Buffered chlorogallate(iii) ionic liquids and electrodeposition of gallium films

Article abstract of DOI:10.1039/c3cp44294b

Buffering of Lewis acidic chlorometallate ionic liquids is a useful tool to modify their properties for electrochemical and catalytic applications. Lewis acidic chlorogallate(iii) ionic liquids containing the 1-octyl-3- methylimidazolium cation, buffered with sodium chloride, were studied using ⁷¹Ga NMR spectroscopy and cyclic voltammetry. All the studied Lewis acidic compositions (0.50 < χ_GaCl3 ≤ 0.75) could be buffered to mild or moderate acidity, but not to neutrality. Electrodeposition of gallium from such buffered systems was possible, yielding deposits of improved morphology over the unbuffered ionic liquids, due to the constant melt composition maintained by the buffer. These findings were in a stark contrast with older studies on chloroaluminate(iii) ionic liquids buffered with sodium chloride.

Full text of DOI:10.1039/c3cp44294b

PCCP
View Article Online
View Journal | View Issue
PAPER
Buffered chlorogallate(III) ionic liquids and
electrodeposition of gallium films
Cite this: Phys. Chem. Chem. Phys., 2013,
1
5, 4518
a
a
a
Kenneth R. Seddon, Geetha Srinivasan,* Małgorzata Swad ´z ba-Kwa ´s ny and
b
Anthony R. Wilson
Buffering of Lewis acidic chlorometallate ionic liquids is a useful tool to modify their properties for
electrochemical and catalytic applications. Lewis acidic chlorogallate(III) ionic liquids containing the
71
1
-octyl-3-methylimidazolium cation, buffered with sodium chloride, were studied using Ga NMR
Received 30th November 2012,
Accepted 28th January 2013
spectroscopy and cyclic voltammetry. All the studied Lewis acidic compositions (0.50 o wGaCl3 r 0.75)
could be buffered to mild or moderate acidity, but not to neutrality. Electrodeposition of gallium from
such buffered systems was possible, yielding deposits of improved morphology over the unbuffered
ionic liquids, due to the constant melt composition maintained by the buffer. These findings were in a
stark contrast with older studies on chloroaluminate(III) ionic liquids buffered with sodium chloride.
DOI: 10.1039/c3cp44294b
www.rsc.org/pccp
1
8
influence of the cation. It was shown that Lewis acidity of
Introduction
the chlorometallate(III) ionic liquids is quantifiable using
1
2
3
19
Group 13 metal(III) chlorides (aluminium, gallium or indium ) the Gutmann Acceptor Number, and the Lewis acidity of the
react with various organic chlorides to form chlorometallate(III) wAlCl3 = 0.67 composition may be almost as high as that of
2
0
ionic liquids. Anionic speciation in such systems depends both trifluorosulfonic acid. Also, the electrochemical window of
on the metal and on the reagent ratio, expressed as the nominal chloroaluminates(III) is strongly composition-dependent: the
mol fraction of metal chloride, wMCl3. Due to the alternation effect, widest for the neutral systems (4 V), and much narrower for
4
21
properties do not change monotonically down the Group 13.
the basic (2.8 V) and acidic r ´e gimes (2.4 V). The exact neutral
Chloroaluminate(III) ionic liquids were the first group of composition is difficult to achieve, since any chemical or
ionic liquids to attract the broad attention of the scientific electrochemical processes that consume or generate acidic or
5
22
community. Being good solvents for a range of inorganic basic species will displace the ionic liquid from neutrality.
materials and organometallic complexes, they have found However, it was discovered that chloroaluminate(III) ionic
6
,7
8–10
many applications in electrochemistry
as Lewis acidic catalysts, co-catalysts and solvents in a number MCl, or strong organic bases (e.g. pyridine). For example,
of reactions, particularly for Friedel–Crafts catalysis).
when Lewis acidic chloroaluminate(III) systems were buffered to
Speciation of chloroaluminates(III) is rather complex, and neutral with NaCl, any deviation from neutrality due to added
has been studied using Al NMR spectroscopy, vibrational or electrogenerated acidic/basic agents was hindered. At the
spectroscopy,
and catalysis
liquids could be buffered using Group 1 metal chlorides,
2
1
23
(
2
7
11
21
1
2,13
14
15
FAB mass spectrometry, potentiometry,
same time, the buffered Lewis acidic chloroaluminates(III)
1
6
24
and vapour pressure measurements. Depending on the com- possess chemical behaviour termed ‘latent acidity’ – when a
position (wAlCl3), chloroaluminates(III) were proven to contain weak base L is added, an adduct with aluminium(III) chloride is
three sequential equilibrated anions from the following list: formed, and the solid MCl precipitates, according to eqn (1).
À
À
À
À
À 16
13
Cl , [AlCl ] , [Al Cl ] , [Al Cl ] and [Al Cl ] . Conse-
4
2
7
3
10
4
+
À
quently, they may be Lewis basic due to the presence of free
L + M + [AlCl ] " [L-AlCl ] + MCl(s)
(1)
4
3
chloride (wAlCl3 o 0.5), neutral (wAlCl3 = 0.5) or acidic due to the
1
7
presence of polynuclear anions (wAlCl3 > 0.5), with negligible The latent acidity of such systems is also quantifiable by the
1
9
Gutmann Acceptor Number, being dependent on the buffer
2
5
used, and always exhibiting a lower Lewis acidity than the
unbuffered composition. Buffered systems were used success-
fully as solvents and co-catalysts for the oligomerisation of
a
b
QUILL, The Queen’s University of Belfast, Belfast, BT9 5AG, UK.
E-mail: g.srinivasan@qub.ac.uk
Astron Advanced Materials Ltd, 2 Newlands Way, Broadstone,
Dorset. BH18 9JH, UK
2
3
1-alkenes, where moderate Lewis acidity is an advantage.
4
518 Phys. Chem. Chem. Phys., 2013, 15, 4518--4526
This journal is c the Owner Societies 2013

View Article Online
Paper
PCCP
Many successful applications of chloroaluminates(III) encouraged buffering of room temperature chlorogallate(III) ionic liquids,
the search for other chlorometallate ionic liquids. Chloro- comparing with the behaviour of the analogous chloro-
2
gallate(III) systems, firstly reported in 1987, were studied aluminate(III) systems.
