Effects of magnesium and fluorine co-doping on the structural and electrochemical performance of the spinel LiMn2O4 cathode materials

Accepted Manuscript Title: Effects of magnesium and fluorine co-doping on the structural and electrochemical performance of the spinel LiMn2 O4 cathode materials Author: Weicheng Wen Bowei Ju Xianyou Wang Chun Wu Hongbo Shu Xiukang Yang PII: DOI: Reference:

S0013-4686(14)01959-8 http://dx.doi.org/doi:10.1016/j.electacta.2014.09.115 EA 23467

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

24-7-2014 21-9-2014 24-9-2014

Please cite this article as: Weicheng Wen, Bowei Ju, Xianyou Wang, Chun Wu, Hongbo Shu, Xiukang Yang, Effects of magnesium and fluorine co-doping on the structural and electrochemical performance of the spinel LiMn2O4 cathode materials, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2014.09.115 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effects of magnesium and fluorine co-doping on the structural and electrochemical performance of the spinel LiMn2O4 cathode materials Weicheng Wen, Bowei Ju, Xianyou Wang, Chun Wu, Hongbo Shu, Xiukang Yang (Key Laboratory of Environmentally Friendly Chemistry and Applications of Ministry of Education，Hunan Province Key Laboratory of Electrochemical Energy Storage and Conversion， School of Chemistry, Xiangtan University, Hunan, Xiangtan 411105, China)

PT

Highlights

The Mg-F co-doping LiMg0.1Mn1.9O3.8F0.2 is firstly synthesized.

2.

LiMg0.1Mn1.9O3.8F0.2 shows the notable cycling performance at 55 ºC.

3.

The effects of Mg-F co-doping on the structure of LiMn2O4 are investigated.

U

Abstract

SC

RI

1.

N

The spinel LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples are synthesized via a

M

A

solid-state reaction route with the MgF2 as the dopant. The structure and performance of the samples are characterized by powder X-ray diffraction, atomic absorption

TE

D

spectroscopy, X-ray photoelectron spectroscopy, scanning electron microscopy, galvanostatic charge-discharge and electrochemical impedance spectroscopy. The

EP

results reveal that the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples possess the

CC

typical octahedral structure with single phase. Especially, LiMg0.1Mn1.9O3.8F0.2 shows

A

the highest initial discharge capacity of 121.1 mAh g-1, and the capacity retention is still as high as 89.2% even after 400 cycles at a rate of 1 C under an elevated temperature of 55 ºC. Besides, it presents an excellent rate capability, a high discharge capacity of over 76.1 mAh g-1 is still obtained even at the high rate of 20 C. 

Corresponding author: Tel: +86-731-58292060 Fax: +86-731-58292061 E-mail address: [email protected] (X. Wang)

Furthermore, it has been confirmed that the Mg and F co-substitution can fundamentally ameliorate the Jahn-teller distortion of the conventional LiMn2O4, thus this is a very effective way for improving the elevated temperature capacity retention of LiMn2O4. Keywords: Lithium-ion battery; Spherical spinel lithium manganese oxide; Mg-F co-substitution; Capacity retention

PT

1. Introduction

RI

Lithium-ion battery has been considered as the most promising alternative for

SC

environmentally friendly power source because of high energy density, low

U

maintenance, and economical and environmental advantages [1-3]. It is well known

N

that the improvement of the positive electrode material plays a critical role in the

A

performance enhancement of lithium-ion battery. Among numerous positive materials,

M

LiMn2O4 has been spotlighted much in respect to their comparatively low cost,

D

natural abundance, high energy density and environmental friendliness. However,

TE

LiMn2O4 electrodes in the 4 V (versus Li/Li+) region suffer from severe capacity

EP

fading at elevated temperature mainly due to the structural instability caused by the

CC

Jahn-teller distortion and the Mn dissolution in electrolyte [4-7]. To overcome these problems, the surface modification technologies of LiMn2O4

A

with a thin coating layer, which can inhibit the manganese dissolution, have been recently proposed [8-12]. Unfortunately, these coating layers can’t fundamentally ameliorate the Jahn-teller distortion and the materials still suffer from the corrosion of the electrolyte, which means that the long term cyclability of LiMn2O4 can’t be

significantly enhanced. Therefore, more recently a lot of studies have focused on substituting manganese ions with alien cations (Al, Co, Ni etc) to resist the Jahn-teller distortion by stabilizing the MnO6 octahedron structure [13-15]. Xiao et al. [16] reported that the cycling performance of LiMn2O4 was improved by Al doping on the Mn sites (16 d octahedral) in the LiMn2 O4 framework. Particularly, the LiAl0.08Mn1.92O4 exhibited a good cyclic performance with the capacity retention of

PT

99.3% after 50 cycles at 1 C under 55 °C. Another work reported by Sigala et al. [17]

RI

described the improvement of the cycle behavior of the chromium substituted spinel

SC

LiCr0.25Mn1.75O4. And the capacity retention of the substituted LiMn2O4 was over

U

90.0% after 50 cycles between 3.0 and 4.4 V at 1 C under room temperature.

N

Although such substitutions often result in an enhanced stability of the spinel

A

LiMn2O4, the discharge capacity is found to be considerably lower than that of the

M

parent spinel. As well known, the discharge capacity of LiMn2O4 depends on its

D

content of the electrochemically active Mn3+. Hence, many researchers have

TE

attempted to introduce monovalent anions to the bulk of LiMn2O4 modified by cations,

EP

which can reduce the Mn oxidation state, resulting in the increase of the specific

CC

capacity. Kang et al. [18] studied the effects of aluminum and fluorine two-fold substitution on the electrochemical performances of LiMn2O4. They found that the

A

Li1.05 Al0.1Mn1.85O3.9F0.1 sample showed a relatively high initial discharge capacity of 110.0 mAh g-1 and the capacity retentions reached 99.0% under the current density of 20 mA g-1 at 55 °C after 50 cycles. Therefore, cation or/and anion co-doping is an effective way to simultaneously improve the cycling stability and the discharge

capacity of LiMn2O4. Up to the present, Mg doping on LiMn2O4 is found to be significantly effective in stabilizing the spinel structure as reported by many authors [19,20]. Nevertheless, the Mg doped LiMn2O4 samples still suffer from the low discharge capacity. Although many investigations have been done in order to enhance the electrochemical performance of LiMn2O4, there is a lack of knowledge about the effects of Mg-F co-doping on the structural and electrochemical performance of the

PT

spinel LiMn2 O4.

RI

In this work, F and Mg are simultaneously introduced into the framework of

SC

LiMn2O4 with the molar ratio 2:1. The effects of Mg-F co-substitution on the

U

structural and electrochemical properties of the stoichiometric spinel LiMn2 O4 are

A

N

investigated in detail.

M

2. Experimental

D

2.1 Preparation of the spherical precursor MnCO3

TE

The precursor MnCO3 was synthesized by the co-precipitation method. An

EP

aqueous solution of MnSO4 with the concentration of 1.6 mol L-1 was pumped into a

CC

continuously stirred tank reactor at the feeding rate of 1.0 mL min-1. At the same time, the 1.6 mol L-1 of Na2CO3 solution (aq) and the desired amount of NH4OH solution

A

(aq) as the chelating agent were also separately fed into the reactor at the feeding rate of 1.03 mL min-1 and 0.5 mL min-1, respectively. The concentration of the solution, the pH (about 7.8), the operation temperature (60 °C) and the stirring rate (800 rpm) in the reactor were carefully controlled. At last, after 6 h of stirring, the spherical

MnCO3 precursor was obtained, and then filtered, washed and dried at 80 °C. 2.2 Preparation of the spherical spinel LiMgxMn2-xO4-2xF2x (x=0, 0.05, 0.1, 0.2) The as-prepared MnCO3 precursor was heated at 560 °C at a heating rate of 2 °C min-1 for 6 h to decompose the carbonate into the spherical Mn3O4 oxide. According to our previous work [21], the spherical Mn3O4 oxide was mixed with the desired amount of Li2CO3, and then the mixture was sintered at 750 °C for 20 h to obtain the

PT

spherical spinel LiMn2O4. At the same time, the spherical Mn3O4 oxide was mixed

RI

with Li2 CO3 and MgF2 based on the desired mole ratio of Li/Mg/Mn. Then the

SC

obtained mixtures were further calcined to obtain LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1,

U

0.2) at 750 °C for 20 h, respectively.

N

2.3 Physical characterizations

A

To analysis the high temperature synthesis process, thermal behavior was studied

M

by thermogravimetric-differential thermal analysis (TG-DTA, Model TAS-200,

D

Rigaku, Tokyo, Japan). The phase identification of the samples was performed with a

TE

diffractometer (D/Max-3C, Rigaku, Japan) using Cu Kα radiation (λ = 0.154178 nm)

EP

and a graphite monochromator at 40 kV, 40 mA. Structural refinement was carried out

CC

by Rietveld analysis using Maud program. The X-ray photoelectron spectroscopy (XPS) measurements were carried on a K-Alpha 1063 spectrometer with Al Kα X-ray

A

radiation and hemispherical electron energy analyzer. The morphology of the sample was observed using scanning electron microscopy (SEM, JEOL JSM-5600LV). The dissolved amounts of transition metal Mn and the total average chemical composition of the LiMgxMn2-xO4-2xF2x were determined by atomic absorption spectroscopy (Vario

6 Analytik Jena AG, Jena, Germany). 2.4 Electrochemical characterizations The electrochemical tests were examined using CR2025 coin-type cells. In all cells, the cathode was consisted of a mixture of active material (80 wt%), acetylene black (10 wt%), graphite (5 wt%) and polyvinylidene fluoride (5 wt%) as the binder agent, lithium metal was used as the counter and reference electrodes, a Celgard 2400

PT

was used as the separator, and the electrolyte was a 1 mol L-1 LiPF6 solution in

RI

ethylene carbonate (EC)-dimethyl carbonate (DMC) (1:1, v/v). The cells were tested

SC

at the rate of 1 C between 3.0 and 4.4 V on a Neware battery test system

U

BTS-XWJ-6.44S-00052 (Newell, Shenzhen, China). In addition, the cells were first

N

charged at the same current density of 0.5 C, and then discharged at 1 C, 2 C, 5 C, 10

A

C, and 20 C between 3.0 V and 4.4 V. The electrochemical impedance spectroscopys

M

(EIS) of the cells were measured on a VersaSTAT3 electrochemical workstation

TE

EP

of 5 mV.

D

(Princeton, America) in the frequency range of 10 kHz to 10 mHz with an AC voltage

CC

3. Results and discussion

A

3.1 Morphology and Structure analysis The SEM images of LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples are shown

in Fig. 1. It can be seen from Fig. 1 that all the samples are made up of micron-sized spherical secondary particles in the range of 10-15 µm formed by the primary particles aggregating together, whose diameters are approximately 200 nm.

Furthermore, with the x increasing from 0 to 0.2, the edges become more and more distinct, and the shapes of the primary particles vary from irregular shape to cubic octahedral crystallite. A similar morphological change has also been reported by Kim and colleagues [22]. The reason may be that the halide salts are the effective mineralizing agents, which are often used as a flux in crystal growth. Therefore, this suggests that F ions are probably successfully substituted into the host structure. In

PT

addition, the actual Li/Mn/Mg ratio of the materials measured by atomic absorption

RI

spectroscopy is given in Table 1. It is confirmed that the chemical compositions of the

U

determined by atomic absorption spectroscopy.

SC

as-prepared powder are stoichiometric, except the anion content which could not be

N

In order to further understand the influences of Mg-F co-doping on

A

crystallization and morphology of LiMn2O4, the TG/DTA was carried out. Fig. 2a and

M

b depict the TG/DTA curves for pristine LiMn2O4 and Mg-F co-doped LiMn2O4. The

D

DTA curve for each sample shows both endothermic and exothermic peak at different

TE

temperatures. The mass losing in TG curves and the endothermic peaks in DTA curves

EP

observed in the range of 30-100 °C for two samples was due to the release of

CC

adsorbed water. Similarly, the big mass losing in TG curves and distinct endothermic peaks in DTA curves are observed in the range of 500-620 °C, which is contributed to

A

the decomposition of Li2CO3. Li2CO3 = Li2O + CO2

(1)

Besides, as seen in Fig. 2a, an exothermic peak is observed at about 708 °C, which is contributed to the formation of LiMn2O4. However, the exothermic peak for

the Mg-F co-doped LiMn2O4 is observed at about 687 °C. Compared with the LiMn2O4, the endothermic and exothermic peaks for the Mg-F co-doped LiMn2O4 are a slightly low to some extent, indicating the preparation of Mg-F co-doped LiMn2O4 is easier than LiMn2O4. Simultaneously, due to the mineralizing action of MgF2, which acts as an effective flux in crystal growth, the Mg-F co-doped LiMn2O4 will show much higher crystallization and much better morphology.

PT

Fig. 3 shows the XRD patterns of the LiMgxMn2-xO4-2xF2x (x=0, 0.05, 0.1, 0.2)

RI

samples. As being seen, all samples display the single crystalline cubic spinel phase

SC

with a space group Fd3m, which is implied that the co-substituted Mg and F ions not

U

only change the crystal structure of the spinel but also probably enter into the lattice

N

rather than forming impurities [23]. Moreover, the diffraction peaks are extremely

A

narrow, indicating high crystallinity. The XRD patterns of the pristine LiMn2O4 and

M

LiMg0.1Mn1.9O3.8F0.2 are further analyzed by Rietveld refinements, and the results are

D

shown in Fig. 4a and 4b. The reliability factor RWp of the refinement is around 8%,

TE

demonstrating that the results are reliable. The corresponding structural parameters

EP

obtained from the refinement for both samples are listed in Table 2. It can be observed

CC

that the lattice parameters and the unit cell volume are increased as incorporation of the Mg and F ions, which is caused by the much bigger ionic radius of Mg2+ (0.072

A

nm) compared to that of Mn3+ (0.064 nm) or Mn4+ (0.054 nm) despite the slightly smaller radius of F- (0.133 nm) than that of O2- (0.140 nm) [22]. Besides, the refinement results further demonstrate that the atoms are located in the following sites: Li atoms in 8a sites, Li, Mg, Mn atoms in 16d sites, and O, F atoms in 32e sites.

To identify the valence states of Mg, F and Mn in the as-prepared LiMg0.1Mn1.9O3.8F0.2, the XPS spectra of LiMg0.1Mn1.9O3.8F0.2 is shown in Fig. 5. The XPS peak at the binding energy of 1303.4 eV in Fig. 5b is attributed to Mg 1s (according to the XPS BE lookup table). As seen from Fig. 5c, it is evident that the XPS core peak of F 1s at 685.7 eV is too weak to be observed clearly because F content is much lower than that of Mn. In Fig. 5d, Mn 2p1/2 and Mn 2p3/2 are

PT

responsible for the bands at 652.8 eV and 641.4 eV, which are slightly decreased

RI

compared to the pristine LiMn2O4 (653.2 and 641.7 eV) [24]. It indicates that the

SC

average valence state of Mn ions is decreased, which will be beneficial to the increase

U

of the discharge capacity.

N

3.2 Electrochemical performance

A

The initial charge-discharge curves of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1,

M

0.2) samples at the rate of 1 C between 3.0 and 4.4 V at the room temperature are

D

shown in Fig. 6 Being similar to usual charge/discharge curve of LiMn2O4, all the

TE

charge-discharge curves in Fig. 6 exhibit two obvious discharge plateaus associated

EP

with the two-stage mechanism of the electrochemical lithium intercalation and

CC

extraction at 3.9-4.0 V and 4.0-4.1 V. This indicates that the Mg and F co-doping does not change the intrinsic structure of LiMn2O4 during the insertion and extraction

A

processes of lithium ions. Moreover, it can be clearly found that the potential interval of the samples is less than one of the pristine LiMn2O4 sample, indicating that the co-substituted LiMgxMn2-xO4-2xF2x samples have higher reaction kinetics than the pristine LiMn2O4. The LiMg0.05Mn1.95O3.9F0.1 and LiMg0.1Mn1.9O3.8F0.2 deliver the

initial discharge capacity of 119.1 mAh g-1 and 121.4 mAh g-1, which are higher than the pristine LiMn2O4 (117.2 mAh g-1). The capacity augment probably attributes to the increase of the molar ratio of Mn3+/Mn4+. Besides, the discharge capacity of the LiMg0.2Mn1.8O3.6F0.4 sample is lower than the pristine LiMn2O4 sample, it is most likely contributed to the overmuch substitution of Mg, which probably leads to the decrease of the electrochemically active Mn3+.

PT

The cycling performance of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2)

RI

samples at 1 C between 3.0 and 4.4 V at room temperature is shown in Fig. 7a. As

SC

seen, the discharge capacity of the pristine LiMn2 O4 fades from 116.6 to 94.0 mAh g-1

U

after 400 cycles, and only 79.0% of the original capacity is reserved. However, the

N

LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1, 0.2) maintain 94.8%, 96.0% and 90.1% of the

A

initial capacity after 400 cycles, which show distinctly better capacity retention than

M

the pristine LiMn2O4. These results suggest that the substituted samples exhibit more

D

stable structure during cycling processes because of the stronger bonding strength of

TE

Mg-O than that of Mn-O. On the other hand, the elevated temperature (55 °C)

EP

performance is usually an important criterion for cathode materials. As shown in Fig.

CC

7b, the improvement of the elevated temperature cycling performance appears to be much notable than those at the room temperature. The discharge capacity of the

A

pristine LiMn2O4 fades from 127.2 to 60.6 mAh g-1 with a capacity retention of 47.6% after 400 cycles between 3.0 and 4.4 V at 1 C. However, the LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1, 0.2) samples maintain 81.8%, 89.2% and 88.4% of the initial capacity after 400 cycles, respectively. Apparently, these data with the same fabrication of electrode

composition are obviously higher than the results reported by other groups [15,23,25]. The significant improvement of the capacity retention at elevated temperature further suggests that the substituted samples possess more stable structure, which is probably due to suppression of the Jahn-teller distortion and reduction of the Mn dissolution. In addition, changes in the structure are usually derived from the dissolution of active materials [26]. Thus, the fully charged electrodes were stored in fresh electrolytes for

PT

2 weeks at 55 °C, and the dissolved amounts of transition metal element was then

RI

measured by atomic absorption spectroscopy. After 2 weeks, the dissolved Mn

SC

concentrations for the LiMn2O4 sample and LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1, 0.2)

U

samples were 80.6 ppm, 20.5 ppm, 18.6 ppm, 16.3 ppm, respectively. Based on the

N

above results, the substituted samples exhibit more stable structure.

A

In order to verify the positive effect of the Mg and F co-substitution on LiMn2O4,

M

the further XRD measurements of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2)

D

samples before and after 50 cycles at 55 °C were finished. The corresponding

TE

comparison on the XRD results was presented in Fig. 8a and b. Obviously, the

EP

conventional LiMn2O4 sample reveals the coexistence of the original cubic phase with

CC

the tetragonal phase after 50 cycles, which is a clear indication of structural degradation of host structure. However, the XRD patterns of the LiMgxMn2-xO4-2xF2x

A

(x= 0.05, 0.1, 0.2) samples show that the original cubic structure was well maintained throughout the cycling test. Based on above results, it is considered that the Mg and F co-substitution LiMn2O4 can reveal much higher structural stability. Good rate capability is usually required for lithium-ion batteries in the field of

high power applications. This demands that the electrode can release the most of the self capacity when discharged at high current densities. For this purpose, the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples are charged at the same current density of 0.5 C and then discharged at different current densities from 1 to 20 C between 3.0 V and 4.4 V. Fig. 9 compares the discharge capacities of the cathodes at various rates. Since the ohmic polarization and electrochemical polarization result

PT

from the limited lithium ion diffusion rate in the spinel particles, the four samples

RI

show the decreased capacities at high rates, but it decreases more slowly after

SC

substitutions. Especially, the LiMg0.1Mn1.9O3.8F0.2 presents the best rate capability

U

among all the samples. It delivers the discharge capacity of 121.1 (1 C), 119.7 (2 C),

N

114.7 (5 C) and 95.0 mAh g−1 (10 C), respectively. Even at the rate of 20 C, the

A

discharge capacity of LiMg0.1Mn1.9O3.8F0.2 is still as high as 76.1 mAh g−1. However,

M

the discharge capacity of the pristine LiMn2O4 sample is abruptly decreased to 45.9

D

mAh g-1 at 20 C. As well known, the polarization is usually caused by limited

TE

electronic conductivity and slow transfer rate of Li ions, it will lead to deterioration in

EP

the performances of cathodes at high current density. The co-substitutions with Mg for

CC

Mn site and F for O site result in a larger cell volume, which probably broaden the transport passageway for Li+, facilitating Li+ ions insertion/extraction, especially in

A

the high rate regimes. In order to further analyze the electrochemical properties and the conductivity of the electrode materials, EIS measurements are employed. Before EIS tests, the cells are charged/discharged for several cycles. Fig. 10a shows the Nyquist plots of the

LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) after cycles at fully discharged state (about 3.0 V). Obviously, all the EIS spectra are consisted of a high-frequency intercept, a broad semicircle and an inclined line. The equivalent circuit, which is used to fit experimental results, is shown in Fig. 10b. The intercept in the high frequency region corresponds to the ohmic resistance (Rs), which combines resistance of the electrolyte and the contacts of the cell. The semicircle at middle frequency is correlated with the

PT

Li ion charge transfer resistance (Rct) at the interface. Meanwhile, the inclined line in

RI

the low frequency region is attributed to the lithium-ion diffusion in the bulk electrode

SC

[26,27]. The constant phase element (CPE) in the Fig. 10b stands for the double layer

U

capacitance. The fitted results are listed in Table 3. It can be found from Table 3 that

N

the Rs values are almost the same throughout the experiments because of the same

A

electrolyte and fabrication parameters. Also, the Rct of the pristine LiMn2O4 is 101.6

M

Ω, while the Rct of the LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1, 0.2) samples are 26.1 Ω,

D

31.0 Ω, 49.4 Ω, respectively. Apparently, the Rct of the Mg-F co-substitution samples

TE

are much less than one of the pristine LiMn2O4 sample, indicating that the

EP

electrochemical reaction resistance is dramatically reduced by Mg-F co-substitution.

CC

On the other hand, the small Rct is favorable to the rapid electrochemical reaction and

A

may result in better electrochemical performance of the active materials [14].

4. Conclusions Mg and F co-doping LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples are synthesized by a solid-state reaction route. The Mg and F ions have been successfully

introduced into the LiMn2O4 lattice. The XRD patterns of the pristine LiMn2O4 and LiMg0.1Mn1.9O3.8F0.2 are also further analyzed by Rietveld refinements, and the atoms in the LiMg0.1Mn1.9O3.8F0.2 crystal are located in the following sites: Li atoms in 8a sites, Li, Mg, Mn atoms in 16d sites, and O, F atoms in 32e sites. Compared with the pristine LiMn2O4, the LiMg0.1Mn1.9O3.8F0.2 exhibits a much higher discharge capacity of 121.4 mAh g-1 between 3.0 V and 4.4 V at a rate of 1 C and excellent lithium

PT

intercalation stability with capacity retention of 96.0% after 400 cycles at room

RI

temperature, while those of pristine LiMn2 O4 are only 117.2 mAh g-1 and 79.0%.

SC

Besides, the LiMg0.1Mn1.9O3.8F0.2 can deliver a remarkably high discharge capacity of

U

76.1 mAh g-1 even at the high charge-discharge rate of 20 C. The outstanding

M

A

material for power lithium-ion battery.

N

electrochemical properties of LiMg0.1Mn1.9O3.8F0.2 make it a promising cathode

D

Acknowledge

TE

This work was funded by the National Natural Science Foundation of China

EP

under project No. 51272221, Scientific and Technical Achievement Transformation

CC

Fund of Hunan Province under project No. 2012CK1006, Key Project of Strategic New Industry of Hunan Province under project No. 2013GK4018, and Science and

A

Technology plan Foundation of Hunan Province under project no. 2013FJ4062.

Reference [1] M. Armand and J.M. Tarascon, Building better batteries, Nature 451 (2008) 652.

[2] P.G. Bruce, B. Scrosati, J.M. Tarascon, Nanomaterials for Rechargeable Lithium Batteries, Angew. Chem. Int. Ed. 47 (2008) 2930. [3] J.B. Goodenough and Y. Kim, Challenges for Rechargeable Li Batteries, Chem. Mater. 22 (2010) 587. [4] D.H. Jang, Y.J. Shin, S.M. Oh, Dissolution of Spinel Oxides and Capacity Losses in 4 V Li / LixMn2O4 Cells, J. Electrochem. Soc. 143 (1996) 2204.

PT

[5]J.B. Jiang, K. Du, Y.B. Cao, Z.D. Peng, G.R. Hu, J.G. Duan, Syntheses of spherical

RI

LiMn2O4 with Mn3O4 and its electrochemistry performance, J. Alloys Compd. 577

SC

(2013) 138.

U

[6] H.G. Tang, Z.R. Chang, H.L. Zhao, X.Z. Yuan, H.J. Wang, S.Y. Gao, Effects of

N

precursor treatment on the structure and electrochemical properties of spinel LiMn2O4

A

cathode, J. Alloys Compd. 566 (2013) 16.

M

[7] Y. Gao and J.R. Dahn, Correlation between the growth of the 3.3 V discharge

D

plateau and capacity fading in Li1+xMn2-xO4 materials, Solid State Ionics 84 (1996) 33.

TE

[8] S.W. Lee, K.S. Kima, H.S. Moona, H.J. Kimb, B.W. Chob, W. Chob, J.B. Ju, J.W.

EP

Park, Electrochemical characteristics of Al2O3-coated lithium manganese spinel as a

CC

cathode material for a lithium secondary battery, J. Power Sources 126 (2004) 150. [9] J.F. Xiang, C.X. Chang, L.J. Yuan, J.T. Sun, A simple and effective strategy to

A

synthesize Al2O3-coated LiNi0.8Co0.2O2 cathode materials for lithium ion battery, Electrochem. Commun. 10 (2008) 1360. [10] W.K. Kim, D.W. Han, W.H. Ryu, S.J. Lim, H.S. Kwon, Al2O3 coating on LiMn2O4 by electrostatic attraction forces and its effects on the high temperature

cyclic performance, Electrochim. Acta 71 (2012) 17. [11] D. Arumugam, G.P. Kalaignan, Synthesis and electrochemical characterizations of Nano-SiO2-coated LiMn2O4 cathode materials for rechargeable lithium batteries, J. Electroanal. Chem. 624 (2008) 197. [12] D. Arumugam, G.P. Kalaignan, Synthesis and electrochemical characterization of nano-La2O3-coated nanostructure LiMn2O4 cathode materials for rechargeable lithium

PT

batteries, Mater. Res. Bull. 45 (2010) 1825.

RI

[13] P. Arora, B.N. Popov, R.E. White, Electrochemical Investigations of

SC

Cobalt-Doped LiMn2O4 as Cathode Material for Lithium-Ion Batteries, J.

U

Electrochem. Soc. 145 (1998) 807.

N

[14] S. Mandal, R.M. Rojas, J.M. Amarilla, P. Calle, N.V. Kosova, V.F. Anufrienko,

A

J.M. Rojo, High Temperature Co-doped LiMn2O4-Based Spinels. Structural, Electrical,

M

and Electrochemical Characterization, Chem. Mater. 14 (2002) 1598.

D

[15] D. Capsoni, M. Bini, G. Chiodelli, P. Mustarelli, V. Massarotti, C.B. Azzoni, M.C.

TE

Mozzati, L. Linati, Inhibition of Jahn-Teller Cooperative Distortion in LiMn2O4

EP

Spinel by Ga3+ Doping, J. Phys. Chem. B 106 (2002) 7432.

CC

[16] L.F. Xiao, Y.Q. Zhao, Y.Y. Yang, Y.L. Cao, X.P. Ai, H.X. Yang, Enhanced electrochemical stability of Al-doped LiMn2O4 synthesized by a polymer-pyrolysis

A

method, Electrochim. Acta 54 (2008) 545. [17] C. Sigala, D. Guyomard, A. Verbaere, Y. Piffard, M. Tournoux, Positive electrode materials with high operating voltage for lithium batteries: LiCryMn2-yO4 (0 ≤ y ≤ 1), Solid State Ionics 81 (1995) 167.

[18] Y.J. Kang, J.H. Kim, Y.K. Sun, Structure and electrochemical study of Li-Al-Mn-O-F spinel material for lithium secondary batteries, J. Power Sources 146 (2005) 237. [19] G.M. Song, W.J. Li, Y. Zhou, Synthesis of Mg-doped LiMn2O4 powders for lithium-ion batteries by rotary heating, Mater. Chem. Phys. 87 (2004) 162. [20] M. Takahashi, T. Yoshida, A. Ichikawa, K. Kitoh, H. Katsukawa, Q. Zhang, M.

PT

Yoshio, Effect of oxygen deficiency reduction in Mg-doped Mn-spinel on its cell

RI

storage performance at high temperature, Electrochim. Acta 51 (2006) 5508.

SC

[21] B.W. Ju, X.Y. Wang, C. Wu, X.K. Yang, H.B. Shu, Y.S. Bai, W.C. Wen, X. Yi,

U

Electrochemical performance of the grapheme/Y2 O3/LiMn2O4 hybrid as cathode for

N

lithium-ion battery, J. Alloys Compd. 584 (2014) 454.

A

[22] G.-H. Kim, S.-T. Myung, H.J. Bang, J. Prakash, Y.-K. Sun, Synthesis and

M

Electrochemical Properties of Li[Ni1/3Co1/3Mn(1/3-x)Mgx]O2-yFy via Coprecipitation,

D

Electrochem. Solid-State Lett. 7 (2004) A477.

TE

[23] X.M. He, J.J. Li, Y. Cai, Y.W. Wang, J.R. Ying, C.Y. Jiang, C.R. Wan, Fluorine

EP

doping of spherical spinel LiMn2O4, Solid State Ionics 176 (2005) 2571.

CC

[24] Y.J. Wei, L.Y. Yan, C.Z. Wang, X.G. Xu, F. Wu, G. Chen, Effects of Ni Doping on [MnO6] Octahedron in LiMn2O4, J. Phys. Chem. B 108 (2004) 18547.

A

[25] K.M. Elsabawy, M.M. Abou-Sekkina, F.G. Elmetwaly, Structure visualization and yttrium(III)-dopings on LiMn2-xYxO4 for promoting structural microstructural and cathodic capacity features of LiMnO-spinel, Solid state sciences 13 (2011) 601. [26] K.-S. Lee, S.-T. Myung, K. Amine, H. Yashiro, Y.-K. Sun, Dual functioned

BiOF-coated Li[Li0.1Al0.05Mn1.85]O4 for lithium batteries, J. Mater. Chem. 19 (2009) 1995. [27] I. Jeong, J. Kim, H. Gu, Electrochemical properties of LiMgyMn2−yO4 spinel phases for rechargeable lithium batteries, J. Power Sources 102 (2001) 55.

RI

PT

Fig. 1 SEM images of high magnification of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2). (a) x= 0; (b) x= 0.05; (c) x= 0.1; (d) x= 0.2. The insets are the relative low magnification of the LiMgxMn2-xO4-2xF2x

SC

Fig. 2 TG-DTA curves of the precursor: (a) undoped spinel and (b) doped spinel

U

Fig. 3 XRD patterns of the LiMn2O4 sample and the LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1, 0.2)

N

samples

A

Fig. 4 Rietveld refinements of X-ray diffraction patterns for the pristine LiMn2O4 and

M

LiMg0.1Mn1.9O3.8F0.2. The experimental data are indicated by red dots and the calculated profile by

TE

intensities at each step

D

the overlaid black line; the lower red curve is the difference between the observed and calculated

EP

Fig. 5 XPS spectra of LiMg0.1Mn1.9O3.8F0.2: survey spectra (a) and core-level spectra of Mg 1s (b),

CC

F 1s (c), Mn 2p1/2 and Mn 2p3/2 (d) Fig. 6 Initial charge-discharge curves of the LiMn2O4 sample and the LiMgxMn2-xO4-2xF2x (x= 0.05,

A

0.1, 0.2) samples at 1 C in the voltage range of 3.0-4.4 V Fig. 7 Cycling performances of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples cycle at 1 C between 3.0 and 4.4 V at (a) room temperature, (b) 55 °C Fig. 8 The comparison on the XRD results of LiMn2O4 and LiMgxMn2-xO4-2xF2x (x= 0.05, 0.1, 0.2)

before and after their cycling Fig. 9 The rate performance of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2) samples in the voltage range of 3.0-4.4 V Fig. 10 (a) Nyquist plots of the LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1, 0.2). (b) Equivalent circuit

PT

used for fitting the experimental EIS data

SC

RI

Table 1 The AAS results of the pristine and Mg-F co-doped LiMn2O4

U

Chemical composition

LiMn2O4

Li1.002Mn2O4 Li1.001Mg0.051Mn1.95O4-2xF2x Li1.002Mg0.097Mn1.905O4-2xF2x Li1.005Mg0.196Mn1.812O4-2xF2x

D

LiMg0.1Mn1.9O3.8F0.2

CC

EP

LiMg0.2Mn1.8O3.6F0.4

A

Experimental

TE

M

LiMg0.05Mn1.95O3.9F0.1

A

N

Theoretical

Table 2 Structural parameters obtained from XRD Rietveld refinement for the pristine LiMn2O4 and LiMg0.1Mn1.9O3.8F0.2

LiMn2O4

LiMg0.1Mn1.9O3.8F0.2

Lattice constant (a) / nm

0.8245

0.8253

Cell volume (v) / nm3

0.5605

0.5622

Mn-O band average

1.970

Li-O band average

1.937

Rwp (%)

8.890

PT

Samples

RI

1.972

8.120

M

A

N

U

SC

1.940

Rs / Ω

Rct / Ω

LiMn2O4

4.6

101.6

LiMg0.05Mn1.95O3.9F0.1

3.5

26.1

LiMg0.1Mn1.9O3.8F0.2

3.2

31.0

LiMg0.2Mn1.8O3.6F0.4

2.8

49.4

CC

Samples

A

0.2)

EP

TE

D

Table 3 The AC impedance fitting results for LiMgxMn2-xO4-2xF2x (x= 0, 0.05, 0.1,

Fig. 1

(b)

(c)

(d)

A

CC

EP

TE

D

M

A

N

U

SC

RI

PT

(a)

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 2

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 3

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 4

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 5

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 6

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 7

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 8

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 9

TE

EP

CC

A D

PT

RI

SC

U

N

A

M

Fig. 10