Regioselectivity of the tri-π-methane rearrangement: Mechanistic and exploratory organic photochemistry

Article abstract of DOI:10.1021/jo034189j

The di-π-methane rearrangement with two π-groups attached to the central "methane carbon" of the reactant and which leads to a π-substituted cyclopropane has been studied intensively. Our present research had the goal of elucidating the regioselectivity of the tri-π-methane counterpart. The reactants with three π groups attached to the central carbon mechanistically are capable of affording both di-π-methane and tri-π-methane photoproducts. In common with the di-π-methane system, bridging of two of the π-systems affords a cyclopropyldicarbinyl diradical intermediate that opens to an allyl carbinyl diradical. This diradical has the option of closing to a three-membered or a five-membered ring. It was found that the regioselectivity of the initial π-π-bridging step and the three-ring opening of the cyclopropyldicarbinyl diradical exhibit regioselectivity parallel to that of the di-π counterpart. Both three-ring and five-ring photoproducts were formed with the ratio varying with conversion. Since the three-ring (i.e. di-π-methane) photoproducts were found to ring expand to the five-ring (i.e. tri-π-methane) products, kinetics were employed to determine to what extent the reaction proceeds in a two-step versus direct formation of the five-ring product. It was found that the direct route was the major one.

Full text of DOI:10.1021/jo034189j

Regioselectivity of th e Tr i-π-m eth a n e Rea r r a n gem en t:
Mech a n istic a n d Exp lor a tor y Or ga n ic P h otoch em istr y1,2
Howard E. Zimmerman* and Tibor Novak
Chemistry Department, University of Wisconsin, Madison, Wisconsin 53706
zimmerman@chem.wisc.edu
Received February 11, 2003
The di-π-methane rearrangement with two π-groups attached to the central “methane carbon” of
the reactant and which leads to a π-substituted cyclopropane has been studied intensively. Our
present research had the goal of elucidating the regioselectivity of the tri-π-methane counterpart.
The reactants with three π groups attached to the central carbon mechanistically are capable of
affording both di-π-methane and tri-π-methane photoproducts. In common with the di-π-methane
system, bridging of two of the π-systems affords a cyclopropyldicarbinyl diradical intermediate that
opens to an allyl carbinyl diradical. This diradical has the option of closing to a three-membered or
a five-membered ring. It was found that the regioselectivity of the initial π-π-bridging step and
the three-ring opening of the cyclopropyldicarbinyl diradical exhibit regioselectivity parallel to that
of the di-π counterpart. Both three-ring and five-ring photoproducts were formed with the ratio
varying with conversion. Since the three-ring (i.e. di-π-methane) photoproducts were found to ring
expand to the five-ring (i.e. tri-π-methane) products, kinetics were employed to determine to what
extent the reaction proceeds in a two-step versus direct formation of the five-ring product. It was
found that the direct route was the major one.
SCHEME 1. Th e Di-π-m eth a n e a n d Tr i-π-m eth a n e
Rea r r a n gem en ts
In tr od u ction
Recently we reported the tri-π-methane rearrange-
ment,3,4 which is related to the well-known di-π-methane
counterpart.5 As with the di-π-methane reaction, the
mechanism (note Scheme 1) begins with π-π bridging
to afford a cyclopropyldicarbinyl diradical. However, on
three-ring opening, the diradical produced is allylic and
may close to either a three-membered or five-membered
ring.
In the case of the di-π-methane rearrangement,5 in-
vestigations of the effect of para substituents on regiose-
lectivity,6 as well as in other unsymmetrical systems,5c
afforded insight into mechanistic details. Thus it was of
special interest in the present study to investigate the
tri-π-methane rearrangement regioselectivity.
Resu lts
Syn th etic Asp ects. In our previous study3 a very
simple approach to the trivinylmethane structure was
developed by using the Prins reaction as shown in eq 1.
(1) This is paper 271 of our general series.
(2) For paper 270 see: Zimmerman, H. E.; Nesterov, E. E. J . Am.
Chem. Soc. 2003, 125, 5422-5430.
(3) Zimmerman, H. E.; Cirkva V. J . Org. Chem. 2001, 66, 1839-
1851.
(4) One early example of the tri-π-methane rearrangement occurred
only in a solid-state photolysis; note: Zimmerman, H. E.; Zuraw, M.
J . J . Am. Chem. Soc. 1989, 111, 7974-7989.
(5) (a) Zimmerman, H. E.; Mariano, P. S. J . Am. Chem. Soc. 1969,
91, 1718-1727. (b) Zimmerman, H. E. Mol. Photochem. 1971, 3, 281-
292. (c) Zimmerman, H. E.; Pratt, A. C. J . Am. Chem. Soc. 1970, 92,
6259-6267.
(6) (a) Zimmerman, H. E.; Gruenbaum, W. T. J . Org. Chem. 1978,
43, 1997-2005. (b) Zimmerman, H. E.; Welter, T. R. J . Am. Chem.
Soc. 1978, 100, 4131-4145.
However, with para substitution in the diphenylacrolein
reactant (i.e. 8) anticipated to afford substitution in one
of the three vinyl moieties, an unanticipated but rather
interesting reaction occurred. In each case the unsubsti-
10.1021/jo034189j CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/23/2003
5056
J . Org. Chem. 2003, 68, 5056-5066

Tri-π-methane Rearrangement
SCHEME 2. Syn th esis of th e Dicya n op h en yl Tr ien e
SCHEME 3. Syn th esis of th e Isop r op ylid en e
Tr i-π-m eth a n e System s
tuted hexaphenyl product 9 resulted. Note eq 2. The
reaction results from the reversibility of the Prins reac-
tion and also the low reactivity of di-p-chlorophenyleth-
ylene;7 however, consideration of the mechanism is
delayed to the Discussion section (vide infra).
Thus different approaches were needed. These are as
outlined in Schemes 2 and 3. The synthesis in Scheme
2, beginning with the known14 tetraphenyldienone 11, is
basically a modified Prins sequence wherein the dienol
12 employed is the first intermediate of a more typical
Prins reaction. This synthesis afforded a 53% yield of tri-
π-methane 13. Also, this result confirms that the problem
in the original approach (note eq 2) was with formation
of diphenylacrolein 8 as a consequence of the Prins re-
versibility, and in the present synthesis 8 is not formed.
Scheme 3 outlines the approach employed where one
isopropylidene moiety was desired. This approach, start-
ing with the known3 carbethoxydiene 16, was employed
since the Prins approach was unsuccessful with di-
methylacrolein and diphenylethylene as reactants.
Another interesting facet was the observation that the
trienyl carbanions derived from trienes 9 and 14 were
uniquely deep blue. The alkylation of these anions went
regioselectively at the center carbon. Similarly, protona-
tion of the delocalized species occurred at the central car-
bon. This regioselectivity is considered in the Discussion
section.
The structures of the tri-π-methane species (15, 20a ,
and 20b) were confirmed by NMR, mass spectral analy-
sis, and X-ray in the case of the di-p-cyano-triene 15.
Exp lor a tor y P h otoch em istr y. (a ) Rea ctivity of
th e Isop r op ylid en e Tr i-π-m eth a n e Rea cta n ts. Equa-
tion 3 outlines the course of the direct and sensitized
photolyses of the isopropylidene tri-π-methane reactants.
Both on direct and acetophenone-sensitized irradiation
photoproducts 21, 22, and 23 were observed. In addition,
photoproduct 24b was encountered in sensitized runs of
(7) (a) Zimmerman, H. E.; English, J ., J r. J . Am. Chem. Soc. 1954,
76, 2285-2290. (b) Zimmerman, H. E.; English, J ., J r. J . Am. Chem.
Soc. 1954, 76, 2291-2294. (c) Zimmerman, H. E.; English, J ., J r. J .
Am. Chem. Soc. 1954, 76, 2294-2300.
(8) Quantum Mechanics for Organic Chemists; Zimmerman, H. E.,
Ed.; Academic Press, New York, 1975; pp 153-155.
(9) Zimmerman, H. E.; Werthemann, D. P.; Kamm, K. S. J . Am.
Chem. Soc. 1973, 95, 5094-5095.
(10) (a) Zimmerman, H. E.; Pratt, A. C. J . Am. Chem. Soc. 1970,
92, 6267-6271. (b) Hixson, S. S.; Mariano, P. S.; Zimmerman, H. E.
Chem. Rev. 1973, 73, 531-551. (c) Zimmerman, H. E.; Schissel, D. N.
J . Org. Chem. 1986, 51, 196-207.
(11) (a) Zimmerman, H. E.; Armesto, D.; Amezua, M. G.; Gannett,
T. P.; J ohnson, R. P. J . Am. Chem. Soc. 1979, 101, 6367-6383. (b)
Zimmerman, H. E.; Factor, R. E. Tetrahedron, 1981, 37, Suppl. 1, 125-
141. (c) Zimmerman, H. E.; Penn, J . H.; J ohnson, M. R. Proc. Natl.
Acad. Sci. U.S.A. 1981, 78, 2021-2025.
(12) (a) Sheldrick, G. M. SHELXTL, Version 5.1; Bruker AXS, Inc.:
Madison, WI, 1997. (b) Obtained from Prof. G. Sheldrick, Go¨ttingen.
(13) Corey, E. J .; Suggs, J . W. Terahedront. Lett. 1975, 31, 2647-
2650.
(14) Marin, G.; Chodkiewicz, W.; Cadiot, P.; Willemart, A. Bull. Soc.
Chim. Fr. 1958, 1594-1597.
