Skoltech Industry-Oriented Computational Discovery Lab
The laboratory is engaged in the computational discovery of new materials and their properties for industrial applications.
We are a part of the global AI for Materials Science movement, pushing the boundaries of what advanced algorithms and simulations can reveal about the world of real materials.
Join us at the intersection of physics, chemistry, materials science, and artificial intelligence, where computational tools and experimental insights work together to illuminate the cutting-edge of the physical world.
IOCD Lab at a Glance
Research Scope
- The use and implementation of AI-powered methods for predicting the properties of functional materials, in the broadest sense of the term
- Catalytic nanoparticles: metallic, bimetallic, high-entropy; DFT in conjunction with AI-powered techniques to construct predictive models
- High-entropy materials: MD simulations with ML potentials; AI-powered techniques for data processing
- High-temperature superconductivity under high pressure: stability; formation; properties; data analysis; ML-based predictive models
- Constructional and functional materials: MD simulations with ML potentials; analysis of degradation of properties; increasing the operating time
- Low-dimensional materials: thermal transport; electronic transport; new ML-based techniques
“We envision the future of materials science as a convergence of computational power, theoretical frameworks, and intelligent algorithms with experimental materials science”
Our Team
Projects
Publication List
Reviews
Courses
Thermodynamics of Materials
Graduate-level overview of thermodynamics applied to computational materials science.
Detailed Description
The course covers stability of materials, phase transitions, defects, surfaces, and thin films with applications in catalysis, sensing, and energy storage.
Computational Methods in Atomistic Simulations
Classical and modern computational approaches in materials science.
Detailed Description
The course spans classical transport theories to modern DFT, molecular dynamics, and machine learning methods with hands-on computational practice.