Related Experiment Video
Updated: May 23, 2025

13:56
Probe Type II Band Alignment in One-Dimensional Van Der Waals Heterostructures Using First-Principles Calculations
Published on: October 12, 2019
7.5K
Scalable Van Der Waals Integration of III-N Devices Over 2D Materials for CMOS-Compatible Architectures
Gyuhyung Lee1, Youngtek Oh2, Junsik Hwang2
1Department of Intelligent Semiconductor, Soongsil University, Seoul, 06938, South Korea.
Advanced Materials (Deerfield Beach, Fla.)
|March 11, 2025
Summary
This study introduces scalable van der Waals (vdW) integration for beyond CMOS scaling, using fluidic-assisted self-alignment transfer (FAST) to integrate III-N devices and 2D materials with CMOS. This breakthrough enables advanced semiconductor functionalities beyond current physical limits.
Area of Science:
- Semiconductor technology
- Materials science
- Nanotechnology
Background:
- Moore's Law limitations in silicon transistor scaling.
- Emerging non-group IV materials for heterogeneous integration.
- Need for advanced integration techniques beyond CMOS.
Purpose of the Study:
- Develop scalable van der Waals (vdW) integration technology.
- Integrate III-N devices, 2D materials (graphene, MoS2), and CMOS using back-end-of-line processes.
- Demonstrate beyond CMOS scaling through novel integration methods.
Main Methods:
- Utilized fluidic-assisted self-alignment transfer (FAST) for precise integration.
- Employed vertical 3D and lateral 2D integration strategies.
- Verified vdW interface fidelity through yield, uniformity, and reliability analysis.
Main Results:
- Achieved process accuracy of ≈ 32.6 nm on a 200 mm wafer scale.
- Successfully integrated freestanding III-N chips onto 2D materials.
- Demonstrated multi-layer graphene as a functional back-gating interconnect line.
- Integrated GaN-based radio-frequency power and cascode GaN/Si transistors on SOI-CMOS.
Conclusions:
- The proposed vdW integration approach surpasses physical scaling limits.
- Facilitates functional diversification in semiconductor devices.
- Advances the
- More than Moore
- paradigm through scalable, high-fidelity integration.

