Molecular thermal hysteresis in helix-dimer formation of sulfonamidohelicene oligomers in solution

Article abstract of DOI:10.1002/chem.201204556

Sulfonamidohelicene oligomers up to the nonamer level were synthesized by the repeated coupling reactions of a building block. A tetramer formed a helix dimer in 1,3-difluorobenzene, which unfolded to a random coil with heating. This structural change exhibited thermal hysteresis in which different thermal responses were observed in the course of temperature increase and decrease. The feature of the hysteresis was examined under different heating/cooling modes, and the mechanisms are discussed on the basis of the population change and the presence of an induction period. A proposal regarding the use of thermal hysteresis for sensing a temperature increase/decrease is also given. Copyright

Full text of DOI:10.1002/chem.201204556

FULL PAPER
DOI: 10.1002/chem.201204556
Molecular Thermal Hysteresis in Helix-Dimer Formation of
Sulfonamidohelicene Oligomers in Solution
Masanori Shigeno, Yo Kushida, and Masahiko Yamaguchi*[a]
Abstract: Sulfonamidohelicene oligom-
which different thermal responses were
observed in the course of temperature
increase and decrease. The feature of
the hysteresis was examined under dif-
ers up to the nonamer level were syn-
thesized by the repeated coupling reac-
tions of a building block. A tetramer
formed a helix dimer in 1,3-difluoro-
benzene, which unfolded to a random
coil with heating. This structural
change exhibited thermal hysteresis in
ferent heating/cooling modes, and the
mechanisms are discussed on the basis
of the population change and the pres-
ence of an induction period. A propos-
al regarding the use of thermal hystere-
sis for sensing a temperature increase/
decrease is also given.
Keywords: dimerization
structures oligomers
mides · thermodynamics
·
helical
sulfona-
·
·
Introduction
obtained under different salt concentrations, and the rela-
tionship between the parameters and nucleic acid structures
has been discussed. Poly(N-isopropylacrylamide) also
showed thermal hysteresis in dilute aqueous solution during
globule to coil transition.[6] Thermal hysteresis of molecules
in nonpolar organic solvents, in which solvent/solvent and
solvent/solute interactions are much weaker than those in
water, is essentially unknown, with a rare exception report-
ed previously in our laboratory.[7] Bis(ethynylhelicene)
oligomers exhibited thermal hysteresis in intramolecular ag-
gregation and disaggregation in solution, which was not ob-
served in intermolecular aggregations. The mechanism of ag-
gregate formation was examined, and the involvement of
different mechanisms in aggregation and disaggregation has
been suggested by the presence of hysteresis.
Thermal hysteresis is a phenomenon in which different re-
sponses occur during heating and cooling. It is observed in
condensed phases of solid states,[1] neat liquids,[2] bulk poly-
mers,[3] or aggregates[4] (Figure 1). It is considered to be in-
duced by strong and multiple cooperative interactions be-
The above studies showed the phenomenon of thermal
hysteresis, but until now no study of molecular thermal hys-
teresis itself has been conducted. Thermal hysteresis exhibit-
ed by molecules in solution can have a broad applicability
as will be proposed in the last part of this work. It is thus
critical to study hysteresis in nonpolar solvents with regard
to intermolecular structural changes in which hysteresis fea-
tures can explicitly appear.
Figure 1. Schematic presentation of a) thermal response and b) thermal
hysteresis.
tween molecules. The thermal hysteresis of molecules in sol-
ution in which such multiple cooperative interactions do not
occur is quite rare. It was observed for oligonucleotides
during triplex formation and dissociation from a duplex in
water.[5] Kinetic and thermodynamic parameters have been
Synthetic oligomeric compounds, which change their con-
formation or aggregate structure in response to external
stimuli, have been developed by connecting small molecular
units with functional groups such as an alkyne or an
amide.[8] The use of these rigid groups has an advantage in
that the formation of ordered structures is more entropically
favored than that of flexible structures. Sulfonamides (i.e.,
sulfur analogues of amides) have not been employed for this
purpose despite their structural similarity to amides. It is
considered that a sulfonamide group tends to interfere with
the formation of ordered structures relative to an amide
group because of conformation flexibility; such an example
was reported for b-peptidosulfonamides with b peptides.[9]
The interference was attributed to the smaller barrier of ro-
[a] Dr. M. Shigeno, Y. Kushida, 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.201204556.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Figure 2. Structures of helicene oligomers connected by two-atom linkers.
À
À
tation at the S N bond than that at the amide C N bond.
Results and Discussion
À
An ab initio study showed barriers of rotation at the S N
À1
À
bond of sulfonamide (6–8.5 kcalmol ) and at the C N
Synthesis: Sulfonamidohelicene oligomers (M)-1 to (M)-9
were synthesized by repeated amide formation and depro-
tection (Scheme 1). The helicenedisulfonyl dichloride (M)-
10 was synthesized from the known helicenedialdehyde
(Scheme S1 in the Supporting Information). The sulfonyl
chloride building block (M)-12 was obtained by the reaction
of (M)-10 with 11 (1 equiv), and isolated in 50% yield by
silica gel chromatography. Oligomers were synthesized by
the reaction of amine-terminated sulfonamidohelicene
oligomers with (M)-12 (2 equiv). The t-butyloxycarbonyl
(Boc) group was then removed with trifluoroacetic acid.
Such a sequence of coupling and deprotection reaction af-
forded oligomers from a monomer to a heptamer and a non-
amer in high yields (Scheme 1).
bond of formamide (18.2–22.5 kcalmolÀ1).[10] Also note that
À
À
sulfonamide N H is a stronger proton donor than amide N
H, but sulfonamide S=O is a weaker proton acceptor than
amide C=O, which weakens hydrogen-bonding networks.[11]
We have been studying the helix dimer formation of heli-
cene oligomers (Figure 2).[12–14] Ethynylhelicene oligomers
form double helixes in hard aromatic solvents such as tri-
fluoromethylbenzene or 1,3-difluorobenzene, and unfold in
soft solvents such as bromobenzene and iodobenzene.[15,16]
Amidohelicene and reverse-amidohelicene oligomers also
form helix dimers in nonpolar solvents such as 1,3-difluoro-
benzene and CHCl3, and unfold in hydrogen-bond-breaking
solvents such as DMSO.[17,18] These results showed the im-
portant role of two-atom linkers in helix dimer formation,
and the modification of two-atom linkers was expected to
provide a variety of oligomers that form helix dimers. Thus,
the use of flexible sulfonamide linkers was considered inter-
esting.