2
6
27
electrochemically, and by DSC studies, Raman spectroscopy
2
8
and FAB mass spectrometry. More recent research, aside from
Experimental
Materials and equipment
2
9
sparse physicochemical studies, involved Raman scattering
3
0
71
combined with pyknometric studies, and Ga NMR spectro-
3
1
scopy combined with EXAFS. Based on these publications, 1-Chlorooctane (99%), Ga(NO ) Á6H O (99.999%), sodium
3
3
2
chlorogallates(III) appear to have speciation analogous to that of chloride (99.999%; anhydrous beads), and gallium(III) chloride
chloroaluminates(III), with the main equilibria in the system (99.999+%; anhydrous beads; ampoules sealed under argon)
given in eqn (2) and (3).
were purchased from Sigma-Aldrich and used without further
purification. 1-Methylimidazole (Sigma-Aldrich) was redistilled
immediately prior to use from solid sodium hydroxide. Dry
ethanenitrile was obtained from the MBraun MB-SPS solvent
À
À
À
2[GaCl
4
]
" [Ga
2
Cl
7
]
+ Cl
+ [GaCl ]
(2)
(3)
À
À
À
4
2[Ga
2
Cl
7
]
" [Ga
3
Cl10
]
purification system. 1-Butyl-3-methylimidazolium nitrate,
In contrast to the chloroaluminates(III), chlorogallate(III)
systems form homogenous room-temperature ionic liquids
for compositions between wGaCl3 = 0.67 and wGaCl3 = 0.75. It is
46
[
C mim][NO ], was prepared by a standard literature method.
4
3
An MBraun Labmaster DP glovebox was used for moisture-
free synthesis and storage of the chlorogallate(III) ionic liquids
À
known that the formation of [Ga Cl ] is thermodynamically
1
2
7
(
o0.1 ppm H
2 2
O, O ). H NMR spectra were acquired using
À
À1
more favourable than the formation of [Al
2
Cl
7
] ; enthalpies of
71
Bruker DRX 300 spectrometer, whereas the Ga NMR ones
were collected with a Bruker DRX 500 spectrometer. All spectra
were collected at 27 1C. Cyclic voltammograms were recorded
with a PC-controlled Autolab Type III Potentiostat. Sodium
analysis of each buffered chlorogallate(III) ionic liquid was
the reaction shown in eqn (4) are À164.8 kJ mol for M = Ga(III)
À1
20
and only À140.6 kJ mol for M = Al(III).
À
À
7
[
MCl4] + [MCl ] " [M Cl ]
(4)
3
2
Although such data are not available in the literature for performed on a Perkin-Elmer Analyst 400 atomic absorption
trimeric and larger species, the existence of homogenous ionic spectrometer operated in the atomic emission mode.
liquids for w
> 0.67 indicates that the formation of
The scanning electron microscopy (SEM) studies on the
GaCl3
À
[
M Cl ] species is much more favoured for the gallium(III) electrodeposits were carried out using a JEOL 6500F Field
3
10
systems than for the aluminium(III) systems.
Emission Scanning Electron Microscope, and the energy dispersive
Compared to chloroaluminates(III), chlorogallates(III) may X-Ray analysis (EDX) was carried out using Oxford Instruments
2
0
provide equal or even stronger Lewis acidity and have a INCA systems.
higher moisture stability, but their applications are relatively
Synthesis of 1-octyl-3-methylimidazolium chloride, [C mim]Cl
8
poorly explored. In catalysis, neutral compositions were used
3
2
3
for palladium-catalysed hydroethoxycarbonylation and acetal Dry ethanenitrile (100 cm ) was poured into a two-necked
3
3
formation,
Brønsted superacidic system catalysed arene round-bottomed flask (equipped with a stirring bar), fitted with
3
4
carbonylation, and the Lewis acidic compositions were used a reflux condenser attached to a dinitrogen line. 1-Methyl-
for oligomerisation of alkenes to lubricant base oils. Recently, imidazole (1.000 eq., 53.40 g, 0.6504 mol) was added under a
an ionic liquid-based synthesis of a new related inorganic positive pressure of dinitrogen, which was then bubbled
3
5
3
6
compound, [Bi
3
GaS
5
]
2
[Ga
3
Cl10
]
2
[GaCl
4
]
2
ÁS
8
, was reported.
through the solution for 30 min. Subsequently, a pressure-
3
Electrodeposition of gallium from chlorogallate(III) and from equalising dropping funnel (250 cm ) was mounted on the side
3
7
chlorometallate systems has been described. Flamini et al. neck of the round-bottomed flask through which 1-chloro-
studied the electrodeposition of gallium from aqueous octane (1.350 eq., 130.7 g, 0.8791 mol) was added dropwise over
solutions, where dihydrogen evolution was observed which 130 min to the vigorously stirred solution at room temperature.
38
can result in rough and non-uniform deposits. Electrochemistry An excess of 1-chlorooctane was used to guarantee the consump-
of a gallium(III) chloride solution in high-temperature molten tion of 1-methylimidazole. Subsequently, the flask was wrapped
salts was also investigated, and the redox properties were in aluminium foil, and the reaction mixture was stirred at room
3
9–41
studied.
Smolenskii reported briefly the electrodeposition temperature for 24 h. The reaction temperature was then
of gallium on tungsten from 1-ethyl-3-methylimidazolium increased to 60 1C, and the reaction was allowed to proceed
42
1
bis(trifluoromethylsulfonyl)amide. Electrochemistry of galium(III) for 9 d, until H NMR spectra failed to detect 1-methylimidazole.
chloride in air-stable ionic liquids, using in situ STM on the
Au(111) electrode were carried out by Endres and co-workers,
The reaction mixture was allowed to cool to room tempera-
ture, and washed with diethyl ether (10 Â 100 cm ) to remove
4
3
3
and nanodeposition of gallium on Au(111) was reported by the excess of 1-chlorooctane; the diethyl ether was then
4
4
Freyland et al.
removed under reduced presure using a rotary evaporator
Aiming at further studies of electrodeposition of gallium (1 h, 50 1C). Subsequently, the ionic liquid was dried under
metal, we have studied buffering of the chlorogallate(III) ionic high vacuum (7 d, 65 1C) to yield a pale yellow, viscous liquid
4
5
liquids. We report here a detailed study focussed on the (95.9%, 144 g) and stored in the glovebox. This synthesis
This journal is c the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 4518--4526 4519

View Article Online
PCCP
Paper
4
7
71
method is slightly different from existing in literature as it
eliminates any contamination from unreacted 1-methylimid-
Ga NMR spectroscopy
69
71
Gallium naturally occurs as two isotopes, Ga and Ga
azole and thereby reduces its influence on electrochemical
processes. H NMR (CDCl , 300 MHz, 27 1C): d/ppm = 10.52
s, 1H, N-CH-N); 7.84 (s, 1H, CH -N-CH-CH); 7.60 (s, 1H, CH -N-
69
71
(relative abundance Ga – 60.1%, Ga – 39.9%), both with a
1
71
3
nuclear spin equal to 3/2. The Ga isotope was chosen for the
NMR spectroscopy experiments due to the higher relative
intensity of the signal (0.024, as opposed to 0.007 for Ga).
