The cyclohexadienyl radical in the thermal syn-anti isomerization of two crossed pentaenes of the type of bis-homofulvalene

Article abstract of DOI:10.1021/ja953416z

To augment the elucidation of the dependence of enthalpy of stabilization on configuration among the pentadienyl radicals, a designer pentaene of the (Z,Z) type, (R,R)-5,5',6,6',7,7',8,8'-octahydro-4a,4a'-dimethyl-2,2'-bi-4a(H)-naph thylidene, has been prepared and subjected to a kinetic study of cis, trans isomerization about its central double bond. The resulting activation parameters and those of a more complicated example, 3,3'-bicholesta-1,4-dienylidene, are essentially identical, their mean values being ΔH(≠) = 36.8 kcal/mol and ΔS(≠) = +1.8 eu. This enthalpy of activation is significantly higher than those of pentaenes of the (E,E) and (E,Z) configuration. Conjugative interaction in coplanar cross-conjugated systems of the bis-homofulvalene type not having been experimentally evaluated or estimated by theoretical calculation, an enthalpy of stabilization for the cyclohexadienyl radical is not definitively to be extracted at the present time. If conjugative interactions are to be equated to four Kistiakowsky butadiene units and the difference in steric energies between educt and 90°-twisted state are to be estimated by the MMEVBH program, an enthalpy of stabilization of the cyclohexadienyl radical of ~20 kcal/mol results.

Full text of DOI:10.1021/ja953416z

+
+
6
660
J. Am. Chem. Soc. 1996, 118, 6660-6665
The Cyclohexadienyl Radical in the Thermal Syn-Anti
Isomerization of Two Crossed Pentaenes of the Type of
Bis-Homofulvalene
W. von E. Doering,* Ludmila Birladeanu, Keshab Sarma, and Li-Sming Shao
Contribution from the Department of Chemistry, HarVard UniVersity,
Cambridge, Massachusetts 02138-2902
X
ReceiVed October 10, 1995. ReVised Manuscript ReceiVed April 1, 1996
Abstract: To augment the elucidation of the dependence of enthalpy of stabilization on configuration among the
pentadienyl radicals, a designer pentaene of the (Z,Z) type, (R,R)-5,5′,6,6′,7,7′,8,8′-octahydro-4a,4a′-dimethyl-2,2′-
bi-4a(H)-naphthylidene, has been prepared and subjected to a kinetic study of cis, trans isomerization about its central
double bond. The resulting activation parameters and those of a more complicated example, 3,3′-bicholesta-1,4-
q
q
dienylidene, are essentially identical, their mean values being ∆H ) 36.8 kcal/mol and ∆S ) +1.8 eu. This
enthalpy of activation is significantly higher than those of pentaenes of the (E,E) and (E,Z) configuration. Conjugative
interaction in coplanar cross-conjugated systems of the bis-homofulvalene type not having been experimentally
evaluated or estimated by theoretical calculation, an enthalpy of stabilization for the cyclohexadienyl radical is not
definitively to be extracted at the present time. If conjugative interactions are to be equated to four Kistiakowsky
butadiene units and the difference in steric energies between educt and 90°-twisted state are to be estimated by the
MMEVBH program, an enthalpy of stabilization of the cyclohexadienyl radical of ∼20 kcal/mol results.
The (Z,Z) is one of three configurations of the pentadienyl
radical. Although it has remained undetected experimentally
In the present work, two crossed pentadienes corresponding to
the (Z,Z) configuration of the pentadienyl radical, shown in
Scheme 1, have been prepared: one in the steroid series
parenthetical to a vain search for an effect of “solvent friction”
in its paradigmatic, alicyclic form, it has been thoroughly
1
scrutinized theoretically by Fort, Hrovat, and Borden. By
7
b,d
contrast, a cyclic relative, the cyclohexadienyl radical, which
is the accepted intermediate in the addition of radicals to benzene
and derivatives, has attracted extensive experimental attention.
Efforts to bring its enthalpy of stabilization to ground include
examination of the kinetics of the reaction of cyclohexa-1,3-
on rate of isomerization;
another, less highly substituted,
consonant with the previously studied, (E,E) and (E,Z) pen-
taenes.
The first, 3,3′-bicholesta-1,4-dienylidene (1), is prepared as
a mixture of stereoisomers, anti-1 and syn-1 (anti: defined with
rings D and D′, trans and 10 and 10′ methyl groups, cis; syn,
analogously), by Mukaiyama-Tyrlik-McMurry reductive cou-
2
diene with nitric oxide by Benson and co-workers; determi-
nation of the heat of reaction of cyclohexa-1,4-diene with tert-
8
9
butyloxyl (photoacoustic calorimetry) by Griller and co-
pling of cholesta-1,4-dien-3-one; the second, anti-2 and syn-
2, similarly, from (R)-(+)-5,6,7,8-tetrahydro-4a-methyl-2(3H)-
naphthalenone (5), as depicted in Scheme 1. Although each
stereoisomer is obtained in essentially equal amounts, in both
instances a single pure stereoisomer can be separated by
crystallization. These are held to be syn-1 and anti-2, respec-
tively, on the basis of nuclear Overhauser enhancements (see
Experimental Section) and, in the case of syn-1, of an X-ray
3
workers; several investigations of the reaction of benzene and
hydrogen atoms, most recently given a unified interpretation
4
by Tsang; and loss of methyl radical from 3,3,6,6-tetrameth-
ylcyclohexa-1,4-diene.5 Perhaps surprisingly, the thermo-
chemistry of the cyclohexadienyl radical seems not to have been
assessed theoretically at the highest, currently affordable level.
Geometrical, thermal, first-order, cis-trans isomerization
about a double bond passes through a 90° conformation, which
is economically formulated as a singlet diradical. This reaction
has been used to gain information about the thermochemistry
,6
1
0
crystallographic analysis.
Auspiciously for study of the
kinetics of thermal interconversion of syn and anti isomers, the
vinyl hydrogen atoms of both isomers of 1 and 2 are quanti-
7
1
of the (E,E) and (E,Z) configurations of the pentadienyl radical.
tatively distinguishable in the H NMR spectra, as can be
inferred from the affiliated chemical shifts included in Scheme
X
Abstract published in AdVance ACS Abstracts, July 1, 1996.
1
and discussed in detail in the Experimental Section.
(1) Fort, R. C., Jr.; Hrovat, D. A.; Borden, W. T. J. Org. Chem. 1993,
Rearrangement of the bicholesta-1,4-dienylidenes (1) is clean
in that spectra taken at the longer times of reaction, at which
the (pseudo)equilibrium constant seems to be reassuringly close
5
8, 211-216.
2) Shaw, R.; Cruikshank, F. R.; Benson, S. W. J. Phys. Chem. 1967,
1, 4538-4543.
3) (a) Hawari, J. A.; Engel, P. S.; Griller, D. Int. J. Chem. Kinet. 1985,
(
7
(
1
7, 1215-1219. (b) Griller, D.; Wayner, D. D. M. ReV. Chem. Intermed.
(7) (a) Doering, W. von E.; Kitagawa, T. J. Am. Chem. Soc. 1991, 113,
4288-4297. (b) Doering, W. von E.; Birladeanu, L.; Cheng, X-h.;
Kitagawa, T.; Sarma, K. J. Am. Chem. Soc. 1991, 113, 4558-4563. (c)
Doering, W. von E.; Sarma, K. J. Am. Chem. Soc. 1992, 114, 6037-6043.
(d) Doering, W. von E.; Shi, Y.-q.; Zhao, D.-c. J. Am. Chem. Soc. 1992,
114, 10763-10766.
1
986, 7, 31-44. (c) Burkey, T. J.; Majewski, M.; Griller, D. J. Am. Chem.
Soc. 1986, 108, 2218-2221.
(4) (a) Tsang, W. J. Phys. Chem. 1986, 90, 1152-1155. (b) Nicovich,
J. M.; Ravishankara, A. R. J. Phys. Chem. 1984, 88, 2534-2541 (q. v. for
other relevant references). (c) Sauer, M. C., Jr.; Mani, I. J. Phys. Chem.
1970, 74, 59-63. (d) James, D. G. L.; Suart, R. D. Trans. Faraday Soc.
1968, 64, 2752-2769. (e) Sauer, M. C., Jr.; Ward, B. J. Phys. Chem. 1967,
71, 3971-3983. (f) Yang, K. J. Am. Chem. Soc. 1962, 84, 3795-3799.
(8) McMurry, J. E. Chem. ReV. 1989, 89, 1513-1524, and references
cited therein.
(9) (a) Pinder, A. R.; Robinson, R. J. Chem. Soc. 1952, 1224-1231.
(b) Wilds, A.; Djerassi, C. J. Am. Chem. Soc. 1946, 68, 1712-1715.
(10) This structure, determined by Mr. Mihai Azimioara, will be
published elsewhere.
(
5) W u¨ stefeld, M., reported in ref 6, Tables 25, 37, 38, and 45.
(6) Roth, W. R.; Staemmler, V.; Neumann, M.; Schmuck, C. Liebigs
Ann. 1995, 1061-1118.
S0002-7863(95)03416-0 CCC: $12 00 © 1996 American Chemical Society

+
+
Pentaenes of the Bis-HomofulValene Type
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6661
Table 1. Kinetics of Thermal Interconversion of Syn- and
Scheme 1
Anti-3,3′-Bicholesta-1,4-dienylidene (1) in Benzene-d
6
: Specific
Rate Constants and Activation Parameters
a
Kb
T, °C
k
1
154.0 ( 0.1
169.4 ( 0.1
186.1 ( 0.1
196.6 ( 0.1
3.10 ( 0.02c
13.75 ( 0.2
63.38 ( 0.5
170.8 ( 2.2
0.94
0.98
0.97
1.00
Arrhenius Parametersd
E ) 37.4 ( 0.5 kcal/mol
a
log A ) 13.6 ( 0.3
Eyring Parameterse
q
∆
∆
H ) 36.5 ( 0.5 kcal/mol
q
S ) 1.0 ( 1.2 eu
Thermodynamicsf
∆
∆
∆H° ) 0.4 ( 0.2 kcal/mol
∆S° ) 0.9 ( 0.5 eu
a
Starting isomer, mp 232 °C, is syn-1. Specific rate constants are
calculated by linear regression of the usual expression for reversible
first-order reactions: (k + k-1) ) (1/t) ln[(Xeq - X )(Xeq - X)]; values
(syn f anti)/k-1] are obtained
1
0
-
6 -1 b
of k in units of 10
s . Values of K[k
1
c
from values of Xeq that lead to the best fit of the data. Double standard
d
errors for 90% confidence limits. By linear regression of the Arrhenius
expression. e Calculated at 175.3 °C. Calculated by linear regression
f
of the usual expression, ln K ) T∆S - ∆H.
6
Table 2. Thermal Isomerization of anti-2 to syn-2 in Benzene-d :
Specific Rate and Equilibrium Constants and Activation Parameters
a
T, °C
k
1
K
1
1
1
1
1
1
1
44.5 ( 0.2
44.5 ( 0.2
52.2 ( 0.2
52.2 ( 0.2
60.9 ( 0.1
69.3 ( 0.1
69.3 ( 0.1
1.08 ( 0.01b
1.09 ( 0.01
2.54 ( 0.01
2.49 ( 0.01
6.10 ( 0.03
13.86 ( 0.17
13.92 ( 0.12
29.80 ( 0.21
30.45 ( 0.21
29.75 ( 0.64
1.07
1.07
1.06
1.07
1.06
1.08
1.05
1.07
1.05
1.07
to one, are uncontaminated, even by hydrogen atoms of the
aromatic type. However, recoveries, as measured against an
internal standard, fall off at the longer times of reaction.
Attempts to suppress this hap have come to little (details in
Experimental Section). However, beyond decreasing the preci-
sion of measured relative concentrations slightly, this complica-
tion is of little consequence provided both stereoisomers
disappear at the same rate, a condition that seems quite likely
satisfied.
177.2 ( 0.1
177.2 ( 0.1
1
77.4 ( 0.1
Arrhenius Parametersb
E ) 37.80 ( 0.11 kcal/mol
log A ) 13.82 ( 0.06
a
Eyring Parametersb,c
H ) 36.9 ( 0.12 kcal/mol
∆S ) 2.0 ( 0.3 eu
Although recoveries are no better at the longest times in the
simpler pentaenes, anti-2 and syn-2, the kinetics in the two
systems are essentially identical, as can be seen in Tables 1
and 2 (basic data given in Tables 1S and 2S, respectively, as
supporting information). The weighted mean of the two sets
q
∆
q
a
Calculated by linear regression of the usual expression for reversible
first-order reactions: (k + k-1) ) (1/t) ln[(Xeq - X )(Xeq - X)]; K )
Double all standard errors for 90%
Calculated at 161.0 °C.
