Porous micrometer-sized MnO cubes as anode of lithium ion battery

Accepted Manuscript Title: Porous micrometer-sized MnO cubes as anode of lithium ion battery Author: Xiaoyong Fan Siheng Li Li Lu PII: DOI: Reference:

S0013-4686(16)30655-7 http://dx.doi.org/doi:10.1016/j.electacta.2016.03.114 EA 26944

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

20-1-2016 16-3-2016 20-3-2016

Please cite this article as: Xiaoyong Fan, Siheng Li, Li Lu, Porous micrometersized MnO cubes as anode of lithium ion battery, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2016.03.114 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.

Porous micrometer-sized MnO cubes as anode of lithium ion battery Xiaoyong Fana, b, Siheng Lib, Li Lub* Abstract In this study, porous micrometer-sized MnO cubes have been designed and synthesized by hydrothermal treatment followed by high temperature annealing. The pore size is controlled by changing annealing temperature in order to achieve good electrochemical performance. The cube edge length is about 10 μm and the pore size changes from mesoporous to macroporous. The presence of pores in the MnO cubes is able to accommodate the volumetric changes during electrochemical cycling, and enables electrolyte easy penetration so that to improve the electrochemical performance. The porous micrometer-sized MnO cubes prepared by hydrothermal treatment at 100 ℃ followed by annealing at 700 ℃ delivers the best long-term and rate cyclability owing to its stable porous structure serving as lithium ion rapid transfer channels and enough pore volume to accommodate volumetric changes during electrochemical cycling. The reversible capacity in the first cycle is 615.9 mAh g-1at 0.2 A g-1, slightly decreases to 404.6 mAh g-1 at 1.0 A g-1in the 6th cycle and remains at 425.5 mAh g-1 at 1.0 A g-1 even after 495 cycles. The same porous micrometer-sized MnO cube electrode delivers high rate reversible specific capacities of 201.8 and 50.4 mAh g-1 at 5.0 and 10.0 A g-1 respectively. Introduction Over the past decades, lithium ion batteries have been widely explored in portable electronic devices due to their high energy densities and environmental benignity [1-3]. However, their energy and power densities are still unable to meet the ever growing demands of the potential application fields such as electrical vehicles and energy storage station using conventional electrode materials. Therefore, the search for new electrode materials with high energy and power densities is urgent for next generation of lithium ion battery [4, 5]. Transition metal oxides (TMOs, M = Co, Ni, Cu, Fe, Mn) have widely been investigated as anodes of lithium ion battery owing to their

high theoretical specific capacities based on multi-electrons reaction [6]. Among them, MnO has attracted much attention recently because of its high theoretical capacity of 756 mAh g-1, that is more than twice the capacity of commercial graphite, a lower electromotive force (1.032 V vs Li/Li+) than other TMO anodes such as Fe3O4, Co3O4,

NiO, and CuO, low voltage hysteresis (<0.8 V), natural abundance and environmentally benign [7-9].

School of Materials Science and Engineering, Chang’an University, Xi’an 710061, China. E-mail: [email protected]; Fax: +86-29-82337340; Tel: +86-29-82337340 b Department of Mechanical Engineering, National University of Singapore, 9Engineering Drive 1, Singapore 117576, Singapore. E-mail: [email protected]; Fax:+65-67791459; Tel: +65-65162236 a

Unfortunately, like other transition metal oxides, poor cyclability caused by drastic volumetric variation during electrochemical cycling hinders its practical application [7-9]. In addition, the intrinsically low electronic conductivity of MnO leads to poor high rate capacity, which also limits its practical application [7-10]. To overcome these issues, carbon coating [11-15], dispersing in carbon matrix [16-21], composite with carbon nanotube [22-25] and graphene [26-28] were used because carbon can provide electron rapid transfer channels for MnO, and also alleviate the strain caused by the considerable volumetric changes during electrochemical cycling. However, this improvement is still limited since carbon is unable to strongly improve lithium ion transfer rate and to remain integrity after suffering repeated expansion/contraction during electrochemical cycling [29-31]. Porous structure is considered to be another effective method to improve the electrochemical performance of MnO since the feature of porous structure can not only provides lithium ion rapid transfer channels, but also accommodate volumetric changes during charging/discharging [32-35]. Combining carbon and porous structure, different porous MnO/C composites have been designed and produced to improve the electrochemical performance [36-43]. For instance, MnO/C microspheres with hollow porous structure have been prepared using biotemplating method by Xia et al [36]. The MnO/C microspheres could deliver 700 mAh g-1 at 0.1 A g-1 with enhanced rate performance of 230 mAh g-1 at 3.0 A g-1. This enhanced electrochemical performance was attributed to the fact that the free place in the hollow interior and the porosity in the shell can favorably accommodate the volumetric changes and provide lithium ion transfer channels during electrochemical cycling, the porous carbon matrix can enhance the electrical connectivity between MnO nanoparticles and also serves as an elastic barrier to buffer the strain caused by volumetric changes. Su et al [37] have produced 3D porous MnO/C microspheres through preparing MnxZn1–xCO3precursors, followed by annealing with glucose, and finally leaching of Zn. The pore size was controlled from 14.9 nm to 31.8 nm by adjusting the Zn/Mn molar ratio in the MnxZn1–xCO3 precursors. This composite shows good rate capacity 846 mAh g-1 at 100 mA g-1 and 406 mAh g-1 at 3200 mA g-1. Although good electrochemical performance has been achieved by constructing different porous MnO/C composites, the gravity and volumetric specific capacities are lower when consider the whole mass and volume of MnO and carbon. In addition, existence of carbon may lead to large first irreversible capacity. Therefore, how to improve the electrochemical performance of pristine MnO electrode is always a challenge to researchers. Inspired by above studies, micrometer-sized MnO cubes with controllable pores are constructed in this work by low temperature hydrothermal treatment followed by high temperature annealing to improve their electrochemical performance. The pores in micrometer-sized MnO cubes can provide lithium ion rapid transfer channels and

accommodate the volumetric changes as well during charging/discharging. In addition, the unique morphology can alleviate aggregation of nanoparticles and the porous structure can remain stable to some degree during charging/discharging. Therefore, the long-term and rate cycle performances are improved. 1. Experimental 1.1 Materials preparation Porous micrometer-sized MnO cubes were produced by low temperature hydrothermal treatment using Mn(NO3)2·4H2O and urea as the raw materials followed by high temperature annealing. In a typical synthesis, 10 mmol Mn(NO3)2·4H2O and 50 mmol urea were firstly dissolved in 40 mL H2O, which was then sealed in Teflon lined stainless steel autoclave and hydrothermally treated at 100 ℃ for 16 h. After hydrothermal treatment, the precipitates was collected and annealed at four different temperatures, namely 350, 400, 500 and 700 ℃ in an Ar atmosphere with a ramp of 2 ℃ min-1 for 4 h. The three batches of powder heat-treated at 400, 500 and 700 ℃ are termed as M400, M500 and M700, respectively. 1.2 Materials characterization The crystal structure was characterized using X-ray powder diffraction (XRD) performed on a Shimadzu XRD-6000 diffractometer using Cu Kα radiation (λ= 0.154 nm)，with a 2θ scan range from 10° to 80°. Morphology of the powder particles was observed using a Hitachi S-4300 scanning electron microscope. Thermogravimetric analysis (TGA, Shimadzu DTG-60H) was used in the range from room temperature to 800 ℃ with a ramp of 10 ℃ /min in Ar atmosphere. The specific surface area was measured by the Bruauer-Emmett-Teller (BET) method using nitrogen adsorption isotherms at 77 K on Micromeritics’ Gemini VII 2390 Series of surface area analyzer. Pore size distribution plots were obtained by Barret-Joyner-Halenda (BJH) method. 2.3 Electrochemical characterization The electrodes were prepared by coating the slurry consisting of 70% active materials, 20% super-P carbon black, 10% polyvinylidene fluoride(PVDF) and appropriate n-methyl-2-pyrrolidone (NMP) on the copper foil and dried at 120 ℃ in vacuum for over 12 h. The mass loading of the active material is about 1.0 mg/cm2. Coin-type cells were assembled in an argon-filled dry glove box using the prepared electrodes as the positive electrodes, Li metal as the negative electrode, a membrane (Celgard 2400) as the separator, and 1M LiPF6 dissolved in ethylene carbonate (EC), dimethyl carbonate (DMC) and diethylene carbonate (DEC) (1:1:1 by volume, provided by Guotaihuarong, Zhangjiagang, China) as the electrolyte. Charging/discharging measurements were carried out on a

Land 2000A battery test system at different current densities in the voltage range from 0.05 to 3.0 V. Electrochemical characterization in terms of cyclic voltammetry (CV) was carried out using Solartron Analytical 1470E CellTest System at different scan rates. Electrochemical impedance was also tested at the potential of 0.7 V (vs. Li/Li+) using Solartron Analytical 1470E CellTest System combined with a Solartron Analytical 1400 CellTest System, over a frequency range of 100 KHz to 10 mHz with a signal amplitude of 5mV.

3. Results and discussion 3.1 Structure and morphology characterization XRD spectra as shown in Fig.1 demonstrate that the samples before and after high temperature annealing at 350 ℃ mainly contain MnCO3 rhombohedral phase with a space group of R-3c (JCPDS no. 44-1472). When the annealing temperature was increased to 400 ℃ and above, all the samples are mainly composed of MnO cubic phase with a space group of Fm-3m (JCPDS no. 07-0230). The reaction is given in Eq. 1, MnCO3

MnO +CO2

(1)

It can be also observed that the diffraction peaks become narrow upon increasing annealing temperature, which indicates increased crystallinity. Fig. 2 reveals the TGA curve of the sample before high temperature annealing. There appear a little weight loss at 400 ℃ and a large weight loss at 500 ℃. The sharp weight loss is associated with dissociation of MnCO3 starting at 400 ℃ and completed at 500 ℃ with total weight loss of about 37.2%, which is in agreement with the weight loss (38.2%) calculated from Eq.1. Fig. 3 shows the SEM images of the samples before and after high temperature annealing at 350 ℃. The sample before high temperature annealing displays solid cube-like shape with an edge length of about 10 m (Fig. 3a and 3b). As shown in Fig. 3c and 3d, the cube-like shape remained after annealing at 350 ℃ in Ar atmosphere but surface became rough. All the samples of M400, M500 and M700 after annealing at 400, 500 and 700 ℃ remained cube-like shape, but a large number of pores with different size formed due to relief of CO2 during annealing. Meanwhile, we can also observe that the crystalline size increases upon increasing annealing temperature. M400 is piled up by a large number of MnO nanoparticles with the size of about 50 nm, results in a lot of mesopores between the MnO nanoparticles (Fig. 4a and 4b). As shown in Fig. 4c and 4d, M500 is composed of a large number of MnO nanoparticles with the size of about 200 nm and a lot of interconnected pores with the size of about 50 nm. Fig. 4e and 4f show morphology of the M700 sample consisting of a large number of MnO particles with the size of about 2 m and a lot of interconnected pores with the size of about 1 m. This morphology has high tap density compare with normal nanoparticles and is beneficial to lithium ion transport as

well. To further evaluate the specific surface area and the porous nature of M400, M500 and M700, the N2 adsorption isotherms at 77 K and BJH adsorption pore size distribution plots are shown in Fig. 5. The BET specific surface areas of M400, M500 and M700 are determined to be 10.0, 4.7 and 1.3 m2 g-1 respectively. For the pore size distribution plots of M400, there mainly appear three peaks at 2.4 nm, 16.1 nm and 31.1 nm respectively, indicating the pores are in mesoporous region, which is in agreement with the SEM images shown in Fig.4b. For the pore size distribution plots of M500, there are mainly two broad pore size distribution regions around 104.5 nm and 172.1 nm, which is consistent of the SEM images shown in Fig. 4d. Seen from Fig. 4f, the pores in M700 are in macroporous region, which is hard to evaluate by N2 adsorption method, therefore there is no pore size distribution peak.

In addition, the pore size distribution curve (inset Fig. 7) based on Barret–Joyner–Halenda (BJH) method clearly indicates that pore is in the mesoporous region, and the pore size distribution is very narrow, at about 12.6 nm which is in good agreement with the TEM observations. These results show that hierarchical flower-like structure has large surface area which is due to the mesoporous embedded in the structure.

3.2 Electrochemical characterization Fig. 5 shows the charge/discharge curves of M400, M500 and M700 and their long-term cyclability at current density of 1.0 A g-1 (initial 5 cycles are tested at 0.2 A g-1 for activation). For the first discharge of the M400 electrode at 0.2 A g-1, a short voltage plateau at about 1.5 V appears followed by a long voltage plateau at about 0.3 V and a voltage slop down to 0.05 V (Fig. 5a). The short voltage plateau at about 1.5 V can be ascribed to electrolyte decomposition on the electrode surface, so that it disappears from the second cycle. The long voltage plateau at about 0.3 V and voltage slop down to 0.05 V can be attributed to the reaction between Li+ and MnO (MnO + 2Li+↔Mn0 +Li2O). A long voltage plateau appeared at about 1.25 V in the first charge. The first discharge and charge capacities of M400 are 821.2 and 585.9 mAhg-1, respectively, with a coulomb efficiency of 71.3%. For the second cycle, the charge voltage plateau remained almost the same but the long discharge voltage plateau at 0.3 V increased to 0.53 V, indicating decrease in the electrochemical polarization due to formation of MnO nanocrystalline after the first electrochemical cycle [8, 29]. The coulomb efficiency increased to 92.9%. When the current density was increased to 1.0A g-1, the discharge voltage plateau slightly decreased to about 0.44 V whereas the charge voltage plateau increased to about 1.45 V due to the electrochemical polarization. Upon

cycling, the discharge voltage plateau gradually decreased, while the charge voltage plateau gradually increased due to further electrochemical polarization. It is also noted that their length became shorter. In the 150th cycle, the discharge and charge voltage plateau became very short with the discharge and charge capacities of 130.1 and 129.3 mAhg-1, respectively, implying rapid capacity failure, and almost remained stable after 150 cycles. The discharge/charge curves of M500 shown in Fig. 5b are similar to those of M400. There is a short voltage plateau at about 0.7 V in the first discharge corresponding to electrolyte decomposition on the electrode surface, and a long voltage plateau at about 0.19 V corresponding to the reaction of Li+ and MnO. The first discharge and charge capacities are 993.1 and 706.1 mAh g-1 respectively, with a coulomb efficiency of 71.1%. The long voltage plateau is lower than that of M400, which can be attributed to their morphology. From the SEM images shown in Fig. 4b and 4d, the sub-particle size of M400 is smaller than that of M500, which results in shorter lithium ion diffusion distance. Therefore the electrochemical polarization of M400 in the first cycle is smaller than that of M500. Similarly, since MnO nanocrytalline formed in the second cycle, the electrochemical polarization became smaller. Therefore the discharge voltage plateau increased to 0.59 V. When the current density was increased to 1.0 A g-1, relatively large polarization can be also observed from the discharge and charge voltage plateau. During long-term cycling, M500 electrode showed very small polarization up to 150 cycles, after which the charge/discharge voltage plateau was polarized accompanied with reduction in length of voltage plateaus. The discharge and charge capacities in the 150th cycle remained 529.5 and 523.9 mAh g-1 respectively. The similar behaviors was also observed from M700. When it was first discharged at 0.2 A g-1, a short voltage plateau at about 0.67 V appeared. This voltage plateau is associated with electrolyte decomposition, followed by a long voltage plateau at about 0.15 V corresponding to the reaction of Li+ and MnO. It shows the first discharge capacity of 1083.3 and first charge capacity of 615.9 mAh g-1, with the coulomb efficiency of 56.9%. The voltage plateau is lower than that of M400 and M500. A possible reason is that the particle size of M700 is larger than that of M400 and M500, resulting in longer lithium ion diffusion distance. In the second cycle, the electrochemical polarization became smaller and the discharge voltage plateau positively shifted to 0.61 V due to formation of MnO nanocrytalline. At high charge current 1.0 A g-1, the voltage plateau increased to about 1.2 V. Different from M400 and M500, the discharge and charge voltage plateaus of M700 almost remained unchanged within 150 cycles. After 500 cycles, the discharge voltage plateau of M700 remained at about 0.26 V, which is higher than that of M400 (almost no voltage plateau) and M500 (about 0.14 V), indicating lower electrochemical polarization. Fig. 5d shows the long-term cyclability of M400, M500 and M700. It can be observed that M700 delivers the best cyclability with gradual increase in the capacities and then cyclability almost remained stable upon cycling.

However, the capacity of M400 continuously decreased whereas the capacity of M500 slightly increased and then continuously decreased upon cycling. As shown in Fig. 4, since M400 has the smallest sub-particle size, it led to electrolyte decomposition on the electrode surface, which can be seen from the lower coulomb efficiency than that of M500 and M700 in initial 150 cycles (Fig. 5d). In addition, the mesopores in M400 are formed by just simply piling up of MnO nanoparticles (Fig. 4b) and easily cracked and blocked during charging/discharging as shown in Fig. 7a and 7b, which also results in capacity failure. M500 has larger interconnected pore size than that of M400 and smaller sub-particle size than that of M700, therefore it has the largest first reversible specific capacity. However, the pores were also filled up by active materials and products of electrolyte decomposition after repeated volumetric expansion and contraction (Fig. 7c and 7d), then the lithium ion transfer channels are blocked, as a result the reversible specific capacity of M500 rapidly decreases after about 150 cycles. M700 has the largest sub-particle size and pore size, resulting in the reversible capacity is lower than other two electrodes in initial several cycles at current density of 1.0 A g-1 due to longer lithium ion transfer distance. However it has enough pore volume for accommodating the volumetric changes of active particles, less electrolyte decomposition, stable pore structure which almost remained after 500 cycles (Fig. 7 e and 7f), therefore it has the best long-term cyclability. Fig. 6 shows the rate capabilities of the M400, M500 and M700. M700 delivers 201.8 mAh g-1 of the reversible capacity at current density of 5.0 A g-1, which is similar to that of M500 (208.4 mAh g-1) and much better than that of M400 (69.8 mAh g-1). Meanwhile, the reversible capacity at 10.0 A g-1 of M700 are 50.4 mAh g-1, which is a little better than that of M500 (38.9 mAh g-1) and much better than that of M400 (4.6 mAh g-1). According to the SEM images shown in Fig. 4, the pores in M400 are the smallest, which may be not enough to accommodate the volumetric changes during electrochemical cycling, the lithium ion transfer channels are blocked due to the pores are filled up by active materials caused by volumetric changes and products of electrolyte decomposition, as a result the rate capacity of M400 is smallest. 3.3 Kinetics Characterization To investigate the electrochemical kinetics, cyclic voltammetry (CV) and electrochemical impedance were tested. Fig. 8a, 8b and 8c show the CV curves at different scanning rates of M400, M500 and M700. It can be found that the cathodic peaks negatively move and anodic peaks positively move a little for all the three electrodes when the scanning rates increased. To compare the lithium ion diffusion coefficient (DLi) of M400, M500 and M700, their anodic peak currents Ip versus square root of sweep rate ν1/2 are shown in Fig. 8d. For the M400, due to the plots at 1.0 and 2.0 mV s-1 seriously deviate from liner relation, so only the plots at 0.1, 0.25 and 0.5 mV s-1

were used for liner fitting. For M500, the plots at 0.1, 0.25 and 0.5 and 1.0 mV s-1 were used for liner fitting because the plot at 2.0 mV s-1 seriously deviate from liner relation. The anodic peak current Ip are proportional to the square root of the scan rate ν1/2, indicating the lithium ion diffusion is the velocity controllable step in the electrochemical process. Thus, the DLi can be estimated based on Eq 2, Ip＝2.69×105An3/2ΔCLi(DLi) 1/2ν1/2.

Eq 2

where Ip is the peak current and n is the number of electrons involved in the electrochemical reaction. For simplicity, 2 is used here according to MnO + 2Li+↔Mn0 +Li2O. A is the real contact area between working electrode and electrolyte. For simplicity, the electrode geometric area, 1.13 cm2, is used here. CLi is the molar concentration of Li+ ions and can be determined via CLi (mol cm-3) = nLi /NA (mol-1) V (cm3). The nLi is the number of Li in the crystal cell

(the lithiated products Mn and Li2O are considered as the crystal cell here for simplicity). Because z=4 for the cubic MnO crystal cell, therefore 8 Li+ ions are in (Mn+Li2O) crystal cell. NA is the Avogadro's constant, 6.02× 1023 mol-1. V is volume of the crystal cell (Mn+Li2O). The volume of the crystal cell for cubic MnO is 8.782×10-23 cm3. There exist 200-300% volume expansion during lithiation of MnO. If 200% volume expansion is used here,

the volume of the crystal cell (Mn+Li2O) is 1.756×10-22 cm3). Therefore CLi is estimated to be 7.566×10-2 mol cm-3. Considering the slops of M400, M500 and M700 are 1.163, 0.9358 and 0.7530 respectively, the DLi of them are estimated via Eq.2 to be 3.193×10-10, 2.068 ×10-10 and 1.339×10-10 cm2 s-1 respectively. Fig. 9 shows the Nyquist plots of M400, M500 and M700 before electrochemical cycling and after 500 cycles. There are two arcs in high and middle frequency respectively, and a straight line in low frequency for all the three electrodes before electrochemical cycling. In generally, high frequency arc is attributed to solid electrolyte interphase (SEI) film caused from the electrolyte decomposition on the electrode surface and increase upon cycling. However, all the high frequency arcs of M400, M500 and M700 become smaller after 500 cycles, so they are mainly associated with the contact problems between active materials and current collector, or the initial presence of absorbed bubbles in porous electrodes and minor associated with SEI film. [44-48] The middle frequency arc presents the charge transfer impedance and the low frequency straight line presents the lithium ion diffusion in the particles. The high frequency and middle arcs are almost equal for all the three electrodes, which indicates their electrochemical impedance are almost equal before electrochemical cycling. The slops of the low frequency straight line are also equal, which indicates the lithium ion diffusion coefficients are equal, which is in

agreement with the results of CV. The middle/low frequency arcs of the M400 and M500 become much larger after 500 cycles (Fig. 9b) indicating increase of charge transfer impedance, resulting in their electrochemical cyclability are worse than that of M700. 4. Conclusion Porous micrometer-sized MnO cubes with pore size from mesoporous to macroporous were fabricated from MnCO3 cubes formed from hydrothermal treatment. The relationship between the pore structure and electrochemical performance was investigated and the results demonstrate that the porous micrometer-sized MnO cube annealed at 700 ℃ has the best long-term cycle performance and high rate capacity due to it has interconnected and stable pores serve as lithium ion transfer channels and enough pore volume for accommodating volumetric changes during electrochemical cycling. The first reversible specific capacity at 0.2 Ag-1 is 615.9 mAhg-1 and remains 425.5 mAh g-1 after 495 cycles at 1.0 A g-1. The reversible specific capacities at 5.0 A g-1 and 10.0 A g-1 remain 201.8 and 50.4 mAh g-1, respectively. Acknowledgement This research is financial supported by the Special Fund for Basic scientific Research of Central Colleges, Chang'an University, China (Grant No. 310831153505). References [1] J. M. Tarascon, M. Armand, Nature 414 (2001) 359-367. [2] J. B. Goodenough and Y. Kim, Chem. Mater. 22 (2010) 587-603 [3] M. Armand, J.M. Tarascon, Nature451 (2008) 652-657 [4] B. L. Ellis, P. Knauth, and T. Djenizian, Adv. Mater. 26 (2014) 3368-3397 [5] H. Wang, H. B. Feng and J. H. Li, Small 10 (2014) 2165-2181 [6] P. Poizot, S. Laruelle, S. Grugeon, L. Dupont and J. M. Tarascon, Nature 407 (2000) 496-499. [7] X. Q. Yu, Y. He, J. P. Sun, K. Tang, H. Li, L.Q. Chen, X. J. Huang, Electrochem.Commun.11(2009) 791-794 [8] K. F. Zhong, X. Xia, B. Zhang, H. Li, , Z. X. Wang, L. Q. Chen, J. Power Sources 195(2010)3300-3308 [9] G. L. Xu,Y. F. Xu, J. C. Fang, F. Fu, H. Sun, L. Huang, S. H. Yang and S. G. Sun, ACS Appl. Mater. Interfaces 5 (2013) 6316-6323 [10] J. Liu, Q. M. Pan, Electrochem. Solid state Lett. 13 (2010) A139-A142 [11] X. W. Li, S. L. Xiong, J. F. Li, X. Liang, J. Z. Wang, J. Bai and Y. T. Qian, Chem. Eur. J. 19 (2013) 11310-11319 [12] S.M. Guo, G.X. Lu, S. Qiu, J.R. Liu, X.Z. Wang, C.Z. He, H.G. Wei, X.R. Yan, Z.H. Guo, Nano Energy 9

(2014) 41-49 [13] S. B. Wang, Y. B. Ren, G. R. Liu, Y. L. Xing and S. C. Zhang, Nanoscale 6 (2014) 3508-3512 [14] B. Liu, X. L. Hu, H. H. Xu, W. Luo, Y. M. Sun and Y. H. Huang, Sci. Rep.4 (2014) 4229 [15] H. Liu, Z.H. Li, Y.R. Liang, R.W. Fu, D.C. Wu, CARBON 84 (2015) 419-425 [16] X.Y. Tao, W.C. Chai, F.Q. Xu, J.M. Luo, H. Xiao, C. Liang, Y.P. Gan, H. Huang, Y. Xia, W.K. Zhang, Electrochim. Acta 169 (2015) 159-167 [17] J. Gan Wang, C.B. Zhang, D.D. Jin, K.Y. Xie and B.Q. Wei, J. Mater. Chem. A3 (2015) 13699-13705 [18] R. C. Lee, Y. P. Lin, Y. T. Weng, H. A. Pan, J. F. Lee, N. L. Wu, J. Power Sources 253 (2014) 373-380 [19] C.Y. Zhu, N. Sheng and T. Akiyama, RSC Adv. 5 (2015) 21066-21073 [20] H. Jiang, Y. J. Hu, S. J. Guo, C. Y. Yan, P. S. Lee and C. Z. Li, ACS. Nano. 8 (2014) 6038-6046 [21] Y. M. Liu, X. Y. Zhao, F. Li, D. G. Xia, Electrochem. Acta 56 (2011) 6448-6452 [22] Y. L. Ding, C. Y. Wu, H. M. Yu, J. Xie, G. S. Cao, T. J. Zhu, X. B. Zhao, Y. W. Zeng,Electrochem. Acta,56 (2011)5844-5848 [23] J. W. Qin, Q. Zhang, Z. Y. Cao, X. Li, C. W. Hu, B. Q. Wei, Nano Energy2(2013) 733-741 [24] X. F. Sun, Y. L. Xu, P. Ding, M. R. Jia, G. Ceder, J. Power Sources 244 (2013) 690-694 [25] X. F. Sun, Y. L. Xu, P. Ding, G. G. Chen, X. Y. Zheng, R. Zhang,L. Li, J. Power Sources 255 (2014) 163-169 [26] K. J. Zhang, P. X. Han, L. Gu, L. X. Zhang, Z. H. Liu, Q. S. Kong, C. J. Zhang, S. M. Dong, Z. Y. Zhang, J. H. Yao, H. X. Xu, G. L. Cui and L. Q. Chen, ACS Appl. Mater. Interfaces 4 (2012) 658-664 [27] J. Zang, H. Qian, Z. K. Wei, Y. Cao, M. S. Zheng, Q. F. Dong, Electrochim. Acta 118 (2014) 112-117 [28] Y. M. Sun, X. L. Hu , W. Luo , F. F. Xia and Y. H. Huang, Adv. Funct. Mater.23(2013) 2436-2444 [29] K. F. Zhong, B. Zhang, S. H. Luo, W. Wen, H. Li, X. J. Huang, L. Q. Chen, J. Power Sources 196 (2011) 6802-6808 [30] Y. M. Sun, X. L. Hu, W. Luo and Y. H. Huang,J. Mater. Chem.22(2012) 19190 [31] Y.F. Deng, L.N. Wan, Y. Xie, X.S. Qin and G.H. Chen, RSC Adv.4(2014) 23914-23935 [32] X. W. Li, D. Li, L. Qiao, X. H. Wang, X. L. Sun, P. Wang and D. Y. He, J. Mater. Chem.22(2012) 9189-9194 [33] X. H. Ma, Q. Y. Wan, X. Huang, C. X. Ding, Y. Jin , Y. B. Guan, C. H. Chen, Electrochimica Acta 121 (2014) 15-20 [34] X.J. Jiang, Y.Y. Wang, L.S. Yang, D.W. Li, H.Y. Xu, Y. Ding, J. Power Sources 274 (2015) 862-868

[35] J.H Yang, X.F. Yang, Y.L Zhong, J. Y. Ying, Nano Energy 13 (2015) 702-708 [36] Y. Xia, Z. Xiao, X. Dou, H. Huang, X. H. Lu, R. J. Yan, Y. P. Gan,W. J. Zhu, J. P. Tu, W. K. Zhang and X. Y. Tao, ACS Nano 7 (2013) 7083-092 [37] K. Su, C. Wang, H. G. Nie, Y. Guan, F. Liu and J. T. Chen, J. Mater. Chem. A2(2014) 10000-10006 [38] S. M. Guo, G. X. Lu, S. Qiu, J. R. Liu, X. Z. Wang, C. Z. He, H. G. Wei, X. R. Yan, Z. H. Guo, Nano Energy9 (2014) 41-49 [39] W. Luo, X. L. Hu, Y. M. Sun and Y. H. Huang, ACS Appl. Mater. Interfaces 5 (2013) 1997-2003 [40] W.J. Zhu, H. Huang, W.K. Zhang, X.Y. Tao, Y.P. Gan, Y. Xia, H. Yang, X.Z. Guo, Electrochim. Acta 152 (2015) 286-293 [41] G. L. Xu, Y. F. Xu, H. Sun, F. Fu, X. M. Zheng, L. Huang, J. T. Li, S. H. Yang and S. G. Sun, Chem. Commun.48 (2012) 8502-8504 [42] T. Y. Wang, Z. Peng, Y. H. Wang, J. Tang and G. F. Zheng, Scientific Reports3 (2013) 2693 [43] J.G. Wang, C.B. Zhang and F.Y. Kang, ACS Appl. Mater. Interfaces 7 (2015) 9185-9194 [44] M. Holzapfel, A. Marinent, F. Allion, B. Le Gorrec, R. Yazami, C. Montella, J. Electroanalytical Chem. 546 (2003) 41 [45] M. Gaberscek, J. Moskon, B. Erjavec, R. Dominko and J. Jamnik, Electrochem. Solid-State Lett., 11 (2008) A170-A174 [46] X.Y. Fan, Q.C. Zhuang, G.Z. Wei, L. Huang, Q.F. Dong, S.G. Sun, J. Appl. Electrochem., 39 (2009) 1323-1330 [47] J. P. Schmidt, T. Chrobak, M. Ender, J. Illig, D. Klotz, E. Ivers-Tiffée, J. Power Sources 196 (2011) 5342-5348 [48] J. Illig, M. Ender, T. Chrobak, J. P. Schmidt, D. Klotz and E. Ivers- Tiffée, J. Electrochem. Soc. 159 (2012) A952-A960

Caption of figures Fig. 1 XRD patterns of the samples before high temperature annealing and after annealing at 350, 400, 500 and 700 ℃in Ar atmosphere. Fig. 2 TGA curves of MnCO3 before high temperature annealing. Fig. 3 SEM images of the samples before annealing (a, b) and after annealing at 350 ℃ (c, d). Fig. 4 SEM images of M400 (a, b), M500 (c, d) and M700 (e, f). Fig. 5 Charging/discharging curves of M400 (a), M500 (b) andM700 (c) and their long-term cyclability curves at 1.0 A g-1 (Initial 5 cycles are tested at 0.2 A g-1 for activation). Fig. 6 Rate cyclability curves of M400, M500 and M700. Fig. 7 SEM images of M400 (a, b), M500 (c, d) and M700 (e, f) after 500 cycles. Fig. 8 Cyclic voltammetry (CV) curves of M400(a), M500(b) and M700 (c)at a scanning rate of 0.1, 0.25, 0.5, 1.0 and 2.0 mV s-1 and their anodic peak current Ip versus square root of sweep rate ν1/2 (d). Fig.9 Nyquist plots of M400, M500 and M700 before electrochemical cycling (a) and after 500 cycles (b).

Fig.1

Fig. 2

Fig.3

Fig.4

Fig. 5

Fig.5

Fig. 6

Fig. 7

Fig. 8

Fig. 9