Department of Chemistry, University of Washington, Box 351700, Seattle, WA, 98195, USA.
Department of Chemistry, University of Washington, Box 351700, Seattle, WA, 98195, USA.
J Chromatogr A. 2019 Nov 8;1605:460368. doi: 10.1016/j.chroma.2019.460368. Epub 2019 Jul 15.
Comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC × GC-TOFMS) is a powerful instrument for the analysis of complex samples. Deconvolution of overlapped analytes using a suitable chemometric data analysis method such as Parallel Factor Analysis (PARAFAC) is often required. However, PARAFAC is designed to require a strict data trilinearity requirement. In this study we examine how strict this requirement is in the context of GC × GC experimental conditions, and demonstrate that under suitable conditions the data is sufficiently trilinear to achieve accurate deconvolution. The term trilinear deviation ratio (TDR) was previously introduced as a quantitative metric to predict the accuracy of PARAFAC deconvolution. Trilinear deviation ratio is defined as the run-to-run retention time shift, Δt, for a given analyte on the second dimension (D) separation, divided by the D analyte peak width-at-base, W. We demonstrate that experimental conditions impact the TDR range produced and PARAFAC performance. Column selection and modulation period, P, are shown to significantly influence the TDR range. Two column sets were evaluated, giving rise to different k' ranges for the D separations. Each column set was used with an optimum P as well as a longer P to demonstrate the effect of P selection on the TDR range and PARAFAC quantification. A P of 6 s produced a Δt range from -19.5 ms to -98 ms and TDRs from 0.157 to 0.439, translating into a PARAFAC bias from +1.6% to -13.5%. However, a P of 1.5 s produced a Δt range of -1.1 ms to -8.8 ms, and significantly lower TDRs from 0.013 to 0.057, translating into PARAFAC errors from +2.1% to -3.9%, with an average of -1.1% ± 1.4. These results validate the idea that a suitable GC × GC experimental design will provide accurate quantification with PARAFAC.
全二维气相色谱飞行时间质谱联用仪（GC×GC-TOFMS）是一种用于分析复杂样品的强大仪器。通常需要使用合适的化学计量数据分析方法（如平行因子分析（PARAFAC））对重叠分析物进行解卷积。然而，PARAFAC 的设计要求严格的数据三线性要求。在本研究中，我们研究了在 GC×GC 实验条件下，这种要求的严格程度如何，并证明在适当的条件下，数据具有足够的三线性，可以实现准确的解卷积。三线性偏差比（TDR）一词以前被引入作为一种定量指标来预测 PARAFAC 解卷积的准确性。三线性偏差比定义为给定分析物在第二维（D）分离上的运行到运行保留时间偏移量Δt，除以 D 分析物峰底宽 W。我们证明了实验条件会影响产生的 TDR 范围和 PARAFAC 性能。柱选择和调制周期 P 被证明会显著影响 TDR 范围。评估了两个柱集，导致 D 分离的 k'范围不同。每个柱集都使用最佳 P 以及较长的 P 来演示 P 选择对 TDR 范围和 PARAFAC 定量的影响。P 为 6 s 产生的 Δt 范围为-19.5 ms 至-98 ms，TDR 范围为 0.157 至 0.439，导致 PARAFAC 偏差为+1.6%至-13.5%。然而，P 为 1.5 s 产生的 Δt 范围为-1.1 ms 至-8.8 ms，并且 TDR 显著降低至 0.013 至 0.057，导致 PARAFAC 误差为+2.1%至-3.9%，平均为-1.1%±1.4。这些结果验证了这样一种观点，即合适的 GC×GC 实验设计将为 PARAFAC 提供准确的定量。
