Journal of Alloys and Compounds 509 (2011) 1897–1900

Contents lists available at ScienceDirect

Journal of Alloys and Compounds

journal homepage: www.elsevier.com/locate/jallcom

Effect of Mn2+ -doping in LiFePO4 and the low temperature electrochemical

performances
Chengfeng Li a,b,c , Ning Hua a,b,c , Chengyun Wang a,b,c , Xueya Kang a,c,∗ ,
Tuerdi Wumair a,c , Ying Han a,c
a
Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi Xinjiang 830011, China
b
Graduate University of Chinese Academy of Sciences, Beijing 100049, China
c
Xinjiang Key Laboratory of Electronic Information Materials and Devices, Urumqi Xinjiang 830011, China

a r t i c l e i n f o a b s t r a c t

Article history: Olivine composites LiFe1−x Mnx PO4 /C (x = 0, 0.05, 0.1) were prepared by chelation assisted mechanochem-
Received 21 June 2010 ical activation method using C2 H2 O4 as the chelating reagent. The structures of the prepared composites
Received in revised form 20 October 2010 were characterized by X-ray diffraction (XRD). Cyclic voltammetry, ac impedance spectroscopy, gal-
Accepted 21 October 2010
vanostatic charge/discharge performances were studied at room temperature. The discharge capacity of
Available online 29 October 2010
LiFe0.95 Mn0.05 PO4 was 155.6 mAh g−1 and 102.9 mAh g−1 at 1 C and 10 C rate, respectively. The results indi-
cated Mn2+ -doping in LiFePO4 could effectively enhance the electrochemical performances of this olivine
Keywords:
compound especially at high charge/discharge rate. In addition, charge/discharge method was also used
Olivine compound
Electrochemical performance
to characterize the low temperature properties of the different samples. The results showed that the
Low temperature performance lower the operation temperature was, the poorer the performances were. Furthermore, ac impedance
Chelation assisted method measurement was used to understand the factors of the poor performances at low temperature.
Mn2+ -doping © 2010 Elsevier B.V. All rights reserved.

1. Introduction ductivity [14]. In the olivine family, LiMnPO4 was also an attractive
cathode material for its high potential of 4.1 V vs. Li+ /Li [15]. But
Since the work of Goodenough and co-workers [1,2], olivine its poor electrochemical performances limited its practical usage.
structured LiFePO4 has been investigated intensively as cathode However, Mn2+ -doping in LiFe1−x Mnx PO4 solid solution could sig-
materials for lithium batteries because of their high stabilities, envi- nificant improve the kinetic properties of LiFePO4 in the region of
ronmental benign and high theoretical capacities (170 mAh g−1 ) 0 ≤ x ≤ 0.75 [16–18]. So it might be an effective way to enhance
[3–5]. However, their intrinsic low electronic conductivity [3] and the electrochemical performances of LiFePO4 . In this work, we
poor ionic transport property [6] have largely hindered the devel- have successfully prepared Mn2+ doped LiFe1−x Mnx PO4 composites
opment of this cathode material. For these reasons, few works by chelation assisted mechanochemical activation method using
about these materials had been reported till the landmark work C2 H2 O4 as the chelating reagent and using water as the media
finished by Chung et al. [3], whom enhanced the electronic con- of ball-mill procedure. These Mn2+ -doping composites showed an
ductivity with a factor of ∼108 by aliovalent doping. Till now, enhancement of high rate capacities and cycling stabilities.
based on the numerous work about these olivine compounds, Further considering the practical usage of LiFePO4 , a very impor-
several concepts have been concluded to increase their electro- tant aspect was the low temperature performances. Several reports
chemical performances [7,8], such as (i) decreasing the particle had already been published about the low temperature perfor-
size to lessen the diffusion distance and increase the surface area mances of this material [19,21]. We also attempted to enhance
by advanced synthesis techniques [9], (ii) increasing the electronic the low temperature performances by Mn2+ -doping. But the results
conductivity by surface coating with conductive carbon or other indicated that there was little improvement which indicated other
electronic conducting metal or oxide [10–13], (iii) selective doping method should be used to solve this problem. Ac impedance
with supervalent cations to increase the intrinsic electronic con- spectroscopy was also used to investigate the low temperature
properties of Mn2+ -doping samples. From the results we could get
that the Rct (correspond to charge transfer resistance) increased
largely while decreasing the operation temperature, which indi-
∗ Corresponding author at: Xinjiang Technical Institute of Physics and Chemistry,
cated electrolyte optimization and surface modification might be
Chinese Academy of Sciences, Urumqi Xinjiang 830011, China.
Tel.: +86 991 3850517; fax: +86 991 3850517. the most effective way to enhance the low temperature perfor-
E-mail address: xueyakang@yahoo.cn (X. Kang). mances.

0925-8388/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2010.10.083
1898 C. Li et al. / Journal of Alloys and Compounds 509 (2011) 1897–1900

LiFe 0.9 Mn 0.1 PO 4

Intensity/a.u.

LiFe 0.95 Mn 0.05 PO 4

LiFePO4

10 20 30 40 50 60 70 80
2θ

Fig. 1. XRD patterns of the synthesized LiFe1−x Mnx PO4 (x = 0, 0.05, 0.1).

2. Experimental

2.1. Sample preparation

Fig. 3. (a) Nyquist plots of the three composites. (b) Equivalent circuit model used
LiFe1−x Mnx PO4 composites were prepared by chelation assisted mechanochem- for EIS fitting of the electrode.
ical activation method using C2 H2 O4 as the chelating reagent. Stoichiometric
amount of NH4 H2 PO4 , Mn(CH3 COO)2 ·4H2 O, FePO4 ·2H2 O and LiOH·H2 O were ball
milled with 15 wt.% C2 H2 O4 for 7 h using deionised-water as the media of ball-
(Fig. 1). All the peaks could be indexed as orthorhombic structure
milling. Polyurethane vessel and ZrO2 balls were used during ball-milling procedure
(QM-3SP4, Nanjing NanDa Instrument Plant) with a ball-to-powder ratio (weight
belonging to the space group Pnma (JCPDS No. 40-1499). The absence
ratio) of 4:1 and a rotation speed of 350 rpm. The milled powder was dried at 70 ◦ C of any other signals indicated there were no high content impu-
in a dry oven. After the powder was dried, 15 wt.% glucose was added while grinding rities related to Mn2+ or Fe3+ compounds and the residual carbon
the powder in mortar. Then the powder was sintered in a tube furnace at 700 ◦ C for decomposed from glucose was amorphous in these composites. The
10 h using nitrogen atmosphere to avoid the oxidation of Fe2+ .
results illustrated the feasibility of our method for the preparation
of LiFe1−x Mnx PO4 .
2.2. Sample characterization
To evaluate the effect of Mn2+ substitution, cyclic voltamme-
The phase structure of the obtained samples were investigated by X-ray diffrac- try was used at room temperature. Fig. 2 was the profiles of the
tion (XRD, RINT-2500 V, Rigaku Co.) using Cu K␣ radiation ( = 1.5418 Å) in the range prepared samples. From Fig. 2(a) a more symmetric curve with
of 5–80◦ with a scanning rate of 2◦ per min. To test the electrochemical perfor- smaller interval between the cathodic and anodic peak current was
mances, the cathode was prepared by spreading the cathode slurry (80 wt.% of the
active material) onto an aluminum foil followed by drying in vacuum at 120 ◦ C for
obtained by Mn2+ substitution. This denoted a decreasing of polar-
12 h. The cells (CR2025) were assembled in an argon filled glove-box using lithium ization with Mn2+ substitution, which illustrated an enhancement
metal foil as the counter electrode. The electrolyte was 1.0 mol dm−3 LiPF6 in a mix- of electrochemical performances. Fig. 2(b) showed the CV curves
ture of ethyl carbonate (EC), diethyl carbonate (DEC) and dimethyl carbonate (DMC) of LiFe0.95 Mn0.05 PO4 at the sweeping rate of 0.1 mV s−1 , 0.2 mV s−1 ,
(volume ratio 1:1:1). Electrochemical performances of LiFePO4 were investigated by
0.5 mV s−1 and 1 mV s−1 , respectively. It could be seen in Fig. 2(b),
using CR2025 coin-type cell. The cells were charged and discharged between 2.3 V
and 4.3 V on a charge/discharge apparatus (BTS-51, Neware, China). Cyclic voltam- as the sweeping rate was raised the oxidation peak shifted higher
metry (CV) and electrochemical impedance spectroscopy (EIS) were conducted by and the reduction peak sifted lower. And the reduction peaks in
using a CHI650 electrochemical working station. The sinusoidal excitation voltage Fig. 2(a) and (b) seemed more vulnerable to the variety of Mn2+
of EIS was 10 mV and the frequency range was between 105 and 10−2 Hz. content and sweeping rate which was coincidence with the work
of Nakamura et al. [16].
3. Results and discussion The electrode impedance of these prepared composites was also
evaluated at room temperature. Fig. 3(a) presented the Nyquist
All the samples were successfully synthesized by this chela- plots of the composites at the same charge/discharge state. The
tion assisted method which could be verified by the XRD patterns impedance spectra were fitted with the fitting equivalent circuit as

-1
0.5 mV s
a LiFe0.95Mn0.05PO4 10 b
4
LiFe0.9Mn0.1PO4 LiFePO4 0.2 mV s
-1
-1
5 1 mV s
2
Current/mA

Current/mA

0 0

-1
-2 0.1 mV s
-5

-4
-10
2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2
Potential/V Potential/V

Fig. 2. (a) CV curves of the composites at 0.1 mV s−1 . (b) CV profiles of LiFe0.95 Mn0.05 PO4 at different sweeping rate.
 C. Li et al. / Journal of Alloys and Compounds 509 (2011) 1897–1900 1899

Table 1 LiFePO4
Impedance parameters of LiFe1−x Mnx PO4 fitted with equivalent circuit.
LiFe0.95Mn0.05PO4
Re / Rf / Rct / 160
LiFePO4 5.283 7.184 21.11 140 LiFe0.9Mn0.1PO4
LiFe0.95 Mn0.05 PO4 3.273 8.484 34.55

Capacity/mAh g-1
LiFe0.9 Mn0.1 PO4 4.06 11.7 54.64 120

100
shown in Fig. 3(b). According to the literature [21], Re represents 80
the solution resistance, Rf signify the diffusion resistance of Li-ions
through the solid electrolyte interface (SEI) layer, Rct correspond 60
to the charge transfer resistance. Zw is related to the solid-state 40
diffusion of Li-ions in the active materials corresponding to the
slopping line at the low frequency. The fitting results were shown 20
in Table 1. It was obviously that Rct was increasing with the rising 0
of the substitution content. This phenomenon illustrated that even -40 -30 -20 -10 0 10 20 30
if the enhanced electrochemical performances could be obtained Temperature/ºC
by Mn2+ substitution, the surface charge transfer resistance repre-
Fig. 6. Capacities of the composites at different temperatures.
sented by Rct still increased. It is well known that LiMnPO4 has very
poor electrochemical performances when compared with LiFePO4 ,
duo to its insulation property [15]. The increased Rct might be the three composites was slightly increasing, which could be seen
ascribed to the existence of Mn2+ substituted olivine structure on in many reports using CTR method [20]. While the raise of the
the surface of the crystals, which hindered the charge transfer reac- C rate, the advantage of Mn2+ substitution became notable. Even
tion on the surface. the capacity of LiFe0.9 Mn0.1 PO4 exceeded the non-substituted one
The charge/discharge curves of LiFe0.95 Mn0.05 PO4 at room tem- at 10 C and stable at this capacity. So the electrochemical perfor-
perature were exhibited in Fig. 4. As shown in Fig. 4, the discharge mances (especially the high rate performances) of olivine LiFePO4
capacities of this composite were 160.8 mAh g−1 , 155.6 mAh g−1 , could be effectively enhanced by Mn2+ substitution, duo to the
146.6 mAh g−1 , 129.6 mAh g−1 and 102.9 mAh g−1 at 0.5 C, 1 C, 2 C, improved electronic conductivity and ionic transportation [16].
5 C, 10 C rate, respectively. Fig. 5 showed the cycling profiles of the Further considering the practical usage of LiFe1−x Mnx PO4 , we
different composites at room temperature. At first, the capacity of investigated the low temperature performances of the prepared
composites by galvanostatic charge/discharge method. The rela-
4500
tionship of capacity to temperature was displayed in Fig. 6.
While lowering the temperature, the capacities of the composites
decreased largely. This phenomenon indicated a poor electrochem-
4000 ical performance at low temperature, which should be improved
for the practical, large-scale usage of these composites. From Fig. 6,
the capacity decreased linearly when lowering the temperature and
Voltage/mV

3500
the slopes were nearly the same, which were irrespective to such
different composites. It was interesting that the kinetic enhanced
3000 properties almost had no impact on the low temperature perfor-
mances. To answer this puzzle, ac impedance spectroscopy was
2500 used to investigate the impedance property of LiFe0.95 Mn0.05 PO4 at
low temperature. The Nyquist plots at different temperatures were
0.
10

presented in Fig. 7. It was obviously that the Rct increased largely

2
5

5
C
C
C

C
C

2000 as the temperature decreased, especially when the temperature fell

0 20 40 60 80 100 120 140 160
Capacity/mAh g-1 below −20 ◦ C. The spectra were also fitted with the equivalent cir-
cuit as shown in Fig. 3(b). The impedance parameters were listed
Fig. 4. Charge/discharge curves of LiFe0.95 Mn0.05 PO4 at different C rates. in Table 2. From the table, the decreasing of temperature showed
much pronounced influence on Rct which represented charge trans-
0.5 C 1C
160 1C
2C 25 ºC
140 5C 0 ºC
12000 -10 ºC
120 -20 ºC
10 C 10000
Capacity/mAh g-1

-40 ºC
100
8000
-Z''/ohm

80
LiFePO4 6000
60

40
LiFe0.95Mn0.05PO4 4000

20 LiFe0.9Mn0.1PO4 2000

0 0
0 5 10 15 20 25 30 0 10000 20000 30000
Cycle number Z'/ohm

Fig. 5. Cycling profiles of the composites. Fig. 7. Nyquist plots of LiFe0.95 Mn0.05 PO4 at different temperature.
1900 C. Li et al. / Journal of Alloys and Compounds 509 (2011) 1897–1900

Table 2 charge/discharge rate. In addition, galvanostatic charge/discharge

Impedance parameters of LiFe0.95 Mn0.05 PO4 at different temperatures.
method was also used to characterize the low temperature proper-
Temperature Re / Rf / Rct / ties of these composites. The results showed that the lower the
25 C ◦
6.321 18.18 69.55 operation temperature was, the poorer the performances were.
0 ◦C 9.212 68.84 376.6 Furthermore, ac impedance measurements were used to under-
−10 ◦ C 7.967 115.6 921.5 stand the factors of the poor performances at low temperature.
−20 ◦ C 8.676 286.5 2233 From the results, the decreasing of temperature showed much
−40 ◦ C 10.94 2481 1.959 × 104
pronounced influence on Rct which represented charge transfer
reaction occurring at the electrode/electrolyte interface. This indi-
cated that the most effective way to enhance the low temperature
Re
performances might be decreasing the Rct by facilitating the charge
20000
Rf transfer reaction which could be solved by surface modification or
electrolyte optimization.
R ct
15000
References
R/ohm

10000 [1] A.K. Padhi, K.S. Nanjundaswamy, J.B. Goodenough, J. Electrochem. Soc. 144
(1997) 1188–1194.
[2] A.K. Padhi, K.S. Nanjundaswamy, C. Masquelier, S. Okada, J.B. Goodenough, J.
5000 Electrochem. Soc. 144 (1997) 1609–1613.
[3] S.Y. Chung, J.T. Bloking, Y.M. Chiang, Nat. Mater. 1 (2002) 123–128.
[4] C. Delacourt, P. Poizot, J.M. Tarascon, C. Masquelier, Nat. Mater. 4 (2005)
254–260.
0 [5] M. Thackeray, Nat. Mater. 1 (2002) 81–82.
-50 -40 -30 -20 -10 0 10 20 30 [6] B. Zhang, X.J. Wang, Z.J. Liu, H. Li, X.J. Huang, J. Electrochem. Soc. 157 (2010)
Temperature/ºC A285–A288.
[7] D. Zhang, X. Yu, Y.F. Wang, R. Cai, Z.P. Shao, X.Z. Liao, Z.F. Ma, J. Electrochem.
Fig. 8. Comparison for temperature dependency of the different resistances. Soc. 156 (2009) A802–A808.
[8] F. Yu, J.J. Zhang, Y.F. Yang, G.Z. Song, J. Mater. Chem. 19 (2009) 9121–9125.
[9] L. Wang, Y. Huang, R. Jiang, D. Jia, Electrochim. Acta 52 (2007) 6778–6783.
[10] Y. Wang, Y. Wang, E. Hosono, K. Wang, H. Zhou, Angew. Chem. Int. Ed. 47 (2008)
fer reaction occurring at the electrode/electrolyte interface (as 7461–7465.
shown in Fig. 8). These results indicated that Mn2+ substitution [11] S. Yang, Y. Song, P.Y. Zavalij, M. Stanley Whittingham, Electrochem. Commun.
4 (2002) 239–244.
had few influences on the low temperature performances. The most
[12] S.L. Bewlay, K. Konstantinov, G.X. Wang, S.X. Dou, H.K. Liu, Mater. Lett. 58 (2004)
effective way to enhance the low temperature performances might 1788–1791.
be decreasing the Rct by facilitating the charge transfer reaction, [13] F. Croce, A.D. Epifanio, J. Hassoun, A. Deptula, T. Olczac, B. Scrosati, Electrochem.
which could be solved by surface modification or electrolyte opti- Solid State 5 (2002) A47–A50.
[14] L.J. Li, X.H. Li, Z.X. Wang, L. Wu, J.C. Zheng, H.J. Guo, J. Phys. Chem. Solids 70
mization. (2009) 238–242.
[15] S.K. Martha, B. Markovsky, J. Grinblat, Y. Gofer, O. Haik, E. Zinigrad, D. Aurbach,
4. Conclusion T. Drezen, D. Wang, G. Deghenghi, I. Exnar, J. Electrochem. Soc. 156 (2009)
A541–A552.
[16] T. Nakamura, K. Sakumoto, M. Okamoto, S. Seki, Y. Kobayashi, T. Takeuchi, M.
In this work, we introduced a chelation assisted mechanochem- Tabuchi, Y. Yamada, J. Power Sources 174 (2007) 435–441.
ical activation method with C2 H2 O4 as the chelating reagent to [17] A. Yamada, H. Koizumi, S.I. Nishimura, N. Sonoyama, R. Kanno, M. Yonemura,
T. Nakamura, Y. Kobayashi, Nat. Mater. 5 (2006) 357–360.
prepare LiFe1−x Mnx PO4 composites. The structures of the compos- [18] A. Yamada, Y. Kudo, K.Y. Liu, J. Electrochem. Soc. 148 (2001) A747–A754.
ites were characterized by X-ray diffraction. The electrochemical [19] X.Z. Liao, Z.F. Ma, Q. Gong, Y.S. He, L. Pei, L.J. Zeng, Electrochem. Commun. 10
properties at room temperature were investigated by galvanos- (2008) 691–694.
[20] H.P. Liu, Z.X. Wang, X.H. Li, H.J. Guo, W.J. Peng, Y.H. Zhang, Q.Y. Hu, J. Power
tatic charge/discharge method, cyclic voltammetry, ac impedance Sources 184 (2008) 469–472.
spectroscope. The results showed that the electrochemical perfor- [21] S.S. Zhang, K. Xu, T.R. Jow, J. Power Sources 159 (2006) 702–707.
mances were enhanced by Mn2+ substitution, especially at high