An experimental and theoretical study of the type C enone rearrangement: Mechanistic and exploratory organic photochemistry

Article abstract of DOI:10.1021/ja028631b

We recently described a new photochemical rearrangement which we termed a Type C process. The reaction involves a δ to α aryl migration in 5-disubstituted cyclohexenones also having bulky C-3 substituents. In contrast to most cyclohexenone rearrangements, the reaction occurs via a twisted π-π* excited triplet rather than the usual n-π* state. The electronic nature of the rearrangement was assessed using migration selectivity with p-anisyl and p-cyanophenyl groups. A synthesis of the reactants was elaborated, and the product structures were established by X-ray and NMR analysis. The reaction mechanism was established further with DFT and CASSCF computations. In the latter, localized NBO basis orbitals permitted proper selection of the active space. The nature of the diradical intermediates as well as the transition states was established computationally. Sensitization experiments with regioselectivities the same as those in direct irradiation confirmed the triplet multiplicity of the process.

Full text of DOI:10.1021/ja028631b

Published on Web 04/11/2003
An Experimental and Theoretical Study of the Type C Enone
Rearrangement: Mechanistic and Exploratory Organic
Photochemistry1
Howard E. Zimmerman* and Evgueni E. Nesterov
Contribution from the Chemistry Department of the UniVersity of Wisconsin,
Madison, Wisconsin 53706
Received September 20, 2002; E-mail: Zimmerman@chem.wisc.edu
Abstract: We recently described a new photochemical rearrangement which we termed a Type C process.
The reaction involves a δ to R aryl migration in 5-disubstituted cyclohexenones also having bulky C-3
substituents. In contrast to most cyclohexenone rearrangements, the reaction occurs via a twisted π-π*
excited triplet rather than the usual n-π* state. The electronic nature of the rearrangement was assessed
using migration selectivity with p-anisyl and p-cyanophenyl groups. A synthesis of the reactants was
elaborated, and the product structures were established by X-ray and NMR analysis. The reaction
mechanism was established further with DFT and CASSCF computations. In the latter, localized NBO
basis orbitals permitted proper selection of the active space. The nature of the diradical intermediates as
well as the transition states was established computationally. Sensitization experiments with regioselectivities
the same as those in direct irradiation confirmed the triplet multiplicity of the process.
Scheme 1. Type C Enone Rearrangement and Its Proposed
Mechanism
Introduction
Our previous study2 uncovered the Type C rearrangement
and suggested a mechanism. According to this mechanism, the
reaction occurs on a triplet hypersurface with the π-π* excited
triplet species 2T being a key intermediate for this reaction.
The subsequent phenyl migration through the half-migrated
species 3T occurs on the triplet hypersurface, and only after
the migration is complete, or near completion, does intersystem
crossing lead the system to the S0 ground-state surface con-
comitant with ring-closure to diastereomeric products 5a and
6a (note Scheme 1).
To check this mechanism experimentally, it was decided to
study relative migratory aptitudes of aryl groups. 4-Methoxy-
phenyl and 4-cyanophenyl substituents have been used previ-
ously to establish mechanisms of other photochemical trans-
formations: for example, the Type A dienone,3 Type B enone,4
and Type B bicyclic rearrangements.5 Thus, it was natural to
use this approach presently in the case of the Type C Enone
Rearrangement. The two enones 1b and 1c were synthesized
with the idea that the relative preference for δ to R migration
of p-substituted phenyl vs phenyl groups in these compounds
would provide information regarding the electronic factors
controlling this reaction.
Synthesis
(1) This is publication 200 of our Photochemical Series and 269 of the General
Papers.
The previously designed2 synthesis of the parent Type C
enone 1a (Scheme 2) has some evident drawbacks. First, it is
not very efficient and convenient since in the key step, allylic
oxidation of cyclohexene 8, the desired cyclohexenone 9 is only
a minor product with the major product being enone 7, the ratio
of 7 to 9 being 4:1. This approach also poses difficulties in
introducing para groups such as cyano.
(2) Zimmerman, H. E.; Nesterov, E. E. J. Am. Chem. Soc. 2002, 124, 2818-
2830.
(3) (a) Zimmerman, H. E.; Schuster, D. I. J. Am. Chem. Soc. 1961, 83, 4486-
4487. (b) Zimmerman, H. E.; Schuster, D. I. J. Am. Chem. Soc. 1962, 84,
4527-4540. (c) Zimmerman, H. E.; J. O. Grunewald, J. O. J. Am. Chem.
Soc. 1967, 89, 3354-3356.
(4) (a) Zimmerman, H. E.; Wilson, J. W. J. Am. Chem. Soc. 1964, 86, 4036-
4042. (b) Zimmerman, H. E.; Rieke, R. D.; Scheffer, J. R. J. Am. Chem.
Soc. 1967, 89, 2033-2047. (c) Zimmerman, H. E.; Hancock, K. G. J. Am.
Chem. Soc. 1968, 90, 3749-3760. (d) Zimmerman, H. E.; Lewin, N. J.
Am. Chem. Soc. 1969, 91, 879-886. (e) Zimmerman, H. E.; Elser, W. R.
J. Am. Chem. Soc. 1969, 91, 887-896.
Hence the new synthetic approach in Scheme 3 was devel-
oped. This synthesis began with the addition of the aryllithium
or phenylmagnesium bromide to 3-ethoxycyclohexenone 10. A
key step of this synthesis is the conjugate addition of lithium
(5) (a) Zimmerman, H. E.; Grunewald, J. O. J. Am. Chem. Soc. 1967, 89, 3354-
3356. (b) Zimmerman, H. E.; Nasielski, J.; Keese, R.; Swenton, J. S. J.
Am. Chem. Soc. 1966, 88, 4895-4903.
9
5422
J. AM. CHEM. SOC. 2003, 125, 5422-5430
10.1021/ja028631b CCC: $25.00 © 2003 American Chemical Society

Type C Enone Rearrangement
A R T I C L E S
Scheme 2. Previous Synthesis of the Type C Enone 1a
Table 1. Calculated and Experimental 1H NMR Chemical Shifts
for Diastereomeric Photoproducts 5a and 6aa,b
atom
δ
calcd, ppm
δ
expt, ppm
atom
δ
calcd, ppm
δ
expt, ppm
H2a
H2b
H4c
H6a
H6b
2.48
3.34
3.48
1.78
0.35
2.73
3.39
3.62
1.87
0.41
H2a
H2b
H4c
H6a
H6b
2.64
2.99
4.31
2.13
1.11
2.73
3.15
4.34
1.92
1.11
a GIAO RHF/6-31G(df,pd)//B3LYP/6-31G(d,p). b Calculated energies
(B3LYP/6-31G(d,p)) for 5a -928.0220560 a.u., for 6a -928.0258215 a.u.
c Note a remarkably high downfield shift for endo proton H4 in 6a compared
with exo proton in 5a.
Scheme 3. Improved Synthesis of 3-tert-Butyl-5,5-diaryl Enones
1a-c
Figure 1. Alternative half-migrated species.
of the two diastereomeric bicyclic ketones 5b and 6b in an
approximate ratio of 1:1. No products corresponding to phenyl
group migration were found for this reaction up to 75%
conversion. The products and their ratio were not changed when
the irradiation was performed in the presence of 10-fold excess
of acetophenone triplet sensitizer. Note Scheme 4.
Scheme 4. Photochemistry of Para-Substituted 5-Aryl-5-phenyl
Cyclohexenones 1b-c
Interestingly, the corresponding photolysis of cyanophenyl
enone 1c again led only to cyanophenyl migration products and,
again, afforded two diastereomers 5c and 6c in an approximately
1:1 ratio (see Scheme 4). Similar to the case of the anisyl
counterpart, acetophenone sensitization led to the same products
in the same ratio as that in direct irradiation.
It turned out to be very easy to experimentally monitor the
1
Type C photochemical rearrangement by H NMR and even
diphenylcuprate to enones 11a-c to form 3,3-diarylcyclohex-
anones 12a-c. Curiously, the reverse sequence with para-
substituted cuprates reacting with enone 11a afforded product
only in poor or vanishing yields.
The enones 14a-c were obtained by bromination-dehydro-
bromination directly under kinetically controlled conditions and
also via the silyl enol ether 13. Finally, the desired â-tert-butyl
group was introduced by conjugate addition of lithium di-tert-
butyl cuprate to enones 14a-c. This was followed by selene-
nylation of the copper enolate followed by oxidative selenoxide
elimination.6
assign the product distribution as the reaction proceeds. This is
possible because a singlet peak corresponding to an endo proton
H4 in 6a-c is deshielded and shifted almost 1 ppm downfield
as compared with exo-H4 in 5a-c (note Table 1). Remarkably,
this deshielding effect is perfectly predicted by GIAO ab initio
computation of NMR shifts for the photoproducts 5a and 6a
(Table 1).
Computational Results: Density Functional and Nem-
ukhin-Weinhold CASSCF/NBO Methods. In view of the new
reaction encountered with its complex mechanism, computa-
tional treatments proved both necessary and inviting. For
example, the preferential migration of the substituted aryl groups
in the Type C reaction of enones 1b and 1c needed to be
understood. It is clear, qualitatively, that the half-migrated triplet
species 3T-1 should be lower in energy than its counterpart 3T-2
due to the greater resonance stabilization (note Figure 1).
However, their relative stabilities must differ enough to account
Results
Irradiation of p-methoxyphenyl 1b and p-cyanophenyl 1c
enones in benzene solution proceeded smoothly to give only
products derived from Type C enone rearrangements, thus
providing further support for the rearrangement generality. See
Scheme 4. For anisyl enone 1b the reaction occurred selectively
with exclusive migration of the anisyl group and with formation
(7) For example, in a photorearrangement of 4-(4-cyanophenyl)-4-phenyl-
cyclohex-2-enone the ratio of cyanophenyl to phenyl migration is 15:1,
see ref 4.
(6) Reich, H. J.; Renga, J. M.; Reich, I. L. J. Am. Chem. Soc. 1975, 97, 5434-
5447.
9
J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003 5423

A R T I C L E S
Zimmerman and Nesterov
(14,14) active space
Table 2. Comparison of Energies of Reacting Intermediates 3T for Two Competing Pathwaysa,b
CASSCF/NBO computationsc
DFT (UB3LYP) computations
(10,10) active space
absolute
migration
absolute
relative
relative
absolute
energy, a.u.
relative
enone
intermediate
energy, a.u.
energy, kcal/mol
energy, a.u.
energy, kcal/mol
energy, kcal/mol
1b
3T-1
3T-2
3T-1
3T-2
-1042.439049
-1042.438111
-1020.165753
-1020.161048
0
0.6
0
-1035.809534d
-1035.808699d
-1013.696226
-1013.692251
0
0.5
0
-
-
1c
-1013.739082
-1013.736363
0
1.7
3.0
2.5
a 6-31G* (6-31+G* for N and O) basis set was used throughout the computations. b Geometries were optimized at UB3LYP level of theory. c Single-
point energy computations. d Computation with (10,9) active space.
for such an absolute migratory preference.7 This is just one of
many questions we needed to consider computationally.
To assess the problem of the relative energies of 3T-1 and
3T-2, we ran unrestricted density functional (i.e., DFT) methods
with a B3 hybrid exchange-correlation functional and a LYP
correlation functional (UB3LYP). This led to the results given
in Table 2 (columns 3 and 4). The energy differences predicted
a lower energy for migration of the para-substituted aryl group
in agreement with experiment.
Considering the complex nature of biradical triplets 3T-1 and
3T-2 with possibly more than one configuration needed for an
adequate description of them, it was of real interest to compare
CASSCF results with those of the density functional. Normally,
multiconfigurational computations on such complex systems as
involved in the present study require an approach with a very
large active space. The basic requirement for a proper computa-
tion is that one should include all of those configurations
involving MOs from which and to which electron excitation is
appreciable. However, current restrictions permit inclusion of
only a limited number of orbitals into the active space (up to14
in Gaussian 98).
that, although the computation starts with localized MOs as an
initial guess, the subsequent MCSCF treatment leads to an
expansion of the MOs throughout 3T-1 and 3T-2. The results
of these computations are given in Table 2. For the cyanophenyl
enone 1c a (10,10) active space was first utilized. One can see
now that this method predicts the cyanophenyl migration to be
a more favorable process compared with phenyl migration
(preference 2.5 kcal/mol) and is quite close to the prediction of
the UB3LYP result of 3.0 kcal/mol. To check the validity of
the active space selection, we performed a rather lengthy
computation with an expanded (14,14) active space. The 14 bond
orbitals used as a basis consisted of the π and π* orbitals of
the overlapping cyano π bond, the three bonds (six orbitals) of
the nonmigrating benzene ring, the two π bonds (four orbitals)
of the migrating ring, and then the two lone pair orbitals. This
led to a 1.7 kcal/mol preference, suggesting that the (10,10)
computation was a reasonable approximation. In the case of
the methoxyphenyl enone 1b the CASSCF/NBO computation
with (10,9) active space gave a 0.5 kcal/mol preference for the
methoxyphenyl migration. Again, this result came very close
to the DFT prediction (0.6 kcal/mol).
Since molecules such as 3T-1 have a much larger number of
potentially important molecular orbitals, one needs a way to
select the orbitals which have the heaviest impact. An intuitive
selection of MOs to comprise the active set is not safe. However,
the dilemma is largely avoided using the Nemukhin-Weinhold
“Complete Orthonormal Set of Natural Bond Orbitals (NBOs)
Localized Basis Orbital” approach. This method uses those
group orbitals which are most involved in delocalization.9
Thus, these orbitals, being placed into the CASSCF active space,
contribute most to the correlation energy.
For example, for a conjugated enone moiety one needs to
consider only five localized molecular orbitals with a total of
six electrons. The MOs consist of four butadiene-like π orbitals
(two bonding and two antibonding) and one nonbonding ny
orbital on oxygen. Thus, if a CASSCF computation starts with
these localized MOs being placed into an active space, it will
require only a (6,5) active space to reliably describe the enone
moiety. Inclusion of a still larger active space adds σ framework
involvement. These localized orbitals nicely correspond to the
chemist’s Lewis structure picture (two-center “bonds” and one-
center “lone pairs”). Using NBOs8 allows starting the CASSCF
computation with those orbitals in the active space which are
of utmost importance and thus ensures getting reliable compu-
tational results with a minimal active space. It should be noted
In addition, for the cyanophenyl enone 1c, DFT computations
were utilized to establish the nature of the reactant excited states
and parts of the excited and ground-state hypersurfaces in
general. These results are given in Table 3. They will be
considered in the context of the reaction mechanism in the
Discussion section below.
Discussion of Computations
There are two aspects to the computation which should be
discussed. One is the practicality of the methodology employed,
and the second is discussion of the results themselves. In the
Results section, we have noted the utility of employing localized
group orbitals as a starting point in obtaining a CASSCF active
space which is within practical limits and yet includes the
important components of the configurational wavefunctions.
As an example, Figure 2 shows molecular orbitals (NBOs)
which were used for CASSCF/NBO(10,10) treatment of the
p-cyanophenyl migration intermediate 3T-1. It should be
additionally noted that, although the computation starts from
localized MOs, a subsequent SCF treatment leads to “expansion”
of these orbitals throughout the entire molecule, which ensures
that all the local factors (e.g., other substituents present) not
included originally into the consideration are now taken into
account. One example of this “expansion” is shown in Figure
3. During the CASSCF treatment an initial localized NBO
corresponding to the radical p-hybridized lone pair on carbon
C3 (referred to as “lone pair 2” in Figure 2) is transformed to
(8) NBO 5.0. Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. Theoretical Chemistry
Institute, University of Wisconsin, Madison, WI, 2001.
