Mechanism of helix induction in poly(4-carboxyphenyl isocyanide) with chiral amines and memory of the macromolecular helicity and its helical structures

Article abstract of DOI:10.1021/ja904128d

An optically inactive poly(4-carboxyphenyl isocyanide) (poly-1-H) changed its structure into the prevailing, one-handed helical structure upon complexation with optically active amines in dimethylsulfoxide (DMSO) and water, and the complexes show a charac

Full text of DOI:10.1021/ja904128d

Published on Web 07/06/2009
Mechanism of Helix Induction in Poly(4-carboxyphenyl
isocyanide) with Chiral Amines and Memory of the
Macromolecular Helicity and Its Helical Structures
Yoko Hase,† Kanji Nagai,†,‡ Hiroki Iida,† Katsuhiro Maeda,† Noriaki Ochi,†
Kyoichi Sawabe,† Koichi Sakajiri,‡,| Kento Okoshi,*,‡,⊥ and Eiji Yashima*,†,‡
Department of Molecular Design and Engineering, Graduate School of Engineering, Nagoya
UniVersity, Chikusa-ku, Nagoya 464-8603, Japan, and Yashima Super-structured Helix Project,
Exploratory Research for AdVanced Technology (ERATO), Japan Science and Technology
Agency (JST), Japan
Received May 21, 2009; E-mail: kokoshi@polymer.titech.ac.jp; yashima@apchem.nagoya-u.ac.jp
Abstract: An optically inactive poly(4-carboxyphenyl isocyanide) (poly-1-H) changed its structure into the
prevailing, one-handed helical structure upon complexation with optically active amines in dimethylsulfoxide
(DMSO) and water, and the complexes show a characteristic induced circular dichroism in the polymer
backbone region. Moreover, the macromolecular helicity induced in water and aqueous organic solutions
containing more than 50 vol % water could be “memorized” even after complete removal of the chiral
amines (h-poly-1b-H), while that induced in DMSO and DMSO-water mixtures containing less than 30 vol
% water could not maintain the optical activity after removal of the chiral amines (poly-1a-H). We now
report fully detailed studies of the helix induction mechanism with chiral amines and the memory of the
macromolecular helicity in water and a DMSO-water mixture by various spectroscopic measurements,
theoretical calculations, and persistence length measurements together with X-ray diffraction (XRD)
measurements. From the spectroscopic results, such as circular dichroism (CD), absorption, IR, vibrational
CD, and NMR of poly-1a-H, h-poly-1b-H, and original poly-1-H, we concluded that the specific configurational
isomerization around the CdN double bonds occurs during the helicity induction process in each solvent.
In order to obtain the structural information, XRD measurements were done on the uniaxially oriented films
of the corresponding methyl esters (poly-1-Me, poly-1a-Me, and h-poly-1b-Me) prepared from their liquid
crystalline polymer solutions. On the basis of the XRD analyses, the most plausible helical structure of
poly-1a-Me was proposed to be a 9-unit/5-turn helix with two monomer units as a repeating unit, and that
of h-poly-1b-Me was proposed to be a 10-unit/3-turn helix consisting of one repeating monomer unit. The
density functional theory calculations of poly(phenyl isocyanide), a model polymer of h-poly-1b-Me, afforded
a 7-unit/2-turn helix as the most possible helical structure, which is in good agreement with the XRD results.
Furthermore, the persistence length measurements revealed that these structural changes accompany a
significant change in the main-chain stiffness. The mechanism of helix induction in poly-1-H and the memory
of the macromolecular helicity are discussed on the basis of these results.
Introduction
of helical polymers have been synthesized, the determination
of their helical structures, including the helical pitch and
handedness (right- or left-handed helix), remains very difficult
but is quite important in order to develop more sophisticated
chiral materials based on their macromolecular helicity.1b,f,i
Optically active helical polymers with a controlled helix-
sense have recently attracted considerable interest not only
for mimicking biological helices but also for developing novel
chiral materials.1 In general, fully synthetic helical polymers
can be prepared either by the polymerization of optically
active monomers, such as isocyanates,1a,c silanes,1d and
acetylenes,1h,j-q or by the helix-sense-selective polymeriza-
tion of achiral or prochiral methacrylates,1f isocyanides,2 and
carbodiimides3 bearing bulky substituents as the pendant
groups using chiral catalysts or initiators. Although a number
Polyisocyanides are among the most extensively studied
helical polymers since the pioneering research by Millich,
Drenth, and Nolte.2,4 The helical structure of polyisocyanides
is considered to be a 4-unit/1-turn (4/1) helix when they have
a bulky side group,2c which was first postulated by Millich
on the basis of an X-ray analysis2b and later proposed
experimentally by Nolte5 and theoretically by Hoffmann.6
The direct chromatographic resolution of poly(tert-butyl
isocyanide) into enantiomeric helices clearly suggested a
stable helical structure of polyisocyanides.5,7 The resolved
polymer with a positive rotation was postulated to have a
left-handed helical conformation on the basis of a circular
dichroism (CD) spectral analysis.7 The stable helical con-
† Nagoya University.
‡ ERATO, JST.
| Present address: Department of Chemistry, Faculty of Engineering, Gifu
University, Yanagido, Gifu 501-1193, Japan.
⊥ Present address: Department of Organic and Polymeric Materials,
Graduate School of Science and Engineering, Tokyo Institute of Technology,
2-12-1, Ookayama, Meguro-ku, Tokyo 152-8552, Japan.
9
10.1021/ja904128d CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 10719–10732 10719

