Abstract
Ethylene (C2H4) purification from multi-component mixtures by physical adsorption is a great challenge in the chemical industry. Herein, we report a GeF62- anion embedded MOF (ZNU-6) with customized pore structure and pore chemistry for benchmark one-step C2H4 recovery from C2H2 and CO2. ZNU-6 exhibits significantly high C2H2 (1.53 mmol/g) and CO2 (1.46 mmol/g) capacity at 0.01 bar. Record high C2H4 productivity is achieved from C2H2/CO2/C2H4 mixtures in a single adsorption process under various conditions. The separation performance is retained over multiple cycles and under humid conditions. The potential gas binding sites are investigated by density functional theory (DFT) calculations, which suggest that C2H2 and CO2 are preferably adsorbed in the interlaced narrow channel with high aff0inity. In-situ single crystal structures with the dose of C2H2, CO2 or C2H4 further reveal the realistic host-guest interactions. Notably, rare C2H2 clusters are formed in the narrow channel while two distinct CO2 adsorption locations are observed in the narrow channel and the large cavity with a ratio of 1:2, which accurately account for the distinct adsorption heat curves.
Similar content being viewed by others
Introduction
Ethylene (C2H4) is the foremost olefin as well as the highest volume product in the petrochemical industry, with an annual production capacity exceeding 214 million tons in 20211. The manufacture of C2H4 and C3H6 accounts for 0.3% of global energy2. Current C2H4 production mainly relies on stream cracking of hydrocarbons3,4,5,6. Alternatively, oxidative coupling of methane (CH4) has emerged as a promising technique to produce C2H4, among which acetylene (C2H2) and carbon dioxide (CO2) are generated as byproducts and need to be deeply removed to produce polymer grade (>99.996%) C2H47. Presently, multi-step purification process is adopted for purification of C2H4 from C2H4/C2H2/CO2 mixtures in industry8. C2H2 is removed by catalytic hydrogenation using expensive noble-metal catalysts or solvent extraction, which is either energy intensive or associated with pollution9,10. CO2 is removed by chemical adsorption using caustic soda, which causes huge waste of costly solvents11.
Physical adsorption offers potential to significantly reduce the energy footprint of separation processes12,13,14,15,16,17,18,19,20,21. Nonetheless, C2H4 purification from ternary C2H4/C2H2/CO2 mixtures remains an unmet challenge due to the similarity in molecular size and polarity (Supplementary Table 2), although separation of C2H2/C2H422,23,24,25,26 or C2H2/CO227,28,29,30,31,32 binary mixtures has been realized by a plethora of porous materials. Besides, single-step purification of C2H4 from ternary C2H2/C2H4/C2H633,34 or quaternary C2H2/C2H4/C2H6/CO235 mixtures has also been realized by several porous materials. To date, less than ten materials have been demonstrated to separate C2H4 from C2H4/C2H2/CO2, including activated carbons, zeolites, covalent organic frameworks and metal organic framework (MOFs)36,37,38,39. TIFSIX-17-Ni36, NTU-6537, and NTU-6738 are so far the three optimal materials. TIFSIX-17-Ni36 exhibits high C2H2/C2H4 and CO2/C2H4 selectivity due to the negligible uptake of C2H4 under ambient condition. However, the capacity of C2H2 (3.30 mmol/g) and CO2 (2.20 mmol/g) is relatively low due to the over-contracted channel. NTU-6537 can selectively capture C2H2 and CO2 by tuning the gate opening. However, the applied temperature must be at 263 K because lower temperatures lead to the adsorption of all the gases while higher temperatures cause the exclusion of CO2. NTU-6738 displays similar C2H2 (3.29 mmol/g) and CO2 (2.04 mmol/g) capacity, but the C2H2/C2H4 and CO2/C2H4 selectivity is greatly reduced as the C2H4 capacity (1.41 mmol/g) is relatively high. Additionally, the separation performance is deteriorated under humid conditions. Therefore, there is still a lack of ideal and stable materials to realize the simultaneous removal of C2H2 and CO2 in C2H2/CO2/C2H4 mixtures.
