Conference article

Carbon dioxide absorption by mixtures of propylene oxide / polypropylene carbonate at 600C was monitored by Raman spectroscopy at 20, 40 and 60 bar in a 2 L autoclave reactor. Multivariate preprocessing techniques were used to process raw Raman spectra and Principal Component Analysis was performed. Simulation data from the Peng- Robinson equation of state were used to model the absorbed CO2 amount and spectroscopic signals. Results showed that Principal Component Analysis can be used to explore the dynamics of the system at different pressure levels and to track the CO2 absorption. A similar analysis was carried out to monitor CO2 absorption by four different amines at room temperature and pressure in a batch reactors. The CO2 content was determined from titration and was used to model the spectroscopic data. Principal Component Analysis proved to be able to identify CO2 absorption capacity in the amines. This feasibility study confirms that Raman spectroscopy together with multivariate analysis can effectively report chemical information and dynamics in these CO2 absorption systems and hence can be used for developing regression models for online monitoring and control.

principal component analysis, CO2 absorption, propylene oxide, amines