Carrier-free silver carbonate catalysis: in situ conversion of low-concentration CO₂ in air and actual exhaust gas to oxazolidinones

Abstract

The increasing concentration of CO₂ in the atmosphere drives global climate change, and efficient methods for CO₂ utilisation are urgently needed. This study introduces a novel carrier-free catalytic approach using cost-effective Ag₂CO₃ to enable the in situ carboxylation of ultra-low-concentration CO₂ with propargylamine under ambient temperature and pressure conditions. This method efficiently produces oxazolidinone compounds, which hold significant industrial value. The catalytic system not only efficiently converts CO₂ at low concentrations ranging from 5 vol% to 35 vol% but also directly converts CO₂ present in air (0.04 vol%) into oxazolidinones, achieving maximum yields of up to 97%, and demonstrates good substrate applicability. A cyclic reaction device reduces CO₂ concentration in simulated gas from 15 vol% to 3 vol%, maintaining an average oxazolidinone yield of 92%.

---

Green foundation

1. Under the urgent situation of global carbon emission reduction, developing more environmentally friendly and efficient pathways for the resource utilization of carbon dioxide can not only effectively reduce CO₂ emissions but also achieve the circular utilization of resources. This holds extremely important strategic significance for promoting the achievement of carbon emission reduction goals.

2. This study innovatively developed a carrier-free and inexpensive Ag₂CO₃ catalytic system that efficiently converts low-concentration CO₂ into industrially valuable oxazolidinones at room temperature and pressure. By skipping CO₂ enrichment and transportation, it reduces energy use and adsorbent reliance, making the process more environmentally friendly and sustainable, while significantly lowering carbon capture costs.

3. Future work will focus on expanding the substrate range and developing an integrated continuous device for CO₂ conversion, product separation, and solvent recovery to enhance practical CO₂ pathways in practical scenarios.

1. Introduction

Amid the escalating challenge of global warming, carbon dioxide (CO₂)—a principal greenhouse gas—has garnered significant attention for its potential in resource utilization.^1–3 Over the past few years, extensive research has been focused on devising innovative processes to transform CO₂ into a range of valuable chemical products, such as carbon monoxide,^4–6 methane,^7–9 methanol,^10,11 oxazolidinones,^12,13 cyclic carbonates,^14,15 and macromolecular polymeric materials,^16,17 thereby enhancing the efficient utilization of CO₂ resources. However, most existing CO₂ conversion technologies rely on high-purity feedstock, yet the industrial process of capturing, purifying, and transporting CO₂ from waste streams is energy-intensive and costly, with estimates ranging from $30 to $200 per ton.¹⁸ These auxiliary steps not only increase the overall complexity of the conversion process but also undermine the economic feasibility of CO₂ utilization, thereby severely limiting its large-scale practical application (Fig. 1a).^19,20 In addition, the extensive use of adsorbents and the operation of high-energy-consuming equipment may lead to secondary pollution issues, further undermining the environmental benefits of CO₂ resource utilization.

Fig. 1 Chemical sequestration pathways.

Therefore, it is imperative to develop technologies capable of directly converting low-concentration CO₂in situ. Such technologies can directly transform low-concentration CO₂ in actual flue gases into target products,^21–23 bypassing numerous cumbersome intermediate steps and potentially reducing carbon sequestration costs significantly.

Although in situ CO₂ conversion offers the advantage of reducing overall carbon sequestration costs and promoting widespread adoption, the conversion of low-concentration CO₂ in flue gases or in air poses significant challenges. The slow mass transfer process, coupled with an increase in interfacial binding energy and a high activation energy requirement, makes the direct conversion of low CO₂ concentrations particularly difficult.^24,25 Moreover, actual gas streams contain impurities such as sulfur dioxide and nitrogen oxides that can deactivate the catalyst and compromise overall performance.²⁶

To address these challenges, previous studies have emphasized that the key lies in developing novel materials that integrate both CO₂ enrichment capability and catalytic function (Fig. 1b). In earlier research, porous organic polymers,²⁷ metal–organic frameworks (MOFs),²⁸ covalent organic frameworks (COFs),²⁹ and hyper-cross-linked polymers (HCPs)³⁰ have been employed for the fixation of low-concentration CO₂. These materials primarily rely on pore confinement to create a high-concentration CO₂ microenvironment within their structure, thereby facilitating the conversion reaction. However, the synthesis of such porous materials is complex and costly.³¹ Furthermore, their conventional preparation often involves energy-intensive processes, which can result in substantial embedded carbon emissions.

During the research process, we fortuitously discovered that silver carbonate (Ag₂CO₃) can serve as an effective mediator for the carboxylation cyclization reactions of low-concentration CO₂. Therefore, in this work, we constructed a carrier-free CO₂ catalytic platform utilizing inexpensive and readily available Ag₂CO₃ as the catalyst to catalyze the carboxylative cyclization of CO₂ with propargylamine, using gases containing low-concentration CO₂ as a carbon source (Fig. 1c). This catalytic platform enables the efficient synthesis of oxazolidinones. Oxazolidinones are a class of high-value molecules, primarily serving as core pharmacophores in antibacterial drugs (e.g., linezolid) and finding broad applications in the fields of agrochemicals and materials science. This catalytic system effectively converts CO₂ with concentrations ranging from 5 vol% to 30 vol% in simulated gases. It can also achieve the in situ chemical conversion of CO₂ from actual exhaust gas and even ultra-low-concentrations of CO₂ present in ambient air. When combined with a cyclic carbon sequestration unit, it is possible to effectively reduce the CO₂ concentration in the simulated gas while also obtaining oxazolidinones in excellent yields.

The Ag₂CO₃-based catalytic platform developed herein demonstrates significant advantages: it exhibits much lower material cost, simpler preparation free from energy- and carbon-intensive synthesis, higher catalytic efficiency and broader substrate adaptability under low-concentration CO₂ conditions, as well as superior tolerance to interfering components in real gas atmospheres, thereby effectively avoiding rapid catalyst deactivation. Furthermore, the system shows excellent scalability and cyclic operation potential. When integrated with a cyclic carbon capture unit, it enables continuous reduction of CO₂ concentration in simulated flue gas while maintaining high yields of oxazolidinones under mild conditions. This approach exhibits promising economic viability and scalability, offering a new technological pathway for practical carbon capture and utilization.

2. Experiments

2.1 Materials

Silver carbonate (Ag₂CO₃, AR >99%), silver oxide (Ag₂O, AR >99%), silver acetate (AgOAc, AR >99%), silver bromide (AgBr), silver nitrate (AgNO₃, AR >99%), copper acetate (Cu(OAc)₂, AR >99%), copper oxide (CuO, AR >99%), cuprous iodide (CuI, AR >99%), tetrakis(triphenylphosphine)palladium (Pd(PPh₃)₄, AR >99%), palladium acetate (Pd(OAc)₂, AR >99%), magnesium chloride hexahydrate (MgCl₂·6H₂O, AR >99%), iron chloride hexahydrate (FeCl₃·6H₂O, AR >99%), iodobenzene (AR >99%), cobalt acetate tetrahydrate (Co(OAc)₂·4H₂O, AR >99%), nickel acetate tetrahydrate (Ni(OAc)₂·4H₂O, AR >99%), aluminum chloride (AlCl₃, AR >99%) and 2-methyl-3-butyn-2-amine (>95%) were obtained from Innochem. Solvents such as toluene, ethyl acetate (EA, AR >99%), dimethyl sulfoxide (DMSO, AR >99%), acetonitrile (MeCN, AR >99%) and methanol (MeOH, AR >99%) were purchased from local vendors. The reagents used in this experiment were of analytical purity and generally did not require additional purification during use unless otherwise specified.

2.2 Carboxylative cyclization of propargylamine with low-concentration CO₂

The carboxylative cyclization of propargylamine with low-concentration CO₂ was performed in a 10 mL branched glass reaction tube. Specifically, the Ag₂CO₃ catalyst (2.5 mol%) was first added to the tube, followed by three vacuum-refill cycles with a CO₂ balloon to completely replace the air in the apparatus. Under the CO₂ atmosphere, a solution of propargylamine (0.2 mmol) in 1 mL of DMSO was injected into the system via a syringe, and the mixture was stirred in an agitator (Scheme 1). The yield of oxazolidinone was determined using a high performance liquid chromatograph (HPLC), with a mobile phase ratio of methanol/water = 80/20, a flow rate of 1 mL min⁻¹, an absorption wavelength of 281 nm.

Scheme 1 Carboxylative cyclization of propargylamine with low-concentration CO₂.

2.3 Computational details

All structural relaxation and electronic property calculations were performed using the Vienna Ab Initio Simulation Package (VASP).^32,33 The electron–electron interactions were treated using the Generalized Gradient Approximation (GGA),³⁴ within the Perdew–Burke–Ernzerhof (PBE) functional framework. The pseudopotentials used in these calculations were based on the Projector Augmented Wave (PAW) method.³⁵ The electronic wave function calculations employed a cutoff energy of 500 eV. During the structural relaxation, all atoms were allowed to relax fully until the residual force on each atom was minimized to below 0.02 eV Å⁻¹, ensuring the energy of the system was converged. The electronic energy relaxation process was carried out to a convergence criterion of 10⁻⁶ eV. For the Brillouin zone sampling³⁶ in structural relaxation, a 1 × 1 × 1 k-point grid was used.

3. Results and discussion

3.1 Catalytic evaluation

The carboxylative cyclization reaction was developed using mixed gas with a CO₂ concentration of 5 vol% CO₂ in N₂ as a carbon source and compound 1a as a reaction substrate (Scheme 1). This carboxylative cyclization was used as a model reaction for the optimization of the reaction conditions. First, the effects of different metals on the reaction were investigated. The experimental results showed that CuO, Cu(OAc)₂, CuI, Pd(PPh₃)₄, Pd(OAc)₂, MgCl₂·6H₂O, FeCl₃·6H₂O, Co(OAc)₂·4H₂O, Ni(OAc)₂·4H₂O and AlCl₃ could not effectively catalyse the carboxylative cyclization reaction of low-concentration CO₂ with alkynylamine (Fig. 2a). It was also observed that while CuI showed no catalytic effect in a 5 vol% CO₂/N₂ mixture, it could catalyze the reaction under a gas atmosphere containing high CO₂ concentration, albeit with a modest product yield of 32% (Fig. S1). This contrast clearly demonstrates the unique effectiveness of the silver-based catalyst in driving the transformation under low CO₂ concentrations. In contrast, silver salts showed significant catalytic activity in the reaction, highlighting the important role of silver in this conversion process. Subsequently, by comparing the catalytic effects of different kinds of silver salts, it was found that the best results were obtained when Ag₂CO₃ was used as the catalyst, resulting in a 75% yield of oxazolidinones (Fig. 2b).

Fig. 2 Optimisation of reaction conditions. (a and b) Type of metal catalyst, (c) solvent type, (d) reaction temperature, (e) dosage of Ag₂CO₃, and (f) CO₂ concentration. General reaction conditions: 1a (0.2 mol), catalyst (5 μmol), solvent (1 mL), balloon containing 5 vol% CO₂/N₂ mixed gas, r.t., 12 h.

To identify the optimal reaction medium, we systematically evaluated a set of ten solvents, including acetonitrile, DMF, DMSO, ethanol, and a deep eutectic solvent (DES, choline chloride : urea = 1 : 2). As shown in Fig. 2c, the reaction proceeded effectively in five of these solvents, yielding the product in the following descending order: DMSO (75%), DMF (47%), acetonitrile (31%), DES (13%), and ethanol (6%). These results clearly identify DMSO as the most conducive medium for this transformation. This may be attributed to the fact that DMSO, as a strong polar solvent, facilitates the dissolution of the reactants and stabilizes the reaction intermediates, thereby increasing the efficiency of the reaction. Subsequently, an investigation into the effect of temperature on the reaction revealed that room temperature was optimal (Fig. 2d), with increasing or decreasing temperatures proving unfavorable for the reaction. This is mainly due to increased temperatures decreasing the solubility of CO₂ in the solvent and decreased temperatures reducing the reactivity of the reaction substrate. The catalyst dosage also significantly affects the reaction. When Ag₂CO₃ dosage was 5 μmol, the oxazolidinone yield reached 75%, the TON value was 30, and the catalyst utilization was high. When the dosage of Ag₂CO₃ was increased to 10 μmol, the yield slightly improved to 82%. However, the TON decreased to 16, indicating a reduced catalytic efficiency. Further increases in catalyst dosage did not significantly improve yield and continued to decrease catalyst utilization (Fig. 2e).

The effect of CO₂ concentration on the yield of the target product was subsequently investigated. The results indicated that under a nitrogen atmosphere, the target product could not be detected, confirming that CO₂ acts as a necessary substrate in the reaction. As the CO₂ concentration increased, there was a clear upward trend in the yield of the target product (Fig. 2f). Specifically, when the CO₂ concentration reached 10 vol%, the yield of oxazolidinone exceeded 95%. In contrast to existing reports (Table S1), the silver carbonate-based catalytic system enables the conversion of propargylamines to oxazolidinones using CO₂ feedstock with a wide range of concentrations. Interestingly, even trace amounts of CO₂ present in the air were found to be sufficient as a carbon source for producing the target product. The catalytic process demonstrated high sensitivity to CO₂ concentration, thus creating favorable conditions for utilizing it in capturing and fixing CO₂ from various types of flue gases.

The concentration of CO₂ affects not only the yield of the target product but also alters the reaction rate. When using 5 vol% CO₂/N₂ mixed gas as the carbon source, it takes approximately 12 h to achieve the optimal yield of the product (Fig. 3a). However, utilizing 10 vol% CO₂/N₂ mixed gas as the carbon source significantly accelerates the reaction rate, enabling a 96% yield of the target product in just 10 h. Under a high-purity CO₂ atmosphere, the reaction achieved a 92% yield of the target product in merely 1 h (Scheme S3). It was encouraging that even trace amounts of CO₂ in the air can be effectively fixed through this process. By appropriately extending the reaction time to 96 h and increasing the catalyst dosage to 15 mol% Ag₂CO₃ while using CO₂ from ambient air as the carbon source, a yield of up to 45% for the target product can also be achieved (Fig. 3b). After that, the yield of the target product does not increase significantly with continuous increase in the amount of Ag₂CO₃ used.

Fig. 3 (a) Yield of 2a at various times. Reaction conditions: 1a (0.2 mol), Ag₂CO₃ (5 μmol), DMSO (1 mL), balloon containing 5 vol% or 10 vol% CO₂/N₂ mixed gas, r.t. (b) The effect of Ag₂CO₃ dosage on the yield of 2a using air as a CO₂ source. Reaction conditions: 1a (0.2 mol), DMSO (1 mL), air, r.t., 96 h.

3.2 Mechanical studies

To further investigate the reaction mechanism, a series of crucial controlled experiments were conducted (Scheme 2). First, it was observed that with the addition of catalytic amounts of Ag₂CO₃, the reaction could not proceed in the absence of CO₂, which served as the carbon source. However, when the amount of Ag₂CO₃ was increased to 2 eq., the target product was obtained with a yield of 32%. In contrast, the addition of 2 eq. of Ag₂O did not result in the formation of the target product, indicating that the carbonate moiety in Ag₂CO₃ can directly participate in the reaction as a carbon source.

Scheme 2 Control experiments.

Raman measurements were also performed on compound 1a before and after the addition of Ag₂CO₃ to probe the interaction between the propargylamine and the catalyst. As shown in Fig. S2, the Raman peak at 2039 cm⁻¹, assigned to the alkyne groups, exhibited a slight redshift after Ag₂CO₃ was introduced, compared to that of pure 1a, confirming that Ag₂CO₃ effectively activates the propargylamine (Fig. S2). To obtain further insights into the carboxylative cyclization of CO₂ with propargylamines, the reaction process catalyzed by Ag₂CO₃ was monitored using in situ FT-IR spectroscopy. As shown in Fig. 4, characteristic peaks emerged at 1650 and 1387 cm⁻¹ after 10 min, corresponding to the stretching vibrations of the carbamate group (NCOO⁻).^37–39 Notably, the difference between the antisymmetric ν_s(COO⁻) and symmetric ν_s(COO⁻) stretching vibrations exceeded 200 cm⁻¹, suggesting the formation of a metallacarboxylate species through coordination between silver and a monodentate carbamate. Furthermore, the intensities of the peaks attributed to the newly formed C=O (1789 cm⁻¹), C=C (1689 cm⁻¹), and C–O–C (1234 cm⁻¹) bonds in the oxazolidinone product progressively increased as the reaction proceeded.

Fig. 4 In situ FT-IR spectra for the reaction process of carboxylative cyclization of 1a with CO₂ into 2a over the Ag₂CO₃ catalyst.

Based on previous reports^40–42 and our observations, we propose a possible mechanism (Scheme 3). Firstly, Ag₂CO₃ is introduced to activate the propargylamine (1a), forming intermediate A. Simultaneously, Ag₂CO₃ acts as a base, facilitating the removal of the hydrogen atom from the amino group of 1a. This facilitates the smooth insertion of CO₂, leading to the formation of intermediate B. Additionally, the resulting HCO₃⁻, in the presence of Ag^I, reacts with 1a to yield intermediate C as well. Subsequently, intermediate B undergoes intramolecular cyclization to form intermediate C. Finally, intermediate C acquires a proton, resulting in the formation of the desired oxazolidinone product 2a. Simultaneously, the generated AgOH reacts with CO₂ to form Ag₂CO₃, thereby enabling the regeneration of the catalyst.

Scheme 3 Possible reaction mechanism.

Theoretical calculations elucidate the core mechanism of the Ag₂CO₃ catalysis: the Ag⁺ active site specifically coordinates with the alkyne bond (C≡C) of the propargylamine (Fig. 5). This coordination facilitates significant π back-bonding, which effectively lowers the energy of the alkyne's π anti-bonding orbital and markedly increases the electrophilicity of the terminal carbon atom. This electronic modification places the entire propargylamine molecule in an “activated” state. It is this activated alkyne that creates a favorable electronic environment for the subsequent nucleophilic attack on the CO₂ molecule, thereby significantly reducing the energy barrier for the carboxylative addition step to 2.0443 eV, substantially lower than the 2.2393 eV required under uncatalyzed conditions. These computational results confirm that Ag₂CO₃ functions by precisely activating the alkyne bond to drive the reaction, establishing it as the indispensable catalytic center for this crucial C–N bond formation step.

Fig. 5 DFT calculation. (a) Carboxylative cyclization of 1a with CO₂ under catalyst-free conditions. (b) Ag₂CO₃-catalyzed carboxylation and cyclization of CO₂ with acetylene 1a.

3.3 Application

To study the substrate scope of the catalysts, we investigated the efficiency of the carboxylative cyclization reactions of alkynylamines with various substituents under optimal reaction conditions in atmospheres of 10 vol% CO₂/N₂ mixed gas and 5 vol% CO₂/N₂ mixed gas (Scheme 4). Alkyl groups (e.g., methyl, ethyl, and methoxy) with electron-donating properties at the para-position on the benzene ring successfully participated in the carboxylative cyclization reaction under a 10 vol% CO₂ atmosphere, resulting in oxazolidinones (2b–2d) with yields exceeding 90%. Similarly, compounds with electron-withdrawing substituents at the para-position on the benzene ring were also effectively converted into the corresponding products (2f–2g). When methyl formate serves as the para-substituent, steric hindrance led to a reduced yield of 76%, although related oxazolidinone 2f was still generated. Furthermore, an acetylenic propylamine containing a 2-thiazolyl heterocyclic structure effectively participated in generating the target heterocyclic product 2i with an 82% yield. Even the terminal acetylenic structure of propylenic amine successfully yielded target product 2j under similar conditions. Additionally, utilizing 1-phenylethynyl-cyclohexylamine as the substrate resulted in a yield of up to 90% for the target product. Under a 5 vol% CO₂ atmosphere, substrates with various substituents were also successfully carboxylated with CO₂, achieving carboxylation cyclization with target product yields over 60%. Thus, the catalytic platform utilizing Ag₂CO₃ as the catalyst demonstrates broad applicability across different substrates.

Scheme 4 Substrate scope for the carboxylative cyclization of propargylamine with low-concentration CO₂. Reaction conditions: ^a 1 (0.2 mmol), Ag₂CO₃ (5 μmol), balloon containing 5 vol% CO₂/N₂ mixed gas, DMSO (1 mL), 12 h, r.t., isolated yields. ^b Balloon containing 10 vol% CO₂/N₂ mixed gas.

Currently, the majority of research on the catalytic conversion of CO₂ is focused on high-purity CO₂ or simulated gases with specific CO₂ concentrations. Relatively little research has been conducted on the in situ conversion of low-concentration CO₂ in actual-production flue gas. However, for practical applications, it is crucial to assess the ability of a catalyst to effectively fix CO₂ under actual flue gas conditions for practical application. Studies have indicated that the efficiency of CO₂ fixation in actual flue gas is not only affected by the concentration of CO₂ but also by the presence of other components in the flue gas. Small amounts of SO₂, NO₂ and other gases can deactivate the catalyst and significantly interfere with the reaction. Therefore, we conducted experiments using simulated gases containing varying concentrations of SO₂ and NO₂ based on their actual content in flue gas, to investigate their effects on carbon sequestration efficiency (Fig. 6). The results demonstrated that a catalytic platform utilizing Ag₂CO₃ as the catalyst exhibited good tolerance of SO₂ and NO₂. Specifically, the yield of the target product did not show significant decreases within the concentration ranges of 48 mg m⁻³ to 285 mg m⁻³ for SO₂ and 34 mg m⁻³ to 149 mg m⁻³ for NO₂.

Fig. 6 Effects of SO₂ and NO₂ contents on reaction yield. Reaction conditions: 1a (0.2 mmol), Ag₂CO₃ (5 μmol), simulated gas, DMSO (1 mL), 12 h, r.t. Sample 1: 10% CO₂, 48 mg m⁻³ SO₂, 34 mg m⁻³ NO₂, 0.67% O₂, 0.033% CO, N₂ balance. Sample 2: 10% CO₂, 89 mg m⁻³ SO₂, 71 mg m⁻³ NO₂, 0.67% O₂, 0.033% CO, N₂ balance. Sample 3: 10% CO₂, 196 mg m⁻³ SO₂, 149 mg m⁻³ NO₂, 0.67% O₂, 0.033% CO, N₂ balance. Sample 4: 10% CO₂, 285 mg m⁻³ SO₂, 0.67% O₂, 0.033% CO, N₂ balance.

In addition to the various gaseous components in flue gas that can affect the catalytic performance, the presence of certain metallic impurities may also interfere with the catalytic process. According to different emission standards, the permitted emission levels of various metals, based on their toxicity, typically range from a few tenths to over ten milligrams per cubic meter. To investigate the impact of these metal concentrations, six representative metals—mercury, cadmium, lead, copper, nickel, and cobalt—were simultaneously introduced into the solution. Under different concentration gradients, the carboxylation reaction between carbon dioxide and propargylamine proceeded smoothly, with the yield of the target product consistently exceeding 90% (Fig. 7), demonstrating the strong interference resistance of the catalytic process.

Fig. 7 Effect of simulated impurity metals (mercury, cadmium, lead, copper, nickel, and cobalt) on catalytic reaction. Reaction conditions: 1a (0.2 mmol), Ag₂CO₃ (5 μmol), balloon containing 10 vol% CO₂/N₂ mixed gas, DMSO (1 mL), 12 h, r.t. Sample 1: 0 mg L⁻¹ mercury, cadmium, lead, copper, nickel, and cobalt. Sample 2: 5 mg L⁻¹ mercury, cadmium, lead, copper, nickel, and cobalt. Sample 3: 10 mg L⁻¹ mercury, cadmium, lead, copper, nickel, and cobalt. Sample 4: 15 mg L⁻¹ mercury, cadmium, lead, copper, nickel, and cobalt. Sample 5: 20 mg L⁻¹ mercury, cadmium, lead, copper, nickel, and cobalt.

To further explore the practical applicability of this catalytic system, we collected representative exhaust gas samples from various typical industrial scenarios, including coal-fired flue gas, automobile exhaust, cement kiln exhaust gas, lime kiln exhaust gas, and anaerobic fermentation gas. The concentration of CO₂ in these exhaust gas samples ranged from 8 vol% to 35 vol% (Fig. 8). Experimental results indicated that when using Ag₂CO₃ as a catalyst, CO₂ in these actual exhaust gases can be directly converted into oxazolidinones without the need for purification or enrichment treatment, achieving target product yields exceeding 88%. This outcome fully demonstrates the efficiency and practicality of this catalytic system in treating exhaust gases containing low-to-moderate concentrations of CO₂.

Fig. 8 Conversion of CO₂ in actual exhaust gas derived from different scenarios. Reaction conditions: 1a (0.2 mmol), Ag₂CO₃ (5 μmol), balloon containing different kinds of exhaust gases, DMSO (1 mL), 12 h, r.t.

Based on coal-fired flue gas as a representative example, the applicability of the catalytic system to substrates was tested by using this exhaust gas as a source of CO₂ (Scheme 5). The coal-fired flue gas can be used for carboxylation and cyclization reactions with other alkynyl propylamines containing various substituents, producing the corresponding oxazolidinone compounds with excellent yields. Furthermore, we investigated the substrate suitability of our catalytic system using air as the carbon source, and it is gratifying to note that oxazolidinones with different substituents can also be obtained in moderate yields.

Scheme 5 Substrate scope for the carboxylative cyclization of propargylamine with coal-fired flue gas (or air). Reaction conditions: ^a 1 (0.2 mmol), Ag₂CO₃ (5 μmol), balloon containing coal-fired flue gas (8 vol% CO₂), DMSO (1 mL), 12 h, r.t., isolated yields. ^b Ag₂CO₃ (0.03 mmol), air, 96 h.

To further validate the feasibility of this method in practical applications, we conducted a scaled-up reaction to assess its applicability in large-scale operations. Our research revealed that by using 5 vol% CO₂/N₂ mixed gas as the carbon source and elevating the substrate concentration in the model reaction to the gram level, we were able to maintain a high yield of target product 2a at 73% (Scheme 6), without any significant yield reduction. This finding suggests that the reaction exhibits excellent scalability and retains high activity during large-scale synthesis.

Scheme 6 A gram-scale carboxylative cyclization. Reaction conditions: 1a (6.5 mmol), Ag₂CO₃ (0.1625 mmol), 5 vol% CO₂/N₂ mixed gas, aeration rate 1 mL min⁻¹, DMSO (33 mL), 12 h, r.t.

In addition, the cycling stability of the catalyst is one of the crucial parameters for evaluating its potential in industrial applications. The experimental results demonstrate that using 10 vol% CO₂/N₂ mixed gas as the CO₂ source, the catalyst maintained stable performance over five consecutive reaction cycles, with the target product yield consistently remaining above 80% (Fig. 9). The slight decrease in yield during the cycling tests can be attributed to the partial conversion of Ag₂CO₃ to Ag₂O. The XRD patterns of the catalyst before and after use are highly consistent with the standard diffraction peaks of silver carbonate, suggesting that silver carbonate may be successfully regenerated after the reaction. In addition, the XRD results indicate the presence of a small amount of silver oxide in the used catalyst (Fig. S3). The analysis results of X-ray photoelectron spectroscopy (XPS) further confirm this point. Specifically, the XPS spectra (Fig. S4 and S5) show characteristic peaks in the C 1s (Fig. S4b and S5b), O 1s (Fig. S4c and S5c), and Ag 3d (Fig. S4d and S5d) regions. The high-resolution O 1s spectrum shown in Fig. S4c displays a single peak centered at 531.1 eV and 532.9 eV, assigned to oxygen in the carbonate (CO₃²⁻) ion.⁴³ In the used catalyst, the O 1s spectrum shown in Fig. S5c reveal three components at binding energies of 529.1 eV, 531.1 eV, 532.9 eV and 534.5 eV, which are assigned to oxygen in carbonate, oxygen in silver oxide (Ag₂O), and oxygen from adsorbed water, respectively.⁴⁴ Furthermore, the high-resolution Ag 3d spectra in both Fig. S4d and S5d exhibit two distinct peaks with Ag 3d_5/2 and Ag 3d_3/2 binding energies centered at 368.1 eV and 374.1 eV, respectively, confirming that silver is present as Ag⁺. ICP-MS analysis of the filtered reaction solution revealed that the silver content in the solution was only about 81 μg, indicating that 94% of the silver had been successfully separated from the reaction mixture.

Fig. 9 Catalyst recycling experiment. Reaction conditions: 1a (0.2 mmol), Ag₂CO₃ (5 μmol), balloon containing 10 vol% CO₂/N₂ mixed gas, DMSO (1 mL), 12 h, r.t.

Subsequently, we developed a cyclic carbon sequestration device that utilizes a peristaltic pump to circulate the gas, thereby reducing gas consumption and enhancing sequestration efficiency (Fig. 10). Experimental results demonstrated that a simulated 15 vol% CO₂/N₂ mixed gas was circulated in the system for 12 h. A portion of the CO₂ dissolved in the solvent and was diluted by the dead volume of the device, while another portion underwent reaction with the substrate to produce compound 2a. Consequently, the CO₂ concentration decreased from 15% to 3%. The average yield of 2a in the four reaction modules reached 92%.

Fig. 10 Cyclic carbon sequestration unit. Reaction conditions: 1a (2 mmol × 4), Ag₂CO₃ (0.05 mmol × 4), 15 vol% CO₂/N₂ mixed gas (2 L), aeration rate: 1 mL min⁻¹, DMSO (10 mL × 4), 12 h, r.t.

4. Conclusions

The cost-effective and readily available Ag₂CO₃ was employed as a catalyst to effectively catalyze the carboxylative cyclization reaction of alkynylamine with a gas mixture of low-concentration CO₂, yielding oxazolidinone derivatives with high efficiency under mild conditions. The catalytic system demonstrates the capability to efficiently convert CO₂ present in actual exhaust gas under ambient temperature and pressure conditions, resulting in yields of oxazolidinone derivatives as high as 97%. Even when reacting with ultra-low concentrations of CO₂ present in air, a yield of oxazolidinone derivatives of 45% can be achieved, while maintaining substrate universality. Furthermore, the reaction is scalable, and the designed cyclic carbon sequestration device achieves effective reduction of CO₂ concentrations in actual flue gas. This study holds great significance for addressing the challenge of resource utilization of CO₂ from industrial flue gas and provides new methods and technical means for the green synthesis of oxazolidinones.

Conflicts of 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.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5gc04212g.

Acknowledgements

We acknowledge the financial support of the Guangxi Natural Science Foundation of China (2024GXNSFDA010013), the Guangxi Special Project for Science and Technology Base and Talents (GuikeAD25069083), and the National Natural Science Foundation of China (52360014 and 52170107).

References

1. J.-P. Yue, J.-C. Xu, H.-T. Luo, X.-W. Chen, H.-X. Song, Y. Deng, L. Yuan, J.-H. Ye and D.-G. Yu, Nat. Catal., 2023, 6, 959–968.
2. Y. Li, L. Zeng, G. Pang, X. Wei, M. Wang, K. Cheng, J. Kang, J. M. Serra, Q. Zhang and Y. Wang, Appl. Catal., B, 2023, 324, 122299.
3. C. P. O’Brien, R. K. Miao, A. S. Zeraati, G. Lee, E. H. Sargent and D. Sinton, Chem. Rev., 2024, 124, 3648–3693.
4. X. Yan, M. Yuan, Y.-L. Yuan, P. Su, Q. Chen and F.-X. Xiao, Chem. Sci., 2024, 15, 10625–10637.
5. Q. Chen, J.-L. Li, Q.-L. Mo and F.-X. Xiao, Chem. Eng. J., 2024, 497, 154584.
6. Q. Chen, Y. Chen, J. Zhu, Z. Li and F. Xiao, Adv. Funct. Mater., 2025, 35, 2417139.
7. P. Yan, F. Ji, W. Zhang, Z. Mo, J. Qian, L. Zhu and L. Xu, J. Colloid Interface Sci., 2023, 634, 1005–1013.
8. C. Ban, Y. Wang, J. Ma, Y. Feng, X. Wang, S. Qin, S. Jing, Y. Duan, M. Zhang, X. Tao, L. Gan and X. Zhou, Chem. Eng. J., 2024, 488, 150845.
9. X. Yan, J.-H. Dong, J.-Y. Zheng, Y. Wu and F.-X. Xiao, Chem. Sci., 2024, 15, 2898–2913.
10. N. Onishi and Y. Himeda, Chem Catal., 2022, 2, 242–252.
11. B. Tang and F.-X. Xiao, ACS Catal., 2022, 12, 9023–9057.
12. T. Xiong, Q. Xia, X. Zhou, S. Li, F. Cui, H. Tang, Y. Pan and Y. Liang, Adv. Synth. Catal., 2023, 365, 2183–2187.
13. P. M. Srinivasappa, D. Prasad, N. K. Chaudhari, A. K. Samal, R. Thapa, E. E. Siddharthan and A. H. Jadhav, ACS Appl. Mater. Interfaces, 2023, 15, 21994–22011.
14. R. Wang, J. Wan, H. Guo, B. Tian, S. Li, J. Li, S. Liu, T. D. James and Z. Chen, Carbon, 2023, 211, 118118.
15. P. Tyagi, D. Singh, N. Malik, S. Kumar and R. S. Malik, Mater. Today, 2023, 65, 133–165.
16. T. Yue, B. Ren, W. Zhang, X. Lu, W. Ren and D. J. Darensbourg, Angew. Chem., 2021, 133, 4361–4367.
17. S. Liuand and X. Wang, Curr. Opin. Green Sustainable Chem., 2017, 3, 61–66.
18. H. Ishaq and C. Crawford, RSC Sustainability, 2025, 3, 3652–3680.
19. C. Fan, R. Wei, T. Cheng, R. Sun, H. Zhang and H. Long, Chem. Eng. J., 2024, 498, 155379.
20. C. Hepburn, E. Adlen, J. Beddington, E. A. Carter, S. Fuss, N. Mac Dowell, J. C. Minx, P. Smith and C. K. Williams, Nature, 2019, 575, 87–97.
21. L. Wang and J. Yu, Chem Catal., 2022, 2, 428–430.
22. P.-B. Chen, J.-W. Yang, Z.-X. Rao, Q. Wang, H.-T. Tang, Y.-M. Pan and Y. Liang, J. Colloid Interface Sci., 2023, 652, 866–877.
23. X. Li, S. Wang, L. Li, Y. Sun and Y. Xie, J. Am. Chem. Soc., 2020, 142, 9567–9581.
24. D. A. Henckel, P. Saha, S. Rajana, C. Baez-Cotto, A. K. Taylor, Z. Liu, M. G. Resch, R. I. Masel and K. C. Neyerlin, ACS Energy Lett., 2024, 9, 3433–3439.
25. K. Takeuchi, H. Koizumi, H. Nagae, K. Matsumoto, N. Fukaya, K. Sato and J.-C. Choi, J. CO₂ Util., 2024, 83, 102814.
26. Y. Xi, C. Su, N. A. Ottinger and Z. G. Liu, Appl. Catal., B, 2021, 284, 119749.
27. Y. Liang, Q. Xia, J.-Y. Yang, X.-J. Meng, X.-M. Hu and Y.-M. Pan, Chem. Eng. J., 2024, 498, 155305.
28. P. Rani, R. Das and C. M. Nagaraja, Inorg. Chem. Front., 2025, 12, 430–478.
29. G. Liu, X. Li, M. Liu, S. Yang, X. Yang, X. Chen, W. Wei, Q. Xu and G. Zeng, ACS Catal., 2024, 14, 11076–11086.
30. X. Liao, Z. Wang, Z. Li, L. Kong, W. Tang, Z. Qin and J. Lin, Chem. Eng. J., 2023, 471, 144455.
31. D. DeSantis, J. A. Mason, B. D. James, C. Houchins, J. R. Long and M. Veenstra, Energy Fuels, 2017, 31, 2024–2032.
32. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50.
33. G. Kresse and J. Furthmüller, Phys. Rev. B:Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
34. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
35. P. E. Blöchl, Phys. Rev. B:Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
36. D. J. Chadi, Phys. Rev. B, 1977, 16, 1746–1747.
37. Y. Zhang, X. Lan, F. Yan, X. He, J. Wang, L. Ricardez-Sandoval, L. Chen and G. Bai, Green Chem., 2022, 24, 930–940.
38. P.-B. Chen, J.-W. Yang, Z.-X. Rao, Q. Wang, H.-T. Tang, Y.-M. Pan and Y. Liang, J. Colloid Interface Sci., 2023, 652, 866–877.
39. Z. Rao, P. Chen, J. Xu, Q. Wang, H. Tang, Y. Liang and Y. Pan, ChemSusChem, 2023, 16, e202300170.
40. J.-F. Qin, B. Wang and G.-Q. Lin, Green Chem., 2019, 21, 4656–4661.
41. K. Sekine, R. Kobayashi and T. Yamada, Chem. Lett., 2015, 44, 1407–1409.
42. K. Yoo, D. G. Jwa, H.-E. Lee, H. J. Kim, C. Kim and M. Kim, Catalysts, 2019, 9, 1032.
43. H. Zeng, Z. Yu, L. Shao, X. Li, M. Zhu, Y. Liu, X. Feng and X. Zhu, Desalination, 2020, 491, 114558.
44. S. Liang, D. Zhang, X. Pu, X. Yao, R. Han, J. Yin and X. Ren, Sep. Purif. Technol., 2019, 210, 786–797.

---