Kinetic and Computational Studies of a Novel Pseudopericyclic Electrocyclization. The First Evidence for Torquoselectivity in a 6-? System

Full text of DOI:10.1021/ja972701a

12366
J. Am. Chem. Soc. 1997, 119, 12366-12367
Kinetic and Computational Studies of a Novel
Pseudopericyclic Electrocyclization. The First
Evidence for Torquoselectivity in a 6-π System
Table 1. Experimental Rate Constants and Calculated Activation
Energies and Heats of Reaction for the Thermal Electrocyclizations
of R- and/or â-Substituted o-Vinylphenyl Isocyanates in C D
6 6
Lian Luo, Michael D. Bartberger, and
William R. Dolbier, Jr.*
Department of Chemistry, UniVersity of Florida
P.O. Box 117200, GainesVille, Florida 32611-7200
calcda
q
∆
G
substituent
(
kcal/
E
a
E
rxn
E
rxn
ReceiVed August 4, 1997
R
â
T, °C
k (s-1)
mol) (1f2) (1f2) (1f3)
-
-
-
-
5
5
6
5
Our research program has long maintained an interest in
structure-reactivity relationships with respect to electrocyclic
reactions. In earlier work, substituent effects in the 4-π
electrocyclic interconversion of cyclobutene and butadiene were
H
H
H
H
H
CH
F
CF
H
H
F
CN
CHO
H
109.4 9.8(0.5) × 10
112.6 7.3(0.5) × 10
178.9 2.0(0.3) × 10
182.3 9.0(0.5) × 10
29.8 29.8 19.2 -23.1
30.1 29.1 20.8 -25.3
38.6 32.7 25.5 -21.1
35.5 32.8 22.2 -24.4
23.2 26.1 16.9 -24.3
33.0 31.9 22.8 -21.3
28.0 29.0 23.0 -22.0
34.7 27.2 -20.8
3
3
-
4 b
CH
3
35 2.4 × 10
1
examined, with this work and that of others culminating in the
-
5
CF
CH
H
H
OCH
3
149.3 6.9(0.2) × 10
2
invention of the concept of torquoselectivity by Houk in 1984.
-4
3
101.6 3.80(0.01) × 10
More recently we have been interested in designing experi-
ments which could provide similar insight to 6-π electrocy-
clizations. Houk has predicted similar, albeit diminished,
32.3 23.5 -23.1
3
21.9 16.3 -23.5
3
torquoelectronic effects for such processes. Unfortunately,
a
MP2/6-31G*//RHF/6-31G* + ZPE. b See footnote 15.
steric effects play a dominant role in determining the dynamics
of the classic 1,3,5-hexatriene to 1,3-cyclohexadiene electro-
cyclization process, to such an extent that the impact of
reaction.10 In view of the novelty of the system and its potential
for polar effects, we believed that it was essential to carry out
a computational study concurrent with the experimental one.
Utilizing ab initio [MP2/6-31G*//RHF/6-31G*] methodology,1
the structures and energies of the ground states, transition
structures, and products of the cyclization/rearrangement se-
quence (1 f 2 f 3) for 1 and a number of its R- and
4
torquoselectivity cannot be assessed via studies of this system.
In pursuit of an electrocyclization system which would allow
1-13
unambiguous evaluation of 6-π electron torquoselectivity, we
initiated an investigation of substituent effects on the rates of
cyclization of o-vinylphenyl isocyanates.5
-7
Lacking cis (or
trans) substituents at the isocyanate terminus of its 6-π system,
this rearrangement system should be free of the usual steric
effects which have impeded past efforts to examine 6-π electron
1
4
â-substituted derivatives were computed.
Table 1 provides the kinetic and computational data which
were obtained in our investigation of the reactivity of such R-
and â-substituted vinylphenyl isocyanates. From this data, it
can be seen that substitution, particularly by polar substituents,
gives rise to strong kinetic effects which can largely be attributed
to polar influences on the thermodynamics of the endothermic,
4
torquoselectivity.
Parent system 1 underwent thermal rearrangement smoothly
to 2-quinolinone product 3,8,9 with the reaction exhibiting good
first-order behavior. Activation parameters were obtained which
1
7,18
rate-determining step of these processes.
