Angewandte
Chemie
stirring. This partial orientation induces an
artificial contribution in the circular dichroism
spectra of both chiral and achiral assemblies.
The shape, sign, and intensity of the CD
spectrum depend on the type of flow, and two
different CD effects are observed; one is due to
convective flow and is based on linear dichro-
ism, and the other is due to vortex flow patterns
and is based on linear dichroism and linear
birefringence. It is clear that the alignment of
supramolecular assemblies in dilute solutions
can create artifacts in the chiroptical properties
of the system, which should be clearly differ-
entiated from the effects originating from
chirality at the molecular length scale. To
reduce the flow, we are currently designing a cuvette with
improved heat transfer to decrease the temperature gradient.
The flow-induced CD, together with hydrodynamic modeling,
will be used to study size and shape of the assemblies.
Figure 4. S-OPV4 dodecane solution (1 cm cuvette, c=310À5 m). a) CD spectra at
293 K without stirring (c) and with clockwise (a) and counterclockwise (g)
stirring (solution obtained by cooling from 363 to 293 K with 60 KhÀ1 without stirring)
b) Temperature-dependent CD measurements at l=420 nm (ÀdT/dt=60 KhÀ1) with-
~
*
out stirring ( ) and with clockwise ( ) and counterclockwise (&) stirring.
the two types of artifacts discussed above and corresponds to
the spectra reported earlier.[1,2] Furthermore, shaking does
not affect the CD spectra.[2] Under stationary conditions, no
LD effects are present. A CD cooling curve, similar to that
reported previously, is shown in Figure 4b; it displays an
elongation temperature Te of 339 K (c = 2.4 10 À5 m).[1] Mon-
itoring the cooling of the S-OPV4 solution with linear
dichroism shows an increase in LD[2] that starts at 334 K,
that is, below the elongation temperature of the self-assembly.
Since the size of the S-OPV4 assemblies is much smaller than
that for A-OPV3, the S-OPV4 fibers are less susceptible to
the flow patterns created in the cuvette; hence, randomization
of the fibers is much faster.[1] Although LD is present during
cooling, the CD response at 293 K is only marginally affected
(typically by 4%). Previously, we suggested on the basis of
AFM data that S-OPV4 forms bundles of fibers at high
concentrations and ascribed the additional increase in CD
intensity during cooling to the lateral interactions of the
fibers.[1] When this S-OPV4 solution (c = 2.4 10 À4 m) is
monitored by LD, an increase in intensity occurs at the
same temperature as the additional increase in CD. This result
suggests that the deviation from the cooperative self-assembly
model is probably caused by the formation of bundled fibers
that are aligned by the convective flow, as visualized with
linear dichroism.[2] Hence, we conclude that for S-OPV4,
convective flows can also give rise to deviations of the optical
spectra; the magnitude is so low that all published data are
only marginally influenced by this phenomenon.
Stirring of the S-OPV4 solutions, on the other hand, again
has a significant influence on the CD effect, the sign of which
depends on the stirring direction (Figure 4a). The absorption
is not affected by the vortex flow.[2] The CD intensity upon
stirring is less than for A-OPV3, which is probably a result of
the different sizes of the assemblies. Furthermore, the CD
intensity was also monitored at l = 420 nm during cooling,
while the S-OPV4 solution was either not stirred or stirred
continuously in a clockwise or counterclockwise direction
(Figure 4b). The effect of stirring becomes visible 5 K below
the elongation temperature. This behavior again suggests that
a certain stack length is needed for the systems to be
influenced by stirring and that chirality at the molecular level
seems unaffected by the stirring.
Received: July 10, 2007
Published online: September 20, 2007
Keywords: alignment · circular dichroism · convective flow ·
.
self-assembly · vortex flow
[1] P. Jonkheijm, P. van der Schoot, A. P. H. J. Schenning, E. W.
[2] See the Supporting Information.
[3] J. R. Welty, C. E. Wicks, R. E. Wilson, G. Rorrer, in Fundamen-
tals of Momentum, Heat, and Mass Transfer, 4th ed., Wiley, New
York, 2001.
[4] C. R. L. P. N. Jeukens, P. Jonkheijm, F. J. P. Wijnen, J. C. Gielen,
P. C. M. Christianen, A. P. H. J. Schenning, E. W. Meijer, J. C. J.
[5] a) B. Norden, Acta Chem. Scand. 1973, 27, 4021 – 4024; b) Y.
[6] A. Tsuda, M. A. Alam, T. Harada, T Yamaguchi, N. Ishii, T.
Aida, Angew. Chem. 2007, 119, 8346 – 8350; Angew. Chem. Int.
Ed. 2007, 46, 8198 – 8202.
[7] After quickly cooling the same solution by quenching in an ice
bath, stirring showed a less intense bisignate CD effect similar in
sign to the slowly cooled solution. See the Supporting Information.
[10] In dye-doped cholesteric LCs, usually monosignate CD effects
arise. In this case, the LB component originates from the host
liquid crystal, which usually does not show a pronounced
wavelength dependence in the region where the dye absorbed,
because the LC is transparent in this region. See: S. F. Mason,
[11] The synthesis of the compound has been reported elsewhere: P.
Jonkheijm, F. J. M. Hoeben, R. Kleppinger, J. van Herrikhuyzen,
[12] The LD contribution in the self-assembly of S-OPV3 (the exact
chiral analogue of A-OPV3 with S-methylbutyloxy side chains)
is negligible. See the Supporting Information.
In conclusion, we observed that supramolecular assem-
blies can be partially oriented by convective flow, shaking, or
Angew. Chem. Int. Ed. 2007, 46, 8203 –8205
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim