Response surface modelling of CO₂ capture by ammonia aqueous solution in a microchannel

Abstract

Post-combustion CO 2 capture is one strategy of greenhouse gases mitigation. Ammonia is a useful option as CO 2 absorbent and an alternative to conventional amine-based solutions. This study deals with CO 2 capture by ammonia aqueous solution in a co-current two-phase flow by utilizing a T-shaped microchannel. Three parameters of temperature, gas flow rate, and ammonia concentration were considered as the main parameters affecting the CO 2 capture efficiency. A response surface methodology based on central composite design (CCD) was used to model the CO 2 capture efficiency as output in terms of the aforementioned input variables. CCD suggested a quadratic model to fit the experimental data. The model validation was implemented by ANOVA. All statistic tools including correlation coefficient, P-value, and F-value of the model, and P-value of lack-of-fit confirmed that the prediction model was significant. It was deduced from F-values that the importance of the input variables followed the sequence of ammonia concentration > gas flow rate > temperature. Ammonia concentration was the most effective input variable because there was a direct correlation between ammonia concentration and the number of absorption sites in the liquid phase. Numerical optimization predicted the best output of 96.48% CO 2 capture under the following optimum conditions: temperature of 20.00 C, gas flow rate of 110.59 mL/min, and ammonia concentration of 0.1382 mL/mL (13.82 vol%). The average CO 2 capture of 95.42% obtained at the input conditions indicates the accuracy of the prediction model.

Green Synthesis of Reduced Graphene Oxide Using the Tinospora cordifolia Plant Extract: Exploring Its Potential for Methylene Blue Dye Degradation and Antibacterial Activity

“…The internal studentized residuals and predicted values for the % of MB dye removal are correlated in Figure c. The randomly distributed plot demonstrated that for all of the obtained responses, the variance of originally collected data remains the same . Internally studentized residuals and run numbers for the % removal of MB dye were correlated and are shown in Figure d.…”

“…The internal studentized residuals and predicted values for the % of MB dye removal are correlated in Figure 7 c. The randomly distributed plot demonstrated that for all of the obtained responses, the variance of originally collected data remains the same. 59 Internally studentized residuals and run numbers for the % removal of MB dye were correlated and are shown in Figure 7 d. This plot revealed that there was no valid correlation between experimental data points and the outcomes observed. Design Expert software generated multiple runs for various independent variables.…”

“…The randomly distributed plot demonstrated that for all of the obtained responses, the variance of originally collected data remains the same. 59 Internally studentized residuals and run numbers for the % removal of MB dye were correlated and are shown in Figure 7d. This plot revealed that there was no valid correlation between experimental data points and the outcomes observed.…”