Synthesis and mesomorphic properties of rigid-core ionic liquid crystals

Article abstract of DOI:10.1021/ja075651a

Ionic liquid crystals combine the unique solvent properties of ionic liquids with self-organization found for liquid crystals. We report a detailed analysis of the structure-property relationship of a series of new imidazolium-based liquid crystals with an extended aromatic core. Investigated parameters include length and nature of the tails, the length of the rigid core, the lateral substitution pattern, and the nature of the counterion. Depending on the molecular structure, two mesophases were observed: a bilayered SmA ₂ phase and the more common monolayered SmA phase, both strongly interdigitated. Most materials show mesophases stable to high temperatures. For some cases, crystallization could be suppressed, and room-temperature liquid crystalline phases were obtained. The mesomorphic properties of several mixtures of ionic liquid crystals were investigated. Many mixtures showed full miscibility and ideal mixing behavior; however, in some instances we observed, surprisingly, complete demixing of the component SmA phases. The ionic liquid crystals and mixtures presented have potential applications, due to their low melting temperatures, wide temperature ranges, and stability with extra ion-doping.

Full text of DOI:10.1021/ja075651a

Published on Web 10/20/2007
Synthesis and Mesomorphic Properties of Rigid-Core Ionic
Liquid Crystals
Paul H. J. Kouwer‡ and Timothy M. Swager*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
77 Massachusetts AVenue, Cambridge, Massachusetts 02139
Received July 28, 2007; E-mail: tswager@mit.edu
Abstract: Ionic liquid crystals combine the unique solvent properties of ionic liquids with self-organization
found for liquid crystals. We report a detailed analysis of the structure-property relationship of a series of
new imidazolium-based liquid crystals with an extended aromatic core. Investigated parameters include
length and nature of the tails, the length of the rigid core, the lateral substitution pattern, and the nature of
the counterion. Depending on the molecular structure, two mesophases were observed: a bilayered SmA2
phase and the more common monolayered SmA phase, both strongly interdigitated. Most materials show
mesophases stable to high temperatures. For some cases, crystallization could be suppressed, and room-
temperature liquid crystalline phases were obtained. The mesomorphic properties of several mixtures of
ionic liquid crystals were investigated. Many mixtures showed full miscibility and ideal mixing behavior;
however, in some instances we observed, surprisingly, complete demixing of the component SmA phases.
The ionic liquid crystals and mixtures presented have potential applications, due to their low melting
temperatures, wide temperature ranges, and stability with extra ion-doping.
Introduction
materials, and as electrochemical components in various devices,
such as lithium ion batteries, fuel cells, solar cells, and
capacitors.7
Over the past decade ionic liquids (ILs) have become a mature
class of materials with applications as reaction media for organic
synthesis. Initial interest developed from the fact that ILs
eliminate the hazards associated with harmful and volatile
conventional solvents. Currently, many ILs can be categorized
as “task-specific” ionic liquids (TSIL), in which one or more
functionalities have been added to their primary role as solvent.1
For example, TSILs may show catalytic activity,2 function as a
ligand1a or as a solvent in chemical synthesis.3 Other important
emerging fields of application are in separation and extraction
processes,4 in the synthesis5 and functionalization6 of nanosized
Liquid crystals (LCs) offer the combination of order and
mobility in one material. When LC molecules only possess
orientational order, a nematic phase is observed. In a smectic
LC phase, the molecules show one-dimensional positional order,
resulting in a layered phase. The lowest-symmetry layered phase
is the smectic A (SmA) phase, with the average orientation of
the mesogen long axis normal to the layers. In a smectic C
(SmC) phase the molecules are tilted with respect to the layer
normal. Higher-ordered smectic phases have some degree of
in-plane and/or between-planes positional order, and many of
these phases are considered crystal or soft-crystal phases.
The unique properties liquid crystals offer have led to their
widespread application in display technology; however, many
other applications in a variety of fields are currently suggested
and investigated. One of those applications is the use of LCs
as a templating medium. Some examples are the alignment of
polymers8 or chromophores9 and controlled metal deposition
or nanoparticle synthesis.10
‡ Current address: Radboud University, Institute for Molecules and
Materials, Toernooiveld 1, 6525 ED Nijmegen, The Netherlands. E-mail:
kouwer@science.ru.nl.
(1) For recent reviews on task-specific ionic liquids: (a) Lee, S. Chem.
Commun. 2006, 1049-1063. (b) Davis, J. D, Jr. Synthesis of task-specific
ionic liquids. In Ionic Liquids in Synthesis; Wasserscheid, P., Welton, T.,
Eds.; Wiley-VCH: Berlin, 2003; pp 33-40.
(2) For examples of Friedel-Crafts reactions: (a) Boon, J. A.; Levisky, J. A.;
Pflug, J. L.; Wilkes, J. S. J. Org. Chem. 1986, 51, 480. Esterifications: (b)
Cole, A. C.; Jensen, J. L.; Ntai, I.; Tran, K. L. T.; Weaver, K. J.; Forbes,
D. C.; Davis, J. H. J. Am. Chem. Soc. 2002, 124, 5962-5963.
(3) (a) Fraga-Dubreuil, J.; Bazureau, J. P. Tetrahedron Lett. 2001, 42, 6097-
6100. (b) Miao, W.; Chan, T. H. Org. Lett. 2003, 5, 5003-5005. (c)
Anjaiah, S.; Chandrasekhar, S.; Gree, R. Tetrahedron Lett. 2004, 45, 569-
571.
(4) (a) Visser, A. E.; Swatloski, R. P.; Reichert, W. M.; Davis, J. H.; Rogers,
R. D.; Mayton, R.; Sheff, S.; Wierzbicki, A. Chem. Commun. 2001, 135-
136. (b) Visser, A. E.; Swatloski, R. P.; Reichert, W. M.; Mayton, R.; Sheff,
S.; Wierzbicki, A.; Davis, J. H.; Rogers, R. D. EnViron. Sci. Technol. 2002,
36, 2523-2529.
The merger of ILs and LCs has yielded a new class of
materials called ionic liquid crystals (ILCs). These self-
(6) (a) Itoh, H.; Naka, K.; Chujo, Y. J. Am. Chem. Soc. 2004, 126, 3026-
3027. (b) Kim, K.-S.; Demberelnyamba, D.; Lee, H. Langmuir 2004, 20,
556-560.
(7) (a) Trulove, P. C.; Mantz, R. A. Electrochemical Properties of Ionic Liquids.
