Novel Chloroimidazolium-Based Ionic Liquids: Synthesis, Characterisation and Behaviour as Solvents to Control Reaction Outcome

Article abstract of DOI:10.1002/cplu.201600099

Novel ionic liquids containing chlorine atoms on the imidazolium cation were synthesised. The physicochemical properties of these ionic liquids were investigated extensively, including glass transition, melting and decomposition temperatures, density, vis

Full text of DOI:10.1002/cplu.201600099

DOI: 10.1002/cplu.201600099
Full Papers
Novel Chloroimidazolium-Based Ionic Liquids: Synthesis,
Characterisation and Behaviour as Solvents to Control
Reaction Outcome
Rebecca R. Hawker, Janjira Panchompoo, Leigh Aldous, and Jason B. Harper*[a]
Novel ionic liquids containing chlorine atoms on the imidazoli-
um cation were synthesised. The physicochemical properties
of these ionic liquids were investigated extensively, including
glass transition, melting and decomposition temperatures,
density, viscosity, miscibility with common solvents and elec-
trochemical window. The behaviour of these ionic liquids as
solvents was examined through temperature-dependent kinet-
ic analyses on two reactions: a nucleophilic aromatic substitu-
tion (SNAr) reaction and a bimolecular nucleophilic substitution
(SN2) reaction. The properties and effects on reaction outcome
of these new ionic liquids were shown to correlate with the
change in the nature of the cation.
Introduction
The range of ionic liquids, defined as salts that have a normal
melting point below 1008C,[1] is vast,[2] and includes those con-
taining imidazolium, pyridinium, pyrrolidinium, ammonium
and phosphonium cations and a variety of organic and inor-
ganic anions.[3] Composed only of charged species, ionic liquids
have particular properties, including negligible vapour pres-
sure.[4] The utility of ionic liquids is increased by the fact that
through changing the constituent ions, physical properties
such as melting point, viscosity and miscibility with other sol-
vents can be tailored;[5] thus, ionic liquids can be considered
“designer solvents”[6] and it is suggested that “task-specific”[7]
ionic liquids may be developed. There is even the potential to
use ionic liquids composed of mixtures of multiple cations and
multiple anions to give a wider range of solvents,[8] though
such mixtures are rarely well characterised.[9]
ionic liquid, particularly the cation,[11] has been examined ex-
tensively (with common modifications including variation of
the alkyl chain length(s) on the heteroatom(s) and, if appropri-
ate, the addition of alkyl groups to the heterocyclic ring) in
terms of their effect on the physicochemical properties; how-
ever, the effect of such changes on reaction control have not
been considered extensively. Note that variation of the anion
of an ionic liquid often complicates the interpretations of any
changes in properties and solvent effects, as the changes to
the anion often lead to multiple features (size, coordination
ability, etc.) of that ion changing simultaneously; this has been
particularly marked in the past for ionic liquids involving the
hexafluorophosphate anion.[10m]
The new ionic liquids targeted here are based on modified
imidazolium cations. This cation type was chosen for modifica-
tion owing to the relative ease of introducing changes, along
with the fact that systematic changes to the cation might be
expected to have predictable effects on the physicochemical
properties of the ionic liquid.[11] Further, interaction of compo-
nents of the solution with the cation of an ionic liquid has
been shown to be responsible for ionic liquid solvent effects in
a range of cases.[10d,g,h,j,l] The chloro substituent was chosen as
an electron-withdrawing group as it was anticipated that it
could be introduced relatively easily and its size is comparable
to that of a methyl group,[12] which had been incorporated in
similar ionic liquids previously. The bis(trifluoromethanesulfo-
nyl)imide anion ([Tf2N]À) was chosen as it is particularly
common in the ionic liquid literature because of the favourable
properties it imparts to ionic liquids, including lower viscosity
and relatively good electrochemical stability.[4] Ionic liquids
containing this anion have also been studied extensively for
their effects on reaction outcome.[10b–j,l,o]
Ionic liquids are also potential alternatives to molecular sol-
vents because they can alter reaction outcomes (rate, selectivi-
ty, etc.). Through systematic investigation of the effects that
ionic liquids have on the outcome of well-described reac-
tions,[10] we have developed a set of principles for predicting
the effects that an ionic liquid might have on reaction out-
come. These principles involve considering the extent of
charge development in the transition state,[10b] the nature of
the starting materials,[10a,b,h] the proportion of the ionic liquid in
the reaction mixture,[10a,f–h,j,k,p] and the structure of the constitu-
ent ions in the solvent.[10i–n]
This developing predictive framework suggests that there is
the possibility to design an ionic liquid to affect reaction out-
come in a given fashion, as well as affecting its physicochemi-
cal properties. Systematic variation of the components of an
[a] R. R. Hawker, Dr. J. Panchompoo, Dr. L. Aldous, Assoc. Prof. J. B. Harper
School of Chemistry,University of New South Wales
Sydney, NSW 2052 (Australia)
As such, this manuscript describes the synthesis of three
new chlorinated ionic liquids (Figure 1): 1-butyl-4-chloro-3-
E-mail: j.harper@unsw.edu.au
methylimidazolium
bis(trifluoromethanesulfonyl)imide
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cplu.201600099.
([b4Clmim][Tf2N], 1), 1-butyl-4,5-dichloro-3-methylimidazolium
ChemPlusChem 2016, 81, 574 – 583
574
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
starting material 6 was observed through H NMR spectrosco-
py after four weeks at room temperature, with no indication of
the side reactions described above.
As 2-chloroimidazole and its derivatives were not readily
available, [b2Clmim][Tf2N] 3 was prepared from the corre-
sponding non-halogenated salt 1-butyl-3-methylimidazolium
chloride ([bmim][Cl], 9) through deprotonation and reaction
with hexachloroethane. After screening of a number of bases,
methylmagnesium bromide was found to be an experimentally
straightforward choice (Scheme 2). The intermediate mixed salt
was not isolated but converted to the desired ionic liquid 3
through anion metathesis; thus, although the mixed salt was
not purified, the final compound was prepared successfully.
Figure 1. Novel chloroimidazolium-based ionic liquids synthesised in this
work.
bis(trifluoromethanesulfonyl)imide ([b45Cl2mim][Tf2N], 2) and
1-butyl-2-chloro-3-methylimidazolium bis(trifluoromethanesul-
fonyl)imide ([b2Clmim][Tf2N], 3). The physical properties, in-
cluding glass transition, melting and decomposition tempera-
tures, density, viscosity, miscibility and electrochemical stability
of ionic liquids 1–3 are then examined extensively and their
utility as solvents is investigated using two well-described or-
ganic processes.
Results and Discussion
Scheme 2. Synthesis of [b2Clmim][Tf2N] 3. i) CH3MgBr, THF, 08C; ii) C2Cl6, THF,
À788C; iii) Li[Tf2N], H2O, 24 h, 47%.
Synthesis of novel ionic liquids
In the cases for which the appropriate chlorinated imidazoles
were commercially available at reasonable cost (namely species
4 and 5), the chlorinated ionic liquids were prepared using
methods analogous to those described in the literature
(Scheme 1);[13] this gave access to the salts [b4Clmim][Tf2N]
1 and [b45Cl2mim][Tf2N] 2.
