Plasma catalysts for preparing products and devices therefor

Abstract

Described are methods of selectively producing a target carbon-containing product from a methane-containing gas, the method including: (a) forming a non-thermal plasma in the presence of the methane-containing gas and a metal catalyst, (i) wherein the methane-containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (iii) wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature, and that dissociatively adsorbs on the metal catalyst; (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, and / or selecting a different metal catalyst; thereby producing hydrogen gas and the target carbon-containing product, wherein the target carbon-containing product is produced in a greater amount than non-target carbon-containing products. Devices for performing the described methods are also disclosed.

Description

[0001] Attorney Docket No. 10046-649W01

[0002] PLASMA CATALYSTS FOR PREPARING PRODUCTS AND DEVICES THEREFOR

[0003] CROSS REFERENCE TO RELATED APPLICATIONS

[0004] This application claims the benefit of priority to U.S. Provisional Application 63 / 755,786, filed on February 7, 2025, which is incorporated by reference herein in its entirety.

[0005] STATEMENT ACKNOWLEDGING GOVERNMENT SUPPORT

[0006] This invention was made with government support under grant no. DE-SC0024437 awarded by the Department of Energy. The government has certain rights in the invention.

[0007] BACKGROUND

[0008] Methane (CH4), a crucial component of various natural gas sources (natural gas, shale gas, methane clathrates, and biogas), is a key hydrogen (H2) vector (Frankenberg, C. et al. Assessing Methane Emissions from Global Space-Borne Observations. Science (1979) 308, 1010- 1014 (2005); Fletcher, S. E. M. et al. Rising methane: A new climate challenge. Science (1979) 364, 932-933 (2019); Yu, X. et al. Stoichiometric methane conversion to ethane using photochemical looping at ambient temperature. Nat Energy 5, 511—519 (2020)). Despite its substantial global reserves (Williams, C. et al. Selective Oxidation of Methane to Methanol Using Supported AuPd Catalysts Prepared by Stabilizer-Free Sol- Immobilization. ACS Catal 8, 2567-2576 (2018); Alvarez-Galvan, M. C. et al. Direct methane conversion routes to chemicals and fuels. Catal Today 171, 15-23 (2011)), the commercialization of natural gas (~ 70 - 90% CH4) is hindered by stringent storage conditions required for liquefied natural gas (LNG, T < -111°C). Current methods of utilizing CH4(oxidative reforming) demand high operating temperatures (> 900 °C) and pressures (up to - 50 bar), large reactors and high capital costs. Approximately, 90% of methane currently burnt in various reforming processes releases ~ 12.5 - 13.5 kgCO2-eq / kgH2into the atmosphere (McFarland, E. Unconventional Chemistry for Unconventional Natural Gas. Science (1979) 338, 340-342 (2012); Dapsens, P. Y. et al. Biobased Chemicals from Conception toward Industrial Reality: Lessons Learned and To Be Learned. ACS Catal 2, 1487-1499 (2012)). While conversion of methane to H2is desirable, several key industrial feedstocks can be derived from methane, such as small hydrocarbons with aliphatic and olefinic content, such as C2 compounds, ethane (C2H6), and ethylene (C2H4) (Riaz, A. et al. A review of cleaner production methods for the manufacture of methanol. J Clean Prod 57, 19—37 (2013); Lange, J.-P. Methanol synthesis: a short revi ew of technology improvements.Attorney Docket No. 10046-649W01

[0009] Catal Today 64, 3-8 (2001); Schwach, P. et al. Direct Conversion of Methane to Value- Added Chemicals over Heterogeneous Catalysts: Challenges and Prospects. ChemRev 117, 8497-8520 (2017); Lallemand, M. et al. Catalytic oligomerization of ethylene over Ni-containing dealuminated Y zeolites. Appl Catal A Gen 301, 196-201 (2006); Geyer, R. et al. Production, use, and fate of all plastics ever made. Sci Adv 3, (2017); Naliapareddy, C. R. et al. Tailoring Vibrational Excitation Pathways for High-Yield Oxidation of Methane to Methanol. ACS Sustain Chem Eng 12, 9144-9155 (2024)). To fully utilize the potential of methane and decarbonize our energy systems, development of carbon-neutral methods that operate efficiently at milder thermodynamic conditions is needed.

[0010] An alternative approach to produce H2at milder thermodynamic conditions is direct conversion of methane to H2and C(s)or higher hydrocarbons. This process is energetically favorable (ΔR= 74.6 kJ / mol for conversion into C(s)), allowing operation at lower pressures (~1 bar) and sequestration of carbon as carbon black or value-added hydrocarbons. However, methane’s inherent stability, arising from its symmetrical tetrahedral geometry, low polarizability, and strong C-H bonds (439 kJ / mol bond energy), requires extreme conditions for its conversion. Consequently, there is a growing demand for innovative methods to activate methane and enhance the rate of dissociative adsorption on a broader spectrum of catalysts. Current direct conversion routes, such as methane aromatization and methane pyrolysis, often necessitate harsh operating conditions (often exceeding 900°C), leading to rapid coke deposition, catalyst sintering, and eventual catalyst deactivation. The coking and poisoning of catalysts degrade production rates and eventually necessitate the regeneration of catalytic surfaces (a process that typically emits CO2).

[0011] Therefore, there is an urgent need to develop emission-free processes that enable the direct, on-demand conversion of CH4into H2and value-added hydrocarbons under mild conditions.

[0012] In addition to the production of H2 from methane, the production of other high-value products like methanol and other liquid fuels often requires processes that are inefficient and / or have low selectivity. Thus, a general strategy and devices that can be used to prepare such products are greatly needed. The compositions, methods, and devices disclosed herein address these and other needs.

[0013] Still further, there is a significant need for the removal of nitrous oxide (N2O) from gas streams. N2O is a potent greenhouse gas with significant environmental impacts on both climate and atmospheric chemistry. N2O has a global warming potential over 250 times higher than carbon dioxide (CO2) and an atmospheric lifetime exceeding 100 years, making oAttorney Docket No. 10046-649W01

[0014] it a major contributor to long-term radiative forcing, comparable to hydrofluorocarbons (HFCs) and perfluorocarbons (PFCs). In addition to its role in climate change, N2O is the leading ozone-depleting molecule emitted in the 21st century. Major anthropogenic sources of N2O include nitrogen-based fertilizers in agriculture, industrial processes such as adipic and nitric acid production, and emerging sectors like semiconductor manufacturing.

[0015] Although catalytic decomposition of N2O to N2and O2has been demonstrated using thermal reactors, these systems require high operating temperatures (430-970 K) and rely on expensive heterogeneous catalytic materials. These limitations highlight the need for novel low-temperature, energy-efficient N2O abatement pathways that utilize abundant catalysts. The compositions, methods, and devices disclosed herein address these needs as well.

[0016] SUMMARY

[0017] In accordance with the purposes of the disclosed materials, compounds, compositions, articles, devices, and methods, as embodied and broadly described herein, the disclosed subject matter, in one aspect, relates to compositions and methods for preparing and using the disclosed compositions. In more specific aspects, methods of forming target carbon-containing products are disclosed. Devices that can be used to perform the disclosed methods are also disclosed. A unique aspect of the disclosed subject matter is that the skilled artisan can selectively obtain one target carbon-containing product over other possible products by choosing a particular combination of adjustable conditions.

[0018] In some aspects, the techniques described herein relate to a method of selectively producing a target carbon-containing product during the production of hydrogen gas from a methane-containing gas, the method including: (a) forming a non-thermal plasma in the presence of the methane-containing gas and a metal catalyst, (i) wherein the methane- containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (iii) wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature, and that dissociatively adsorbs on the metal catalyst; (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, and / or selecting a different metal catalyst; thereby producing hydrogen gas and the target carbon-containing product, wherein the target carbon-containing product is produced in a greater amount than non¬ target carbon-containing products.

[0019] In some other aspects, the techniques described herein relate to a method of selectively producing a target carbon-containing product during the production of a liquidAttorney Docket No. 10046-649W01

[0020] fuel from a methane-containing gas and a vibrationally active oxygen-containing co¬ reactant, the method including: (a) forming a non-thermal plasma in the presence of the methane-containing gas, the vibrationally active oxygen-containing co-reactant, and a metal catalyst; (i) wherein the methane-containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (hi) wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature and vibrationally excited oxygen-containing co-reactant having a oxygen-containing co-reactant vibrational temperature, and that each dissociatively adsorbs on the metal catalyst; (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, adjusting tlie oxygen-containing co-reactant vibrational temperature, and / or selecting a different metal catalyst, thereby producing a liquid fuel and the target carbon-containing product, wherein the target carbon-containing product is produced in a greater amount than non-target carbon-containing products.

[0021] In some further aspects, the techniques described herein relate to methods of removing nitrous oxide from a gas stream, by (a) forming a non-thermal plasma in the presence of a nitrous oxide-containing gas, an additional co-reactant, and a metal catalyst, (i) wherein the nitrous oxide-containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (iii) wherein the non-thermal plasma produces vibrationally excited nitrous oxide having a nitrous oxide vibrational temperature, a vibrationally excited co-reactant having a co-reactant vibrational temperature, and that each dissociatively adsorbs on the metal catalyst; and (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the nitrous oxide vibrational temperature, adjusting the co-reactant vibrational temperature, and / or selecting a different metal catalyst; thereby removing nitrous oxide from the gas stream, and producing nitrogen and oxygen gases.

[0022] In some further aspects, described herein is a device for producing a product from a gaseous reactant, including: (a) a reaction chamber having an inlet for the gaseous reactant and outlet for the product; (b) a valve connected to the inlet that controls flow rate of gaseous reactant into the reaction chamber and a valve connected to the outlet that controls flow rate of product out of the chamber; (c) two electrodes, wherein the reaction chamber is positioned between the two electrodes, (d) an electrical energy supply connected to the two electrodes to supply sufficient electrical energy to produce a non-thermal plasma from the gaseous reactants; (e) a metal catalyst positioned inside the reaction chamber; and (f) an optional electrical heater connected to tlie device to heat the reaction chamber, heat theAttorney Docket No. 10046-649W01

[0023] metal catalyst, and / or heat the gaseous reactant.

[0024] Additional advantages of the disclosed subject matter will be set forth in part in the description that follows and the Figures, and in part will be obvious from the description, or can be learned by practice of the aspects described below. The advantages described below will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive.

[0025] BRIEF DESCRIPTION OF FIGURES

[0026] The accompanying figures, which are incorporated in and constitute a part of this specification, illustrate several aspects described below.

[0027] Figs. 1A-1C show plasma-catalytic design space for the direct conversion of methane. (Fig. 1A) The plasma-catalytic design space in non -equilibrium (i.e., vibrational) plasmas is an intersection of conditions that are governed by the plasma, catalyst, and gaseous state. Key parameters identified for the design space are vibrational temperature, gas temperature, surface temperature, binding energy of carbon, and pressure. (Fig. IB) Plasma-catalytic conversion consists of multiple reaction types: gas-phase conversion into primary products via electron impact excitation, Langmuir- Hinshelwood (LH) and Eley- Rideal (ER) type reactions involving catalytic surfaces, and conversion of desorbed catalytic gas products into secondary gas products via electron impact excitations. (Fig. 1C) Low-temperature plasma-catalysis breaks catalytic scaling constraints, achieves tunable product speciation, and eliminates coking through independent control of plasma- induced properties and surface / gas-phase properties.

[0028] Figs. 2A-2E show catalytic scaling laws constrain the performance of direct methane conversion. (Fig. 2A) Direct methane conversion on catalytic surface is a multistep process. This includes dehydrogenation of methane to H2in reactions R1-R5 and its conversion to C? hydrocarbons in reactions R6-R10. (Fig. 2B) Linear scaling relationships for R1-R10 on transition metal surfaces (111 and 211) for activation and reaction energies as a function of the binding energy of carbon. (Fig. 2C) Production rates (i.e., TOFs) of H2and C2hydrocarbons on 111 metal sites at 10% initial methane conversion, Tsur= 873 K, P = 1 atm. (Fig. 2D) Production rates of H2and C2hydrocarbons on 211 metal sites at 10% initial methane conversion, Tsur= 873 K, P = 1 atm. (Fig. 2E) Direct methane conversion via thermal catalysis in a flow reactor with fixed 10% initial methane conversion on Ni (111 ) at Tsra= 873 K, P = 1 atm results in early coking (~ 100minutes) that inactivate the catalyst.Attorney Docket No. 10046-649W01

[0029] Figs. 3A-3F show plasma-catalysis breaks catalytic scaling laws on transition metal catalysts. (Fig. 3A) Plasma excitations can populate methane in stretch mode (ν(1,3)) and bend mode (ν(2,4)) Methane is populated in 11 vibrational levels (ground + ν(2,4) − 10ν(2,4)) using a two temperature Treanor distribution in the microkinetic model (Figs. 7A-7B, Table 1)). (Fig. 3B) Plasma-catalysis of direct methane conversion on 111 metal sites at 10% initial methane conversion, Tsur= 473 K, Tvib= 4200 K, and P = 1 atm increases TOF of H2at reduced temperature and shifts optimal catalyst towards negative AEc compared to thermal catalytic pathway. H2TOF for 111 metal sites are at least 103times of H2TOF for 211 metal sites for Δc> −1. (Fig. 3C) Vibrational temperature of CH radical is determined by fitting the measured band at 430 nm with a synthetic spectrum (Fig. 3C, Fig.

[0030] 27). (Fig. 3D) Variation of yield of H2with specific input energy (SEI) for transition metal catalysts loaded on Al2O3. Hydrogen fuel production efficiency (77) are overlaid. (Fig. 3E) Variation of TOF ratio with specific input energy (SEI) for transition metal catalysts loaded on Al2O3. TOF ratio = 1 indicates contribution of catalyst has no impact on the production rate. (Fig. 3F) Measured H2yield at ~ 10% initial methane conversion agrees with the trend observed from the microkinetic model (Pt > Ni > Cu) on 111 metal sites at 10% initial methane conversion, Tsur= 473 K, Tvib= 4200 K, and P = 1 atm.

[0031] Figs. 4A-4D show plasma excitations enable tunable product speciation in direct methane conversion. (Fig. 4A) The design space in direct methane conversion involves the intersection of properties from catalysts (Ea), surface (Tsur), and gas state (P, Xi, T residence). Changing one of these parameters can lead to variations in the type of C2 hydrocarbons including, C2H6, C2H4, and C2H2. (Fig. 4B) Variation of yield with Δcfor H2, C2H6, and C2H4on 111 metal sites at 10% initial methane conversion, Tsur= 473 K, Tvib= 4200 K, and P = 1 atm. The trends are bounded by Tsur= 453 K and Tsur= 503 K. (Fig. 4C) Plasma- catalytic yield of C2H6, and C2H4 on Al2O3, Pt / AhOs, Ni / Al2O3, and CU / Al2O3at ~ 10% initial methane conversion. Plasma + Al2O3yield is subtracted to represent only the plasma- catalytic yield. (Fig. 4D) Influence of resident timescale (Tresidence) on the upgraded form of methane. At SEI >15 eV / initial CH4, desorbed C2H4from the catalytic surface undergoes secondary conversion in plasma phase to form C2H2.

[0032] Figs. 5A-5D show plasma and catalytic design can synergize to limit surface inactivation through coking. (Fig. 5 A) Plasma activation of CH 4 enhances the rate of dissociation adsorption which enables reaction to proceed to at lower Tsur. The rate of subsequent steps are set by surface properties (i.e., independent of plasma properties) thatAttorney Docket No. 10046-649W01

[0033] enable control over the removal of catalyst poisons (CHX) through recombination. This process facilitates the use of noble catalyst that resist coking. (Fig. 5B) Yield of H2on Ni / Al2O3is higher than C11 / Al2O3initially, but it eventually converges to the yield of Al2O3only in time on steam (TOS) experiments. Yield of H2 on C11 / Al2O3is constant till 14 hrs. (Fig. 5C) Long term effects in production rate (TOF H2) are captured by the microkinetic model for Ni (111) and Cu (111). The simulations are done in a flow reactor configuration at fixed 10% initial methane conversion (Tsur= 473 K, Tvib= 4200 K, and P = 1 atm). (Fig.

[0034] 5D) Cu / Al2O3overtakes Ni / Al2O3for cumulative H2yield at ~ 6 hrs. of runtime at fixed 10% initial methane conversion.

[0035] Figs. 6A-6D show probing reaction pathways that overcome surface inactivation. (Fig. 6A) Variation of normalized surface coverage of adsorbates over reaction time on Ni (111) and Cu (111) simulated in a flow reactor configuration at fixed 10% initial methane conversion (Tsur= 473 K, Tvib= 4200 K, and P = 1 atm). (Fig. 6B) XPS spectra of Ni / Al2O3and CU / Al2O3samples extracted from the TOS experiments at - 6 hrs. Surface carbon coverage is determined by normalizing C Is area to Al 2p3area. Coverage of functional groups of adsorbates (CH3*, CH2* / C1H4*, and C*) are normalized to C 1s area (Figs. 32 A- 32B, 33A-33D, 34A-34D, Table 6). (Fig. 6C) Variation of normalized surface carbon over reaction time on Ni (111) and Cu (111) (Tsur= 473 K, Tvib= 4200 K, and P = 1 atm).

[0036] Surface carbon coverage is defined as the sum of coverages of all carbon-containing adsorbates. (Fig. 6D) TEM / EDS scans Ni / Al2O3and CU / A12O3samples extracted from the TOS experiments at ~ 6 hrs. Scans are performed where Ni and Cu catalysts are identified. Carbon coverage on Ni / Al2O3active sites is 100%, while it is ~ 60% on CU / A12O3active sites, confirmed by simulations.

[0037] Figs. 7A-7B show the distribution of energies across polyads. (Fig. 7 A) The average polyad energy scales approximately linearly with polyad number (Pn). The dissociation energy of CH4is ~35000 cm-1. (Fig. 7B) The spread of energies among states in the bending mode (n2, 114) is - 16%’ of the average energy of a given polyad.

[0038] Fig. 8 shows a comparison between population fractions for the lumped Boltzmann and Treanor distribution functions of the CH4ν(2,4) mode. Results presented in the main text use the Treanor distribution.

[0039] Fig. 9 shows a detailed methane excitation and relaxation scheme for the dyad (Pi) and pentad (P2) polyads. CH4 excitation to bend and stretch modes takes place via electron impact and vibrational-vibrational (V-V) coupling. Vibrational-translational (V-T)Attorney Docket No. 10046-649W01

[0040] relaxation towards the ground state of CH4 (p(0)) takes place via molecular collisions.

[0041] Fig. 10 shows timescales and diffusion length scales of bend and stretch modes of vibrating methane and methyl radical. The bending mode has the highest lifetime of ~2 [is, and the diffusion length scale of ~12 ptn among the species of interest.

