当前位置：文档库 › Design of electrolyte solutions for Li and Li-ion batteries

Design of electrolyte solutions for Li and Li-ion batteries

Electrochimica Acta50(2004)

247–254

Design of electrolyte solutions for Li and Li-ion batteries:a review Doron Aurbach a,?,1,Yosef Talyosef a,Boris Markovsky a,Elena Markevich a,

Ella Zinigrad a,Liraz Asraf a,Joseph S.Gnanaraj a,Hyeong-Jin Kim b

a Department of Chemistry,Bar-Ilan University,Ramat-Gan52900,Israel

b Battery Research Institute,LG Chem.Research Park,Daejeon305-380,Korea

Received2June2003;received in revised form15January2004;accepted30January2004

Available online3August2004

Abstract

This paper reviews approaches to the design of advanced electrolyte solutions for Li and Li-ion batteries.Important challenges are wide electrochemical windows,a wide temperature range of operation,acceptable safety features,and most important,appropriate surface reactions on the electrodes that induce ef?cient passivation,but not on the account of low impedance.We describe research tools,quick tests,and discuss some selected examples and strategies for R&D of solutions of improved performance.

Keywords:Additives;Surface reactions;Elevated temperature;Safety;Rechargeable lithium batteries

1.Introduction

Li-ion batteries are one of the most successes of mod-ern electrochemistry.These batteries,which became a com-mercial reality about a decade ago,are conquering the mar-kets with increasingly wider applications[1–5].Present chal-lenges are to extend their use to high power and large size applications(e.g.,propulsion,EV)[6–10].The current sys-tems use graphitic carbons as the anode material,LiCoO2 as the major cathode materials,mixtures of alkyl carbon-ates including ethylene carbonate(a mandatory component for suf?cient negative electrode passivation),dimethyl,di-ethyl,and ethyl-methyl carbonates(EC,DMC,DEC,EMC, respectively),and LiPF6as the electrolyte solution[1–10]. The alkyl carbonates were chosen due to their acceptable anodic stability for the4V cathodes used in Li-ion batter-ies,as well as lithiated graphite,together with other prop-erties,such as high polarity(i.e.,good conductivity of their solutions),a reasonable temperature range between freez-ing and boiling points,suf?ciently low toxicity,and accept-?Corresponding author.Tel.:+97235318317;fax:+97235351250.

E-mail address:aurbach@mail.biu.ac.il(D.Aurbach).

1ISE member.able safety features.The LiPF6salt is,to some extent,also a compromise.Other commercially available Li salts have too many disadvantages:LiAsF6is poisonous,LiClO4is ex-plosive,LiBF4is problematic on the negative side(reactions of the BF4?anion on the anode’s surface interfere badly with passivation),LiSO3CF3forms solutions of too low con-ductivity,LiN(SO2CF3)2and LiC(SO2CF3)3are problem-atic on the cathode side—the aluminum current collector currently used for the positive electrodes is not well passi-vated in the solutions and corrodes[11].LiPF6,the compro-mised salt,decomposes to LiF and PF5,and the latter readily hydrolyzes to form HF and PF3O[12].These two hydrol-ysis products are highly reactive on both the negative and positive sides,and their unavoidable presence in LiPF6solu-tions has a detrimental impact on the electrodes’performance [13].

Li-ion battery systems have a very limited performance at elevated temperatures,and their cycle life is also limited, due to surface phenomena on both electrodes that increase their impedance upon cycling[14].The safety features of commercially produced Li-ion batteries are not suf?cient for large size applications[15,16].Hence,in recent years we have seen intensive efforts to introduce new solvents,salts and ad-ditives that may lead to improvements in the performance

248 D.Aurbach et al./Electrochimica Acta50(2004)247–254

of these systems.As examples for new solvents,we should mention organo sulfur compounds,such as propylene sul-?te[17],?uorine-substituted compounds(e.g.,?uoro alkyl carbonates[18])and organo phosphorous compounds[19]. As new salts we mention Li bis(oxalato)borate(LiBOB)[20] and LiPF3(C2F5)3(LiFAP)[21,22].In addition to these salts, we should mention a wide variety of attempts to introduce new additives to the electrolyte solutions,including organo nitrates[23],sulfates[24],phosphates[25],active gases,such as SO2[26]and CO2[27],organoboran complexes[28], surface polymerizable agents,such as ole?ns[29],vinylene carbonate[30],aromatic compounds as over-charge protec-tion agents(by a shuttle mechanism,e.g.,biphenyl)[31]and more.

