Tyrosine-Based Ionic Liquid Crystals: Switching from a Smectic A to a Columnar Mesophase by Exchange of the Spherical Counterion

Article abstract of DOI:10.1002/chem.201602937

Synthetic strategies were developed to prepare l-tyrosine-based ionic liquid crystals with structural variations at the carboxylic and phenolic OH groups as well as the amino functionality. Salt metathesis additionally led to counterion variation. The liquid-crystalline properties were investigated by differential scanning calorimetry (DSC), polarizing optical microscopy (POM) and X-ray diffraction (WAXS, SAXS). The symmetrical ILC chlorides bearing the same alkyl chain at both the ester and ether but either an acyclic or cyclic guanidinium group displayed enantiotropic SmA₂mesophases with phase widths of 31–88 K irrespective of the head group. It was particularly the replacement of chloride in the acyclic guanidinium ILC by hexafluorophosphate that induced a phase change from SmA₂to Col_r. This phase change was attributed to a higher curvature of the interface due to the larger anion, which increased the effective head group cross-sectional area of the amphiphilic ILC. The unsymmetrical acyclic guanidinium chlorides, bearing a constant C₁₄ester and variable alkyl chains on the phenolic position, formed enantiotropic SmA₂phases. The derivative with the largest difference in chain lengths, however, displayed a Col_rphase, resulting from discoid aggregates of the cone-shaped guanidinium chloride. The results are discussed in terms of the packing parameters, which indicate that the phase behaviour of the thermotropic tyrosine-based ILCs shows analogies to those of lyotropic liquid crystals.

Full text of DOI:10.1002/chem.201602937

DOI: 10.1002/chem.201602937
Full Paper
&
Materials Chemistry
Tyrosine-Based Ionic Liquid Crystals: Switching from a Smectic A
to a Columnar Mesophase by Exchange of the Spherical
Counterion
Manuel M. Neidhardt,[a] Manpreet Wolfrum,[a] Stuart Beardsworth,[a] Tobias Wçhrle,[a]
Wolfgang Frey,[a] Angelika Baro,[a] Cosima Stubenrauch,[b] Frank Giesselmann,[b] and
Sabine Laschat*[a]
Abstract: Synthetic strategies were developed to prepare l-
tyrosine-based ionic liquid crystals with structural variations
at the carboxylic and phenolic OH groups as well as the
amino functionality. Salt metathesis additionally led to coun-
terion variation. The liquid-crystalline properties were inves-
tigated by differential scanning calorimetry (DSC), polarizing
optical microscopy (POM) and X-ray diffraction (WAXS,
SAXS). The symmetrical ILC chlorides bearing the same alkyl
chain at both the ester and ether but either an acyclic or
cyclic guanidinium group displayed enantiotropic SmA2
mesophases with phase widths of 31–88 K irrespective of
the head group. It was particularly the replacement of chlo-
ride in the acyclic guanidinium ILC by hexafluorophosphate
that induced a phase change from SmA2 to Colr. This phase
change was attributed to a higher curvature of the interface
due to the larger anion, which increased the effective head
group cross-sectional area of the amphiphilic ILC. The un-
symmetrical acyclic guanidinium chlorides, bearing a con-
stant C14 ester and variable alkyl chains on the phenolic posi-
tion, formed enantiotropic SmA2 phases. The derivative with
the largest difference in chain lengths, however, displayed
a Colr phase, resulting from discoid aggregates of the cone-
shaped guanidinium chloride. The results are discussed in
terms of the packing parameters, which indicate that the
phase behaviour of the thermotropic tyrosine-based ILCs
shows analogies to those of lyotropic liquid crystals.
Introduction
layer type (smectic), columnar or complex 3D (cubic) meso-
phases.[18–23] Both thermotropic and lyotropic amino acid-de-
rived liquid crystals are well known. The earliest reports date
back to the 1970s, when thermotropic amino acid derivatives
were used as chiral stationary phases for gas chromatogra-
phy.[24] More recent examples include metallomesogens with
amino acid ligands,[25] amino-acid-based functionalized nano-
particles,[26] dendrons,[27] discotic hexakis(phenylethynyl)ben-
zenes,[28] hexacatenar bisamides,[29–31] and furanone liquid crys-
tals.[32,33] On the other hand, the lyotropic behaviour of pep-
tides was discovered in 1977 for collagen[34] and glycoprotein
plant toxins.[35] Lyotropic phases for low molecular weight
amino-acid derivatives were first reported in 1980.[36,37] More
recent reports have dealt with amino-acid derivatives as dop-
ants for lyotropic chromonic liquid crystals[38] to study chiral
amplification, lyotropic reverse hexagonal liquid crystals,[39] lyo-
tropic dendrons,[27] cationic lipids for gene delivery[40] and to
modify self-assembled nomoglyceride-based nanostructures.[41]
However, amino-acid-derived ILCs have been much less ex-
plored. Among the few reported examples are lyotropic ILCs
consisting of imidazolium cations and amino acid anions.[42] Al-
ternatively, amino acids could also be used as cationic unit in
such charged liquid crystals.[43] Another example of thermo-
tropic amino acid-derived ILCs are chiral perylene diimides ob-
tained by ionic self-assembly.[44]
Task-specific ionic liquids derived from the chiral pool have
gained increased attention in recent years due to their broad
range of applications, including catalysis, asymmetric synthesis,
spectroscopy, chromatography, detoxification and antimicrobial
activity.[1–17] Amino acids, as naturally occurring chiral com-
pounds, represent readily available, prominent building blocks
for the development of task-specific ionic liquids. Their scope
can be significantly broadened if the ionic liquid properties,
such as adjustable polarity, solubility, and viscosity, as well as
fluidity and low vapour pressure can be combined with the
anisotropic properties of liquid crystals, resulting in functional
ionic liquid crystals (ILCs). The nanosegregation of immiscible
cationic head groups and side chains and the Coulomb inter-
action of counterions lead to supramolecular self-assembly in
[a] Dr. M. M. Neidhardt, M. Wolfrum, S. Beardsworth, T. Wçhrle, Dr. W. Frey,
Dr. A. Baro, Prof. Dr. S. Laschat
Institut fꢀr Organische Chemie, Universitꢁt Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
E-mail: sabine.laschat@oc.uni-stuttgart.de
[b] Prof. Dr. C. Stubenrauch, Prof. Dr. F. Giesselmann
Institut fꢀr Physikalische Chemie, Universitꢁt Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
Supporting information and ORCID from the author for this article are
available on the WWW under http://dx.doi.org/10.1002/chem.201602937.
Chem. Eur. J. 2016, 22, 1 – 12
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Results and Discussion
Synthesis of the guanidinium tyrosinates
The synthetic route to tyrosine derivatives 3 and 4 with side
chains of the same length at the carboxylic and phenolic OH
group is shown in Scheme 2. Known starting material N-Boc-l-
tyrosine 7, prepared in 97% following
a procedure by
Rosen,[54] was submitted to a simultaneous esterification and
Williamson etherification with alkyl bromides in the presence
of K2CO3 and NaI in refluxing MeCN, providing the N-Boc-tyro-
sine derivatives 8a–g in 28–89% yield. Subsequent cleavage of
the Boc protecting group with trifluoroacetic acid (TFA) gave
the corresponding amines 9a–g in good yields of 79–98%,
which were treated with either N,N,N’,N’-tetramethylchloro-
formamidinium chloride 6a or its cyclic counterpart 6b[55] in
the presence of NaHCO3 to yield the desired acyclic and cyclic
guanidinium chlorides 3(n,n)Cl and 4(n,n)Cl, respectively. Salt
metathesis of chloride 3(14,14)Cl in MeCN gave the guanidini-
um salts 3(14,14)X (X=Br, I, OTf, BF4, PF6) quantitatively
(Scheme 2).
Scheme 1. Previous work on imidazolium ILCs 2 based on natural amino
acids[45] and our envisioned synthetic strategy to double-tailed ILCs 3 and 4
starting from l-tyrosine 5.
We recently reported imidazolium ILCs 2, which were pre-
pared from amino acids 1 (Scheme 1). These ILCs formed SmA
mesophases with phase widths up to 70 K and clearing points
below 1208C,[45] but the synthesis was strenuous and had
a poor overall yield.
Furthermore, the scope was strictly limited to glycine, ala-
nine and phenylalanine derivatives and some ILCs suffered
from thermal decomposition despite their relatively low clear-
ing points. Thus, there is a need for thermotropic amino-acid-
derived ILCs that would allow for greater flexibility of the
design parameters whilst keeping the synthesis as short and
efficient as possible. We anticipated that l-tyrosine-derived
guanidinium ILCs 3(m,n)X, 4(m,n)X might be suitable candi-
dates for such purposes because they should require fewer
synthetic steps and provide four different sites for structural
variations: the alkyl chain at the carboxylic and phenolic OH
group, respectively, the counterion, and the amino functionali-
ty to attach the ionic head group (Scheme 1). The guanidinium
moiety was chosen because of its already shown beneficial
effect on ILCs.[46,47] In addition, the double-tailed molecular
structure of the envisaged compounds 3(m,n)X and 4(m,n)X
with a cationic head group resembles phospholipids such as
1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), which form
liposomes useful for drug and gene delivery.[48–51] In particular,
peptide-based cationic liposomes have shown many advantag-
es over other cationic liposomes.[52,53] As amino-acid-derived
ILCs that mimic such liposome building blocks seem to be at-
tractive as target species, we synthesized two series of guanidi-
nium tyrosine-based compounds 3(m,n)X and 4(m,n)X and an-
alysed them with respect to their thermotropic mesomor-
phism.
Scheme 2. Synthesis of the symmetrical guanidinium tyrosinates 3(n,n)Cl
and 4(n,n)Cl and subsequent salt metathesis to 3(14,14)X.
Our initial synthetic strategy to introduce side chains of dif-
ferent length on the ester and ether moiety is outlined in
Scheme 3. The synthesis commenced with the esterification of
N-Boc-l-tyrosine 7 with acetyl chloride in MeOH[56] to give the
known methyl ester 10 in 71% (Scheme 3). A Williamson ether-
ification with bromotetradecane led to derivative 11 in 48%
yield. After saponification,[56] the resulting carboxylic acid 12
was esterified with bromooctane to provide the N-Boc-tyrosine
derivative 13(8,14) in 95% yield. The latter was deprotected as
described above to the corresponding amine 14(8,14) in an
overall yield of 27% starting from 7.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 4. Alternative route to unsymmetrical guanidinium chlorides
3(14,n)Cl starting from l-tyrosine 5.
Scheme 3. Initial synthetic route to unsymmetrical tyrosine derivative
14(8,14).
exceeds the molecular length by a factor of 1.26 indicating the
degree of interdigitation. The lateral distances of the atoms of
the interdigitated alkyl layers ranged between 3.8 and 4.0 ꢁ.
Although this synthetic sequence worked quite well, it still
required six steps to amine 14. An alternative route was there-
fore developed starting directly with an esterification of l-tyro-
sine 5 using tetradecanol and octadecanol (m=14, 18), respec-
tively, in the presence of an excess of p-TsOH (Scheme 4).[57]
The resulting esters 15a[57] and 15b were protected to the re-
spective tyrosinates 16a,b with Boc2O under basic conditions
using NaHCO3 in a water/dioxane mixture. Williamson etherifi-
cation of 16 with the appropriate 1-bromoalkane in MeCN and
subsequent cleavage of the Boc protecting group with TFA in
CH2Cl2 as described above provided the desired amines
14(14,n)Cl in 48–90% yield over both steps. Following this
route, amines 14 were isolated in 41–77% overall yield starting
from 5 (Scheme 4). Final treatment with N,N,N’,N’-tetramethyl-
chloroformamidinium chloride 6a under basic conditions using
NaHCO3 in CH2Cl2 gave the target tyrosine chlorides 3(14,n)Cl
(Scheme 4). In the case of 16b, decomposition was observed
in the follow-up reaction to 14 upon deprotection with TFA
and subsequent neutralization.
Mesomorphic properties of symmetrical guanidinium salts
3(n,n)X and 4(n,n)Cl
The mesomorphic properties of guanidinium ILCs 3(n,n)X and
4(n,n)Cl were first investigated using differential scanning calo-
rimetry (DSC). Typical DSC curves of both types of symmetrical
tyrosine derivatives are shown for compound 3(14,14)Cl and
4(16,16)Cl in Figure 2.
In the second heating cycle of 3(14,14)Cl an endothermic
crystal-to-crystal transition at 588C, a melting transition at
618C, and a clearing transition at 1098C was visible (Figure 2a).
Upon second cooling from the isotropic phase an exothermic
peak was detected at 1118C, whereas the mesophase-to-crystal
transition was considerably shifted due to a supercooling
effect, thus appearing at 458C, and the crystal-to-crystal transi-
tion at 408C (Figure 2a). In the case of 4(16,16)Cl, a cold crys-
tallization at 478C followed by melting and clearing transitions
at 758C and 1048C, respectively, were observed upon second
heating (Figure 2b). In the second cooling cycle, the isotropic-
to-mesophase transition temperature slightly deviated from
that upon heating. However, as already found for 3(14,14)Cl,
the mesophase-to-crystal transition was again shifted and
could be detected at 528C (Figure 2b). Further heating and
cooling cycles of both derivatives did not lead to any changes
of the curves (Figure 2). The other chlorides 3(n,n)Cl and
4(n,n)Cl behaved similarly (Figures S23–S31 in the Supporting
Information). The results of the DSC measurements of 3(n,n)Cl
and 4(n,n)Cl are summarized in Table 1 and Figure 3.
Single crystals of derivative 16b that were suitable for X-ray
crystal structure analysis (Figure 1) were obtained.[58] In the
solid state, tyrosinate 16b crystallized as a pure enantiomer
with one molecule in the asymmetric unit of the acentric
space group P212121 and with a remarkable long c-axis of
about 72 ꢁ. The stereogenic centre at C2 was identified to be
(S)-configured. Both the amino functions N1ꢀH1 and the phe-
nolic p-hydroxy group O3ꢀH3 act as donors building intermo-
lecular hydrogen bonds with the oxo functions O4 and O1 as
acceptors. The bc-view of the packing diagram of 16b shows
a layer-type stacking of alternate tyrosine head groups and
alkyl chains along the c-axis (Figure 1b,c). The Boc tert-butyl
moieties form a bilayer by nonpolar interactions. Along the b-
axis an antiparallel, interdigitated stacking of the alkyl chains is
evident. The total length (O3···C32) of 16b is 30.37 ꢁ. Accord-
ing to Figure 1b, the alternating packing of two molecules re-
sulted in a bilayer arrangement. The bilayer distance of 38.35 ꢁ
The symmetrical tyrosine chlorides 3(n,n)Cl with acyclic tet-
ramethylguanidinium head group formed an enantiotropic
mesophase (Table 1, entries 1–7). In the case of chlorides
3(6,6)Cl and 3(8,8)Cl, the phase transitions were determined
only by polarizing optical microscopy (POM) (entries 1, 2). As
can be seen from Table 1 and Figure 3a, in the series of chlo-
rides 3(n,n)Cl the melting points increased steadily with in-
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 2. a) DSC curves of symmetrical ILC 3(14,14)Cl with acyclic guanidini-
um head group and b) DSC curves of symmetrical ILC 4(16,16)Cl with cyclic
guanidinium head group (heating/cooling rate 10 Kminꢀ1; H=heating,
C=cooling).
Figure 1. a) X-ray structure of precursor 16b (m=18) in the solid state
(atomic numbering only for X-ray diffraction studies). Cell parameters
a=5.0920(3) ꢁ, b=8.7874(5) ꢁ, c=71.955(4) ꢁ. The H3···O4 distance is
2.11(3) ꢁ with an N1ꢀH1···O4 angle of 162(2)8. The H3···O1 distance is
1.96(3) ꢁ and the corresponding angle O3ꢀH3···O1 shows with 169(3)8 also
a strong, nearly linear interaction. b,c) Packing diagram of 16b showing the
interdigitation of the octadecyl chains (H atoms omitted for clarity).
creasing chain lengths except 3(6,6)Cl and 3(10,10)Cl, whereas
no comparable trend was visible for the clearing temperatures.
For ILCs 3(n,n)Cl (n=6, 8, 10) clearing points ranged between
79–958C before reaching a plateau of approximately 1108C
(n=12, 14, 16) and finally decreased to 1028C for 3(18,18)Cl.
Chloride 3(12,12)Cl displayed the broadest mesophase (DT=
85 K) followed by 3(8,8)Cl (DT=75 K).
Figure 3. Mesophase ranges for the acyclic guanidinium chlorides
3(n,n)Cl (a) and cyclic guanidinium chlorides 4(n,n)Cl (b) upon second cool-
ing. Mesophase widths are given in K.
The corresponding symmetrical guanidinium chlorides
4(n,n)Cl (n=12, 14, 16, 18) with cyclic guanidinium head
group also showed enantiotropic mesomorphism (Table 1, en-
tries 8–12). The phase transition of 4(10,10)Cl could be deter-
mined only by POM (Table 1, entry 8). Both melting and clear-
ing points increased with increasing chain length for the series
4(n,n)Cl except for the melting point of 4(10,10)Cl and the
clearing point of 4(18,18)Cl. The mesophase widths showed
a maximum value of 67 K for 4(14,14)Cl (Figure 3b).
Under POM all symmetrical guanidinium chlorides 3(n,n)Cl
showed a strong tendency for homeotropic alignment with
Maltese crosses and small areas of fan-shaped textures, sug-
gesting a smectic A mesophase. In Figure 4a the texture of
3(14,14)Cl is given as a typical example for ILCs 3 with acyclic
head group. ILCs 4(n,n)Cl with cyclic head group also tend to
homeotropic alignment and showed similar textures. For ex-
ample, Maltese crosses were observed for 4(18,18)Cl at 908C
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 1. Phase transition temperatures (T) and enthalpies (DH) of acyclic
and cyclic guanidinium tyrosine ionic liquid crystals 3(n,n)X and 4(n,n)Cl
determined by DSC upon second cooling for better comparison.[a,b]
Entry Cmpd
Phase Tm [8C] (DH
Phase Tc [8C] (DH
Phase
[kJmolꢀ1])
[kJmolꢀ1])
1
2
3
4
5
6
7
8
3(6,6)Cl[c]
G
G
G
30 (Tg)
20 (Tg)
SmA2 83
SmA2 95
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3(8,8)Cl[c]
3(10,10)Cl
30 (0.8)
27 (29.0)
45 (38.2)
56 (46.1)
71 (56.1)
24 (Tg)
18 (7.2)
30 (11.1)
52 (38.6)
57 (36.1)
35 (8.3)
64 (24.6)
19 (17.8)
31 (10.0)
24 (14.1)
SmA2 79 (1.5)
SmA2 112 (1.9)
SmA2 111 (0.9)
SmA2 114 (1.0)
SmA2 102 (1.0)
SmA2 58
SmA2 81 (0.9)
SmA2 97 (1.0)
SmA2 104 (0.8)
SmA2 98 (1.0)
SmA2 90 (1.2)
SmA2 93 (0.6)
SmA2 32 (1.0)
SmA2 87 (1.1)
Colr 78 (1.2)
3(12,12)Cl Cr
3(14,14)Cl Cr
3(16,16)Cl Cr
3(18,18)Cl Cr
Figure 5. Mesophase ranges for the acyclic guanidinium ILCs 3(14,14)X.
Mesophase widths are given in K.
4(10,10)Cl[c]
G
9
4(12,12)Cl Cr
4(14,14)Cl Cr
4(16,16)Cl Cr
4(18,18)Cl Cr
3(14,14)Br Cr
10
11
12
13
14
15
16
17
3(14,14)I
Cr
3(14,14)OTf Cr
3(14,14)BF4 Cr
3(14,14)PF6 Cr
[a] Heating/cooling rate 10 Kminꢀ1; melting point (Tm), clearing point (Tc),
glass transition temperature (Tg). [b] Observed phases: glass G, crystalline
Cr, smectic bilayer SmA2, columnar rectangular Colr, isotropic I. [c] Phase
transition could not be detected by DSC; transition temperatures were
measured with POM.
Figure 6. POM images of bromide 3(14,14)Br at 758C (a) and 3(14,14)PF6 at
558C (b) as seen between crossed polarizers upon cooling from the isotropic
liquid (magnification ꢂ100; cooling rate 10 Kminꢀ1).
(SAXS) and wide-angle regions (WAXS) were carried out. The
experimentally observed d values, lattice spacings and indices
are summarized in Table 2.
The diffraction patterns of all guanidinium chlorides 3(n,n)Cl
and 4(n,n)Cl as well as of the salts 3(14,14)X with different
counterions (X=Br, I, BF4, OTf) displayed a sharp reflection
(001) in the small-angle region and a diffuse wide-angle halo
between 4.0 and 4.7 ꢁ, which is characteristic for a smectic
mesophase.[65] Figure 7 shows the diffractogram of chloride
3(14,14)Cl as a typical example.
Figure 4. POM images of 3(14,14)Cl at 908C (a) and 4(18,18)Cl at 908C (b)
as seen between crossed polarizers upon cooling from the isotropic liquid
(magnificationꢂ100; cooling rate 10 Kminꢀ1).
upon cooling from the isotropic liquid (Figure 4b) indicating
the formation of a SmA mesophase.
The variation of the counterion is reported to have a pro-
found effect on the mesomorphism of ILCs.[47,59–62] We thus in-
vestigated the symmetrical acyclic guanidinium salts 3(14,14)X
with different counterions by DSC in comparison with the chlo-
ride 3(14,14)Cl (Table 1, entries 5, 13–17, Figure 5). For all
counterions (X=Cl, Br, I, BF4, PF6, OTf) enantiotropic mesomor-
phism was detected. Except for the iodide 3(14,14)I, a decrease
of both melting and clearing points with increasing ionic radii
was observed. In the series of halides, 3(14,14)I displayed the
smallest mesophase width of 29 K. Moreover, the mesophase
ranges of 3(14,14)Br and 3(14,14)BF4 were similar, whereas
the one of the triflate 3(14,14)OTf was only 13 K.[63]
When the salts 3(14,14)X (X=Cl, Br, I, BF4, OTf, PF6) were
studied by POM, either Maltese crosses or fan-like textures
were observed irrespective of the counterion, as exemplarily
shown for 3(14,14)Br and 3(14,14)PF6 in Figure 6.