J . Org. Chem, Vol. 68, No. 13, 2003 5057

Zimmerman and Novak
of allylic coupling. The structures of all products were
similar to those reported earlier.3 The structures of 21a
and 25 were established by X-ray and thus confirmed the
validity of the NMR spectral assignment methodology.
(b) Kin etic Asp ects. In our previous study3 of the tri-
π-methane rearrangement we encountered two routes to
the five-membered-ring products: a direct route and an
indirect route proceeding via the three-ring photoproduct
as outlined in Scheme 4 for the present cases. Using the
kinetic approach of our previous study, the partition be-
tween these routes was determined. For this it was
necessary to obtain the product distributions as a func-
tion of extent of conversion. The detailed data are given
in the Supporting Information but examples are plotted
in Scheme 5.
Direct irradiations were used for the kinetic runs. In
addition, runs were made with a naphthalene sensitizer
under concentration conditions where naphthalene ab-
sorbs minimally 99% of the light and with reactant
concentrations high enough to permit singlet energy
transfer. The purpose of the naphthalene was to have
constant light absorption as the reaction proceeded and
the distribution of compounds, with slightly different
absorbtivities, varied.
In Scheme 4a,b, the reactants are labeled “A”, the di-
π-methane products are labeled “B”, and the tri-π-meth-
ane products are labeled “C”. We note here that the cis
and trans stereoisomers of the di-π-methane products are
taken as one kinetic entity, since they are rapidly equili-
brated. The analytic expressions for an A-B-C sequence
in which “A” may also proceed directly to “C” had not
been solved, but were derived in our earlier publication.3
Equation 5 gives the relative concentration of species “B”
as a function of time. Then eq 6a gives the deviation of
B from experiment at a single time corresponding to
variations of k1, k2, and k3. For measurements at a series
of times, we obtain eqs 6b, giving ∆B1, ∆B2, ∆B3, etc., as
deviations from the theoretical B as given in eq 5 and at
times t1, t2, t3, etc. The partial derivatives in 6 are
obtained from derivatization of eq 5 and thus are known.
Hence we have a large series of simultaneous eqs 6b with
deviations of B as a function of variations in the three
rate constants. It is necessary to adjust the ∆k values to
minimize the ∆Bt values. The problem is overdetermined
and one can solve for the deviations of the rate constants.
20b. Table 1 summarizes the product distributions from
both the benzyl reactant 20a and the isopropyl reactant
20b in high conversion runs. The product distribution
varied with the extent of conversion; and product distri-
butions as a function of conversion are given below in
connection with the reaction kinetics. One other point is
that photoproduct 23 was a very minor product in the
acetophenone-sensitized runs while 24b was encountered
only in the sensitized photolysis of 20b.
(i) Rea ctivity of th e Tr is-d ia r ylvin yl Tr i-π-m eth -
a n e Rea cta n ts. In the case of the cyano-substituted tri-
ene 15, the direct and sensitized photochemistry, again,
proceeded to afford both di-π-methane and tri-π-methane
products. As with the isopropylidene reactions, acetophe-
none sensitization did not lead to the tri-π-methane
product. These runs are described in eq 4. Table 2 sum-
marizes typical photochemical product distributions.
B ) (k1A0)[e-(k +k )t - e-k t]/(k3 - k1 - k2)
(5)
1
2
3
∆Bt ) (∂Bt/∂k1)∆k1 + (∂Bt/∂k2)∆k2 + (∂Bt/∂k3)∆k3
(6)
(ii) Assa y of P h otop r od u cts for Kin etic a n d Id en -
tifica tion P u r p oses. In the reactions described above,
the photoproducts were separated by column chroma-
tography and crystallization as described in the Experi-
mental Section. The mass balances were generally close
to 80%. However, for kinetic purposes, NMR analysis
proved most helpful. Thus, the di-π-methane and tri-π-
methane photoproducts were easily identified as a con-
sequence of their characteristic NMR spectra. For ex-
ample, the AB quartet of the di-π-methane products (21,
22, and 25, 26) derived from the vicinal cyclopropyl
methine and diphenylvinyl hydrogen. Cyclopentenes (23
and 27) have a very typical broad doublet, a broad triplet,
and a sharp doublet coupling pattern as a consequence
These equations can be formulated more neatly in
matrix form as in eqs 7a and 7b. Then, solving for the
unknown deviations of the rate constants needed to
minimize error in the calculated B versus time, we obtain
eq 8.
(∂B1/∂k1) (∂B1/∂k2) (∂B1/∂k3)
(∂B2/∂k1) (∂B2/∂k2) (∂B2/∂k3)
(∂B3/∂k1) (∂B3/∂k2) (∂B3/∂k3)
∆B1
∆B2
∆B3
l
∆k1
∆k2
)
(7a)
[
]
∆k
[
]
[
]
3
l
l
l
∆B ) F ∆k
(7b)
(8)
∆k ) (F tF )-1F t∆B
5058 J . Org. Chem., Vol. 68, No. 13, 2003

Tri-π-methane Rearrangement
TABLE 1. P h otop r od u cts Obta in ed on Dir ect a n d Sen sitized Ir r a d ia tion of 20a a n d 20b
cis-di-π-
photoproduct
21
trans-di-π-
photoproduct
22
tri-π-
photoproduct
23
secondary
photoproduct
24b
irrad. time
(min)
conv
(%)
reactant
additive
none
none
none
none
acetophenone
acetophenone
20a
20a
20b
20b
20a
20b
12
50
9
35
18
8
66
100
75
100
100
100
24
35
21
20
67
45
19
23
20
17
28
10
23
42
34
63
5
0
0
0
0
0
trace
45
TABLE 2. P h otop r od u cts Obta in ed on Dir ect a n d Sen sitized Ir r a d ia tion of 15
cis-di-π-
trans-di-π-
tri-π-
photoproduct
27
irrad. time
(min)
conv.
(%)
photoproduct
photoproduct
reactant
additive
none
none
acetophenone
25
26
15
15
15
25
50
20
82
100
100
32
34
53
32
27
47
18
39
0
SCHEME 4. Alter n a tive Rea ction P a th w a ys Lea d in g to Tr i-π-m eth a n e P r od u cts
SCHEME 5. Kin etics of Dir ect Solu tion P h otolysis of Tr i-π-m eth a n e Rea cta n ts (15 a n d 20b)a
a
Diamonds are for the tri-π-methane reactant, squares are for di-π-methane, and triangles are for the tri-π-methane photoproducts:
left, 15; right, 20b.
Further details of the derivation are given in our
previous publication3 and there is a description of our
program “ABC_kinetics”, which obtains the rate con-
stants iteratively. Thus one starts with a set of assumed
rate constants and the vector of eq 7 listing the differ-
ences in experimental and theoretical concentrations of
B as a function of time. Equation 8 solves for the errors
in the three rate constants corresponding to these errors.
Then the program “ABC_kinetics” in each step calculates
the change in the correction in each rate constant to
diminish the error in the concentration of B. Table 3 gives
the rate constants obtained in this way.
In ter p r eta tive Discu ssion
Gen er a l. Having dealt with the experimental observa-
tions, we now need to interpret these results. Thus, one
aspect of the synthesis is of particular interest, namely
the regioselectivity of electrophilic attack on a delocalized
trivinyl methane system. However, most of our efforts
have been photochemical and are to be interpreted.
Hence, we note that in the direct irradiations the tri-
π-methane and di-π-methane rearrangements occur com-
petitively and that the ratio of the two types of photo-
products is dependent on the extent of conversion. At
higher conversions, we find an increasing amount of five-
J . Org. Chem, Vol. 68, No. 13, 2003 5059

Zimmerman and Novak
TABLE 3. Rep r esen ta tive Rela tive Ra te Con sta n ts
obsd relative rate constants
tri-π-methane reactant
sensitizer
k1
k2
k3
3-benzyl tri-π-methane (20a )
none
naphthalene
none
naphthalene
none
naphthalene
0.02396
0.02658
0.03109
0.04271
0.04850
0.04677
0.01117
0.00365
0.01851
0.01399
0.01236
0.01197
0.00149
0.00160
0.00609
0.00592
0.00081
0.00096
3-isopropyl tri-π-methane (20b)
3-benzyl dicyano-π-methane (15)
“Mariano compound” 35 is 1.4 × 1011 while that for S1 of
the ”Pratt compound” 37 is 6.9 × 109; note Scheme 7,
thus illustrating the effect of extra delocalization.
In the case of the triplet of 20 with R ) i-Pr it appears
that steric congestion inhibits bonding of the two diphen-
ylvinyl moieties to the extent that Path B becomes
accessible. Another point is that the triplet of 20 is
reactive only where the central substituent is isopropyl
but not benzyl. In the simpler di-π-methane systems the
triplets of acyclic dienes generally are either nonreactive
or minimally so. One exception is those with bulky
substituents, as isopropyl, on the central (i.e. “methane”)
carbon.10c The lack of reactivity has been attributed to a
“free-rotor effect”,9,10 wherein triplet excited π-bonds tend
to dissipate energy by rotation with concomitant inter-
system crossing to the ground state. However, in the case
of the tri-π-methane system presently studied with
isopropyl central substitution (i.e. 30), the bulky nature
of the groups bonded to the “methane carbon” inhibits
free-rotor energy dissipation. With the benzyl group at
the central carbon, the triplet is unreactive. This is not
unexpected, since the bulky phenyl part of this group is
more remote.
