Paul Albertus

Identifying Capacity Limitations in the Li/Oxygen Battery Using Experiments and Modeling

The Li/oxygen battery may achieve a high practical specific energy as its theoretical specific energy is 11,400 Wh/kg Li assuming Li 2 O 2 is the product. To help understand the physics of the Li/oxygen battery we present the first physics-based model that incorporates the major thermodynamic, transport, and kinetic processes. We obtain a good match between porous-electrode experiments and simulations by using an empirical fit to the resistance of the discharge products (which include carbonates and oxides when using carbonate solvents) as a function of thickness that is obtained from flat-electrode experiments. The experiments and model indicate that the discharge products are electronically resistive, limiting their thickness to tens of nanometers and their volume fraction in one of our discharged porous electrodes to a few percent. Flat-electrode experiments, where pore clogging is impossible, show passivation similar to porous-electrode experiments and allow us to conclude that electrical passivation is the dominant capacity-limiting mechanism in our cells. Although in carbonate solvents Li 2 O 2 is not the dominant discharge product, we argue that the implications of this model, (i.e., electrical passivation by the discharge products limits the capacity) also apply if Li 2 O 2 is the discharge product, as it is an intrinsic electronic insulator.

Batteries for Electric and Hybrid-Electric Vehicles

Batteries have powered vehicles for more than a century, but recent advances, especially in lithium-ion (Li-ion) batteries, are bringing a new generation of electric-powered vehicles to the market. Key barriers to progress include system cost and lifetime, and derive from the difficulty of making a high-energy, high-power, and reversible electrochemical system. Indeed, although humans produce many mechanical and electrical systems, the number of reversible electrochemical systems is very limited. System costs may be brought down by using cathode materials less expensive than those presently employed (e.g., sulfur or air), but reversibility will remain a key challenge. Continued improvements in the ability to synthesize and characterize materials at desired length scales, as well as to use computations to predict new structures and their properties, are facilitating the development of a better understanding and improved systems. Battery research is a fascinating area for development as well as a key enabler for future technologies, including advanced transportation systems with minimal environmental impact.

Experiments on and Modeling of Positive Electrodes with Multiple Active Materials for Lithium-Ion Batteries

We adapt a previously developed lithium-ion mathematical model to treat multiple types of active materials in a single electrode; our model treats both direct (galvanostatic) and alternating (impedance) currents. We compare our simulations to experimental data from coin cells built with two positive-electrode materials (compositions based on Li y Ni 0.80 Co 0.15 Al 0.05 O 2 and Li y Mn 2 O 4 ) mixed in five different molar ratios and develop a model parameter set that qualitatively matches both the galvanostatic and impedance data. We found that to match the behavior of the high rate discharge curves and the impedance data (which showed a similar width of the positive-electrode kinetic arc for any composition containing Li y Mn 2 O 4 ), multiple types of electronic connections between the active material and the conductive matrix were required. Our experiments showed that at high powers the specific energy from an electrode with pure Li y Mn 2 O 4 exceeded that from an electrode with pure Li y Ni 0.80 Co 0.15 Al 0.05 O 2 , while at low specific powers the Li y Ni 0.80 Co 0.15 Al 0.05 O 2 electrode had a higher specific energy. Mixing these active materials combined power and energy characteristics. We discuss other applications in which a mixed active-material electrode may be beneficial. For example, combining a sloped-potential system with a flat-potential system may assist in electrode state-of-charge determination.

Two-Dimensional Modeling of Lithium Deposition during Cell Charging

Empirical observations have shown that lithium deposition occurs preferentially at electrode edges and that extending the negative electrode beyond the edge of the positive by approximately 1 mm can prevent deposition from occurring. In this work, we use a simplified COMSOL Multiphysics model to explain this behavior and to investigate the conditions under which deposition occurs, paying particular attention to the magnitude of edge effects. Model results show that geometric effects generate overpotential at the edge of the electrode and create conditions which favor plating, despite unused capacity in the center of the electrode. Extending the negative electrode beyond the edge of the positive provides excess capacity where it is needed and prevents deposition from occurring before the cutoff potential is reached. Under the assumptions of this model, an extension of 0.4 mm is sufficient to prevent the onset of lithium deposition until after the cutoff potential is reached.

Optimizing the Performance of Lithium Titanate Spinel Paired with Activated Carbon or Iron Phosphate

Here we describe a model for lithium titanate spinel paired with an activated carbon electrode: an asymmetric hybrid supercapacitor. The model is compared to experimental results. The performance of this system is compared to a lithium titanate spinel/lithium iron phosphate battery. The model is used to study the performance of these chemistries and to assist in cell optimization. A Ragone plot is generated for various cell designs in order to assess the ability of the chemistries to achieve the U.S. Department of Energy goals for hybrid-electric vehicles. The specific energy of a cell is maximized by optimizing the design for a fixed time of discharge. The thickness and porosity of both electrodes are varied, while holding constant the capacity ratio for the two electrodes, as well as the properties of the separator. The capacity ratio can also be optimized for each time of discharge. A 41% increase in specific power is seen when one optimizes the capacity ratio of a lithium titanate spinel/iron phosphate battery. The optimized designs derived here can be used as a starting point for battery manufacturers and to help decrease the time to commercialization.

