Answers are grounded in this paper and cite their sources.
Electronic synapses featuring conduction modulation for synaptic weight update play a central role in implementing neuromorphic computing. Though memristors and memtransistors operating via ion exchange for conductance modulation have demonstrated synaptic functionality, their planar architectures suffer from a reduced interfacial area during lateral scaling, limiting their dynamic range of conductance modulation. Here, an electrochemical memtransistor is presented that replaces conventional planar layout with a precisely engineered 3D structure composed of an SU‐8 organic reservoir and ZnO channel, fabricated using proximity‐field nanopatterning. This design increases the effective oxygen ion‐exchange interfacial area available for conductance modulation without enlarging the device footprint. The 3D device achieves a dynamic range nearly 34‐fold larger than a planar counterpart, withstands over 10 4 updating pulses, and exhibits key biological synaptic dynamics. Deep neural network simulations based on the measured weight‐update characteristics show improved classification accuracy over the planar control, and the inherent fading memory of the device supports reservoir computing. These findings establish deterministic 3D structural engineering as a scalable strategy for decoupling device footprint from synaptic modulation capability, thereby addressing a scaling bottleneck in neuromorphic hardware.
No takes yet. Share an insight, caveat, or question.
Ryu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: