J. Chem. T ech. Biotechnol.

1997, 68, 351È356

Review

Ionic Liquids for Clean Technology*

Kenneth R. Seddon
School of Chemistry, The QueenÏs University of Belfast, Stranmillis Road, Belfast BT9 5AG, UK
(Received 20 June 1996 ; accepted 5 August 1996)

Abstract : The use of room-temperature chloroaluminate(III) ionic liquids, speciÐ-

cally 1-butylpyridinium chlorideÈaluminium(III) chloride and 1-ethyl-3-methyl-
imidazolium chlorideÈaluminium(III) chloride, as solvents for clean synthesis and
catalytic processes, particularly those applicable to clean technology, is becoming
widely recognised and accepted. The design principles for room-temperature
ionic liquids, some of their properties, and the rationale for using these neoteric
solvents, are discussed here, and an indication of the scope of these solvents for
future industrial processes is given.

Key words : ionic liquids, molten salts, clean technology, clean synthesis, neoteric
solvents, chloroaluminates(III)

1 PREAMBLE ous support from BP Chemicals, Unilever, ICI, BNFL,

Ionic liquids ? Neoteric solvents ? Molten salts for clean
technology and catalysis ? Are you serious ? ? Well, yes, I
am serious, but yes there are questions to be answered.
In 1982, I submitted a proposal to the then SERC
which concerned the use of room-temperature ionic
liquids to study catalytic processes which might have
some industrial application. The proposal was rejected :
referee 1 stated that the systems were far too compli-
cated, and therefore would never have any application,
referee 2 thought that the systems were far too simple,
and the work was not worth doing, and referee 3
(clearly not having the correct proposal in front of him)
wondered why we wanted to study the neutron di†rac-
tion patterns of vanadium bronzes ! ! Fortunately, the
BP Venture Research Unit (headed by Prof. Don
Braben) was more imaginative, and shortly afterwards
awarded us a substantial grant to enable us to investi-
gate the chemistry of ionic liquids. In 1996, thanks to
this crucial grant, and with subsequent extremely gener-
* This paper is dedicated to the memory of Prof. Gleb
Mamantov, whose recent death has deprived the ionic liquids
community of its most respected, and best loved, expert.
Presented in part at Clean Tech Ï96, London, 19È21 June
1996.
351
J. Chem. T ech. Biotechnol. 0268-2575/97/$17.50 ( 1997 SCI. Printed in Great Britain
IRTU and EPSRC (ROPA and Clean Synthesis
awards), combined with the blood, sweat and tears of
dozens of graduate students and postdoctoral fellows,
we believe that the initial concepts and speculation have
been vindicated, and that the industrial relevance of
ionic liquids, especially for clean synthesis and clean
catalysis, has been demonstrated (vide infra).
The purpose of this brief review is to give an intro-
duction to the nature and properties of room-
temperature ionic liquids, with particular emphasis
being placed upon their potential as solvents for indus-
trially relevant catalytic reactions, and (more generally)
for clean technology. This account is neither com-
prehensive nor exhaustive, but is intended to whet the
appetite of the reader, and perhaps encourage further
interest in a subject that, until recently, had been gener-
ally considered as rather esoteric. This account centres
around the use of room-temperature chloroaluminate
ionic liquids as solvents, especially the systems 1-
butylpyridinium chlorideÈaluminium(III) chloride,
[NBupy]ClÈAlCl and 1-ethyl-3-methylimidazolium
3
chlorideÈaluminium(III) chloride, [emim]ClÈAlCl (see
3
Scheme 1) : these are by no means the only systems
which may be used for catalytic studies, but they are
undoubtedly the archetypal systems. Much greater
352
Scheme 1.
detail concerning these solvents may be discovered in
two exceptional reviews by Hussey.1,2 An excellent
review of the chemistry in the related pyrophoric
system, [bmim]ClÈAlEtCl ([bmim]` \ 1-butyl-3-
2 reactions, and the ionic environment tends to signiÐ-
