Boosting Ethylene/Ethane Separation within Copper(I)‐Chelated Metal–Organic Frameworks through Tailor‐Made Aperture and Specific π‐Complexation

Abstract

The development of new materials for separating ethylene (C₂H₄) from ethane (C₂H₆) by adsorption is of great importance in the petrochemical industry, but remains very challenging owing to their close molecular sizes and physical properties. Using isoreticular chemistry in metal–organic frameworks (MOFs) enables the precise design and construction of target materials with suitable aperture sizes and functional sites for gas separations. Herein, it is described that fine‐tuning of pore size and π‐complexation simultaneously in microporous copper(I)‐chelated MOFs can remarkably boost the C₂H₄/C₂H₆ adsorption selectivity. The judicious choice of organic linkers with a different number of carboxyl groups in the UiO‐66 framework not only allows the fine tuning of the pore size but also immobilizes copper(I) ions onto the framework. The tailor‐made adsorbent, Cu^I@UiO‐66‐(COOH)₂, thus possesses the optimal pore window size and chelated Cu(I) ions to form π‐complexation with C₂H₄ molecules. It can rapidly adsorb C₂H₄ driven by the strong π‐complexation interactions, while effectively reducing C₂H₆ uptake due to the selective size‐sieving. Therefore, this material exhibits an ultrahigh C₂H₄/C₂H₆ selectivity (80.8), outperforming most previously described benchmark materials. The exceptional separation performance of Cu^I@UiO‐66‐(COOH)₂ is validated by breakthrough experiments for 50/50 v/v C₂H₄/C₂H₆ mixtures under ambient conditions.

Introduction

Purification of olefins alone accounts for 0.3% of global energy consumption. Olefin/paraffin separation has been thus highlighted as one of seven most important chemical separations. Ethylene (C₂H₄), as the most important olefins, is the mainstay of petrochemical industry, with a global annual production of exceeding 170 million tonnes per year. “Polymer‐grade” specification of ethylene is required for the manufacture of polyethylene plastic. The industrial separation of ethylene from ethylene/ethane (C₂H₄/C₂H₆) mixtures highly relies on the repeated distillation–compression cycling at the temperature as low as −160 °C. Such heat‐driven separation involving in the phase change of isolated fractions, is highly energy‐ and capital‐intensive. Finding energy‐efficient alternatives to distillation would widely lower global energy consumption, carbon emissions, and pollution. It is feasible in principle to separate C₂H₄/C₂H₆ mixtures based on porous solid materials via the energy‐efficient and environmentally friendly adsorption technology. In this context, development of suitable porous adsorbents for ethylene/ethane separation is of highly commercial significance.

A number of porous materials including zeolites, carbon molecular sieves, and alumina, have been explored for the separation of ethylene and ethane. However, the limits on deliberately designing the structure of such purely inorganic materials make them hardly meet the requirement of industrial implement. As an emerging class of microporous materials, metal–organic frameworks (MOFs) hold particular promise for the separation of light hydrocarbons, because of their powerful reticular chemistry that enables them more readily tuning of their pore aperture and functionality. The most popular cases of C₂H₄/C₂H₆ separation at present have been mostly achieved by thermodynamic driven separation. One of the effective strategies is to immobilize the metal ions (such as Cu(I) and Ag(I)) on the pores to form the selective π‐complexation with ethylene, which has been well proved by porous materials. Another analogous π‐complexation effect is typically contributed from open metal sites (OMSs) among MOFs. Such effect is capable of discriminating ethylene from ethane, and thus perform well for the separation of ethylene and ethane. Unfortunately, coadsorption of ethane commonly exists in these materials, originated from dynamic diffusion in oversize apertures and/or polarization of OMSs, thus delimiting the separation performance in most cases. On the other hand, C₂H₄/C₂H₆ separation can also be achieved by the controlled size‐sieving effect. However, considering their nearly identical sizes (only 0.028 nm difference in kinetic diameter), a few MOFs show the selective separation of C₂H₄/C₂H₆ based on this strategy with moderate high selectivities, and only one MOF (UTSA‐280) reported so far has shown the complete exclusion of ethane from C₂H₄/C₂H₆ mixtures with benchmark selectivity. Therefore, it still remains very challenging for MOF materials to separate ethylene from ethylene/ethane mixtures with a satisfied high selectivity based on a single separation mechanism. We speculate that if we combined the effect of π‐complexation and aperture sieving into a single MOF material, the resulting adsorbent may exhibit a superior separation performance for C₂H₄/C₂H₆. However, such synergistic effect has yet to be systematically studied within MOFs for this challenging separation.

