Direct Construction of Peaks from Free Induction Decay Curves for Gas Chromatography-Molecular Rotational Resonance Spectroscopy without Fourier Transforms.
Direct Construction of Peaks from Free Induction Decay Curves for Gas Chromatography-Molecular Rotational Resonance Spectroscopy without Fourier Transforms.
复制标题
无需傅立叶变换即可从气相色谱-分子旋转共振光谱的自由感应衰减曲线直接构建峰。
DOI：
10.1021/acs.analchem.2c02535
复制
发表时间：
2022
影响因子：
7.4
通讯作者：
Armstrong,DanielW
中科院分区：
文献类型：
--
作者：
Wahab,MFarooq;Neill,JustinL;Armstrong,DanielW
2020年引入了气相色谱与分子旋转共振光谱(GC-MRR)联用的概念，结合了GC的分离能力和MRR无与伦比的专属性。在这项研究中，我们解决了MRR光谱仪高数据吞吐量的挑战，因为GC-MRR光谱仪每秒可以产生数千到数百万个数据点。在以前的GC-MRR研究中，对自由感应衰变(FID)测量进行傅立叶变换，以生成色谱图上的每个点。这种大规模的计算限制了GC-MRR的性能、灵敏度和速度。本文提出了一种利用Gram-Schmidt矢量正交化方法从FID提取峰值强度的直接方法。首先，不含分析物的FID被用来构建代表仪器背景噪声的基集，然后剩余的FID被正交化到该固定基集。在Gram-Schmidt正交化之后，每个FID产生一个单独的强度值。正交化分析物FID的大小是色谱图中绘制的信号强度。这种方法在计算上比传统的傅里叶变换算法快得多(高达10倍)，至少和FT算法一样灵敏，并且保持或改善了色谱峰的形状。我们使用合成的FID和仪器数据，比较了傅立叶变换和格拉姆-施密特方法的灵敏度、线性和色谱峰形状。这种方法将允许基本上实时地显示总和的峰强度，然后可以进一步调查识别的峰，以识别和量化与每个峰相关联的物种。
The concept of coupling gas chromatography with molecular rotational resonance spectroscopy (GC-MRR) was introduced in 2020, combining the separation capabilities of GC with the unparalleled specificity of MRR. In this study, we address the challenge of the high data throughput of MRR spectrometers, as GC-MRR spectrometers can generate thousands to millions of data points per second. In the previous GC-MRR studies, a free induction decay (FID) measurement was Fourier transformed to generate each point on the chromatogram. Such extensive calculations limit the performance, sensitivity, and speed of GC-MRR. A direct approach is proposed here to extract peak intensity from FID using the Gram–Schmidt vector orthogonalization method. First, analyte-free FIDs are used to construct a basis set representing the instrument’s background noise, and then the remaining FIDs are orthogonalized to this fixed basis set. Each FID yields a single intensity value after Gram–Schmidt orthogonalization. The magnitude of the orthogonalized analyte FID is the signal intensity plotted in the chromatogram. This approach is computationally much faster (up to 10 times) than the conventional Fourier transform algorithm, is at least as sensitive as the FT algorithm, and maintains or improves the chromatographic peak shape. We compare the sensitivity, linearity, and chromatographic peak shapes for the Fourier transform and Gram–Schmidt approaches using both synthetically generated FIDs and instrumental data. This approach would allow the summed peak intensity to be displayed essentially in real-time, following which identified peaks can be further investigated to identify and quantify the species associated with each.