In search of ionic liquids incorporating azolate anions

Article abstract of DOI:10.1002/chem.200500840

Twenty-eight novel salts with tetramethyl-, tetraethyl-, and tetrabutylammonium and 1-butyl-3-methylimidazolium cations paired with 3,5-dinitro-1,2,4-triazolate, 4-nitro-1,2,3-triazolate, 2,4-dinitroimidazolate, 4,5-dinitroimidazolate, 4,5-dicyanoimidazolate, 4-nitroimidazolate, and tetrazolate anions have been prepared and characterized by using differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), and single-crystal X-ray crystallography. The effects of cation and anion type and structure on the physicochemical properties of the resulting salts, including several ionic liquids, have been examined and discussed. Ionic liquids (defined as having m.p. <100°C) were obtained with all combinations of the 1-butyl-3- methylimidazolium cation ([C₄mim]+) and the heterocyclic azolate anions studied, and with several combinations of tet raethyl or tetrabutylammonium cations and the azolate anions. The [C₄mim] ⁺ azolates were liquid at room temperature exhibiting large liquid ranges and forming glasses on cooling with glasstransition temperatures in the range of -53 to -82°C (except for the 3,5-dinitro-1,2,4-triazolate salt with m.p. 33°C). Six crystal structures of the corresponding tetraalkylammonium salts were determined and the effects of changes to the cations and anions on the packing of the structure have been investigated.

Full text of DOI:10.1002/chem.200500840

DOI: 10.1002/chem.200500840
In Search of Ionic Liquids Incorporating Azolate Anions
Alan R. Katritzky,*[a] Shailendra Singh,[a] Kostyantyn Kirichenko,[a] Marcin Smiglak,[b]
John D. Holbrey,*[b, c] W. Matthew Reichert,[b] Scott K. Spear,[b] and Robin D. Rogers*[b]
Abstract: Twenty-eight novel salts with
tetramethyl-, tetraethyl-, and tetrabuty-
lammonium and 1-butyl-3-methylimi-
dazolium cations paired with 3,5-dini-
tro-1,2,4-triazolate, 4-nitro-1,2,3-triazo-
late, 2,4-dinitroimidazolate, 4,5-dini-
troimidazolate, 4,5-dicyanoimidazolate,
4-nitroimidazolate, and tetrazolate
anions have been prepared and charac-
terized by using differential scanning
calorimetry (DSC), thermogravimetric
analysis (TGA), and single-crystal X-
ray crystallography. The effects of
cation and anion type and structure on
the physicochemical properties of the
resulting salts, including several ionic
liquids, have been examined and dis-
cussed. Ionic liquids (defined as having
m.p.<1008C) were obtained with all
combinations of the 1-butyl-3-methyli-
midazolium cation ([C4mim]+) and the
heterocyclic azolate anions studied,
and with several combinations of tet-
raethyl or tetrabutylammonium cations
and the azolate anions. The [C4mim]+
azolates were liquid at room tempera-
ture exhibiting large liquid ranges and
forming glasses on cooling with glass-
transition temperatures in the range of
ꢀ53 to ꢀ828C (except for the 3,5-dini-
tro-1,2,4-triazolate salt with m.p.
338C). Six crystal structures of the cor-
responding tetraalkylammonium salts
were determined and the effects of
changes to the cations and anions on
the packing of the structure have been
investigated.
Keywords: azolates
structures imidazolium salts
ionic liquids · phase transitions
·
crystal
·
·
Introduction
high academic and industrial interest in IL technologies.[1–10]
Ionic liquids make a unique architectural platform on
which, at least potentially, the properties of both anion and
cation components can be independently modified, enabling
tunability in the design of new functional materials. This
design approach can be used as a platform strategy to deliv-
er different functional attributes in the corresponding cati-
onic and anionic components of an IL, while retaining the
core desirable features of the IL state of matter, and has ap-
plicability in many different materials areas.
The growing social pressure for new green technologies and
the promise of ionic liquids (ILs) to deliver such has led to
[a] Prof. A. R. Katritzky, Dr. S. Singh, Dr. K. Kirichenko
Center for Heterocyclic Compounds
University of Florida
Department of Chemistry
Gainesville, FL 32611-7200 (USA)
Fax : (+1)352-392-9199
The desire to develop energetic salts[11] as advanced mate-
rials with reduced environmental footprints and overall ex-
posure to hazardous and potentially toxic materials includ-
ing hydrazine, metals, and perchlorates, has led to renewed
efforts to explore this area. Drake and co-workers utilized
one aspect of the platform approach to IL design to prepare
nonvolatile, thermally stable, liquid energetic materials[12] by
combining energetic cations (based on 1,2,4-triazole, 4-
amino-1,2,4-triazole, and 1,2,3-triazole cores[13]) and anions
E-mail: Katritzky@chem.ufl.edu
[b] M. Smiglak, Dr. J. D. Holbrey, Dr. W. M. Reichert, Dr. S. K. Spear,
Prof. R. D. Rogers
Center for Green Manufacturing and Department of Chemistry
The University of Alabama
Tuscaloosa, AL 35487 (USA)
Fax : (+1)205-348-0823
E-mail: j.holbrey@qub.ac.uk
RDRogers@bama.ua.edu
([NO3]ꢀ, [ClO4]ꢀ, and [N(NO2)2]ꢀ). It was suggested that
A
[c] Dr. J. D. Holbrey
Current address: The QUILL Research Centre
Queenꢀs University of Belfast
BT9 5AG, Northern Ireland (UK)
the ready formation of ILs using these cations was due to
their similar topographical shape and charge distribution to
more conventional imidazolium cations.
Our interest in functionalized heterocycles in this context
comes from 1) the ongoing study of the formation of ILs
Supporting information for this article (showing the chemical struc-
tures of the azolate salts explored for IL behavior) is available on the
WWW under http://www.chemeurj.org/ or from the author.
4630
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4630 – 4641

FULL PAPER
containing energetic components, in which heterocycles can
furnish either the cationic or anionic portion of the resultant
salt, and 2) investigation of the influence of heterocycle sub-
stitution patterns on the ability of these materials to form
IL phases. That is, using the platform strategy to deliver the
energetic characteristics of nitro- and amine-substituted het-
erocycles in an IL material in which the liquid characteris-
tics are controlled through the complementary counterion.
Whereas nitro- and cyano-functionalized imidazoles can
be used to prepare imidazolium-based ILs,[14] the addition of
electron-withdrawing substituents leads to an overall desta-
bilization of the heterocyclic cation relative to the corre-
sponding azole, thus limiting their applicability in the forma-
tion of conventional ILs. However, the anionic azolate form
of the heterocycles are stabilized further by the introduction
of such electron-withdrawing substituents and presents op-
portunities to investigate the formation of ILs containing
heterocyclic anions.
ment in the basic understanding of the key physical and
chemical properties of these complex liquids can be made.
Here we present the synthesis and characterization, in-
cluding elucidation of solid-state crystal structures, of a sys-
tematic series of twenty-eight novel tetraalkylammonium
and 1-butyl-3-methylimidazolium ([C4mim]+) salts prepared
by combining the four cations with seven heterocyclic tetra-
zolate, triazolate, and imidazolate anions. Figure 1 illustrates
Surprisingly, the investigation of potential ILs containing
planar heterocyclic anions has been largely neglected. A
small number of tetraalkylammonium and -phosphonium
1,2,4-triazolates and imidazolates (formed from onium hal-
ides and azolates) have been described as solvents and
highly active catalysts for oligomerization of isocyanates in
the patent literature.[15] Ohno et al. described the prepara-
tion and electrochemical properties of two ILs 1-ethyl-3-
methylimidazolium tetrazolate and triazolate, which were
prepared through the reaction of imidazolium hydroxide
(generated by anion exchange from the halide) with the cor-
responding azole.[16,17]
This formation of low-melting, relatively low-viscosity ILs
containing both heterocyclic planar cations and anions was
somewhat unexpected and prompted us to investigate
whether comparable low-melting ILs could be prepared in-
corporating highly functionalized azolate anions. The salt,
tetrabutylammonium 4-nitroimidazolate[18] (m.p. 1208C),
had been reported as an intermediate for the synthesis of N-
alkyl-4-nitroimidazoles, and following this, we described the
formation of 1-butyl-3-methylimidazolium 3,5-dinitro-1,2,4-
triazolate, a crystalline low-melting solid (m.p. ꢁ338C),
which demonstrated that ILs could be readily obtained even
with more highly functionalized heterocyclic anions[19] by
using the appropriate selection of complementary cations
and anions.
Figure 1. Cations and anions explored in this work(note: compounds
1a–g are covalent, all others are ionic salts).
the diversity of azolate salts that can be prepared by using
simple procedures (also see Figure 1 in the Supporting Infor-
mation). Many of the azoles used to furnish the anion com-
ponent of these salts are energetic heterocyclic compounds
and have been studied extensively as potential energetic ma-
terials. The ability to readily form IL examples with en-
hanced stability relative to the starting heterocycles (con-
trolling melting points, higher thermal decomposition tem-
perature, and reduced volatility), and the generic observa-
tion that IL materials can be formed from a wide range of
flat heterocyclic anions by modification or selection of the
appropriate ions, sets out principles by which ILs containing
a new class of anions can be developed and exploited.
Results and Discussion
Organic tetraalkylammonium (methyl, ethyl, butyl) and 1-
butyl-3-methylimidazolium salts were prepared with seven
different azolate anions (Figure 1), 3,5-dinitro-1,2,4-triazo-
late (a), 4-nitro-1,2,3-triazolate (b), 2,4-dinitroimidazolate
(c), 4,5-dinitroimidazolate (d), 4,5-dicyanoimidazolate (e), 4-
nitroimidazolate (f), and tetrazolate (g), by metathesis reac-
tion of the corresponding potassium azolate salts with the
tetraalkylammonium or [C4mim]+ halides in acetone/di-
chloromethane. The cation choice was made in order to es-
tablish the broad relative potentials of the organic azolate
salts prepared to form IL phases, using the progression in
size from tetramethylammonium to tetrabutylammonium to
identify the range of melting points that might be anticipat-
ed when using other cations that have a higher tendency to
form IL salts, and to isolate crystalline salts from which
solid-state data could be obtained by using X-ray crystallog-
raphy analysis. Salts with [C4mim]+ cations were then pre-
pared and examined to determine the validity of these pre-
Most recently, Shreeve and co-workers[20,21] have reported
that a wide range of energetic azolium azolate salts can be
prepared with melting points approaching the definition of
“ionic liquids”, however to date, a systematic study of the
complementarity of cation and anion selection, which can
be used to obtain room-temperature energetic ILs, has not
been reported.
To extend the scope of these promising results, the availa-
bility of more organic anions (rather than the common inor-
ganic anions now in vogue) that provide the same architec-
tural flexibility as the commonly used organic cations cur-
rently studied (e.g., substituted imidazolium cations) needs
to be achieved. With diverse anions available, an improve-
Chem. Eur. J. 2006, 12, 4630 – 4641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4631

A. R. Katritzky, J. D. Holbrey, R. D. Rogers et al.
dictions. ([C4mim]+ is currently the most investigated
common IL supporting cation and it was used to give ther-
mal data directly comparable to the body of existing IL sys-
tems.)
tetramethylammonium chloride and potassium tetrazolate in
acetone/dichloromethane.
The structures of salts (3–6)a–g were supported by data
1
from H and 13C NMR spectra and elemental analyses (see
The potassium azolate salts 2a–d were initially prepared
by treatment of the corresponding azoles 1a–d with potassi-
um carbonate in acetone (Scheme 1). Reaction of potassium
the Experimental Section).
Thermal characterization: The salts varied in their physical
state from high-melting-point solids (with tetramethylammo-
nium cations) to ILs at room temperature, forming glassy
solids on cooling ([C4mim]+ cations). The thermal behaviors
of the salts (Table 1) were characterized by using differential
scanning calorimetry (DSC). The crystalline salts all dis-
played a sharp melting transition on heating, and crystal-
lized on cooling from the melt. However, in the bulk, sever-
al of the lower melting salts only crystallized very slowly on
cooling from the melt. The melting transitions of many of
the salts were poorly defined in the DSC data, commonly
having a characteristic broad transition with a strong “lead-
ing edge”. In contrast, the crystallization peaks were excep-
tionally sharp. This behavior has been observed for other
IL-forming salts.[22]
The salts with the 3,5-dinitro-1,2,4-triazolate (a), 2,4-dini-
troimidazolate (c), and 4-nitroimidazolate (f) anions all
showed simple thermal behavior, with melting and crystalli-
zation or glass transitions (6c and 6 f) that were constant
and independent of the sampleꢀs thermal history. The re-
maining salts, with 4-nitro-1,2,3-triazolate (b), 4,5-dinitroimi-
dazolate (d), 4,5-dicyanoimidazolate (e), and tetrazolate (g)
anions, showed more extensive and complicated behavior
exhibiting both reversible solid–solid and irreversible transi-
tions.
Scheme 1. Synthesis strategies.
azolates 2a–d with equivalent amounts of appropriate tet-
raalkylammonium chlorides or bromides or [C4mim]+ chlor-
ide in acetone/dichloromethane (1:1 v/v) at 20–258C gave
the corresponding tetraalkylammonium and [C4mim]+ azo-
lates (3–6)a–d in high yields (78–99%) after removal of the
precipitated potassium halide and evaporation of the sol-
vent.
It proved difficult to isolate pure potassium 4,5-dicyano-
imidazolate (2e), potassium 4-nitroimidazolate (2 f), and po-
tassium tetrazolate (2g) due to their low solubility. For the
synthesis of the organic 4,5-dicyanoimidazolate (3–6)e, 4-ni-
troimidazolate (3–6)f, and tetrazolate (3–6)g salts, the re-
spective potassium azolate intermediates 2e–g were generat-
ed in situ in the presence of potassium carbonate and not
isolated. Thus, treatment of azoles 1e–g with tetraalkylam-
monium or [C4mim]+ halides in the presence of potassium
carbonate in acetone/dichloromethane gave good yields of
ILs (3–6)e–g except salt 3g, which was obtained in only
15% yield, perhaps as a result of the low solubility of both
Salts (3–4)a–g, 5a–f, and 6a were crystalline or microcrys-
talline solids, which were isolated by precipitation from solu-
tion, whereas 5g and all the [C4mim]+ cation salts 6b–g
were isolated as room-temperature liquids. The tetramethy-
lammonium salts 3a–g were high-melting solids with melting
points ranging from 157 to 2168C, with only small, or no
liquid ranges and with decomposition generally being initiat-
Table 1. Thermal transitions for 3–6
ACHTREUNG
Et4N+ (4)
Tdecomp
Bu4N+ (5)
C
Anion
m.p.
m.p.
m.p.
Tdecomp
m.p.
33 (77)
3,5-dinitro-1,2,4-triazolate (a)
4-nitro-1,2,3-triazolate (b)
2,4-dinitroimidazolate (c)
4,5-dinitroimidazolate (d)
4,5-dicyanoimidazolate (e)
4-nitroimidazolate (f)
tetrazolate (g)
bistriflimide
triflate
tetrafluoroborate
214 (97)
235 (275)
180
114 (156)[c]
205 (296)
194
139 (186)
87 (113)[c]
81 (101)
85 (38)[c]
79 (85)[b]
106 (122)
ꢀ66[d]
219 (298)
192
157 (170)[b]
181 (132)
215[b]
82 (101)[c]
86 (132)
84 (97)
ꢀ73[d]
ꢀ53[d]
ꢀ64[d]
ꢀ74[b]
ꢀ63[b]
ꢀ82[d]
222 (274)
225 (258)
202 (232)
165 (215)
198 (241)
–
216 (271)
215 (253)
212 (255)
188 (230)
189 (231)
–
221 (297)
222 (287)
215 (258)
193 (232)
180 (220)
–
197[b]
118 (142)
124 (201)
114 (138)[c]
109[34]
185
216[b]
133[34]
96[34]
>300[37]
–
–
160–161[38]
–
111–112[38]
159–162
–
–
>300
>300[39]
–
ꢀ85–
A
[a] Melting (m.p.) or glass transition (Tg) points [8C] were measured from transition onset temperature and transition enthalpy (Jgꢀ1, in parentheses) de-
termined by using DSC from the second heating cycle at 58C minꢀ1, after initially melting and then cooling samples to ꢀ1008C unless otherwise indicat-
ed. Decomposition temperatures (Tdecomp) were determined by TGA, heating at 58C minꢀ1 under argon, and are reported as onset to 5 wt% mass loss
and are shown with the inflection point for total mass loss in parentheses. Salts meeting the definition of ionic liquids (m.p. <1008C) have values given
in bold. [b] Reversible solid–solid transitions: 3b, 118 (13); 3d, 51 (7) and 86 (50); 3e, 15 (14) and 60 (64); 3g, 83 (82); 5e, 52 (28). [c] Irreversible solid–
solid transition, from first heating: 4a, 102 (19); 4b, 32 (83); 4g, 49 (97); 5b, 59 (16); 5d, 55 (47). [d] Glass transitions, measured on cooling. [e] Glass
transition/crystallization point/melting point.
4632
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4630 – 4641

Ionic Liquids
FULL PAPER
ed upon melting. For each of the anions here, increased
cation size is connected with a significant decrease in melt-
ing point. The tetraethylammonium salts 4a–g all exhibited
melting points around 1008C (82–1248C), which correspond-
ed to a destabilization of the crystal state of around 1008C
relative to the tetramethylammonium analogues. When the
larger tetrabutylammonium cation was used (5a–g), a great-
er variation in the melting point with anion was observed,
from 1398C for 5a to room-temperature liquid with sub-am-
bient glass formation for 5g. The characteristic heating and
cooling profiles for the tetrabutylammonium salts 5a,c–f are
shown in Figure 2.
The tetrazolate salt, 3g, also displayed a reversible solid-
solid transition at 838C. A solid–solid transition at 498C was
observed in the first heating scan only for 4g.
The most striking evidence for polymorphism was found
in the 4-nitro-1,2,3-triazolate salt 4b, which was obtained as
a low-temperature monotropic polymorph from solution.
On heating, the salt appears to melt at 308C to form a clear
viscous phase, which transforms on standing to a more
stable solid polymorph. This transition is observed in the
DSC trace as a first-order endothermic peakon the first
heating cycle. After cooling and on the second heating, a
more stable polymorph (reversible m.p. 828C, crystallizing
at 638C) is obtained (Figure 3).
Figure 2. Characteristic DSC traces for the tetrabutylammonium salts
5a,c–f showing the characteristic broad melting and sharp crystallization
transitions, and the relative changes in melting point with anion type.
Transitions are shown from the second heating and cooling cycles.
Figure 3. Characteristic DSC traces for the tetraethylammonium 4-nitro-
1,2,3-triazolate (4b) showing the polymorphic behavior of the character-
ized salt. Transitions are shown from the first and second heating and
cooling cycles to indicate the differences in the melting points of the
polymorphs.
As anticipated, introduction of the asymmetric [C4mim]+
cation (6) led to a further reduction in the melting points of
the salts. Salts 6b–g were liquids at room temperature, and
6a, the only [C4mim]+ salt isolated as a crystalline solid, had
a melting point of only 338C. The ILs with [C4mim]+ cations
6b–g showed only a glass transition point in the range be-
tween ꢀ53 and ꢀ828C, and no freezing transition was ob-
served. This behavior, widely observed in IL samples, is
comparable to the results reported for 1-ethyl-3-methylimi-
dazolium triazolate and tetrazolate by Ohno and co-work-
ers.[16]
In the 4,5-dinitroimidazolate (d) series, reversible solid–
solid transitions were observed for 3d at 51 and 868C. In
contrast, 5d showed a sharp solid–solid transition at 558C in
the first heating cycle followed by a broad melting transition
at ꢁ958C. On cooling, the sample of 5d crystallized at
658C, and on subsequent heating scans, melted at 858C.
For the 4,5-dicyanoimidazolates (e), two reversible crys-
tal–crystal transitions at 15 and 608C were observed for 3e
(Table 1). 5e showed a reversible crystal–crystal transition
that was observed at 608C on first heating, and at 528C on
the second heating cycle (shown in Figure 2) followed by
melting at 798C. Only a single crystallization peak, at 208C,
was observed in the cooling scan.
Thermal stability was measured by using thermogravimet-
ric analysis (TGA), with isocratic heating at 58Cminꢀ1
under an inert argon atmosphere. All the salts prepared
were thermally stable to greater than 2008C, except for the
4-nitro-1,2,3-triazolates (3–6)b, which started to decompose
between 180 and 2198C (Table 1). The decomposition tem-
peratures shown in Table 1 were determined from both the
onset to 5 wt% mass loss in an isocratic TGA experiment,
and from the inflection point for complete decomposition,
and all showed a single-step decomposition profile. The salts
show a slight increase in upper stability temperature on in-
creasing the cation alkyl chain from methyl to butyl, and on
changing from ammonium to imidazolium cations. However,
the increased stability is relatively small and is probably not
significant. The high-melting tetramethylammonium salts all
showed the onset of decomposition around the melting
points, whereas appreciable liquid ranges could be obtained
with the larger cations, most notably with [C4mim]+, with
liquidus ranges from sub-ambient to above 1508C.
The melting points (and Tg points for noncrystallizing
salts) varied most significantly with change in the size and
shape of the organic cation, as anticipated, in that they de-
creased with increasing cation size and with much less varia-
Chem. Eur. J. 2006, 12, 4630 – 4641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4633

A. R. Katritzky, J. D. Holbrey, R. D. Rogers et al.
tion than was found with anion type. The liquid ranges for
each salt, determined from the lower melting transition
through to the decomposition temperatures, are shown in
Figure 4 grouped by anion type. It can be clearly seen that
lates versus triflates and tetrafluoroborates in the tetraethyl-
and tetramethylammonium salts.
It can be clearly seen that the liquid range for these ILs is
governed principally by the choice of cation, which has the
greatest influence on the lower melting (or glass) transition
temperature, with the [C4mim]+ salts 6a–g showing the
lowest temperatures of all.
X-ray crystallography: Crystals suitable for X-ray structure
analysis were obtained for six of the anions with tetraalky-
lammonium cations (4a, 3c, 5c, 5d, 4e, and 4 f) by tritura-
tion and slow crystallization from dichloromethane/diethyl
ether (Table 2). All non-hydrogen atoms were fully refined
anisotropically and all hydrogen atoms were refined isotrop-
ically, except for the hydrogen atoms in the disordered 3c.
The six structurally characterized salts are depicted in Fig-
ures 5 and 6 with an ORTEP view of each asymmetric unit
(displayed such that the hydrogen-bonding contact most
under the van der Waals separation is noted) and a packing
diagram.
Figure 4. Liquid range from melting or glass-transition points (open sym-
bols) to onset of decomposition (filled symbols) for (3–6)a–g, showing
the significant changes in thermal range and stability with variation in
cation ([NMe4]+ (3), diamond; [NEt3]+ (4), square; [NBu4]+ (5), triangle;
[C4mim]+ (6), circle) and the relatively small variation with anion.
for each cation (3–6), the melting and decomposition tem-
peratures measured are largely invariant with anion. Two
observations are worth noting, first that all the anions stud-
ied could be used to form low-melting ILs, and that with
each cation type, the 4-nitro-1,2,3-triazolate anion (b) yield-
ed salts with the lowest melting points.
Salt 3c was the only structure to exhibit disorder, and
only one of the cations in this structure is disordered. Three
of the four methyl groups on the disordered cation are rota-
tionally displaced with occupancies of 70:30. As in 4a, the
cations and anions form discrete ionic columns. In 3c the
two positions for the disordered carbon atoms have different
ꢀ
Comparison of the thermal properties of azolate ILs (3–
6)a–g with data for the corresponding bistriflimides, tri-
flates, and tetrafluoroborate salts (Table 1) shows that intro-
duction of the flat heterocyclic azolate anion tends to lead
to salts with melting points that are either comparable to or
lower than those for the bistriflimide salts. The most signifi-
cant depression of melting points was observed for the azo-
close contacts: 2.38 Š for C72 H72B···O32 and 2.50 Š for
ꢀ
C72A H72E···O12 (Table 3).
Structures 3c and 5c were the only structures to crystal-
lize with more than one cation and anion per unit cell. As
mentioned above, one difference in the two cations in 3c is
the disorder, however, there is also a difference in the close
contacts between the ions. The nondisordered cation in 3c
Table 2. Crystallographic data.
4a
3c
5c
5d
4e
4 f
color/shape
formula
Mr
colorless/platelet
C10H20N6O4
288.3
yellow/platelet
C7H13N5O4
231.22
colorless/fragment
C19H37N5O4
399.54
yellow/fragment
C19H37N5O4
399.54
colorless/fragment
C13H21N5
247.35
colorless/fragment
C11H22N4O2
242.33
crystal system orthorhombic
orthorhombic
Pnma
triclinic
monoclinic
P21/c
orthorhombic
Pbcm
monoclinic
P21/n
¯
space group
P212121
(no. 19)
P1
(no. 2)
N
(no. 62)
G
(no. 14)
(no. 57)
(no. 12)
T [K]
a [Š]
b [Š]
c [Š]
a [8]
173
173
173
173
173
173
6.9962(17)
12.148(3)
16.572(4)
90
90
90
1408.5(6)
4
1.360
9.810(5)
19.469(9)
16.745(8)
90
90
90
3198(3)
12
1.441
12.278(4)
13.596(5)
13.628(5)
90.477(6)
95.765(7)
91.929(7)
2262.0(14)
4
12.428(3)
11.967(3)
15.047(4)
90
91.644(4)
90
2237.0(10)
4
1.186
7.902(2)
15.154(5)
12.082(4)
90
90
90
1446.8(8)
4
1.136
14.391(4)
12.516(3)
7.5669(17)
90
106.380(3)
90
1307.6(6)
4
1.231
b [8]
g [8]
V [Š3]
Z
1calcd [gcmꢀ3
]
1.173
independent/ 2021 (Rint = 0.0145)/ 2377 (Rint = 0.0285)/ 6508 (Rint = 0.0188)/ 3216 (Rint = 0.0138)/ 1095 (Rint = 0.0276)/ 1763 (Rint = 0.0393)/
obsd reflns
GooF
R1, wR2
1957 ([I>2s])
1.073
0.0198, 0.0486
1962 ([I>2s)])
1.072
0.0804, 0.2202
4726 ([I>2s])
1.062
0.0525, 0.1445
2871 ([I>2s])
1.033
0.0312, 0.0774
1016 ([I>2s])
1.030
0.0285, 0.0705
865 ([I>2s])
1.018
0.0509, 0.1575
[a]
[I>2s(I)]
[a]
R1, wR2
(all data)
0.0207, 0.0493
0.0909, 0.2325
0.0710, 0.1626
0.0360, 0.0807
0.0311, 0.0725
0.0806, 0.1855
ACHTREUNG
[a] R = ꢀjjFo jꢀjFc jj/ꢀjFo j. R2 = {ꢀ[w(F2oꢀFc2)2]/ꢀ(w(Fo2)2]}1/2
.
4634
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4630 – 4641

Ionic Liquids
FULL PAPER
Figure 5. ORTEP views of the asymmetric units and packing diagrams for tetraethylammonium 3,5-dinitro-1,2,4-triazolate (4a), tetramethylammonium
2,4-dinitroimidazolate (3c), and tetrabutylammonium 2,4-dinitroimidazolate (5c).
has a longer cation–anion shortest contact at 2.62 Š, versus
the 2.38 and 2.50 Š close contacts for the disordered cation.
On close examination of the cations and anions in 5c,
there is little difference in conformation of the individual
cations and anions, but as shown in Figure 5, there is a dif-
ference in the close contacts between the cation and anion
of the two unique IL formula units. The two closest anion/
ꢀ
ꢀ
cation contacts are C91 H91A···O41 (2.62 Š) and C182
H18C···O32 (2.35 Š). This difference in close contacts dis-
rupts any higher-order symmetry that would be possible in
Chem. Eur. J. 2006, 12, 4630 – 4641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4635

A. R. Katritzky, J. D. Holbrey, R. D. Rogers et al.
Table 3. Cation···anion close-contact geometries.[a]
ꢀ
ꢀ
ꢀ
Hydrogen
type
D H···A
D H [Š]
H···A [Š]
ꢀDvdW
D···A [Š]
D H···A [8]
Symmetry code
ꢀ
ꢀ
4a
N CH2CH3
C10 H10B···O2
1.010(16)
0.948(13)
2.408(16)
2.552(12)
ꢀ0.31
ꢀ0.17
3.3952(17)
3.2001(18)
165.5(11)
125.8(9)
ꢀ
C12 H12A···O1
1ꢀx, ꢀ0.5+y, 1.5ꢀz
3c[b]
C72 H72B···O32
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
2.38
2.41
2.49
2.50
2.53
2.56
2.58
2.62
2.62
ꢀ0.34
ꢀ0.31
ꢀ0.23
ꢀ0.22
ꢀ0.19
ꢀ0.16
ꢀ0.14
ꢀ0.11
ꢀ0.11
2.828(11)
3.072(11)
2.983(12)
3.386(6)
3.299(5)
3.440(13)
3.177(11)
3.520(5)
3.486(6)
106.9
124.4
110.8
150.7
134.9
148.9
119.4
152.8
146.8
ꢀ
N CH3
ꢀ
ꢀ
C92 H92B···O32
ꢀx, 1ꢀy, 1ꢀz
ꢀ
C82 H82A···O41
ꢀ0.5ꢀx, 1ꢀy, ꢀ0.5+z
ꢀ1+x, y, z
ꢀ
C72A H72E···O12
ꢀ
C62 H62C···O22
ꢀ
C92 H92B···O12
ꢀ1+x, y, z
ꢀ
C82 H82B···O31
ꢀ1ꢀx, 1ꢀy, 1ꢀz
ꢀ
C81 H81C···O11
ꢀ
C82A H82E···O12
ꢀ1+x, y, z
ꢀ
ꢀ
5c
N CH2-
C182 H18C···O32
0.98(2)
0.98(2)
0.97(2)
0.98(2)
0.98(2)
1.02(2)
1.01(3)
1.00(3)
2.35(2)
2.43(2)
2.45(2)
2.52(2)
2.55(2)
2.65(2)
2.58(3)
2.62(3)
ꢀ0.37
ꢀ0.29
ꢀ0.27
ꢀ0.19
ꢀ0.18
ꢀ0.10
ꢀ0.12
ꢀ0.10
3.325(3)
3.403(3)
3.400(3)
3.459(3)
3.458(3)
3.669(3)
3.411(4)
3.457(3)
174.0(16)
173.3(16)
166.8(16)
158.9(15)
155.6(15)
172.5(15)
140(2)
ꢀ
C142 H14C···O31
x, y, 1+z
ꢀ
C182 H18D···O32
1ꢀx, 1ꢀy, 2ꢀz
ꢀ
C62 H62B···O31
x, 1ꢀy, 1ꢀz
ꢀ
C142 H14D···O31
ꢀ
C101 H10A···N12
1ꢀx, 1ꢀy, 1ꢀz
ꢀx, ꢀy, 1ꢀz
ꢀ
-CH3
C131 H13A···O42
ꢀ
C171 H17C···O41
141.0(19)
ꢀ
ꢀ
5d
N CH2-
C18 H18A···O1
0.961(14)
0.986(13)
0.968(14)
0.972(14)
2.440(15)
2.462(14)
2.466(14)
2.498(14)
ꢀ0.28
ꢀ0.26
ꢀ0.26
ꢀ0.25
3.3311(19)
3.4202(19)
3.3362(18)
3.4550(19)
154.2(11)
164.0(10)
149.5(10)
168.4(10)
ꢀ
C14 H14B···O3
x, 0.5ꢀy, 0.5+z
ꢀ
C6 H6A···O2
ꢀx, ꢀ0.5+y, 1.5ꢀz
ꢀ
C10 H10A···N3
ꢀ
ꢀ
4e
4 f
N CH2CH3
C10 H10B···N3
0.977(12)
0.961(12)
0.987(12)
2.613(12)
2.641(13)
2.650(13)
ꢀ0.14
ꢀ0.12
ꢀ0.11
3.5043(17)
3.5089(18)
3.6013(18)
151.6(8)
150.4(8)
162.0(8)
ꢀ
C8 H8A···N7
2ꢀx, ꢀ0.5+y, 0.5ꢀz
1ꢀx, ꢀ0.5+y, 0.5ꢀz
ꢀ
C10 H10A···N1
ꢀ
ꢀ
N CH2CH3
C12 H12B···N3
0.99(3)
0.90(3)
1.05(3)
1.03(4)
1.08(5)
2.50(3)
2.58(3)
2.54(3)
2.61(4)
2.64(5)
ꢀ0.24
ꢀ0.18
ꢀ0.20
ꢀ0.11
ꢀ0.11
3.443(5)
3.459(6)
3.441(6)
3.354(6)
3.711(7)
158(3)
164(3)
144(2)
129(3)
169(4)
ꢀ
C8 H8B···N3
ꢀ2ꢀx, 2ꢀy, 1ꢀz
ꢀ
ꢀ
N CH2CH3
C11 H11B···O2
ꢀ2.5ꢀx, ꢀ0.5+y, 1.5ꢀz
ꢀ
C11 H11A···O1
ꢀ
C9 H9C···O2
ꢀ3ꢀx, 2ꢀy, 1ꢀz
[a] The contacts included in this table are those at least 0.1 Š under the van der Waals (vdW) separation. [b] The hydrogen atoms of 3c were constrained
due to the presence of disorder.
these highly symmetric ions, resulting in a low-symmetry
(triclinic) space group.
Bond lengths and angles of the azolate anions are shown
in Table 4. A comparison of the imidazolate anions (3c, 4e,
4 f, 5c, and 5d) indicates that there is little effect of substitu-
tion on the bond lengths and angles in the ring.
The nitro substituents in the structures of 4a, 3c, 5c, 5d,
and 4 f exhibit some twisting out of the plane of the ring
(Table 5) ranging from a slight twist in 4a, 3c, 5c, and 4 f to
32.73 and 17.958 in 5d. The observation of larger twist
angles for the nitro groups in 5d is consistent with the steric
hindrance inherent in the 4,5-substituted ring.
Table 4. Bond lengths [Š] and angles [8] in the azolate rings.
4a
3c
5c
5d
4e
4 f
ꢀ
ꢀ
N(1) N(2)
1.3585(16)
1.3326(16)
1.3329(17)
1.3258(16)
1.3323(17)
N(1) C(2)
1.343(5)
1.325(5)
1.349(6)
1.381(6)
1.346(6)
1.340(4)
1.324(4)
1.344(4)
1.386(5)
1.346(5)
1.354(3)
1.324(3)
1.335(3)
1.386(3)
1.337(3)
1.359(3)
1.329(3)
1.336(3)
1.392(3)
1.345(3)
1.348(2)
1.342(2)
1.340(2)
1.3595(19)
1.389(2)
1.358(2)
1.371(4)
1.326(4)
1.360(3)
1.388(4)
1.343(4)
ꢀ
ꢀ
N(2) C(3)
C(2) N(3)
1.3370(19)
1.3438(17)
1.3907(19)
1.3409(18)
ꢀ
ꢀ
C(3) N(4)
N(3) C(4)
ꢀ
ꢀ
N(4) C(5)
C(4) C(5)
ꢀ
ꢀ
C(5) N(1)
C(5) N(1)
N(1)-N(2)-C(3)
N(2)-C(3)-N(4)
C(3)-N(4)-C(5)
N(4)-C(5)-N(1)
C(5)-N(1)-N(2)
103.93(10)
117.10(11)
97.64(10)
117.35(11)
103.98(10)
N(1)-C(2)-N(3)
C(2)-N(3)-C(4)
N(3)-C(4)-C(5)
C(4)-C(5)-N(1)
C(5)-N(1)-C(2)
118.8(4)
99.4(4)
112.3(4)
107.5(4)
102.0(4)
119.2(3)
99.5(3)
111.9(3)
107.8(3)
101.6(3)
118.8(2)
99.59(19)
112.3(2)
107.9(2)
101.40(19)
118.5(2)
99.9(2)
112.4(2)
107.7(2)
101.5(2)
116.91(13)
102.15(11)
109.43(11)
109.63(12)
101.87(12)
116.59(14)
102.64(13)
109.00(13)
109.35(13)
102.42(13)
116.5(3)
101.4(2)
111.2(2)
107.9(2)
103.0(2)
4636
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4630 – 4641

Ionic Liquids
FULL PAPER
Table 5. Torsion angles [8] for the NO2 groups in the anions.
4a
3c
5c
5d
4 f
N(2)-C(3)-N(6)-O(1)
N(2)-C(3)-N(6)-O(2)
N(1)-C(5)-N(7)-O(3)
N(1)-C(5)-N(7)-O(4)
ꢀ5.97(17)
173.53(11)
ꢀ178.67(12)
2.72(17)
N(11)-C(21)-N(61)-O(11)
N(11)-C(21)-N(61)-O(21)
N(31)-C(41)-N(71)-O(31)
N(31)-C(41)-N(71)-O(41)
0.000(1)
ꢀ0.1(4)
180.0
179.5(3)
0.000(1)
ꢀ1.9(4)
180.000(1)
178.2(3)
179.9(2)
ꢀ5.3(3)
0.4(3)
175.6(2)
ꢀ4.7(3)
1.1(3)
N(1)-C(5)-N(7)-O(4)
N(1)-C(5)-N(7)-O(3)
N(3)-C(4)-N(6)-O(1)
N(3)-C(4)-N(6)-O(2)
ꢀ32.73(18)
–
145.36(13)
ꢀ17.95(19)
162.87(12)
–
1.4(6)
174.77(19)
ꢀ178.6(2)
ꢀ178.9(4)
Three related crystal structures of simple nitroazolate
anions were found in the Cambridge Crystallographic Data-
A unique aspect of the anions in these structures is their
potential to form anion–anion hydrogen-bonding interac-
tions, and the salts 3c, 4e, and 4 f exhibit such close contacts
(Table 6). In 3c the interactions are between the C5 hydro-
gen atoms on both anions to a nitro group oxygen of an ad-
jacent anion (Figure 5). The anions in the tetraethylammoni-
um salts 4e and 4 f interact in a head to tail fashion forming
base:[23]
a
sodium 2-nitroimidazolate complex with
[15]crown-5,[24] potassium 4,5-dinitroimidazolate,[25] and 1,8-
bis(dimethylamino)naphthalene 2,4-dinitroimidazolate.[26] In
the 1,8-bis(dimethylamino)naphthalene salt of 2,4-dinitroi-
ꢀ
midazolate (anion c), the C(2) NO2 group of the anion re-
ꢀ
ꢀ
ꢀ
sides in the plane of the ring, however, the C(4) NO2 group
planes with C2 H···N (cyano) or C2 H···O (nitro) close
contacts, respectively (Figure 6).
is twisted out of the anion plane (dihedral angle = 13.618),
whereas the maximum deviation of either nitro group from
the plane of the anion in 5c is 5.628. The anion ring in potas-
sium 4,5-dinitroimidazolate has both nitro groups twisted
out of the heterocycle ring plane, with dihedral angles of
14.0 and 14.318 (compared with 14.13 and 32.198 in 5d).
What is most obvious from the packing diagrams of 5c,
4e, and 4 f (Figures 5 and 6) is that the salts crystallize in
layered structures, forming sheets of anions separated by
layers of tetraethylammonium or tetrabutylammonium cati-
ons that close-packat the van der Waals separation distan-
ces. The cations flatten out into an oblate conformation that
enables maximum space filling within the layers, and a
number of short hydrogen-bonding contacts from the most
polar hydrogen-bond donor positions on the cations, adja-
cent to the hetero atom, to the most basic sites of the
anions.
The anions in 4a, 5c, and 5d have no similar close anion–
anion interactions. The anion in 4a lacks a hydrogen-bond
donor. In the structures 5c and 5d, the larger tetrabutylam-
monium cations appear to more effectively separate the
anions in the lattice.
Conclusion
Challenges facing the preparation of energetic ILs based on
ring-substituted heterocyclic cations result from the destabi-
lization of the aromatic core by substitution with electron-
withdrawing groups (such as nitro or nitrile). In contrast,
the electron-withdrawing characteristics of these same func-
tional groups can be utilized to prepare correspondingly sta-
bilized anionic azolate anions, which leads to opportunities
to form ILs containing heterocyclic anions. A wide range of
heterocyclic anions can be used to prepare ILs and fifteen
such examples have been illustrated here: seven from all
combinations of [C4mim]+ and the seven heterocyclic azo-
late anions studied, six from combinations of tetraethyl- or
tetrabutylammonium cations and anions b, c, and d, and two
additional ionic liquids from the combination of tetrabuty-
lammonium with e and g.
In contrast to the layered structures described above, 5d
forms a close-packed crystalline phase with each ion sur-
rounded by five counterions. The out of plane nitro groups
are good hydrogen-bond acceptors and contribute to rela-
tively short hydrogen bonds with a-hydrogen atoms on the
cations. The tetrabutylammonium cations have extended
ꢀ
alkyl chains, closer to tetrahedral symmetry (terminal CH3
CH3 distances are 8.25ꢂ1.03 Š) relative to the other struc-
tures here.
The smaller tetramethylammonium cation in 3c prevents
efficient formation of an insulating cation layer between
sheets of anions (in contrast to 5c). Instead, a layered struc-
ture with both cations and anions in each layer is observed.
In 4a, the nitrogen and carbon atoms of the triazolate
anion ring are planar, with both nitro groups also in the
plane of the ring. The cations and anions form an ordered
alternating array held together primarily by Coulombic in-
teractions. The three shortest and most significant cation–
It is most interesting that ILs can be readily formed with
N-heterocyclic cations and anions. The [C4mim]+ azolates
are liquid at room temperature (except for the low melting
3,5-nitrotriazolate), with tetraalkylammonium azolates
showing melting points lower than those of the correspond-
ing triflates and tetrafluoroborates, especially in the tetra-
methyl- and tetraethylammonium series, and comparable
with corresponding bistriflimides. Efficient packing, and sub-
sequent crystallization, presumably is suppressed by the in-
troduction of N-alkyl substituents, which block close p–p
stacking of cations and anions. In the structures of the tet-
ꢀ
ꢀ
anion close contacts are C4 H···N3 (2.60 Š), C5 H···O2
ꢀ
(2.41 Š), and C7 H···O1 (2.54 Š).
Chem. Eur. J. 2006, 12, 4630 – 4641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4637

A. R. Katritzky, J. D. Holbrey, R. D. Rogers et al.
Figure 6. ORTEP views of the asymmetric units and packing diagrams for tetrabutylammonium 4,5-dinitroimidazolate (5d), tetraethylammonium 4,5-di-
cyanoimidazolate (4e), and tetraethylammonium 4-nitroimidazolate (4 f).
Table 6. Anion···anion close-contact geometries.[a,b]
Experimental Section
ꢀ
ꢀ
ꢀ
Hydrogen
type
D H···A
D H [Š]
H···A [Š]
ꢀDvdW
D···A [Š]
D H···A [8]
Caution! Although no problems were
encountered during this work, the in-
troduction of energetic functionalities
into heterocyclic compounds has in-
herent risks and appropriate proce-
dures must be taken and all materials
must be handled with extreme care.
ꢀ
ꢀ
3c
C H
C51 H51A···O41
0.97(4)
0.99(4)
0.996(16)
1.01(3)
2.55(4)
2.54(3)
2.420(17)
2.36(3)
ꢀ0.17
ꢀ0.17
ꢀ0.33
ꢀ0.36
3.227(6)
3.236(5)
3.406(2)
3.356(5)
127(3)
127(2)
170.6(12)
172(2)
ꢀ
C52 H52B···O42
ꢀ
ꢀ
4e
4 f
C H
C2 H2A···N6
ꢀ
ꢀ
C H
C2 H2A···O2
[a] The contacts included in this table are those at least 0.1 Š under the van der Waals (vdW) separation.
[b] Compounds 4a, 5c, and 5d have no anion–anion close contacts.
General methods: Melting points
were determined on a hot-stage appa-
ratus and are uncorrected. NMR spectra were recorded in CDCl3 (unless
otherwise stated) at 258C with tetramethylsilane (TMS) as the internal
raalkylammonium salts, flattening of the cations into an
oblate form allows enhanced ion–ion packing.
4638
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4630 – 4641

Ionic Liquids
FULL PAPER
1
standard for H (300 MHz) or the solvent as the internal standard for 13C
Tetramethylammonium 4,5-dinitroimidazolate (3d): Microcrystals from
acetone (98%); m.p. 215–2188C; 1H NMR ([D6]DMSO): d = 3.11 (s,
12H), 6.99 ppm (s, 1H); 13C NMR ([D6]DMSO): d = 54.4 (t, J = 4.0 Hz,
4C), 139.4, 140.3 ppm (2C); elemental analysis calcd (%) for C7H13N5O4
(231.2): C 36.36, H 5.67, N 30.29; found: C 36.51, H 5.71, N 30.46.
(75 MHz). All of the chemicals were employed as supplied.
Materials: Potassium 3,5-dinitro-1,2,4-triazolate (2a) was prepared (in
46% yield) according to the published one-pot procedure by the reaction
of cyanoguanidine with hydrazine followed by diazotization of intermedi-
ate 3,5-diamino-1,2,4-triazole and substitution at the diazonium group.[27]
4-Nitro-1,2,3-triazole (1b) was prepared according to a published proce-
dure from 2-phenyl-1,2,3-triazole by trinitration followed by treatment of
intermediate 2-(2,4-dinitrophenyl)-4-nitro-1,2,3-triazole with sodium
methoxide in methanol under reflux.[28] 2,4-Dinitroimidazole (1c) was
prepared by N-nitration[29,30] of 4-nitroimidazole (1 f) followed by isomer-
ization of the intermediate 1,4-dinitroimidazole in chlorobenzene at
1158C.[31] 4,5-Dinitroimidazole (1d) was prepared by nitration of 4-nitroi-
midazole with fuming nitric acid in concentrated sulfuric acid under
reflux.
Tetramethylammonium 4,5-dicyanoimidazolate (3e): Microcrystals from
acetone (98%); m.p. 197–1998C; 1H NMR ([D6]DMSO): d = 3.11 (s,
12H), 7.31 ppm (s, 1H); 13C NMR ([D6]DMSO): d = 54.4 (t, J = 4.0 Hz,
4C), 116.9, 117.5, 148.8 ppm; elemental analysis calcd (%) for C9H13N5
(191.2): C 56.53, H 6.85, N 36.62; found: C 54.95, H 7.18, N 35.79.
Tetramethylammonium 4-nitroimidazolate (3 f): White microcrystals
(91%); m.p 170–1728C; 1H NMR ([D6]DMSO): d = 3.11 (s, 12H), 7.09
(s, 1H), 7.71 ppm (s, 1H); 13C NMR ([D6]DMSO): d
= 54.4 (t, J =
4.0 Hz, 4C), 131.6, 137.4, 146.0 ppm; elemental analysis calcd (%) for
C7H16N4O3 (186.2): C 41.17, H 7.90, N 27.43; found: C 41.54, H 7.21, N
28.41.
Procedure for the preparation of 4,5-dinitroimidazole (1d): A mixture of
concentrated sulfuric acid (10 mL) and fuming nitric acid (90%, 8 mL,
0.17 mol) was added to 4-nitroimidazole (1 f) (2.0 g, 18 mmol) dissolved
in a minimum quantity of concentrated sulfuric acid. The mixture was
heated under reflux for 5 h. After cooling, the mixture was poured into
ice, and the pH was adjusted to 2 by the addition of sodium bicarbonate.
The product was extracted with ethyl acetate. The extract was concen-
trated to give 4,5-dinitroimidazole (1d) as yellow crystals from water
(1.55 g, 55%). M.p. 166–1698C (ref. [30]: 188–1898C); 1H NMR
Tetramethylammonium tetrazolate (3g): Microcrystals from acetone
(15%); m.p. 216–2188C; 1H NMR ([D6]DMSO): d
= 3.10 (s, 12H),
8.03 ppm (s, 1H); 13C NMR ([D6]DMSO): d = 54.4 (t, J = 4.0 Hz, 4C),
148.6 ppm; elemental analysis calcd (%) for C5H13N5 (143.2): C 41.94, H
9.15, N 48.91; found: C 42.36, H 9.53, N 47.67.
Tetraethylammonium 3,5-dinitro-1,2,4-triazolate (4a): Colorless crystals
from dichloromethane/diethyl ether (94%); m.p. 104–1068C; 1H NMR: d
=
1.40 (tt, J = 7.3, 1.6 Hz, 12H), 3.46 ppm (q, J = 7.3 Hz, 8H);
([D6]DMSO):
([D6]DMSO): d = 135.5, 134.9 ppm.
d =
12.37 (brs, 1H), 8.06 ppm (s, 1H); 13C NMR
13C NMR: d = 7.5 (4C), 52.6 (t, J = 2.9 Hz, 4C), 163.3 ppm (2C); ele-
mental analysis calcd (%) for C10H20N6O4 (288.3): C 41.66, H 6.99, N
29.15; found: C 41.79, H 7.12, N 27.34.
General procedure for the preparation of potassium azolates 2a–d: Po-
tassium carbonate (0.85 g, 6 mmol) was added to a solution of the appro-
priate azole (4 mmol) in acetone (40 mL) (Figure 1 and Scheme 1). The
mixture was stirred at 208C for 6 h, filtered, and the precipitate was
washed with acetone. The solvent was evaporated and the residue was
dried under vacuum to give potassium azolates 2a–d.
Tetraethylammonium 4-nitro-1,2,3-triazolate (4b): Microcrystals from
acetone (99%); m.p. 30–318C; 1H NMR: d = 1.30 (tt, J = 7.3, 1.7 Hz,
12H), 3.31 (q, J = 7.3 Hz, 8H), 8.16 ppm (s, 1H); 13C NMR: d = 7.3
(4C), 52.3 (t, J = 2.9 Hz, 4C), 130.0, 154.5 ppm; elemental analysis calcd
(%) for C10H21N5O2 (243.3): C 49.36, H 8.70, N 28.78; found: C 48.26, H
8.84, N 27.95.
General procedure for the preparation of azolates (3–6)a–d: A solution
of the appropriate tetraalkylammonium or imidazolium halide (3 mmol)
in dichloromethane (30 mL) was added to a solution of the potassium
azolate 2a–d (3 mmol) in acetone (30 mL) at 20–258C. The reaction mix-
ture was stirred for 3 h (24 h for the preparation of tetramethylammoni-
um salts 3a–d). The resultant precipitate of potassium halide was re-
moved by using filtration and the solvent was evaporated under vacuum.
The residue was dried under vacuum. The product was extracted with
acetone, the extract was filtered, and the solvent was removed to give (3–
6)a–d.
Tetraethylammonium 2,4-dinitroimidazolate (4c): Microcrystals (97%);
1
m.p. 76–788C; H NMR: d = 1.36 (tt, J = 7.3, 1.5 Hz, 12H), 3.40 (q, J =
7.3 Hz, 8H), 7.79 ppm (s, 1H); 13C NMR: d = 7.3 (4C), 52.5 (t, J =
2.9 Hz, 4C), 130.7, 147.3, 154.3 ppm; elemental analysis calcd (%) for
C11H21N5O4 (287.3): C 45.98, H 7.37, N 24.37; found: C 45.54, H 7.57, N
23.85.
Tetraethylammonium 4,5-dinitroimidazolate (4d): Microcrystals from
1
acetone (96%); m.p. 101–1038C; H NMR: d = 1.32 (tt, J = 7.3, 1.6 Hz,
12H), 3.28 (q, J = 7.3 Hz, 8H), 7.09 ppm (s, 1H); 13C NMR: d = 7.3
(4C), 52.5 (t, J = 2.9 Hz, 4C), 139.9, 140.1, 140.6 ppm; elemental analysis
calcd (%) for C11H21N5O4 (287.3): C 45.98, H 7.37, N 24.37; found: C
45.78, H 7.57, N 23.60.
General procedure for the preparation of azolates (3–6)e–g: Anhydrous
potassium carbonate (0.7 g, 5 mmol) was added to a solution of the ap-
propriate azole 1e–g (3 mmol) in acetone (30 mL) and the mixture was
stirred for 10 min at 20–258C. Then a solution of the appropriate tetraal-
kylammonium or imidazolium halide (3 mmol) in dichloromethane
(30 mL) was added and the reaction mixture was stirred for 3 h (24 h for
the preparation of tetramethylammonium salts 3e–g). The mixture was
filtered and concentrated under vacuum. The product was extracted with
acetone, the extract was filtered, and the solvent was removed to give (3–
6)e–g.
Tetraethylammonium 4,5-dicyanoimidazolate (4e): Microcrystals from
1
acetone (85%); m.p. 129–1318C; H NMR: d = 1.33 (tt, J = 7.3, 1.8 Hz,
12H), 3.25 (q, J = 7.3 Hz, 8H), 7.46 ppm (s, 1H); 13C NMR: d = 7.4
(4C), 52.5 (t, J = 2.9 Hz, 4C), 117.1, 118.4, 149.2 ppm; elemental analysis
calcd (%) for C13H21N5 (247.3): C 63.13, H 8.56, N 28.31; found: C 63.04,
H 8.79, N 28.32.
Tetraethylammonium 4-nitroimidazolate (4 f): Microcrystals (78%); m.p.
124–1268C; 1H NMR: d = 1.24 (tt, J = 7.3, 1.6 Hz, 12H), 3.18 (q, J =
7.3 Hz, 8H), 7.33 (s, 1H), 7.93 ppm (s, 1H); 13C NMR: d = 7.2 (4C),
52.3 (t, J = 2.9 Hz, 4C), 132.2, 146.7, 148.3 ppm; elemental analysis calcd
(%) for C11H22N4O2 (242.3): C 54.52, H 9.15, N 23.12; found: C 53.43, H
9.27, N 22.68.
Tetramethylammonium 3,5-dinitro-1,2,4-triazolate (3a): Microcrystals
1
from acetone (84%); m.p. 219–2218C; H NMR: d = 3.13 ppm (s, 12H);
13C NMR: d = 54.4 (t, J = 4.0 Hz, 4C), 162.9 ppm (2C); elemental anal-
ysis calcd (%) for C6H12N6O4 (232.2): C 31.04, H 5.21, N 36.19; found: C
31.41, H 5.17, N 36.20.
Tetramethylammonium 4-nitro-1,2,3-triazolate (3b): Microcrystals from
acetone (92%); m.p. 124–1268C; 1H NMR ([D6]DMSO): d = 3.13 (s,
12H), 8.05 ppm (s, 1H); 13C NMR ([D6]DMSO): d = 54.4 (t, J = 4.0 Hz,
4C), 129.5, 154.1 ppm; elemental analysis calcd (%) for C6H13N5O2
(187.2): C 38.50, H 7.00, N 37.41; found: C 38.81, H 7.08, N 37.36.
Tetraethylammonium tetrazolate (4g): Microcrystals from acetone
(81%); m.p. 94–968C; 1H NMR: d = 1.21 (tt, J = 7.3, 1.8 Hz, 12H),
3.12 (q, J = 7.3 Hz, 8H), 8.35 ppm (s, 1H); 13C NMR: d = 7.3 (4C), 52.1
(t, J = 2.9 Hz, 4C), 149.8 ppm; elemental analysis calcd (%) for C9H21N5
(199.3): C 54.24, H 10.62, N 35.14; found: C 53.55, H 11.18, N 33.60.
Tetramethylammonium 2,4-dinitroimidazolate (3c): Microcrystals (83%);
m.p. 178–1808C; 1H NMR ([D6]DMSO): d = 3.13 (s, 12H), 7.75 ppm (s,
1H); 13C NMR ([D6]DMSO): d = 54.4 (t, J = 4.0 Hz, 4C), 130.1, 146.8,
154.1 ppm; elemental analysis calcd (%) for C7H13N5O4 (231.2): C 36.36,
H 5.67, N 30.29; found: C 36.69, H 5.76, N 29.79.
Tetrabutylammonium 3,5-dinitro-1,2,4-triazolate (5a): Microcrystals from
acetone (95%); m.p. 136–1388C; 1H NMR: d = 0.96 (t, J = 7.3 Hz,
12H), 1.39 (sext, J = 7.3 Hz, 8H), 1.65–1.75 (m, 8H), 3.30–3.36 ppm (m,
8H); 13C NMR:
d = 13.4 (4C), 19.6 (4C), 23.8 (4C), 58.8 (4C),
Chem. Eur. J. 2006, 12, 4630 – 4641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4639

A. R. Katritzky, J. D. Holbrey, R. D. Rogers et al.
163.3 ppm (2C); elemental analysis calcd (%) for C18H36N6O4 (400.5): C
53.98, H 9.06, N 20.98; found: C 53.36, H 9.07, N 21.51.
(%) for C11H16N6O4 (296.3): C 44.59, H 5.44, N 28.36; found: C 44.21, H
5.44, N 28.45.
Tetrabutylammonium 4-nitro-1,2,3-triazolate (5b): Microcrystals from
acetone (82%); m.p. 60–618C; 1H NMR: d = 0.97 (t, J = 7.3 Hz, 12H),
1.33–1.45 (m, 8H), 1.55–1.66 (m, 8H), 3.20–3.26 (m, 8H), 8.19 ppm (s,
1H); 13C NMR: d = 13.4 (4C), 19.5 (4C), 23.7 (4C), 58.6 (4C), 130.0,
154.5 ppm; elemental analysis calcd (%) for C18H39N5O3 (355.5): C 57.88,
H 10.52, N 18.75; found: C 58.63, H 10.28, N 20.96.
1-Butyl-3-methylimidazolium 4,5-dicyanoimidazolate (6e): Colorless oil
(98%); 1H NMR: d = 0.94 (t, J = 7.3 Hz, 3H), 1.35 (sext, J = 7.4 Hz,
2H), 1.81–1.91 (m, 2H), 3.98 (s, 3H), 4.19 (t, J = 7.4 Hz, 2H), 7.40 (s,
1H), 7.41 (s, 1H), 7.46 (s, 1H), 9.26 ppm (s, 1H); 13C NMR: d = 13.1,
19.2, 31.6, 36.4, 49.8, 116.8, 118.1, 122.3, 123.6, 135.9, 148.8 ppm; elemen-
tal analysis calcd (%) for C13H16N6 (256.3): C 60.92, H 6.29, N 32.79;
found: C 59.72, H 6.46, N 32.36.
Tetrabutylammonium 2,4-dinitroimidazolate (5c): Microcrystals (84%);
1
1-Butyl-3-methylimidazolium 4-nitroimidazolate (6 f): Colorless oil
(96%); 1H NMR: d = 0.92 (t, J = 7.4 Hz, 3H), 1.32 (sext, J = 7.3 Hz,
2H), 1.81 (quin, J = 7.4 Hz, 2H), 3.94 (s, 3H), 4.17 (t, J = 7.4 Hz, 2H),
7.32–7.34 (m, 3H), 7.36 (s, 1H), 7.96 ppm (s, 1H); 13C NMR: d = 13.2,
19.3, 31.9, 36.1, 49.6, 121.8, 123.3, 131.9, 137.3 (m, 1C), 146.0, 148.3 ppm;
elemental analysis calcd (%) for C11H19N5O3 (251.3): C 49.06, H 7.11, N
26.01; found: C 49.47, H 7.30, N 26.43.
m.p. 79–818C; H NMR: d = 0.93–0.98 (m, 12H), 1.37 (sext, J = 7.3 Hz,
8H), 1.62–1.72 (m, 8H), 3.26–3.31 (m, 8H), 7.80 ppm (s, 1H); 13C NMR:
d = 13.4 (4C), 19.6 (4C), 23.7 (4C), 58.8 (4C), 130.7, 147.3, 154.3 ppm;
elemental analysis calcd (%) for C19H39N5O5 (399.5): C 54.65, H 9.41, N
16.77; found: C 54.64, H 9.10, N 18.19.
Tetrabutylammonium 4,5-dinitroimidazolate (5d): Microcrystals from
acetone (92%); m.p. 64–658C; 1H NMR: d = 0.97 (t, J = 7.3 Hz, 12H),
1-Butyl-3-methylimidazolium tetrazolate (6g): Colorless oil (89%);
1H NMR: d = 0.89 (t, J = 7.3 Hz, 3H), 1.22–1.34 (m, 2H), 1.73–1.83 (m,
2H), 3.91 (s, 3H), 4.15 (t, J = 7.3 Hz, 2H), 7.39 (s, 1H), 7.44 (s, 1H),
8.39 (s, 1H), 9.72 ppm (s, 1H); 13C NMR: d = 13.1, 19.1, 31.7, 36.0, 49.4,
121.9, 123.4, 137.1, 149.7 ppm; elemental analysis calcd (%) for C9H16N6
(208.3): C 51.90, H 7.74, N 40.35; found: C 50.26, H 8.25, N 39.73.
1.38 (sext, J
= 7.2 Hz, 8H), 1.56–1.68 (m, 8H), 3.15–3.21 (m, 8H),
7.08 ppm (s, 1H); 13C NMR: d = 13.4 (4C), 19.5 (4C), 23.7 (4C), 58.6
(4C), 139.9, 140.7, 141.7 ppm; elemental analysis calcd (%) for
C19H39N5O5 (399.5): C 54.65, H 9.41, N 16.77; found: C 54.93, H 9.26, N
18.23.
Tetrabutylammonium 4,5-dicyanoimidazolate (5e): Microcrystals from
acetone (93%); m.p. 69–708C; 1H NMR: d = 1.00 (t, J = 7.3 Hz, 12H),
Analysis: Melting points of the isolated salts were determined by using
differential scanning calorimetry (MDSC) using a TA Instruments model
2920 Modulated differential scanning calorimeter (New Castle, DE)
cooled with a liquid-nitrogen cryostat. The calorimeter was calibrated for
1.41 (sext, J
= 7.3 Hz, 8H), 1.56–1.67 (m, 8H), 3.13–3.19 (m, 8H),
7.45 ppm (s, 1H); 13C NMR: d = 13.4 (4C), 19.5 (4C), 23.7 (4C), 58.7
(4C), 117.1, 118.4, 149.1 ppm; elemental analysis calcd (%) for C21H37N5
(359.6): C 70.15, H 10.37, N 19.48; found: C 70.44, H 10.66, N 20.07.
temperature and cell constants using indium (m.p. 156.618C, DHA
=
28.71 Jgꢀ1). Data were collected at constant atmospheric pressure, using
samples between 10 and 40 mg in aluminum sample pans sealed by using
pin-hole caps. Experiments were performed by heating the samples at a
rate of 58Cminꢀ1. The calorimeter was adjusted so that zero heat flow
was between 0 and ꢀ0.5 mW, and the baseline drift was less than 0.1 mW
over the temperature range 0–1808C. An empty sample pan was used as
reference.
Tetrabutylammonium 4-nitroimidazolate (5 f): Microcrystals (95%); m.p.
103–1058C; 1H NMR: d = 0.97 (t, J = 7.3 Hz, 12H), 1.37 (sext, J =
7.3 Hz, 8H), 1.49–1.60 (m, 8H), 3.08–3.13 (m, 8H), 7.36 (s, 1H),
7.96 ppm (s, 1H); 13C NMR: d = 13.4 (4C), 19.5 (4C), 23.7 (4C), 58.5
(4C), 131.9, 146.4, 148.4 ppm; elemental analysis calcd (%) for
C19H38N4O2 (354.5): C 64.37, H 10.80, N 15.80; found: C 63.33, H 11.06,
N 16.52.
Thermal decomposition temperatures were measured in the dynamic
heating regime using a TGA 2950 TA instrument under argon. Samples
between 2 to 10 mg were heated from 40–5008C under constant heating
1
Tetrabutylammonium tetrazolate (5g): Colorless oil (95%); H NMR: d
= 0.96 (t, J = 7.2 Hz, 12H), 1.30–1.42 (m, 8H), 1.47–1.57 (m, 8H), 3.06–
3.11 (m, 8H), 8.33 ppm (s, 1H); 13C NMR: d = 13.1 (4C), 19.1 (4C),
23.3 (4C), 58.0 (4C), 149.2 ppm; elemental analysis calcd (%) for
C17H37N5 (311.5): C 65.55, H 11.97, N 22.48; found: C 64.66, H 12.92, N
21.67.
at 58Cminꢀ1
.
X-ray crystallographic studies: Samples were recrystallized from metha-
nol by trituration with diethyl ether at 258C. Single crystals suitable for
X-ray analysis were isolated in air, mounted on fibers, and transferred to
the goniometer. The crystals were cooled to ꢀ1008C with a stream of ni-
trogen gas and data were collected on a Siemens SMART diffractometer
with a charge-coupled device (CCD) area detector, using graphite mono-
chromated MoKa radiation. The SHELXTL software (version 5) was used
for solutions and refinements.[32] Absorption corrections were made with
SADABS.[33] Each structure was refined by using full-matrix least-squares
methods on F2.
1-Butyl-3-methylimidazolium 3,5-dinitro-1,2,4-triazolate (6a): Pale yellow
microcrystals (78%); m.p. 35–368C; 1H NMR: d = 0.91 (t, J = 7.4 Hz,
3H), 1.35 (sext, J = 7.3 Hz, 2H), 1.90 (quin, J = 7.4 Hz, 2H), 4.13 (s,
3H), 4.36 (t, J = 7.4 Hz, 2H), 7.56 (s, 1H), 7.61 (s, 1H), 9.72 ppm (s,
= 13.1, 19.2, 31.9, 36.4, 49.8, 122.3, 123.8, 136.9,
163.0 ppm (2C); elemental analysis calcd (%) for C10H15N7O4 (297.3): C
40.80, H 5.09, N 32.98; found: C 40.40, H 5.10, N 32.58.
1H); 13C NMR: d
1-Butyl-3-methylimidazolium 4-nitro-1,2,3-triazolate (6b): Colorless oil
(99%); H NMR: d = 0.90 (t, J = 7.3 Hz, 3H), 1.25–1.38 (m, 2H), 1.78–
In each structure, the atoms were readily located; the positions of all
non-hydrogen atoms were refined anisotropically. The hydrogen atoms
were added in approximated positions and allowed to refine unconstrain-
ed in order to obtain proper close-contact interactions. In only one case
(3c) was disorder observed. The disorder present in the structure was re-
solved with an occupancy of the two conformations of the disordered
methyl group of 70:30. Due to the disorder, the hydrogen atoms of 3c
were added in calculated positions and refined by using a constrained
riding model.
1
1.88 (m, 2H), 4.04 (s, 3H), 4.26 (t, J = 7.3 Hz, 2H), 7.43 (brs, 1H), 7.47
(brs, 1H), 8.19 (s, 1H), 10.01 ppm (s, 1H); 13C NMR: d = 13.1, 19.2,
31.8, 36.2, 49.7, 122.0, 123.4, 130.0, 137.4, 154.6 ppm; elemental analysis
calcd (%) for C10H16N6O2 (252.3): C 47.61, H 6.39, N 33.31; found: C
46.77, H 6.56, N 33.01.
1-Butyl-3-methylimidazolium 2,4-dinitroimidazolate (6c): Colorless oil
(90%); 1H NMR: d = 0.89 (t, J = 7.3 Hz, 3H), 1.32 (sext, J = 7.3 Hz,
2H), 1.81–1.91 (m, 2H), 4.07 (s, 3H), 4.30 (t, J = 7.3 Hz, 2H), 7.52 (t, J
= 1.6 Hz, 1H), 7.56 (t, J = 1.5 Hz, 1H), 7.80 (s, 1H), 9.97 ppm (s, 1H);
13C NMR: d = 13.1, 19.1, 31.8, 36.2, 49.7, 122.1, 123.5, 130.6, 137.3, 147.2,
154.0 ppm; elemental analysis calcd (%) for C11H16N6O4 (296.3): C 44.59,
H 5.44, N 28.36; found: C 44.24, H 5.51, N 28.38.
CCDC-252733 and 278498–278502 contain the supplementary crystallo-
graphic data for salts 4a, 3c, 5c, 5d, 4e, and 4 f, respectively. These data
can be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
1-Butyl-3-methylimidazolium 4,5-dinitroimidazolate (6d): Colorless oil
(95%); 1H NMR ([D6]DMSO): d = 0.90 (t, J = 7.3 Hz, 3H), 1.26 (sext,
J = 7.4 Hz, 2H), 1.77 (quin, J = 7.3 Hz, 2H), 3.86 (s, 3H), 4.17 (t, J =
7.1 Hz, 2H), 6.96 (s, 1H), 7.71 (t, J = 1.6 Hz, 1H), 7.77 (t, J = 1.6 Hz,
1H), 9.13 ppm (s, 1H); 13C NMR ([D6]DMSO): d = 13.2, 18.8, 31.4, 35.7,
48.5, 122.3, 123.6, 136.5, 139.4, 140.3 ppm (2C); elemental analysis calcd
4640
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 4630 – 4641

Ionic Liquids
FULL PAPER
[22] J. D. Holbrey, W. M. Reichert, R. P. Swatloski, G. A. Broker, W. R.
Pitner, K. R. Seddon, R. D. Rogers, Green Chem. 2002, 4, 407.
[23] F. H. Allen, Acta Crystallogr. Sect. B 2002, 58, 380–388.
[24] P. Duncanson, D. V. Griffiths, J. R. Miller, Tetrahedron 1997, 53,
13177.
Acknowledgements
This research was supported by the Air Force Office of Scientific Re-
search (grant F49620 03 1 0357).
[25] A. Kowalski, Z. Kristallogr. 1993, 208, 244.
[26] T. Glowiak, Z. Malarski, L. Sobczyk, E. Grech, J. Mol. Struct. 1987,
157, 329.
[27] V. M. Chernyshev, N. D. Zemlyakov, V. B. Ilꢀin, V. A. Taranushich,
Russ. J. Appl. Chem. 2000, 73, 839; [Zh. Prikl. Khim. (Russ.) 2000,
73(5), 791].
[28] T. E. Eagles, M. A. Khan, B. M. Lynch, Org. Prep. Proced. 1970, 2,
117.
[29] S. S. Novikov, L. I. Khmelꢀnitskii, O. V. Lebedev, V. V. Sevastꢀyano-
va, L. V. Epishina, Chem. Heterocycl. Compd. 1970, 6, 465.
[30] M. R. Grimmett, S.-T. Hua, K.-C. Chang, S. A. Foley, J. Simpson,
Aust. J. Chem. 1989, 42, 1281.
[31] S. Bulusu, R. Damavarapu, J. R. Autera, R. Behrens, L. M. Minier,
J. Villanueva, K. Jayasuriya, T. Axenrod, J. Phys. Chem. 1995, 99,
5009.
[32] G. M. Sheldrick, SHELXTL, version 5.05, Siemens Analytical X-ray
Instruments Inc., 1996.
[33] G. M. Sheldrick, Program for Semiempirical Absorption Correction
of Area Detector Data, University of Gçttingen, Gçttingen (Ger-
many), 1996.
[34] J. Sun, M. Forsyth, D. R. MacFarlane, J. Phys. Chem. B 1998, 102,
8858.
[1] M. T. Carter, C. L. Hussey, S. K. D. Strubinger, R. A. Osteryoung,
Inorg. Chem. 1991, 30(5), 1149.
[2] T. Welton, Chem. Rev. 1999, 99, 2071.
[3] J. D. Holbrey, K. R. Seddon, Clean Prod. Proc. 1999, 1, 223.
[4] P. Wasserscheid, W. Keim, Angew. Chem. 2000, 112, 3926; Angew.
Chem. Int. Ed. 2000, 39, 3772.
[5] R. Sheldon, Chem. Commun. 2001, 2399.
[6] C. M. Gordon, Appl. Catal. A 2001, 222, 101.
[7] H. Olivier-Bourbigou, L. Magna, J. Mol. Catal. A 2002, 182–183,
419.
[8] J. Dupont, R. F. de Souza, P. A. Z. Suarez, Chem. Rev. 2002, 102,
3667.
[9] K. N. Marsh, J. A. Boxall, R. Lichtenthaler, Fluid Phase Equilib.
2004, 219, 93.
[10] S. A. Forsyth, J. M. Pringle, D. R. MacFarlane, Aust. J. Chem. 2004,
57, 113.
[11] T. M. Klapçtke, G. Holl, Green Chem. 2001, 3, G75.
[12] G. Drake, T. Hawkins, A. Brand, L. Hall, M. Mckay, A. Vij, I.
Ismail, Propellants, Explos., Pyrotech. 2003, 28, 174.
[13] Y. X. Ou, B. R. Chen, J. R. Li, S. Dong, J. J. Li, H. P. Jia, Heterocy-
cles 1994, 38, 1651.
[14] A. R. Katritzky, H. Yang, D. Zhang, K. Kirichenko, J. D. Holbrey,
W. M. Reichert, M. Smiglak, R. D. Rogers, New J. Chem. 2006, 30,
349.
[35] P. Bonhꢂte, A.-P. Dias, M. Armand, N. Papageorgiou, K. Kalyana-
sundaram, M. Grꢃtzel, Inorg. Chem. 1996, 35, 1168.
[36] W. Xu, L.-M. Wang, R. A. Nieman, C. A. Angell, J. Phys. Chem. B
2003, 107, 11749.
[37] S. J. Kubisen, F. H. Westheimer, J. Am. Chem. Soc. 1979, 101, 5985.
[38] K. Rousseau, G. C. Farrington, D. Dolphin, J. Org. Chem. 1972, 37,
3968.
[15] H.-J. Laas, R. Halpaap, F. Richter, J. Kocher, US Pat. App. Pub.
20030204041, 2003.
[16] W. Ogihara, M. Yoshizawa, H. Ohno, Chem. Lett. 2004, 33, 1022.
[17] H. Ohno, M. Yoshizawa, W. Ogiwara, H. Ogawa, H. Taguchi, Jpn.
Pat. 2004331521, 2004; [Chem. Abstr. 2004, 141, 424197].
[18] M. Stearcey, P. L. Pye, J. B. Lee, Synth. Commun. 1989, 19, 1309.
[19] A. R. Katritzky, S. Singh, K. Kirichenko, J. D. Holbrey, M. Smiglak,
W. M. Reichert, R. D. Rogers, Chem. Commun. 2005, 868.
[20] H. Xue, Y. Gao, B. Twamley, J. M. Shreeve, Inorg. Chem. 2005, 44,
5068.
[39] K. Xu, M. S. Ding, T. R. Jow, J. Electrochem. Soc. 2001, 148, A267.
[40] J. D. Holbrey, K. R. Seddon, J. Chem. Soc. Dalton Trans. 1999, 2133.
Received: July 18, 2005
Revised: January 23, 2006
Published online: April 4, 2006
[21] C. Ye, J. Xiao, B. Twamley, J. M. Shreeve, Chem. Commun. 2005,
2750.
Chem. Eur. J. 2006, 12, 4630 – 4641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4641