Interestingly, these transition structures (i.e., Figure 1) proved
not to be of a classic, disrotatory, 6-π electrocyclic nature.
Instead, the isocyanato function remained essentially coplanar
with the benzene ring (-3.7° out of plane) in the transition state,
(
10) The rates of electrocyclization were similar with and without the
added pyridine ((10%), but the Arrhenius behavior was somewhat better
were consistent with a concerted, pericyclic pathway for the
in the presence of pyridine.
(
11) All calculations were performed with the GAMESS12 and Gaussian
3
(
1) Dolbier, W. R., Jr.; Koroniak, H.; Burton, D. J.; Bailey, A. R.; Shaw,
941 program systems, on Silicon Graphics Power Challenge XL and IBM
RS/6000 SP compute servers, respectively. Stationary points were optimized
at the restricted Hartree-Fock (RHF) level and characterized as minima or
transition structures by harmonic frequency analysis. RHF vibrational
frequencies and zero-point energy corrections are scaled by 0.8929. MP2
energy calculations on the RHF geometries utilized the frozen-core approach.
(12) General Atomic and Molecular Electronic Structure System, 22
November 1995 and 18 March 1997 versions: Schmidt, M. W.; Baldridge,
K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. H.; Koseki, S.;
Matsunaga, N.; Nguyen, K. A.; Su, S.; Windus, T. L.; Dupuis, M.;
Montgomery, J. A., Jr. J. Comput. Chem. 1993, 14, 1347-1363.
(13) Gaussian 94 (Revision C.3): Frisch, M. J.; Trucks, G. W.; Schlegel,
H. B.; Gill, P. M. W.; Johnson, 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.; Peng, C.
Y.; Ayala, P. A.; 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.
(14) B3LYP and BLYP investigations of this system yield very similar
transition structures, as do calculations of the transition structure of the
non-benzo analogue.
(15) The R-methyl derivative rearranged at a rate competitive with that
of the Curtius rearrangement, and thus its rate could only be estimated using
the standard procedure for such two-step consecutive, irreversible pro-
cesses.16
(16) Moore, J. W.; Pearson, R. G. Kinetics and Mechanism, John Wiley
& Sons: New York, 1981; p 290.
G. S.; Hansen, S. W. J. Am. Chem. Soc. 1984, 106, 1871-1872. Dolbier,
W. R., Jr. J. Am. Chem. Soc. 1987, 109, 219-225.
(
2) Kirmse, W.; Rondan, N. G.; Houk, K. N. J. Am. Chem. Soc. 1984,
1
1
06, 7989-7991. Rondan, N. G.; Houk, K. N. J. Am. Chem. Soc. 1985,
07, 2099-2111. Dolbier, W. R., Jr.; Koroniak, K.; Houk, K. N.; Sheu, C.
Acc. Chem. Res. 1996, 29, 471-477.
(
3) Evanseck, J. D.; Thomas, B. E., IV; Spellmeyer, D. C.; Houk, K. N.
J. Org. Chem. 1995, 60, 7134-7141.
4) Dolbier, W. R., Jr.; Palmer, K.; Koroniak, H.; Zhang, H. W.; Goedken,
(
V. J. Am. Chem. Soc. 1991, 113, 1059-1060. Dolbier, W. R., Jr.; Palmer,
K. W. J. Am. Chem. Soc. 1993, 115, 9349-9350. Dolbier, W. R., Jr.; Palmer,
K. W. Tetrahedron Lett. 1993, 34, 6201-6204.
(
5) There is literature precedent for 6-π electrocyclizations involving the
6
isocyanato group, although not specifically of an o-vinylphenyl isocyanate
type.
(
6) Overman, L. E.; Tsuboi, S. J. Am. Chem. Soc. 1977, 99, 2813-2815.
Eloy, F; Deryckere, A. HelV. Chim. Acta 1969, 52, 1755-1762. Eloy, F.;
Deryckere, A. J. Heterocycl. Chem. 1970, 7, 1191-1193. MacMillan, J.
H.; Washburne, S. S. J. Org. Chem. 1973, 38, 2982-2984.
