Metal–organic frameworks with two different-sized aromatic ring-confined nanotraps for benchmark natural gas upgrade

Abstract

Recovery of light alkanes from natural gas is of great significance in petrochemical production. Herein, a promising strategy utilizing two types of size-complementary aromatic ring-confined nanotraps (called bi-nanotraps here) is proposed to efficiently trap ethane (C₂H₆) and propane (C₃H₈) selectively at their respective sites. Two isostructural metal–organic frameworks (MOFs, SNNU-185/186), each containing bi-nanotraps decorated with six aromatic rings, are selected to demonstrate the feasibility of this method. The smaller nanotrap acts as adsorption sites tailored for C₂H₆ while the larger one is optimized in size for C₃H₈. The separation is further facilitated by the large channels, which serve as mass transfer pathways. These advanced features give rise to multiple C–H⋯π interactions and size/shape-selective interaction sites, enabling SNNU-185/186 to achieve high C₂H₆ adsorption enthalpy (43.5/48.8 kJ mol⁻¹) and a very large thermodynamic interaction difference between C₂H₆ and CH₄. Benefiting from the bi-nanotrap effect, SNNU-185/186 exhibits benchmark experimental natural gas upgrade performance with top-level CH₄ productivity (6.85/6.10 mmol g⁻¹), ultra-high purity and first-class capture capacity for C₂H₆ (1.23/0.90 mmol g⁻¹) and C₃H₈ (2.33/2.15 mmol g⁻¹).

---

Introduction

Natural gas is mainly composed of methane (CH₄, 85% by volume) which is an important clean energy source and essential chemical feedstock. However, the presence of ethane (9% C₂H₆) and propane (3% C₃H₈) not only reduces the combustion efficiency and conversion rate of CH₄, but also affects the safety of CH₄ storage.^1–5 In addition, C₂H₆ and C₃H₈ are valuable petrochemical feedstocks for the manufacture of alkenes and polymers.^6–11 Therefore, efficient separation and recovery of C₂H₆ and C₃H₈ from natural gas are important for both CH₄ upgrading and full energy utilization. The current separation process is mainly based on cryogenic distillation technology which is energy intensive and environmentally unfriendly.^12–14 By contrast, adsorption-based separation using solid adsorbents is cost- and energy-efficient.^15–20

With guest accessible porosity, and a variety of different components contributing to the tunability of pore structures and surface properties, porous coordination polymers (PCPs) or metal–organic frameworks (MOFs) are a promising class of solid adsorbents capable of overcoming the performance bottleneck resulting from imprecise pore control, few structural building units, and a limited number of coordination pathways of traditional adsorbents.^21–25 To date, many MOFs have been investigated for C₃H₈/C₂H₆/CH₄ separation based on the thermodynamic separation mechanism. Generally, creating a polar pore surface (C–H⋯O/N/F hydrogen bonds)^26–30 or non-polar pore environments (aromatic C–H⋯π bonds or confined aliphatic C–H⋯C hydrogen bonds), and simultaneously regulating pore size to provide a confined space for enhanced MOF–gas interaction,^31–34 are effective strategies. One difficulty is that the performance of MOF materials is limited by the C₂H₆/CH₄ separation step as shown in Zn-BPZ-SA,⁵ LIFM-ZZ-1,⁹ BSF-2,¹⁴ MIL-101,³⁰ UiO-66-NaPh³⁴ and CFA-1,³⁵ largely due to the greater similarity in molecular size and chemical properties between C₂H₆ and CH₄.^4,34 Another often-neglected but crucial reason is the competitive adsorption between C₃H₈ and C₂H₆ in the ternary gas separation system.³⁶ C₃H₈ molecules preferentially occupy adsorption sites to form stronger interaction with the framework due to their larger polarizability and molecular size compared to C₂H₆, which further increases the difficulty of the C₂H₆/CH₄ separation step. Therefore, the key to improving the performance of C₂H₆/CH₄ separation is to increase the thermodynamic difference between C₂H₆ and CH₄ while simultaneously installing size-selective sites for C₂H₆ and C₃H₈ to minimize the competitive adsorption between C₂H₆ and C₃H₈.

Fortunately, the difference in molecular polarizability and the number of H-donors between C₃H₈, C₂H₆, and CH₄ could enable thermodynamic preferential adsorption of C₃H₈ or C₂H₆ by creating polar/non-polar pore surfaces. Compared to single adsorption sites, nanotraps or molecular traps that allow for the selective capture of specific gas molecules are more effective and attractive.^37–45 With multiple and gas-specific adsorption sites, nanotraps provide stronger binding interactions and recognition capabilities for target molecules, which is promising for widening the thermodynamic gap between C₃H₈, C₂H₆ and CH₄. However, the construction of nanotraps is rare and challenging for MOFs.

In addition, the combination of strong binding affinity and molecular sieving should have great potential in preventing competitive adsorption and achieving the most effective purification and separation. Its high efficiency and application potential have been demonstrated in multi-component separation.^36,46–48 The construction of coexistent C₂H₆- and C₃H₈-selective adsorption sites in one MOF system is difficult because C₂H₆ and C₃H₈ tend to occupy the same sites, with C₃H₈ being preferred. However, utilizing the difference in kinetic diameter between C₃H₈ (5.1 Å) and C₂H₆ (4.4 Å) to discriminate between them could be an effective method to eliminate competitive adsorption (Table S1†), leading to enhanced MOF performance in the key C₂H₆/CH₄ step. Overall, the combination of nanotraps with the molecular sieving effect is expected to facilitate multiple and strong interactions and widen the thermodynamic difference between C₃H₈, C₂H₆ and CH₄. It will also help install sites targeting selective adsorption for C₃H₈ and C₂H₆ to reduce their competitive adsorption and, therefore, maximize the separation performance (Scheme 1).