The series shown in Figure 1 might be extended with the
salt 1-butyl-3-methyl-2,4,5-trichloroimidazolium bis(trifluoro-
methanesulfonyl)imide ([b245Cl3mim][Tf2N]). The correspond-
ing imidazole was not commercially available, and reaction of
the ionic liquid [b45Cl2mim][I] 8 in a fashion analogous to that
shown in Scheme 2 was unsuccessful with reaction of the
iodide counterion implied (see Supporting Information for fur-
ther details). This was not investigated further.
Characterisation
It is important to consider the properties of these novel salts
1–3, and particularly, to compare these properties to those of
the related ionic liquids shown in Figure 2: the parent ionic
Scheme 1. Synthesis of the novel ionic liquids [b4Clmim][Tf2N] 1 and
[b45Cl2mim][Tf2N] 2. i) KOH, CH3CN, 2 h; ii) CH3I, 24 h, 88%;
iii) CH3CH2CH2CH2X, X=I, r.t. 4 weeks, 79%, X=Cl, 708C, 9 d, 64%; iv) LiTf2N,
H2O, 24 h, R=H 60%, R=Cl 84%.
Note that butylation of chloroimidazole 5 was initially at-
tempted by using the same conditions as for the formation of
salt 7, using n-chlorobutane as the alkylating agent. The reac-
tion was found to proceed slowly, presumably because of the
greatly reduced nucleophilicity of reagent 6 compared with
imidazole 5. Replacing the electrophile with the n-bromobu-
tane also resulted in an impractical reaction rate at room tem-
perature (ꢀ5% conversion in four days); however, increasing
the temperature resulted in the formation of a mixture of
products consistent with the desired [b45Cl2mim]+ undergoing
a reverse Menschutkin reaction to give the 1-butylimidazole,
which subsequently reacted further (see Supporting Informa-
tion for further details). To avoid this unwanted side reaction,
we used n-iodobutane, and complete consumption of the
Figure 2. Non-halogenated ionic liquids 10–12 considered in this work for
comparison.
liquid 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)-
imide ([bmim][Tf2N], 10) and the methylated ionic liquids 1-
butyl-2,3-dimethylimidazolium bis(trifluoromethylsulfonyl)imide
([bm2im][Tf2N], 11) and 1-butyl-3,4,5-trimethylimidazolium bis-
(trifluoromethylsulfonyl)imide ([bm3im][Tf2N], 12). These latter
ionic liquids were chosen because of the similarity of the
methyl and chloro groups in terms of size. The comparisons
allow any effects of chlorination of the cation to be seen, as
well as giving some indication of the relative importance of
steric and electronic effects.
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
575
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Ideally, the methyl analogue of ionic liquid 1, 1-butyl-3,4-di-
methylimidazolium bis(trifluoromethanesulfonyl)imide, would
also be used. The corresponding 1,5-dimethylimidazole is not
readily available commercially, and attempted preparation of
this intermediate from 4(5)-methylimidazole (analogous to that
shown in Scheme 1) gave an inseparable mixture of regioisom-
ers (see Supporting Information for further details); therefore,
the preparation was not pursued further.
uids 1–3 are quite high, which, combined with the glass transi-
tion data above, means they may be useful as solvents over
a wide range of temperatures. For each of the salts 1–3, the
decomposition temperature was slightly lower than that of the
non-chlorinated ionic liquid 10, (Td =712 K[13a]) suggesting that
addition of chlorine atoms to the ionic liquid cation facilitates
decomposition; this is unsurprising given that a CÀCl bond is
weaker than a CÀH bond.[15] The order of the measured de-
composition temperatures is [b45Cl2mim][Tf2N] 2 <[b2Clmim]
[Tf2N] 3 <[b4Clmim][Tf2N] 1, which suggests that the greater
the number of chlorine atoms on the cation, the lower the
thermal stability of the resultant ionic liquid; this, again, is rea-
sonable given the relative CÀCl and CÀH bond strengths. Fur-
ther, the trend shown demonstrates that the position of the
chlorine atom is important, with ionic liquid 1 more stable
than ionic liquid 3; the relative reactivity of the chlorine group
at each of these positions in different contexts will also be dis-
cussed further below.
Thermophysical properties of ionic liquids 1–3
It is important to understand the thermal properties of an
ionic liquid as they demonstrate the temperature range over
which the ionic liquid may be used as a solvent. The broader
the temperature range, the more useful the ionic liquid be-
comes.
Each of the ionic liquids 1–3 was analysed by differential
scanning calorimetry (DSC) and thermogravimetric analysis
(TGA) to determine the glass transition temperature (Tg), melt-
ing temperature (Tm) and decomposition temperature (Td);
these data are given in Table 1, with original scans supplied in
the Supporting Information.
Density and viscosity of ionic liquids 1–3 and 10–12
The density and viscosity of an ionic liquid are important prop-
erties to know to determine the utility of the salt as a solvent.
For example, optimal physical properties for an ionic liquid to
act as an effective solvent might include a low viscosity to fa-
cilitate mixing, and a large density compared with other pro-
cess fluids to hasten phase separation.[13a]
Table 1. Glass transition, melting and decomposition temperatures of
ionic liquids 1–3.
Ionic liquid
Tg [K]
Tm [K]
Td [K][b]
[b4Clmim][Tf2N] 1
[b45Cl2mim][Tf2N] 2
[b2Clmim][Tf2N] 3
200.5Æ0.2[a]
211.1Æ0.1[b]
199.0Æ0.2[a]
316
671.9Æ1.5
619.4Æ2.3
634.7Æ1.3
Density and viscosity measurements were made for each of
the ionic liquids 1–3, 10 and 12 at both 258C (for comparison
with literature data) and 508C (for comparison with ionic liquid
1; the melting point of the chlorinated ionic liquid 1 (41–448C)
precluded the examination of density and viscosity at the stan-
dard temperature of 258C as used in the literature). The results
of the density and viscosity measurements are shown in
Table 2, along with literature data for ionic liquid 11.
[c]
–
[c]
–
[a] Uncertainties quoted are the standard deviation of triplicate measure-
ments. [b] Uncertainties quoted are the standard deviation of duplicate
measurements. [c] No melting transitions could be observed.
All three of the novel ionic liquids 1–3 vitrified at low tem-
peratures and underwent a glass transition. The order of the
glass transition temperatures was ionic liquid [b2Clmim][Tf2N]
3 < [b4Clmim][Tf2N] 1 < [b45Cl2mim][Tf2N] 2. These data dem-
onstrate that addition of chlorine atoms to the ionic liquid
cation increases the glass transition temperature. This can be
seen through initial comparison with the non-chlorinated ionic
liquid 10, which has a glass transition temperature of 186 K,[14]
but also through comparison of the data for species 1–3.
Those species with a single chlorine atom added (1 and 2)
have comparable glass transition temperatures, whereas the
data for ionic liquid 3 show that multiple chlorine atoms on
the cation increase the glass transition temperature further.
Only glass transitions could be discerned clearly for ionic liq-
uids 2 and 3. Ionic liquid 1 displayed a clear melting point for
the initial solid but also demonstrated significant supercooling,
such that melting features were not observed on subsequent
temperature cycles.