In Ionic Liquids in Synthesis; Wasserscheid, P., Welton, T., Eds.; Wiley-
VCH: Berlin, 2003; pp 103-126. (b) Ohno, H., Ed. Electrochemical
Aspects of Ionic Liquids; Wiley: Hoboken, NJ, 2005. (c) Buzzeo, M. C.;
Evans, R. G.; Compton, R. G. ChemPhysChem. 2004, 5, 1106-1120. (d)
Bonhoˆte, P.; Dias, A.-P.; Papageorgiou, N.; Kalyanasundaram, K.; Gra¨tzel,
M. Inorg. Chem. 1996, 35, 1168-1178. (e) Forsyth, S. A.; Pringle, J. M.;
MacFarlane, D. R. Aust. J. Chem. 2004, 57, 113-119.
(5) (a) Gadenne, B.; Hesemann, P.; Moreau, J. J. E. Chem. Commun. 2004,
1768-1769. (b) Dobbs, W.; Suisse, J.-M.; Douce, L.; Welter, R. Angew.
Chem., Int. Ed. 2006, 45, 4179-4182. (c) Zhu, Y. J.; Wang, W. W.; Qi,
R. J.; Hu, X. L. Angew. Chem., Int. Ed. 2004, 43, 1410. (d) Wang, Y.;
Yang, H. J. Am. Chem. Soc. 2005, 127, 5316.
9
14042
J. AM. CHEM. SOC. 2007, 129, 14042-14052
10.1021/ja075651a CCC: $37.00 © 2007 American Chemical Society

Synthesis of Rigid-Core Ionic Liquid Crystals
A R T I C L E S
organizing systems offer great advantages over isotropic sys-
tems. For instance, ion conduction was enhanced in the SmA
and columnar phases compared to that in the isotropic phase.11
ILCs have also been used as an anisotropic medium for
electrochemistry, where the ILC acted as the solvent, the
electrolyte, and the template. Applications in dye-sensitized solar
cells have been proposed for these systems.12 Other efforts have
reported the polymerization of conducting polymers in lyotropic
phase,13 although the latter was a multicomponent system, to
which a common electrolyte was added.
Beginning in the late 1980s, an increasing number of reports
on ILCs have appeared in the literature. Early papers focused
on the mesomorphic properties of small aromatic systems
(imidazolium, pyridinium, etc.), substituted by one or multiple
long aliphatic tails.14 The SmA phase displayed by such systems
often requires a minimum length of the tail (usually C12), and
the observed mesophases strongly depend on the counterion.
Replacement of one tail with a threefold substituted benzylgroup
gives a wedge-shaped ILC that forms columnar (Col) meso-
phases.11b,c In these types of imidazolium- and pyridinium-based
materials, the liquid crystalline properties originate from their
strong amphiphilic character. Excluded volume effects induced
by a rigid anisotropic molecular core, which are one of the main
driving forces for mesophase formation in conventional liquid
crystals, play no significant role. As a result, no nematic
mesophases can be expected from such systems.
Figure 1. Generic structure of the prepared series of ILCs with different
parameters. A total of 47 molecules have been prepared.
date, few mesogens have been described where charged imi-
dazolium groups are incorporated in the rigid core of the
mesogen.17
We report herein a series of ILCs that bridge the classes of
ILs and LCs. Through recent developments in aromatic ami-
nation chemistry,18 we were able to conveniently integrate
imidazolium groups in the core of the mesogen, which now
has the shape of a conventional rod-shaped mesogen. The
molecular structures have implications for the organization in
the material, and the extended aromatic core increases the order
parameter in the LC phase.
A wide variety of parameters were examined for their
influence on the mesomorphic properties (Figure 1). These
include the tail structure (length, branching, chirality, position),
core size, lateral substituents, and counterion. The liquid
crystalline properties have been investigated by optical polar-
izing microscopy (OPM), differential scanning calorimetry
(DSC), and X-ray diffraction (XRD). Our conclusions are
supported by molecular modeling experiments.
More recently, ILCs have been described in the literature with
a conventional liquid crystal mesogen linked via a flexible linker
to a conventional ionic liquid, most commonly a 1-methylimi-
dazolium (MIm) group. MIm groups attached to rod-shaped LCs
have yielded SmA and SmE phases, even when the liquid crystal
used was a strong nematogen.15 When multiple MIm groups
were attached on the tail ends of discotic liquid crystals,
stabilized columnar phases were observed.16 The architecture
of these molecules is unconventional, as the imidazolium group
is a polarizable rigid group that is suitable as part of a mesogenic
core rather than being placed at the ends of flexible tails. To
Results and Discussion
(8) Zhu, Z.; Swager, T. M. J. Am. Chem. Soc. 2002, 124, 9670-9671.
(9) (a) Long, T. M.; Swager, T. M. AdV. Mater. 2001, 13, 601-604. (b) Long,
T. M.; Swager, T. M. J. Am. Chem. Soc. 2002, 124, 3826-3827. (c)
Amemiya, K.; Shin, K. C.; Takanishi, Y.; Ishikawa, K.; Azumi, R.; Takezoe,
H. J. Appl. Phys. 2004, 43, 6084-6087.
Experimental. The synthesis and chemical characterization
(1H, 13C NMR) of compounds 1-36 are described in the
Supporting Information. After purification by crystallization and/
or column chromatography, the materials were dissolved, filtered
(0.2 µm pores), and then dried for at least 24 h in vacuum before
analysis. Mixtures of different ILCs were prepared by dissolving
the appropriate amounts of ILC in a common solvent, which
was then removed by evaporation. Mixtures were dried for 24
h under vacuum prior to analysis. ILCs with relatively hydro-
phobic counterions did not show different phase behavior after
exposure to atmospheric conditions.
(10) (a) Gray, D. H.; Gin, D. L. Chem. Mater. 1998, 10, 1827. (b) Dag, O¨ .;
Alayoglu, S.; Tura, C.; C¸ elik, O¨ . Chem. Mater. 2003, 15, 2711-2717. (c)
Yamauchi, Y.; Momma, T.; Yokoshima, T.; Kuroda, K.; Osaka, T. J. Mater.
Chem. 2005, 15, 1987-1994.
(11) (a) Hoshino, K.; Yoshio, M.; Mukai, T.; Kishimoto, K.; Ohno, H.; Kato,
T. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 3486-3492. (b) Yoshio,
M.; Mukai, T.; Ohno, H.; Kato, T. J. Am. Chem. Soc. 2004, 126, 994-
995. (c) Yoshio, M.; Kagata, T.; Hoshino, K.; Mukai, T.; Ohno, H.; Kato,
T. J. Am. Chem. Soc. 2006, 128, 5570-5577.
(12) Yamanaka, N.; Kawano, R.; Kubo, W.; Kitamura, T.; Wada, T.; Watanabe,
M.; Yanagida, S. Chem. Commun. 2005, 740-742.
(13) (a) Hulvat, J. F.; Stupp, S. I. Angew. Chem., Int. Ed. 2003, 42, 778. (b)
Hulvat, J. F.; Stupp, S. I. AdV. Mater. 2004, 16, 589-592.
