UC Berkeley

A major challenge facing larger-scale and more widespread use of electric vehicles is the practical energy density of current battery technology. While significant improvements in lithium-ion technology have been achieved over the past few decades, a dramatic increase in battery capacities and decrease in cost will be required to increase market penetration for electric vehicles. This dissertation focuses on deepening the fundamental understanding of two systems, the sodium-oxygen battery and the magnesium metal electrochemistry, which are highly relevant in potential low-cost, high-energy-density battery applications.

Metal-air, or metal-O2 batteries, have been intensely studied in recent decades due to their high theoretical energy densities. Of these, the nonaqueous sodium-oxygen (Na-O2) battery offers improved stability, higher full-cycle efficiencies, and higher reversible capacities on both discharge and charge than similar nonaqueous metal-O2 technologies. However, the Na- O2 battery is afflicted by a “sudden death” during discharge at a capacity significantly lower than that predicted from complete conversion of the active materials. This sudden death effectively limits the achievable battery capacity on discharge, and has been previously linked to the electrochemistry occurring at the Na- O2 cathode, where the sodium superoxide (NaO2) discharge product is formed. I studied the dependence of this sudden death on the discharge current density under constant-pressure conditions, and found that at high current densities, the maximum capacity on discharge was limited by passivation of the cathode surface by insulating NaO2 films. The capacity on discharge could be enhanced by decreasing the current density, and at low current densities the capacity was limited by pore clogging by large NaO2 crystals.

I further examined the dependence of the sudden death on other operating and design parameters of the cell, and in particular explored the influence of O2 pressure on the maximum discharge capacity. I observed that, at a given current density, there exists a transition between the mechanisms of sudden death with O2 pressure, as a result of phenomena related to the deposition of the NaO2 discharge product. Cells operated at low O2 pressures were more susceptible to failure due to surface passivation by thin NaO2 films; increasing the O2 pressure at the same current density caused an increase in capacity and a transition to failure due to pore clogging from NaO2 crystal deposition. I correlated the transition between failure mechanisms with the spatial deposition of NaO2 through the cathode, and associated it with a combination of electron and mass transfer effects.

The Na-O2 battery is also subject to a sudden death during charge, which typically occurs at a capacity lower than the prior discharge capacity of the cell. I observed that the discharge and charge current densities both influenced the attainable charge capacity prior to sudden death. These variables were associated with changes in the deposition and oxidation of the NaO2 discharge product. I proposed a charge mechanism consistent with my data, where a concerted surface oxidation mechanism and dissolution-oxidation mechanism contributed to the observed potentials. Sudden death on charge resulted when these two pathways could not support the applied current rate. Informed with this understanding of the Na-O2 capacity limitations on charge, I explored the utility of both redox mediation and modified charging schemes in preventing sudden death and enhancing the achievable charge capacity. While I found the introduction of redox mediators typically did not result in a significant enhancement in cell performance, the use of a combined constant-current/constant-potential charging scheme led to improved charge capacity, reversibility, and overall performance.

A separate metal-O2 chemistry, magnesium-O2, represents one of several hypothetical magnesium (Mg) batteries. Mg metal offers numerous advantages over other related battery materials, including high abundance, light weight, low toxicity, and ease of safe handling. Both rechargeable and non-rechargeable Mg batteries are of interest for energy storage applications. However, a particular challenge affecting aqueous Mg batteries is the corrosion of Mg in the presence of water, which decreases the battery's efficiency. Unusually, this corrosion is exacerbated during oxidative polarization of the Mg metal, such as during discharge of a Mg-O2 battery, where increasing currents result in increasing hydrogen (H2) evolution. This phenomenon is referred to as the negative difference effect (NDE), and has been the subject of research for more than a century, with no consensus on its precise mechanisms and cause.

To investigate the NDE, I designed an electrochemical cell that enabled quantitative study of the NDE under various pH and electrolyte compositions. I quantified H2 evolution due to the NDE in high pH conditions, and observed that in electrolytes of a sufficiently low bulk pH, the NDE disappears. This observation has significant implications with respect to the possible mechanism of the NDE.