Considerable variation in the measurement of polymer–solvent interactions using gc retention data may occur due to kinetic factors, surface excess concentrations of probe vapour, and non-linear partition isotherms. The kinetic factors, which appear as a flow rate dependence of the retention volume, are analysed in terms of a previously reported theoretical model for retention on polymeric stationary phases passing through the glass transition. The predicted linear extrapolations to zero flow rate are obtained for the retention of n-tetradecane on polystyrene. The variation of this flow rate dependence with temperature and with the thickness of the stationary phase are also in qualitative agreement with the theory. A simplified model for the effect of loading on the retention diagram is presented. Non-linear absorption and bulk sorption isotherms result in a dependence of retention volume on sample size, necessitating an extrapolation of the measured retention volumes to zero peak height. The temperature variation of the flow rate dependence, the effect of loading, and the effect of sample size on retention volume are all further complicated by uneven distribution of polymer on support. From scanning electron micrographs of the beads it is evident that 'beading up' of the polystyrene on the glass surface may occur at low loadings, resulting in a non-uniform coating with large areas of the beads uncoated. It is concluded that in order to obtain reliable data on polymer–solvent interactions using gas chromatography, all of the above-mentioned factors must be considered.
Substrate-independent three-dimensional polymer nanosheets induced by solution casting
