Towards room temperature ionic liquid crystals: Linear versus bent imidazolium phenylpyrimidines

Article abstract of DOI:10.1039/c2jm34595a

The synthesis and the mesomorphic properties of novel imidazolium salts with mesogenic 2-phenylpyrimidine or 2-alkylpyrimidinecarboxylic acid central cores are reported. The mesogenic units are connected to the imidazolium head groups via an alkoxy spacer

Full text of DOI:10.1039/c2jm34595a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 21987
www.rsc.org/materials
PAPER
Towards room temperature ionic liquid crystals: linear versus bent
imidazolium phenylpyrimidines†
*
€
Gundula F. Starkulla, Simon Klenk, Martin Butschies, Stefan Tussetschlager and Sabine Laschat
Received 13th July 2012, Accepted 17th August 2012
DOI: 10.1039/c2jm34595a
The synthesis and the mesomorphic properties of novel imidazolium salts with mesogenic 2-
phenylpyrimidine or 2-alkylpyrimidinecarboxylic acid central cores are reported. The mesogenic units
are connected to the imidazolium head groups via an alkoxy spacer. In order to adjust the mesomorphic
properties the length of the alkoxy spacer and the terminal alkyl group, the counter ion, the substitution
pattern of the imidazolium head group and the molecular geometry (linear vs. bent) are varied and the
corresponding compounds were investigated in detail using differential scanning calorimetry (DSC),
polarized optical microscopy (POM) and X-ray scattering (WAXS, SAXS). Whereas SmA phases with
monolayer orientation were observed for imidazolium salts with short N-substituents (R ¼ CH3, C4H9)
at the imidazolium head group, the corresponding derivatives with longer N-substituents (R ¼ C12H25,
C12H25OC6H4) displayed SmA phases with bilayer orientation irrespective of a linear or bent geometry.
For two derivatives, p-5(10,8) and p-5(12,8), a SmC phase was observed. Indeed, bending of the
mesogenic core led to a shift of the mesophases towards lower temperatures. Several of the meta-2-
phenylpyrimidine derivatives as well as 2-pyrimidine carboxylates displayed melting points below
50 ꢀC. For 2-pyrimidine carboxylates replacement of a bromide anion by triflate resulted in a further
decrease of the melting transition close to ambient temperature.
5-phenylpyrimidines
1 and 2 (Scheme 1) are prominent
Introduction
members.4 The nitrogen atoms introduce lateral polarity which is
favourable for the formation of SmC phases while retaining a
rod-like structure resulting in low viscosity materials. Thus,
phenylpyrimidines are widely applied as ferroelectric host
materials.4,5
Ionic liquids (ILs) are molten salts, in which either the cation or
the anion or both are organic compounds and which are firmly
established for both laboratory as well as industrial applications
due to their promising properties such as high chemical and
thermal stability, non-volatility, non-flammability, high ionic
conductivity and broad electrochemical window.1 It has been
fully recognized only recently that ionic liquid crystals (ILCs), i.e.
ionic liquids with liquid crystalline properties, not only combine
the best of two worlds, i.e. ILs and liquid crystals (LCs),2 but
display additional features which are not observed in ILs or LCs
alone, e.g., one dimensional ionic conductivity.3 Thus, ILCs can
be considered as charged supramolecular ordered systems whose
mesomorphic properties, i.e. mesophase type, packing symmetry,
and phase transition temperatures, can be adjusted by the type of
anions/cations, mesogenic subunits, substitution patterns and
additives although a de novo design of certain properties is still
out of reach.
Furthermore, 2- and 5-phenylpyrimidines, 1 and 2, are not just
regioisomers, but they differ in their conformation.6,7 For
2-phenylpyrimidine 1 a planar conformation is favoured while
the corresponding 5-phenylpyrimidine 2 is twisted due to steric
hindrance between lateral ortho-hydrogen atoms. Therefore, it is
not surprising that the regioisomers also differ in their meso-
morphic properties, as observed by Lemieux and Tschierske.8,9
Despite ample precedence of ionic liquid crystals containing
biphenyls as calamitic units,3 the corresponding phenyl-
pyrimidines 3 (Scheme 2) have only been rarely employed for
such purposes.10
We have recently reported the mesomorphic properties of
5-phenylpyrimidines 3 with a tethered imidazolium moiety with
respect to chain lengths, tether lengths and counter ions.10
Calamitic thermotropic liquid crystals consist of a large variety
of structurally different mesogenic cores, in which both 2- and
€
€
Institut fur Organische Chemie, Universitat Stuttgart, Pfaffenwaldring 55,
70569 Stuttgart, Germany. E-mail: sabine.laschat@oc.uni-stuttgart.de;
Fax: +49-711-685-64285; Tel: +49-711-685-64565
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2jm34595a
Scheme 1
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 21987–21997 | 21987

View Article Online
Scheme 3
commercially available16a or can be easily prepared from 4-cya-
nophenol via ammonolysis,16b to afford 4-(5-alkylpyrimidin-2-yl)
phenols p-16(n). Further Williamson etherification gave the
respective bromides p-17(n,m). The latter was treated with
Scheme 2
Although N-methylimidazolium bromides 3 (n ¼ 12, m ¼ 4–9)
displayed broad mesophase ranges (SmA), the high clearing
temperatures affected the thermal stability considerably and led
to decomposition after repetitive heating/cooling cycles. In
addition melting points were relatively high, thus limiting the
potential applications of these novel materials.
In order to shift the mesophase towards ambient temperature
we anticipated that a bent geometry of the calamitic phenyl-
pyrimidine via a meta- rather than a para-substitution pattern
should be a suitable approach for such purposes. From ionic
liquids it is well known that reduction of symmetry leads to a
decrease of melting points.11 In addition the corresponding linear
2-phenylpyrimidines p-5(n,m)–p-8(n,m) and bent 2-phenyl-
pyrimidines m-5(n,m)–m-8(n,m) were investigated for compar-
ison. The results are reported below.
Results and discussion
Following our previously established procedure, enamine 912 was
submitted to condensation with dodecanamidine 10a13 to give
3-(2-dodecylpyrimidin-5-yl)phenol 11.10,14 Subsequent ether-
ification with 1,9-dibromononane gave bromide 12, which was
alkylated with N-methylimidazole 13a to yield the desired bent
5-phenylpyrimidine 4 (Scheme 3).
However, due to the very poor yield of key intermediate 11 we
abandoned this route and decided to look into more detail of the
corresponding linear and bent 2-phenylpyrimidines p-5(n,m)–p-
8(n,m) and m-5(n,m)–m-8(n,m), respectively. For the synthesis of
2-phenylpyrimidines p-5(n,m)–p-8(n,m) the inverse route was
chosen as compared to 5-phenylpyrimidines (Scheme 4).
Enolethers 14(n), being accessible from dimethyl acetals
according to a literature procedure,15 were condensed with
4-hydroxybenzenecarboximidamide hydrochloride p-15, which is
Scheme 4 Preparation of 2-phenylpyrimidines p-5(n,m)–p-8(n,m) and m-
5(12,m)–m-8(12,m) (numbering for NMR assignment).
21988 | J. Mater. Chem., 2012, 22, 21987–21997
This journal is ª The Royal Society of Chemistry 2012

View Article Online
N-alkylimidazoles 13a–d to afford the desired target compounds
p-5(n,m)–p-8(n,m). In the case of meta-substituted 2-phenyl-
pyrimidines, only derivatives m-5(12,m)–m-8(12,m) with the most
promising chain length C12 at the pyrimidine moiety were
prepared in an analogous manner by condensation of 14(12) and
m-15, followed by etherification and reaction with N-alkylimi-
dazoles 13a–d (Scheme 4).
(98–177 ꢀC and 116–197 ꢀC) (entries 7 and 8). It should be noted
that SmC phases are relatively uncommon for ionic liquid
crystals.17
One possible reason for the observation of SmC phases of
compounds p-5(10,8) and p-5(12,8) might be that, for these
particular spacer and chain length combinations, the ‘‘SmC
promoting effect’’ of phenylpyrimidines, reported by
Lemieux,4,8a seems to overrule any other mesophase-controlling
effects. The finding that compound p-5(10,8) has a lower melting
point as compared to compound p-5(8,8) (entries 6 and 7) might
be explained as follows. In the case of p-5(10,8) the melting
transition from Cr to SmC presumably requires less energy as
compared to the melting transition from Cr to SmA in the case of
p-5(8,8).