For [bmim][Tf2N] 10, the density determined was consistent
with that reported in the literature.[16] The addition of the
methyl groups to the cation causes a decrease in density,
which is observed in both the reported density[16b,17] of
[bm2im][Tf2N] 11 and the density determined here for [bm3im]
[Tf2N] 12. The differences in density as a result of C2-methyla-
tion have been attributed previously to the looser interionic
packing in the parent ionic liquid 10 than in the methylated
case 11.[18]
Addition of chloro substituents to the cation, as in ionic liq-
uids [b4Clmim][Tf2N] 1 and [b2Clmim][Tf2N] 3, causes the den-
sity to increase. The position of the chlorine atom does not
appear to have a significant effect, as the densities of ionic liq-
uids 1 and 3 are similar (at 508C). With two chloro substitu-
ents, the density is increased even further, as seen in the case
of [b45 L2mim][Tf2N] 2. Hence, the addition of a chlorine atom
has the opposite effect to a methyl group, which can be attrib-
uted to the introduction of increased dispersion forces owing
to incorporation of the chlorine centre.[11] This effect is en-
hanced further by the larger atomic weight of the chlorine
atom relative to that of carbon and hydrogen.
The decomposition temperature was determined using the
onset method, that is, the intercept of the extrapolated base-
line and the tangent of the first significant decomposition fea-
ture. The decomposition temperatures for each of the ionic liq-
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
576
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. Densities and viscosities of ionic liquids 10–12 and 1–3.
Ionic liquid
Density [gcmÀ3
]
Viscosity [mPas]
258C
508C
258C
508C
[bmim][Tf2N] 10
[bm2im][Tf2N] 11
[bm3im][Tf2N] 12
[b4Clmim][Tf2N] 1
[b45Cl2mim][Tf2N] 2
[b2Clmim][Tf2N] 3
1.4373Æ0.0018
1.4147Æ0.0005
50.34Æ0.82
20.670Æ0.071
[b]
[b]
1.4159[a]
–
93.33[a]
–
1.3892Æ0.0007
1.3661Æ0.0005
1.4703Æ0.0002
1.5209Æ0.0005
1.4761Æ0.0005
70.7Æ1.6
25.54Æ0.40
33.61Æ0.17
49.76Æ0.49
38.40Æ0.52
[c]
[c]
–
–
1.5467Æ0.0004
189.8Æ2.5
1.5005Æ0.0002
120.52Æ0.73
[a] From Katsuta et al.[16b] with no errors reported. [b] Not reported. [c] Salt 1 is a solid at this temperature.
For [bmim][Tf2N] 10, the viscosity determined here was also
consistent with that reported in the literature.[16b,17] The pres-
ence of methyl groups on the cations of the ionic liquids in-
creases the viscosity, which is observed in both the reported
value for [bm2im][Tf2N] 11[16b] and the viscosity determined
here for [bm3im][Tf2N] 12. The viscosity of [bm2im][Tf2N] 11 is
higher than that of [bm3im][Tf2N] 12, which suggests that the
position of the methyl group is important in determining the
viscosity; methylation at the C2 position is particularly signifi-
cant, more so than the number of methyl groups.
atoms to the cation, the viscosity increases even further, with
ionic liquid [b45Cl2mim][Tf2N] 2 having a viscosity almost four
times that of the parent ionic liquid 10 and almost double that
of the di-methylated ionic liquid 12.
Miscibility of ionic liquids 1–3, 10 and 12 with common sol-
vents
The miscibility of each of the ionic liquids 1–3, 10 and 12 in
common solvents was examined; the solvents chosen provided
a range of properties, including polarity, hydrogen-donating
ability, extent of halogenation and aromaticity. Ionic liquid 1 is
solid at room temperature, so rather than its miscibility, its sol-
ubility in each of the solvents was examined.
The importance of the alkylation site has been studied previ-
ously by Bonhôte et al.,[5] who compared the series of ionic liq-
uids 13–15 (Figure 3). The viscosity of the C2 methylated ionic
liquid 14 was 88 MPas (at 208C), which is significantly higher
than that of the C5 methylated ionic liquid 15 (37 MPas) and
ionic liquid 13 (34 MPas) at the same temperature. Subsequent
ab initio studies[19] suggested that the increased viscosity is
caused by the high potential energy barriers between ion pair
conformations, which limit the movement of ions.
It was found that irrespective of the ionic liquid used (in-
cluding in the solubility studies of the solid 1), all of the ionic
liquids 1–3, 10 and 12 were miscible (or soluble) with most of
the solvents considered (for full details, see Supporting Infor-
mation); immiscibility (or insolubility) was observed only for
the solvents that have extreme polarity. All the ionic liquids
were partially soluble in both diethyl ether and toluene. This
may be because these have similar but low polarities;[20] how-
ever, there is likely to be a contribution in the toluene case
from cation–multipole interactions, as used previously to de-
scribe the anomalous solubility of benzene in ionic liquids.[21]
(Induced fluctuations in the electrostatics of the solute may en-
hance the solubility of these species; however, the methodolo-
gy used in the initial molecular dynamics studies[21] demon-
strated that this was not necessary to explain the observed sol-
ubility).
Figure 3. Methylated ionic liquids 13–15 considered previously by Bonhôte
et al.[5]
Upon chloro substitution of the cation, the viscosity of the
ionic liquid increases significantly. For [b2Clmim][Tf2N] 3, the
viscosity was about twice that of the parent ionic liquid
[bmim][Tf2N] 10 and also greater than that reported for the C2
methylated ionic liquid 11.[16b,17] In a similar fashion to the 2-
substituted case 3, the viscosity of ionic liquid [b4Clmim][Tf2N]
1 is higher than that of [bmim][Tf2N] 10 at the only common
temperature (508C). Upon comparison with [b2Clmim][Tf2N] 3,
the C4 chlorinated ionic liquid 1 was found to have a slightly
lower viscosity, demonstrating the importance of the site of
substitution. This is consistent with the Bonhôte study dis-
cussed above,[5] in which it was shown that blocking of the C2
position caused a higher viscosity than the equivalent substitu-
tion at the C4 position, though it is not clear if the source of
the change is the same. Upon addition of multiple chlorine
These results demonstrate that the addition of chlorine
atoms to the cation does not affect significantly the interaction
of an ionic liquid with a range of solvents, suggesting that, on
miscibility grounds, the novel ionic liquids 1–3 might be used
as solvents, replacing the ionic liquid 10 with little effect.
Electrochemical windows of ionic liquids 1–3 and 10
The electrochemical window measures the relative electro-
chemical potentials an electrolyte can withstand before the
onset of (often unwanted) oxidation and reduction reactions.
Chemical stability is often mirrored by electrochemical stability,
such that some exceptionally stable ionic liquids have been
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
577
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
demonstrated to be superior electrolytes (both neat and in
dilute solution) compared with “conventional” electrolytes.
The electrochemical experiments were undertaken using
both platinum and glassy carbon (GC) working electrodes.
These electrode materials are common in the literature,[5,22]
and sometimes have different electrochemical windows owing
to their ability to facilitate (or electrocatalyse) processes to dif-
ferent degrees.[23] For example, the electrochemical reduction
of imidazolium cations is believed to result in an unstable radi-
cal (centred on the C2 carbon), which can dimerise[24] and po-
tentially yield imidazol-2-ylidene carbenes and molecular hy-
drogen.[25] Platinum electrodes are known to electrocatalyse
this process if the C2 carbon contains a labile proton, such
that cation reduction at platinum electrodes occurs at poten-
tials about 0.5 V more positive than at GC electrodes (i.e., plati-
num electrodes have smaller electrochemical windows than
GC electrodes in imidazolium ionic liquids with C2 protons).
