Mechanism of Helix Induction on a Stereoregular Poly((4-carboxyphenyl)acetylene) with Chiral Amines and Memory of the Macromolecular Helicity Assisted by Interaction with Achiral Amines

Article abstract of DOI:10.1021/ja0318378

Cis-transoidal poly((4-carboxyphenyl)acetylene) (poly-1) is an optically inactive polymer but forms an induced one-handed helical structure upon complexation with optically active amines such as (R)-(1-(1-naphthyl)ethyl) amine ((R)-2) in DMSO. The complexes show a characteristic induced circular dichroism (ICD) in the UV-visible region of the polymer backbone. Moreover, the macromolecular helicity of poly-1 induced by (R)-2 can be "memorized" even after complete replacement of (R)-2 by various achiral amines. We now report fully detailed studies on the mechanism of the helicity induction and memory of the helical chirality of poly-1 by means of UV-visible, CD, and infrared spectroscopies. We have found that a one-handed helix is cooperatively induced on poly-1 upon the ion pair formation of the carboxy groups of poly-1 with optically active amines and that the bulkiness of the chiral amines plays a crucial role for inducing an excess of a single-handed helix. On the other hand, the free ion formation was found to be essential for the macromolecular helicity memory of poly-1 after the replacement of the chiral amine by achiral amines, since the intramolecular electrostatic repulsion between the neighboring carboxylate ions of poly-1 significantly contributes to reduce the atropisomerization process of poly-1. On the basis of the mechanism of helicity induction and the memory of the helical chirality drawn from the present studies, we succeeded in creating an almost perfect memory of the induced macromolecular helicity of poly-1 with (R)-2 by using 2-aminoethanol as an achiral chaperoning molecule to assist in maintaining the memory of helical chirality.

Full text of DOI:10.1021/ja0318378

Published on Web 03/13/2004
Mechanism of Helix Induction on a Stereoregular
Poly((4-carboxyphenyl)acetylene) with Chiral Amines and
Memory of the Macromolecular Helicity Assisted by
Interaction with Achiral Amines
Katsuhiro Maeda,† Kazuhide Morino,† Yoshio Okamoto,‡ Takahiro Sato,§ and
Eiji Yashima*,†
Contribution from the Department of Molecular Design and Engineering and Department of
Applied Chemistry, Graduate School of Engineering, Nagoya UniVersity, Chikusa-ku,
Nagoya 464-8603, Japan, and Department of Macromolecular Science, Osaka UniVersity,
Machikaneyama-cho 1-1, Toyonaka, Osaka 560-0043, Japan
Received December 19, 2003; E-mail: yashima@apchem.nagoya-u.ac.jp
Abstract: Cis-transoidal poly((4-carboxyphenyl)acetylene) (poly-1) is an optically inactive polymer but forms
an induced one-handed helical structure upon complexation with optically active amines such as (R)-(1-
(1-naphthyl)ethyl)amine ((R)-2) in DMSO. The complexes show a characteristic induced circular dichroism
(ICD) in the UV-visible region of the polymer backbone. Moreover, the macromolecular helicity of poly-1
induced by (R)-2 can be “memorized” even after complete replacement of (R)-2 by various achiral amines.
We now report fully detailed studies on the mechanism of the helicity induction and memory of the helical
chirality of poly-1 by means of UV-visible, CD, and infrared spectroscopies. We have found that a one-
handed helix is cooperatively induced on poly-1 upon the ion pair formation of the carboxy groups of poly-1
with optically active amines and that the bulkiness of the chiral amines plays a crucial role for inducing an
excess of a single-handed helix. On the other hand, the free ion formation was found to be essential for
the macromolecular helicity memory of poly-1 after the replacement of the chiral amine by achiral amines,
since the intramolecular electrostatic repulsion between the neighboring carboxylate ions of poly-1
significantly contributes to reduce the atropisomerization process of poly-1. On the basis of the mechanism
of helicity induction and the memory of the helical chirality drawn from the present studies, we succeeded
in creating an almost perfect memory of the induced macromolecular helicity of poly-1 with (R)-2 by using
2-aminoethanol as an achiral chaperoning molecule to assist in maintaining the memory of helical chirality.
Introduction
interesting and valuable not only to mimic biopolymers but also
to develop novel chiral materials in areas such as liquid crystals,
Proteins and nucleic acids are typical optically active,
biological polymers with a one-handed helicity.1 Their helical
structures seem to be essential for their sophisticated and
fundamental functions in nature. After discovery of the helical
structures in these biopolymers, significant attention has been
paid to developing artificial helical polymers2 and oligomers
(foldamers)3 with a controlled helix-sense. Such studies are
membranes, and chiral selectors.2,3 To date, a large number of
optically active polymers have been prepared, but optically
active polymers, whose optical activity is largely governed by
the helical chirality of the polymer backbone, are still limited.2
Fully synthetic helical polymers prepared so far can be classified
into two types on the basis of the nature of their helical
conformations; one is a stable (or static) helical polymer even
in solution, and the other is a dynamic helical polymer. Poly-
(triphenylmethyl methacrylate),2b,4 polychloral,5 polyiso-
cyanides,2c,2i,6 and poly(2,3-quinoxalines)7 belong to the former
category. The one-handed helical polymers have been prepared
by the screw-sense selective polymerization of achiral or
† Department of Molecular Design and Engineering, Nagoya University.
‡ Department of Applied Chemistry, Nagoya University.
§ Osaka University.
(1) (a) Saenger, W. Principles of Nucleic Acid Structure; Springer-Verlag: New
York, 1984. (b) Schulz, G. E.; Schirmer, R. H. Principles of Protein
Structure; Springer-Verlag: New York, 1979.
(2) Reviews: (a) Farina, M. In Topics in Stereochemistry; Ernest, L. E., Samuel,
H. W., Eds.; John Wiley & Sons: New York, 1987; Vol. 17. (b) Okamoto,
Y.; Nakano, T. Chem. ReV. 1994, 94, 349-372. (c) Nolte, R. J. M. Chem.
Soc. ReV. 1994, 23, 11-19. (d) Green, M. M.; Peterson, N. C.; Sato, T.;
Teramoto, A.; Cook, R.; Lifson, S. Science 1995, 268, 1860-1866. (e)
Pu, L. Acta Polym. 1997, 48, 116-141. (f) Rowan, A. E.; Nolte, R. J. M.
Angew. Chem., Int. Ed. 1998, 37, 63-68. (g) Green, M. M.; Park, J.-W.;
Sato, T.; Teramoto, A.; Lifson, S.; Selinger, R. L. B.; Selinger, J. V. Angew.
Chem., Int. Ed. 1999, 38, 3138-3154. (h) Nakano, T.; Okamoto, Y. Chem.
ReV. 2001, 101, 4013-4038. (i) Cornelissen, J. J. L. M.; Rowan, A. E.;
Nolte, R. J. M.; Sommerdijk, N. A. J. M. Chem. ReV. 2001, 101, 4039-
4070. (j) Fujiki, M. Macromol. Rapid Commun. 2001, 22, 539-563.
(3) Reviews: (a) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173-180. (b)
Hill, D. J.; Mio, M. J.; Prince, R. B.; Hughes, T. S.; Moore, J. S. Chem.
ReV. 2001, 101, 3893-4011. (c) Cheng, R. P.; Gellman, S. H.; DeGrado,
W. F. Chem. ReV. 2001, 101, 3219-3232.
(4) (a) Okamoto, Y.; Suzuki, K.; Ohta, K.; Hatada, K.; Yuki, H. J. Am. Chem.
Soc. 1979, 101, 4763-4765. (b) Okamoto, Y.; Yashima, E. Prog. Polym.
Sci. 1990, 15, 263-298. (c) Nakano, T.; Okamoto, Y.; Hatada, K. J. Am.
Chem. Soc. 1992, 114, 1318-1329.
(5) (a) Corley, L. S.; Vogl, O. Polym. Bull. 1980, 3, 211-217. (b) Ute, K.;
Hirose, K.; Kashimoto, H.; Nakayama, H.; Hatada, K.; Vogl, O. Polym. J.
1993, 25, 1175-1186.
9
10.1021/ja0318378 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 4329-4342
4329

A R T I C L E S
Maeda et al.
Chart 1
prochiral monomers with chiral catalysts or initiators. These
polymers have bulky substituents at the side chains, so that a
one-handed helical conformation is formed and stabilized by
steric hindrance during the polymerization under kinetic control.
On the other hand, polyisocyanates2d,g,8 and polysilanes2j,9
are typical helical polymers belonging to the latter category;
they have long, alternate sequences of left- and right-handed
helices. An equilibrium exists in solution between the helices
separated by the helix reversal points that move along the
polymer backbone; therefore, they are called dynamic helical
polymers. However, in these polymers the helix reversals occur
infrequently so that they have a long persistence length. Due to
this remarkable feature, optically active polyisocyanates and
polysilanes with a predominantly one-handed helix-sense have
been successfully prepared by introducing a tiny amount of a
chiral factor to the polymers, such as the copolymerization of
achiral monomers with a small amount of optically active
monomers. The helix-sense is determined under thermodynamic
control.8,9
temperature dependence of the CD spectra for the homopoly-
mers of optically active phenylacetylenes and their copolymers
with achiral phenylacetylenes in detail and found that the helical
conformations are dynamic in nature like polyisocyanates
because the ICD magnitudes of several optically active poly-
(phenylacetylene)s composed of chiral and achiral phenylacety-
lenes significantly increased with decreasing temperature. On
the basis of the theoretical analysis of the CD data, we succeeded
in estimating the thermodynamic stability parameters for the
helical conformations.12
Poly(phenylacetylene)s bearing an optically active substituent
are also considered to take a helical conformation with a
predominant one-handedness in solution, because they show a
characteristic induced circular dichroism (ICD) in the π-π*
electronic transition region due to the conjugated double bonds
in the polymer main chain.10,11 Recently, we investigated the
Optically active helical polymers described above can be
prepared either by polymerization of optically active monomers
or by screw-sense selective polymerization; their helical struc-
tures and helix-senses are determined by chiral or bulky
substituents covalently bonded to the polymer main chains,
thermodynamically or kinetically, during the polymerization
process.
Besides these helical polymers, we recently succeeded in
inducing a predominantly one-handed helical conformation in
optically inactive polymers bearing functional groups upon
noncovalent complexation with optically active small molecules
capable of interacting with the functional groups of the
polymers. Cis-transoidal poly((4-carboxyphenyl)acetylene) (poly-
1, Chart 1) is the first example of such a noncovalent, one-
handed helicity induction.13 In the presence of optically active
amines, poly-1 forms complexes with the amines through
noncovalent, chiral acid-base interaction;14 a one-handed helical
structure can be induced on poly-1, resulting in a characteristic
ICD in the UV-visible region of the polymer backbone (Figure
(6) (a) Nolte, R. J. M.; van Beijnen, A. J. M.; Drenth, W. J. Am. Chem. Soc.
1974, 96, 5932-5933. (b) Kamer, P. C. J.; Nolte, R. J. M.; Drenth, W. J.