Scheme 1 A proposed strategy for paraffin separation with the synergistic effect of C–H⋯π interactions and nanotraps.

Herein, a promising example of bi-nanotraps is demonstrated. In two newly constructed MOFs (SNNU-185/186), the smaller type of nanotraps with appropriate size and shape is ideally suited for accommodating C₂H₆ based on the thermodynamic-molecular sieving mechanism and the larger nanotraps are more advantageous for trapping C₃H₈ thanks to the thermodynamic interaction difference. In the meantime, the large channels serve as mass transfer pathways, promoting gas molecules to enter the adsorption sites from pore walls. As a result, multiple C–H⋯π interactions and highly discriminating interaction sites are achieved in one unprecedented MOF system, contributing to benchmark −Q_st for C₂H₆ and the exceptionally large −Q_st difference between C₂H₆ and CH₄. The overall effect is greatly increased thermodynamic difference and weakened competitive adsorption. Together with excellent adsorption capability and high stability, SNNU-185 and SNNU-186 can produce ultra-high purity CH₄ (>99.9999%) at flow rates of 4/6 mL min⁻¹ with top-level productivities for CH₄ (6.85 and 6.10 mmol g⁻¹), and top-notch capture capacities for C₂H₆ (1.23 and 0.90 mmol g⁻¹) and C₃H₈ (2.33 and 2.15 mmol g⁻¹) in breakthrough experiments. GCMC simulation provides a molecular level insight and mechanistic explanation of the role of bi-nanotraps. This work not only provides promising materials for natural gas upgrade, but also reveals an effective design philosophy toward the development of porous coordination polymers for challenging multi-component separation processes.

Results and discussion

Hydrothermal reactions of 1,3,5-tris(4-pyridyl)-benzene (TPB) or 2,4,6-tri(4-pyridinyl)-1-pyridine (TPP), 2,5-pyridinedicarboxylic acid (2,5-PDC) and cobalt acetate hydrate were used to synthesize SNNU-185 (with TPB) and SNNU-186 (with TPP) (Fig. S1 and S2†). From the single crystal analysis, they are found to be isostructural and crystallize in the hexagonal space group P6̄c2 with the formula of {[Co₃(μ₃-OH)][Co(2,5-PDC)₂]₃(TPB/TPP)₃}_n (Fig. S3 and Table S2†), which is isostructural with our reported SNNU-54 (ref. 49) synthesized under different conditions (Fig. S4 and S5†). SNNU-185 and SNNU-186 were selected to demonstrate the feasibility of the aromatic ring-confined bi-nanotrap strategy for efficiently and separately trapping C₂H₆ and C₃H₈ based on a thermodynamic-molecular sieving coupling mechanism. As shown in Fig. 1, both SNNU-185 and SNNU-186 contain two distinct Co atoms that form two types of secondary building units (SBU I and SBU II). The Co₁ center is six-coordinated by four O-donors from four different 2,5-PDC ligands, one N-donor from TPB/TPP and one central μ₃-OH. Three Co₁ atoms form a [Co₃(μ₃-OH)(COO)₆] trimer (Fig. 1a and b), acting as a 9-connected node. The Co₂ atom (Fig. 1c and d) is hexacoordinated in a distorted octahedral configuration formed by two carboxylate O and two pyridine N atoms from two 2,5-PDC ligands, and two N atoms from TPB/TPP ligands. Chelate rings are on the same side, forming [Co(2,5-PDC)₂(4-pyridine)₂] MOLs (metal–organic linkers) in a cis-configuration, which is considered a 4-connected node. Two Co₁ trinuclear clusters and three Co₂ MOLs connect with each other to build a trigonal bipyramid-type cage along the c-axis, which is further extended into 1D {[Co₃(μ₃-OH)][Co(PDC)₂]₃}_n chains (Fig. 1g).

Fig. 1 Schematic representation showing the assembly of SNNU-185/186: (a–d) two types of clusters and their simplified representations in SNNU-185/186. (e) Triangular ligands TPB and TPP used to construct SNNU-185 and SNNU-186, respectively; (f) 3D structure of SNNU-185/186 viewed along the c-axis direction. Schematic diagram of nanotraps in (g) large-type nanotrap chains and (i) small-type nanotrap channels: (h) aromatic ring-confined nanotrap 1 and (j) aromatic ring-confined nanotrap 2 in SNNU-185/186. Schematic diagram of nanotraps: thermodynamic C₃H₈-selective nanotrap 1 and thermodynamic-molecular sieving C₂H₆-selective nanotrap 2. (k). Schematic diagram of 3D structures of SNNU-185/186 viewed along the (l) c-axis direction and (m) b-axis direction.