Herein, we selected the UiO‐66 family from the abundant MOF gene pool for the study of C₂H₄/C₂H₆ separation due to their typically predictable and readily designable pore size and functionality. In this work, we carefully single out two organic linkers with different number of carboxyl groups, namely benzene‐1,2,4‐tricarboxylic acid (1,2,4‐BTC) and benzene‐1,2,4,5‐tetracarboxylic acid (1,2,4,5‐BTEC), to construct two different fcu‐MOFs (termed as UiO‐66‐COOH and UiO‐66‐(COOH)₂) by using isoreticular chemistry (Figure 1). The judicious choice of organic linkers in the UiO‐66 framework allows us to not only finely tune the pore size but also immobilize copper(I) ions onto the framework. The tailor‐made copper(I)‐chelated adsorbent, Cu^I@UiO‐66‐(COOH)₂, thus possesses the optimal pore window size and specific π‐complexation for C₂H₄/C₂H₆ separation. Gas sorption studies show that Cu^Ι@UiO‐66‐(COOH)₂ can rapidly capture ethylene via the strong π‐complexation affinity, while effectively reduce ethane adsorption due to its size‐sieving effect. As a result, its ideal adsorbed solution theory (IAST) selectivity for 50/50 C₂H₄/C₂H₆ mixtures at ambient conditions can reach up to 80.8, only lower than the benchmark UTSA‐280 but outdistancing all the other previously top‐performing materials, such as Zeolite 13X (13.4), NaETS‐10 (14.7), NOTT‐300 (48.7), Co‐gallate (52) and FeMOF‐74 (13.5), PAF‐SO₃Ag (26.9). Its exceptional separation performance was further validated by the breakthrough experiments on 50/50 v/v C₂H₄/C₂H₆ mixtures under ambient conditions.

X‐ray single crystal structure of UiO‐66‐type MOFs, indicating that 2‐connected linkers bridge 12‐connected [Zr6(µ3‐OH)8(O2C)12] molecular building blocks (MBBs) to form the 3D fcu‐topology frameworks. The pore window size can be systemically modulated via the judicious choice of organic linkers, and it can be further contracted after the configuration of copper(I) ions.

Results and Discussion

Encouraged by the successful construction of isoreticular UiO‐66 with deliberately fine‐tuned apertures, we elected to explore the promise of this fcu‐MOF platform for the adsorptive separation of ethylene form ethane. As shown in Figure , MOF materials with fcu topology hold a unique desired feature that the entrance into the inner pores of the frameworks is only via the ligand‐delimited triangular windows. Construction of isoreticular fcu‐MOF with relatively bulkier linkers will permit the anticipated contraction of pore apertures, and subsequently realize the efficient size‐sieving effect. With this in mind, we selected 2‐connected linkers with different number of carboxyl groups (1,2,4‐BTC and 1,2,4,5‐BTEC) to construct two different fcu‐MOFs (UiO‐66‐COOH and UiO‐66‐(COOH)₂) by using isoreticular chemistry. Both of the synthesized MOFs have the prospective fcu topology as determined by the PXRD analyses (Figure 2b), in which the patterns of both UiO‐66‐(COOH)₂ and UiO‐66‐COOH match well with the theoretical ones of UiO‐66 derived from the crystal structure. Topological analysis of the UiO‐66 series indicates that the carbon atoms from the coordinated carboxylates act the extension points of network, bridging the 2‐connected linkers and the 12‐connected [Zr₆(µ₃‐OH)₈(O₂C−)₁₂] molecular building blocks (MBBs) to produce the 3D framework (Figure a,b). The window size of UiO‐66 was found to be around 6.0 Å. Obviously, both ethylene and ethane molecules can readily diffuse into the internal pores of UiO‐66. When using the bulkier linker of 1,2,4‐BTC instead of 1,4‐BDC, the resulting UiO‐66‐COOH shows a contracted pore window size of 5.3 Å. Further reduction in window size can be fulfilled by using 1,2,4,5‐BTEC to construct the framework. The corresponding size is reduced to be 4.8 Å for UiO‐66‐(COOH)₂. In principle, such pore aperture is still much larger than the sizes of both C₂H₄ (4.1 Å) and C₂H₆ (4.4 Å), and cannot provide the obvious size‐sieving effect for C₂H₄/C₂H₆ separation. However, these functionalized linkers with free carboxyl groups provide us a platform to immobilize some targeted metal sites and thus to further reduce the pore window size (Figure c).

a) XPS for Cu(I) sites of CuΙ@UiO‐66‐(COOH)2 after the surface etching; b) The PXRD patterns for the synthesized UiO‐66 series MOF materials along with the simulated XRD pattern of UiO‐66 (black) derived from the simulated crystal structure; c) The pore size distribution of UiO‐66 series MOF materials; d) Element maps for Cuprum (Cu) of CuΙ@UiO‐66‐(COOH)2; e) N2 sorption isotherms at 77 K of UiO‐66 series MOF materials.

It has been well documented that the incorporated Cu(Ι) and Ag(Ι) ions into porous materials can form the specific π‐complexation with alkenyl units of olefin molecules to fulfill high separation performance. Owing to the existing free carboxyl groups in Cu^Ι@UiO‐66‐COOH and Cu^Ι@UiO‐66‐(COOH)₂, it is reasonable to speculate that the incorporation of such transition metals into these MOFs may be highly effective to discriminate ethylene and ethane significantly. On the other hand, this process can also reduce their pore window sizes and thus make them provide more efficient size‐sieving effect to further improve the separation performance. Therefore, we herein introduced the Cu(I) ions into the aforementioned fcu‐MOF materials by the coordination with bare carboxyl groups to form two Cu(I) immobilized materials, namely Cu^Ι@UiO‐66‐COOH and Cu^Ι@UiO‐66‐(COOH)₂. First, we confirmed the presence of Cu(I) ions in these metalated materials by X‐ray photoelectron spectroscopy (XPS) analyses (Figure a; Figure S5, Supporting Information). There exists a characteristic signal from copper(I) at binding energies of 952.9 and 933.1 eV, corresponding to the peaks of Cu 2p^1/2 and 2p^3/2, respectively. Then, their phase purity of bulk materials was determined by the PXRD analysis (Figure 2b and S1, Supporting Information). FE‐SEM was performed to characterize the morphology of MOF materials (Figure S2, Supporting Information), all of which feature the homogeneous particles with a size of 100–200 nm. Apparently, the immobilization process shows no significant effect on the morphology of these robust MOFs. Fourier infrared spectrum analysis (FTIR) indicated that a strong band at 1715.7 cm⁻¹ can be clearly observed in UiO‐66‐COOH and UiO‐66‐(COOH)₂, mainly attributed to the CO stretching vibration of uncoordinated ‐COOH groups (Figure S3, Supporting Information). After the metalation of two MOFs, this peak almost disappeared in Cu^Ι@UiO‐66‐(COOH)₂ and Cu^Ι@UiO‐66‐COOH, further confirming the success of chelating the Cu(I) ions into carboxyl groups. The ICP‐MS (Table S1, Supporting Information) and energy‐dispersive X‐ray (EDS) analyses (Figure d; Figure S6, Supporting Information) comprehensively determined that there are ≈47% and ≈52% uncoordinated –COOH groups transformed into –COOCu for Cu^Ι@UiO‐66‐COOH and Cu^Ι@UiO‐66‐(COOH)₂, respectively.

Permanent porosity studies of these four UiO‐66‐type MOFs were performed by nitrogen (N₂) sorption at 77 K, and all isothermals show typically reversible type‐I isothermals (Figure c,e). The Brunauer–Emmett–Teller (BET) surface area was estimated to be 712.8 and 622.3 m² g⁻¹ for UiO‐66‐COOH and UiO‐66‐(COOH)₂ respectively, notably lower than that of the parent UiO‐66 (1110 m² g⁻¹) due to the bulkier linkers. With the copper ions chelated into the MOFs, the porosity can be further reduced and the BET surface area of Cu^Ι@UiO‐66‐COOH and Cu^Ι@UiO‐66‐(COOH)₂ decreased to be 437.7 and 319.7 m² g⁻¹, respectively. The variation of pore sizes among these MOFs, determined by Horvath–Kawazoe method, corresponds well to the results of the porosities. As shown in Figure c, the pore size is also gradually reduced from 5.6 Å in UiO‐66‐COOH to 4.1 Å in Cu^Ι@UiO‐66‐(COOH)₂, with the increased number of functional carboxyl groups and the chelation of copper ions. These experimental results are consistent well with the aforementioned pore sizes calculated from the crystal structures. The optimal aperture (4.1 Å) of Cu^Ι@UiO‐66‐(COOH)₂ just falls in the range of the kinetic diameter of C₂H₄ (4.1 Å) and C₂H₆ (4.4 Å), which may provide an efficient size‐sieving effect on C₂H₄/C₂H₆ separation.

Single‐component adsorption isotherms of C₂H₄ and C₂H₆ including low‐pressure adsorption data for all four UiO‐66‐type MOFs were collected and shown in Figure 3a,b; Figures S9 and S10 (Supporting Information). Evidently, both of UiO‐66‐COOH and UiO‐66‐(COOH)₂ without Cu(I) ions show the very similar adsorption capacities toward C₂H₄ and C₂H₆, indicating that neither of them can discriminate the two gases due to their oversize aperture and the absence of specific recognition sites. Conversely, both copper‐chelated materials show distinct ethylene‐selective sorption behaviors. Especially, Cu^Ι@UiO‐66‐(COOH)₂ adsorbs C₂H₄ rapidly at low‐pressure region, with an appreciable uptake capacity of 1.86 mmol g⁻¹ at 298 K and 1.0 bar. This C₂H₄ uptake at 1.0 bar approaches the stoichiometric quantity (2.14 mmol g⁻¹) expected if one gas molecule is adsorbed per Cu(I) sites, indicating that the Cu(I) ions mainly account for its C₂H₄ uptake. Besides the increased C₂H₄ uptake, Cu^Ι@UiO‐66‐COOH and Cu^Ι@UiO‐66‐(COOH)₂ also prove the notably reduced ethane uptake, indicating an efficient size‐selective effect. Cu^Ι@UiO‐66‐COOH reduces ≈35.6% uptake for C₂H₆ at ambient conditions compared with UiO‐66‐COOH, and about 51.3%‐decreased uptake of C₂H₆ occurs on Cu^Ι@UiO‐66‐(COOH)₂ at the same conditions. As illustrated in Figure c, when the pore size of fcu MOFs gradually decreases from UiO‐66‐COOH to Cu^Ι@UiO‐66‐(COOH)₂, the uptake capacity of C₂H₆ at 0.01 bar becomes fewer and fewer. In contrast, the C₂H₄ uptake at 0.01 bar increases in the order of UiO‐66‐COOH < UiO‐66‐(COOH)₂ < Cu^Ι@UiO‐66‐COOH <Cu^Ι@UiO‐66‐(COOH)₂. Thus, the tailor‐made Cu^Ι@UiO‐66‐(COOH)₂ features the highest C₂H₄ adsorption (0.71 mmol g⁻¹) but the lowest C₂H₆ adsorption (0.02 mmol g⁻¹), thus offering an ultrahigh C₂H₄/C₂H₆ uptake ratio of 35.5 at 0.01 bar and 298 K. In comparison to other promising MOFs, both the C₂H₄ capture capacity and the low C₂H₆ uptake at low pressure are also very significant (Figures S11 and S12, Supporting Information). These results clearly indicate that the immobilization of Cu(I) ions into Cu^Ι@UiO‐66‐(COOH)₂ can efficiently improve the C₂H₄ capture capacity via the strong π‐complexation interactions, while the contracted aperture size provides the size‐sieving effect to reduce C₂H₆ uptake simultaneously, thus affording the excellent C₂H₄/C₂H₆ separation.

a) Single‐component adsorption isotherms for C2H4 and C2H6 of UiO‐66‐COOH and CuΙ@UiO‐66‐COOH at 298 K; b) Single‐component adsorption isotherms for C2H4 and C2H6 of UiO‐66‐(COOH)2 and CuΙ@UiO‐66‐(COOH)2 at 298 K; c) Experimental C2H4 and C2H6 adsorption uptake of UiO‐66 series MOF materials at 0.01 bar; d) The isosteric heat (Qst) of C2H4 adsorption in the UiO‐66‐type MOFs; e) IAST calculations of activated UiO‐66‐type MOFs for the C2H4/C2H6 separation at 298 K; f) IAST calculations of representative materials explored for C2H4/C2H6 separation at room temperature.

Such C₂H₄ adsorption behavior of four fcu‐MOFs can be explained by the isosteric heat of adsorption (Q_st), calculated by adsorption isotherms at different temperatures. As shown in Figure d and Figure S15 (Supporting Information), the Q_st results are in good agreement with the experimental C₂H₄ uptakes. Higher Q_st values of C₂H₄ were found in both Cu(I)‐chelated MOFs, thus validating that Cu(I) ions indeed enhance the binding affinity of C₂H₄. The Q_st value (48.5 kJ mol⁻¹) for C₂H₄ at close to zero loading in Cu^Ι@UiO‐66‐(COOH)₂ is the highest, followed by Cu^Ι@UiO‐66‐COOH (32.9 kJ mol⁻¹), UiO‐66‐(COOH)₂ (27.4 kJ mol⁻¹), and UiO‐66‐COOH (24.3 kJ mol⁻¹). This C₂H₄ Q_st value of Cu^Ι@UiO‐66‐(COOH)₂ is even comparable to that of M₂(dobdc) with high density of OMSs, implying the strong affinity between Cu^Ι@UiO‐66‐(COOH)₂ and C₂H₄. We speculate that such strong C₂H₄ affinity is mainly attributed to the synergistic effect of the specific π‐complexation interactions and the smaller pores in Cu^Ι@UiO‐66‐(COOH)₂. Instead, the incorporation of Cu(I) ions shows a negative effect on the affinity of ethane, in which the Q_st values of C₂H₆ for Cu^Ι@UiO‐66‐COOH and Cu^Ι@UiO‐66‐(COOH)₂ are even lower than the unmetalized MOFs. It is worth to note that the Q_st value of C₂H₄ in Cu^Ι@UiO‐66‐(COOH)₂ is much lower than that of most Ag^Ι‐chelated porous adsorbents, including PAF‐1‐SO₃Ag (106 kJ mol⁻¹) and (Cr)‐MIL‐101‐SO₃Ag (120/63 kJ mol⁻¹), affording a relatively low regeneration energy among these kinds of materials.

The C₂H₄/C₂H₆ adsorption selectivity (S_ads) of these UiO‐66 materials was calculated by IAST based on the measured sorption isotherms. Figure e shows the data obtained at 298 K. Due to the very similar C₂H₄ and C₂H₆ adsorption isotherms, the S_ads value of UiO‐66‐COOH is as low as 0.9. Unlike UiO‐66‐COOH, the Cu(I)‐chelated Cu^Ι@UiO‐66‐COOH shows a significantly improved S_ads value up to 14.5. With the tailor‐made apertures and Cu(I) ions, we found that Cu^Ι@UiO‐66‐(COOH)₂ exhibits an ultrahigh S_ads up to 225.7 at 0.01 bar and decreases down to 80.8 at 1.0 bar, which is orders of magnitude higher than all the other UiO‐66 materials. We note that this S_ads value of Cu^Ι@UiO‐66‐(COOH)₂ is only next to UTSA‐280, outperforming all the other benchmark materials including Zeolite 13X (13.4), NaETS‐10 (14.7), Co‐gallate (52), NOTT‐300 (48.7) (Figure f). In addition, it is also significantly higher than Ag(I)‐chelated adsorbents like PAF‐1‐SO₃Ag (27) and (Cr)‐MIL‐101‐SO₃Ag (16/9.6). Considering its lower C₂H₄ Q_st compared with these Ag(I)‐chelated adsorbents, the size‐sieving effect on Cu^Ι@UiO‐66‐(COOH)₂ also contributes significantly to its exceptional selectivity apart from the π‐complexation interactions. Moreover, the sieving effect of C₂H₄/C₂H₆ separation in Cu^Ι@UiO‐66‐(COOH)₂ can be strengthened at low temperature of 273 K, wherein the corresponding S_ads value can be further enhanced to 110 at 1 bar (Figure S20, Supporting Information).

The strong C₂H₄ capture capacity and ultrahigh IAST selectivity of Cu^Ι@UiO‐66‐(COOH)₂ prompted us to perform the breakthrough experiments in order to evaluate its actual separation efficiency. Such experiments were conducted in a packed column filled with the activated Cu^Ι@UiO‐66‐(COOH)₂ powder, under 1.0 mL min⁻¹ feed gas of equimolar C₂H₄/C₂H₆ mixture at 298 K (Figure 4; Figure S27, Supporting Information). As shown in Figure a, efficient separation of C₂H₄ from 50/50 C₂H₄/C₂H₆ mixtures can be successfully fulfilled by using the activated Cu^Ι@UiO‐66‐(COOH)₂. C₂H₆ gas was first eluted through the separation bed, while no C₂H₄ was detected until about 86 min. The adsorbed amount of C₂H₄, enriched from the equimolar C₂H₄/C₂H₆ mixtures, was calculated to be 1.92 mol per kg for a given cycle. Subsequently, the captured C₂H₄ gas in the column can then be recovered with high purity during the regeneration desorption step, which was carried out by the sweeping He gas (10 mL min⁻¹) at 413 K. As shown in Figure b, the regeneration of Cu^Ι@UiO‐66‐(COOH)₂ revealed that the adsorbed gas can be almost completely recovered within 15 min and no detectable C₂H₄ and C₂H₆ were found after 15 min, notably faster than that of Ag‐doped adsorbent (typically thousands of minutes). Moreover, high pure (92.5%) ethylene can be obtained during one cycle of the adsorption–desorption procedures. Multiple cycling column breakthrough tests under the same conditions showed that the breakthrough times of Cu^Ι@UiO‐66‐(COOH)₂ for both C₂H₄ and C₂H₆ remains almost unchanged within three continuous cycles, confirming its good recyclability for C₂H₄/C₂H₆ separation (Figure c). Furthermore, Cu^Ι@UiO‐66‐(COOH)₂ was proved to be insensitive to moisture, since its C₂H₄ adsorption is not affected after soaking it into oxygen‐free water (Figure S29, Supporting Information).

Multicomponent column breakthrough results for CuΙ@UiO‐66‐(COOH)2 at 298 K. a) The breakthrough curves of CuΙ@UiO‐66‐(COOH)2 for the C2H4/C2H6 (50/50, v/v) separation; b) The desorption curves of CuΙ@UiO‐66‐(COOH)2 under 10.0 mL min−1 sweeping He gas at 413 K; c) Multiple cycles of breakthrough tests of CuΙ@UiO‐66‐(COOH)2 for the C2H4/ C2H6 (50/50, v/v) separation.

Conclusions

Herein, we precisely designed and constructed two copper(I)‐chelated metal–organic frameworks by using isoreticular chemistry, for the very challenging ethylene/ethane separation. The tailor‐made adsorbent, Cu^I@UiO‐66‐(COOH)₂, possesses the optimal pore window size and specific π‐complexation. Our foregoing results indicated that this material can rapidly adsorb ethylene driven by the strong affinity of π‐complexation, while notably lessen ethane uptake due to its size‐sieving effect. This rare synergistic effect of the specific π‐complexation interactions and efficient size‐sieving effect in Cu^I@UiO‐66‐(COOH)₂ thus led to an ultrahigh IAST selectivity of 80.8 at ambient conditions for 50/50 C₂H₄/C₂H₆ mixture, outdistancing most of previously benchmark porous materials reported. The exceptional separation performance was further confirmed by the detailed breakthrough experiments on 50/50 v/v C₂H₄/C₂H₆ mixtures. In light of its attractive design strategy and ultrahigh selectivity, we believe that Cu^Ι@UiO‐66‐(COOH)₂ is placed among the best‐performing porous materials for the challenging C₂H₄/C₂H₆ separation, and this work may provide some guidance to develop new porous materials for boosting olefin/paraffin separation performance.

Experimental Section

Materials and Methods: All reagents and solvents including the organic ligands, 1,2,4‐BTC and 1,2,4,5‐BTEC, used to construct the UiO‐66‐type MOFs, were commercially available and used without further purification. Powder X‐ray diffraction (PXRD) patterns were collected in the 2θ = 5°–50° range on an X'Pert PRO diffractometer with Cu Kα (λ = 1.542 Å) radiation at room temperature. Inductively coupled plasma‐mass spectrometry (ICP‐MS) was performed on a Thermo Scientific XSERIES 2 ICP‐MS system. Fourier transform infrared (FT‐IR) spectrum was recorded on Thermo Fisher Nicolet iS10 spectrometer using KBr pallets in the 500–4000 cm⁻¹ range. The morphology was investigated using a field‐emission scanning electron microscopy (FE‐SEM, Hitachi S4800). All gas sorption isotherms of UiO‐66‐type MOFs were obtained from the Micromeritics ASAP 2020 surface area analyzer and pore size analyzer. An ice‐water bath and water bath were used for C₂H₄ and C₂H₆ gases adsorption isotherms at 273 and 298 K, respectively.

Preparation of Cu^I@UiO‐66‐(COOH)₂ and Cu^I@UiO‐66‐COOH: The activated UiO‐66‐(COOH)₂ (≈200 mg, 0.09 mmol) and cuprous chloride (CuCl, about 108 mg, 1.08 mmol) were dispersed in acetonitrile (3 mL) in a 20 mL Teflon‐capped vial which was tightly wrapped with polytetrafluoroethylene sealing tape. The foregoing procedures were carefully operated in a glovebox with positive N₂ pressure. Then, the vials were heated in an oven at 80 °C for two weeks to afford the resultant color‐changed powder Cu^I@UiO‐66‐(COOH)₂. The preparation of Cu^I@UiO‐66‐COOH was referred to the preparation of Cu^I@UiO‐66‐(COOH)₂, just the replace of the amount of cuprous chloride (≈62 mg, 0.62 mmol) and the activated UiO‐66‐COOH (≈200 mg, 0.10 mmol).

Activation of Cu^I@UiO‐66‐(COOH)₂ and Cu^I@UiO‐66‐COOH: Before the gas sorption measurements, the obtained powder solids were directly solvent‐exchanged by methanol for at least ten times in a glovebox with positive N₂ pressure. After solvent‐exchange, the powder materials were carefully transferred into in the adsorption tube, then were evaluated from the Micromeritics ASAP 2020 surface area analyzer at 373 K for 12 h and 413 K for another 12 h to yield the activated Cu^I@UiO‐66‐COOH and Cu^I@ UiO‐66‐(COOH)_2.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 51472217, 51432001, 21606163, and 51803179), the Zhejiang Provincial Natural Science Foundation of China (Nos. LR13E020001 and LZ15E020001), and Fundamental Research Funds for the Central Universities (Nos. 2015QNA4009, 2016FZA4007, and 2018QNA4010), and partly supported by Welch Foundation (AX‐1730).

Conflict of Interest

The authors declare no conflict of interest.