Effect of Zn doping on the performance of LiMnPO4 cathode for lithium ion batteries

Article abstract of DOI:10.1016/j.electacta.2012.03.160

In this work, effect of Zn doping on the performance of LiMnPO₄ is revisited. Samples of pure and Zn-doped LiMnPO₄ are synthesized by a new solid-state method, and their structure, morphology and electrochemical behavior are characterized and compared. The results reveal that a small amount of Zn doping (2 at.%) is highly beneficial for the performance of LiMnPO ₄ due to the reduced charge transfer resistance, the increased lithium ion diffusion and phase conversion, but this effect is remarkably traded off at a high level of Zn doping (10at.%). Compared with LiMnPO₄, LiMn_0.98Zn_0.02PO₄ has a much higher capacity and a much better rate capability. After 2 at.% Zn doping, the discharge capacity increases from 101 to 139 mAhg^-1 at 0.1 C and 56 to 105 mAhg^-1 at 2 C.

Full text of DOI:10.1016/j.electacta.2012.03.160

Electrochimica Acta 71 (2012) 266–269
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Effect of Zn doping on the performance of LiMnPO4 cathode for lithium ion
batteries
Haisheng Fanga,b,c,∗ , Huihua Yia,b,c , Chenglin Hua,b,c , Bin Yanga,b,c , Yaochun Yaoa,b,c , Wenhui Maa,b,c
,
Yongnian Daia,b,c
a Faculty of Metallurgical and Energy Engineering, Kunming University of Science and Technology, Kunming 650093, China
b National Engineering Laboratory for Vacuum Metallurgy, Kunming University of Science and Technology, Kunming 650093, China
c Key Laboratory of Nonferrous Metals Vacuum Metallurgy of Yunnan Province, Kunming University of Science and Technology, Kunming 650093, China
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work, effect of Zn doping on the performance of LiMnPO4 is revisited. Samples of pure and Zn-doped
LiMnPO4 are synthesized by a new solid-state method, and their structure, morphology and electrochem-
ical behavior are characterized and compared. The results reveal that a small amount of Zn doping (2 at.%)
is highly beneficial for the performance of LiMnPO4 due to the reduced charge transfer resistance, the
increased lithium ion diffusion and phase conversion, but this effect is remarkably traded off at a high
level of Zn doping (10 at.%). Compared with LiMnPO4, LiMn0.98Zn0.02PO4 has a much higher capacity
and a much better rate capability. After 2 at.% Zn doping, the discharge capacity increases from 101 to
139 mAh g−1 at 0.1 C and 56 to 105 mAh g−1 at 2 C.
Received 23 December 2011
Received in revised form 25 March 2012
Accepted 31 March 2012
Available online 6 April 2012
Keywords:
Lithium ion batteries
Cathode
Lithium manganese phosphate
© 2012 Elsevier Ltd. All rights reserved.
Doping
Zinc
Mg2+, Co2+, Ni2+, Cu2+, Zn2+, Ca2+, Ti4+, Zr4+ and their combination
have been attempted [14–21]. Among these cations, negative effect
1. Introduction
Lithium ion batteries are widely used in portable electronic
devices, but the commercial cobalt-containing cathode materials
are expensive, unsafe and toxic, which largely limit their applica-
tions in portable power tools, power supplies and hybrid electric
vehicles. Therefore, developing low-cost, safe and nontoxic cath-
ode materials to substitute cobalt-containing cathodes is highly
desired for the lithium ion batteries community. Among various
researched cathode materials, the cheap and environment-friendly
LiMnPO4 olivine appears to be quite attractive due to its high poten-
tial plateau of 4.1 V which is compatible with that of the present
cobalt-containing cathode materials [1,2]. Despite low electro-
chemical activity reported in the early literature owing to its low
electronic conductivity and slow lithium ion diffusion [3–5], the
performance of LiMnPO4 has recently been significantly improved
by the approaches including carbon coating [6–9], particle size
reduction [10–13] and cation doping or substitution [14–21].
