• Energy Research
• US close
• DE close
• KR close
• Journal of Power Sources close

Abstract Thin microfibrous substrates were built using a low cost, high speed, wet lay papermaking process with nickel and copper microfibers. High quality microfiber sheets with designed thickness and void volumes were fabricated at the Center for Microfibrous Materials Manufacturing at Auburn University. Products include thin microfibrous nickel and copper substrates with or without suitable expanded metal meshes, either sandwiched inside or on the backside. An improved method for fabrication of zinc electrodes was accomplished by electro-depositing active zinc material on 6 and/or 9 μm diameter microfibrous copper substrates. These zinc electrodes with various additives were examined for shape change and tested for performance. A new structure of electrode design was employed using an accordion-fold cell. By using five individual segments sintered onto an expanded metal mesh, an accordion-fold cell (670 mAh rated capacity) was assembled and had 98.0% coulomb efficiency, 88.6% energy efficiency at 72% depth-of-discharge (DOD) and 73.2% state-of-recharge (SOR) in the 79th cycle. A Ni–Zn cell consisting of one thick nickel and one thick zinc electrode was tested for rate performance in comparison with the 670 mAh accordion-fold cell. The accordion-fold cell (1279 mAh rated capacity) was scaled-up and tested for 835 cycles at 38% DOD at a 0.3 C charge and discharge rates, and continuously operated at 38% DOD until the 1050th cycle was finished at 0.16 C charge/discharge rates. A 7115 mAh accordion-fold demonstration cell was tested at 0.28 C charge and discharge rates for 608 cycles and still had a 4.7 Ah capacity at a 0.08 C discharge rate.

Abstract Integrating PEM fuel cells effectively with liquid hydrocarbon reforming requires careful system analysis to assess trade-offs associated with H 2 production, purification, and overall water balance. To this end, a model of a PEM fuel cell system integrated with an autothermal reformer for liquid hydrocarbon fuels (modeled as C 12 H 23 ) and with H 2 purification in a water–gas-shift/membrane reactor is developed to do iterative calculations for mass, species, and energy balances at a component and system level. The model evaluates system efficiency with parasitic loads (from compressors, pumps, and cooling fans), system water balance, and component operating temperatures/pressures. Model results for a 5-kW fuel cell generator show that with state-of-the-art PEM fuel cell polarization curves, thermal efficiencies >30% can be achieved when power densities are low enough for operating voltages >0.72 V per cell. Efficiency can be increased by operating the reformer at steam-to-carbon ratios as high as constraints related to stable reactor temperatures allow. Decreasing ambient temperature improves system water balance and increases efficiency through parasitic load reduction. The baseline configuration studied herein sustained water balance for ambient temperatures ≤35 °C at full power and ≤44 °C at half power with efficiencies approaching ∼27 and ∼30%, respectively.

Abstract The accuracy of Peukert's equation for battery capacity lessens under dynamic loading and varying temperature conditions. Previous attempts by others endeavor to overcome the current shortfall; however, many still neglect the inclusion of a temperature dependency. This paper investigates the feasibility of Peukert's equation for practical automotive situations and expands upon the equation in order to include both variable current and temperature effects. The method proposed captures these requirements based on the theory that a battery contains a relative maximum absolute capacity, different from manufacturer 20-h specifications, and the specific discharge conditions determine the rate at which to remove this capacity. Experimental methods presented in the paper provide an economical testing procedure capable of producing the required coefficients in order to build a high-level, yet accurate state of charge prediction model. Finally, this work utilizes automotive grade lithium-based batteries for realistic outcomes in the electrified vehicle realm.

Abstract Most electrochemical models fail to accurately simulate lithium-ion battery behaviors at high C-rates (generally above 2C) and thus limit lithium-ion battery usage in many of today's applications, including electric vehicles and hybrid electric vehicles. To address this issue, the non-uniform concentration distribution effects that occur within the electrodes at higher C-rates must be included in the electrochemical model. The essential modifications to the model must incorporate solid-phase diffusion, liquid-phase diffusion, and reaction polarization. This paper develops an electrochemical model that considers high C-rate performance and assesses the model's performance for LiCoO2 batteries with charge/discharge rates up to 4C, and LiFePO4 batteries up to 5C.

Water management is of critical importance in a proton exchange membrane (PEM) fuel cell, in particular, those based on a sulfonic acid polymer, which requires water to conduct protons. Yet there are limited in situ studies of water transfer through the membrane and no data are available for water transfer due to individual mechanisms through the membrane in an operational fuel cell. Thus it is the objective of this study to measure water transfer through the membrane due to each individual mechanism in an operational PEM fuel cell. The three different mechanisms of water transfer, i.e., electro-osmotic drag, diffusion and hydraulic permeation are isolated by specially imposed boundary conditions. Therefore water transfer through the membrane due to each mechanism is measured separately. In this study, all the data is collected in an actual assembled operational fuel cell. The experimental results show that water transfer due to hydraulic permeation, i.e. the pressure difference between the anode and cathode is at least an order of magnitude lower than those due to the other two mechanisms. The data for water transfer due to diffusion through the membrane are in good agreement with some of the ex situ data in the literature. The data for electro-osmosis show that the number of water molecules dragged per proton increases not only with temperature but also with current density, which is different from existing data in the literature. The methodology used in this study is simple and can be easily adopted for in situ water transfer measurement due to different mechanisms in other PEM fuel cells without any cell modifications.

Abstract We present an efficient water recirculation method for air-breathing direct methanol fuel cells (DMFC) by utilizing so-called electroosmotic (EO) pumps. The fuel management system includes three EO pumps for the delivery of methanol solution, pure methanol, and pure water, respectively. Water recirculates from the fuel cell back to the fuel supply stream with the aid of these pump systems. We characterized the performance of the air-breathing DMFC for 2 M and 4 M methanol solutions using a syringe pump and EO pumps, respectively. The DMFC performance is similar for both types of pumps as long as the EO pump operates with the applied voltage of 6 V or higher. The maximum net power density (fuel cell power generation minus pump power consumption) was 50 mW cm −2 for 2 M methanol solution and the applied pump potential of 8 V. The minimum parasitic power ratio (pump power consumption divided by fuel cell power generation) was merely 2.1% for 4 M methanol solution and the applied pump potential of 6 V. We successfully demonstrated that the air-breathing DMFC integrated with the EO pumps operated in a stable condition in 1-h galvanostatic measurement.

Abstract Simulation results of lithium diffusion in Li(Ni1/3Mn1/3Co1/3)O2 under galvanostatic discharge (intercalation) conditions are presented for a set of cathode active material particles imaged using X-ray nanotomography. A range of operating conditions from 0.5C to 10C are considered. Behavior of these real particles is compared to idealized spherical particles on the basis of dimensionless parameters that link particle morphology, lithium diffusion and reaction kinetics. These parameters include the particle sphericity, a mass transfer Biot number and a mass transfer Fourier number. At lower C-rates, lower Biot numbers, and higher sphericities the spherical particle model proves sufficient, and the spherical particle model predicts minimum concentrations within one standard deviation of the average behavior of the particle ensemble that is analyzed. However, at higher C-rates, higher Biot numbers, and intermediate Fourier numbers the spherical particle model significantly overestimates the state of charge in the active material particles, an error that may underestimate lithium available for plating at the anode and state of charge dependent degradation in cathode materials. The results presented highlight a need and means for more accurate assessment of the influence of microstructural geometry on battery performance under aggressive current conditions.

Abstract A mathematical model of a lithium-ion cell is used to analyze pulse and relaxation behavior in cells designed for hybrid-electric-vehicle propulsion. Predictions of cell voltage show good agreement with experimental results. Model results indicate the ohmic voltage loss in the positive electrode is the dominant contributor to cell overvoltage in the first instances of a pulse. The concentration overvoltage associated with the reduced lithium in the solid phase of the positive is of secondary importance through pulse duration, but dominates after current interruption. Effects of anisotropy in the particle diffusion coefficient are also studied. Heaviside mollification functions are utilized to describe the thermodynamic open-circuit voltage of lithiated graphite, and the “pleated-layer model” is extended to realize the phase behavior of primary-particle aggregates during cell operation. The negative electrode contributes little to the cell overvoltage, and two-phase behavior results in a reaction front within the electrode. No voltage relaxation is associated with the negative electrode, and after full relaxation, a stable composition gradient of lithium exists throughout the solid phase. Internal galvanic coupling removes the composition gradients in the positive electrode during relaxation.