Electrochemistry Communications 8 (2006) 1553–1557

www.elsevier.com/locate/elecom

Doping effects of zinc on LiFePO4 cathode material

for lithium ion batteries
H. Liu, Q. Cao, L.J. Fu, C. Li, Y.P. Wu *, H.Q. Wu
Department of Chemistry and Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, Fudan University, Shanghai 200433, China

Received 7 July 2006; accepted 11 July 2006

Available online 17 August 2006

Abstract

Alien atom doping has been adopted to modify the electrochemical performance of olivine type LiFePO4 for cathode material of the
lithium ion batteries. Here, we first report that zinc-doping can improve the performance of LiFePO4. The effects of zinc-doping have
been studied by the measurements of X-ray diffraction pattern, scanning electronic microscopy, electrochemical impedance spectroscopy
and cyclic voltammetry. The results indicate that the zinc atoms do not destroy the lattice structure of LiFePO4, and enlarge the lattice
volume. During de-intercalation and intercalation process of lithium ions, the doped zinc atoms protect the LiFePO4 crystal from shrink-
ing. This kind of ‘‘pillar’’ effect provides larger space for the movement of lithium ions. Consequently, the conductivity is enhanced and
the lithium ion diffusion coefficient is boosted after doping. These favourable changes are beneficial to the improvement of the electro-
chemical performance of LiFePO4 including discharge capacity and rate capability.
 2006 Elsevier B.V. All rights reserved.

Keywords: LiFePO4; Doping; Zn; Lithium ion battery

1. Introduction In the early period of research, its reversible capacity

reaches only 60% of the theoretic one, namely, the de-inter-
Since the first report that the olivine structure LiFePO4 calation of only 0.6 Li atom/LiFePO4. The limited interca-
could be used as cathode material for lithium ion batteries lation/de-intercalation dynamics is mainly due to its poor
in 1997 [1,2], it has been widely investigated as a promising conductivity [1,3]. Among previous attempts to enhance
cathode to substitute LiCoO2. In addition, it is attractive the electronic conductivity, carbon coating is the common
for its cheapness, rich source and environmental benig- way to overcome the limited rate capacity because the dis-
nancy [3]. Besides, it has high theoretic capacity as persed carbon provides pathways for electron transferring,
170 mAh/g [2], high charge/discharge potential (3.4 V vs. resulting in improvement of the conductivity and electro-
Li+/Li) [4] and excellent thermal stability [5,6]. chemical properties [7,9,18–22]. In addition, small metal
One natural form of LiFePO4 is triphylite, which is particles [12,23] and electronically conductive polymers
impure and contains other elements such as manganese [24] are also available as coating materials. Recently, it
atoms. There are a lot of methods to synthesize purified was found that doping alien cations such as Mn, Mg, Ti
LiFePO4, for instance, solid state reaction [7–9], hydrother- and Zr was another feasible way to enhance the intrinsic
mal method [10,11], sol–gel synthesis [12–14], and other unu- conductivity [25–28].
sual approaches such as spray solution technology [15], In this paper, we reported for the first time a Zn-doped
pulsed laser deposition [16], and template technology [17]. LiFePO4 material to form homogeneous LiZn0.01-
Fe0.99PO4, and the doping effects were investigated. Pri-
*
Corresponding author. Tel./fax: +86 21 55664223. mary results show that this is a promising method to
E-mail address: [email protected] (Y.P. Wu). improve the electrochemical performance of LiFePO4.

1388-2481/$ - see front matter  2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.elecom.2006.07.014
1554 H. Liu et al. / Electrochemistry Communications 8 (2006) 1553–1557

2. Experimental
a
Virginal and doped LiFePO4 samples were prepared by

Intensity (a.u.)
a solid state route. The starting material Li2CO3,
FeC2O4 Æ 2H2O, ZnO, NH4H2PO4 and 5 wt% acetylene
black were mixed in atomic ratio of Li: (Fe + Zn):
PO4 = 1:1:1, the mixtures were milled in the inert atmo-
sphere to avoid the transformation of Fe2+ to Fe3+ in a
planet mixer (QM-BP) for 24 h. After milling, the mixtures
were calcined in a tube furnace at 400 C for 10 h in an
inert atmosphere to decompose the oxalate and phosphate. 10 20 30 40 50 60 70 80
Then the precursors were sintered at 750 C for 15 h in an 2θ (Degree)
inert atmosphere.
The cathodes were prepared by mixing 90 wt% active b
material, 5 wt% acetylene black, and 5 wt% polyvinylidene
fluoride (PVDF) binder in n-methylpyrrolidone (NMP) sol-
vent to form a homogeneous slurry. Then the mixtures

Intensity (a.u.)
were coated on an aluminum foil and cut to pieces. After
drying under ambient condition, the pieces were further
dried under vacuum at 120 C for 12 h. Finally, coin-type
cells were assembled in a glove box, which used lithium foil
as the counter and reference electrode, Celgard 2400 as the
separator, and LIB315 (Guotai Huarong Chemical Plant)
as the electrolyte.
Powder samples were identified using a powder X-ray 30 35 40

diffractometer (D8, Bruker) with monochromatized Cu 2θ (degree)

Ka radiation. The surface morphology was observed with Fig. 1. (a) X-ray diffraction patterns of the virginal and LiZn0.01-
scanning electronic microscope (SEM, Philips XL 300). Fe0.99PO4. (b) The dilation of (a) in the range of 25–45.
Both electrochemical impedance spectroscopy (EIS, EG
& G5210) and cyclic voltammetry (CV, CHI400) were eval-
uated in a two-electrode cell at room temperature. The CV Table 1
curves were carried out at the range of 2.1–4.5 V. In EIS Calculated crystal parameters of LiFePO4 and LiZn0.01Fe0.99PO4
measurement, the excitation voltage applied to the cells Sample LiFePO4 LiZn0.01Fe0.99PO4
was 10 mV and the frequency range was between
Space group Pmnb Pmnb
100 kHz and 10 mHz. a (Å) 5.9606 6.0013
b (Å) 10.2043 10.3206
3. Results and discussion c (Å) 4.7192 4.7394
Volume of unit (Å3) 287.04 293.54
Fig. 1a shows the X-ray diffraction patterns of the vir-
ginal and LiZn0.01Fe0.99PO4. Both samples exhibit single
phase of LiFePO4 with an ordered olivine structure amount of Zn is below 5%, the total volume of the lattice
indexed to the orthorhombic Pmnb space group. The Zn always increases. However, when the doping amount is
doping does not destruct the lattice structure of LiFePO4 too high, perhaps the structure of the formed LiFePO4
because of the low doping concentration. On the other will be different, and unfavourable results will appear.
hand, the distortion of the lattice is negligible because The expansion in lattice crystal could provide more space
of the almost similar ion radius of Zn2+ (Goldschmidt for lithium intercalation/de-intercalation. As to the rea-
radius: 0.083 nm) and Fe2+. Fig. 1b shows the dilation son, it is still not accurate. Perhaps the introduced
of Fig. 1(a) in the range of 25–45. The intensity of the Zn2+ forms LiZnPO4 and present larger crystal lattice.
diffraction peaks are enhanced and become some kind By the way, some lithium was evaporated during the
of acute after doping, indicating larger crystal size. Fur- heat-treatment process [3], and zinc can also move to
thermore, the curve of the doped material exhibits a left- the left Li sites and acting as another doping effect. Dur-
ward shift in the axis of 2h compared with the virginal ing the de-intercalation process, the lattice would be pre-
one, indicating an influence on the lattice parameters. vented from shrinking by the doped Zn2+ due to the
The lattice parameters are calculated and summarized unchangeable radius of Zn2+. In the case of Fe2+, it will
in Table 1. The Zn doping increases the lattice parame- be changed to Fe3+ during de-intercalation, which is
ters of a, b and c, and enlarges the total volume of the smaller. As a result, the lattice will collapse and limit dif-
lattice for about 3 wt%. We found that when the doping fusion of lithium ions. In the doped LiZn0.01Fe0.99PO4,
 H. Liu et al. / Electrochemistry Communications 8 (2006) 1553–1557 1555

Zn2+ in the lattice will act as a pillar to prevent the col- 3000
lapse of crystal during cycling as doping of heteroatoms a virginal
doped
in LiMn2O4 [29,30]. 2500
The morphology of Zn-doped sample is shown in Fig. 2.
The SEM micrograph shows that the product particles are 2000

Zim(Ω )
spherical, their diameter is about hundreds of nanometers,
1500
and the acetylene black is homogeneously distributed in the
composite [9]. The coating of conductive acetylene black 1000
on the surface of LiFePO4 crystals can reduce the parti-
cle-to-particle contact resistance, significantly enhancing 500
the electrical conductivity of the composites [9,22]. It is
believed that the unique morphology and the small size 0
1000 2000 3000 4000 5000 6000
of the particles are favorable for the intercalation/de-inter-
Zre(Ω )
calation process.
Electrochemical impedance spectroscopy was measured 6000
in the fully discharged state. Fig. 3a shows typical Nyquist b virginal
plots of the doped and virginal LiFePO4 composite elec- 5000 doped
trodes. Both profiles exhibit a semicircle in the high fre-
quency region and a straight line in the low-frequency 4000
region. The numerical value of the diameter of the semicir-

Zre(Ω )
cle on the Zreal axis is approximately equal to the charge
3000
transfer resistance (Rct), therefore, it can be seen that there
is a marked decrease in Rct after doping. The straight line is
2000
attributed to the diffusion of the lithium ions into the bulk
of the electrode material or so-called Warburg diffusion.
1000
The lithium ion diffusion coefficient could be calculated
from the low frequency plots according to the following 0.5 1.0 1.5 2.0 2.5 3.0 3.5
equation [31]: -1/2
w
D ¼ R2 T 2 =2A2 n4 F 4 C 2 r2 ð1Þ
Fig. 3. (a) The Nyquist plots of untreated and doped LiFePO4 in the
where R is the gas constant, T is the absolute temperature, frequency range between 100 kHz and 10 mHz. (b) The relationship
A is the surface area of the cathode (0.125 cm2), n is the between Zre and x1/2 at low frequency.
number of electrons per molecule during oxidization, F is
the Faraday constant, C is the concentration of lithium Fig. 3b shows the relationship between Zre and square
ion (7.69 · 103 mol/cm3), r is the Warburg factor which root of frequency (x1/2) in the low frequency region. The
is relative with Zre. lithium ion diffusion coefficients of the virginal and the
doped LiFePO4 were calculated, and are 9.98 · 1014 and
Z re ¼ RD þ RL þ rx1=2 ð2Þ
1.58 · 1013 cm2/s, respectively. The lithium ion diffusion
coefficient was increased slightly by the doping due to the
pillar effect in the crystal lattice. In the case of carbon coat-
ing, it can markedly increase the lithium ion diffusion coef-
ficient, at least one order of magnitude, in comparison with
the LiFePO4 without carbon coating since the intercalation
and de-intercalation is also a redox process [9]. Here since
both samples are from carbon coating, the increase degree
is not so marked.
Fig. 4 shows the CV curves of the virginal and doped
LiFePO4. The CV of doped LiFePO4 shows more symmet-
rical and sharper shape of the anodic/cathodic peaks, and
the peak current of doped sample is as twice as the virginal
LiFePO4, which also indicating the enhancement of the
lithium ion diffusion. In the case of the LiZn0.01Fe0.99PO4,
the potential separation between anodic and cathodic
peaks is 0.590 V, whereas that of the virginal LiFePO4
was 0.922 V. The well-defined peaks and narrower peak
Fig. 2. SEM photograph of Zn doped LiFePO4 synthesized from the solid separation suggest that the reversibility of the electrode
state reaction. reaction is markedly ameliorated by zinc doping.
1556 H. Liu et al. / Electrochemistry Communications 8 (2006) 1553–1557

0.6
virginal
between capacity and discharge rate of the virginal and
doped the doped LiFePO4. It is completely consistent with the
0.4 results from EIS and CV. After doping the charge capacity
increases from 100 to 163 mAh/g, and the discharge capac-
0.2 ity from 88 to 133 mAh/g. Since both LiFePO4 products
Current (mA)

present good olivine structures as shown in Fig. 1, their

0.0
cycling behavior is, of course, very satisfactory. Due to
-0.2
the larger crystal unit and the increase of conductivity from
doping, the movement of lithium ions and the electronic
-0.4 transferring process in the doped LiFePO4 are faster. As
a result, its discharge capacity under higher rate is larger
-0.6 than that of the virginal LiFePO4.
2.0 2.5 3.0 3.5 4.0 4.5
Voltage (V)
4. Conclusion
Fig. 4. CV profiles of the virginal and the doped LiFePO4 at the scanning
rate 0.5 mV/s. Zinc-doped LiZn0.01Fe0.99PO4 was synthesized by solid
state route. The particles are spherical and the size is
The data from EIS and CV measurements show that around hundreds of nanometers. The Zn doping favours
better electrochemical performance will be achieved by the formation of crystal structure, expands the lattice vol-
the zinc doping due to the more stable lattice structure ume and provides more space for lithium ion intercala-
and enhanced conductivity. Fig. 5 shows (a) the charge tion/de-intercalation. In addition, the doping decreases
and discharge curves at the current density of 0.2 mA/ the charge transfer resistance, improves the reversibility
cm2 (0.1 C) in the first cycle and (b) the relationship of lithium intercalation and de-intercalation, and increases
the diffusion of lithium ions due to the pillar effect of the
doped Zn atoms. As a result, zinc doping is very promising
a for the improvement of the electrochemical performance of
4.0 LiFePO4. Both discharge capacity and rate capability are
greatly ameliorated.
3.5 If the crystal structure of LiFePO4 is not destroyed after
doping, other kinds of heteratoms with similar or larger
Voltage (V)

3.0 radius will act in the same way to improve the electrochem-
ical performance of LiFePO4, and further study is going
2.5 on.

2.0
Acknowledgement
1.5
0 20 40 60 80 100 120 140 160 Financial support from China Natural Science Founda-
Capacity (mAh/g) tion (Key Project No. 20333040) and Shanghai Science and
Technology Committee (04QMX1406) is greatly
b appreciated.
140

References
120
Capacity(mAh/g)

[1] A.K. Padhi, K.S. Nanjundaswamy, J.B. Goodenough, J. Electro-

100 chem. Soc. 144 (1997) 1188.
[2] A.K. Padhi, K.S. Nanjundaswamy, C. Masquelier, S. Okada, J.B.
80 Goodenough, J. Electrochem. Soc. 144 (1997) 1609.
[3] Y.P. Wu, X.B. Dai, J.Q. Ma, Y.J. Chen, Lithium Ion Batteries:
Practice and Application, Chemical Industry Press, Beijing, 2004.
60 [4] A. Yamada, S.C. Chung, K. Hinokuma, J. Electrochem. Soc. 148
(2001) A224.
40 [5] D.D. MacNeil, Z.H. Lu, Z.H. Chen, J.R. Dahn, J. Power Sources 108
(2002) 8.
C/10 C/5 C/2 1C [6] M. Takahashi, S. Tobishima, K. Takei, Y. Sakurai, Solid State Ionics
Discharge rate (C) 148 (2002) 283.
[7] H. Huang, S.C. Yin, L.F. Nazar, Electrochem. Solid State Lett. 4
Fig. 5. (a) Charge and discharge curves at the current density of 0.2 mA/ (2001) A170.
cm2 (0.1 C) in the first cycle. (b) The relationship between capacity and [8] M. Takahashi, S. Tobishima, K. Takei, Y. Sakurai, J. Power Sources
discharge rate of the virginal and the doped LiFePO4. 97–98 (2001) 508.
 H. Liu et al. / Electrochemistry Communications 8 (2006) 1553–1557 1557

[9] H. Liu, C. Li, H.P. Zhang, L.J. Fu, Y.P. Wu, H.Q. Wu, J. Power [21] A.D. Spong, G. Vitins, J.R. Owen, J. Electrochem. Soc. 152 (2005)
Sources, in press, doi:10.1016/j.jpowsour.2005.10.098. A2376.
[10] S.F. Yang, P.Y. Zavalij, M.S. Whittingham, Electrochem. Commun. [22] H. Liu, L.J. Fu, H.P. Zhang, J. Gao, C. Li, Y.P. Wu, H.Q. Wu,
3 (2001) 505. Electrochem. Solid State Lett., in press.
[11] S.F. Yang, Y.N. Song, P.Y. Zavalij, M.S. Whittingham, Electrochem. [23] K.S. Park, J.T. Son, H.T. Chung, S.J. Kim, C.H. Lee, K.T. Kang,
Commun. 4 (2002) 239. H.G. Kim, Solid State Commun. 129 (2004) 311.
[12] F. Croce, A.D. Epifanio, J. Hassoun, A. Deptula, T. Olczac, B. [24] G.X. Wang, L. Yang, Y. Chen, J.Z. Wang, Steve Bewlay, H.K. Liu,
Scrosati, Electrochem. Solid State Lett. 5 (2002) A47. Electrochim. Acta 50 (2005) 4649.
[13] J.S. Yang, J.J. Xu, Electrochem. Solid State Lett. 7 (2004) A515. [25] S.Y. Chung, J.T. Bloking, Y.M. Chiang, Nat. Mater. 1 (2002)
[14] L.J. Fu, H. Liu, Y.P. Wu, E. Rahm, R. Holze, H.Q. Wu, Prog. Mater. 123.
Sci. 50 (2005) 881. [26] D.X. Gouveia, V. Lemos, J.A.C. de Paiva, A.G. Souza Filho, J.
[15] S.L. Bewlay, K. Konstantinov, G.X. Wang, S.X. Dou, H.K. Liu, Mendes Filho, S.M. Lala, L.A. Montoro, J.M. Rosolen, Phys. Rev. B
Mater. Lett. 58 (2004) 1788. 72 (2005) 24105.
[16] S.W. Song, R.P. Reade, R. Kostecki, K.A. Striebel, J. Electrochem. [27] M. Abbate, S.M. Lala, L.A. Montoro, J.M. Rosolen, Electrochem.
Soc. 153 (2005) A12. Solid State Lett. 8 (2005) A288.
[17] C.R. Sides, F. Croce, V.Y. Young, C.R. Martin, B. Scrosati, [28] G.X. Wang, S. Bewlay, S.A. Needham, H.K. Liu, R.S. Liu, V.A.
Electrochem. Solid State Lett. 8 (2005) A484. Drozd, J.F. Lee, J.M. Chen, J. Electrochem. Soc. 153 (2006)
[18] N. Ravet, Y. Chouinard, J.F. Magnan, S. Besner, M. Gauthier, M. A25.
Armand, J. Power Sources 97–98 (2001) 503. [29] J. Kawakita, K. Makino, Y. Katayama, T. Miura, T. Kishi, J. Power
[19] S.T. Myung, S. Komaba, N. Hirosaki, H. Yashiro, N. Kumagai, Sources 75 (1998) 244.
Electrochim. Acta 49 (2004) 4213. [30] Y.P. Wu, E. Rahm, R. Holze, Electrochim. Acta 47 (2002) 3491.
[20] K. Zaghib, J. Shim, A. Guerfi, P. Charest, K.A. Striebel, Electro- [31] A.J. Bard, L.R. Faulkner, Electrochemical Methods, second ed.,
chem. Solid State Lett. 8 (2005) A207. Wiley, 2001, p. 231.