1
6
0
-
-1 b
1.05 - 1.08; k × 10
confidence limits.
s .
q
c
of activation parameters affords an enthalpy of activation, ∆H
)
q
36.8 kcal/mol, and an entropy of activation, ∆S ) +1.8 eu.
indicates a small empirical stabilization; it is 1.1 kcal/mol higher
than that of (E,Z) pentaene 8, 0.7 kcal/mol above that of 2,2′-
bicholesta-3,5-dienylidene; and 4.4 kcal/mol above that of the
isomeric (E,E) pentaene 7.
least stabilizing of the three configurations of the pentadienyl
radicals, having an empirical stabilizing effect 2.2 kcal/mol less
than that of the (E,E)-pentadienyl radical. This observation is
qualitatively in agreement with the theoretical calculations of
Fort, Hrovat, and Borden of the unsubstituted (E,E) and (Z,Z)
pentadienyl radicals, the latter having been found to be less
well stabilized by 5.6 kcal/mol. Note, however, that (Z,Z)-penta-
1,4-dien-3-yl is sterically a far cry from the cyclohexa-1,4-dien-
-yl radical.
Although the precision of these values seems satisfactory, their
accuracy is open to some question owing to the entropy of
activation being more positive in value than those previously
7
d
7
b
7
d,12
q
On these bases, the (Z,Z) is the
encountered in isomerizations of this type (∆S -3 to -5 eu).
If the explanation lies in a more negative entropy of formation
in the starting pentadienes, there is little cause for concern. Such
an interpretation is consistent with a crystallographic study of
syn-1,10 which points to a near planar, quite rigid structure for
the cyclohexadienylidencyclohexadiene (4,4′-dihydrodiphenyl,
1
bis-homofulvalene) system.11
The experimental enthalpy of activation is 3.3 kcal/mol lower
than that of closely related triene 67d (see Scheme 2) and
3
(
11) (a) Traetteberg, M.; Bakken, P.; Almenningen, A.; L u¨ ttke, W.;
Behind the wide, long-standing interest in the perturbing effect
of substituents on the enthalpy of the carbon free radical lies
Janssen, J. J. Mol. Struct. 1982, 81, 87-103. L u¨ tke, J. Mol. Struct. 1982,
8
1
1, 87-00. (b) Noltemeyer, M.; Janssen, J.; L u¨ ttke, W. J. Mol. Struct.
982, 81, 105-112. This structure has been redetermined by Professor
(12) Two other (E,E) pentaenes7b (data on all previously studied trienes
and pentaenes in the series of semirigid polyenes are collected in Table 4
of ref 7d) are essentially indistinguishable.
Roland Boese, Institut f u¨ r Anorganische Chemie, Universit a¨ t Essen. and
will be published elsewhere.

+
+
6
662 J. Am. Chem. Soc., Vol. 118, No. 28, 1996
Von Eggers Doering et al.
Scheme 2
from one example to another. Disentwinement of these factors
is the object of much theoretical and experimental effort.
In the series of all-trans trienes, pentaenes, heptaenes, and
7
a,c
nonaenes, transannular interactions in the ground state were
assumed to dominate steric interactions and, within the pairs
of anti and syn isomers and to an acceptable approximation, to
be unchanging within the series of polyenes. They could safely
be neglected as significant contributors to observed differences
1
4
in enthalpies of activation. Transannular steric factors at the
0°-twisted conformation were assumed to be negligible owing
9
to the large increase in distance of separation between “ortho”
hydrogen atoms of the two conjoined ring systems. A further
factor may have been the influence on steric energy of bond
angles as they changed from those in the starting olefins to those
in the model free radicals, but no datum exists to our knowledge
on which to base a parametrization of force field programs, for
example, that might have been looked to as a means of
addressing the question.
In the comparison between pentaenes of the (E,E) and (E,Z)
type (cf. 7 and 8 in Scheme 2), these assumptions appear still
to be acceptable. But in the present comparison of (Z,Z)-2 with
(E,E)-7 and (E)-6, attenuation of transannular interactions must
differ significantly in its contribution to observed enthalpies of
activation.
Molecular mechanical or force field programs seem to offer
a straightforward way to disentwine the strands of steric
interactions in the educts from the four-stranded thread.15 Three
programs have been applied: the MM2 of Allinger;16 the
MM2ERW, presently the most extensively parametrized pro-
1
7
gram for hydrocarbons; and the MMEVBH, a program
combining the MM2ERW force field program with the valence
bond Hamiltonian of the Heisenberg type introduced by Maynau
the hope of establishing values for “enthalpies of stabilization”,
which might be useful in the prediction of the enthalpic part of
the rate constant in a variety of reactions mechanistically thought
to involve the free radical as an approximation to the rate-
determining transition state. Of paramount importance in this
exercise is the oft repeated, recognition that enthalpy of
stabilization does not exist independent of its definition; that
there is no absolute enthalpy of stabilization in the sense of
an enthalpy of formation; and that definitions may be created
ad libitum but survive only through their usefulness (cf.
and collaborators and further extended by Roth et al.6 The
resulting steric energies, tabulated in Table 3S (supplementary
information), are encouraging at the level of accuracy required
18
19
for significance. Among the three, MMEVBH affords the best
agreement with experimental differences in heats of formation
between pairs of anti and syn isomers. In particular, it predicts
the greater stability of the anti isomers, which the other two
programs do not.
1
3
6
The MMEVBH program, painstakingly constructed to handle
“
aromaticity”!).
steric factors in hydrocarbons, conjugated and unconjugated
olefins, radicals and diradicals, appears to be the best available
for estimation of the steric corrections needed in both the educt
and the 90°-twisted conformation. Applied to triene 6, the
calculated decrease in “steric energy” driving the rearrangement
is calculated to be 7.0 kcal/mol. When coupled with the intrinsic
Primary interest in this and past works has been confined to
evaluation of the magnitude of the effect of perturbations by
olefinic substituents on the paradigmatic thermal interconversion
of cis- and trans-ethene.7 These perturbations may result in an
empirical lowering or raising of the enthalpy of activation
20
rotational barrier of ethene (65.9) and conventional correction
for conjugative interaction by two Kistiakowsky units, the
experimental enthalpy of activation leads to an enthalpy of
stabilization (RSE) for the allyl radical of 13.2 kcal/mol.2
When applied to pentaene 2, a smaller steric correction of 4.2
(stabilizing or destabilizing effect, respectively). They are
considered to operate on the educt through steric and electronic
interactions and on the transition state (approximated by a 90°
twisted geometry) likewise through steric and electronic interac-
tions; that is, each of four strands of a thread need explicitly to
be considered in the construction of differences between starting
state and transition state. In the paradigm, the four factors are
set to zero by definition. In perturbations of the paradigm, the
so-called (electronic) enthalpy of stabilization defines the
electronic interaction of the perturbing substituent(s) and the
transient free radicals assumed to describe the transition state.
