-
ORIGINAL PAPER
Polymer template-assisted microemulsion synthesis of
largesurface area, porous Li2MnO3 and its characterizationas a
positive electrode material of Li-ion cells
Tirupathi Rao Penki & D. Shanmughasundaram &N.
Munichandraiah
Received: 20 June 2013 /Revised: 6 August 2013 /Accepted: 9
August 2013# Springer-Verlag Berlin Heidelberg 2013
Abstract Lithium-rich manganese oxide (Li2MnO3) is pre-pared by
reverse microemulsion method employing Pluronicacid (P123) as a
soft template and studied as a positiveelectrode material. The
as-prepared sample possesses goodcrystalline structure with a
broadly distributed mesoporositybut low surface area. As expected,
cyclic voltammetry andchargedischarge data indicate poor
electrochemical activity.However, the sample gains surface area
with narrowly distrib-uted mesoporosity and also electrochemical
activity aftertreating in 4 M H2SO4. A discharge capacity of
about160 mAh g1 is obtained. When the acid-treated sample isheated
at 300 C, the resulting porous sample with a largesurface area and
dual porosity provides a discharge capacity of240 mAh g1. The rate
capability study suggests that the sampleprovides about 150 mAh g1
at a specific discharge current of1.25 A g1. Although the cycling
stability is poor, the high ratecapability is attributed to porous
nature of the material.
Keywords Lithium-ion cell . Mesoporous . Lithium excessmanganese
oxide .Microemulsion route . Polymer template .
High rate capability
Introduction
Lithium-ion batteries have attracted global interest from
bothconsumers and researchers during the past a couple of
decades[1, 2]. The interest has arisen because of the extended
appli-cations, which are successful in small sizes at present
andanticipated in large sizes in the future. Although the
energydensity of the present Li-ion batteries is greater than that
of
Pb-acid batteries by about four times, future requirementssuch
as electric vehicle applications require still greater
energydensity. The next generation Li-ion batteries thus need
novelelectrode materials which can provide greater discharge
ca-pacity than the materials in use at present, in addition to
theneed that they should be safe, inexpensive, non-toxic,
andenvironmental-friendly.
The present Li-ion batteries employ positive electrode
ma-terials of the categoryLiCoO2, LiMn2O4, and LiFePO4either in
their pure state or with partial substitutions of thetransitional
metals. The discharge capacity values of LiCoO2,LiMn2O4,and LiFePO4
are 140, 130, and 170 mAh g
1, re-spectively [3]. Li-ion batteries with greater energy
densitythan the present batteries require positive electrode
materialsof greater discharge capacity. Compounds which can
storemore than one lithium atom per transition metal atom
areexpected to provide enhanced discharge capacity. Li2MnO3belongs
to this category of materials [4]. Li2MnO3 is consid-ered
isostructural to layered LiCoO2 and its formula can alsobe
represented as Li(Li0.33Mn0.67)O2. One third of the octahe-dral
sites meant for Mn in the crystal lattice are occupied by Liatoms.
On the basis of extraction of the total available Li inLi2MnO3, a
discharge capacity of 456 mAh g
1 is expected,provided the compound is electrochemically active.
Li2MnO3was synthesized in single-phase from the reaction of LiOHand
MnO2 [5]. By treating the compound with H2SO4 orHNO3, a discharge
capacity of 199 mAh g
1 was obtainedin the first chargedischarge cycle, which
decreased rapidly to143mAh g1 in the eighth cycle. Following this
report, severalpublications have appeared with varying capacity
values[614]. Initial discharge capacity values are generally
highfor the activated phases of Li2MnO3, but cycling instability
isobserved in all reports.
In addition to the high discharge capacity, an electrodematerial
needs to possess high rate capability for the purposeof fast charge
or/and discharge. Porous materials are expected
T. R. Penki :D. Shanmughasundaram :N. Munichandraiah
(*)Department of Inorganic and Physical Chemistry, Indian Institute
ofScience, Bangalore 560012, Indiae-mail:
[email protected]
J Solid State ElectrochemDOI 10.1007/s10008-013-2221-1
-
to possess high rate capability because the electrolyte cancreep
into particles and enhance the contact area of theelectroactive
surface with the electrolyte [15]. As a result,the material can
withstand an enhanced specific current duringchargedischarge
cycling. To the best of the authors' knowl-edge, there are no
reports on the synthesis of porous Li2MnO3.In the present study,
the lithium excess oxide is prepared byinverse microemulsion route
assisted by soft polymer tem-plate, namely, Pluronic acid (P123).
By treatment in diluteacid and heating, the compound gains a large
surface area anddual mesoporosity, which result in providing a high
initialdischarge capacity and also a high rate capability.
Experimental
High purity or analytical grade chemicals, namely lithiumnitrate
(Aldrich), manganese nitrate tetrahydrate (Aldrich),
Pluronic acid [P123, poly(EO)20-poly(PO)70-poly(EO)20,where EO
and PO are ethylene oxide and propylene oxideunits, respectively;
molecular weight, 5,800] (Aldrich), lith-ium dodecylsulfate (LDS,
Aldrich), cyclohexane (Merck), n -butanol (SD Fine Chemicals),
H2SO4 (SD Fine Chemicals),lithium ribbon (0.75 mm thickness,
Aldrich), acetylene black(AB, Alfa Aesar), poly(vinylidene
fluoride) (PVDF, Aldrich),1-methyl-2-pyrrolidinone (NMP, Aldrich)
and 1 M LiPF6dissolved in ethylene carbonate, diethyl carbonate
anddimethyl carbonate (2:1:2v /v ) electrolyte (Chameleon) wereused
as received.
Li2MnO3 was prepared by reverse microemulsion routeemploying
P123 as a soft template. The oil and aqueousphases were prepared
separately. For the oil phase, 1.0 gP123 was dissolved in a mixture
consisting of 51.2 ml cyclo-hexane and 6.2 ml n -butanol by
stirring for 2 h. Then, 0.225 gLDS was added and stirred for 3 h to
get a transparentsolution. Lithium nitrate (0.84 g) and manganese
nitratetetrahydrate (1.074 g) were dissolved in 15 ml
double-distilled water. About 20 % of excess of lithium nitrate
thanthe stoichiometric quantity was used. The aqueous phase
wastransferred to the oil phase and stirred for 12 h at
ambienttemperature. The emulsion was slowly evaporated at 110
C.Awhite gel was obtained. Samples of gel were calcined in airat
400, 500, 600, 800 C for 6 h. Red colored powder sampleswere
obtained.
For activation of Li2MnO3, 1.0 g of a sample was added to100 ml
of 4 M H2SO4 and stirred at ambient conditions fordifferent
durations from 2 to 24 h. The powder was separatedfrom the acid by
centrifugation and washed with double-distilled water thrice,
finally rinsed with acetone, and driedat 110 C for about 12 h. A
black colored powder wasobtained. The acid-treated Li2MnO3 samples
were heated ateither 300 or 500 C in air for 4 h. The color of the
samplesremained black.
The powder X-ray diffraction (XRD) patterns wererecorded using a
Bruker AXS D8 Advance X-ray diffractom-eter at 40 kVand 30 mA using
Cu Ka ( =1.5418 ) radiationsource. Nitrogen adsorptiondesorption
isotherms wererecorded at 196 C by using Micromeritics surface
areaanalyzer model ASAP 2020. The specific surface area
wascalculated using the BrunauerEmmettTeller (BET) method
100 200 300 400 500 600 700 8000
20
40
60
80
100
170 C
9 % loss
455 C
8 % loss
II region
240 C
I region
59 % loss
(iv)
(iii)
(ii)
Weig
ht lo
ss / %
Temperature / C
(i)
10 20 30 40 50 60 70 80
Inte
nsity
/ a.
u.
2 / degree
(iv)
(iii)
(ii)
(i)
(330)/
(061)
(060)
(202)
(132)
(201)
(131)
(130)
(110)
(020)
(001)
a
b
Fig. 1 a Thermogravimetry recorded at a heating rate of 10 C
min1 ofprecursor gel (i ), sample S5 (ii ), sample S5A6 (iii ), and
sampleS5A6H3 (iv ); and b powder XRD pattern of samples S4 (i ), S5
(ii ),S6 (iii ), and S8 (iv )
Table 1 Unit cell parameters obtained from XRD pattern
Sample a () b () c () () Cell volume (3)
S4 4.942 (3) 8.523 (4) 5.008 (4) 109.14 (4) 199.29 (2)
S5 4.934 (4) 8.534 (5) 5.015 (4) 109.20 (2) 199.35 (2)
S6 4.929 (2) 8.530 (2) 5.022 (2) 109.12 (1) 199.49 (1)
S8 4.935 (3) 8.537 (5) 5.026 (4) 109.27 (2) 199.87 (2)
S5A6 5.042 (4) 8.672 (6) 5.045 (4) 110.29 (3) 206.90 (3)
J Solid State Electrochem
-
in the relative pressure (p /p0) range 0.050.25 from adsorp-tion
branch of the isotherm. The pore size distribution wascalculated by
BarrettJoynerHalenda (BJH) method fromdesorption isotherm. The
morphology was examined using aFEI Co scanning electron microscope
(SEM) model Sirion.The chemical composition was analyzed by
inductive coupledplasma atomic emission spectroscopy using Varion
inductive-ly couple atomic emission spectrometer model
Vista-PRO.The elemental analysis for C and H was carried out by
usinga Thermo Finnigan FLASH EA 1112 CHN analyzer.Thermogravimetric
analysis (TGA) was recorded from ambi-ent temperature to 800 C at a
heating rate of 10 C min1
under the flow of O2 gas by using thermal analyzerNETZSCH model
TG 209 FI.
For fabrication of electrodes, the active material (80 wt%),AB
(15 wt%) and PVDF (5 wt%) were mixed in a mortar andfew drops of
NMP were added to obtain a slurry. Stainlesssteel disks (16 mm
diameter) were cleaned with water, etchedin 30 % dilute HNO3,
rinsed with double-distilled waterfollowed by acetone and
air-dried. The slurry was applied ona pre-treated stainless steel
disk and dried at 110 C undervacuum for 12 h. The mass of active
material was 35 mg cm2. Lithium metal foil was used as a counter
cumreference electrode and Celgard porous polypropylene
membrane (2400) was used as a separator. A commercialelectrolyte
of 1 M LiPF6 dissolved in ethylene carbonate,diethyl carbonate and
dimethyl carbonate (2:1:2v /v ) was usedas the electrolyte.
Coin-type cells CR2032 (Hohsen Corpora-tion, Japan) were assembled
in an argon-filled MBraun glovebox.
The cells were galvanostatically cycled in the voltage rangefrom
1.5 to 4.4 V at different current densities at room tem-perature.
Cyclic voltammetry and chargedischarge cyclingexperiments were
carried out using an EG&G potentiostatmodel Versastat and
Biologic potentiostat/galvanostat modelVMP3. Rate capability with
different current densities wasexamined by using Bitrode battery
cycling unit in an air-conditioned room at 221 C.
Results and discussion
Soft chemical synthesis by inverse microemulsion route pro-vides
a control over particle size of the product. By dispersinga small
volume of aqueous phase consisting of the reactants ina large
volume of non-aqueous phase, the reactants are con-fined to
micrometer sized reaction zones and particles of theproducts are
limited to the size of the aqueous droplets, which
Fig. 2 Scanning electronmicroscopy images of Li2MnO3samples a
S4, b S5, c S6, and dS8
J Solid State Electrochem
-
are stabilized by surfactant molecules. Sub-micrometer/nanosized
product particles are synthesized by this route[16]. The presence
of polymeric templates such as P123 inthe reaction medium
facilitates the product particles to devel-op porosity. The
presence of hydrophilic EO block and hy-drophobic PO block is
considered to be responsible for gen-erating porosity on the
product particles [17]. By combiningthe salient features of inverse
microemulsion and polymerictemplates, synthesis of porous,
sub-micrometer sizes cathodematerials, namely, LiFePO4 and
LiNi1/3Mn1/3Co1/3O2, weresynthesized in our laboratory [1820]. A
similar procedurewas adopted for preparation of Li2MnO3 in the
present work.
The gel obtained after evaporation of solvents at 110 Cwas
subjected to thermal analysis (Fig. 1a, curve i). There is
acontinuous loss of mass between ambient and about 240 Cdue to the
removal of solvents and decomposition of nitratesand organic
matter. About 58 % of weight loss is observed at240 C. The mass of
the sample is fairly constant between 240
and 450 C. There is about 8 % loss of mass between 240 and450 C
followed by stability up to 800 C. Therefore, samplesof the gel
were heated at several temperatures from 400 to800 C for 6 h. The
samples prepared at 400, 500, 600, and800 C are hereafter referred
to as S4, S5, S6 and S8, respec-tively. Thermogravimetric analysis
of the heated samples(Fig. 1a, curve ii shown typically for sample
S5) indicatesthermal stability of the compounds in the temperature
rangefrom ambient to 800 C.
Powder XRD patterns of the samples prepared at
differenttemperatures are shown in Fig. 1b, which are similar for
allsamples. The structure of Li2MnO3 was determined by usingsingle
crystal X-ray diffraction by Strobel and Lamber-Andron [21].
Li2MnO3 was described as Li[Li1/3Mn2/3]O2
Table 2 Conditions of preparation, BET surface area and pore
diameterof different samples
Sample Conditions of preparation BET surfacearea (m2 g1)
porediameter(nm)
S4 Microemulsion400 C heating 5.8 1045S5 Microemulsion500 C
heating 5.6 1025S6 Microemulsion600 C heating 5.3 1030S8
Microemulsion800 C heating 2.6 1050S5A2 Microemulsion500 C
heating
2 h acid treatment115 3.9
S5A6 Microemulsion500 C heating6 h acid treatment
89 3.9
S5A12 Microemulsion500 C heating12 h acid treatment
78 3.7
S5A24 Microemulsion500 C heating24 h acid treatment
74 3.5
S5A2H3 Microemulsion500 C heating2 h acid treatment3 h heatingat
300 C
49 3.6 and 6.1
S5A6H3 Microemulsion500 C heating6 h acid treatment3 h heatingat
300 C
61 3.6 and 7
S5A12H3 Microemulsion500 C heating12 h acid treatment3 hheating
at 300 C
44 2.4 and 3.7
S5A24H3 Microemulsion500 C heating24 h acid treatment3 hheating
at 300 C
42 3.7
Table 3 Elemental analyses. Mn and Li were estimated by
inductivecoupled plasma atomic emission spectroscopy; C and H by
CHNSanalysis, and O is the balance
Sample Weight % of elements
Mn Li C H O
S5 39.09 7.78 1.28 0.78 51.07
S5A6 49.06 4.70 0.894 1.31 44.03
S5A6H3 51.79 4.55 0.407 0.67 42.57
0 30 60 90 120 150
0.0000
0.0003
0.0006
0.0009
dVol
/dD
ia /
cm3
g-1
nm
-1
Pore diameter / nm
(ii) (iii)(iv) (i)
0.0 0.3 0.6 0.9
0
20
40
(iv)
(ii)
(iii)
Quan
tity
of N
2 ads
orb
ed /
cm3
g-1
(i)
Relative pressure (p/po)
a
b
Fig. 3 a Nitrogen adsorption/desorption isotherms and b pore
sizedistribution BJH curves of Li2MnO3 samples S4 (i), S5 (ii), S6
(iii),and S8 (iv) samples. In a , curves (ii), (iii), and (iv) are,
respectively,vertically shifted by 5, 10, and 15 units of y-axis
scale relative to theposition of curve (i)
J Solid State Electrochem
-
structure of O3-type where the octahedral sites of inter-slabare
occupied by lithium-ion and octahedral sites of the slab bylithium
and manganese ions in 1:2 ratio. The XRD patterns(Fig. 1b) agree
well with the standard pattern of layeredstructure (JCPDS file No.
841634). The (020) and (110)reflections in 2 range 2023 indicate
LiMn ordering inthe mixed cation layer and these superstructure
reflections aresignatures for Li2MnO3. The XRD patterns of all
samples(Fig. 1b) were indexed to Li2MnO3 and lattice constants
wereobtained (Table 1). Lattice constants are close to those
report-ed in the JCPDS file 841634 (a =4.937 ; b =8.532 ; andc
=5.03 ). Similar agreement in lattice constants wasreported for
Li2MnO3 prepared from aqueous sol gel meth-od [22]. The average
crystallite size of Li2MnO3 sampleswere calculated from diffraction
peaks of (001), (201), and(131) planes using Scherrer equation [23]
and the averagecrystallite size was 140 nm. The unit cell
parameters andalso the crystallite size were nearly the same for
all S4S8samples (Table 1).
SEM micrographs of the as-prepared samples of Li2MnO3are
presented in Fig. 2. The S5 sample appears to have layer-like
morphology with several layers aggregated together andedges
projecting upwards. The thickness of layer is about23 nm and length
is about 190 nm. With an increase intemperature of preparation,
morphology changes to poroussponge-like at 600 C (S6 sample) and
plate-like morphologyat 800 C (S8 sample). Thus, the temperature of
preparationinfluences the morphology, particle nature, and as well
as thesize.
Nitrogen adsorption/desorption isotherms and BJH poros-ity
curves are presented in Fig. 3. The adsorption and desorp-tion
branches do not merge in the pressure region p /p0 be-tween 0.50
and 0.99 for all samples suggesting porous natureof the samples.
The amount of N2 adsorbed at p /p
0=0.99 isabout 40 cm3 g1, which is considered as high. This is
attrib-uted to porosity of the materials. The porous nature is
alsoreflected in BJH curves (Fig. 3b). There is a broad
distributionof pores around 1040 nm diameter. The BETsurface area
andaverage pore diameter obtained for all samples are listed
inTable 2. The surface area of S5 samples is 5.5 m2 g1 withpore
diameter of 1020 nm. There is a marginal decrease insurface area by
increasing the temperature of preparation(Table 2). The porosity
acquired by the Li2MnO3 samples isattributed to the presence of the
polymeric template in thereaction medium of preparation.
Results of elemental analysis and the calculated composi-tion
for S5 sample, typically, are provided in Table 3. It islikely that
the origin for C and H is the polymer P123. Thecomposition is
calculated assuming that Mn is present as perthe intended
composition of Li2MnO3. The deficiency of Li(1.57 against intended
2.0) is probably due to loss of Li in theprocess of synthesis. It
is reported that the XRD patterns ofcompounds with Li/Mn ratio less
than 2 also correspond to thepattern of stoichiometric Li2MnO3
[24].
Cyclic voltammograms (not shown) of Li2MnO3 preparedat different
temperatures suggested poor electrochemical ac-tivity of the
compound. In general, the positive electrodematerials of Li-ion
cells exhibit well-defined reduction andoxidation current peaks of
cyclic voltammograms [25]. In thepresent study, redox current peaks
were absent for all as-prepared samples. Poor electrochemical
activity is alsoreflected in galvanostatic charge/discharge cycling
(Fig. 4a).The electrodes were subjected to chargedischarge
cyclingbetween 1.50 and 4.40 V at a specific current of 33 mA
g1.Although Li2MnO3 was reportedly [14] cycled between 1.50and 4.80
V, the potentials greater than 4.50 V are expected tobe undesirable
due to the possibility of decomposition of theelectrolyte and
evolution of gases inside sealed cells. Hence,the upper limit of
cycling is limited to 4.40 V in the presentstudy, similar to the
studies reported on Li2MnO3 electrodeswhich were cycled between
1.50 and 4.50 V in order to avoiddecomposition of the electrolyte
[26]. The discharge capacity
0 1 2 3 4 5 6 7 8 9 10 11
20
40
60
80
100
120
(i)
(iii)
(iv)Spec
ific d
ischa
rge c
apac
ity /
mAh
g-1
Cycle number
(ii)
0 20 40 60 80 100 1201.5
2.0
2.5
3.0
3.5
4.0
4.5
2st
ch
1st
dis
Pote
ntia
l / V
V
s Li
/Li+
Specific discharge capacity / mAh g-1
1st
ch
a
b
Fig. 4 a Chargedischarge curves of Li2MnO3 sample S5 at100 A cm2
(specific current 33 mA g1) and b discharge capacityvariation of
Li2MnO3 samples S4, S5, S6, and S8 on repeated chargedischarge
cycling at a specific current of 3033 mA g1
J Solid State Electrochem
-
obtained at a specific current of 33 mA g1 rate is in the
range95115 mAh g1 for S4, S5 and S6 samples (Fig. 4). Thecapacity
of S8 sample is only about 22 mAh g1. On repeatedchargedischarge
cycling (Fig. 4b), the discharge capacity ofS5 sample is fairly
stable at 100 mAh g1, whereas thecapacity of S4 and S6 samples
decreases gradually. Thecapacity of S8 sample is stable at about 10
mAh g1. Thesevalues are considerably lower than the theoretically
expectedvalue of 456 mAh g1. The poor electrochemical of
Li2MnO3samples is because Mn is already in +4 oxidation state
anddelithiation of it during charging process necessitates an
in-crease in the oxidation state to +5, which is unlikely to
exist[5]. As the discharge capacity to the extent of 100 mAh g1
isobtained from the S4, S5, and S6 samples, it is presumed
thatthese compounds are non-stoichiometric (Table 3) althoughthe
XRD patterns (Fig. 1b) match with the standard pattern. Itis likely
that the temperature of preparation influences stoichi-ometry of
the compound. It was reported that discharge ca-pacity of Li2MnO3
depended on the method of synthesis [25].An initial capacity of
about 70 mAh g1 was obtained whenLi2MnO3 was prepared from Mn3O4 at
900 C whereas lessthan 20 mAh g1 was obtained when it was prepared
from -MnOOH precursor [26].
It is known that Li2MnO3 can be converted into
electro-chemically active phase by treatment in acid [5]. By
treatmentin acid, the Li2MnO3 undergoes a partial dissolution of
Li2Othereby facilitating the insertion of Li+ ion into the
resultingsample during discharge. Thus, the sample gains
electrochem-ical activity. As the quantity of removable Li2O
depends onduration of acid treatment, it was attempted to activate
themesoporous Li2MnO3 samples by treatment in 4 M H2SO4solution for
different durations. The S5 and S8 samples were
treated in 4 M H2SO4 for a few hours, and then tested
forelectrochemical activity after washing and drying. It wasfound
that both the samples delivered higher discharge capac-ity than the
as-prepared samples, but the S5 sample deliveredhigher capacity
than the S8 sample after acid treatment.Hence, detailed
investigations were carried out with S5 sam-ple of Li2MnO3. Sample
S5 was subjected to treatment in 4 MH2SO4 for different durations.
Samples of S5, which weretreated for 2, 6, 12, and 24 h are
hereafter referred to asS5A2, S5A6, S5A12, and S5A24, respectively.
Thethermogravimetry data S5A6 sample (Fig. 1a, curve iii) sug-gests
a mass loss of 10 wt% at about 170 C. It is thus inferredthat the
sample gains protons or H2O to the extent of about10wt%. There is a
gradual mass loss between 170 and 800 C.The sample retains about 75
% of its initial mass at 800 C.Powder XRD pattern of S5A6 sample is
shown in Fig. 5a.There are some changes observed in the XRD pattern
(Fig. 5a)in comparison with the patterns of as-prepared
samples(Fig. 1b). The (001) reflection exhibits a split, the
superlatticestructure is slightly altered at 2 =23, a new peak is
devel-oped next to (130) reflection and the (131) peak is
diminished(Fig. 5a). Nevertheless, the unit cell parameters of
S5A6sample calculated on the basis of Li2MnO3 structure are
listedin Table 1. There is a marginal increase in the values of a ,
b , , and unit cell volume. Significant changes are not observedin
morphology (Fig. 5b) when compared with the data of theas-prepared
S5 sample (Fig. 2). However, marked changes areobserved in N2
adsorption/desorption isotherms and BJHporosity curves (Fig. 6).
BET surface area measured fromadsorption isotherms are
significantly greater than the valuesmeasured for the as-prepared
sample (Table 2). The loopbetween adsorption and desorption
isotherms (Fig. 6a) is
10 20 30 40 50 60 70 80
Inte
nsity
/ a.u
.
2 / degree
a bFig. 5 a Powder XRD patternand b SEM image of sampleS5A6
J Solid State Electrochem
-
wider in relation to the data of the as-prepared samples(Fig.
3). The quantity of N2 adsorbed by the acid-treatedsamples at p
/p0=0.99 is about 120 cm3 g1 (Fig. 6a), whichis three times greater
than the corresponding values for the as-prepared samples (Fig. 3).
Pore diameter decreases to a narrowrange at about 4 and pore volume
is also greater (Fig. 6b)than the as-prepared samples (Fig. 3). The
increased surfacearea and pore volume are thus attributed to the
acid treatmentof Li2MnO3, which is already mesoporous before
subjectingto the acid treatment. The chemical analysis of
S5A6sample (Table 3) indicates a decrease in the Li contentand also
in C and O contents. The quantity of H has increaseddue to acid
treatment, which is also reflected in TGA data(Fig. 1a, curve
iii).
The electrochemical results of acid-treated samples ofLi2MnO3
are presented in Fig. 7. Cyclic voltammogram ofS5A6 sample (Fig.
7a) shows an oxidation current peak at3.20 Vand a reduction current
peak at 2.80 V. Thus, the peak
potential separation is about 0.40 V, which is an indication
ofan irreversible nature of electrode process. In addition to
themajor oxidation current peak observed at 3.20 V, thereis a minor
oxidation peak at 4.20 V. Thus, the cyclicvoltammogram indicates
that the inactive phase of the as-prepared Li2MnO3 is converted
into electrochemically activephase on treating in 4 M H2SO4 for a
few hours, although thepeak potential separation is 0.40 V. The
chargedischargecurves (Fig. 7b) contain potential plateaus at 3.20
and2.80 V for charge and discharge process, respectively.
Thedischarge capacity calculated from Fig. 7b is 196 mAh g1
forsample S5A6. This value is significantly greater than the
valueobtained from the as-prepared sample (Fig. 4b). The
variationsof discharge capacity of acid-treated samples on
repeatedcycling at a specific current of 30 mA g1 are shown inFig.
7c. The discharge capacity values of samples S5A2,S5A6, S5A12, and
S5A24 in the first cycle are 179, 196,183, and 182 mAh g1,
respectively, and the correspondingvalues in the tenth cycle are
135, 146, 161, and 110 mAh g1.The coulombic chargedischarge
efficiency throughout thecycle-life test is greater than 95 % (Fig.
7c curve v, typicallyfor the sample S5A6). After mild acid
treatment, thus,Li2MnO3 samples gain electrochemical activity, but
the cy-cling stability is poor. A gradual change in
crystallographicstructure is perhaps responsible for the cycling
instability.
As thermogravimetry of the acid-treated sample (S5A6)indicated
the presence of impurities which were removed at170 C (Fig. 1a,
curve iii), attempts were made to heat thissample and to examine
the electrochemical properties. Sam-ples of S5A6were heated for 4 h
at 300 and 500 C, and testedfor chargedischarge capacity. The
sample heated at 300 Cprovided greater discharge capacity than the
sample heated at500 C. Therefore samples of S5A2, S5A6, S5A12,
andS5A24 were heated at 300 C for 4 h. The resulting samplesare
hereafter referred as S5A2H3, S5A6H3, S5A12H3, andS5A24H3,
respectively. The thermogravimetry (Fig. 1a, curveiv) of sample
S5A6H3 indicates that the sample is stable up to800 C and the
impurities present in sample S5A6 wereremoved by heating at 300 C
for 4 h.
Powder XRD pattern of the S5A6H3 sample, typically, isshown in
Fig. 8a. The patterns of the other heated samples aresimilar to
this pattern. The patterns of these samples aredifferent from the
patterns of the as-prepared samples(Fig. 1b). Yu and Yanagida
reported detailed structural analy-sis of Li2MnO3 and related
compounds, recently [24]. Duringacid treatment, there was a gradual
reduction of O3 peaks(ABCABC stacking) and an increase in the P3
peaks(AABBCC stacking) as supported by a shift of the mainXRD peak
from 2 =18.7 to 19.15, and also by the emer-gence of a peak at 2
=38.3 [24]. On the basis of XRDpatterns, Raman spectra, and TGA
results, it was concludedthat after acid treatment and heating,
Li2MnO3 transforms intospinel Li4Mn5O12 phase [24]. The SEM image
(Fig. 8b)
0 2 4 6 8 10 12 14
0.00
0.01
0.02
0.03
dVol
/dD
ia /
cm3
g-1
nm
-1
(iii)
(i)
(ii)
Pore diameter / nm
(iv)
0.0 0.2 0.4 0.6 0.8 1.00
20
40
60
80
100
120
140
(iii)(ii)
(iv)
Quan
tity
of N
2 ads
orbe
d / c
m3
g-1
Relative pressure (p/p0)
(i)
a
b
Fig. 6 a Nitrogen adsorption/desorption isotherms and b BJH
curves ofsamples S5A2 (i), S5A6 (ii), S5A12 (iii), and S5A24 (iv).
In a , curves(ii), (iii), and (iv) are, respectively, vertically
shifted by 5, 10, and 15 unitsof y-axis scale relative to the
position of curve (i)
J Solid State Electrochem
-
shows that the morphology of S5A6H3 sample is nearly thesame as
S5 and S5A6 samples (Fig. 2 and 5b). Nevertheless,significant
changes are observed in N2 adsorption/desorption
isotherms and BJH curves (Fig. 9). The gap between theadsorption
and desorption isotherms has increased (Fig. 9a).The quantity of N2
adsorbed at p /p
0=0.99 by S5A6H3
1.5 2.0 2.5 3.0 3.5 4.0 4.5
150
100
50
0
-50
-100
Curr
ent /
mA
cm-2
Potential / V vs Li/Li+
0 50 100 150 2001.5
2.0
2.5
3.0
3.5
4.0
4.5
2nd dis 1st
dis
2nd
ch
Pote
ntial
/ V V
s Li/L
i+
Specific discharge capacity / mAh g-1
1st
ch
0 2 4 6 8 100
50
100
150
200
0 1 2 3 4 5 6 7 8 9 10 110
20
40
60
80
100(v)
(iv)
(i)(ii)(iii)
Coul
ombi
c effi
cien
cy /
%
Spec
ific d
ischa
rge c
apac
ity /
mAh
g-1
Cycle number
a
b
c
Fig. 7 a Cyclic voltammogramof sample S5A6 at a sweep rate
of0.05 mV s1, b chargedischargecurves of sample S5A6 at aspecific
current of 30mA g1, andc variation of specific dischargecapacity of
samples S5A2 (i),S5A6 (ii), S5A612 (iii), andS5A24 (iv). Variation
ofcoulombic efficiency is shown ascurve (v), typically, for
sampleS5A6
J Solid State Electrochem
-
sample is about 160 cm3 g1, which is greater than the volume(120
cm3 g1) adsorbed by S5A6 sample (Fig. 6a) and signif-icantly
greater than the volume (40 cm3 g1) adsorbed by theas-prepared S5
sample (Fig. 3a). Furthermore, the presence oftwo kinds of pores is
observed in BJH curves (Fig. 9b). Poresof narrow size distribution
around 4 nm are present on allsamples S5A2H3S5A24H3, similar to the
acid-treated sam-ples (Fig. 6b). Additionally, another pore with
broad distribu-tion around 510 nm has evolved. Initiation of the
secondarypore is clearly visible for S5A2H3 sample (Fig. 9b, curve
i).Formation of the secondary pore around 7 nm is clearlyobserved
for S5A6H3 sample (Fig. 9, curve ii). For the sampleS5A12H3, the
secondary pore diameter decreases to 6 nmwith decreased pore volume
(Fig. 9b, curve iii) and it de-creases further for the S5A24H3
sample (Fig. 9b, curve iv).Dual porosity is beneficial for
electrode materials because thepores allow the electrolytes to
creep and tolerate volumeexpansion/contraction during
chargedischarge cycling.Thus, both the time of acid treatment and
heating influenceto the formation of dual porosity. The chemical
analysis ofS5A6H3 sample (Table 3) indicates a decrease in H and
Ccontent in relation to S5A6 sample. However, the Mn and Licontents
in the samples are nearly the same.
Electrochemistry results are presented in Fig. 10.
Cyclicvoltammogram of S5A6H6 sample recorded at a sweep rate of0.05
mV s1 indicates sharp reductionoxidation pair of peaksin the
potential region 2.803.20 V (Fig. 10a). In addition tothis pair of
peaks, there is another pair of small broad peaksappearing at
4.204.50 V region. This is perhaps due to thepresence of some
quantity of LiMn2O4 phase in the sample.The discharge profiles
(Fig. 10b) of S5A6H3 sample providesa minor potential plateau at
about 4.0 V and a major constant
10 20 30 40 50 60 70 802 / degree
Inte
nsity
/ a.u
a bFig. 8 a Powder XRD patternand b SEM image of
sampleS5A6H3
0 2 4 6 8 10 12 14 16 18 20
0.000
0.007
0.014
dVol
/dD
ia /
cm3
g-1
nm
-1
(iii)
(ii)
(iv)
Pore diameter / nm
(i)
-
0.0 0.2 0.4 0.6 0.8 1.00
50
100
150
(iv)(iii)(ii)Qu
antit
y of
N2
ads
orb
ed /
cm3 g
-1
Relative pressure (p/p0)
(i)
a
b
Fig. 9 a Nitrogen adsorption/desorption isotherm and b BJH
curves ofsamples S5A2H3 (i), S5A6H3 (ii), S5A12H3 (iii), and
S5A24H3 (iv). Ina , curves (ii), (iii), and (iv) are, respectively,
vertically shifted by 5, 10,and 15 units of y-axis scale relative
to the position of curve (i)
J Solid State Electrochem
-
0 50 100 150 200 2501.5
2.0
2.5
3.0
3.5
4.0
4.5
2nd dis
2nd ch
1st dis
Pote
ntia
l / V
V
s Li
/Li+
Specific discharge capacity / mAh g-1
1st ch
0
50
100
150
200
250
300
0 5 10 15 20 25 30 35 40 45 500
10
20
30
40
50
60
70
80
90
100
0 1 2 3 4 5 6 7 8 9 10 110
50
100
150
200
250
300
(iv)
(iii)
(ii)
Spec
ific
disc
harg
e ca
pacit
y / m
Ah g
-1
Cycle number
(i)
Coul
ombi
c effi
cienc
y / %
Sp
ecifi
c disc
harg
e cap
acity
/ m
Ah g-
1
Cycle number
1.5 2.0 2.5 3.0 3.5 4.0 4.5
-400
-200
0
200
400
600
Potential / V vs Li/Li+
Curr
ent d
ensit
y / m
A cm
-2
a
b
c
Fig. 10 a Cyclic voltammogramof at a sweep rate of 0.05 mV s1,b
chargedischarge curves at aspecific current of 30mA g1, andc
cycle-life data of sampleS5A6H3. Cycle-life data ofS5A2H3 (i),
S5A6H3 (ii),S5A12H3 (iii), and S5A24H3(iv) samples for 20 cycles at
aspecific current of 30 mA g1 areshown as inset in c
J Solid State Electrochem
-
potential plateau at 2.80 V, which are followed by a
gradualpotential fall from 2.80 to 1.50 V. The major charge
anddischarge plateaus observed in 2.803.00 V region agree withthe
results reported by Yu and Yanagida [24] for acid-treatedand heated
samples. The discharge capacity obtained from thefirst cycle is
about 240 mAh g1. However, there is a rapidcapacity decrease on
repeated chargedischarge cycling(Fig. 10c). Similar results are
obtained from all heated samples(Fig. 10c inset).
The results of rate capability study are presented in Fig.
11.For each current density, a fresh cell was employed and it
wassubjected to five chargedischarge cycles. At each current,there
is a decrease in capacity similar to the data presented inFig. 10c.
There is a gradual capacity decrease by increasingthe specific
current. It is interesting to observe that about150 mAh g1 of
initial capacity is delivered as a specificcurrent as high as 1.25
A g1. This high rate capacity isattributed to the porous nature of
the samples.
Conclusions
Lithium-rich manganese oxide (Li2MnO3) was prepared byreverse
microemulsion method employing P123 as a softtemplate and studied
as a positive electrode material. The as-prepared sample possessed
good crystalline structure with abroadly distributed mesoporosity
with poor electrochemicalactivity. However, the sample gained
surface area with nar-rowly distributed mesoporosity and also
electrochemical ac-tivity after treating in 4 M H2SO4. A discharge
capacity ofabout 160 mAh g1 was obtained. When the
acid-treatedsample was heated at 300 C, the resulting sample with a
largesurface area and dual porosity provided a discharge capacityof
240 mAh g1. The rate capability study suggested that the
sample provides about 150 mAh g1 at a specific dischargecurrent
of 1.25 A g1. Further work is in progress on dual-porosity
lithium-rich electrochemically stable composites ofLi2MnO3.
Acknowledgments The authors thank Renault Nissan Technology
andBusiness Centre India Pvt. Ltd. for financial support, and Dr.
Subramaniand Dr. Arockia Vimal for helpful discussions. The authors
also thank Dr.C. Shivakumara for his help in analysis of XRD
patterns.
References
1. Winter M, Brodd RJ (2004) What are batteries, fuel cells,
andsupercapacitors. Chem Rev 104:42454269
2. Matsuki K, Ozawa K (2009) Lithium ion batteries. Wiley,
Tokyo3. Wittingham MS (2004) Lithium batteries and cathode
materials.
Chem Rev 104:427143014. Thackeray MM (1997) Manganese oxides for
lithium batteries. Prog
Solid State Chem 25:1715. Rossouw MH, Thackeray MM (1991)
Lithium manganese oxides
from Li2MnO3 for rechargeable lithium battery applications.
MaterRes Bull 26:463473
6. Kalyani P, Chitra S, Mohan T, Gopukumar S (1999) Lithium
metalrechargeable cells using Li2MnO3 as the positive electrode. J
PowerSources 80:103106
7. Johnson CS, Korte SD, Vaughey JT, Thackeray MM, Bobinger
TE,Shao YH, Hackney SA (1999) Structural and electrochemical
anal-ysis of layered compounds from Li2MnO3. J Power Sources
8182:491495
8. Tang W, Kanoh H, Wang KOY (2000) Preparation of a new type
ofmanganese oxide by selective lithium extraction from
monoclinicLi2MnO3 for lithium rechargeable batteries. J Mater Sci
Lett19:13611363
9. Robertson AD, Bruce PG (2002) The origin of
electrochemicalactivity in Li2MnO3. Chem Comm 23:27902791
10. Armstrong AR, Robertson AD, Bruce PG (2005)
Overchargingmanganese oxides: extracting lithium beyondMn4+. J
Power Sources146:275280
11. Pasero D, McLaren V, De Souza S, West AR (2005)
Oxygennonstoichiometry in Li2MnO3: an alternative explanation for
itsanomalous electrochemical activity. Chem Mater 17:345348
12. Park SH, Sato Y, Kim JK, Lee YS (2007) Powder property
andelectrochemical characterization of Li2MnO3 material. Mater
ChemPhys 102:225
13. Yu DYW, Yanagida K, Kato Y, Nakamura H (2009)
Electrochemicalactivities in Li2MnO3. J Electrochem Soc
156:A417A424
14. Lim J, Moon J, Gim J, Kim S, KimK, Song J, Kang J, ImWB, Kim
J(2012) Fully activated Li2MnO3 nanoparticles by oxidation
reaction.J Mater Chem 22:1177211777
15. Yang XY, Li Y, Lemaire A, Yu BJG, Su L (2009)
Hierarchicallystructured functional materials: synthesis strategies
for multimodalporous networks. Pure Appl Chem 81:22652307
16. Boutonnet M, Kizling J, Marie PSG (1982) The preparation
ofmonodisperse colloidal metal particles from microemulsions.
Col-loids and Surfaces 5:209
17. Kleitz F, Choi SH, Ryoo R (2003) Cubic Ia3d large
mesoporoussilica: synthesis and replication to platinum nanowires,
carbonnanorods and carbon nanotubes. Chem Comm 17:21362137
18. Sinha NN, Munichandraiah N (2009) Synthesis and
characterizationof carbon-coated LiNi1/3Co1/3Mn1/3O2 in a single
step by an inversemicroemulsion route. ACS Appl Mater Interface
1:12411249
0
50
100
150
200
250
300
5,05,05,05,05,05,05,05,0 5
125011217535183812491128457
5,0
Spec
ific d
ischa
rge c
apac
ity /
mAh
g-1
Cycle number0
28
Fig. 11 Rate capability of sample S5A6H6. Current density in
milliam-pere per gram is indicated. For each current density, a
fresh cell was used
J Solid State Electrochem
-
19. Sinha NN, Shivakumara C, Munichandraiah N (2010) High
ratecapability of a dual-porosity LiFePO4/C composite. ACS
ApplMaterInterface 2:20312038
20. Sinha NN, Munichandraiah N (2010) High rate capability of
porousLiNi1/3Co1/3Mn1/3O2 synthesized by polymer template route.
JElectrochem Soc 157:A647A653
21. Strobel P, AndronBL (1988) Crystallographic andmagnetic
structureof Li2MnO3. J Solid State Chem 75:9098
22. Jain G, Yang J, Balasubramanian M, Xu JJ (2005)
Synthesis,electrochemistry, and structural studies of lithium
intercalationof a nanocrystalline Li2MnO3-like compound. Chem Mater
17:38503860
23. Burton AW, Ong K, Rea T, Chan IY (2009) On the estimation
ofaverage crystallite size of zeolites from the Scherrer equation:
a criticalevaluation of its application to zeolites with
one-dimensional poresystems. Microporous Mesoporous Mater
117:7590
24. Yu DYW, Yanagida K (2011) Structural analysis of Li2MnO3
andrelated LiMnO materials. J Electrochem Soc 158:A1015A1022
25. Park B, Kim YJ and Cho J (2004) in: Lithium batteries:
Science andTechnology, G. A. Nazri, G. Pistoia. Kluwer, Boston
26. Park SH, Ahn HS, Park GJ, Kim J, Lee YS (2008)
Cyclemechanism and electrochemical properties of lithium
manganeseoxide prepared using different Mn sources. Mater Chem
Phys112:696701
J Solid State Electrochem
Polymer...AbstractIntroductionExperimentalResults and
discussionConclusionsReferences