Properties of new low melting point quaternary ammonium salts with bis(trifluoromethanesulfonyl)imide anion

Article abstract of DOI:10.1016/j.molstruc.2010.08.036

Eight new monocationic quaternary ammonium (QA) salts with the bis(trifluoromethanesulfonyl)imide (TFSI) anion were prepared by metathesis using our previously reported QA halides as precursors. New salts were characterized both in liquid and solid state

Full text of DOI:10.1016/j.molstruc.2010.08.036

Journal of Molecular Structure 983 (2010) 82–92
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Properties of new low melting point quaternary ammonium salts
with bis(trifluoromethanesulfonyl)imide anion
⇑
Minna Kärnä, Manu Lahtinen , Anna Kujala, Pirkko-Leena Hakkarainen, Jussi Valkonen
Department of Chemistry, University of Jyväskylä, P.O. Box 35, FI-40014 Jyväskylä, Finland
a r t i c l e i n f o
a b s t r a c t
Article history:
Eight new monocationic quaternary ammonium (QA) salts with the bis(trifluoromethanesulfonyl)imide
(TFSI) anion were prepared by metathesis using our previously reported QA halides as precursors. New
salts were characterized both in liquid and solid state using 1H and 13C NMR techniques, mass spectros-
copy and elemental analysis together with X-ray diffraction and thermoanalytical methods. In addition,
residual water content, viscosity and conductivity measurements were made for three of the room-
temperature ionic liquids (RTILs). The crystal structures of three compounds were determined by
X-ray single crystal diffraction. Powder diffraction was used to study the crystallinity of the solid salts
and to compare structural similarities between the single crystals and the microcrystalline bulk powders.
Three of the salts are liquid below room temperature, having broad liquid ranges (ꢀ300 °C), and in total
five out of eight salts melt below 100 °C. Moreover, powder diffraction data of the two RTILs were able to
be measured at sub-ambient temperatures using in situ low-temperature powder X-ray diffraction
revealing high crystallinity on both RTILs below their freezing point. The RTILs presented relatively high
conductivities (ꢀ0.1–0.2 S mꢁ1) and moderate to relatively low viscosities. The determined physicochem-
ical properties of the reported ILs suggest their applicability on various applications such as heat transfer
fluids, high temperature synthesis and lubricants.
Received 12 August 2010
Accepted 12 August 2010
Available online 22 August 2010
Keywords:
Quaternary ammonium salt
Bis(trifluoromethanesulfonyl)imide (TFSI)
Ionic liquids
Thermal analysis
X-ray diffraction
Physicochemical properties
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
efficient in palladium-catalyzed Suzuki reactions than the non-
substituted ones [27].
Room-temperature ionic liquids (RTILs) have already been
extensively studied by several authors due to their unique charac-
teristics which have enabled their use in numerous applications
[1–6]. Most of the previous studies still focus on costly imidazoli-
um based ionic liquids to which ammonium and phosphonium
based compounds are an alternative since they can be produced
from readily available low cost materials [7–10]. Applications of
ammonium-based ILs include their use as a reaction solvent
[11,12] and a recent study indicates that certain ammonium-based
ILs might possess anti-cancer activity [13]. Due to their wider elec-
trochemical window and higher cathodic stability QA ILs are
attractive alternatives to traditional aromatic ILs in electrochemi-
cal applications [14–17].
Building new types of ionic liquids with different functionalities
prepares the way for a broader range of applications. The aim of
our research in general is to develop new QA based salts that
encompass ionic liquid nature – favorably having a broad liquid
range, low viscosity, good electrochemical properties and low
melting point combined with good thermal stability and desired
solubility properties. This will be achieved by methods of crystal
engineering: pairing systematically newly designed functionalized
QA based cations with various anions, and performing systematic
structural and thermoanalytical evaluations of the novel materials
and exploring their physical properties. In our previous studies we
have reported several series of QA salts with different countera-
nions [28–33]. In this study, some of the previously synthesized
QA halides have been paired with the bis(trifluoromethanesulfo-
nyl)imide as the melting point lowering effect of the TFSI anion
is a well known fact. Because our study focuses on structural char-
acterization and thermal analysis of the QA salts, the performed
synthesis methods were not fully optimized, although majority of
the salts were already afforded in moderate to good yields.
Introducing an oxygen-containing side chain has been proven to
reduce the toxicity of certain ionic liquids [18,19] and it modifies
their solvent properties and thermal behaviour. It also has a ten-
dency to reduce the viscosity of QA-cation type ILs [20–26]. More-
over, ether-functionalized ILs have been proven to be more
The 1H and 13C NMR spectroscopy, elemental analysis and ESI
TOF MS measurements were used to verify chemical composition
and purity of the compounds. The potential presence of residual
⇑
Corresponding author. Fax: +358 14 2602501.
E-mail address: manu.k.lahtinen@jyu.fi (M. Lahtinen).
0022-2860/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2010.08.036

M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
83
bromide content in the TFSI salts were examined by inspecting
corresponding mass regions of relevant mass spectra. X-ray pow-
der diffraction was used to study the crystallinity of the salts and
to compare structural properties between the single crystal struc-
tures and their powdery forms, when applicable. The thermal
properties of the salts were examined by thermogravimetry (TG/
DTA) and differential scanning calorimetry (DSC) and the effect
of different anions to the characteristics of the salts were viewed
by comparing these results. Viscosity and conductivity, as well as
the UV–Vis spectra of the three room-temperature ionic liquids
were obtained by capillary electrophoresis and their residual water
content was determined with Karl Fischer titration. Densities for
the fluid salts 5, 7 and 8 were determined using a pycnometer.
Ar-C). ESI–TOF MS: m/z calculated for C18H20N2O4S2F6 [M-TFSI]+:
226.16; found [M-TFSI]+: 226.14. Elemental analysis: calculated
for C18H20N2O4S2F6: C, 42.68; H, 3.98; N, 5.53. Found C, 42.63; H,
3.98; N, 5.58.
2.1.3. (2) Dimethyldi(3-methylbenzyl) ammonium bis(trifluorometh-
ylsulfonyl)imide
Reagents: dimethyldi(3-methylbenzyl) ammonium chloride
(1.0 g, 3.5 mmol) and C2F6NO4S2 Li (0.93 g, 3.2 mmol). The yield
(white powder) was 1.8 g (95%).
1H NMR (CDCl3, 400 MHz, ppm): 2.37 (6 H, s, Ar-CH3), 2.89 (6 H, s,
N–CH3), 4.48 (4 H, s, N–CH2–), 7.24–7.34 (8 H, m, Ar-H). 13C NMR
(CDCl3, 100 MHz, ppm): 21.15 (2 C, Ar-CH3), 48.43 (2 C, N–CH3),
69.07 (2 C, N–CH2–), 115.12, 118.32, 121.51, 124.70 (2 C, –CF3),
126.35, 129.30, 129.98, 131.86, 133.60, 139.58 (12 C, Ar-C). ESI–
TOF MS: m/z calculated for C20H24N2O4S2F6 [M-TFSI]+: 254.19;found
[M-TFSI]+: 254.16. Elemental analysis: calculated for C20H24N2O4S2
F6: C, 44.94; H, 4.53; N, 5.24. Found C, 44.97; H, 4.67; N, 5.23.
