Wine Science Programme, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand.
Kim Crawford Winery, Constellation Brands NZ, 237 Hammerichs Road, Blenheim 7273, New Zealand.
J Agric Food Chem. 2021 Dec 15;69(49):15020-15035. doi: 10.1021/acs.jafc.1c06411. Epub 2021 Dec 7.
A new quantitative method based on static headspace-gas chromatography-ion mobility spectrometry (SHS-GC-IMS) is proposed, which enables the simultaneous quantitation of multiple aroma compounds in wine. The method was first evaluated for its stability and the necessity of using internal standards as a quality control measure. The two major hurdles in applying GC-IMS in quantitation studies, namely, nonlinearity and multiple ion species, were also investigated using the Boltzmann function and generalized additive model (GAM) as potential solutions. Metrics characterizing the model performance, including root mean squared error, bias, limit of detection, limit of quantitation, repeatability, reproducibility, and recovery, were investigated. Both nonlinear fitting methods, Boltzmann function and GAM, were able to return desirable analytical outcomes with an acceptable range of error. Potential pitfalls that would cause inaccurate quantitation, that is, effects of ethanol content and competitive ionization, were also discussed. The performance of the SHS-GC-IMS method was subsequently compared against that of a currently established method, namely, GC-MS, using commercial wine samples. These findings provide an initial validation of a GC-IMS-based quantitation method, as well as a starting point for further enhancing the analytical scope of GC-IMS.
提出了一种基于静态顶空-气相色谱-离子迁移谱(SHS-GC-IMS)的新定量方法,可同时定量葡萄酒中的多种香气化合物。该方法首先评估了其稳定性和使用内标作为质量控制措施的必要性。还使用 Boltzmann 函数和广义加性模型(GAM)研究了将 GC-IMS 应用于定量研究中的两个主要障碍,即非线性和多种离子种类。研究了表征模型性能的指标,包括均方根误差、偏差、检测限、定量限、重复性、再现性和回收率。Boltzmann 函数和 GAM 这两种非线性拟合方法都能够返回具有可接受误差范围的理想分析结果。还讨论了可能导致定量不准确的潜在陷阱,即乙醇含量和竞争电离的影响。随后使用商业葡萄酒样品将 SHS-GC-IMS 方法的性能与当前建立的方法(即 GC-MS)进行了比较。这些发现为基于 GC-IMS 的定量方法提供了初步验证,并为进一步扩展 GC-IMS 的分析范围提供了起点。