(9) Nemukhin, A. V.; Weinhold, F. J. Chem. Phys. 1992, 97, 1095-1108.
9
5424 J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003

Type C Enone Rearrangement
A R T I C L E S
Table 3. Energies of the Intermediate Species Included in the Photochemistry of Enone 1c
triplet hypersurface
singlet hypersurfaceb
abs energy, a.u. relative energy, kcal/mol
speciesa
abs energy, a.u.
relative energy, kcal/mol
Cyanophenyl Migration
A, π-π* twisted triplet 2T′
B, TS 2T f 3T-1
C, half-migrated species 3T-1
-1020.15387825
-1020.13144910
-1020.1531002
26.2
40.2
26.7
-1020.1955741
-1020.1533007
-1020.1339865
0
26.5
38.6
Phenyl Migration
D, π-π* twisted triplet 2T”
E, TS 2T f 3T-2
F, half-migrated species 3T-2
-1020.1539451
-1020.1297017
-1020.1483233
26.2
41.4
29.7
-1020.1956917
-1020.1462874
-1020.1267297
0
31.0
43.3
a Geometry of the triplet species was optimized with UB3LYP/6-31G* level. b B3LYP/6-31G* single-point computations on the corresponding triplet
geometries. Note: These computations used a less intricate basis than those in Table 2.
Figure 2. Natural bond orbitals (NBOs) used for CASSCF/NBO (10,10)
computation of cyanophenyl migration intermediate 3T-1. (14,14) computa-
tion additionally includes nonmigrating phenyl orbitals.
Figure 4. Schematic representation of two alternative reaction pathways
for the Type C rearrangement of cyanophenyl enone 1c. Solid lines
correspond to triplet hypersurface, and dased lines correspond to S0.
phenyl- and phenyl-migrated intermediates. Inclusion of the
methoxy lone pair from the methoxy group oxygen of the
nonmigrating aryl superficially seems superfluous but ensures
contribution to the correlation energies in both cases. That both
CASSCF/NBO and DFT computations gave almost the same
energy preference for the para-substituted phenyl migration (see
Table 1) provides support for the treatment.
The second important consideration is whether one can use
energies of half-migrated diradical intermediates 3T alone to
predict computationally the relative efficiencies of aryl migra-
tions. These intermediates represent local minima on triplet
hypersurfaces, while the relative rates and thus the migratory
aptitudes are determined by the heights of corresponding
activation barriers (i.e. transition state energies). However, for
many well-understood photochemical reactions of enones (e.g.
the Type B enone rearrangement) computational studies revealed
certain parallelisms between the energies of the half-migrated
intermediates and the corresponding transition states.11a Thus,
the much easier to compute half-migrated biradicals were used
instead of actual transition states for theoretical modeling of
the various photochemical processes.11b-d
Figure 3. Expansion and delocalization of localized NBOs during CASSCF
treatment: an orbital corresponding to lone pair on carbon C3 before (left)
and after (right) CASSCF/NBO computation.
a fully delocalized MO, which still, to some extent, keeps its
original character.
To be able to compare energies of two different species
obtained with CASSCF computations, one needs to use the same
active space size as well as to employ same basis orbitals in
the active space. This ensures accessibility of similar configura-
tions in both computations and inclusion of comparable cor-
relation effects. For this reason when computing the phenyl
migration intermediate 3T-2 (which lacks a cyano group on the
migrating aryl) corresponding π and π* molecular orbitals of
the cyano group from the second nonmigrating aryl were
included in the active space, thus keeping the same (10,10)
active space size which was used for the intermediate 3T-1 (see
Figure 2).10 Similarly, we used a (10,9) active space size during
the methoxyphenyl enone 1b computations both for methoxy-
To validate our use of the half-migrated intermediates 3T,
we ran DFT computations, for both T1 and S0, of part of the
reaction pathway for phenyl vs cyanophenyl migrations for
(10) The computations with extended (14,14) active space additionally included
the π system of the nonmigrating aromatic ring in the active space (see
Experimental Section for more details).
(11) (a) Zimmerman, H. E.; Kutateladze, A. G. J. Org. Chem. 1995, 60, 6008-
6009. (b) Zimmerman, H. E.; Sebek, P. J. Am. Chem. Soc. 1997, 119,
3677-3690. (c) Zimmerman, H. E.; Sebek, P.; Zhu, Z. J. Am. Chem. Soc.
1998, 120, 8549-8550. (d) Zimmerman, H. E.; Alabugin, I. V.; Smolen-
skaya, V. N. Tetrahedron 2000, 56, 8621-6831.
9
J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003 5425

A R T I C L E S
Zimmerman and Nesterov
Scheme 5. Two Alternative Migration Pathways for the Para-Substituted Enones 1b-c
enone 1c. The results are given in Table 3 and graphically
presented in Figure 4. We note that consideration of transition-
energy barriers gives a 1.2 kcal/mol preference for the cy-
anophenyl migration in parallel, but with somewhat smaller
results than those obtained with half-migrated biradicals 3T-1
and 3T-2.
Interestingly, these computations also support our suggestion
that the key stepsaryl migrationsoccurs on triplet hypersurface.
From Figure 4 we see that the initial π-π* triplet can
intersystem-cross into a ground-state surface only after it has
moved past the transition-state barrier. A possible intersystem-
crossing to S0 might occur at the point after the transition state
where the triplet and ground-state surfaces cross (see Figure
4). However, were it to occur, this would revert the molecule
downhill back to the starting ground-state enone. Thus, for the
rearrangement to occur, it must proceed through the triplet
intermediate 3T. A subsequent intersystem crossing leading to
final products probably further occurs somewhere near the ring-
opened intermediate 4T.
species. In the present study the source of the difference seems
to arise from steric enhancement of enone twisting about the
R,â-bond of the â-tert-butyl substituted enones. For these
compounds, in contrast to the ordinary n-π* triplets, the π-π*
triplet 2T has the excitation primarily localized on the CdC
double bond with strong diradical character of this bond and
high electron density on the R-carbon. That this electronic
feature is indeed a consequence in twisting of the π CdC bond
is supported by the CASSCF/NBO computations of the triplet.2
Moreover, the Type C rearrangement requires a bulky group
(as tert-butyl) attached to the â-carbon, which facilitates this
π-bond twisting and makes the π-π* triplet more energetically
favorable compared with the n-π* triplet.16
A priori, while there is no doubt that the reaction starts as a
triplet process, the rearrangement key stepsaryl migrations
might occur either completely on the triplet hypersurface, or at
some point with intersystem crossing to the ground-state S0
surface. Both possibilities are known in conjugated enone
photochemistry. For example, in a case of a typical Type B
enone rearrangement the aryl migration occurs in triplet state,4
while for the Type A dienone rearrangement, an aryl migration
occurs in a ground-state singlet zwitterion.3 Our test of relative
migratory aptitudes was designed to address this problem. In
the Type B enone rearrangement, it is known4d,e that the rate-
limiting step is formation of a triplet-bridged species which is
the counterpart to the one involved presently.
Discussion: Mechanistic Aspects
The first point to be made is that in our previous report of
the Type C rearrangement, the triplet excited state was noted
to undergo the reaction. That the regioselectivity is the same
for sensitized and direct irradiations, strongly suggests that the
triplet is the reactive species in the direct irradiations as well.
