Abstract Separation of amino acids (AAs) and their simple and inexpensive determination/identification is an interesting topic in biological and protein science, different food industries, and drug factories. Also, the presentation of the chromatographic behavior of compounds in a predictive model can be effective to estimate the structural/chemical properties of analyte and mobile phases. In this work for the first time, retardation factor (RF) of 42 AAs in reversed-phase thin layer chromatography (RP-TLC) was modeled. Acetonitrile-sodium azide solution and 1,2 dioxane-sodium azide solution were two mobile phases which have been studied in this work. Results showed that the values of RF are correlated with the structural properties of AAs and these properties had some similarities and differences in two noted mobile phases. For the TLC data in two mobile phases, five parametric linear models were suggested (R2train = 0.93 and 0.97; R2test = 0.93 and 0.99). The models were also evaluated with different statistical approaches. It was shown that increasing the sum of geometrical distances between N and O in AAs causes decreasing their RF in RP-TLC using both mobile phases. Other structural effects of AAs on their separation in the desired RP-TLC system were also discussed. Graphical Abstract

中文翻译：

---

RP-TLC 两种不同流动相中天然氨基酸的结构-延迟因子关系

摘要 氨基酸 (AA) 的分离及其简单、廉价的测定/鉴定是生物和蛋白质科学、不同食品工业和制药厂的一个有趣话题。此外，在预测模型中呈现化合物的色谱行为可以有效地估计分析物和流动相的结构/化学特性。在这项工作中，首次模拟了反相薄层色谱 (RP-TLC) 中 42 种 AA 的延迟因子 (RF)。乙腈-叠氮化钠溶液和 1,2 二恶烷-叠氮化钠溶液是本工作中研究的两种流动相。结果表明，RF 值与 AA 的结构特性相关，并且这些特性在两个著名的流动相中具有一些相似之处和不同之处。对于两种流动相中的 TLC 数据，建议使用五个参数线性模型（R2train = 0.93 和 0.97；R2test = 0.93 和 0.99）。还使用不同的统计方法对模型进行了评估。结果表明，在使用两种流动相的 RP-TLC 中，增加 AA 中 N 和 O 之间几何距离的总和会导致其 RF 降低。还讨论了 AA 在所需 RP-TLC 系统中对其分离的其他结构影响。图形概要 还讨论了 AA 在所需 RP-TLC 系统中对其分离的其他结构影响。图形概要 还讨论了 AA 在所需 RP-TLC 系统中对其分离的其他结构影响。图形概要