Efficient CO2 Desorption Catalysts: from Material Design to Kinetics Analysis and Application Evaluation

Lei Xing , Zhen Chen , Guoxiong Zhan , Zhoulan Huang , Lidong Wang , Junhua Li

Engineering ›› 2025, Vol. 49 ›› Issue (6) : 251 -259.

PDF (3290KB)
Engineering ›› 2025, Vol. 49 ›› Issue (6) :251 -259. DOI: 10.1016/j.eng.2024.08.024
Research
Article

Efficient CO2 Desorption Catalysts: from Material Design to Kinetics Analysis and Application Evaluation

Author information +
History +
PDF (3290KB)

Abstract

Catalytic amine-solvent regeneration has been validated as an energy-saving strategy for CO2 chemisorption by boosting reaction kinetics under mild conditions. The upscale performance evaluation and long-term durability are indispensable steps for industrial application but have been scarcely reported thus far. Here, we report a ZrO2/Al2O3 pack catalyst that possesses strong metal oxide-support interactions, a porous structure, active and stable Zr–O–Al coordination, promoted proton transfer and a 40.7% decrease in the energy activation of carbamate decomposition, which significantly accelerates CO2 desorption kinetics. The upscale experiment and cost evaluation based on industrial flue gas revealed that the use of packing catalysts can reduce energy consumption by 27.56% and optimize the overall cost by 10.49%. The active sites present excellent stability in alkaline solvents. This work is the first to investigate the ability of high-technology readiness (technology readiness level at 6 (TRL 6)) for catalytic amine-solvent regeneration, providing valuable insights for potential applications involving efficient CO2 capture with catalyst assistance.

Graphical abstract

Keywords

CO2 capture / Catalytic CO2 desorption / Upscale evaluation / Process simulation / Technical economics

Cite this article

Download citation ▾
Lei Xing, Zhen Chen, Guoxiong Zhan, Zhoulan Huang, Lidong Wang, Junhua Li. Efficient CO2 Desorption Catalysts: from Material Design to Kinetics Analysis and Application Evaluation. Engineering, 2025, 49(6): 251-259 DOI:10.1016/j.eng.2024.08.024

登录浏览全文

4963

注册一个新账户 忘记密码

1. Introduction

Extensive greenhouse gas carbon dioxide (CO2) emissions from fossil fuel consumption are the major source of CO2 and lead to global warming [1]. CO2 chemisorption from flue gas via amine scrubbing contributes to CO2 reduction because fossil fuel combustion has remained significant for a long time [2], [3]. In the amine scrubbing process, CO2 reversibly reacts with the –NH2 of the amine through the generation and rupture of Csingle bondN bonds. Notably, Csingle bondN bond rupture is a thermodynamic-driven reaction that requires significant amounts of high-enthalpy steam for thorough and fast regenerating CO2-spent solvents, resulting in energy-intensive and costly CO2 capture (3.6–4.0 gigajoule per tonnes CO2 (GJ·tCO2−1) for the benchmarking 30 wt% monoethanolamine (MEA) solvent) [4], [5]. Consequently, achieving amine scrubbing in a sustainable, commercially viable, and affordable way is highly important for decarbonization. Some advanced amine blends or biphasic solvents are advantageous for reducing regeneration to 2.0–2.8 GJ·tCO2−1; however, the practical application remains challenging due to severe amine loss, weak water balance robustness, and low phase splitting ratios [6], [7], [8], [9], [10].

Heterogeneous catalysis has been validated as an efficient and widely used strategy for boosting reaction kinetics under relatively mild conditions, typically for thermodynamic-driven reactions at high temperature or pressure [11], [12], [13]. Continuous efforts in catalyst design have greatly advanced energy-intensive industrial processes, such as, synthetic ammonia, alkylation/olefin synthesis, pyrolysis/cracking processes, and so forth [14], [15]. Recently, the regeneration of amine-based solvents assisted by solid acid catalysts (SACs) has become recognized as a burgeoning route to effectively reduce energy consumption. High-performance SACs with accessible acid sites can serve as proton donors or electron acceptors to accelerate proton transfer and weaken C–N bonds [16], [17]. Acid materials, such as metal oxides (e.g., Al2O3, TiO2, ZrO2, NiO, and CeO2), transition metal oxide-modified molecular sieves (e.g., zeolite socony mobil-5 (ZSM-5), santa barbara amorphous-15 (SBA-15), mobil composition of matter-41 (MCM-41), and montmorillonite) and even metal–organic framework or covalent–organic framework derivatives, present excellent acceleration of the CO2 desorption rate by more than 30%, an increase in the CO2 cyclic capacity of approximately 40% and a decrease in the operational temperature/pressure [18], [19], [20], [21]. Notably, the present studies [18], [19] are in the powder catalyst synthesis phase, while the development of advanced SACs with low cost, high activity, durability, and feasibility is challenging. Moreover, the combination of SACs and packed towers is considerably important but has received little attention, as has the long-term durability and technical economy; thus, this strategy is still in experimental proofs of concept (technology readiness level at 3 (TRL 3)), which are far from industrial conditions [22]. Thus, the preparation and performance evaluation of packed SACs at the laboratory scale, covering these potential issues for actual systems and practical applications, plays a pivotal role in increasing the maturity of this technology.

ZrO2 and γ-Al2O3, which are inexpensive and have excellent durability, are promising candidates for preparing packing catalysts. The integration of Zr species with a high valence in the γ-Al2O3 skeleton can lead to an increase in the positive charge of Al atoms and improve their adsorption and activation for nucleophilic H2O molecules and proton transfer, while Zr sites with improved electron intensity can lead to a high affinity for carbamate adsorption and Csingle bondN bond activation [23], [24]. To this end, a robust and active packing catalyst with high surface activity and acidity was designed by modifying stable γ-Al2O3 with environmentally friendly ZrO2 (Fig. 1 and Fig. S1 (a) in Appendix A). The strong interaction between the metal oxide and support through Zr–O–Al coordination results in the uniform and stable immobilization of active Zr species in the robust skeletal framework of γ-Al2O3 and drives an improved surface structure, electron intensity, acidity, and hydrophilicity of γ-Al2O3. We described the long-life CO2 absorption and catalytic desorption cycle of MEA using a ZrO2/Al2O3 catalyst and validated its feasibility via laboratory- and bench-scale demonstrations under industrial conditions. This packing catalyst enables greater proton intermediate transfer to reduce the energy barrier by 40.7% for Csingle bondN bond rupture, thus decreasing energy consumption by 27.56% and overall cost by 10.49%. This strategy with high TRL 6 can support amine scrubbing more technically and economically.

