Revealing the physics of switching mechanism in 2D materials based memristor devices (2DMEM)

Projektinformationen

2DMEM

ID Finanzhilfevereinbarung: 101109772

DOI

10.3030/101109772

Projekt abgeschlossen

EK-Unterschriftsdatum 14 Juli 2023

Startdatum 1 Oktober 2023

Enddatum 31 März 2026

Finanziert unter

• Marie Skłodowska-Curie Actions (MSCA)

Gesamtkosten

Keine Daten

EU-Beitrag

€ 199 694,40

Koordiniert durch

UNIVERSITY COLLEGE CORK - NATIONAL UNIVERSITY OF IRELAND, CORK
Ireland

Periodic Reporting for period 1 - 2DMEM (Revealing the physics of switching mechanism in 2D materials based memristor devices (2DMEM))

Berichtszeitraum: 2023-10-01 bis 2026-03-31

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

The rapid expansion of artificial intelligence, data-intensive technologies, and edge computing is creating an urgent demand for energy-efficient and scalable hardware. Conventional semiconductor memory technologies are approaching fundamental limits, forming a critical bottleneck for future computing systems. Two-dimensional material (2DM)-based memristors offer a promising alternative, enabling ultra-low-power operation, high integration density, and neuromorphic functionality. However, their deployment is limited by a lack of predictive understanding of the atomic-scale mechanisms governing resistive switching.

The 2DMEM project addresses this challenge by developing a physics-based framework for resistive switching in transition-metal dichalcogenide (TMD) memristors under realistic conditions. It focuses on the roles of atomic defects, ion migration, and structural effects, which critically influence device performance and reliability. The project combines atomistic modelling, first-principles simulations, and quantum transport methods to move beyond simplified descriptions toward predictive modelling. It develops experimentally grounded atomic models, quantifies electric-field-driven migration processes, and determines how strain and device geometry affect stability and functionality. By linking atomic-scale processes to measurable device characteristics, the project bridges the gap between materials physics and device engineering.

This approach enables the rational design of reliable, energy-efficient memristor technologies. The results provide strategies to reduce variability, improve reliability, and optimize switching performance, supporting industrial adoption. In the medium to long term, these advances contribute to low-power computing architectures capable of addressing the growing energy demands of digital infrastructure. The project aligns with European priorities in semiconductor innovation, advanced materials, and energy-efficient digital technologies, strengthening technological competitiveness.

Overall, 2DMEM represents a shift from descriptive to predictive modelling of resistive switching, laying the foundation for future advances in memory devices and next-generation computing systems.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

During the reporting period, the 2DMEM project systematically investigated resistive switching in 2DM-based memristors using first-principles modelling, atomistic simulations, and quantum transport calculations. The work focused on understanding how defects, ion migration, and structural effects govern device behavior in TMD systems, and was structured around three core objectives.

RO1: Realistic atomic-scale models of MoS2-based memristors were developed using experimentally measured defect and impurity concentrations, including sulfur vacancies (~2 × 10¹² cm⁻²). Both monolayer and multilayer structures were optimized using density functional theory (DFT), while electrode effects were captured through representative gold configurations. These validated models provided a reliable platform for studying switching and transport.

RO2: Dominant diffusion processes were identified by calculating migration pathways and energy barriers for vacancies and metal ions using nudged elastic band methods. Simulations under equilibrium and applied electric fields revealed how defect mobility evolves under bias and clarified mechanisms driving conductive filament formation. Quantum transport simulations of high- and low-resistance states established a direct link between atomic processes and electrical behavior.

RO3: The impact of structure-induced local strain was analyzed, particularly in crossbar geometries where the 2DM layer bends over electrodes. Results showed that strain significantly modifies electronic structure and defect mobility, influencing switching stability. While key effects were demonstrated and integrated into transport analysis, this objective remains partially completed.

Main outcomes: The project established an experimentally grounded, predictive framework for resistive switching in 2DM memristors by quantifying diffusion mechanisms and identifying the interplay between defects, strain, and transport. By linking atomic-scale processes to device performance, it provides a foundation for designing reliable, energy-efficient memristor technologies.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The 2DMEM project has advanced the state of the art in 2D-material-based memristors by moving from simplified descriptions of resistive switching to a predictive, physics-based understanding grounded in atomistic modeling.

Key results:
• Developed experimentally validated atomic-scale models of TMD memristors, incorporating realistic defect densities and impurities.
• Quantified dominant diffusion mechanisms, including sulfur vacancies and metal ions, under applied fields, determining migration energy barriers accurately.
• Demonstrated that resistive switching arises from combined effects of defect dynamics, electric fields, and local structural factors.
• Identified local strain as a key factor influencing defect mobility, conductive filament formation, and switching stability.
• Integrated first-principles simulations with quantum transport modelling, linking atomic-scale processes directly to device-level electrical behavior.

Advancement beyond the state of the art: The project establishes a comprehensive framework capturing interactions between defects, ion migration, electric fields, and strain, addressing limitations of existing models. It provides measurable parameters for device design and simulation, bridging fundamental materials physics and practical device operation.

Potential impacts: Scientifically, the results offer a reliable framework for studying resistive switching in 2DM systems. Technologically, they enable more energy-efficient and reliable memristors, supporting neuromorphic computing and next-generation memory. Industrially, findings are directly relevant to semiconductor R&D and advanced memory integration. At the European level, the project strengthens capabilities in nanoelectronics, materials, and semiconductor innovation. Impacts are medium-term at the device level, guiding design and optimization, and long-term at the system level, supporting energy-efficient computing and AI hardware.

Key needs for uptake and success: Further research should extend the framework to more materials and device architectures and validate predictions experimentally. Demonstrating mechanisms in prototypes will confirm performance improvements. Integrating models into industrial simulation tools and engaging with industry partners will support commercialization. Continued international collaboration and alignment with European semiconductor initiatives will enhance funding, infrastructure access, and innovation pathways.