2. Experimental section
2.1. Synthesis and analysis
2.1.1. General procedure
All reagents were used as achieved from manufacturers. The
precursors (QA halides) for compounds 1–8 (Scheme 1) were pre-
pared as described in our previous studies [29–33].
2.1.4. (3) Dimethyldi(4-methylbenzyl) ammonium bis(trifluoro
methylsulfonyl)imide
Reagents: dimethyldi(4-methylbenzyl) ammonium chloride (1.0 g,
3.5 mmol) and C2F6NO4S2 Li (1.11 g, 3.9 mmol). The yield (white
powder) was 1.7 g (84%).
The final TFSI salts were prepared by dissolving the QA halide in
a minimum volume of water and adding the stoichiometric
amount of lithium bis(trifluoromethanesulfonyl)-imide (C2F6NO4-
S2Li) dissolved in a minimum amount of water. The mixture was
stirred overnight at room temperature. The product, QA TFSI, pre-
cipitated from the solution or formed a separate liquid layer. The
raw product was filtered or decanted and washed with water until
the AgNO3-test was negative. If necessary, the product was recrys-
tallized from an appropriate solvent, i.e. dichloromethane. The
purified products were finally dried in vacuo.
1H NMR (CDCl3, 400 MHz, ppm): 2.37 (6 H, s, Ar-CH3), 2.85 (6 H,
s, N–CH3), 4.46 (4 H, s, N–CH2–), 7.23–7.35 (8 H, m, Ar-H). 13C NMR
(CDCl3, 100 MHz, ppm): 21.25 (2 C, Ar-CH3), 48.12 (2 C, N–CH3),
68.74 (2 C, N–CH2–), 115.11, 118.31, 121.50, 123.40 (2 C, –CF3),
123.23, 130.15, 132.86, 141.53 (12 C, Ar-C). ESI–TOF MS: m/z calcu-
lated for C20H24N2O4S2F6 [M-TFSI]+: 254.19; found [M-TFSI]+:
254.16. Elemental analysis: calculated for C20H24N2O4S2F6: C,
44.94; H, 4.53; N, 5.24. Found C, 45.00; H, 4.64; N, 5.20.
2.1.2. (1) Dibenzyl dimethylammonium bis(trifluoromethylsulfonyl)
imide
2.1.5. (4) Dimethyldi(4-methoxybenzyl) ammonium bis(trifluorometh
ylsulfonyl)imide
Reagents: dibenzyl dimethylammonium bromide (1.0 g, 3.2
mmol) and C2F6NO4S2 Li (0.84 g, 2.9 mmol). The yield (white pow-
der) was 1.4 g (88%).
Reagents: dimethyldi(4-methoxybenzyl) ammonium bromide
(1.0 g, 2.7 mmol) and C2F6NO4S2 Li (0.71 g, 2.5 mmol). The yield
(white powder) was 1.5 g (99%).
1H NMR (CDCl3, 400 MHz, ppm): 2.89 (6 H, s, N–CH3), 4.54 (4 H,
s, N–CH2–), 7.44–7.51 (10 H, m, Ar-H). 13C NMR (CDCl3, 100 MHz,
ppm): 48.35 (2 C, N–CH3), 69.02 (2 C, N–CH2–), 115.10, 118.30,
121.49, 124.67 (2 C, –CF3), 126.39, 129.52, 131.17, 133.01 (12 C,
1H NMR (CDCl3, 400 MHz, ppm): 2.82 (6 H, s, O–CH3), 3.82 (6 H,
s, N–CH3), 4.44 (4 H, s, N–CH2–), 6.93–7.39 (8 H, m, Ar-H). 13C NMR
(CDCl3, 100 MHz, ppm): 47.77 (2 C, O–CH3), 55.41 (2 C, N–CH3),
68.48 (2 C, N–CH2–), 114.85, 118.18, 134.38, 161.65 (12 C, Ar-C),
A–
A–
A–
A–
+
+
N
+
+
N
N
N
O
O
1
2
3
4
A–
A –
+
N
A–
A–
N+
+
N
+
N
O
O
O
O
8
7
6
5
Scheme 1. Molecular structures of the compounds 1–8. ‘‘Aꢁ” represents the TFSI-anion.

84
M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
115.14, 118.30, 121.50, 124.73 (2 C, –CF3). ESI–TOF MS: m/z calcu-
lated for C20H24N2O6S2F6 [M-TFSI]+: 286.18; found [M-TFSI]+:
286.16. Elemental analysis: calculated for C20H24N2O6S2F6: C,
42.40; H, 4.27; N, 4.96. Found C, 42.63; H, 4.30; N, 4.80.
CH2–), 3.50–3.54 (4 H, m, N–CH2–CH2–O–CH2–CH3), 3.77 (2 H, m,
N–CH2–CH2–O–). 13C NMR (CDCl3, 100 MHz, ppm): 13.63 (3 C,
–CH2–CH2–CH3), 14.81 (1 C, O–CH2–CH3), 21.53 (3 C, N–CH2–
CH2–CH2), 22.00, 28.13 (6 C, –CH2–CH2–CH3), 58.38, 67.15 (2 C,
N–CH2–CH2–O–CH2), 59.63 (3 C, N–CH2–CH2–O), 67.15 (1 C, N–
CH2–CH2–O–), 115.07, 118.27, 121.46, 124.66 (2 C, –CF3). ESI–TOF
MS: m/z calculated for C21H42N2O5S2F6 [M-TFSI]+: 300.33; found
[M-TFSI]+: 300.28. Elemental analysis: calculated for C21H42N2O5-
S2F6: C, 43.44; H, 7.29; N, 4.82. Found C, 42.65; H, 7.16; N, 5.05.
2.1.6. (5) Triethyl ethoxyethyl ammonium bis(trifluoro
methylsulfonyl)imide
Reagents: triethyl ethoxyethyl ammonium bromide (2.0 g,
6.8 mmol) and C2F6NO4S2 Li (2.03 g, 7.1 mmol). The yield (colour-
less liquid) was 2.6 g (84%).
1H NMR (CDCl3, 400 MHz, ppm): 1.18 (3 H, t, O–CH2–CH3), 1.31
(9 H, t, N–CH2–CH3), 3.35 (6 H, q, N–CH2–CH3), 3.39–3.42 (2 H, m,
N–CH2–CH2–O), 3.51 (2 H, q, O–CH2–CH3), 3.75–3.76 (2 H, m, N–
CH2–CH2–O). 13C NMR (CDCl3, 100 MHz, ppm): 7.41 (3 C, N–CH2–
CH3), 14.72 (1 C, O–CH2–CH3), 53.95 (3 C, N–CH2–CH3), 56.87 (1
C, N–CH2–CH2–O), 63.42 (1 C, N–CH2–CH2–O), 67.11 (1 C, O–CH2–
CH3), 115.07, 118.27, 121.46, 124.65 (2 C, –CF3). ESI–TOF MS: m/z
calculated for C12H24N2O5S2F [M-TFSI]+: 174.19; found [M-TFSI]+:
174.15. Elemental analysis: calculated for C12H24N2O5S2F6: C,
31.71; H, 5.32; N, 6.16. Found C, 31.52; H, 5.22; N, 6.29.