This difference is reduced significantly by replacing the C2
proton with a methyl group.[5,22]
Acetonitrile is relatively electrochemically stable,[26] and the use
of a dilute solution of ionic liquid allows comparisons that are
unbiased by differences in the ionic liquid properties such as
melting points, density and viscosity.
The results for the parent salt 10 are consistent with litera-
ture reports;[27] both types of working electrode display large
electrochemical windows (ꢀ5 V), with the GC case having
a larger window than platinum owing to the reduction process
described above. Oxidation (of the [Tf2N]À anion[28]) is observed
at similar potentials.
Upon replacing the C2 hydrogen with a chlorine atom to
give the ionic liquid [b2Clmim][Tf2N] 3, a smaller electrochemi-
cal window is observed. No significant change in the oxidation
process is seen, but reduction of the cation now occurs at simi-
lar potentials for both electrodes. This is reasonable because
the CÀCl bond is weaker than both the CÀH bond and the
C=N bond.[15] This would account for the electrochemical
window remaining the same irrespective of the electrode used,
as only electrode materials such as silver are known to electro-
catalyse alkyl halide cleavage reactions, but platinum and GC
are classed as relatively “inert” (i.e., equally poor electrocata-
lysts).[29] The precise reduction products are not known, but
are likely to involve either a C2-centred imidazole radical or an
imidazol-2-ylidene-based carbene and a chloride anion; elec-
trochemical reduction of the non-chlorinated analogue has
been reported to form radicals or carbenes at GC[24] and plati-
num[25] electrodes, respectively.
The electrochemical windows of each of the ionic liquids 1–
3 and 10 were evaluated at both GC and platinum electrodes
by performing cyclic voltammetry in approximately 0.1m solu-
tions of the ionic liquids dissolved in acetonitrile (Figure 4).
Regarding the ionic liquid [b4Clmim][Tf2N] 1, there is no ob-
servable difference in the electrochemical window upon com-
parison with the parent ionic liquid 10. As ionic liquid 1 has
a C2ÀH bond that is activated at low potentials by the electro-
catalytic platinum electrode, it is reasonable that the reduction
process observed is similar to that in the case of the ionic
liquid 10. For the ionic liquid [b45Cl2mim][Tf2N] 2, reduction
occurs at the platinum electrode at slightly more positive po-
tentials, either because of the electron-withdrawing effect of
the two chlorine atoms, or owing to some electrochemical pro-
cess involving the chlorine atoms on adjacent carbon centres.
The reduction processes of each of the ionic liquids 1 and 2
at the GC electrode are considerably different from both the
parent ionic liquid 10 and the results observed at the platinum
electrode. For the mono-chlorinated ionic liquid 1, one reduc-
tion process is observed, whereas the di-chlorinated ionic
liquid 2 exhibits two processes. Clear hysteresis is also seen,
consistent with passivation of the GC electrode through the
formation of a dense organic layer that blocks the electrode
surface.[30] This is consistent with direct cleavage of the CÀCl
bond resulting in reactive organic radicals that react with the
sp2-hybridised GC surface. A similar process has been noted for
the reduction of alkylferrocenes with a terminal iodo substitu-
ent, for which reduction at À1.7 V versus Ag/AgCl at a GC elec-
trode in acetonitrile resulted in the covalent immobilisation of
the alkylferrocene species at the GC surface through CÀI cleav-
age.[31]
Figure 4. Electrochemical window of ionic liquids 1–3 and 10 in 0.1m solu-
tions in acetonitrile, recorded at platinum (red) and GC (black) electrodes.
Arrows indicate the direction of the scan, and dashed lines indicate the ap-
proximate electrochemical window of [bmim][Tf2N] 10. Scans were recorded
Interestingly, the oxidation features of ionic liquids 1 and 2
also change from those of the parent ionic liquid 10 at both
GC and platinum electrodes. Recent computational investiga-
under a nitrogen atmosphere at a scan rate of 5 mVsÀ1
.
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
578
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
tions have demonstrated that the oxidation potentials of
[Tf2N]-based ionic liquids are limited by the [Tf2N]À anion,
whereas those of imidazolium-based ionic liquids with [PF6]À
and [BF4]À anions are limited by the relative oxidative (in)stabil-
ity of the cation.[28] Hysteresis and passivation were observed
for cases 1 and 2 at the GC electrode, consistent with the elec-
trochemistry of the cation rather than the [Tf2N]À anion; oxida-
tion was also more facile at the platinum electrode. Although
multiple authors have reported anions limiting the reductive
electrochemical window of ionic liquids,[32] the results de-
scribed here are only the second example after that of the cy-
clopropenium salts reported by Curnow et al.[33] of an ionic
liquid cation limiting the oxidative window of an ionic liquid.
Ultimately, all the ionic liquids 1–3 prepared here have large
electrochemical windows (from ꢀ4.5 V at platinum before
chlorination to ꢀ4.0 V after dichlorination), indicating that they
are pure, relatively stable and unlikely to undergo unwanted
side reactions. Additionally, new electrochemical functionality
can be introduced; ionic liquid layers have been formed delib-
erately on GC to introduce functionality,[34] and ionic liquids
1 and 2 present the possibility to do this at a GC electrode by
potentially direct connection through the imidazolium back-
bone, through both reduction and oxidation processes, intro-
ducing additional versatility to this process.
here with the previous cases (salts 10–12), we can assess how
chlorination of the cation affects ionic liquid solvent effects.
Nucleophilic aromatic substitution (SNAr)
The reaction between the fluorodinitrobenzene 16 and ethanol
(Scheme 3) is a well-known reaction that proceeds through
a nucleophilic aromatic substitution mechanism.[10f] It has been
found that the key interaction responsible for ionic liquid sol-
vent effects on this process is that between the fluorodinitro-
benzene 16 and the anion of the ionic liquid.[10f] This interac-
tion is responsible for the enthalpic cost and entropic benefit
(the balance of which, in turn, results in an increase in rate
constant) on moving from ethanol to the parent ionic liquid
10.[10f] Upon further study of this rate enhancement, it has
been shown that there is a general ionic liquid effect, with the
rate enhancement virtually independent of the nature of either
the cation or the anion.[10i] It was of interest to determine
whether the chlorinated systems 1 and 2 would have the
same effect as the ionic liquids used previously (the 2-chloro
system 3 was not considered as initial studies showed that it
underwent reaction with ethanol under these reaction condi-
tions; for more information, see Supporting Information).