In this conceptual scheme, experimental enthalpy of activation
is composed of two electronic interactions of the substituent
with the paradigmsenthalpy of conjugation in the educt and
enthalpy of stabilization in the transition state, both hoped to
be constantssand two steric factors, which are expected to vary
1,22
(
14) In all cases, the observation of values of ∆∆fH° (favoring the anti
isomer) varying narrowly between 0.3 and 1.0 kcal/mol was considered
reassuring.
(
15) But see Lipkowitz, K. B. J. Chem. Educ. 1995, 72, 1070-1075.
(16) As made available in the CSC Chem3D Plus program by Cambridge
Scientific Computing, Inc.
17) Roth, W. R.; Adamczak, O.; Breuckmann, R.; Lennartz, H.-W.;
(
Boese, R. Chem. Ber. 1991, 124, 2499-2521.
18) See ref 6 for references.
tions.
(20) Doering, W. v. E.; Roth, W. R.; Bauer, F.; Breuckmann, R.;
Ebbrecht, T.; Herbold, M.; Schmidt, R.; Lennartz, H.-W.; Lenoir, D.; Boese,
R. Chem. Ber. 1989, 122, 1263-1275.
(
21) [(65.9) - (40.1 + 7.0) + (2 × 3.75)]/2 ) 13.2 kcal/mol.
(13) See, for example: Clark, K. B.; Wayner, D. D. M.; Demirdji, S.
(22) Doering, W. v. E.; Roth, W. R.; Bauer, F.; Boenke, M.; Breuckmann,
H.; Koch, T. H. J. Am. Chem. Soc. 1993, 115, 2447-2453.
R.; Ruhkamp, J.; Wortmann. O. Chem. Ber. 1991, 124, 1461-1470.

+
+
Pentaenes of the Bis-HomofulValene Type
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6663
kcal/mol emerges and leads, in combination with a total of four
Kistiakowsky units of conjugative interaction for the crossed
system, to an enthalpy of stabilization for the cyclohexadienyl
radical of 20.0 kcal/mol.2
Scheme 3
3-25
Electronic effects in the starting conjugated polyenes are
represented by conjugative interactions between the paradigm
and the perturber. In the polyene series, this factor is held to
enter only once and to have the value of a Kistiakowsky unit
(
3.75 kcal/mol) invariant with the order of the perturbing
7c,26-28
polyene.
This procedure would be incorrect if, as now
indicated by Roth’s MMEVBH calculations, the unit were to
increase (or decrease) with the number of double bonds in the
conjugated polyene.2 With this reservation, handling conjuga-
tive interaction in pentaene anti-7 and higher polyenes presents
no major difficulty.
7a
That is not true of the twice crossed pentaene anti-2, for which
there is neither any experimentally based enthalpy of conjugation
available nor any immediate prospect of having the heats of
combustion and sublimation determined for 1-(4′-4′-dimethyl-
cyclohexa-2,5-dienylidene)-4,4-dimethylcyclohexa-2,5-diene, the
simplest stable and available model for the bis-homofulvalene
system present in 1 and 2. Catalytic hydrogenation, an
alternative possibility, is suspect in polyenes containing tetra-
substituted double bonds.29
resulting heat of formation of the triene (+26.0 kcal/mol) is
.7 kcal/mol lower than that calculated by MMEVBH, 28.7 kcal/
mol. Not withstanding, to accept the operation of a total of
four Kistakowsky units in the starting, crossed pentaene 2 is a
simple, if possibly ingenuous, expedient, which leads to the
stabilization energy of 20.0 kcal/mol noted above.
2
Direct comparison of “enthalpies of stabilization” derived for
7
a
anti-7 (16.9 kcal/mol) and anti-2 (20.0 kcal/mol) is not
warranted because their stabilization enthalpies are defined in
different ways. The need for explicit recognition of this point
Indeed, not even the heat of formation of the simplest planar
crossed triene, methylenecyclohexa-1,4-diene (p-isotoluene), is
known with sufficient accuracy. From a study of its gas-phase
ion chemistry, Bartmess30 has deduced a heat of formation of
has been most clearly explicated by Roth in terms of the
34
shorthand notations, RSE1,3 and RSE1,4.
In Scheme 3 are
shown what would have been proper comparisons from this
point of view (not from some others!): anti-7 of type (E,E)-
RSE1,3 might have been compared with (Z,Z)-RSE1,3 or anti-2
3
6 ( 3 kcal/mol, 24 ( 3 kcal/mol above that of toluene (12.1
kcal/mol). This value is 10.7 kcal/mol below that of the heat
of formation, 46.7 kcal/mol, of a strain-free model calculated
(Z,Z)-RSE1,4 with (E,E)-RSE1,4. In the reality of this paper,
3
1
by means of Benson group equivalent values uncorrected for
conjugative interaction. Application of Bartmess’ proposed
uncertainty in the least favorable direction decreases the
discrepancy to 7.7 kcal/mol, a value that would be fully
accommodated by two Kistiakowsky units. Also bearing on
the problem is a preliminary result from Professor Rogers on
the heat of hydrogenation of 3-methylene-6,6-dimethylcyclo-
hexa-1,4-diene (∆HH2 -76.8 kcal/mol)32 to 1,1,4-trimethylcy-
anti-2 affords a stabilization energy defined in terms of a type
of RSE1,4, anti-7 in terms of a type of RSE1,3. Correction of
this dissimilarity is simply, if dubitatively, achieved by adding
one Kistiakowsky unit to the RSE1,3 value, 16.9 kcal/mol,
7a
derived for the (E,E)-pentadienyl radical anti-7, whence, all
other factors remaining equal, the hypothetical value RSE1,4 for
pentadienyl radical, (E,E)-RSE1,4, would become 20.7 kcal/mol.
Thus, little difference in stabilization energies is discernible
when both are similarly defined as stabilization energies of the
RSE1,4 type. By an identical correction, the stabilization
enthalpy of anti-2 could have been translated into an RSE1,3
value of 16.3 kcal/mol.
33
clohexane (estimated heat of formation, -50.8 kcal/mol). The
(23) We express many thanks to Professor Wolfgang R. Roth, Ruhr
Universit a¨ t, Bochum, for suggesting and making these calculations.
(
(
24) [(65.9) - (36.8 + 4.2) + (4 × 3.75)]/2 ) 20.0 kcal/mol.
25) Because the derivation includes correction for steric factors by the
In comparison to the several other values already in the
literature (RSE1,4) for the enthalpy of stabilization of the
MMEVBH program, the resulting value for the enthalpy of stabilization
can be applied to other systems for the purpose of predicting enthalpies of
activation involving the cyclohexadienyl radical only if this same program
is used to estimate any steric corrections.
2
3
4
5,6
cyclohexadienyl radical, 24.6; 25.4; 22.9; and 26.7 kcal/mol,
the value, 20.0 kcal/mol, from this work is low. Within
(
26) Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of
Organic Compounds, 2nd ed.; Chapman and Hall: London, 1986.
27) (a) As indicated in Table 17 of ref 6. The final effect would be an
reasonable uncertainties, it is in fair agreement with the value,
6
2
1.4 kcal/mol, recommended by Roth.
(
7c
Another sighting on the W and U configurations of the
increase in the stabilization energies of the polyenyl radicals as summarized.
Beyond this theoretical indication, there is no experimental evidence bearing
on the question. (b) With reference to the espousal of 5 kcal/mol for a
pentadienyl radicals is obtained by examination of the cyclo-
reversion of cyclobutane 9 and comparison with 10. As shown
35
1
Kistiakowsky unit, the reader is referred to the following exercises upon
2
6
in Scheme 4, the cyclobutane 9 can be produced by sensitized
irradiation at -75 °C of 4,4-dimethylmethylenecyclohexa-2,5-
diene. Although no distinction can be made between head-to-
head or head-to-tail structure on the basis of the NMR spectrum,
the rapid rate of disappearance of 9 and reappearance of the
monomeric triene at -31.6 °C is better reconciled by the head-
heats of formation: (i) butadiene, butene-1, n-butane; (ii) (E)- and (Z)-
pentadiene-1,3, (E)- and (Z)-pentene-2, pentene-1, n-pentane; (iii) 2-
methylbuta-1,3-diene, 2- and 3-methylbutene-1,2-methylbutane; (iV) hexa-
2
8
1
,3,5-triene, hexa-1,5-diene, (Z)-hexene-3, n-hexane; (V) 2,3-dimethyl-
butadiene, 2,3-dimethylbutene-1,2,3-dimethylbutane.
28) (a) Fang, W.; Rogers, D. W. J. Org. Chem. 1992, 57, 2294-2297.
b) Turner, R. B.; Mallon, B. J.; Tichy, M.; Doering, W. v. E.; Roth, W.
R.; Schr o¨ der, G. J. Am. Chem. Soc. 1973, 95, 8605-8610.
29) Turner, R. B.; Meador, W. R.; Doering, W. v. E.; Knox, L. H.;
Mayer, J. R.; Wiley, D. W. J. Am. Chem. Soc. 1957, 79, 4127-4133.
30) (a) Bartmess, J. E. J. Am. Chem. Soc. 1982, 104, 335-337; (b)
Bartmess, J. E.; Griffith, S. S. J. Am. Chem. Soc. 1990, 112, 2931-2936.
31) Benson, S. W. Thermochemical Kinetics, 2nd ed.; Wiley: New York,
976.
32) We are grateful to Professor D. W. Rogers, Long Island University,
for the communication of this preliminary result.
(
(
(
(33) Derived from experimental heats of formation of cyclohexane (-29.5
kcal/mol), 1-methylcyclohexane (-37.0 kcal/mol), and 1.1-dimethylcyclo-
26
(
hexane (-43.3 kcal/mol); in good agreement with MMEVBH calculated
17
value of -51.0 kcal/mol and Benson uncorrected value of -51.7 kcal/
(
mol.31
1
(34) See Scheme 6, page 1075 of ref 6.
(35) Doering, W. von E.; Belfield, K. D.; He, J.-n. J. Am. Chem. Soc.
1993, 115, 5414-5421.
(

+
+
6
664 J. Am. Chem. Soc., Vol. 118, No. 28, 1996
Von Eggers Doering et al.
Scheme 4
Experimental Section
General Methods. Unless explicitly noted, H and 13C NMR spectra
1
are measured in C
Spin-lattice relaxation times (T
recovery method with use of vacuum-sealed solutions in C
6
D
6
solution by a Bruker AM-500 spectrometer.
1
) are determined by the inversion-
6
6
D .
Chemical shifts are reported in parts per million from TMS (δ). Infrared
spectra are recorded on a Nicolet Impact 400D FT IR instrument and
-1
reported in cm . Liquid samples are observed as thin films on a NaCl
plate; solid samples are measured as thin layers prepared by evaporating
CHCl solutions on a NaCl plate. UV-visible electronic spectra are
3
determined with a Varian Cary 1E UV-visible spectrophotometer and
are reported as λmax in nm and coefficients of extinction (ꢀ). Optical
rotations are measured on a Perkin-Elmer Polarimeter 241; enantiomeric
excess (ee) is determined with a Hewlett Packard 5890 Series II gas
Chromatograph using a Chiral-dex G-TA column (20 m × 0.25 mm
ID; Advanced Separation Technology, Inc.). High resolution mass
spectra are measured on a Jeol AX 505 spectrometer equipped with a
data recovery system.
to-head structure.36,37 On being warmed to 0 °C, 9 reverts
rapidly and completely to the monomer. By monitoring the
disappearance of 9 at -31.6 °C, an approximate specific rate
-
5
-1
constant, k ) 4.4 × 10 s , can be estimated (Scheme 4).
From this quantity and the assumption of a log A factor (14.47)
identical to that determined for cyclobutane 10, an estimated
activation energy of ∼21 kcal/mol can be calcualted. An
informative comparison of 9 with the Ekmanis dimer 11 reveals
a decrease of ∼11 kcal/mol in enthalpy of activation resulting
from the formal change from two allyl radicals to the two
cyclohexadienyl radicals assumed to approximate the transition
(R)-(-)-4,4a,5,6,7,8-Hexahydro-4a-methyl-2(3H)-naphthale-
none (4). This compound is synthesized by the method of Revial and
Pfau40 from 2-methylcyclohexanone (23.0 g, 0.205 mol), methyl vinyl
ketone (15.1 g, 0.216 mol), and (S)-(-)-R-methylbenzylamine (25.0
g, 0.206 mol) in >99% of purity and 55% of theoretical yield: 18.5 g;
2
0
1
[
(
1
R]
s, 1 H), 2.50-1.34 (m, 12 H), 1.23 (s, 3H); IR 2927, 2859, 1677,
627, 1448, 1431, 1326, 1261, 1225, 1186, 858.
R)-(+)-5,6,7,8-Tetrahydro-4a-methyl-2(3H)-naphthalenone (5).
D
) -228° (c ) 1.20, ethanol); ee >99%; H NMR (CDCl
3
) 5.72
(
3
8
state. A similar comparison of 10 with 11 reveals a lowering
of ∼7 kcal/mol.
A solution of 3.00 g (18.3 mmol) of ketone (R)-4 and 5.00 g (22.9
mmol) of dichlorodicyanoquinone (DDQ) in 200 mL of anhydrous
dioxane is boiled under nitrogen and reflux for 24 h. After filtration
and removal of dioxane in Vacuo, the residual oil is diluted with 200
mL of ethyl ether, washed with 5% aqueous NaOH (2 × 30 mL),
Conclusions
The experimental values of the activation parameters for the
thermal cis-trans isomerizations of 1 and 2 agree very closely
with each other. Translation into an estimation of the enthalpy
of stabilization of the cyclohexadienyl radical is presently
hampered by the absence of apposite thermochemical informa-
tion about crossed trienes and the bis-homofulvalene type of
saturated NaHCO
3
(2 × 30 mL), water, and dried over Na
2 4
SO . Solvent
is removed in Vacuo leaving a yellow oil (3.00 g), which GLC shows
41
to be a mixture of the desired ketone 5 and 3,4,5,6-tetrahydro-4a-
methyl-2(3H)-naphthalenone7d in a ratio of 4:1. Chromatography on
silica gel on elution with mixtures of hexane-ethyl acetate in ratios
varying from 20:1 to 5:1 affords 1.66 g (56% of theoretical yield) of
20
(
R)-5 as a colorless oil of 98% purity: [R]
D
) +61.9° (c ) 1.25,
ethanol); ee > 99% (determined by GLC); H NMR (CDCl ) 6.75 (d,
H, J ) 9.9 Hz, H-4), 6.18 (dd, 1 H, J ) 9.9 Hz, J ′ ) 1.7 Hz, H-3),
.07 (s, 1 H, H-1), 2.46-2.40 (m, 1H), 2.01-1.98 (m, 1 H), 1.83-
q
q
pentaene. Nonetheless, their means∆H ) 36.8 kcal/mol, ∆S
+1.8 eusmay be compared with the parameters for the triene
1
3
)
7
1
6
q
q
s∆H ) 40.1 kcal/mol, ∆S ) -4.3 eusby, first, correcting
the former for extra enthalpy of conjugation in the eductstaken
arbitrarily to be -7.5 kcal/mol (two Kistiakowsky units)sand,
second, correcting for a difference between the two in the
contribution made by relief of steric energy -2.8 kcal/mol. The
resulting corrected difference in enthalpies of activation is 12.8
kcal/mol, corresponding to an enthalpy of stabilization of the
RSE1,4 type of 20.0 kcal/mol for the cyclohexadienyl radical.
The discrepancy between this value and many of those in the
literature would be ameliorated by an augmentation of conjuga-
tive interaction in crossed trienes. At the moment, two pieces
of evidence, but no high level quantum mechanical calculations,
point suggestively in that direction.39
1.68 (m, 3 H), 1.35-1.28 (m, 2 H); IR 2936, 2860, 1663, 1627, 1605,
1444, 1395, 1397, 1268, 942, 880, 815.
Cholesta-1,4-dien-3-one (3). This ketone is prepared from 2,4-
dibromocholestan-3-one by a procedure slightly modified from that of
9a
Pinder and Robinson in turn modified from that of Wilds and
Djerassi9b [7.2 g, mp 189.5-190 °C (lit.9a mp 193 °C)] by boiling in
freshly distilled collidine (31 mL) for 80 min. The cooled solution is
filtered from collidine hydrobromide (4.9 g, 91.8% theoretical yield),
which is washed with 5 20-mL portions of ether. The filtrate is
concentrated in Vacuo to a dark residue, which is boiled in methanol
for 15 min. The cooled solution is filtered and concentrated in Vacuo
to a residue (2.7 g, brown oil), which is chromatographed on silicagel
(elution with 5% ether-hexane) to give colorless, crystalline ketone 3:
2
.5 g (49% of theoretical yield): mp 110-110.5 °C [lit. mp 111-
9a 9b 1
(
36) The behavior of this methylenecyclohexadiene is quite normal on
111.5 °C ; 110-112 °C ]; H NMR (400 MHz) 6.47 (d, 1 H, J )
sensitized irradiation and only under conditions of rapid thermal cyclo-
reversion of its corresponding cyclobutane is the vinylog of the di-π-methane
rearrangement realized in an uncharacteristically low quantum yield
1
1
0.2 Hz), 6.32 (dd, 1 H, J ) 10.2 Hz, J ′ ) 1.7 Hz), 6.19 (s, 1 H),
.89-1.70 (m, 4 H), 1.60-1.40 (m, 5 H), 1.41-1.10 (m, 9 H), 1.0-
3
7
0.9 (m, 5 H), 0.77-0.60 (m, 4 H), 0.94 (d, 3 H, J ) 1.2 Hz), 0.92 (d,
3 H, J ) 1.2 Hz), 0.77 (s, 3 H), 0.58 (s, 3 H); IR 2938, 2868, 2853,
1666, 1627, 1603, 1466, 1441, 1402, 1240, 929, 886.
Reductive Coupling. Dienones 3 and 5 are converted to the
corresponding ylidenes 1 and 2, respectively, by the Mukaiyama-Tyrlik-
(
0.003)!
(37) Zimmermann, H. E.; Hackett, P.; Juers, D. F.; McCall, J. M.;
Schr o¨ der, B. J. Am. Chem. Soc. 1971, 93, 3653-3662; Zimmermann, H.
E.; Juers, D. F.; McCall, J. M.; Schr o¨ der, B. J. Am. Chem. Soc. 1971, 93,
3
662-3674.
38) Ekmanis, J. L., Ph.D. Dissertation, Harvard University, Cambridge,
MA, 1976. Diss. Abstr. Int. B 1976, 37, 224B (Order No. 76-14,401).
39) Suggestions for further clarification include application of the
(
8
42
McMurry reaction following published procedure. Activated zinc
dust and dry pyridine are added slowly to a solution of TiCl in THF,
(
4
30b
calorimetric method of Bartmess and Griffith to methylenecyclohexadiene
p-isotoluene); determination of heats of hydrogenation of further model
compounds such as the mono- and di-olefins related to 3-methylene-6,6-
dimethylcyclohexa-1,4-diene; establishment of the temperature dependence
of the equilibrium constants among the several isomers of a triene such as
(
(40) Revial, G.; Pfau, M. Organic Synthesis 1992, 70, 35-45.
(41) (a) Naya, K.; Okayama, T.; Fujiwara, M.; Nakata, M.; Ohtsuka, T.;
Kurio, S. Bull. Chem. Soc. Jpn. 1990, 63, 2239-2245. (b) Alunni, S.;
Minuti, L.; Taticchi, A. J. Org. Chem. 1991, 56, 5353-5356.
(42) Lenoir, D. Synthesis 1977, 553-554.
4
,4-dimethylcyclohexa-1,4-dienylidene-4′,4′-dimethylcyclohexane.

+
+
Pentaenes of the Bis-HomofulValene Type
freshly distilled from LiAlH , under nitrogen at 0 °C with stirring. This
titanium reagent is then treated with the solution of the dienone in THF
and stirred at 0 °C for 30 min, after which starting material is shown
by TLC (silica gel, eluted with hexane) to be completely reacted.
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6665
4
ppm are assigned to anti-1. The correctness of this assignment is
supported by the striking similarity of the spectra of 1 and 2 in the
vinyl regions.
For quantitative analysis, the H-1 singlets at 6.67 ppm (anti-1) and
6.74 (syn-1) are used. Recovery is determined from the ratio of the
sum of these two signals to that of the internal standard (18-crown-6;
3.52 ppm).
3 2 2 4
Quenching with NaHCO , extracting with CH Cl , drying over MgSO ,
filtering through a short silica gel column, and removing solvent in
Vacuo affords the product as a mixture of syn and anti isomers.
Assignments of structure are based on nuclear Overhauser enhance-
ments determined by the gated decoupling method by using degassed
Kinetic Measurements. Kinetics are determined essentially as
described before.7 Samples in sealed NMR tubes are heated in the
1
solutions in C
pulses five times the longest T
R,R)-5,5′,6,6′,7,7′,8,8′-Octahydro-4a,4a′-dimethyl-2,2′-bi-(3H)-
D
6 6
and a relaxation delay and a saturation period between
vapors of solvents of appropriate boiling point. Analysis is by H NMR
1
’s of the relevant protons.
employing in both instances the singlet olefinic hydrogen atoms, as
noted above. In both instances, recovery remains satisfactory until the
longest reaction times, as may be seen in greater detail in Tables 1S
and 2S, which contain the experimental data for all runs of both 1 and
2, and are available as supporting information.
(
naphthylidene (2). Application of the Mukaiyama-Tyrlik-McMurry
reaction to (R)-(+)-5,6,7,8-tetrahydro-4a-methyl-2(3H)-naphthalenone
4
[(R)-5] (0.63 g, 4 mmol in 10 mL of THF), TiCl (6.0 mL, 55 mmol),
zinc (7.02 g, 110 mmol), and pyridine (3.6 mL) in 150 mL of THF
affords, after flash chromatography on silica gel (hexane), 0.53 g (91%
theoretical yield) of 2 as a 50:50 mixture of syn and anti isomers. Four
recrystallizations from pentane afford (R)-anti-2 (Vide infra), 98% of
By the time equilibrium is reached, recovery in the case of 1 has
dropped as low as 34%. In probing for possible means of suppressing
the factors responsible for the low recovery, several experiments were
carried out with variations on a standard set of conditions: T, 196 °C;
2
0
1
purity: mp 142 °C; [R]
dd, 2 H, J ) 10 Hz, J ′ ) 1.9 Hz, H-3, H-3′), 6.55 (s, 2 H, H-1, H-1′),
.55 (d, 2 H, J ) 10 Hz, H-4, H-4′), 2.32-2.25 (m, 2 H), 2.04-2.01
m, 2 H), 1.69-1.64 (m, 2 H), 1.50-1.41 (m, 8 H), 1.28-1.23 (m, 2
H), 1.12 (s, 6 H); IR 3040, 2970, 2923, 2849, 1657, 1549, 1464, 1439,
D
) -148° (c ) 0.11, hexane); H NMR 6.79
time, 3.0 h; internal standard: ∼0.05% 18-crown-6; solvent: C
6 6
D ;
(
5
(
reactant: ∼0.1% syn isomer. The following recoveries were ob-
tained: no additive, 77%; 9,10-dihydroanthracene (∼90 mol%), 91%;
4,4′-thio-bis-(6-tert-butyl-3-methylphenol) (∼30 mol%), 84%; tri-
phenylmethane (∼40 mol%), 80%; benzoylperoxide (∼135 mol%), 0%.
No solution to the problem was found.
4
4
1
175, 974, 858, 767; UV-vis: 330 (5.26 × 10 ), 346.5 (5.14 × 10 );
m/z calcd for C22 28, 292.2191; found, 292.2193.
H
In the case of 2, recovery was not as bad but still not good, varying
from 76-88% by the time equilibrium had been reached. In calculating
specific rate constants and Arrhenius parameters, it made no significant
difference, however, whether all data were used or the last couple of
points were omitted.
Partial assignment of NMR frequences to the three hydrogen atoms
in the vinyl region of the other isomer (syn-2, Vide infra) can easily be
1
made by subtraction: H NMR 6.71 (dd, 2 H, J ) 10.2 Hz, J ′ ) 1.4
Hz, H-3, H-3′), 6.61 (s, 2 H, H-1, H-1′), 5.52 (d, 2 H, J ) 10.2 Hz,
H-4, H-4′), 1.14 (s, 6 H).
Photodimerization of 1-Methylene-4,4-dimethylcyclohexa-2,5-
diene. The starting triene is prepared following a published procedure:
Distinction between the stereoisomeric structures relies on nuclear
Overhauser enhancement: irradiation at 6.55 ppm shows an enhance-
ment at 6.79 ppm of 30%, while irradiation at 6.61 ppm shows no
enhancement at 6.71 ppm. Because symmetry in the syn isomer places
two identical vinyl hydrogen atoms in proximity, no enhancement
between hydrogen atoms H-1 and H-3′ (and Vice Versa) can be seen.
Consequently, the doublet at 6.71 ppm and the singlet at 6.61 ppm are
assigned to syn-2, while the doublet, 6.79 ppm, and the singlet, 6.55
ppm, are assigned to anti-2.
3
7 1
H NMR (toluene-d
H, J ) 9.9 Hz), 4.83 (s, 2 H), 0.95 (s, 6 H). Irradiations are carried
out in toluene-d at -75 °C in Pyrex NMR tubes that had been degassed
8
, 500MHz) 6.13 (d, 2 H, J ) 9.9 Hz), 5.49 (d,
2
8
and sealed. Analysis was directly by NMR spectroscopy as noted
above. After irradiation for 2 h, no change was observed. In the
presence of benzophenone as triplet sensitizer [10 mol% (2 h) and 30
mol% (8 h)], ∼5% and ∼37%, respectively, of peaks corresponding
to a new compound 9 are seen. When either tube is warmed to room
temperature for 5 min, peaks only of the monomer remain. Assignment
For quantitative analysis, the H-1 singlets at 6.55 ppm (anti-2) and
of a symmetrical cyclobutane structure is based on NMR spectrum:
6
.61 (syn-2) are used. Recovery is determined from the ratio of the
1
H NMR (toluene-d , 500MHz) 5.70 (d, 4 H, J ) 10.1 Hz), 5.55 (d, 4
8
sum of the H-4 doublets at 5.55 ppm (anti-2) and 5.52 (syn-2) to that
of the internal standard (18-crown-6; 3.52 ppm).
H, J ) 10.1 Hz), 1.95 (s, 4 H), 1.06 (s, 6 H), 1.01 (s, 6 H). Kinetics
are followed at -31.6 °C by NMR at that temperature with the results
shown in Table 4S.
3,3′-Bicholesta-1,4-dienylidene (1). This compound is obtained by
the procedure above from cholesta-1,4-dien-3-one (3) [0.310 g (0.8
mmol) in 5 mL of THF] and titanium reagent prepared from 1.2 mL
of TiCl
4
(11 mmol), 1.4 g of Zn (22 mmol) and 1 mL of pyridine in 30
mL THF at 0 °C: 0.250 g (92% theoretical yield) of light yellow
crystals consisting of a 50:50 mixture of syn and anti isomers. Three
recrystallizations from pentane afford the less soluble isomer (syn-1,
Vide infra) in >99% of purity: mp 231-232 °C: 1H NMR 6.84 (dd,
Acknowledgment. We express great gratitude to Professor
Wolfgang R. Roth for many constructively critical discussions
and his generous assistance with the molecular mechanical
calculations. This investigation has been supported by the
National Science Foundation, Grants CHE-88 16186, 91 23207,
and 94 11248. We thank the Norman Fund in Organic
Chemistry, Harvard University, in memory of Ruth Alice
Norman Weil Halsband, for its partial support of this work.
Funding of the Harvard University Mass Spectroscopy Labora-
tory by the National Institutes of Health (RR06716) and the
National Science Foundation (CHE 90 20043) is acknowledged.
2
5
)
H, J ) 10.2 Hz, J ′ ) 1.4 Hz, H-2, H-2′), 6.74 (s, 2 H, H-4, H-4′),
.98 (d, 2 H, J ) 10.2 Hz, H-1, H-1′), 2.41 (dt, 2 H, J ) 13.2 Hz, J ′
3.9 Hz), 2.14, (d, 2 H, J ) 12.8 Hz), 2.01 (d, 2 H, J ) 8 Hz),
1
1
1
.88-1.77 (m, 2 H), 1.73 (d, 2 H, J ) 11.3 Hz), 1.63 (dd, 2 H, J )
3.1 Hz, J ′ ) 2.6 Hz), 1.59-1.49 (m, 4 H), 1.49-1.33 (m, 10 H),
.32-1.15 (m, 10 H), 1.14 (s, 6 H, H-19, H-19′), 1.13-0.99 (m, 10
H), 1.00 (d, 6 H, J ) 6.3 Hz, H-21, H-21′), 0.96-0.90 (m, 14 H), 0.71
4
4
(s, 6 H, H-18, H-18′); UV-vis 352 (4.18 × 10 ), 335 (3.83 × 10 ).
Supporting Information Available: Tables 1S and 2S
contain the unprocessed experimental data from which Tables
Although the more soluble isomer (anti-1, Vide infra) has not been
enriched beyond 60:40, a partial assignment of NMR frequences can
be made cleanly for three hydrogen atoms in the vinyl region (but for
1
and 2 are derived, respectively, Table 3S contains a com-
1
parison of heats of formation and steric energies calculated for
compounds 2, 6, 7, and 8 by MM2, MMERW, and MMEVBH
methods, and Table 4S contains the data for the kinetics of the
8
none in the allylic and aliphatic region): H NMR 6.92 (dd, 2 H, J )
1
0.2 Hz, J ′ ) 1.3 Hz, H-2, H-2′), 6.67 (s, 2 H, H-4, H-4′), 5.92 (d, 2
H, J ) 10.2 Hz, H-1, H-1′).
Syn and anti isomers are distinguished by nuclear Overhauser
enhancement of vinylic proton resonances at 322 K (400 Mz).
Irradiation at 6.63 ppm shows an enhancement at 6.88 ppm of +16%,
while irradiation at 6.69 ppm shows no enhancement at 6.82 ppm.
Protons resonating at 6.84 and 6.74 ppm (300 K; 500 Mz) are
consequently assigned to syn-1, while those resonating at 6.92 and 6.67
thermal cycloreversion of compound 9 in toluene-d at -31.6
°C to 1-methylene-4,4-dimethylcyclohexa-2,5-diene (11 pages).
See any current masthead page for ordering and internet access
instructions.
JA953416Z