A R T I C L E S
Hase et al.
formation of the polyisocyanides in solution was further
proved by the helix-sense-selective polymerization of achiral
bulky isocyanides by Nolte8 and Novak.9 Since then, wide
varieties of helical polyisocyanides with functional pendant
groups have been synthesized, and some of them have been
used for optoelectronic and liquid crystalline materials.2,10
However, Green11 and Salvadori12 claimed that polyisocya-
nides bearing less bulky aliphatic, aralkyl, and aromatic pendant
groups might not have a regular 4/1 helical conformation
because of a possible complicated configurational isomerism
(syn-anti isomerization) around the CdN double bond and s-cis
and s-trans conformational isomerism around the CsC bond
of the main-chain. In addition, Euler and Rosen13 experimentally
showed a slow interconversion of poly(phenyl isocyanide) from
an “as-prepared” 4/1 helical conformation to a more extended
s-trans, zigzag conformation; an analogous s-trans structure as
one of the most stable structures was also theoretically proposed
by Clericuzio and Salvadori.14 As a consequence, helical
polyisocyanidessin particular, helical poly(phenyl isocya-
nide)sslack unquestionable direct evidence for their helical
structures.15
Recently, we reported that a helical polyisocyanide with a
controlled helix-sense could be produced on the basis of the
noncovalent “helicity induction and chiral memory effect”.1h,k,o,p,16
An optically inactive poly(4-carboxyphenyl isocyanide) (poly-
1-H) and its sodium salt (poly-1-Na) were found to form a
preferred-handed helical conformation upon noncovalent com-
plexation with chiral amines in water.17 The induced helix
remained after complete removal of the chiral amines, and
further modifications of the pendant groups to esters and amide
residues were possible without loss of the macromolecular
(10) For liquid crystalline polyisocyanides: (a) 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. (b) Cornelissen, J. J. L. M.; Granswinckel, W. S.;
Rowan, A. E.; Sommerdijk, N. A. J. M.; Nolte, R. J. M. J. Polym.
Sci., Part A: Polym. Chem. 2003, 41, 1725–1736. (c) Tian, Y.; Kamata,
K.; Yoshida, H.; Iyoda, T. Chem.-Eur. J. 2006, 12, 584–591. (d)
Kajitani, T.; Okoshi, K.; Sakurai, S.-i.; Kumaki, J.; Yashima, E. J. Am.
Chem. Soc. 2006, 128, 708–709. (e) Wezenberg, S. J.; Metselaar,
G. A.; Rowan, A. E.; Cornelissen, J. J. L. M.; Seebach, D.; Nolte,
R. J. M. Chem.-Eur. J. 2006, 12, 2778–2786. (f) Metselaar, G. A.;
Adams, P. J. H. M.; Nolte, R. J. M.; Cornelissen, J. J. L. M.; Rowan,
A. E. Chem.-Eur. J. 2007, 13, 950–960. (g) Metselaar, G. A.;
Wezenberg, S. J.; Cornelissen, J. J. L. M.; Nolte, R. J. M.; Rowan,
A. E. J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 981–988. (h)
Onouchi, H.; Okoshi, K.; Kajitani, T.; Sakurai, S.-i.; Nagai, K.;
Kumaki, J.; Onitsuka, K.; Yashima, E. J. Am. Chem. Soc. 2008, 130,
229–236. (i) Kajitani, T.; Lin, H.; Nagai, K.; Okoshi, K.; Onouchi,
H.; Yashima, E. Macromolecules 2009, 42, 560–567. For optoelec-
tronic polyisocyanides, see: (j) Kauranen, K.; Verbiest, T.; Boutton,
C.; Teerenstra, M. N.; Schouten, A. J.; Nolte, R. J. M.; Persoons, A.
Science 1995, 270, 966–969. (k) Hida, N.; Takei, F.; Onitsuka, K.;
Shiga, K.; Asaoka, S.; Iyoda, T.; Takahashi, S. Angew. Chem., Int.
Ed. 2003, 42, 4349–4352. For other leading references on helical
polyisocyanides prepared by the polymerization of the corresponding
optically active monomers, see: (l) Cornelissen, J. J. L. M.; Fischer,
M.; Sommerdijk, N. A. J. M.; Nolte, R. J. M. Science 1998, 280, 1427–
1430. (m) Amabilino, D. B.; Ramos, E.; Serrano, J.-L.; Sierra, T.;
Veciana, J. J. Am. Chem. Soc. 1998, 120, 9126–9134. (n) Amabilino,
D. B.; Ramos, E.; Serrano, J.-L.; Veciana, J. AdV. Mater. 1998, 10,
1001–1005. (o) Hasegawa, T.; Kondoh, S.; Matsuura, K.; Kobayashi,
K. Macromolecules 1999, 32, 6595–6603. (p) Takei, F.; Yanai, K.;
Onitsuka, K.; Takahashi, S. Chem.-Eur. J. 2000, 6, 983–993. (q)
Cornelissen, J. J. L. M.; Graswinckel, W. S.; Adams, P. J. H. M.;
Nachtegaal, G. H.; Kentgens, A. P. M.; Sommerdijk, N. A. J. M.;
Nolte, R. J. M. J. Polym. Sci., Part A: Polym. Chem. 2001, 39, 4255–
4264. (r) Yamada, Y.; Kawai, T.; Abe, J.; Iyoda, T. J. Polym. Sci.,
Part A: Polym. Chem. 2002, 40, 399–408. (s) Takei, F.; Onitsuka,
K.; Takahashi, S. Macromolecules 2005, 38, 1513–1516. (t) Kajitani,
T.; Okoshi, K.; Yashima, E. Macromolecules 2008, 41, 1601–1611.
(11) Green, M. M.; Gross, R. A.; Schilling, F. C.; Zero, K.; Crosby, C.
Macromolecules 1988, 21, 1839–1846.
(1) Reviews: (a) Green, M. M.; Peterson, N. C.; Sato, T.; Teramoto, A.;
Cook, R.; Lifson, S. Science 1995, 268, 1860–1865. (b) Okamoto,
Y.; Yashima, E. Angew. Chem., Int. Ed. 1998, 37, 1020–1043. (c)
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. (d) Fujiki, M. Macromol. Rapid Commun. 2001, 22, 539–563.
(e) Hill, D. J.; Mio, M. J.; Prince, R. B.; Hughes, T. S.; Moore, J. S.
Chem. ReV. 2001, 101, 3893–4011. (f) Nakano, T.; Okamoto, Y. Chem.
ReV. 2001, 101, 4013–4038. (g) Brunsveld, L.; Folmer, B. J. B.; Meijer,
E. W.; Sijbesma, R. P. Chem. ReV. 2001, 101, 4071–4097. (h)
Yashima, E.; Maeda, K.; Nishimura, T. Chem.-Eur. J. 2004, 10, 42–
51. (i) Reggelin, M.; Doerr, S.; Klussmann, M.; Schultz, M.; Holbach,
M. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5461–5466. (j) Lam,
J. W. Y.; Tang, B. Z. Acc. Chem. Res. 2005, 38, 745–754. (k) Maeda,
K.; Yashima, E. Top. Curr. Chem. 2006, 265, 47–88. (l) Aoki, T.;
Kaneko, T.; Teraguchi, M. Polymer 2006, 47, 4867–4892. (m) Masuda,
T. J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 165–180. (n) Rudick,
J. G.; Percec, V. New J. Chem. 2007, 31, 1083–1096. (o) Yashima,
E.; Maeda, K. Macromolecules 2008, 41, 3–12. (p) Yashima, E.;
Maeda, K.; Furusho, Y. Acc. Chem. Res. 2008, 41, 1166–1180. (q)
Rudick, J. G.; Percec, V. Acc. Chem. Res. 2008, 41, 1641–1652. (r)
Pijper, D.; Feringa, B. L. Soft Matter 2008, 4, 1349–1372. (s) Kim,
H.-J.; Lim, Y.-B.; Lee, M. J. Polym Sci., Part A: Polym. Chem. 2008,
46, 1925–1935.
(2) Reviews: (a) Millich, F. Chem. ReV. 1972, 72, 101–113. (b) Millich,
F. AdV. Polym. Sci. 1975, 19, 117–141. (c) Drenth, W.; Nolte, R. J. M.
Acc. Chem. Res. 1979, 12, 30–35. (d) Millich, F. J. Polym. Sci.
Macromol. ReV. 1980, 15, 207–253. (e) Nolte, R. J. M. Chem. Soc.
ReV. 1994, 23, 11–19. (f) Takahashi, S.; Onitsuka, K.; Takei, F. Proc.
Jpn. Acad. Ser. B 1998, 74, 25–30. (g) Cornelissen, J. J. L. M.; Rowan,
A. E.; Nolte, R. J. M.; Sommerdijk, N. A. J. M. Chem. ReV. 2001,
101, 4039–4070. (h) Suginome, M.; Ito, Y. AdV. Polym. Sci. 2004,
171, 77–136. (i) Amabilino, D. B.; Serrano, J.-L.; Sierra, T.; Veciana,
J. J. Polym. Sci., Part A: Polym. Chem. 2006, 44, 3161–3174.
(3) (a) Tian, G.; Lu, Y.; Novak, B. M. J. Am. Chem. Soc. 2004, 126,
4082–4083. (b) Tang, H.-Z.; Novak, B. M.; He, J.; Polavarapu, P. L.
Angew. Chem., Int. Ed. 2005, 44, 7298–7301. (c) Tang, H.-Z.; Boyle,
P. D.; Novak, B. M. J. Am. Chem. Soc. 2005, 127, 2136–2142.
(4) (a) Millich, F.; Sinclair, R. G. J. Polym. Sci., Part C 1968, 22, 33–43.
(b) Nolte, R. J. M.; Zwikker, J. W.; Reedijk, J.; Drenth, W. J. Mol.
Catal. 1978, 4, 423–426. (c) van Beijnen, A. J. M.; Nolte, R. J. M.;
Zwikker, J. W.; Drenth, W. J. Mol. Catal. 1978, 4, 427–432.
(5) Nolte, R. J. M.; van Beijnen, A. J. M.; Drenth, W. J. Am. Chem. Soc.
1974, 96, 5932–5933.
(12) Pini, D.; Iuliano, A.; Salvadori, P. Macromolecules 1992, 25, 6059–
6062.
(13) (a) Spencer, L.; Kim, M.; Euler, W. B.; Rosen, W. J. Am. Chem. Soc.
1997, 119, 8129–8130. (b) Euler, W. B.; Huang, J-.T.; Kim, M.;
Spencer, L.; Rosen, W. Chem. Commun. 1997, 257–258. (c) Huang,
J.-T.; Sun, J.; Euler, W. B.; Rosen, W. J. Polym. Sci., Part A, Polym.
Chem. 1997, 35, 439–446. (d) Spencer, L.; Euler, W. B.; Traficante,
D. D.; Kim, M.; Rosen, W. Magn. Reson. Chem. 1998, 36, 398–402.
(14) (a) Clericuzio, M.; Alagona, G.; Ghio, C.; Salvadori, P. J. Am. Chem.
Soc. 1997, 119, 1059–1071. Molecular orbital calculations of poly-
(iminomethylene) using semi-empirical methods were also conducted.
See: (b) Young, V. Y.; Hellmuth, E. W. J. Mol. Struct.: THEOCHEM
2002, 578, 1–17.
(15) Recently, we have reported the helix-sense-selective living polymer-
ization of an optically active phenyl isocyanide bearing an L-alanine
pendant with a long decyl chain with the achiral µ-ethynediyl Pt-Pd
catalyst that unprecedentedly produced both right- and left-handed
helical, rigid-rod polyisocyanides at once with a different molecular
weight and a narrow molecular weight distribution, which could be
separated into each helix in a facile fractionation with acetone. The
structures of both right- and left-handed helical polyisocyanides were
proposed to be a 15-unit/4-turn (15/4) helix on the basis of X-ray
analyses of the corresponding oriented films.10h However, the detailed
structural analysis, including a possible configurational and confor-
mational isomerism of the main-chain, has never been pursued.
(6) Kollmar, C.; Hoffmann, R. J. Am. Chem. Soc. 1990, 112, 8230–8238.
(7) van Beijnen, A. J. M.; Nolte, R. J. M.; Drenth, W.; Hezemans, A. M. F.
Tetrahedron 1976, 32, 2017–2019.
(8) (a) Kamer, P. C. J.; Nolte, R. J. M.; Drenth, W. J. Chem. Soc., Chem.
Commun. 1986, 1789–1791. (b) Kamer, P. C. J.; Nolte, R. J. M.;
Drenth, W. J. Am. Chem. Soc. 1988, 110, 6818–6825.
(9) Deming, T. J.; Novak, B. M. J. Am. Chem. Soc. 1992, 114, 7926–
7927.
9
10720 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009