Second, our previous2 and present computations showed that
the reaction originates from a twisted π-π* triplet 2T (see
Scheme 5). The twisting of the π CdC bond minimizes energy
with the computed 35° dihedral angle for the triplet. Although
for some of the known enone photochemical processes12,13
participation of the twisted π-π* triplet has also been suggested
on experimental14 and theoretical15 grounds, this is quite
different from the more common n-π* triplet reactions of
conjugated enones where the n-π* triplet is the lower-energy
The experimentally found exclusive preference for the para-
substituted aryl groups to migrate independently of the electronic
character of the substituent supports the conclusion that this
migration occurs in triplet state, and not on a singlet, zwitterionic
hypersurface. Two different migration possibilities are depicted
in Scheme 5. With the reaction occurring via the intermediacy
of the triplet half-migrated intermediate 3T-1, both cyano and
methoxy substituents in the migrating group qualitatively are
expected to stabilize the developing odd-electron center in the
aromatic ring, thus lowering the energy of the intermediate 3T-
1. This explains why it is the para-substituted groups which
exclusively migrate in a course of Type C enone rearrangement.
(12) (a) Schuster, D. I.; Brown, R. H.; Resnick, B. M. J. Am. Chem. Soc. 1978,
100, 4504-4512. (b) Schuster, D. I. In Rearrangements in Ground and
Excited States; de Mayo, P., Ed.; Academic Press: London, 1980; Vol. 3,
p 167-279.
(13) (a) Rudolph, A.; Weedon, A. C. J. Am. Chem. Soc. 1989, 111, 8756-
8757. (b) Schuster, D. I.; Yang, J.-M.; Woning, J.; Rhodes, T. A.; Jensen,
A. W. Can. J. Chem. 1995, 73, 2004-2010.
(14) Schuster, D. I.; Dunn, D. A.; Heibel, G. E.; Brown, P. B., Rao, J. M.;
Woning, J.; Bonneau, R. J. Am. Chem. Soc. 1991, 113, 6245-6255.
(15) (a) Reguero, M.; Olivucci, M.; Bernardi, F.; Robb, M. A. J. Am. Chem.
Soc. 1994, 116, 2103-2114. (b) Reguero, M.; Bernardi, F.; Olivucci, M.;
Robb, M. A. J. Org. Chem. 1997, 62, 6897-6902.
(16) This is a necessary requirement for the Type C rearrangement to occur in
solution. However, in solid-state photochemistry this reaction may occur
even without a bulky substituent, when crystalline environment facilitates
it. One known example is the photoreaction of 4,5,5-triphenylcyclohex-2-
enone (see ref 2).
9
5426 J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003

Type C Enone Rearrangement
A R T I C L E S
as colorless crystals, mp 110-113 °C (lit.19 mp 114-115 °C). 1H NMR
(CDCl3) δ 7.33-7.14 (m, 10H), 2.96 (s, 2H), 2.58 (t, J ) 5.8 Hz, 2H),
2.36 (t, J ) 6.8 Hz, 2H), 1.75-1.62 (m, 2H).
Conclusions
The Type C enone rearrangement which includes long-range
δ to R aryl migration occurs as a triplet reaction. Preference of
para-substituted aryl groups to migrate independently of the
nature of the substituent supports the mechanism including a
key stepstransannular aryl migrationsto occur in triplet state
followed by intersystem crossing to a singlet ground state. The
CASSCF/NBO computational approach was shown to be a
useful and reliable way for treatment of medium-sized open-
shell biradical species where a multiconfigurational method is
desirable.
1-(Trimethylsilyloxy)-5,5-diphenylcyclohex-1-ene 13a. A solution
of n-BuLi (51.1 mL of 1.2 M solution in hexanes, 61.3 mmol) was
added at -70 °C to a stirred solution of 7.74 g (76.6 mmol) of
diisopropylamine in 40.0 mL of THF, followed by a stirring for 10
min at this temperature. To the resulting solution of LDA a solution of
33.4 g (38.8 mL, 0.306 mol) of trimethylchlorosilane in 40.0 mL of
THF was added followed by the dropwise addition of a solution of
7.66 g (30.6 mmol) of 3,3-diphenylcyclohexanone (12a) in 60.0 mL
of THF. The resulting mixture was stirred at -70 °C for 5 min,
quenched with 45.0 mL of triethylamine followed by the addition of
saturated aqueous solution of NaHCO3. The further workup included
extracting with pentane, washing with water and 0.1 M aqueous solution
of citric acid, and drying over Na2SO4. Concentration in vacuo afforded
crude solid product, which was recrystallized from pentane to give 7.42
g (75%) of 13a as colorless crystals, mp 102-104 °C. 1H NMR (CDCl3)
δ 7.29-7.11 (m, 10H), 4.81 (narrow m, 1H), 2.58 (s, 2H), 2.26 (t, J )
6.3 Hz, 2H), 1.81-1.71 (m, 2H), 0.22 (s, 9H).
Experimental Section
General Procedures. All reactions were performed under an
atmosphere of dry nitrogen. Melting points were determined in open
capillaries and are uncorrected. Column chromatography was performed
on silica gel (Aldrich, 60 Å, 200-400 mesh, or Silica gel 60 Geduran
35-75 µm) mixed with Sylvania 2282 phosphor and slurry packed
into quartz columns to allow monitoring with a hand-held UV lamp.
All solvents were additionally purified and dried by standard techniques.
1H and 13C NMR spectra were recorded at 300 and 75 MHz,
respectively, and are reported in ppm downfield from tetramethylsilane.
3-Ethoxycyclohex-2-enone 10 was prepared according to a pub-
lished procedure17 from 60.0 g (0.54 mol) of 1,3-cyclohexandione. This
gave 75.0 g (100%) of the product as a colorless oil, bp 110-110 °C/
4.5 mmHg (lit.17 bp 88 °C/2.5 mmHg).
3-Phenylcyclohex-2-enone 11a. A solution of phenylmagnesium
bromide was prepared from 55.0 g (0.35 mol) of bromobenzene in 200.0
mL of ether and 8.6 g (0.36 mol) of Mg in 100.0 mL of ether. To the
solution of this Grignard reagent a solution of 40.5 g (0.29 mol) of
3-ethoxycyclohex-2-enone (10) in 200.0 mL of ether was added
dropwise at room temperature. The resulting mixture was stirred at
reflux conditions for 3 h and then allowed to cool to room temperature.
It was carefully quenched with 150.0 mL of 10% HCl, followed by
addition of 200.0 mL of concentrated HCl. The resulting mixture was
stirred at room temperature for 2.5 h. Then 400.0 mL of ether and
800.0 mL of water were added, the organic fraction was separated and
washed successively with water, a saturated solution of NaHCO3, water,
and brine, and dried over Na2SO4. Concentration in vacuo gave crude
solid product which was recrystallized from ether-hexane mixture to
yield 37.5 g (75%) of 11a as a yellow crystalline material, mp 59-61
°C (lit.18 mp 64-65 °C).
3,3-Diphenylcyclohexanone 12a. A solution of PhLi (222.7 mL of
0.73 M solution in ether-hexane, 162.6 mmol) was added dropwise
to a stirred at 0 °C suspension of 15.53 g (81.3 mmol) of CuI in 200.0
mL of ether. The resulting mixture was stirred for 20 min, and a solution
of 10.0 g (58.1 mmol) of 3-phenylcyclohex-2-enone (11a) in 100.0
mL of ether was added dropwise, followed by stirring for 2.5 h at 0
°C. The reaction mixture was poured into a separation funnel filled
with a mixture of ice - concentrated aqueous NH4Cl and ammonia
solutions, and shaken vigorously for 10 min. The organic layer was
separated, the aqueous layer was extracted with ether, and the combined
organic fractions were washed with water and concentrated NaCl
solution, and dried over Na2SO4. After concentrating in vacuo, the oily
residue was crystallized from ether-hexane to give 5.3 g of crude
crystalline product. A mother liquor left after the crystallization was
concentrated and subjected to chromatography on silica gel (column
16.0 cm × 4.0 cm, eluent ethyl acetate-hexane 1:5). Fraction with Rf
0.40 gave additional 2.6 g of the product. Combined material was
recrystallized from ether-hexane mixture to yield 7.4 g (51%) of 12a
5,5-Diphenylcyclohex-2-enone 14a. A mixture of 3.43 g (9.7 mmol)
of 1-(trimethylsilyloxy)-5,5-diphenylcyclohex-1-ene (13a) and 2.81 g
(20.4 mmol) of K2CO3 in 250.0 mL of ether was stirred and cooled to
-75 °C, and a solution of 1.68 g (10.2 mmol) of bromine in 56.0 mL
of CCl4 was added via a dropping funnel as rapidly as possible
(temperature of the reaction mixture should remain below -65 °C).