As it is known that there is a general ionic liquid effect for
this reaction, it was anticipated that addition of chloro sub-
stituents to the imidazolium cation would have little effect on
the rate constant. That is, the novel ionic liquids 1 and 2 were
expected to behave in a similar fashion to previously examined
ionic liquids, including 10–12.[10i]
Kinetic analyses
For investigation of the effects of the novel ionic liquids 1–3
on the outcome of organic processes, they were examined as
solvents for two well-described reactions. Chosen because
they each proceed to give a single product (so there are no
issues with selectivity), are readily monitored and have been
investigated previously in ionic liquid media, particularly the
parent ionic liquid 10 and the methylated species 11 and 12,
these standard reactions were the nucleophilic aromatic substi-
tution reaction between 1-fluoro-2,4-dinitrobenzene 16 and
ethanol (Scheme 3),[10f,i] and the bimolecular nucleophilic sub-
stitution reaction between benzyl bromide 18 and pyridine
(Scheme 4).[10d,g,h,j,m] By comparing the solvent effects observed
It should be noted that no attempt was made to isolate the
product 17, nor determine a yield. This is because any isolated
yield would incorporate a component from the extent of con-
version (the effect of the ionic liquid on the rate) and a compo-
nent from the efficacy of isolation (which will be complicated
by interactions of aromatic 17 with the solvent[21]). This is con-
sistent with previous reports.[10f,i]
The temperature-dependent kinetic analyses in this study
were performed in the same manner as previous studies[10f,i]
using 19F NMR spectroscopy, allowing the construction of an
Eyring plot (Figure 5) for the reaction in ethanol and each of
the ionic liquids 2 and 3 and the determination of activation
parameters in each of the solvent systems (Table 3). Also in-
cluded to allow direct comparison of solvent effects are data
for the non-chlorinated systems 10–12. The mole fraction of
the novel ionic liquids 1–3 used was approximately 0.5, to
enable direct comparison with the previous studies.[10f,i]
The second-order rate constants and activation parameters
determined in ethanol were consistent with those reported in
the literature.[10f] As well as enabling determination of the acti-
vation parameters, the Eyring plot (Figure 2) can be thought of
as a representation of the relative rate constant data, and
shows that each of the ionic liquids 1 and 2 results in larger
rate constants for the reaction over the range of temperatures
used than those observed in the molecular solvent ethanol.
Further, the effect of chlorination compared to methylation on
the rate constants was minimal (Eyring plots comparing the
chlorinated and methylated ionic liquids are presented in Fig-
Scheme 3. Nucleophilic aromatic substitution reaction of 1-fluoro-2,4-dinitro-
benzene 16 to give the ether 17. i) EtOH, NEt3.
Scheme 4. Menschutkin reaction between benzyl bromide 18 and pyridine
to give the salt 19. i) Pyridine.
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
579
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
ionic liquid 10.[10d] The extent of this interaction, and hence the
rate enhancement, has been shown to be dependent on the
ease of accessing the charged nitrogen centre of the cation.[10j]
Hence, it is of interest here to consider how changing the elec-
tronic nature of the substituent affects the interaction, and
hence the rate enhancement, so the process was considered in
the chlorinated ionic liquids 1–3 (note that in this case the 2-
chloro derivative 3 was found to be stable under the reaction
conditions and was hence investigated as a solvent). It was an-
ticipated that the electron-withdrawing nature of the substitu-
ents on the novel ionic liquids 1–3 would cause an increase in
the key interaction identified previously, and hence, in the rate
constant for the process, relative to their methylated counter-
parts.
Figure 5. The bimolecular Eyring plot, from which the activation parameters
in Table 3 were determined, for the reaction between the fluorobenzene 16
and ethanol in either ethanol (black) or one of the ionic liquids [b4Clmim]
[Tf2N] 1 (blue) or [b45Cl2mim][Tf2N] 2 (red) at cIL ꢀ0.5. Exact mole fractions
are outlined in Table 3.
For the same reasons as in the case above, no attempt was
made to isolate the product 19, nor determine a yield. Once
again, this is consistent with previous reports.[10d,g,h,j,m]
Temperature-dependent kinetic analyses were performed as
1
described previously[10d,g,j] using H NMR spectroscopy, allowing
the construction of an Eyring plot (Figure 6) for the reaction in
acetonitrile and each of the ionic liquids 1–3, and the determi-
nation of activation parameters in each of the solvent systems
(Table 4). The mole fraction of the novel ionic liquids 1–3 used
Table 3. Activation parameters for the reaction between the fluoroben-
zene 16 and ethanol in either ethanol or one of the ionic liquids 1, 2 or
10–12, at cIL ꢀ0.5.
[a]
[a]
Solvent
cIL
DH° [kJmolÀ1
]
DS° [JKÀ1 molÀ1
]
ethanol
0
48.7Æ2.1
49.6Æ0.5
49.0Æ3.9
51.5Æ3.2
54.0Æ1.3
51.5Æ2.1
À249Æ7
À229Æ2
À223Æ12
À211Æ10
À213Æ4
À220Æ6
[bmim][Tf2N] 10[b]
[bm2im][Tf2N] 11[c]
[bm3im][Tf2N] 12[c]
[b4Clmim][Tf2N] 1
[b45Cl2mim][Tf2N] 2
0.54
0.52
0.50
0.53
0.52
[a] Uncertainties quoted are from the fit of the linear regression. [b] Data
reproduced from Jones et al.[10f] [c] Data reproduced from Tanner et al.[10i]
ures S6 and S7, Supporting Information). This was consistent
with the general ionic liquid effect observed previously, and
was as expected given the minimal changes seen upon modifi-
cation of the cation.[10i]
Figure 6. The bimolecular Eyring plot, from which the activation parameters
in Table 4 were determined, for the reaction between benzyl bromide 18
and pyridine in either acetontitrile (black) or one of the ionic liquids
[b4Clmim][Tf2N] 1 (blue), [b45Cl2mim][Tf2N] 2 (red) or [b2Clmim][Tf2N] 3
(green), at cIL ꢀ0.9. Exact mole fractions are outlined in Table 4.
The activation parameters (Table 3) demonstrate that the
rate constant enhancement observed in all the ionic liquids rel-
ative to that in ethanol is caused by an increase in the entropy
of activation, which again is consistent with previous observa-
tions, and was as anticipated.[10f,i] From these data it is clear
that the chlorinated ionic liquids 1 and 2 behave in a predicta-
ble fashion as solvents for this nucleophilic aromatic substitu-
tion reaction.
Table 4. Activation parameters for the reaction between benzyl bromide
18 and pyridine in either acetonitrile or one of the ionic liquids 1–3 and
10–12, at cIL ꢀ0.9.
[a]
[a]
Solvent
cIL
DH° [kJmolÀ1
]
DS° [JKÀ1 molÀ1
]
Bimolecular nucleophilic substitution (SN2)
ethanol
0
44.7Æ0.8
49.9Æ0.8
53.5Æ2.8
53.5Æ1.1
54.4Æ1.4
53.7Æ1.8
53.8Æ1.0
À220Æ3
À195Æ3
À183Æ10
À189Æ4
À180Æ5
À181Æ6
À183Æ3
The Menschutkin reaction (Scheme 4) between benzyl bromide
18 and pyridine proceeds through an SN2 mechanism.[35] This
reaction has been studied previously in ionic liquids.[10d,g,h,j,m] It
has been found that the key interaction in ionic liquid solu-
tions is that between pyridine and the cation of the ionic liq-
uid.[10g] This interaction is responsible for the enthalpic cost
and entropic benefit (which, in turn, results in an increase in
rate constant) upon moving from acetonitrile to the parent
[bmim][Tf2N] 10[b]
[bm2im][Tf2N] 11[c]
[bm3im][Tf2N] 12[c]
[b4Clmim][Tf2N] 1
[b45Cl2mim][Tf2N] 2
[b2Clmim][Tf2N] 3
0.86
0.88
0.85
0.86
0.85
0.86
[a] Uncertainties quoted are from the fit of the linear regression. [b] Data
reproduced from Yau et al.[10d] [c] Data reproduced from Tanner et al.[10j]
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
580
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
was about 0.9, enabling direct comparison with previous stud-
ies.[10d,g,j]
Conclusion
The second-order rate constants and activation parameters
determined in acetonitrile were consistent with those reported
previously.[10j] Performed in each of the ionic liquids [b4Clmim]
[Tf2N] 1, [b45Cl2mim][Tf2N] 2 and [b2Clmim][Tf2N] 3, an in-
crease in the rate constant of the Menschutkin reaction was
observed over the range of temperatures used compared with
the molecular solvent, acetonitrile.
