Parameter Estimation and Optimization of a PSA Process for CO2 Capture

The PSA process is one of the viable technologies for CO2 capture from large CO2 generating sources. It is generally known that a two-stage PSA process is required to recover 99% CO2 from flue gas containing 10-15% CO2. The first stage concentrates CO2 up to 50-60% while the second stage is used to further enrich the first-stage product up to 99% or higher. The operating cost of a PSA process for CO2 capture is mostly contributed by the operation of vacuum pumps. Hence, how to reduce the operation cost of vacuum pumps becomes an important issue in the concerned PSA process. 

The aim of this research is placed in re-evaluating the economy of the two-stage PSA process for CO2 capture using zeolite 13X as adsorbent through numerical simulation and optimization. The performance curves of commercial vacuum pumps were used for realistic calculation of the operating cost. In addition, a new mass transfer model was proposed to represent the adsorption behavior on an energetically heterogeneous surface, and the coefficients of the rate model were estimated through the experiments. For numerically stable calculation, a novel numerical procedure using gradient-directed adaptive predictive collocation with a cubic spline interpolation function and far-side boundary conditions were adopted. First, effects of the process variables such as the P/F ratio, desorption pressure, bed utilization factors, and so forth on the operating cost and CO2 recovery were investigated to select the decision variables for optimization. Then, the economy of the PSA process was evaluated for the optimized process conditions as a function of CO2 contents of the inlet flue gas and CO2 recovery rate.