Pyrrolidinium ionic liquid crystals

Article abstract of DOI:10.1002/chem.200801566

N-Alkyl-N-methylpyrrolidinium cations have been used for the design of ionic liquid crystals, including a new type of uranium-containing metallomesogen. Pyrrolidinium salts with bromide, bis(trifluoromethylsulfonyl) imide, tetrafluoroborate, hexafluorophosphate, thiocyanate, tetrakis(2- thenoyltrifluoroacetonato)europate(III) and tetrabromouranyl counteranions were prepared. For the bromide salts and tetrabromouranyl compounds, the chain length of the alkyl group C_nH_2n+1 was varied from eight to twenty carbon atoms (n = 8, 10-20). The compounds show rich mesomorphic behaviour: highly ordered smectic phases (the crystal smectic E phase and the uncommon crystal smectic T phase), smectic A phases, and hexagonal columnar phases were observed, depending on chain length and anion. This work gives better insight into the nature and formation of the crystal smectic T phase, and the molecular requirements for the appearance of this highly ordered phase. This uncommon tetragonal mesophase is thoroughly discussed on the basis of detailed powder X-ray diffraction experiments and in relation to the existing literature. Structural models are proposed for self-assembly of the molecules within the smectic layers. In addition, the photophysical properties of the compounds containing a metal complex anion were investigated. For the uranium-containing mesogens, luminescence can be induced by dissolving them in an ionic liquid matrix. The europium-containing compound shows intense red photoluminescence with high colour purity.

Full text of DOI:10.1002/chem.200801566

DOI: 10.1002/chem.200801566
Pyrrolidinium Ionic Liquid Crystals
Karel Goossens, Kathleen Lava, Peter Nockemann, Kristof Van Hecke,
Luc Van Meervelt, Kris Driesen, Christiane Gçrller-Walrand, Koen Binnemans, and
Thomas Cardinaels*[a]
Abstract: N-Alkyl-N-methylpyrrolidini-
um cations have been used for the
design of ionic liquid crystals, including
a new type of uranium-containing me-
highly ordered smectic phases (the
crystal smectic E phase and the uncom-
mon crystal smectic T phase), smectic
nal mesophase is thoroughly discussed
on the basis of detailed powder X-ray
diffraction experiments and in relation
to the existing literature. Structural
models are proposed for self-assembly
of the molecules within the smectic
layers. In addition, the photophysical
properties of the compounds contain-
ing a metal complex anion were inves-
tigated. For the uranium-containing
mesogens, luminescence can be in-
duced by dissolving them in an ionic
liquid matrix. The europium-containing
compound shows intense red photolu-
minescence with high colour purity.
A
phases, and hexagonal columnar
A
phases were observed, depending on
chain length and anion. This work
gives better insight into the nature and
ACHTUNGTRENNUNG
phosphate, thiocyanate, tetrakis(2-
thenoyltrifluoroacetonato)europate(III)
and tetrabromouranyl counteranions
were prepared. For the bromide salts
and tetrabromouranyl compounds, the
chain length of the alkyl group CnH2n+1
was varied from eight to twenty carbon
atoms (n=8, 10–20). The compounds
show rich mesomorphic behaviour:
formation of the crystal smectic T
phase, and the molecular requirements
for the appearance of this highly or-
dered phase. This uncommon tetrago-
Keywords: ionic liquids
· liquid
crystals · metallomesogens · pyrroli-
dinium cations · uranium
Introduction
duction and are of interest as low-dimensional ion-conduc-
tive materials.[8,9] Recent studies have shown that ionic me-
Ionic liquid crystals are a fascinating class of molecular ma-
terials, both from fundamental and from applied points of
view.[1] The ionic nature of these compounds allows uncom-
mon mesophases like the crystal smectic T phase[2–6] and the
nematic columnar phase to be obtained.[7] Aligned samples
of ionic liquid crystals show pronounced anisotropic ion con-
ACHTUNGTRENtNGNU allomesogens can be used as an anisotropic reaction
medium for shape-selective synthesis of nanoparticles.[10–12]
Different types of ionic liquid crystals have been investigat-
ed, among others imidazolium,[13–18] pyridinium,[13,16,19–22]
quaternary ammonium[2–4,23] and quaternary phosphonium
salts.[24,25] Recently, we reported on nematic ionic liquid crys-
tals and a luminescent tetrakis b-diketonate lanthanidome-
AHCTUNGTRENNUNG
sogen with mesogenic imidazolium cations.[26] The liquid-
[a] K. Goossens, K. Lava, Dr. P. Nockemann, Dr. K. Van Hecke,
Prof. Dr. L. Van Meervelt, Dr. K. Driesen,
Prof. Dr. C. Gçrller-Walrand, Prof. Dr. K. Binnemans,
Dr. T. Cardinaels
crystalline behaviour of the ionic salts strongly depends on
the nature of both the cation and the anion, but also on the
length of the alkyl chain of the organic cation. Pyrrolidinium
salts are used as ionic liquids with a large electrochemical
window,[27–29] as surfactants,[30] and as mesostructuring agents
in the preparation of oriented thin films.[31] To the best of
our knowledge, no investigation of the mesomorphic proper-
ties of thermotropic liquid-crystalline pyrrolidinium salts has
appeared in the literature yet. However, the pyrrolidinium
core offers advantages for studying the metal-centred lumi-
nescence of ionic metallomesogens, because it shows no in-
convenient background fluorescence. Here we describe the
Katholieke Universiteit Leuven, Department of Chemistry
Celestijnenlaan 200F, PO Box 2404, 3001 Leuven (Belgium)
Fax : (+32)16 327992
E-mail: thomas.cardinaels@chem.kuleuven.be
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801566. It contains synthetic
procedures and characterisation data for UO2Br2·xH2O and 13; stack
column graphs showing the evolution of the transition temperatures
in the series 2a–12a and 2g–12g; powder X-ray diffraction data for
uranyl complexes 6g–12g; and powder X-ray diffractograms of 12a
and 12g in the SmA phase.
656
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

FULL PAPER
thermotropic liquid-crystalline properties of N-alkyl-N-
methylpyrrolidinium salts with bromide, bis(trifluoromethyl-
sulfonyl)imide, tetrafluoroborate, hexafluorophosphate, thio-
cyanate, tetrakis(2-thenoyltrifluoroacetonato)europateACTHNUTRGNE(NUG III)
and tetrabromouranyl counteranions. In addition, the lumi-
nescence properties of the mesogens that contain a metal
complex anion were studied.
Results and Discussion
Synthesis: Synthesis of the pyrrolidinium salts is outlined in
Scheme 1. Compounds 1a–12a were prepared by quaterni-
sation (Menschutkin reaction) of 1-methylpyrrolidine with
CnH2n+1Br (n=8, 10–20).[32] Yields were in the range of 72–
88% (except for 1a and 2a: 47 and 51%, respectively).
Under the experimental conditions used, 1a and 2a were
obtained as monohydrate salts, as confirmed by CHN ele-
mental analysis, while 3a–12a were obtained as water-free
compounds and stored in a desiccator. Compounds 10b–10e
were synthesised by a metathesis reaction between bromide
salt 10a and lithium bis(trifluoromethylsulfonyl)imide
(LiNTf2, Tf=SO2CF3), sodium tetrafluoroborate, potassium
Scheme 1. Synthesis of the pyrrolidinium compounds. i) CnH2n+1Br (n=8,
10–20), Ar, dry toluene, RT (1a and 2a) or 808C (3a–12a); ii) 10b: 10a,
LiNTf2, H2O/MeOH, 608C; ii) 10c: 10a, NaBF4, acetone, reflux; ii) 10d:
10a, KPF6, H2O/MeOH, 608C; ii) 10e: 10a, KSCN, Ar, acetone, 558C;
ii) 10 f: 10a, Htta, NaOH, EuCl3·6H2O, H2O/EtOH, 508C;
iii) UO2Br2·xH2O, EtOH, reflux.
hexafluorophosphate and potassium thiocyanate, respective-
ly. Compound 10 f was prepared by a reaction between bro-
mide salt 10a, 2-thenoyltrifluoroacetone (Htta), sodium hy-
droxide and europiumACTHNUGRTENUNG(III) chloride hexahydrate. Com-
pounds 1g–12g were synthesised by reaction of bromide
salts 1a–12a with uranyl bromide hydrate (UO2Br2·xH2O)
in refluxing ethanol. Due to its hygroscopic character, it is
difficult to determine the exact number of water molecules
x in the uranyl bromide salt precursor. The uranyl bromide
salt was carefully dried, and when weighing, x was assumed
to be zero so that compounds 1g–12g would not be conta-
minated with an excess of uranyl bromide. The uranyl com-
plexes were obtained as anhydrous yellow solids (as con-
firmed by CHN elemental analysis), although no special
Abstract in Dutch: Het N-alkyl-N-methylpyrrolidinium-
ACHTUNGTRENNUNGkation werd gebruikt voor de bereiding van ionische vloeiba-
re kristallen, waaronder een nieuw type uraniumhoudend me-
tallomesogeen. Er werden pyrrolidiniumzouten met bromide-,
bis(trifluoromethylsulfonyl)imide-, tetrafluoroboraat-, hexa-
fluorofosfaat-,
thiocyanaat-,
tetrakis(2-thenoyl-trifluoro-
acetonato)europaatACHTUNGTRENNUNG(III)- en tetrabromouranyl-anionen gesyn-
thetiseerd. In het geval van de bromidezouten en de tetrabro-
mouranylverbindingen werd de ketenlengte gevarieerd van
acht tot twintig koolstofatomen (n=8, 10, 11, 12 ,13, 14, 15,
16, 17, 18, 19, 20). De verbindingen vertonen een rijk mesofa-
segedrag, van hooggeordende smectische fasen (de kristal-
smectische E fase en de zeldzame kristal-smectische T fase)
over smectische A fasen tot hexagonaal kolomvormige fasen,
afhankelijk van de ketenlengte en het anion. Dit werk geeft
een beter inzicht in de aard en vorming van de kristal-smecti-
sche T fase, en de moleculaire vereisten voor het voorkomen
van deze hooggeordende fase. Er wordt een nauwkeurige be-
schrijving gegeven van de mesofasestructuur, op basis van ge-
detailleerde poeder X-stralendiffractie-experimenten en door
het verband te leggen met de bestaande literatuur omtrent
deze ongewone tetragonale mesofase. Er worden structurele
modellen voorgesteld die de zelforganisatie van de moleculen
in de smectische lagen verduidelijken. Daarnaast werden de
fotofysische eigenschappen van de verbindingen met een
anionisch metaalfragment onderzocht. Voor de uraniumhou-
dende metallomesogenen werd aangetoond dat luminescentie
geꢀnduceerd kan worden door deze op te lossen in een ioni-
sche vloeistof. De europiumhoudende verbinding vertoont
een intense rode fotoluminescentie met een hoge kleurzuiver-
heid.
1
care was taken to prevent hydration of the salts. The H and
13C NMR spectra of the uranyl complexes in CD3OD (not
reported here) are similar to those obtained for the parent
halide salts in CD3OD. The uranyl complexes were obtained
in relatively low yields because of their rather good solubili-
ty in ethanol, from which they were recrystallised, and the
difficulty of working with the hygroscopic uranyl bromide
salt precursor (see above).
Single-crystal X-ray diffraction: Crystals of uranyl complex
6g suitable for single-crystal X-ray structure determination
could be obtained by slowly evaporating a solution of the
compound in ethanol. The crystal structure consists of two
crystallographically independent N-methyl-N-tetradecylpyr-
rolidinium cations and a [UO2Br4]2ꢀ ion in the asymmetric
unit (Figure 1, top). In the first coordination sphere, the
uranyl ion is surrounded by four bromide ions in the equato-
rial plane. Weak interactions of the protons of the pyrrolidi-
nium cations with the bromine atoms (ranging from 2.75 to
3.04 ꢁ) and with the oxygen atoms (ranging from 2.43 to
2.59 ꢁ) of the [UO2Br4]2ꢀ ions are found (Figure 1, bottom).
In the packing of the crystal structure of 6g, the cationic
head groups and the [UO2Br4]2ꢀ anions are arranged in
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
657