Helix dimer formation: The CD spectra of sulfonamidoheli-
cene oligomers were obtained in CHCl3 at a concentration
of 5.0ꢁ10À4 or 1.0ꢁ10À3 m (Figures S1 and S2 in the Sup-
porting Information). The Cotton effect of the monomer
(M)-1 to the heptamer (M)-7 and the nonamer (M)-9
showed a monotonic increase with the number of helicenes,
and slight deviations were observed for the tetramer (M)-4
and the pentamer (M)-5. Very small changes were observed
in the CD spectra between 5 and 508C for these compounds.
Vapor pressure osmometry (VPO) studies of (M)-4 at 5.0ꢁ
10À4 m (CHCl3, 358C) showed a monomeric nature of (M)-4:
observed molecular weight, (3.1Æ0.2)ꢁ103; calculated mon-
omer weight, 3184; N=(0.97Æ0.06). In CHCl3, the sulfona-
mide oligomers were in the monomeric and random coil
states.
Described in this study are the synthesis of sulfonamido-
helicene oligomers (M)-n (n=1–7, 9) and helix dimer for-
mation of (M)-4. Among oligomers up to the nonamer level,
only a tetramer forms an ordered structure of a helix dimer
in a fluorinated aromatic solvent, which unfolds with heat-
ing. This is an interesting example of the formation of an or-
dered structure using a sulfonamide oligomer.
Notably, the intermolecular structural change between a
helix dimer and a random coil exhibits thermal hysteresis.
The feature of molecular thermal hysteresis was revealed
under different cooling/heating modes. The phenomena
were explained by different populations of helix dimers and
random coils at the same temperature during heating and
cooling. A model was provided to explain the phenomenon
using the molecular feature. A proposal on the use of ther-
mal hysteresis for sensing a temperature increase/decrease is
also given.
In 1,3-difluorobenzene, the same spectra as those in
CHCl3 were obtained for the hexamer (M)-6 and the hep-
tamer (M)-7 (Figure 3 and Figure S3 in the Supporting In-
formation). The spectra of the trimer (M)-3 and the pentam-
er (M)-5 were slightly different from those in CHCl3. In con-
trast, the tetramer (M)-4 provided quite a different spec-
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Thermal Hysteresis
FULL PAPER
CD analysis (258C) of (M)-4
at
a concentration of 2.5ꢁ
10À4 m showed a helix dimer in
a steady state (Figure S6 in the
Supporting Information). At
lower concentrations of 5ꢁ10À5
and 5ꢁ10À6 m, the CD spectra
indicated a random coil state of
(M)-4 (Figure S7 in the Sup-
porting Information).
The CD spectra (258C, 5.0ꢁ
10À4 m) of (M)-4 in fluoroben-
zene showed modest intensity
of the Cotton effects, and the
spectra in benzene, toluene,
bromobenzene, and DMSO
showed random coil states (Fig-
ures S8 and S9 in the Support-
ing Information).
On the basis of the above
study, VPO analysis was con-
ducted at 45 and 558C in 1,3-di-
Scheme 1. Synthesis of sulfonamidohelicene oligomers (M)-1–7 and 9.
fluorobenzene
(5.0ꢁ10À4 m).
The solution of (M)-4 was
heated at 708C for 30 min, cooled to 458C for 720 min, and
a steady state was reached (Figure S4 in the Supporting In-
formation). The VPO analysis (458C) of this sample showed
dimeric aggregate formation: observed molecular weight,
(6.2Æ0.4)ꢁ103; calculated molecular weight of dimer, 6368
(N=(1.9Æ0.1)). VPO analysis at 558C was also conducted.
The solution of (M)-4 in 1,3-difluorobenzene (5.0ꢁ10À4 m)
was heated at 708C for 60 min, cooled to 558C, and allowed
to settle for 60 min (Figure S10 in the Supporting Informa-
tion). The monomeric state was obtained for this sample:
observed molecular weight, (3.9Æ0.2)ꢁ103; calculated mo-
lecular weight of monomer, 3184 (N=(1.2Æ0.1)). These re-
sults indicated the helix dimer formation of (M)-4 at 5.0ꢁ
10À4 m in 1,3-difluorobenzene below 458C. Also note that
(M)-4 showed a sharp structure transition between 45 and
508C.
trum as exhibited by the strong negative and positive
Cotton effects at 302 and 320 nm, respectively. This suggest-
ed that (M)-4 formed an ordered structure in this solvent.
To analyze the ordered structure, CD spectra were ob-
tained at different temperatures in 1,3-difluorobenzene
(Figure 4). A solution of (M)-4 (5.0ꢁ10À4 m) was first heated
at 708C for 25 min, and then cooled to 508C for 25 min. The
De (320 nm) values at 70 and 508C were confirmed not to
change after 10–20 min, which were considered to be steady
states. When the solution was cooled from 50 to 458C, it
took 540 min to reach a steady state with a De value of
740mÀ1 cmÀ1 (Figure S4 in the Supporting Information).
Then the solution was heated to 508C, and a steady state
was obtained after 200 min with a De of 620mÀ1 cmÀ1. The
CD spectra at 70 and 508C were random coils (Figure 4 and
Figure S1 in the Supporting Information). At 458C, the CD
spectra in the steady state showed an ordered structure (Fig-
ures 3 and 4), which is a helix dimer as will be noted later
by VPO analysis. When the solution was heated from 45 to
508C, the CD spectra remained a helix dimer. Thus, two dif-
ferent CD spectra of random coil and helix dimer were ob-
tained at 508C during cooling and heating, respectively. The
results suggested thermal hysteresis, which will be discussed
later.
VPO analyses were also conducted under the following
conditions to confirm the dimeric aggregate formation with-
out the formation of higher aggregates. The solution of (M)-
4 in 1,3-difluorobenzene was heated at 708C for 60 min,
cooled to 358C, and allowed to settle for 720 min. The di-
meric states were obtained by VPO: observed molecular
weight, (6.6Æ0.4)ꢁ103 (N=(2.1Æ0.1)) at 5.0ꢁ10À4 m and
(5.9Æ0.1)ꢁ103 (N=(1.8Æ0.1)) at 3.5ꢁ10À4 m. The solution
of (M)-4 in 1,3-difluorobenzene (5.0ꢁ10À4 m) was heated at
708C for 60 min, cooled to 458C for 600 min, cooled to
258C for 120 min, warmed to 458C, and allowed to settle for
60 min: observed molecular weight, (6.7Æ0.5)ꢁ103 (N=
(2.1Æ0.2)).
A structure-switching experiment was conducted using
the De (320 nm) values under several heating and cooling
cycles (Figure S5 in the Supporting Information). The solu-
tion of (M)-4 (5.0ꢁ10À4 m) in 1,3-difluorobenzene was
heated to 708C for 40 min, cooled to 508C for 20 min, and
cooled to 458C. At 45 8C, De started to increase, and
reached a steady state after 8 h with De= +800mÀ1 cmÀ1.
