Ionic liquids based on diethylmethyl(2-methoxyethyl)ammonium cations and bis(perfluoroalkanesulfonyl)amide anions: Influence of anion structure on liquid properties

Article abstract of DOI:10.1039/c1cp21783f

A series of diethylmethyl(2-methoxyethyl)ammonium (DEME)-based ionic liquids were prepared using bis(perfluoroalkanesulfonyl)amide (C _nF_2n+1SO₂)₂N anions with different perfluoroalkyl chain lengths (n = 0, 1, 2, 3, and 4), and the influence of the structural variation on their thermal, ion-diffusive (ionic conductivity and viscosity), ion-concentration (molar concentration and ion association), and solvatochromic (polarity and hydrogen-bond acceptor ability) properties was investigated. The elongation of the perfluoroalkyl chain causes the pronounced suppression of ionic conductivity, fluidity, and polarity. According to the crystallographic study of the corresponding (C_nF _2n+1SO₂)₂N salts formed with high-symmetrical tetramethylammonium cations, the decreased ion diffusivity must be a consequence of the increased contribution of the interionic van der Waals interactions of F...F type and hydrogen-bonding interactions of C-H...F type in addition to C-H...O type. The Kamlet-Taft π*-scale (polarity) is under the control of the ion concentration, associated with the perfluoroalkyl chain length in the anions. The larger Kamlet-Taft β-scale (hydrogen-bond acceptor ability) of DEME-based ionic liquids with a longer perfluoroalkyl chain appears to be responsible for the larger degree of ion association of oppositely charged ions, which was manifested in the Walden rule deviation.

Full text of DOI:10.1039/c1cp21783f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
PCCP
Cite this: Phys. Chem. Chem. Phys., 2011, 13, 20302–20310
www.rsc.org/pccp
PAPER
Ionic liquids based on diethylmethyl(2-methoxyethyl)ammonium cations
and bis(perfluoroalkanesulfonyl)amide anions: influence of anion structure
on liquid propertiesw
Yukihiro Yoshida* and Gunzi Saito
Received 1st June 2011, Accepted 9th September 2011
DOI: 10.1039/c1cp21783f
A series of diethylmethyl(2-methoxyethyl)ammonium (DEME)-based ionic liquids were prepared
using bis(perfluoroalkanesulfonyl)amide (CnF2n+1SO2)2N anions with different perfluoroalkyl
chain lengths (n = 0, 1, 2, 3, and 4), and the influence of the structural variation on their
thermal, ion-diffusive (ionic conductivity and viscosity), ion-concentration (molar concentration
and ion association), and solvatochromic (polarity and hydrogen-bond acceptor ability) properties
was investigated. The elongation of the perfluoroalkyl chain causes the pronounced suppression
of ionic conductivity, fluidity, and polarity. According to the crystallographic study of the
corresponding (CnF2n+1SO2)2N salts formed with high-symmetrical tetramethylammonium
cations, the decreased ion diffusivity must be a consequence of the increased contribution of the
interionic van der Waals interactions of Fꢀ ꢀ ꢀF type and hydrogen-bonding interactions of
C–Hꢀ ꢀ ꢀF type in addition to C–Hꢀ ꢀ ꢀO type. The Kamlet–Taft p*-scale (polarity) is under the
control of the ion concentration, associated with the perfluoroalkyl chain length in the anions.
The larger Kamlet–Taft b-scale (hydrogen-bond acceptor ability) of DEME-based ionic liquids
with a longer perfluoroalkyl chain appears to be responsible for the larger degree of ion
association of oppositely charged ions, which was manifested in the Walden rule deviation.
for dye-sensitized solar cells.4,5 In this context, the (CF3SO2)2N
Introduction
anion, with largely delocalized charge6 and flexibility,7 gives
Ionic liquids, which are entirely composed of ions and melt
thermally and electrochemically stable hydrophobic ionic liquids
with a relatively high ionic conductivity.2a,8 Considering the
fact that (CH3SO2)2N-based ionic liquids have lower ionic
conductivity and fluidity (reciprocal of viscosity) than those
of the corresponding (CF3SO2)2N-based ionic liquids,9 it is
immediately apparent that the fluorination plays an impor-
tant role in reducing the interionic interactions. Among the
(CF3SO2)2N-based ionic liquids, the combination with a
diethylmethyl(2-methoxyethyl)ammonium (DEME; Scheme 1)
cation is currently attracting considerable attention as a
potentially benign solvent for high-energy devices such as
electrochemical capacitors,8 lithium batteries,10 and field-effect
transistors.11 However, studies on fundamental properties of
[DEME][(CF3SO2)2N] (or DEME-1,1; in this paper, we use an
abbreviation ‘‘DEME-n,m’’ to represent the ionic liquids com-
posed of DEME cations and (CnF2n+1SO2)(CmF2m+1SO2)N
anions) and its analogues are limited to thermal and electro-
chemical stabilities, viscosity, and ionic conductivity for
DEME-1,1,12 and thermal and electrochemical stabilities for
[DEME][(FSO2)2N] (DEME-0,0), [DEME][(FSO2)(CF3SO2)N]
(DEME-0,1) and [DEME][(CF3SO2)(C2F5SO2)N] (DEME-1,2).13
As part of our interest to explore new ionic liquids, we herein
systematically report the synthesis and characterization of a
below room temperature (RT) or 100 1C, appear to be one of
the most grown functional materials in chemistry. A special
fascination of the ionic liquids is that the selection of compo-
nent ions can tailor the liquid properties such as melting point,
thermal and electrochemical stabilities, miscibility, solubility,
polarity, viscosity, and ionic conductivity, over a wide range.1
In particular, it has been known that the sort of anions exerts
a drastic effect on the liquid properties. For example, the
combination with PF6 or (CF3SO2)2N anion gives the hydro-
phobic ionic liquids,2 whereas CH3COO or EtOSO3 anion
stabilizes the liquid state even at low temperatures.3 These
properties could find a range of synthetic and separation
applications. Highly conducting ionic liquids formed with
N(CN)2, C(CN)3, or B(CN)4 anion have served as electrolytes
Research Institute, Meijo University, Shiogamaguchi 1-501 Tempaku-ku,
Nagoya 468-8502, Japan. E-mail: yyoshida@meijo-u.ac.jp;
Fax: +81-52-833-7200; Tel: +81-52-838-2552
w Electronic supplementary information (ESI) available: Numerical
data of ionic conductivity and viscosity of DEME-n,n as a function of
temperature, and the best-fit parameters of the Litovitz equation.
CCDC 808116–808118. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c1cp21783f
c
20302 Phys. Chem. Chem. Phys., 2011, 13, 20302–20310
This journal is the Owner Societies 2011

View Article Online
although no endothermic peak corresponding to the melting
event was observed for DEME-0,0 in contrast to the earlier
report.13 Whereas DEME-1,1 also exhibits no melting event,
DEME-2,2 exhibits an exothermic crystallization peak at ꢁ61 1C
followed by an endothermic melting peak at ꢁ40 1C. DEME
salts with n Z 2 exhibit a melting event, and the Tm value
increases with the elongation of the perfluoroalkyl chain in the
anions. This trend allows us to assert that the melting event in
these ionic liquids is under the control of the vdW interaction
between the perfluoroalkyl chains. In relation, as reported
for several (FSO2)2N- and (CF3SO2)2N-based ionic liquids,15
DEME-3,3 and DEME-4,4 exhibit several successive endo-
thermic peaks, which correspond to solid–solid phase transition(s)
and melting point. The observed solid–solid transition can
be ascribed to the conformational transition of anions.16 TGA
traces revealed that the present DEME-based ionic liquids
have an excellent thermal stability up to 370 1C, except for
DEME-0,0 with Td = 253 1C, which provides a wide liquid
region over 350 1C.
Scheme 1 Main component ions present in this text.
series of ionic liquids combining bis(perfluoroalkanesulfonyl)-
amide ((CnF2n+1SO2)2N; Scheme 1) anions having different
perfluoroalkyl chain lengths (n = 0, 1, 2, 3, and 4) with DEME
cations. To elucidate the interionic interactions between
ammonium cations and amide anions from crystallographic
viewpoints, we examined the crystal structure determination
of (CnF2n+1SO2)2N-based salts (n = 0, 1, and 2) formed
with tetramethylammonium (TMA) cations (abbreviated as
‘‘TMA-n’’; Scheme 1) instead of DEME cations.14 The high-
symmetrical TMA cation facilitates the crystallographic study
of the (CnF2n+1SO2)2N-based salts due to the stabilization of
the solid state. Elucidation of the type of interionic forces
(Coulomb, hydrogen-bonding, and van der Waals (vdW)) and
the influence of the perfluoroalkyl chain length on their liquid
properties will provide a valuable design guide for forth-
coming selection and synthesis of ionic liquids as solvents
for electrochemical devices.
2. Structural properties
The crystallographic data of three TMA-n salts (n = 0, 1, and 2)
at 0 1C are summarized in footnote z. We note that the S–N
bond lengths of the anions in the crystals (1.459–1.645 A) are
significantly shorter than typical S–N single bonds (1.74 A17), as
a consequence of the delocalized negative charge within the
S–N–S backbone.
Colorless rhombic plate crystal TMA-0 with a Tm of 72 1C
crystallizes in monoclinic lattice, and a unit cell contains two
ion pairs. Nitrogen atoms of both ions locate on a special
position, and half of the TMA cations and half of the (FSO2)2N
anions are crystallographically independent. The (FSO2)2N
anion with disordered F(1) and O(2) atoms has a cis conforma-
tion as observed for M[(FSO2)2N] (M = Li+, K+, and Cs+).18
The salt shows alternating two-dimensional sheets composed
of cations and anions along the bc plane (Fig. 1a). This seems
to be a consequence of the attractive Coulomb interactions
between oppositely charged ions with similar vdW volumes
(TMA: 93.9 A3, (FSO2)2N: 95.3 A3), estimated based on an
ab initio restricted Hartree–Fock (RHF) method with the 6-31G*
basis set.19 Namely, four TMA cations surround a (FSO2)2N
anion in a tetragonal fashion through C–HꢀꢀꢀO hydrogen bond-
ing with the distance being 3.306–3.589 A (Fig. 1b), which is
Results and discussion
1. Thermal properties
DEME salts prepared in this study are hydrophobic transparent
liquids at RT. Table 1 summarizes glass transition (Tg), crystal-
lization (Tc), solid–solid transition (Ts–s), and melting (Tm)
temperatures determined by the analysis of differential scanning
calorimetry (DSC) traces during the heating scan (10 K minꢁ1),
as well as decomposition temperatures (Td) determined by the
thermogravimetric analysis (TGA) during the heating scan
(10 K minꢁ1). The low-temperature phase of the DEME salts
with n r 2 is in the glass state. The Tg values of DEME-0,0 and
DEME-1,1 are quite similar to those reported previously,12,13
z Crystal data for [TMA][(FSO2)2N] (TMA-0): C4H12F2N2O4S2. M =
254.28, monoclinic, space group P21/m (#11), a = 5.8107(5) A,
b = 11.132 (1) A, c = 8.4816(7) A, b = 90.683(1)1, V = 548.61(8) A3,
Z = 2, dcalc = 1.539 g cmꢁ3, m(Mo Ka) = 0.507 mmꢁ1, F(000) = 264,
1392 independent reflections, 91 refined parameters, Rgt = 0.0443 (for I 4
2s(I)), wR2 = 0.1247 (for all data), GOF = 1.061. Crystal data for
[TMA][(CF3SO2)2N] (TMA-1): C6H12F6N2O4S2. M = 354.29, triclinic,
Table 1 Thermal properties of [DEME][(CnF2n+1SO2)2N]
n
Tga/1C
Tca/1C
Ts–sa/1C
Tma/1C
Tdb/1C
0c
1d
2
ꢁ111
ꢁ93
ꢁ93
253
382
376
373
374
%
space group P1 (#2), a = 8.343(2) A, b = 12.798(3) A, c = 13.793(4) A,
a = 91.360(4)1, b = 90.644(4)1, g = 93.478(4)1, V = 1469.5(7) A3, Z = 4,
ꢁ61
ꢁ40
14
dcalc = 1.601 g cmꢁ3, m(Mo Ka) = 0.441 mmꢁ1, F(000) = 720, 5529
3
8
independent reflections, 538 refined parameters, Rgt
= 0.0803
4
ꢁ75, ꢁ65, ꢁ16, 12
18
(for I 4 2s(I)), wR2 = 0.2577 (for all data), GOF = 1.020. Crystal
data for [TMA][(C2F5SO2)2N] (TMA-2): C8H12F10N2O4S2. M =
454.31, monoclinic, space group P21/n (#14), a = 10.1328(8) A, b =
17.324(1) A, c = 10.2870(8) A, b = 105.360(1)1, V = 1741.3(2) A3,
Z = 4, dcalc = 1.733 g cmꢁ3, m(Mo Ka) = 0.424 mmꢁ1, F(000) = 912,
4208 independent reflections, 239 refined parameters, Rgt = 0.0556
(for I 4 2s(I)), wR2 = 0.1658 (for all data), GOF = 1.030.
a
Onset temperatures of a heat capacity change (Tg), an exothermic
peak (Tc), and an endothermic peak (Ts–s, Tm) determined by DSC
b
on heating process. Temperatures of 10% weight loss determined
c
by TGA on heating process. Literature values: Tg = ꢁ112 1C, Tm
=
ꢁ22 1C.13 d Literature value: Tg = ꢁ95 1C.12
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 20302–20310 20303

View Article Online
sites at around (0, 1/4, 1/2). It should be noted that, in contrast
to TMA-0, both F and O atoms in anions are responsible for
the hydrogen-bonding-type interactions with cations (Fig. 2a).
Based on the ab initio calculation,19 elongation of the per-
fluoroalkyl chain has little impact on the electron density of
oxygen atoms (ꢁ0.66e to ꢁ0.64e for cis-(FSO2)2N and ꢁ0.68e
to ꢁ0.67e for trans-(CF3SO2)2N). Therefore, the appearance
of C–Hꢀ ꢀ ꢀF hydrogen bonding would stem from the sterically-
bulky trifluoromethyl groups of the (CF3SO2)2N anion,
although the electron density of each fluorine atom decreases
with the elongation of the perfluoroalkyl chain length (ꢁ0.43e
for cis-(FSO2)2N and ꢁ0.35e to ꢁ0.34e for trans-(CF3SO2)2N).
The anions in the x = 0.5 plane are tetragonally surrounded by
four TMA cations, while the anions at the x = 0 plane are
surrounded by eight cations in a CsCl-type manner. Such a
molecular packing also indicates that the attractive Coulomb
interactions between oppositely charged ions are a factor deter-
mining the crystal packing. Notably, some short Fꢀ ꢀ ꢀF atomic
contacts (2.622–2.910 A vs. vdW distance: 2.94 A20) between
anions were observed in contrast to TMA-0.
Fig. 1 (a) Crystal packing of [TMA][(FSO2)2N] viewed along the a
axis (red: TMA cation, blue: (FSO2)2N anion). (b) Interionic contacts
between TMA cation and (FSO2)2N anion in [TMA][(FSO2)2N],
where short C–Hꢀ ꢀ ꢀO contacts are shown by dashed lines. In both
figures, one of the disordered F(1) and O(2) atoms in (FSO2)2N anions
are drawn.
significantly shorter than the sum of vdW radii (3.72 A20). The
anion serves as a bi-dentate ligand in the interionic contacts
along the c direction and a mono-dentate ligand along the b
direction. Notably, there is no short C–HꢀꢀꢀF hydrogen-bonding
interaction, which would be associated with the smaller electron
density of fluorine atoms (ꢁ0.43e) than those of oxygen atoms
(ꢁ0.66e to ꢁ0.64e) calculated based on the ab initio method.19
It has been known that the (CF3SO2)2N anion is highly flexible
and has two low-energy conformers of C1 (cis) and C2 (trans)
The (C2F5SO2)2N anion is also highly flexible with two low-
energy C1 (cis) and C2 (trans) conformations, the latter being
more stable by 5.4–7.3 kJ molꢁ1 22
Colorless elongate plate
.
crystal TMA-2 with a Tm of 130 1C belongs to the monoclinic
system, and the asymmetric component is comprised of one
ion pair. The (C2F5SO2)2N anion is fully ordered in a transoid
configuration (torsion angle C–S–S–C: 174.61) as seen in Fig. 3a.
Fig. 3b shows the crystal structure viewed along the a axis.
The (C2F5SO2)2N anions produce a herring bone-like arrange-
ment at x E 0.25 and 0.75 planes, and the TMA cations locate
the interstitial sites within the planes. TMA-2 also has signi-
ficant C–Hꢀ ꢀ ꢀF and C–Hꢀ ꢀ ꢀO hydrogen bonding between
cations and anions (C–Hꢀ ꢀ ꢀF: 3.235–3.658 A and C–Hꢀ ꢀ ꢀO:
3.266–3.720 A; Fig. 3c), and consequently, each anion is
surrounded by six cations in a distorted octahedral manner.
symmetries,7,16 the latter being more stable by 2.3 kJ molꢁ1 7a
.
Colorless elongate plate crystal TMA-1 with a Tm of 134 1C
has four ion pairs in a unit cell, and two full TMA cations and
two half and one full (CF3SO2)2N anions are crystallographi-
cally independent. All anions have orientationally disordered
F and O atoms, and adopt a trans configuration (torsion
angles C–S–S–C: 178.1–180.01; Fig. 2a), as observed for most
of the (CF3SO2)2N-based salts with organic cations.21 In the
crystal, there are two unique anionic layers at x = 0 and 0.5
(Fig. 2b), and the (CF3SO2)2N anions in one layer are oriented
in a nearly perpendicular geometry with respect to the anions
in another layer (Fig. 2c). TMA cations locate the interstitial
Fig. 2 (a) Typical interionic contacts between TMA cation and
(CF3SO2)2N anion in [TMA][(CF3SO2)2N], where short C–Hꢀ ꢀ ꢀF
and C–Hꢀ ꢀ ꢀO contacts are shown by dashed lines. Crystal packing
of [TMA][(CF3SO2)2N] viewed along the b axis (b) and the a axis (c).
TMA cations and (CF3SO2)2N anions are indicated in red and blue,
respectively. Deep-colored cations and anions are located at y E 0.75
and 0, respectively, while pale-colored cations and anions are located
at y E 0.25 and 0.5, respectively. In both figures, one of the disordered
F and O atoms in (CF3SO2)2N anions are drawn.
Fig. 3 (a) Molecular structure of (C2F5SO2)2N anion in
[TMA][(C2F5SO2)2N]. (b) Crystal packing of [TMA][(C2F5SO2)2N]
viewed along the a axis. TMA cations and (C2F5SO2)2N anions are
indicated in red and blue, respectively. Deep-colored and pale-colored
ions are located at x E 0.75 and 0.25, respectively. (c) Typical
interionic contacts between TMA cation and (C2F5SO2)2N anion in
[TMA][(C2F5SO2)2N], where short C–Hꢀ ꢀ ꢀF and C–Hꢀ ꢀ ꢀO contacts
are shown by dashed lines.
c
20304 Phys. Chem. Chem. Phys., 2011, 13, 20302–20310
This journal is the Owner Societies 2011

View Article Online
In the case of TMA-1, a short Fꢀ ꢀ ꢀF atomic contact (2.946 A)
between anions, which is comparable to the corresponding
vdW distance (2.94 A20), was observed.
From the above crystallographic study, it appears that
hydrogen-bonding interaction of C–Hꢀ ꢀ ꢀF type between
cations and anions and vdW interaction of Fꢀ ꢀ ꢀF type between
anions become marked as the perfluoroalkyl chain in the anions
elongates. Such interionic interactions must have considerable
impact on the ion-diffusive properties in ionic medium of the
(CnF2n+1SO2)2N-based ionic liquids, although DEME cations
are of larger size (169.7 A3) than TMA cations (93.9 A3).
Fig. 4 Temperature dependence of ionic conductivity (s) for
[DEME][(CnF2n+1SO2)2N] with n = 0 (J), n = 1 (n), n = 2 (,),
n = 3 (&), and n = 4 (ꢂ).
3. Ion-diffusive properties
The present DEME-n,n show a semicircle-like impedance Cole–
Cole plot in the whole temperature region (25–80 1C). The RT
ionic conductivity s decreases as the perfluoroalkyl chain in
the anions elongates (Table 2 and Fig. 4), and the s value of
DEME-1,1 (2.7 ꢂ 10ꢁ3 S cmꢁ1 at 25 1C) is similar to that
reported previously (2.6 ꢂ 10ꢁ3 S cmꢁ1 at 25 1C).12 Such an
elongation must cause the depressed Coulomb cohesive energy
due to the interionic separation, which exerts an effect in ways
that lead to the increased diffusivity. However, this is not the case
for the present ionic liquids, and it is thus most likely that the
cationꢀꢀꢀanion hydrogen-bonding attraction and anionꢀꢀꢀanion
vdW attraction as confirmed by the crystallographic study
of TMA-n are factors governing the ion diffusivity. Mole-
cular dynamics simulations, which predicted the formation of
nanometre-sized nonpolar domains of alkyl groups in the ionic
liquids,23 also support that the observed liquid properties,
especially ion diffusivity, are more under the control of the
vdW interactions than of the Coulomb interactions.
where R, z, and F are the universal gas constant, the charge of
ions, and the Faraday constant, respectively, the D value at
25 1C decreases with the elongation of the perfluoroalkyl chain
in the anions in going from DEME-0,0 (3.2 ꢂ 10ꢁ7 cm2 sꢁ1) to
DEME-4,4 (9.4 ꢂ 10ꢁ9 cm2 sꢁ1). However, it should be noted
that these values were underestimated since the C value used
for eqn (1) is just the number of ion pairs in a unit volume. In
the next section, we will discuss the actual number of charged
components in a unit volume of DEME-n,n.
Irrespective of ion association (vide infra), the viscosity Z of
ionic liquids is an important issue to serve as solvents for
chemical and electrochemical processes, because of its impact
on the mass transport. As seen in Fig. 5, the viscosity steadily
increases in going from DEME-0,0 (39.5 mPa s at 25 1C) to
DEME-4,4 (739 mPa s at 25 1C).26 The anion effect in the
present work also indicates that the ionic conduction described
above can be qualitatively explained based on the Stokes–
Einstein equation (eqn (2)) for the relationship between the ion
diffusivity of component ions and the viscosity of the liquid,
connected with the degree of ion dissociation (0 r y r 1)
Arrhenius plots of s show a slight upward curvature as depicted
in Fig. 4. Since the measured temperature range (25–80 1C) is
far from their Tg, the temperature dependencies were fitted
the two-parameter Litovitz equation,24 s = A exp(ꢁB/T3), on
account of interionic hydrogen-bonding interactions, instead of
the three-parameter Vogel–Fulcher–Tammann (VFT) equation,25
s = A exp(ꢁB/(T ꢁ T0)) with an ideal glass transition tem-
perature (T0), where A and B, as well as T0 are fitting parameters
(see ESIw). According to the Nernst–Einstein equation (eqn (1))
for the relationship between the ion diffusivity (D) of charged
components including ion aggregates/clusters in ionic liquids
and their molar conductivity (L = s/C; molar concentration C
was estimated simply from d/M, where d and M are measured
density and formula weight, respectively)
D = yRT(rꢁc 1 + raꢁ1)/fpNAZ
(2)
where f is a constant (4 r f r 6; f = 4 for perfect slip and
f = 6 for perfect stick), NA is the Avogadro number, and rc
and ra are cation and anion radii, respectively. It is immedi-
ately apparent that the increase in both the Z and ra values
with the elongation of the perfluoroalkyl chain in the anions
leads to the depressed ion diffusivity.
Temperature dependence of Z of the present DEME-n,n,
as depicted in Fig. 5, can also fit the Litovitz equation, Z =
Aexp(B/T3), over the measured temperature range (see ESIw).
These results clearly indicate that both the ionic conductivity
D = RTL/z2F2
(1)
Table 2 Physical properties of [DEME][(CnF2n+1SO2)2N]a
n
s/10ꢁ3 S cmꢁ1
Z/mPa s
d/g cmꢁ3
C/10ꢁ3 mol cmꢁ3
L/S cm2 molꢁ1
LZ/S cm2 mPa s molꢁ1
C0/10ꢁ3 mol cmꢁ3
0
4.9
2.7
0.62
0.22
0.076
39.5
64.0
180
447
739c
1.34
1.42
1.48
1.56
1.59
4.10
3.33
2.81
2.49
2.19
1.2
47
52
40
39
26c
1.9
1.7
1.1
0.97
0.57
1b
2
0.82
0.22
0.086
0.035
3
4
a
s: ionic conductivity at 25 1C, Z: viscosity at 25 1C, d: density at 25 1C, C: molar concentration at 25 1C, L: molar conductivity at 25 1C,
C0: effective ion concentration at 25 1C (see the text). Literature values: s = 2.6 ꢂ 10ꢁ3 S cmꢁ1, Z = 69 mPa s, d = 1.42 g cmꢁ3 12 c
.
The Z value
b
was evaluated on the basis of the Litovitz equation.24,26
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 20302–20310 20305

View Article Online
both of which seem to be a dominant factor determining
the interionic interactions. Ionic conductivity and fluidity of
[C2MI][(CnF2n+1SO2)2N] are significantly high in comparison
with those of [DEME][(CnF2n+1SO2)2N], as a consequence of
the largely delocalized charge on C2MI cations. However, it
would be premature to compare the alkyl chain length depen-
dencies given that there are a small number of data points for
[C2MI][(CnF2n+1SO2)2N].
4. Effective ion concentration
The molar concentration C decreases steadily in going from
DEME-0,0 (4.10 ꢂ 10ꢁ3 mol cmꢁ3) to DEME-4,4 (2.19 ꢂ
10ꢁ3 mol cmꢁ3), while the d value increases as the perfluoro-
alkyl chain in the anions elongates, as seen in Table 2.
Fig. 5 Temperature dependence of viscosity (Z) for [DEME]-
[(CnF2n+1SO2)2N] with n = 0 (J), n = 1 (n), n = 2 (,), n = 3
(&), and n = 4 (ꢂ).
It has been found that the empirical Walden rule, namely
that the molar conductivity L is inversely proportional to the
viscosity Z of the liquid29 based on a wide range of aqueous
solution systems, can apply to the ionic liquid system.4b–d,30–33
According to the rule, the discrepancy from the Walden
product LZ of infinitely diluted KCl aqueous solution
(100 S cm2 mPa s molꢁ1), in which the component ions with
nearly equal ion sizes are completely dissociated (y = 1), is
strongly indicative of the existence of a significant fraction of
association of oppositely charged ions,30 which accounts for
the reduced ionic conductivity at a given viscosity.34 Here we
compare the Walden products for the present DEME-n,n at
25 1C. The Walden plot, as depicted in Fig. 7, demonstrates
that all of the present DEME-n,n reside below the ‘‘ideal’’
and viscosity of the present DEME-n,n could be finely tuned in
a wide range, by simply varying the perfluoroalkyl chain length.
Because one of the attractive features of ionic liquids is the
controllability of liquid properties, it is important to compare
the alkyl chain length dependence of such properties with
those of related salts. In Fig. 6, we replot the ionic conductivity
and viscosity of DEME-n,n (red closed circles), together with
reported [DEME][BF3(CnF2n+1)] (Scheme 1; Fig. 6, blue open
circles),12 [C2MI][(CnF2n+1SO2)2N] (C2MI: 1-ethyl-3-methyl-
imidazolium, Scheme 1; Fig. 6, red open triangles),2a,13,27 and
[CnMI][(CF3SO2)2N] (CnMI: 1-alkyl-3-methylimidazolium,
Scheme 1; Fig. 6, green open squares)28 for comparison,
against the number of carbon atoms in alkyl or perfluoroalkyl
chain(s) (n0; for instance, the value of [DEME][(C2F5SO2)2N]
(DEME-2,2) is 4 as the anion has two pentafluoroethyl groups).
Unexpectedly low ion diffusivity of [DEME][BF4], [DEME]-
[BF3CF3], and [C1MI][(CF3SO2)2N] (indicated by black arrows
in Fig. 6) would arise from the relatively high symmetry of BF4,
BF3CF3, and C1MI ions, respectively. Except for them, the
variations of DEME-n,n (red closed circles) are pronounced in
comparison with those of [DEME][BF3(CnF2n+1)] (blue open
circles) and [CnMI][(CF3SO2)2N] (green open squares). It is
thus apparent that the ion diffusivity of the present DEME-n,n
system can be controlled by modifying the alkyl chain length more
effectively than [DEME][BF3(CnF2n+1)] and [CnMI][(CF3SO2)2N]
systems. The little chain length effect might be caused by the
presence of the BF3 moiety in BF3(CnF2n+1) anions and the
hydrogen atom at the 2-position in CnMI cations, respectively,
Walden line; namely the LZ values (26–47 S cm2 mPa s molꢁ1
)
are apparently less than that of the ideal Walden product.
Moreover, it decreases with the elongation of the perfluoro-
alkyl chains in the anions (inset of Fig. 7). Although the LZ value
is originally anticipated to vary in proportion to (rꢁc 1 + raꢁ1) as
the combination of eqn (1) and (2), the difference in (rꢁc 1 + raꢁ1),
being 0.644 Aꢁ1 for DEME-0,0 and 0.507 Aꢁ1 for DEME-4,4,35
is not large enough to account for a 45% decrease in LZ
going from DEME-0,0 (47 S cm2 mPa s cmꢁ1) to DEME-4,4
(26 S cm2 mPa s cmꢁ1). A similar alkyl chain length effect on
LZ was seen earlier for cationic variations of [CnMI][FeX4]
(X: Cl, Br)32b and [CnMI][N(CN)2] salts by us,4c and [CnMI]-
[(CF3SO2)2N] salts by Watanabe and coworkers,28,36 and
Fig. 6 Alkyl or perfluoroalkyl chain length (n0) dependence of (a)
Fig. 7 A plot of the molar conductivity against the reciprocal of visco-
sity for [DEME][(CnF2n+1SO2)2N]. The dotted line indicates the ideal
Walden line (see the text), and the numbers are the perfluoroalkyl chain
length (n) in the anions. Inset is the Walden product as a function of n.
The dotted line indicates the ideal Walden product (see the text).
ionic conductivity and (b) viscosity of [DEME][(CnF2n+1SO2)2N]
( ; n = 2n), [DEME][BF3(Cn F2n +1)] ( ),12 [C2MI][(CnF2n+1SO2)2N]
0
0
0
0
(
; n0 = 2n),2a,13,27 and [Cn MI][(CF3SO2)2N] ( ).28 For black arrows,
see the text.
c
20306 Phys. Chem. Chem. Phys., 2011, 13, 20302–20310
This journal is the Owner Societies 2011

View Article Online
anionic variations of [C2MI][CnH2n+1BF3] by Zhou and
coworkers.37
Table 3 Kamlet–Taft parameters of [DEME][(CnF2n+1SO2)2N]
Ionic liquid
p*
b
As mentioned above, the molecular dynamics simulations
predicted the heterogeneous phase separation between the
polar domains composed of anions and charged groups of
the cations and the nonpolar domains of alkyl groups,23 and
such a phase separation would promote the ion association of
oppositely charged ions. Liquid-state X-ray diffraction study
confirmed the formation of nonpolar domains through vdW
attractions between the alkyl groups.38 Although there has
been no evidence for the formation of nonpolar domains of
perfluoroalkyl groups in ionic liquids, the present solid-state
diffraction study, which revealed the increasing importance
of the vdW attractions of Fꢀ ꢀ ꢀF type with the elongation of
the perfluoroalkyl chain in the anions, suggests a possible
aggregation of perfluoroalkyl groups in DEME-n,n. Assuming
that the ratio LZ(in S cm2 mPa s molꢁ1)/100 corresponds to the
y value,1e,36 the product yC gives the effective ion concentration
C0 (Table 2), which steadily decreases in going from DEME-0,0
(1.9 ꢂ 10ꢁ3 mol cmꢁ3) to DEME-4,4 (5.7 ꢂ 10ꢁ4 mol cmꢁ3).
We note that these values must be closer to the actual ion
concentration (the number of pairs of charged components in
a unit volume). According to these values, the actual D values for
charged components were estimated to be 6.9 ꢂ 10ꢁ7 cm2 sꢁ1 for
DEME-0,0, 4.2 ꢂ 10ꢁ7 cm2 sꢁ1 for DEME-1,1, 1.5 ꢂ 10ꢁ7 cm2 sꢁ1
for DEME-2,2, 6.1 ꢂ 10ꢁ8 cm2 sꢁ1 for DEME-3,3, and 3.7 ꢂ
10ꢁ8 cm2 sꢁ1 for DEME-4,4, based on eqn (1).
DEME-0,0
DEME-1,1
DEME-2,2
DEME-3,3
DEME-4,4
[C2MI][(CF3SO2)2N]
[C4MI][(CF3SO2)2N]
1.018
0.972
0.915
0.886
0.857
0.970a, 0.980c
0.964a, 0.984b, 0.971c
0.195
0.263
0.277
0.297
0.317
0.233c
0.243b, 0.248c
a
b
Ref. 36. Ref. 39e. Ref. 42.
c
Fig. 8 Dependence of Kamlet–Taft p* parameter of [DEME]-
[(CnF2n+1SO2)2N] on effective ion concentration (C0). The numbers
are the perfluoroalkyl chain length (n) in the anions. The dotted line
through the points is a guide to the eye.
that the parameters are anion-driven. As seen in Fig. 8, the p*
value steadily increases with increasing the effective ion concen-
tration C0 (= yC). The linear variation clearly indicates that the
dipolarity/polarizability of the ionic liquids can be controlled by
changing the perfluoroalkyl chain length, based on the fact that
the elongation results in the decreased degree of ion dissociation
(y) and the molar concentration (C).4d,39h The elongation of
the perfluoroalkyl chain also results in the increased b value, in
going from DEME-0,0 (0.195) to DEME-4,4 (0.317). We note that
the b values of DEME-0,0 and DEME-4,4 are comparable to those
of water (0.14) and acetonitrile (0.370), respectively, both of which
are commonly-used solvents for electrochemical devices such as
dye-sensitized solar cells, electrochemical capacitors, and lithium
batteries. As described above, the appearance of hydrogen bonding
of C–HꢀꢀꢀF type in addition to C–HꢀꢀꢀO one would affect the
HBA ability in the same sense as the ion-association property.
5. Solvatochromic properties
So far, a wide range of solvatochromic probes have been used
to evaluate the polarity of ionic liquids,2a,39 by comparison
with well-established empirical solvent polarity scales based on
molecular liquids. In this study, we estimate the dimensionless
p* and b values, which were initially defined by Kamlet, Abboud,
and Taft,40 since it has been known that these polarity scales
depend largely on the anions. The polarity scale of dipolarity/
polarizability characteristics p* can be readily estimated on the
basis of the intramolecular CT transition energy of N,N-diethyl-
p-nitroaniline (DENA) with both electron-withdrawing (e.g.
NO2) and -donating (e.g. NEt2) groups. The p* value is normal-
ized by taking cyclohexane (p* = 0.00) and dimethyl sulfoxide
(p* = 1.00) by definition, as shown in eqn (3),41
p* = 8.649 ꢁ 0.314nDENA (in 103 cmꢁ1
)
(3)
Conclusions
where nDENA is the wavelength maximum of the absorption
band. On the other hand, the Kamlet–Taft b value is more
diagnostic for the hydrogen-bond acceptor (HBA) ability of
anions in ionic liquids, and is related to the nNA and p* values
through eqn (4),41
In this study, our research interest is centered on the effects of
the structural variation of (CnF2n+1SO2)2N anions on liquid
properties of the DEME-based ionic liquids, and we found
that the elongation of the perfluoroalkyl chain exerts a drastic
effect on the properties such as ion-diffusive, ion-concentration,
and polar properties. Since the crystallographic data for
[TMA][(CnF2n+1SO2)2N] confirm the increasing importance
of the hydrogen-bonding interaction of C–Hꢀ ꢀ ꢀF type and vdW
interaction of Fꢀ ꢀ ꢀF type with the elongation of the perfluoro-
alkyl chain in the anions, it is possible that these interactions are
dominantly correlated in a way that leads to diminished ion
diffusivity in liquid media with weak interionic Coulomb inter-
actions. In the same manner, the ion-association of oppositely
charged ions is under the control of the hydrogen-bonding
b = ꢁ0.357nNA (in 103 cmꢁ1) ꢁ 1.176p* + 11.12
(4)
where nNA is the wavelength maximum of the intramolecular
CT absorption band of p-nitroaniline (NA).
Table 3 presents the p* and b values of the present
DEME-n,n, together with those of [C2MI][(CF3SO2)2N] and
[C4MI][(CF3SO2)2N].36,39e,42 Both polarity scales of DEME-1,1
are similar to those of the C2MI and C4MI salts formed with
(CF3SO2)2N anions, providing us with further confirmation
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 20302–20310 20307

View Article Online
interactions between neighboring cations and anions. Ionic
conductivity (7.6 ꢂ 10ꢁ5–4.9 ꢂ 10ꢁ3 S cmꢁ1 at 25 1C), viscosity
(40–739 mPa s at 25 1C), and effective ion concentration (5.6 ꢂ
10ꢁ4–1.9 ꢂ 10ꢁ3 mol cmꢁ3 at 25 1C) of the present ionic liquids
cover a very wide range by varying the perfluoroalkyl chain
length (0 r n r 4). The present finding will be beneficial to the
design, selection, and blend of ionic liquids (and their compo-
nent ions) as ‘‘designer solvents’’ for many applications.
Synthesis
Solvents (water, methanol, ethanol, and acetone) were distilled
prior to use. Colorless crystalline solid [DEME]Br was synthe-
sized by the Menschutkin reaction of diethylmethylamine
(Tokyo Kasei) with 2-bromoethyl methyl ether (Tokyo Kasei)
in water according to the literature procedure,12 followed by
the decolorization by activated charcoal powder in dry methanol,
recrystallization twice from dry acetone, and dried in vacuo at 80 1C
for 2 days. K[(FSO2)2N] (Kanto Chemical), Li[(CF3SO2)2N]
(Aldrich), Li[(C2F5SO2)2N] (Kishida Chemical), Li[(C3F7SO2)2N]
(Wako Chemical), and Li[(C4F9SO2)2N] (Wako Chemical)
were used without purification. Colorless plate-shaped crystals
[TMA][(CnF2n+1SO2)2N] (n = 0, 1, and 2) were synthesized
according to the literature procedure for n = 0 salt,45 and
recrystallized twice from water for TMA-0 and TMA-1, and
ethanol/H2O (1 : 2) for TMA-2 (see ESIw). p-Nitroaniline
(NA) and N,N-diethyl-p-nitroaniline (DENA) were obtained
from Wako Chemical and used without purification.
Experimental section
Measurements
All measurements were carried out under ambient atmosphere,
since the present DEME- and TMA-based salts are highly
hydrophobic. 1H NMR measurements were conducted on a
JEOL ECA-600 spectrometer operating at 600 MHz, and
acetone-d6 was used as solvent. FT-IR spectra were taken in
compressed KBr pellets with a Perkin-Elmer 1000 Series
spectrophotometer. EDS experiments were conducted with a
JEOL JSM-5510LVN Scanning Electron Microscope operated
at 20 kV. The samples were mounted on carbon tape. DSC
thermograms were measured on a Shimadzu DSC-60 instru-
ment equipped with nitrogen cryostatic cooling. The samples
were heated to 80 1C followed by cooling to ꢁ130 1C (n = 0) or
ꢁ110 1C (n Z 1) and then heated to 400 1C at a cooling and
heating rate of 10 K minꢁ1. The glass transition (Tg), crystal-
lization (Tc), solid–solid transition (Ts–s), and melting (Tm)
temperatures were determined from the DSC traces during
the heating scans. Decomposition temperature (Td) was deter-
mined by thermogravimetric analyses (from RT to 600 1C at a
heating rate of 10 K minꢁ1) under a nitrogen atmosphere on a
Shimadzu DTG-60M instrument. Conductance measurements
were performed in a two platinum electrode conductivity cell
(cell constant is 35.0 cmꢁ1). The conductance of the samples
was measured using a Wayne Kerr impedance analyzer 6440B
over a frequency range of 20 Hz to 3 MHz. Viscosities were
measured using a cone-type Tokyo Keiki RE-80L rotational
viscometer. Density values were obtained by measuring the
weight of the sample in a 1 cm3 pycnometer. UV-Vis absorption
spectra of NA and DENA dissolved in ionic liquids (ca. 10ꢁ3 M)
were taken in a quartz cell with a light path length of 1 mm on a
Shimadzu UV-3100 spectrophotometer (190–800 nm).
[DEME][(FSO)2N] (DEME-0,0)
A slight excess of K[(FSO)2N] (5.74 g, 26.2 mmol) and
[DEME]Br (5.37 g, 23.8 mmol) was dissolved in distilled water
(50 cm3). The resultant suspension was stirred for 2 days under
dark conditions at RT, followed by repeatedly washing with
water. The resultant transparent liquid was finally dehydrated
under vacuum at 80 1C for 2 days (5.39 g, 16.5 mmol, 70%
yield), and was stored under an inert atmosphere of argon gas
in a glovebox (H2O, O2 o 1 ppm). IR (KBr) nmax: 434w, 455w,
483w, 573s, 642w, 753m, 832m, 872w, 951w, 1027w, 1107m,
1182s, 1219m, 1261w, 1362w, 1383s, 2828w, 2901w, 2945w,
2997w cmꢁ1. H NMR (600 MHz, acetone-d6) d: 1.46 (t, J =
1
7.2 Hz, 2 ꢂ 3H), 3.26 (s, 3H), 3.42 (s, 3H), 3.65 (q, J = 7.6 Hz,
2 ꢂ 2H), 3.74 (t, J = 4.8 Hz, 2H), 3.95 (s, 2H) ppm. Anal.
calcd for C8H20F2N2O5S2: C, 29.44; H, 6.18; N, 8.58%.
Found: C, 29.33; H, 6.16; N, 8.57%. No traces of K and Br
were detected for energy-dispersive X-ray spectroscopy (EDS).
[DEME][(CF3SO)2N] (DEME-1,1)
Transparent liquid was prepared by the procedure described
above for [DEME][(FSO)2N] except that Li[(CF3SO)2N] was
used instead of K[(FSO)2N]. Yield: 97%. IR (KBr) nmax
513m, 571m, 602w, 618m, 654w, 741w, 763w, 790w, 872w,
:
X-Ray diffraction data for [TMA][(CnF2n+1SO2)2N] (n = 0, 1,
and 2) were collected on a CCD type diffractometer (Bruker
SMART APEX II) with graphite monochromated Mo Ka
radiation at 0 1C. The absorption correction was carried out
using the program SADABS in Bruker APEX2 software. The
crystal structure was solved by a direct method of SIR200443
and refined by a full matrix least-squares method of F2 by
means of SHELXL.44 The positional parameters of hydrogen
atoms were calculated under a fixed C–H bond length of
0.96 A for the methylene group with an sp3 configuration of
the bonding carbon atoms. CCDC-808116 for TMA-0, 808117
for TMA-1, and 808118 for TMA-2.z Elemental analyses
(C, H, and N) were carried out at the Center for Organic
Elemental Microanalysis of Kyoto University.
951w, 1058s, 1139m, 1197s, 1228w, 1332w, 1352s, 1399w,
2830w, 2904w, 2946w, 2998w cmꢁ1 1H NMR (600 MHz,
.
acetone-d6) d: 1.44 (t, J = 7.2 Hz, 2 ꢂ 3H), 3.24 (s, 3H), 3.39
(s, 3H), 3.63 (q, J = 7.6 Hz, 2 ꢂ 2H), 3.72 (t, J = 4.8 Hz, 2H),
3.93 (s, 2H) ppm. Anal. calcd for C10H20F6N2O5S2: C, 28.18;
H, 4.73; N, 6.57%. Found: C, 28.02; H, 4.60; N, 6.59%. No
trace of Br was detected for EDS. This procedure is essentially
the same as that reported previously.12
[DEME][(C2F5SO)2N] (DEME-2,2)
Transparent liquid was prepared by the procedure described
above for [DEME][(FSO)2N] except that Li[(C2F5SO)2N] was
used instead of K[(FSO)2N]. Yield: 94%. IR (KBr) nmax: 525w,
536w, 617m, 640w, 740w, 756w, 777w, 810w, 873w, 979m, 992w,
1028w, 1087m, 1176s, 1226s, 1332m, 1355s, 1400w, 2830w,
Geometry optimizations of TMA and DEME cations and
(CnF2n+1SO2)2N anions (n = 0 and 1) were performed by
RHF/6-31G* calculations using HyperChem 8.0.4.19
1
2900w, 2950w, 2999w cmꢁ1. H NMR (600 MHz, acetone-d6)
c
20308 Phys. Chem. Chem. Phys., 2011, 13, 20302–20310
This journal is the Owner Societies 2011

View Article Online
d: 1.44 (t, J = 7.2 Hz, 2 ꢂ 3H), 3.24 (s, 3H), 3.39 (s, 3H), 3.63
(q, J = 7.4 Hz, 2 ꢂ 2H), 3.72 (t, J = 4.8 Hz, 2H), 3.93 (s, 2H)
ppm. Anal. calcd for C12H20F10N2O5S2: C, 27.38; H, 3.83; N,
5.32%. Found: C, 27.41; H, 3.69; N, 5.37%. No trace of Br was
detected for EDS.
R. D. Rogers, Green Chem., 2002, 4, 407–413; (c) E. F. Borra,
O. Seddiki, R. Angel, D. Eisenstein, P. Hickson, K. R. Seddon and
S. P. Worden, Nature, 2007, 447, 979–981.
4 (a) D. R. MacFarland, J. Golding, S. Forsyth, M. Forsyth and
G. B. Deacon, Chem. Commun., 2001, 1430–1431; (b) Y. Yoshida,
K. Muroi, A. Otsuka, G. Saito, M. Takahashi and T. Yoko, Inorg.
Chem., 2004, 43, 1458–1462; (c) Y. Yoshida, O. Baba and G. Saito,
J. Phys. Chem. B, 2007, 111, 4742–4749; (d) Y. Yoshida,
O. Baba, C. Larriba and G. Saito, J. Phys. Chem. B, 2007, 111,
12204–12210.
[DEME][(C3F7SO)2N] (DEME-3,3)
Transparent liquid was prepared by the procedure described
above for [DEME][(FSO)2N] except that Li[(C3F7SO)2N] was
5 (a) P. Wang, S. M. Zakeeruddin, J.-E. Moser and M. Gratzel,
¨
J. Phys. Chem. B, 2003, 107, 13280–13285; (b) R. Kawano,
H. Matsui, C. Matsuyama, A. Sato, M. A. B. H. Susan, N. Tanabe
and M. Watanabe, J. Photochem. Photobiol., A, 2004, 164, 87–92;
(c) P. Wang, B. Wenger, R. Humphry-Baker, J.-E. Moser,
J. Teuscher, W. Kantlehner, J. Mezger, E. V. Stoyanov, S. M.
used instead of K[(FSO)2N]. Yield: 98%. IR (KBr) nmax
:
517w, 528w, 541w, 575m, 630m, 655w, 682w, 728w, 735w,
747w, 753w, 765w, 810w, 866m, 945w, 1028w, 1052w, 1071m,
1111m, 1172s, 1200m, 1217m, 1234m, 1283m, 1339m, 1356s,
Zakeeruddin and M. Gra
6850–6856; (d) D. Kuang, P. Wang, S. Ito, S. M. Zakeeruddin and
M. Gratzel, J. Am. Chem. Soc., 2006, 128, 7732–7733.
¨
tzel, J. Am. Chem. Soc., 2005, 127,
1
1400w, 2831w, 2852w, 2904w, 2947w, 2999w cmꢁ1. H NMR
(600 MHz, acetone-d6) d: 1.43 (t, J = 7.5 Hz, 2 ꢂ 3H), 3.23
(s, 3H), 3.39 (s, 3H), 3.63 (q, J = 7.4 Hz, 2 ꢂ 2H), 3.71 (t, J =
4.8 Hz, 2H), 3.92 (s, 2H) ppm. Anal. calcd for C14H20F14N2O5S2:
C, 26.84; H, 3.22; N, 4.47%. Found: C, 26.88; H, 3.14; N, 4.48%.
No trace of Br was detected for EDS.
¨
6 D. Benrabah, R. Arnaud and J.-Y. Sanchez, Electrochim. Acta,
1995, 40, 2437–2443.
7 (a) J. Foropoulos Jr. and D. D. DesMarteau, Inorg. Chem., 1984,
23, 3720–3723; (b) R. Arnaud, D. Benrabah and J.-Y. Sanchez,
J. Phys. Chem., 1996, 100, 10882–10891; (c) P. Johansson,
S. P. Gejji, J. Tegenfeldt and J. Lindgren, Electrochim. Acta,
1998, 43, 1375–1379; (d) J. D. Holbrey, W. M. Reichert and R. D.
Rogers, Dalton Trans., 2004, 2267–2271.
[DEME][(C4F9SO)2N] (DEME-4,4)
8 T. Sato, G. Masuda and K. Takagi, Electrochim. Acta, 2004, 49,
3603–3611.
9 J. M. Pringle, J. Golding, K. Baranyai, C. M. Forsyth, G. B. Deacon,
J. L. Scott and D. R. MacFarlane, New J. Chem., 2003, 27,
1504–1510.
10 (a) T. Sato, T. Maruo, S. Marukane and K. Takagi, J. Power Sources,
2004, 138, 253–261; (b) S. Seki, Y. Kobayashi, H. Miyashiro,
Y. Ohno, Y. Mita, A. Usami, N. Terada and M. Watanabe,
Transparent liquid was prepared by the procedure described
above for [DEME][(FSO)2N] except that Li[(C4F9SO)2N] was
used instead of K[(FSO)2N]. Yield: 99%. IR (KBr) nmax
:
476w, 516m, 532w, 576w, 592m, 615w, 651w, 684w, 696w,
737w, 747w, 757w, 803w, 845w, 859w, 874w, 953w, 1010w,
1030w, 1076m, 1140m, 1172m, 1200m, 1218m, 1238m, 1288w,
1334w, 1355s, 1400w, 2831w, 2903w, 2948w, 3001w cmꢁ1
.
Electrochem.
Solid-State
Lett.,
2005,
8,
A577–A578;
1H NMR (600 MHz, acetone-d6) d: 1.44 (t, J = 7.2 Hz, 2 ꢂ 3H),
3.24 (s, 3H), 3.39 (s, 3H), 3.63 (q, J = 7.4 Hz, 2 ꢂ 2H), 3.72
(t, J = 4.8 Hz, 2H), 3.93 (s, 2H) ppm. Anal. calcd for
C16H20F18N2O5S2: C, 26.45; H, 2.77; N, 3.86%. Found: C,
26.01; H, 2.67; N, 3.92%. No trace of Br was detected
for EDS.
(c) Y. Kobayashi, Y. Mita, S. Seki, Y. Ohno, H. Miyashiro and
N. Terada, J. Electrochem. Soc., 2007, 154, A677–A681.
11 (a) T. Fujimoto, M. M. Matsushita and K. Awaga, Chem. Phys.
Lett., 2009, 483, 81–83; (b) H. Yuan, H. Shimotani, A. Tsukazaki,
A. Ohtomo, M. Kawasaki and Y. Iwasa, J. Am. Chem. Soc., 2010,
132, 6672–6678; (c) J. T. Ye, S. Inoue, K. Kobayashi, Y. Kasahara,
H. T. Yuan, H. Shimotani and Y. Iwasa, Nat. Mater., 2010, 9,
125–128.
12 Z.-B. Zhou, H. Matsumoto and K. Tatsumi, Chem.–Eur. J., 2005,
11, 752–766.
Acknowledgements
13 H. Matsumoto, N. Terasawa, T. Umecky, S. Tsuzuki, H. Sakaebe,
K. Asaka and K. Tatsumi, Chem. Lett., 2008, 1020–1021.
14 At present, structural refinement of [TMA][(C3F7SO2)2N] (TMA-3;
Tm = 122 1C) was not completed due to the orientational disorder of
heptafluoropropyl groups. However, the following lattice parameters
were determined; monoclinic C2/c (#15), a = 14.579(3) A, b =
11.454(3) A, c = 12.418(3) A, b = 97.270(3)1, V = 2057.0(8) A3,
Z = 4. Half of the TMA cations and (C3F7SO2)2N anions are
The authors are grateful to Dr Takaaki Hiramatsu and
Prof. Hideki Yamochi (Kyoto University) for technical sup-
1
port for H NMR and X-ray diffraction measurements. Y.Y.
also acknowledges the financial support of Grants-in-Aid for
Young Scientists (No. 20750107, 22685015) from Japan
Society for the Promotion of Science (JSPS).
crystallographically independent, and anions adopt a transoid
configuration as (CF3SO2)2N and (C2F5SO2)2N anions in the TMA
salts.
Notes and references
15 (a) D. R. MacFarlane, P. Meakin, J. Sun, N. Amini and M. Forsyth,
J. Phys. Chem. B, 1999, 103, 4164–4170; (b) H. Tokuda, K. Ishii,
M. A. B. H. Susan, S. Tsuzuki, K. Hayamizu and M. Watanabe,
J. Phys. Chem. B, 2006, 110, 2833–2839; (c) Q. Zhou, W. A.
Henderson, G. B. Appetecchi, M. Montanino and S. Passerini,
J. Phys. Chem. B, 2008, 112, 13577–13580; (d) A. Fernicola,
F. C. Weise, S. G. Greenbaum, J. Kagimoto, B. Scrosati and
A. Soleto, J. Electrochem. Soc., 2009, 156, A514–A520.
16 (a) M. Herstedt, W. A. Henderson, M. Smirnov, L. Ducasse,
L. Servant, D. Talaga and J. C. Lassegues, J. Mol. Struct., 2006,
783, 145–156; (b) W. A. Henderson, M. Herstedt, V. G. Young Jr.,
S. Passerini, H. C. De Long and P. C. Trulove, Inorg. Chem., 2006,
45, 1412–1414.
1 For selected reviews, see: (a) Ionic Liquids in Synthesis,
ed. P. Wasserscheid and T. Welton, Wiley-VCH, Weinheim,
2003; (b) P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed.,
2000, 39, 3772–3789; (c) M. C. Buzzeo, R. G. Evans and R. G.
¨
Compton, ChemPhysChem, 2004, 5, 1106–1120; (d) H. Weingartner,
Angew. Chem., Int. Ed., 2008, 47, 654–670; (e) K. Ueno, H. Tokuda
and M. Watanabe, Phys. Chem. Chem. Phys., 2010, 12, 1649–1658;
(f) Y. Yoshida and G. Saito, Phys. Chem. Chem. Phys., 2010, 12,
1675–1684.
2 (a) P. Bonhote, A.-P. Dias, M. Armand, N. Papageorgiou,
K. Kalyanasundaram and M. Gratzel, Inorg. Chem., 1996, 35,
¨
1168–1178; (b) P. A. Z. Suarez, S. Einloft, J. E. L. Dullius, R. F. de
Souza and J. Dupont, J. Chim. Phys. Phys.-Chim. Biol., 1998, 95,
1626–1639.
17 L. Pauling, The Nature of the Chemical Bond, Cornell University
Press, New York, 3rd edn, 1960.
18 (a) O. Hiemisch, D. Henschel, A. Blaschette and P. G. Jones,
Z. Anorg. Allg. Chem., 1997, 623, 324–332; (b) B. Krumm, A. Vij,
R. L. Kirchmeier and J. M. Shreeve, Inorg. Chem., 1998, 37,
3 (a) J. S. Wilkes and M. J. Zaworotko, J. Chem. Soc., Chem.
Commun., 1992, 965–966; (b) J. D. Holbrey, W. M. Reichert,
R. P. Swatloski, G. A. Broker, W. R. Pitner, K. R. Seddon and
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 20302–20310 20309

View Article Online
6295–6303; (c) M. Beran, J. Prı
´
Polyhedron, 2006, 25, 1292–1298.
19 HyperChem Professional 8.0, Hypercube, Inc., Gainesville, FL,
2007.
20 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
hoda, Z. Za
´
k and M. Cernı
´
k,
the value less than unity clearly means that the deviation from the
Walden line of the infinitely diluted KCl aqueous solution by
elongating the perfluoroalkyl chain in the anions becomes more
marked when used the recent data.
35 The rc and ra values are evaluated by (3Vc/4p)1/3 and (3Va/4p)1/3
,
21 (a) K. Matsumoto and R. Hagiwara, J. Fluorine Chem., 2007, 128,
317–331 and references therein; (b) J. A. Schlueter, U. Geiser,
H. H. Wang, A. M. Kini, B. H. Ward, J. P. Parakka, R. G.
Daugherty, M. E. Kelly, P. G. Nixon, R. W. Winter, G. L. Gard,
L. K. Montgomery, H.-J. Koo and M.-H. Whangbo, J. Solid State
Chem., 2002, 168, 524–534.
22 P. Johansson, J. Tegenfeldt and J. Lindgren, J. Phys. Chem. A,
2000, 104, 954–961.
23 (a) Y. Wang and G. A. Voth, J. Am. Chem. Soc., 2005, 127,
12192–12193; (b) J. N. Canongia Lopes, M. F. Costa Gomes and
A. A. H. Padua, J. Phys. Chem. B, 2006, 110, 16816–16818.
´
where the vdW volumes Vc for cations and Va for anions are
calculated based on the ab initio restricted Hartree–Fock (RHF)
method with the 6-31G* basis set.19Vc of DEME: 169.7 A3, Va of
(FSO2)2N: 95.4 A3, and Va of (CF3SO2)2N: 148.6 A3. For the
anions with n Z 2, we assume that the addition of two CF2 units
to the anion increases the anion radius by 0.451 A based on the
difference in ra between (FSO2)2N (2.835 A) and (CF3SO2)2N
(3.286 A) anions. Va of (C2F5SO2)2N: 218.6 A3, Va of
(C3F7SO2)2N: 307.7 A3, and Va of (C4F9SO2)2N: 418.2 A3.
36 H. Tokuda, S. Tsuzuki, M. A. B. H. Susan, K. Hayamizu and
M. Watanabe, J. Phys. Chem. B, 2006, 110, 19593–19600.
37 Z.-B. Zhou, H. Matsumoto and K. Tatsumi, ChemPhysChem,
2005, 6, 1324–1332.
24 T. A. Litovitz, J. Chem. Phys., 1952, 20, 1088–1089.
25 (a) H. Vogel, Phys. Z., 1921, 22, 645–646; (b) G. S. Fulcher, J. Am.
Ceram. Soc., 1923, 8, 339–355; (c) G. Tamman and W. Hesse,
Z. Anorg. Allg. Chem., 1926, 156, 245–257.
38 A. Triolo, O. Russina, H.-J. Bleif and E. Di Cola, J. Phys. Chem. B,
2007, 111, 4641–4644.
26 The Z value of DEME-4,4 at 25 1C (739 mPa s) was estimated by
extrapolating the measured data to the Litovitz equation,24 since
the viscosity data of DEME-4,4 are only available above 30 1C for
technical reason.
27 A. B. McEwen, H. L. Ngo, K. LeCompte and J. L. Goldman,
J. Electrochem. Soc., 1999, 146, 1687–1695.
28 H. Tokuda, K. Hayamizu, K. Ishii, M. A. B. H. Susan and
M. Watanabe, J. Phys. Chem. B, 2005, 109, 6103–6110.
29 P. Walden, Z. Phys. Chem., 1906, 55, 207–249.
30 (a) W. Xu, E. I. Cooper and C. A. Angell, J. Phys. Chem. B, 2003,
107, 6170–6178; (b) W. Xu, L.-M. Wang, R. A. Nieman and C. A.
Angell, J. Phys. Chem. B, 2003, 107, 11749–11756; (c) W. Xu and
C. A. Angell, Science, 2003, 302, 422–425.
31 R. Hagiwara, K. Matsumoto, Y. Nakamori, T. Tsuda, Y. Ito,
H. Matsumoto and K. Momota, J. Electrochem. Soc., 2003, 150,
D195–D199.
32 (a) Y. Yoshida, A. Otsuka, G. Saito, S. Natsume, E. Nishibori,
M. Takata, M. Sakata, M. Takahashi and T. Yoko, Bull. Chem.
Soc. Jpn., 2005, 78, 1921–1928; (b) Y. Yoshida and G. Saito,
J. Mater. Chem., 2006, 16, 1254–1262.
33 (a) Z.-B. Zhou, H. Matsumoto and K. Tatsumi, Chem.–Eur. J.,
2004, 10, 6581–6591; (b) Z.-B. Zhou, H. Matsumoto and K. Tatsumi,
Chem.–Eur. J., 2006, 12, 2196–2212.
39 (a) A. J. Carmichael and K. R. Seddon, J. Phys. Org. Chem., 2000,
13, 591–595; (b) M. J. Muldoon, C. M. Gordon and I. R. Dunkin,
J. Chem. Soc., Perkin Trans. 2, 2001, 433–435; (c) S. N. Aki,
J. F. Brennecke and A. Samanta, Chem. Commun., 2001, 413–414;
(d) J. L. Anderson, J. Ding, T. Welton and D. W. Armstrong,
J. Am. Chem. Soc., 2002, 124, 14247–14254; (e) L. Crowhurst,
P. R. Mawdsley, J. M. Perez-Arlandis, P. A. Salter and T. Welton,
Phys. Chem. Chem. Phys., 2003, 5, 2790–2794; (f) A. Kawai,
T. Hidemori and K. Shibuya, Chem. Lett., 2004, 1464–1465;
(g) C. Reichardt, Green Chem., 2005, 7, 339–351; (h) T. Fujisawa,
M. Fukuda, M. Terazima and Y. Kimura, J. Phys. Chem. A, 2006,
110, 6164–6172.
40 (a) M. J. Kamlet and R. W. Taft, J. Am. Chem. Soc., 1976, 98,
377–383; (b) R. W. Taft and M. J. Kamlet, J. Am. Chem. Soc.,
1976, 98, 2886–2894; (c) T. Yokoyama, R. W. Taft and
M. J. Kamlet, J. Am. Chem. Soc., 1976, 98, 3233–3237; (d) M. J.
Kamlet, J. L. Abboud and R. W. Taft, J. Am. Chem. Soc., 1977,
99, 6027–6038.
41 (a) M. J. Kamlet, J. L. Abboud, M. H. Abraham and R. W. Taft,
J. Org. Chem., 1983, 48, 2877–2887; (b) A. Sarkar and S. Pandey,
J. Chem. Eng. Data, 2006, 51, 2051–2055.
42 C. Chiappe and D. Pieraccini, J. Phys. Chem. A, 2006, 110,
4937–4941.
34 In a recent report, infinitely diluted KCl aqueous solution obeys
the ‘‘fractional’’ Walden rule LZa = const. instead of conventional
Walden rule LZ = const., where the exponent a was evaluated to
be 0.92(2). C. Schreiner, S. Zugmann, R. Hartl and H. J. Gores,
J. Chem. Eng. Data, 2010, 55, 1784–1788. Since the exponent a
corresponds to the slope of the logarithmic Walden plot as Fig. 7,
43 M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini, G. L.
Cascarano, L. De Caro, C. Giacovazzo, G. Polidori and
R. Spagna, J. Appl. Crystallogr., 2005, 38, 381–388.
44 G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinement, University of Gottingen, Germany, 1997.
¨
45 J. K. Ruff, Inorg. Chem., 1965, 4, 1446–1449.
c
20310 Phys. Chem. Chem. Phys., 2011, 13, 20302–20310
This journal is the Owner Societies 2011