Inkjet‐Printed h‐BN Memristors with a Record Endurance under Low‐Currents
ABSTRACT Memristors exhibiting both volatile and non‐volatile resistive switching (RS) behaviors hold significant potential for their utilization as artificial neurons and synapses in neuromorphic computing systems, respectively. Inkjet printing presents an economical, low‐material‐waste, and compatible process for the fabrication of solution‐based memristors, which are essential for flexible, wearable and biocompatible applications. However, inkjet‐printed memristors typically exhibit low endurance (below 10 4 cycles) and high switching currents (above 100 µA), both of which represent critical limitations for the practical implementation and low‐power operation of neuromorphic computing systems. In this work, a hexagonal boron nitride (h‐BN) based memristor formulated via liquid phase exfoliation (LPE) and fabricated with inkjet printing demonstrates exceptional electrical performance with a remarkable endurance of 2.2 million cycles at very low currents (below 1 µA) in a volatile regime under a current‐driven approach. This result surpasses the highest endurance reported to date on inkjet‐printed‐based memristors, surpassing the threshold of one million cycles. At higher currents (above 10 µA), the same devices exhibit stable non‐volatile RS with an endurance of 70 000 cycles. This study not only demonstrates the remarkable performance of the designed memristors, but also opens new avenues for current‐driven memristive electrical characterization and neuromorphic applications.
Authors
- Giovanni Vescio (ORCID: https://orcid.org/0000-0002-2418-249X)
- A. Cirera (ORCID: https://orcid.org/0000-0002-2075-1536)
- S. Hernández (ORCID: https://orcid.org/0000-0002-2226-1020)
- F. Palacio (ORCID: https://orcid.org/0000-0002-2210-1801)
- Juan Castillo
- Pere Aran Vila
- Blas Garrido
Institutions
- University of Nariño (CO)
- Universitat de Barcelona (ES)
- Universitat Politècnica de Catalunya (ES)
Publication Details
- Journal
- Advanced Electronic Materials
- Published
- 2026-09-04
- DOI
- https://doi.org/10.1002/aelm.70557
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00