Anion effects on ionogel formation in N,N′-dialkylimidazolium-based ionic liquids

Article abstract of DOI:10.1016/j.ica.2004.06.042

Addition of an appropriate concentration of water to either 1-decyl-3-methylimidazolium bromide or its nitrate analog is shown to trigger the self-assembly of the ionic liquid (IL) and the concomitant formation of gel ( ionogel ) phases adopting a rich variety of structural motifs. Infrared and ¹H nuclear magnetic resonance studies indicate that water addition disrupts hydrogen bonding between the imidazolium ring and the IL anion, and that gelation involves the partial replacement of these bonds with H-bonds between the anion and water. Using small angle X-ray scattering, the relationship of the water content and the nature of the IL anion to the mesoscopic structure of the ionogels has been determined. Of particular note is the observation that by adjustment of the ionogel composition (water content or anion), near-monodomain materials can be formed with either lamellar, 2-D hexagonal or 3-D cubic structures with tunable lattice dimensions.

Full text of DOI:10.1016/j.ica.2004.06.042

Inorganica Chimica Acta 357 (2004) 3991–3998
www.elsevier.com/locate/ica
Anion effects on ionogel formation in
N,N0-dialkylimidazolium-based ionic liquids
q
a,
b
c
b
Millicent A. Firestone *, Paul G. Rickert , So¨nke Seifert , Mark L. Dietz
a
Materials Science Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439, USA
b
Chemistry Division, Argonne National Laboratory, Argonne, IL 60439, USA
c
Advanced Photon Source Division, Argonne National Laboratory, Argonne, IL 60439, USA
Received 4 May 2004; accepted 26 June 2004
Available online 29 July 2004
Dedicated to Professor T.J. Marks on the occasion of his 60th birthday
Abstract
Addition of an appropriate concentration of water to either 1-decyl-3-methylimidazolium bromide or its nitrate analog is shown
to trigger the self-assembly of the ionic liquid (IL) and the concomitant formation of gel (‘‘ionogel’’) phases adopting a rich variety
1
of structural motifs. Infrared and H nuclear magnetic resonance studies indicate that water addition disrupts hydrogen bonding
between the imidazolium ring and the IL anion, and that gelation involves the partial replacement of these bonds with H-bonds
between the anion and water. Using small angle X-ray scattering, the relationship of the water content and the nature of the IL
anion to the mesoscopic structure of the ionogels has been determined. Of particular note is the observation that by adjustment
of the ionogel composition (water content or anion), near-monodomain materials can be formed with either lamellar, 2-D hexagonal
or 3-D cubic structures with tunable lattice dimensions.
Published by Elsevier B.V.
Keywords: Ionic liquid; Ionogel; Anion effects; Mesoscopic structure; SAXS
1. Introduction
nitrogen- or phosphorus-containing organic salts charac-
terized by sufficient asymmetry to inhibit their crystalliza-
tion (and thus reduce their melting temperature to 6100
ꢀC), exhibit numerous properties that distinguish them
from ordinary (i.e., molecular) organic solvents. For ex-
ample, like classical molten salts, ionic liquids exist in a
fully ionized state. As a result, they typically exhibit little
or no vapor pressure [3]. In addition, they exhibit a wide
liquidus range and electrochemical window [2,11]. Finally,
and perhaps most importantly, ILs are extraordinarily
tunable, with even minor changes in the structure of
the anion or cation comprising the IL often having a
substantial effect on its physicochemical properties [2].
This tunability is of obvious potential utility in the ap-
plication of ILs in the design and synthesis of novel ma-
terials, and recently, there has been growing recognition
of the possibilities offered by ILs in this area [12–19]. Of
Since the introduction of the first air- and water-stable
ionic liquids (ILs) more than a decade ago [1], there has
been increasing interest in their potential as substitutes
for conventional organic solvents in a wide range of cat-
alytic [2–4], electrochemical [5,6], and synthetic [7–10] ap-
plications. These compounds, which typically consist of
q
The submitted manuscript has been created by the University of
ChicagoasOperatorofArgonneNationalLaboratory(‘‘Argonne’’)under
Contract No. W-31-109-ENG-38 with the US Department of Energy. The
USGovernmentretainsforitself, andothersactingonitsbehalf,apaid-up,
non-exclusive, irrevocable worldwide license in said article to reproduce,
prepare derivative works, distribute copies to the public, and perform
publicly and display publicly, by or on behalf of the Government.