(14) (a) Bowlas, C. J.; Bruce, D. W.; Seddon, K. R. Chem. Commun. 1996,
1625. (b) Gordon, C. M.; Holbrey, J. D.; Kennedy, A. R.; Seddon, K. R.
J. Mater. Chem. 1998, 8, 2627. (c) Holbrey, J. D.; Seddon, K. R. J. Chem.
Soc., Dalton Trans. 1999, 2133. (d) Tosoni, M.; Laschat, S.; Baro, A. HelV.
Chim. Acta 2004, 87, 2742. (e) Knight, G. A.; Shaw, B. D. J. Chem. Soc.
1938, 682. (f) Sudholter, E. J. R.; Engberts, J. B. F. N.; de Jeu, W. J.
Phys. Chem. 1982, 86, 1908.
Synthesis. The synthesis of the ILCs studied is outlined in
Schemes 1 and 2. Alkoxy-tails were introduced at the far end
of the mesogen under standard Williamson etherification
conditions. To prepare 9, the phenol group was protected with
a tetrahydropyranyl group. However, it was recently published
(15) For example: (a) Hoshino, K.; Yoshio, M.; Mukai, T.; Kishimoto, K.; Ohno,
H.; Kato, T. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 3486. (b)
Navarro-Rodriguez, D.; Frere, Y.; Gramain, P.; Guillon, D.; Skoulios, A.
Liq. Cryst. 1991, 9, 321. (c) Cui, L.; Sapagovas, V.; Lattermann, G. Liq.
Cryst. 2002, 29, 1121. (d) Yoshizawa, H.; Mihara; Koide, N. Mol. Cryst.
Liq. Cryst. 2004, 423, 61. (e) El Hamaoui, B.; Zhi, L.; Pisula, W.; Kolb,
U.; Wu, J.; Mu¨llen, K. Chem. Commun. 2007, 2384.
(17) (a) Kosaka, Y.; Kato, T.; Uryu, T. Liq. Cryst. 1995, 18, 693-698. (b) Suisse,
J. M.; Bellemin-Laponnaz, S.; Douce, L.; Maisse-Francois, A.; Welter, R.
Tetrahedron Lett. 2005, 46, 4303-4305.
(18) (a) Altman, R. A.; Buchwald, S. L. Org. Lett. 2006, 8, 2779-2782. (b)
Zhang, H.; Cai, Q.; Ma, D. J. Org. Chem. 2005, 70, 5164-5173. (c) Ma,
D.; Cai, Q.; Zhang, H. Org. Lett. 2003, 5, 2453.
(16) Motoyanagi, J.; Fukushima, T.; Aida, T. Chem. Commun. 2005, 101.
9
J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007 14043

A R T I C L E S
Kouwer and Swager
Scheme 1. Synthesis of the ILC Precursorsa
a Key: (i) RBr, K2CO3, KI, butanone, 16 h reflux; (ii) K2CO3, CuI, N,N-dimethylglycine, DMSO, 24-48 h at 110 °C; (iii) Ce2CO3, CuI, 4,7-dimethoxy-
1,10-phenanthroline, PEG-600, butyronitril 4-24 h reflux; (iV) K2CO3, CuI, L-proline, DMSO, 16 h at 110 °C; (V) tetrahydropyran, p-toluenesulfonic acid,
CH2Cl2, 2 h at room temperature; (Vi) MeOH, 1 N HCl, 2 h at room temperature; (Vii) (a) 5N KOH, EtOH, 2 h reflux; (b) concd HCl to neutral; (Viii)
1,3-dicyclohexylcarbodiimide (DCC); 4-(N,N-dimethylamino)pyridine (DMAP), THF, 40 h at 50 °C.
Scheme 2. Synthesis of the ILCsa
a Key: (i) Propyl or dodecyl iodide, neat, 4-24 h at 100 °C; (ii) 12, 13, or 14, toluene, 16-40 h at 100 °C; (iii) ion exchange: BF4- and PF6- by multiple
-
-
washes with concentrated aqueous NaBF4 or NH4PF6 solutions, respectively; ClO4 through a AgClO4 solution; H25C12SO3 (dodecylsulfonate) through
prefunctionalized beads.
that imidazole can be coupled directly to 4-iodophenol in an
excellent yield.18a
The alkylation of 1-arylimidazoles needed optimization, and
we found that alkyliodides are required to give full conversion
under reasonable conditions. Therefore, the (commercially)
available alkylbromides and the tosylates20 were converted into
iodides 12-14 by means of Finkelstein-type reactions. The
alkylations with alkyliodides were conducted either without any
solvent (n-propyliodide and n-dodecyliodide) or in toluene
solution. The reduced concentration when a solvent is used
generally leads to longer reaction times. After the alkylation
reaction, pure products were obtained by multiple crystalliza-
tions, if necessary preceded by column chromatography. Iodide
ions were exchanged by multiple washings of a dichloromethane
solution of the ILC with concentrated aqueous sodium salt
The key step in the synthesis was an Ullman-type coupling
of an aryl iodide or bromide with imidazole and 2-substituted
imidazole derivatives. Recent literature shows that the copper-
catalyzed reaction can be facilitated by the use of specific
ligands, which makes this type of reaction a particularly useful
tool in the formation of aryl heterocyclic compounds.18 The
procedures described by Ma et al.18b,c were used for unsubsti-
tuted imidazoles. In their work, they describe good to excellent
yields with commercially available, inexpensive ligands. The
reactions are relatively slow, but minimal byproducts are
produced, and any unreacted arylhalide is easily recovered in
the purification procedure. For 2-alkyl- or 2-phenyl-substituted
imidazoles, these reaction conditions gave low conversions, and
the dimethoxyphenanthroline ligand following the methods of
Buchwald et al. gave superior results.18a Under Buchwald’s
conditions, we were able to obtain moderate yields for imida-
zoles with bulky substituents (pentyl and phenyl) at the
2-position.19
-
solutions (BF4 and PF6-), by precipitation of silver iodide
(using AgClO4), or by treatment with pre-functionalized beads
(sulfonates). Tables 1 and 2 show the variation in the materials
that were prepared.
Mesomorphic Properties. To systematically investigate the
influence of specific chemical parameters on the mesophase
behavior, we will discuss the materials in separate series in
(19) Note that for reactions with the bulky 2-phenylimidazole, we were not able
to obtain the high yields reported in literature (ref 18a).
(20) Jullien, L.; Canceill, J.; Lacombe, L.; Lehn, J. M. J. Chem. Soc., Perkins
Trans. 2 1995, 417-426.
9
14044 J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007

Synthesis of Rigid-Core Ionic Liquid Crystals
A R T I C L E S
Table 1. Prepared ILCs with Yields of the Alkylation Step and the
Ion Exchange
Table 2. Prepared ILCs with Yields of the Alkylation Step and the
Ion Exchange
R
′′ a
X-
y/%
mesogen
R
′
R
′′
X-
y/%
mesogen
26a
26b
27a
27b
28a
28b
29a
29b
30a
30b
31a
31b
32a
32b
CH3
CH3
CH3
CH3
C2H5
C2H5
C2H5
C2H5
C5H11
C5H11
C5H11
C5H11
Ph
n-C3H7
n-C3H7
n-C12H25
n-C12H25
n-C3H7
I-
92
92
90
93
87
97
82
97
83
98
76
93
67
72
15a
15b
15c
15d
15e
16a
16b
17a
17b
18a
18b
19a
19b
n-C3H7
n-C3H7
n-C3H7
n-C3H7
n-C3H7
n-C12H25
n-C12H25
TMD
TMD
S-citronellyl
S-citronellyl
Me(OCH2CH2)3-
Me(OCH2CH2)3-
I-
92
97
98
88
71
86
98
84
98
91
96
70
92
-
-
BF4
BF4-
I-
PF6
-
-
-
-
-
-
-
BF4
I-
ClO4
-
H25C12SO3
I-
n-C3H7
BF4
I-
-
n-C12H25
n-C12H25
n-C3H7
BF4
I-
BF4
I-
-
BF4
I-
n-C3H7
BF4
I-
-
n-C12H25
n-C12H25
n-C12H25
n-C12H25
BF4
I-
BF4
I-
BF4
-
BF4
Ph
mesogen
Ar
R
′′ a
X-
y/%
mesogen
R
′′ a
X-
y/%
20a
20b
21a
21b
22a
22b
23a
24a
24b
25a
25b
PhCO2Ph
PhCO2Ph
PhCO2Ph
PhCO2Ph
PhCO2Ph
PhCO2Ph
Ph
Ph
Ph
Ph
Ph
n-C3H7
n-C3H7
n-C12H25
n-C12H25
TMD
I-
89
95
92
94
90
96
65
78
87
75
90
-
-
-
33a
33e
34a
34b
35e
35a
35b
36a
36b
n-C3H7
n-C3H7
I-
80
52
90
96
62
92
93
59
67
BF4
I-
-
-
H25C12SO3
n-C12H25
n-C12H25
n-C12H25
TMD
I-
BF4
BF4
I-
-
TMD
BF4
H25C12SO3
I-
I-
n-C3H7
n-C12H25
n-C12H25
TMD
I-
-
TMD
BF4
-
-
Me(OCH2CH2)3-
Me(OCH2CH2)3-
I-
BF4
I-
-
BF4
TMD
BF4
a TMD ) 3,7,11-trimethyldodecyl; Me(OCH2CH2)3- ) 3,6,9-trioxade-
cyl.
a TMD ) 3,7,11-trimethyldodecyl; S-DMO ) S-3,7-dimethyloctyl;
Me(OCH2CH2)3- ) 3,6,9-trioxadecyl.
some materials show a cold crystallization process preceding
the melting transition (for instance 15c), whereas in others, no
crystallization is observed (for instance 20b). Because the glass
transition temperatures are generally around room temperature,
the SmA phase of the latter group shows an unusually broad
temperature window, which is beneficial for applications. The
supercooled state of the latter materials is stable for months.
Some materials in the series displayed crystal polymorphism
(e.g., traces 18b and 21b). At lower temperatures, multiple
transitions, probably (soft) crystal-crystal transitions, are visible
in the DSC traces. Compound 22a shows three such transitions
before melting into the SmA phase. Some laterally substituted
ILCs show transitions to another smectic phase on decreasing
temperatures.
which only one structural element is varied. Compounds 15 and
16 have been used as “internal standards” in all series. Tables
with the mesomorphic properties and XRD results of all syn-
thesized materials are included in the Supporting Information.21
OPM analysis of the LC samples consistently showed the
characteristic focal conical textures of a SmA phase (Figure 2a).
The textures are often combined with large spontaneous
homeotropically aligned areas that are optically isotropic when
viewed normal to the glass surfaces, particularly on cooling from
the isotropic phase, even between unprepared glass slides (Figure
2b). At lower temperature, crystal and soft crystal phases are
observed. Optical textures of these lower temperature phases,
shown in Figure 2c,d, are not conclusive, and no attempt has
been made to further elucidate their identity.
X-ray diffraction studies in the mesophase show the charac-
teristic SmA pattern with a strong fundamental (001) reflection,
a small first order (002) reflection, and a diffuse reflection of
the alkyl tails around 4.5 Å (Figure 4). Typically, in the (soft)
crystal phases at lower temperatures, the layered organization
is conserved. The presence of sharper reflections at lower
temperatures in the wide angle region indicates the increased
order within the layers of these phases.
DSC measurements were performed on all liquid crystalline
materials and mixtures. Transition temperatures were recorded
as the local maxima of the endotherm. Five characteristic
examples are shown in Figure 3. Generally, the DSC traces were
simple and showed two transitions, the melting and the clearing
point. Some materials showed suppression of crystallization.
In the first heating, the solid, as obtained from the synthesis,
shows a clear melting transition. However, the reverse process
is not observed in the cooling run. Instead, a glass transition is
observed around room temperature. In the second heating run,
The (001) reflections were fit with a Gaussian distribution
to obtain the layer spacings d001 over a wide temperature range
(see the inset in Figure 4). Depending on the alkyl substitution
pattern of the ILC, the molecules organize either in monolayers
SmAd with d001 < Lcalc, or in bilayers SmA2 with Lcalc < d001
(21) Note that more ionic liquid crystals have been synthesized than discussed
in the body of the paper. All synthesized materials are in the Supporting
Information.
9
J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007 14045

A R T I C L E S
Kouwer and Swager
Figure 2. Optical polarizing microscopy images: (a) 24b at T ) 130 °C
(focal conic SmA phase); (b) 15e at T ) 178 °C; (c) 22a at T ) 182 °C
(CrZ phase); (d) 24b at T ) 50 °C (CrX phase). All pictures are taken with
crossed polarizers. The scale-bar is shown in the figures.
Figure 3. Normalized DSC traces of selected mesogens: 16b, characteristic
trace of an ILC; 15c, suppression of crystallization with cold crystallization;
20b complete suppression of crystallization; 18b and 21b polymorphism
at lower temperatures. All traces are second heating runs and first cooling
runs. Third cooling runs were identical to the second, provided that no
degradation took place (T > 230 °C).
< 2 Lcalc, where Lcalc is the molecular length of the fully ex-
tended mesogens with the aliphatic tails in an all-trans config-
uration. Hence, the experimentally determined molecular length
LXRD equals the layer spacing d001 for monolayered materials
and equals half of the layer spacing in the case of bilayers.
