Characterization of methylenepropenylidenecyclohexadiene derivatives and their competing 1,6-electrocyclic reaction and 1,7-hydrogen at room temperature

Full text of DOI:10.1021/ja9621455

J. Am. Chem. Soc. 1996, 118, 10311-10312
10311
Scheme 1
Characterization of
Methylenepropenylidenecyclohexadiene Derivatives
and Their Competing 1,6-Electrocyclic Reaction and
1,7-Hydrogen Shift at Room Temperature
Mariella Mella, Mauro Freccero, and Angelo Albini*
Dipartimento Chimica Organica
UniVersita` di PaVia, Viale Taramelli 10
27100 PaVia, Italy
ReceiVed June 25, 1996
The interconversion between 1,3,5-hexatriene and 1,3-cyclo-
hexadiene and related electrocyclic reactions have been the
subject of many experimental and theoretical investigations.1
Thermal processes have been thermodynamically and kinetically
characterized and occur at relatively high temperatures.1b,c It
is interesting to characterize a system where the process occurs
at room temperature or below. Pentaene 2, expected to be
generated by photolysis of 1,2-dihydronaphthalene (1, Scheme
1), is an appropriate model, since there is a strong thermody-
namic drive toward rearomatization. The photochemistry of
compound 12 and of several of its derivatives3-8 has been
thoughroughly investigated, and the many products reported
have all been rationalized as arising from (thermal and
photochemical) ring closure, cycloaddition, and (whith alkyl-
substituted derivatives) sigmatropic rearrangements of o-xy-
lylenes of type 2.2-8 However, the generation of polyenes of
this type by the ring opening of fused cyclohexadienes has been
directly documented only in the particular case of triphenyl-
9,9a-dihydroanthracene.9 It occurred to us that incorporation
of conjugated substituents would allow direct observation of
this compound and a better discrimination between the following
reactions. We now report that polyenes of type 2 can be
spectroscopically and chemimically characterized starting from
some cyano derivatives of 1.
acetonitrile solution of 3 resulted in a conspicuous end of pulse
absorption between 370 and 500 nm (maximum at 450 nm, see
Figure 1a). This was red-shifted by 60 nm with respect to parent
o-xylylene12 and closely correponded to that reported for its
1,8-diphenyl derivative.9 Therefore, it was assigned to the
5-methylene-6-(2-propenylidene)-1,3-cyclohexadiene 4 (see
Scheme 1). Assuming that the absorbtivity of this compound
is close to that of o-xylylene12 allowed a rough evaluation of
the quantum yield for ring opening (Φe ca 0.1); this rearrange-
ment is expected to occur in the picosecond range.1e The
transient absorption showed a first-order decay, with k ) 4.2
s-1 at 25 °C. Since no irreversible reaction occurs under this
condition, this can be safely attributed to the thermally allowed
cyclization (k ) kc, Scheme 1).
Irradiation of the 1-benzyl derivative 5 caused an efficient
reaction leading to the alkenes 6 as a mixture of the four (two
main) stereoisomers (Scheme 1). No attempt to draw any
conclusion for the isomeric distribution was made due to the
easy E/Z isomerization of the alkene in steady state irradiations.
When the R,R-D2 derivative was used,10 selective deuterium
incorporation at the methyl group was observed (see Scheme
1). The quantum yield of reaction (see Table 1) was affected
little by deuteration. Flash photolysis evidenced a transient with
shape and intensity similar to the previous case (Figure 1b) but
much shorter-lived and with a marked deuterium effect [k(H)
) 440 s-1, k(D) ) 80 s-1, D effect of 5.5]. These data show
that the polyene 7 is formed and is the intermediate for the end
products, which arise through a [1,7]-sigmatropic hydrogen shift.