*
Corresponding author. Tel.: +630-252-8298; fax: +630-252-9151.
E-mail address: firestone@anl.gov (M.A. Firestone).
0020-1693/$ - see front matter. Published by Elsevier B.V.
doi:10.1016/j.ica.2004.06.042

3992
M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
particular interest has been the use of ILs as templates in
the fabrication of micro- and mesoporous inorganic and
hybrid organic–inorganic materials. Adams et al. [12], for
example, noting the surfactant-like nature of long-chain
dialkylimidazolium salts, examined the use of 1-alkyl-3-
methylimidazolium halides ([Cnmim+][Xꢀ], with n=6–18
and Xꢀ =Brꢀ or Clꢀ) as replacements for the alkyltri-
methylammonium salts normally employed as templates
in the preparation of MCM molecular sieves. As is the
case in conventional routes to MCM synthesis using
[CnNMe3]+Xꢀ templates, best results were obtained for
the hexadecyl-substituted ILs (i.e., n=16). Interestingly,
however, MCM-41 of acceptable quality could be pre-
pared only from the chloride-based ILs, suggesting a sig-
nificant anion effect on template viability. More recently,
Zhou et al. [15] have exploited hydrogen bond formation
between an IL anion (BF4ꢀ) and the silano group of silica
gel and the p–p stack interactions between neighboring
lished procedures [25], characterized by 1H and 13C
NMR, dried in vacuo at 80 ꢀC, and stored in a desicca-
tor until use. The corresponding nitrate was synthesized
by a metathesis reaction involving treatment of an aque-
ous solution of [C10mim+][Brꢀ] with silver nitrate. The
accumulated solids (AgBr) were removed by filtration
and the IL recovered by evaporation of the water. Tests
of the recovered IL showed no additional precipitate
formation upon addition of AgNO3 solution. (It is im-
portant to note that this does not imply the complete ab-
sence of bromide ion, as the minimum achievable level
of bromide contamination is dictated by the solubility
product constant of AgBr in water, 5.2·10ꢀ13 [26].
Ion chromatographic determination of the bromide con-
tent of the [C10mim+][NO3ꢀ], however, showed it to be
less than the limit of detection for the method, 0.05%
(w/w).) Nanopure water was used in the preparation
of all solutions. All other materials were A.C.S. reagent
grade and were used as received.
imidazolium rings to form a network of [C4mim+][BF4
]
ꢀ
that served to template monolithic mesoporous silica.
These same investigators have demonstrated [16] that
highly ordered, monolithic ‘‘super-mesoporous’’ (pore
size: 1–2 nm) lamellar silica can be produced using as a
template the liquid crystalline (LC) phase of various
long-chain (e.g., n=16) 1-alkyl-3-methylimidazolium
chlorides. It remains unclear in either case if the results
are anion-specific.
2.1.2. Ionogel preparation
Ionogels were prepared by combining weighed quan-
tities of water and the desired ionic liquid. To ensure
that the gels produced were homogeneous, repeated vig-
orous mixing was required. Gentle warming, by reduc-
ing the viscosity of the mixture, facilitated the
homogenization in most instances.
Our own work with ILs has concerned the preparation
and characterization of liquid crystalline [Cnmim+][Xꢀ]
‘‘ionogels’’, physical gels comprising mesoscopically or-
dered assemblies of relatively short-chain (n=10) 1-al-
kyl-3-methylimidazolium salts. Recent work in our
laboratory has shown, for example, that the addition of
an appropriate concentration of water to 1-decyl-3-meth-
ylimidazolium bromide results in its spontaneous self-or-
ganization and the formation of a LC gel, thus
demonstrating that ionogel formation requires neither
the functionalization of the IL [20] nor the addition of
an exotic mesogen [21] or organogelator [22]. As part of
our ongoing efforts to exploit soft matter in the design
and preparation of new materials [23,24] and, in particu-
lar, as a first step in an examination of the utility of
ionogels as templates for the synthesis of mesoporous in-
organic materials, we have undertaken an investigation
of the origins of ionogel formation and of the relationship
ofthe natureof theIL aniontothemesoscopicstructureof
ionogels. In this report, we describe our initial findings.
2.2. Methods
2.2.1. Infrared characterization
All IR spectra (64 scans; 4 cmꢀ1 resolution) were ob-
tained using a Nicolet Nexus Model 870 infrared spec-
trometer with samples placed between ZnSe windows.