The aim of this paper is to provide some guidelines for R&D of new,improved electrolyte solutions for Li and Li-ion batteries,to suggest quick reliable tests,and to provide some new examples from our recent studies.

2.Experimental

We obtained standard solutions,including EC–DEC–DMC,EC-DMC and EC-DEC with LiPF61M from Merck (Selectipure series,which could be used as received).Li-FAP solutions were obtained from Merck.We used commer-cial graphitic and LiCoO2electrodes from LG.Experiments were carried out in glove boxes from V AC and MBrown.The tools for these studies included FTIR(Nicolet-Magna860, glove box operation,grazing angle re?ectance,and diffuse re-?ectance attachments),XPS(HS-Axis Kratos),ARC(Arthur de Little),DSC(Mettler),NMR(Bruker),XRD(Bruker), and standard electrochemical tools(EIS,SSCV,CV,PITT, galvanostatic cycling)with equipment from Solartron,Eco Chemie,EG&G,Maccor,and Arbin.Measurements were performed in home-made thermostats at a temperature range 25–80?C.(See references13,14,22,27,and30for details on the experimental aspects).

3.Results and discussion

The present arsenal of solution components that can be produced at high enough purity and reasonable prices in-cludes three families of solvents,namely,ethers,esters,and alkyl carbonates,and salts from the following list:LiPF6, LiBF4,LiN(SO2CF2CF3)2,(LiBETI),LiBC4O8(LiBOB), LiPF3(CF2CF3)3(LiFAP),and LiN(SO2CF3)2(LiTFSI). When selecting electrolyte solutions for battery application, the key features to look at are:

1.Evaluation of transport properties:what in?uences con-

ductivity and what parameters should be measured in or-der to evaluate the solvents properly.

2.The electrochemical stability,i.e.,the electrochemical

window.3.Temperature range of operation.

4.Safety features.

Evaluation of solution properties related to transport phe-nomena(conductivity,diffusion)has been dealt with thor-oughly in the literature.In general,important parameters that were developed include donor and acceptor numbers,and solvatochromic response(e.g.,ET30),in addition to triv-ial parameters,such as dielectric constant and dipole mo-ment[32](see detailed discussion in Refs.[33,34]).In gen-eral,high solvent polarity usually goes together with strong solvent-solute interactions.This means good solubility,but also high viscosity and high friction for ionic mobility as well.Hence,mixtures of solvents of high polarity and high viscosity with solvents of low polarity and low viscosity may provide optimal conductivity of Li salts(e.g.,alkyl carbon-ates plus ethers or esters)[35].In this respect,the salt con-centration should also be optimized,since too high a salt concentration means a high concentration of charge carriers, but strong solvent–solute and solute–solute interactions that may be detrimental to high conductivity(see discussion and examples in Refs.[33–35]).

The next points for discussion are the electrochemical win-dows of electrolyte solutions for Li-ion batteries.Fig.1shows schematically a summary of EQCM and voltammetric stud-ies of a typical Li-salt/alkyl carbonate solution(LiClO4/PC in this particular case)with a noble metal working electrode (Au in this case)[36].Major irreversible processes of inter-est are solvent oxidation of potentials>3.5V(Li/Li+),trace oxygen reduction(around2V versus Li/Li+),trace water re-duction(around1.5V versus Li/Li+),and gradual,continu-ous solvent and salt anion reduction at potentials below1.5V (Li/Li+).Other reversible processes,such as gold oxidation or Li UPD,which also appear in the scheme,are irrelevant to our discussion.

The study of the electrochemical windows of the Li bat-tery electrolyte solutions by both us and by others can be summarized as follows:

1.The order of oxidation potentials is alkyl carbonates>

esters>ethers.The oxidation of the solvents is usually the limiting anodic reaction.

