Eu3 + as a dual probe for the determination of IL anion donor power: A combined luminescence spectroscopic and electrochemical approach

Article abstract of DOI:10.1016/j.molliq.2014.03.005

This work is aimed at giving proof that Eu(Tf₂N)₃ (Tf₂N = bis(trifluoromethanesulfonyl)amide) can act as both an optical and electrochemical probe for the determination of the Lewis acidity of an ionic liquid anion. For that reason the luminescence spectra and cyclic voltammograms of dilute solutions of Eu(Tf₂N)₃ in various ionic liquids were investigated. The Eu^{2 +/3 +} redox potential in the investigated ILs can be related to the Lewis basicity of the IL anion. The IL cation had little influence. The lower the determined halfwave potential, the higher the IL anion basicity. The obtained ranking can be confirmed by luminescence spectroscopy where a bathochromic shift of the ⁵D ₀ → ⁷F₄ transition indicates a stronger Lewis basicity of the IL anion.

Full text of DOI:10.1016/j.molliq.2014.03.005

Journal of Molecular Liquids 192 (2014) 191–198
Contents lists available at ScienceDirect
Journal of Molecular Liquids
journal homepage: www.elsevier.com/locate/molliq
Eu3+ as a dual probe for the determination of IL anion donor power:
A combined luminescence spectroscopic and electrochemical approach
Arash Babai, Gabriel Kopiec, Anastasia Lackmann, Bert Mallick, Slawomir Pitula,
⁎
Sifu Tang, Anja-Verena Mudring
Inorganic Chemistry III — Materials Engineering and Characterization, Ruhr-Universität Bochum, D-44780 Bochum, Germany
Materials Science and Engineering, Iowa State University, Critical Materials Institute, Ames Laboratory, Ames, IA 50011, USA
a r t i c l e i n f o
a b s t r a c t
Available online 22 March 2014
This work is aimed at giving proof that Eu(Tf2N)3 (Tf2N = bis(trifluoromethanesulfonyl)amide) can act as both
an optical and electrochemical probe for the determination of the Lewis acidity of an ionic liquid anion. For that
reason the luminescence spectra and cyclic voltammograms of dilute solutions of Eu(Tf2N)3 in various ionic
liquids were investigated. The Eu2+/3+ redox potential in the investigated ILs can be related to the Lewis basicity
of the IL anion. The IL cation had little influence. The lower the determined halfwave potential, the higher the IL
anion basicity. The obtained ranking can be confirmed by luminescence spectroscopy where a bathochromic shift
Keywords:
Acidity
Electrochemistry
Europium
Ionic liquids
Rare-earth elements
of the 5D0
→
7F4 transition indicates a stronger Lewis basicity of the IL anion.
© 2014 Published by Elsevier B.V.
1. Introduction
solvatochromic dyes or transition metal complexes in the solvents
under investigation. The absorption or emission bands of the probe
show a strong shift in their optical spectra (solvatochromic shift)
which can be related to the polarity of the solvent in which they are dis-
solved [4,5]. However, it has been realized that it is difficult to analyze
and separate specific interactions in the liquid which becomes even
more important in non-classical, highly structured solvents such as
ionic liquids.
For ionic liquids Kamlet–Taft parameters derived from solvatochromic
studies of organic dyes have been associated with different solvent prop-
erties, such as π* with dipolarity and polarizability, α with the hydrogen
bond acceptor capability and β with the hydrogen bond acceptor capabil-
ity. Although it was noticed that α is largely dependent on the IL anion
and β on the IL cation, an analysis of separate influences of cation and
anion that would ultimately allow for a designed choice of the cation–
anion combination was not possible. But the unique power of ionic liquids
is that through a smart choice of the cation and anion a solvent with
designed properties can be created. For that reason it would be interest-
ing to have probes that allow testing specifically for certain solvent
properties.
Ionic liquids (ILs) are receiving growing attention as solvent replace-
ments for conventional organic solvents [1]. Often they are considered
to be environmentally more benign and chemically safe because many
ILs have low flammabilities and flash points. However, the most im-
portant advantage that ionic liquids offer over conventional organic
solvents (and water) is the tuneability of their solvent properties
through the choice of the respective cation/anion combination. It is
estimated that roughly 1018 ionic liquids are accessible [2]. Each
ionic liquid with a specific cation–anion combination offers a new
set of chemical and physical properties. In consequence, ILs can be
truly designed for a certain application. However, to do so, it is im-
portant to be able to know and predict fundamental solvent proper-
ties such as polarity and acidity.
Solvent polarity and acidity will determine the solubility of sub-
stances and are able to influence the outcome of chemical reactions,
the position of chemical equilibria as well as reaction rates [3]. It is not
an easy task to determine the acidity of non-aqueous media, but several
methods have been used to describe the solvent power of ionic liquids
addressing their polarity. A common approach for determining the sol-
vent polarity is to evaluate the UV–Vis spectra of optical probes such as
Solvent polarity as such is rather complex. IUPAC defines solvent
polarity as being “the overall solvation capability (or solvation power)
for (i) starting materials and products, which influences chemical equi-
librium; (ii) reactants and activated complexes (“transition states”),
which determines reaction rates; and (iii) ions or molecules in their
ground and first excited state, which is responsible for light absorptions
in the various wavelength regions. This overall solvation capability de-
pends on the action of all, non-specific and specific, intermolecular
⁎
Corresponding author.
E-mail address: anja.mudring@rub.de (A.-V. Mudring).
http://dx.doi.org/10.1016/j.molliq.2014.03.005
0167-7322/© 2014 Published by Elsevier B.V.

192
A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
solute–solvent interactions, excluding such interactions leading to defi-
nite chemical alterations of the ions or molecules of the solute” [6].
