observed for the cis and trans isomers of photoresponsive
azobenzene containing polymers.8 The differences in size
were experimentally apparent from gel permeation chroma-
tography (GPC) (Figure 1). Despite the identical molecular
Scheme 3a
Figure 1. GPC traces of anti-2 and syn-2. Conditions: Waters
Styragel HMW 2 column at 1 mL/min, in chloroform.
weight of the isomers, GPC was able to differentiate syn-
and anti-2, yielding estimated molecular weights (Mn) of
2665 and 2457, respectively, on the basis of polystyrene
standards. The unfolded anti isomer eluted first, demonstrat-
ing that it had a larger hydrodynamic volume and giving
additional support for the initial dipole-based assignments.
In addition, the GPC measured molecular weights of both
syn-2 and anti-2 were inflated in comparison to their absolute
molecular weights (2286), which is consistent with what has
been observed for other shape-persistent systems.9
a (a) 90% HNO3, 23 °C (49%). (b) H2, 10% Pd/C, EtOH (92%).
(c) 0.5 equiv of 1,4,5,8-naphthalenetetracarboxylic dianhydride,
CH2Cl2, 23 °C then neat, 150 °C in vacuo, 3 days (44%). (d) CH2Cl2
23 °C, then neat 170 °C in vacuo, 12 h (36%). (e) H2, 10% Pd/C,
EtOH (97%). (f) phthalic anhydride, DMF, reflux 12 h (59%). (g)
6 equiv of 1,4,5,8-naphthalenetetracarboxylic dianhydride, DMF,
reflux.
The unique aspect of octaimide 2 is its dynamic shape-
persistent and shape-adaptable qualities. Control of the anti/
syn equilibria at elevated temperatures would enable the
oligomers to be “molded” into one shape or the other and
then to hold that shape. A possible strategy was suggested
from studies of the model compound 1 of using differences
in dipole moment of the conformers to control morphology.10
These differences should be expressed to a different extent
depending upon the dielectric constant of the medium. To
test this strategy, oligomer 2 was equilibrated in a range of
different solvents, by heating for greater than 10 half-lives.
On cooling to room temperature, the anti/syn equilibria ratios
were preserved as a result of the re-establishment of restricted
rotation even when removed from the promoting environ-
ment. The anti/syn ratios were then measured by integration
of the central naphthalene proton by 1H NMR in CDCl3. The
heating experiments demonstrated that in more polar solvents
such as acetonitrile the differences in dipole are minimized,
and the anti/syn ratio approaches 1:1. In nonpolar solvents
at lower temperatures (140 °C) no reaction was observed.
Reaction of the first amine of 8 apparently reduces the
nucleophilicity of the second such that oligomerization is
suppressed, eliminating the need for protecting groups. The
final condensation reaction between diamine 9 and monoan-
hydride 10 was carried out under more forcing conditions
(170 °C, neat, in vacuo) because of lower reactivity of the
amines in 9, and the product, octaimide 2, was isolated in
36% yield.
Like model system 1, oligoimide 2 was isolated as a
mixture of stable syn and anti conformers. The syn and anti
conformers were separated by preparative TLC and assigned
on the basis of polarity arguments and also by differences
in size and shape (vide infra). The rotational barrier of 2
(26.5 kcal/mol) was similar to that of model system 1 (27
kcal/mol) and was calculated from the rate of reequilibration
of the pure syn and anti isomers in acetonitrile at 80 °C.
Molecular models predicted that the two conformational
isomers would have drastically different shapes. The rigid
framework coupled with the limited conformational freedom
enabled an accurate measure of the dimensions of syn- and
anti-2. The syn isomer has a curved compact shape; whereas,
the anti isomer has a larger extended shape (Scheme 1).
Similar differences in hydrodynamic volume have been
(8) (a) Junge, D. M.; McGrath, D. V. J. Am. Chem. Soc. 1999, 121,
4912-4913. (b) Izumi, A.; Teraguchi, M.; Nomura, R.; Masuda, T.
Macromolecules 2000, 33, 5347-5352. (c) Beattie, M. S.; Jackson, C.;
Jaycox, G. D. Polymer 1998, 39, 2597-2605.
(9) (a) Rader, H. J.; Spickermann, J.; Mullen, K. Macromol. Chem. Phys.
1995, 196, 3967-3978. (b) Rader, H. J.; Spickermann, J.; Kreyenschmidt,
M.; Mullen, K. Macromol. Chem. Phys. 1996, 197, 3285-3296.
(10) Dipole moments of anti- and syn-2 were calculated as 0.66 and
6.05 D using MacSpartan (semiemperical, AM1).
Org. Lett., Vol. 3, No. 23, 2001
3759