Hydrophobic Silica Gels and Aerogels for Direct Air Capture: Hybrid Grafting To Suppress Water Uptake and Capillary Condensation

Abstract

Direct air capture of CO₂ (DAC) is critical to achieving climate goals, requiring the removal of approximately 220 000 Mt of CO₂ by 2100. Amine-functionalized sorbents are promising for CO₂ capture but face challenges under high humidity, where water sorption reduces the CO₂ selectivity and capillary condensation may degrade the sorbent, especially the delicate silica aerogel backbone, compromising structural integrity. To address this issue, we introduce superhydrophobic silica gels functionalized by hybrid grafting of amine, long-aliphatic, and trimethylsilyl groups. These materials drastically reduce water uptake, from >50 wt % for nonhydrophobized gels to 5 wt % at 95% RH, and maintain high CO₂/water selectivity at high relative humidity. This shows how to overcome a limitation of silica sorbents, which have shown poor performance up to now under realistic high-humidity DAC conditions. The hydrophobic functional groups shield the hydrophilic amine sites and prevent capillary condensation, preserving pore hydrophobicity, preventing pore clogging with water, and ensuring rapid CO₂ diffusion, essential for sustained performance. We demonstrate that conventional single-component and dry-condition CO₂ sorption tests overlook the critical role of moisture, underestimating the detrimental effect of presorbed water on nonhydrophobized silica gels, previously reported in the literature. The ability to apply ambient pressure drying without significant selectivity loss, along with the affordability of the silica matrix, enhances the feasibility of real-world implementation for silica-based DAC sorbents. Given the versatility of amorphous sorbents, this study paves the way for further material optimization tailored to DAC applications.

Keywords: CO2/water selectivity; amine-functionalized sorbents; direct air capture; hybrid grafting functionalization; silica aerogel; superhydrophobic; surface modification.

1

(A) General scheme of gel synthesis and functionalization. Scanning electron microscopy images on (B) xerogel (2amino-16C-TMS-3Vol-Xero) and (C) aerogel (2amino-16C-TMS-3Vol-Aero). Details of the imaging can be found in the Supporting Information.

2

Breakthrough experiment on the 2amino-Xero sample using 400 ppm of CO₂ in nitrogen. (A) First cycle: relative humidity is gradually increased from 0 to 80%. (B) Second cycle: the sample is first saturated with 80% of RH, and then the breakthrough with 400 ppm of CO₂ starts. For clarity, sample (re)­activation, calibration stage, and blank experiments are not shown in the figure.

3

Example nitrogen adsorption isotherms measured on silica gels at 77 K.

4

Solid-state NMR spectra: (A) nonquantitative one-dimensional ¹H–²⁹Si cross-polarization; (B) quantitative ¹³C spectra. Spectra are normalized to the same number of scans and same mass of the sample in the rotor.

5

Water adsorption isotherms measured on silica gels and a Lewatit VP OC 1065. For Lewatit and 2-amino-Xero, only the adsorption branch was measured. For clarity, samples are separated between high mass uptake (A) and low-to-moderate mass uptake (B).

6

Examples of the pore size distribution calculated using the BJH method. Solid and dashed lines represent incremental and cumulative pore volume, respectively.

7

Single component CO₂ adsorption isotherms measured on silica gels at 25 °C.

8

CO₂ adsorption was measured during the breakthrough experiments (400 ppm of CO₂) under varying relative humidity. The CO₂ sorption was evaluated for each relative humidity step in the 1st breakthrough cycle, Figure A.

9

Comparison of adsorbed (summarized for all relative humidity steps) and total desorbed CO₂ amount in the first and second cycles of the breakthrough (400 ppm of CO₂). Note that the 2nd cycle was carried out after saturation with water (80% relative humidity, without CO₂).

10

CO₂/H₂O adsorption selectivity on silica gels and Lewatit VP OC 1065 (400 ppm of CO₂). The data are calculated based on the CO₂ adsorption measured in the breakthrough experiments for each individual RH step in the first cycle. Water adsorption was quantified during the separate dynamic vapor sorption experiments. Yellow and blue colors indicate usual RH values in Europe in summer and winter, respectively (New et al. 2002 ). Data points above 80% of RH for silica gels were extrapolated, assuming constant CO₂ uptake measured at 80%, utilizing relatively constant CO₂ sorption as a function of RH (the symbols without filling; cf. Figure ).

11

Experimental and simulated CO₂ breakthrough curves under dry conditions and after sample saturation with 80% relative humidity prior to CO₂ breakthrough for (a) 2amino-Xero, (b) 2amino-16C-TMS-Xero, and (c) 2xConc.-2amino-TMS-Xero.