Helical Structures of Poly(phenyl isocyanide) Derivatives
A R T I C L E S
helicity memory.18 However, the helical structures of poly-1-H
induced by chiral amines in water and aqueous organic solutions
containing more than 50 vol % water can be effectively
memorized after complete removal of the chiral amines (h-poly-
1b-H), while the helical structures of poly-1-H induced by chiral
amines in dimethylsulfoxide (DMSO)19 and DMSO-water
mixtures containing less than 30 vol % water cannot maintain
their helicity after removal of the chiral amines (poly-1a-H;
Figure 1).20 These results raise questions as to what different
helical structures the polyisocyanide can form in DMSO and
water in the presence of chiral amines and why the helical
structure induced in DMSO cannot be memorized after removal
of the chiral amines.
To address these questions, we synthesized a series of
isotopically labeled poly-1-H’s and their methyl esters (poly-
1-Me’s) with deuterium (D), 13C, and 15N (Chart 1) and
nonlabeled poly-1-H and poly-1-Me, and their structures before
and after the helicity induction with chiral amines in water and
organic solvent-water mixtures and subsequent memory were
investigated in detail by absorption, CD, IR, vibrational CD
(VCD), and NMR spectroscopies together with X-ray diffraction
(XRD) of the oriented films prepared from nematic and
cholesteric liquid crystalline (LC) poly-1-Me’s, persistence
length measurements, and theoretical calculations. These results
Figure 1. Schematic illustration of helicity induction in poly-1-H in DMSO
and water with chiral amines (2 or 3) and after subsequent removal of the
chiral amines. The macromolecular helicity induced in water is memorized
after complete removal of the chiral amines (h-poly-1b-H), whereas that
induced in DMSO cannot be memorized (poly-1a-H).
shed light on the mechanism of helicity induction in poly-1 and
subsequent memory of the macromolecular helicity.
Results and Discussion
Synthesis and Polymerization of Isotopically Labeled
Monomers. Four novel isotopically labeled sodium 4-isocy-
anobenzoates (1-Na-d4, 1-Na-13C, 1-Na-d4-13C, and 1-Na-15N)
were prepared as outlined in Schemes S1-S4 (Supporting
Information). The structures of the obtained isocyanides were
characterized by IR, NMR, and mass spectroscopy, and they
were in fair agreement with their assigned structures (see the
Supporting Information). The polymerization of 1-Na, 1-Na-
d4, 1-Na-13C, 1-Na-d4-13C, and 1-Na-15N was performed in water
with NiCl2 ·6H2O as the catalyst as previously reported, afford-
ing the corresponding polyisocyanides in good yields, poly-1-
H, poly-1-H-d4, poly-1-H-13C, poly-1-H-d4-13C, and poly-1-H-
15N, respectively (Scheme 1A, Chart 1, and Table 1).17 The
number-average molecular weights (Mn) of their methyl esters,
as determined by size exclusion chromatography (SEC) with
polystyrene standards using chloroform (CHCl3) as the eluent,
were relatively low (2.6 × 104-3.5 × 104) (runs 1-5 in Table
1). However, when the polymerization of 4-(ethoxycarbon-
yl)phenyl isocyanide (1-Et) was performed in dry tetrahydro-
furan (THF) containing a small amount of alcohols, such as
methanol (MeOH) (THF-MeOH ) 99:1, v/v), with the same
NiCl2 ·6H2O catalyst, we found that a higher molecular weight
poly-1-Et (HMW-poly-1-Et; Mn ) 8.6 × 104) was obtained in
88% yield (run 6 in Table 1 and Scheme 1B).21 The obtained
HMW-poly-1-Et was further alkaline hydrolyzed, followed by
esterification with CH2N2 to give HMW-poly-1-H and HMW-
poly-1-Me, respectively (Scheme 1B).
(16) (a) Yashima, E.; Maeda, K.; Okamoto, Y. Nature 1999, 399, 449–
451. (b) Maeda, K.; Morino, K.; Okamoto, Y.; Sato, T.; Yashima, E.
J. Am. Chem. Soc. 2004, 126, 4329–4342. (c) Morino, K.; Watase,
N.; Maeda, K.; Yashima, E. Chem.-Eur. J. 2004, 10, 4703–4707.
(d) Onouchi, H.; Kashiwagi, D.; Hayashi, K.; Maeda, K.; Yashima,
E. Macromolecules 2004, 37, 5495–5503. (e) Onouchi, H.; Miyagawa,
T.; Furuko, A.; Maeda, K.; Yashima, E. J. Am. Chem. Soc. 2005, 127,
2960–2965. (f) Miyagawa, T.; Furuko, A.; Maeda, K.; Katagiri, H.;
Furusho, Y.; Yashima, E. J. Am. Chem. Soc. 2005, 127, 5018–5019.
(g) Onouchi, H.; Hasegawa, T.; Kashiwagi, D.; Ishiguro, H.; Maeda,
K.; Yashima, E. Macromolecules 2005, 38, 8625–8633. (h) Hasegawa,
T.; Morino, K.; Tanaka, Y.; Katagiri, H.; Furusho, Y.; Yashima, E.
Macromolecules 2006, 39, 482–488. (i) Hasegawa, T.; Maeda, K.;
Ishiguro, H.; Yashima, E. Polym. J. 2006, 38, 912–919. (j) Kawauchi,
T.; Kumaki, J.; Kitaura, A.; Okoshi, K.; Kusanagi, H.; Kobayashi,
K.; Sugai, T.; Shinohara, H.; Yashima, E. Angew. Chem., Int. Ed. 2008,
47, 515–519. (k) Kawauchi, T.; Kitaura, A.; Kumaki, J.; Kusanagi,
H.; Yashima, E. J. Am. Chem. Soc. 2008, 130, 11889–11891. For chiral
memory effect of other macromolecular and supramolecular systems,
see: (l) Furusho, Y.; Kimura, T.; Mizuno, Y.; Aida, T. J. Am. Chem.
Soc. 1997, 119, 5267–5268. (m) Mizuno, T.; Takeuchi, M.; Hamachi,
I.; Nakashima, K.; Shinkai, S. J. Chem. Soc., Perkin Trans. 2 1998,
2281–2288. (n) Bellacchio, E.; Lauceri, R.; Gurrieri, S.; Scolaro, L. M.;
Romeo, A.; Purrello, R. J. Am. Chem. Soc. 1998, 120, 12353–12354.
(o) Prins, L. J.; Jong, F. D.; Timmerman, P.; Reinhoudt, D. N. Nature
2000, 408, 181–184. (p) Kubo, Y.; Ohno, T.; Yamanaka, J.-i.; Tokita,
S.; Iida, T.; Ishimaru, Y. J. Am. Chem. Soc. 2001, 123, 12700–12701.
(q) Lauceri, R.; Raudino, A.; Scolaro, L. M.; Micali, N.; Purrello, R.
J. Am. Chem. Soc. 2002, 124, 894–895. (r) Mammana, A.; D’Urso,
A.; Lauceri, R.; Purrello, R. J. Am. Chem. Soc. 2007, 129, 8062–
8063. (s) Rosaria, L.; D’Urso, A.; Mammana, A.; Purrello, R. Chirality
2008, 20, 411–419. (t) Ousaka, N.; Inai, Y.; Kuroda, R. J. Am. Chem.
Soc. 2008, 130, 12266–12267.
(17) Ishikawa, M.; Maeda, K.; Mitsutsuji, Y.; Yashima, E. J. Am. Chem.
Soc. 2004, 126, 732–733.
Macromolecular Helicity Induction in Poly-1-H with Chiral
Amines and Isolation of the Polymers. Poly-1-H and the isoto-
pically labeled poly-1-H’s (poly-1-H-d4, poly-1-H-13C, poly-
1-H-d4-13C, and poly-1-H-15N) were then annealed with
(1R,2S)-2 in DMSO-water (9:1 or 4:1, v/v) and (R)- or (S)-3
in water, acetonitrile (CH3CN)-water (1:1, v/v), or 2-propanol-
water (1:1, v/v) at 50 °C for 3-72 days, as reported
previously.17,20 Figure 2 shows the typical CD and absorption
spectra of the poly-1-H-(1R,2S)-2 complex in DMSO-water
(18) (a) Hase, Y.; Mitsutsuji, Y.; Ishikawa, M.; Maeda, K.; Okoshi, K.;
Yashima, E. Chem. Asian J. 2007, 2, 755–763. (b) Miyabe, T.; Hase,
Y.; Iida, H.; Maeda, K.; Yashima, E. Chirality 2009, 21, 44–50.
(19) Ishikawa, M.; Maeda, K.; Yashima, E. J. Am. Chem. Soc. 2002, 124,
7448–7458.
(20) Hase, Y.; Ishikawa, M.; Muraki, R.; Maeda, K.; Yashima, E.
Macromolecules 2006, 39, 6003–6008.
(21) The effect of MeOH on the polymerization of 1-Et with NiCl2 ·6H2O
remains obscure, but this method may be useful to obtain high-
molecular-weight polyisocyanides.
9
J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009 10721

A R T I C L E S
Hase et al.
Chart 1. Structures of Poly(phenyl isocyanide) Derivatives
Scheme 1. Synthesis of Poly(phenyl isocyanide) Derivatives
Table 1. Polymerization Results of 1-Na, 1-Na-d4, 1-Na-13C,
1-Na-d4-13C, 1-Na-15N, and 1-Et with NiCl2 ·6H2Oa
effect), while that in the main-chain imino chromophore region
(λ ) ca. 360 nm) was increased by ca. 20% (hyperchromic
effect), as compared with the values for poly-1a-H. These results
suggested intramolecular stacking of the aromatic pendant
groups of h-poly-1b-H and also a difference in their main-chain
structure, which will be discussed later in detail.
run
monomer
yield (%)b
Mn × 10-4 (Mw/Mn)c
polymer
1
2
3
4
5
6
1-Na
90
66
82
78
62
88
3.5 (3.2)
2.6 (2.4)
2.7 (2.8)
2.8 (2.5)
2.7 (5.8)
8.6 (5.0)e
poly-1-H
1-Na-d4
poly-1-H-d4
1-Na-13C
1-Na-d4-13C
1-Na-15N
1-Etd
poly-1-H-13C
poly-1-H-d4-13C
poly-1-H-15N
HMW-poly-1-H
The chiroptical properties of the poly-1-H’s after helicity
induction in different solvents followed by removal of the chiral
amines are summarized in Table 2. All the helical h-poly-1b-
H’s induced in water and CH3CN- or 2-propanol-water (1:1,
v/v) maintained their induced helical structures after complete
removal of the chiral amines, and the memory efficiencies thus
estimated on the basis of the ∆ε values of the polymers before
isolation were greater than 85%; their absolute ∆ε values at ca.
360 nm after isolation were in the range of 9.28-11.9 in alkaline
water (Table 2), indicating that the isotope effects on helicity
induction and subsequent memory of the helicity could be
negligible. The CD intensity tended to increase with an increase
in the molecular weight of h-poly-1b-H and HMW-h-poly-1b-H
(runs 2-4, 15, and 16 in Table 2). A similar molecular weight
effect on the helicity induction was also observed for the
dynamic helical poly((4-carboxyphenyl)acetylene) upon com-
plexation with chiral amines in DMSO.16b As previously
reported,17 h-poly-1b-H’s can be converted to the corresponding
methyl esters (h-poly-1b-Me’s) without loss of macromolecular
helicity. On the other hand, poly-1-H, poly-1-H-13C, poly-1-
H-15N, and HMW-poly-1-H induced in DMSO-water (9:1 or
4:1, v/v) with (1R,2S)-2 could not retain their optical activity
after removal of the chiral amine (runs 1, 7, 12, and 14 in Table
2).
a Polymerized in water under air at room temperature for 1 day.
[1-Na] ) 0.5 M; [1-Na]/[NiCl2 ·6H2O] ) 100. b MeOH-insoluble part.
c Determined by SEC (polystyrene standards) with CHCl3 as the eluent
as its methyl ester. d Polymerized in THF-MeOH (99:1, v/v) under
nitrogen at room temperature for 1 day. [1-Et] ) 0.1 M; [1-Et]/
[NiCl2 ·6H2O] ) 100. e Determined by SEC (polystyrene standards) with
CHCl3 as the eluent as its ethyl ester.
(9:1, v/v) (a and c in Figure 2A) and the poly-1-H-(R)-3
complex in CH3CN-water (1:1, v/v) (e and g in Figure 2B).
Both polymers formed a predominantly one-handed helix and
exhibited an induced CD (ICD) in the imino chromophore region
of the polymer backbones (300s450 nm), and their CD spectral
patterns and λmax of the Cotton effects were similar to each other,
although their absorption spectra (c and g in Figure 2) were
slightly different from one another. As previously reported,17,20
the helicity of poly-1-H’s (h-poly-1b-H’s) induced by (R)-3 in
water and aqueous organic solutions containing more than 50
vol % water was automatically memorized during the helicity
induction process, thus showing an intense ICD after complete
removal of the chiral amine (see f in Figure 2B and run 4 in
Table 2), whereas the optical activity induced in the poly-1-
H’s by (1R,2S)-2 in DMSO-water (9:1, v/v) almost completely
disappeared after removal of the chiral amine (poly-1a-H’s) (see
b in Figure 2A and run 1 in Table 2).20 We observed a
significant difference in the absorption spectra between h-poly-
1b-H and poly-1a-H (d and h in Figure 2 and Figure 2C); the
absorption intensity of h-poly-1b-H in the pendant aromatic
region (λmax ) 251 nm) was reduced by 18% (hypochromic
1
NMR Studies. The H, 13C, and 15N NMR spectra of the
original poly-1-H, poly-1-H-13C, and poly-1-Me-15N derived
from poly-1-H-15N and those of the corresponding helical
polymers during the helicity induction with chiral amines in
different solvents and after removal of the chiral amines were
then measured to gain insight into the changes in the structures
9
10722 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009