When the mixture was heated to 708C, De rapidly decreased
to À120mÀ1 cmÀ1.
Dynamic light scattering (DLS) analysis in chloroform
and 1,3-difluorobenzene (258C, 5.0ꢁ10À4 m) provided aver-
age diameters of 2 nm (Figure S11 in the Supporting Infor-
mation). No change was observed in the CD spectra by fil-
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
Figure 4. CD spectra of (M)-4 (5ꢁ10À4 m, 1,3-difluorobenzene). The solu-
tion was heated at 708C (25 min; brown line), cooled to 508C (25 min;
red line), cooled to 458C (540 min; green line), and warmed to 508C
(200 min; blue line). The CD spectra were obtained at each temperature
after allowing the solution to stand for the times shown in parentheses,
when the De change was less than 2mÀ1 cmÀ1 within a 10 min period.
in the Supporting Information). The same heating/cooling
experiment as that shown in Figure 4 was conducted. The
solution of (M)-4 in 1,3-difluorobenzene (5.0ꢁ10À4 m) was
heated at 708C for 25 min, cooled to 508C for 20 min,
cooled to 458C for 540 min, warmed to 508C for 200 min,
and warmed to 708C for 20 min. At 708C and the initial
508C, at which (M)-4 was in a random coil state, the intensi-
ty at 430 nm was not strong. At 458C and the second 508C,
at which a helix dimer was formed, fluorescence ten times
stronger at 430 nm was observed. Also note that two differ-
ent fluorescence spectra were obtained at 508C during cool-
ing and heating, which indicated thermal hysteresis.
Figure 3. CD (top) and UV/Vis (bottom) spectra of (M)-n (n=3–6 and 7)
in 1,3-difluorobenzene. Concentrations: (M)-3 (5ꢁ10À4 m), (M)-4 (5ꢁ
10À4 m), (M)-5 (5ꢁ10À4 m), (M)-6 (2.5ꢁ10À4 m), and (M)-7 (2.5ꢁ10À4 m).
For (M)-3, the solution was heated at 708C (40 min) and then cooled to
50 (40 min), 40 (40 min), 25 (100 min), 15 (200 min), and 58C (50 min).
For (M)-5, the solution was heated at 708C (50 min) and then cooled to
50 (240 min), 45 (240 min), 40 (240 min), and 258C (440 min). For (M)-6
and (M)-7, the solutions were heated at 708C (40–50 min) and then
cooled to 50 (30–35 min), 45 (20 min), 40 (20–30 min), 25 (25–40 min), 15
(30 min), and 58C (30 min). The CD and UV/Vis spectra were obtained
at each temperature after allowing the solution to stand for the times
shown in parentheses, when the De (320 nm) and e (304 nm) changes
were less than 2 and 1ꢁ103 mÀ1 cmÀ1 within a 10 min period, respectively
(Figure S3 in the Supporting Information). For (M)-4, the solution was
first heated at 708C (25 min) and then cooled to 65 (15 min), 60 (10 min),
55 (15 min), 50 (40 min), 45 (450 min), 40 (115 min), 25 (200 min), and
58C (45 min). This experiment is part of the low-rate experiment in
Figure 5. The CD and UV/Vis spectra were obtained at each temperature
after allowing the solution to stand for the times shown in parentheses,
when the De (320 nm) and e (305 nm) changes were less than 2 and 1ꢁ
103 mÀ1 cmÀ1 within a 10 min period, respectively.
IR analysis of (M)-4 in 1,3-difluorobenzene (5.0ꢁ10À4 m)
À
at room temperature provided the SO2N H absorption at
3358 and 3261 cmÀ1, with the intensity of the latter consider-
ably stronger than the former (Figure S15 in the Supporting
Information). In chloroform (5.0ꢁ10À4 m), absorptions ap-
peared at 3378 and 3254 cmÀ1 with comparable intensities.
The absorptions at approximately 3360 cmÀ1 were assigned
À
to non-hydrogen-bonded SO2 NH, and those at approxi-
mately 3260 cmÀ1 were assigned to hydrogen-bonded SO2
À
NH.[19] The present results indicated the increase in hydro-
À
gen-bonded SO2N H in the helix dimer state.
tration of the 1,3-difluorobenzene solution (5.0ꢁ10À4 m)
through a membrane filter (200 nm) (Figure S12 in the Sup-
porting Information). The results confirmed no higher ag-
gregate formation.
The sulfonamidohelicene tetramer (M)-4 formed a helix
dimer in 1,3-difluorobenzene. Structural transformation be-
tween a helix dimer and a random coil took place with
changes in temperature, concentration, and solvent. Note
that the methodology of helix-dimer formation using heli-
cene oligomers connected by a two-atom linker was applica-
ble to a flexible sulfonamide group. Various helix-dimer-
forming oligomers might be obtained by this method.
The 1H NMR spectrum of (M)-4 in [D6]DMSO (258C,
5.0ꢁ10À4 m) showed sharp peaks consistent with a random
coil. The spectrum in [D5]fluorobenzene (5.0ꢁ10À4 m) was
extremely broadened at 258C, and some peaks appeared at
608C (Figure S13 in the Supporting Information).
The reason for helix-dimer formation only for (M)-4 and
not for other oligomers is not clear at present. It might be
due to the balance between enthalpic and entropic costs:
The trimer (M)-3 did not aggregate on account of the insuf-
Fluorescence analysis of (M)-4 in 1,3-difluorobenzene
(5.0ꢁ10À4 m) showed different intensities between helix
dimer and random coil at different temperatures (Figure S14
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Thermal Hysteresis
FULL PAPER
ficiency in enthalpic gain, and oligomers longer than the tet-
ramer (M)-4 also did not aggregate on account of a large en-
tropic loss during aggregation.[20] Alternatively, it can be as-
cribed to very slow aggregation.[21]
cooling rate of 0.5 KminÀ1 provided another thermal hyste-
resis curve with a smaller area in which De changed between
150 and À140mÀ1 cmÀ1 (0.5 KminÀ1 experiment). Irrespec-
tive of the cooling/heating rate, the same De value of
À140mÀ1 cmÀ1 was reached at 658C, at which all the mole-
cules are random coils. The De reached at 58C is dependent
on the heating/cooling rate, and decreased with an increase
in the heating/cooling rate. The states should be frozen in
terms of equilibrium, and the helix dimer and random coil
could not interconvert at the low temperatures. The experi-
ments can also be described by triangular waves of thermal
input and De output (Figure S19 in the Supporting Informa-
tion).