The syntheses of three novel imidazolium-based ionic liquids
containing chlorine atoms have been described. The thermo-
physical and physicochemical properties of each of the ionic
liquids have been measured and compared with those of the
corresponding protiated and methylated systems. The chlori-
nated ionic liquids have a higher density and viscosity than
their non-chlorinated counterparts, along with the same misci-
bility, whereas the introduction of a chlorine atom reduces the
electrochemical window of the system slightly and decreases
the temperature at which it decomposes. Interestingly these
chlorinated cations demonstrate both reductive and oxidative
instability. Importantly, these changes are as expected on the
basis of the structural changes made and trends reported in
the literature, demonstrating that modification of the cation
can be used to alter predictably the properties of an ionic
liquid.
The ionic liquids [b4Clmim][Tf2N] 1 and [b2Clmim][Tf2N] 3
exhibited almost identical rate constants over the range of
temperatures used, indicating that the position of the chlorine
is not important in determining solvent effects on the rate
constant. The dichlorinated ionic liquid 2 showed a slightly in-
creased rate of reaction over the range of temperatures com-
pared with the monochlorinated cases 1 and 3. The origin of
all of these rate constant enhancements can be considered by
examining the activation parameters (Table 4). The activation
parameters for all three ionic liquids 1–3 show that the origin
of the rate enhancement is an entropic benefit, which over-
comes the enthalpic cost. This is consistent with previous ob-
servations,[10j] and suggests that there is ordering of the cation
around the starting material 18.
Once again, it is useful to compare the solvent effects on
rate data for chlorinated and non-chlorinated ionic liquids by
using an Eyring plot (see Figures S8 and S9, Supporting Infor-
mation). The rate constant enhancement for ionic liquids
1 and 3 was consistent with that observed for the parent ionic
liquid 10 and the methylated ionic liquid 11. The rate con-
stants for the C2ÀCl ionic liquid 3, the C2ÀCH3 ionic liquid 11
and the parent ionic liquid 10 cases are the same within error,
which suggests that the steric and electron-withdrawing
nature of the chlorine atom has no effect on the rate constant
of the reaction at that position. To a certain extent, this dem-
onstrates a limitation of the predictive tools available, as the
chlorinated cases 1 and 3 might be expected to result in
a faster rate constant, but they were the same within the un-
certainties of measurement across the range of temperatures
considered.
The effects of these salts on the outcomes of organic pro-
cesses were as expected on the basis of the developing under-
standing of the microscopic interactions that drive ionic liquid
solvent effects. The “general ionic liquid effect” seen previously
for an SNAr reaction is replicated, and the balance of introduc-
ing substituents with competing steric and electronic effects is
demonstrated well in the case of an SN2 reaction. In addition,
a rate constant enhancement is seen, demonstrating the po-
tential utility of these species as solvents for organic processes.
The results demonstrate that the cation of an ionic liquid can
be modified in a rational fashion, not only to control the prop-
erties of the solvent, but also reaction outcomes.
The limitations of the existing understanding of ionic liquid
effects are also demonstrated. Quantification of such effects is
not possible (though it might be argued that this is limited
even for molecular solvents[36]) and uncertainties in data may
limit the ability to determine subtle solvent effects.
Although any replacement of existing solvents with novel
ionic liquids must take into account other parameters (includ-
ing cost, stability and appropriateness of physical properties,
relative to both molecular solvents and existing ionic liquids),
overall, the key outcome is that the effects of functionalization
observed on both ionic liquid properties and how they affect
reaction outcome could both be correlated directly to the
structure of the cation and its associated interactions with
other species. With this understanding, there is the potential
to extend this work to the rational design of ionic liquids with
appropriate physical and solvent properties for a given applica-
tion.
The ionic liquid [b45Cl2mim][Tf2N] 2 had a very different rate
constant profile compared with the previously studied [bm3im]
[Tf2N] 12, with greater rate constants observed across the
range of temperatures used. This demonstrates that changing
from two electron-donating methyl substituents to two elec-
tron-withdrawing chlorine substituents (of comparable size) on
the cation of the ionic liquid alters the solvent effect signifi-
cantly. The different effects of methylation compared with
chlorination are reasonable given the argument presented
above: the electron-withdrawing chloro substituents would be
expected to increase the interaction of the cation with the nu-
cleophile pyridine. These electronic effects compete with the
steric interactions, which were used previously to explain the
decrease in rate constants observed on moving between the
parent ionic liquid 10 and the dimethylated case 12.[10j] In the
case of the dichloro system 2, the electronic effects counteract
the negative steric effects, giving a similar rate constant en-
hancement to that seen in the parent ionic liquid 10.
Experimental Section
General experimental details; syntheses of ionic liquids 1, 2 and 3;
reverse Menschutkin reaction in mixtures containing imidazole 6
and n-bromobutane; deprotonation of salt 8 and reaction with
hexachloroethane; alkylation reaction of 4(5)-methylimidazole with
n-iodobutane; raw DSC plots for ionic liquids 1–3; raw TGA curves
of ionic liquids 1–3; complete miscibility data for the salts 1–3, 10
and 12; reaction of ionic liquid 3 in ethanol; Eyring plots compar-
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
581
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
ing the effects of the salts 1 and 2 with the methylated ionic liq-
uids 11 and 12 on the reaction shown in Scheme 3; Eyring plots
comparing the effects of the salts 1–3 with the methylated ionic
liquids 11 and 12 on the reaction shown in Scheme 4; preparation
of the imidazole 6 and the ionic liquids 9, 11 and 12; details of the
electrochemical studies; rate data for all of the plots presented in
this work; and NMR spectra for all new compounds are provided
in the Supporting Information.