2.2. Characterization in the liquid state
The formation of the desired salts was confirmed by 1H and 13
C
NMR spectroscopy and ESI TOF mass spectroscopy and elemental
analysis was used to verify the purity of the salts. 1H and 13C
NMR spectra were measured in CDCl3 (or appropriate solvent) at
30 °C by using a Bruker Avance DRX 400 NMR spectrometer oper-
ating at 400 MHz for 1H and at 100 MHz for 13C. Mass spectra were
obtained by using the Micromass LCT time of flight (TOF) mass
spectrometer with electrospray ionization (ESI). The measure-
ments were made using the positive ion mode with a sample con-
centration of 25 mg/l in a methanol solution. The elemental
analyses were carried out with Vario EL III CHN elemental analyzer
using sample weights of 2–8 mg. For salts 5, 7 and 8, the residual
water contents were measured with Mettler Toledo C30 Coulomet-
ric Karl-Fischer titrator and densities were determined using a 5 ml
capillary pycnometer.
Temperature dependent dynamic viscosities of 5, 7 and 8 were
measured with Brookfield DV-III rheometer using conical geometry
(CPE-51 spindle) at temperatures 25–60 °C with a Brookfield TC-
500 water circulator. The sample volume was 0.5 ml. UV–Vis spec-
tra, viscosities of IL/molecular solvent mixtures and conductivity
measurements were performed with a HP CE3D G1600 AX appara-
tus (Agilent, Waldbronn, Germany) equipped with a diode array
detector (DAD) and an air cooling unit for the capillary. More de-
tailed descriptions of the measurements including calibration data
can be found in the Supplementary material.
2.1.7. (6) Tripropyl ethoxyethyl ammonium bis(trifluoromethylsulfo
nyl)imide
Reagents: tripropyl ethoxyethyl ammonium bromide (1.0 g,
3.0 mmol) and C2F6NO4S2 Li (0.87 g, 3.0 mmol). The yield (yellow
powder) was 1.2 g (80%).
1H NMR (CDCl3, 400 MHz, ppm): 1.01 (9 H, t, N–CH2–CH2–CH3),
1.20 (3 H, t, O–CH2–CH3), 1.68–1.74 (6 H, m, N–CH2–CH2–CH3),
3.18–3.23 (6 H, m, N–CH2–CH2–CH3), 3.50–3.53 (4 H, m, N–CH2–
CH2–O–CH2–CH3), 3.78 (2 H, m, N–CH2–CH2–O).
13C NMR (CDCl3, 100 MHz, ppm): 10.44 (3 C, N–CH2–CH2–CH3),
14.87 (1 C, O–CH2–CH3), 15.54 (3 C, N–CH2–CH2–CH3), 58.63, 67.21
(2 C, N–CH2–CH2–O–CH2–CH3), 61.23 (3 C, N–CH2–CH2–O), 63.66
(1 C, –CH2–O–CH2–) 115.10, 118.29, 121.49, 124.68 (2 C, –CF3).
ESI–TOF MS: m/z calculated for
C
15H30N2O5S2F6 [M-TFSI]+:
216.23; found [M-TFSI]+: 216.23. Elemental analysis: calculated
for C15H30N2O5S2F6: C, 36.28; H, 6.09; N, 5.64. Found C, 35.95; H,
6.11; N, 5.65.
2.3. X-ray single crystal diffraction structures
2.1.8. (7) Tributyl ethoxyethyl ammonium bis(trifluoromethylsulfonyl)
imide
Reagents: tributyl ethoxyethyl ammonium bromide (1.0 g,
2.9 mmol) and C2F6NO4S2 Li (0.76 g, 2.6 mmol). The yield (yellow
liquid) was 1.5 g (96%).
The crystal structures for compounds 2, 3 and 4 were determined
by X-ray single crystal diffraction. Colourless single crystals were
obtained by diffusion crystallization from suitable solvent, i.e. mix-
ture of methanol and diethyl ether or mixture of ethanol and
pentane.
1H NMR (CDCl3, 400 MHz, ppm): 0.99 (9 H, t, –CH2–CH2–CH3),
1.18 (3 H, t, O–CH2–CH3), 1.36–1.44 (6 H, m, –CH2–CH2–CH3),
1.58–1.66 (6 H, m, –CH2–CH2–CH3), 3.20–3.25 (6 H, m, N–CH2–
CH2–CH2–), 3.48–3.53 (4 H, m, N–CH2–CH2–O–CH2–CH3), 3.75 (2
H, m, N–CH2–CH2–O–). 13C NMR (CDCl3, 100 MHz, ppm): 13.39
(3 C, –CH2–CH2–CH3), 14.81 (1 C, O–CH2–CH3), 19.48 (3 C, –CH2–
CH2–CH3), 23.78 (3 C, –CH2–CH2–CH3), 58.41 (1 C, N–CH2–CH2–
O), 61.23 (3 C, N–CH2–CH2–CH2–),63.68 (1 C, N–CH2–CH2–O–),
67.16 (1 C, O–CH2–CH3), 115.07, 118.29, 121.48, 124.68 (2 C, –
CF3). ESI–TOF MS: m/z calculated for C18H36N2O5S2F6 [M-TFSI]+:
258.28; found [M-TFSI]+: 258.27. Elemental analysis: calculated
for C18H36N2O5S2F6: C, 40.14; H, 6.74; N, 5.20. Found C, 39.75; H,
6.55; N, 5.50.
The crystallographic data were recorded with a Kappa APEX II
diffractometer at ꢁ120 °C using graphite monochromatized Mo K
a
(k = 0.71073 Å) radiation. The data were processed with Denzo-
SMN v0.95.373 [34,35] and the absorption corrections were per-
formed using SADABS2008 [36]. The structures were solved by
using direct methods (SHELXS-97 [37] or SIR2004 [38] and refined
on F2 by full matrix least squares techniques (SHELXL-97 [39]) by
using anisotropic displacement parameters for all non-hydrogen
atoms. Hydrogen atoms, except those of solvents, were calculated
to their positions as riding atoms by using isotropic displacement
parameters. The isotropic displacement parameters were fixed to
be 1.2–1.5 times larger than those of the attached host atom. The
programs Diamond [40] and Mercury [41] were used for depicting
the crystal structures.
2.1.9. (8) Tripentyl ethoxyethyl ammonium bis(trifluoromethyl
sulfonyl)imide
Reagents: tripentyl ethoxyethyl ammonium bromide (1.0 g,
2.6 mmol) and C2F6NO4S2 Li (0.68 g, 2.4 mmol). The yield (yellow
liquid) was 1.3 g (86%).
2.4. X-ray powder diffraction analyses
The X-ray powder diffraction data was measured with PANalyt-
ical X’Pert PRO diffractometer in Bragg–Brentano geometry using
step-scan technique, and Johansson monochromator to produce
pure Cu Ka1 radiation (1.5406 Å; 45 kV, 30 mA). Hand ground
1H NMR (CDCl3, 400 MHz, ppm): 0.94 (9 H, t, –CH2–CH2–CH3),
1.21 (3 H, t, O–CH2–CH3), 1.34–1.43 (12 H, m, –CH2–CH2–CH3),
1.63–1.69 (6 H, m, N–CH2–CH2–), 3.21–3.25 (6 H, m, N–CH2–CH2–

M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
85
powder samples were prepared either on standard steel plate hold-
ers (having 10 or 16 mm diameter sample cavity) or on a silicon-
made zero-background holder (petrolatum jelly was used as an
adhesive). The data was collected by X’Celerator detector using con-
tinuous scanning mode in 2h-range of 4–70° with a step size of
0.017° and sample dependently counting time of 40–100 s per step.
Programmable divergence slit (PDS) was used to set irradiated
length on sample to 10 mm together with 10 or 15 mm incident
beam mask. Soller slits of 0.02° rad were used both on incident
and diffracted beam side together with anti-scatter slits 4° and
13 mm(4°and 6.6 mm for non-ambient measurements), respectively.
For in situ low-temperature measurements, the default sample
stage was changed to an Anton Paar TTK 450 low-temperature
chamber with automated height-controller of the stage. Vacuum
pump system and liquid N2 cooling unit were used to control mea-
suring conditions. The sample was prepared on a sample cavity in-
side a TTK 450 chamber and cooled under vacuum (0.129 mbar)
from 25 to ꢁ110 °C with a cooling rate of 10 °C/min by controlled
liquid nitrogen cooling. And then heated at a rate of 10 °C/min to
selected measurement temperatures such as ꢁ30 and ꢁ110 °C, at
which diffraction patterns were recorded isothermically.
bromide-ion was confirmed by the mass spectra. The residual
water content of RTILs 5, 7 and 8 was found to be <150 ppm after
drying in vacuo.
The temperature dependent dynamic viscosities of TFSI salts 5,
7 and 8 were made by rheometer and the results are presented in
Fig. 1. In the literature the viscosities of aliphatic monocationic
ammonium-based ILs with the TFSI anion have been reported to
vary between 50 and 200 mPa s at 25 °C [6]. Compared to those,
viscosities of these new ILs fall in the same category being
70 mPa s, 315 mPa s and 312 mPa s, respectively. At 60 °C the vis-
cosities drop to 18 mPa s, 43 mPa s and 48 mPa s. The densities of
compounds 5, 7 and 8 at room temperature were 1.36 g cmꢁ3
,
1.23 g cmꢁ3 and 1.19 g cmꢁ3, respectively.
Glass-forming liquids can be classified as either ‘‘strong” or
‘‘fragile”, depending on their behaviour [42]. One approach to
study this property is by using the so called Angell plot, log
g vs.
glass transition scaled temperature, Tg /T (Fig. 2). Liquids falling
on the right hand side of the plot, showing the largest deviation
from the Arrhenius behaviour, are considered as fragile. In the
Fig. 2 the unit Poise is used instead of the standard mPa s to facil-
itate comparisons to ILs found in the literature [43–48]. When
compared to other ILs the fragilities of these new compounds are
similar being rather fragile and therefore the properties have
strong temperature dependence near the Tg. Comparing the two
new salts, differences in their behaviour are very small which
could be expected since the compounds are very similar.
All diffraction data was converted from automatic slit mode
(ADS) to a fixed slit mode (FDS) data in X’Pert HighScore Plus v.
2.2d software package before further analyses. The simulated pow-
der diffraction patterns were calculated by the program Mercury
[41] from the CIFs of the QA salts under investigation.
In addition, the fluidities (gꢁ1
, g is viscosity) of compounds 5, 7
2.5. Thermal properties
and 8 were studied using Arrhenius plots (Fig. 3). To elucidate the
fluidity over longer liquid range the glass transition temperature is
included in the second plot (salts 5 and 8). At the glass transition
temperature, the value of fluidity is approximated to be 1011 [49].
Conductivity measurements were performed on RTILs 5, 7 and 8
using capillary electrophoresis. Details can be found in the elec-
tronic supplementary information (ESI), as well as the UV–Vis
spectra and the parallel viscosity measurements using electropho-
resis instead of rheometer. The conductivities of pure ILs were
The thermal transitions of the QA salts were examined on a
power compensation type Perkin Elmer Diamond DSC. Measure-
ments were carried out under nitrogen atmosphere (flow rate
50 ml/min) using 50 ll sealed aluminum pans. To ensure good
thermal contact between a sample and a pan and to minimize
the free volume inside the pan, the sealing was made using a
30 ll aluminum pan with pinholes. The temperature and energy
j
= 0.26 S mꢁ1
, j j
= 0.14 S mꢁ1
= 0.11 S mꢁ1 for 5, 7 and 8, respec-
= 0.90–
calibration was made using In metal standard (156.6 °C; 28.45 J/
g). The samples were heated at a rate of 5–10 °C/min either from
ꢁ60 °C or ꢁ50 °C to a temperature close to the observed decompo-
sition temperature of each salt and then cooled down to the start-
ing temperature at a rate of 5 °C/min. The cycle was repeated twice
and the sample weights used were about 2–20 mg. The extreme
low temperature DSC-scans were made using a Mettler Toledo
DSC829 DSC. The samples were heated under nitrogen atmosphere
(50 ml/min) at a rate of 10 °C/min from ꢁ120 °C to 50 °C and
cooled back to ꢁ120 °C at a rate of 5 °C/min. The cycle was re-
peated twice. The sample pans were cold pressed and holes were
punctured into the lid to release any forming pressure, sample
weights of 5–10 mg were used. The calibration of the instrument
was made using hexane, water, indium and zinc.
The thermal decomposition paths for compounds 1–8 were ob-
tained with PerkinElmer TGA 7 thermogravimetric analyzer. Mea-
surements were carried out in open platinum pan under air
atmosphere (flow rate of 50 ml/min) with heating rate of 5 °C/min
on temperature range of 25–900 °C. The temperature calibration of
the analyzerwas made usingCurie-point calibrationtechnique (Alu-
mel, Ni, Perkalloy, Fe). The weight balance was calibrated by mea-
suring the standard weight of 50 mg at room temperature. The
sample weights used in the measurements were about 5–20 mg.
tively. The conductivities could be noticeably increased (
j
2.14 S mꢁ1) by using IL-methanol mixtures in the solvent molar
fraction range of 0.89–0.97 (see ESI). Determined values are lower
than that of traditional conductive solvents but fall into the same
range with some previously reported similar monocationic QA ILs
[6].
Moreover, combining the data from viscosity, conductivity and
density measurements we find the molar conductivities of these
5
7
8
400
350
300
250
200
150
100
50
Temp (°C)
Viscosity (mPa s)
5
7
8
25
30
35
40
45
50
55
60
71
55
43
35
29
24
20
18
315
222
159
117
88
312
230
167
126
97
68
75
54
59
43
48
η
3. Results and discussion
0
25
30
35
40
45
50
55
60
3.1. Characterization in liquid and solid state
Temp (°C)
According to the 1H, 13C NMR spectra and mass spectra the salts
were free of solvents and any other impurities. The absence of the
Fig. 1. Temperature dependent viscosities of 5, 7 and 8 obtained by rheometer.

