MAVENs

Materials Advancing a Viable ENergy future

Governing Quantum Materials

When does local electronic structure decide a material’s function, and when does something larger take over? We track how symmetry breaking — a vacancy, a Janus stacking, a substitutional site — propagates from local descriptor to collective spin-dependent response.
doi: 10.1016/j.jssc.2024.124602
doi: 10.1063/5.0238199
doi: 10.1016/j.jpcs.2024.111914

Energy Materials by Design

Which features of the electronic structure actually decide whether a material solves an energy problem? We isolate electronic-structure control variables — d-band filling, site-resolved d-band centre, Fermi-level position — to make compositional design predictive rather than exploratory.
doi: 10.1016/j.physb.2025.417148

You may say we are dreamers
— John Lennon, Imagine (slightly modified)

MAVENs (Materials Advancing a Viable ENergy Future), led by Dr. Rudra Banerjee, is based in the Department of Physics and Nanotechnology at the SRM Institute of Science and Technology.

Bloch’s theorem rewards a perfect crystal with clean, classifiable states — and almost no real material obliges. Substitution, vacancies, and chemical mixing are not blemishes on an ideal lattice; they are what the material actually is. Disorder is intrinsic, and it is often the thing that decides whether a material works.

Predicting how disorder controls functional behaviour — from first principles — is the central problem of our research.

We combine DFT, Monte Carlo, spin-dynamics, and machine learning — using each where the physics demands it: DFT for electronic structure, Monte Carlo and spin dynamics for finite-temperature collective behaviour, ML where compositional spaces become too large for direct first-principles study.

Research Themes

Disorder and Magnetism How does chemical disorder decide whether a magnet orders, at what temperature, and how much magnetisation survives?

Disorder and Coherence What lets a chemically generated spin centre hold quantum phase information, and how does the disorder around it set the limit?

Disorder and Catalysis In a compositionally complex catalyst no two sites are alike — so what sets the distribution of adsorption energies, and where it sits?

Machine Learning Driven Discovery Machine-learning discovery that keeps the physical insight which makes a prediction transferable.

Explore Our Research

Research Highlights

May 2026 Mensis Mirabilis of MAVENs

May 2026 New Paper in PRB: Half-metallicity in 2D TMDCs

May 2026 New Paper in PRM: Machine Learning for Quantum Defect Hosts

May 2026 IonVis Wins QtHack

Nov 2025 Karunakaran defended his thesis

Open Positions

We welcome M.Sc. and PhD students in physics, chemistry, or materials science. Projects span ab-initio DFT, Monte Carlo, and ML-driven screening.

See Open Projects