When the addition was complete, the cold reaction mixture was poured
into saturated aqueous solutions of NaHCO3 and NaCl. The organic
layer was washed with water and dried over Na2SO4. This reaction
was repeated with another 3.43 g portion of the silyl ether 13a. The
combined product solutions from these two runs were concentrated and
dried in vacuo, which afforded 7.95 g of the bromide as a yellow oil.
This bromide was dissolved in 90.0 mL of anhydrous DMF, and to
this solution were added freshly vacuum-dried LiBr (6.34 g, 72.9 mmol)
and Li2CO3 (3.60 g, 48.6 mmol). The resulting mixture was stirred at
135 °C for 5 h. Then it was allowed to cool to room temperature, poured
into 600 mL of water, and extracted with two 250-mL portions of
ether-pentane (1:1); the combined organic fraction was washed with
water and dried over Na2SO4. Concentration in vacuo gave solid
product, which was recrystallized from CH2Cl2-hexane mixture to
afford 4.2 g (75% based on 13a) of 14a as colorless crystals, mp 104-
106 °C (lit.2 mp 104-105 °C).
3-tert-Butyl-5,5-diphenylcyclohex-2-enone 1a. The reaction of 3.50
g (14.1 mmol) of 5,5-diphenylcyclohex-2-enone (14a) was performed
according to a previously published procedure2, and gave 1.96 g (46%)
of 1a as colorless crystals, mp 102-104 °C (lit.2 mp 102-104 °C).
3-(4-Methoxyphenyl)cyclohex-2-enone 11b. A solution of 16.0 g
(85.7 mmol) of 4-bromoanisole in 100.0 mL of ether was added
dropwise to a stirred at 0 °C solution of n-BuLi (84.0 mL of 1.02 M
solution in hexanes, 85.7 mmol), and the resulting mixture was stirred
for 2 h at this temperature. Then the temperature was decreased to
-70 °C, and a solution of 10.0 g (71.4 mmol) of 3-ethoxycyclohex-
2-enone (10) in 60.0 mL of ether was added dropwise. After the addition
was complete, the reaction mixture was stirred for additional 2 h at
-70 °C, and the cold mixture was quenched with 70.0 mL of 10%
HCl. The reaction mixture was allowed to warm to room temperature
followed by addition of 70.0 mL of concentrated HCl. After stirring
for 1.5 h at room temperature, water (100 mL) and ether (200 mL)
were added, and the organic fraction was separated and washed
successively with water, saturated aqueous NaHCO3, water, and brine
and was dried over Na2SO4. Concentration in vacuo afforded a yellow
(17) Mattay, J.; Banning, A.; Bischof, E. W.; Heidbreder, A.; Runsink, J. Chem.
Ber. 1992, 125, 2119-2128.
(18) Okano, T.; Satou, Y.; Tamura, M.; Kiji, J. Bull. Chem. Soc. Jpn. 1997, 70,
1879-1886.
(19) Zimmerman, H. E.; Epling, G. A. J. Am. Chem. Soc. 1972, 94, 7806-
7811.
9
J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003 5427

A R T I C L E S
Zimmerman and Nesterov
solid, which was recrystallized from ether-hexane mixture to yield
9.5 g (66%) of 11b as colorless crystals, mp 81-83 °C (lit.20 mp 83-
84 °C).
reaction mixture was stirred at 0 °C for 3 h. A solution of PhSeBr was
prepared by dropwise addition of 2.53 g (0.81 mL, 15.8 mmol) of Br2
to a solution of 5.10 g (16.4 mmol) of diphenyldiselenide in 50.0 mL
of ether at intensive agitation.6 The resulting solution of PhSeBr was
rapidly added to the reaction mixture. After the stirring for an additional
5 min at 0 °C, the reaction mixture was poured into a mixture of 200.0
mL of 5% HCl and 200.0 mL of ether-pentane (1:1). The organic
fraction was washed with saturated NaHCO3 solution and water and
dried over Na2SO4. After concentrating in vacuo, the oily residue was
dissolved in 80.0 mL of CH2Cl2. Pyridine (5.17 g, 65.5 mmol) was
added, and a solution of H2O2 (prepared from 9.37 mL of 30% H2O2
and 8.0 mL of water, 91.7 mmol) was carefully added at intensive
stirring with ice-water cooling (the temperature of the reaction mixture
should be kept below 30 °C). After the addition was complete, the
resulting mixture was intensively stirred at room temperature for 30
min, and it was then poured into a mixture of 150.0 mL of 7% NaHCO3
and 150.0 mL of CH2Cl2. The aqueous layer was extracted with CH2-
Cl2, the combined organic fraction was washed successively with 10%
HCl, water, and brine and dried over Na2SO4. Concentration in vacuo
gave a dark-orange oil, which was separated by chromatography on
silica gel (column 42.0 cm × 4.0 cm, eluent ethyl acetate-hexane 1:3).
A fraction with Rf 0.36 was collected, which afforded 1.08 g of crude
1b as an orange oil. It was crystallized and further recrystallized from
hexane to give 0.83 g (35%) of colorless crystals, mp 122-123 °C. 1H
NMR (CDCl3) δ 7.30-7.13 (m, 5H), 7.09 (d, J ) 9.0 Hz, 2H), 6.79
(d, J ) 9.0 Hz, 2H), 5.94 (br s, 1H), 3.76 (s, 3H), 3.16 (s, 2H), 3.06
(s, 2H), 1.05 (s, 9H); UV (benzene), nm: 334 (ꢀ ) 48), 286 (ꢀ )
1000); HRMS m/e 334.1927 (calcd for C23H26O2 334.1933).
3-(4-Methoxyphenyl)-3-phenylcyclohexanone 12b. To a stirred at
0 °C suspension of 11.92 g (62.4 mmol) of CuI in 150.0 mL of ether
a solution of PhLi (145.1 mL of 0.86 M solution in ether-hexanes,
124.8 mmol) was added via a dropping funnel, and the resulting mixture
was stirred for 20 min at this temperature. A solution of 9.0 g (44.6
mmol) of 3-(4-methoxyphenyl)cyclohex-2-enone (11b) in 150.0 mL
of ether was added dropwise, followed by stirring for 3 h at 0 °C. The
reaction mixture was quenched with aqueous ammonia and concentrated
NH4Cl solutions, and the resulting mixture was vigorously stirred for
1.5 h at room temperature. The organic layer was separated, the aqueous
layer was extracted with ether, and the combined organic fraction was
washed with water and brine and dried over Na2SO4. After concentrating
in vacuo, the oily residue was separated by chromatography on silica
gel (column 48.0 cm × 4.0 cm, eluent ethyl acetate-hexane 1:3) to
give the following fractions:
(a) 4.0 g of diphenyl as a yellow solid, Rf 0.83;
(b) 1.2 g of unidentified product as a yellowish solid, Rf 0.74;
(c) 7.4 g (59%, or 76% based on consumed starting material) of
1
12b as a yellowish oil, Rf 0.43. H NMR (CDCl3) δ 7.31-7.16 (m,
5H), 7.11 (d, J ) 9.0 Hz, 2H), 6.81 (d, J ) 9.0 Hz, 2H), 3.77 (s, 3H),
2.93 (dd, J1 ) 18.0, J2 ) 15.6 Hz, 2H), 2.58-2.50 (m, 2H), 2.34 (t, J
) 7.2 Hz, 2H), 1.74-1.63 (m, 2H); HRMS m/e 280.1467 (calcd for
C19H20O2 280.1463);
(d) 1.9 g of the starting enone 11b as a yellowish solid, Rf 0.16.
