(524b) Dynamic Modeling Studies On Absorption of CO2 in Rotated Packing Bed

Absorption by chemical solvents is one of the most mature technologies developed for CO2 capture from flue gas.  Traditional devices, such as packed towers, required huge volume because of mass transfer limitation，which is a critical factor contributing to the cost of the process.   The rotating packed bed (RPB) was proposed to relax mass transfer limitations Lin et al 2003, Chen and Tan 2006, Jassim et al. 2007, Cheng and Tan 2009, 2011.  However, there are a limited number of studies on modeling of CO2 capture by alkanoamines using RPB.  For example Yu et al 2012 proposed a model with an assumption of 6 stirred tanks in series followed by a gas–liquid contactor. 

        In this work, a model of absorption of CO2 by mono-ethanolamine (MEA) in RPB was developed using two-film theory for mass transfer.  Traditionally, in RPB studies, the Onda’s correlation (Onda et al 1968) was used for predicting gas phase mass transfer coefficients.  The correlation proposed by Tung and Mah 1985 was used for predicting liquid phase mass transfer coefficients.  The gas-liquid interfacial area was predicted by the Onda’s correlation.    The only dependence on gravity is in the gas-liquid interface area correlation, which depends on the Froude number (Fr).   However the Onda correlation has only an almost negligible dependence upon Fr (Fr^-0.05); leading to under prediction of capture efficiency.  Recently, Hanley and Chen (2012), in a new study on mass transfer characteristics of trays and packing, proposed a new correlation for the gas-liquid interfacial area, which has a more prominent dependence on gravity (Fr^-0.161).  We found that this modified expression can lead to better agreement with experimental data using 30 wt% MEA solutions without the need to introduce an extra void volume.

        In CO2 capture by alkano-amine solutions, mass transfer is accompanied by reaction.  A pseudo-first order reaction constant was used to calculate enhancement factor of mass transfer rate.  In this study, we found that this model significantly underestimated capture efficiency for experiments conducted with high concentrations of MEA.  Aboudheir et al 2003 suggested that pseudo-first order reaction is not appropriate and suggested a new kinetic model to account for improved enhancement at high MEA concentrations.   However, the model showed some systematic errors with respect to MEA concentrations.  An empirical correction was introduced in this work to improve the prediction of enhancement factors.  These corrections of enhancement factors enable us to predict experiment data provided by Jassim et al 2007 at high MEA concentrations.

        In summary, an improved model of CO2 capture by MEA solutions by RPB was developed by modifying the gravity dependence of gas-liquid interfacial area and the enhancement factor at high MEA concentrations.  By integrating these improvements obtained from independent studies, we can substantially improve the predictions of CO2 capture efficiency provided in the literature by various research groups.