(7) All isocyanate substrates (1) were prepared in situ via thermal Curtius
rearrangements of the respective acyl azides at room temperature in C6D6.
8) Rates of rearrangement of the isocyanates to the respective 2-quino-
(
linones (3) were determined by measuring the decrease of an appropriate
1
19
NMR ( H or F) signal with respect to an internal standard. Quinolinone
products were isolated in >85% yield and fully characterized.
(
9) 2-Quinolinones are often erroneously referred to as their less stable
tautomers, 2-hydroxyquinolines.
S0002-7863(97)02701-7 CCC: $14.00 © 1997 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12367
Table 2. Comparison of the Rate Constants for Electrocyclization
of the E- and Z-Isomers of Some â-Substituted o-Vinyl Phenyl
Isocyanates2
6,27
calcda,b
substituent
∆
∆Erxn
q
R
â
T, °C
k
E
/k
Z
∆∆G
∆∆E
a
(1f2)
H
H
CH
H
CH
3
112.6
182.3
101.6
11.7
3.2
82.1
-1.9
-1.0
-3.3
-2.0
-0.9
-7.4
+2.2
+2.0
+2.9
+0.6
+3.2
CF
F
3
3
Figure 1. Transition structure for the electrocyclization of o-vinylphe-
CHO
nyl isocyanate.
a
kcal/mol. b MP2/6-31G*//RHF/6-31G* + ZPE.
with the vinyl group twisted (28.2° out of plane),20 thus orienting
its terminal p-orbital for favorable overlap with the carbon
p-orbital of the carbonyl π-bond, rather than with that of the
CdN π-bond, which would have been the case for the classic,
disrotatory process. The short bond length (1.883 Å) of the
forming bond in the isocyanate cyclization is indicative of the
lateness of this transition state, whereas the length of the forming
bond (2.118 Å) in the pericyclic cyclization of o-divinylbenzene
is typical of such pericyclic reactions.2
isocyanates would thus not appear to be classic, disrotatory
pericyclic processes, they should still have the potential to
exhibit torquoselectivity in their monorotatory processes. In-
deed, the results given in Table 2, which provide a measure of
the torquoelectronic influence of E versus Z â-substituents on
the electrocyclization of 1, are such that there can be little doubt
that their source must be torquoelectronic in nature.
The kE/kZ ratios are clearly not steric in origin, since the
smallest substituent (F) gives rise to the largest, whereas the
largest (CF3) gives rise to the smallest observed kinetic effect!
Second, the results are consistent with both Houk’s general
expectations for a smaller degree of torquoselectivity in 6-π
1,22
Such a transition structure permits the maintenance of
conjugation within the phenyl isocyanate system, where one of
the carbonyl nonbonded pairs becomes a bonding pair as the
new σ-bond is formed. This description is consistent with
Lemal’s definition of a pseudopericyclic process as one in which
3,28
electrocyclizations and our own calculations for these specific
11-14,29
2
3
systems (Table 2).
Another factor which probably serves
“
nonbonding and bonding atomic orbitals interchange roles”.
to diminish the influence of torquoelectronics in this specific
system is the significant endothermicity of the reaction,30 which
to a significant extent is derived from the loss of amide-like
The above-described process of σ-bond formation gives rise to
a “disconnection” in the cyclic array of overlapping orbitals
because the atomic orbitals which are switching functions are
mutually orthogonal.
3
2
stabilization of the isocyanate group.
The potential importance of pseudopericyclic reactions, while
In conclusion, computational and experimental evidence has
been presented for a novel pseudopericyclic mechanism being
involved in the electrocyclizations of o-vinylphenyl isocyanates,
such reactions also providing the platform for obtaining the first
evidence of torquoselectivity in a 6-π electrocyclization process.
24
being acknowledged occasionally since Lemal’s invention, has
recently been given new emphasis through a series of papers
by Birney,2 who has concluded that, as in the electrocyclization
of 1, “when a pericyclic reaction has even one possible orbital
disconnection, this may result in a pseudopericyclic transition
structure.”
5
Acknowledgment. Support of this research in part by the National
Science Foundation is acknowledged with thanks. We are also grateful
to Academic Computing and Network Services at Florida State
University. Insightful discussions with Professor Birney regarding
pseudopericyclic reactions and early communication of his results in
this area are greatly appreciated.
Although the 6-π electrocyclizations of our o-vinylphenyl
(17) The rate-determining (and thus irreversible) nature of the electro-
cyclization step (1 f 2) was demonstrated by (a) the lack of significant
kinetic effect of added base, pyridine and (b) a study of the cyclization of
cis- and trans-â-deuterio 1, which gave no indication of reversibility. We
believe the proton-transfer step which converts 2 f 3 to be a bimolecular
process because, according to our calculations, the concerted [1.5]-H
migration transition structure is much too high in energy (Ea(2f3) ) 31 kcal/
mol) to be responsible for this process.
JA972701A
(
18) The recognized inductive influences of substituents such as F and
(26) The rate constants for rearrangement of the respective E- and
1
9
8
CF3 on carbonyl stability are sufficient to understand the impact of these
Z-isomers were obtained as described earlier using a ∼50:50 mixture of
substituents on the stability of cyclization products (2).
isomers.
(
19) Chambers, R. D. Fluorine in Organic Chemistry; John Wiley &
(27) Although selectivity was certainly observed for the E- vs Z-
cyclization of the simple â-fluoro system, the system did not exhibit good
Arrhenius behavior at the high temperatures required for their rearrangement,
and we therefore do not feel comfortable presenting such data at this time.
The R-methyl-â-fluoro system, rearranging at a much lower temperature,
had much more reliable kinetic behavior.
Sons: New York, 1973.
(20) This can be compared with our comparable computed dihedral angles
of 33.0° for both vinyl groups in the “normal” pericyclic transition structure
for rearrangement of o-divinylbenzene.
(
21) Houk, K. N.; Evanseck, J. D. Angew. Chem., Int. Ed. Engl. 1992,
3
1, 682-708.
(28) Our kE/kZ ratio of 11 for the â-methyl substituent can be compared
with the torquoselective factor of 120 which was observed for outward
versus inward rotation of a 3-methyl substituent in the 4-π electrocyclic
ring opening of cyclobutene.2
(29) Whereas the experimental and calculated values for kE/kZ correlate
well for â-methyl and â-trifluoromethyl substituents, for reasons we do
not yet understand, this ratio seems to be overestimated in calculations for
the â-fluoro substituent.
(30) The more endothermic a process, the later its transition state and
the more likely that the kinetics of the process will be determined by the
thermodynamics of the reaction, rather than by orbital interactions, which
generally exert their greatest influence in exothermic reactions which have
early transition states.31
(22) Further calculations have indicated that similar transition structures
should be involved for the ketene and, perhaps surprisingly, even for the
allene analogues. In the allene transition structure the terminal CH2 of the
allene has twisted 60.5° from its beginning orthogonal position to provide
the overlap required for exo-methylene double-bond formation!
(
23) Ross, J. A.; Seiders, R. P.; Lemal, D. M. J. Am. Chem. Soc. 1976,
8, 4325-4327. Bushweller, C. H.; Ross, J. A.; Lemal, D. M. J. Am. Chem.
Soc. 1977, 99, 629-631.
24) Okamura, W. H.; Peter, R.; Reischl, W. J. Am. Chem. Soc. 1985,
9
(
1
5
1
07, 1034-1041. Elnagar, H. Y.; Okamura, W. H. J. Org. Chem. 1988,
3, 3060-3066. Henriksen, U.; Snyder, J. P.; Halgren, T. A. J. Org. Chem.
981, 46, 3767-3768. Burke, L. A.; Elguero, J.; Leroy, G.; Sana, M. J.
Am. Chem. Soc. 1976, 98, 1685-1690.
(31) Houk, K. N. Acc. Chem. Res. 1975, 8, 361-391.
(32) Apropos, the computed ∆Erxn(1f2)’s for the ketene and allene
analogues are exothermic (-3.7 and -11.1 kcal/mol, respectively).
(
25) Birney, D. M.; Ham, S.; Unruh, G. R. J. Am. Chem. Soc. 1997,
1
19, 4509-4517. Birney, D. M. J. Org. Chem. 1996, 61, 243-251.