1-(Trimethylsilyloxy)-5-(4-methoxyphenyl)-5-phenylcyclohex-1-
ene 13b. The reaction was performed according to a procedure similar
to the one for compound 13a. Reaction of 5.99 g (59.3 mmol) of
diisopropylamine in 30.0 mL of THF, 39.5 mL of 1.20 M solution of
n-BuLi in hexanes (47.4 mmol), 25.83 g (237.0 mmol) of chlorotri-
methylsilane in 50.0 mL of THF, and 6.64 g (23.7 mmol) of 3-(4-
methoxyphenyl)-3-phenylcyclohexanone (12b) in 50.0 mL of THF gave
a yellow oil, which was dissolved in pentane (80.0 mL), and insoluble
crystalline material was removed by filtration. The filtrate was
concentrated in vacuo to afford 7.67 g (92%) of 13b as a yellowish
oil. 1H NMR (CDCl3) δ 7.30-7.10 (m, 5H), 7.09 (d, J ) 9.0 Hz, 2H),
6.78 (d, J ) 9.0 Hz, 2H), 4.83-4.77 (m, 1H), 3.77 (s, 3H), 2.56 (s,
2H), 2.22 (t, J ) 6.0 Hz, 2H), 1.80-1.71 (m, 2H), 0.22 (s, 9H).
5-(4-Methoxyphenyl)-5-phenylcyclohex-2-enone 14b. The com-
pound was synthesized by a procedure similar to the one for compound
14a. Two separate runs with each run including 3.43 g (9.7 mmol) of
1-(trimethylsilyloxy)-5-(4-methoxyphenyl)-5-phenylcyclohex-1-ene (13b)
and 2.81 g (20.4 mmol) of K2CO3 in 250.0 mL of ether, and 1.63 g
(10.2 mmol) of bromine in 55.0 mL of CCl4 afforded 7.05 g of the
bromide as an orange oil. This bromide was brought into reaction with
5.12 g (58.8 mmol) of LiBr and 2.90 g (39.2 mmol) of Li2CO3 in 90.0
mL of DMF, and the crude product was purified by chromatography
on silica gel (column 43.0 cm × 4.0 cm, eluent ethyl acetate-hexane
1:3). A fraction with Rf 0.31 was collected and yielded, after
concentration in vacuo, 3.1 g (57%) of 14b as an orange oil. 1H NMR
(CDCl3) δ 7.31-7.13 (m, 5H), 7.09 (d, J ) 8.7 Hz, 2H), 6.99 (dt, J1
) 10.0, J2 ) 4.1 Hz, 1H), 6.80 (d, J ) 8.7 Hz, 2H), 6.04 (dt, J1 )
10.0, J2 ) 1.8 Hz, 1H), 3.76 (s, 3H), 3.19 (s, 2H), 3.16 (dd, J1 ) 4.1,
J2 ) 1.8 Hz, 2H); HRMS m/e 278.1304 (calcd for C19H18O2 278.1307).
3-tert-Butyl-5-(4-methoxyphenyl)-5-phenylcyclohex-2-enone 1b.
A solution of t-BuLi (16.9 mL of 1.7 M solution in pentane, 28.8 mmol)
was added to a stirred suspension of 1.29 g (14.4 mmol) of CuCN in
50.0 mL of ether at 0 °C, and the resulting mixture was stirred at this
temperature for 20 min, followed by the dropwise addition of a solution
of 2.00 g (7.2 mmol) of 5-(4-methoxyphenyl)-5-phenylcyclohex-2-
enone 14b in 50.0 mL of ether. After the addition was complete, the
Preparative Solution Photolysis of 3-tert-Butyl-5-(4-methoxy-
phenyl)-5-phenylcyclohex-2-enone 1b (Direct Irradiation). A solution
of 200 mg (0.6 mmol) of enone 1b in 80.0 mL of benzene was irradiated
in a Pyrex tube placed next to a water-cooled 400-W medium-pressure
mercury lamp equipped with a 2-mm Pyrex filter for 8 h. Nitrogen
was passed through the solution for 1 h before and during the photolysis.
After concentrating in vacuo, the oily residue was separated by column
chromatography on silica gel (column 50.0 cm × 2.5 cm, eluent ethyl
acetate-hexane 1:5) to give the following fractions:
(a) 30 mg (15%) of unidentified product as a yellow oil, Rf 0.58;
(b) 61 mg (31%) of 1-tert-butyl-exo-2-(4-methoxyphenyl)-5-
phenylbicyclo[3.1.0]hexan-3-one (5b) as a colorless solid, Rf 0.52. It
was recrystallized from ether-hexane mixture to give colorless prisms,
mp 171-174 °C. The structure of the product 5b was confirmed by a
single-crystal X-ray analysis. 1H NMR (CDCl3) δ 7.59-7.53 (m, 2H),
7.43-7.23 (m, 7H), 6.94 (d, J ) 8.1 Hz, 2H), 3.83 (s, 3H), 3.57 (s,
1H), 3.36 (ddd, J1 ) 18.3, J2 ) 3.3, J3 ) 1.1 Hz, 1H), 2.70 (d, J )
18.3 Hz, 1H), 1.85 (dd, J1 ) 5.9, J2 ) 3.3 Hz, 1H), 0.62 (s, 9H), 0.38
(d, J ) 5.9 Hz, 1H); HRMS m/e 334.1926 (calcd for C23H26O2
334.1933);
(c) 54 mg (27%) of 1-tert-butyl-endo-2-(4-methoxyphenyl)-5-
phenylbicyclo[3.1.0]hexan-3-one (6b) as a colorless solid, Rf 0.45. It
was recrystallized from ether-hexane mixture to give colorless crystals,
mp 136-137 °C. 1H NMR (CDCl3) δ 7.49-7.43 (m, 2H), 7.39-7.23
(m, 3H), 7.20 (d, J ) 8.7 Hz, 2H), 6.88 (d, J ) 8.7 Hz, 2H), 4.29 (s,
1H), 3.81 (s, 3H), 3.14 (dt, J1 ) 19.1, J2 ) 2.3 Hz, 1H), 2.72 (d, J )
19.1 Hz, 1H), 1.87 (dt, J1 ) 6.3, J2 ) 2.3 Hz, 1H), 1.05 (d, 6.3 Hz,
1H), 0.67 (s, 9H); HRMS m/e 334.1935 (calcd for C23H26O2 334.1933);
(d) 35 mg (18%) of the starting material as a yellowish oil,
crystallizing upon standing, Rf 0.36.
Preparative Solution Photolysis of 3-tert-Butyl-5-(4-methoxy-
phenyl)-5-phenylcyclohex-2-enone 1b (Sensitized Irradiation). A
solution of 100 mg (0.3 mmol) of enone 1b and 0.36 g (3.0 mmol) of
acetophenone in 40.0 mL of benzene was irradiated in a Pyrex tube
placed next to a water-cooled 400-W medium-pressure mercury lamp
equipped with 2-mm Pyrex filter for 4 h. Nitrogen was passed through
the solution for 1 h before and during the photolysis. After concentrating
in vacuo, acetophenone was removed by vacuum distillation (1 mmHg)
(20) Cieplak, A. S.; Tait, B. D.; Johnson, C. R. J. Am. Chem. Soc. 1989, 111,
8447-8462.