Helical Structures of Poly(phenyl isocyanide) Derivatives
A R T I C L E S
Figure 2. (A,B) CD and absorption spectra of poly-1-H-(1R,2S)-2 complex in DMSO-water (9:1, v/v) (run 1 in Table 2) (a,c), poly-1-H-(R)-3 complex
in CH3CN-water (1:1, v/v) (run 4 in Table 2) (e,g), and poly-1a-H (run 1 in Table 2) (b,d) and h-poly-1b-H (run 4 in Table 2) (f,h) after complete removal
of (1R,2S)-2 and (R)-3, respectively, measured in alkaline water ([polymer]/[NaOH] ) 1) at ambient temperature; [polymer] ) 1.0 mg/mL. (C) Absorption
spectra of poly-1a-H (run 1 in Table 2) (i) and h-poly-1b-H (run 4 in Table 2) (j), measured in alkaline water ([polymer]/[NaOH] ) 1) at ambient temperature;
[polymer] ) 1.0 mg/mL.
Table 2. CD Spectral Data of Helical Poly-1-H’s Induced by (1R,2S)-2 in DMSO-Water (9:1 or 4:1, v/v) and by 3 in Water and Organic
Solvent-Water (1:1, v/v) and Isolated Polymers and Their Methyl Esters after Removal of Chiral Aminesa
after removal of chiral amines
poly-1-Me
∆ε (λ)e
run
polymer
chiral amine
solvent (v/v)
∆ε (λ, time)b
∆ε (λ)c
memory efficiency (%)d
1
2
3
4
5
6
7
8
9
poly-1-H
poly-1-H
poly-1-H
poly-1-H
(1R,2S)-2 DMSO-water (9/1)
+8.53 (362, 3)
poly-1a-H
+0.04 (360)
-10.3 (357)
+10.6 (357)
-10.6 (357)
-9.65 (356)
+10.4 (357)
+0.05 (360)
-11.9 (357)
-10.8 (358)
0.5
92
88
89
90
89
0.5
94
88
93
93
0.5
92
1.6
88
85
ca. 0
(R)-3
(S)-3
(R)-3
(R)-3
(S)-3
water
-11.1 (360, 46) h-poly-1b-H
+12.0 (360, 55) h-poly-1b-H
-11.9 (360, 15)f h-poly-1b-H
-10.7 (360, 17)f h-poly-1b-H-d4
+11.7 (360, 17)f h-poly-1b-H-d4
+9.48 (362, 33)g poly-1a-H-13C
-12.7 (360, 33) h-poly-1b-H-13C
-12.3 (360, 16)f h-poly-1b-H-13C
water
+11.6 (360)
-11.8 (360)
-9.16 (360)
+9.59 (359)
ca. 0
CH3CN-water (1/1)
CH3CN-water (1/1)
CH3CN-water (1/1)
poly-1-H-d4
poly-1-H-d4
poly-1-H-13C
poly-1-H-13C
poly-1-H-13C
(1R,2S)-2 DMSO-d6-D2O (9/1)
(R)-3
(R)-3
D2O
CH3CN-water (1/1)
CH3CN-water (1/1)
CH3CN-water (1/1)
-12.8 (360)
-8.88 (359)
+8.08 (360)
ca. 0
-11.0 (360)
+0.04 (360)
-12.0 (360)
-14.9 (359)
10 poly-1-H-d4-13C (R)-3
11 poly-1-H-d4-13C (S)-3
-9.97 (360, 17)f h-poly-1b-H-d4-13C -9.28 (357)
+10.4 (360, 17)f h-poly-1b-H-d4-13C +9.63 (357)
12 poly-1-H-15N
13 poly-1-H-15N
14 HMW-poly-1-H (1R,2S)-2 DMSO-water (4/1)
15 HMW-poly-1-H (R)-3
16 HMW-poly-1-H (R)-3
(1R,2S)-2 DMSO-water (9/1)
(R)-3 CH3CN-water (1/1)
+8.79, (362, 3)
-11.8 (359, 17)f h-poly-1b-H-15
+7.21 (362, 5) HMW-poly-1a-H
poly-1a-H-15N
+0.04 (360)
-10.9 (357)
+0.12 (362)
N
CH3CN-water (1/1)
-14.7 (360, 21)f HMW-h-poly-1b-H -12.9 (357)
2-propanol-water (1/1) -17.9 (357, 72)f HMW-h-poly-1b-H -15.2 (360)
a The concentration of polymers is 1.0 mg/mL. b Measured at ambient temperature (20-25 °C) after the samples had been allowed to stand at 50 °C
for days; molar ratio of an amine to monomeric units of poly-1-H’s is 10; ∆ε (M-1 cm-1), λ (nm), and time (days). c Measured in alkaline water
([NaOH]/[polymer] ) 1 and [polymer] ) 1.0 mg/mL); ∆ε (M-1 cm-1) and λ (nm). d Estimated on the basis of the ICD values of poly-1-amine
complexes before isolation. e Measured in CHCl3 ([polymer] ) 1.0 mg/mL); ∆ε (M-1 cm-1) and λ (nm). f Molar ratio of an amine to monomeric units
g
of poly-1-H is 20. After 14 days, the poly-1-H-13C-(1R,2S)-2 complex showed ∆ε362 ) +9.33.
of the poly-1-H’s (Figure 3). The 1H, 13C, and 15N NMR spectra
showed considerable differences in their patterns. In particular,
the changes in the structures of poly-1-H-13C before and after
the helix formation upon complexation with (1R,2S)-2 in
DMSO-d6-D2O (9:1, v/v) (e) and (R)-3 in D2O (g) were most
clearly observed in their 13C NMR spectra (Figure 3B). The
original poly-1-H-13C exhibited rather broad and complicated
signals composed of at least three signals with different
intensities centered at 160.7, 161.9, and 162.8 ppm due to the
main-chain CdN carbon resonances, which can be ascribed to
some stereoirregularity of the main-chain for triad tacticity due
to the syn-anti isomerism and/or conformationally different
segments (s-cis and s-trans) (Chart 2). However, interestingly,
the 13C NMR spectra of the poly-1-H-13C complexed with
(1R,2S)-2 in DMSO-d6-D2O (9:1, v/v) and (R)-3 in D2O after
helicity induction (the molar ellipticity at the first Cotton effect
(∆εfirst) was +9.33 and -12.7, respectively) exhibited sharp
signals centered at 162.8 and 160.9 ppm, respectively, suggest-
ing selective configurational isomerization around the CdN
double bonds (syn-anti isomerization) (Chart 2C) which may
take place during the helix formation.17,19 More interestingly,
these sharp signals remained even after removal of the chiral
amines (f and h), although the former polymer immediately lost
its optical activity, while the latter maintained its optical activity.
1
Similar spectral changes were also observed in the H NMR
spectra (Figure 3A). The peaks of the original poly-1-H (a) in
1
its H NMR spectrum were rather broad and complicated but
became sharper upon complexation with chiral amines in
DMSO-d6-D2O (9:1, v/v) (b) and D2O (c) and subsequent
removal of the amines. In addition, the aromatic proton
resonances of the poly-1a-H significantly shifted downfield (b),
whereas those of the h-poly-1b-H (c) shifted upfield, indicating
intramolecular stacking of the aromatic pendant groups of
h-poly-1b-H in water, which are aligned in a one-handed helical
array, thus showing an ICD.17 The 15N NMR spectra of poly-
1a-Me-15N and h-poly-1b-Me-15N also showed similar changes,
1
but the changes were not as clear as those observed in the H
and 13C NMR spectra.
In addition, we recently found that the “as-prepared” poly-
1-H in the absence of chiral amines also very slowly changed
9
J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009 10723

A R T I C L E S
Hase et al.
Figure 3. (A) 1H NMR spectra of poly-1-H (a), poly-1a-H (run 1 in Table 2) (b), and h-poly-1b-H (run 2 in Table 2) (c) in D2O containing 2 mM NaOD
at 30 °C. (B) 13C NMR spectra of poly-1-H-13C before (d) and after helicity induction with (1R,2S)-2 in DMSO-d6-D2O (9:1, v/v) (run 7 in Table 2) (e)
and (R)-3 in D2O (run 8 in Table 2) (g). The samples had been allowed to stand at 50 °C for 14 (e) and 33 days (g), and their NMR spectra were measured
at 30 °C. 13C NMR spectra after removal of the chiral amines measured in D2O containing 2 mM NaOD at 30 °C are also shown (f and h) (runs 7 and 8
in Table 2). (C) 15N NMR spectra of poly-1-Me-15N (i), poly-1a-Me-15N (run 12 in Table 2) (j), and h-poly-1b-Me-15N (run 13 in Table 2) (k) at 23 °C in
CDCl3.
Chart 2. Possible Triad Structures of Polyisocyanides (A), s-Cis
and s-Trans Conformations (B), and Configurational Syn-Anti
Isomerization of Polyisocyanide Dyad (C)
its structure with time at ambient temperature in alkaline
water.17,19 Figure 4 shows the 13C NMR spectra of the main-
chain CdN region of poly-1-H-13C (a) after it had been allowed
to stand at ambient temperature for 5 months in D2O containing
2 mM NaOD (poly-1′-H-13C (b)); those of poly-1a-H-13C (c)
and h-poly-1b-H-13C (f) are also shown for comparison. Quite
interestingly, the 13C NMR spectrum of poly-1′-H-13C (b) was
almost identical to that of poly-1a-H-13C (c), showing a sharp
signal centered at 162.8 ppm.22 A similar 13C NMR spectral
change took place in h-poly-1b-H-13C (f) after storing the
sample solution at ambient temperature for 2 months (i). On
the other hand, upon heating at 80 °C for 2 h, poly-1a-H-13C
(c) and h-poly-1b-H-13C (f) showed almost the same 13C NMR
Figure 4. 13C NMR spectra of poly-1-H-13C (a), poly-1a-H-13C (run 7 in
Table 2) (c), h-poly-1b-H-13C (run 8 in Table 2) (f), and poly-1′-H-13C (b
and i) obtained from poly-1-H-13C and h-poly-1b-H-13C after storing their
solutions at room temperature for 5 and 2 months, respectively. 13C NMR
spectra (d, e, g, and h) were obtained from poly-1a-H-13C (c) and h-poly-
1b-H-13C (f) after heating the samples at 80 °C for 2 h (d and g,
respectively), followed by storing the samples (d and g) at room temperature
for 1 month (e) and 3 months (h), respectively. The 13C NMR spectrum of
h-poly-1b-H-13Cs(R)-3 complex obtained from poly-1a-H-13C after an-
nealing the sample (c) with (R)-3 ([(R)-3]/[poly-1a-H-13C] ) 10) in D2O
at 50 °C for 1 month is also shown (j). These NMR spectra were measured
in D2O containing 2 mM NaOD at 30 °C.
(22) Similar changes in the 13C NMR spectra of the main-chain CdN region
of poly-1-Me-13C were also observed (see Figure S6, Supporting
Information).
spectra (d and g, respectively) as that of the “as-prepared” poly-
1-H-13C (a),17 and the ICD of h-poly-1b-H-13C completely
9
10724 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009

