Conference Agenda

The growing concern over global warming, which is triggered by the widespread burning of fossil fuels, has captured the attention of communities around the world. Screening ionic liquids (ILs) with low viscosity, low toxicity, and high CO₂ absorption potential through machine learning models is critical for mitigating global warming. Previous studies in the application of machine learning in this field have primarily concentrated on the development of models using various descriptors and algorithms. The evaluation of the performance of these models is usually based on metrics such as the coefficient of determination (r²) and mean squared error (MSE). In this study, we used machine learning models to screen over 50,000 ILs by representing their chemical structures with molecular fingerprints, molecular descriptors, molecular images, and molecular graphs. We then built four models using these descriptors and identified 16 ILs with low predicted viscosity, low toxicity, and high predicted CO₂ absorption performance, including 7 novel ILs. During the model development process, multiple algorithms were considered and employed. The selection of the final algorithm was based on its ability to achieve the highest r² value. To validate the model's accuracy, we compared the experimental results of 10 ILs absorbing CO₂ at room temperature and atmospheric pressure with the models’ predictions. The results show a consistent trend, with longer side chains and more stable cations lead to a greater capacity for CO₂ absorption. The success of this study demonstrates that machine learning models may have a great potential to screen high CO₂ absorption ILs.

Converting CO₂ to value-added chemicals using surplus light alkanes such as ethane is an attractive opportunity to move toward a circular carbon economy without requiring H₂ as a feedstock. Currently, the production of valuable oxygenated hydrocarbons such as alcohols, aldehydes, and acids from ethane involves either multistep, high-pressure heterogeneous catalysis processes or homogeneous catalytic reactions that entail significant product separation challenges. One-step conversion of ethane and CO₂ to oxygenates is not thermodynamically feasible under mild conditions and has not been previously achieved as a one-step process. To circumvent thermodynamic limitations, nonequilibrium plasma may be employed to overcome the activation barriers of the reaction under room temperature conditions. Furthermore, modular plasma-activated reactions are more easily adaptable to renewable electricity and small-scale CO₂ capture than large-scale thermally activated processes.

Nonthermal plasma was used to demonstrate one-step production of alcohols, aldehydes, and acids as well as C1–C5+ hydrocarbons under ambient pressure, with a maximum oxygenate selectivity of 12%. The effects of plasma power, feed gas ratio, and catalysts on activity and selectivity were investigated in an atmospheric pressure flow reactor using time-on-stream results. Isotope-labeling experiments were combined with plasma chemical kinetic modeling to reveal the reaction pathways. The reaction proceeded primarily via oxidation of activated ethane derivatives by CO₂-derived oxygen-containing species, demonstrating a mechanism that is fundamentally different from thermocatalytic alcohol synthesis. Results from this study illustrate the potential to use plasma for the direct synthesis of value-added alcohols, aldehydes, and acids from the greenhouse gas CO₂ and underutilized ethane under ambient pressure.

Vast quantities of H₂ gas will be produced, stored, and retrieved in the zero-carbon global energy future. The underground storage options for H₂ are salt caverns and porous sedimentary formations. A major concern is bacterial sulfate reduction, utilizing H₂ as the electron donor and generating H₂S gas, which is highly toxic and dangerous. This concern is fueled by a widespread perception that the mere presence of mineral sulfur would preclude a formation outright, even though only oxidized sulfur can be an electron acceptor. Predominantly, mineral sulfur occurs in the +6 oxidation state, such as in gypsum, or the -1 reduced state, such as in pyrite. We hypothesize that the sulfur minerals in the two types of geological formations are very different in oxidation state and are categorically different in terms of H₂S risk. We acquired rock samples from both rock types and used the TES beamline (8BM) at the NSLS-II synchrotron for x-ray absorption spectroscopy (XAS) and x-ray absorption near edge structure (XANES). This research is the first sulfur oxidation analysis of salt formation rocks, and it is the first comparative analysis of sulfur oxidation in salt vs sedimentary rocks. We found that salt formation rocks uniformly contain sulfate minerals, and we conclude that safe and reliable storage of H₂ in salt caverns may be jeopardized if the risks are not thoroughly characterized. In contrast, we found that sedimentary formation rocks contain mostly disulfide minerals. If H₂ is stored in these formations, the risk of H₂S gas is minimal.

Pathways for coupled carbon removal and permanent sequestration are indispensable to climate action. The most secure form of CO₂ trapping is mineralization, where CO₂ is immobilized as thermodynamically stable carbonate minerals. However, no studies to date have monetized the benefit of elevated storage security in mafic basins that support mineralization compared to traditional physical or solubility trapping in ‘non-reactive’ sedimentary basins. Here, we developed the first economic framework for valorizing the permanence of CO₂ trapping mechanisms and associated risk in geologic sequestration systems. We capture the value of mineralization as reduced costs of monitoring, measurement, and verification; insurance on meeting storage performance standards; risk aversion metrics; and avoided costs of CO₂ loss and potential remediation events, based on a stochastic range of prescribed leakage scenarios. On the upstream end, we also compare the economics of subsurface CO₂ mineralization coupled with direct air capture vs. capture from various CO₂ point sources. For the representative project (1 MtCO₂/year for 50 years) considered across the range of scenarios, the results demonstrate that CO₂ mineralization has significant cost benefits driven by enhanced storage security that are not captured in current CO₂ storage incentives. In particular, mineralization is most valuable for direct air capture technologies, where higher capture costs and available 45Q tax credits place additional incentive on the permanence of CO₂ storage and complete immobilization. This should motivate greater near-term development of CO₂ storage operations in mafic reservoirs with mineral trapping capacity to support lower-risk deployment of negative emissions technologies.

Carbon capture and storage (CCS) is essential if global warming mitigation scenarios are to be met. However, today's maturing thermochemical capture technologies have exceedingly high energy requirements and rigid form factors that restrict their versatility and limit scale. Using renewable electricity, rather than heat, as the energy input to drive CO₂ separations provides a compelling alternative to surpass these limitations. Although electrochemical technologies have been extensively developed for energy storage and CO₂ utilization processes, the potential for more expansive intersection of electrochemistry with CCS is only recently receiving growing attention, with multiple scientific proofs-of-concept and a burgeoning pipeline with numerous concepts at various stages of technology readiness. In this presentation, I describe the emerging science and research progress underlying electrochemical CCS processes and assess their current maturity and trajectory. I also highlight emerging ideas that are ripe for continued research and development, which will allow the impact of electrochemical CCS to be properly assessed in coming years.