physrevb.44.9170 phase diagram of li„c6.pdf

7/27/2019 PhysRevB.44.9170 Phase diagram of LiC6.pdf

1/8

PHYSICAL REVIEW B VOLUME 44, NUMBER 17 1 NOVEMBER 1991-IPhase diagram of LiC6

J.R. DahnDepartment ofPhysics, Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6(Received 28 June 1991)

Using lithium/graphite electrochemical cells and in situ x-ray diffraction, we have determined thephase diagram of Li C6 for 0&x & 1 and O'C & T&70 C. We find several differences between our re-sults and previous work. For x & 0.04, the intercalated Li is randomly distributed throughout the graph-ite host in a dilute stage-1 phase. As x increases past 0.04, a clear coexistence between this phase and astage-4 phase is observed. As x increases further at 21 C, transitions to stage 3, to a "liquidlike" stage-2phase (denoted 2L), to stage 2, and finally to filled stage 1 occur sequentially as expected on the basis ofprevious work. We have been able to accurately determine the ranges in x and in T of the single-phaseand the coexisting-phase regions. Clear coexisting-phase regions are observed (indicating first-ordertransitions) for all the transitions except for the stage-4 to -3 transition. We discuss the difference be-tween this transition and the others. The average layer spacing of Li C6 has been measured as a functionof x.

INTRODUCTIONThe intercalation of lithium in graphite and studies ofthe staged phases which form have been extensively stud-ied both theoretically and experimentally. ' More re-cently, experiments have focused on the determination ofthe phase diagram of the staged phases ' using samplesprepared chemically having fixed overall composition.The sample temperature is varied and the structure isprobed using diffraction methods to determine the phasespresent. The phase diagram is determined by combiningthe phase-temperature results from numerous samples

having different Li concentration x in LiC6.Phase diagrams of Li intercalation compounds are wellprobed using electrochemical methods and in situ x-raydiffraction. "' Electrochemical cells are used to inter-calate Li into graphite at a fixed temperature and thevoltage of these cells is used to measure the chemical po-tential of the intercalated Li. ' Because the chemical po-tential of Li in coexisting phases is equal, coexistingphase regions appear as plateaus in plots of the voltageV(x, T) of Li/LiCs cells versus x at fixed temperature T.The derivative, dx/d V)T, shows peaks at the voltagesand compositions where coexisting phases occur.Structural information is simultaneously acquired usingcells with Be x-ray windows. ' Phase diagrams are deter-mined by combing the results obtained during a number

of fixed temperature intercalations.Early attempts to intercalate Li into graphite usingelectrochemical cells' ' have been plagued with side re-actions (caused by electrolyte decomposition) and there-fore are not reliable measurements of the chemical poten-tial variation. Recently, we learned how to overcomethese experimental difhculties' ' and studied the inter-calation of Li within disordered carbons. In Ref. 20 wereported preliminary studies on Li intercalated graphitewhich showed that the coexisting phase regions between

the staged phases in Li C6 were easily observed inV(x, T) and (dx/dV)z. Here we concentrate entirelyon an accurate determination of the phase diagram of Liintercalated graphite in the temperature range (0 C& T & 70 C) accessible to our electrochemical cells.

EXPERIMENTCrystalline synthetic graphite powder (99.9% pure),

designated KS-44, was obtained from Lonza Corporation.The mean particle size of this powder is 44 pm and thespacing between (002) planes, d(002), is 3.355 A indicativeof well-graphitized material. ' Electrodes were preparedfrom this powder as described previously. In some elec-trodes we added 7% by weight of Super S Battery Black(Chemetals Inc. Baltimore Md. , USA) a carbon blackwhich ensures good electrical contact between particlesin the electrode. Super S black reversibly intercalatesabout b,x =,' in LiCs. For the Li/(Super S BatteryBlack) cells described in Ref. 22, V(x, T) varies almostlinearly from x =0 at V=1.4 V to x =0.5 at V=0.01 V.Thus, for cells containing Super S Battery Black, we mustremember that a sma11 fraction of the total Li(0.5 X0.07/( 1 X0.93 )=0.037) is intercalating into theSuper S Battery Black, not into the graphite.Two electrode electrochemical cells were constructedusing the electrodes described above, Li foil (LithiumCorporation of America), porous polypropylene separa-tors (Celgard 2502 from Celanese Corp. ), and a nonaque-ous electrolyte containing a dissolved Li salt and asequestering agent which prevents the cointercalation ofthe electrolyte solvent along with the Li atom. ' Theelectrolyte was a 1M solution of LiN(CF3SOz)z (3M Cor-poration) and 12-crown-4 ether (Aldrich Chemical Co.)dissolved in a 50:50 volume mixture of propylene car-bonate and ethylene carbonate (Texaco). The salt wasvacuum dried at 140'C and the solvents were vacuum9170 1991 The American Physical Society


2/8

P IASE DIAGRAM Qp J j~ ~dist~lied prior to use. The 12-crown-4 ether was used asth 100received. The moisture content of th 1 1an ppm. e e ectro yte was lessTwo types of electrochemical cells were used here.Hermetically sealed test cells of the type described in Ref.18 were used to determine V(x, T) of Li/Li C6 ceHs.Electrochemical cells with Beso that struo t at structural changes in the intercalated graphitechar 'cou e measured in situ as x in L Cc arging or discharging the cells. Our diff. ur i ractometerTo determine V(x, T) and dx /d Vchar ed and x, cells werege an discharged using constant cur t bren s etweenvo age limits. The temperature of the celled constant in a thermostat to +0.1'C. The cellcyclers maintain stable currents to +0 5P . Care calculated from the cath d0 . o. hanges in xca o e mass, the constantcurrent, and the time of curr t fI . Dren ow. Data were mea-sured whenever V changed by +0.001 V. For the in siturac ion experiment, cells were equilibrated at fixed

tiostat; once the cell current fell below 1 A x-rawere measured. ThThe value of x corresponding to eachmeasured on ascan was then obtained from the cell voltameasure on a different cell using a current correspond-ing to a change b,x =1 in 800 h (fn or a typical cell with-mg graphite, this current is 6.9 pA). X-ra rwere collected at 66 p . -ray profilesa 66 voltages over a period of about 2months while the voltage of th 11e ce was graduallystepped down (intercalatin L') di an then graduallys eppe back up again (deintercalating Li).The x-ray experiment is used to identify the hp ase tran-responding to the various plateaus in V(x, T)e p asethe compositions corresponding to the be-ginning and the end of each plat Thcan often a eau. ese compositionscan o ten be determined directly from V(x), but are

w en x Vis plottedometimes more easily observed when xRESULTS

Figure 1 shows the first discharge and ch fLi/ ra hite c arge o a' g p i e cell using a current corresp d'chan e bx =1 ' espon ing to age x = in 100 h. The graphite electrode in thisThe latecell contained 7% by weight of S S Buper Battery Black.e p ateau near 0.85 V is associated with the dsition of the electrolyte on the surface of the carbons inwi e ecompo-the formation of a passivating la er ' This 1er. is plateau is noto serve on subsequent cycles because once thce e passivat-g ayer is formed, it protects the lithiated b fa e car on romer reaction. The other plateaus below 0.52 V hcorres onond to a transition between staged phases of~ eacdifferent composition. For exork or examp e, in our previouswor, we showed that the lowest volt 1ge p ateau (near

1.2

0.8E 0.6}D+ 0.4

0.0

0.251.0

hl}05C)

~ 0.20C3

0.100.0 0.1 0.2 P.Sx in Li.c, 0.4 0,50.0 0 40 80 120 160 BQO 240TIME (h)FIG. 1. Voltage vs time for the first discharge and char ea Li/graphite cell at 30 C Thchange Ax =1 ' L C e currents used corin i in 100h. respond to a

FIG. 2. Volta e vs x foror a Li/graphite cell measured at 30 Cusing currents corresponding to a change b,x = 1 in Li C inin (a) and in (b) is for the discharge of thecell and the upper curve is for the chare ween t e plateaus in V xor e c arge. The correspondence'x, T) and transitions between thestaged phases has been indicated for the discharge


3/8

9172 J.R. DAHN70 mV during discharge) corresponds to the stage-2tage-1 coexistence region. The discharge of the celltook longer than the subsequent charge because the elec-trolyte decomposition associated with the formation ofthe passivating film consumes Li which has beentransferred to the graphite electrode. Although this pro-cess is clearly evident at 0.85 V, it occurs slowlythroughout the rest of the first discharge.Figure 2 shows the second discharge-charge cycle of aLi/graphite cell at 30'C using currents corresponding toan 800-h rate. The cell was discharged only to 0.1 V,which corresponds to x =0.5, and then charged. Hadthe cell been allowed to discharge further, to 0.02 V,another plateau extending from x =,' to 1 would havebeen observed. Figure 2(b) clearly shows the plateaus0which mark coexisting phase regions. We have labeledthe plateaus on discharge based on the x-ray work to bepresented - below. The offset between the charge anddischarge is due to the IR drop from the internal resis-

40

SOLID DISCHARGEDASH CHARGE

~ 20I

A lgII

tance of the cell. If the cell could be measured under tru-ly equilibrium conditions, V(x, T) would lie midway be-tween the charge and discharge data shown.Figure 3 shows dx /d V)z calculated from the dataof Fig. 2. The derivative is obtained simply by takingAx/hV for adjacent data points. The peaks labeled 2,C, and D in Fig. 3 correspond to the coexistence ranges:dilute stage-1 tage-4; stage-3 tage-21. ; and stage-2I.tage-2, respectively, as seen by comparing to F' 220 ig.In a previous work the voltage resolution (0.01 V inRef. 20) was 10 times poorer than in the present workand we did not observe C. The peaks labeled 8 are notunderstood and will be discussed further later.Using I,n situ x-ray diffraction, we examined thestructural changes taking place in different regions of th ei/graphite voltage curve. The experiments reportedhere were done while Li was being inserted into the car-bon; similar results were obtained during deintercalation.Figure 4 shows portions of x-ray scans in the region ofthe graphite (002) peak taken for V) 0.230 V, above thefirst plateau in Fig. 2. The (002) peak corresponds to theaverage spacing between carbon layers. For V) 0.230 V,this peaks shifts smoothly to lower angle as the cell volt-

age decreases (as Li intercalates within the carbon) Thh ift corresponds to an increase of the (002) plane spacingdue to the presence of Li between the layers. All theBragg peaks arising from the intercalated graphite can beindexed assuming that it is a single phase at each voltage(Li concentration), but whose x axis varies smoothly withvoltage. No evidence of superlattice peaks indicative ofstaged phases were observed for V) 0.230 V. This evi-dence, coupled with the smooth variation of V(x) forV)0.230 V proves that Li is intercalated uniforml

w i00000

6

I

SOLID D ISCHARGEDASH CHARGE

I

j

)

t

80000

~ 60000-M40000

80000-

/0.10 I0.15 l0.20 0.25VOLTAGE (V)FIG. 3. (dx/dV)T vs Vfor the data of Fig. 2. The data in

(b) is the same as in (a) except plotted on an expanded verticalscale. The peaks labeled A, 8, C, and D in (a) are described inthe text.

086.0 P,6.5 87.0SCATTERING ANGLE (deg)FICz. 4. X-ray intensity vs scattering angle (Cu Ea radiation)for the (002) region of Li C6 in the dilute stage-1 phase mea-sured by in situ diffraction at 21'C. From the right, the scanswere measured at the following cell voltages: 2.00 (fresh cell),0.400, 0.295, 0.257, 0.257, and 0.230 V. One of the scans at0.257 V was measured as the cell voltage was stepped down, andthe other while the cell voltage was stepped up.


4/8

PHASE DIACTRAM OF L'Li C6 9173within host at these concentrationsstage-1 phase. ons forming a diluteNext we examined the plateau in VFigure 5(a) shows thes e region near the dilu g-

h t -4 pea [in this case, thef '"h'xpected to appear. Th(00) 1 tf 't-n p ase is given b theh h (002) 1cep or gra hitep ne spacing is us dstage-n phase co t ' anesains n carbon lanescept for the dilute sta -1, ' ros age-1 phase, which r

8 1 'nk f""the intercalant is onl re p es appear becauseon y present between ever thery nt pair of

carbon planes for stwhich are observables age n. In our not ation, (OOI) peaks,erva e only because of thcalled superlattice peaks. The Miller iI hh'l ' P g y

t 4 h 1 1en e dilute sta e-1's c ear y observed. In tdisplayed in Fig. 5( ) ( 11co ected at 0.202n the third scancorresponding to th 2 V), two peakse average la er scoexisting phas y spacings of the twoases, are observed. Theat the lower volt age is t e stage-4 hae phase which formspage- p ase appears exactly where expected

K+ 80-

w 60-

025.5~3K4+g

26.0 26.5(b)

025.5 26.0 26.5

130 I I ISCATTERING ANGLE i esi u diffraction results for a Li ' eute stage-1 tage-4 coexistence re ions or a i/Li C6 cell in the

d'-edf o th bo ttom in (b), the scans were in a)

(a) shows thebon l p cing between cing to the avera e s a

'

(b h b off ' llows t e stage-4 (005ver ica y by 0.3 for clarity. ]

l

SCATTERING ANGLE DEGREES)FIG. 6. In situu infraction results in theF he nght in (a) and fre sure at the followin g o g0, and 0.160 V. a rad' h g p gla 0 t e avera e s

raggcar onpea {near 33 ) and thoft'set vertically b 0 3 f. or clarity. ] ave een


5/8

917& J.R. DAHNown in Fig 5(b)]As further Li is added, we n

h h 11 1o tage varies a r o a regiappreciably with ~. F't k f 0201V3. Fi greater than peak B (onigure 6(a) shows tha on discharge) in Fiavera e

s at the peak corrure 6

p gy as Li is added. F's i ts smoothl

of h (005) peak of the sta e-4 est epreh -3 hshift ofh 1 p ase. There aice intensity from the mootpsta e-age-4 packets rando 1 i a ' opose, , oi a raction, 1 f,o

0--P5.05 26.05.5

stage-3 packets As L' ' se tede6 he superlattice peak oV, (005) o 3 (004) in a contin

m stage 4e o not full un muc like

g. stage-3 transitio bage- tounderstand the st -4

dff t fro th d' mar edly'b d bo th age- L and the stage-2L to

d'ff io d 14 t -3 t itio dF 7re shows sn situ x-ra -dre s o ' -ray- iff'raction data 11co ectedd d In this ran e of ge is step-t 2L d t 2L g-eo spo d g to the aveig. 7 b) shows averagewhich probes the (0 e sta e-kofbo h -2 hFi. 7() d hg e third scan fromsta e-3 h d 01 y the fifth scan, thet 8 th fifththrou h pcan we have mo

two scans e in Fig. 3(a . urmg the next3() d fo h -2 i erence between the aver-g p

fb th k Ca s and Din Fi e pres-ko h hep asediagramof Li C. ' '6ho h hK 8,24 phase disa ear

'xppears below about 260tern er

asurements of dxe is. Figure 8

Us'g g shows dx/d V)T

16

8I

I

II I

III

I II I

I

III

'I

IIII 40'C

32 I34 86 38DEGREES)ATTERING ANGLEFIG. 7. In situ diftraction rtage-2 regions From

in (b), therom the right in (a

e asure at the folio)g olg

~ y y andgg peaks correspondin tbon la e ( ho ing

to the avera e s() ws t e sta e-3 car-pea (near 38.7') Th ~ aildh 11' L' dano e in the cell. Th

is romo8' t t 11 b 0.3fo

55'C

0.0965'C

0.13 0.17 0.21OLTAGE ~vFIG 8 dh/dV)T vs volta e fatures indicated g e t correspo d. The corres ond ence between thon ing to a

data has x 6er ica ly by 2.0 V ' or clarity.


6/8

PHASE DIAGRAM OF L~x C6 9175plotted versus Vmeasured at 30, 40, 55, and 65'C for thedischarge of a Li/graphite cell using a current where achange, Ax =1 in Li C6 occurs in 200 h. The separationbetween the peaks in (dx/dV)z, corresponding to thestage-3 tage-2L and the stage-2L tage-2 transitions,decreases as the temperature decreases. Once these peaksmerge into a single peak, as the temperature is furtherlowered, then a transition directly from stage 3 to stage 2will occur. Figure 9 shows the separation between thepeaks plotted versus temperature. Assuming that theseparation varies linearly with temperature, we expectthat the 2L phase will disappear below 10 +2'C. Mea-surements at 10'C confirm the absence of the stage-3tage-21. peak in (dx /d V) z .The peaks labeled 8 in Fig. 3(a) are still a mystery. Ourexperiments show that this peak is located in a rangewhere Li C6 is in the stage-3 phase. Perhaps this peak isassociated with some ordering of the intercalated Liwithin each occupied layer. Our di6'raction experimentsdo not probe the (hk0) reflections well because thegraphite electrodes in our cells have been compressed'causing (00l) reflections to be preferentially favored.Finally, we constructed some cells using electrodeswithout Super S Battery Black to probe the width of thedilute stage-1 region. Figure 10 shows V(x, T) for such acell in the dilute stage-1 region, measured using a currentcorresponding to hx =1 in Li C6 in 400 h. The single-phase region extends from x =0 to 0.04+0.005 at 30 C.The results in Fig. 10 agree well with those in Fig. 2 ifthe contributions of the Super S Battery Black present inthe cell of Fig. 2 are taken into consideration. The dataof Fig. 2 suggest the single-phase region has a width ofAx =0.07. The Super S Battery Black has a total capaci-ty which corresponds to Ax =0.037 between 1.4 and 0.01V. Assuming this capacity to be distributed uniformlyover this voltage range, the capacity of the graphiteabove 0.23 V in the data of Fig. 2 is

bx =0.07.037X1.2/1. 4=0.039,25

20

Olw10-02

12

C3~ 0.4

0.00.00 0.01 0.02 0.03x in LiCs 0.04 0.05FIG. 10. Voltage vs x for a Li/Li C6 cell at 30 C withoutSuper S Battery Black in the electrode for x in the dilute stage-1region. The current corresponds to a change hx = 1 in 400 h.

I I II I II 0I I I6Q- I II I I

I I II I II I II I II I If

I I II II I I 31 I1+4I4 3 I+ ' 2L+2I2LII I I

I I II I II I II I II II I II I I 3+2I I I

0.0 0.1 0.2 0.3 0.4x in LiCs

A 40-

C4& 2O-

IIII

IIIIIIIIIII', 2+1I

IIIIIIIIII I0.5

in good agreement with Fig. 10.To construct the Li C6 phase diagram shown in Fig.11,we used the data of Figs. 2, 3, and 810 as well as thex-ray information presented above. For 0.5(x (1.0, acoexistence between the stage-2 and stage-1 phases is ob-served. The single-phase stage-2L and stage-2 phases arenarrow, of maximum width Ax =0.03; these have beendenoted as line phases in Fig. 11. There is little tempera-ture dependence in the phase diagram apart from thedisappearance of the 2L phase below 10'C. We have la-

0 0 800 40 60TEMPERATURE ('C)FIG. 9. Separation between peaks in (dx/dV)T corre-sponding to the 3+2J and 2L +2 transitions vs temperature.

FIG. 11. The phase diagram of LiC6 in the range O'C(T(70'C. The phase designations are described in the text.Data points are plotted where the error in x and T of the phaseboundary is less than +0.015 and +2'C, respectively. Else-where, the errors in x at a given T are estimated to be less than+0.03.


7/8

9176 J.R. DAHNoW 3.55~ 3.53~ 3.51P 3493.47~ 3.45~ 3.43

~ 3.41+ 3.39~ 3.37

I I I II I I II I I II II I Iu-gI I II I I I

I II I I II I I I

oI I I II I II I II I I

I II I

I I II I l3 I1'' 1'+4' 43 '+ 'I I 7 I Io I2LI

I I II I Ip I I-O I I I I

I I I IE) I I II ~ I I I0.0 0.1 0.2 0.3x in Lics

0

2L+2

0.4

III

o+IIIII

IIIIIIIII 2+I 1II

IIIII0.5

FIG. 12. Variation of the average layer between carbonplanes as a function of x in Li C6 measured at 21'C. The phasediagram information from Fig. 11 for 21'C is superimposed.beled the region of the phase diagram corresponding tostages 3 and 4 as "3,4." At the extreme left of the "3,4"phase, Li C6 is pure stage 4 while at the right it is purestage 3. In the discussion regarding Fig. 6, we showedthat distinct coexisting phases were not observed duringthe stage-3 tage-4 transition and that the transition ap-peared to be continuous. This is in disagreement with theresults reported in Ref. 8 (see Fig. 14 in that reference)where a clear two-phase coexistence region is observedbetween stage 3 and stage 4 LiC62 (Lio Q97C6) as the tem-perature is increased, although the temperatures are notgiven in that reference. We emphasize that further workis needed to fully understand the stage-3 tage-4 transi-tion.The major differences between our Li C6 phase dia-gram and those previously determined (for example, seeFig. 3.6 in Ref. 24) are now summarized. First, the dilutestage-1 phase, labeled 1' in Fig. 11, extends to from x =0to 0.04 at 300 K, and does not have zero width as impliedby previous work. Second, the width of the two-phasecoexistence region between the stage-2L phase and thestage-2 phase is much larger than implied by previouswork. Third, the stage-3 tage-4 transition has marked-ly different behavior than the other transitions which areall clearly first-order transitions. Finally, there is no evi-dence for the stability of phases with a stage indexgreater than 4 above 10 C.The lattice expansion induced by the intercalant can beused to probe the elastic energies associated with inter-

calation. ' Furthermore, phenomenological modelsused to predict the lattice expansion of Li C6 havesuffered in that precision data for comparison has notbeen available. Figure 12 shows the average spacing be-tween carbon layers plotted as a function of x from our insitu experiments. The regions of the phase diagram for21'C from Fig. 11 have been indicated in Fig. 12. In thetwo-phase regions, two sets of Bragg peaks are observedexcept in the 2I.+2 region where we cannot resolve theBragg peaks from each phase. The layer spacing forstage-1 LiC6 from in situ experiments has been reportedpreviously.

DISCUSSIQNThe qualitative features of our phase diagram and thatdetermined for Li C6 previously (e.g., see Ref. 8) aresimilar. Theories for the phase diagram of LiC6 requireaccurate experimental data for comparison. We havenow provided an accurate phase diagram determination,although only within a narrow temperature range.

Equally important are the measurements of the voltage ofLi/LiC6 cells which we have presented. These are adirect measure of the variation of the chemical potentialof the intercalated Li within the graphite as a function ofx.' Any theory for the phase diagram of Li C6 based onlattice-gas models allows for the calculation of this chem-ical potential variation. Clearly, if a theory is to be con-sidered reasonable, it must reproduce not only the phasediagram, but also the chemical potential variation. It isour opinion that the current theories ' are unable todo both. We are now exploring methods to improve thetheoretical description of intercalation in LiC6.The electrochemical methods outlined here are verypowerful methods for obtaining phase diagram informa-tion. However, the use of liquid electrolytes in Li/LiC6cells limits the temperature range we can probe. It is ouropinion that experiments using solid polymer electrolytesare warranted because the range of cell operation can beextended up to at least 1SO'C. The stage-3 tage-4 tran-sition in Li C6 could then be studied under conditionswhere the Li diffusion is much faster. If kinetics play arole in the apparent complexity of this phase transition,then this approach may allow the separation of the kinet-ic effects from the equilibrium thermodynamics.

ACKNO%'LEDGMENTSWe thank the Natural Sciences and EngineeringResearch Council of Canada and Moli Energy (1990) Ltd.for funding. Moli Energy (1990) Ltd. performed the vac-uum distillation of the electrolyte solvents. Discussionswith W. R.McKinnon regarding this work were useful.

R. Juza and V.Wehle, Nature (London) 52, 560 (1965).2M. Bagoin, D. Guerard, and A. Herold, C. R. Acad. Sci. Ser. C262, 557 (1966).D. Geurard and A. Herold, C. R. Acad. Sci. Ser. C 275, 571(1972).4P. Delhaes and J.P.Manceau, Synth. Met. 2, 277 (1980)

~J. E. Fischer, C. D. Fuerst, and K. C. Woo, Synth. Met. 7, 17(1983).J.E. Fischer and H. J.Kim, Synth. Met. j.2, 137 (1985).7J. E. Fischer and H. J.Kim, Synth. Met. 23, 121 (1988).J. E. Pischer, in Chemical Physics of Intercalation, edited by A.P. Legrand and S. Flandrois (Plenum, New York, 1987).


8/8

PHASE DIAGRAM OF Li C6 9177H. J. Kim, A. Magerl, J. L. Soubeyroux, and J. E. Fischer,Phys. Rev. B 39, 4760 (1989).K. C. Woo, H. Mertwoy, J. E. Fischer, W. A. Kamitakahara,and D. S. Robinson, Phys. Rev. B 27, 7831 (1983).J. R. Dahn and W. R. McKinnon, J. Electrochem. Soc. 1311823 (1984)~J.R. Dahn, W. R.McKinnon, and S. T. Coleman, Phys. Rev.B31,484 (1985).'3W. R. McKinnon and R. R. Haering, in Modern Aspects ofElectrochemistry, edited by R. E.White, J.O'M. Bockris, andB.E.Conway (Plenum, New York, 1983),No. 15, p. 235.~4J. R. Dahn, M. A. Py, and R. R. Haering, Can. J. Phys. 60,307 (1982).~5A. N. Dey and B. P. Sullivan, J. Electrochem. Soc. 117, 222(1970).

I Y.Takada, R. Fujii, and K.Matsuo, Tanso 114, 120 (1983).M. Arakawa and J. Yamaki, J. Electroanal. Chem. 219, 273(1987).R. Fong, U. Von Sacken, and J.R.Dahn, J.Electrochem. Soc.

137, 2009 (1990).J. R. Dahn and D. P. Wilkinson, U.S. Patent Application 602497 (pending).J. R. Dahn, R. Fong, and M. J. Spoon, Phys. Rev. B 42, 6424(1990).2~W. Ruland, Acta Crystallogr. 18, 992 (1965).R. S.McMillan (unpublished).S.Hendricks and E. Teller, J.Chem. Phys. 10, 147 (1942).G. Kirczenow, in Graphite Intercalation Compounds I, Vol. 14of Springer Series in Materials, edited by H. Zabel and S. A.Solin (Springer-Verlag, New York, 1990).J.R. Dahn, D. C. Dahn, and R. R. Haering, Solid State Com-mun. 42, 179 (1982).J.E. Fischer and H. J.Kim, Phys. Rev. B 35, 3295 (1987).S.A. Safran, Phys. Rev. Lett. 44, 937 (1980).8D. P. DiVincenzo and T. C. Koch, Phys. Rev. B 30, 7092(1984).9S. E. Millman and G. Kirczenow, Phys. Rev. 8 28, 3482(1983).

physrevb.44.9170 phase diagram of li„c6.pdf

Documents