Development and Characterization of Polyethylenimine-Infiltrated Mesoporous Silica Foam Pellets for CO₂ Capture

Polyethylenimine (PEI) has been shown to be promising for direct air capture (DAC) of carbon dioxide and has potential for commercial scale-up globally. Laboratory scale processes include multiple steps, such as mixing, solvent extraction, vacuum application, sonication, and various flushes and activation steps. It is critical to properly control these operating parameters to achieve higher capture capacity as a result of the optimized material configuration. This study adopts previously published pelletization processes for PEI-infiltrated mesoporous foam silica (mesoporous silica foam) to uncover the adsorption mechanisms and optimize the associated fabrication steps, such as sonication, to achieve higher sorbent productivity. A high capture capacity was achieved at 46 °C for 75 wt % PEI loading (2.27 mmol/g) followed by PEI_MSF 70 (1.81 mmol/g) and PEI_MSF 80 (1.44 mmol/g). As part of the optimization, sonication parameters of frequency, amplitude, and time were modified for PEI_MSF 75 sorbent, which resulted in the highest uptake capacity of 3.04 mmol/g (sonicated at 40 kHz and a wave amplitude of 50% for 30 s). These preliminary results would tend to prove that sonication energy affects carbon capture capacity, although there is still a lack of understanding regarding the exact underlying mechanism, suggesting the need for further investigation. It is important to note that the present work is focused on the adsorption mechanisms and not desorption or durability of the capture performance. Ongoing research addresses these factors. This paper is intended to establish baseline DAC behavior of a promising capture medium and begins probing the optimization spectrum by considering the effects of sonication energy on adsorption. Ongoing work intends to address potential abbreviations of the full range of process steps and furthers the understanding of kinetics by considering the desorption and resorption attributes.