Modeling Side Reactions and Nonisothermal Effects in Nickel Metal-Hydride Batteries

Previous modeling of the NiMH chemistry has not comprehensively covered the phenomena occurring in NiMH batteries, and model validation has focused on the low rates characteristic of electric vehicles. We have expanded a previously developed Fortran battery model to treat the NiMH system rigorously. The model now includes material balances (in the superimposed solid, liquid, and gas phases), Ohms law (in the solid and liquid phases), kinetic expressions (for the main insertion and side reactions), and a charge balance. Other features include a rigorous energy balance, appropriate temperature-dependent properties, and the ability to treat a variable solid-phase diffusion coefficient. The model can now simulate the following experimentally observed features of the NiMH system: overcharge protection, self-discharge, and pressure-voltage-temperature coupling. Model validation has been carried out at a variety of rates (charge/discharge times from 6 min to 20 h) on a battery module removed from a 2005 Toyota Prius. We have constructed an optimized Ragone plot that allows comparison of the performance of the NiMH chemistry to lithium-ion chemistries and shows the NiMH chemistry to be a capable high-power system. Our simulations also show how aging of the NiMH system may lead to significant generation and venting of hydrogen gas.

Overview of LiO2 Battery Systems, with a Focus on Oxygen Handling Requirements and Technologies

The reactions of Li and O2 to form Li2O2, and of Li, O2, and H2O to form LiOH·H2O, have exceptional energy content but are adversely affected by components of air such as CO2 (for both cases) and H2O (for the Li2O2 case). Hence, a method is required to supply O2 while excluding contaminants. In this chapter we focus on O2 supply for both a closed system (in which tanks store pure O2 at pressures up to 350 bar) and an open system (in which CO2 and possibly H2O are removed through a series of unit operations). In particular, we consider the implications of the O2 supply method on the specific energy and energy density at the system level, as well as other system attributes such as cost. For the closed (tank) system we find that with the use of a carbon fiber tank, for the reaction forming Li2O2, the specific energy is twice that of a comparison cell (one pairing Li metal with an advanced intercalation metal oxide), but the energy density is about 30 % lower. For the reaction forming LiOH·H2O, the specific energy is about 40 % above that of a Li/metal oxide cell, but the energy density is 50 % lower. A unique challenge for the closed system is the need for high-pressure compression. An open system may be enabled through the combined use of several gas separation steps (including a membrane and solid adsorption) as well as a compressor to drive the air. The required purity of an O2 supply stream remains unclear, but for a reduction of CO2 and H2O to levels of 1 ppm, the mass and volume of the O2 supply equipment for the open system is comparable to that of the closed system. A unique challenge for the open system is safely charging in closed environments where the O2 emitted does not quickly dissipate. For both types of systems, handling any volatile cell components (e.g., solvents) may be a challenge (for the closed system they may enter the high-pressure O2 tanks, while in the open system they may be lost to the atmosphere), and potential technologies to address volatiles are not included in this analysis. We encourage Li/O2 researchers to investigate sets of nonvolatile materials that may improve the robustness of the cell chemistry to the presence of air contaminants.

Battery Size and Capacity Use in Hybrid and Plug-In Hybrid Electric Vehicles

This chapter focuses on the issue of battery size and capacity used in hybrid and plug-in hybrid electric vehicles (PHEVs). Several battery chemistries such as nickel–metal hydride (Ni/MH) and lithium ion are either in use or being considered for use in HEVs and PHEVs. The fundamental properties of a pair of electrode materials is related to the size of a battery required for a particular application and the fraction of the capacity in the battery that can be accessed during a given driving cycle is discussed. The three main electrode properties, which include the specific capacity (in mAh/g), the magnitude of the cell equilibrium potential (in V), and the shape of the equilibrium potential, are also discussed. A simple battery model to show the relationships among the major design variables by neglecting the details of an actual battery operation is presented. The combined battery and vehicle model captures the details of the complex relationships between battery chemistry and performance. These models show that a large battery energy and maximum pulse-power capability decrease battery size and increase capacity use, but the influence of the shape of the equilibrium potential is more subtle. In the case of a cell with a flat equilibrium potential, there is no driving force for the relaxation of concentration gradients through the depth of the electrodes and the persistence of concentration gradients can reduce performance by shifting the current distribution, which results in more polarization for consecutive charge or discharge pulses.