Department of Chemistry and Biochemistry, The University of Texas at Arlington, Arlington, TX, 76019, USA.
Anal Bioanal Chem. 2020 Mar;412(8):1925-1932. doi: 10.1007/s00216-020-02444-8. Epub 2020 Jan 28.
A primary focus in liquid chromatography analysis of complex samples is high peak capacity separations. Using advanced instrumentation and optimal small, high-efficiency columns, complex multicomponent mixtures can now be analyzed in relatively short times. Despite these advances, chromatographic peak overlap is still observed. Recently, attention has shifted from improvements in chromatographic efficiency and selectivity to enhancing data processing after collection. Curve fitting methods can be used to trace underlying peaks, but do not directly enhance chromatographic resolution. Methods based on the properties of derivatives and power transform were recently shown to enhance chromatographic peak resolution while maintaining critical peak information (peak areas and retention times). These protocols have been extensively investigated for their fundamental properties, advantages, and limitations, but they have not been evaluated with complex chromatograms. Herein, we evaluate the use of deconvolution via Fourier transform (FT), even-derivative peak sharpening, and power law with the fast separation (< 60 s) of a 101-component mixture using ultra-high-pressure liquid chromatography. High noise and peak overlap are present in this gradient separation, which is representative of fast chromatography. Chromatographic resolution enhancement is demonstrated and described. Further, accurate quantitation is maintained and shown with representative examples. Enhancements in peak capacity and peak-to-peak resolutions are discussed. Finally, the statistical theory of overlap is used for 101 peaks and predictions are made for the number of singlet, doublet, and multiplets analyte peaks. The effect of increasing peak capacity by FT even derivative sharpening and power laws leads to a decrease in the number of peak overlaps and an increase in total peak number. Graphical abstract.
在复杂样品的液相色谱分析中，主要关注点是高的峰容量分离。使用先进的仪器和最佳的小尺寸、高效率柱，现在可以在相对较短的时间内分析复杂的多组分混合物。尽管有了这些进步，但仍然观察到色谱峰重叠。最近，人们的注意力已经从提高色谱效率和选择性转移到了收集后的数据处理上。曲线拟合方法可用于跟踪潜在的峰，但不能直接增强色谱分辨率。最近，基于导数和幂变换特性的方法被证明可以在保持关键峰信息（峰面积和保留时间）的同时增强色谱峰分辨率。这些方法已经在其基本性质、优点和局限性方面得到了广泛的研究，但尚未用复杂的色谱图进行评估。在此，我们评估了使用傅里叶变换（FT）反卷积、偶数导数峰锐化和幂律法，在超高压液相色谱中分离 101 组分混合物的快速分离（<60s）。在这种梯度分离中存在高噪声和峰重叠，这是快速色谱的代表。本文展示并描述了色谱分辨率的增强。此外，还保持并展示了代表性示例的准确定量。讨论了峰容量和峰峰分辨率的增强。最后，使用重叠的统计理论对 101 个峰进行了预测，并对单峰、双峰和多峰分析物峰的数量进行了预测。通过 FT 偶数导数锐化和幂律增加峰容量的效果导致峰重叠数量减少，总峰数量增加。