Mesomorphic properties of compounds 4, p-5(n,m)–p-8(n,m)
were investigated by differential scanning calorimetry (DSC),
polarizing optical microscopy (POM) and X-ray diffraction
(WAXS and SAXS). The DSC results are summarized in Table 1.
Comparison of the known para-substituted 5-phenyl-
pyrimidine 3a10 with the corresponding meta-substituted 5-phe-
nylpyrimidine 4 revealed a decrease of both melting and clearing
temperatures from 72 ꢀC and 169 ꢀC to 34 ꢀC and 114 ꢀC,
respectively (entries 1 and 2). As mentioned above, the poor yield
of 4 precluded the access to other derivatives of this series and
thus, we focused on the 2-phenylpyrimidines. Para-substituted 2-
phenylpyrimidines p-5(8,4)–p-5(12,4) with a methylimidazolium
unit and a C4 spacer displayed melting points between 89 and
112 ꢀC and clearing points between 190–255 ꢀC (entries 3–5).
Increasing the spacer length to C8 had only a little effect on the
melting temperatures, but decreased the clearing transitions to
148–197 ꢀC (entries 6–8). However, the most surprising effect
was that compounds p-5(10,8) and p-5(12,8) displayed a SmC
phase (71–98 ꢀC and 104–116 ꢀC) in addition to the SmA phase
Extending the lengths of the substituent at the imidazolium
unit has considerable influence on the mesomorphic properties.
Within the N-butylimidazolium series derivatives p-6(8,4),
p-6(10,4) and p-6(8,8) are non-mesomorphic, while compounds
p-6(12,4), p-6(10,8) and p-6(12,8) display SmA mesophases with
ꢀ
much lower melting (46–69 C) and clearing temperatures (87–
113 ꢀC) (entries 9–14) as compared to the corresponding
N-methylimidazolium salts p-5(12,4), p-5(10,8) and p-5(12,8). A
further extension of the lengths of the N-substituent gives a
different result. While melting points either decreased only
slightly for the C4 spacer on the N-dodecylimidazolium series
p-7(n,m), and increased for the C8 spacer (entries 15–20), clearing
Table 1 Phase transition temperatures (ꢀC) and transition enthalpies (kJ molꢁ1) of compounds 3, 4 and p-5(n,m)–p-8(n,m) with different side chains,
tether lengths and N-substituents
Entry
Compound
R1
Cr
SmC
SmA
I
Heating/cooling
1
2
3
4
5
6
7
8
3aa
4
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
C4H9
C4H9
C4H9
C4H9
C4H9
C4H9
C12H25
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
72 (25.7)
34 (16.6)
99 (9.6)
—
—
—
—
—
—
ꢂ
ꢂ
169 (0.6)
114 (0.3)
190 (0.8)
220 (0.5)
255 (0.7)
148 (0.9)
177 (0.7)
197 (0.7)
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
2. Heating
2. Heating
2. Heating
1. Heating
1. Heating
2. Heating
2. Heating
2. Heating
1. Heatingc
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Cooling
2. Heating
2. Cooling
2. Heating
2. Heating
2. Heating
2. Cooling
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Cooling
2. Heating
ꢂ
p-5(8,4)
p-5(10,4)b
p-5(12,4)b
p-5(8,8)
p-5(10,8)
p-5(12,8)
p-6(8,4)
p-6(10,4)
p-6(12,4)
p-6(8,8)
p-6(10,8)
p-6(12,8)
p-7(8,4)
ꢂ
89 (6.4)
ꢂ
112 (21.5)
88 (18.7)
71 (25.1)
104 (48.0)
56 (13.2)
97 (15.1)
46 (2.7)
82 (3.7)
54 (21.7)
69 (4.0)
110 (23.9)
93 (ꢁ27.0)
112 (54.9)
106 (ꢁ30.4)
111 (30.9)
74 (42.1)
78 (54.9)
65 (ꢁ50.7)
81 (53.2)
158 (35.2)
156 (38.7)
143 (34.2)
158 (37.8)
164 (40.7)
137 (ꢁ38.6)
99 (6.9)
ꢂ
ꢂ
98 (0.4)
116 (0.6)
ꢂ
ꢂ
ꢂ
9
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
ꢂ
10
11
12
13
14
15
113 (0.4)
—
ꢂ
87 (6.1)
106 (1.0)
ꢂ
—
ꢂ
100 (ꢁ1.8)
16
p-7(10,4)
C12H25
—
ꢂ
116 (ꢁ1.8)
125 (1.7)
79 (1.7)
17
18
19
p-7(12,4)
p-7(8,8)
p-7(10,8)
C12H25
C12H25
C12H25
ꢂ
ꢂ
—
ꢂ
82 (ꢁ3.6)
20
21
22
23
24
25
p-7(12,8)
p-8(8,4)
p-8(10,4)
p-8(12,4)
p-8(8,8)
C12H25
Ar
Ar
Ar
Ar
—
ꢂ
190 (2.4)
204 (2.3)
195 (1.9)
ꢂ
ꢂ
—
—
ꢂ
p-8(10,8)
Ar
142 (ꢁ1.3)
26
p-8(12,8)
Ar
—
a
Values from ref. 10. b Decomposition after 1. heating. c No mesophases were detected. Ar ¼ C12H25OC6H4, Cr ¼ crystalline, SmC ¼ smectic C, SmA ¼
smectic A, and I ¼ isotropic. Dot ¼ phase was observed; dash ¼ phase was not observed.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 21987–21997 | 21989

View Article Online
points were shifted significantly to lower temperatures, resulting
in the complete loss of mesophases for p-7(8,4), p-7(10,4),
p-7(10,8), p-7(12,8) and a decrease of the mesophase width for
p-7(12,4) and p-7(8,8). For p-7(8,4), p-7(10,4) and p-7(10,8) SmA
mesophases were observed only upon second cooling from the
isotropic liquid (entries 15, 16, and 19). This counterbalance of
spacer lengths versus the N-substituent size can also be observed
for the series p-8(n,m) with a more rigidified N-(4-dodecyloxy-
phenyl)imidazolium unit (entries 21–26). Melting points are
significantly increased to 99–164 ꢀC as compared to the
N-methylimidazolium series p-5(n,m) regardless of the spacer.
Clearing points, however, remained high for the C4 spacer as
observed for the N-methyl series p-5(n,4), while a decrease of
clearing temperatures for the C8 spacer resulted in the complete
loss of mesomorphism. Only derivative p-8(10,8) displayed a
SmA mesophase over a small temperature range of 5 K upon
second cooling.
some cases small second-order (002) and third-order (003) peaks,
ꢀ
and the diffuse scattering from the alkyl chains around 4.8 A.
The first order peaks (001) were fit with a Gaussian distribution
to obtain the exact d001 values (the whole XRD data can be found
in the ESI†). Two typical diffractograms are shown in Fig. 3
and 4.
Compounds p-5(8,8)–p-5(12,8) were also investigated in
temperature-dependent XRD experiments to confirm the above-
mentioned observations under POM. While both compounds
p-5(10,8) and p-5(12,8) displayed slightly increasing layer spac-
ings within the SmC phase and strongly decreasing layer spacings
in the SmA phases upon raising the temperature, only a SmA
phase with much less temperature-dependent layer spacing was
observed for compound p-5(8,8) (Fig. 5). The negative thermal
expansion within the SmA phase results from the loss of orien-
tational order upon raising the temperature.19,20 It should be
noted that this behavior already has been observed
previously.21,22
The effect of the spacer versus the N-substituent for meta-
substituted 2-phenylpyrimidines m-5(12,m)–m-8(12,m) is given in
Table 2.
Further temperature-dependent measurements for compounds
p-6(12,8), p-7(12,4), p-8(12,4), m-5(12,4) and m-7(12,4) have been
done (see ESI†). To assign whether the molecules are packed in
mono- or bilayers within the smectic mesophases, the experi-
mental data (see ESI†) were interpolated to determine the d001
values at a reduced temperature near the clearing point (T ¼ 0.95
ꢃ Tiso) and the results were compared to the molecular length
calculated with Chem3D23 (Table 3).
Comparison of the results for these compounds with the cor-
responding para-substituted series p-5(12,m)–p-8(12,m) clearly
revealed a decrease of both melting and clearing points with the
effect being most pronounced for compound m-5(12,8) albeit at
the loss of the SmC phase, which is present in p-5(12,8). In the
case of m-6(12,4), any mesomorphism totally disappeared and in
m-7(12,8) the mesophase width was decreased to a range of only
4 K. Typical DSC traces are given in Fig. 1.
For derivatives p-5(10,8) and p-5(12,8) displaying an addi-
tional SmC mesophase the experimentally determined d001 values
were also interpolated to temperatures near the SmA/SmC phase
transition (T ¼ 1.05 ꢃ TSmC/SmA and T ¼ 0.95 ꢃ TSmC/SmA). As
expected the d001 values significantly increase from 5724 pm
for p-5(10,8) and 5982 pm for p-5(12,8) (both at T ¼ 0.95 ꢃ
TSmC/SmA) to 6432 pm for p-5(10,8) and 6486 pm for p-5(12,8)
(both at T ¼ 1.05 ꢃ TSmC/SmA) at the transition from the smectic
C to the smectic A phase (Table 3).
Under POM, samples of compounds p-5(10,8) and p-5(12,8),
which were slowly cooled from the isotropic mixture, showed
homeotropic alignment and dichroism was only visible at the air–
liquid interface of air bubbles. By shearing of the homeotropic
textures lamellar textures could be generated. For compounds
p-5(10,8) and p-5(12,8) the phase transition from SmA to SmC
was clearly visible under POM. Upon cooling from the isotropic
phase, homeotropic textures of the SmA phase were observed,
which turned into a non-characteristic texture at 110 ꢀC and
became a Schlieren texture at 105 ꢀC typical for a SmC phase
(Fig. 2).18
For all mesophases of compounds p-5(12,4), p-5(8,8),
p-5(10,8), p-8(12,8), p-6(12,8) and p-5(12,4) with short tails at the
imidazolium head group (N-Me, N-Bu) the quotient of the
observed layer spacing and the calculated molecular length is
larger than 1, while it is smaller than 1 for compounds p-7(12,4),
p-8(12,4) and m-7(12,4) with larger tails (N-n-C12H25, N-4-
C6H4OC12H25) attached to the head group. We therefore
propose smectic bilayers with interdigitated alkyl chains for the
former ones and smectic monolayers for the latter ones. The
formation of smectic bilayers where the ionic head groups of the
molecules are stacked towards each other seems to be only
favored for compounds with small chains (R1 ¼ CH3, C4H9)
attached to the ionic heads, while compounds bearing larger
chains (R ¼ C12H25, 4-C6H4OC12H25) at the imidazolium head
group favor the formation of smectic monolayers, possibly due
to steric hindrance. A similar behavior has been observed for
5-phenylpyrimidine derivatives.24 The tightest packing of the
molecules within the smectic bilayers was observed for
compound p-6(12,8) (d001/Lcalc. ¼ 1.19) with the N-butylimida-
zolium head group. A similar observation was made for the
corresponding 5-phenylpyrimidine isomer.10
XRD studies of selected compounds p-5(12,4), p-5(8,8),
p-5(10,8), p-5(12,8), p-6(12,8), p-7(12,4), p-8(12,4), m-5(12,4) and
m-7(12,4) revealed diffraction patterns typical for smectic mes-
ophases, with a strong fundamental diffraction peak (001), in
Table 2 Phase transition temperatures (ꢀC) and transition enthalpies (kJ
molꢁ1) of compounds m-5(12,m)–m-8(12,m) with different side chains,
tether lengths and N-substituentsa
Entry Compd
R1
Cr
SmA
I
Heating
1
2
3
4
5
6
7
8
m-5(12,4) CH3
m-5(12,8) CH3
m-6(12,4) C4H9
m-6(12,8) C4H9
m-7(12,4) C12H25
m-7(12,8) C12H25
m-8(12,4) Ar
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
81 (3.4)
52 (8.4)
38 (7.1)
40 (33.3)
55 (0.9)
50 (1.7)
125 (11.7)
91 (21.0)
ꢂ
185 (0.6)
134 (0.4)
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
2. Heating
ꢂ
—
ꢂ
89 (2.2)
88 (1.5)
54 (3.3)
157 (1.2)
108 (1.5)
ꢂ
ꢂ
ꢂ
m-8(12,8) Ar
ꢂ
Although the use of meta-substituted 2-phenylpyrimidines
m-5(12,m)–m-8(12,m) rather than para-substituted ones
p-5(n,m)–p-8(n,m) indeed shifted the mesophases to lower
a
Ar ¼ C12H25OC6H4, Cr ¼ crystalline, SmA ¼ smectic A, and I ¼
isotropic. Dot ¼ phase was observed; dash ¼ phase was not observed.
21990 | J. Mater. Chem., 2012, 22, 21987–21997
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 1 DSC traces of (a) bent compound m-5(12,8) and (b) compound p-5(12,8), displaying a smectic C phase (heating/cooling rate: 10 K minꢁ1).
Fig. 3 SAXS plot and WAXS diffractogram (inset) of compound p-
5(12,8) within the SmC phase (both at 110 ꢀC).
Fig. 2 Compounds p-5(10,8) and p-5(12,8) under crossed polarizers
upon cooling from the isotropic melt (magnification ꢃ200): (a) homeo-
tropic alignment and dichroism of p-5(10,8) at the air–liquid interface at
160 ꢀC (cooling rate 10 K minꢁ1); (b) lamellar texture generated by
shearing of p-5(10,8) at 160 ꢀC (cooling rate 10 K minꢁ1); (c) unspecific
texture of compound p-5(12,8) observed at the phase transition from
SmA to SmC at 110 ꢀC (cooling rate 5 K minꢁ1); (d) Schlieren texture of
p-5(12,8) typical for a SmC phase (at 105 ꢀC, cooling rate 5 K minꢁ1).
temperatures, we anticipated that this beneficial effect of tilting
the mesogenic unit might be further extended, if the phenyl
moiety is replaced by a bent and more flexible carboxylic ester
moiety.
Fig. 4 SAXS plot and WAXS diffractogram (inset) of compound p-
5(12,8) within the SmA phase (both at 190 ꢀC).
Thus, the sodium salt of 3,3-dimethoxy-2-methoxy-
carbonylpropen-1-ol 1825 was treated with tetradecanamidine
10b to give the desired methyl-2-tetradecylpyrimidine-5-
carboxylate 19, which was submitted to saponification,26
Again, mesomorphic properties were studied by DSC (Table 4).
Indeed, 2-pyrimidine carboxylates 21(m) and 22(m) showed
melting points in the range of 41 to 55 ꢀC independent of the
spacer length or the N-substituent and decreased clearing points
with increasing spacer lengths and/or N-substituent (entries 1–4).
followed by esterification with u-bromo-a-alcohols under Steg-
lich conditions27 to give bromoalkyl pyrimidine-5-carboxylates
20(m). Subsequent treatment with N-alkylimidazole 13a,c,d
yielded the target products 21(m)–26(m) (Scheme 5).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 21987–21997 | 21991

View Article Online
Fig. 5 Temperature dependence of the layer spacing for compounds p-
5(8,8)–p-5(12,8).
In the case of N-4-dodecyloxyphenylimidazolium derivatives
23(4) and 23(8) the transition temperatures are still relatively
high (entries 5 and 6). However, the high clearing temperatures
of 21(4), 21(8) and 23(4) resulted in decomposition after pro-
longed heating or repetitive heating/cooling cycles.
Scheme 5
In order to avoid this thermal decomposition we decided to
replace bromide by triflate, because literature precedence
attributed this counter ion a beneficial effect on the thermal
stability.28 We were pleased to find that the N-methyl and N-
dodecylimidazolium triflates 23(4), 24(8) and 25(4) indeed
showed the SmA phase at or slightly above roꢀom temperature
and clearing temperatures between 73 and 142 C (entries 7–9).
Table 4 also shows the balanced influence between the spacer and
the N-substituent. While short C4 spacer/N-methyl resulted in
clearing transition at 142 ꢀC, the combination C8 spacer/N-
methyl decreased the clearing transition to 79 ꢀC. The combi-
nation C4 spacer/N-dodecyl decreased the clearing temperature
to 73 ꢀC. Again the long/long combination, i.e. C8 spacer/N-
dodecyl turned out to be unfavourable for mesophase formation
(entry 10). Both N-(dodecyloxyphenyl)imidazolium triflates
26(4) and 26(8) also displayed SmA mesophases albeit with
higher melting points than those of 24(m) and 25(m), but
comparable clearing points (entries 11 and 12).
^
Under POM typical Batonnet textures together with homeo-
tropic textures were observed upon cooling. As an example
textures of compound 26(4) are shown in Fig. 6.
The temperature-dependency of the layer spacing was inves-
tigated for 30b, 31a and 32a upon cooling from the isotropic melt
(Fig. 7).
Upon increasing temperature the layer spacings of 24(8), 25(4)
and 26(4) became smaller due to the simultaneous decrease of
orientational order. This behaviour is typical for SmA phases and
confirms the above-proposed assignment of the mesophase typeof
the triflate compounds.29 As described above, the d001 values at a
reduced temperature were determined in order to compare them
with the calculated length of the molecules (Table 5).
The quotient of experimentally determined layer spacing and
the calculated molecular length is smaller than 1 for compounds
25(4) and 26(4). SmA monolayers with strongly interdigitated
alkyl chains are therefore proposed for both compounds (Fig. 8).
In contrast the corresponding value of 1.28 for compound
24(8) reveals the formation of a bilayer structure within the
mesophase (Fig. 9).
Table 3 XRD data of representative 2-phenylpyrimidines p-5(n,m)–p-
8(n,m) and m-5(12,m)–m-8(12,m) assigned type of mesophase and
proposed packing within the mesophases
b
f
LXRD /
pm
Phase
typea
Lcalc
pm
/
d001
pm
/
d001/
Lcalc.
Compound
R1
p-5(12,4)
p-5(8,8)
p-5(10,8)
CH3
CH3
CH3
SmA2
SmA2
SmA2
SmA2
SmC2
SmA2
SmA2
SmC2
SmA2
SmA
3200
3200
3400
3400
4501c
4416c
4633c
6432d
5724e
5115c
6486d
5982e
4746c
3540c
3675c
4053c
2767
1.41
1.38
1.35
1.89
1.68
1.39
1.75
1.62
1.19
0.79
0.74
1.27
0.95
2250
2208
2317
3216
2862
2558
2558
2991
2373
3540
3675
2027
2767
In agreement with the above-discussed 2-phenylpyrimidines
the type of packing pattern depends on the N-substituent. For N-
methyl a smectic bilayer is proposed and for N-4-dodecyloxy-
phenyl and N-dodecyl, smectic monolayers are proposed in
agreement with the experimental data.
p-5(12,8)
CH3
3700
3700
p-6(12,8)
p-7(12,4)
p-8(12,4)
m-5(12,4)
m-7(12,4)
C4H9
C12H25
Ar
Me
C12H25
4000
4500
5000
3200
2900
SmA
SmA2
SmA
Experimental
a
The subscript 2 in SmA2 and SmC2 indicates the formation of bilayers.
Molecular length calculated for all-trans configuration (calculated with
b
General information
c
Chem3D).23 d001 at T ¼ 0.95 ꢃ Tiso (after linear interpolation of data).
d
d001 at T ¼ 1.05 ꢃ TSmC/SmA (after linear interpolation of data). e d001 at
Solvents were dried and distilled under nitrogen prior to use and
generally all reactions were carried out under a nitrogen atmo-
sphere with Schlenk-type glassware. Flash chromatography was
T ¼ 0.95 ꢃ TSmC/SmA (after linear interpolation of data). f Layer spacing
LXRD ¼ d001 for monolayers, LXRD ¼ 0.5 ꢃ d001 for bilayers.
21992 | J. Mater. Chem., 2012, 22, 21987–21997
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Table 4 Phase transition temperatures (ꢀC) and transition enthalpies (kJ molꢁ1) of carboxylates 21(m)–26(m)
Entry
Compound
R1
X
Cr
SmA
I
Heating
1
2
3
4
5
6
7
8
21(4)
21(8)
22(4)
22(8)
23(4)
23(8)
24(4)b
24(8)b
25(4)c
25(8)
26(4)c
26(8)c
CH3
CH3
C12H25
C12H25
C12H25OC6H4
C12H25OC6H4
CH3
CH3
C12H25
Br
Br
Br
Br
Br
Br
CF3SO3
CF3SO3
CF3SO3
CF3SO3
CF3SO3
CF3SO3
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
44 (35.3)
42 (53.0)
41 (18.4)
55 (35.2)
77 (78.5)
101 (18.2)
35 (19.2)
32 (19.9)
25 (21.0)
22
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
—
ꢂ
ꢂ
237 (0.5)
219 (1.0)
164 (1.8)
76 (1.3)
229 (2.6)
172 (2.6)
142 (0.6)
79 (0.3)
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
1sta
2nda
2nd
2nd
1sta
2nd
2nd
2nd
2nd
9
73 (1.1)
10
11
12
C12H25
C12H25OC6H4
C12H25OC6H4
90 (19.0)
49 (8.5)
142 (2.1)
72 (1.6)
2nd
2nd
Decomposition after 1. heating. b Packing as smectic bilayers. c Packing as smectic monolayers. Cr ¼ crystalline, SmA ¼ smectic A, and I ¼ isotropic.
a
Dot ¼ phase was observed; dash ¼ phase was not observed.
performed on silica gel, grain size 40–63 mm (Fluka). Thin layer
chromatography (TLC) was done on aluminium sheets pre-
coated with silica gel 60 F254 (Merck). The obtained chromato-
grams were visualized by UV-light (l ¼ 254 nm) to determine the
retention factors (Rf).
The following instruments were used for characterization of
the compounds. NMR: Bruker Avance 250 (1H, 250 MHz; 13C,
63 MHz) and Bruker Avance 500 (1H, 500 MHz; 13C, 126 MHz).
1H and 13C NMR spectra were referenced to TMS (Me4Si dH
¼
0.0 ppm, dC ¼ 0.0 ppm) as an internal standard. Unless otherwise
stated, spectra were recorded at room temperature. Chemical
shift calculations and 2D experiments (COSY and HMBC)
supported the assignments of the resonances. Elemental anal-
yses: Carlo Erba Strumentazione Elemental Analyzer, Modell
1106. IR: Bruker Vector 22 FT-IR Spectrometer with MKII
golden gate single reflection Diamant ATR system. MS (EI):
Varian MAT 711 spectrometer. MS (ESI): Bruker Daltonics
microTOF-Q spectrometer. Differential scanning calorimetry
(DSC): Mettler-Toledo DSC 822e (heating/cooling rates were 5 K
minꢁ1). Polarizing optical microscopy: Olympus BX50 polarizing
microscope combined with a Linkam TP93 central controller.
X-ray diffraction (WAXS, SAXS regions): Bruker AXS Nano-
star C diffractometer employing Ni-filtered CuKa radiation (l ¼
^
Fig. 6 Batonnet textures of compound 26(4) under crossed polarizers
upon cooling from the isotropic melt (magnification 200): (a) at 45 ꢀC and
(b) at 35 ꢀC (cooling rate 10 K minꢁ1).
ꢀ
1.5418 A). Melting points: Bu€chi SMP-20.
Full characterization of selected new compounds p-5(12,8),
m-5(12,8), 21(8) and 24(8) is given below. For all other
compounds see the ESI.†
General procedure for the preparation of alkylimidazolium
bromides 4, p-5(n,m)–p-6(n,m), m-5(n,m)–m-6(n,m) and 21(m)
Fig. 7 Temperature dependent layer spacings of compounds 24(8), 25(4)
and 26(4).
To
a solution of the respective bromide 12, p-17(n,m),
m-17(12,m) or 20(m) (1.00 mmol) in abs. acetonitrile was added
Table 5 X-ray data for 2-pyrimidine carboxylates 24(8), 25(4) and 26(4)
Comp.
m
R1
X
Phase typea
Lcalc.b/pm
d001c/pm
d001/Lcalc.
LXRDd/pm
24(8)
25(4)
26(4)
8
4
4
Me
C12H25
Ar
CF3SO3
CF3SO3
CF3SO3
SmA2
SmA
SmA
3400
4200
4700
4388
3167
3316
1.29
0.75
0.71
2194
3167
3316
a
b
The subscript 2 in SmA2 indicates the formation of bilayers. Molecular length calculated for all-trans configuration (calculated with Chem3D).21
c
d001 at T ¼ 0.95 ꢃ Tiso (after linear interpolation of data). d Layer spacing LXRD ¼ d001 for monolayers, LXRD ¼ 0.5 ꢃ d001 for bilayers.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 21987–21997 | 21993

View Article Online
Fig. 8 Schematic illustration of a smectic monolayer e.g. formed by compounds 25(4) and 26(4).
freshly distilled N-methylimidazole 13a or N-butylimidaꢀzole 13b
(10.0 mmol) and the reaction mixture was stirred at 60 C for a
given time. Then the mixture was added dropwise in EtOAc (10
mL), the precipitate was filtered off and washed with EtOAc (3 ꢃ
10 mL). The precipitate was dried over P4O10 to give the imi-
dazolium salts as colourless solids.
1108 (w), 1064 (w), 1036 (m), 996 (w), 946 (w), 909 (w), 845 (m),
818 (w), 799 (m), 786 (w), 772 (w), 756 (w), 729 (m), 652 (w), 623
(w), 611 (w). dH (250 MHz, CDCl3) 0.81 (t, 3H, J ¼ 6.6 Hz, CH3),
1.11–1.46 (m, 26H, CH2), 1.50–1.64 (m, 2H, NCH2CH2), 1.66–
1.78 (m, 2H, 5000-CH2CH2), 1.79–1.94 (m, 2H, OCH2CH2), 2.45–
2.60 (m, 2H, 5000-CH2), 3.95 (t, 2H, J ¼ 6.4 Hz, OCH2), 4.05 (s,
3H, NCH3), 4.25 (t, 2H, J ¼ 7.5 Hz, CH2N), 6.85–6.97 (m, 2H,
300-H, 500-H), 7.15–7.32 (m, 2H, 4-H, 5-H), 8.22–8.33 (m, 2H, 200-
H, 600-H), 8.50 (s, 2H, 4000-H, 6000-H), 10.59 (s, 1H, 2-H) ppm. dC
(125 MHz, CDCl3) 14.1 (CH3), 22.7, 25.9, 26.1, 28.8, 29.0, 29.1,
29.3, 29.5, 29.6, 30.2, 30.8, 31.9 (CH2) 36.8 (NCH3), 50.2
(CH2N), 67.9 (OCH2), 114.4 (C-300, C-500), 121.5 (C-4), 123.1
(C-5), 129.4 (C-200, C-600), 130.2 (C-100), 132.2 (C-2), 138.2
(C-5000), 157.0 (C-4000, C-6000), 161.1 (C-400), 162.4 (C-2000) ppm. MS
(ESI): m/z ¼ 533.4 [M ꢁ Br]+. HRMS (ESI): m/z calcd for
C34H53N4O+: 533.4214, found: 533.4210 [M ꢁ Br]+. DSC: Cr
3-Methyl-1-{4-[4-(5-dodecylpyrimidin-2-yl)phenoxy]octyl}
imidazolium bromide (p-5(12,8))
From p-17(12,8) (0.19 mmol, 100 mg) and 13a (1.88 mmol, 154
mg) in abs. DMF (2 mL), reaction time: 120 h; yield: 84.0 mg
(73%), colourless solid. Found: C, 66.60; H, 8.49; N, 9.11%.
C34H53BrN4O requires C, 66.54; H, 8.70; N, 9.13%; M, 585.66.
FT-IR (ATR): nmax/cmꢁ1 3141 (w), 3032 (w), 2920 (s), 2849 (m),
2193 (w), 1608 (m), 1582 (m), 1540 (w), 1514 (w), 1469 (w), 1430
(s), 1395 (w), 1330 (w), 1303 (w), 1251 (s), 1179 (m), 1167 (m),
ꢁ1
ꢁ1
ꢀ
ꢀ
ꢀ
104 C [48.0 kJ mol ] SmC 116 C [0.6 kJ mol ] SmA 197 C
Fig. 9 Schematic illustration of a smectic bilayer, e.g. formed by compound 24(8).
21994 | J. Mater. Chem., 2012, 22, 21987–21997
This journal is ª The Royal Society of Chemistry 2012

View Article Online
[0.7 kJ molꢁ1] I ꢁ(21nd heating); I 197 ꢀC [ꢁ0.7 kJ molꢁ1] SmA 118
ꢀC [ꢁ0.6 kJ mol ] SmC 64 ꢀC [ꢁ47.1 kJ molꢁ1] Cr (2nd cooling).
FT-IR (ATR): nmax/cmꢁ1 3046 (w), 2920 (s), 2851 (m), 1720 (s),
1590 (s), 1551 (m), 1516 (w), 1469 (m), 1442 (m), 1394 (w), 1289
(s), 1257 (m), 1136 (m), 1075 (w). dH (500 MHz, CDCl3) 0.88 (t,
6H, J ¼ 6.9 Hz, 2 ꢃ CH3), 1.19–1.50 (m, 48H, CH2), 1.73–1.87
(m, 6H, CH2), 1.96–2.05 (m, 2H, CH2), 2.98–3.06 (m, 2H, 2-
CH2), 3.98 (t, 2H, J ¼ 6.6 Hz, OCH2), 4.35 (t, 2H, J ¼ 6.9 Hz,
NCH2), 4.60 (t, 2H, J ¼ 7.3 Hz, CH2N), 7.01–7.06 (m, 2H, 3000-
H, 5000-H), 7.39–7.52 (m, 2H, 4-H, 5-H), 7.66–7.70 (m, 2H, 2000-
H, 6000-H), 9.17 (s, 2H, 400-H, 600-H), 11.15 (s, 1H, 2-H) ppm. dC
(125 MHz, CDCl3) 14.1 (CH3), 22.7, 25.8, 26.0, 26.2, 28.5, 28.6,
28.91, 28.94, 29.1, 29.36, 29.42, 29.5, 29.57, 29.60, 29.62, 29.66,
29.68, 30.4, 31.90 (CH2), 39.8 (200-CH2), 50.4 (CH2N), 65.6, 68.7
(2 ꢃ OCH2), 116.1 (C-3000, C-5000), 120.3 (C-500), 121.5, 122.0 (C-
4, C-5), 123.3 (C-2000, C-6000), 127.1 (C-1000), 136.5 (C-2), 158.1
(C-400, C-600), 160.5 (C-4000), 164.1 (CO2), 175.3 (C-200) ppm. MS
(ESI): m/z ¼ 759.6 [M ꢁ Br]+, 591.4, 439.4, 329.3. HRMS (ESI):
m/z calcd for C48H79N4O3+: 759.6147, found: 759.6152 [M ꢁ
1-{4-[3-(5-Dodecylpyrimidin-2-yl)phenoxy]octyl}-3-
methylimidazolium bromide (m-5(12,8))
From m-17(12,8) (0.081 mmol, 43.0 mg) and 13a (0.81 mmol,
67.0 mg), 72 h; yield: 35.0 mg (71%). Found: C, 64.83; H, 8.83; N,
8.54%. C34H53BrN4O2 requires C, 66.54; H, 8.70; N, 9.13%; M,
613.71. FT-IR (ATR): nmax/cmꢁ1 3081 (w), 2953 (w), 2919 (s),
2849 (m), 1608 (w), 1581 (m), 1542 (m), 1469 (m), 1447 (m), 1426
(s), 1395 (w), 1328 (m), 1296 (m), 1252 (w), 1234 (w), 1205 (m),
1169 (m), 1087 (w), 1045 (m), 948 (w), 902 (w), 882 (w), 825 (w),
811 (w), 778 (m), 755 (w), 722 (m), 692 (m), 651 (w), 619 (w), 561
(w), 544 (w), 520 (w). dH (500 MHz, CDCl3) 0.81 (t, 3H, J ¼ 6.5
Hz, CH3), 1.03–1.47 (m, 6H, CH2), 1.80–1.94 (m, 2H, CH2),
2.49–2.63 (m, 2H, 5000-CH2), 3.99 (t, 2H, J ¼ 6.4 Hz, OCH2), 4.05
(s, 3H, NCH3), 4.25 (t, 2H, J ¼ 7.4 Hz, NCH2), 6.89–7.00 (m, 1H,
600-H), 7.13–7.24 (m, 2H, 4-H, 5-H), 7.31 (t, 1H, J ¼ 7.9 Hz, 500-
H), 7.83–7.99 (m, 2H, 200-H, 400-H), 8.55 (s, 2H, 4000-H, 6000-H),
10.62 (s, 1H, 2-H) ppm. dC (125 MHz, CDCl3) 14.1 (CH3), 22.7,
25.9, 26.1, 28.8, 29.0, 29.1, 29.2, 29.3, 29.5, 29.6, 30.2, 30.8, 31.9
(CH2), 36.8 (NCH3), 50.2 (CH2N), 67.9 (OCH2), 113.1 (C-200),
117.3 (C-600), 120.3, 121.4, 123.0 (C-400, C-4, C-5), 129.6 (C-500),
133.2 (C-300), 138.3, 139.1 (C-2, C-5000), 157.0 (C-4000, C-6000), 159.5
(C-100), 162.2 (C-2000) ppm. MS (ESI): m/z ¼ 533.4 [M ꢁ Br]+.
HRMS (ESI): m/z calcd for C30H53N4O+: 533.4214, found:
533.4204 [M ꢁ Br]+. DSC: Cr 81 ꢀC [3.4 kJ molꢁ1] SmA 185 ꢀC
[0.6 kJ molꢁ1] I (2nd heating), I 184 ꢀC [ꢁ0.7 kJ molꢁ1] SmA (2nd
cooling); no recrystallization was observed upon cooling.
Br]+. DSC: Cr1 31 ꢀC [7.9 kJ molꢁ1] Cr2 101 ꢀC [18.2 kJ molꢁ1
]
SmA 172 ꢀC [2.6 kJ molꢁ1] I (2nd heating); I 172 ꢀC [ꢁ2.3 kJ
molꢁ1] SmA 76 ꢀC [ꢁ14.4 kJ molꢁ1] 36 ꢀC [ꢁ8.3 kJ molꢁ1] (2nd
cooling).
General procedure for the preparation of
trifluoromethanesulfonates 24(m)–26(m)
To a solution of the respective compounds 21(m)–23(m) (1.00
mmol) in acetonitrile (5 mL) at room temperature was added
potassium trifluoromethanesulfonate (1.10 mmol), whereby
colourless KBr precipitated. The reaction mixture was stirred for
2 h (tlc control). After completion of the reaction, the solvent was
removed under vacuum. The residue was taken up in CH2Cl2 (5
mL) and filtered. The products were obtained as colourless solids
without further purification.
3-Methyl-1-(8-{[(2-tetradecylpyrimidin-5-yl)carbonyl]oxy}octyl)
imidazolium bromide (21(8))
From 20(8) (0.16 mmol, 80.0 mg) and 13a (1.60 mmol, 131 mg),
72 h; yield: 50.0 mg (49%). Found: C, 61.42; H, 8.88; N, 9.04%.
C31H53BrN4O2 requires C, 62.72; H, 9.00; N, 9.44%; M, 593.34.
FT-IR (ATR): nmax/cmꢁ1 3068 (w), 2923 (s), 2853 (m), 1723 (s),
1590 (s), 1551 (m), 1466 (m), 1442 (m), 1390 (m), 1290 (s), 1169
(m), 1134 (m), 1037 (w), 909 (s). dH (500 MHz, CDCl3) 0.88
(t, 3H, J ¼ 7.1 Hz, CH3), 1.21–1.46 (m, 24H, CH2), 1.73–1.87
(m, 4H, CH2), 1.90–1.99 (m, 2H, CH2), 2.99–3.05 (m, 2H, 2-
CH2), 4.14 (s, 3H, CH3), 4.35 (t, 2H, J ¼ 6.9 Hz, CH2N), 7.29–
7.39 (m, 2H, 4-H, 5-H), 9.18 (s, 2H, 40-H, 60-H), 10.67 (s, 1H, 2-
H) ppm. dC (125 MHz, CDCl3) 14.1 (CH3), 22.7, 25.8, 26.2, 28.5,
28.9, 29.36, 29.38, 29.41, 29.5, 29.6, 29.65, 29.68, 29.69, 30.3,
31.9, 36.8 (CH2), 39.8 (20-CH2), 50.2 (CH2N), 65.6 (OCH2),
121.4, 121.5 (C-4, C-5), 123.1 (C-50), 138.3 (C-1), 158.1 (C-40, C-
60), 164.1 (CO2), 175.3 (C-20) ppm. MS (ESI): m/z ¼ 513.4 [M ꢁ
Br]+, 321.3, 193.2, 83.1. HRMS (ESI): m/z calcd for
C31H53N4O2+: 513.4163, found: 513.4153 [M ꢁ Br]+. DSC: Cr
42 ꢀC [53.0 kJ molꢁ1] SmA 219 ꢀC [1.0 kJ molꢁ1] I (1st heating);
decomposition upon clearing.
3-Methyl-1-(8-{[(2-tetradecylpyrimidin-5-yl)carbonyl]oxy}octyl)
imidazolium trifluoromethanesulfonate (24(8))
From 21(8) (0.068 mmol, 40.0 mg); yield: 45.0 mg (quant.). FT-
IR (ATR): nmax/cmꢁ1 2924 (s), 2853 (m), 1724 (s), 1590 (s), 1442
(m), 1391 (m), 1259 (s), 1161 (m), 1031 (w), 904 (s)1. dH (500
MHz, CDCl3) 0.88 (t, 3H, J ¼ 6.9 Hz, CH3), 1.19–1.47 (m, 28H,
CH2), 1.73–1.95 (m, 6H, CH2), 2.98–3.07 (m, 2H, 2-CH2), 4.00
(s, 3H, CH3), 4.21 (t, 2H, J ¼ 7.6 Hz, OCH2), 4.35 (t, 2H, J ¼
6.6 Hz, OCH2), 7.24–7.32 (m, 2H, 4-H, 5-H), 9.17 (s, 2H, 40-H,
60-H), 9.28 (s, 1H, 2-H) ppm. dC (125 MHz, CDCl3) 14.1 (CH3),
22.7, 25.7, 26.1, 28.5, 28.6, 28.8, 28.9, 29.36, 29.38, 29.42, 29.5,
29.6, 29.65, 29.69, 30.0, 31.9, 36.6 (CH2), 39.8 (20-CH2), 50.2
(CH2N), 65.6 (OCH2), 121.5, 121.8 (C-4, C-5), 121.9 (CF3),
123.3 (C-50), 137.4 (C-1), 158.1 (C-40, C-60), 164.1 (CO2), 175.3
(C-20) ppm. MS (ESI, positive): m/z ¼ 513.4 [M ꢁ CF3SO3]+,
321.3, 193.2, 83.1. MS (ESI, negative): m/z ¼ 149.0 [CF3SO3]ꢁ,
+
99.0, 80.0. HRMS (ESI, positive): m/z calcd for C31H53N4O2
:
513.4163, found: 513.4177 [M ꢁ CF3SO3]+. HRMS (ESI,
1-(8-{[(2-Tetradecylpyrimidin-5-yl)carbonyl]oxy}octyl)-3-[4-
(dodecyloxy)phenyl]imidazolium bromide (23(8))
negative):ꢁm/z calcd for CF3SO3ꢁ: 148.9515, found: 148.9527
ꢁ1
ꢀ
ꢀ
[CF3SO3] . DSC: Cr1 9 C [12.8 kJ mol ] Cr2 32 Cꢀ[19.9 kJ
molꢁ1] SmA 79 C [0.3 kJ molꢁ1] I ꢁ(21nd heating), I 84 C [ꢁ0.3
ꢀ
From 20(8) (0.14 mmol, 70.0 mg) and 13d (0.27 mmol, 90.0 mg),
72 h; yield: 105 mg (91%). Found: C, 68.61; H, 9.25; N, 6.61%.
C48H79BrN4O3 requires C, 68.63; H, 9.48; N, 6.67%; M, 840.07.
kJ molꢁ1] SmA 35 ꢀC [ꢁ20.3 kJ mol ] Cr2 1 ꢀC [ꢁ6.6 kJ molꢁ1
]
Cr1 (2nd cooling).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 21987–21997 | 21995

View Article Online
1-(8-{[(2-Tetradecylpyrimidin-5-yl)carbonyl]oxy}octyl)-3-[4-
(dodecyloxy)phenyl]imidazolium trifluoromethanesulfonate
(26(8))
X-ray investigations of selected compounds of both series,
2-phenylpyrimidines and 2-alkylpyrimidinecarboxylic acids,
suggest a general trend for the packing of the molecules within
the smectic mesophases depending on the substitution pattern of
the imidazolium head groups. Smectic bilayers comprised
compounds with N-methyl or N-butyl groups attached to the
imidazolium head groups while N-4-dodecyloxyphenyl and
N-dodecyl derivatives displayed smectic monolayers.
From 23(8) (0.071 mmol, 60.0 mg); yield: 65.0 mg (quant.). FT-IR
(ATR): nmax/cmꢁ1 2921 (s), 2852 (m), 1725 (s), 1591 (s), 1554 (m),
1514 (s), 1468 (m), 1391 (m), 1256 (s), 1160 (m), 1031 (w). dH (500
MHz, CDCl3) 0.87 (t, 3H, J ¼ 7.0 Hz, 2 ꢃ CH3), 1.18–1.51 (m,
48H, CH2), 1.73–1.87 (m, 6H, CH2), 1.89–1.99 (m, 2H, CH2),
2.98–3.06 (m, 2H, 2-CH2), 3.99 (t, 2H, J ¼ 6.3 Hz, CH2N), 4.35 (t,
2H, J ¼ 6.6 Hz, OCH2), 4.39 (t, 2H, J ¼ 7.6 Hz, CH2N), 7.01–7.06
(m, 2H, 3000-H, 5000-H), 7.40–7.51 (m, 2H, 2000-H, 6000-H), 7.52–7.56
(m, 2H, 4-H, 5-H), 9.17 (s, 2H, 400-H, 600-H), 9.61 (s, 1H, 2-H) ppm.
dC (125 MHz, CDCl3) 14.1 (CH3), 22.7, 25.7, 26.0, 26.1, 28.5, 28.6,
28.8, 28.9, 29.1, 29.36, 29.42, 29.52, 29.57, 29.60, 29.65, 29.7, 30.2,
31.9 (CH2), 39.8 (200-CH2), 50.6 (CH2N), 65.6, 68.7 (OCH2), 116.1
(C-3000, C-5000), 121.1 (C-500), 121.4 (CF3), 121.3, 122.5 (C-4, C-5),
123.5 (C-2000, C-6000), 127.0 (C-1000), 135.4 (C-2), 158.1 (C-400, C-600),
160.7 (C-4000), 164.1 (CO2), 175.3 (C-200) ppm. MS (ESI, positive):
m/z ¼ 759.6 [M ꢁ CF3SO3]+, 591.4, 439.4, 329.3, 271.2, 161.1. MS
(ESI, negative): m/z ¼ 149.0 [CF3SO3]ꢁ, 99.0, 80.0. HRMS (ESI,
positive): m/z calcd for C48H79N4O3+: 759.6147, found: 759.6131
[M ꢁ CF3SO3]+. HRMS (ESI, negative): m/z calcd for CF3O3Sꢁ:
148.9515, found: 148.9523 [CF3SO3]ꢁ. DSC: Cr1 21 ꢀC [ꢁ17.7 kJ
In conclusion, the results revealed that for both 5- and
2-phenylpyrimidine ILCs bending of the mesogenic core unit, i.e.
meta rather than para substitution at the phenyl ring indeed
resulted in mesophases near to room temperature with sufficient
thermal stability. Furthermore, the carboxylate seems to be a
suitable surrogate for meta-substituted phenyl units in 2-phe-
nylpyrimidines leading also to significantly decreased melting
points particularly if triflate anions were used rather than
bromides. Future work must demonstrate whether bending of
the mesogenic core unit leads to a general approach towards
room temperature ILCs.
Acknowledgements
Generous financial support by the Deutsche For-
schungsgemeinschaft, the Fonds der Chemischen Industrie, the
ꢁ1
molꢁ1] Cr2 44 C [36.1 kJ mol ] Cr3 49 C [8.5 kJ molꢁ1] SmA
72 ꢀC [1.6 kJ molꢁ1] I (2nd heating); I 72 ꢀC [ꢁ1.7 kJ molꢁ1] SmA
29 ꢀC [ꢁ16.7 kJ molꢁ1] Cr2 17 ꢀC [ꢁ10.6 kJ molꢁ1] Cr1 (2nd
cooling).
ꢀ
ꢀ
€
Bundesministerium fur Bildung und Forschung (shared instru-
mentation grant), the Studienstiftung des Deutschen Volkes
€
(fellowship for M.B.) and the Ministerium fur Wissenschaft,
€
Forschung und Kunst des Landes Baden-Wurttemberg is
gratefully acknowledged.
Conclusions
References
Two series of new compounds, imidazolium bromides with
mesogenic 2-phenylpyrimidine p-5(n,m)–p-8(n,m) and 2-alkyl-
pyrimidinecarboxylic acid cores 21(m)–23(m), were investigated.
As a general trend the liquid crystalline 2-phenylpyrimidines p-
5(n,4)–p-8(n,4) with a C4 spacer exhibited smectic A mesophases.
Elongation of the spacer to C8 generally leads to a small decrease
of melting and a significant decrease of clearing temperatures.
Surprisingly two of the derivatives with the longer spacer (p-
5(10,8), p-5(12,8)) also displayed smectic C phases. 2-Phenyl-
pyrimidines p-5(n,m)–p-8(n,m) exhibited smectic mesophases
over a broad temperature range (up to DT ¼ 143 K), but the
melting and clearing points are quite high. We therefore changed
the molecular geometry from linear (para-substituted 2-phenyl-
pyrimidines p-5(n,m)–p-8(n,m)) to bent and synthesized a series
of meta-substituted 2-phenylpyrimidines m-5(12,m)–m-8(12,m)
with terminal C12 alkyl chains. The reduction of symmetry
delivered the desired effect as all compounds m-5(12,m)–m-
8(12,m) displayed lower melting and clearing temperatures
compared to their linear isomers p-5(12,m)–p-8(12,m).
1 Reviews on ILs: (a) Electrodeposition from Ionic Liquids, ed. F.
Endres, A. P. Abbott and D. R. MacFarlane, Wiley-VCH,
Weinheim, 2008; (b) Ionic Liquids IIIA: Fundamentals, Progress,
Challenges, and Opportunities, ed. K. R. Seddon and R. D. Rogers,
ACS Symp. Ser., American Chemical Society, Washington DC,
2005, vol. 901; (c) Ionic Liquids IIIB: Fundamentals, Progress,
Challenges, and Opportunities–Transformations and Processes, ed. K.
R. Seddon and R. D. Rogers, ACS Symp. Ser., American Chemical
Society, Washington DC, 2005, vol. 902; (d) Ionic Liquids in
Synthesis, ed. P. Wasserscheid and T. Welton, Wiley-VCH,
Weinheim, 2nd edn, 2007; (e) J. Dupont, R. F. de Souza and
P. A. Z. Suarez, Chem. Rev., 2002, 102, 3667–3692.
2 Reviews on LCs: Handbook of Liquid Crystals, ed. D. Demus, J.
Goodby, G. W. Gray, H. W. Spiess and V. Vill, Wiley-VCH,
Weinheim, 2008, vol. 1–3.
3 Reviews on ILCs: (a) K. V. Axenov and S. Laschat, Materials, 2011,
4, 206–259; (b) T. Kato, N. Mizoshita and K. Kishimoto, Angew.
Chem., 2006, 118, 44–74; Angew. Chem., Int. Ed., 2006, 45, 38–68;
(c) K. Binnemans, Chem. Rev., 2005, 105, 4148–4204.
4 R. P. Lemieux, Acc. Chem. Res., 2001, 34, 845–853.
5 Review: (a) M. Hird, Liq. Cryst., 2011, 38, 1467–1493; recent
examples: (b) A. Katsuragi, E. Inoue and T. Sasaki, Mol. Cryst.
Liq. Cryst., 2011, 548, 107–119; (c) K. Takeuchi, Y. Takanishi,
J. Yamamoto and A. Yoshizawa, Liq. Cryst., 2010, 37, 507–515; (d)
P. K. Mandel, S. Haldar, A. Lapanik and W. Haase, Jpn. J. Appl.
Phys., 2009, 48, 011501.
6 C. Adamo and V. Barone, J. Chem. Phys., 1998, 108, 664–675.
7 V. Barone, L. Commisso, F. Lelj and N. Russo, Tetrahedron, 1985,
41, 1915–1918.
8 (a) J. C. Roberts, N. Kapernaum, Q. Song, D. Nonnenmacher,
K. Ayub, F. Giesselmann and R. P. Lemieux, J. Am. Chem. Soc.,
2010, 132, 364–370; (b) J. C. Roberts, N. Kapernaum,
F. Giesselmann and R. P. Lemieux, J. Am. Chem. Soc., 2008, 130,
13842–13843; (c) J. C. Roberts, N. Kapernaum, F. Giesselmann,
M. D. Wand and R. P. Lemieux, J. Mater. Chem., 2008, 18,
The second investigated series, the 2-alkylpyrimidinecarbox-
ylic acid imidazolium bromides, exhibited smectic A mesophases,
most of them over a very broad temperature range (up to DT ¼
193 K). Unfortunately some of the compounds (21(m), 23(4))
decomposed at higher temperatures. To overcome this problem
the bromide anion was exchanged with a triflate anion in a salt
metathesis reaction. The resulting thermally stable derivatives
24(m)–26(m) exhibited smaller mesophase ranges but melted near
room temperature.
21996 | J. Mater. Chem., 2012, 22, 21987–21997
This journal is ª The Royal Society of Chemistry 2012

View Article Online
5301–5306; (d) M. P. Thompson and R. P. Lemieux, J. Mater.
Chem., 2007, 17, 5068–5076; (e) L. Li, C. D. Jones, J. Magolan
and R. P. Lemieux, J. Mater. Chem., 2007, 17, 2313–2318; (f)
M. P. Thompson, T. Hegmann, M. D. Wand and R. P. Lemieux,
Liq. Cryst., 2007, 34, 987–994; (g) C. S. Hartley, N. Kapernaum,
J. C. Roberts, F. Giesselmann and R. P. Lemieux, J. Mater.
Chem., 2006, 16, 2329–2337.
9 (a) C. Damm, G. Israel, T. Hegmann and C. Tschierske, J. Mater.
Chem., 2006, 16, 1808–1816; (b) B. Neumann, T. Hegmann,
C. Wagner, P. R. Ashton, R. Wolf and C. Tschierske, J. Mater.
Chem., 2003, 13, 778–784.
Meervelt, P. Pattison, K. Binnemans and T. Cardinaels, Langmuir,
ꢁ
2009, 25, 5881–5897; (c) R. G. Santos-Martell, A. Ceniceros-Olguın,
ꢁ
ꢁ
ꢁ
ꢁ
L. Larios-Lopez, R. J. Rodrıguez-Gonzalez, D. Navarro-Rodrıguez,
B. Donnio and D. Guillon, Liq. Cryst., 2009, 36, 787–797; (d)
Q. Zhang, C. Shan, X. Wang, L. Chen, L. Niu and B. Chen, Liq.
Cryst., 2008, 35, 1299–1305; (e) D. Tsiourvas, T. Felekis,
Z. Sideratou and C. M. Paleos, Liq. Cryst., 2004, 31, 739–744; (f)
S. Ujiie and K. Iimura, Chem. Lett., 1994, 17–20; (g) L. Cui,
V. Sapagovas and G. Lattermann, Liq. Cryst., 2002, 29, 1121–
1132.
18 Liquid Crystals: Experimental Study of Physical Properties and Phase
Transitions, ed. S. Kumar, Cambridge University Press, Cambridge,
2001.
10 G. Starkulla, M. Kaller, W. Frey, K. V. Axenov and S. Laschat, Liq.
Cryst., 2011, 38, 1515–1529.
11 (a) N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37,
123–150; (b) T. Payagala, J. Huang, Z. S. Breitbach, P. S. Sharma
and D. W. Armstrong, Chem. Mater., 2007, 19, 5848–5850; (c)
K. N. Marsh, J. A. Boxall and R. Lichtenthaler, Fluid Phase
Equilib., 2004, 219, 93–98.
19 J. P. F. Lagerwall and F. Giesselmann, Phys. Rev. E: Stat., Nonlinear,
Soft Matter Phys., 2002, 66, 031703.
20 S. Urban, J. PrzedmoJski and J. Czub, Liq. Cryst., 2005, 32, 619–624.
21 P. H. J. Kouwer and T. M. Swager, J. Am. Chem. Soc., 2007, 129,
14042–14052.
€
12 D. Lloyd, C. Reichardt and M. Struthers, Liebigs Ann. Chem., 1986,
1368–1379.
13 (a) C. E. Kwartler and P. Lucas, J. Am. Chem. Soc., 1943, 65, 354–
355; (b) Commercially available from Aurora Fine Chemicals Ltd.
14 (a) G. F. Starkulla, E. Kapatsina, A. Baro, F. Giesselmann,
22 S. Sauer, S. Saliba, S. Tussetschlager, A. Baro, W. Frey,
F. Giesselmann, S. Laschat and W. Kantlehner, Liq. Cryst., 2009,
36, 275–299.
23 Molecular modelling was done with Chem3D Pro 2008 software.
24 G. Starkulla, Ph.D. thesis, University of Stuttgart, 2012, submitted.
25 P. Zhichkin, D. J. Fairfax and S. A. Eisenbeis, Synthesis, 2002, 720–
722.
26 P. Schenone, L. Sansebastiano and L. Mosti, J. Heterocycl. Chem.,
1990, 27, 295–305.
27 G. Hofle, W. Steglich and H. Vorbruggen, Angew. Chem., 1978, 90,
602–615; Angew. Chem., Int. Ed., 1978, 17, 569–583.
€
S. Tussetschlager, M. Kaller and S. Laschat, Beilstein J. Org.
Chem., 2009, 5, 63; (b) E. Kapatsina, M. Lordon, A. Baro and
S. Laschat, Synthesis, 2008, 2551–2560.
15 V. Nair, D. E. Vietti and C. S. Cooper, J. Am. Chem. Soc., 1981, 103,
3030–3036.
16 (a) Commercially available from Alfa Aesar A Johnson Matthey
Company; (b) A. D. Turner, S. V. Pizzo, G. Rozakis and
N. A. Porter, J. Am. Chem. Soc., 1988, 110, 244–250.
€
€
€
28 M. Butschies, J. C. Haenle, S. Tussetschlager and S. Laschat, Liq.
Cryst., submitted.
17 Examples of ILC with SmC phases: (a) S. Kokmoto, Y. Hara and
K. Kishikawa, Tetrahedron Lett., 2010, 51, 1508–1511; (b)
K. Goossens, K. Lava, P. Nockemann, K. Van Hecke, L. Van
29 (a) S. Urban, J. Przedmojski and J. Czub, Liq. Cryst., 2005, 32, 619–
624; (b) J. P. F. Lagerwall, F. Giesselmann and M. D. Radcliffe, Phys.
Rev. E: Stat., Nonlinear, Soft Matter Phys., 2002, 66, 031703.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 21987–21997 | 21997