Fusing generative AI and ML-FFs for inverse design and high-fidelity dynamic simulation of fused-ring aromatic hydrocarbons
- 1National Technology Innovation Center for Digital Construction, Huazhong University of Science and Technology, Wuhan 430074, China
- 2School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
Abstract
Rational design of fused-ring aromatic hydrocarbons is hindered by vast chemical space, costly quantum simulations, and a disconnect between generation and dynamic validation. Here, we report an integrated computational platform that unifies physics-constrained generative AI, interpretable multi-objective screening, and high-fidelity machine-learning force fields (ML-FFs) to enable inverse design and dynamic simulation of fused-ring aromatic hydrocarbons. A physical consistency module incorporating the Brown-Ladner method and iterative molecular weight inversion enforces chemical and structural feasibility from the outset, generating diversity, novelty, and theoretically synthesizable structures validated against experimental benchmarks, while XGBoost predictors accurately forecast key electronic properties. We further developed DeepAsphalt-26, ML-FFs specifically optimized for aromatic systems, which attains density-functional-theory-level accuracy while accelerating molecular dynamics by two orders of magnitude, accurately reproducing π-π stacking dynamics and free energy landscapes. By bridging generative design and high-fidelity simulation, this workflow offers an end-to-end in silico route to engineering carbonaceous materials with targeted properties.
Related
- Oxidative Nucleophilic Aromatic ipso-Substitution for the Late-Stage Functionalization of Polycyclic Aromatic Hydrocarbons with Five-Membered Rings
- Extensive TD-DFT investigation of the absorption wavelength in single and multilayer polycyclic aromatic hydrocarbons/graphene quantum carbon dots
- Mechanochemical Birch Reductive Alkylation of Polycyclic Aromatic Hydrocarbons
- Cyclization of a 1,8‐Diphenylanthracene by Scholl Reactions to Form Five‐ and Seven‐Membered Ring Embedded Polycyclic Aromatic Hydrocarbons
- Synthesis of silylated polycyclic aromatic hydrocarbons by lithium-mediated mechanochemical Birch reductive silylation† † Dedicated to the memory of Dr Kazuhiko Sato at the National Institute of Advanced Industrial Science and Technology (AIST), Japan, in recognition of his outstanding contributions to silicon chemistry and catalysis, and in mourning of his untimely passing.
BibTeX
@article{long2026fusing,
title = {{Fusing generative AI and ML-FFs for inverse design and high-fidelity dynamic simulation of fused-ring aromatic hydrocarbons}},
author = {Zhengwu Long and Lingyun You},
journal = {Cell Reports Physical Science},
year = {2026},
month = jul,
doi = {10.1016/j.xcrp.2026.103452},
url = {https://doi.org/10.1016/j.xcrp.2026.103452},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Fusing generative AI and ML-FFs for inverse design and high-fidelity dynamic simulation of fused-ring aromatic hydrocarbons
AU - Zhengwu Long
AU - Lingyun You
PY - 2026
DA - 2026/07/23
JO - Cell Reports Physical Science
DO - 10.1016/j.xcrp.2026.103452
UR - https://doi.org/10.1016/j.xcrp.2026.103452
AB - Rational design of fused-ring aromatic hydrocarbons is hindered by vast chemical space, costly quantum simulations, and a disconnect between generation and dynamic validation. Here, we report an integrated computational platform that unifies physics-constrained generative AI, interpretable multi-objective screening, and high-fidelity machine-learning force fields (ML-FFs) to enable inverse design and dynamic simulation of fused-ring aromatic hydrocarbons. A physical consistency module incorporating the Brown-Ladner method and iterative molecular weight inversion enforces chemical and structural feasibility from the outset, generating diversity, novelty, and theoretically synthesizable structures validated against experimental benchmarks, while XGBoost predictors accurately forecast key electronic properties. We further developed DeepAsphalt-26, ML-FFs specifically optimized for aromatic systems, which attains density-functional-theory-level accuracy while accelerating molecular dynamics by two orders of magnitude, accurately reproducing π-π stacking dynamics and free energy landscapes. By bridging generative design and high-fidelity simulation, this workflow offers an end-to-end in silico route to engineering carbonaceous materials with targeted properties.
SN - 2666-3864
ER -