Lithium Batteries

The exponential growth in portable electronic devices such as cellular phones and laptop computers during the past decade has created enormous interest in compact, light-weight batteries offering high energy densities. Also, growing environmental concerns around the globe are driving the development of advanced batteries for electric vehicles. Lithium-ion batteries are appealing for these applications as they provide higher energy density compared to the other rechargeable battery systems such as lead-acid, nickel-cadmium, and nickel-metal hydride batteries as shown in Figure l.l.¹ The higher volumetric and gravimetric energy densities of the lithium-ion cells in Figure 1.1 are due to the higher cell voltages (~ 4 V) achievable by the use of non-aqueous electrolytes, which also allow a wider temperature of operation. Lithium-ion cells have become a commercial reality after the initial announcement by Sony in the early 1990s because of an intense world-wide activity on lithium insertion compounds (electrode materials) during the past three decades. This chapter provides an overview of the electrode materials aspects of rechargeable lithium batteries. A more detailed discussion of the specific cathode and anode materials systems as well as the electrolytes is presented by various authors in the subsequent chapters.

Lithium transition metal oxides are remarkable intercalation compounds, exhibiting an array of intriguing electronic and phase transformation phenomena. Their ability to undergo large variations in lithium concentration, often without suffering irreversible changes, makes these materials ideal electrodes for rechargeable lithium batteries. Lithium transition metal oxides consist of a metal oxide host with a crystal structure in which lithium ions occupy a relatively open network of interstitial sites. While some lithium transition metal oxides can serve as an anode in rechargeable lithium batteries, most are more suited for the role of cathode as they typically exhibit a high voltage with respect to a metallic lithium anode, the ideal theoretical reference electrode. In a rechargeable lithium battery, lithium ions are shuttled between an anode and a cathode, whereby lithium ions are removed from and inserted into the electrodes. During deintercalation of a lithium transition metal oxide, vacancies are created on the lithium sites of the host. During intercalation, these sites are refilled by lithium. Removal and insertion of lithium ions can lead to several phenomena that can significantly affect the electrochemical properties of the compound. For one, variations in lithium concentration alter the electronic properties of the transition metal oxide host. The valence electron of each lithium ion is generally donated to the host where it can either shift the valence state of the transition metal ion and/or alter the nature of the bonds between the transition metal and the oxygen ions. Simultaneously, lithium removal from the host may structurally destabilize the metal oxide structure or may lead to order-disorder phase transitions between lithium and vacancies once a critical vacancy concentration is reached. These phenomena often affect the voltage characteristic or the lattice parameters of the compound in important ways.

Within the last decade, much attention has been devoted to understanding the properties of Li _x Co0₂, Li_xNi0₂, Li _x Mn0₂, Li _x Mn₂0₄ ¹ and Li _x FeP0₄ ², currently among the most important candidate cathode materials for rechargeable lithium batteries. Other compounds receiving attention are typically doped variants of the Co, Ni and Mn compounds. While many lithium transition metal oxides have similar crystal structures, either a layered form or one derived from the spinel structure (Li _x FeP0₄ has the olivine crystal structure), they are often characterized by very different electrochemical properties owing to the unique electronic structure of each transition metal. Li _x Co0₂, for example undergoes a concentration driven metal-insulator phase transformation whereby the electrons over much of the lithium concentration range (x>0.75) are delocalized and exhibit metallic properties.3 In contrast, the Li _x Mn₂0₄ compound in important lithium concentration regions is characterized by more localized electrons with large magnetic moments leading to phenomena such as charge ordering and Jahn-Teller distortions.^4–6 Often, phenomena such as these can be rationalized with crude crystal-field and molecular orbital models. More subtle properties such as the relative stability of similar crystal structures, stable ordered lithium-vacancy arrangements and charge-ordered and magnetic-ordered electronic states are more difficult to predict or rationalize with intuition based on simple electronic structure models. It is here that first-principles numerical-schemes for solving the Schrödinger equation of solids are proving invaluable. The record of the last decade of first principles studies of lithium transition metal oxides has demonstrated that these methods are an important new tool in predicting and understanding the properties of these materials.^7–40

