3D integration of 2D electronics for AI hardware

Abstract The relentless demand for faster, smaller and more energy‐efficient integrated circuits has pushed the semiconductor industry to the performance, power and area (PPA) wall, where interconnect delay and energy dissipation dominate system efficiency. While 2.5D and TSV‐based 3D integration offer partial relief, they suffer from long interconnects, thermal hotspots and reliability concerns. Monolithic 3D integration (M3DI) provides a transformative TSV‐free paradigm, enabling nanoscale vertical interconnects between front‐end‐of‐line logic and back‐end‐of‐line memory tiers, thereby maximizing PPA efficiency and reducing latency. Here we demonstrate M3DI of two‐dimensional (2D) electronics, integrating single‐crystalline MoS 2 field‐effect transistors with nanocrystalline WS 2 memristors in a vertically stacked 1T1R architecture. These atomically thin devices are BEOL‐compatible and scalable, offering uniform threshold and Ion/Ioff distributions for logic, and stable filamentary switching with narrow SET/RESET variability for memory. The resulting arrays achieve ultra‐dense integration, reproducible operation and low‐latency performance within a compact footprint. Beyond binary switching, the M3DI‐1T1R chip exhibits synaptic functionalities including multistate non‐volatile memory, spike‐timing‐dependent plasticity and array‐level uniformity, essential for neuromorphic and AI hardware. This work establishes 2D‐material‐based M3DI as a compelling foundation for next‐generation intelligent and energy‐efficient computing systems beyond CMOS scaling and von Neumann architectures. image

Authors

Institutions

Publication Details

Journal
InfoMat
Published
2026-09-18
DOI
https://doi.org/10.1002/inf2.70191
Primary Topic
2D Materials and Applications
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

3D integration of 2D electronics for AI hardware

Hyunho Seok, Taesung Kim, Jin-Hyoung Lee, Dongho Lee et al.
InfoMat
2D Materials and Applications
article

3D integration of 2D electronics for AI hardware

Hyunho Seok, Taesung Kim, Jin-Hyoung Lee, Dongho Lee, Seowoo Son, Geonwook Kim, Gunhyoung Kim, Sihoon Son, Hyunbin Choi
article en

Abstract

Abstract The relentless demand for faster, smaller and more energy‐efficient integrated circuits has pushed the semiconductor industry to the performance, power and area (PPA) wall, where interconnect delay and energy dissipation dominate system efficiency. While 2.5D and TSV‐based 3D integration offer partial relief, they suffer from long interconnects, thermal hotspots and reliability concerns. Monolithic 3D integration (M3DI) provides a transformative TSV‐free paradigm, enabling nanoscale vertical interconnects between front‐end‐of‐line logic and back‐end‐of‐line memory tiers, thereby maximizing PPA efficiency and reducing latency. Here we demonstrate M3DI of two‐dimensional (2D) electronics, integrating single‐crystalline MoS 2 field‐effect transistors with nanocrystalline WS 2 memristors in a vertically stacked 1T1R architecture. These atomically thin devices are BEOL‐compatible and scalable, offering uniform threshold and Ion/Ioff distributions for logic, and stable filamentary switching with narrow SET/RESET variability for memory. The resulting arrays achieve ultra‐dense integration, reproducible operation and low‐latency performance within a compact footprint. Beyond binary switching, the M3DI‐1T1R chip exhibits synaptic functionalities including multistate non‐volatile memory, spike‐timing‐dependent plasticity and array‐level uniformity, essential for neuromorphic and AI hardware. This work establishes 2D‐material‐based M3DI as a compelling foundation for next‐generation intelligent and energy‐efficient computing systems beyond CMOS scaling and von Neumann architectures. image

InfoMat
Massachusetts Institute of Technology (US), Sungkyunkwan University (KR)
Korea Institute for Advancement of Technology
Affordable and clean energy
Openalex Percentile: Top 24%
2D Materials and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.