For decades, scientists have pursued two ambitious goals: moving beyond silicon by building electronics from molecules, and creating neuromorphic hardware that learns the way the brain does. A new study from the Indian Institute of Science (IISc) suggests these challenges may converge in a single solution.
Researchers at IISc have developed molecular devices that can be tuned to perform multiple functions, including memory storage, logical operations, signal selection, analog processing, and synapse-like learning. Unlike conventional systems that imitate intelligence through rigid architectures, these devices rely on the natural, complex interactions within molecular networks.
Supported by a robust theoretical framework grounded in quantum chemistry and many-body physics, the team can predict device behavior directly from molecular structure. This work demonstrates how chemical design can become a foundation for computation itself, paving the way for compact, energy-efficient neuromorphic hardware where learning and processing are embedded directly in the material.