Synthesis is critical not only to the generation of new materials for exploration of new structures and fundamental properties, but also for the formation of materials with the optimum electrochemical behavior for commercial devices. The technique used is quite often very different for the above two cases. The initial synthesis of a material should provide it in a pure enough state that its intrinsic behavior and properties can be determined; it should for example be possible to accurately determine whether its structure is cubic spinel, layered or some other form. There is presently much confusion in the literature because the synthesized material was insufficiently well-characterized. Once well characterized, the material needs to be synthesized in a form suitable to be used in a battery, for example it may be doped or coated to enhance the conductivity, its particle size and morphology will be optimized for maximum reactivity but minimum corrosivity and side-reactions.

Many battery materials are metastable phases, and therefore non-traditional synthesis methods must be devised to allow kinetics to over-ride thermodynamics. Hence, many soft chemistry techniques have come to the fore such as hydrothermal, ion-exchange, intercalation, etc. In addition, to optimize the formation of the desired material frequently the reactants are finely mixed prior to final reaction by for example sol-gel formation or co-precipitation as hydroxides.

The past years have witnessed significant improvements in the nature and capability of energy storage systems. A combination of fundamental and technologically driven studies were responsible for these advances, which have been reviewed in Chapter 1. Motivation for the advancement has been inspired by the ever-increasing demands that a multitude of applications are placing on the energy storage battery. In addition to the ever-popular needs for portable energy electronics, future demands lie in rechargeable batteries for hybrid electric vehicles, miniaturized electronics; space exploration, uninterrupted power supplies, and medical devices,^1,2,3 These various applications differ in their requirements of the energy storage cell: some needing high power, and others needing high capacity. All demand safety, however, and most require excellent stability of the cell over long-term usage. As fuel cells are still many years away from meeting the needs in most of these areas, lithium-ion rechargeable batteries offer the only technological solution at present, and the best long-term solution for the foreseeable future in many of the areas. Developments in the positive electrode arena have produced materials capable of gravimetric energy densities in the regime of 200 mAh/g; therefore anode materials are sought to match these high-capacity cathodes. This chapter is devoted to a review of the new technologies in the field that may address these issues.

In rechargeable lithium ion batteries the main¹ application of carbon is the use as a host anode accessible by lithium cations via electrochemical insertion, the lithium being provided by another insertion material used as host cathode (Figure 5.1 A). This “electro-insertion” reaction is basically a host/guest solid state redox reaction involving electrochemical charge transfer coupled with insertion of mobile guest ions (in lithium ion batteries, these are lithium ions) into the structure of a solid host, which is a mixed electronic and ionic conductor. The major structural features of the host are kept after the insertion of the guests. A common convention in the literature is that “intercalation” is regarded as a special case of “insertion”. The term “intercalation” implies the restricting condition that a layered host takes up guests within its interlayer gaps (“galleries”), which may result in a volume change perpendicular to the layers, but which causes no other structural changes. However, according to this strict definition, even graphite would be not a pure intercalation host, as during Li accomodation the stacking sequence changes by sliding of the graphene layers. Therefore, in this review the term intercalation will be used where the historical conventions or the present practice make its use appropriate, i.e.,we will use the term lithium-graphite intercalation compounds for lithiated graphites (cf. Refs. 1–16).

The recent commercialization of advanced lithium batteries is mainly due to the breakthrough in the stabilization of the anode-electrolyte interface. Although metallic lithium has an energy density (3860 mAh/g) higher than that of other alternative anodes, its poor performance and safety issues related to the low melting point of lithium (180 °C), dendritic growth during lithium deposition (charge), and high reactivity toward the electrolytes have hindered the commercialization of rechargeable lithium-anode batteries.

There have been several approaches to solve the problem of lithium anode. The development of lithium alloys, particularly the binary and ternary alloys, has received considerable attention (see Chapter 9 of this book). However, the performance of these alloys is unsatisfactory mainly due to the large volume change (100–200%) during lithiation and delithiation processes. This expansion and contraction processes may cause the alloy particles to crack and lose contact with the electrode substrate. In addition, the lithium alloys with high concentration of lithium are very reactive toward the electrolytes and cause decomposition. Therefore a problem similar to that of the metallic lithium also exists for Li-alloy anodes. Recently, some success has been made using intermetallic alloys such as Cu₆Sn₅ that insert lithium topotactically over a wide composition range Li _x Cu₆Sn₅ (0<x<13).¹ Despite the good volumetric energy density of these alloys, their gravimetric energy density is poor and there is a significant capacity loss during multiple cycling.

In response to the needs of today's mobile society and the emergence of ecological concerns such as global warming, one of the major technological challenges in this new century is undoubtedly energy generation and storage. Ninety percent of today's electrical power generation still comes from fossil fuels, and we are constantly struggling to reduce the carbon dioxide emissions per unit of electric power so as to help curtail global warming. It is now mandatory that new and environmentally friendly energy/storage sources be found. Hence, the fast developing research in that field involving, among others, fuel cells, primary and rechargeable batteries, and supercapacitors. As a result of this worldwide ecological priority, political concerns have come into play, and science has suffered from prioritisation based on both industrial pressure and media reports, rather than on the clear and rigorous scientific identification of technological stoppers inherent in each storage system. Needless to say, this applies to battery systems as well.

In the past two decades, intensive efforts have given birth to the rechargeable Li-ion battery technology that has dominated the market place, and can be regarded as one of the great successes in modern electrochemistry to date. But these Li-based systems still suffer from the lack of suitable electrode and electrolyte materials, which they require if they are ever to accommodate the increasing user's demands. Aware of this limitation, chemists have been acting at several levels to incrementally improve the Li-ion performance. They have followed a dual approach, dealing with either positive or negative electrode materials, with efforts centered around: 1) the modification of existing materials through cationic/anionic substitution, texture modification and surface treatments, 2) the making of composite electrodes or electrolytes made of several chemical components, and 3) the design of new electrode materials. Such approaches were pursued at the macroscopic scale on electrode materials^1–3 having a dual electronic-ionic conductivity, a void structure to insert/de-insert Li ions, or the ability to alloy with Li. They led to the identification of layered LiMn_1−x Cr _x 0₂ oxides^4–5 or three-dimensional iron phosphates (LiFeP0₄)⁶, that stand as a possible alternative to LiCo0₂ or negative electrode materials such as tin-based oxides (Sn0₂, SnO),^7–8 intermetallics (CuSb⁹, Cu₆Sn₅ ¹⁰, ...), nitrides¹¹ and phosphides,^12'13 which could be used as alternatives to carbonaceous materials, once their initial large irreversibility and poor cycle life have been overcome.

The search for negative electrode materials which exhibit improved electrochemical properties relative to the widely used graphite intercalation materials has lead to a renaissance of research on inorganic negative electrodes.

The intercalation of Li⁺ into graphite follows the basic topotactic insertion process:

Early work on the commercial development of rechargeable lithium batteries to operate at or near ambient temperatures involved the use of elemental lithium as the negative electrode reactant. Binary phases, generally involving a solid solution of lithium in one of the forms of carbon, are currently employed on the negative side of lithium cells.

There is considerable interest in finding alternative materials that might be more attractive than the lithium-carbons. Improvements might involve the ability to operate safely at higher current densities, less first cycle irreversible capacity loss, better cycling behavior, reduced specific volume, and lower cost.

Metallic lithium has the highest theoretical specific capacity (3860 mAh/g) and the most negative redox potential among all metals. These features have attracted the interest of battery investigators, who have attempted to put lithium metal into practical use in rechargeable battery systems. Its poor charge-discharge cycleability, however, and its potential fire hazards have hindered thus far the development of cells with a metallic lithium anode.¹

This chapter will not provide an exhaustive review of the work on cathodes for rechargeable Li batteries, but will try to focus on the most significant recent advances in both fundamentals and applications.

Commercial Li-ion batteries were successfully introduced into the market by Sony in 1991.¹ They used LiCo0₂ as a cathode and more than a decade later this is still the cathode of choice. Nonetheless, the research on cathode materials has been more intense than ever and remarkable success has been met both at the practical and at the theoretical level. Improving the LiCoCO₂ synthesis has contributed to double the specific energy of commercial batteries in 10 years: from 80 to 165 Wh/kg. A new cathode, LiMn₂0₄, has been introduced by NEC^2–l4a in 1996 and Sanyo^4b in 2001, although for a limited market. Some other materials, derived from LiCo0₂, e.g. LiNi_1-x-y Co _x M _y 0₂ (x+y<0.25, M=Mg or AI, preferably), have become serious candidates as future cathodes.

Secondary lithium batteries have long been studied because they exhibit the highest specific energy among all rechargeable batteries. Nevertheless, the limited safety and rechargeability associated with the use of metallic lithium has prevented their widespread acceptance in the market. Unsatisfactory results with lithium metal have directed the research towards the development of the so called “rocking-chair” or “lithium-ion” batteries (LIB), based on a transition metal oxide cathode and a carbon anode.^1–3 In 1990, Sony Energytec Inc. commercialized this type of batteries.³ Today, lithium ion batteries are used in cellular phones, notebook-size personal computers, video cameras, and, to a limited extent, in electric vehicle (EV) applications. Most commonly, the negative electrode acts as a “lithium sink” and the positive LiA_zB_y electrode acts as a “lithium source”. Layered lithium transition metal oxides and spinel lithium manganese oxide have been selected as preferred cathode materials for lithium-ion batteries. Table 12.1 summarizes the cathode materials that have been, or possibly will be, used in LIB. LiCo0₂ is the cathode of choice for small-size batteries. In view of the economic and environmental advantages, spinel-structured manganese oxides have demonstrated to be the most promising positive electrode materials for large-size batteries for EV and industrial applications, as they are cheaper, less toxic, and show higher safety.

The first commercial lithium-ion batteries used LiCo0₂ as the cathode active material, and this material continues to be used in most lithium-ion batteries manufactured despite the high cost and safety hazards associated with cobalt. Apart from LiCo0₂, only the isostructural nickelate LiNi0₂, and more particularly the Co-substituted nickelate LiCo _x Ni_1−x 0₂, have been considered to have sufficient energy density and cycling stability to be of commercial interest. However the nickelates present safety and toxicity concerns which are still greater than LiCo0₂. Manganese oxides offer lower cost and toxicity than cobalt or nickel, and have been demonstrated to be safer on overcharge. A lithium manganese oxide based cathode should therefore, at least in principle, provide significant technological advantages in a lithium-ion system over LiCo0₂, LiNi0₂, or LiCo _x Ni_1−x 0₂.

One of the reasons that LiCo0₂ functions so well as a lithium-ion cathode material is that it has a well-ordered and stable layered crystal structure which is easily prepared and handled in air, and enables a fast and reversible lithium intercalation. Nickelates also have a layered crystal structure, although the synthesis conditions are more difficult, requiring calcination and sintering under a controlled oxygen-rich atmosphere. The most readily prepared lithium manganese oxide is LiMn₂0₄, which does not have a layered crystal structure but a spinel structure (refer previous chapters in this volume). Although the spinel crystal structure of LiMn₂0₄ permits rapid intercalation of lithium ions, its stoichiometry means that it has a lower capacity than LiCo0₂. Furthermore, the stoichiometric spinel LiMn₂0₄ shows a large capacity fade with cycling. Lithium-rich spinel compositions Li[Li _x Mn_2−x .]0₄ or materials substituted with cations such as Co, Cr, Al,

Li-ion cells consisting of LiMe0₂ (Me: a 3d-transition metal element) and carbon materials have been of interest because of their capability to be safely operated for thousands of cycles whilst retaining a high energy density. Materials with the above formula more extensively examined for positive electrodes include LiCo0₂, LiNi0₂, LiCoi_1-x Ni _x O₂ and LiMn0₂. Among these, more research has been done on LiCo0₂ because of its high energy density and good cycling performance. The electrochemical properties of cathodes based on LiCo0₂ are shown in Table 14.1.^1–5 LiCo0₂ has served as an archetypal cathode material for secondary Li batteries ever since the discovery by Mizushima et al ⁶ that Li can be reversibly removed (deintercalated) from and reinserted (intercalated) into Li _x Co0₂. The layered form of LiCoO₂, which has a rhombohedral symmetry belonging to the space group R3m, is ideally suited to accommodate large changes of the Li content, x. This crystal structure consists of close-packed oxygen layers stacked in an ABCABC sequence with Co and Li ions residing in octahedral sites in alternating layers between the oxygen planes.⁷ Figure 14.1 illustrates the crystal structure of LiCo0₂. As the Li concentration is changed in Li _x Co0₂, vacancies are either created or filled within the Li planes.

Over the last 15 years the vast majority of fundamental and technological contributions on the search for better positive electrode materials has been devoted to transition metal oxides such as Li _x M0₂ (M= Co, Ni, Mn), Li _x Mn₂0₄, Li _x V₂0₅ or Li_xV₃O₈.1 The first two classes of materials, built on close-packed oxygen stacking adopt bi-dimensional and tridimensional crystal structures, respectively. Lithium ions may be easily intercalated or extracted from these structures in a reversible manner. These oxides are reasonably good ionic and electronic conductors and lithium insertion/extraction proceeds while operating on the M⁴⁺/M³⁺ redox, located at around 4V vs. Li+/Li. Cost considerations brought some special attention to the possible use of LiFe0₂ that may be prepared in the same crystallographic arrangement as of LiCo0₂ through ion exchange from α-NaFe0₂.2 Electrochemical extraction of lithium would give access to the Fe⁴⁺/Fe³⁺ redox couple but no stability of Li_1−x Fe0₂ was successfully demonstrated yet.

The peculiar structure of the two-dimensional oxides Li_1−x M0₂ (M= Co, Ni, Fe, Mn) lead to structural instabilities when the number x of extracted lithium is high (end of charge). Irreversible motion of transition metals within the lithium layers may occur and lead to important capacity loss on cycling. In practical use, for instance, only 150 mAh/g out of the theoretical 273 mAh/g is used for LiCo0₂ in commercial Li-ion batteries.

Downsizing of electronic components demands energy storage systems with high energy and power densities, and thus continuously drives the research and development efforts. The most commonly used batteries in portable computers and cellular phones are lithium rechargeable batteries¹. In these batteries, chemical energy stored in the positive electrode is released and converted into electrical energy during discharge through an intercalation (insertion) process by which lithium ions are incorporated within the host structure of the positive electrode materials. During charge, the process is reversed (de-intercalation) and electric energy is applied to remove lithium ions from the positive electrode. Lithium transition metal oxides that host mobile lithium ions on the interstitial sites have been studied as positive electrode materials and the intercalation and de-intercalation processes of lithium ions are accompanied by redox of transition metal ions. The energy output of lithium rechargeable batteries is dependent on the voltage upon which lithium ions are inserted and the number of interstitial sites that can accommodate lithium ions in the host structure. Ideally, intercalation and de-intercalation of lithium ions and redox of transition metal ions, should leave the host structure intact. In practice, however, variation in lithium contents in the host structure leads to lattice expansion or contraction, migration of transition metal ions and local or global lattice distortion as a result of ordering of lithium and vacancies or Jahn-Teller distortion of transition metal ions. These structural changes can 1) induce stresses and strains within lithium transition metal oxide crystals, 2) affect the electronic conductivities of lithium battery materials, and 3) lead to irreversible phase changes and thus decrease in energy outputs of lithium batteries. Therefore, much research has been focused on characterizing structural changes of lithium transition metal oxides associated with lithium intercalation and de-intercalation, and modifying material chemistry to suppress phase transformations and to optimize lithium battery performance.

The basic requirements of a suitable electrolyte for electrochemical devices are high ionic conductivity, low melting and high boiling points, chemical and electrochemical stability, and safety. Electrolyte conductivity and electrochemical stability are key parameters in selecting an electrolyte for modern electrochemical devices such as advanced batteries, fuel cells, super-capacitors, sensors, and electrochromic displays. These parameters, conductivity and electrochemical stability, will receive particular attention in this chapter. Although progress has been made in enhancing the conductivity of solid electrolytes, particularly the polymeric ones, liquid electrolytes are still used in most electrochemical systems. The solvent properties, and dynamics of ion solvent interactions, must be understood in designing new electrolytes. In this chapter, a short but general introduction to properties of solvents and ion-solvent dynamics is discussed.

The history of electrolyte development goes as far back as the work of Greek philosophers in search for a universal solvent, the so-called “Alkahest”. In search of Alkahest, many solvents and chemical rules were discovered such as “like dissolves like” (similia similibus solvuntur) as shown in Table 17.1. Later, the theory of osmotic pressure by van't Hoff (1852–1911), and the theory of electrolyte dissociation by Arrhenius (1859– 1927) were discovered. Many speculations about the nature of solute-solvent interactions and the influence of solvent media on the rate of chemical reaction were proposed in the early eighteen-century. The role of solvents on chemical equilibrium, on tautomerism (i.e. keto-enol tautomerism), and the phenomenon of solvatochromism (shift of UV/Vis absorption bands due to the changes of the index of refraction) were discovered.^1,2 Scheibe et al. have correlated the solvating ability of solvents to their degree of influence on reaction rate, chemical equilibrium, and shift in absorption spectra.³

The major families of this class are ethers, esters, and alkyl carbonates, which are highly important for the field of high energy density batteries. These solvents and their formulas are summarized in Figure 18.1. Other important solvents are acetonitrile (AN), dimethylsulfoxide (DMSO), N-N dimethyl formamide (DMF), methylene chloride and nitromethane. The latter five solvents are mainly important for electroanalytical use and organic synthesis. As discussed later in this chapter, the type of salt plays a major role in determining the electrochemical window of a polar aprotic system. Commonly used anions, in conjunction with Li⁺, are PF₆ ⁻, CIO₄ ⁻ and BF₄ ⁻. All polar aprotic solutions may contain atmospheric contaminants such as O₂, H₂0 and C0₂, which are reactive and may play an important role in determining the electrochemical behavior.

Ionically conducting polymers have been the focus of much fundamental and applied research for many years. Polyelectrolyte membranes have found significant technological use in the production of chlorine and caustic soda,¹ as separators in fuel cells^2,3 and in electrodialysis^4,5 for example. The discovery of ionic conductivity in polyethylene oxide solutions of alkali metal salts^6,7 led the way for the introduction of polymer electrolyte in devices such as lithium batteries and electrochromic windows.8,9 Since those early days many books,^10–13 book chapters⁸ and reviews^9,14–19 have been published on these materials and the reader is referred to these for more detailed information.

In recent years there have been important advances in the stability, safety, and performance of lithium and lithium-ion batteries. Many of the electrolyte materials being examined are based on organic liquids or polymers, although solid inorganic electrolytes still have an important role for a variety of applications. There are numerous reports of new compositions, advanced synthesis routes, and novel cell architecture for the glass and ceramic electrolytes.

For any rechargeable lithium battery, the electrolyte material must permit the repeated and rapid transfer of Li⁺ ions between the anode and cathode over the expected range of operating conditions (voltage, temperature, and current), without significant deterioration. Additionally, the ideal electrolyte material would be an electronic insulator, ultra-thin, lightweight, free of hazards and inexpensive. The inorganic solid electrolytes offer both advantages and disadvantages over liquid and organic polymer electrolytes. For the required rapid transport of Li⁺ across the electrolyte, the product of the resistivity and electrolyte thickness must be minimized. Typical room temperature conductivities are 10⁻¹ S/cm for liquids, 10⁻² S/cm for superionic conductors such as β-alumina, 10⁻³ to 10⁻⁶ S/cm for various gel and solvent-free (dry) polymers, and 10⁻⁴ to 10⁻⁸ S/cm for typical glass and ceramic solid electrolytes. If formed as a very thin film of less than about 1 μm, even a rather resistive material may compete favorably when compared to a much thicker cast polymer or liquid-filled porous separator membrane. An ultra-thin electrolyte also provides a considerable savings in terms of volume and mass for the battery, if not offset by the need for a thick inactive support material. It comes as a surprise to many that 1–10 μm thick glass and ceramic sheets are quite flexible.

So called “batteries” are generally sized in the range from one to hundreds of kWh, addressing a market roughly divided in two segments: transportation and stationary applications. This market is mainly addressed by lead-acid and NiCd systems. Large special batteries for space and defense generally use more advanced and sophisticated chemistries, such as nickel-hydrogen, silver-zinc, silver-aluminum oxide, etc. One of the most exciting questions in this battery business today is: will the “Li-ion revolution” which occurred in portable batteries extend to larger battery systems? Obviously, that would only occur if two main conditions were satisfied: better answer to a real need, and in a cost-effective way. This chapter will review the most important applications in which Li-ion might bring benefits, and will describe the technical status today.

Going back a little over the last 5 to 10 years, the possible use of Liion in large batteries was first considered in electric vehicles, for which high specific energy is crucial to grant a sufficient range. In parallel with the development of portable Li-ion, very important R&D programs started about 8 years ago aiming at demonstrating the feasibility of such batteries in electric vehicles. These programs were supported by a strong financial effort all over the world: European Commission and french government organizations in Europe, DOE / USABC in USA, MITI / NEDO / LIBES in Japan (addressing first the stationary batteries for home individual power sources). Saft was one of the few companies involved from the beginning. Then, with the fast developments of HEV's, Li-ion also had the opportunity to demonstrate its very high potential in terms of specific power and power density.

This chapter on lithium batteries for medical applications is not meant to be an exhaustive review, but rather a broad overview of some of the different types of lithium batteries that power implantable medical devices. The battery systems described in this chapter fall into two major categories, primary or single use cells containing lithium metal anodes, and secondary or rechargeable systems utilizing lithium ion chemistry. Primary lithium batteries have been used for implantable devices such as cardiac pacemakers, drug pumps, neurostimulators and cardiac defibrillators. Secondary lithium ion batteries have been used with left ventricular assist devices, total artificial hearts, and implantable hearing assist devices.

The first human implant of a lithium battery, a lithium/iodine cell that powered an implantable cardiac pacemaker, was conducted thirty years ago.¹Since that time several different lithium anode batteries have been developed and used successfully in a diverse set of implantable medical devices. The cells used in these devices are typically developed for the application and have used various combinations of cathode, electrolyte and separator, to meet the specific requirements of a device. Despite the various approaches that have been used, there are several power source characteristics that are desirable across all applications.

The focus of this chapter is Li-Ion batteries in consumer applications. There is a long history and a large market of non Li-Ion batteries-many of them primary lithium batteries—but this other field is too extensive for inclusion here except to mention the recent apparent success to commercialize, after a long gestation period, the lithium metal-polymer electrolyte technology in the telecommunications backup application. The reader should also note that Li-Ion batteries are now finding their way into many non-consumer applications. This, too, is outside the scope of the present chapter.

The subject begins with the lithium metal/molybdenum disulfide system introduced by Moli Energy in the late 1970, early 1980 period. Other rechargeable lithium systems were introduced for consumer applications about this same time but the Moli cell was the first to reach wide spread use. As is very well known, the product encountered a safety issue. Was it just faulty construction or was the lithium metal-liquid electrolyte system so unstable as to be unacceptable for consumer applications? (And does the “dry” polymer electrolyte solve this problem?) Whatever the technical explanation, the technology gained a perceived image as unsafe and there was a need to move on. Sony researchers (and others) saw the opportunity and developed a high capacity carbon based anode. (Research into lithium intercalation in graphite has a much longer history but the capacities needed for practical use had generally not been observed.) This anode, along with a high voltage metal oxide cathode was key to what came to be called the Li-Ion battery, first officially described in 1991. The name was intended to dissociate the new system from the “unsafe” lithium battery. It is, of course, also to imply a “rocking chair” type battery (or the physicist's battery as early on mentioned). The researchers, themselves, cannot be blamed for the ensuing confusion from the unfortunate use of “anode” and “cathode” in a rechargeable system. Not incidental to Sony's development of Li-Ion technology was a need to leapfrog the nickel-metal hydride technology in the growing portable electronics market.