Shao Shaofeng, Yan Liangwei, Li Jiale, Zhang Yizhou, Zhang Jun, Kim Hyoun Woo, Kim Sang Sub
Institute of Advanced Materials and Flexible Electronics (IAMFE), School of Chemistry and Materials Science, Nanjing University of Information Science & Technology, Nanjing 210044, China.
College of Physics, Centre for Marine Observation and Communications, Qingdao University, Qingdao 266071, China.
ACS Sens. 2025 Mar 28;10(3):1930-1947. doi: 10.1021/acssensors.4c03085. Epub 2025 Mar 3.
Gas sensing materials based on transition metal perovskite oxides (TMPOs) have garnered extensive attention across various fields such as air quality control, environmental monitoring, healthcare, and national defense security. To overcome challenges encountered in traditional research, a deep learning framework combining natural language processing technology (Word2Vec) and crystal graph convolutional neural network (CGCNN) was adopted in this study, proposing a predictive method that incorporates a comprehensive data set consisting of 1.2 million literature abstracts and 110,000 crystal structure data entries. This method assessed the optimal combination of zinc-cobalt bimetallic ions complexed with ligands as precursors for perovskite oxides. The regulatory function of ligand concentration on the p-n transformation of zinc-cobalt oxide sensing performance was identified, and optimization strategies were provided. The Zn(II)/Co(III)/1-methyl-1-imidazole-2-carboxylic acid complex was synthesized and demonstrated exceptional sensitivity and selectivity toward volatile organic compounds (VOCs), particularly 3-hydroxy-2-butanone (3H-2B). The p-n transformation mechanism of sensing performance was deeply analyzed through the construction of the hyper-synergistic ligand interaction matrix model for n-type sensors (HSLIM-n) and the parametrized surface-ligand resonance model for p-type sensors (PSLRM-p), enhancing the fundamental understanding of the sensing material properties. Even in highly interfering environments, the functionalized perovskite oxides exhibited outstanding sensitivity and selectivity toward 3H-2B gas, with a low detection limit of 25 parts per billion (ppb). This comprehensive research approach has facilitated the construction of a transfer learning-enhanced deep learning framework, which has shown high efficiency in predicting the performance and precise design of perovskite oxides, and its effectiveness was meticulously verified through detailed experimental validation.
基于过渡金属钙钛矿氧化物（TMPOs）的气敏材料在空气质量控制、环境监测、医疗保健和国防安全等各个领域都受到了广泛关注。为了克服传统研究中遇到的挑战，本研究采用了一种将自然语言处理技术（Word2Vec）和晶体图卷积神经网络（CGCNN）相结合的深度学习框架，提出了一种预测方法，该方法纳入了一个由120万篇文献摘要和11万个晶体结构数据条目组成的综合数据集。该方法评估了与配体络合的锌钴双金属离子作为钙钛矿氧化物前驱体的最佳组合。确定了配体浓度对氧化锌钴传感性能的p-n转变的调节作用，并提供了优化策略。合成了Zn(II)/Co(III)/1-甲基-1-咪唑-2-羧酸络合物，并证明其对挥发性有机化合物（VOCs），特别是3-羟基-2-丁酮（3H-2B）具有优异的灵敏度和选择性。通过构建n型传感器的超协同配体相互作用矩阵模型（HSLIM-n）和p型传感器的参数化表面配体共振模型（PSLRM-p），深入分析了传感性能的p-n转变机制，增强了对传感材料性能的基本理解。即使在高度干扰的环境中，功能化钙钛矿氧化物对3H-2B气体也表现出出色的灵敏度和选择性，检测限低至十亿分之二十五（ppb）。这种综合研究方法促进了一个迁移学习增强的深度学习框架的构建，该框架在预测钙钛矿氧化物的性能和精确设计方面显示出高效率，并通过详细的实验验证精心验证了其有效性。
