Multiple states of dimeric aggregates formed by (amido-ethynyl)helicene bidomain compound and (amido-ethynyl-amido)helicene tridomain compound

Article abstract of DOI:10.1002/chem.201201929

An (amido-ethynyl)helicene bidomain compound and an (amido-ethynyl-amido) helicene tridomain compound were synthesized. The multidomain compounds were designed on the basis of previous findings that amido and ethynyl oligomers form dimeric aggregates with properties orthogonal to each other. Four aggregate states of multidomain compounds, namely, all-dimer, amido-dimer, ethynyl-dimer, and random-coil states, were obtained in different solvents, which were analyzed by circular dichroism (CD), UV/Vis, ¹H NMR, and IR spectroscopy; vapor pressure osmometry (VPO); dynamic light scattering (DLS); and atomic force microscopy (AFM). The amido and ethynyl domains independently aggregated and disaggregated in a two-state manner. Reversible structural changes occurred for a tridomain compound between the ethynyl-dimer/random-coil state and the all-dimer/amido-dimer state with heating and cooling. Two structural change processes with different properties were obtained using a single compound. Domain switching! An (amido-ethynyl)helicene bidomain compound and an (amido-ethynyl-amido)helicene tridomain compound were synthesized. Four aggregate states of multidomain compounds, namely, all-dimer, amido-dimer, ethynyl-dimer, and random-coil states, were obtained in different solvents. Each domain independently aggregated and disaggregated in a two-state manner. Two different reversible structural changes occurred between the ethynyl-dimer/random-coil and the all-dimer/amido-dimer with heating and cooling. Copyright

Full text of DOI:10.1002/chem.201201929

FULL PAPER
DOI: 10.1002/chem.201201929
Multiple States of Dimeric Aggregates Formed by (Amido–ethynyl)helicene
Bidomain Compound and (Amido–ethynyl–amido)helicene Tridomain
Compound
Wataru Ichinose, Masanori Shigeno, and Masahiko Yamaguchi*[a]
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Abstract: An (amido–ethynyl)helicene
bidomain compound and an (amido–
obtained in different solvents, which
were analyzed by circular dichroism
(CD), UV/Vis, H NMR, and IR spec-
troscopy; vapor pressure osmometry
(VPO); dynamic light scattering
(DLS); and atomic force microscopy
(AFM). The amido and ethynyl do-
mains independently aggregated and
disaggregated in a two-state manner.
Reversible structural changes occurred
for a tridomain compound between the
ethynyl-dimer/random-coil state and
the all-dimer/amido-dimer state with
heating and cooling. Two structural
change processes with different proper-
ties were obtained using a single com-
pound.
1
ethynyl–amido)helicene
tridomain
compound were synthesized. The mul-
tidomain compounds were designed on
the basis of previous findings that
amido and ethynyl oligomers form di-
meric aggregates with properties or-
thogonal to each other. Four aggregate
states of multidomain compounds,
namely, all-dimer, amido-dimer, ethyn-
yl-dimer, and random-coil states, were
Keywords: aggregation
·
helical
structures · independent structure
change · molecular switching · mul-
tidomain structures
Introduction
stimuli, can be obtained. Such synthetic compounds, howev-
er, are not easily obtained, because the structure at each
domain must be well-defined and not perturbed by other
domains. Partially cooperatively, various degrees of aggrega-
tion, intermediate formation in structural changes, and other
undetermined factors complicate structural analysis.
Proteins are organic macromolecules constructed by a linear
and sequential connection of amino acids; they fold into
programmed three-dimensional structures.[1] Folded proteins
contain domains with different structures and properties
such as a-helices and b-sheets. Each domain plays important
roles in maintaining the three-dimensional structures of pro-
teins and in exhibiting functions such as protein–protein in-
teractions or interactions with small molecules. It is there-
fore critical to understand the roles of all domains and their
combinations in biological functions. Protein folding, howev-
er, is a complex process, and its analysis is not straightfor-
ward. It is therefore interesting to examine synthetic multi-
domain compounds with well-defined structures and proper-
ties to understand how the combinations of domains can
affect the bulk properties of macromolecules. In addition,
such multidomain compounds can exhibit functions not ob-
served in single-domain compounds.
Synthetic bidomain compounds with oligomer structures
form layers,[2] particles,[3] fibers,[4] and gels,[5] which employ
conjugates of peptide/peptide, alkylcarboxylate/carbohy-
drate, carbohydrate/polymer, alkylcarboxyamide/peptide,
and peptide/polymer. In such cases one of the domains has
been used as a platform, and the other domain has shown
different aggregate properties from the corresponding
single-domain compounds. We noted that bidomain com-
pounds, in principle, have four structure states, as each
domain can contribute two states. When each domain aggre-
gates and disaggregates independently, multiple molecular-
switch systems, such as proteins that respond to external
Dimeric aggregate formation, particularly double-helix
formation is considered attractive for the development of
such well-defined multidomain compounds. This is because
double helix formation occurs without forming higher aggre-
gates (selective dimerization). Note that the aggregation and
disaggregation of a double helix can occur in a two-state
manner, in which no partly aggregated molecular intermedi-
ates are formed (two-state structural change). These proper-
ties make the design and analysis of double-helix-forming
multidomain compounds straightforward. In addition, the
structural changes of dimeric aggregates have the advantag-
es of being 1) large, reaching several nanometers; 2) sensi-
tive to the environment; 3) detectable by various methods;
and 4) reversible. No such synthetic compounds with a mon-
odispersed multidomain, capable of forming dimeric aggre-
gates at each domain, were examined before.
We previously synthesized oligomers containing optically
active helicene and m-phenylene that form a double helix in
organic solvents. They disaggregate to a random coil with
heating,[6] and their double-helix stability is affected by the
hardness and softness of aromatic solvents. Amidohelicene
oligomers, in which the helicene and m-phenylene are con-
nected by an amide linkage, also form dimeric aggregates.
They disaggregate in hydrogen-bond-breaking solvents such
as 2,2,2-trifluoroethanol or dimethylsulfoxide (DMSO).[7]
The stability of helix dimers, however, is not affected by
temperature changes. Two helix-dimer-forming compounds
that show properties orthogonal to each other were ob-
tained by changing the two-atom linkage from amido to
ethynyl groups. The differences were attributed to the differ-
ent driving forces for the aggregation, p–p interactions in
ethynylhelicene oligomers, and hydrogen bonding in amido-
helicene oligomers. Also note that these compounds do not
form higher aggregates, and aggregate/disaggregate in a
two-state manner.
