DISCOTIC LIQUID CRYSTALS OF TRANSITION METAL COMPLEXES 52 1451
that perfect homeotropic alignment was formed both
as illustrated in Fig. 1c. At the first time in 2011, we
adopted a new monodomain method, together with
the conventional polydomain method [15–18, 20–22],
illustrated in Figs 6a and 6b, respectively, and thoroughly
established this peak H as a helical pitch in Z-axis
direction [23, 30–32] from these two different methods.
Also by using these methods, we could establish here the
helical structure of C60 moieties around the Pc columns
for the present dyad (C14S)6PcCu-VAN-C60 (3a).
Figure 6 also illustrates the X-ray diffraction patterns
of (C14S)6PcCu-VAN-C60 (3a) by using these two
methods [A] and [B]. The red X-ray diffraction pattern
in this figure was measured by Polydomain Method [A].
As can be seen from this red pattern, both Peak H and
(1 0 0) reflection could be clearly observed. The black
and blue X-ray diffraction patterns in this figure were
measured by Method [B]. The black diffraction pattern
was measured for only two cover glass plates without
the sample. It gave no peak. The blue X-ray diffraction
pattern was measured for the (C14S)6PcCu-VAN-C60 (3a)
dyad sandwiched between two cover glass plates. As can
be seen from this blue diffraction pattern, it gave (1 0 0)
reflection but Peak H disappeared. This means that Peak
H is the periodicity in Z-direction.
the Colho1 and Colho2 mesophases. When the film in the
dark field was scratched with a spatula, the scratched
part became birefringent. In the bottom of Fig. 5a are
illustrated the hoemotropic alignment models. When
disk-like molecules face-to-face stack one after another
to form columns perpendicular to the glass substrate,
homeotropic alignment is achieved to show the dark field
under crossed Nicols. When the columns perpendicular
to the glass substrate is disturbed by scratching with a
spatula, the disturbed area may become birefringent.
Thus, we could certify the homeotropic alignment of the
(C14S)6PcCu-VAN-C60 (3a) dyad, as can be seen from
the photomicrographs in Fig. 5a.
As can be seen also from Table 1, the (C14S)6PcCu-
OPh-C60 (3b) dyad having no methoxy group at the
phenoxy group also showed two different hexagonal
columnar mesophases, Colho1 and Colho2. When the
Colho1 mesophase was heated from rt, it transformed into
to Colho2 mesophase at 96.5°C. On further heating, the
Colho2 mesophase cleared into I.L. at 176.7°C. When the
I.L. was cooled down to 120°C, a sanded texture appeared
for the Colho2 mesophase, as can be seen from Fig. 5b;
on further cooling into the Colho1 mesophase temperature
region, the Colho1 mesophase also gave birefringence, as
can be seen from Fig. 5b. Thus, the dyad 3b does not
show homeotropic alignment.
Moreover, Peak H did not appear in the Colho
mesophases of 4a and 4b, but appeared only in the Colho
mesophases of 3a and 3b. Therefore, Peak H could be
attributed to the fullerene moieties. Unless the fullerenes
would be helically staked in the Z-direction, the fullerene
balls having diameter = 10 Å could not pile up around
the column having the stacking distance between the flat
Pc disks = ca. 3.5 Å. Hence, Peak H could be assigned to
one pitch in a helical structure of the fullerenes.
It is very noteworthy that the dyad (C14S)6PcCu-
VAN-C60 (3b) having a methoxy at the phenoxy group
shows homeotropic alignment, whereas the dyad
(C14S)6PcCu-OPh-C60 (3a) having no methoxy group at
the phenoxy group does not show homeotropic alignment.
Peak H height difference. Figure 7a shows X-ray
diffraction patterns of the Colho mesophases at rt
for the present dyads, (C14S)6PcCu-VAN-C60 (3a)
and (C14S)6PcCu-OPh-C60 (3b), and the previous
dyad, (C14S)6PcCu-C12-C60 (2c in Fig. 2a) having a
long n-alkylene spacer. When the Peak H height was
normalized by the (1 0 0) reflection height for these dyads
3a, 3b and 2c, we noticed interesting differences in these
Peak H heights. The Peak H height of 2c is nearly same as
the (1 0 0) reflection height, whereas the Peak H heights
of 3a and 3b are 1.5~2.5 times larger than those of the
(1 0 0) reflections. This implies that the helical
structures of fullerene moieties are formed in much
longer distance for 3a and 3b than 2c. Therefore,
the fullerenes may more strongly aggregate for
3a and 3b than 2c. Furthermore, the helical pitch
values of Peak H are smaller for 3a and 3b (H =
61.3 Å and H = 58.7 Å) than that of 2c (H = 71.5 Å).
This means that the helical diameters are smaller for 3a
and 3b than 2c. Accordingly, the fullerenes are closer to
the phthalocyanine cores for 3a and 3b than 2c.
Temperature-dependent X-ray diffraction study
In Table 2 are summarized the X-ray data of the
Pc precursors, (C14S)6PcCu-VAN-CHO (4a) and
(C14S)6PcCu-OPh-CHO (4b), and the corresponding
Pc-C60 dyads, (C14S)6PcCu-VAN-C60 (3a) and
(C14S)6PcCu-OPh-C60 (3b). As can be seen from this
table, each of the mesophases in 3a, 3b, 4a and 4b
could be identified from the X-ray diffraction study as a
hexagonal ordered columnar (Colho) mesophase.
Helical structure of fullerene moieties in 3a and
3b established by SAXS. Very interestingly, these Colho
mesophases in 3a and 3b gave an additional large peak
in a very low angle region of 2q ≈ 1.5° (ca. 60 Å) of their
SAXS (small angle X-ray scattering) patterns. We named
this additional peak as Peak H which we found also in
our previous PcCu-C60 dyads at the first time in 2009
[15–18, 20–22]. This peak H could not be assigned to a
reflection from any two-dimensional lattices in columnar
mesophases known up to date, so that we assigned this
peak as a helical pitch of C60 moieties around the Pc
column, and we described a schematic model of the C60
helical structure around the Pc column [15–18, 20–22],
Figure 7b shows X-ray diffraction patterns of the
isotropic liquid (I.L.) at high temperatures for the present
dyads, 3a and 3b, and the previous dyad 2c. As can
Copyright © 2016 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2016; 20: 1451–1456
Watarai, Ayumi