In this work, we reported a GeF62− anion embedded MOF ZNU-6 (ZNU = Zhejiang Normal University) with large cages (~8.5 Å diameter) connected by narrow interlaced channels (~4 Å diameter) for benchmark one-step C2H4 recovery from C2H2 and CO2. ZNU-6 is constructed by CuGeF6 and tri(pyridin-4-yl)amine (TPA) and exhibits excellent chemical stability. Static gas adsorption isotherms showed that ZNU-6 takes up 1.53/8.06 mmol/g of C2H2 and 1.46/4.76 mmol/g of CO2 at 0.01 and 1.0 bar (298 K), respectively. The calculated IAST selectivities for C2H2/C2H4 (1/99) and CO2/C2H4 (1/99) are 43.8–14.3 and 52.6–7.8 (0.0001–1.0 bar), respectively. The calculated Qst values at near-zero loading for C2H2 and CO2 are 37.2 and 37.1 kJ/mol, indicative of its facility for material regeneration but much higher than that of C2H4 (29.0 kJ/mol). Modeling study indicates that there are two potential binding sites for C2H2, C2H4, and CO2. One is in the interlaced channel and the other locates in the large cage. Moreover, all gas molecules prefer to be adsorbed in the interlaced channel with higher affinity. The realistic binding sites and host–guest interactions under normal conditions (298 K and 1.0 bar) were further demonstrated by in-situ single crystal structures with the saturated dose of gases. Notably, rare C2H2 clusters formed by π···π packing and C-H···C≡C interactions are observed in the interlaced channel with a small proportion of C2H2 molecules adsorbed in the large cage additionally. In sharp contrast, only 1/3 of CO2 molecules are located in the narrow channel while 2/3 of CO2 molecules are accommodated in the large cavity. This distinct gas distribution is highly consistent with the difference of adsorption heat curves. The practical C2H4 purification performance is further demonstrated by dynamic breakthroughs and record high C2H4 productivity is achieved from ternary C2H2/CO2/C2H4 mixtures in a single adsorption process under various conditions. The separation performance is retained over multiple cycles and under humid conditions.
Results
Violet single crystals of ZNU-6 (Supplementary Fig. 1) were produced by layering a MeOH solution of TPA onto an aqueous solution of CuGeF6 (Fig. 1a). X-ray crystal analysis revealed that ZNU-6 [Cu6(GeF6)6(TPA)8] crystallizes in a three-dimensional (3D) framework in the cubic Pm-3n space group. Every unit cell consists of six Cu2+ ions, six GeF62- anions, and eight tridentate TPA ligands (Supplementary Table 1). The combination of Cu2+ and TPA produces a cationic pto network first (Fig. 1b), which determines the main pore size. The network is further embedded by GeF62− pillar to give a ith-d topology framework with optimal pore chemistry (Fig. 1c). The frameworks are composed of large icosahedral cage-like pores (~8.5 Å) and interlaced narrow channels (~4 Å) (Fig. 1d–f). Each large cage is surrounded by 12 channels and every interlaced channel connects 4 cages. The adjacent two cages and two channels share the same GeF62- anions at the edge. Both large pores and interlaced channels are abundant of Lewis basic F functional sites on the surface for gas binding. Such interconnected large cages and narrow channels are distinct from previous straight 1D channels of anion pillared MOFs (e.g., SIFSIX-1-Cu, SIFSIX-3-Ni). Besides, the narrow channel size may provide kinetic selectivity for C2H2 (3.3 Å) and CO2 (3.3 Å) given their small molecular size compared to C2H4 (4.2 Å). Thus, ZNU-6 with abundant functional GeF62− binding sites, high porosity for C2H2 and CO2 accommodation and narrow channel for kinetic preference features the promising characteristics for efficient purification of C2H4 from ternary C2H2/CO2/C2H4 mixture.
The intrinsic porosity of ZNU-6 was investigated by N2 adsorption at 77 K. As shown in Fig. 2a, ZNU-6 exhibited a type I adsorption isotherm. The Brunauer–Emmett–Teller surface area and pore volume were calculated to be 1330.3 m2/g and 0.554 cm3/g (Supplementary Fig. 10). The calculated pore size ranges from 8.22 to 10.76 Å with the summit in 9.0 Å, highly close to the pore aperture of ~8.5 Å evaluated from the single crystal structure (Fig. 2a). Then, single-component adsorption isotherms of C2H2, CO2, and C2H4 were collected at 298 K (Fig. 2b). At 1.0 bar, the C2H2 and CO2 uptakes are 8.06 and 4.76 mmol/g, higher than those of most APMOFs (Fig. 2c). The capacities are equal to 4.68 and 2.77 gas molecules per GeF62− anion. Such high C2H2/anion and CO2/anion uptakes have never been realized in anion pillared MOFs (Supplementary Table S7)36,37,38,40,41,42,43,44. Particularly, C2H2/anion and CO2/anion uptakes in benchmark TIFSIX-17-Ni36, SIFSIX-17-Ni36 and NTU-6738 are only 1.36/0.91, 1.29/0.9 and 2.06/1.28, respectively (Supplementary Fig. 25). So far, isomorphic SIFSIX-Cu-TPA40 displays the ever highest C2H2/anion (4.44) uptake while SIFSIX-1-Cu41 displays the ever highest CO2/anion (2.72) uptake. It is worth mentioning that these records have been marginally surpassed by ZNU-6’s (Supplementary Fig. 25). Notably, the uptakes of C2H2 and CO2 on ZNU-6 at 0.01 bar are as high as 1.53 and 1.46 mmol/g, superior to those of all the porous materials in the context of ternary C2H2/CO2/C2H4 separation, such as TIFSIX-17-Ni (1.38/0.32 mmol/g)36, SIFSIX-17-Ni (0.91/0.20 mmol/g)36, NTU-67 (0.47/0.65 mmol/g)38, and TpPa-NO2 (0.17/0.03 mmol/g)39. At 0.1 bar, the capacities of C2H2 and CO2 reach up to 4.64 and 2.21 mmol/g (Fig. 2b), even higher than the uptakes of many porous materials at 1 bar and 298 K, for example, TIFSIX-17-Ni (3.30/2.20 mmol/g)36. In the meantime, the C2H4 uptakes on ZNU-6 at 0.01 and 0.1 bar are only 0.15 and 1.07 mmol/g, much lower than those of C2H2 and CO2 under the same conditions. The C2H2, CO2, and C2H4 adsorption isotherms were further collected at 278 and 308 K (Fig. 2d). The adsorption capacities of C2H2 and CO2 at 1 bar increase to 8.74 and 6.26 mmol/g at 278 K. As selectivity is also an important parameter to assess the separation performance, we further calculated the C2H2/C2H4 and CO2/C2H4 selectivities on ZNU-6 using ideal adsorbed solution theory (IAST) after fitting isotherms into dual site Langmuir or single site Langmuir equation with excellent accuracy. The IAST selectivity for 1/99 C2H2/C2H4 is 43.8–14.3 (Fig. 2e), higher than those of NTU-67 (8.1)38 and TpPa-NO2 (5.9)39. The IAST selectivities for 1/99 CO2/C2H4 mixture is also as high as 52.6-7.8 (Fig. 2e). Besides, both C2H2/C2H4 and CO2/C2H4 selectivity on ZNU-6 is improved with the pressure decrease or the increase of C2H4 ratios (from 90% to 99%) in the binary mixtures (Supplementary Figs. 13, 14), indicating ZNU-6 is favored for trace C2H2 and CO2 capture from bulky C2H4 mixtures. Apart from the IAST selectivity, the Henry coefficients were also calculated to evaluate the Henry’s selectivity of ZNU-6 (Supplementary Figs. 15–17), the Henry’s selectivity for C2H2/C2H4 and CO2/C2H4 is 8.2 and 7.8, respectively, superior to those of NTU-67 (2.4/4.2)38 and TpPa-NO2 (4.0/1.8)39 (Supplementary Tables 4, 5). We further calculated the isosteric enthalpy of adsorption (Qst) for ZNU-6 by using the Clausius-Clapeyron equation. Qst values at near-zero loading for C2H2, CO2, and C2H4 are 37.2, 37.1, and 29.0 kJ/mol (Fig. 2f), respectively, indicative of the preferred affinity of C2H2 and CO2 over C2H4. Notably, the Qst values for C2H2 and CO2 on ZNU-6 are only modestly high and slightly lower than those of many top-performing materials in the context of C2H4 purification, such as SIFSIX-17-Ni (44.2/40.2 kJ/mol)36, TIFSIX-17-Ni (48.3/37.8 kJ/mol)36, and NTU-67 (44.1/41.5 kJ/mol)38. Such moderate Qst endows facile regeneration of ZNU-6 under mild conditions.
To gain more insights into the gas adsorption behavior, density functional theory (DFT)-based calculations (see Method section) were applied to identify the adsorption configuration and binding energies of C2H2, CO2, and C2H4. For all gases, two different binding sites were observed. Site I is in the interlaced channel and Site II is in the large cavity (Fig. 3). For C2H2 in Site I, the two hydrogen atoms interact strongly with three F atoms with the distances of 1.80, 1.93, and 2.37 Å. The calculated binding energy is 57.1 kJ/mol (Fig. 3a). As for C2H2 adsorbed in Site II, only one hydrogen atom can interact with the adjacent F atoms with the distance of 2.23 and 2.24 Å, and the corresponding binding energy decreases to 37.9 kJ/mol (Fig. 3b), indicating that C2H2 is preferentially adsorbed in the narrow channel. The same results are also observed for CO2 and C2H4, the binding energies in the channel are much higher than those in the large cage. In Site I, CO2 is trapped by two strong and two weak electrostatic F···C=O interactions in the distance of 2.89, 3.02, 3.60, and 3.90 Å, the binding energy is 52.8 kJ/mol (Fig. 3c); C2H4 is adsorbed via two F···H interactions (2.29 and 2.37 Å) with the binding energy of 43.3 kJ/mol (Fig. 3e). In Site II, the binding energy of CO2 drops to 40.7 kJ/mol with the number of electrostatic F···C=O interactions (2.74 and 2.87 Å) decreasing to two (Fig. 3d); the binding energy of C2H4 reduces to 25.3 kJ/mol with the length of F···H extending to 2.55 and 2.32 Å (Fig. 3f). In addition, it is notable that either in Site I or II, the binding energy of C2H2 or CO2 is superior to that of C2H4, confirming that the adsorption of C2H2 or CO2 in ZNU-6 is more preferable than that of C2H4.