(
3
3
CH-CH); 4.34 (t, 2H, J = 7.5 Hz, N-CH
2
); 4.15 (s, 3H, N-CH
); 1.25 (m, 10H, (CH -CH
3
).
3
); 1.92
69
48
(
(
t, J = 6.9 Hz, 2H, N-CH
t, J = 6.3 Hz, 3H, (CH
2
-CH
-CH
2
2
)
5
3
), 0.87
Neat ionic liquids were placed in borosilicate NMR tubes
5 mm diameter), with an external reference (saturated
2
)
5
(
Ga(NO ) solutions in D O, 27 1C) in a sealed capillary tube.
3
3
2
All experiments were performed using a Bruker DRX 500
spectrometer (152.53 MHz, 27 1C).
Synthesis and buffering of [C mim]Cl–GaCl ionic liquids
8
3
A range of compositions (0.33 r wGaCl3 r 0.75) (see Table 1) of
chlorogallate(III) ionic liquids based on the 1-octyl-3-methyl-
imidazolium cation was prepared on a 4 g scale.
Electrochemical experiments
Voltammetric experiments were carried out using 4 g of ionic
liquid in a three-electrode arrangement with platinum (1.6 mm
diameter) as the working electrode, a bright platinum coil as
the counter electrode, and all potentials measured with respect
In the glovebox, an appropriate amount of [C mim]Cl was
8
3
weighed into a sample vial (10 cm ) equipped with a stirring
bar. Leaving the vial on the balance (Æ0.0002 g), an appropriate
amount of gallium(III) chloride was added slowly and carefully
to achieve the desired composition. An exothermic reaction
always occurred. When the addition was completed, the sample
vial was closed with a cap, and the mixture was stirred
vigorously for 1 h; homogenous, transparent ionic liquids were
obtained in all cases. Samples were stored in the glovebox prior
to study.
+
to a Ag /Ag reference (a silver wire immersed in 0.01 M AgNO3
4 3
solution in [C mim][NO ], and separated from the bulk
solution via a glass frit). Voltammograms were reproducible
by the use of this reference electrode and no ionic leakage
detected as demonstrated by mutiple experiments. The IR-drop
was uncompensated. The platinum (Pt – BASi, US) electrode
was polished using an alumina slurry (Kemet, UK) of decreas-
ing particle sizes (6.0–0.1 mm) on soft lapping pads, washed
with distilled water, sonicated for 10 min, and dried with
compressed air prior to each experiment. Electrodeposition
experiments were carried out on glassy carbon substrates of
The buffered ionic liquids were prepared by adding a small
molar excess of dry sodium chloride, in very small portions, to
the vigorously stirred chlorogallate(III) ionic liquid. After each
addition, if no undissolved sodium chloride was visible, more
was added, and the stirring was continued until saturation was
achieved, whence the mixtures were stirred vigorously (24 h, 60 1C).
Neutralisation was ensured by noting the reduction of the
2
surface area 1 cm (Alfa Aesar). All experiments were conducted
inside the glovebox.
À
voltammetric reduction current of [Ga Cl ] . The presence of
2
7
undissolved sodium chloride was considered as evidence for
saturation. The ionic liquids were filtered through fine-fritted
syringe filter to remove the excess of sodium chloride, and the
clear liquid was analysed for sodium.
Results and discussion
Synthesis
The precursor of the chlorogallate(III) ionic liquid, [C mim]Cl,
8
was prepared by a standard methodology, and was pure by
1
H NMR spectroscopy and elemental analysis. The absence of
Table 1 Weights of reactants for the synthesis of [C
8
mim]Cl–GaCl
3
ionic liquids,
free amine was assured by using an excess of chloroalkane for
synthesis. Water content, as determined by cyclic voltammetry,
was less than 10 ppm. For all compositions of the chloro-
and the mass of the added NaCl
Mass/g
w
GaCl3
[C
8
mim]Cl
GaCl
3
NaCl
gallate(III) systems (0.33 r w
GaCl3
r 0.75), transparent, pale yellow
to yellow, very fluid room temperature ionic liquids were obtained.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.3300
.4500
.4600
.4700
.4800
.4900
.5000
.5100
.5200
.5300
.5500
.5800
.6000
.6300
.6700
.7000
.7200
.7500
2.9074
2.4627
2.4243
2.3858
2.3470
2.3081
2.2689
2.2295
2.1899
2.1501
2.1101
1.9509
1.8653
1.7429
1.5692
1.4416
1.3533
1.2162
1.0926
1.5373
1.5757
1.6142
1.6530
1.6919
1.7311
1.7705
1.8101
1.8499
1.8899
2.0491
2.1347
2.2571
2.4308
2.5584
2.6467
2.7838
0.3804
0.5523
0.5668 Buffering
0.5712
0.5836
Buffering of the chlorogallate(III) ionic liquids was ensured by
0.5900 preparing saturated solutions of sodium chloride for a range of
compositions (0.33 r wGaCl3 r 0.75); saturation was guaranteed
when undissolved sodium chloride could be observed in a
0.6304 given solution.
Recently, Welton and co-workers re-emphasised that salts
dissolved in ionic liquids form randomly dispersed mixtures of
0.7625 anions and cations, which should not be treated in terms
of traditional solute–solvent relationships (as for molecular
solvent), but rather as new materials incorporating multiple
0.9520 types of cations and anions. This is in agreement with the
0.6126
0.6121
0.6223
0.6465
0.7009
0.7212
49
0.8303
0.8738
0.9040
4
520 Phys. Chem. Chem. Phys., 2013, 15, 4518--4526
This journal is c the Owner Societies 2013

View Article Online
Paper
PCCP
For 0.50 o wGaCl3 r 0.60, the solubility of sodium chloride
increased rapidly, to produce effective compositions of values
close to neutral, by analogy with chloroaluminate(III) systems in
2
1
the same composition range. Although the solubility of
sodium chloride increased further for wGaCl3 = 0.67, the effective
wGaCl3 (X) did not decrease to neutrality, but reached X D 0.55.
