Protonic Nickelate Device Networks for Spatiotemporal Neuromorphic Computing

2025-12-27

AI Summary

This study addresses the challenge of replicating biologically realistic nonlinear spatiotemporal responses and dynamic spatial coupling of neural circuits in neuromorphic computing. We propose a homogeneously integrated platform based on hydrogenated NdNiO₃ nickelate—a perovskite material—enabling co-fabrication of symmetric and asymmetric protonation junction devices within a single material system. For the first time, this platform unifies nanosecond-scale short-term spatiotemporal dynamics, multilevel long-term programmable synaptic weights, and proton-mediated network-level spatial coupling, overcoming the limitation of conventional devices that model only isolated neurons or synapses. Leveraging an on-chip proton redistribution network and CMOS-compatible thin-film fabrication, the platform achieves superior accuracy over time-domain or decoupled architectures on speech digit classification and early epileptic seizure detection tasks, with energy consumption as low as 0.2 nJ per inference and nanosecond-scale operation.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Computation in biological neural circuits arises from the interplay of nonlinear temporal responses and spatially distributed dynamic network interactions. Replicating this richness in hardware has remained challenging, as most neuromorphic devices emulate only isolated neuron- or synapse-like functions. In this work, we introduce an integrated neuromorphic computing platform in which both nonlinear spatiotemporal processing and programmable memory are realized within a single perovskite nickelate material system. By engineering symmetric and asymmetric hydrogenated NdNiO3 junction devices on the same wafer, we combine ultrafast, proton-mediated transient dynamics with stable multilevel resistance states. Networks of symmetric NdNiO3 junctions exhibit emergent spatial interactions mediated by proton redistribution, while each node simultaneously provides short-term temporal memory, enabling nanoseconds scale operation with an energy cost of 0.2 nJ per input. When interfaced with asymmetric output units serving as reconfigurable long-term weights, these networks allow both feature transformation and linear classification in the same material system. Leveraging these emergent interactions, the platform enables real-time pattern recognition and achieves high accuracy in spoken-digit classification and early seizure detection, outperforming temporal-only or uncoupled architectures. These results position protonic nickelates as a compact, energy-efficient, CMOS-compatible platform that integrates processing and memory for scalable intelligent hardware.

Problem

• Develop integrated neuromorphic platform with spatiotemporal processing
• Combine ultrafast proton dynamics with stable multilevel memory states
• Enable real-time pattern recognition and classification in single material

Innovation

• Integrated perovskite nickelate platform for neuromorphic computing
• Proton-mediated ultrafast dynamics with stable multilevel resistance states
• Combined spatiotemporal processing and memory in single material system