2. Methods

2.1. Catalyst synthesis

ZrO2/Al2O3 catalysts were synthesized using a straightforward hydrothermal method, as illustrated in Fig. 2(a). Specifically, γ-Al2O3 (10 g) and urea (3 g) were introduced into a ZrOCl2·8H2O (0.01–0.4 mol·L−1, 20 mL) solution, which was then heated to 180 °C for 20 h in a 50 mL Teflon reactor. Subsequently, the obtained materials were rinsed three times with deionized water, dried at 120 °C for 12 h, and then calcined in air at 500 °C for 6 h at a heating rate of 5 °C·min−1. Various concentrations of γ-Al2O3 with different ZrO2 dosages were prepared and designated as ZrO2-x/Al2O3, where x represents 0.2, 0.4, 1.0, 2.0, 4.0, or 8.0 wt%. The synthesis procedure for the M3O4/Al2O3 (M = Co, Mn, Fe) samples followed the same method as that for ZrO2/Al2O3.

2.2. Catalytic CO2 absorption and desorption procedure

CO2 desorption procedure. Catalytic CO2 desorption from CO2-rich MEA solvent was carried out on a customized batch reactor as previously described. Typically, CO2-rich solvent (CO2 loading reached (2.75 ± 0.05) mol·L–1) was obtained under a bubble of flue gas (12 vol% CO2 in N2, 1 L·min–1) in MEA solution (30 wt%, 150 mL) at 40 °C and atmospheric pressure. Subsequently, the solvent was heated to (90 ± 0.1) °C with a 1.0 wt% ZrO2-x/Al2O3 catalyst. The desorbed pure CO2 was condensed through a circulating water bath ((10 ± 0.1) °C), and the desorption rate (vCO2, mol·min–1) was measured through an online mass flow detector. The CO2 desorption amount (nCO2(t), mol·L−1) was calculated by converting the desorption rate, as shown in Eq. (1), where VM is molar volume of gas (L·mol−1), Vsol is the volume of solution (L) and t is the reaction time (min).

nCO2(t)=VM-1·Vsol-1·0tvCO2dt

Ea calculation. According to Eq. (2), CO2 desorption from carbamate (RNHCOO) and protonated amine (RNH3+) is a typical secondary reaction, where R is –OHCH2CH2, t is time, and the rate constant (k) can be obtained as Eq. (3), where CMEACOO- is the change in RNHCOO concentration and 0.56 (mol CO2 per mol amine) is the initial concentration of RNHCOO. The Ea can be calculated from the Arrhenius equation at different temperatures for the catalytic and blank samples, as shown in Eq. (4), where A is pre-exponential factor, R0 is molar gas constant (KJ·(mol·K)−1), and T is reaction temperature (°C).

MEACOO-+MEAH+CO2+2MEA
10.56-CMEACOO-=10.56+k·t
Ink=InA-EaR·T

Bench-scaled CO2 desorption procedure. Continuous CO2 absorption and desorption from CO2-rich MEA solvent with ZrO2-2/Al2O3 catalysts was carried out on a customized bench-scale system, as shown in Fig. 1. The scale-up experimental system was composed of two stainless steel kettles (left-absorber and right-stripper) equipped with a packing column (ϕ60 mm × 2000 mm) filled with Dixon packing (800 g, ϕ4 mm × 4 mm, 316 L). Under catalytic desorption conditions, the packing in the stripper was replaced by a uniform mixture of Dixon packing (600 g) and a scattered ZrO2-2/Al2O3 catalyst (200 g), which did not greatly change the original flow form of the solvent or influence the pressure drop (Fig. S1(b) and Table S1 in Appendix A), thus enabling the comparable evaluation of the blank and catalytic conditions. The volume of amine solvent in this system was 11 L. During the bench-scale test, simulated flue gas (12% CO2 and 10% O2 in N2, 20 L·min–1) was piped into the bottom of the absorber and reverse contacted and reacted with the amine solution from the top of the absorber in the filler. The decarbonized flue gas was condensed through a circulating water bath ((10 ± 0.1) °C), and the concentration of outlet CO2 was measured by a UV-induced flue gas analyzer. After CO2 absorption, the CO2-rich MEA solution in the absorber was pumped into the top of the stripper through a heat exchanger with a cyclic flow rate of 50 mL·min–1. The amine solvent was gradually heated to 95–120 °C to desorb CO2 by regulating the pressure of the stripper and the heat from the circulating oil bath on the stripper side. The desorbed CO2 was released from the top of the stripper, and the desorption rate (vCO2, L·min–1) was also measured online by a mass flow detector. The regenerated CO2-lean amine solvent was pumped into the top of the absorber through a heat exchanger. In this system, the heat duty (H(t)) was evaluated based on the monitored electrical energy (EL), as shown in Eq. (5).

H(t)=EL(t)0tvCO2dt

2.3. Simulation and economic–environmental evaluation

The electrolyte nonrandom two-liquid equation was employed as the thermodynamic model for the process simulation. The main components in the flue gas included 12% CO2, 76% N2, 5% O2, and 7% H2O, with a flow rate of 1.2 × 105 N·m3·h−1. The flow rate range of the absorption solvent was approximately 300 to 550 t·h−1. The inlet temperature of the flue gas and absorption solvent was 40 °C. The range of the operational regeneration pressure was approximately 1.0–2.0 bar (1 bar = 100 000 Pa). The absorption pressure was set to atmospheric pressure. The diameters of the absorption and regeneration columns are 7.5 and 6 m, respectively. The CO2 capture process was simulated using the rate-based RadFrac module. In the process simulation, the reaction kinetic parameters Ea of the blank and catalytic regeneration columns were regulated to simulate the catalytic decomposition of carbamate by Aspen Plus software, while Ea was calculated via regression of the reaction kinetics experimental data. Process economic analysis was employed to investigate profit performance by checking the index total capture cost (TCC; USD per tonnes CO2 (USD·(tCO2)−1). The TCC is made up of the annual capital cost (ACC; USD·(tCO2)−1) and total operating cost (TOC; USD·(tCO2)−1). Furthermore, the net CO2 emission reduction (NCOER) was also investigated by process carbon accounting. The detailed calculations of the TCC and NCOER can be found in Appendix A.