[a] Dr. W. Ichinose, Dr. M. Shigeno, Prof. Dr. M. Yamaguchi
Department of Organic Chemistry
Graduate School of Pharmaceutical Sciences
Tohoku University, 6-3 Aoba, Sendai 980-8578 (Japan)
Fax : (+81)22-795-6811
E-mail: yama@m.tohoku.ac.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201929.
&
2
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Multidomain Structures
FULL PAPER
dimers with orthogonal properties,[6,7] the bidomain com-
pound 1 containing the ethynyl heptamer and amido tetram-
er domains was designed. The tridomain compound 2 with
an ethynyl pentamer domain at the central part and an
amido tetramer domain at both terminals was also designed.
The ethynyl pentamer domain in 2 should aggregate more
weakly than the ethynyl heptamer domain in 1. The design,
however, had the advantage of the amido tetramer and
ethynyl pentamer domains exhibit comparable Cotton ef-
fects with regard intensity, which facilitated the CD analysis
of different aggregate states of 2. It was also considered in-
teresting to induce thermal structural changes of 2 at the
central ethynyl pentamer domain without inducing a change
at the terminal amido domain.
Compounds 1 and 2 were synthesized by the Sonogashira
coupling of an amidohelicene oligomer and ethynylhelicene
oligomers. The synthesis of the iodo-terminated amidoheli-
cene tetramer 17 was conducted by
a one-directional
method starting from iodobenzene 9,[8] which in turn was
synthesized from the amine 7.[7] Compound 9 was subjected
to a sequence of couplings with the acid chloride 10 fol-
lowed by deprotection, giving 17 (Scheme 1).
Figure 1. Molecular structure of bidomain 1, tridomain 2, amidohelicene
oligomer 3, and ethynylhelicene oligomers 4–6.
In this study, we examined multidomain compounds con-
taining an amido and an ethynyl domain (Figure 1). The bi-
domain compound 1 was synthesized by the coupling of an
amidohelicene tetramer and an ethynylhelicene heptamer,
and the tridomain compound 2 was obtained by the coupling
of two amidohelicene tetramers and an ethynylhelicene pen-
tamer. Both 1 and 2 formed four well-defined aggregate
states, namely, all-dimer, amido-dimer, ethynyl-dimer, and
random-coil states. In the all-dimer state, both the amido
and ethynyl domains formed dimeric aggregates. In the
amido-dimer state, only the amido domain formed dimeric
aggregate. In the ethynyl-dimer state, only the ethynyl
domain formed dimeric aggregate. The random-coil state
was monomeric. Four aggregate states were obtained in dif-
ferent solvents, which were analyzed by circular dichroism
1
(CD), UV/Vis, H NMR, and IR spectroscopy; vapor pres-
sure osmometry (VPO); dynamic light scattering (DLS);
and atomic force microscopy (AFM). In these multidomain
compounds, the amido and ethynyl domains independently
aggregated and disaggregated. On the basis of the results of
these studies, reversible structure switching between the
ethynyl-dimer/random-coil state and the all-dimer/amido-
dimer state were conducted for 2. Two different thermal re-
sponses appeared in nonpolar and polar solvents using the
single compound 2, which originated from the well-defined
aggregation and disaggregation properties of 2 at the amido
domain.
Scheme 1. Synthesis of iodo-terminated amidohelicene oligomers 11–17.
Then, 1 was obtained by the coupling of an equimolar
amount of 17 and the ethynylhelicene heptamer 6[6] in 53%
yield. Polymer 2 was synthesized by the coupling of the dia-
lkyne 18[6] and two equivalents of 17 in 71% yield
(Scheme 2).
The four aggregate states of bidomain compound 1: The bi-
domain compound 1 formed four aggregate states, that is,
all-dimer, amido-dimer, ethynyl-dimer, and random-coil
states, in different solvents (Figure 2). In this study, condi-
tions in which the equilibrium was shifted for each structure
were explored. For example, the all-dimer state containing
Results and Discussion
Design and synthesis of multidomain compounds: On the
basis of previous observations that the ethynylhelicene hep-
tamer 5 and the amidohelicene tetramer 3 formed helix
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
Figure 3. CD spectra of 1 at different a) temperatures (2.5ꢁ10À6 m, tri-
fluoromethylbenzene) and b) concentrations (258C, trifluoromethylben-
zene and fluorobenzene). The calculated spectrum (Figure S4 in the Sup-
porting Information) is also shown.
Scheme 2. Synthesis of bidomain compound 1 and tridomain compound
2.
ined by CD spectroscopy in these solvents. In trifluorome-
thylbenzene, the CD spectra (258C, 2.5ꢁ10À6 m) of 1 (Fig-
ure 3a) are in good agreement with the calculated spectra
(Figure S4 in the Supporting Information) obtained by
adding the spectra of 3 (chloroform, 258C, 5ꢁ10À6 m) and
the ethynyl heptamer 5 (trifluoromethylbenzene, 258C, 5ꢁ
10À6 m), both of which formed helix dimers under the above-
mentioned conditions.[6,7] The spectrum did not change by
heating at 1008C or cooling at À58C. Since 1 was not solu-
ble in trifluoromethylbenzene concentrations higher than
2.5ꢁ10À5 m, its CD spectrum at this concentration was ob-
tained in fluorobenzene, which coincided with that in tri-
fluoromethylbenzene at 2.5ꢁ10À6 m (Figure 3b). These re-
sults indicate that 1 is in the S-all-dimer state in trifluorome-
thylbenzene and fluorobenzene at 258C between 2.5ꢁ10À5
and 2.5ꢁ10À6 m.
The dimeric aggregation of 1 in the all-dimer state was ex-
amined by VPO in chloroform (1ꢁ10À3 m, 358C, benzyl
standard). It was confirmed that the same structure was
formed in chloroform (1ꢁ10À4 m, 258C) and trifluoromethyl-
benzene (2.5ꢁ10À6 m, 258C), as indicated in the CD spec-
trum (Figure S5 in the Supporting Information). The ob-
tained apparent molecular weight of 1.29ꢁ104 was in ac-
cordance with the calculated molecular weight of the dimer-
ic aggregate, that is, 1.319ꢁ104. The experiment also con-
firmed the lack of higher-aggregate formation.
It was previously observed that the addition of DMSO
promoted the unfolding of the helix dimer of 3.[7] Therefore,
a CD analysis of 1 in the ethynyl-dimer state was conducted
in 5% DMSO/trifluoromethylbenzene. The CD spectrum
(258C, 2.5ꢁ10À6 m) of 1 in the mixed solvent (Figure 4) was
in good agreement with the calculated spectra (Figure S6 in
the Supporting Information) obtained by adding the spec-
trum of 3 in the random-coil state[7] (THF, 258C, 5ꢁ10À6 m)
and of 5 in the helix dimer state[6] (trifluoromethylbenzene,
258C, 5ꢁ 10À6 m). The CD spectra also did not change be-
tween 5 and 608C, and the same spectrum was obtained at a
higher concentration of 2.5ꢁ10À5 m (Figure 4b). The spectra
did not change by adding DMSO between concentrations of
3 and 10% (Figure S7 in the Supporting Information). The
results indicate that 1 is in the S-ethynyl-dimer state in 5%
Figure 2. Four aggregate states of 1.
no amido-dimer, ethynyl-dimer, or random-coil state was ex-
amined, and such a state was denoted as the S-all-dimer
state (S=shifted).
Since both 3 and 5 were in the random-coil state in
THF,[6,7] the random-coil state of 1 was examined in this sol-
vent. The CD spectrum (THF, 258C, 2.5ꢁ10À6 m) of 1 (Fig-
ure S1a in the Supporting Information) coincided with the
calculated spectrum (Figure S2 in the Supporting Informa-
tion) obtained by adding the spectra of 3 (THF, 5ꢁ10À6 m,
258C) and 5 (THF, 5ꢁ10À6 m, 258C).[6,7] The spectrum
showed no change with heating at 608C or cooling at À58C
(Figure S1a in the Supporting Information), and the same
spectrum was obtained at a higher concentration of 2.5ꢁ
10À5 m (Figure S1b in the Supporting Information). The in-
sensitivity of the CD spectra to temperature and concentra-
tion indicated that 1 was in the S-random-coil state in THF
at 258C between 2.5ꢁ10À5 and 2.5ꢁ10À6 m. Then, the CD
spectrum obtained here was that of the S-random-coil state
of 1, not containing the all-dimer, amido-dimer, or ethynyl-
dimer state.
Trifluoromethylbenzene and fluorobenzene are solvents
that promote double-helix formation of the ethynyl heptam-
er 5.[6] Moreover, the amidohelicene tetramer 3 exists in the
helix dimer state in trifluoromethylbenzene, fluorobenzene,
and chloroform (Figure S3 in the Supporting Information).[7]
The formation of the all-dimer state of 1, in which both
amido and ethynyl domains formed helix dimers, was exam-
&
4
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Multidomain Structures
FULL PAPER
Figure 6. The syn-configuration of 1 in the all-dimer and amido-dimer
formation.
showed reversible change between all-dimer and amido-
dimer on heating and cooling (Figure S9 in the Supporting
Information). The all-dimer state should possess the syn
configuration. Therefore, the amido-dimer state, which inter-
converted with the all-dimer state, should be the syn config-
uration.
Figure 4. CD spectra of 1 in 5% DMSO/trifluoromethylbenzene at differ-
ent a) temperatures (2.5ꢁ10À6 m) and b) concentrations (258C). The cal-
culated spectrum (Figure S6 in the Supporting Information) is also
shown.
DMSO/trifluoromethylbenzene at 258C at concentrations
between 2.5ꢁ10À6 and 2.5ꢁ10À5 m.
Four aggregate S-states of 1 were obtained in different
solvents exhibiting different CD spectra (Figure 7a). How-
ever, the spectrum of the S-ethynyl-dimer state in 5%
The amido-dimer state of 1 was examined in bromoben-
zene on the basis of our observations that the ethynylheli-
cene heptamer 5 was in the random-coil state[6] and the ami-
dohelicene tetramer 3 was in the helix dimer[7] in this sol-
vent (Figure S3 in the Supporting Information). The CD
spectrum of 1 (2.5ꢁ10À6 m, 258C) (Figure 5) was similar to
Figure 5. CD spectra of 1 in bromobenzene at different a) temperatures
(2.5ꢁ10À6 m) and b) concentrations (258C). The calculated spectrum (Fig-
ure S8 in the Supporting Information) is also shown.
the calculated spectrum (Figure S8 in the Supporting Infor-
mation) obtained from the helix dimer of 3 (chloroform,
258C, 5ꢁ10À6 m) and the random coil of 5 (THF, 258C, 5ꢁ
10À6 m), with slight deviations between 320 and 390 nm. Sim-
ilar deviations in the experimental and calculated spectra
for the amido-dimer state were observed in the tridomain
compound 2 (vide infra), which could be due to some struc-
tural perturbation at the ethynyl domain caused by the
dimer formation at the amido domain. The CD spectra did
not change between 60 and À58C, and the same spectra
were obtained at a higher concentration of 2.5ꢁ10À5 m (Fig-
ure 5b). Thus, it is considered that 1 in bromobenzene is in
the S-amido-dimer state under these conditions.
Figure 7. Spectra of four aggregate S-states of 1 (2.5ꢁ10À6 m, 258C): a)
CD and b) UV/Vis spectra.
DMSO/trifluoromethylbenzene was similar to that of the
S-all-dimer state in trifluoromethylbenzene, and small differ-
ences were observed only at the 290–320 nm region. Amido-
helicene oligomers showed changes in this region by aggre-
gation and disaggregation,[7] and the small differences be-
tween the S-ethynyl-dimer and the S-all-dimer states were
due to the extremely strong Cotton effect of the ethynyl
heptamer domain on double-helix formation compared with
the Cotton effect of the amido tetramer domain. As will be
discussed later, apparently different CD spectra of four ag-
gregate S-states were obtained from the tridomain com-
pound 2, in which the ethynyl pentamer domain instead of
the ethynyl heptamer domain was employed to adjust CD
Two isomeric structures are conceivable in the amido-
dimer state: syn, in which the amido- and ethynyl domain
were arranged in the same direction and anti, in which the
two domains were arranged in the opposite direction
(Figure 6). The CD spectra of 1 in toluene (2.5ꢁ10À6 m)
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
intensity. Four aggregate S-states of 1 were also observed
from the UV/Vis spectra (Figure 7b). The absorptions at
about 340 nm for the S-amido-dimer (bromobenzene) and
S-random-coil (THF) states were enhanced compared with
those for the S-all-dimer (trifluoromethylbenzene) and
IR analysis of 1 was conducted in solution (3ꢁ10À3 m) at
room temperature. The all-dimer in fluorobenzene and
amido-dimer in bromobenzene showed amide carbonyl ab-
sorptions at 1653 and 1648 cmÀ1, respectively (Figure S12a
and b in the Supporting Information). In THF, 1 was in
random-coil state, and provided a carbonyl absorption at
1682 cmÀ1 (Figure S12c in the Supporting Information). The
presence of such a hydrogen-bonded amide carbonyl in the
all-dimer and amido-dimer state was previously observed in
the helix-dimer formation of amidohelicene tetramer 3.[7]
The aggregation of 1 at the amide domain was confirmed in
the all-dimer and amido-dimer states.
