phosphorus as anode for lithium-ion batteries

phosphorus as anode for lithium-ion batteries

Journal of Power Sources 289 (2015) 100e104 Contents lists available at ScienceDirect Journal of Power Sources journal homepage:

700KB Sizes 8 Downloads 101 Views

Journal of Power Sources 289 (2015) 100e104

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage:

Short communication

Composite of graphite/phosphorus as anode for lithium-ion batteries Aojun Bai a, Li Wang a, b, Jiaoyang Li a, Xiangming He a, c, *, Jixian Wang a, Jianlong Wang a a

Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing, 100084, China State Key Laboratory of New Ceramics & Fine Processing, Beijing, 100084, China c State Key Laboratory of Automotive Safety and Energy, Beijing, 100084, China b

h i g h l i g h t s  Graphite/Phosphorus composite anode shows promising performance.  The composite anode presents high reversible capacity of 500 mAh g1 and good cycleability.  Performance of a compositejLiFePO4 full-cell shows easy SOC evaluation.  This paves a new way for exploring new battery chemistry.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 January 2015 Received in revised form 16 April 2015 Accepted 28 April 2015

Graphite/Phosphorus composite anodes are prepared by mixing graphite and the phosphorus/carbon material, which prepared by heating the mixture of red phosphorus and porous carbon. Their electrochemical performances are evaluated as anodes for Li-ion batteries. A graphite/Phosphorus compositejLiFePO4 full-cell is also attempted. When the phosphorus/carbon content in the composite anode is 28.6 wt.%, the composite anode presents high reversible capacity of 500 mAh g1 and considerable cycleability comparable to that of graphite anode, showing promising performance. © 2015 Elsevier B.V. All rights reserved.

Keywords: Graphite Phosphorus/carbon Composite anode Full cell Lithium ion batteries

1. Introduction Lithium-ion batteries play important roles in the field of energy storage and electric vehicles, and the increasing demand on the energy density encourages exploration on new battery chemistry [1e6]. Therefore, it is a hot topic to investigate novel anode or cathode materials with high energy density [7e10]. Elemental phosphorus is a promising anode candidate due to its high gravimetric and volumetric energy density [11e16]. Cheol-Min Park and Hun-Joon Sohn [17] prepared black-P/carbon black composite using a high energy mechanical milling technique. The composite shows considerably high coulombic efficiency during the first cycle (90%) and a good cycling performance (600 mAh g1) between 0.78 and 2 V vs. Liþ/Li. An amorphous P/C nanocomposite

* Corresponding author. Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing, 100084, China. E-mail address: [email protected] (X. He). 0378-7753/© 2015 Elsevier B.V. All rights reserved.

prepared by ball-milling red phosphorus (red-P) and conductive carbon powders was reported by Jiangfeng Qian et al. [18], and shows high lithium storage capacity near the theoretical (2592 mAh g1) and perfect cycling stability. In detail, its reversible capacity is 2355 mAh g1 at the 2nd discharge, and the capacity retention is 90% after 100 cycles. Weihan Li et al. [19] prepared porous carbon nanofibers encapsulated with crystalline red P by electrospinning of polyacrylonitrile/poly(methyl methacrylate) and thermal carbonization process. The P-PCNFs delivered a high cyclability of 2030 mAh g1 at 0.1 C rate after 100 cycles and outstanding rate capability of 380 mAh g1 at 11C for 10 cycles. Using the sublimation of red-P, our group [20] designed thermal vapour deposition to realize the embedding of red-P nanoparticles into the pores of porous carbon, and the as-prepared composite is named as [email protected] The structure of [email protected] composite is effective to avoid the disadvantages of phosphorus, in terms of poor conductivity, large volume change during lithiation/delithiation, chemical instability in atmosphere and possible safety issue. Furthermore, the composite particle in micro-size is important to fit the state-of-

A. Bai et al. / Journal of Power Sources 289 (2015) 100e104

art for mass production of Li-ion batteries, as well as ensuring considerably high volumetric energy density. However, similar to all the novel electrode material (i.e. Si-based anode material), the application of [email protected] material in a commercial battery still needs more engineering research. On the other hand, though graphite anode has been well studied and widely used in commercial Li-ion batteries, it still encounters many problems. As we all known, the over-charging and overdischarging of a battery is considered to shorten the service life of a power source. Therefore, it is necessary to regulate the charge and discharge potentials. However, graphite anode delivers relatively flat charging and discharging platforms, so the detectors have to be sensitive and accurate when judging battery charging boundary by voltage. Since the lithiation/delithiation potential of [email protected] is slightly higher than that of graphite, the capacity boundary of the battery may be tailored by the composite anode consisting of graphite and [email protected] Though the higher lithiation potential of [email protected] leads to lower working voltage of the battery, the increase in capacity may compensate the energy loss. In this sense, graphite/[email protected] composite anode will facilitate the state of charge management of the battery. In this paper, we explored the preparation of graphite/[email protected] composite anode, as well as its electrochemical performance. The results indicate that the novel composite anode presents high specific capacity and merit for easy SOC (State-Of-Charge) evaluation. In addition, the electrochemical performances of graphite/ [email protected] full-cell are also investigated.


2.3. Electrochemical characterization The charge and discharge performance of the cells was tested with a LAND cycler (Wuhan, China) at room temperature. For the investigation of graphite/[email protected] composite anode, the half-cells were cycled at 0.1 C in the voltages range between 0.005 V and 2 V. For the graphite/[email protected] full-cells, the cells were cycled between 1.5 V and 4.2 V. AC impedance measurements were performed on Zahner Im6ex electrochemical workstation. 3. Results and discussions The [email protected] composite was characterized by scanning electron microscopy (SEM) and powder X-ray diffraction (XRD). The SEM images (Fig. S1a, b) and the carbon (Fig. S1c) and phosphorus (Fig. S1d) elemental mapping images show that phosphorus is homogeneously distributed in the framework of the porous carbon, while a little on the external surface. The XRD patterns (Fig. S2) show that all the typical lines for red-phosphous are disappeared except the line near 15 , indicating that the phosphous in [email protected] material is almost amorphous. On the contrary, the diffraction patterns of carbon remains unchange, implying that the carbon matrix provides mostly physical support for [email protected] material. The results are consistent with our group's work before [20]. Fig. 1 shows charging (delithiation) curves and coulombic efficiencies of graphite and the series of graphite/[email protected] composite

2. Experimental 2.1. Synthesis of the [email protected] material The [email protected] material was prepared by thermal vapour deposition as reported [11]. Activated carbon and excess red phosphorus were placed into a vessel separately, then the vessel was heated to 450  C. The phosphorus sublimated and diffused into the pores by capillary force and pressure difference, whereupon it was adsorbed and deposited on the internal surface of the activated carbon. The phosphorus in the composite was measured to be 43.47 wt.%. 2.2. Preparation of electrodes and batteries A series of graphite/[email protected] mixtures, with [email protected] content of 10 wt.%, 20 wt.%, 28.6 wt.%, 41.8 wt.% and 100 wt.%, were mixed by grinding. Then 70 wt.% graphite/[email protected] mixture, 20 wt.% acetylene black, 10 wt.% polyvinylidene fluoride (PVDF) were blend with N-methyl2-pyrrolidone (NMP) as solvent to form an uniform slurry. The composite anode was prepared by casting the slurry on copper foil, and cut into disk for cell assembling. The loading amount of the anode materials was 6 mg cm2. LiFePO4-based cathode was prepared using the similar procedure. In particular, LiFePO4 content is 85 wt.% in the final dry cathode. The evaluation of both the graphite/[email protected] composite anode and graphite/[email protected] full-cell were performed using CR2025-type coin-type cell. all the electrodes were dried at 120  C under vacuum for 24 h before assembling. CR2025-type coin cells were assembled in a glove box (M. Braun GmbH, Germany) with H2O and O2 content below 1 ppm. Celgard 2400 microporous film was used as separator, and lithium foils were used as counter electrodes to investigate the electrochemical performance of graphite/[email protected] composite anode and LiFePO4-based cathode. The electrolyte used was 1 M LiPF6/EC (ethylene carbonate)/DMC (dimethyl carbonate)/EMC (ethylene methyl carbonate).

Fig. 1. Delithiation curves (a) and coulombic efficiencies (b) of the graphite/[email protected] composite in the half-cell with different [email protected] content.


A. Bai et al. / Journal of Power Sources 289 (2015) 100e104

Fig. 2. (a) Electrolyte holdup and surface resistance of the graphite/[email protected] composite anode with different [email protected] content. The electrolyte holdup and surface resistance take those of graphite as 100%. Nyquist plotting of the half-cells: (b) before the first cycle; (c) after the first cycle; (d) equivalent circuit.

anodes. In Fig. 1a, it can be seen that there are two voltage platforms at 0.1 V and 0.9 V, which represent the delithiation of graphite and phosphorus respectively. Especially, the voltage platform for graphite/[email protected] composite anodes with different [email protected] contents are similar, indicating that the graphite and [email protected] material in the all the composite anode are only physically mixed. With increase of [email protected] content in the graphite/[email protected] composites from 0 wt.% to 28.6 wt.%, the capacity of the graphite/[email protected] composite anode increases gradually. However, when [email protected] content is up to 41.8 wt.%, obvious decrease in capacity can be observed not only for the graphite but also for the [email protected] material. Their initial coulombic efficiencies also show the similar results (Fig. 1b). In detail, the coulombic efficiency is approximately 70% and over 94% at the 2nd cycle and afterwards, for the graphite/[email protected] composite anode with [email protected] material of less than 28.6 wt.%. It should be mentioned that pristine graphite shows initial coulombic efficiency of only 71% and reach 94% at the 2nd cycle. Though this data may be not satisfying compared with many reports, based on the fact that all the materials are evaluated with the same process, the data is still reasonable as a reference. In this sense, when [email protected] material in graphite/[email protected] composite anode is less than 41.8 wt.%, comparable with the graphite, the irreversible capacity led by [email protected] material is neglectable. When [email protected] content reaches 41.8 wt.%, the remarkable irreversible capacity, relating with phosphide dissolution, volume change and absorbed water led by

[email protected] material, is considerable and the cycle stability becomes worse. The above observations can be explained by the further analysis on electrode in terms of electrolyte holdup, surface resistance and AC impedance. Fig. 2a indicates the electrolyte holding ability (taking that of graphite as 100%) and electronic conductivity (taking that of graphite as 100%) of the electrodes with different compositions. With [email protected] content increasing, the electrolyte holdup decreases, while the surface resistance increases. On one hand, electrolyte holdup is generally related with porosity of the electrode and wettability of the electrode material surface to electrolyte. According to the large average size (about 20 mm, Fig. S3a) of the [email protected] material, which also could be seen in the anode (Fig. S3b), higher [email protected] content means lower porosity of the electrode, so the porosity may be responsible more for the decrease in electrolyte holdup. On the other hand, graphite behaves very good electronic conductivity due to perfect lamellar conductive structure, while the conductivity of [email protected] material is offered by the porous activated carbon matrix and red-P is an insulator. The surface resistance increases with [email protected] content going up implies that too much addition of [email protected] material may deteriorate the rate performance of the composite electrode. Fig. 2 also shows the -Nyquist plotting of electrochemical impedance spectroscopy (EIS) of Lijgraphite/[email protected] composite halfcell. Fig. 2b and c shows the curves before and after the 1st cycle

A. Bai et al. / Journal of Power Sources 289 (2015) 100e104


Fig. 3. The electrochemical performance of graphite/[email protected] composite, with [email protected] content of 28.6 wt.%. (a) Charge/discharge profiles of the initial cycles. (b) CV curve of the first cycle; (c) Variations of coulombic efficiency and charge/discharge capacity with cycles.

respectively. Fig. 2d shows the equivalent circuit. The semicircle at high frequency can be attributed to the interface impedance between active materials and electrolyte, as well as the impedance between electrode materials and collector. It can be observed that the radii of the semicircles rise along with the growth in [email protected] content. After the 1st cycle, the all impedances decrease. Especially, the impedances for electrode with [email protected] content lower than 41.8 wt.% are much similar. In Fig. 2d, the equivalent circuit is composed of the solution resistance Re, interfacial capacitance Csf (simulated with a constant phase element (CPE) due to the dispersion effect of actual electrode), interface impedance Rsf and Warburg impedance Zw. The simulation results show that the Rsf for all the samples fall in the range from 284 U cm2 to 1790 U cm2 before first cycle, and from 65 U cm2 to 173 U cm2 after first cycle (Table S1 shows other Rsf). Higher [email protected] content leads to larger Rsf. On the basis of above results, it can be deduced that as [email protected] material is poor in electronic conductivity and porosity when compared with graphite, an optimized content needs to be find to balance the capacity, rate capability, coulombic efficiency and cycleablility. And graphite/[email protected] composite anode with [email protected] content of 28.6 wt.% seems to be the best among them, according to its high capacity and stability. Fig. 3a shows charge(delithiation)/discharge(lithiation) curves between 0.005 V and 3 V of the anode electrode, which electrode material is graphite/[email protected] composite with [email protected] content of 28.6 wt.%. For the first discharge, the electrode shows lithium storage capacity of 740 mAh g1, and the initial coulombic efficiency is 71.6%. During the second round, the charge and discharge capacities are 530 mAh g1 and 560 mAh g1, respectively. The first charge experiences two platforms, one lies between 0.005 V and 0.25 V lasting 230 mAh g1, the other lies between 0.8 V and 1.2 V. These two platforms can be attributed to the delithiation of graphite and

phosphorus, respectively. Fig. 3a also shows charge/discharge curves of a graphite electrode. During the first discharge, the electrode shows capacity of 505 mAh g1, while the initial charge capacity is 360 mAh g1. Then the initial coulombic efficiency is 71%, and the platform located between 0.005 V and 0.25 V delivers capacity of 310 mAh g1 in the first charge. For the composite electrode material, the graphite part is 71.4 wt.% and will contribute 221 mAh g1 accordingly. The initial coulombic efficiency of [email protected] composite is about 70%. The capacity of phosphorus can also be calculated as 2415 mAh g1 with the first charges of the two electrodes. A conclusion then can be drown that mixing of graphite and [email protected] material does not change their charge/discharge properties in terms of platforms potential, capacity and initial coulombic efficiency. And the cyclic voltammetry (CV) curve of the first cycle shown in Fig. 3b further confirms the same conclusion. In addition, the reaction of formation of SEI also contributes to the initial irreversible coulombic efficiency. The electrode with 28.6 wt.% [email protected] content also shows good cycling performance, as shown in Fig. 3c. It can be observed that the charge capacity at the 2nd cycle is 530 mAh g1, and is 485 mAh g1 at the 51st cycle, indicating the capacity retention of 91.5%. The reduced capacity is mainly from the [email protected] material. In detail, the capacity calculated based on phosphorus changes from 2415 mAh g1 to 2100 mAh g1 from the 2nd cycle to the 50th cycle. Fig. 3c shows a stable charge capacity, while the coulombic efficiency keeps going up until 30 cycles due of the decreasing discharge capacity. It may be related to the continuous formation of SEI film on the composite surface. The increase in Rsf with cycling, as shown in Fig. S4, is coincide with this observation. Fig. 4 shows the charge/discharge profile of the graphite/ [email protected] full-cell, as well as the calculated profile on the basis of LijLiFePO4 and Lijgraphite/[email protected] half-cells. The red lines describe the


A. Bai et al. / Journal of Power Sources 289 (2015) 100e104

Acknowledgements This work is supported by NFSC (No. 20901046), MOST (Grant No. 2011CB935902), Beijing Municipal Program (Grant No. YETP0157).

Appendix A. Supplementary data Supplementary data related to this article can be found at http://


Fig. 4. State of charge profile of graphite/[email protected] full-cell.

charge/discharge profiles of LiFePO4-based cathode. The black lines belong to the composite anode with 28.6 wt.% [email protected] material. The green lines represent the calculated curves of graphite/[email protected] cell. The blue lines are experimental charge/discharge curves of graphite/ [email protected] cell. It can be observed that the experimental result is matched well with the calculated one. The graphite/[email protected] composite anode provides obvious voltage stage for easy evaluation of state of charge by voltage, compared with flat-platform cathode materials. 4. Conclusion [email protected] material is pushed forward to practical application for Liion batteries by the composite of graphite/[email protected] as anode electrode in this study. The results show that the capacity, potential, columbic efficiency and cycling stability can be reasonably improved by tuning the content of the graphite/[email protected] composite anode. In particular, when the content of [email protected] material is 28.6 wt.%, the composite anode presents considerably high reversible capacity (500 mAh g1) and cycleability when comparing with graphite anode. However, too much [email protected] material will lead to decrease in electrolyte uptake, electronic conductivity, as well as poor electrochemical performances of the composite anode. Furthermore, the attempt on graphite/[email protected] full-cell is also successful. The full-cell provides obvious potential stages for easy SOC evaluation. This paves a new way for exploring new battery chemistry.

[1] J.M. Tarascon, M. Armand, Nature 414 (2001) 359e367. k, Adv. Mater. [2] Martin Winter, Jürgen O. Besenhard, Michael E. Spahr, Petr Nova 10 (1998) 725e763. [3] B. Scrosati, J. Garche, J. Power Sources 195 (2010) 2419e2430. [4] Zhe Li, Jun Huang, Bor Yann Liaw, Viktor Metzler, Jianbo Zhang, J. Power Sources 254 (2014) 168e182. [5] John B. Goodenough, Youngsik Kim, Chem. Mater. 22 (2010) 587e603. [6] R. Marom, S.F. Amalraj, N. Leifer, D. Jacob, D. Aurbach, J. Mater. Chem. 21 (2011) 9938e9954. [7] Hailiang Wang, Li-Feng Cui, Yuan Yang, Hernan Sanchez Casalongue, Joshua Tucker Robinson, Yongye Liang, Yi Cui, Hongjie Dai, J. Am. Chem. Soc. 132 (2010) 13978e13980. [8] Jing Bai, Xiaogang Li, Guangzeng Liu, Yitai Qian, Shenglin Xiong, Adv. Funct. Mater. 24 (2014) 3012e3020. [9] Jun Liu, Hui Xia, Dongfeng Xue, Li Lu, J. Am. Chem. Soc. 131 (2009) 12086e12087. [10] Jian Lin, Abdul-Rahman O. Raji, Kewang Nan, Zhiwei Peng, Zheng Yan, Errol L.G. Samuel, Douglas Natelson, James M. Tour, Adv. Funct. Mater. 24 (2014) 2044e2048. [11] L. Wang, X.M. He, J.G. Ren, W.H. Pu, J.J. Li, J. Gao, United States Patent Application Publication, US20100239905, 2010. [12] C. Marino, A. Debenedetti, B. Fraisse, F. Favier, L. Monconduit, Electrochem. Commun. 13 (2011) 346e349. [13] R. Yazami, Electrochim Acta 45 (1999) 87e97. [14] Li-Qun Sun, Ming-Juan Li, Kai Sun, Shi-Hua Yu, Rong-Shun Wang, HaiMing Xie, J. Phys. Chem. C 116 (2012) 14772e14779. [15] Youngjin Kim, Yuwon Park, Aram Choi, Nam-Soon Choi, Jeongsoo Kim, Junesoo Lee, Ji Heon Ryu, Seung M. Oh, Kyu Tae Lee, Adv. Mater. 25 (2013) 3045e3049. [16] Marian Cristian Stan, Jan von Zamory, Stefano Passerini, Tom Nilges, Martin Winter, J. Mater. Chem. A 1 (2013) 5293e5300. [17] Cheol-Min Park, Hun-Joon Sohn, Adv. Mater. 19 (2007) 2465e2468. [18] Jiangfeng Qian, Dan Qiao, Xinping Ai, Yuliang Cao, Hanxi Yang, Chem. Comm. 71 (2012) 8931e8933. [19] Weihan Li, Zhenzhong Yang, Yu Jiang, Zirui Yu, Lin Gu, Yan Yu, Carbon 78 (2014) 455e462. [20] Li Wang, Xiangming He, Jianjun Li, Wenting Sun, Jian Gao, Jianwei Guo, Changyin Jiang, Angew. Chem. Int. Ed. 51 (2012) 9034e9037.