[0042] Fig. 11 shows that a direct methane conversion involves a combination of plasma and catalytic processes. Initially, plasma-induced reactions occur in the gas phase, generating hydrogen and higher hydrocarbons. Subsequently, these gaseous products interact with the catalyst surface, undergoing further transformations. A secondary plasma conversion step can occur in the gas phase for specific products (e.g., C2H4 to C2H2), primarily at lower flow rates.

[0043] Fig. 12 shows global sensitivity coefficients for elementary reaction (R1-R10). Tsuris varied from 373 K-473 K. Simulations are performed at P= 1 atm, Tvib= 4200 K till 10% initial of CH4conversion is achieved.

[0044] Fig. 13 shows the yields of C2H6, C2H4, and H2at P = 1 atm, Tvib = 4200 K, and 10%’ initial CH4 concentration, illustrating the bounds influenced by the mean average error (MAE) in the activation energy (Ea,6) for reaction R6. The shaded bands indicate the range of variation in product speciation attributable to the MAE.

[0045] Figs. 14A-14C show vibrational temperature sensitivity (Fig. 14A), shortterm speciation (Fig. 14B), and long term catalyst performance (Fig. 14C) results at P = 1 atm, Tsui- = 473 K.

[0046] Figs. 15A-15C show a residual refinement study for absolute tolerance.

[0047] Figs. 16A-16C show a reducing the residual tolerance of the microkinetic model decreased error in TOF.

[0048] Fig. 17 shows the production rates (i.e., turnover frequencies (TOFs)) of H2, C2H2, C2H4, and C2H6for 111 sites using Treanor distribution at 10% initial CH4 conversion, P=1 atm, Tsur= 473 K and Tvlb= 4200 K.

[0049] Fig. 18 shows the production rates (i.e., turnover frequencies (TOFs)) of H2, C2H2, C2H4, and C2H6for 211 sites using Treanor distribution at 10% initial CH 4 conversion, P=1 atm, Tsur = 473 K and Tvlb = 4200 K.

[0050] Fig. 19 shows a setup for plasma-catalytic experiments. Features of the setup include a modular reactor for plasma-catalytic reactions, online gas chromatography (GC) with TCD and FID, and a spectrometer for OES.

[0051] Figs. 20A-20B show (Fig. 20A) applied voltage and (Fig. 20B) measured currentAttorney Docket No. 10046-649W01

[0052] traces for a DBD plasma operating at Q = 7.5 seem, V = 8 kV, and f = 20kHz.

[0053] Figs. 21 A-21E show calibration curves for (Fig. 21A) methane, (Fig. 21B) hydrogen, (Fig. 21C) ethylene, (Fig. 21D) ethane, and (Fig. 21E) acetylene. Each sample (250 / / I) was injected into the GC through gas sampling valves (GSVs). Calibration relates the number of moles of an analyte to the area of the peak on the GC chromatogram.

[0054] Fig. 22 shows sample GC chromatograms for FID and TCD loops. A TCD, equipped with Hayesep-Q column, was used to detect H2and CH4. A FID, equipped with PLOT-Q column, was used to detect C2H6, C2H4, C2H2, and H2.

[0055] Fig. 23 shows a comparison of H2yield with and without catalytic pretreatment using a H2DBD plasma. Plasma-treated catalyst resulted in ~1.4% more H2yield than the untreated catalyst.

[0056] Fig. 24 shows a time-averaged image of DBD plasma with quartz wool as a catalyst carrier taken at ISO =100, exposure time = 2 seconds and f / 2.

[0057] Figs. 25A-25F show an image sequence of showing the evolution of streamers within a packed bed reactor (200 fps, ~ 0.4 cm field of view).

[0058] Figs. 26A-26B show (Fig. 26A) simulated yields of gaseous species (C2H6, C2H4, C2H2, and H2) and (Fig. 26B) surface carbon coverage as a function of surface temperature for plasma-catalytic and thermal -catalytic direct conversion of methane over Ni(l 11) and Cu(lll) at 1 atm and 10% CH4 conversion.

[0059] Fig. 27 shows a measurement of vibrational temperature (Tvib) for CH radical. CH A-X spectrum is measured from OES. Simulate spectrum from massiveOES is fitted on the measured spectrum. State-specific populations for CH radical are fitted with non-Boltzmann (Treanor) distribution.

[0060] Fig. 28 shows the variation of vibrational temperature approximation with the ratio of first vibrationally excited population to ground state population of the species for CH and C2 radicals, and CH4 molecule.

[0061] Fig. 29 shows SEM / EDS of Cu / AEOs and Ni / AEOs after exposure to DBD plasma operating at 8 kV for ~ 360 min. A uniform flow rate of 3.75 seem was maintained during the experiment. SEM / EDS scans were done on at least 10 different locations on a given sample.

[0062] Fig. 30 shows a TEM of Ni / AbOa after exposure to DBD plasma operating at 8 kV for ~ 360 min. A uniform flow rate of 3.75 seem was maintained during the experiment. The lattice structure of the metal catalyst is observed.

[0063] Fig. 31 shows a TEM of Cu / Al2O3after exposure to DBD plasma operating at 8 kVAttorney Docket No. 10046-649W01

[0064] for ~ 360 min. A uniform flow rate of 3.75 seem was maintained during the experiment. Elemental maps were obtained on the C11 / Al2O3particle trapped in a broken lacy carbon grid.

[0065] Figs. 32A-32B show normalized XPS survey spectra of (Fig. 32A) Ni / Al2O3, and (Fig. 32B) C11 / Al2O3. The samples were exposed to CH4 plasma for -360 min.

[0066] Figs. 33A-33D show normalized XPS high resolution spectra of Ni / Al2O3sample for Ni 2p3, O 1s, C 1s, and Al 2p3. The samples were exposed to CH4 plasma for -360 min.

[0067] Figs. 34A-34D show normalized XI’S high resolution spectra of CU / Al2O3sample for Cu 2ps, O Is, C Is, and Al 2ps. The samples were exposed to CPU plasma for -360 min.

[0068] Fig. 35 shows the plasma catalytic design space for controlled partial oxidation of methane with the addition of a vibrationally active co-reactant. Traditional pathways for the production of liquid fuels from methane suffer from two major challenges, catalytic inactivation and over-oxidation of reactants to produce CO. By combining plasmas, catalysts, and gas state properties together, the production pathway can be controlled by increasing the selectivity of key intermediates to form liquid fuels.

[0069] Fig. 36 shows the yield of liquid fuels in the partial oxidation of methane with CO2in this work. While the current yield is - 1%, the yield can be increased up to - 12 % with optimized reaction engineering. With process optimization, the energy-adjusted yield of liquid fuels done at Tg= 300 K, and P = 1 atm can be closer to the yield from conventional methods done at Tg> 1000 K, and P > 30 atm.

[0070] Fig. 37 shows a schematic of the experimental setup. Features of the setup include a hot plate, a cold water trap maintained at - 1 °C, a modular reactor for plasma-catalytic reactions, online gas chromatography (GC / MS) with TCD, and a spectrometer for OES.

[0071] Fig. 38 shows how plasma activation can break the kinetics of the first step by lowering the energy barrier for catalytic reactions. The chemisorbed species on the surface are in equilibrium with the surface; hence, the subsequent reaction rates are set by surface properties.

[0072] Fig. 39 shows the production of key intermediate CH3O* (methanol precursor) on the catalytic surfaces. From the DRIFTS spectra, it is observed that CH3O* is formed on the Cu / AhOs, while Ni / Al2O3and AbOsdo not show evidence of CH3O* on the surface. From the DFT calculated energies, Cu (111 ) is more favorable to produce CH₃O* than Ni (111).

[0073] Fig. 40 shows the evolution of H2O over time under different reaction conditions. First, the reactor is flushed with CH4 / CO2 / H2, then the plasma is ignited. At - 90 min, the flow of CH4 / CO2 is stopped while plasma remains ignited. C11 / Al2O3produces the largestAttorney Docket No. 10046-649W01

[0074] amount of H2O at t > 90 min due to surface inactivation by O*.

[0075] Figs. 41 A-41D show plasma-catalytic liquid fuel production pathways for different classes of catalysts. (Fig. 41A) Classification of transition metal catalysts based on oxygen and carbon binding energies on metal surfaces, identifying weakly and strongly binding regimes for CH4-CO2 conversion. (Fig. 41B) Comparison of reported methanol selectivity versus CH4 conversion for metal-loaded catalysts (e.g., Cu / Al₂O₃, Ni / Al₂O₃) versus bare supports (e.g., Al2O3), highlighting the role of transition metal catalysts. Dashed lines indicate a constant methanol yield. (Fig. 41C) Schematic of plasma-assisted activation following a Langmuir-Hinshelwood mechanism, in which plasma excitation enhances dissociative chemisorption, enabling subsequent surface reactions. Product formation is governed by competing pathways, where catalyst deactivation or intermediate over¬ oxidation can suppress rates of conversion or liquid fuel selectivity, respectively. (Fig. 41D) DFT-calculated energy diagrams for Cu (111) and Ni (111) reveal that Cu favors stabilization of oxygenates like CH₃O*, enabling methanol formation, whereas Ni stabilizes CH fragments, facilitating carbon buildup and coking.

[0076] Figs. 42A-42H show deactivation mechanisms of weakly and strongly binding catalysts in plasma-catalytic CH4-CO2 conversion. (Fig. 42A) Schematic showing deactivation on Cu (weakly binding catalyst) through surface oxygen poisoning and on Ni (strongly binding catalyst) through carbon accumulation. (Fig. 42B) Temporal evolution of total liquid fuel selectivity reveals that Cu / Al₂O₃ and Ni / Al₂O₃ converge to the baseline performance of Al₂O₃ supports, indicating early-stage catalyst deactivation(~ 2-3 min) under non thermal plasma conditions (p = 1 atm, Tgas~ 473 K, Tv* ~ 4200 K. (Fig. 42C) High-resolution C Is spectra comparing the reference and spent Ni / AbOs samples (after 200 min plasma exposure with CH₄:CO₂ = 1:1) reveal an ~ 50-60% increase in both sp3- and sp2-hybridized carbon species, indicating coke accumulation on the catalyst surface. (Fig. 42D) High-resolution O Is spectra of reference and spent Cu / Al₂O₃ samples (200 min plasma exposure with CH₄:CO₂ = 1:1) show increased adatom (Ad-O) contribution (49%), indicating higher surface oxygen coverage. (Figs. 42E-42F) Operando DRIFT’S spectra showing time-resolved evolution (t = 0-30 min) of surface-bound poisons (mono- and bidentate carbonates) on Cu / Al₂O₃ and Ni / Al₂O₃ during plasma-catalysis. (Figs. 42G-42H) Quantitative analysis of operando DRIFTS spectra reveals a ~ 200%’ higher accumulation of mono- and bidentate carbonate species on Ni / Al₂O₃ compared to Cu / Al₂O₃, indicating surface poisoning by carbon -containing intermediates on the strongly binding catalyst.

[0077] Figs. 43A-43C show a comparison of vibrationally assisted Langmuir-HinshelwoodAttorney Docket No. 10046-649W01

[0078] and radical-assisted activation pathways as a function of mean electron temperature (Te) for plasma-enabled CH₄-CO₂ (1:1) catalysis on Ni(111). (Fig. 43A) Plasma-activated reactants can couple to the surface by vibrational excitations and radical pathways. Calculated reaction rates for (Fig. 43B) CH4 activation to CH₃* and H* and (Fig. 43C) CO2 activation to CO* and O*, comparing vibrationally excited dissociative adsorption with radical- mediated pathways. Rates are evaluated at 1 atm, 3% initial CH₄ conversion and Tgas= Tsur= 473 K using steady-state plasma compositions as inputs to the coupled microkinetic model.

[0079] Figs. 44A-44D show plasma-enabled active site regeneration. (Fig. 44A) Plasma- excited H2, an example co-reactant, enables active site regeneration at p = 1 atm, Tgas~ 473 K, Tvib ~ 4200 K by reacting with surface- bound intermediates, removing oxygen from weakly binding catalysts as H2O, and converting carbonaceous deposits on strongly binding catalysts into C1 / C2 hydrocarbons. (Fig. 44B) Experimental sequence for catalyst regeneration studies showing three phases: (i) purge with CH₄ / CO₂ (or CH₄ / CO₂ / H₂ for H₂ co-feed) (Q = 20 seem), (ii) plasma-catalytic reaction (P = 3-4 W), and (iii) H2-only plasma exposure to quantify regeneration effects. (Fig. 44C) Evolution of regenerated active sites during the H2 plasma phase: on Cu / Al₂O₃ through cumulative H₂O formation (removal of surface oxygen (Ad-O)), and on Ni / Al₂O₃ via C₁ / C₂ hydrocarbon production (removal of surface carbon). (Fig. 44D) DFT-calculated energy diagrams for Cu (111) and Ni (111) surfaces at Tsur = 473 K showing how O* and CHx* fragments recombine when H* is supplied.

[0080] Figs. 45 A-45D show the evolution of catalyst surface composition during plasma-enabled active-site regeneration. (Fig. 45 A) Deconvolved C is spectra of Ni / Al₂O₃ after 200 minutes of plasma exposure (p = 1 atm, Tgas~ 473 K, Tvib~ 4200 K, Q = 20 sccm) under CH4+CO2 (1:1) and CH4+CO2+H2 (1:1:4) conditions. (Fig. 45B) Deconvolved C Is spectra of Cu / Al₂O₃ after 200 minutes of plasma exposure under CH4+CO2 and CH4+CO2+H2 conditions. (Fig. 45C) Deconvoluted Cu 2p spectra of Cu / AhOs and distribution of Cu2+ / Cu° on Cu / Al₂O₃ after 200 minutes of plasma exposure under CH4+CO2 and CH4+CO2+H2 conditions. (Fig. 45D) Deconvolved O ls spectra of Cu / Al₂O₃ after 200 minutes of plasma exposure under CH4+CO2 and CH₄+CO₂+H₂ conditions. Surface oxygen adatoms accumulated on CU / Al2O3are removed under plasma-H₂ conditions, restoring the catalyst to its initial reduced state.

[0081] Figs. 46A-46C show a comparison of hydrogen activation and oxygen removal pathways as a function of mean electron temperature (Te) for CH4 / CO2 / H2 (1:1:4) plasmaAttorney Docket No. 10046-649W01

[0082] catalysis on Ni(111). (Fig. 46A) Reaction rates for vibrationally assisted H2 dissociation compared with radical-assisted hydrogen activation on vacant sites. (Fig. 46B) Reaction rates for OH formation from surface-bound H and O compared with direct reactions between gas-phase H and adsorbed O. (Fig. 45C) Reaction rates for H2O formation from surface-bound OH and H compared with direct reactions between gas-phase H and adsorbed OH. Rates are evaluated at 1 atm, 3% initial CH4 conversion, and Tgas= Tsur= 473 K using steady-state plasma compositions as inputs to the coupled microkinetic model.

[0083] Figs. 47A-47E show the evolution of reaction intermediates during plasma-enabled active site regeneration. (Fig. 47 A) Schematic representation of reaction pathways under thermal conditions (top), plasma without regeneration (middle), and plasma with active-site regeneration (bottom), highlighting intermediate formation and stabilization. In the absence of regeneration, operando DRIFTS on Cu / Al₂O₃ (p = 1 atm, Q = 20 sccm) shows a delayed onset of CO(g) signal, appearing after t = 5 min and intensifying over time, indicating catalyst deactivation (normalized integrated area ~ 0.11 at t = 30 min). In contrast, under regeneration conditions, no CO(g) (normalized integrated area - 0.02 at t = 30 min) signal is detected throughout the experiment, suggesting sustained surface activity. (Fig. 47B) Operando DRIFTS spectra showing progressive accumulation of OH* species on Cu / Al₂O₃, in contrast to transient formation and suppression on Ni / Al₂O₃. (Fig. 47C) DRIFTS spectra capturing the growth of CH3O* bands on Cu / Al₂O₃, while on Ni / Al₂O₃ these species appear briefly and decay rapidly (decay after t = 7 min). (Figs. 47D-47E) Quantified time evolution of OH* and CH3O* intermediates highlight sustained formation on Cu and limited stabilization on Ni.

[0084] Figs. 48A-48D show plasma-Catalytic Design Space for Liquid Fuel Synthesis across Catalyst Binding Regimes (Fig. 48A) Schematic Venn diagram defining the multidimensional plasma-catalytic design space as the intersection of three control knobs: plasma properties (electron temperature, electron density, excitation energy distribution, catalytic surface properties (binding energy, activation barriers), and gas-phase parameters including co-reactant composition and total pressure. Regions ①-④ represent distinct plasma-catalyst-gas coupling regimes influencing reaction pathways and product distributions. (Fig. 48B) Time-resolved evolution of total liquid oxygenate selecti vity for Al₂O₃, Cu / Al₂O₃, and Ni / Al₂O₃ during CH₄-CO₂ plasma operation (1:1 ratio, p =1 atm, Tgas~ 473 K, Tvib~ 4200 K, Q = 20 sccm), showing an initial decline in selectivity due to catalyst deactivation (t < 3 min), followed by performance recovery enabled by in situ regeneration. The shaded orange region marks the onset of catalytic passivation, while theAttorney Docket No. 10046-649W01

[0085] white region indicates sustained active-site availability under regeneration. (Fig. 48C) DFT- calculated energy diagrams illustrating how co-reactant addition shifts surface reaction pathways on weakly binding (Cu) and strongly binding (Ni) catalysts during CH4-CO2 conversion. (Fig. 48D) Carbon balance plot. Transparent bars represent CH4-CO2 plasma. Solid bars represent CH4-CO2-H2 plasma. Selectivity toward liquid oxygenates-methanol (MeOH), ethanol (EtOH), and acetic acid (AtAc) show enhanced formation of MeOH on Cu / Al₂O₃ with plasma-enabled active site regeneration. Selectivity toward gas-phase carbon products, CO, C₂H₆, and C₂H₄, reveals that Ni / Al₂O₃ strongly favors C-C coupling and hydrocarbon formation, whereas Cu / Al₂O₃ suppresses hydrocarbon generation and CO formation, consistent with oxygenate-favoring reaction pathways.

[0086] Fig. 49 shows the experimental setup used to study plasma-assisted catalysis for N₂O abatement. A dielectric barrier discharge was used to form plasma near a porous packed bed of CU / Al2O3polycrystalline catalysts. Reactant gas mixtures, including combinations of N₂O, H₂, and CH₄, were introduced into the reactor using mass flow controllers. Process performance was evaluated by analyzing the composition of downstream products using gas chromatography-mass spectrometry. Input energy costs and plasma properties were characterized using voltage and current probes, along with optical emission spectroscopy (OES).