Molecular modeling studies (Supporting Information) suggest
that the long axis of the mesogens is considerably larger than
the experimental layer thickness in the SmA phase, which
indicates that the aliphatic tails are strongly interdigitated. To
quantify the extent of interdigitation the ratio LXRD/Lcalc was
calculated. Since the layer spacing in many materials is strongly
dependent on the temperature, LXRD was determined at reduced
temperature T ) 0.95 TIso. ILC 17b is an indicative example
(Figure 4) with Lcalc ) 41.2 Å and LXRD ) 30.9 Å; the flexible
tails are largely interdigitated at this temperature. The inter-
digitation is likely the result of an increased effective volume
of the mesogen core due to the presence of the counterion.
All of SmA layer spacings display a strong linear temperature
dependence (d001 ) R + âT). Interestingly, 17b, LXRD extrapo-
lated to T ) 0 K gives a value of 42.5 Å, which is close to the
calculated value of 41.2 Å. An analysis of the temperature
dependence and the interdigitation of the different series of
materials will be discussed later. The layer spacings of the
crystal phase(s) were either similar to the SmA phase or smaller
with a continued reduction in length with decreasing temper-
ature. The example shown in Figure 4 indicates a tilt of the
mesogens in the layers below the SmA phase, although no (soft)
crystal phase assignments have been made from X-ray patterns.
Figure 4. XRD results from 17b, recorded cooling from the isotropic phase
(diffractograms every 5 °C recorded with exposure times of 10 min per
measurement). (Inset) Temperature dependence of the layer spacing. The
dashed lines are linear fits of the SmA phase and the soft crystal phase.
Fitting results (°C) are indicated in the figure.
Series 1. Evaluation of the Influence of the N-Alkyl
Substituent. We investigated ILCs by first modifying the length
of the aliphatic tails. Shorter tails generally give higher clearing
temperatures and mesophases with no positional order (i.e.,
nematic phases). Increasing the tail length initially reduces
9
14046 J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007

Synthesis of Rigid-Core Ionic Liquid Crystals
A R T I C L E S
The mesogen carrying the short propyl tail (15), displays a
different behavior. Although the transition temperature of 15 is
in the range of 16-18, the latent heat at the clearing point is
much lower, indicating a lower degree of organization in the
-
SmA phase. Interestingly, upon ion exchange from I- to BF4
crystallization is suppressed, and a very wide mesophase is
formed, ranging from room temperature to nearly 200 °C.
The glycol-functionalized mesogens are very different from
other members of the series. The transition temperatures are
very low: 19a shows glass transition at -1 °C, 19b melts at
8 °C, and both show clearing temperatures that are 100 °C lower
than those of the rest of the series. The deviant behavior of 19
is likely associated with the presence of dipolar interactions
between the imidazolium ion and the oligoglycol tail. When
one equivalent of LiBF4 is added to 19b, the clearing temper-
ature jumps up to values similar to those of the other members
of the series. We suggest that Li+ ions in the mixture compete
with the imidazolium group in complexing the glycol tails. This
allows for a more rod-shaped conformation of the mesogen,
which in turn increases the clearing temperatures. This argument
is in line with preliminary molecular modeling results, discussed
in the Supporting Information.
-
The BF4 salts of 15-19 were subjected to XRD at
temperatures throughout the SmA phase. Characteristic SmA
patterns were observed in all cases with a strong sharp
fundamental layer reflection d001 and diffuse reflections in the
wide-angle region. The layer spacings, obtained after fitting the
(001) reflections, are shown in Figure 6. Materials with long
aliphatic tails 16b, 17b, and 18b, formed monolayers with layer
spacings of the order of 3 nm. The degree of interdigitation at
0.95 TIso is comparable for all three compounds. Moreover, the
bilayers formed by propyl-substituted 15b are interdigitated to
an extent similar to that of the monolayers formed by 16-18.
Compound 19b with the glycol tails forms bilayers as well but
shows much stronger interdigitation than any other member of
the series. We attribute this behavior and the large temperature
dependence of the layer spacing (15% over only 30 °C) for
19b to the imidazolium-glycol interaction. Addition of LiBF4
to 19a reduces this strong temperature dependence to trends in
accord with those observed for 15b; however, the interdigitation
remains.
Figure 5. Mesomorphic properties of 15-19 (generic structures shown
-
above the diagrams): (a) Iodide salts and (b) BF4 salts. Note that 16a
starts decomposing above 250 °C. Upon fast heating a clearing temperature
of approximately 270 °C was observed in microscopy experiments.
clearing temperatures and promotes smectic phases. We limited
ourselves to short (C3) or long (C12) linear tails and investigated
introduction of multiple small branches by using racemic (3,7,-
11-trimethyldodecyl) and chiral ((S)-citronellyl) tails. We also
prepared an oligo(ethyleneglycol) derivative to study the
interactions between the imidazolium salt and a polar tail. The
synthesis and subsequent ion-exchange reactions yielded both
Series 2. The Influence of the Counterion. When compared
to conventional liquid crystals, ILCs have an additional param-
eter for mesophase manipulation: the counterion. The counterion
has a large effect on both the melting and the clearing tempera-
tures.22 We chose to investigate a small selection of counterions,
-
iodide and BF4 salts for parallel analysis. The transition
temperatures are summarized in Figure 5. Latent heat values
for the BF4 salts are shown in Table 3.
-
including BF4-, PF6-, and ClO4 ions. These ions have
hydrophobic tendencies, and a low sensitivity to atmospheric
humidity was anticipated. The trends are mesogen dependent.
ILCs with long aliphatic tails 16-18 show very similar
mesomorphic properties, both in transition temperatures and
latent heat. The presence of small branches has limited influence
on the mesophase stability. In other reports on non-ionic liquid
crystals, it has been found that the use of the trimethyldodecyl
tail, with disordered methyl groups, widens the mesophases by
suppression of crystallization. It has also been reported that the
chiral branches in the citronellyl group cause an increase in the
melting temperature. However, both modifications have limited
effects in our prepared series of ILCs. The phase behavior of
the BF4- salts is comparable to that of the corresponding iodide
salts. Further details of the influence of the counterion are
discussed later in Series 2.
However, for most materials, halides show higher clearing
- 22
temperatures than the lipophilic anions PF6-, BF4-, and ClO4
.
In the series 15a-d, the clearing points decrease in the
order: I- > BF4- > ClO4- > PF6- (Figure 7 and Table 3). In
addition, we observed that the crystallization of 15b and 15d is
suppressed, and hence, the mesophase window can be extended
to room temperature to give a SmA phase width of 150 °C.
Table 4 summarizes the DSC and XRD data for 15a-15e.
The ILCs 15a-15d, with small counterions, show strong
similarities both in latent heat at the clearing temperature (the
(22) Binnemans, K. Chem. ReV. 2005, 105, 4148-4204.
