Electrochemical Applications Of Room Temperature Ionic Liquids
Electrochemical Applications of
Room-Temperature Ionic Liquids
by Tetsuya Tsuda and Charles L. Hussey
There is good evidence that the first room-temperature Unfortunately for electrochemists and despite these
ionic liquid or RTIL was identified and characterized
opportunities for structural engineering, there are only
by Walden in 1914, who examined the electrical
a limited number of RTILs that are really suitable for
conductivity of ethylammonium nitrate.1 Thus, it would
electrochemistry. This is because most exhibit low electrical
not be an exaggeration to say that the very beginnings of
conductivity (< 5 mS cm-1) and high viscosity (> 50 cP).
electrochemistry involved room-temperature ionic liquids!
In some cases, RTILs that do have sufficient conductivity
Although once considered to be a curious subset of the
for electrochemical applications exhibit meager
general class of substances called “molten salts,” RTILs have
electrochemical windows and/or show limited chemical
taken on a life of their own and now occupy the attention
stability toward reactive materials and ions such as Li, H- or
of scientists worldwide. In fact, scientific interest in these
O -2. Thus, it is sometimes difficult to identify RTILs having
interesting materials continues to grow exponentially.
all of the properties necessary to guarantee the success of a
Today a literature search with SciFinder Scholar (Chemical
specific application. Relatively speaking, the combination
Abstracts Online) employing the keywords ionic liquid
of imidazolium- or sulfonium-based cations and complex
or room-temperature ionic liquid produces hundreds
halide anions such AlCl -
-
-
-
-
4 , BF4 , CF3CO2 , CF3SO3 , NTf2 ,
of articles, whereas just ten years ago a search would
N(SO
-
-
2F)2 , or F(HF)2.3 results in RTILs with good working
produce only a handful of papers. From the viewpoint of
conductivities. From the relationship known as Walden’s
electrochemistry, RTILs are under investigation as solvents
rule, we know that the conductivity of an electrolyte is
for technological applications such as metal surface
approximately correlated with its viscosity. Thus, one
finishing, batteries, capacitors, fuel cells, electrosynthesis,
cannot expect viscous RTILs prepared from cations with
and nuclear waste treatment.
long-chain substituents to show appreciable conductivities,
The question as to what exactly constitutes an ionic
and this is universally the case. With regard to the
liquid or RTIL is discussed in the preceding article by
electrochemical window of RTILs, tertraalkylammonium- ,
Keith Johnson and will not be considered further herein.
dialkylpyrrolidinium-, and dialkylpiperidinium-based
Although a large number of different RTILs have been
ionic liquids typically exhibit superior electrochemical
reported in the literature, most can be classified into one of
stability relative to imidazolium-based RTILs, owing to
seven families on the basis of their cationic structures, as
their superior resistance toward reduction compared to
depicted in Fig. 1. Some of the common anions on which
cations based on aromatic heterocyclic species. This is true
RTILs are based are also given in this figure. One advantage
provided that the accompanying anions are not reduced
of having such a diverse selection of anions and cations
before the cations. But unfortunately, the conductivities of
is that it is possible to “engineer” the physicochemical
these systems are usually inferior to the imidazolium- and
properties of RTILs by the choice of the ionic constituents.
sulfonium-based RTILs, illustrating the trade-off between
For example, the combination of an imidazolium cation,
stability and favorable transport properties. There are
such as 1-ethyl-3-methylimidazolium ion with BF -
4 results
several excellent monographs2 and review articles3 that
in a hydrophilic RTIL, whereas the combination of
describe the synthesis, purification, physicochemical, and
this cation and N(SO
-
-
2CF3)2 (NTf2 ) produces a strongly
electrochemical properties of RTILs. In many cases, they
hydrophobic ionic liquid. Although other factors may
also provide guidance about the applications of various
play a role, the viscosity and conductivity of such ionic
RTILs to specific electrochemical problems.
liquids is largely a function of the chain length of the alkyl
Although “historical” RTILs, i.e., ionic liquids composed
substituents, a factor that also carries over to the other
of an anhydrous metal chloride combined with an organic
classes of cations in Fig. 1.
chloride salt, e.g., AlCl3–1-ethyl-3-methylimidazolium
Common Cations:
R2
R1
R1
R1
+
+ R3
N
N
N
+
R4
N R2
R3
R4
Tetraalkylammonium
Di, Tri, and Tetraalkylimidazolium
Alkylpyridinium
R1
+ R2
R1
R1
+ R2
N
N
+
R1
R4
P R2
+
S
R3
R3
R2
Dialkylpyrrolidinium
Dialkylpiperidinium
Tetraalkylphosphonium
Trialkylsulfonium
Common Anions:
BF -
-
-
-
-
-
-
-
-
-
-
-
4 , B(CN)4 , CH3BF3 , CH2CHBF3 , CF3BF3 , C2F5BF3 , n-C3F7BF3 , n-C4F9BF3 , PF6 , CF3CO2 , CF3SO3 , N(SO2CF3)2 ,
N(COCF
-
-
-
-
-
-
3)(SO2CF3)-, N(SO2F)2 , N(CN)2 , C(CN)3 , SCN-, SeCN-, CuCl2 , AlCl4 , F(HF)2.3 etc.
Fig. 1. Common cations and anions for room-temperature ionic liquids.
42
The Electrochemical Society Interface • Spring 2007
Electrochemical Investigations
(a)
of Inorganic Complexes
Although they are environmentally unstable and require
manipulation in an inert atmosphere glove box or sealed
cell, chloroaluminate RTILs have proven to be especially
versatile solvents for electrochemical studies of inorganic
chloride complexes and low-valent metal species. This is
because the Lewis acidity of these RTILs can be adjusted
by changing the AlCl3/RCl ratio, where RCl is the organic
chloride salt, e.g., EtMeImCl. Ionic liquids containing
a molar excess of AlCl3 are Lewis acidic because they
contain coordinately unsaturated species such as Al
-
2Cl7
that are strong chloride acceptors. On the other hand,
chloroaluminate RTILs containing a molar excess of RCl
are Lewis basic because they contain chloride that is not
bound to aluminum. Acidic chloroaluminate RTILs readily
dissolve and stabilize low oxidation state transition metal
species, e.g., polynuclear metal-metal bonded clusters
such as [Nb6Cl12]z+ and [Ta6Cl12]z+. They have proven
to be excellent, if not unique solvents for probing the
electrochemistry of these interesting inorganic species.6
Basic chloroaluminate RTILs support the formation
of anionic chloride complexes similar to those found
when metal ions are introduced into molten mixtures
(b)
of alkali chlorides such as LiCl–KCl. They have been
used as solvents to investigate the room-temperature
electrochemistry of a large number of transition metal and
main group chloride complexes of the type
[MpClq](q-pn)-, where M is the metal with an oxidation state
of n. An interesting example is the electrochemistry of
[Re2Cl9]- in Lewis basic 44.4–55.6 mol% AlCl3–EtMeImCl.7
Figure 2a shows a series of cyclic voltammograms for
[Re2Cl9]- that were recorded at a glassy carbon electrode in
this RTIL. Three reduction waves with approximate peak
potentials of 0.52, -0.16, and -0.61 V, respectively, appear
on the negative scan in the fresh solution, but only the
first and last waves exhibit an associated oxidation current.
The first wave corresponds to the one electron reversible
reduction of [Re2Cl9]- to [Re2Cl9]2-. The latter two reduction
waves were also present in the RTIL when it contained
only [Re2Cl9]2- and were found to arise from the chemically
irreversible reduction of this species to [Re2Cl8]2- and the
subsequent one-electron reversible reduction of [Re2Cl8]2-
to [Re2Cl8]3-. After a period of a few hours, the original
Fig. 2. (a) Cyclic voltammograms of 5.07 mmol L-1 [n-Bu4N][Re2Cl9]
waves for [Re2Cl9]- and its progeny begin to disappear,
in a Lewis basic 44.4–55.6 mol % AlCl3–EtMeImCl at a GC electrode:
and a new wave attributed to the reversible one-electron
(A) fresh solution; (B) after 70 min, (C) after 197 min, (D) solution in
reduction of [ReCl6]2- to [ReCl6]3- becomes evident. On the
C after being heated to 363 K for 60 min and cooled. The sweep rates
other hand, [Re
were 0.050 V s-l, and the temperatures were 301 K. (Reproduced from
2Cl9]2- disproportionates into a mixture
Ref. 7 with permission from The Electrochemical Society.) (b) Cyclic
of this monomeric species and dimeric [Re2Cl8]2-. In both
voltammogram of 5 mmol L-1 [n-Bu4N]4[S2W18O62] in n-BuMeImPF6
cases, the conversion process can be expedited by heating
at a GC electrode. The sweep rate was 0.10 V s-l, and the temperature
the solution. Combined with electronic spectroscopic
was 293 K. (Reproduced from Ref. 8 with permission from the American
data, these electrochemical investigations reveal a rich and
Chemical Society.)
diverse chemistry for [Re2Cl9]- in this chloride-rich RTIL as
summarized below:
chloride (EtMeImCl), AlCl3–1-(1-butyl)pyridinium chloride
+e
+e
+e
(BuPyCl), and CuCl–EtMeImCl, are highly sensitive
[Re2Cl9]-
[Re2Cl9]2-
[Re2Cl8]2-
[Re2Cl8]3-
to moisture, these RTILs are still the preferred choice
-e
-Cl-
-e
Green
Purple
Blu
B e-gre
lue-gr en
een
Blue
for applications involving metal surface finishing, e.g.,
electroplating. This is because it is much easier to dissolve
Δ +3Cl-
Δ +Cl-
+e
salts of the plating metals in these RTILs compared to the
2[ReCl6]2-
[ReCl6]2-
[ReCl6]3-
more inert fluoroanion-based systems such as EtMeImBF
-e
4,
Pa
P le-y
ale-yel o
ellow
+
n-BuMeImPF6, or EtMeImNTf2. In addition, it has been
½[Re
demonstrated many times that metallic films on the order
2Cl8]2-
of μm-thicknesses can be readily electrodeposited from
Non-chloroaluminate RTILs have also proven to be
these RTILs.4,5 On the other hand, there are technologies
valuable for exploring the electrochemistry of novel
that can take advantage of the chemical inertness of these
inorganic species. For example, tetrabutylammonium salts
fluoroanion-based systems. One example is the use of these
of the polyoxometallate anions [M6O19]2-, [a-SiM12O40]4- ,
liquids as electrolytes for Li batteries and low-temperature
and [a-S2M18O62]4-, where M = Mo or W, dissolve readily
fuel cells.
in n-BuMeImPF6 with heating.8 Figure 2b shows a cyclic
We describe below some selected research highlights
voltammogram recorded at a glassy carbon electrode
involving current electrochemical applications of
for a solution of the latter anionic species in this RTIL.
RTILs. We also provide literature citations to guide the
This voltammogram exhibits six well-defined reduction
interested reader.
(continued on next page)
The Electrochemical Society Interface • Spring 2007
43
waves on the negative scan, labeled I to VI in the original
requires an electrolyte solution with certain favorable
figure reproduced herein, and each reduction wave has
properties. Foremost, the polymer used in the actuator must
an associated oxidation wave. These waves were ascribed
be completely stable in the electrolyte solution during the
to the six reversible one-electron reactions given in the
oxidation/reduction cycle, and the solution must exhibit
scheme below.
reasonable conductivity and a large electrochemical window.
Also, in order to achieve fast switching speeds, the doping
[α-S
ions should have high ionic mobilities in the electrolyte.
2W18O62]4- +e [α-S2W18O62]5- +e [α-S2W18O62]6- +e [α-S2W18O62]7-
-e
-e
e-e
Clearly, environmentally stable RTILs are ideally suited
+e [α-S
for this application due to their chemical inertness, high
2W18O62]8- +e [α-S2W18O62]9- +e [α-S2W18O62]10-
-e
-e
-e
intrinsic conductivities, large electrochemical windows, and
negligible vapor pressures. Furthermore, such RTILs should
permit the operation of actuators at higher temperatures and
Electrochemical Mechanical Actuator Devices
larger voltages than is possible with conventional molecular
solvent/electrolyte systems.
There is considerable on-going interest in using
The application of RTILs as electrolytes for
conjugated polymers such as polyaniline (PANI), polypyrrole
electrochemical mechanical actuator devices has been
(PPy), and polythiophene (PT) to fabricate lightweight,
explored by several groups, with the first report appearing
highly efficient electrochemical mechanical actuator devices
in the journal Science in 2002.9 In this archetypical
or artificial muscles. This proposed application is based on
investigation, PANI and PPy were cycled in n-BuMeImBF4
the observation that these polymers physically deform when
or n-BuMeImPF6 under ambient conditions and exhibited
undergoing an oxidation/reduction cycle in an electrolyte
no failure after 1 million cycles. For PANI, the doping
solution. This deformation is due to the contraction/
mechanism involves the insertion/expulsion of n-BuMeIm+
expansion resulting from the reversible expulsion/
cations. However, for PPy, the nature of the doping
incorporation of electrolyte ions or solvent from the solution
mechanism was reported by the authors to be potential
(electrochemical doping). However, the construction of an
dependent, involving the insertion of n-BuMeIm+ cations
effective mechanical actuator device from these polymers
at low potentials and PF -
6 anions at high potentials. Since
the appearance of this original publication, several new
approaches and different materials have been proposed for
(a)
electrochemical actuators based on RTILs.10 But in each case,
the basic concept is the same in that the main components
are an electrochemically-active polymeric material and a
suitably conductive, but inert RTIL.
The instrumentation used to study the deformation of
polymer films during electrochemical doping in RTILs
is designed to examine both the strain and electrical
impedance of the actuator material. A state-of-the-art
experimental set-up used for this purpose is shown in
Fig. 3a.11 The experiments used to characterize the polymer
films consist of applying a random potential with a RMS
magnitude of ~1 V to the sample and measuring the
displacement of the sample as a function of frequency with
(b)
the laser vibrometer. By measuring the induced current
as a function of the frequency of the applied potential,
the impedance of the sample can also be determined. The
deflected shapes of model actuators prepared from Nafion
membranes painted with Au and/or RuO2 powder, soaked
with EtMeImTfO (TfO- = CF
-
3SO3 ), and hot pressed between
gold foils are shown in Fig. 3b.11 It can be seen in this figure
that the range of motion increases as the amplitude of the
square wave input is increased.
Figure 3c illustrates the basic structure of the device
proposed by Lu, et al.12 This device is fabricated from
(c)
PANI that is doped with trifluoromethane sulfonic acid.
The polymer has been formed into the shape of a fiber,
designated by the authors as a “panion triflate fiber.” Several
such fibers were wound into a yarn and threaded into a
hollow tube also made from PANI; this tube serves as the
counter electrode. An electrically insulating layer of non-
woven polyacrylonitrile nanofiber serves as the separator
and was soaked with a RTIL, in this case,
n-BuMeImBF4. With the application of a suitable potential,
this actuator generated a stress of 0.42 ~ 0.82 MPa, which
exceeds the stress generated by skeletal muscle (0.1 ~ 0.5
MPa). This example illustrates that the practical application
of such devices may not be far in the future.
Dye-Sensitized Photoelectrochemical Cells
Fig. 3. (a) Experimental apparatus for free deflection and impedance
Dye-sensitized electrochemical photovoltaic or
tests of polymer film electrochemical actuators. (b) Photographs of the
photoelectrochemical solar cells consist of a dye-sensitized film
deflected shapes as a function of amplitude for a square wave potential
input to a polymer film electrochemical actuator. (Reproduced from Ref.
of a mesoporous oxide semiconductor, e.g., TiO2, which has been
11 with permission from Elsevier B.V.) (c) A SEM image of a yarn-in-
deposited on a conductive transparent surface and coated with a
fiber electrochemical linear actuator fabricated by threading eight Panion
chemisorbed layer of a photoactive dye such as cis-RuL2(SCN)2,
triflate yarns into a Panion triflate hollow fiber. (Reproduced from Ref.
where L = 2,2'-bipyridyl-4,4'-dicarboxylate.13,14 The surface
12 with permission from CSIRO Publishing.)
44
The Electrochemical Society Interface • Spring 2007
(a)
(b)
(d)
(c)
Fig. 4. (a) Diagram showing the fabrication of an ion-gel sheet-type dye-sensitized solar cell. (b) 100 mm × 100 mm array of the dye-sensitized
solar cells. (Reproduced from Ref. 17 with permission from Elsevier B.V.) (c) Modified TiO
-
2 nanoparticles. (d) Diffusion of I3 through a matrix of (A)
modified and (B) unmodified TiO2 nanoparticles. (Reproduced from Ref. 18 with permission from The Electrochemical Society.)
of the semiconductor, which serves as the photoanode, is in
randomly as shown in “B” in Fig. 4d. In contrast, ordered
contact with a solution containing a redox mediator, usually
structures on the TiO2 resulting from modification with longer
I-/I -3. Upon exposure to light of a suitable wavelength, the dye
chain alkyl groups provide paths for rapid diffusion by the
injects electrons into the semiconductor, becoming oxidized.
mediator ions as depicted in “A” in Fig. 4d.
The oxidized dye layer accepts electrons from the mediator in
Overall, the conversion efficiency of dye-sensitized
the electrolyte, which is itself regenerated by electrons from the
photoelectrochemical solar cells is only about 50% of
external circuit provided by the cathode.14
that typically seen for conventional Si-based photovoltaic
Typically, volatile organic solvents such as acetonitrile are
cells. However, the former are much cheaper to produce.
used to prepare the electrolyte/mediator solution used in
Although the energy conversion efficiency for dye-sensitized
dye-sensitized photoelectrochemical cells. However, the high
photoelectrochemical cells based on RTILs is somewhat inferior
vapor pressure associated with conventional organic solvents,
to cells prepared with conventional organic solvent-based
especially under conditions where considerable solar heating
electrolytes, the benefits of replacing the volatile organic solvent
may take place, is problematic. As a result, much research has
component far outweigh this disadvantage. Do not be surprised
been devoted to identifying and characterizing RTILs that
to see commercially viable cells of this type that utilize RTILs in
can be used to replace the organic solvents normally used
some form or another in the near future.
in these photoelectrochemical cells.15 One of the first RTILs
to be investigated was n-HexMeImI (Hex = n-hexyl),16 but as
Electrochemical Supercapacitors
research in this area progressed, RTILs such as EtMeImNTf2
have been used successfully.17 By dissolving EtMeImNTf2 and
Electrochemical supercapacitors are simple energy storage
poly(vinylidenefluoride-co-hexafluoropropylene) together
devices with high rate charging-discharging capabilities and
in a volatile organic solvent, it is possible to prepare an ion-
high power density. The latter property greatly exceeds that
gel electrolyte that can be surface cast.17 Figure 4a shows the
of conventional electrochemical energy storage devices, e.g.,
construction of a typical ion-gel based cell using a cast polymer
batteries and fuel cells. Furthermore, supercapacitors can
electrolyte sheet. A 100 mm × 100 mm array of such cells is
store a much greater charge per unit volume of mass than
illustrated in Fig. 4b.
conventional dielectric-based capacitors. An electrochemical
Recently, Hayase, et al.18 have succeeded in preparing RTIL-
supercapacitor is based on the electrochemical double layer
based gel electrolytes by chemically modifying 21 nm TiO2
resulting from the electrostatic adsorption of ionic species
particles with various 1-alkyl-3-methylimidazolium salts via the
at the electrode-solution interface, i.e., no actual redox
formation of Ti-O-CO-R bonds, where
reaction is supposed to take place during the charging-
R = 1-alkyl-3-methyl-imidazolium bromide or iodide. The
discharging of these devices. To obtain the maximum
structures of the modified nanoparticles are illustrated in
possible capacitance, supercapacitor electrodes must have
Fig. 4c. Clay-like gel electrolytes synthesized from regular
a high surface area; the standard material used in these
TiO2 nanoparticles and RTILs generally lead to a decrease
devices is typically high surface area carbon. Because these
in photoelectrochemical performance, especially at higher
devices are based on the electrosorption of ionic species, the
nanoparticle content. Fortunately, this problem can be avoided
region between the electrodes of the capacitor must contain
using these modified TiO2 nanoparticles. The improved
an electrolyte with mobile ions. To obtain the maximum
performance seen in this case depended on the alkyl chain
operating voltage without solvent decomposition, it is
lengths. For shorter alkyl groups or the regular unmodified
necessary to use aprotic solvents such as acetonitrile.
nanoparticles, the mediator ions, I -3 and I- , likely diffuse
(continued on next page)
The Electrochemical Society Interface • Spring 2007
45
(a)
(b)
(c)
Fig. 5. (a) Schematic diagram of a pouch-type electrochemical double layer supercapacitor. (Reproduced from Ref. 20 with permission from Elsevier Ltd.)
(b) A commercialized pouch-type cell. (Reproduced from Japan Radio Co., Ltd. Website: http://www.jrc.co.jp/jp/whatsnew/20061018/index.html with
permission.) (c) A 2032 coin-type electrochemical double layer supercapacitor. (Reproduced from Ref. 21 with permission from The Electrochemical Society.)
Carbon-based supercapacitors based on conventional
have also found their way into hybrid supercapacitors.
aprotic electrolytes are commercially available.
In fact, a hybrid supercapacitor based on activated carbon,
Not surprisingly, because of the numerous favorable
poly(3-methylthiophene) and n-BuMePyrNTf2 (Pyr =
properties described above, RTILs are considered to be
pyrrolidinium) may be the first viable supercapacitor based
promising electrolytes for electrochemical supercapacitors.
on a RTIL, producing 24 Wh kg-1 and 14 kW kg-1.24
EtMeImBF4 and EtMeImNTf2 dissolved in alkyl carbonate
solvents were among the first RTILs to be investigated.19
Lithium-Ion Batteries
More recently, an electrochemical supercapacitor based
on N,N-diethyl-N-methyl-N-(2-methoxyethyl)ammonium
Lithium-ion batteries are now ubiquitous in society and
tetrafluoroborate was shown to have superior properties
serve as the power sources in almost all portable electronic
compared to supercapacitors based on conventional aprotic
devices that are marketed to today’s consumer. Articles
electrolytes such as mixtures of Et4NBF4 in propylene
devoted to or that discuss some aspects of lithium-ion
carbonate.20 A diagram of a commercial supercapacitor that
batteries have appeared in Interface on several occasions.25
utilizes this electrolyte is shown in Figs. 5a and 5b. This
Therefore, no effort is made herein to discuss either the
device is available commercially from Japan Radio Co., Ltd.