The above experiments exhibited kinetic aspects of a
structural change. To probe the equilibrated state, a steady-
state experiment was conducted at each temperature
(Figure 5). The solution of (M)-4 (5.0ꢁ10À4 m, 1,3-difluoro-
benzene) was first heated at 708C, and then cooled to 65,
60, 55, 50, 45, 40, 25, and 58C in a stepwise manner (low-
rate experiment). De (320 nm) at each temperature was ob-
tained after the times shown in parentheses (Figures S17
and S18 in the Supporting Information) when the De change
was less than 2mÀ1 cmÀ1 within a 10 min period. During cool-
ing from 70 to 508C, the De value of approximately
À150mÀ1 cmÀ1 remained constant; at 458C, a De value of
760mÀ1 cmÀ1 was obtained, which took 9 h to reach a con-
stant value; at 408C, a De value of 830mÀ1 cmÀ1 was ob-
tained; at 25 and 58C, a De value of 970mÀ1 cmÀ1 was ob-
tained (cooling curve). Then, the solution was warmed from
5 to 25, 30, 40, 45, 46, 48, 50, 55, 60, 65, and 708C again in a
stepwise manner. Between 5 and 308C, the De value of
980mÀ1 cmÀ1 remained constant; between 40 and 608C, the
Thermal hysteresis: As shown in Figure 4, two different
structures, or De values at 320 nm, appeared in the steady
state at 508C depending on the thermal process. Cooling
from 70 to 508C caused the formation of a random coil with
De=À150mÀ1 cmÀ1; cooling from 70 to 458C and warming to
508C caused the formation of a helix dimer with De=
620mÀ1 cmÀ1. The results suggested thermal hysteresis in
which cooling and heating of (M)-4 occurred in different
manners. The De value at 320 nm was used in the following
analysis of thermal hysteresis.
To obtain hysteresis curves, responses under a constant
De value decreased; above 658C,
a
De value of
Figure 5. Profiles of De (320 nm)/T for (M)-4 (1,3-difluorobenzene, 5ꢁ
10À4 m). In the 0.25 KminÀ1 experiment, the temperature was changed
À150mÀ1 cmÀ1 was reached (heating curve). The results
showed apparent thermal hysteresis between 50 and 608C
under the steady-state conditions (Figure 5). The results are
unusual, since the same heating and cooling curve should be
obtained under equilibrium conditions; otherwise, two equi-
libria appear under the same conditions, which is against the
first law of thermodynamics. As will be noted below, the
rates to reach equilibrium were extremely low in cooling
from 65 to 458C because of long induction periods. This
steady-state experiment was then designated the low-rate
experiment to compare it with 0.25 and 0.5 KminÀ1 experi-
ments.
from 65 to 58C and then to 658C at
a heating/cooling rate of
0.25 KminÀ1 In the 0.5 KminÀ1 experiment, the temperature was
.
changed from 65 to 58C and then to 658C at a heating/cooling rate of
0.5 KminÀ1. In the low-rate experiment, the solution was first heated to
708C (25 min) and then cooled to 65 (15 min), 60 (15 min), 55 (15 min),
50 (40 min), 45 (450 min), 40 (115 min), 25 (200 min), and 58C (45 min).
Then the solution was warmed to 25 (75 min), 30 (65 min), 40 (115 min),
45 (120 min), 46 (75 min), 48 (120 min), 50 (160 min), 55 (115 min), 60
(240 min), 65 (55 min), and 708C (20 min). The red lines are drawn be-
tween points. Values of De were obtained at each temperature after al-
lowing the solution to stand for the times shown in parentheses, when the
De change was less than 2mÀ1 cmÀ1 within a 10 min period (Figures S17
and S18 in the Supporting Information).
cooling/heating rate were examined (Figure 5). A solution
of (M)-4 in 1,3-difluorobenzene (5.0ꢁ10À4 m) was first
heated at 658C, and was cooled from 65 to 58C at a rate of
0.25 KminÀ1 (0.25 KminÀ1 experiment). Between 65 and
408C, the De of À140mÀ1 cmÀ1 remained unchanged; be-
tween 40 and 208C, De increased; at 58C, De reached
490mÀ1 cmÀ1 (cooling curve). The solution was then warmed
from 5 to 658C at the same rate. Between 5 and 308C, the
De of 490mÀ1 cmÀ1 did not change; between 30 and 608C, De
decreased; at 658C, the De value reached À140mÀ1 cmÀ1
(heating curve). The experiment using a constant heating/
Induction period: The hysteresis was related to the presence
of induction periods at 45 and 508C during the structural
change from a random coil to a helix dimer. The De values
(320 nm) at 458C were taken from the CD data in Figure S4
in the Supporting Information, and provided the De
(320 nm)/t profiles, which showed an induction period of
30 min (Figure 6). To detect the induction period at 508C,
(M)-4 (5.0ꢁ10À4 m) in 1,3-difluorobenzene was heated at
708C for 30 min, and then cooled to 508C. The De value re-
mained approximately À150mÀ1 cmÀ1 for 6 h, and then grad-
ually increased. After 10 h, De reached À60mÀ1 cmÀ1.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
508C-2 state, De changed from 280 to 670mÀ1 cmÀ1 within
4 h, and reached 690mÀ1 cmÀ1 after 12 h (blue arrow). A
clearly higher reaction rate was observed in the 508C-2 state
at the later stage of the reaction compared to the 508C-1
state at the earlier stage. The results are consistent with the
presence of a long induction period.
Thus, the apparent thermal hysteresis under the low-rate
conditions at 45 and 508C does not imply thermal hysteresis
under equilibrium, and is due to a long induction period in
the transition of a random coil to a helix dimer (Figure 5).
The cooling curve (thin red line) of the low-rate experiment
is therefore not the equilibrated curve, but the heating curve
(thick red line) can be approximated to the equilibrated
curve.[20] The comparison of the equilibrium curve and con-
stant-rate curves indicates that (M)-4 was mostly in the
random-coil state during cooling between 55 and 408C, al-
though the helix dimer is thermodynamically more stable
than the random coil.
Figure 6. Profile of De (320 nm)/t for (M)-4 (1,3-difluorobenzene, 5ꢁ
10À4 m) at the initial state of aggregation. Blue line: The solution was
heated at 708C (40 min), cooled to 508C (20 min), and 458C, and allowed
to settle for 240 min. The De values in Figure S4 of the Supporting Infor-
mation were used. Red line: The solution was heated at 708C (30 min),
cooled to 508C, and allowed to settle for 10 h. The De in Figure S10 of
the Supporting Information were used. The inset shows the expansion of
the profile between 0 and 80 min. The lines are drawn between points.
Stop-return experiment: The De/T profiles were also ob-
tained by stop-return experiments, in which cooling was
stopped at an intermediate temperature and returned to
heating (Figure 8). The solution of (M)-4 in 1,3-difluoroben-
Figure 7. Profiles of De (320 nm)/T for (M)-4 (1,3-difluorobenzene, 5ꢁ
10À4 m) to show the presence of an induction period at 508C. The solution
was heated at 708C (45 min), cooled to 508C (designated 508C-1, green
arrow; 20 min and 4 h), and 458C (10 h), warmed to 608C (40 min), and
cooled to 508C (designated 508C-2, blue arrow; 20 min, 4 h, and 12 h).
The values of De at each temperature were obtained after allowing the
solution to stand for the times shown in parentheses, when the De change
was less than 2mÀ1 cmÀ1 within a 10 min period. Exceptionally at 608C,
De was obtained after 40 min before reaching a steady state, and the solu-
tion was cooled to 508C. The lines are drawn between points. The inset
shows the expansion of the profile in the 508C-2 state.
Figure 8. Profiles of De (320 nm)/T for (M)-4 (1,3-difluorobenzene, 5ꢁ
10À4 m) in stop-return experiments 1, 2, and 3 with heating/cooling rates
of 0.25 KminÀ1. Temperature was changed in the following manner. Stop-
return 1: decrease from 70 to 508C, increase to 708C. Stop-return 2: de-
crease from 70 to 258C, increase to 458C, decrease to 58C, increase to
708C. Stop-return 3: decrease from 70 to 438C, increase to 508C. Three
stop-return experiments were conducted twice to confirm reproducibility.
zene (5.0ꢁ10À4 m) was first heated at 708C. Then, at a rate
of 0.25 KminÀ1, the mixture was cooled to 508C and
warmed to 708C (stop-return 1). No change in De was ob-
served. Another experiment (stop-return 2) was conducted
in which the solution at 708C was cooled to 258C, warmed
to 458C, cooled to 58C, and warmed to 708C. When the sol-
ution was warmed from 25 to 458C, the De showed a small
increase and a small decrease and reached 340mÀ1 cmÀ1.
Then, cooling the solution from 45 to 58C increased De to
430mÀ1 cmÀ1. When the solution was heated to 70 from 58C,
the heating curve obtained was close to the cooling curve
To further confirm the presence of an induction period at
508C, the following experiment was conducted (Figure 7). A
solution of (M)-4 (5.0ꢁ10À4 m) in 1,3-difluorobenzene was
heated at 708C for 45 min, cooled to 508C, allowed to settle
for 4 h (designated 508C-1), cooled to 458C, allowed to
settle for 8 h, warmed to 608C, allowed to settle for 40 min,
cooled to 508C, and allowed to settle over 12 h (designated
508C-2). In the 508C-1 state, the De value of À150mÀ1 cmÀ1
did not essentially change within 4 h (green arrow). In the
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Thermal Hysteresis
FULL PAPER
between 45 and 58C. The temperature of stop-return experi-
ment 3 returned from 43 to 508C, during which De increased
from À120 to 130mÀ1 cmÀ1.
In stop-return experiment 1, the De/T profile followed the
same curve during cooling and heating. In stop-return ex-
periments 2 and 3, De showed a small increase on heating,
and did not return to that in the cooling curve. The results
indicated that 458C was the threshold of the return to the
cooling curve, and that once helix-dimer formation started,
the profile approached the heating curve.
To summarize the experiments, thermal hysteresis was ob-
served for (M)-4 during heating and cooling in 1,3-difluoro-
benzene, in which (M)-4 changed its structure between a
random coil and a helix dimer. The De/T profiles exhibited
different curves during heating and cooling under constant-
rate conditions, which was due to the presence of long in-
duction periods during cooling. An equilibrium curve was
then approximated to the heating curve in the low-rate ex-
periment, in which the cooling and heating curves should
emerge. The stop-return experiments showed that the De
value did not follow the cooling curve but approached the
heating curve after a helix dimer started to form. Note that
the thermal hysteresis of (M)-4 occurred in a nonpolar or-
ganic solvent, and that VPO and DLS analysis indicated the
absence of enhanced aggregation (Figure S11 in the Sup-
porting Information). Thus, the present system exhibits the
feature of thermal hysteresis by molecules in solution.
Figure 9. Model of the thermal hysteresis shown with De/T profiles. An
equilibrium curve was taken from the heating curve determined in the
low-rate experiment in Figure 5. A constant-rate curve was taken from
the cooling/heating curve at a rate of 0.25 KminÀ1 in Figure 5. The
states A–H in the heating and cooling curves are shown with their energy
diagram models. In the diagrams, the relative populations of molecules in
the X (i.e., helix dimer) and Y (i.e., random coil) structures are shown in
blue bars. Green arrows indicate low-barrier paths and red arrows high-
barrier paths. In the states B, D, and H, both high- (red arrow) and low-
barrier paths (green arrow) are shown. The relative populations in states
A/D/E and B/G/F are the same. The energy diagrams in states B/D/H
and C/E/G are the same. In state D, the low-barrier path is closed, and
the high-barrier path is open. In the states E and D, induction periods
were observed in the structural change from Y to X.
Mechanistic model for molecular thermal hysteresis: A
model to explain the thermal hysteresis of molecules in solu-
tion is provided, since no such attempt has been made previ-
ously. Thermal hysteresis in condensed phases of solid
states, neat liquid states, or bulk polymers occurs by strong
and multiple cooperative interactions between molecules. In
contrast, the thermal hysteresis shown in this study occurs
with a relatively nonpolar organic compound in a nonpolar
organic solvent. The phenomena must therefore be ex-
plained by the nature of molecules in the absence of multi-
ple cooperative interactions.
For a discussion of the model, the following assumptions
were made: 1) The energy diagrams are identical at the
same temperature. In other words, the energy diagrams are
not affected by a temperature increase or decrease. 2) Ther-
mal hysteresis should originate purely from the differences
between the relative populations of molecules in the X (i.e.,
helix dimer) and Y (i.e., random coil) structures during
heating and cooling, which are expressed by De.
rate experiment, states H and D possess the same energy di-
agrams as B, as in the case of C, E, and G, and the thermal
hysteresis is explained by the relative populations of mole-
cules with the X and Y structures. In high-temperature state
A, all the molecules have a Y structure. When the tempera-
ture is decreased to that in state D, all the molecules still
have a Y structure despite the fact that EX =EY, which is
due to the high-energy barrier (red arrow) between X and
Y. In state E, molecules of Y gradually change to those of X
(EX <EY), because the low-energy barrier path opens (green
arrow). In low-temperature state F, molecules have either an
X or Y structure despite the fact that EX <EY and they
cannot interconvert from each other in a frozen manner.
When the temperature increases to that in state G, the mole-
cules start to interconvert because the low-barrier path is
opened (green arrow). In state H, some molecules become
Y (EX >EY), and finally state A is reached. This model ex-
plains the De change by population, that is, the same relative
populations of X and Y in states A/D/E and B/F/G.