9
J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007 14047

A R T I C L E S
Kouwer and Swager
Table 3. Latent Heat and XRD Data of N-Alkyl Substituent Variation
1
1
e
LC
R
′′
∆HCr
a /kJ mol-
∆HSmA
b /kJ mol-
Lcalc c /Å
LXRD d,e/Å
LXRD/Lcalc
-SmA
-Iso
15b
16b
17b
18b
19b
n-propyl
n-dodecyl
3,7,11-tridodecyl
(S)-citronellyl
3,6,9-trioxadecyl
-
0.6
30.5
41.5
41.2
36.7
38.0
38.0
38.0
44.5
31.7
30.9
29.5
48.4
45.5
45.2
0.73
0.76
0.75
0.80
0.64
0.60
0.59
13.1
14.3
14.4
10.4
-
3.4
3.7
2.5
0.2
0.2
0.2
19b + 0.5 equiv LiBF4
19b + 1.0 equiv LiBF4
-
a Latent heat at the crystal to SmA transition (taken from the second heating trace). b Latent heat at the SmA to isotropic transition. c Length of the
stretched molecule, determined by molecular modeling. d Layer spacing, determined by XRD experiments: for monolayers LXRD ) d001, for bilayered materials
LXRD ) 1/2 × d001
.
e LXRD was determined at T ) 0.95TIso (after interpolation of the data).
Table 4. Latent Heat and XRD Data of 15a-15e
a
b
c
d.e
∆
HCr
/
∆
HSmA
/
Lcalc
Å
/
LXRD
Å
/
-
SmA
-
Iso
1
1
e
LC
X-
kJ mol-
kJ mol-
LXRD/Lcalc
15a
15b
15c
15d
15e
I-
BF4
PF6
38.8
-
31.0
-
50.6
0.6
0.6
0.6
0.5
3.0
30.5
30.5
30.5
30.5
44.2
41.0
44.4
45.5
43.5
32.0
0.67
0.73
0.75
0.71
0.72
-
-
-
ClO4
25C12SO3
-
H
a Latent heat at the crystal to SmA transition (taken from the second
heating trace). b Latent heat at the SmA to isotropic transition. c Length of
the stretched molecule, determined by molecular modeling. d Layer spacing,
determined by XRD experiments, for monolayers LXRD ) d001, for bilayered
materials LXRD ) 1/2 × d001
.
e LXRD was determined at T ) 0.95TIso (after
interpolation of the data).
Figure 6. XRD-determined time dependence of the layer spacing in the
SmA phase of the BF4- salts of 15-19. Note that 15b and 19b form bilayers,
whereas for 16b, 17b, and 18b, monolayers are observed.
Figure 8. Temperature dependence of the layer spacing in the SmA phase
of 15a-e as determined by XRD experiments. Note that only 15e forms
monolayers, while bilayers are observed for 15a-d.
To contrast the small hydrophobic counterions, we also
prepared a derivative with a dodecylsulfonate group 15e,
wherein the counterion bears another hydrocarbon tail. Interest-
ingly, this material behaves completely differently from that of
the other members of this series. The latent heat at the clearing
transition is much larger than that of 15a-d; in fact, it is nearly
the same as that observed for 16b, the dodecyl-substituted ILC.
XRD experiments reveal monolayer formation for 15e.
Series 3 and 4. The Influence of the Core Size and Lateral
Substituents. In (conventional) liquid crystals, one of the driving
forces for mesophase formation is excluded volume. The energy
of a rigid anisotropic body can be decreased by aligning with
Figure 7. Mesomorphic properties of 15a-15e (generic structures shown
above the diagrams).
amount of SmA order lost at the transition) and their layer
spacings. They form highly interdigitated SmA2 phases and
-
similar layer spacings with iodine being the largest (I- > BF4
> PF6 > ClO4-). The temperature dependencies of the layer
-
spacings (Figure 8) are nearly identical for all these materials.
9
14048 J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007

Synthesis of Rigid-Core Ionic Liquid Crystals
A R T I C L E S
Figure 9. Mesomorphic properties of Series 3 (ILCs with different
-
cores): (a) series R′′ ) propyl, X- ) I- (top) and X- ) BF4 (bottom);
-
(b) same series with R′′ ) dodecyl and X- ) I- (top) and X- ) BF4
Figure 10. Mesomorphic properties of series 4 (ILCs with lateral
-
substituents): (a) series R′′ ) propyl, X- ) I- (top) and X- ) BF4
(bottom). The generic structures are shown above the diagram.
(bottom); (b) same series with R′′ ) dodecyl and X- ) I- (top) and X-
)
-
BF4 (bottom). The generic structures are shown in the diagram.
its environment. This explains why liquid crystal phases are
observed predominantly in anisometric molecules. Aspect ratio
has a major effect on the phase stability as well as the type of
mesophase. Lateral substitution of the core with (small) side
groups decreases the aspect ratio and often prevents dense
molecular packing in the mesophase, which usually leads to
decreased order in the mesophase and lower clearing temper-
atures.
We prepared ILCs with different core sizes and decided to
use the imidazolium group for lateral substitution in order to
(i) facilitate synthesis, as 2-substituted imidazoles are com-
mercially available or readily accessible and (ii) improve
stability since it removes the potentially labile proton on the
imidazolium ring.
Figure 9 compares the transition temperatures of four groups
of ILCs with cores differentially substituted with: (a) propyl
substituents and (b) dodecyl substituents. The results of the I-
salts are in the top of the diagrams, and those of the BF4- salts
are at the bottom. An increase in the core size by the introduction
of an ester group between the two phenyl rings results in a
decreased range of SmA stability due to an increase in the
melting temperatures. In addition, all of the prepared esters
display polymorphism at low temperatures with up to three (soft)
crystal phases. A decreased core size obtained by omitting one
of the phenyl rings has a large effect on the mesomorphic
properties. For the propyl-substituted material, 23a (I- salt), the
clearing temperature drops significantly to 57 °C and the
corresponding BF4 salt does not exhibit liquid crystalline
properties. By increasing the tail length from propyl to dodecyl
24 (or trimethyldodecyl, see Supporting Information) stable
mesophases are obtained, but the clearing temperatures remain
much lower than those of 16.
-
The influence of lateral core substitutions is shown in Figure
10. The sterically undemanding methyl group has limited effect
on the phase stability, and small differences relative to the
unsubstituted mesogens 15 and 16 are observed. As the size of
the lateral group increases, the mesophase and crystal phase
stabilities are considerably decreased, and 28b displays a Tg of
-2 °C. In other cases, such as 29b, the suppression of
crystallization by the ethyl group allows for the formation of
other smectic mesophases. We also prepared larger n-pentyl and
phenyl side groups, but for these materials no mesomorphic
properties were observed. The latent heat values at the clearing
transitions and the XRD results of the BF4- salts are summarized
in Table 5.
Temperature-dependent XRD experiments (Figure 11) show
the layer spacing of the ILCs from series 3 and 4 as a function
of temperature. As expected, the observed layer spacings are
largest for the extended core material 20 and smallest for the
shortened 23. When compared to their molecular length (Table
4), the extent of interdigitation is largely mesogen independent.
9
J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007 14049

A R T I C L E S
Kouwer and Swager
Table 5. Latent Heat and XRD Data of Series 3 and 4
1
1
e
LC
core (Ar)
R
′
∆HCr
a/ kJ mol-
∆
HSmA b/ kJ mol-
Lcalcc/ Å
LXR d,e/ Å
LXRD/Lcalc
-SmA
-
Iso
15b
20b
23b
26b
28b
30b
32b
PhPh
PhCO2Ph
Ph
PhPh
PhPh
PhPh
PhPh
H
H
H
13.1
40.3
7.5
3.4
3.8
3.0
3.2
1.6
41.5
44.0
37.5
41.5
41.5
31.7
32.9
26.4
30.5
30.9
0.76
0.75
0.70
0.73
0.74
CH3
41.8
C2H5
n-C5H11
Ph
0.7f
not mesomorphic
not mesomorphic
a
Latent heat at the crystal to SmA transition (taken from the second heating trace). b Latent heat at the SmA to isotropic transition. c Length of the
stretched molecule, determined by molecular modeling. d Layer spacing, determined by XRD experiments, for monolayers LXRD ) d001, for bilayered materials
LXRD ) 1/2 × d001
e LXRD was determined at T ) 0.95TIso (after interpolation of the data). f ∆H at SmX to SmA transition.
.
Figure 13. Schematic representation of materials with one or two tails.
The monolayer SmA phases formed by the C12-substituted
mesogens are grouped together in the bottom of the plot. They
are characterized by interdigitations of 25-30% of the calculated
molecular length with relatively low-temperature dependence.
The citronellyl derivative 18b shows similar behavior, albeit
interdigitation is slightly reduced. The temperature dependence
of the bilayered systems, however, is significantly larger. The
presence of the Li+ ions in the mixtures of 19b with LiBF4
causes interdigitation to increase but also results in decreased
temperature dependence as compared to the other bilayer
structures.
Figure 11. Temperature dependence of the layer spacing of the monolayer
forming series 3 and 4 (dodecyl-substituted materials only).
Series 5: Influence of the Position of the Aliphatic Tails.
Considering that 15b and 34b are isomers, we anticipated
comparable mesomorphic properties. Similarly, 35b, in which
the dodecyl tail is replaced by the trimethyldodecyl tail, was
expected to show behavior comparable to that of 15b, the parent
material. However, the liquid crystalline properties for this series
of materials were found to be very different.
Figure 12. Temperature dependence of the layer spacing plotted against
the degree of interdigitation. The equations defining these quantities are
given in the text. For convenience, points are grouped together for the 19b/
LiBF4 mixture, 18b (C8-citronellyl tail), the bilayer SmA phases of 15a-d
and 19b and the C12-substituted monolayer SmA phases of 15e, 16b, 17b,
21b, 24b, 27b, and 29b.
We have investigated the effect of the substitution pattern
with one or two long (C12) tails on the mesogenic core and the
surfactant anion. As the cationic charge in the mesogenic core
creates highly asymmetric ILCs, the influence of the tail
substitution pattern on the mesophase behavior is substantial.
Figure 13 schematically depicts the materials investigated, and
their mesophase behavior is summarized in Figure 14.
Laterally substituted mesogens 26 and 28 display a similar layer
spacing and degree of interdigitation as the parent 16.
Mixtures of Ionic Liquid Crystals. Mixtures of liquid
crystals are widely used in commercial applications. This
approach eliminates the need for design and synthesis of a
mesogen with all desired properties (which possibly may not
exist). An additional advantage of mixtures is that crystallization
is often suppressed and mesophases can be stable at lower
temperatures. To investigate the mixing behavior of our newly
When comparing the XRD results between different series
of ILCs, we identified two important variables: the degree of
interdigitation, defined as 1 - {LXRD(T ) 0.95TIso)}/{Lcalc}, and
the temperature dependence of the layer spacing {LXRD(T )
0.90TIso) - LXRD(T ) 0.95TIso)}/{Lcalc}, determined over a small
temperature window (with T/K). Figure 12 shows all materials
for which XRD data was collected in a master plot.
9
14050 J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007

Synthesis of Rigid-Core Ionic Liquid Crystals
A R T I C L E S
Figure 14. Comparison of the mesomorphic properties of ILCs with one
and two (branched) C12 tails. The structures are represented schematically
in Figure 13.
prepared ILCs, we prepared binary and ternary mixtures (Table
6) that were investigated by DSC and OPM.
Contact samples of two different mesogens were prepared
by allowing two small drops of ILC in the isotropic phase to
mix under a cover slip. The interface between the two materials
was studied by OPM by varying the temperature. Contact
samples of structurally similar mesogens always showed
complete (linear) miscibility of the two components. Interesting
behavior was observed for the contact sample prepared from
15b and 34b (isomers with a different position of the propyl
tail). Both materials mixed homogeneously in the isotropic
phase, but the mixture did not permit mesophase formation. The
microscope image showed two different textures of the meso-
phases of both materials separated by an isotropic region as
shown in Figure 15a. Compound 34b showed a clear, focal,
conical texture, while 15b showed a grainy texture with a
tendency for homeotropic alignment in the bulk of the material.
The isotropic region did not disappear upon cooling to room
temperature. Similarly, the SmA phases of compounds 26b and
36b, both substituted with methyl groups on the tail and the
core, respectively, are immiscible in the SmA phase. The two
(different) textures are separated by an isotropic region (Figure
15b), that reduced in size when cooling the sample to room
temperature, but never disappeared. Annealing the structure (for
multiple days) gave rise to complete phase separation into the
two (smectic) components.
Figure 15. Contact samples of (a) 15b/34b at 125 °C and (b) 26b/35b at
80 °C (right) and 30 °C (left). Both pictures show immiscibility of the pairs
of SmA mesogens. Mesogens 34b and 35b show a focal conic texture;
15b and 26b a grainy texture or homeotropic on cooling. The dark
homeotropic SmA areas are labeled SmAh. This is not a separate phase
assignment but is used to distinguish the black texture from the isotropic
areas. The area in between the SmA phases is isotropic. (c) Mixture 20a/
23a at 60 °C: 20a is homeotropically aligned (bottom left) and at the
interface with 23a a focal conic texture appears. All pictures were taken
between crossed polarizers.
aligned homeotropically. On cooling, the SmA interface slowly
shifts toward the bulk of 23a. At the interface, a classical focal
conic texture appears. The picture in Figure 15c was recorded
just above the clearing temperature of 23a and further cooling
of this sample resulted in the formation of a focal conic texture
in the bulk of 23a.
