Development of CO2 Selective Poly（Ethylene Oxide）-Based Membranes： From Laboratory to Pilot Plant Scale

Membrane gas separation is one of the most promising technologies for the separation of carbon dioxide （CO2） from various gas streams. One application of this technology is the treatment of flue gases from combustion processes for the purpose of carbon capture and storage. For this application, poly（ethylene oxide）-containing block copolymers such as Pebax or PolyActiveTM polymer are well suited. The thin-film composite membrane that is considered in this overview employs PolyActiveTM polymer as a selective layer material. The membrane shows excellent CO2 permeances of up to 4 m^3（STP）.（m^2·h·bar）^-1 （1 bar = 105 Pa） at a carbon dioxide/nitrogen （CO2/N2） selectivity exceeding 55 at ambient temperature. The membrane can be manufactured reproducibly on a pilot scale and mounted into fiat-sheet membrane modules of different designs. The operating performance of these modules can be accurately predicted by specifically developed simulation tools, which employ single-gas permeation data as the only experimental input. The performance of membranes and modules was investigated in different pilot plant studies, in which flue gas and biogas were used as the feed gas streams. The investigated processes showed a stable separation performance, indicating the applicability of PolyActiveTM polymer as a membrane material for industrialscale gas processing.

Absorption linewidth inversion with wavelength modulation spectroscopy

For absorption linewidth inversion with wavelength modulation spectroscopy（WMS）, an optimized WMS spectral line fitting method was demonstrated to infer absorption linewidth effectively, and the analytical expressions for relationships between Lorentzian linewidth and the separations of first harmonic peak-to-valley and second harmonic zero-crossing were deduced. The transition of CO_2 centered at 4991.25 cm^（-1） was used to verify the optimized spectral fitting method and the analytical expressions. Results showed that the optimized spectra fitting method was able to infer absorption accurately and compute more than 10 times faster than the commonly used numerical fitting procedure. The second harmonic zero-crossing separation method calculated an even 6 orders faster than the spectra fitting without losing any accuracy for Lorentzian dominated cases. Additionally, linewidth calculated through second harmonic zero-crossing was preferred for much smaller error than the first harmonic peak-to-valley separation method. The presented analytical expressions can also be used in on-line optical sensing applications, electron paramagnetic resonance, and further theoretical characterization of absorption lineshape.