Journal of The Electrochemical Society, 158 (8) A883-A889 (2011) A883

0013-4651/2011/158(8)/A883/7/$28.00 V
C The Electrochemical Society

Electrochemical Behavior of Layered Solid Solution

Li2MnO32LiMO2 (M 5 Ni, Mn, Co) Li-Ion Cathodes
with and without Alumina Coatings
W. C. West,a,*,z J. Soler,a M. C. Smart,a,* B. V. Ratnakumar,a,* S. Firdosy,a V. Ravi,a,b,*
M. S. Anderson,a J. Hrbacek,c E. S. Lee,c and A. Manthiramc,*
a
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91109, USA
b
Department of Chemical and Materials Engineering, California State Polytechnic University, Pomona, California
91768, USA
c
Electrochemical Energy Laboratory and Materials Science and Engineering Program, The University of Texas at
Austin, Austin, Texas 78712, USA

Thin films of oxides, phosphates, fluorides and other analogous materials on lithium-ion cathode particles are well known to
improve cathode performance in terms of cycle life and rate performance. Explanations for this phenomenon abound, but the
underlying mechanisms that dictate the nature of these effects are still in question, which motivates the work herein. We have car-
ried out systematic PITT, EIS, Tafel, and cycling experiments as a function of temperature for Al2O3-coated and uncoated layered
solid solution Li2MnO3LiMO2 (M ¼ Mn, Co, Ni) cathode materials and shown that we can reproduce the well-documented
improvement in performance with surface coatings. In particular the effects are most pronounced at reduced temperatures and after
temperature cycling (23 to 0 C to 30 to 0 C). Interestingly, we find the activation energies for the diffusion coefficients estimated
from PITT data are nearly identical to the activation energy for exchange current measured from Tafel polarization data. This find-
ing may provide some insight into the relative control of the mass transfer and the charge transfer processes on the overall cathode
reaction. Alternately, it may be the due to inadequate correction for the mass transfer effects in the Tafel and PITT analyses.
C 2011 The Electrochemical Society.
V [DOI: 10.1149/1.3597319] All rights reserved.

Manuscript submitted March 22, 2011; revised manuscript received May 5, 2011. Published June 6, 2011.

The layered solid solution Li2MnO3LiMO2 (M ¼ Mn, Co, Ni) spectroscopy (EIS), Potentiostatic Intermittent Titration Technique
materials system is an attractive Li-ion battery cathode given its (PITT), and Tafel polarization studies as a function of temperature
high specific capacity of upwards of 240–280 mAh/g between to help elucidate the contributions of Liþ diffusion, charge transfer
4.8–2 V.1–4 However, these cathode materials exhibit large irrevers- and other factors that impact cathode performance with and without
ible capacity on the first charge-discharge cycle and are somewhat a cathode surface coating.
rate limited with comparatively low cycle life relative to conven-
tional cathodes such as LiCoO2 and LiNi0.8Co0.15Al0.05O2. In order
to improve the rate capability and cycle life, many groups have Experimental
demonstrated that adding a thin coating of metal phosphates, fluo-
rides, oxides, or other analogous materials onto the cathode particle The cathode powders were prepared using a carbonate route syn-
results in reduced irreversible capacity, improved rate capability, thesis method. A 0.6 M (Mn0.675Ni0.1625Co0.1625)SO4 solution was
and cycle life.5–8 Yet this effect seems counterintuitive: how can prepared by dissolving stoichiometric quantities of manganese,
AlF3, AlPO4, Al2O3 or other poor ionically and/or electronically nickel, and cobalt sulfates in deionized water. This solution was
conducting films improve the performance of the mixed ion-electron added dropwise into a stirred aqueous solution of Li2CO3
conducting electrode material when notionally such surface films maintained at 50 C by an oil bath for 24 h. The precipitate,
should only hinder charge transfer and diffusion at the electrode (Mn0.675Ni0.1625Co0.1625)CO3, was then filtered and washed with de-
particle? ionized water. After washing, the precipitate was dried overnight at
The improvements in performance of these lithium ion cathodes 100 C in air. The dried carbonate precursor was fired in air at 500 C
as well as other cathodes such as LiCoO2 and LiMn2O4 with the for 10 h in an alumina crucible at a heating rate of 5 C/min and was
addition of poorly conducting films have been attributed to a diverse allowed to cool in the furnace. The carbonate precursor decomposed
range of mechanisms, such as the coating promoting the retention of into an oxide. The resulting oxide precursor was mixed with
oxide ion vacancies in the crystal lattice after the first charge,9 sup- LiOHH2O at a ratio of 0.76 g LiOHH2O per 1 g of oxide precursor.
pression of the decomposition of the electrolyte,10 suppression of The mixture was ground thoroughly with an agate mortar and pestle
cathode electrolyte interphase (CEI) formation,11 maintenance of until a fine and homogenous powder was obtained. The ground mix-
low microstrain for better structural integrity and crystallinity during ture was added to an alumina crucible. The mixture was heated at a
cycling,12 and scavenging of HF from the electrolyte.13 Clearly, rate of 5 C/min until 500 C was reached. The furnace was held at
with the multitude of hypotheses presented in literature to explain 500 C for 3 h before increasing the temperature to 900 C at a rate
this very important but rather inexplicable efficacy in cathode coat- of 5 C/min. After holding at 900 C for 18 h, the sample was
ings, further study of the fundamental underlying mechanisms is quenched in air by removing the crucible from the furnace and plac-
warranted. ing it on a ceramic brick. For samples prepared with a 1.5 wt %
In this study we have performed a systematic study of the overli- Al2O3 coating, the cathode powder was dispersed in deionized
thiated layered cathode Li1.17Mn0.56Ni0.135Co0.135O2 with and with- water, and a stoichiometric amount of aluminum nitrate nonahydrate
out a 1.5 wt % Al2O3 coating. This cathode materials system has was added to the suspension. Ammonium hydroxide was added to
been shown to clearly benefit from an alumina coating,14 though the suspension until the pH reached 12. The suspension was stirred
again, the fundamental nature of the beneficial effects of the coating for 2 h, filtered, rinsed, and dried at 100 C for 12 h. The powder
are as yet not well understood. In addition to two and three electrode was heated at 300 C for 8 h at a heating rate of 2 C/min and cooled
cell level studies to validate the performance improvements with at a cooling rate of 5 C/min.
coatings, we have carried out detailed electrochemical impedance The X-Ray diffraction (XRD) measurements were carried out
using a Siemens D500 diffractometer run in the theta-2 theta geome-
try, with a Cu anode (k ¼ 1.541 Å) at an accelerating voltage of 40
* Electrochemical Society Active Member. kV and a tube current of 20 mA. Surface morphology was studied
z
E-mail: william.c.west@jpl.nasa.gov using a FEI Nova NanoSEM 600 field-emission electron

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 A884 Journal of The Electrochemical Society, 158 (8) A883-A889 (2011)

microscope. Energy Dispersive Spectroscopy (EDS) measurements hardware with Al clad stainless steel cases at the cathode terminal
were collected, displayed and interpreted using the EDAX Genesis with 100 l of the above electrolyte, with multiple cell replicants
Analyzer System. The Fourier Transform Infrared (FTIR) experi- used to validate the reproducibility of the cell data. All spiral wound
ments were performed using a BioRad FTS 6000 infrared spectrom- and coin cells were first cycled through five formation cycles at
eter with a Pike Diffuse Reflectance attachment. The samples were C/20 charge/discharge rates between 2 and 4.8 V, with a 1 h
analyzed using Diffuse Reflectance Fourier Transform Infrared potentiostatic current taper step at 4.8 V.
(DRIFT) spectroscopy.15–17 The DRIFT method involves mixing The spiral wound cells were tested as a function of temperature
the sample powder with infrared transparent potassium bromide. by EIS, PITT, and Tafel Polarization measurements using a potentio-
The spectrometer beam is then reflected off the sample and is stat/galvanostat/frequency response analyzer (Biologic VMP2). For
collected over a wide solid angle. Tap density measurements were the EIS measurements, the applied AC signal was 5 mV peak-to-
performed with a Quantachrome Autotap instrument, with 1000 taps peak over a frequency range of 200 kHz–5 mHz at potentiostatic con-
per sample. ditions at 4.0 V vs. Li/Liþ. Impedance spectra were fitted using
Electrodes were prepared by spray coating Al foil substrates Zview software (Scribner). PITT measurements were performed by
with slurries of 80 wt % cathode powder, 10 wt % C black (Shawini- applying 10 mV steps from 4.00 to 3.95 V vs. Li/Liþ. Tafel Polariza-
gan), and 10 wt % poly(vinylene difluoride) (PVDF) binder (Sigma tion measurements were performed by applying a 1 mV/sec potentio-
Aldrich, MWavg ¼ 534,000) in N-methyl-2-pyrrolidinone (NMP) dynamic sweep from 4.0 V to 3.85 V vs. Li/Liþ. The cells were
(Sigma Aldrich). The electrode active mass loading was 5–6 mg/cm2. soaked at the desired temperature in a forced convection Tenney ther-
Three electrode spiral wound cells were prepared by winding the mal chamber for at least 1 h prior to carrying out the measurements.
spray-coated 3.8  15.9 cm cathodes, 3.8  19 cm Li foil counter
electrodes around a poly(tetrafluoroethylene) (PTFE) mandrel. A
Results and Discussion
3.8  0.5 cm Li foil reference electrode was inserted between the
counter and working electrodes toward the beginning of the wind- Cathode powder materials analyses.— The morphology of the
ing. The electrodes were heat sealed in 20 m Tonen separators. uncoated Li1.17Mn0.56Ni0.135Co0.135O2 and Al2O3-coated
The spiral wound electrodes were placed in a glass vial, and 5.00 ml Li1.17Mn0.56Ni0.135Co0.135O2 cathode powders are shown in the
1 M LiPF6 in ethylene carbonate:dimethyl carbonate:diethyl carbon- SEM micrographs in Fig. 1 at 5 and 400 kX magnifications. The
ate (EC:DMC:DEC) (1:1:1 vol %) electrolyte was pipetted into the two powders were generally indistinguishable from one another in
inner glass vial. The glass vial containing the electrodes and electro- the micrographs, indicating that the Al2O3 coating did not lead to
lyte was then sealed in an outer glass assembly with a Viton o-ring clumping or any other observable change in the microstructure of
clamped outer glass assembly. Coin cell studies were performed by the cathode particles. The uniformity of the Al2O3 coating was not
assembling the above electrodes in stainless steel CR2032 coin cell quantitatively assessed in this study. A small difference in the tap

Figure 1. SEM micrographs of (a, b)

uncoated Li(Li0.17Mn0.56Ni0.135Co0.135)O2
and (c,d) Al2O3-coated Li(Li0.17Mn0.56-
Ni0.135Co0.135)O2 cathode powders.

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 158 (8) A883-A889 (2011) A885

Figure 4. FTIR data of Li(Li0.17Mn0.56Ni0.135Co0.135)O2 and Al2O3-coated

Li(Li0.17Mn0.56Ni0.135Co0.135)O2 cathode powders.

Figure 2. XRD data of uncoated Li(Li0.17Mn0.56Ni0.135Co0.135)O2 and at 2h of 21 to 25 that were not indexed to the R3m symmetry were
Al2O3-coated Li(Li0.17Mn0.56Ni0.135Co0.135)O2 cathode powders. Asterisks consistent with cation ordering that occurs in the transition metal
indicate reflections associated with cation superlattice ordering.
layer.18 The two diffraction patterns overlay nearly identically,
indicating that the heat treatment or other processes involved with
densities was measured, corresponding to 1.14 g/cm3 (std. the Al2O3 coating did not result in distortion of the crystal lattice.
dev. ¼ 0.008 g/cm3) and 1.17 g/cm3 (std. dev. ¼ 0.02 g/cm3) for the EDS data revealed the presence of aluminum in the Al2O3-coated
uncoated and Al2O3-coated powders, respectively. While these tap cathode powder and no aluminum in the uncoated powder as expected.
densities were well below the tap density of conventional Li-ion However, for the uncoated cathode powder, the presence of sulfur was
cathode powders, subsequent studies have demonstrated that detected (Fig. 3). To better understand the nature of the sulfur, FTIR
changes in the synthesis conditions can yield tap densities greater measurements were conducted, which identified the sulfur species as
than 2.0 g/cm3, and will be reported in a follow-on report. SO42 (Fig. 4). Almost certainly, the sulfate contaminant was a result
The X-ray diffraction patterns of both the uncoated and Al2O3- of the incomplete reaction of the transition metal sulfates with Li2CO3.
coated powders were well indexed to the layered solid solution Both EDS and FTIR measurements of the Al2O3-coated cathode pow-
Li2MnO3LiMO2 (M ¼ Ni, Mn, Co), with no additional diffraction der failed to detect any sulfur species, which indicated the coating pro-
peaks associated with alumina phases discerned (Fig. 2). Weak peaks cess was effective in removing the contaminant through either the

Figure 3. EDS data of (a) Li(Li0.17Mn0.56

Ni0.135Co0.135)O2 and (b) Al2O3-coated
Li(Li0.17Mn0.56Ni0.135Co0.135)O2 cathode
powders.

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 A886 Journal of The Electrochemical Society, 158 (8) A883-A889 (2011)

Figure 6. Coin cell cycling data with uncoated Li(Li0.17Mn0.56Ni0.135-

Co0.135)O2 and Al2O3-coated Li(Li0.17Mn0.56Ni0.135Co0.135)O2 electrodes.

pares the specific discharge capacity of the two cathodes as the cells
were subjected to formation cycling at 23 C at C/20, followed by C/
10 rate cycling at 23, 0, 30 C, and then returned to 0 C cycling.
Under mild conditions of formation cycling and room temperature
C/10 cycling, the Al2O3-coated material slightly outperformed the
uncoated material. However, after cycling at C/10 at 0 C, 30 C,
then returning to 0 C cycling, the specific capacity of the uncoated
material fell sharply compared to the Al2O3-coated material. A com-
parison of the specific capacity at 0 C before and after the 30 C cy-
cling showed that the uncoated material capacity fell by approxi-
mately a factor of three whereas the Al2O3-coated material
experienced almost no loss in capacity following the 30 C cycling.
Although the uncoated material appeared to recover in capacity with
additional cycling, it was clear that the Al2O3-coated material out-
performed the uncoated material under all cycling conditions partic-
ularly during and after 30 C cycling.

Figure 5. Room temperature coin cell charge/discharge data at C/20 for (a) EIS studies.— Three electrode cells were used to carry out EIS
Li(Li0.17Mn0.56Ni0.135Co0.135)O2 and (b) Al2O3-coated Li(Li0.17Mn0.56- measurements as a function of temperature. Following room tempera-
Ni0.135Co0.135)O2 electrodes. ture C/20 formation cycling, the cells were charged to 4.0 V (approxi-
mately 80% state of charge) and cathode impedances as a function of
temperature were measured. It should be noted that the cell impedance
rinse or post-coating heating steps. The lithium content and pH of the and the charge transfer resistance are known to change as a function
powders were not assessed before or after the coating process. of state of charge, especially at lower states of charge. Thus, effort
was devoted to ensuring the cells were in a similar state of charge to
Electrochemical capacity studies.— Coin cell studies were per- minimize these effects. The complex plane plots are shown in Fig. 7.
formed to assess the specific capacity and cycleability of the cathode The impedance spectra were characterized by a high frequency relaxa-
materials under various conditions. Room temperature formation tion, a second mid-frequency relaxation, as well as a low frequency
charge/discharge profiles are shown in Fig. 5. For both the uncoated Warburg-like tail that was not well resolved at lower temperatures.
and Al2O3-coated cathode cells, the characteristic first cycle charge The equivalent circuit used to interpret such an impedance pattern
plateau was observed at about 4.5 V, corresponding to the removal with two relaxation loops is comprised of a series resistance which
of oxygen from the cathode accompanied by diffusion of transition represents a sum of the electronic resistance from the electrodes,
metal ions from surface to bulk where they occupy vacancies cre- leads, and the electrolyte resistance, a parallel resistor–constant phase
ated by lithium removal.19 While the first discharge capacities of element (CPE) network in series for the high-frequency relaxation
both the coated and uncoated cells were measured to be about 275 loop typically associated with the film impedance, another resistor–
mAh/g, the first cycle irreversible capacity for the uncoated cathode CPE network parallel in series for the mid-frequency relaxation loop
cell was 69 mAh/g compared with 46 mAh/g for the Al2O3-coated associated with charge transfer impedance, and a Warburg impedance
cathode cell, as expected based upon other reports in literature that corresponding to solid-state diffusion of lithium in the cathode.24
have demonstrated that various cathode coatings can reduce first The magnitude of impedance of the high frequency relaxation
cycle irreversible capacity losses.20–23 was only weakly dependent on temperature for both the uncoated
More significant differences in the performance of the two cath- and Al2O3-coated cathode. This high frequency relaxation is gener-
ode materials were observed in cycling experiments. Figure 6 com- ally ascribed to lithium-ion diffusion through the surface layer

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 158 (8) A883-A889 (2011) A887

Figure 8. Fitted mid-frequency x-axis intercept data from EIS data of

uncoated Li(Li0.17Mn0.56Ni0.135Co0.135)O2 and Al2O3-coated Li(Li0.17Mn0.56-
Ni0.135Co0.135)O2 electrodes.

is consistent with previous studies in which reduced charge transfer

resistances have been observed with the incorporation of coatings; the
phenomena generally being attributed to suppression of CEI growth
and improved surface electronic conductivity.11,24 It should be noted
that one of the perceived benefits associated with the electrode coat-
ings is that they provide a protective role, resulting in both decreased
impedance growth and decreased electrolyte decomposition with pro-
longed cycling.27
PITT studies.— Three electrode cells were used to carry out PITT
measurements as a function of temperature. Following equilibration
at the desired temperature and voltage (i.e., 4.0 V), 10 mV steps were
applied and the current recorded until the cell current tapered to 100
A (0.00165 mA/cm2). Following Wen et al.,28,29 the diffusion coef-
ficient D for an intercalation for a small voltage step is given by

Figure 7. EIS data as a function of temperature of (a) uncoated Li

(Li0.17Mn0.56Ni0.135Co0.135)O2 and (b) Al2O3-coated Li(Li0.17Mn0.56Ni0.135-
Co0.135)O2 electrodes.

(i.e., the CEI layer), often referred to as the film resistance.11,25 It was
also observed that the Al2O3-coated cathode at ambient temperatures
displayed a somewhat higher film resistance compared to the
uncoated sample, possibly due to the contribution from the surface
coating itself. The mid-frequency relaxation response is typically
ascribed to the charge transfer process, which is generally observed to
be associated with higher impedance compared to the film resistance,
thus being rate determining.26 In contrast to the high frequency relax-
ation response, the magnitude of impedance associated with the mid-
frequency relaxation as a function of temperature was much more
dramatic. For example, the charge transfer resistance of the uncoated
cathode grew from 3.5 to 78.2 X (a factor of 22) from room tempera-
ture to 12 C while the Al2O3-coated cathode grew from 1.8 to 11.0
X (a factor of 6) over the same temperature range. The fitted resist-
ance (x-axis intercept) of the mid-frequency relaxation as a function
of temperature is shown in Fig. 8. It is clear that the Al2O3 cathode
coating imparts a beneficial effect in terms of mitigating the charge Figure 9. Lithium ion diffusion coefficient data as a function of temperature
transfer impedance particularly at reduced temperature. This finding as measured by PITT.

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 A888 Journal of The Electrochemical Society, 158 (8) A883-A889 (2011)

D ¼ ½ðIt0:5 Þmax p0:5 r=ðDQÞ2 [1] ature. Tafel plots in terms of specific current (mA/g) are shown in
Fig. 10 from room temperature to 39 C. Consistent with the PITT
where t is time, r is particle radius, DQ ¼ $I(t)dt, and measurements of Liþ diffusion coefficients, at room temperature the
(It0.5)max ¼ slope of tangent of I(t) vs. t0.5 plot. The particle radii uncoated cathode sustains a greater specific current than the Al2O3-
for both cathode powders were taken as 50 nm. coated cathode, but at reduced temperatures the trend reverses.
At room temperature, the Liþ diffusion coefficient for the The specific exchange current, taken as the specific current ex-
uncoated cathode was slightly higher than that of the Al2O3-coated trapolated from the Tafel plots to 4.0 V (0 V vs. open circuit), is
cathode, at 6.1  1012 cm2/s and 4.4  1012 cm2/s, respectively. plotted as a function of temperature in Fig. 11. The activation ener-
However, at lower temperatures, the Liþ diffusion coefficient of the gies of the specific exchange current cannot be fitted to single val-
uncoated cathode fell much more steeply than the Al2O3-coated ues. However, in examination of the exchange current in the upper
cathode, with the temperature dependencies of the diffusion coeffi- temperature range (Fig. 11 inset), a noteworthy trend may be
cients following single activation energies of 0.45 and 0.17 eV, observed. The activation energies of the specific exchange current –
respectively (Fig. 9). It is important to consider that PITT measures 0.45 eV for uncoated cathode and 0.17 eV for the Al2O3-coated
a composite diffusion coefficient that includes bulk crystalline diffu- cathode – track very closely to the activation energies of the Liþ dif-
sion, diffusion through the cathode coating, and diffusion through fusion coefficients (0.45 eV for uncoated cathode and 0.19 eV for
the CEI. the Al2O3-coated cathode).
This close correlation between the activation energy for dif-
Tafel polarization studies.— Utilizing three electrode cells, Tafel fusion and exchange current reveals an important insight as to
polarization measurements were performed as a function of temper- the nature of mass transport through the cathodes and charge
transfer at the electrode/electrolyte interface in that one of the
two mechanisms is the rate limiting step during charge and dis-
charge of the cathode. The exchange current is strongly depend-
ent on the concentration of lithium ions at reaction zone, which
is the CEI-electrode interface. In other words, lithium ions will
have to diffuse through the CEI and the surface coatings
(unless the coating is porous) to be available for intercalation
into the cathode. The diffusion thus includes two components,
diffusion through the CEI and the solid state diffusion in the
bulk of the cathode (diffusion in the electrolyte phase if rela-
tively rapid) and cannot be distinguished experimentally. If
diffusion of the Liþ to the CEI-cathode interface is the rate
limiting step, then this term will dictate the exchange current
and as such the exchange current will follow the same tempera-
ture dependency as that for diffusion. Yet a key assumption for
the validity of diffusivity data generated from PITT measure-
ments is that diffusion is the rate limiting step.28 Thus, the data
suggest one of two scenarios: either the net diffusivity of the
cathode is the rate limiting step, or charge transfer is the rate
limiting step in which case the PITT data cannot be considered
valid based on the violation of the rate limiting step assump-
tion. If the slow diffusion is occurring through the CEI and sur-
face coating, it is likely that the assumed boundary conditions

Figure 10. Tafel polarization data as a function of temperature, from 23 to Figure 11. Specific exchange current as a function of temperature for
39 C in approximately 5 C increments for (a) uncoated Li(Li0.17Mn0.56- uncoated Li(Li0.17Mn0.56Ni0.135Co0.135)O2 and Al2O3-coated Li(Li0.17Mn0.56-
Ni0.135Co0.135)O2 and (b) Al2O3-coated Li(Li0.17Mn0.56Ni0.135Co0.135)O2 Ni0.135Co0.135)O2 electrodes. Inset highlights the data from the temperature
electrodes. range of 23 to 5 C, fitted to a single activation.

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 158 (8) A883-A889 (2011) A889

may not be applicable, as with the bulk diffusion. In any event, References
it is clear that the alumina coating results in a demonstrable 1. Z. Lu, D. D. MacNeil, and J. R. Dahn, Electrochem. Solid-State Lett., 4, A191
improvement in cathode performance particularly at reduced (2001).
temperatures as measured by EIS, PITT, and Tafel polarization 2. Z. Lu, D. D. MacNeil, and J. R. Dahn, Electrochem. Solid-State Lett., 4, A200
(2001).
measurements. 3. S.-H. and K. Amine, J. Power Sources, 146, 654 (2005).
4. C. S. Johnson, N. Li, C. Lefief, and M. M. Thackeray, Electrochem. Commun., 9,
Conclusions 787 (2007).
5. Y. Wu and A. Manthiram, Solid State Ionics, 180, 50 (2009).
The efficacy of alumina coatings on Li2MnO3-LiMO2 (M ¼ Mn, 6. Y. Wu, A. V. Murugan, and A. Manthiram, J. Electrochem. Soc., 155, A635
Co, Ni) for improving cathode performance has been confirmed, (2008).
7. S.-H. Kang and M. M. Thackeray, Electrochem. Commun., 11, 748 (2009).
with the most pronounced differences in performance being 8. S. H. Lee, B. K. Koo, J.-C. Kim, and K. M. Kim, J. Power Sources, 184, 276
observed when cycling half cells at varying temperatures. EIS data (2008).
show much steeper charge transfer resistance increases as a function 9. Y. Wu and A. Manthiram, Solid State Ionics, 180, 50 (2009).
of temperature for uncoated Li1.17Mn0.56Ni0.135Co0.135O2 compared 10. J. Ying, C. Wan, and C. Jiang, J. Power Sources, 102, 162 (2001).
11. J. Liu and A. Manthiram, J. Electrochem. Soc., 156, A66 (2009).
with 1.5 wt % Al2O3-coated Li1.17Mn0.56Ni0.135Co0.135O2. PITT 12. A. M. Kannan and A. Manthiram, Electrochem. Solid State Lett., 5, A167 (2002).
measurements demonstrate that at room temperature Liþ diffusivity 13. Y. K. Sun, K. J. Hong, J. Prakash, and K. Amine, Electrochem. Commun., 4, 344
is greater for the uncoated material versus the Al2O3-coated mate- (2002).
rial, but at reduced temperatures the Liþ diffusivity is significantly 14. Y. Wu and A. Manthiram, Electrochem. Solid State Lett., 9, A221 (2006).
15. K. L. Norton, A. J. Lange, and P. R. Griffiths, J. High Resolut. Chromatogr., 14,
higher for the Al2O3-coated material. Tafel polarization measure- 225 (1991) and references therein.
ments reveal a similar trend in that specific exchange currents are 16. S. A. Yeboah, W. J. Yang, and P. R. Griffiths, Proc. SPIE-Int. Soc. Opt. Eng., 289,
higher at room temperature for the uncoated material, yet at reduced 118 (1981) and references therein.
temperature the Al2O3-coated material sustains greater specific 17. P. R. Griffiths and J. A. de Haseth, Fourier Transform Infrared Spectrometry, John
Wiley & Sons, New York (1986).
exchange currents. The activation energies of Liþ diffusion and spe- 18. M. M. Thackeray, S.-H. Kang, C. S. Johnson, J. T. Vaughey, R. Benedek, and
cific exchange current are nearly identical. This suggests that either S. A. Hackney, J. Mater. Chem., 17, 3052 (2007).
the charge transfer reaction or diffusion of the Liþ through the cath- 19. A. R. Armstrong, M. Holzapfel, P. Novak, C. S. Johnson, S.-H. Kang, M. M.
ode particles and cathode coatings is the rate limiting step for cur- Thackeray and P. G. Bruce, J. Am. Chem. Soc., 128, 8694 (2006).
20. Y. Wu and A. Manthiram, Solid State Ionics, 180, 50 (2009).
rent flow. If the charge transfer reaction is the rate limiting step, 21. D. Y. W. Yu, K. Yanagida, and H. Nakamura, J. Electrochem. Soc., 157, A1177
then the diffusion data may be spurious due to the violation of a key (2010).
assumption in the diffusion measurement interpretation. 22. Y. Wu, A. V. Murugan, and A. Manthiram, J. Electrochem. Soc., 155, A635
(2008).
23. C. Li, H. P. Zhang, L. J. Fu, H. Liu, Y. P. Wu, E. Rahm, R. Holze, and H. Q. Wu,
Acknowledgments Electrochem. Acta, 51, 3872 (2006).
This work was carried out at the Jet Propulsion Laboratory, Cali- 24. M. C. Smart, B. L. Lucht, and B. V. Ratnakumar, J. Electrochem. Soc., 155, A557
(2008).
fornia Institute of Technology, under contract with the National 25. J. Liu, Q. Wang, B. Reeja-Jayan, and A. Manthiram, Electrochem. Commun., 12,
Aeronautics and Space Administration. The authors thank J. Kulleck 750–753 (2010).
for carrying out the SEM, EDS, and XRD measurements. The 26. J. Liu and A. Manthiram, Chem. Mater., 21, 1695 (2009).
authors acknowledge the funding support of NASA’s Exploration 27. R. Alcantara, M. Jarabe, P. Lavela, and J. L. Tirado, J. Electroanal. Chem., 566,
187 (2004).
Technology Development Program. In addition, the authors wish to 28. C. J. Wen, B. A. Boukamp, R. A. Huggins, and W. Weppner,, J. Electrochem.
acknowledge the useful discussions with Prof. S. R. Narayanan of Soc., 126, 2258 (1979).
the University of Southern California. 29. E. Deiss, Electrochim. Acta, 47, 4027 (2002).

Downloaded on 2014-07-04 to IP 128.6.218.72 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).