According to Reichardt solvent–solute interactions can be divided
into specific and non-specific interactions [7]. Non-specific interactions
are instantaneous/induced dipole forces (dispersion or London forces),
dipole/induced-dipole forces (induction or Debye forces), dipole/dipole
forces (orientation or Keesom forces) and ion/dipole forces (Coulomb
forces) which might quite well get addressed through measurements
of continuum properties such as dielectric constants. Specific interac-
tions include hydrogen bond donor and/or hydrogen bond acceptor in-
teractions, electron pair donor/electron pair acceptor or charge–transfer
interactions, solvophobic interactions (which can become important
only in highly structured solvents). As ionic liquids contain, in contrast
to conventional molecular solvents, charged ions it is expected that
electron–donor–acceptor interactions have a rather high influence. In
consequence, an important property of ionic liquids as solvents is
their Lewis acidity and basicity or, in other words, their electron accep-
tor and donor capability. It is anticipated that the electron acceptor
properties of an IL depend largely on the cation whereas the IL anion
governs the IL electron donor properties.
Methylimidazole (98%), N-methylpyrrolidine (98%), the respective hal-
ogen alkanes (N98%) were purchased from Acros, Geel, B and distilled
prior to use.
2.2. General purification procedure for ILs
The respective IL was diluted with CH2Cl2 or CHCl3 and washed sev-
eral times with deionized water to remove any excess of halides or alkali
metal salts (AgNO3 test) as well as unreacted starting material. After
filtration of the solution over a column with neutral Al2O3 and active
charcoal the solvent was removed under high vacuum and the ILs
dried under dynamic vacuum for 1–2 days at 80–90 °C.
2.3. Synthesis of HTf2N
HTf2N was obtained by sublimation at 70 °C from a solution of LiTf2N
in concentrated sulfuric acid under reduced pressure (10−3 mbar). The
crude product was resublimed for further purification. Yield: 90%.
1H−NMRðD2OÞ : δðppmÞ ¼ 4:77ðs; 1HÞ
We could confirm this hypothesis through evaluation of spectro-
scopic properties of certain transition metal compounds in different
ionic liquids. They made it possible to separate the anionic influ-
ence, hence Lewis basicity. We were able to rank various IL anions
with respect to their Lewis basicity by evaluating the absorption
spectra of the indicator complex (Ni(tmen)(acac)][B(Ph)4] (tmen =
tetramethylethylendiamine, acac = acetylacetonate) [8]. In a second
step we transferred Duffy's concept of the optical basicity [9] to ILs by
using Mn(Tf2N)2 as a spectroscopic probe [10].
19 F−NMRðD2OÞ : δðppmÞ ¼ −79:16ðs; 6FÞ
n
o
13C 19 F −NMRðD2OÞ : δðppmÞ ¼ 19:27ðs; 2CÞ
When investigating in a different context the solvation and li-
gand exchange processes of ytterbium(III) salts in ionic liquids
such as (C4mpyr)(Tf2N) (C4mpyr = N-butyl-N-methylpyrrolidinium)
or (C4mpyr)(TfO) we made an interesting observation: We found that
the Yb2+/Yb3+ redox potential strongly depends on the local chemical
environment and coordination of Yb3+ by the IL anion. It was found that
the more coordinating the IL anion is, the less negative the redox
potentials (vs. Fc/Fc+) [11]. Thus, it also should be possible to elec-
trochemically determine the Lewis basicity of an IL with a given
anion by investigating the redox potential of a dissolved lanthanide
bis(trifluoromethanesulfonyl)amide, Ln(Tf2N)3. An additional ad-
vantage of using lanthanide salts as a probe is, that many of the tri-
valent lanthanides show distinct photoluminescence. As Yb3+ is
spectroscopically inactive in the visible range of the electromagnetic
spectrum, we choose for this study Eu3+ which is commonly used as
a structural probe. To date several Eu3+ containing ionic liquids and
liquid crystals are known [12]. In case of Eu3+ it is possible to monitor
the electron donation power of the anion via the shift of (hyper-) sensi-
tive transitions in the luminescence spectra. Thus, by using the same
probe, Eu(Tf2N)3, but different analytical tools, namely cyclic voltamm-
etry and luminescence spectroscopy, it should be possible to access the
electron donor power of an IL anion. To check whether Eu(Tf2N)3 can
work as both, an optical and electrochemical probe for the anion
Lewis basicity we have investigated solutions of Eu(Tf2N)3 in a set of
ionic liquids with anions of weak (e.g. Tf2N−), middle (e.g. TfO−) and
strong (e.g. DCA−) Lewis basicity and recorded the respective lumines-
cence spectra and cyclic voltammograms.
2.4. Synthesis of Eu(Tf2N)3
Eu2O3 was suspended in deionized water and five-fold molar excess
of an aqueous HTf2N solution was added dropwise. After complete dis-
solution of Eu2O3 the water was boiled off until a slurry solid appeared.
This slurry was transferred to a Schlenk tube and dried at 140–160 °C
under high vacuum. The residual solid was sublimed for purification
under reduced pressure at 270 °C.
Elemental analysis : EuðTf2NÞ3
calc:N 4:23%:C 7:26%; H 0:00%; S 19:39 %
found N 4:19%:C 7:25%; H 0:10%; S 20:34%
2.5. Synthesis of 1-alkyl-3-methylimidazolium bromides
(CnmimBr) bromides (n = 2, 3, 4 and 6) were obtained by alkylation
of 1-methylimidazole with the respective halogen alkane [13]. All bro-
mides except C6mimBr, which is liquid, were obtained as crystalline
white powders.
ðC2mimÞBr:1H−NMR ðCDCl3Þ : δ ðppmÞ
¼ 1:23 ðt; 3H; J ¼ 7:4 HzÞ; 3:75 ðs; 3HÞ; 4:06 ðq; 2H; J ¼ 7:2 HzÞ;
7:39 ðs; 2HÞ; 9:82 ðs; 1HÞ
2. Experimental section
2.1. Chemicals and synthesis
LiTfO (98%), Eu(TfO)3 (98%) and Eu2O3 (99.9%) were purchased from
Sigma-Aldrich, Schnelldorf, D. LiTf2N, (N4441)(Tf2N), (C2mim)(MeSO3),
(C2mim)(TfO) and (C4pyr)(Tf2N) (all 99%) were purchased from
IoLiTec, Heilbronn, D. (P666 14)Cl (95%) was provided by Cytec, Stamford,
USA. All commercial ILs were purified prior to the experiments.
ðC3mimÞBr:1H−NMR ðCDCl3Þ : δ ðppmÞ
¼ 0:44 ðt; 3H; J ¼ 7:4 HzÞ; 1:45 ðhex; 2H; J ¼ 7:2 HzÞ; 3:61 ðs; 3HÞ;
3:82 ðt; 2H; J ¼ 7:2 HzÞ; 7:82 ðd; 2H; J ¼ 1:7 HzÞ; 9:67 ðs; 1HÞ

A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
193
ðC4mimÞBr:1H−NMR ðCDCl3Þ : δ ðppmÞ ¼ 0:85 ðt; 3H; J ¼ 7:4 HzÞ;
ðC6mimÞðTf2NÞ:1H−NMR ð; CDCl3Þ : δðppmÞ ¼ 0:83 ðt; 3H; J ¼ 7:4 HzÞ;
1:26 ðm; 6HÞ; 1:81 ðpen; 2HÞ; 3:87 ðs; 3HÞ; 4:11 ðt; 2H; J ¼ 7:4 HzÞ;
7:30 ðd; 2H; J ¼ 1:7 HzÞ; 8:61 ðs; 1HÞ
1:27 ðhex; 2H; J ¼ 7:1; 7:4 HzÞ; 1:81 ðpen; 2H; J ¼ 7:1; 7:4 HzÞ;
4:02 ðs; 3HÞ; 4:24 ðt; 2H; J ¼ 7:1 HzÞ; 7:56 ðd; 2H; J ¼ 1:7 HzÞ; 10:19 ðs; 1HÞ
ðC6mimÞBr:1H−NMR ðCDCl3Þ : δ ðppmÞ ¼ 0:75 ðt; 3H; J ¼ 7:4 HzÞ;
1:20 ðm; 6HÞ; 1:81 ðpen; 2HÞ; 2:05 ðs; 2HÞ; 4:03 ðs; 3HÞ;
19 F−NMR ðCDCl3Þ : δðppmÞ ¼ −79:27 ðs; 6HÞ
ðC3mpyrÞðTf2NÞ1H−NMR ð; CDCl3Þ : δðppmÞ ¼ 0; 91 ðt; 3H; J ¼ 7:3 HzÞ;
1:68 ðm; 2HÞ; 2:08 ðm; 4HÞ; 2:97 ðs; 3HÞ; 3:25 ðm; 2HÞ; 3:34 ðm; 4HÞ
4:22 ðt; 2H; J ¼ 7:4 HzÞ; 7:55 ðd; 2H; J ¼ 1:6 HzÞ; 10:24 ðs; 1HÞ
2.6. Synthesis of 1-propyl-1-methylpyrrolidinium bromide
19 F−NMR ðCDCl3Þ : δðppmÞ ¼ −79; 43 ðs; 6HÞ
n-Propyl bromide was dropwise added to a solution of 1-
methylpyrrolidin and dichloromethane. The mixture was stirred
for 24 h, washed with ethyl acetate and concentrated. The crude prod-
uct was dropwise added to cold ethyl acetate and the white crystalline
product was filtered off. The product was dried under dynamic vacuum
for 1–2 days. Yield: 50.8%.
ðP666 14ꢀðTf2NÞ1H−NMR ð; CDCl3Þ : δðppmÞ ¼ 0:83–0:95 ðm; 12HÞ;
1:22–1:60 ðm; 48HÞ; 2:01–2:19 ðm; 8HÞ
ðC3mpyrÞBr:1H−NMR ðCDCl3Þ : δðppmÞ ¼ 1; 05 ðt; 3H; J ¼ 7:3 HzÞ;
1; 84 ðm; 2HÞ; 2; 28 ðs; 4HÞ; 3; 27 ðs; 3HÞ; 3; 59 ðm; 2Þ; 3; 80 ðs; 4HÞ
19 F−NMR ðCDCl3Þ : δðppmÞ ¼ −78:93 ðs; 6FÞ
31P−NMR ðCDCl3Þ : δ ¼ 32:89 ðs; 1PÞ
2.7. Synthesis of 1-alkyl-3-methylimidazolium, ammonium and phosphoni-
um bis(trifluoromethane)sulfonylamides
2.8. Synthesis of 1-butyl-3-methylpyrrolidinium dicyanamide
The respective halide IL was dissolved in deionized water (pH = 6)
and after an equimolar amount of LiNTf2 in water had been added
dropwise, the reaction mixture was stirred for 1 day at 70 °C. Then
CH2Cl2 was added and the aqueous phase was removed. The organic
phase was washed halide-free with deionized water (AgNO3 test). The
solution was filtered over a column filled with neutral Al2O3 and activat-
ed charcoal. The organic solvent was removed under reduced pressure
and the reaction product finally dried under dynamic vacuum for 1–
2 days at 80–90 °C.
1-Butyl-3-methylimidazolium bromide was dissolved in deionized
water and a slight molar excess of silver dicyanamide suspended in de-
ionized water was added. The mixture was stirred for 2 h at 60 °C, then
filtered and dried. The crude product was dissolved in CH2Cl2, upon
cooling to 5 °C most of the silver bromide precipitated. After separation
from silver halide by filtration, the remaining solution was filtered over
a column filled with neutral Al2O3 and activated charcoal. The solvent
was removed under reduced pressure and the IL dried under dynamic
vacuum for 1–2 days at 80–90 °C.
ðC2mimÞðTf2NÞ:1H−NMR ðCDCl3Þ : δðppmÞ ¼ 1:58 ðt; 3H; J ¼ 9:0 HzÞ;
4:06 ðpen; 2H; J ¼ 9:0 HzÞ; 7:73 ðd; 2H; J ¼ 2:2 HzÞ; 9:00 ðs; 1HÞ
ðC4mimÞðDCAÞ:1H−NMR ðCDCl3Þ : δðppmÞ ¼ 0:85 ðt; 3H; J ¼ 7:2 HzÞ;
1:27 ðhex; 2H; J ¼ 7:6 HzÞ; 1:81 ðpen; 2H; J ¼ 7:6 HzÞ; 4:02 ðs; 3HÞ;
4:24 ðt; 2H; J ¼ 7:4 HzÞ; 7:56 ðd; 2HÞ; 10:19 ðs; 1HÞ
19 F−NMR ð; CDCl3Þ : δðppmÞ ¼ −79:96 ðs; 6HÞ
2.9. Synthesis of 1-butyl-1-methylimidazolium triflate
ðC3mimÞðTf2NÞ:1H−NMR ðCDCl3Þ : δðppmÞ ¼ 0:87 ðt; 3H; J ¼ 7:4 HzÞ;
1:82 ðhex; 2H; J ¼ 7:2 HzÞ; 3:84 ðs; 3HÞ; 4:05 ðt; 2H; J ¼ 7:2 HzÞ;
7:29 ðd; 2H; J ¼ 1:9 HzÞ; 8:53 ðs; 1HÞ
1-Butyl-3-methylimidazolium bromide was dissolved in acetone/
acetonitrile (50:50) and an equimolar amount of LiTfO in acetone was
added. The mixture was stirred for 1 day at 60 °C and then filtered.
The solvent was removed under reduced pressure and the crude prod-
uct dissolved in CH2Cl2. Upon cooling to 5 °C most of the Li halide pre-
cipitated and the precipitate was filtered off. The remaining solution
was washed halide-free with deionized water (AgNO3 test) and filtered
over a column filled with neutral Al2O3 and activated charcoal. The re-
sidual organic phase was freed from solvent under reduced pressure
and dried under dynamic vacuum for 1–2 days at 80–90 °C.
19 F−NMR ðCDCl3Þ : δðppmÞ ¼ −79:43 ðs; 6HÞ
ðC4mimÞðTf2Nꢀ:1H−NMR ðCDCl3Þ : δðppmÞ ¼ 0:89 ðt; 3H; J ¼ 7:1 HzÞ;
1:31 ðhex; 2H; J ¼ 7:1; 7:4 HzÞ; 1:80 ðpen; 2H; J ¼ 7:2; 7:4 HzÞ; 3:87 ðs; 3HÞ;
4:11 ðt; 2H; J ¼ 7:2 HzÞ; 7:30 ðd; 2HÞ; 8:59 ðs; 1HÞ
ðC4mimÞðTfOÞ1H−NMR ðacetone−d6Þ : δðppmÞ ¼ 0:96 ðt; 3H; J ¼ 7:4 HzÞ;
1:40 ðhex; 2HÞ; 1:84 ðm; 2HÞ; 2:25 ðs; 4HÞ; 3:20 ðt; 3H; J ¼ 7:3 HzÞ;
3:55 ðm; 2HÞ; 3:70 ðm; 4HÞ
19 F−NMR ðCDCl3Þ : δðppmÞ ¼ −79:32 ðs; 6HÞ

194
A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
resulting mixture was stirred for 1 day at 60 °C. After cooling to room
temperature the mixture was diluted with CH2Cl2 and filtered over a
column filled with neutral or acidic Al2O3 and active charcoal. The sol-
vent was removed under vacuum and the residual dried under high vac-
uum for 1–2 days at 80–90 °C. Caution: Although we have never
observed any sudden decomposition, organic perchlorates are known
615 nm
593 nm
690/695 nm
Wavelength/nm
614 nm
Fig. 1. Electrochemical cell used for cyclic voltammetry measurements under inert
conditions.
19 F−NMR ð:Acetone−d6Þ : δ ¼ −78:88 ðs; 3FÞ
591 nm
2.10. Synthesis of 1-butyl-3-methylimidazolium perchlorate
689/695 nm
1-Butyl-3-methylimidazolium bromide was dissolved in CH2Cl2 in a
flask and an equimolar amount of NaClO4∙H2O suspended in CH2Cl2 was
added. Na2SO4 was added to remove the crystal water of the sodium
salt. The resulting mixture was stirred for 3 days at RT. The precipitated
alkali halide was filtered off and the remaining reaction mixture then
washed halide free (AgNO3 test). After filtration of the solution over a
column filled with neutral or acidic Al2O3 and active charcoal the resid-
ual organic phase was freed from solvent and dried under dynamic vac-
uum for 1–2 days at 80–90 °C. General caution should be taken as
perchlorates are know to violently decompose when dry or heated to el-
evated temperatures.
Wavelength/nm
614/618.5 nm
ðC4mimÞðClO4Þ:1H‐NMR ðDMSO−d6Þ : δðppmÞ ¼ 0:86 ðt; 3H; J ¼ 7:4 HzÞ;
1:24 ðhex; 2H; J ¼ 7:4 HzÞ; 1:76 ðpen; 2H; J ¼ 7:2 HzÞ; 3:84 ðs; 3HÞ;
4:14 ðt; 2H; J ¼ 7:2 HzÞ; 7:61 ðd; 2H; J ¼ 2:0 HzÞ; 8:99 ðs; 1HÞ
692.5/700 nm
592.5 nm
13C−NMR ðDMSO−d6Þ : δ ¼ 12:5 ðsÞ; 18:1 ðsÞ; 30:6 ðsÞ; 35:0 ðsÞ; 47:9 ðsÞ;
121:5 ðsÞ; 122:8 ðsÞ; 135:7 ðsÞ
2.11. Synthesis of 1-ethyl-3-methylimidazolium ethylsulfate
Wavelength/nm
Et2SO4 was added dropwise to an equimolar amount of dry
methylimidazole. After complete addition of the starting materials the
Fig. 2. From top to bottom: Emission spectra of Eu(Tf2N)3 in (C4mim)(Tf2N), (C4mpyr)(Tf2N)
and in (C4mpyr)(DCA).

A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
195
Table 1
5D0
various ionic liquids.
2.12. Sample preparation for luminescence measurements and cylic volt-
ammetry studies
→
7FJ (J = 0–4) transition maxima in nm for 0.025 M solutions of Eu(Tf2N)3 in
5D0
→
7F1
5D0
→
7F2
5D0 7F4
→
Eu(Tf2N)3 and Eu(TfO)3 were dissolved in the respective IL under
inert conditions to obtain 0.025 M solutions in a glovebox. The dissolu-
tion process was facilitated by ultrasound and heating to 70 °C.
(C2mim)(Tf2N)
(C3mim)(Tf2N)
(C4mim)(Tf2N)
(C6mim) (Tf2N)
(C3mpyr)(Tf2N)
(C4mpyr)(Tf2N)
(N444 1)(Tf2N)
(P666 14)(Tf2N)
(C2mim)(ClO4)
(C2mim)(MeSO3)
(C4mim)(TfO)
(C4mpyr)(DCA)
(C2mim)(EtSO4)
592
592
593
592
592
591
592
588/596
591
591/594
591
615
614
615
614
615
614
615
688/695
689/695
690/695
690/695
689/695
689/695
690/695
690/695/699.5*
689/697
688/698
699
3. Instrumentation
3.1. Elemental Analyses
612.5/615/619.5
614/618
612
611/616
614/618.5 (616)
612/615/619
Elemental analyses were performed using a Euro Vektor CHNS-O
Euro EA3000 analyzer. The samples were sealed in small tin tubes in
the glove box.
592.5
592
692.5/700
697/702
3.2. NMR
All NMR spectra were recorded on a Bruker DPX AC 300 spectrome-
ter with an automatic sample changer.
to decompose violently especially at elevated temperatures. Cautious
handling of the substance is highly recommended.
3.3. Photoluminescence
ðC2mimÞðEtSO4Þ:1H−NMR ð; CDCl3Þ : δðppmÞ ¼ 0:85 ðt; 3H; J ¼ 7:2 HzÞ;
1:27 ðhex; 2H; J ¼ 7:2 HzÞ; 1:81 ðpen; 2HÞ; 4:02 ðs; 3HÞ;
4:24 ðt; 2H; J ¼ 7:5 HzÞ; 7:56 ðs; 2HÞ; 10:19 ðs; 1HÞ
Excitation and emission photoluminescence spectra were recorded
at room temperature on a Fluorolog 3 (Horiba Jobin Yvon GmbH,
München, D) with a 450 W xenon lamp as a steady state excitation
Fig. 3. Cyclic voltammograms of Eu(Tf2N)3 in (C4mim)(Tf2N) vs. Fc/Fc+ at scan rates of a) 10 mV/s, b) 50 mV/s, c) 100 mV/s and d) 200 mV/s.

196
A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
source, a double excitation monochromator, an emission monochroma-
tor and a photomultiplier for detection. For excitation spectra the 5D0 →
7F2 transition of Eu3+ was monitored, to obtain the emission spectra the
level of Eu3+ was excited into the most intense absorption which is the
4. Results and discussion
For a proof of concept that Eu(Tf2N)3 can act as both, a spectroscopic
and electrochemical probe for the Lewis basicity of the ionic liquid anion
three prominent IL anions of different Lewis acidity were chosen. Previ-
ous investigations indicate that bis(trifluoromethanesulfonyl)amide,
(Tf2N)−, is an anion of weak Lewis basicity, trifluorosulfonate, (TfO)−,
an anion of medium Lewis basicity and dicyanamide, (DCA)− an anion
of strong Lewis basicity. As counter cations typical IL cations such
as 1-alkyl-3-methylimidzolium, quarternary ammonium and phos-
phonium cations with different alkyl chain lengths were chosen. A
prerequisite for the study was that the chosen ILs had to have a
low viscosity to ensure that the thermodynamic equilibrium can be
reached fast and to come as close as possible to a reversible electro-
chemical reaction.
The basis for the detection of the ionic liquid anions' Lewis basicity
through Eu(Tf2N)3 is that the Tf2N anion is an extremely weakly co-
ordinating anion (WCA) [15]. Any stronger coordinating IL anion
will expel Tf2N- from the coordination sphere of Eu3+ upon disso-
lution of Eu(Tf2N)3. Such ligand exchange reactions for Ln(Tf2N)3
(Ln = lanthanide) have been evidenced in several ILs by multiple
techniques. It was observed that upon dissolution of ytterbium(III)
bis(trifluoromethanesulfonyl)amides in (C4mpyr)(TfO) the TfO−
anion, as the stronger Lewis base, replaces the Tf2N− anions completely
in the coordination sphere of Yb(III) [16]. The respective complex an-
ions [Yb(TfO6)]3− and [Yb(Tf2N)5]2− could be characterized unambigu-
ously by X-ray crystal structure analysis.
7F0
→
7L6 transition. All spectra were recorded at room temperature.
Electronic transitions were assigned according to the Dieke energy
level diagram for trivalent 4f-elements. [14]
3.4. Cyclic voltammetry
The measurements carried out under N2 as a protection gas in a
custom-made electrochemical cell (Fig. 1) with a three electrode align-
ment. An Ag rod reference electrode (Metrohm AG, Filderstadt, D), a
1 mm Pt wire counter electrode (VWR) and a 2 mm diameter Pt disk
working electrode (Metrohm AG, Filderstadt, D) were used.
All cyclic voltammograms were referenced against the ferrocene/
ferrocenium redox couple. The redox potential of this redox couple
was determined in the IL under investigation and the respective correc-
tion applied to the collected data. The solutions of Eu3+ in ferrocene as
well as the solutions of Eu(Tf2N)3 in the ILs were equilibrated 5 s prior to
measurement and the third cycle scan was used for data analysis. Be-
tween two measurements the built Helmholtz electrical double layer
was destroyed by stirring and gently heating the sample above room
temperature and checked for a steady state potential as an indicator
for the thermodynamic equilibrium after cooling to room temperature.
The steady state potential was used as the starting potential.
Fig. 4. Cyclic voltammograms of Eu(Tf2N)3 in (C4mim)(DCA) vs. Fc/Fc+ at different scan rates of a) 10 mV/s, b) 50 mV/s, c) 100 mV/s and d) 200 mV/s.

A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
197
Gradual exchange of Tf2N− in the ligand sphere of transition metals
by stronger coordinating ligands has also been investigated by mo-
lecular dynamic simulations [17]. EXAFS studies on ligand exchange
processes confirm that ligand exchange of the weakly coordinating
Tf2N− by stronger coordinating anions takes place for Eu(Tf2N)3
[18]. It is assumed that Eu(Tf2N)3 will show a behavior similar to
that of Yb(Tf2N)3 and that e.g. [Eu(Tf2N)5]2− as found in (C4mpyr)2
[Eu(Tf2N)5] [19] will form when Eu(Tf2N)3 is dissolved in a Tf2N-IL
and [Eu(TfO)6]3− forms when Eu(Tf2N)3 is dissolved in a TfO-IL. Sim-
ilarly, ligand exchange will occur when Eu(Tf2N)3 is dissolved in a
DCA-IL like (C4mim)(DCA).
the main origin of this behavior. In the more viscous ionic liquids like
(P666 14](Tf2N) (η@20 °C = 165 cP [21]) and (C4mpyr)(TfO)(η@20 °C =
158 cP [22]) as well as (C4mim)(TfO)(η@20 °C = 90 cP [23]) the oxida-
tion and reduction peaks are farther apart than in less viscous ILs like
(C4pyr)(Tf2N) (η@20 °C = 77 cP [24]), (C4mim)(Tf2N) (η@20 °C =
63.5 cP17), (C2mim)(TfO)(η@20 °C = 42.7 cP [25]) and (C4mim)(DCA)
(η@20 °C = 33.2) [26]. Furthermore, the dependency of the half-wave
potential with the scan rate is less in the latter ILs. Comparing the half-
wave potential of Eu2+/3+ at low scan rates (which allows for the best
approximation of a reversible chemical reaction) in Tf2N ionic liquids
with different cations, it becomes obvious that E1/2 is found for all ILs
roughly at −0.4 V and can be seen as independent from the IL cation.
If Eu(Tf2N)3 is added to a TfO IL E1/2 gets more negative and assumes
5. Luminescence spectroscopic studies
Thus, we have started investigating some ionic liquids by dissolving
Eu(Tf2N)3 and recording the luminescence spectra upon excitation into
the most intense 7F0
luminescence spectra and Table 1 for the compiled data).
→
7L6 transition (see Fig. 2 for representative
Table 2
Half-wave potentials, of the Eu3+/2+, reduction and oxidation potential in various ILs at
different scan rates. Data referenced against Fc/Fc+
.
For Eu3+ there exist two transitions that are quite sensitive towards
changes in the coordination sphere. These are the hypersensitive 5D0 →
7F2 transition (usually found around 610–625 nm) and the sensitive
5D0 → 7F4 transition (typically found around 680–710 nm). Amongst
those two the second one shows larger wavelength shifts depending on
the local surrounding of the emitting ion. A bathochromic shift of the
5D0 → 7F4 transition is observed in the presence of strongly electron do-
nating ligands [20]. Consequently a hypsochromic shift points to weak
IL
E
/V
Reduction potential/
V vs. Fc/Fc+
Oxidation potential/
V vs. Fc/Fc+
½
scan rate/
(Eu3+/2+
)
mV s−1
(P666 14)(Tf2N)
Eu(Tf2N)3
−0.62
200
100
50
−1.19
−0.97
−0.79
−0.65
−0.04
−0.07
−0.08
−0.13
−0.52
−0.44
−0.39
10
donor properties of the ligand. The 5D0
to a magnetic dipole interaction and is typically quite unaffected by
the chemical surrounding of Eu3+. Both the 5D0 7F0 and 5D0 7F3
→
7F1 transition corresponds
(C4pyr)(Tf2N)
Eu(Tf2N)3
−0.43
200
100
50
−0.91
−0.84
−0.80
−0.72
−0.05
−0.08
−0.10
−0.17
→
→
−0.46
transitions correspond to generally forbidden electric dipole transitions.
However, the 5D0 → 7F3 transition might gain some intensity through
spin–orbit contributions. The 5D0 → 7F0 transition cannot be observed
for Eu3+ at a site with Cn, Cnv and Cs symmetry. Mixing with other
configurations may lead to an appreciable intensity of this transition.
First we wanted to the check the cation influence and looked at var-
ious Tf2N-based ILs with different cations (Table 1). Indeed, the 5D0 →
7F4 transition seems to be quite unaffected by the respective IL cation,
regardless whether it is a 1-alkyl-3-methylimidzolium, a quarternary
ammonium or a phosphonium cation. However, a shift of the maximum
of the 5D0 → 7F4 transition is noticeable when the anion is exchanged
(Table 1 and Fig. 2) and, as expected, the DCA anion leads to a significant
−0.45
−0.44
10
(C4mim)(Tf2N)
Eu(Tf2N)3
−0.57
200
100
50
−1.08
−0.90
−0.83
−0.77
0.13
−0.05
−0.16
−0.10
−0.53
−0.46
−0.39
10
(C2mim)(ClO4)
Eu(Tf2N)3
−0.57
200
100
50
−0.98
−0.90
−0.86
−0.71
−0.15
−0.17
−0.21
−0.25
−0.53
−0.53
−0.48
10
bathochromic shift of the 5D0
which points to its higher Lewis basicity.
→
7F4 transition compared to Tf2N−
(C4mim)(TfO)
Eu(Tf2N)3
−0.51
200
100
50
−1.32
−1.21
−1.19
−1.09
0.31
0.18
0.16
0.05
Based on the maxima of the 5D0 → 7F4 transition the investigated IL
anions can be ranked according to increasing Lewis basicity as follows
Tf2N− b ClO−4 b MeSO3− b TfO− b DCA− b EtSO−4 .
−0.52
−0.52
−0.52
10
(C2mim)(TfO)
Eu(Tf2N)3
−0.63
6. Electrochemical studies
200
100
50
−1.11
−0.95
−0.97
−0.89
−0.14
−0.29
−0.37
−0.52
−0.62
Initial cyclic voltammetry revealed for Yb(Tf2N)3 that the more
Lewis basic and stronger coordinating the anion present in the ligand
sphere of Ln3+ was the more negative the redox potential became.
This suggests that it might be possible to establish a Lewis basicity
scale for IL anions by electrochemical means. To check whether such
an electrochemically determined redox potential of Eu2+/3+ in the re-
spective IL can be used as a measure of its anion's electron donor
power we dissolved Eu(Tf2N)3 in various ILs and determined the
redox half-wave potentials by means of cyclic voltammetry at different
scan rates (see Figs. 3–4 for representative data and Table 2 for a compi-
lation of data).
When looking at the cyclic voltammograms of Eu(Tf2N)3 in the var-
ious ionic liquids at different scan rates it becomes obvious that reduc-
tion and oxidation peaks are in all cases far apart, indicating a poor
reversibility of the redox reaction. At lower scan rates the separation
of both peaks becomes less but never gets so low that reversibility is
achieved. It seems that IL viscosity and, hence, also IL conductivity, are
−0.67
−0.71
10
(C4mpyr)(TfO)
Eu(TfO)3
−0.63
−0.70
−0.72
−0.73
200
100
50
−1.42
−1.36
−1.29
−1.14
0.17
−0.04
−0.15
−0.32
10
(C4mim)(EtSO4)
Eu(Tf2N)3
−0.96
200
100
50
−1.55
−1.45
−1.40
−1.34
−0.37
−0.52
−0.53
−0.66
−0.98
−0.97
10
−1.00
(C4mim)(DCA)
Eu(Tf2N)3
−1.08
200
100
50
−1.58
−1.54
−1.44
−1.37
−0.57
−0.63
−0.70
−0.79
−1.09
−1.07
10
−1.08

198
A. Babai et al. / Journal of Molecular Liquids 192 (2014) 191–198
(c) T. Welton, P. Wasserscheid, Ionic liquids in Synthesis, VCH-Wiley, Weinheim,
2002.;
(d) K.R. Seddon, Kinet. Katal. 37 (1996) 743;
values of about −0.7 V which is similar to that found for Eu(TfO)3 in
TfO-ILs. The Eu2+/3+ redox potential is observed to be the most nega-
tive in (C4mim)(DCA). From this it can be concluded that the redox po-
tentials are mainly affected by the IL anion. The larger the Lewis basicity
of the anion the more negative the Eu2+/Eu3 redox potential becomes.
The spectroscopic results can be basically confirmed.
C.L. Hussey, Adv. Molten Salt Chem. 5 (1983) 185;
(f) M.J. Earle, P.B. McCormac, K.R. Seddon, Green Chem. 1 (1) (1999) 23.
[2] K.R. Seddon, in: S. Boghosian, V. Dracopoulos, C.G. Kontoyannis, G.A. Voyiatzis
(Eds.), The International George Papatheodorou Symposium (Proceedings),
Institute of Chemical Engineering and High Temperature Chemical Processes,
Patras, 1999, p. 131.
[3] Ch. Reichardt, Pure Appl. Chem. 76 (2004) 1903.
[4] (a) Ch. Reichardt, Chem. Soc. Rev. 21 (1992) 147;
7. Conclusions
(b) Ch. Reichardt, Chem. Rev. 94 (1994) 2319;
(c) E. Buncel, S. Rajagopal, Acc. Chem. Res. 23 (1990) 226;
(d) Y. Marcus, Chem. Soc. Rev. 22 (1993) 409;
(e) S. Spange, A. Reuter, W. Linert, Langmuir 14 (1998) 3479;
(f) H. Jin, B. O'Hare, J. Dong, S. Arzhantsev, G.A. Baker, J.F. Wishart, A.J. Benesi, M.
Maroncelli, J. Phys. Chem. B 112 (2008) 81.
Irreversible one-electron transfer processes corresponding to the
Eu2+/3+ redox couple were found for solutions of Eu(Tf2N)3 in various
ionic liquids. The irreversibility of the electron transfer could be related
to the high viscosity of the investigated ILs. At lower scan rates the sep-
aration between oxidation and reduction process becomes less indicat-
ing the approximation of a quasi-reversible electrochemical reaction.
The redox potentials at low scan rates depend also strongly on the IL
anion. The cation's influence appears to be small. For the DCA ionic liq-
uid the most negative redox potential was found, followed by the TfO
and Tf2N ionic liquids. Thus, the redox potential can be related to the
Lewis basicity of the IL anion. From cyclovoltammetric investigation
the IL anions can be ordered with respect to rising Lewis basicity as fol-
lows: Tf2N− N TfO− N EtSO−4 N N DCA−.
[5] A.-V. Mudring, in: B. Kirchner (Ed.), Topics in Current Chemistry, 2010, p. 209.
[6] P. Müller, Pure Appl. Chem. 66 (1994) 1077.
[7] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, 2nd ed. Wiley-VCH,
Weinheim, 1988.
[8] J. Bartosik, A.-V. Mudring, Phys. Chem. Chem. Phys. 12 (2010) 4005.
[9] J.A. Duffy, Bonding, Energy levels & Bands in Inorganic Solids, New York, 1990.
[10] S. Pitula, A.-V. Mudring, Chem. Eur. J. 16 (2010) 3355.
[11] A. Babai, S. Pitula, A.-V. Mudring, Eur. J. Inorg. Chem. 31 (2010) 4933.
[12] (a) A. Babai, A.-V. Mudring, Z. Anorg. Allg. Chem. 632 (2006) 1956;
S.-F. Tang, J. Cybinska, A.-V. Mudring, Helv. Chim. Acta 92 (2009) 2375;
S.-F. Tang, A. Babai, A.-V. Mudring, Angew. Chem. Int. Ed. 47 (2008) 7631;
A.-V. Mudring, S.-F. Tang, Eur. J. Inorg. Chem. 18 (2010) 2569;
A. Getsis, A.-V. Mudring, Eur. J. Inorg. Chem. 14 (2010) 2172;
S.-F. Tang, A.-V. Mudring, Growth Des. 11 (2011) 1437;
Luminescence spectroscopy confirms these results. The used lumi-
nescent Eu3+ ion doped in several ILs showed to be a very sensitive
probe to its coordination sphere and hence to its surrounding ligands
with different donation properties.
(g) P. Nockemann, B. Thijs, N. Postelmans, K. Van Hecke, L. Van Meervelt, K.
Binnemans, J. Am. Chem. Soc. 128 (2006) 13658;
(h) M. Pellens, B. Thijs, K. Van Hecke, L. Van Meervelt, K. Binnemans, P. Nockemann,
Acta Crystallogr. E 64 (2008) m945;
In summary, the determination of the Lewis basicity of IL anions by
investigation of the optical and electrochemical properties of Eu3+ ap-
pears to be promising and a proof of concept for this method could be
given. At this point a larger set of ionic liquids with different cation
and anion combinations needs to be investigated to be able to compare
the methods with other established techniques for the determination of
ionic liquids' acidity and basicity.
(i) P. Nockemann, B. Thijs, K. Lunstroot, T.N. Parac-Vogt, C. Goerller-Walrand, K.
Binnemans, K. Van Hecke, L. Van Meervelt, S. Nikitenko, J. Daniels, C. Hennig, R.
Van Deun, Chem. Eur. J. 15 (2009) 1449.
[13] (a) A. Babai, A.-V. Mudring, Chem. Mater. 17 (2005) 6230;
(b) A. Babai, S. Arenz, R. Giernoth, K. Driesen, P. Nockemann, A.-V. Mudring, J. Alloys
Compd. 418 (2005) 204.
[14] G.H. Dieke, Spectra and Energy Levels of Rare Earth Ions in Crystals, Interscience
Publishers, New York, 1968.;
W.T. Carnall, H.M. Crosswhite, H. Crosswhite, Special Report, Chemistry Division, Ar-
gonne National Laboratory, Argonne, IL, 1977.
[15] S.H. Strauss, Chem. Rev. 93 (1993) 927.
Acknowledgments
[16] 4 A. Babai, Dissertation, Universität zu Köln, 2006, and ref. 11.
[17] A. Chaumont, G. Wipff, J. Phys. Chem. B 112 (2008) 12014.
[18] A. Tomasawa, PhD Thesis, Queen's University, Belfast, 2010; S. Pitula, Dissertation,
Universität zu Köln, 2010.
[19] A. Babai, A.-V. Mudring, Angew. Chem. 44 (2005) 5485.
[20] S.P. Sinha, J. Inorg. Nucl. Chem. 28 (1966) 189.
[21] R. Hagiwara, T. Hirashige, T. Tsuda, Y. Tio, Solid State Sci. 4 (2002) 23.
[22] H. Shirota, A.M. Funstou, J.F. Wishart, E.W. Castner, J. Chem. Phys. 122 (2005)
184512.
This work has been supported by the Deutsche Forschungsgemeinschaft
within the priority program SPP 1191 “Ionic Liquids”, the DFG Cluster of
Excellence RESOLV and the Critical Materials Institute, an Energy Inno-
vation Hub funded by the U.S. Department of Energy, Office of Energy
Efficiency and Renewable Energy, Advanced Manufacturing Office.
[23] H. Olivier-Bourbigou, L. Magna, J. Mol. Catal. A Chem. 182 (2002) 419.
[24] R. Bini, M. Malvadi, W.R. Pitner, C. Chiappe, J. Phys. Org. Chem. 21 (2008) 622.
[25] C. Nanjundiah, F. McDevitt, V.R. Koch, J. Electrochem. Soc. 144 (1997) 3392.
[26] L. Galán Sánchez, J.R. Expel, F. Onink, G.W. Meindersma, A.B. de Haan, J. Chem. Eng.
Data 54 (2009) 2803.
References
[1] (a) T. Welton, Chem. Rev. 99 (1999) 2071;
(b) P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. Engl. 39 (2000) 3772;