Ortho-Substituted Diaryl Ethylene and Diaryl Ketone
obtained, nor did any line (both at the 1H and 13C
frequencies) display additional broadening and decoa-
lescence when the sample was further cooled from -152
to -170 °C. The absence of minor signals suggests that
only one conformer is essentially populated and the
observation of a single 13C line for each pair of carbons
at any temperature seems to indicate that its structure
cannot be asymmetric as that of C1.
SCHEME 5. Stereomutation Pathways Computed
for 2
Ab initio calculations actually show that the C2-syn
conformer of 2 is the most stable species, with C1 and
C2-anti having much higher energies (1.6 and 5.3 kcal
mol-1, respectively). The 6.35 kcal mol-1 value obtained
from the analysis of the methyl signal of 2 thus corre-
sponds to the interconversion barrier of the two enantio-
mers of C2-syn (Scheme 1). This process, as mentioned,
requires the passage through the C1 conformer but, being
the energy of the latter quite high in 2, its proportion is
too low to be experimentally detected (the computed
relative energy of 1.6 kcal mol-1 entails in fact a popula-
tion as low as 0.1% at -152 °C, and, of course, the
population expected for C2-anti is even lower).
It seems quite surprising that the measured enantio-
merization barrier, which interchanges the methyl sig-
nals of 1 (6.45 kcal mol-1), is essentially equal to that
observed for the same type of exchange in 2 (6.35 kcal
mol-1): in both cases, in fact, a passage through the
transition state TS-A (Scheme 2) is required, which seems
more facile for the less hindered ketone 2 than for the
ethylene 1. Indeed this barrier was computed to be much
lower (3.3 kcal mol-1) for 2 than for 1 (8.0 kcal mol-1, as
in Table 1). This suggests that the barrier measured in
2 does not correspond, as in 1, to the transition state TS-A
but reflects another type of transition state.
The pathway for interconverting the C2 enantiomers
is the same, as mentioned, as that required to obtain the
C1 conformer (invisible in the case of 2 because it has
too low a population): the latter conformer, however,
must itself enantiomerize to bring about the overall
enantiomerization of C2-syn. Whereas in 1 this process,
which takes place through the transition state TS-C
(Scheme 3), is computed to have a lower energy than that
of TS-A of Scheme 2 (5.4 vs 8.0 kcal mol-1, as in Table
1), the opposite occurs in the case of 2. Calculations
predict, in fact, a barrier of 9.4 kcal mol-1 for passage
through the transition state TS-C, i.e., a value much
larger than the 3.3 kcal mol-1 for the passage through
the transition state TS-A. The computed interconversion
pathways for 2 are displayed in Scheme 5.
larger than that of the former (10.3 vs 9.4 kcal mol-1 for
TS-D and TS-C, respectively) the stereomutation is
expected to occur via the more facile pathway comprising
the TS-C transition state. Contrary to the case of 1,
however, the difference between these two computed bar-
riers (0.9 kcal mol-1) is quite small and, for this reason,
the choice of the preferred interconversion pathway
proposed for derivative 2 cannot be considered completely
unambiguous, since this conclusion might be affected by
the approximations involved in the computational ap-
proach.
Conclusions
Derivatives comprising two equal aryl rings bonded to
an sp2 carbon give rise, in principle, to three types of
conformers if each ring bears a single substitutent in the
ortho position (in the present example an isopropyl
group). In the case of the ethylenic derivative 1, experi-
mental evidence for the existence of such conformers has
been obtained and the corresponding interconversion
barriers determined. These results are in agreement with
the ab initio calculations describing their stereomutation
pathways. In the analogous ketone 2 only one conformer
was experimentally observed because the relative ener-
gies of the other two are computed to be exceedingly high.
Whereas in 1 the most stable conformer is asymmetric
(C1 point group, as confirmed also by X-ray diffraction),
that of 2 has C2 symmetry, in agreement with the
theoretical prediction.
The measured barrier of 6.35 kcal mol-1 in 2 corre-
sponds, therefore, to the passage through the TS-C
transition state, which has an energy comparably high
to that (8.0 kcal mol-1) computed for the TS-A transition
state of 1. This might account for the similarity of the
two measured barriers: 6.45 (experimental) corresponds
in fact to 8.0 kcal mol-1 (computed) in the case of 1 and
6.35 (experimental) corresponds to 9.4 kcal mol-1 (com-
puted) in the case of 2.
It should be pointed out that the interconversion of the
enantiomers of C2-syn, which has been indicated to take
place (full line of Scheme 5) via the transition states
TS-A, TS-C, and TS-A′, might alternatively occur via the
transition states TS-A, TS-B, and TS-D′, as represented
by the dashed lines of Scheme 5. Since the rate deter-
mining step of this second process involves a barrier
Experimental Section
Materials: Bis(2-isopropylphenyl)methanone (2). To a
cooled (-78 °C) solution of 2-isopropylbenzene lithium, pre-
pared by addition of butyl lithium (14 mmol) to a solution of
1-bromo-2-isopropylbenzene (2.98 g, 15 mmol in 50 mL of THF)
was added a solution of 2-isopropylbenzaldeyde (2.0 g, 13.5
mmol in 10 mL of THF). After 30 min the solution was allowed
to warm and quenched with aqueous NH4Cl. The product was
extracted with Et2O and dried (Na2SO4) and the solvent was
removed at reduced pressure. The crude was washed with Et2O
to give 3.25 g (90%) of pure bis(2-isopropylphenyl)methanol.
Oxidation with pyiridinium chlorochromate (4.70 g, 18.2 mmol
in 50 mL of CH2Cl2) at room temperature, followed by filtration
on silica, yielded 3.50 g (88%) of bis(2-isopropylphenyl)-
methanone, which yielded 3.20 g of the product pure (purified
by chromatography on silica gel, petroleum ether/Et2O 10/1).
1
Mp 38.2-39.0 °C. H NMR (400 MHz, CDCl3, 22 °C, TMS) δ
J. Org. Chem, Vol. 70, No. 2, 2005 461