Am. Chem. Soc. 1988, 110, 6818-6825. (c) Deming, T. J.; Novak, B. M.
J. Am. Chem. Soc. 1992, 114, 7926-7927. (d) Takei, F.; Yanai, K.;
Onitsuka, K.; Takahashi, S. Angew. Chem., Int. Ed. Engl. 1996, 35, 1554-
1556. (e) Cornelissen, J. J. L. M.; Fischer, M.; Sommerdijk, N. A. J. M.;
Nolte, R. J. M. Science 1998, 280, 1427-1430. (f) Amabilino, D. B.;
Ramos, E.; Serrano, J.-L,; Sierra, T.; Veciana, J. J. Am. Chem. Soc. 1998,
120, 9126-9134. (g) Takei, F.; Yanai, K.; Onitsuka, K.; Takahashi, S.
Chem.sEur. J. 2000, 6, 983-993. (h) Cornelissen, J. J. L. M.; Donners,
J. J. J. M.; de Gelder, R.; Graswinckel, W. S.; Metselaar, G. A.; Rowan,
A. E.; Sommerdijk, N. A. J. M.; Nolte, R. J. M. Science 2001, 293, 676-
680.
(7) (a) Ito, Y.; Ihara, E.; Murakami, M. Angew. Chem., Int. Ed. Engl. 1992,
31, 1509-1510. (b) Ito, Y.; Miyake, T.; Hatano, S.; Shima, R.; Ohara, T.;
Suginome, M. J. Am. Chem. Soc. 1998, 120, 11880-11893.
(8) (a) Green, M. M.; Andreola, C.; Mun˜oz, B.; Reidy, M. P.; Zelo, K. J. Am.
Chem. Soc. 1988, 110, 4063-4065. (b) Green, M. M.; Reidy, M. P.;
Johnson, R. J.; Darling, G.; O’Leary, D. J.; Willson, G. J. Am. Chem. Soc.
1989, 111, 6452-6454. (c) Lifson, S.; Andreola, C.; Peterson, N. C.; Green,
M. M. J. Am. Chem. Soc. 1989, 111, 8850-8858. (d) Okamoto, Y.;
Matsuda, M.; Nakano, T.; Yashima, E. J. Polym. Sci., Part A: Polym.
Chem. 1994, 32, 309-315. (e) Maeda, K.; Matsuda, M.; Nakano, T.;
Okamoto, Y. Polym. J. 1995, 27, 141-146. (f) Mu¨ller, M.; Zentel, R.
Macromolecules 1996, 29, 1609-1617. (g) Maeda, K.; Okamoto, Y.
Macromolecules 1998, 31, 1046-1052. (h) Li, J.; Schuster G. B.; Cheon,
K.-S.; Green, M. M.; Selinger, J. V. J. Am. Chem. Soc. 2000, 122, 2603-
2612.
(9) (a) Fujiki, M. J. Am. Chem. Soc. 1994, 116, 6017-6018. (b) Fujiki, M. J.
Am. Chem. Soc. 1994, 116, 11976-11981. (c) Fujiki, M. J. Am. Chem.
Soc. 2000, 122, 3336-3343.
(11) (a) Yashima, E.; Huang, S.; Okamoto, Y. J. Chem. Soc., Chem. Commun.
1994, 1811-1812. (b) Yashima, E.; Huang, S.; Matsushima, T.; Okamoto,
Y. Macromolecules 1995, 28, 4184-4193. (c) Yashima, E.; Matsushima,
T.; Nimura, T.; Okamoto, Y. Korean Polym. J. 1996, 4, 139-146. (d)
Yashima, E.; Maeda, Y.; Okamoto, Y. J. Am. Chem. Soc. 1998, 120, 8895-
8896. (e) Yashima, E.; Maeda, Y.; Okamoto, Y. Polym. J. 1999, 31, 1033-
1036. (f) Yashima, E.; Maeda, K.; Sato, O. J. Am. Chem. Soc. 2001, 123,
8159-8160. (g) Nishimura, T.; Takatani, K.; Sakurai, S.; Maeda, K.;
Yashima, E. Angew. Chem., Int. Ed. 2002, 41, 3602-3604. (h) Morino,
K.; Maeda, K.; Yashima, E. Macromolecules 2003, 36, 1480-1486.
(12) Morino, K.; Maeda, K.; Okamoto, Y.; Yashima, E.; Sato, T. Chem.sEur.
J. 2002, 8, 5112-5120.
(13) (a) Yashima, E.; Matsushima, T.; Okamoto, Y. J. Am. Chem. Soc. 1995,
117, 11596-11597. (b) Yashima, E.; Matsushima, T.; Okamoto, Y. J. Am.
Chem. Soc. 1997, 119, 6345-6359. (c) Yashima, E. Anal. Sci. 2002, 18,
3-6. (d) Maeda, K.; Morino, K.; Yashima, E. J. Polym. Sci., Part A: Polym.
Chem. 2003, 41, 3625-3631. (e) Yashima, E.; Maeda, K.; Nishimura, T.
Chem.sEur. J. 2004, 10, 42-51.
(10) (a) Yamaguchi, M.; Omata, K.; Hirama, M. Chem. Lett. 1992, 2261-2262.
(b) Aoki, T.; Kokai, M.; Shinohara, K.; Oikawa, E.; Chem. Lett. 1993,
2009-2012. (c) Matsuura, K.; Furuno, S.; Kobayashi, K. Chem. Lett. 1998,
847-848. (d) Kwak, G.; Masuda, T. J. Polym. Sci., Part A: Polym. Chem.
2001, 39, 71-77. (e) Li, B. S.; Cheuk, K. K. L.; Salhi, F.; Lam, J. W. Y.;
Cha, J. A. K.; Xiao, X.; Bai, C.; Tang, B. Z. Nano Lett. 2001, 1, 323-328.
For screw-sense selective polymerization, see: (f) Aoki, T.; Kaneko, T.;
Maruyama, N.; Sumi, A.; Takahashi, M.; Sato, T.; Teraguchi, M. J. Am.
Chem. Soc. 2003, 125, 6346-6347. For optically active aliphatic poly-
acetylenes, see: (g) Ciardelli, F.; Benedetti, E.; Pieroni, O. Makcromol.
Chem. 1967, 103, 1-18. (h) Ciardelli, F.; Lanzillo, S.; Pieroni, O.
Macromolecules 1974, 7, 174-179. (i) Moore, J. S.; Gorman, C. B.;
Grubbs, R. H. J. Am. Chem. Soc. 1991, 113, 1704-1712. (j) Kishimoto,
Y.; Itou, M.; Miyatake, T.; Ikariya, T.; Noyori, R. Macromolecules 1995,
28, 6662-6666. (k) Aoki, T.; Shinohara, K.; Kaneko, T.; Oikawa, E.
Macromolecules 1996, 29, 4192-4198. (l) Akagi, K.; Piao, G.; Kaneko,
S.; Sakamaki, K.; Shirakawa, H.; Kyotani, M. Science 1998, 282, 1683-
1686. (m) Nakako, H.; Nomura, R.; Tabata, M.; Masuda, T. Macromol-
ecules 1999, 32, 2861-2864. (n) Nomura, R.; Fukushima, Y.; Nakako,
H.; Masuda, T. J. Am. Chem. Soc. 2000, 122, 8830-8836. (o) Mitsuyama,
M.; Kondo, K. Macromol. Chem. Phys. 2000, 201, 1613-1618. (p) Tabei,
J.; Nomura, R.; Masuda, T. J. Am. Chem. Soc. 2002, 124, 5405-5409.
(14) For related counterion induction of supramolecular chirality, see: (a) Oda,
R.; Huc, I.; Candau, S. J. Angew. Chem., Int. Ed. 1998, 37, 2689-2691.
(b) Oda, R.; Huc, I.; Schmutz, M.; Candau, S. J. Nature 1999, 399, 566-
569. (c) Berthier, D.; Buffeteau, T.; Le´ger, J.-M.; Oda, R.; Huc, I. J. Am.
Chem. Soc. 2002, 124, 13486-13494 and references therein.
9
4330 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Helix Induction and Macromolecular Helicity Memory
A R T I C L E S
Figure 1. Schematic illustration of one-handed helicity induction on poly-1 upon complexation with chiral amines. The original poly-1 is drawn as a
mixture of right- and left-handed helices (top), which change to a predominantly one-handed helix with chiral amines (bottom), thus exhibiting an ICD in
the π-conjugated polymer backbone region.
1).13 The split type of Cotton effect reflects the absolute
configuration of the chiral amines; all primary amines of the
same configuration gave the same Cotton effect signs. Therefore,
the Cotton effect signs can be used as a novel probe for the
assignments of the absolute configuration of the chiral amines.13b,15
Poly-1 is not a stiff but a flexible polymer because of its short
persistent length (4.2 nm in dimethyl sulfoxide (DMSO)),16 and
therefore, a one-handed helicity induction may occur if the twist
of adjacent double bonds around a single bond takes place
favorably.13b However, our recent viscometric studies on the
solution structure of poly-1 in DMSO combined with theoretical
analysis indicated that poly-1 can be regarded as an analogue
of a dynamic helical polymer of which right- and left-handed
helical conformations are rapidly interconvertible.16 Therefore,
the appearance of ICDs of poly-1 in the presence of chiral
amines is considered to be due to a change in population of the
dynamic helices (Figure 1).
as well as aliphatic functional polyacetylenes,21 also respond
to the chirality of optically active compounds and form an
induced one-handed helical structure, thus exhibiting a similar
ICD in the UV-visible region in organic solvents.13,18,20 Similar
helicity induction can be possible in water through electrostatic
interactions with a variety of biomolecules including amino
acids, amino sugars, and peptides on water-soluble poly-
(phenylacetylene)s including poly-1.19,22
Besides polyacetylene derivatives, similar helicity induction
on an optically inactive polymer through noncovalent bonding
interaction with chiral compounds has been reported for
polyphosphazene,23 polyisocyanide,24 polyisocyanate,25 polyguani-
dine,26 polyaniline,27 and polypyrrole.28
The macromolecular helicity induced on these polymers might
be dynamic in nature. However, interestingly, we recently found
that such induced helical chirality of poly-1 by an optically
(19) Onouchi, H.; Maeda, K.; Yashima, E. J. Am. Chem. Soc. 2001, 123, 7441-
7442.
Related stereoregular poly(phenylacetylene)s bearing other
functional groups, such as a boronic acid residue,17 an amino
group,18 a phosphonate group,19 and a crown ether pendant20
(20) (a) Nonokawa, R.; Yashima, E. J. Am. Chem. Soc. 2003, 125, 1278-1283.
(b) Nonokawa, R.; Yashima, E. J. Polym. Sci., Part A: Polym. Chem. 2003,
41, 1004-1013. (c) Nonokawa, R.; Oobo, M.; Yashima, E. Macromolecules
2003, 36, 6599-6606.
(15) For recent reviews of chirality recognition of chiral amines based on ICDs,
see: (a) Canary, J. W.; Holmes, A. E.; Liu, J. Enantiomer 2001, 6, 181-
188. (b) Borovkov, V. V.; Lintoluoto, J. M.; Inoue, Y. J. Am. Chem. Soc.
2001, 123, 2979-2989. (c) Kurta´n, T.; Nesnas, N.; Li, Y.-Q.; Huang, X.;
Nakanishi, K.; Berova, N. J. Am. Chem. Soc. 2001, 123, 5962-5973. (d)
Tsukube, H.; Shinoda, S. Chem. ReV. 2002, 102, 2389-2403. (e) Allenmark,
S. Chirality 2003, 15, 409-422.
(21) (a) Yashima, E.; Goto, H.; Okamoto, Y. Polym. J. 1998, 30, 69-71. (b)
Maeda, K.; Goto, H.; Yashima, E. Macromolecules 2001, 34, 1160-1164.
(22) Saito, M. A.; Maeda, K.; Onouchi, H.; Yashima, E. Macromolecules 2000,
33, 4616-4618.
(23) Yashima, E.; Maeda, K.; Yamanaka, T. J. Am. Chem. Soc. 2000, 122,
7813-7814.
(24) (a) Ishikawa, M.; Maeda, K.; Yashima, E. J. Am. Chem. Soc. 2002, 124,
7448-7458. (b) Ishikawa, M.; Maeda, K.; Mitsutsuji, Y.; Yashima, E. J.
Am. Chem. Soc. 2004, 126, 732-733.
(16) Ashida, Y.; Sato, T.; Morino, K.; Maeda, K.; Okamoto, Y.; Yashima, E.
Macromolecules 2003, 36, 3345-3350.
(17) (a) Yashima, E.; Nimura, T.; Matsushima, T.; Okamoto, Y. J. Am. Chem.
Soc. 1996, 118, 9800-9801. (b) Kawamura, H.; Maeda, K.; Okamoto, Y.;
Yashima, E. Chem. Lett. 2001, 58-59.
(25) (a) Green, M. M.; Kartri, C.; Peterson, N. C. J. Am. Chem. Soc. 1993, 115,
4941-4942. (b) Maeda, K.; Yamamoto, N.; Okamoto, Y. Macromolecules
1998, 31, 5924-5926.
(18) (a) Yashima, E.; Maeda, Y.; Okamoto, Y. Chem. Lett. 1996, 955-956. (b)
Yashima, E.; Maeda, Y.; Matsushima, T.; Okamoto, Y. Chirality 1997, 9,
593-600. (c) Maeda, K.; Okada, S.; Yashima, E.; Okamoto, Y. J. Polym.
Sci., Part A: Polym. Chem. 2001, 39, 3180-3189.
(26) Schlitzer, D. S.; Novak, B. M. J. Am. Chem. Soc. 1998, 120, 2196-2197.
(27) Majidi, M. R.; Kane-Maguire, L. A. P.; Wallace, G. G. Polymer 1995, 36,
3597-3599.
(28) Zhou, Y.; Yu, B.; Zhu, G. Polymer 1997, 38, 5493-5495.
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4331

A R T I C L E S
Maeda et al.
Table 1. Molecular Weight Effect of Poly-1 on Helicity Induction with Chiral Amines in DMSO
second Cotton (10-4[θ] (λ))c
(R)-3e
b
f
poly-1
10-4M (M /M )a
DP
ꢀ400
(R)-2d
(S)-4
n
w
n
poly-1a
poly-1b
poly-1c
1.7 (2.1)
3.3 (2.8)
13 (4.3)g
106
206
812
2840
3130
3180
-2.82 (376)
-2.95 (377)
-3.08 (376)
-2.88 (374)
-3.53 (374)
-3.61 (373)
0.89 (376)
1.37 (377)
1.41 (377)h
a Determined by SEC as its methyl ester using polystyrene standards with THF (poly-1a and poly-1b) or chloroform (poly-1c) as the eluent. b Degree of
polymerization of poly-1. c The molar ratio of (R)-2, (R)-3, and (S)-4 to monomer units of poly-1 was 10. The concentrations of poly-1 were 3 ((R)-2) and
1 mg/mL ((R)-3 and (S)-4), respectively; [θ] in deg cm2 dmol-1 and λ in nm. d Measured just after the preparation of samples. e Measured after the samples
had been allowed to stand at ambient temperature for 24 (poly-1a), 30 (poly-1b), and 45 h (poly-1c). f Measured after the samples had been allowed to stand
at 80 °C for 5 (poly-1a), 6 (poly-1b), and 20 h (poly-1c). g Chloroform-soluble part. h The molar ratio of (S)-4 to monomer units of poly-1c was 3. The
concentration of poly-1c was 5 mg/mL.
active amine could be maintained, namely “memorized”, when
the chiral amine was replaced by various achiral amines in
DMSO.29,30 The memory of macromolecular helicity was not
transient but lasted for an extremely long time. The memory
efficiency was influenced by small structural changes in the
achiral amines.29
This work is concerned with the mechanism of helix
formation of poly-1 in the presence of chiral amines and the
memory of the helicity process with achiral amines by means
of CD and IR spectroscopies. The results will contribute to the
construction of novel helical polymeric architectures with a
desired helix-sense and supramolecular helical assemblies with
desired achiral molecules in a one-handed helical array.
know that the magnitude of the ICD of poly-1 corresponding
to an excess of a single-handed helix (right- and left-handed
helices) increases with an increase in the bulkiness of the chiral
amines, whereas chiral amino alcohols generate a very intense
ICD irrespective of the bulkiness.13b Although the assignments
of the splitting Cotton effects in the ICD of poly-1 with chiral
amines have not yet been done, the presumed molecular model
of the poly-1-(R)-primary amines or -amino alcohols suggests
that the poly-1 may have a left-handed helix in the presence of
(R)-amines.13d
The ICD intensities tended to increase with an increase in
the molecular weight of poly-1; especially for the poly-1-(R)-3
complexes, the tendency was remarkable. The ICD intensity of
poly-1 with the less bulky amino alcohol (R)-3 increased slowly
with time, while the complex with the bulky amine (R)-2 showed
almost no change in the ICD intensity with time regardless of
the molecular weight of poly-1 (see Figure S-1A in the
Supporting Information). We also followed the changes in the
ICD of poly-1 with different molecular weights using the less
bulky chiral amine ((S)-4) and found a similar tendency but
with an unusually slow increase in the ICD intensity. The ICD
intensity increased very slowly with time at ambient temperature
and continued to increase even after 50 days (see Figure S-1B
in the Supporting Information). The increase in the ICD
remarkably accelerated at 80 °C and reached an almost constant
value after 5-20 h (see Table 1). However, poly-1 showed a
slight decrease in the absorption intensity, probably due to
decomposition or isomerization of poly-1 at 80 °C.
Results and Discussion
Molecular Weight Effect of Poly-1 on Helicity Induction.
In the previous studies, we used rather low molecular weight
poly-1s (Mn ) 17 000 and 46 000) prepared by the polymeri-
zation of (4-((triphenylmethoxy)carbonyl)phenyl)acetylene with
[Rh(nbd)Cl]2 (nbd ) norbornadiene), followed by hydrolysis
of the ester group, for the ICD experiments.13a,b,29 Our recent
finding of the direct polymerization of (4-carboxyphenyl)-
acetylene (1) in water with a water-soluble rhodium catalyst in
the presence of bases, such as diethylamine, made it possible
to prepare a higher molecular weight poly-1.22 Therefore, we
first investigated the effect of molecular weights of poly-1 on
ICD with chiral amines ((R)-2 and (S)-4) and an amino alcohol
((R)-3) in DMSO (Chart 1). The molecular weights of cis-
transoidal poly-1 (poly-1a-c) prepared with rhodium catalysts
are summarized in Table 1 together with the ICD results with
(R)-2, (R)-3, and (S)-4 and molar absorptivities (ꢀ400) of poly-
1s in DMSO at ambient temperature (23-25 °C).13b,22,29 These
amines were selected as a chiral bulky ((R)-2) and a less bulky
amine ((S)-4) and as a chiral amino alcohol ((R)-3), since we
The molar absorptivities of poly-1 in DMSO also increased
with increasing the molecular weight. Judging from the ratio
of the CD intensity (exactly, the molar ellipticity, ∆ꢀ) and molar
absorptivity (ꢀ), which corresponds to the Kuhn dissymmetric
ratio (g-factor ) ∆ꢀ/ꢀ),31 the molecular weight of poly-1 seems
to have almost no effect on the ICD upon complexation with
(R)-2; the g-factor values of poly-1a-poly-1c complexed with
(29) Yashima, E.; Maeda, K.; Okamoto, Y. Nature 1999, 399, 449-451.
(30) For chirality memory effect in the fields of host-guest and supramolecular
chemistry, see: (a) Furusho, Y.; Kimura, T.; Mizuno, Y.; Aida, T. J. Am.
Chem. Soc. 1997, 119, 5267-5268. (b) Mizuno, T.; Takeuchi, M.; Hamachi,
I.; Nakashima, K.; Shinkai, S. J. Chem. Soc., Perkin Trans. 2 1998, 2281-
2288. (c) Sugasaki, A.; Ikeda, M.; Takeuchi, M.; Robertson, A.; Shinkai,
S. J. Chem. Soc., Perkin Trans. 1 1999, 3259-3264. (d) Rivera, J. M.;
Craig, S. L.; Martin, T.; Rebek, J., Jr. Angew. Chem., Int. Ed. 2000, 39,
2130-2132. (e) Prins, L. J.; de Jong, F.; Timmerman, P.; Reinhoudt, D.
N. Nature 2000, 408, 181-184. (f) Mizuno, Y.; Aida, T.; Yamaguchi, K.
J. Am. Chem. Soc. 2000, 122, 5278-5285. (g) Kubo, Y.; Ohno, T.;
Yamanaka, J.; Tokita, T.; Iida, T.; Ishimaru, Y. J. Am. Chem. Soc. 2001,
123, 12700-12701. (h) Lauceri, R.; Raudino, A.; Scolaro, L. M.; Micali,
N.; Purrello, R. J. Am. Chem. Soc. 2002, 124, 894-895. (i) Prins, L. J.;
Verhage, J. J.; de Jong, F.; Timmerman, P.; Reinhoudt, D. N. Chem.s
Eur. J. 2002, 8, 2302-2313. (j) Ishi-i, T.; Crego-Calama, M.; Timmerman,
P.; Reinhoudt, D. N.; Shinkai, S. J. Am. Chem. Soc. 2002, 124, 14631-
14641. (k) Zieglar, M.; Davis, A. V.; Johnson, D. W.; Raymond, K. N.
Angew. Chem., Int. Ed. 2003, 42, 665-668. (l) Purrello, R. Nat. Mater.
2003, 2, 216-217.
(R)-2 are 2.69 × 10-3, 2.65 × 10-3, and 2.76 × 10-3
,
respectively. On the other hand, the dissymmetric ratio for the
poly-1-(R)-3 and poly-1-(S)-4 complexes was dependent on
the molecular weight and increased with increasing the molec-
ular weight; the g-factors of poly-1a-poly-1c complexed with
(R)-3 are 2.68 × 10-3, 3.24 × 10-3, and 3.37 × 10-3, and those
complexed with (S)-4 are 0.82 × 10-3, 1.17 × 10-3, and 1.17
× 10-3, respectively. Similar effect of molecular weight on the
optical activity was also observed for rodlike helical polymers,
(31) (a) Kuhn, W. Trans. Faraday Soc. 1930, 46, 293-308. (b) Dekkers, H. P.
J. M. In Circular Dichroism: Principles and Applications; Nakanishi, K.,
Berova, N., Woody, R. W., Eds.; VCH: New York, 1994; Chapter 6.
9
4332 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Helix Induction and Macromolecular Helicity Memory
A R T I C L E S
Scheme 1. Series of Acid-Base Equilibria for the Complexation of Poly-1 with Chiral Amines in DMSO
Table 2. Binding Constants (Ks) for the Complexation of Amines
with Poly-1c and VBA in DMSO
K (M-1 a
)
amines
VBA
poly-1c
(R)-2
(R)-3
(S)-4
95 (57 ( 2b)
48
2003
1723
974 (509 ( 30b)
949 (671 ( 57b,c
)
a Estimated by IR titrations (see the Supporting Information). b Cited from
ref 13b, and these values were estimated by 1H NMR titrations in DMSO-
d6. c Revised value in this work.
appeared; their intensities increased with increasing the con-
centration of (R)-2 (Figure 2). These IR spectral changes indicate
that the poly-1c forms a complex with (R)-2 through acid-
base interaction and the carboxy groups of poly-1c change to
carboxylate ions in DMSO in the presence of (R)-2.
Previously, we tried to estimate the binding constants (Ks)
of poly-1 to chiral amines using 1H NMR titrations, but it was
difficult because of the broadening of the peaks of chiral amines
and amino alcohols.13b Therefore, 4-vinylbenzoic acid (VBA)
(Chart 1) was used as a model compound of poly-1 to estimate
K values for chiral amines by using 1H NMR titrations.
However, we can now follow the complexation of poly-1 with
amines in DMSO by means of IR spectroscopy, so that their K
values upon complexation with poly-1 could be determined
(Table 2).
Figure 2. IR spectral changes of the poly-1c-(R)-2 complex in DMSO.
The IR spectra were measured in a 0.01 cm CaF2 cell at ambient temperature
with a poly-1c concentration of 5 mg/mL.
polyisocyanates32 and polysilanes;33 this was explained theoreti-
cally by using the Ising model. A similar model may be
applicable to explain the molecular weight effect on the ICD
of poly-1 depending on the structure of chiral amines on the
assumption that poly-1-amine complexes are dynamic helical
polymers consisting of right- and left-handed helices separated
by the helix reversal points.16 However, such a theoretical
analysis has not yet been done because of a limited number of
data. We then used the high molecular weight poly-1c through-
out for all the experiments of this study except for the “memory”
experiments unless otherwise noted.
The K values of (R)-2, (R)-3, and (S)-4 upon complexation
with poly-1c, determined by the IR titration experiments, are
listed in Table 2 together with those with VBA. IR titrations
were performed under the conditions of constant poly-1c or
VBA concentration (5 mg/mL) with varying amine or amino
alcohol concentration at ambient temperature (ca. 25 °C), and
the changes in absorbance at 1703 cm-1 were followed. The
titration curves were analyzed by a nonlinear least-squares
method by using a binding model with a 1:1 stoichiometry37,38
and are in agreement with 1:1 complexation (see Figure S-2 in
the Supporting Information). We observed no time-dependent
IR spectral changes. Comparison of the K values indicates a
higher affinity of the amine ((S)-4) and amino alcohol ((R)-3)
to poly-1c and VBA than that of the bulky amine ((R)-2). We
Complexation Behavior of Poly-1 with Chiral Amines in
DMSO through Acid-base Interaction. In the acid-base
complexation of poly-1 with chiral amines in DMSO, there exist
equilibria between the free acid and base, the ion pair, and the
free ions as shown in Scheme 1.13b,34 Among these three species,
the ion pairs have been shown to be predominant in DMSO35
and should be important for the helicity induction on poly-1,
resulting in the appearance of an ICD in the polymer backbone.
The IR spectral changes of poly-1 in DMSO with an increasing
amount of (R)-2 support this ion pair formation. Figure 2 shows
the changes in the IR spectra of poly-1c upon complexation
with (R)-2 in DMSO. Poly-1c showed absorptions at 1703 and
1261 cm-1 corresponding to the free carboxylic acid bands
(COOH),36 whose intensities decreased with increasing the
concentration of (R)-2 and almost completely disappeared at
[(R)-2]/[poly-1c] ) 20, and new bands at 1583 and 1371 cm-1
assigned to the ion-paired and/or dissociated acid band (COO-)
(36) In nonpolar solvents such as benzene and the solid state, benzoic acid is
known to exist as an intermolecular hydrogen-bonded dimer which shows
a peak due to the dimeric COOH band at 1688 cm-1, while in the gaseous
state it exists as a monomer, showing a peak due to the free COOH band
at 1762 cm-1. On the other hand, in polar DMSO, benzoic acid exists as
a monomer because of solvation with DMSO through hydrogen bonding,
and therefore, it shows a peak due to the COOH band at 1704 cm-1. See:
Novak, P.; Vikic-Topic, D.; Meic, Z.; Sekusak, S.; Sabljic, A. J. Mol. Struct.
1995, 356, 131-141. In the solid state, the COOH group of benzoic acid
(32) (a) Gu, H.; Nakamura, Y.; Sato, T.; Teramoto, A.; Green, M. M.; Andeola,
C.; Peerson, N. C.; Lifson, S. Macromolecules 1995, 28, 1016-1024. (b)
Okamoto, N.; Mukaida, F.; Gu, H.; Nakamura, Y.; Sato, T.; Teramoto, A.;
Green, M. M.; Andeola, C.; Peerson, N. C.; Lifson, S. Macromolecules
1996, 29, 2878-2884. (c) Gu, H.; Sato, T.; Teramoto, A.; Varichon, L.;
Green, M. M. Polym. J. 1997, 29, 77-84. (d) Jha, S. K.; Cheon, K.-S.;
Green, M. M.; Selinger, J. V. J. Am. Chem. Soc. 1999, 121, 1665-1673.
(33) Teremoto, A.; Terao, K.; Terao, Y.; Nakamura, N.; Sato, T.; Fujuki, M. J.
Am. Chem. Soc. 2001, 123, 12303-12310.
(34) (a) Izutsu, K. Acid-Base Dissociation Constants in Dipolar Aprotic
SolVents; Blackewll: Oxford, U.K., 1990. (b) Reichardt, C. SolVents and
SolVent Effects in Organic Chemistry; VCH: Weinheim, Germany, 1990.
(35) Zingg, S. P.; Arnett, E. M.; Mcphail, A. T.; Bothner-By, A. A.; Gilkerson,
W. R. J. Am. Chem. Soc. 1988, 110, 1565-1580.
hydrogen bonded with amines also exhibits a vibration at ca. 1720 cm-1
.
Nema´k, K.; AÄ cs, M.; Ja´szay, Z. M.; Kozma, D.; Fogassy, E. Tetrahedron
1996, 52, 1637-1642.
(37) (a) Yashima, E.; Yamamoto, C.; Okamoto, Y. J. Am. Chem. Soc. 1996,
118, 4036-4048. (b) Yamamoto, C.; Yashima, E.; Okamoto, Y. J. Am.
Chem. Soc. 2002, 124, 12583-12589.
(38) (a) Sawada, M.; Okumura, Y.; Shizuma, M.; Takai, Y.; Hidaka, Y.; Yamada,
H.; Tanaka, T.; Kaneda, T.; Hirose, K.; Misumi, S.; Takahashi, S. J. Am.
Chem. Soc. 1993, 115, 7381-7388. (b) Albert, J. S.; Goodman, M. S.;
Hamilton, A. D. J. Am. Chem. Soc. 1995, 117, 1143-1144.
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4333

A R T I C L E S
Maeda et al.
Figure 3. (A) ICD ([θ]2nd) intensity changes of poly-1c with (R)-2 (b, method 1; O, method 2) and (R)-3 (9, method 1; 0, method 2) versus poly-1c
concentration in DMSO at 25 °C. The ICD ([θ]2nd) intensities of the poly-1c-(R)-3 complexes were measured after the sample had been allowed to stand
for 10 h. The concentrations of (R)-2 and (R)-3 were kept constant at 0.34 M and 68 mM, respectively. (B) ICD ([θ]2nd) intensity changes of poly-1c with
(R)-2 versus (R)-2‚HCl concentration in DMSO. The initial dilute poly-1c solution with (R)-2 (0.34 M) in DMSO was indicated by an arrow in part A.
note that there seems to be a macromolecular effect in the acid-
base complexation depending on the structure of the chiral
amines; the carboxy group of poly-1c has a higher binding
affinity to the less bulky chiral amines ((R)-3 and (S)-4) than
that of its polymer model, VBA, although the K value of the
bulky amine (R)-2 upon complexation with poly-1c is lower
than that with VBA. The decrease in binding affinity of (R)-2
to poly-1c may be due to the bulkiness of (R)-2, which might
prevent further binding of (R)-2 to poly-1c.
Although acid-base complexations in organic solvents have
been studied extensively,39 a complete structural interpretation
of the equilibrium species in solution is still not completely
solved.35 Particularly, it is difficult to distinguish between ion
pairs and free ions and to determine the equilibrium constant
between them by means of IR and 1H NMR spectroscopies.35,39a
However, we have recently confirmed the existence of the free
ions by measuring viscosity of a poly-1c solution with or without
(R)-2 in DMSO,16 since viscosity is very sensitive to a change
in global conformation of polymers, particularly polyelectrolytes.
The reduced viscosity ([η]sp/c) value of poly-1c in the presence
of (R)-2 in DMSO increased monotonically with the decrease
in the polymer concentration, while the poly-1c in the absence
of (R)-2 in DMSO showed a linear relationship in the Huggins
plot. The dependence of the [η]sp/c on the concentration of poly-
1c for the poly-1c-(R)-2 complex in DMSO became normal
after the addition of the hydrochloride of (R)-2 ((R)-2‚HCl) (0.02
M) into the solution. These viscosity changes are a characteristic
feature of polyelectrolytes, suggesting that the ion pair complex
of poly-1c with (R)-2 is at least partly dissociated into the free
ions (carboxylate and ammonium ions) in DMSO and the
dissociation into the free ions can be suppressed in the presence
of a common salt, (R)-2‚HCl.16 This implies that poly-1c and
(R)-2 can be complexed more tightly through ion pairing in the
presence of a small amount of (R)-2‚HCl, resulting in a full
ICD with a small amount of (R)-2 with (R)-2‚HCl (see below).
Mechanism of Helicity Induction on Poly-1 with Chiral
Amines. As described above, the ion pairing is considered to
be important for the helicity induction on poly-1 with chiral
amines in DMSO. To further elucidate the effect of the ion pair
and the free ions on the helicity induction on poly-1, dilution
experiments of poly-1c in DMSO containing a large excess of
(R)-2 or (R)-3 to poly-1c were performed to correlate the
concentration of poly-1c and the ICD intensity (Figure 3).