Significantly, each cage is decorated with six aromatic rings (from 2,5-PDC) and each aromatic ring layer has three aromatic rings which are distributed in a staggered pattern from top to bottom (Fig. 1h). The available inner cavity is ∼7.2 Å × 7.2 Å and the window size is ∼5.3 Å × 4.8 Å (Fig. 1h and S3c†). This cage size and environment match well with the size and shape of C₂H₆ and C₃H₈, acting as “aromatic ring-confined nanotrap 1” which is expected to promote the formation of strong host–guest interactions. Moreover, because the inner cavity of nanotrap 1 is more compatible with C₃H₈, this type of large nanotrap can act as C₃H₈-selective interaction sites. The 3D framework of SNNU-185/186 is formed when each cage chain connects six neighboring chains via six TPB/TPP ligands. The resulting small-sized channels (Fig. 1i) are also modified by six aromatic rings (from peripheral pyridine rings of TPB/TPP) in a staggered pattern from top to bottom, which are referred to as “aromatic ring-confined nanotrap 2” (Fig. 1j). This small nanotrap 2 has a pore size of about 4.8 Å × 4.8 Å and a window size of about 4.3 Å × 4.8 Å (Fig. S3d†). By summarizing and analyzing MOF materials with high C₂H₆/CH₄ separation performance such as Ni(TMBDC)(DABCO)_0.5 (ref. 2) (5.0 Å), ZUL-C₂ (ref. 4) (5.3 Å), Ni-MOF 1 (ref. 50) (5.7 Å), and SNNU-Bai₆₉ (ref. 3) (6.4 Å), it can be concluded that such pore sizes favor the formation of strong interactions with C₂H₆ through C–H⋯π bonds and can amplify the thermodynamic gap between C₂H₆ and CH₄ to the maximum extent. Furthermore, considering the size-exclusion potential of C₃H₈ as shown in KAUST-7,^51,52 Y-abtc,⁵³ Co-gallate,⁵⁴ JNU-3a,⁵⁵ and NTU-85-WNT⁵⁶ which have aperture sizes of about 4.7 Å, 4.7 Å, 5.2 Å, 5.3 Å and 4.6 Å respectively, this small-size channel is expected to limit C₃H₈ entry to some extent, thus creating C₂H₆-selective interaction sites based on a molecular sieving mechanism. Finally, the large channel decorated with oxygen atoms from uncoordinated carboxylic acids can also interact with gas molecules. However, considering its large pore size, the main role of the large channel might be to facilitate gas diffusion, allowing gas molecules to enter size-selective adsorption sites from pore walls. It can be concluded that the construction of C₂H₆-selective nanotraps, C₃H₈-selective nanotraps and mass transfer channels is achieved in SNNU-185 and SNNU-186 (Fig. 1k–m). Such a structural arrangement lays the foundation for efficient separation and recovery of C₂H₆ and C₃H₈ from natural gas.

PXRD patterns of the as-synthesized SNNU-185 and SNNU-186 samples align well with the calculated patterns obtained from the single crystals, indicating their successful synthesis with high purity (Fig. S6†). Also, a decagram scale synthesis of SNNU-186 was carried out under reflux conditions for 3 days. As shown in Fig. S7 and S8,† impurity-free SNNU-186 (∼12.7 g) could be easily obtained without loss of crystallinity, demonstrating its scalability. The TG analysis data showed that the as-synthesized and solvent-exchanged SNNU-185 and SNNU-186 are stable up to around 573 K, indicating their high thermal stability (Fig. S9†). Overall, the architecture of shape/size-matched bi-nanotraps, combined with size selectivity based on molecular sieving mechanisms and high stability inspired us to further investigate their C₃H₈/C₂H₆/CH₄ separation performance.

The permanent porosity of activated SNNU-185 and SNNU-186 was confirmed using N₂ adsorption–desorption isotherms at 77 K. As shown in Fig. S10,† both MOFs exhibit microporous type I sorption isotherms with calculated Brunauer–Emmett–Teller (BET) surface areas of 886 m² g⁻¹ and 875 m² g⁻¹ for SNNU-185 and SNNU-186, respectively. Single component adsorption isotherms for CH₄, C₂H₆, and C₃H₈ on SNNU-185 and SNNU-186 were measured at different temperatures (273, 283 and 298 K) and at pressures up to 1 bar (Fig. 2a, b and S11†). Taking advantage of the bi-nanotrap structure, SNNU-185/186 adsorbed much more C₃H₈ and C₂H₆ than CH₄ under the same conditions, indicating their potential for C₃H₈/C₂H₆/CH₄ separation. At 298 K and 1.0 bar, the C₂H₆ storage capacity of SNNU-185 and SNNU-186 can reach 69.8 cm³ g⁻¹ (3.12 mmol g⁻¹) and 74.3 cm³ g⁻¹ (3.32 mmol g⁻¹), respectively. These values exceed those of many well-known reported MOF adsorbents, such as Zn-BPZ-SA⁵ (2.97 mmol g⁻¹), ZUL-C₁ (ref. 4) (2.95 mmol g⁻¹), ZUL-C₂ (ref. 4) (2.82 mmol g⁻¹), BSF-3 (ref. 38) (2.35 mmol g⁻¹), SNNU-Bai₆₉ (ref. 3) (2.0 mmol g⁻¹), ECUT-Th-10a¹¹ (1.72 mmol g⁻¹) and UiO-66-Naph³⁴ (1.24 mmol g⁻¹). The C₃H₈ isotherms of SNNU-185 and SNNU-186 at 273/298 K exhibited saturated uptakes of 98.6/94.0 cm³ g⁻¹ (4.40/4.20 mmol g⁻¹) and 108.0/97.1 cm³ g⁻¹ (4.82/4.33 mmol g⁻¹), respectively, surpassing those of most MOF materials such as ZUL-C₁ (ref. 4) (2.72 mmol g⁻¹), ZUL-C₂ (ref. 4) (2.52 mmol g⁻¹), BSF-3 (ref. 38) (2.98 mmol g⁻¹), Ni-MOF 1 (ref. 50) (3.56 mmol g⁻¹) and LIFM-ZZ-1 (ref. 9) (4.06 mmol g⁻¹). Thanks to strong interactions from thermodynamic C₃H₈-selective nanotraps, the C₃H₈ uptake shows steep adsorption at low pressure, which is beneficial for capturing C₃H₈. For C₂H₆, steep adsorption at low pressure especially at 0–50 mm_Hg can also be observed, which might be attributed to strong interaction with C₂H₆-selective nanotraps. In addition, considering the presence of water and acidic gases such as H₂S and SO₂ in raw natural gas, detailed stability tests were further performed. After being treated under different conditions including soaking in water, exposure to aqueous solutions with different pH values or exposed to air for an extended period, satisfactory water stability and pH stability of these two MOFs were verified by adsorption/desorption tests (Fig. 2c).

Fig. 2 C₃H₈, C₂H₆ and CH₄ sorption isotherms of (a) SNNU-185 and (b) SNNU-186 at 273/283/298 K. (c) Stability tests: 77 K N₂ adsorption/desorption isotherms and pore size distributions of SNNU-186 after being treated under different conditions, and C₂H₆ sorption isotherms of SNNU-185 and SNNU-186 before and after exposure to air for 14 days. (d) −Q_st plots of C₃H₈, C₂H₆ and CH₄. (e) Summary of −Q_st (C₂H₆) and the difference between −Q_st (C₂H₆) and −Q_st (CH₄) among all the reported MOF materials used for C₃H₈/C₂H₆/CH₄ separation. (f) IAST selectivity for C₂H₆/CH₄ mixtures (50/50) and C₃H₈/CH₄ mixtures (50/50) of SNNU-185 and SNNU-186 at 298 K.

To measure the binding affinities between the host surface and guest gas molecules, the adsorption enthalpy (−Q_st) of C₃H₈, C₂H₆ and CH₄ in SNNU-185/186 was calculated (Fig. 2d, S12 and Table S3†). Significantly, SNNU-186 shows the highest −Q_st value of 48.8 kJ mol⁻¹ for C₂H₆ compared to all reported MOF materials used for C₃H₈/C₂H₆/CH₄ separation such as ZUL-C₂ (ref. 4) (45 kJ mol⁻¹), Ni(TMBDC)(DABCO)_0.5 (ref. 2) (36 kJ mol⁻¹), ZUL-C1 (ref. 4) (33 kJ mol⁻¹) and SNNU-Bai₆₉ (ref. 3) (30.6 kJ mol⁻¹) (Table S4†). Importantly, SNNU-185 and SNNU-186 exhibit the largest −Q_st difference between C₂H₆ and CH₄ among MOF materials used for natural gas upgrading (Fig. 2e). This benchmark −Q_st for C₂H₆ and the largest thermodynamic interaction difference between C₂H₆ and CH₄ could be attributed to the bi-nanotrap structure which fully takes advantage of the synergistic effects of C–H⋯π interactions and nanotraps. As a result, multiple and strong C–H⋯π interactions and an increased thermodynamic interaction difference between gas molecules were achieved, which are beneficial for improving the performance in the key C₂H₆/CH₄ step. Due to strong C–H⋯π interactions in C₃H₈-nanotraps and the rejection of C₃H₈ by C₂H₆-nanotraps, the −Q_st values for C₃H₈ in SNNU-185/186 are moderate (48.1/47.2 kJ mol⁻¹).

Ideal adsorbed solution theory (IAST) was used to further evaluate the separation potential of SNNU-185/186 for 50/50 C₂H₆/CH₄ mixtures and 50/50 C₃H₈/CH₄ mixtures at 298 K (Fig. 2f, S13–15 and Table S5†). At 1 kPa, for C₂H₆/CH₄, the IAST selectivities of SNNU-185 and SNNU-186 are 43.4 and 52.1, respectively. For 50/50 C₃H₈/CH₄, the selectivity values of SNNU-185 and SNNU-186 at 298 K and 100 kPa are 132.5 and 126.0, respectively. These values are not top-level but still higher than those of many well-known MOF materials, such as MIL-101-Cr³⁰ (84.3), ZUL-C₁ (ref. 4) (73), UiO-66 (ref. 34) (65) and ECUT-Th-10a¹¹ (54.5) under the same conditions.

Considering that the relatively small window size of the nanotraps might influence the gas diffusion behaviour, kinetic mass transfer factors were investigated. The adsorption kinetics of C₂H₆ and C₃H₈ were evaluated using the time-dependent uptake profile. As shown in Fig. S16,† both C₂H₆ and C₃H₈ with similar slopes could achieve complete desorption within similar timeframes, indicating their similar diffusion behaviour, thus excluding their diffusion rate differences as a key factor in their sorption properties. Demonstrating the extent of exclusiveness of bi-nanotraps is crucial and the key is to prove that C₂H₆ and C₃H₈ do not affect each other during the separation process. Since the selectivity of “bi-nanotraps” results from both “thermodynamics” and the “molecular sieving” mechanism rather than thermodynamics alone, and the effectiveness of “bi-nanotraps” in weakening competitive adsorption can be demonstrated when C₂H₆ and C₃H₈ coexist, two-component breakthrough tests were performed to provide evidence for the “bi-nanotrap” effect (Fig. 3a, b and S17–19†). As shown in Fig. 3a, whether mixed with CH₄ or C₂H₆, the breakthrough time of C₃H₈ was not affected (∼115 min g⁻¹), implying that C₂H₆ does not affect the adsorption of C₃H₈. Moreover, whether mixed with CH₄ or C₃H₈, the breakthrough time of C₂H₆ was not affected (∼62 min g⁻¹, Fig. 3b), implying that C₃H₈ does not affect the adsorption of C₂H₆ as well. Therefore, once gases enter the “bi-nanotrap” structure, it is expected that C₃H₈ will be adsorbed in C₃H₈-selective nanotrap 1 and C₂H₆ will be adsorbed in C₂H₆-selective nanotrap 2. Clearly, “bi-nanotraps” play a crucial role in removing the competitive adsorption between C₂H₆ and C₃H₈, thus improving C₃H₈/C₂H₆/CH₄ separation performance.

Fig. 3 Breakthrough curves for (a) C₃H₈/CH₄ (50/50, v/v) and C₃H₈/C₂H₆ (50/50, v/v) of SNNU-186 at 298 K; (b) C₂H₆/CH₄ (50/50, v/v) and C₃H₈/C₂H₆ (50/50, v/v) of SNNU-186 at 298 K; (c) C₂H₆/CH₄ (20/80, v/v) and C₃H₈/CH₄ (20/80, v/v) of SNNU-185 and SNNU-186; (d) C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) of SNNU-185; (e) C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) of SNNU-186; (f) C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) of SNNU-186 with different total flow rates of 4 mL min⁻¹ and 6 mL min⁻¹. (g) CH₄ purity in the C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) outlet gas of SNNU-186. (h) Comparison of the separation performance among all reported MOFs used for C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) separation. (i) Comparison of breakthrough curves of SNNU-186 and mixed MOFs (SNNU-26-Co + SNNU-28-Co) for C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) with a total flow rate of 6 mL min⁻¹.

Inspired by the increased thermodynamic interaction difference and exclusive interaction sites, and encouraged by the satisfactory gas uptake and potential separation ability of activated SNNU-185/186, further experimental dynamic breakthrough experiments were performed to evaluate their C₃H₈/C₂H₆/CH₄ separation performance. As shown in Fig. 3c and S20,† CH₄ eluted out first due to its lowest adsorption capacity and weakest affinity with the frameworks, while C₂H₆ and C₃H₈ were trapped until their saturation sorption. For 20/80 C₂H₆/CH₄ and 20/80 C₃H₈/CH₄ mixtures with a total flow rate of 2 mL min⁻¹ at 298 K, C₂H₆/C₃H₈ was retained for additional 66.0/204.0 min g⁻¹ on SNNU-185, and 58.0/193.6 min g⁻¹ on SNNU-186. Considering the practical composition of natural gas, experimental breakthrough tests with a feed gas of ternary C₃H₈/C₂H₆/CH₄ (5/10/85, v/v/v) mixtures at flow rates of 4/6 mL min⁻¹ were carried out at 298 K. As shown in Fig. 3d–g and S21,† SNNU-185 and SNNU-186 can produce ultra-high purity CH₄ (>99.9999%) with exceptional productivities for CH₄. The CH₄ productivity of SNNU-185/186 was calculated to be 6.85/6.10 mmol g⁻¹, surpassing those of most top-performing MOFs such as SNNU-Bai₆₉ (ref. 3) (5.93 mmol g⁻¹), ZUL-C₁ (ref. 4) (5.42 mmol g⁻¹), BSF-1/2/3 (ref. 1, 14 and 38) (3.75/3.79/4.60 mmol g⁻¹), UiO-66-NaPh³⁴ (2.25 mmol g⁻¹), MIL-101-Cr³⁰ (2.66 mmol g⁻¹), and Zn-BPZ-SA⁵ (1.56 mmol g⁻¹); it is comparable to that of MOF-303 (ref. 57) (7.97 mmol g⁻¹), and is only lower than those of ZUL-C₂ (ref. 4) (1 mL min⁻¹, 11.4 mmol g⁻¹) and Ni(TMBDC)(DABCO)_0.5 (ref. 2) (4 mL min⁻¹, 12.6 mmol g⁻¹) (Table S6†). When the experimental breakthrough tests were performed at a high flow rate of 6 mL min⁻¹, the CH₄ purity still reached 99.9999%, which can be attributed to the multiple interactions between C₂H₆ and MOF frameworks, as well as the increased interaction difference between C₂H₆ and CH₄.

Furthermore, considering the importance of C₂H₆ and C₃H₈ recovery, the breakthrough capture capacities of SNNU-185 and SNNU-186 for C₂H₆ and C₃H₈ were calculated accordingly. SNNU-185/186 possess outstanding C₂H₆ and C₃H₈ capture capacities of 1.23/0.90 mmol g⁻¹ and 2.33/2.15 mmol g⁻¹, respectively, which are superior to those of most MOF materials and are comparable to those of top-level MOF materials such as ZUL-C₂ (ref. 4) (2.13/1.66 mmol g⁻¹), ZUL-C₁ (ref. 4) (0.98/1.19 mmol g⁻¹) and Ni-MOF 1 (ref. 50) (0.78/2.10 mmol g⁻¹) (Table S6†). As shown in Fig. 3h, when considering CH₄ productivity, breakthrough capture capacities for C₃H₈, and C₃H₈ uptake, SNNU-185 and SNNU-186 exhibit the best performance for CH₄ purification as well as for C₂H₆ and C₃H₈ recovery. Notably, the excellent separation performance of SNNU-185/186 is based on both “thermodynamics” and the “bi-nanotrap effect” in contrast to other MOFs that rely only on thermodynamics. As a result, although the thermodynamics-based IAST selectivities of SNNU-185/186 for C₂H₆/CH₄ are moderate, thanks to the guest-specific interactions, competitive adsorption between C₂H₆ and C₃H₈ is weakened and the practical separation performance is improved (Table S6†). Furthermore, considering the presence of CO₂ in raw natural gas and the challenges associated with its removal,²⁷ breakthrough experiments were conducted to provide an assessment of the impact of CO₂ contaminants. As shown in Fig. S22 and S23,† SNNU-186 could effectively separate C₂H₆/CO₂/CH₄ = 15/4/81 (v/v/v) and C₃H₈/CO₂/CH₄ = 4/4/92 (v/v/v) containing 4% CO₂. Overall, SNNU-185 and SNNU-186 are highly competitive candidates for natural gas upgrade.