F IGURE 1. Bond orders of type 1,2 are larger than those of
type 2,3.
membered-ring formation. Thus there are two modes of
formation of the vinylcyclopentenes: a direct formation
and a secondary reaction of the vinylcyclopropanes. The
sensitized reactions lead to little or none of the tri-π-
methane products.
Syn t h et ic: R egioselect ivit y of Alk yla t ion a n d
P r oton a tion . It was noted in the Results section that
both alkylation and protonation of the trivinylmethyl
carbanions proceeded regioselectivity at the central
carbon. This phenomenon has been encountered previ-
ously in the case of pentadienyl-type systems where,
again, protonation occurs at the central carbon. The effect
results from the highest electron density being at the
central carbon. The same effect is now seen in the
trivinylmethyl carbanions. Interestingly, simple Hu¨ckel
computations of these odd-alternant systems predict
equal density at the central and allylic carbons. However,
as has been noted,8 even Hu¨ckel theory is not na¨ıve
regarding the bond orders of such systems. The bond
orders of the bonds allylic to the central carbon (type 1,2
in Figure 1) are larger than those of type 2,3.
With the pentadienyl 1,2-bond orders in these systems
compared with the 2,3-bond orders, the corresponding
1,2-resonance integrals (and the Fock Matrix elements
in SCF theory) become larger than the 2,3-counterparts.
This then leads to concentration of the electron density
at the central carbon.
R egioselect ivit y of t h e Isop r op ylid en e Tr i-π-
Meth a n e Rea ction . Several aspects are of special
interest in the reactions of the isopropylidene reactants
20a and 20b. The first is the partition between Paths A
and B as outlined in Scheme 6. In the direct irradiation
there is a preference for bridging between the two
diphenylvinyl groups to afford diradical 28 compared
with bridging between one diphenylvinyl and one iso-
propylidene to give diradical 30. This is readily under-
stood on consideration of the greater odd-electron delo-
calization of diradical 28 in Path A compared to diradical
30 in path B; the preferred diradical 28 has benzhydryl
delocalization for both odd-electron centers while species
30 has only one benzhydryl group.
The mechanism of formation of vinylcyclopentene 24b
is relevant to the triplet reactivity. This product appears
to be formed via the di-π-methane precursor 34, which
is seen in the NMR spectrum of the reaction mixture at
lower conversion but is not formed in sufficient quantities
to permit isolation (note the Experimental Section).
Regioselectivity of th e Bis-p-Cya n op h en ylvin yl
Tr i-π-m eth a n e Rea cta n t. Turning next to consideration
of the reactivity of the cyanophenyl tri-π-methane reac-
tant, we find in Scheme 8 that the regioselectivity, again,
is definitive with preferential bridging to give that
diradical which has the greater odd-electron delocaliza-
tion (i.e. 39 rather than 41). Both the di-π-methane (25
and 26) and tri-π-methane (27) photoproducts are formed
with the former being preferred. Again, the amount of
five-membered-ring product increased with the extent of
conversion suggesting a contribution from a secondary
reaction of the initial di-π-methane product. This point
is considered further in connection with discussion of the
reaction kinetics. As in the case of the isopropylidene
reactant, the triplet led only to the di-π-methane photo-
products.
Di-π-m eth a n e Ver su s Tr i-π-m eth a n e P h otop r od -
u ct F or m a tion . We have seen that the tri-π-methane
photoproducts are formed only in the direct (i.e. S1)
irradiations. This is understood when one considers that
triplet diradicals prefer to have their odd-electron centers
as remote as possible. The effect relates to our “large K
- small K” exchange integral postulate,11 which sug-
gested that in general singlets prefer “small K” transition
Additionally, the S1 rates for di-π-methane reactants
355a and 375c (Scheme 7) are known.9 That for S1 of the
5060 J . Org. Chem., Vol. 68, No. 13, 2003

Tri-π-methane Rearrangement
SCHEME 6. Mech a n ism of th e Tr i-π-m eth a n e Rea r r a n gem en t in Com p etition w ith th e Di-π-m eth a n e
P r ocess in th e Ca se of 20
SCHEME 7. Com p a r ison of Tw o Sin glets P r ocesses
SCHEME 8. Mech a n ism of th e Tr i-π-m eth a n e Rea r r a n gem en t in Com p etition w ith th e Di-π-m eth a n e
P r ocess in th e Ca se of 15
structures while triplets prefer “large K” reaction routes.
Note Figure 2. The quantum mechanical exchange inte-
gral, 2K, increases with odd-electron separation. Thus,
the transoid open diradicals 29 and 40 have larger K
values than the cisoid counterparts and are preferentially
formed from triplet precursors. However, these transoid
allylic systems cannot close to five-membered-ring prod-
ucts. Conversely, the cisoid allylic diradicals 29 and 40,
with lesser odd-electron separations, are preferentially
formed from the singlet (i.e. S1) precursors. These can
close to the five-membered-ring products.
Discu ssion of th e Kin etic Resu lts
Experimentally, the three-membered-ring photoprod-
ucts (21, 22, 25, and 26) dominate, and reference to Table
3 reveals that in general the rate constant k1 for direct
formation of the di-π-methane product is the largest of
J . Org. Chem, Vol. 68, No. 13, 2003 5061

Zimmerman and Novak
60 Å, 200-400 mesh) mixed with 1% (v/v) of fluorescent
indicator (Sylvania 2282 green phosphor, UV254) and slurry
packed into quartz columns to allow monitoring with a hand-
held UV lamp. Tetrahydrofuran, diethyl ether, and 1,4-dioxane
were purified by storage over potassium hydroxide, followed
by successive distillation under a nitrogen atmosphere from
calcium hydride and sodium benzophenone ketyl. Dimethyl
sulfoxide was distilled from calcium hydride prior to use.
Photograde benzene (1 L) was prepared by washing two times
with a mixture of 100 mL of saturated potassium permanga-
nate and 10 mL of sulfuric acid, water, saturated sodium
bicarbonate, and brine, drying over calcium chloride, and
distilling from calcium hydride.
1H NMR spectra were recorded at 300 MHz and are reported
in ppm downfield from tetramethylsilane. UV spectra were
measured at 25 °C in hexane. The high-resolution electron
impact mass spectra were run at 150 °C source temperature
on a MS50TC ultra-high-resolution mass spectrometer manu-
factured by Kratos, Inc. (Manchester, England). The samples
were run via direct insertion probe and the data collected via
a Kratos DS-55 data acquisition system.
F IGURE 2. Energy levels as a function of singlet-triplet
separation and the exchange integral.
SCHEME 9. Alter n a tive Rea ction P a th w a ys
Lea d in g to Tr i-π-m eth a n e P r od u cts
Gen er a l P r oced u r e for Kin etic P h otolyses. Kinetic
solution photolyses were carried out in sealed NMR tubes (17
cm × 0.5 cm o.d.) in d6-benzene under nitrogen at 25 °C. The
tube was irradiated in a Black-box apparatus.5b Three filter
cells were used (2.0 M NiSO4 in 5% H2SO4, 0.004 M SnCl2 in
15% HCl, and 0.8 M CoSO4 in 5% H2SO4) providing a 300-
350-nm band-pass.
Gen er a l P r oced u r e for P r ep a r a tive Solu tion P h otoly-
ses. Preparative photolyses were carried out with an immer-
sion well apparatus and a Hanovia 450 W medium-pressure
mercury lamp equipped with a 5-mm recirculating filter
solution of 0.020 M copper sulfate. The solution was purged
with deoxygenated and dried nitrogen for 1 h prior to and
during photolysis.
Gen er a l P r oced u r e for X-r a y Cr ysta llogr a p h y An a ly-
sis. X-ray diffraction data were collected on a Bruker-AXS P4
diffractometer with a Smart 1000 CCD area detector for single
crystals of each compound. Lorentz and polarization correction
were applied, and each structure was solved with the ap-
propriate space group symmetry by direct methods with use
of SHELXTL12a and SHELXS.12b Hydrogen atom positions were
calculated at idealized positions and included in the structure
factor calculation with fixed isotropic displacement param-
eters. Full-matrix least-squares refinement on F2 was carried
out employing anisotropic displacement parameters for all non-
hydrogen atoms. The Crystallographic Information Files (cif
files) for all compounds studied by X-ray crystallography are
provided as Supporting Information.
the three involved (note Scheme 9) in both the direct and
the naphthalene-sensitized runs. However, the question
was whether there is a direct route for formation of the
tri-π-methane photoproduct rather than merely a route
proceeding via initial formation of the di-π-methane
product. Comparison in Table 3 of k2 with k3 shows that
the direct route has an effective rate constant k2 that is
from 2.3- to 15-fold larger than k3, the rate constant for
the secondary process. We need to recognize that these
rate constants are “operational” and include effects due
to differential light absorption in the direct irradiation
runs and effects due to differential energy transfer in the
naphthalene singlet-sensitized runs. In both the direct
and naphthalene sensitized S1 reactions we can conclude
that the direct route for formation of the tri-π-methane
photoproduct is the dominant one.
