Delivered by Publishing Technology to: University of New South Wales IP: 149.171.232.34 On: Wed, 27 Feb 2013 03:01:32Copyright American Scientific Publishers RESEARCH ARTICLECopyright©2013American Scientific Publishers All rights reservedPrinted in the United States of AmericaJournal of Nanoscience and Nanotechnology V ol.13,1535–1538,2013Research on High Rate CapabilitiesB-Substituted LiFePO4Fu Wang,Yun Zhang∗,and Chao ChenCollege of Materials Science and Engineering,Sichuan University,Chengdu610064,P.R.China LiFePO4is currently recognized as one of the most promising electrode materials for large-scale application of lithium ion batteries.However,the limitation of rate capability is believed to be intrinsic to this family of compounds due to the existence of larger tetrahedral(PO4 3−unit and quasi-hexagonal close-packed oxygen array.This paper report here a systematic investigation of the enhancement of rate performance by partly substitution of light small triangle oxyanion,(BO3 3−, for the larger tetrahedral(PO4 3−units in LiFePO4.Cathode electrode materials LiFeB X P 1−X O4− , in which X=0 3 6and9,mol%,were synthesized by solid-state method.The as-synthesized products were characterized by X-Ray Diffraction(XRD),Scanning Electron Microscope(SEM)and Electrochemical Measurements.The results showed that6mol%of boron substitution had no effect on the structure of LiFePO4material,but significantly improved its rate performance.The initial discharge capacity of the LiFeB0 06P0 94O4− sample was145.62mAh/g at0.1C,and the capacity retention ratios of81%at2C and76%at5C were obtained,demonstrating that a proper amount of boron substitution(lower than6mol%)could significantly improve the rate performance of LiFePO4 cathode material.Keywords:LiFePO4,High Rate Capability,Li-Ion Battery,Nano-Particles,Boron.1.INTRODUCTIONLiFePO4has recently received a great deal of attentionowing to their advantages of competitive high theoreti-cal capacity,good cycle stability,excellent thermal stabil-ity and low toxicity,1–3aimed at utilizing it as a cathodematerial for large-scale application of lithium ion batter-ies,such as electric vehicle and hybrid electric vehicle.Moreover,its voltage,about3.5V versus lithium,is com-patible with the window of a solid-polymer Li-ion elec-trolyte.However,this kind of compound is a wide-gapsemiconductor(3.7eV)and has an inherently extremelylow electronic conductivity(∼10−9S cm−1 at room tem-perature because of the existence of larger tetrahedral(PO4 3−units and quasi-hexagonal close-packed oxygen array.1Various material processing approaches have beenadopted to overcome this drawback,including methods ofcarbon coating,4reducing particle size to nano level,5 6and doping with super valence cations.7The aforemen-tioned methods for improving electronic conductivity andrate capability are not the most optimistic choice and havetheir intrinsic limitations:the shortcomings of carbon coat-ing including the lower content of active materials in thecathode material and no actual improvement in conductiv-ity for the core of LiFePO4particles.The preparation of ∗Author to whom correspondence should be addressed.nano-sized particles with a uniform size distribution areextremely difficult for industrial scale production.And thequantity of Fe3+/Fe2+redox couples is reduced by supervalence cations substitution.LiFeBO3,as a new potential cathode material with atheoretical capacity of220mAh/g which is much largerthan that of LiFePO4,has been reported to have the actualspecific capacity of over190mAh/g at1/20C.8In addi-tion,from the thermodynamic study performed in the caseof LiFeBO3,the Fe3+/Fe2+reduction couple lies between3.1V and2.9V(vs.Li/Li+ ,demonstrating an impor-tant inductive effect of BO3group,and the electrical con-ductivity of LiFeBO3is reported to be1 5×10−4S/cm,9which is also much higher than that of LiFePO4.10Thus,it is believed that partly replacing the tetrahedral anionunits,(PO4 3−,to plane triangle oxyanion,(BO3 3−,could be significantly increasing the electronic conductivity of the LiFePO4particles because of the smaller and lighter (BO3 3−and the controlled off-stoichiometry of oxygen element formed.In this regard,we proposed a new method,partly sub-stitution of boron element for phosphorus element inLiFePO4,for improving the rate capability of the cathodematerial.We report here a systematic investigation of theenhancement of capacity at high rates of charge and dis-charge by partly substitution of light small plane triangleJ.Nanosci.Nanotechnol.2013,Vol.13,No.21533-4880/2013/13/1535/004doi:10.1166/jnn.2013.59811535Delivered by Publishing Technology to: University of New South Wales IP: 149.171.232.34 On: Wed, 27 Feb 2013 03:01:32Copyright American Scientific Publishers R E S E A R C H A R T I C L Eoxyanion,(BO 3 3−,for the larger tetrahedral (PO 4 3−unit in LiFePO 4.The effects of boron content on the struc-ture and electrochemical performances of LiFePO 4were discussed and characterized by XRD,SEM and Electro-chemical Measurements.2.EXPERIMENTAL DETAILSLiFeB X P 1−X O 4− material was prepared by a solid-state reaction consisting of a mixture of Li 2CO 3,Fe(C 2O 4 ·2H 2O,NH 4H 2PO 4in the molar ratio of Li:Fe:B:P =1.01:1:X :1−X ,in which the X =0 3 6 9,mol%.All chemicals are of analytical grade from Changzheng (Chengdu,China).The synthesized samples were marked by B0,B3,B6and B9,respectively.The stoichiomet-ric precursors were mixed by ball milling in anhydrous ethanol for 4h.After some carbon source citric acid wasadded into the ball-milled precursors,it was ball-milled for another 4h.Then the resulting gel was dried at 40Cunder vacuum,thoroughly reground again.The dry mix-tures were put into a tube furnace full with purified N 2gas,and subjected initially to a lower calcination at 350C for 8h and subsequently to 700C for 14h.The X-ray diffraction (XRD)patterns were analyzed on a Bruker DX1000diffractometer,using Cu K radiation,operating at 40kV and 25mA in the angular range 10≤2theta ≤70 with an acquisition step of 0.02with 1s/step of continue time.A scanning electron microscope (Hitachi S-4800)was used to examine the particle size,microstruc-ture and morphology of the prepared products.The positive electrode slurry,with a ratio of LiFePO 4:acetylene black:poly(vinylidene difluoride)=80:15:5(wt%),was made by mixing the samples powder with acetylene black and PVDF in a solvent of N -methyl pyrrolidone (NMP).The slurry was coated on an alu-minum foil and dried at 120 C for 12h in a vacuum oven to obtain a cathode electrode.The formed cath-ode foil was assembled into a CR2032battery in an argon-filled glove box,with Li foil as anode,1M LiPF 6in ethylene carbonate (EC):dimethyl carbonate (DMC)(1:1(vol%))as electrolyte and a Celgard-2400separa-tor.The electrochemical performance of the cell samples was tested by a high precision battery performance test-ing system (Neware,Shenzhen,China).The galvanostatic charge/discharge cycling performance of the cell samples was tested at different C rates in the range of 2.5–4.3V at room temperature.3.RESULTS AND DISCUSSIONFigure 1shows the XRD pattern of the samples.All of thepeaks are indexed on the basis of an orthorhombic olivine-type structure in the Pnma space group (no.83-2092inJCPDS database).No second phase was found except forthe sample B9,of which the peak (220)wasabnormally Fig.1.XRD patterns of the samples.increased.The analysis of the abnormal peak showed that vonsenite (Fe 2Fe(BO 3 O 2 impurity could be existed in the sample,demonstrating that the substitution of boronto about 6mol%phosphorus allows for the synthesis ofsingle-phase LiFePO 4without unwanted impurity phases.The lattice parameters,calculated from the refined XRD patterns,of the pure LiFePO 4sample were a =10.2894Å,b =5.9933Å,c =4.6720Åand volume =288.11Å.These values,which were very close to but slightly lowerthan the standard data (a =10.347Å,b =6.019Å,c =4.704Å),probably due to the sample’s single-crystal-likecharacteristics.11Figure 2shows the continuous changein lattice dimensions as a function of the substituted boron content in LiFeB X P 1−X O 4− .The lattice parame-ter of a and b decreased with the increasing substituted boron element content probably because more tetrahedral anion units,(PO 4 3−,in LiFePO 4were replaced by smaller and plane triangle oxyanion,(BO 3 3−.However,the lat-tice dimension of c increased with the increase of substi-tuted boron content,this could be explained by that the plane direction of the substituted plane triangle oxyanion,(BO 3 3−is parallel to the plane formed by a axis and b axis.Figure 3(a)shows the voltage profile of the four sam-ples in the first cycle at different C rates in voltage range of 2.5–4.3V .The observed reversible initial dis-charge capacities of 153.99mAh/g,147.78mAh/g and 145.62mAh/g at 0.1C for sample B0,B3and B6were obtained,respectively.The results revealed that the spe-cific capacities of the synthesized samples were slightlydecreased when boron element was used to partly replace the phosphorus element in the LiFePO 4.The possible rea-son of this phenomenon is that some boron element are in the lattice position of lithium,but we really believe that it would be overcame or improved by changing the solid-state method to other synthesizing method such as sol–gel method,hydrothermal method or ion exchange method.When the amount of substituted boron reached 1536J.Nanosci.Nanotechnol.13,1535–1538,2013Delivered by Publishing Technology to: University of New South Wales IP: 149.171.232.34 On: Wed, 27 Feb 2013 03:01:32Copyright American Scientific Publishers RESEARCH ARTICLEttice dimensions as a function of boron content in LiFeB X P 1−X O4−.9mol%,only an initial specific capacity of139.21mAh/g at0.1C was obtained because the second phase vonsenite (Fe2Fe(BO3 O2 impurity might be appeared in the sample from the results of XRD analysis(Fig.1).The cycling performances of all samples at the constant current density of0.1C,0.2C,0.5C,1C,2C and5C are shown in Figure3(b).We know that3mol%and6mol% of boron substitution had negligible affect on the discharge performance of the products,but significantly improved the cycling performance from the Figure3(b).At0.2C, the sample B0,B3and B6had a discharge specific capac-ity of147.47mAh/g,143.37mAh/g and142.30mAh/g,(a)(b)Fig.3.Electrochemical performances of the samples.(a)Initial charge/discharge curves of the samples at0.1C.(b)Cycling perfor-mances of the samples.respectively,and the95.77%,97.02%and97.73%of the initial discharge capacity at0.1C were retained,respec-tively.At0.5C,the discharge capacities of the sample B3 and B6can equal to that of the sample B0,with a specific capacity of about138mAh/g.The initial discharge capac-ity of the sample B3and B6at1C was136.27mAh/g and125.39mAh/g respectively,which exceeded that of the sample B0at the same discharge rate.At2C and5C, the discharge capacity of the sample B0was decreased to 108.65mAh/g and90.90mAh/g,only a capacity retention ratio of70.56%and59.03%were kept.However,the sam-ple B3and B6can retain the initial discharge capacity of 119.62mAh/g and118.79mAh/g at2C,109.98mAh/g and111.11mAh/g at5C,respectively.The capacity reten-tion ratios of81.23%at2C,74.42%at5C for sample B3 and the retention ratios of81.13%at2C,76.30%at5C for sample B6,respectively,were retained.All abovemen-tioned analysis demonstrates that appropriate amount of boron substitution for the phosphorus(less than6mol%) in LiFePO4can significantly ameliorate the rate perfor-mance of LiFePO4cathode material.The beneficial effect of proper amounts of boron substitution on rate performance of LiFePO4could be explained as follows:Firstly,a proper amount of boron substitution have no effect on the basic orthorhombic olivine-type structure of the LiFePO4material and the extraction/insertion of Li+are not impeded;Secondly, lighter small plane triangle oxyanion,(BO3 3−,with a molecular weight of58.8,replace the larger tetrahedral (PO4 3−unit with a molecular weight of95.0may facili-tate Li+transportation for more of larger Li+channels were achieved;Thirdly,the controlled off-stoichiometry of oxygen formed through the substitution of boron for phosphorus,which break the quasi-hexagonal close-packed oxygen array,significantly improve the electronic conductivity of the material.However,the lower initial charge/discharge capacity and rapid capacity fading of the sample B9at high rate demonstrates that the Fe2Fe(BO3 O2impurity not only have bad effect on the discharge capacity of LiFePO4material,but also signifi-cantly affect the material’s rate performance.Figures4(a)and(b)represent the sample B0(LiFePO4 and the sample B6(LiFeB0 06P0 94O4− composite pow-ders consisting of nanometric particles with a partially irregular morphology or quasi-spherical shapes.A very fine powder with an average size of about120–150nm for sample LiFePO4were obtained(Fig.4(a)).The particles of the LiFeB0 06P0 94O4− sample with a homogeneous parti-cle sizes between80–110nm except for a very few abnor-mal particles that are larger than150nm were smaller than that of the sample pure LiFePO4(Fig.4(b)).Com-paring with the pure LiFePO4sample which had two or more particles partly connected each other through melted substance,the particles of LiFeB0 06P0 94O4− sample were well distributed and almost with no agglomeration,indicat-ing that the adding of boron contained compound,H3BO3,J.Nanosci.Nanotechnol.13,1535–1538,20131537Delivered by Publishing Technology to: University of New South Wales IP: 149.171.232.34 On: Wed, 27 Feb 2013 03:01:32Copyright American Scientific Publishers R E S E A R C H A R T I C L EFig.4.SEM micrographs of the sample (a)B0and (b)B6.into raw material could further stop the LiFePO 4parti-cles growing to bigger size and facilitate particle distri-bution during synthesizing process.From this perspective,the excellent rate performance of the LiFeB 0 06P 0 94O 4−sample could also be explained that the average particlesize is smaller than that of the sample LiFePO 4and the particle dispersion is better.4.CONCLUSIONS The B -substituted LiFePO 4were successfully synthe-sized by solid-state method using Li 2CO 3,FeC 2O 4·2H 2O,H 3BO 3and NH 4H 2PO 4as raw material.Up to 6mol%of boron substitution had no effect on the structure of LiFePO 4material,but significantly improved the rate per-formance due to the improved electronic conductivity caused by off-stoichiometry of oxygen and the facilitated extraction/insertion of Li +channels produced by substi-tuting of lighter small plane triangle oxyanion,(BO 3 3−,for the larger tetrahedral (PO 4 3−unit.The initial dis-charge capacity of the LiFeB 0 06P 0 94O 4− sample was 145.62mAh/g at 0.1C,and the capacity retention ratios of about 81%at 2C and 76%at 5C were obtained,demon-strating that proper amount of boron substitution could improve the rate performance of LiFePO 4material.Acknowledgments:This work was financially sup-ported by The Sichuan Province Key Technology Support Program (2011GZ0131)and Innovation Fund for Technol-ogy Based Firm from Ministry of Science and Technology (11C26215103354).References and Notes1. A.K.Padhi,K.S.Nanjundaswamy,and J.B.Goodenough,J.Elec-trochem.Soc.144,1188(1997).2.J.B.Goodenough and Y .Kim,Chem.Mater.22,587(2010).3. A.Manthiram,J.Phys.Chem.Lett.2,176(2011).4.M.M.Doeff,Y .Hu,F.McLarnon,and R.Kostecki,Electrochem.Solid.State.Lett.6,A207(2003).5.S.Ferrari,vall,D.Capsoni,E.Quartarone,A.Magistris,P.Mustarelli,and P.Canton,J.Phys.Chem. C.114,12598(2010).6. B.Kang and G.Ceder,Nature 458,190(2009).7.S.Chung,J.T.Bloking,and Y .Chiang,Nat.Mater.1,123(2002).8. A.Yamada,N.Iwane,Y .Harada,S.Nishimura,Y .Koyama,and I.Tanaka,Adv.Mater.22,3583(2010).9.Y .Z.Dong,Y .M.Zhao,Z.D.Shi,X.N.An,P.Fu,and L.Chen,Electrochim.Acta 53,2339(2008).10.Y .N.Xu,S.Y .Chung,J.T.Bloking,Y .Chiang,and W.Y .Ching,Electrochem.Solid.State Lett.7,A131(2004).11. D.H.Kim and J.Kim,Electrochem.Solid.State.Lett.9,A439(2006).Received:12September 2011.Accepted:30November 2011.1538J.Nanosci.Nanotechnol.13,1535–1538,2013。