Extending the Determination of Collision Cross Section Parameters from Experimental Gaseous Diffusion Coefficients

The temperature variation in gaseous diffusion coefficients has been previously employed for estimating collision cross sections (potential maxima, collision diameters). Rather than simply reporting new diffusivity data, the primary aim is to identify and quantify the major error-inducing factors in these parameters, with dichlorodifluoromethane in nitrogen, oxygen and synthetic air as model systems, using two different correlations (extended principle of corresponding states, Lennard-Jones 12-6 potential). Diffusivity measurements by reversed-flow gas chromatography were fitted to the general relation of gaseous diffusivity with temperature, removing outliers and permitting the study of the effect of different temperature pairs on the cross-sectional uncertainties. It was found that small temperature ranges yield potential maxima with large uncertainties, as the smaller slope of the collision integral ratio increases the contribution of experimental error (temperature and gaseous diffusivity), while collision diameters are less affected by the uncertainty of potential maxima. Increased precision can be achieved by utilizing accurate diffusion coefficients over a wide temperature range, thus obtaining a clearer image of the interactions between dissimilar molecules. Comparisons revealed the unsuitability of the Lennard-Jones model in predicting the intermolecular interactions in the studied systems. Large deviations were observed in parameters obtained by scaling factors using Lorentz–Berthelot combining rules, while collision diameters from viscosity data showed the closest agreement (2.4–3.3%).