AFM analysis was conducted on the aggregates of 1. The
AFM images of the samples prepared from fluorobenzene
solution showed particles about 30 nm diameter for all-
dimer state. The samples prepared from bromobenzene sol-
ution provided 20–30 nm diameter for amido-dimer state
(Figure S13 in the Supporting Information). Particles of uni-
form sizes were formed in wide surfaces for both experi-
ments.
S-ethynyl-dimer
(5%
DMSO/trifluoromethylbenzene)
states. The results were ascribed to disaggregation at the
ethynyl domain, in accordance with our previous observa-
tions.[6]
1
Quite different H NMR spectra were obtained for 1 in
CDCl3, [D5]bromobenzene, 20% [D6]DMSO/CDCl3, and
[D8]THF, which reflected different aggregate states
(Figure 8). In [D8]THF (2ꢁ10À4 m, 258C), 1 was in the
It is shown that four aggregate states, namely, S-all-dimer,
S-amido-dimer, S-ethynyl-dimer, and S-random-coil states,
were obtained for 1 by the judicious choice of solvents,
which indicated that the amido and ethynyl domains of 1 in-
dependently aggregated and disaggregated.
The four aggregate states of tridomain compound 2: Four
aggregate states of tridomain compound 2, namely, S-all-
dimer, S-amido-dimer, S-ethynyl-dimer, and S-random-coil
states, were examined by the same methods. (Figure 9)
Figure 8. 1H NMR spectra of 1 in different solvents (2ꢁ10À4 m, 258C); a)
[D8]THF, b) CDCl3, c) [D5]PhBr, and d) 20% [D6]DMSO/CDCl3.
S-random-coil state, and, accordingly, two sharp amide pro-
tons at d=10.0 and 10.2 ppm as well as well-resolved aro-
matic protons were observed owing to rapid molecular mo-
tions. In CDCl3 (2ꢁ10À4 m, 258C), 1 was in the S-all-dimer
1
state, as indicated by CD and VPO, and its H NMR spectra
showed a serious broadening of amide and aromatic protons,
which was attributed to restricted molecular motions.
Broadening was previously observed for the amido tetramer
3 (Figure S10 in the Supporting Information) and the ethyn-
yl heptamer 5[6] in CDCl3 (1ꢁ10À3 m, 258C). In
[D5]bromobenzene (2ꢁ10À4 m, 258C), an S-amido-dimer was
formed, and the amide protons were again broadened. In
20% [D6]DMSO/CDCl3 (2ꢁ10À4 m, 258C), two relatively
sharp amide peaks were observed at d=10.3 and 10.5 ppm,
which were in accordance with the S-ethynyl-dimer forma-
tion observed by CD spectroscopy (Figure S11 in the Sup-
porting Information). The broadening of amide protons in
CDCl3 and [D5]bromobenzene was ascribed to hydrogen
bonding in helix dimer formation, which should be different
from hydrogen bonding in solvents such as in [D8]THF and
[D6]DMSO/CDCl3.
Figure 9. Four aggregate states of 2.
The CD spectrum of 2 (258C, 2.5ꢁ10À6 m) in THF (Fig-
ure S14a in the Supporting Information) coincided with the
calculated spectrum (Figure S15 in the Supporting Informa-
tion) obtained by adding the spectra of the random-coil
amido tetramer 3[7] (THF, 5ꢁ10À6 m, 258C) and the random-
coil ethynyl pentamer 4[6] (chloroform, 5ꢁ10À6 m, 258C) in
2:1 ratio. The spectra did not change by heating at 608C or
cooling at 58C at a higher concentration of 2.5ꢁ10À5 m (Fig-
ure S14b). Hence, compound 2 is in the S-random-coil state
in THF under the above-mentioned conditions.
&
6
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Multidomain Structures
FULL PAPER
ty between the experimental and calculated spectra may be
partly due to the relatively unstable nature of the helix
dimer of 4, which contains only five helicenes. The CD spec-
tra did not change between 25 and 58C. The intensity, how-
ever, considerably decreased with heating at 808C, which
then recovered with cooling to 58C. The similar spectra
were obtained at concentrations up to 1ꢁ10À4 m (258C) (Fig-
ure 11b). The DMSO content was changed from 6 to 16%,
and essentially the same CD spectrum was obtained be-
tween 10 and 16% (Figure S19 in the Supporting Informa-
tion). Since the amount of hydrogen-bonding-breaking sol-
vent had no effect on the CD spectra above 10%, the
amido domain should be fully disaggregated in these states.
The results indicate the S-ethynyl-dimer state of 2 under the
above-mentioned conditions.
Figure 10. CD spectra of 2 in fluorobenzene at different a) temperatures
(2.5ꢁ10À6 m) and b) concentrations (258C). The calculated spectrum (Fig-
ure S16 in the Supporting Information) is also shown.
Since 2 was not soluble enough in trifluoromethylbenzene
for CD analysis, the experiments on the all-dimer state were
conducted in fluorobenzene (258C, 2.5ꢁ10À6 m). The CD
spectrum of 2 (Figure 10a) coincided with the calculated
spectrum (Figure S16 in the Supporting Information) ob-
tained by adding the spectra of 3 in the helix dimer state[7]
(chloroform, 258C, 5ꢁ10À6 m) and of 4 in the double helix
state[6] (trifluoromethylbenzene, 258C, 1ꢁ10À3 m) in a ratio
of 2:1. The spectra did not change between À5 and 258C.
However, when the mixture was heated at 808C, the Cotton
effect was minimized, which is due to partial disaggregation
at the central ethynyl pentamer domain (Figure S17 in the
Supporting Information). The difference in thermal stability
between 1 (Figure 3) and 2 in ethynyl domain aggregation is
due to the weaker aggregate formation of the ethynyl pen-
tamer domain of 2 than of the heptamer domain of 1. The
same spectrum of 2 was obtained at concentrations of 2.5ꢁ
10À6 and 2.5ꢁ10À5 m (Figure 10b). The results indicated that
2 is in the S-all-dimer state in fluorobenzene at 258C.
