Macroscopic spinning chirality memorized in spin-coated films of spatially designed dendritic zinc porphyrin J-aggregates

Article abstract of DOI:10.1002/anie.200461431

Supramolecular polymerization of dendritic zinc porphyrins with two carboxylic acid functionalities leads to the formation of J-aggregates through multipoint π-stacking interactions. Spin-coating of the J-aggregate solutions gives optically active films,

Full text of DOI:10.1002/anie.200461431

Communications
Chirality
Macroscopic Spinning Chirality Memorized in
Spin-Coated Films of Spatially Designed
Dendritic Zinc Porphyrin J-Aggregates**
Tatsuya Yamaguchi, Tatsumi Kimura, Hiro Matsuda,
and Takuzo Aida*
The breaking of chiral symmetry can occur under nonequili-
brated conditions.[1–4] Such a spontaneous induction of optical
activity from achiral entities has attracted much attention in
relation to the origin of chirality in nature. Representative
examples can be seen in growth processes of large electro-
static assemblies of achiral chromophoric compounds such as
cyanine dyes and porphyrin derivatives in aqueous media.[5]
In these cases, the assemblies are considered to adopt helical
architectures, where either the P or M form is selected only
accidentally at the initial stage of the self-organization event
and develops predominantly in the subsequent stage. Hence,
handedness of emerging chirality is unpredictable. Ribó
et al.[5c] have discovered, through studies on electrostatic J-
aggregation of a 4-sulfonatophenylporphyrin in aqueous
media, that the sense of such an optical activity can be
selected by the spinning direction of vortex stirring initially
applied to the solutions. Herein we report an interesting
finding that spin-coated films of dendritic zinc porphyrin J-
aggregates chiroptically memorize the spinning directions.
The optical activities of the spin-coated films are thermally
stable and preserved up to the melting temperatures of the J-
aggregates. In contrast with the previous examples,[5] the J-
aggregates are optically inactive in solution.
Porphyrin J-aggregates have attracted attention because
of their potential application as nonlinear optical materials.[6]
Although several examples of porphyrin J-aggregates have
been reported, they are formed mostly in aqueous media;[5b,c]
however, there are few examples of J-aggregation in organic
media[7] which occur because of rather weak p-stacking
interactions. We found that zinc porphyrins bearing carbox-
ylic acid (-CO2H) functionalities at the opposite meso posi-
tions undergo supramolecular polymerization by dimeriza-
tion of the -CO2H groups to give J-aggregates in organic
media. The J-aggregation takes great advantage of a multi-
[*] Dr. T. Yamaguchi, Prof. Dr. T. Aida
Aida Nanospace Project
ExploratoryResearch for Advanced Technology(ERATO)
Japan Science and TechnologyAgency(JST)
2-41 Aomi, Koto-ku, Tokyo 135-0064 (Japan)
Fax: (+81)3-5841-7310
E-mail: aida@macro.t.u-tokyo.ac.jp
Dr. T. Kimura, Dr. H. Matsuda
Photonics Research Institute
National Institute of Advanced Industrial Science and
Technology(AIST)
Tsukuba Central 5, 1-1-1 Higashi Tsukuba, Ibaraki 305-8565 (Japan)
[**] T.Y. thanks the JSPS Young Scientist Fellowship.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
6350
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200461431
Angew. Chem. Int. Ed. 2004, 43, 6350 –6355

Angewandte
Chemie
point p-stacking interaction between the polymeric zinc
porphyrin molecules. Our molecular design strategy also
includes introduction of [G2] or [G3] dendritic wedges at the
residual two meso positions (Gm/nacid, where m is the
generation number of the dendritic wedges (2 or 3) and n is
the number of benzyl ether units (0—2) in the spacer parts of
the dendritic wedges; Scheme 1). The dendritic wedges are
essential not only for providing the resulting supramolecular
polymers with sufficiently high solubilities but also for steric
control over the p-stacking interaction among the polymeric
focal cores. For comparison, we synthesized a zinc porphyrin
dicarboxylic acid (G0C18/1acid) bearing long alkyl chains, and
investigated the aggregation behavior of this nondendritic
reference as well as those of ester versions Gm/nester, which
1acid. Other dendritic zinc porphyrins such as G2/0acid–G2/2acid,
G3/0acid, and nondendritic G0C18/1acid (Scheme 1) were synthe-
sized in a similar manner to the above and unambiguously
characterized.[8] As the benzyl ether spacers between the zinc
porphyrin unit and the dendritic wedges become longer, the
zinc porphyrin core should possess a larger spatial freedom
for the p-stacking interaction (G2/0acid < G2/1acid < G2/2acid
;
G3/0acid < G3/1acid). On the other hand, the spatial freedom
should become smaller as the dendritic wedges become larger
(G3/0acid < G2/0acid; G3/1acid < G2/1acid). In contrast with the
case of these dendritic versions Gm/nacid, the zinc porphyrin
unit in G0C18/1acid is omitted because of the absence of
dendritic wedges.
We found that [G2] dendritic zinc porphyrin dicarboxylic
acids such as G2/0acid–G2/2acid all form stable J-aggregates in
CHCl3. For example, the electronic absorption spectrum of a
CHCl3 solution of G2/2acid (5 mm) at 258C displayed an
intense red-shifted Soret band at 453 nm and a blue-shifted
band at 413 nm (Figure 1c) which is characteristic of zinc
porphyrin J-aggregates. G2/1acid with shorter benzyl ether
spacers exhibited a virtually identical absorption spectral
profile to G2/2acid (Figure 1b). Although G2/0acid without
were the synthetic precursors of Gm/nacid
.
For the synthesis of G3/1acid (Scheme 1), the zinc complex
of 5,15-bis(4-methoxycarbonylphenyl)-10,20-bis(4-hydroxy-
phenyl)porphyrin was treated under alkaline conditions
with [G3] poly(benzyl ether) dendron bromide, and the
resulting ester (G3/1ester) was hydrolyzed with KOH. After
neutralization, the reaction mixture was subjected to prepa-
rative size-exclusion chromatography (SEC) to isolate G3/
Scheme 1. Structures of the dendritic and nondendritic zinc porphyrin dicarboxylic acids Gm/nacid (m=0, 2, 3, n=0–2) and their esters Gm/nester
(m=0, 2, 3, n=0–2).
Angew. Chem. Int. Ed. 2004, 43, 6350 –6355
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6351

Communications
compounds are hydrogen bonded to one another to form
supramolecular polymers. In contrast, their ester versions G2/
nester (n = 0–2) showed a single Soret band at 423 nm
(Figure 1a–c; green curves), which is characteristic of non-
aggregated zinc porphyrins. Therefore, it is concluded that the
J-aggregation of zinc porphyrins can be induced in organic
media by their hydrogen-bonding interactions. We assume
that one-dimensional zinc porphyrin polymers formed by the
dimerization of the CO2H moieties stack up together, through
a multipoint p-electronic interaction, to form a large, two-
dimensional (2D) sheetlike assembly (Scheme 2). The differ-
Figure 1. Electronic absorption spectra of solutions (5 mm) of Gm/nacid
(m=0, 2, 3, n=0—2; red curves) and Gm/nester (m=0, 2, 3, n=0—2;
green curves) in CHCl3 at 258C; a) G2/0acid and G2/0ester, b) G2/1acid
and G2/1ester, c) G2/2acid and G2/2ester, d) G3/0acid and G3/0ester, e) G3/
1acid and G3/1ester, and f) G0C18/1acid and G0C18/1ester. CD spectra of thin
films of Gm/nacid (m=2, 3, n=0–2) prepared byspin-coating in clock-
wise (red curves) and counterclockwise (blue curves) spinning direc-
Scheme 2. Structures of zinc porphyrin J-aggregates with a short,
oblique slip (a) and a long, non-oblique slip (b).
ence in the absorption spectral profile between G2/nacid (n = 1,
2) and G2/0acid (Figure 1a–c; red curves) is attributable to a
difference in the p-stacking geometry of the zinc porphyrin
units. Namely, the J-aggregates of G2/1acid and G2/2acid both
involve a short, oblique slip of the p-stacked zinc porphyrin
units (Scheme 2a), typical of J-aggregated tetraarylporphyrin
derivatives, since such a geometry can reduce the steric
interference from the meso-aryl groups on the zinc porphyrin
units.[9] On the other hand, the J-aggregate of G2/0acid, which is
devoid of any spacers between the dendritic wedges and the
zinc porphyrin core, involves a long, non-oblique slip of the p-
stacked zinc porphyrin units (Scheme 2b), possibly because of
limited spatial freedom for the p-stacking interaction. The
dendritic side chains, which are located presumably on both
sides of the 2D sheet, are responsible for the high solubilities
of the J-aggregates. For reference, the absorption spectral
tions are also shown; a’) G2/0acid, b’) G2/1acid, c’) G2/2acid, d’) G3/0acid
e’) G3/1acid, f’) G0C18/1acid
,
.
benzyl ether spacers formed a J-aggregate, it displayed a
slightly different absorption spectrum, with two red-shifted
Soret bands at 451 and 432 nm (Figure 1a). The infrared
spectra of solutions of G2/nacid (n = 0—2; 5 mm) in CHCl3 all
showed a carbonyl stretching vibration at 1688 cmÀ1 arising
from a dimeric form of CO2H, with only a negligibly small
shoulder attributable to free CO2H at 1725 cmÀ1 [8]
Dynamic
.
light scattering (DLS) studies on a dilute solution of J-
aggregated G2/1acid in CHCl3 (0.05 mm), for example, showed
the presence of large assemblies with an average radius of
200 nm.[8] Thus, the zinc porphyrin cores of these dendritic
6352
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 6350 –6355

Angewandte
Chemie
profile of nondendritic G0C18/1acid in hot CHCl3 (Figure 1 f)
was analogous to those of J-aggregated G2/1acid and G2/2acid
crystalline particles by polarized microscopy (Figure 3a–c).
,
The particles from G2/0acid were smaller than those from the
other two compounds. Quite interestingly, these films were
optically active, although their solutions were silent in circular
but the compound gradually precipitated on standing at 258C.
In contrast with the [G2] dendritic zinc porphyrins G2/nacid
(n = 0–2), solutions of the one-generation higher systems G3/
0acid and G3/1acid (5 mm) in CHCl3 both displayed a single
Soret band at 428 nm, similar to their ester versions, thus
indicating that their zinc porphyrin cores are not assembled
through p-stacking interactions (Figure 1d,e). Infrared spec-
troscopic analysis of these compounds showed a considerable
amount of free CO2H groups at 1725 cmÀ1, along with dimeric
CO2H groups at 1688 cmÀ1 [8]
Therefore, it is likely that the
.
large dendritic wedges attached to the zinc porphyrin core
suppress both the hydrogen-bonding and p-stacking inter-
actions. On the other hand, we also found that G3/1acid gives a
J-aggregate in an apolar solvent such as C6H6, where a dimeric
form of CO2H is stable. The electronic spectrum of a solution
of G3/1acid in C6H6 (5 mm) showed a red-shifted Soret band at
453 nm and a blue-shifted band at 414 nm (Figure 2b; red
Figure 2. Electronic absorption spectra of solutions (5 mm) of G3/nacid
(n=0, 1; red curves) and G3/nester (n=0, 1; green curves) at 258C in
C6H6; a) G3/0acid and G3/0ester, b) G3/1acid and G3/1ester. CD spectra of
thin films of G3/nacid (n=0, 1) prepared byspin-coating in the clock-
wise (red curves) and counterclockwise (blue curves) directions are
Figure 3. Polarized-light micrographs of spin-coated films from solu-
tions of Gm/nacid (m=2, 3, n=0–2) in CHCl3 and solutions of G3/nacid
also shown; a’) G3/0acid, b’) G3/1acid
.
(n=0, 1) in C6H6: a) G2/0acid, b) G2/1acid, c) G2/2acid, d) G3/0acid
e) G3/1acid, f) G3/0acid, g) G3/1acid; scale bar, 50 mm.
,
curve), whereas ester G3/1ester, under identical conditions,
displayed an ordinary Soret band at 429 nm (Figure 2b; green
curve). Infrared spectroscopic analysis of G3/1acid in C6H6
(5 mm), in contrast with the case in CHCl3, showed a carbonyl
stretching vibration predominantly at 1688 cmÀ1 arising from
the dimeric form of CO2H. These observations again support
our hypothesis that the zinc porphyrin units, when polymer-
ized through hydrogen-bonding interactions, have an
enhanced capability for p stacking. In contrast, G3/0acid, with
a sterically encumbered zinc porphyrin core, formed a weak
gel in C6H6 whose absorption spectrum (Figure 2a; red curve)
was similar to that of non-aggregated G3/0ester (Figure 2a;
green curve).
dichroism (CD) spectroscopy. We also found that the spin-
coated films displayed a chirality dominance in the statistical
distributions that was dependent on the “spinning direction”,
whereas the cast films did not show any dominance in
chirality. For example, when a solution of G2/2acid in CHCl3
(5 mm) was spin-coated in a clockwise direction at 6000 rpm,
the resulting film displayed intense CD bands at 413, 440, and
453 nm with positive, negative, and positive signs, respectively
(Figure 1c’; red curve). On the other hand, spin-coating of the
same solution in a counterclockwise direction resulted in the
appearance of a mirror-image CD spectrum (Figure 1c’; blue
curve). The intensities of the CD bands hardly changed when
observed by rotating the samples along an axis perpendicular
to the substrate surface,[8] thus indicating a negligibly small
contamination with linear dichroism. Furthermore, we pre-
pared 10 samples for each spinning direction and confirmed a
In the course of the above study we noticed that J-
aggregated G2/nacid (n = 0–2) gave birefringent films when
cast from their CHCl3 solutions on to glass plates. Spin-
coating of these solutions also resulted in the formation of
birefringent films, which showed the presence of rodlike
Angew. Chem. Int. Ed. 2004, 43, 6350 –6355
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6353

Communications
complete dominance of the emerging chirality sense.[8] Such a
because of a large steric repulsion between the dendritic
wedges, along with the one-dimensionality of the backbone
polymers. However, such 2D molecular sheets, when trans-
ferred to the solid state, are known to role up to form coiled
architectures.[4,10] If this happens to the J-aggregated 2D
sheet, the zinc porphyrin units are forced to adopt a twisted
(chiral) geometry relative to one another, and the assembly
can eventually be optically active when the parity of chirality
is broken. On the other hand, when a large rotational sheer
force is applied to this roll-up event by spin-coating, either a
right-handed or left-handed helical coil, depending on the
spinning direction, may be selected. In other words, the spin-
coated films can chiroptically memorize the macroscopic
spinning direction. We found that the chiroptical memory
thus fixed is thermally stable. For example, the spin-coated
film of J-aggregated G2/2acid still preserved its optically
activity when heated at 2008C for five minutes. Further
heating the film up to 2608C resulted in it losing its
birefringence and becoming optically inactive.
In summary, we have demonstrated that spin-coating of
hydrogen-bonded dendritic zinc porphyrin J-aggregates gives
optically active films, where either of the two enantiomeric
forms is selected by the spinning direction. This is the first
successful example of the transformation of a macroscopic
spinning chirality into a stable supramolecular chirality in the
solid state. Extension of this finding to other self-assembling
systems and application of the resulting optically active
materials to absolute asymmetric synthesis and chiral sepa-
ration are the challenging subjects worthy of further inves-
tigation.
spinning direction dependent chirality dominance was also
observed for spin-coated films of J-aggregated G2/1acid
(Figure 1b’). As already described, the J-aggregates of G2/
1acid and G2/2acid both involve a short, oblique slip of the p-
stacked zinc porphyrin units (red curves in Figure 1b and c,
respectively). In sharp contrast, J-aggregated G2/0acid, which
is proposed to involve a long, non-oblique slip of the zinc
porphyrin p stacks (Figure 1a; red curve), gave a spin-coated
film that exhibited negligibly weak CD bands (Figure 1a’) in
the visible region. Similar weak CD bands resulted when
nondendritic G0C18/1acid, which is J-aggregated in CHCl3
(Figure 1 f), was spin-coated (Figure 1 f’). These observations
indicate the importance of spatial design around the hydro-
gen-bonded zinc porphyrin chromophores for the emergence
of chirality on spin-coating.
One-generation higher G3/0acid, which is not associated
through p interactions in CHCl3 (Figure 1d) or even in C6H6
(Figure 2a), gave neither birefringent (Figure 3d, f) nor
optically active films (Figure 1d’, Figure 2a’) on spin-coating.
A nonbirefringent, optically inactive film (Figure 3e, Fig-
ure 1e’) also resulted when non-assembled G3/1acid in CHCl3
(Figure 1e) was spin-coated. In contrast, spin-coating of a
solution of J-aggregated G3/1acid in C6H6 (Figure 2b) resulted
in the formation of a birefringent, optically active film
(Figure 3g, Figure 2b’) that exhibited a spinning direction
dependent CD response. On the other hand, non-assembled
ester versions such as Gm/nester (m = 2, 3, n = 0–2) and G0C18
ester (Figure 1a–f; green curves) gave CD-silent films on spin-
coating, as expected.
/
1
Figure 4 shows a schematic representation of the sug-
gested J-aggregated 2D sheet. This is composed of an offset
stacking of hydrogen-bonded dendritic zinc porphyrin poly-
mers (Scheme 2). A helical J-aggregate structure, proposed
for the precedent examples,[5b,c] seems unlikely in solutions,
Received: July 26, 2004
Keywords: chirality· dendrimers · J-aggregate · porphyrinoids ·
.
self-assembly
[1] J. Jacques, A. Collet, S. H. Wilen, Enantiomers, Racemates,
Resolutions, Wiley, New York, 1998.
[2] a) D. K. Kondepudi, R. J. Kaufman, N. Singh, Science 1990, 250,
975 – 976; b) D. K. Kondepudi, K. L. Bullock, J. A. Digits, J. K.
Hall, J. M. Miller, J. Am. Chem. Soc. 1993, 115, 10211 – 10216;
c) D. K. Kondepudi, J. Laudadio, K. Asakura, J. Am. Chem. Soc.
1999, 121, 1448 – 1451.
[3] D. R. Link, G. Natale, R. Shao, J. E. Maclennan, N. A. Clark, E.
Körblava, D. M. Walba, Science 1997, 278, 1924 – 1927.
[4] X. Huang, C. Li. S. Jiang, X. Wang, B. Zhang, M. Liu, J. Am.
Chem. Soc. 2004, 126, 1322 – 1323.
[5] a) U. De Rossi, S. Dꢀhne, S. C. J. Meskers, H. P. J. M. Dekkers,
Angew. Chem. 1996, 108, 827 – 830; Angew. Chem. Int. Ed. Engl.
1996, 35, 760 – 763; b) O. Ohno, Y. Kaizu, H. Kobayashi, J. Chem.
Phys. 1993, 99, 4128 – 4139; c) J. M. Ribó, J. Crusats, F. Sagues,
J. M. Claret, R. Ruvires, Science 2001, 292, 2063 – 2066.
[6] a) C. Halvorson, A. Hays, B. Kraabel, R. Wu, F. Wudl, A. J.
Heeger, Science 1994, 265, 1215 – 1216; b) H. S. Nalwa, Adv.
Mater. 1993, 5, 341 – 358; c) K. Misawa, T. Kobayashi, J. Chem.
Phys. 1999, 110, 5844 – 5850.
[7] a) S. Okada, H. Segawa, J. Am. Chem. Soc. 2003, 125, 2792 –
2796; b) M. Shirakawa, S. Kawano, N. Fujita, K. Sada, S. Shinkai,
J. Org. Chem. 2003, 68, 5037 – 5044.
Figure 4. A proposed mechanism for the formation of a chiral zinc
porphyrin J-aggregate.
[8] See Supporting Information.
6354
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 6350 –6355

Angewandte
Chemie
[9] J.-H. Furhop, C. Demoulin, C. Boettcher, J. König, U. Siggel, J.
Am. Chem. Soc. 1992, 114, 4159 – 4165.
[10] a) J. M. Schnur, Science 1993, 262, 1669 – 1676; b) R. Oda, I.
Huc, M. Schmutz, S. J. Candau, F. C. MacKintosh, Nature 1999,
399, 566 – 569; c) E. D. Sone, E. R. Zubarev, S. I. Stupp, Angew.
Chem. 2002, 114, 1781 – 1785; Angew. Chem. Int. Ed. 2002, 41,
1706 – 1709; d) J. P. Hill, W. Jin, A. Kosaka, T. Fukushima, H.
Ichihara, T. Shimomura, K. Ito, T. Hashizume, N. Ishii, T. Aida,
Science 2004, 304, 1481 – 1483.
Angew. Chem. Int. Ed. 2004, 43, 6350 –6355
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6355