Galea Charlene, Mangelings Debby, Heyden Yvan Vander
Vrije Universiteit Brussel (VUB), Center for Pharmaceutical Research (CePhaR), Department of Analytical Chemistry and Pharmaceutical Technology (FABI), Laarbeeklaan 103, B-1090 Brussels, Belgium.
J Pharm Biomed Anal. 2015;111:333-43. doi: 10.1016/j.jpba.2014.12.043. Epub 2015 Jan 5.
Supercritical fluid chromatography (SFC) is drawing considerable interest as separation technique in the pharmaceutical industry. The technique is already well established in chiral separations both analytically and on a preparative scale. The use of SFC as a technique for drug impurity profiling is examined here. To define starting conditions in method development for drug impurity profiling, a set of dissimilar stationary phases is screened in parallel. The possibility to select a set of dissimilar columns using the retention factors (k-values) for a set of 64 drugs measured on 27 columns in SFC was examined. Experiments were carried out at a back-pressure of 150 bar and 25 °C with a mobile phase consisting of CO2 and methanol with 0.1% isopropylamine (5-40% over 10 min) at a flow rate of 3 mL/min. These k-values were then used to calculate correlation coefficients on the one hand and to perform a principal component analysis on the other. The Kennard and Stone algorithm, besides dendrograms and correlation-coefficient colour maps were used to select a set of 6 dissimilar stationary phases. The stationary phase characterization results from this study were compared to those from previous studies found in the literature. Retention mechanisms for compounds possessing different properties were also evaluated. The dissimilarity of the selected subset of 6 stationary phases was validated using mixtures of compounds with similar properties and structures, as one can expect in a drug impurity profile.
超临界流体色谱法（SFC）作为制药行业的一种分离技术正引起人们极大的兴趣。该技术在分析和制备规模的手性分离方面已经相当成熟。本文研究了将SFC用作药物杂质剖析技术的情况。为了确定药物杂质剖析方法开发的起始条件，并行筛选了一组不同的固定相。研究了使用在SFC中27根色谱柱上测得的64种药物的保留因子（k值）来选择一组不同色谱柱的可能性。实验在150 bar的背压和25°C下进行，流动相由二氧化碳和甲醇组成，并含有0.1%的异丙胺（在10分钟内从5%到40%），流速为3 mL/分钟。然后，一方面使用这些k值来计算相关系数，另一方面进行主成分分析。除了树形图和相关系数彩色图外，还使用肯纳德和斯通算法来选择一组6种不同的固定相。将本研究的固定相表征结果与文献中先前研究的结果进行了比较。还评估了具有不同性质的化合物的保留机制。使用具有相似性质和结构的化合物混合物验证了所选6种固定相子集的差异性，正如在药物杂质剖析中所预期的那样。