Under equilibrium conditions, heating and cooling should
give the same De/T profile, as shown by a red line
(Figure 9). In state A, Y is thermodynamically more stable
than X (EX >EY), and all the molecules have a Y structure.
In state C with the reversed relative thermodynamic stability
(EX <EY), all molecules have the X structure. In state B, in
which the thermodynamic stabilities of X and Y are compa-
rable (EX =EY), a 1:1 equilibrium mixture is formed.
The model needs to be consistent with experimental re-
sults. First, induction periods (Figures 6 and 7) can be ex-
plained by the presence of high- and low-barrier paths (red/
green arrows) in the formation of X from Y and by the
switch of the paths under certain conditions (Figure 9).
During the induction period (state D), the low-barrier path
(green arrow) is closed, and molecules cannot override the
high-barrier path (red arrow). The low-barrier path opens
after some time (state E), and molecules start to change
their structure from Y to X. In states G and H, the low-bar-
As noted above, it was assumed that the energy diagrams
are identical at the same temperature. Then, in the constant-
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
rier path is open, and molecules change their structure from
X to Y. A possible reason for the change of the path is self-
catalysis. A small amount of X is formed during induction
period and catalyzes the structural change. No X exists in
states E or D, in which the low-barrier path is closed, and
the low-barrier path is open in states G and H, in which X
exists in the reaction system.
Second, the results of the stop-return experiments
(Figure 8) can be explained by this model (Figure 9), in
which cooling was stopped during the temperature decrease,
and returned to heating. In stop-return experiment 1, De fol-
lows the same curve in cooling and heating, because the
low-barrier path (green arrow) is closed. In stop-return ex-
periments 2 and 3, De starts to increase and approaches the
heating curve, because the low-barrier path (green arrow)
opens. The results are again consistent with self-catalysis.
No X exists in stop-return experiment 1, but X is formed in
stop-return experiments 2 and 3. This model provides a
rough sketch of the molecular thermal hysteresis on a mo-
lecular basis.
Third, thermal hysteresis is related to the “memory
effect.”[22] Two states, D and H, appeared at the same tem-
perature of 558C (Figure 9). This means that the molecules
in state D are in the course of cooling, and that the mole-
cules were previously at high temperatures. In the state H,
they are in the course of heating, and they were previously
at low temperatures. The molecules memorize their thermal
history, that is, whether they were heated or cooled previ-
ously.
construct complex systems to detect temperature changes.
This is an advantage of the molecular thermal hysteresis in
sensing a state of temperature increase/decrease. In the case
that the X/Y structural change is transferred to other chemi-
cal events, various temperature-dependent molecular sys-
tems can be constructed. For example, plants discriminate
between morning and evening and between spring and
autumn in flowering. The presence of thermal hysteresis
molecules in plants might be suggested.
Conclusion
Sulfonamidohelicene oligomers up to the nonamer level
were synthesized, and the tetramer (M)-4 formed a helix
dimer in a fluorinated aromatic solvent, which unfolded
with heating. The thermal structural change between a helix
dimer and a random coil in a nonpolar organic solvent ex-
hibited hysteresis. Molecular thermal hysteresis was exam-
ined under different heating/cooling modes. The mechanism
was discussed on the basis of a population model of mole-
cules with helix dimer and random coil structures, which is
consistent with the presence of induction periods, stop-
return experiments, and memory effect. A proposal for ther-
mal hysteresis molecules to serve as sensors for detecting
temperature changes has also been made.
Experimental Section
The above model explains the “memory effect” in thermal
hysteresis by the different relative populations of X and Y
between states D and H, although their energy diagrams are
the same at states D and H. In state D, all the molecules
have a Y structure despite the relative thermodynamic sta-
bility EX =EY; this is because all the molecules had a Y
structure in original state A. The relative populations in
state H are also affected by the population in state F. This is
the origin of the “memory effect,” in which the population
of the past states affects the present states. A single mole-
cule cannot memorize its history because molecular phe-
nomena are reversible.
Finally, proposals on the use of thermal hysteresis mole-
cules as a sensor of a temperature increase/decrease state
are given. Such molecules respond not only to high/low tem-
peratures but also to a temperature increase/decrease. A
temperature increase at 558C can be exhibited by the helix
dimer (X) of the molecule using a De of state H, and a tem-
perature decrease at 558C by the random coil (Y) structure
using a De of state D. A notable feature of this detection
method is that a temperature increase/decrease state can be
determined by observing the molecular structure or De only
at one time point using the “memory effect” without the
need to observe the time dependences of the molecular
structure or De change at more than one time point.
Also note that detection can be conducted at the molecu-
lar level with a large structural change between a helix
dimer (X) and a random coil (Y); that is, there is no need to
Tetramer (M)-4: Pyridine (13 mL, 0.16 mmol) and (M)-12 (59 mg,
73 mmol) in dichloromethane (2 mL) were added to a solution of (M)-2H
(44 mg, 27 mmol) in dichloromethane (2 mL) under an argon atmosphere
at 08C. The mixture was then stirred at room temperature for 16 h. The
solvent was evaporated under reduced pressure. Purification by silica gel
chromatography (hexane/ethyl acetate=2:1 and 1:1) gave (M)-4 (73 mg,
23 mmol, 85%). M.p. 163–1658C (CH2Cl2/hexane); [a]2D4 =À32 (c=2.8 in
THF); 1H NMR (400 MHz, [D6]DMSO): d=0.62–0.67 (m, 9H), 0.71 (t,
J=7 Hz, 6H), 0.91–1.27 (m, 70H), 1.27 (m, 6H), 1.33 (s, 18H), 1.47–1.49
(m, 16H), 1.54 (s, 6H), 1.58 (s, 6H), 3.87 (brs, 6H), 4.05 (t, J=6 Hz,
4H), 7.13–7.16 (m, 6H), 7.24–7.32 (m, 11H), 7.52–7.56 (m, 10H), 7.70 (s,
2H), 7.80 (t, J=8 Hz, 2H), 8.59 (s, 6H), 8.78–8.88 (m, 10H), 9.49 (s,
2H), 11.07 (s, 2H), 11.12 (s, 2H), 11.17 ppm (s, 4H); 13C NMR
(100 MHz, [D6]DMSO): d=13.8, 13.9, 22.0 (2 peaks), 22.6, 22.7 (2
peaks), 25.0, 25.2, 27.7, 28.0, 28.2, 28.5, 28.6 (2 peaks), 28.8 (2 peaks),
31.1, 31.2 (2 peaks), 64.6, 79.5, 112.2, 112.9, 113.7, 121.7, 121.9, 126.7,
127.0, 127.1, 128.3, 128.6, 129.6, 129.8, 130.4 (2 peaks), 130.9, 131.2, 131.5,
133.7, 134.1, 136.8, 136.9, 138.2, 138.6, 140.7, 152.4, 164.3, 165.0 ppm; UV/
Vis (CHCl3, 5ꢁ10À4 m): lmax (e)=304 nm (2.3ꢁ105 mÀ1 cmÀ1); CD (CHCl3,
5ꢁ10À4 m):
l
(De)=282 (146), 321 (À132), 368 nm (69mÀ1 cmÀ1); IR
(KBr): n˜ =3369, 3259, 2925, 2854, 1701, 1336, 1151 cmÀ1; MS (MALDI-
TOF): m/z: calcd for C175H204N10O30S8: 3181.3; found: 3181.3; elemental
analysis calcd (%) for C175H204N10O30S8: C 66.01, H 6.46, N 4.40; found: C
65.72, H 6.66, N 4.23.
Acknowledgements
We are grateful to Professor Hiroyuki Isobe and Dr. Syunpei Hitosugi
(Graduate School of Science and Faculty of Science, Tohoku University)
for helpful discussions on DLS. We also thank Professor Kazue Kurihara
&
8
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Molecular Thermal Hysteresis
FULL PAPER
and Dr. Motohiro Kasuya (Institute of Multidisciplinary Research for
Advanced Materials, Tohoku University) for helpful discussions on re-
fractive index measurements. This work was financially supported by a
Grant-in-Aid for Scientific Research (no. 21229001), and the GCOE pro-
gram from the Japan Society for the Promotion of Science (JSPS). Finan-
cial support in the form of a Grant-in-Aid for Scientific Research for
Young Scientists (B) (no. 23790003) from the Ministry of Education, Cul-
ture, Sports, Science and Technology, Japan to M.S. is also acknowledged.
[10] a) Y. K. Kang, H. S. Park, J. Mol. Struct. 2004, 676, 171–176; b) J.
Heyd, W. Thiel, W. Weber, J. Mol. Struct. 1997, 391, 125–130; c) J.
Elguero, P. Goya, I. Rozas, J. Mol. Struct. 1989, 184, 115–129.
[11] a) J. M. Langenhan, J. D. Fisk, Org. Lett. 2001, 3, 2559–2562; b) C.
Gennari, M. Gude, D. Potenza, U. Piarulli, Chem. Eur. J. 1998, 4,
1924–1931.
[12] For recent reviews of synthetic double helices, see: a) S. E. Howson,
P. Scott, Dalton Trans. 2011, 40, 10268–10277; b) Y. Furusho, E. Ya-
shima, Macromol. Rapid Commun. 2011, 32, 136–146; c) D. Haldar,
C. Schmuck, Chem. Soc. Rev. 2009, 38, 363–371; d) Y. Furusho, E.
Yashima, J. Polym. Sci. Part A 2009, 47, 5195–5207; e) M. Albrecht,
Chem. Rev. 2001, 101, 3457–3497.
[1] For examples, see: a) K. Shuvaev, A. Decken, F. Grein, T. S. M.
Abedin, L. K. Thompson, J. Passmore, Dalton Trans. 2008, 4029–
4037; b) D. Kiriya, H.-C. Chang, S. Kitagawa, J. Am. Chem. Soc.
2008, 130, 5515–5522; c) F. Volatron, L. Catala, E. Riviꢂre, A.
Gloter, O. Stꢃphan, T. Mallah, Inorg. Chem. 2008, 47, 6584–6586;
d) J. Larionova, L. Salmon, Y. Guari, A. Tokarev, K. Molvinger, G.
Molnꢄr, A. Bousseksou, Angew. Chem. 2008, 120, 8360–8364;
Angew. Chem. Int. Ed. 2008, 47, 8236–8240; e) O. Kahn, C. J. Marti-
nez, Science 1998, 279, 44–48.
[2] For examples, see: a) H. Du, R. C. Haddon, I. Krossing, J. Passmore,
J. M. Rawson, M. J. Schriver, Chem. Commun. 2002, 1836–1837;
b) K. Terao, N. Kikuchi, T. Sato, A. Teramoto, M. Fujiki, T. Dobashi,
Langmuir 2006, 22, 7975–7980. Review of antifreeze proteins: c) Y.
Yeh, R. E. Feeney, Chem. Rev. 1996, 96, 601–618.
[13] For recent examples of double-helix formation by synthetic oligo-
meric compounds, see: a) H. Yamada, Z.-Q. Wu, Y. Furusho, E. Ya-
shima, J. Am. Chem. Soc. 2012, 134, 9506–9520; b) H.-B. Wang,
B. P. Mudraboyina, J. A. Wisner, Chem. Eur. J. 2012, 18, 1322–1327;
c) Y. Ferrand, Q. Gan, B. Kauffmann, H. Jiang, I. Huc, Angew.
Chem. 2011, 123, 7714–7717; Angew. Chem. Int. Ed. 2011, 50, 7572–
7575; d) Y. Haketa, H. Maeda, Chem. Eur. J. 2011, 17, 1485–1492;
e) D. Hu, Z. Yang, G. Zhang, M. Liu, J. Xiang, T. Liang, J. Ma, G.
Yang, Crystengcomm 2011, 13, 6021–6023; f) Q. Gan, F. Li, G. Li, B.
Kauffmann, J. Xiang, H. Jiang, Chem. Commun. 2010, 46, 297–299.
[14] For reviews of our helicene-based double helixes, see: a) R. Ame-
miya, M. Yamaguchi, Org. Biomol. Chem. 2008, 6, 26–35; b) R.
Amemiya, M. Yamaguchi, Chem. Rec. 2008, 8, 116–127.
[15] 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) R. Amemiya, N. Saito,
M. Yamaguchi, J. Org. Chem. 2008, 73, 7137–7144; d) N. Saito, R.
Terakawa, M. Shigeno, R. Amemiya, M. Yamaguchi, J. Org. Chem.
2011, 76, 4841–4858; e) W. Ichinose, M. Shigeno, M. Yamaguchi,
Chem. Eur. J. 2012, 18, 12644–12654.
[16] For studies of the absolute hardness of arenes, see: R. G. Pearson, J.
Org. Chem. 1989, 54, 1423–1430.
[17] R. Amemiya, W. Ichinose, M. Yamaguchi, Bull. Chem. Soc. Jpn.
2010, 83, 809–815.
[18] W. Ichinose, M. Miyagawa, J. Ito, M. Shigeno, R. Amemiya, M. Ya-
maguchi, Tetrahedron 2011, 67, 5477–5486.
[19] a) C. Gennari, M. Gude, D. Potenza, U. Piarulli, Chem. Eur. J. 1998,
4, 1924–1931; b) J. M. Langenhan, J. D. Fisk, S. H. Gellman, Org.
Lett. 2001, 3, 2559–2562.
[20] According to the comments of a reviewer, enthalpy DH, and entro-
py DS values were obtained by using the equilibrium curve: DH=
(267Æ4) kJmolÀ1 and DS=(0.75Æ0.01) kJmolÀ1 KÀ1 (Table S1 and
Figure S16 in the Supporting Information), assuming the heating
curve of the low-rate experiments to be under equilibrium. The
values are similar to those reported in the studies of bis(ethynylheli-
cene)s (see Ref. [7]), and the unusual aggregation property of (M)-4
could not directly be related to the data.
[21] a) E. A. Archer, M. J. Krische, J. Am. Chem. Soc. 2002, 124, 5074–
5083; b) H. Jiang, V. Maurizot, I. Huc, Tetrahedron 2004, 60, 10029–
10038; c) G. E. Job, R. J. Kennedy, B. Heitmann, J. S. Miller, S. M.
Walker, D. S. Kemp, J. Am. Chem. Soc. 2006, 128, 8227; d) H. Goto,
Y. Furusho, K. Miwa, E. Yashimma, J. Am. Chem. Soc. 2009, 131,
4710–4719; e) B. Baptiste, J. Zhu, D. Haldar, B. Kauffmann, J. M.
Lꢃger, I. Huc, Chem. Asian J. 2010, 5, 1364–1375.
[22] Memory effect under a static condition was reported. For examples,
see: a) Y. Hase, K. Nagai, H. Iida, K. Maeda, N. Ochi, K. Sawabe,
K. Sakajiri, K. Okoshi, E. Yashima, J. Am. Chem. Soc. 2009, 131,
10719–10732; b) N. Ousaka, Y. Inai, R. Kuroda, J. Am. Chem. Soc.
2008, 130, 12266–12267; c) K. Toyofuku, M. A. Alam, A. Tsuda, N.
Fujita, S. Sakamoto, K. Yamaguchi, T. Aida, Angew. Chem. 2007,
119, 6596–6600; Angew. Chem. Int. Ed. 2007, 46, 6476–6480.
[3] For examples, see: a) X. Liu, S. Luo, J. Ye, C. Wu, Macromolecules
2012, 45, 4830–4838; b) H. Horiuchi, T. Ishihara, S. Tsunoda, D.
Morishita, T. Takei, T. Okutsu, T. Seko, H. Sugino, Y. Kawakami, H.
Watanabe, M. Abe, S. Takigami, H. Hiratsuka, J. Phys. Chem. B
2006, 110, 9072–9078.
[4] For examples, see: a) M. R. Berber, H. Mori, I. H. Hafez, K. Mina-
gawa, M. Tanaka, T. Niidome, Y. Katayama, A. Maruyama, T.
Hirano, Y. Maeda, T. Mori, J. Phys. Chem. B 2010, 114, 7784–7790;
b) Y. Tang, Y. Ding, G. Zhang, J. Phys. Chem. B 2008, 112, 8447–
8451; c) A. K. Dikshit, T. Jana, S. Malik, A. K. Nandi, Polym. Chem.
2003, 52, 925–931; d) R. Tadmor, R. L. Khalfin, Y. Cohen, Lang-
muir 2002, 18, 7146–7150. Thermal hysteresis by spin crossover:
e) P. N. Martinho, Y. Ortin, B. Gildea, C. Gandolfi, G. McKerr, B.
OꢅHagan, M. Albrecht, G. G. Morgan, Dalton Trans. 2012, 41, 7461–
7463; f) J. R. Galꢄn-Mascarꢆs, E. Coronado, A. Forment-Aliaga, M.
Monrabal-Capilla, E. Pinilla-Cienfuegos, M. Ceolin, Inorg. Chem.
2010, 49, 5706–5714; g) K. Kuroiwa, H. Kikuchi, N. Kimizuka,
Chem. Commun. 2010, 46, 1229–1231.
[5] For examples, see: a) M. P. B. Nagata, K. Yamashita, M. Miyazaki,
H. Nakamura, H. Maeda, Anal. Biochem. 2009, 390, 38–45; b) D. P.
Arya, T. C. Bruice, Proc. Natl. Acad. Sci. USA 1999, 96, 4384–4389;
c) M. Rougꢃe, B. Faucon, J. L. Mergny, F. Barcelo, C. Giovannange-
li, T. Garestier, C. Hꢃlꢂne, Biochemistry 1992, 31, 9269–9278; d) D.
Barszcz, D. Shfflgar, Eur. J. Biochem. 1968, 5, 91–100.
[6] C. Wu, X. Wang, Phys. Rev. Lett. 1998, 80, 4092–4094.
[7] H. Sugiura, R. Amemiya, M. Yamaguchi, Chem. Asian J. 2008, 3,
244–260.
[8] For reviews and highlights of synthetic oligomeric compounds that
can form helical structures, see a) E. Yashima, K. Maeda, H. Iida, Y.
Furusho, K. Nagai, Chem. Rev. 2009, 109, 6102–6211; b) D. Haldar,
C. Schmuck, Chem. Soc. Rev. 2009, 38, 363–371; c) H.-J. Kim, Y.-B.
Lim, M. Lee, J. Polym. Sci. Part A 2008, 46, 1925–1935; d) Foldam-
ers: Structure, Properties, and Applications (Eds.: S. Hecht, I. Huc),
Wiley-VCH, Weinheim, 2007; e) M. Albrecht, Angew. Chem. 2005,
117, 6606–6609; Angew. Chem. Int. Ed. 2005, 44, 6448–6451; f) I.
Huc, Eur. J. Org. Chem. 2004, 17–29; g) A. R. Sanford, B. Gong,
Curr. Org. Chem. 2003, 7, 1649–1659; h) D. J. Hill, M. J. Mio, R. B.
Prince, T. S. Hughes, J. S. Moore, Chem. Rev. 2001, 101, 3893–4011.
[9] a) R. M. J. Liskamp, J. A. W. Kruijtzer, Mol. Div. 2004, 8, 79–87;
b) R. De Jong, D. T. S. Rijkers, R. M. J. Liskamp, Helv. Chim. Acta
2002, 85, 4230–4243.
Received: December 12, 2012
Revised: April 22, 2013
Published online: && &&, 0000
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

M. Yamaguchi et al.
Know when to unfold ꢀem: Sulfonami-
dohelicene oligomers up to the non-
amer level were synthesized. The tet-
ramer formed a helix dimer in a fluori-
nated aromatic solvent, which
unfolded with heating. This structural
change exhibited molecular thermal
hysteresis (see figure).
Helical Structures
M. Shigeno, Y. Kushida,
M. Yamaguchi* ............... &&&& — &&&&
Molecular Thermal Hysteresis in
Helix-Dimer Formation of Sulfonami-
dohelicene Oligomers in Solution
&
10
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!