Capacity Fade Mechanisms and Side Reactions in Lithium-Ion Batteries Pankaj Arorat and Ralph E. White* 作者：Pankaj Arorat and Ralph E. White* Center For Electrochemical Engineering, Department of Chemical Engineering, University of South Carolina,Columbia, South Carolina 29208, USA 美国，南卡罗来纳，年哥伦比亚29208，南卡罗来纳大学，化学工程系，中心电化学工程 | |
ABSTRACT The capacity of a lithium-ion battery decreases during cycling. This capacity loss or fade occurs due to several different mechanisms which are due to or are associated with unwanted side reactions that occur in these batteries. These reactions occur during overcharge or overdischarge and cause electrolyte decomposition, passive film formation, active material dissolution, and other phenomena. These capacity loss mechanisms are not included in the present lithium-ion battery mathematical models available in the open literature. Consequently, these models cannot be used to predict cell performance during cycling and under abuse conditions. This article presents a review of the current literature on capacity fade mechanisms and attempts to describe the information needed and the directions that may be taken to include these mechanisms in advanced lithium-ion battery models. Introduction The typical lithium-ion cell (Fig. 1) is made up of a coke or graphite negative electrode, an electrolyte which serves as an ionic path between electrodes and separates the two materials, and a metal oxide (such as LiCoO2, LiMn2O4, or LiNiO2) positive electrode. This secondary (rechargeable) lithium-ion cell has been commercialized only recently.Batteries based on this concept have reached the consumer market, and lithium-ion electric vehicle batteries are under study in industry.The lithium-ion battery market has been in a period of tremendous growth ever since Sony introduced the first commercial cell in 1990.With energy density exceeding 130 Wh/kg (e.g., Matsushita CGR 17500) and cycle life of more than 1000 cycles (e.g., Sony 18650) in many cases, the lithium-ion battery system has become increasingly popular in applicationssuch as cellular phones, portable computers, and camcorders.As more lithium-ion battery manufacturers enter the market and new materials are developed,cost reduction should spur growth in new applications. Several manufacturers such as Sony Corporation, Sanyo Electric Company, Matsushita Electric Industrial Company, Moli Energy Limited, and A&T Battery Corporation have started manufacturing lithium-ion batteries for cellular phones and laptop computers. Yoda1 has considered this advancement and described a future battery society in which the lithium-ion battery plays a dominant role. Several mathematical models of these lithium-ion cells have been published. Unfortunately, none of these models include capacity fade processes explicitly in their mathematical description of battery behavior. The objective of the present work is to review the current understanding of the mechanisms of capacity fade in lithium-ion batteries. Advances in modeling lithium-ion cells must result from improvements in the fundamental understanding of these processes and the collection of relevant experimental data. Some of the processes that are known to lead to capacity fade in lithium-ion cells are lithium deposition (overcharge conditions), electrolyte decomposition, active material dissolution, phase changes in the insertion electrode materials, and passive film formation over the electrode and current collector surfaces. Quantifying these degradation processes will improve the predictive capability of battery models ultimately leading to less expensive and higher quality batteries. Significant improvements are required in performance standards such as energy density and cycle life, while maintaining high environmental, safety, and cost standards. Such progress will require considerable advances in our understanding of electrode and electrolyte materials, and the fundamental physical and chemical processes that lead to capacity loss and resistance increase in commercial lithium-ion batteries. The process of developing mathematical models for lithium ion cells that contain these capacity fade processes not only provides a tool for battery design but also provides a means of understanding better how those processes occur. Present Lithium-Ion Battery Models The development of a detailed mathematical model is important to the design and optimization of lithium secondary cells and critical in their scale-up. West developed a pseudo two-dimensional model of a single porous insertion electrode accounting for transport in the solution phase for a binary electrolyte with constant physical properties and diffusion of lithium ions into the cylindrical electrode particles. The insertion process was assumed to be diffusion limited, and hence charge-transfer resistance at the interface between electrolyte and active material was neglected. Later Mao and White developed a similar model with the addition of a separator adjacent to the porous insertion electrode. These models cover only a single porous electrode; thus, they do not have the advantages of a full-cell-sandwich model for the treatment of complex, interacting phenomena between the cell layers. These models confine themselves to treating insertion into TiS2 with the kinetics for the insertion process assumed to be infinitely fast. Spotnitz accounted for electrode kinetics in their model for discharge of the TiS2, intercalation cathode. The galvanostatic charge and discharge of a lithium metal/solid polymer separator/insertion positive electrode cell was modeled using concentrated-solution theory by Doyle. The model is general enough to include a wide range of separator materials, lithium salts, and composite insertion electrodes. Concentrated-solution theory is used to describe the transport processes, as it has been concluded that ion pairing and ion association are very important in solid polymer electrolytes. This approach also provides advantages over dilute solution theory to account for volume changes. Butler-Volmer-type kinetic expressions were used in this model to account for the kinetics of the charge-transfer processes at each electrode. The positive electrode insertion process was described using Pick's law with a constant lithium diffusion coefficient in the active material. The volume changes in the system and film formation at the lithium/polymer interface were neglected and a very simplistic case of constant electrode film resistances was considered. Long-term degradation of the cell due to irreversible reactions (side reactions) or loss of interfacial contact is not predictable using this model. Fuller developed a general model for lithium ion insertion cells that can be applied to any pair of lithium- ion insertion electrodes and any binary electrolyte system given the requisite physical property data. Fuller work demonstrated the importance of knowing the dependence of the open-circuit potential on the state of charge for the insertion materials used in lithium-ion cells. The slopes of these curves control the current distribution inside the porous electrodes, with more sloped open-circuit potential functions leading to more uniform current distributions and hence better utilization of active material. Optimization studies were carried out for the Bellcore plastic lithium-ion system. The model was also used to predict the effects of relaxation time on multiple charge-discharge cycles and on peak power. Doyle modified the dual lithium-ion model to include film resistances on both electrodes and made direct comparisons with experimental cell data for the LiC6-LiPF6, ethylene carbonate/dimethyl carbonate (EC/ DMC), Kynar FLEX-ILiyMn2O4 system. Comparisons between data and the numerical simulations suggested that there is additional resistance present in the system not predicted by present models. The discharge performance of the cells was described satisfactorily by including either a film resistance on the electrode particles or by contact resistances between the cell layers or current-collector interfaces. One emphasis of this work was in the use of the battery model for the design and optimization of the cell for particular applications using simulated Ragone plots. Thermal modeling is very important for lithium batteries because heat produced during discharge may cause either irreversible side reactions or melting of metallic lithium, Chen and Evans carried out a thermal analysts of lithiumion batteries during charge-discharge and thermal runaway using an energy balance and a simplified description of the electrochemical behavior of the system. Their analysis of heat transport and the existence of highly localized heat sources due to battery abuse indicated that localized heating may raise the battery temperature very quickly to the thermal runaway onset temperature, above which it may keep increasing rapidly due to exothermic side reactions triggered at high temperature. Pals and Newman developed a model to predict the thermal behavior of lithium metal-solid polymer electrolyte cells and cell stacks.This model coupled an integrated energy balance to a fullcell- sandwich model of the electrochemical behavior of the cells. Both of these models emphasized the importance of considerations of heat removal and thermal control in lithium polymer battery systems. Verbrugge and Koch developed a mathematical model for lithium intercalation processes associated with a cylindrical carbon microfiber. They characterized and modeled the lithium intercalation process in single-fiber carbon microelectrodes including transport processes in both phases and the kinetics of charge transfer at the interface. The primary purpose of the model was to predict the potential as a function of fractional occupancy of intercalated lithium. The overcharge protection for a Li/TiS2 cell using redox additives has been theoretically analyzed in terms of a finite linear diffusion model by Narayanan . Darling and Newman modeled a porous intercalation cathode with two characteristic particle sizes.They reported that electrodes with a particle size distribution show modestly inferior capacity-rate behavior and relaxation on open circuit is substantially faster when the particles are uniformly sized. Nagarajan modeled the effect of particle size distribution on the intercalation electrode behavior during discharge based on packing theory. They observed that during pulse discharge, an electrode consisting of a binary mixture displays higher discharge capacity than an electrode consisting of single sized particles. The current from the smaller particles reverses direction during the rest period which cannot be observed in the case of an electrode comprised of the same-sized particles. Recently Darling and Newman made a first attempt to model side reactions in lithium batteries by incorporating a solvent oxidation side reaction into a lithium-ion battery model, Even though a simplified treatment of the oxidation reaction was used, their model was able to make several interesting conclusions about self-discharge processes in these cells and their impact on positive electrode state-of-charge。 A number of models having varying degrees of sophistication have been developed for lithium rechargeable batteries. For the most part, these models consider the ideal behavior of the systems, neglecting the phenomena that lead to losses in capacity and rate capability during repeated charge-discharge cycles. Fundamental models of these latter phenomena are less common because these processes are not as well understood. Also, models of failure modes in batteries do not usually have general applicability to a wide range of systems. However, the importance of these phenomena in the safe and efficient operation of high-energy lithium-ion batteries requires that they be in corporated into future battery models。 Capacily Fading Phenomenon Side reactions and degradation processes in lithium-ion batteries may cause a number of undesirable effects leading to capacity loss. Johnson and White have shown that the capacities of commercial lithium-ion cells fade by ca.10-40% during the first 450 cycles.A flow chart describing many of the processes leading to capacity fade is shown in Fig. 2. In Fig. 3, the capacity fade processes are shown on half-cell discharge curves. This gives a clearer picture of the processes by demonstrating where each is expected to manifest itself during operation of the battery Below, we discuss each of these processes in some detail, after first discussing the general topic of capacity balance. Capacity Balancing in Lithium-Ion Cells Lithium-ion cells operate by cycling lithium ions between two insertion electrode hosts having different insertion energies.For optimum performance, the ratio of the lithium-ion capacities of the two host materials should be balanced. Capacity balancing refers to the optimization of the mass loading in the two electrodes to achieve the maximum capacity (or energy) from the battery under conditions of steady cycling. Due to the practical importance of this subject for maximizing cell performance, as well as the safety implications with poorly balanced cells, this subject has been discussed in the literature by several authors. The condition for balanced capacities in a lithium-ion cell can be written in terms of a ratio γ of active masses in the electrodes. Written as a ratio of positive to negative electrode masses, this expression is This equation says that the desired mass ratio depends on the relative coulombic capacities of the two electrodes (C is in units of mAh/g) and the amount of cyclable lithium in each. The cyclable lithium is quantified in terms of the range of lithium stoichiometry in the insertion electrode that can be cycled reversibly, with the notation that Δx refers to the range of negative electrode stoichiometry and Δy to the positive electrode. For some insertion materials, which have several plateaus over which lithium can be inserted and deinserted, one may choose to cycle over only a limited range of stoichiometry for reversibility or safety reasons. In these cases, the stoichiometric range entered in the above formula would be reduced from its maximum value. For example, consider the case of a lithium-ion cell having a petroleum coke negative electrode and a lithium manganese oxide spinel positive electrode. By choice, we can assign useful ranges of stoichiometries for the two electrode materials of 0.61 for the coke and 0.83 for the lithium manganese oxide. These stoichiometric ranges correspond to the following electrochemical processes The active mass ratio needed to cycle these two materials in the manner shown here is equal to 1.85. This is calculated by using the theoretical capacities of both positive and negative electrode (C÷ = 148 mAh/g and C = 372 mAh/g), equal to F divided by the molecular）. The situation above describes an "ideal" lithium-ion cell in which the capacity balance does not change over the life of the cell. For an ideal cell, the initial lithium capacity available for cycling is constant over the life of the battery .Unfortunately the true case in actual lithium-ion batteriesis more complicated than this, and side reactions and secondary processes are able to perturb the capacity balance from its ideal state. The actual optimized active mass ratiois 2.05-2.15 for the coke/LiMn2O4 system, which corresponds to 14% excess capacity in the positive electrode.This excess capacity is a measure of the amount of lithium needed to form a stable film over the electrode surfaces. A major process that affects the capacity balance is the initial formation period needed to passivate carbon-based electrodes. It is now well known that carbonaceous lithium insertion electrodes have irreversible capacity associated with the initial charging cycles. This irreversible capacity loss is thought to result in the formation of a lithium conducting solid electrolyte interface (SEI) layer on the surface of the carbon, while in the process consuming some portion of the cyclable lithium ions in the cell. The loss of cyclable lithium to create this passivation layer has a profound impact on the capacity balance in the cell because it can remove a significant portion of the cyclable lithium depending on the type of carbon used. If the cyclable lithium in the cell is reduced due to side reactions of any type, the capacity balance is changed irreversibly and the degree of lithium insertion in both electrodes during cell cycling is changed. Consider the case of the initial carbon passivation process that occurs on all lithium-ion cells using carbon-based electrodes. The cell is assembled initially in the discharged state, with the carbon free of lithium and the metal oxide positive electrode at its maximum lithium content. The amount of lithium in either electrode can be represented as shown in Fig. 4, which illustrates the difference between the ideal and actual carbon/LiMn2O4 lithium-ion system during the first few cycles. In an ideal lithium-ion cell (Fig. 4a), all of the lithium should be intercalated into the negative electrode from the positive electrode during the first charge. Similarly all of the lithium ions should be intercalated back into the positive electrode during the first discharge. In an actual lithium- ion cell, upon charging the cell for the first time, some portion of the lithium removed from the LiMn2O4 positive electrode goes into the irreversible film formation reaction while the remainder inserts into the carbon structure. The capacity due to the irreversible reaction is represented schematically in Fig. 4b by the smaller box below the negative electrode. After the cell is finished charging to some arbitrary cut off voltage, the positive electrode has been delithiated to the extent possible under the charging conditions and the negative electrode is as full of lithium as possible given the amount of positive electrode mass available. Ideally the lithium content in the carbon at this point is at its maximum safe value. Also, we can imagine that the passivation layer is fully formed on the initial charging cycle, having consumed a certain amount of cyclable lithium irreversibly. When this cell is now discharged for the first time, the total quantity of lithium available for discharge is equal only to the amount of lithium reversibly inserted into the carbon electrode. Hence, the initial irreversible lithium lost cannot be recovered or utilized. The discharge proceeds until all of the reversible lithium is removed from the carbon electrode. At this time, the stoichiometry in the positive electrode will not reach its initial value upon cell assembly due to the capacity lost on the initial charging cycle. This situation is reflected in Fig. 4 in the bottom diagram. If the cell operates without any additional side reactions for the rest of its life, it will still never utilize the full range of stoichiometry available in the positive electrode. Thus for the above carbon/LiMn2O4 system it is safe to cycle within the limits of Δx = 0.61 (x varying from 0 to 0.61) and Ay = 0.83 (y varying from 0.17 to 1.0) as shown in Fig. 4. It should be remembered that these Ax and Ay values are cell and material specific. The range of stoichiometries accessed in the negative electrode in this example depends on the positive to negative mass ratio parameter γ. If the ideal value of γ had been used to fabricate this example cell, the initial loss of lithium due to the irreversible passivation process would prevent the carbon electrode from being fully utilized to an extent that depended directly on the amount of irreversible capacity that the particular carbon electrode material exhibited. Rather than let this happen, the common procedure is to assemble cells having a greater than theoretical amount of positive-electrode mass, thus allowing for losses of cyclable lithium during operation by initially providing extra lithium. One method of providing the extra lithium without increasing the cathode mass is to use overlithiated manganese oxide (Li+Mn2O4) spinel electrodes as proposed by Tarascon and Peramunage. Even with side reactions and irreversible capacity losses, the desired mass ratio can still be calculated via a formula analogous to the above one, although we must now include in the negative electrode capacity an additive contribution due to the passivation process. Referring to this contribution as C1.. (mAh/g), the capacity balancing condition can be expressed as For example, in the case of a lithium-ion cell fabricated using a carbon (petroleum coke) negative electrode and a lithium manganese oxide spinel positive electrode, the actual mass ratio desired for optimum utilization of the two electrodes is about 14% larger than its theoretical value. This excess capacity is a measure of the amount of lithium needed to form a stable film over the electrode surfaces. The active mass ratio for the graphite/LiMn2O4 system is ca. 2.4-2.45. Smaller mass ratios will prevent full utilization of the negative electrode whereas larger mass ratios present a safety hazard because the negative electrode can be overcharged (more lithium is available to insert into the electrode than is desirable). Overall cell performance such as energy density is maximized at the optimum mass ratio only. It should also be apparent that there is a relationship between the expected overcharge and overdischarge processes and the cell's capacity balance. For example, in the case of the lithium manganese oxide spinel material discussed above, overcharge reactions involving solvent oxidation depend on the capability of the cell to fully oxidize the positive electrode during normal cycling conditions. For cells with high mass ratios, this may not be possible because the negative electrode becomes fully charged before allowing the positive to become fully charged (i.e., before complete removal of lithium from the positive). Overdischarge of high-mass-ratio cells will affect the negative electrode by emptying the carbon of lithium completely and then driving the negative electrode potential up to an undesirably high value. In other cases, the mass ratio may be lower than desired leading to overcharge of the positive electrode. For example, in the case of the coke/LiMn2O4 system, mass ratios higher than 2.1 can lead to overlithiation of the negative electrode during charge. Mass ratios lower than 2.1 will have less lithium available than needed and will thus result in overdischarge of the negative electrode with accompanying negative safety or performance consequences. The carbon passivation process is the most common and well-studied example of a side reaction in the lithium-ion cell that will change the capacity balance.However, a number of other processes are also capable of having this effect. Any side reaction that either produces or consumes lithium ions or electrons will lead to a change in the cell's capacity balance, with the potential to impact negatively the cell's performance. In addition, once the capacity balance is changed from its desired state, the changes are generally irreversible and may accumulate over many cycles to generate a hazardous condition in the cell. Although difficult to quantify experimentally, it is straightforward to follow these effects using battery models and computer simulations under dynamic conditions if the relevant phenomena are included in the models. Formation Cycles Lithium-ion cells exhibit a sharp decay in capacity during the first few cycles. This period is known as the formation period during which cells are conditioned prior to use. It is generally desirable for the capacity decay observed after the formation period to be very small compared to the total cell capacity, after which the charge-discharge reactions are nearly 100% efficient. The sharp decay in capacity is due primarily to the solid electrolyte interface layer formation on the negative electrode. Passivation of the carbon electrode during the formation period and subsequent capacity loss are highly dependent on specific properties of the carbon in use, such as degree of crystallinity, surface area, pretreatments, and other synthesis and process details.After the first few cycles, the cell stabilizes and exhibits a constant capacity. The formation cycles are one of the critical steps in the manufacture of lithium-ion systems. For graphitic materials such as Osaka Gas mesocarbon micobeads (MCMB), the irreversible capacity is as low as 8 to 15%, whereas for hard carbons it can be as high as 50% of the reversible capacity. Fong demonstrated that irreversible reactions occur on carbon-based electrodes during the first discharge in carbonate-based electrolytes prior to the reversible insertion- deinsertion of lithium ions. These irreversible reactions are associated with electrolyte decomposition and cause the formation of a passivating film or solid electrolyte interface on the surface of the carbon. When all the available surface area is coated with a film of decomposition products, further reaction stops. In subsequent cycles, these cells exhibit excellent reversibility and can be cycled without capacity loss for many cycles. These authors first showed that the reversible insertion of lithium into graphitic carbons was possible as long as the proper passivating solvent was present. Gas evolution was observed by Gozdz et al.53 during the formation of the passivation layer on the carbon electrode during the first charge of a lithium-ion cell. The gas evolved correlated well with the irreversible capacity loss observed during the formation cycle. More details of carbon passivation in various solvent systems and the mechanisms of the passivation process are reviewed in later sections on electrolyte reduction and film formation. 自动皂液器The formation period is critical in lithium-ion battery manufacture because of its economic impact. First, it obligates manufacturers to invest in battery cycling stations to cycle cells several times before sending them to market, consuming both time and resources. Second, irreversible capacity consumed during the formation period is lost to the battery, directly subtracting from the system's energy. Last, the formation period generates gases which under some conditions may need to be vented prior to further operation of the cell. Research efforts worldwide continue to generate very high capacity carbon electrode materials having high irreversible capacities. To utilize these materials in the most efficient manner requires a prepassivation or prelithiation scheme not involving the sacrifice of a substantial quantity of the cyclable lithium available in the positive electrode. Although several research groups have been studying these processes and potential alternative approaches, there is no known solution for eliminating the formation period in an economically feasible manner. Overcharge Phenomena Under conditions of overcharge, major capacity losses have been observed in all types of lithium-ion cells. The poor overcharge resistance of commercial lithium-ion cells and the safety issues that result from overcharge have led to tight co:ntrol over charging and discharging of commercial cells using built-in electronic circuitry. The future application cf lithium-ion cells in new areas would be facilitated by advances in understanding and controlling overcharge. In particular the use of lithium-ion cells in multicell bipolar stacks requires a greater degree of overcharge tolerance due to the difficulty in achieving uniform utilization of all cells in series stacks. Overcharge losses can be classified into three main types at present: (i) overcharge of coke and graphite-based negative electrodes, (ii) overcharge reactions for high-voltage positive electrodes, and (iii) overcharge/high-voltage electrolyte oxidation processes. These side reactions lead to loss of the active material and consumption of electrolyte, both of which can lead to capacity loss in the cell. Overcharge of Coke and Graphite-Based Negative Electrodes During overcharge of lithium-ion cells, metallic lithium may be deposited on the negative electrode surface as the primary side reaction. This reaction is expected for cells with excess cyclable lithium due to either higher than desired initial mass ratio or lower than expected lithium losses during the formation period. The freshly deposited lithium covers the active surface area of the negative electrode leading to a loss of the cyclable lithium and consumption of electrolyte because of the highly reactive nature of metallic lithium. This phenomenon may occur at high charge rates even for cells with the correct mass ratio because of the polarization at the negative electrode under these conditions.However,the more common circumstance leading to lithium deposition is poorly balanced cells having too much positive electrode mass initially. The primary side reaction involved in the overcharge process is Li++e-= Li(s) and the intercalation-deintercalation reaction on the negative 多媒体教室讲台electrode (coke or graphite) may be written Lithium metal deposited on the negative electrode reacts quickly with the solvent or salt molecules in the vicinity giving Li2CO3, LiF, or other products.The lithium metal is expected to form near the electrode-separator boundary where the electrode potential is more negative. The products formed may block the pores, leading to a loss of rate capability as well as capacity losses. Formation of lithium metal is also a safety hazard due to its extreme reactivity with the liquid solvents. Lithium metal deposition may be more of a concern with graphitic carbon electrodes than with coke electrodes due to the lower average open-circuit potential of the former. For this reason, mass ratios in cells using graphite are usually chosen to be smaller than the optimum in order to provide a buffer against lithium deposition, with the negative consequence that the full 372 mAh/g capacity of the graphite is not attained. Overcharge Reactions for High-Voltage Positive Electrodes Overcharging the positive electrodes in lithium-ion cells can lead to a wide variety of electrochemical reactions depending on the details of the system chemistry. As with the negative electrode, the extent to which overcharging is expected at the positive electrode depends on the system's capacity balance. For cells with too low a mass ratio, the positive electrode is stressed to a greater extent during charging and overcharge becomes a possibility. Overcharging the positive electrode can lead to capacity loss due to inert material formation (e.g.Co3O4) or solvent oxidation due to the high positive electrode potential. Formation of electrochemically inactive electrode decomposition products leads to a capacity imbalance between the electrodes.Thermal abuse of the positive electrode can lead to oxygen loss from the metal oxide lattice. This oxygen can increase the pressure inside the cell and represents a potential safety concern. Dahn proposed the following decomposition reactions for the three main positive electrode materials in their charged states under abuse conditions. The reaction proposed for LiCoO2 can be expressed as They observed that γ-MnO2 is more tolerant toward electrical and thermal abuse than LiyNiO2 and LiyCoO2. Oxygen loss from the metal oxides was observed when y < 1 and increased with decreasing stoichiometry. This loss of oxygen began at 200°C for Li0.3NiO2，240°C for Li0.4CoO2, and about 385°C forγ-MnO2, when heated at a rate of 1°C/mm. The higher the heating rate, the higher is the 02 onset temperature and vice versa. Formation of oxygen in the sealed cell in the absence of any recombination mechanism (as exists in Ni-Cd, Pb-acid, and Ni-MH cells using aqueous electrolytes) is a safety concern because of the accumulation of flammable gas mixtures in the cell. Also, the finalmetal oxide products such as Co3O4, LiNi2O4, and Mn2O3 are inert to lithium insertion-deinsertion, and hence capacity is lost irreversibly. Staniewicz proposed an overcharge mechanism for their LiNiO2 electrodes by accounting for all the lithium ions in LiNiO2-based cells. They divided the cycling process for the LiNiO2 electrode into three phases. Phase 1—Lithium ions used to passivate carbon irreversibly and/or not able to cycle back into the LiNiO2 LiNiO2—0.15Li + + Li0.85NiO2 + 0.15e- Phase 2.—Reversible cycling Li0.85NiO2 —0.5Li + + Li0.35NiO2 + 0.5e- Phase 3.—Overcharge Li0.35NiO2 —0.35Li + + NiO2 + O.35e- The first phase accounts for the lithium ions used to form the passive film, the second phase for reversible cycling, and the third phase accounts for the lithium ions removedduring overcharge. The overcharge reaction proposed by Staniewic does not agree with the mechanism proposed by Dahn for Li2NiO2 electrodes under conditions of thermal abuse. Moreover, NiO2 is not usually thought to be stable due to the Ni(IV) oxidation state.No experimental data were provided by the author to support the above hypotheses. The formation of low lithium content Liy NiO2(y ＜0.2) has been stated as a cause of cell failure during cycling. In addition, the material becomes highly catalytic toward electrolyte oxidation, and some of the nickel ions may migrate to lithium sites. The first cycle irreversibility is primarily related to the amount of Ni2+ between the slabs of NiO2. which requires extra charge for oxidation to a higher valence state. The stability of the structure of LiNiO2 at low lithium content can be improved by substituting Ni with Al or B. Recently, the thermal stability of LiNiO2 cathodes has been studied in detail by substituting a portion of the Ni or Li with other elements (Co, Mn, Mg, Ca, Sr, Ba). The substitution of Ni with Co improves the cycling behavior at room temperature, but the cycling characteristics at high temperatures remain unsatisfactory.Substituting a part of the Ni in LiNiO2 with alkaline earth metals (Mg, Ca, Sr, Ba) and with Al improves the high-temperature and high-rate performance of these electrodes. At high temperatures, the deterioration in the performance during charging and discharging is greatly influenced by a change in the chemical reactivity of the active material. In the presence of other substituents in the crystal structure, the reactivity decreases during deep charging and discharging and at high temperatures, which leads to a more stable material. In the case of LiCoO2, the high-temperature performance is deteriorated by the introduction of other elements. When the Ni in LiNiO2 is substituted with Ca, Nb, or In, the structural changes observed during the charging process are very small, which leads to better cyclability of these doped materials. The thermal stability of lithium manganese oxide spinel phases was studied by Thackeray .They proposed that Li2MnO3 is formed and oxygen is evolved when the spinel is heated in the temperature range of 780 to 915℃. The rate of oxygen evolution increased above 915℃, and around 1200℃, both O2 and Li2O were lost. The above reactions are not electrochemical reactions and occur when the active material is heated to a particular temperature. Including thermally induced electrode decomposition reactions in a phenomenological battery model may not be necessary because battery failure will likely have occurred already at lower temperatures. Gao and Dahn showed a correlation between the capacity fade of the spinel and the growth of the 3.3 V discharge plateau upon cycling.The 3.3 V discharge plateau increased each time the cell was charged to a higher voltage, suggesting that LiMn2O4 tends to lose O2 when overcharged. The following mechanism was proposed El —(Oxid El) + +e- and LiMn2O4 + 2δe —LiMn2O4-δ+ δO2- where El is the electrolyte solvent molecule and (Oxid El) + denotes a positively charged electrolyte solvent molecule (radical cation). The radical cation (Oxid El) +can be assumed to be very unstable and will participate in further side reactions immediately upon formation. One possible process would be dimerization of the radical with accompanying expulsion of two protons. If this cationic species is sufficiently stable to reach the negative electrode, it would undergo reduction either back to the original solvent species or to other products. Highly delocalized saltanions such as PF6 may help to stabilize cationic species such as these. These authors state that the electrolyte may act as an electron donor to the partially delithiated spinel, inducing the oxygen loss from the oxide structure. It is possible that a second phase (similar to what is formed during heating) with the rock-salt structure forms at the surface of LiMn2O4 when it loses oxygen over the course of cycling. The loss of oxygen from the sample during cycling is undesirable, not only because it could induce structural damage to the sample surface impacting the cycling ability, but also because it tends to oxidize the electrolyte which also reduces the cell's life. Overcharge/High Voltage Electrolyte Oxidation Processes The electrolytes used in lithium-ion cells are mixtures of organic solvents and one or more lithium salts. The most popular electrolytes currently being used include mixtures of the linear and cyclic carbonates such as propylene carbonate (PC), ethylene carbonate (EC), dimethyl carbonate (DMC), diethyl carbonate (DEC), and ethyl methyl carbonate (EMC) and salts such as LiPF6, LiBF4, LiAsF6 and LiC1O4. Sony reportedly uses a mixture of PC, DMC, and EMC with LiPF6 salt, whereas Sanyo and Matsushita use mixtures of EC, DMC, and DEC and EC, DMC, DEC, and EMC, respectively, with LiPF6. High voltage positive electrodes used in lithium-ion batteries present a stringent requirement for electrolyte stability and purity. The electrolyte choice is a limiting factor in lithium-ion batteries because the maximum voltage of the cell is limited by the decomposition potential of the electrolyte. Common electrolytes in use today decompose at high voltages (>4.5 V) forming insoluble products (Li2CO3, etc.)which block the pores of the electrodes and cause gas generation in the cell. These effects can cause both capacity loss upon further cycling of the cell and can also be an extreme safety hazard. One particular solvent combination, EC/DMC, is in use in many systems alone and in combination with other solvents and is claimed to have the highest oxidation resistance among the common carbonate mixtures.Campbell reported that the oxidation potential of pure PC is higher than that of PC containing electrolyte salts. This suggests that the electrochemical oxidation of nonaqueous electrolytes is enhanced by the presence of electrolyte salts. Decomposition potentials are assessed experimentally by performing cyclic voltammetry either on inert metal surfaces or on actual insertion electrode materials and setting an arbitrary criterion on the current density above which solvent breakdown is assumed to be occurring. For irreversible electrochemical side reactions such as these, no thermodynamic open-circuit potential exists, and hence the decomposition potential does not have a firm meaning. Instead, these side reactions may often be described with Tafel equations which lead to a finite rate of decomposition at all voltages, increasing exponentiallywith increasing voltage. The decomposition potentials of many electrolytes are reported in Table I; however, it is not always clear whether the solvent or the salt or both are involved in the oxidation processes.In addition, the ambiguity in reporting values for solvent oxidation potentials is not always appreciated. These data are only well defined if the value of the current density at which the decomposition potential is assessed is given as well as the voltage scan rate and the electrode material used. We have taken the cyclic voltammetry data of Kanamura measured at a sweep rate of 50 mV/s and used the criterion of 0.1 mA/cm2 as the threshold current density to define the oxidation potentials given in Table I. Data of Tarascon were used as given in their paper because the actual cyclic voltammagrams were not provided by these authors. Christie and Vincent measured the oxidation potential using cyclic voltammetry at a sweep rate of 200 mV/s and used the criterion of 1 mA/cm2. Ossolo used linear sweep voltammetry at a scan rate of 2mV/s to determine the oxidation potential (E) of various electrolytes on Li1+xV3O8 electrodes. They assumed E to be the potential at which a current density of 0.5mA/cm2 was recorded. The solvent oxidation process can be stated in general as follows solvent — oxidized products (gases, solution, andsolid species) + ne Any solvent (for example, PC or EC) oxidized will be lost, eventually leading to an increase in the salt concentration and a drop in the electrolyte level which will adversely affect the cell capacity. Also, the solvent oxidation products such as gases or other species will build up in the cell and cause a variety of problems. The rate of solvent oxidation depends ort the surface area of the positive electrode material, current collectors, and the carbon black additive. In fact, the choice of carbon black and its surface area are critical variables because solvent oxidation may occur more on the carbon black than on the metal oxide electrode due to the higher surface area of the former. If a small part of the electrolyte is consumed during each charge, more electrolyte needs to be used when the cell is assembled. This implies less active material for a constantvolume container and consequently less initial capacity. Also, the solid products may form passivating layers on the electrodes that increase the polarization of the cell and thereby lower the output voltage of the battery. Novak found that PC oxidizes at potentials as low as 2.1V vs. Li/Li on Pt and that the rate of oxidation increases substantially above 3.5V. Depending on the electrode material, PC oxidation can begin at potentials as low as 2Vvs. Li/Lit; however, a much greater degree of stability (up to and exceeding 4.5V) is often exhibited by PC in practice. Cattaneo and Ruch analyzed the volatile gaseous products from the decomposition of LiClO4/PC and LiAsF6/PC on heat-treated MnO2 electrodes using online mass spectroscopy. Bulk oxidation of the electrolyte takes place above 4.0 V vs. Li/Li.CO2 evolution was observed at low potentials (3.15 and 3.4 V vs. Li/Lit) depending on the state of charge of the electrodes. No CO2 was observed in the reverse (cathodic) scan. Chlorinated species were formed from the decomposition of ClO- ions above 4.5 V.The species identified were CO2 and HCL, which were assumed to be formed by the following mechansm ClO4-→ e- + ClO4 →ClO2+2Oad + e- ClO2 + H + +e -→ HC1 + O2(g) 2Oad — O2(g) Eggert and Heitbaum71 also observed the oxidation of perchlorate anions on a Pt electrode at potentials above 4.6 V vs. Li/Lit using differential electrochemical mass spectrometry (DEMS). The instability of ClO2 in the presence of protons from the oxidation of PC will produce HCl. Oxygen evolution is also observed in the decomposition of LiC1O4 electrolytes. Christie and Vincent67 reported the oxidation potential for 1 M LiPF6 in PC at a Ni microelectrode. Kanamura et al.7273 studied the ring opening of PC on Pt, Al, Au, and Ni electrodes. The PC oxidation potentials on these materials varied from 4.5 V for Ni to 6 V for Cu vs. Li/Lit. It was shown by Kanamura et a].73 that the anodic behavior of Ni electrodes in various propylene carbonate electrolytes depends strongly on the type of electrolyte salt used. The occurrence of decomposition products depends on the type of anion in the high electrode potential range. Electrolyte oxidation during overcharge of lithium-ion cells has been verified experimentally by Tarascon et al.37 by cyclic voltammetry experiments. However, none of these studies provided mechanisms for the decomposition processes or used analytical techniques to study the products formed. Considering the large number of studies on solvent reduction in lithium batteries, and the relative importance of solvent oxidation to cell performance and safety, the lack of fundamental knowledge in this area is surprising. A detailed discussion of electrolyte decomposition (reduction) mechanisms is given in a later section. Overcharge Protection Successful commercialization of lithium-ion batteries depends very much on their safety during operation under normal and especially under abusive conditions. An abuse condition generally leads to an increase in cell temperature which can initiate self-heating of the cell and eventually lead to thermal runaway. The organic solvents commonly used to prepare electrolytes for lithium-ion batteries undergo irreversible oxidation at the positive electrode, which deleteriously affects the cycling performance of lithium batteries. A common way to avoid this is by including an additive in the electrolyte, an internal "chemical shuttle," to provide a current bypass mechanism when the cell exceeds a certain voltage.The ideal chemical shuttle operates at or near the voltage of the fully charged cell and takes up the extra charge passed during overcharge, thereby preventing damaging reactions from proceeding. For typical lithium-ion cells, the desired potential of the redox process is approximately 4.5 to 5.0 V vs. Li/Li+or 1.5 to 1.0 V vs. HVH2. In such a case, the electrochemical reactions sustained during overcharge at the positive and negative electrodes are R → O + ne- O+ ne- → R Species O is generated at the positive electrode and diffuses to the negative electrode where it is reduced to H. The redox couple shuttles between the two electrodes to take up the excess charge input during overcharge and continues until charging is terminated. For example, LiI functions as a good redox additive for overcharge protection at voltages close to the charged cell potential for certain 3 V lithium cells.Its use in 1 M LiPF6/THF solutions has been shown to avoid the oxidation of the electrolyte on Pt surfaces during overcharge. At anodic potentials (less positive than the fully charged cell potential), Lii undergoes a two-step process of oxidation of iodide ion to triiodide ion (3.2 V vs. Li/Li+) and further oxidation of tritodide ion to iodine (3.65 V) LiI—I-+ Li+ 3I—I3- +2e 2I3- —3I2 + 2e The iodine generated by the oxidation of LiI reacts chemically with lithium metal to regenerate LiI. The reduction of iodine occurs via a similar stepwise process of reduction of iodine to trhodide ion (3.55 V) and trilodide ion to iodide ion (2.75 V). In addition to chemical shuttles, other methods of overcharge protection which are being used in commercial cells are 1. Separators with melting points of about 140℃can help in overcharge protection.The purpose is to have a polymer membrane that melts and shuts off the current when the cell temperature rises above a given value during short-circuit conditions. The cells are prevented from reaching a high internal temperature by ceasing the reactions (and hence heat generation) when current is prevented from flowing across the separator. A number of polyolefin polyethylene (130℃), and polypropylene (155℃) based separators that can be used as internal safety devices by closing down the pores during short-circuit conditions have been developed.Sony uses a polypropylene-based separator (161.7℃ mp), whereas Sanyo and Matsushita use polyethylene-based separators with melting points of 135.4 and 133.4℃, respectively The shut down temperature for Celgard® 2300 FSM is 131℃。 2. Additives to the cathode mix, such as LiCoO2 will decompose during overcharge and increase the pressure inside the cell. This pressure activates the vent present at the top of the cell, and the pressure is released and the circuit broken. The presence of excess additives does not greatly affect the current flow during normal operation of the cell, as shown in the Sony cell.Moli Energy has used 2 wt % biphenyl in their graphite/LiCoO2 cells for overcharge protection. Solid biphenyl decomposition products deposit on the cathode resulting in high internal impedance and low rate capability。 3. Explosion-proof valves become deformed upon increase of internal pressure of the cell to cut a connection lead contained in the cell. The supply of charging current is cut off when the pressure increases abnormally。 Electrolyte Decomposition (Reduction) Processes Electrolyte Reduction can jeopardize the capacity and cycle life of the cell by consuming salt and solvent species, and compromises the safety of the system by generating gaseous products which increase the internal pressure of the cell. Minimizing the electrolyte reduction reactions and the capacity losses related to these processes is a major requirement for enhancing cycle life and improving the high-temperature performance of lithium-ion batteries. Electrolyte reduction is an expected feature of all cells using carbon-based insertion electrodes due to the instability of the electrolyte to the carbon electrode under cathodic conditions. The process of carbon passivation during the initial cell cycling is referred to as the formation period as discussed earlier. Ideally, electrolyte reduction is confined to the formation period and does not continue during the steady cycling of the cell。 Electrolyte reduction reactions on carbon surfaces are similar to those on lithium metal because the difference in potential between the metallic lithium and fully lithiated carbon is very small.For this reason, a large amount of literature on electrolyte reduction processes on metallic lithium can be utilized to understand these processes on carbon insertion electrodes. A large number of experimental techniques including X-ray photoelectron spectroscopy (XPS), Auger electron spectroscopy energy dispersive Xray analysis (EDAX), Haman spectroscopy on-line mass spectroscopy in situ and ex situ Fourier transform infrared (FTIR) spectroscopy atomic force microscopy (AFM), and electron spin resonance have been used to determine the reduction mechanisms and to identify the products formed on carbon electrode surfaces. Dey studied the electrochemical decomposition of PC on graphite and proposed a decomposition mechanism 2Li++2e-+ (PC/EC) —[propylene(g)/ethylene(g)]+ Li2CO3(s) The above reaction occurs during the first discharge when the potential of the electrode is near 0.8V vs. Li/Li+'. Fong proposed a similar mechanism for the irreversible capacity loss during the first cycle on graphitic carbon electrodes with mixed EC/PC electrolytes. Aurbach and coworkers have performed extensive studies of solvent and salt reduction processes and their products on metallic lithium and carbon-based electrodes in a variety of electrolytes. They observed ROCO2Li species [probably CH3CH(OCO2Li)CH2OCO2Li] and propylene resulting from a one-electron reduction process of PC. Previously Dey stated that PC reduction on graphite is a two-electron process with Li2CO3 and propylene as major products. Aurbach has argued that ROCO2Li species are highly sensitive to trace water and react rapidly with it to form Li2CO3, suggesting that previous studies were not conducted under sufficiently dry conditions. In the presence of crown ethers,carbon electrodes retain their graphitic structure and undergo reversible intercalation because the solvent is not co-intercalated and reduced within the carbon, but rather on the surface. When the reduction of PC takes place on the surface, the charge transfer occurs mostly through existing films, and one would expect a one-electron reduction to be favorable because the driving force for PC reduction diminishes. In previous studies, graphite electrodes were destroyed and exfoliated in the absence of crown ethers, and did not reach intercalation stages when PC was the solvent. Thus, most of the PC reduction in that case could take place within the carbon structure after coin tercalation of PC molecules into the graphite. In such a case, two-electron processes to form Li2CO3 may become favorable. Matsumara observed that the irreversible capacity loss during the first cycle is not only due to PC decomposition to Li2CO3, but also because of additional side reactions. They concluded that during the first charge, it is possible that there are two paths for the decomposition of PC. In one, PC directly undergoes reduction to form propylene and Li2CO3. In the other, PC undergoes reduction to form a radical anion, and then forms lithium alkyl carbonates by radical termination. These are unstable and can be reduced to form unstable radical compounds that then react with propylene to form oligomer radicals, and finally oxidize to compounds which contain C-H bonds and COOH groups。 Shu studied the electrochemical intercalation of lithium into graphite in a 1M LiCIO4 PC/EC (1:1) electrolyte. They suggested that two processes are involved, namely, a two-electron reduction of PC and EC to propylene and ethylene gases and one-electron reduction to form lithium alkyl carbonates. The two-electron reduction can be further divided into direct electrochemical and chemical reduction. The initial step for both electrochemical reduction and solid electrolyte interphase (SEI) film formation processes involves formation of lithium carbonate complexes followed by one-electron reduction to a radical anion. The radical anions undergo further one electron reduction yielding gaseous products or radical termination to form an SEI film. Chu used in situ electrochemical atomic force microscopy to study the surface films formed on highly ordered pyrolytic graphite (HOPG) electrodes during cathodic polarization in 1MLiClO4 EC/DMC (1:1) and 1ML1PF6 EC/DMC (1:1) electrolytes. They found the reduction reactions to be irreversible and suggested that these reactions occur on edge surfaces of HOPG at a higher potential (1.6 and 2.0 V vs. Li/Li+) than on the basal surface (0.8 and 1.0 V). Peled has also shown that more solvent reduction occurs on the basal surface and more salt reduction occurs on the edge surface. Chu showed that the surface film formed over the electrode is much thicker (a few hundred nanometers) than previously believed (10 nm at the basal surface and 50% thicker on the edge surface). A number of film-forming additives with beneficial properties have been discussed in the literature. The addition of large amounts of SO2 promotes the reversible intercalation-cleintercalation of lithium ions into graphite in selected nonaqueous electrolytes.SO2 offers the advantage of forming fully developed passive films on the graphite electrode at potentials much higher than that of electrolyte reduction itself. These passive layers are composed of Li2S and Li-oxysulfur compounds as follows SO2 + 6Li ++ 6e -→2Li2O+Li2S Li2O +SO2→(LiO)2SO Or Li2O +2SO2-(LiO)OSOSO(LiO) Carbonate-based mixed electrolytes containing DEC and DMC were found to undergo ester exchange reactions, which lead to the spontaneous formation of methyl ethyl carbonate.This solvent has been disclosed in the patent literature as having desirable passivation properties for lithium-ion cells.A recent study conducted on graphite electrodes in methyl propyl carbonate (MPC) solutions containing LiPF6 or LiAsF6 showed that the reduction of this solvent is initiated at potentials at or below 1.5 V vs. Li/Li+.The authors observed Li2CO3 as a major surface species on the graphite electrode. A new solvent mixture composed of chloroethylene carbonate (CEC), PC, and EC has been proposed for lithiumion batteries.CEC forms a stable passive film on the negative electrode which is insoluble in the electrolyte. This solvent allows the use of PC-based electrolytes with graphitic electrodes without increasing electrolyte decomposition.The recent patent literature contains references to other carbonate-based solvents, including other halogen-substituted carbonates and a variety of unsaturated carbonates, claimed to have desirable properties for lithium-ion batteries. Often these carbonates are added to the cell in small quantities and are involved and consumed in the initial reduction and film formation process at the carbon electrode. It is expected that work in this area will continue and that the understanding of the passive film composition and its relationship to battery performance will improve in the future. A number of mechanisms have been proposed for the reduction of carbonate-based electrolytes (solvents and salts). These mechanisms (Eq. 28 to 46) are grouped based on solvents, salts, and contaminants. Solvent reduction—Propylene carbonate (PC)—The two-electron reduction mechanism of Dey is PC + 2e-→ propylene+ CO2-3 The one-electron mechanism for PC reduction given by Aurbach is PC + e - →PC- radical anion 2PC radical anion →Propylene +CH3CH(CO3-)CH2(CO3-) CH3CH(CO3-)CH2(CO3-)+2Li→ CH3CH (OCO2Li)CH2OCO2Li(s) (Li alkyl carbonate) Ethylene carbonate (EC).—The two-electron reduction of EC is similar to that for PC EC + 2e -→ethylene + CO2-3 and the one-electron mechanism is also analogous to the PC case EC + e→ EC- radical anion 2EC- radical anion →ethylene + CH2(OCO2)- CH2(OCO2)- CH2(OCO2)- CH2(OCO2)- + 2Li+→ CH2(OCO2Li)CH2OCO2Li(s) (Li alkyl carbonate) The EC reduction product CH2(OCO2Li)CH2OCO2Li(s) acts as an efficient passivating layer and is comparable to Li2CO3 in this respect. Dimethyl carbonate (DMC).—This mechanism can be written as follows or CH3OCO2CH3 + e- + Li→ CH3OCOOLi+ CH3+ or CH3OCO2CH3 + e- + Li+→ CH3OLi + CH3OCO+ Both CH3OLi and CH3OCO2Li are formed by a nudeophilic attack on the DMC. The radicals formed (CH and CH3OCO) are converted to CH3CH2OCH3 and CH3CH2OCO2CH3 as shown by Aurbach. Diethyl carbonate (DEC).—This mechanism can be written as follows CH3CH2OCO2CH2CH3 + 2e- + 2Li+→ CH3CH2OLi+CH3CH2OCO+ or CH3CH2OCO2CH2CH3 + 2e + 2Li +→ CH3CH2OCO2Li+CH3CH2+ The radicals formed (CH3CH+ and CH3CH2OCO+) by the decomposition of DEC are converted to CH3CH2OCH2CH3 and CH3CH2OCO3CH2CH3 as shown by Aurbach Imhof and Novak observed propylene and ethylene evolution on graphite electrodes in four different solvent mixtures (EC/DMC, PC/DMC, EC/PC, and EC/PC/DMC), but neither propylene nor ethylene were detected on nickel electrodes. Salt reduction—The salt used and its concentration should also affect the performance of carbon insertion electrodes, because salt reduction has been shown to participate in the buildup of surface films. In certain cases, salt reduction may contribute to stabilization of the surface and the formation of a desirable, passivating interface. In other cases, precipitation of salt reduction products may interfere with solvent reduction products. According to Jean reduction of the lithium salt LiCF3SO3 occurs before solvent (PC/EC/DMC) reduction on the negative electrode. The reduction reactions for the salts are as follows LiAsF6 LiAsF9 + 2e -+ 2Li+→ 3LiF + AsF3 AsF3 + 2xe- + 2xLi+→ LixAsF3-x + xLiF LiC1O4 LiC1O4 + 8e -+ 8Li+→ 4Li2O + LiC1 or LiCIO4 + 4e- + 4Li+→ 2Li2O + LiC1O2 or LiClO4 + 2e- + 2Li+ → Li2O + LiCIO3 LiPF6 LiPF6→ LiF + PF5 PF5 + H2O-→ 2HF + PF3O PF5 + 2xe- + 2xLi+→ xLiF + LixPF5-x PF3O + 2xe- + 2xLi+→ xLiF + LixPF3-xO and PF6- + 2e- + 3Li+→ 3LiF + PF3 LiBF4 (similar to LiPF6) BF4 + xe -+ 2xLi+→ xLiF + LixBF4-x Contaminant reduction—The electrolyte often contains contaminants such as oxygen and water. Oxygen can be reduced to form lithium oxide 1/2O2+ 2e +2Li+→ Li2O(s) The performance of graphite electrodes is unaffected by small amounts of water (100 to 300 ppm) present in the solvents. For higher concentrations of water. LiOH is formed on reduction of water on graphite in the presence of Lit, which precipitates on the surface of the carbon and acts as a blocking agent with a high interfacial resistance. Thus, LiOH can prevent further intercalation of Li into graphite H2O+e-→ OH- +1/2H2 Li +OH-→ LiOH(s) LiOH(s) + e- + Li+→Li2O(s) + 1/2H2 In the presence of CO2、Li2CO3 is formed as a passive layer on the negative electrode 2CO2 + 2e- + 2Li+→ Li2CO3 + CO or CO2 + e- + Li+→ CO2-Li+ CO2Li + CO2→ OCOCO2Li OCOCO2Li + e- + Li+→ CO + Li2CO3 Secondary reaction—Lithium carbonate can also be formed by a secondary reaction 2ROCO2Li + H2O → 2ROH + CO2 + Li2CO3 where R = ethyl or propyl group. LiAsF6 and LiPF6 reduction (Eq. 36 and 40) occur at potentials less than 1.5 V (vs.Li/Li+) Self-discharge Processes Self-discharge refers to the drop in cell voltage under open-circuit conditions that occurs spontaneously while batteries are left standing. Lithium-ion batteries undergo self-discharge which, although less significant than those of the competing Ni-Cd and Ni-NH batteries, is still relatively rapid and temperature dependent. Self-discharge phenomena inevitably occur in oxidized LiMn2O4, LiCoO2, and LiNiO2 electrodes. The extent of this self-discharge depends on factors such as cathode and cell preparation, nature and purity of the electrolyte, temperature, and time of storage. Self-discharge losses in lithium-ion cells have been classified according to whether they are reversible or irreversible.Reversible capacity losses were defined as those that can be recovered by charging the cell again while irreversible capacity losses were not recoverable. While this is a useful practical distinction to make, the extent to which capacity loss is irreversible depends on the charge and discharge rates used on the subsequent cycling. Hence, capacity losses due to self-discharge should preferably be accompanied by statements of the rates at which the data were obtained. For purposes of discussing self-discharge mechanisms, we attempt to separate processes that might lead to true irreversible capacity losses (capacity that cannot be regained at any charge rate) from processes that should be readily reversible and lead to no permanent capacity loss. Johnson and White have reported the self-discharge behavior of Sony and Matsushita cells. They monitored the open-circuit potential of these cells for 30 days and observed that all the cells maintained capacities greater than 97% of their initial capacity. Thus, they concluded that the effect of self-discharge on capacity fade during cell cycling is insignificant. The self-discharge rate is very high (10%/month at 55℃) at high temperature compared to 2-3%/month at room temperature. The capacity loss due to self-discharge is mostly recoverable as reported in the literature. The self-discharge mechanism for LiMn2O4/organic electrolyte cells has been stated to involve irreversible electrolyte oxidation at the positive electrode and the reversible insertion of lithium into the LiyMn3O4 spinel structure. The insertion process is reversible and the extent of delithiation of the electrode can be returned by recharging the cell. In general, charged lithium-ion cells can self-discharge by coupling the electrolyte-decomposition reaction to the primary lithium-intercalation reaction. The superficial oxidation of the electrolyte at the positive electrode surface can be written as EI→ e- + EI+ where El can be any solvent (EC, PC, etc.) used in lithium-ion batteries. The released electrons are used by the metal oxides (positive electrodes) to intercalate lithium according to the reaction yLi+ + ye- + MOz→LiyMOz which is the bulk intercalation of lithium into the positive electrode structure leading to a decrease in the state of charge of the electrode. For LixMn2O4 LiyMn2O4 + xLi+ + xe- → Liy+xMn2O4 The above reactions (Eq. 47 and 49) can occur simultaneously at the composite positive electrode without any external electron source. The overall reaction is LiyMn2O4 + xLi+ xEI→Liy+xMn2O4+ xEI+ The rate of self-discharge is limited primarily by the rate of solvent oxidation, emphasizing the importance of solvent stability in long shelf-life batteries. Guyomard have demonstrated that solvent oxidation occurs mainly on the carbon black surface, and recommended low surface area carbon black for controlling self-discharge rates. However, reducing the surface area of the active material has also been stated as important in this regard in the case of LiMn2O4, and the role of the current collector surface in carrying out solvent oxidation cannot be dismissed. The self-discharge of the positive electrode as written above would cause a permanent loss of capacity if the anode retained its charged state. This is a result of the perturbation of the capacity balance in the cell implicit in these mechanisms. Fortunately, self-discharge data recorded with a lithium reference electrode indicated that each electrode in the lithium-ion cell may self-discharge at a similar rate as in the manganese oxide-carbon case.The salt concentration in the cell can also be changed irreversibly by these processes. Either of these phenomena will lead to capacity or rate capability losses over the life of the cell. After long or repeated self-discharges, the lithium-ion cell will have unbalanced capacities in the two electrodes with an increased risk of lithium plating on carbon during charging. Further studies of self-discharge phenomena in lithium-ion cells were undertaken by Pistoia. The selfdischarge rates of the three main metal oxide cathodes were compared to one another in various electrolyte systems. Electrolyte oxidation was again involved in the selfdischarge mechanism, although this process alone could not explain all of the experimental findings. Self-discharge rates varied widely in different electrolytes aswould be expected from the electrolyte oxidation mechanism. In addition, the consequences of pore pluggage byoxidation products after self-discharge periods were seen in some cases, including higher internal resistance and losses of rate capability. Two additional mechanisms for self-discharge were proposed, one being the spontaneous lithium reinsertion into the positive electrode driven by the negative electrode and the second being electrode dissolution.The former process was ascribed to the partial instability of the oxidized positive electrode. Interestingly, this process halted when the lithium-metal negative electrode was replaced by a platirLum electrode, leading the authors to conclude that lithium ions from the negative electrode were involved in the self-discharge process. Because no net flow of current can exist during self-discharge, a flux of lithium ions from the negative to positive electrode must be compensated by an equal and opposite flux of another ionic species between the electrodes. This could be the result of either solvent oxidation at the positive electrode (leading to generation of cations) or solvent reduction at the lithium-metal electrode. The second mechanism, which is discussed in more detail in a later section, can be controlled by the proper choice of electrolyte. During self-discharge, delithiation of lithiated coke (and graphite) can be explained by the following redox process, which is obtained by coupling the electrolyte decomposition reaction and lithium insertion reaction at the negative electrode. EI+ ye-→passivating layer LixC6 → ye-→ yLi+→Lix-yC6 In this case, the electrolyte reduction reaction is thermodynamically possible due to the very reducing potential (a few hundreds of millivolts vs. Li/Lit) but is kinetically slow due to the already existing passivating layer on the negative electrode surface. As stated above, the rates of the selfdischarge processes on the two electrodes in this study were similar leading to little permanent capacity loss. The majority of the self-discharge observed in commercial lithium-ion cells is reversible with only a very small fraction irreversible. The mechanisms proposed in the literature (coupled electrolyte decomposition, discussed above) lead to irreversible losses in many cases because of the consumption of cyclable lithium (to form products such as lithium carbonate) or the physical blockage of active electrode surface area. Self-discharge mechanisms which do not lead to permanent capacity loss are needed. One possible process would be the transport of oxidized solvent species from the positive to negative electrode where reduction would occur, in other words, a redox shuttle process. These species could be reversibly oxidized and reduced similar to internal shuttle mechanisms discussed earlier, or could be destroyed upon reduction adding to the carbon electrode's natural passivation layer. As long as the same number of electrons were involved in both the oxidation and reduction processes, the cell's capacity balance would survive these processes and the self-discharge would be recoverable. Another reversible contribution to self-discharge in lithium-ion cells can be attributed to the leakage currents through the separator of the cell. This leakage current may be increased due to any number of imperfections in the manufacturing process, such as pinholes in the separator. Because of the need to conserve current flow, leakage currents due to finite separator electronic conductivities are balanced by the electrochemical discharge of the cell. This process would likely occur at very low rates, limited only by the electronic resistance of the separator. Because this process is expected to be only weakly dependent on temperature, the fact that experimental self-discharge data on lithium-ion cells are strongly temperature dependent suggests this mechanism is not a primary one. Interfacial Film Formation A passive film is formed at the negative electrode/electrolyte interface because of irreversible side reactions that occur between lithium ions and/or the solvent and electrode surface (see Eq. 28 to 46). Certain aspects of these processes were discussed in previous sections. These side reactions will ideally form a stable, protective film on the negative electrode allowing the electrode to continue to operate without further reaction. The initial loss of lithium ions in forming this film causes the capacity balance between the two electrodes to change. This may result in a diminished utilization and hence a decreased specific energy for the entire battery. The irreversible capacity loss with carbon electrodes can vary between 10 and 100% of the reversible capacity for different types of carbon. The capacity lost depends on the type of carbon used, the components of the electrolytic solution, and additives to the electrode or solution. Passive film formation is different than lithium deposition which occurs during negative electrode overcharge. The passive film can form on either electrode and consists of products (Li2CO3etc.) formed by electrolyte decomposition. Peled has explained many of the fundamental processes taking place at the lithium and lithiated carbon electrode/electrolyte interfaces and has put forth models to explain these interfacial phenomena. The solid electrolyte interphase (SEI) passivation layer on the carbon surface plays a major role in determining electrode and battery behavior and properties including cycle life, shelf life, safety, and irreversible capacity loss. The major role of the SEI is to separate the negative electrode from the electrolyte, to eliminate (or to reduce) the transfer of electrons from the electrodes to the electrolyte, and also the transfer of solvent molecules and salt anions from the electrolyte to the electrodes. The actual morphology of the SEI is Very complex and changes with time and with electrolyte composition. It is best described as a thin heterogeneous film of a mosaic of numerous individual particles of different chemical compositions in partial contact with each other at grain boundaries.The SEI passivating layer over metallic lithium is a more porous or structurally open layer of corrosion products which, to some extent, blocks the surface of the anode and do not take part in the deposition-dissolution process. When a carbonaceou electrode is cathodically polarized to potentials lower than 2Vvs. Li/Li+, many side reactions take place simultaneously as shown in Fig. 5. Peled studied the lithium metal/polymer electrolyte interface in detail The polymer electrolytes used were LiI-PEO-A12O3 based composites. He found that a single parallel RC element representing the apparent resistance (RSEI), apparent capacitance (CSEI), and apparent thickness (LSEI) of the SEI layer could be used to fit the impedance response of the interface and predicted a parabolic growth rate for SEI films.The proposed model cannot be generalized to conditions outside the ones on which they have been derived and therefore cannot be used to develop detailed design specifications. A more comprehensive and general model based on first principles needs to be developed which will be valid for lithium-ion electrodes under varying conditions. The deposition-dissolution process of a solid electrolyte interface film involves three consecutive steps: (1) electron transfer at the metal/solid electrolyte interface; (2) migration of cations from one interface to the other; (3) ion transfer at the solid electrolyte/solution interface. M — ne-n-苯基咔唑→Mn+M/SE Mn+M/SE→Mn+SE/sol m(solv) + Mn+SE/sol →Mn+SE.(solv) The passive film formation over the electrode can be explained by a simple heterogeneous reaction between the electrolyte solvent (El) and LixC6 LixC6 + δEI= δLiEI + Lix-δC6 The SEI model assumes that the passive layer formed over the surface prevents further reaction but still allows lithium ions to pass through. The above reaction can be separated into two steps LixC6= Lix-δC6+δLi++δe- δLi++δe-+EI1iquid→δLiEI (solid) The above two reactions show that the process will proceed only if electrons are transferred through the film from the carbon to the SEI/electrolyte interface, or if solvent molecules are transferred from the electrolyte to the SET/carbon interface. Solvent molecules could penetrate the film either through imperfections, such as cracks, or if they are sufficiently mobile through the bulk of the SEI film. Garreau have proposed a polymer electrolyte interphase (PET) model to describe the behavior of the lithium metal electrode covered by a porous nonconducting polymeric film. The charge-transfer reaction is limited by the surface coverage of the lithium electrode by the PET and with the diffusion of electrolyte through the porous structure of the interface. The composition of the electrolyte is a determining factor in the nature of the passivating film formed. The carbonate-based solvents react with the carbon, lithium, or lithiated carbon to form alkene gases and Li2CO3 as the primary film-forming material, as discussed earlier in the solvent decomposition section. Current Collectors Copper and aluminum are the most commonly used current collectors for negative and positive electrodes, respectively In addition to these metals, nickel and stainless steel have also been tested for current collectors in lithium-ion cells. The main issues related to current collectors are passive film formation, adhesion, localized corrosion such as pitting, and general corrosion. These phenomena increase the internal impedance of the cell during cycling and can lead to capacity and rate capability losses. There are still relatively few studies of corrosion processes in lithium-ion cells; however, as the challenges in this area receive more attention, more work and advances in understanding can be expected. Both current collectors in lithium-ion cells are susceptible to degradation, with Al to pitting corrosion and Cu to environmentally assisted cracking.The pitting of aluminum in PC/DEC and EC/DMC electrolytes was studied using impedance spectroscopy, XPS, and Auger techniques by Braithwaite and coworkers. Al became more passive on cycling, and Li and P were the predominant surface species observed on the Al surface. These authors demonstrated that chromate conversion coatings provide good protection to Al in lithium-ion cells. Environmental cracking of Cu can occur at or near the lithium potential if specific metallurgical conditions exist such as work hardening. The passive film on the copper current collector in these environments was relatively thin (<150 A) and did not appear to thicken during cycling. Recently Pistoia reported that LiPF6-EC/DMC electrolytes corrode Al nets at 3.1 to 3.2 V and Al foils at 4.2 V vs. Li/Lit depending on their HF content. In LiBF4-EC/PC and LiC1O4-EC/DMC, Al foils do not corrode below 4.9 V and on graphitized Al. several electrolytes can withstand potentials above 4.5 V. Chen examined Al current collectors using scanning electron microscopy (SEM) after charging at various potentials in lithium/poly(ethylene oxide) LiN(CF2SO3)2/V6O3 or TiS cells. They observed pitting corrosion on Al current collectors which affects the long-term reliability of lithium-polymer batteries. An alternate corrosion-resistant W-Al alloy was proposed which forms a more protective corrosion product film over the current collector surface. Both of the current collectors in commercial lithium-ion cells are pretreated (acid-base etching, corrosion-resistant coatings, conductive coatings, etc.) to improve their adhesion properties and to reduce corrosion rates. These pretreatments help significantly for both Al and Cu current collectors. In the case of Al, operation without any prior surface treatments leads to substantial increases in interfacial resistance over the life of the battery. Loss of current collector adhesion can dramatically impact cell capacity because portions of the electrode may become totally disconnected from the conducting matrix. With Cu, disconnected regions can even promote lateral variations in electrode potential which may force lithium deposition to occur. Current collector corrosion can lead to an increase in the battery's internal resistance over many cycles due to the formation of an insulating film of corrosion products on the surface of the current collector. The increase in internal resistance depends critically on the treatments used on the current collector interface prior to cell assembly and causes the battery to lose some power capability later in its life. A loss in rate capability (or increase in internal resistance) can lead indirectly to capacity loss when the capacity is assessed at a given rate. The dissolution of the copper current collector is a possible overdischarge reaction at the negative electrode(Fig. 3) Cu - Cu + e Univalent copper is more likely than divalent in a nonaqueous environment.The thermodynamic equilibrium potential for this reaction in an aqueous solution under standard conditions is 0.521 V vs. SHE or 3.566 V vs. Li/Lit. The potential of carbon-based negative electrodes near the end of discharge or under overdischarge conditions may reach in excess of 1.5 V vs. Li/Li+. Apparently, this reaction occurs much more readily than expected because of the nonaqueous environment in the cell, where standard thermodynamic data taken in aqueous media no longer hold. The copper ions that are formed on overdischarge can redeposit later as copper metal at the negative electrode, forming dendrites which may penetrate the separator and cause cells to fail catastrophically. These processes generally prevent lithium-ion cells from being discharged below approximately 2.5 V. This is a serious problem in bipolar stacks of cells as it is difficult to control each individual cell voltage. Thus, a cell may be over discharged because it has a lower capacity, and this could destroy the whole stack. For other consumer electronics applications, it would be advantageous for lithium-ion cells to have better overdischarge tolerance. Positive Electrode Dissolution Positive electrode dissolution phenomena are both electrode and electrolyte specific, and limited data on these processes are available for most materials. The factors determining positive electrode dissolution are structural defects in the positive active material, high charging potentials, and the carbon content in the composite positive electrode. Oxygen defects in the LiMn2O4 and LiNiO2 structures may weaken the bonding force between the transition metal and oxygen, resulting in Mn and Ni metal dissolution. The transition metal ions which have weak bonding forces with oxygen could be driven into the electrolyte, especially when polarized to high potentials.Electrolyte oxidation on the carbon black surface may generate catalytic species which can increase the rates of metal-ion dissolution. Of the three main high-voltage metal oxide positive electrode materials, cation dissolution has been most studied to date on the lithium manganese oxide spinel. Dissolution of the LiMn2O4 active material occurs through a disproportionation mechanism leading eventually to manganese deposition on the negative electrode. This causes a loss of positive active material and blocking of pores in the negative electrode. The Mn2+ dissolution reaction is believed to occur when the LiMn2O4 electrode is fully discharged, and can be a major problem in the shelf life of discharged cells. These processes have been studied by a number of authors including Thackeray and coworkers and the Beilcore research group. The capacity fading upon cycling was first ascribed to the dissolution of Mn by Thackeray.Tarascon and coworkers detected the presence of Mn on the surface of the negative electrode by Rutherford backscattering spectroscopy (RBS). Wen reported that the capacity fade on cycling in the higher voltage region was attributed to the fact that the active electrode material was gradually converted to a lower voltage defect spinel phase via the dissolution of Mn into the electrolyte. Recently Xia reported that capacity loss caused by the simple dissolution of Mn3+ accounted for only 23 and 34% of the overall capacity loss observed at room temperature and 50℃, respectively. They suggested that the rest of the capacity fade originated from structural changes and decomposition of the electrolyte solution. According to Bellcore and Dahn,25% of the Mn2 dissociated ends up depositing on the negative electrode surface. This occurs via the following mechanism 4H++ 2LiMn3+Mn4+O4-→3λ—MnO2+Mn2+ 2Li+ +2H2O which is the acid-induced decomposition of the spinel.This disproportionation process is due to the instability of the Mn3+ oxidation state, which reacts spontaneously to form Mn2+ and Mn4+ Mn2 then goes into solution and redeposits as Mn(s) at the negative electrode Mn2++ 2e-→ Mn(s) It could also be the case that colloidal LiMn2O4 particles move toward the negative electrode by electrophoresis and deposit manganese at the negative electrode. However, if electrophoresis were taking place, the amount of manganese determined at the negative electrode should depend strongly on the particle size of the electrodes, which was not the case. The above reaction (Eq. 59) is greatly accelerated with increasing temperature. The protons arise from HF, which originates with the hydrolysis of LiPF6 salt and thus epends critically on LiPF6 purity H2O + LiPF6→POF3+ 2HF + LiF The water produced by the Eq. 59 can also generate more protons, making the manganese dissolution process autocatalytic in nature. Free protons could also be consumed at the negative electrode via 2H+ + 2e-→H2(g) The Mn2+ dissolution reaction can be reduced or slowed by using high purity LiPF6 and low surface area LiMn2O4 (<1 m2/g). Limiting the surface area also reduces the catalysis of side reactions such as solvent oxidation. There is experimental evidence that the hexafluoropho-sphate anion is more directly involved in the manganese dissolution process, perhaps through an anion-assisted mechanism. Jang stated that manganese dissolution is the primary reason for capacity losses in LiMn2O4 electrodes. In this study, manganese dissolution brought about an increase in contact resistance at the manganese-depleted_ spinel/carbon interface and also increased the electrode reaction resistances for lithium insertion-deinsertjon They reported experimental data for the dependence of manganese dissolution on applied potential. The dissolution rate was not appreciable when the applied potential was below ca. 4.0 V vs. Li/Li, but it became notably high above 4.0 V. According to these authors, the disproportionation mechanism (Eq. 59) seems unlikely to be a cause for the dissolution because dissolution was seen predominantly at the end of the charging process, in which potential range the Mn3 content in the spinel is minimal. In another study, Jang and Oh reported that spinel dissolution is induced by acids that are generated as a result of electrochemical oxidation of solvent molecules on composite cathodes. The spinel dissolution was much higher in the ether-containing electrolytes such as DME and THF as compared to the carbonates. In the initial stages only Li and Mn ions are extracted, while in the later stages oxygen loss becomes dominant. They found that solvent-derived acid generation was not significant in electrolytes containing fluorinated salts (LiPF6, LiBF4, and LiAsF6), yet the spinel dissolution in these electrolytes was appreciable because the acids are generated by the reaction between the F containing anions and impurity water (such as Eq. 61 above). Although the results of appear contradictory to the earlier work, it is likely that a similar mechanism is operating in all cases. Jang have emphasized the important link between the generation of protons in the cell and manganese dissolution. By using ether-based solvents, a much larger quantity of protons is generated thus leading to higher rates of manganese dissolution than would have been seen in earlier studies with oxidation-resistant carbonate solvents. The link between manganese dissolution and electrode potential is due to this potential-dependent solvent oxidation process, not the state-of-charge of the positive electrode as suggested by Jang. Robertson also proposed an alternate mechanism for spinel dissolution. They proposed that a modified proton-catalyzed redox mechanism is responsible for Mn extraction from the cathode and the concurrent formation of LiyMnOx species which were electrochemically inactive at 4 V l2LiMn2O4→Li2MnO3 + 5Li2Mn4O9 + 3Mn2+ solve 12Mn3+→9Mn4++ 3Mn2+ solve+ 6e- oxidation 3Mn2+ solve+ 6e- →3Mn ↓ reduction Li2MnO3 is electrochemically inert, andLi2Mn4O9 has virtually no capacity in the 4 V region. From Eq. 63 to 65, 12 M of LiMn2O4 become inactive for every 3 M of Mn2+ that dissolve into the electrolyte. Moreover, Li2Mn4O9 cycles in the 2 to 3 V domain. They suggested that low levels of Cr3- in the spinel framework can substantially reduce cathodic manganese dissolution into the electrolyte. Based on the evidence for the dissolution of manganese and the change in crystal structure of the spinel during cycling, Xia proposed that manganese dissolves via two routes: (1) LiMn2O4 is transformed to LiMn2+xO4-x, via loss of MnO, and, (ii) LiMn2O4 is transformed to Li1+xMn2-xO4, via loss of Mn3O4 For the first case, a portion of Mn3+ transforms to Mn4+ together with dissolution of MnO into solution, i.e. Mn3+(LiMn2O4) → Mn4+(Li2Mn3O7) + Mn2+(MnO) Amatucci reported a strong and direct relationship between capacity loss and percentage of cobalt detected on the negative electrode for LiCoO2-based lithiurn-ion cells charged above 4.2 V. The capacity loss (or Co dissolution) depended on the thermal treatment during synthesis of the active material. Cobalt ions present in the electrolyte after dissolution are reduced at the negative electrode. The rate of dissolution increases with cutoff voltage, with a steep increase when the cutoff voltage is 4.5 V A quantitative relationship between the percentage of cobalt dissociated and capacity loss for LiCoO2 cells with a cutoff (charging) voltage of greater than 4.2 V was given by these authors. In certain lithium-vanadium oxide (LiyV2O5) cells based on an LiAsF6-EC/PC/2Me-THF (15:70:15) electrolyte, vanadium was found to dissolve partially and then plate on the lithium electrode leading to an increase in cycle life.The vanadium incorporated into the lithium surface film was shown to be beneficial to the passive film properties in this electrolyte. This work demonstrates the electrode material and cell-specific nature of the dissolution process and its impact on battery performance. Phase Changes in Insertion Electrodes The understanding of the relationship between phase changes in insertion electrodes and capacity loss is weak even though this is widely quoted as a mechanism of capacity loss. The basic mechanism stated is that phase changes or large changes in lattice parameters leads to fracture of particles and loss of contact from the electrode matrix. In general, one might believe that good electrodes showing high reversibility and cycle life probably are not accompanied by significant phase changes or lattice expansion or contraction during operation. The three main metal oxide insertion compounds most studied currently for lithium-ion batteries are in this category. However, the constant search for higher capacity materials makes phase changes and structural changes difficult to avoid, and the effects of these processes on battery performance is only beginning to be quantified and studied in a fundamental manner. The phase changes occurring in lithium-ion cells can be classified into two types: those that occur during normal insertion-deinsertion of lithium, and others that occur when the positive electrode is overcharged or overdischarged (for example, the Jahn-Teller distortion in the case of overdischarge of LiyMn2O4 above y = l).Xia and Ohzuku have studied the phase changes occurring in spinel LiMn2O4 electrodes. Xia concluded that the transformation of unstable two-phase to a more stable one-phase structure occurs via loss of MnO (Mn3+→Mn4+→MnO), which dominates the capacity fading during the room temperature cycling of cells. Amatucci described the insertion of lithium into LiMn2O4 as single phase and biphase. The biphase insertion scheme as proposed by them is Delithiation of LiyMn2O4 leads to formation of λ-Mn2O4 that decomposes to β-MnO2, at temperatures ranging from 100 to 250℃, and in some cases β-MnO2 may be detected as an intermediate in the process of decomposition. Rutile β-MnO2 is electrochemically inactive at 4 V and leads to the formation of orthorhombic LiMnO2 on lithium insertion. These phase changes during cycling may lead to capacity losses in practice. Recently, Cairns have used in situ X-ray adsorption techniques to determine that upon delithiation of LiMn2O4, the lattice contracts to a structure similar to λ-Mn02, whereas lithiating LiMn2O4 transforms the cubic spinel into the tetragonal phase Li2Mn2O4. Another major reason for capacity fade in LiMn2O4 spinel-based cells is overdischarge leading to a lower than 3.5 average oxidation state for manganese.This may cause a Jahn-Teller distortion of the LiMn2O4 structure that occurs when the oxidation state drops below During the Jahn-Teller distortion, the z axis stretches by 15% and the x and y axes contract by 6%. The large anisotropic expansion (16%) of the unit cell is too severe for the tetragonal Li2Mn2O4 phase on the surface and cubic LiMn2O4 in the bulk to remain as one inter grown structure within a single crystallite. Particle fragmentation may result (giving higher surface area), causing the Mn3t dissolution reaction to become more troublesome and leading to loss of contact between particles. This structural damage to the spinel electrode may lead to slow capacity loss on Cycling. The Jahn-Teller distortion process is greatly reduced by using an excess of lithium in the LiMn2O4 starting material (x = 1.05 or higher). Lithium substitution into manganese sites leads to an increase in the average oxidation state for the manganese. This has also been accomplished by doping LiMn2O4 with other atoms such as cation substitution with Fe or Co, or anion substitution with F. Lithium manganese oxide spinel electrodes stabilized in some manner to prevent overdischarge can exhibit constant capacity during cycling for several thousand cycles at room temperature. Ohzuku concluded that the reduction of LiyMn2O4 proceeds in three steps, which consist of a cubic/cubic two-phase reaction, a cubic one-phase reaction, and a cubic/tetragonal two-phase reaction as shown in Table II. A large voltage drop of 1 V is observed for y＞1, which is due to a Jahn-Teller distortion of MnO6-octahedron from Oh symmetry to D4h symmetry. According to these authors, the apparent stabilization energy of MnO6-octahedron (Oh D46) results in a loss of 1 V. The abrupt change in unit-cell volume (6.5%) which accompanies the cubic-to-tetragonal phase transition in the spinel has a deleterious effect on the cyclability.Ideally, tetragonal LiMn2O4 should be present at the end of each discharge. The cubic LiMn2O4 accumulated at the end of discharge is the primary cause of the significant loss in capacity. A number of other lithium manganese oxide structures have been synthesized and tested for insertion-deinsertion capacity and reversibility over the past several years.In general, these compounds revert to the spinel structure over the course of cycling due to the strong driving force to achieve the thermodynamically most stable phase. These phase changes are often also accompanied by capacity losses, although the mechanisms for these losses probably vary from case to case. Phase changes are also observed in the case of LiCoO2 and LiNiO2 electrodes. The different phases and the corresponding y values for both electrode materials are given in Table II. In the case of LiCoO2, various hexagonal and monoclinic phases are observed, whereas in the case of LiNiO2, monoclinic and rhombohedral phases are formed. Although phase changes occur in these electrodes, they have not been associated with capacity losses. The LiyNiO2 electrode is usually cycled between y = 0.3 and 0.9, whereas LiyCoO2 is cycled between p = 0.5 and 1.0, to avoid significant phase changes during cycling. Gummow and Thackeray synthesized LT-LiCo0.9Ni0.1O2 with a structure that is intermediate between an ideal lithiated spinel and a layered structure.They reported improvements in electrochemical performance and attributed it to the formation of a defect spinel phase Li0.8[Co1.6Ni0.2]O4 in which the lithium ions adopt the tetrahedral A sites and the cobalt and nickel ions the B sites of an A[B2]O4 spinel. In carbon insertion electrodes, during the charging cycle a reversible expansion of interlayer spacing occurs along the c axis.Lithium intercalation in graphite shows a structural transformationfrom the ABA to the AAA structure. These structural changes in graphite electrodes during intercalation/ deintercalation of lithium have not usually been associated with capacity losses. Incorporation of capacity Fade Mechanisms into Battery Modeling Modeling capacity fading and failure mechanisms requires that side reactions be incorporated into a general lithium-ion battery model. The basic procedure for accomplishing this is discussed here; followed later by some elaboration within the context of particular capacity fade mechanisms. Either heterogeneous or homogeneous reactions can be included, although only heterogeneous electrochemical side reactions are treated here. Past battery modeling work on other systems has included many of the phenomena discussed here, such as the incorporation of the oxygen evolution side reaction into nickel-cadmium and lead-acid battery models.The basic approach to incorporating side reactions is to write a rate expression for the side reaction in question and material balances, for each species involved in the reaction. Various approximations might be used to reduce the number of equations beyond this point. A thorough discussion of the macroscopic approach to full-cell sandwich battery modeling can be found in the literature. A general electrochemical side reaction can be expressed as where st is the stoichiometric coefficient of species i and n is the number of electrons transferred in the reaction. Because the rate of side reactions will often be kinetically limited, we first discuss the formulation of kinetic rate expressions for electrochemical reactions. A Butler-Volmer type rate expression is written for each reaction as Where the exchange-current density o,k has the functional Form and the overpotential (liok) driving force for the reaction is where The potential variables Φ1 and Φ2 represent the potentials in the solid and solution phases, respectively, of either electrode. Uk is the thermodynamic open-circuit potential of the side reaction, and io泡面碗o,k,γ1,αak and αck are the kinetic parameters. The kinetic data will depend on the specific electrode materials in use and the composition of the electrolyte solution. U corresponds to the thermodynamic potential of reaction k under standard conditions. An arbitrary number of side reactions can be included, each having its own overpotential defined above.If the reactions are treated as occurring independently, the total current density at each electrode is a summation over the rates of all reactions Using this treatment, depending on the value of the local potential and the thermodynamics of each reaction, cases can result where both anodic and cathodic reactions will occur on the same surface. This treatment will also predict the occurrence of corrosion or self-discharge processesunder open-circuit conditions. Obviously, the rates of all of the reactions are coupled through the potentials of each phase in the porous electrode and a numerical solution of the governing equations is necessary. For many side reactions, the assumption of an irreversible reaction will hold, in which case the above rate expression can be simplified. A Tafel expression can be inserted in place of the Butler-Volmer equation For a side reaction obeying a Tafel equation, the value of the open-circuit potential becomes arbitrary because there is no longer a backward reaction. Thus, the only kinetic parameters of interest are the exchange-current density and Tafel slope (or transfer coefficient). Once kinetic equations have been developed and rate expressions written, the number of species existing in the system and their relative importance can be assessed. Material balances can be written on each species existing in the system, although the concentrations of species involved in homogenous chemical reactions might be related through reaction equilibria. The material balance for species i can be written in the form where c1 and Nt are the concentration and molar flux of species i. The net rate of production of species i is given by Ri. This quantity applies to either bulk homogenous reactions or heterogeneous reactions if the latter are treated using a pseudohomogenous averaging approach. Under porous electrode theory, the rate of production of a given species can be related to the partial current densities of the electrochemical reactions according to where ak is the specific interfacial area per unit volume of electrode. If the electrochemical reaction is localized onto a particular surface, such as the current collector the generation of material can be included in a boundary condition. Different approximations will be valid depending on the phase of the species (solid, liquid, gases) and the concentrations present. In many cases, the rate of a side reaction will be low enough or the quantity of a species generated small enough that a material balance can be neglected entirely. Often transport processes in the cell are described using concentrated solution theory which accounts for interactions between each pair of species. With a larger number of species present due to side reactions, simplifications may be made to this rigorous treatment to reduce the number of transport properties required in the model. For example, the electrolyte solution might be treated as composed of a primary solvent, a single salt, and an arbitrary number of solution-phase pseudo-dilute species. Interactions between the primary solvent and salt are treated rigorously as a binary electrolyte system, while the dilute species are treated as interacting with the primary solvent only. This simplified treatment generates (3 + n) transport properties or Dij's where a is the number of dilute species rather than the (3 + n)(2 + n)/2 required in the more rigorous treatment. Neglected in this treatment are interactions between the various dilute species and interactions of dilute species with the salt. Under these simplifications, the flux of a dilute nonionic component might be written simply as where Do1 is an interaction parameter or diffusion coefficient representing motion of species i through the solvent. The flux of a dilute ionic component is composed of the above diffusive flux as well as a migration components If the concentration of a dilute ionic component is low enough, migration can be neglected entirely because the primary salt will act as a supporting electrolyte. In both of the above expressions, convective fluxes are neglected which is usually a good approximation. These flux expressions are substituted into the material balance above (Eq. 76) to provide an equation for the concentration of a dilute species as a function of time and position across the cell. Initial and boundary conditions are needed to complete the problem. The phase of a given species is an important consideration in determining the level of sophistication required in the model. For example, gaseous species might be treated as existing purely in a vapor space above the cell, and their concentrations might be assumed uniform because of fast transport of gas-phase species. Under these assumptions, the pressure developed inside the cell could be assessed (assuming the system is at constant volume) and the relative partial pressures of each species determined. In other cases, the solubility of gases in the electrolyte may be an important factor in cell performance or self-discharge processes, in which case these assumptions would not be made. The three phases (solid, solution, and gas) can be treated as superimposed continuous phases without regard to the geometric details of the pore structure using porous electrode theory. Solid-phase species with low solubility might be treated as localized, such as with the precipitation of salts in previous battery modeling literature.Reactive intermediates, such as radical species resulting from solvent oxidation or reduction processes, are often treated using the pseudo-steady-state approximation, wherein the net rate of formation of the species is assumed to be zero.Further consequences of these assumptions will be discussed in the remainder of this section. Film growth.—Side reactions that lead to growth of a film on the surface of the electrode particles are seen in several instances within the lithium-ion battery. Passivation of the carbon-based negative electrode with associated film or SEI layer formation is a standard process in all cells. Abuse conditions such as overcharge can lead to film formation from the deposition of metallic lithium onto the negative electrode. Any lithium metal formed in the cell will probably undergo secondary reactions leading to more thick reaction product layers or secondary films. Other side reactions can also lead indirectly to film formation because of the formation of products with low solubility in the nonaqueous solvents. We start by discussing film formation due to overcharge of the negative electrode and deposition of metallic lithium because this process is possible in all lithium-ion batteries in theory and presents an important safety concern with commercial cells. A number of different approximations can be used to include the lithium metal deposition side reaction in lithium-ion battery models. For this reason, the detail of a model is driven primarily by its intended usage, with models having more predictive power needing a greater level of detail. Before decisions can be made on the sophistication needed, overcharge experimental data for lithium-ion cells using coke and graphite as negative electrodes are needed. Based on these data, the effects of overcharge on the discharge and cycle life of the battery can be assessed, and the phenomena that must be incorporated into models determined. Also, kinetic data for the lithium deposition reaction are needed under the appropriate conditions. Because the lithium deposition reaction is a facile process under many conditions, the surface overpotential will be low and the reaction can be described adequately using the linear approximation to the Butler-Volmer equation For solids such as metallic lithium, the flux will be zero to a very good approximation. By integrating the material balance over the volume of the negative electrode, the rate of the side reaction can be related to the growth of a film on the surface of the electrode particles where δ is the film thickness composed of solid lithium and other lithium products, and L- represents the thickness of the negative electrode. This film thickness, along with assumed properties for the film such as conductivity and dielectric constant, can be incorporated into a battery model to predict the impact on discharge performance or electrochemical impedance data. A similar approach was taken by Pollard to treat precipitation of salt films in electrochemical cells.As the film thickens dynamically during simulation of the cell charge and discharge, the interfacial resistance will increase and the current distribution in the electrodes will change. Also, the capacity balance in the cell is modified due to the side reaction current density. Other effects can then be incorporated such as porosity changes in the electrode and secondary side reactions with the deposited lithium. Another side reaction of general interest in lithium-ion batteries is passive film formation on the negative electrode during the initial cycling or formation period. Modeling passive film formation is similar to the modeling of lithium deposition during overcharge conditions. However, the reduction reactions taking place which lead to the deposition of solid products are much less well understood, larger in number, and varied in their nature depending on the composition of the electrolyte solution as seen in an earlier section. For these reasons, a great simplification might be made where the passivation process is treated as the consumption of solvent and lithium ions to form a single homogeneous product such as lithium carbonate. The passivation process at the electrode/electrolyte interface can be modeled to allow the film thickness and other relevant properties to be tracked during cycling. Other intermediate species formed such as radicals could be either neglected completely or treated in a simplified manner (for example, these might be included to track the rate of selfdischarge). By following the rate of solvent, lithium ion, and electron consumption by the passivation process, the important effects of the side reaction on the capacity balance in the cell can still be assessed. Under the assumptions given above, the film is modeled as a one-dimensional layer having a time-dependent film thickness. The film thickness is calculated from the partial current density of the reduction side reaction. Because the exact nature of the side reaction is either not known or is being simplified greatly, the rate expression used would probably be obtained by empirical fits to experimental data such as charge and discharge curves during the formation period. The film thickness calculated from the model is related to a film resistance and capacitance using assumed physical properties. Because the resistance of the film changes with time, an increase in interfacial resistance and a capacity loss in the cell will result. Assuming that the rate-limiting step in the charge-transfer process is bulk migration of lithium through this passive layer as stated earlier, then the treatment of the film as a pure resistor is justified. In other cases, the finite rate of the charge-transfer process at either interface may also be an important consideration. Finally, if the film has some porosity or if it is not a pure cation conductor, then diffusion limitations inside the film might also be included. Electrolyte decomposition reactions—A large number of different solvent and salt combinations are used in lithium-ion batteries. Because of the complexity of the mechanisms of salt and solvent oxidation and reduction reactions, it is not possible at present to create a general model that can treat all possible systems. Future modeling work in this area will almost certainly be confined to specific and well-studied systems. The modeling of electrolyte decomposition reactions is not difficult mathematically, but for complicated multistep mechanisms, the number of equations and unknowns make the solution procedure computationally demanding. Kinetic data for each reaction and transport property data for each species are in theory needed. It should be possible to treat these systems by making use of standard approaches such as the psedo steady-state or rate-limiting-step approximate methods. A rigorous treatment of side reactions involving the solvent and salt species in lithium-ion cells would require substantial amounts of fundamental data on the rates of these processes so that information such as rate-limiting steps can be assessed. Fortunately, for the practical purpose of treating side reactions in battery modeling and predicting their consequences to cell performance, this level of detail is probably unnecessary, and advances in the understanding of these processes can be made using the present state of knowledge in the literature. Side reactions involving both oxidation and reduction of components of the electrolyte solution are lumped together in this section because similar issues exist in modeling both situations, such as the loss of solvent or salt from the system, formation of product species which may take part in secondary processes or reactions, and perturbation of the cell's capacity balance. In many current lithium-ion battery models, the opencircuit potential of the positive electrode is forced to inf inity as the value of y reaches its minimum value (e.g., U→∞ as y→0.4 for LiyCoO2) and no side reactions involving electrolyte oxidation are considered. Applying this condition to the open-circuit potential for the insertion electrode stops the insertion process at the desired stoichiometry in an artificial manner by forcing the electrode polarization to infinity. Side reactions (both chemical and electrochemical) could be included in a battery model as secondary processes that proceed in parallel to the main reaction with their own separate kinetic expressions. The electrochemical side reactions depend on the local electrode potential and are often treated adequately using Tafel equations due to the irreversible nature of many of these processes.Chemically or thermally induced degradation processes do not depend on electrode potential and might be more important under conditions of abuse. The presence of these side reactions can have a number of dramatic consequences in a battery model. By tracking the rates of the oxidation or reduction processes occurring as side reactions, the actual lithium stoi chiometry in the two insertion electrodes is predictable even during abuse conditions. This allows capacity fading to be predicted as well as the potential hazards associated with accumulated loss of capacity in one electrode. In many cases, the loss of salt or solvent from the cell will represent a very small fraction of the total electrolyte, and this effect on the cell's performance might be neglected entirely. However, over long-term cycling of the cell, it is possible that the accumulated loss of electrolyte from the system will begin to manifest itself in losses of rate capability or cell capacity. Products formed in the side reactions can also be monitored and their role in secondary processes such as self discharge followed. These products might be solution phase, gaseous, or solid depending on the nature of the side reaction, and in each case different approximations are warranted as discussed earlier. In some cases, the active electrode material itself may take part in the side reaction. For example, the acidinduced delithiation and dissolution of the manganese oxide spinel material is a case where solvent oxidation or salt hydrolysis processes lead to the formation of products (protons) that participate in secondary reactions with the electrode material. Often these processes will lead to formation of electrochemically inactive products either due to blockage of the electrode particle surface due to precipitation of solid-phase species or because of conversion to materials with either limited reversibility or no lithium insertion capacity in the voltage range of interest. Modeling these processes requires a mass balance for each species involved in the side reactions, especially those related to consumption of active materials and formation of inert products. A more simplistic method of accounting for the formation of inert or nonelectrochemically active metal oxide compounds is to adjust the volume fraction of active material as the side reaction proceeds, increasing the volume of inert filler in proportion to the rate of active material consumption. For example, by tracking the production of protons inside the cell under abuse conditions (such as solvent oxidation), the rate of electrode dissolution and conversion to nonactive materials can also be followed. These nonactive materials could be treated as uniformly distributed throughout the composite electrode similar to the standard treatment of polymer binders and carbon-black additives under porous electrode theory. Corrosion and dissolution processes—Additional ionic species can make their way into the cell due to the processes of electrode dissolution and current collector corrosion. Cationic species other than lithium can create problems for cell performance because they are more easily reducible than lithium ions and will generally end up depositing on the negative electrode during charging conditions. In the worst case, these processes lead to dendritic growth through the separator and short circuit of the cell. The presence of deposited manganese and cobalt have been mentioned in the literature, and corrosion of current collector materials such as copper and aluminum is also well documented. Modeling these phenomena appears challenging due to our limited understanding of the fundamental processes occurring and the fate of metallic impurities inside the nonaqueous environment of the cell (other than their eventual likely fate at the negative electrode). For example, it would be useful to predict the onset of copper dissolution at the negative electrode under overdischarge conditions. This might allow more optimum design of cells to limit the maximum negative electrode potential and prevent significant loss of copper. Unfortunately, the copper dissolution-deposition voltage is difficult to predict because very little data are available for nonaqueous systems. For any given nonaqueous electrolyte system, the thermodynamic open-circuit potential for copper dissolution must be assessed experimentally. If these data were known or assumed, the copper dissolution reaction could be modeled in a manner similar to the treatment of lithium deposition during negative-electrode overcharge by using a Butler-Volmer-type rate equation and a material balance on solution-phase copper species. However, the growth of dendrites leading to cell short circuit is more difficult to predict because this is a function of deposition morphology. The related topic of electrode dissolution in LiMn2O4-based cells can be included in battery models by adding the acid-assisted dissolution mechanism discussed earlier as a side reaction. Equation 59 occurs on the surface of the positive electrode, LiPF6 hydrolysis is a homogenous chemical reaction, and H2 evolution occurs at the surface of the negative electrode. Note that these processes are potentially autocatalytic as water generated by the dissolution of the lithium manganese oxide spinel can create more protons from hydrolysis of the salt. These reactions can be included in a cell model by writing material balances for each species, although a number of simplifications are sure to be warranted. The rate equation (Butler-Volmer expression for electrochemical reactions) for each reaction could be written in the same manner as shown previously (Eq. 70). Positive active material is lost when electrode dissolution occurs leading to capacity loss. Also, dissolved manganese is reduced at the negative electrode and blocks pores or surface area again leading to capacity loss. Self-discharge processes—Although self-discharge is an important phenomena for practical commercial cells, it has received relatively little attention in the battery modeling literature. However, the modeling of self-discharge processes is straightforward once a full-cell-sandwich battery model exists and once the necessary side reactions have been incorporated into the model. Self-discharge processes can be separated into those involving a single electrode such as the coupled electrolyte decomposition and lithium insertion-deinsertion reactions discussed earlier, and those that involve interactions between the two electrodes such as redox processes and short-circuiting. Self-discharge processes involving a single electrode can be treated by simply adding the kinetics for the electrolyte decomposition reaction as well as Eq. 74 to the overall mathematical model. For example, self-discharge of a high-voltage positive electrode would be modeled by including the oxidation of solvent species such as the carbonates using a Tafel rate expression.35 Because of the behavior of the Tafel equation, the solvent oxidation process would occur at all voltages but would increase dramatically at higher voltages dependent on the values used for the exchange-current density and transfer coefficient for the oxidation reaction. When the current is interrupted in a model of this kind, rather than attaining a constant voltage, the cell voltage will decrease spontaneously as lithium intercalation into the positive electrode is balanced against the solvent oxidation process. Other self-discharge mechanisms could also be included in a macroscopic battery model without much trouble. Leakage currents are included in modeling by carrying out the constant resistance discharge of the cell under otherwise open-circuit conditions. It might be possible to measure the proper value of resistance to use for the discharge or this value might be fit against experimental self-discharge data. Redox-shuttle-type self-discharge mechanisms require a source for the redox species, such as a solvent decomposition reaction, as well as a material balance on the species in question to predict its transport rate across the cell and concentration (both of which impact the self-discharge rate). Self-discharge processes have been verified experimentally in the literature described earlier, but more experimental data will be needed to validate model predictions using well-characterized systems. Acknowledgments The authors acknowledge the financial support from the Office of Research and Department of the United States Central Intelligence Agency for this project under contract no. 93-F148100-l00. The authors also acknowledge useful discussions on capacity fade mechanisms with Dr. A. S. Gozdz of Bellcore. Manuscript submitted November 18, 1997; revised manuscript received April 17, 1998. The University of South Carolina assisted in meeting the publication costs of this article. | 摘要 锂离子电池容量随着循环衰减。容量损失或者衰减的发生主要是由于以下几种反应机理，这些机理起因于或者关联于一些我们不希望发生在电池里的副反应。这些反应发生在过充或者过放中，导致了电解液分解、钝化膜的形成、活性物质溶解和其他现象形成。这些容量损失机理并没有包含在目前我们可接触到的公开的锂离子电池数学模型中。因此，这些模型并不能用在预测电池循环或者滥用条件下的电化学行为。这篇文章提出了当前锂离子电池容量衰减机理的观点，并且试图描述我们需要的信息和方向，这些信息和方向有可能被引入先进的锂离子电池模型的机理中。 前言 典型的锂离子电池主要由以下三大部分组成：碳(石墨）负极；电解液，主要提供锂离子传送通道并且分隔开两种材料；过渡金属氧化物正极材料（例如LiCoO2、LiMn2O4或者LiNiO2）。这种二次电池最近已经商业化。这种理念下的电池已经进入消费市场。在工业上，交通工具使用的动力电池已经在研究。自从1990年，索尼首次引进商业化电池，锂离子电池市场在一段时期内取得了巨大增长。在许多条件下，锂离子电池的体积能量密度超过130Wh/kg，循环次数超过1000次，锂离子电池体系在手机、笔记本电脑、便携式摄像机的使用越来越普遍。随着越来越多的电池制造商进入市场，新材料得到发展、成本降低加速了电池的新应用。一部分制造商例如索尼、三洋、松下、Moli能源、A&T电池公司已经开始制作锂离子电池用于移动电话和掌上电脑。Yoda 已经考虑到这些进步并且描述了一个未来的电池社会，在这个社会里锂离子电池扮演非常重要的角。 已经有一部分锂离子电池的数学模型被出版。不幸的是，他们的模型里所描述锂离子电池的行为中，没有一种模型明确的描述容量衰减过程。他们的主要目的是回顾现有容量衰减机理模型的相关理解认识。要建立新模型，必须建立在对上述这些基础过程的充分理解和收集整合相关数据的的基础上。 大家所知的导致容量衰减的一些过程是锂沉积（过充）、电解液分解、活性物质溶解、插入电极材料的相变，电极和集流体表面钝化膜形成；定量的分析这些退化反应过程将提高电池模型预测能力，最终我们会设计出更便宜质量更好的电池。在电池能量密度、循环性能等方面需要有重大的提升，尽管我们已经达到了环境友好、成本低的标准。取得这些改进，需要我们对电解液、电极材料、以及导致锂离子电池容量衰减和电阻增加的物理化学反应的基本原理的最新的理论知识有足够的了解。锂离子电池容量衰减模型的发展不仅提供了设计电池的工具并且提供了这些过程如何发生的思维方式。 目前锂电池模型 精细的数学模型的发展对电池后续设计和优化非常重要并且对电池模型规模扩充非常关键。West 提出了类似二维多孔电极模型，该模型解释了具有稳定物理性能的二元电解质的迁移和锂离子扩散到圆柱形的电极；假定嵌入过程的扩散是有限的，因此电解质和活性物质界面的电荷转移阻抗被忽略。随后，Mao和White提出了一个类似的模型， 但是增加了一个多孔电极的分割器。这些模型仅仅覆盖了一个单一的电极；因此，它们并没有优势处理复杂“三明治”型的电池模型里正负两个电极之间复杂的相互作用的现象。这个模型限制了它自己本身，因为在分析锂离子嵌入TiS2的过程的动力学，模型假定嵌入行为是非常快。Spotnitz解释了插入型正极TiS2放电过程中电极的电化学动力学。 Doyle根据浓溶液理论建立了金属锂负极/固体聚合物隔膜/嵌入式正极的恒流充放电模型。这是个通用模型，包括（正负极）分离材料、锂盐、复合嵌入式电极。浓溶液理论用来描述传导过程，因为已经得出结论：离子配对和交联在固体聚合物电解质中非常重要。这种方法也为体积变化提供了稀浓度理论；这个模型中使用Butler-Volmer-type 动力学公式解释每个电极电荷转移的过程。正极锂离子嵌入过程采用Pick's 定律描述，锂离子以恒定速率在活性物质中扩散迁移。在锂/聚合物界面的系统和钝化膜的形成过程中，体积变化忽略不计，但是考虑到一个简化了的恒定的电极界面膜阻抗。该模型不能预测电池长久的退化是由于由于不可逆反应（副反应）或接触面损失。 Fuller 提出了一个普遍的锂离子电池模型，该模型可以应用于任何一对锂离子嵌入式电极和二元电解质体系，并给出必要物理性能参数的。Fuller的模型说明了解锂离子电池嵌入式材料充电状态对开路依赖非常重要。这些曲线的斜率控制了多孔电极材料内部电流分布，开路电压方程曲线的斜率越高，电流分布越均一，因此活性物质利用率更高。Bellcore通信研究所做了一些塑料锂离子电池体系的优化研究。 这些模型也用来预测松弛时间对多个充电放电周期和峰值功率的影响。 Doyle改进了双电极模型，包含了双电极的界面阻抗，对比了LiC6-LiPF6（碳酸乙烯酯/碳酸二甲酯（EC / DMC）），聚偏氟乙烯-LiyMn2O4 体系的实验数据。对比的实验数据和和拟合数据表明体系还有目前电池模型不能预测稍微额外的阻抗存在。电池的放电行为描述比较满意，包括电极材料颗粒表面的膜阻抗，电池电极层之间以及电极和接流体之间的接触阻抗。这项工作的一个重点是为特殊用途的的 Ragone 模块电池模型的设计和优化的运用。 电池热建模对锂离子电池非常重要，因为放电过程中产生的热量可能引起不可逆的化学副反应或者熔融的金属锂，陈和Evans利用能量守恒进行了锂离子电池充放电过程的热分析实验以及热散失实验，并且简单描述了体系的电化学行为。他们对热传递以及局部高热量来源的分析，表明局部高热引起电池温度很快升高到热失控温度，由于放热副反应发生，高于这个温度后，电池温度上升更快。Pals 和 Newman提出一个模型预测聚合物电解质电池和电池组的热行为。这个模型联合了一三明治模型电池的热平衡的电化学行为。这两个模型强调了热传递和热控制在聚合物锂离子电池体系中的重要性。 Verbrugge和Koch 提出了一个与圆柱型碳纤维电极的 锂离子嵌入过程的数学模型。他们模拟锂离子嵌入单纤维碳电极的过程，包括锂离子在两相的迁移和界面点和传导的动力学。这个模型最初的目的是为了预测电势与嵌入锂的函数关系。Narayanan采用线性扩散模型对Li/TiS的过充保护，电池的氧化还原添加剂进行了理论分析。 Darling和 Newman建立了具多孔电极以及有两个粒度特征分布的电池模型。他们电极的颗粒粒度只有一个分布的电池显示了较差的倍率性能，有大小分布的开路电压的弛豫时间更快。Nagarajan 以包装理论为基础，建立模型说明了颗粒分布对嵌入电极的放电的影响。他们观察到在脉冲放电的过程，二元混合物电极的放点容量比一个单一粒子组成的电极的容量更高。小颗粒的电流调转方向，但是在单一粒度的电极中并没有观察到。最近， Darling和Newman首次模拟电池的副反应，通过引入一种溶剂的氧化还原副反应到电池模型中。尽管氧化还原简单处理，他们的模型仍然能够得出自放电过程的结论，以及它们对正极充电的影响。 许多复杂程度不同的模型已经被被提出来描述可充电锂离子电池。大多数情况下，这些模型考虑这个体系的理想行为，忽略了在充放电循环过程中导致容量和倍率性能损失的现象。这些现象的基础模型比较少见，因为这些过程更加不容易理解。同样，锂离子电池的衰败模型一般不适用于较为广泛的体系。然而，在安全和电池能效方面的现象的重要性需要与未来的电池模型结合。 容量衰减现象 锂离子电池发生的副反应和衰败过程可能引起一系列会导致容量损失的不希望发生的结果。Johnson和 White已经证明商业锂离子电池容量在开始的450次循环衰减10-40%，导致容量衰减的一些过程的流程图表如图2所示；图3 中，显示了半电池放电曲线衰减过程。这个图清晰的演示了每个过程，每个预期的电池运行的过程。在讨论了容量平衡的主题之后，我们会讨论了每个过程的细节。 锂离子电池中容量的平衡 锂离子电池的运行通过锂离子在两个嵌入式电极之间循环，这两个电极嵌入锂的能量不同。为了优化性能，两种电极的锂离子容量比例应该保持平衡。容量平衡是指在循环稳定的条件下，正负极处的物质量必须获得最高的容量（或者能量）。这个题目的实际意义非常重要，如同不平衡电池的安全性一样重要。并且这些主题已经被很多作者在文献中讨论。 锂离子电池的容量平衡条件可以用方程式表达，与两个电极中的活性物质比例γ 有关。用正极与负极的比例写出，这个表达式是 这个方程式说明理想的比例依两个电极相对的库伦容量和可循环的锂离子的数量而定，可循环的锂离子总数与嵌入电极的可逆的循环锂离子化学计量数有关，Δx符号是指负极化学计量数的变化，Δy是指正极化学计量数的变化。对于一些可嵌入材料，它有几个平台，这些平台超出了锂可逆的平台，考虑到安全和可逆，这样你可以选择在可逆或者安全的化学计量比的平台内。在这种情况下，化学计量数的变化将会减小它的最值。 举个例子，考虑这种情况，锂离子电池是碳负极和尖晶石锰酸锂正极材料，通过选择，我们可以计算两个电极的化学计量比范围，正极尖晶石锰酸锂0.83，碳负极0.61.这些化学计量比对应下面的两个电化学反应过程： 足够循环所需要的活性物质的比是1.85.这是通过理论计算的两种正负极材料的容量(C = 148 mAh/g and C = 372 mAh/g)，等于法拉第F除以分子量。 以上情况所描述的理想型的锂离子电池，这个理想模型在整个寿命中平衡容量始终没有改变。对这种理想模型，最初循环过程可以利用的锂容量是恒定。不幸的是，真实情况是，实际的电池体系比这个更加复杂，副反应和二级反应过程打乱了理想模型的容量平衡。对于碳负极/LiMn2O4 电池体系，实际最佳化的活性物质的比是2.05-2.15，也就是正极容量过量了14%，这些过量锂容量的量刚好是形成电极表面的一层稳定的膜的需要量。碳负极首次钝化过程是影响容量平衡的主要过程。周所周知，嵌锂碳负极电极在开始的几次循环会产生一部分不可逆容量。虽然消耗了一些循环锂，但是这些损失的不可逆容量导致了固体电解质钝化膜的形成。损失锂形成的钝化膜对电池的容量平衡有很重要的意义，因为它可以根据使用的负极调节钝化膜用锂量。 电池中任何形式的副反应导致的循环锂的减少，容量平衡都将会发生不可逆的改变，嵌入电极的锂量也会在循环过程中改变。我们需要考虑到碳负极钝化过程发生在所有使用碳负极的锂电池。假定电池开始处于放电状态，碳负极有自由锂，正极过渡金属氧化物在最佳量。任何一个电极的锂量可以用图4表示，这个情况下忽视了碳负极/LiMn2O4电池体系理想电池体系和实际电池体系的最初循环的区别。 在理想的模型中（图4a)，首次充电，所有正极的锂都应该进入负极的夹层中。同样地，在首次放电中，所有负极的锂应该嵌入到正极材料。在实际锂离子电池中，在首次充电的过程中，一部分锂从正极 LiMn2O4脱出，参与了不可逆薄膜的形成，剩余的进入了负极碳。不可逆反应的容量如图4b，负极下面的小盒子。当电池充电到一定终止电压，正极脱锂，尽可能使负极充满锂。理想的负极容纳锂量是安全下的。同时，我们可以想象，钝化层在首次充电过程中完全生成，已经消耗了一部分不可逆锂量。 当电池首次放电，可以放出的锂的总数等于嵌入碳负极的可逆锂量。因此，最初损失的不可逆锂无法恢复。放电一直持续到可逆锂量全部从负极脱出。这时候，脱出锂的化学计量数没有达到最初的数，因为首次充电出现容量损失。这种情况如图4反应。如果电池循环过程中没有副反应，它仍然不会全部利用正极嵌入的锂量。因此，在Δx = 0.61 (x 0 - 0.61) 和Δy = 0.83 (y 0.17 - 1.0) 条件下，碳负极/LiMn2O4电池体系是安全的。我们应该明确，Δx 和 Δy值是由电池和材料决定的。 在这个例子中，负极的嵌入的锂的化学计量数的变化依据正极的活性物质与负极活性物质比的参数γ。理想的γ用于实例中的电池，最初因为不可逆钝化膜造成的锂损失会阻碍碳负极全部利用正极的锂，为了不让这发生，一般的过程是组装电池的正极材料超过理论容量，这样，最初过量的锂允许循环锂的损失。Tarascon 和Peramunage提出一种提供过量锂的方法是采用富锂的正极材料（富锂的尖晶石金属氧化物(Li+Mn2O4) ）。 尽管有副反应和不可逆的容量，设计的物质的量比例仍然能够通过类似于上面的公式计算出来，但是由于钝化反应，我们要把负极添加剂的贡献考虑在内。关于这一容量计做C1，容量平衡可表示为： 举个例子，在使用碳负极和锂锰氧化物尖晶石正极的锂离子电池情况下，实际所需要的正极的量比理论大14%多。这些过量的容量主要用来形成电极表面的钝化膜。活性物质碳负极的比例是2.4-2.45。较小物质量比会阻止负极的利用率，然而较大的质量比会阻止安全隐患，因为负极可以过充（过多的锂可以插入负极）。总之，电池性能包括能量密度只有在最佳的物质量比下才最大化。 显而易见，电池预料之内的过充过放和电池的容量平衡之间有关系。例如，以上讨论的尖晶石锂锰氧化物，在循环过程中，过充反应里的溶剂氧化，取决于电池本身的情况。有较高物质比的电池，这可能不会发生，因为在负极完全被充满电之前，正极也是完全充满电的。高质量比的电池过放使负极的锂完全被充出，负极电位更负，到不理想的负极电位。其他情况下，质量比可能低于设计，导致正极过充。例如，在碳/LiMn2O4 体系，质量比高于2.1将会导致充电过程中负极有过量的锂。质量比低于2.1将会导致锂不够，这会引起负极的过放，伴随负极安全和影响电池的性能。 碳负极的钝化过程是最常见和研究最广泛的影响电池容量平衡的锂离子电池副反应。然而，也有几个其他过程同样具有汇总效果。任何副反应，产生或者消耗锂离子或者电子，都会导致电池容量平衡的改变。另外，一旦容量平衡从理想状态改变，这些改变一般都是不可逆的，可能在循环过程积累，会在电池中产生危害。尽管这些无法用实验量化，如果相关的现象出现在这些模型中，是可以通过实验模型和计算机模拟动态条件，从而直截了当的跟踪这些影响。 化成循环 锂离子电池在最初几次循环过程中出现容量急剧衰减。这个过程就是众所周知的锂离子电池前期必须先发生的化成过程 。通常，化成以后的容量循环一般是比较满意，因为相比较整个电池的容量，衰减容量比较小，基本上放电效率是100%。容量急剧衰减主要是由于负极表面固体钝化膜的引起。碳负极在化成的过程形成钝化膜，随后的容量损失主要是跟碳负极的特定性质有关，包括结晶度、比表面积、预处理以及其他合成工艺细节。最初几次循环之后，电池开始稳定并且展现出稳定的容量。化成循环过程是锂离子电池体系形成非常关键的过程，石墨材料，例如MCMB，不可逆容量是8-15%，然而，碳黑高达50%的不可逆容量。 芳等人证实了不可逆反应发生在碳负极处，主要是在首次放电时碳酸基电解液里发生，然后是可逆锂的嵌入和脱出。这些不可逆反应与电解液分解以及负极表面形成钝化膜或者固体电解质界面有很大关系。当所有的可接触的面积都有分解产物覆盖时，进一步的反应将会停止。在接下来的循环中，这些电池表现出了较好的可逆性并且循环稳定几乎没有容量衰减。这些学者们首次展示了可逆锂离子插入负极是有可能的，只要是负极表面有钝化剂。Gozdz发现在电池首次充电负极钝化膜形成过程中有气体释放出，气体释放与首次化成不可逆容量的损失有关系。更多关于在不同溶剂中负极钝化膜形成以及钝化过程的机理将会在后续电解液还原以及钝化膜形成的部分重新提到。 由于经济方面的影响，化成在电池制作过程非常关键，因为制造商有义务花钱对电池进行化成循环直到电池容量稳定，消耗时间物力。其次，化成过程不可逆容量损失，直接将会在电池体系能量里减去。最后，化成过程产生了气体，需要被释放，然后电池才能运作。世界范围内的研究工作都在继续寻高容量的碳负极材料，有更高的不可逆容量。为了高效利用这些材料，需要预先钝化和锂化，并且不消耗正极材料循环需要的锂量。尽管一些研究组已经研究了这些过程和潜在的代替方法，但是仍然没有到经济上可行的能够消除化成过程的方法。 过充现象 很多电池在过充条件下都观察到容量损失的现象。由于过充引起的商用锂离子电池的耐过充性和安全性，已经导致了严格控制过充和过放。未来锂离子电池在不同领域的运用，将会促进对锂离子电池的理解以及对电池过充的控制。尤其是锂电池在多单元双栈方面的应用更需求电池的耐过充性，这导致了所有规格电池在多双栈利用有相当大的困难。 过充损失可以归为三种类型，(i) 碳负极和石墨负极电极的过充；(ii)高电压下正极的过充反应； (iii)高电压条件下电解质的氧化过充反应。这些副反应导致了活性物质的损失和电解质的消耗，这两种都会导致电池的容量衰减。 碳和石墨负极电极的过充 在锂离子电池过充过程，最初发生的副反应是金属锂沉积在负极表面。在化成过程中，循环锂过量，不管是高于理想的物质量比还是低于最初锂损失量，这个反应是人们所期望的。最开始的沉积锂覆盖在负极活性物质表面的区域导致了循环锂量的损失和电解质的消耗，因为金属锂的活性较高。在正确的质量比的条件下，这种现象在高倍率电池更容易发生，因此高倍率电极更容易极化。普通情况下的锂沉积导致的不平衡需要更多的正极活性物质。参与充电的副反应主要是： Li++e-= Li(s) 负极的脱嵌反应可以写成： 金属锂在负极沉积，与溶剂和盐类快速反应生成 Li2CO3，LiF或者其他产物。人们期望金属锂在电极附近形成边界物，使负极电位更负。形成的产物可能阻塞孔隙，导致倍率性能和容量降低。金属锂的形成也是一种安全隐患，因为这是与溶剂的极端反应。金属锂沉积在石墨电极比在碳黑负极更让人担忧，因为会形成较低的开路电压。为了提供锂沉积的缓冲区，石墨的电池的质量比较低，负极372 mAh/g 的克容量是到不到效果的。 高压正极过充反应 锂离子电池正极电极的过充可能导致一系列电化学反应，但是取决于系统化学。与负极电极一样，预期的正极过充很大程度上取决于体系的容量平衡。质量比低的电池，正极电极更有可能发生过充和过放。过充产生惰性物质，导致容量损失（例如四氧化三钴）或者电解质溶剂氧化，因为正极的电位更高。电极电化学活性物质的分解导致容量不平衡。正极材料热滥用导致正极材料释放出氧气。氧气产生导致电池内压增大，这预示存在一个潜在的危险。 Dahn就正极材料在滥用条件下过充的主要三个方面提出了以下三个反应过程。以钴酸锂为例，可以描述成以下： 他们观察到，γ-MnO2比LiyNiO2和LiyCoO2更耐电和热的滥用。在y＜1时，氧气从材料中脱出，随着电化学计量的减少而增加。升温速率是 1℃/mm，从200℃开始， Li0.3NiO2氧气开始脱出；240°C 时 Li0.4CoO2氧气开始脱出；385°Cγ-MnO2氧气开始脱出。加热速度越快，氧气的脱出温度越高，反之亦然。电池内部氧气产生，在缺乏任何重组机制的条件下，对电池都是一种安全隐患，因为电池内部可燃气体在积累。同样，金属化合物的产物例如 Co3O4、LiNi2O4 和Mn2O3在电池脱嵌锂过程是无效果的，因此容量发生不可逆的损失。 Staniewicz通过研究所有以LiNiO2电池，提出了关于LiNiO2电极的过充机理，他们把LiNiO2电池的循环过程分为三个阶段。 第一阶段：一部分锂离子用于碳负极钝化，不可重新嵌入到LiNiO2 LiNiO2—0.15Li++ Li0.85NiO2 + 0.15e- 第二阶段：可逆循环 Li0.85NiO2—0.5Li+ + Li0.35NiO2 + 0.5e- 第三阶段：过充 Li0.35NiO2 —0.35Li+ + NiO2 + O.35e- 第一部分说明锂离子参与了负极钝化膜的形成，第二部分说明了可逆容量的循环，第三部分说明了锂离子过充。Staniewic提出的过充反应与Dahn提出的在滥用条件下的过充机理不一致。另外NiO2并不稳定，因为 Ni(IV)是氧化态。但是没有实验数据支持作者所提出的以上假设。 低锂量的 Liy NiO2 (y ＜0.2) 的形成是电池循环过程衰退的主要原因。另外，该材料成为电解质氧化的催化剂，并且一些二价镍离子迁移到锂离子的位置。首次不可逆的容量与以 NiO2形式存在的Ni量有关系，这些镍需要额外充电到更高的价态。可以通过 Al 和 B代替Ni提高低锂量的 LiNiO2 的稳定性。 最近，LiNiO2电极的热稳定性被详细研究，通过其他元素代替Ni 和 Li (Co, Mn, Mg, Ca, Sr, Ba)。Co代替Ni提高了常温下的循环性能，但是高温循环性能仍然没有改善。用碱金属(Mg, Ca, Sr, Ba)和Al代替 LiNiO2的 Ni提高了高温和高倍率性能。在高温条件下，充放电过程的衰减主要是由活性物质材料的化学反应的影响。晶体结构存在其他取代时，深度充放电的反应程度会降低，在高温条件下，材料更加稳定。LiCoO2材料高温性能主要是由引进其他元素参杂。LiNiO2 中的Ni由 Ca, Nb 或者 In 代替时，充电过程结构变化微小，这保证了参杂后材料的循环稳定性。 Thackeray研究了尖晶石锂锰氧化物的热稳定性。他们提出，尖晶石锂锰氧化物在加热到780 到915℃范围时，Li2MnO3形成，并且释放出氧气。在温度范围在915℃到1200℃，氧气释放的速率会增加。O2和Li2O 都会析出。以上反应不是电化学反应，而是活性物质被加热到一个特定温度时才发生的。包括在现象电池模型里由温度诱发的电极分解反应都是没有必要的。因为电池故障在较低的温度下就会发生。 Gao和Dahn提出了尖晶石容量衰减和3.3V电压平台随着循环而升高的之间的联系。每次电池充电到更高电压后，3.3V放电平台都会增加，预示着尖晶石LiMn2O4过充后 开始析出氧气，反应机理如下： El —(Oxid El) + +e- 和 LiMn2O4 + 2δe —LiMn2O4-δ+ δO2- El是电解质溶剂，(Oxid El) +带一个正电荷的电解质分子基团（自由基阳离子）。自由基阳离子(Oxid El) +被认为是不稳定的物质，一旦形成就会参与后续的反应。一个伴随着两个质子的二聚反应将可能发生。如果这个阳离子能够稳定到达负极，它将会还原成原始溶剂或者其他产物。高度游离的盐例如PF6可以帮助稳定这些阳离子。 这些学者认为电解质可以作为电子的供体，导致部分物质脱离尖晶石，包括结构里的氧脱出。很有可能这是第二个阶段（类似于加热后生成什么物质）。岩盐结构在LiMn2O4表面形成，当循环过程中它失去氧的时候。氧从结构脱出是不可逆，不仅是因为它产生结构破坏引起循环性能能降低，而且因为它氧化了电解质，降低了电池的循环寿命。 过充/高电压电解液氧化过程 锂离子电池的电解液是由有机溶剂和一种或者多种锂盐混合物。目前比较常见的电解质是线性或者环状的碳酸酯混合物，比如碳酸丙烯酯（PC），碳酸乙烯酯（EC），碳酸二甲酯（DMC），二乙基碳酸酯（DEC），和碳酸甲乙酯（EMC）和盐如LiPF6，LiBF4，LiAsF，与LiClO4。据报道索尼用的是PC, DMC, EMC 与LiPF6锂盐的混合物。三洋用的是EC和DMC 、DEC、EC, DMC,和DEC、EMC分别于LiPF6的混合物。 锂离子电池的高压正极电极对电解液的纯度和稳定性提出较高的要求，锂离子电池的电解质选择是一个限制因素，因为电压上限是由电解质的分解电势决定的。现今使用的一般的电解液在＞4.5V时分解会产生不溶性产物，堵住电极的空隙并且产生气体。这些影响导致在循环过程容量损失，并且是极端的安全隐患。一种比较特殊的溶质 溶剂，EC/DMC，在很多体系单独使用或者与其他溶剂混合使用，被认为在碳负极具有比较高的抗氧化性。Campbell报道称纯的PC的氧化电位低于混有锂盐的氧化电位。这也预示锂盐增强了非水质电解液的电化学氧化。 分解电势通过试验进行评估，在惰性金属表面上或在实际插入电极材料进行循环伏安法测试，设置一个任意的标准电流密度，该电流密度可以使溶剂击穿。对于不可逆的电化学反应，没有热力学开路电压存在。因此分解电势没有实际意义。相反，这些副反应可以用塔菲尔方程描述，在所有电压下，都有限定的分解速率，随着电压增长，分解速率呈指数增长。 表I列了一些电解液的分解电势；但是，人们并不清楚溶剂、锂盐或者二者混合时是否涉及到了氧化反应。另外，部分报告中提到的氧化电位含糊不清。只有当电流密度、电压扫描速率、使用才材料给定，分解电势测定才算是明确测定的数值.我们获取了 Kanamura 的循环伏安数据，如表1所示，扫描速率是50 mV/s ，电路密度 0.1 mA/cm2 。Tarascon 的文章里的数据被人们采用，因为实际的循环伏安图没有被一些作者提供。Christie 和 Vincen通过循环伏安法测量了氧化电位，扫描速率200 mV/s ，电流密度是1 mA/cm2。Ossolo用现行循环伏安法测不同种类的电解液的氧化电位，电极是Li1+xV3O8。他们假定氧化电位的电流密度是0.5 mA/cm2。 溶剂的氧化过程如下所示： 溶剂—氧化产物（气体、溶液、固体物质）+ne。任何溶剂（PC 或者EC）被氧化后将会丢失，最后总会导致盐浓度升高电解质质量下降。同样，溶剂氧化产物例如气体或者其他物质会在电池中积累造成很多严重的问题。溶剂氧化速率与活性物质面积、电流和负极材料有关系。事实上，负极选择和比表面积是很关键的因素，因为溶剂氧化更容易在碳负极发生而不是正极，因为碳负极的比表面积更大。 每次充电都消耗一部分电解液，所以电池组装时需要更多的电解液，这意外着额定体积电池装更少的活性物质，从而初始容量也减少了。同时，固体产物很可能形成钝化膜，增加电池的极化，导致电池输出电压降低。 Novak发现，在Pt 为参比电极的Li/Li电极里，PC氧化发生在电势低于2.1V。在高于3.5V时，氧化速率增加。根据正极材料，Pc的氧化在低于2V时就可以发生。然而，在实践中，高于4.5V时，PC会达到一个更大程度的稳定。Cattaneo 和Ruch分析，MnO2 电极在加热处理后，对其中的物质在线进行光谱测试， LiClO4/PC and LiAsF6/PC分解释放了挥发性物质。大部分电极的氧化反应发生在4V以上。CO2释放在较低的电压就可以观察到，再电极充电过程。反向扫描阴极没有观察到二氧化碳。 氯化物主要是由于ClO-在4.5V以上分解形成。一般会生成CO2和 HCL，生成机理一般认为通过一下反应： ClO4-→ e- + ClO4 →ClO2+2Oad + e- ClO2 + H + +e -→ HC1 + O2(g) 2Oad — O2(g) Eggert 和Heitbaum 通过光谱测定法，也发现在铂金电极，当电势高于4.6V时高氯酸盐离子发生氧化反应。PC氧化后在ClO2存在下不稳定会产生HCl。氧化反应可以在LiClO4电解液分解观察到。 Christie和 Vincent报道称1M的LiPF6 在 PC 的镍电极的氧化电位。Kanamura研究了PC在 Pt, Al, Au, 和Ni电极的开路电压。这些材料的氧化电位从镍的4.5V到Cu的6V，在Li/Li电极。Kanamura提出，镍电极在各种碳酸丙烯酯电解质的氧化行为强烈依赖所用的电解液盐。分解产物主要由高压电极变化范围内的阴离子决定。锂离子电池过充电解液氧化已经被Tarascon采用循环伏安法验证。但是所有的这些研究都没有提供分解过程的机理和研究的分析技术。考虑到大量的锂离子电池研究，并且溶剂氧化对电池性能和安全的重要性，在这方面的知识缺乏令人感到惊讶。 关于电解液分解机理的更详尽的讨论在后面一部分讨论。 过充保护 商业电池成功主要取决于它的安全性，尤其是恶劣条件下。恶劣条件会导致电池温度上升，继而导致电池内部自我产生热量，最终电池热失控。 常用的锂离子电池正极材料电解液会发生不可逆的氧化反应，会影响电池的循环性能。一种常见的有效方式是往电解液里加入添加剂，电解液添加剂在电池超过一定电压时提供电流通道。理想的化学往复是在接近完全充电的电池电压或者在过充电时占用额外的电荷，从而防止有害的反应继续进行。对于典型的锂离子电池，氧化还原反应的潜在电压是4.5-5V（Li/Li+）或者1.5-1V（HVH2）。在这种情况下，正极持续的过充反应是： R → O + ne- O+ ne- → R O在正极产生，扩散到负极，当遇到H时，发生反应，量也减少。两极之间的氧化还原反应让过充过程多余的电流导入继续充电，直至充电截止。 例如，LiI作为一种氧化还原添加剂，可以作为3V锂电池过充的保护。在1M LiPF6/THF电解液里添加，已经证明可以阻止电解液在Pt电极的表面氧化反应。阳极电位（低于完全充电电位），LiI 经过两步反应过程，一个是3.2V碘离子氧化成三碘化物，另外一个是3.65V时三碘离子氧化成碘单质。 LiI—I-+ Li+ 3I—I3- +2e 2I3- —3I2 + 2e LiI 氧化反应生成I-与金属锂反应生成LiI。碘的减少通过两个相似的过程发生，3.55V碘生成碘离子，2.75V碘离子到碘离子。 除了“化学往复”，其他一些过充保护也运用于商业电池： 1.熔点在140℃的分离器可以用作过充保护。其目的是存在一种聚合物膜，短路时电池的温度高达一个既定的值时，聚合物膜会融化阻断电流。通过中断阻止电流通过反应，阻止电池达到一个更高的内部温度。一些以聚烯烃聚乙烯（130℃），聚丙烯（155℃）为基础的聚合物分离器可以作为内部安全阀，当短路的时候关闭隔膜孔。索尼使用聚丙烯基分离器（161.7℃MP），而三洋和松下使用135.4和133.4℃熔点聚乙烯通过分离器，分别关闭Celgard®2300 FSM的温度是131℃。 2.阴极材料添加剂，例如 LiCoO2过充后容易分解，增加电池内部气压。这个气压会冲破电池顶部的减压阀，压力被释放，电池电路也被破坏。索尼电池显示，内部的添加剂不会影响电池运行的电流。Moli 能源使用了2%联苯在石墨/LiCoO2做过充保护。联苯分解产物会沉积在负极表面导致较高的内阻以及低倍率性能。 3.电池内部气压升高，电池防爆阀门变形，将会切断电池内部的连接。当压力增大时，充电电流被切断。 电解质分解（还原）过程 电解质还原会危害锂电池的容量和循环寿命，主要是通过消耗盐类和溶剂。并且产生气体产物增加电池内部电压降低了电池的安全性能。减小电解液的还原反应以及容量损失过程是提高电池循环寿命以及高温性能的一个关键需求。电解液还原是所有使用碳插入式电池的普遍特征，电解液相对于碳负极不稳定。电极最初化成过程形成的钝化膜被认为是早期的化成过程。理想情况下，电解液还原仅在化成过程形成。在后期循环过程不会再发生。 发生在碳负极表面的电解质还原反应类似于金属锂表面发生的反应。因为碳负极和金属锂负极之间的区别很小。因为这个原因，大量关于金属锂负极电解质还原的过程可以用来理解碳负极电解质还原的过程。大量的实验技术X射线冠电子能谱 (XPS)、X射线俄歇电子能谱学（EDAX）、哈曼谱在线质谱原位及非原位傅里叶变换红外（FTIR）光谱、原子力显微镜（AFM），和电子自旋共振技术已被用来确定的电解质还原机理和识别在碳电极表面形成的产品。 Dey 研究了PC电解质在碳负极的电化学分解反应，他们认为反应机理如下： 2Li++2e-+ (PC/EC) —[propylene(g)/ethylene(g)]+ Li2CO3(s) 以上反应发生在首次放电，当电极电位接近0.8V。在电解质是EC/PC混合物时，Fong为首次循环碳负极发生的不可逆容量提出相似了反应机理。 Aurbach和他的同事对电解质的溶剂和盐类还原过程以及它们在金属锂负极和碳负极的产物做了广泛的研究。它们发现了有机物质 ROCO2Li （CH3CH(OCO2Li)CH2OCO2Li）和PC单电子还原生成的丙烯。前期Dey就阐述了碳负极处PC是一个双电子过程，生成了碳酸锂和丙烯。Aurbach认为ROCO2Li 对微量的水非常敏感，会与水快速反应生成 Li2CO3，表明前期的研究工作没有在充分的干燥的环境下进行。 在冠状醚的存在下，碳负极保持了石墨结构，可以经受可逆脱嵌，因为溶剂不共插层，碳并没减少，只是发生在表面。当PC在表面减少时，转移电荷大部分穿过表面的薄膜，一个电子减少将有利于驱动PC的还原。在前期的研究中，当没有冠醚存在的时候，碳负极结构被破坏，并且被一层一层剥落。溶剂减少的时候，并没有到脱嵌的阶段。因此，当PC分子进入到碳结构里，大部分PC的还原发生在碳结构里，在这种情况下，双电子过程形成的 Li2CO3是有利的。 Matsumara观察到首次循环过程的不可逆容量不仅仅是PC分解成 Li2CO3，同时也因为其他的副反应。他得出结论，在首次充电的过程中，PC很有可能使通过两个途径分解的。一个是，PC直接还原生成 Li2CO3和丙烯。另一个是，PC经过还原，生成一个阴离子。然后生成烷基锂碳酸盐，自由基也同时终止。这些物质都不稳定，并且还原生成低聚合不稳定产物，然后与丙烯反应生成低聚物自由基，最终氧化生成含有C-H键的化合物和COOH基团。 舒研究了锂在石墨负极的电化学脱嵌，电解质是1MLiCIO4 PC/EC (1:1)。他们认为主要涉及两个过程，也就是，一个PC的双电子还原和EC生成丙烯和乙烯气体，另一个单电子过程形成烷基碳锂酸盐。双电子还原过程可以细分为电化学反应和化学反应。电化学还原和固体电解质界面形成的初始步骤都涉及到通过单电子还原生成阴离子从而形成碳酸锂化合物，阴离子自由基进一步发生单电子还原产生的气态产物或SEI膜。 楚通过原位电化学原子力显微镜研究热解高序石墨电极表面形成薄膜的极化过程，电解质是MLiClO4 EC/DMC (1:1) 和1ML1PF6 EC/DMC (1:1)。他们发现还原反应不可逆，表明这些反应发生在高电位(1.6/2.0V)相比基底表面在表面（0.8/1.0V)。Peled也表明更多的溶剂还原发生在基底表面，盐的还原发生在边缘的表面。楚表明电极表面薄膜形成的厚度更薄（几百个纳米）（在基地表面有10纳米左右，边缘表面薄50%） 大部分的成膜助剂的优势性能在一些文献中已经讨论。SO2添加剂促进了石墨负极锂离子在非水电解质里的不可逆脱嵌。SO2提供了石墨负极处形成发育完全的钝化膜的有利优势，电势高于电解质的还原电势。这些钝化膜层主要由Li2S 和锂氧硫化合物组成，反应过程如下： SO2 + 6Li ++ 6e -→2Li2O+Li2S Li2O +SO2→(LiO)2SO Li2O+2SO2-(LiO)OSOSO(LiO) 由DEC和DMC混合而成的碳酸盐电解质被发现会发生酯交换反应，这导致了碳酸甲乙酯的自发生成。这种溶剂在最近 的文献中已经报道，对锂离子电池有理想的钝化性能。最近一份关于石墨电极在含有 LiPF6 或者LiAsF6的MPC的研究显示这种溶剂的还原反应最开始在1.5V发生。作者观察到Li2CO3是主要的电极表面生成物质。 一种新的混合溶剂由氯乙烯碳酸酯（CEC）、PC、EC组成，ECE在锂离子在负极形成钝化膜，不会溶于电解质。这种溶剂允许石墨电极使用PC电解质，而不会增加电解质的分解。最近的专利文献包含其他碳酸酯类溶剂的引用，包括其他卤素取代的碳酸酯和多种不饱和碳酸盐，声称具有理想锂离子电池的性能。通常在电池里添加少计量的碳酸酯，在碳负极最初的还原反应里和钝化膜形成的过程被消耗。预计这方面的工作将继续被研究，包括钝化膜性能的了解、与电池性能的关系，在将来都会得到发展。 碳酸盐类电解质还原反应的机理已经被提出了多种（溶剂和盐类）。这些机理（方程28-46）包括溶剂、盐类、污染物的反应。 溶剂还原反应-碳酸丙烯酯（PC)- Dey 双电子还原机理： PC + 2e-→ propylene+ CO2-3 Aurbach给出的PC的单电子还原机理： PC + e - →PC- radical anion （自由阴离子） 2PC自由基负离子 →Propylene +CH3CH(CO3-)CH2(CO3-) CH3CH(CO3-)CH2(CO3-)+2Li→ CH3CH (OCO2Li)CH2OCO2Li(s) （锂烷基碳酸） 乙烯碳酸酯 - EC的双电子还原机理与PC的还原机理类似， EC + 2e-→ethylene + CO2-3 EC的单电子还原机理与PC的还原机理类似， EC + e-→ EC- radical anion 自行葫芦2EC- radical anion →ethylene + CH2(OCO2)- CH2(OCO2)- CH2(OCO2)- CH2(OCO2)- + 2Li+→ CH2(OCO2Li)CH2OCO2Li(s) (锂烷基碳酸) EC的还原产物CH2(OCO2Li)CH2OCO2Li(s)与碳酸锂作为一种有效的钝化层。 Dimethyl carbonate (DMC)—This mechanism can be written as follows or CH3OCO2CH3 + e- + Li→ CH3OCOOLi+ CH3+ 碳酸二甲酯(DMC)-还原机理如下方程： CH3OCO2CH3 + e- + Li→ CH3OCOOLi+ CH3+ 或者 CH3OCO2CH3+e-+Li+→ CH3OLi + CH3OCO+ Aurbach表示DMC的亲核反应生成了CH3OLi和CH3OCO2Li . 自由基(CH and CH3OCO) 转化为CH3CH2OCH3和CH3CH2OCO2CH3。 二乙基碳酸酯（DEC），这种机制可以写成如下 CH3CH2OCO2CH2CH3 + 2e- + 2Li+→ CH3CH2OLi+CH3CH2OCO+ 或者 CH3CH2OCO2CH2CH3 + 2e + 2Li +→ CH3CH2OCO2Li+CH3CH2+ 形成的自由基 (CH3CH+和CH3CH2OCO+)由DEC分解转化成CH3CH2OCH2CH3和CH3CH2OCO3CH2CH3。 Novak 发现丙烯和四种不同的混合溶剂对石墨电极的乙烯释放量不同，但无论是丙烯或乙烯，镍电极上检测都没有。 盐类还原反应-盐的种类和浓度会影响碳插入电极的性能，因为盐的还原作用参与了表面膜的构建。这些特定的情况下，盐的还原作用促进了表面的稳定性和理想的钝化膜的形成。在其他情况下，盐的还原作用产物的沉积会干预溶剂的还原反应。根据Jean 的锂盐LiCF3SO3的还原发生于溶剂 (PC/EC/DMC) 在负极还原之前。 盐类的还原作用如下： LiAsF6 LiAsF9 + 2e -+ 2Li+→ 3LiF + AsF3 AsF3 + 2xe- + 2xLi+→ LixAsF3-x + xLiF LiC1O4 LiC1O4 + 8e -+ 8Li+→ 4Li2O + LiC1 或者 LiCIO4 + 4e- + 4Li+→ 2Li2O + LiC1O2 或者 LiClO4 + 2e- + 2Li+ → Li2O + LiCIO3 LiPF6 LiPF6→ LiF + PF5 PF5 + H2O-→ 2HF + PF3O PF5 + 2xe- + 2xLi+→ xLiF + LixPF5-x PF3O + 2xe- + 2xLi+→ xLiF + LixPF3-xO 并且 PF6- + 2e- + 3Li+→ 3LiF + PF3 LiBF4 (similar to LiPF6) BF4 + xe-+ 2xLi+→ xLiF + LixBF4-x 污染物还原-电解质经常包含一些污染物质，比如说氧气和水。氧气被还原生成锂氧化物 1/2O2+ 2e +2Li+→ Li2O(s) 石墨负极的性能不受溶剂里的少量的水的影响（100-300ppm)。对于高浓度的水。水在石墨电极处被还原生成LiOH，沉积在碳负极表面，作为一种具有高界面电阻的阻断剂。因此，LiOH会阻止锂进一步嵌入层状石墨电极。 H2O+e-→ OH- +1/2H2 Li +OH-→ LiOH(s) LiOH(s) + e- + Li+→Li2O(s) + 1/2H2 激光快速成型机CO2存在，会生成Li2CO3 ，成为负极电极表面的钝化层。 2CO2 + 2e- + 2Li+→ Li2CO3 + CO 或者 CO2 + e- + Li+→ CO2-Li+ CO2Li + CO2→ OCOCO2Li OCOCO2Li + e- + Li+→ CO + Li2CO3 二次反应-碳酸锂也可以由二次反应生成 2ROCO2Li + H2O → 2ROH + CO2 + Li2CO3 R是乙烯或者丙烯基团 LiAsF6和LiPF6还原反应发生在电势低于1.5V (方程36 -40) (vs.Li/Li+) 自放电过程 自放电是指在开路条件下，电池静置时候，电池电压自发的降低。锂离子电池会自放电，虽然低于镍镉和镍氢电池，但是还是比较快，并且对温度比较依赖。自放电不可避免的发生在氧化物锰酸锂、钴酸锂、镍酸锂电极。自放电的程度取决于以下因素，诸如阴极和电池的制备，性质和电解质的纯度，温度，存储时间。 锂离子电池的自放电按照可逆和不可逆来分类。可逆容量损失是指电池重新充电时容量可以恢复，不可逆容量不可以恢复。这是一个有用的实际存在的区别，不可逆容量的损失程度依赖接下来循环的充放电倍率。因此，自放电引起的容量损失最好是在放电倍率数据前提条件下陈述。为了讨论自放电机理，我们试图分开可能导致不可逆容量损失的过程（在任何倍率条件下，容量都不能够恢复）和可逆容量的过程（没有导致永久性容量损失）。 Johnson 和White报道索尼和松下电池的自放电行为。他们监测了电池30天开路条件下的电压，电池的容量保持率是超过最初容量的97%。因此，他们得出结论，电池自放电对循环过程的容量损失影响更大。自放电率在高温55℃下（10%/月）比常温的自放电率高（2-3%/月）。自放电容量损失在这些文献报道中都是可以恢复的。 LiMn2O4/有机电解质电池自放电机理和正极电极的电解质的不可逆氧化以及可逆嵌入尖晶石 LiyMn3O4 结构有关。插入电极过程是可逆的，以及电极脱锂的程度是放电返回的锂。一般而言，充电电池可以自放电，主要是通过电解质沉积耦合反应到最初锂插入反应。正极电极表面的氧化可以写成 EI→ e- + EI+ 此处的El是指锂离子电池溶剂(EC, PC,等等.)释放的电子通过插入锂离子反应进入金属氧化物： yLi+ + ye- + MOz→LiyMOz 块状插层是正极氧化物插层结构，降低了该电极的充电状态。 对于LixMn2O4 LiyMn2O4 + xLi+ + xe- → Liy+xMn2O4 上反应同时发生在正极复合物，并且不需要外部电子。总反应式： LiyMn2O4 + xLi+ xEI→Liy+xMn2O4+ xEI+ 自放电速率是有限的，因为溶剂的氧化，强调了溶剂在电池循环过程稳定性的重要性。Guyomard 表明溶剂氧化主要发生在碳负极表面，建议降低比表面积来降低自放电率。然而降低活性物质的比表面积在LiMn2O4条件下已经被证明很重要，电流集流体与溶剂接触的表面积不能够消除。 根据上述锂离子电池自放电，如果电池保持充电状态，这将会导致容量永久损失。这是这些机理中对锂离子电池容量平衡的扰动的结果。幸运的是，电池自放电速率与锂锰氧化物的放电速率类似。电池电解质盐浓度也会通过这些过程产生不可逆容量。这些现象的任何过程都会导致电池循环过程中容量或者倍率性能的损失。长时间或者重复的自放电，电池会出现两电极容量不平衡，增加充电过程锂嵌入负极的风险。 Pistoia对锂离子电池自放电现象进行路进一步研究。在不同电解质体系，针对三个主要阴极金属氧化物的自放电速率进行了相互对比。电解质氧化又一次涉及到电池自放电机理中，仅仅通过这个过程不能解释所有的实验发现。在不同的电解质，自放电速率变化比较大，预计从电解质氧化机理开始。另外，自放电过程产生的污染物在一些情况下可以看到，包括更高阻抗和倍率性能损失。 两个额外的自放电机理也被提出，一个是负极驱动锂离子重新自发的插入阴极电极，第二个是电极溶解。第一个过程主要由于部分阴极材料不稳定。有趣的是，当金属锂负极被铂电极取代时，这个过程就会停止，导致一些学者认为负极处的锂离子同样参与了自放电反应。因为自放电过程没有电流存在，锂离子从负极到正极必须有同样的电子从正极到负极补偿。这可能导致溶剂在任何一电极氧化（产生阳离子）或者溶剂在锂金属电极还原。第二个机理，在接下来的一部分会详细讲述，可以通过适当的选择电解质控制。 自放电过程，富锂的负极脱锂可以通过以下的氧化还原反应来解释，主要是由耦合电解质分解反应和锂嵌入反应组成。 EI+ ye-→passivating layer LixC6 → ye-→ yLi+→Lix-yC6 这种情况，由于热力学还原电位电解质的还原反应是可能的，但是动力学非常缓慢，主要是负极表面已经存在的钝化膜原因。如上面描述，本文研究的两个电极间的自放电过程速率是相似的，导致了部分永久容量损失。 商业锂离子电池的自放电大部分是可逆的，只有一小部分是不可逆的。文献中提出的自放电机理（耦合电解液分解，讨论以上）在大多数情况下会导致不可逆的损失，主要是因为可逆循环锂的消耗（形成其他碳酸锂盐产物）或者堵塞活性物质表面孔。形成可逆容量的自放电机理还是有必要的。一种可能的过程是通过氧化还原反应将氧化溶剂从正极转移到负极，换句话说，发生了氧化还原往复的过程。这些物质发生可逆的氧化还原，类似于前面提出的往复迁移机理，或者是被破坏后增加到负极表面的钝化层。一旦有同样数量的电子参与氧化还原反应，电池平衡容量就会保持，自放电可以恢复。 锂离子电池自放电的另一个可逆的贡献，可以归因于通过隔板的电池的漏电流。这个漏电流可能是由于制造过程中的一些缺陷增加，如在机针孔。为了需要保护电流通过，有限隔膜没有隔开的漏电流由电化学放点平衡。这个过程的速率很低，受到隔膜的阻碍。预计到这个过程对温度的依赖较低，实际上自放电的实际数据对温度的依赖比较大，证明这个机理不是最主要的一个。 界面膜的形成 负极和电解质界面生成的钝化膜主要是由发生在锂离子电池或者溶剂里的或者电极表面的不可逆界面反应引起。文章前面已经探讨过这个过程的某些方面。这些界面反应会在负极表面形成一个稳定的有保护作用的膜，这个钝化膜会保护电极继续运作而不发生更进一步的反应。最初为了形成钝化膜锂离子损失引起了容量平衡的改变，这可能导致效率降低，从而降低整个电池的能量。负极不可逆容量损失随着不同类型的碳变化范围是10%到100%。容量损失依赖于碳负极的类型、电解质溶剂成分、电解质或者溶剂的添加剂。钝化膜的形成与负极锂的沉积不同，负极锂沉积主要发生在过充电。钝化膜形成在电极表面，产物(碳酸锂等等）由电解质分解形成。 Peled解释了发生在锂和锂化碳负极/电解质表面的基本过程，提出模型解释这些界面现象。碳表面的固体电解质界面膜（SEI）在决定电极和电池行为以及性能方面有很重要的作用，包括循环寿命、电池寿命、安全性、不可逆容量。固体电解质界面膜主要作用是隔开负极和电解质，消除或者减少电极到电解质的电荷转移，或者溶剂分子和盐阴离子由电解质转移到电极。SEI膜的形态很复杂。随着时间和电解质的成分改变。最好把它描述成薄的多相不均匀膜，镶嵌在不同物质单个粒子界面。金属锂的固体电解质界面膜是一个多孔的开放的腐蚀产物，很大程度上，堵塞了阳极的表面，但是不会参与沉积和溶解过程。 碳化负极极化电势低于2V（Li/Li+），许多副反应会同时发生，图5所示。Peled 详细研究了金属锂/聚合物电解质界面。使用的聚合物电解质是LiI-PEO-A12O3混合物，他发现并联的RC元件体现了固体电解质膜明显的电阻、电容、和厚度，这个可以适用界面阻抗，预测电解质膜厚度呈抛物线增长。提出的模型不能概括他们产生的外部条件，因此不能制定详细的设计规程。一个更全面和更普遍的模型需要开发，该模型对任何条件下的锂离子电极都有效。 固体电解质的沉积溶解过程主要涉及三个步骤：(1)电子在金属/固体电解质界面转移; (2) 阳离子从一个迁移到另外一个； (3)离子在固体电解质/溶剂界面迁移。 M — ne-→Mn+M/SE Mn+M/SE→Mn+SE/sol m(solv) + Mn+SE/sol →Mn+SE.(solv) 沉积溶解过程的速率主要是锂离子穿过钝化层到金属锂表面。 电极表面生成的钝化膜可以通过一个简单的非均相反应解释，该反应发生在电解质(El) 和 LixC6，固体电解质模型假设表面形成的过渡层，阻止了进一步的反应，并且允许锂离子通过。上述反应可以分解为两步 LixC6= Lix-δC6+δLi++δe- δLi++δe-+EI liquid→δLiEI (solid) 以上两个反应显示反应过程只有电子通过碳膜转移到SEI /电解质界面进行的，或如果溶剂分子从溶液转移到设置/碳界面。如果他们有足够的移动通过SEI膜体溶剂分子通过膜通缺陷透过薄膜，如裂缝。 Garreau提出聚合物电解质模型描述由多孔不导电薄膜覆盖的金属锂电极的性能。由于有PET覆盖并且通过多孔扩散电解质，导致电荷转移反应是有限的。电解质的成分是形成的钝化膜的性能的决定因素，碳酸类溶剂与碳、锂、碳化负极反应生成烯烃气体和碳酸锂作为最初成膜物质，这是前面电解质分解部分已经讨论过的内容。 集流体 铜和铝是目前最常用的用于负极和正电极的电流集流体。另外，镍和不锈钢也可以用来做锂离子电池的集流体。与集流体相关的问题主要是钝化膜的形成、吸附力和局部腐蚀，如点蚀，和一般腐蚀。这些现象增加了电池的界面电阻，导致容量和倍率的损失。锂离子电池中的腐蚀过程的研究还比较少，但是，在这方面还需要得到更多的关注，更多的工作和更多地认知，这都是人们面临的挑战。 两种电流集流体很容易退化，铝容易腐蚀，铜容易裂开。Braithwaite和他的同事通过阻抗图谱法、XPS、俄歇技术研究铝在 PC/DEC 和EC/DMC 的电解质里点腐蚀。铝在循环过程容易钝化，在铝的表面发现了锂和磷。作者表明铬酸盐包覆处理可以保护铝集流体。 如果发生硬化， 铜在锂电位附近发生环境应力龟裂。铜集流体在这些环境中相对比较薄，但在循环过程没有变厚。 最近Pistoia报道LiPF6-EC/DMC电解质腐蚀铝网的电位是3.1 到3.2 V ，铝带的电位是4.2 V，主要取决于 HF的含量。在LiBF4-EC/PC和 LiC1O4-EC/DMC电解质里，铝带腐蚀低于4.9V。石墨化几种电解质可以忍受的电位高于4.5V。陈检测铝集流体在锂/聚合物LiN(CF2SO3)2/V6O3 或者TiS充电后在不同电位下通过SEM扫描。他们观察到点腐蚀会影响聚合物电池长期的粘接性。另外一种耐腐蚀合金W-A可以在集流体表面形成保护膜。 商业锂离子电池的两种集流体都是通过预处理（酸-碱腐蚀，耐腐蚀涂层，导电涂层等）来提升他们的粘附力以及耐腐蚀速率。这些预处理明显的提高了铜和铝集流体性能。对于铝而言，没有任何表面处会导致电池的界面电阻增加，集流体的粘附力损失，会影响电池因为电子可能会不连续。对铜而言，不连续区域影响后面电极电位，会导致锂沉积。 集流体腐蚀会导致电池的内阻增大，在多次循环以后，腐蚀产物会形成一个绝缘膜在集流体表面。界面电阻的增加依赖于集流体堵塞预处理，界面电阻会导致电池在后期损失一些容量。在一定倍率下，倍率性能损失间接导致容量损失。 铜集流体溶解可能在负极过充时应发生， Cu - Cu + e 在非水环境下一价铜比二价更稳定。在热力学标准条件下，该反应在水溶液的平衡电位是0.521V，在Li/Li+的平衡电位3.566V。碳化负极接近放电截止电压或者过充条件下可以达到1.5V。显然，这个反应更容易发生，因为在非水环境，水的标准热力学数据不存在。 过放过程形成的铜离子在循环后期会形成金属铜沉积在负极表面，形成的枝晶会穿透隔膜，引起电池衰退。为了阻止这些过程，电池不能放电到低于2.5V。这对于正负极相叠电池是很严重的问题，因为很难控制每一个电池的电压。因为，电池有可能过放，因为容量太低。这回破坏整个电池。对于其他一些消费电子产品，这将有利于电池有更好的耐过放性。 阳极溶解 正极溶解现象是电极和电解质共同的性质，大部分材料的该过程的数据已经获取。决定材料溶解性能的因素主要是材料结构缺陷、充电电压高、正极材料里的碳含量。锰酸锂和镍酸锂的氧缺陷减弱了材料过渡金属和氧的键合力，导致锰和镍溶解。过渡金属离子和氧之间的键合力比较弱，会被电解质催化驱动，尤其是极化到高电位时。电解质在负极的氧化产生催化活性促进金属离子溶解。 作为三种主要的正极材料，对尖晶石锰酸锂阳离子溶解进行了研究。锰酸锂活性物质的溶解通过歧化反应机理，导致锰最后沉积在负极。这会导致正极活性物质减少，并且阻塞负极表面的空隙。Mn2+ 溶解确定发生在锰酸锂过放过程，电池放电过程是一个很关键的问题。这些过程已经被许多学者研究，包括Thackeray和他同事的研究机构。 首次将锰酸锂容量衰减归因于Mn 溶解的是Thackeray。Tarascon 和同事通过卢瑟福背散射（RBS）发现负极存在锰。文报道容量在高电压区间会衰减主要是因为活性物质通过溶解渐渐的转变为一个低电压的缺陷尖晶石相。最近夏报道容量损失，由于Mn3+ 溶解引起的仅仅占23和24%（常温和50℃）。他提出，其余的容量损失主要是由结构变化以及电解质溶剂分解引起。 根据 Bellcore和Dahn的研究，25%的Mn2+解离沉积在负极表面。这个过程通过以下机理发生： 4H++ 2LiMn3+Mn4+O4-→3λ—MnO2+Mn2+ 2Li+ +2H2O 这是尖晶石酸分解歧化过程。这个歧化反应发生是因为Mn3+存在，与 Mn2+ 和 Mn4+ 反应。Mn2+进入溶剂，在负极以Mn(s) 沉积。 Mn2++ 2e-→ Mn(s) 也有一种情况，胶状的锰酸锂粒子通过电泳迁移到负极，在负极形成锰沉积。然而，如果电泳发生，负极处的锰量强烈依赖与锰酸锂粒子的颗粒大小，这基本不会发生。 上述反应随着温度升高而加快。形成氟化氢的质子来源于电解质的 LiPF6盐。因此，很大程度上依赖LiPF6 的纯度。 H2O + LiPF6→POF3+ 2HF + LiF 上述方程产生的水生成更多地质子，会催化锰的溶解。自由质子也可以通过以下机理在负极处被消耗 2H+ + 2e-→H2(g) LiPF6的纯度高或者 LiMn2O4的比表面＜1 m2/g 会降低或者减少Mn2+溶解反应。限制比表面积可以减少副反应的催化作用，例如溶剂氧化。已经有实验证据证明六氟磷酸盐磷酸阴离子直接参与锰溶解过程，或者通过粒子辅助机理。 姜表示，锰溶解四锰酸锂电池衰减的最主要原因。这篇研究中，锰溶解带来的是缺锰尖晶石/碳界面接触电阻的增大以及锂离子嵌入脱出的电极反应电阻。他们已经报道了关于锰溶解对电压依赖的实验数据。溶解速率在电压低于4V时增加的并不明显，但是你高于4V时，增加比较明显。根据这些作者，歧化反应机理似乎不是引起锰溶解的主要原因，因为在充电过程结束时，Mn3的含量是最低的。 在两外一篇研究中，Jang和Oh报道称尖晶石溶解是由电解质溶剂分子电化学氧化产生的酸诱发。相对于碳酸盐电解质，尖晶石在乙醚存在的电解质中，溶解更大，例如 DME 和THF。在最初阶段，只有 Li 和 Mn参与反应，在后期，氧损失占主导地位。他们发现，溶剂衍生酸 在电解质里并不显著，包括(LiPF6, LiBF4和LiAsF6)，然而，尖晶石在这些电解质的溶解是相当大，因为产生的酸包含质子和水。 尽管前期的工作貌似出现了矛盾，好像是各种机理在所有情况下都运行。姜强掉了电子产生与锰溶解之间的重要联系。通过使用乙醚溶剂，更多地质子产生了，导致更多地锰溶解，相比之前的碳酸盐电解质。锰溶解于电极电压之间的关系潜在依赖溶剂氧化过程，不是姜之前所说的充电电压。 Robertson 提出一个代替尖晶石溶解的机理。他们提出一个改进了的质子催化氧化还原机理，锰从电极析出同时生成在4V时没有的活性的LiyMnOx 物质 l2LiMn2O4→Li2MnO3 + 5Li2Mn4O9 + 3Mn2+ solve 12Mn3+→9Mn4++ 3Mn2+ solve+ 6e- oxidation 3Mn2+ solve+ 6e- →3Mn ↓ reduction Li2MnO3是电化学惰性物质，Li2Mn4O9 在4V几乎没有容量。从方程63到65，12 M 的 LiMn2O4将失去活性，因为3 M 的Mn2+ 溶解进入电解质。另外Li2Mn4O9 在2-3V循环。他们提出在尖晶石参杂少量的Cr3-大幅度减少锰溶解进入电解质。 基于锰溶解以及循环过程尖晶石结构的变化的实验证据，夏提出尖晶石溶解通过两条途径：(1)LiMn2O4 转化成 LiMn2+xO4-x, 损失一个 MnO， (ii) LiMn2O4 Li1+xMn2-xO4, 损失一个 Mn3O4 。在这个情况下，一部分Mn3+ 转化车工Mn4+伴随 MnO溶解进入溶剂；也就是 Mn3+(LiMn2O4) → Mn4+(Li2Mn3O7) + Mn2+(MnO) Amatucci报道了LiCoO2电池充电到4.2V时，容量衰减与负极处探测到的钴含量之间的紧密和直接的关系。容量损失依赖于活性物质合成过程中的热处理。钴离子在溶解后会出现在电解质里，负极处的钴含量就会减少。溶解速率随着截止电压增加，当电压增加到4.5V时，速率迅速增加。分离溶解的钴含量和充电截止电压大于4.2V后的容量损失量之间的少量关系已经由这些作者给出。 某些钒氧化物(LiyV2O5)电池基于LiAsF6-EC/PC/2Me-THF (15:70:15) 电解质，钒氧化物也被发现有溶解现象，会覆盖在负极表面，会增加循环寿命。钒覆盖在锂表面是有利于负极表面钝化膜的性能。这项工作证明电极材料的溶解过程对电池的性能有影响。 插入电极的相变 我们对插入电极的相变与容量损失之间联系的理解是比较薄弱的，尽管这是广泛引用的衰减机理。基本的机理是相变或者结构变化会导致颗粒断裂以及接触电极损失。一般而言，我们可以理解，好的电极显示； 高可逆容量和循环寿命，不可能伴随着明显的结构变化或者劲歌膨胀或者收缩。目前对锂离子电池的三种主要的金属氧化物的插入化合物都在这一类。然而，对对高容量材料的研究使得相变和结构变化都难以避免。这些过程对电池的影响被量化研究，并且是基本的研究方式。 发生在锂离子电池的相变可以分为两个类型：发生在正常脱嵌过程以及在过充条件下发生的相变（例如歧化反应LiyMn2O4 ），夏和 Ohzuku研究了尖晶石锰酸锂的相变。夏认为稳定到不稳定两相变化是通过损失一个MnO (Mn3+→Mn4+→MnO),容量损失主要发生在常温电池循环过程。 Amatucci 描述了LiMn2O4 单相插入锂和两相插入锂，两相插入锂是： 在温度变化为100- 250℃，LiyMn2O4脱锂生成了λ-Mn2O4，然后分解成β-MnO2。在一些条件下，β-MnO2 可以检测到，作为分解过程的一种中间物质。金红石 β-MnO2t 在4 V时没有电化学活性，导致嵌锂形成正交LiMnO2。循环过程中的这些变化导致容量损失。最近Cairns 采用X-ray 射线吸收原理技术发现LiMn2O4脱锂，结果收缩形成类似结构λ-MnO2,，但是锂化的LiMn2O4 转变为四方尖晶石结构Li2Mn2O4。 尖晶石锰酸锂容量衰减的另外一个原因是过充导致3.5V锰氧化态出现。这可能导致锰酸锂的结构变化，产生Jahn-Teller畸变。在Jahn-Teller畸变溶解过程，晶体结构在Z轴延伸15%和X轴和Y轴的同时延伸6%。每个晶胞各向异性膨胀16%直到表面部分转变为四方相Li2Mn2O4，立方相的LiMn2O4的单个晶粒仍然在膨胀。颗粒破碎导致比表面积增大， Mn3+溶解反应变的比较严重，导致颗粒之间的接触被破坏。尖晶石的结构破坏导致循环容量降低。 Jahn-Teller畸变的过程可以通过使用过量的锂大大降低（x = 1.05或更高）。锂代替锰的晶格位置会增加锰的平均价态。这个过程也可以通过掺杂其他金属离子完成，例如阳离子Fe 或者 Co，阴离子 F。尖晶石锂锰氧化物电极在一些情况下比较稳定（可以阻止过充），显示了稳定的循环容量，室温条件下循环几千次容量稳定。 一种立方相的插入反应，立方相和四方相间的变化，如图2所示。当y＞1时，电压降低1 V 。这是由Jahn-Teller效应引起，形成的MnO6八面体由对称的Oh转化为对称的D4h。根据这些作者提出的，MnO6八面体由(Oh D46) 导致电压下降1V.晶胞体积突然变化(6.5%) 伴随的立方相转变为四方相阶段对电池的性能有很严重的影响。理想的情况是，放电结束，四方LiMn2O4积累，是最初引起容量损失的原因。 近几年来，一部分其他结构的锂锰氧化物已经被合成，并且对脱嵌容量以及可逆性做了检测。一般而言，这些化合物与可以通过循环过程中强烈的驱动力达到热力学稳定。这些过程的改变伴随着容量的损失，尽管反应机理可能在不同的情况下不同。 在LiCoO2 和LiNiO2 电极也发现了相变。图2给出了不同相变以及不同材料对应的y值。对钴酸锂而言，可以观察到六方相和单斜相，但是在镍酸锂可以观察到单斜和斜方六面体。尽管相变发生在这些电极，他们并没有引起容量损失。为了避免在循环过程中的容量衰减，LiyNiO2电极在y在0.3-0.9循环。LiyCoO2在 y在 0.5 -1.0,循环，Gummow 和Thackeray 合成LT-LiCo0.9Ni0.1O2 ,结构介于理想的尖晶石和层状结构之间。他们报道提高了电极的电化学性能，对尖晶石Li0.8[Co1.6Ni0.2]O4合成油促进作用，锂离子在A[B2]O4的A，钴离子和镍离子在B处。 在碳负极，在充电循环中，层之间的不可逆的膨胀发生在c轴。锂嵌入石墨负极，负极由ABA转化为 AAA 结构。石墨的这些结构变化在电极嵌入/脱嵌锂过程中通常是与容量损失没有关系。 容量衰减机理的电池模型 设计电池模型容量衰减和失败机理，必须要将副反应纳入电池模型考虑。完成这个的基本过程在该文章讨论，随后，会完善在特定背景下容量衰减的机理。包括异构和均相反应，但是只有异构电化学反应在该处讨论。过去的在其他体系建立的电池模型包括许多在此要讨论的现象，例如镍镉电池和铅酸电池、的析氧反应。基本的结合副反应的方法是写出一个副反应速率和物质平衡的方程，包括参与反应的每一种物质。因为存在各种近似，可以减少方程的数量。在文献中可以到三明治型电池模型完整的宏观研究方法。 一般的电化学侧反应可以表示为 公式si物质i额化学计量系数，n是参与反应的电子数。反应速率往往受动力学限制，我们首先讨论电化学反应动力学速率表达公式。 动力学Butler-Volmer 速率方程式可以表达每个反应速率： 交流电流密度 o,k 有自己的函数关系： 过电势驱动力的反应方程式 可变因素是 Φ1 和 Φ2，表示两个电极的固相和液相的电位，Uk 是反应的热力学开路电压。ioo,k,γ1,αak和αck是热力学参数。热力学数据是特定材料和电解液条件下。U对应的标准条件下的热力学电势。 任何一个副反应都可以包括在内，每个反应都拥有自己的特定的过电势。如果每个反应被认为是独立发生的，总体的电流密度就是单个反应电流密度的综合： 采用这种处理，主要是依据每个反应的局部电势以及热力学值，也有正极负极电极反应发生在同一个表面上。这种处理预测腐蚀或者自放电过程的短路情况。显然，所有反应的速率通过每个阶段的多孔电极的电势耦合在一起，并且控制一些反应数值是有必要的。 对于许多副反应来说，假定一个不可逆的反应成立，在这种情况下，反应方程可以简化。塔费尔表达式可以用电极过程动力学方程表示： 服从塔费尔方程的副反应，开路电压的电势是任意的，因为没有后续反应。因此，唯一的动力学参数是交换电流密度和塔菲尔斜率（或转移系数）。 一旦动力学方程和速率写入确定，体系存在的物质以及它们质检的相对重要性都可以评估。物质平衡可以写在该物质存在的体系。尽管同质化的化学反应与平衡反应有关系。 物质i的平衡反应方程可以写成一下形式： 其中c1和Nt分贝时物质浓度以及物质通量，R1是净生产率。这个量适用于任何均相反应或者非均相反应，如果后面的反应任然按照类似于均相反应来处理。在多孔电极理论，生成物的反应速率与反应的局部电流有关系，根据下面方程： ak是电极特定的单位界面面积，如果将电化学反应定位于特定的界面，例如电流集流体，生成的物质包括边界产生的物质。可以根据物质在不同的阶段的种类（气体、固体、液体）和浓度分为近似的物质。在许多情况下，副反应速率很小，或者说生成的物质足够小，甚至可以完全忽略。 电池内部传输过程通常用浓度理论描述，该理论考虑了每一对物质之间的相互作用。因为副反应产生大量的产物，简化处理可以减少传输特性需要的模型。例如，电解质溶液可能会假定为单一的溶剂，只有一种盐类，其他任意数量的溶剂认为是浓度稀的溶剂。元溶剂和盐类被认为是二元电解质体系，尽管稀释物质与主溶剂之间有相互作用。这种简化的处理产生了（3+n)型传输性能，n是稀释物质，而不是(3 + n)(2 + n)/2 ，是更严格的处理。在该理论忽视的是不同的稀释物质之间的相互作用以及稀释溶剂与盐类之间的相互作用。 在简单化的条件下，低浓度离子成分的的物质通量可以表示为： Do1 是物质i的相互作用参数或者在溶剂里的扩散系数。低浓度的离子通量由以上的扩散流量和迁移组成： 如果低浓度离子的浓度太低，迁移可以完全忽略，因为盐类可以作为主要的电解质支持物。在以上的两个方程，对流通量被忽略，通常是一个很好的略计。这些通量表达式代入上表面的物料平衡方程，可以得到一个低浓度离子的方程，低浓度离子在电池的时间和位置。这个问题需要给出初始和界面条件。 给定物质的相是电池模型中复杂程度考虑的重要因素。例如，气相物质纯粹在电池里以蒸汽态存在，他们的浓度是均匀的，因为其他物质传输比较快。在这些假定，电池内部的压强可以评估，每部分物质的压强可以测定。在其他情况下，气体在电解质的溶解度是电池性能或者自放电的重要影响因素，但是在某些情况下，这些假定并不成立。三种物质的相，可以认为是连续叠加相，在不考虑多孔电极的几何细节。 低容量物质被视为局部反应，例如前期电池模型里的盐的沉积。反应中间体，例如溶剂氧化或者还原产生的自由基，被视为伪稳态处理，生成速率为零的物质。这些假定的影响在后面章节会进一步讨论。 钝化膜的生长——副反应引起电极颗粒表面的钝化膜生长，在电池模型的实例中都可以观察到。碳化负极的钝化膜或者固体电解质钝化膜（SEI)的形成是所有电池的一个基础过程。滥用条件下，例如过充会导致负极有金属锂沉积。沉积的金属锂会导致二次反应，生产产物或者二次膜。其他的副反应可能会间接的形成膜，因为其在非水电解质的浓度低。 我们从过充负极产生钝化膜和负极产生金属锂沉积开始讨论，因为这个过程在哦所有的锂离子电池都发生，提供了商业电池的关键的安全性问题。一些不同的近似反应可以包括在锂离子电池的金属锂沉积的副反应模型。考虑到这个原因，模型的细节主要是由它的预测能力决定，具有更高的预测能力就需要更多的细节设计。在决定所需的复杂性之前，碳负极以及石墨负极的过充的实验数据都是必要的。在这些数据之上，过充对放电以及循环的影响可以评估，并且实验现象可以引进模型。另外，在适当的条件下，锂离子电池沉积的动力学数据也是必需要有的。 因为锂沉积反应是比较容易发生的过程，因此表面电位比较低，反应过程可以用线性动力学方程表示： 对于固体物质金属锂，物质通量基本上是零时一个近似值。综合负极的物质平衡，副反应速率与电极颗粒表面的钝化膜的生长有关系 δ固体锂产物的厚度，L代表负极的厚度。 膜的厚度以及假定的特性例如导电性以及介电常数，可以纳入电池模型来预测挤压对放电的影响以及电化学电阻。Pollard 采用同样的方法处理电化学电池的盐膜沉积。随着充放电过程膜的厚度的增加，界面电阻增加，电极的电流分布也会改变。同样，由于副反应的电流密度，电池的容量平衡被改变。其他的影响，例如电极的孔隙度以及沉积金属锂的二次反应也可以一并考虑在内。 另外一种普遍存在在锂离子电池模的是最初循环或者化成阶段负极表面生成钝化膜的过程。钝化膜形成模型和金属锂过充沉积的过程类似。然而，导致固体沉积物的还原反应更不好理解，数量更大，而且性质根据电解质的组分而变化，这在早期的部分都可以观察到。因为这些原因，需要一个更加简化的模型，钝化过程可视为溶剂的消耗以及锂离子单一产物的形成，例如碳酸锂。电极/电解质界面的钝化模型可以允许膜的厚度和相关的性能可以追踪。其他中间产物的形成，例如自由基，要么是完全忽略，要么是简单化处理（例如，自放电速率的追踪）。通过跟踪钝化过程溶剂、锂离子、电子的消耗，副反应对电池容量平衡的重要影响既可以被评估。 根据以上假定，薄膜模仿一个由时间控制厚度的单层膜。薄膜的厚度根据局部还原反应的电流密度计算。因为副反应的确切性能要么是未知要么是简单化了，速率表达式经验拟合实验数据确定，例如化成阶段的充放电曲线。薄膜厚度的计算模型与薄膜阻抗和假定的物理性质电容有关系。由于薄膜阻抗随着时间变化，界面电阻增加和容量损失在电池里会发生。假定电荷转移过程的限制速率通过钝化膜层大量的锂迁移控制，然后钝化膜作为纯电阻处理是合理的。在其他情况下，电荷转移的限制速率是一个很关键的考虑因素。最后，如果该薄膜有一定的孔隙，或者不是一个纯粹的阳离子道题，则要考虑薄膜内部的锂离子扩散。 电解质分解反应-大量不同种类的溶剂和盐类混合物被运用到锂离子电池。因为盐类和溶剂的氧化还原机理非常复杂，如果建立一个单一通用的模型处理不同体系是不可行的。未来这方面的模型基本上是限定在特定环境和体系的模型。电解质分解还原反应模型虽然不是很难的数学，但是因为复杂的反应机理，大量的反应方程以及不确定，需要计算整个过程。电化学动力学数据，包括每一个还原反应以及每种物质的转移特性数据在模型里是必要的。通过标准方法处理这些体系是非常必要的，例如，拟稳态或限速步骤近似方法。 一个严格的副反应处理，涉及到溶剂和盐类，需要大量的关于这些过程速率的基础数据，所以限制速率步骤可以被评估。幸运的是，在实际电池建模以及预测其对电池的影响后果应运中，这些细节是没有必要的，对这些过程的理解可以采用目前阶段的文献的知识。副反应涉及电解质溶剂的氧化和还原主要集中在这个部分，因为相似的问题以及位置存在这个模型，例如，溶剂或者盐类的损失，此生产物的生成过程和反应，电池容量平衡的干扰。 在目前的锂离子电池模型，正极的开路电压当y最小的时候达到无穷大(e.g., U→∞ as y→0.4 for LiyCoO2) ，电解质没有发生氧化副反应为前提条件。将这个条件应用于插入电极的开路电压制止了插入过程，在所需的化学计量下，热内停止，通过把电极极化到无穷大。副反应（化学反应和电化学反应）包含在电池模型中，包括二次反应过程，进行的平行反应通过独立的动力学防长。电化学副反应取决于局部电极电势，通常采用塔菲尔方程组表示，因为许多是不可逆的反应。电化学或者热诱导降解过程基本不依赖电极电势，很大可能是滥用条件引起。 这些副反应的存在对电池模型有剧烈的影响。通过跟踪氧化或者还原速率，即使在滥用的条件下，实际的化学计量的锂是可以预测的。这就允许电池容量衰减，以及衰减的危害，与每个电极损失容量的积累有关。在许多情况下，盐类和溶剂的损失只占总电解质的一小部分，因此，对电视性能的影响可以忽略。然而，在长期循环条件下，体系电解质损失积累会显示容量损失挥着电池容量。副反应生成的产物可以被监控，以及它们在二次反应的角，例如自放电。这些产物可能是液相、气相或者固相，取决于反应的性质，每种情况下，都会出现近似，如前面讨论。 在某些情况下，电极活性物质自己发生副反应。例如，脱锂和尖晶石锰酸锂溶解就是例子，溶剂氧化或者盐类水解导致副产物（质子）生成，参与了电极材料的二次反应。经常这些过程非活性物质的产生，这些产物产生的方法：阻塞电极颗粒表面孔隙，固相沉淀，或者物质相互转化生成不可逆锂，或者在i电压范围内无锂插入容量。 这些过程建模需要每个参与副反应的物质的质量平衡，尤其是与活性物质消耗以及惰性物质生成有关的物质平衡反应。一个相对简单的理论解释了惰性或者非电化学活性物质混合物是调节活性物质体积分数，增加了非活性物质的体积比例，当活性物质是一定比例时。例如，跟踪滥用条件下电池内部的质子产物，电极分解和转化为非活性物质额速率可以跟踪。这些非活性物质可以被认为是均匀分布在复合电极的表面，类似于非活性粘结剂均匀的分布在加了碳黑的多孔电极表面理论 腐蚀和溶解过程——外部的阳离子可以同各国电极溶解和集流体腐蚀过程进入电池内部。阳离子物质除了锂都会对电池的性能产生危害，因为它们很容易被还原，最后会在充电过程沉积在负极。最坏的条件下，这些过程会导致枝晶生长，穿过隔膜导致内部短路。这些锰钴沉积物的存在可以被检测，集流体的腐蚀物质也可以被检测到，例如铜和铝，同样有文件记录。这些过程建模似乎有更大的挑战，因为一些发生的基础过程以及金属内部杂质的电池内部会如何我们并不是非常清楚（而不是他们的最终会在负电极）。 例如，它可以预测放电条件下的负电极铜溶解作用。这就需要优化电池设计，限制负极最大电极电位，防止铜的重大损失。不幸的是，铜溶解沉积电位很难预测，因为非水体系的数据很少。对于任何给定的非水电解质体系，铜溶解热力学开路电位必须通过实验评估。如果这些数据是已知的或已经假定，负极充电过程铜的溶解反应采用Butler-Volmer速率方程和液相铜物种物质平衡类似模型处理。然而，枝晶的生长导致电池短路很难预测，这是一个沉积形态的函数。 相关的话题，正极锰酸锂电极溶解可以被包括在电池模型里，但是得增加副反应酸溶解机理。方程59发生在正极电极表面， LiPF6水解是一个简单化了的化学反应，H2释放发生在负极表面。注意，这些过程自发的发生，只要尖晶石锰酸锂溶解产生水，可以从电解质盐类的溶剂热产生更多地质子。这些反应包括在电池模型，通过每种物质的物质平衡，虽然大量的简化是有必要的。每种反应速率可以写成方程70,。当正极溶解正极活性物质会损失，导致容量损失。同样，锰溶解发生在负极，会阻塞负极孔隙，同样导致容量损失。 自放电过程——尽管自放电是锂离子商业电池的一个重要现象，但是在目前的模型里却没有引起足够的重视。然而，自放电过程是比较简单的，一旦引入三明治模型，并且将一些必要的副反应引入模型。自放电过程可以分为设计到单个电极和涉及到双电极的若干个部分，例如耦合电解质分解，锂离子嵌入和脱出的反应，双电极之间的相互反应，包括氧化还原反应和电池短路反应。 单电极的自放电过程可以通过简单的增加电化学动力学方程（方程74）到电解质分解的整体数学模型来处理。例如，高电压的正极电极自放电模型包括溶剂的氧化，例如碳酸盐的氧化是通过塔菲尔方程（方程35）。由于塔菲尔方程的特性，溶剂氧化过程发生在整个电压过程但是在高电压条件下明显增加，取决于交换电流密度和热传递系数。当模型里的电流中断，不是到达一个稳定的电压，而是电池电压自发的降低，由于锂嵌入正极与溶剂氧化过程对立。 其他一些自放电机理也可以被简单地引入到一个宏观电池模型。泄漏电流被引入到模型，在其他开路条件下，通过恒定的放电的电阻。测量一个适当的电阻值，运用于放电，或者说这个电阻值是合理值，在自放电实验数据的基础上。氧化还原往复型自放电机理需要一个氧化还原的源物质，例如溶剂分解反应，以及物质平衡预测转移速率和浓度（二者都影响自放电速率）。在前期的文献中，自放电过程已经被实验证实，但是仍然需要更多的实验数据，通过特性体系证实模型的预测能力。 鸣谢 作者感谢美国中央情报局的研究部办公室对该项目的财务支持，合同号93-f148100-l00。作者也感谢与美国博士Dr. A. S. Gozdz of Bellcore.对容量衰减机制的有益讨论。 手稿于1997年11月18日提交；1998年4月17日修订稿。 南卡罗来纳州大学在会议上协助了这篇文章的出版费用。 |
本文发布于:2023-05-23 17:08:43，感谢您对本站的认可！
本文链接：https://patent.en369.cn/patent/3/111211.html
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系，我们将在24小时内删除。
| 留言与评论（共有 0 条评论） |