An atom-level interaction design between amine and support: achieving efficient and stable CO2 capture

doi:10.21203/rs.3.rs-3837453/v1

Download PDF

Article

An atom-level interaction design between amine and support: achieving efficient and stable CO₂ capture

https://doi.org/10.21203/rs.3.rs-3837453/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 13 Jun, 2024

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

Amine-functionalized adsorbents hold great promise for CO₂ capture due to their exceptional selectivity and diverse applications scenarios. However, their applications are impeded by low efficiency and unstable cyclic performance. Here, a novel amine-support system is synthesized to achieve an efficient and stable CO₂ capture. Based on an atom-level design, each polyethyleneimine (PEI) molecule is precisely impregnated into cage-like pore of MIL-101(Cr), forming stable composites through strong coordination with unsaturated Cr acid sites in crystal internalization. The developed adsorbent exhibits low regeneration energy (39.6 kJ/mol_CO2), excellent cyclic stability (0.18% decay per cycle under dry CO₂ regeneration) high CO₂ adsorption capacity (4.0 mmol/g), and rapid adsorption kinetics (15 min for saturation at room temperature). These outstanding properties result from the unique electron-level interaction between amine and support, which prevents the dehydration of carbamate products. This work provides a feasible and promising strategy for cost-effective and sustainable CO₂ capture.

Physical sciences/Energy science and technology/Carbon capture and storage

Physical sciences/Engineering/Chemical engineering

Physical sciences/Chemistry/Materials chemistry/Metal–organic frameworks

Physical sciences/Chemistry/Green chemistry/Sustainability

Increasing CO₂ emissions have triggered a series of environmental catastrophes. To tackle this global challenge, over 100 nations have pledged to achieve carbon neutrality before 2050 and limit the average global temperature increase to 1.5 − 2°C above pre-industrial levels^1,2. Carbon capture, utilization and storage (CCUS) is the most promising and effective short-term solution for reducing CO₂ levels³. As projected by the International Energy Agency, CCUS is expected to cumulatively capture 27 billion tons of CO₂ by 2050⁴. Currently, the utilization of aqueous amine solutions for CO₂ capture from industrial flue gas is prevalent worldwide⁵. However, this technology is susceptible to oxidation, degradation, and secondary pollution, while also entailing significant energy consumption during amine solution regeneration⁶. Therefore, it is crucial to explore viable alternatives. Among these, porous solid adsorbents stand at the forefront of next-generation carbon capture technology owing to their non-corrosive nature and lower energy requirements for regeneration^7,8.

Porous solid adsorbents such as amine-functionalized adsorbents^9,10, zeolites¹¹, and metal-organic frameworks¹², have been widely investigated for CO₂ capture. Amine-functionalized adsorbents, in particular, are highly selective and resistant to water vapor, making them suitable for diverse applications scenarios involving humid and low-concentration CO₂ environments^2,13. Generally, they are prepared by impregnating polymeric amines such as polyethyleneimine (PEI) into porous materials^14,15, grafting aminosilanes¹⁶ onto support surfaces, or via in situ polymerization¹⁷ of amine monomers within supports. High loading of amines into large-porosity materials enable high CO₂ uptakes (≥ 3 mmol/g)^7,18. Previous studies have mostly focused on enlarging the support mesopores (2 − 50 nm) to improve amine loading¹⁹. However, aggregation frequently occurs due to poor amine dispersion within the support. Large pore spaces allow multiple polymeric amine molecules to impregnate a single pore, which greatly increases CO₂ diffusion resistance, resulting in slow CO₂ adsorption rates²⁰. These adsorbents often require higher temperatures to adsorb/desorb CO₂, leading to increased energy consumption and capture costs²¹. Another challenge is the rapid deactivation of amines during regeneration in a CO₂-rich atmosphere, a critical step for separating and recycling pure CO₂²². Most amine-functionalized adsorbents lose their CO₂ capture ability after just a few cycles because of dehydration reactions between the amines and CO₂ that form urea²³. Addressing amine deactivation is thus crucial when considering associated costs and scalability challenges²⁴. Therefore, the development of amine-functionalized adsorbents that minimize CO₂ diffusion restrictions during the adsorption process and exhibit strong resistance to deactivation in a CO₂-rich atmosphere is critical.

To enhance amine dispersion within supports and alleviate aggregation, using smaller micropores for amine impregnation can be an effective strategy²⁵. However, previous studies have highlighted a significant challenge in impregnating amines into micropores due to the close size match between amine molecules and micropores, which leads to considerable hindrance in the impregnation process^7,26,27. Inspirationally, designing an atom-level interaction amine-support system with two key characteristics holds promising potential. First, from the perspective of atomic model, dispersion-focused characteristic involves precise impregnation of individual amine molecules into specific micropores of the support material, possessing unique active sites, thus overcoming impregnation resistance. Second, stability-focused characteristic relates to the active sites enabling interactions with amines, resulting in the formation of a stable structure capable of resisting deactivation caused by urea formation^10,13. Most supports are unsuitable for this amine-support system because their primary function is enhancing the accessibility of amines. Metal-organic frameworks offer exceptional control over pore size and surface chemistries¹². Specifically, MIL-101(Cr) is noteworthy due to its high porosity, customized pore structure, suitable pore dimensions, and the interconnected cages within MIL-101(Cr) exhibiting a high density of Cr acid sites²⁸. This renders it a promising support for the specialized amine-support system. Despite limited reports on MIL-101(Cr) as a porous support for amine impregnation^14,29, specific designs of an amine-support system to address hindered diffusion and amine deactivation have been lacking.

Here, based on above designed amine-support system, we develop a novel approach of impregnating amines into crystal internalization (IACI) for uniform dispersions and highly stability. To demonstrate CCUS applicability, we measured the CO₂ adsorption performance, regeneration energy, and cyclic stability under different temperatures, pressures, and regeneration atmospheres. Additionally, we examined the structure and deactivation resistance of IACI using time-of-flight secondary-ion mass spectrometry and density functional theory (DFT). For the first time, we uncovered the electronic-level origin of IACI urea inhibition under a dry CO₂ regeneration atmosphere. Our findings demonstrate that IACI exhibits negligible diffusion resistance (saturation adsorption within a mere 15 min at 30°C), exceptional stability (decay rates of 0.11% and 0.18% per cycle in Ar and pure CO₂ atmospheres, respectively), and high adsorption capacity (4.0 mmol/g at 1 bar and 303 K).

Synthesis and characterization of IACI

The porous supports in MIL-101(Cr) were synthesized according to previous reports (Supplementary Fig. 1). It comprises cages with uniform 1.6-nm intra-crystalline micropores²⁸ (Fig. 1a) and numerous Lewis acid sites (Supplementary Fig. 2), making it an ideal candidate for the designed amine-support system. For pore-amine matching, we used various molecular-weight PEIs (Supplementary Note 1) and calculated their size (Supplementary Fig. 3). Among these amines, the PEI-1200 with planar dimensions of 1.3 nm is considered the optimal type, as it perfectly meets the requirement of precisely accommodating a single PEI-1200 molecule within each pore to pass through the hexagonal windows in MIL-101(Cr) (Fig. 1a). To confirm our hypothesis on the role of acid sites in MIL-101(Cr) cages as a driving force for the amine-support system within micropores, we synthesized PEI-functionalized MIL-101(Cr) using PEI-1200 and MIL-101(Cr) via moderate impregnation (Supplementary Fig. 4). As shown in Fig. 1b and c, porosity analysis was used to evaluate changes in Brunauer-Emmett-Teller specific surface areas (S_BET) and pore volumes at various impregnation time intervals. In the initial 5 min, S_BET decreased from 3468 to 1180 m²/g, with a reduction in pore volume from 1.75 to 0.63 cm³/g. This indicated that PEI-1200 could be rapidly loaded into the MIL-101(Cr) support and occupy a significant portion of its pore volume. After the initial stage, the amine loading slowed. From 5 minutes to 4 h, minimal changes in S_BET and the pore volume were observed. Hence, it took approximately 4 h to fill the remaining pore volume of 0.63 cm³/g, a rate 85 times slower than that during the first 5 min. Overall, PEI-1200 could penetrate the internal crystals (cages) of MIL-101(Cr) through the hexagonal windows, and the entire synthesis was completed within 4 h.

To confirm that PEI-1200 impregnated the MIL-101(Cr) cages, we investigated the relationship between loaded PEI and MIL-101(Cr). Figure 1d displays a series of high-magnification transmission electron microscopy (TEM) images, scanning TEM (STEM) images, and energy dispersive X-ray spectroscopy (EDX) maps of 50-nm-thick cuts of PEI-functionalized MIL-101(Cr) at specific time intervals. In each case, elemental Cr and N respectively corresponded to the support and the amines. All the sample cuts exhibited uniform distributions of Cr and N, with marked variations in the elemental ratios. During impregnation, the N-to-Cr element ratio reached 49:51 within the first 5 min, and then the rate of increase slowed, with a slight shift to 51:49 after an additional 5 min. This observation aligned with the porosity results and provided confirmation that PEI penetrated the crystal interior. It also verified the rapid diffusion and slow filling stages. To understand the amine distribution within the crystal throughout the loading process, EDX line scans were collected for all the slices (Supplementary Fig. 5). The N content remained nearly constant at different positions within the samples at each stage, indicating a uniform distribution of PEI-functionalized MIL-101(Cr) and the absence of aggregation. In contrast to the approach of impregnating amine-functionalized adsorbents via mesoporous supports^27,30, we observed amine impregnation into the microporous support driven by the abundance of acid sites in the internal crystals, establishing IACI in MIL-101(Cr).

CO₂ capture properties

The amine dispersion significantly impacts CO₂ uptake, adsorption rates, and regeneration energy consumption of amine-functionalized adsorbents³¹. To demonstrate the CO₂ capture properties of PEI-functionalized MIL-101(Cr), we conducted experiments under varied conditions. The CO₂ mass isothermal adsorption curves (Fig. 2a) show that PEI-functionalized MIL-101(Cr) exhibited rapid adsorption, reaching equilibrium within 15 min at different temperatures. Moreover, the CO₂ uptake decreased as the temperature increased (from 3.2 mmol/g at 30°C to 1.4 mmol/g at 90°C). Hence, there was no significant diffusion resistance and the CO₂ adsorption was thermodynamically controlled. The CO₂ volumetric isothermal adsorption curves (Fig. 2b) confirmed the isothermal mass adsorption curves, with CO₂ uptake reaching 4.0 mmol/g at 5°C, further highlighting the remarkable CO₂ adsorption performance of PEI-functionalized MIL-101(Cr). Additionally, differential scanning calorimetry indicated a regeneration energy consumption of approximately 39.6 kJ/mol of CO₂ (Fig. 2c), which was significantly lower than that required for other amine-functionalized adsorbents^15,32,33. Therefore, energy consumption associated with the adsorption-desorption cycles for CCUS would be significantly reduced. In summary, the PEI-functionalized MIL-101(Cr) exhibited rapid adsorption equilibrium at room temperature and required low energy for desorption.

Another crucial consideration is amine deactivation during the cyclic process. We conducted 90 cycles of CO₂ adsorption-desorption using a thermogravimetric analyzer regenerated under pure Ar or CO₂ (Fig. 2d). The PEI-functionalized MIL-101(Cr) demonstrated very stable cycles, with decay rates of only 0.11% and 0.18% per cycle in Ar and CO₂, respectively. This was in contrast to other adsorbents that were deactivated after a few cycles under CO₂ regeneration because of urea formation^34–36. It will allow PEI-functionalized MIL-101(Cr) to desorb high-purity CO₂ continuously under CO₂ regeneration atmospheres for subsequent utilization or storage. Table 1 lists comparisons of its CO₂ capture properties with other amine-functionalized adsorbents. It exhibited advantages in three aspects: i) less amine was consumed to capture the same amount of CO₂; ii) low regeneration energy consumption; and iii) stable cycles for a longer service life at a reduced cost.

The PEI-functionalized MIL-101(Cr) cyclic stability is likely attributed to the Lewis acid sites in the MIL-101(Cr), similar to those in Al₂O₃^12,16. Fourier-transform infrared spectra of MIL-101(Cr) and PEI-functionalized MIL-101(Cr) provided evidence for this relationship (Supplementary Fig. 6), by revealing a clear decrease in Cr site concentration with amine loading. Given the amine affinity of the support surface, a small fraction of amines could have remained outside the MIL-101(Cr) crystals. These would be less affected by Lewis acid sites, resulting in a lower cycle stability and possible urea formation. Thus, the faster decay rate under the CO₂ regeneration atmosphere relative to that under Ar atmosphere suggests low chemical deactivation via urea formation.

Table 1

**Comparison of CO**₂ **capture properties between PEI-functionalized MIL-101(Cr) and previously reported amine-functionalized adsorbents**
Adsorbents synthesis			Ability to sustainable CCUS							Consumption comparison
Amine loading	Support	Amine type and content	X^a (mmol/g)	T_a^b /^oC	t_a^b /min	T_d^c/ ^oC	t_d^c/ min	Purge gas^d	η^e/%	Amine consumption^f (g/mol_CO2)	Energy consumption^g (kJ/mol_CO2)	Cycle time /min	ref.
impregnated	mesoporous (SiO₂)	PEI (40 wt.%)	3.1	90	30	165	15	CO₂	8.65	20.40	~ 81.7	45	¹³
impregnated	mesoporous (SiO₂)	PEI (40 wt.%)	3.0	90	30	150	30	CO₂	8.13	20.00	~ 80.22	60	³⁷
impregnated	mesoporous (Al₂O₃)	PEI (55 wt.%)	3.0	90	30	165	15	CO₂	0.33	2.71	~ 92.6	45	¹⁰
impregnated	mesoporous (carbons)	PEHA^h (78 wt.%)	3.03	75	120	110	120	N₂	1.6	8.23	\	240	³⁸
grafted	mesoporous (SiO₂)	ethane-1,2-diamine (33%)	1.37	75	60	110	10	N₂	0.24	1.15	\	70	³⁹
grafted	mesoporous (geopolymer)	APTES^h (77 wt.%)	1.17	60	60	110	60	N₂	0.50	6.52	\	120	⁸
grafted	mesoporous (zeolite)	EDA^h (16 wt.%)	1.4	40	30	130	30	CO₂	1.05	2.37	\	60	⁴⁰
grafted	microporous (Mg₂(dobpdc))	een^h (36 wt.%)	3.1	80	20	140	10	CO₂	0.30	0.70	~ 74	30	⁹
IACI	microporous (MIL-101(Cr))	PEI (55 wt.%)	3.2	30	15	150	15	CO₂	0.18	0.56	39.6	30	This study
IACI	microporous (MIL-101(Cr))	PEI (55 wt.%)	3.2	30	15	150	15	Ar	0.11	0.38	39.6	30	This study
^aX refers to the maximum CO₂ uptake; ^bT_a and t_a represent the adsorption temperature and time, respectively; ^cT_d and t_d represent the desorption temperature and time, respectively; ^dThe regeneration atmosphere; ^eThe average inactivation efficiency per cycle; ^f The amount of amine reagent consumed per mole of CO₂ captured when the CO₂ capacity drops by half; ^gThe regeneration energy consumption at an adsorption and desorption cycle; ^hPEHA, APTES, EDA and een represent pentaethylenehexamine, (3-Aminopropyl)triethoxysilane, ethylenediamine, and N-ethylethylenediamine respectively. (More details on consumption comparison are provided in Supplementary Note 2)

Difference between IACI and amine impregnation outside crystals

To validate the proposed hypothesis, a series of experiments were performed to provide robust evidence for IACI stability. We first removed the amines on the MIL-101(Cr) surface via washing with deionized water. As shown in Fig. 3a, the washed samples exhibited a slight increase in S_BET from 7 m²/g to 14 m²/g relative to the unwashed samples. Pore-size distributions indicated almost no increase in 1–2 nm pores after washing (Fig. 3b), which suggested that the majority of amines remained bound to the internal supports, with only a negligible fraction being removed during the washing process. Furthermore, changes in consumed pore volume (from 1.76 to 1.71 cm³/g) and N content (17.2–16.9%) confirmed that the post-washed samples retained approximately 98% of the total amine content (Supplementary Fig. 7), with only washing out by 2%. For direct observation, TEM and EDX were used to image surface-adsorbed amines. Before washing, the PEI-functionalized MIL-101(Cr) displayed a series of aggregated particles with a size of approximately 500 nm (Fig. 3c). And the high-magnification TEM image shows a thin film composed of only N elements, indicating amine adsorbed on the surface, at the edge of the particles. In contrast, for the post-washed particles, the thin film composed of only N elements at the edge of the particles totally disappeared (Fig. 3d). These findings support the notion that in the designed amine-support system, less than 2% of the amines remained outside of the crystals, with the remaining 98% bound within the cages as IACI.

\(2RN{H_2}+{}^{{13}}C{O_2}\xrightarrow{{>135{{\kern 1pt} ^o}C}}{(RNH)_2}{}^{{13}}CO+{H_2}O{\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots {\kern 1pt} \ldots 1\)

Furthermore, to compare the anti-urea formation properties of these two types of amines, isotope labeling experiment was conducted during urea synthesis (Eq. 1). The unwashed PEI-functionalized MIL-101(Cr) was placed in a pure ¹³CO₂ atmosphere for 24 h (Supplementary Fig. 8) and synthesized urea was detected via time-of-flight secondary-ion mass spectrometry in a scanning electron microscope. As illustrated in Fig. 3e, the aggregate consisted of regular 500-nm octahedral particles, similar to MIL-101(Cr). Subsequently, the ¹³C intensity was observed in both planar and vertical directions of the aggregate (Fig. 3f, g). For the surface profile, a homogeneous distribution of ¹³C was observed on particle outlines, indicating that urea was formed in the adsorbent, and resulting in a small amount of residual adsorbed ¹³CO₂ consistent with Fig. 2d. For the depth profile, pronounced aggregation was observed, with the ¹³C signal concentrated on the sample surface, indicating that the urea was primarily formed via amines adsorbed on the particle surfaces. To further confirm this conclusion, the relationship between the ¹³C signal intensity and depth was plotted in Supplementary Fig. 9. In the first five frames, the ¹³C signal rapidly decreased from 1.5 to 0.5 and then fluctuated within a certain range. Based on these findings, the cyclic stability of IACI was much better than those amines that remained outside the crystals; hence there was a mechanism in IACI that inhibited urea formation.

Urea inhibition mechanism in IACI

DFT simulations were performed to model the urea inhibition mechanism in IACI; the model was simplified to reduce the computational complexity (see Methods). As shown in Fig. 4a, the coordinatively unsaturated metal sites (CUSs) acted as Lewis acid sites and exhibited a − 1.79 eV (− 173 kJ/mol) binding energy with the PEI primary amine. This indicated that amines readily entered and anchored within the crystal cages of the Cr (III) CUSs sites, resulting in the high IACI dispersion. Electron transfer most likely occurred when the acid and basic sites combined. Figure 4b shows the differential charge density distribution upon the binding of primary, secondary, and tertiary amines with Cr sites, indicating significant PEI electron density redistributions. (The yellow iso-surface represents increased charge density, while the light blue iso-surface represents decreased charge density). Interestingly, although electrons transferred from the amine molecules to the MIL-101(Cr) support, resulting in an overall positive charge, the calculation of Bader charges revealed that not all amine groups lose electrons. Instead, a new electronic rearrangement phenomenon occurs: the primary amines lost 0.023 e⁻ while the secondary amines gained 0.3 e⁻ (Fig. 4c). This led to discernible fluctuations in the electron cloud density, necessitating solid-state ¹³C nuclear magnetic resonance (NMR) to analyze the chemical shift of the amine nucleus⁴¹. Relative to solid-state ¹³C NMR spectra of 55%PEI@SiO₂, the proportions of primary and secondary amines in PEI-functionalized MIL-101(Cr) decreased to 27% and 37%, respectively (Supplementary Fig. 10).

The effect of electron rearrangement on urea inhibition is shown schematically in Fig. 5a. When the primary amine loses an electron, we simplify this state to that original R group replaced by an electron-withdrawing group R' (δ−). In mechanism A, the electron withdrawal results in a partial positive charge (δ+) on the amine nitrogen (N), which then attracts electron density from the N − C bond, creating a positive charge (δ+) on the carbon (C) atom. This, in turn, decreases interactions between ammonium carbamates, inhibiting dehydration that forms urea. The effect of electron gain in the secondary amines is depicted in mechanism B. When the R group is substituted by an electron-donating group R* (δ+), this results in a partial negative charge (δ−) on the secondary amine nitrogen. This facilitates electron donation, where the electron density of the N − C bond shifts towards the carbon (C) atom, leading to increased negative charge (δ−). Consequently, the covalent N-H bond is strengthened, inhibiting breakage during dehydration. Mechanism A was confirmed with reaction-energy diagrams from DMol3 calculations, where the electron-withdrawing group − CCl₃ as R' resulted in a 2.42 eV transition-state energy, ten times that with the normal R group (CH₃), confirming the significant inhibitory effect on urea formation (Fig. 5b). Similarly, mechanism B was validated by considering − SiH₃ as R* in Fig. 5c, where the transition-state energy of the intermediate species increased to 5.65 eV, indicating a pronounced inhibitory effect.

Amine-functionalized adsorbents possess desirable CO₂ capture performance with easy preparation, but these face the challenges of poor amine dispersion and rapid deactivation, resulting in inefficient and substantial costs. To address these issues, we developed an interaction amine-support system that exhibited negligible diffusion resistance, a stunning high stability among amine-functionalized adsorbents, low regeneration energy consumption, and high CO₂ adsorption capacity. While some previous amine-impregnated materials have also achieved excellent properties in some respects, there typically tend to be mediocre in other respects^7,42. These adsorbents are not compatible in adsorption capacity, adsorption rate, regeneration energy consumption and stability, such as high adsorption capacity but low adsorption rate^20,43, low diffusion resistance but poor stability^23,44, high stability but moderate CO₂ uptake^45,46, etc. Furthermore, we firstly elucidated that the interaction of amines with Lewis acid sites would induce a special electron rearrangement within aminopolymer, enabling a notable urea inhibition in amine-functionalized adsorbents. Therefore, the innovative amine-support system opens up opportunities to design high-efficiency and stable amine-functionalized adsorbents for CO₂ capture. Finally, further research may be to expand the range of specialized amine-support systems, particularly for small molecule amines and readily available porous supports.

Materials. The metal–organic frameworks MIL-101(Cr) was synthesized based on the previous literature procedure⁴⁷. First, 800 mg of Cr (NO₃)₃·9H₂O and 332 mg of H₂BDC (terephthalic acid) were blended in 14 mL of deionized (DI) water. After shaking fully, the mixture was added 1 mL of 40% HF (hydrofluoric acid). The solution was placed in a Teflon-lined autoclave and kept at 220°C in an oven for 8 h, then naturally cooling to room temperature. After the synthesis, the MIL-101(Cr) was separated using a centrifuge and washed repeatedly with MeOH and N, N-Dimethylformamide (DMF) for three times, respectively. The MIL-101(Cr) support was finally obtained by drying in a high vacuum (–0.1 Mpa) at 150°C for 12 h. In the synthetic process of amine-functionalized adsorbent, the physical impregnation method was adopted to load amine into the MIL-101(Cr) support, and an airflow control procedure was implemented to control the amine impregnation process. First, 0.2 g of MIL-101(Cr) support and 0.24 g of polyethyleneimine were separately placed in two 25 mL beakers and added with 10 mL of methanol. Subsequently, the two mixtures were sonicated at room temperature (25°C) for 30 min to fully disperse. Then the two mixtures were mixed and stirred in the fume hood for 4 h with a constant airflow of 1.5 m/s, allowing the methanol to evaporate. Finally, the mixture was dried in a vacuum at 60°C for 12 h to ensure the methanol removing completely. For comparison, the same synthetic process of amine-functionalized was also treated by replacing the MIL-101(Cr) with same weight SiO₂, under otherwise identical conditions.

Characterization. The N₂ adsorption-desorption analysis for all samples was carried out at − 196°C by a gas adsorption analyzer (Micromeritics, ASAP 2460, USA). The degassing processes of MIL-101(Cr) supports and PEI-functionalized MIL-101(Cr) adsorbents were conducted at 150°C for 6 h and 60°C for 18 h, respectively. The specific surface area (S_BET) was calculated according to the Brunauer-Emmett-Teller (BET) equation, whereas the total pore volume (V_pore) and the pore size distribution were determined by the Single Point model and Density-Functional-Theory (DFT) model, respectively. The micromorphology of samples was observed using a field emission scanning electron microscope (SEM; Zeiss, Merlin, Germany) and a transmission electron microscope (TEM, TALOSF200, FEI, USA). The TEM instrument was equipped with a high-angle annular dark-field detector and a bright-field detector (1nA@1nm, 200 kV). The distribution of ¹³C in PEI-functionalized MIL-101(Cr) adsorbents was analyzed using an ultra-high resolution scanning electron microscope equipped with time-of-flight secondary ion mass spectrometry (TOFSIMS, GAIA3 GMU Model, Czech Republic). The sample structures and characteristics, including phase volume fraction, grain size, and microstructural features, were determined using electron backscattering diffraction (EBSD) analysis. The regeneration energy consumption of the PEI-functionalized MIL-101(Cr) adsorbents was measured using a differential scanning calorimetry analyzer (Discovery DSC, USA), which was tested under the ramping temperature of 30–150°C (20°C/min) and a CO₂ flow of 50 mL/min. The solid-phase ¹³C spectra were acquired on a 600 MHz nuclear magnetic resonance (NMR) spectrometer (Agilent, 600 MHz, America) with an operating speed of 10 kHz. The ¹³C magic-angle-spinning (MAS) NMR spectra were obtained under the operation parameters of an acquisition time of 4 µs, a recycle delay of 5 s and 300 scans. Fourier transform infrared (FT-IR) spectroscopy analysis was performed on a Nicolet iS50 spectrometer (Thermo Fisher Scientific, USA).

CO ₂ adsorption tests. The CO₂ uptakes of PEI-functionalized MIL-101(Cr) adsorbents were measured using a thermogravimetric analyzer. First, 15–20 mg adsorbent was placed in an alumina crucible and heated from room temperature to 150°C at a heating rate of 20°C/min, and then maintained 30 min at 150°C for degassing under an Ar flow of 50 mL/min. After degassing, the temperature was decreased to the adsorption temperature of 30–90°C at a rate of − 5°C/min. Next, the Ar flow was switched to a CO₂ flow of 50 mL/min and maintained 60 min for adsorbing CO₂. The volumetric adsorption isotherms for CO₂ were conducted using a gas adsorption analyzer (Micromeritics, ASAP 2460, USA) at 5–90°C. In a general procedure, ~ 150 mg adsorbent was loaded into the glass analysis tube. The sample was heated under vacuum at 60°C for 18 h to remove moisture and adsorbed CO₂ from the air. Subsequently, the sample was backfilled with N₂ before being transferred to the analysis port. Finally, the sample was degassed under vacuum conditions for 4 h, and a high purity CO₂ flow (99.998%) was introduced for CO₂ adsorption. For the cyclic stability tests, 15–20 mg adsorbent was first degassed under the same conditions of the above degassing process. Once the temperature was dropped (–5°C/min) to 30°C, the Ar atmosphere transformed into the CO₂ (50 mL/min), and the CO₂ adsorption process began as well as maintained for 15 min. Then the temperature was heated (10°C/min) to the expected temperature of 150°C, and a pure CO₂ purge (50 mL/min) was introduced to regenerate the adsorbent for 15 min at 150°C. Subsequently, a new cycle was conducted after cooling the temperature to 30°C (–5°C/min).

Computational details. For all density functional theory (DFT) calculations within the generalized gradient approximation (GGA) using the PBE formulation, we utilized the Vienna Ab Initio Package (VASP)⁴⁸. The projected augmented wave (PAW) potentials⁴⁹ were employed to represent the ionic cores and account for valence electrons, while a plane wave basis set with a kinetic energy cutoff of 400 eV was used. To consider partial occupancies of the Kohn-Sham orbitals, we employed the Gaussian smearing method with a width of 0.05 eV. The electronic energy was considered self-consistent when the energy change was less than 10^− 5 eV. For geometry optimization, convergence was achieved when the force change was below 0.02 eV/Å. To account for dispersion interactions, we employed Grimme's DFT-D3 methodology⁵⁰. The MIL-101(Cr) cluster comprises of 3 Cr, 1 O and 6 ligands, contained in a cubic box of 22 Å. During the optimization of the structure, we used the Γ point in the Brillouin zone for k-point sampling, allowing all atoms to relax. The adsorption energy (E_ads) of adsorbate A was calculated using the formula E_ads = E_A/surf – E_surf – E_A(g), where E_A/surf represents the energy of adsorbate A adsorbed on the surface, E_surf is the energy of the clean surface, and E_A(g) is the energy of an isolated A molecule enclosed in a cubic periodic box with a side length of 20 Å and a 1 × 1 × 1 Monkhorst-Pack k-point grid for Brillouin zone sampling. This approach ensures the dissimilarity to previous literature while maintaining alignment with the Methods style of Nature Sustainability.The DMol3 code was employed for performing electrostatic potential computations. The BLYP exchange-correlation functional within the generalized gradient approximation (GGA) was used to describe electron interactions. For the expansion of molecular orbitals, we employed the double numerical polarization (DNP) function basis set with an orbital cutoff of 3.6 Å. All core electrons were included in the calculations. During geometry optimizations, relaxation of atomic positions was allowed. Convergence thresholds for energy change, maximum force, and maximum displacement between optimization cycles were set to 10^− 5 Hartree, 0.002 Hartree/Å, and 0.005 Å, respectively. The electronic self-consistent field convergence threshold was set to 10^− 6 Hartree. To search for minimum energy paths, we utilized the climbing image nudged elastic band method (CI-NEB).

Competing interests

The authors declare no competing interests.

Additional information

Correspondence and requests for materials should be addressed to Zuotai Zhang.

Author contributions

The project was conceived and supervised by Z.Z. The amine-support system was designed by Z.Z. and X.S., who also proposed the concept of impregnation Amine of Crystal Internalization (IACI). Experimental work and data analysis were carried out by X.S. and X-H.S., while density functional theory (DFT) computational simulations were performed by H.W. and F.Y. The manuscript was primarily written by X.S. and Z.Z. Editing and supervision of the research were provided by X.S., X-H.S., W.H., J.H., and Z.Z.

Acknowledgements

The authors gratefully acknowledge the generous financial support provided by the National Science Fund for Distinguished Young Scholars (Grant No. 52225407) and the National Natural Science Foundation of China (Grant No. 22008104). Additionally, the authors would like to express their gratitude to the Shenzhen Science and Technology Innovation Committee for their financial support through grants JSGG20210713091810035 and KCXFZ20211020174805008. These funding sources have played a crucial role in facilitating the research presented in this paper.

Xia, Y. et al. Future reductions of China's transport emissions impacted by changing driving behaviour. Nat. Sustain., (2023).
Sanz-Pérez, E. S., Murdock, C. R., Didas, S. A. & Jones, C. W. Direct capture of CO₂ from ambient air. Chem. Rev. 116, 11840–11876 (2016).
Rekker, S. et al. Evaluating fossil fuel companies’ alignment with 1.5°C climate pathways. Nat. Clim. Chang. 13, 927–934 (2023).
Chen, D. & Jiang, M. Assessing the socio-economic effects of carbon capture, utility and storage investment from the perspective of carbon neutrality in China. Earth's Future. 10, (2022).
Hu, X. E. et al. NMR techniques and prediction models for the analysis of species formed in CO₂ capture processes with Amine-Based Sorbents: A Critical Review. Acs Sustain. Chem. Eng. 8, 6173–6193 (2020).
Dutcher, B., Fan, M. & Russell, A. G. Amine-Based CO₂ Capture technology development from the beginning of 2013—a review. Acs Appl. Mater. Interfaces. 7, 2137–2148 (2015).
Siegelman, R. L., Kim, E. J. & Long, J. R. Porous materials for carbon dioxide separations. Nat. Mater. 20, 1060–1072 (2021).
Chen, H., Dong, S., Zhang, Y. & He, P. Robust structure regulation of geopolymer as novel efficient amine support to prepare high-efficiency CO₂ capture solid sorbent. Chem. Eng. J. 427, 131577 (2022).
Choe, J. H. et al. Functionalization of diamine-appended MOF-based adsorbents by ring opening of epoxide: long-term stability and CO₂ recyclability under humid conditions. J. Am. Chem. Soc. 144, 10309–10319 (2022).
Shen, X. et al. Biogas upgrading via cyclic CO₂ adsorption: application of highly regenerable PEI@nano-Al₂O₃ adsorbents with anti-urea properties. Environ. Sci. Technol. 55, 5236–5247 (2021).
Du, P. et al. Control of zeolite framework flexibility for ultra-selective carbon dioxide separation. Nat. Commun. 13, (2022).
Lin, J. et al. A scalable metal-organic framework as a durable physisorbent for carbon dioxide capture. Science. 374, 1464–1469 (2021).
Li, C. et al. Highly efficient and stable PEI@Al₂O₃ adsorbents derived from coal fly ash for biogas upgrading. Chemical Engineering Journal. 409, 128117 (2021).
Darunte, L. A., Oetomo, A. D., Walton, K. S., Sholl, D. S. & Jones, C. W. Direct air capture of CO₂ using amine functionalized MIL-101(Cr). Acs Sustain. Chem. Eng. 4, 5761–5768 (2016).
Li, K., Jiang, J., Yan, F., Tian, S. & Chen, X. The influence of polyethyleneimine type and molecular weight on the CO₂ capture performance of PEI-nano silica adsorbents. Appl. Energy. 136, 750–755 (2014).
Qi, G., Fu, L. & Giannelis, E. P. Sponges with covalently tethered amines for high-efficiency carbon capture. Nat. Commun. 5, (2014).
Zhang, Z., Zhang, L., Gao, Q., Sun, N. & Wei, W. Robust nitrogen-doped microporous hollow carbon spheres for energy-efficient CO₂ capture from flue gas. J. Co2 Util. 75, 102570 (2023).
Nguyen, T. S., Dogan, N. A., Lim, H. & Yavuz, C. T. Amine chemistry of porous CO₂ adsorbents. Acc. Chem. Res. 56, 2642–2652 (2023).
Wang, L. et al. Amine-modified ordered mesoporous silica: The effect of pore size on CO₂ capture performance. Appl. Surf. Sci. 324, 286–292 (2015).
Rong, X. et al. Liquid marble-derived solid-liquid hybrid superparticles for CO₂ capture. Nat. Commun. 10, (2019).
Raganati, F., Miccio, F. & Ammendola, P. Adsorption of carbon dioxide for post-combustion capture: A Review. Energy Fuels. 35, 12845–12868 (2021).
Sayari, A., Heydari-Gorji, A. & Yang, Y. CO₂-induced degradation of amine-containing adsorbents: reaction products and pathways. J. Am. Chem. Soc. 134, 13834–13842 (2012).
Lashaki, M. J., Khiavi, S. & Sayari, A. Stability of amine-functionalized CO₂ adsorbents: a multifaceted puzzle. Chem. Soc. Rev. 48, 3320–3405 (2019).
Patel, H. A., Byun, J. & Yavuz, C. T. Carbon dioxide capture adsorbents: chemistry and methods. Chemsuschem. 10, 1303–1317 (2017).
Zhu, X. et al. Recent advances in direct air capture by adsorption. Chem. Soc. Rev. 51, 6574–6651 (2022).
Shi, X. et al. Sorbents for the direct capture of CO₂ from ambient air. Angewandte Chemie International Edition. 59, 6984–7006 (2020).
Zhang, S. et al. Design and optimization of solid amine CO₂ adsorbents assisted by machine learning. Acs Sustain. Chem. Eng. 10, 13185–13193 (2022).
Ferey, G. et al. A chromium terephthalate-based solid with unusually large pore volumes and surface area. Science. 309, 2040–2042 (2005).
Zhong, R. et al. Amine-grafted MIL-101(Cr) via double-solvent incorporation for synergistic enhancement of CO₂ uptake and selectivity. Acs Sustain. Chem. Eng. 6, 16493–16502 (2018).
Moon, H. J. et al. Understanding the impacts of support–polymer interactions on the dynamics of poly(ethyleneimine) confined in Mesoporous SBA-15. J. Am. Chem. Soc. 144, 11664–11675 (2022).
Li, W., Fu, D. & Pan, J. Adsorption of CO₂ in flue gas by polyethylenimine-functionalized adsorbents: effects of the support morphology and dispersion mode of amines. Energy Fuels. 37, 15879–15895 (2023).
McDonald, T. M. et al. Capture of carbon dioxide from air and flue gas in the alkylamine-appended metal-organic framework mmen-Mg₂(dobpdc). J. Am. Chem. Soc. 134, 7056–7065 (2012).
Zhou, C. et al. Amine-functionalized mesoporous monolithic adsorbents for post-combustion carbon dioxide capture. Chem. Eng. J. 413, 127675 (2021).
Shen, X., Yan, F., Li, C., Zhang, Z. & Zhang, Z. A green synthesis of PEI@nano-SiO₂ adsorbent from coal fly ash: selective and efficient CO₂ adsorption from biogas. Sustain. Energ. Fuels. 5, 1014–1025 (2021).
Liu, L. et al. Alkyl-functionalization of (3-Aminopropyl)triethoxysilane-grafted zeolite beta for carbon dioxide capture in temperature swing adsorption. Chem. Eng. J. 382, 122834 (2020).
Jeon, S., Min, J., Kim, S. H. & Lee, K. B. Introduction of cross-linking agents to enhance the performance and chemical stability of polyethyleneimine-impregnated CO₂ adsorbents: Effect of different alkyl chain lengths. Chem. Eng. J. 398, 125531 (2020).
Meng, Y. et al. Comprehensive study of CO₂ capture performance under a wide temperature range using polyethyleneimine-modified adsorbents. J. CO₂ Util. 27, 89–98 (2018).
Peng, H. et al. Chitosan-derived mesoporous carbon with ultrahigh pore volume for amine impregnation and highly efficient CO₂ capture. Chem. Eng. J. 359, 1159–1165 (2019).
Bai, F. et al. Amine functionalized mesocellular silica foam as highly efficient sorbents for CO₂ capture. Sep. Purif. Technol. 299, 121539 (2022).
Kim, C., Cho, H. S., Chang, S., Cho, S. J. & Choi, M. An ethylenediamine-grafted Y zeolite: a highly regenerable carbon dioxide adsorbent via temperature swing adsorption without urea formation. Energy Environ. Sci. 9, 1803–1811 (2016).
Barfield, M. & Fagerness, P. Density functional theory GIAO studies of the C-13, N-15, and H-1 NMR chemical shifts in aminopyrimidines and aminobenzenes: Relationships to electron densities and amine group orientations. J. Am. Chem. Soc. 119, 8699–8711 (1997).
Lin, L. et al. Characteristics, application and modeling of solid amine adsorbents for CO₂ capture: A review. J. Environ. Manage. 325, 116438 (2023).
Chen, C., Xu, H., Jiang, Q. & Lin, Z. Rational design of silicas with meso-macroporosity as supports for high-performance solid amine CO₂ adsorbents. Energy. 214, 119093 (2021).
Jiao, J., Cao, J., Xia, Y. & Zhao, L. Improvement of adsorbent materials for CO₂ capture by amine functionalized mesoporous silica with worm-hole framework structure. Chem. Eng. J. 306, 9–16 (2016).
Choi, W. et al. Epoxide-functionalization of polyethyleneimine for synthesis of stable carbon dioxide adsorbent in temperature swing adsorption. Nat. Commun. 7, 12640 (2016).
Li, X. et al. Amine-impregnated porous carbon–silica sheets derived from vermiculite with superior adsorption capability and cyclic stability for CO₂ capture. Chem. Eng. J. 464, 142662 (2023).
Rim, G. et al. Sub-ambient temperature direct air capture of CO₂ using amine-impregnated MIL-101(Cr) enables ambient temperature CO₂ Recovery. Jacs Au. 2, 380–393 (2022).
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B. 59, 1758–1775 (1999).
Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurateab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. The Journal of Chemical Physics. 132, (2010).

There is NO Competing Interest.

SI.docx

Download PDF