Although DFT calculations have identified two different binding sites for each gas, it is still difficult to understand the distinct adsorption heat curves. Therefore, we further studied the in-situ structures of ZNU-6 with gas loading (Fig. 4). We found that averagely 25.78 C2H2, 18 CO2, or 13.07 C2H4 molecules can be adsorbed per unit cell of ZNU-6 (Supplementary Table 1), corresponding to 4.3 C2H2, 3.0 CO2, and 2.2 C2H4 molecules for each GeF62- anion, which are close to the saturated values from gas adsorption isotherms (4.63 C2H2, 2.77 CO2, and 2.75 C2H4). Both of C2H2 and CO2 have two binding sites, i.e., Site I in the interlaced channel and Site II in the large cage. Notably, the amount of C2H2 molecules distributed to the two locations is 3.8 and 0.5 per GeF62- anion while that for CO2 is 1 and 2 per GeF62− anion (Fig. 4a, b). Such different gas distribution can precisely account for the C2H2 Qst curve with a modest decrease and the CO2 Qst curve with a sharp decrease along the gas loading. Specifically, C2H2 molecules adsorbed in Site I bind to F atoms on the surface of the channels via multiple cooperative hydrogen bonds (C-H···F = 1.97–2.55 Å), and the others in Site II interact F atoms via single H···F hydrogen bond with the distance of 2.51 Å (Fig. 4a and Table 1). Besides, the C2H2 molecules in Site I aggregate to form a stacked gas cluster by π···π packing and C-H···C≡C interactions, which has rarely been observed previously. Regarding CO2, it is trapped by F···C=O electrostatic interaction in Site I and II (Fig. 4b). The only difference is that the C···F distance is 2.64 Å in Site I and 2.80 Å in Site II (Table 1). From the single crystal structure, two different CO2 molecules that are very close and opposite to each other in the narrow channel (site I) are observed. However, these two CO2 molecules cannot exist in the same narrow channel at the same time and thus both CO2 molecules display the occupancy of 50%. In Site II, the C atom of CO2 is ordered while the O atoms are disordered to two perpendicular positions with the occupancy of 50% for each configuration. Besides, the linear CO2 molecules are slightly bent due to the strong attraction from GeF62− anion. The bent angle of 157.5° (in Site I) and 170.8° (in Site II) are consistent with the interaction strength. In term of C2H4, only one site in the narrow channel is found. The C atoms of C2H4 molecule are ordered while the H atoms are disordered. The distances of C-H···F interactions between C2H4 and framework are 2.31–2.64 Å (Fig. 4c and Table 1). Considering the slight lower C2H4/GeF62− ratio observed in the single crystal structure, there should be some C2H4 molecules adsorbed in the large cage. However, due to the probable disorder of C2H4 molecules over the whole cage, the C2H4 molecules in Site II were not solved. Nonetheless, this uniform adsorption configuration is consistent with the flat Qst curve for C2H4.
Apart from the C2H2, CO2, or C2H4 molecules, some water molecules were also identified in the framework (Supplementary Fig. 4). As there is still a lot of space in the large cavity after saturated adsorption of C2H2, CO2, or C2H4 gases at 100 kPa, the water adsorption behavior probably occurred during the single crystal measurement, which is exposed to air. Interestingly, these water molecules are distant from GeF62-, indicating that these H2O molecules do not occupy the binding sites for the targeted gases. Instead, some unique interactions are observed between the gas molecules and water molecules, e.g., O-H···C=O hydrogen bonds between CO2 and H2O. Notably, our resolved single crystal structures show completely different C2H2 and CO2 adsorption configurations from those of isomorphic SIFSIX-Cu-TPA for C2H2/CO2 separation in Wu’s work39.
Motivated by the high adsorption capacity and selectivity in single-component adsorption as well as the in-situ single crystal structure analysis, breakthrough experiments were conducted for C2H2/C2H4, CO2/C2H4, and C2H2/CO2/C2H4 mixtures. The results showed that highly efficient separations can be accomplished by ZNU-6 for all the gas mixtures under various conditions. For 1/99 C2H2/C2H4 mixtures, C2H4 is eluted at 12 mins while C2H2 is detected until 192 min. For 10/90 CO2/C2H4 mixtures, C2H4 and CO2 are detected at 12 and 43.5 min, respectively (Fig. 5a). For 1/1/98 C2H2/CO2/C2H4 mixtures, C2H2 and CO2 broke out simultaneously and 64.42 mol/kg of polymer grade C2H4 is produced by single adsorption process (Fig. 5b). The productivity is improved to 80.89 mol/kg when decreasing the temperature to 283 K (Supplementary Fig. 46). The CO2 breakthrough time becomes shortened with the increase of CO2 ratio, which is 72 and 52 min for 1/5/94 (Figs. 5c) and 1/9/90 (Fig. 5d) C2H2/CO2/C2H4 mixtures. The polymer grade C2H4 productivity is 21.37 and 13.81 mol/kg, respectively. As most reported C2H4 productivity from C2H2/CO2/C2H4 mixtures are compared under 1/9/90, a comparison plot of the C2H4 productivity and dynamic C2H2 capacity from 1/9/90 C2H2/CO2/C2H4 mixtures is presented in Fig. 5e. ZNU-6 displays the record high C2H4 productivity and second highest C2H2 dynamic capacity. The C2H4 productivity of ZNU-6 is >2.5 folds of the previous benchmark of NTU-67 (5.42 mol/kg)38. C2H4 productivity with the unit of mol/kg/h is also calculated for comparison (Supplementary Table S10). ZNU-6 with the productivity of 15.93 mol/kg/h is the highest reported value.
In view of the importance of the recyclability and stability of porous materials for practical applications, the water and thermal stability of ZNU-6 was investigated. There was no noticeable loss in the CO2 adsorption capacity after six cycles of adsorption/desorption experiments (Supplementary Fig. 26). Long time soaking of ZNU-6 in water or polar organic solvents such as DMSO, DMF and MeCN did not change the porous structure of ZNU-6, as demonstrated by the PXRD patterns as well as the gas adsorption isotherms (Supplementary Fig. 7). Thermogravimetric analysis (TGA) and temperature varied PXRD indicated ZNU-6 is stable below 200 °C (Supplementary Figs. 8, 9). Breakthroughs under humid conditions or over four cycles preserved nearly the identical separation performance (Fig. 5f). Although many water molecules can be adsorbed in ZNU-6, as described in in-situ crystals and water adsorption isotherms (Supplementary Fig. 27), the presence of humid has negligible influence on the separation performance (Fig. 5f). This is probably due to the co-adsorption of water and target gases as well as the fast C2H2/CO2/C2H4 diffusion kinetics (Supplementary Fig. 29–31).
Discussion
In conclusion, we reported a GeF62− anion embedded metal organic framework ZNU-6 with optimal pore structure and pore chemistry for benchmark one-step C2H4 purification by simultaneous removal of C2H2 and CO2. ZNU-6 exhibits remarkably high C2H2 and CO2 capacity under both low and high pressures. The C2H2/anion and CO2/anion uptakes are the highest among all the anion pillared MOFs. 64.42, 21.37, 13.81 mol/kg polymer grade C2H4 can be produced from C2H2/CO2/C2H4 (1/1/98, 1/5/94, 1/9/90) mixtures, all superior to the previous benchmarks. The separation performance is sustained over multiple cycles or under humid conditions. The potential gas binding sites are investigated by DFT calculation, which indicate that C2H2 and CO2 are preferentially adsorbed in the interlaced narrow channel with high affinity. In-situ single crystal structures with the dose of C2H2, CO2 or C2H4 further reveal the realistic host–guest interactions, accounting for the distinct shapes of the adsorption heat curves. In general, our work highlights the significance of regulating pore structure and pore chemistry in porous materials to construct multiple cooperative functionalities for gas separation.
Methods
Synthesis of ZNU-6
To a 5 mL long thin tube was added a 1 mL of aqueous solution with Cu(NO3)2·3H2O (~1.3 mg) and (NH4)2GeF6 (~1.0 mg). 2 mL of MeOH/H2O mixture (v:v = 1:1) was slowly layered above the solution, followed by a 1 mL of MeOH solution of TPA (~1.0 mg). The tube was sealed and left undisturbed at 298 K. After ~1 week, purple single crystals were obtained.
Preparation of gas loaded ZNU-6
The crystalline sample of ZNU-6 was filled into a glass tube and heated at 120 °C under vacuum for 24 h. After the sample cooling down, CO2, C2H2, or C2H4 was introduced into the sample respectively with Builder SSA 7000 (Beijing) instrument until the pressure reach to 1 bar at 298 K and the state is maintained for another hour. Then, the crystals were picked out, covered with the degassed oil, and single crystal X-ray diffraction measurements were then carried out at 298 K as soon as possible.
Single-crystal X-ray diffraction
Single-crystal X-ray diffraction studies were conducted on the BrukerAXS D8 VENTURE diffractometer equipped with a PHOTON-100/CMOS detector (GaKα, λ = 1.34139 Å). Indexing was performed using APEX2. Data integration and reduction were completed using SaintPlus 6.01. Absorption correction was performed by the multi-scan method implemented in SADABS. The space group was determined using XPREP implemented in APEX2.1 The structure was solved with SHELXS-97 (direct methods) and refined on F2 (nonlinear least-squares method) with SHELXL-97 contained in APEX2, WinGX v1.70.01, and OLEX2 v1.1.5 program packages. All non-hydrogen atoms were refined anisotropically. The contribution of disordered solvent molecules was treated as diffuse using the Squeeze routine implemented in Platon.
Powder X-ray diffraction
Powder X-ray diffraction (PXRD) data were collected on the SHIMADZU XRD-6000 diffractometer (Cu Kαλ = 1.540598 Ǻ) with an operating power of 40 kV, 30 mA and a scan speed of 4.0°/min. The range of 2θ was from 5° to 50°.
Thermal gravimetric analysis
Thermal gravimetric analysis was performed on the TGA STA449F5 instrument. Experiments were carried out using a platinum pan under nitrogen atmosphere which conducted by a flow rate of 60 mL/min nitrogen gas. The data were collected at the temperature range of 50 °C to 600 °C with a ramp of 10 °C /min.
The static gas/vapor adsorption equilibrium measurements
The static gas adsorption equilibrium measurements were performed on the Builder SSA 7000 instrument. The water vapor adsorption equilibrium measurements were performed on the BeiShiDe DVS instrument. Before measurements, the sample of ZNU-6 (~100 mg) was evacuated at 25 °C for 2 h firstly, and then at 120 °C for 12 h until the pressure dropped below 7 μmHg. The sorption isotherms were collected at 77 K, 278, 298, and 308 K on activated samples. The experimental temperatures were controlled by liquid nitrogen bath (77 K) and water bath (278, 298, and 308 K), respectively.
Breakthrough experiments
The breakthrough experiments were carried out on a dynamic gas breakthrough equipment. The experiments were conducted using a stainless steel column (4.6 mm inner diameter × 50 mm length). The weight of ZNU-6 powder packed in the columns were 0.5806 g. The column was activated at 75 °C for 2 h under vacuum, and then raised to 120 °C for overnight. The mixed gas of C2H2/C2H4 (1/99, v/v), CO2/C2H4 (10/90, v/v), or C2H2/CO2/C2H4 (1/9/90, 1/5/94, 1/1/98, 5/5/90, v/v/v) was then introduced. C2H2/CO2/C2H4 mixtures are produced by mixing three pure gases or mixing binary mixture with pure gas. Every flowrate was calibrated by self-made soap film flowmeter. Outlet gas from the column was monitored using gas chromatography (GC-9860-5CNJ) with the thermal conductivity detector TCD. After the breakthrough experiment, the sample was regenerated with an Ar flow of 5 mL min−1 under 120 °C for 8 h or under vacuum at 120 °C for 8 h.
Fitting of experimental data on pure component isotherms
The unary isotherms for C2H2 and CO2 measured at three different temperatures 278 K, 298 K, and 308 K in ZNU-6 were fitted with excellent accuracy using the dual-site Langmuir model, where we distinguish two distinct adsorption sites A and B:
In Eq (S1), the Langmuir parameters \({b}_{A},{b}_{B}\) are both temperature dependent
In Eq. (2), \({E}_{A},{E}_{B}\) are the energy parameters associated with sites A, and B, respectively.
The corresponding unary isotherms for C2H4 measured at three different temperatures 278 K, 298 K, and 308 K in ZNU-6 were fitted with excellent accuracy using the single-site Langmuir model.
The unary isotherm fit parameters for C2H2, CO2, and C2H4 are provided in Table S1.
IAST calculations
The adsorption selectivity for separation of binary mixtures of species 1 and 2 is defined by
where q1, q2 are the molar loading (units: mol kg-1) in the adsorbed phase in equilibrium with a gas mixture with partial pressures p1, p2 in the bulk gas.
Calculation of isosteric heat of adsorption (Q st)
The isosteric heat of adsorption, Qst, is defined as
where the derivative in the right member of Eq. (5) is determined at constant adsorbate loading, q. The calculations are based on the Clausius-Clapeyron equation.
Density functional theory calculation
In this work, the DFT-based calculations were carried out using the CP2K package45. The Perdew-Burke-Ernzerhof (PBE) exchange functional46, Gaussian plane wave (PAW) pseudopotentials47 and DZVP basis sets48 for carbon, oxygen, fluorine, nitrogen, germanium and copper atoms, were used to describe the exchange–correlation interactions and electron–ion interaction, respectively. At the same time, the PBE-D3 method49 with Becke–Jonson damping for all atoms and Hubbard U corrections for the open-shell 3d transition metal (Cu) was used for geometry optimizations. The U value of 5.0 eV was used in this study. In all calculations, the net charges of simulation systems were set to zero. The adsorption energy can be obtained from formula below:
where Eadsorbate+substrate and Esubstrate were the total energies of the substrate with and without adsorbate, and Eadsorbate was the energy of the adsorbate.
Data availability
The authors declare that the data supporting the findings of this study are available within the article and Supplementary Information. The X-ray crystallographic data related to ZNU-6 have been deposited at the Cambridge Crystallographic Data Centre (CCDC), under deposition numbers 2192744–2192747, respectively. These data can be obtained free of charge from the CCDC via www.ccdc.cam.ac.uk/data_request/cif. The data that support the findings of this study are available from the corresponding author. Besides, Source data are provided with this paper.
References
Fernández, L. Global production capacity of ethylene 2018–2021, https://www.statista.com/statistics/1067372/global-ethylene-production-capacity/ (2022).
Sholl, D. S. & Liverly, R. P. Seven chemical separations to change the world. Nature 532, 435–437 (2016).
Wang, Q. et al. One-step removal of alkynes and propadiene from cracking gases using a multi-functional molecular separator. Nat. Commun. 13, 2955 (2022).
Suo, X. et al. Synthesis of ionic ultramicroporous polymers for selective separation of acetylene from ethylene. Adv. Mater. 32, 1907601 (2020).
Li, J. et al. Metal-organic framework containing planar metal-binding sites: efficiently and cost-effectively enhancing the kinetic separation of C2H2/C2H4. J. Am. Chem. Soc. 141, 3807–3811 (2019).
Sahoo, R. et al. C2s/C1 hydrocarbon separation: the major step towards natural gas purification by metal-organic frameworks (MOFs). Coord. Chem. Rev. 442, 213998 (2021).
Farrell, B. L. et al. A viewpoint on direct methane conversion to ethane and ethylene using oxidative coupling on solid catalysts. ACS Catal. 6, 4340–4346 (2016).
Chen, K.-J. et al. Synergistic sorbent separation for one-step ethylene purification from a four-component mixture. Science 366, 241–246 (2019).
Zhang, R. et al. The cost-effective Cu-based catalysts for the efficient removal of acetylene from ethylene: the effects of Cu valence state, surface structure and surface alloying on the selectivity and activity. Chem. Eng. J. 351, 732–746 (2018).
Huang, Y. et al. Separation of light hydrocarbons with ionic liquids: a review. Chin. J. Chem. Eng. 27, 1374–1382 (2019).
Ren, T. et al. Olefins from conventional and heavy feedstocks: energy use in steam cracking and alternative processes. Energy 31, 425–451 (2006).
Chai, Y. et al. Control of zeolite pore interior for chemoselective alkyne/olefin separations. Science 368, 1002–1006 (2020).
Li, H. et al. Porous metal-organic frameworks for gas storage and separation: status and challenges. EnergyChem 1, 100006 (2019).
Adil, K. et al. Gas/vapour separation using ultra-microporous metal-organic frameworks: insights into the structure/separation relationship. Chem. Soc. Rev. 46, 3402–3430 (2017).
Yang, L. et al. Energy-efficient separation alternatives: metal-organic frameworks and membranes for hydrocarbon separation. Chem. Soc. Rev. 49, 5359–5406 (2020).
Wang, H. & Li, J. Microporous metal-organic frameworks for adsorptive separation of C5-C6 alkane isomers. Acc. Chem. Res 52, 1968–1978 (2019).
Barnett, B. R. et al. Recent progress towards light hydrocarbon separations using metal-organic frameworks. Trends Chem. 1, 159–171 (2019).
Zhao, X. et al. Metal-organic frameworks for separation. Adv. Mater. 30, 1705189 (2018).
Zeng, H. et al. Orthogonal-array dynamic molecular sieving of propylene/propane mixtures. Nature 595, 542–548 (2021).
Zhang, P. et al. Ultramicroporous material based parallel and extended paraffin nano-trap for benchmark olefin purification. Nat. Commun. 13, 4928 (2022).
Jiang, Y. et al. Comprehensive pore tuning in an ultrastable fluorinated anion cross-linked cage-like MOF for simultaneous benchmark propyne recovery and propylene purification. Angew. Chem. Int. Ed. 61, e202200947 (2022).
Peng, Y.-L. et al. Robust ultramicroporous metal-organic frameworks with benchmark affinity for acetylene. Angew. Chem. Int. Ed. 57, 10971–10975 (2018).
Zhang, Y. et al. Rational design of microporous MOFs with anionic boron cluster functionality and cooperative dihydrogen binding sites for highly selective capture of acetylene. Angew. Chem. Int. Ed. 59, 17664–17669 (2020).
Lin, R.-B. et al. Optimized separation of acetylene from carbon dioxide and ethylene in a microporous material. J. Am. Chem. Soc. 139, 8022–8028 (2017).
Zhang, Z.-Q. et al. Hexafluorogermanate (GeFSIX) anion-functionalized hybrid ultramicroporous materials for efficiently trapping acetylene from ethylene. Ind. Eng. Chem. Res. 57, 7266–7274 (2018).
Ke, T. et al. Molecular sieving of C2-C3 alkene from alkyne with tuned threshold pressure in robust layered metal-organic frameworks. Angew. Chem. Int. Ed. 59, 12725–12730 (2020).
Yang, L. et al. Adsorption site selective occupation strategy within a metal-organic framework for highly efficient sieving acetylene from carbon dioxide. Angew. Chem. Int. Ed. 60, 4570–4574 (2021).
Niu, Z. et al. A MOF-based ultra-strong acetylene nano-trap for highly efficient C2H2/CO2 separation. Angew. Chem. Int. Ed. 60, 5283–5288 (2021).
Zhang, L. et al. Benchmark C2H2/CO2 Separation in an Ultra-Microporous Metal-Organic Framework via Copper(I)-Alkynyl Chemistry. Angew. Chem. Int. Ed. 60, 15995–16002 (2021).
Wang, L. et al. Interpenetration symmetry control within ultramicroporous robust boron cluster hybrid MOFs for benchmark purification of acetylene from carbon dioxide. Angew. Chem. Int. Ed. 60, 22865–22870 (2021).
Di, Z. et al. Cage-like porous materials with simultaneous High C2H2 storage and excellent C2H2/CO2 separation performance. Angew. Chem. Int. Ed. 60, 10828–10832 (2021).
Lou, W. et al. Screening Hoffman-type metal organic frameworks for efficient C2H2/ CO2 separation. Chem. Eng. J. 452, 139296 (2023).
Xu, Z. et al. A robust Th-azole framework for highly efficient purification of C2H4 from a C2H4/C2H2/C2H6 mixture. Nat. Commun. 11, 3163 (2020).
Gu, X.-W. et al. Immobilization of lewis basic sites into a stable ethane-selective mof enabling one-step separation of ethylene from a ternary mixture. J. Am. Chem. Soc. 144, 2614–2623 (2022).
Cao, J.-W. et al. One-step ethylene production from a four-component gas mixture by a single physisorbent. Nat. Commun. 12, 6507 (2021).
Mukherjee, S. et al. Amino-functionalised hybrid ultramicroporous materials that enable single-step ethylene purification from a ternary mixture. Angew. Chem. Int. Ed. 60, 10902–10909 (2021).
Dong, Q. et al. Tuning gate-opening of a flexible metal-organic framework for ternary gas sieving separation. Angew. Chem. Int. Ed. 59, 22756–22762 (2020).
Dong, Q. et al. Shape- and size-dependent kinetic ethylene sieving from a ternary mixture by a trap-and-flow channel crystal. Adv. Funct. Mater. 32, 2203745 (2022).
Xiong, X.-H. et al. Nitro-decorated microporous covalent organic framework (TpPa-NO2) for selective separation of C2H4 from a C2H2/C2H4/CO2 mixture and CO2 capture. ACS Appl. Mater. Interfaces 14, 32105–32111 (2022).
Li, H. et al. An unprecedented pillar-cage fluorinated hybrid porous framework with highly efficient acetylene storage and separation. Angew. Chem. Int. Ed. 60, 7547–7552 (2021).
Cui, X. et al. Pore chemistry and size control in hybrid porous materials for acetylene capture from ethylene. Science 353, 141–144 (2016).
Li, B. et al. An ideal molecular sieve for acetylene removal from ethylene with record selectivity and productivity. Adv. Mater. 29, 1704210 (2017).
Xu, N. et al. A TIFSIX pillared MOF with unprecedented zsd topology for efficient separation of acetylene from quaternary mixtures. Chem. Eng. J. 450, 138034 (2022).
Wang, J. et al. Optimizing pore space for flexible-robust metal-organic framework to boost trace acetylene removal. J. Am. Chem. Soc. 142, 9744–9751 (2020).
Hutter, J. et al. CP2K: atomistic simulations of condensed matter systems. WIREs Comput. Mol. Sci. 4, 15–25 (2014).
Perdew, J. P. et al. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Hartwigsen, C. et al. Relativistic separable dual-space Gaussian pseudopotentials from H to Rn. Phys. Rev. Lett. 58, 3641–3663 (1998).
VandeVondele, J. et al. Gaussian basis sets for accurate calculations on molecular systems in gas and condensed phases. J. Chem. Phys. 127, 114105 (2007).
Klimeš, J. et al. Perspective: Advances and challenges in treating van der Waals dispersion forces in density functional theory. J. Chem. Phys. 137, 120901 (2012).
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Nos. 21908193 and 22205207) and Jinhua Industrial Key Project (2021A22648). We thank the help of Dr. Yunlei Peng from CUP.
Author information
Authors and Affiliations
Contributions
Y.J. and Y.H. contributed equally to this work. Y.Z. designed and guided the project. Y.J. and Y.H. designed and synthesized the materials, performed the majority of the structural characterization, collected gas sorption data and conducted breakthrough experiments. B. L. conducted the DFT calculations. L.W. and H.N. collected X-ray diffraction data and solved the structures. R.K. performed the IAST and Qst calculation. Y.J. and Y.Z. draft the paper. X.H. provided important advice and revised the paper. All authors contributed to the discussion of results and commented on the paper.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Zhang-Wen Wei, Soumya Mukherjee and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Jiang,, Y., Hu,, Y., Luan,, B. et al. Benchmark single-step ethylene purification from ternary mixtures by a customized fluorinated anion-embedded MOF. Nat Commun 14, 401 (2023). https://doi.org/10.1038/s41467-023-35984-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-023-35984-5
- Springer Nature Limited