[0087] Figs. 50A-50B show conversion of pure N₂O under plasma catalytic conditions (p = 1 atm, Teas = 360 K, Tvib = 2750 K) using polycrystalline Cu / Al₂O₃ as the active transition metal catalyst and Al₂O₃ as a control, measured after 40 min of plasma exposure (Fig. 50A). Comparable conversion was measured for both polycrystalline Cu and the control, indicating no enhancement from the presence of Cu. (Fig. 50B) Measurements of time-resolved deactivation following catalyst pretreatment in an H2 plasma (4 kV, 1.3 W, 30 min). The Cu / Al₂O₃ catalyst exhibited rapid deactivation over ~90 s of exposure to plasma, whereas the Al₂O₃ control maintained stable conversion.

[0088] Figs. 51A-51D: DFT calculated snapshots of N₂O decomposition on a Cu(111) surface in reference to molecular ground state N2O at p = 1 atm, Tsur = Teas = 360 K, showing representative bonding configurations along the reaction pathway (Fig. 51A). (Fig.

[0089] 51B) The activation energies (Ea) and free energy changes (AG) plotted for steps 1, 2, and 3 are (0.47, -1.83 eV), (2.34, 2.33 eV), and (0.87, 0.87 eV), respectively. The recombination of surface-bound oxygen atoms (O) to form adsorbed O2 (O2*) is the rate-limiting step at Tsur = 360 K and p = 1 atm. While gas phase reaction pathways are influenced by plasma conditions, surface reaction rates are governed by the physicochemical properties and theAttorney Docket No. 10046-649W01

[0090] state of the catalyst. (Figs. 51C-51D) XPS spectra of Cu-based polycrystalline catalysts before and after plasma exposure for 30 min, with fitted components corresponding to Cu(II) and Cu(0) / Cu(I) oxidation states. After plasma treatment in N2O for 30 min, the Cu(0) / Cu(I) surface fraction decreased from 43 to 1.34%, indicating a near-complete oxidation to Cu(II), which becomes the predominant surface species.

[0091] Figs. 52A-52B show mitigation of catalytic deactivation of polycrystalline Cu by adding H₂ as a co-reactant at mild thermodynamic conditions. (Fig. 52A) N₂O conversion under plasma catalytic conditions (p = 1 atm, TGas= 360 K, TVib= 2750 K) with increasing H₂ mole fractions (up to 5%). The introduction of H2 enhanced performance, resulting in up to a 2× increase in conversion and a 60% reduction in energy cost for N₂O conversion. (Fig.

[0092] 52B) N₂O conversion at higher H₂ concentrations, up to 85% by volume, beyond the lower and upper explosive limits (LEL and UEL) for N₂O / H₂ mixtures. While conversion efficiency continues to rise with increasing H₂ content, the energy cost plateaus, indicating diminishing returns at high H₂ loading.

[0093] Figs. 53A-53B show mitigation of catalytic deactivation of polycrystalline Cu by Water formation was monitored over a 2 h period to determine adding CH₄ as a co-reactant under mild thermodynamic conditions. (Fig. 53A) N₂O conversion under plasma catalytic conditions (p = 1 atm, TGas= 360 K, TVib= 2750 K) with increasing CH₄ mole fractions, up to 80% by volume, outside the lower and upper explosive limits (LEL and UEL.) for N2O / CH4 mixtures. Conversion increased with CH₄ addition; however, energy costs plateaued around 16 g / kWh at higher CH4 loadings. (Fig. 53B) Carbon balance and product selectivity for 50% CH4 by volume showed that the major carbon-containing products were C₂H₆ (41%), CO₂ (28%), C₂H₄ (17%), and C₂H₂ (7%), with C2 hydrocarbons representing valuable industrial compounds.

[0094] Figs. 54A-54C show mechanistic insights into O removal through water formation using polycrystalline Cu catalysts. (Fig. 54A) Time-resolved measurements of H2O formation during sequential exposure to an N₂O plasma, H₂ purge, and H₂ plasma over polycrystalline Cu at p = 1 atm, TGas= 360 K. Mass spectrometry signals showed no water formation during the N2O plasma or H2 purge phases. Upon ignition of the H₂ plasma, a sharp increase in the H₂O signal was observed, followed by a decay to baseline, indicating the removal of surface-bound O* as H₂O. (Fig. 54B) Integrated II2O signals for Cu / Al₂O₃ and Al₂O₃-only catalysts. The Cu catalyst produced 1.05 μmol of H₂O, while the Al₂O₃ control produced 0.2 μmol. The difference (0.85 μmol) is attributed to O* species associated with Cu sites, corresponding to the removal of ~ 1 / 3 of the initially available active sites.Attorney Docket No. 10046-649W01

[0095] (Fig. 54C) Proposed mechanism for O* accumulation and its subsequent removal via water formation.

[0096] Figs. 55A-55D: DFT calculated reaction pathway of N2O decomposition on a Cu(lll) surface in reference to molecular ground state N₂O at p = 1 atm, TSur= TGas~ 360 K with the addition of H2 as a co-reactant. (Fig. 55 A) Multiple pathways are feasible with O₂ formation, H₂ surface recombination (orange), and water formation and dissociation, being the most favorable. Activation and total energies for reactions (in eV): Rxn 1 (0.47, -1.83), Rxn 4 (0.34, -0.53), Rxn 2 (2.34, 2.33), Rxn 3 (0.87, 0.87), Rxn 5 (0.87, 0.53), Rxn 6 (0.72, -0.62), Rxn 7 (1.05, -0.15), Rxn 8 (0.17, 0.17). (Fig. 55B) XPS surface characterization focusing on the Cu region of catalyst samples for before reactions, post N2O plasma (30 min), post H2plasma cleaning (30 min), and post N₂O / H₂ (5% H₂ by vol) plasma (30 min). Pretreatment of Cu surfaces with plasma-assisted H2 cleaning produced similar XPS features for both the reference and the N2O / H2 plasma-treated samples, with the latter showing the emergence of Cu(OH)₂ species, absent in prior states, and a reduction in Cu(II) compared to exposure to pure N O plasma. (Fig. 55C) Quantified XPS areas, the use of II 2 in the gas stream reduces the amount of Cu (II) states by 25% and provides evidence for the predicted mechanism with the Cu(OH)₂ signal. (Fig. 55D) Oxygen and water production as a function of H2%, as the H2 percentage increases, oxygen is no longer formed, and the mechanism for O* removal is shifted to water production. Mechanism for water formation on the catalyst surface compared to O2 production leading to deactivation.

[0097] Figs. 56A-56E: DFT calculated reaction pathway of N2O decomposition on a Cu(111) surface in reference to the molecular ground state N₂O at p = 1 atm, TSur= TGas= 360 K with the addition of CH4 as a co-reactant. Multiple pathways are possible with the formation of C2 hydrocarbons specific to the use of methane as a co-reactant (Fig. 56A). (Fig. 56B) XPS spectra of the Cu 2p region for reference, post-N2O plasma (30 min), and post-N2O / CH4 plasma- treated samples (30 min). Introduction of CH4 shifted the Cu binding energy, and peak fitting revealed significant Cu(0 / I) character and a reduction in Cu(II), indicating removal of surface O* species. (Fig. 56C) Quantified XPS peak areas, showing an increase in Cu(0 / I) states after CH4 plasma treatment, suggesting regeneration of active sites for continued N2O decomposition. (Fig. 56D-56E) XPS spectra of the C 1 s region, fitted with peaks corresponding to C-C, C-O-C, and O-C=O bonding environments. The reference sample features predominant C-C species attributed to adventitious carbon from the atmosphere. N2O plasma exposure enhances C-O-C contributions, reflecting oxygen incorporation from N2O. Following N2O / CH4 plasma exposure, a more balancedAttorney Docket No. 10046-649W01

[0098] distribution of carbon species is observed, consistent with the formation of C₂ hydrocarbons (C-C) and oxygenates (C-O-C, O-C=O), in agreement with detected gas-phase products.

[0099] DETAILED DESCRIPTION

[0100] The materials, compounds, compositions, articles, and methods described herein may be understood more readily by reference to the following detailed description of specific aspects of the disclosed subject matter and the Examples and Figures included therein.

[0101] Before the present materials, compounds, compositions, and methods are disclosed and described, it is to be understood that the aspects described below are not limited to specific synthetic methods or specific reagents, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular aspects only and is not intended to be limiting.

[0102] Also, throughout this specification, various publications are referenced. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which the disclosed matter pertains. The references disclosed are also individually and specifically incorporated by reference herein for the material contained in them that is discussed in the sentence in which the reference is relied upon.

[0103] General Definitions

[0104] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill in the art to which this disclosure belongs.

[0105] The term “comprising” and variations thereof as used herein is used synonymously with the term “including” and variations thereof and are open, non-limiting terms.

[0106] Although the terms “comprising” and “including” have been used herein to describe various aspects, the terms “consisting essentially of’ and “consisting of’ can be used in place of “comprising” and “including” to provide for more specific aspects and are also disclosed. Throughout the description and claims of this specification the word “comprise” and other forms of the word, such as “comprising” and “comprises,” means including but not limited to, and is not intended to exclude, for example, other additives, components, integers, or steps.

[0107] As used in the specification and claims, the singular form “a”, “an”, and “the” include plural references unless the context clearly dictates otherwise. For example, the term “a material” includes a plurality of materials, including mixtures thereof.Attorney Docket No. 10046-649W01

[0108] As used herein, the terms "may," "optionally," and "may optionally" are used interchangeably and are meant to include cases in which the condition occurs as well as cases in which the condition does not occur. Thus, for example, the statement that a formulation "may include a hydrocarbon" is meant to include cases in which the formulation includes a hydrocarbon as well as cases in which the formulation does not include a hydrocarbon.

[0109] It is understood that throughout this specification, the identifiers "‘first” and “second” are used solely to aid in distinguishing the various components and steps of the disclosed subject matter. The identifiers “first” and “second” are not intended to imply any particular order, amount, preference, or importance to the components or steps modified by these terms.

[0110] By “reduce,” or other forms of the word, such as “reducing” or “reduction,” is meant the lowering of an event or characteristic. It is understood that this is typically in relation to some standard or expected value; in other words, it is relative, but it is not always necessary for the standard or relative value to be referred to.

[0111] By “increase” or other forms of the word, such as “increasing,” is meant raising or elevating. It is understood that this is typically in relation to some standard or expected value; in other words, it is relative, but it is not always necessary for the standard or relative value to be referred to.

[0112] “Plasma” refers to a state of matter comprising ionized gas with positive ions and electrons in any proportion. A non-thermal plasma (also known as cold plasma or non¬ equilibrium plasma) is a type of plasma that is not in thermodynamic equilibrium, where the electrons are at a much higher temperature compared to the ions and neutral atoms. Non-thermal plasmas can be created by, for example, dielectric barrier discharge, diffused plasma discharge, corona discharge, or plasma jets. A thermal plasma (also known as hot plasma) is a type of plasma where the temperatures of the electrons, ions, and neutral particles are nearly equal, meaning that the plasma is in thermodynamic equilibrium.

[0113] Thermal plasmas can be generated at high temperatures by, for example, electric arcs or inductively coupled plasma.

[0114] Reference will now be made in detail to specific aspects of the disclosed materials, compounds, compositions, articles, and methods, examples of which are illustrated in the accompanying Examples and Figures.

[0115] Methods

[0116] The subject matter disclosed herein demonstrates how non-thermal (non-Attorney Docket No. 10046-649W01

[0117] equilibrium) plasma and catalysts can be codesigned to produce various products by converting molecules like CH4 into value-added products like H2, ethylene, acetylene, and oxygenated hydrocarbons like methanol, ethanol, and acetic acid (when co-reactants are present) at mild thermodynamic conditions. The disclosed subject matter can also be used to limit coking or catalyst deactivation. As an example, Figs. 1A-1C show the direct conversion of CH₄. The production of oxygenated hydrocarbons is shown in Fig. 35.

[0118] Electrical energy induces vibrational excitation in a gaseous starting material (e.g., CH4), promoting dissociative adsorption on the metal catalyst by lowering the activation energy barrier. This permits operation at reduced temperatures and pressures, breaking catalytic scaling laws by shifting the optimal catalyst to more noble surfaces and expanding the design space for plasma-catalysis.

[0119] In reference to the direct conversion of CH4, by employing model experiments and a microkinetic model, a packed bed dielectric barrier discharge (DBD) achieves CII4 conversion into hydrogen (H2) at near room temperature (473 K) and pressure (1 atm) without forming CO: 2. Experiments herein show that plasma activation with 20% Ni / Al₂O₃ produces 45 times more H₂ than gas-phase plasma alone (Al₂O₃ supports only). Also shown is how independent control of plasma properties (vibrational temperature (Tvib)), catalyst choice, surface temperature (Tsui), and residence timescales (^residence) allows selective formation of upgraded C2 hydrocarbons by altering the energies of catalytic branching reactions for C2 hydrocarbons. Ethene (C2H4) and ethane (C2H6) are the primary upgraded C2 hydrocarbons on 20% Pt / Al₂O₃ and 20% Cu / Al₂O₃ at Tsur= 473 K and Tvib= 4200 K, respectively. It is shown herein that subsequent catalytic reactions (R2-R10) are set by surface and gaseous state properties (dependent of Tsur), independent of plasma properties (which controls R1 dependent on Tvib), allowing the use of noble catalysts for sustained 11 ■ and C2 hydrocarbon production. Noble catalysts, like Cu / Al₂O₃, mitigate coking by facilitating CH₃* recombination on the surface while maintaining stable H₂ and C₂H₆ production for over 15 hours. Experiments show that while 20% Ni / Al₂O₃ produces more H₂ initially, it deactivates rapidly due to coking. Simulations and ex-situ surface analysis (SEM / EDS, XPS, TEM / EDX) confirm the correlation between simulated and observed functional groups of the carbon coverage on the catalytic surfaces.

[0120] Plasma-catalysis presents a transformative solution for direct methane conversion by breaking the constraints imposed by the catalytic scaling laws. The synergistic co-design of plasma and catalysts not only lowers energy barriers but also bypasses limitations of conventional methods, opening new avenues for catalyst and process design (Figs. 1A-1C).Attorney Docket No. 10046-649W01

[0121] This permits methane conversion under mild thermodynamic conditions and facilitates waste gas conversion at low temperature and pressure. The ability to selectively control branching reactions in plasma-catalytic systems allows for precise tuning of product distribution, enabling the upgrade of methane into higher- value hydrocarbons. Plasma activation initiates the reaction by overcoming the rate -limiting step, while the rate of subsequent steps are set by surface properties (i.e., independent of plasma properties) that enable control over the removal of catalyst poisons (CHX) through recombination. This process facilitates the use of noble catalysts that resist coking. Moreover, low-temperature plasmas can actively modify catalyst surfaces, mitigating coke formation and enhancing catalyst stability over extended operational periods. In summary, plasma-catalysis represents a breakthrough in sustainable chemical processing, offering a greener, more economically viable pathway for direct methane conversion.

[0122] In a first aspect, disclosed are methods of selectively producing a target carbon-containing product during the production of hydrogen gas from a methane-containing gas, the method comprising: (a) forming a non-thermal plasma in the presence of the methane- containing gas and a metal catalyst, (i) wherein the methane-containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (iii) wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature and that dissociatively adsorbs on the metal catalyst; (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, and / or selecting a different metal catalyst; thereby producing hydrogen gas and the target carbon -containing product, wherein the target carbon-containing product is produced in a greater amount than non-target carbon-containing products.

[0123] The target carbon-containing product can be any one of C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, graphene, carbon nanotubes, or fullerenes. As shown herein, by independently selecting or adjusting any one or more of the following parameters, namely, the gas temperature, the surface temperature of the metal catalyst, the methane vibrational temperature, and type of metal catalyst, it is possible to selectively produce one target carbon-containing product over other non-target carbon -containing products. That is, if the target carbon-containing product is:

[0124] (a) C2H6, then the non-target carbon-containing products are C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;

[0125] (b) C₂H₄, then the non-target carbon-containing products are C₂H₆, C₂H₂, carbonAttorney Docket No. 10046-649W01

[0126] black, graphite, graphene, carbon nanotubes, and fullerenes;

[0127] (c) C2H2, then the non-target carbon-containing products are C₂H₆, C₂H₄, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;

[0128] (d) carbon black, then the non-target carbon-containing products are C₂H₆ C2H4, C2H2, graphite, graphene, carbon nanotubes, and fullerenes;

[0129] (e) graphite, then the non-target carbon-containing products are C₂H₆, C₂H₄, C₂H₂, carbon black, graphene, carbon nanotubes, and fullerenes;

[0130] (f) graphene, then the non-target carbon-containing products are C₂H₆ C2H4, C2H2, carbon black, graphite, carbon nanotubes, and fullerenes;

[0131] (g) carbon nanotubes, then the non-target carbon-containing products are C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, graphene, and fullerenes; or

[0132] (h) fullerenes, then the non-target carbon-containing products are C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, graphene, and carbon nanotubes.

[0133] In the disclosed method, the amount of CO2 produced can be limited, e.g., the method produces less than 1% of CO2.

[0134] The non-thermal plasma can be formed by a dielectric barrier discharge, diffused plasma discharge, corona discharge, RF coupled plasma, microwave reactor, or plasma jets. I'he non-thermal plasma can be formed with gas temperatures ranging from 300 K to 800K and vibrational temperatures greater than 1500 K (e.g., 1500 to 5500 K). The non-thermal plasma can be formed at 1 atm. In other examples, the non-thermal plasma can be formed with surface temperatures ranging from 373K to 1000K and vibrational temperatures ranging from 800 K to 6000K.

[0135] The methane-containing gas can be any gas that contains methane. For example, the methane-containing gas can be liquified natural gas. In other examples, the methane- containing gas can be from distributed methane sources, such as from wetland emissions, termite activity, ocean release, natural gas reserves, dumps, pump jacks, or tail gas streams from an industrial process.

[0136] The disclosed methods and devices utilize a metal catalyst. In some examples, the metal catalyst can comprise a transition metal catalyst. In other examples, the metal catalyst can comprise a metal oxide. In other examples, the metal catalyst can comprise a mixture of more than one metal. In still other examples, the metal catalyst can be a bimetallic catalyst. In yet further examples, the metal catalyst can have a binding energy of carbon to a metal atom in the metal catalyst of greater than -0.5 eV. That is, the metal catalyst can be a weakly-binding catalyst. While not wishing to be bound by theory, a weakly bindingAttorney Docket No. 10046-649W01

[0137] catalyst can minimize inactivation. In specific examples, the metal catalyst can be Cu, Ag, Au, or any combination thereof. In other examples, the metal catalyst can be Re, Pt, Rh, Ru, Pd, Ni, Co, Zn, or any combination thereof. The metal catalyst can be on a support, such as a dielectric support selected from the group consisting of Al₂O₃, SiO₂, TiO₂, zeolites, activated carbon, glass, and ceramic.

[0138] The metal catalyst has a surface temperature, which can be adjusted or selected by the practitioner. Heating or cooling the metal catalyst so as to adjust its surface temperature can be performed with electric heaters, lamps, or fans. In some examples, the surface temperature of the metal catalyst can be adjusted to from 300K to 1000 K, e.g., from 400 K to 900 K, from 500 K to 800K, from 600 K to 700K, from 300 K to 800K, from 400 K to 700 K, from 500 K to 600 K, from 300 K to 700 K, from 400 K to 600 K, from 300 K to 600K, from 400 K to 500 K, from 300 K to 500 K, from 300 K to 400 K, from 400 K to 1000K, from 400 K to 800 K, from 500 to 1000 K, from 500 K to 900 K, from 500 K to 700 K, from 600 K to 1000 K, from 600 K to 900 K, from 600 K to 800K, from 700 K to 1000K, from 700 K to 900 K, from 700 K to 800 K, from 800 K to 1000 K, from 800 K to 900 K, or from 900 K to 1000 K. In specific examples, the surface temperature of the metal catalyst can be 300 K, 350 K, 400 K, 450 K, 500 K, 550 K, 600 K, 650 K, 700 K, 750 K, 800 K, 850 K, 900 K, 950 K, or 1000 K, where any of the stated values can be an upper or lower endpoint of a range.

[0139] The methane- containing gas has a gas temperature, which can be adjusted or selected by the practitioner. Heating or cooling the methane containing gas so as to adjust its gas temperature can be performed with electric heaters, lamps, fans, or heat exchangers. In some examples, the gas temperature is adjusted to from 300K to 800K, e.g., from 400 K to 700 K, from 500 K to 600 K, from 400 K to 800 K, from 300 K to 700 K, from 400 K to 600 K, from 300 K to 600 K, from 400 K to 500 K, from 500 K to 800 K, from 500 K to 700 K, from 600 K to 800 K, or from 600 K to 700 K. In specific examples, the gas temperature can be 300 K, 350 K, 400 K, 450 K, 500 K, 550 K, 600 K, 650 K, 700 K, 750 K, or 800 K, where any of the stated values can be an upper or lower endpoint of a range.

[0140] In still other examples, the vibrational temperature of the reactants, e.g., methane, can be adjusted or selected. The vibrational temperature of the reactant can be from 1500 K to 5500 K, e.g., from 2000 K to 5000 K, from 2500 K to 4500 K, from 3000 K to 4000 K, from 2000 K to 5500 K, from 1500 to 5000 K, from 2000 K to 4500 K, from 2500 K to 4000 K, from 3000 K to 3500 K, from 1500 K to 4500 K, from 2000 K to 4000 K, from 2500 K to 3500 K, from 1500 K to 4000 K, from 2000 K to 3500 K, from 2500 K to 3000Attorney Docket No. 10046-649W01

[0141] K, from 1500 K to 3500 K, from 2000 K to 3000 K, from 1500 K to 3000 K, or from 2000 K to 2500 K. In specific examples, the vibrational temperature of a reactant can be 1500 K, 2000 K, 2500 K, 3000 K, 3500 K, 4000 K, 4500 K, 5000 K, or 5500 K, wherein any of the stated values can form an upper or lower endpoint of a range. While not wishing to be bound by theory, it is believed that high vibrational temperatures can prevent catalytic inactivation.

[0142] In some specific examples, when the metal catalyst is Cu, the surface temperature is greater than 973 K, and the gas temperature is from 300K to 800K, the target carbon- containing product is C2H2. In some other specific examples, when the metal catalyst is Cu, the surface temperature is from 573 to 973 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C2H4. In still some other specific examples, when the metal catalyst is Cu, the surface temperature is less than 573 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C₂H₆.

[0143] In some specific examples, when the metal catalyst is Ni, the surface temperature is greater than 640 K, and the gas temperature is from 300K to 800K, the target carbon- containing product is C H -. In some other specific examples, when the metal catalyst is Ni, the surface temperature is from 473 to 640 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C2H4. In still some other specific examples, when the metal catalyst is Ni, the surface temperature is less than 473 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C₂H₆.

[0144] In more specific examples, wherein the target product is ethane, the vibrational temperature is from 1500 K to 6000K, the surface temperature is from 350K to 450 K, and the catalyst binding energy of from 0.1 eV to 1 eV. In another example, wherein the product is ethene, the vibrational temperature is from 800 K to 6000K, the surface temperature is from 500K to 1000 K, and the catalyst binding energy of from -1.2 eV to -0.8 eV. In another example, wherein H2 is produced at least at 14 %, the vibrational temperature is from 800 K to 6000K, the surface temperature is from 550 K to 1000 K, and the catalyst binding energy of from -1.5 eV to 0 eV. In other examples, wherein acetylene is the target carbon-containing product, the surface temperature is from 600 K to 1000 K, and the catalyst binding energy is from -1.4 eV to -0.2 eV.

[0145] When the catalyst is weakly binding (Ec > 0 eV), higher II2 yields can be achieved at higher surface temperature (> 550 K). When the catalyst is strongly binding (Ec < -1 eV), higher II2 yields can be achieved at even at lower surface temperatures (> 373 K). Cu achieves lower II 2 yields at Tvib / TSUr of from -11-12. Ni achieves higher II - yields atAttorney Docket No. 10046-649W01

[0146] Tvib / Tsurat from ~7.5–9.5.

[0147] Carbon accumulation can occur most severely at low vibrational temperature (e.g., from 800 K to 2000 K), strongly binding catalyst (e.g., Ec < -1.0 eV). Weakly binding (Ec > 0.2 eV) can reduce carbon accumulation.

[0148] In some specific examples, when the metal catalyst is Cu and it is not coked or deactivated at a surface temperature of less than 500 K and gas temperature of greater than 2000K for up to 840 minutes.

[0149] Electrical energy is used to heat the methane-containing gas, the metal catalyst, and / or make the non-thermal plasma.

[0150] Use of co-reactants

[0151] In a second aspect, disclosed herein is a method of selectively producing a target carbon-containing product during the production of a liquid fuel from a methane-containing gas and a vibrationally active oxygen-containing co-reactant, the method comprising: (a) forming a non-thermal plasma in the presence of the methane-containing gas, the vibrationally active oxygen-containing co-reactant, and a metal catalyst; (i) wherein the methane-containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (iii) wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature and vibrationally excited oxygen-containing co-reactant having a oxygen-containing co-reactant vibrational temperature that dissociatively adsorb on the metal catalyst; (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, adjusting the oxygen-containing co-reactant vibrational temperature, and / or selecting a different metal catalyst, thereby producing a liquid fuel and the target carbon-containing product, wherein the target carbon-containing product is produced in a greater amount than non-target carbon-containing products. The method can further produce water and / or hydrogen.

[0152] The vibrationally active oxygen-containing co-reactant can be CO2, CO, NO2, and / or O2.

[0153] In another aspect, the reactants can also include a hydrogen-containing co-reactant, such as H₂ and / or H₂O.

[0154] The target carbon-containing product can be any one of methanol, ethanol, acetic acid, C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, or fullerenes. As shown herein, by independently selecting or adjusting any one or more of the following parameters, namely, the gas temperature, the surface temperature of the metal catalyst, theAttorney Docket No. 10046-649W01

[0155] methane vibrational temperature, and type of metal catalyst, it is possible to selectively produce one target carbon-containing product over other non-target carbon-containing products. That is, if the target carbon-containing product is:

[0156] (a) methanol, then the non-target carbon-containing products are ethanol, acetic acid C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;

[0157] (b) ethanol, then the non-target carbon-containing products are methanol, acetic acid C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, graphene, carbon nanotubes, and fullerenes; or (c) acetic acid, then the non-target carbon-containing products are methanol, ethanol, C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, and fullerenes.

[0158] The non-thermal plasma can be formed by a dielectric barrier discharge, diffused plasma discharge, corona discharge, RF coupled plasma, microwave reactor, or plasma jets. The non-thermal plasma can be formed with gas temperatures ranging from 300 K to 800K and vibrational temperatures greater than 1500 K (e.g., 1500 to 5500 K). The non-thermal plasma can be formed at 1 atm.

[0159] The methane-containing gas can be any gas that contains methane. For example, the methane-containing gas can be liquified natural gas. In other examples, the methane- containing gas can be from distributed methane sources, such as from wetland emissions, termite activity, ocean release, natural gas reserves, dumps, pump jacks, or tail gas streams from an industrial process.

[0160] The disclosed methods and devices utilize a metal catalyst. In some examples, the metal catalyst can comprise a transition metal catalyst. In other examples, the metal catalyst can comprise a metal oxide. In other examples, the metal catalyst can comprise a mixture of more than one metal. In still other examples, the metal catalyst can be a bimetallic catalyst. In yet further examples, the metal catalyst can have a binding energy of carbon to a metal atom in the metal catalyst of greater than -0.5 eV. In specific examples, the metal catalyst can be Cu, Ag, Au, or any combination thereof. In other examples, the metal catalyst can be Re, Pt, Rh, Ru, Pd, Ni, Co, Zn, or any combination thereof. The metal catalyst can be on a support, such as a dielectric support selected from the group consisting of Al₂O₃, SiO₂, TiO₂, zeolites, activated carbon, glass, and ceramic.

[0161] The metal catalyst has a surface temperature, which can be adjusted or selected by the practitioner. Heating or cooling the metal catalyst so as to adjust its surface temperature can be performed with electric heaters, lamps, or fans. In some examples, the surface temperature of the metal catalyst can be adjusted to from 300K to 1000 K, e.g., from 400 KAttorney Docket No. 10046-649W01

[0162] to 900 K, from 500 K to 800K, from 600 K to 700K, from 300 K to 800K, from 400 K to 700 K, from 500 K to 600 K, from 300 K to 700 K, from 400 K to 600 K, from 300 K to 600K, from 400 K to 500 K, from 300 K to 500 K, from 300 K to 400 K, from 400 K to 1000K, from 400 K to 800 K, from 500 to 1000 K, from 500 K to 900 K, from 500 K to 700 K, from 600 K to 1000 K, from 600 K to 900 K, from 600 K to 800K, from 700 K to 1000K, from 700 K to 900 K, from 700 K to 800 K, from 800 K to 1000 K, from 800 K to 900 K, or from 900 K to 1000 K. In specific examples, the surface temperature of the metal catalyst can be 300 K, 350 K, 400 K, 450 K, 500 K, 550 K, 600 K, 650 K, 700 K, 750 K, 800 K, 850 K, 900 K, 950 K, or 1000 K, where any of the stated values can be an upper or lower endpoint of a range.

[0163] The methane- containing gas has a gas temperature, which can be adjusted or selected by the practitioner. Heating or cooling the methane-containing gas so as to adjust its gas temperature can be performed with electric heaters, lamps, fans, or heat exchangers. In some examples, the gas temperature is adjusted to from 300K to 800K, e.g., from 400 K to 700 K, from 500 K to 600 K, from 400 K to 800 K, from 300 K to 700 K, from 400 K to 600 K, from 300 K to 600 K, from 400 K to 500 K, from 500 K to 800 K, from 500 K to 700 K, from 600 K to 800 K, or from 600 K to 700 K. In specific examples, the gas temperature can be 300 K, 350 K, 400 K, 450 K, 500 K, 550 K, 600 K, 650 K, 700 K, 750 K, or 800 K, where any of the stated values can be an upper or lower endpoint of a range.

[0164] In still other examples, the vibrational temperature of the reactants, e.g., methane, oxygen-containing co-reactant, or hydrogen-containing co-reactant, can be adjusted or selected. The vibrational temperature of the reactant can be from 1500 K to 5500 K, e.g., from 2000 K to 5000 K, from 2500 K to 4500 K, from 3000 K to 4000 K, from 2000 K to 5500 K, from 1500 to 5000 K, from 2000 K to 4500 K, from 2500 K to 4000 K, from 3000 K to 3500 K, from 1500 K to 4500 K, from 2000 K to 4000 K, from 2500 K to 3500 K, from 1500 K to 4000 K, from 2000 K to 3500 K, from 2500 K to 3000 K, from 1500 K to 3500 K, from 2000 K to 3000 K, from 1500 K to 3000 K, or from 2000 K to 2500 K. In specific examples, the vibrational temperature of a reactant can be 1500 K, 2000 K, 2500 K, 3000 K, 3500 K, 4000 K, 4500 K, 5000 K, or 5500 K, wherein any of the stated values can form an upper or lower endpoint of a range. While not wishing to be bound by theory, it is believed that high vibrational temperatures can prevent catalytic inactivation.

[0165] In some specific examples, when the metal catalyst is Cu, the surface temperature is greater than 973 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C₂H₂. In some other specific examples, when the metal catalyst is Cu,Attorney Docket No. 10046-649W01

[0166] the surface temperature is from 573 to 973 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C2H4. In still some other specific examples, when the metal catalyst is Cu, the surface temperature is less than 573 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C2H6.

[0167] In some specific examples, when the metal catalyst is Ni, the surface temperature is greater than 640 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C₂H₂. In some other specific examples, when the metal catalyst is Ni, the surface temperature is from 473 to 640 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C2H4. In still some other specific examples, when the metal catalyst is Ni, the surface temperature is less than 473 K, and the gas temperature is from 300K to 800K, the target carbon-containing product is C2H6.

[0168] In some specific examples, when the metal catalyst is Cu and it is not coked or deactivated at a surface temperature of less than 500 K and gas temperature of greater than 2000K for up to 840 minutes.

[0169] In some specific examples, wherein the target carbon-containing product is methanol, the metal catalyst is Cu, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In some examples, wherein the target carbon- containing product is ethanol, the metal catalyst is Cu, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In some examples, wherein the target carbon-containing product is ethane, the metal catalyst is Cu, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In some examples, wherein the target carbon -containing product is ethanol, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In some examples, wherein the target carbon-containing product is acetic acid, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In other examples, wherein the target carbon-containing product is ethane, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In still other examples, wherein the target carbon-containing product is ethene, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K. In any of these examples the Cu or Ni catalyst can be on an Al₂O₃ support. The ratio of CH₄ to CO₂ co-reactant can be 1:3. In other examples, the ratio of CH₄ to CO₂ and H₂ co-reactants can be 1:1:4 to 1:1:1.

[0170] Electrical energy is used to heat the methane-containing gas, the metal catalyst, and / or make the non-thermal plasma.Attorney Docket No. 10046-649W01

[0171] In certain examples, the flow rate can be from 15 seem to 60 seem. Higher methanol selectivity can be observed at 30 seem. Higher ethanol and acetic acid selectivity can be observed at from 30 to 50 seem.

[0172] N2O abatement

[0173] In a third aspect, the techniques disclosed herein relate to methods of removing nitrous oxide from a gas stream, by (a) forming a non-thermal plasma in the presence of a nitrous oxide-containing gas, an additional co-reactant, and a metal catalyst, (i) wherein the nitrous oxide-containing gas has a gas temperature, (ii) wherein the metal catalyst has a surface temperature, and (iii) wherein the non-thermal plasma produces vibrationally excited nitrous oxide having a nitrous oxide vibrational temperature, a vibrationally excited co-reactant having a co-reactant vibrational temperature, and that each dissociatively adsorbs on the metal catalyst; and (b) adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the nitrous oxide vibrational temperature, adjusting the co-reactant vibrational temperature, and / or selecting a different metal catalyst; thereby removing nitrous oxide from the gas stream, and producing nitrogen and oxygen gases.

[0174] The co-reactant can be CH₄, CO₂, and / or H₂. For example, the co-reactant can be a 1:1:4 mixture of CH₄, CO₂, and H₂. The use of a co-reactant here, e.g., like H. can be targeted to clean the surfaces in-situ. Without this, then weakly binding or strongly binding catalysts can either get poisoned by oxygen or coke. Thus, the plasma can activate a coreactant and clean it in-situ.

[0175] In these methods, the method can additionally produce C₂H₆, C₂H₄, C₂H₂, methanol, ethanol, acetic acid, carbon black, graphite, graphene, carbon nanotubes, or fullerenes.

[0176] The non-thermal plasma can be formed by a dielectric barrier discharge, diffused plasma discharge, corona discharge, RF coupled plasma, microwave reactor, or plasma jets. The non-thermal plasma can be formed with gas temperatures ranging from 300 K to 800K and vibrational temperatures greater than 1500 K (e.g., 1500 to 5500 K). The non-thermal plasma can be formed at 1 atm.

[0177] The nitrous oxide-containing gas can be any gas that contains nitrous oxide.

[0178] The disclosed methods and devices utilize a metal catalyst. In some examples, the metal catalyst can comprise a transition metal catalyst. In other examples, the metal catalyst can comprise a metal oxide. In other examples, the metal catalyst can comprise a mixture of more than one metal. In specific examples, the metal catalyst can be Cu, Ag, Au, or any combination thereof. In other examples, the metal catalyst can be Re, Pt, Rh, Ru, Pd, Ni,Attorney Docket No. 10046-649W01

[0179] Co, Zn, or any combination thereof. The metal catalyst can be on a support, such as a dielectric support selected from the group consisting of Al₂O₃, SiO₂, TiO₂, zeolites, activated carbon, glass, and ceramic. In other examples, the dielectric support can be used without a transition metal catalyst.

[0180] The metal catalyst has a surface temperature, which can be adjusted or selected by the practitioner. Heating or cooling the metal catalyst so as to adjust its surface temperature can be performed with electric heaters, lamps, or fans. In some examples, the surface temperature of the metal catalyst can be adjusted to from 300K to 1000 K, e.g., from 400 K to 900 K, from 500 K to 800K, from 600 K to 700K, from 300 K to 800K, from 400 K to 700 K, from 500 K to 600 K, from 300 K to 700 K, from 400 K to 600 K, from 300 K to 600K, from 400 K to 500 K, from 300 K to 500 K, from 300 K to 400 K, from 400 K to 1000K, from 400 K to 800 K, from 500 to 1000 K, from 500 K to 900 K, from 500 K to 700 K, from 600 K to 1000 K, from 600 K to 900 K, from 600 K to 800K, from 700 K to 1000K, from 700 K to 900 K, from 700 K to 800 K, from 800 K to 1000 K, from 800 K to 900 K, or from 900 K to 1000 K. In specific examples, the surface temperature of the metal catalyst can be 300 K, 350 K, 400 K, 450 K, 500 K, 550 K, 600 K, 650 K, 700 K, 750 K, 800 K, 850 K, 900 K, 950 K, or 1000 K, where any of the stated values can be an upper or lower endpoint of a range.

