About this webinar
Density-functional theory (DFT) is one of the most widely used first-principles methods for understanding and predicting the properties of materials. However, obtaining reliable quantitative predictions for complex materials remains challenging, particularly when localized electronic states, magnetism, defects, or excited-state phenomena are involved. Addressing these challenges requires not only theoretical advances but also the computational power and scalable software infrastructure provided by high-performance computing (HPC).
In this webinar, I will present a decade of theoretical and computational developments aimed at making DFT more accurate, systematic, and predictive through Hubbard-corrected functionals, including DFT+U and DFT+U+V. I will introduce methods for calculating on-site U and inter-site V Hubbard parameters self-consistently from first principles using density-functional perturbation theory (DFPT), and discuss their implementation in the open-source Quantum ESPRESSO suite.
I will show how these developments enable quantitative simulations of transition-metal oxides, battery materials, defects in perovskites, and photocatalysts, addressing properties such as structural stability, electronic band gaps, magnetic ordering, and battery voltages. I will also discuss recent extensions beyond ground-state calculations to lattice dynamics, electron-phonon coupling, and magnetic excitations.
Throughout the webinar, I will highlight the role of HPC in making these advances possible. We will explore how parallel electronic-structure calculations, iterative linear-response methods, and the efficient use of large computing resources enable systematic parameter calculations, more demanding simulations, and the study of increasingly complex materials. The webinar will illustrate how the combination of theoretical innovation, open-source scientific software, and HPC transforms computational materials science from qualitative modelling into a more predictive tool, with direct connections to experimentally measurable properties.
Who is the webinar for
This webinar is intended for:
- Computational materials scientists and condensed-matter physicists
- Researchers using DFT and first-principles methods to study materials
- HPC users interested in scientific applications and computational physics
- Research software engineers and developers of scientific computing codes
- PhD students and postdoctoral researchers working in materials science, physics, chemistry, and computational science
- HPC specialists interested in how large-scale computing enables new scientific capabilities
Key takeaways
By the end of this webinar, participants will:
- Understand the main challenges in obtaining accurate and predictive DFT simulations of complex materials.
- Learn how DFT+U and DFT+U+V improve the description of materials with localized electronic states.
- Discover how Hubbard parameters can be calculated self-consistently from first principles using density-functional perturbation theory.
- Understand how HPC enables computationally demanding linear-response calculations and systematic first-principles studies.
- See how parallel scientific software, particularly Quantum ESPRESSO, supports simulations ranging from ground-state properties to lattice dynamics and magnetic excitations.
- Explore applications to transition-metal oxides, battery materials, defects, and photocatalysts, with connections to experimental measurements.
- Gain insight into how theoretical developments and HPC infrastructure work together to make computational predictions more reliable and scientifically useful.
Speaker, moderators and helper
Dr. Iurii Timrov (Paul Scherrer Institute, Switzerland)
Dr. Iurii Timrov is a scientist at the Paul Scherrer Institute (PSI), in the Center for Scientific Computing, Theory and Data, where he works on the development of first-principles methods for computational materials science. His research focuses on density-functional theory and its extensions, particularly Hubbard-corrected functionals, density-functional perturbation theory, and time-dependent density-functional perturbation theory.
He has developed methods for calculating Hubbard parameters from first principles and contributed to their implementation in the open-source Quantum ESPRESSO suite. His work spans electronic-structure theory, lattice dynamics, electron-phonon coupling, and magnetic excitations, with applications to functional materials, including battery electrodes and transition-metal oxides.
A central theme of his research is the development of systematic and computationally efficient theoretical methods that improve the predictive power of first-principles simulations. By combining theoretical innovation, scientific software development, and high-performance computing, he aims to enable reliable simulations of increasingly complex materials and connect computational predictions with experimental observations.
Find out more about his work on his PSI page and on the Timrov Research Group website.
Moderators: Yonglei Wang and Karim Elgammal (ENCCS)
Helper: Wei Li (ENCCS)
More events & contact
Check out more upcoming events from ENCCS and our European network at https://enccs.se/events/, as well as available ENCCS lesson materials, suitable also for self-learning.
For questions regarding this webinar or general questions about ENCCS training events, please contact training@enccs.se.
Fees
Free of charge