3. Results and Discussion

3.1. Synthesis and characterization of pelletized ZrO2-x/Al2O3

To verify the practicability of the proposed catalytic technology for the CO2 desorption reaction, we designed pelletized catalysts composed of γ-Al2O3 and metal oxide species. A series of ZrO2-x/Al2O3 and M3O4/Al2O3 (M =Co, Mn, Fe) samples were synthesized via a hydrothermal method (Fig. 2(a)). Various synthetic sequences (e.g., active site dosages ranging from 0.1 to 8.0 wt% and calcination temperatures ranging from 100 to 1100 °C), which can cause noticeably different surface structures and acidity properties, were regulated prior to identifying the optimal synthesis strategy by comparing the structure and catalytic CO2 desorption using 5 mol·L−1 MEA solvent. As depicted in Fig. S2 in Appendix A, an increase in CO2 desorption kinetics was achieved using the above four different metal catalysts compared with that of the blank sample. In contrast, the introduction of Zr species resulted in predominant CO2 desorption, the preferred calcination temperature was 500 °C, and the preferred catalyst concentration during the test was 1.0 wt%, suggesting a greater efficiency of Zr species in alkaline environments (Figs. S3 and S4 in Appendix A). Accordingly, subsequent property tests were performed on the ZrO2-x/Al2O3 sample prepared after calcination at 500 °C.

To investigate the activity of ZrO2-x/Al2O3, structural characterizations of various catalysts with different ZrO2 dosages were performed. Scanning electron microscopy (SEM; JSM-7401, JEOL, Japan) revealed a homogeneous nanosheet-like structure with an average size of approximately 40 µm for γ-Al2O3, and these Zr-doped samples exhibited negligible changes in morphology (Fig. S5 in Appendix A). The ZrO2-x/Al2O3 samples were further examined by high-resolution transmission electron microscopy (HRTEM; JEM-2100, JEOL, Japan) and energy dispersive spectroscopy (EDS; JEM-2100, JEOL) (Fig. 2(b)). Specifically, the interplanar spacing of 0.181 nm in the HRTEM image of ZrO2–0.2/Al2O3 was consistent with the (440) lattice plane of γ-Al2O3, while the lattice spacing increased to 0.185 and 0.218 nm when the Zr species dose reached 2.0 and 8.0 wt%, respectively [25]. No obvious ZrO2 lattice fringes were found in the HRTEM images of the ZrO2-0.2/Al2O3 and ZrO2-2.0/Al2O3 samples. However, well-dispersed and increasing amounts of Zr species can be observed in the EDS mapping images, suggesting that the ZrO2 species were successfully introduced into the γ-Al2O3 skeleton. For ZrO2-8.0/Al2O3 with a higher Zr content, the ZrO2 particles had a (111) lattice plane with a 0.273 nm interplanar space oriented on γ-Al2O3 [26], [27]. These results suggested that the ZrO2 species attached to the Al2O3 substrate through the Zr–O–Al units, and then, the ZrO2 particles gradually grew as the amount of ZrO2 decoration increased. Additionally, regions without lattice fringes can be observed in all of these samples, indicating the presence of an amorphous structure and unsaturated coordination sites on the ZrO2-x/Al2O3 composite, thereby increasing the number and activity of the Lewis acid sites (LASs).

The HRTEM and mapping results can be confirmed by the phenomenon observed in the X-ray diffraction (XRD; D8 advance, Bruker, Germany) curves. As presented in Fig. 2(c) and Fig. S6 in Appendix A, the XRD patterns display the well-resolved characteristic diffraction peaks of γ-Al2O3 with weak crystallization in the range of 10°–85° (joint committee on powder diffraction standards (JCPDS) NO. 10–0425) [23]. A weak and broad peak at 28.7° in the ZrO2-x/Al2O3 pattern was detected and enhanced with increasing ZrO2 dosage, which was mainly attributed to the associated ZrO2 particles [28]. Notably, the peak at 31.9° for γ-Al2O3 parallel to the basal (220) plane obviously shifted to a lower degree with increasing Zr dosage in the samples, suggesting that Zr atoms with a high radius were introduced and existed as amorphous Zr–O–Al units in the γ-Al2O3 skeleton, resulting in an increase in the overall skeletal morphology [23], [24]. Notably, the obtained Zr–O–Al units undergo strong metal oxide support interactions, which are favorable for tuning the properties and structure of catalysts [29]. The coordination of Zr with Al atoms in lower valence and the amorphous structure increases the negative charge of Zr sites, thus resulting in large numbers of unsaturated coordination Zr sites on ZrO2-x/Al2O3, thereby providing ample active LAS for CO2 desorption [30]. This result can be verified by the parameters of the surface chemical states of Zr, Al, and O species in the ZrO2-x/Al2O3 catalysts via X-ray photoelectron spectroscopy (XPS; PHI Quantro SXM, ULVAC-PHI, Japan) (Fig. 2(d)). The peaks at 181.7 and 184.2 eV in the Zr 3d spectrum were attributed to Zr 3d5/2 and 3d3/2, respectively. The Zr 3d3/2 peak can be assigned to two peaks at 184.9 and 184.0 eV for the ZrO2 and Zr–O–Al species, respectively [31]. The typical peak of Al3+ can be observed at 74.3 eV [32]. The O 1 s spectrum exhibits two peaks at 530.6 and 531.7 eV, which can be attributed to the lattice oxygens bonded with Zr or Al and to the surface adsorbed oxygen (e.g., –OH, H2O) [33]. Compared with those of ZrO2–0.2/Al2O3, the peaks of Zr 3d of the other two catalysts are slightly negatively shifted, while those of Al 2p are positively shifted, and the shifts are enhanced with increasing Zr content. Moreover, the Zr 3d and O 1s XPS spectra displayed an increase in the peak intensity of Zr species with low valence and surface-adsorbed oxygen. A more negative Bader charge for the Zr sites and a more positive charge for the Al sites were also observed for the Zr–O–Al species than for the Al2O3 and ZrO2 structures (Fig. S7 in Appendix A). Combining the above results, it can be concluded that electron accumulation occurred on the Zr species along the Al–O–Zr interface at the ZrO2/Al2O3 interface [24]. The positive Al sites are favorable for the adsorption of H2O, thus inducing abundant oxygen groups to maintain the hydrophilicity of the catalyst, while the modified Zr sites with higher electron intensities and unsaturated states can significantly promote the activation of carbamate and H2O [34]. As such, ZrO2-doped Al2O3 has a remarkable affinity for polar H2O molecules, in which the contact angle decreased from 84.1° to 21.6° as the ZrO2 dose increased from 0.2 to 8.0 wt% (Fig. 2(e)).

The improved hydrophilicity of catalysts is apparently propitious for matching their polarity with carbamate, H2O, and protonated amine reactants, thus significantly facilitating the diffusion and adsorption of reactants and the subsequent activity toward CO2 desorption. Figs. 2(f) and Fig. S8 in Appendix A indicate that the catalytic CO2 desorption amount exhibited a volcanic distribution with increasing ZrO2 dosage. The optimized ZrO2-2.0/Al2O3 exhibited a maximum CO2 desorption amount and rate of 0.49 mol·L–1 and 10.28 mmoL·min–1, respectively, which were 53% and 88% greater than those of the blank sample. The tendency for discrepancies between the hydrophobic and CO2 desorption properties is mainly attributed to the accessibility and acidity of the catalyst surface, which also has a critical influence on the CO2 desorption property. Thus, we further determined the variations in the CO2 desorption properties associated with changes in the surface structure and acidity of ZrO2-x/Al2O3. All the samples exhibited typical Langmuir IV adsorption–desorption isotherms and a microporous structure (Fig. S9 and Table S2 in Appendix A). Similar to the change in the CO2 desorption property, the surface area of ZrO2-x/Al2O3 increased with increasing ZrO2 dosage (from 0.1 to 1.0 wt%) and then decreased with increasing ZrO2 dosage (Fig. 2(g)). ZrO2-2.0/Al2O3 possessed a relatively high surface area and pore diameter, which suggested that the introduced Zr species might exist as amorphous structures and broaden the original pore channel structure, thus improving the efficiency of both diffusion and mass transfer [31]. Unlike the surface area of ZrO2-x/Al2O3, the total acidity of these catalysts was comparable, as observed by NH3 temperature-programmed desorption (NH3-TPD, Auto Chem II 2920, Micromeritics, USA) analysis (Figs. 2(g) and Fig. S10 in Appendix A). However, after ZrO2 introduction, more LAS were observed in the Pyridine-adsorption infrared spectroscopy (Py-IR; Nicolet 380, Thermo Fisher Scientific, USA), and the LAS peaked at a ZrO2 content of 2.0 wt%, which was similar to the change in the CO2 desorption property (Fig. S10 and Table S3 in Appendix A), implying that strong LAS originating from Zr species can cause high CO2 desorption activity [35]. Consequently, by providing γ-Al2O3 support with well-modified active Zr–O–Al coordination sites, the catalyst surface area, hydrophilicity, and acidity of SACs can be optimized, and CO2 desorption can be regulated.

3.2. Investigation of the structure-properties and catalytic mechanism

Although various SACs with promising catalytic CO2 desorption properties have been prepared, the majority of previous studies have investigated their catalytic properties via intermittent experiments at atmospheric pressure, which is different from actual industrial conditions (at relatively low TRL3). Few studies [22] have addressed the operational lifetime and feasibility of this strategy for industrial applications. Therefore, to further reveal the feasibility of a catalytic CO2 desorption strategy under industrial conditions, a bench-scale ZrO2-2.0/Al2O3 packing experiment under typical industrial conditions was investigated in this work, in which the packing temperature was controlled from 70 to 95 °C while the corresponding kettle temperature was 95–120 °C. The operation results revealed an increasing CO2 desorption rate with gradually increasing temperature because CO2 desorption is an endothermic reaction (Fig. 3(a)). As expected, ZrO2-2.0/Al2O3 had a positive effect on the CO2 desorption rate. The CO2 desorption rate increased and flattened as the packing temperature exceeded 90 °C. The maximum increase in the CO2 desorption rate was observed at 90 °C with the catalyst, which was greater than that of the blank sample by approximately 17.1%. This is mainly attributed to the fact that endothermic CO2 desorption is hampered by fewer active molecules at lower temperatures, while the catalysts exhibit a weak effect due to the high intrinsic kinetics of CO2 desorption at higher temperatures. Thus, an enhanced desorption rate led to full CO2 release, relatively high cyclic CO2 capacity and low CO2 loading in the CO2-lean solvent, resulting in a significantly greater CO2 capture efficiency under catalytic conditions (Fig. 3(b)). Notably, the CO2 capture efficiency of the blank sample reached 90.8% when the packing temperature was 90 °C and increased to 99.5% (9.6% improvement) when the temperature increased to 95 °C. Therefore, the regeneration temperature of the packing section must be maintained above 90 °C to guarantee an industrial CO2 capture efficiency greater than 90%. When ZrO2-2.0/Al2O3 was introduced, a 99.5% CO2 capture efficiency could be achieved at 90 °C, and a comparable efficiency of 90.8% could be achieved at 88 °C to that of the blank case at 90 °C, indicating a decrease in the regeneration temperature at the packing section from 90–95 to 88–90 °C. A decrease in the packing temperature reduces the regeneration temperature of the reactor and the energy consumption from electricity or high-grade steam. In this work, the heat duty of catalytic regeneration decreased by 11.2% compared to that of the blank sample at 90 °C (Fig. 3(c) and Table S4 in Appendix A). Combined with the decreased regeneration temperature, it can be calculated that a reduction of 18.4% of the high-grade steam quantity can be realized under catalytic conditions.