Con clu sion
Eth yl 2-Alk yl-2-(2,2-d ip h en ylvin yl)-4,4-d ip h en yl-3-bu -
ten oa te (17). To a solution of 3.2 mL (22.8 mmol, 2 equiv) of
diisopropylamine in 90 mL of anhydrous THF at -10 °C was
added dropwise 10.6 mL (15.77 mmol, c 1.49, 1.4 equiv) of
butyllithium in hexane. After the mixture was stirred for 30
min, a solution of 5.0 g (11.25 mmol) of ethyl 2-(2,2-diphen-
ylvinyl)-4,4-diphenyl-3-butenoate3 (16) in 30 mL of anhydrous
THF was added dropwise. The solution was allowed to warm
to 10 °C, when 3.0 equiv of alkyl bromide was added. The
resulting solution was stirred 2 h at room temperature before
being quenched with addition of 10% hydrochloric acid. The
mixture was diluted with ethyl ether, successively washed with
saturated NaHCO3, water, and brine, and dried over anhy-
drous magnesium sulfate. Concentration in vacuo afforded a
brown-yellow oil, which was crystallized from ethanol.
Ethyl 2-benzyl-2-(2,2-diphenylvinyl)-4,4-diphenyl-3-butenoate
(17a ): 85.6%; mp 118-119 °C (lit.3 mp 117-119 °C) 1H NMR
(CDCl3, 300 MHz) δ 0.96 (t, J ) 7.2 Hz, 3H), 3.22 (s, 2H), 3.42
(q, J ) 7.1 Hz, 2H), 6.18 (s, 2H), 7.12-7.27 (m, 25H).
Ethyl 2-isopropyl-2-(2,2-diphenylvinyl)-4,4-diphenyl-3-bu-
tenoate (17b): 83%; mp 121 °C; 1H NMR (CDCl3, 300 MHz) δ
0.99 (d, J ) 6.9 Hz, 6H), 1.03 (t, J ) 7.2 Hz, 3H), 1.23 (heptet,
J ) 6.8 Hz, 1H), 3.48 (q, J ) 7.1 Hz, 2H), 6.29 (s, 2H), 7.15-
Three new examples of the tri-π-methane rearrange-
ment have been found and their regioselectivity studied.
There is a parallelism with the di-π-methane rearrange-
ment in the initial π-π-bridging step with a preference
for the more delocalized cyclopropyldicarbinyl diradical.
The dependence on multiplicity appears to result from
the tendency of triplets to afford the transoid conformer
of the allylic-carbinyl diradical resulting from three-ring
opening of the cyclopropyldicarbinyl diradical. Finally,
the tri-π-methane product is shown to arise by two routes,
a direct one and a two-step one that proceeds via a di-
π-methane product initially formed. A kinetic treatment
shows that the direct route is the predominant one.
Exp er im en ta l Section
Gen er a l P r oced u r es. All reactions were performed under
an atmosphere of dry nitrogen. Melting points were deter-
mined in open capillaries with a Meltemp heating block.
Column chromatography was performed on silica gel (Aldrich,
5062 J . Org. Chem., Vol. 68, No. 13, 2003

Tri-π-methane Rearrangement
7.26 (m, 20H); MS-EI 486 (2), 443 (12), 291 (12), 263 (100),
3-Isopropyl-3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-
pentadiene (20b): 45%; mp 91 °C; 1H NMR (CDCl3, 300 MHz)
δ 0.93 (d, J ) 6.6 Hz, 6H), 1.31 (d, J ) 1.5 Hz, 3H), 1.64 (d, J
) 1.5 Hz, 3H), 1.87 (h, J ) 6.6 Hz, 1H), 4.89 (t, J ) 1.2 Hz,
1H), 6.17 (s, 2H), 6.88-7.28 (m, 20H); 1H NMR (C6D6, 300
MHz) δ 1.01 (d, J ) 6.6 Hz, 6H), 1.37 (d, J ) 1.5 Hz, 3H), 1.70
(d, J ) 1.5 Hz, 3H), 2.03 (h, J ) 6.6 Hz, 1H), 5.09 (t, J ) 1.2
Hz, 1H), 6.42 (s, 2H), 7.00-7.26 (m, 20H); UV (hexane) λ(max)
246 (ꢀ 32 867), λ 278 (ꢀ 16 667); MS-EI 468 (8), 425 (100), 291
(18), 167 (56); HRMS (EI) (M+calc) 468.2817, (M+found) 468.2829.
Syn th esis of 1,1,5,5-Tetr aph en yl-1,4-pen tadien -3-ol (12).
To the cooled (0 °C) slurry of 11 g (28.5 mmol) of 1,1,5,5-
tetraphenyl-1,4-pentadien-3-one14 (11) in 100 mL of ethanol
was added 1.14 g (30 mmol) of NaBH4 in portions. The mixture
was kept at reflux for an hour. The resulting solution was
poured into 300 mL of 1 M NaOH solution. The phases were
separated and the water phase was washed by ether. The
combined organic phases were washed with water and dried
over MgSO4. Concentration in vacuo gave 10.3 g of slightly
yellow oil. 1H NMR spectroscopy showed the oil was 95% pure.
The product was not purified further because of the decom-
position on column or during crystallization.
167 (15); HRMS (ESI) (M + Na+calc) 509.2457, (M + Na+
)
found
509.2433.
2-Alk yl-2-(2,2-d ip h en ylvin yl)-4,4-d ip h en yl-3-b u t en ol
(18). To the suspension of 2.5 g (67 mmol, 4.27 equiv) of
lithium aluminum hydride in 100 mL of anhydrous THF was
added 15.7 mmol of ethyl 2-alkyl-2-(2,2-diphenylvinyl)-4,4-
diphenyl-3-butenoate (17) in 50 mL of anhydrous THF. The
reaction mixture immediately turned blue-green and was
stirred for 2-5 h at 25 °C (described bellow). The excess
lithium aluminum hydride was quenched by cautious addition
of 3.0 mL of water and 5.0 mL of 1.0 M NaOH. The gray solid
was filtered and washed with 100 mL of ether. The filtrate
was dried and concentrated in vacuo to afford a clear oil, which
was crystallized from ethanol.
2-Ben zyl-2-(2,2-diph en ylvin yl)-4,4-diph en yl-3-bu t en ol
(18a ): 2 h at 25 °C; 81.2%; mp 137-139 °C (lit.3 mp 137-139
°C); 1H NMR (CDCl3, 300 MHz) δ 1.27 (t, J ) 6.0 Hz, 1H),
2.98 (s, 2H), 3.12 (d, J ) 6.6 Hz, 2H), 5.80 (s, 2H), 6.92-7.33
(m, 25H).
2-Isopropyl-2-(2,2-diphenylvinyl)-4,4-diphenyl-3-butenol
(18b): 5 h at 25 °C; 88%; oil; 1H NMR (CDCl3, 300 MHz) δ
1.07 (d, J ) 7.2 Hz, 6H), 1.27 (t, J ) 6.0 Hz, 1H), 2.13 (heptet,
J ) 6.9 Hz, 1H), 3.29 (d, J ) 6.6 Hz, 2H), 5.92 (s, 2H), 6.95-
7.33 (m, 20H); MS-EI 444 (1), 401 (100), 305 (76), 191 (57),
1,1,5,5-Tetraphenyl-1,4-pentadien-3-ol (12): 89%; 1H NMR
(CDCl3) δ 1.62 (d, J ) 3.0 Hz, 1H), 4.88 (dt, J 1 ) 9.3 Hz, J 2
3.3 Hz, 1H), 6.18 (d, J ) 9.0 Hz, 2H), 7.36-6.95 (m, 20H).
)
167 (85); HRMS (EI) [(M - iPr)+calc] 401.1905, [(M - iPr)+
401.1899.
]
Ethyl-3-(1,1,5,5-tetraphenyl-1,4-pentadienyl) ether (12a )
was formed during recrystallization from ethanol. The satu-
rated solution of 1,1,5,5-tetraphenyl-1,4-pentadien-3-ol in
ethanol was cooled to -20 °C. After weeks yellow crystals
found
2-Alk yl-2-(2,2-d ip h en ylvin yl)-4,4-d ip h en yl-3-b u t en a l
(19). To the suspension of 4.62 g (21.4 mmol) of pyridinium
chlorochromate13 in 60 mL of anhydrous CH2Cl2 was added
10.72 mmol of 2-alkyl-2-(2,2-diphenylvinyl)-4,4-diphenyl-3-
butenol (18) in one portion. After 2 h 60 mL of ether was added
and the supernatant decanted from the black gum. The
insoluble residue was washed thoroughly 3 times with 30-mL
portions of ether whereupon it became a black granular solid.
The combined organic solution was passed through a short pad
of silica gel. Concentration in vacuo afforded a green-yellow
solid, which was recrystallized from hexane.
1
precipitated: mp 112 °C; H NMR (CDCl3) δ 1.10 (t, J ) 7.2
Hz, 3H), 3.31 (q, J ) 7.2 Hz, 2H), 4.50 (t, J ) 9.3 Hz, 1H),
6.14 (d, J ) 9.3 Hz, 2H), 7.36-6.93 (m, 20H); HRMS (ESI) (M
+ Na+calc) 439.2038, (M + Na+found) 439.2038.
Syn th esis of 3-(2,2-Bis(4-br om op h en yl)eth en yl)-1,1,5,5-
tetr a p h en ylp en ta -1,4-d ien e (13). To the solution of 7.76 g
(0.02 mol) of 1,1,5,5-tetraphenyl-1,4-pentadien-3-ol (12) and
0.02 mol of 1,1-bis(4-bromophenyl)ethylene15 in 50 mL of
anhydrous dioxane was added 0.50 mL of concentrated sulfuric
acid. The solution turned blue and then was heated to reflux
for 5 h. The mixture was cooled, successively washed with
saturated NaHCO3, water, and brine, and dried over anhy-
drous magnesium sulfate. The dioxane was removed at
reduced pressure to afford oil. Chromatography on silica gel
column eluted with hexane gave white crystals.
