Purification of Propylene and Ethylene by a Robust Metal–Organic Framework Mediated by Host–Guest Interactions

Abstract Industrial purification of propylene and ethylene requires cryogenic distillation and selective hydrogenation over palladium catalysts to remove propane, ethane and/or trace amounts of acetylene. Here, we report the excellent separation of equimolar mixtures of propylene/propane and ethylene/ethane, and of a 1/100 mixture of acetylene/ethylene by a highly robust microporous material, MFM‐520, under dynamic conditions. In situ synchrotron single crystal X‐ray diffraction, inelastic neutron scattering and analysis of adsorption thermodynamic parameters reveal that a series of synergistic host–guest interactions involving hydrogen bonding and π⋅⋅⋅π stacking interactions underpin the cooperative binding of alkenes within the pore. Notably, the optimal pore geometry of the material enables selective accommodation of acetylene. The practical potential of this porous material has been demonstrated by fabricating mixed‐matrix membranes comprising MFM‐520, Matrimid and PIM‐1, and these exhibit not only a high permeability for propylene (≈1984 Barrer), but also a separation factor of 7.8 for an equimolar mixture of propylene/propane at 298 K.


S2
temperatures were controlled by system programmed water bath. Before measurement of the isotherm, the as-synthesized MFM-520 was activated at 120 °C under dynamic high vacuum (10 -10 bar measured at the pump) for 24 h to give the fully desolvated MFM-520. C2H4, C2H6, C3H6, and C3H8 were ultra-pure research-grade (99.999%) purchased from BOC. C2H2 was purified through an activated carbon filter before introduction to the IGA system.

Gas Separation Breakthrough Experiments
Breakthrough experiments were performed on a Hiden Isochema IGA-003 with ABR attachments in combination with a Hiden Analytical mass spectrometer and FTIR spectrometer to detect the gases as they are released from the sample bed. The temperature was controlled by a temperatureprogrammed system. Breakthrough experiments were carried out in a 7 mm diameter fixed-bed of 120 mm length packed with ~1.0 g of MFM-520 powder. The sample was heated at 120 ºC under a flow of He for 16h for activation. The fixed-bed was then cooled to 318 K using a temperature programmed water bath and the breakthrough experiment was performed with a stream of hydrocarbon gases at atmospheric pressure and 318 K. The flow rate of the entering gas mixture was maintained at 4-6 mL min -1 , and the gas concentration, C, of hydrocarbons at the outlet was determined by mass spectrometry and compared with the corresponding inlet concentration C0, where C/C0 = 1 indicates complete breakthrough.

Inelastic Neutron Scattering
Inelastic neutron scattering (INS) experiments were undertaken using the TOSCA spectrometer at the ISIS facility. TOSCA is an indirect geometry crystal analyser instrument that provides a wide dynamic range (-26-4000 cm −1 ) with resolution optimised in the 50-2000 cm −1 range. In this region TOSCA has a resolution of 1.25% of the energy transfer. The instrument is comprised of 130 3 He detectors in the forward and backscattering geometry located 17 m downstream of a 300 K Gd poisoned water moderator. A temperature of 7 ± 0.2 K was maintained during data collection by two He closed cycle refrigerators with 30 mbar He as an exchange gas.
MFM-520 was loaded into an 11 mm diameter vanadium sample can and degassed at 10 -7 mbar and 120 °C for 1 day. The loading of gases was performed at room temperature and the sample can loaded into a closed cycle refrigerator (CCR) He cryostat and cooled to (7 ± 0.1 K) for data collection. C2H2, C2H4, and C2H6 were introduced by warming the sample to 290 K, and the gas dosed volumetrically from a calibrated volume. The gas-loaded sample was then cooled to 7 K over a period of 2 h to ensure good mobility of adsorbed gases within the crystalline structure of MFM-520. The sample was kept at 7 K for an additional 30 mins before data collection to ensure the thermal equilibrium.

DFT Modelling and Simulation
Modelling by Density Functional Theory (DFT) of the bare and C2H2/C2H4/C2H6-loaded MFM-520 was performed using the Vienna Ab initio Simulation Package (VASP) [2] . The calculation used the Projector Augmented Wave (PAW) method [3,4] to describe the effects of core electrons, and Perdew-Burke-Ernzerhof (PBE) [5] implementation of the Generalized Gradient Approximation (GGA) for the exchange-correlation functional. Energy cutoff was 800 eV for the plane-wave basis of the valence electrons. The lattice parameters and atomic coordinates determined by synchrotron X-ray single crystal diffraction in this work were used as the initial structure. The electronic structure was calculated on a 4×4×2 Monkhorst-Pack mesh for the unit cell, and at the Γ point only for the 2×2×1 supercell. The total energy tolerance for electronic energy minimization was 10 -8 eV, and for structure optimization it was 10 -7 eV. The maximum interatomic force after relaxation was below 0.001 eV/Å. The optB86b-vdW functional [6] for dispersion corrections was applied. The vibrational eigen-frequencies and modes were then calculated by solving the force constants and dynamical matrix using Phonopy [7] . The OClimax software [8] was used to convert the DFT-calculated phonon results to the simulated INS spectra.

Fabrication of MMM Membrane
A solution-casting method was employed for the fabrication of the mixed matrix membrane (MMM) of PIM-1/Matrimid and PIM-1/Matrimid/MFM-520. A powder of Matrimid (600 mg) was dispersed in CHCl3 (60 mL) with sonication and stirring for 12 h, and this was followed by addition of PIM-1 (600 mg) and stirring for another 12 h. When a homogenous solution was obtained, MFM-520 (60 mg) as ground powder was added and stirred for 12 h. The mixture was evaporated to 15-30 mL and cast evenly onto three PTFE substrates and the solvent evaporated at room temperature. The freestanding MMM was dried in a vacuum oven at 120 °C for 12 h to remove trace solvents from membrane prior to the separation measurements. The PIM-1/Matrimid membrane was fabricated using the same method without adding any MOF. The Matrimid was used as received from commercial suppliers without purification and the PIM-1 was synthesised following the reported method [9] . Synthesized PIM-1: GPC (in chloroform): Mw = 38,026 g mol -1 , and Mw/Mn = 2.309. 1 (4.24); N, 6.08 (6.13).

Gas Separation by Mixed Matrix Membrane (MMM)
Gas permeability was evaluated using the constant-pressure method with a Bruker Matrix MG5 infrared spectrometer as gas detector. The membranes of PIM-1/Matrimid and PIM-1/Matrimid/MFM-520 were activated at 120 C under vacuum for 24 h. The permeabilities of an equimolar mixture of C3H6/C3H8 were measured at 1.5 bar and 298 K. All measurements were repeated three times to allow the standard errors to be estimated, and the membrane was activated by flushing with dry He between measurements to ensure the integrity of the membrane. A circular membrane of diameter 5.0 cm was placed in the cell. The thickness of the membrane was measured using a micrometre and multiple measurements from different sites on the membrane taken and averaged to determine the thickness. The average thickness of the PIM-1/Matrimid and PIM-1/Matrimid/MFM-520 membranes were determined to be 44 and 48 µm, respectively. The permeability was calculated by (1): where P is the permeability coefficient in Barrer, j is the flux of the gas in 10 -10 cm 3 (STP)/(cm 2 ·s), l is the thickness of the membrane in cm, and ∆p is the pressure difference between the feed side and the permeate side in cmHg -1 .The selectivity for C3H6/C3H8 was calculated by equation (2) [10] :

In Situ Synchrotron Single Crystal X-Ray Diffraction
In situ synchrotron X-ray single-crystal diffraction data were collected at beamline 11.3.1 and 12.2.1 of the Advanced Light Source in Berkeley using monochromated radiation [λ = 0.7749 and 0.7288 Å]. These in situ diffraction measurements were carried out in a 50 micro quartz tube coupled to gas handling and vacuum equipment. In a typical experiment, a single crystal of synthesized MFM-520 was selected and glued onto a MiTeGen loop with Loctite Double Bubble epoxy, and the quartz tube loaded. The sample system was connected to a high vacuum (10 -4 mbar) and heated to 420 K for ~3 h to generate desolvated MFM-520. After cooling the temperature to 273 K, the X-ray data for desolvated MFM-520 were collected as a background. No residual electron density was found in the pores of desolvated MFM-520. Upon loading of hydrocarbon gases into MFM-520 at 1 bar and 273 K for 1 h, the data were re-collected. Significant residual electron densities appeared in the pore and were sequentially assigned as adsorbed hydrocarbon molecules. Hydrogen atoms were placed and refined using a riding model, and the crystallographic data are summarized in Table S2.

Calculation of Isosteric Heats of Adsorption
To estimate the differential enthalpies (ΔHn) and (ΔSn) for C2H2, C2H4, C2H6, C3H6 and C3H8 adsorption, the isotherms at different temperatures were fitted to the van't Hoff isochore: where P is pressure, T is the temperature, R is the real gas constant. Selected linear fitting plots at 1.2, 1.3, 1.4, 1.5 mmol/g are shown in Figs S7, S9, S11, S13 and S15. All linear fittings show R 2 above 0.9, indicating consistency in the isotherm data. A plot of ln(p) versus 1/T at constant amount adsorbed allows the differential enthalpy and entropy of adsorption and also the isosteric enthalpy of adsorption (Qst, n) to be determined.

S11
Summary of Crystallographic Data for Hydrocarbon-loaded MFM-520         PXRD Analysis of PIM-1/Matrimid/MFM-520 (w/w/w =10: 10: 1) MMM PXRD analysis of the membrane was conducted on a Philips X'Pert XRD using Cu-Kα radiation to confirm the retention of the crystal structure of the MOF after incorporation into the MMM. Analysis of cross-sectioned membrane was conducted using scanning electron microscopy (SEM). The MMM was cross-sectioned mechanically and adhered to the sample stub via a carbon tab. The deformed region highlighted by the orange square in Fig. S17a contains surface dips from mechanical damage induced during cross-sectioning. The undeformed region, as highlighted via the blue square in Fig. S17a, is homogeneous in texture. Fig. S17b shows a magnified image of this undeformed area. Even in the deformed MMM region seen in Fig. S17c, unperturbed areas, such as the green square highlighted and magnified via Fig. S17d, show further homogeneous textures within the MMM. The images were acquired at 15 kV with spotsize 3.5 on the Quanta 650 FEG SEM at magnifications ranging from 1000X to 26,000X.