-
Electrochimica Acta 55 (2010) 832837
Contents lists available at ScienceDirect
Electrochimica Acta
journa l homepage: www.e lsev ier .com
Electro -sivia the ic L
X. Fang, YCAS Key Labora ing, U
a r t i c l
Article history:Received 4 JulReceived in re15 SeptemberAccepted
15 SAvailable onlin
Keywords:Lithium nickelEutecticLow temperatCapacity retenLithium
batter
partcetatC. Aft
perd at 9hemacity
ized Lt 10capac
1. Introduction
Lithium-ion batteries are one of the most successful
powersources and have dominated the portable electronic device
mar-ket for thefast develowide appliccontinuedcapacities arials,
such aspinel LiNi0LiNi0.5Mn1.5Amine et avoltage
plaLiNi0.5Mn1.5tial, the eneLiCoO2. Thufor use in el
A varieLiNi0.5Mn1.5tion [4], soelectrophorKimet al. [1by the
mo
CorresponE-mail add
and Li/Ni/Mn hydroxides; the synthesized LiNi0.5Mn1.5O4 pow-ders
show excellent cycling performance. In this study, we choseto use
acetates as starting materials; these can form a ternaryeutectic
LiNiMn acetate below 80 C. Though acetates have been
0013-4686/$ doi:10.1016/j.past two decades. Nonetheless, to keep
up with thepment of the laptop central processing unit and theation
of 3G techniques to cell phones, people haveto search for new
electrode materials with higherndmore power. Compared to
traditional cathodemate-s LiCoO2 (3.9V), LiMn2O4 (4.1V) and LiFePO4
(3.5V),.5Mn1.5O4 has a higher voltage (4.7V) [1]. In 1996,O4 was
rst reported to be a 3V cathode material by
l. [2]. Later, Dahn and coworkers discovered the 4.7Vteau of
LiNi0.5Mn1.5O4 [3]. The theoretical capacity ofO4 is 146.7mAhg1;
due to its high working poten-rgy density of LiNi0.5Mn1.5O4 is 20%
higher than that ofs, LiNi0.5Mn1.5O4 is seen as a potential
cathodematerialectric vehicles and energy storage systems in the
future.ty of synthetic methods for the preparation ofO4 have been
reported; these include solid state reac-lgel [5], co-precipitation
[6,7], spray pyrolysis [8,9],etic deposition [10] and pulsed laser
deposition [11].2] prepared thewell-dened octahedral
LiNi0.5Mn1.5O4lten salt method starting with the mixture of
LiCl
ding author. Tel.: +86 551 3606971; fax: +86 551 3601592.ress:
[email protected] (C.H. Chen).
widely used in the literature, most studies used a
wet-chemicalroute assisted by organic materials, such as acrylic
acid [13], citricacid [14], poly(ethylene glycol) [15] and
poly(methyl methacry-late) [16]. There are also some reports of
studies using the so-calledsolgelmethod. Thismethod involvesrst
dissolving the acetatesin water and then evaporating the water to
obtain a gel [17,18].Webelieve that the abovemethod shouldproduce
the same ternaryLiNiMn acetate eutectic instead of the real gel. It
should be men-tioned thatprior toourwork, Lafont et al.
[19]observed this eutecticphenomenonof LiNiMnacetate,which
theycalled greenslurry,but no extensive investigationwas conducted.
In Lafonts work, thecapacity at 2Cwas observed to be only about
60mAhg1. Ourworkshows that by simply optimizing the sintering
temperature of theternary eutectic LiNiMn acetate, we can obtain
nano- and micro-sized LiNi0.5Mn1.5O4 particleswith a capacity of
about 100mAhg1
at 8C.
2. Experimental procedures
We mixed 2.488g (10mmol) nickel acetate (Ni(Ac)24H2O),7.352g
(30mmol) manganese acetate (Mn(Ac)24H2O) and 2.146g(21mmol) lithium
acetate (LiAc2H2O) and milled the mixture byhand in a mortar. Then,
the mixture was calcined at 300 C for 5h.Aftermilling byhand again,
the powderswere sintered in air at 300,
see front matter 2009 Elsevier Ltd. All rights
reserved.electacta.2009.09.046chemical properties of nano- and
micrormal decomposition of a ternary eutect
. Lu, N. Ding, X.Y. Feng, C. Liu, C.H. Chen
tory of Materials for Energy Conversion, Department of Materials
Science and Engineer
e i n f o
y 2009vised form2009eptember 2009e 22 September 2009
manganese oxide
ure performancetiony
a b s t r a c t
Nano- and micro-sized LiNi0.5Mn1.5O4eutectic LiNiMn acetate.
Lithium aeutectic LiNiMn acetate below 80
be obtained at an extremely low tem700 C, the particle size
increases, anof about 4m) are obtained. Electroc(sinteredat900
C)exhibit thebest capcapacity can still be reached. Nano-sat low
temperatures; when cycled aLiNi0.5Mn1.5O4 powders can deliver a/
locate /e lec tac ta
zed LiNi0.5Mn1.5O4 synthesizediNiMn acetate
niversity of Science and Technology of China, Anhui, Hefei
230026, China
icles are prepared via the thermal decomposition of a ternarye,
nickel acetate and manganese acetate can form a ternaryer further
calcination, nano-sized LiNi0.5Mn1.5O4 particles canature (500 C).
When the sintering temperature goes above00 C micro-sized
LiNi0.5Mn1.5O4 particles (with a diameter
ical tests show that the micro-sized LiNi0.5Mn1.5O4
powdersretentionat25 C, andafter100cycles, 97%of initial
dischargeiNi0.5Mn1.5O4 powders (sintered at 700 C) perform the
bestC and charged and discharged at a rate of 1C, nano-sizedity as
high as 110mAhg1.
2009 Elsevier Ltd. All rights reserved.
-
X. Fang et al. / Electrochimica Acta 55 (2010) 832837 833
Fig. 1. Images f LiNportion); (c) th bottle
400, 500, 6rate of 3 Cstructurestion (XRD,the range fthe
powder(SEM, JEOL-ucts were ein ethyleneweight ratiglovebox (Mwas
compoand poly(vion a multi-cbetween 2.8
3. Results
3.1. Ternary
The meand 80 C,NiAc24H2Owith an incthat amixtua ternary euand
d). On tput in a botmelts at an(Fig. 1b andof the formation of a
ternary eutectic LiNiMn acetate: (a) mechanical mixture oe
mechanical mixture (left) and MnNi acetates without mixing at 80 C;
(d) lying00, 700, 800, 900 and 1000 C for 10h (with a heatingmin1)
and allowed to cool naturally. The crystallineof the powders were
characterized by X-ray diffrac-Philips XPert Pro Super, Cu K
radiation) with 2 inrom 10 to 80. The morphology and composition
ofs were determined by scanning electronic microscopy6970). The
electrochemical characteristics of the prod-valuated with coin
cells (CR2032 size) of Li/1M LiPF6carbonate (EC) and dimethyl
carbonate (DMC, with ao of 1:1)/LiNi0.5Mn1.5O4 assembled in an
argon-lledBraunLabmaster 130). Thepositive electrode laminate
sed of LiNi0.5Mn1.5O4 (84wt.%), acetylene black (8wt.%)nylidene
uoride) (PVDF, 8wt.%). The cells were testedhannel battery test
system (Shenzhen Neware Co. Ltd.)and 5.1V (vs. Li+/Li).
and discussion
eutectic LiNiMn acetate
lting points of LiAc2H2O and MnAc24H2O are 70respectively.
Unlike lithium and manganese acetates,can be directly decomposed,
rather than being melted,
rease in temperature. In our experiment, we observedreof
LiAc2H2O,MnAc24H2OandNiAc24H2Ocan formtectic solution of LiNiMn
acetate at 80 C (Fig. 1a, che other hand, when MnAc24H2O and
NiAc24H2O aretle one after another without mixing, only
MnAc24H2Oelevated temperaturewhileNiAc24H2O remains a solidc).
Thus, the melting of the LiNiMn acetate-mixture
can be attrwhich guarlevel.
The TG-in Fig. 2. That 54 C is athe endothetion water,weight
lossto the decoreticalweigthe temper
Fig.iMn acetates; (b) Mn acetate (bottom portion) and Ni acetate
(topof LiNiMn acetate-mixture at 80 C.ibuted to the formation of a
ternary eutectic system,antees the mixing of Li, Ni and Mn atoms at
the atomic
DTA curves of the LiNiMn acetate-mixture are shownere are
several thermal steps: (i) the endothermic peakresult of the
formation of a ternary eutectic solution; (ii)rmic peak at 118 C
results from the loss of crystalliza-accompanied by a weight loss
of 26% (the theoreticalis 27.7%); (iii) the exothermic peak at 345
C is related
mposition of acetate, with a 41.4% weigh loss (the theo-ht loss
is 37.2%); (iv) theweight is nearly constantwhenature is above 370
C; (v) the small endothermic reac-
2. TG-DTA curves of the ternary eutectic LiNiMn acetate.
-
834 X. Fang et al. / Electrochimica Acta 55 (2010) 832837
tion at 750 C is a result of the emission of oxygen in
LiNi0.5Mn1.5O4(LiNi0.5Mn1.5O4) [20].
3.2. Structure and morphology
The XRD patterns of the samples sintered at different
tempera-tures (from 200 to 1000 C) are shown in Fig. 3. Due to the
similardiffraction patterns of the spinel and layered structures,
it is dif-cult to differentiate these structures for the samples
sintered at lowtemperatures (below 500 C). We believe that the
samples sinteredat high temperatures (beyond 500 C) are of a pure
spinel structure.This conclusion is supported by the
electrochemical test, as shownbelow. The peak width at half-height
decreases with an increase inthe sintering temperature, which
indicates improved crystallinity.
The SEM images of the as-synthesized samples are shown inFig. 4.
When the temperature is below 700 C, nanoparticles areobtained.
When the temperature increases to 800 C, the particlesgrow; at 900
C, the particle size is about 4m. When the tem-perature increases
to 1000 C, the particle size further increasesto about 10m. There
are some small particles on the surface ofthe sample sintered at
1000 C; these particles may have arisen
Fig. 3. XRD patterns of the samples sintered at different
temperatures (from 200 to1000 C).
ized samples.Fig. 4. SEM images of the as-synthes
-
X. Fang et al. / Electrochimica Acta 55 (2010) 832837 835
Fi ples (c
from the d[21].
3.3. Electro
The galvas-synthesiat 300 C haode (e.g. LiThis result i(LiMnO2)
a400 C, a voof LiNi0.5Mthat might[24,25]. TheMn ions. It sshould be
tto 500 C, thimpurity inshort voltaThis plateauin LiNi0.5Mnten as
LiNi0in the sampthe plateaudeliver a catemperaturto a submic800 C
canbeenrepeatat 800 C isthe lengththe plateauions to Mn3
+ ionto ae ofadveredu, theg. 5. Galvanostatic chargedischarge
curves (2nd cycle) of the as-synthesized sam
ecomposition of LiNi0.5Mn1.5O4 at high temperature
chemical performance
anostatic chargedischarge curves (2nd cycle) of the
of Mn4
leadingincreasfor thewhich900 Czed samples are shown in Fig. 5.
The sample sintereds a typical voltage prole of a layered-structure
cath-
MnO2) with a voltage plateau at around 4.0V [22,23].ndicates
that Mnn+ ions are only oxidized to trivalencet 300 C. When the
sintering temperature increases toltage plateau at 4.7V appears,
indicating the formationn1.5O4. At around 3V another voltage
plateau appearsbe due to the transition from LiMn2O4 to
Li2Mn2O4powder sintered at 400 Cmay still have some trivalenthould
be mentioned that in LiNi0.5Mn1.5O4, all Mn ionsetravalent. When
the sintering temperature increasese plateau at 3V almost
disappears, indicating that theLiMn2O4 has been converted into
LiNi0.5Mn1.5O4. A
ge plateau at around 4.1V is still observed, however.may be
attributed to the transition from Mn3+ to Mn4+
1.5O4 [26]; thus, the real composition should be writ-.5Mn1.5O4.
The plateau at 3V completely disappearsle sintered at 600 C and due
to the oxidation of Mn3+,at 4.1V shortens. The sample sintered at
700 C canpacity of as high as 127mAhg1. When the sinteringe is
above700 C, theparticle size increases signicantlyron-size (as
shown in Fig. 4). The sample sintered atonly deliver a capacity of
121mAhg1. This result hasedbysintering thesampleagain. Thecapacity
reductionlikely due to the growth of particles, which increasesof
the lithium diffusion path. It is also observed thatat 4.1V
lengthens, due to the reduction of some Mn4++ ions at high
temperatures [21]. During the reduction
129.2mAhAs a result,the samplefrom the 4.similar volt125mAhg
Fig. 6 shLiNi0.5Mn1.discharged
Fig. 6. Cycled at 25 C, charged and discharged at a rate of
1/3C).
s, the concentration of oxygen vacancy also increases,n increase
in electronic conductivity. Nevertheless, theelectronic
conductivity cannot completely compensaterse effect of increasing
the length of the diffusion path,ces capacity. As the temperature
further increases toas-synthesized sample delivers the highest
capacity of1g , yet 16.6% capacity comes from the 4.1V plateau.the
overall output energy is still lower than that fromsintered at 700
C, for which only 7.2% capacity comes1V plateau. The sample
sintered at 1000 C exhibits aage prole as that sintered at 900 C,
with a capacity of1.ows the cycling performance of the nano- and
micro-5O4 sintered at different temperatures, charged andat a rate
of 1C, cycled between 2.8 and 5.1V at room
ycling performance of nano- and micro-LiNi0.5Mn1.5O4 (at
1C).
-
836 X. Fang et al. / Electrochimica Acta 55 (2010) 832837
Fig.
temperaturexhibits theing cycling;still be reaccycle. Althoers a
high cmicro-900.dissolutionLiNi0.5Mn1.5during cyc600. As
narelativelybeinvestigated900 (chargemicro-sizedbility,
withmicro-900increases tofactor for thcan be enhasintering tetion;
at theLiNi0.5Mn1.5path, is bettcapacity ret
As a higseen as a poelectric vehformance atnano-700
achargediscThe voltagenicant drorate performat a rate ofcells
werewas foundwas still abLiNi0.5Mn1.5traditionalLiFePO4 [31to be
fast Lwas foundthat nano-stures.
alvanled at
Fig. 9. Rate performance of nano-700 and micro-900 at 10 C.
clusions
o- and micro-sized LiNi0.5Mn1.5O4 particles are
successfullysized via the thermal decomposition of a ternary
eutecticMnacetate.When the sintering temperature is below800 C,ized
LiNi0.5Mn1.5O4 powders are obtained. When the sinter-perature
further increases, the particles grow, and nally,
sized LiNi Mn O powders are obtained. Electrochemical7. Rate
performance of nano-700 and micro-900 at 25 C.
e. Obviously, the sample sintered at 900 C (micro-900)highest
capacity and the best capacity retention dur-after 100 cycles, 97%
of initial discharge capacity canhed, with a capacity loss of less
than 0.04mAhg1 perugh the sample sintered at 700 C (nano-700)
alsodeliv-apacity, the capacity retention is not as good as that
ofThe large surface area of the nano-700 accelerates theofNi andMn
ions,which results in the loss of capacity ofO4 during cycling
[27,28]. The ability to retain capacity
ling is: micro-900 submicro-800>nano-700>nano-no-700 and
micro-900 exhibit higher capacities andtter capacity retention,
their rate capabilitywas further. The rate performance at 25 C of
nano-700 andmicro-d at a rate of 1C) is shown in Fig. 7. Both nano-
andLiNi0.5Mn1.5O4 particles exhibit an excellent rate capa-a
capacity of about 100mAhg1 at 8C. At the low rate,displays a better
rate capability, though when the rate6C,nano-700 isbetter. At the
lowrate, the rate-limitinge rate performance is the electronic
conductivity,whichnced by the formation of oxygen vacancy at the
highermperature; thus, micro-900 is better under this condi-high
rate, the limiting factor is the lithium diffusion inO4, and thus
nano-700, which has a shorter diffusioner. Again, for long-time
cycling, micro-900 has a betterention than nano-700.h energy
density cathode material, LiNi0.5Mn1.5O4 istential cathodematerial
for electric vehicles and hybridicles in the future. Thus, it is
necessary to test its per-
Fig. 8. G900 (cyc
4. Con
NansyntheLiNinano-sing temmicro-low temperatures.We investigated
the performance ofnd micro-900 at 10 C. Fig. 8 shows the
galvanostaticharge curves (2nd cycle) of nano-700 and
micro-900.plateau still remains at about 4.6V, without a sig-
p like that found in LiMn2O4 [29]. Fig. 9 shows theance of the
two samples cycled at 10 C, charged1C (except for the rst 10
cycles, during which the
charged and discharged at a rate of 1/3C). Nano-700to have a
higher capacity, and at 1C the capacityout 110mAhg1. The capacity
retention of nano-sizedO4 at low temperatures is much better than
thosecathode materials such as LiMn2O4 [29], LiCoO2 [30],,32],
andevenV2O5 nanoberswhichhasbeen reportedi-ion conductor [33,34].
For micro-900, the capacityto be only about 70mAhg1 at 1C. Thus, we
believeized LiNi0.5Mn1.5O4 perform better at low tempera-
tests at roopowders sincycled at 2ity can stillpowders
siperatures,110mAhg
particles ex100mAhg
Acknowled
This stuChina (granProvince (gPlan of Acaostatic chargedischarge
curves (2nd cycle) of nano-700 and micro-10 C, charged and
discharged at a rate of 1/3C).0.5 1.5 4m temperature show that
micro-sized LiNi0.5Mn1.5O4tered at 900 C have the best capacity
retention when
5 C; after 100 cycles, 97% of initial discharge capac-be
reached. Nevertheless, nano-sized LiNi0.5Mn1.5O4
ntered at 700 C exhibit a higher capacity at low tem-and at a
rate of 1C it can still deliver a capacity of1 at 10 C. Both nano-
and micro-sized LiNi0.5Mn1.5O4hibit excellent rate capabilities,
with a capacity of about1 at 8C (at room temperature).
gements
dy was supported by National Science Foundation oft no.
20971117), the Education Department of Anhuirant no. KJ2009A142)
and the Solar Energy Operationdemia Sinica.
-
X. Fang et al. / Electrochimica Acta 55 (2010) 832837 837
References
[1] S. Patoux, L. Daniel, C. Bourbon, H. Lignier, C. Pagano, F.
Le Cras, S. Jouanneau,S. Martinet, J. Power Sources 189 (2009)
344.
[2] K. Amine, H. Tukamoto, H. Yasuda, Y. Fujita, J. Electrochem.
Soc. 143 (1996)1607.
[3] Q.M. Zhong, A. Bonakdarpour, M.J. Zhang, Y. Gao, J.R. Dahn,
J. Electrochem. Soc.144 (1997) 205.
[4] H. Fang, Z. Wang, X. Li, H. Guo, W. Peng, J. Power Sources
153 (2006) 174.[5] H. Liu, Y.P. Wu, E. Rahm, R. Holze, H.Q. Wu, J.
Solid State Electrochem. 8 (2004)
450.[6] Y.S. Lee, Y.K. Sun, S. Ota, T. Miyashita, M. Yoshio,
Electrochem. Commun. 4
(2002) 989.[7] Y. Fan, J. Wang, X. Ye, J. Zhang, Mater. Chem.
Phys. 103 (2007) 19.[8] S.-H. Park, Y.-K. Sun, Electrochim. Acta 50
(2004) 431.[9] D. Li, A. Ito, K. Kobayakawa, H. Noguchi, Y. Sato,
Electrochim. Acta 52 (2007)
1919.[10] A. Caballero, L. Hernn, M. Melero, J. Morales, R.
Moreno, B. Ferrari, J. Power
Sources 158 (2006) 583.[11] H. Xia, S.B. Tang, L. Lu, Y.S. Meng,
G. Ceder, Electrochim. Acta 52 (2007) 2822.[12] J.-H. Kim, S.-T.
Myung, Y.-K. Sun, Electrochim. Acta 49 (2004) 219.[13] S.B. Park,
W.S. Eorn, W.I. Cho I, H. Jang, J. Power Sources 159 (2006)
679.[14] N. Amdouni, K. Zaghib, F. Gendron, A. Mauger, C.M. Julien,
Ionics 12 (2006) 117.[15] J.C. Arrebola, A. Caballero, M. Cruz, L.
Hernn, J. Morales, E.R. Castelln, Adv.
Funct. Mater. 16 (2006) 1904.[16] J.C. Arrebola, A. Caballero,
L. Hernn, J. Morales, J. Power Sources 180 (2008)
852.
[17] T. Nakamura, H. Demidzu, Y. Yamada, J. Phys. Chem. Solids
69 (2008) 2349.[18] G. Du, Y. NuLi, J. Yang, J. Wang, Mater. Res.
Bull. 43 (2008) 3607.[19] U. Lafont, C. Locati, E.M. Kelder, Solid
State Ionics 177 (2006) 3023.[20] M.G. Lazarraga, L. Pascual, H.
Gadjov, D. Kovacheva, K. Petrov, J.M. Amarilla,
R.M. Rojas, M.A. Martin-Luengo, J.M. Rojo, J. Mater. Chem. 14
(2004) 1640.[21] M.I. Zaki, M.A. Hasan, L. Pasupulety, K. Kumari,
Thermochim. Acta 2 (1997) 171.[22] T. Nohma, H. Kurokawa, M.
Uehara, M. Takahashi, K. Nishio, T. Saito, J. Power
Sources 54 (1995) 522.[23] A.R. Armstrong, P.G. Bruce, Nature
381 (1996) 499.[24] M.M. Thackeray, W.I.F. David, P.G. Bruce, J.B.
Goodenough, Mater. Res. Bull. 18
(1983) 461.[25] T. Ohzuku, M. Kitagawa, T. Hirai, J.
Electrochem. Soc. 137 (1990) 769.[26] J.-H. Kim, S.-T. Myung, C.S.
Yoon, S.G. Kang, Y.-K. Sun, Chem. Mater. 16 (2004)
906.[27] Y. Talyosef, B. Markovsky, G. Salitra, D. Aurbach,
H.-J. Kim, S. Choi, J. Power
Sources 146 (2005) 664.[28] X. Fang, N. Ding, X.Y. Feng, Y. Lu,
C.H. Chen, Electrochim. Acta,
doi:10.1016/j.electacta.2009.09.046.[29] R.A. Marsh, S. Vukson,
S. Surampudi, B.V. Ratnakumar, M.C. Smart, M. Manzo,
P.J. Dalton, J. Power Sources 97 (2001) 25.[30] Y. Ein-Eli, R.C.
Urian, W. Wen, S. Mukerjee, Electrochim. Acta 50 (2005) 1931.[31]
S.S. Zhang, K. Xu, T.R. Jow, J. Power Sources 159 (2006) 702.[32]
X.Z. Liao, Z.F. Ma, Q. Gong, Y.S. He, L. Pei, L.J. Zeng,
Electrochem. Commun. 10
(2008) 691.[33] C.R. Sider, C.R. Martin, Adv. Mater. 17 (2005)
125.[34] C.K. Chan, H. Peng, R.D. Twesten, K. Jarausch, X.F. Zhang,
Y. Cui, Nano Lett. 7
(2007) 490.