Professor Alexander V. Mironenko and his team at the University of Illinois Urbana-Champaign have developed a new theoretical framework that may change how engineers predict the energetics of chemical reactions. Their work focuses on reducing the cost and complexity of quantum mechanical calculations without lowering accuracy, an issue that has long limited the routine use of high-level computational tools in chemical engineering. The model is grounded in density functional theory but redefines the reference state from independent electrons to independent atoms. This shift allows the core equations to be expressed in a simpler way while still preserving the physics needed to describe bond breaking and formation.
Mironenko, A. v., Leung, L., & Zhuang, J. (2025). Self-consistent equations for nonempirical tight-binding theory. The Journal of Chemical Physics, 163(16). https://doi.org/10.1063/5.0276043
In current quantum chemistry, the challenge lies in tracking interactions between many electrons. Mironenko has described this difficulty using an analogy: trying to follow the movement of particles in a bag of crushed candy. When everything moves at once, it becomes almost impossible to track individual trajectories. The new method instead treats atoms as the fundamental units, similar to watching full pieces of candy rather than powder. This reduces the complexity of the calculations while retaining the essential information required to model chemical processes.
Professor Alexander V. Mironenko and his team at the University of Illinois Urbana-Champaign stated,
“This is career-defining work. If each subsequent developmental step proves as successful as our initial efforts, we may be on the verge of a revolution in quantum mechanical calculations.”
The research team tested the framework using oxygen, nitrogen, fluorine and other commonly studied molecules. These systems are frequently used as benchmarks in computational chemistry because they highlight where models often fail, especially when atoms are far apart. The results from the new approach matched the predictions of established high-accuracy quantum methods and, in certain cases, performed better. The improvements were especially visible in scenarios involving stretched bonds, a region where many approximate quantum methods struggle.
This work builds on earlier research from Mironenko’s group, including a 2023 study on hydrogen clusters. That study identified limitations in standard tight-binding approaches and helped set the stage for a more general and physically grounded alternative. The new framework extends beyond simple systems to molecules more relevant to chemical engineering applications. Reactions involving plastics, fuels and dyes all depend on the accurate prediction of bond energetics, and reliable tools for modelling these systems are important for catalyst design and reaction optimization.
A key motivation behind the project is the growing need for predictive methods that do not rely on large numbers of empirical parameters. Many modern computational techniques, including neural-network models, have grown in popularity but often require extensive training datasets generated from high-level quantum calculations. Mironenko has pointed out that while these models can be powerful, they are not derived from quantum mechanical equations and may lose predictive accuracy when applied to systems outside the range of their training data. By contrast, the independent-atom approach retains a stronger connection to the underlying physics.
The researchers emphasize that this development is an early step. The current formulation addresses the foundational equations needed for a nonempirical tight-binding theory. Future studies will explore how well the method performs for larger molecules, catalytic surfaces, and condensed-phase environments, all of which present additional challenges for existing models. Nonetheless, the early results suggest that the method may offer a more practical route to performing accurate quantum mechanical calculations in settings where computational cost has been a barrier.
If the approach continues to show strong performance across more complex systems, it may become a useful tool for chemical engineers who rely on computational predictions to guide experimental work. Reaction mechanisms, catalyst screening, and materials design are areas where accurate and affordable models can accelerate progress. Mironenko has described the research as career-defining, noting that each successful step brings the field closer to a more efficient and physically grounded alternative to current quantum methods.
Adrian graduated with a Masters Degree (1st Class Honours) in Chemical Engineering from Chester University along with Harris. His master’s research aimed to develop a standardadised clean water oxygenation transfer procedure to test bubble diffusers that are currently used in the wastewater industry commercial market. He has also undergone placments in both US and China primarely focused within the R&D department and is an associate member of the Institute of Chemical Engineers (IChemE).
