Volumetric Frequency Response Investigation of the Mass Transfer Mechanisms of N2, O2, and Ar in Carbon Molecular Sieve

Despite decades of research, the dominant mass transfer mechanisms of N2, O2, and Ar in carbon molecular sieve (CMS) have yet to be well understood. Hence, the objective of this work was to collect experimental data for the uptake and release of N2, O2, and Ar in Shirasagi MSC 3K 172 CMS by volumetric frequency response (VFR) and then contrast these data against several micropore mass transfer models. Data were collected for N2, O2, and Ar in this CMS at 100 and 200 Torr at 25 degrees C, and at 750 Torr at 20, 30, 40, and 50 degrees C. Data were measured at 32 different frequencies, spanning 5 x 10-5 to 10 Hz, for each temperature and pressure. The resulting intensity curves, comprising 192 data points, were fit to six different micropore models to identify the dominant mass transfer mechanisms of N2, O2, and Ar in this CMS. These progressively more complex micropore models included the following resistances: single site micropore (SSM) always with Darken loading dependence, single site micropore mouth (SSPM), single site micropore and single site micropore mouth combined (SSC), SSC with Darken loading dependence also applied to the micropore mouth (SSCD), SSCD with an additional more complex Qinglin empirical loading dependence (SSCDQ), and dual site combined with Qinglin loading dependence (DSCDQ). The resulting phase lag curves comprising 192 data points were subsequently predicted by each model. The model that best described the behavior of N2, O2, and Ar in Shirasagi MSC 3K 172 CMS was the DSCDQ model, which consisted of two micropore sites, each with different micropore diffusion resistances but with the same micropore mouth resistance and each resistance with both Darken and Qinglin loading dependencies. The evidence for these two micropore sites came from unique features exhibited only by O2 in the intensity and especially in the phase lag curves and only by the 100 and 200 Torr data. The DSCDQ model was the only one that could capture these features. This result illuminated the power of the VFR technique for discriminating transport mechanisms and the importance of experimentally examining a broad range of conditions with VFR.