Conductivities and electrochemical stabilities of lithium salts of polyfluoroalkoxyaluminate superweak anions

Article abstract of DOI:10.1149/1.1776589

Six lithium salts of tris- and tetrakis(polyfluoroalkoxy)aluminate superweak anions have been studied for their potential use as battery electrolytes. Four of the six are based on the formula LiAl(OCR(CF ₃)₂)₄ (R = H, Me,

Full text of DOI:10.1149/1.1776589

A1418
Journal of The Electrochemical Society, 151 ͑9͒ A1418-A1423 ͑2004͒
0013-4651/2004/151͑9͒/A1418/6/$7.00 © The Electrochemical Society, Inc.
Conductivities and Electrochemical Stabilities of Lithium Salts
of Polyfluoroalkoxyaluminate Superweak Anions
Shoichi Tsujioka,a,b Benjamin G. Nolan,a Hironari Takase,b Benjamin P. Fauber,a
,z
and Steven H. Straussa,
*
aDepartment of Chemistry, Colorado State University, Fort Collins, Colorado 80523 USA
bCentral Glass Company, Tokyo, Japan
Six lithium salts of tris- and tetrakis͑polyfluoroalkoxy͒aluminate superweak anions have been studied for their potential use as
battery electrolytes. Four of the six are based on the formula LiAl͑OCR͑CF3)2)4 (R ϭ H, Me, CF3 , Ph); the other two are
LiAl͑OCH2CF3)4 and LiAlF͑OCPh͑CF3)2)3 . The thermally stable electrolytes LiAl͑OCH͑CF3)2)4 and LiAl͑OCPh͑CF3)2)4 were
not oxidized at potentials less than or equal to 5.0 V vs. Liϩ/0 in dimethoxyethane ͑DME͒ or in 50:50% ethylene carbonate:dim-
ethylcarbonate ͑EC:DMC͒. The LiAl͑OCH͑CF3)2)4 electrolyte was not reduced at
0
V
vs. Liϩ/0 in DME. Neither
The electrolyte
LiAl͑OCH͑CF3)2)4 nor LiAl͑OCPh͑CF3)2)4 promoted the corrosion of aluminum at 5.0 V vs. Liϩ/0
.
LiAl͑OCH͑CF3)2)4 underwent efficient, reversible reductive intercalation of Liϩ with MCMB carbon or LiCoO2 electrodes over
the potential ranges 0-2 and 2.4-4.8 V, respectively, vs. Liϩ/0, but did not react in any other way with these electrode materials. The
conductivities of some of the LiAl͑ORF)4 electrolytes in DME or in EC:DMC were high enough for them to be considered as
potential replacements for LiPF6 in primary and secondary lithium batteries.
© 2004 The Electrochemical Society. ͓DOI: 10.1149/1.1776589͔ All rights reserved.
Manuscript submitted August 18, 2003; revised manuscript received March 18, 2004. Available electronically August 18, 2004.
We recently reported the conductivity and electrochemical be-
havior of lithium salts of new bis͑polyfluorodiolato͒borates such as
B͑OC(2-O-C6H4)(CF3)2)Ϫ2 .1 These salts satisfied a number of re-
quirements as potential replacements for LiPF6 in nonaqueous pri-
mary and secondary lithium batteries.2-5 As far as electrolytes are
concerned, suitable lithium salts must be ͑i͒ readily available at a
reasonable cost, (ii) highly conductive in solution ͑у5 mS cmϪ1͒,
(iii) thermally more stable than LiPF6 , which decomposes at tem-
LiAlF(HFPP)3.—To a solution of H͑HFPP͒ ͑70.0 g, 287 mmol͒
in toluene ͑150 mL͒ was added LiAlH4 ͑2.7 g, 71.1 mmol͒ at room
temperature. This mixture was stirred under reflux for 4 h. Heat was
removed, and crystals formed upon cooling to room temperature.
These were isolated by filtration and washed with hexane. Toluene
͑600 mL͒ was then added, the resulting suspension filtered, and the
filtrate cooled to Ϫ40°C. The crystals that formed were collected by
filtration and dried under vacuum to yield LiAlF͑HFPP)3 as a white
peratures well below 100°C in the solid state and in solution,6,7 (i )
1
v
powder. Yield 13.1 g ͑24% based on LiAlH4). H NMR (C6D6) ␦
7.68 ͑m, 6 H͒, 7.06 ͑m, 9 H͒. 19F NMR (C6D6) ␦ Ϫ75.9 ͑s͒. The
fluorine atom attached to the aluminum atom was not observed in
the 19F NMR spectrum, presumably because of quadrupolar broad-
ening by 27Al. Low resolution mass spectrum ͑negative ion electro-
spray, CH3CN solution͒ m/z 774.9 ((M-Li)Ϫ; calcd. for
C27H15AlF19OϪ3 , m/z 775.0͒.
electrochemically stable in a suitable solvent when in contact with
commonly used anodes and cathodes, and ( ) must not cause sig-
v
nificant corrosion of aluminum current collectors at high positive
potentials.
In this paper we report a parallel study of six lithium salts of
polyfluoroalkoxyaluminates. We8-12 and others13 have reported
the synthesis and properties of a series of lithium salts of
superweak
͑i.e.,
extremely
weakly
coordinating14-17
͒
LiAl(TFE)4 • 0.1DME.—The fluoroalcohol H͑TFE͒ ͑1.6 kg, 16
mol͒ was added with stirring to LiAlH4 ͑38 g, 1 mol͒ at room
temperature. The exothermic reaction heated the reaction mixture to
80°C. After 6 h of stirring, all volatiles were removed under
vacuum, leaving a white solid. The solid was dissolved in DME,
forming a cloudy mixture which was filtered to remove insoluble
material. The solvent was removed from the clear, colorless filtrate
under vacuum, and the white solid that remained was dried under
tetrakis͑polyfluoroalkoxy͒ aluminate anions. The first example,
LiAl͑OC͑Ph͒͑CF3)2)4(LiAl͑HFPP)4), was reported by us in
1996.12 One of the salts we previously synthesized,
LiAl͑OCH͑CF3)2)4(LiAl͑HFIP)4),8 has recently been studied as a
component of a nano-composite electrolyte.18
1
Experimental
vacuum at 60°C for 24 h. Yield 440 g ͑81% based on LiAlH4). H
NMR (CD3CN) ␦ 3.98 ͑s, 8 H͒, 3.30 ͑s, OCH3), 3.46 ͑s, OCH2).
19F NMR (CD3CN) ␦ Ϫ76.9 ͑s͒. The 1H NMR spectrum demon-
strated that the Li:DME molar ratio was 10:1.
Materials.—All syntheses, manipulations, and measurements
were done under an inert atmosphere of purified argon or helium
using Schlenk, glovebox, or high-vacuum techniques.19 Battery
grade dimethoxyethane ͑DME͒, ethylene carbonate ͑EC͒, propylene
carbonate ͑PC͒, and dimethyl carbonate ͑DMC͒ were stored in a
glovebox and used as received from Mitsubishi Chemical. The fluo-
roalcohols HOCH2CF3 ͑H͑TFE͒, Aldrich͒ and HOCPh͑CF3)2 ͑H͑H-
FPP͒, Central Glass͒ were dried with 4 Å molecular sieves. The
compounds LiAlH4 ͑Aldrich͒ and LiPF6 ͑Central Glass͒ were used
as received. MCMB carbon ͑Osaka Gas Chemicals͒ and LiCoO2
͑Nippon Chemical͒ were used as received. The electrolytes
Apparatus and measurements.—NMR spectra were recorded us-
ing a JEOL AL-400 spectrometer. DSC measurements were made
using a Rigaku DSC8230 ͑20°C minϪ1 heating rate͒. Electrolytic
conductivities were measured in an argon- or helium-filled glovebox
at 24 Ϯ 1°C using either a Kyoto Electronics Model K-111 conduc-
tivity cell (k ϭ 0.9878 cmϪ1͒ and a Kyoto Electronics Model CM-
115 conductivity bridge operated at 1.2 KHz or a YSI Model 3403
conductivity cell (k ϭ 0.9988 cmϪ1͒ and a YSI model 32 conduc-
tivity bridge operated at 1 KHz. Measurements for the same com-
pounds taken in Japan and in Fort Collins agreed to within experi-
mental error, which were generally Ϯ2%. All of the lithium salts
LiAl͑HFIP)4 ,8
LiAl͑OC͑CH3)(CF3)2)4
(LiAl͑HFTB)4),10,13
LiAl͑HFPP)4 ,12 and LiAl͑OC͑CF3)3)4 (LiAl͑PFTB)4)10,13 were
prepared as described in the literature.
1
were anhydrous as determined by H NMR spectroscopy. Solutions
of them were prepared in volumetric flasks in the glovebox. Al-
though fixed-frequency conductivity measurements may differ by as
* Electrochemical Society Active Member.
z E-mail: strauss@lamar.colostate.edu

Journal of The Electrochemical Society, 151 ͑9͒ A1418-A1423 ͑2004͒
A1419
Table I. Thermal and electochemical stabilities and maximum conductivities of LiPF6 and LiAlFu„ORF…4Àn electrolytesa.
a All data from this work unless otherwise noted. Abbreviations: HFIP ϭ OCH͑CF3)2 ; HFTB ϭ OC͑CH3)(CF3)2 ; HFPP ϭ OCPh͑CF3)2 ; PFTB
ϭ OC(CF3)3; TFE ϭ OCH2CF3 ; DME, 1,2-dimethoxyethane; EC:DMC, 50:50 mol % ethylene carbonate:dimethylcarbonate; PC, 1,2-propylene
carbonate; Eox, potential at which the aluminate anion is oxidized in the indicated solvent; ␴max , maximum conductivity in the solvent and at the
concentration ͑conc.͒ indicated in parentheses.
b Duration of experiment was 1 h.
c From Ref. 6.
d The conductivity values in italics do not represent maximum conductivities.
much as 10% from variable-frequency complex impedance mea-
surements, the relative conductivities of the new salts, an important
issue in this work, differ by as much as four times the lowest value.
Voltammetric and chronoamperometric experiments were performed
at 24 Ϯ 1°C in the glovebox using an ALS Model 600 electro-
chemical analyzer. The glassy carbon working electrode ͑0.008 cm2͒
was polished with alumina, rinsed and sonicated with distilled water,
and dried before each use. The aluminum working electrodes
͑99.997%, 2.0 cm2͒ were cleaned and dried before use. A new alu-
minum electrode was used for each experiment. The counter and
reference electrodes were lithium foil ͑Honjo Chemical, 99.9%͒.
The MCMB carbon anode was prepared by coating a drying an
N-methylpyrolidone ͑NMP͒ paste containing the MCMB carbon ͑95
wt %͒ and polyvinylidene fluoride ͑PVDF, 5 wt %͒ on copper foil.
The LiCoO2 cathode was prepared by coating and drying an NMP
paste containing LiCoO2 ͑85 wt %͒ and PVDF ͑15 wt %͒ on alumi-
num foil. All voltammetric experiments were performed at a scan
Results and Discussion
Minimum thermal stabilities.—Table I lists the formulas, abbre-
viations, and structures of the six electrolytes we have studied. In
addition, our results for LiPF6 have been added for comparison.
DSC experiments revealed that two of the aluminate salts,
LiAl͑HFIP)4 and LiAl͑HFPP)4 , undergo thermal decomposition at
100°C, significantly higher than the 40°C thermal decomposition
point of solid LiPF6 .6 In addition, when ethylene carbonate:di-
methyl carbonate ͑EC:DMC͒ solutions of these two lithium alumi-
nates were heated to 100°C for 1 day, the room temperature conduc-
tivities, 19F NMR spectra, and appearance ͑colorless solutions͒ was
unchanged. Solutions of LiPF6 in EC:DMC are reported to decom-
pose at 85°C.7 We found that an EC:DMC solution of LiPF6 decom-
posed when heated to only 70°C for 1 day ͑the evidence for decom-
position was a color change and the formation of a precipitate͒. We
did not investigate the thermal stabilities of the other lithium alumi-
nate salts listed in Table I.
rate of 10 mV sϪ1
.

A1420
Journal of The Electrochemical Society, 151 ͑9͒ A1418-A1423 ͑2004͒
Figure 2. Molar conductivity vs. the square root of the concentration
for DME solutions of LiAl͑HFTB)4 and LiAl͑HFPP)4 (HFTBϪ
ϭ OC͑CH3) (CF3)2Ϫ ; HFPPϪ ϭ OC͑C6H5)(CF3)Ϫ2 ). The lines drawn
through the data points are a visual aid.
Figure 1. Molar conductivity vs. the square root of the concentration for
DME solutions of LiAl͑HFIP)4 (HFIPϪ ϭ OCH͑CF3)2Ϫ). The lines drawn
through the data points are a visual aid. The molecular species shown is
the
Al͑HFIP) -Li-Al͑HFIP) ]Ϫ triple ion in the structure of
͓
4
4
1-Et-3-Me-1,3-C H N Li͑Al͑HFIP) ) ] ͑redrawn from Ref. 8͒. The un-
labeled shaded gray spheres are oxygen atoms. The unlabeled unshaded and
shaded white spheres represent carbon and fluorine atoms, respectively.
͓
͔͓
3
3
2
4 2
the dimeric structure of LiAl͑HFIP)4 , in which the Al͑HFIP)Ϫ4 an-
ions bridge the two Liϩ ions.8
The conductivity of a lithium salt of a molecular anion in a given
solvent is dependent, to varying degrees at different concentrations,
on ͑i͒ the coordinating and/or ion-pairing ability of the anion and
(ii) the mobility of the anion, which in turn is closely correlated
with the size of the anion. Figure 3 shows ␴ vs. DME concentration
curves for five of the six lithium aluminate electrolytes in this study.
The 0.1 and 0.2 M DME ␴ values for LiAl͑PFTB)4 , which are not
shown, are virtually the same as the corresponding ␴ values for
LiAl͑HFIP)4 and LiAl͑HFTB)4 ͑see Table I͒. Solutions of
LiAl͑PFTB)4 in DME more concentrated than 0.2 M could not be
prepared because of the limited solubility of this perfluorinated elec-
trolyte. It is significant that the ␴ values for LiAl͑HFPP)4 are lower
Conductivities.—The DME ␴
minate electrolytes and for LiPF6 are listed in Table I, along with
50:50 mol % EC:DMC ␴max values for LiAl͑HFIP)4 , LiAl͑HFPP)4
values for the six lithium alu-
max
and LiPF6 , and PC ␴
values for LiAl͑HFIP)4 and LiPF6 . Also
max
listed are 0.1 and 0.2 M DME ␴ values for the seven electrolytes.
Figures 1 and 2 display plots of molar conductivity ͑⌳͒ vs. the
square root of the concentration for DME solutions of LiAl͑HFIP)4 ,
LiAl͑HFPP)4 , and LiAl͑HFTB)4 . A similar plot ͑not shown͒ was
obtained for LiAl͑PFTB)4 . The shape of the plots for LiAl͑HFIP)4
and LiAl͑HFTB)4 , with a local minimum in ⌳ at ca. 0.01 M and a
local maximum in ⌳ at ca. 0.1 M, is common for electrolytes in
low-dielectric solvents ͑a similar curve for LiBF4 in DME has been
reported20͒ and has been interpreted in terms of significant triple-ion
formation at concentrations above the local minimum.21 The X-ray
structure of the Li͑Al͑HFIP)4)Ϫ2 inner-sphere triple ion8 is repro-
duced in Fig. 1. The Liϩ ion is coordinated to two polyfluoroalkox-
ide oxygen atoms from each of the two Al͑HFIP)Ϫ4 anions. The plot
in Fig. 2 for LiAl͑HFPP)4 does not have a distinct local minimum
and maximum. Nevertheless, the shape of the plot suggests that
triple ions are also formed in DME solutions of this electrolyte. The
lack of a local maximum for LiAl͑HFPP)4 suggests that the concen-
tration of Li͑Al͑HFPP))4)Ϫ2 inner-sphere triple ions in DME may be
lower than for the other two lithium aluminate electrolytes. This
may be due to steric hindrance of the bulky phenyl groups in the
Al͑HFPP)Ϫ4 anion. In support of this hypothesis, we note that the
solid state structure of LiAl͑HFPP)4 is monomeric,12 in contrast to
at all concentrations than the corresponding
␴ values for
LiAl͑HFIP)4 , LiAl͑HFTB)4 , or LiAl͑PFTB)4 . Since all four an-
ions are very weakly coordinating,8 we do not think that differences
in anion basicity or ion-pairing ability are responsible for the differ-
ence in ␴ values. We propose that the difference is due to the
significantly larger size, and hence significantly increased solution
viscosity and concomitant decreased mobility, of the Al͑HFPP)4Ϫ
anion relative to the other three Al͑OCR͑CF3)2)Ϫ4 anions (R
ϭ H, CH3 , CF3). This is consistent with the fact that ␴
occurs
max
at only 0.3 M for LiAl͑HFPP)4 instead of at 0.5 M for LiAl͑HFIP)4
and LiAl͑HFTB)4 . It is also consistent with the observation that
␴
for LiAl͑HFIP)4 is 17% higher than ␴
for LiAl͑HFTB)4 ,
max
max
which contains the ͑slightly͒ larger anion. Not surprisingly, anion
mobility does not affect ␴ as much at low concentrations as it does
at high concentrations: the 0.05 M DME ␴ values of LiAl͑HFIP)4 ,
LiAl͑HFTB)4 , and LiAl͑HFPP)4 are very similar because their ion-
pairing abilities ͑as well as their basicities8͒, are very similar.

Journal of The Electrochemical Society, 151 ͑9͒ A1418-A1423 ͑2004͒
A1421
Figure 5. Conductivities of 50:50 mol
LiAl͑HFIP)4 , LiAl͑HFPP)4 , and LiPF6 (HFIP ϭ OCH͑CF3)2 ; HFPP
ϭ OCPh͑CF3)2).
%
EC:DMC solutions of
Figure 3. Conductivities of DME solutions of LiAl͑HFIP)4 , LiAl͑HFTB)4 ,
LiAl͑HFPP)4 , LiAl͑TFE)4 , and LiAlF͑HFPP)3 (HFIP
ϭ OCH͑CF3)2 ; HFTB ϭ OC͑CH3)(CF3)2 ; HFPP ϭ OCPh͑CF3)2 ; TFE
ϭ OCH2CF3).
Al͑OCR͑CF3)2)Ϫ4 anions (R ϭ H, CH3 , CF3 , Ph). It has only one
CF3 group per alkoxide substituent and is the smallest of the five
Al͑ORF)Ϫ4 anions. For both of these reasons, we conclude that the
Al͑TFE)Ϫ4 anion is ion-paired with Liϩ to a much greater extent than
are the four Al͑OCR͑CF3)2)Ϫ4 anions, leading to the relatively low ␴
However, the LiAl͑TFE)4 electrolyte, which contains the small-
est of the five tetrakis͑polyfluoroalkoxy͒aluminates, also has the
lowest DME ␴ values at 0.1 to 0.4 M. The anion in this electrolyte
is electronically and sterically different than the four
values. On the other hand, the small size of the Al͑TFE)Ϫ4 anion is
responsible for the fact that ␴ continues to increase at concentrations
between 0.5 and 0.8 M, presumably because the viscosity of a 0.5 M
DME solution of LiAl͑TFE)4 is lower than the viscosity of a 0.5 M
DME solution of either LiAl͑HFIP)4 or LiAl͑HFTB)4 .
The LiAlF͑HFPP)3 electrolyte is unique in that it contains a very
polar Al-F bond with a strongly coordinating fluorine atom. The
12
22
X-ray structures of LiAl͑HFPP)4 and LiAlF͑HFPP)3 are com-
pared in Fig. 4. The latter structure contains a diamond-shaped Li2F2
core involving the fluorine atoms that are bonded to the Al atoms.
One of the two Li-F͑Al͒ bonds is 1.821͑8͒ Å, shorter than any of the
Li-O͑Al͒ bonds in either structure, which range from 1.978͑8͒ to
2.017͑8͒ Å. Therefore, the presence of the Al-F bond in the
AlF͑HFPP)Ϫ3 anion renders this anion much more strongly coordi-
nating ͑and, presumably, more strongly ion-pairing͒ than the
Al͑HFPP)Ϫ4 anion. This explains why the conductivity of
LiAlF͑HFPP)3 in DME is so much lower than that of LiAl͑HFPP)4 ,
despite the fact that the AlF͑HFPP)Ϫ3 anion is smaller than
Al͑HFPP)Ϫ4 . Consistent with this, ␴
for LiAlF͑HFPP)3 , while
max
Figure 4. Drawings of the structures of the monomeric structure
LiAl͑HFPP)4 ͑Ref. 12͒, left, and the centrosymmetric dimeric structure of
LiAlF͑HFPP)3 ͑Ref. 22͒, right. The unlabeled shaded circles are fluorine
atoms and the unlabeled plain circles are carbon atoms. Hydrogen atoms
lower than ␴
for LiAl͑HFPP)4 , occurs at a higher concentration
max
than for LiAl͑HFPP)4 .
The 0.1, 0.2, and 0.3 M DME ␴ values for LiPF6 , which are also
not shown, are lower than the corresponding values for four of the
have been omitted for clarity. HFPPϪ ϭ OCPh͑CF3)2Ϫ
.

A1422
Journal of The Electrochemical Society, 151 ͑9͒ A1418-A1423 ͑2004͒
Figure 7. CVs of 0.3 M propylene carbonate solutions of LiAl͑HFIP)4 ,
LiAl͑HFPP)4 , and LiPF6 . Conditions: aluminum working electrode, lithium
foil counter and reference electrodes, 10 mV sϪ1͒.
between 0 and 2 V vs. Liϩ/0 ͑5 mV sϪ1 scan rate͒. In both cases,
efficient reductive intercalation of Liϩ was observed. There were no
other reactions of the electrolyte with the carbon electrode. The
same was true for both electrolyte solutions when a LiCoO2 elec-
trode was used instead of the carbon electrode ͑the potential range
investigated was 2.4-4.8 V vs. Liϩ/0).
The lack of any tendency of LiAl͑HFIP)4 and LiAl͑HFPP)4 to
promote the corrosion of aluminum was investigated by cyclic vol-
tammetry, chronoamperometry, and scanning electron microscopy
͑SEM͒. CVs for these two electrolytes and for LiPF6 in PC using an
aluminum working electrode are shown in Fig. 7. In each experi-
ment, potentials between 2 and 5 V vs. Liϩ/0 were scanned five
times. Passivation of the electrode surface was evident for all three
electrolytes during the first scan. By the fifth scan, the current den-
sity at 5 V was р2.2 ␮A cmϪ2. After the fifth scan, each aluminum
electrode was examined by SEM, and no evidence of corrosion ͑i.e.,
pitting͒ was observed. In addition, new aluminum working elec-
trodes were held at 5 V vs. Liϩ/0 in each electrolyte solution for one
hour. Again, no pitting of any of the three electrodes was observed
by SEM.
Figure 6. Conductivities of 0.5 M LiAl͑HFIP)4 solutions as a function of
mol % EC in EC:DMC mixtures (HFIP ϭ OCH͑CF3)2).
six lithium aluminates ͑see Table I͒. We suggest that the PFϪ6 anion,
like the AlF͑HFPP)Ϫ3 and Al͑TFE)Ϫ4 anions, is more strongly asso-
ciated with Liϩ in DME than are the four Al͑OCR͑CF3)2)4 anions.
In contrast, the EC:DMC ␴ values for LiPF6 are higher than for
LiAl͑HFIP)4 and LiAl͑HFPP)4 , as shown in Fig. 5. This is probably
due to both the higher viscosity and the higher dielectric constant of
the EC:DMC mixture relative to DME. Nevertheless, the EC:DMC
␴
for LiAl͑HFIP)4 , 6.3 mS cmϪ1, is high enough for this ther-
max
mally stable electrolyte to be considered as a replacement for LiPF6
in secondary lithium-ion batteries.
Acknowledgment
We investigated whether 50:50 mol % EC:DMC was the opti-
mum blend of these solvents for the LiAl͑HFIP)4 electrolyte. The
results, shown in Fig. 6, show that a blend between 20:80 and 30:70
mol % EC:DMC is marginally better than the standard 50:50 mol %
mixture.
This research was supported by U.S. National Science Founda-
tion grant CHE-9905482.
Colorado State University assisted in meeting the publication costs of
this article.
References
Electrochemical stability.—The electrochemical stabilities of the
Al͑HFIP)Ϫ4 and Al͑HFPP)Ϫ4 anions were investigated using cyclic
voltammetry ͑CV͒. Negligible faradaic current was observed be-
tween 0 and 5.2 V vs. Liϩ/0 ͑conditions: 0.1 M Liϩ salt in DME,
glassy carbon working electrode, 5 mV sϪ1͒. At potentials higher
than 5.2 V, irreversible DME oxidation, and possible anion oxida-
tion, occured. Below 0 V, plating of lithium was observed. The same
results were obtained for a 0.1 M DME solution of LiPF6 . In addi-
tion, a 0.5 M EC:DMC solution of LiAl͑HFIP)4 did not undergo
oxidation at potentials less than or equal to 5 V vs. Liϩ/0. At poten-
tials higher than 5 V, irreversible oxidation of EC, DMC, and/or the
Al͑HFIP)Ϫ4 anion occured.
1. B. G. Nolan and S. H. Strauss, J. Electrochem. Soc., 150, A1726 ͑2003͒.
2. Handbook of Batteries, 3rd Ed., D. Linden and T. B. Reddy, Editors, McGraw-Hill,
New York ͑2002͒.
3. J. Barthel and H. J. Gores, in Handbook of Battery Materials, J. O. Besenhard,
Editor, Wiley-VCH, Weinheim ͑1999͒.
4. M. Morita, M. Ishikawa, and Y. Matsuda, in Lithium Ion Batteries: Fundamentals
and Performance, M. Wakihara and O. Yamamoto, Editors, p. 156, Wiley-VCH,
Weinheim ͑1998͒.
5. G. E. Blomgren, in Lithium Batteries, J. P. Gabano, Editor, Academic Press, New
York ͑1983͒.
6. T. C. Ehlert and M. M. Hsia, J. Chem. Eng. Data, 17, 18 ͑1972͒.
7. S. E. Sloop, J. K. Pugh, S. Wang, J. B. Kerr, and K. Kinoshita, Electrochem.
Solid-State Lett., 4, A4 ͑2001͒.
8. S. M. Ivanova, B. G. Nolan, Y. Kobayashi, S. M. Miller, O. P. Anderson, and S. H.
Strauss, Chem.-Eur. J., 7, 503 ͑2001͒.
9. T. J. Barbarich, B. G. Nolan, S. Tsujioka, S. M. Miller, O. P. Anderson, and S. H.
Strauss, J. Fluorine Chem., 112, 335 ͑2001͒.
10. S. H. Strauss, B. G. Nolan, T. J. Barbarich, and J. J. Rockwell, U.S. Pat. 6,221,941
B1 ͑2001͒.
Lack of reactivity with MCMB carbon, LiCoO2 , and alumi-
num.—CVs ͑not shown͒ of EC:DMC solutions of LiAl͑HFIP)4 ͑0.5
M͒ and LiPF6 ͑1 M͒ using an MCMB carbon electrode were similar

Journal of The Electrochemical Society, 151 ͑9͒ A1418-A1423 ͑2004͒
A1423
11. T. J. Barbarich, S. M. Miller, O. P. Anderson, and S. H. Strauss, J. Mol. Catal. A:
18. H. Tokuda and M. Watanabe, Electrochim. Acta, 48, 2085 ͑2003͒.
Chem., 128, 289 ͑1998͒.
12. T. J. Barbarich, S. T. Handy, S. M. Miller, O. P. Anderson, P. A. Grieco, and S. H.
Strauss, Organometallics, 15, 3776 ͑1996͒.
13. I. Krossing, Chem.-Eur. J., 7, 490 ͑2001͒.
14. I. Krossing and I. Raabe, Angew. Chem. Int. Ed., 43, 2066 ͑2004͒.
15. C. A. Reed, Acc. Chem. Res., 31, 133 ͑1998͒.
19. D. F. Shriver and M. A. Drezdzon, The Manipulation of Air-Sensitive Compounds,
2nd Ed., Wiley-Interscience, New York ͑1986͒.
20. J. Barthel, R. Gerber, and H. J. Gores, Ber. Bunsenges. Phys. Chem., 88, 616
͑1988͒.
21. R. M. Fuoss and C. A. Kraus, J. Am. Chem. Soc., 55, 2387 ͑1933͒.
22. B. G. Nolan, T. J. Barbarich, S. Tsujioka, S. M. Ivanova, B. P. Fauber, S. M. Miller,
O. P. Anderson, and S. H. Strauss, In preparation.
16. S. H. Strauss, Chem. Rev. (Washington, D.C.), 93, 927 ͑1993͒.
17. S. H. Strauss, Chemtracts: Inorg. Chem., 6, 1 ͑1994͒.