2.In fact,even solvents of apparent relatively high anodic

stability,such as alkyl carbonates,undergo slow scale an-odic reactions on noble metal(Au,Pt)electrodes at po-tentials below4V versus Li/Li+[37].Nevertheless,these solvents are stable with4V cathodes(LiNiO2LiCoO2, LiMn2O4,etc.),whose charging potentials may reach

4.5V(Li/Li+),due to passivation phenomena.The com-

monly used4.V cathodes react with solution species and become covered by surface?lms[38].These surface?lms seem to inhibit massive solvent oxidation at potentials be-low4.5V.Consequently,alkyl carbonate solutions may be stable with cathode materials at potentials as high as5V

[39].Any negative electrode that operates at potentials be-

low1.5V(Li/Li+)should react with solution species and become covered by surface?lms comprising insoluble Li

D.Aurbach et al./Electrochimica Acta50(2004)247–254

249

Fig.1.A schematic presentation of the voltammetric behavior of an alkyl carbonate/Li salt solution with a noble metal electrode,studied by EQCM and spectroscopy[36].The various processes are sketched as peaks or waves,as they appear in the voltammograms.The m.p.e.numbers listed are the theoretical mass per electron values expected for the various surface?lm formation processes(EQCM)[36].

salts.Hence,Li or Li–C electrodes are obviously covered by surface?lms and should be de?ned as SEI electrodes

[40].Therefore,negative electrodes that operate at poten-

tials>1.5V(e.g.,Li x TiO y compounds)[41]may not be controlled by surface?lms.

3.High oxidation potential requires a high oxidation state of

the solvents’atoms are good for cathodes,e.g.,alkyl car-bonate solvents.However,the high oxidation state of the solvent atoms means high reactivity at the negative side.

As a result,with highly reactive anodes,such as Li metal-based systems,low oxidation state solvents/systems(e.g., ethers,PEO derivatives)should be used.This means a penalty on the positive side:with ethers/PEO derivatives one cannot use4V cathodes.Thus,with Li metal-based systems cathodes,such as V2O5,MV2O5(bronze),3V

Li x MnO2,etc.should be used,which are compatible with ethers and PEO derivatives.

The next point for discussion relates to electrolyte solu-tions for Li(metal)electrodes.The behavior of Li electrodes was intensively studied over the years(see for example,Refs. [42,43],and other references therein).Li electrodes are al-ways covered by spontaneously formed surface?lms com-prising insoluble Li salts,which are products of reduction of solution species by the active metal.

Since,the surface?lms on lithium are comprised of Li salts,they are not suf?ciently?exible to accommodate the changes on the active metal surface/volume during Li deposition–dissolution.Moreover,the surface?lms are very non-uniform on the microscopic and nanoscopic level.The

250 D.Aurbach et al./Electrochimica Acta50(2004)247–254

surface?lms on Li have a multilayer structure and later-ally they are mosaic-like,comprising different clusters of compounds,including Li salts,polymeric species,etc[42]. Hence,there is no way that Li deposition-dissolution can be uniform.In most of the electrolyte solutions,Li depo-sition is dendritic.Therefore,most of the commonly used electrolyte solutions,i.e.,esters,ethers,alkyl carbonates with LiPF6,LiBF4,LiClO4,LiN(SO2CF3)2,etc.are not suitable for rechargeable Li batteries with Li metal anodes.There are only a few electrolyte solutions in which Li deposition is not dendritic.One of them is LiAsF6/1-3dioxolane,stabi-lized with a tertiary amine[44],in which Li is deposited in a?ake-like formation.However,even solutions in which Li deposition is not dendritic,are not suitable for rechargeable Li batteries,because Li passivation can never be hermetic. Hence,if the charging rates of the Li anodes are not very low(C/9–C/12),Li is deposited in small grains,which un-avoidably react with solution species due to their high sur-face area.Consequently,there is a continuous depletion of the electrolyte solutions in practical rechargeable Li batteries whenever operated at practical charging rates(rate>C/3h), due to reactions between the solution components and high surface area Li deposits[44].

Over the years we have worked on Li surface modi-?cations,using Li–Al,Li3N,Li–Mg,and Li–Ga surface alloys[45].We also tested additives including alkenes and alkanes,surfactants,polymerizable agents,and active gases(CO2,SO2,N2O).The solvents that we tried in-cluded ethers,alkyl carbonates and esters,while the salts included:LiClO4,LiPF6,LiBF4,LiAsF6,LiC(SO2CF3)3, LiN(SO2CF3)2,LiSO3CF3,LiBr,and LiI.We concluded that there is no future for Li(metal)secondary batteries containing liquid solutions either because of dendrite formation,which means severe safety problems,or because of short cycle life if the charging rates are too high,since the solution disappears by reacting with Li deposits[44].

