jp908741d
Reversible Three-Electron Redox Behaviors of FeF3Nanocrystals as High-Capacity Cathode-Active Materials for Li-Ion Batteries
Ting Li,Lei Li,Yu L.Cao,Xin P.Ai,and Han X.Yang*
Department of Chemistry,Wuhan Uni V ersity,Wuhan430072,China
Recei V ed:September9,2009;Re V ised Manuscript Recei V ed:January7,2010
Three types of FeF3nanocrystals were synthesized by different chemical routes and investigated as a
cathode-active material for rechargeable lithium batteries.XRD and TEM analyses revealed that the
as-synthesized FeF3samples have a pure ReO3-type structure with a uniformly distributed crystallite
size of～10to20nm.Charge-discharge experiments in combination with cyclic voltammetric and
XRD evidence demonstrated that the FeF3in the nanocomposite electrode can realize a reversible
electrochemical conversion reaction from Fe3+to Fe0and vice versa,enabling a complete utilization of
its three-electron redox capacity(～712mAh·g-1).Particularly,the FeF3/C nanocomposites can be well
cycled at very high rates of1000-2000mA·g-1,giving a considerably high capacity of～500mAh·g-1.
These results seem to indicate that the electrochemical conversion reaction can not only give a high
capacity but also proceed reversibly and rapidly at room temperature as long as the electroactive FeF3
particles are suf?ciently downsized,electrically wired,and well-protected from aggregation.The high-
rate capability of the FeF3/C nanocomposite also suggests its potential applications for high-capacity
rechargeable lithium batteries.
Introduction
Li-ion batteries(LIBs)represent the most advanced battery
technology with the highest speci?c energy among all the rechargeable batteries currently commercialized,but they are still dif?cult to satisfy the fast-growing demand for light-weight and high-capacity electrical storage,such as in future wireless communications,electric vehicles,or power storage from renewable energy resources.Therefore,the search for new energetic materials for LIBs has been highlighted in battery chemistry and a wide range of intercalation com-pounds have been developed for high-capacity LIBs in the past decade.1-3Though many types of metal oxides and phosphates have been tested as alternative cathode materials,4,5 no real breakthrough has been achieved in capacity,especially for intercalation cathodes,the capacity-determining electrode in the present LIBs systems.This embarrassment probably originates from the intrinsic chemistry of intercalation reactions that allows no more than one Li+ion per formula unit to insert into the cathodic host lattice(i.e,less than one-electron redox).Thus,it is necessary to develop new redox mechanisms and feasible materials in order to build future rechargeable lithium batteries with dramatically increased energy densities.
Electrochemical conversion reactions seem to provide an alternative way to realize the signi?cant breakthrough in the storage capacity of the cathode materials for LIBs by full utilization of all the oxidation states of a high-valence transition-metal compound.In recent years,a number of metal ?uorides,6,7oxides,8,9sul?des,10and nitrides11,12have been demonstrated to produce a large multielectron redox capacity though reversible electrochemical conversion reactions:
Here,M stands for the transition-metal ions(M)Fe+3,Ni+2, Cu+2,etc.)and X denotes?uoride,oxide,or sul?de anions(X )F-,O-2,S-2,etc.).In such conversion reactions,M n X m is
electrochemically reduced,with lithium uptake,to M/Li n X at discharge,which reconverts to M n X m at subsequent charge, utilizing all the alterable valence states of the metal cations for reversible electrical storage.Particularly,many transition-metal ?uorides have their metallic cations in high oxidation states and a strong ionic character of M-F bonds,which are expected to give a high reversible capacity and high redox voltage when used as cathode-active materials.
However,metal?uorides have long been ignored as recharge-able cathode-active materials due to their insulating nature and apparent irreversibility in structural conversion.Until the late 1990s,Arai et al.?rst reported a high-voltage charge/discharge behavior of FeF3with a reversible capacity of80mAh·g-1, which is even far below the theoretical1e-transfer reaction capacity of the Fe3+/Fe2+couple(237mAh·g-1).13Recently, Badway et al.revealed a reversible conversion process of FeF3, leading to a cycleable capacity of～600mAh·g-1at a very low current rate from carbon-FeF3nanocomposites at70°C.14 At the same time,Li et al.reported the reversible phase-transition reactions of TiF3and VF3giving a Li-storage capacity of500-600mAh·g-1.6It has been well-recognized from these pioneering work that the electrochemical conversion reactions of these metal?uorides are extremely rate-and temperature-sensitive and full capacity utilization of the valent electrons of the?uorides is only attainable at low rates and elevated temperatures.6,14-17This phenomenon is understandable because these conversion reactions must involve structural decomposition and reconstruction of the metal?uorides along with reversible lithium insertion and transport in the bulk phases of metal ?uorides.To improve the capacity utilization and rate capability,
*To whom correspondence should be addressed.Tel:086-27-68754526. Fax:086-27-87884476.E-mail:hxyang@https://www.wendangku.net/doc/671235618.html,.mn Li++M
n
X
m
+mn e-T m Li
n
X+n M0(1)
J.Phys.Chem.C2010,114,3190–3195
3190
10.1021/jp908741d 2010American Chemical Society
Published on Web01/28/2010
it is favorable to downsize the metal?uorides so as to minimize the lithium insertion pathways and create the abundant active interfaces for reversible conversion of the metal?uorides.
In recent studies of reversible metal?uoride conversion materials,a FeF3-based nanocomposite was selected as a model system because of its potentially high capacity and output voltage as a cathode material for rechargeable Li batteries. However,the FeF3nanocomposites reported so far were mostly prepared by mechanical ball-milling with primary particle sizes on the scale of100-1000nm14,19and their electrochemical performances are far from satisfaction in terms of cycleability and rate capability.18-20
In this work,we synthesized three kinds of FeF3nanocrystals by different chemical routes and compared their electrochemical performances as cathode materials with our focus on the high-rate utilization and cycleability at room temperature.Also,the
three-eletron redox mechanism and structural evolution of the FeF3nanocrystals during the discharge/charge process were also described.
Experimental Section
Three types of FeF3nanocrystals were synthesized through different chemical routes.The?rst synthetic method was to precipitate the FeF3nanoparticles from a solution reaction of Fe(NO3)3ethanol solution(0.25mol·L-1)and NH4HF2aqueous solution(1.5mol·L-1)by thoroughly stirring,using polyeth-ylene glycol(PEG,20000)as a surfactant(0.005mol·L-1). The powder sample so prepared was marked as P-FeF3.The second method was to synthesize the FeF3nanoparticles by mixing a0.048mol·L-1Fe(NO3)3hexanol+H2O(1.06:1mol %)solution with a0.143mol·L-1NH4F hexanol+H2O(1.06:1 mol%)solution together,with a0.327mol·L-1cetyltrimethyl ammonium bromide(CTAB)as a surfactant,with vigorous stirring for2h and then separating the powder sample by the centrifugation of the reaction solution.The sample so prepared was denoted as C-FeF3.The third synthetic route was so-called as a liquid-solid-solution phase-transfer reaction already used in the preparation of rare-earth?uoride nanocrystals.21,22A typical experimental procedure is to mix1g of octadecylamine, 8mL of linoleate acid,and32mL of ethanol together at stirring to form a homogeneous solution and then add aqueous Fe(NO3)3 solution(1.25g/15mL distilled water)and NH4HF2solution (0.53g/15mL distilled water)one after the other into the mixed organic solution.This reaction mixture was stirred for about 10min and then transferred to a100mL autoclave,sealed,and hydrothermally treated at120°C for about6h.After they cooled down naturally to room temperature,the products were deposited at the bottom of the vessel.The?nal products were puri?ed with ethanol several times and denoted as L-FeF3.All the powder samples were dried under vacuum at80°C and then calcined at400°C for2h under high-purity argon to remove the organic residues.The FeF3/C nanocomposites were prepared by mechanical ball-milling of the as-prepared FeF3nanopowders with graphite for2h(FeF3/graphite)1:1by weight).
The phase analysis of the synthesized samples was performed using a Shimadzu XRD-6000system with Cu K R radiation.The electrode samples at different depths of charge and discharge for XRD analysis were made by disassembling the experimental cells in an Ar-?lled glovebox and taking out and rinsing the electrode in dimethyl carbonate(DMC).The dried electrode was sealed in a polyethylene pouch and taken out immediately for XRD characterizations.The morphology and microstructure of the synthesized FeF3nanocrystals were examined with a high-resolution JEM-2010FEF transmission electron microscopy system(HRTEM).The samples for HRTEM analysis were prepared by dispersing the powders in ethanol and releasing a few drops of the dispersed solution on a carbon?lm supported on a copper grid.
For electrochemical evaluation,the FeF3/C cathode was prepared by mixing80wt%active material,12wt%acetylene black,and8wt%polytetra?uoroethylene(PTFE)into ca.～0.1 mm thick?lms and pressing the cathode?lms onto an aluminum net.The charge-discharge experiments were carried out on the test cells of a three-electrode design with a Li sheet as a counter electrode and a small piece of Li metal as a reference electrode. The cells were assembled in an argon-?lled glovebox with a Celgard2400microporous membrane as the separator and1 mol·L-1LiPF6dissolved in a mixed solvent of ethylene carbonate(EC),dimethyl carbonate(DMC),and ethylene methyl carbonate(EMC)(1:1:1by wt)as the electrolyte.The cells were controlled by the BTS-55Neware Battery Testing System (Shenzhen,China)in a voltage range of4.5-1V at different current densities.Cyclic voltammograms(CV)were measured using a CHI660a electrochemical station(Shanghai,China)at a scanning rate of0.1mV s-1with the voltage ranges of4.5-1 V and4.5-2.5V.
Results and Discussion
Structural Characterization.XRD patterns of the FeF3 nanocrystallites synthesized by different chemical routes,as compared with pure FeF3powder,are shown in Figure1.For all the as-prepared samples,the entire diffraction patterns are very similar to those of pure commercial FeF3crystal and can be well-indexed to a ReO3-type structure(R3j c space group, JCPDS No.33-0647).14,20Except for this similarity,there ap-peared a few very weak peaks(26.8°and51.5°)in the XRD patterns of the synthesized samples corresponding to FeF2,which is possibly resulted from chemical reduction of a trace of FeF3 in the calcining process.By Lorentzian?tting of the XRD lines of the nanocrystal samples to obtain the2θandλvalues of the FeF3(012)peak and using the Scherrer equation(d)0.9λ/B cosθ),the average size of the as-synthesized FeF3nanocrystals was calculated to be about10-20nm.
This can also be clearly visualized from TEM images of the samples.Figure2shows the TEM images of the three FeF3 samples(P-FeF3,C-FeF3,and L-FeF3).Though slight aggrega-tion appeared in the case of C-FeF3,a single crystallite with a diameter of～10nm is evidently seen in the image for this sample.In contrast,the P-FeF3and L-FeF3samples were shown as homogeneously distributed crystallites with a uniform size of10-20nm,which agrees very well with the calculated values from the XRD data.Particularly,all the P-FeF3
crystallites Figure1.XRD patterns of the as-synthesized FeF3samples:(a)P-FeF3, (b)C-FeF3,(c)L-FeF3,and(d)pure FeF3(Alfa,ReO3-type).
Reversible High-Capacity FeF3Cathode Materials J.Phys.Chem.C,Vol.114,No.7,20103191
appear in a well-dispersed state and have an almost similar size distribution of ～15nm,suggesting that polyethylene glycol as a surfactant can effectively prevent the FeF 3nanocrystals from agglomeration.In comparison with the FeF 3nanocomposites previously reported,18-20the chemically synthesized FeF 3nano-crystals have a considerably reduced size and appear in a better-dispersed state.These morphological features of the FeF 3nanocrystals may be bene?cial to accelerate the phase-transform process,as well as lithium transport in the ?uoride phase,during the electrochemical conversion reaction.
Electrochemical Performance Characterization.The spe-ci?c capacity and cycleability of the FeF 3/C nanocomposites were directly evaluated by charge -discharge measurements at constant currents and at room temperature.Because carbon has no contribution to the speci?c capacity in this voltage range,7,14,19the charge and discharge capacities observed from the composite electrodes can entirely be attributed to electroactive FeF 3nanocrystals.Figure 3gives the voltage pro?les of the FeF 3/C electrodes cycled at a current density of 100mA ·g -1,which is a moderately high rate for conventional inserting cathodes in LIBs but ten times higher than those used for capacity evaluation of the ball-milled FeF 3/C samples recently reported.14,20Even though the charge and discharge were conducted at a quite high rate,the FeF 3/C nanocomposites showed a well-de?ned two-staged discharge with a higher voltage plateau at 3.5-2.0V,followed with a low voltage discharge at 1.7-1.0V,demon-strating a two-step electrochemical reduction process.These discharge features in the V -I curves of Figure 3are very similar to those observed from the ball-milled FeF 3/C samples at small current drain 19,20and at elevated temperature 18and are probably attributed to a two-step electrochemical conversion reaction,which proceeds through two successive steps:?rst,a one-electron reduction of FeF 3with Li insertion to form LiFeF 3and,subsequently,a two-electron reduction of LiFeF 3to produce Fe and LiF,as revealed in recent studies.14,20The total discharge capacities of these samples are all around 700mAh ·g -1,agreeing very well with the theoretical capacity (712mAh ·g -1)of FeF 3expected from a complete three-electron reduction.From the viewpoint of battery applications,only the discharge capacity of the FeF 3samples at the high voltage plateau is usable as a cathode-active material.In comparison,the P-FeF 3nano-crystals can deliver a quite high capacity of ～300mAh ·g -1in the high-voltage region of 4.5-2V,obviously higher than those of the C-FeF 3(～240mAh ·g -1)and L-FeF 3(～150mAh ·g -1),possibly because the P-FeF 3crystallites are highly dispersed with an uniform size distribution,which are kinetically favorable for the lithium insertion and phase-transformation reactions.Even taking only the high-voltage capability into account,the P-FeF 3and C-FeF 3nanocrystals can still realize much higher capacities (～300and 240mAh ·g -1,respectively)than currently commercialized high-capacity LiCoO 2(～150mAh ·g -1)and LiFePO 4(～160mAh ·g -1).These data demonstrate a potential feasibility that a high redox capacity can be practically realized from the FeF 3/C nanocomposite at room temperature and at a considerable high rate by an electrochemical conversion
reaction.
Figure 2.TEM images of (a)P-FeF 3,(b)magni?ed picture of (a),(c)C-FeF 3,and (d)L-FeF 3.
3192J.Phys.Chem.C,Vol.114,No.7,2010Li et al.
Figure 4shows the cycling performance of the FeF 3/C nanocomposite https://www.wendangku.net/doc/671235618.html,pared with the A-FeF 3/C sample made from commercial FeF 3(Alfa reagent),the chemically synthesized FeF 3/C nanocrystals display good capacity retention
and give a reversible capacity of 742,615,and 547mAh ·g -1,with slight capacity decay after 10cycles.This good cycleability suggests that the electrochemical conversion reaction of FeF 3could occur very reversibly as long as the FeF 3particles are suf?ciently downsized,electrically wired,and well-protected from aggregation.
To test their practical availability as a cathode-active material,we cycled the FeF 3nanocomposites at different current rates.As an example,Figure 5shows the high-rate performance of the P-FeF 3/C electrode cycled at various current densities.At a considerable high rate of 500mA ·g -1,the P-FeF 3/C electrode delivered an initial discharge capacity of 712mAh ·g -1and remained ～600mAh ·g -1after 10cycles.When the current density was increased to a very high value of 1000mA ·g -1,the discharge capacity of the P-FeF 3/C electrode could still reach ～500mAh ·g -1during the cycles.In recent studies of the electrochemical conversion reaction of tri?uorides,a full three-electron redox capacity can only be observed from FeF 3/C nanocomposites at a very low charge -discharge rate of ～7.6mA ·g -1.14,20The dramatic increase in the high-rate capability observed from the FeF 3nanocrystals chemically synthesized in this work implies that slow kinetics may not be an intrinsic nature of the conversion reaction,which could proceed rapidly and reversibly if appropriate nanodomains in the ?uoride electrodes can be created to facilitate the Li insertion and electron conduction in the insulative ?uoride particles.It should be pointed out that high loading of carbon (52%of the electrode mass),which may greatly dilute the active FeF 3within the conductive matrix,may also contribute greatly to the observed higher rate capability due to increased electroactive surface area.Electrochemical Conversion Mechanism.To further con-vince of the reversible conversion of the FeF 3nanocrystals electrochemically,we measured the cyclic voltammetric re-sponse (CV)of the FeF 3/C nanocomposite by a powder microelectrode technique.Figure 6shows a typical CV curve of the P-FeF 3/C powder at a slow scan of 0.1mV s -1in the voltage range of 4.5-1.0V.In the ?rst negative scan,there are two obvious reduction peaks at 2.8and 1.6V,respectively,in accordance with the two discharge plateaus in Figure 3,corresponding to Li +insertion into the FeF 3to form LiFeF 3and successive reductive decomposition of LiFeF 3into LiF and Fe 0.In the reversed scan,three distinct oxidation peaks appeared at 2,3.2,and 3.9V.The former two peaks are reasonably attributed to the reverse oxidation reactions of the FeF 3formation through two steps:electrochemical conversion of the LiF/Fe 0nanomixture into a LiFeF 3phase and consecutive Li deintercalation from the LiFeF 3phase to regenerate FeF 3nanocrystals.The third oxidative peak at 3.9V may be due to the surface ?lm formation on the ?uoride cathode by
electrolyte
Figure 3.Discharge/charge pro?les of the tri?uoride electrodes:(a)P-FeF 3/C,(b)C-FeF 3/C,and (c)L-FeF 3/C at a constant current of 100mA ·g -1.The inset in (a)displays a discharge curve of the P-FeF 3/C at the high-voltage region of 4.5-2.5
V.
Figure https://www.wendangku.net/doc/671235618.html,parison of the discharge capacities of the FeF 3/C electrodes at the ?rst ten cycles and at a constant cycling current of 100mA ·g -1
.
Figure 5.Cycling performance of the P-FeF 3/C electrodes at various high rates,as labeled in the ?gure.
Reversible High-Capacity FeF 3Cathode Materials J.Phys.Chem.C,Vol.114,No.7,20103193
oxidation because this oxidation current peak appeared only in the ?rst scan and disappeared since the second scan.
In a very recent study of the structural conversion of the FeF 3/C nanocomposite,20it was found by high-resolution XRD and MAS NMR spectroscopic analysis that the ?rst reduction step of Fe 3+to Fe 2+has comprised two processes:?rst,a half mole of Li-insertion reaction leading to a structural transforma-tion of the ReO 3-phase FeF 3to a lithiated rutile phase Li 0.5FeF 3with the Fe oxidation state close to +2.5and then a single-phase Li insertion to form LiFeF 3.This detailed mechanism can also be con?rmed electrochemically from the magni?ed volta-mmgrams of the as-prepared FeF 3/C nanocrystals,scanning in the potential region of 4.5-2.5V (Figure 7).As shown in Figure 7,two pairs of reversible redox peaks appear symmetrically in the potential region of 3.2-2.9V,indicative of the stepwise Li-insertion/deinsertion processes for the ?rst one-electron redox reaction of the FeF 3nanocrystals.
To further con?rm the assignments of the CV features and related structural conversions,we also performed an XRD analysis of the FeF 3electrode discharged and charged in different depths.As shown in Figure 8,the FeF 3electrode discharged to 2.4V shows a very different XRD pattern from its initial pristine phase with the appearance of a number of new diffraction peaks at 32.5,35,53.2,and 63°,while the strongest XRD peak (012)of the FeF 3phase at 23.8°decreases its intensity considerably and other XRD signals at 33.4°(104),34.4°(110),48.8°(024)and 54.2°(116/211)disappear com-pletely.Once discharged to 2.0V,the main peak (012)at 23.8°from the electrode is no longer visible.Instead,there appear a group of new peaks at 38.7,45.1,65.6,44.6,and 65°,which can be well-indexed to LiF and Fe 0,indicating the onset of the conversion reaction of Fe 2+to Fe 0.When the electrode is fully
discharged to 1.0V,only the XRD signals of LiF and Fe are observed,while the XRD peaks characteristic of Fe 3+and Fe 2+vanish completely.In reversed charge,the XRD lines of the FeF 3phase reappear and become dominant at a complete charge of 4.5V.This reversible change in the XRD pattern of the ?uoride electrode evidently demonstrates the reversible structural conversion of the tri?uoride during charge -discharge cycles.Conclusions
In summary,we prepared the FeF 3/C nanocomposites by chemical synthesis of FeF 3nanocrystals,followed by ball-milling the nanocrystals with graphite,and investigated these nanocomposites as cathode-active materials for high-capacity rechargeable lithium batteries.XRD and TEM analyses revealed that the as-synthesized FeF 3samples have a pure ReO 3-type structure with a quite uniformly distributed crystallite size of ～10to 20nm.Charge -discharge experiments demonstrated that all the FeF 3/C samples can realize a reversible electro-chemical conversion reaction from Fe 3+to Fe 0and vice versa,as con?rmed by CV and XRD evidence,capable of utilizing completely a very high three-electron redox capacity (～700mAh ·g -1).In addition,the FeF 3/C nanocomposites can be well cycled at a very high rate of 1000-2000mA ·g -1,indicating that the electrochemical conversion reaction could also proceed very reversibly and rapidly as conventional electrochemical intercalation reactions.The experimental data given in this study suggest a potential feasibility to use metal ?uorides as a high-capacity cathode material for lithium-ion batteries.
Acknowledgment.The authors acknowledge the ?nancial support by the 973Program,China (Grant No.2009CB220100).References and Notes
(1)Tarascon,J.M.;Armand,M.Nature 2001,414,359.(2)Armand,M.;Tarascon,J.M.Nature 2008,451,652.
(3)Bruce,P.G.;Scrosati,B.;Tarascon,J.M.Angew.Chem.,Int.Ed.2008,47,2930.
(4)Ohzuku,T.;Brodd,R.J.J.Power Sources 2007,174,449.(5)Whittingham,M.S.Chem.Re V .2004,104,4271.
(6)Li,H.;Richter,G.;Maier,J.Ad V .Mater.2003,15,736.
(7)Amatucci,G.G.;Pereira,N.J.Fluorine Chem.2007,128,243.(8)Morcrette,M.;Rozier,P.;Dupont,L.;Mugnier,E.;Sannier,L.;Galy,J.;Tarascon,J.M.Nat.Mater.2003,2,755.
(9)Poizot,P.;Laruelle,S.;Grugeon,S.;Dupont,L.;Tarascon,J.M.Nature 2000,407,496.
(10)Poizot,P.;Laruelle,S.;Grugeon,S.;Tarascon,J.M.J.Electrochem.Soc.2002,149,A1212.
(11)Pereira,N.;Klein,L.C.;Amatucci,G.G.J.Electrochem.Soc.2002,149,
A262.
Figure 6.Cyclic voltammograms of the P-FeF 3/C sample in 1mol ·L -1LiPF 6+EC -DMC -EMC.Scan rate )0.1mV s -1
.
Figure 7.Cyclic voltammograms of the P-FeF 3/C sample in 1mol ·L -1
LiPF 6+EC -DMC -EMC in a narrow potential range of 4.5-2.5V.Scan rate )0.1mV s -1
.
Figure 8.Ex situ XRD patterns of the P-FeF 3/C electrodes at different discharge and charge states,as indicated in the ?gure.
3194J.Phys.Chem.C,Vol.114,No.7,2010Li et al.
(12)Pereira,N.;Balasubramanian,M.;Dupont,L.;McBreen,J.;Klein, L.C.;Amatucci,G.G.J.Electrochem.Soc.2003,150,A1118.
(13)Arai,H.;Okada,S.;Sakurai,Y.;Yamaki,J.J.Power Sources1997, 68,716.
(14)Badway, F.;Cosandey, F.;Pereira,N.;Amatucci,G.G.J. Electrochem.Soc.2003,150,A1318.
(15)Doe,R.E.;Persson,K.A.;Meng,Y.S.;Ceder,G.Chem.Mater. 2008,20,5274.
(16)Badway, F.;Mansour, A.N.;Pereira,N.;Al-Sharab,J. F.; Cosandey,F.;Plitz,I.;Amatucci,G.G.Chem.Mater.2007,19,4129.
(17)Li,H.;Balaya,P.;Maier,J.J.Electrochem.Soc.2004,151,A1878.
(18)Plitz,I.;Badway,F.;Al-Sharab,J.;DuPasquier,A.;Cosandey,F.; Amatucci,G.G.J.Electrochem.Soc.2005,152,A307.
(19)Badway, F.;Pereira,N.;Cosandey, F.;Amatucci,G.G.J. Electrochem.Soc.2003,150,A1209.
(20)Yamakawa,N.;Jiang,M.;Key,B.;Grey,C.P.J.Am.Chem.Soc. 2009,131,10525.
(21)Wang,X.;Zhuang,J.;Peng,Q.;Li,Y.D.Nature2005,437,121.
(22)Wang,X.;Zhuang,J.;Peng,Q.;Li,Y.D.Inorg.Chem.2006,45, 6661.
JP908741D
Reversible High-Capacity FeF3Cathode Materials J.Phys.Chem.C,Vol.114,No.7,20103195