基于化学环境自适应学习的掺杂石墨氮化碳纳米片光学带隙预测  

Prediction of optical band gap of doped graphitic Carbon nanosheets based on chemically adaptive learning environment

作　　者：陈宸 张继勇 侯佳 CHEN Chen;ZHANG Jiyong;HOU Jia(Hangzhou Dianzi University of China,Hangzhou 310037,China;Lishui Institute of Hangzhou Di‐anzi University,Lishui 323060,China)

机构地区：[1]杭州电子科技大学,杭州310037 [2]杭州电子科技大学丽水研究院,丽水323060

出　　处：《中国传媒大学学报（自然科学版）》2024年第1期74-80,共7页Journal of Communication University of China：Science and Technology

摘　　要：一直以来,从新药物的发现到最终应用的过程被认为是非常耗时且消耗资源密集的。在化学领域,经典传统方法密度泛函理论(DFT)使用非常广泛,其计算出分子的密度泛函并推导出各种性质。然而,传统的量子模拟技术既昂贵又难以探索潜在大范围的掺杂分子结构。为了降低成本并提高效率,提出了一种基于化学环境的图神经网络模型,希望能在新型材料和药物的研发上推动发展。本文探索领域聚焦于石墨氮化碳(g-C3N4)及其掺杂变体。鉴于石墨氮化碳(g-C3N4)的分子性质带隙在现实中的重要性,准确预测材料的光学带隙成为了本研究的目标。本文使用基于化学环境的图神经网络有效地捕捉了分子的复杂结构,即使同时探索具有多个变体的掺杂g-C3N4结构,它也能精确预测它们的带隙,相比于传统的图神经网络有极大的提升,提供了一种方便快捷且精确的工具。The process from the discovery of new drugs to their final application has always been considered very time-consuming and resource intensive.In the field of chemistry,the classical traditional method density functional theory(DFT)is widely used,which calculates the density functional of molecules and derives various properties.However,traditional quantum simulation techniques are both expensive and difficult to explore potential large-scale doped molecular structures.In order to reduce costs and improve efficiency,this article proposes a graph neural network model based on chemical environment,hoping to promote development in the research and development of new materials and drugs.The field explored in this article focuses on graphite nitride carbon(g-C3N4)and its doped variants.Given the importance of the molecular properties and bandgap of graphite nitride carbon(g-C3N4)in reality,accurately predicting the optical bandgap of the material has become the research objective of this paper.This article effectively captures the complex structure of molecules using a graph neural network based on chemical environment.Even when exploring doped g-C3N4 structures with multiple variants simultaneously,it can accurately predict their band gaps,which is greatly improved compared to traditional graph neural networks,providing a convenient,fast,and accurate tool.

关 键 词：图神经网络 自适应聚合器 光学带隙 石墨氮化碳化合物 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]