Aside from activity, the long-term stability of a catalyst is another critical metric for practical CO2 capture techniques. The durability of the ZrO2-2.0/Al2O3 catalyst was validated by a 168 h continuous experiment at 88 °C with 90% CO2 capture efficiency. As shown in Figs. 3(d) and (e), there was a negligible decrease in the CO2 desorption rate, and the CO2 capacity at 24 and 48 h exhibited a comparable value of approximately 1.5 mol·L–1, outperforming most other reported SACs for amine regeneration [36], [37]. The leaching concentrations of Zr and Al ions from ZrO2-2.0/Al2O3 (catalyst concentration of 1.0 wt%) after 30 d of immersion in MEA at 120 °C were characterized through inductively coupled plasma optical emission spectroscopy. Traces of Zr and Al ions at 0.9 and 6 parts per billion (ppb) were detected after 14 d but slightly increased to 1.0 and 7 ppb after 30 d (Fig. 3(f) and Table S5 in Appendix A). The calculated extraction ratio of the metal species was lower than 0.004 wt%. Notably, previous research has indicated that metal species increase the risk of amine degradation. Therefore, the 13C nuclear magnetic resonance (NMR; AVANCE III 600 M, Bruker) spectra of the amine solvent cotreated with the catalyst under the above conditions are displayed in Fig. 3(g). The good structural integrity of the amine molecule further verified that the introduced catalyst had no poisoning or decomposition effects on the amine solvent. Additionally, the durability of ZrO2-2.0/Al2O3 can also be confirmed by the structural variation results obtained through XRD, NH3-TPD, surface area, and XPS analyses, in which negligible changes in the structure, acidity, surface area, and XPS spectra of Zr 3d and Al 2p can be observed (Figs. S11, Fig. S12, and Table S6 in Appendix A). A slight peak in the N 1s spectrum was detected after immersion, which was mainly ascribed to the residual amine-related species present during the catalyst recovery process and did not affect the catalytic properties (Fig. S12(d)). Based on the high stability of ZrO2-2.0/Al2O3 packing catalysts, a structured pack was prepared by coupling ZrO2-2.0/Al2O3 packing (150 kg) with 250Y structural packing and then arranged in a stripper in an industrial amine-based CO2 capture system in Jiantao (Hebei, China) (Fig. S1(c)), which can capture 200 000 tonnes of CO2 from coal-fired flue gas every year. Notably, negligible amounts of Zr (0.0045 mg·L−1) and Al (0.0095 mg·L−1) species were detected after ten months (Table S7 in Appendix A). Thus, the results revealed the excellent stability of ZrO2-2.0/Al2O3 in an alkaline environment at high temperature, which was mainly attributed to the strong metal oxide-support interactions of the Zr–O–Al units.

In our previous work [19], we verified that a catalyst can reduce the CO2 desorption energy barrier. A lower reaction energy barrier can yield a comparable catalytic desorption rate at low temperature, thereby enabling a decrease in temperature and converting the CO2 desorption reaction into a kinetically controlled process. Thus, to investigate the reduction in the energy barrier of CO2 desorption, the activation energy Ea of this reaction was estimated on the basis of Arrhenius plots, while the reaction kinetic constants were calculated from the desorption kinetic data at different temperatures with or without (W/O) catalysts (Fig. 3(h) and Fig. S13 in Appendix A) [19], [38]. The results indicated that the CO2 desorption rate increased with increasing temperature, which was primarily attributed to the inherent features of the thermodynamically controlled reaction. ZrO2-2.0/Al2O3 exhibited a considerably lower Ea of 51.64 kJ·mol–1 for CO2 desorption, whereas the derived Ea was 87.16 kJ·mol–1 without catalyst, implying a 40.7% decrease in energy activation (Fig. 3(i)). A reduced temperature will significantly reduce energy (electricity and steam) consumption and thermal degradation in amine scrubbing technologies. Consequently, the good stability and decreased energy consumption clearly confirmed the advance and feasibility of a catalytic desorption strategy for energy-saving amine scrubbing, thus enabling this technology to be applied to TRL 6.

Based on the results of the structural variation and the significant effect of LAS on CO2 desorption, the changes in the acid properties of the recycled ZrO2-2.0/Al2O3 were measured to understand the mechanism of this reaction. As depicted in Fig. S14 in Appendix A, after treatment in an amine solution, the peak intensity of adsorbed oxygen and Bronsted acid sites (BAS) on ZrO2-2.0/Al2O3 increased, and the content ratio of BAS/LAS increased from 0.19 to 0.29. Density functional theory (DFT) calculations revealed the adsorption and activation of H2O at Al sites using two optimized models, γ-Al2O3 and ZrO2/Al2O3 (Fig. 3(j) and Fig. S15 in Appendix A) [39]. These calculations revealed that the H2O molecule was more readily adsorbed on the Al atom of the Zr–O–Al site in the ZrO2/Al2O3 model, exhibiting a higher adsorption energy of –1.76 eV compared to –1.62 eV in γ-Al2O3. Additionally, the Osingle bondH bond length of the adsorbed *H2O increased from 0.99 to 1.03 Å over the γ-Al2O3 and ZrO2/Al2O3 models, respectively, indicating enhanced activation of *H2O. Figs. 3(j) and Fig. S16 in Appendix A further present Gibbs free energy diagrams of H2O dissociation at these sites, where the conversion of *H2O to *OH–H was identified as the rate-determining step for proton feeding. *H2O dissociation on the Al atom of the Zr–O–Al site required a lower energy barrier of 0.66 eV compared to 1.29 eV for γ-Al2O3. This reduced energy barrier suggests that the adsorbed *H2O on the Zr–O–Al site could primarily produce protons to support the protonation process for carbamate decomposition, which is advantageous for lowering the reaction energy barrier for CO2 desorption and facilitating the CO2 desorption reaction at a lower temperature.

Combining the above HRTEM, XRD, XPS, and DFT results and experimental results, it can be concluded that the Zr/Al correlation and its effect on catalysis were as follows. As shown in Fig. 3(k), the constructed Zr–O–Al structure demonstrates superior acidity and stability and increases the surface accessibility of the material. The doped ZrO2 in Al species with a low valence and radius increases the porosity of the original γ-Al2O3 and, in turn, enhances its electron intensity and acidity, which promotes the adsorption and activation of H2O molecules into proton and –OH groups. The proton originating from H2O activation on Al LASs thus reacts with R1NHCOO active on Zr LASs to boost amine regeneration, CO2 desorption, and subsequent deprotonation of RNH3+. Moreover, considering the diversity of amines available for CO2 capture, the universality of the ZrO2-2.0/Al2O3 catalyst over two blends and biphasic amines (e.g., MEA–MDEA and MEA–MDEA–TMS, where MDEA is N-methyldiethanolamine and TMS is sulfolane) was investigated. As shown in Fig. S17 in Appendix A, except for the easy regeneration of tertiary amines, ZrO2-2.0/Al2O3 mainly regenerated primary amines with a high energy barrier; thus, a comparable increase in the amount of CO2 desorption of approximately 0.12–0.17 mol·L–1 was observed. ZrO2-2.0/Al2O3 has been widely utilized for amine regeneration. Taken together, the above results indicated that the ZrO2-2.0/Al2O3 catalyst has remarkable durability and activity for CO2 desorption and can thus be readily applied to practical amine scrubbing technology to achieve viable, energy-efficient, and long-life CO2 capture cycles.

3.3. Process simulation and technologic–economic evaluation

After the experimental investigation, process simulation using Aspen Plus was carried out to verify the key operational parameters and economic-environmental efficiency [40], [41]. Before the industrial-scale simulation, rigorous model validation was carried out to simulate the bench-scale system (Fig. S18 in Appendix A). The lower average absolute relative deviation (AARD), 1.39% (catalytic), and 2.73% (blank) for the temperature and 2.07% (catalytic) and 3.06% (blank) for the CO2 capture efficiency verified the reliability of the process simulation and indicated that it could be used for further industrial simulations. Based on these results, the process flowchart of CO2 capture from flue gas is shown in Fig. 4(a). The main components in the flue gas included 12% CO2, 76% N2, 5% O2, and 7% H2O, with a flow rate of 1.2 × 105 N·m3·h−1. The detailed process specifications can be found in the methods section. Several scenarios of different catalytic filler replacements have been investigated to evaluate the capture performance. The effects of the operational parameters on the CO2 capture efficiency and specific energy consumption were investigated via sensitivity analysis. The results are shown in Fig. S19 in Appendix A. With increasing solvent flow rate and desorption pressure, the heat duty first decreases and then increases steadily, while the capture efficiency shows the opposite trend (Fig. S19(a) and (b)). Generally, for the solvent flow rate, the initial decrease may be due to an improvement in the efficiency of vapor–liquid contact in the absorption column, while the subsequent increase may be the main result of a decrease in the regeneration effect. For desorption pressure, this can be attributed to the fact that the higher pressure of the desorption tower makes the solvent temperature rise, conducive to the decomposition of the carbamate, while the higher pressure makes it difficult for CO2 to leave the solution and reduces the desorption efficiency. From Fig. S19(c) and (d), as the tower plate and feed solvent temperature increase, the heat duty decreases, and the capture efficiency increases gradually. For tower plates, increasing the number of tower plates facilitates vapor–liquid contact, in which the solvent touches the flue gas more fully. With respect to the solvent temperature, decreasing the temperature changes the vapor–liquid equilibrium (VLE) to a higher level of CO2 solubility in the solvent. These factors all boost the CO2 capture efficiency. The heat power showed the opposite trend for the tower plate number and feed solvent temperature (Fig. S19(e)). A large amount of heat power was input to the solvent regeneration tower and helped to enhance the regeneration efficiency of the rich liquid.

From Fig. 4(b), with increasing catalytic dosage in the regeneration column, the desorption amount and capture efficiency first increase and then decrease, while the heat duty shows the opposite trend. A catalytic dose of 60% had the highest desorption amount of 27.22 (tCO2)·h−1, with the highest capture efficiency of 95.48% and the lowest heat duty of 3.01 GJ·tCO2−1. The reference scenario of a catalytic dosage of 0 is introduced to conduct a comparative study. For the desorption amount, capture efficiency, and heat duty, the scenario of 60% catalytic dosage has a great cost, being approximately 121.64%, 121.64%, and 82.46% of those of the reference scenario, respectively. Hence, the catalytic filler can improve CO2 capture performance, and the optimal catalytic dosage can reach 60%. The capture cost of those scenarios was calculated for the total project life. The catalytic regeneration process with an optimal catalytic dosage is also advantageous for reducing the cutting capture cost, with the TCC being approximately 21.26% of that of the reference scenario. The TCC and heat duty of conventional and catalytic scenarios under the same capture efficiency (approximately 95%) are shown in Fig. 4(c). Compared with those of the conventional scenario, the TCC and heat duty of the catalytic scenario can decrease by approximately 10.5% and 27.6%, respectively. The potential CO2 emission reduction can be regarded as a key environmental indicator of the carbon capture process. The NCOER of the conventional and catalytic processes was investigated, and the results are shown in Fig. S20 in Appendix A. For NCOER, the catalytic process can significantly improve CO2 emission reduction, reducing CO2 emission by approximately 13.0%. Technologically speaking, catalytic regeneration technology is largely subject to economic and environmental feasibility.

4. Conclusions

In summary, acidic ZrO2/Al2O3 catalytic amine regeneration for energy-saving CO2 capture via chemisorption was comprehensively evaluated via the integration of upscale experiments and process simulations. The investigations confirmed that the doped ZrO2 species facilitated electron transfer along Zr–O–Al units and enhanced the interaction and sequential proton transfer between H2O and carbamate on the surface of ZrO2/Al2O3. The reduced energy activation by 40.7% and optimized CO2 desorption rate enable a reduction in energy consumption of 27.6% and a decrease in overall cost of 10.5%. Moreover, the Zr–O–Al units also endow the ZrO2/Al2O3 packing catalyst with good structural integrity during long-term treatment, thus leading to the complete verification of the feasibility of the use of high TRL 6 for catalytic solvent regeneration. Our approach paves the way for regulating CO2 desorption behaviors to realize energy-efficient, and long-life cycled amine scrubbing and, by extension, provides insights into the efficient chemisorption of CO2 from fossil fuel consumption in the industry to address concerns about global warming issues.

CRediT authorship contribution statement

Lei Xing: Writing – original draft, Validation, Methodology, Funding acquisition, Conceptualization. Zhen Chen: Writing – original draft, Validation, Methodology, Formal analysis, Conceptualization. Guoxiong Zhan: Writing – original draft, Validation, Methodology, Conceptualization. Zhoulan Huang: Visualization, Software, Investigation. Lidong Wang: Writing – review & editing, Funding acquisition. Junhua Li: Writing – review & editing, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

This work was supported by the National Natural Science Foundation of China (52300134 and 22106084) and the China Postdoctoral Science Foundation (2022TQ0175, 2023M741931, and 2022T150350).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.eng.2024.08.024.

References

[1]

Massen-Hane M, Diederichsen KM, Hatton TA.Engineering redox-active electrochemically mediated carbon dioxide capture systems.Nat Chem Eng 2024; 1(1):35-44.

[2]

Wei K, Guan H, Luo Q, He J, Sun S.Recent advances in CO2 capture and reduction.Nanoscale 2022; 14(33):11869-11891.

[3]

Liu C, Jing G, Zhu Z, Fan Y, Mo S, Zhang Y, et al.Inhibiting effect of 2-amino-2-methyl-1-propanol on gelatinous product formation in non-aqueous CO2 absorbents: experimental study and molecular understanding.Chem Eng J 2024; 481:148545.

[4]

Sullivan I, Goryachev A, Digdaya IA, Li X, Atwater HA, Vermaas DA, et al.Coupling electrochemical CO2 conversion with CO2 capture.Nat Catal 2021; 4(11):952-958.

[5]

Yu Y, Shen Y, Zhou X, Liu F, Zhang S, Lu S, et al.Relationship between tertiary amine’s physical property and biphasic solvent’s CO2 absorption performance: quantum calculation and experimental demonstration.Chem Eng J 2022; 428:131241.

[6]

Liu X, Niu X, Zhan G, Xing L, Huang Z, Yuan B, et al.Dynamic phase-splitting behaviour of biphasic solvent for carbon capture in a novel annular phase separator.Appl Energy 2024; 360:122788.

[7]

Zhang X, Zhu Z, Sun X, Yang J, Gao H, Huang Y, et al.Reducing energy penalty of CO2 capture using Fe promoted SO42−/ZrO2/MCM-41 catalyst.Environ Sci Technol 2019; 53(10):6094-6102.

[8]

Wang N, Peng Z, Gao H, Sema T, Shi J, Liang Z, et al.New insight and evaluation of secondary amine/N-butanol biphasic solutions for CO2 capture: equilibrium solubility, phase separation behavior, absorption rate, desorption rate, energy consumption and ion species.Chem Eng J 2022; 431:133912.

[9]

Gautam A, Mondal MK.Review of recent trends and various techniques for CO2 capture: special emphasis on biphasic amine solvents.Fuel 2023; 334:126616.

[10]

Chen Z, Yuan B, Zhan G, Li Y, Li J, Chen J, et al.Energy-efficient biphasic solvents for industrial carbon capture: role of physical solvents on CO2 absorption and phase splitting.Environ Sci Technol 2022; 56(18):13305-13313.

[11]

Stern MC, Simeon F, Herzog H, Hatton TA.Post-combustion carbon dioxide capture using electrochemically mediated amine regeneration.Energy Environ Sci 2013; 6(8):2505.

[12]

Lin X, Wegner B, Lee KM, Fusella MA, Zhang F, Moudgil K, et al.Beating the thermodynamic limit with photo-activation of N-doping in organic semiconductors.Nat Mater 2017; 16(12):1209-1215.

[13]

Espinosa MR, Ertem MZ, Barakat M, Bruch QJ, Deziel AP, Elsby MR, et al.Correlating thermodynamic and kinetic hydricities of rhenium hydrides.J Am Chem Soc 2022; 144(39):17939-17954.

[14]

Foster SL, Bakovic SIP, Duda RD, Maheshwari S, Milton RD, Minteer SD, et al.Catalysts for nitrogen reduction to ammonia.Nat Catal 2018; 1(7):490-500.

[15]

Xia P, Pan X, Jiang S, Yu J, He B, Ismail PM, et al.Designing a redox heterojunction for photocatalytic “overall nitrogen fixation” under mild conditions.Adv Mater 2022; 34(28):2200563.

[16]

Lai Q, Toan S, Assiri MA, Cheng H, Russell AG, Adidharma H, et al.Catalyst-TiO(OH)2 could drastically reduce the energy consumption of CO2 capture.Nat Commun 2018; 9(1):2672.

[17]

Xing L, Chen Z, Zhan G, Huang Z, Li M, Li Y, et al.Sulfur migration enhanced proton-coupled electron transfer for efficient CO2 desorption with core-shelled C@Mn3O4.Environ Sci Technol 2024; 58(10):4606-4616.

[18]

Li X, Xing L, Zhan G, Huang Z, Chen Z, Chang H, et al.Oxygen vacancy enhanced proton transfer to boost carbamate decomposition kinetics with tunable heterostructure Ni/NiO.ACS Catal 2024; 14(2):1083-1092.

[19]

Li M, Xing L, Xu Z, Liang Z, Qi T, Li Y, et al.Embedded Mo/Mn atomic regulation for durable acidity-reinforced HZSM-5 catalyst toward energy-efficient amine regeneration.Environ Sci Technol 2023; 57(41):15465-15474.

[20]

Xing L, Li M, Li M, Xu T, Li Y, Qi T, et al.MOF-derived robust and synergetic acid sites inducing C–N bond disruption for energy-efficient CO2 desorption.Environ Sci Technol 2022; 56(24):17936-17945.

[21]

Bhatti UH, Nam S, Park S, Baek IH.Performance and mechanism of metal oxide catalyst-aided amine solvent regeneration.ACS Sustain Chem Eng 2018; 6(9):12079-12087.

[22]

Chen Z, Xing L, Zhan G, Huang Z, Bai X, Wang R, et al.Integration of strong oxide-support interaction and mesoporous confinement to engineer efficient and durable Zr/Al2O3 for catalytic solvent regeneration in CO2 capture.Appl Catal B 2024; 351:124000.

[23]

Dang C, Zhou J, Xia H, Cai W.A hierarchical hollow Ni/γ-Al2O3 catalyst derived from flower-like Ni–Al layered double hydroxide with stable catalytic performance for CO2 methanation.J Mater Chem A 2024; 12(14):8281-8290.

[24]

Zhu N, Zhang X, Chen N, Zhu J, Zheng X, Chen Z, et al.Integration of MnO2 nanosheets with Pd nanoparticles for efficient CO2 electroreduction to methanol in membrane electrode assembly electrolyzers.J Am Chem Soc 2023; 145:24852-24861.

[25]

Wang S, Borisevich AY, Rashkeev SN, Glazoff MV, Sohlberg K, Pennycook SJ, et al.Dopants adsorbed as single atoms prevent degradation of catalysts.Nat Mater 2004; 3(3):143-146.

[26]

Liu J, Shi X, Shan Y, Yan Z, Shan W, Yu Y, et al.Hydrothermal stability of CeO2–WO3–ZrO2 mixed oxides for selective catalytic reduction of NOx by NH3.Environ Sci Technol 2018; 52:11769-11777.

[27]

Zhang Y, Zhao Y, Otroshchenko T, Lund H, Pohl MM, Rodemerck U, et al.Control of coordinatively unsaturated Zr sites in ZrO2 for efficient C–H bond activation.Nat Commun 2018; 9(1):3794.

[28]

Peng B, Rapp KGé, Cui Y, Gao F, Szanyi J, Olszta MJ, et al.Enhancement of high-temperature selectivity on Cu-SSZ-13 towards NH3-SCR reaction from highly dispersed ZrO2.Appl Catal B 2020; 263:118359.

[29]

Zhao ZJ, Wu T, Xiong C, Sun G, Mu R, Zeng L, et al.Hydroxyl-mediated non-oxidative propane dehydrogenation over VOx/gamma-Al2O3 catalysts with improved stability.Angew Chem Int Ed 2018; 57(23):6791-6795.

[30]

Park SY, Park JS, Kim BJ, Lee H, Walsh A, Zhu K, et al.Sustainable lead management in halide perovskite solar cells.Nat Sustain 2020; 3(12):1044-1051.

[31]

Xing L, Wei K, Li Q, Wang R, Zhang S, Wang L, et al.One-step synthesized SO42−/ZrO2-HZSM-5 solid acid catalyst for carbamate decomposition in CO2 capture.Environ Sci Technol 2020; 54(21):13944-13952.

[32]

Li W, Nie X, Jiang X, Zhang A, Ding F, Liu M, et al.ZrO2 support imparts superior activity and stability of Co catalysts for CO2 methanation.Appl Catal B 2018; 220:397-408.

[33]

Zhu X, Yuan J, Peng Y, Chen J, Yuan X, Li J, et al.In situ S-doping engineering for highly efficient NH3-SCR over metal-free carbon catalysts: a novel synergetic promotional mechanism.Environ Sci Technol 2024; 58(1):371-380.

[34]

Wu Y, Cai T, Zhao W, Chen X, Liu H, Wang Y, et al.First-principles and experimental studies of [ZrO(OH)]+ or ZrO(OH)2 for enhancing CO2 desorption kinetics—imperative for significant reduction of CO2 capture energy consumption.J Mater Chem A 2018; 6(36):17671-17681.

[35]

Xiong S, Chen J, Huang N, Yan T, Peng Y, Li J, et al.The poisoning mechanism of gaseous HCl on low-temperature SCR catalysts: MnO−CeO2 as an example.Appl Catal B 2020; 267:118668.

[36]

Bhatti UH, Kazmi WW, Min GH, Haider J, Nam S, Baek IH, et al.Facilely synthesized M-montmorillonite (M = Cr, Fe, and Co) as efficient catalysts for enhancing CO2 desorption from amine solution.Ind Eng Chem Res 2021; 60(36):13318-13325.

[37]

Bairq ZAS, Gao H, Murshed FAM, Tontiwachwuthikul P, Liang Z.Modified heterogeneous catalyst-aided regeneration of CO2 capture amines: a promising perspective for a drastic reduction in energy consumption.ACS Sustain Chem Eng 2020; 8(25):9526-9536.

[38]

Ighodalo KO, Chen W, Liang Z, Shi Y, Chu S, Zhang Y, et al.Negligible ion migration in tin-based and tin-doped perovskites.Angew Chem Int Ed 2023; 62(5):e202213932.

[39]

Wang F, Ma J, Xin S, Wang Q, Xu J, Zhang C, et al.Resolving the puzzle of single-atom silver dispersion on nanosized γ-Al2O3 surface for high catalytic performance.Nat Commun 2020; 11(1):529.

[40]

Zhan G, Yuan B, Duan Y, Bai Y, Chen J, Chen Z, et al.Simulation and optimization of carbon dioxide capture using water-lean solvent from industrial flue gas.Chem Eng J 2023; 474:145773.

[41]

Zhan G, Cao F, Bai L, Chang F, Zhou B, Duan Y, et al.Process simulation and optimization of ammonia-containing gas separation and ammonia recovery with ionic liquids.ACS Sustain Chem Eng 2021; 9(1):312-325.

AI Summary AI Mindmap
PDF (3290KB)

1643

Accesses

0

Citation

Detail

Sections
Recommended

AI思维导图

/