saswata@physics.iitd.ac.in
+91-11-26591359
At DISCERE, our mission is to combine methodological innovation with application-driven materials discovery. Our research program is built upon advanced first-principles and machine-learning frameworks, deployed to discover and manipulate next-generation optoelectronic, semiconductor, and quantum materials as primary domains.
The future of electronic materials design is driven by a unifying paradigm—the Quantum Blueprint—which integrates multiple quantum degrees of freedom into a single, coherent architecture. Rather than treating optoelectronics, spintronics, and valleytronics as independent domains, this framework connects them through a central data-driven AI core, enabling coordinated discovery and control of complex quantum phenomena.
This perspective transforms device engineering from isolated functionalities into deeply interconnected, multifunctional quantum systems, where charge, spin, and momentum-space degrees of freedom are engineered simultaneously. In this architecture, artificial intelligence does not merely assist materials discovery—it serves as the organizing principle that links fundamental physics to device realization.
At the foundation of this architecture lies an accelerated discovery pipeline based on the concept of an inverted funnel of computational screening. The process begins with an enormous theoretical materials space—on the order of 10¹⁰ possible structural permutations—representing a vast landscape of unexplored compounds.
To navigate this space efficiently, machine learning and high-throughput screening are employed as the first filtering layer. Using physics-informed descriptors such as phonon stability, tolerance factors, and exfoliation energies, this stage rapidly eliminates unviable candidates. The small fraction of materials that survive this screening are then subjected to high-accuracy density functional theory (DFT) calculations and, ultimately, experimental validation.
This hierarchical approach drastically compresses the search space, enabling the identification of optimized materials spanning multiple functionalities, including ferrovalley systems such as Cr₂CSF, optoelectronic absorbers such as Mg₃NSb, and twisted antiferromagnets such as MnPSe₃. By embedding AI into this discovery loop, materials design becomes predictive, scalable, and targeted, marking a fundamental shift away from traditional trial-and-error approaches.
Once optimal materials are identified, the Quantum Blueprint provides a systematic pathway to harness their functionality by ascending through the electronic degrees of freedom. At the most fundamental level lies charge, which governs conventional optoelectronic behavior and enables light–matter interaction in photovoltaic and photonic materials.
Building on this, control over electron spin introduces a new layer of functionality, allowing information processing through magnetic and symmetry-driven effects. At the highest level, the valley degree of freedom operates in momentum space, enabling information processing with minimal scattering and dissipation.
This hierarchy represents a conceptual transition from classical electronics to quantum-functional architectures, where multiple degrees of freedom are not only accessible but can be engineered in a coordinated manner.
A central challenge in spintronics is overcoming symmetry-protected spin degeneracy, which suppresses spin polarization in many materials. In centrosymmetric antiferromagnets, combined parity–time (PT) symmetry enforces this degeneracy by pairing opposite spins at the same momentum, effectively locking spin degrees of freedom and limiting their functionality.
Our work demonstrates that this limitation can be overcome through symmetry engineering, particularly by introducing controlled geometric perturbations such as twisting layered systems. While perfectly aligned structures preserve PT symmetry and spin degeneracy, a finite twist breaks the underlying symmetry across the Brillouin zone, giving rise to momentum-dependent spin splitting even in the absence of spin–orbit coupling. This leads to the emergence of altermagnetic spin textures, where spin polarization is governed by crystal symmetry rather than relativistic effects.
More generally, this behavior can be understood within a broader framework in which external perturbations—such as twist, strain, or electric fields—systematically lift degeneracies and enable tunable spin textures. This establishes a new paradigm where geometry and symmetry serve as the primary tools for controlling spin, opening pathways toward robust, low-dissipation, and scalable spintronic devices.
Importantly, this perspective also reveals that many materials previously considered inactive for spintronics can, in fact, host hidden or emergent spin functionality when their symmetries are subtly broken. By treating symmetry not as a constraint but as a design variable, it becomes possible to access a wider class of materials and engineer spin responses with high precision. This shift—from relying on intrinsic material properties to actively designing symmetry and geometry—provides a powerful route toward realizing next-generation spintronic technologies.
At the highest level of functionality, valleytronics enables direct manipulation of the electronic structure in momentum space. The key mechanism underlying this control is the valley funnel, which arises from the interplay between an in-plane electric field and valley-specific Berry curvature.
These Berry curvatures act as highly selective sorting fields, directing carriers based on their spin and valley index. As a result, spin-up electrons are funneled into one valley (+K), while spin-down electrons are directed into the opposite valley (–K). This spatial separation leads to the generation of pure spin currents with minimal scattering, enabling highly efficient information transport.
This mechanism establishes a direct link between excited-state physics, Berry curvature engineering, and device-level functionality, positioning valleytronics as a powerful platform for next-generation low-dissipation electronics.
Our research focuses on understanding how quantum confinement, dielectric screening, and many-body interactions govern excitons and carrier dynamics in emerging materials. By combining many-body perturbation theory (GW-BSE), nonadiabatic dynamics, and experiment–theory integration, we uncover how excitonic effects evolve across dimensionality, composition, and interfaces.
A central theme of our work is the transition from weakly bound Wannier–Mott excitons to strongly bound excitons in low-dimensional systems, driven by reduced screening and confinement effects. We further demonstrate how exciton binding energy, dielectric response, and polaron formation can be simultaneously tuned in antiperovskite nitrides, enabling high carrier mobility and efficient photovoltaic absorption.
Beyond intrinsic materials, we explore exciton–plasmon coupling and recombination engineering in hybrid systems. In TMDC–metal nanostructure interfaces, we show how midgap states and localized surface plasmon resonance govern radiative and nonradiative pathways, enabling controlled enhancement or suppression of photoluminescence. Complementary studies on halide perovskites reveal how organic spacer engineering suppresses self-trapped excitons and enhances carrier delocalization, directly impacting device sensitivity.
More recently, our work addresses nonradiative recombination from a first-principles perspective, identifying the interplay between bandgap fluctuations, nonadiabatic coupling, and decoherence in determining carrier lifetimes. Overall, our goal is to develop a predictive understanding of excited-state processes linking atomistic physics to device-relevant optoelectronic performance.