, 2H), 2.08 (m, 2H), 1.90
(m, 2H) ppm. 13C NMR (CDCl3, 200 MHz) 154.8 (s, 1C), 135.6
(s, 1C), 129.5 (d, 1C), 128.4 (d, 2C), 125.5 (d, 2C), 104.8 (s,
1C), 68.1 (t, 1C), 35.8 (t, 1C), 26.3 (t, 1C), 24.2 (t, 1C), 20.4 (t,
1C) ppm. MW for C13H15NO2: calcd 217.1103, found 217.1100.
(3r,3a r,6a r)-3-Ben zoyl-2-p h en ylcyclop en t o[2,1-d ]t et -
r a h yd r oisoxa zole (7a ): yield ) 46.2%, mp 153-154 °C. IR
(thin film): 3063, 2957, 2870, 1694, 1672, 1597, 1580, 1491,
1275, 1180, 1026, 756, 692 cm-1
.
1H NMR (400 Hz, CDCl3):
1.60-1.67 (m, 2H), 1.74-1.91 (m, 3H), 2.02-2.08 (m, 2H),
3.24-3.29 (m, 1H), 4.20 (d, J ) 6.1 Hz, 1H), 4.95 (dd, J ) 6.0,
5.2 Hz, 1H), 6.94 (t, J ) 7.4 Hz, 1H), 6.99 (dd, J ) 8.7, 1.1 Hz,
2H), 7.21 (dd, J ) 8.7, 7.4 Hz, 2H), 7.48 (dd, J ) 8.4, 7.4 Hz,
2H), 7.59 (t, 7.4 Hz, 1H), 8.19 (dd, J ) 8.4, 1.2 Hz, 2H) ppm.
13C NMR (400 Hz, CDCl3): 24.1 (t, 1C), 31.5 (t, 1C), 32.1 (t,
1C), 50.3 (d, 1C), 78.1 (d, 1C), 84.0 (d, 1C), 116.4 (d, 2C), 122.6
(d, 1C), 128.8 (d, 2C), 128.8 (d, 2C), 129.2 (d, 2C), 133.7 (d,
1C), 135.2 (s, 1C), 149.2 (s, 1C), 197.3 (s, 1C) ppm. Anal.
Calcd. for C19H19NO2: C, 77.79; H, 6.53; N, 4.77. Found: C,
77.54; H, 6.46; N, 4.72.
(3â,3a r,6a r)-3-Ben zoyl-2-p h en ylcyclop en t o[2,1-d ]t et -
r a h yd r oisoxa zole (7b): yield ) 37.8%, mp 138-139 °C. IR
(thin film): 3063, 2957, 2872, 1688, 1597, 1579, 1491, 1449,
1275, 1236, 1211, 1182, 1159, 1026, 988, 752, 691 cm-1
.
1H
NMR (400 Hz, CDCl3): 1.38-1.43 (m, 1H), 1.50-1.60 (m, 2H),
1.71-1.89 (m, 2H), 2.01-2.07 (m, 1H), 3.46 (dddd, J ) 9.0,
5.9, 5.8, 3.0 Hz, 1H), 4.79 (ddd, J ) 7.0, 5.8, 1.6 Hz, 1H), 5.23
(d, J ) 9.0 Hz, 1H), 6.93 (t, J ) 7.3 Hz, 1H), 6.97 (dd, J ) 8.8,
1.1 Hz, 2H), 7.23 (dd, J ) 8.8, 7.3 Hz, 2H), 7.51 (dd, J ) 8.4,
7.5 Hz, 2H), 7.61 (t, 7.5 Hz, 1H), 8.02 (dd, J ) 8.4, 1.2 Hz, 2H)
ppm. 13C NMR (400 Hz, CDCl3): 25.5 (t, 1C), 27.9 (t, 1C), 30.6
(t, 1C), 52.5 (d, 1C), 75.0 (d, 1C), 84.8 (d, 1C), 114.6 (d, 2C),
121.8 (d, 1C), 128.5 (d, 2C), 128.9 (d, 2C), 128.9 (d, 2C), 133.6
(d, 1C), 136.1 (s, 1C), 151.3 (s, 1C), 196.0 (s, 1C) ppm. Anal.
Calcd. for C19H19NO2: C, 77.79; H, 6.53; N, 4.77. Found: C,
77.68; H, 6.67; N, 4.58.
(3r,3a r,6a r)-3-Ben zoylh exa h yd r o-2-p h en ylfu r o[3,2-d ]-
isoxa zole (7c): yield ) 51%, mp 133-134 °C. IR (thin film):
3062, 2969, 2894, 1692, 1597, 1489, 1219, 1078, 1016, 922, 789,
754 cm-1 1H NMR (400 Hz, CDCl3): 1.97-2.02 (m, 1H), 2.14-
.
2.17 (m, 1H), 3.50-3.52 (m, 1H), 3.72 (dt, J ) 9.2, 7.3 Hz, 1H),
3-P h en yl-2-a za -1-oxa bicyclo[4.3.0]n on -2-en e (5a ): yield
) 14.3%, colorless oil. IR (KBr): 3063.8, 2939.9, 1597.3,
1446.8, 1014.7, 908.6, 760.1, 694.5 cm-1 1H NMR (CDCl3, 200
.
(25) (a) Faragher, R.; Gilchrist, T. L. J . Chem. Soc., Perkin Trans.
1 1979, 249. (b) Gilchrist, T. L. Chem. Soc. Rev. 1983, 79, 53.
MHz): 7.63 (m, 2H), 7.33 (m, 3H), 4.00 (ddd, J ) 8.5, 3.7, 1.9

Experimental Tests of Conformational Switches
J . Org. Chem., Vol. 63, No. 4, 1998 1073
3.85 (ddd, J ) 9.2, 8.6, 2.2 Hz, 1H), 5.07 (d, J ) 2.2 Hz, 1H),
6.01 (d, J ) 5.3 Hz, 1H), 6.93 (t, J ) 7.3 Hz, 1H), 7.11 (dd, J
) 8.7, 1.0 Hz, 2H), 7.25 (dd, J ) 8.7, 7.3 Hz, 2H), 7.46 (dd, J
) 8.4, 7.4 Hz, 2H), 7.57 (t, 7.4 Hz, 1H), 8.02 (dd, J ) 8.4, 1.2
Hz, 2H) ppm. 13C NMR (400 Hz, CDCl3): 31.1 (t, 1C), 50.3 (d,
1C), 68.3 (t, 1C), 73.8 (d, 1C), 108.9 (d, 1C), 114.4 (d, 2C), 121.7
(d, 1C), 128.8 (d, 2C), 128.8 (d, 2C), 128.9 (d, 2C), 133.6 (d,
1C), 135.0 (s, 1C), 149.6 (s, 1C), 196.0 (s, 1C) ppm. Anal. Calcd
for C18H17NO3: C, 73.20; H, 5.80; N, 4.75. Found: C, 72.87;
H, 5.65; N, 4.57.
3061, 2984, 2957, 2890, 1696, 1674, 1582, 1489, 1449, 1213,
1180, 1159, 1094, 1082, 1026, 982, 783, 729, 694 cm-1
.
1H
NMR (400 Hz, CDCl3): 1.96-2.04 (m, 1H), 2.11-2.28 (m, 2H),
2.30-2.39 (m, 1H), 2.82 (d, J ) 8.3 Hz, 2H), 3.89-3.99 (m,
2H), 5.17 (t, J ) 8.3 Hz, 1H), 6.92 (t, J ) 7.3 Hz, 1H), 6.96 (d,
J ) 7.8 Hz, 2H), 7.24 (dd, J ) 7.8, 7.3 Hz, 2H), 7.48 (dd, J )
7.7, 7.4 Hz, 2H), 7.59 (t, 7.4 Hz, 1H), 8.14 (d, J ) 7.7 Hz, 2H)
ppm. 13C NMR (360 Hz, CDCl3): 24.5 (t, 1C), 32.3 (t, 1C), 42.4
(t, 1C), 67.8 (t, 1C), 71.1 (d, 1C), 113.6 (d, 2C), 114.6 (s, 1C),
121.4 (d, 1C), 128.7 (d, 2C), 128.8 (d, 2C), 129.2 (d, 2C), 133.7
(d, 1C), 134.7 (s, 1C), 152.7 (s, 1C), 197.3 (s, 1C) ppm. Anal.
Calcd for C19H19NO3: C, 73.77; H, 6.18; N, 4.53. Found: C,
73.63; H, 6.09; N, 4.44.
(3â,3a r,6a r)-3-Ben zoylh exa h yd r o-2-p h en ylfu r o[3,2-d ]-
isoxa zole (7d ): yield ) 47%, mp 176-178 °C. IR (thin
film): 3064, 2976, 2892, 1690, 1597, 1491, 1252, 1076, 1024,
926, 799, 750 cm-1 1H NMR (400 Hz, CDCl3): 1.74-1.85 (m,
.
cis-3-Ben zoyl-N-p h en yl-2-a za -1,6-d ioxzsp ir o[4.4]n o-
n a n e (8c): yield ) 23.0%, mp 133-134 °C. IR (thin film):
3063, 2955, 2886, 1698, 1671, 1491, 1339, 1277, 1211, 1182,
2H), 3.68-3.70 (m, 1H), 3.98 (ddd, J ) 10.9, 7.7, 2.0 Hz, 1H),
4.25 (ddd, J ) 10.9, 7.9, 2.4 Hz, 1H), 4.96 (d, J ) 8.1 Hz, 1H),
5.96 (d, J ) 5.3 Hz, 1H), 7.04 (t, J ) 7.3 Hz, 1H), 7.09 (dd, J
) 8.7, 1.1 Hz, 2H), 7.24 (dd, J ) 8.7, 7.3 Hz, 2H), 7.46 (dd, J
) 8.6, 7.4 Hz, 2H), 7.57 (t, 7.4 Hz, 1H), 8.02 (dd, J ) 8.6, 1.4
Hz, 2H) ppm. 13C NMR (400 Hz, CDCl3): 28.8 (t, 1C), 51.1 (d,
1C), 69.6 (t, 1C), 72.9 (d, 1C), 105.1 (d, 1C), 118.7 (d, 2C), 124.3
(d, 1C), 128.4 (d, 2C), 128.7 (d, 2C), 129.2 (d, 2C), 134.2 (d,
1C), 135.6 (s, 1C), 148.5 (s, 1C), 194.2 (s, 1C) ppm. Anal. Calcd
for C18H17NO3: C, 73.20; H, 5.80; N, 4.75. Found: C, 72.90;
H, 5.82; N, 4.61.
1159, 1088, 954, 758, 692 cm-1
.
1H NMR (400 Hz, CDCl3):
1.89-2.00 (m, 1H), 2.03-2.20 (m, 2H), 2.23-2.34 (m, 1H), 2.85
(dd, J ) 13.1, 6.2 Hz, 1H), 2.91 (dd, J ) 13.1, 10.1 Hz, 1H),
3.96-4.01 (m, 1H), 4.10-4.15 (m, 1H), 4.78 (dd, J ) 10.1, 6.2
Hz, 1H), 6.96 (t, J ) 7.3 Hz, 1H), 7.00 (dd, J ) 8.7, 1.0 Hz,
2H), 7.23 (dd, J ) 8.7, 7.3 Hz, 2H), 7.45 (dd, J ) 8.6, 7.4 Hz,
2H), 7.56 (t, 7.4 Hz, 1H), 8.17 (d, J ) 8.6 Hz, 2H) ppm. 13C
NMR (360 Hz, CDCl3): 25.1 (t, 1C), 35.0 (t, 1C), 43.3 (t, 1C),
69.1 (t, 1C), 72.5 (d, 1C), 113.0 (s, 1C), 116.6 (d, 2C), 123.4 (d,
1C), 128.6 (d, 2C), 128.8 (d, 2C), 129.3 (d, 2C), 134.2 (d, 1C),
135.6 (s, 1C), 150.8 (s, 1C), 197.2 (s, 1C) ppm. Anal. Calcd
for C19H19NO3: C, 73.77; H, 6.18; N, 4.53. Found: C, 73.83;
H, 5.92; N, 4.26.
3-B e n zo y l-N -p h e n y l-2-a za -1-o x a s p ir o [4.4]n o n a n e
(8a ): yield ) 85%, mp 84-85 °C. IR (thin film): 3063, 2961,
2872, 1696, 1674, 1597, 1449, 1337, 1219, 1182, 910, 752, 731,
692 cm-1 1H NMR (400 Hz, CDCl3): 1.41-1.50 (m, 1H), 1.60-
.
1.71 (m, 2H), 1.80-1.93 (m, 4H), 2.12-2.18 (m, 1H), 2.69 (dd,
J ) 12.0, 8.5 Hz, 1H), 2.72 (dd, J ) 12.0, 8.3 Hz, 1H), 4.90
(dd, J ) 8.5, 8.3 Hz, 1H), 6.87 (t, J ) 7.3 Hz, 1H), 6.94 (dd, J
) 8.8, 1.0 Hz, 2H), 7.22 (dd, J ) 8.8, 7.3 Hz, 2H), 7.47 (dd, J
) 8.4, 7.4 Hz, 2H), 7.58 (t, 7.4 Hz, 1H), 8.20 (dd, J ) 8.4, 1.2
Hz, 2H) ppm. 13C NMR (400 Hz, CDCl3): 24.0 (t, 1C), 24.4 (t,
1C), 35.8 (t, 1C), 37.2 (t, 1C), 44.4 (t, 1C), 71.9 (d, 1C), 92.8 (s,
1C), 113.4 (d, 2C), 120.9 (d, 1C), 128.6 (d, 2C), 128.8 (d, 2C),
129.3 (d, 2C), 134.4 (d, 1C), 134.4 (s, 1C), 151.8 (s, 1C), 198.0
(s, 1C) ppm. Anal. Calcd for C20H21NO2: C, 78.15; H, 6.89;
N, 4.56. Found: C, 78.41; H, 7.02; N, 4.43.
Ack n ow led gm en t. We are grateful to the National
Institute of General Medical Sciences, National Insti-
tutes of Health, for financial support of this research,
to Professor Hans-Ulrich Reissig for bringing this
problem to our attention, and to the UCLA Office of
Academic Computing and the National Center for
Supercomputing Applications (NCSA) for the computer
time used for this work.
tr a n s-3-Ben zoyl-N-p h en yl-2-a za -1,6-d ioxa zsp ir o[4.4]-
n on a n e (8b): yield ) 72%, mp 99-100 °C. IR (thin film):
J O971408Q