Mesomorphic [2]rotaxanes: Sheltering ionic cores with interlocking components

Article abstract of DOI:10.1021/ja309558p

Two types of liquid crystalline [2]rotaxanes based on a conventional tetracatenar motif (a rod-shaped molecule with two side chains at each end) have been prepared. Dicationic compounds with ester stoppers and tetracationic materials with pyridinium stoppers are compared to each other and their dumbbell shaped analogs. Since the ionic core contributes about 70% to the overall length and molecular weight of the molecules, sheltering the ionic cores with an interlocked neutral macrocycle has considerable effect on the mesomorphism and thermal stability of the materials. The influence of the sheltering macrocycle, the numbers of charges on the core and the size and nature of the side chains (aliphatic vs siloxane) were probed. [2]Rotaxanes with linear side chains and minimum ratios of chain-to-core volumes of about 0.35 and 0.30 for tetra- and dicationic compounds, respectively, display smectic liquid crystal phases. Larger ratios increase the temperature range of the smectic A phases beyond the decomposition temperatures; a disadvantage for processing because no stable isotropic liquid phase is available. The change from tetra- to dicationic [2]rotaxanes increased not only the fluidity of their smectic A phases but also their thermal and chemical stability. Branched side chains (2-hexyldecyl) disfavor the formation of lamellar mesophases and, instead, induce higher ordered soft crystal phases. No liquid crystal phases but soft crystal phases are observed for the analogous di- and tetracationic compounds without an ion sheltering interlocked macrocycle (dumbbells).

Full text of DOI:10.1021/ja309558p

Article
pubs.acs.org/JACS
Mesomorphic [2]Rotaxanes: Sheltering Ionic Cores with Interlocking
Components
Natalie D. Suhan, Stephen J. Loeb,* and S. Holger Eichhorn*
Department of Chemistry and Biochemistry, University of Windsor, Windsor, Ontario, N9B 3P4, Canada
S
* Supporting Information
ABSTRACT: Two types of liquid crystalline [2]rotaxanes
based on a conventional tetracatenar motif (a rod-shaped
molecule with two side chains at each end) have been
prepared. Dicationic compounds with ester stoppers and
tetracationic materials with pyridinium stoppers are compared
to each other and their dumbbell shaped analogs. Since the
ionic core contributes about 70% to the overall length and
molecular weight of the molecules, sheltering the ionic cores
with an interlocked neutral macrocycle has considerable effect
on the mesomorphism and thermal stability of the materials. The influence of the sheltering macrocycle, the numbers of charges
on the core and the size and nature of the side chains (aliphatic vs siloxane) were probed. [2]Rotaxanes with linear side chains
and minimum ratios of chain-to-core volumes of about 0.35 and 0.30 for tetra- and dicationic compounds, respectively, display
smectic liquid crystal phases. Larger ratios increase the temperature range of the smectic A phases beyond the decomposition
temperatures; a disadvantage for processing because no stable isotropic liquid phase is available. The change from tetra- to
dicationic [2]rotaxanes increased not only the fluidity of their smectic A phases but also their thermal and chemical stability.
Branched side chains (2-hexyldecyl) disfavor the formation of lamellar mesophases and, instead, induce higher ordered soft
crystal phases. No liquid crystal phases but soft crystal phases are observed for the analogous di- and tetracationic compounds
without an ion sheltering interlocked macrocycle (dumbbells).
academic and commercial interest,8 and the idea of molecular
switches based on liquid crystalline rotaxanes represents a new
approach to these types of materials.
Despite the impressive syntheses of these first examples of
INTRODUCTION
■
Over the past 20 years, the concept of mesomorphism has been
extended well beyond the classical structural motifs of liquid
crystals (LCs). Judicious application of shape anisotropy,
directed molecular interactions, self-assembly, and microphase
liquid crystalline materials containing MIMs, no predictive
1
knowledge has been generated that would allow for the rational
design of liquid crystalline interlocked molecules. In fact, very
little is known about the mesomorphism of ionic liquid crystals
that have multiple charges on a rigid rod core, and completely
unexplored is the effect of an interlocked macrocyclic ring on
mesomorphism in general. An investigation of these structure−
property relationships for ionic liquid crystalline [2]rotaxanes is
also of fundamental interest because it adds to the body of work
that has been produced on thermotropic ionic liquid crystals.9
All reported thermotropic ionic liquid crystals with multiple
charges located on a rigid-rod core are based on doubly charged
viologen salts,10 except for single reports on dimeric 2,4,6-
triarylpyrylium11 and bis(benzimidazolium)12 derivatives.
Presented herein is the synthesis and characterization of two
classes of liquid crystalline ionic [2]rotaxanes and their
dumbbell analogs (stoppered axles without a macrocycle
component) based on a conventional tetracatenar motif. In
this design, the charged core contributes about 70% to the
overall length and molecular weight of the entire molecule.
segregation have converted typical nonmesogens, such as C60
and other molecules with low or unfavorable aspect ratios,
including ionic molecules, into compounds that display
mesophases over a wide range of temperatures.2
Mechanically interlocked molecules (MIMs), such as [2]-
rotaxanes, are interesting candidates for inclusion into
mesogens due to the complex nature of their component
building blocks, their size, and often ionic character.3 It was
only recently that the first liquid crystalline [2]rotaxane and
[2]catenane were reported by Kato in collaborations with
Stoddart4 and Sauvage,5 respectively, and the first example of a
mesophase forming polyrotaxane reported by Kidowaki et al.6
The feasibility of molecular switching in a mesophase was also
clearly demonstrated by Kato and Stoddart who showed that a
bistable [2]rotaxane could be electrochemically switched
between two different molecular states in a liquid crystal
phase.7 These initial studies were undoubtedly motivated by the
intriguing prospect of preparing self-organizing molecular
switches since it may eventually be possible to amplify the
molecular switching process in a mesophase by cooperative
motion, thereby producing a directed (anisotropic) macro-
scopic response. Stimuli responsive materials are of both
Received: October 1, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of [2]Rotaxanes R14+−R16b4+
Consequently, structural changes to the core considerably affect
the mesomorphism of the material. Of the investigated
structural variations: (i) addition of the neutral macrocycle
(comparison of dumbbells and [2]rotaxanes); (ii) two versus
four charges (positive) on the core, (iii) size, and (iv) nature of
the side chains; the addition of the neutral macrocycle is most
effective in the introduction of mesomorphism. This result is
attributed to a sheltering of the charges on the core by the
neutral macrocycle that significantly reduces intermolecular
ionic interactions. Thus, [2]rotaxanes based on a small
molecule design that form stable liquid crystal and isotropic
liquid phases are obtained for the first time. Although not
investigated here, this new design provides a blueprint for the
preparation of analogous switchable [2]rotaxanes3o,s that might
be expected to display a macroscopic response if MIM
switching were to be achieved in a liquid crystalline phase.
Containing DB24C8 as the Macrocycle and Their
Corresponding Dumbbells D14+−D16b4+ without the
a
Macrocycle
RESULTS AND DISCUSSION
■
Design. Thermotropic ionic liquid crystals form mesophases
when heated above their melting temperature and, in contrast
to lyotropic ionic liquid crystals, do not require the presence of
a solvent. The majority of known thermotropic ionic liquid
crystals consist of singly charged surfactant-like molecules, and
their self-organization is primarily controlled by amphiphilic
interactions that often promote both thermotropic and
lyotropic mesomorphism. More recently, structural features of
amphiphilic ionic liquid crystals were combined with mesogenic
rod-shaped rigid cores of conventional calamitic liquid crystals,
but in many of these approaches, the charged units were
attached via flexible spacers and were not part of the rigid core.9
Only a few examples of thermotropic ionic liquid crystals
consisting of a charged rigid-rod core centered between side
chains have been reported.11 Examples are the thermotropic
ionic liquid crystals based on rigid-rod metal complexes
developed by Bruce13 and others,14 doubly charged viologens,10
and a range of imidazolium containing rigid-rod ionic liquid
crystals recently studied by Swager15 and others.12 The
mesomorphism of all these compounds was altered by a
change of anions and attached side chains to optimize
amphiphilic interactions and excluded volume effects.
The alteration of the size and type of side chains is employed
here for the creation of LC [2]rotaxanes. Also varied is the core
charge between 4+ in the ionic core based on two viologen-like
4,4′-bipyridinium groups linked by an ethylene chain (Scheme
1) and 2+ in the analogous 4-phenyl-pyridinium derivative
(Scheme 2). All [2]rotaxanes have two side chains attached to
each aromatic end group (tetracatenar motive), as this was
predicted to provide the best match between the packing
volumes of the cores and the side chains for the formation of
lamellar mesophases. A branched 2-hexyldecyl and a siloxane
chain were tested in addition to linear alkyl chains of various
lengths, as they have been shown to promote tilted lamellar
mesophases such as SmC.16 Most importantly, all these
compounds can be prepared with and without a dibenzo[24]-
crown-8 ether (DB24C8) macrocycle that wraps around the
majority of the pyridinium core charges allowing direct
comparison of the “naked” charged species (dumbbells) to
the analogous [2]rotaxanes as depicted in Figure 1. The use of
an uncharged macrocycle and a charged core, in contrast to the
charged macrocycle employed by Stoddard and Kato,4 is also
expected to promote the formation of liquid crystal phases by
sheltering the ionic cores and reducing strong ionic interactions
between the di- or tetracationic [2]rotaxane molecules.
a
(i) CH3NO2/CHCl3 60 °C, 3 days; (ii) 8 equiv DB24C8, CH3NO2/
CHCl3 50 °C, μW 200 W, 5h followed by NaOTf(aq) and a further 50
°C, μW 200 W, 5−15 h. Reported yields are isolated after multiple
runs of column chromatography (see SI for details).
Scheme 2. Synthesis of [2]Rotaxanes R12+, R122+, and RSi2+
containing DB24C8 as the Macrocycle and the Dumbbell
a
D122+
a
(i) 7:3 CHCl3/CH3CN, Bu3P cat., RT, 3 h; 5 equiv DB24C8 for
[2]rotaxanes, no crown for the dumbbell. Reported yields are isolated
after multiple runs of column chromatography (see SI for details).
B
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Cartoon depiction of dumbbell (top) and [2]rotaxane
(bottom) featuring a tetracatenar structural motif designed to induce
mesomorphism.
Synthesis. The 4+ charged [2]rotaxanes R14+−R16b4+ and
their corresponding dumbbells D14+−D16b4+ were prepared by
alkylation of the terminal pyridine groups of axle 1,2-bis(4,4′-
bipyridinium)ethane, 1[OTf]2 with the appropriate 3,5-diester-
substituted benzylbromide, in the presence (rotaxane) or
absence (dumbbell) of the macrocycle DB24C8, as outlined
in Scheme 1.17
The preparation of the dumbbells is straightforward, while
preparation and purification of the [2]rotaxanes is more
challenging. Traditionally, [2]rotaxane formation with this
templating motif involves prolonged reaction at room temper-
ature until all terminal pyridine groups are alkylated,17 ranging
from days to weeks with reaction times increased for this study
due to extreme solubility differences between the ionic axles
and aliphatic stoppers. Efforts to decrease reaction times by
heating were thwarted by a marked decrease in pseudorotaxane
formation and correspondingly poorer yields. To decrease
reaction times and increase product yields, microwave
irradiation was employed.17a Temperatures as high as 50 °C
yielded nearly complete [2]rotaxane formation (>80%, 1H
NMR) in 5−15 h (SI).
Figure 2. X-ray crystal structures of R14+ (top) and MR12+ (bottom)
showing the relative positioning of the axle (blue), macrocycle (red),
stoppering groups (green), and anions (gold) in the solid state.
longer chains are similar to those observed for the X-ray
structures of these model compounds.
Both, R14+ and MR12+ show a typical S-shaped conformation
for the macrocycle DB24C8, and the stopper groups are
oriented in a zigzag shape, i.e., up and down relative to the
rotaxane plane with the anions positioned close to the ends of
the core.19 One of the main differences in overall structure is
that for R14+ the benzylic stopper groups are almost orthogonal
to the core axis, while for MR12+ the benzoic ester stopper
groups are twisted outward at an angle of about 120° to the
core.17a
As outlined in Scheme 2, dumbbell D122+ and [2]rotaxanes
(R12+, R122+, and RSi2+) were synthesized by ester formation
using conditions reported by Takata et al.18 This system offered
a significant synthetic improvement as [2]rotaxane formation
was complete in ca. 3 h, with no formation of the usual
dumbbell side products. Purification was accomplished by
multiple runs of column chromatography employing fully end-
capped RP-C18 silica gel and methanol as the eluent. For both
4+ and 2+ charged systems, the isolated yield depended upon
the effectiveness of the chromatography and the number of
runs required to achieve sufficient purity for LC experiments.
Single-Crystal X-ray Structures of Model Compounds.
[2]Rotaxanes R14+ and R12+, containing R = Me groups, were
prepared as compounds that would be expected to be
crystalline versions of the LC [2]rotaxanes in these series; as
designed, they do not melt before they start to decompose at
temperatures >200 and 260 °C, respectively. Unfortunately,
single crystals of R12+ were too small for X-ray analysis,
however suitably sized crystals of the closely related derivative
MR12+ (Me groups in place of COOMe groups) could be
grown. Thus, single-crystal X-ray structures were obtained for
R1[CF3SO3]4 and the derivative MR1[BF4]2.
Importantly these structures, as depicted in Figure 2, provide
an idea of the overall arrangement of the major structural
features in these compounds, for example, the position of the
anions relative to the cationic axle/wheel core and the
conformation of the stopper groups with respect to the
rotaxane core. Compounds with longer chains either do not
form single crystals of sufficient quality or are not crystalline. It
is assumed for the purpose of estimating molecular dimension
(see Table 1) that the ionic cores of the compounds with
The basic shape of the molecular components and the
positioning of the cations and anions might be expected to
persist throughout a series of related compounds in which only
changes to peripheral R-groups have been made. However, the
lowest energy conformation of the molecule in a mesophase
would likely differ greatly from that observed in the crystalline
phase and may be altered significantly by the exchange of
methyl groups for longer and more flexible alkyl chains.
Nevertheless, the differences in observed mesomorphism for
R124+and R122+ (vide infra) are consistent with the basic
structural features observed for the [2]rotaxanes in the crystal
structures of model compounds R14+and MR12+, justifying the
use of their crystallographic parameters to estimate the
molecular dimensions summarized in Table 1.
Estimation of Molecular Dimensions. Space filling
calculations were performed for all dumbells and [2]rotaxanes
to compare molecular dimensions with layer spacings and other
packing distances obtained by PXRD (see Table 1). Core
volumes were calculated as cylindrical excluded volumes and
imply a free rotation of the molecules over their long axis, as is
the case in smectic liquid crystal phases. The dimensions of the
2+ and 4+ charged cores are based on their conformations in
the crystal structures of model compounds R14+and MR12+ and
include all anions, the aromatic parts of the stopper groups, and
for [2]rotaxanes, the crown ether macrocycle. Core lengths of
2+ and 4+ charged cations differ (29.8 and 27.4 Å,
respectively), while their average diameters (average of
C
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Table 1. Estimated Molecular Dimensions of Dumbbells and [2]Rotaxanes, Inclusive of CF3SO3 Anions
Article
−
a
b
c
core dimensions
chain dimensions
Vchains (Å3)
totals
Lmax (Å)
experimental values
d-spacing (Å), (T (°C), phase)
32.8 (30, SC)
d
e
compd
Lcore (Å)
Wcore (Å)
Vcore (Å3)
Lchains (Å)
Ltot (Å)
AR
D64+
D94+
27.4
27.4
27.4
27.4
27.4
27.4
27.4
27.4
27.4
27.4
29.8
29.8
29.8
8.1
8.1
1350
1350
1350
1350
1350
2400
2400
2400
2400
2400
1536
2630
2630
440
640
8.5
12.4
16.5
21.7
22.3
5.0
42.6
49.1
55.6
64.2
51.7
42.6
49.1
55.6
64.2
51.7
54.9
54.9
53.8
35.9
39.8
43.9
49.1
49.6
32.4
34.7
37.0
40.1
40.4
46.3
39.4
43.4
4.4
4.9
5.4
6.1
6.1
3.1
3.3
3.5
3.8
3.8
5.7
3.7
4.1
45.0 (30, SC1), 38.8 (100, SC2)
45.0 (25, SC1), 45.5 (90, SC2)
48.9 (25, SC7L), 50.3 (130, SC5L)
37.3 (25, SC)
D124+
D164+
D16b4+
R64+
8.1
850
8.1
1120
1150
440
8.1
10.6
10.6
10.6
10.6
10.6
8.1
30.5 (25, SC)
R94+
640
7.3
35.9 (30, SC)
R124+
R164+
R16b4+
D122+
R122+
RSi2+
850
9.6
39.6 (100, SmA), 38.5 (150, SmA)
44.6 (25, SmX1), 42.9 (100, SmA)
33.7 (25, SC1), 32.8 (130, SC2)
39.3 (30, SC1), 40.8 (140, SC2)
39.8 (25, SmA)
1120
1150
850
12.7
13.0
16.5
9.6
10.6
10.6
850
1200
13.6
42.2 (25, SmA)
a
b
From single-crystal X-ray structures of R14+ and MR12+. Vcore = π(Wcore/2)2Lcore. Established values for the molten state from ref 20. Lchain = Vchain
/
c
π(Wchain/2)2 and assuming Wcore = Wchain. From AM1 calculations on models with extended chains where X-ray structures of R14+ and MR12+ were
used as the basis of the model. Ltot = Lcore + Lchain. Smallest angle reflection measured by PXRD at given temperature. Aspect ratio Ltot/Wcore.
d
e
maximum and minimum core width) are identical but differ
between dumbbells (8.1 Å) and [2]rotaxanes (10.6 Å). The
overall length of the cylindrical space the molecule occupies is
calculated based on the assumption that both core and side
chains fill a cylindrical (excluded) volume of the same diameter.
A cylindrical volume occupied by side chains is then calculated
based on the established packing volumes of hydrocarbon and
siloxane groups in their isotropic liquid phases.20 Consequently,
the calculated packing dimensions are estimations of the
minimum layer spacing and in-plane intermolecular distances
for a uniaxial lamellar liquid crystalline phase (e.g., SmA) and
are independent of the degree of interdigitation. Estimated
molecular lengths for all trans extended side chains (Lmax) are
also provided in Table 1 for comparison.
the least stable of the five [2]rotaxanes because it already
reaches the 0.3% marker at 165 °C, but this is likely due to
decomposition of the siloxane groups (vide infra).
All compounds decompose in several distinct weight loss
steps (Figure 3). The % weight loss of the first step correlates
Thermal stability. Thermal gravimetric analysis (TGA) of
all [2]rotaxanes and dumbbells was performed to probe their
thermal stability and the presence of any residual volatile
compounds (Table S2). Samples dried in vacuum below their
melting temperatures show a small weight loss of up to 0.5%
between 50 and 100 °C, which is attributed to the loss of
acetonitrile that is also present in the single crystals of R14+ and
other related [2]rotaxanes. No further measurable weight loss
occurs until the onset of decomposition at temperatures >150
°C. Decomposition temperatures are estimated not only based
on weight loss in TGA but also degradation observed by
differential scanning calorimetry (DSC) and NMR measure-
ments of heated samples. Thermal stability values measured by
TGA at a heating rate of 2 °C/min under He(g) best agree with
the values obtained by DSC and NMR when the limit of
thermal stability is defined as the temperature at which 0.3% of
weight loss has occurred, rather than the extrapolated onset, a
common procedure for polymers.21
Figure 3. TGA curves for the 4+ charged [2]rotaxane R124+ and 2+
charged [2]rotaxanes R122+ and RSi2+ obtained at a heating rate of 2
°C/min under He(g). Double-headed arrows show the calculated
weight loss for the removal of both stopper groups. The onset of
weight loss is defined as the temperature at which 0.3% weight loss has
occurred and is marked in the enlarged view at the top of the graph.
All samples were preheated in vacuum to remove remaining solvent.
All dumbbells reach a weight loss of 0.3% at about 170 °C,
whereas their analogous 4+ and 2+ charged [2]rotaxanes,
except for R4Si2+, do not reach the 0.3% marker <190 and 250
°C, respectively (Table S2). Clearly, complexation with a
macrocycle increases the thermal stability of the axle and is
more effective for the 2+ charged compounds. The stabilization
can be reasoned with a better delocalization of the positive
charges on the axle due to charge-transfer interactions between
the electron-deficient pyridinium groups and the electron-rich
catechol groups of the crown. Surprisingly, [2]rotaxane RSi2+ is
with mass contributions from the benzylic and ester stopper
groups in all cases but R4Si2+ (Table S2). TGA of R14+ coupled
with mass spectrometric analysis of the evolved gas supports
thermal cleavage of the C−N+ bond as a high concentration of
fragment ion peaks characteristic for the benzylic stopper was
detected during the first weight loss event. Consequently, the
thermal stability of the 4+ charged compounds is limited by the
C−N+ bond between the pyridinium and benzylic groups and
the stability of the 2+ charged compounds by the benzoate
esters.
D
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Similar decomposition temperatures have been reported for
other pyridinium compounds and benzyl benzoates.22 RSi2+
shows an unexpectedly low thermal decomposition temperature
of only 175 °C, 95 °C below the thermal decomposition of
R122+. This difference is attributed to an initial decomposition
of the siloxane chains since this represents the only structural
difference between RSi2+ and R122+. The combined % weight
losses of the first and second steps account for two ester
stopper groups suggesting that ester cleavage of the benzyl
benzoate constitutes the second weight loss − the onset
temperature of about 260 °C is similar to R122+.
The diffraction pattern of the SC phase of D164+ is shown as
an example in Figure 4. The softness of these crystal phases was
confirmed by POM studies that showed the crystallites deform
like viscous fluids under pressure (Figure 5). Crystallites were
obtained only by precipitation from solution and did not form
in the bulk phase upon cooling from higher temperature
phases.
Mesomorphism: General Observations. All [2]-
rotaxanes and dumbbells, except the crystalline compounds
R14+ and D14+, are soft birefringent materials at room
temperature. The temperature-dependent (thermotropic)
mesomorphism of these materials was studied by polarized
optical microscopy (POM), differential scanning calorimetry
(DSC), and variable temperature X-ray diffraction (vt-PXRD)
(see SI for details). Many of the phase transitions occur slowly
because the mesophases of these high molecular weight ionic
compounds have high viscosity. To circumvent measurements
on kinetically trapped rather than thermodynamically stable
phases, POM images and XRD patterns were monitored at
specific temperatures for up to several hours to identify
nonequilibrium states by changes that occurred over time.
Many of the [2]rotaxanes and dumbbells displayed rich
poly(meso)morphism, but only compounds R124+, R164+,
R122+, and RSi2+ formed mesophases that are liquid crystal
phases. All other phases were categorized, based on their
diffraction patterns, as soft-crystal phases (SC) of unknown
structure or lamellar structure (SCL) (Figure 4). Phases were
Figure 5. POM photomicrographs (crossed polarizers, 25 °C) of D64+.
(a) As crystallites precipitated from solution. (b) The same area after
pressure was applied to the cover slide to generate a birefringent
viscous fluid-like appearance.
Many of the SC phases show a single intense small angle
reflection at a d-spacing close to the calculated layer spacing for
a SmA-type packing of the molecules (Table 1), but a simple
lamellar structure can be excluded because of additional small
angle reflections. Lamellar structures (SCL), however, are
proposed for SC phases of D164+ (Figure 4) and R124+ because
their diffraction patterns contain a second-order reflection of
the layer spacing as the only other reflection below 5° 2-θ.
All our attempts of obtaining good quality 2D diffraction
patterns of well-aligned SC phases failed, and no attempt was
made to index the reflections based solely on 1D diffraction
patterns. However, the diffraction patterns did not agree with
patterns of crystal smectic B, E, and T phases that have been
regularly observed in ionic smectic liquid crystals. This is not
surprising because the formation of crystal smectic B, E, and T
phases is usually driven by ordering of ionic head groups within
bilayers.23 Bilayer formation is unlikely to occur in the
dumbbells and [2]rotaxanes presented herein, since the charged
groups are associated with the rod-shaped core in the center of
the molecules. A bilayer packing would only be possible if the
cores form dimers and the side chains are oriented to both sides
orthogonally to the cores, similar to what has been proposed
for disc-shaped mesogens with polyionic cores.24 However, the
expected layer spacings of bilayers are much smaller than the
measured spacings because the four side chains on each side
would fully interdigitate with the side chains of the next layer to
fill the available space.
Figure 4. Example XRD patterns of the (lamellar) soft crystal phase
(SC) and lamellar soft crystal phases (SCL) of D164+ and the SmX and
SmA liquid crystal phases of R164+.
categorized as SC phases when their diffraction patterns
contained several additional reflections to what is observed
for liquid crystal phases and could not be indexed based on
plane groups. However, the degree of crystalline order varies
widely between different SC phases, and differences are
particularly obvious in the angle range of 15−22° 2-θ (d-
spacings of 6 − 4 Å) that is indicative of the degree of
crystallinity of the aliphatic side chains. A crystalline state of the
aliphatic side chains generates sharp reflections and an
amorphous state a broad diffuse reflection (halo).
A strong reflection at spacings between 14.5 and 17 Å
observed in the diffraction patterns of all phases of 4+ charged
dumbbells and [2]rotaxanes (Figure 4) must originate from the
lateral packing of the molecules, as it is independent of the
molecular length and was the only consistent indication of
partial in-plane order. The d-spacing is about twice the average
diameter of the dumbbells and [2]rotaxanes including all anions
of about 8−10 Å and likely originates from short-range, ordered
packing of the anions. A comparison with the single crystal
structure of R14+ confirmed that the packing distances of the
E
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
anions along the c-axis (10.7 and 9.8 Å) are of this order
(Figure 2).
Mesomorphism: 4+ Charged Dumbbells. Character-
ization of the mesomorphism of the 4+ charged dumbbells was
mainly based on DSC and XRD studies because defect textures
observed by POM were unspecific, except for focal-conic-type
textures displayed by D124+. Dumbbells D64+−D16b4+ all form
soft-crystal phases that decrease in degree of crystallinty with
increasing length of side chains, but no liquid crystalline phases
are formed (Figure 6).
Figure 7. Phase transition temperatures in °C and enthalpies in kJ/
mol of 4+ charged [2]rotaxanes based on POM, DSC, and XRD
analysis (SC = soft crystal phase; SCL = lamellar SC phase; SmX =
unknown smetic liquid crystal phases; SmA = smectic A liquid crystal
phase; Iso = isotropic liquid phase). (a) Broad transition; (b)
determined by POM (not observed by DSC); (c) observed only in the
first heating run of a compound precipitated from solution;(d)
observed by DSC only in the first heating run but observed by POM in
subsequent heating runs; (e) this transition consists of one large and
two small overlapping transitions; (f) sample remains isotropic on
cooling and forms the SC phase upon heating at 124 °C.
Figure 6. Phase transition temperatures in °C and enthalpies in kJ/
mol of 4+ charged dumbbells based on POM, DSC, and XRD analysis
(SC = soft crystal phase; SCL = lamellar SC phase; numbers (e.g., SC1,
SC2) designate different phases of the same category for each
compound). (a) Transition temperatures (enthalpies): (1) 43.7(−1.5),
(2) 69.8(−5.8), (3) 113.1(−21.3); and (b) two overlapping
transitions.
than the spacing for an expanded molecule (Table 1). XRD
patterns of the initial SCL phase are almost identical with the
patterns of the SmA phase, except for some additional weak
and broad reflections that indicate a higher degree of order than
in the SmA phase. These small changes in the diffraction
patterns are contrasted by a large enthalpy of 40 kJ/mol for the
SCL to SmA transition. A possible explanation for the large
enthalpy is the concomitant loss of solvent molecules at this
phase transition, which also explains its irreversibility.
The assignment of a SmA phase is also supported by the
observation of characteristic fan-shaped textures of focal
domains upon cooling from the isotropic liquid in cells with
planar alignment layer and large pseudo-isotropic areas in cells
treated with cetyl-trimethylammonium bromide for vertical
(homeotropic) alignment (Figure 8). The latter observation
confirms the optically uniaxial character of the phase. 2D XRD
studies of a mechanically aligned fiber of R124+ in its SmA
phase confirm the expected alignment of the smectic layers
parallel to the fiber axis (Figure 8). Unfortunately, the reflection
at 15.5 Å and the halo of the amorphous side chains are not
observed in the 2D pattern because of too low intensity. Both
reflections are expected to be orthogonal to the layer
reflections. All observations but the fan-shaped defect texture
would also agree with a SmB phase, and in fact, short-range in-
plane positional correlations indicated by the reflection at 15.5
Å are typically observed in SmB and not in SmA phases.
Consequently, the presence of SmB phases is excluded solely
based on the absence of mosaic defect textures.26
Compound D164+ is the only dumbbell derivative that
displays SC phases of lamellar structures, while its analogous
dumbbell D16b4+ with branched side chains does not, possibly
because the increased lateral packing volume of the branched
side chains disfavors a lamellar structure. A similar change from
smectic to columnar and cubic mesophases has been reported
for catenar liquid crystals when the lateral packing volume of
the side chains is increased.25
Mesomorphism: 4+ Charged [2]Rotaxanes. The 4+
charged [2]rotaxanes R64+−R164+ are less crystalline than their
parent dumbbells and clear into isotropic liquid phases below
their decomposition temperature of about 190 °C, except for
R164+ (Figure 7). [2]Rotaxanes R64+ and R94+ form highly
viscous SC phases that clear into isotropic liquids at 81 and 118
°C, respectively. Attachment of longer aliphatic chains in
[2]rotaxanes R124+ and R164+ expectedly promotes lamellar
packing order and introduces liquid crystal phases. Compound
R124+ forms a lamellar SC phase when precipitated from
solution but irreversibly converts to a SmA phase at 53 °C
based on POM, DSC, and XRD (Figure 8). The SmA phase
reversibly clears into an isotropic liquid at 137 °C.
Attachment of longer C16 aliphatic chains in R164+ stabilizes
smectic mesomorphism and increases the fluidity of the phases.
R164+ displays two unknown smectic phases SmX1 and SmX2
at low temperature and a SmA phase at higher temperature that
is stable beyond the onset of decomposition at 190 °C.
Diffraction patterns of the SmA phases of R164+ and R124+ are
virtually identical, except for the larger layer spacing of 43 Å for
R164+ that is 3 Å larger than the estimated minimum spacing of
40 Å (Table 1).
The presence of first- and second-order reflections of the
layer structure in the diffraction patterns of R124+ indicates a
high periodicity of the layer spacing (Figure 8). A broad
reflection at 15.5 Å is assigned to the packing of the anions
within layers discussed above, and the diffuse reflection at 4.5 Å
is characteristic for an amorphous state of the aliphatic side
chains. The measured d-spacing of about 39 Å is 2 Å larger than
the estimated minimum layer spacing of 37 Å but 16 Å shorter
F
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Diffraction patterns and textures of the SmX1 and SmX2
phases are virtually identical, but the SmX1 to SmA transition
occurs at 35 °C higher temperature, and its transition enthalpy
is twice as large (see SI). Both, the higher transition enthalpy
and temperature are reasoned with a concomitant loss of
remaining solvent that could not be removed in high vacuum at
room temperature. The loss of solvent is also consistent with
the irreversibility of this transition and the assignment as two
different phases.
[2]Rotaxane R16b4+ with four-branched 2-hexyl-decyl
chains, like its parent dumbbell D16b4+, does not form lamellar
mesophases but two different unknown SC phases. d-Spacings
of their smallest angle reflections are 6−8 Å below the
estimated minimum layer spacing of 40 Å for SmA type
packing, and the diffraction patterns of both phases contain a
large number of weak reflections that indicate a high degree of
order, although the broad and intense halos in all XRD patterns
of the SC1 and SC2 phases suggest that the side chains are
predominantly in an amorphous state (see SI). The SC1 phase
is best distinguished from the SC2 phase by the much higher
intensity of its reflection at 4.2 Å. DSC measurements show
that the SC2 to isotropic liquid transition is reversible on
heating but that the isotropic liquid remains stable upon
cooling down to −40 °C at scan rates of 5−10 °C/min. The
transition from the isotropic phase into the SC2 phase occurs
on the subsequent heating run at 124 °C, similar to a cold
crystallization.
In summary, attachment of linear aliphatic side chains of 12
or more carbon atoms converts the 4+ charged [2]rotaxanes
into smectic liquid crystals, but their smectic phases are still
highly viscous and have short-range in-plane order. An increase
in length of the aliphatic chains from 12 to 16 carbon atoms
stabilizes the SmA phase but generates materials that do not
form isotropic liquid phases because of their limited thermal
stability. Attachment of branched aliphatic chains in R16B4+
lowers transition temperatures but is not compatible with a
lamellar packing of the molecules, most likely because of their
increased lateral packing volume.
Mesomorphism: 2+ Charged Materials. Compounds
D122+, R122+, and RSi2+ based on a 2+ charged design were
prepared to generate more fluid and thermally stable
mesophases. Dodecane chains were chosen for the 2+ charged
compounds D122+and R122+ because [2]rotaxane R124+
showed the most promising phase behavior of the 4+ charged
derivatives. Attachment of siloxane chains in RSi2+ was tested as
an alternative method to branched aliphatic chains for the
lowering of transition temperatures (Figure 9).
Figure 8. Optical micrographs of R124+ in (a) a liquid crystal cell
coated with cetyl-trimethylammonium bromide for vertical (homeo-
tropic) alignment (6 μm gap, 70 °C) and (b) a cell coated with rubbed
polyimide for parallel (homoge-neous) alignment (4 μm gap, 70 °C).
Images were taken at crossed polarized conditions and identical
magnification. (c) Variable-temperature diffraction patterns of R124+
as free-standing bulk material. (d) Overlaid images of optical
micrographs of a drawn fiber of R124+ aligned parallel to one of the
crossed polarizers and rotated ccw by 45°. (e) 2D-XRD pattern of the
drawn fiber at 25 °C; overlaid image of drawn fiber indicates
orientation with regard to the detector and orthogonal to the X-ray
beam.
The reversible SmX2 to SmA transition at about 13 °C is
observed by POM, DSC, and XRD, but the textural changes are
small, and both phases are optically uniaxial (see SI). The main
differences between the diffraction patterns of the SmX2 and
the SmA phases (Figure 3) are a more intense reflection at 15 Å
and an additional broad reflection at 4.2 Å for the SmX2 phase.
Both reflections are likely caused by a short-range order in the
lateral packing of the cores and aliphatic side chains. The
observed diffraction patterns resemble those of SmB phases,
but the few reflections are insufficient for a definitive
assignment, and the large enthalpic change of 32 kJ/mol for
the reversible phase transitions between the SmX2 and SmA
phases is inconsistent with a SmB to SmA transition.
Figure 9. Phase transition temperatures in °C and enthalpies in kJ/
mol of 2+ charged dumbbell and [2]rotaxanes based on POM, DSC,
and XRD analysis (SC = soft crystal phase ; SmA = smectic A liquid
crystal phase; Iso = isotropic liquid phase). (a) Determined by POM.
G
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
To our surprise dumbbell D122+ is more crystalline than the
4+ charged dumbbell D124+ and forms 3 soft-crystal phases
SC1 to SC3. The thermal transitions at 70 °C are reversible
based on POM and DSC, but XRD patterns of SC1 obtained
by precipitation from solution are different than the pattern of
SC3 obtained on cooling (see SI). Not fully reversible is the
clearing transition at 166 °C as it coincides with the onset of
thermal decomposition. A small domain focal conic-type
texture is observed by POM, similar to that observed for the
mesophases of D124+, and remains unchanged across the phase
transition at 72 °C. What is observed at the phase transition is a
softening of the material and a partial melting of the side chains
based on XRD, which is consistent with the transition enthalpy
of 22 kJ/mol measured by DSC.
In contrast to the results for dumbbells D122+ and D124+, the
expected increase in fluidity is observed for the SmA phase of
[2]rotaxane R122+ when compared to the SmA phase of R124+.
No transitions are observed by DSC, but POM studies reveal a
clearing temperature of 111 °C (see SI). Small domain focal
conic defect textures are consistent with a smectic mesophases,
and the pseudoisotropic appearance as a free-standing film
confirms its optically uniaxial character.
Figure 10. Diffraction patterns of RSi2+ at 125 and 25 °C upon
cooling from the isotropic phase. POM photomicrographs of RSi2+ at
25 °C upon cooling from the isotropic phase. (a,b) A fan-shaped focal
conic texture is obtained between untreated glass slides and a small
domain focal conic texture between cleaned, more hydrophilic, glass
slides.
The assignment as a SmA phase is based on the diffraction
patterns that consist of only two reflections, an intense small
angle peak at a d-spacing of about 39 Å and a broad halo at
about 4.6 Å that indicates an amorphous state of the side chains
(see SI). A second-order reflection of the layer spacing is
observed in the diffraction pattern of a free-standing film. Both,
POM and XRD results are consistent with the presence of a
SmA phase, and this assignment is also supported by the
agreement between measured and predicted layer spacings
(Table 1). The only change observed in the diffraction patterns
with increasing temperature is a small decrease in layer spacing
that is common in SmA phasesan increase in layer spacing
would be expected for a SmC phase.27
CONCLUSIONS
■
It has been demonstrated, for the first time, that ionic liquid
crystalline [2]rotaxanes based on a small-molecule design can
be prepared in gram quantities and good purity by the
optimization of design as well as synthetic and purification
procedures. Of the two presented designs, the 2+ charged
[2]rotaxanes are superior to the 4+ charged [2]rotaxanes
because their preparation is less time consuming, they are more
easily purified, and they are chemically and thermally more
stable. The change from 4+ to 2+ charged [2]rotaxane also
improves mesomorphism because (1) the fluidity of the
mesophases increases due to the reduction in ionic character
and (2) higher temperature phase transitions can be tolerated
because of the increased thermal stability.
A comparison of the mesomorphism of dumbbells and
[2]rotaxanes provides clear evidence that the interlocked
macrocycle promotes the formation of thermotropic ionic
liquid crystal phases and increases thermal stability. A reduction
in ionic interactions between cores (sheltering of the ions) is
proposed as the major reason for the observed change from soft
crystal phases for the dumbbells to SmA liquid crystal phases
for the [2]rotaxanes. This sheltering of ions is also an advantage
of the chosen [2]rotaxane design based on a charged core and a
neutral macrocycle, in contrast to the previously reported liquid
crystalline [2]rotaxane that consists of a neutral core and a
charged macrocycle.
Incorporation of siloxane chains in compound RSi2+ was
tested because siloxane chains usually lower viscosity and often
introduce SmC phases. However, the mesomorphism of RSi2+
is very similar to the mesomorphism of R122+, although the
incorporation of siloxane chains stabilizes the SmA phase by
increasing the clearing temperature by 40 to 151 °C (Figures 9
and 10). Again, no transitions are observed by DSC, and the
clearing temperature was determined by POM studies. Also
consistent with SmA mesomorphism are fan-shaped focal conic
textures and a pseudo-isotropic appearance of free-standing
films. Diffraction patterns contain an intense small angle peak at
a d-spacing of 42 Å and two broad halos at about 6.6 and 4.6 Å
for the amorphous siloxane and aliphatic side chains,
respectively. The observed layer spacing is in good agreement
with the estimated value and slightly decreases with temper-
ature as it is common in SmA phases.
In summary, a reduction in charge from 4+ to 2+ for
[2]rotaxanes R124+ and R122+ lowers transition temperatures
and generates a less viscous SmA phase. Exchange of dodecyl
by siloxane chains in [2]rotaxane RSi2+ stabilizes the SmA
phase compared to R122+; siloxane chains are more powerful
promoters of microphase segregation than aliphatic chains.16
Although, siloxane chains are known for their high propensity
to induce SmC phases, the SmA phase promoting character of
ionic liquid crystals appears to be more influential.9
Disappointing, however, is the low thermal stability of the
siloxane chains in these ionic compounds which limits their
utility.
ASSOCIATED CONTENT
* Supporting Information
Synthesis of all new compounds and their characterization by
1H and 13C NMR, HR-MS, IR, vt-PXRD, TGA, DSC and
POM. SC-XRD analyses of two model compounds with R =
CH3. This material is available free of charge via the Internet at
http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
eichhorn@uwindsor.ca; loeb@uwindsor.ca
Notes
■
The authors declare no competing financial interest.
H
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Bielawski, C. W. J. Mater. Chem. 2010, 20, 5709. (e) Causin, V.; Saielli,
G. J. Mater. Chem. 2009, 19, 9153. (f) Tanabe, K.; Yasuda, T.; Yoshio,
M.; Kato, T. Org. Lett. 2007, 9, 4271.
(11) Veber, M.; Berruyer, G. Liq. Cryst. 2000, 27, 671.
(12) Wiggins, K. M.; Kerr, R. L.; Chen, Z.; Bielawski, C. W. J. Mater.
Chem. 2010, 20, 5709.
ACKNOWLEDGMENTS
■
S.J.L. and S.H.E. are grateful for the awarding of NSERC of
Canada Discovery grants in support of this research. Additional
support was provided to S.J.L. through the NSERC Canada
Research Chair program.
(13) Bruce, D. W. Acc. Chem. Res. 2000, 33, 831.
(14) (a) Marcos, M.; Ros, M. B.; Serrano, J. L.; Esteruelas, M. A.;
Sola, E.; Oro, L. A.; Barbera. J. Chem. Mater. 1990, 2, 748.
(b) Gallardo, H.; Magnago, R.; Bortoluzzi, A. Liq. Cryst. 2001, 28,
1343.
(15) Kouwer, P. H. J.; Swager, T. M. J. Am. Chem. Soc. 2007, 129,
14042.
(16) Li, L.; Jones, C. D.; Magolan, J.; Lemieux, R. P. J. Mater. Chem.
2007, 17, 2313. Roberts, J. C.; Kapernaum, N.; Song, Q.;
Nonnenmacher, D.; Ayub, K.; Giesselmann, F.; Lemieux, R. P. J.
Am. Chem. Soc. 2010, 132, 364.
(17) (a) Mercer, D. J.; Vella, S. J.; Guertin, L.; Suhan, N. D.;
Tiburcio, J.; Vukotic, V. N.; Wisner, J. A.; Loeb, S. J. Eur. J. Org. Chem.
2011, 1763. (b) Loeb, S. J.; Tiburcio, J.; Vella, S. J.; Wisner, J. A. Org.
Biomol. Chem. 2006, 4, 667. (c) Mercer, D. J.; Yacoub, J.; Zhu, K.;
Loeb, S. K.; Loeb, S. J. Org. Biomol. Chem. 2012, 10, 6094. (d) Mercer,
D. J.; Loeb, S. J. Dalton Trans. 2011, 40, 6386. (e) Davidson, G. J. E.;
Loeb, S. J.; Passaniti, P.; Silva, S.; Credi, A. Chem.Eur. J. 2006, 12,
3233. (f) Loeb, S. J.; Tramontozzi, D. A. Org. Biomol. Chem. 2005, 3,
1393. (g) Loeb, S. J.; Wisner, J. A. Chem. Commun. 1998, 2757.
(18) (a) Suzuki, S.; Nakazono, K.; Takata, T. Org. Lett. 2010, 12, 712.
(b) Makita, Y.; Kihara, N.; Takata, T. J. Org. Chem. 2008, 73, 9245.
(c) Makita, Y.; Kihara, N.; Takata, T. Chem. Lett. 2007, 36, 102.
(d) Kawasaki, H.; Kihara, N.; Takata, T. Chem. Lett. 1999, 1015.
(19) (a) Davidson, G. J. E.; Sharma, S.; Loeb, S. J. Angew. Chem., Int.
Ed. 2010, 49, 4938. (b) Loeb, S. J.; Tiburcio, J.; Vella, S. J. Chem.
Commun 2006, 1598. (c) Suhan, N. D.; Allen, L.; Gharib, M. T.;
Viljoen, E.; Vella, S. J.; Loeb, S. J. Chem. Commun. 2011, 47, 5991.
(20) (a) Kitaigorodsky, A. I. Molecular Crystals and Molecules;
Academic Press: New York, 1973. (b) Ubbelohde, A. R. Melting and
Crystal Structure; Clarendon Press: Oxford, U.K., 1965. (c) Doolittle,
A. K. J. Appl. Phys. 1951, 22, 1471.
REFERENCES
■
(1) (a) Sawamura, M.; Kawai, K.; Matsuo, Y.; Kanic, K.; Kato, T.;
Nakamura, E. Nature 2002, 419, 702. (b) Donnio, B.; Buathong, S.;
Bury, I.; Guillon, D. Chem. Soc. Rev. 2007, 36, 1495.
(2) (a) Kato, T.; Mizoshita, N.; Kishimoto, K. Angew. Chem., Int. Ed.
2006, 45, 38. (b) Kato, T.; Yasuda, T.; Kamikawa, Y.; Yoshio, M.
Chem. Commun. 2009, 729. (c) Nealon, G. L.; Greget, R.; Dominguez,
C.; Nagy, Z. T.; Guillon, D.; Gallani, J. L.; Donnio, B. Beilstein J. Org.
Chem. 2012, 8, 349. (d) Rosen, B. M.; Wilson, C. J.; Wilson, D. A.;
Peterca, M.; Imam, M. R.; Percec, V. Chem. Rev. 2009, 109, 6275.
(e) Saez, I. M.; Goodby, J. W. In Liquid Crystalline Functional
Assemblies and Their Supramolecular Structures; Kato, T., Ed.; Springer-
Verlag: Berlin, Germany, 2008; Vol. 128, p 1. (f) Tschierske, C. Chem.
Soc. Rev. 2007, 36, 1930.
(3) (a) Fahrenbach, A. C.; Bruns, C. J.; Cao, D.; Stoddart, J. F. Acc.
Chem. Res. 2012, 45, 1581. (b) Rambo, Brett M.; Gong, Han-Yuan;
Oh, Moonhyun; Sessler, Jonathan L. Acc. Chem. Res. 2012, 45, 1390.
(c) Friscic, T. Chem. Soc. Rev. 2012, 41, 3493. (d) Beves, J. E.; Blight,
B. A.; Campbell, C. J.; Leigh, D. A.; McBurney, R. T. Angew. Chem.,
Int. Ed. 2011, 50, 9260. (e) Haenni, K. D.; Leigh, D. A. Chem. Soc. Rev.
2010, 39, 1240. (f) Gassensmith, J. J.; Baumes, J. M.; Smith, B. D.
Chem. Commun. 2009, 42, 6329. (g) Harada, A.; Hashidzume, A.;
Yamaguchi, H.; Takashima, Y. Chem. Rev. 2009, 109, 5974.
(h) Stoddart, J. F. Chem. Soc. Rev. 2009, 38, 1802. (i) Mullen, K.
M.; Beer, P. D. Chem. Soc. Rev. 2009, 38, 1701. (j) Crowley, J. D.;
Goldup, S. M.; Lee, A-L; Leigh, D. A.; McBurney, R. T. Chem. Soc. Rev.
2009, 38, 1530. (k) Faiz, J. A.; Heitz, V.; Sauvage, J.-P. Chem. Soc. Rev.
2009, 38, 422. (l) Li, S.; Liu, M.; Zheng, B.; Zhu, K.; Wang, F.; Li, N.;
Zhao, X.-L.; Huang, F. Org. Lett. 2009, 11, 3350. (m) Dichtel, W. R.;
Miljanic, O. S.; Zhang, W.; Spruell, J. M.; Patel, K.; Aprahamian, I.;
Heath, J. R.; Stoddart, J. F. Acc. Chem. Res. 2008, 41, 1750.
(n) Chmielewski, M. J.; Beer, P. D. Ed.; Atwood, J. L.; Steed, J. W.
Org. Nanostruct. 2008, 63. (o) Loeb, S. J. Ed.; Atwood, J. L.; Steed, J.
W. Org. Nanostruct. 2008, 33. (p) Champin, B.; Mobian, P.; Sauvage,
J.-P. Chem. Soc. Rev. 2007, 36, 358. (q) Vickers, M. S.; Beer, P. D.
Chem. Soc. Rev. 2007, 36, 211. (r) Saha, S.; Stoddart, J. F. Chem. Soc.
Rev. 2007, 36, 77. (s) Loeb, S. J. Chem. Commun. 2005, 1511.
(t) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.;
Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 899. (u) Kim, K. Chem.
Soc. Rev. 2002, 31, 96.
(21) Wunderlich, B. Thermal analysis; Academic Press: New York,
1990.
(22) (a) Rieger, A. L.; Edwards, J. J. Org. Chem. 1988, 53, 1481.
(b) Farring, J. A.; Ledwith, A.; Stam, M. F. Chem. Commun. 1969, 259.
(c) Novakovic, V.; Hoffman, M. Z. J. Am. Chem. Soc. 1987, 109, 2341.
(23) (a) Neve, F.; Crispini, A.; Armentano, S.; Francescangeli, O.
Chem. Mater. 1998, 10, 1904. (b) Ohta, K.; Sugiyama, T.; Nogami, T.
J. Mater. Chem. 2000, 10, 613. (c) Arkas, M.; Tsiourvas, D.; Paleos, C.
M.; Skoulios, A. Chem.Eur. J. 1999, 5, 3202. (d) Goossens, K.; Lava,
K.; Nockemann, P.; Van Hecke, K.; Van Meervelt, L.; Driesen, K.;
Gorller-Walrand, C.; Binnemans, K.; Cardinaels, T. Chem.Eur. J.
2009, 15, 656.
(24) (a) Kadam, J.; Faul, C. F. J.; Scherf, U. Chem. Mater. 2004, 16,
3867. (b) Sorensen, T. J.; Hildebrandt, C. B.; Elm, J.; Andreasen, J. W.;
Madsen, A. O.; Westerlund, F.; Laursen, B. W. J. Mater. Chem. 2012,
22, 4797.
(25) Nguyen, H. T.; Destrade, C.; Malthete, J. In Handbook of Liquid
Crystals; Demus, D., Goodby, J., Gray, G. W., Spiess, H.-W., Vill, V.,
Eds.; Wiley-VCH: Weinheim, Germany, 1998; Vol. 2B, p 865.
(26) Gray, G. W.; Goodby, J. W. Smectic liquid crystals: Textures and
structures; Heyden & Son: London, 1984.
(4) Aprahamian, I.; Yasuda, T.; Ikeda, T.; Saha, S.; Dichtel, W. R.;
Isoda, K.; Kato, T.; Stoddart, J. F. Angew. Chem., Int. Ed. 2007, 46,
4675.
(5) Baranoff, E. D.; Voignier, J.; Yasuda, T.; Heitz, V.; Sauvage, J.-P.;
Kato, T. Angew. Chem., Int. Ed. 2007, 46, 4680.
(6) Kidowaki, M.; Nakajima, T.; Araki, J.; Inomata, A.; Ishibashi, H.;
Ito, K. Macromolecules 2007, 40, 6859.
(7) Yasuda, T.; Tanabe, K.; Tsuji, T.; Coti, K. K.; Aprahamian, I.;
Stoddart, J. F.; Kato, T. Chem. Commun. 2010, 46, 1224.
(8) (a) Roy, D.; Cambre, J. N.; Sumerlin, B. S. Prog. Polym. Sci. 2010,
35, 278. (b) Stuart, M. A. C.; Huck, W. T. S.; Genzer, J.; Muller, M.;
̈
Ober, C.; Stamm, M.; Sukhorukov, G. B.; Szleifer, I.; Tsukruk, V. V.;
Urban, M.; Winnik, F.; Zauscher, S.; Luzinov, I.; Minko, S. Nat. Mater.
2010, 9, 101. (c) Yerushalmi, R.; Scherz, A.; van der Boom, M. E.;
Kraatz, H.-B. J. Mater. Chem. 2005, 15, 4480.
(27) Dierking, I. Textures of Liquid Crystals; Wiley-VCH: Weinheim,
Germany, 2003.
(9) (a) Binnemans, K. Chem. Rev. 2005, 105, 4148. (b) Axenov, K.
V.; Laschat, S. Materials 2011, 4, 206. (c) Chen, S.; Eichhorn, S. H. Isr.
J. Chem. 2012, 52, 830.
(10) (a) Tabushi, I.; Yamamura, K.; Kominami, K. J. Am. Chem. Soc.
1986, 108, 6409. (b) Bhowmik, P. K.; Han, S.; Nedeltchev, I. K.; Cebe,
J. J. Mol. Cryst. Liq. Cryst. 2004, 419, 27. (c) Veber, M.; Berruyer, G.
Liq. Cryst. 2000, 27, 671. (d) Wiggins, K. M.; Kerr, R. L.; Chen, Z.;
I
dx.doi.org/10.1021/ja309558p | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX