Polarity of tetraalkylammonium-based ionic liquids and related low temperature molten salts

Article abstract of DOI:10.1039/c7nj00443e

Solvatochromic measurements and ¹H NMR spectroscopy have been used to investigate the overall polarity of tetraalkylammonium (R₃N⁺-CH₃ with R = n-C₄H₉; n-C₈H₁₇) based ionic liquids (IL) and low temperature molten salts (LTMS), respectively, as function of the counter anion. Solvent acidity (SA), solvent basicity (SB), polarizability (SP) and dipolarity (SdP) parameters have been determined according to the Catalán scale. [Fe^II(1,10-phenanthroline)₂(CN)₂] (Fe), 3-(4-amino-3-methylphenyl)-7-phenylbenzol[1,2-b:4,5-b′]difuran-2,6-dione (ABF), 3-(4-N,N-dimethylaminophenyl)-7-phenylbenzol[1,2-b:4,5-b′]-difuran-2,6-dione (DMe-ABF), 4-tert-butyl-2-(dicyanomethylene)-5-[4-(dimethylamino)benzylidene]-Δ³-thiazoline (Th), and 2-[4-(N,N-dimethylamino)benzylidene]malononitrile (BMN) have been used as solvatochromic polarity indicators. Significant correlations of the SB parameter of anion X^- as a function of the chemical shift of the H-atoms of R₃N⁺-CH₃X^- are presented and discussed. Thus, IL solvent basicity (SB) of the anion determines the interaction strength towards those CH-bonds in the direct vicinity of the ammonium nitrogen atom.

Full text of DOI:10.1039/c7nj00443e

NJC
View Article Online
View Journal
PAPER
Polarity of tetraalkylammonium-based ionic
liquids and related low temperature molten salts†
Cite this:DOI: 10.1039/c7nj00443e
Gabi Thielemann
and Stefan Spange*
Solvatochromic measurements and 1H NMR spectroscopy have been used to investigate the overall
polarity of tetraalkylammonium (R3N+–CH3 with R = n-C4H9; n-C8H17) based ionic liquids (IL) and low
temperature molten salts (LTMS), respectively, as function of the counter anion. Solvent acidity (SA),
solvent basicity (SB), polarizability (SP) and dipolarity (SdP) parameters have been determined according
to the Catalan scale. [FeII(1,10-phenanthroline)2(CN)2] (Fe), 3-(4-amino-3-methylphenyl)-7-phenylbenzol[1,2-
´
b:4,5-b0]difuran-2,6-dione (ABF), 3-(4-N,N-dimethylaminophenyl)-7-phenylbenzol[1,2-b:4,5-b0]-difuran-2,6-
dione (DMe-ABF), 4-tert-butyl-2-(dicyanomethylene)-5-[4-(dimethylamino)benzylidene]-D3-thiazoline (Th),
and 2-[4-(N,N-dimethylamino)benzylidene]malononitrile (BMN) have been used as solvatochromic polarity
indicators. Significant correlations of the SB parameter of anion Xꢀ as a function of the chemical shift of the
H-atoms of R3N+–CH3Xꢀ are presented and discussed. Thus, IL solvent basicity (SB) of the anion determines
the interaction strength towards those CH-bonds in the direct vicinity of the ammonium nitrogen atom.
Received 7th February 2017,
Accepted 5th July 2017
DOI: 10.1039/c7nj00443e
rsc.li/njc
Introduction
The impact of ionic liquids (ILs) on chemical reactivity is
fascinating due to the perspectives in synthesis and material
research.1,2 There are several concepts that semi-quantitatively
explain the effect of IL structures on chemical reaction kinetics,
thermodynamics and product formation.3,4 At the moment,
there are two major concepts that could explain these effects.
The results were either explained in terms of the cumulative
effects of the IL or by specific influences of the anion and
cation, respectively.5,6 For ILs, it has been shown that the anion
structure especially determines their impact on chemical reactivity
rather than the cation, which seems of minor importance.6 However,
the appropriate cation structure is momentous for the occurrence of
the liquid state at low temperatures. Furthermore, functional groups
can be integrated in the periphery of the cation, which can change
the influence of the IL on the reaction dramatically.3
A series of papers has shown that the anion properties of ILs
can be described well by their hydrogen bond accepting (HBA)
property (b) or basicity (SB). Both items, b and SB, can be
Scheme 1 Cations and anions selected for this study.
classified using the LFE (Linear Free Energy) relationships of
7–9
´
Kamlet–Taft and Catalan, respectively (see below).
The objective of this study is to determine the empirical solvent This type of organic salt can be synthesized more easily compared
polarities of tetraalkylammonium based ionic liquids (Scheme 1). to the established imidazolium based ionic liquids. Tri(n-butyl)-
methylammonium (N4441
) and tri(n-octyl)methylammonium
Department of Polymer Chemistry, Institute of Chemistry, Chemnitz University of
Technology, 09107 Chemnitz, Germany. E-mail: Stefan.spange@chemie.tu-chemnitz.de;
Fax: +49 (0)371-531-21239; Tel: +49 (0)371-531-21230
(N8881) are selected as cations because of their wide possible
applications. These cation-based salts are, for example, under
investigation as an electrolyte for batteries,10–12 in fuel cells,13,14
and for pharmaceutical and biomedical applications.15–17 Due
to the differences in melting points of these salts, they can be
† Electronic supplementary information (ESI) available: Information to UV/Vis
measurements, Catalan and Kamlet–Taft parameter set, 1H NMR spectroscopic
´
studies of all compounds in solution. See DOI: 10.1039/c7nj00443e
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017
New J. Chem.

View Article Online
Paper
NJC
classified into two categories of substances – melting point
o100 1C is an ionic liquid (IL) or 4100 1C is a low temperature
molten salt (LTMS). Motivation of this study is to investigate
whether related correlations between solvatochromic measurements
and 1H NMR spectroscopic results can be discovered as already
found for imidazolium based ionic liquids.18
As shown in several review articles, empirical solvent para-
meters for ILs have been well established and utilized for the
classification of various ILs.4,5,18,19 The investigation of ionic
liquid (IL) or low temperature molten salts (LTMS) by the solvato-
chromic method using Linear Solvation Energy Relationships
(LSER) according to Kamlet–Taft has been well accepted in the
literature because various physicochemical properties of the
liquid can be simultaneously determined and classified.9,20 In a
series of papers, the Kamlet–Taft approach was successfully
applied to measure the following three parameters of ILs: hydrogen
bond acidity (a), hydrogen bond basicity (b), and dipolarity/
polarizability (p*).5,18,19,21–23
The use of appropriate solvatochromic probes for the accurate
determination of the Kamlet–Taft parameter of IL is indispensable.
Unfortunately, in the literature the longstanding solvatochromic
method, originally introduced by Kamlet–Taft, is uncritically used
sometimes. For instance, 4-nitroanisole cannot be used as probe for
p* when hydrogen-bond donating ILs are investigated because this
probe is also sensitive to a. This problem has been treated by us
recently19 and already accentuated by Omar A. El Seoud24 in a
highly important but undervalued work on the interpretation of
solvatochromic data.
Scheme 2 Solvatochromic dyes (Fe, ABF, DMeABF, Th and BMN) used in
this study for determination of the Catalan parameters SA, SB, SP and SdP
´
of ILs or LTMS.19
and 3-(4-N,N-dimethylaminophenyl)-7-phenylbenzol[1,2-b:4,5-b0]-
difuran-2,6-dione (DMe-ABF), and the SP and SdP parameters by
using the homomorphic pair of 4-tert-butyl-2-(dicyanomethylene)-
5-[4-(dimethylamino)benzylidene]-D3-thiazoline (Th) and 2-[4-(N,N-
dimethylamino)benzylidene]malononitrile (BMN) (Scheme 2).
The LSER equations for calculating the individual solvent para-
meters are explained in the experimental section of this work.
´
The reliability of the determined Kamlet–Taft or Catalan
parameters can be verified by correlation with independent
physicochemical measures.25,26 Especially, the 1H NMR methodology
has been successfully applied to investigate the impact of the anion
structure on the cation anion interaction of imidazolium ionic
liquids.25,26 Several authors have shown that the specific effect of
the anion on the individual hydrogen atoms of the imidazolium
ion can be measured. The extent of the chemical shift of the
H-atom in the 2-position as well as of those in the 3 and 4
positions of imidazolium increases with increasing hydrogen
bond basicity b or SB of the anion. For this conclusion, it makes
no difference whether the empirical polarity parameters of IL
are measured as pure liquid via UV/Vis spectroscopy or in a
solution of dichloromethane (DCM) via 1H NMR spectroscopy.26
Moreover, knowledge of the SP of IL seems of importance
because the anion polarizability additionally contributes to the
overall anion effect on IL properties, as shown recently.19
In this work, the solvent parameters of the IL and LTMS were
determined as pure salts by a solvatochromic methodology.
Moreover, the chemical shift of the salts was investigated,
depending on the variation of the anion and the concentration
in CD2Cl2, by 1H NMR spectroscopy. The choice of the measuring
methods is aligned to the special requirements of the used salts,
which is a subject of the discussion.
As an improvement compared to the Kamlet–Taft scale, the
´
Catalan LSER enables the independent determination of the
dipolarity and polarizability parameter of a solvent of each
other. Accordingly, the solute–solvent interactions are described
by four solvent parameters: solvent acidity (SA), solvent basicity
(SB), solvent polarizability (SP), and solvent dipolarity (SdP).
´
Eqn (1) shows the LSER approach according to Catalan, which
takes into account these four parameters.
(XYZ) = (XYZ)0 + aSA + bSB + dSP + eSdP
(1)
The result (XYZ) denotes a solvent-dependent physiochemical
property in a given solvent and (XYZ)0 is the statistical quantity
corresponding to the value of the property in the gas phase as
reference. a, b, d, and e are regression coefficients describing
the sensitivity of the solvent-affected physicochemical property
XYZ in relation to the particular individual solvent property.
Of course, there exist reasonable correlations between a and
SA or b and SB, as shown by us recently, because of the
apparent relation of the Kamlet–Taft and Catalan concept.19
For describing solvent acidity or basicity, it makes no crucial
´
difference whether Kamlet–Taft or Catalan parameters are
Results and discussion
used. It was demonstrated that our developed novel solvato-
chromic methodology is suitable to determine the SA, SB, SP The solvatochromic dyes were dissolved in particular ionic
and SdP of ILs.19
liquids or molten salts and the UV/Vis spectra were recorded
Accordingly, the SA parameter was determined solely by immediately (see the Experimental part). The results of UV/Vis
using [FeII(1,10-phenanthroline)2(CN)2] (Fe), the SB parameter measurements are compiled in the ESI† (Table S1).
by using the pair of structurally comparable dyes 3-(4-amino-3-
The UV/Vis measurements were performed in the pure salts
methylphenyl)-7-phenylbenzol[1,2-b:4,5-b0]difuran-2,6-dione (ABF) and not in solutions of DCM, because DCM is a stronger
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017

View Article Online
NJC
Paper
acceptor (SA parameter 0.0427) than [N4441]Cl. Thereby, the
interaction of the solvatochromic dye Fe with DCM is much
stronger than with [N4441]Cl. In contrast, the SB parameter of
DCM (0.17827) is significantly smaller than those of the salts
and, as expected, highly dependent on the concentration (ESI,†
Fig. S2). For this reason, UV/Vis measurements have been
performed only in the pure IL or LTMS.
´
In this part, the Catalan parameters are used exclusively to
prevent confusion. However, the Kamlet–Taft solvent para-
meters can be likewise calculated using the measured UV/Vis
absorption maxima of the used solvatochromic probes Fe, ABF and
Th (Scheme 2). The Kamlet–Taft solvent parameters a, b and p* were
also computed and are provided in the ESI† (Table S2).
Fig. 1 (a) Correlation of SB with SA of [N4441]X and [N8881]X; (b) comparison
of SP (filled and half-filled symbols) and SdP (empty symbols) as a function of
the SB of tri-n-alkylmethylammonium cations with various anions.
Due to the obvious chemical reaction of Fe with some LTMS
at higher temperatures, it was not possible to directly deter-
mine SA values for all LTMS. It seems that the Fe(phen)2(CN)2
complex reacts with the ammonium salt and proton elimination
takes place at a higher temperature because the cyano-ligands
show a relatively high basicity.28,29 The SA values for the LTMS
that cause Fe decomposition were estimated using eqn (2) and (3),
presented later in the text.
According to eqn (1), the significant UV/Vis data have been
used to calculate the empirical solvent parameters. The established
LSER eqn (12)–(15) used are explained in the experimental section.
The results are compiled in Table 1.
regression, eqn (2) and (3) were calculated. These equations
allow us to estimate the missing SA values for such LTMS, and
induce the Fe decomposition. These values are also given in
Table 1, marked with an asterisk.
Using linear regression (Fig. 1(a)), eqn (2) was calculated for
[N4441]X and eqn (3) for [N8881]X.
SA = ꢀ0.31SB + 0.32
where n = 4, r2 = 0.96, sd = 0.071.
SA = ꢀ0.14SB + 0.19
(2)
(3)
The data set in Table 1 is ordered with decreasing SB values. where n = 4, r2 = 0.93, sd = 0.091.
It is remarkable that the determined SA values are rather
and the lowest for the hexafluorophosphate anion (bottom of low, which is in line with the intrinsically weaker interaction
the list). strength of tetraalkylammonium ions compared to imidazolium
Fig. 1(a) shows a correlation between SB and SA for the based cations.18 Moreover, the more sterically shielded [N8881
The largest SB is found for the chloride anion (top in the list)
]
solvents that allows a direct measurement of SA. Using linear cations show larger SB values in comparison to [N4441] salts due
to the weakened interaction between cation and anion. A
mutual dependence between the SA and SB parameters was
found (Fig. 1a), eqn (2) for [N4441]X and eqn (3) for [N8881]X.
These results demonstrate the significant connection of the
acidity and basicity, which is similar to that observed for
imidazolium based ionic liquids.8,9,18 This is a crucial result
because it shows that even bulky tetraalkylammonium ions
evidently interact with the anion in spite of their low hydrogen
bond acidity.
The polarizabilities of the IL do not vary significantly as a
function of anion structure [compare Fig. 1b]. However, large
anions, such as CF3SO3 or Ntf2, clearly show the trend for
increasing the SP parameter. SdP values increase slightly with
SB values because the increase of the acid–base interaction has
an impact on increase of SdP. In contrast to the imidazolium
based IL, no correlation of SP or SdP as a function of SA ꢁ SB
was found [ESI,† Fig. S3b].18 This proves the complex influence
of the cation structure.
By reason of the strong distribution of the SP and SdP
values, it is assumed that the variation of anion does not
linearly influence these parameters (Fig. 1a and ESI,† Fig. S3).
The low hydrogen bond acidity of the tetraalkylammonium
salts is not surprising because bulky tetraalkylammonium
cations are known for their relatively weak specific interaction
strength. The question arises, which molecular aspects are
Table 1 Melting points (mp) and empirical Catalan solvent parameters SA,
SB, SP, and SdP of tri-n-alkylmethylammonium cations with different
´
anions according to Scheme 1
Substance
Cation
[N4441]
Anion
mp [1C]
SA
SB
SP
SdP
Cl
CF3CO2
Br
113b
orta
125b
123b
orta
88a
0.009
0.070
0.998
0.854
0.786
0.658
0.631
0.489
0.382
0.304
0.298
0.661
0.747
0.722
0.645
0.825
0.840
0.668
0.754
0.710
1.255
1.037
0.997
1.193
1.043
0.905
1.031
0.968
0.977
0.072*
0.112*
0.096
0.164*
0.198*
0.227
NO3
N(CN)2
CF3SO3
BF4
Ntf2
PF6
158b
orta
132b
0.224*
[N8881
]
Cl
CF3CO2
Br
30a
0.033
0.054
1.070
0.913
0.905
0.756
0.673
0.588
0.514
0.403
0.386
0.638
0.700
0.661
0.638
0.769
0.730
0.700
0.806
0.710
1.106
1.002
1.101
1.106
1.032
0.927
1.002
0.888
0.977
orta
57a
0.057*
0.078*
0.105
0.102*
0.113*
0.118
NO3
45a
N(CN)2
CF3SO3
BF4
Ntf2
PF6
orta
48a
54a
orta
67a
0.132*
*Marked values are calculated using eqn (2) for [N4441]X and 3 for
[N8881]X. These salts react with Fe and discoloration takes place at T 4
b
60 1C.a Ionic liquids. Low temperature molten salts.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017
New J. Chem.

View Article Online
Paper
NJC
responsible for the mutual dependency of acidity and anion
basicity. Therefore, apart from the solvatochromic studies,
1H NMR spectra of all the salts were measured as a function
of concentration.
At room temperature, the pure tetraalkylammonium based
ionic liquids are mostly solids or highly viscous, in particular.
1
Therefore, a H NMR measurement in its pure form according
to literature29,30 was not directly possible.
The 1H NMR spectroscopic studies were performed in
CD2Cl2 as the solvent using various concentrations of the salt.
According to previous studies,18,19 CD2Cl2 was chosen as the
NMR solvent because of its low relative permittivity of 9.8.
Hence, ion pair formation should dominate in concentrated
solutions and dissociation into solvated ions should only take
place to a low extent at low concentrations.30–32
Fig. 3 1H NMR spectra of [N4441]X with various anions in CD2Cl2 at the
concentration of 1.8 M.
two different structural groups have to be considered from the
NMR measurements.
Exemplary, the influence of [N4441]Cl-concentration on the
chemical shift in the range of 0.2 M to 2 M in 0.2 M steps is
demonstrated. The NMR spectra series are shown in Fig. 2.
An upfield shift for all the hydrogen atoms was observed
with increasing concentration of salt. This is a clear indication
of anion coordination due to the increased electron density of
the hydrogen atoms in the vicinity of the positively charged
nitrogen atom. Thus, the other salts were measured in 1.8 M,
1.0 M, and 0.2 M solutions, and the hydrogen atoms of the
cation are monitored at the highest concentration of the salt.
The 1H NMR spectra of all the salts (Fig. 3) show an upfield
shift of the hydrogen atoms neighboring to the nitrogen atom
(CH3–N and N–CH2–) with a simultaneously downfield shift of
the alkyl chain at a high concentration. The downfield shift of
the alkyl chain is alleviated with decreasing concentration (see
ESI,† Fig. S4 and S5). Therefore, the variation of concentration
influences the alkyl chain periphery, but not the moieties
directly linked to the positively charged nitrogen atom.
In the following part, the 1H NMR data and the results from
UV/Vis measurements were combined in such a way to locate
correlations between the chemical shift d so that the SB or SA
parameters can be investigated. In general, for the discussion
On the one hand, the CH-moieties directly bonded to the
nitrogen atom of the cation show an upfield shift with increasing
SB values and a downfield shift with increasing SA values for all
three concentrations (Fig. 3 and ESI,† Fig. S6, S7, eqn (4)–(7)).
Using linear regression (Fig. 4a), eqn (4) was calculated for
[N4441]X and eqn (5) for [N8881]X. Only the SA values were used
for correlations, which were directly measurable.
SA = ꢀ0.69d + 2.24
where n = 4, r = 0.93, sd = 0.041.
SA = ꢀ0.31d + 1.05
where n = 4, r = 0.99, sd = 0.014.
(4)
(5)
Using linear regression (Fig. 4b), eqn (6) was calculated for
[N4441]X and eqn (7) for [N8881]X.
SB = 2.50d ꢀ 7.05
where n = 8, r = 0.97, sd = 0.066.
SB = 2.38d + 6.61
where n = 8, r = 0.97, sd = 0.061.
(6)
(7)
On the other hand, for the groups of the alkyl chain in the
outer periphery the inverse correlation is observed for a high
concentration (1.8 M, Fig. 5 and eqn (8)–(11)) and no sensible
correlation at a low concentration (0.2 M, ESI,† Fig. S8 and S9).
Fig. 4 SA as a function of the chemical shift of the hydrogen atoms
(CH3–N) (a) and SB as a function of the chemical shift of the hydrogen
atoms (CH3–N) (b) measured in CD2Cl2 0.2 M.
Fig. 2 1H NMR spectra of [N4441]Cl in CD2Cl2 with increasing concentra-
tions from 0.2 M to 2.0 M.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017

View Article Online
NJC
Paper
Scheme 3 Schematic illustration of possible arrangement of the ILs/LTMS
at low and high concentrations.
ꢀ
Fig. 5 SA (a) and SB (b) as a function of the chemical shift of the hydrogen Ntf2 for the tetraalkylammonium salts is in line with previously
reported b values for related compounds.19 Altogether, these
atoms in the outer periphery (CH3–CH2) measured in CD2Cl2 1.8 M.
determined correlations are of very good quality and allow
significant conclusions. However, the influence of the anion
Furthermore, there is no significant change of the chemical
and the concentration on the chemical shift is much more
shift of CH2-groups along the alkyl chain, which is not directly
complex compared to imidazolium based ILs.18
bonded to the nitrogen observed at a low concentration. This
result shows that the strength of the ion pair aggregation
depends on the concentration of the salt.
In general, no correlation between SP or SdP and d of the
salts was found. This was expected due to the weak dependence
of SP or SdP as function of the anion.
Using linear regression [Fig. 5a], eqn (8) was calculated for
[N4441]X and eqn (9) for [N8881]X.
Conclusions
SA = 0.78d ꢀ 0.57
where n = 4, r = 0.90, sd = 0.049.
SA = 0.33d ꢀ 0.18
where n = 4, r = 0.86, sd = 0.136.
(8)
(9)
Empirical solvent parameters for tetraalkylammonium based
ILs and low temperature molten salts according to the Catalan
scale or the Kamlet–Taft scale can be readily determined by the
solvatochromic methodology. The significant correlation of the
SB parameter with the independently determined H chemical
´
1
Using linear regression [Fig. 5b], eqn (10) was calculated for
[N4441]X and eqn (11) for [N8881]X.
shift of the –CH3 and –CH2– groups of tetraalkylammonium
salts as a function of the anion is a promising result. Thus, the
linear correlations of d = f(SB) evidently support the reasonability
of the determined SB parameter. Thus, the NMR methodology
allows the direct comparison of IL and MLTS compounds regard-
less of which melting point, because the effect of the anion works
independently of the solvent CD2Cl2. It enables us to localize the
interaction sites of anions with complex cations. Thus, this
methodology can be extended in future work to other cation types,
which may contain functional and polar groups in their periphery.
As a consequence, ILs and LTMS can be classified within one
concept, which makes this methodology useful for a wide
application to determine polarity parameters of cations and anions.
SB = ꢀ2.79d + 3.04
where n = 9, r = 0.92, sd = 0.103.
SB = ꢀ2.56d + 2.73
where n = 9, r = 0.90, sd = 0.111.
An important issue is that the strength of the interaction
depends not only on the basicity of the anion, but also on the
nature of the counter-cation, its concentration dependent
aggregation behavior as well as the solvent CD2Cl2. Therefore,
the explanation of the evident correlations in terms of molecular
aspects is not trivial. Altogether, these results indicate that a
strong electrostatic interaction exists between the anion and
cation independently of the concentration. However, this inter-
action does not influence the chemical shift of the alkyl chain at
a low concentration since no change in chemical shift can be
(10)
(11)
Experimental section
Materials and instrumentation
observed in this case. There seems to be another effect at a high All the chemicals used were purchased in the highest available
concentration (1.8 M), because a significant change in the purity from commercial sources and used as received, unless
chemical shift of the alkyl chain is monitored. We conclude stated otherwise. The established dyes Fe,28 ABF,33 and Th34,35
that at a high concentration the positively charged ammonium as well as DMeABF36 and BMN37 were synthesized and purified
centers repulse each other up to the maximum distance. This as described in the literature. Dichloromethane (DCM) was
process is still assisted by the counter anion and likely leads to dried over CaH2 and freshly distilled before use. All salts were
an ion pair system with stacked alkyl chains (Scheme 3).
dissolved in dry DCM, molecular sieve 4 Å was added and the
Despite this difficulty to explain all observed effects of mixture was stirred for 24 h. The solvent was removed by
concentration on NMR chemical shifts, significant correlations distillation and was subsequently evaporated under a reduced
are found between d and SB. The results from Fig. 4 as well as pressure of 2 mbar at 60 1C for 8 h.
eqn (6), (7), (10) and (11) clearly show the impact of the anion
All NMR spectra were recorded at 25 1C on a Bruker Avance
basicity on the electron density distribution of the corresponding DRX-250 spectrometer (1H: 250.13 MHz) using CD2Cl2 (dried
tetraalkylammonium cation. The order of SB values of N(CN)2ꢀ 4 over molecular sieve 4 Å) as solvent. The chemical shift was
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017
New J. Chem.

View Article Online
Paper
NJC
referenced internally for the 1H NMR using the solvent residual
peak of CD2Cl2 (d = 5.31 ppm). The UV/Vis absorption spectra
were measured on an MCS 400 diode-array spectrometer from
Carl Zeiss Jena with a resolution of 1 nm in 2 mm quartz cells.
Spectral analyses were performed with Aspect Plus (version
1.76, Carl Zeiss Jena). All spectroscopic data are available in
Tables S1 and S2 in the ESI.†
Acknowledgements
Financial support by the Deutsche Forschungsgemeinschaft
(DFG SP 392/45-1) is gratefully acknowledged. The authors
thank J. Kronawitt for her experimental work and Dr A. Seifert
for the fruitful discussions.
Notes and references
General synthesis of tri-n-alkylmethylammonium ILs/LTMSs
1 J. P. Hallett and T. Welton, Chem. Rev., 2011, 111, 3508–3576.
2 J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D. Willauer,
G. A. Broker and R. D. Rogers, Green Chem., 2001, 3, 156–164.
3 P. Wasserscheid, M. Sesing and W. Korth, Green Chem.,
2002, 4, 134–138.
4 C. Reichardt, Green Chem., 2005, 7, 339–351.
5 A. J. Carmichael and K. R. Seddon, J. Phys. Org. Chem., 2000,
13, 591–595.
The ILs/LTMS were synthesized from the respective chloride
salt according to established literature procedures.38–42 In
general, tri-n-butylmethylammonium chloride (5 g) was dis-
solved in water (150 mL) and a solution of equimolar silver
salt (AgNO3, AgCF3CO2, AgCF3SO3, AgN(CN)2 [freshly synthe-
sized]); sodium salt (NaBF4) or lithium salt (LiNtf2) was added
dropwise. For the synthesis of tri-n-octylmethylammonium
ILs/LTMS tri-n-octylmethylammonium chloride was dissolved
in a H2O/DCM solution (50 : 50% v/v). When using silver salts,
the formed solid was filtered and the solvent removed by rotary
distillation. Insoluble products that were generated by using
sodium salts were filtered off and washed several times with the
reaction solvent. All the products were dried as described above
and characterized via NMR spectroscopy, IR spectroscopy,
quantitative elemental analysis and DSC measurements.
¨
6 H. Weingartner, Angew. Chem., Int. Ed., 2008, 47, 654–670.
´
´
7 J. Catalan, J. Palomar, C. Dıaz and J. L. G. de Paz, J. Phys.
Chem. A, 1997, 101, 5183–5189.
´
´
´
´
8 J. Catalan, C. Dıaz, V. Lopez, P. Perez, J.-L. G. De Paz and
´
J. G. Rodrıguez, Liebigs Ann., 1996, 11, 1785–1794.
9 R. Lungwitz, M. Friedrich, W. Linert and S. Spange, New
J. Chem., 2008, 32, 1493–1499.
10 V. Selvamani, V. Suryanarayanan, D. Velayutham and
S. Gopukumar, J. Solid State Electrochem., 2016, 20, 2283–2293.
11 N. Terasawa and K. Asaka, Sens. Actuators B Chem., 2014,
193, 851–856.
Methodology for determination of the empirical polarity
parameters (theoretical background see ref. 19)
´
The innovation of the Catalan scale in comparison to the
12 N. L. Nguyen and D. Rochefort, Electrochim. Acta, 2014, 147,
96–103.
Kamlet–Taft scale is the differentiation between solvent polarity
and polarizability. Therefore, it is necessary to use dyes which
are unaffected by hydrogen bonds and respond in different
degrees to polarity and polarizability. The chromophore pair
Th and BMN have been developed by us as appropriate dyes
for this.
´
13 U. Domanska and M. Wlazło, Fuel, 2014, 134, 114–125.
14 B. Qiu, B. Lin, L. Qiu and F. Yan, J. Mater. Chem., 2011, 22,
1040–1045.
15 V. Kumar, V. S. Parmar and S. V. Malhotra, Biochimie, 2010,
92, 1260–1265.
´
The Catalan acidity parameter (SA) can be calculated by the
´
16 W. L. Hough, M. Smiglak, H. Rodrıguez, R. P. Swatloski,
sole use of eqn (12). For this purpose, the UV/Vis absorption
S. K. Spear, D. T. Daly, J. Pernak, J. E. Grisel, R. D. Carliss,
M. D. Soutullo, J. H. Davis and R. D. Rogers, New J. Chem.,
2007, 31, 1429–1436.
˜
˜
maxima vmax of the Fe indicator are measured in the ILs. Dv
values are used in 10ꢀ3 cmꢀ1
v Fe solv: ꢁ 10ꢀ3 cmꢀ1 ꢀ 16:255
ꢀ3:62047
Þ
~ð Þð
17 B. S. Sekhon, Ars Pharm., 2013, 54, 37–44.
18 S. Spange, R. Lungwitz and A. Schade, J. Mol. Liq., 2014, 192,
137–143.
SA ¼
(12)
The SB parameter can be determined by simultaneous use of
19 A. Schade, N. Behme and S. Spange, Chem. – Eur. J., 2014,
20, 2232–2243.
20 L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton,
Phys. Chem. Chem. Phys., 2001, 3, 5192–5200.
˜
ABF and DMeABF. vmax (DMeABF) is not affected by the solvent
basicity. Therefore, the SB paramꢀe3ter can be calculated by
eqn (13), where Dv values are in 10 cmꢀ1
.
˜
~
~
DvðABFÞ ꢀ DvðDMeABFÞðsolv:Þ ꢀ 2:84647
SB ¼
(13)
¨
21 K. Bica, M. Deetlefs, C. Schroder and K. R. Seddon, Phys.
ꢀ3:62047
Chem. Chem. Phys., 2013, 15, 2703–2711.
22 P. G. Jessop, D. A. Jessop, D. Fu and L. Phan, Green Chem.,
2012, 14, 1245–1259.
23 C. Chiappe, C. S. Pomelli and S. Rajamani, J. Phys. Chem. B,
2011, 115, 9653–9661.
The solvent polarity SdP and polarizability SP can be separately
calculated from eqn (14) and (15) using the UV/Vis absorption
maxima of Th and BMN measured in the particular IL.
˜
˜
SP = 0.387vmax(Th)(solv.) ꢀ 0.605vmax(BMN)(solv.) + 8.500
24 L. P. Novaki and O. A. El Seoud, Ber. Bunsenges. Phys. Chem.,
1996, 100, 648–655.
(14)
˜
˜
¨
SdP = 0.556vmax(BMN)(solv.) ꢀ 0.980vmax(BMN)(solv.) + 3.791
25 T. Cremer, C. Kolbeck, K. R. J. Lovelock, N. Paape, R. Wolfel,
(15)
P. S. Schulz, P. Wasserscheid, H. Weber, J. Thar, B. Kirchner,
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017

View Article Online
NJC
Paper
F. Maier and H.-P. Steinru¨ck, Chem. – Eur. J., 2010, 16, 35 S. Spange, R. Sens, Y. Zimmermann, A. Seifert, I. Roth,
9018–9033. S. Anders and K. Hofmann, New J. Chem., 2003, 27, 520–524.
26 K. Noack, P. S. Schulz, N. Paape, J. Kiefer, P. Wasserscheid 36 G. Hallas and C. Yoon, Dyes Pigm., 2001, 48, 107–119.
¨
¨
and A. Leipertz, Phys. Chem. Chem. Phys., 2010, 12, 37 M. Mantri, O. de Graaf, J. van Veldhoven, A. Goblyos,
14153–14161.
J. K. von Frijtag Drabbe Ku¨nzel, T. Mulder-Krieger, R. Link,
H. de Vries, M. W. Beukers, J. Brussee and A. P. IJzerman,
J. Med. Chem., 2008, 51, 4449–4455.
´
27 J. Catalan, J. Phys. Chem. B, 2009, 113, 5951–5960.
28 A. A. Schilt, J. Am. Chem. Soc., 1960, 82, 3000–3005.
29 J. Burgess, Spectrochim. Acta, Part A, 1970, 26, 1957–1962.
30 A. G. Avent, P. A. Chaloner, M. P. Day, K. R. Seddon and
38 R. K. Blundell and P. Licence, Phys. Chem. Chem. Phys.,
2014, 16, 15278–15288.
`
T. Welton, J. Chem. Soc., Dalton Trans., 1994, 23, 39 M. Fabris, V. Lucchini, M. Noe, A. Perosa and M. Selva,
3405–3413. Chem. – Eur. J., 2009, 15, 12273–12282.
31 R. Lungwitz and S. Spange, ChemPhysChem, 2012, 13, 40 P. S. Kulkarni, L. C. Branco, J. G. Crespo, M. C. Nunes,
1910–1916.
A. Raymundo and C. A. M. Afonso, Chem. – Eur. J., 2007, 13,
32 R. Lungwitz and S. Spange, New J. Chem., 2008, 32, 392–394.
8478–8488.
33 A. A. Gorman, M. G. Hutchings and P. D. Wood, J. Am. 41 J.-P. Mikkola, P. Virtanen, H. Karhu, T. Salmi and
Chem. Soc., 1996, 118, 8497–8498. D. Y. Murzin, Green Chem., 2006, 8, 197–205.
34 S. Spange, S. Prause, E. Vilsmeier and W. R. Thiel, J. Phys. 42 J.-P. Mikkola, P. Virtanen and R. Sjoholm, Green Chem.,
Chem. B, 2005, 109, 7280–7289. 2006, 8, 250–255.
¨
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2017
New J. Chem.