T. Cardinaels et al.
Figure 2. Packing in the crystal structure of 6g, viewed along the a axis.
Figure 1. Top: asymmetric unit of the crystal structure of 6g. Bottom:
weak interactions of protons of the pyrrolidinium cations with the sur-
rounding anions, shown for one of the cations in the crystal structure of
6g.
weight loss, while 10g had not started to decompose yet.
The alkyl chain length does not seem to have any influence
on the thermal stability of the pyrrolidinium bromides. On
the other hand, the type of anion strongly determines the
thermal stability (Figure 4b). Bis(trifluoromethylsulfonyl)-
ionic double layers with the alkyl chains pointing in opposite
directions (Figure 2). These double layers are separated
from one another by the alkyl chains, which are arranged in
a parallel and interdigitated fashion. The horizontal U···U
distance within the double layers is 10.52 ꢁ, and the vertical
U···U distance is 8.57 ꢁ. The shortest U···U distances be-
tween two parallel double layers are 22.14 and 35.64 ꢁ be-
tween the inner and outer uranyl ions, respectively.
AHCTUNGTREGiNNUN mide compound 10b is thermally stable up to 3658C. The
thermal stability order of the singly charged anions can be
ꢀ
summarised as: NTf2ꢀ >PF6ꢀ >[Eu
(tta)4]ꢀ >Brꢀ ꢁBF4
>
SCNꢀ. Compound 10c shows a two-stage decomposition
process. The weight loss for compound 10 f occurring be-
tween 265 and 3408C corresponds to the gradual decomposi-
tion of the metal-containing anion [EuACTHNUTRGNEUNG
(tta)4]ꢀ.
Thermal behaviour: The thermal properties of all com-
pounds were examined by polarising optical microscopy
(POM), differential scanning calorimetry (DSC) and X-ray
diffraction on powder samples. The transition temperatures
and thermal data for all the pyrrolidinium salts are collected
in Table 1. Representative DSC traces of 6a, 10a and 10g
are shown in Figure 3. In general, the second heating run
was not analogous to the first one, because the pyrrolidini-
um compounds are not stable at high temperatures (above
ca. 2008C). This was confirmed by POM, by the fact that
the baseline of the DSC thermogram alters above approxi-
mately 2008C, and by thermogravimetry (Figure 4). If the
compounds were heated to only 1808C during the first heat-
ing run, a matching and reproducible cooling run and a re-
producible second heating run were obtained (Figure 5). Ac-
cording to thermogravimetry, the pyrrolidinium bromide
compounds are thermally less stable than the corresponding
uranyl complexes (Figure 4a). At 2508C, 10a showed 28%
For some compounds, two or more crystalline phases
were observed. This is not unusual, as crystal polymorphism
and the existence of conductive plastic crystal phases (rota-
tionally disordered phases) of pyrrolidinium-based ionic liq-
uids have been reported in the past.[27,29,33–35] A minimum
alkyl chain length of eleven carbon atoms is required for the
pyrrolidinium bromide salts to exhibit mesomorphism. In-
vestigation of the thermal properties of 1a and 2a was com-
plicated by their hygroscopicity and by the fact that differ-
ent crystal forms melt at different temperatures in the case
of samples without a thermal history. As a consequence,
crystals of a high-melting polymorph are scattered in the
melt of a lower-melting crystal polymorph when viewed by
POM. The first and second DSC heating runs differed from
one another when samples that had not been kept under an
inert atmosphere were measured in DSC pans with a small
hole in the lid. The transition temperatures for the third and
fourth heating run were identical to the values observed for
658
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
Table 1. Transition temperatures and thermal data for the pyrrolidinium salts.
Compd n[a] Anion
Transition[b] T [8C][c] DH [kJmolꢀ1
]
Compd n[a] Anion
Transition[b] T [8C][c] DH [kJmolꢀ1
(DS [JKꢀ1 molꢀ1])
]
A
ACHTUNGTRENNUNG
1a
2a
8
Brꢀ
Cr!I
159
7.1 (16)
1g
2g
8
10
N
67
57
84
61
108
45
11.7 (34)
4.0 (12)
5.8 (16)
11.6 (35)
7.0 (18)
48.9 (154)
5.4 (14)
50.0[e]
[d]
10 Brꢀ
Cr1!Cr2
Cr2!I
Cr!T1
T1!T2
T2!I
41 18.6 (59)
162
27 10.2 (34)
70
6.7 (15)
Cr2!I
Cr2!I
Cr2!I
Cr2!E
3a
4a
5a
11 Brꢀ
4g
5g
6g
12
13
14
0.5 (1.4)
5.7 (13)
171
12 Brꢀ
13 Brꢀ
Cr!T1
T1!T2
T2!I
53 32.7 (100)
78
109
49
[d]
0.8 (2)
6.7 (14)
191
55
[h]
[h]
Cr1!Cr2
Cr2!T1
T1!T2
T2!I
53 13.8[e]
G
–
–
59
79
201
133
60
6.4 (16)
74.2 (223)
0.9 (3)
7.6 (16)
7g
8g
9g
15
16
17
[h]
[h]
E
–
–
[d]
6a
7a
8a
14 Brꢀ
15 Brꢀ
16 Brꢀ
Cr!T1
62 41.6 (124)
80
208
147
64
6.7 (16)
T1!T2
T2!I
0.9 (3)
8.7 (18)
47.0 (139)
[h]
A
151[h]
160
69
–
Cr!T2
74 17.2 (50)
E!I
7.3 (17)
(T2!T1)
83[f]
214
0.9 (3)[f]
9.3 (19)
80.9[e]
T2!I
Cr2!E
72
[d]
[h]
[h]
Cr1!Cr2
Cr2!T2
66 31.4[e]
(I!Colh)
–
–
75
77[f]
217
38[g]
82 24.9 (70)
218
42
E!I
169
62
72
172
75
80
185
227
41
69
83
7.6 (17)
11.2 (33)
41.2 (118)
10.0 (22)
96.8[e]
(T2!T1)
0.8 (2)[f]
9.7 (20)
1.1 (4)
10g
11g
18
19
T2!I
Cr2!E
E!I
9a
17 Brꢀ
18 Brꢀ
Cr1!Cr2
Cr2!T
T!I
9.3 (19)
2.4 (8)
Cr2!E
10a
Cr1!Cr2
Cr2!T
E!SmA
SmA!I
9.2 (20)
0.5 (1.1)
2.8 (9)
11.2 (33)
50.2 (141)
8.5 (19)
0.8 (2)
82 26.4 (74)
217
T!SmA
SmA!I
Cr1!Cr2
Cr2!T
T!SmA
SmA!I
Cr!T
8.1 (17)
1.6 (3)
1.1 (4)
12g
20
229
Cr2!Cr3
Cr3!E
11a
12a
19 Brꢀ
35[g]
89 29.3 (81)
216
233
89 32.7 (90)
214
E!SmA
SmA!I
186
267
8.3 (17)
1.5 (3)
20 Brꢀ
T!SmA
SmA!I
Cr1!Cr2
Cr2!I
8.9 (18)
1.3 (3)
1.5 (5)
233
ꢀ
10b
10c
18 NTf2
34[g]
72 32.3 (94)
47 7.8 (24)
92 20.8 (57)
ꢀ
18 BF4
Cr1!Cr2
Cr2!T
T!I
166
5.5 (13)
ꢀ
10d
10e
18 PF6
18 SCNꢀ
Cr!T
95 33.1 (90)
145
37
86 29.9 (83)
164
T!I
4.7 (11)
3.6 (12)
Cr1!Cr2
Cr2!SmA
SmA!I
0.9 (2)
10 f
18 [Eu
(tta)4]ꢀ Cr!I
121 46.5 (118)
[a] n=number of carbon atoms in the alkyl chain. [b] Abbreviations: Cr, Cr1, Cr2 and Cr3 =crystalline or partially crystalline phase; T, T1 and T2 =crystal
smectic T phase; E=crystal smectic E phase; SmA=smectic A phase; Colh =hexagonal columnar phase; I=isotropic liquid. Transitions in parentheses
indicate transitions to a monotropic mesophase. [c] Onset temperatures obtained by DSC at heating/cooling rates of 108Cminꢀ1 (He atmosphere). For
1a–12a, 10c and 10d, values were taken from the second heating run (first heating run up to 1808C). For the other compounds, values were taken from
the first heating run. During the first cooling run, 1a, 2a and 1g–12g were cooled to ꢀ508C, and the other compounds to ꢀ208C. [d] Before this endo-
thermic transition, an exothermic recrystallisation process took place. For 2a: at 98C (peak temperature) (DH=ꢀ15.2 kJmolꢀ1; DS=ꢀ54 JKꢀ1 molꢀ1);
for 4a: at 238C (peak temperature) (DH=ꢀ9.7 kJmolꢀ1; DS=ꢀ33 JKꢀ1 molꢀ1); for 6a: at 338C (peak temperature) (DH=ꢀ18.5 kJmolꢀ1; DS=
ꢀ60 JKꢀ1 molꢀ1); for 8a: at 538C (peak temperature) (DH=ꢀ5.6 kJmolꢀ1; DS=ꢀ17 JKꢀ1 molꢀ1). [e] As the transition peaks in the DSC thermogram
were not fully resolved, the total enthalpy value DH for both transitions is given. [f] Values taken from the first cooling run (first heating run up to
1808C). [g] Peak temperature. [h] Not detected by DSC. If possible, the temperature at which the monotropic mesophase transformed into the isotropic
liquid phase on heating was determined by POM.
the second heating run. When the compounds were dried in
vacuo at 1808C (i.e., in their liquid state) for three hours
and put in completely sealed DSC pans in a glove box, the
first and second DSC heating run were almost identical to
one another (slightly lower melting point for the second
heating run) and the same peaks as for the second heating
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
659

T. Cardinaels et al.
Figure 4. Thermograms of selected pyrrolidinium salts (heating rate of
108Cminꢀ1; N2 atmosphere). a) 4a, 10a and 10g; b) 10b, 10c, 10d, 10e
and 10 f.
sition, as indicated by the change in colour from white to
pale brown.
Bromide salts 3a–12a showed one or more enantiotropic
smectic phases. Compounds 3a, 4a, 5a and 6a (n=11, 12,
13 and 14, respectively) formed two types of crystal smectic
T phases (for a discussion on the T phase, see the section
about X-ray diffraction experiments), designated as T1 and
T2 for the lower- and higher-temperature phase, respectively.
The T1 and T2 phases appeared both on heating and on cool-
ing (Figure 5a). Compounds 7a and 8a (n=15 and 16, re-
spectively) showed only one type of enantiotropic T phase
(designated as T2). However, on cooling, another T phase
(designated as T1) appeared beneath the enantiotropic T
phase, as indicated by the appearance of an additional,
small peak in the DSC thermograms. Compound 9a (n=17)
exhibited only one type of T phase. Compounds 10a, 11a
and 12a (n=18, 19 and 20, respectively) showed a SmA
phase at higher temperatures and a T phase at lower tem-
Figure 3. DSC traces of a) 6a, b) 10a and c) 10g (heating/cooling rate of
108C minꢀ1; He atmosphere). The first heating/cooling cycle is shown by
a solid line, and the second heating/cooling cycle by a dashed line. Ab-
breviations: Cr, Cr1 and Cr2 =crystalline or partially crystalline phase; T,
T1 and T2 =crystal smectic T phase; E=crystal smectic E phase; SmA=
smectic A phase; I=isotropic liquid. Endothermic peaks point upwards.
run of the first series of measurements were observed. The
melting temperatures were, however, lower (by 3–58C) after
the drying procedure. This might be due to slight decompo-
660
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
The SmA phases could be easily identified by their defect
textures: focal conic fan and oily streak textures were ob-
served (Figure 6, top right and middle left). For both the
SmA and T phases, observation of large homeotropic do-
mains confirmed that they are orthogonal, optically uniaxial
mesophases. For the more viscous T phases, the observed
lancet-like texture (Figure 6, top left) or mosaic texture
points at a highly ordered smectic phase. On the basis of
POM observations alone, these phases could not be un-
equivocally assigned as T phases (e.g., SmB and crystal B
phases also show a lancet-like texture). X-ray diffraction
measurements on powder samples were required for full
characterisation of these phases (see below). A completely
non-birefringent texture such as reported by Nogami
et al.[41] and Ohta et al.[5] for the T phases of 1,4-dialkyl-1,4-
diazoniabicycloACTHNUTRGNEUNG[2.2.2]octane dibromides was not observed
for the T phases of the pyrrolidinium compounds described
here (except for compound 10c). Tetrabromouranyl salts
6g–12g showed viscous mesophases for which lancet-like
and mosaic textures, as well as large pseudo-homeotropic
domains, were observed (Figure 6, middle right and bottom
left). For the monotropic Colh phase of 6g-9g, a pseudo
focal conic texture was seen (Figure 6, bottom right). In the
case of 6g and 9g, the Colh phase was observed upon cool-
ing over a narrow temperature range for a very short period
of time, while it was more stable in the case of 7g and 8g.
A contact sample of compound 8a and N,N-bis(hexadec-
yl)-N,N-dimethylammonium bromide (13) was prepared by
allowing a small droplet of 8a in its T phase and a small
droplet of 13 in its isotropic liquid state to mix under a
cover slip at 1808C. Compound 13 was one of the first com-
pounds that was reported to exhibit a T phase (see below).[2]
In the first heating run, it shows a crystal-to-crystal transi-
tion at 648C (DH=5.6 kJmolꢀ1; DS=17 JKꢀ1 molꢀ1); its
melting point and clearing point are 76 and 1718C, respec-
tively (DHm =54.8 kJmolꢀ1
; ; DHc =
DSm =157 JKꢀ1 molꢀ1
Figure 5. DSC traces of a) 6a and b) 10a (heating/cooling rate of 108C
minꢀ1; He atmosphere). The first heating/cooling cycle is shown by a
solid line, and the second heating/cooling cycle by a dashed line. Abbre-
viations: Cr1 and Cr2 =crystalline or partially crystalline phase; T, T1 and
T2 =crystal smectic T phase; SmA=smectic A phase; I=isotropic liquid.
Endothermic peaks point upwards.
7.6 kJmolꢀ1; DSc =17 JKꢀ1 molꢀ1).[42] The interface between
the two materials was studied by POM on cooling the con-
tact sample from the isotropic liquid. Interestingly, the two
compounds mixed homogeneously, and at the interface a
SmA phase was formed (Figure 7). Apparently, the two dif-
ferent mesogens, both showing a T phase in their pure form,
cannot form this highly ordered phase when they are mixed;
a disordered SmA phase is formed instead. The difference
in molecular structure (one chain per cation for 8a; two
chains per cation for 13) and the difference in molecular
packing in the T phases shown by both mesogens (ionic
double layers for 8a (see below); ionic monolayers for 13[2])
preclude ordering of a mixture of the compounds in a
square two-dimensional lattice.
peratures. The bis(trifluoromethylsulfonyl)imide salt 10b
was not liquid-crystalline. The tetrafluoroborate salt 10c and
the hexafluorophosphate salt 10d showed a T phase, while
the thiocyanate salt 10e showed a SmA phase. Of the com-
pounds containing metal complex anions, the tetrakis(2-
thenoyltrifluoroacetonato)europateACTHNUTRGNEUNG(III) salt 10 f was not
liquid-crystalline, while the tetrabromouranyl salts with a
chain length of at least fourteen carbon atoms (6g–12g)
showed crystal smectic E phases. For compounds 6g–9g
(n=14–17), an additional monotropic hexagonal columnar
phase (Colh) was observed on cooling from the isotropic
liquid. Compounds 11g and 12g (n=19 and 20, respectively)
showed an additional SmA phase above the highly ordered
smectic phase. Only a very limited number of uranium-con-
taining liquid crystals have been reported till now.[36–40]
In the series 3a–12a and 6g–12g, the melting and clearing
temperatures increase on going from shorter to longer chain
lengths. Such behaviour was also observed for mesomorphic
imidazolium and pyridinium salts.[13,14,16,17,19,22] However,
compared to the imidazolium and pyridinium salts, which
only exhibit SmA phases, our pyrrolidinium salts show a
rich polymorphism, ranging from several types of smectic
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
661

T. Cardinaels et al.
Figure 6. Top left: lancet-like texture of the T2 phase of 6a at 1658C (100ꢂ magnification). Top right: focal conic fan texture of the SmA phase of 12a at
2158C (200ꢂ magnification). Middle left: oily streak texture of the SmA phase of 10e at 1308C (200ꢂ magnification). Middle right: lancet-like texture
of the E phase of 6g at 1158C (100ꢂ magnification). Bottom left: mosaic texture of the E phase of 6g at 998C (200ꢂ magnification). Bottom right:
pseudo focal conic texture of the Colh phase of 8g at 1508C (100ꢂ magnification).
ꢀ
phases to even a (monotropic) Colh phase for some uranyl
complexes. The difference can be attributed to the different
shapes of the cationic head groups, the absence of charge
delocalisation for pyrrolidinium cations and the absence of
strong hydrogen bonding for N-alkyl-N-methylpyrrolidinium
halides. Although short contacts exist in the crystalline solid
state of N-methyl-N-propylpyrrolidinium iodide (between
the halide anions and the b-protons of the pyrrolidinium cat-
ions), these interactions cannot be assigned as true hydrogen
bonds because of the large distance between them.[43] On
the other hand, the presence of non-classical C H···X hy-
drogen bonding in halide salts with heteroaromatic cations
has been described by several authors in the past.[44–47] The
fact that a minimum alkyl chain length is required to induce
mesomorphism has been commonly observed for ionic
liquid crystals.[1] In the series 1a–12a, an alkyl chain length
of at least eleven carbon atoms is needed to obtain micro-
phase separation between the ionic head groups and the hy-
drophobic alkyl chains, which leads to the formation of
(highly ordered and disordered) lamellar phases. The micro-
662
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
shorter chain lengths than fourteen carbon atoms. It was ob-
served that in the homologous series of [Cnmim]2ACHTUNGTRENNUNG[PdCl4]
complexes ([Cnmim]+ =1-methyl-3-alkylimidazolium) a min-
imum of 14 carbon atoms (n=14) is required to show
liquid-crystalline properties, while [C8C8im]2ACHTUNTGRNE[NUG PdCl4] already
shows a SmA phase over a broad temperature range.[50,51]
In the series 10a–10g, the anion clearly has great influ-
ence on the thermal behaviour.[14,52] The large NTf2 and
ꢀ
[EuACHTNUTGRNEUNG
(tta)4]ꢀ anions (compounds 10b and 10 f, respectively)
destroy the liquid-crystalline properties, while for the small-
er anions, the anion type has a large effect on the mesomor-
phic behaviour (mesophase type and transition tempera-
tures). Compound 10a exhibits both T and SmA phases,
while compounds 10c and 10d only exhibit a T phase. Fur-
thermore, when the bromide anion is replaced by a thiocya-
nate anion, as in compound 10e, the highly ordered T phase
is replaced by a disordered SmA phase. Apparently, the
larger, non-spherical thiocyanate anion is not suitable for as-
sembly in an ordered, square, two-dimensional network (see
below). Combination of the N-methyl-N-octadecylpyrrolidi-
nium cation with a [UO2Br4]2ꢀ dianion, as in 10g, leads to
formation of a crystal smectic E phase.
Figure 7. Contact preparation of 8a and 13 at 1638C (200ꢂ magnifica-
tion) after cooling from 1808C. Lower part: 8a in its T phase. Upper
part: 13 in its T phase. Middle part (interface between the two T phases):
SmA phase (mostly homeotropically aligned; a focal conic fan texture
appears when the sample is compressed with a needle).
phase separation is a result of the strong van der Waals in-
teractions between the long aliphatic alkyl chains.
For compounds 10a, 11a and 12a, a SmA phase is formed
above the T phase. Such a phase sequence was also ob-
served for N,N-bis(2-hydroxyethyl)-N-methyl-N-alkylammo-
nium bromides.[4] However, for the latter compounds, the
SmA phase was observed for a whole series of compounds
(n=12, 14, 16 and 18 carbon atoms), whereas for our pyrro-
lidinium compounds, the SmA phase is only seen for the
long-chain alkyl analogues (n=18, 19 or 20 carbon atoms).
The T–SmA transition is a first-order transition which is ac-
Powder X-ray diffraction: Most of the liquid-crystalline pyr-
rolidinium compounds were studied by X-ray diffraction on
powder samples for further identification of the enantiotrop-
ic mesophases and to obtain more information about the
molecular packing in the mesophases. Table 2 gives an over-
view of the Bragg reflections collected from the X-ray dif-
fractograms of compounds 4a, 6a, 8a, 10a, 12a, 10c, 10d
and 10e. Diffraction data for the uranyl complexes 6g-12g
are given in the Supporting Information.
companied by
a
considerable enthalpy change (8–
9 kJmolꢀ1). It appears that (very) long alkyl chains are not
compatible with an ordered packing of the ions in the ionic
sublayers at higher temperatures, and hence a disordered
SmA phase results (see below).
Before describing and discussing the X-ray diffractograms
of the mesophases formed by our pyrrolidinium compounds,
we briefly discuss the nature of the T phase. The T phase is
a highly ordered mesophase which has not often been ob-
served and which has an unusual symmetry. The T phase
was first observed in 1993 by Skoulios and co-workers, when
they examined the thermal behaviour of N,N-dialkyl-N,N-di-
methylammonium bromides.[2] On the basis of powder X-ray
diffraction measurements, they concluded that, within the
smectic layers, the lateral packing of the cationic head
groups and bromide anions was ordered and tetragonal in
symmetry. Therefore, the mesophase was named smectic T
(according to Przedmojski and Dynarowicz-Ła¸tka, this
phase was only shown by the homologues with two hexadec-
yl or two octadecyl chains; the compounds with two decyl,
two dodecyl or two tetradecyl chains were said to exhibit
crystal smectic E phases[53]). More specifically, the ammoni-
um and bromide ions were arranged periodically in a square
two-dimensional network, and the ionic sublayers were sep-
arated from one another by sublayers containing the disor-
dered alkyl chains. The T phases are rare, and until now
they have only been observed for ionic liquid crystals: oxy-
nitrostilbene derivatives of N,N-dialkyl-N,N-dimethylammo-
nium bromides,[3] N,N-bis(2-hydroxyethyl)-N-methyl-N-alkyl-
For the series 6g–9g, the appearance of columnar (besides
smectic) mesomorphism was surprising. However, a Colh
phase was also observed for several N-alkylpyridinium salts
with [CuCl4]2ꢀ as counteranion (n=12, 13 or 14 carbon
atoms).[22] The octahedral [UO2Br4]2ꢀ dianion has a different
coordination geometry compared to the tetrahedral
[CuCl4]2ꢀ dianion and consequently has a different shape
and size. The importance of the coordination geometry is il-
lustrated by the fact that N-alkylpyridinium salts of the
square-planar [PdCl4]2ꢀ dianion only show smectic meso-
phases, while the corresponding [CuCl4]2ꢀ salts exhibit col-
umnar, cubic and smectic A mesophases.[22,48] It was also
found that salts of the [CoCl4]2ꢀ, [NiCl4]2ꢀ, [CuCl4]2ꢀ,
[ZnCl4]2ꢀ and [CdCl4]2ꢀ complex anions (tetrahedral coordi-
nation geometry) with N-hexadecylpyridinium cations have
similar transition temperatures and mesophase temperature
ranges, which on the other hand differ from those of the cor-
responding [PdCl4]2ꢀ salt.[13,49] Although the pyrrolidinium
compounds with two long alkyl chains per cation have not
been synthesised yet, it can be expected that the corre-
sponding uranyl complexes would be mesomorphic for
ammonium
bromides,[4]
1,4-dialkyl-1,4-diazoniabicyclo-
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
663