Given that there are two types of pores in SNNU-185/186, comparative experiments were conducted to demonstrate the advantage of using a single material with two types of pores (∼5 Å and ∼7 Å) over using a mixture of two MOFs with one type of pore each. First, since many factors such as metal centers, open metal sites (OMSs), functional groups can strongly influence the adsorption behaviour of MOFs, it is necessary to ensure the same metal center (Co center) and a similar chemical environment (decorated with aromatic rings, N sites, no OMSs). Bearing the above factors in mind, two reported MOFs, SNNU-26-Co⁵⁸ (Co-BDC-TPP, with a pore size of ∼5 Å) and SNNU-28-Co⁵⁸ (Co-2,6-NDC-TPP, with a pore size of ∼7 Å) were selected (Table S7†). As shown in Fig. 3i and S24,† under the same conditions, SNNU-186 exhibited better practical separation performance, confirming that using one MOF with two types of pores is more favourable for the C₃H₈/C₂H₆/CH₄ separation process.

To give a mechanistic explanation of the role and effectiveness of the bi-nanotrap structure, and to gain a molecular-level insight into the host–guest interactions and adsorption behaviors of C₃H₈, C₂H₆ and CH₄, Grand Canonical Monte Carlo (GCMC) simulations were performed (Fig. 4 and S25†). As shown in Fig. 4a–c, the large-type nanotrap with a pore size of 7.2 Å can trap CH₄, C₂H₆ and C₃H₈via multiple C–H⋯π bonds with distances between 3.535 Å and 3.818 Å (3 bonds) for CH₄, 3.265 Å and 3.987 Å (10 bonds) for C₂H₆, and 3.232 Å and 3.964 Å (12 bonds) for C₃H₈. Thanks to the higher number of H atoms in C₃H₈ and better size matching, these large nanotraps are more favorable for C₃H₈, forming more and stronger C–H⋯π bonds, and are thus considered thermodynamic C₃H₈-selective nanotraps. As for the small-type nanotraps with a pore size of 4.8 Å, they do not allow C₃H₈ molecules to enter due to the pore size limitation (Fig. 4d–f and S25†). However, C₂H₆ molecules can enter and bind to aromatic rings of TPP ligands on the surface of the nanotraps via a large number of strong and shape-matching C–H⋯π interactions with short distances (2.702–3.957 Å, 12 bonds), implying the exceptionally strong interactions between C₂H₆ and frameworks as well as preferential adsorption selectivity for C₂H₆ (Fig. 4e). As a result, C₂H₆-selective nanotraps are successfully constructed based on the dual integrated thermodynamic-molecular sieving mechanism. CH₄ molecules interact with both kinds of nanotraps via fewer and weaker interactions (Fig. 4a and d). Clearly, the construction of thermodynamic C₃H₈-selective nanotraps and coupled thermodynamic-molecular sieving C₂H₆-selective nanotraps in the bi-nanotrap structure provides a reasonable explanation for the benchmark performance of SNNU-185 and SNNU-186 for C₃H₈/C₂H₆/CH₄ separation. When C₂H₆ and C₃H₈ molecules coexist, they tend to preferentially occupy different and size-matching sites to form multiple and strong interactions, thus leading to a performance breakthrough.

Fig. 4 GCMC simulated adsorption binding sites for (a) CH₄, (b) C₂H₆ and (c) C₃H₈ in nanotrap 1, (d) CH₄ and (e) C₂H₆ in nanotrap 2 and (f) the size limitation of nanotrap 2 for C₃H₈. The unit of interaction distance is angstrom (Å).

Conclusions

In summary, a promising aromatic ring-confined bi-nanotrap strategy for excellent natural gas upgrading has been demonstrated here. The perfectly size/shape-matched C₂H₆-selective nanotraps and C₃H₈-selective nanotraps enable C₂H₆ and C₃H₈ to be preferentially trapped via abundant and extra-strong C–H⋯π bonds. Such a combination of thermodynamic-based nanotraps with molecular sieving-based size exclusion enables multiple, powerful and shape-matched interactions, and selective interaction sites, which is unprecedented. As a result, the goal of increasing the thermodynamic difference and reducing competitive adsorption was achieved. With excellent thermal/chemical stability and satisfactory gas sorption properties, the two MOFs reported here can produce high purity CH₄ at high flow rates along with achieving first-class productivities for CH₄, C₂H₆ and C₃H₈. This work not only creates highly ideal adsorbents with benchmark practical performance for natural gas upgrading, but also introduces a design concept of installing selective bi-nanotraps and fully exploiting the integrated thermodynamic-molecular sieving mechanism for the development of high-performance absorbents for more challenging multi-component gas systems.

Data availability

All the associated data are available in the ESI.†

Author contributions

Q.-G. Z. and S.-Y. L. conceived the idea of this research. S.-Y. L. carried out the experiments, analyzed the results and wrote the manuscript. Q.-G. Z. led the project and edited the manuscript. X. B. edited the manuscript. All authors participated in and contributed to the preparation of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (22071140), the Natural Science Foundation of Shaanxi Province (2021JLM-20), the Youth Innovation Team of Shaanxi Universities (2023), and the Fundamental Research Funds for the Central Universities (GK202307009).

Notes and references

1. Y. Zhang, L. Yang, L. Wang, S. Duttwyler and H. Xing, Angew. Chem., Int. Ed., 2019, 58, 8145–8150.
2. Y. Wu, Z. Liu, J. Peng, X. Wang, X. Zhou and Z. Li, ACS Appl. Mater. Interfaces, 2020, 12, 51499–51505.
3. M. Ding, Q. Wang, H. Cheng and J. Bai, CrystEngComm, 2022, 24, 2388–2392.
4. J. Zhou, T. Ke, F. Steinke, N. Stock, Z. Zhang, Z. Bao, X. He, Q. Ren and Q. Yang, J. Am. Chem. Soc., 2022, 144, 14322–14329.
5. G.-D. Wang, R. Krishna, Y.-Z. Li, Y.-Y. Ma, L. Hou, Y.-Y. Wang and Z. Zhu, ACS Mater. Lett., 2023, 5, 1091–1099.
6. Y. He, W. Zhou, G. Qian and B. Chen, Chem. Soc. Rev., 2014, 43, 5657–5678.
7. J. Shen, A. Dailly and M. Beckner, Microporous Mesoporous Mater., 2016, 235, 170–177.
8. J. Li, X. Luo, N. Zhao, L. Zhang, Q. Huo and Y. Liu, Inorg. Chem., 2017, 56, 4141–4147.
9. Z. Zeng, W. Wang, X. Xiong, N. Zhu, Y. Xiong, Z. Wei and J.-J. Jiang, Inorg. Chem., 2021, 60, 8456–8460.
10. Z. Ke, H. Xiao, Y. Wen, S. Du, X. Zhou, J. Xiao and Z. Li, Ind. Eng. Chem. Res., 2021, 60, 4668–4676.
11. L. Wang, W. Zhang, J. Ding, L. Gong, R. Krishna, Y. Ran, L. Chen and F. Luo, Nano Res., 2023, 16, 3287–3293.
12. J.-R. Li, R. J. Kupplera and H.-C. Zhou, Chem. Soc. Rev., 2009, 38, 1477–1504.
13. K. Adil, Y. Belmabkhout, R. S. Pillai, A. Cadiau, P. M. Bhatt, A. H. Assen, G. Maurinb and M. Eddaoudi, Chem. Soc. Rev., 2017, 46, 3402–3430.
14. Y. Zhang, L. Yang, L. Wang, X. Cui and H. Xing, J. Mater. Chem. A, 2019, 7, 27560–27566.
15. B. Li, M. Chrzanowski, Y. Zhang and S. Ma, Coord. Chem. Rev., 2016, 307, 106–129.
16. X. Zhao, Y. Wang, D.-S. Li, X. Bu and P. Feng, Adv. Mater., 2018, 30, 1705189.
17. E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown and J. R. Long, Science, 2012, 335, 1606–1610.
18. S. Yang, A. J. Ramirez-Cuesta, R. Newby, V. Garcia-Sakai, P. Manuel, S. K. Callear, S. I. Campbell, C. C. Tang and M. Schröder, Nat. Chem., 2015, 7, 121–129.
19. Q. Liu, S. G. Cho, J. Hilliard, T.-Y. Wang, S.-C. Chien, L.-C. Lin, A. C. Co and C. R. Wade, Angew. Chem., Int. Ed., 2023, e202218854.
20. L. Wang, H. Huang, X. Zhang, H. Zhao, F. Li and Y. Gu, Coord. Chem. Rev., 2023, 484, 215111.
21. H.-C. Zhou, J. R. Long and O. M. Yaghi, Chem. Rev., 2012, 112, 673–674.
22. H.-C. Zhou and S. Kitagawa, Chem. Soc. Rev., 2014, 43, 5415–5418.
23. B. Li, H.-M. Wen, W. Zhou, J. Q. Xu and B. Chen, Chem, 2016, 1, 557–580.
24. S. Rupam and C. D. Madhab, Coord. Chem. Rev., 2021, 442, 213998.
25. X. Han and S. Yang, Angew. Chem., Int. Ed., 2023, e202218274.
26. H. Wang, D. Luo, E. Velasco, L. Yu and J. Li, J. Mater. Chem. A, 2021, 9, 20874–20896.
27. P.-Q. Liao, W.-X. Zhang, J.-P. Zhang and X.-M. Chen, Nat. Commun., 2015, 6, 8697.
28. L. Li, R.-B. Lin, R. Krishna, H. Li, S. Xiang, H. Wu, J. Li, W. Zhou and B. Chen, Science, 2018, 362, 443–446.
29. H. Zeng, X.-J. Xie, M. Xie, Y.-L. Huang, D. Luo, T. Wang, Y. Zhao, W. Lu and D. Li, J. Am. Chem. Soc., 2019, 141, 20390–20396.
30. L.-Z. Qin, X.-H. Xiong, S.-H. Wang, L. Zhang, L.-L. Meng, L. Yan, Y.-N. Fan, T.-A. Yan, D.-H. Liu, Z.-W. Wei and C.-Y. Su, ACS Appl. Mater. Interfaces, 2022, 14, 45444–45450.
31. R.-B. Lin, H. Wu, L. Li, X.-L. Tang, Z. Li, J. Gao, H. Cui, W. Zhou and B. Chen, J. Am. Chem. Soc., 2018, 140, 12940–12946.
32. J. Pei, J.-X. Wang, K. Shao, Y. Yang, Y. Cui, H. Wu, W. Zhou, B. Li and G. Qian, J. Mater. Chem. A, 2020, 8, 3613–3620.
33. Y. Ye, Y. Xie, Y. Shi, L. Gong, J. Phipps, A. M. Al-Enizi, A. Nafady, B. Chen and S. Ma, Angew. Chem., Int. Ed., 2023, e202302564.
34. L. Zhang, X.-H. Xiong, L.-L. Meng, L.-Z. Qin, C.-X. Chen, Z.-W. Wei and C.-Y. Su, J. Mater. Chem. A, 2023, 11, 12902–12909.
35. J. Peng, J. Zhong, Z. Liu, H. Xi, J. Yan, F. Xu, X. Chen, X. Wang, D. Lv and Z. Li, ACS Appl. Mater. Interfaces, 2023, 15, 41466–41475.
36. Q. Dong, Y. Huang, K. Hyeon-Deuk, I.-Y. Chang, J. Wan, C. Chen, J. Duan, W. Jin and S. Kitagawa, Adv. Funct. Mater., 2022, 32, 2203745.
37. M. Wriedt, J. P. Sculley, A. A. Yakovenko, Y. Ma, G. J. Halder, P. B. Balbuena and H.-C. Zhou, Angew. Chem., Int. Ed., 2012, 51, 9804–9808.
38. L. Wang, W. Sun, S. Duttwyler and Y. Zhang, J. Solid State Chem., 2021, 299, 122167.
39. L. Yang, X. Cui, Q. Yang, S. Qian, H. Wu, Z. Bao, Z. Zhang, Q. Ren, W. Zhou, B. Chen and H. Xing, Adv. Mater., 2018, 30, 1705374.
40. O. T. Qazvini, R. Babarao, Z.-L. Shi, Y.-B. Zhang and S. G. Telfer, J. Am. Chem. Soc., 2019, 141, 5014–5020.
41. Z. Niu, X. Cui, T. Pham, P. C. Lan, H. Xing, K. A. Forrest, L. Wojtas, B. Space and S. Ma, Angew. Chem., Int. Ed., 2019, 58, 10138–10141.
42. Y.-Y. Xue, S.-N. Li, Y.-C. Jiang, M.-C. Hu and Q.-G. Zhai, J. Mater. Chem. A, 2019, 7, 4640.
43. Z. Niu, X. Cui, T. Pham, G. Verma, P. C. Lan, C. Shan, H. Xing, K. A. Forrest, S. Suepaul, B. Space, A. Nafady, A. M. Al-Enizi and S. Ma, Angew. Chem., Int. Ed., 2021, 60, 5283–5288.
44. Y. Ye, S. Xian, H. Cui, K. Tan, L. Gong, B. Liang, T. Pham, H. Pandey, R. Krishna, P. C. Lan, K. A. Forrest, B. Space, T. Thonhauser, J. Li and S. Ma, J. Am. Chem. Soc., 2022, 144, 1681–1689.
45. H. Zhu, Y. Wang, X. Wang, Z.-W. Fan, H.-F. Wang, Z. Niu and J.-P. Lang, Chem. Commun., 2023, 59, 5757.
46. Y. Wang, N.-Y. Huang, X.-W. Zhang, H. He, R.-K. Huang, Z.-M. Ye, Y. Li, D.-D. Zhou, P.-Q. Liao, X.-M. Chen and J.-P. Zhang, Angew. Chem., Int. Ed., 2019, 58, 7692–7696.
47. L. Yang, S. Qian, X. Wang, X. Cui, B. Chen and H. Xing, Chem. Soc. Rev., 2020, 49, 5359–5406.
48. F. Zheng, R. Chen, Z. Zhang, Q. Yang, Y. Yang, Q. Ren and Z. Bao, Cell Rep. Phys. Sci., 2022, 3, 100903.
49. Y.-Y. Xue, S.-N. Li, Y.-C. Jiang, M.-C. Hu and Q.-G. Zhai, J. Mater. Chem. A, 2019, 7, 4640.
50. X.-X. Zhang, X.-Z. Guo, S.-S. Chen, H.-W. Kang, Y. Zhao, J.-X. Gao, G.-Z. Xiong and L. Hou, Chem. Eng. J., 2023, 466, 143170.
51. A. Cadiau, K. Adil, P. M. Bhatt, Y. Belmabkhout and M. Eddaoudi, Science, 2016, 353, 137–140.
52. Y. Cheng, B. Joarder, S. J. Datta, N. Alsadun, D. Poloneeva, D. Fan, R. Khairova, A. Bavykina, J. Jia, O. Shekhah, A. Shkurenko, G. Maurin, J. Gascon and M. Eddaoudi, Adv. Mater., 2023, 2300296.
53. H. Wang, X. Dong, V. Colombo, Q. Wang, Y. Liu, W. Liu, X.-L. Wang, X.-Y. Huang, D. M. Proserpio, A. Sironi, Y. Han and J. Li, Adv. Mater., 2018, 30, 1805088.
54. B. Liang, X. Zhang, Y. Xie, R.-B. Lin, R. Krishna, H. Cui, Z. Li, Y. Shi, H. Wu, W. Zhou and B. Chen, J. Am. Chem. Soc., 2020, 142, 17795–17801.
55. H. Zeng, M. Xie, T. Wang, R.-J. Wei, X.-J. Xie, Y. Zhao, W. Lu and D. Li, Nature, 2021, 595, 542–548.
56. Q. Dong, Y. Huang, J. Wan, Z. Lu, Z. Wang, C. Gu, J. Duan and J. Bai, J. Am. Chem. Soc., 2023, 145, 8043–8051.
57. S. Xian, J. Peng, H. Pandey, T. Thonhauser, H. Wang and J. Li, Engineering, 2023, 23, 56–63.
58. Y.-Y. Xue, X.-Y. Bai, J. Zhang, Y. Wang, S.-N. Li, Y.-C. Jiang, M.-C. Hu and Q.-G. Zhai, Angew. Chem., Int. Ed., 2021, 60, 10122–10128.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2268660 and 2268661. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4sc04387a

---