Virtual Chromatography for the Effective Method Development in Liquid Chromatography of Polymers

Virtual chromatography approach was applied for method development in polymer separations. For this purpose, the suitability of three different retention models, the linear (LSSM) and the quadratic solvent strength model (QSSM), and a model specific for polymers (PCM) was investigated to describe and predict the retention behaviour of polymers and copolymers as a function of eluent composition, gradient time, and gradient shape. The PCM was extended to account for linear and multi-step gradients. An application in Origin software was written that can be used to predict retention times in gradients of virtually any shape. Poly(ethylene glycol)s (PEGs) were used as model polymers and the results were verified for poly(methyl methacrylate)s and also for copolymers of styrene with ethyl acrylate (SEA). It was found that all models are suited to predict properly gradient retention volumes when the analyte specific model parameters were extracted from gradient experiments. The LSSM and QSSM, in principle, cannot describe retention behaviour under critical or SEC conditions. Since the PCM is designed to cover all three modes of polymer chromatography, it is therefore superior to the other models. However, the determinations of analyte specific model parameters strongly depend on the selection of initial experiments. A rational strategy for a purposeful selection of the initial calibration experiments is proposed. This selection is based on the gradient elution of high molar mass homopolymers as well as statistical copolymers that occurs very close to the critical composition. The close relationship between composition at elution and the critical composition was also utilized for fast estimations of critical eluent compositions of different homopolymers for different stationary phase/mobile phase combinations. This method is also valid for statistical SEA copolymers and hence provides a simple novel procedure for the determination of the critical composition for statistical copolymers. In order to predict polymer separations, the complete retention behaviour peak shapes must also be predicted. Therefore, PCM was extended to incorporate the effect of band broadening due to molar mass polydispersity of the polymer samples employing a simple scaling relationship describing the dependence of R/D parameter on molar mass. Interestingly, no a priori information about the molar mass and polydispersity is needed to simulate broadening due to polydispersity. The instrumental broadening was accounted for by using column efficiency. Thus, knowing the column parameters, i.e. void and interstitial volume, number of theoretical plates, the complete prediction of the chromatograms in various conditions can be performed based on only three simple but rationally selected initial experiments. Using this approach suitability of a certain chromatographic system to obtain optimum separations can be evaluated in short time. This can save substantially the precious time in screening the mobile phases and stationary phases. The suitability of virtual chromatography approach as an effective tool for method development was demonstrated using model mixtures of homopolymers and statistical copolymers