T. Cardinaels et al.
Table 2. Bragg reflections collected from the X-ray diffractograms of the different enantiotropic mesophases.
Compd
dmeas [ꢁ][a]
I[b]
hkl[c]
dcalcd [ꢁ][a]
Parameters of
smectic
Compd
dmeas [ꢁ][a]
I[b]
hkl[c]
dcalcd [ꢁ][a]
Parameters of
smectic
phase[d–g]
phase[d–g]
4a
27.80
–
VS
001
002
003
004
005
006
110
111
200
210
001
002
003
004
005
006
110
111
200
210
001
002
003
004
005
006
110
111
200
210
001
002
003
004
005
006
110
111
200
210
001
002
003
004
005
006
110
111
200
210
001
002
003
004
005
006
007
008
110
27.75
T1: T=708C
12a
38.99
19.50
–
VS
M
001
002
003
004
005
006
007
008
110
200
210
001
002
38.92
19.46
T: T=1508C
[h]
V
A
M =574 ꢁ3
V
M =811 ꢁ3
[h]
9.25
6.93
5.55
4.62
M
M
M
W
9.25
6.94
5.55
4.63
M =41.4 ꢁ2
A
M =41.6 ꢁ2
a=6.38 ꢁ
L=29.32 ꢁ
L=19.13 ꢁ
9.72
7.77
6.48
5.56
4.87
4.51
3.20
2.85
35.74
17.98
M
M
M
M
W
M
W
W (br)
VS
W
9.73
7.78
6.49
5.56
4.87
4.51
3.19
2.85
35.85
17.92
[i]
–
–
–
–
[i]
[i]
[i]
27.29
VS
27.28
T2: T=1308C
[h]
–
V
A
M =599 ꢁ3
SmA: T=2008C[j]
9.08
6.82
5.46
4.55
4.52
4.46
M
M
M
W
W
W
9.09
6.82
5.46
4.55
4.52
4.46
M =44.0 ꢁ2
V
M =839 ꢁ3
a=6.39 ꢁ
L=19.13 ꢁ
A
M =46.8 ꢁ2
L=29.32 ꢁ
10c
32.63
16.32
10.88
8.17
6.54
5.46
4.88
4.84
4.68
4.47
4.20
3.94
3.62
3.45
VS
M
W
M
M
W
M
M
W
W
W
W
W
W
001
002
003
004
005
006
110
111
112
113
114
115
116
200
210
001
002
003
004
005
110
006
111
112
113
114
115
200
210
001
32.67
16.34
10.89
8.17
6.53
5.45
4.88
4.82
4.67
4.45
4.19
3.91
3.63
3.45
T: T=1308C
V
A
M =762 ꢁ3
M =46.6 ꢁ2
[i]
–
–
a=6.90 ꢁ
L=26.73 ꢁ
[i]
6a
30.02
15.01
10.01
7.52
6.02
5.02
4.44
4.40
3.13
2.77
30.02
15.09
10.05
7.52
6.03
5.01
4.50
4.47
3.18
2.71
33.36
16.50
11.00
8.28
6.63
5.53
4.51
4.49
3.19
2.82
35.96
17.98
VS
W
M
M
M
W
W
W
W
W
VS
W
W
M
30.06
15.03
10.02
7.52
6.01
5.01
4.44
4.39
3.14
2.81
30.11
15.05
10.04
7.53
6.02
5.02
4.50
4.45
3.18
2.85
33.13
16.56
11.04
8.28
6.63
5.52
4.51
4.47
3.19
2.85
35.92
17.96
T1: T=758C
V
A
M =625 ꢁ3
M =41.6 ꢁ2
a=6.28 ꢁ
L=21.67 ꢁ
[i]
–
T2: T=1308C
10d
30.64
15.40
10.25
7.67
6.14
5.16
VS
M
M
M
M
W
W
30.71
15.36
10.24
7.68
6.14
5.16
T: T=1308C
V
A
M =649 ꢁ3
V
M =866 ꢁ3
M =43.2 ꢁ2
A
M =56.4 ꢁ2
a=6.37 ꢁ
L=21.67 ꢁ
a=7.29 ꢁ
L=26.73 ꢁ
M
W
W
W
W
W
VS
M
W
M
M
W
M
M
W
W (br)
VS
M
5.12
5.12
[k]
–
4.87
4.60
4.27
3.94
3.63
W
W
W
W
4.89
4.60
4.28
3.94
3.65
8a
T2: T=1308C
V
A
M =700 ꢁ3
M =42.2 ꢁ2
W (br)
[i]
a=6.38 ꢁ
L=24.23 ꢁ
–
10e
34.12
VS
34.12
SmA: T=1258C
V
M =708 ꢁ3
A
M =41.6 ꢁ2
L=26.73 ꢁ
10a
T: T=1508C
V
M =760 ꢁ3
[h]
–
A
M =42.4 ꢁ2
9.00
7.18
5.98
5.13
4.48
4.51
M
M
M
W
W
M
8.98
7.18
5.99
5.13
4.49
4.51
a=6.37 ꢁ
L=26.73 ꢁ
664
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
Table 2. (Continued)
FULL PAPER
Compd
dmeas [ꢁ][a]
I[b]
hkl[c]
dcalcd [ꢁ][a]
Parameters of
smectic
Compd
dmeas [ꢁ][a]
I[b]
hkl[c]
dcalcd [ꢁ][a]
Parameters of
smectic
phase[d–g]
phase[d–g]
3.20
2.84
33.36
16.77
W
W (br)
VS
200
210
001
002
3.19
2.85
33.45
16.73
SmA: T=1958C[j]
W
V
M =784 ꢁ3
A
M =46.8 ꢁ2
L=26.73 ꢁ
[a] dmeas and dcalcd are the measured and calculated diffraction spacings, respectively. [b] I is the intensity of the reflections: VS, very strong; M, medium;
W, weak; br, broad reflection. [c] hkl are the Miller indices of the reflections. [d] T is the temperature at which the X-ray diffractogram was recorded.
[e] VM is the molecular volume, and AM the molecular area. [f] a is the lattice parameter of (the square basis of) the tetragonal unit cell in the T phase
(a=b, c=d). For compound 4a, a could not be determined due to the absence of signals in the wide-angle region. [g] L is the calculated length of the
relevant pyrrolidinium cation in its most extended conformation (estimated with Chem3D; the structure of the pyrrolidinium cation was energy-mini-
mised by an MM2 calculation within Chem3D). [h] Not detected because of overlap with the diffraction signal produced by the covering foil used in the
experimental set-up (this signal occurs at 2q=5.7–7.68). [i] Not detected. [j] During the temperature scan of the X-ray diffraction measurement, the SmA
phase was observed at lower temperatures than indicated by DSC (Table 1). This might point to slight decomposition of the sample at high temperatures.
[k] This reflection overlaps with the (006) reflection.
A
absence of more small-angle reflections for the SmA phases
confirms that the positions of the strongly diffracting bro-
mide anions in the SmA layers are not as well-defined as in
the crystal smectic T layers. In the wide-angle region, a dif-
fuse signal, centred at about 4.7 ꢁ, was observed, corre-
sponding to the lateral short-range order of the disordered
(molten) aliphatic chains. In addition, several relatively
sharp peaks were seen in the wide-angle region for the T
and E phases. For the E phases, additional peaks were ob-
served in the medium-angle region (see Supporting Informa-
tion). The nature of the T phase of 4a could not be unequiv-
ocally confirmed by X-ray diffraction due to the absence of
signals in the wide-angle region. Nevertheless, we believe
that this assignment is justified because 4a shows identical
defect textures and similar enthalpy values compared to 6a.
In addition, Skoulios et al. also observed less and less in-
tense wide-angle reflections for the shortest chain com-
pound in comparison to the other ammonium bromides.[4]
The T phases shown by the pyrrolidinium bromide com-
pounds were identified by the presence of two rather sharp
peaks (located at about 4.5 and 3.2 ꢁ) and one low-intensity
peak (located around 2.8 ꢁ) in the wide-angle region
(Figure 8). These peaks were indexed as the (110), (200) and
(210) reflections of a square lattice [reciprocal spacings of
the reflections in the characteristic ratio 1:21/2:(51/2/21/2)].[2]
The low-intensity (210) reflection broadens with increasing
chain length. For compounds 4a, 6a and 8a, the diffracto-
grams additionally show the (111) reflection of a tetragonal
lattice. This indicates that the smectic layers are slightly
three-dimensionally correlated. Due to the shorter chains,
the ionic sublayers are separated by a smaller distance (see
below).[5,6]
fates.[6] For the highly ordered smectic (and crystal smectic)
phases formed by non-ionic calamitic mesogens, only hexag-
onal (hexatic phases) and rectangular two-dimensional net-
works have been observed so far, in which the molecules
can be tilted or non-tilted within the smectic layers.[54,55]
Considering the type and structure of the cation, T phases
have only been observed for salts with cations carrying one
or two non-delocalised charges, similar to the pyrrolidinium
compounds described in this paper (on the contrary, unfunc-
tionalised 1-methyl-3-alkylimidazolium and N-alkylpyridini-
um bromides invariably form disordered SmA phases). For
these compounds, localised coulombic interactions result in
an ordered packing of the ions in the ionic sublayers; no
other relevant interactions must be taken into account (the
cations do not carry any functional groups). Ordering of the
ions in a square two-dimensional network with cations and
anions in the corners and centres, respectively, allows strict
alternation of positive and negative charges (in a manner
similar to that found in the faces of the face-centred cubic
crystals of alkali metal halides). In this way, positive and
negative charges compensate each other most efficiently.
The following section describes the X-ray diffraction data
of our pyrrolidinium compounds. The crystalline nature of
the solid phases at room temperature was confirmed by the
presence of several sharp Bragg reflections at both small
and wide angles. In addition, the equidistant small-angle re-
flections indicated that the molecules are arranged in a la-
mellar fashion.
The diffractograms of the smectic mesophases show sever-
al sharp and equidistant reflections at small angles in the
case of the highly ordered smectic phases (crystal smectic T
and crystal smectic E), or only two sharp and equidistant re-
flections at small angles in the case of the SmA phases
(except for 10e, for which only one sharp reflection was ob-
served). These reflections are related to the consecutive
smectic layers of a specific layer thickness d (Table 2). The
In contrast to what is commonly observed for SmA
phases, the layer thickness d of the T phase is independent
of temperature (Figure 9). However, d increases linearly
with the number of carbon atoms n in the alkyl chain
[Eq. (1), Figure 10].
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
665

T. Cardinaels et al.
Figure 8. X-ray diffractogram of 12a in the T phase at 1508C, represented
as the logarithm of the scattering intensity (in counts) versus the scatter-
ing angle 2q (the broad diffraction signal centred at 2qꢁ6.58 is due to
the covering foil used in the experimental set-up).
Figure 10. Evolution of the layer thickness d of the T phase at 1308C as a
function of the number of carbon atoms n in the alkyl chain for the
series 3a–12a: d=9.9
A
G
hand, Skoulios et al. explain in their paper why they prefer
to consider the T phase, from a morphological point of view,
as a truly smectic liquid-crystalline phase (with abbreviation
SmT).[4] They state that the three-dimensional correlations
in the T phase arise from long-range coulombic interactions
of the ionic charges, which are periodically distributed
inside the ionic sublayers according to a tetragonal lattice.
However, as the coulombic interactions are stronger inside
the polar layers than between them (especially when longer
alkyl chains separate the ionic layers, which explains the fact
that no (111) reflection was observed for 10a and 12a in
their T phases), the smectic layers can slide over each other
rather easily. We indeed observed that the T phases, al-
though appreciably viscous, are soft and flowing. We pro-
pose to consider the T phase as a crystal smectic phase, and
denote it by a capital letter T without the prefix “Sm”. This
type of abbreviation is also used for the other known crystal
smectic (or anisotropic plastic crystal) phases, namely, the B,
J, G, E, K and H phases.[54,55]
Figure 9. Evolution of the layer thickness d of the T phase of 8a as a
function of temperature.
Interestingly, the unit-cell parameter a of the higher-tem-
perature T2 phase of compound 6a differs only slightly from
that of the lower-temperature T1 phase (unfortunately, a
could not be calculated for the T1 phase of compound 4a).
This indicates that there are only subtle structural differen-
ces between the two T phases. This is supported by the very
small enthalpy change that accompanies the T1–T2 phase
transition of 4a and 6a (Table 1). Whereas the lamellar
period of compound 4a decreases on going from T1 to T2, it
remains constant for 6a. It is not justified to make any state-
ments about the thickness of the ionic sublayers of the T1
phases, as only two lamellar periods are available to plot as
a function of the chain length n.
ð1Þ
d=ꢁ ¼ 9:9ðꢂ0:1Þ þ 1:448ðꢂ0:008Þn
As a consequence, the thickness of the ionic sublayers d0
can be estimated to be approximately 9.9 ꢁ. The unit cell
parameter a of the square lattice (for a two-dimensional
square lattice, 1/d2hk =(h2 +k2)/a2) only slightly increases with
temperature (e.g., for compound 8a: a=6.31 ꢁ at 908C, a=
6.45 ꢁ at 1708C). However, at a given temperature it is
nearly constant for all chain lengths, and is on average equal
to 6.37 ꢁ at 1308C. As noted by Skoulios et al., this lattice
can only describe the periodic arrangement of the cationic
head groups and the bromide anions in three-dimensional
space, since the alkyl chains are disordered.[2,4] From a
formal point of view, the T phase is a genuine crystal phase
exhibiting three-dimensional positional order. On the other
To provide a more precise description of the molecular ar-
rangement, a quantitative discussion based on the measured
periodicities and calculated (partial) molecular volumes is
666
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
presented. As the layer thickness d=V/a2 =(ZV0 +ZVCH n)/
are not juxtaposed in single
layers, but superposed on
top of one another to form
double layers. A single-layer
molecular arrangement, as
represented schematically in
Figure 12, is not possible,
because of the dimensions
of the unit cell of the tetrag-
onal lattice: one side of the
square basis of the unit cell
measures only 6.37 ꢁ (as de-
duced above), while the
ionic radius of one bromide
anion is 1.95 ꢁ and the
2
a2 (where V is the volume of the tetragonal unit cell, Z the
number of molecules per unit cell, V0 the volume of one
ionic head group and VCH the volume of one methylene
2
unit; at 1308C VCH ꢁ29.2 ꢁ3;[56] the volume contribution of
2
the terminal methyl group is somewhat larger than that of a
methylene group, but this is not taken into account), Z can
be deduced from the slope of the linear d versus n plot
(Figure 10). This calculation leads to a value of Zꢁ2.0.
Therefore, each tetragonal unit cell contains two molecules,
and consequently the ionic sublayer consists of two cationic
head groups and two bromide anions per unit cell. The mo-
lecular volume VM can be estimated by using the relation
VM =(M/0.6022)f, where M is the molecular mass (gmolꢀ1)
and f is a temperature-correcting factor (f=0.9813+7.474ꢂ
10ꢀ4T with T in 8C).[26] For 8a, this gives a molecular volume
of VM ꢁ699 ꢁ3 at 1308C. This allows us to calculate the mo-
lecular area AM as 2VM/dꢁ42.2 ꢁ. This value closely corre-
sponds to the area of the square basis of the tetragonal lat-
tice (a2 =40.6 ꢁ2). Consequently, the unit cell contains
indeed two molecules, in accordance with Zꢁ2.0.
Figure 12. Top view of a theo-
retically possible single-layer
molecular arrangement for the
T phase, which satisfies the re-
quirements of charge alterna-
tion and the appearance of
two molecules per unit cell.
However, this arrangement
should be discarded for the T
phases exhibited by com-
width of
ring is approximately 4.3 ꢁ
(as estimated with
Chem3D).
a pyrrolidinium
pounds 3a–12a, because the
dimensions of the unit cell are
too small.
2) The thickness of an ionic sublayer d0, estimated to be
9.9 ꢁ, is large enough to accommodate two pyrrolidini-
um cations on top of one another, with the positively
charged nitrogen atoms located in the upper and lower
plane of the ionic sublayer, respectively. In this way, the
alkyl chains extend above and below the ionic sublayer
to create the aliphatic sublayers.
Based on the parameters calculated above, a structural
model is proposed for the molecular arrangement within the
T phases of 3a–12a (Figure 11). For this model, the follow-
ing issues were taken into account:
1) The coulombic interactions demand strict alternation of
positive and negative charges, and this can only be ach-
ieved by placing the cations and anions in the corners
and centres of the square basis of the tetragonal lattice.
Since there are two molecules (two cations and two
anions) per unit cell, the pyrrolidinium and bromide ions
3) The surface area of the square basis (a2 =40.6 ꢁ2) indeed
corresponds to the area of two long alkyl chains and thus
to two molecules per unit cell. (Although they do not
contain an elongated rigid core, the molecules are con-
sidered to be in an “upright” position. In fact, tilted fluid
phases have never been found in amphiphilic liquid crys-
tals without rigid cores).[57]
4) The alkyl chains are in their extended conformation and
are interdigitated. For all compounds, the layer thickness
d is indeed equal to about 1.37 times the calculated
length L of a cation in its most extended conformation
(Table 2). The ratio d/L slightly decreases on going from
n=12 (1.43) to n=20 (1.33), that is, interdigitation of the
longer chains is greater.
The molecular packing in the model was also tested with
regard to the space-filling requirements by elementary mo-
lecular modelling with Chem3D using the cell parameters
obtained from X-ray diffraction (Figure 13). The molecular
packing found is plausible, as no molecules or ions are sus-
pended in empty space and none overlap.
The proposed model thus strongly resembles the structure
of the T phases shown by the N,N-bis(2-hydroxyethyl)-N-
methyl-N-alkylammonium bromides (without the additional
hydrogen bonding of the hydroxyl groups within the ionic
sublayers).[4] The calculated density 1 within the ionic sub-
layers of our pyrrolidinium compounds is 1.37 gcmꢀ3, which
is consistent with the value found by Skoulios et al.
(1.40 gcmꢀ3) (1/molꢁꢀ3 =NAꢀ1V0ꢀ1 =NAꢀ1Z/a2d0, where NA is
the Avogadro constant; 1/gcmꢀ3 =166.06ꢂ1024 1/molꢁꢀ3,
where 166.06 gmolꢀ1 is the molecular weight of one ionic
Figure 11. Structural model for the T phases exhibited by 3a–12a. Left:
unit cell of the tetragonal lattice (the carbon and hydrogen atoms of the
pyrrolidinium ring and the methyl group on the nitrogen atom of the pyr-
rolidinium ring are omitted for clarity). Right: two ionic sublayers sepa-
rated by the aliphatic continuum; d is the layer thickness of the T layers
(the length of the alkyl chains is underestimated for clarity).
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
667

