一种面向多组分定量的通用谱图解析方法

在一个多组分系统中各组分的定量分析是非常重要的。现代光谱如拉曼光谱(RS)、傅里叶变换红外(FT-IR)光谱、紫外-可见(UV-Vis)光谱、核磁共振(NMR)、质谱(MS)等,可通过获取丰富峰信号的谱图对样品进行各角度的详尽描述。然而,由于谱图的复杂性及其解析工作的繁重,使得仅通过样品谱图来同时量化混合物中的每个组分成为很具挑战性的工作。在这项研究中,我们首次介绍了一个名为定量主成分分析(q PCA)的可靠策略,快速计算混合物中每个成分的比例,而不需要任何手动解谱。通过使用纯组分的谱图作为参考,多组分系统的谱图可以通过PCA自动分辨并解析,然后就可以使用我们的计算方法来计算每个组分的比例。计算机建模实验和RS、FT-IR、UV-Vis、NMR、MS实验都证明了这一策略胜任多组分系统的快速定量工作。

A Universal Strategy for Quantitative Decomposition of Multicomponent System Via Spectra

Quantification of each constituent within a multicomponent system is very important.Modern spectrometers such as Raman spectrometry (RS),Fourier transform infrared (FT-IR) spectroscopy,ultraviolet-visible(UV-Vis) spectroscopy,nuclear magnetic resonance (NMR),mass spectrometry (MS) help to provide a colorful view inside a sample by acquiring spectra with abounding peaks.However,it has been extremely challenging for them to simultaneously quantify each constituent in a mixture by sample spectra,due to the complexity of the spectra and the burden of attributing peaks for each constituent.In this study,we introduce an effective and reliable strategy named quantitative principal component analysis(qPCA) to calculate the proportion of each constituent within a mixture without any manual spectral inspection for the first time.By using spectra of pure constituent as references,a spectrum of multicomponent system can be automatically resolved by PCA and then proportion of each constituent can be calculated with our calculating method.Both computer modeling and RS,FT-IR,UV-Vis,1H-NMR,ESI-MS experiments have demonstrated this strategy suitable to accomplish quantification task for multicomponent systems.