Finally, for wGaCl3 = 0.75, the solubility of sodium chloride
decreased compared to wGaCl3 = 0.67, and the X remained
strongly acidic, as shown in Fig. 1 and Table 2.
One limitation of the study of the chloroaluminate(III)
system was the impossibility to obtain a clear liquid upon
2
1
buffering beyond wAlCl3 = 0.60. In contrast, the analogous
chlorogallate(III) system formed clear homogenous liquids up
to wGaCl3 = 0.75, permitting a much more extensive study of the
acidic r ´e gime. Of interest, according to the detailed phase
1
6
Fig. 1 Solubility of sodium chloride in various compositions of the [C
8
mim]Cl– diagram of the [C
mim]Cl–AlCl
AlCl3 = 0.60 is the most acidic composition where only [AlCl ]
3
system published by Øye et al.,
2
GaCl
3
system,
, and the achieved effective compositions, m, plotted as a
À
w
4
function of the unbuffered (starting) composition at 80 1C. The insert shows
the same solubility data, but with the y-axis (solubility) plotted on a logarithmic
scale, which allows for the observation of solubility the trend observed for the
basic compositions at 80 1C.
À
and [Al
2
À
Cl
7
]
are present, with no measurable amounts of
[
Al Cl ] yet in the system.
3 10
Analysing the solubility pattern in terms of the buffering
reaction presented in eqn (5), two factors must be considered:
the ability of the chlorogallate(III) ionic liquid to distrupt the
sodium chloride crystal lattice to produce sodium and chloride
ions, and the Lewis acidity of the heptachlorodigallate(III) anion.
original studies on the buffered chloroaluminate(III) system by
2
1,50
Wilkes and co-workers,
who treated such buffered ionic
liquids as ternary systems, e.g. [C mim]Cl–AlCl –NaCl, where
2
3
À
À
+
NaCl(s) + [Ga
2
Cl
7
]
" 2[GaCl
4
]
+ Na
(5)
reactive dissolution leads to formation of complex anions, and
effectively new ionic liquid phases. Therefore, in this work, the
The increase in the solubility of NaCl from wGaCl3 = 0.50 to
wGaCl3 r 0.60 can be easily understood in terms of the increasing
solubility of sodium chloride in the [C
expressed in terms of mole equivalents of NaCl in moles
equivalents of [C mim]Cl–GaCl composition (see Fig. 1 and
8 3
mim]Cl–GaCl system is
À
amounts of [Ga Cl ] , available for the reaction given in eqn (5).
2
7
8
3
The subsequent solubility decrease may be related to the for-
mation of the decachlorotrigallate(III) anion, which would have a
much lower ability to distrupt/solvate the sodium chloride lattice.
To confirm the buffering effect of the sodium chloride
addition, i.e. the expected change in the anionic speciation
corresponding to the effective wGaCl3 values, selected composi-
tions of the [C mim]Cl–GaCl system (buffered and unbuffered)
Table 2), to enhance better understanding of the underpinning
reactive dissolution.
Upon addition of a small molar excess (Na : Ga) of sodium
chloride to a basic or neutral chlorogallate(III) ionic liquid,
no visible dissolution was observed (Fig. 1 and Table 2).
8
3
7
1
were studied using Ga NMR spectroscopy (Fig. 2).
a
Table 2 Solubility of sodium chloride in different compositions of chlorogallate
For the unbuffered compositions, Fig. 2(c)–(h), the chemical
ionic liquid
7
1
shifts of the Ga NMR signals did not vary significantly, but the
peak widths changed dramatically with wGaCl3, which is in
agreement with the prior studies of the chlorometallate(III)
Solubility of NaCl/
(mol NaCl)(mol IL)
Electrochemical
window/V
À1
w
GaCl3
X
GaCl3
À3
31
0.3300
0.4000
0.4500
0.4700
0.4800
0.5000
0.5100
0.5200
0.5500
0.5800
0.6000
0.6300
0.6700
0.7000
0.7200
0.7500
1.60 Â 10
0.3296
0.3999
0.4499
0.4699
0.4799
0.4999
0.5081
0.5008
0.4947
0.5089
0.4867
0.5139
0.5524
0.6069
0.6501
0.6981
2.4
2.4
2.43
2.41
2.41
3.7
—
—
—
—
—
ionic liquids. These changes were related to the three main
À4
6.01 Â 10
À4
factors: viscosity of the medium, symmetry of the species, and
rapid exchange between anionic species. The sharpest peak was
found for the neutral sample, where a highly symmetrical
4.86 Â 10
À4
5.59 Â 10
À4
3.83 Â 10
À4
4.55 Â 10
À
31
À3
[GaCl
was in accordance to the chemical shift found for this anion in
4
] is the only anion. The chemical shift (B273 ppm)
7.64 Â 10
À2
7.98 Â 10
4
8
0.25
0.33
0.58
0.61
0.65
0.51
0.38
0.30
other solvents (e.g. 240 to 275 ppm in water). Viscosity, and
hence the linewidth of the Ga NMR signals, increased with
71
—
—
—
—
increasing basicity, due to increasing amounts of the chloride
anion in the samples. On the other hand, a rapid broadening
(and eventual disappearance) of the signal on the acidic side
—
could not be accounted solely for the viscosity changes, but it
À
resulted predominantly from [GaCl
4
]
remaining in dynamic
a
Sodium chloride solubility was determined from the sodium concen-
equilibria with chlorogallate(III) species of lower symmetry, like
trations in the buffered solutions, by atomic absorption spectroscopy in
21
À
À
emission mode, using the method of Wilkes and coworkers.
[Ga Cl ] and [Ga Cl10] , as shown in eqn (3).
2 7 3
This journal is c the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 4518--4526 4521

View Article Online
PCCP
Paper
71
Fig. 2
Ga NMR stacked spectra (117.6 MHz, neat liquid, 27 1C) of the
system. Chemical shift, d/ppm, is presented as a function of
the samples composition, wGaCl3: (a) 0.67 (buffered), (b) 0.55 (buffered), (c) 0.55,
d) 0.51, (e) 0.50, (f) 0.49, (g) 0.48 and (h) 0.47.
8 3
[C mim]Cl–GaCl
(
7
1
Ga NMR spectra of buffered samples are shown in Fig. 2(a)
and (b). For the buffered wGaCl3 = 0.67 composition, Fig. 2(a), the
71
Fig. 3
Ga NMR stacked spectra (117.6 MHz, neat liquid, 27 1C) of
effective wGaCl3 value was calculated to be around X = 0.55 chlorogallate(III) systems: (a) unbuffered, wGaCl = 0.50, (b) buffered, wGaCl3
=
3
0
.67 and (c) unbuffered, wGaCl = 0.67.
(
acidic), and a ‘flat’ spectrum was obtained, as expected for
strongly acidic chlorogallate(III) ionic liquid. For the buffered
= 0.55 composition, Fig. 2(b), with the effective w
3
w
GaCl3
GaCl3
7
1
value calculated to be X = 0.4947, a sharp Ga NMR signal, current was chosen to be 0.1 mA. The anodic and cathodic
characteristic of such a nearly-neutral composition, was potential limits observed in pure ionic liquids are due to the
expected. Nevertheless, the obtained broad signal was more oxidation and reduction of their anionic and cationic compo-
1
,52
similar to that of the slightly acidic wGaCl3 = 0.51 composition, nents, respectively.
Linear sweep DC cyclic voltammetry is
Fig. 2(d), than to the sharp peak found in the neutral sample, the simplest and the most commonly used technique to deter-
5
1
Fig. 2(e). Worth mentioning, an identical spectrum has been mine the electrochemical window.
obtained for a wGaCl3 = 0.55 sample that has been stirred with
The electrochemical responses of the unbuffered chloro-
NaCl for a period of one week. This discrepancy between the gallate(III) systems (basic, neutral and acidic), obtained at a
7
1
calculated wGaCl eff value and the Ga NMR spectrum, obviously platinum electrode, are illustrated in Fig. 4. The potentials were
3
À
indicating the presence of a small amount of [Ga
indicates, that in the ternary [C mim]Cl–GaCl
the equilibrium in the eqn (2) is shifted slightly to the right. In anodic direction up to +2 V. Fig. 4(a) shows the cyclic voltamo-
2
Cl
7
]
scanned initially towards the negative direction down to À2 V
8
3
–NaCl system, from the open circuit potential and then reversed towards the
À
À
other words, unlike for the binary [C
8
mim]Cl–GaCl
3
system, in gram for w
GaCl3 4
= 0.47, where two anions, Cl and [GaCl ] , are
31
the ‘nearly neutral’ X = 0.495 composition, there are significant present. Under these basic conditions, the cathodic background
À
À
7
amounts of Cl and [Ga Cl ] anions present.
process is the reduction of the 1-octyl-3-methylimidazolium
2
5
3,54
21
Apparently, sodium chloride has a buffering effect on the cation.
acidic chlorogallate(III) ionic liquids. The buffering does not
lead to neutral ionic liquids, but to compositions of lower predominant anion, and the concentrations of Cl and [Ga Cl ]
acidity. Consequently, electrodeposition of gallium metal from are undetectably low. In such a neutral environment, reduction
such buffered systems should be possible. To test this hypoth- of the cation remained as the primary cathodic process – see
esis, three chlorogallate(III) systems were selected for the elec- Fig. 4(b). However, with no free chloride present, the anodic
trodeposition studies: unbuffered wGaCl3 = 0.50, as well as limit here was shifted towards the more difficult oxidation of
In this system, chloride oxidation is the anodic limit.
À
For the neutral (wGaCl3 = 0.50) composition, [GaCl ] is the
4
À
À
2
7
3
1
7
1
À
buffered and unbuffered wGaCl3 = 0.67. Ga NMR spectra of [GaCl ] . The electrochemical window for the neutral system,
4
these samples are compared in Fig. 3.
shown in Fig. 4(b) is 3.7 V, which is in agreement with findings
of Gale and co-workers (3.6 V), who justified this wide electro-
2
Cyclic voltammetry
chemical window for an exactly neutral [C mim]Cl–GaCl system
4
3
À
The electrochemical window is the potential difference between in terms of absence of either Cl to shorten the anodic limit, or
À
the oxidation and reduction potentials of an electrolyte used in [Ga
a voltammetric experiment, where the cut-off limiting current is
2 7
Cl ]
to shorten the cathodic limit.
Fig. 4(c) shows the cyclic voltammetric response at the
6
,51
arbitrarily chosen.
In the present work, the cut-off limiting platinum electrode from an acidic composition (wGaCl3 = 0.55),
4
522 Phys. Chem. Chem. Phys., 2013, 15, 4518--4526
This journal is c the Owner Societies 2013

View Article Online
Paper
PCCP
Fig. 5 Cyclic voltammetry of unbuffered and buffered [C
8
mim]Cl–GaCl
3
systems
À1
at a platinum electrode, 50 mV s , 20 1C; (a) unbuffered wGaCl = 0.51;
3
(
b) buffered wGaCl = 0.51.
3
The redox peaks associated with the reduction of Ga(III) to Ga(0)
and with its oxidation back to Ga(I) and then to Ga(III) were still
À
present. This suggests that the buffering reaction of [Ga Cl ]
2
7
with the excess of NaCl, given with eqn (5), does not lead to
À
complete regeneration of the neutral [GaCl
4
]
but, in contrast
to the chloroaluminate(III) systems, NaCl buffers chloro-
gallates(III) to milder acidity.
Fig. 4 Cyclic voltammetry of unbuffered [C
8
mim]Cl–GaCl
3
systems at a platinum
À1
electrode, 50 mV s , 20 1C; (a) wGaCl = 0.47; (b) wGaCl = 0.50 and (c) wGaCl = 0.55.
3
3
3
Fig. 5 shows the cyclic voltammetry of the wGaCl3 = 0.51
composition before buffering, Fig. 5(a), and after buffering,
are expected to be the predomi- Fig. 5(b). In the buffered system, although the peak intensities
À À
where [GaCl ] and [Ga Cl ]
nant anions.
4 2 7
2
6–31
In this case, the cathodic limit (À0.5 V) are reduced drastically, the gallium reduction around À0.5 V is
À
corresponds to the reduction of [Ga
2
Cl
7
]
to deposit gallium not eliminated in the cathodic region and the subsequent
upon the electrode at a more positive potential than that oxidation peaks for Ga(I) and Ga(III) in the anodic region are
necessary to reduce the imidazolium cation. The anodic also present clearly, showing that the system is not buffered to
processes are the oxidation of Ga(0) to Ga(I) at ca. À0.37 V, absolute neutrality, as shown in the inset, but to a lower acidity.
4
0
and Ga(I) to Ga(III) at ca. 1.1 V.
Whereas, in chloroaluminate systems, it was impossible to
While dichlorine was generated at the anodic limits in all maintain a liquid buffered system for acidic compositions
5
5
compositions, it was produced at different potentials and in
various amounts, as it was sourced from different electroche-
mical oxidations.
It has been demonstrated that the electrochemical windows
of chlorogallate(III) systems vary with composition, by analogy
to the behaviour observed for chloroaluminate(III) ionic
2
1
liquids, with the widest window identified for the neutral
composition. Following this analogy, chlorogallates(III) were
expected to behave similarly to chloroaluminates(III) when
buffered with anhydrous sodium chloride.
Buffering with sodium chloride left the electrochemical
windows of 0.33 r wGaCl3 r 0.50 compositions unaffected
(see Table 2), as its solubility is too small to change the ionic
liquid composition significantly.
For the acidic compositions, a significant dissolution of
sodium chloride was followed by the observable change in
the electrochemical window. Interestingly, the buffered acidic
compositions of chlorogallate(III) ionic liquids were not
completely neutralised, i.e. the expected wide electrochemical
window of the exactly neutral composition was not achieved.
Fig. 6 Cyclic voltammetry of unbuffered and buffered [C
8
mim]Cl–GaCl
3
systems
À1
on platinum electrode, 50 mV s , 20 1C; (a) unbuffered wGaCl = 0.67; (b) buffered
3
w
GaCl = 0.67 and (c) unbuffered wGaCl = 0.50.
3
3
This journal is c the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 4518--4526 4523

View Article Online
PCCP
Paper
2
1
higher than wAlCl3 = 0.60, the highly acidic chlorogallate stripping of gallium(0). However, the cyclic voltammogram was
system viz. wGaCl3 = 0.67 composition could be buffered not typical for metal deposition/stripping. A classical deposi-
to milder acidity; Fig. 6 shows the cyclic voltammetry of the tion/stripping pattern, observed for example when aluminium
3
7
wGaCl3 = 0.67 composition before buffering, Fig. 6(a), and after wire in chlorogallate(III) was used as the reference electrode,
buffering, Fig. 6(b), compared with the unbuffered, neutral involves a sharp reduction of Ga(III) to Ga(0), followed by
(w
GaCl3 = 0.50) chlorogallate(III) system, Fig. 6(c).
The electrochemical behaviour of the unbuffered wGaCl3
immediate sharp oxidations. In Fig. 6(a), a broad reduction
wave, Ga(III) to Ga(0), and two broad oxidation waves, Ga(0) to
=
0
.67 system, Fig. 6(a), clearly demonstrated the deposition and Ga(I), followed by Ga(I) to Ga(III), are observed. The second
anodic wave, for Ga(I) to Ga(III), was complex, as the oxidation
kinetics were faster.
After buffering, Fig. 6(b), a drastic reduction (but not elim-
ination) in the current for the peaks corresponding to the
deposition and stripping of gallium(0) was observed. The cyclic
voltammogram was similar to that of the unbuffered wGaCl3
=
0.55 composition, but not to that of the neutral chlorogallate(III)
ionic liquid, Fig. 6(c). Changes in temperature, or prolonged
stirring with sodium chloride (even several weeks), had no
effect on this behaviour of the buffered acidic system.
In the cyclic voltammetry of the buffered wGaCl3 = 0.67
composition, Fig. 6(b), the gallium reduction/deposition is
strongly shifted into the cathodic direction, by 0.5 V from that
observed in the unbuffered system, Fig. 6(a), but the complete
elimination of electro-reduction of gallium was not observed.
Fig. 7(b) shows the electrochemical behaviour of the buffered
w
GaCl3 = 0.67 system compared with the electrochemical beha-
Fig. 7 Cyclic voltammetry of unbuffered and buffered [C
8
mim]Cl–GaCl
3
systems
À1
viour of unbuffered wGaCl3 = 0.55 system, Fig. 7(a). A similar
on platinum electrode, 50 mV s , 20 1C; (a) unbuffered wGaCl = 0.55, (b) buffered
3
w
GaCl = 0.67.
pattern of redox peaks in the cyclic voltammetry was observed,
3
Fig. 8 Scanning electron microscopy (SEM) images and energy-dispersive X-ray spectroscopy (EDX) analysis of the gallium deposits, obtained on glassy carbon
substrate from chlorogallate(III) systems; (a) unbuffered wGaCl = 0.50; (b) unbuffered wGaCl = 0.67; (c) buffered wGaCl = 0.67 and (d) EDX spectrum of the deposits.
3
3
3
4
524 Phys. Chem. Chem. Phys., 2013, 15, 4518--4526
This journal is c the Owner Societies 2013

View Article Online
Paper
PCCP
which complements the results from sodium chloride solubility
studies (Table 2).
From these buffered systems, it was possible to electro-
deposit metallic gallium, which was subsequently studied
In order to compare the morphology of gallium deposited using SEM and EDX. The morphlogy of the deposit from the
from various compositions, electrodeposition experiments on buffered system was more uniform than that obtained from the
glassy carbon substrates were carried out by holding the unbuffered one, due to stable anionic speciation of the former
+
potential at À0.5 V vs. Ag/Ag for a period of 15 min at 20 1C system, maintained by the buffer.
(
Fig. 8). Deposition experiments were repeated under similar
These findings are in stark contrast with the behaviour of
2
1,50
conditions in triplicate and the results were reproducible.
chloroaluminate(III) ionic liquids,
which can be buffered to
Depositions from buffered, Fig. 8(c), and from unbuffered neutrality with sodium chloride, but only up to wAlCl3 = 0.60
wGaCl3 = 0.67 chlorogallate(III) ionic liquids, Fig. 8(b), were composition, and display only latent acidity when buffered
studied, and compared with the deposition from an unbuffered (i.e. it is not possible to electrodeposit aluminium from such
2
1
wGaCl3 = 0.50 composition, Fig. 8(a). The deposit from the systems).
unbuffered wGaCl3 = 0.67 composition was visibly denser than
that from the unbuffered system (N.B. for chloroaluminates(III),
Notes and references
no aluminium metal electrodeposition was observed from the
2
1
buffered wAlCl3 = 0.55 composition, here or elsewhere). The
uniform morphology of the deposit obtained from the buffered
system can be attributed to the constant anionic speciation of
the ionic liquid, maintained by the buffer.
1 J. S. Wilkes, J. A. Levisky, R. A. Wilson and C. L. Hussey,
Inorg. Chem., 1982, 21, 1263–1264.
2 S. P. Wicelinski, R. J. Gale and J. S. Wilkes, J. Electrochem.
Soc., 1987, 134, 262–263.
The deposits were identified as gallium metal using EDX
analysis, Fig. 8(d). Elemental gallium was observed as an
enhanced peak, along with traces of oxygen and carbon. The
peak for oxygen was present due to the thin film of oxide
formed on the surface of the gallium, while the traces of carbon
originated from the glassy carbon substrate.
3 M. K. Carpenter and M. W. Verbrugge, J. Mater. Res., 1994, 9,
2584–2588.
4 A. Earnshaw and N. N. Greenwood, Chemistry of the
Elements, Butterworth-Heinemann, Oxford, 2nd edn, 1997.
5 C. L. Hussey, Adv. Molten Salt Chem., 1983, 5, 185–229.
6 H. Ohno, Electrochemical Aspects of Ionic Liquids, John Wiley
& Sons, Inc., Hoboken, New Jersey, 2nd edn, 2011.
7 F. Endres, ChemPhysChem, 2002, 3, 145–154.
8 Ionic Liquids in Synthesis, ed. P. Wasserschied and
T. Welton, Wiley Interscience, Weinheim, 2nd edn, 2007.
9 H. Olivier-Bourbigou, L. Magna and D. Morvan, Appl. Catal.,
A, 2010, 373, 1–56.
Combination of the results from the electrodeposition experi-
7
1
ments (Fig. 8) and the Ga NMR spectroscopy study (Fig. 2) gave
a new insight to electrochemical properties of chlorogallate(III)
ionic liquids. For both chlorometallate(III) systems, sodium
chloride is only marginally soluble in wMCl3 r 0.50 ionic liquids,
as there is no reaction which would drive passing of the salt
À
into the ionic liquid. When [M
2
Cl
7
]
appears in the system 10 J. P. Hallett and T. Welton, Chem. Rev., 2011, 111,
3508–3576.
solubility, and yields buffered systems. Since the stability of 11 (a) J. L. Gray and G. E. Maciel, J. Am. Chem. Soc., 1981, 103,
(wMCl3 > 0.50), the reaction given by eqn (5) drives the increased
À
À
[
Ga Cl ] is higher than that of [Al Cl ] , the equilibrium of this
7147–7151; (b) S. Takahashi, M.-L. Saboungi, R. J. Klingler,
M. J. Chen and J. W. Rathke, J. Chem. Soc., Faraday Trans.,
1993, 89, 3591–3595; (c) K. Ichikawa and T. Matsumoto,
J. Magn. Reson., 1985, 63, 445–453.
2
7
2
7
reaction is shifted more to the left for the former system, hence
lowering the solubility of sodium chloride, resulting in incom-
plete buffering. Finally, for wMCl3 > 0.60, when polynuclear
species appear, the situation becomes less trivial, with solidifica- 12 (a) R. J. Gale, B. Gilbert and R. A. Osteryoung, Inorg. Chem.,
tion of chloroaluminate(III) systems, and an unexpected decrease
in the solubility of sodium chloride in the chlorogallate(III) ionic
liquids. Of importance, this finding is in contrast to the prevail-
1978, 17, 2728–2729; (b) S. Takahashi, L. A. Curtiss,
D. Gosztola, N. Koura and M.-L. Saboungi, Inorg. Chem.,
1995, 34, 2990–2993.
ing assumptions present in the literature, where it is common to 13 C. J. Dymek, J. S. Wilkes, M.-A. Einarsrud and H. A. Øye,
perceive the electrochemical properties of chloroaluminate(III)
and chlorogallate(III) systems as analogous.
Polyhedron, 1988, 7, 1139–1145.
14 A. K. Abdul-Sada, A. M. Greenway, K. R. Seddon and
T. Welton, Org. Mass Spectrom., 1993, 28, 759–765.
3
1
1
5 C. L. Hussey, T. B. Scheffler, J. S. Wilkes and A. A. Fannin,
Jr., J. Electrochem. Soc., 1985, 133, 1389–1391.
Conclusions
Buffering of a range of compositions of a chlorogallate(III) 16 H. A. Øye, M. Jagatoyen, T. Oksefjel and J. S. Wilkes, Molten
system, [C
gated. It was demonstrated by Ga NMR spectroscopy and
cyclic voltammetry that 0.50 o w r 0.60 compositions,
8
mim]Cl–GaCl
3
, with sodium chloride, was investi-
Salt Chemistry and Technology, in Materials Science Forum,
ed. M. Chemla and D. Devilliers, Trans. Techn. Publ.
Aldermannsdorf, Switzerland, 1991, vol. 73–75, pp. 183–189.
7
1
GaCl3
although nominally containing enough sodium chloride to 17 T. Tsuda and C. L. Hussey, Electrochemistry of Room-
buffer them to neutrality, in reality display mild acidity.
Furthermore, strongly acidic samples (wGaCl3 = 0.67 and 0.75)
can be buffered to moderate acidity.
Temperature Ionic Liquids and Melts, in Modern Aspects of
Electrochemistry, ed. R. E. White, Springer, Dordrecht, 2009,
vol. 45, pp. 63–174.
This journal is c the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 4518--4526 4525

View Article Online
PCCP
Paper
1
1
8 R. A. Mantz, P. C. Trulove, R. T. Carlin, T. L. Theim and
R. A. Osteryoung, Inorg. Chem., 1997, 36, 1227–1232.
9 (a) V. Gutmann, The Donor-Acceptor Approach to Molecular
Interactions, Plenum Press, New York, 1978; (b) T. A.
Zawodzinski, Jr. and R. A. Osteryoung, Inorg. Chem., 1989,
(c) Y. Yoshida, O. Akihiro, S. Gunzi, N. Seiichi, N. Eiji,
T. Masaki, S. Makoto, T. Masahide and Y. Toshinobu, Bull.
Chem. Soc. Jpn., 2005, 78, 1921–1928; (d) P.-Yu. Chen,
Y.-F. Lin and I.-W. Sun, J. Electrochem. Soc., 1999, 146,
3290–3294.
2
8, 1710–1715.
38 (a) D. O. Flamini, S. B. Saidman and J. B. Bessone, J. Appl.
Electrochem., 2007, 37, 467–471; (b) D. O. Flamini,
S. B. Saidman and J. B. Bessone, Thin Solid Films, 2007,
515, 7880–7885.
39 J. M. Shafir and J. A. Plambeck, Can. J. Chem., 1970, 48,
2131–2132.
2
2
2
2
2
0 J. Estager, A. A. Oliferenko, K. R. Seddon and M. Swad ´z ba-
Kwa ´s ny, Dalton Trans., 2010, 39, 11375–11382.
1 T. J. Melton, J. Joyce, J. T. Maloy, J. A. Boon and J. S. Wilkes,
J. Electrochem. Soc., 1990, 137, 3865–3869.
2 M. Lipstajn and R. A. Osteryoung, J. Electrochem. Soc., 1983,
1
30, 1968–1969.
3 B. Ellis, W. Keim and P. Wasserscheid, Chem. Commun.,
999, 337–338.
4 (a) I. C. Quarmby and R. A. Osteryoung, J. Am. Chem. Soc.,
40 S. P. Yatsenko, V. N. Danilin and L. N. Volkova, Zh. Fiz.
Khim., 1968, 42, 874–877.
41 U. Anders and J. A. Plambeck, Can. J. Chem., 1969, 47,
3055–3060.
1
1994, 116, 2649–2650; (b) R. A. Osteryoung, I. C. Quarmby, 42 V. V. Smolenskii, A. L. Bove, A. A. Khokhryakov and A. G.
R. A. Mantz and L. M. Goldenberg, Anal. Chem., 1994, 66,
558–3561.
5 P. Koronaios, D. King and R. A. Osteryoung, Inorg. Chem.,
998, 37, 2028–2032.
6 S. P. Wicelinski, R. J. Gale and J. S. Wilkes, Thermochim. 44 G. B. Pan, O. Mann and W. Freyland, J. Phys. Chem. C, 2011,
Acta, 1988, 126, 255–263. 115, 7656–7659.
7 P. Wicelinski, R. J. Gale, S. D. Williams and G. Mamantov, 45 K. Seddon, G. Srinivasan and A. Wilson, Electrodeposition of
Spectrochim. Acta, Part A, 1989, 45, 759–762.
8 S. P. Wicelinski, R. J. Gale, K. M. Pamidimukkala and
R. A. Lane, Anal. Chem., 1988, 60, 2228–2232.
Osipenko, Radiochemistry, 2003, 45, 449–452.
3
43 L. H. S. Gasparotto, N. Borisenko, O. H ¨o fft, R. Al-Salmana,
W. Maus-Friedrichs, N. Bocchi, S. Zein El Abedin and
F. Endres, Electrochim. Acta, 2009, 55, 218–226.
2
2
2
2
2
1
elemental zirconium from a mixture of a Lewis acid and and an
ionic liquid, World Pat., WO2010136814, 2010.
46 A. E. Bradley, C. Hardacre, M. Nieuwenhuyzen, W. R. Pitner,
D. Sanders, K. R. Seddon and R. C. Thied, Inorg. Chem.,
2004, 43, 2503–2514.
9 (a) J. Tong, Q.-S. Liu, W. Guan and J.-Z. Yang, J. Phys. Chem.
B, 2007, 111, 3197–3200; (b) W. G. Xu, X.-M. Lu, J.-S. Gui,
Q.-G. Zhang and J.-Z. Yang, Chin. J. Chem., 2006, 24, 47 J. G. Huddleston, A. E. Visser, W. M. Reichert,
3
31–335; (c) W. Martino, J. Fernandez de la Mora,
Y. Yoshida, G. Saito and J. Wilkes, Green Chem., 2006, 8,
90–397.
H. D. Willauer, G. A. Broker and R. D. Rogers, Green Chem.,
2001, 3, 156–164.
48 J. Mason, Multinuclear NMR, Plenum Press, New York, 1987.
3
3
3
3
3
3
3
3
3
0 H. Z. Yang, Y. Jin, W.-G. Xu, Q.-G. Zhang and S.-L. Zang, 49 M. Lui, L. Crowhurst, J. Hallett, P. Hunt, H. Niedermeyer
Fluid Phase Equilib., 2005, 227, 41–46. and T. Welton, Chem. Sci., 2011, 2, 1491–1496.
1 C. Hardacre, R. W. Murphy, K. R. Seddon, G. Srinivasan and 50 A. M. Elias and J. S. Wilkes, J. Chem. Eng. Data, 1994, 39, 79–82.
M. Swad ´z ba-Kwa ´s ny, Aust. J. Chem., 2010, 63, 845–848.
2 G. Rangits and L. Kollar, J. Mol. Catal. A: Chem., 2006, 246,
51 A. J. Bard and L. R. Faulkner, Electrochemical Methods:
Fundamentals and Applications, John Wiley & Sons, New
York, 1980.
59–64.
3 Y. J. Kim and R. S. Varma, Tetrahedron Lett., 2005, 46, 52 M. Ue, A. Murakami and S. Nakamura, J. Electrochem. Soc.,
447–7456. 2002, 149, A1572–A1577.
4 E. J. Angueira and M. G. White, J. Mol. Catal. A: Chem., 2007, 53 P. Bonh ˆo te, A.-P. Dias, N. Papageorgiou, K. Kalyanasundaram
77, 164–170.
and M. Gr ¨a tzel, Inorg. Chem., 1996, 35, 1168–1178.
5 M. P. Atkins, K. R. Seddon and M. Swadzba-Kwasny, Pure 54 L. E. Barrosse-Antle, A. M. Bond, R. G. Compton and
Appl. Chem., 2011, 83, 1391–1406. A. M. O’Mahony, Chem.–Asian J., 2010, 5, 202–230.
6 D. Freudenmann and C. Feldmann, Dalton Trans., 2011, 40, 55 T. Schubert, S. Z. E. Abedin, A. P. Abbott, K. J. Mckenzie,
7
2
4
52–456.
K. S. Ryder and F. Endres, Electrodeposition of metals, in
Electrodeposition from Ionic Liquids, ed. F. Endres,
A. P. Abbott and D. R. MacFarlane, Wiley-VCH, Weinheim,
2008, p. 105.
7 (a) M. K. Carpenter and M. W. Verbrugge, J. Electrochem.
Soc., 1990, 137, 123–129; (b) M. W. Verbrugge and
M. K. Carpenter, AIChE J., 1990, 36, 1097–1106;
4
526 Phys. Chem. Chem. Phys., 2013, 15, 4518--4526
This journal is c the Owner Societies 2013