(ꢀ0.5 molLÀ1). In all cases the reaction was monitored in situ with
the spectrometer set to the desired temperature for the duration
of the reaction. All reactions were followed until more than 95% of
the starting material (either 16 or 18) was consumed, and all kinet-
ic analyses were performed in triplicate. NMR spectra were pro-
cessed using the MestReNova 8.1.1 software. The pseudo-first-
order rate constants for the reactions were calculated using either
integrations of the signal attributed to the starting material 16 at
dꢀÀ110 (obtained from the processed 19F NMR spectra of the re-
action in Scheme 3) or integrations of the signal attributed to the
benzyl bromide 18 at dꢀ4.6 (obtained from the processed
1H NMR spectra of the reaction in Scheme 4), by fitting the natural
logarithm of the integrations to a linear function using the Micro-
soft Excel 14.4.3 LINEST function. Bimolecular rate constants were
obtained from the pseudo-first-order constants by dividing by the
initial nucleophile concentration in the reaction mixture. The acti-
vation parameters were then determined by fitting the data ob-
tained using the Microsoft Excel 14.4.3 LINEST function to the bi-
molecular Eyring equation shown in Equation (1).[10d,37]
Characterisation
Prior to characterisation and kinetic analyses, the ionic liquids were
dried under reduced pressure (0.1 mbar) at room temperature for
about 4 h, and were found to have <200 ppm water by using Karl
Fischer titration methodology. Ion chromatography was used to
determine if there were residual halide anions in the ionic liquids;
all were <0.2 mol%. The glass transition temperatures and melting
points were measured on
a Mettler Toledo DSC823 with
a TS0800GC1 gas controller. The analysis consisted of heating and
cooling the sample at a rate of 108CminÀ1 under a 50 mLminÀ1
flow of nitrogen. All measurements were performed in either dupli-
cate or triplicate, with uncertainties reported as the standard devia-
tion. The decomposition temperatures were determined by using
a Mettler Toledo TGA/DSC 1 with a GC 200 gas controller. The anal-
ysis consisted of heating the sample from 508C to 5008C at a rate
of 108CminÀ1 under a 50 mLminÀ1 flow of argon. All measure-
ments were performed in duplicate, with uncertainties reported as
the standard deviation. Decomposition temperatures were deter-
mined through the onset method using the Mettler Toledo soft-
ware. The density and viscosity measurements were made on an
Anton Paar DMA 4100m Density Meter and a Lovis 2000 ME Micro-
viscometer, respectively. All measurements were performed in trip-
licate, and uncertainties are reported as the standard deviation of
the triplicate data. Miscibility studies were performed by adding
samples of the ionic liquid (1 mL) and the appropriate solvent
(1 mL) to a measuring cylinder, inverting several times, and then al-
lowing the mixture to settle; the number of phases and their pro-
portions was then observed. Full details are provided in the Sup-
porting Information. The cyclic voltammetry studies were per-
formed on a m-Autolab III potentiostat/galvanostat system in
a three- electrode setup. The data were handled by using Nova
1.10 software. The experiments were performed using either
a glassy carbon (diameter 3 mm) or platinum (diameter 1.4 mm)
working electrode, an AgjAg+ reference electrode and a platinum
counter electrode. Potentials were subsequently referenced to the
FcjFc+ redox couple by adding ferrocene to the solution at the
end of the experiment, repeating the scan, estimating E8(FcjFc+)
from the cyclic voltammogram and referencing the data such that
E8(FcjFc+)=0 V. Further information is given in the Supporting In-
formation.
ꢀ
ꢁ
k2h
kBRT2
DS° DH°
ð1Þ
ln
¼
À
R RT
Tables containing the kinetic data for all the systems described can
be found in the Supporting Information.
Acknowledgements
J.B.H. acknowledges financial support from the Australian Re-
search Council Discovery Project Funding Scheme (DP130102331)
and L.A. acknowledges the Australian Research Council Discovery
Early Career Research Award Scheme (DE130100770). We ac-
knowledge Dr. Hank De Bruyn from the University of Sydney for
DSC support and Jeffrey J. Black from the University of New
South Wales for TGA support. We would also like to acknowledge
the NMR Facility within the Mark Wainwright Analytical Centre
at the University of New South Wales for NMR support.
Keywords: electrochemistry · ionic liquids · kinetics · solvent
design · solvent effects
[1] S. Z. E. Abedin, F. Endres, Acc. Chem. Res. 2007, 40, 1106–1113.
[2] L. Crowhurst, P. R. Mawdsley, J. M. Perez-Arlandis, P. A. Salter, T. Welton,
Phys. Chem. Chem. Phys. 2003, 5, 2790–2794.
[3] J. P. Hallett, T. Welton, Chem. Rev. 2011, 111, 3508–3576.
[4] C. L. Hussey, Pure Appl. Chem. 1988, 60, 1763–1772.
[5] P. Bonhôte, A.-P. Dias, N. Papageorgiou, K. Kalyanasundaram, M. Grätzel,
Inorg. Chem. 1996, 35, 1168–1178.
[6] a) M. Freemantle, Chem. Eng. News 1998, 76, 32–37; b) J. B. Harper,
M. N. Kobrak, Mini-Rev. Org. Chem. 2006, 3, 253–269.
[7] A. E. Visser, R. P. Swatloski, W. M. Reichert, R. Mayton, S. Sheff, A. Wierz-
bicki, J. H. Davis, Jr., R. D. Rogers, Chem. Commun. 2001, 135–136.
[8] G. Chatel, J. F. B. Pereira, V. Debbeti, H. Wang, R. D. Rogers, Green Chem.
2014, 16, 2051–2083.
[9] M. T. Clough, C. R. Crick, J. Grasvik, P. A. Hunt, H. Niedermeyer, T. Welton,
O. P. Whitaker, Chem. Sci. 2015, 6, 1101–1114.
[10] a) B. Y. W. Man, J. M. Hook, J. B. Harper, Tetrahedron Lett. 2005, 46,
7641–7645; b) H. M. Yau, S. A. Barnes, J. M. Hook, T. G. A. Youngs, A. K.
Croft, J. B. Harper, Chem. Commun. 2008, 3576–3578; c) H. M. Yau, S. J.
Chan, S. R. D. George, J. M. Hook, A. K. Croft, J. B. Harper, Molecules
2009, 14, 2521–2534; d) H. M. Yau, A. G. Howe, J. M. Hook, A. K. Croft,
J. B. Harper, Org. Biomol. Chem. 2009, 7, 3572–3575; e) S. R. D. George,
G. L. Edwards, J. B. Harper, Org. Biomol. Chem. 2010, 8, 5354–5358;
Kinetic analyses
1
The H NMR and 19F NMR kinetic experiments were performed on
either a Bruker Advance III 400, Bruker Advance III 500 or a Bruker
Advance III 600 spectrometer with a TBI, BBO or BBFO probe. The
results were shown to be reproducible between the different spec-
trometers and probes. Kinetic analyses for the reaction described
in Scheme 3 were performed in solutions containing fluorodinitro-
benzene 16 (ꢀ0.015 molLÀ1), triethylamine (ꢀ0.35 molLÀ1) and
ethanol (ꢀ2 molLÀ1) at a given temperature in the range 304–
334 K. For the reaction outlined in Scheme 4, solutions were made
up containing benzyl bromide 18 (ꢀ0.04 molLÀ1) and pyridine
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
582
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
f) S. G. Jones, H. M. Yau, E. Davies, J. M. Hook, T. G. A. Youngs, J. B.
Harper, A. K. Croft, Phys. Chem. Chem. Phys. 2010, 12, 1873–1878;
g) H. M. Yau, A. K. Croft, J. B. Harper, Faraday Discuss. 2012, 154, 365–
371; h) S. T. Keaveney, J. B. Harper, RSC Adv. 2013, 3, 15698–15704;
i) E. E. L. Tanner, R. R. Hawker, H. M. Yau, A. K. Croft, J. B. Harper, Org.
Biomol. Chem. 2013, 11, 7516–7521; j) E. E. L. Tanner, H. M. Yau, R. R.