Helical Structures of Poly(phenyl isocyanide) Derivatives
A R T I C L E S
Figure 5. (A) IR spectra of poly-1-Me (a), poly-1a-Me (run 1 in Table 2) (b), poly-1′-Me (c), h-poly-1b-Me (run 4 in Table 2) (d), h-poly-1b-Me-d4 (run
5 in Table 2) (f), h-poly-1b-Me-13C (run 9 in Table 2) (g), h-poly-1b-Me-d4-13C (run 10 in Table 2) (h), and h-poly-1b-Me-15N (run 13 in Table 2) (i)
measured in CHCl3 (800-1150 and 1330-2000 cm-1) and CDCl3 (1150-1330 cm-1) (a, b, d-i) and in CHCl3 (c) at ambient temperature; [polymer] ) 50
mg/mL. Poly-1′-Me (c) was obtained from poly-1-Me (a) after storing its CHCl3 solution at room temperature for 12 days. The IR spectrum after heating
the CHCl3 solution of h-poly-1b-Me (d) at 80 °C for 15 h is also shown (e). Magnified IR spectra corresponding to the areas indicated by the squares in
panel A are shown in panels B and C.
disappeared. Further storage of the samples (d and g) at room
temperature for 1 month (poly-1′-H-13C (e)) and 3 months (poly-
1′-H-13C (h)) produced significant changes in their 13C NMR
spectra, showing a sharp signal centered at 162.8 ppm. These
results revealed that poly-1′-H and poly-1a-H may have the
thermodynamically most favorable structure, and such a struc-
tural change will slowly take place via configurational isomer-
ization around the CdN double bonds in solution. A similar
change in the structure, accompanied by a preferred-handed
helicity induction, appears to occur slowly under predominantly
thermodynamic control in the “as-prepared” poly-1-H during
the complexation with chiral amines in DMSO,19 although the
optical activity due to the excess one helical sense is tentative
and instantly disappears after removal of the chiral amines
(poly-1a-H),17,20 suggesting that this helicity induction process
in DMSO involves a helical conformational change. Interest-
ingly, the thermodynamically most stable poly-1a-H further
changed its structure into a preferred-handed helical polymer
with a macromolecular helicity memory (h-poly-1b-H) upon
complexation with chiral amines in water (j). As a consequence,
the polyisocyanides with different main-chain structures, poly-
1-H, poly-1a-H, and h-poly-1b-H, can interconvert into each
other as revealed by the NMR studies.23
IR and VCD Studies. Additional evidence of the main-chain
structural changes during the helicity induction in poly-1-H as
well as its one-handed helical structure was obtained from the
IR and VCD spectra. A problem in any approach based on
analyzing vibrational spectra is the difficulty in resolving the
contributions from individual residues. To this end, a series of
isotopically labeled poly-1-H’s with D, 13C, or 15N before and
after the helicity induction with chiral amines in different
solvents and after removal of the chiral amines were prepared
and esterified with CH2N2 (Table 2), and these polymers were
used to assign the peaks in the vibrational spectra.
Figure 5 shows the IR spectra of poly-1-Me, poly-1a-Me,
poly-1′-Me, h-poly-1b-Me, and the isotopically labeled h-poly-
1b-Me-d4, h-poly-1b-Me-13C, h-poly-1b-Me-d4-13C, and h-poly-
1b-Me-15N measured in CHCl3 or CDCl3. Noticeable changes
in the IR spectra after helicity induction and memory were
(23) A similar dynamic helical model has also been proposed by Corne-
lissen, Nolte, and co-workers for helical poly(isocyanopeptide)s derived
from ꢀ-amino acids. See ref 10e.
9
J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009 10725

A R T I C L E S
Hase et al.
the polyisocyanides. The enantiomeric helical h-poly-1b-Me’s
obtained from the corresponding h-poly-1b-H’s induced by (R)-
and (S)-3 showed virtually mirror-image Cotton effects in the
CdN stretching and the main-chain C-C vibration regions in
their VCD spectra (Figure 6).17 In addition, h-poly-1b-Me’s
exhibited an intense split-type Cotton effect in the main-chain
C-C bond vibration region around 940 cm-1, a characteristic
band for the helical polyisocyanides as mentioned above, and
the positive and negative VCD signals go through near the
absorption maximum of the C-C bond vibration. Another clear
VCD was also observed at 852 cm-1, which may provide
additional structural information about the helical structure of
h-poly-1b-Me, but peak assignment was difficult due to the
complexity of its vibrational mode, even though the labeled
polymers were used for their assignments. The isotopically
labeled helical polyisocyanides also showed virtually identical
VCD spectral patterns in the CdN stretching and CsC bond
vibration regions (Figure S8).
Figure 6. VCD (a, b) and IR (c) spectra of h-poly-1b-Me prepared by
(R)-3 (run 4 in Table 2) (a, c) and (S)-3 (run 3 in Table 2) (b), measured
in CHCl3 at ambient temperature ([polymer] ) 120 (a, b) and 50 mg/mL
(c)).
observed in the CdN stretching and the main-chain C-C
vibration regions (a and d in Figure 5B,C). The CdN stretching
(1647 cm-1) and the main-chain helical C-C vibrations (940
cm-1) of h-poly-1b-Me observed in its IR spectrum were
significantly different from those of poly-1-Me, poly-1a-Me,
and poly-1′-Me (a-c in Figure 5B). The former band was
shifted to lower wavenumbers by ca. 30 cm-1 relative to the
original poly-1-Me, poly-1a-Me, and poly-1′-Me, and the latter
band appeared as a new band after the helix induction and
memory. These bands are characteristic for the helical poly-
isocyanide because h-poly-1b-Me showed distinct Cotton effects
in the CdN stretching region as well as the main-chain C-C
bond vibration region in its VCD spectrum (see Figure 6). These
assignments were supported by the facts that the CdN stretching
vibrations were shifted to lower wavenumbers by ca. 30 cm-1
relative to that of h-poly-1b-Me in the IR spectra of h-poly-
1b-Me-13C, h-poly-1b-Me-d4-13C, and h-poly-1b-Me-15N (g-i
in Figure 5B), and that the main-chain C-C vibration at 940
cm-1 observed in the IR spectrum of h-poly-1b-Me was
significantly shifted to lower wavenumbers when the carbon or
nitrogen of the CdN double bonds was labeled with 13C (h-
poly-1b-Me-13C and h-poly-1b-Me-d4-13C) or 15N (h-poly-1b-
Me-15N) (g-i in Figure 5C). In addition, poly-1-Me, poly-1a-
Me, and poly-1′-Me have an additional weak but apparent
shoulder peak around 1026 cm-1. It seems likely that the
shoulder peak almost completely disappeared after the helicity
induction and memory in water (h-poly-1b-Me), accompanied
by appearance of the main-chain C-C bond vibration at 940
cm-1 (a-d in Figure 5C). This shoulder peak was assigned to
The advantages of VCD are not only its sensitivity for chiral
nonchromophoric molecules, but also reliable theoretical cal-
culations of its spectrum using methods available in the Gaussian
program.24 Taking advantage of the available calculation
method, calculations of poly(phenyl isocyanide) (PPI), a model
polymer of the helical h-poly-1b-Me, were then performed to
simulate its VCD spectrum, which was further used to determine
the helix-sense (right- or left-handed helix) of h-poly-1b-Me
by comparison with the experimental VCD spectra (see below).
Persistence Length Measurements. We next estimated the
backbone stiffness (persistence length q) of HMW-poly-1-Me,
HMW-poly-1a-Me, and HMW-h-poly-1b-Me in CHCl3.25 The
q values of these polymers were estimated using SEC equipped
with multiangle laser light scattering (MALS) and refractive
index detectors in series along with the wormlike chain model.26
This model can be described as an analytical function of the
molecular weight (Mw) and the radius of gyration (S) if the q
values and the molar mass per unit contour length (ML), which
eventually leads to the monomer unit height (h), are given. In
this way, the dependences of the Mw on the radii of gyration of
these polymers were explored by using the SEC-MALS system
and CHCl3 as the eluent (Figure 7). The solid curves in the
plots were calculated using the parameters determined from the
(24) Frisch, M. J.; et al. Gaussian 03, Revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
(25) The persistence length (q) is a useful measure to evaluate the stiffness
of rodlike polymers, but only a limited number of q values have been
determined for synthetic helical polymers. See: (a) Sato, T.; Teramoto,
A. AdV. Polym. Sci. 1996, 126, 85–161. (b) Nieh, M.-P.; Goodwin,
A. A.; Stewart, J. R.; Novak, B. M.; Hoagland, D. A. Macromolecules
1998, 31, 3151–3154. (c) Gu, H.; Nakamura, Y.; Sato, T.; Teramoto,
A.; Green, M. M.; Andreola, C. Polymer 1999, 40, 849–856. (d)
Teramoto, A.; Terao, K.; Terao, Y.; Nakamura, N.; Sato, T.; Fujiki,
M. J. Am. Chem. Soc. 2001, 123, 12303–12310. (e) Nomura, R.; Tabei,
J.; Nishiura, S.; Masuda, T. Macromolecules 2003, 36, 561–564. (f)
Ashida, Y.; Sato, T.; Morino, K.; Maeda, K.; Okamoto, Y.; Yashima,
E. Macromolecules 2003, 36, 3345–3350. (g) Okoshi, K.; Sakajiri,
K.; Kumaki, J.; Yashima, E. Macromolecules 2005, 38, 4061–4064.
(h) Okoshi, K.; Sakurai, S.-i.; Ohsawa, S.; Kumaki, J.; Yashima, E.
Angew. Chem., Int. Ed. 2006, 45, 8173–8176. Persistence length values
of certain polyisocyanide derivatives have been measured. See ref 11
and the following: (i) Biegle´, A.; Mathis, A.; Galin, J.-C. Macromol.
Chem. Phys. 2000, 201, 113–125. (j) Samor´ı, P.; Ecker, C.; Go¨ssl, I.;
de Witte, P. A. J.; Cornelissen, J. J. L. M.; Metselaar, G. A.; Otten,
M. B. J.; Rowan, A. E.; Nolte, R. J. M.; Rabe, J. P. Macromolecules
2002, 35, 5290–5294. (k) Okoshi, K.; Nagai, K.; Kajitani, T.; Sakurai,
S.-i.; Yashima, E. Macromolecules 2008, 41, 7752–7754.
a type of C-C bond vibration, since after enrichment with 13
C
(poly-1-Me-13C) the peak disappeared (e in Figure S7, Sup-
porting Information). These assignments were further supported
by the fact that the typical peaks at 1647 and 940 cm-1 for
h-poly-1b-Me disappeared after heating the sample at 80 °C
for 15 h (d and e in Figure 5); the resulting IR spectrum was
almost identical to that of poly-1-Me (a and e in Figure 5). We
noted that poly-1a-Me and poly-1′-Me showed almost identical
IR spectra, as expected from the NMR results. These IR results
revealed a configurational isomerization around the CdN double
bonds that surely takes place during the helicity induction and
memory; that is, an imino configurational mixture of syn and
anti forms of the original poly-1 is transformed into an excess
of a single configuration upon complexation with the chiral
amines.
The VCD spectra of h-poly-1b-Me and the isotopically
labeled polymers (Figures 6 and S8 (Supporting Information))
provided strong evidence of a one-handed helical structure of
(26) (a) Yamakawa, H.; Fujii, M. Macromolecules 1974, 7, 128–135. (b)
Yamakawa, H.; Yoshizaki, T. Macromolecules 1980, 13, 633–643.
9
10726 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009