For a more conclusive characterization of the mesophase ge-
ometries, X-ray diffraction (XRD) measurements in the small-
Figure 7. X-ray scattering profile of 3(14,14)Cl at 808C after cooling from
the isotropic liquid (cooling rate 10 Kminꢀ1). Inset: the corresponding dif-
fraction image.
The temperature-dependent layer spacings d of the smectic
mesophases of symmetrical guanidinium chlorides 3(n,n)Cl
and 4(n,n)Cl with acyclic and cyclic head group, respectively,
were determined using the distinct layer reflection (001) in the
small-angle region. For each member in both series a decrease
of the layer spacing was observed with increasing tempera-
tures, which is typical for SmA phases (Figure 9).[66]
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 2. X-ray diffraction data for the symmetrical ILCs 3(n,n)X and
4(n,n)Cl.
Entry Cmpd
Mesophase Lattice Layer
Miller
Z
T [8C]
[ꢁ] spacings [ꢁ] indices
1
3(10,10)Cl SmA2
30.4
4.6
31.4
4.0
33.6
4.4
36.6
4.6
37.5
4.7
29.0
4.3
33.9
4.6
32.9
4.5
35.9
4.7
39.6
4.7
37.8
4.7
38.0
4.3
34.7
4.6
36.3
4.7
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(001)
(halo)
(01)
68
60
80
80
88
47
72
85
89
88
88
84
31
76
2
3(12,12)Cl SmA2
3(14,14)Cl SmA2
3(16,16)Cl SmA2
3(18,18)Cl SmA2
4(10,10)Cl SmA2
4(12,12)Cl SmA2
4(14,14)Cl SmA2
4(16,16)Cl SmA2
4(18,18)Cl SmA2
3(14,14)Br SmA2
3
4
5
6
7
8
9
Figure 8. Molecular lengths of symmetrical guanidinium chlorides 3(n,n)Cl
calculated with ChemDraw 3D[67] as exemplified for 3(10,10)Cl and proposed
bilayer packing.
10
11
12
13
14
15
3(14,14)I
SmA2
cal layer spacings (Figure 9a). Within the series of cyclic guani-
dinium chlorides 4(n,n)Cl, again a continuous increase of the
layer spacing with respect to increasing side chain length
could be found (Figure 9b).
3(14,14)OTf SmA2
3(14,14)BF4 SmA2
However, 4(14,14)Cl with C14 side chain displayed smaller
layer spacings than the shorter chloride 4(12,12)Cl. To explain
this finding, the orientational order parameter S, a measure for
the quality of orientational ordering, must be considered. We
assume that the smectic mesophase of ILCs 3(10,10)Cl and
3(14,14)PF6 Colr
a=49.5 57.8
b=57.8 37.6
10 70
p2mm
(11)
19.0 (18.8)[a] (22)
14.6 (14.4)[a] (40)
4.6 (halo)
[a] Calculated values using the software by Godbert[64] in parenthesis.
To rationalize the mesophase formation, smectic layer spac-
ings d from SAXS experiments were compared with the fully
stretched all-trans molecular lengths Lcalcd obtained by using an
energy minimization based on the MM2 basis set (Chem-
Draw 3D).[67] The ester and benzyl units differ in length despite
having the same alkyl chain lengths. The longer benzyl unit
was used for the calculation because this limits the maximum
possible interdigitation (Figure 8). The resulting ratios d001/Lcalcd
ranged between 1.1–1.3 (Supporting Information, Table S1).
For example, Lcalcd of 3(10,10)Cl is roughly 23 ꢁ, leading to
a ratio d001/Lcalcd of 1.3. Considering the length of a second
head group (4.5 ꢁ) a molecular length of around 28 ꢁ and
a ratio d001/Lcalcd of 1.1 was obtained. Thus, a smectic A bilayer
(SmA2) arrangement with interdigitated stacking of molecules
in a similar manner as seen in the single crystal structure of
precursor 16b appears to be reasonable (Figure 8).
Regarding the relative effects of the side chain length on
the temperature-dependent layer spacing, clear differences
emerge between the series. Whereas a continuous increase of
the layer spacing with increasing side chain lengths was de-
tected for the acyclic compounds 3(12,12)Cl up to 3(18,18)Cl,
derivatives 3(n,n)Cl with C10 and C12 side chains showed identi-
Figure 9. Temperature-dependent layer spacings in the SmA2 phases of:
a) acyclic guanidinium chlorides 3(n,n)Cl, and b) cyclic guanidinium chlorides
4(n,n)Cl with side chain lengths of n=10, 12, 14, 16, 18.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
4(12,12)Cl has a higher S value than the mesophases of the
other chlorides. Alternatively, a small deviation during calibra-
tion might cause this reversed order. The similar d spacings of
3(10,10)Cl and 3(12,12)Cl agree with earlier work by Berger-
on.[68] As is the case for the thickness of surfactant monolayers
at the water–air interface the layer distance remains constant
for short alkyl chains. Upon increase of the chain lengths the
degree of interdigitation increases while the layer spacing re-
mains constant, until a certain chain length at which the maxi-
mum degree of interdigitation is reached and the d values in-
crease continuously by 2.5 ꢁ per C2 unit.[69]
The experimental values of the layer spacing d for the com-
pounds with the halide counterions 3(14,14)Cl, 3(14,14)Br and
3(14,14)I were 33.6, 37.8, and 38.0 ꢁ, respectively (Table 2, en-
tries 3, 11, 12). As expected, the layer spacing increased with
increasing ion radius of the halide counterion. In case of tyrosi-
nate 3(14,14)BF4 with complex tetrafluoroborate anion, a layer
spacing comparable to the corresponding bromide 3(14,14)Br
or iodide 3(14,14)I would be expected due to the larger anion
size. However, the opposite was observed (Table 2, entry 14).
This result might be rationalized as follows. In previous studies
on guanidinium salts we observed that hydrogen bonds be-
tween the NH of the guanidinium moiety and the counterion
are beneficial for stabilization of the bilayer arrangement.[47]
Presumably due to the stronger hydrogen bond acceptor tet-
rafluoroborate as compared to bromide a tighter packing of
the ionic sublayer (consisting of cationic head groups and
anions) within the SmA2 bilayer is possible resulting in lower
layer spacings. Furthermore, despite the pseudo spherical
charge distribution of tetrafluoroborate, its tetrahedral geome-
try might also contribute to more efficient packing in the
SmA2 phase as compared to the spherical bromide.
Figure 10. a) X-ray diffraction pattern of a sample of 3(14,14)PF6 which was
partially oriented by fibre extrusion at 708C upon cooling from the isotropic
liquid. Inset: the corresponding diffraction image. b) Columnar rectangular
mesophase with plane group p2mm. A disc consists of ten self-assembled
molecules (schematic illustration).
halides, which cannot be accommodated in a lamellar organi-
zation of the SmA mesophase but rather prefers the formation
of a disc-like aggregate, leading to a columnar packing. To ex-
plain the mesophase type change of ILCs 3(14,14)X, we pro-
pose a similar packing mechanism as compared to micelle for-
mation in lyotropic liquid crystals. For this purpose, we applied
the packing model by Israelachvili[72] for hydrocarbon amphi-
philes and by Tschierske[73] for lyotropic liquid crystals to ILCs 3
(Figure 11).
In contrast to ILCs 3(14,14)X (X=Cl, Br, I, BF4, OTf), the
small-angle X-ray scattering profile of the mesophase of hexa-
fluorophosphate 3(14,14)PF6 revealed a set of clearly visible re-
flections upon cooling from the isotropic liquid (Figure 10a),
indicating a columnar phase geometry. Analysis of the diffrac-
tion data using a software developed by Godbert and co-work-
ers[64] resulted in a columnar rectangular (Colr) mesophase with
p2mm symmetry and lattice parameters a=49.5 ꢁ and b=
57.8 ꢁ (Table 2, entry 15). With a density of 1=1.03 gcmꢀ3 the
number of molecules per unit cell Z for the columnar meso-
phase was calculated to be Z=10.[64] To rationalize the Colr
phase, we assume that ten molecules self-assemble into dis-
coid aggregates stacking in columns which organize into a rec-
tangular–columnar lattice (Figure 10b).
Figure 11. Schematic illustration of the mesophase formation of ILCs
3(14,14)X for: a) X=chloride, b) X=iodide, and c) X=hexafluorophosphate.
Increasing anion radii decrease the packing parameter and thus lead to
a transition from planar bilayers via postulated undulated smectic bilayers to
cylindrical aggregates.[72,74] Guanidinium cation: grey spheres, anion: orange
spheres; phenyl: grey ovoids.
The experimentally observed lattice parameters a and b are
in good agreement with a calculated molecular length of
around 28 ꢁ for 3(14,14)Cl (Supporting Information, Table S1)
and an estimated diameter of the aggregate of approximately
56 ꢁ. This behaviour is in contrast to other calamitic guanidini-
um ILCs for which anion exchange to hexafluorophosphate re-
sulted in a considerable reduction[70] or complete loss of meso-
morphism.[71]
When the effective area of the head group and the halide
counterion is similar to the effective area of the core unit and
hydrophobic alkyl chains, a non-curved planar aggregate and
thus a SmA2 phase is formed (Figure 11a). With increasing size
of the anion the curvature of the interface changes, resulting
The change of mesophase type from SmA to Colr in the case
of 3(14,14)PF6 might be due to the larger ion radius of the
spherical PF6 as compared to the smaller tetrafluoroborate or
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
first in postulated undulated smectic bilayers (Figure 11b).
Such a packing model would also explain why anions such as
tetrafluoroborate and triflate with larger anionic radii com-
pared to bromide and iodide have smaller layer spacing values
than the corresponding halides. Increasing the anion size even
further finally leads to a strongly curved interface and thus the
formation of cylindrical aggregates, that is, a columnar meso-
phase (Figure 11c).[73] It should be noted that Yoshizawa pro-
posed a related packing model to explain the experimentally
observed lamellar–columnar–micellar cubic (SmA–Colh–CubI)
phase sequence for amphiphilic liquid crystalline oligomers.[75]
Another related case was reported by Ohno for amphiphilic 4-
acylpyridinium zwitterions.[74] Depending on the lithium salt
added, layered phase structures were detected for hard anions
(e.g. chloride) and the resulting cylindrically shaped mesogen
salt complexes. For large, soft anions (e.g. bis(trifluoromethane-
sulfonyl)imide), bicontinuous cubic phases were observed, re-
sulting in circular cone-shaped mesogen salt complexes.
Cheng studied the self-assembly of rod-like imidazolium ILCs
which showed a transition from lamellar to micellar organiza-
tion by variation of the spacers as well as the lengths and
number of side chains while keeping the head group and
counterion constant.[76]
Figure 12. Mesophase ranges of the unsymmetrical guanidinium chlorides
3(14,n)Cl. Mesophase width is given in K.
Under the POM on plain-glass slides all chlorides 3(14,n)Cl
showed homeotropic textures indicating the presence of
a SmA mesophase. By using TMS-coated (TMS=tetramethylsi-
lane) glass plates, on which strong interactions between the
polar glass surface and the ionic guanidinium group is hin-
dered, homeotropic alignment could be effectively eliminated
and enabled the observation of Maltese cross textures as ex-
emplarily shown for ILC 3(14,16)Cl and fan-shaped textures for
3(14,8)Cl (Figure 13).
Mesomorphic properties of unsymmetrical ILCs 3(m,n)Cl
(m¼n)
To investigate the effect of unsymmetrical core substitution on
the mesomorphic properties, the unsymmetrical ionic liquid
crystals 3(14,n)Cl with the acyclic guanidinium head group
(Scheme 4) were investigated by DSC, POM and XRD measure-
ments. The DSC results are summarized in Table 3 and depict-
Figure 13. Maltese cross textures of 3(14,16)Cl at 1108C (a) and fan-shaped
textures of 3(14,8)Cl at 988C (b) on TMS-coated glass plates as seen be-
tween crossed polarizers upon cooling from the isotropic liquid (magnifica-
tionꢂ100; cooling rate 10 Kminꢀ1).
Table 3. Phase transition temperatures (T) and enthalpies (DH) of acyclic
guanidinium tyrosine ionic liquid crystals 3(14,n)Cl determined by DSC
upon second cooling.[a,b]
Entry Cmpd
Phase Tm [8C] (DH
Phase Tc [8C] (DH
Phase
[kJmolꢀ1])
[kJmolꢀ1])
With longer chains on both the ester and ether, hydrophobic
interactions between interdigitating alkyl chains become more
pronounced, thus driving nanosegregation so that chain
length differences have only a little effect on the mesophase
ranges.
1
2
3
4
5
3(14,8)Cl Cr
3(14,10)Cl Cr
3(14,12)Cl Cr
3(14,16)Cl Cr
3(14,18)Cl Cr
26 (1.3)
Colr 71 (1.7)
SmA2 87 (0.6)
SmA2 106 (1.1)
SmA2 105 (1.3)
SmA2 109 (1.1)
I
I
I
I
I
36 (2.6)
36 (6.3)
36 (23.3)
39 (2.0)
[a] Heating/cooling rate 10 Kminꢀ1; melting point (Tm), clearing point (Tc).
[b] Observed phases: crystalline Cr, smectic bilayer SmA2, columnar rec-
tangular Colr, isotropic I.
The results of XRD measurements of unsymmetrical guanidi-
nium ILCs 3(14,n)Cl are summarized in Table 4. In contrast to
chlorides 3(14,n)Cl (n=12, 16, 18), in the small-angle X-ray
scattering profile of 3(14,8)Cl a set of reflections was observed
that were indexed as (01), (10), (20), (33), and (40) indicating
the presence of a columnar mesophase (Figure 14a).
Analysis of the diffraction data[64] of 3(14,8)Cl led to a Colr
mesophase with p2mm symmetry and the number of mole-
cules per unit cell was calculated to be Z=5 (Table 3, entry 1
and Figure 14b).
ed in Figure 12. In the series 3(14,n)Cl, alkyl chain lengths nꢁ
12 on the phenolic ether had only a small effect on both the
melting and clearing transitions, resulting in similar phase
widths of 66–70 K (Table 3, entries 3–5, Figure 12). For ILC
3(14,10)Cl, the melting point was less influenced, but the
clearing transition decreased, leading to a mesophase width of
51 K (Figure 12). ILC 3(14,8)Cl, however, formed a columnar
(Colr) mesophase with a 45 K phase width.
