Direct Conversion of Methane to Ethylene and Acetylene over an Iron-Based Metal–Organic Framework

Conversion of methane (CH4) to ethylene (C2H4) and/or acetylene (C2H2) enables routes to a wide range of products directly from natural gas. However, high reaction temperatures and pressures are often required to activate and convert CH4 controllably, and separating C2+ products from unreacted CH4 can be challenging. Here, we report the direct conversion of CH4 to C2H4 and C2H2 driven by non-thermal plasma under ambient (25 °C and 1 atm) and flow conditions over a metal–organic framework material, MFM-300(Fe). The selectivity for the formation of C2H4 and C2H2 reaches 96% with a high time yield of 334 μmol gcat–1 h–1. At a conversion of 10%, the selectivity to C2+ hydrocarbons and time yield exceed 98% and 2056 μmol gcat–1 h–1, respectively, representing a new benchmark for conversion of CH4. In situ neutron powder diffraction, inelastic neutron scattering and solid-state nuclear magnetic resonance, electron paramagnetic resonance (EPR), and diffuse reflectance infrared Fourier transform spectroscopies, coupled with modeling studies, reveal the crucial role of Fe–O(H)–Fe sites in activating CH4 and stabilizing reaction intermediates via the formation of an Fe–O(CH3)–Fe adduct. In addition, a cascade fixed-bed system has been developed to achieve online separation of C2H4 and C2H2 from unreacted CH4 for direct use. Integrating the processes of CH4 activation, conversion, and product separation within one system opens a new avenue for natural gas utility, bridging the gap between fundamental studies and practical applications in this area.


Preparation of the materials
All the reagents were used as received from commercial suppliers without further purification.Synthesis of MFM-300(Fe).MFM-300(Fe) was synthesised using our previously reported method. 1 Typically, 244 mg of biphenyl-3,3′,5,5′-tetracarboxylic acid and 800 mg of FeCl 3 •6H 2 O were dispersed in a mixture solution of N,N-dimethylformamide (DMF, 20 mL) and concentrated hydrochloric acid (0.75 mL) in a 50 mL round-bottom flask under ambient pressure.The reaction was heated at 120 °C for 72 h, and the yellow powder product was collected and washed with DMF and acetone several times.The as-synthesised MFM-300(Fe) was dried and activated by heating at 150 °C under dynamic vacuum before further use.

Synthesis of MIL-53(Fe)
. MIL-53(Fe) was synthesised by a solvothermal method according to the reported method. 2 Typically, 0.66 g of terephthalic acid (H 2 BDC) and 1.08 g of iron(III) chloride hexahydrate (FeCl 3 •6H 2 O) were dissolved in 20 mL of DMF.The mixture was transferred into a teflon lined bomb, sealed and heated at 150 °C for 24 h.The product was collected by centrifugation and washed with DMF and deionized water several times.The as-synthesised MIL-53(Fe) was dried and activated by heating at 150 °C under dynamic vacuum before further use.

Synthesis of MIL-100(Fe)
. MIL-100(Fe) was synthesised by a solvothermal method according to the reported method. 3Typically, 0.32 g of trimesic acid (H 3 BTC) and 0.68 g of iron(III) nitrate nonahydrate [Fe(NO 3 ) 3 •9H 2 O] were dissolved in a mixture solution containing 10 mL of deionized water and 0.18 mL of hydrofluoric acid.The mixture was transferred into a teflon lined bomb, sealed and heated at 150 °C for 12 h.The product was collected by centrifugation and washed with deionized water and EtOH several times.The as-synthesised MIL-100(Fe) was dried and activated by heating at 150 °C under dynamic vacuum before further use.
Synthesis of Fe/ZSM-5.Fe/ZSM-5 was synthesised according to a reported method. 4Typically, 500 mg of ZSM-5 was added into a solution containing Fe(NO 3 ) 3 •9H 2 O (800 mg) dissolved in deionized water (20 mL), and the mixture was then heated at 85 °C for 24 h.The product was centrifuged, washed with deionized water several times, dried overnight at 80 °C and finally calcined at 550 °C under air flow for 3h.

Synthesis of HKUST-1.
HKUST-1 was synthesised through solvothermal method according to the reported method. 5,6Typically, 0.84 g of H 3 BTC and 1.45 g of copper(II) nitrate trihydrate [Cu(NO 3 ) 2 •(H 2 O) 3 ] were dissolved in a mixture solution containing DMF (20 mL), EtOH (20 mL)   and deionized water (20 mL).The mixture was transferred into a round bottom flask, then heated at 85 °C for 12 h.The product was collected by centrifugation and washed with DMF and EtOH several times.The as-synthesised HKUST-1 was dried and activated by heating at 150 °C under dynamic vacuum before further use.