9
5428 J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003

Type C Enone Rearrangement
A R T I C L E S
1
at 80 °C, leaving 100 mg of oily residue. H NMR analysis of the
residue showed 70% conversion of the starting material with the ratio
of products 5b to 6b to be 1:1.2. No formation of any other products
was observed.
to a stirred at -70 °C solution of 0.74 g (7.4 mmol) of diisopropylamine
in 30.0 mL of THF, and the resulting solution was stirred for an
additional 20 min at this temperature. A solution of 1.30 g (4.7 mmol)
of 3-(4-cyanophenyl)-3-phenylcyclohexanone (12c) in 20.0 mL of THF
was added to the resulting LDA solution, followed by stirring for 1 h
at -70 °C. Then a solution of 1.18 g (7.4 mmol) of Br2 in 44.0 mL of
CCl4 was slowly added at this temperature, and the resulting mixture
was further stirred for 2 h. After quenching with 20.0 mL of water, the
reaction mixture was warmed to room temperature, extracted with ether,
washed successively with saturated aqueous NaHCO3, water, and brine,
and dried over Na2SO4. It was concentrated and dried in vacuo to afford
2.0 g of the crude bromoketone as an orange oil. This bromoketone
was dissolved in 30.0 mL of anhydrous DMF, and freshly vacuum-
dried LiBr (1.47 g, 17.0 mmol) and Li2CO3 (0.84 g, 11.3 mmol) were
added to this solution. The resulting mixture was stirred at 135 °C for
5 h. Then it was allowed to cool to room temperature, poured into 150
mL of water, extracted with two 100-mL portions of ether-pentane
(2:1), and the combined organic fractions were washed with water and
dried over Na2SO4. After concentration in vacuo, the oily residue was
purified by chromatography on silica gel (column 54.0 cm × 2.5 cm,
eluent ethyl acetate-hexane 1:2). A fraction with Rf 0.34 afforded 0.75
g (58%) of 14c as a colorless solid, which was further recrystallized
from a mixture of hexanes-ether-dichloromethane to give colorless
3-(4-Cyanophenyl)cyclohex-2-enone 11c. A solution of n-BuLi
(47.6 mL of 1.2 M solution in hexanes, 57.1 mmol) was added slowly
to a stirred at -70 °C solution of 10.0 g (54.9 mmol) of 4-bromoben-
zonitrile in 300.0 mL of THF (temperature of the reaction mixture
should remain below -65 °C during the addition). After the addition
was complete, the resulting mixture was stirred for 1 h at -70 °C,
followed by dropwise addition of a solution of 6.99 g (49.9 mmol) of
3-ethoxycyclohex-2-enone (10) in 60.0 mL of THF. The resulting
solution was further stirred for 2 h at this temperature and then carefully
quenched with 40.0 mL of 10% HCl. The reaction mixture was allowed
to warm gradually to room temperature, concentrated HCl (15.0 mL)
was added, and the resulting mixture was additionally stirred for 1.5
h. Then it was poured into 300.0 mL of water and extracted with two
300.0 mL portions of ether. The organic fraction was washed succes-
sively with water, saturated NaHCO3, water, and brine and dried over
Na2SO4. Concentration in vacuo afforded a mixture of products as a
dark oil. One of the byproducts, 3-butylcyclohex-2-enone, was removed
by distillation in vacuo (0.5 mmHg) at 110 °C, and the oily residue
was crystallized from methanol to give 3.2 g of crude 11c as yellow
crystals. A mother liquor left after the crystallization was separated by
chromatography on silica gel (column 47.0 cm × 4.0 cm, eluent ethyl
acetate-hexane 1:2), and the following fractions were collected:
(a) 1.1 g (9%) of 3-(4-pentanoyl-phenyl)cyclohex-2-enone 17 as a
yellowish solid, Rf 0.42. It was recrystallized from ether-hexane
mixture to give colorless crystals, mp 74-76 °C. 1H NMR (CDCl3) δ
7.99 (d, J ) 8.7 Hz, 2H), 7.61 (d, J ) 8.7 Hz, 2H), 6.46 (t, J ) 1.5
Hz, 1H), 2.98 (t, J ) 7.5 Hz, 2H), 2.79 (td, J1 ) 6.5, J2 ) 1.5 Hz, 2H),
2.51 (t, J ) 6.5 Hz, 2H), 2.25-2.13 (m, 2H), 1.79-1.66 (m, 2H), 1.48-
1.35 (m, 2H), 0.96 (t, J ) 6.9 Hz, 3H); 13C NMR (CDCl3) δ 199.76,
199.52, 158.31, 142.96, 137.75, 128.40 (2C), 126.66, 126.21 (2C),
38.40, 37.18, 27.99, 26.33, 22.69, 22.39, 13.89; HRMS (ESI) m/e
257.1548 ([M + H]+) (calcd for C17H21O2 257.1541);
1
crystals, mp 156-157 °C. H NMR (CDCl3) δ 7.57 (d, J ) 8.4 Hz,
2H), 7.34-7.19 (m, 5H), 7.16-7.11 (m, 2H), 7.01 (dt, J1 ) 10.2, J2 )
4.2 Hz, 1H), 6.07 (dt, J1 ) 10.2, J2 ) 1.8 Hz, 1H), 3.24-3.16 (m,
4H); HRMS m/e 273.1157 (calcd for C19H15NO 273.1154).
3-tert-Butyl-5-(4-cyanophenyl)-5-phenylcyclohex-2-enone 1c. The
compound 1c was synthesized by a procedure similar to the one for
compound 1b. Reaction of 0.60 g (2.2 mmol) of 5-(4-cyanophenyl)-
5-phenylcyclohex-2-enone (14c) in 20.0 mL of THF, 5.16 mL of 1.7
M solution of t-BuLi in pentane (9.5 mmol), 0.39 g (4.3 mmol) of
CuCN in 10.0 mL of ether, and a solution of PhSeBr prepared from
1.56 g (5.0 mmol) of diphenyl diselenide in 5.0 mL of ether and 0.78
g (4.9 mmol) of Br2 gave the intermediate selenoketone. This was
dissolved in 30.0 mL of methylene chloride and oxidized with 3.21 g
of 30% H2O2 (28.3 mmol) in the presence of 1.59 g (20.1 mmol) of
pyridine. The material after the reaction was purified by chromatography
(column 50.0 cm × 2.5 cm, eluent ethyl acetate-hexane 1:3), and a
fraction with Rf 0.42 afforded 0.28 g (39%) of 1c as a yellowish solid.
An additional recrystallization from ether-hexane mixture gave 1c as
a colorless crystals, mp 161-165 °C. 1H NMR (CDCl3) δ 7.56 (d, J )
9.0 Hz, 2H), 7.31-7.17 (m, 5H), 7.15-7.10 (m, 2H), 5.97 (s, 1H),
3.19 (s, 2H), 3.07 (s, 2H), 1.05 (s, 9H); UV (benzene), nm: 334 (ꢀ )
52), 282 (ꢀ ) 550); HRMS (ESI) m/e 352.1683 ([M + Na]+) (calcd
for C23H23NONa 352.1677).
Preparative Solution Photolysis of 3-tert-Butyl-5-(4-cyanophenyl)-
5-phenylcyclohex-2-enone 1c (Direct Irradiation). A solution of 100
mg (0.3 mmol) of enone 1c in 30.0 mL of benzene was irradiated in a
Pyrex tube placed next to a water-cooled medium-pressure mercury
lamp equipped with 2-mm Pyrex filter for 3.5 h. Nitrogen was passed
through the solution for 1 h before and during the photolysis. After
concentrating in vacuo, the oily residue was separated by column
chromatography on silica gel (column 39.0 cm × 1.5 cm, eluent ethyl
acetate-hexane 1:3) to give the following fractions:
(b) 2.1 g of 11c as yellowish solid, Rf 0.28.