Helical Structures of Poly(phenyl isocyanide) Derivatives
A R T I C L E S
Figure 8. Polarized optical micrographs of LC phases of nematic HMW-
poly-1-H (run 6 in Table 1) (A) and cholesteric HMW-h-poly-1b-H (run
15 in Table 2) (B) in alkaline water and HMW-h-poly-1b-Me (run 15 in
Table 2) (C) in 1,1,2,2-tetrachloroethane. Concentrations of the polymers
(A-C) were 20, 15, and 20 wt %, respectively.
ester (HMW-poly-1-Me) also showed a nematic LC phase in
CHCl3. Furthermore, the corresponding helical polymers, HMW-
h-poly-1b-H and HMW-h-poly-1b-Me, exhibited cholesteric LC
phases in alkaline water (15 wt %) and 1,1,2,2-tetrachloroethane
(20 wt %), respectively, showing a fingerprint texture (B and
C in Figure 8, respectively). The spacings of the fringes
corresponding to the half-pitch of the cholesteric helical structure
were 6.0 (B) and 1.8 µm (C), respectively. This finding of
nematic and cholesteric LC formations of the polyisocyanides
before and after the helicity induction and memory makes
possible the analysis of their helical structures by XRD, because
we can make shear-oriented films from the LC samples.10h,27
Figure 9B shows the wide-angle X-ray diffraction (WAXD)
pattern with different sensitivity ranges of the uniaxially oriented
HMW-h-poly-1b-Me film prepared from its concentrated cho-
lesteric LC CHCl3 solution. The oriented HMW-h-poly-1b-Me
film showed broad but apparent equatorial and near-meridional
reflections. The five equatorial reflections, 14.19, 10.56, 6.46,
5.13, and 4.03 Å, can be indexed with a simple two-dimensional
tetragonal lattice of a ) 14.37 Å (Table S1, Supporting
Information). Next, we determined the fiber period to be 10.30
Å () c) from the layer lines (1/10.30 Å-1 apart) attributed to
the helical structure, where the weak meridional reflection of
nearly 4.1 Å was excluded from this analysis because the
reflection cannot be assumed for the regular helical structure
of HMW-h-poly-1b-Me, as described below for HMW-poly-
1-Me. The observed density (1.259 g/cm3) indicates that this
unit cell includes 10 monomer units/1 chain along the c-axis,
in agreement with the calculated density (1.258 g/cm3).28
Although we could not observe a meridian reflection on the
10th layer line (1.03 Å) (see Table S1) corresponding to the
unit height (h) of the helical structure, even when the X-ray
measurements were performed using a cylindrical camera with
the samples tilted ca. 48° normal to the beam, the most plausible
helical structure of the HMW-h-poly-1b-Me is proposed to be
a 10-unit/3-turn helix (10/3) (Figure 10A) on the basis of
the strong reflection on the third layer line (3.41 Å) and the
reflection on the fourth layer line (nearly 2.6 Å), which
Figure 7. Double-logarithmic plots of the radius of gyration versus the
molecular weight of HMW-poly-1-Me (a), HMW-poly-1a-Me (run 14 in
Table 2) (b), and HMW-h-poly-1b-Me (run 16 in Table 2) (c) in CHCl3
obtained by SEC-MALS measurements at 25 °C. Solid curves were obtained
on the basis of the wormlike chain theory and fit well with the experimental
data. The evaluated parameters are as follows: HMW-poly-1-Me, q ) 59
nm, ML ) 1618 nm-1, h ) 0.10 nm; HMW-poly-1a-Me, q ) 43 nm, ML
) 1483 nm-1, h ) 0.11 nm; HMW-h-poly-1b-Me, q ) 88 nm, ML ) 1592
nm-1, h ) 0.10 nm.
fits of the unperturbed wormlike chain model over the entire
Mw studied range and are represented by the theoretical values
of 〈S2〉0.5.
The calculated q value of the “as-prepared” HMW-poly-1-
Me is 59 nm, which is slightly smaller than that of a helical
polyisocyanopeptide (76 nm).25j Interestingly, HMW-h-poly-
1b-Me (run 16 in Table 2) with a macromolecular helicity
memory exhibited the significantly higher q value of 88 nm. In
contrast, HMW-poly-1a-Me (run 14 in Table 2), which lost
optically activity after helicity induction, showed a dramatic
decrease in its q value to 43 nm, leading to a rather semirigid
polymer. These changes in the q values clearly demonstrate the
specific configurational isomerization around the CdN double
bonds (syn-anti isomerization) that likely takes place during
the helicity induction process, resulting in the changes of the
main-chain stiffness of HMW-poly-1-Me.
Helical Structures. Although the helical structure of poly-
isocyanides has been postulated to be a 4-unit/1-turn (4/1) helical
conformation on the basis of the X-ray analysis of poly(R-
phenylethyl isocyanide) (PPEI)2b,d,4a due to using unoriented
samples, the exact helical structure was difficult to determine
on the basis of their X-ray diffractions. Nolte and co-workers
successfully synthesized a series of rigid-rod helical polyiso-
cyanopeptides with a controlled helicity, some of which showed
cholesteric LC phases.10a,b,e,g X-ray diffractions of their unori-
ented cast films were then measured, but only an orthorhombic
arrangement of the polymers in the solid could be proposed.10b
Recently, we found that some helical poly(phenyl isocya-
nide)s derived from low-molecular-weight h-poly-1b-H (Mn )
3.3 × 104) after modification of the carboxy pendant groups
with bulky primary and secondary amines or primary amines
with a long ethylene oxide trimer or a pyridylmethyl group
formed lyotropic LC phases in concentrated solutions due to
their main-chain stiffness, thus showing a fingerprint texture.18
In contrast, the original h-poly-1b-H and h-poly-1b-Me did not
show LC phases in concentrated solutions. However, as we
anticipated, HMW-poly-1-H was found to form a nematic LC
phase in concentrated alkaline water solution (15-20 wt %),
thus showing a clear Schlieren texture (Figure 8A). Its methyl
(27) (a) Nagai, K.; Sakajiri, K.; Maeda, K.; Okoshi, K.; Sato, T.; Yashima,
E. Macromolecules 2006, 39, 5371–5380. (b) Sakurai, S.-i.; Okoshi,
K.; Kumaki, J.; Yashima, E. Angew. Chem., Int. Ed. 2006, 45, 1245–
1248. (c) Sakurai, S.-i.; Ohsawa, S.; Nagai, K.; Okoshi, K.; Kumaki,
J.; Yashima, E. Angew. Chem., Int. Ed. 2007, 46, 7605–7608.
(28) The densities of the HMW-poly-1-Me, HMW-poly-1a-Me, and HMW-
h-poly-1b-Me films were measured by the standard flotation method
in a KBr saturated aqueous solutionswater mixture at ambient
temperature (20-25 °C).
9
J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009 10727

A R T I C L E S
Hase et al.
Figure 9. X-ray diffraction patterns of oriented films obtained from optically inactive HMW-poly-1-Me (A), optically active HMW-h-poly-1b-Me with
helicity memory (run 16 in Table 2) (B), optically inactive HMW-poly-1a-Me (run 14 in Table 2) (C), and HMW-poly-1′-Me (D) with different ranges of
sensitivities taken from the edge-view position with a beam parallel to the film surface using a curved imaging plate; the vertical direction nearly corresponds
to the helical axis. HMW-poly-1′-Me was obtained from HMW-poly-1-Me after its CHCl3 solution was stored at room temperature for 3 months. The
representative layer lines and the main reflections are labeled. The reflections around 2.1 Å indicated by asterisks are probably not attributed to the reflections
due to helical structures judging from WAXD measurements of the same polymer films using a flat imaging plate (see Figure S9, Supporting Information).
correspond to the first- or second-order Bessel functions,
respectively (Table S1), consistent with the helical diffraction
theory.29
rather sharp proton and carbon resonances with different
chemical shifts in their NMR spectra (Figures 3 and S6), and
the IR spectrum of poly-1a-Me is significantly different from
that of h-poly-1b-Me, particularly in the CdN stretching and
the main-chain C-C bond vibration regions (Figure 5). As
expected, the thermodynamically most stable HMW-poly-1′-
Me showed an XRD pattern almost identical to that of HMW-
poly-1a-Me. We first determined the fiber period to be 20.70
Å () c) from the layer lines (1/20.70 Å-1 apart) attributed to
the helical structure. The observed density (1.255 g/cm3)
indicates that this unit cell includes 18 monomer units/1 chain
along the c-axis, in agreement with the calculated density (1.266
g/cm3).28 Although the number of meridian reflections is limited,
the structure of poly-1a-Me and poly-1′-Me could be proposed
to be an 18-unit/5-turn helix (18/5) on the basis of the strong
reflections on the fifth layer line (4.14 Å) and the sixth layer
line (3.45 Å) (Table S1). However, we assumed that the structure
of poly-1a-Me and poly-1′-Me may be different from an s-cis-
isotactic structure like h-poly-1b-Me because of the essentially
different IR and NMR spectral patterns from those of h-poly-
1b-Me. Poly-1a-Me and poly-1′-Me most likely take the
thermodynamically most stable structures. We then performed
molecular mechanics (MM) calculations (see the Supporting
Information) in order to obtain the optimized helical structure
Next, we tried to elucidate the structures of HMW-poly-1a-
Me and HMW-poly-1′-Me, which were anticipated to be the
same on the basis of their essentially identical NMR and IR
spectra, by XRD in the same way in order to address the
questions concerning a difference in the helical structures of
the polyisocyanides induced in DMSO and water in the presence
of chiral amines, and why the preferred-handed helical structure
induced in DMSO cannot be memorized after removal of
the chiral amines. The XRD data of the oriented optically
inactive HMW-poly-1a-Me film indicated that the polymer
chains are packed with a smaller tetragonal lattice of a ) 13.56
Å (Figure 9C and Table S1). A clearly strong and sharp
reflection of 4.14 Å, which was also observed in the HMW-h-
poly-1b-Me film as a weak reflection (Figure 9B), suggests that
the HMW-poly-1a-Me may have another regular helical struc-
ture, different from the 10/3 helical structure proposed for
HMW-h-poly-1b-Me. This speculation is supported by the IR
and NMR results; poly-1a-Me and h-poly-1b-Me exhibited
(29) Cochran, W.; Crick, F. H. C.; Vand, V. Acta Crystallogr. 1956, 5,
581–586.
9
10728 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009

