Nanometer-scale ordering in cast films of columnar metallomesogen as revealed by STM observations

Article abstract of DOI:10.1039/b419429b

STM observations were performed on a cast film of a columnar metallomesogen ([Cr(5C₈)₃]; 5C₈ = 1-(3,4,5-trioctyloxyphenyl) -3-(3,4-dioctyloxyphenyl)propane-1,3-dionate anion) on a graphite surface, revealing the nanometer-scale surface ordering into an oblique lattice (a = 10.5 nm, b = 11.5 nm, α = 55°) possibly due to the ΔΛ-chiral interactions. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b419429b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Nanometer-scale ordering in cast films of columnar metallomesogen as
revealed by STM observations{
Norishige Kakegawa,*a Naomi Hoshino,b Yuki Matsuoka,a Noboru Wakabayashi,c Shin-ichiro Nishimurac
and Akihiko Yamagishiad
Received (in Cambridge, UK) 5th January 2005, Accepted 7th March 2005
First published as an Advance Article on the web 17th March 2005
DOI: 10.1039/b419429b
STM observations were performed on a cast film of a columnar
metallomesogen ([Cr(5C8)3]; 5C8 5 1-(3,4,5-trioctyloxyphenyl)-
3-(3,4-dioctyloxyphenyl)propane-1,3-dionate anion) on a gra-
phite surface, revealing the nanometer-scale surface ordering
into an oblique lattice (a 5 10.5 nm, b 5 11.5 nm, a 5 55u)
possibly due to the DL-chiral interactions.
Microscopic ordering in molecular assemblies has attracted
extensive interest because it plays a central role in a number of
chemical and physical processes such as nonlinear optics, ferro-
electric properties, and catalytic functions. Surface observation
techniques such as atomic force microscopy (AFM) and scanning
tunneling microscopy (STM) have been favorably employed to
detect these ordering phenomena.1,2
When liquid crystalline (LC) materials are subject to such
observations, nano-scale ordered structures, if observed in the LC
Fig. 1 Model structure of [Cr(5C8)3].
films, may be thought of as a two-dimensional (2D) manifestation
of intricate interactions in their bulk phases. Previous studies on
LC films of columnar triphenylene derivatives3–5 have even
steric considerations (a total of 15 octyloxy chains must be
unraveled surface induced chirality in suitably alkylated cases. In
distributed as in 8/7 instead of 9/6 along the pseudo-C3 axis).
contrast, metallomesogenic molecules contain characteristically
Accordingly, its DSC thermogram displayed sharp transitions
rigid coordination structures such as planar, tetrahedral, and
both on heating and cooling for Colh–I at 115 uC (DH 5
3.5 kJ/mol), which is reasonable compared to the literature data
octahedral ones. Of particular interest to us are the octahedral
(ca. 100 uC for the dodecyl homologue).7,8 Focal conic fan texture
mesogens which are expected to express a certain regularity arising
from their stereoselective packing involving the DL-chirality.6–8
An interesting proposal by Swager6 states that, in a hexagonal
was observed under a polarized light microscope at 109 uC, which
is also consistent with the mesophase identification. On the other
columnar (Colh) mesophase of labile complexes, microdomains
hand, no clear transition was observed for crystal-to-Colh.
may form where a D- or L-isomer dominates. This is based on an
According to the previous studies on analogous compounds
([Fe(5C12)3]),6 this class of metallomesogen has a tendency for
assumption that the phase consists of persistently homochiral
columns, which would lead to a typically frustrated system. It is
supercooling due to their large masses and the relatively low
suggestive of segregation of chiral domains under favorable
clearing temperature. Thus, no solidification exotherm was
conditions, but no direct verification has been presented. It came
recorded in cooling runs.
to us that exploration in the 2D hexagonal framework might serve
STM observations were performed with a Nanoscope III
as a test for the feasibility. The present paper reports on the first
(Digital Instruments) scanning tunneling microscope.
A
example of surface ordering in the cast film of [Cr(5C8)3] (Fig. 1)
on highly ordered pyrolytic graphite (HOPG). The formation of a
nanometer-scale structure found at a vertical resolution of less
than 0.05 nm is interpreted as a possible manifestation of the chiral
interactions.
sample was prepared by casting a chloroform solution of
[Cr(5C8)3] (1 6 1023 M) onto a freshly cleaved HOPG
(0.8 cm 6 0.8 cm). The tunneling current was IT 5 2.0 nA.
The images were recorded in the constant current mode at a
scanning rate of 20 lines per second at air and room temperature.
Fig. 2A shows a 25 nm 6 25 nm STM image at a sample tunnel
voltage of Vs 5 766 mV. Bright spots in a 2D lattice of estimated
dimensions 2.0–2.3 nm are clearly visible, and this should indicate
the molecular arrangement on the HOPG surface considering the
molecular dimension (ca. 2 nm, Fig. 1). The unit cell can be
regarded as hexagonal with a 5 2.28 nm, while the image appears
as somewhat compressed towards a centered rectangular lattice.
[Cr(5C8)3] was prepared following literature procedures with
some modifications.{ The product could be a mixture of mer- and
fac-isomers, but the isolated sample is likely to be the former from
{ Electronic supplementary information (ESI) available: focal conic fan
texture and DSC thermogram. See http://www.rsc.org/suppdata/cc/b4/
b419429b/
*kawahagi@eps.s.u-tokyo.ac.jp
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 2375–2377 | 2375

View Article Online
Fig. 3 XRD peaks of the cast film of [Cr(5C8)3].
showing a peak at 2.23 nm, which is close to the above lattice
constant. This peak is assignable to the (100) diffraction from the
hexagonal lattice, as observed in the Colh phases of a higher
homologue.8 This experiment gives an assurance that the
molecules of [Cr(5C8)3] will sit on HOPG with their pseudo-C3
axes parallel to the substrate normal. The image in Fig. 2A
suggests that the monomolecular layer of [Cr(5C8)3] is rotationally
disordered as in its Colh phase.
Fig. 2B shows a 100 nm 6 100 nm STM image for a different
sample at tunnel voltage of Vs 5 766 mV. No molecular images
were obtained in this region. On the right-hand side, a large rough
area was observed due to multi-layer formation. On the left-hand
side, however, there emerged a regular array of bright domains of
nanometer scale. When a 50 nm 6 50 nm area is enlarged as in
Fig. 2C, a definite periodicity in the 2D arrangement of bright
spots shows up. It fits to an oblique lattice (a 5 10.5 nm, b 5
11.5 nm, a 5 55u), and can also be approximated as hexagonal.
From the measurements of height profiles, the bright regions were
only 0.05 nm higher than the dark regions. Such periodic regions
span over the length of 50–70 nm.
The structure as observed in Fig. 2C may be interpreted on the
basis of a triangular Ising net. We start with a chiral lowest-energy
state (Fig. 4a). A hexagon in the lower part of Fig. 4a indicates
seven molecules arranged in the hexagonal base network (six at
peripheral sites and one at the center). The triangle marking for the
Fig. 2 STM images of the cast film of [Cr(5C8)3] on HOPG. The
conditions were V 5 766 mV, I 5 2.0 nA.
Fig. 4 Schematics showing (a) a chiral 2D hexagonal arrangement of
[Cr(5C8)3] octahedra with thick lines indicating the ligands, and (b) a
possible arrangement of them on HOPG. The rhombic lattice drawn in
solid line is the !3 6 !3 superlattice and the lattice drawn in dashed line is
three times larger than the !3 6 !3 superlattice.
X-ray diffraction measurements were then performed with a
Rigaku Rint Ultima System. Fig. 3 shows the diffraction pattern,
2376 | Chem. Commun., 2005, 2375–2377
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Norishige Kakegawa,*a Naomi Hoshino,b Yuki Matsuoka,a
peripheral six molecules is to be viewed as the top plane of
[Cr(5C8)3], and then the up and down triangles correspond to the
D- and L-enantiomers, respectively. They must alternate for close
packing. The central molecule, however, cannot be fixed as either
one of the enantiomers, since it is impossible for that molecule to
form racemic pairs with all its neighbours. Because of this
frustration, the molecule at the hexagonal site marked by a double
circle can have random orientations. Fig. 4b shows the 2D
hexagonal network with such a double circle site at the center, and
!3 6 !3 superlattice sites relative to it are indicated by circles (the
rhombic lattice is drawn as a solid line in Fig. 4b). In fact, in the
case of a triangular triphenylene derivative with suitable alkyl
chains, the !3 6 !3 superlattice was observed as bright spots in
STM images.4 In order to explain the present nanometer scale
periodicity as observed in Fig. 2C, it is further assumed that a
rhombic unit as indicated by dashed lines in Fig. 4b provides the
contrast in STM images. Then three times expansion would result
in a superlattice that is 3!3 (5 5.2) times greater in length than the
original triangular net, and this dimension is close to what was
observed in Fig. 2C.§ At present, however, it is not clear why the
ordering is suppressed on such a large scale, unlike in triphenylene
derivatives. The system also involves vertical frustration related to
the packing of 15 octyloxy chains, and the small contrast reported
above may reflect surface undulation due to the partitioning of 8 vs.
7 chains along the normal. Thus the problem may be better treated
by a coupled XY-Ising model.9 Further experiments are now in
progress to obtain more STM images hopefully with (sub)mole-
cular resolution. Preparation of enantiomerically enriched
materials is also under way.
Noboru Wakabayashi,c Shin-ichiro Nishimurac and Akihiko Yamagishiad
aThe University of Tokyo, Graduate School of Science, 7-3-1,
Bunkyo-ku, 113-0033, Tokyo, Japan.
E-mail: kawahagi@eps.s.u-tokyo.ac.jp; Fax: +81-5841-4553;
Tel: +81-5841-4553
bDivision of Chemistry, Graduate School of Science, Hokkaido
University, Sapporo, 060-0810, Japan
cDivision of Biological Sciences, Graduate School of Science, Hokkaido
University, Sapporo, 060-0810, Japan
dCREST, Japan Science and Technology Agency, Japan
Notes and references
{ [Cr(5C8)3] was prepared by reacting 1.0 g of [Cr(acac)3] (10 mmole) and
3.0 g of 5C8H (30 mmole) in a Teflon vessel at 180 uC for 12 hours. The
brown product was dissolved in methanol and eluted on a column packed
with silica gel (C500, Wako). The eluted fraction was further purified
chromatographically on an HPLC column (Capcel Pack, Shiseido). The
eluting solvent was 1 : 1 (v/v) acetonitrile/benzene. The collected fraction
was analyzed by MS: m/e 5 2645 (calc: 2644.92).
§ We have a reservation that smaller lattices may emerge if the images are
processed. However, the results reported herein are what we have observed
repeatedly and to the best of our resolution.
1 U. Mazur, M. Leonetti, W. A. English and K. W. Hipps, J. Phys. Chem.
B, 2004, 108, 17003.
2 A. Miura, Z. J. Chen, H. Uji-i, S. De Feyter, M. Zdanowska,
P. Jonkheijm, A. P. H. J. Schenning, E. W. Meijer, F. Wurthner and
F. C. De Schryver, J. Am. Chem. Soc., 2003, 125, 14968.
3 L. Askadskaya, C. Boeffel and J. P. Rabe, Ber. Bunsenges. Phys. Chem.,
1993, 97, 517.
4 F. Charra and J. Cousty, Phys. Rev. Lett., 1998, 80, 1682.
5 H. Schonherr, M. Manickam and S. Kumar, Langmuir, 2002, 18,
7082.
6 S. T. Trzaska, H. F. Hsu and T. M. Swager, J. Am. Chem. Soc., 1999,
121, 5418.
7 T. M. Swager and H. Zheng, Mol. Cryst. Liq. Cryst., 1995, 260,
301.
We thank Dr K. Fujii (NIMS, Japan) for assistance in DSC
measurements. This work has been financially supported by a
Grant-in-Aid for Scientific Research on Priority Areas (417) from
the Ministry of Education, Culture, Sports, Science, and
Technology (MEXT) of the Japanese Government.
8 H. Zheng and T. M. Swager, J. Am. Chem. Soc., 1994, 116,
761.
9 M. He´bert and A. Caille´, Phys. Rev. E, 1995, 51, R1651.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 2375–2377 | 2377