Origins of Nonideal Coulombic Efficiency in Magnesium Electrodeposition and Electrodissolution

Controlling the morphology of metal anodes is required for the energy storage community to push beyond the limiting energy density of Li ion battery technology. Lithium metal and its tendency to form dendrites is one example where morphological control restricts the replacement of limited energy density graphite intercalation anodes (818 Ah/L theoretical).  Increased energy density and elimination of the dendrite problem are two often argued benefits for moving to a battery chemistry based on Mg as opposed to Li (3833 vs. 2026 Ah/L theoretical). For Mg anodes to be successful, 99.9% coulombic efficiency must be achieved at relevant rates, capacities, and cycling profiles necessary for a viable transportation battery.

Structural change within the Mg metal anode during repetitive deposition and dissolution is a complicated process that is not well understood and that can impact the overall cell performance of a rechargeable Mg battery. Our results from cycling the outermost region of a well-defined, 6 micron thick Mg film in a continuous mode in a variety of Mg chloro complex forming electrolytes (i.e., Magnesium Chloro Complex, All Phenyl Complex, and several MgCl₂ modified organomagnesium complexes) demonstrates that the average coulombic efficiency is maintained at the electrolyte’s intrinsic, near 100% single cycle value for 50 to 100 cycles. In contrast, discontinuous Mg cycling where an open circuit equilibration period is inserted prior to the deposition half-cycle (see Figure 1a), produces as much as a 2% decrease in the average cycle efficiency to 98%. Such an efficiency decrease requires that a significant excess of Mg be incorporated into a cell resulting in a substantial reduction in energy density. The origin of this efficiency decrease is the result of several key phenomena that have not been fully explored within the literature. We first show that interfacial films form on the Mg surface during equilibration and these films direct further growth and dissolution of deposited metal. Films force Mg to re-nucleate onto itself resulting in discontinuous growth producing buried interfaces that range from nanoscopic discontinuity to microscale voiding between the Mg base layer and the cycled overlayer of the deposit. We demonstrate that Mg dissolution in Mg chloro complex forming electrolytes is rate limited by the formation of the complex required to solvate the Mg²⁺ cation, which leads to both local and longer range roughening of the Mg base layer. Cross-section transmission electron microscopy (TEM) imaging shows that discontinuous cycling, film formation, and dissolution roughening produce regions of net accumulation of Mg due to inhibited dissolution and regions of net loss where dissolution becomes enhanced. Over as few as 50 cycles, local variation in dissolution and deposition rates produce microns of accumulated height difference, as dimensional control of the Mg is essentially lost. More significant is that regions of enhanced dissolution tend to form a cumulative porous Mg overlayer during the deposition half cycles, as seen in the contrast between a continuous and discontinuous deposit in Figure 1b,c. We hypothesize that the observed 2% CE decrease is a result of the formation of an Mg overlayer that becomes increasingly structurally and electrically isolated from the Mg base layer, resulting in a local fraction of deposited Mg whose charge is not recovered during the final discharge of the full film. In this presntation, we discuss the impact of rate, capacity, time of equilibration, and electrolyte composition on the coulombic efficiency, The role of local mass transport as imposed by a separator will also be compared to the case of the unobstructed Mg anode, explored in half-cell experiments.  Finally, we will demonstrate and explain a successful strategy for mitigating the effects of equilibrium films formation and maintaining the electrolyte’s intrinsic coulombic efficiency.  

This work is supported as part of the Joint Center for Energy Storage Research, an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Science. Sandia is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Company, for the U.S. DOE’s NNSA under contract DE-AC04-94AL85000.