The efficiency of the photochemical ring opening is not greatly
changed with respect to the conversion of compound 3 into 4
(Φe ca 0.1 in both cases). With 5, this is near to the quantum
yield for the irreversible reaction. Since there is no reason to
expect that electrocyclic ring closure is significantly accelerated
in this case, the fast decay observed for 7 is attributed to
sigmatropic shift (k ) kc+ ks ≈ ks) in Scheme 1) with little
contribution from the electrocyclic reaction. In accord with this
mechanism, the deuterium effect shows up only on the decay
rate of the intermediate, not on the steady state quantum yield
of the final products [Φ ) Φeks/(kc+ ks), where ks . kc] and is
The dinitrile 310 underwent no irreversible reaction by
irradiation in both polar and apolar solvents (λ 254 nm, Φ-3 e
3 × 10-3). However, flash photolyzing (λ 254 nm)11 an
(1) (a) Woodward, R. B.; Hoffmann, R. The ConserVation of Orbital
Symmetry; Verlag Chemie, GmbH/Academic Press: Weinheim, 1970. (b)
Marvel, E. N. Thermal Electrocyclization Reactions in Organic Chemistry;
Academic Press: New York, 1980. (c) Bakulev, V. A. Russ. Chem. ReV.
1995, 64, 99. (d) Baldwin, J. E.; Reddy, V. P.; Schaad, L. J.; Hess, B. A.
J. Am. Chem. Soc. 1988, 110, 8554, 8555. (e) Reid, P. J.; Doig, S. J.;
Wickham, S. D.; Mathies, R. A. J. Am. Chem. Soc. 1993, 115, 4754. (f)
Evanseck, J. D.; Thomas, B. E., IV; Spellmeyer, D. C.; Houck, K. N. J.
Org. Chem. 1995, 60, 7134. (g) Dolbier, W. R.; Koroniak, H.; Burton, D.
J.; Bailey, A. R.; Shaw, G. S.; Hansen, S. W. J. Am. Chem. Soc. 1984,
106, 1871 (h) McCullough, J. J. Acc. Chem. Res. 1980, 13, 270. (i) Okamura,
W. M.; De Lara, A. K. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 5, p 699.
(2) Salisbury, K. Tetrahedron Lett. 1971, 737.
(3) Kleinhuis, H.; Wijting, R. L. C.; Havinga, E. Tetrahedron Lett. 1971,
255.
(4) Widmer, U.; Heimgartner, H.; Schmid, H. HelV. Chim. Acta 1975,
58, 2210.
(5) Lamberts, J. J. M.; Laarhoven, W. H. J. Am. Chem. Soc. 1984, 106,
1736.
(6) Lamberts, J. J. M.; Laarhoven, W. H. Recl. TraV. Chim. Pays-Bas
1984, 103, 131. Lamberts, J. J. M.; Cuppen, T. J. H. M.; Laarhoven, W. H.
J. Chem. Soc. 1985, 1819. Woning, J.; Lijten, F. A. T.; Laarhoven, W. H.
J. Org. Chem. 1991, 56, 2427.
(7) Laarhoven, W. H. L. Org. Photochem. 1987, 9, 129.
(8) Seeley, D. A. J. Am. Chem. Soc. 1972, 94, 4378.
(9) Grellmann, K. H.; Palmowski, J.; Quinkert, G. Angew. Chem., Int.
Ed. Engl. 1971, 10, 196.
(10) (a) The dihydronaphthalenedicarbonitriles 3, 5, and 8 were photo-
chemically prepared from 1,4-naphthalenedicarbonitrile, refs 10b,c. The
deuterated derivatives were prepared using R,R-D2-trimethylbenzylsilane.
(b) Albini, A.; Fasani, E.; Oberti, R. Tetrahedron 1982, 38, 1034. (c)
Sulpizio, A.; Albini, A.; d’Alessandro, A.; Fasani, E.; Pietra, S. J. Am. Chem.
Soc. 1989, 111, 5773.
(12) Flynn, C. R.; Michl, J. J. Am. Chem. Soc. 1974, 96, 3280.
(13) Baldwin, J. E.; Reddy, V. P. J. Am. Chem. Soc. 1987, 109, 8051.
(14) Spangler, C. W. Chem. ReV. 1976, 76, 187.
(15) (a) More O’Ferral, R. A. J. Chem. Soc. B 1970, 785. (b) Kwart, H.
Acc. Chem. Res. 1982, 15, 401.
(11) By means of a Nd-YAG laser, λ ) 266 nm, pulse duration 10 ns,
power 20 mJ.