background or the electrochemistry of these devices. With
and has a rated potential of 3 V, a capacity of 1000 F, and a
such widespread use and in view of the safety issue of
power density of 13.5 kW L-1.
lithium-ion batteries, there are considerable ongoing efforts
A coin-type supercapacitor fabricated from the very
by battery manufacturers to improve the performance of
conductive fluoride-based RTIL, EtMeImF(HF)2.3, that
these devices. As a result of the many attractive aspects
utilizes carbon cloth electrodes is shown in Fig. 5c.21 In
of RTILs, there is a modest, but continuing interest in
spite of limitations imposed by the potential stability of
using them as electrolytes for these cells. Until several
this RTIL, this device was found to exhibit better energy
years ago, the RTIL electrolyte of choice was some variety
storage capabilities than supercapacitors based on aqueous
of chloroaluminate,26 but interest in the use of non-
H2SO4. A RTIL-based supercapacitor has even been prepared
chloroaluminate ionic liquids has gradually increased.
from carbon nanotubes. This device utilizes carbon gel
It is difficult to pinpoint the first instance where non-
electrodes fabricated by combining EtMeImNTf2 with
chloroaluminate RTILs were used in lithium-ion batteries.
pulverized single-walled carbon nanotubes.22
An early report describes a successful Li/LiMn2O4
Hybrid supercapacitors obtain energy storage from
cell prepared with 1,2-dimethyl-4-fluoropyrazolium
the electrostatic double layer capacitance obtained at a
tetrafluoroborate + LiBF4 or LiAsF6.27 Li/LiCoO2 cells
high surface area carbon electrode as discussed above
utilizing n-PrMePipNTf2 (Pip = piperidinium) show good
and from a rapid, reversible charge-transfer process that
cyclic efficiency,28,29 and it is clear that RTILs based on
occurs at a dopable conjugated polymer, e.g., poly(3-
those anions that offer good anodic stability, e.g., NTf -
2 or
methylthiophene).23 This charge-transfer process is
N(SO
-
2F)2 , give the best performance.29,30 At the present
designated as a pseudocapacitance. Not surprisingly, RTILs
time, the main problem is the incompatibility of the
46
The Electrochemical Society Interface • Spring 2007
anode, e.g., Li metal, and the RTILs. That is, the solid
A schematic of the process developed for the extraction of
electrolyte interphase film that is produced on the anode
Cs+ from tank waste utilizing BOBCalixC6 dissolved in n-
during the charge/discharge process is less stable than
Bu3MeNNTf2 is shown in Fig. 6.34 After extraction of Cs+ into
that obtained in conventional organic solvents. This
the hydrophobic RTIL, the extraction solvent is dried and the
incompatibility problem limits the cycling efficiency of
oxygen is removed by sparging with dry N2 or by heating.
the cell. MacFarlane, et al.31 have succeeded in elucidating
The complexed Cs+ is then selectively removed from the
the mechanism of film formation on Li in RTILs based on
extraction solvent by reduction at a Hg electrode. The Cs(Hg)
N-alkyl-N-methylpyrrolidinium ions and NTf -
2 .31 Perhaps,
produced during the last step can be transported to another
future research of this nature will lead to resolution
cell, where the Cs can be chemically or electrochemically
of this problem, enabling the practical use of RTILs as
stripped from the Hg into a minimum volume of aqueous
electrolytes in Li batteries.
solution. This proposed electrochemical process for
remediation of the extraction solvent preserves both the
Treatment of Nuclear Waste
ionophore and the RTIL.
When spent nuclear reactor fuel is reprocessed, large
Lab-on-a-Chip System
volumes of liquid waste are produced, called “tank
waste.” This liquid waste, which may be acidic or basic
Although it is not based on classical electrochemistry,
depending on the treatment process, contains many
a novel and noteworthy application of RTILs is their
long-lived radioactive products, including the -emitters
use by Dubois, et al.35 as solvents in electrowetting on
137Cs, 129I, 90Sr, and 99Tc, and a variety of transuranic
dielectric (EWOD) microreactors or lab-on-a-chip systems.
elements, which are normally a-emitters. The problems
Microreactors are devices for carrying out microscale
associated with this waste have been described previously
synthesis, and as such, are very useful for preparing small
in Interface.32 The 137Cs+ and 90Sr2+ in tank waste are
quantities of rare, expensive, or dangerous materials.
normally present at low concentrations in a large
In the case of the microreactor based on a RTIL, EWOD
volume of liquid, and any treatment process for this
actuation is used as a fluidic motor to move and combine
waste must therefore produce a significant reduction in
ionic liquid droplets containing the reagents of interest on
the volume of this liquid. A number of strategies have
two-dimensional, addressable chips by the application of a
been employed to selectively extract 137Cs+ and 90Sr2+
potential. Each segment of the chip is fabricated from gold
from aqueous tank waste with ionophores specific for
electrodes coated with a dielectric of Si3N and a hydrophobic
137Cs+ and 90Sr2+, such as calix[4]arene-bis(t-octylbenzo-
layer of Teflon (Fig. 7). The negligible vapor pressure of
crown-6), BOBCalixC6, and dicyclohexano-18-crown-6,
the RTIL that was employed by these workers makes the
respectively, dissolved in hydrophobic solvents, including
manipulation of small liquid volumes (< 1 μL) of reagents
hydrophobic RTILs.33 However, in order to formulate an
possible in an open system without complications due to
economically-viable treatment process based on RTILs, a
the loss of the solvent. The viability of this RTIL-based
process must be devised to recycle the expensive RTIL/
microreactor was demonstrated by carrying out a synthetic
ionophore mixtures. We found in our own work that
reaction involving multiple components that resulted in the
electrochemistry can play a useful role in this recycling
preparation of tetrahydroquinolines, i.e., Grieco’s reaction.
process.
(continued on next page)
a
s
t
e
u
l
a
t
e
d
t
a
n
k
w
i
m
6
p
h
a
s
e
|
S
a
l
i
x
C
C
B
O
T
I
L
+
B
R
Fig. 6. Proposed process for the extraction of cesium from aqueous tank waste using n-Bu
-
3MeNNTf2 + BOBCalixC6.
(Reproduced from Ref. 34 with permission from The Electrochemical Society.)
The Electrochemical Society Interface • Spring 2007
47
Fig. 7. (A) Ionic liquid drop on a dielectric electrode. (B) Chemical synthesis by merging ionic liquid droplets on an e-microreactor. (C) Left:
multiplexed chip for performing parallel synthesis. Right: fluidic processor for parallel syntheses: (a) electrode bus for communication; (b) reservoir
entry; (c) reagent storage and dispensing, and (d) reservoir for dispensing task-specific ionic liquid. (Reproduced from Ref. 35 with permission from the
American Chemical Society.)
(a)
(b)
(c)
Control of Experimental Conditions during Electrochemical
Experiments with RTILs
Although RTILs have many attractive features for
electrochemistry, they can be exceedingly difficult to purify.
The expectation is that RTILs will be clear, colorless liquids
with a wide, clean electrochemical window, as assessed with
voltammetry. Unlike conventional aprotic organic solvents, the
fact that RTILs exhibit virtually no vapor pressure eliminates
the most powerful purification technique: distillation. Thus,
the real business of purification must begin with the reagents
that are used in the preparation of the RTIL of interest. The
various procedures that are used to prepare and purify these
components are too numerous to list here, and they tend to be
specific for the RTIL that is being prepared. See for example the
recent paper by Appetecchi, et al.36 Articles appearing in the
literature wherein the procedures used to purify the starting
materials, to eliminate undesirable reaction byproducts, and
to assess the impurity levels of the resulting RTIL are absent or
unclear should be viewed with skepticism.
One particularly insidious contaminant is water. It is well
established that even small amounts of adventitious water can
strongly affect both the physicochemical and electrochemical
properties of RTILs. Figure 8 shows photographs of three
Pyrex test tubes containing different types of RTILs before
and after the intentional addition of water. Included in this
figure are a classical moisture-reactive ionic liquid, 60.0–40.0
mol % AlCl3–EtMeImCl; an air stable hydrophilic ionic
liquid, EtMeImF(HF)2.3; and a hydrophobic ionic liquid, n-
Fig. 8. Photographs of (a) 60.0–40.0 mol% AlCl3–EtMeImCl; (b)
Bu
EtMeImF(HF)
3MeNNTf2. The appearance of each RTIL before the addition
2.3; (c) n-Bu3MeNNTf2: (top) before the addition of water,
of water is shown in the top picture. The addition of water to an
and (bottom) 20 s after the addition of water.
open test tube of the Lewis acidic AlCl3–EtMeImCl results in a
violent reaction with the evolution of heat and HCl gas and an
In fact, the exposure of this supposedly hydrophobic liquid to
obvious change in the appearance of the RTIL (Fig. 8, bottom
atmospheric moisture for several days raises the water content
picture). (Please don’t try this test without using proper
from an initial value of ~ 4 ppm to 1000 ppm
safety precautions!) In contrast, the addition of water to
(= 5.55 × 10-2 mol kg-1)34! Fortunately, in this case, the water is
EtMeImF(HF)2.3 and n-Bu3MeNNTf2 results in no obvious
not strongly bound and can be removed with heating under
chemical change. To the eye, the former remains unchanged,
vacuum, restoring the RTIL to its initial condition. Thus,
whereas the latter exhibits a simple phase separation as expected
all RTILs, even those thought to be hydrophobic, should be
due to its strong hydrophobicity. However, in reality the
handled under an inert gas atmosphere or in a glove box if water
hydrophobic RTIL also dissolves a considerable quantity of water.
contamination is to be avoided.
48
The Electrochemical Society Interface • Spring 2007
Conclusion
Schmidt, S. M. Zakeeruddin, and M. Grätzel, J. Mater. Chem., 16,
2978 (2006); Z. Fei, D. Kuang, D. Zhao, C. Klein, W. H. Ang, S. M.
In this article, we have attempted to present a short overview
Zakeeruddin, M. Grätzel, and P. J. Dyson, Inorg. Chem., 10.1021/
of some of the many ways that RTILs have been applied to
ic061232n (published on Web).
electrochemical problems. A comprehensive review devoted to
16. N. Papageorgiou, Y. Athanassov, M. Armand, P. Bonhôte, H.
the use of RTILs as electrolytes for supercapacitors would alone
Pettersson, A. Azam, and M. Grätzel, J. Electrochem. Soc., 143, 3099
(1996).
occupy the entire space allotted for this article. Obviously, RTILs
17. H. Matsui, K. Okada, T. Kawashima, T. Ezure, N. Tanabe, R.
have great potential for electrochemical applications in science
Kawano, and M. Watanabe, J. Photochem. Photobiol., A, 164, 129
and technology. Some of the readers may hold the opinion that
(2004).
