Comparative ion transport in several polymer electrolytes

Comparative ion transport in several polymer electrolytes

JOURNALOF PgWlg ELSEV|ER SURGES Journal of Power Sources 68 { 1997 ) 372-376 Comparative ion transport in several polymer electrolytes F. Alloin *...

317KB Sizes 0 Downloads 96 Views

JOURNALOF

PgWlg ELSEV|ER

SURGES

Journal of Power Sources 68 { 1997 ) 372-376

Comparative ion transport in several polymer electrolytes F. Alloin *, D. Benrabah, J.-Y. Sanchez Laboratotre d'Electrochmue et de P h y s u o¢ himie des Matdriaux et des" lnterfa~ es, D o m a m e Universitaire B P 75, 38402 Saint-Martm-d'Hbres Cedex, France

Accepted 25 November 1996

Abstract A comparative transport number study on a wide variety of lithium salts was performed, confirming the prevalence of an anionic transport in most of the usual salt.,, dissolved in a polyether network. A catiomc transport close to 0.5 was determined for several perfluorosulfonatebased polymer electrolytes. The method used confirmed a transport number close to unity for a perfluorosulfonate-based kmomer. © 1997 Elsevier Science S.A. K e v u ords: Polymer electrolytes; Transport number; Poly ( oxyethylene ) network; LJthmm salts: lonomers

1. Introduction

Contrary to glass or ceramic electrolytes and except for ionomers, both cations and anions contribute towards the conductivity of polymer electrolytes. Therefore, in addition to the determination of their ionic conductivities and electrochemical stability range, a good method to compare the performance of polymer electrolytes consists in measuring the transport numbers and diffusion coefficients. Functional groups, i.e polar protic groups, which induce anionic solvation through hydrogen bonds, may modify the mobility of anions, cations, or both anions and cations. Protic and some aprotic groups react in presence of lithium, while others are unstable versus most of the cathode materials. The polymer matrix must be therefore accurately selected. In this contribution we will compare the lithium transport number for several salts dissolved in high molecular weight linear poly(oxyethylene) (POE) and cross-linked polyethers NPC I network poly (isobuthenyloligooxyethylene). The cross-linked polyethers were obtained by a stepgrowth polymerization between poly( ethylene glycol) 1000 and a dihalide unsaturated compound 3-chloro-2-chloromethyl- 1-propene [ 1]. The Williamson-type polycondensation proceeds [2], according to the following scheme, in NaOH or KOH solutions *

Corresponding author.

0378-7753/97/$17.00 © 1997 Elsevier Science S.A. All rights reserved PIIS0378-7753( 97 )02536-6

HC~,CH,-C. ~-%-cn,,-CH~-OH

{CH2-CHz-())~C}t2-CI L2-Ok

i

cI~,/cI,<, cII~l:

..M~(CH2-CH2-O)~CH2-CH~,-O~

sO ~

II

II

('H,

(H2

Due to the incorporation of the isobutenyl moiety, the resulting polyethers show a clear decrease in crystallinity and melting point as compared with the linear POE and even to the starting poly(ethylene glycol). The isobutenyl groups allow further free-radical cross-linking, again decreasing crystallinity and melting point when compared with the linear polycondensate. Finally, the isobutenyl moiety may copolymerize with a monomer bearing an ionophoretic function, resulting in a single cation conductor polyether network. We compared to.+, at several concentrations, for LiTFSI, (CF3SO2)2NLi, LiClO4, CF3SO3Li, TriTFSMLi, (CF3SO2)3CLi, as well as for non-bonded unsaturated salts such as a diallyl amide by-product DaaRtSO3Li, ( C H 2 = C H CH2)2NCO-CF(CF3)SO3Li, and a allyloxy by-product A1RISO3Li, CH2=CH-CH2-O-CF2-CF:,SO3Li [ 3 ]. 2. Experimental

2.1. Polycondensation reaction The linear unsaturated polycondensates were prepared in bulk, according to the typical procedure already described for

F AIIom et al./Journal of Power Sources 68 (1997) 372-376

blocking electrodes and an HP4192A impedance analyser over the frequency range 5 Hz to 13 MHz.

c~,oJ-dihydroxyoligo(oxyethylene) [1], P E G I 0 0 0 (from Aldrich) and 3-chloro-2-chloromethyl-1-propene were used in a molar stoichiometric ratio. The average tool. wt. Mw and M., were determined in tetrahydrofuran ( T H F ) by GPC, Waters 590 and a data module Waters 745B, and expressed in a polystyrene equivalent. We found M ~ = 105000, M, = 47000 and 1 = M J M , = 2.2.