The combined product material was recrystallized from methyl tert-
butyl ether to give 4.5 g (46%) of 11c as colorless crystals, mp 90-91
°C (lit.21 mp 79-81 °C).
3-(4-Cyanophenyl)-3-phenylcyclohexanone 12c. To a stirred at 0
°C suspension of 2.66 g (13.9 mmol) of CuI in 80.0 mL of ether was
added dropwise a solution of PhLi (30.6 mL of 0.91 M solution in
ether, 27.9 mmol), and the resulting mixture was stirred for 1 h at 0
°C. Then the reaction mixture was allowed to warm to room
temperature, and a solution of 1.83 g (9.3 mmol) of 3-(4-cyanophenyl)-
cyclohex-2-enone (11c) in a mixture of 7.0 mL of THF and 7.0 mL of
ether was slowly added, followed by the stirring at room temperature
for additional 3 h. Concentrated solutions of NH4Cl (50 mL) and
ammonia (50 mL) were added, and the reaction mixture was transferred
into a separation funnel and shaken vigorously until all inorganic
material dissolved in the aqueous layer and the organic layer became
clear and transparent. The organic fraction was washed with water and
brine and dried over Na2SO4. After concentrating in vacuo, the oily
residue was separated by chromatography on silica gel (column 43.0
cm × 4.0 cm, eluent ethyl acetate-hexane 1:2), and a fraction with Rf
0.44 was collected which afforded 1.4 g (55%) of 12c as a very viscous,
(a) 13 mg (13%) of unidentified product as a yellowish oil, Rf 0.75;
(b) 18 mg (18%) of 1-tert-butyl-exo-2-(4-cyanophenyl)-5-phenyl-
bicyclo[3.1.0]hexan-3-one (5c) as a colorless oil, Rf 0.64. 1H NMR
(CDCl3) δ 7.59-7.10 (m, 9H), 3.66 (s, 1H), 3.36 (dm, J ) 17.9 Hz,
1H), 2.74 (d, J ) 17.9 Hz, 1H), 1.91 (dd, J1 ) 6.4, J2 ) 2.1 Hz, 1H),
0.61 (s, 9H), 0.43 (d, J ) 6.4 Hz, 1H); HRMS m/e 329.1776 (calcd for
C23H23NO 329.1780);
(c) 22 mg (22%) of 1-tert-butyl-endo-2-(4-cyanophenyl)-5-phen-
ylbicyclo[3.1.0]hexan-3-one (6c) as a colorless solid, Rf 0.50. It was
recrystallized from ether-hexane mixture to give colorless prisms, mp
165-168 °C. The structure of the product 6c was confirmed by a single-
1
yellow oil. H NMR (CDCl3) δ 7.58 (d, J ) 8.7 Hz, 2H), 7.35-7.13
(m, 7H), 2.99 (d, J ) 15.0 Hz, 1H), 2.90 (d, J ) 15.0 Hz, 1H), 2.65-
2.53 (m, 2H), 2.38 (t, J ) 6.9 Hz, 2H), 1.75-1.63 (m, 2H); HRMS
(ESI) m/e 298.1204 ([M + Na]+) (calcd for C19H17NONa 298.1208).
5-(4-Cyanophenyl)-5-phenylcyclohex-2-enone 14c. A solution of
n-BuLi (5.3 mL of 1.12 M solution in hexanes, 5.9 mmol) was added
(21) Klement, I.; Rottlaender, M.; Tucker, C. E.; Majid, T. N.; Knochel, P.;
Venegas, P.; Cahiez, G. Tetrahedron 1996, 52, 7201-7220.
9
J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003 5429

A R T I C L E S
Zimmerman and Nesterov
1
crystal X-ray analysis. H NMR (CDCl3) δ 7.61-7.11 (m, 9H), 4.41
(s, 1H), 3.24 (dt, J1 ) 17.5, J2 ) 2.2 Hz, 1H), 2.77 (d, J ) 17.5 Hz,
1H), 1.99 (dt, J1 ) 6.6, J2 ) 2.2 Hz, 1H), 1.15 (d, J ) 6.6 Hz, 1H),
0.68 (s, 9H); HRMS m/e 329.1787 (calcd for C23H23NO 329.1780);
a checkpoint (*.chk) file using a command “AONBO)c” in the NBO
command line. Then the NBO output was analyzed, and the orbitals
of interest were chosen. The final CASSCF run was performed starting
from the checkpoint file with these NBOs having been placed into the
active space (using a command “guess)(check,alter)” in the command
line). The size of the active space was varied, depending on the
particular species to analyze. Thus, the computations of cyanophenyl
enone 1c half-migrated species 3T-1 and 3T-2 utilized active space
(10,10). It included the following orbitals: two (π and π*) orbitals of
CdO group, five (two π, two π* and a p-type lone pair) orbitals on a
migrating aryl group, p-type lone pair on carbon C3, and two conjugated
with aromatic π system π and π* orbitals of a cyano group (for the
intermediate 3T-2 with phenyl group migrated, these two orbitals were
included from the cyano group of a nonmigrating aryl to keep
consistency of the active space). Similar to that, computations of the
methoxyphenyl enone 1b case involved (10,9) active space. Here one
p-type lone pair orbital of methoxy group was used instead of two
orbitals of cyano group in the enone 1c case. The computations of the
cyanophenyl enone intermediates 3T-1 and 3T-2 with extended
(14,14) active space involved all the same orbitals as mentioned above
for the (10,10) active space computation except for the two (π and
π*) orbitals of CdO group which were not included (total eight
orbitals). In addition to them, six π orbitals (three π and three π*) of
the nonmigrating aromatics were added, thus bringing the active space
to (14,14). An example of the CASSCF/NBO input and output files is
given in Supporting Information.
(d) 38 mg (38%) of the starting material 1c as a colorless solid, Rf
0.36.
Preparative Solution Photolysis of 3-tert-Butyl-5-(4-cyanophenyl)-
5-phenylcyclohex-2-enone 1c (Sensitized Irradiation). A solution of
50 mg (0.15 mmol) of enone 1c and 0.18 g (1.5 mmol) of acetophenone
in 15.0 mL of benzene was irradiated in a Pyrex tube placed next to a
water-cooled medium-pressure mercury lamp equipped with 2-mm
Pyrex filter for 1.5 h. Nitrogen was passed through the solution for 1
h before and during the photolysis. After concentrating in vacuo,
acetophenone was removed by vacuum distillation (1 mmHg) at 80
1
°C, leaving 50 mg of oily residue. H NMR analysis of the residue
showed conversion of the starting material 60% with the ratio of
products 5c to 6c to be 1:1.4. No formation of any other products was
observed.
Computational Procedures. Ab initio and DFT computations were
performed on Intel Pentium III-based multiprocessor cluster with Linux
OS using the Gaussian 9822 computational package. Geometry opti-
mization of intermediates 2T and 3T was performed with UB3LYP
method. Geometry of the transition states between 2T and 3T was
obtained by QST3 computations followed by IRC analysis. The
CASSCF/NBO computations were done by the following way. A
preliminary CASSCF(4,4) single-point computation was accompanied
by NBO analysis,8 with the resulting localized NBOs been written into
Acknowledgment. Support of this research by the National
Science Foundation is gratefully acknowledged with special
appreciation for its support of basic research.
(22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, Rev. A.9; Gaussian, Inc.:
Pittsburgh, PA, 1998.
Supporting Information Available: Examples of Gaussian
98 input and output files for CASSCF/NBO treatment; X-ray
coordinates of compounds 5b and 6c (PDF). X-ray crystal-
lographic files in CIF format. This material is available free of
charge via the Internet at http://pubs.acs.org.
JA028631B
9
5430 J. AM. CHEM. SOC. VOL. 125, NO. 18, 2003