86
M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
13
11
9
0.6
0.4
0.2
Strong
0.0
-0.2
-0.4
-0.6
-0.8
5
8
7
5
0.55
0.60
0.65
0.70
3
1
-1
-3
-5
Fragile
0.8
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.9
1.0
Tg/T
Fig. 2. Angell plot of ionic liquids 5 and 8.
5
5
8
-1.2
-1.4
-1.6
-1.8
-2.0
-2.2
-2.4
-2.6
-2
7
8
-4
-6
-8
-10
-12
-14
3.0
3.1
3.2
3.3
3.4
3.0
3.5
4.0
4.5
5.0
5.5
6.0
1000/T (K-1)
1000/T (K-1)
Fig. 3. Left: Arrhenius plots of fluidities of salts 5, 7 and 8. Right: Arrhenius plot of the fluidities of 5 and 8 including the values in glass transition temperature.
compounds to be 0.87 S cm2 molꢁ1
,
0.61 S cm2 molꢁ1 and
tions as well as several weak hydrogen–fluorine interactions (C–
Hꢂ ꢂ ꢂF < 2.66 Å) (Fig. 5a). Each cation is connected to three different
anions by these contacts and the hydrogen to fluorine distances
vary from 2.60 to 2.65 Å with bond angles from 123° to 134°,
respectively. The shortest distance between nitrogen and fluorine
is 4.38(3) Å and between cation nitrogen and anion nitrogen atoms
is 4.45(5) Å. Cations are also packed via intermolecular methyl
0.54 S cm2 molꢁ1, and the Walden products 60.8 S Pa s cm2 molꢁ1
,
193.1 S Pa s cm2 molꢁ1 and 167.5 S Pa s cm2 molꢁ1, respectively.
Compared to previously reported QA ILs [6] the molar conductivi-
ties for all three ILs are similar. The Walden product for compound
5 is again similar to previously reported, but for compounds 7 and
8 the values are noticeably higher, due to higher viscosities.
C–H-
phenyl groups as can be seen in Fig. 5b. The cations form pairs
via C–H- interactions arranging in layers along the crystallo-
graphic a-axis. These pairs then form infinite chains via face to face
interactions. Similarly the anions are packed in ‘‘planar”
p and face to face p–p interactions between the substituted
3.2. X-ray single crystal diffraction structure analysis
p
The crystallographic data of compounds 2, 3 and 4 are pre-
sented in Table 1 and the selected bond angles and distances in Ta-
ble 2. The solidified salts other than above did either not form
single crystals in re-crystallization or the quality of the crystals
was inadequate for structure determination.
Compound 2 crystallizes in a monoclinic space group P21/n with
one cation and one anion in the asymmetric unit being structurally
similar to the analogous chloride salt [33]. Cation appears in a W-
conformation whereas the anion appears in trans-conformation
having the trifluoromethyl groups on the opposite sides of the
imide core (Fig. 4). The packing is affected by Coulombic interac-
p–p
zigzag chains along the (ꢁ1 0 1) plane.
Compound 3 crystallizes in triclinic spacegroup P-1 having one
cation and a disordered anion in the asymmetric unit (Fig. 6).
Again, the cations appear in W-conformation and the anions are
in trans-conformation as in the structure 2. The cation is connected
to one anion with a weak C–Hꢂ ꢂ ꢂF interaction (<2.61 Å) with bong
angle of 145°. The cation also has weak C–Hꢂ ꢂ ꢂO interactions
between six anions, one of which is bifurcated. The lengths of
these interactions vary from 2.34 to 2.68 Å with bond angles of

M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
87
Table 1
Crystallographic data for compounds 2, 3 and 4.
Compound
2
3
4
Formula
C20H24F6N2O4S2 C20H24F6N2O4S2 C20H24F6N2O6S2
Mr (g/mol)
Crystal system
Space group
543.53
Monoclinic
P21/n (No. 14)
534.53
Triclinic
P-1 (No. 2)
566.53
Orthorhombic
P212121 (No.
19)
a (Å)
b (Å)
c (Å)
12.078(2)
15.200(3)
13.338(3)
90
103.85(3)
90
2377.6(8)
4
1.493
8.7581(4)
9.0398(7)
16.4596(7)
100.897(2)
96.011(3)
110.487(2)
1177.83(12)
2
14.204(3)
17.094(3)
20.323(4)
90
90
90
4934.5(17)
8
1.525
a
(°)
b (°)
c
(°)
V (Å3)
Z
qcalcd (g/cm3)
1.507
0.303
l
(mmꢁ1
)
0.301
0.300
F (0 0 0)
1104
552
2336
Crystal size (mm3)
h range (°)
0.2 ꢃ 0.2 ꢃ 0.3
2.04–28.46
30,067
5959
0.08 ꢃ 0.3 ꢃ 0.4 0.3 ꢃ 0.3 ꢃ 0.36
2.53–25
15,101
4134
2.74–28.78
37,756
12,368
Reflections collected
Independent
reflections
Fig. 4. The molecular structure and labelling scheme of compound 2.
Data/restraints/
parameters
5959/0/307
4134/0/313
12,368/0/649
GooF
1.096
1.062
1.010
being pointing either to the same side or on opposite sides of the
cation. The anions are in trans-conformation as in the structure 2
also forming helical chains (Fig. 9a). The packing is affected by Cou-
lombic interactions and also by several weak interactions. Each
cation is connected to one anion with a weak C–Hꢂ ꢂ ꢂF interaction
(<2.67 Å) with bong angles of 120° and 127°, respectively. Weak
C–Hꢂ ꢂ ꢂO interactions between two cations and eight anions also
exist, some of which are bifurcated. The length of these interac-
tions varies from 2.38 to 2.66 Å with bond angles of 129–170°,
respectively. The shortest distance between cation nitrogen and
anion nitrogen atoms is 4.01(3) Å. Cations are also packed via
R (int)
Final R indices
[I > 2r(I)]
0.0637
R1 = 0.0665
0.0643
R1 = 0.1039
0.0516
R1 = 0.0533
wR2 = 0.1590
R1 = 0.1028
wR2 = 0.3073
R1 = 0.1240
wR2 = 0.0914
R1 = 0.0847
R indices (all data)
wR2 = 0.1801
wR2 = 0.3189
wR2 = 0.1015
0.37 and ꢁ0.30
Largest diff. peak/
hole (e/Å3)
Flack parameter
0.38 and ꢁ0.39 1.57–0.58
–
–
0.114(51)
Table 2
Selected bond lengths (Å) and bond angles (°) with e.s.d.s for salts 2, 3 and 4.
intermolecular face to face
p–p interactions between the substi-
tuted phenyl groups as can be seen in Fig. 9b.
Compound
2
3
4
N(1)–C(11)
N(1)–C(21)
N(1)–C(31)
N(1)–C(41)
C(11)–N(1)–C(21)
C(11)–N(1)–C(31)
C(11)–N(1)–C(41)
C(21)–N(1)–C(31)
C(21)–N(1)–C(41)
C(31)–N(1)–C(41)
1.496(4)
1.501(4)
1.531(4)
1.533(4)
109.4(2)
109.6(2)
110.8(2)
110.65(2)
109.5(2)
106.8(2)
1.527(9)
1.537(9)
1.492(9)
1.506(9)
107.0(5)
110.3(6)
110.1(6)
110.7(6)
109.6(6)
109.1(5)
1.532(4)
1.526(4)
1.497(3)
1.494(3)
105.8(2)
110.5(2)
109.9(2)
110.9(2)
110.4(2)
109.2(2)
3.3. X–ray powder diffraction analysis
The powder X-ray diffraction (PXRD) patterns for compounds 2, 3
and 4 are presented in Fig. 10 together with the simulated patterns
from the corresponding single crystal structures. For salts 2 and 3,
the similarity between the bulk powder and the single crystal struc-
ture is obvious. Whereas on compound 4 few unidentified peaks for
example at 10.5°, 11.7°, 12.1°, 13.1°, 14.7°, 15.9° and 21.0° in 2h can
be seen. These peaks indicate presence of either a polymorphor a hy-
drated form, from which the polymorph is the most likely candidate
(see thermal analysis part). Compounds 1 and 6 are also crystalline
(see ESI, Fig. S4) but did not produce any good quality single crystals
suitable for structure determination.
To get further insight of the structural properties on the RTILs 5
and 7, powder diffraction patterns were measured also at sub-
ambient temperatures; ꢁ30 and ꢁ110 °C for 5 and ꢁ30 °C for 7
(compound 8 was not measured since it formed only a glassy solid).
As tentatively suggested by the sharp melting/crystallization tran-
sitions obtained with DSC, both ILs proved to be highly crystalline
in the solid phase. The powder diffraction patterns along with the
corresponding DSC-scans are presented in Figs. 11 and 12, respec-
tively. Salt 5 is completely amorphous at ꢁ110 °C whereas highly
crystalline above cold crystallization that occurs about at ꢁ30 °C
then melting sharply at ꢁ1.6 °C. In case of 7, diffraction pattern
have been presented using square rooted y-axis due to dramatic
preferred orientation effect that can be observed on diffraction
peaks 8.2° and 16.5° 2h. The visual inspection of the sample while
in the chamber revealed that the preferred orientation is induced
by diffraction occurring from a highly oriented film-surface instead
136–172°, respectively. The shortest distance between cation
nitrogen and anion nitrogen atoms is 3.99(2) Å. Along with
Coulombic interactions the cations are packed via face to face
interactions between the substituted phenyl groups as can be
seen in Fig. 7. The cations then form infinite chains along the crys-
tallographic c-axis via face to face interactions. The residual
p–
p
p–p
electron density with highest peak 1.57 e Åꢁ3 is caused by the dis-
ordered anion. However, the observed disorder was too severe to
be modelled properly around the defined anion coordinates but
it is clearly caused by anion packing in helical chains in tunnel like
loose void along c-axis (Fig. 7), so that the trifluoromethyl groups
of the adjacent anions are against each other. Slight positional
variation of these chains in the crystal lattice causes the observed
residual density around the anion.
Compound 4 crystallizes in orthorhombic space group P212121
having two cations and two anions in the asymmetric unit
(Fig. 8). The structure is analogous with the corresponding bromide
[29]. The cations yet again appear in W-conformation, the two cat-
ions mainly differing by the orientation of the methoxy groups

88
M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
Fig. 5. (A) The packing of compound 2 left: viewed along the crystallographic c–axis. Right: along the (ꢁ1 0 1) plane (B) C–H-
p and face to face p–p interactions forming the
network of connections between cations. The anions have been removed for clarity.
space reserved for each atom including some space for hydrogen
atom) addressing also potential correctness of the found unit cell
The final unit cell refinement resulted good Pawley fit indicating
correctness of the determined cell settings and space group P-1
(Pawley v
2 = 4.52%, Rwp = 0.15 with 403 reflections). The structure
determination was attempted by DASH [52] program that is based
on simulated annealing method in which known structure moiety/
moieties are needed to structure determination. The needed struc-
ture moieties for a cation and an anion was retrieved from the
known structures and introduced to the DASH program. Unfortu-
nately the data quality of the low-temperature diffraction pattern
was still inadequate for successful structure determination as only
the spatial position of the anion was able to be found whereas the
position of the cation remained ambiguous. In case of compound
7, few tentative monoclinic and orthorhombic unit cell settings
were obtained but structure determination attempts failed on each
time with tested cell and space group settings.
Fig. 6. The molecular structure and labelling scheme of compound 3.
of a powder, as sample resided in the sample cavity in a form of
transparent coating. Similar transparent coat-formation was ob-
served also for 5, but diffraction peaks are clearly less affected by
crystal orientation effects. Diffraction patterns were used to inves-
tigate whether the low-temperature powder structures could be
determined for 5 and 7. In case of 5 first 19 diffraction peak posi-
tions were used as an input for the auto-indexing procedure made
by DICVOL04 [50]. All the peaks can repeatedly be indexed by tri-
clinic unit cell setting a = 9.2376(21) Å, b = 8.4410(39) Å, c =
3.4. Thermal properties
The summarized results of the DSC measurements together
with the decomposition temperatures obtained by TG/DTA are pre-
sented in Table 3. The examples of the DSC-scans are in Fig. 12 (see
ESI for DSCs) and TG curves for all compounds are presented in
Fig. 13. To emphasize the crystallization properties of the salts
when they are cooled from a liquid state and then reheated, results
of the second heating scans are also included in the Table 3. Many
of the ILs, including one salt reported in this study, can form glassy
states via a supercooled liquid state and thus can be cooled down
below expected freezing point without initiation of crystallization
[53]. Some of the ILs in glassy state can be recrystallized via a
cold crystallization when heated above the glass transition
13.9017(69) Å,
a = 84.833(29)°, b = 92.900(32)°, c = 103.945(29)°
and V = 1047.28 Å3 with reasonably high figure of merits
(M19 = 43.1, F19 = 125.8 (0.0058, 26)) [51]. Similar unit cell settings
were acquired also by comparative auto-indexing in X’pert High-
Score Plus. The given unit cell dimension could adequately be filled
by two ion pair units of 5 (2 ꢃ (27 non-hydrogen atoms ꢃ 20 Å3

M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
89
Fig. 7. (A) The packing of compound 3 left: along (1 ꢁ1 0) plane, right: along c-axis and (B) face to face
p–p interactions forming the network of connections between cations.