RTILs are “messy, intractable” solvents, i.e., hard to make and
18. T. Kato, T. Kado, S. Tanaka, A. Okazaki, and S. Hayase, J.
hard to purify, and to some extent this is true. But, there is no
Electrochem. Soc., 153, A626 (2006).
longer any doubt about the attractiveness of RTILs as solvents for
19. A. B. McEwen, S. F. McDevitt, and V. R. Koch, J. Electrochem. Soc.,
144
electrochemistry because some have excellent physicochemical
, L84 (1997); A. B. McEwen, H. L. Ngo, K. LeCompte, and J. L.
Goldman, J. Electrochem. Soc., 146, 1687 (1999).
properties for this purpose, including ample intrinsic
20. T. Sato, G. Masuda, and K. Takagi, Electrochim. Acta, 49, 3603
conductivities, wide electrochemical windows, and negligible
(2004).
vapor pressures. Among the various classes of known solvents,
21. M. Ue, M. Takeda, A. Toriumi, A. Kominato, R. Hagiwara, and Y.
only RTILs have this combination of desirable characteristics. n
Ito, J. Electrochem. Soc., 150, A499 (2003).
22. T. Katakabe, T. Kaneko, M. Watanabe, T. Fukushima, and T. Aida, J.
References
Electrochem. Soc., 152, A1913 (2005).
23. A. Laforgue, P. Simon, J. F. Fauvarque, M. Mastragostino, F.
1. P. Walden, Bulletin de l’Académie Impériale des Sciences de St.-
Soavi, J. F. Sarrau, P. Lailler, M. Conte, E. Rossi, and S. Saguatti, J.
Pétersbourg, 405 (1914).
Electrochem. Soc., 150, A645 (2003).
2. For example, Ionic Liquids in Synthesis, P. Wasserscheid and T.
24. A. Balducci, W. A. Henderson, M. Mastragostino, S. Passerini, P.
Welton, Editors, Wiley-VCH, Weinheim (2003); Electrochemical
Simon, and F. Soavi, Electrochim. Acta, 50, 2233 (2005).
Aspects of Ionic Liquids, H. Ohno, Editor, Wiley-Interscience, New
25. For example, S. Megahed and B. Scrosati, Interface, 4(4), 34 (1995);
Jersey (2005); Metal Catalysed Reactions in Ionic Liquids, P. J. Dyson
D. A. Scherson, Interface, 5(3), 34 (1996); R. J. Brodd, Interface, 8(3),
and T. J. Geldbach, Springer, The Netherlands (2006).
20 (1999); J. Dahn, J. Jiang, L. Moshurchak, C. Buhrmester, and R.
3. For example, S. Takahashi, N. Koura, S. Kohara, M.-L. Saboungi,
L. Wang, Interface, 14(4), 27 (2005).
and L. A. Curtiss, Plasmas & Ions, 2, 91 (1999); R. Hagiwara and Y.
26. For example, A. Webber and G. E. Blomgren, “Ionic Liquids for
Ito, J. Fluorine Chem., 105, 221 (2000); C. F. Poole, J. Chromatogr., A,
Lithium Ion and Related Batteries,” in Advances in Lithium-Ion
1037, 49 (2004); P. J. Scammells, J. L. Scott, and R. D. Singer, Aust.
Batteries, W. A. van Schalkwijk and B. Scrosati, Editors, p. 185,
J. Chem., 58, 155 (2005); J. Zhang and A. M. Bond, Analyst, 130,
Kluwer Academic / Plenum Publishers, New York (2002); J. S.
1132 (2005); H. Xue, R. Verma, and J. M. Shreeve, J. Fluorine Chem.,
Wilkes, “The Past, Present and Future of Ionic Liquids as Battery
127, 159 (2006); J. L. Anderson, D. W. Armstrong, and G.-T. Wei,
Electrolytes,” in Green Industrial Applications of Ionic Liquids, R. D.
Anal. Chem., 78, 2892 (2006); M. Galinski, A. Lewandowski, and I.
Rogers, K. R. Seddon, and S. Volkov, Editors, p. 295, NATO Science
Stepniak, Electrochim. Acta, 51, 5567 (2006).
Series, Vol. 92, Kluwer Academic Publishers, Dordrecht, The
4. T. Tsuda and C. L. Hussey, “Electrochemistry of Non-
Netherlands (2002).
Chloroaluminate Room-Temperature Ionic Liquids,” in Modern
27. J. Caja, T. D. J. Dunstan, D. M. Ryan, and V. Katovic, in Molten
Aspects of Electrochemistry, R. E. White, Editor, to be published
Salts XII, P. C. Trulove, H. C. De Long, G. R. Stafford, and S. Deki,
(2007).
Editors, PV 99-41, p. 150, The Electrochemical Society Proceedings
5. G. R. Stafford and C. L. Hussey, “Electrodeposition of Transition
Series, Pennington, NJ (1999).
Metal-Aluminum Alloys from Chloroaluminate Molten Salts,” in
28. H. Matsumoto, H. Sakaebe, K. Tatsumi, M. Kikuta, E. Ishiko, and
Advances in Electrochemical Science and Engineering, R. C. Alkire and
M. Kono, J. Power Sources, 160, 1308 (2006).
D. M. Kolb, Editors., Vol. 7, p. 275, Wiley-VCH, Weinheim (2002);
29. M. Ishikawa, T. Sugimoto, M. Kikuta, E. Ishiko, and M. Kono, J.
T. Tsuda, C. L. Hussey, and G. R. Stafford, “Progress in Surface
Power Sources, 162, 658 (2006).
