Vol. 43, No. 9, pp. 1105-1107, 1998 :q 1998 ElsevierScienceLtd. All rights reserved Printed in Great Britain
ool 3m686/98 $t9.00 + 0.00
PRELIMINARY NOTE A lithium ion polymer battery G. B. Appetecchi and B. Scrosati* Dipartimento di Chimica, Universit/t '"La Sapienza", 00185 Rome, Italy
(Received 25 July 1997; in revised form 17 September 1997)
Abstrac~ With the final goal of developing lithium ion batteries having a plastic configuration, we have prepared and characterized highly conducting polymer electrolytes, as well as graphite anode and chromium-stabilized manganese spinel film electrodes. Laboratory prototypes of lithium ion batteries formed by the direct lamination of these electrode and electrolyte componen:s were assembled and tested. The preliminary results of this work are presented and discussed. ,~-) 1998 EL,;evier Science Ltd. All rights reserved
INTRODUCTION Lithium technology is the most promising for assuring the proper development of advanced, high energy batteries for the portable electronic consumer market and for electric vehicle operation [1,2]. Indeed, a first generation of lithium batteries exploiting the rocking-chair concept [3, 4] and generally named "lithium-ion" batteries  is already a commercial success. Lithium ion batteries are presently produced at a rate of several millions units per month, since they are rapidly replacing the bulkier and less energetic nickel-cadmium and nickel metal hydride batteries in popular portable devices such as cellular phones and computers. In addition, lithium ion batteries are scaled-up in view of their use in electric vehicles . The next important step in lithium battery techology will be the production of prototypes having a full plastic configuration which assures modularity in design and reduction in fabrication costs. Many attempts to reach this goal are presently underway. The main requirements for a successful result are: (i) the availability of polymer electrolyte membranes having lithium ion conductivity approaching that of common liquid electrolytes; and (ii) the optimisation and control of the electrode/polymer electrolyte interfacial properties. The *Author to whom correspondence should be addressed. E-mail: scrosati(a axrma.uniroma I .it
first requirercent has been satisfactorily achieved and today many types of highly conductive polymer electrolytes have been developed and characterised 117,8]. The remaining problem is linked to the second requirement since difficulties are still met in assuring smooth interfacial contacts. In this paper we report preliminary, yet promising, results obtained by cells formed by the direct lamination of the components, namely a graphite anode film, a gel-type polymer electrolyte membrane and a Crstabilised lithiJm manganese spinel cathode film. EXPERIMENTAL The gel-type, electrolyte membrane was prepared by immobilising a lithium liquid solution in a PAN matrix. The preparation procedure was that commonly used irk our laboratory for the synthesis of PAN-based membranes . Basically', the membrane was prej~ared by dissolving LiClOa (1 M) in a ethylene carbonate-diethyl carbonate, E C - D E C (2:1 w~lume ratio) mixture, stirring at 50'(2 until complete dissolution. Poly(acrylonitrile) PAN is then added and the temperature slowly increased to 90100"C to achi,,~ve gelification. Dimensionally stable membranes, or thickness ranging between 100 and 2(10 ~m, were typically obtained by cast between two glass sheels. For reasons of simplicity, the membranes are hereafter indi,zated by writing in sequence the lithium salt, the liquid solvent mixture and the
immobilising polymer, i.e. LiCIO4-EC-DEC-PAN. The composition of the membrane is (in molar ratio) 4.5:53.5:23:19, respectively. The stable (in excess of two months as tested in closed cells) room temperature conductivity of the membrane is of the order of 4 × 10-3 Scm -! and its electrochemical stability exceeds 4.7 V vs. Li. Both these values are appropriate for allowing the use of the membrane as separator in high rate and high voltage lithium ion batteris . The anode film electrode was prepared by spreading on a copper foil current collector a mixture of graphite (Lonza KS 44) and a polylvinyl chloride) poly(tetra ethylene fluoride) P V C - P T F E binder, using tetrahydrofuran as the carrier solvent. The cathode film was similarly formed by spreading a mixture of the active component (a Crstabilised lithium manganese spinel, of formal LiCr0.o2MNI.9804 composition) carbon and the PVC PTFE binder on an aluminium foil current collector. The electrode characteristics were investigated in three electrode cells using a lithium counter and lithium reference electrodes. An EG&G P.A.R. mod. 362 potentiostat/galvanostat, was used for the measurements. The lithium ion battery prototypes were assembled in an environmentally controlled dry box by pressing into appropriate coin-type containers a laminated cell formed by the graphite anode, the electrolyte membrane and the spinel film cathode. The cells were sealed and tested using a Maccor cycling equipment. RESULTS AND DISCUSSION As is well known [2,5], a lithium ion battery implies in its basic concept the use of two Li-intercalation electrodes, i.e. a carbonaceous (e.g. graphite) LixC6 anode and a lithium metal oxide (e.g. Litl _x)Mn204 spinel) cathode, separated by a Li + ion conducting liquid solution (e.g. a solution of LiC104 in an organic E C - D E C liquid solvent mixture). The electrochemical process is the cycling transfer of lithium ions from one electrode to the other. The replacement of the liquid solution with a polymer electrolyte is not straightforward. The challenging requirement is to achieve smooth interfacial contact, with particular concern to the negative, carbon electrode side. The major problem is to assure full wettability throughout the entire electrode bulk. A successful approach has been reported by the Bellcore Laboratory in the United States, where a multistep technology based on the use of a membrane electrolyte formed as a copolymer of vinylidene fluoride and hexafluoropropylene ( P V D F HFP), which is capable of absorbing large quantities of liquid electrolyte, is presently under development [9, 10].
An alternative process is that which involves the direct lamination of the battery components. The validity of this approach has been demonstrated by K. M. Abraham et al. [11, 12]. We have further explored this approach by considering the development of batteries formed by laminating in sequence a grahite anode film, a LiC104 E C - D E C - P A N geltype membrane electrolyte separator and a chromium-stabilized manganese spinel cathode film. To test the validity of this battery design, we have first evaluated the electrochemical response of tlhe graphite film electrode by monitoring its poteatial capacity response during the lithium intercalation-deintercalation process (i.e. xLi ~ +6C<=~ Li×C6) in a LiCIO4-EC-DEC PAN elect::olyte cell. Figure 1 illustrates the result. The response is similar to that expected in liquid electrolyte cells: the voltage profile of the Li-intercalation process decreases along a series of distinguishable plateaus which correspond to the progressive staging process and a similar trend is reproduced in the following Li-deintercalation process, with a total cyclable capacity approaching the theorectical 372 raAhg -I (x = l) value . As is typical in liquid electrolyte cells, in this case there is a formal excess in capcity in the course of the initial charging cyles, this being due to the decomposition of the liquid organic solvents with the formation of a passivation film (mainly formed by a Li2CO3 layer) on the graphite electrode surface. The occurrence of this film is essential for assuring cyclability and stability to the electrode . In fact, the clharge-discharge efficiency of the graphite film electrode approaches 100% after the second cycle and remains at this value for all the subsequent cycling. Onze having ascertained the compatability of the graphite film electrode, we have used the LiCIO4E C - D E C - P A N polymer electrolyte for the fabrication and test of a complete Li-ion battery using a LiCr002Mnl.g804 film as the cathod. The favourable response of this Cr-stabilized manganese spinel elec3.0 li
': i ----" "°
I---- rch..,, l i ......... ! .......... :'r
~ 1.5 =~
2 = DL,~.~, / ...... ~'c'.'"
,o :I o.s O0
............... ;..........,, 600 800 1000
Specific capacity [mAhg"l] Fig. l. Voltage vs. capacity response of a Li×C6 electrode in a I_iCIO4-EC-DMC-PAN polymer electrolyte cell. Counter electrode: Li. Initial cycling. Cycling rate: C/10. Room ;emperature.
• 1 ~ . - - ~ . . . . . . . . . . . . . . ... . . . . . ... . . . . . ... . ~,Oe
2o . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i .. . . . . . ... . . !
Discharge . .
: • 'v- ,,r ~++o+.o + ~. . . . . . . . . . . + 'q' w t I I i I Y v v v • ~
The author~; would like to thank Prof. M. Wakihara o:~ Tokyo Institute of Technology in Japan and Covalent Associates (USA) for having kindly provided the Cr-stabilized manganese spinel samples. Financial support from NEDO (International Joint Research Program), Japan and the C.N.R., Italy (Progetto Strategico "Batterie Leggere", Contract N. 96.01008.ST74 is also acknowledged.
Cycle number Fig. 2. Percent of graphite full capacity versus cycle number for a C/LiCIO4-EC-DEC PAN/LiCr0.02Mn]+9804 plastic coin type lithium-ion battery. Cycling rate: C/20. Room temperature. trode in PAN-based polymer electrolyte cells was demonstrated in a previous paper . The expected electrochemical process of the Liion polymer battery is the cycling transfer of lithium ions from the manganese spinel electrode to the graphite electrode: 6C + LiCro+o2Mnl.9804 ,~ LixC6+LiO-x)LiCro.02Mnl.9804 Fig. 2 shows the cycling behaviour of the battery reported in terms of percent of graphite full capacity versus cycle number. Although with some progressive decay, probably due to some residual disuniformities in the electrode and electrolyte formulation, the cell delivers a high fraction of the maximum graphite capacity. Much higher cycling stability is expected following optimization of the electrode configuration and structure. The promising results reported here suggest that PAN-based, gel-type polymer electrolyte membranes may indeed be suitable for the fabrication of plastic lithium ion batteries for practical use. This expectation is presently under further test in our laboratory.
1. B. Scrosati, Nature 373, 557 (1995). 2. B. Scrosati, Chim. & Ind. 79, 463 (1997). 3. B. DiPietra, M. Patriarca anti B. Scrosati+ J. Power Sources 8, 289 (1982). 4. B. Scrosati, J. Electrochem. So,'+'. 139, 2276 (1992). 5. S. Megahed and B. Scrosati, J. Power Sources 8, 289 (1982). 6. Y. Nishi and K. Katayama 8th Intl. Meeting on Lithium Batteries, IMLB-9. Nagoya, Japan, June 1996, Abstr. MON-04. 7. K. M. Abraham, in Applications of Conductive Polymers, (Edited by B+ Scrosati) 75, Chapman & Hall, London (1993). 8. G. B. Appetecchi, F. Croce, G. Dautzenber, F. Gerace, S. Panero, F. Ronci, E. Spila and B. Scrosati, Gazz. Chint. It. 126, 415 (1996). 9. J-M. Tarascon, A. S. Gozdz, C. Schmutz, F. Shokoohi and P. C. Warren, Solid State lonics 86--88, 49 (1996). 10. F. Shokoohi, P. C. Warren. S. J. Greaney, J-M. Tarascon, A. S. Gozdz and G. C. Amatucci, Proceeding~ 35th Power Sources Conference, Cherry Hill, New Jersey, (June 1996). I1. M. Alamgir and K. M. Abraham, J. Power Sources 54, 40 (1995). 12. Z. Jiang and K. M. Abrahara, J. Electrochem. Soc. 142, 333 (1995). 13. G. B. Appetecchi and B. Scrosati, J. Electrochem. Soc. 144, 138 (1997).