The anions have been removed for clarity.
respondingly but the re-crystallization on subsequent scans is
incomplete which can be seen from the magnitude of melting en-
thalpy. Compounds of the second group, salts 4 and 5, supercool on
cooling and turn into a glass via glass transitions at low tempera-
ture. However, on reheating a cold crystallization occurs above
the glass transition, followed by re-melting at temperature similar
to that observed on the first heating scan. The re-crystallization of
compound 4 on subsequent scans is incomplete and the presence
of minor fraction of one or several polymorphs can be found in
the DSC-scan. Salt 5 also exhibits polymorphism depending on
the cooling rate (see ESI Fig. S7 for additional DSC-scans). Com-
pound 8 is liquid at and below room temperature having a liquid
range of 347 °C. By cooling it turns into glass but shows no other
transitions. In all, five compounds 2, 5–8, are liquid below 100 °C
and two salts somewhat above that.
The thermal decomposition of compounds 1–8 occur in one
stage without identifiable cleavages. The decomposition of the aro-
matic compounds (1–4) starts at temperatures 244–306 °C and of
the RR03NþAꢁ compounds (5–8) at 279–320 °C. Compared to the
previously published analogous compounds with different anions
the thermal stability is significantly increased. Only the PFꢁ6 anion
results in similar thermal stabilities, but compounds with the TFSI
anion also exhibit lower melting points [29–31,33].
Fig. 8. The molecular structure and labelling scheme of compound 4.
4. Conclusions
temperature, whereas certain salts will enter to a supercooled
liquid state and simply form new glassy state when cooled again
below the glass forming temperature.
Eight new small low melting point quaternary ammonium salts
having TFSI as an anionic part were synthesized and characterized
both in liquid and solid state. Together with determining various
physicochemical properties of these salts, single crystal structures
of three salts were determined. Five out of eight salts melt below
100 °C whereas three of the salts are fluid at room temperature.
Viscosities of 5, 7 and 8 at room temperature are 70 mPa s,
315 mPa s and 312 mPa s, respectively and decrease rapidly with
increase of temperature being below 50 mPa s already at 60 °C.
All salts except 8 exhibit a melting transition on both heating
scans. The aromatic salts (1–4) have melting points between 79
and 105 °C and the RR03NþAꢁ salts between <ꢁ60 and 62 °C.
The thermal behaviour of the compounds, apart from com-
pound 8, can roughly be divided in two groups. The first group is
formed by salts 1, 3, 6 and 7. They crystallize from a melt on cool-
ing and re-melt on the second heating (Table 3). Both the melting
points and their enthalpies observed on the second heating corre-
spond well with those found on the first heating indicating good
re-crystallization properties to form a same structural form with
equal crystallinity from a melted state. Compound 2 behaves cor-
These salts also exhibit rather good conductivities (0.1–0.2 S mꢁ1
)
which can be increased substantially (up to 2.1 S mꢁ1) using IL/
molecular solvent mixtures. It is expected that these new ther-
mally stable and low melting ionic liquids may have many new

90
M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
Fig. 9. (A) The packing of compound 4 up: viewed along the crystallographic c-axis. Down: along the b-axis. (B) Face to face
p
–p
interactions forming the network of
connections between cations. The anions have been removed for clarity.
2 measured
2 simulated
5 at -110 °C
5 at -30 °C
7 at -30 °C
3 measured
3 simulated
4 measured
4 simulated
4
8
12
16
20
24
28
32
36
2θ
40
(°)
5
10
15
20
25
30
35
40
2θ (°)
Fig. 10. Experimental PXDR patterns of 2 and 3 and 4 compared with simulated
data, which are from the single crystal structure parameters.
Fig. 11. The powder diffraction patterns of compound 5 (ꢁ110 and ꢁ30 °C) and 7
(at ꢁ30 °C) measured by in situ low-temperature X-ray diffraction.
perspectives in various fields of applications such as, heat transfer
fluids, electrolytes (for instance dye-sensitized solar cells on which
they will be tested in near future), high temperature synthesis sol-
vents and lubricants.

M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
91
perä for the ESI TOF MS-measurements. M.K. gratefully acknowl-
edges the financial support of the Inorganic Materials Chemistry
Graduate Program.
3
Appendix A. Supplementary material
4
7
Supporting information includes details of capillary electropho-
resis measurements, additional PXRD patterns and DSC-scans.
CCDC 775,442, 775,443 and 780,593 contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif. Supplementary data
associated with this article can be found, in the online version, at
doi:10.1016/j.molstruc.2010.08.036.
8
-50
0
50
100
150
Temp. (°C)
Fig. 12. DSC-curves for salts 3, 4, 7 and 8.
References
[1] K.R. Seddon, J. Chem. Technol. Biotechnol. 68 (1997) 351–356.
[2] T. Welton, Chem. Rev. 99 (1999) 2071–2083.
[3] J.D. Holbrey, K.R. Seddon, Clean Prod. Proc. 1 (1999) 223–236.
[4] M.J. Earle, K.R. Seddon, Pure Appl. Chem. 72 (2000) 1391–1398.
[5] R. Sheldon, Chem. Commun. (2001) 2399–2407.
Table 3
Thermal properties of compounds 1–8.
Comp M.W. 1st Heating scan 2nd Heating scan
Decomp.
[6] P. Wasserscheid, T. Welton (Eds.), Ionic Liquids in Synthesis, second ed., Wiley-
VCH, Germany, 2008.
Tm and
D
H
Tg and
D
Cp
Tm and
D
H
Td
[7] S.T. Handy, Chem. Eur. J. 9 (2003) 2938–2944.
1
2
3
4
506.48 101.71 (23.7)
534.54 78.66 (33.2)
534.54 132.99 (31.5)
566.54 105.31 (23.1)
–
–
–
101.42 (24.3) 306
77.60 (25.7) 295
131.05 (30.0) 277
105.59 (4.2) 244
[8] L. Ferguson, P. Scovazzo, Ind. Eng. Chem. Res. 46 (2007) 1369–1374.
[9] A. Grosse Boewing, A. Jess, Chem. Eng. Sci. 62 (2007) 1760–1769.
[10] G.J. Kemperman, T.A. Roeters, P.W. Hilberink, Eur. J. Org. Chem. (2003) 1681–
1686.
[11] A. Marra, A. Vecchi, C. Chiappe, B. Melai, A. Dondoni, J. Org. Chem. 73 (2008)
2458–2461.
[12] A. Vecchi, B. Melai, A. Marra, C. Chiappe, A. Dondoni, J. Org. Chem. 73 (2008)
6437–6440.
[13] V. Kumar, S.V. Malhotra, Bioorg. Med. Chem. Lett. 19 (2009) 4643–4646.
[14] T. Sato, G. Masuda, K. Takagi, Electrochim. Acta. 49 (2004) 3603–3611.
[15] K. Yuyama, G. Masuda, H. Yoshida, T. Sato, J. Power Sources 162 (2006) 1401–
1408.
ꢁ18.51
(0.2)
5
454.45 ꢁ1.65 (21.4)
ꢁ84.10
(0.1)
ꢁ1.65 (20.9)a 320
14.98 (21.7)b
496.53 62.08 (30.4)
538.61 9.76 (9.0)
ꢁ84.2 (0.1)b 15.11(21.5)b
6
7
8
ꢁ
58.72 (31.8)
10.29 (00.8)
–
294
282
279
ꢁ
580.69
–
ꢁ68.66
(0.2)
[16] H. Zheng, G. Liu, V. Battaglia, J. Phys. Chem. C. 114 (2010) 6182–6189.
[17] T. Sato, T. Maruo, S. Marukane, K. Takagi, J. Power Sources 138 (2004) 253–261.
[18] C. Samori, A. Pasteris, P. Galletti, E. Tagliavini, Environ. Toxicol. Chem. 26
(2007) 2379–2382.
[19] S. Stolte, J. Arning, U. Bottin-Weber, A. Mueller, W. Pitner, U. Welz-Biermann,
B. Jastorff, J. Ranke, Green Chem. 9 (2007) 760–767.
Tm = melting temperature, Tg = glass transitions temperature,
D
H = enthalpy of a
transition (kJ molꢁ1),
D
Cp = heat capacity change [kJ molꢁ1 °Cꢁ1], Td = decomposi-
tion onset by TG/DTA.
a
Exhibits cold crystallization (crystallization on heating) between glass transi-
tion and melting.
b
[20] L.C. Branco, J.N. Rosa, J.J.M. Ramos, C.A.M. Afonso, Chem. Eur. J. 8 (2002) 3671–
3677.
Show also two polymorphs depending on the cooling rate.
[21] Q. Liu, M.H.A. Janssen, F. van Rantwijk, R.A. Sheldon, Green Chem. 7 (2005) 39–
42.
[22] Z. Zhou, H. Matsumoto, K. Tatsumi, Chem. Eur. J. 11 (2005) 752–766.
[23] W.A. Henderson, V.G. Young Jr., D.M. Fox, H.C. De Long, P.C. Trulove, Chem.
Commun. (2006) 3708–3710.
[24] H.S. Schrekker, M.P. Stracke, C.M.L. Schrekker, J. Dupont, Ind. Eng. Chem. Res.
46 (2007) 7389–7392.
[25] H.S. Schrekker, D.O. Silva, M.A. Gelesky, M.P. Stracke, C.M.L. Schrekker, R.S.
Goncalves, J. Dupont, J. Braz. Chem. Soc. 19 (2008) 426–433.
[26] L.J.A. Siqueira, M.C.C. Ribeiro, J. Phys. Chem. B. 113 (2009) 1074–1079.
[27] X. Yang, Z. Fei, T.J. Geldbach, A.D. Phillips, C.G. Hartinger, Y. Li, P.J. Dyson,
Organometallics 27 (2008) 3971–3977.
[28] M. Kärnä, M. Lahtinen, P. Hakkarainen, J. Valkonen, Aust. J. Chem. 63 (2010)
1122–1137.
[29] M. Kärnä, M. Lahtinen, J. Valkonen, J. Mol. Struct. 922 (2009) 64–76.
[30] S. Busi, M. Lahtinen, H. Mansikkamäki, J. Valkonen, K. Rissanen, J. Solid State
Chem. 178 (2005) 1722–1737.
[31] S. Busi, M. Lahtinen, J. Ropponen, J. Valkonen, K. Rissanen, J. Solid State Chem.
177 (2004) 3757–3767.
[32] J. Ropponen, M. Lahtinen, S. Busi, M. Nissinen, E. Kolehmainen, K. Rissanen,
New J. Chem. 28 (2004) 1426–1430.
1
100
75
2
3
4
50
25
0
100
5
6
75
7
8
50
25
0
[33] S. Busi, M. Lahtinen, M. Kärnä, J. Valkonen, E. Kolehmainen, K. Rissanen, J. Mol.
Struct. 787 (2006) 18–30.
25
100
200
300
400
500
600
700
Temp (°C)
[34] Z. Otwinowski, W. Minor, Methods Enzymol. 276 (1997) 307–326.
[35] Z. Otwinowski, D. Borek, W. Majewski, W. Minor, Acta Crystallogr. Sect. A:
Found. Crystallogr. A59 (2003) 228–234.
[36] G.M. Sheldrick, SADABS, Multi-Scan Absorption Correction Program, Version 2,
2001.
Fig. 13. TG curves of compounds 1–8 measured under air with heating rate of 5 °C/
min.
[37] G.M. Sheldrick, Acta Crystallogr. Sect. A: Found. Crystallogr. A46 (1990) 467–
473.
Acknowledgements
[38] M.C. Burla, R. Caliandro, M. Camalli, B. Carrozzini, G.L. Cascarano, L. De Caro, C.
Giacovazzo, G. Polidori, R. Spagna, J. Appl. Crystallogr. 38 (2005) 381–388.
[39] G.M. Sheldrick, Acta Crystallogr. Sect. A: Found. Crystallogr. A64 (2008) 112–
122.
We thank Ms. Elina Hautakangas for the elemental analyses, Mr.
Reijo Kauppinen for the NMR-measurements and Ms. Mirja Lahti-
[40] K. Brandenburg, Crystal Impact GbR, Bonn, Germany, 2005.

92
M. Kärnä et al. / Journal of Molecular Structure 983 (2010) 82–92
[41] C.F. Macrae, P.R. Edgington, P. McCabe, E. Pidcock, G.P. Shields, R. Taylor, M.
Towler, J. van de Streek, J. Appl. Crystallogr. 39 (2006) 453–457.
[42] C.A. Angell, J. Phys. Chem. Solids 49 (1988) 863–871.
[43] W. Xu, E.I. Cooper, C.A. Angell, J. Phys. Chem. B 107 (2003) 6170–6178.
[44] J. Belieres, C.A. Angell, J. Phys. Chem. B 111 (2007) 4926–4937.
[45] M. Anouti, M. Caillon-Caravanier, Y. Dridi, H. Galiano, D. Lemordant, J. Phys.
Chem. B 112 (2008) 13335–13343.
[46] M. Anouti, M. Caillon-Caravanier, C. Le Floch, D. Lemordant, J. Phys. Chem. B
112 (2008) 9412–9416.
[47] I. Bandres, B. Giner, I. Gascon, M. Castro, C. Lafuente, J. Phys. Chem. B 112
(2008) 12461–12467.
[48] C. Brigouleix, M. Anouti, J. Jacquemin, M. Caillon-Caravanier, H. Galiano, D.
Lemordant, J. Phys. Chem. B 114 (2010) 1757,1758–1766.
[49] C.T. Moynihan, P.B. Macedo, C.J. Montrose, P.K. Gupta, M.A. DeBolt, J.F. Dill, B.E.
Dom, P.W. Drake, A.J. Easteal, et al., Ann. N. Y. Acad. Sci. 279 (1976) 15–35.
[50] A. Boultif, D. Louer, J. Appl. Crystallogr. 37 (2004) 724–731.
[51] P.M. de Wolff, J. Appl. Crystallogr. 1 (1968) 108–113.
[52] W.I.F. David, K. Shankland, J. van de Streek, E. Pidcock, W.D.S. Motherwell, J.C.
Cole, J. Appl. Crystallogr. 39 (2006) 910–915.
[53] P.S. Kulkarni, L.C. Branco, J.G. Crespo, M.C. Nunes, A. Raymundo, C.A.M. Afonso,
Chem. Eur. J. 13 (2007) 8478–8488.