Helical Structures of Poly(phenyl isocyanide) Derivatives
A R T I C L E S
form nearly 10/3 helices having an average helical pitch of
nearly 3.5 Å, but it is postulated that other HMW-poly-1-Me
segments likely form another helical structure corresponding
to that of HMW-poly-1a-Me, that is, a nearly 9/5 helical
structure with two monomer units as the repeating unit because
of a diffuse but relatively strong meridional reflection of nearly
4.1 Å, which may correspond to the fifth layer line of the 9/5
helical structure, as observed for the HMW-poly-1a-Me. The
intensity of this reflection is definitely strong for the stereoir-
regular HMW-poly-1-Me but weak for the regular HMW-h-
poly-1b-Me (Figure 9B). These observations are quite consistent
with the IR and NMR results. In addition, there results imply
that the regularity of the helical backbone of HMW-h-poly-1b-
Me is still imperfect and the reflection in the region of nearly
4.1 Å for HMW-h-poly-1b-Me will disappear when the regular-
ity of the helical backbone increases.
Consequently, the h-poly-1b-Me backbone, with a preferred-
handed helical conformation, consists of the regular 10/3 helix
as the major structural element, and the poly-1a-Me and poly-
1′-Me chains have predominantly 9/5 helices with two monomer
units as a repeating unit whose helical structure appears to be
completely different from that of the h-poly-1b-Me with respect
to their main-chain conformations and configurations. The poly-
1-Me chains may be composed mainly of a mixture of a nearly
10/3 helix, a nearly 9/5 helix, and some other disordered
segments.
Figure 10. Top and side views of a 10/3 right-handed helical structure of
s-cis-isotactic h-poly-1b-Me (20-mer) (A) and a 9/5 right-handed helical
structure of s-cisss-trans poly-1a-Me (20-mer) with two monomer units
having alternative s-cis and s-trans conformations as a repeating unit (B).
The main-chain carbon and nitrogen atoms are shown in yellow and blue,
respectively, using the ball-and-stick model for clarity. The phenyl groups
of the side chains are shown in gray using the cylinder model, and hydrogen
atoms and the methyl ester groups are omitted for clarity.
Computational Studies on Helical Structure. To gain further
information regarding the helical structure of h-poly-1b, PPI
oligomers with a forced periodic structure were employed as a
model polymer, and their structures were calculated according
to a procedure developed by Sawabe and co-workers which is
useful for calculating the structure and its related property of a
periodic polymer based on a small sized oligomer (see the
Supporting Information).32 Calculations were performed using
the semiempirical PM3 method33 and the hybrid density
for poly-1a-Me and poly-1′-Me that satisfies the 18/5 helix;
the optimized 18/5 helical structure was found to alternate s-cis
and s-trans (Figure 10B) and thus can be regarded as a 9/5 helix
with two continuous s-cis and s-trans monomer units as a
repeating unit.30,31 A similar helical polyisocyanide structure
was theoretically proposed by Clericuzio and Salvadori.14a
The structural determination for HMW-poly-1-Me was also
attempted using the same strategy as for the HMW-h-poly-1b-
Me and HMW-poly-1a-Me, measuring the WAXD of an
oriented optically inactive HMW-poly-1-Me film prepared from
a concentrated nematic LC CHCl3 solution (Figure 9A). The
XRD data indicate that the HMW-poly-1-Me chains are packed
with a slightly smaller tetragonal lattice of a ) 13.96 Å (Table
S1). However, we could not detect the reflection of nearly 2.6
Å, corresponding to the fourth layer line of the 10/3 helical
structure as observed for HMW-h-poly-1b-Me, which suggests
that there is no correlation with the long-range order as observed
for the HMW-h-poly-1b-Me. The remaining third layer line of
the 10/3 helix indicates that some poly-1-Me segments may
functional theory (DFT) method at the B3LYP level.34-37
A
mixed basis set (MIX), consisting of the 6-31G(d) basis38,39
for nitrogen and carbon atoms of the main-chain and the 6-31G
basis39 for other atoms, was used. All calculations were carried
out using Gaussian 03.24
At the PM3 level, we performed the full optimization for
the s-cis-isotactic PPIs of 4-, 10-, 20-, 30-, and 40-mers. The
increase in the main-chain length led to the convergence of
the dihedral angle (C1-C2-C3-C4) of the central part of the
oligomers along the main-chain to 80°. On the basis of the
converged structure, the repeating structure unit was con-
firmed to be a monomer (>CdNsPh) for the s-cis isotactic
PPI. A similar procedure was performed for the s-trans
isotactic PPI, and its repeating structure unit for the periodic-
ity was found to be two monomers.
(30) The MM calculations of the s-cis-isotactic 18/5 helical structure of
poly-1a-Me and poly-1′-Me with a repeating structure unit (1-mer)
for the periodicity like h-poly-1b-Me revealed that the energy-
minimized conformation had a much higher energy than the 10/3
helical h-poly-1b-Me. We then considered a plausible helical structure
of poly-1a-Me and poly-1′-Me that has a repeating structure unit (2-
mer) for the periodicity (Chart S3, Supporting Information) and found
that the energy-minimized 9/5 helical structure with an alternating s-cis
and s-trans conformation had a much lower energy than h-poly-1b-
Me (see the Supporting Information).
(31) Although the IR and NMR spectral results suggest that the main-
chain configuration of poly-1a-Me may differ from that of h-poly-
1b-Me (Figures 3 and S6), the optimized helical structure of poly-
1a-Me obtained by the empirical MM calculations appears to be
isotactic; a syndiotactic structure may likely be more plausible for
poly-1a-Me to explain the differences in their IR and NMR spectra.
Apparently, more detailed nonempirical calculations are required to
determine the main-chain configuration of poly-1a-Me.
At the B3LYP/MIX level, geometry optimizations for the
n-mer (n ) 1-14) oligomers of the s-cis isotactic PPI were
then carried out under the forced periodic condition (see the
(32) Ochi, N.; Hase, Y.; Maeda, K.; Yashima, E.; Sawabe, K. Manuscript
in preparation.
(33) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209–220.
(34) Becke, A. D. Phys. ReV. A 1988, 38, 3098–3100.
(35) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5642.
(36) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785–789.
(37) Stephens, P. J.; Devlin, J. F.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623–11627.
(38) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56,
2257–2261.
(39) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta. 1973, 28, 213–222.
9
J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009 10729

A R T I C L E S
Hase et al.
Figure 11. Top and side views of the optimized 7/2 right-handed helical
structure of s-cis isotactic PPI (14-mer). The main-chain carbon and nitrogen
atoms are shown in yellow and blue, respectively, using the ball-and-stick
model for clarity. The phenyl groups of the side chains are shown in gray
using the cylinder model, and hydrogen atoms are omitted for clarity.
Figure 12. Comparison between calculated and observed IR and VCD
spectra. (A) VCD spectra of h-poly-1b-Me prepared by (R)-3 (run 4 in
Table 2) (b) and (S)-3 (run 3 in Table 2) (d), measured in CHCl3 at ambient
temperature ([polymer] ) 120 mg/mL), and the calculated left-handed (a)
and right-handed helical PPIs (c). (B) IR spectra of h-poly-1b-Me prepared
by (R)-3 (run 4 in Table 2) (e) measured in CHCl3 at ambient temperature
and the calculated left-handed helical PPI (f). The calculated IR and VCD
spectra were constructed from calculated dipole and rotational strengths
assuming Lorentzian band shape with a half-width at half-maximum of 3
cm-1. The calculated frequencies were scaled by a frequency-independent
factor of 0.9614.44
Supporting Information). As a result, the 14-mer helical PPI
was sufficient to yield the physical and chemical properties of
the true periodic polymer. The optimized s-cis isotactic PPI has
a repeating structure unit (1-mer) for the periodicity and a helical
conformation with the bond length (C1-C2, 1.528 Å), the bond
angle (C1-C2-C3, 114.8°), and the dihedral angle (C1-C2-
C3-C4, 85.6°), giving a 7-unit/2-turn (7/2) helix (see Figure 11
for a 7/2 right-handed helical structure of PPI). For comparison,
the 10/3 right-handed helical structure of h-poly-1b-Me pos-
tulated by the X-ray analysis is shown in Figure 10A; the
pendant methyl ester groups were omitted for clarity. The
calculated helical structure of PPI (7/2 helix) appears to be
analogous to that of a PPI derivative, h-poly-1b-Me (10/3 helix)
experimentally determined by XRD measurements; the differ-
ence in their helical structures (7/2 versus 10/3 helix) may be
attributed to the difference in their pendant structures.40
observations of the self-assembled helical poly(phenyl isocya-
nide) derivatives.10d,h
Mechanism of Helicity Induction and Memory. Combining
the present experimental observations, theoretical calculations,
and persistence length measurements allowed us to propose the
mechanism of helix induction in poly-1-H with chiral amines
in DMSO and water and the subsequent memory of the
macromolecular helicity in water, as well as possible structures
of the relevant polyisocyanides (Figure 13).
The “as-prepared” poly-1-H may be composed of a mixture
of a nearly 10/3 helix and a nearly 9/5 helix, with two monomer
units as a repeating unit as well as disordered segments arising
from the syn and anti states of each monomer unit, as indicated
by the IR and NMR results, lacking a long-range order in its
structure. In the presence of chiral amines in water or aqueous
organic solutions containing more than 50 vol % water, the poly-
1-H folds into a rigid preferred-handed 10/3 helical structure
accompanied by transformation of an imino configurational
mixture of syn- and anti-poly-1-H to one of a single configu-
ration, and this selective configurational isomerization, assisted
by hydrophobic and chiral ionic interactions in water, forces
the helical conformation on the polymer backbone to take an
excess helical sense. Therefore, the induced helix is supposed
to be kinetically controlled and remains after removal of the
chiral amines at ambient temperature, while at high temperature
the helical conformation unravels due to the syn-anti configu-
The IR and VCD spectra for the optimized 7/2 helical PPI
(14-mer) were then calculated at the B3LYP/MIX level. We
noted that the observed and calculated IR and VCD41 spectra
in the main-chain CdN stretching vibration regions were in
excellent agreement with each other (Figure 12).42 The calcu-
lated VCD spectra for the left- and right-handed helical 7/2 PPIs
(a and c in Figure 12, respectively) were then compared with
the experimental spectra of the helical h-poly-1b-Me’s showing
positive and negative Cotton effect signs (b and d in Figure
12A, respectively), revealing that the helical h-poly-1b-Me,
showing a negative Cotton effect in the main-chain CdN
stretching vibration region, has a right-handed helix. This
assignment of the helical sense in the poly(phenyl isocyanide)s
agrees with those determined by the exciton-coupled CD
method43 and by the direct atomic force microscopy (AFM)
(43) (a) Takei, F.; Hayashi, H.; Onitsuka, K.; Kobayashi, N.; Takahashi,
S. Angew. Chem., Int. Ed. 2001, 40, 4092–4094. The helical sense of
a one-handed helical polyguanidine is also assigned on the basis of
its observed and calculated VCD spectra; see ref 3b. For other
examples of helical sense assignments of polyisocyanides, see ref 7
and the following: (b) Cornelissen, J. J. L. M.; Sommerdijk,
N. A. J. M.; Nolte, R. J. M. Macromol. Chem. Phys. 2002, 203, 1625–
1630.
(40) The optimized s-trans isotactic PPI has a 5/2 helix with two monomer
units as a repeating unit for the periodicity (see Figure S2 and the
Supporting Information). The result also proves the validity of the
assigned s-cis isotactic helical structure of h-poly-1b-Me.
(41) Cheeseman, J. R.; Frisch, M. J.; Devlin, F. J.; Stephens, P. J. Chem.
Phys. Lett. 1996, 252, 211–220.
(42) Because PPI and h-poly-1b-Me possess different pendant groups, the
calculated IR and VCD of PPI in other vibration regions were not in
agreement with the observed IR and VCD spectra of h-poly-1b-Me.
(44) Merrick, J. P.; Moran, D; Radom, L. J. Phys. Chem. A 2007, 111,
11683–11700.
9
10730 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009

Helical Structures of Poly(phenyl isocyanide) Derivatives
A R T I C L E S
and after the helicity induction with chiral amines in water and
DMSO and subsequent memory were investigated in detail by
absorption, CD, IR, VCD, and NMR spectroscopies together
with XRD of the oriented films prepared from the nematic and
cholesteric LC poly-1-Me’s, persistence length measurements,
and theoretical calculations. From the spectroscopic measure-
ment results, it has been concluded that the specific configu-
rational isomerization around the CdN double bonds (syn-anti
isomerism) occurs during the helicity induction processes in each
solvent. XRD of the uniaxially oriented films of the correspond-
ing methyl esters prepared from their LC polymer solutions
suggests that the most plausible helical structure of poly-1a-
Me is a 9/5 helix with two monomer units as a repeating unit,
while that of h-poly-1b-Me is a 10/3 helix consisting of one
repeating monomer unit. The DFT calculations of poly(phenyl
isocyanide), a model polymer of h-poly-1b-Me, afford a 7/2
helix as the most possible helical structure, which is in good
agreement with the XRD results. Furthermore, the persistence
length (q) of poly-1-Me significantly increases from 59 to 88
nm once a preferred-handed helicity is induced and memorized
(h-poly-1b-Me), whereas the q value of poly-1a-Me, which lost
its optical activity after helicity induction, dramatically decreases
to 43 nm, indicating a rather semirigid polymer. The variable-
temperature NMR and CD spectra reveal that the polyisocya-
nides with different main-chain structures, poly-1-H, poly-1a-
H, and h-poly-1b-H, can interconvert in solution. The h-poly-
1b-H and poly-1a-H, with and without a helicity memory,
respectively, favorably changed their helical structures to the
original poly-1-H upon heating in solution via configurational
isomerization of the main-chain. In addition, the original poly-
1-H and h-poly-1b-H also changed their structures into a
structure analogous to that of poly-1a-H after standing in
solution for an extremely long time without heat treatment. Upon
complexation with chiral amines in water, a preferred-handed
helix with a macromolecular helicity memory can also be
induced in poly-1a-H in water. On the basis of these results
together with the remarkable changes in the main-chain stiffness
and the difference in the helical structures of poly-1a-Me and
h-poly-1b-Me, the mechanism of helix induction in poly-1-H
with chiral amines in DMSO and water and the subsequent
memory of the macromolecular helicity in water have been
proposed as follows. In the presence of chiral amines in water,
the original achiral poly-1-H folds into a rigid preferred-handed
10/3 helical structure, accompanied by transformation of an
imino configurational mixture of syn- and anti-poly-1-H to one
of a single configuration, and this selective configurational
isomerization, driven by hydrophobic and chiral ionic interac-
tions in water, assists the helical conformation on the polymer
backbone to take an excess helical sense. Therefore, the induced
helix is likely kinetically controlled and is retained after removal
of the chiral amines at ambient temperature, while at high
temperature the helical conformation unfolds due to the syn-anti
configurational isomerization of the CdN double bonds, going
back to the analogous structure of poly-1-H.
Figure 13. Possible structures of poly-1-H, poly-1a-H, poly-1′-H, and
h-poly-1b-H and their mutual interconversions.
rational isomerization of the CdN double bonds, going back
to an analogous structure of poly-1-H.
The poly-1-H also folds into a preferred-handed rather
semirigid 9/5 helical conformation composed of two monomer
units as a repeating unit with an excess helical sense upon
complexation with chiral amines in organic solutions and
aqueous organic solutions containing less than 30 vol % water,
as evidenced by the appearance of an ICD similar to that of
h-poly-1b-H,19 arising from the similar syn-anti configurational
isomerization. In sharp contrast to h-poly-1b-H, however, the
resulting helical conformation is dynamic in nature, different
from that of h-poly-1b-H, and may spontaneously racemize after
removal of the chiral amines, as observed for the dynamically
racemic polyacetylenes.16a-i Therefore, the induced helix in
poly-1a-H is predominantly controlled by thermodynamics.19
At high temperature, the poly-1a-H favorably changes to a
structure similar to the original poly-1-H via configurational
isomerization of the main-chain. The poly-1a-H structure may
be among the thermodynamically most stable structures, since
the “as-prepared” poly-1-H and kinetically produced h-poly-
1b-H also change their structures into a similar structure after
being stored in solution for an extremely long time, and this
was confirmed by following the changes in its 13C NMR
spectrum with time without heat treatment. In addition, upon
complexation with chiral amines in water, a preferred-handed
helix with a macromolecular helicity memory can also be
induced in poly-1a-H.
Conclusions
The poly-1-H also folds into a preferred-handed, semirigid
9/5 helical conformation with an excess helical sense upon
complexation with chiral amines in DMSO, arising from the
similar syn-anti configurational isomerization. In sharp contrast
to h-poly-1b-H, however, the resulting 9/5 helical conformation
appears to be dynamic and spontaneously racemizes after
removal of the chiral amines, as observed for dynamically
racemic polyacetylenes. Therefore, the induced helix in poly-
1-H is predominantly controlled by thermodynamics in DMSO.
In order to unravel the helicity induction mechanisms of the
optically inactive poly(4-carboxyphenyl isocyanide) (poly-1-
H) in DMSO and water with chiral amines and to address the
question of why the helical structure induced in DMSO cannot
be memorized after removal of the chiral amines (poly-1a-H),
while that induced in water can be retained (h-poly-1b-H), a
series of isotopically labeled poly-1-H’s and their methyl esters
(poly-1-Me’s) with labeled with deuterium, 13C, and 15N and
nonlabeled polymers were prepared, and their structures before
9
J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009 10731

A R T I C L E S
Hase et al.
Agency (JST), and the Global COE Program “Elucidation and
Design of Materials and Molecular Functions” of the Ministry of
Education, Culture, Sports, Science, and Technology, Japan. We
thank Professors T. Yamane, A. Suzuki, and T. Hikage (Nagoya
University) for their help in the X-ray diffraction measurements.
We are deeply grateful to Dr. H. Kusanagi (ERATO, JST) for
valuable discussions. We thank Mr. T. Miyabe for his assistance
in the NMR measurements. K.N. expresses his thanks for a JSPS
Research Fellowship for Young Scientists (No. 6683).
The fact that the original poly-1-H and h-poly-1b-H transform
into the poly-1a-H at the end after standing in solution for an
extremely long time supports the fact that the poly-1a-H
structure may be the thermodynamically most stable structure.
As a consequence, optically inactive poly(4-carboxyphenyl
isocyanide) can adopt either a static or a dynamic helical
conformation upon complexation with chiral amines in water
or organic solvents, respectively, accompanied by selective
configurational isomerization around the CdN double bonds,
and these different helical conformations can interconvert. These
findings combined with the previous discovery that the static
helical polyisocyanides, once modified with achiral primary
amines,18 showed a significant improvement in thermal stability
in water and organic solvents may be useful to design and
construct novel functional chiral materials for chiral recognition
and enantioselective catalysis18b due to the supramolecular
helical array of the pendant functional groups.
Supporting Information Available: Full experimental details;
XRD data of HMW-poly-1-Me, HMW-h-poly-1b-Me, HMW-
poly-1a-Me, and HMW-poly-1′-Me; 13C NMR spectra of poly-
1-Me-13C, poly-1′-Me-13C, h-poly-1b-Me-13C, and poly-1a-Me-
13C; IR spectra of poly-1-Me-13C and poly-1-Me-15N; VCD and
IR spectra of h-poly-1b-Me-d4, h-poly-1b-Me-13C, h-poly-1b-
Me-d4-13C, and h-poly-1b-Me-15N and complete ref.24 This
material is available free of charge via Internet at http://
pubs.acs.org.
Acknowledgment. This work was supported in part by a Grant-
in-Aid for Scientific Research from the Japan Society for the
Promotion of Science (JSPS), Japan Science and Technology
JA904128D
9
10732 J. AM. CHEM. SOC. VOL. 131, NO. 30, 2009