2.2.2. NMR spectroscopy
NMR measurements were performed using a Varian
Unity/INOVA system incorporating a 300 MHz magnet
(Oxford) and a Nalorac Z-SPEC 5 mm broadband
probe. The nature of the water–IL interactions leading
to gelation was probed by examining the effect of water
addition upon the chemical shift of the protons of the
[C10mim+] cation for [C10mim+][Xꢀ] (X=Brꢀ and
NO3ꢀ). In an initial series of experiments, increasing
quantities of water were added to weighed samples of
[C10mim+][Brꢀ] in a series of NMR tubes, and the mix-
tures repeatedly warmed and mixed in an effort to ob-
tain uniform samples. For mixtures containing either
low (approximately 60.16) or high ( P 0.8) ratios of wa-
ter to IL, the sample viscosity was such that a homoge-
neous sample could eventually be obtained and
satisfactory NMR spectra could be acquired. For inter-
mediate water contents, however, this was not the case,
and efforts to obtain spectra directly on the ionogels
were eventually abandoned. Instead, a concentrated (1
M) solution of the IL in deuterated acetonitrile was pre-
2. Experimental
2.1. Materials
2.1.1. Ionic liquid synthesis
The ionic liquid 1-decyl-3-methylimidazolium bro-
mide was synthesized and purified according to the pub-

M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
3993
pared and placed in an NMR tube. This solution was
then titrated with water added using a metered pipette
and the chemical shift of the [C10mim+] protons meas-
ured as a function of water concentration. The small
amount of non-deuterated acetonitrile present in
CD3CN was used as the internal reference for the chem-
ical shift determination. (CH3CN has a pentuplet cen-
tered at 1.93 ppm.) Because the IL contains a trace of
water even after prolonged oven drying, the total water
content of samples containing low added-water-to-IL
ratios was determined by integration of the NMR spec-
trum.
gelation. Detailed experimental evidence in support of
this contention was not provided, however. In an effort
to obtain such evidence, we have carried out both infra-
red and H NMR investigations directed at the elucida-
tion of the nature of the interaction between water
molecules and the IL leading to gel formation.
1
Consistent with the literature reports describing the
infrared spectra of various [C2mim+][Xꢀ] salts [27–29],
the infrared spectrum of oven-dried [C10mim+][Brꢀ] ex-
hibits a series of three prominent peaks at 2854, 2925,
and 2956 cmꢀ1 associated with aliphatic C–H stretching,
a peak at 3137 cmꢀ1 attributable to aromatic C–H
stretching, and most notably, a peak at 3058 cmꢀ1
,
2.2.3. Small-angle X-ray scattering studies
Small-angle X-ray scattering studies (SAXS) weꢀre
which has been attributed to C–H stretching for C–
Hꢁ ꢁ ꢁBrꢀ. In addition, a series of weak peaks, referred
to as ‘‘Sheppard effect bands’’ and arising from combi-
nations of the three fundamental vibrational modes of
C–Hꢁ ꢁ ꢁBrꢀ [29] are observed in the region between
carried out on [C10mim+][Brꢀ] and [C10mim+][NO3
]
ILs containing either 17% or 40% (w/w) water. Viscous
samples (e.g., [C10mim+][Brꢀ] containing 17% (w/w) wa-
ter) were loaded from the melt and held in glass capillar-
ies fashioned from Pasteur pipettes. Low viscosity
samples (e.g., [C10mim+][NO3ꢀ] containing 40% (w/w)
water) were loaded into 1.5 mm quartz capillaries by a
syringe. SAXS measurements were performed on an
undulator beamline (12ID-C) of the advanced photon
source (APS) at Argonne National Laboratory. The
scattering profiles were recorded with either a Mar-
CCD-165 detector (Mar USA, Evanston, IL), which fea-
tures a circular, 165 mm diameter active area and
2048·2048 pixel resolution or on a custom-built mosaic
detector (gold) consisting of 9 CCD chips with an image
are of 15·5 cm and 1536·1536 pixel resolution. The
area detector images were corrected for background
scattering of the glass and quartz capillaries by subtract-
ing from the recorded images an area detector image of
the respective capillaries with the same exposure time as
the ionogel/IL (1–5 s for the MarCCD and 0.1–0.5
gold). The sample-to-detector distance was set such that
the detecting range for momentum transfer was
2500 and 2800 cmꢀ1
.
Upon uptake of water (ca. 1% w/w), a strong, broad
band consistent with the O–H stretch for intermolecu-
larly hydrogen-bonded water appears at 3430 cmꢀ1. At
the same time, the peaks associated with the aromatic
C–H stretch and the C–H stretch of the C–Hꢁ ꢁ ꢁBrꢀ moi-
ety shift significantly (6 and 16 cmꢀ1, respectively) to
higher wavenumbers. According to Pimentel and McC-
lellan [30], the stretching mode of an A–H moiety is
shifted to lower frequencies upon H-bond formation.
The upward shift in the stretching frequency for C–
Hꢁ ꢁ ꢁBrꢀ is thus consistent with disruption/diminution
of H-bond formation between the imidazolium ring
and the bromide ion upon addition of water. In contrast
to these shifts, most of the bands for the [C10mim+][Brꢀ]
remain essentially unchanged upon an increase in water
content. The positions of the bands associated with ali-
phatic C–H stretching (2854, 2925 and 2956 cmꢀ1), for
example, are indistinguishable from those of the dried
compound. The same can be said for the methylene scis-
soring and twisting bands (1466 and 1338 cmꢀ1, respec-
tively) and for the methyl symmetrical bending band
(1378 cmꢀ1). Thus, as might be expected, the alkyl side-
arms, unlike the imidazolium ‘‘headgroup’’, are essen-
tially unaffected by the presence of water.
0.004<q<1.5 Aꢀ1. The collected scattering data were
˚
calibrated on the basis of the known positions of silver
behenate powder Bragg reflections. All samples were
measured within 2 days of preparation.
Upon further addition of water and subsequent gela-
tion, several additional changes in the spectrum become
apparent. First, the position of the O–H stretching band
falls to 3421 cmꢀ1, a shift of ꢀ9 cmꢀ1 relative to the cor-
responding peak in the spectrum of the fluid phase. At
the same time, the position of the aromatic C–H stretch
for C–Hꢁ ꢁ ꢁBrꢀ shifts upward by 8 cmꢀ1 (for a total of 24
cmꢀ1 versus the dried compound). In addition, the in-
tensity of this band decreases significantly relative to
that observed for the fluid phase. According to Pimentel
and McClellan [30], the absorption coefficient of a
stretching band will increase when a hydrogen bond
forms. This decrease is thus indicative of a lessening of
3. Results and discussion
3.1. Origins of ionogel formation
In a prior communication [13], we noted that the for-
mation of an ionogel upon addition of water to
[C10mim+][Brꢀ] likely involved the formation of an H-
bonded network comprising water, bromide ion, and
the imidazolium cation. Such network formation could
serve to enhance the segregation of the hydrophobic
and hydrophilic segments of [C10mim+], thus leading
to regions of confined water and ultimately, to physical

3994
M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
H-bonding between C–H and Brꢀ upon addition of wa-
ter/gelation. This is supported by the observation that
the Sheppard effect bands of C–Hꢁ ꢁ ꢁBrꢀ weaken to the
point of becoming barely discernible upon gel forma-
tion. In contrast, no effect on the peaks associated with
aliphatic C–H stretching (methyl or methylene), bending
(methyl), or scissoring (methylene) is evident.
According to LeGrange et al. [31] and Firestone et al.
[32], the position of the asymmetric methylene stretching
band is indicative of the degree of conformational order
of the alkyl chains, with ordered, all-trans alkyl chains
and disordered chains bearing gauche defects exhibiting
peaks at 2918 and 2924 cmꢀ1, respectively. This suggests
that the C10 alkyl chains of [C10mim+][Brꢀ], apparently
disordered in the fluid state, remain so in the gel state.
That is, the process of gelation has no discernible effect
on the state of the alkyl chains of the imidazolium cati-
on. Thus, gelation in this system manifests itself prima-
rily by changes in the aromatic headgroup of the
compound and the associated anion. Specifically, water
addition appears to disrupt hydrogen bond formation
between the imidazolium ring and the bromide ion, sug-
gesting that gelation involves the partial replacement of
these bonds with H-bonds between Brꢀ and water (I).
(The existence of H-bonding between water and a halide
ion in an IL is not, it should be noted, unprecedented.
Lee et al. [33], for example, present X-ray crystallo-
graphic data indicating the presence of such interactions
in various 1,3-dialkylbenzimidazolium salts, such as
[(C12)2–Bim+][Clꢀ], and describe their significance in de-
termining the structure of the compound.) That gel for-
mation appears to arise from a balance between IL
cation–anion and IL anion–water interactions is consist-
ent with the recent studies by Cammarata et al. [34] of
the state of water in [C4mim+][Tf2Nꢀ] (where Tf2Nꢀ
represents the bis[(trifluoromethyl)sulfonyl]imide anion)
and related ILs, for which competition between the im-
idazolium cation and water for H-bonding with the IL
anion was noted.
Fig. 1. Effect of water on the chemical shift of the C-2 proton of
[C10mim+][Brꢀ] (open circles) and [C10mim+][NO3ꢀ] (solid squares) in
CD3CN.
significantly deshielded (d=9.712 ppm and 9.254 ppm,
respectively, for the bromide and nitrate ILs), gradually
becomes less so as the water content of the solution ris-
es. (Although the chemical shifts associated with all of
the imidazolium ring protons are impacted by water ad-
dition, the effect is most pronounced for the C-2 proton,
consistent with the reported greater acidity of this pro-
ton and the accompanying greater sensitivity of its
chemical shift to changes in counteranion and solvent
properties [35].) That the initial chemical shift is greater
for the [C10mim+][Brꢀ] indicates that the interaction of
bromide ion with the IL cation (in particular, with its
C-2 proton) is stronger than for the nitrate anion, as ex-
pected given the greater basicity of Brꢀ [36]. The absence
of any discontinuity in this plot at water:IL ratios corre-
sponding to gelation is consistent with the view of gel
formation suggested by the infrared results: that gela-
tion results from the gradual establishment of a balance
between IL cation–anion (specifically, C-2–H-anion)
and anion–water interactions as the water concentration
is increased.
Although we do not consider here the relationship be-
tween the molecular/ionic composition of ionogels/ILs
and their macroscopic properties, it is interesting to note
that [C10mim+][Brꢀ], whose anion is apparently capable
of stronger interaction with the IL cation, forms a ro-
bust (i.e., rigid) gel compared to the [C10mim+][NO3ꢀ],
which is a viscous liquid at room temperature. This is al-
so supported by thermal analysis (differential scanning
calorimetry, DSC), which reveals an endothermic transi-
tion, Tm, at 44.6 ꢀC for the 16% (w/w) [C10mim+][Brꢀ]
[13] versus 24.7 ꢀC for [C10mim+][NO3ꢀ].
+
N
N
CH3
H
H
H
X-
O
1
The results of H NMR studies in which the effect of
water addition upon the observed chemical shift for the
imidazolium ring protons of [C10mim+][Brꢀ] and the cor-
responding nitrate was examined are presented in Fig. 1.
As can be seen, in both cases, addition of water to an
acetonitrile solution of the IL results in a significant de-
crease in the chemical shift associated with the proton
on C-2 (the carbon between the two nitrogen atoms in
the imidazolium ring). Thus, this proton, which is initially
3.2. Anion effects on ionogel/IL structure
The structural ordering of ILs on the mesoscopic
length scale (nm–lm) makes them well suited for study
by SAXS [13]. Moreover, this technique, unlike either
electron microscopy or scanning probe methods, is

M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
3995
non-destructive and can probe the structure of soft ma-
terials without significantly perturbing them. Our prior
SAXS studies of [C10mim+][Brꢀ] ionogels with water
contents between 11% and 16% (w/w) showed that their
mesoscopic structure is lamellar (i.e., a stacked, layered
structure) (Fig. 2, LAM), as indicated by the appearance
of two strong Bragg peaks at integral order spacing in
the azimuthally averaged data [13]. The 2-D scattering
pattern, however, displayed strong anisotropic scatter-
ing, with the scattered X-ray intensity predominantly di-
rected along the equatorial axis, thus suggesting either
directionally ordered lamellae or the onset of a structure
with modulation (with hexagonal symmetry) of the la-
mellae, designated hexagonal modulated layer (HML)
[37].
block copolymers, the HPL structure is characterized
by penetration of layers of the minority component by
the majority component, forming hexagonally arranged
channels that serve to connect the layers [41].
The results presented here do not, it should be noted,
suggest that the gel phase consists solely of an HPL struc-
ture under the experimental conditions. In fact, closer ex-
amination of Fig. 3(a) (making particular note of the
higher order reflections in the 2-D images that are diffi-
cult to observe without saturating the first-order reflec-
tions) indicates that it possess features that may be
consistent with the remnants of the characteristic 10-spot
pattern of a gyroid (Ia3d) phase (Fig. 2, G), a phase char-
acterized by a bicontinuous (interpenetrating) cubic net-
work of two phases (i.e., water/anion and [C10mim+]).
The absence of Bragg reflections that index to ratios of
At a slightly higher water content than those employed
previously, 17% (w/w), a structural transition to a higher
order symmetry phase is observed (Fig. 3(a)). Specifi-
cally, the averaged 1-D data (Fig. 3(b)) display a single
strong Bragg reflection and three weaker reflections (at
p
p
6 and 8 preclude the assignment of a true gyroid
structure (G) to this phase, however. Thus, the structure
of the gel at this water content is best described as HPL
with some cubic character, perhaps indicative of conver-
sion of the structure to a stable G phase at slightly higher
water contents. It should also be noted that factors other
than anion and water content, among them sample han-
dling (e.g., loading), temperature, and sample history,
could influence the mesophase structure.
When the water content of the C10mim+Brꢀ is in-
creased sufficiently (to 40% (w/w)), the sample converts
back to a low viscosity state and the 2-D scattering pat-
tern becomes an isotropic ring, indicating a randomly
oriented polydomain sample (Fig. 3(c)). Azimuthally av-
q=0.218 and 0.378, 0.430, and 0.572 Aꢀ1, respectively)
˚
p
p
From the position of the first-order peak, a d-spacing
that readily index to a hexagonal lattice (1: 3:2: 7).
˚
of 28.8 A can be determined. The 2-D scattering pattern
shows that, as was the case at lower water contents, the
scattered X-ray intensity is directed predominately in
the equatorial direction. Nonetheless, the arcs show var-
iations in intensity that make a six spot pattern readily
apparent (Fig. 3(a)), indicating significant long-range,
‘‘single-crystal-like’’ ordering and suggesting a structural
conversion to a hexagonally perforated layer mesophase
(Fig. 2, HPL). Such mesophases have been reported for
both surfactant-water systems [38,39] and diblock copol-
ymers, and are often observed as intermediate structural
states during the transition from a lamellar to a hexago-
nal phase associated with temperature jumps [40]. In di-
eraging the 2-D data (Fig. 3(d)) yields two broad diffrac-
p
tion peaks at a ratio of 1: 3 (q=0.189, 0.320 Aꢀ1),
˚
signaling conversion of the mesophase to a disordered
hexagonal structure.
To determine the effect of a change in anion on the
mesophase structure, analogous SAXS investigations
Fig. 2. Schematic illustrating the structural phases identified by SAXS on samples prepared with no additional water added [C10mim+][Brꢀ]: lamellar
structure (LAM); 17% (w/w) added water [C10mim+][Brꢀ] and [C10mim+][NO3ꢀ]: hexagonal perforated layers (HPL) and gyroid structure (G); 40%
(w/w) added water [C10mim+][Brꢀ] and [C10mim+][NO3ꢀ]: hexagonal structure (HEX).

3996
M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
Fig. 3. (a) 2-D small-angle X-ray scattering pattern for a [C10mim+][Brꢀ] phase prepared with 17% (w/w) water added. (b) Azimuthally integrated
scattering data presented in (a). (c) SAXS pattern from a [C10mim+][Brꢀ] phase prepared with 40% (w/w) water added. (d) Azimuthally integrated
scattering data presented in (c).
Fig. 4. (a) 2-D small-angle X-ray scattering pattern from a [C10mim+][NO3ꢀ] phase prepared with 16% (w/w) water added. (b) Azimuthally
integrated scattering data presented in (a). (c) SAXS pattern from a [C10mim+][NO3ꢀ] phase prepared with 40% (w/w) water added. (d) Azimuthally
integrated scattering data presented in (c).

M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
3997
of [C10mim+][NO3ꢀ] were carried out. The 1-D scatter-
ing curve for a nitrate ionogel containing 16% (w/w) wa-
ter is presented in Fig. 4(b). Three diffraction peaks
Department of Energy, under contract number W-31-
109-ENG-38. This manuscript is dedicated to Prof. To-
bin Marks on the occasion of his 60th birthday.
(q=0.058, 0.100, 0.116 Aꢀ1), which can be indexed to
˚
p
ingly, although the thermochemical radii of nitrate and
a hexagonal structure (1: 3:2), are observed. Interest-
References
˚
bromide ions are comparable (1.96 and 1.88 A, respec-
tively) [42], the d-spacing for nitrate-based ionogel is
˚
108.3 A, more than three times that observed for the
[1] J.S. Wilkes, M.J. Zaworotko, J. Chem. Soc., Chem. Commun.
(1992) 965.
[2] H. Olivier-Bourbigou, L. Magna, J. Mol. Catal. A 182 (2002) 419.
[3] K.R. Seddon, Kinet. Catal. 37 (1996) 693.
˚
bromide-based gel (28.8 A), implying that anion proper-
ties other than simply size (e.g., symmetry, bonding di-
rectionality) can exert a significant influence on the
mesoscopic structure of ionogels. The 2-D SAXS pat-
tern (Fig. 4(a)) at this water content is anisotropic, fea-
turing a spot pattern similar to that observed for the
corresponding bromide system, again suggestive of for-
mation of a HPL type structure. In this case, however,
preliminary data collected on a gel containing a lower
water content (9% w/w; results not shown) suggest that
3-D cubic structures comprising a gyroid phase are
formed (Fig. 2, G). Investigations on block copolymers
have also demonstrated the epitaxial growth of oriented
gyroid phases from hexagonal phases [43]. When the wa-
ter concentration is increased to 40% (w/w), the struc-
ture of the nitrate ionogel, like that of the
corresponding bromide, is found to convert to a disor-
[4] D. Zhao, M. Wu, Y. Kou, E. Min, Catal. Today 74 (2002) 157.
[5] J. Fuller, R.T. Carlin, R.A. Osteryoung, J. Electrochem. Soc. 144
(1997) 3881.
[6] R.T. Carlin, J. Fuller, H.A. Frank, E.T. Seo, in: Proceedings of
the 12th Annual Battery Conference on Applications and
Advances – 1997, Institute of Electrical and Electronics Engineers,
New York, 1997, pp. 261–266.
[7] K.R. Seddon, J. Chem. Technol. Biotechnol. 68 (1997) 351.
[8] J.D. Holbrey, K.R. Seddon, Clean Prod. Process. 1 (1999) 223.
[9] M.J. Earle, K.R. Seddon, Pure Appl. Chem. 72 (2000) 1391.
[10] A.E. Visser, W.M. Reichert, R.P. Swatloski, H.D. Willauer, J.G.
Huddleston, R.D. Rogers, in: R.D. Rogers, K. Seddon (Eds.),
Ionic Liquids: Industrial Applications for Green Chemistry,
American Chemical Society, Washington, DC, 2002, pp. 289–308.
[11] J.F. Brennecke, E.J. Maginn, AIChE J. 47 (2001) 2384.
[12] C.J. Adams, A.E. Bradley, K.R. Seddon, Aust. J. Chem. 54
(2001) 679.
[13] M.A. Firestone, J.A. Dzielawa, P. Zapol, L.A. Curtiss, S. Seifert,
M.L. Dietz, Langmuir 18 (2002) 7258.
[14] M.L. Dietz, J.A. Dzielawa, M.P. Jensen, M.A. Firestone, in:
R.D. Rogers, K.R. Seddon (Eds.), Ionic Liquids as Green
Solvents: Progress and Prospects, American Chemical Society,
Washington, DC, 2003, pp. 526–543.
dered hexagonal structure, with the azimuthally aver-
p
aged data exhibiting two broad reflections at 1: 3
(Fig. 4(d)). The appearance of both a diffuse isotropic
ring in the 2-D image, as well as the breadth of the dif-
fraction peaks in the 1-D curves, are indicative of a mes-
ophase of poor spatial coherence.
[15] Y. Zhou, J.H. Schattka, M. Antonietti, Nano Lett. 4 (2004) 477.
[16] Y. Zhou, M. Antonietti, Chem. Mater. 16 (2004) 544.
[17] T. Nakashima, N. Kimizuka, J. Am. Chem. Soc. 125 (2003) 6386.
[18] M. Yoshio, T. Mukai, H. Ohno, T. Kato, J. Am. Chem. Soc. 126
(2004) 994.
[19] B. Lee, H. Luo, C.Y. Yuan, J.S. Lin, S. Dai, Chem. Commun.
(2004) 240.
4. Conclusions
[20] N. Kimizuka, T. Nakashima, Langmuir 17 (2001) 6759.
[21] M. Yoshio, T. Mukai, K. Kanie, M. Yoshizawa, H. Ohno, T.
Kato, Adv. Mater. 14 (2002) 351.
The results presented here, in addition to elucidating
the origins of ionogel formation, clearly demonstrate
that the mesophase structure (both symmetry and lattice
dimensions) of IL-based gels can be ‘‘tuned’’ simply by
careful selection of the anion and adjustment of the wa-
ter concentration. Given that progress in the design of
templated inorganic and hybrid organic–inorganic ma-
terials is dependent on the ability to control the struc-
ture of the underlying template, the ease with which
the structure of [C10mim+][Xꢀ] ionogels can be varied
suggests that they may provide a versatile platform for
the preparation of nano- and mesoporous materials.
Work addressing these opportunities is now underway
in this laboratory.
[22] A. Ikeda, K. Sonoda, M. Ayabe, S. Tamaru, T. Nakashima, N.
Kimizuka, S. Shinkai, Chem. Lett. (2001) 1154.
[23] M.A. Firestone, P. Thiyagarajan, D.M. Tiede, Langmuir 14
(1998) 4688.
[24] M.A. Firestone, D.E. Williams, S. Seifert, R. Csencsits, Nano
Lett. 1 (2001) 129.
[25] J.G. Huddleston, H.D. Willauer, R.P. Swatloski, A.E. Visser,
R.D. Rogers, Chem. Commun. (1998) 1765.
[26] D.A. Skoog, D.M. West, Fundamentals of Analytical Chemistry,
third ed., Holt, Rinehart, and Winston, New York, 1976.
[27] A.K. Abdul-Sada, A.M. Greenway, P.B. Hitchcock, T.J. Mo-
hammed, K.R. Seddon, J.A. Zora, J. Chem. Soc., Chem.
Commun. (1986) 1753.
[28] J. Fuller, R.T. Carlin, H.C. DeLong, D. Haworth, J. Chem. Soc.,
Chem. Commun. (1994) 299.
[29] A. Elaiwi, P.B. Hitchcock, K.R. Seddon, N. Srinivasan, Y.-M.
Tan, T. Welton, J.A. Zora, J. Chem. Soc., Dalton Trans. (1995)
3467.
Acknowledgement
[30] G.C. Pimentel, A.L. McClellan, The Hydrogen Bond, W.H.
Freeman, San Francisco, 1960.
This work was performed under the auspices of the
Office of Basic Energy Sciences, Division of Chemical
(MLD, PGR) and Materials (MAF, SS) Sciences, US
[31] J.D. LeGrange, J.L. Markham, C.R. Kurkjian, Langmuir 9
(1993) 1749.

3998
M.A. Firestone et al. / Inorganica Chimica Acta 357 (2004) 3991–3998
[32] M.A. Firestone, M.L. Shank, S.G. Sligar, P.W. Bohn, J. Am.
Chem. Soc. 118 (1996) 9033.
[38] Y. Rancon, J. Charvolin, J. Phys. Chem. 92 (1988) 2646.
[39] M. Imai, A. Kawaguchi, A. Sacki, K. Nakaye, T. Kato, K. Ito,
Y. Amemiya, Phys. Rev. E 62 (2000) 6865.
[33] K.M. Lee, C.K. Lee, I.J.B. Lin, Chem. Commun. (1997) 899.
[34] L. Cammarata, S.G. Kazarian, P.A. Salter, T. Welton, Phys.
Chem. Chem. Phys. 3 (2001) 5192.
[40] S. Qi, Z.G. Wang, Phys. Rev. E 55 (1997) 1682.
[41] C.-Y. Wang, T.R. Lodge, Macromolecules 35 (2002) 6997.
[42] B.A. Moyer, P.V. Bonnesen, in: A. Bianchi, K. Bowman-James,
E. Garcia-Espana (Eds.), Supramolecular Chemistry of Anions,
Wiley-VCH, New York, 1997, pp. 1–44.
[35] A.D. Headley, N.M. Jackson, J. Phys. Org. Chem. 15 (2002) 52.
[36] R. Shukla, T. Kida, B.D. Smith, Org. Lett. 2 (2000) 3099.
[37] M.E. Vigild, K. Almdal, K. Mortensen, I.W. Hamley, J.P.A.
Fairclough, A.J. Ryan, Macromolecules 31 (1998) 5702.
[43] J.-H. Ahn, W.-C. Zin, Macromol. Res. 11 (2003) 153.