EUVIMEDEUVIMED entdecken
Crossref

LLM-driven discovery for carbon allotropes with bond-network entropy

Yuzhou Hao, Yujie Liu, Xuejie Li, Turab Lookman, Xiangdong Ding, Jun Sun, Zhibin Gao

Applied Physics Letters · 09.03.2026 · 128 (10)

DOI: 10.1063/5.0314279 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Physics Letters, online erschienen am 09.03.2026. Verfasst von Yuzhou Hao und 6 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 58 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Key Research and Development Program of the Ministry of Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The discovery of novel carbon allotropes with tailored thermal and mechanical properties is critical for advanced thermal management. However, exploring the vast configurational space of carbon using ab initio calculations remains computationally prohibitive. Driven by the rich topological landscape of carbon, where the competition between sp, sp2, and sp3 hybridization states dictates material performance, we establish a closed-loop artificial intelligence (AI) framework to explore this complex configurational space. We introduce a hybridization entropy descriptor to guide the search beyond conventional forms. Here, we establish a closed-loop AI framework that synergizes a Large Language Model (LLM) for structural generation with a Machine Learning Potential (MLP) for accelerated evaluation. Leveraging CrystaLLM to generate candidates and an iteratively refined MLP for high-fidelity validation, we screened thousands of structures to identify several stable allotropes with exotic properties. Specifically, we report “yne-diamond C12” and “yne-hex-diamond C8,” which exhibit extreme thermal anisotropy and ultralow in-plane shear stiffness arising from their mixed sp–sp3 hybridization. Furthermore, we discovered a complex sp–sp2–sp3 hybridized C12 phase that combines metallic conductivity with an anomalous negative Poisson's ratio. Notably, we identified a superhard phase (C16_3) possessing a calculated Vickers hardness (103.3 GPa) exceeding that of diamond {96 GPa [R. A. Andrievski, Int. J. Refract. Met. Hard Mater. 19, 447–452 (2001)]}. Microscopic analysis reveals that thermal transport in these materials is governed by the interplay between rigid frameworks and flexible linkers. This work expands the known carbon phase space and demonstrates the efficacy of coupling generative AI with MLPs for the accelerated inverse design of functional materials.

Förderung

  • National Natural Science Foundation of China
  • Key Research and Development Program of the Ministry of Science and Technology

Zitieren

APA

Hao, Y., Liu, Y., Li, X., Lookman, T., Ding, X., Sun, J., & Gao, Z. (2026). LLM-driven discovery for carbon allotropes with bond-network entropy. Applied Physics Letters, 128(10). https://doi.org/10.1063/5.0314279

Vancouver

Hao Y, Liu Y, Li X, Lookman T, Ding X, Sun J, et al. LLM-driven discovery for carbon allotropes with bond-network entropy. Applied Physics Letters. 2026;128(10). doi:10.1063/5.0314279

BibTeX
@article{Hao2026,
  author = {Hao, Yuzhou and Liu, Yujie and Li, Xuejie and Lookman, Turab and Ding, Xiangdong and Sun, Jun and Gao, Zhibin},
  title = {LLM-driven discovery for carbon allotropes with bond-network entropy},
  journal = {Applied Physics Letters},
  year = {2026},
  volume = {128},
  number = {10},
  doi = {10.1063/5.0314279}
}
Kostenlos starten →