S0002-7863(96)02145-2 CCC: $12.00 © 1996 American Chemical Society

10312 J. Am. Chem. Soc., Vol. 118, No. 42, 1996
Communications to the Editor
intensity to the previous cases, and, thus Φe was ca 0.1 also in
this case. However, the decay was slower than with 4 (k ) 0.8
s-1) and the deuterium effect was negligible within the
experimental error. The steady state quantum yield (sum of
that for isomerization and that for ring opening to 9) was again
close to Φe. In this case both reactions, thermally allowed
disrotatory ring closure and sigmatropic H shift, lead to distinct
products, respectively the trans 1,2-disubstituted dihydronaph-
thalene 10 and the alkenes 9.16 These occurred with a
comparable efficiency. Thus, both kc and ks contributed
significantly to the observed transient decay rate in this case,
viz., electrocyclic cyclization competed with sigmatropic rear-
rangement with the polyene 11 but not with isomeric 7.
Furthermore, the latter reaction was ca 103 times slower with
11 than with 7 (see insets in Figure 1b,c). This could be
rationalized as due to different conformations of the two
compounds. MO calculations (semiempirical AM1)17 showed
that the energy minimum for 7 lay at a conformation well suited
for hydrogen transfer with the benzylic hydrogen colinear with
the relevant π orbital (in accord with the observed deuterium
effect), whereas the minimun for 11 had the central C-C bond
in the butadiene moiety rotated outward. Thus both electrocyclic
and sigmatropic reactions were subjected to a conformational
barrier.
In conclusion, methylenepropylidenecyclohexadienes are
generated by photoinduced ring opening of 1,2-dihydronaph-
thalenes. There structure is supported by correspondence of
the visible spectrum with expectation, as well as by the observed
thermal reactions (ring closure with 4, various rearrangement
with 7 and 11). Ring opening is a general reaction, since
formation of a color upon irradiation of other derivatives of 1
has been previously reported.3,7,8 Under the chosen condition,
secondary photochemical reactions are avoided and thermal
processes of these polyenes are cleanly observed. Thermal
reclosure occurs at room temperature in tenths of seconds
through one of the fastest known hexatriene ring closure
reactions.18 When there is an allylic hydrogen atom, a [1,7]-
sigmatropic shift occurs and is by 2 orders of magnitude faster
than the electrocyclic reactions. This is similar to the difference
measured in non-benzo-fused trienes,2b,c,14 but occurs here at
room temperature and is many orders of magnitude faster. As
it appears for 11, in this temperature range a conformational
barrier strongly hinders both reactions. The thermal and
photochemical reactivity of the polyenes 4, 7, and 1 will be the
subject of further quantitative studies.
Figure 1. End of pulse spectra obtained by 266 nm flashing of the
dihydronaphthalenes 3 (a), 5 (b), and 8 (c) in acetonitrile. Insets: kinetic
trace for the decay at the absorbance maximum (a) for 3 at 450 nm,
100 ms per division; (b) for 5 at 430 nm, 10 ms per division; and (c)
for 8 at 450 nm, 200 ms per division.
Table 1. Rate Constants (k) for the Decay of the
Flash-Photolytically Observed Transients and Quantum Yield for
Irreversible Quantum Yield (Φ)
substrate
k, s-1
4.2
440 (80)a
Φ-(substrate)
product proportion
3
5
8
Φ
-3 e 3 × 10-3
Φ-5 ) 0.13 (0.115)a 6 (100%)
0.8 (0.75)a Φ-8 ) 0.21 (0.19)a
9 (62%), 10 (38%)
a For 5- or 8-d2.
Acknowledgment. Partial suppport of this work by CNR and
MURST is gratefully acknowleged.
close to that observed for the corresponding shift of octatrienes
(6.4-7.7).13,14 This is actually the expected value for a linear
C-H-C geometry in the excited state.15
JA9621455
Irradiation of the cis 2-benzyl derivative 8 gave again a
mixture of alkenes 9, along with the isomeric trans dihy-
dronaphthalene 10. With the R,R-D2 derivative, deuterium was
selectively incorporated at the methylene group R to the cyano
function (Scheme 1). Again, a transient was detected and
attributed to polyene 11 (see Figure 1c). It was similar in
(16) The measured quantum yield for isomerization to 10 was a minimal
value, since some E/Z isomerization of the alkene may have taken place
during the steady state irradiation before reclosure..
(17) Austin M1.
(18) The only fastest related ring closure is that observed with the
o-xylylene obtained from 9,9,10-triphenyl-9,9a-dihydroanthracene, see ref
9.