General characterisation of the materials
Powder X-ray diffraction (PXRD) patterns were recorded on a Philips X'pert X-ray diffractometer (40kV and 30 mA) using Cu-Kα radiation (λ = 1.5406Å).Nitrogen adsorption isotherms were collected on a Micromeritics 3Flex analyzer at 77 K.The samples were activated under dynamic vacuum before measuring N 2 isotherms.The specific surface area was calculated using the Brunauer-Emmett-Teller (BET) method.Scanning electron microscopy (SEM) imaging and energy dispersive X-ray spectroscopy (EDX) analysis was performed using the FEI/Thermofisher Quanta 650 field emission gun SEM at the University of Manchester.The SEM was equipped with a Bruker X Flash 6 | 30 silicon drift detector with Bruker ESPRIT EDX software v2.2.For highresolution imaging, beam deceleration was employed to achieve a landing energy of 1 kV.For EDX analysis, beam conditions were set to 15 kV.High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images and EDX elemental maps were collected on a Thermo Fisher Titan STEM (G2 80-200) equipped with a Cs probe corrector (CEOS), high-angle annual dark-field (HAADF) detector and ChemiSTEM Super-X EDX detector, operating at 200 kV.TEM samples were prepared by dispersing the powders in methanol and drop-cast onto a copper grid coated with an amorphous holey carbon film.X-ray photoelectron spectroscopy (XPS) was performed using an Axis Ultra Hybrid spectrometer (Kratos Analytical, Manchester, United Kingdom) using monochromated Al-Kα radiation (1486.6 eV, 10 mA emission at 150 W, spot size 300 x 700 μm) with a base vacuum pressure of ~5×10 −9 mbar.Charge neutralisation was achieved using a filament.Binding energy scale calibration was performed using C-C in the C 1s photoelectron peak at 284.8 eV.Analysis and curve fitting was performed using Voigt-approximation peaks using CasaXPS.

Inelastic neutron scattering (INS)
Direct visualisation of the interactions between adsorbed CH 4 molecules and the active sites in MFM-300(Fe) is crucial to understanding the molecular details of adsorption, activation and conversion of CH 4 over catalyst.INS is a powerful neutron spectroscopy technique 7

Experimental setup for catalytic testing
For the NTP-assisted reaction, the reactor comprised of two coaxial quartz tubes; the outside diameters of the outer tube and inner tube was 6 mm and 3 mm, respectively, giving a discharge gap of 0.5 mm. 8,9The outer tube was covered by a metal mesh electrode that was connected to a high-voltage output, and a metal wire electrode (ground electrode) was placed inside the inner tube.The catalyst was packed in the discharge region to ensure that plasma was generated around the catalyst.An alternating current plasma generator (CTP-2000K, 0-25 kV, 10 kHz) was used to ignite the plasma, and an oscilloscope (Tektronix TDS 2022B) was used to monitor the electrical parameters.The discharge power used for reaction was about 2 W and the specific energy input was calculated to be around 2 kJ L −1 , with the AC peak-to-peak voltage (Vpk-pk) at around 20 kV and a frequency of 10 kHz.At a high CH 4 conversion of ~10%, the specific energy input was calculated to be around 8 kJ L −1 .
The gaseous products were detected using (i) a Bruker MATRIX-MG5 FTIR spectrometer (resolution = 0.5 cm −1 ) for CH 4 , C 2+ products, CO and CO 2 and (ii) a mass spectrometer (Hiden QGA quantitative gas analysis system, Hiden Analytical Ltd.) for H 2 .The Bruker MATRIX-MG5 features a 5 m multi-reflection gas cell and is designed for the high-precision quantification of gas compounds from very low concentrations on the ppb level up to one hundred percent.The gas analysis system uses the certain sections of FTIR spectrum that are unique to a given gas to first identify the gas is present and then uses a fitting algorithm to quantity the amount of gas present.
If other gases interfere with the signal, then the system also fits these interfering gases and takes them into account when performing the quantification.The system was calibrated and set up before use.

In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS)
To investigate the surface chemistry of the NTP-assisted CH 4 conversion reaction, in situ DRIFTS measurements were performed using a Bruker Vertex 70 FTIR spectrometer equipped with a liquid N 2 -cooled detector, and a custom-made fixed-bed NTP-DRIFTS reactor. 10Approximately 25 mg of MFM-300(Fe) sample was placed in a crucible in an in situ DRIFTS cell.The gas mixture was controlled by independent mass flow controllers and introduced to an in situ DRIFTS cell via a 4way valve that allows a switch between He pre-treatment and CH 4 gas feed.
Prior to DRIFTS measurement, MFM-300(Fe) was pre-treated to remove any adsorbed water by heating up to 120 C in He flow with a total flow rate of 50 mL min −1 .After 1 h, the temperature was decreased to room temperature and the IR spectrum of MFM-300(Fe) recorded as a background.Subsequently, the gas feed containing 1% CH 4 (diluted in He balance) was introduced to the NTP-DRIFTS reactor for 5 min before switching on the plasma.The power electrode was driven at ~ 27 kHz with an applied peak voltage of 5.0 kV.The discharge electrical parameters were measured using a resistor (10 Ω) for the current and a high-impedance probe (Tektronix, P6015) for the applied voltage, connected to a digital oscilloscope (Tektronix TBS1072B).The cycling switches between NTP on and off were carried out every 10 min to monitor the effect of plasma on the changes of surface species.All DRIFT spectra were recorded with a resolution of 4 cm −1 and an accumulation of 128 scans every 60 s and analysed by the OPUS software.

Experimental setup for C2+ products separation
A cascade system was designed for the separation of the C 2+ hydrocarbon products from unreacted CH 4 .Typically, the feed gas (CH 4 diluted in He) was passed through the fixed bed plasma reactor (fixed bed-1, see Figure S1 for details of the experimental setup), and CH 4 was partially converted to C 2+ products.When the gas mixture passes through line 1 (without the fixed bed-2 for C 2+ collection), the gas mixture is analysed using a Bruker Matrix MG5 FTIR spectrometer.When the gas mixture passes through line 2 where the fixed bed-2 is packed with porous materials (HKUST-1 or ZSM-5), C 2+ products can be separated and collected and the unreacted CH 4 detected by FTIR spectroscopy.The INS spectra in Figure S24a show excellent agreement between simulated and measured bare MFM-300(Fe), confirming that the sample is clean and the structural model for simulation is correct.This affords the assignment of INS peaks.The strongest peaks in the low-energy region (<40 meV) are all due to CH 4 libration, including translational and rotational motions.In Figure S24c, the difference spectra show sharp peaks for adsorbed CH 4 that are very different from solid CH 4 .The peaks are sharp because the CH 4 are pinned to the adsorption site with very little recoil, whereas in solid CH 4 the intermolecular interaction is much weaker and the incident neutrons cause significant recoil.In order to detect the presence of bound CH 4 , the sample cell was evacuated under dynamic vacuum with a turbo pump.The remaining CH 4 appears to be in a similar adsorption status (Figure S24c, blue line).The sample cell was then evacuated with a turbo pump for another 45 minutes.Surprisingly, this process did not remove all CH 4 , with some INS signal for CH 4 still detected (Figure S24c, green line), confirming that some CH 4 molecules are bound to MFM-300(Fe).Interestingly, the integrated intensity of the entire spectrum (including both elastic and inelastic scattering), which scales roughly with the amount of CH 4 , has a ratio of 1:0.66:0.26among the three curves in Figure S24c (red, blue and green curves).The initial loading was about 4 mmol g −1 of CH 4 (1 CH 4 per Fe site).Therefore, there was ~1 mmol g −1 of CH 4 retained within the pores after desorption, consistent with the CH 4 isotherm data.Surface −OH species that are relevant to the activation of CH 4 and stabilisation of intermediates will show changes at a similar rate to that observed for the gas phase reactants.The relative correlation between (OH) at 3649 cm −1 (MOF) and (CH) at 3016 cm −1 as a function of ToS during plasma on-off is shown in Figure 3b and Figure S25.Interestingly, it is observed that the rate of decrease/increase of (CH) from CH 4 (g/ads.)matches very closely with that of surface (OH) upon switching plasma on/off (Figure 3b and Figure S25).This close correlation between the two species indicates that CH 4 molecules are activated and converted at the  2 −OH sites.

Figure S1 .
Figure S1.View of the experimental setup for the cascade system.Fixed bed-1 is packed with catalyst for conversion of CH 4 to C 2+ products, and fixed bed-2 is packed with porous materials (HKUST-1 or ZSM-5) for separation of C 2+ products from unreacted CH 4 .

Figure S6 .
Figure S6.Views of the structure of MFM-300(Fe) determined by NPD at 7 K (C, grey; O, red; Fe, light orange; H, white).The Fe−O(H)−Fe hydroxyl sites are partially highlighted in ball-and stick model.

Figure S8 .
Figure S8.Normalised Fe K-edge XANES spectrum of as-synthesised MFM-300(Fe) and standard XANES spectrum of Fe 2 O 3 reference (left).Plot of non-phase corrected Fourier transformed Fe K-edge EXAFS data for as-synthesised MFM-300(Fe) shown against a Fe 2 O 3 standard reference (right).

Figure S14 .
Figure S14.Comparison of the catalytic activity (time yield for total C 2+ products) and product selectivity over different catalysts under activation of NTP.Reaction conditions: specific energy input (SEI) of 8 kJ L −1 for MFM-300(Fe) at high SEI, and 2 kJ L −1 in other experiments; 2% CH 4 in He for MFM-300(Fe) at high SEI and 1% CH 4 in He for other experiments, with a total flow rate for the gas feed of 60 mL min −1 ; 60 mg of catalyst, 25 °C and 1 atm.

Figure S15 .
Figure S15.Time yield of C 2+ hydrocarbon products and H 2 over MFM-300(Fe) under activation of NTP.Reaction conditions: specific energy input of 2 kJ L −1 , 1% CH 4 in He as gas feed at a total flow rate of 60 mL min −1 , 60 mg of catalyst, 25 °C and 1 atm.

Figure S17 .
Figure S17.Comparison of C 2+ selectivity and CH 4 conversion over 3 cycles performed at a high specific energy input of 8 kJ L −1 with a CH 4 conversion of ~10%; 2% CH 4 in He with a total flow rate of 60 mL min −1 as feed gas, 60 mg of catalyst, 25 °C and 1 atm.

Figure S21 .
Figure S21.Normalised Fe K-edge XANES spectrum of used MFM-300(Fe) and standard XANES spectrum of Fe 2 O 3 reference (left).Plot of non-phase corrected Fourier transformed Fe K-edge EXAFS data for used MFM-300(Fe), shown against a Fe 2 O 3 standard reference (right).

Figure S22 .
Figure S22.(a) In situ DRIFT spectra of MFM-300(Fe) in NTP-assisted CH 4 conversion as a function of plasma on and off, showing the red shift of the υ(O−H) stretching mode of the  2 −OH in MFM-300(Fe).All the DRIFT spectra were recorded at a resolution of 4 cm −1 , with the spectrum of KBr subtracted as the background.(b) Mass spectrometric (MS) signals measured at the exit of the DRIFTS cell during NTP-assisted CH 4 conversion (plasma is turned on at 5 mins, and turned off at 15 mins).

Figure S24 .
Figure S24.(a) Comparison of experimental and simulated INS spectra for bare MFM-300(Fe).(b) Comparison of experimental and simulated difference INS spectra for CH 4 loaded MFM-300(Fe).(c) Comparison of the difference INS spectra for CH 4 loaded MFM-300(Fe), the difference INS spectra upon evacuating under dynamic vacuum with a turbo pump, and the INS spectra for solid CH 4 .

Figure S25 .
Figure S25.In situ DRIFT spectra showing (a, c) the (OH) stretching vibration in MFM-300(Fe), and (b, d) the (CH) stretch of CH 4 in NTP-assisted CH 4 conversion reactions as a function of plasma on-off and reaction time.The correlation between relative intensities of the (OH) peak at 3649 cm −1 and (CH) peak at 3016 cm −1 is shown in Figure 3b.

Figure S26 .
Figure S26.In situ DRIFT spectra of catalysts [top to bottom: MFM-300(Fe), MIL-53(Fe) and Fe/ZSM-5] in NTP-assisted CH 4 conversion reaction as a function of plasma off and on.CH 4 was flowing for both plasma off and on.All the DRIFT spectra were recorded at a resolution of 4 cm −1 , and the spectra of bare catalysts have been subtracted as the background.

Table S1 .
Summary of BET surface areas determined from N 2 sorption isotherms.

Table S2 .
Fe K-edge EXAFS fitting parameters of fresh and used MFM-300(Fe).

Table S3 .
Comparison of the reported catalysts for CH 4 conversion to C 2+ products.