From the XRD data of symmetrical tyrosine chlorides
3(n,n)Cl and the unsymmetrical counterparts 3(14,n)Cl
(Tables 2 and 4) it becomes clear that the layer distance in
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Considering the different ionic radii of hexafluorophosphate
and chloride, a self-assembled discoid aggregate as proposed
for the symmetrical 3(14,14)PF6 doesn’t seem reasonable for
the unsymmetrical 3(14,8)Cl. However, we argue that in the
case of 3(14,8)Cl, the effective head group cross-sectional area
is much smaller than the rest of the molecule that is why an
inversed cone shape is proposed for 3(14,8)Cl. Thus, five mole-
cules self-assemble into a discoid aggregate and these aggre-
gates then organize into a rectangular lattice (Figure 14).
Table 4. X-ray diffraction data for the unsymmetrical ILCs 3(14,n)Cl.
Entry Cmpd
Mesophase Lattice [ꢁ] Layer
Miller
Z
5
T [8C]
spacings [ꢁ] indices
1
3(14,8)Cl Colr
a=30.7
b=39.8
40.0 (39.8)[a] (01)
49
p2mm
30.8 (30.7)[a] (10)
15.4 (15.4)[a] (20)
8.1 (8.1)[a]
7.7 (7.7)[a]
4.4
4.1
33.0
4.2
36.2
4.0
36.3
(33)
(40)
(halo)
(001)
(001)
(halo)
(001)
(halo)
(001)
(halo)
2
3
4
3(14,12)Cl SmA2
3(14,16)Cl SmA2
3(14,18)Cl SmA2
78
88
89
Conclusions
Effective routes have been developed that provide access to l-
tyrosine-derived symmetrical ILCs 3(n,n)X and 4(n,n)Cl with
acyclic and cyclic guanidinium head groups, respectively, bear-
ing the same alkyl chain length on ester and ether moiety, as
well as unsymmetrical ILCs 3(m,n)Cl which bear alkyl side
chains of different length. Chain length variation for symmetri-
cal ILCs 3(n,n)Cl and 4(n,n)Cl as well as guanidinium head
group variation had some impact on mesophase stability and
temperature range but did not affect the mesophase type.
From the XRD data of 3(n,n)Cl and 4(n,n)Cl bilayered SmA2
mesophases with interdigitated alkyl chains were found. How-
ever, the most pronounced effect on the mesomorphic proper-
ties was obtained by salt metathesis of chloride 3(14,14)Cl,
which led to ILCs 3(14,14)X (X=halide, OTf, BF4, PF6). Symmet-
rical guanidinium halides 3(14,14)Br and 3(14,14)I also dis-
played SmA2 phases with a bilayer structure. The mesophase
widths decreased in the halide series with increasing ion radius
from 66 K for chloride to 29 K for iodide. No such size depend-
ence was observed for non-halide counterions. Mesophase
widths of approximately 55 K were found for BF4 and PF6 salts,
whereas the triflate salt had a phase range of only 13 K. Al-
though ILCs 3(14,14)BF4 and 3(14,14)OTf also showed SmA2
bilayers, albeit with smaller layer spacing values compared to
the halides, ILC 3(14,14)PF6 surprisingly formed a Colr meso-
phase. This phase change from SmA2 to Colr via a postulated
undulated SmA phase could be explained by an increased cur-
vature of the interface caused by the large anion which in-
creases the effective head group size in a similar fashion as ob-
served for lyotropic liquid crystals. In contrast, for unsymmetri-
cal chlorides 3(14,n)Cl broad mesophases of 45–70 K were ob-
tained that were less affected by the phenolic alkyl substituent.
The Colr phase of ILC 3(14,8)Cl was rationalized by an inverse
micellar-type self-assembled aggregate. Thus, thermotropic ty-
rosine-derived ILCs 3(m,n)X provide a useful model to study
self-assembly processes ranging from micellar-like via lamellar-
like to inverse hexagonal-like aggregates by simple exchange
of the counterion or reduction of the side-chain symmetry.
This is another example of the many analogies between ther-
motropic ionic and lyotropic liquid crystals[77] and comple-
ments the work by Cheng and Tschierske in which the lamel-
lar—columnar–micellar cubic transitions were achieved by var-
iation of the number and lengths of C-terminal chains as well
as N-terminal chains at an imidazolium unit.[76] Future work
must explore whether a full lyotropic analogue phase se-
4.4
[a] Calculated values using the software by Godbert[64] in parenthesis.
Figure 14. a) Small-angle X-ray scattering profile of 3(14,8)Cl at 498C upon
cooling from the isotropic liquid. Inset: the corresponding diffraction image.
b) Plane group of the Colr mesophase and schematic illustration of a disc
consisting of 5 molecules.
SmA2 phases are comparable, with the exception of 3(14,12)Cl,
which shows a larger d-value than 3(12,12)Cl. In this case, in-
terdigitation of the SmA2 bilayers seemed to be less efficient
for 3(14,12)Cl. For tyrosine chloride 3(14,8)Cl and ILC
3(14,14)PF6 with Colr mesophases the same space symmetry
(p2mm) was observed, however, both the lattice parameters
and Z clearly differ. Symmetrical ILC 3(14,14)PF6 has values of
a=49.5 ꢁ, b=57.8 ꢁ, Z=10, whereas for unsymmetrical ILC
3(14,8)Cl values of a=30.7 ꢁ, b=39.8 ꢁ, Z=5 were found.
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[16] L. Chen, G. E. Mullen, M. Le Roch, C. G. Cassity, N. Gouault, H. Y. Fada-
miro, R. E. Barletta, R. A. O’Brien, R. E. Sykora, A. C. Stenson, K. N. West,
H. E. Horne, J. M. Hendrich, K. R. Xiang, J. H. Davis, Angew. Chem. Int. Ed.
2014, 53, 11762–11765; Angew. Chem. 2014, 126, 11956–11959.
quence SmA–Col–Cub could be obtained for ionic liquid crys-
tals by simple exchange of counterions.
´
[17] P. Borowiecki, M. Milner-Krawczyk, D. Brzezinska, M. Wielechowska, J.
Experimental Section
Plenkiewicz, Eur. J. Org. Chem. 2013, 712–720.
[18] K. V. Axenov, S. Laschat, Materials 2011, 4, 206–259.
[19] L. Douce, J.-M. Suisse, D. Guillon, A. Taubert, Liq. Cryst. 2011, 38, 1653–
1661.
[20] H. Ohno, Electrochemical Aspects of Ionic Liquids, Wiley, 2011.
[21] T. Kato, N. Mizoshita, K. Kishimoto, Angew. Chem. Int. Ed. 2006, 45, 38–
68; Angew. Chem. 2006, 118, 44–74.
[22] K. Binnemans, Chem. Rev. 2005, 105, 4148–4204.
[23] a) K. Goossens, K. Lava, C. W. Bielawski, K. Binnemans, Chem. Rev. 2016,
116, 4643–4807; b) B. Ringstrand, P. Kaszynski, Acc. Chem. Res. 2013, 46,
214–225.
Mesomorphic properties
Polarizing optical microscopy (POM): Olympus BX50 polarizing mi-
croscope combined with a Linkam TP93 central controller. X-ray
diffraction (WAXS, SAXS regions): Bruker AXS Nanostar C diffrac-
tometer employing Ni-filtered CuKa radiation. Differential scanning
calorimetry (a=1.5418 ꢁ) (DSC): Mettler-Toledo DSC 822e.