2-Ben zyl-2-(2,2-diph en ylvin yl)-4,4-diph en yl-3-bu t en a l
(19a ): 92.2%; mp 139-140 °C (lit.3 mp 139-141 °C). 1H NMR
(CDCl3, 300 MHz) δ 3.15 (s, 2H), 6.03 (s, 2H), 6.97-7.33 (m,
25H), 8.98 (s, 1H).
2-Isopropyl-2-(2,2-diphenylvinyl)-4,4-diphenyl-3-butenal
1
(19b): 88.6%; mp 144 °C; H NMR (CDCl3, 300 MHz) δ 1.00
3-(2,2-Bis(4-bromophenyl)ethenyl)-1,1,5,5-tetraphenylpenta-
(d, J ) 7.5 Hz, 6H), 2.24 (heptet, J ) 7.1 Hz, 1H), 6.17 (s, 2H),
7.10-7.27 (m, 20H), 9.09 (s, 1H); MS-EI 442 (38), 413 (50),
1
1,4-diene (13): 53%; mp 193 °C; H NMR (CDCl3) δ 4.24 (q,
360 (60), 323 (73), 233 (94), 167 (100); HRMS (EI) (M+
442.2297, (M+found) 442.2306.
)
calc
1H, J ) 9.8 Hz), 6.03 (d, 2H, J ) 9.9 Hz), 6.04 (d, 1H, J )
10.2 Hz), 6.40-6.49 (m, 2H), 6.65-6.72 (m, 4H), 6.97-7.40 (m,
22H); HRMS (EI) ([M - H]-calc) 705.0793, ([M - H]-
705.0800.
)
found
3-Alkyl-(2,2-dim eth ylvin yl)-1,1,5,5-tetr a p h en yl-1,4-pen -
ta d ien e (20). To the suspension of 5.34 g (12.36 mmol, 1.25
equiv) of triphenylisopropyl phosphonium iodide in 90 mL of
anhydrous THF at -10 °C was added 7.7 mL (11.5 mmol, c
1.448, 1.125 equiv) of butyllithium dropwise. The resulting red
solution was allowed to stir for 30 min and then 9.89 mmol of
2-alkyl-2-(2,2-diphenylvinyl)-4,4-diphenyl-3-butenal (19) in 40
mL of anhydrous THF was quickly added. The solution
immediately turned blue and was stirred for 3 h at room
temperature. After the reaction was completed, 1.0 mL of
acetone was added, and the solvent was removed in vacuo.
The resulted mixture was filtered through a short plug of silica
gel (2 × 2 cm) with 20% ether in hexane as eluent. Evaporation
of the solvents in vacuo resulted yellow oil, which was
crystallized from hexane.
3-Benzyl-3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-pen-
tadiene (20a ): 60%; mp 110 °C; 1H NMR (CDCl3, 300 MHz) δ
1.25(d, J ) 1.2 Hz, 3H), 1.58 (d, J ) 1.2 Hz, 3H), 2.99 (s, 2H),
4.81 (m,1H), 6.02 (s, 2H), 6.90-7.26 (m, 25H); 1H NMR (C6D6,
300 MHz) δ 1.27 (d, J ) 1.5 Hz, 3H), 1.59 (d, J ) 1.5 Hz, 3H),
3.10 (s, 2H), 4.96 (t, J ) 1.4 Hz, 1H), 6.22 (s, 2H), 6.80-7.36
(m, 25H); UV (hexane) λ(max) 254 (ꢀ 31 720), λ 218 (ꢀ 31 600),
λ(shoulder) 232 (ꢀ 29 300); MS-EI 516 (3), 425 (100), 291 (16),
167 (52); HRMS (EI) (M+calc) 516.2817, (M+found) 516.2822.
Syn th esis of 3-(2,2-Bis(4-cya n op h en yl)eth en yl)-1,1,5,5-
tetr a p h en ylp en ta -1,4-d ien e (14). A mixture of 11 g (15.5
mmol) of 3-(2,2-bis(4-bromophenyl)ethenyl)-1,1,5,5-tetra-
phenylpenta-1,4-diene (13), 20 g (15 equiv) of cuprous cyanide,
and 100 mL of N-methyl-2-pyrrolidinone was heated to reflux
for 5.5 h. After cooling, the mixture was shaken with a solution
of 8.0 g of sodium cyanide in 200 mL of water. This was
benzene extracted, washed with 10% sodium cyanide and then
water, dried over anhydrous sodium sulfate, and concentrated
in vacuo to give oil. Chromatography on silica gel column
eluted with hexane and dichloromethane gave white crystals.
3-(2,2-Bis(4-cyanophenyl)ethenyl)-1,1,5,5-tetraphenylpenta-
1,4-diene (14): 77.4%; mp 206 °C; 1H NMR (CDCl3) δ 4.11 (q,
1H, J ) 9.8 Hz), 6.05 (d, 2H, J ) 9.6 Hz), 6.22 (d, 1H, J )
10.2 Hz), 6.65-6.75 (m, 6H), 6.95-7.15 (m, 4H), 7.15-7.30 (m,
16H), 7.53-7.60 (m, 2H); HRMS (EI) (M+calc) 600.2565, (M+
600.2569.
)
found
Syn th esis of 3-Ben zyl-3-(2,2-bis(4-cya n op h en yl)eth e-
n yl)-1,1,5,5-tetr a p h en ylp en ta -1,4-d ien e (15). To a suspen-
(15) Matsumoto, T.; Ishida, T.; Koga, N.; Iwamura, H. J . Am. Chem.
Soc. 1992, 114 (25), 9952-9959.
J . Org. Chem, Vol. 68, No. 13, 2003 5063

Zimmerman and Novak
sion of 0.42 g (17.5 mmol, 1.5 equiv) of sodium hydride in 150
mL of DMSO was added 7.11 g (11.8 mmol) of 3-(2,2-bis(4-
cyanophenyl)ethenyl)-1,1,5,5-tetraphenylpenta-1,4-diene (14)
in 50 mL of a mixture of anhydrous DMSO-THF (1:1). After
the deep blue solution was stirred for 6 h at room temperature,
4.0 mL (34 mmol, 3 equiv) of benzyl bromide was added
quickly. The mixture was diluted with water and extracted
with ether. Concentration of the extract in vacuo afforded
brown oil, which was purified on a silica gel column with
hexane-dichloromethane eluent.
MHz) δ 1.37 (d, J ) 0.9 Hz, 3H), 1.50 (d, J ) 0.9 Hz, 3H), 2.32
(d, J ) 13.8 Hz, 1H), 2.82 (d, J ) 13.8 Hz, 1H), 2.98 (d, J )
11.1 Hz, 1H), 5.71 (m, 1H), 5.95 (d, J ) 11.1 Hz, 1H), 6.86-
7.62 (m, 25H); MS-EI 516 (7), 425 (100), 291 (15), 167 (15);
HRMS (EI) (M+calc) 516.2817, (M+found) 516.2811. The structure
assignment was conformed by X-ray crystallography (see
Supporting Information).
trans-2-Benzyl-2-(2-methylpropenyl)-1,1-diphenyl-3-(2,2-
diphenylvinyl)cyclopropane (22a ): 18% after 50 min of ir-
radiation; mp 162-163 °C from ethanol; 1H NMR (CDCl3, 300
MHz) δ 1.00 (s, 3H), 1.25 (s, 3H), 2.59 (d, J ) 10.8 Hz, 1H),
2.88 (d, J ) 14.7 Hz, 1H), 3.25 (d, J ) 14.4 Hz, 1H), 4.95 (m,
3-Benzyl-3-(2,2-bis(4-cyanophenyl)ethenyl)-1,1,5,5-tetra-
phenylpenta-1,4-diene (15): 30.5%; mp 178 °C; 1H NMR
(CDCl3) δ 3.04 (s, 2H), 5.76 (s, 2H), 6.12 (s, 1H), 6.54-6.66
(m, 6H), 6.84-6.90 (m, 2H), 7.00-7.40 (m, 20H); 1H NMR
(C6D6) δ 2.96 (s, 2H), 5.80 (s, 2H), 6.04 (s, 1H), 6.24-6.28 (m,
2H), 6.40-6.51 (m, 2H), 6.72-7.20 (m, 24H); UV (hexane)
λ(max) 204 (ꢀ 91 736), λ 262 (ꢀ 32 500); MS (EI) 599 (11), 382
(20), 197 (68), 181 (100); MS (MALDI, Ag+, anthracene) 972
(4, M + Ag+ + anthracene-5H), 970 (5, M + Ag+ + anthracene-
5H), 799 (1, M + Ag+), 797 (1, M + Ag+), 109 (88, Ag+), 107
(100, Ag+); HRMS (EI) ([M - CH2Ph]calc) 599.2487, ([M -
CH2Ph]found) 599.2489. The structure assignment was con-
firmed by X-ray crystallography (see Supporting Information).
P r in s Rea ction of 1,1-Dip h en yleth ylen e (7) w ith 3,3-
Bis(4-ch lor op h en yl)-2-p r op r en a l (10-Cl). To the solution
of 1.0 g (5.56 mmol) of 1,1-diphenylethylene16a,b (7) and 1.54 g
(5.56 mmol) of 3,3-bis(4-chlorophenyl)-2-proprenal16a (10-Cl)
in 10 mL of dry 1,4-dioxane was added 0.025 g (0.13 mmol) of
p-toluenesulfonic acid monohydrate. The solution was heated
to reflux for 30 h in N2 atmosphere. The solvent was evapo-
rated in vacuo and the oil was filtered through a short plug of
Florisil (3 × 1) and then through a plug of Alumina (3 × 1)
with dichloromethane as eluent. The solvent was evaporated
in vacuo. The mixture was purified by column chromatography
(5% ethyl acetate in hexane, silica gel). The concentration in
vacuo afforded crystals, which were recrystallized from ethanol
to yield 52% of 3-(2,2-diphenylvinyl)-1,1,5,5-tetraphenyl-1,4-
pentadiene (9): mp 226-228 °C (lit.17 mp 228-230 °C (n-
1
1H), 6.02 (d, J ) 11.1 Hz, 1H), 6.95-7.63 (m, 25H); H NMR
(C6D6, 300 MHz) 1.08 (d, J ) 0.9 Hz, 3H), 1.20 (d, J ) 0.9 Hz,
3H), 2.93 (d, J ) 10.8 Hz, 1H), 3.05 (d, J ) 14.4 Hz, 1H), 3.31
(d, J ) 14.7 Hz, 1H), 5.15 (m, 1H), 6.33 (d, J ) 11.1 Hz, 1H),
6.85-7.65 (m, 25H); MS-EI 516 (8), 425 (100), 291 (15), 167
(64); HRMS (EI) (M+calc) 516.2817, (M+found) 516.2819.
1-Benzyl-3,3-dimethyl-4,4-diphenyl-5-(2,2-diphenylvinyl)-1-
cyclopentene (23a ): 31% after 50 min of irradiation; mp 107
°C from ethanol; 1H NMR (CDCl3, 300 MHz) δ 0.97 (d, J ) 1.5
Hz, 3H), 1.57 (d, J ) 0.9 Hz, 3H), 3.62 (d, J ) 15.3 Hz, 1H),
3.82 (d, J ) 15.3 Hz, 1H), 4.31 (t, J ) 9.6 Hz, 1H), 5.18 (dm,
J ) 9.6 Hz, 1H), 5.92 (d, J ) 10.2 Hz, 1H), 6.75-7.35 (m, 25H);
1H NMR (C6D6, 300 MHz) δ 1.14 (d, J ) 1.2 Hz, 3H), 1.63 (d,
J ) 1.2 Hz, 3H), 3.68 (d, J ) 15.6 Hz, 1H), 3.93 (d, J ) 15.3
Hz, 1H), 4.63 (t, J ) 9.6 Hz, 1H), 5.39 (dm, J ) 9.3 Hz, 1H),
6.18 (d, J ) 9.9 Hz, 1H), 6.85-7.40 (m, 25H); MS-EI 516 (26),
425 (64), 181 (80), 119 (100); HRMS (EI) (M+calc) 516.2817,
(M+found) 516.2816.
Kin etics: Dir ect Solu tion P h otolysis of 3-Ben zyl-3-
(2,2-d im et h ylvin yl)-1,1,5,5-t et r a p h en yl-1,4-p en t a d ien e
(20a ). A solution of 10 mg (0.019 mmol) of 3-benzyl-3-(2,2-
dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-pentadiene (20a ) in 2.5
mL of d6-benzene in a NMR tube was irradiated for varying
times (see Supporting Information). The course of the reaction
was monitored by 1H NMR spectroscopy. The spectral char-
acteristics of 21a , 22a , and 23a were the same as before (see
above).
Kin etics: Sen sitized (Na p h th a len e) Solu tion P h otoly-
sis of 3-Ben zyl-3-(2,2-d im eth ylvin yl)-1,1,5,5-tetr a p h en yl-
1,4-p en ta d ien e (20a ). A solution of 10 mg (0.019 mmol) of
3-benzyl-3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-penta-
diene (20a ) and 90 mg (0.71 mmol) of naphthalene in 2.5 mL
of d6-benzene in a NMR tube was irradiated for varying times
(see Supporting Information). The course of the reaction was
monitored by 1H NMR spectroscopy. The spectral character-
istics of 21a , 22a , and 23a were the same as before (see above).
1
BuOH)); H NMR (CDCl3) δ 4.33 (q, J ) 10 Hz, 1H), 6.05 (d,
J ) 10 Hz, 3H), 6.70-7.40 (m, 30H).
Dir ect Solu tion P h otolysis of 3-Ben zyl-3-(2,2-d im eth -
ylvin yl)-1,1,5,5-tetr a p h en yl-1,4-p en ta d ien e (20a ). A solu-
tion of 150 mg (0.29 mmol) of 3-benzyl-3-(2,2-dimethylvinyl)-
1,1,5,5-tetraphenyl-1,4-pentadiene (20a ) in 200 mL of benzene
was irradiated. Concentration in vacuo yielded yellow oil.
Conversion and the composition of the reaction mixture were
1
determined by H NMR spectroscopic analysis. The mixtures
at 66% conversion (12 min of irradiation) contained 34% of
starting material (20a), 19% of trans-2-benzyl-2-(2-methylpro-
penyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)cyclopropane (22a ),
24% of cis-2-benzyl-2-(2-methyl-propenyl)-1,1-diphenyl-3-(2,2-
diphenylvinyl)cyclopropane (21a ), and 23% of 1-benzyl-3,3-
dimethyl-4,4-diphenyl-5-(2,2-diphenylvinyl)-1-cyclopentene (23a).
The mixtures at 100% conversion (50 min of irradiation)
contained 23% of trans-2-benzyl-2-(2-methylpropenyl)-1,1-
diphenyl-3-(2,2-diphenylvinyl)cyclopropane (22a ), 35% of cis-
2-benzyl-2-(2-methylpropenyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)-
cyclopropane (21a ), and 42% of 1-benzyl-3,3-dimethyl-4,4-
diphenyl-5-(2,2-diphenylvinyl)-1-cyclopentene (23a ). The com-
pounds were purified by column chromatography (silica gel,
hexane as eluent) and by crystallization from ethanol.
cis-2-Benzyl-2-(2-methyl-propenyl)-1,1-diphenyl-3-(2,2-di-
phenylvinyl)cyclopropane (21a ): 30% after 50 min of irradia-
tion; mp 178-179 °C from ethanol; 1H NMR (CDCl3, 300 MHz)
δ 1.36 (s, 3H), 1.63 (s, 3H), 2.14 (d, J ) 14.1 Hz, 1H), 2.66 (d,
J ) 10.8 Hz, 1H), 2.78 (d, J ) 14.1 Hz, 1H), 5.51 (m, 1H), 5.59
(d, J ) 10.8 Hz, 1H), 6.83-7.56 (m, 25H); 1H NMR (C6D6, 300
Sen sitized (Acetop h en on e) Solu tion P h otolysis of
3-Ben zyl-3-(2,2-d im et h ylvin yl)-1,1,5,5-t et r a p h en yl-1,4-
p en ta d ien e (20a ). A solution of 150 mg (0.29 mmol) of
3-benzyl-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-pentadi-
ene (20a ) and 5.0 mL (43 mmol) of acetophenone in 200 mL of
benzene was irradiated for 18 min. Concentration in vacuo
1
yielded 180 mg of yellow oil. H NMR spectroscopic analysis
showed 100% conversion and the composition of the reaction
mixture was 28% of trans-2-benzyl-2-(2-methylpropenyl)-1,1-
diphenyl-3-(2,2-diphenylvinyl)cyclopropane (22a ), 67% of cis-
2-benzyl-2-(2-methylpropenyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)-
cyclopropane (21a ), and 5% of 1-benzyl-3,3-dimethyl-4,4-
diphenyl-5-(2,2-diphenylvinyl)-1-cyclopentene (23a ). The com-
pounds were purified by column chromatography (silica gel,
hexane as eluent) and by crystallization from ethanol. The
spectral characteristics of 21a , 22a , and 23a were the same
as before (see above), and the yields were 58%, 19%, and 2%,
respectively.
Dir ect Solu tion P h otolysis of 3-Isop r op yl-3-(2,2-d i-
m eth ylvin yl)-1,1,5,5-tetr a p h en yl-1,4-p en ta d ien e (20b). A
solution of 150 mg (0.32 mmol) of 3-isopropyl-3-(2,2-dimeth-
ylvinyl)-1,1,5,5-tetraphenyl-1,4-pentadiene (20b) in 200 mL of
benzene was irradiated. Concentration in vacuo yielded yellow
oil. Conversion and the composition of the reaction mixture
(16) (a) Schmidle, C. J .; Barnett, P. G. J . Am. Chem. Soc. 1956, 78,
3209-3210. (b) Bellassoued, M.; Majidi, A. J . Org. Chem. 1993, 58,
2517-2522.
(17) Ho¨ft, E.; Gru¨ndemann, E.; Gross, H. Liebigs Ann. Chem. 1969,
722, 225-227.
