Conrard Giresse Tetsassi Feugmo

Dr. Conrard G. Tetsassi Feugmo is an Assistant Professor in the Department of Chemistry at the University of Waterloo. He holds a master's degree in Nanotechnologies from the Louvain School of Engineering and a Ph.D. in Computational Chemistry from the University of Namur, both in Belgium.His laboratory stands at the intersection of physics, machine learning, and electrochemical science. Trained in nanotechnologies and computational chemistry across Belgium's leading institutions, he brings a rare cross-disciplinary fluency to some of the most pressing challenges in energy and materials science.

His vision is bold: to replace empirical guesswork with fundamental, equation-level understanding of how materials live, degrade, and fail in the most demanding environments on earth—from the molten salt coolants of next-generation nuclear reactors to the electrochemical interfaces dictating the lifetime of tomorrow's batteries and fuel cells. To realize this vision, his group builds physics-grounded "digital twins" that think across scales—from the quantum motion of individual atoms all the way up to device-level performance—bridging worlds that have long remained disconnected.

At the heart of this enterprise is a deep conviction that machine learning should not merely fit data, but understand physics. Dr. Feugmo's group embeds governing physical laws directly into neural architectures, creating models that are interpretable, thermodynamically consistent, and capable of uncovering hidden mechanisms that neither experiment nor simulation alone could reveal. Where others see noisy electrochemical data, his team sees governing equations waiting to be discovered.

From developing neural-parameterized density functional theories that capture the intricate dance of ions at electrified interfaces, to building open-source differentiable frameworks that decode the atomic structure of disordered solids, the Feugmo lab is constructing a fully integrated, end-to-end pipeline for materials discovery—one where atoms, interfaces, and devices are all part of the same living, learnable model.

• Physics-Informed Neural Networks · Machine Learning Interatomic Potentials · Molecular Dynamics · Density Functional Theory · Automatic Differentiation

• Classical Density Functional Theory · Multiscale Modeling · Neural Functional Theory · Differentiable Optimization

• Electrochemical Interfaces · Solid Oxide Electrolysis · Molten Salt · Solid-State Batteries · Fuel Cells · High-Entropy Alloys · Nuclear Materials

• Ion Transport · Degradation Mechanisms · Corrosion · Charge Transfer · Electric Double Layer · Defect Chemistry · Phase Behavior

Dr. Feugmo's research group develops interpretable, multiscale computational frameworks that unite atomistic simulation, continuum theory, and data-driven discovery to understand and predict electrochemical and thermochemical phenomena in energy materials. His team constructs physically grounded "digital twins" of complex systems—spanning atomic, interfacial, and device length scales—to enable rational design of corrosion-resistant materials, high-temperature electrochemical devices, and ultra-long-life energy storage technologies.

A central pillar of the group's work is the integration of Physics-Informed Neural Networks (PINNs) into electrochemical modeling. By embedding governing laws—such as the Butler–Volmer, Nernst, and Nernst–Planck equations—directly into neural architectures, his team builds models capable of inferring kinetic and transport parameters from experimental signals like cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS), without relying on large labeled datasets. Complementary sparse regression and symbolic learning methods extract closed-form degradation laws from noisy data, replacing empirical assumptions with mechanistic, equation-level discovery.

The group also develops high-fidelity machine learning interatomic potentials (MLIPs) that achieve near–ab initio accuracy at a fraction of the computational cost. Through GPU-accelerated molecular dynamics simulations, they extract macroscopic transport coefficients—viscosity, thermal conductivity, diffusivity—for applications including corrosion in molten salt environments relevant to advanced nuclear reactors, ion transport in solid electrolytes, and defect-mediated kinetics in energy materials.

On the theoretical side, the group advances Classical and Dynamic Density Functional Theory (cDFT/DDFT) using neural-parameterized functionals to model inhomogeneous ionic fluids, electric double layer dynamics, and ion transport in concentrated electrolytes beyond the limits of classical Poisson–Boltzmann theory. Additionally, Dr. Feugmo's open-source framework TorchDisorder leverages automatic differentiation and constrained optimization to resolve atomic structures in amorphous and disordered solids directly from experimental scattering data—providing high-quality structural inputs for downstream ML and DFT workflows.

Across all these themes, the group applies its methods to batteries, solid oxide electrolysis cells (SOECs), fuel cells, molten salt systems, and high-entropy alloys (HEAs)—materials at the frontier of energy, nuclear, and extreme-environment technologies.

A central theme of his work is the integration of Physics-Informed Neural Networks (PINNs) and Machine learning Potential to the modeling of electrochemical interface and predictive, interpretable, and computationally efficient models for energy materials. By embedding fundamental physical laws—such as the Butler–Volmer, Nernst, and diffusion equations—into neural network architectures, his team develops Electrochemistry-Informed Neural Networks capable of directly inferring kinetic and transport parameters from experimental data. This approach allows accurate modeling of processes such as charge transfer, mass transport, and localized corrosion, providing deep insight into the electrochemical interface that dictates material lifetime and energy efficiency.

Over the next decade, Dr. Feugmo aims to establish these AI-augmented, physics-grounded models as a foundation for predictive design and lifetime assessment of advanced energy materials. His long-term vision is to unify quantum-to-continuum modeling, machine learning, and experimental data into accountable, generalizable computational tools supporting technologies critical to a sustainable future—including corrosion-resistant alloys, clean energy and molten salt reactor systems.

Vianode Canada :

In collaboration with Vianode Canada and supported by the Mitacs Accelerate program, Dr. Feugmo's group is advancing the science of fast-charging lithium-ion batteries for electric vehicles. Vianode operates the large-scale Via TWO facility in St. Thomas, Ontario—one of North America's leading producers of low-emission synthetic anode graphite—and this partnership brings together their materials expertise with the group's computational capabilities to model lithium-ion transport within graphite microstructures. By combining crystallographic data from XRD and electron microscopy with physics-informed diffusion models, the team is establishing quantitative links between atomic-scale structural features—grain boundaries, defects, lattice spacing—and macroscopic electrochemical performance, directly accelerating the development of next-generation fast-charging anode materials.

Nissan Technical Center North America

In partnership with Nissan Technical Center North America, Dr. Feugmo's group is developing physics-informed machine learning models to predict the long-term mechanical behavior of polymer interlayer materials used in solid-state battery modules for next-generation electric vehicles. As battery stacks expand and contract during charge-discharge cycles, these polymer layers undergo sustained creep that can compromise both structural integrity and electrical performance—yet validating a single candidate material currently demands six to twelve months of testing. By integrating classical viscoelastic constitutive theory with data-driven learning, the team aims to predict material performance over 10,000+ hours and cycles from just weeks of experimental data, dramatically compressing Nissan's development timeline.

• 2018, PhD, Computational Chemistry, University of Namur, Belgium

• 2011, Mphil, Nanotechnologies, Louvain School of Engineering, Belgium

• 2010, MSc, Chemistry, University of Yaounde I, Cameroon

• 2007, BSc, Chemistry, University of Yaounde I, Cameroon

• 2022, Talent Bursaries Alberta AI-week

• 2019, Center for Nonlinear Studies at Los Alamos National Laboratory travel grants

• 2018, Western’s Postdoctoral Fellowships Program

• 2014, C.G.B. (Comité de Gestion du Bulletin) - C.B.B travel grants

• 2014, Gordon Research Conferences travel grants

• 2012, Institutional PhD CERUNA grants

• 2011, Special Research Fund (FSR) Scholarship

• Member, Waterloo Artificial Intelligence Institute

• Member, Waterloo Institute for Nanotechnology

• CHEM 120 - General Chemistry 1
  • Taught in 2025
• CHEM 123 - General Chemistry 2
  • Taught in 2023, 2024
• CHEM 400 - Special Topics in Chemistry
  • Taught in 2026
• CHEM 740 - Selected Topics in Theoretical Chemistry
  • Taught in 2026
• NE 451 - Simulation Methods
  • Taught in 2023, 2024, 2025
• NE 452 - Special Topics in Nanoscale Simulations
  • Taught in 2026

* Only courses taught in the past 5 years are displayed.

• Reimer C., Saidi P., Casert C., Beeler C., Tetsassi Feugmo C.G., Whitelam S., Mansouri E., Martinez A., Beland L., and Tamblyn I., Prediction of vacancy defect diffusion paths in high entropy alloys via machine learning on molecular dynamics data, Journal of Applied Physics, Volume 127, Canada.

• Roy A., and Feugmo C.G.T., Optimizing bio-imaging with computationally designed polymer nanoparticles, Journal of Materials Chemistry B, Volume 13, 12502-12515, Canada.

• Sun Y., Chen L., Jing Z., Li Y.Y., Kim D., Gao J.-Y., Noroozi R., Yi G.Y., Tetsassi Feugmo C.G., Klinkova A., Sask K., Kristiadi A., Wang B., Gillies E.R., Lu K.P., Shi H.H., and Hu P., Generative AI for the Design of Molecules: Advances and Challenges, Journal of Chemical Information and Modeling, Volume 65, 12668-12690, Canada.

• Gariepy Z., Chen Z., Tamblyn I., Singh C.V., and Tetsassi Feugmo C.G., Automatic graph representation algorithm for heterogeneous catalysis, APL Machine Learning, Volume 1, Canada.

• Chen Z.W., Gariepy Z., Chen L., Yao X., Anand A., Liu S.-J., Tetsassi Feugmo C.G., Tamblyn I., and Singh C.V., Machine-Learning-Driven High-Entropy Alloy Catalyst Discovery to Circumvent the Scaling Relation for CO2Reduction Reaction, ACS Catalysis, Volume 12, 14864-14871, Canada, 2022.

• Chen Z.W., Gariepy Z., Chen L., Yao X., Anand A., Liu S.-J., Tetsassi Feugmo C.G., Tamblyn I., and Singh C.V., Machine-Learning-Driven High-Entropy Alloy Catalyst Discovery to Circumvent the Scaling Relation for CO2Reduction Rea, ACS Catalysis, Volume 12, 14864-14871, Canada, 2022.

• Tetsassi Feugmo C.G., Ryczko K., Anand A., Singh C.V., and Tamblyn I., Neural evolution structure generation: High entropy alloys, Journal of Chemical Physics, Volume 127, Canada, 2021.