PERGAMON Carbon38(2000)183–197
Recent development of carbon materials for Li ion batteries
*
M.Endo,C.Kim,K.Nishimura,T.Fujino,K.Miyashita
Faculty of Engineering,Shinshu University,500Wakasato,Nagano380-8553,Japan
Received19April1999;accepted24June1999
Abstract
Lithium ion secondary batteries are currently the best portable energy storage device for the consumer electronics market. The recent development of the lithium ion secondary batteries has been achieved by the use of selected carbon and graphite materials as an anode.The performance of lithium ion secondary batteries,such as the charge/discharge capacity,voltage pro?le and cyclic stability,depend strongly on the microstructure of the anode materials made of carbon and graphite.Due to the contribution of the carbon materials used in the anode in last?ve years,the capacity of the typical Li ion battery has been improved1.7times.However,there are still active investigations to identify the key parameters of carbons that provide the improved anode properties,as carbon and graphite materials have large varieties in the microstructure,texture,crystallinity and morphology,depending on their preparation processes and precursor materials,as well as various forms such as powder,?bers and spherule.In the present article,we describe the correlation between the microstructural parameters and electrochemical properties of conventional and novel types of carbon materials for Li ion batteries,namely,graphitizable carbons such as milled mesophase pitch-based carbon?bers,polyparaphenylene-based carbon heat-treated at low temperatures and boron-doped graphitized materials,by connecting with the market demand and the trends in Li ion secondary batteries.The basic scienti?c theory can contribute to further developments of the Li ion batteries such as polymer batteries for consumer electronics,multimedia technology and future hybrid and electric vehicles.?2000Published by Elsevier Science Ltd.All rights reserved.
Keywords:A.Intercalation compounds;D.Electrochemical properties
1.Introduction[2–5].The intercalation compound of lithium metal into
graphite by vapor transport was?rst synthesized by Herold Among the metals,lithium has great promise as an[6]as a graphite intercalation compound(GIC)with stage electrode material of batteries that can combine the lightest structure.Since then,extensive study has been performed weight with high voltage and high energy density.Because to investigate the staging structure and charge transfer lithium possesses the lowest electronegativity of the stan-phenomena of the Li-GIC compounds,which have the dard cell potential23.045V in the existing metals,it is composition of Li C,where0 x6 the anode material that donates electrons the most easily to pressure,into order and disordered host materials[7–12]. form positive ions[1].However,the negative electrode of In the rechargeable lithium ion batteries based on the lithium metal has serious problems as secondary battery‘‘rocking chair’’or‘‘shuttle cock’’concepts,the lithium use,since it does not have a long enough cyclic life and ions intercalates have to shift back and forth easily there are safety aspects that need to be considered due to between the intercalation hosts of the cathode and anode. the dendrite formation on the surface of lithium metal Thus,the lithium ion secondary battery mainly consists of electrode during charge/discharge cycles.In order to solve a carbonaceous anode and a lithium transition metal oxide these problems a‘‘locking-chair’’concept has been estab-such as LiCoO,LiNiO and LiMn O as the cathode,as 2224 lished,in which the intercalation phenomena has been used demonstrated in Fig.1a.The anode on Cu foil and the as an anode reaction for lithium ion secondary batteries cathode on Al foil are formed into spiral or plate folded shapes that give the US18650cylindrical type(18mm f and650mm high,Fig.1b)and rectangular cells,respec-*Corresponding author.Tel.:181-26-2269-5201;fax:181- tively.Between these two electrodes is placed a porous 26-223-7754. E-mail address:endo@endomoribu.shinshu-u.ac.jp(M.Endo)polymer separator of polyole?n of about25m m thickness, 0008-6223/00/$–see front matter?2000Published by Elsevier Science Ltd.All rights reserved. PII:S0008-6223(99)00141-4 186M.Endo et al./Carbon38(2000)183–197 battery have been supporting the recent large developments corresponding to the apparatus life.The battery enable to in cellular phones,personal computers with color liquid be settled in PC(personal computer)and cellular phone crystal device(LCD)and high speed CPU as shown in will make us forget the battery charging.Li ion polymer Fig.4a and b[27,28].For example,95g weight of cellular type battery,described later and which will become the phones with46mm width,22mm length and11mm main type in these market?elds,has the possibility to thickness rectangular Li cell can operate during6h for talk challenging for such a request as well as for future full EV and600h for stand-by.application. Fig.5is the recent transition in production amount of small size secondary batteries in Japanese market.Espe- cially,the Li ion battery produced in Japan have about 99%of the world market,and the yearly production has 3.Voltage pro?les of carbon electrodes reached2.0billion dollar in1997which is about twice those for NiMH and NiCd’s.Because of the strong demand In the electrochemical cell used in the present article,the from consumer electronics market and also electric ve-electrodes of carbon materials are positive electrodes since hicle,Li ion battery technologies have been required to the counter electrode is lithium metal,therefore lithium achieve further development in energy density,output intercalation to carbon corresponds to the discharge pro-current,safety and cost.It has been expected from the PC cess,whereas the deintercalation of lithium ions is a charge market that the10h of operation and1000cycle life process. Fig.4.(a)Number of cells sold in the world market in1995and1996,and(b)the use of Li ion battery(data from Battery Association of Japan(1997)). M.Endo et al./Carbon38(2000)183–197187 Fig.5.The transition of production amount(in dollars)of the various types of secondary batteries(data from International Trade and Industry Statistics of MITI,1997). Fig.6shows the voltage pro?les for lithium/carbon electrochemical cells made from representative carbon and graphite materials[21].The graphite electrode cell gives a reversible capacity of280|330mAh/g,and lithium discharge/charge plateau at below0.2V was reproduced [9,23].In the?rst cycle,all types of carbon materials show the irreversible capacity at about0.8V due to electrolyte decomposition and formation of solid electrolyte inter- phase[29].Then,after2nd cycles the irreversible capacity is much reduced,and the electrode exhibits stable cyclic properties.Among many types of carbon electrodes,well- ordered graphites are currently becoming one of the representatives for the industrial standard because of its long plateau in voltage pro?le and for low electrode potential relative to lithium metal.However,a major disadvantage of the graphite system has the limited lithium storage capacity as about310mAh/g in current commer- cial cell,which is much less than372mAh/g corre- sponding to LiC.On the other hand,soft carbons heat- 6 treated at temperature(500|10008C)give a reversible capacity of near700mAh/g,and indicate characteristic plateau in discharge/charge properties at about1.0V,as well as hysteresis in voltage pro?le about0.0V[21,22,30]. Furthermore,hard carbons heat-treated at temperatures around11008C give a reversible capacity of600mAh/g, but have the small irreversible capacity and hysteresis Fig.6.Plots of voltage vs.reversible capacity for(a)the second discharge and(b)charge cycle of representative carbon and graphite samples,graphitizable carbon heat-treated at30008C, Fig.7.Carbon materials used for commercial Li ion secondary graphitizable carbon heat-treated at20008C,non-graphitiz- batteries used in the market in1996. able carbon heated at7008C. 190M.Endo et al./Carbon38(2000)183–197 Fig.10.(continued) 5.Effect of morphologies,carbon?bers for Li ion other hand,Endo[16]and Tatsumi[37]reported the battery battery characteristics of VGCF with wooden annual ring structure,and showed the interesting potential application Carbon?bers used for electrodes in lithium ion batteries as anode.Fig.12a and b[38]demonstrates the SEM are roughly classi?ed into two types such as milled pictures of the thin VGCF obtained by the?oating catalysis mesophase pitch-based carbon?bers and gas phase grown process and the graphite anode mixed with the?bers.The carbon?ber commonly called as vapor grown carbon conductivity and cyclic life has been improved and this is ?bers(VGCFs)[35].The former is contributing as one of very promising as one of the important technology of the the practical and promising anode material with high battery. density of electrode,larger discharge capacity and better The Li charge capacity and cyclic ef?ciency depends output current performances.The latter can serve as strongly on the cross sectional structure of the?bers,such conductive?ller in anode and cathode electrodes.On the as onion,radial and random structures[36,37].In par- M.Endo et al./Carbon38(2000)183–197191 ticular,mesophase pitch-based carbon?bers(MPCFs) exhibit a high degree of anisotropy with regard to me- chanical,electrical,magnetic,thermal as well as chemical properties[19,39].These anisotropies are directly related to the layered structure with strong interlayer interactions and very weak van der Waals interplanar interactions between adjacent graphene sheets aligned parallel to the ?ber axis[40].The novel chemical and physical properties of the mesophase pitch-based carbon?bers have also been Fig.11.The voltage dependences for the high frequency Raman modes of the PPP-700electrode obtained by?ts to Lorentzian line shape. Fig.12.(a)SEM photographs of the thin VGCF obtained by the Fig.13.FE–SEM photographs of milled MPCFs with HTT(a)?oating catalysis process[35,38]and(b)the graphite anode mixed10008C,(b)30008C,and(c)high modulus mesophase pitch-based with the?bers in a commercial cell.graphite?ber(P-100). M.Endo et al./Carbon38(2000)183–197193 tion of CVD pyrolysis of organic molecules containing boron atoms and the substitutional boron doping mecha- nism into carbon structures. Fig.16illustrates the typical voltage pro?les of the second discharge and charging for pristine graphitized and boron-doped electrodes.These samples were prepared from a mixture of pristine material and boron carbide (B C)by heat treatment at28008C in Ar atmosphere.The 4 wide plateaus below0.2V correspond to the reversible intercalation of Li in graphitized and boron-doped samples. These electrochemical behaviors of the graphitized sam- ples are almost the same as that of other highly graphitized electrodes[46,51].It should be noted that the2nd dis- charge-charge capacity of boron-doped graphite I(petro- leum coke)and graphite II(carbon spheres)slightly decreased relative to those of the pristine samples.How- ever,in the case of boron-doped MPCFs,the2nd charge capacity is larger than that of undoped pristine MPCFs. The reduced charge capacity of the boron-doped samples may be related to boron atoms occupying the lithium Fig.15.The relationship linking the charge capacity of MCMBs during the?rst cycle and the volume ratio of the structures:(a)the charge capacity in the potential range of0–0.25V vs.P plots;(b) 1 the charge capacity in the potential range of0.25–1.3V vs. (12P)plots,referred from Tatsumi et al.[46]. 1 pounds consist of layered structure[37,47,48].In par- ticular,boron-doped carbon materials have been ex- perimentally and theoretically investigated from different points of view,not only from fundamental scienti?c aspects,such as electronic properties,but also towards potential applications,such as the high temperature oxida- tion protector for carbon/carbon(C/C)composite and an anode material for Li ion batteries.Because boron-doping is inducing the creation of electron acceptor level[37,47– 50],the enhanced capacity has been expected.Many Fig.16.Change in potential during the second discharging and researchers[37,47–50]have reported and suggested pre-charging cycle of the graphitized and boron-doped samples for parative methods for boron-doped carbons by co-deposi-various types of graphite hosts. 194M .Endo et al ./Carbon 38(2000)183–197insertion active site,such as an edge-type site in the carbon layers and thus the presence of the boron will inhibit the lithium insertion process.Also,the discharge capacities of graphite I and graphite II are larger than that of MPCFs, because lithium insertion is easily accomplished in the wholly exposed edge surface of these carbons,relative to MPCFs.It is worthwhile to note that the voltage pro?les of boron-doped samples are higher than those of the undoped samples at about 40mV [36],and this could be useful for practical cell applications.On the discharging cycle for the boron-doped samples,shoulder plateau are characteristical- ly observed at about 1.3V ,which may be caused by inducing an electron acceptor level so that lithium insertion yields a higher voltage compared to undoped samples [36,47].The irreversible capacity is calculated as the average ratio of the capacities for the discharge and charge process.It is interesting that the irreversible capacity loss for boron-doped samples is lower than that of the corre- sponding undoped samples.These results may be related to the redistribution of the Fermi level of the boron-doped samples,which is lowered by boron-doping by introducing an electron acceptor in the lattice [36,47]. Fig.17shows the carbon 1s peak at high resolution XPS spectra.It is interesting that the C peak of the boron-1s doped samples is located at slightly lower binding energy compared with the undoped samples.The lowering of the binding energy of the C peak for the boron-doped 1s samples might be due to the lowering of the Fermi level, because of the redistributed p -electrons in the graphite layer planes [52,53].This result may be related to lowering the density of p electrons in the graphite layers,because of the chemical bond formation of carbon atoms with the Fig.17.XPS C spectra of graphitized and boron-doped samples 1s electron de?cient boron atom.Therefore,the C peak of 1s for various types of graphite hosts.boron-doped samples moves to lowering the binding energies. Figs.18and 19illustrate the boron and nitrogen 1s peak in the XPS spectrum.As expected,the B peak appears in 1s the boron-doped samples,but their forms and positions are different,depending on the samples.In particular,the B 1s peak of the boron-doped graphite I was split into three peaks at 185.6,187.7,and 189.8eV ,which were assigned to boron in boron carbide,in a boron cluster or boron bound to incorporated nitrogen atoms,respectively.And also,in Fig.19the N peak is observed only in boron-1s doped Graphite I.From these results,the appearance of the B near |190eV and N near |398eV peaks of boron-1s 1s doped Graphite I is corresponding to the substitutionally incorporated boron atoms into the graphite lattice which are preferentially bonding with nitrogen atoms existing in the heat treatment atmosphere.It is also possible that the residual nitrogen atoms in raw materials react with the substitutionally incorporated boron atoms into the graphite lattice during the carbonization step and then formed the boron nitride and/or BC N compounds during the graphiti-x zation step.Konno et al.[54]demonstrated the B–N bonding in boron doped graphites by XPS and suggested Fig.18.XPS B spectra of boron-doped samples. 1s M.Endo et al./Carbon38(2000)183–197195 dimension,L,due to the borons acting as a graphitization a catalyst.On the other hand,boron-doping in MPCFs can induce the homogeneous crystallite ordering,approaching to the ideal graphite structure,as indicated by the XRD, XPS,and Raman results[55].Fig.20shows a proposed schematic model of a graphite plane in which the substitu- tional boron atoms show different bond lengths between the B–C and C–C bonds.Data from a molecular simula- tion are shown in the?gure[49].In order to better employ the effects of boron doping,depending on the carbon materials obtained from a different precursor with a wide variety of shapes and microstructure,the doping conditions including the atmosphere and heating rate should be carefully selected.This could provide one of the new types of material design of carbon or graphite electrodes. 7.Polymer battery Fig.19.XPS N spectra of boron-doped samples. 1s The name of‘‘polymer battery’’is de?ned like the battery which has a polymer electrode and/or electrolyte. Very recently,a Li ion battery with gel type polymer the possible nitrogen source as air occluded in raw electrolyte has been commercialized which is only3.6mm materials that are packed into a graphite crucible.These thick,weighs15g and has a capacity of500mAh,as phenomena should be taken into consideration seriously shown in the Table1[56].The gel electrolyte consists of for industrial process using an Acheson type furnace.poly(ethylene oxide)(PEO),poly(acrylonitrile)(PAN),and As mentioned above,the degradation of the lithium poly(vinylidene?uoride)(PVdF)as the host polymer,EC/ insertion capacity observed in some kinds of boron-doped PC as the plasticizer and LiPF as the electrolyte salt.This 6 graphite might be highly related to the presence of borons?lm or card shaped battery has energy density in volume in forms of boron nitride and boron carbide.Also the and weight of250Wh/l and120Wh/kg,respectively, unexpected opposite effects of boron doping could be which is almost equivalent with those of the conventional related to the heterogeneous growth of the crystallites cylindrical cell.No inclusion of liquid electrolyte increases the safety and can contribute to reducing the cell thickness and weight.The same types of graphitic anode materials as in the conventional cylindrical cell have been used. Formation of the anode?lm with suitable density and porosity to permit a high degree of penetration of gel in the electrodes and enough conductivity should be achieved. These polymer batteries could reduce the thickness of cellular phones and portable PCs(see Table1,Fig.21a and b,and Ref.[56]). Table1 Speci?cations of a commercialized Li polymer battery[56] Production Lithium polymer battery Model No.SSP356236 Shape Prismatic Size35(W)362(L)33.6(T)mm Voltage 3.7V Average electric capacity500mAh(charged at4.2V) Weight Approximately15.0g Fig.20.Schematic representation of a graphene plane of the charge–discharge cycles500cycles or more boron doped MPCFs.The difference in bond length between a Working temperature range210|1608C B–C and C–C bond is indicated[49]. M.Endo et al./Carbon38(2000)183–197197 [31]Dahn JR,Sleigh AK,Shi H,Way BM,Weycanz WJ,Reimers References JN,Zhong Q,von Sacken U.In:Pistoia G,editor,Lithium batteries,new materials and perspectives,Amsterdam: [1]Gabano JP.In:Lithium batteries,New York:Academic Elsevier,1993,p.1. Press,1983. [32]Yoshino A.Tanso1999;186:45–9,In Japanese. [2]Armand MB.In:Van Gool W,editor,Fast ion transport in [33]Endo M,Kim C,Karaki T,Fujino T,Matthews MJ,Brown solids,Elsevier,1973,p.665. SDM,Dresselhaus MS.Synthetic Metals1998;98:17–24. [3]Murphy DW,Carides JN.J Electrochem Soc1979;126:349. [34]Inaba M,Yoshida H,Ogumi Z.J Electrochem Soc [4]Lazzari M,Scrosati B.J Electrochem Soc1980;127:773. 1996;143:2572–8. [5]Auborn JJ,Barberio YL.J Electrochem Soc1987;134:638. [35]Endo M.Chemtech1988;18:568–76. [6]Herold A.Bull Soc Chim France1955;187:999. [36]Nishimura Y,Yakahashi T,Tamaki T,Endo M,Dresselhaus [7]Juza R,Wehle V.Nature1965;52:560. MS.Tanso1996;172:89–94,In Japanese. [8]Guerard D,Herold A,Acad CR.Sci Ser1972;C275:571. [37]Tatsumi K,Zaghib K,Sawada Y,Abe H,Ohsaki T.J [9]Dahn JR,Fong R,Spoon JJ.Phy Rev1990;B42:6424–32. Electrochem Soc1995;142:1090–6. [10]Dresselhaus MS,Dresselhaus G.Adv in Phys1981;30:139. [38]Showa denko’s catalog,Fine Carbon,V.G.C.F.,1997. [11]Semenenko KN,Avdeev VV,Mordkovich VZ,Doklady AN. [39]Matthews MJ,Dresselhaus MS,Dresselhaus G,Endo M, SSSR1983;271:1402,In Russian. Nishimura Y,Hiraoka T,Tamaki T.Appl Phys Lett [12]Avdeev VV,Nalimova V A,Semenenko KN.High Pressure 1996;69:430. Research1990;6:11. [40]Endo M,Kim C,Karaki T,Kasai T,Matthaus MJ,Brown [13]Sony’s catalog,Lithium ion rechargeable battery,ACG- SDM,Dresselhaus MS,Tamaki T,Nishimura Y.Carbon 4012-N-9707-P3-002,1997. 1998;36:1633–41. [14]Endo M,Karaki T,Fujino T.New ceramics1998;4:46–52, [41]Tamaki N,Satoh A,Hara M,Ohsaki T.J Electrochem Soc (In Japanese). 1995;142:2564–71. [15]Yata S,Kinoshita H,Komori M,Ando N,Anekawa A, [42]Satoh A,Tamaki N,Ohsaki T.Solid State Ionics Hashimoto T.In:Extended abstract of60th annual meeting 1995;80:291. of the electrochemical society of Japan,Tokyo,2G09,1993. [43]Ohsaki T,Kanda M,Aoki Y,Shiroki H,Suzuki S.J Power [16]Endo M,Nishimura Y,Takahashi T,Takeuchi K,Dreselhaus Sources1997;68:102. MS.J Phys Chem Solids1996;57:725–8. [44]Endo M,Nakayama J,Sasabe Y,Takahashi T,Inagaki M. [17]Oberlin A.In:Thrower PA,editor,Chemistry and physics of Tanso1994;165:282–7,In Japanese. carbon,Vol.22,New York:Marcel Dekker,Inc,1989,p.1. [45]Fujimoto H,Mabuchi A,Tokumitsu K,Kasuh T,Akuzawa [18]Imanishi N,Kashiwagi H,Ichikawa T,Takeda Y,Yamamoto N.Carbon1994;32:193–8. O,Inagaki M.J Electrochem Soc1993;140:315–20. [46]Tasumi K,Iwashita N,Sakaebe H,Shioyama H,Higuchi S, [19]Dresselhasu MS,Dresselhaus G,Sugihara K,Spain IL, Mabuchi A,Fujimoto H.J Electrochem Soc1995;142:716–Goldberg HA.In:Cardona Manuel,editor,Graphite?bers 20. and?laments,New York:Springer-Verlag,1988. [47]Way BM,Dahn JR.J Electrochem Soc1994;141:907–12. [20]Sato K,Noguchi M,Demachi A,Oki N,Endo M.Science [48]Nakajima T,Koh K,Takashima M.Electrochimica Acta 1994;264:556–8. 1998;43:883–91. [21]Dahn JR,Zheng T,Liu Y,Xue JS.Science1995;270:590–3. [49]Endo M,Kim C,Karaki T,Tamaki T,Nishimura Y, [22]Mabuchi A,Tokumitsu K,Fujimoto H,Kasuh T.J Electro- Matthews MJ,Brown SDM,Dresselhaus MS.Phy Rev chem Soc1995;142:1041–6. 1998;B58:8991–6. [23]Dahn JR.Phys Rev1991;B44:9170–7. [50]Hagio T,Nakamizo M,Kobayashi K.Carbon1989;27:259–[24]Flandrois S,Simon B.Carbon1999;37:165–80. 63. [25]Funabiki A,Inaba M,Ogumi Z,Yuasa S,Otsuji J,Tasaka A. J Electrochem Soc1998;145:172–8.[51]Koh K,Nakajima T.Carbon1998;36:913–20. [26]Nagaura T,Tozawa K.Prog Batt Solar Cells1990;9:209.[52]Henning G.J Chem Phy1965;42:1167–72. [27]CMC,Development of Li-ion Rechargeable Battery Materi-[53]Jones LE,Thrower PA.Carbon1991;29:251–69. als,Tokyo,CMC,1998,p.149(in Japanese).[54]Konno H,Nakahashi T,Inagaki M,Sogabe T.Carbon [28]Data from Battery Association of Japan,1997,1998.1999;37:471–5. [29]Fong F,Sacken Kvon,Dahn JR.J Electrochem Soc[55]Endo M,Kim C,Fujino T,Dresselhaus MS,to be submitted. 1990;137:2009–13.[56]National/Panasonic catalog,1999(courtesy of Matsushita [30]Buiel E,George AE,Dahn JR.J Electrochem Soc Battery Industrial Co.,Ltd.). 1998;145:2252–7.