Microporous Metal–Organic Framework with Dual Functionalities for Efficient Separation of Acetylene from Light Hydrocarbon Mixtures

Abstract

Separating acetylene from light hydrocarbon mixtures like ethylene is a very important process for downstream industrial applications. Herein, we report a new MOF [CuL₂(SiF₆)] (UTSA-220, L = (1E,2E)-1,2-bis(pyridin-4-ylmethylene)hydrazine) with dual functionalities featuring optimal pore size with strong binding sites for acetylene. UTSA-220 exhibits apparently higher uptake capacity for C₂H₂ than those for other light hydrocarbons. The potential of this material for trace C₂H₂ removal from C₂H₄ has also been demonstrated by a dynamic breakthrough experiment performed with C₂H₂/C₂H₄ (1/99 v/v) under simulated industrial conditions. According to the dispersion-corrected density functional theory (DFT-D) simulation, SiF₆²⁻ and azine moieties serve as preferential binding sites for C₂H₂, indicating the feasibility of the dual functionalities incorporated in UTSA-220 for adsorbent-based C₂H₂ separations.

Graphical Abstract

INTRODUCTION

The separation of light hydrocarbons is widely regarded as important processes in petrochemistry.^1,2 Among them, acetylene (C₂H₂) is the source of many organic chemicals in industry, such as α-ethynyl alcohols, vinyl compounds, and acrylic acid derivatives.³ C₂H₂ usually coexists with carbon dioxide (CO₂), ethylene (C₂H₄) or other light hydrocarbons, because C₂H₂ is mainly manufactured by cracking of hydrocarbons or partial combustion of methane (CH₄). Apart from this, C₂H₂ is a common impurity in the production of olefins like C₂H₄. The presence of C₂H₂ in C₂H₄ largely affects the downstream C₂H₄ polymerization reaction by poisoning the catalyst.² Thus, the significance of C₂H₂/C₂H₄ separation is comparable to that of some important separations such as C₂H₄/C₂H₆, C₃H₄/C₃H₆ and C₃H₆/C₃H₈.^4–8 Considerable difficulty has been encountered in the removal of C₂H₂ from the coexisting species for industrial requirements, because these small molecules are similar in physical properties such as molecular sizes, electronic structures, and volatilities.⁹ Traditional cryogenic distillation method suffers from the extraordinarily high energy and capital input, which propels us to develop alternative approaches to addressing the issue more efficiently.¹ Separation by means of adsorption with porous materials is perceived as a promising way to replace the traditional distillation method and has experienced significant development in recent years.^{4–8,10–13}

As a new generation of porous materials, metal–organic frameworks (MOFs) have attracted vast attention in exploring their application potential in various fields,^14–21 including but not limited to gas storage,^22–25 separation,^6,26–34 catalysis,^35–38 sensing,^24,39,40 and drug delivery.^41,42 The diverse properties and functions of MOFs stem from their highly modifiable pore structure and surface.^43,44 With rational choice of the building blocks (metal ions/clusters and organic ligands), the topology, pore size, surface functionality of MOFs can be tailored specifically to fit into application scenarios like gas separation.^45,46 For C₂H₂ removal and enrichment, the control over pore size and pore functionalities are found effective in developing MOFs with desired performance.^8,47,48

Recently, a series of MOFs containing SiF₆²⁻ moieties have been reported to have excellent abilities to separate C₂H₂ from C₂H₄ as a combined outcome of the molecular sieving effect and strong hydrogen bonding interactions between SiF₆²⁻ and C₂H₂ molecules.^8,49 Herein, we report a new MOF [CuL₂(SiF₆)] (termed UTSA-220) based on an azine based ligand L (L = (1E,2E)-1,2-bis(pyridin-4-ylmethylene)-hydrazine), featuring a bifunctionalized environment for C₂H₂ accommodation in the framework. This MOF shows a three-dimensional (3D) network with one-dimensional (1D) channels that provide optimal pore size and strong binding sites for C₂H₂. Gas sorption studies reveal that UTSA-220 selectively adsorbs C₂H₂ over other light hydrocarbons at ambient temperature. A dynamic breakthrough experiment performed with a C₂H₂/C₂H₄ (1/99 v/v) mixture further demonstrates UTSA-220 is feasible to separate C₂H₂ from C₂H₄ under practical conditions.

EXPERIMENTAL SECTION

Synthesis of UTSA-220.

Attempts to obtain single crystals were unsuccessful. Thus, bulk powder of this MOF was synthesized as follows. A methanolic solution of copper hexafluorosilicate was made by dissolving 22.4 mg (0.1 mmol) CuSiF₆·xH₂O into 10 mL methanol. The CuSiF₆ solution was added dropwise with constant stirring into a methanolic ligand solution prepared by dissolving 42 mg (0.2 mmol) L into 10 mL methanol. The suspension was stirred at room temperature for 30 min before the powder product was separated from the solution and washed with methanol for 3 times with a centrifuge. The product was transferred to a gas sorption tube and activated under vacuum at room temperature for 24 h before gas sorption measurements.

Breakthrough Experiment.

The breakthrough experiments were conducted on a self-built instrument (see Supporting Information) with a gas mixture of C₂H₂/C₂H₄ (1/99 v/v) at room temperature (298 K) and 1 bar. The MOF solid was packed into a ϕ2 × 70 mm stainless-steel column under the protection of N₂ gas in a glovebox. The packed column was flushed with helium gas at a rate of 40 mL/min for 2 h at room temperature to further activate the sample prior to measurements. The flow rate of the C₂H₂/C₂H₄ mixture was set at 2 mL/min and was first directed to a blank column to stabilize the gas flow before being switched to the adsorbent column to initiate the breakthrough experiment. The effluent composition was evaluated by a gas chromatography (GC) with a thermal conductivity detector (detection limit 0.1 ppm). Between two tests, the adsorbent was regenerated by helium flow (40 mL/min) for 12 h to guarantee a complete removal of the adsorbed gases. The desorption test on breakthrough instrument was performed at room temperature by switching the feed gas to He gas. The He flow was maintained at 20 mL/min. The effluent composition was measured by GC until no detectable C₂H₂ and C₂H₄ was found from the effluent.

RESULTS AND DISCUSSION

Structure and Synthesis.

The dropwise addition of a methanolic solution of L ligand into a CuSiF₆·xH₂O methanolic solution at room temperature with constant stirring affords the light purple crystalline powder of UTSA-220. Based on its powder X-ray diffraction pattern, the rough structure of UTSA-220 was determined to crystallize in the space group of C2/m and has one Cu atom, in which two L ligands and one SiF₆ group were found in the asymmetric unit (Figure S1 and Table S1). In this MOF, Cu atom is coordinated by four N atoms equatorially and two F atoms axially (Figure 1a–c). The four N donors come from four bidentate L ligands while the two F atoms are from two bridging SiF₆²⁻ anions. The obtained structure is a 2-fold interpenetrated 3D network (Figure S2). Since the organic linker L is not parallel but tilted with respect to the axial direction of coordinated Cu²⁺ ion, UTSA-220 features two kinds of 1D channels along the a-axis (Figure 1b,d). Owing to interpenetration, the sizes of two channel windows are narrowed to 3.0 × 3.2 Ų and 4.0 × 6.5 Ų (Figure 1d), with the width of the larger channel between 4.5 and 5.5 Å and that of smaller one 3.1–4.8 Å (Figure 1e,f). The total guest-accessible volume for UTSA-220 is estimated to be 40%. The thermogravimetric analysis of the as-synthesized sample shows that the solvent molecules can be easily removed around ambient temperature, and framework decomposition is at around 250 °C (Figure S3).

Figure 1.

Building units (a) and structure (b) of UTSA-220. The channels of UTSA-220 are viewed along the a-axis. (c) The coordination sphere of Cu atom in UTSA-220. (d) Illustration of the channel dimensions of UTSA-220 viewed along the a-axis. Cross section of the larger (e) and smaller (f) channels viewed along b-axis. Color code: Cu, turquoise; F, bright green; Si, orange; C, gray; N, blue; H, white.

Gas Sorption Properties.

To investigate the porosity and channels in UTSA-220, the N₂ sorption isotherms at 77 K and CO₂ sorption at 195 K were collected. The obtained N₂ sorption curve is a typical type I isotherm with the Brunauer–Emmett–Teller (BET) and Langmuir surface areas as 577 and 825 m²/g, respectively. With the aid of nonlocal density functional theory (NLDFT), the pore size distribution diagram was also obtained from the N₂ adsorption data, showing a narrow distribution of approximately 8.8 Å (Figure 2, inset). Notably, the measured pore volume of UTSA-220 is 0.35 cm³/g based on 77 K N₂ isotherm (at P/P₀ = 0.85), which is consistent with the theoretical value of 0.33 cm³/g from crystal structure, implying the well retention of its porosity. In contrast, the zinc analogue of UTSA-220 shows far lower measured pore volume,⁵⁰ which might be attributed to its high humidity sensitivity.

Figure 2.

Gas sorption isotherms of UTSA-220. N₂ at 77 K (violet) and CO₂ at 195 K (olive). The inset graph shows its incremental pore size distribution obtained from N₂ isotherms at 77 K.

Because of the suitable pore size and favorably functionalized channels, UTSA-220 is suitable for selective adsorption of C₂H₂ over other light hydrocarbons. Gas sorption tests were performed on UTSA-220 at 298 K (Figure 3a) and 273 K (Figure S4) with C₂H₂, CO₂, C₂H₄, C₂H₆, CH₄ and N₂. UTSA-220 show distinct adsorption amount of these gases. At 298 K and 100 kPa (1 bar), C₂H₂ has the highest uptake (3.40 mmol/g) among the series, followed by CO₂ (3.38 mmol/g), C₂H₄ (2.53 mmol/g), C₂H₆ (2.14 mmol/g), CH₄ (0.59 mmol/g) and N₂ (0.18 mmol/g) (Figure 3a). At 273 K, the same sequence is also observed in corresponding gas uptakes (Figure S4). Dual-site Langmuir isotherm model was employed to fit the two sets of data at different temperatures to calculate the isosteric heat of adsorption (denoted as −Q_st) of these gases (Table S2). At zero-coverage, the −Q_st for C₂H₂, CO₂, C₂H₄ and C₂H₆ are calculated to be 29, 27, 24 and 28 kJ/mol, respectively (Figure 3b). It should be noted that UTSA-220 adsorbs C₂H₂ rapidly at low pressure, with C₂H₂ uptake at 15 kPa reaching 2.40 mmol/g at 298 K. By contrast, the uptakes of CO₂ (1.56 mmol/g), C₂H₄ (1.22 mmol/g), C₂H₆ (1.10 mmol/g) and CH₄ (0.12 mmol/g) at the same condition are considerably lower. Such pronounced difference in the uptakes between C₂H₂ and other gas species reveals that UTSA-220 is very promising for C₂H₂ related separation. To assess the separation ability of UTSA-220, the ideal adsorbed solution theory (IAST) selectivity was calculated on the C₂H₂/C₂H₄ (1/99 v/v) and C₂H₂/CH₄ (1/99 v/v) mixtures (Figure 3c, Figure S5). The C₂H₂/C₂H₄ (1/99 v/v) adsorption selectivity at 100 kPa is 10, suggesting that UTSA-220 has the potential to effectively remove trace C₂H₂ from C₂H₄ to yield high quality monomer for polyethylene production. More impressively, the C₂H₂/CH₄ (1/99 v/v) selectivity is as high as 358 at 100 kPa (Table 1). We also calculate the equimolar IAST selectivity for C₂H₂/C₂H₄, C₂H₂/CO₂ and C₂H₂/C₂H₆ to explore the potential to perform C₂H₂ separation from these gas mixtures. Based on the results, UTSA-220 has moderately high selectivities in these separation scenarios, with the selectivity values as 8.8 (C₂H₂/C₂H₄), 4.4 (C₂H₂/CO₂) and 14 (C₂H₂/C₂H₆), respectively (Figure 3d, Table 1).

Figure 3.

(a) C₂H₂, CO₂, C₂H₄, C₂H₆, CH₄ and N₂ adsorption isotherms of UTSA-220 at 298 K. (b) Isosteric heat of adsorption of C₂H₂, CO₂, C₂H₄, C₂H₆ in UTSA-220. (c) IAST selectivity of C₂H₂/CH₄ (1/99 v/v) and C₂H₂/C₂H₄ (1/99 v/v) at 298 K in UTSA-220. (d) IAST selectivity of C₂H₂/C₂H₆ (50/50 v/v), C₂H₂/C₂H₄ (50/50 v/v) and C₂H₂/CO₂ (50/50 v/v) at 298 K in UTSA-220.

Table 1.

Summary of the IAST Selectivities of C₂H₂/C₂H₄ (1/99 v/v), C₂H₂/CH₄ (1/99 v/v), C₂H₂/C₂H₄ (50/50 v/v), C₂H₂/CO₂ (50/50 v/v) and C₂H₂/C₂H₆ (50/50 v/v) in UTSA-220 at 100 kPa

mixture	component proportion	IAST selectivity
C2H2/C2H4	1/99	10
C2H2/CH4	1/99	358
C2H2/C2H4	50/50	8.8
C2H2/CO2	50/50	4.4
C2H2/C2H6	50/50	14

To delve into the adsorption mechanism of C₂H₂ and preferential C₂H₂ adsorption sites in UTSA-220, we carried out computational investigations using dispersion-corrected density functional theory (DFT-D). Two C₂H₂ adsorption sites are found within the framework, which are located at two different channels, respectively (Figure 4a). In Site I, C₂H₂ molecule is oriented almost parallel to the c-axis, with both hydrogen atoms pointing at two F atoms from two SiF₆²⁻ units. On both ends of the C₂H₂ molecule, the H···F distances are 1.994 Å (Figure 4b), suggesting C₂H₂ forms hydrogen bonds with the surrounding F atoms. In Site II, C₂H₂ molecule is almost parallel with the ab plane, and is in an orientation with both H in proximity to two F atoms. Since H···F distances are 2.420 Å (Figure 4c), the C₂H₂ molecule in Site II also has hydrogen bonds formed. Each C₂H₂ molecule interacts with two F atoms from SiF₆²⁻ units located on opposite sides of the channel simultaneously. The calculated static binding energies for two sites are −48.5 and −45.8 kJ/mol, respectively, which accounts for the highly selective adsorption of C₂H₂ over other light hydrocarbons. The adsorption of C₂H₂ follows the similar mechanism as those in previously reported SIFSIX MOFs,^8,49,51 which highly relies on the optimal pore structure that matches well with the molecular shape of C₂H₂. Moreover, the simulation at ambient pressure reveals that there exists an extra interaction between C₂H₂ and azine groups with modest strength in UTSA-220 (Figure S6). These simulation results serve as a good evidence for our rational design of the framework to achieve desired performance.

Figure 4.

(a) DFT-D simulated adsorption sites for C₂H₂ in UTSA-220 and (b, c) their close contacts with the framework. Atom color code: F, bright green; Si, orange; Cu, turquoise; C in framework, gray; C in Site I, violet; C in Site II, golden; H, white; and N, blue.

Breakthrough Experiment.

To evaluate whether the material is capable to separate C₂H₂/C₂H₄ mixture in simulated industrial conditions, dynamic breakthrough experiments were conducted on UTSA-220 with a mixed gas of C₂H₂/C₂H₄ (1/99 v/v) using a homemade apparatus (Figure S7) at a flow rate of 2 mL/min at 298 K and 1 bar. C₂H₄ was first monitored at the column outlet after a short period of time since the start of the experiment, while C₂H₂ was still fully captured by the adsorbent and retained in the column (Figure 5). At this stage, a polymer-grade pure C₂H₄ was generated from the column with no measurable C₂H₂ (<0.1 ppm), indicating UTSA-220 is effective to remove trace amount of C₂H₂ from C₂H₄. As the adsorbent was close to saturation, C₂H₂ emerged from the column and gradually increased in concentration until all C₂H₂ in the gas mixture can be probed. The retention time for C₂H₂ and C₂H₄ are 48 min/g and 10 min/g, respectively. In order to test whether UTSA-220 can be utilized as a reusable adsorbent, the column was successively regenerated using a He flow at a rate of 20 mL/min at room temperature and put to breakthrough experiments for four times (Figures S8 and S9). Each time, breakthrough time is around 48 min/g, suggesting the recyclable feature of UTSA-220.

Figure 5.

Experimental column breakthrough curves of UTSA-220 material for C₂H₂/C₂H₄ (1/99 v/v) separation at 298 K and 1 bar.

CONCLUSION

To sum up, a new MOF with dual functionalities was successfully synthesized based on an azine ligand (1E,2E)-1,2-bis(pyridin-4-ylmethylene)hydrazine for efficient C₂H₂ removal and purification. Thanks to the dual functionalities, this MOF exhibits selective adsorption of C₂H₂ over other light hydrocarbons like C₂H₄. Breakthrough experiments further demonstrate the feasibility to remove trace amount of C₂H₂ from a mixture of C₂H₂ and C₂H₄ with this MOF as the column adsorbent. This work affords a functionalized approach of novel porous materials to advancing important gas separations in relevant energy-intensive petrochemical processes.