2.5. Electrochemical study The transport number to,+ determination in polymer electrolytes was performed, by means of a combination of complex impedance measurements and potentiostatic polarization measurements, sandwiching the samples between two metallic lithium electrodes [2 ]. The amplitude of the impedance measurements and the polarization voltage were kept at 10 mV. Polarization measurements were carried out with a Mac Pile computer device, allowing 0.25 IxA and 1.25 mV accuracy on intensity and voltage, respectively.

2.2. Cross-linking o f the membrane The films were prepared by free-radical polymerization at 70 °C, using benzoyl peroxide as an initiator. The resulting membranes were washed several times in methanol, and dried under vacuum. The electrolyte films were obtained by swelling the films in an acetonitrile preweighted salt solution. In order to obtain a single-cation conductor, DaaRfSO~Li salt has been attached, by covalent bonding, to the unsaturated polyether by free-radical copolymerization of the salt and the polyether double bonds. The polycondensate cross-linking is therefore achieved, simultaneously with the salt grafting. In order to avoid any contamination by non-grafted salt, the membranes were thoroughly washed in methanol to eliminate residual free salt.

3. Results and discussion

3.1. Ionic conductivit3' The ionic conductivities of NPC1000-based cross-linked electrolytes with several salts were investigated. Fig. 1 presents the best conductivities obtained with these different salts. The best were obtained with LiTFSI and LiCIO4 reaching 2 X 10 -5 S c m - ~at room temperature and 10 -3 S c m - ~near 80 °C. As for CF3SO3Li, TriTFSMLi and DaaR,SO3Li showed ambient temperature conductivities (25 °C) close to 7 × 10 6 S cm ~. At higher temperature, 80 °C, TriTFSMLi appeared more conductive than CF3SO3Li and DaaRfSO3Li reaching 3 × 10 -4 S c m - ~ but remained markedly less conductive than LiTFSI complexes (factor 5). Although the conductivity study of TriTFSMLi complexes was performed in a somewhat limited salt concentration range 6 < O / L i < 33. this result was surprising and not in agreement with a previous study performed on linear POE: 5 × 106 [3] which established that L i T F S I - P O E and TriTFSMLi-POE complexes reached about the same conductivity maxima. Regarding the single-cation conductor NPC1000/DaaRtSO3Li, its Arrhe-

2.3. Thermal analysis Glass transition temperature (Tg) and melting temperature were measured in helium using a Netzsch STA409 thermal analyser. In the typical protocol, samples are first cooled from room temperature to - 120 °C, then heated again at 10 °C rain ~ to 150 °C. After this first heating cycle the samples are quenched to - 120 °C and then heated at a rate of 10 °C min 1 2.4. Ionic" conductivity The ionic conductivities of various complexes were obtained by impedance spectroscopy using stainless-steel 97 10 2

84

72

60

373

~)

40

30

20

13

!

l0



D

#



C F ~ S O 3 I J O/Ll=18



LI('lO40'LI=14 5



DaaRISO3LI lonomer O;LJ= 12

*

(OF~SO3)3('LI

#

D a a R i S O f l a hbre O,qA=12 #

,-

104

,~,

Z.

'

*



,~

~'

t ~

'

105

..



~

O'L]=I6

5

(Ct' ~SO2)2NLt 0¢L1=9 ('CF3SC)sLCHLt O,'1 u=19



b



,

~

':'.

i0~, L

10 ~ 2.6

J 2.7

I

2.8

2.9

3

3.1

3.2

3.3

3.4

3.5

3.6 3.7 3.8 1000~T ( K i ) Fig. 1 Comparison of best conductivity levels for several NPC 1000/lithium salt complexes

F. Alloin et a l . / Journal ol Power Sources 68 (1997) 372-376

374

transport numbers when the salt is dissolved in an N P C I 0 0 0 host polymer. As the single-ion conductor is prepared from NPC1000, we selected therefore to perform the transport number comparison in the NPC 1000 host polymer rather than in POE. Cationic transport number tL,* evolutions with salt concentrations are given at the same temperature ~80 °C) for different lithium salts, in Fig. 3. The lowest tL, + was obtained with LiTFSI, while tel. is about twice as high in the methide salt, TriTFSMLi, but remained very lov,. As for LiCIOa, we found tL,+ ~ 0.2, which was higher than that, 0. I 1, reported by Watanabe and Nishimoto [10]. The best results were obtained with the perfluorosulfonate salts, both unsaturated salts DaaR~SO3Li and A1R~SO3Li exhibiting ILl+ values higher than those determined for LiCF~SO3. We must emphasize the tL,- = 0.6 value obtained for DaaRtSO3Li, but also the values close to 0.5 obtained with the.' other perfluorosulfonate salts, rE,+ values are slightly sensitive to the polymer electrolyte composition, te,+ decreasing generally at high concentration. If DaaR~SO3Li provides the highest tL, + values when the salt is only dissolved in NPC1000, it allows us to reach a cationic transport number close to unity when attached to the macromolecular backbone. Indeed the experimental values, ranging between 0.92 and 0.96, are in full agreement with the theoretical values and support the effectiveness of this electrochemical method. Table 1 reports the lithium conductivities, ~r + =O'tL,+, for the different salts. The highest cationic conductivity, 2 × 10 4 S cm ~ at 80 °C, was obtained for N P C I 0 0 0 / LiCIOa complex. Surprisingly, the cationic conductivities of LiTFS1, non-bonded DaaRISO3Li and CF3SO3Li are very close at 80 °C. From this comparison we may be optimistic about the potentiality of DaaRtSO3Li, since the electrolyte composition has not been optimized. Nernst-Einstein relationship D, = cr,RT/F2C, allows the diffusion coefficient to be calculated, at 80 °C, for the different ions (Table 2 ). Obviously NPC 1000/LiCIO4 electrolyte

nius plot showed a conductivity value close to 4 × 10 7 S cm ~ at 25 °C, so one order of magnitude smaller than that observed with the same concentration of non-attached salt, dissolved in NPC 1000 network. The Arrhenius plots of NPC1000-based polymer electrolytes exhibit free volume behaviour, above room temperature ( 25 °C). This behaviour is in agreement with the differential scanning calorimetry ( D S C ) recording which shows that all the lithium salts complexes are, independently of their concentration, completely amorphous above 25 °C. T~ of TriTFSMLi complexes are very close to those of LiTFSI with, for instance, T~ = - 44.2 °C for the former, as compared with - 4 3 . 5 °C obtained for N P C I 0 0 0 / L i T F S I at the same salt concentration ( O / L i = 15). Single-ion conductor electrolytes ( NPC 1000/DaaR~SO3Li ~ exhibited very low Tg ( T~ = - 67.6 °C ). This phenomenon has often been observed with cross-linked ionomers [4,51.

3.2. Li + transport number It has been noted [6,7] that this method over-estimates a cationic transport number in the situation where aggregates or ionic associations exist in the electrolyte. The ionic association occurs in such low polarity media in accordance with spectroscopy studies [ 8,9]. Most of the experiments were carried out at 80 °C. Transport number measurements were performed on LiTFSI-based polymer electrolytes, using two host polymers, namely linear poly(oxyethylene) P O E = 5 × 106 g tool ~ and the crosslinked polycondensate, NPCI000. These electrolytes exhibited the same behaviour with a cationic transport number close to 0.1 (Fig. 2), these values were close to those determined by Watanabe and Nishimoto [ 10] in different polyether networks. A slight increase in cationic transport number is observed when decreasing the salt concentration. As observed, architectural modifications, with respect to linear crystalline POE, do not induce a modification in LiTFSI 0,2 +

•:~

NPCI000 laTFS1

A

POLSM I aTFSI

0,15 0

03 f ~

0,05

A

J

10

.

.

.

.

.

i

20

i

I

30

,

,

,

,

I

40

i

,

,

J

,

i

,

i

50

60 O/Li

Fig 2. Comparison of hthium transport number m LiTFSI salt using two different host polymers

375

F AIIom et al /Journal of Power Sourt es 68 (1997) 372 376 + 1

~

laTFSI OF SOb

--

. la('lO 4

A _

DaaRiSOfla

0,8 •

0

[..



I'nTFS M DaaRtSO la Ionomer



A1RfSOJ.t

0,6 tll i 0,4

0,2

#

,

i

,

i

, }

i

10

2O

,

i

i

i

I

.N)

40

30

O/Li

Fig 3 Lahmm transport number In several NPClO00/salt complexes Table 1 Cationic conductivity values fol several salts at 80 °C o-+(~ LiTFSIO/LI= 14 CF,SO~LI O/Li= 18 LICLOaO/LI= 14 TriTFSM O / L l = 15

T F S I anion, a l l o w us to d e t e r m i n e , f r o m the o p t i m i z e d g e o m etries, a m i n i m u m b u l k i n e s s b e t w e e n 6 and 9 A,. In a d d i t i o n , local electrostatic r e p u l s i o n s s h o u l d take place b e t w e e n the f r e e - a n i o n s a n d the ' i m m o b i l e p i l l a r s ' , as e x h i b i t e d b e l o w

'cm 'I

4.85× 10 s 6× 10 • l.SX If) 4 3 24× 10 ~ 5 7× 10 s 1.3 × 10 s

DaaR,SO~LIO/LI= 12

DaaRtSO~Ll mnomer O/LJ = 12

Table 2 Anmnlc and catlomc dfffusmn coefficient~ tor complexes NPC 1000/hthmm salt

J',)_#

-,

......

I C B O

I

C F - CF Dhmmm ( cm 2 h

I]

D.,r,mn ( Cl][12 h

I} SO 3

LITFSIO/LI=I4 CF~SO~L10/Ll- 18 LICLO40/Li=I4 Tr1TFSMLIO/Ll=15 DaaR,SO~LI O/kl = 12 DaaR,SO~L lonomer O/LI = 12

I 32",<10 s 9.76"( 10 '~ 49XI0 ~ 7 6 × 1 0 "' 1 23 ," 10 s 3 >, 10 '~

2 1×10 7 2.9× I0 s I 5x10 7 56x'10 s 1.45 X 10 ~

e x h i b i t s the h i g h e q l i t h i u m c o e f f i c i e n t with 4 × 1 0 - ~ c m z s at 80 °C, this in c o n t r a s t to N P C I 0 0 0 / L i T F S I e l e c t r o l y t e s p r e s e n t i n g the h i g h e s t a n i o n i c m o b i l i t y with an a n i o n diffusion coefficient o f 2 × 10-- v c m 2 s - i. In s i n g l e - c a t i o n c r o s s - l i n k e d n e t w o r k s the a n i o n m o b i l i t y is restricted to the s e g m e n t a l fluctations a r o u n d its equilibrium position. W e m a y therefore a s s u m e that the grafted a n i o n s b e h a v e as " i m m o b i l e pillars', a rapid c a l c u l a t i o n , taking into a c c o u n t the specific m a s s o f the electrolyte, s h o w i n g that the a v e r a g e d i s t a n c e b e t w e e n t w o grafted a n i o n s s h o u l d be close to 6 A for O / L i = 12 a n d s h o u l d not e x c e e d 10 A for O / L i = 60. In case o f a m i x t u r e o f free-salt a n d s i n g l e - c a t i o n c o n d u c t o r we m i g h t a s s u m e that, due to steric h i n d r a n c e , the " i m m o b i l e pillars" s h o u l d not f a v o u r the T F S I - a n i o n m o b i l ity. Indeed, r e c e n t a b initio c a l c u l a t i o n s p e r f o r m e d on the

Electrostattc repulsion w t t h g r a f t anion sterlc h i n d r a n c e /

,,,,o

d~O~'

~

i+

t~lO ~

A -- 4, SO~ L i

~ ~ .

3

: ~

C----O I

,%

/N\ 3

HC - - ~ H

B o t h steric h i n d r a n c e and electrostatic r e p u l s i o n s s h o u l d therefore n o t a b l y d e c r e a s e the f r e e - a n i o n s m o b i l i t y . To c h e c k the validity o f o u r a s s u m p t i o n , we p e r f o r m e d a m e a s u r e m e n t on a c r o s s - l i n k e d s i n g l e - c a t i o n n e t w o r k N P C I 0 0 0 / DaaR,SQ , Li +, m w h i c h E i T F S I was i n c o r p o r a t e d . T h e m e m b r a n e N P C 1 0 0 0 / D a a R t S O ~ L i , p r e v i o u s l y w a s h e d in m e t h a n o l to r e m o v e n o n - b o n d e d DaaRrSO3Li, has b e e n dried a n d w e i g h t e d , before s w e l l i n g in acetonitrile solution of LiTFSI. A f t e r the r e m o v a l o f a c e t o n i t r i l e the m e m b r a n e has b e e n w e i g h t e d , the w e i g h t d i f f e r e n c e a l l o w i n g the T F S I L i c o m p o s i t i o n to be calculated. T a b l e 3 g i v e s the c a t i o n i c

376

F Allom et al. / Journal of Power Sources 68 (1997) 372-376

Table 3 Transport numbers for mixed 'free' / "grafted' hthium salts Z +

DaaR~SO~Li O/Li = 12 + LtTFSI O/Ll = 8 DaaRtSO~Ll O/Li = 12 + LiTFSI O/L~ = 12 DaaRtSO~Li O/LI = 25 + LITFSI O/Lt = 17

0.119 0.12 0.113

DaaRfSO3Li c o m p l e x e s . On the o t h e r h a n d , /L,+ close to unity, has b e e n f o u n d for DaaR~SO~Li grafted in the N P C 1 0 0 0 n e t w o r k , c o n f i r m i n g the s i n g l e - i o n c o n d u c t i v t y o f t h e s e m e m b r a n e s . O t h e r t r a n s p o r t n u m b e r m e t h o d s are at p r e s e n t in p r o g r e s s to c h e c k these values.

References t r a n s p o r t n u m b e r s d e t e r m i n e d for b o t h salts. U n f o r t u n a t e l y , a l t h o u g h tL,+ v a l u e s i n c r e a s e f r o m 50 to 100%, they still r e m a i n very low a n d c l o s e to 0.1. A n i o n i c diffusion, c o n t r a r y to o u r a s s u m p t i o n , h a s not b e e n d e c r e a s e d b y the p r e s e n c e of the q m m o b i l e pillars'.

4. Conclusions T h i s p r e l i m i n a r y study a l l o w e d an ion t r a n s p o r t c o m p a r i son to be p e r f o r m e d in a wide variety o f l i t h i u m salts, u s i n g a p o l y ( o x y e t h y l e n e ) n e t w o r k as the host p o l y m e r . A prevalent a n i o n i c t r a n s p o r t h a s b e e n f o u n d for m o s t o f the l i t h i u m salts, n o t a b l y for L i T F S I a n d L i T r i T F S M . O n the contrary, p e r f l u o r o s u l f o n a t e - b a s e d salts e x h i b i t e d tL,- v a l u e s close to 0.5, a n d the h i g h tL,+ value, close to 0.6, c a l c u l a t e d for the

[ 1 ] F Allom, J.-Y Sanchez and M Armand, J Electrochem. Soc , 141 (1994) 1915 121 J. Evans, C.A. Vmcent and P.G. Bruce, Polymer, 28 (1987) 2324 13] D. Benrabah, D. Baril, J.-Y. Sanchez, M. Armand and G. Gard. J. Chem. Soc. Faraday Tran,~., 89 (1993) 355. [4J D Benrabah, S Sylla, F Allom. J.-Y. Sauchez and M Armand, Electroclum. Acta, 40 ( 1995 ) 2259. 151 D Benrabah, S. Sylla, J.-Y Sanchez and M Armand, J Power Sotqrees, 54 { 1995) 456. [ 6] P.G Bruce and C A Vincent, Faraday Dtscus.~ Chem. So¢., 8 ( 1989 ) 43 171 G.G. Cameron, J.-L. Harvie and M.D. Ingram, Solid State lont¢s, 34 (1989) 65. l 8] G Petersen, L M Torell, S. Panero, B Scrosah, C.J Da SiIwt and M. Smith, Solid State lomcs, 60 11993 ) 55. [9] M Kaklhana, S Schantz. L.M. Torell and .I.R. Stevens. Sohd State lomcs, 40/41 (1990) 641. [ 10] M Watanabe and A. Nlshimoto. Solid State Iome~, 79 11995 ) 306.