Laboratory for Forensic Chemistry, Department of Analytical Chemistry, Faculty of Chemistry, Jagiellonian University in Kraków, Ingardena 3, 30-060 Kraków, Poland.
Laboratory for Forensic Chemistry, Department of Analytical Chemistry, Faculty of Chemistry, Jagiellonian University in Kraków, Ingardena 3, 30-060 Kraków, Poland.
Talanta. 2017 May 15;167:714-724. doi: 10.1016/j.talanta.2017.02.017. Epub 2017 Feb 20.
The aim of this study was to develop the first CE-based method enabling separation of 20 structurally similar coumarin derivatives. To facilitate method optimization a series of three consequent Doehlert experimental designs with the response surface methodology was employed, using number of peaks and the adjusted time of analysis as the selected responses. Initially, three variables were examined: buffer pH, ionic strength and temperature (No. 1 Doehlert design). The optimal conditions provided only partial separation, on that account, several buffer additives were examined at the next step: organic cosolvents and cyclodextrin (No. 2 Doehlert design). The optimal cyclodextrin type was also selected experimentally. The most promising results were obtained for the buffers fortified with methanol, acetonitrile and heptakis(2,3,6-tri-O-methyl)-β-cyclodextrin. Since these additives may potentially affect acid-base equilibrium and ionization state of analytes, the third Doehlert design (No. 3) was used to reconcile concentration of these additives with optimal pH. Ultimately, the total separation of all 20 compounds was achieved using the borate buffer at basic pH 9.5 in the presence of 10mM cyclodextrin, 9% (v/v) acetonitrile and 36% (v/v) methanol. Identity of all compounds was confirmed using the in-lab build UV-VIS spectra library. The developed method succeeded in identification of coumarin derivatives in three real samples. It demonstrates a huge resolving power of CE assisted by addition of cyclodextrins and organic cosolvents. Our unique optimization approach, based on the three Doehlert designs, seems to be prospective for future applications of this technique.
本研究的目的是开发首个基于毛细管电泳（CE）的方法，实现 20 种结构相似的香豆素衍生物的分离。为了便于方法优化，采用了三连续的 Doehlert 实验设计与响应面法，以峰数和分析调整时间作为选择的响应。最初，考察了三个变量：缓冲液 pH 值、离子强度和温度（第 1 个 Doehlert 设计）。优化条件仅提供了部分分离，因此在下一步中检查了几种缓冲添加剂：有机溶剂和环糊精（第 2 个 Doehlert 设计）。也通过实验选择了最佳的环糊精类型。结果表明，添加甲醇、乙腈和七（2,3,6-三-O-甲基）-β-环糊精的缓冲液最有希望。由于这些添加剂可能潜在地影响分析物的酸碱平衡和离解状态，因此使用了第 3 个 Doehlert 设计（第 3 个）来协调这些添加剂的浓度与最佳 pH 值。最终，在碱性 pH 值 9.5 的硼酸盐缓冲液中，添加 10mM 环糊精、9%（v/v）乙腈和 36%（v/v）甲醇，实现了所有 20 种化合物的完全分离。使用实验室构建的 UV-VIS 光谱库确证了所有化合物的身份。所开发的方法成功地鉴定了三个实际样品中的香豆素衍生物。它展示了添加环糊精和有机溶剂辅助的 CE 的巨大分辨能力。我们基于三 Doehlert 设计的独特优化方法似乎很有前途，可用于该技术的未来应用。