Hawker, A. K. Croft, J. B. Harper, Org. Biomol. Chem. 2013, 11, 6170–
6175; k) H. M. Yau, S. T. Keaveney, B. J. Butler, E. E. L. Tanner, M. S. Guerry,
S. R. D. George, M. H. Dunn, A. K. Croft, J. B. Harper, Pure Appl. Chem.
2013, 85, 1979–1990; l) S. T. Keaveney, K. S. McHale Schaffarczyk, R. S.
Haines, J. B. Harper, Org. Biomol. Chem. 2014, 12, 7092–7099; m) S. T.
Keaveney, D. V. Francis, W. Cao, R. S. Haines, J. B. Harper, Aust. J. Chem.
2015, 68, 31–35; n) S. T. Keaveney, R. S. Haines, J. B. Harper, Org. Biomol.
Chem. 2015, 13, 3771–3780; o) S. T. Keaveney, R. S. Haines, J. B. Harper,
Org. Biomol. Chem. 2015, 13, 8925–8936; p) S. T. Keaveney, R. S. Haines,
J. B. Harper in Encyclopedia of Physical Organic Chemistry (Ed.: U. Wille),
Wiley, Chichester, UK, 2016, in press; q) S. T. Keaveney, B. P. White, R. S.
Haines, J. B. Harper, Org. Biomol. Chem. 2016, 14, 2572–2580.
[11] R. R. Hawker, R. S. Haines, J. B. Harper, Targets Heterocycl. Syst. 2014, 18,
141–213.
[22] P. R. Gifford, J. B. Palmisano, J. Electrochem. Soc. 1987, 134, 610–614.
[23] L. Aldous, A. Khan, M. M. Hossain, C. Zhao in Catalysis in Ionic Liquids:
From Catalyst Synthesis to Application (Eds.: C. Hardacre, V. Parvulescu),
RSC Publishing, Cambridge, 2014.
[24] M. C. Kroon, W. Buijs, C. J. Peters, G.-J. Witkamp, Green Chem. 2006, 8,
241–245.
[25] M. Orsinia, I. Chiarotto, M. N. Elinsonc, G. Sotgiua, A. Inesi, Electrochem.
Commun. 2009, 11, 1013–1017.
[26] M. C. Buzzeo, C. Hardacre, R. G. Compton, ChemPhysChem 2006, 7,
176–180.
[27] A. M. O’Mahony, D. S. Silvester, L. Aldous, C. Hardacre, R. G. Compton, J.
Chem. Eng. Data 2008, 53, 2884–2891.
[28] S. P. Ong, O. Andreussi, Y. Wu, N. Marzari, G. Ceder, Chem. Mater. 2011,
23, 2979–2986.
[29] C. A. Paddon, F. L. Bhatti, T. J. Donohoe, R. G. Compton, J. Phys. Org.
Chem. 2007, 20, 115–121.
[30] G. Liu, M. Chockalingham, S. M. Khor, A. L. Gui, J. J. Gooding, Electro-
analysis 2010, 22, 918–926.
[31] V. Jouikov, J. Simonet, Langmuir 2012, 28, 931–938.
[32] a) E. Markevich, R. Sharabi, V. Borgel, H. Gottlieb, G. Salitra, D. Aurbach,
G. Semrau, M. A. Schmidt, Electrochim. Acta 2010, 55, 2687–2696; b) R.
Wibowo, L. Aldous, R. M. J. Jacobs, N. S. A. Manan, R. G. Compton,
Chem. Phys. Lett. 2011, 517, 103–107; c) N.-C. Lo, H.-Y. Chen, W.-J.
Chuang, C.-Y. Lu, P.-Y. Chen, P.-Y. Chen, J. Phys. Chem. B 2013, 117,
13899–13905; d) P. C. Howlett, E. I. Izgorodina, M. Forsyth, D. R. MacFar-
lane, Z. Phys. Chem. (München Ger.) 2006, 220, 1483.
[33] K. J. Walst, R. Yunis, P. M. Bayley, D. R. MacFarlane, C. J. Ward, R. Wang,
O. J. Curnow, RSC Adv. 2015, 5, 39565–39579.
[34] a) J. Ghilane, G. Trippe-Allard, J. C. Lacroix, Electrochem. Commun. 2013,
27, 73–76; b) M. Opallo, A. Lesniewski, J. Electroanal. Chem. 2011, 656,
2–16; c) C. Liang, J.-F. Huang, Z. Li, H. Luo, S. Dai, Eur. J. Org. Chem.
2006, 586–589.
[12] V. Hathwar, S. M. Roopan, R. Subashini, F. N. Khan, T. N. Guru Row, J.
Chem. Sci. 2010, 122, 677–685.
[13] a) J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D. Willauer, G. A.
Broker, R. D. Rogers, Green Chem. 2001, 3, 156–164; b) K. M. Hindi, T. J.
Siciliano, S. Durmus, M. J. Panzner, D. A. Medvetz, D. V. Reddy, L. A.
Hogue, C. E. Hovis, J. K. Hilliard, R. J. Mallet, C. A. Tessier, C. L. Cannon,
W. J. Youngs, J. Med. Chem. 2008, 51, 1577–1583.
[14] H. Tokuda, S. Tsuzuki, M. A. B. H. Susan, K. Hayamizu, M. Watanabe, J.
Phys. Chem. B 2006, 110, 19593–19600.
[15] J. E. Huheey, E. A. Keiter, R. L. Keiter, O. K. Medhi, Inorganic Chemistry:
Principles of Structure and Reactivity, Pearson Education, London, 2006.
[16] a) G. McHale, C. Hardacre, R. Ge, N. Doy, R. W. K. Allen, J. M. MacInnes,
M. R. Bown, M. I. Newton, Anal. Chem. 2008, 80, 5806–5811; b) S. Katsu-
ta, Y. Shiozawa, K. Imai, Y. Kudo, Y. Takeda, J. Chem. Eng. Data 2010, 55,
1588–1593.
[35] N. Menschutkin, Z. Phys. Chem. 1890, 5, 589.
[36] M. B. Smith, J. March, March’s Advanced Organic Chemistry: Reactions,
Mechanisms, and Structure, 7th ed., Vol. 7, John Wiley & Sons, Hoboken,
New Jersey, 2013.
[17] F. F. C. Bazito, Y. Kawano, R. M. Torresi, Electrochim. Acta 2007, 52, 6427–
6437.
[37] H. Eyring, J. Chem. Phys. 1935, 3, 107–115.
[18] Z. J. Chen, J.-M. Lee, J. Phys. Chem. B 2014, 118, 2712–2718.
[19] E. I. Izgorodina, R. Maganti, V. Armel, P. M. Dean, J. M. Pringle, K. R.
Seddon, D. R. MacFarlane, J. Phys. Chem. B 2011, 115, 14688–14697.
[20] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Wiley-
VCH Verlag GmbH & Co. KGaA, Weinheim, 2004, pp. 57–91.
[21] C. G. Hanke, A. Johansson, J. B. Harper, R. M. Lynden-Bell, Chem. Phys.
Lett. 2003, 374, 85–90.
Manuscript received: March 6, 2016
Revised: April 6, 2016
Accepted Article published: April 28, 2016
Final Article published: May 27, 2016
ChemPlusChem 2016, 81, 574 – 583
www.chempluschem.org
583
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim