// Research

What we do

From quantum mechanics to reactor scale, VDLab bridges atomistic understanding and process engineering — resolving plasma–material interactions and turning them into predictive, AI-driven virtual process platforms for semiconductor manufacturing.

01 — Core identity

Physical AI & Digital Twin for Manufacturing

Physics-informed virtual process platforms that combine simulation data with experimental databases. Bridging the virtual and physical worlds, our Physical AI and digital twins target real-time process evaluation, predictive control, and interpretable decision-making — with ultra-precision cryogenic etching as the primary proving ground.

More detail →
Digital twin for semiconductor manufacturing (virtual–physical integrated AI)
AI-based virtual-process & physical-system integration platform (cryogenic etching)
Surrogate models for feature-scale process prediction & acceleration
AI sensor monitoring (OES / QMS → CNN + MLP, interpretability / SHAP)
LLM / Agentic AI for process automation and the machine–AI–human interface
02 — Foundational capability

Multiscale / Multiphysics Simulation

Data-driven computational frameworks bridging atomistic to continuum scales. We resolve plasma–surface interactions at the atomic level, combining machine-learning force fields with molecular dynamics to explore the theoretical limits of challenging processes.

More detail →
Quantum → atomistic → continuum → reactor multiscale modeling
Plasma–material / plasma–surface interaction (etching, plasma annealing, sputtering)
Molecular dynamics (including reactive MD), density functional theory / first-principles
Machine-learning force fields (MLFF) & neural network potentials
Feature-scale / DSMC / level-set process modeling
03 — Applied track

Computational & Data-Driven Materials & Process Modeling

First-principles and continuum simulations of semiconductor processes — ALD and CVD — and functional materials, integrating deep learning to accelerate process and materials design for next-generation semiconductor materials and selective thin-film processes.

More detail →
Atomic layer deposition (ALD) & area-selective ALD; CVD process modeling
Selective metal thin-film processes (ASD; Ru / Mo; inhibitors, 100% selectivity)
Low-temperature chalcogenide-based high-mobility semiconductor materials
Oxide semiconductor materials for next-generation DRAM
Deep-learning / Bayesian-optimization-driven materials & process design