Machine learning the ab initio potential energy surface of condensed phase systems
Recent advancements in machine learning applied to molecular sciences have significantly extended the length and time scales of atomistic simulations using quantum mechanics-based potential energy surfaces. Machine learning models offer computationally efficient predictions of interatomic interactions while preserving the accuracy of the electronic structure methods on which they were trained. Our group leverages these models to simulate the real-time dynamics of condensed phase systems, offering a realistic depiction of materials at finite temperature and pressure, and under explicit solvation. This approach allows us to directly compute experimental observables, such as free energy and spectroscopy, thereby creating a direct link between atomistic simulations and experimental data. Below are some of the areas our group is actively exploring.
Physics and chemistry of water near solid surfaces
The study of solid-liquid interfaces is crucial for understanding a wide range of natural and technological processes. These interfaces play a pivotal role in phenomena such as corrosion, catalysis, adsorption, and biocompatibility. By investigating the interactions between solids and water, we can develop new materials with tailored properties, improve industrial processes, and gain insights into fundamental scientific questions. For example, understanding the behavior of water at solid surfaces is essential for developing efficient materials for hydrogen production and designing materials that resist corrosion. Additionally, studying the interactions between biological molecules and solid surfaces is crucial for advancing fields like biomaterials and drug delivery.
Direct air capture of carbon dioxide
The development of new technologies for carbon capture is imperative to mitigate the adverse effects of climate change. Carbon dioxide emissions from human activities, primarily the burning of fossil fuels, are a major contributor to global warming. Carbon capture technologies aim to capture and store carbon dioxide before it enters the atmosphere, thereby reducing greenhouse gas emissions and helping to stabilize the climate. These technologies are essential for transitioning to a low-carbon economy and ensuring a sustainable future for generations to come.
Water under extreme confinement
Understanding the fundamentals of water under extreme nanoconfinement is crucial for advancing various fields of science and technology. Water’s behavior at the nanoscale is significantly different from its bulk properties, with implications for fields such as materials science, nanotechnology, and biology. By studying the behavior of water in confined spaces, we can develop new materials with enhanced properties, design more efficient energy storage devices, and gain insights into the fundamental principles governing biological processes. Additionally, understanding the behavior of water under extreme conditions can provide valuable information for applications in desalination, water purification, and environmental remediation.
Proton and electrons transfer reactions at photocatalyst-water interfaces
Solar energy conversion, a promising alternative to fossil fuels, often involves the interaction of light with solid materials immersed in liquids. By studying the photochemical processes occurring at these interfaces, we can optimize the design and performance of solar cells, photocatalysts, and other energy-harvesting devices. This knowledge can lead to improvements in energy efficiency, cost-effectiveness, and long-term durability, ultimately contributing to a more sustainable and environmentally friendly energy future.