Finishing with Lewis Acidic Room-Temperature Chloroaluminate
30. B. Garcia, S. Lavallée, G. Perron, C. Michot, and M. Armand,
Ionic Liquids,” in ECS Transactions, Molten Salts XV, in Memory of
Electrochim. Acta, 49, 4583 (2004).
Robert A. Osteryoung, In press (2007).
31. P. C. Howlett, N. Brack, A. F. Hollenkamp, M. Forsyth, and D. R.
6. R. Quigley, P. A. Barnard, C. L. Hussey, and K. R. Seddon, Inorg.
MacFarlane, J. Electrochem. Soc., 153, A595 (2006).
Chem., 31, 1255 (1992); C. L. Hussey, R. Quigley, and K. R. Seddon,
32. J. Janata and R. E. Gephart, Interface, 4(4), 46 (1995).
Inorg. Chem., 34, 370 (1995).
33. H. Luo, S. Dai, and P. V. Bonnesen, Anal. Chem., 76, 2773 (2004);
7. S. K. D. Strubinger, I-W. Sun, W. E. Cleland, Jr., and C. L. Hussey,
H. Luo, S. Dai, P. V. Bonnesen, A. C. Buchanan III, J. D. Holbrey,
Inorg. Chem., 29, 4246 (1990).
N. J. Bridges, and R. D. Rogers, Anal. Chem., 76, 3078 (2004); P.-Y.
8. J. Zhang, A. M. Bond, D. R. MacFarlane, S. A. Forsyth, J. M. Pringle,
Chen and C. L. Hussey, Electrochim. Acta, 50, 2533 (2005).
A. W. A. Mariotti, A. F. Glowinski, and A. G. Wedd, Inorg. Chem.,
34. T. Tsuda, C. L. Hussey, H. Luo, and S. Dai, J. Electrochem. Soc., 153,
44, 5123 (2005).
D171 (2006).
9. W. Lu, A. G. Fadeev, B. Qi, E. Smela, B. R. Mattes, J. Ding, G. M.
35. P. Dubois, G. Marchand, Y. Fouillet, J. Berthier, T. Douki, F.
Spinks, J. Mazurkiewicz, D. Zhou, G. G. Wallace, D. R. MacFarlane,
Hassine, S. Gmouh, and M. Vaultier, Anal. Chem., 78, 4909 (2006).
S. A. Forsyth, and M. Forsyth, Science, 297, 983 (2002).
36. G. B. Appetecchi, S. Scaccia, C. Tizzani, F. Alessandrini, and S.
10. For example, J. Ding, D. Zhou, G. Spinks, G. Wallace, S. Forsyth,
Passerini, J. Electrochem. Soc., 153, A1685 (2006).
M. Forsyth, and D. MacFarlane, Chem. Mater., 15, 2392 (2003);
D. Zhou, G. M. Spinks, G. G. Wallace, C. Tiyapiboonchaiya, D.
R. MacFarlane, M. Forsyth, and J. Sun, Electrochim. Acta, 48, 2355
About the Authors
(2003); F. Vidal, C. Plesse, D. Teyssié, and C. Chevrot, Synth. Met.,
t
142, 287 (2004); W. Lu and B. R. Mattes, Synth. Met., 152, 53
etsuya tsuda is a research assistant professor in the Department of
Chemistry and Biochemistry at the University of Mississippi and a
(2005); H. L. Ricks-Laskoski and A. W. Snow, J. Am. Chem. Soc.,
guest research scientist in the Department of Industrial Chemistry,
128, 12402 (2006).
Tokyo University of Science. He received his PhD in Energy Science
11. B. J. Akle, M. D. Bennett, and D. J. Leo, Sens. Actuators, A, 126, 173
from Kyoto University, Japan in 2001 with Professors Yasuhiko Ito
(2006).
and Rika Hagiwara. His research interests are energy science and
12. W. Lu, I. D. Norris, and B. R. Mattes, Aust. J. Chem., 58, 263 (2005).
materials science related to electrochemistry in molten salts (ionic
13. B. O’Regan and M. Grätzel, Nature, 353, 737 (1991).
liquids). He may be reached at ttsuda@olemiss.edu.
14. M. Grätzel, Nature, 414, 338 (2001).
15. For example, H. Matsumoto, T. Matsuda, T. Tsuda, R. Hagiwara,
Charles l. hussey is a professor of chemistry and Chair of the
Y. Ito, and Y. Miyazaki, Chem. Lett., 26 (2001); P. Wang, S. M.
Department of Chemistry and Biochemistry at the University
Zakeeruddin, J.-E. Moser, R. Humphry-Baker, and M. Grätzel, J.
of Mississippi. He received his BS in chemistry in 1971 and PhD
Am. Chem. Soc., 126, 7164 (2004); P. Wang, S. M. Zakeeruddin, R.
in chemistry in 1974 from the University of Mississippi. He is a
Humphry-Baker, and M. Grätzel, Chem. Mater., 16, 2694 (2004);
Fellow of The Electrochemical Society and serves as an Associate
S. Ito, S. M. Zakeeruddin, R. Humphry-Baker, P. Liska, R. Charvet,
Editor for the Journal of The Electrochemical Society. He has carried
P. Comte, M. K. Nazeeruddin, P. Péchy, M. Takata, H. Miura, S.
out research on the electrochemistry and transport properties
Uchida, and M. Grätzel, Adv. Mater., 18, 1202 (2006); D. Kuang,
of molten salts/ionic liquids for more than 33 years. He may be
P. Wang, S. Ito, S. M. Zakeeruddin, and M. Grätzel, J. Am. Chem.
reached at chclh@chem1.olemiss.edu.
Soc., 128, 7732 (2006); N. Mohmeyer, D. Kuang, P. Wang, H.-W.
The Electrochemical Society Interface • Spring 2007
49