T. Cardinaels et al.
Figure 13. Molecular packing in the T phases exhibited by compounds
3a–12a, as obtained by elementary molecular modelling (Chem3D soft-
ware). Left: side view of a unit cell of the tetragonal lattice. Right: top
view of a unit cell of the tetragonal lattice. For clarity, the long alkyl
chains on the cations are replaced by single methyl groups. Before
modelACHTUNGTRENNUNGling, the structure of a pyrrolidinium cation was energy-minimised
by an MM2 calculation within Chem3D. The pyrrolidinium ring is non-
planar, with the nitrogen atom sitting above the plane of the four ring
carbon atoms.[29]
Figure 14. Evolution of the layer thickness d of the SmA phase of 10e as
a function of temperature.
head group).[4] On the contrary, the proposed structure dif-
fers from those of all other reported T phases, which contain
only one molecule per tetragonal unit cell, although the
latter phases were all formed by compounds with two long
alkyl chains per cation.[2,5,6] Considering the dimensions of
the tetragonal unit cell of our pyrrolidinium compounds, the
pyrrolidinium cations are theoretically free to rotate around
the axis normal to the smectic layer planes. However, the
space-filling model suggests cooperative rotation, whereby
rotating cations must share space (as in SmB and crystal B
phases).[54]
In the case of 10a, 11a and 12a (n=18, 19 and 20, respec-
tively), a SmA phase exists above the T phase at higher
temperatures, corresponding to the loss of ordering within
the smectic layer planes. This phase, however, only exists
over a narrow temperature range. The T–SmA phase transi-
tion is accompanied by a decrease in the smectic layer thick-
ness and by an increase of the molecular area (Table 2). The
transition can be explained as follows. When the tempera-
ture is increased, the volume of the long alkyl chains in-
creases. To counterbalance this increase in volume and mo-
lecular area of the alkyl chains, the molecular area of the
ionic moieties increases as well (this is necessary for a stable
smectic phase). At the T–SmA transition, however, the alkyl
chains (and ionic head groups) occupy such a large molecu-
lar cross section that the high degree of ordering within the
ionic sublayers of the T phase cannot be sustained anymore
(as is confirmed by an increase in the thickness of the ionic
sublayers, d0).[4] Nevertheless, for the long-chain compounds
10a, 11a and 12a, the van der Waals interactions between
the aliphatic alkyl chains are strong enough to retain the mi-
crophase separation between ionic sublayers and aliphatic
continuum, which leads to the formation of a disordered
SmA phase. For the shorter alkyl chains (n<18), this is not
the case (the T phase transforms directly into the isotropic
liquid, not via a SmA phase), nor is it the case for 10c and
10d (n=18). It may, nevertheless, be expected that for
The T phases of 10c and 10d have a smaller layer thick-
ness d in comparison to the T phase of bromide analogue
10a (Table 2). Furthermore, the molecular area AM and lat-
tice parameter a are larger (at 1308C, AMꢀ=41.6 ꢁ2 for 10a).
ꢀ
This is due to the larger size of the BF4 and PF6 anions,
respectively. The molten alkyl chains are more folded to
compensate for the larger molecular area of the ionic head
groups in the smectic layers. This explains the smaller d
values. In general, however, we propose a similar model for
ꢀ
these T phases to that depicted in Figure 11. The BF4 and
ꢀ
PF6 anions can indeed be considered to be negatively
charged spheres that resemble a single bromide anion, but
with a larger radius. The diffractograms of 10c and 10d ad-
ditionally show the (111), (112), (113), (114) and (115) re-
flections of a square lattice (Table 2), that is, the smectic
layers are correlated (see above). The (210) reflection, how-
ever, could not be detected. For 10c and 10d, the layer
thickness d of the T phase increases slightly with increasing
temperature (10c: d=32.28 ꢁ at 1008C, d=32.99 ꢁ at
1608C; 10d: d=30.33 ꢁ at 1008C, d=30.95 ꢁ at 1408C).
ꢀ
longer alkyl chains (n>18), the corresponding BF4 and
ꢀ
PF6 compounds will show an additional SmA phase, as the
van der Waals interactions between the aliphatic alkyl
chains will then be sufficiently strong to preserve a micro-
phase separation between ionic sublayers and aliphatic con-
tinuum.
ꢀ
ꢀ
In contrast to the spherical BF4 and PF6 anions, the
linear thiocyanate anion seems to be incompatible with te-
A
Indexing the diffractograms of uranyl complexes 6g–12g
was not straightforward (Figure 15). However, for all these
salts an orthorhombic lattice could be obtained, with unit
cell parameter c equal to the layer thickness d of the meso-
pected, the layer thickness d in the SmA phase of 10e de-
creases with increasing temperature (Figure 14). The alkyl
chains are interdigitated (L<d<2L).
668
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
Thus, while the cationic head groups and anions of com-
pounds 3a–12a are arranged in a square two-dimensional
lattice in the T phase, the ionic fragments of uranyl com-
plexes 6g–12g are ordered in a rectangular two-dimensional
lattice. The alkyl chains are interdigitated (L<d<2L). The
layer thickness d of the E phase increases slightly with in-
creasing temperature (Figure 16). However, it is much more
dependent on the number of carbon atoms in the alkyl
chain n; the layer thickness increases linearly with n
(Figure 17).
Figure 15. X-ray diffractogram of 6g in the E phase at 958C, represented
as the logarithm of the scattering intensity (in counts) versus the scatter-
ing angle 2q (the broad diffraction signal centred at 2qꢁ6.58 is due to
the covering foil used in the experimental set-up).
phase (diffraction data and unit-cell parameters are present-
ed in the Supporting Information). Unfortunately, no rea-
sonable unit-cell parameters a and b could be found for
compound 12g. The orthorhombic symmetry is consistent
only with an orthogonal crystal smectic E phase, which be-
longs to the class of anisotropic plastic crystal phases.[54,55]
This phase has already been observed for N-alkylpyridinium
salts of [PdCl4]2ꢀ with hexadecyl and octadecyl chains.[48] In-
terestingly, the corresponding [PdBr4]2ꢀ salts only showed a
disordered SmA phase. This was attributed to a lower
degree of hydrogen bonding and to size effects. Although no
hydrogen bonding can be expected for our [UO2Br4]2ꢀ com-
plexes, these compounds form a highly ordered E phase.
Weak interactions of the protons of the pyrrolidinium cat-
ions with the [UO2Br4]2ꢀ dianions, as observed in the crystal
structure of compound 6g (Figure 1), may be of importance
for the formation of the E phase.
Figure 16. Evolution of the layer thickness d of the E phase of 10g as a
function of temperature.
Heating the E phase destroys the long-range (intralayer
and interlayer) positional order and leads directly to forma-
tion of the isotropic liquid in the case of the short-chain
compounds (6g–10g). For the long-chain analogues (11g
and 12g), the transition to the isotropic liquid occurs gradu-
ally via a SmA phase, by analogy with the T–SmA–I phase
sequence observed for the bromide compounds (see above).
The diffractogram of the SmA phase of uranyl complexes
11g and 12g shows two sharp, equidistant peaks in the
small-angle region (corresponding to the lamellar structure),
and a diffuse signal in the wide-angle region, centred at
about 4.85 ꢁ (corresponding to the lateral short-range order
of the molten aliphatic chains). In addition, a diffuse signal
centred at about 8.1 ꢁ could be observed (see Supporting
Information). This broad reflection is associated with the
short-range positional correlation of the [UO2Br4]2ꢀ dianions
within the smectic layers.
Figure 17. Evolution of the layer thickness d of the E phase at 1158C as a
function of the number of carbon atoms n in the alkyl chain for the
series 6g–12g: d=13.5
N
U
Luminescence properties: EuropiumACTHNUTRGNEUNG(III) complex 10 f ex-
hibits intense red photoluminescence (Figure 18). The lumi-
5
nescence spectrum consists of transitions from the D0 excit-
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
669

T. Cardinaels et al.
Figure 19. Photoluminescence spectra of uranyl complex 6g at room tem-
perature (excitation wavelength: 435 nm). Black line: 6g dissolved in ace-
tonitrile. Grey line: 6g dissolved in [C4mpyrr]ACTHNUTRGNENG[U NTf2].
Figure 18. Photoluminescence spectrum of europiumACTHNUGRTENUNG(III) complex 10 f in
the solid state at room temperature (excitation wavelength: 393 nm).
7
ed state to the different J levels of the F ground term (J=
length and anion type have great influence on the mesomor-
phic behaviour. The most interesting feature of (some of)
these pyrrolidinium salts is the formation of T phases on
heating. The absence of charge delocalisation in the cation
seems to be of crucial importance for packing of the ions
into a tetragonal lattice. Moreover, large, asymmetric coun-
terions do not seem to be compatible with the highly or-
dered ionic sublayers of the T phase. To induce disordered
phases above the T phase, sufficiently long alkyl chains and
small anions are required for the pyrrolidinium compounds.
The pyrrolidinium salts containing the [UO2Br4]2ꢀ fragment
are not photoluminescent in the pure, undiluted form, while
photoluminescence can be induced by dissolving them in an
ionic liquid. In this way, autoquenching is avoided. The
5
7
0, 1, 2, 3, 4). The spectrum is dominated by the D0! F2
line, with emission maximum at 614.3 nm. The intensity
7
7
ratio I
N
ACHTUNGTRENNUNG
emission peaks, is 19.9. This value is remarkably high, but is
not unusual for europiumACTHNUTRGENU(GN III) tetrakis b-diketonate com-
5
7
plexes.[58] Thanks to the intense D0! F2 line, the emitted
5
7
light has high colour purity. The very weak D0! F0 line at
580.2 nm (only 3% of the intensity of the magnetic dipole
5
7
transition D0! F1) may indicate that the symmetry of the
eight-coordinate europium(III) complex is close to an ideal
AHCTUNGTRENNUNG
5
7
dodecahedron with D2d symmetry (the D0! F0 line is for-
bidden in D2d symmetry).
Tetrabromouranyl salts 1g–12g are not luminescent in
pure form, probably due to autoquenching.[59,60] However,
when these salts were dissolved in acetonitrile or in the
ionic liquid N-butyl-N-methylpyrrolidinium bis(trifluoro-
europ
ACHUTGTNRNEUGiN umHCATUNGTRNE(NUGN III) complex 10 f shows intense red photolumines-
cence with high colour purity.
ACHTUNGTRENNUNGmethylsulfonyl)imide ([C4mpyrr]ACHUTGNTREN[NUNG NTf2]), photoluminescence
in the visible region of the electromagnetic spectrum could
be observed. The luminescence spectra of uranyl complex
6g, recorded at room temperature, are shown in Figure 19.
For both solutions, a broad and intense emission peak was
found between 16000 and 21000 cmꢀ1. In contrast to the
[UO2Cl4]2ꢀ species, which shows a distinct vibrational fine
structure when dissolved in ionic liquids, no fine structure
could be observed for the [UO2Br4]2ꢀ species, neither in
acetonitrile, nor in the ionic liquid.[61–63]
Experimental Section
General: Nuclear magnetic resonance (NMR) spectra were recorded on
a Bruker Avance 300 spectrometer (operating at 300 MHz for 1H). Ele-
mental analyses were obtained on a CE Instruments EA-1110 elemental
analyser. Electrospray ionisation (ESI) mass spectra were recorded on a
Thermo Finnigan LCQ Advantage mass spectrometer. Defect textures of
the mesophases were observed with an Olympus BX60 polarising micro-
scope equipped with a LINKAM THMS600 hot stage and a LINKAM
TMS93 programmable temperature controller. Microscope glass slides
were soaked in concentrated HNO3 for an hour, washed with distilled
water, and finally rinsed with acetone, which was then allowed to evapo-
rate. DSC traces were recorded with a Mettler-Toledo DSC822e module
(heating/cooling rate of 108Cminꢀ1; He atmosphere). Thermal stability
was assessed by thermogravimetry in the range from 25 to 4008C at a
heating rate of 108C minꢀ1 under nitrogen with a TA Instruments SDT
Q600 thermal analyser. Powder X-ray diffractograms were recorded with
a Bruker AXS D8 Discover diffractometer mounted with a copper X-ray
ceramic tube, working at 1.6 kW. The emitted CuKa radiation (l=
1.5418 ꢁ) was focussed on the sample by a Gçbel mirror. All samples
(without a thermal history) were prepared by spreading the powders on a
thin cleaned Si wafer. Diffractograms were collected in the Bragg–Bren-
Conclusion
We have synthesised the first series of thermotropic ionic
liquid crystals based on the pyrrolidinium cation. The pyrro-
lidinium compounds show a rich mesomorphism, including
highly ordered smectic phases such as the crystal smectic T
phase and the crystal smectic E phase, disordered SmA
phases and hexagonal columnar phases. Both the alkyl chain
670
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
tano reflection geometry (q/2q set-up) at an angular resolution (in 2q) of
0.038 per step. The deviation between the temperature on the surface of
the sample holder and the set temperature was about 3%. The scattering
signal was recorded with a one-dimensional detector (LynxEye detector).
Indexation of the powder X-ray diffractograms was performed with the
WinXPOW program package with the Index & Refine program by using
the Werner TREOR algorithm program (allowed error in matching the
experimentally observed peaks of 2q=0.058).[64,65] Molecular models
were obtained with the Chem3D software package from CambridgeSoft.
Photoluminescence spectra were recorded on an Edinburgh Instruments
FS900 spectrofluorimeter. This instrument is equipped with a xenon arc
lamp, a microsecond flashlamp (pulse length: 2 ms) and a red-sensitive
photomultiplier (300–850 nm). The excitation wavelength was 393 nm for
10 f and 435 nm for 6g. The ionic liquid N-butyl-N-methylpyrrolidinium
bis(trifluoromethylsulfonyl)imide was obtained from IoLiTec.
yellow precipitate was filtered off and washed carefully with toluene, n-
hexane and ethyl acetate. The residue was vigorously stirred in n-hexane
for 15 min, filtered off and washed with n-hexane and diethyl ether. The
1
pure, white compound was dried in vacuo at 508C. Yield: 72%. H NMR
(300 MHz, CDCl3, 258C, TMS): d=0.90 (t, 3J
ACTHNUTRGNE(UNG H,H)=6.5 Hz, 3H; CH3),
1.19–1.46 (m, 16H; CH2), 1.71–1.85 (m, 2H; NCH2CH2), 2.25–2.38 (m,
4H; pyrr. H-3 and H-4), 3.30 (s, 3H; NCH3), 3.60–3.69 (m, 2H; NCH2),
3.77–3.91 ppm (m, 4H; pyrr. H-2 and H-5); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=14.02, 21.59, 22.57, 24.03, 26.33, 29.16, 29.28, 29.37,
29.43, 31.76, 48.61, 64.15, 64.37 ppm; ESIMS (acetonitrile/water (80/20)+
0.1% formic acid): m/z (%): 240.5 (100) [MꢀBrꢀ]+, 559.5 (38)
[M+MꢀBrꢀ]+; elemental analysis calcd (%) for C16H34BrN: C 59.99, H
10.70, N 4.37; found: C 59.68, H 11.07, N 4.34.
4a–12a: The appropriate 1-bromoalkane (1 mol), or for n>16 a solution
of the 1-bromoalkane in dry toluene, was added dropwise to a stirred so-
lution of 1-methylpyrrolidine (1 mol) in dry toluene under an argon at-
mosphere. The mixture was stirred for 48 h at 808C under an argon at-
mosphere. A pale yellow precipitate formed and was filtered off and
washed carefully with toluene and ethyl acetate. The precipitate was re-
crystallised from methanol/diethyl ether to obtain a white solid. The pure
compounds were dried in vacuo at 508C.
Yellow single crystals of compound 6g were obtained after three days by
slowly evaporating a solution of the compound in ethanol. X-ray intensity
data were collected on a SMART 6000 diffractometer equipped with a
CCD detector using CuKa radiation (l=1.5418 ꢁ). The images were in-
terpreted and integrated with the program SAINT from Bruker.[66] 6g:
C38H80Br4N2O2U, M=1154.70 gmolꢀ1
8.5731(3), b=56.8084(2), c=10.5237(4) ꢁ, b=113.8850(1)8, V=
4686.4(3) ꢁ3, T=100(2) K, Z=4, 1calcd =1.637 gcmꢀ3
(CuKa)=
13.966 mmꢀ1, F
(000)=2280, crystal size 0.3ꢂ0.2ꢂ0.1 mm, 45996 inde-
pendent reflections (Rint =0.0760). Number of parameters=428. Absorp-
tion correction: multi-scan (SADABS, Bruker 2004). Tmin =0.094; Tmax
, monoclinic, P21/n (No. 14), a=
4a: Yield: 88%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.88 (t,
3J
ACHTNUGTRNEG(UN H,H)=6.4 Hz, 3H; CH3), 1.26–1.37 (m, 18H; CH2), 1.74–1.75 (m, 2H;
,
mACHTUNGTRENNUNG
ACHTUNGTRENNUNG
NCH2CH2), 2.23–2.39 (m, 4H; pyrr. H-3 and H-4), 3.33 (s, 3H; NCH3),
3.64–3.69 (m, 2H; NCH2), 3.85–3.89 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.51, 22.06, 23.06, 24.51,
26.81, 29.64, 29.70, 29.76, 29.85, 29.98, 32.29, 49.05, 64.55, 64.82 ppm;
ESIMS (acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 254.5
(100) [MꢀBrꢀ]+, 587.4 (22) [M+MꢀBrꢀ]+; elemental analysis calcd (%)
for C17H36BrN: C 61.06, H 10.85, N 4.19; found: C 60.74, H 11.25, N 4.14.
=
0.247. Final R=0.0486 for 8254 reflections with I>2s(I) and wR2 =
0.1182 for all data. The structure was solved by direct methods and re-
fined by full-matrix least-squares techniques on F2 by using the
SHELXTL program package.[67] Non-hydrogen atoms were refined aniso-
tropically and hydrogen atoms in riding mode with isotropic temperature
factors fixed at 1.2U(eq) of the parent atoms (1.5U(eq) for methyl
groups). CCDC 695524 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
5a: Yield: 80%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.89 (t,
3J
ACHTNUGTRNEG(UN H,H)=6.4 Hz, 3H; CH3), 1.19–1.46 (m, 20H; CH2), 1.71–1.84 (m, 2H;
NCH2CH2), 2.25–2.38 (m, 4H; pyrr. H-3 and H-4), 3.31 (s, 3H; NCH3),
3.61–3.69 (m, 2H; NCH2), 3.77–3.92 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.13, 21.68, 22.66, 24.13,
26.44, 29.25, 29.34, 29.40, 29.47, 29.60, 31.89, 48.69, 64.20, 64.46 ppm;
ESIMS (acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 268.5
(100) [MꢀBrꢀ]+, 615.4 (27) [M+MꢀBrꢀ]+; elemental analysis calcd (%)
for C18H38BrN: C 62.05, H 10.99, N 4.02; found: C 61.85, H 11.14, N 3.98.
1a and 2a: The appropriate 1-bromoalkane (1 mol) was added dropwise
to an ice-cooled, stirred solution of 1-methylpyrrolidine (1 mol) in dry
toluene under an argon atmosphere. The mixture was allowed to warm
to room temperature and was stirred for 48 h under an argon atmos-
phere. The solvent was removed under reduced pressure. The residue
was washed carefully with toluene, ethyl acetate and diethyl ether until a
white, hygroscopic powder was obtained. The pure compounds were
dried in vacuo at 708C.
6a: Yield: 78%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.88 (t,
3J
ACHTNUGTRNEG(UN H,H)=6.5 Hz, 3H; CH3), 1.26–1.36 (m, 22H; CH2), 1.69–1.79 (m, 2H;
NCH2CH2), 2.25–2.37 (m, 4H; pyrr. H-3 and H-4), 3.32 (s, 3H; NCH3),
3.63–3.69 (m, 2H; NCH2), 3.81–3.90 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.13, 21.69, 22.69, 24.13,
26.44, 29.27, 29.37, 29.48, 29.64, 31.92, 48.70, 64.19, 64.46 ppm; ESIMS
(acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 282.5 (100)
[MꢀBrꢀ]+, 643.5 (25) [M+MꢀBrꢀ]+; elemental analysis calcd (%) for
C19H40BrN: C 62.96, H 11.12, N 3.86; found: C 62.76, H 10.71, N 3.83.
1a: Yield: 47%. 1H NMR (300 MHz, CD3OD, 258C, TMS): d=0.91 (t,
3J
ACHTUNGTRENNUNG(H,H)=6.4 Hz, 3H; CH3), 1.28–1.50 (m, 10H; CH2), 1.75–1.90 (m, 2H;
NCH2CH2), 2.19–2.32 (m, 4H; pyrr. H-3 and H-4), 3.09 (s, 3H; NCH3),
3.35–3.45 (m, 2H; NCH2), 3.51–3.66 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CD3OD, 258C, TMS): d=14.54, 22.68, 23.75, 24.91,
27.64, 30.30, 32.97, 48.90, 65.49, 65.67 ppm; ESIMS (acetonitrile): m/z
(%): 198.4 (100) [MꢀBrꢀ]+, 477.2 (10) [M+MꢀBrꢀ]+; elemental analysis
calcd (%) for C13H28BrN·H2O: C 52.70, H 10.21, N 4.73; found: C 52.29,
H 10.33, N 4.53.
7a: Yield: 74%. 1H NMR (300 MHz, CD2Cl2, 258C): d=0.90 (t,
3J
ACHTNUGTRNEG(UN H,H)=6.3 Hz, 3H; CH3), 1.22–1.47 (m, 24H; CH2), 1.72–1.85 (m, 2H;
NCH2CH2), 2.24–2.36 (m, 4H; pyrr. H-3 and H-4), 3.26 (s, 3H; NCH3),
3.56–3.64 (m, 2H; NCH2), 3.70–3.88 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CD2Cl2, 258C): d=13.95, 21.68, 22.75, 24.03, 26.44,
29.24, 29.42, 29.55, 29.71, 31.98, 48.66, 64.30, 64.51 ppm; ESIMS (acetoni-
trile/water (80/20)+0.1% formic acid): m/z (%): 296.6 (100) [MꢀBrꢀ]+,
671.6 (46) [M+MꢀBrꢀ]+; elemental analysis calcd (%) for C20H42BrN: C
63.81, H 11.25, N 3.72; found: C 63.36, H 11.66, N 3.54.
2a: Yield: 51%. 1H NMR (300 MHz, CD3OD, 258C, TMS): d=0.90 (t,
3J
ACHTUNGTRENNUNG(H,H)=6.4 Hz, 3H; CH3), 1.26–1.44 (m, 14H; CH2), 1.74–1.89 (m, 2H;
NCH2CH2), 2.19–2.28 (m, 4H; pyrr. H-3 and H-4), 3.08 (s, 3H; NCH3),
3.34–3.43 (m, 2H; NCH2), 3.50–3.62 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CD3OD, 258C, TMS): d=14.55, 22.66, 23.80, 24.90,
27.64, 30.33, 30.48, 30.64, 33.11, 48.94, 65.49, 65.68 ppm; ESIMS (acetoni-
trile/water (80/20)+0.1% formic acid): m/z (%): 226.4 (100) [MꢀBrꢀ]+,
533.3 (10) [M+MꢀBrꢀ]+; elemental analysis calcd (%) for
C15H32BrN·H2O: C 55.55, H 10.57, N 4.32; found: C 55.61, H 10.30, N
4.18.
8a: Yield: 76%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.88 (t,
3J
ACHTNUGTRNEG(UN H,H)=6.4 Hz, 3H; CH3), 1.15–1.48 (m, 26H; CH2), 1.70–1.87 (m, 2H;
NCH2CH2), 2.22–2.40 (m, 4H; pyrr. H-3 and H-4), 3.32 (s, 3H; NCH3),
3.61–3.72 (m, 2H; NCH2), 3.76–3.96 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.19, 21.74, 22.75, 24.20,
26.50, 29.33, 29.43, 29.55, 29.67, 29.74, 31.98, 48.74, 64.23, 64.50 ppm;
ESIMS (acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 310.6
(100) [MꢀBrꢀ]+, 699.6 (30) [M+MꢀBrꢀ]+; elemental analysis calcd (%)
for C21H44BrN: C 64.59, H 11.36, N 3.59; found: C 64.33, H 11.69, N 3.52.
3a: 1-Bromoundecane (1 mol) was added dropwise to an ice-cooled
stirred solution of 1-methylpyrrolidine (1 mol) in dry toluene under an
argon atmosphere. The mixture was allowed to warm to room tempera-
ture and stirred for 12 h under an argon atmosphere. The temperature
was increased to 508C and stirring was continued for 36 h. The pale
9a: Yield: 80%. 1H NMR (300 MHz, CD2Cl2, 258C): d=0.91 (t,
3J
ACHTNUGTRNEG(UN H,H)=6.5 Hz, 3H; CH3), 1.18–1.51 (m, 28H; CH2), 1.71–1.85 (m, 2H;
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
671

T. Cardinaels et al.
NCH2CH2), 2.23–2.37 (m, 4H; pyrr. H-3 and H-4), 3.26 (s, 3H; NCH3),
3.54–3.64 (m, 2H; NCH2), 3.70–3.87 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CD2Cl2, 258C): d=14.16, 21.88, 22.96, 24.24, 26.65,
29.45, 29.63, 29.76, 29.97, 32.19, 48.89, 64.57, 64.76 ppm; ESIMS (acetoni-
trile/water (80/20)+0.1% formic acid): m/z (%): 324.6 (100) [MꢀBrꢀ]+,
727.5 (24) [M+MꢀBrꢀ]+; elemental analysis calcd (%) for C22H46BrN: C
[M+MꢀBF4ꢀ]+; elemental analysis calcd (%) for C23H48BF4N: C 64.93,
H 11.37, N 3.29; found: C 64.71, H 11.03, N 3.23.
10d: A solution of 10a (1.19 mmol) in methanol/water (60/40, 10 mL)
was heated to 608C. An aqueous solution of KPF6 (2.38 mmol) was
added dropwise and the reaction mixture was stirred for 3 h at 608C. A
white precipitate formed, and was filtered off and washed with water. A
test with AgNO3 confirmed that no bromide starting compound was pres-
ent anymore. The pure product (which completely dissolved in dry
chloroform, that is, there was no residual KPF6) was dried in vacuo at
508C. Yield: 98%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.88 (t,
65.32,
10a: Yield: 81%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.88 (t,
3J
(H,H)=6.4 Hz, 3H; CH3), 1.26–1.36 (m, 30H; CH2), 1.72–1.75 (m, 2H;
H
11.46,
N
3.46; found:
C
64.91,
H
11.73,
N
3.33.
ACHTUNGTRENNUNG
NCH2CH2), 2.25–2.37 (m, 4H; pyrr. H-3 and H-4), 3.32 (s, 3H; NCH3),
3.63–3.69 (m, 2H; NCH2), 3.85–3.88 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.51, 22.06, 23.07, 24.49,
26.81, 29.65, 29.77, 29.89, 30.05, 30.10, 46.05, 64.55, 64.82 ppm; ESIMS
(acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 338.6 (100)
[MꢀBrꢀ]+, 755.6 (28) [M+MꢀBrꢀ]+; elemental analysis calcd (%) for
C23H48BrN: C 66.00, H 11.56, N 3.35; found: C 65.54, H 11.25, N 3.45.
3J
ACHTNUGTRNEG(UN H,H)=6.5 Hz, 3H; CH3), 1.25–1.34 (m, 30H; CH2), 1.73–1.77 (m, 2H;
NCH2CH2), 2.22–2.32 (m, 4H; pyrr. H-3 and H-4), 3.05 (s, 3H; NCH3),
3.26–3.32 (m, 2H; NCH2), 3.51–3.53 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.53, 21.67, 23.09, 24.23,
26.68, 29.51, 29.77, 29.91, 29.92, 30.13, 32.32, 48.77, 64.82, 64.99 ppm;
ESIMS (acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 338.6
ꢀ
ꢀ
(100) [MꢀPF6ꢀ]+, 821.4 (35) [M+MꢀPF6ꢀ]+, 628.1 (12) [M+PF6
] (ob-
11a: Yield: 75%. 1H NMR (300 MHz, CD2Cl2, 258C): d=0.91 (t,
3J
ACHTUNGTRENNUNG(H,H)=6.4 Hz, 3H; CH3), 1.20–1.49 (m, 32H; CH2), 1.72–1.85 (m, 2H;
served in negative mode); elemental analysis calcd (%) for C23H48F6NP:
C 57.12, H 10.00, N 2.90; found: C 56.90, H 9.98, N 2.99.
NCH2CH2), 2.24–2.36 (m, 4H; pyrr. H-3 and H-4), 3.26 (s, 3H; NCH3),
3.55–3.64 (m, 2H; NCH2), 3.69–3.87 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CD2Cl2, 258C): d=14.15, 21.88, 22.96, 24.24, 26.65,
29.45, 29.63, 29.76, 29.97, 32.19, 48.89, 64.58, 64.76 ppm; ESIMS (acetoni-
trile/water (80/20)+0.1% formic acid): m/z (%): 352.6 (100) [MꢀBrꢀ]+,
783.5 (12) [M+MꢀBrꢀ]+; elemental analysis calcd (%) for C24H50BrN: C
66.64, H 11.65, N 3.24; found: C 66.22, H 11.97, N 3.12.
10e: KSCN (36.00 mmol) was added to a solution of 10a (18.00 mmol) in
acetone (150 mL) at 558C. The mixture was stirred for 48 h at 558C
under an argon atmosphere with exclusion of light. The warm reaction
mixture was filtered to remove the precipitated KBr. The precipitate was
washed with warm acetone and warm chloroform. The solvent of the fil-
trate was removed under reduced pressure, and the residue was dissolved
in dry chloroform. The precipitated residual KBr and the excess of
KSCN were filtered off. The solvent of the filtrate was removed under re-
duced pressure, and the residue was dissolved in dry chloroform again.
This solution was left to stand in a freezer overnight. The formed precipi-
tate was filtered off. The solvent of the filtrate was removed under re-
duced pressure. This procedure was repeated once. The pure, white to
pinkish product was dried in vacuo at 508C. Yield: 84%. 1H NMR
12a: Yield: 76%. 1H NMR (300 MHz, CDCl3, 258C, TMS): d=0.88 (t,
3J
ACHTUNGTRENNUNG(H,H)=6.5 Hz, 3H; CH3), 1.25–1.40 (m, 34H; CH2), 1.69–1.77 (m, 2H;
NCH2CH2), 2.25–2.37 (m, 4H; pyrr. H-3 and H-4), 3.32 (s, 3H; NCH3),
3.63–3.68 (m, 2H; NCH2), 3.82–3.88 ppm (m, 4H; pyrr. H-2 and H-5);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=14.17, 21.72, 22.73, 24.17,
26.47, 29.31, 29.40, 29.45, 29.55, 29.75, 31.96, 48.70, 64.19, 64.46 ppm;
ESIMS (acetonitrile/water (80/20)+0.1% formic acid): m/z (%): 366.7
(100) [MꢀBrꢀ]+, 813.5 (20) [M+MꢀBrꢀ]+; elemental analysis calcd (%)
for C25H52BrN: C 67.24, H 11.74, N 3.14; found: C 67.20, H 11.50, N 3.10.
(300 MHz, CDCl3, 258C, TMS): d=0.88 (t, 3J
ACTHNUTRGNE(UNG H,H)=6.5 Hz, 3H; CH3),
1.18–1.48 (m, 30H; CH2), 1.75–1.88 (m, 2H; NCH2CH2), 2.31–2.42 (m,
4H; pyrr. H-3 and H-4), 3.25 (s, 3H; NCH3), 3.49–3.57 (m, 2H; NCH2),
3.67–3.82 ppm (m, 4H; pyrr. H-2 and H-5); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=14.26, 22.05, 22.82, 24.28, 26.57, 29.33, 29.50, 29.64,
29.82, 32.05, 48.74, 64.82, 131.48 ppm; ESIMS (acetonitrile/water (80/
20)+0.1% formic acid): m/z (%): 338.7 (100) [MꢀSCNꢀ]+, 734.5 (40)
[M+MꢀSCNꢀ]+; elemental analysis calcd (%) for C24H48N2S: C 72.66, H
12.20, N 7.06; found: C 72.28, H 11.82, N 7.03.
10b: A solution of 10a (1.19 mmol) in methanol (6 mL) was heated to
608C. An aqueous solution of LiNTf2 (2.38 mmol) was added dropwise
and the reaction mixture was stirred for 3 h at 608C. Methanol was then
removed under reduced pressure. The white precipitate was filtered off,
washed with water and dried in vacuo at 508C. Yield: 88%. 1H NMR
(300 MHz, CDCl3, 258C, TMS): d=0.88 (t, 3J
ACTHNUTRGNE(UNG H,H)=6.4 Hz, 3H; CH3),
1.25–1.35 (m, 30H; CH2), 1.76–1.78 (m, 2H; NCH2CH2), 2.23–2.35 (m,
4H; pyrr. H-3 and H-4), 3.08 (s, 3H; NCH3), 3.30–3.35 (m, 2H; NCH2),
3.52–3.56 ppm (m, 4H; pyrr. H-2 and H-5); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=14.25, 21.62, 22.81, 23.99, 26.29, 29.15, 29.23, 29.48,
29.57, 29.73, 29.81, 32.04, 48.54, 64.67, 64.94, 74.11, 113.56, 117.82, 122.06,
126.32 ppm; ESIMS (acetonitrile/water (80/20)+0.1% formic acid): m/z
(%): 338.6 (100) [MꢀNTf2ꢀ]+, 956.3 (33) [M+MꢀNTf2ꢀ]+, 280.2 (100)
10 f: 2-Thenoyltrifluoroacetone (1.46 mmol) was dissolved in ethanol
(4 mL), and NaOH (1.46 mmol, 1 molLꢀ1 aqueous solution) was added.
The mixture was stirred for 10 min. Then a solution of 10a (0.37 mmol)
in ethanol (5 mL) was added. The mixture was heated to 508C. A solu-
tion of EuCl3·6H2O (0.37 mmol) in water (2.5 mL) was added dropwise
and the mixture was stirred for 1.5 h at 508C with exclusion of light. A
yellow precipitate formed and was filtered off and washed with cold
water and with cold ethanol. The pure product was dried in vacuo at
508C. Yield: 56%. ESIMS (acetonitrile/water (80/20)+0.1% formic
ꢀ
ꢀ
[NTf2]ꢀ (observed in negative mode), 898.1 (68) [M+NTf2
] (observed
in negative mode); elemental analysis calcd (%) for C25H48F6N2O4S2: C
48.53, H 7.82, N 4.53; found: C 48.61, H 8.09, N 4.35.
acid): m/z (%): 338.6 (100) [Mꢀ[Eu
(tta)4]ꢀ]+, 1037.1 (100), [Eu(tta)4]ꢀ
ACHUTGTNRENNUG ACHTUNGTRENNUNG
(observed in negative mode); elemental analysis calcd (%) for
C55H64EuF12NO8S4: C 48.03, H 4.69, N 1.02; found: C 48.00, H 4.67, N
1.09.
10c: A solution of 10a (1.19 mmol) in acetone (10 mL) was heated to
558C. NaBF4 (2.38 mmol) was added and the reaction mixture was
heated to reflux for 3 h. Inorganic salts were filtered off and washed with
acetone. The filtrate was evaporated under reduced pressure to give a
white powder. A test with AgNO3 confirmed that no bromide starting
compound was present anymore. Residual NaBF4 was removed by dis-
solving the crude product in dry chloroform, leaving this solution in a re-
frigerator and filtering the cold solution. The solvent of the filtrate was
removed under reduced pressure. This procedure was repeated once. The
pure product was dried in vacuo at 508C. Yield: 70%. 1H NMR
1g–12g: A solution of UO2Br2·xH2O (1 mol) in ethanol was added drop-
wise to a stirred solution of the appropriate N-alkyl-N-methylpyrrolidini-
um bromide 1a–12a (2 mol) in ethanol. The reaction mixture was heated
to reflux for 2 h. Then the solvent was removed under reduced pressure.
The crude product was recrystallised twice from a minimal amount of 1-
butanol/ethanol (2/1; 1g–2g) or from a minimal amount of ethanol (4g–
12g) to obtain yellow crystals (crystals formed in the refrigerator for the
short-chain compounds). The pure compounds were dried in vacuo at 40
or 508C.
(300 MHz, CDCl3, 258C, TMS): d=0.88 (t, 3J
ACTHNUTRGNE(UNG H,H)=6.4 Hz, 3H; CH3),
1.25–1.34 (m, 30H; CH2), 1.72–1.77 (m, 2H; NCH2CH2), 2.23–2.31 (m,
4H; pyrr. H-3 and H-4), 3.10 (s, 3H; NCH3), 3.33–3.34 (m, 2H; NCH2),
3.52–3.61 ppm (m, 4H; pyrr. H-2 and H-5); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=14.53, 22.03, 23.09, 24.31, 26.75, 29.57, 29.77, 29.92,
30.10, 32.32, 48.73, 64.73, 64.78 ppm; ESIMS (acetonitrile/water (80/20)+
0.1% formic acid): m/z (%): 338.6 (100) [MꢀBF4ꢀ]+, 763.5 (25)
1g: Yield: 17%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 198.3 (100) [MꢀC13H28N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C26H56Br4N2O2U: C
31.66, H 5.72, N 2.84; found: C 31.78, H 5.89, N 2.72.
672
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674

Pyrrolidinium Ionic Liquid Crystals
FULL PAPER
2g: Yield: 18%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
[7] F. Artzner, M. Veber, M. Clerc, A. M. Levelut, Liq. Cryst. 1997, 23,
27–33.
[8] T. Kato, N. Mizoshita, K. Kishimoto, Angew. Chem. 2006, 118, 44–
74; Angew. Chem. Int. Ed. 2006, 45, 38–68.
m/z (%): 226.3 (100) [MꢀC15H32N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C30H64Br4N2O2U: C
34.56, H 6.19, N 2.69; found: C 34.52, H 6.24, N 2.52.
[9] M. Yoshio, T. Mukai, H. Ohno, T. Kato, J. Am. Chem. Soc. 2004,
126, 994–995.
[10] W. Dobbs, J. M. Suisse, L. Douce, R. Welter, Angew. Chem. 2006,
118, 4285–4288; Angew. Chem. Int. Ed. 2006, 45, 4179–4182.
[11] A. Taubert, Angew. Chem. 2004, 116, 5494–5496; Angew. Chem. Int.
Ed. 2004, 43, 5380–5382.
[12] A. Taubert, Z. Li, Dalton Trans. 2007, 723–727.
[13] C. J. Bowlas, D. W. Bruce, K. R. Seddon, Chem. Commun. 1996,
1625–1626.
[14] A. E. Bradley, C. Hardacre, J. D. Holbrey, S. Johnston, S. E. J.
McMath, M. Nieuwenhuyzen, Chem. Mater. 2002, 14, 629–635.
[15] W. Dobbs, L. Douce, L. Allouche, A. Louati, F. Malbosc, R. Welter,
New J. Chem. 2006, 30, 528–532.
4g: Yield: 20%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 254.5 (100) [MꢀC17H36N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C34H72Br4N2O2U: C
37.17, H 6.61, N 2.55; found: C 37.38, H 6.33, N 2.55.
5g: Yield: 23%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 268.6 (100) [MꢀC18H38N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C36H76Br4N2O2U: C
38.38, H 6.80, N 2.49; found: C 38.22, H 6.99, N 2.40.
6g: Yield: 31%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 282.5 (100) [Mꢀ
G
E
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C38H80Br4N2O2U: C
39.53, H 6.98, N 2.43; found: C 39.46, H 7.15, N 2.38.
7g: Yield: 34%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
[16] C. M. Gordon, J. D. Holbrey, A. R. Kennedy, K. R. Seddon, J.
Mater. Chem. 1998, 8, 2627–2636.
[17] J. D. Holbrey, K. R. Seddon, J. Chem. Soc. Dalton Trans. 1999,
2133–2139.
m/z (%): 296.5 (100) [MꢀC20H42N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C40H84Br4N2O2U: C
40.62, H 7.16, N 2.37; found: C 40.41, H 7.09, N 2.45.
8g: Yield: 40%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
[18] I. J. B. Lin, C. S. Vasam, J. Organomet. Chem. 2005, 690, 3498–3512.
[19] G. A. Knight, B. D. Shaw, J. Chem. Soc. 1938, 682–683.
[20] C. Cruz, B. Heinrich, A. C. Ribeiro, D. W. Bruce, D. Guillon, Liq.
Cryst. 2000, 27, 1625–1631.
m/z (%): 310.5 (100) [MꢀC21H44N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C42H88Br4N2O2U: C
41.66, H 7.33, N 2.31; found: C 41.59, H 7.47, N 2.26.
[21] L. Cui, V. Sapagovas, G. Lattermann, Liq. Cryst. 2002, 29, 1121–
1132.
[22] F. Neve, O. Francescangeli, A. Crispini, J. Charmant, Chem. Mater.
2001, 13, 2032–2041.
9g: Yield: 38%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 324.5 (100) [MꢀC22H46N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C44H92Br4N2O2U: C
42.66, H 7.49, N 2.26; found: C 42.49, H 7.17, N 2.31.
[23] L. Lu, N. Sharma, G. A. N. Gowda, C. L. Khetrapal, R. G. Weiss,
Liq. Cryst. 1997, 22, 23–28.
[24] D. J. Abdallah, A. Robertson, H. F. Hsu, R. G. Weiss, J. Am. Chem.
Soc. 2000, 122, 3053–3062.
[25] H. Chen, D. C. Kwait, Z. S. Gonen, B. T. Weslowski, D. J. Abdallah,
R. G. Weiss, Chem. Mater. 2002, 14, 4063–4072.
[26] K. Goossens, P. Nockemann, K. Driesen, B. Goderis, C. Gçrller-Wal-
rand, K. Van Hecke, L. Van Meervelt, E. Pouzet, K. Binnemans, T.
Cardinaels, Chem. Mater. 2008, 20, 157–168.
[27] S. Forsyth, J. Golding, D. R. MacFarlane, M. Forsyth, Electrochim.
Acta 2001, 46, 1753–1757.
[28] K. M. Johansson, J. Adebahr, P. C. Howlett, M. Forsyth, D. R. Mac-
Farlane, Aust. J. Chem. 2007, 60, 57–63.
10g: Yield: 43%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 338.6 (100) [MꢀC23H48N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C46H96Br4N2O2U: C
43.61, H 7.64, N 2.21; found: C 43.21, H 7.80, N 2.12.
11g: Yield: 45%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 352.6 (100) [Mꢀ
G
E
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C48H100Br4N2O2U: C
44.52, H 7.78, N 2.16; found: C 44.47, H 7.60, N 2.23.
12g: Yield: 41%. ESIMS (acetonitrile/water (80/20)+0.1% formic acid):
m/z (%): 366.6 (100) [MꢀC25H52N+ꢀ
ACHTUNGTRENNUNG
[UO2Br4]2ꢀ+Brꢀ]+; elemental analysis calcd (%) for C50H104Br4N2O2U: C
45.39, H 7.92, N 2.12; found: C 45.22, H 8.10, N 2.07.
[29] D. R. MacFarlane, P. Meakin, J. Sun, N. Amini, M. Forsyth, J. Phys.
Chem. B 1999, 103, 4164–4170.
[30] G. A. Baker, S. Pandey, S. Pandey, S. N. Baker, Analyst 2004, 129,
890–892.
Acknowledgements
[31] A. M. Dattelbaum, S. N. Baker, G. A. Baker, Chem. Commun. 2005,
939–941.
[32] N. Menschutkin, Z. Phys. Chem. (Leipzig), 1890, 6, 41–57.
[33] J. Golding, N. Hamid, D. R. MacFarlane, M. Forsyth, C. Forsyth, C.
Collins, J. Huang, Chem. Mater. 2001, 13, 558–564.
[34] A. J. Hill, J. Huang, J. Efthimiadis, P. Meakin, M. Forsyth, D. R.
MacFarlane, Solid State Ionics 2002, 154, 119–124.
[35] J. Efthimiadis, S. J. Pas, M. Forsyth, D. R. MacFarlane, Solid State
Ionics 2002, 154, 279–284.
K.G. is a Research Assistant of the FWO-Flanders (Belgium). T.C. and
K.D. are postdoctoral fellows of the FWO-Flanders. Financial support by
the K.U.Leuven (projects GOA 08/03 and IDO/05/005) and by the FWO-
Flanders (project G.0508.07) is gratefully acknowledged. Part of the
uranyl bromide salt was kindly provided by Diederik Huys. C,H,N ele-
mental analyses were performed by Dirk Henot. Mass spectra were mea-
sured by Bert Demarsin. Thermogravimetric traces were recorded by Di-
mitri Soccol. Powder X-ray diffractograms were recorded by Dr. Jan
D’Haen and Bart Ruttens (UHasselt, Institute for Materials Research).
[36] S. Clark, J. M. Elliott, J. R. Chipperfield, P. Styring, E. Sinn, Inorg.
Chem. Commun. 2002, 5, 249–251.
[37] J. M. Elliott, J. R. Chipperfield, S. Clark, E. Sinn, Inorg. Chem. 2002,
41, 293–299.
[38] J. L. Sessler, W. B. Callaway, S. P. Dudek, R. W. Date, D. W. Bruce,
Inorg. Chem. 2004, 43, 6650–6653.
[39] I. Aiello, M. Ghedini, A. Grisolia, D. Pucci, O. Francescangeli, Liq.
Cryst. 2005, 32, 763–769.
[40] T. Cardinaels, J. Ramaekers, D. Guillon, B. Donnio, K. Binnemans,
J. Am. Chem. Soc. 2005, 127, 17602–17603.
[1] K. Binnemans, Chem. Rev. 2005, 105, 4148–4204.
[2] E. Alami, H. Levy, R. Zana, P. Weber, A. Skoulios, Liq. Cryst. 1993,
13, 201–212.
[3] F. Tittarelli, P. Masson, A. Skoulios, Liq. Cryst. 1997, 22, 721–726.
[4] M. Arkas, D. Tsiourvas, C. M. Paleos, A. Skoulios, Chem. Eur. J.
1999, 5, 3202–3207.
[41] J. Shimizu, T. Nogami, H. Mikawa, Solid State Commun. 1985, 54,
1009–1011.
[5] K. Ohta, T. Sugiyama, T. Nogami, J. Mater. Chem. 2000, 10, 613–
616.
[6] A. Nikokavoura, D. Tsiourvas, M. Arkas, Z. Sideratou, C. M.
Paleos, Liq. Cryst. 2002, 29, 1547–1553.
[42] P. Dynarowicz, M. Godlewska, W. Witko, Mol. Cryst. Liq. Cryst.,
Sect. C: Molec. Mater. 1997, 8, 309–317.
Chem. Eur. J. 2009, 15, 656 – 674
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
673

T. Cardinaels et al.
[43] A. Babai, A. V. Mudring, Acta Crystallogr. Sect. E: Struct. Rep.
Online 2005, 61, o2913–o2915.
[44] C. B. Aakerçy, K. R. Seddon, Chem. Soc. Rev. 1993, 22, 397–407.
[45] T. Steiner, Acta Crystallogr. Sect. B: Struct. Sci. 1998, 54, 456–463.
[46] C. B. Aakerçy, T. A. Evans, K. R. Seddon, I. Palinko, New J. Chem.
1999, 23, 145–152.
[56] B. Donnio, B. Heinrich, H. Allouchi, J. Kain, S. Diele, D. Guillon,
D. W. Bruce, J. Am. Chem. Soc. 2004, 126, 15258–15268.
[57] C. Tschierske, J. Mater. Chem. 1998, 8, 1485–1508.
[58] K. Lunstroot, K. Driesen, P. Nockemann, C. Gçrller-Walrand, K.
Binnemans, S. Bellayer, J. Le Bideau, A. Vioux, Chem. Mater. 2006,
18, 5711–5715.
[59] P. Benson, A. Cox, T. J. Kemp, Q. Sultana, Chem. Phys. Lett. 1975,
35, 195–197.
[60] M. D. Marcantonatos, Inorg. Chim. Acta 1978, 26, 41–46.
[61] P. Nockemann, K. Servaes, R. Van Deun, K. Van Hecke, L. Van
Meervelt, K. Binnemans, C. Gçrller-Walrand, Inorg. Chem. 2007, 46,
11335–11344.
[62] R. G. Denning, T. R. Snellgrove, D. R. Woodwark, Mol. Phys. 1976,
32, 419–442.
[63] C. Gçrller-Walrand, S. De Jaegere, J. Chim. Phys. 1973, 70, 360–366.
[64] WinXPOW 1.06, STOE & Cie GmbH, Darmstadt (Germany), 1999.
[65] P. E. Werner, L. Eriksson, M. Westdahl, J. Appl. Crystallogr. 1985,
18, 367–370.
[47] G. R. Desiraju, Acc. Chem. Res. 2002, 35, 565–573.
[48] F. Neve, A. Crispini, S. Armentano, O. Francescangeli, Chem. Mater.
1998, 10, 1904–1913.
[49] F. Neve, O. Francescangeli, A. Crispini, Inorg. Chim. Acta 2002, 338,
51–58.
[50] C. Hardacre, J. D. Holbrey, P. B. McCormac, S. E. J. McMath, M.
Nieuwenhuyzen, K. R. Seddon, J. Mater. Chem. 2001, 11, 346–350.
[51] C. K. Lee, H. H. Peng, I. J. B. Lin, Chem. Mater. 2004, 16, 530–536.
[52] M. Yoshio, T. Ichikawa, H. Shimura, T. Kagata, A. Hamasaki, T.
Mukai, H. Ohno, T. Kato, Bull. Chem. Soc. Jpn. 2007, 80, 1836–
1841.
[53] J. Przedmojski, P. Dynarowicz-Ła¸tka, Phase Transitions 1999, 70,
133–146.
[54] G. W. Gray, J. W. Goodby, Smectic Liquid Crystals: Textures and
Structures, Leonard Hill, Glasgow, 1984.
[55] I. Dierking, Textures of Liquid Crystals, Wiley-VCH, Weinheim,
2003.
[66] SAINT, Version 5/6.0, Bruker Analytical X-ray Systems Inc., Madi-
son, Wisconsin, 1997.
[67] SHELXTL-PC, Version 5.1, Bruker Analytical X-ray Systems Inc.,
Madison, Wisconsin, 1997.
Received: July 31, 2008
Published online: November 26, 2008
674
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 656 – 674