5064 J . Org. Chem., Vol. 68, No. 13, 2003

Tri-π-methane Rearrangement
were determined by 1H NMR spectroscopic analysis. The
mixtures at 75% conversion (9 min of irradiation) contained
25% of starting material, 20% of trans-2-isopropyl-2-(2-meth-
ylpropenyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)cyclopropane (22b),
21% of cis-2-isopropyl-2-(2-methylpropenyl)-1,1-diphenyl-3-
(2,2-diphenylvinyl)cyclopropane (21b), and 34% of 1-isopropyl-
3,3-dimethyl-4,4-diphenyl-5-(2,2-diphenylvinyl)-1-cyclopen-
tene (23b). The mixtures at 100% conversion (35 min of
irradiation) contained 17% of trans-2-isopropyl-2-(2-methyl-
propenyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)cyclopropane (22b),
20% of cis-2-isopropyl-2-(2-methylpropenyl)-1,1-diphenyl-3-
(2,2-diphenylvinyl)cyclopropane (21b), and 63% of 1-isopropyl-
3,3-dimethyl-4,4-diphenyl-5-(2,2-diphenylvinyl)-1-cyclopen-
tene(23b).Thecompoundswerepurifiedbycolumn chromatography
(silica gel, hexane as eluent) and by crystallization from
ethanol.
cis-2-Isopropyl-2-(2-methylpropenyl)-1,1-diphenyl-3-(2,2-
diphenylvinyl)cyclopropane (21b): 9% after 35 min of irradia-
tion; mp 161-162 °C from ethanol; 1H NMR (CDCl3, 300 MHz)
δ 0.79 (s, 3H), 0.86 (m, 1H), 0.89 (s, 3H), 1.64 (d, J ) 1.2 Hz,
3H), 1.70 (d, J ) 1.2 Hz, 3H), 2.46 (d, J ) 13.2 Hz, 1H), 5.65
(d, J ) 12.9 Hz, 1H), 5.67 (m, 1H), 6.97-7.58 (m, 20H); 1H
NMR (C6D6, 300 MHz) δ 0.86 (d, J ) 6.6 Hz, 3H), 1.01 (d, J )
6.3 Hz, 3H), 1.12 (m, 1H), 1.56 (d, J ) 1.2 Hz, 3H), 1.71 (d, J
) 0.9 Hz, 3H), 2.82 (d, J ) 11.1 Hz, 1H), 5.85 (m, 1H), 6.05 (d,
J ) 11.1 Hz, 1H), 6.85-7.63 (m, 20H); MS-EI 468 (25), 425
(100), 167 (74); HRMS (EI) (M+calc) 468.2817, (M+found) 468.2822.
(CDCl3, 300 MHz) δ 0.65 (d, J ) 6.6 Hz, 3H), 1.04 (d, J ) 6.6
Hz, 3H), 1.28 (d, J ) 0.9 Hz, 3H), 1.50 (d, J ) 0.9 Hz, 3H),
2.20 (heptett, J ) 6.6 Hz, 1H), 4.57 (d, J ) 11.1 Hz, 1H), 4.69
(d of m, J ) 11.1 Hz, 1H), 6.29 (s, 1H), 6.90-7.86 (m, 20H);
13C NMR (CDCl3, 500 MHz) δ 18.40 (dC-CH3), 24.30 (CH-CH3),
24.41 (CH-CH3), 25.88 (dC-CH3), 28.15 (C-CH3), 59.86 (C-H),
62.61 (CPh2), 76.56 ((iPr)C-CPh2), 125.33 (CHd), 131.92 (Ph2C-
CHd), 132.40 ((CH3)2Cd), 155.72 ((iPr)Cd); HRMS (EI) (M+
)
calc
468.2817, (M+found) 468.2814.
Sen sitized (Acetop h en on e) Solu tion P h otolysis of
3-Isop r op yl-3-(2,2-d im eth ylvin yl)-1,1,5,5-tetr a p h en yl-1,4-
p en ta d ien e (20b). A solution of 10 mg (0.021 mmol) of
3-isopropyl-3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-pen-
tadiene (20b) and 90 mg (0.75 mmol) of acetophenone in 2.5
mL of d6-benzene in a NMR tube was irradiated for 4 min.
Concentration in vacuo yielded 17 mg of yellow oil. The
mixture was purified by flash column chromatography (silica
gel, hexane as eluent). 1H NMR spectroscopic analysis showed
45% conversion and the composition of the mixture was 4% of
trans-2-isopropyl-2-(2-methylpropenyl)-1,1-diphenyl-3-(2,2-
diphenylvinyl)cyclopropane (22b), 21% of cis-2-isopropyl-2-(2-
methylpropenyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)cyclopro-
pane (21b), 16% of 4-(2-methylpropenyl)-1-isopropyl-3,3,5,5-
tetraphenyl-1-cyclopentene (24b), a trace of 1-isopropyl-3,3-
dimethyl-4,4-diphenyl-5-(2,2-diphenylvinyl)-1-cyclopentene (23b),
and 4% of cis-2-isopropyl-2-(2,2-diphenylvinyl)-1,1-diphenyl-
3-(2-methylpropenyl)cyclopropane (cis-34) (1H NMR (CDCl3,
300 MHz) δ 5.49 (d, J ) 11.4 Hz, 1H, CHdC(CH3)), 5.55 (s,
1H, CHdCPh2); cis-34 did not form in sufficient quantities to
permit isolation and its assignment was based on NMR).
Kin etics: Dir ect Solu tion P h otolysis of 3-Isop r op yl-
3-(2,2-d im e t h ylvin yl)-1,1,5,5-t e t r a p h e n yl-1,4-p e n t a d i-
en e (20b). A solution of 10 mg (0.021 mmol) of 3-isopropyl-
3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-pentadiene (20b)
in 2.5 mL of d6-benzene in a NMR tube was irradiated for
varying times (see Supporting Information). The course of the
reaction was monitored by 1H NMR spectroscopy. The spectral
characteristics of 21b, 22b, and 23b were the same as before
(see above).
trans-2-Isopropyl-2-(2-methylpropenyl)-1,1-diphenyl-3-(2,2-
diphenylvinyl)cyclopropane (22b): 10% after 35 min of ir-
radiation; mp 147-148 °C from ethanol; 1H NMR (CDCl3, 300
MHz) δ 0.77 (s, 3H), 0.85 (m, 1H), 0.91 (s, 3H), 1.29 (d, J )
1.2 Hz, 3H), 1.35 (d, J ) 1.5 Hz, 3H), 2.51 (d, J ) 11.4 Hz,
1H), 5.08 (m, 1H), 5.84 (d, J ) 11.4 Hz, 1H), 7.00-7.59 (m,
1
20H); H NMR (C6D6, 300 MHz) δ 0.86 (d, J ) 6.6 Hz, 3H),
1.03 (d, J ) 6.6 Hz, 3H), 1.08 (m, 1H), 1.26 (d, J ) 1.2 Hz,
3H), 1.36 (d, J ) 1.2 Hz, 3H), 2.83 (d, J ) 11.4 Hz, 1H), 5.29
(m, 1H), 6.12 (d, J ) 11.4 Hz, 1H), 6.85-7.66 (m, 20H); MS-
EI 468 (20), 425 (100), 167 (85); HRMS (EI) (M+calc)468.2817,
(M+found) 468.2809.
1-Isopropyl-3,3-dimethyl-4,4-diphenyl-5-(2,2-diphenylvinyl)-
1-cyclopentene (23b): 52% after 35 min of irradiation; oil; 1H
NMR (CDCl3, 300 MHz) δ 0.98 (t (dd), J ) 7.05 Hz, 6H), 1.11
(d, J ) 0.9 Hz, 3H), 1.61 (d, J ) 0.9 Hz, 3H), 2.77 (h, J ) 7.2
Hz, 1H), 4.05 (t, J ) 9.3 Hz, 1H), 5.25 (d of m, J ) 9.3 Hz,
Kin etics: Sen sitized (Na p h th a len e) Solu tion P h otoly-
sis of 3-Isop r op yl-3-(2,2-d im eth ylvin yl)-1,1,5,5-tetr a p h en -
yl-1,4-p en ta d ien e (20b). A solution of 10 mg (0.021 mmol)
of 3-isopropyl-3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-
pentadiene (20b) and 90 mg (0.71 mmol) of naphthalene in
2.5 mL of d6-benzene in a NMR tube was irradiated for varying
times (see Supporting Information). The course of the reaction
was monitored by 1H NMR spectroscopy. The spectral char-
acteristics of 21b, 22b, and 23b were the same as before (see
above).
Dir ect Solu tion P h otolysis of 3-Ben zyl-3-(2,2-bis(4-
cya n op h en yl)et h en yl)-1,1,5,5-t et r a p h en ylp en t a -1,4-d i-
en e (15). A solution of 150 mg (0.22 mmol) of 3-benzyl-3-(2,2-
bis(4-cyanophenyl)ethenyl)-1,1,5,5-tetraphenylpenta-1,4-
diene (15) in 200 mL of benzene was irradiated for 50 min.
Concentration in vacuo yielded yellow oil. Conversion and the
composition of the reaction mixture were determined by 1H
NMR spectroscopic analysis. The mixtures contained 27% of
trans-2-benzyl-1,1-bis(4-cyanophenyl)-2,3-bis(2,2-diphenylethe-
nyl)cyclopropane (26), 34% of cis-2-benzyl-1,1-bis(4-cyanophen-
yl)-2,3-bis(2,2-diphenylethenyl)cyclopropane (25), and 39% of
1-benzyl-4,4-bis(4-cyanophenyl)-3,3-diphenyl-5-(2,2-diphenylvi-
nyl)-1-cyclopentene (27). The compounds were purified by
column chromatography (silica gel, hexane-dichloromethane
100:10 as eluent) and by crystallization from ethanol to a
constant mp of 170-171 °C.
1
1H), 6.13 (d, J ) 9.6 Hz, 1H), 6.87-7.45 (m, 20H); H NMR
(C6D6, 300 MHz) δ 1.08 (d, J ) 2.7 Hz, 3H), 1.10 (d, J ) 2.7
Hz, 3H), 1.24 (d, J ) 1.2 Hz, 3H), 1.65 (d, J ) 1.2 Hz, 3H),
2.93 (h, J ) 7.05 Hz, 1H), 4.34 (t, J ) 9.2 Hz, 1H), 5.46 (d of
m, J ) 9.3 Hz, 1H), 6.36 (d, J ) 9.9 Hz, 1H), 6.83-7.47 (m,
20H); MS-EI 468 (31), 425 (100), 167 (90); HRMS (EI) (M+
468.2817, (M+found) 468.2828.
)
calc
Sen sitized (Acetop h en on e) Solu tion P h otolysis of
3-Isop r op yl-3-(2,2-d im eth ylvin yl)-1,1,5,5-tetr a p h en yl-1,4-
p en ta d ien e (20b). A solution of 150 mg (0.32 mmol) of
3-isopropyl-3-(2,2-dimethylvinyl)-1,1,5,5-tetraphenyl-1,4-pen-
tadiene (20b) and 5.0 mL (43 mmol) of acetophenone in 200
mL of benzene was irradiated for 8 min. Concentration in
vacuo yielded 167 mg of yellow oil. 1H NMR spectroscopic
analysis showed 100% conversion and the composition of the
reaction mixture was 10% of trans-2-isopropyl-2-(2-methyl-
propenyl)-1,1-diphenyl-3-(2,2-diphenylvinyl)cyclopropane (22b),
45% of cis-2-isopropyl-2-(2-methylpropenyl)-1,1-diphenyl-3-
(2,2-diphenylvinyl)cyclopropane (21b), 45% of 4-(2-methylpro-
penyl)-1-isopropyl-3,3,5,5-tetraphenyl-1-cyclopentene (24b), and
a trace of 1-isopropyl-3,3-dimethyl-4,4-diphenyl-5-(2,2-diphen-
ylvinyl)-1-cyclopentene (23b). The compounds were purified
by column chromatography (silica gel, hexane as eluent) and
by crystallization from ethanol. The spectral characteristics
of 21b and 22b were the same as before (see above), and the
yields were 33% and 5%, respectively.
trans-2-Benzyl-1,1-bis(4-cyanophenyl)-2,3-bis(2,2-diphen-
ylethenyl)cyclopropane (26): 23%; mp 190 °C; 1H NMR (CDCl3,
300 MHz) δ 2.75 (d, J ) 15.3 Hz, 1H), 2.76 (d, J ) 9.9 Hz,
1H), 3.26 (d, J ) 15.3 Hz, 1H), 5.70 (d, J ) 9.0 Hz, 1H), 5.71
(s, 1H), 6.30-6.36 (br d, J ) 8.1 Hz, 2H), 6.48-6.55 (br d, J )
6.6 Hz, 2H), 6.91-7.65 (m, 29H); 1H NMR (C6D6, 300 MHz) δ
2.69 (d, J ) 14.1 Hz, 1H), 2.73 (d, J ) 7.4 Hz, 1H), 3.11 (d, J
4-(2-Methylpropenyl)-1-isopropyl-3,3,5,5-tetraphenyl-1-cy-
clopentene (24b): 36%; mp 164 °C from ethanol; 1H NMR
J . Org. Chem, Vol. 68, No. 13, 2003 5065

Zimmerman and Novak
) 14.7 Hz, 1H), 5.62 (s, 1H), 5.86 (d, J ) 9.9 Hz, 1H), 6.25-
6.31 (br d, J ) 7.2 Hz, 2H), 6.49-6.56 (br d, J ) 8.4 Hz, 4H),
6.85-7.28 (m, 27H); MS (MALDI, Ag+, anthracene) 972 (4, M
+ Ag+ + anthracene-5H), 970 (4, M + Ag+ + Anthracene-5H),
799 (1, M + Ag+), 797 (1, M + Ag+), 109 (98, Ag+), 107 (100,
Ag+). Anal. Calcd for C52H38N2: C, 90.40; H, 5.54; N, 4.05.
Found: C, 90.32; H, 5.55; N, 4.12.
cis-2-Benzyl-1,1-bis(4-cyanophenyl)-2,3-bis(2,2-diphenylethe-
nyl)cyclopropane (25): 28%; mp 233-234 °C; 1H NMR (CDCl3,
300 MHz) δ 2.21 (d, J ) 5.5 Hz, 2H), 2.72 (d, J ) 10.7 Hz,
1H), 5.72 (d, J ) 10.7 Hz, 1H), 6.18 (s, 1H), 6.76-6.82 (m,
2H), 6.97-7.56 (m, 27H), 7.65 (s, 4H); 1H NMR (C6D6, 300
MHz) δ 2.20 (br s, 2H), 2.79 (d, J ) 10.7 Hz, 1H), 5.86 (d, J )
10.7 Hz, 1H), 6.37 (s, 1H), 6.72-7.45 (m, 33H); MS (MALDI,
Ag+, anthracene) 972 (13, M + Ag+ + anthracene-5H), 970 (14,
M + Ag+ + anthracene-5H), 799 (3, M + Ag+ ), 797 (3, M +
Ag+), 109 (99, Ag+), 107 (100, Ag+). Anal. Calcd for C52H38N2:
C, 90.40; H, 5.54; N, 4.05. Found: C, 90.42; H, 5.49; N, 4.08.
The structure assignment was confirmed by X-ray crystal-
lography (see Supporting Information).
Kin etics: Sen sitized (Na p h th a len e) Solu tion P h otoly-
sis of 3-Ben zyl-3-(2,2-bis(4-cya n op h en yl)eth en yl)-1,1,5,5-
tetr a p h en ylp en ta -1,4-d ien e (15). A solution of 10 mg (0.015
mmol) of 3-benzyl-3-(2,2-bis(4-cyanophenyl)ethenyl)-1,1,5,5-
tetraphenylpenta-1,4-diene (15) and 90 mg (0.71 mmol) of
naphthalene in 2.5 mL of d6-benzene in a NMR tube was
irradiated for varying times (see Supporting Information). The
course of the reaction was monitored by 1H NMR spectroscopy.
The spectral characteristics of 25, 26, and 27 were the same
as before (see above).
Sen sitized (Acetop h en on e) Solu tion P h otolysis of
3-Ben zyl-3-(2,2-b is(4-cya n op h en yl)et h en yl)-1,1,5,5-t et -
r a p h en ylp en ta -1,4-d ien e (15). A solution of 150 mg (0.22
mmol) of 3-benzyl-3-(2,2-bis(4-cyanophenyl)ethenyl)-1,1,5,5-
tetraphenylpenta-1,4-diene (15) and 5.0 mL (43 mmol) of
acetophenone in 200 mL of benzene was irradiated for 20 min.
1
Concentration in vacuo yielded 175 mg of yellow oil. H NMR
spectroscopic analysis showed 100% conversion and the com-
position of the reaction mixture was 47% of trans-2-benzyl-
1,1-bis(4-cyanophenyl)-2,3-bis(2,2-diphenylethenyl)cyclopro-
pane (26) and 53% of cis-2-benzyl-1,1-bis(4-cyanophenyl)-2,3-
bis(2,2-diphenylethenyl)cyclopropane (25). The compounds
were purified by column chromatography (silica gel, hexane-
dichloromethane 100:10 as eluent) and by crystallization from
ethanol. The spectral characteristics of 26 and 25 were the
same as before (see above), and the yields were 41% and 43%,
respectively.
1-Benzyl-4,4-bis(4-cyanophenyl)-3,3-diphenyl-5-(2,2-diphen-
1
ylvinyl)-1-cyclopentene (27): 30%; mp 170-171 °C; H NMR
(CDCl3, 300 MHz) δ 3.57 (br d, J ) 3.7 Hz, 2H), 4.79-4.88
(m, 2H), 6.29 (br d, J ) 3.6 Hz, 1H), 6.43-6.47 (m, 2H), 6.72-
7.56 (m, 31H); 1H NMR (C6D6, 300 MHz) 4.30 (br s, 2H), 4.75-
4.84 (m, 2H), 6.15 (br s,1H), 6.50-6.57 (m, 2H), 6.59-7.70 (m,
4H), 6.82-7.17 (m, 27H); MS (MALDI, Ag+, anthracene) 972
(17, M + Ag+ + anthracene-5H), 970 (18, M + Ag+
+
anthracene-5H), 799 (5, M + Ag+ ), 797 (6, M + Ag+), 109 (97,
Ag+), 107 (100, Ag+). Anal. Calcd for C52H38N2: C, 90.40; H,
5.54; N, 4.05. Found: C, 90.52; H, 5.41; N, 4.06.
Ack n ow led gm en t. Support of this research by the
National Science Foundation is gratefully acknowledged
with special appreciation for its support of basic re-
search.
Kin etics: Dir ect Solu tion P h otolysis of 3-Ben zyl-3-
(2,2-bis(4-cyan oph en yl)eth en yl)-1,1,5,5-tetr aph en ylpen ta-
1,4-d ien e (15). A solution of 10 mg (0.015 mmol) of 3-benzyl-
3-(2,2-bis(4-cyanophenyl)ethenyl)-1,1,5,5-tetraphenylpenta-1,4-
diene (15) in 2.5 mL of d6-benzene in a NMR tube was
irradiated for varying times (see Supporting Information). The
course of the reaction was monitored by 1H NMR spectroscopy.
The spectral characteristics of 25, 26, and 27 were the same
as before (see above).
Su p p or tin g In for m a tion Ava ila ble: Kinetic data, X-ray
coordinates for three compounds (15, 21a , and 25) and CIF
files, ABC kinetic iterative details for two typical runs. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O034189J
5066 J . Org. Chem., Vol. 68, No. 13, 2003