During the dilution experiments, the concentrations of (R)-2 and
(R)-3 in DMSO were kept constant (0.34 M and 68 mM,
respectively) and these DMSO solutions were used as diluents
to avoid a shift of the complexation to the free acid and base
from the ion pair in the first equilibrium in Scheme 1. In the
course of the dilution experiments, we found that the ICD
intensities of the poly-1c solutions were highly dependent on
the dilution methods (methods 1 and 2).
In method 1, a series of dilute poly-1c solutions in DMSO
(0.005-10 mg/mL) and a stock solution of (R)-2 in DMSO (0.68
M) were prepared and equal volumes of the solutions were
mixed at once to give the desired concentrations of poly-1c
complexed with (R)-2 in DMSO. As shown in Figure 3A, the
ICD intensity of the poly-1c-(R)-2 complex decreased as the
dilution was increased and became less than one-third the
original maximum value at log([poly-1c]) ) -4.8 as marked
by an arrow in Figure 3A. A similar decrease in the ICD
intensity was also observed for the poly-1c-(R)-3 complex in
method 1.40 Since the degree of dissociation of the ion pair into
the free ions increases with decreasing the poly-1c concentration,
the observed decrease in the ICD intensity of poly-1c under
the dilute conditions indicates that the ion pairing with chiral
amines rather than the free ion plays a central role for the helicity
induction on poly-1c.
(39) (a) Wilcox, C. S. In Frontiers in Supramolecular Organic Chemistry and
Photochemistry; Schneider, H.-J., Du¨rr, H., Lehn, J.-M., Ed.; VCH:
Weinheim, Germany, 1988; pp 123-143. (b) Rebek, J., Jr. Acc. Chem.
Res. 1990, 23, 399-404. (c) Zimmerman, S. C. Top. Curr. Chem. 1993,
165, 71-101. (d) MacDonald, J. C.; Whitesides, G. M. Chem. ReV. 1994,
94, 2383-2420. (e) Kato, T.; Fre´chet, J. M. J. Macromol. Symp. 1995, 98,
311-326. (f) Paleos, C. M.; Tsiouvas, D. Angew. Chem., Int. Ed. Engl.
1995, 34, 1696-1711. (g) Desiraju, G. R. Angew. Chem., Int. Ed. Engl.
1995, 34, 2311-2327. (h) Seel, C.; Gala´n, A.; Mendoza, J. Top. Curr.
Chem. 1995, 175, 101-132. (i) Schmidtchen, F. P.; Berger, M.; Metzger,
A.; Gloe, K.; Stephan, H. In Molecular Design and Bioorganic Catalysis;
Wilcox, C. S., Hamilton, A. D., Ed.; Kluwer: Dordrecht, The Netherlands,
1996; pp 191-210. (j) Collet, A.; Ziminski, L.; Garcia, C.; Vigne´-Maeder,
F. In Supramolecular Stereochemistry; Siegel, J. S., Ed.; Kluwer: Dor-
drecht, The Netherlands, 1995; pp 90-110. (k) Hirose, T.; Naito, K.;
Shitara, H.; Nohira, H.; Baldwin, B. W. Tetrahedron: Asymmetry 2001,
12, 375-380.
As described in the previous section, the addition of the (R)-
2‚HCl is expected to suppress the degree of the dissociation of
the ion pair into the free ions. The (R)-2‚HCl was then added
to the diluted poly-1c solution showing a weak ICD, marked
(40) The ICD intensity of the poly-1c with (R)-2 in the dilute concentration
region of poly-1c gradually decreased with time, while that with (R)-3 hardly
changed.
9
4334 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Helix Induction and Macromolecular Helicity Memory
A R T I C L E S
by an arrow in Figure 3A. As expected the ICD intensity of the
diluted poly-1c solution gradually increased by the addition of
an increasing amount of the (R)-2‚HCl and recovered to the
original maximum value at [(R)-2‚HCl] ) 0.92 M (Figure 3B).
It should be noted that poly-1c showed no ICD in the same
absorption region in the presence of excess (R)-2‚HCl alone.
These results indicate that the ion pair formation between the
carboxy groups of poly-1 and chiral amines is essential for the
helicity induction in one-handedness excess.
Contrary to the results in method 1, the ICD intensities of
poly-1c in DMSO solutions of (R)-2 and (R)-3 remained almost
constant over the concentration range examined when the
samples were prepared by method 2; the concentrated poly-1c
samples in DMSO solutions of (R)-2 (0.34 M) and (R)-3 (68
mM) were initially prepared separately and then diluted with
increasing volumes of the DMSO solutions of (R)-2 (0.34 M)
and (R)-3 (68 mM), respectively; CD spectra were measured
for each dilution (Figure 3A). These unusual results suggest
that the induced helical chirality of poly-1c is maintained
regardless of the increase in the amount of the free ions41 under
dilution once a one-handed helix is induced on poly-1c assisted
by interaction with (R)-2 and (R)-3. This phenomenon is closely
correlated to the memory of macromolecular helicity of poly-
1c, which will be discussed later in detail.
To correlate the ICD intensity and the amount of the chiral
amine interacting with poly-1c, CD titration experiments were
conducted under the same conditions as the IR titrations (Figure
2) because the complexation of poly-1c with amines in DMSO
can be followed by IR as described above. In Figure 4, the ratios
([θ]/[θ]max) of the observed ICD intensity of the second Cotton
Figure 4. Plots of [θ]/[θ]max (second Cotton) for poly-1c complexed with
(R)-2 (9), (R)-3 (b), and (S)-4 (2) against the content of the carboxylate
ion (COO-) of poly-1c in DMSO at ambient temperature (ca. 24-26 °C),
where [θ] and [θ]max represent the observed and the maximum second Cotton
values for the poly-1c-(R)-2 (-3.06 × 104), poly-1c-(R)-3 (-3.53 × 104),
and poly-1c-(S)-4 (1.41 × 104) complexes, respectively, with [poly-1c] )
5 mg/mL. The ICD intensities ([θ]) of the poly-1c-(R)-2 and poly-1c-
(R)-3 were measured after the samples had been allowed to stand for 42
and 47 h at ambient temperature, respectively. The [θ] values of the poly-
1c-(S)-4 were the maximum values at 80 °C.
On the other hand, the complex with the less bulky amino
alcohol (R)-3 showed almost no ICD in the region of [COO-]
< 40%. However, the ICD intensity of the poly-1c-(R)-3
complex started to increase in a sigmoidal fashion at [COO-]
) ca. 40% and reached a maximum value at [COO-] ) ca.
80%. The less bulky (R)-3 can also induce a one-handed helix
on poly-1 cooperatively, but which requires almost complete
ion pair formation of the carboxy groups of poly-1 with the
chiral amino alcohol. The poly-1c-(S)-4 complex also showed
an increase in the ICD intensity with weak cooperativity. In
this less bulky amine, most of the carboxy groups of poly-1c
should be complexed with (S)-4 to saturate the ICD; neverthe-
less, the complex could not exhibit a full ICD under the present
conditions.
These results together with the CD titration results clearly
indicate that both basicity and bulkiness of the chiral amines
are important factors for the helicity induction and its excess
of a single-handed helix (the ICD intensity). The importance
of bulkiness for the helicity induction was already demonstrated13b
and in good agreement with the previous results for poly-
(phenylacetylene)s bearing an optically active substituent.11b,12
Dynamic Nature of Induced Helical Conformation of
Poly-1 with Chiral Amines. The stability and dynamic nature
of the helical poly-1 induced by chiral amines were investigated
by means of CD spectroscopy. The ICD of the poly-1-(R)-2
complex instantly disappeared when the complex was exposed
to a stronger acid such as trifluoroacetic acid, which liberates
the poly-1 so it reverts to the original, optically inactive
polymer.13,29 This suggests that the original poly-1 with free
carboxylic acid residues cannot maintain the helical conforma-
tion in DMSO.
([θ]) during the titrations to the maximum ICD values ([θ]max
)
for the poly-1c-(R)-2 (-3.06 × 104), poly-1c-(R)-3 (-3.53
× 104), and poly-1c-(S)-4 (1.41 × 104) were plotted against
the content of the carboxylate ion ([COO-]) in the poly-1c
determined by the IR titrations.
The ICD intensity of the poly-1c-(R)-2 complex dramatically
increased with an increase in the amount of the carboxylate ion
generated by the complexation with (R)-2 at around the region
where the [COO-] was from 20 to 50% and reached an almost
maximum value at [COO-] ) ca. 50%. This sudden onset and
rapid increase in the optical activity of poly-1c with a sigmoidal
fashion suggest that the polymer forms a slight excess of one-
handed helical structure at around this region ([COO-] = 20%)
and further change in the population of the right- and left-handed
helices of the polymer main chain into a one-handedness takes
place cooperatively on poly-1c upon the ion pair formation with
(R)-2. We note that poly-1c can form an almost one-handed
helix when half the carboxy groups of poly-1c complex with
(R)-2 form the ion pair, and the remaining half of carboxy groups
are not necessary to form the ion pairs.43
(41) The amounts of the free ions could not be determined quantitatively using
spectroscopies, but they can be calculated by considering the effect of ion
condensation according to Manning42 by using the IR titration data and
the dissociation constant for the corresponding low-molar-mass salts such
as a VBA-amine complex. The dissociation constant (0.0125 M-1) for
mandelic acid with amines in DMSO, which was determined by means of
conductance measurements, is reported by Zingg et al.35 By using this
dissociation constant (0.0125 M-1) on assumption that the dissociation
constant is almost constant regardless of the amines used, we can estimate
the amounts of the ion pair and free ions. The degrees of the ion pair and
free ions (carboxylate of poly-1) in DMSO at 25 °C were estimated to be
0.743 and 0.196, respectively, at the infinite dilution condition of poly-1
in the presence of (R)-2‚HCl (0.02 M) at [(R)-2] ) 0.35 M.16
(43) The ion pairs of poly-1 with amines partly dissociate into the free ions in
DMSO, so the obtained [COO]- value is the sum of the ion pair and the
free ions. Therefore, a practical amount of the ion pair to induce an almost
one-handed helix on poly-1 with (R)-2 must be less than 50%. The amounts
of the ion pair and free ions can be estimated by the method described in
ref 41. These results together with the theoretical analyses by the Ising
model will be published elsewhere.
(42) Manning, G. S. J. Chem. Phys. 1969, 51, 924-933.
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4335

A R T I C L E S
Maeda et al.
Figure 5. (A) CD spectral changes of the poly-1c-(R)-3 complex ((R)-3; 5 equiv of monomer units of poly-1c) by the addition of (S)-3 (method B). Molar
ratio of (S)-3 to monomer units of poly-1c is 0 (a), 5 (b), and 20 (c). The CD spectra were measured in DMSO in a 0.01 cm quartz cell at ambient
temperature (ca. 24-26 °C) with a poly-1c concentration of 3 mg/mL. (B) CD spectra of the poly-1a-(R)-2 complex ([(R)-2]/[poly-1a] ) 10) (d), the
mixture of the poly-1a-(R)-2 complex with 5 ([5]/[poly-1] ) 50) (e), and the isolated poly-1a (f) by SEC using a DMSO solution of 5 (0.8 M) as the mobile
phase, in DMSO at ambient temperature (20-25 °C). The concentration of poly-1a is 3.0 mg/mL for (d) and (e) and 0.13 mg/mL for (f).
The ICD intensity of the poly-1c-(R)-2 complex in DMSO
([(R)-2]/[poly-1c] ) 10) decreased by the addition of increasing
amount of (S)-2, and the ICD completely disappeared at [(R)-
2]/[(S)-2] ) 1. Further addition of (S)-2 induced the ICD with
an opposite Cotton effect sign, whose ICD magnitude almost
corresponded to the enantiomeric excess (ee) of the amine (see
Figure S-3A in the Supporting Information). This indicates that
the induced helix of the poly-1c is dynamic in nature, the (R)-2
complexed with poly-1 can exchange rapidly with (S)-2, and
the helix-sense is controlled by chirality of 2. However, upon
the addition of a small amount of the chiral amino alcohol (S)-3
([(S)-3]/[(R)-2] ) 0.2 ([(S)-3]/[poly-1c] ) 2)) to the poly-1c-
(R)-2 solution ([(R)-2]/[poly-1c] ) 10), the ICD spontaneously
changed and the signs inverted to give mirror images (see Figure
S-3B in the Supporting Information). In this case, the helix-
sense was determined by a rather small amount of (S)-3 with
the opposite configuration. Since an acid-base binding constant
(K (M-1)) of (S)-3 with poly-1c in DMSO obtained by the IR
titration experiments is much larger than that of (R)-2 (see Table
2), the (R)-2 is considered to be completely replaced by the
(S)-3 during the titration.
In the previous study, we found that the complex formation
of poly-1 with partially resolved amines including 2 and 3
displayed a unique, positive nonlinear relationship (chiral
amplification or “majority rule”)2d,g,32d,44 between the ee of
amines and the observed ellipticity of the Cotton effects; ICD
intensities of poly-1, corresponding to the helical sense excesses,
are out of proportion to the ee’s of amines, showing a convex
deviation from linearity through a wide range of ee values of
the chiral amines in DMSO.13b For example, when poly-1 (Mn
) 4.6 × 104) was dissolved in DMSO with a 5-fold excess 3
of 60% ee (R rich) (method A), the complex showed as intense
an ICD as that of 100% ee. The excess enantiomer bound to
the polymer may induce an excess of a single-handed helix
(right- or left-handed helix) despite its proportion, which may
result in a more intense ICD than that expected from the ee of
3. However, the departure from linearity was found to be
sensitive to the mixing manner of the (R)- and (S)-amines with
poly-1; the addition order of (S)- and (R)-3 to the poly-1 solution
determined the Cotton effect sign of the poly-1 solution
irrespective of the ee of 3.13b
Figure 5A shows the changes of CD spectra of the poly-1c-
(R)-3 complex in DMSO ([(R)-3]/[poly-1c] ) 5) after addition
of 5 and 20 equiv of (S)-3 (method B). Interestingly, even though
an excess amount of (S)-3 ([(S)-3]/[(R)-3] ) 4; ee ) 60% (S
rich)) was present in the solution, the Cotton effect sign did
not change and the ICD intensity only slightly decreased just
after the addition of excess (S)-3. The ICD intensity decreased
very slowly with time at ambient temperature; even after 2
months poly-1c still exhibited the same Cotton effect sign. At
higher temperature (80 °C), however, the ICD intensity de-
creased rapidly and the sign inverted after 13 h at 80 °C (see
Figure S-4 in the Supporting Information). These results suggest
that the exchange rate between the amino alcohol enantiomers
may be very slow or the inversion of helicity of poly-1c may
be occurring very slowly after the fast exchange reaction; the
latter was found to be the case on the basis of the ee
determination results of the bound 3 to poly-1c (see the
Experimental Section (Enantiomeric Excess Determination of
Bound 3 to Poly-1c), Figure S-5, and Table S-1 in the
Supporting Information). That is, once one of the helicities is
induced on poly-1c with the chiral amino alcohol, the polymer
maintains its helical conformation even after the bound amino
alcohol could be randomly replaced by the opposite antipode.
We then added an excess amount of an achiral amino alcohol,
2-aminoethanol (5 in Chart 2) ([5]/[poly-1] ) 50), instead of
racemic or nonracemic 3 to the poly-1-(R)-2 complex ([(R)-
2]/[poly-1] ) 50) in DMSO (Figure 5B); in this experiment,
we used poly-1a instead of poly-1c. As described below, poly-
1b and poly-1c also showed almost the same results. The achiral
amino alcohol 5 is a strong base similar to (S)-3 (K ) 2003
(M-1)), so that the (R)-2 (K ) 48 (M-1)) complexed with poly-
1a will be completely replaced by an excess of 5. However,
quite interestingly, the ICD signal was still observed as such
with only a slight decrease in the CD intensity (trace e in Figure
5B). This clearly indicates that the one-handed helical confor-
mation of the poly-1a induced by (R)-2 can be retained even
after the (R)-2 is replaced by achiral 5.
Memory of the Macromolecular Helicity of Poly-1. As
described above, the one-handed helicity of the poly-1 induced
(44) Green, M. M.; Garetz, B. A.; Chang, H. J. Am. Chem. Soc. 1995, 117,
4181-4182.
9
4336 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Helix Induction and Macromolecular Helicity Memory
A R T I C L E S
Chart 2
by (R)-2 is considered to be maintained after removal and
replacement of (R)-2 by achiral 5. To gain further concrete
evidence, we isolated the poly-1a from the poly-1a-(R)-2
complex with excess 5 by size exclusion chromatography (SEC)
using a DMSO solution of 5 (0.8 M) as the mobile phase (Figure
6). The poly-1a eluted first, followed by the (R)-2. Each fraction
was collected and subjected to CD and absorption measure-
ments. On the basis of the UV spectrum of the (R)-2 fraction,
more than 99% of the (R)-2 was recovered. Nevertheless, the
poly-1a fraction (0.13 mg/mL) containing a large excess of
achiral 5 ([5]/[poly-1a] ) ca. 900) showed an intense ICD
comparable to that measured before the SEC fractionation (trace
f in Figure 5B), which indicates the memory of the induced
helicity of poly-1a. To check if a small amount of (R)-2 is still
present in the poly-1a fraction after the SEC purification,45 the
fractionated, optically active poly-1a was injected again into
the SEC system using the same mobile phase. But we could
not detect any trace amounts of (R)-2, and the refractionated
poly-1a exhibited ICD without any loss of the macromolecular
helicity. This procedure was further repeated, but no change in
the CD spectrum of the poly-1a was observed.
A standard solution of poly-1a-(R)-2 complex (the amount
of (R)-2 is 0.01 equiv of monomer units of poly-1a which
corresponds to 0.1% (R)-2 based on the initial amount of the
(R)-2) was then prepared, and the solution was also injected
into the same SEC system using the same mobile phase. The
peak due to 0.1% (R)-2 was clearly detected, whereas, in the
chromatogram of the fractionated poly-1a solution, the peak
due to (R)-2 was hardly detected (see Figure S-6 in the
Supporting Information). These results indicate that the fraction-
ated poly-1a solution exhibiting the intense ICD scarcely
contains (R)-2 (less than 0.01% based on the detection limit).
Moreover, we found that the addition of (R)-2 ([(R)-2]/[poly-
1a] ) 1) to a solution of poly-1a-5 complex ([5]/[poly-1a] )
50) did not induce any CD at all at ambient temperature (ca.
20-22 °C) even after 9 days and at 50 °C for 14 h. We further
added 100 equiv of (R)-2 to the solution, but no CD was
observed. These results strongly support that the macromolecular
helicity of the poly-1a induced by the (R)-2 is undoubtedly
retained by the replacement with achiral 5. On the basis of the
ratio of the ICD intensity of the second Cotton ([θ]2nd) just after
the SEC fractionation to the original ICD intensity of the poly-
1a-(R)-2 complex, the memory efficiency with achiral 5 was
estimated to be 87%. When (S)-2 was used for the induction of
a one-handed helix on poly-1a, poly-1a of the opposite
macromolecular helicity was obtained.
We further investigated the macromolecular helicity memory
of the poly-1 induced by (R)-2 using a series of achiral amino
alcohols (6-13) and amines (14-25) and chiral amines 4 and
(45) It should be noted that the poly-1a fraction separated from the poly-1a-
(R)-2 complex using pure DMSO as the mobile phase showed no ICD,
indicating that the (R)-2 complexed with poly-1a was completely removed
during the SEC fractionation, and free poly-1a cannot maintain the induced
helical conformation.
Figure 6. SEC chromatogram of the separation of the poly-1a-(R)-2
complex with DMSO containing 5 (0.8 M) as the mobile phase. Each
fraction marked by brackets was collected to measure CD and UV-visible
spectra.
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4337

A R T I C L E S
Maeda et al.
Figure 7. Memory efficiencies of the macromolecular helicity of poly-1 induced by (R)-2 using a series of achiral amino alcohols (5-13) and chiral and
achiral amines (4, 14-26). Memory efficiencies (%) were estimated on the basis of the ICD values of the second Cotton effect ([θ]2nd), just after the SEC
fractionation of the poly-1-(R)-2 complex solution ([poly-1] ) 3 mg/mL, [(R)-2]/[poly-1] ) 10) in the presence of excess amino alcohols (5-12) and
amines (17, 19) ([5-12, 17, 19]/[poly-1] ) 50) or in the absence of an amino alcohol (13) and amines (4, 14-16, 18, 20-26) using 0.8 (5-8, 10, 11, 17,
19), 0.4 (9), 0.2 (12), or 0.008 M (4, 13-16, 18, 20-26) amines in DMSO as the mobile phase.
26 (Chart 2). Higher molecular weight poly-1b and poly-1c were
also used for comparison (see below).
rather high memory efficiency for poly-1a (82%), and other
primary (15-17, 20, 26) and secondary (25) amines also showed
moderate helicity memory effects on poly-1a. We anticipate
that this helicity memory process can be used to construct new
supramolecular assemblies with macromolecular helicity, for
instance the crown ethers (25, 26), which may exist in a helical
array along the poly-1 strand.
Effects of Molecular Weight of Poly-1 and Concentration
of Achiral Amines on Macromolecular Helicity Memory. The
molecular weight of poly-1 also affects the memory efficiency
as well as the helicity induction on poly-1. As shown in Figure
7, the memory efficiencies using the achiral amino alcohol 5
and amines 14 and 16 increased with an increase in the
molecular weight of poly-1 in the order poly-1a < poly-1b <
poly-1c. It is worth noting that as for the memory using the
amino alcohol 13, high molecular weight poly-1c showed only
an apparent memory effect (9%), although low molecular weight
poly-1a lost its chiral memory.
