CO2 Adsorption on Supported Molecular Amidine Systems on Activated Carbon
W. Richard Alesi Jr., McMahan Gray and John R. Kitchin, “CO2 Adsorption on Supported Molecular Amidine Systems on Activated Carbon”, ChemSusChem 3(8), 948-956 (2010). https://doi.org/10.1002/cssc.201000056
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Note on the source text: the extracted text is two-column interleaved, figures are not available, and minus signs were lost in extraction. The Gibbs free energies below are reported as negative values, as the equilibrium constants greater than one and the surrounding discussion require.
Question
Solid sorbents could avoid the cost of heating water in each regeneration cycle of aqueous amine capture, but polymeric and grafted amine sorbents mix primary, secondary and tertiary sites, so the thermodynamics of the capture reaction are hard to extract. This paper uses molecular tertiary amidines, where every molecule carries a single basic site and the number of active sites is known from the loading, to measure capture capacity and efficiency, to establish whether water is needed, and to derive a reaction free energy that can predict performance under other capture and regeneration conditions (Alesi et al. 2010).
Approach
Two bicyclic derivatives of N-methyltetrahydropyrimidine (MTHP), 1,5-diazabicyclo[4.3.0]non-5-ene (DBN) and 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), were wet-impregnated from methanol onto a Calgon activated carbon (BET area 1197 m2/g, 320 m2/g micropore area). Loading was capped near 2.7 mol amidine/kg (40 wt%) because higher loadings left a sticky, undryable solid. Capacities were measured at NETL in a 0.375 inch tubular packed bed with mass spectrometer detection, adsorbing from about 10% CO2 in He with controlled humidity (set by a temperature-controlled sparger) at 26 to 50 C, then regenerating by switching to an inert sweep and ramping to about 80 C. A void space correction measured on bare carbon was subtracted. Thermogravimetric analysis probed water uptake. The capture efficiency (mol CO2 per mol amidine) was fed into a single-site, two-component Langmuir isotherm in which CO2 and H2O react with the amidine to form bicarbonate, giving an equilibrium constant and a reaction Gibbs energy, which was corrected to the standard state using JANAF gas-phase data and the assumption that the amidine and its bicarbonate salt contribute similar free energies (Alesi et al. 2010).
Key findings
At 26 C in roughly 10% CO2 and 2.5% water, DBN adsorbed 0.78 and released 0.72 mol CO2/kg, while DBU adsorbed 0.34 and released 0.32 mol CO2/kg, giving efficiencies of 0.28 and 0.17 mol CO2 per mol amidine. DBN outperformed DBU in every sample, consistent with its reported higher basicity. With about 7% water, DBN capacity fell from 0.86 mol/kg (efficiency 0.30) at 29 C to 0.52 mol/kg (0.18) at 50 C. The sorbent mass barely changed (1.9 to 1.8 g), so the loss reflects weaker thermodynamic driving force rather than amidine evaporation, and no diffusion-related rise in capacity with temperature was seen. Capacity was constant over four cycles with desorption at 80 C.
Water proved essential: a dried feed (0.1% water) gave no desorbed CO2, ruling out direct zwitterion formation and supporting bicarbonate formation. Raising water from 2.5% to 3.9% increased efficiency, but at 7.4% water it dropped from 0.3 to 0.16 mol/mol. Water desorbing from the sorbent was nearly an order of magnitude larger than CO2, and the authors attribute the loss to bulk water uptake by the hydrophilic, microporous carbon blocking CO2 diffusion.
The derived standard reaction free energies were about -25 to -26 kJ/mol for DBU and about -28 kJ/mol for DBN, smaller in magnitude than the values the authors cite for converting solid Na2CO3 and K2CO3 to bicarbonates (about -35 and -41 kJ/mol). Mapping efficiency against standard reaction free energy for capture at 30 C and 0.1 bar CO2 with regeneration at 75 C and 0.0001 bar CO2 gives a feasibility window: amines weaker than about -20 kJ/mol have no working capacity and those stronger than about -110 kJ/mol cannot be regenerated. Because the amidines sit near the weak end, the model predicts that higher CO2 pressure should raise efficiency (from 0.3 to 0.8 and 0.95 at 0.24 and 0.54 bar). Experiments at 50% CO2 raised DBN efficiency from 0.3 to 0.53, the right trend but less than predicted, which the authors suggest may reflect interactions between neighboring bicarbonates or a spread of site energies (Alesi et al. 2010).
Significance
The work shows how a molecular, single-site sorbent turns breakthrough data into a reaction free energy and a simple thermodynamic screen for whether an amine suits a given capture and regeneration cycle, including the point that a sorbent suited to a packed bed may differ from one suited to a fluidized bed. It also identifies the support’s water uptake as a practical limit for bicarbonate-forming chemistry, and the authors note the amidine loading here (and capacities of 0.4 to 1.0 mol/kg) are well short of the 3 to 4 mol/kg they cite as needed to compete with liquid amines (Alesi et al. 2010).