We foresee the future of rechargeable Li batteries in the use of PEO-based solid electrolytes at elevated temperatures. It should be noted that with gel electrolytes,the surface chem-istry of lithium electrodes is dominated by reduction of the solvents(usually alkyl carbonates)that are used as plasticiz-ers[46].Thereby,similar problems of poor passivation of Li electrodes may be encountered with gels,as is the case with liquid solutions.In the case of PEO-based electrolytes,the Li surface chemistry may be affected mostly by salt anion reduc-tion[47].However,in spite of the fact that Li attacks ethers and surface ROLi species are formed[47],the Li/PEO-based electrolyte interface is relatively stable and Li electrodes may behave very reversibly when in contact with electrolyte sys-tems based on derivatives of PEO.

The next subject is the effect of temperature.There are three separate issues to be dealt with.

1.The temperature range of operation primarily relates to

physical properties,such as freezing,boiling,and con-ductivity.It should be noted that high boiling points may

go together with high polarity,viscosity,and high freez-ing points(e.g.,as with cyclic alkyl carbonates,EC,PC), while low freezing points may go together with high volatility and low boiling points(esters,ethers,linear alkyl carbonates).It appears that relatively high temperature ranges can be obtained with standard alkyl carbonate solu-tions comprising ternary mixtures,e.g.,EC–DEC–DMC

[48].In general,the use of ternary and quaternary mix-

tures of alkyl carbonates,and alkyl carbonates with esters, allows the attainment of an impressive temperature range of operation and extends the applicability of Li batteries to very low temperatures(

2.There is a pronounced effect of elevated temperatures

on the electrodes’performance.The electrodes’surface chemistry depends on the temperature.At elevated tem-peratures,the electrodes’passivation may be lost(see later discussion).

3.As the temperature increases,safety issues emerge,i.e.,

thermal runaway,dangerous electrode-solution interac-tions,etc.These points are dealt with below.

Safety issues regarding Li and Li-ion battery systems have been intensively dealt with in recent years[49,50].When dealing with safety there are many subjects to consider,in-cluding?ammability,short circuit,overcharge and overdis-charge,dangerous heat dissipation,thermal runaway due to red–ox reactions of the electrolyte solutions,and thermal runaway/explosion due to electrode/solution interactions at elevated temperatures.The?ammability of the commonly used electrolyte solutions was dealt with in recent years,and organophosphorous compounds[19]and?uorinated com-pounds[18]were suggested as co-solvents in order to de-crease?ammability.However,when designing any new elec-trolyte solution for these batteries,one should remember that all electrolyte solutions in Li and Li-ion batteries are reactive with the electrodes.The key factor for battery performance is a surface chemistry that leads to a suf?cient passivation with good Li-ion transports at the electrodes’surface.All the other factors should be considered after this?rst condition is ful?lled.Hence,any introduction of a new co-solvent should involve a rigorous study of its impact on the electrodes sur-face chemistry,especially at elevated temperatures.

When dealing with safety features,critical issues are pos-sible thermal runaway scenarios for Li-ion batteries.Most of the electrolyte solutions for Li batteries are comprised of red–ox couples,with the solvent as a reductant and the salt as an oxidant.Hence,Li battery electrolyte solutions may undergo self-heating processes in which pressure is de-veloped and heat is liberated.Fig.2shows as an example results of thermal studies of LiPF6and LiPF3(C2F5)3(Li-FAP)solutions,using accelerating rate calorimetry(ARC) [51].This?gure shows self-heating rates versus the tem-perature of LiPF6,LiFAP,and LiPF6–LiFAP solutions in EC–DEC–DMC mixtures,measured by ARC.As seen from this?gure(upper part),LiPF6solutions have a relatively low onset for thermal reactions(<200?C).LiFAP solutions have

D.Aurbach et al./Electrochimica Acta50(2004)247–254

251

Fig.2.Self-heating rates of1M LiPF6,1M LiPF3(C2F5)3(LiFAP)and 0.5M LiPF6+0.5M LiP(F3C2F5)3solutions in EC–DEC–DMC(2:1:2)in ARC experiments(upper chart).Speci?c capacity vs.cycle number mea-sured in galvanostatic processes(C/10)of graphite electrodes(vs.Li elec-trodes)at80?C with LiPF6,LiPF3(C2F5)3,LiN(SO2C2F5)2(Li Beti),and LiPF6–LiFAP solutions(EC–DEC–DMC,2:1:2).

a higher onset(>200?C),but their self-heating rate is very high.

It was interesting to discover that solutions containing both LiFAP and LiPF6behave differently from the single salt so-lutions,as can be seen in the?gure.It was also interesting to discover that while all the single salt solutions that we tested cannot be suitable for graphite electrodes at elevated tem-peratures,graphite electrodes behave highly reversibly with LiPF6–LiFAP solutions at elevated temperatures(80?C),as is also seen in Fig.2(lower chart).The special properties of these solutions are currently being investigated.In addition to the red–ox reactions of the solutions,which lead to self-heating phenomena,as demonstrated in Fig.2,Li salt/alkyl carbonate solutions react vigorously at elevated temperatures with both lithiated graphite and delithiated cathode materials (e.g.,Li x CoO2(x<0.5)[52,53].

At elevated temperatures,the passivation of graphite elec-trodes is destroyed,and hence the lithium stored in them can react directly with solution species.Delithiated cathodes can oxidize the alkyl carbonates at elevated temperatures in exothermal reactions.Hence,any R&D of electrolyte solu-tions for Li batteries has to include calorimetric studies in order to map both the internal thermal red–ox reactions of the solutions(solvent/salt)and their high temperature reac-tions with the electrodes.

After the above review,the following question arises. How do we optimize solutions for Li-ion batteries?It should be noted that the commonly used solutions(e.g., EC–DEC–DMC,etc.and LiPF6)are a compromise!

The problems with these standard solutions include:HF contamination,which worsens passivation and induces ca-pacity fading(cathodes),the use of DEC in the mixtures for low T conductivity worsens passivation and thermal sta-bility,and the high temperature performance of these solu-tions is poor(capacity fading).In addition,these solutions are ?ammable and their electrochemical window may be limited to4.5V.

It is dif?cult to foresee a major revolution in the arse-nal of solvents for Li-ion batteries.We do not see any rea-sonable substitutes for the alkyl carbonate solvents that are currently in use.There are alternatives to LiPF6that mostly include,LiPF3(C2F5)3(LiFAP)[22]and LiBC4O8(LiBOB) [20].The former is still too expensive,while the latter is still being examined in several laboratories.We should note that the electrode–solution systems in Li-ion batteries are very complicated.By the introduction of a new component,one may gain in one aspect,but lose in another.We emphasize that the surface chemistry of the electrodes in solutions is the critical factor.

We suggest that the cheapest and easiest way to improve the performance of currently used electrolyte solutions for Li-ion batteries is by the use of surface reactive additives(at low concentration in solutions)that can react predominantly on the electrodes’surfaces and form highly protecting surface ?lms,which are stable at elevated temperatures.

During recent years,we have been studying several fami-lies of additives that could improve the performance of Li-ion battery systems.These include pyrocarbonates,dicarbonates, strain organic compounds,and organo silicon compounds.A key question here is how to test the impact of additives in the most ef?cient way.One approach may be to use elec-trolyte solutions in which graphite or LiCoO2electrodes show low performance,and to demonstrate how additives improve the electrodes’behavior.We,however,prefer an-other approach,where the reference solutions are all standard commercial solutions,and the tests are designed to demon-strate how the performance of a standard solution is improved by the use of certain additives.Fig.3shows typical results from such testing.Coin-type cells comprised of graphite and LiCoO2electrodes at the appropriate balance and a standard EC–EMC/1M LiPF6solution were cycled at60?C.This tem-perature is too high,and cells containing standard solutions fail very rapidly,as shown in the?gure.However,when we added some selected organo silicon compounds(still propri-etary,denoted as additives1and2)to the solution at relatively

252 D.Aurbach et al./Electrochimica Acta50(2004)247–254

Fig.3.Speci?c capacity vs.cycle number measured with coin-type cells containing LiCoO2and graphite electrodes and EC–EMC/1M LiPF6solutions, additive-free and with small amounts(%by weight indicated)of additive1from organo siloxane family and of additive2from alkoxysilanes(still proprietary), as indicated.60?C,galvanostatic cycling at C/10rates.

Fig.4.Description of self-discharge experiments and calculation of self-discharge currents with graphite electrodes in EC–DMC/LiPF6solutions at different temperatures(indicated).Graphite electrodes were stabilized and fully lithiated by slow scan rate voltammetry,and were then stored for50h at the temperatures indicated,followed by delithiation(by anodic,slow linear potential scan).Upper charts:The relevant cyclic voltammograms of the lithiation–storage–delithiation experiments.Lower charts:A typical plot of OCV vs.t of a graphite electrode upon storage(45?C)from which d E/d t is calculated,as illustrated(left),and a plot of Q(the electrodes’accumulative charge)during delithiation of the fully lithiated electrode,obtained by integration of the anodic voltammogram(I vs. E,t),from which Q(d x/d E)is calculated.

D.Aurbach et al./Electrochimica Acta50(2004)247–254253

low percentage(indicated),the performance of the cells im-proved considerably in terms of both speci?c capacity and stability,as demonstrated in Fig.3.It should be noted that the impact of organo silicon compounds is both on the solution bulk(scavengers of problematic contaminants)and on the electrodes’surface.This impact is currently being studied. Hence,we suggest a basic approach in which surface active additives are examined in tests at elevated temperatures,in which the electrodes/cells usually fail.If the additive pre-vents this failure at high temperature,it provides a relatively quick indication that the modi?ed solution is worth further investigation and optimization.

When using such tests,we recently found that the use of pyrocarbonate and dicarbonate solvents as additives,con-siderably improves the performance of both graphite and LiCoO2electrodes at elevated temperatures[54].

The last point dealt with in this paper also relates to sig-ni?cant tests for new solutions that can provide a suf?ciently good indication on the performance of modi?ed solutions. We propose a new testing approach that measures long-term impact in terms of self-discharge current.This approach is de-scribed in Fig.4,which provides data of graphite electrodes tested in standard EC–DMC/LiPF6solutions at different tem-peratures(three-electrode coin-type cells).This?gure shows typical slow scan rate voltammograms of graphite electrodes that were stored for50h at a fully lithiated state.The anodic branch of the voltammograms was measured after storage.It is clear from the charts that as the temperature is higher,more stored lithium is lost.The anodic voltammograms measured after storage starts at higher potentials.The self-discharge current is I s.d=|Q|(d E/d t)×(d x/d E)where|Q|is the total electrode capacity,d E/d t is the derivative of potential with time(measured during storage by the continuous changes in the OCV)and d x/d E is the derivative of the intercalation level (O

4.Conclusions

The major challenges in R&D of improved electrolyte so-lutions for Li-ion batteries are to obtain high anodic stability for5V systems,to improve low T conductivity and high T performance with minimal capacity-fading,and to improve safety features,meaning no thermal runaway and no?amma-bility.The key behavior factor is the electrodes’surface chem-istry:good passivation at a wide temperature range.We do not foresee a possible revolution with the currently used alkyl carbonate solvents,There are suggestions for new salts,e.g., a combination of LiPF6+LiPF3(CF2CF3)3is promising(sur-face and solution bulk effect),as well as the new LiBOB salt. The easiest,cheapest and most effective route for the im-provement of the currently used electrolyte solutions is the use of surface-active additives.However,it is important to carry out reliable tests for new solutions.We suggest tests at elevated temperatures and measurements of self-discharge currents during storage as a suitable basis for the selection of new additives that improve the electrode/solution interac-tions.

Acknowledgment

Partial support for these studies was obtained from LG Inc.,Merck KGaA,and GM(USA),and the German Ministry of Science,in the framework of the DIP Program. References

[1]J.M.Tarascon,M.Armand,Nature414(2001)359.

[2]M.M.Thackeray,Handbook of Battery Materials,Part III,

Wiley–VCH,Weinheim and NY,1999(Chapter1).

[3]B.Scrosati,in:W.A.van Schalkwijk,B.Scrosati(Eds.),Advances

in Lithium Ion Batteries,Kluwer Academic/Plenum Publishers,NY and London,2002,p.251(Chapter8).

[4]D.Aurbach,Special Issue on Li Insertion into Host Materials,Elec-

trochim.Acta45(1,2)1999(and papers therein).

[5]G.M.Ehrlich,Handbook of Batteries,third ed.,McGraw-Hill,NY

and London,2002(Chapter35).

[6]T.Tanaka,K.Ohta,N.Arai,J.Power Sources97–98(2001)2.

[7]R.Hamlen,G.Au,M.Brundage,M.Handrickson,E.Plichta,S.

Slane,J.Barbarello,J.Power Sources97–98(2001)22.

[8]R.A.March,S.Vukson,S.Surampudi, B.V.Ratnakumar,M.C.

Smart,M.Manzo,P.J.Dalton,J.Power Sources97–98(2001)236.

[9]A.Chu,P.Braatz,J.Power Sources112(2002)236.

[10]T.Takamura,Solid State Ionics152–153(2002)19.

[11]K.Kanamura,W.Hoshikawa,T.Umegaki,J.Electrochem.Soc.149

(2002)339.

[12]E.Sloop,J.K.Pugh,S.Wang,J.B.Kerr,K.Kinoshita,Electrochem,

Solid State Lett.4(2001)42.

[13]D.Aurbach,K.Gamolsky,B.Markovsky,G.Salitra,Y.Gofer,J.

Electrochem.Soc.147(2000)1322.

[14]D.Aurbach,B.Markovsky,A.Rodkin,E.Levi,Y.S.Cohen,H-J.

Kim,M.Schmidt,Electrochim.Acta47(2002)4291.

[15]M.Broussely,S.Herreyre,P.Biensan,P.Kasztejna,K.Nechev,R.J.

Staniewicz,J.Power Sources97–98(2001)13.

[16]H.Maleki,S.A.Hallij,J.R.Selman,R.B.Dinwiddie,H.Wang,J.

Electrochem.Soc.146(1999)947.

[17]G.H.Wrodnigg,T.M.Wrodnigg,J.O.Besenhard,M.Winter,Elec-

https://www.wendangku.net/doc/273716312.html,m.1(1999)148;

G.H.Wrodnigg,J.O.Besenhard,M.Winter,J.Electrochem.Soc.

146(1999)470.

[18]T.Nakajima,K.Dan,M.Koh,T.Ino,T.Shimizu,J.Fluorine Chem.

111(2001)167.

[19]X.M.Wang, E.Yasukawa,S.Kasuya,J.Electrochem.Soc.148

(2001)1066.

[20]K.Xu,S.S.Zhang,T.R.Jow,W.Xu,C.A.Angell,Electrochem.

Solid State Lett.5(2002)A26.

[21]M.Schmidt,U.Heider,A.Kuehner,R.Oesten,M.Jungnitz,N.

Ignat`e v,P.Sartori,J.Power Sources97–98(2001)557.

[22]J.S.Gnanaraj,M.D.Levi,Y.Gofer,D.Aurbach,J.Electrochem.

Soc.150(2003)445.

254 D.Aurbach et al./Electrochimica Acta50(2004)247–254

[23]Nitrate Additives for Nonaqueous Electrolyte Rechargeable Cells,

Wilson Greatbatch Ltd.,USA,Application for US Patent,An.

2000:752090.

[24]H.Gan,E.S.Takeuchi,Sulfate Additives for Nonaqueous Electrolyte

Rechargeable Cells,Wilson Greatbatch Ltd.,USA Eur.Pat.Appl Ep./2001,An.2001:4510457.

[25]H.Gan,E.S.Takeuchi,Phosphonate Additives for Nonaqueous Elec-

trolyte in Rechargeable Electrochemical Cells,Wilson Greatbatch Ltd.,USA,US patent application,US2001,An.2001:453468;

H.Gan,E.S.Takeuchi,Phosphonate Additives for Nonaqueous Elec-

trolyte in Rechargeable Electrochemical Cells,Wilson Greatbatch Ltd.,USA,US patent6,200,701.;

H.Gan,E.S.Takeuchi,Phosphate Additives for Nonaqueous Elec-

trolyte in Rechargeable Lithium Ion Batteries,Wilson Greatbatch Ltd.,USA,Eur.patent application,Ep./2000,An.2000:790241. [26]Y.Ein-Eli,S.R.Thomas,V.Koch,J.Electrochem.Soc.143(1996)

195.

[27]D.Aurbach,Y.Gofer,M.Ben-Zion,P.Aped,J.Electroanal.Chem.

339(1992)451.

[28]X.Sun,H.S.Lee,X.Q.Yang,J.McBreen,J.Electrochem.Soc.149

(2002)355.

[29]Y.Matsuda,T.Takemitsu,T.Tanigawa,T.Fukushima,J.Power

Sources97–98(2001)589.

[30]D.Aurbach,K.Gamolsky,B.Markovsky,Y.Gofer,Electrochim.

Acta47(2002)1423.

[31]H.Mao,Polymerizable Aromatic Additives for Overcharge Pro-

tection in Secondary Nonaqueous Lithium Batteries,Moli En-ergy(1990)Limited,Can.,Eur.patent application,Ep./1997,An.

1997:257375.;

H.Mao and D.S.Wainwright,Polymerizable Additives for Making

Secondary Nonaqueous Lithium-Ion Batteries Safe After Overcharge, Moli Energy Limited,Can.,Eur.patent application,Ep./1998,An.

1998:764156.;

H.Mao,and U.V on Sacken,Aromatic Monomer Gassing Agents

for Protecting Nonaqueous Lithium Batteries Against Overcharge, Moli Energy Limited,Can.,Eur.patent application,Ep./1997,An.

1997:443251.

[32]C.Reichardt,Angew.Chem.Int.Ed.Engl.18(1979)98.

[33]D.Aurbach,I.Weissman,Nonaqueous Electrochemistry,Marcel

Dekker,NY,1999(Chapter1).

[34]G.Blomgren,Nonaqueous Electrochemistry,Marcel Dekker,NY,

1999(Chapter2).[35]J.F.Cotezee,C.D.Ritchi(Eds.),Solute–Solvent Interactions,vol.2,

Marcel Dekker,NY,1976(and references therein).

[36]D.Aurbach,M.Moshkovich,Y.Gofer,J.Electrochem.Soc.148

(2001)155.

[37]M.Moshkovich,M.Cojocaru,H.E.Gottlieb,D.Aurbach,J.Elec-

troanal.Chem497(2001)84.

[38]D.Aurbach,J.Power Sources89(2000)206.

[39]K.Kanamura,J.Power Sources81–82(1999)123.

[40]E.Peled,in:J.P.Gabano(Ed.),Lithium Batteries,Academic Press,

London and NY,1983,p.43(Chapter3).

[41]M.Manickam,M.Takata,J.Power Sources114(2003)298.

[42]E.Peled, D.Golodnitsky,J.Peneiner,in:J.O.Besenhard(Ed.),

Handbook of Battery Materials,Part III,Wiley–VCH,Weinheim and NY,1999,p.419(Chapter6).

[43]D.Aurbach,Nonaqueous Electrochemistry,Marcel Dekker,NY,

1999(Chapter6).

[44]D.Aurbach,E.Zinigrad,H.Teller,P.Dan,J.Electrochem.Soc.147

(2000)2486.

[45]D.Aurbach,E.Zinigrad,H.Teller,Y.Cohen,G.Salitra,J.Elec-

trochem.Soc.149(2002)1267.

[46]D.Aurbach,O.Chusid,Y.Gofer,M.Watanabe,T.Momma,T.

Osaka,J.Power Sources97–98(2001)632.

[47]D.Aurbach,E.Granot,Electrochim.Acta42(1997)697.

[48]M.C.Smart,B.V.Ratnakumar,S.Surampudi,J.Electrochem.Soc.

149(2002)361.

[49]A.Funahaski,Y.Kida,K.Yanagida,T.Nohma,I.Yonezu,J.Power

Sources104(2002)248.

[50]H.Arai,M.Tsuda,K.Saito,M.Yahayaski,Y.Sakurai,J.Elec-

trochem.Soc.(2002)A401.

[51]J.S.Gnanaraj,E.Zinigrad,L.Asraf,H.E.Gottlieb,M.Sprecher,

D.Aurbach,M.Schmidt,J.Power Sources119–121(2003)

794.

[52]M.N.Richard,J.R.Dahn,J.Electrochem.Soc.146(1999)2068;

D.D.MacNeil,https://www.wendangku.net/doc/273716312.html,rcher,J.R.Dahn,J.Electrochem.Soc.146

(1999)3596.

[53]D.D.MacNeil,L.Christensen,https://www.wendangku.net/doc/273716312.html,nducci,J.M.Paulsen,J.R.Dahn,

J.Electrochem.Soc.147(2000)970;

D.D.MacNeil,J.R.Dahn,J.Phis.Chem.A105(2001)4430.

[54]D.Aurbach,M.D.Levi,C.Wang,E.Markevich,Z.Berkovich,J.

Electrochem.Soc.(submitted for publication).