Jesse S. Ko

The development of electrochemical materials is constrained by extensive formulation spaces and the low throughput of manual screening. Here, we report a self-driving electrochemical laboratory integrating liquid-handling robotics, online electroanalysis, and Bayesian optimization to accelerate electrolyte discovery. Operating in a 96-well microtiter format, the platform automatically formulates candidate electrolytes, performs temperature-controlled electrochemical measurements, and iteratively updates surrogate models to explore high-dimensional chemical spaces. This framework was applied to the development of high-entropy aqueous electrolytes for low-temperature zinc metal batteries. The system screened hundreds of formulations, and multiobjective optimization yielded an electrolyte with a conductivity of 134.1 mS cm–1 and an electrochemical stability window of 2.50 V, with ∼2019 h of cycling stability in symmetric cells at –20 °C. This approach significantly reduces experimental labor while accelerating the development of electrochemical materials.

Closed-loop design of multi-principal element alloys using Bayesian optimization and arc melting.

Multi-principal element alloys (MPEAs) are a unique class of alloys that typically consist of three or more principal elements in near-equimolar ratios. These alloys are often sought-after due to their exceptional mechanical properties, such as high hardness, strength, and thermal stability, which make them highly desirable for various applications. However, a significant challenge in the discovery and design of MPEAs lies in the vast and complex compositional space they occupy, which is both high-dimensional and sparsely explored. Traditional methods for identifying MPEAs with desirable properties tend to rely heavily on trial-and-error experimentation, which is time-consuming and inefficient. In this work, we apply an active learning approach, PAL 2.0, utilizing Bayesian optimization to significantly accelerate the discovery of MPEAs with high hardness.[1] PAL 2.0 operates in a closed-loop framework, closely integrating physics-based Gaussian process models with experi- mental validation. Our methodology enables the model to intelligently navigate the compositional space and make informed decisions about the most promising alloys to synthesize and test. Based on recommendations made by PAL 2.0, we successfully synthesized 20 new MPEAs through a rapid arc-melting process.[2] Among these, we identified two alloys with exceptionally high Vickers hardness values of 1269 and 1263. While the original training dataset had only three MPEAs with hardness above 1000, our method recognized five additional compositions with a hardness over 1000, thereby doubling the number of very hard MPEAs. The most striking discovery is the appearance of silicon and tantalum together in the alloys, a combination not seen in any high hardness alloy within the training dataset. Our study demonstrates the power of PAL 2.0 as a fast, efficient, and scalable tool for the discovery of materials with optimal properties, significantly reducing the parameter space for promising new materials candidates. This work not only accelerates the development of high-performance MPEAs but, since PAL’s use is materials-agnostic, it offers a pathway to explore other complex, high-dimensional material spaces, paving the way for future advancements in materials science.

The development of advanced anode materials is critical for enhancing the energy density and safety of lithium-ion batteries.

In aqueous alkaline electrolyte, why use platinum in the air cathode to electrocatalyze oxygen reduction (ORR) when the 2×2 tunnel polymorph of MnO₂ (cryptomelane) is effective, more sustainable, and a lower cost catalyst? A polymorphic cousin, lamellar birnessite MnO₂, also catalyzes ORR and is the morphology attained by sacrificially oxidizing the walls of carbon nanofoam papers (CNFPs) with permanganate thereby creating an architected catalyzed air cathode. Birnessite-painted CNFP (MnOx@CNFP) brings high areal capacity for pulse power through Mn^4+/3+ redox-based capacitance, even while catalyzing ORR in primary zinc–air cells. The oxide spontaneously recovers the +4 charged state as residual molecular in the meso/macroporous electrode re-oxidizes the discharged +3 state. In powder-composite air cathodes, cryptomelane expressed as an ultraporous, covalently networked aerogel architecture sustains continuous high load demands in a primary zinc–air cell that the denser xerogel form of cryptomelane cannot. But in rechargeable zinc–air cells, manganese oxides are less effective electrocatalysts for oxygen evolution (OER), requiring high charging voltage that degrades the oxide coupled with parasitic oxidation reactions of electrolyte and carbon that limit cycle life. Formulating powder-composite air cathodes with cryptomelane aerogel and an OER electrocatalyst, such as Ni ferrite expressed as an aerogel, keeps charging voltage < 2 V while cryptomelane still dominates ORR catalysis for a higher discharge cell voltage. Versus a dendrite-suppressing zinc sponge anode, the cell using an air electrode formulated with two architected oxygen electrocatalysts delivers state-of-the-art low voltage hysteresis and power performance. The interdependent role catalytic activity and carbon stability play on cycle life point to the importance of cell metrics to rethink catalyst and electrode design. Preventing carbon corrosion, whether by replacing carbon with other electronic conduits or improving electrocatalyst activity, needs greater emphasis as a strategy for rechargeable zinc–air cells to achieve high total areal energy density.

Prelithiation of LiMn ₂ O ₄ balances the low first cycle efficiency of silicon anodes for improvements in energy density and prospective cell chemistry enabled by naturally abundant elements.