In bromobenzene, the CD spectrum (2.5ꢁ10À6 m, 258C) of
2 (Figure 12a) was similar to the calculated spectrum (Fig-
ure S20) obtained by adding the spectra of the helix dimer
Figure 12. CD spectra of 2 in bromobenzene at different a) temperatures
(2.5ꢁ10À6 m) and b) concentrations (258C). The calculated spectrum (Fig-
ure S20 in the Supporting Information) is also shown.
In DMSO/trifluoromethylbenzene, 2 is considered to be
in the ethynyl-dimer state. The CD spectrum of 2 (5ꢁ10À5 m,
258C) in 12% DMSO/trifluoromethylbenzene (Figure 11a)
coincided in shape with the calculated spectrum (Figure S18
in the Supporting Information), which was obtained by
adding the spectra of the random coil of 3[7] (THF, 5ꢁ
10À6 m, 258C) and the helix dimer of 4[6] (trifluoromethylben-
zene, 1ꢁ10À3 m, 258C) in 2:1 ratio. The difference in intensi-
of 3[7] (chloroform, 258C, 5ꢁ10À6 m) and the random coil of
4[6] (chloroform, 258C, 5ꢁ10À6 m) in 2:1 ratio. Some devia-
tions at 320 and 360 nm, however, were observed as for 1 in
bromobenzene (Figure 5a). The CD spectra did not change
between 5 and 608C, and at a higher concentration at 2.5ꢁ
10À5 m (Figure 12b). Compound 2 is in the S-amido-dimer
state in bromobenzene under these conditions.
The amido and ethynyl domains of 2 aggregated and dis-
aggregated in different solvents, and four aggregate S-states
were obtained (Figure 13). Compared with the case of 1,
quite different CD spectra were obtained for 2 in the four
S-states (Figure 13a), which were used for the reversible
structural changes, as will be described later. Different
UV/Vis spectra were also obtained for the four S-states
(Figure 13b). A characteristic absorption at about 340 nm
appeared for the S-random coil in THF and S-amido-dimer
in bromobenzene, in which the ethynyl domain was disag-
gregated. Stronger absorptions were observed at about
300 nm for the S-ethynyl-dimer in 12% DMSO/trifluorome-
thylbenzene and the S-random coil in THF, which were dis-
aggregate states at the amido domain.
Figure 11. CD spectra of 2 in 12% DMSO/trifluoromethylbenzene at dif-
ferent a) temperatures (5ꢁ10À5 m) and b) concentrations (258C). The cal-
culated spectrum (Figure S18 in the Supporting Information) is also
shown.
1H NMR analysis of 2 in [D8]THF at 2ꢁ10À4 m showed
sharp amide peaks at d=10.0 and 10.2 ppm and well-re-
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
Figure 15. Size distributions of 2 by DLS experiments in different sol-
vents (5ꢁ10À5 m, 208C).
1
amide peaks were sharp. The H NMR analysis of 2 exhibit-
ed the same trends as 1.
Four aggregate states of 2 were examined by DLS (208C,
5.0ꢁ10À5 m) (Figure 15). In THF, an average diameter of
6 nm was obtained for the random-coil state. The ethynyl-
dimer state in 12% DMSO/trifluoromethylbenzene showed
a 7 nm diameter, similar to the random-coil state. The all-
dimer state in fluorobenzene gave a peak at 50 nm, and the
amido-dimer state in bromobenzene gave a peak at 35 nm.
The sizes of the random-coil monomer and ethynyl-dimer
states for 2 were 6–7 nm, and those of the amido-dimer and
all-dimer states were 35–50 nm. This may be due to the rod-
like rigid structure of the amido-dimer state, which in-
creased the apparent diameters of aggregates of the all-
dimer and amido-dimer states. A larger diameter was ob-
tained for the all-dimer state in fluorobenzene than for the
amido-dimer state in bromobenzene. This may be ascribed
to the folding of the latter at the disaggregated ethynyl
domain.
Figure 13. Spectra of four aggregate S-states of 2 (258C): a) CD and b)
UV/Vis spectra. Concentration of 2 in 12% DMSO/trifluoromethylben-
zene, 5ꢁ10À6 m, and in other solvents, 2.5ꢁ10À6 m.
AFM analysis of the aggregates of 2 was conducted
(Figure 16). AFM images of the samples prepared from
chloroform showed dispersed particles of about 30 nm diam-
eter, which should indicate an all-dimer state. The samples
prepared from bromobenzene provided particles of an
amido-dimer of approximately 20 nm diameter. Particles of
uniform sizes were formed in wide surfaces for both experi-
ments. The results were in good accordance with those ob-
tained by DLS. The sizes were similar with the all-dimer of
1, which was confirmed to be dimeric by VPO (vide supra).
The observations are consistent with dimeric aggregate for-
mation of 2. The random-coil and ethynyl-dimer states of 2
prepared in THF and 12% DMSO/trifluoromethylbenzene
solution produced no evident particles, which may be due to
the small size of the particles and/or rapid molecular mo-
tions.
Figure 14. 1H NMR spectra of 2 in different solvents (2ꢁ10À4 m, 258C): a)
[D8]THF, b) CDCl3, c) [D5]PhBr, and d) 12% [D6]DMSO/[D5]PhCF3.
Compounds 1 and 2 formed four dimeric aggregate states,
and the ethynyl and amido domains independently aggregat-
ed and disaggregated. It was also noted that 1 and 2 aggre-
gated and disaggregated in a two-state manner. This is a no-
table example of synthetic multidomain compounds with
well-defined aggregate structures.
solved aromatic peaks (Figure 14), which are due to the
rapid molecular motions. It was confirmed by CD that the
all-dimer state was formed in chloroform and fluorobenzene
(Figure 10 and Figure S21 in the Supporting Information).
In CDCl3 and [D5]bromobenzene, broad amide protons
were observed by 1H NMR spectroscopy. In 12%
[D6]DMSO/[D5]CF3Ph, in which CD (258C, 1ꢁ10À4 m) anal-
ysis indicated ethynyl-dimer formation (Figure 11), the
Structural change of tridomain compound: It was considered
interesting to examine the structural switch of 2 between
different states, particularly, the thermoresponding structural
&
8
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Multidomain Structures
FULL PAPER
Figure 17. a) CD spectra (12% DMSO/trifluoromethylbenzene, 5.0ꢁ
10À5 m) of 2 at 5 and 808C. A calculated spectrum obtained by adding the
spectra of 2:10 of the ethynyl-dimer and 8:10 of the random coil is also
shown. b) The profiles of De at 367 nm and the time of 2 (12% DMSO/
trifluoromethylbenzene, 5.0ꢁ10À5 m) for repeating cycles of 80/58C every
15 min.
Figure 16. AFM images (height mode) of 2 on mica: a)–c) obtained in
chloroform (2.5ꢁ10À8 m), d)–f) obtained in bromobenzene (2.5ꢁ10À8 m).
c) Section analysis along line A–B. f) Section analysis along line C–D.
change at the ethynyl domain. As noted in the experiment
using 12% DMSO/trifluoromethylbenzene, 2 changed its
structure between the ethynyl-dimer and random-coil mono-
mer states on heating (Figure 11), and this structural change
was examined in detail. At 58C, 2 formed the S-ethynyl-
dimer, as indicated by the CD (5ꢁ10À5 m) with a De value at
367 nm of +1700 cmÀ1 mÀ1 (Figure 11a). When the solution
was heated at 808C for 30 min, the De value decreased to
+80 cmÀ1 mÀ1. A ratio of 2:8 of the ethynyl-dimer to the
random coil was estimated at 808C on the basis of the De
value of +1730 cmÀ1 mÀ1 of the S-ethynyl-dimer (Figure 11)
and that of À221 cmÀ1 mÀ1 of the S-random coil (Figure S14
in the Supporting Information). The calculated CD spectra
obtained by adding those of the ethynyl-dimer and the
random coil in a 2:8 ratio was in good agreement with the
experimental spectra (Figure 17a). When the solution was
cooled to 58C, the CD spectra returned to the S-ethynyl-
dimer state with a De value of +1700 cmÀ1 mÀ1. The process
of heating at 808C and cooling at 58C was reversible and re-
producible, as indicated by the De time profiles (Fig-
ure 17b). The results indicate that the structural change in-
volved the ethynyl-dimer and random-coil states, and not
the all-dimer or amido-dimer state (Figure 18). In the polar
solvent, the amido domain of 2 remained disaggregated, and
Figure 18. Structure change of 2 between ethynyl-dimer and random coil.
only two of the four structural states of 2 were mutually
switched.
As noted before, 2 with the all-dimer state changed its
structure on heating (Figure S17 in the Supporting Informa-
tion). When 2 in fluorobenzene (2.5ꢁ10À6 m) was heated at
808C and then cooled to 58C, a De value of +720 cmÀ1 mÀ1
at 367 nm was obtained. Using the De values of the all-
dimer (+871 cmÀ1 mÀ1) and amido-dimer (À228 cmÀ1 mÀ1)
states, which were obtained from the CD spectra in fluoro-
benzene and bromobenzene, respectively (Figure 13), the
mixture was calculated to be an 8:2 mixture of the all-dimer
and amido-dimer states. The experimental CD spectra (2.5ꢁ
10À6 m, 258C) were in good agreement with the calculated
spectra obtained by adding those of the all-dimer and
amido-dimer states in an 8:2 ratio (Figure 19a). When the
mixture was heated to 808C, the De value rapidly decreased,
reaching a steady state within 5 min with a De value of
+270 cmÀ1 mÀ1. The equilibrium ratio of the all-dimer state
to the amido-dimer state was calculated to be 4:6, and the
calculated spectra were in good agreement with the experi-
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
rates, concentration effects, and aggregate sizes were also
different. In other words, two types of thermal response ap-
peared using a single compound, 2. The differences originat-
ed from the well-defined aggregate and disaggregate states
at the amido domain. It may be worth noting that these two
structural changes are orthogonal to each other, and that
each process proceeds only between two structures and does
not involve other structures.
Conclusion
In summary, the bidomain compound 1 and the tridomain
compound 2 were designed and synthesized. Both 1 and 2
form four well-defined aggregate states, namely, all-dimer,
amido-dimer, ethynyl-dimer, and random-coil states. The
domains in 1 and 2 independently aggregate and disaggre-
gate in a two-state manner. The reversible and reproducible
structural changes of 2 between the ethynyl-dimer/random-
coil state and the all-dimer/amido-dimer state occur with
heating and cooling. The well-defined aggregate and disag-
gregate states at the amido domain provide two different
thermal processes using a single compound, 2 (Figure 21).
Since ethynyl oligomers exhibit various phenomena such as
gelation[9] and vesicle formation,[10] multidomain compounds
should exhibit various switch behavior in higher-assembly
formation.
Figure 19. a) CD spectra (fluorobenzene, 2.5ꢁ10À6 m) of 2 at 5 and 808C.
The calculated spectra, obtained by adding the spectra of 8:10 of the all-
dimer state and 2:10 of the amido-dimer state, and 4:10 of the all-dimer
state and 6:10 of the amido-dimer state, are also shown. b) The profiles
of De at 367 nm and the time of 2 (fluorobenzene, 2.5ꢁ10À6 m) for repeat-
ing cycles of 80/58C every 15 min.
mental results (Figure 19a). The heating–cooling cycle be-
tween 80 and 58C was repeated, and De/time profiles were
obtained in a reproducible manner (Figure 19b). The results
showed that the thermal process involved the all-dimer and
amido-dimer states, not the ethynyl-dimer or random-coil
state (Figure 20). The amido domain remained aggregated
Figure 20. Structure change of 2 between all-dimer and amido-dimer.
Figure 21. Two structure change systems of 2 between all-dimer/amido-
dimer and ethynyl-dimer/random coil.
Experimental Section
during the process in the nonpolar solvent. The all-dimer/
amido-dimer change is a switching phenomenon between
the double helix and random-coil states at the central
domain with aggregation at both terminal domains. As sug-
gested by the results of the DLS (Figure 15) and AFM
(Figure 16) measurements, this heating and cooling process
should induce the structure change of 2 on a 10 nm scale.
The ethynyl-dimer/random-coil switch in the polar solvent
(Figure 17) and all-dimer/amido-dimer switch in the nonpo-
lar solvent between 5 and 808C (Figure 19) exhibited differ-
ent dynamic properties. The CD spectral changes in the
processes were quite different, and the thermal response
Bidomain oligomer 1: Under an argon atmosphere, a mixture of 17
(9 mg, 0.0031 mmol), a tris(dibenzylideneacetone)dipalladium(0) chloro-
form adduct (0.81 mg, 0.00078 mmol), cuprous iodide (1.7 mg, 0.0093
mol), trimesitylphosphine (1.8 mg, 0.0047 mol), tetrabutylammonium
iodide (11 mg, 0.031 mmol), triethylamine (0.3 mL), and N,N-dimethyl-
formamide (3 mL) was also freeze-evacuated three times. A solution of 6
(12 mg, 0.0031 mmol) in THF (3 mL) was freeze–evacuated three times
and then added dropwise to the above solution. The mixture was stirred
for 9 h at 458C. The reaction was quenched by adding saturated aqueous
ammonium chloride (20 mL), and the organic materials were extracted
with toluene (30 mL). The organic layer was washed with water (15 mL)
and brine (15 mL), and dried over magnesium sulfate. The solvents were
&
10
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Multidomain Structures
FULL PAPER
13C5H554N18O48
12
evaporated under reduced pressure. Separation by recycling GPC and
silica gel chromatography gave 1 (11 mg, 0.0016 mmol, 53%). M.p.: 222–
2268C (hexane/chloroform); [a]2D6 = +2790 (c=0.18 in chloroform within
1 h after dissolution); 1H NMR (600 MHz, [D8]THF, 1 mm at 608C): d=
0.85 (t, J=6 Hz, 33H), 1.27–1.53 (m, 154H), 1.75–1.87 (m, 22H), 1.98 (s,
66H), 3.91 (s, 3H), 4.27 (quint, J=6 Hz, 4H), 4.33 (m, 6H), 4.41 (m,
12H), 7.47 (d, J=5 Hz, 8H), 7.51 (d, J=5 Hz, 14H), 7.60–7.74 (m, 22H),
7.89 (d, J=7 Hz, 1H), 7.93 (s, 1H), 8.03 (d, J=7 Hz, 1H), 8.07 (s, 1H),
8.11 (s, 1H), 8.17–8.40 (m, 40H), 8.53 (d, J=8 Hz, 1H), 8.58 (d, J=8 Hz,
12H), 8.70 (s, 1H), 8.80 (s, 2H), 8.81 (d, J=9 Hz, 1H), 9.81 (s, 1H),
10.0 ppm (s, 8H); 13C NMR (151 MHz, [D8]THF): d=14.4, 23.4, 23.5 (2
peaks), 27.0, 27.8, 28.7, 29.7, 29.8, 30.2 (2 peaks), 30.3, 30.5, 32.8 (2
peaks), 52.3, 65.5, 65.7, 66.3, 80.0, 90.2, 93.7, 93.8, 115.4, 115.8, 115.8,
116.9, 120.9, 121.0, 124.1, 124.6, 125.4, (2 peaks) 127.7 (2 peaks), 127.9,
128.0, 128.4, 129.6, 129.7, 129.8, 130.9, 131.1, 131.2, 132.0, 132.2, 132.3,
132.6, 132.8, 133.0, 133.3, 136.2, 136.3 (3 peaks), 136.4 (2 peaks), 136.5
137.6 (2 peaks), 137.7, 137.8 (2 peaks), 138.9, 139.0, 141.2, 141.4, 141.6,
153.7, 165.4, 166.5 (2 peaks), 168.1 ppm; IR (KBr): n˜ =3284, 2922, 2852,
a-cyano-4-hydroxycinnamic acid): m/z calcd for
C :
539
8111.16; found: 8110.47 [MÀH]À; elemental analysis calcd (%) for
C544H554N18O48: C 80.54, H 6.88, N 3.11; found: C 80.40, H 7.07, N 2.96.
Acknowledgements
We thank Professor Hiroyuki Isobe (Department of Chemistry, Graduate
School of Sciences, Tohoku University) for DLS use. This work was fi-
nancially supported by
a Grant-in-Aid for Scientific Research (No.
21229001). Financial support and a grant for research assistantship from
the WPI-AIMR Fusion Research and the G-COE program from JSPS
are acknowledged. W. Ichinose thanks the JSPS for a Fellowship for
Young Japanese Scientists.
2206, 1724, 1681, 1651 cmÀ1
; UV/Vis (trifluoromethylbenzene, 2.5ꢁ
[1] G. A. Petsko, Protein Structure and Function, New Science Press,
London, 2004.
10À6 m): lmax (e)=310 nm (3.8ꢁ105 cmÀ1 mÀ1); CD (trifluoromethylben-
zene, 2.5ꢁ10À6 m):
l
(De)=307 (À1519), 328 (À1523), 366 nm
[2] For examples, see: a) S. Lecommandoux, M.-F. Achard, J. F. Langen-
walter, H.-A. Klok, Macromolecules 2001, 34, 9100–9111; b) A.
Sꢂnchez-Ferrer, R. Mezzenga, Macromolecules 2010, 43, 1093–1100.
[3] For examples, see: a) S. Dai, P. Ravi, C. Y. Leong, K. C. Tam, L. H.
Gan, Langmuir 2004, 20, 1597–1604; b) E. P. Holowka, D. J.
Pochan, T. J. Deming, J. Am. Chem. Soc. 2005, 127, 12423–12428;
c) X.-D. Xu, Y. Jin, Y. Liu, X.-Z. Zhang, R.-X. Zhuo, Colloids Surf.
B 2010, 81, 329–335; d) I. Otsuka, K. Fuchise, S. Halila, S. Fort, K.
Aissou, I. Pignot-Paintrand, Y. Chen, A. Narumi, T. Kakuchi, R.
Borsali, Langmuir 2010, 26, 2325–2332; e) S. de Medeiros Modolon,
I. Otsuka, S. Fort, E. Minatti, R. Borsali, S. Halila, Biomacromole-
cules 2012, 13, 1129–1135.
[4] For examples, see: a) J. D. Hartgerink, E. Beniash, S. I. Stupp, Sci-
ence 2001, 294, 1684–1688; b) G. John, J. H. Jung, H. Minamikawa,
K. Yoshida, T. Shimizu, Chem. Eur. J. 2002, 8, 5494–5500; c) Y.-B.
Lim, K.-S. Moon, M. Lee, Angew. Chem. 2009, 121, 1629–1633;
Angew. Chem. Int. Ed. 2009, 48, 1601–1605; d) R. N. Shah, N. A.
Shah, M. M. D. R. Lim, C. Hsieh, G. Nuber, S. I. stupp, Proc. Natl.
Acad. Sci. USA 2010, 107, 3293–3298.
[5] For examples, see: a) S. Bhattacharya, Y. Krishnan-Ghosh, Chem.
Commun. 2001, 185–186; b) A. P. Nowak, V. Breedveld, L. Pakstis,
B. Ozbas, D. J. Pine, D. Pochen, T. J. Deming, Nature 2002, 417,
424–428; c) P. K. Vemula, J. Li, G. John, J. Am. Chem. Soc. 2006,
128, 8932–8938.