Crystallographic details of 16b
[24] a) C. H. Lochmueller, R. W. Souter, J. Phys. Chem. 1973, 77, 3016–3020;
b) C. H. Lochmueller, R. W. Souter, J. Chromatogr. 1973, 88, 41–54;
c) C. H. Lochmueller, J. V. Hinshaw, J. Chromatogr. 1979, 171, 407–410;
d) C. H. Lochmueller, J. V. Hinshaw, J. Chromatogr. 1980, 202, 363–368.
[25] D. J. Saccomando, C. Black, G. W. V. Cave, D. P. Lydon, J. P. Rourke, J. Or-
ganomet. Chem. 2000, 601, 305–310.
[26] a) J.-Y. Chane-Ching, F. Cobo, D. Aubert, H. G. Harvey, M. Airiau, A.
Corma, Chem. Eur. J. 2005, 11, 979–987; b) Y. Zhou, R. L. Marson, G.
van Anders, J. Zhu, G. Ma, P. Ercius, K. Sun, B. Yeom, S. C. Glotzer, N. A.
Kotov, ACS Nano 2016, 10, 3248–3256.
[27] a) G.-C. Kuang, Y. Ji, X.-R. Jia, Y. Li, E.-Q. Chen, Y. Wei, Chem. Mater. 2008,
20, 4173–4175; b) G. Kuang, Y. Ji, X. Jia, E. Chen, M. Gao, J. Yeh, Y. Wei,
Chem. Mater. 2009, 21, 456–462; c) G.-C. Kuang, X.-R. Jia, M.-J. Teng, E.-
Q. Chen, W.-S. Li, Y. Ji, Chem. Mater. 2012, 24, 71–80.
[28] K. Sakajiri, H. Yoshida, K. Moriya, S. Kutsumizu, Chem. Lett. 2009, 38,
1066–1067.
C32H55NO5; Fw =533.77; 0.36ꢂ0.09ꢂ0.08 mm3; orthorhombic;
P212121; a=5.0920(3) ꢁ, b=8.7874(5) ꢁ, c=71.955(4) ꢁ, a =b =
g=908; V=3219.7 ꢁ3; Z=4; 1calcd =1.101 mgmꢀ3; m=0.572 mmꢀ1
;
22128 reflections were measured at a temperature of 100(2) K;
5421 reflections were unique (Rint =0.0422), of which 5122 were
observed [I>2s(I)]. R1/wR2 [I>2s(I)]: 0.0404/0.1012. R1/wR2 [all
data]: 0.0430/0.1024. S=1.051; residual electron density between
0.195 and ꢀ0.149 eꢁꢀ3. For further data see the Supporting Infor-
mation.
Acknowledgements
Generous financial support by the Deutsche Forschungsge-
meinschaft (LA 907/17-1 and GI 243/8-1), the Bundesministeri-
um fꢃr Bildung und Forschung (shared instrumentation grant
01 RI05177), the Fonds der Chemischen Industrie, the Ministeri-
um fꢃr Wissenschaft, Forschung und Kunst des Landes Baden-
Wꢃrttemberg, and the Gesellschaft Deutscher Chemiker (travel
grant for M.N.) is gratefully acknowledged. We would like to
thank E. Cunningham, University of Leicester, UK, for initial ex-
periments and I. Hartenbach, University of Stuttgart, for valu-
able discussion.
[29] G. Shanker, D. S. Shankar Rao, S. K. Prasad, C. V. Yelamaggad, RSC Adv.
2012, 2, 1592–1597.
[30] G. Shanker, D. S. Shankar Rao, S. Krishna Prasad, C. V. Yelamaggad, Tetra-
hedron 2012, 68, 6528–6534.
[31] C. V. Yelamaggad, G. Shanker, Tetrahedron 2008, 64, 3760–3771.
[32] S.-H. Luo, Q.-F. Wang, Z.-Y. Wang, P. Peng, Res. Chem. Intermed. 2013, 39,
2513–2526.
[33] S.-H. Luo, J.-F. Xiong, Z.-Y. Wang, G.-Z. Mo, Res. Chem. Intermed. 2013,
39, 1865–1876.
[34] D. W. L. Hukins, J. Woodhead-Galloway, Mol. Cryst. Liq. Cryst. 1977, 41,
33–39.
[35] S. Takai, W. C. Richards, J. Phytopathol. 1978, 91, 129–146.
[36] Y. Murakami, A. Nakano, K. Fukuya, J. Am. Chem. Soc. 1980, 102, 4253–
4254.
[37] B. Gallot, H. Haj Hassan, Mol. Cryst. Liq. Cryst. 1989, 170, 195–214.
[38] a) M. Lavrentovich, T. Sergan, J. Kelly, Liq. Cryst. 2003, 30, 851–859;
b) C. K. McGinn, L. I. Laderman, N. Zimmermann, H.-S. Kitzerow, P. J. Col-
lings, Phys. Rev. E 2013, 88, 062513/1-062513/9; c) C. Peng, O. D. Lavren-
tovich, Soft Matter 2015, 11, 7257–7263.
Keywords: amino acids · ionic liquid crystals · mesophases ·
self-assembly · X-ray diffraction
[1] C. Chiappe, C. S. Pomelli, Eur. J. Org. Chem. 2014, 6120–6139.
[2] T. Payagala, D. W. Armstrong, Chirality 2012, 24, 17–53.
[3] R. Giernoth, Angew. Chem. Int. Ed. 2010, 49, 2834–2839; Angew. Chem.
2010, 122, 2896–2901.
[39] D. Libster, P. B. Ishai, A. Aserin, G. Shoham, N. Garti, Langmuir 2008, 24,
2118–2127.
[40] S. R. Sarker, S. Arai, M. Murate, H. Takahashi, M. Takata, T. Kobayashi, S.
Takeoka, Int. J. Pharm. 2012, 422, 364–373.
[41] A. Chemelli, B. Conde-Valentin, F. Uhlig, O. Glatter, Langmuir 2015, 31,
10377–10381.
[42] a) K. Fujimura, T. Ichikawa, M. Yoshio, T. Kato, H. Ohno, Chem. Asian J.
2016, 11, 520–526; b) T. Ichikawa, K. Fujimura, M. Yoshio, T. Kato, H.
Ohno, Chem. Commun. 2013, 49, 11746–11748.
[43] J. Wang, T. L. Greaves, D. F. Kennedy, A. Weerawardena, G. Song, C. J.
Drummond, Aust. J. Chem. 2011, 64, 180–189.
[44] G. Echue, G. C. Lloyd-Jones, C. F. J. Faul, Chem. Eur. J. 2015, 21, 5118–
5128.
[45] M. Mansueto, W. Frey, S. Laschat, Chem. Eur. J. 2013, 19, 16058–16065.
[46] M. Butschies, J. C. Haenle, S. Tussetschlꢄger, S. Laschat, Liq. Cryst. 2013,
40, 52–71.
[47] M. Butschies, S. Sauer, E. Kessler, H.-U. Siehl, B. Claasen, P. Fischer, W.
Frey, S. Laschat, ChemPhysChem 2010, 11, 3752–3765.
[48] C. R. Safinya, K. K. Ewert, R. N. Majzoub, C. Leal, New J. Chem. 2014, 38,
5164–5172.
[4] K. Zalewska, L. C. Branco, Mini-Rev. Org. Chem. 2014, 11, 141–153.
[5] Z. Fei, P. J. Dyson, Chem. Commun. 2013, 49, 2594–2596.
[6] M. Pereira, Mini-Rev. Org. Chem. 2012, 9, 243–260.
[7] S. Tang, G. A. Baker, H. Zhao, Chem. Soc. Rev. 2012, 41, 4030–4066.
[8] T.-K.-T. Truong, O. Nguyen Van Buu, A. Aupoix, B. Pegot, G. Vo-Thanh,
Curr. Org. Synth. 2012, 9, 53–64.
[9] M. Pucheault, M. Vaultier, in Ionic Liquids (Ed.: B. Kirchner), Springer, Hei-
delberg, 2009, pp. 83–126.
[10] M. H. G. Prechtl, J. D. Scholten, B. A. D. Neto, J. Dupont, Curr. Org. Chem.
2009, 13, 1259–1277.
[11] J.-C. Plaquevent, J. Levillain, F. Guillen, C. Malhiac, A.-C. Gaumont, Chem.
Rev. 2008, 108, 5035–5060.
[12] K. Bica, P. Gaertner, Eur. J. Org. Chem. 2008, 3235–3250.
[13] A. Winkel, P. Reddy, R. Wilhelm, Synthesis 2008, 999–1016.
[14] C. M. Gordon, M. J. Muldoon, M. Wagner, C. Hilgers, J. H. Davis, P. Was-
serscheid, in Ionic Liquids in Synthesis (Eds.: P. Wasserscheid, T. Welton),
Wiley-VCH, Weinheim, 2007, pp. 7–55.
[15] H. Ohno, K. Fukumoto, Acc. Chem. Res. 2007, 40, 1122–1129.
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[49] M. T. Vitor, P. C. Bergami-Santos, J. A. M. Barbuto, L. G. de La Torre,
Recent Pat. Drug Deliv. Formul. 2013, 7, 99–110.
[64] N. Godbert, A. Crispini, M. Ghedini, M. Carini, F. Chiaravalloti, A. Ferrise,
J. Appl. Crystallogr. 2014, 47, 668–679.
[50] G. Caracciolo, H. Amenitsch, Eur. Biophys. J. 2012, 41, 815–829.
[51] C. R. Safinya, K. K. Ewert, C. Leal, Liq. Cryst. 2011, 38, 1715–1723.
[52] Y. Zhao, S. Zhang, S. Cui, B. Wang, S. Zhang, Expert Opin. Drug Delivery
2011, 9, 127–139.
[65] J. M. Seddon, D. Demus, J. Goodby, G. W. Gray, H.-W. Spiess, V. Vill, in
Handbook of Liquid Crystals, Wiley-VCH, Weinheim, 2008, pp. 635–679.
[66] J. P. F. Lagerwall, F. Giesselmann, M. D. Radcliffe, Phys. Rev. E 2002, 66,
031703.
[53] W. Dobbs, B. Heinrich, C. Bourgogne, B. Donnio, E. Terazzi, M.-E. Bonnet,
F. Stock, P. Erbacher, A.-L. Bolcato-Bellemin, L. Douce, J. Am. Chem. Soc.
2009, 131, 13338–13346.
[54] B. M. Rosen, M. Peterca, K. Morimitsu, A. E. Dulcey, P. Leowanawat, A.-M.
Resmerita, M. R. Imam, V. Percec, J. Am. Chem. Soc. 2011, 133, 5135–
5151.
[55] W. Kantlehner, E. Haug, W. W. Mergen, P. Speh, T. Maier, J. J. Kapassakali-
dis, H.-J. Brꢄuner, H. Hagen, Liebigs Ann. Chem. 1984, 1984, 108–126.
[56] R. Moumne, S. Lavielle, P. Karoyan, J. Org. Chem. 2006, 71, 3332–3334.
[57] N. Joondan, S. Jhaumeer-Laulloo, P. Caumul, Microbiol. Res. 2014, 169,
675–685.
[58] CCDC 1478863 (16b) contains the supplementary crystallographic data
for this paper. These data are provided free of charge by The Cam-
bridge Crystallographic Data Centre.
[59] L. M. Antill, M. M. Neidhardt, J. Kirres, S. Beardsworth, M. Mansueto, A.
Baro, S. Laschat, Liq. Cryst. 2014, 41, 976–985.
[60] M. R. Schenkel, J. B. Hooper, M. J. Moran, L. A. Robertson, D. Bedrov,
D. L. Gin, Liq. Cryst. 2014, 41, 1668–1685.
[61] M. J. Pottage, T. L. Greaves, C. J. Garvey, S. T. Mudie, R. F. Tabor, Soft
Matter 2015, 11, 261–268.
[67] Chem. 3D from ChemBioDraw Ultra 13.0.0.3015 (PerkinElmer).
[68] V. Bergeron, Langmuir 1997, 13, 3474–3482.
[69] F. Giesselmann, R. Germer, A. Saipa, J. Chem. Phys. 2005, 123, 034906-
034906-6.
[70] S. Sauer, S. Saliba, S. Tussetschlꢄger, A. Baro, W. Frey, F. Giesselmann, S.
Laschat, W. Kantlehner, Liq. Cryst. 2009, 36, 275–299.
[71] J. C. Haenle, M. M. Neidhardt, S. Beardsworth, J. Kirres, A. Baro, S. La-
schat, Aust. J. Chem. 2014, 67, 1088–1099.
[72] J. N. Israelachvili, D. J. Mitchell, B. W. Ninham, J. Chem. Soc. Faraday
Trans. 2 1976, 72, 1525–1568.
[73] C. Tschierske, J. Mater. Chem. 1998, 8, 1485–1508.
[74] T. Matsumoto, T. Ichikawa, J. Sakuda, T. Kato, H. Ohno, Bull. Chem. Soc.
Jpn. 2014, 87, 792–796.
[75] A. Yamaguchi, Y. Maeda, H. Yokoyama, A. Yoshizawa, Chem. Mater. 2006,
18, 5704–5710.
[76] X. Cheng, X. Bai, S. Jing, H. Ebert, M. Prehm, C. Tschierske, Chem. Eur. J.
2010, 16, 4588–4601.
[77] C. Stubenrauch, F. Gießelmann, Angew. Chem. Int. Ed. 2016, 55, 3268–
3275; Angew. Chem. 2016, 128, 3324–3332.
[62] A. Pana, F. Badea, M. Ilis, T. Staicu, M. Micutz, I. Pasuk, V. Circu, J. Mol.
Struct. 2015, 1083, 245–251.
Received: June 20, 2016
[63] S. Zhang, N. Sun, X. He, X. Lu, X. Zhang, J. Phys. Chem. Ref. Data 2006,
35, 1475–1517.
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Materials Chemistry
M. M. Neidhardt, M. Wolfrum,
S. Beardsworth, T. Wçhrle, W. Frey,
A. Baro, C. Stubenrauch, F. Giesselmann,
S. Laschat*
The role of the anion: Hexafluorophos-
phate induces a mesophase change
from SmA to Colr for novel l-tyrosine-
based guanidinium ionic liquid crystals.
Due to the large PF6 anionic radius, the
formation of discoid aggregates of ten
self-assembled molecules is preferred,
leading to a columnar packing. This ar-
gument is further rationalized by con-
sidering a strongly increased interfacial
curvature in analogy to the packing
model of lyotropic liquid crystals.
&& – &&
Tyrosine-Based Ionic Liquid Crystals:
Switching from a Smectic A to
a Columnar Mesophase by Exchange
of the Spherical Counterion
&
&
Chem. Eur. J. 2016, 22, 1 – 12
www.chemeurj.org
12
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!