Porcar Iolanda, García-Lopera Rosa, Abad Concepción, Campos Agustín
Departament de Química Física & Institut de Ciencia dels Materials (ICMUV), Universitat de València, Burjassot, València, Spain.
J Sep Sci. 2007 Aug;30(13):2037-45. doi: 10.1002/jssc.200700063.
The size-exclusion chromatographic (SEC) behaviour of different solvent/polymer systems in three packing sets has been analysed from fractal considerations. The three-column sets studied are specifically formed by: (i) 'pure' micro-styragel, (ii) 'mixed' TSK Gel H(HR + XL + HR) and (iii) mixed TSK Gel H(XL + HR + XL). The experimental data reveals that in most of the systems assayed the classical universal calibration (UC) is not fulfilled, denoting the existence of secondary effects accompanying the main SEC mechanism. In order to obtain an accurate characterization of different polymers eluted in solvent mixtures and/or mixed packings, the use of a reliable and trusted calibration curve is required. In this sense, two alternative procedures have been analysed: the specific (SC) and the fractal (FC) calibrations. The results have evidenced that the use of the FC instead of the classical universal method diminishes up to nine times (in the case of the micro-styragel set) the mean deviation on the calculated molar mass with respect to the value given by the supplier. In the case of TSK Gel-based sets, the mean deviation is reduced to the half. The SC curve made with standards of the sample under study also reduces the mean deviation values but needs a broad set of narrow standards, whereas the fractal approach only needs one polymeric sample to build up the calibration curve.
从分形学角度分析了不同溶剂/聚合物体系在三种填料柱中的尺寸排阻色谱(SEC)行为。所研究的三组色谱柱具体由以下组成:(i)“纯”微Styragel,(ii)“混合”TSK Gel H(HR + XL + HR),以及(iii)混合TSK Gel H(XL + HR + XL)。实验数据表明,在大多数被测体系中,经典的普适校准(UC)并不成立,这表明在主要的SEC机制之外还存在次级效应。为了准确表征在溶剂混合物和/或混合填料中洗脱的不同聚合物,需要使用可靠且可信的校准曲线。从这个意义上讲,分析了两种替代方法:特定校准(SC)和分形校准(FC)。结果表明,使用FC而非经典的普适方法,可将计算得到的摩尔质量相对于供应商给出的值的平均偏差降低多达九倍(在微Styragel柱的情况下)。对于基于TSK Gel的色谱柱组,平均偏差降至一半。用所研究样品的标准品制作的SC曲线也降低了平均偏差值,但需要大量的窄分布标准品,而分形方法仅需一个聚合物样品来建立校准曲线。