High-efficiency green catalytic conversion for waste CS₂ by non-noble metal cage-based MOFs: an access pathway to high-value thiazolidine-2-thione

Abstract

Green and effective disposal of carbon disulfide (CS₂) waste into high-valued chemicals under mild conditions is meaningful yet challenging. Herein, a novel 3D cluster-based metal–organic framework (MOF) {(Me₂NH₂)₂[Co₃(μ₃-O)(XN)(BDC)₃]·4DMF·5MeOH}_n (compound 1) (XN = 4′-(4-pyridine)4,2′:2′,4′′-terpyridine, H₂BDC = terephthalic acid) assembled by [Co₁₅] and [Co₁₈] nano-cages was harvested, presenting excellent stability. Catalytic characterization demonstrated that compound 1 can efficiently promote the cycloaddition reaction of CS₂ with aziridines to form sole high-valued thiazolidine-2-thione upon 30 °C and 0.1 MPa for 6 h, which matches well with the atom economy and the sustainable development intention. Noteworthily, compound 1 is the mildest and most efficient catalyst for CS₂ treatment and can be reused at least ten times without significant activity degradation; it also retains excellent catalytic capacity in both gram-scale reaction and simulated CS₂ waste liquid, which lays a solid foundation for its practical application. Additionally, density functional theory (DFT) calculations further confirm the synergistic effect of the nanocage characteristic and the Me₂NH₂⁺ cation, which can significantly reduce the reaction energy barrier in this CS₂/aziridine coupling reaction system.

---

Introduction

Carbon disulfide (CS₂), a cheap and readily available industrial raw material, is usually used in the industrial production of cellulosic viscose fibers and CCl₄ as well as a precursor in the production of thiourea.¹ However, CS₂ is also an environmental toxin, and its neurological and reproductive toxicity following long-term industrial use have been demonstrated.^2,3 To date, the effective treatment for CS₂ waste in industrial production has always been a tough question, with incineration being the most common disposal method.⁴ However, the CS₂ incineration products which include SO₂ and H₂S are released into the air, leading to fog, haze and acid rain, which poses serious threats to the environment and human health. Considering that CS₂ is a convenient C1 building block for the syntheses of S and C-containing compounds, the CS₂ waste in exhaust gases or liquids may be employed and converted into high-value-added chemicals.⁵ Up to now, a lot of effort has been made to synthesize high-valued chemicals such as sulfur-containing polymers,^6–9 thiocarbamate intermediates,^10,11 cyclic thiocarbonates,¹² thioketones,^13–15 and others from CS₂. Among these chemicals, thiazolidine-2-thiones have received extensive attention due to their applications as intermediates in both organic syntheses and medicinal chemistry. Up to now, efficient synthesis methods of thiazolidine-2-thiones mainly include the multicomponent reaction of CS₂, amines, and sulfoxonium ylides;¹⁶ the reaction of primary amines, carbon disulphide, with nitro epoxides;¹⁷ or the cycloaddition of CS₂ with aziridines to thiazolidine-2-thione.¹⁸ Significantly, the coupling of CS₂ and aziridine has become one of the most promising strategies in consideration of its atomic economy. Nevertheless, the catalysts reported so far all require large doses (10% mmol), long reaction time (48 h), or high reaction temperatures (100 °C).^19–21 Therefore, the development of an efficient and mild catalyst is necessary yet challenging.

Metal-organic frameworks (MOFs) are active in many fields, including catalysis, luminescence sensing, lithium-ion battery, supercapacitor and gas adsorption due to their designable structure, tunable pore size, large specific surface area and characteristic functional groups.^22–27 Particularly, cluster-based MOFs with richer metal clusters tend to form large nanocages, which can help in gathering catalytic substrates, exhibiting superior catalytic performances through domain-limiting effects.^28–31 Moreover, the multinuclear metal clusters of cluster-based MOFs can bring higher stability and more metal-active sites to the framework, which provides strong support for the recyclability and activity of the catalytic reaction.^32–35 Many cluster-based MOFs with different SBUs and organic ligands have been designed to meet the requirements of various organocatalytic reactions. Although cluster-based MOFs have demonstrated good catalytic activity as non-homogeneous catalysts in the preparation of cyclic carbonates,³⁶ oxazolidinones,^37,38 methyl phenyl sulfone³⁹ and tertiary amines,⁴⁰ only one example has been employed in the synthesis of thiazolidine-2-thiones with harsh terms.

In this study, a 3D cluster-based MOF {(Me₂NH₂)₂[Co₃(μ₃-O)(XN)(BDC)₃]·4DMF·5MeOH}_n (1) with [Co₁₅] and [Co₁₈] nanocages has been harvested by solvothermal method and structurally characterized, presenting a large specific surface area and good chemical stability. Catalytic studies revealed that compound 1 can efficiently catalyze the cycloaddition reaction of CS₂ and aziridine into thiazolidine-2-thiones (99%) as a non-noble metal catalyst under the mildest conditions of 30 °C and 0.1 MPa for 6 h without by-products. Moreover, catalyst 1 can be recycled at least ten times without significant decrease in catalytic activity, showing good turnover frequency (TOF = 157 h⁻¹) in gram scale-up experiments. More importantly, compound 1 still maintained a high catalytic efficiency in the simulated CS₂ effluent, suggesting the great potential to treat CS₂ contamination in practical application. It is worth mentioning that compound 1 is among the mildest and most efficient catalysts for CS₂ treatment at the current times. Finally, the catalytic reaction mechanism has been speculated and further elucidated through both ¹H NMR analyses and density functional theory (DFT) calculations, providing an experimental and theoretical guidance for the preparation of a novel catalytic material for CS₂ conversion.

Experimental

Materials and general methods

All the chemicals and solvents used in this work for crystal synthesis were purchased without further purification, including cobalt(II) chloride hexahydrate (CoCl₂·6H₂O), XN (4′-(4-pyridine)4,2′:2′,4′′-terpyridine), H₂BDC (terephthalic acid), DMF (N,N-dimethylformamide), and MeOH (methanol). All aziridine substrates were synthesized according to the references reported previously.⁴¹ Powder X-ray diffraction (PXRD) data were carried out on a PANalytical B.V. Empyrean diffractometer in 5°–50° 2θ range by using Cu-Kα radiation. Fourier transform infrared (FT-IR) spectroscopy was recorded in the range of 4000–400 cm⁻¹ on a Bruker VERTEX70 spectrophotometer as KBr pellets. Thermogravimetric analysis (TGA) was conducted using a TA-TGA 550 analyzer at a heating rate of 10 °C min⁻¹ from room temperature to 800 °C under N₂ atmosphere. X-ray photoelectron Spectrometer (XPS) analysis was performed on a Thermo SCIENTIFIC K-Alpha spectrometer, and used the Al-Kα radiation as X-ray source with monochromatized X-ray radiation. The ¹H NMR spectra were carried out by a Bruker AVANCE NEO 600 M spectrometer and 1,3,5-trimethoxybenzene served as an internal reference with chemical shifts in 6.08 ppm. The Co²⁺ content of filtrate after the tenth cycle was detected by ICAP-7400 spectral analysis.

Synthesis of compound 1

A mixture of CoCl₂·6H₂O (0.1 mmol, 0.0238 g), XN (0.04 mmol, 0.0124 g), H₂BDC (0.1 mmol, 0.0166 g), DMF (4 mL), MeOH (4 mL), and tetrafluoroboric acid (100 μL, AR > 40%) was transferred to a Teflon-lined reactor and kept at 160 °C for 96 h. Then the reactor was cooled to room temperature within 24 h and Orange-red hexagonal crystals were obtained (Fig. S1 and S2†). Then, these samples were washed with DMF, and air-dried to give a yield of 66% (based on CoCl₂·6H₂O).

Single crystal X-ray diffraction analysis

The single-crystal data of compound 1 was corrected on a Bruker APEX-II CCD X-ray single-crystal diffractometer equipped with graphite-monochromic Mo Kα radiation (λ = 0.71073) at 296.15 K by selecting the suitable crystal sample. The diffraction absorption data were gathered by the SADABS program. The structure of compound 1 was solved directionally using the SHELXS and SHELXL programs in Olex2,⁴² and convergence was refined using F²-based full-matrix least squares techniques. Anisotropic thermal parameters were assigned to all non-hydrogen atoms. All hydrogen atoms were placed in the computational positions and isotropically refined by riding the model. The diffraction contributions of all solvent molecules in compound 1 were removed using SQUEEZE in PLATON and the composition and content of solvent molecules were determined in combination with TGA.⁴³ Table S1† lists the detailed crystal data and structural improvement results for 1. The CCDC number of compound 1 is 2337117.†

Gas adsorption measurements

Gas adsorption measurement of compound 1 was performed on a Micrometrics 3-Flex gas adsorption analyzer using ultrapure (99.995%) N₂ gas. Before the adsorption measurement, 100 mg sample was activated by heating at 120 °C under vacuum for 24 h. Then, N₂ adsorption tests were performed at 77 K for full pore testing. The temperature was achieved by using liquid N₂.

General procedure for the catalytic reactions

The cycloaddition reaction of CS₂ with aziridines: in a typical experiment, a mixture of aziridine, tetrabutylammonium bromide (TBAB), CS₂ and catalyst 1 was sequentially added to a 20 mL Schlenk tube. The reactor was then placed in an oil bath and stirred continuously for a certain time at a corresponding temperature. At the end of the reaction, the reactor was cooled to room temperature. The yield of the corresponding product was determined by ¹H NMR analysis using 1,3,5-trimethoxybenzene as an internal standard.

Catalyst cycling experiment

After the catalytic reaction, 1,3,5-trimethoxybenzene was added to the reaction tubes as the internal standard substance and the mixture in the tubes was separated by centrifugation. Then, a ¹H NMR analysis of the liquid product was conducted to calculate the catalytic yield. The solid was then washed three times using fresh CH₂Cl₂ and centrifuged sufficiently. Finally, this solid was dried under vacuum at 60 °C for 6 h before proceeding to the next cycle of the catalytic experiment.

Computational details

All calculations in this work were performed using Gaussian 09 program package.⁴⁴ Full geometry optimizations were performed to locate all the stationary points, using the B3LYP⁴⁵ with the 6-31G(d,p)⁴⁶ basis for C, H, O, N, and S; and Lanl2dz basis for Co.⁴⁷ Dispersion corrections were computed with Grimme's D3(BJ) method in optimization.⁴⁸ Frequency analysis was carried out to verify the optimized geometry to be a minimum or transition structure and to obtain the thermal corrections for the Gibbs free energy. Intrinsic reaction coordinate (IRC) calculations were performed to confirm the transition state (TS) structures connecting the expected minima.⁴⁹ Single-point energy calculations were obtained at the B3LYP-D3(BJ) with 6-311+G(d,p) for C, H, O, N, and S; and SDD for Co of theory. Harmonic vibrational frequency was performed at the same level to guarantee that there was no imaginary frequency in the molecules, i.e. they locate on the minima of potential energy surface. Convergence parameters of the default threshold were retained (maximum force within 4.5 × 10⁻⁴ Hartrees per Bohr and root mean square (RMS) force within 3.0 × 10⁻⁴ Hartrees per Radian) to obtain the optimized structure. The optimal structure was identified given that all calculations for structural optimization were successfully converged within the convergence threshold of no imaginary frequency, during the process of vibration analysis.

Results and discussion

The orange polyhedral crystals of compound 1 were obtained by hydrothermal reaction of CoCl₂ ligands H₂BDC and XN at 160 °C for 4 days. Single-crystal X-ray diffraction analyses show that the compound 1 belongs to the hexagonal crystal system with P6₃/mmc space group (ESI, Table S1†). The minimal repeat unit of framework 1 includes a [Co₃(μ₃-O)] cluster, an XN ligand, three deprotonated BDC²⁻ anions, two free Me₂NH₂⁺, four DMF, and five MeOH molecules (Fig. 1a, b and Fig. S3a†). Me₂NH₂⁺ is derived from the decomposition of DMF (N,N-dimethylformamide) under acidic conditions and distributed as a positive counter-ion in the three-dimensional cage-based framework of compound 1. The center Co²⁺ ions are six-coordinated, and each three Co²⁺ form a trinuclear [Co₃(μ₃-O)] secondary building units (SBUs) with one μ₃-O²⁻ ion. The O–Co–O angles are in the range from 89.50 to 175.14°, the Co(1)–N(1) distance is 2.163 Å, while the Co–O bonds range from 2.033 to 2.072 Å (Table S2†).⁵⁰ A 3D structure was assembled through continuous connection between the ligands and [Co₃(μ₃-O)] clusters, presenting a triangular cavity with pores of about 6 × 12 Ų along an a direction (Fig. 1c). The solvent accessible free volume of compound 1 was estimated to be 58.0% by calculation using PLATON software.⁵¹

Fig. 1 Schematic representation of compound 1. (a) and (b) a [Co₃(μ₃-O)]-cluster connects three XN ligands and six H₂BDC ligands in different directions; (c) view of the 3D porous structure of compound 1; (d) the skeleton structures of [Co₁₅] and [Co₁₈] nanocages and their 3D crystal packing, insets: the optical microscopic image of crystal 1 prepared freshly; (e) the tiling representation of [Co₁₅] and [Co₁₈] nanocages as well as the simplifying form of 3D framework. Color codes for atoms: cyan, Co; yellow, S; gray, C; red, O; blue, N.

Interestingly, the different SBUs are connected by a triangular XN ligand and a linear H₂BDC ligand to construct two types of nanocages (Fig. 1d). Thereinto, the larger triangular biconical nanocage has a spherical cavity with a diameter about 12 Å, while the smaller nanocage has a cylindrical cavity with a diameter of 12 Å and a height of 6.0 Å. The two nanocages are interconnected and expanded by sharing vertices and faces, presenting a 3D porous cluster-based framework of compound 1. From the topological point of view, each [Co₃(μ₃-O)]-cluster is coordinated to six H₂BDC ligands and three XN ligands, and thus the [Co₃(μ₃-O)]-cluster can be regarded as a 9-connected node. Also, each XN ligand and H₂BDC ligand can be considered as 3-connected and 2-connected nodes to coordinate different Co²⁺ ions. Because of such simplification, compound 1 can be described as corresponding to a pacs-type topology architecture (Fig. S3b†). Additionally, the tiling representation of framework 1 with different cages is shown in Fig. 1e.

The valence state of Co ion in compound 1 has been determined by the X-ray photoelectron spectroscopy (XPS). The binding energies of Co 2p_1/2 and 2p_3/2 in compound 1 are 796.8 eV and 781.0 eV, respectively, which are accompanied by two vibrational satellite peaks (801.8 and 785.2 eV), confirming the presence of Co²⁺ ion (Fig. S4†). The FT-IR spectra of compound 1 and ligands are presented in Fig. S5.† The characteristic peaks of H₂BDC ligand in the range of 3200–2600 cm⁻¹ should be assigned to the stretching vibration of –OH in –COOH. The absorption intensities of the characteristic peaks in this range decreased significantly when the metal ions were coordinated with the carboxyl oxygen atoms in the ligand to form compounds. Meanwhile, the asymmetric stretching vibration peaks in –COOH attributed to C=O shift from 1689 cm⁻¹ and 1423 cm⁻¹ to 1660 cm⁻¹ and 1381 cm⁻¹ respectively after the formation of compounds. The powder X-ray diffraction (PXRD) patterns of compound 1 are in good agreement with the simulated ones, indicating the high phase purity of compound 1 (Fig. S6†). The thermostability and chemical stability of compound 1 were also investigated. Thermogravimetric analysis (TGA) demonstrated that compound 1 maintains stability up to about 300 °C, which should be attributed to the removal of free DMF and MeOH molecules (Fig. S7†). Then, the framework collapsed gradually after 350 °C, proving good thermal stability. Furthermore, sample 1 was immersed in different organic solvents (DMF, dichloromethane, methanol, ethanol, isopropanol, acetonitrile or tetrachloromethane) for 24 h, and their PXRD patterns still matched well with the simulated patterns, indicating excellent solvent stability of compound 1 (Fig. S8†).

To further investigate the permanent void ratio of compound 1, solvent exchange was performed with CH₂Cl₂ for 2 days and then the solvent was degassed under high vacuum at 100 °C for 6 h. The N₂ adsorption isotherm at 77 K has been determined. The adsorption measurements showed that the maximum N₂ adsorption of compound 1 at different relative pressures was 217.2 cm³ g⁻¹. The BET comparison area was 404.4 m² g⁻¹, while the adsorption average pore size was about 5.9 Å, which coincided with the pore sizes of the two types of nano-cages, confirming its microporous structure (Fig. S9 and S10†).

Inspired by the porous nanocage architecture characteristic of compound 1 and the free dimethylamine cation in the channels, the catalytic cycloaddition reaction of CS₂ and aziridine to form thiazolidine-2-thiones was carried out. First, a series of aziridine substrates were prepared according to the reported literatures.⁴¹ Then, the optimal reaction conditions were explored by selecting 1-ethyl-2-phenylaziridine as a model substrate. Subsequently, the reaction was carried out at 60 °C under atmospheric pressure with 10 mg 1 and 0.05 mmol tetrabutylammonium bromide (TBAB) as catalyst and co-catalyst respectively, as shown in Table 1. Catalyst 1 could catalyze the CS₂ transformation reaction efficiently with 3-ethyl-5-phenylthiazolidine-2-thione yield of 99% as the temperature dropping from 60 °C to 30 °C (Table 1, entries 1–4). As the temperature further fall to 25 °C, a 91% yield of catalytic product could still be obtained (Table 1, entry 5), indicating that the optimal temperature for the reaction is 30 °C. Subsequently, the effects of reaction time, catalyst dosage and co-catalyst amount on the reaction catalytic activities were explored at 30 °C.

Table 1 Optimization of reaction conditions for the cycloaddition of aziridine with CS₂ ^a

Entry	Time (h)	Co-catalyst (mmol)	T (°C)	Yield^b (%)
a Reaction conditions: 1-ethyl-2-phenylaziridine (1.0 mmol), CS2 (5 mmol), compound 1 (10 mg), TBAB (0.05 mmol), solvent-free, 6 h. b The total yield of the product was determined by ^1H NMR using 1,3,5-trimethoxybenzene as an internal standard. c 5 mg 1. d 0 mg 1. e 1-Ethyl-2-phenylaziridine (10 mmol), CS2 (50 mmol). f 1-Ethyl-2-phenylaziridine (1.0 mmol), CS2 (5 mmol), 2 mL MeOH.
1	6	0.05	60	>99
2	6	0.05	50	>99
3	6	0.05	40	>99
4	6	0.05	30	>99
5	6	0.05	25	91
6	4	0.05	30	87
7	2	0.05	30	47
8	6	0.025	30	88
9^c	6	0.05	30	89
10^d	6	0.05	30	51
11^e	6	0.05	30	94
12^f	6	0.05	30	86

---

When the reaction time reduced from 6 h to 4 h and then to 2 h, the corresponding yields decreased from 99% to 87% or 47%, respectively (Table 1, entries 6–7). Similarly, the yields of 3-ethyl-5-phenylthiazolidine-2-thione correspondingly decreased to 88% and 89% when the amounts of catalyst or co-catalyst were halved, suggesting the optimal dose for catalyst 1 (10 mg) and TBAB (0.05 mmol), as well as the synergistic effect between catalyst and cocatalyst. The synergistic action mentioned above was further verified by thermal filtration experiments. The substrate was barely able to be converted to the corresponding thiazolidine-2-thione product when catalyst 1 was isolated, suggesting that the presence of 1 efficiently facilitates this conversion process (Fig. S11†). Therefore, the catalytic reaction can be performed effectively under optimal conditions with 10 mg catalyst 1 and 0.05 mmol TBAB at 30 °C and 0.1 MPa for 6 h (thiazolidine-2-thione yield: 99%) with no by-products, which is in accordance with the concept of green chemistry and sustainable development.

In order to assess the possible practical application, the catalytic performance of catalyst 1 in the reaction of CS₂ with aziridines was further explored in scale-up experiments, and in a simulated industrial CS₂ waste liquid. In the first place, gram-scale tests were carried out by scaling up the amounts of 1-ethyl-2-phenylaziridine and CS₂ by a factor of ten without change in the reaction conditions for the catalyst and co-catalyst amounts at 30 °C and atmospheric pressure for 6 h (Table 1, entry 11). The corresponding product yield was 94% and the TOF reached 157 h⁻¹, which is much higher than that of other reported catalysts for this catalytic reaction, such as Dy₂₄-MOF (TOF = 51.1 h⁻¹), basic ion-exchange resins (TOF = 7.1 h⁻¹), and amidato lanthanide amides {tBuC₆H₄CONC₆H₃(iPr)₂Eu[N(SiMe₃)₂] THF}₂ (TOF = 1.3 h⁻¹) (Fig. S12†).^18–20

CS₂ has the advantages of being inexpensive, stable and easy to obtain, making it suitable for application in industrial production, especially in the manufacture of cellophane and viscose fibers.³ However, a large amount of CS₂ exhaust or waste liquid is released into the atmosphere after industrial production, resulting in haze and acid rain, and further posing a hazard to the environment and human health.⁵² Traditionally, industrial CS₂ exhaust and waste liquids can be treated by the physical adsorption of CS₂ by methanol at low temperature. Nonetheless, the subsequent treatment of the resulting mixed solution is costly and requires complex processing. In comparison, the cycloaddition reaction of CS₂ with aziridines into high-valued thiazolidine-2-thiones is more efficient and more environmentally friendly, helping to not only dispose CS₂ waste conveniently, but to also realize the reuse of resources. As shown in entry 12 in Table 1, 86% catalytic yield of the product thiazolidine-2-thione can be gained for simulated CS₂ waste liquid (MeOH: 2 mL, CS₂: 0.3 mL), indicating the practical potential of catalyst 1 in industry.

To further explore the catalytic properties of compound 1, the catalytic reactivities between CS₂ and aziridines with various substituents have been investigated under optimized conditions. The three-dimensional molecular modeling and size of all catalytic substrates and products have been given in Fig. S13 to S21.† As shown in Table 2, the corresponding yields of products decreased from 99% to 29% when the spatial site resistance of the substituent group on the N atom in aziridine substrates gradually increases (-ethyl, -propyl, -benzyl and isobutyl) (Table 2, entries 1a–1d). This experimental result above may be ascribed to the large size of the substrates, which hinders the activation of the substrate by catalyst 1 and the nucleophilic attack of Br⁻ from TBAB on the N atoms in aziridine. Then, similar yields of 3-methoxyethyl-5-phenylthiazolidine-2-thione (84%, Table 2, 1f) and 3-butyl-5-phenylthiazolidine-2-thione (80%, Table 2, 1e) could be obtained when the R1 group on substituent was replaced by butyl or methoxyethyl with similar spatial resistances, respectively. Subsequently, we investigated the electronic effects of the CS₂ cycloaddition reaction when the benzene ring on R₂ carries an electron-withdrawing or electron-giving group, respectively (Table 2, 1g–1j). The corresponding product yield with a –CH₃ electron-donating group on R₂ can reach 90%, which is higher than that of –F (83%) or –Cl (55%) electron-withdrawing substituted substrates. At the same time, the stronger the electron-withdrawing ability of the substituent group on the aryl ring, the higher the yield of the target product, as evidenced by the higher yield of the substrate with the –F substitution compared to that of the substrate with the –Cl substitution. All experimental results to date indicate that compound 1 can be used as an effective heterogeneous catalyst with practical potential for CS₂ conversion with large amounts of aziridines.

Table 2 Substrate scope for the cycloaddition reaction of CS₂ with different aziridines^a

a Reaction conditions: substrate (1.0 mmol), CS₂ (5 mmol), compound 1 (10 mg), TBAB (0.05 mmol), solvent-free, 6 h.

---

In practical applications, the evaluation of the recoverability of non-homogeneous catalysts and the decay of catalytic activity is indispensable. For recyclability, catalyst 1 can be easily separated from the mixed reaction system by centrifugation or filtration and used for the next catalytic reaction. Notably, the morphology of compound 1 during the reaction has some change in particle size after ten cycles, which may be attributed to the magnetic stir. Only a small portion of the crystals remained in regular form (Fig. S22†). As shown in Fig. 2, catalyst 1 can be reused at least ten times and its catalytic efficiency hardly decreases as the cycle number increases. Meanwhile, the high consistency of the PXRD patterns of sample 1 before and after the catalytic reaction, as well as the trace Co²⁺ leakage (0.16%) in the reaction filtrate and the XPS spectrum after the tenth cycle all proved that 1 still can maintain an intact skeletal structure and good crystallinity (Fig. S23, S24 and Table S3†). Furthermore, the FT-IR spectra before and after the catalytic reaction further corroborated this conclusion (Fig. S25†). The excellent cyclicity and high atomic economic efficiency of catalyst 1 in this reaction without the production of by-products and contaminants illustrate the concept of green chemistry and sustainable development.

Fig. 2 Recycling test of catalyst 1 in the CS₂ cycloaddition with 1-ethyl-2-phenylaziridine.

To gain insights into the mechanism of the cycloaddition reaction of CS₂ with aziridines catalyzed by compound 1, the interactions between aziridine substrate (1a), the catalyst and co-catalyst were explored by ¹H NMR analyses. As demonstrated in Fig. 3a, the proton signal peak of 1a is slightly lower with the addition of 1 or TBAB alone, whereas the signal peaks at δ = 1.65 (1.88) and 2.29 ppm were more broadened and dwarfed when 1 and TBAB were added simultaneously, proving the synergistic effect of catalyst 1 and TBAB in efficiently activating the substrate and catalysing its reaction with the CS₂. Moreover, the whole reaction process was monitored by ¹H NMR and the characteristic peaks of 1a gradually decreased until they disappeared entirely with the increase of time, being replaced by the characteristic peaks of 3-ethyl-5-phenylthiazolidine-2-thione (Fig. 3b).

Fig. 3 (a) The activation of 1-ethyl-2-phenylaziridine in different systems (in CDCl₃); (b) ¹H NMR monitoring of the cycloaddition reaction of aziridines with CS₂ in CDCl₃.

Then, we hypothesized that the high catalytic activity of compound 1 might also be related to the free Me₂NH₂⁺ in the pores to some extent. Firstly, the presence of counterbalancing ions in anionic/cationic frameworks may cause the pores to be highly polarized in comparison with the neutral framework, thus increasing their adsorption capacity for nonpolar molecules with high polarizability.^53,54 Secondly, there is likely to be an interaction between the Me₂NH₂⁺ as a hydrogen bond donor and CS₂ as a hydrogen bond acceptor. As shown in Fig. S26,† the signal peak attributed to –NH₂⁺ shifted from 7.98 ppm to 7.93 ppm in the system with the addition of CS₂, whereas the signal peaks of –CH₃ and DMSO-d6 did not show any displacement, which proved that a hydrogen bond has been formed in the imine group in Me₂NH₂⁺ with the CS₂ molecules.^55–57 Thus, the Me₂NH₂⁺ cation in the pores of compound 1 is likely to play a role in the adsorption and activation of CS₂ in the catalytic reaction. Thirdly, in order to confirm the essential function of anion MOF frame and Me₂NH₂⁺ cation in this catalytic reaction, an example of similar [Co₃]-MOF: Co(II/III) with neutral framework and mixed metal valences state was synthesized according to previous literature.⁵⁸ Then, the catalytic performance was evaluated in the cycloaddition reaction of CS₂ with 1-ethyl-2-phenylaziridine to form thiazolidine-2-thione under optimal conditions (30 °C, 0.1 MPa, 6 h) with only 81% yield (Fig. S27 and Table S4†), which was inferior to that catalyzed by compound 1, revealing the significance of anion framework. Fourthly, five supplement catalytic cycle experiments of neutral framework Co(II/III) were conducted, presenting a low average catalytic yield of 79% (81%, 78%, 80%, 78% and 78%) (Fig. S28†). In comparison, compound 1 still maintains about 99% catalytic yield after several cycle experiments, which is much higher than that of the Co(II/III) framework (Fig. 2). Fifth, the leaching experiments for the coupling reaction between CS₂ and aziridine with compound 1 or Co(II/III) as catalyst also have been performed. As shown in Fig. S29,† compound 1 demonstrates higher catalytic efficiency. Notably, the catalytic reaction rate decreased significantly when the compound 1 was filtered out after 3 hours of reaction, and the catalyst was replaced by Co(II/III) framework material. Additionally, the catalytic yields of this CS₂ conversion reaction at different points in time with Co(II/III) framework as catalyst have been gained in Fig. S29.† Compared with compound 1, the catalytic efficiency of Co(II/III) framework is lower at the same reaction time. Therefore, the catalytic cycle reaction test and the leaching experiments supplied above all sufficiently reveal that the free Me₂NH₂⁺ in the pores will enhance the catalytic activity of compound 1.

Moreover, we further assessed and probed the effect of the interaction between the framework with and without Me₂NH₂⁺ in the pore channel and CS₂ on the catalytic reaction activity and mechanism by DFT calculations (Fig. S30†). According to the calculations, the cycloaddition of CS₂ with aziridines occurs over a two-step reaction in the absence of the Me₂NH₂⁺. The catalytic path and Gibbs free energy profile of the reaction are listed in Fig. 4a and Fig. 5. Initially, the porous framework of compound 1 effectively enriched aziridine and CS₂ molecules with a slight exotherm of 5.9 kcal mol⁻¹. Subsequently, one of the S-atom in the trapped CS₂ nucleophilically attacked the substituted carbon of the aziridine, and a C–S bond was formed while the C–N bond was broken. The energy barrier of the transition state TS₁ was 18.8 kcal mol⁻¹ in the step above. Finally, the thiazolidine-2-thione product was generated by an intramolecular ring-closing reaction, corresponding to an energy barrier of 12.3 kcal mol⁻¹ for the transition state TS₂. In contrast, the initial exotherm of this system reduced to 1.7 kcal mol⁻¹, and the reaction energy barriers of the corresponding transition states TS₁ as well as TS₂ reduced to 15.8 and 10.3 kcal mol⁻¹ respectively in the presence of dimethylamine ions in the reaction. Additionally, the formation of N–H⋯N H-bonding and N–H⋯S H-bonding interactions between Me₂NH₂⁺ and aziridine intermediates can effectively stabilize the intermediates and lower the energy barrier of the reaction, as shown in Fig. 4b.

Fig. 4 DFT optimization of various geometries in the catalytic process is performed through two paths, respectively. (a) The reaction system in the absence of Me₂NH₂⁺ ions; (b) the reaction system in the presence of Me₂NH₂⁺ ions. Color codes for atoms: bluish violet, Co; yellow, S; gray, C; red, O; blue, N; white, H.

Fig. 5 Gibbs free energies of 1-catalyzed CS₂-aziridine cycloaddition reactions under two catalytic systems.

Based on the above studies and previous related reports,^18,59 the reaction mechanism in the case of 1/TBAB catalysis has been speculated (Fig. 6). First, the unsaturated Lewis acid active site in the 3D cage framework of compound 1 can enrich the aziridine substrate and activates the N atom therein. Secondly, Br⁻ from TBAB makes a nucleophilic attack on the C atom and a ring-opening reaction occurs. Subsequently, CS₂ molecules adsorbed and activated by Me₂NH₂⁺ bind to the ring-opened substrate. Finally, the intramolecular loop closure is completed by the attack of the S⁻ ion on the C atom, accompanied by the regeneration of catalyst 1 and TBAB.

Fig. 6 A possible mechanism for catalyzed reaction. TBAB: tetrabutylammonium bromide; [Co₃]: {(Me₂NH₂)₂[Co₃(μ₃-O)(XN)(BDC)₃]·4DMF·5MeOH}_n.

Conclusions

In summary, we synthesized and characterized a porous 3D cobalt-organic framework assembled by [Co₁₅] and [Co₁₈] nanocages. Benefitting from abundant channels and free Me₂NH₂⁺ ions, compound 1 can catalyze the conversion reaction of CS₂ with aziridines efficiently at near room temperature (30 °C) and 1 bar with a yield of up to 99%, presenting excellent catalytic capacity and recursivity. Worthily, the reaction TOF reached a record of 157 h⁻¹, which is well ahead of other catalysts reported about this catalytic reaction. Importantly, compound 1 can also directly catalyze the conversion of CS₂ derived from simulated industrial waste streams and aziridines with a high yield of 94%. The reaction mechanism of the CS₂ coupling process has been proposed by both experimental analyses and DFT calculations, revealing the synergistic activation effects of aziridine and CS₂ by [Co₃]-cluster and Me₂NH₂⁺. This work will provide valuable insights into methods and ideas for the design of efficient MOF-based catalysts for CS₂ green treatment and promising practical applications.

Data availability

All data needed to evaluate this work are available in the main manuscript and/or in the ESI.†

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

This work was supported by NSFC (22201148 and 21761023), the Program for Young Talents of Science and Technology in Universities of Inner Mongolia Autonomous Region (NJYT23038).

References

1. M. Luecke, L. Giarrana, A. Kostenko, T. Gensch, S. Yao and M. Driess, Angew. Chem., Int. Ed., 2022, 61, e202110398.
2. J. Gao, M. Li, H. Zhao, Y. Wu, Q. Gao, X. Wu, Y. Zhang and Y. Zhang, J. Environ. Chem. Eng., 2023, 11, 110815.
3. J. Song, D. Wang, M. Zhou, X. You, Q. Tan, W. Liu, L. Yu, B. Wang, W. Chen and X. Zhang, J. Hazard. Mater., 2023, 454, 131464.
4. A. W. DeMartino, D. F. Zigler, J. M. Fukuto and P. C. Ford, Chem. Soc. Rev., 2017, 46, 21–39.
5. Y. Ishida, S. Hasegawa and H. Kawaguchi, Angew. Chem., Int. Ed., 2023, 62, e202304700.
6. X. Zhao, L. Wang, G. Zhou, S. Feng and L. Li, Polym. Chem., 2023, 14, 4898–4905.
7. J. Stephan, M. R. Stühler, S. M. Rupf, S. Neale and A. J. Plajer, Cell Rep. Phys. Sci., 2023, 4, 101510.
8. D. B. Schwarz, A. Patil, S. Singla, A. Dhinojwala and J. M. Eagan, Front. Chem., 2023, 11, 1287528.
9. C. Fornacon-Wood, B. R. Manjunatha, M. R. Stühler, C. Gallizioli, C. Müller, P. Pröhm and A. J. Plajer, Nat. Commun., 2023, 14, 4525.
10. D. Patra and A. Saha, Org. Chem. Front., 2023, 10, 1686–1693.
11. Q. Wang, X. J. Meng, H. T. Tang, Y. M. Pan, W. G. Duan and M. X. He, Green Chem., 2023, 25, 2572–2576.
12. M. Gupta, N. Chatterjee, D. De, R. Saha, P. K. Chattaraj, C. L. Oliver and P. K. Bharadwaj, Inorg. Chem., 2020, 59, 1810–1822.
13. T. L. Wang, X. J. Liu, C. D. Huo, X. C. Wang and Z. J. Quan, Chem. Commun., 2018, 54, 499–502.
14. A. Sudo, Y. Morioka, E. Koizumi, F. Sanda and T. Endo, Tetrahedron Lett., 2003, 44, 7889–7891.
15. A. Khalaj and M. Khalaj, J. Chem. Res., 2016, 40, 445–448.
16. N. Kumar, A. Sharma, U. Kumar and S. K. Pandey, J. Org. Chem., 2023, 88, 6120–6125.
17. A. Ziyaei Halimehjani and Y. Lotfi Nosood, Org. Lett., 2017, 19, 6748–6751.
18. Y. Shi, B. Tang, X. L. Jiang, Y. E. Jiao, H. Xu and B. Zhao, J. Mater. Chem. A, 2022, 10, 4889–4894.
19. Y. Xie, C. Lu, B. Zhao, Q. Wang and Y. Yao, J. Org. Chem., 2019, 84, 1951–1958.
20. A. Liu, L. He, S. Peng, Z. Pan, J. Wang and J. Gao, Sci. China: Chem., 2010, 53, 1578–1585.
21. J. Y. Wu, Z. B. Luo, L. X. Dai and X. L. Hou, J. Org. Chem., 2008, 73, 9137–9139.
22. H. Jiang, X. Zhao, W. Zhang, Y. Liu, H. Li and Y. Cui, Angew. Chem., Int. Ed., 2023, 62, e202214748.
23. X. Han, W. Zhang, Z. Chen, Y. Liu and Y. Cui, Mater. Horiz., 2023, 10, 5337–5342.
24. S. Hou, J. Dong and B. Zhao, Adv. Mater., 2020, 32, 1806163.
25. X. L. Jiang, F. Y. Ren, Y. Shi, Y. Xie, S. L. Hou and B. Zhao, CCS Chem., 2024, 1–13.
26. Z. L. Liang, Z. H. Zhang, Y. E. Jiao, H. Xu, H. S. Hu and B. Zhao, J. Am. Chem. Soc., 2024, 146, 10776–10784.
27. S. Zhang, L. Lu, J. Jiang, N. Liu, B. Zhao, M. Xu, P. Cheng and W. Shi, Adv. Mater., 2024, 2403464.
28. C. Cao, S. Xia, Z. Song, H. Xu, Y. Shi, L. He, P. Cheng and B. Zhao, Angew. Chem., Int. Ed., 2020, 59, 8586–8593.
29. S. Hou, J. Dong, X. Zhao, X. Li, F. Ren, J. Zhao and B. Zhao, Angew. Chem., Int. Ed., 2023, e202305213.
30. W. Xuan, C. Ye, M. Zhang, Z. Chen and Y. Cui, Chem. Sci., 2013, 4, 3154.
31. J. Liu, T. A. Goetjen, Q. Wang, J. G. Knapp, M. C. Wasson, Y. Yang, Z. H. Syed, M. Delferro, J. M. Notestein, O. K. Farha and J. T. Hupp, Chem. Soc. Rev., 2022, 51, 1045–1097.
32. W. Li, X. Liu, X. Yu, B. Zhang, C. Ji, Z. Shi, L. Zhang and Y. Liu, Inorg. Chem., 2023, 62, 18248–18256.
33. M. Zhao, S. Huang, Q. Fu, W. Li, R. Guo, Q. Yao, F. Wang, P. Cui, C. Tung and D. Sun, Angew. Chem., Int. Ed., 2020, 59, 20031–20036.
34. L. Zeng, Y. Cao, Z. Li, Y. Dai, Y. Wang, B. An, J. Zhang, H. Li, Y. Zhou, W. Lin and C. Wang, ACS Catal., 2021, 11, 11696–11705.
35. B. An, Z. Li, Y. Song, J. Zhang, L. Zeng, C. Wang and W. Lin, Nat. Catal., 2019, 2, 709–717.
36. X. Si, Q. Yao, X. Pan, X. Zhang, C. Zhang, Z. Li, W. Duan, J. Hou and X. Huang, Inorg. Chem., 2023, 62, 15006–15014.
37. Z. Jiao, X. Zhao, J. Zhao, Y. Xie, S. Hou and B. Zhao, Acta Phys. -Chim. Sin., 2023, 39, 2301018.
38. X. Kang, Z. Jiao, X. Shi, Y. Tian and Z. Liu, J. Mater. Chem. C, 2022, 10, 16078–16087.
39. Y. Shi, Q. Chang, T. Song, M. Yin, L. Zhang, X. Cao, Z. Wei and X. Cao, Chem. Eng. J., 2024, 479, 147394.
40. X. Tian, Z. Jiang, S. Hou, H. Hu, J. Li and B. Zhao, Angew. Chem., Int. Ed., 2023, e202301764.
41. H. Xu, X. Liu, C. Cao, B. Zhao, P. Cheng and L. He, Adv. Sci., 2016, 3, 1600048.
42. G. M. Sheldrick, Acta Crystallogr., 2015, 71, 3–8.
43. A. L. Spek, Acta Crystallogr., 2015, 71, 9–18.
44. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson and H. Nakatsuji, Gaussian 09 (revision D.01), I. Gaussian, Wallingford, CT, 2013.
45. R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys., 1980, 72, 650–654.
46. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
47. S. Xu, T. He, J. Li, Z. Huang and C. Hu, Appl. Catal., B, 2021, 292, 120145.
48. P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283.
49. C. Gonzalez and H. B. Schlegel, J. Chem. Phys., 1989, 90, 2154–2161.
50. Y. Sun, W. Xu, F. Lang, H. Wang, F. Pan and H. Hou, Small, 2024, 20, 2305879.
51. X. Kang, Z. Wang, X. Shi, X. Jiang, Z. Liu and B. Zhao, Small, 2024, 2311511.
52. X. F. Jiang, H. Huang, Y. F. Chai, T. L. Lohr, S. Y. Yu, W. Lai, Y.-J. Pan, M. Delferro and T. J. Marks, Nat. Chem., 2017, 9, 188–193.
53. Y. Zhou, C. Chen, R. Krishna, Z. Ji, D. Yuan and M. Wu, Angew. Chem., Int. Ed., 2023, 62, e202305041.
54. W. Wang, G. Wang, B. Zhang, X. Li, L. Hou, Q. Yang and B. Liu, Small, 2023, 19, 2302975.
55. A. Ray, B. Martín-García, A. Moliterni, N. Casati, K. M. Boopathi, D. Spirito, L. Goldoni, M. Prato, C. Giacobbe, C. Giannini, F. Di Stasio, R. Krahne, L. Manna and A. L. Abdelhady, Adv. Mater., 2022, 34, 2106160.
56. H. Meng, K. Mao, F. Cai, K. Zhang, S. Yuan, T. Li, F. Cao, Z. Su, Z. Zhu, X. Feng, W. Peng, J. Xu, Y. Gao, W. Chen, C. Xiao, X. Wu, M. D. McGehee and J. Xu, Nat. Energy, 2024, 1–12.
57. W. Hui, L. Chao, H. Lu, F. Xia, Q. Wei, Z. Su, T. Niu, L. Tao, B. Du, D. Li, Y. Wang, H. Dong, S. Zuo, B. Li, W. Shi, X. Ran, P. Li, H. Zhang, Z. Wu, C. Ran, L. Song, G. Xing, X. Gao, J. Zhang, Y. Xia, Y. Chen and W. Huang, Science, 2021, 371, 1359–1364.
58. T. Zhang, Y. Q. Hu, T. Han, Y. Q. Zhai and Y. Z. Zheng, ACS Appl. Mater. Interfaces, 2018, 10, 15786–15792.
59. C. H. Zhang, Z. L. Wu, R. X. Bai, T. D. Hu and B. Zhao, ACS Appl. Mater. Interfaces, 2023, 15, 1879–1890.

---

Footnotes

† Electronic supplementary information (ESI) available. CCDC 2337117. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4gc04541f

‡ Those authors contributed equally.

---