Developing Versatile Contactors for Direct Air Capture of CO2 through Amine Grafting onto Alumina Pellets and Alumina Wash-Coated Monoliths

Abstract

The optimization of the air-solid contactor is critical to improve the efficiency of the direct air capture (DAC) process. To enable comparison of contactors and therefore a step toward optimization, two contactors are prepared in the form of pellets and wash-coated honeycomb monoliths. The desired amine functionalities are successfully incorporated onto these industrially relevant pellets by means of a procedure developed for powders, providing materials with a CO₂ uptake not influenced by the morphology and the structure of the materials according to the sorption measurements. Furthermore, the amine functionalities are incorporated onto alumina wash-coated monoliths that provide a similar CO₂ uptake compared to the pellets. Using breakthrough measurements, dry CO₂ uptakes of 0.44 and 0.4 mmol g_sorbent^-1 are measured for pellets and for a monolith, respectively. NMR and IR studies of CO₂ uptake show that the CO₂ adsorbs mainly in the form of ammonium carbamate. Both contactors are characterized by estimated Toth isotherm parameters and linear driving force (LDF) coefficients to enable an initial comparison and provide information for further studies of the two contactors. LDF coefficients of 1.5 × 10^-4 and of 1.2 × 10^-3 s^-1 are estimated for the pellets and for a monolith, respectively. In comparison to the pellets, the monolith therefore exhibits particularly promising results in terms of adsorption kinetics due to its hierarchical pore structure. This is reflected in the productivity of the adsorption step of 6.48 mol m^-3 h^-1 for the pellets compared to 7.56 mol m^-3 h^-1 for the monolith at a pressure drop approximately 1 order of magnitude lower, making the monoliths prime candidates to enhance the efficiency of DAC processes.

Figure 1

Grafting and polymerization mechanisms of triamine on alumina.

Figure 2

XRD spectra of pristine alumina pellets, wash-coat, and functionalized alumina pellets. A pure, calcined wash-coat material was measured so as to avoid overlapping with the spectrum of mullite.

Figure 3

N₂ isotherms at 77 K of support materials normalized to the mass of alumina.

Figure 4

N₂ isotherms at 77 K of pellets functionalized with (a) different amounts of amine and a constant amount of 50 μL g_alumina^–1 water added and (b) different amounts of water and a constant amount of 1000 μL g_alumina^–1 of amine added.

Figure 5

DRIFT spectra of pristine and functionalized pellets, one with an octyl chain (SA-OTMS-1000-75) and one with triamine (SA-TRI-930-75).

Figure 6

Dry CO₂ adsorption capacity at 400 ppm sorbents prepared with different amounts of triamine and water. The surface is interpolated between the measurement points, indicated as black dots.

Figure 7

Spectroscopic analysis of CO₂ adsorption: (a) ¹³C DNP-enhanced NMR spectra of crushed SA-TRI-1000-75 pellets (black) and SA-TRI-1000-75 pellets + adsorbed ¹³CO₂ (red) and (b) time-resolved DRIFT spectra of SA-TRI-1000-75 pellets during CO₂ adsorption at 400 ppm; the spectra are shown without scaling after subtracting from the original spectra the regenerated SA-TRI-1000-75 pellet spectrum obtained at t = 0.

Figure 8

Effect of the amount of water added during preparation on the CO₂ uptake and mass transfer kinetics. Samples produced with a constant amine addition of 1000 μL g_alumina^–1.

Figure 9

CO₂ capacity at 40 Pa different sections of the monolith measured using the volumetric technique and normalized to the mass of active sorbent (alumina + triamine). Inner (green) and outer (blue) channels are distinguished.

Figure 10

Two breakthrough curves of a monolith at 400 ppm CO₂ in N₂ at a flow rate of 0.001 mol s^–1; the volume of the monolith is 117 cm³. The flow is stopped after 4 h 10 min due to very slow diffusion in the last part of the curve.

Figure 11

Isotherms of upscaled batch of pellets (left) and whole monolith (right) depicted in Figure 9. A temperature-dependent Toth isotherm was fitted to both sorbents, whose values are shown in Table 3.

Figure 12

Relative capacity of pellets and of a monolith after 10 repeated cycles using volumetric measurements for the pellets and breakthrough measurements for a monolith at 400 ppm, normalized to the first measurement. The breakthroughs in cycles 6 and 7 were performed using 5.62% CO₂ in N₂ and are therefore not comparable. The shading of the monolith breakthrough data indicates the flow rates used: lightest blue = 0.18 mmol s^–1, light blue = 1.0 mmol s^–1, dark blue = 2.0 mmol s^–1, and darkest blue = 3.0 mmol s^–1.

Figure 13

SEM picture of a cross-section of the channel wall of a monolith. The darker patches correspond to the γ-alumina wash-coat integrated in the pores, and the light patches are mullite.