(634g) Batch Adsorber Analogs for Rapid Screening of Adsorbents for CO2 Capture

The adsorptive CO₂ capture arena has seen tremendous growth with the ability to synthesize new materials, e.g., MOFs, ZIFs, etc. With a wide range of materials that are now available there is a need for effective tools to rank materials according to their potential to perform at a process scale. Our recent work has indicated that traditional metrics such as selectivity, working capacity, etc. show a poor ability to predict process performance, since they do not take into account the complexities, of adsorption kinetics, isotherm shapes, heat effects and process configurations into consideration [1]. With the access to inexpensive computational power, it is now possible to combine rigorous optimization methods with detailed models to compare material performance in a quantitative manner. However, these approaches are time-consuming and require more sophisticated numerical schemes. Maring and Webley introduced the analog model in which the different steps of a P/VSA process was simulated by considering a batch adsorber without axial gradients [2] . The key advantage of this approach is that the Partial differential equations (PDEs) are converted to an Ordinary differential equation (ODE); and the requirement to repeated simulations to reach cyclic steady state (CSS) is eliminated. They also showed that the prediction of process performance indicators such as purity and recovery when compared to detailed models were reasonable compared to the savings in computational time. In their work, they considerd a simple 4-step cycle which resulted in low value of purity and recovery.

In this presentation, we extend the analysis of Maring and Webley by considering more complex cycles that include a three pressure-level operation; reflux, pressure equalization and light-product pressurization steps. This model is then used to predict the performance of 4 adsorbents, namely Zeolite 13X, Mg-MOF-74, UTSA-16, a type of coconut-shell derived activated carbon for post-combustion CO2 capture. The model assumes instantaneous equilibrium and isothermal operations. The impact of making these assumptions is first verified and then the batch-analog model is first validated by comparing the CO2 purity and recovery with that predicted by detailed models. By varying the operating conditions, the Pareto curve for maximizing purity and recovery is obtained and is comapred to the ones from detailed optimization [1]. The results show that although the Pareto curves from the two modeling approaches are different, the relative ranking of adsorbents is indeed correct. Based on this observation, the effect of operating variables, e.g., intermediate and low pressure on the process performance is evaluated and the specific energy consumption for the different materials is calculated. Finally the by considering the Langmuir adsorption isotherm to represent both CO2 and N2, the effect of the isotherm parametrs, equilibirum constant, saturation capacity and heats of adsorption, on the process performance is mapped in the form of contour plots. These plots provide guidance for the development of new materials that could potentially show superior performance at the process scale.

References

[1] Rajagopalan, A. K.; Avila, A. M. & Rajendran, A.Do adsorbent screening metrics predict process performance? A process optimisation based study for post-combustion capture of CO₂ Int. J. Greenhouse Gas Control, Elsevier, 2016, 46, 76-85.

[2] Maring, B. J. & Webley, P. A. A new simplified pressure/vacuum swing adsorption model for rapid adsorbent screening for CO₂ capture applications Int. J. Greenh. Gas Con., Elsevier, 2013, 15, 16-31