[6] a) H. Sugiura, Y. Nigorikawa, Y. Saiki, K. Nakamura, M. Yamagu-
chi, J. Am. Chem. Soc. 2004, 126, 14858–14864; b) H. Sugiura, M.
Yamaguchi, Chem. Lett. 2007, 36, 58–59; c) H. Sugiura, R. Ame-
miya, M. Yamaguchi, Chem. Asian J. 2008, 3, 244–260; d) R. Ame-
miya, N. Saito, M. Yamaguchi, J. Org. Chem. 2008, 73, 7137–7144.
[7] R. Amemiya, W. Ichinose, M. Yamaguchi, Bull. Chem. Soc. Jpn.
2010, 83, 809–815.
(+2663 cmÀ1 mÀ1); MS (MALDI-TOF, a-cyano-4-hydroxycinnamic acid):
12
m/z calcd for
C
13C4H443N9O34: 6592.33; found: 6590.57 [MÀH]À; ele-
452
mental analysis calcd (%) for C456H443N9O34: C 83.07, H 6.77, N 1.91;
found: C 82.95, H 6.96, N 2.07.
Tridomain oligomer 2: Under an argon atmosphere, a mixture of 17
(25 mg, 0.0086 mmol), a tris(dibenzylideneacetone)dipalladium(0) chloro-
form adduct (2.3 mg, 0.0043 mmol), cuprous iodide (4.8 mg, 0.052 mmol),
trimesitylphosphine (5.0 mg, 0.026 mmol), triphenylphosphine (3.4 mg,
0.026 mmol), tetrabutylammonium iodide (16 mg, 0.043 mmol), triethyla-
mine (0.1 mL), and N,N-dimethylformamide (1 mL) was freeze-evacuat-
ed three times. A solution of 6 (11 mg, 0.0043 mmol) in THF (1 mL) was
also freeze-evacuated three times and then added dropwise to the above
solution. The mixture was stirred for 5 h at 458C. The reaction was
quenched by adding saturated aqueous ammonium chloride (20 mL), and
the organic materials were extracted with ethyl acetate (20 mL) and tol-
uene (20 mL). The organic layer was washed with water (20 mL) and
brine (20 mL), and dried over magnesium sulfate. The solvents were
evaporated under reduced pressure. Separation by recycling GPC and
silica gel chromatography gave 2 (25 mg, 0.0031 mmol, 71%). M.p.: 251–
2558C (hexane/dichloromethane); [a]2D7 =À349 (c=0.25 in THF);
1H NMR (400 MHz, [D8]THF): d=0.84 (m, 42H), 1.27–1.51 (m, 196H),
1.76–1.89 (m, 28H), 1.96 (s, 39H), 1.97 (s, 39H), 4.26–4.43 (m, 28H),
7.48–7.55 (m, 26H), 7.62–7.70 (m, 16H), 7.71–7.78 (m, 10H), 7.97 (s,
2H), 8.10 (s, 2H), 8.11 (s, 2H), 8.22–8.45 (m, 66H), 8.58–8.63 (m, 9H),
8.78–8.85 (m, 9H) 10.02 (s, 2H), 10.20 ppm (s, 16H); 13C NMR
(100 MHz, [D8]THF): d=14.4, 23.5, 23.5, 23.6, 27.0, 28.6, 29.7 (2 peaks),
29.8, 30.3 (2 peaks), 30.5, 32.8, 32.9, 65.5, 65.8, 66.0, 66.3, 80.0, 89.1, 90.1,
90.2, 93.7, 93.8, 94.8, 114.0, 115.1, 115.6, 116.7, 120.8, 120.9, 121.1, 121.4,
124.1, 124.6, 125.1, 125.4, 127.1, 127.5, 127.7 (2 peaks), 127.9, 128.0, 128.1,
128.2, 128.3, 128.4, 128.5, 129.8, 130.3, 130.7, 130.9, 131.0, 131.1 (2 peaks),
131.9, 132.1, 132.2, 132.3, 132.4, 132.6, 132.7, 132.8, 132.9, 133.0, 133.2 (2
peaks), 133.3, 135.8, 136.2 (2 peaks), 136.3, 137.5 (2 peaks), 137.8 (2
peaks), 139.0, 141.2, 141.4, 141.6, 153.6, 165.4, 165.8, 166.5 (2 peaks),
167.9, 168.1, 168.2 ppm; IR (KBr): n˜ =3319, 2924, 2852, 2207, 1724, 1681,
1652 cmÀ1; UV/Vis (fluorobenzene, 2.5ꢁ10À6 m): lmax (e)=308 nm (5.3ꢁ
105 cmÀ1 mÀ1); CD (fluorobenzene, 2.5ꢁ10À6 m): l (De)=300 (À980), 323
(À169), 337 (À365), 367 nm (+871 cmÀ1 mÀ1); MS (MALDI-TOF,
[8] See Supporting Information.
[9] M. Mizutani, R. Amemiya, M. Yamaguchi, Angew. Chem. 2010, 122,
2039–2043; Angew. Chem. Int. Ed. 2010, 49, 1995–1999.
[10] N. Saito, M. Shigeno, M. Yamaguchi, Chem. Eur. J. 2012, 18, 8994–
9004.
Received: May 31, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
Molecular Switching
W. Ichinose, M. Shigeno,
M. Yamaguchi* ............... &&&& — &&&&
Multiple States of Dimeric Aggregates
Formed by (Amido–ethynyl)helicene
Bidomain Compound and (Amido–
ethynyl–amido)helicene Tridomain
Compound
Domain switching! An (amido–ethy-
nyl)helicene bidomain compound and
an (amido–ethynyl–amido)helicene tri-
domain compound were synthesized.
Four aggregate states of multidomain
compounds, namely, all-dimer, amido-
dimer, ethynyl-dimer, and random-coil
states, were obtained in different sol-
vents. Each domain independently
aggregated and disaggregated in a two-
state manner. Two different reversible
structural changes occurred between
the ethynyl-dimer/random-coil and the
all-dimer/amido-dimer with heating
and cooling.
Molecular Switching
Well-defined aggregate formation of multidomain linear
oligomers is described by M. Yamaguchi et al. in their Full
Paper on page && ff. Multidomain compounds designed
on the basis of amido and ethynyl oligomers to form
dimeric aggregates with properties orthogonal to each
other. Four aggregate states and two reversible structural
change processes were obtained using a single compound.
&
12
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!