J Solid State Electrochem (2012) 16:481–490

DOI 10.1007/s10008-011-1356-1

ORIGINAL PAPER

Effects of complexants on [Ni1/3Co1/3Mn1/3]CO3 morphology

and electrochemical performance of LiNi1/3Co1/3Mn1/3O2
Shunyi Yang & Xianyou Wang & Quanqi Chen &
Xiukang Yang & Jiaojiao Li & Qiliang Wei

Received: 11 November 2010 / Revised: 18 February 2011 / Accepted: 28 February 2011 / Published online: 6 April 2011
# Springer-Verlag 2011

Abstract LiNi1/3Co1/3Mn1/3O2 cathode materials for the Introduction

application of lithium ion batteries were synthesized by
carbonate co-precipitation routine using different ammonium In recent years, Li[NixCo1−2xMnx]O2(0<×<1/2) has been
salt as a complexant. The structures and morphologies of the reported as a promising cathode material for lithium ion
precursor [Ni1/3Co1/3Mn1/3]CO3 and LiNi1/3Co1/3Mn1/3O2 batteries [1–3]. In particular, LiNi1/3Co1/3Mn1/3O2 (x=1/3)
were investigated through X-ray diffraction, scanning elec- shows the best cell performance. The main advantages of
tron microscope, and transmission electron microscopy. The LiNi1/3Co1/3Mn1/3O2, compared with Li–M (M=Co, Mn, or
electrochemical properties of LiNi1/3Co1/3Mn1/3O2 were Ni) oxides, are its higher reversible capacity, improved
examined using charge/discharge cycling and cyclic voltam- thermal stability, and lowered toxicity [4–6]. Therefore, there
mogram tests. The results revealed that the microscopic is a great deal of interests in using LiNi1/3Co1/3Mn1/3O2 as
structures, particle size distribution, and the morphology the positive electrode material for high energy density and
properties of the precursor and electrochemical performance high power density Li-ion secondary batteries. Generally,
of LiNi1/3Co1/3Mn1/3O2 were primarily dependent on the LiNi1/3Co1/3Mn1/3O2 has mainly been prepared by solid-
complexant. Among all as-prepared LiNi1/3Co1/3Mn1/3O2 state reaction and hydroxide co-precipitation method.
cathode materials, the sample prepared from Na2CO3– The solid-state reaction method often results in an impure
NH4HCO3 routine using NH4HCO3 as the complexant and inhomogeneous product with inferior electrochemi-
showed the smallest irreversible capacity of 19.5 mAh g−1 cal performance [7–9]. This leads to the widespread
and highest discharge capacity of 178.4 mAh g−1 at the first adoption of the hydroxide co-precipitation method in
cycle as well as stable cycling performance (98.7% of the which a homogeneous precursor of [Ni1/3Co1/3Mn1/3]
initial capacity was retained after 50 cycles) at 0.1 C (OH)2 is prepared [10–14]. But the mixed transition
(20 mA g−1) in the voltage range of 2.5–4.4 V vs. Li+/Li. metal hydroxide is unstable in hydroxide co-precipitation
Moreover, it delivered high discharge capacity of over process; Mn(OH)2 (Mn2+) can easily be oxidized to
135 mAh g−1 at 5 C (1,000 mA g−1). MnOOH (Mn3+) or MnO2 (Mn4+) upon precipitation
conditions, which can decrease the homogeneity of the final
Keywords Lithium ion batteries . LiNi1/3Co1/3Mn1/3O2 . product. Therefore, preparing the precursor reproducibly is
Carbonate co-precipitation method . Complexant . Spherical very difficult [15].
particles Compared with solid-state and hydroxide co-precipitation
methods, the carbonate co-precipitation method results in a
homogeneous and phase-pure precursor of [Ni1/3Co1/3Mn1/3]
CO3 in which the manganese remains as Mn2+, and then it
S. Yang : X. Wang (*) : Q. Chen : X. Yang : J. Li : Q. Wei can be used for the synthesis of a more homogeneous and
Key Laboratory of Environmentally Friendly Chemistry pure LiNi1/3Co1/3Mn1/3O2 with high electrochemical perfor-
and Applications of Ministry of Education, School of Chemistry,
mance [4, 16–18]. During the carbonate co-precipitation
Xiangtan University,
Hunan Xiangtan 411105, China process, the morphology and size of the [Ni1/3Co1/3Mn1/3]
e-mail: [email protected] CO3 precursor were affected by operating conditions such as
482 J Solid State Electrochem (2012) 16:481–490

reaction temperature (T), crystallization time (t), stirring analyze the correlation lengths (the apparent crystallite sizes)
speed (r), and pH of the solution [19–23]. In addition, it has of the [Ni1/3Co1/3Mn1/3]CO3 and LiNi1/3Co1/3Mn1/3O2
been believed that the complexant is a critical factor to crystals in the 2D confinement, the Scherrer’s equation
influence the properties of [Ni1/3Co1/3Mn1/3]CO3 precursors was used:
and LiNi1/3Co1/3Mn1/3O2 final product.
In this work, the effects of the complexants (including Kl
NH4OH, (NH4)2SO4, and NH4HCO3 as sources of NH3 Dhkl ¼ ð1Þ
bhkl cos q
ligand) on the particle size distribution and morphology of
the [Ni1/3Co1/3Mn1/3]CO3 precursor were discussed. The
structure, morphology, and electrochemical performance of Where Dhkl is the mean crystallite size along the (hkl)
the as-prepared LiNi1/3Co1/3Mn1/3O2 were also studied in direction, K is the shape factor (the Scherrer constant, a
detail. value of 0.94 was used in this study [24]), βhkl is the line
breadth, and θ is the half-scattering angle. Usually, βhkl is
taken as the half width at half maximum of the (hkl)
Experimental diffraction. The morphology of the sample was observed
using scanning electron microscopy (SEM; JSM-5600LV,
Spherical [Ni1/3Co1/3Mn1/3]CO3 carbonate precursor was JEOL, Japan) and transmission electron microscopy (TEM;
first synthesized by carbonate co-precipitation method. A Tecnai G2, FEI, America). The particle size and particle
1.8-M mixture solution of NiSO4, CoSO4, and MnSO4 (Ni/ size distribution were measured by Mastersizer-2000
Mn/Co=1:1:1 M ratio) was pumped into a continuous- (Malvern Instruments Ltd, England).
flow-stirred tank reactor with a 3-L capacity. At the same The electrochemical tests of LiNi1/3Co1/3Mn1/3O2
time, a 1.8-M aqueous solution of Na2CO3 and a desired were carried out using coin cells assembled in an
amount of complexant (NH 4 OH, (NH 4 ) 2 SO 4 , and argon-filled glove box. In all cells, the cathode com-
NH4HCO3 solutions) with the same concentration of prised a mixture of active material (80 wt.%), acetylene
NH4+ ions were separately fed into the reactor. The mixture black (10 wt.%), graphite (5 wt.%), and polyvinylidene
solution was stirred at a speed of 1,000 rpm for 12 h at 60 ° fluoride (5 wt.%) as binder agent, lithium was served as
C and pH=7.5 to produce the precipitation. Then, the counter and reference electrodes, a Celgard 2400 was
precipitation was filtered, washed, and dried in a vacuum used as separator, and the electrolyte was a 1 M LiPF6
oven at 40 °C for several hours. Thus, the prepared powder solution in ethylene carbonate–dimethyl carbonate (1:1
was further dried at 110 °C to remove the adsorbed water v/v). Charge–discharge measurement was carried out in
and to obtain precursor carbonate powder [Ni1/3Co1/3Mn1/3] Neware battery test system BTS-XWJ-6.44S-00052
CO3. The [Ni1/3Co1/3Mn1/3]CO3 was fired at 500 °C for 5 h (Newell, Shenzhen, China) at different current densities
to decompose the carbonate into an oxide compound. at room temperature. Cyclic voltammogram was mea-
Finally, the obtained de-carbonated oxide compound sured at a scan rate of 0.1 mV s−1 between 2.5 and 4.8 V
powder was mixed thoroughly with 5% excess amounts of using a CHI600A Electrochemical Analyzer (Chinstr,
LiOH·H2O (molar ratio of 1.05) to compensate the Shanghai, China).
calcining loss. The mixture was preheated at 500 °C for
5 h to melt the lithium salt and calcined at 900 °C for 12 h
in air to obtain LiNi1/3Co1/3Mn1/3O2. The [Ni1/3Co1/3Mn1/3] Results and discussion
CO 3 prepared from Na 2 CO 3 –NH 4 OH, Na 2 CO 3 –
(NH4)2SO4, and Na2CO3–NH4HCO3 systems were, respec- Usually, during the precipitation process the crystals go
tively, denoted as precursor AH, AS, and AB, and the through two stages: the formation of crystal nuclei and their
corresponding LiNi1/3Co1/3Mn1/3O2 were labeled as sample subsequent growth which includes each single-crystal
AH, AS, and AB, respectively. nucleus development and the agglomeration of some nuclei.
The chemical compositions of the resulting powders The above two processes determine the size and morphology
were analyzed by atomic absorption spectroscopy (AAS; of crystals. In order to get the homogeneous micron-spherical
Vario 6 Analytik Jena AG, Jena, Germany). The phase crystals [Ni1/3Co1/3Mn1/3]CO3, NH3 ligand was introduced
identification of the samples was performed with a to carbonate co-precipitation reaction system to keep the
diffractometer (D/Max-3C, Rigaku, Japan) using Cu Kα balance of crystal nucleation rate and crystal growth rate
radiation (l=1.54178 Å) and a graphite monochromator at [18–21]. In this work, NH4OH, (NH4)2SO4, and NH4HCO3
36 kV, 20 mA. The scanning rate was 8°/min, and the scanning were separately used as complexants to provide the NH3
range of diffraction angle (2θ) was 10°≤2θ≤80°. In order to ligand; the mechanism of formation of the crystal nucleus
J Solid State Electrochem (2012) 16:481–490 483

[Ni1/3Co1/3Mn1/3]CO3 can be explained according to the

following two-step process [18]:

1=3Ni2þ ðaq:Þ þ 1=3Co2þ ðaq:Þ þ 1=3Mn2þ ðaq:Þ

þ nNH3  H2 Oðaq:Þ
 2þ
! Ni1=3 Co1=3 Mn1=3 ðNH3 Þn ðaq:Þ þ nH2 O ð2Þ

 2þ
Ni1=3 Co1=3 Mn1=3 ðNH3 Þn ðaq:Þ þ CO3 2 þ nH2 O
 
! Ni1=3 Co1=3 Mn1=3 CO3 ðs:Þ þ nNH3  H2 Oðaq:Þ ð3Þ

This mechanism suggests the formation of an interme- Fig. 2 Particle size distribution of the precursors [Ni1/3Co1/3Mn1/3]
diary, ammoniacal-mixed transition metal ligand, during the CO3
first step. The ligand becomes unstable in the presence of
an increasing concentration of carbonate, and replacement distribution of [Ni1/3Co1/3Mn1/3]CO3 precursors are shown
of the ammoniacal groups takes place during the second in Fig. 2. The average particle size of the precursor AH,
step. The uniformity was attributed to the action of the precursor AS and precursor AB is 19.54, 19.89, and
complexant, which basically prevents the phase separation 16.87 μm, respectively, which is consistent with the SEM
and leads to the formation of homogeneous and uniform observations. Compared with precursors AH and AS,
carbonate particles. In the previous reports [25, 26], the precursor AB prepared by Na2CO3–NH4HCO3 system has
morphology and homogeneity of the carbonate powders a much more narrow size distribution (diameter at 10% is
prepared in the absence of complexant were significantly 12.54 μm and the diameter at 90% is 22.93 μm), indicating
inferior. While in this work, homogeneous and spherical that in Na2CO3–NH4HCO3 system, a good pH buffer
[Ni1/3Co1/3Mn1/3]CO3 was successfully synthesized by solution can keep the balance of crystal nucleation rate
carbonate precipitation method using Na2CO3 precipitant and crystal growth rate, which is much better than in the
and different complexants (NH4OH, (NH4)2SO4, and other two precipitator-complexant systems.
NH4HCO3) as shown in Fig. 1. The microstructures of Figure 3 shows the X-ray diffraction (XRD) patterns of
[Ni1/3Co1/3Mn1/3]CO3 particles are strongly influenced by the [Ni1/3Co1/3Mn1/3]CO3 powders prepared from Na2CO3–
the complexants. SEM images of precursors presented in NH4OH, Na2CO3–(NH4)2SO4, and Na2CO3–NH4HCO3
Fig. 1 show that all [Ni1/3Co1/3Mn1/3]CO3 precursors are systems. The co-precipitated carbonate is single phased
spherical and homogeneous. However, precursor AH and has a typical structure corresponding to those of
prepared by Na2CO3–NH4OH system was more round NiCO3, CoCO3, and MnCO3, all of which have divalent
and smoother than precursors AS and AB, suggesting that transition metals in their formal charge. This structure was
the different complexants are responsible for the difference indexed to a hexagonal structure with a space group of
of precursors in shape. The particle size and particle size R3c[18]. The diffraction peaks were quite broad due to the

a b c

Fig. 1 Scanning electron micrographs of the precursors [Ni1/3Co1/3Mn1/3]CO3 a precursor AH, b precursor AS, and c precursor AB
484 J Solid State Electrochem (2012) 16:481–490

(0.69 Å) ions are similar, a partial disordering among the 3a

and 3b sites is expected and termed as “cation mixing”
[28]. It has been established that the cation mixing
deteriorates the electrochemical performance of the layered
oxide materials. Lattice parameters of the samples were
summarized in Table 2. It is well known that the c/3a ratio
of the lattice constants is a direct measure of deviation of
the lattice from a perfect cubic closed-packed lattice. A
pure layered lattice without transition metal in the
lithium layer has a c/3a ratio of 1.672, 1.639, and 1.648
for LiTiS2, LiNiO2, and LiNi0.5Mn0.5O2, respectively. The
LiMn1/3Ni1/3Co1/3O2 materials synthesized by NH4OH,
(NH4)2SO4, and NH4HCO3 have a c/3a ratio of 1.657,
1.656, and 1.657, respectively, implying that the material
has a more ideal ccp lattice than the previously synthe-
sized materials [28, 29]. According to Dahn’s report [30],
Fig. 3 XRD patterns of the [Ni1/3Co1/3Mn1/3]CO3 precursors and
pure MnCO3 (JCPDS:86–0173) the rI-factor defined by (I006 +I102)/I101 is an indicator of
the hexagonal ordering, the lower the rI, the better the
hexagonal ordering. Sample AB has the lowest rI, which
small particle size. According to the XRD patterns, the indicates better hexagonal ordering of its lattice than that
lattice parameters of a, c, and V were calculated by Bragg’s of the samples AH and AS. In addition, the ratio of I003/
equation and the D104 (the mean crystallite size along the I104 is also a parameter to characterize the cation mixing
(104) direction) was analyzed using Scherrer’s equation; and the decrease of this ratio indicates the increase of the
the results were listed in Table 1. The apparent mean structural deviation from hexagonal to cubic symmetry. It
crystallite sizes D104 of [Ni1/3Co1/3Mn1/3]CO3 powders was reported that the undesirable cation mixing would
calculated using Scherrer’s equation are around 7–10 nm, appear when the ratio of I003/I104 is smaller than 1.2, but
indicating that all the micron-spherical [Ni1/3Co1/3Mn1/3] the ratios of I003/I104 for the samples AH, AS, and AB are
CO3 particles are composed of nano-sized primary particles. 1.497, 1.508, and 1.548, respectively, which are larger than
From the TEM images of precursor [Ni1/3Co1/3Mn1/3]CO3 (as those reported by Li et al. [7], Shaju et al. [11], and Cho et al.
seen in Fig. 4), it can be seen that the size of the primary [25], suggesting less cation mixing in all LiNi1/3Mn1/3Co1/3O2
particles are around 5–15 nm; these results are in good samples prepared by our different carbonate precipitator-
agreement with the calculated values. complexant systems [31]. From the above analysis, it is
Figure 5 illustrates the XRD patterns of all Li[Mn1/3 concluded that the sample AB prepared by Na2CO3–
Ni1/3Co1/3]O2 materials prepared by Na2CO3–NH4OH, NH4HCO3 system displays a better structural integrity (lower
Na2CO3–(NH4)2SO4, and Na2CO3–NH4HCO3, respective- rI and higher ratio of I003/I104) than the sample AH, sample
ly. All XRD patterns can be indexed on the basis of the α- AS, and other materials reported in references [7, 11, 25].
NaFeO2 structure (space group: 166, R3m). The observed The chemical compositions of the resulting powders
peak splitting at (006)/(102) and (108)/(110), and the were analyzed by atomic absorption spectroscopy; the
corresponding XRD patterns are magnified in the inset of results were listed in Table 2. The AAS analysis identified
Fig. 5. The peaks are clearly split, indicating the formation that the element ratios of Li/Ni/Mn/Co for the samples AH,
of a well-ordered α-NaFeO2-type layered structure [27]. In AS, and AB are 1.011:0.336:0.330:0.334,
the layered structure, Li+, transition metal ions (Ni, Co, and 1.006:0.341:0.326:0.333, and 1.008:0.335:0.330:0.335,
Mn), and oxygen ion are situated in 3a, 3b, and 6c sites, respectively, the element ratio of Ni/Mn/Co is close to
respectively. Since the ionic radii of Li+ (0.76 Å) and Ni2+ 1:1:1, and the molar ratio of Li/M (M=Ni, Co, and Mn) is

Table 1 Lattice parameters of

the spherical precursors Precursor Lattice parameters
[Ni1/3Co1/3Mn1/3]CO3
a-Axis (Å) c-Axis (Å) Unit volume (Å)3 D104 (nm)

AH 4.79110 15.01674 298.514 8.907

AS 4.82572 14.88315 300.149 9.148
AB 4.75644 15.15505 296.920 7.753
J Solid State Electrochem (2012) 16:481–490 485

a b c

Fig. 4 TEM images of the precursor [Ni1/3Co1/3Mn1/3]CO3 a precursor AH, b precursor AS, and c precursor AB

also approaching 1:1, which are almost the same as the Normally, the crystal structure and morphology are two
values we designed. Therefore, we could successfully main factors which affect the electrochemical performance
prepare Li[Mn 1 / 3 Ni 1 / 3 Co 1 / 3 ]O 2 by carbonate co- of the materials. Galvanostatic charge/discharge experi-
precipitation routine. ments of the synthesized LiMn1/3Ni1/3Co1/3O2 materials
Figure 6 shows the scanning electron micrographs of the have been performed at a current density of 20 and
LiMn1/3Ni1/3Co1/3O2 materials synthesized from different 100 mA g−1 in the voltage range of 2.5–4.4 V vs. Li+/Li.
precipitator-complexant systems. It can be seen that the The charge/discharge curves of the LiNi1/3Co1/3Mn1/3O2
secondary particle of the prepared LiNi1/3Co1/3Mn1/3O2 has samples are presented in Figs. 7 and 8. It can be seen that
the same shape and size as those of the precursors [Ni1/ the first cycle of LiNi1/3Co1/3Mn1/3O2 electrodes differs
3Co1/3Mn1/3]CO3 even after the precursor was re-crystallized notably from successive cycles and shows more irreversible
with lithium salt during high-temperature calcination. Each of capacity loss. This might be attributed to the formation of a
the spherical particles is made up of a large number of nano- solid electrolyte interface on the surface of the electrode
sized primary grains as shown in Fig. 6a–c. From Scherrer’s and insufficient soaking of the electrode material during the
equation, the mean crystallite sizes was evaluated to be about first cycle [32]. The first charge/discharge specific capac-
70–90 nm as shown in Table 2. From Fig. 6d–f, it can also ities are 198.9/174, 197.3/171.4, 197.9/178.4 mAh g−1 with
be seen clearly that the primary grains of the sample AB a ratio of irreversible capacity loss of 12.52%, 13.13%, and
stacked more closely than those of the samples AH and AS, 9.85% for the samples AH, AS, and AB, respectively,
close contact between the primary grains will likely improve suggesting that the sample AB prepared by NH4HCO3
interparticle lithium ion movement, thereby enhancing the complexant exhibits the highest discharge capacity and
rate capability of the material. coulombic efficiency among all prepared samples. Further-
more, the sample AB exhibits a little smaller IR drop
between charge and discharge in the whole voltage range in
Fig. 8, compared with those at the samples AH and AS, the
cell exhibits very stable cyclability, the capacity retention
after 50 cycles reaches 98.70%. Figure 9 depicts the cycle
performance of the synthesized LiNi1/3Co1/3Mn1/3O2 mate-
rials at 100 mA g−1 current rates between 2.5 and 4.4 V. It
can be seen that the samples AH, AS, and AB exhibit the
initial discharge capacities of 164.8, 161.9, and
171.8 mAh g−1, respectively. The capacity retention ratios
of the samples AH, AS, and AB are 91.23%, 88.26%, and
93.68% after 100 cycles, respectively, the sample AB
synthesized from Na2CO3–NH4HCO3 system appears to
exhibit the most impressive electrochemical performance.
The best results reported by Shaju et al. [11] for Layered Li
[Ni1/3Co1/3Mn1/3]O2 prepared by mixed hydroxide method
do not exceed 160 mAh g−1 at a specific current of
Fig. 5 XRD patterns of the prepared LiNi1/3Co1/3Mn1/3O2 a sample 30 mA g−1 in the range of 2.5 and 4.4 V, and the retained
AH, b sample AS, and c sample AB capacity after 40 cycles was only 143 mAh g−1. In other
486 J Solid State Electrochem (2012) 16:481–490

Table 2 Chemical analysis, calculated structure parameters for LiNi1/3Co1/3Mn1/3O2 compounds

Sample Composition Lattice parameters c/3a (I006 +I102)/I101 I003/I104 D003 (nm)

a-Axis (Å) c-Axis (Å) Unit volume (Å)3

AH Li1.011 Ni0.336Mn0.330Co0.334O2 2.86018 14.21407 100.698 1.657 0.489 1.497 76.32

AS Li1.006 Ni0.341Mn0.326Co0.333O2 2.86126 14.20824 100.733 1.656 0.516 1.508 82.48
AB Li1.008 Ni0.335Mn0.330Co0.335O2 2.85944 14.21607 100.660 1.657 0.474 1.548 76.14
Literature [7] – 2.852 14.208 100.09 – – 1.25 –
Literature [11] – 2.864 14.233 101.102 – – 0.8 –
Literature [25] – 2.8604(6) 14.2453 (29) 100.9353 – – 1.45 –

respects, attractive properties have been recently reported current density of 16 mA g−1, and the capacity retention
with high tap density of Li[Ni1/3Co1/3Mn1/3]O2 synthesized after 30 cycles was 93% [10].
from uniform co-precipitated spherical metal hydroxide In order to evaluate the effect of complexant on the rate
[Ni1/3Co1/3Mn1/3](OH)2, the initial discharge capacity of capability of LiNi1/3Co1/3Mn1/3O2, the cells were charged
166.99 mAh g−1 was obtained between 3 and 4.3 V at a galvanostatically with a 0.2 C (40 mA g−1) rate before each

Fig. 6 SEM images of the

prepared LiNi1/3Co1/3Mn1/3O2 a d
a, d sample AH, b, e sample
AS, and c, f sample AB at
different magnifications

b e

c f
J Solid State Electrochem (2012) 16:481–490 487

Fig. 8 The second charge–discharge curves for LiNi1/3Co1/3Mn1/3O2

the samples at a current density of 20 mA g−1 between 2.5 and 4.4 V

discharge testing, then discharged at different C rates from

0.2 to 5 C stepwise, and finally returned to 0.2 C. The
discharge curves of Li/LiNi1/3Co1/3Mn1/3O2 cells at various
current densities (40–1,000 mA g−1) are shown in Fig. 10,
and the corresponding cycling performance are shown in
Fig. 11. The discharge capacity and average discharging
voltage of the sample AH prepared from Na2CO3–NH4OH
system decreased with increasing current density. The
capacity reached 91.9%, 83.3%, and 71.2% at 1, 2, and 5
C, respectively, compared with the specific capacity of
169.6 mA g−1 at 0.2 C. At the same time, the discharge
capacity of the sample AS prepared from Na2CO3–
(NH4)2SO4 system decreased more quickly than the sample
AH at high currents, only 80% and 60.2% at 2 and 5 C,
respectively, remained when compared with the specific

Fig. 7 First and 50th charge–discharge curves of LiNi1/3Co1/3Mn1/3O2

cathode a sample AH, b sample AS, and c sample AB at a current Fig. 9 Cycling performance of the prepared LiNi1/3Co1/3Mn1/3O2
density of 20 mA g−1 between 2.5 and 4.4 V sample AH (squares), sample AS (circles), and sample AB (triangles)
at a current density of 100 mA g−1 between 2.5 and 4.4 V
488 J Solid State Electrochem (2012) 16:481–490

Fig. 11 Cycle performance of Li/LiNi1/3Co1/3Mn1/3O2 cell at various

current densities between 2.5 and 4.4 V

to 5 C. It could be found that the sample AB delivers a

discharge capacity of 169.9 mAh g−1 at 0.5 C (the capacity
retention rate is about 97.6% as that of 0.2 C),
163.8 mAh g−1 at 1 C (the capacity retention rate is about
93.1% as that of 0.2 C), and 150.6 mAh g−1 at 2 C (the
capacity retention rate is about 86.5% as that of 0.2 C).
Even at 5 C (1,000 mA g−1), the capacity of the sample AB
is still as high as 135 mA g−1 and the capacity retention rate
is about 78.1% as that of 0.2 C. Obviously, it is clear that
the sample AB retains a high percentage of discharge
capacity at high charge/discharge rate during cycling
process, and its capacity could be recovered quickly after
the high-rate discharging process (98.5% capacity was
retained when compared with the first specific capacity of
174 mA g−1 at 0.2 C). This excellent capacity retention at a
high current density is mainly attributed to the lessened
extent of cationic mixing and good packing properties of
the secondary particles.
LiNi1/3Co1/3Mn1/3O2 is only partially de-intercalated and
intercalated when cells are charged and discharged in the
voltage range of 2.5–4.4 V. Higher discharge capacities can
be obtained when the voltage range is broadened, but more
rapid capacity fading will be brought. To illustrate these
phenomena and show the differences of the three samples
operated in broader voltage, the voltage range of 2.5–4.6 V
is adopted. Here, 4.6 V is chosen as the upper limit because
the critical upper limit of LiNiO2 and LiCoO2 is 4.6 V [33].
Figure 12 shows the cycling performance of the samples
Fig. 10 Discharge curves of Li/LiNi1/3Co1/3Mn1/3O2 cell at various AH, AS, and AB at a current density of 100 mA g−1
current densities between 2.5 and 4.4 V between the voltage range of 2.5 and 4.6 V. It is noticed
that the initial discharge capacities for the three samples are
capacity of 167.7 mA g−1 at 0.2 C. As shown in Fig. 10c, remarkably increased. However, their speed of capacity
the capacity of the sample AB prepared from Na2CO3– fading also increases. The initial discharge capacities of the
NH4HCO3 system decreases less than that of the first two samples AH, AS, and AB reach 185.1, 178.3, and
samples when the discharging current is increased from 0.2 193.4 mA g−1, respectively; however, the capacity retention
J Solid State Electrochem (2012) 16:481–490 489

of the sample AB, which in turn, ensures a reduced capacity

fade during battery cycling process.
From the above electrochemical charge/discharge, rate
capability and cyclic voltammogram tests, we conclude
that, of the three complexants tested herein, NH4HCO3
corresponds to the best complexant for preparation of
[Ni1/3Co1/3Mn1/3]CO3 as the precursor of LiNi1/3Co1/3Mn1/3O2
cathode material with excellent performance. Compared
with the layered Li[Ni1/3Mn1/3Co1/3]O2 prepared by
carbonate co-precipitation method without using complex-
ants reported by Cho [25], the sample AB prepared from
Na2CO3–NH4HCO3 system not only shows a better
structural integrity (higher ratio of I003/I104) but also some
improvement on electrochemical performance (higher
initial discharge capacity and better capacity retention).
Fig. 12 Cycling performance of the prepared LiNi1/3Co1/3Mn1/3O2
sample AH (squares), sample AS (circles), and sample AB (triangles)
at a current density of 100 mA g−1 between voltage range 2.5 and
4.6 V Conclusions

Layered LiNi1/3Co1/3Mn1/3O2 cathode materials were pre-

after 100 cycles are only 81.48%, 75.37%, and 85.97%. pared by carbonate co-precipitation methods using
These phenomena indicate that the different charge/dis- Na2CO3–NH4OH, Na2CO3–(NH4)2SO4, and Na2CO3–
charge voltage range may have different influence on the NH4HCO3 precipitator-complexant systems, respectively.
electrochemical performance of electrode material, which The XRD patterns and SEM images show that different
correspond to research results reported by Shaju et al. [11] complexants result in the differences in morphology and
and Ren et al. [19]. structure of LiNi1/3Co1/3Mn1/3O2. The LiNi1/3Co1/3Mn1/3O2
Cyclic voltammogram was carried out to monitor the prepared from Na 2 CO 3 –NH 4 HCO 3 routine using
electrochemical reactions during charge/discharge process NH4HCO3 as complexant has the highest capacity, best
in the voltage range of 2.5–4.8 V at a scan rate of cycle performance, and excellent rate capability. The initial
0.1 mV s−1; the first cycle curves of voltammogram for discharge capacity reaches 178.4 mA g−1, and the capacity
the samples AH, AS, and AB are shown in Fig. 13. It can retention after 50 cycles is 98.7% at a current of 20 mA g−1
be seen that all the three samples show the two major in the voltage range of 2.5–4.4 V. When the discharging
anodic peaks at 3.8–4.0 V and 4.5–4.7 V, whereas the current is increased from 0.2 to 5 C, it delivers a discharge
corresponding major cathodic peak is at 3.5–3.7 V and 4.4– capacity of 174 (0.2 C), 169.9 (0.5 C), 163.8 (1 C), 150.6
4.6 V. The redox processes of the first range roughly
correspond to the expected value for Ni2+/Ni4+ reaction,
and the latter redox range can be conclusively assigned to
the Co3+/Co4+ couple. This is a quite unusual value for the
Co3+/Co4+ redox couple in a layered compound, because in
similar layer-structured compounds LiCoO2, the Co3+/Co4+
couple usually appears at about 4 V and shows better
reversibility [34, 35]. However, in the layered compound Li
(Mn1−xCox)O2 and Li(Ni1−x−yCoxMny)O2, the Co3+/Co4+
couple appears at 4.4–4.6 V. This unusually high value for
the redox couple could have possibly been responsible for
the compounds to undergo structural transformation more
easily, exhibiting the associated capacity fading during
cycling in the upper voltage range near 4.6 V [36, 37]. In
addition, it also can be noticed that the sample AB displays
the sharper redox peak and smaller peak potential differences
(ΔE) between oxidation and reduction peaks positions than
Fig. 13 Cyclic voltammograms for the prepared LiNi1/3Co1/3Mn1/3O2
the samples AH and AS. This suggests that the better a sample AH, b sample AS, and c sample AB at scan rate of
reversibility of Li de-intercalation processes occur in the cell 0.1 mV s−1 between 2.5 and 4.8 V
490 J Solid State Electrochem (2012) 16:481–490

(2 C), and 135.9 mA g−1 (5 C), respectively. Therefore, 15. Lee M-H, Kanga Y-J, Myung S-T, Sun Y-K (2004) Electrochim
Acta 50:939–948
NH4HCO3 is a good choice of complexant for the
16. Lavela P, Sanchez L, Tirado JL, Bach S, Pereira-Ramos JP (1999)
preparation of [Ni1/3Co1/3Mn1/3]CO3 as the precursor of J Power Sources 84:75–79
LiNi 1/3Co1/3Mn1/3O 2 cathode material with excellent 17. Yoncheva M, Stoyanova R, Zhecheva E, Alcántara R, Tirado JL
performance. (2009) J Alloys Compd 475:96–101
18. Park SH, Kang SH, Belharouak I, Sun YK, Amine K (2008) J
Power Sources 177:177–183
Acknowledgments This work is funded by the National Natural
19. Ren HB, Huang YH, Wang YH, Li ZJ, Cai P, Peng ZH, Zhou YH
Science Foundation of China under project no. 20871101, Scientific
(2009) Mater Chem Phys 117:41–45
Research Fund of Hunan Provincial Education Department no.
20. Zhang S, Deng C, Fu BL, Yang SY, Ma L (2010) Powder Technol
09C947, Key Project of Science and Technology Department of
198:373–380
Hunan Province Government under project no. 2009WK2007, and
21. Zhang S, Deng C, Yang SY, Niu H (2009) J Alloys Compd
Colleges and Universities in Hunan Province plans to graduate
484:519–523
research and innovation under project no. CX2009B133.
22. Cho TH, Park SM, Yoshio M, Hirai T, Hideshima Y (2005) J
Power Sources 142:306–312
23. He P, Wang HR, Qi L, Osaka T (2006) J Power Sources 160:627–
References 632
24. Alexander LE (1969) X-ray diffraction methods in polymer
science. Wiley, New York
1. Ohzuku T, Makimura Y (2001) Chem Lett 30:642–643 25. Cho TH, Park SM, Yoshio M (2004) Chem Lett 33:704–705
2. MacNeil DD, Lu Z, Dahn JR (2002) J Electrochem Soc 149: 26. Park SM, Cho TH, Yoshio M (2004) Chem Lett 33:748–749
A1332–A1336 27. Park KS, Cho MH, Jin SJ, Nahm KS (2004) Electrochem Solid-State
3. Cheralathan KK, Kang NY, Park HS, Lee YJ, Choi WC, Ko YS, Lett 7:A239–A241
Park YK (2010) J Power Sources 195:1486–1494 28. Rougier A, Gravereau P, Delmas C (1996) J Electrochem Soc
4. Park SH, Shin HS, Myung ST, Yoon CS, Amine K, Sun YK 143:1168–1175
(2005) Chem Mater 17:6–8 29. Shin HS, Park SH, Bae YC, Sun YK (2005) Solid State Ionics
5. Sun YK, Myung ST, Park BC, Prakash J, Belharouak I, Amine K 176:2577–2581
(2009) Nat Mater 8:320–324 30. Reimers JN, Rossen E, Jones CD, Dahn JR (1993) Solid State
6. Shaju KM, Bruce PG (2006) Adv Mater 18:2330–2334 Ionics 61:335–344
7. Li DC, Muta T, Zhang LQ, Yoshio M, Noguchi H (2004) J Power 31. Ohzuku T, Ueda A, Nagayama M (1993) J Electrochem Soc
Sources 132:150–155 140:1862–1869
8. Kim JM, Chung HT (2004) Electrochim Acta 49:937–944 32. Wang ZX, Sun YC, Chen LQ, Huang XJ (2004) J Electrochem
9. Shizuka K, Kobayashi T, Okahara K, Okamoto K, Kanzaki S, Soc 151:A914–A921
Kanno R (2005) J Power Sources 146:589–593 33. Paulsen JM, Thomas CL, Dahn JR (2000) J Electrochem Soc
10. Luo XF, Wang XY, Liao L, Gamboa S, Sebastian PJ (2006) J 147:861–868
Power Sources 158:654–658 34. Ohzuku T, Ueda A, Nagayama M, Iwakoshi Y, Komori H (1993)
11. Shaju KM, Subba Rao GV, Chowdari BVR (2002) Electrochim Electrochim Acta 38:1159–1167
Acta 48:145–151 35. Madhavi S, Subba Rao GV, Chowdari BVR, Li SFY (2001) J
12. Zhang XY, Jiang WJ, Mauger A, Lu Q, Gendrond F, Julien CM Electrochem Soc 148:A1279–A1286
(2010) J Power Sources 195:1292–1301 36. Armstrong AR, Robertson AD, Gitzendanner R, Bruce PG (1999)
13. Zhang S (2007) Electrochim Acta 52:7337–7342 J Solid State Chem 145:549–556
14. Deng C, Liu L, Zhou W, Sun K, Sun D (2008) Electrochim Acta 37. Kajiyama A, Takada K, Inada T, Kouguchi M, Kondo S,
53:2441–2447 Watanabe M (2001) J Electrochem Soc 148:A981–A983