For cation doping or substitution, most works focused on the
of Zn substitution (10 at.%) on the performance of LiMnPO4 was
reported by Chen et al. [14] and Wang et al. [18]. However, this
result is more or less out of expectation according to the mecha-
nism of substitute effect proposed by Chen et al. [14]. In fact, Chen
et al. [14] pointed out that there was some unidentified impurity
present in their Zn-substituted sample which might influence the
activity of the material and thus mask the effect of Zn doping. They
also pointed out that the impurity formation resulted from the limit
of synthesis reaction rather than the limit of solid solubility [14].
This is the case since 10 at.% Zn doping could be completely incorpo-
rated into the lattice of LiMnPO4 by a solid-state reaction according
to the report by Wang et al. [18]. These results indicate the impor-
tance of the synthesis method for cation doping or substitution. In
addition, as an electrochemically inactive ion, the Zn doping level
of 10 at.% in their experiments may be a little bit high [14,18]. Take
electrochemically inactive Mg doping for example, the optimum
doping level is less than 5 at.%, and further increase will worsen the
electrochemical performance of LiMnPO4 [16,19]. More recently, an
interesting paper was published on LiMn0.95Zn0.05PO4/C electrode
material which exhibited better electrochemical performance as
compared to LiMnPO4/C [22]. This result suggested a positive effect
of Zn doping on the electrochemical behavior of LiMnPO4.
Mn-site doping or substitution, and various cations such as Fe2+
,
∗
Corresponding author at: Faculty of Metallurgical and Energy Engineering, Kun-
ming University of Science and Technology, Kunming 650093, China.
Tel.: +86 871 5107208; fax: +86 871 5107208.
In the present work, to clarify the effect of Zn doping, Zn doped
LiMnPO4 materials with low (2 at.%) and high (10 at.%) level were
E-mail address: hsfang1981@yahoo.com.cn (H. Fang).
0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2012.03.160

H. Fang et al. / Electrochimica Acta 71 (2012) 266–269
267
Fig. 1. XRD patterns and cell volumes of LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4.
smaller than that of Mn2+ (0.80 A), the Zn doping causes an
obvious lattice contraction. As shown in Fig. 1b, the change of
the cell volume with the amount of Zn dopant is approximately
linear and follows Vergard’s law. These observations prove that
the doped Zn was successfully incorporated into the Mn-site of
the lattice to form LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4 solid
solutions. The AAS measurement shows that the atomic ratio of
Li:Mn:Zn is 1:0.968:0.02 for LiMn0.98Zn0.02PO4 and 1:0.888:0.097
for LiMn0.9Zn0.1PO4, which are much closer to the nominal values.
˚
prepared by a new method and the effect of Zn doping on the
electrochemical performance was revisited. Our electrochemical
results demonstrate that the performance of LiMnPO4 could be
significantly improved by a small amount of Zn doping, but a too
high level of Zn doping would reduce the reversible capacity of the
material.
2. Experimental
LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4 were synthe-
sized by a new solid-state method which was recently developed
for synthesizing LiMn0.9Fe0.05Mg0.05PO4 in our group [21]. Chem-
icals of LiH2PO4, MnC4H6O4·4H2O, H2C2O4·2H2O and ZnO was
mixed with 14 wt.% of sucrose by ball-milling for 6 h. The milled
mixture was dried and then heated at 700 ◦C for 10 h under an Ar
atmosphere.
The obtained samples were identified by X-ray diffraction (XRD)
utilizing Cu K␣ radiation to identify the crystalline phase. Lattice
parameters were refined using Rietica program. The morphology
and particle size of the samples were observed by scanning electron
microscopy (SEM, XL30, Philips) equipped with energy-dispersive
spectrometry (EDS, Phoenix). The particle size distribution is mea-
sured by a laser particle size analyzer. The chemical compositions
of the resulting material were determined using an atomic absorp-
tion spectroscopy (AAS). The specific surface areas of the samples
were determined by Brunauer–Emmett–Teller (BET) method with
nitrogen as adsorption gas.
Electrochemical measurement was conducted by assembly of
2025 coin-type cell with a lithium metal anode. The cathode was
made by mixing active material, Super P and polyvinylidene flu-
oride (PVDF) in a weight ratio of 8:1:1. The typical loading of
cathode with a diameter of 1.3 cm was about 2.2 mg of active mate-
rials. The electrolyte was 1 M LiPF6 in EC/DMC (1/1) solution. All
cells were assembled in an Ar-filled glove box. For rate test, the
cells were charged in a constant current–constant voltage (CC–CV)
mode (charged at 0.1 C to 4.5 V, held at 4.5 V until the current
decreased to 0.01 C) and then discharged at various rates to 2.0 V at
25 ◦C. The rate of 1 C is equal to a current density of 150 mA g−1 or
0.25 mA cm−2. Electrochemical impedance spectroscopy (EIS) was
carried out in a frequency range from 0.1 Hz to 100 kHz with an AC
signal of 5 mV.
Fig.
and LiMn0.9Zn0.1PO4, and EDS mappings of carbon for
LiMn0.98Zn0.02PO4 and zinc for LiMn0.98Zn0.02PO4 and
2 shows SEM images of LiMnPO4, LiMn0.98Zn0.02PO4
LiMn0.9Zn0.1PO4. The SEM images at low magnification (not
shown) exhibit that no significant differences in particle size and
morphology are observed for the three samples, and the particles
are mainly secondary particles composed of primary particles.
These observations are validated by the particle size distribution
measurement. The agglomerated particles of the three samples all
showed a bimodal particle size distribution. The first peak is sharp
and located at around 1.4 ␮m, while the second peak is broad and
located at around 6.4 ␮m. Fig. 2a–c shows SEM images at high
magnification and it is seen that the primary particles of all samples
are submicron crystals with an average size around 200 nm. EDS
mappings (Fig. 2d–f) indicate a uniform distribution of C and Zn in
the doped sample. The EDS mapping of Zn for LiMn0.9Zn0.1PO4 is
more dense than that for LiMn0.98Zn0.02PO4, indicative of a higher
content of Zn in the LiMn0.9Zn0.1PO4 sample. In addition, the
specific surface area determined by BET measurement is estimated
to be 42.9 m2 g−1 for LiMnPO4, 44.5 m2 g−1 for LiMn0.98Zn0.02PO4
and 42.0 m2 g−1 for LiMn0.9Zn0.1PO4. From the SEM observation,
particle size distribution and BET measurements, we can conclude
that the particle morphology and size are unaffected by Zn doping,
consistent to what was observed by Wang et al. [22].
To evaluate the effect of Zn doping, the electrochemical behavior
of LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4 was measured
and compared using 2025 coin-type cells. Fig. 3 shows electro-
chemical impedance spectra of LiMnPO4, LiMn0.98Zn0.02PO4 and
LiMn0.9Zn0.1PO4 recorded at the fully discharged state after 5 cycles
at 0.2 C (1 C equal to a current density of 150 mA g−1). The spec-
tra of the three samples have a similar profile composed of a
semicircle in a high-to-medium frequency region and an inclined
line in the low frequency region. The semicircle is approximately
related to the charge transfer process. Clearly, reduced semicir-
cles were observed for the two Zn doped LiMnPO4 samples, which
indicates decreased charge transfer resistances after Zn doping.
However, it is noted that the semicircle of the LiMn0.9Zn0.1PO4 lies
between those of LiMnPO4 and LiMn0.98Zn0.02PO4. In other words,
the charge transfer resistance of LiMn0.9Zn0.1PO4 is larger than that
of the LiMn0.98Zn0.02PO4, but smaller than that of LiMnPO4. The
inclined line of the spectra is attributed to the Warburg impedance
3. Results and discussion
XRD
patterns
of
LiMnPO4,
LiMn0.98Zn0.02PO4
and
LiMn0.9Zn0.1PO4 are shown in Fig. 1a. All samples have a similar
pattern which can be fully indexed into an orthorhombic struc-
ture with a space group of Pmnb. The absence of carbon peak
in the patterns indicates the amorphous nature of the carbon
pyrolyzed from sucrose. As the ionic radius of Zn2+ (0.74 A) is
˚

268
H. Fang et al. / Electrochimica Acta 71 (2012) 266–269
Fig. 2. SEM images of (a) LiMnPO4, (b) LiMn0.98Zn0.02PO4 and (c) LiMn0.9Zn0.1PO4; EDS mappings of (d) carbon for LiMn0.98Zn0.02PO4, zinc for (e) LiMn0.98Zn0.02PO4 and (f)
LiMn0.9Zn0.1PO4.
associated with the lithium ion diffusion in bulk of the electrode
materials, which is evidenced by linear relationship between Zꢀ
and ω−1/2 in the low frequency region as shown in the inset of
Fig. 3 [23]. According to the Warburg behavior in the low frequency,
lithium ion diffusion coefficient can be roughly determined (calcu-
lation equation and related parameters description can be found
elsewhere [23]). In the present work, the effective surface area
used for calculating diffusion coefficient is taken as one-third of the
total surface area because there is only one-dimensional channel
for lithium diffusion for LiMnPO4 based materials [24]. In addi-
tion, Li concentration used for calculation is 0.022 mol cm−3. On the
basis of calculation equation, the diffusion coefficient is estimated
to be 1.69 × 10−15 cm2 s−1 for LiMnPO4, 1.94 × 10−14 cm2 s−1 for
LiMn0.98Zn0.02PO4 and 2.64 × 10−15 cm2 s−1 for LiMn0.9Zn0.1PO4.
The above observations demonstrate that the Zn doping can facil-
itate charge transfer and lithium ion diffusion, but this effect is
remarkably traded off at a high level of Zn doping.
current decreased to 0.01 C and then discharged at various rates
to 2.0 V. Fig. 4a shows typical charge and discharge curves of
LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4 cycled at 0.1 C.
As shown in Fig. 4a, flat voltage plateaus around 4.1 V observed
in all curves indicate that the charge and discharge reaction pro-
ceeds in a two-phase manner. Fig. 4b exhibits rate capability of
LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4. It is evident that
the LiMn0.98Zn0.02PO4 sample has a much better rate capability as
compared to the LiMnPO4 sample. At 0.1 C, the discharge capacity
can reach 139 mAh g−1 for LiMn0.98Zn0.02PO4, while only a capac-
ity of 101 mAh g−1 is observed for LiMnPO4. More importantly, as
the discharge rate increases from 0.1 to 2 C, the LiMn0.98Zn0.02PO4
sample shows a much slower decrease in capacity. At 2 C, the
discharge capacity is 56 mAh g−1 for LiMnPO4 and 105 mAh g−1
for LiMn0.98Zn0.02PO4, which results in correspondingly a capacity
retention of 55% and 76%, respectively. Therefore, the electrochem-
ical performance of LiMnPO4 could be significantly enhanced by
a small amount of Zn doping, which is clearly different from the
observations reported by Chen et al. [14] and Wang et al. [18], and
confirms the positive effect of Zn doping reported by Wang et al.
[22].
Coupled with the results of EIS and the mechanism of substi-
tute effect proposed by Chen et al. [14], improved electrochemical
performance of Zn-doped LiMnPO4 is highly expected, which is
confirmed by the rate performance as shown in Fig. 4. For rate
test, cells were charged at 0.1 C to 4.5 V, held at 4.5 V until the
However, we should note that the Zn doping level was 10 at.%
in the experiments reported by Chen et al. [14] and Wang et al.
[18], which is higher than that (5 at.%) used by Wang et al. al. [22].
To address the question whether the too high level of the inac-
tive Zn doping causes the worsened electrochemical performance
of LiMnPO4, the electrochemical performance of LiMn0.9Zn0.1PO4
was also measured and comparatively shown in Fig. 4. It is seen
that the discharge capacities of LiMn0.9Zn0.1PO4 decrease dramat-
ically as compared with those of LiMn0.98Zn0.02PO4 but still higher
than those of LiMnPO4 especially at high rates. This demonstrates
that a too large amount of Zn doping will degrade the reversible
capacity due to the inactive nature of Zn doping, which is simi-
lar to the result observed for inactive Mg doping [16,19]. Wang
et al. [18] reported that 10 at.% Zn substitution led to a discharge
capacity decrease of more than 20 mAh g−1 at all rates. For exam-
ple, the discharge capacity at 0.1 C decreased from 95 to 62 mAh g−1
after 10 at.% Zn substitution. It is noted that the performance of
our LiMnPO4 is roughly similar to that reported by Wang et al.
(a precise comparison is difficult because the carbon content in
the composite, the cathode fabrication and the cell test condition
Fig. 3. Impedance spectra of LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4
recorded at the fully discharged state after 5 cycles at 0.2 C (1 C equal to a current
density of 150 mA g−1 or 0.25 mA cm−2). Inset: the relationship between Zꢀ and ω−1/2
in the low frequency region.

H. Fang et al. / Electrochimica Acta 71 (2012) 266–269
269
Fig. 4. (a) Typical charge/discharge curves of LiMnPO4, LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4 cycled at 0.1 C and (b) rate capability of LiMnPO4, LiMn0.98Zn0.02PO4 and
LiMn0.9Zn0.1PO4. Cells were charged at 0.1 C to 4.5 V, held at 4.5 V until the current decreased to 0.01 C and then discharged at various rates to 2.0 V.
are different) [18], but our LiMn0.9Zn0.1PO4 exhibits much higher
capacities at all rates as compared with their LiMn0.9Zn0.1PO4. We
speculate that such a difference may be highly due to the different
method used for preparation of Zn doped LiMnPO4. In addition, it is
interesting that the discrepancy of the discharge capacity between
LiMn0.98Zn0.02PO4 and LiMn0.9Zn0.1PO4 became smaller when the
discharge current density became larger. This behavior may possi-
bly be understood by the mechanism proposed by Chen et al. [14].
Similar to Mg doping, the presence of Zn ion may create a more
favorable boundary between the two phases for increased conver-
sion of lithiated phase to delithiated phase, and thus may enhance
the electrochemical kinetics and improve the electrochemical per-
formance of LiMnPO4. From this mechanism, LiMn0.9Zn0.1PO4 may
have a higher conversion rate of lithiated phase to delithiated phase
than LiMn0.98Zn0.02PO4, and then the increased phase conversion
of LiMn0.9Zn0.1PO4 may partly or completely counterbalance its
increased charge transfer resistance and reduced lithium ion dif-
fusion as revealed by EIS. As a result, the LiMn0.9Zn0.1PO4 still has a
good rate performance. Anyway, the above electrochemical results
prove that a proper amount of Zn doping is highly beneficial for the
electrochemical performance of LiMnPO4.
References
[1] A.K. Padhi, K.S. Nanjundaswamy, J.B. Goodenough, Journal of the Electrochem-
ical Society 144 (1997) 1188.
[2] G.H. Li, H. Azuma, M. Tohda, Electrochemical Solid-State Letters 5 (2002) A135.
[3] A. Yamada, Y. Kudo, K.Y. Liu, Journal of the Electrochemical Society 148 (2001)
A1153.
[4] A. Yamada, M. Hosoya, S.C. Chung, Y. Kudo, K. Hinokuma, K.Y. Liu, Y. Nishi,
Journal of Power Sources 119–121 (2003) 232.
[5] C. Delacourt, P. Poizot, M. Morcrette, J.M. Tarascon, C. Masquelier, Chemistry of
Materials 16 (2004) 93.
[6] S.K. Martha, B. Markovsky, J. Grinblat, Y. Gofer, O. Haik, E. Zinigrad, D. Aurbach, T.
Drezen, D. Wang, G. Deghenghi, I. Exnar, Journal of the Electrochemical Society
156 (2009) A541.
[7] S.M. Oh, S.W. Oh, C.S. Yoon, B. Scrosati, K. Amine, Y.K. Sun, Advanced Functional
Materials 20 (2010) 3260.
[8] Z. Bakenov, I. Taniguchi, Electrochemistry Communications 12 (2010) 75.
[9] S.M. Oh, S.W. Oh, S.T. Myung, S.M. Lee, Y.K. Sun, Journal of Alloys and Com-
pounds 506 (2010) 372.
[10] N.H. Kwon, T. Drezen, I. Exnar, I. Teerlinck, M. Isono, M. Graetzel, Electrochem-
ical Solid-State Letters 9 (2006) A277.
[11] T. Drezen, N. Kwon, P. Bowen, I. Teerlinck, M. Isono, I. Exnar, Journal of Power
Sources 174 (2007) 949.
[12] D.Y. Wang, H. Buqa, M. Crouzet, G. Deghenghi, T. Drezen, I. Exnar, N.H.
Kwon, J.H. Miners, L. Poletto, M. Grätzel, Journal of Power Sources 189 (2009)
624.
[13] D.W. Choi, D.H. Wang, I.T. Bae, J. Xiao, Z.M. Nie, W. Wang, V.V. Viswanathan,
Y.J. Lee, J.G. Zhang, G.L. Graff, Z.G. Yang, J. Liu, Nano Letters 10 (2010) 2799.
[14] G.Y. Chen, J.D. Wilcox, T.J. Richardson, Electrochemical Solid-State Letters 11
(2008) A190.
4. Conclusions
[15] S.K. Martha, J. Grinblat, O. Haik, E. Zinigrad, T. Drezen, J.H. Miners, I. Exnar, A.
Kay, B. Markovsky, D. Aurbach, Angewandte Chemie International Edition 48
(2009) 8559.
[16] T. Shiratsuchi, S. Okada, T. Doi, J.I. Yamaki, Electrochimica Acta 54 (2009) 3145.
[17] J. Kim, D.H. Seo, S.W. Kim, Y.U. Park, K. Kang, Chemical Communications 46
(2010) 1305.
[18] D.Y. Wang, C.Y. Ouyang, T. Drézen, I. Exnar, A. Kay, N.H. Kwon, P. Gouerec,
J.H. Miners, M.K. Wang, M. Grätzel, Journal of the Electrochemical Society 157
(2010) A225.
[19] Z. Bakenov, I. Taniguchi, Journal of the Electrochemical Society 157 (2010) A430.
[20] J.W. Lee, M.S. Park, B. Anass, J.H. Park, M.S. Paik, S.G. Doo, Electrochimica Acta
55 (2010) 4162.
Zn doped LiMnPO4 was synthesized by a new solid-state
reaction method, and we demonstrated that the electrochemical
performance of LiMnPO4 can be significantly improved by a small
amount of Zn doping due to the reduced charge transfer resis-
tance, the increased lithium ion diffusion and phase conversion.
The results are of great importance to the performance promotion
of LiMnPO4 and helpful to further understanding of the effect of
cation doping.
[21] C.L. Hu, H.H. Yi, H.S. Fang, B. Yang, Y.C. Yao, W.H. Ma, Y.N. Dai, Electrochemistry
Communications 12 (2010) 1784.
[22] Y.R. Wang, Y.F. Chen, S.Q. Cheng, L.N. He, Korean Journal of Chemical Engineer-
ing 28 (2011) 964.
Acknowledgments
This work was supported by the Natural Science Foundation
of Yunnan Province (no. 2009ZC001X) and analysis and testing
foundation of Kunming University of Science and Technology (no.
2010199).
[23] H. Liu, C. Li, H.P. Zhao, L.J. Fu, Y.P. Wu, H.Q. Wu, Journal of Power Sources 159
(2006) 717.
[24] D. Morgan, A. Van der Ven, G. Ceder, Electrochemical Solid-State Letters 7
(2004) A30.