You Xiao, Huang Chiung-Wei, Vinodgopal Kizhanipuram, Atkin Joanna M
Department of Applied Physical Science, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, USA.
Department of Engineering, Westlake University, Hangzhou, China.
J Chem Phys. 2024 Sep 7;161(9). doi: 10.1063/5.0222228.
Surface functionalization of low-dimensional nanomaterials offers a means to tailor their optoelectronic and chemical characteristics. However, functionalization reactions are sensitive to the inherent surface features of nanomaterials, such as defects, grain boundaries, and edges. Conventional optical characterization methods, such as Raman spectroscopy, have limited sensitivity and spatial resolution and, therefore, struggle to visualize reaction sites and chemical species. Here, we demonstrate the capability of spatially and chemically sensitive tip-enhanced Raman spectroscopy imaging to map the distribution of molecules in covalently functionalized graphene. Hyperspectral vertex component analysis and density functional theory are necessary to interpret the nature of binding sites and extract information from the spatially and spectrally heterogeneous datasets. Our results clarify the origin of heterogeneous surface functionalization, resolving preferential binding at edges and defects. This work demonstrates the potential of nanospectroscopic tools combined with unsupervised learning to characterize complex, partially ordered optoelectronic nanomaterials.
低维纳米材料的表面功能化提供了一种调整其光电和化学特性的方法。然而，功能化反应对纳米材料的固有表面特征敏感，如缺陷、晶界和边缘。传统的光学表征方法，如拉曼光谱，灵敏度和空间分辨率有限，因此难以可视化反应位点和化学物种。在这里，我们展示了空间和化学敏感的针尖增强拉曼光谱成像绘制共价功能化石墨烯中分子分布的能力。高光谱顶点成分分析和密度泛函理论对于解释结合位点的性质以及从空间和光谱异质数据集中提取信息是必要的。我们的结果阐明了异质表面功能化的起源，解析了边缘和缺陷处的优先结合。这项工作展示了纳米光谱工具与无监督学习相结合来表征复杂的、部分有序的光电纳米材料的潜力。
