Department of Chemistry, University of Alberta, Edmonton, Alberta, T6G2G2, Canada.
Department of Chemistry, University of Alberta, Edmonton, Alberta, T6G2G2, Canada; Department of Medicinal Chemistry, School of Pharmacy, Guilan University of Medical Sciences, Rasht, Iran.
J Chromatogr A. 2014 Feb 21;1330:69-73. doi: 10.1016/j.chroma.2014.01.019. Epub 2014 Jan 15.
A new method for calibrating thermodynamic data to be used in the prediction of analyte retention times is presented. The method allows thermodynamic data collected on one column to be used in making predictions across columns of the same stationary phase but with varying geometries. This calibration is essential as slight variances in the column inner diameter and stationary phase film thickness between columns or as a column ages will adversely affect the accuracy of predictions. The calibration technique uses a Grob standard mixture along with a Nelder-Mead simplex algorithm and a previously developed model of GC retention times based on a three-parameter thermodynamic model to estimate both inner diameter and stationary phase film thickness. The calibration method is highly successful with the predicted retention times for a set of alkanes, ketones and alcohols having an average error of 1.6s across three columns.
提出了一种新的方法来校准热力学数据,以便用于预测分析物的保留时间。该方法允许在同一固定相的不同柱上使用在一根柱子上收集的热力学数据进行预测,但柱的几何形状有所不同。这种校准是必不可少的,因为柱子内径和固定相膜厚度之间的微小差异或柱子老化会对预测的准确性产生不利影响。校准技术使用 Grob 标准混合物以及 Nelder-Mead 单纯形算法和先前开发的基于三参数热力学模型的 GC 保留时间模型来估计内径和固定相膜厚度。该校准方法在三组烷烃、酮和醇的预测保留时间方面非常成功,三个柱子的平均误差为 1.6s。