We also investigated the concentration effect of achiral amines
in the mobile phase on the memory efficiency, but we found
no significant changes in the memory efficiency in the concen-
tration ranges of 0.8-0.008 M for 5, 14, and 16 (see Figure
S-7 in the Supporting Information). On the other hand, a
remarkable concentration effect was observed for 21; in a dilute
concentration of 21 (0.008 M), 21 could not assist in the
macromolecular helicity memory at all, but at higher concentra-
tions the macromolecular helicity was memorized and its
efficiency increased with an increase in the concentration of
21 in the mobile phase (from 0 to 38%) (Figure S-7). These
results might be closely related to the difference in the affinity,
that is, the basicity of amines to poly-1. The acid-base binding
constant for benzylamine 21 with poly-1 (K ) 151 M-1) was
much smaller than those of 5 and 14. In a lower concentration
(less than 0.008 M) of weak bases such as 21, the acid-base
equilibria may shift to the free acid and base (see Scheme 1).
Once free carboxy groups (COOH) of poly-1 are generated after
the SEC fractionation of the helical poly-1 induced by (R)-2,
poly-1 could not maintain the induced helical conformation,
resulting in the decrease in the memory efficiency or complete
loss of the memory. However, at higher concentrations of the
weak achiral amines, the macromolecular helicity memory can
The results of the memory efficiency experiments are
summarized in Figure 7. The magnitude of the ICD of the
memorized poly-1a is highly dependent on the structures of the
amines and amino alcohols used. As for a series of achiral amino
alcohols (5-9), the memory efficiency was dependent on the
number of methylene groups of the amino alcohols and the
longer amino alcohols (8 and 9) showed relatively low memory
efficiencies, although their binding affinities (Ks) to a carboxy
group estimated by the 1H NMR titrations with VBA in DMSO-
d6 were not significantly different from each other (K (M-1) )
1522 ( 90 (6), 867 ( 113 (7), and 1241 ( 112 (8)).29 However,
rather surprisingly, upon complexation with analogous amino
alcohols 10 and 11, poly-1a lost its chiral memory completely
despite their high affinities to poly-1a (K (M-1) ) 470 ( 10
(10) and 774 ( 129 (11)). An aniline analogue 12 and a phenol
derivative 13 also resulted in no memory effect for the low
molecular weight poly-1a. A possible explanation for these no
memory effects will be described later.
Besides the amino alcohols, various primary amines also work
as small, achiral chaperoning molecules to assist in the memory
of the macromolecular helicity of poly-1, although the memory
efficiency is also dependent on the structure of amines used.
Less bulky chiral amines such as (S)-, (R)-, and (RS)-4 also
showed an apparent macromolecular helicity memory with
similar memory efficiencies (50-54%) independent of the
absolute configuration of 4. This clearly indicates that the
memory effect is controlled kinetically, since the (S)- and (R)-4
can induce opposite helices on the original poly-1 depending
on the configuration of 4. Nevertheless, the macromolecular
helicity of poly-1 can be memorized by the (S)- and (R)-4. As
described above, less bulky (S)- or (R)-4 could not induce a
single handed helix on the whole poly-1 chain and required a
long time for helicity induction with a one-handedness excess
(see Figure S-1B in the Supporting Information). This is
completely different from (R)-2 and (R)-3; helicity induction
power (or chiral twisting power for poly-1) of 4 is weaker than
that of 2 and 3, so that (S)- and (R)-4 just work to maintain the
helical conformation like achiral amines in this particular
memory experiment. A simple n-butylamine (14) exhibited
9
4338 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Helix Induction and Macromolecular Helicity Memory
A R T I C L E S
Figure 8. Stability of the macromolecular helicity memory. ICD intensity changes (second Cotton) of the poly-1a isolated by SEC using DMSO containing
5 (b) and 14 (O) as the mobile phase component, in DMSO at ambient temperature (A) and 50 °C (B) with time. [θ]0max represents the ICD intensity (second
Cotton) of the fractionated poly-1a with 5 and 14 just after the SEC separation.
be attained, since the equilibria move to the ion pair and free
ions formations.
14 at the high temperatures; therefore, the memory may rapidly
fade (see Figure 8B and Figure S-8 in the Supporting Informa-
tion). The achiral amine 14 is good for the short-term memory
of the macromolecular helicity, but for long-term memory, the
amino alcohols 5-9 may be better.
Weak bases such as 12, 22, and 23 showed no memory effect
in low and high (0.8 M for 22 and 23, 0.2 M for 12)
concentrations in DMSO as the mobile phase for the SEC
separation. This is because even at these amine concentrations,
a large amounts of the free carboxy groups (COOH) of poly-1
are generated after the SEC separation due to their quite low
basicity (K < 2 M-1). The phenol derivative 13 was also a poor
achiral amine for the macromolecular helicity memory of poly-
1. 13 has an acidic phenol OH group as well as a basic amino
group, so that 13 interacts from each other through an
intermolecular acid-base reaction, which must disturb the
interaction of 13 with poly-1, resulting in no memory effect
for poly-1a or a very low memory efficiency for poly-1c. This
speculation was supported by the fact that 24 derived from 13
in which the acidic phenol OH group was replaced by the
methoxy group showed relatively high memory efficiency (66%)
for poly-1c using a dilute DMSO solution of 24 (0.008 M) as
the mobile phase.
Repair and Stability of the Macromolecular Helicity
Memory. Although the memory of the macromolecular helicity
of poly-1 assisted by 5 is not perfect just after the SEC
fractionation (87-94%), we found that it was automatically
amplified (“repaired”) with time in the presence of achiral 5.
The ICD intensity of the poly-1 increased and recovered to the
original maximum values (for poly-1a, see Figure 8A). Similar
macromolecular helicity amplification (repair) of the memorized
poly-1a was also observed for the achiral amino alcohols 6-9
but not observed for the other achiral amines (for 14, see Figure
8A).
The memory with 5 lasted for an extremely long time over
2 years at ambient temperature (20-25 °C). At higher temper-
atures (50-80 °C), the poly-1a also kept its memory for at least
300 h at 50 °C and for 10 h at 80 °C, although, at higher
temperatures, poly-1a gradually decomposes or isomerizes
accompanied by a decrease in the UV-visible intensity.
n-Butylamine (14) also showed a high memory efficiency
(82-98%), but the memory steadily declined at ambient
temperature (20-25 °C) and the CD intensity showed a decrease
with time as shown in Figure 8A. At 50-80 °C, the memory
was lost rapidly with nonlinearity during the initial stage. The
initial half-lives were roughly estimated to be about 30 h, 2 h,
4 min, and 1 min at 50, 60, 70, and 80 °C, respectively. This
may be due to the fast exchange between the free and bound
To investigate the difference in the stability of the macro-
molecular helicity memory of poly-1 with 5 and 14, the hydroxy
group of 5 was converted to the methoxy group (17) and its
stability of the memorized poly-1a was examined. As shown
in Figure 7, the memory efficiency significantly decreased with
17 and the memory was lost rapidly at higher temperature (50
°C) in a way similar to that for the amine 14 (see Figure S-9 in
the Supporting Information). The cooperative hydrogen bond
formation of the hydroxy group of 5 to a carboxy residue of
poly-1 as well as the acid-base interaction46 must contribute
to the good memory of the one-handed helical conformation of
poly-1 with respect to efficiency and stability. We postulate that
during the exchange reaction between the bound (R)-2 and 5, 5
comes into a chiral binding site (carboxy groups of poly-1) with
a chiral gauche-staggered conformation so as to keep the helical
conformation with the same helix-sense induced by the (R)-2.
The bound 5 might have a chiral conformation similar to that
of the bound (R)-3 which can induce the same helix-sense of
poly-1 (Figure 9A,B). Amino alcohols might have a strong
ability to maintain the helical conformation after the exchange
reaction (strong memory effect or holding power), and therefore,
the memory lasted for an extremely long time even at higher
temperatures.
The memory of macromolecular helicity should be governed
kinetically, and therefore, the helical conformation induced by
(R)-2 may be changed to some extent through the exchange
process with 5. The conformation formed immediately after the
exchange reaction between the (R)-2 and 5 may be conforma-
tionally unstable. However, if the bound 5 could have a chiral
conformation, which could act as an optically active amino
alcohol like (R)- or (S)-3, the kinetically formed unstable helical
poly-1 will be repaired to a stable one with an increase in the
CD intensity. The primary amine 14 may not be able to have
such a chiral conformation as 5, and helicity repair cannot be
possible.
In connection with the chiral conformation of 5 described
above, the reason an achiral amino alcohol 11 did not show
(46) Manabe, K.; Okamura, K.; Date, T.; Koga, K. J. Am. Chem. Soc. 1993,
115, 5324-5325.
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4339

A R T I C L E S
Maeda et al.
contribution to helicity memory, since 16 with similar bulkiness
showed high memory efficiency (67%).
Achiral amines having no hydroxy group cannot exist as such
a chiral conformation as proposed for the achiral amino alcohol
5 (Figure 9B); the conformation of the amines bound to poly-1
may be close to the anti-staggered form rather than to the
gauche-staggered one. However, achiral amines such as 14 also
nicely work as small, achiral chaperoning molecules to assist
in the memory of the macromolecular helicity, which indicates
that such a chiral conformation, like the amino alcohol 5, is
not necessarily required for a good-memory effect but may be
essential for long-term memory and repair.
Mechanism of the Macromolecular Helicity Memory. The
question is raised regarding what the main driving force is for
the memory of the helical chirality of poly-1 in DMSO despite
the absence of chiral amines. The steric effect of the chaperoning
molecules (amines and amino alcohols) appears not to be the
central factor for the memory effect because the helical chirality
of poly-1 can be maintained independent of the bulkiness of
the achiral amines. As described above, in the presence of
amines poly-1 forms an ion pair complex, which at least partly
dissociates into the free ions (carboxylate and ammonium ions)
in DMSO; therefore, the intramolecular electrostatic repulsions
between the carboxylate groups of poly-1 might play an
important role for the maintenance of the helical chirality of
poly-1. To confirm if the electrostatic repulsive interactions
could really contribute to the macromolecular helicity memory,
a common salt (hydrochloride of achiral amines), which is
expected to suppress the dissociation of the ion pair to the free
ions, was added to the memorized poly-1-achiral amine
solutions after fractionation by SEC. The stabilities of the
induced memory were followed by CD, and the results were
compared with that in the absence of the common salts (Figure
10).
