Bicarbonate-Carbonate Selectivity through Nanofiltration for Direct Air Capture of Carbon Dioxide

[Geoengineering News]

https://pubs.acs.org/doi/epdf/10.1021/acsestengg.4c00150

Authors 

Anatoly Rinberg and Michael J. Aziz

Publication Date:June 6, 2024

https://doi.org/10.1021/acsestengg.4c00150

Abstract 

Direct air capture (DAC) of carbon dioxide is one approach among many proposed that is capable of offsetting hard-to-avoid emissions. In previous work, we developed the alkalinity concentration swing (ACS) method, which is driven through concentrating an alkaline solution that has been loaded with atmospheric CO2 by desalination technologies, such as reverse osmosis or capacitive deionization. Though the ACS is promising in terms of energy usage and implementation, its absorption rate and water requirements are infeasible for a large-scale DAC process. Here, we propose an improvement on the ACS, the bicarbonate-enriched alkalinity concentration swing (BE-ACS), which selects bicarbonate ions from a stream of aqueous alkaline solution that has absorbed atmospheric CO2. The bicarbonate-rich stream is then concentrated, which greatly increases its CO2 partial pressure, and then CO2 is extracted from solution. We experimentally investigate the use of pressure-driven nanofiltration (NF) membrane-based separation to select bicarbonate ions over carbonate ions. We screen commercial membranes and select one high-performance membrane for detailed studies, quantifying its bicarbonate-carbonate selectivity factor and bicarbonate-passage factor. Feed pH, the combined concentration of aqueous CO2, bicarbonate, and carbonate species (or dissolved inorganic carbon), alkalinity, and permeation flux are systematically varied to study NF separation properties. We find that the selectivity factor, which exceeds 30 times in certain regimes, increases with higher feed pH and higher alkalinity. The performance metrics of the selected NF membrane are input into a theoretical BE-ACS cycle analysis, and the required energy input and cycle capacity output are evaluated. Ideal cycle energy is found to be as low as around 250 kJ/mol, with opportunities identified for further decreases through process engineering and forward osmosis energy recovery.

Source: ACS PUBLICATIONS