Advances in materials innovation, availability, processing, and volumetric and gravimetric utilization, have lead to dramatic changes in the technologies and markets for batteries, and other electrochemical storage devices, in recent years. The increased ability to cycle of the new materials and preparations has also resulted in a more dominant proportion of secondary batteries in the market. The growth of the various technology sectors of the battery and energy storage device industry will be compared to the relevant improvements in materials science. The developments needed in materials, composites, and processing techniques is reviewed, in order to match the ongoing changes in electronic circuits, power devices, transportation, and the growth of portable appliances. An analysis of the technical/market segments of the battery industry is presented.

Two crucial characteristics are expected for batteries. One is large capacity (large energy density) and the other high power. In the last decade, the batteries with large capacity and energy density have been strongly required in the field of portable electronics and communication equipments and, hence, the new batteries such as Ni-MH and Li-ion ones have been developed and commercialized. At present, Ui-ion battery has the largest energy density of all the available batteries and its energy density has come up to 300WM/l, twice as large as that of Ni-Cd one. In general, electric power of batteries signifies the amount of energy put out at unit time. Thus the output power discharged from a battery, storing a certain amount of capacity, in Ihour is ten times as much as that discharged in 10 hours. It is indeed an excellent ability to go on walking all the way with a heavy knapsack on one's back but so is it to output a tremendous power at a moment just as in the weightlifting. It is well known that Ni-Cd battery can output the highest power of all the batteries. It can be discharged at about 10C, namely in about 6 minutes. Large energy density and high output power seem to be similar, but they are essentially different.

Capacity recovery after long term storage and loaded storage is a critical issue with the NiMH system since its inception. A measurable loss in capacity is observed when cells are stored for long periods of time or discharged deeply to zero volts. The different mechanisms that are known to cause self discharge and capacity loss after storage and loaded storage will be the focus of this paper. Capacity loss after long term storage involves two main events. One is self discharge which causes the open circuit voltage(OCV) of the cell to drop. Self discharge is caused by decomposition of NiOOH, migration of metal ions and possible degradation of separator. Self discharge can be prevented by using separators which are stable at high temperatures and pH and have good ion trapping capability. Various separator types and treatments can play an important role in inhibiting metal ions from migrating thus reducing self discharge. Self discharge during storage causes a severe suppression in the voltage of the foam positive electrode. This drop in voltage causes a breakdown of the cobalt conductive network in the nickel positive electrode. Reduction of high valence cobalt(III) which forms the electrode's conductive network takes place at these low voltages. A permanent breakdown in the conductive network results in low efficiency of the cell on consecutive charge and discharge cycles. In addition, the cobalt in its lower valence states can migrate away from the electrode into the separator causing shorts. These events effect the charge and discharge efficiency of these cells thereby resulting in capacity loss. Various mechanisms causing self discharge which affect capacity recovery after long term storage and loaded storage are discussed in this paper.

A high pressure gas atomization approach to rapid solidification has been employed to investigate simplified processing of Sn modified LaNi5 powders that can be used for advanced Ni/metal hydride (Ni/MH) batteries. The current industrial practice involves casting large ingots followed by annealing and grinding and utilizes a complex and costly alloy design. This investigation is an attempt to produce powders for battery cathode fabrication that can be used in an as-atomized condition without annealing or grinding. Both Ar and He atomization gas were tried to investigate rapid solidification effects. Sn alloy additions were tested to promote subambient pressure absorption/desorption of hydrogen at ambient temperature. The resulting fine, spherical powders were subject to microstructural analysis, hydrogen gas cycling, and annealing experiments to evaluate suitability for Ni/MH battery applications. The results demonstrate that a brief anneal is required to homogenize the as-solidified microstructure of both Ar and He atomized powders and to achieve a suitable hydrogen absorption behavior. The Sn addition also appears to suppress cracking during hydrogen gas phase cycling in particles smaller than about 25μm. These results suggest that direct powder processing of a LaNi5−xSnx alloy has potential application in rechargeable Ni/MH batteries.

Rechargeable zinc-air batteries are a long run time solution for portable electronics. Cells with a nominal voltage of 1.05 V have a specific energy of 169 Wh kg−1 and aenergy density of 219 Wh L1. This cell is capable of delivering 16 – 18 Ah of capacity at 2 A (16.7 mA cm−2) discharge rate. However, the cycle life of these cells is artificially shortened because of the generation of carbon dioxide in the alkaline electrolyte during the cycling of the cell. The carbon dioxide has the effect of removing the OH− (equation 1) needed by the anode

half reaction (equations 2 and 3). The need for hydroxide is demonstrated in equation 2.

There is literature evidence [1,2] that the CO2 was the result of corrosion of the carbon matrix of the air electrode. We report here the results of our investigation of this corrosion reaction and what properties of the carbon contribute to the corrosion reaction rate.

In this study, the evolution of the stress-states ahead of the penny shaped internal cracks in both spherical and disk shaped ReNi5 particles during hydrogen charging and discharging cycles were investigated using coupled diffusion/deformation FEM analyses. The results indicate that large tensile stresses, on the order of 20–50% of the modulus of elasticity, develop in the particles. The disk shaped particles, in addition to having faster charging/discharging cycles, may offer better resistance to fracture than the spherical particles.

It was noted at the earlier study and at the longer observations of the novel various types of superstoichiometric fluorocarbon materials CF1+x, where x = 0.1–0.33 (FCM) and their behavior, that despite of their known hygroscopity during a storage of samples in laboratory and technological utensils nevertheless occurs an appreciable sorption of atmospheric moisture. The color of samples does not change but sometimes there is appears a smell of hydrogen fluoride and even corrosion of glasswares at a long storage. On the basis of these facts was assumed that at a long storage the slow reactions of HF producing with a sorption moisture can proceed. This phenomena is necessary to take into account for successful manufacturing of long life lithium cells based on superstoichiometric fluorocarbon composite cathodes (FCC).

Ab initio methods have started to be widely used in materials science for the prediction of properties of metals, alloys and compounds. These methods basically require only the atomic numbers of the constituent species. Such methods not only provide us with predictions of some of the properties of the material (even before synthesizing it) but also help us in understanding the phenomena that control those properties.

The use of ab initio methods in the field of electrochemistry is, however, quite recent and rare [1–4]. In this study, we demonstrate how ab initio methods can be used to investigate the properties of secondary lithium batteries. Particular examples will be given in predicting average insertion voltages in spinel Li-Mn and Li-Co oxides and in layered LiMO2 (M = Ti, V, Mn, Fe, Co and Al) compounds. Additionally, the stability of these compounds to metal reduction and structural stability of LiCoO2 upon lithium removal is investigated. We find that the oxygen anion plays an active role in the electrochemical intercalation of lithium. The amount of electron transfer to oxygen occurring upon lithium intercalation correlates strongly with the cell voltages. The more electron transfer to oxygen occurs, the higher lithium intercalation potential is obtained.

Several types of cation- and vacancy-ordering exist in the LixCoO2 battery material. The ordering patterns are of interest due to the fact that they can control the voltage in rechargeable Li batteries. We present a first-principles total energy theory which can predict both cation-and vacancy-ordering patterns at both zero and finite temperatures. Also, by calculating the energetics of the Li intercalation reaction, this theory can provide first-principles predictions of battery voltages of Lix;CoO2/Li cells. Our calculations allow us to search the entire configurational space to predict the lowest-energy ground state structures, search for large voltage cathodes, explore metastable low-energy states, and extend our calculations to finite temperatures, thereby searching for order-disorder transitions and states of partial disorder.

Semi-empirical and ab initio calculations [1], as well as inelastic neutron spec-troscopy [2], demonstrate that Li can bind to protonated “edge carbons” to create a moiety analogous to the organolithium monomer C2H2Li2. This provides a possible additional channel for Li uptake in high capacity Li-ion battery anodes based on low-T pyrolyzed soft carbons. Here we show that similar reactivity is exhibited by polyaro-matic hydrocarbons with the protons removed (taken as surrogates for the structural units in hard carbons). In the deprotonated PAH'es the Li serves to saturate dangling bonds, maintaining sp2 hybridization, whereas Li added to PAH'es creates sp3 carbons at the edges. In both cases this extra reactivity occurs in parallel with the usual intercalation. These findings have implications for further development in Li-ion rechargeable battery technology.

Disordered carbons that deliver high reversible capacity in electrochemical cells have been synthesized by using inorganic clays as templates to control the pore size and the surface area. The capacities obtained were much higher than those calculated if the resultant carbon had a graphitic-like structure. Computational chemistry was used to investigate the nature of lithium bonding in a carbon lattice unlike graphite. The lithium intercalated fullenere Lin-C60 was used as a model for our (non-graphitic) disordered carbon lattice. A dilithium-C60 system with a charge and multiplicity of (0,1) and a trilithium-C60 system with a charge and multiplicity of (0,4) were investigated. The spatial distribution of lithium ions in an electrochemical cell containing this novel disordered carbon material was investigated in situ by Li-7 NMR using an electrochemical cell that was incorporated into a toroid cavity nuclear magnetic resonance (NMR) imager. The concentration of solvated Li+ ions in the carbon anode appears to be larger than in the bulk electrolyte, is substantially lower near the copper/carbon interface, and does not change with cell charging.

We report modeling studies of polyethylene oxide which are carried out with the goal of elucidating the mechanism of ion conduction in the temperature range of interest to battery applications. We review our previous work in which the amorphous regions of the polymer between its glass and melting temperatures is modeled by a molecular dynamics algorithm in which the model system is polymerized from a model monomeric liquid. We describe new work in which the hydrogen centers are added to the model in order to permit comparison with recent neutron work. We compare our simulations of frequency dependent conductivity with experiment and end with a brief discussion of possibilities for improved conductivity which our current understanding suggests.

A series of molecular dynamics computer simulations of the self-diffusion of lithium in pure and several doped lithium-manganese spinel materials has been completed. The theoretical approach is part of an effort to understand the mechanisms and rates of lithium diffusion, and to evaluate the structural control of the cathode materials upon lithium intercalation (charge-discharge) process. The molecular dynamics approach employs a fully ionic forcefield that accounts for electrostatic, repulsive, and dispersion interactions among all ions. A reference unit cell comprised of 56 ions (Li8Mn3+8Mn4+8O32) is used to perform the simulations under constant volume and constant pressure constraints. All atomic positions are allowed to vary during the simulation. Simulations were completed for the undoped and doped LiMn2O4 at various levels of lithium content (based on the number of lithium ions per unit cell and manganese oxidation state). The molecular dynamics results indicate an activation energy of approximately 97 kJ/mole for self-diffusion of lithium in the undoped material. Lithium ion trajectories from the simulations provide diffusion coefficients that decrease by a factor of ten as the cathode accumulates lithium ions during discharge. Molecular dynamics results for the doped spinel suggest a decrease in the diffusion rate with increasing dopant ion.

The structure and electrochemical potential of monoclinic Li1+xV3O8 were calculated within the local-density-functional-theory framework by use of plane-wave-pseudopotential methods. Special attention was given to the compositions 1+x=1.2 and 1+x=4, for which x-ray diffraction structure refinements are available. The calculated low-energy configuration for 1+x=4 is consistent with the three Li sites identified in x-ray diffraction measurements and predicts the position of the unobserved Li. The location of the tetrahedrally coordinated Li in the calculated low-energy configuration for 1+x=1.5 is consistent with the structure measured by x-ray diffraction for Li1.2V3O8. Calculations were also performed for the two monoclinic phases at intermediate Li compositions, for which no structural information is available. Calculations at these compositions are based on hypothetical Li configurations suggested by the ordering of vacancy energies for Li4V3O8 and tetrahedral site energies in Li1.5V3O8. The internal energy curves for the two phases- cross near 1+x=3. Predicted electrochemical potential curves agree well with experiment.

The electrochemical properties of the layered intercalation compound LiCoO2 used as a cathode in Li batteries have been investigated extensively in the past 15 years. Despite this research, little is known about the nature and thermodynamic driving forces for the phase transformations that occur as the Li concentration is varied. In this work, the phase diagram of LixCoO2 is calculated from first principles for x ranging from 0 to 1. Our calculations indicate that there is a tendency for Li ordering at x = 1/2 in agreement with experiment [1]. At low Li concentration, we find that a staged compound is stable in which the Li ions selectively segregate to every other Li plane leaving the remaining Li planes vacant. We find that the two phase region observed at high Li concentration is not due to Li ordering and speculate that it is driven by a metal-insulator transition which occurs at concentrations slightly below x < 1.

Lanthanum gallate is a promising material for “monolithic” fuel cells or oxygen pumps, i.e. one in which the electrolyte and electrodes are formed from a common phase. We have investigated La1−xSrxGa1–yNiyO3 (LSGNx-y) with x=0.1 and y=0.2 and 0.5 as a potential cathode material for such an electrochemical device. The σ(PO2,T) for LSGN10–20 points to a p-type electronic conductivity at high PO2 and predominantly ionic conductivity at low PO2. LSGN10–50 has an electronic conductivity suitable for SOFC applications of approximately 50 S/cm in air at high temperature. AC impedance spectroscopy on an electron blocking cell of the form M/LSG/LSGN/LSG/M was used to isolate the ionic conductivity in the LSGN10–20 material. The ionic conductivity was found to have a similar magnitude and activation energy to that of undoped LSG material with σi= 0.12 S/cm at 800°C and EA= 1.0 ± 0.1 eV. Thermal expansion measurements on the LSGN materials were characterized as a function of temperature and dopant level and were found to match that of the electrolyte under opeating conditions.

Carbon and fluorinated carbon materials are major constituents of phosphoric acid fuel cells and PEM fuel cells and the stability of these materials is critical for long life operation. Laboratory corrosion studies of separator plate materials were correlated with separator plate changes in commercial PAFC fuel cells. The addition of thin films of Teflon™ to the separator plates extends the life of the separator plates to 60,000+ hours through the formation of a temporary hydrophobic barrier. ESCA studies show the loss of hydrophobicity with time of PAFC electrodes to be a result of delamination of the Teflon from the carbon and not corrosion of the Teflon by phosphoric acid. The projected life of PAFC power plants has been confirmed by commercial operation of power plants for over 40,000 hours.

The structure, thermal expansion coefficients and ionic/electronic conductivity of (Ce1−xSmx)1-y(Tb/Pr)yO2-x/2)+δ (x=0–0.30; y=0–0.10) and Ce1−xCaxO2-x ( x= 0–0.17) solid electrolytes prepared hydrothermally were investigated. The uniformly small particle size (7–68 nm) of the hydrothermally prepared materials allows sintering of the samples into highly dense ceramic pellets at 1400°C, a significantly lower temperature, compared to that at 1600–1650°C required for samples prepared by solid state techniques. The maximum ionic conductivity in Ce1−x(Sm/Ca)xO2-δ was found at x= 0.17 for the Sm (σ600°c = 5.7×10−3 S/cm). In (Ceo0.83Sm0.17)1-y(Tb/Pr)yO1.915+δ the maximum conductivity was found at y= 0.17 for the Pr and Tb substituted ceria (σ600°c=7.6x10−3S/cm, Ea= 0.55 eV and σ600°c = 10−2 S/cm, Ea= 0.72 eV respectively) with electronic contribution to total conductivity around 20–30 %. When the Tb or Pr substitution in Ce0.83Sm0.17OO1.915 is reduced, the conductivity becomes more ionic, and is purely ionic at 2 %. However the conductivity at this lower level doping is not significantly lower.

Electrode materials for solid-oxide fuel cells are developed using sputter deposition. A thin film anode is formed by co-deposition of nickel and yttria-stabilized zirconia. This approach is suitable for composition grading and the provision of a mixed-conducting interfacial layer to the electrolyte layer. Similarly, synthesis of a thin film cathode proceeds by co-deposition of silver and yttria-stabilized zirconia. The sputter deposition of a thin film solid-oxide fuel cell is next demonstrated. The thin film fuel cell microstructure is examined using scanning electron microscopy whereas the cell perfomance is characterized through current-voltage measurement and corresponding impedance spectroscopy.

Lithium aluminate, LiA1O2 is a ceramic powder which is used as the porous solid support for the electrolyte in molten carbonate fuel cells (MCFCs). It has previously been reported that gamma LiAlO2 will convert to lithium aluminum carbonate hydroxide hydrate, Li2Al4(CO3)(OH)12 3H2O and Li2CO3 when exposed to water vapor and carbon dioxide. We compare three techniques, weight gain, carbonate content and x-ray diffraction to measure the amount of conversion. The reaction may involve amorphous intermediates and no one technique by itself is satisfactory to study the conversion.