methylimidazolium cation), has recently been published
by Chauvin and Olivier-Bourbigou3 and the same
group has just reported some novel catalytic reactions
in air-stable ionic liquids,4 derived from those devel-
oped by Wilkes and Zaworotko.5
2 TERMINOLOGY
Ionic liquids are, quite simply, liquids that are com-
prised entirely of ions. Thus, molten sodium chloride is
an ionic liquid : a solution of sodium chloride in water
(a molecular solvent) is an ionic solution. The term ionic
liquids was selected with care, as it is my belief that the
more commonly used phrase molten salts (or simply
melts) is referential, and invokes a Ñawed image of these
solvents as being high-temperature, corrosive, viscous
media (cf. molten cryolite). The reality is that room-
temperature ionic liquids can be Ñuid at temperatures
as low as [ 96¡C, and are colourless, have low vis-
cosities, and are easily handled. To use the term molten
salts to describe these novel systems is as archaic as
describing a car as a horseless carriage.
The abstract of this review includes a term new to
chemistry : neoteric solvents. The word neoteric¤ is well
established in the English language. It is used here to
indicate a class of novel solvents that have remarkable
new properties, that “break new groundÏ, and that o†er
a huge potential for industrial application. It is applied
not only to ionic liquids, but also to supercritical
Ñuids,6,7 another type of solvent that shows huge
promise for clean synthesis. Thus the term neoteric sol-
vents covers both ionic liquids and supercritical Ñuids,
and brings together under one banner the two most
promising solvent systems for cleaning up the modern
chemical industry.
¤ Neoteric is deÐned in the Oxford English Dictionary as
meaning : recent, new, modern.
K. R. Seddon
3 RATIONALE : WHY USE IONIC LIQUIDS ?
The question as to why use ionic liquids at all has
recently been addressed in detail elsewhere, in a com-
panion review,8 which has been made available on a
World-Wide Web site,9 as a section of a dedicated ionic
liquids database.10 The relative merits of using ionic
liquids, instead of the more common molecular solvents,
are rehearsed there, and will not be repeated here.
Moreover, the case for using room-temperature ionic
liquids rather than high-temperature ionic liquids is
also argued. To brieÑy summarise, ionic liquids
suppress conventional solvation and solvolysis pheno-
mena, and provide media capable of dissolving a vast
range of inorganic molecules to very high concentra-
tions. The low (i.e. ambient) temperatures suppress dis-
sociation, disproportionation and degradation
cantly extend the lifetimes of species (such as
[RuCl ]3~)11 which are unstable in conventional
6
molecular solvents.
4 ROOM TEMPERATURE IONIC LIQUIDS
4.1 Lowering the melting point
The melting points of the Group 1 chlorides are given
in Table 1.12 As can be seen, these are signiÐcantly
above room temperature, and far too high to form a
generic medium for reactive chemistry. However, the
eutectic compositions of mixed ternary systems do melt
at signiÐcantly lower temperatures than either binary
component (see Fig. 1). Applying the understanding of
lattice energies gained from the Kapustinskii equa-
tion,13 the e†ect of increasing the size of the anion can
be seen in Table 2. The melting points of these simple
tetrachloroaluminate(III) salts are in the range of the
boiling points of high-boiling organic solvents. Further-
more, as the sodium chlorideÈaluminium(III) chloride
system shows compound formation at the 1 : 1 composi-
tion (viz. Na[AlCl ]), the idealised phase diagram (Fig.
4
2) now exhibits two eutectic points. However, as these
represent the lowest temperatures at which these
TABLE 1
Selected Melting Points of the Group 1 Chlorides
System Mole% m.p. (¡C)
LiCl 100 610
NaCl 100 803
KCl 100 772
CsCl 100 646
LiClÈCsCl 60È40 355
NaClÈKCl 50È50 658
CsClÈKCl 35È65 610
Ionic liquids for clean technology 353