The ICD intensity of the poly-1c-5 solution fractionated by
SEC increased with time at the initial stage and did not change
for a long time as demonstrated above. However, upon the
addition of the hydrochloride of 5 (5‚HCl) to the same solution,
the ICD intensity significantly decreased with time and the
decreasing rate was accelerated with increasing the amount of
5‚HCl, although the changes in the ICD intensity after the
addition of 5‚HCl could not be followed completely because
of precipitation of the polymer (Figure 10A). The remarkable
repair of macromolecular helicity was not observed in the
presence of the salt. Similar rapid decrease in the ICD intensity
was also observed for the fractionated poly-1c-14 solution by
the addition of 14‚HCl, and the decreasing rate was much faster
as compared with that for the poly-1c-5 solution under the same
concentrations of the salts (Figure 10B). In the presence of salts,
the intramolecular electrostatic repulsion between the carboxy-
late ions of poly-1 is weakened so that the atropisomerization
of poly-1 easily proceeds when the dissociation of the ion pairs
into the free ions is suppressed by the common salt. The
importance of the intramolecular electrostatic repulsions for the
memory effect is also evidenced by the results of the dilution
experiments of poly-1 by method 2 as described above (see
Figure 3); once a one-handed helix was induced on poly-1 with
chiral amines, the induced helical chirality of poly-1 was
maintained in a highly dilute solution in which the amount of
the free ions increases. Consequently, the intramolecular
Figure 9. Possible chiral conformations of chiral 3 (A) and achiral 5 (B)
and 11 (C).
any memory effect can be explained as follows (Figure 9C);
the amine has two enantiotopic hydroxy groups and may have
two chiral conformations such as (S)-3 and (R)-3. When binding
with poly-1, each enantiomeric conformer can interact with
poly-1 like chiral amino alcohols such as (S)- and (R)-3;
therefore, the helical chirality of poly-1 induced by (R)-2 could
not be memorized, since each conformer of 11 might show a
strong chiral twisting power to force the poly-1 into a racemic
helical conformation. The two hydroxy groups of 11 were then
converted to the methoxy group (19), which, however, showed
a weak memory effect (5%) (Figure 7). The methoxy groups
may also participate in the formation of hydrogen bonding with
the carboxy groups of poly-1.
In the same way, chiral gauche-staggered conformers may
be possible for achiral 10. If this is the case, the helical
conformation with the same helix-sense induced by the (R)-2
may be able to be memorized when 10 forms such a chiral
conformation during the exchange reaction with (R)-2. However,
10 did not show any memory effect at all. The reason seems to
be difficult to explain at the present stage, but the fact that the
methoxy derivative 18 exhibited moderate memory efficiency
(27%) indicates that the hydroxy group of 10 makes a negative
9
4340 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Helix Induction and Macromolecular Helicity Memory
A R T I C L E S
Figure 10. Changes in the ICD intensities (second Cotton) of the isolated poly-1c having macromolecular helicity memory assisted by 5 and 14 upon
addition of 5‚HCl (A) and 14‚HCl (B), respectively, in DMSO at ambient temperature. The concentrations of 5‚HCl and 14‚HCl were 0 ([), 3 (2), 4 (b),
and 5 (9) mM.
Figure 11. Schematic illustration of the mechanism of macromolecular helicity memory of poly-1 with 5 (A) and 14 (B).
electrostatic repulsions between the carboxylate groups play the
central role for the maintenance of the helical chirality of poly-1
to prevent the atropisomerization of poly-1 as illustrated in
Figure 11. Upon complexation with amines in DMSO, poly-1
becomes more stiff with a longer persistent length (from 4.2 to
8.6 nm),16 which results in the difficulty of the helix reversal
in the poly-1 chains. This structural change observed in poly-1
also contributes more or less to the memory effect. In the case
of amino alcohols (Figure 11A, 5 for example), the hydrogen
bond formation of the hydroxy group to a carboxy residue of
poly-1 must contribute to the maintenance of the helical chirality
of poly-1 and also for the fixation of the chiral conformation
of the bound amino alcohols at the same time. These cooperative
functions of the achiral amino alcohols lead to the good long-
term memory of the one-handed helical conformation of poly-1
and also to the repair of helical chirality. The longer amino
alcohols (8 and 9) may not be able to interact cooperatively
and showed relatively low memory efficiency compared to that
of 5-7 (Figure 7). A similar model can be possible for primary
amines (Figure 11B, 14 for example), and again the electrostatic
repulsion between the neighboring carboxylate groups can be
ascribed to the memory effect. Since the amines have no
hydroxy groups and exist as achiral conformations, their
complexes with the isolated poly-1 may not be able to be
repaired but can keep the helical chirality for a rather short time
as compared with amino alcohols.
The most important conclusions drawn from the present
studies are (1) the ion pairing of poly-1 with chiral amines is
essential for the helicity induction and (2) the electrostatic
repulsion between the carboxylate groups of poly-1 derived from
the dissociation of the ion pairs plays a central role for the
macromolecular helicity memory of poly-1 in DMSO. The
memory of helicity process must be governed kinetically; the
relative rate of the exchange reaction of the bound (R)-2 with
achiral amines and the atropisomerization process of poly-1
might be responsible for the memory efficiency. Since poly-1
and chiral amines can be complexed more efficiently through
ion pairing in the presence of the common salt of the chiral
amines, memory efficiency may be expected to improve when
the replacement of the chiral amine with achiral amines is
performed in the presence of the hydrochloride of the chiral
amine. However, after the exchange reaction, the coexistence
of the common salt prevents the maintenance of the helical
chirality of poly-1 as shown in Figure 10, and therefore, the
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4341

A R T I C L E S
Maeda et al.
(R)-2 fractions were collected. The amount of the recovered (R)-2 was
estimated on the basis of the UV spectrum of the (R)-2 fraction using
the ꢀ of (R)-2 (ꢀ284 ) 6150 or ꢀ300 ) 2520 cm-1 M-1). The excess
amino alcohols ([amino alcohol]/[poly-1] ) 50) were added to the
solution of poly-1-(R)-2 complex before the SEC fractionation unless
otherwise stated. The CD spectra of the fractionated poly-1s were
measured in 2-, 4-, or 5-mm quartz cell. The addition of excess amines
or amino alcohols to the poly-1a-(R)-2 complex before the SEC
fractionation is not necessary for the macromolecular helicity memory,
but it affects the memory efficiency depending on the structures of the
amines and amino alcohols. For instance, the addition of excess amino
alcohols 5-9 ([5-9]/[poly-1a] ) 50) before the SEC fractionation
tended to increase the memory efficiency, while the achiral amine 14
caused a slight decrease in the memory efficiency (from 82 to 68%).
Therefore, achiral amino alcohols (5-12) ([5-12]/[poly-1a] ) 50)
except for 13 were added before the SEC fractionation, but an achiral
amino alcohol 13 and other achiral and chiral amines except for 17
and 19 were not added to the solution of the poly-1a-(R)-2 complex
prior to SEC fractionation; the amine 17 was added before the SEC
fractionation to compare its memory efficiency with an analogous amino
alcohol 5. The concentration of achiral amines and amino alcohols in
the mobile phase influences the memory efficiency, but we found no
significant changes in the memory efficiency for a typical amino alcohol
(5) and amine (14) in the concentration ranges of 0.8-0.016 M for 5
and 0.8-0.008 M for 14.
common salt should be removed by SEC as soon as possible
just after the exchange reaction to attain high memory efficiency.
To examine this possibility, we carried out the following
experiments. The poly-1c-(R)-2 complexes containing various
amount of (R)-2‚HCl in DMSO were prepared, all of which
showed a full ICD ([θ]2nd ) ca. -3.1 × 104) regardless of the
amount of (R)-2‚HCl. Each sample was then injected into the
SEC system using a DMSO solution containing 5 (0.008 M) or
16 (0.008 M) as the mobile phase to remove (R)-2 and (R)-2‚
HCl simultaneously. The poly-1c was isolated in the same way
as described above. After the SEC fractionation, CD spectra of
the fractionated poly-1c were taken and the memory efficiencies
were calculated. By this method, the (R)-2 complexed with poly-
1c can be completely replaced with the achiral amines in the
presence of (R)-2‚HCl, and the common salt as well as the (R)-2
can also be removed at once during the SEC fractionation. The
memory efficiency markedly increased with increasing (R)-2‚
HCl concentration. The maximum memory efficiencies im-
proved from 68 to 99% for 5 and from 57 to 79% for 16 under
the present conditions (see Figure S-10 in the Supporting
Information). These values were higher than those in the absence
of (R)-2‚HCl by a factor of ca. 1.5 and 1.4, respectively.
Especially, in the case of 5, almost perfect memory could be
achieved at [(R)-2‚HCl] ) 10 mM.
Acknowledgment. This work was partially supported by a
Grant-in-Aid for Scientific Research from the Japan Society for
the Promotion of Science, the Ministry of Education, Culture,
Sports, Science, and Technology, Japan, and also PRESTO of
JST (E.Y.). K.M. thanks the Inamori Foundation for financial
support. This paper is dedicated to the memory of the late
Professor Emeritus Kousaku Maruyama of Chiba University and
Research Mentor of PRESTO (Form and Function), JST.
The remarkable procedure developed in this study will be
useful to attain similar macromolecular helicity memory with
perfect memory efficiency in other induced helical polymers.
Experimental Section
Full experimental details are available in the Supporting Information.
Materials. Three cis-transoidal poly-1s with different molecular
weights were synthesized according to the previously reported
method.13b,22,29 The molecular weights (Mn) of poly-1s were estimated
as its methyl ester by SEC with polystyrene standards using tetrahy-
drofuran (THF) or chloroform as the eluent; Mn and Mw/Mn ) 1.7 ×
104 and 2.1 (poly-1a), 3.3 × 104 and 2.8 (poly-1b), and 13.0 × 104
and 4.3 (poly-1c), respectively. Conversion of poly-1s into the methyl
esters was carried out using diazomethane in ether solution or
(trimethylsilyl)diazomethane in hexane solution according to the method
reported previously.13b,22
Memory of Macromolecular Helicity: SEC Fractionation of
Poly-1. Stock solutions of poly-1 (6 mg/mL, 41 mM) and (R)-2 (0.41
M) were prepared. A 400-µL aliquot of the stock solution of poly-1
was transferred to a vessel equipped with a screwcap using a Hamilton
microsyringe, and to this was added 400 µL of the stock solution of
(R)-2; the initial CD spectrum was taken using a 0.01-cm quartz cell.
SEC fractionation was performed using a Jasco PU-980 liquid chro-
matograph equipped with a UV (300 nm; Jasco UV-970) detector. A
Shodex KF-806L SEC column was connected, and DMSO solutions
containing appropriate amounts of amines were used as the mobile phase
at a flow rate of 1.0 mL/min. A 100 µL of the solution of the poly-
1-(R)-2 complex was injected to the SEC system, and the poly-1 and
Supporting Information Available: Full experimental details,
time-dependent ICD changes of poly-1 with (R)-3 and (S)-4,
plots of IR titration data of poly-1 and VBA with (R)-2, (R)-3,
and (S)-4, CD spectral changes of poly-1-(R)-2 complex with
(S)-2 and (S)-3, ICD intensity changes of poly-1-(R)-3 complex
with (S)-3 with time and their CD spectral changes, CD spectral
changes of poly-1-(R)-2 complex by the addition of (RS)-3,
SEC chromatograms of the refractionated poly-1 and the poly-1
solution containing 0.01 equiv of (R)-2, concentration effects
of achiral amines on memory efficiency, CD intensity changes
of the fractionated poly-1-14 complex at various temperatures
and the fractionated poly-1-5, -14, and -17 complexes at 50
°C, and effect of (R)-2‚HCl concentration on memory efficiency
of the macromolecular helicity of poly-1c induced by (R)-2 in
the presence of (R)-2‚HCl (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
JA0318378
9
4342 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004