Mesogens 20a and 23a (same substitution pattern, but
different core size) are an example of a completely miscible
system. In the contact sample (Figure 15c), the bulk of 20a is
Table 6. Composition and Phase Behavior of Mixtures of ILCs
phase behaviorb,c T/
°
C and
M /
g mol-
1
1
mixture
composition (molar fractions)
miscibilitya
(∆
H/ kJ mol-
)
I
II
15a (0.501)
15b (0.502)
26b (0.500)
20a (0.477)
16a (0.337)
20a (0.336)
20b (0.331)
34a (0.499)
34b (0.498)
35b (0.500)
23a (0.523)
21a (0.340)
31a (0.332)
24b (0.331)
-
i
i
i
m
m
m
m
n/a
n/a
n/a
n/a
n/a
n/a
593.6
717.0
667.2
577.9
-
III
IV
V
VI
VII
-
-
G
SmA 14 SmA 132 (0.4) Iso
Cr 127 (5.7) SmA 216 (2.6) Iso
SmA 6 SmA 164 (1.7) Iso
Cr 21 (broad) SmA 83 (0.7) Iso
24a (0.323)
35a (0.332)
34b (0.338)
G
b
a i ) immiscible, m ) miscible. GSmA ) glass with SmA phase frozen in, SmA ) smectic A, Iso ) isotropic. c To determine the latent heat in kJ mol-1
,
the average molecular weight calculated from the molecular weight of the constituents, weighted by the mol fraction x: M ) ∑iMixi.
9
J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007 14051

A R T I C L E S
Kouwer and Swager
Table 7. Observed Clearing Temperatures and Latent Heat,
Compared to Calculated Values
templating medium for the electrosynthesis of materials. Despite
the potential impact, this is the first extensive parameter study
to be performed on these types of mesogens.
a
a
TIso,exp
/
∆
HIso,exp
/
TIso
/
∆HIso /
kJ mol-
1
1
mixture
°
C
kJ mol-
°
C
Using newly developed aryl-amination chemistry, we have
prepared a series of ILCs wherein the charged (imidazolium)
group has been incorporated in the rigid part of the mesogen.
The increased core size, as anticipated, has a positive effect on
the order parameter of these new materials when compared to
simple long-chain 1-alkyl-3-methylimidazolium salts. This me-
sogenic core allowed us to conduct a full investigation of how
the structural parameters influence mesophase formation. In-
vestigated parameters included: (i) the size and nature (branch-
ing effects) of the aliphatic tails and the substitution of those
tails with corresponding ethylene glycol tails, (ii) the size of
the mesogenic core. (iii) lateral substitution on the mesogenic
core, and (iv) the nature of the counterion.
In summary, our results indicate that mesogens with a larger
core display more stable SmA phases (higher clearing temper-
atures), but also higher melting temperatures. The latter can be
reduced by introducing small laterally attached substituents.
When lateral substituents become too bulky, the mesogenic
properties are lost. Mesogens substituted with two long flexible
aliphatic tails form interdigitated SmAd phases. Branching of
the tails (either in racemic or chiral form) has minimal influence
on the phase behavior. Short tail substituents introduce the
formation of SmA2 phases (interdigitated bilayers). Mesogens
substituted with ethyleneglycol tails show deviant behavior
(reduced mesophase stability), due to the interaction of the tail
with the imidazolium cation. Doping these systems with LiBF4
stabilizes the mesophase and has the additional advantage that
the Li+ ions can improve rates of anisotropic ion conductivity.
Mixtures of different ILCs are completely miscible in some
cases, and ideal mixing behavior has been observed. The use
of such mixtures offers advantages over the use of pure
compounds by suppressing crystallization and giving low-
viscosity SmA phases, which is important for potential applica-
tions. In other ILC mixtures incompatibility of the two SmA
phases was observed, thereby indicating that additional structural
investigations of the mesophases are necessary.
IV
V
VI
VII
132
216
154
83
0.4
2.6
1.7
0.7
132
230b
151c
146
1.0
4.1b
2.4c
1.9
a To predict the transition temperatures and the latent heat, we used the
weighted average of the components: TIso ) ∑iTIso,ixi and ∆HIso
∑iHIso,ixi. b Estimated values for 16a: TIso ) 270 °C, ∆HIso ) 4 kJ mol-1
c Estimated values for 31a: TIso ) 50 °C, ∆HIso ) 0.3 kJ mol-1
)
.
.
Discrete binary and ternary mixtures (Table 6) were analyzed
by DSC. Immiscible samples (mixtures I-III) showed DSC
traces with multiple broad peaks. DSC traces of miscible
samples, as judged by OPM (mixtures IV-VII), displayed well-
defined SmA to isotropic transitions that are slightly broadened
with respect to the pure compounds. Some of the miscible
samples showed partial crystallization after extended periods
at room temperature.
Table 7 shows the clearing temperatures and corresponding
latent heat values of mixtures IV-VII, compared to the average
values (weighted by mole fractions) of the separate components.
The close correspondence of the clearing temperatures of
mixtures IV and V indicates nearly linear miscibility and thus,
close to ideal behavior. Observed values of the latent heat at
the clearing temperature are lower than the calculated average
values. This can be readily understood because a SmA phase
containing a mixture of mesogens of different dimensions would
be expected to show reduced order.
On the basis of OPM studies, we found that binary mixture
IV shows complete suppression of crystallization and the
viscosity of the sample was judged (qualitatively) to be lower
than that of the pure components. The sample could easily be
sheared in the SmA phase at 50 °C, a temperature 80 °C below
the clearing temperature.
We were surprised to find that mixture VI, which contains
the non-liquid crystalline 31a and is very similar to the
immiscible mixture III, displayed liquid crystal properties. In
fact, for mixture VI the experimental data is very close to the
average values of the components. Mixture VII with similar
components shows a much lower clearing temperature than
would be expected. Clearly, mixing ILCs can be successful in
creating new properties.
Acknowledgment. This work was supported by the National
Science Foundation and European Union through the Marie
Curie International Outgoing Fellowship of PHJK.
Supporting Information Available: Experimental details,
synthesis, structural characterization, liquid crystal properties
of all 47 final products and intermediates and molecular
modeling results. This material is available free of charge via
the Internet at http://pubs.acs.org.
Conclusions
Liquid crystals with ionic rigid cores constitute a relatively
new class of materials in which self-organization and macro-
scopic anisotropy are imparted to ionic liquids. The anisotropic
properties that these materials offer may find applications in
functional devices, such as photovoltaic fuel cells, or as a
JA075651A
9
14052 J. AM. CHEM. SOC. VOL. 129, NO. 45, 2007