4.2 Tetrachloroaluminate(III) ionic liquids

The development of ionic liquids that are Ñuid at

Fig. 1. An idealised simple binary phase diagram (e.g. for
waterÈsodium chloride mixtures), illustrating the eutectic
point.
room temperature by the groups of Osteryoung,14
Wilkes15 and Hussey16 has provided an ideal resolution
to all of the problems presented above. Although the
earlier work was based around the N-butylpyridinium
cation, [NBupy]`,14 research now is almost exclusively
based upon the 1-ethyl-3-methylimidazolium cation,
[emim]`,¤ or simple derivatives thereof. Undoubtedly
the most studied system is 1-ethyl-3-methylimidazolium
chlorideÈaluminium(III) chloride, [emim]ClÈAlCl , and
3
its experimental phase diagram is shown in Fig. 3.17
The unsymmetrical nature of the cation is crucial ;
[NBupy]` possesses a mirror plane missing from
[emim]`, which only has C symmetry (the ethyl group
1
is not co-planar with the imidazolium ring).18 The 1-
ethyl-3-methylimidazolium salts melt some 100¡C lower
than their N-butylpyridinium analogues.
These chloroaluminate systems are governed by the
following primary equilibrium, eqn (1) :19

systems are liquids, it is necessary to bring the melting
point down even further. This has been done by increas-
ing the size of the cations : replacing the simple inorga-
nic cations with unsymmetrical organic cations
depresses the melting point to temperatures at or below
room temperature, as illustrated in the following
section.
TABLE 2
Melting Points of Selected Tetrachloroaluminate(III) Salts
2[AlCl ]~ H [Al Cl ]~ ] Cl~ ; K B 10~16.3 (1)
4 2 7
The chemical and physical properties of the liquids are
determined by the percentage of aluminium(III) chlo-
ride : those which have X(AlCl ) greater than 50 mol%
3
are referred to as acidic, whilst those with X(AlCl ) less
3
than 50 mol% are called basic. The exactly 50 mol%
composition is referred to as neutral. It should be

System Mole% m.p. (¡C)

AlCl 100 192

3
LiClÈAlCl 50È50 144
3
NaClÈAlCl 50È50 151
3
KClÈAlCl 50È50 256
3

Fig. 3. The experimental phase diagram for the

Fig. 2. An idealised phase diagram for a system showing the
formation of a compound with a congruent melting point.
[emim]ClÈAlCl system, showing the formation of a com-
3
pound, [emim][AlCl ], with a congruent melting point.17
4
¤ In the older literature, this cation is almost exclusively
referred to (incorrectly) as 1-methyl-3-ethylimidazolium, and
represented by [MeEtim]`. Its full, correct, systematic name
is the 1-ethyl-3-methyl-1H-imidazolium cation.
354
TABLE 3
Principal Anions Present in the [emim]ClÈAlCl Ionic Liquid System, as a Function of Composi-
3
tion
Basic ; X(AlCl ) \ 0È0É5 Acidic ; X(AlCl ) \ 0É5È0É67
3 3
Cl~
[AlCl ]~ [AlCl ]~
4 4
[Al Cl ]~
2 7
emphasised here that the terms “acidÏ and “baseÏ refer to
Franklin acidity and basicity, and not (as frequently
assumed) to Lewis acidity and basicity. The ionic com-
position of the [emim]ClÈAlCl system is displayed on
3
a logarithmic scale ( [ pX) in Fig. 4, and on a linear
scale in Fig. 5. There are three key features of note.
Firstly, a clear parallel (both quantitative and
qualitative) exists between the curves illustrated in Fig.
4, and standard pH curves in water. Secondly, the con-
Fig. 4. The concentration of anions (represented on a
logarithmic scale) present in the [emim]ClÈAlCl system, as a
function of composition. 3 “dirtyÏ end of the chemical industry, oil reÐning and
Fig. 5. The concentration of anions (represented on a linear
scale) present in the [emim]ClÈAlCl system, as a function of
composition.3
K. R. Seddon
V ery acidic ; X(AlCl ) \ 0É67È0É75
3
[Al Cl ]~
2 7
[Al Cl ]~
3 10
centration of the tetrachloroaluminate(III) ion is
approximately constant, and very high, across the
whole compositional range. Finally, there is always
exactly 50 mol% of the 1-ethyl-3-methylimidazolium
cation present. Table 3 summarises the principal anions
present as a function of composition : deÐnitive evidence
now exists for the existence of [Al Cl ]~ in very acidic
3 10
systems.20,21
There is not the space here to enlarge further on the
physical and chemical properties of these unique
systems : the interested reader is referred to Hussey.1,2
5 IONIC LIQUIDS FOR CLEAN CATALYSIS
Clean technology is concerned with reducing the waste
from an industrial chemical process to zero. Its imple-
mentation will lead to a cleaner environment, and more
cost-e†ective use of starting materials. It requires, in the
majority of cases, the rethinking and redesign of many
chemical processes that we currently think of as the
industrial norm. Catalysis clearly has a central role to
play in this process, with its potential for both high
yields and high speciÐcity.
The E-factor of a process is the ratio (by weight) of
the by-products to the desired product(s).22 Table 4
illustrates that, contrary to popular perception, the
bulk chemicals, is remarkably waste conscious. It is,
surprisingly, the Ðne chemicals and pharmaceutical
companies that are using inefficient, dirty, processes,
albeit on a much smaller scale, and there will be
increasing demand (political, economic, social and
environmental) for the introduction of new clean tech-
nology. Industries that do not respond will die.
TABLE 4
By-products as a Proportion of Production for the Chemical
Industry22
Industry Production (tons p.a.) E-factor
Oil reÐning 106È108 0É1
Bulk chemicals 104È106 1È5
Fine chemicals 102È104 5È50
Pharmaceuticals 101È103 25È100
Ionic liquids for clean technology
Volatile organic solvents are the normal media for
the industrial synthesis of organic chemicals
(petrochemical and pharmaceutical), with a current
worldwide usage estimated at £4 000 000 000 p.a.
However, the environmental impact of these solvents is
signiÐcant, and the Montreal Protocol has resulted in a
compelling need to re-evaluate many chemical processes
that have proved otherwise satisfactory for much of this
century. A conspicuous example is the recent closure of
one of the DuPont Hypalon' plants in Northern
Ireland, which had been operating with chlorinated
hydrocarbon solvents. There appear to be four main
alternative strategies :
(1) solvent-free synthesis
(2) the use of water as a solvent
(3) the use of supercritical Ñuids as solvents
(4) the use of ionic liquids as solvents
It is the purpose of this review to stimulate interest in
option (4), and to incite enough activity in this area to
allow a meaningful evaluation. Ionic liquids possess,
inter alia, the following desirable properties :
1. They have a liquid range of 300¡C (cf. water, which
has a liquid range of 100¡C, and ammonia, with a
liquid range of only 44¡C), allowing tremendous
kinetic control.
2. They are outstandingly good solvents for a wide
range of inorganic, organic and polymeric materials
(but, fortunately, they do not dissolve polythene,
PTFE or glass !) : high solubility implies small reactor
volumes.
3. They exhibit BrÔnsted, Lewis and Franklin acidity,
as well as superacidity.23,24
4. They have no e†ective vapour pressure.
5. Their water sensitivity does not restrict their indus-
trial applications.
6. They are thermally stable up to 200¡C.
7. They are relatively cheap, and easy to prepare.
Unlike water and other hydroxylic solvents, they will
dissolve a wide range of organic molecules to an appre-
ciable extent (benzene, for example, will form up to 50%
(v/v) solutions), meaning much lower volumes of solvent
are required for a given process. Exploratory work in
our own laboratories (carried out in collaboration with
BP Chemicals and Unilever Port Sunlight Research
Laboratory) has demonstrated that a wide range of
catalysed organic reactions (including oligomerisations,
polymerisations, alkylations and acylations) occurs in
room-temperature ionic liquids, and that these are
serious candidates for commercial processes.25,26 For
example, the oligomerisation of butenes to polyisobu-
tene in a chloroaluminate(III) ionic liquid25 is a strong
competitor for the Cosden process, which uses a sup-
ported or liquid phase aluminium(III) chloride cata-
lyst.27
355
The reactions we have observed represent the tip of
an icebergÈall the indications are that room-
temperature ionic liquids are the basis of a new indus-
trial technology. They are truly designer solvents : either
the cation26 or the anion5 can be changed, if not at will,
then certainly with considerable ease, in order to opti-
mise such phenomena as the relative solubilities of the
reactants and products, the reaction kinetics, the liquid
range of the solvent, the cost of the solvent, the intrinsic
catalytic behaviour of the media, and air-stability of the
system. For the Ðrst time, it is possible to design a
solvent to optimise a reaction (with control over both
yield and selectivity), rather than to let the solvent
dictate the course of the reaction. For example, we have
just reported upon a new series of ionic liquids that are
also liquid crystals, with a phenomenal liquid range (up
to 188¡C) of smectic A phase stability, which o†er a sig-
niÐcant new ionic medium for studying the stereocon-
trol of reactions (e.g. polymerisation reactions) in liquid
crystals.28 This, quite literally, revolutionises the meth-
odology of synthetic organic chemistry : it will never be
the same again !
ACKNOWLEDGEMENT
I am indebted to all my co-workers, but especially to
my past and present postdoctoral assistants (Drs J. E.
Turp, A. K. Abdul-Sada, T. Welton, A. C. Lees, M.
Earle, C. J. Bowlas, C. M. Gordon, Y. Patell and T. A.
Evans), graduate students (D. Appleby, L. Dutta, A.
Elaiwi, L. Agocs and S. A. Boyle) and structural charac-
terisation colleagues (Dr P. B. Hitchcock, Dr A. J. Dent
and M. Nieuwenhuyzen), without whose hard labour
this research would not have been possible. I am partic-
ularly grateful to Prof. D. W. Braben and Dr D. Ray
(Venture Research International), Prof. C. J. Adams
(Unilever Port Sunlight Research Laboratory), Dr C.
Sell (Quest International) and Prof. M. J. Green and
Drs M. P. Atkins and B. Ellis (all BP Chemicals) for
both their Ðnancial and intellectual input to this Ðeld,
and especially for their support when ionic liquids were
not as well accepted as they are now. Finally, the active
collaboration and friendship of Prof. C. L. Hussey
throughout this period is warmly appreciated.
REFERENCES
1. Hussey, C. L., Room temperature molten salt systems.
Adv. Molten Salt Chem., 5 (1983) 185È230.
2. Hussey, C. L., Room-temperature haloaluminate ionic
liquidsÈnovel solvents for transition-metal solution
chemistry. Pure Appl. Chem., 60 (1988) 1763È72.
3. Chauvin, Y. & Olivier-Bourbigou, H., Nonaqueous ionic
liquids as reaction solvents. CHEMT ECH, 25(9) (1995)
26È30.
356
4. Chauvin, Y., Mussmann, L. & Olivier, H., A novel class of
versatile solvents for 2-phase catalysisÈhydrogenation,
isomerization, and hydroformylation of alkenes catalyzed
by rhodium complexes in liquid 1,3-dialkylimidazolium
salts. Angew. Chem. Int. Edit. Engl., 34 (1996) 2698È700.
5. Wilkes, J. S. & Zaworotko, M. J., Air and water stable
1-ethyl-3-methylimidazolium based ionic liquids. J. Chem.
Soc., Chem. Commun., (1992) 965È7.
6. Poliako†, M. & Howdle, S., Supercritical chemistryÈ
synthesis with a scanner. Chem. Brit., 31 (1995) 118È21.
7. Poliako†, M., Howdle, S. M. & Kazarian, S. G., Vibra-
tional spectroscopy in supercritical ÑuidsÈfrom analysis
and hydrogen-bonding to polymers and synthesis. Angew.
Chem. Int. Edit. Engl., 34 (1995) 1275È95.
8. Seddon, K. R., Room-temperature ionic liquids : neoteric
solvents for clean catalysis. Kinet. Katal., (1997) (in press).
9. http ://www.ch.qub.ac.uk/resources/ionic/review/review.html.
10. http ://www.ch.qub.ac.uk/resources/ionic/ionic.html.
11. Hussey, C. L., Barnard, P. A., Sun, I. W., Appleby, D.,
Hitchcock, P. B., Seddon, K. R., Welton, T. & Zora, J. A.,
An electrochemical study of the ruthenium(III) and
ruthenium(IV) hexachlorometallates in a basic room-
temperature chloroaluminate molten salt. J. Electrochem.
Soc., 138 (1991) 2590È4.
12. Lide, D. R. (ed.) CRC Handbook of Chemistry and Physics,
73rd edn. CRC Press, Boca Raton, 1992.
13. Kapustinskii, A. F., Lattice energy of ionic crystals. Quart.
Rev., 10 (1956) 283È94.
14. Robinson, J. & Osteryoung, R. A., An electrochemical and
spectroscopic study of some aromatic hydrocarbons in the
room temperature molten salt system aluminum chlorideÈ
n-butylpyridinium chloride. J. Am. Chem. Soc., 101 (1979)
323È27.
15. Wilkes, J. S., Levisky, J. A., Wilson, R. A. & Hussey, C. L.,
Dialkylimidazolium chloroaluminate meltsÈa new class
of room-temperature ionic liquids for electrochemistry,
spectroscopy and synthesis. Inorg. Chem., 21 (1982) 1263È
4.
16. Sanders, J. R., Ward, E. H. & Hussey, C. L., Aluminum
bromide-1-methyl-3-ethylimidazolium bromide ionic
liquids. 1. Densities, viscosities, electrical conductivities,
and phase-transitions. J. Electrochem. Soc., 133 (1986)
325È30.
17. Fannin, A. A., Floreani, D. A., King, L. A., Landers, J. S.,
Piersma, B. J., Stech, D. J., Vaughn, R. L., Wilkes, J. S. &
Williams, J. L., Properties of 1,3-dialkylimidazolium chlo-
K. R. Seddon
ride aluminum chloride ionic liquids. 2. Phase-transitions,
densities, electrical conductivities, and viscosities. J. Phys.
Chem., 88 (1984) 2614È21.
18. Abdul-Sada, A. K., Greenway, A. M., Hitchcock, P. B.,
Mohammed, T. J., Seddon, K. R. & Zora, J. A., Upon the
structure of room-temperature halogenoaluminate ionic
liquids. J. Chem. Soc., Chem. Commun., (1986) 1753È4.
19. Hussey, C. L., Scheffler, T. B., Wilkes, J. S. & Fannin, A.
A., Chloroaluminate equilibria in the aluminum chloride-
1-methyl-3-ethylimidazolium chloride ionic liquid. J. Elec-
trochem. Soc., 133 (1986) 1389È91.
20. Abdul-Sada, A. K., Greenway, A. M., Seddon, K. R. &
Welton, T., Upon the existence of [Al Cl ]~ in room-
temperature chloroaluminate ionic liquids. 3 10 Org. Mass
Spectrom., 24 (1989) 917È18.
21. Abdul-Sada, A. K., Greenway, A. M., Seddon, K. R. &
Welton, T., A fast-atom-bombardment mass-spectrometric
study of room-temperature 1-ethyl-3-methylimidazolium
chloroaluminate(III) ionic liquidsÈevidence for the exis-
tence of the decachlorotrialuminate(III) anion. Org. Mass
Spectrom., 28 (1993) 759È65.
22. Sheldon, R. A., The role of catalysis in waste mini-
mization. In Precision Process T echnology : Perspectives
for Pollution Prevention, eds M. P. C. Weijnen & A. A. H.
Drinkenburg. Kluwer, Dordrecht, 1993, p. 125È38.
23. Smith, G. P., Dworkin, A. S., Pagni, R. M. & Zingg, S. P.,
BrÔnsted superacidity of HCl in a liquid
chloroaluminateÈAlCl -1-ethyl-3-methyl-1H-imidazolium
chloride. J. Am. Chem.3Soc., 111 (1989) 525È30.
24. Smith, G. P., Dworkin, A. S., Pagni, R. M. & Zingg, S. P.,
Quantitative study of the acidity of HCl in a molten chlo-
roaluminate system (AlCl 1-ethyl-3-methyl-1H-imid-
3 of HCl pressure and melt
azolium chloride) as a function
composition (51É0È66É4 mol% AlCl ). J. Am. Chem. Soc.,
3
111 (1989) 5075È7.
25. Abdul-Sada, A. K., Atkins, M. P., Ellis, B., Hodgson, P. K.
G., Morgan, M. L. M. & Seddon, K. R., World Patent,
WO 95/21806, 1995.
26. Abdul-Sada, A. K., Ambler, P. W., Hodgson, P. K. G.,
Seddon, K. R. & Stewart, N. J., World Patent, WO 95/
21871, 1995.
27. Weissermel, K. & Arpe, H.-J., Industrial Organic Chem-
istry, 2nd revised edn. VCH, Weinheim, 1993.
28. Bowlas, C. J., Bruce, D. W. & Seddon, K. R., Liquid-
crystalline ionic liquids. J. Chem. Soc., Chem. Commun.,
(1996) 1625È26.