[0181] The nitrous oxide-con taini ng gas has a gas temperature, which can be adjusted or selected by the practitioner. Heating or cooling the nitrous oxide-containing gas so as to adjust its gas temperature can be performed with electric heaters, lamps, fans, or heat exchangers. In some examples, the gas temperature is adjusted to from 300K to 800K, e.g., from 400 K to 700 K, from 500 K to 600 K, from 400 K to 800 K, from 300 K to 700 K, from 400 K to 600 K, from 300 K to 600 K, from 400 K to 500 K, from 500 K to 800 K, from 500 K to 700 K, from 600 K to 800 K, or from 600 K to 700 K. In specific examples, the gas temperature can be 300 K, 350 K, 400 K, 450 K, 500 K, 550 K, 600 K, 650 K, 700 K, 750 K, or 800 K, where any of the stated values can be an upper or lower endpoint of a range.

[0182] In still other examples, the vibrational temperature of the reactants, e.g., nitrous oxide or co-reactant can be adjusted or selected. The vibrational temperature of the reac tant can be from 1500 K to 5500 K, e.g., from 2000 K to 5000 K, from 2500 K to 4500 K, from 3000 K to 4000 K, from 2000 K to 5500 K, from 1500 to 5000 K, from 2000 K to 4500 K, from 2500 K to 4000 K, from 3000 K to 3500 K, from 1500 K to 4500 K, from 2000 K to 4000 K, from 2500 K to 3500 K, from 1500 K to 4000 K, from 2000 K to 3500 K, fromAttorney Docket No. 10046-649W01

[0183] 2500 K to 3000 K. from 1500 K to 3500 K, from 2000 K to 3000 K, from 1500 K to 3000 K, or from 2000 K to 2500 K. In specific examples, the vibrational temperature of a reactant can be 1500 K, 2000 K, 2500 K, 3000 K, 3500 K, 4000 K, 4500 K, 5000 K, or 5500 K, wherein any of the stated values can form an upper or lower endpoint of a range. While not wishing to be bound by theory, it is believed that high vibrational temperatures can prevent catalytic inactivation.

[0184] Device

[0185] Also disclosed herein is a device for producing a product from a gaseous reactant, comprising: (a) a reaction chamber having an inlet for the gaseous reactant and outlet for the product; (b) a valve connected to the inlet that controls flow rate of gaseous reactant into the reaction chamber and a valve connected to the outlet that controls flow rate of product out of the chamber; (c) two electrodes, wherein the reaction chamber is positioned between the two electrodes, (d) an electrical energy supply connected to the two electrodes to supply sufficient electrical energy to produce a non-thermal plasma from the gaseous reactants; (e) a metal catalyst positioned inside the reaction chamber; and (f) an optional electrical heater connected to the device to heat the reaction chamber, heat the metal catalyst, and / or heat the gaseous reactant. In specific examples, the catalyst can be on a film in the device, or alternatively can be in a packed bed, which in this case can be a transition metal heterogeneous catalyst.

[0186] The device can be located near distributed sources for the reactants. For example, the device can be in an automobile where methane can be used to generate hydrogen.

[0187] Indeed, the device can be used in any application where methane is used as a storage vector for hydrogen. The device can be in a plant where methane is flared before being released to the atmosphere. In this example, the methane gas can be passed through the device and converted into hydrogen or liquid fuels. Further, the device can be near, connected to an untreated tail gas stream. The device can be used in, or the disclosed methods can use, methane -containing gas streams from well jacks.

[0188] EXAMPLES

[0189] To further illustrate the principles of the present disclosure, the following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how the compositions, articles, and methods claimed herein are made and evaluated. They are intended to be purely exemplary of the invention and are not intended to limit the scope of what the inventors regard as their disclosure. Efforts have been made to ensure accuracy with respect to numbers (e.g., amounts, temperatures, etc.); however, someAttorney Docket No. 10046-649W01

[0190] errors and deviations should be accounted for. Unless indicated otherwise, temperature is °C or is at ambient temperature, and pressure is at or near atmospheric. There are numerous variations and combinations of process conditions that can be used to optimize product quality and performance. Only reasonable and routine experimentation will be required to optimize such process conditions.

[0191] Example 1:

[0192] Limitations of thermal catalysis for direct CH4 utilization

[0193] Direct methane conversion is currently realized by a thermal catalytic (pyrolysis) pathway. To illustrate the challenges of conventional direct conversion pathways, a global microkinetic model was developed to mimic the direct conversion of CH4 to H2 and C2 hydrocarbons in a thermal catalytic pathway (Fig. 2A). The model contains 10 elementary reactions (R1-R10) that tracks the conversion of CH4 into adsorbates (i.e., CH*, CH2*, CH *. C*. C2II2*, C2H4*, and H*) and stable gaseous products (i.e., H2(g), C2H6(g),

[0194] C2H4(g), and C2H2(g)). Concentration of the species i (partial pressure of gases, pi, and surface coverage of adsorbates, Qi) was solved through a system of coupled ordinary differential equations (ODEs). The model was adapted to represent a flow reactor configuration where fixed conversion of CH4is maintained. Direct conversion of CH4was tracked to evaluate both short-term and long-term effects on the production of H2 and C2 hydrocarbons (Figs.

[0195] 2C-2E). The model was initialized at t = 0 with 100% CH4 in the reaction volume, pressure was maintained at 1 atm, and surface temperature (TSUr) was maintained at 873 K to represent methane pyrolysis conditions. The model assumed a uniform catalyst surface with identical active sites (~ 4.2 x 1019). Short-term effects were analyzed at 10% initial CH 4 conversion, while long-term effects on select catalytic surfaces were assessed at fixed 10% initial CH4conversion. A residual refinement study ensured converged results for all simulated cases.

[0196] Table 1. Elementary reactions that are considered for the conversion of CH4into H2and C2hydrocarbons. R1-R5 describe an elementary reaction pathway for methane dehydrogenation into H2(g). R6-R10 describe branching reactions where CHx* can recombine on catalysts to form different C2 hydrocarbons.Attorney Docket No. 10046-649W01

[0197] Reaction Number Reaction

[0198] _ RI _ Cvipm 4 2#* w CH* py

[0199] R2 cm 4 r 0 cm 4- m

[0200] R3 cm 4- r cm 4- m _ R4 _ cm % m c 4 m _

[0201] R5 2m T

[0202] R6 2cm

[0203] R7 2cm mm 4 r

[0204] RS 2cm mm 4 ir

[0205] R9 mm * cMm + r

[0206]

[0207] RIO mm m m ($) m Reaction rates are controlled by rate constants for forward and back ward reaction (kif, kib). These rate constants depend on the Tsur, activation energies (AEa.a), and reaction energies (AH RS) of an elementary reaction Ri. AE'^ / and AWa are con-elated with the binding energy of carbon (A£2) on transition metal catalysts through linear Brpnsted-Evans-Polanyi (BEP) relationships (scaling laws) (Hummelsh0j, J. S. et al. CatApp: A Web Application for Surface Chemistry and Heterogeneous Catalysis. Angewandte Chemie 124, 278-280 (2012); Engelmann, Y. et al. Predicted Influence of Plasma Activation on Nonoxidative Coupling of Methane on Transition Metal Catalysts. ACS Sustain Chem Eng 8, 6043-6054 (2020); Wang, S. et al. Universal transition state scaling relations for (de)hydrogenation over transition metals. Phys Chem Chem Phys 13, 20760 (2011)). The scaling laws used in these examples were derived from density functional theory (DFT)-calculated energies on 111 and 211 crystal planes for direct methane conversion reported in the literature (Fig. 2B). These scaling relations establish a theoretical basis for generating volcano plots, which are used in choosing catalysts that optimize the balance between reactant adsorption and product desorption rates. Production rates of gaseous products (turnover frequency (TOE)) on 111 and 211 sites are normalized to the initial number of active sites.

[0208] Under representative pyrolytic conditions (Tsur= 873 K), short-term effects at 10% initial CFU conversion reveal a maximum H2 TOP of ~ 2.5 x 105site'1s'1on 111 sites for AEc< 0.5 eV (Fig. 2C). H2 TOE on 211 sites is found to be consistently lower than H2 TOP on 111 sites (Fig. 2D). Production of C2 hydrocarbons is negligible (TOFs « 1 site'1s"1), with C2H4 being the sole upgraded C2 hydrocarbon form on all evaluated transition metal catalysts (e.g., Pt, Rh, Ni, etc.), suggesting inadequate surface CHx* recombination in the branching reactions (R6-R10). Long-term effects observed on Ni (111 ) at Tsur= 873 K, representing an extrapolated scenario of a flow reactor with a fixed 10% initial CILtAttorney Docket No. 10046-649W01

[0209] conversion, demonstrate rapid deactivation of 100% of initial active sites ( ~ 100minutes) due to accumulation of coke (CHx*) on the catalytic surfaces over time (Fig. 2E).

[0210] The thermal catalytic pathway for direct methane conversion presents significant challenges. High operating temperatures (e.g., Tsur~ 873 K) are required to achieve meaningful H2yields, which limit the ability to utilize CH4 locally over the scales it is produced. Additionally, catalyst selection for optimal H2production is inherently constrained by scaling laws, which limit the types of transition metals that can be used. While generating H2, this process lacks a mechanism to utilize the accumulating surface CHx* for producing value-added C2 hydrocarbons, nor does it offer control over the specific type of C2 hydrocarbons formed. The rapid deactivation of the catalytic surface requires either frequent catalyst replacement (every few minutes) or in-situ regeneration methods, often involving high- temperature reduction (> 1200 K) to remove surface CHx*.

[0211] These issues highlight the need for an alternative solution that enables on-demand methane conversion to increase the production rates of H2and C2hydrocarbons, with selectivity control over the desired C2products. Distributed methane utilization will need processes that can operate at reduced target temperatures, near atmospheric pressure, and exhibit resistance to coke formation for sustainable H2and C2hydrocarbon production.

[0212] Breaking catalytic scaling laws for direct CH4 con version

[0213] Low-temperature plasmas exhibit high electron and vibrational temperatures yet maintain a low gas temperature. This is due to the rapid movement of electrons, which efficiently excite vibrational modes but not translational modes. Plasma-induced vibrational excitations in reactants effectively lower the activation energy of the rate-limiting step, enhancing reaction rates and enabling new reaction pathways. Experiments were performed to study the plasma-catalytic direct conversion of methane to H2under target conditions (Tsur= 300 K - 500 K, p = 1 atm). A dielectric barrier discharge (DBD) plasma was generated to induce vibrational excitations in methane molecules and drive reactions within a modular reactor (Fig. 19). The DBD plasma source operated at a frequency of ~ 20 kHz between a pair of 1 mm thick glass dielectrics separated by 2.5 mm, achieving a discharge voltage (V) of ~ 5 - 15 kV and a peak current (I) of ~ 200 mA (Figs. 20A-20B). Specific energy input (SEI) per molecule was calculated by dividing average power ( P ) by the initial number of molecules of CH4 (ncH4,d). The reactant composition contained methane only (99.97% purity). A total of 700 mg of each catalyst (Al2O3, 20% (wt) Ni / Al2O3, 20% (wt) Pt / Al2O3, 20% (wt) Cu / Al2O3) was loaded into the reactor in between a pair of 1 mm thick glass dielectrics. Flow rate (Q) was varied (Q = 1.875 - 37.5 seem) to control the SEI andAttorney Docket No. 10046-649W01

[0214] residence timescale ( residence) of gaseous species (reactants + products) in the reaction volume. Methane conversion and product composition were tracked for different flow rates and catalysts using online gas chromatography (GC) (Fig. 22). Optical emission spectroscopy (OES) of methane DBD plasma was used to quantify the vibrational and rotational temperature of CH and C2 (Fig. 27).

[0215] Plasma is assumed to couple to the methane dissociative adsorption step only.

[0216] Electron impact excitation on methane molecules in ground state and collisional relaxation of methane stretch mode (v(l,3)) populates the methane bend mode (nv(2,4)) in the polyads Po-Pn. Plasma activation is modeled by considering 11 (ground + 10 excited) bending mode vibrational excitations (vz and V4 lumped in the 11thpolyad (Pn) are lumped as nv(2,4)(Figs. 7A-7B, Table 1)) for dissociation of CH4 on catalyst surfaces (Fig. 3A) (Butterworth, T. et al. Plasma induced vibrational excitation of CH4- A window to its mode selective processing. Plasma Sources Sci Technol 29, (2020)). Stretch mode of methane (v(l,3)) and radical contributions are neglected due to their negligible relaxation timescales and diffusion length scales (Figs. 9-10, Table 2) (Maitre, P. A. et al. Plasma-Catalysis of Nonoxidative Methane Coupling: A Dynamic Investigation of Plasma and Surface Microkinetics over Ni(111). J Phys Chem C 126, 19987-20003 (2022); Juurlink, L. B. F. et al. Comparative Study of Stretch and Bend Vibrations in Methane Activation on Ni(100) and Ni(111). Phys Rev Lett 94, 208303 (2005); Bal, K. M. et al. Quantifying the impact of vibrational nonequilibrium in plasma-catalysis: insights from a molecular dynamics model of dissociative chemisorption. J Phys DAppl Phys 54, 394004 (2021); Nakashima, N. et al. Oscillator Strengths of the Ultraviolet Bands of Hot and Relaxed Methallyl, Allyl, and Methyl Radicals. Laser Chem 7, 177-196 (1987); Somers, W. et al. Plasma Species Interacting with Nickel Surfaces: Toward an Atomic Scale Understanding of Plasma- Catalysis. J Phys Chem C 116, 20958-20965 (2012)).

[0217] Table 2. Combination of ethane bending mode quantum numbers in polyad levels 1-10.Attorney Docket No. 10046-649W01

[0218] NotaiiaH

[0219] £ '^1 Bending aside Qsaaiitnt

[0220] VC.4) i: (i, C) (o,l); 2x0,4} 2 j _ (2, ti), fl. '■), % 2) _ | ""3Ecur

[0221] 402.4} 4 (4, 01. (3. 1). (2,270, A. (t>.4)

[0222] 5YCM) 5 1 (5. Oi. (4, 17 (3, 27 C2. Si. O, 47 (0, 5) | <5v(2.4i _ 6 _ j _ (N _ j 7 1 (7 {>7 (< S. 1). (5, 27 (4, 3f (3.41. (2, 57 (1, UX ~i | Sv(2.4) 8 1 (8, 0i. (7, 1), (6, 2). (5. Si, (4, 4), (3, 5). (2. <5). 0, 7), (0, 8) M2O- 9 1 (9, <17 (g, 1 >. (7, 2i. (6, 3), (5. 47 (4. Si. O, <51, (2, 7>. Cl. &). • 0, 9) |

[0223]

[0224] RM'2.4>

[0225] Table 3. Timescales and diffusion length scales of bend and stretch modes vibrating methane, and methyl radical.

[0226] „ • imtiuion Lengili Scaie ■ lime Sc ale i

[0227] Process ■... ■. ■

[0228] : ft) i (0:

[0229] „ „ i = 12 urn j x&<l= 1.34 us!

[0230] Stretch fe» i. „ i „!

[0231] Beud V-V | | rsi«O.0OM gSRadical i ~ 2.2 gm j 0.05. us

[0232]

[0233] Elementary reaction R1 in thermal catalytic pathway can be expanded into 11 different elementary steps for each vibrational excitation of CH4(R11-R111, Table 3). A generalized two temperature Treanor distribution is developed to populate CH4v(2,4) in polyads P0-P10(Fig. 8). The input parameters, gas temperature (Tgas) and vibrational temperature (Tvib), are obtained from OES measurements, where the emission spectra of CH(A2Δ − X2Π) (425 nm – 435 nm) and C2swan (d3Πg− a3Πa) (510 nm – 520 nm) were fitted with a simulated spectrum using massiveOES (Fig. 3C, Fig. 27) (Vorac, J. et al. State-by-state emission spectra fitting for non-equilibrium plasmas: OH spectra of surface barrier discharge at argon / water interface.. J Phys D Appl Phys 50, 294002 (2017); Vorac, J. et al. Batch processing of overlapping molecular spectra as a tool for spatio-temporal diagnostics of power modulated microwave plasma jet. Plasma Sources Sci Technol 26, 025010 (2017)). The bounds for vibrational temperature (Tvib) of methane (CH4 v(2,4)) are estimated from the measured Tvibof CH and C2 radicals (Fig. 28). Simulations were repeated to capture short-term effects at 10% initial CH4 conversion and long-term effects on select catalytic surfaces at fixed 10% initial CH4 conversion for target conditions (Tgas= 473 K, Tvtb =4200 K, and p = 1 atm).Attorney Docket No. 10046-649W01

[0234] Table 4. Elementary reactions that are considered for the dissociation of vibrationally excited CH4 on catalytic surfaces. V2 and V4 lumped in the nth polyad are lumped as mf2,4)). Stretch mode of methane ( if 1,3)) is not considered and radical contribution to surface reactions are neglected.

[0235] Reaction

[0236] RP CtfJ' (p'i 28‘ CH) T

[0237] Rp Q?) F 2t:r CH; E ir

[0238] RP+ 2ir CH;. y,

[0239] RE CH?{2A}'+ H,.

[0240] RP CH^,(ZA)($) + CH; A ir

[0241] KF+ 2fr CH.

[0242] (<f) + 2<y CH; ~ i-r

[0243] RP (g) 20* CH; 4- H'

[0244] KP+ 2tr <-> CH; *

[0245] Rl19ctCi2A>+ CH.+ H, Rpf(,g) -i- ’ CH; 4- H'

[0246]

[0247] Plasma-catalytic short-term effects at 10% initial CH4conversion, under target conditions (TSUr = 473 K, Tvib=4200 K, and p = 1 atm), demonstrate an increase in TOF H2of at least 10 orders for strongly binding catalysts (AEC< -0.5 eV) and an increase of at least 5 orders of magnitude for weakly binding catalysts (AEC> -0.5 eV) compared to thermal catalysis results for TSUr = 473 K (Fig. 3B). This increase in TOF H2suggests the potential of reduced operating temperatures for direct conversion of ( H i The increased TOF H2at target conditions is characterized by a shift of optimal catalyst from weakly binding catalysts to strongly binding catalysts (e.g., Ni, Pt, etc.) on 111 compared to thermal catalytic pathway at Tsur= 473 K. TOF H 2 in plasma-catalysis at target condition also increases by at least 2 orders for optimum catalyst compared to representative pyrolytic conditions (Tsur= 873 K). TOF H2for 111 sites is consistently higher than 211 sites across the range of AECvalues (2 eV - 1.4 eV). A similar trend was observed for C2 hydrocarbons (Figs. 17-18). Consequently, further simulation results are presented exclusively for 111 sites.

[0248] Experimental results are compared at approximately 10% initial CH4 conversion, achieved by controlling the flow rates for different catalysts, while maintaining a constant P. Experimentally measured TOF H2ratios corroborate the simulation results, demonstrating the following trend: TOF H2Pt > TOF H2Ni > TOF H2Cu on 111 sites (Fig. 3E). TOF H2ratios observed for 20% (wt) Ni / Al2O3, 20% (wt) Pt / Al2O3, and 20% (wt) Cu / Al2O3exhibit a non-linear increase with increase in SEI (0 - 45 eV / initial CH4 molecule). This non-linear increase in TOF H2 ratios is attributed to the interplay between conversion and production,Attorney Docket No. 10046-649W01

[0249] which are constrained by the equilibrium limitations of the process. Notably, 20% (wt) Pt / Al2O3demonstrates a ~ 45 fold improvement on production rate of H2compared to plasma alone case at SEI ~ 37 eV / initial CH4molecule. The trends and magnitudes of H yield from simulations and experiments align closely: Yield H2 Pt > Yield H2 Ni > Yield H2 Cu (Fig. 3F). A peak H2yield of ~ 10.5 % is observed for 20% (wt) Pt / Al2O3at ~ 10% initial CH4conversion. 20% (wt) Pt / AbCE exhibits a - 8 % H? fuel production efficiency (TJ / H?) for low SETs ( < 5 eV / initial CH4 molecule) (Fig. 3D, Methods). ηfH2increases from 2% to 8% as SEI decreases from 45 eV / initial CH4molecule to 3 eV / initial CH4molecule for all catalysts, suggesting stronger plasma-catalytic coupling at low SEIs or high flow rates for a given plasma input power.

[0250] Plasma-catalysis demonstrates the ability to overcome scaling laws in the direct conversion of CH4. This is evidenced by both an increase in yield H2and TOF H2under reduced target temperatures (Tgas= Tsur= 473 K) compared to thermal catalysis, as well as a shift in the optimal catalyst towards stronger binding materials (Fig. 3B). Plasma coupling of the dissociative adsorption step facilitates lower operating temperatures (from Tgas= Tsur= 873 K to Tgas= Tsur= 473 K) by reducing the activation energy barrier of this elementary step. (Rl, Fig. 3A). It permits noble catalyst with higher Ea, Ri to be used. Furthermore, the plasma coupling to dissociative adsorption step and the associated reaction rates can be modulated through plasma and gas state properties (p, Tgas, Tvib), independent of surface properties (ΔEc) (Figs. 1 A, 1C). This will be shown that with different ΔEa,R1in a plasma-catalytic system, subsequent elementary reactions R2-R10 can be tuned.

[0251] Enabling product speciation

[0252] Selective control of the C2 hydrocarbon production is desirable owing to its higher value compared to methane. Shown herein is how plasma-catalytic design space can be tuned to enable product speciation in direct conversion of CII4 through independent control of the plasma and surface properties. Experiments were conducted to quantify the production speciation of C2 hydrocarbons in the direct conversion of CH4 under mild thermodynamic conditions (Tgas= Tsur = 473 K, p = 1 atm). These experiments were performed at approximately 10% initial CH4conversion, utilizing 700 mg of each catalyst (Al2O3, 20% (wt) Ni / Al2O3, 20% (wt) Pt / Al2O3, 20% (wt) Cu / Al2O3) and a discharge voltage (V) of 7-8 kV across a discharge gap of 2.5 mm. Further experiments were conducted to quantify the influence of τresidenceon the C2hydrocarbon product speciation on 20% (wt) Ni / Al2O3by changing the flow rates (Q = 1.875 - 37.5 seem, Tresidence = 206 s - 6.45 s). At a constant PAttorney Docket No. 10046-649W01

[0253] 4.5 W, changes in SEI and τresidenceare influenced only by the flow rate. To isolate the yield specifically attributed to plasma-catalytic activity, the yields of C2H6, C2H4, and C2H2produced by plasma alone are deducted from the net yield. This approach operates under the assumption of additive effects between plasma and catalyst contributions in a convectionlimited regime (Fig. 11).

[0254] The microkinetic model tracked the production of C2 hydrocarbons in the gas phase through branching reactions (R6-R10). These reactions involved the recombination of surface -bound CHx* species and the subsequent desorption of hydrocarbon products from the surface into the gas phase. Only short-term effects at a given initial CH4 conversion on the product speciation are discussed (Tsur= 473 K, Tvib = 4200 K, and p = 1 atm). The influence of varying residence time was incorporated into the model by adjusting the initial methane conversion. At a constant P̄ ~ 4.5 W, the experimentally observed CH4conversion was found to have an exponential relationship with τresidence / SEI. This correlation was then utilized in the model to quantify simulated C2 hydrocarbons yield as a function of T residence ZSEI.

[0255] Product speciation experiments at ~ 10 % initial CH4conversion demonstrate C2H6is the major C2hydrocarbon for 20% (wt) Cu / Al2O3(~ 2% yield), while C2H4 is the major C2hydrocarbon for 20% (wt) Pt / Al2O3(~ 0.5 % yield) (Fig. 4C). Both experiments and the model show that C2H6 is the major C2 hydrocarbon for Cu (111) where associative desorption of CH3* is favored among the branching reactions (R6, Table 4) (Fig. 4B). Similarly, both experiments and the model agree that C2H4is the major C2hydrocarbon for Ni (111) where CH2* recombination leads to C2H4* formation, which then desorbs as C2H4in gas phase. While the trends in product selectivity align between the model and experiments, the quantitative agreement in terms of yield magnitudes is less clear at Tsur = 473 K. A sensitivity analysis of the microkinetic model identified Tsurwith the highest global sensitivity coefficient among Tsur, Tvib, and ΔEa,R1for reactions R1-R10 (Figs.

[0256] 12, 13, I4A-14C). The model showed good agreement with experimental results at Tsur = 453 K for Ni (111) and at TSUr = 503 K for Cu (111) (Fig. 4B). Extrapolating the microkinetic model to higher TSUr (373 K - 1073 K) revealed ideal temperature ranges for specific C2 hydrocarbon production: Ni (111) favors C IL (Tsur < 473 K), C2II4 (473 K < Tsur < 640 K), C2H2 (Tsur > 640 K); while Cu (111) favors Ts«r < 573 K for C2H6, 573 K < Tsur < 973 K for C2H4, and Tsur > 973 K for C H (Fig. 26A). In plasma-catalysis with Ni (111), decreased surface carbon coverage coincided with increased total C2 hydrocarbonAttorney Docket No. 10046-649W01

[0257] production at Tsur> 700 K, opposite to thermal catalysis (Fig. 26B).

[0258] Plasma-catalytic simulations were performed to capture the C2hydrocarbon production composition at observed experimental CH4 conversions (~ 0.5 % - 13 %). The variation of residence timescales yielded interesting results (Fig. 4D). At high flow rates (Q > 7.5 seem, SEI < 10 eV / initial CH4 molecule), the observed C2H6, and C2H4yields aligned with the simulation results for Tsur= 473 K, Tvib= 4200 K, and p = 1 atm. However, at low flow rates (Q < 7.5 seem, SEI > 15 eV / initial CII4 molecule), a deviation was observed in the experimental C2H6yield. This deviation was successfully captured by the simulated yields bounded by Tsur= 453 K and Tsur = 503 K. Interestingly, the C2H4 showed a drop off in the yield and deviated from the model at low flow rates (Q < 7.5 seem, SEI > 15 eV / initial CH4 molecule), unlike C2H6. C2H4is relatively an unstable gas compared to C2H6and C2H2and at low flow rates (Q < 7.5 seem, SEI > 15 eV / initial CH molecule, T residence > 100 s), it likely undergoes secondary plasma conversion in the gas phase to form stable C2H2. This hypothesis is consistent with the experimental observation of a decrease in C2H4 yield accompanied by a 1:1 stoichiometric increase in ( H yield at low flow rates (Q < 7.5 seem, SEI > 15 eV / initial CII4 molecule, Tr idenc > 100 s).

[0259] Plasma-catalysis demonstrates the potential for tunable product speciation in methane conversion through careful manipulation of material properties, surface temperature, and residence timescales (Fig. 4A). Reaction R1 is impacted by plasma properties and not by surface / gas properties. This approach offers independent control over product distribution, decoupled from plasma properties, enabling new catalyst to be used to tune production of C2 hydrocarbons. This capability contrasts with thermal catalysis, where net C2 hydrocarbon production is severely limited and fails to achieve representative yields.

[0260] Overcoming surface inactivation and reaction mechanism

[0261] As previously shown, conventional methods of direct conversion of methane suffer from catalytic inactivation due to coke formation on the catalytic surface (Figs. 2A-2E). While maintaining tunable product speciation, it is shown here how noble catalysts with higher ΔEa,R1can be leveraged to facilitate longevity of the catalysts in the process. Time on stream (TOS) experiments were performed to evaluate catalytic performance and stability in direct conversion of CH4. The DBD plasma operated at V = 7-8 kV, Tgas= Tsur= 473 K, and p = 1 atm, maintaining a constant P ~ 4.5 W by tuning the voltage of the DBD power supply. 20% (wt) Ni / Al2O3(~ optimal catalyst wrt TOF H2) and 20% (wt) Cu / Al2O3(noble catalyst wrt TOF H2) were chosen as two catalysts to perform TOS experiments. AAttorney Docket No. 10046-649W01

[0262] constant Q = 3.75 seem ensured fixed conversion (~ 10 % initial CH4 conversion) throughout the experiment. A baseline yield H? was established using only dielectric supports (Al2O3) under similar plasma conditions. H2 yield was sampled every ~ 15 min for ~ 14 hrs using GC / TCD. Raw values of yield H2, calculated TOF H2, and cumulative yield H2are reported. The model was adapted to make it analogous to the flow mode experimental setup. Flow mode behavior was captured by assuming a fixed conversion (10% initial CH4conversion) in the reactor at p = 1 atm, Tsur= 473 K, and Tvib= 473 K. Solution from the short-term effect (10% initial CH4 conversion) results were used to initialize the mole fraction of hydrogen and C2 hydrocarbons. Simulations were run for ~ 105seconds.

[0263] TOS experiments reveal that while yield II2 on Ni / AhCh is initially constant and higher than on Cu / AhCh (t < 75 min), it subsequently drops off, falling below the yield H - on CU / Al2O3att - 1 0 min and eventually matching the baseline II? yield measured on Al2O3only at t ~ 250 min (Fig. 5B). This initial increase in H2 yield followed by a subsequent decline suggests that the Ni / Al2O3catalyst experiences deactivation over time. The cause of Ni / Al2O3deactivation is coke formation, which is the buildup of carbon deposits (poisons) on the catalyst surface (Fig. 5A). Coke blocks active sites and hinders the progress of surface reactions, leading to a decrease in performance. The fact that the final yield H2matches that of Al2O3alone implies the active sites on Ni / Al2O3are no longer contributing to the reaction. In contrast, CU / Al2O3, although initially less active (t < 150 min), exhibits a stable H2production rate for a longer duration (t ~ 840 min) (Fig. 5B). Experimental TOFs calculated by dividing the baseline subtracted yield II2 with T residence and normalizing by the number of active sites for each catalyst (Table 5), align with the TOFs predicted by the microkinetic model within the bounds of TSUr = 453 K and TSUr ~ 503 K (Fig. 5C).

[0264] Table 5. Active site characteristics of powder catalysts from pulsed chemisorption measurements.

[0265]

[0266] These findings illustrate that while utilizing a more active catalyst like Ni / Al2O3can lead to increased H2yield in the short term, it ultimately faces deactivation due to catalyst poisoning. In contrast, noble metal catalysts like Cu / Al2O3, while initially less active,Attorney Docket No. 10046-649W01

[0267] demonstrate superior long-term stability, offering sustained production of both H₂ and valuable C₂H₆ for extended periods (up to - 14 hours). This suggests the possibility of achieving in-situ catalyst regeneration through surface recombination by strategically designing plasmas in conjunction with noble metal catalysts (Fig. 5 A). The advantage of noble catalysts becomes evident as early as ~ 360 minutes into the process, where the cumulative H₂ yield for Cu / Al₂O₃ surpasses that of Ni / Al₂O₃ (Fig. 5D).

[0268] Surface diagnostic techniques (SEM / EDS, TEM / EDX, and XPS) were employed to evaluate the reaction mechanisms linked to noble metal catalysts and the types of carbon deposits on catalytic surfaces. The findings were subsequently compared with a microkinetic model (Figs. 6A-6D). XPS spectra of Ni / Al₂O₃ and Cu / Al₂O₃ samples extracted from the TOS experiments at - 6 hrs identified surface functional groups that aligned with the microkinetic model. The model indicated that Cu / Al₂O₃ facilitated the formation of CH₃* as the primary surface carbon type (~ 40%), which subsequently recombined on the surface to produce C₂H₆ in the gas phase. In contrast, Ni / Al₂O₃ exhibited a predominant presence of (CH₂-) on the surface (> 45%) (Figs. 6A-6B).

[0269] TEM / EDS elemental maps of Ni / Al₂O₃ and Cu / Al₂O₃ samples, collected after approximately 6 hours of TOS experiments, revealed the state of the active sites. In Ni / Al₂O₃, complete deactivation was observed at - 6 hours, with 100% of the surface covered by carbon-containing species. This observation aligns with the plateauing H₂ yield on Ni / Al₂O₃, which eventually matches the H₂ yield on Al₂O₃ alone (Fig. 5B). Conversely, Cu / Al₂O₃ maintained consistent catalytic performance. This can be attributed to the abundance of active sites, with only ~ 60% deactivated as early as ~ 15 minutes, and no further deactivation due to the continuous recombination of CH₃* to form C₂H₆. These experimental findings demonstrate good agreement with the simulation results (Figs. 6C- 6D).

[0270] Conclusions

[0271] These experiments demonstrate direct methane conversion to hydrogen and C2 hydrocarbons under mild thermodynamic conditions by synergizing the properties of plasmas and catalysts. It is shown herein that plasma activation can target specific steps in the reaction pathway to initiate the reaction while the subsequent reaction steps are governed by the surface and gaseous state properties independent of the plasma properties. This synergistic co-design of non-equilibrium plasma and catalysts facilitates selective activation of methane at near-ambient temperatures and pressures, allowing for enhanced hydrogen production and selective upgrading of methane into higher hydrocarbons, such as ethyleneAttorney Docket No. 10046-649W01

[0272] and ethane. The ability to independently control plasma properties, catalyst selection, and operating conditions unlocks new design spaces, resulting in tunable product distributions. Additionally, the use of noble metal catalysts in this plasma-catalytic system mitigates coke formation and extends catalyst longevity by promoting surface recombination reactions that prevent deactivation.

[0273] The integration of plasma-catalysis not only improves process feasibility but also aligns with carbon-neutral goals by enabling on-demand methane conversion with minimized CO2 emissions. By harnessing renewable electrical energy to drive vibrational excitations, this approach offers a sustainable and scalable pathway for methane valorization. The findings demonstrate that plasma-catalysis can achieve sustained hydrogen and C2 hydrocarbon production, addressing critical challenges in catalyst stability and product selectivity. By combining the unique advantages of plasma-catalysis with carefully chosen catalysts, we pave the way for a more sustainable and selective approach to methane upgrading, unlocking its full potential as a feedstock for valuable chemicals and fuels.

[0274] Materials and Methods

[0275] Reactor Design

[0276] A modular reactor was built to study the methane plasma-catalytic conversion of CH₄ into H₂ and higher C₂ hydrocarbons. It featured an aluminum frame, o-rings (dash number 022), and removable transparent windows (quartz and acrylic) for optical accessibility. Compression bolts were used to assemble the reactor and maintain a sealed atmosphere (Fig. 19). A sleeve made of polytetrafluoroethylene (PTFE) was inserted within the aluminum frame to establish an inert reaction environment. The reactor was used to operate at atmospheric pressure in this study but could also operate in vacuum conditions (~ 1 Torr). Two electrical feed-throughs (KJL Part EFT0513055, 1 / 8” NPT fitting) were used to deliver electrical energy into gas streams and generate a plasma. The reactor was operated in a flow mode configuration. Reactants were flowed into the reactor using a digital mass flow controller (MFC, Brooks Instrument, Model GF40). The residence timescale τ_residence = 2 s – 120 s) of the gas components in the reactor was tuned by changing the flow rate (Q) over a range of 1.875 seem - 50 seem. The reactor was purged for ~ 15 - 20 mins and the reactor composition was sampled with a gas chromatography (GC) system before plasma was ignited. This was done to initialize the reactor with a uniform gas composition and establish a baseline reactant composition before plasma was formed.

[0277] Plasma Generation

[0278] A dielectric discharge barrier (DBD) plasma was ignited between a pair of 1 mmAttorney Docket No. 10046-649W01

[0279] thick quartz dielectrics using two aluminum electrodes (34 mm x 24 mm x 8 mm) (Figs. 24, 25A-25F). The current and voltage signals of the plasma were measured using a high-voltage probe (Tektronix 6015a) and a current probe (Pearson Model 6585) using a mixed signal oscilloscope (Tektronix MSO44, 200 MHz bandwidth) (Figs. 20A-20B). The discharge gap was fixed at 2.5 mm and Q was varied (1.875-50 seem, 99.97% CH4) using a digital mass flow controller (Brooks Instrument, Model GF40) to tune the specific energy input (SEI) of the plasma. The discharge voltage (V) was varied from 5 kV-15 kV at a frequency (f) of ~ 20 kHz and the average peak current was observed to be - 200 mA during discharges. The flow rate did not affect the power characteristics of the DBD reactor ( < 2% change in SEI with Q = 1.875-50 seem).

[0280] Powder Catalysts

[0281] Al₂O₃ powder (Item# 1600-RG-GAL1-Pdr-25G, Riogen Inc), and three catalysts, 20% wt Ni / Al₂O₃ (Item# 0040-ALNiA20-Pdr-25G, Riogen Inc), 20% wt CU / Al2O3(Item# 0036- ALCuA20-Pdr-25G, Riogen Inc), and 20% wt Pt / Al₂O₃ (Item# 1005-ALPtA20-Pdr-10G, Riogen Inc), were used to perform plasma-catalytic experiments. The polycrystalline transition metal catalysts (catalyst particle size ~ 10 - 100 nm) were loaded on the Al 2O3 supports (support size ~ 10 - 500 pm). The transition metal catalyst dispersion varied from 1 -8 % and the active site density varied from 60-110 μmol / gcatalyst(Table 5).

[0282] Catalytic Loading

[0283] A total of - 700 mg of each catalyst was loaded into the reactor between a pair of 1 mm thick quartz dielectric plates. The catalysts were layered alternately (4 layers) with 40 mg of quartz wool (1-5 μm thickness, CAS# 14808-60-7, Thermo-Fischer) to achieve a homogeneous distribution. This method ensured a packing fraction of - 21%. A hydrogen plasma pretreatment of the catalysts was found to have negligible change (~ 1.4 %) on the yields of gaseous products (Fig. 23).

[0284] Gas Chromatography

[0285] An Agilent 8890 GC instrument, equipped with a flame ionization detector (FID) and thermal conductivity detector (TCD), was used to measure CH4 conversion and product yields of H₂, C₂H₆, C₂H₄ and C₂H₂ (Fig. 19). The GC used two gas sampling valves (GSV s) with a 250 pl volume to take samples from the flow reactor. A HP-PLOT (a bonded polystyrene- divinylbenzene column) with 0.320 mm column thickness and 20 pm film thickness was used to separate compounds of interest (e.g., C₂H₆, C₂H₄ and C₂H₂) for the FID (Fig. 22). A three-column gas-separation setup with two 6-port valves (i.e., column isolation and pre-column backflush valves) were used to separate species of interest (e.g., H₂Attorney Docket No. 10046-649W01

[0286] and CH4) from permanent gases for the TCD (Fig. 22). The pre-column backflush valve was attached with two HayeSep Q columns (1 / 8 inches in diameter) to backflush all compounds heavier than C₂ while the isolation valve connected to a Molsieve 5A column (1 / 8 inches in diameter) was used to increase the separation between CH₄, H₂ and permanent gases.

[0287] Helium (99.999% pure) was used as a carrier gas with a flow rate of 1 ml / min and N2 (99.998% pure) was used as a make-up gas for the FID. The GC temperature program for FID was set to 40°C at the time of sampling followed by a linear ramp of 5 °C / min to 160°C (0 min hold time) and a second linear ramp of 10°C / min to 220°C (10 mins hold time) while the temperature program for TCD was set to 45 °C at the time of sampling for 5 mins followed by a linear ramp of 10°C / min to 160°C (0 min hold time).

[0288] Calibration procedure

[0289] Calibration curves were obtained using GC to estimate CH4 conversion along with the product yield of H₂, C₂H₆, C₂H₄, and C₂H₂ (Figs. 21A-21E). External calibration standards of varying purity were used to measure a linear relationship between GC areas and moles of injected species (calibration curves). Calibration curves were obtained using an external standard calibration method due to absence of solvent effects. For hydrogen and methane calibrations, a 250 ql sample of 99.999%’ pure hydrogen was mixed with 99.97% pure methane at various compositions relevant to the experiment (e.g., H2: CH4 = 5:95 - 40:60) was used. These gases were mixed using MFCs, and they were sampled into a TCD by a GSV. These compositions were then plotted against their corresponding GC areas obtained using TCD. The data points were fit with a linear curve using least square method to obtain the required calibration curves. A similar procedure was implemented for ethane and ethylene calibrations with the FID. A 0.1% v / v (or 1000 ppm) C2II4 balanced with N2 was used for C2II4 calibration while a 10%’ v / v C2II6 balanced with N2 was used for C2II6 calibration. The gases were introduced separately mixed with Ar at various compositions into GC-FID using a GSV to obtain areas corresponding the amounts introduced into the GC. These areas were then plotted against the number of moles of these known samples introduced and fit with a linear curve to obtain the required calibration curves.

[0290] Powder Catalyst Characterization

[0291] CO pulse chemisorption experiments were performed on Ni / Al₂O₃ and Pt / Al₂O₃ using a Micromeritics AutoChem III (LOT / SN 446533 / 1) (Table 5). N₂O pulse chemisorption experiments were performed on Cu / Al₂O₃ using a Micromeritics AutoChem II 2920 (Table 5). Ni / Al₂O₃ and Pt / Al₂O₃ (~ 100 mg) were reduced first at 773 K using a 10% H₂ / Ar flown at 50 ml / min for 30 min to clean and activate the metal sites. HeAttorney Docket No. 10046-649W01

[0292] (99.999%) was flown at 50 ml / niin to reduce the temperature for analysis (300 K) and purge the remaining H₂ / Ar in the sample tube and instrument. He flow was maintained until a stable TCD baseline was established. 10% CO / He pulses (injection volume ~ 0.54 ml) were injected into the sample until the peak sizes were equal or 10 peaks were observed. ~ 50 mg of Cu / Al₂O₃ was reduced at 773 K using a 10% H₂ / Ar flown at 50 ml / min for 30 min to clean and activate the metal sites. He was flown to cool the samples to the analysis temperature of 353 K. N2O pulse chemisorption was done in conjunction with mass spectrometry to monitor the consumption of N2O. N2O pulses were fed in the loop till equal peaks were recorded or 50 peaks were observed. Data was recorded every 0.1 seconds. After the analysis, samples were re-weighed. Mass spectrometer data was imported into MicroActive software and integration was performed at m / z = 28 and m / z = 44 to calculate the number of active sites.

[0293] Surface diagnostics

[0294] Scanning Electron Microscopy (SEM) with Energy Dispersive X-Ray Analysis (EDX) Scanning electron microscopy (SEM) images and EDX elemental maps were acquired using a Thermo Scientific Apreo 2 microscope to study the elemental composition of catalytic surfaces and carbon deposits. A small sample (~ 0.5 - 1 mg) was placed onto carbon tape-covered SEM stubs using a sterilized spatula, then transferred to the SEM chamber for imaging. SEM imaging was conducted at high vacuum (~ 5×10-6Torr), and features a 4 nm spot size, 0.40 nA current, and 15 kV accelerating voltage. Raw images of SEM / EDX were saved in TIFF format (Fig. 29). EDX elemental maps were captured at high vacuum (~ 5xl0‘6Torr), 0.40 nA current, and 10 - 15 kV accelerating voltage to ensure at least ~ 1000 counts per seconds were recorded. Transmission Electron Microscopy(TEM) with EDX High angle annular dark field(HAADF)-STEM images and EDX elemental maps were measured using a probe aberration-corrected JEOL NEOARM TEM, operating at 200 kV and an average beam current of 10 μA (Figs. 30-31). A dwell time of 10 μs was used for imaging. EDX was conducted in a JEOL NEOARM instrument equipped with a JEOL large-angle silicon drift detector (0.96 sr). HAADF-STEM images were analyzed using the Gatan Microscopy Suite. For HAADF and EDX analysis, samples were prepared as follows: 10 mg of catalysts exposed to plasma were mixed in 10 ml of 99.99% pure CH3COOH. Next, 1 ml of the mixed solution was diluted further with 10 ml of CH3COOH to obtain a net concentration of 0.1 nig / nil. 5 pl of the prepared sample was then dropcast onto the Cu grids (Ted Pella, ultrathin C film on Lacey carbon, 400 mesh) and dried for 5 minutes. Next, the TEM grids were placed under high vacuum conditions (~ 1×10-7Torr) for at least 60Attorney Docket No. 10046-649W01

[0295] hours before transporting it to the TEM chamber for analysis. Fractures in the C film during preparation and transport exposed embedded particles. These exposed particles were chosen for analysis to minimize background signal from the intact film.

[0296] X-ray Photoelectron Spectroscopy (XPS)

[0297] X-ray photoelectron spectroscopy (XPS) analysis was performed using a VersaProbe 4 microprobe (Physical Electronics) equipped with a monochromatic, microfocused, scanning Al-Ka X-ray source (hv = 1486.5 eV). A charge neutralizer filament ensured minimal surface charging during analysis. Survey scans were acquired at a step size of 2 eV and a sweep time of 10 minutes per sweep to identify the elemental composition of the sample surface (Figs. 32A-32B). The base pressure of the instrument was maintained at approximately 10-9Torr. High-resolution spectra were collected over a focused area of ~ 250 μm x 250 μm using a pass energy of 10 eV, a step size of 0.5 eV, and a sweep time of 45 minutes per sweep for each elemental photoelectron peak corresponding to the detected elements (Figs. 33A-33D, 34A-34D). CasaXPS software was employed for spectral analysis. Fitting components were modeled using a combination of Gaussian (70%) and Lorentzian (30%) profiles (GL(30)), while a standard Shirley-type baseline with variable offset levels at the high binding energy endpoint was used for fitting transition metal peaks (Table 6).

[0298] Table 6. Measured areas (pre and post run) and sensitivity factors of C 1s and Al 2p₃ from the high resolution spectra for Ni / Al₂O₃, and Cu / Al₂O₃.

[0299] Al 2p₃ area | Al 2p₃ sensitivity factor | Source | Area RatioNi-Pre Run | 152 | 1 | 585.2 | 0.5735 | 35 | 0.12964847Cu-Pre Run | 194 | 1 | 1056 | 0.5735 | 35 | 0.09531646Ni-Post Run | 325 | 1 | 452.2 | 0.5735 | 35 | 0.29187465Cu-Post Run | 189.3 | 1 | 714.3 | 0.5735 | 35 | 0.13193384

[0300]

[0301] Process Evaluation

[0302] Plasma Input Energy

[0303] Voltage and current traces were measured using a voltage probe (Tektronix P6015A, 75 MHz bandwidth) and Pearson probe (Model 6585, 250 MHz bandwidth), respectively.Attorney Docket No. 10046-649W01

[0304] The traces were recorded using an oscilloscope (Tektronix MSO44, 200 MHz bandwidth) (Figs. 20A-20B). Average power (P) input into the system is calculated as,

[0305] P̄ = (∫t₁t₂ VI dt) / (t₂ − t₁)

[0306]

[0307] (1) where V is the applied voltage, I is the measured current and t₂-t₁ is the time period over which the signals were averaged. A minimum of 10 cycles in the V and I were used to quantify P̄.

[0308] Performance Metrics

[0309] Plasma chemical processes were evaluated by comparing the moles of products and reactants, produced along with their rates, during reactions. Conversion of CH 4, (i.e., the reactant species) is evaluated as,

[0310] NCH4, O conversion =

[0311] NCH4, O

[0312]

[0313] (2) where N̂CH₄,0 is the initial number of moles of methane and N̂CH₄,f is the final number of moles of methane. Yield of product X(g) is evaluated as,

[0314] Yieldx=

[0315]

[0316] (3) where N̂ₓ is the number of moles of a product X(g) in the reactor. Yield due to catalytic conversion in plasmas (YieldM / Ai2O3) is evaluated as,

[0317] YieldM / Al₂O₃= YieldAl₂O₃ + M / Al₂O₃− YieldAl₂O₃,

[0318]

[0319] (4) where YieldAi2O3 + M / A1203 is the net yield from reactions in the plasma phase and through catalytic conversion, and YieldAi2O3 is the yield from reactions in plasma phase only where no catalytic reactions occur. TOF and TOF ratio are evaluated as,

[0320] TOF = (YieldX,Al₂O₃+M / Al₂O₃− YieldX,Al₂O₃) / (τresidence× nactive sites) (5) TOF ratio = TOFX,Al₂O₃+M / Al₂O₃ / TOFX,Al₂O₃= 1 + TOFX,M / Al₂O₃ / TOFX,Al₂O₃

[0321]

[0322] TOFX,Al₂O₃(6) where nactive sitesis the number of active sites of a metal catalyst, TOFAl₂O₃ + M / Al₂O₃is the netAttorney Docket No. 10046-649W01

[0323] TOF due to gas phase and catalytic conversion, and TOFAl₂O₃is the TOF due to gas phase conversion only.

[0324] Fuel Production Efficiency

[0325] Fuel production efficiency was used to quantify performance of plasma-catalytic methane conversion. H₂ was chosen as the target fuel and other higher C₂ hydrocarbons were not considered. H₂ fuel production efficiency (ηfH2) is defined as,

[0326] YieIdH2* LHVH2SEI ’

[0327]

[0328] (7) where LHVH₂refers to lower heating value of pure hydrogen, SEI = Average power / input methane flow rate, and hydrogen yield = Molar flow rate of hydrogen / Initial molar flow rate of methane. This definition for fuel production efficiency is used to provide a performance metric for plasma-catalytic methane dehydrogenation results.

[0329] refers to lower heating value of pure hydrogen,

[0330] Optical Emission Spectroscopy

[0331] A Horiba iHR 550 spectrometer with an EMCCD camera (f / # = 6.4) recorded emission spectra from the plasma (300 – 1000 nm) (Figs. 19, 27). The setup included a 1800 grooves / mm diffraction grating (blaze wavelength: 500 nm, angle: 26°45’). A biconvex lens (focal length: 125 mm, f / # = 6.4) focused plasma emissions into the spectrometer's slit (50 pm). Emission spectra of DBD (~ V = 7 kV, 2.5 cm discharge gap, Q = 50 seem, ~ 700 mg Al2O3powder, 99.999% CH4) were recorded using Is exposures and 700 CCD accumulations.

[0332] Microkinetic Modelling

[0333] A global microkinetic model was developed to simulate the direct conversion of methane on different transition metal catalysts. The time evolution of gaseous species (i.e., partial pressure) or adsorbed species coverage (i.e., surface coverage) is evaluated using a series of rate equations,

[0334] ∂θᵢ / ∂t = Σⱼ cᵢⱼrⱼ (8)

[0335] dpi V dt j

[0336]

[0337] (9)Attorney Docket No. 10046-649W01

[0338] where θᵢ is the surface coverage of adsorbate i or free active sites, pᵢ is the partial pressure of gaseous species i, cᵢⱼ is the stoichiometric coefficient for the species in the surface reaction j, and rⱼ is the rate corresponding to reaction j. rⱼ is defined as,

[0339]

[0340] where kⱼ,f and kⱼ,r are the forward and backward reaction rate coefficients, θₛ is the surface coverage in case of surface species and pₛ is the partial pressure for gas-phase species, and cₛⱼ,f and cₛⱼ,r are the stoichiometric coefficients of the reactants of forward and backward reactions, respectively. The direct methane conversion to H2 and C2 hydrocarbons involves 10 reactions (R1-R10) that track the CH₄, adsorbed surface intermediates (e.g., CH*, CH₂*, CH₃*, C*, C₂H₂*, C₂H₄*, and H*), and stable products (e.g., H₂(g), C₂H₆(g), C₂H₄(g), and C₂H₂(g)) (Table 4). In total the mechanism tracks 5 gaseous species and 8 adsorbed species. kⱼ,f and kⱼ,r are calculated from the transition state theory (TST) as,

[0341] kⱼ = (k_B T_sur / h) exp(−ΔĜⱼ / R̂T_sur)

[0342]

[0343] (11) where k_B is the Boltzmann constant, T_sur is the surface temperature, h is the Planck constant; R̂ is the ideal gas constant, and ΔĜⱼ is the Gibbs free energy. ΔĜⱼ of an elementary reaction j is calculated as,

[0344] ΔĜⱼ = ΔÊ_a,j − T_sur ΔŜⱼ + N̂_a ΔZPE_j

[0345]

[0346] (12) Where AEa,j is the activation energy, ASj is the change in entropy, ZPE is the change in the zero-point energy for the elementary reaction j ('Table 7), Nais Avogadro’s number. AEaj and reaction enthalpy (Affj) vary depending on the adsorption energy of carbon ( A Ec) through catalytic scaling laws (Table 9). A Ecfunctions as a universal descriptor to correlate all elementary steps within the reaction mechanism across transition metal catalysts. The plateau behavior observed at A Ec< 0.5 eV arises from a model constraint imposed to ensure thermodynamic feasibility (Fig. 2C). This constraint mandates that activation energies must be greater than or equal to the maximum of the enthalpy change and zero (ΔE_a,Ri ≥ max (ΔH_Ri,0)).

[0347] Table 7. ZPE values of all the gaseous and surface species considered in the microkinetic model.Attorney Docket No. 10046-649W01

[0348] Species ZPE [eV] Source i

[0349] 1 H.h ]

[0350] 1.23

[0351] 1 H2C<?) o.<r H33: 1 M.13: CMf) 0.70

[0352] _ _ H. J3:

[0353] _ 1.33..............................................

[0354] | CH) ii.ii ]

[0355] 1.06

[0356] I CH) 0.74!!.i’ 1 | CH* H.12:

[0357] 0.49

[0358] ! C* 0.15

[0359] _ if _ U 12:

[0360] _ 136 _

[0361] 057

[0362]

[0363] 1 QHf 1.42

[0364] Table 8. Forward rate coefficients of methane electron impact excitation and collisional relaxation scheme. BOLSIG+ rates are determined for T_gas = 473K, and excitation temperature = 4200 K. k_SB, k_BG, k_3,1 and k_2,4 are reported for T_gas = 300 K. k_eb and k_es correspond to second order reaction. K_Sb, k_bg, k_s,1 and k_2,4 correspond to pseudo-first order reaction, where the pressure of one of the reactant species (CH₄) is kept constant.

[0365] Schem? Rssactsea Symbol Rato VaSto Source Eiia-Uea

[0366] Isjsgwf

[0367] 4 e" v(3(4)}

[0368] (Readissg 8 « >:?to-' at'V BOLSKH M-3sfe> '

[0369] EWITOB

[0370] ■SSiy.’K-i

[0371] (StrctdwBg J- e~ -i e- fc<;. 3.2 x K;: SmV BOWGto Mr®}*)

[0372] Tms&r (V-Vj CMfafa + CH* -* CHs(v.) -T kst 4.2 x IF

[0373] (v-;to V,)

[0374] Vfera&taai

[0375] Transfer t V-V) £'0,(^1 to CtT CITOa) to Ct-M fe / . D.t S-'TOTA '* { V; M V4)

[0376] Vii»ats<®s4

[0377] Trnwier (V-V; to Ctl„ -■> CilfaiPX)} to

[0378] X* 26 x IO- s:T0tr- V!2,4) - stetch Ctfaifaip to Ctf

[0379] to

[0380] Vtoratitoist

[0381] Tmsfer (V-Tl Ci7.j{v(2>4 }) 4 CtV / . 4 &;SS55

[0382] <v(2.4? to

[0383] gxwa'sd)

[0384]

[0385] Table 9. Surface reaction steps and. associated microkinetic parameters for fa on (111) sites of transition metal catalysts.Attorney Docket No. 10046-649W01

[0386] Reaction Reaction Reaction Enthalpy. Alfa

[0387] Number Source bf leV j

[0388] R1 cnfaf 4 2r ~ 4 fa 0.422? 0.2341.3?.

[0389] 82 Cfa 4 8’ w CH) T fa 0.4250 0.0630. D. R3 cfa 4 r cr 4 fa 0.4325 -0.5403 2, S R4 cfa 4 tr c* 4 fa 0.4175 0.3463 2. S R5 2fa ~ fa(tf) 4 29" 0.8515 -0.3450. 23. R6 2CH) fafafp) 4 2fa ■'0.5 ODD -0.6500 j.? R7 ■'b $ -0.5025 -0.3739

[0390] R8 2Cfa C2H* y fa -0.5025 -O.37S9 24 R9 fafa ~ CMf) 4 fa 0 161 i -0.50.25 2. S. D.

[0391]

[0392] RIO 1 9457 -09675

[0393] Table 10. Surface reaction steps and associated microkinetic parameters for A fam on (Ill sites of transition metal catalysts.

[0394] Reaction Activation Energy, & fasiNumber Reaction S«HJxe ya[eVJ

[0395] Ri Cfafa) 4 29- Cfa 4- fa 0.3975 0.9751.23. R2 Cfa 4 fa Cfa 4 fa 0.4375 0.8675.23. R3 Cfa 4 fa Cfa 4 fa 0.4450 0.2342. NT. R4 Cfa 4 fa * fa 4 fa 0.2975 1.4891. Ts. R5 2fa ** fa fa) 4 29" 0.9452 -00000

[0396] R6 2Cfa <■■> fa faXp) 4 29" 1.5488 -0.4700

[0397] R7 2Cfa C / fa 4 6' 1.67 H -0.2525. m. RS 2Cfa fa fa 4 6" 2.4110 -0.2750 2.3 R9 If fa -fa fad fa 4 fa 0,0377 -0.0175 2.3

[0398]

[0399] R10 _ _ -0 5675 0.3203 3.3 Table 11. Surface reaction steps and associa ted microkinetic pa rameters for zl on (211) sites of transition metal catalysts.

[0400] Reaction Reaction Enthalpy. &fa;

[0401] Nunibes Reaction SOBICC. &.. A±Y1.

[0402] Ri Cfa fa) 4- 2 fa 4 fa 0.4325 0.3597. D'. R2 Cfa 4 fa ~ Cfa 4 H" 0.3375 0.5195 23 R3 Cfa 4 fa Cfa 4 fa 0.3750 0.2250 'n R4 Cfa 4 fa fa 4 fa 0.2975 0,5391 5.? R5 2 fa ~ faffa 4 21T -0.3650 0.6106

[0403] R6 2Cfa C,fa(.fa 4 2fa -0.5000 -0.6500

[0404] R7 2C Hy< 4* -0.7075 -0.9287

[0405] RS 2Cfa 4 9" -0.6650 -0.4674

[0406] R9 fafa CMs) 4 fa -0.2000 0.0430. X?. R10 fall) (bH-Tyf) 4 fa -0.6300 0.3772.??.

[0407]

[0408] Table 12. Surface reaction steps and associated microkinetic parameters for 2a, mon (211) sites of transition metal catalysts.Attorney Docket No. 10046-649W01

[0409] ~RS. T.. |. 4L2450. J. L3J78

[0410] R9. T. "&*'.1.4ML50. J. Golds'

[0411]

[0412] RIO. \..1. D3AD.1. o5m

[0413] Table 13. Calibration fits for ethane, methane, ethylene, and hydrogen.

[0414] ■ i. 1 x Species Vstese t pl) 1 Linear Fit (yfamdesk xt'GCArexh R:1 CHAg) TCD i y ~ 0. §901 x - 9.59M J 0 $99 ~ FID |

[0415] | OHogj? FID i 0. W49 [ OWgT™

[0416]

[0417] i

[0418] Plasma activation is assumed to couple to the dissociation adsorption step (Rl) of methane decomposition on transition metal catalysts (Table 4). Methane ground state is vibrationally excited to a lumped bending mode (CH4 v(2,4)) that interact with catalytic reaction pathway. The stretch modes of methane and radical species are neglected in the model. A Treanor distribution was used to calculate the populations of 11 vibrational states (the ground state and 10 excited states) of the CH₄ v(2,4) mode for a given Tvib = 4200 K and Tsur= 473 K (Fig. 27). We assume Trot ~ Tgas« Tsur. A Tgas= 473 K is used to compute the forward reaction rates of adsorption reactions and backward reaction rates of desorption of gaseous species. Contributions of the primary and secondary gas phase conversion due to plasma excitations are not considered in the microkinetic model. Interactions of radicals with catalytic surfaces are also neglected. Vibrational excitations of gaseous products are neglected. The partial pressure of an excited state methane in vibration level vis defined as,

[0419] ~nvPcn4>

[0420]

[0421] (13) where nvis the population fraction of vibrational state v and pciu is the total partial pressure of methane.

[0422] Estimating Population Fractions

[0423] The vibrational excitation of methane in the nthvibrationally excited lumped v(2,4) state (Pn(nthpolyad)) is denoted by nv(2,4), where n = 1...10. The bend modes, vo (0, n2, 0, 0) and vi (0, 0, 0, 114), are lumped together as bending mode v(2,4) due to their small energyAttorney Docket No. 10046-649W01

[0424] difference ( ~ 0.16 eV) and vibrational-vibrational (V-V) transition (TB =^30-50 ns at 25 mbar translating to ~ 3 ns at 1 bar). Quantum numbers ni and 113 are not considered (ni = 0 and = 0). The vibrational population fraction of the lumped v(2,4) mode is modeled using a two temperature Treanor distribution,

[0425] . a724 (xn^241 / eexxpP ( | - ( IE(jlT2’n*)h+ " TE^2’ I \ \£vib2rot / /

[0426] w

[0427]

[0428] here E(n2, ri4)h - vmi + V4TI4 and E(n2, TU)^ - X22(n22 - n.2) + %44(n24 - n.4), with X22 = 0 cm-1, and X44 = -6 cm-1. Level degeneracy, $24(2124), is given as, g₂₄(n₂₄) = Σ_{n₂+n₄=n₂₄} (n₂ + 1)((n₄+1)(n₄+2) / 2). Q24(Tvib, Trot) is the vibrational energy partition function which is truncated at n₂₄ = 10. The bend mode of methane was distributed across 11 vibrational states (the ground state and 10 excited states, n₂₄ = 0-10) in microkinetic simulations.

[0429] Measuring Plasma Temperatures

[0430] A database consisting of line strengths for rotational and vibrational transitions was constructed for CH(A2Δ — X0Π) using LIFBASE. Emission spectra for the CH(A2Δ — X0Π) (425 nm - 435 nm) band was recorded inside the reaction volume along the center line of a DBD plasma (Fig. 27 ). Measurements were acquired with an exposure time 1 s with 700 CCD accumulations. The reaction volume was loaded with A12O3 particles (~ 700 mg). The spectrometer was aligned by passing a laser beam (He-Ne, 632 nm) through the acrylic windows closer to the surface of dielectrics. This alignment ensured the plasma properties near the dielectric surfaces were captured using OES. The flow rate of CH4 was maintained at 50 seem. Estimation of rotational (Trot) and vibrational temperatures (Tvib) were done by employing massiveOES to fit the observed CH(A2Δ — X0Π) (425 nm - 435 nm) band with a simulated spectrum. The simulated spectrum undergoes a convolution with a Voigt profile, characterized by Gaussian and Lorentzian half-widths at half-maximum (HWHM). A two-temperature parametric model for non-Boltzmann population distribution was fit with the measured spectrum by following a state-by-state approach. Tvibfor CH(A2Δ — X0Π) was used to calculate the non-Boltzmann methane population distribution in the reactor. A bounded Tvib for CH4 was estimated by considering rates of equilibration between multiple species in a mixture and the average time of our OES measurements. Tvib for CH(A2A — X°fl) was found to be - 4200 K using the state-by-state approach. Therefore, an average Tvibvalue of ~ 4200 K was used in Treanor distribution to determine population densities distribution of excited methane species. The OES measurements captured cumulative emissions overAttorney Docket No. 10046-649W01

[0431] longer periods (~ minutes), measuring a net effect on population fractions of excited species rather than dynamic relaxation of individual species. This suggested sufficient accumulation time for vibrational temperature equilibration among hydrocarbons in the CH₄ plasma. Trotfor CH(A2Δ — X0Π) was found to be ~ 473 K. We assume Trot~ Tgas~ TSUr within microkinetic calculations.

[0432] High Speed Imaging

[0433] A Shimadzu Hyper Vision HPV-X2 high-speed camera and a with LaVision HS- IRO external intensifier (8-bit, 60 gain on a scale of 0-100, configured as a two-stage microchannel plate) was used to capture the interaction between catalyst particles and a DBD plasma (Figs. 25 A-25F). This setup was capable of recording broadband imaging of 256 frames at 10 million frames per second and 50,000 pixels. With a pixel size of 32 pm, images resolved the interaction between plasmas and supported catalytic particles (1- 50 pm) with 1-100 ps exposure at 200 frames per second.

[0434] Statistical Analysis

[0435] Experiments were repeated at least three times to quantify the random variations (standard deviation) in the data (N > 3). TEM / EDX and SEM / EDS scans were performed on at least 10 locations in a sample. XPS spectra were recorded at one location in a given sample. A sensitivity analysis of the microkinetic model identified Tsurwith the highest global sensitivity coefficient among Tsur, Tvib, and Ea,Rifor reactions R1-R10 (Figs. 12, 13, 14A-14C). The best fit Tsur for C2H6 for Ni (111) is 453 K, and for Cu (111) is 503 K. To address uncertainty in the measurement of Tsur, simulated C2 hydrocarbon yields were bounded between the values obtained at Tsur= 453 K and Tsur= 503 K (Fig. 4B).

[0436] Example 2:

[0437] Methane Partial Oxidation Experiments

[0438] Experiments for partial oxidation of methane were conducted with CH4 (99.97% pure), CO2 (99.9% pure), and H₂ (99.999% pure) flown into the reactor at various compositions. CH₄, CO₂, and H₂ were flown into the reactor at a range of flow rates (6-60 seem) through MFCs. The reactants were premixed upstream of the reactor with the help of a turbulent mixer. The reactor was placed on a hot plate, and the temperature of the hot plate was maintained at ~ 70°C to prevent condensation of the liquid fuels on the walls of the reactor. A cold water trap with 6 ml de-ionized water maintained at ~ 4 °C was placed downstream to capture methanol, ethanol, and acetic acid that was produced in the gas phase. The discharge gap was maintained at ~ 1 mm for these experiments. Catalyst powderAttorney Docket No. 10046-649W01

[0439] (-100 mg) dispersed on quartz wool was placed in the discharge gap. A DBD plasma was generated at - 4-5 kV, 20 kHz with - 30 % duty cycle. The gas phase reactant and product composition was sampled and quantified with an online GC / MS equipped with PoraBond Q. The downstream line of reactor going into the GC / MS inlet was heated to - 120 °C.

[0440] Experiments were initially performed with CH₄:CO₂ = 1:1 only using three catalysts (Al2O3, Ni / Al2O3, and CU / Al2O3). It was observed that while the selectivity of methanol was - 8% for Al₂O₃, the selectivities for Ni / Al₂O₃, and Cu / AhCh did not show any improvement due to early-on catalytic inactivation. With the addition of a vibrationally active co-reactant, like H2, the selectivity of methanol for Cu / Al₂O₃ increased from ~ 8% to 23% while it remained ~ 8% Al2O3. This enhancement is attributed to the reaction of H2 with adsorbed oxygen on the Cu surface, forming water that desorbs, thereby regenerating active sites. DRIFTS spectra confirmed that Cu / Al₂O₃, a noble and cost-effective catalyst, facilitates the formation of CH3O, a key intermediate in methanol synthesis.

[0441] Understanding reaction mechanisms on catalytic surfaces and whether they contribute to product formation in plasma environments is necessary to improve process performance in the conversion of CH₄-CO₂. Unlike thermal systems, where deactivation arises from well-defined mechanisms like sintering or coking, plasma environments can introduce distinct pathways governed by how plasma-excited reactants interact with the catalyst surface and the rate kinetics of surface-bound steps. After dissociative chemisorption, the strength of metal -adsorbate interactions determines whether the reaction intermediates participate in reaction pathways forming liquid oxygenates or lead to surface poisoning. To probe this, we classify catalysts by the DFT-calculated carbon binding energies (AEC, Fig. 41A) (Zeigarnik, A. V., et al. C-C bond scission in ethane hydrogenolysis. J Phys Chem B 104, 10578-10587 (2000); Nørskov, J. K., et al. Density functional theory in surface chemistry and catalysis. Proc Nat Acad Sci 108, 937-943 (2011)), and select Ni and Cu as model systems representing strongly and weakly binding regimes, respectively. This comparison allows us to link catalyst properties mechanistically to plasma-induced deactivation behavior and to evaluate whether plasma enables new modes of selectivity control across distinct catalyst regimes (Fig. 41B).

[0442] DFT calculations are performed on slab models of Cu (111) and Ni (111) to quantify surface reaction energetics relevant to CH4-CO2 plasma-catalytic conversion. In this analysis, we identify what each surface tends to accumulate once CH4 and CO2 species undergo dissociative chemisorption prior to subsequent surface reactions (Fig. 41C). DFT calculations reveal that the energetically preferred surface pathways differ significantly across catalystAttorney Docket No. 10046-649W01

[0443] classes but converge toward deactivation (Fig. 41D). Plasma-assisted dissociative chemisorption initiates surface reactions by coupling vibrational energy into gas-phase reactants, which lowers effective activation barriers and increases the rate initial bondbreaking steps irrespective of the catalyst binding strength (k ∝ e-(Ea-αEv)), where Eais the activation barrier of the forward reactions, Evis the energy present in the vibrational mode, a is the Fridman factor) (Mehta, P. et al. Overcoming ammonia synthesis scaling relations with plasma-enabled catalysis. Nat Catal 1, 269-2.75 (2018); Subhankar, V. S., et al. Plasma- Enabled Catalytic Design to Upgrade Methane Directly and Limit Coking. ACS Catal 15, 10433-10447 (2025)). After dissociative chemisorption, on strongly binding catalysts like Ni (111), the dehydrogenation of CH₄ (CH₄ → CH₃* → CH₂* → CH* → C* is favored due to the strong interaction between Ni and CH * intermediates (Fig. 41D). This leads to a rapid buildup of carbon on the surface, which blocks active sites and reduces catalytic activity. The resulting carbon deposits are stable and do not desorb unless the surface is regenerated. In contrast, CO2 activation on Cu (111) yields surface -bound CO* and O*, but O* is favored energetically to remain adsorbed and is not readily removed (2O* → O₂*, Ea = 2.05 eV) at the mild temperatures (TSUr = 473 K, 1 atm) used in plasma catalysis. This leads to oxygen poisoning of the surface. In both cases, DFT shows that the thermodynamically favorable pathways drive the system toward catalyst deactivation, either through carbon fouling on Ni or through oxygen accumulation on Cu, rather than enabling productive turnover (Fig. 42A). These distinct deactivation pathways highlight a shared limitation: both catalyst types ultimately lose access to key reaction intermediates needed for selective oxygenate formation.

[0444] Experiments were performed to demonstrate the catalytic deactivation in the transition metal catalysts during CH4-CO2 conversion. These studies were designed to validate the distinct deactivation mechanisms predicted from DFT that were associated with strongly (Ni) and weakly binding (Cu) polycrystalline surfaces. A dielectric barrier discharge (DBD) plasma reactor was used to activate the reactants-CH₄ (99.97% pure) and CO₂ (99.99% pure)-and drive surface reactions under nonthermal conditions (Fig. 37). The DBD plasma operated at a frequency of - 20 kHz across two 1 mm-thick glass dielectric barriers separated by a 1 mm gap, with a discharge voltage of 4-5 kV (~ 30% duty cycle) and peak current of ~ 150 mA. The specific energy input (SEI) was varied between 0 and 16 eV per initial molecule. A total of 100 mg of supported polycrystalline transition metal catalysts (Al2O3, 20% (wt) Ni / AhCh, 20% (wt) Cu / Al₂O₃) were loaded into the reactor between glass dielectrics (~ 55% porosity). X-ray diffraction (XRD) and transmissionAttorney Docket No. 10046-649W01

[0445] electron microscopy (TEM) measurements confirmed a high abundance of (111) surface facets in both Ni and Cu catalysts. Before initiating plasma-catalytic reactions, the catalysts were pretreated in an H₂ plasma for ~ 30 minutes at a discharge voltage of 4-5 kV (30% duty cycle) and a peak current of - 150 mA. The reactant ratio was fixed at CH₄:CO₂ = 1:1 to match conditions typical of industrial reforming processes. The total flow rate (Q = 4-60 seem) was varied to control SEI and conversion. A 6 mL ice-water trap, maintained at ~ 4 °C, was placed downstream of the reactor to collect liquid products (Fig. 37). Gaseous species (reactants and products) were quantified using an online gas chromatograph and mass spectrometer (GC / MS) installed after the reactor and bypassing the liquid trap. Liquid¬ phase oxygenates were analyzed offline using mass spectrometry through direct syringe injection for different reaction times. Additional experiments were conducted by varying the feed composition, flow rate, and SEI to map the performance variation.

[0446] Surface-bound species formed during CH₄-CO₂ plasma-catalysis were characterized using X-ray photoelectron spectroscopy (XPS) and operando diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS). XPS measurements of reference samples (H₂ plasma- treated) and spent samples (recovered after plasma catalytic reactions for 200 min) were performed to quantify elemental composition and identify the chemical state of the adsorbed species. High-resolution XPS scans were collected for the C Is and O ls regions. The C Is spectra were deconvoluted into distinct components corresponding to saturated carbon (C-C, ~ 283.8 eV), sp²-hybridized graphitic carbon (C=C, ~ 284.4-284.6 eV), and oxygen-containing functional groups such as alcohols or carbonates (C-O / C=O, ~ 286.0-289.5 eV). The O ls spectra were resolved into signals attributed to adsorbed oxygen adatoms (~ 530.0 eV), surface lattice oxygen (~ 531.5 eV), and bulk oxygen or chemisorbed carbonate species (~ 533.0-533.5 eV) (Hayashida, K., etal. Revisit of XPS Studies of Supersonic O 2 Molecular Adsorption on Cu(lll): Copper Oxides. ACS Omega 6, 26814- 26820 (2021)). Peak areas were integrated following Shirley background correction and fit using Gaussian-Lorentzian profiles. Quantitative surface compositions were obtained by normalizing the peak areas using standard elemental sensitivity factors. Operando DRIFTS measurements were used to identify adsorbed reaction intermediates and deactivating surface species. Methoxy (1460 and 1480 cm'1), surface hydroxyl groups (3200-3700 cm'1), and mono- or bidentate carbonate (1384 cm'1and 1443 cm'1) groups were tracked (Taifan, W. E., et al. Surface chemistry of MgO / SiO 2 catalyst during the ethanol catalytic conversion to 1,3-butadiene: in-situ DRIFTS and DFT study. Catal Set Technol 7, 4648-4668 (2017); Meng, H. et al. Designing Cu0-Cu+ dual sites for improved C-H bondAttorney Docket No. 10046-649W01

[0447] fracture towards methanol steam reforming. Nat Commun 14, 7980 (202.3)). Spectra were baseline-corrected and normalized to the catalyst loading.

[0448] The selectivity toward liquid oxygenates (methanol, ethanol, and acetic acid) during exposure to plasma followed the trend C11 / Al2O3> Ni / AhCh > A 1 ( ) • but all systems converged rapidly to the baseline Al2O3performance within the first 3 minutes of plasma operation (Fig. 42B). For t > 3 min, no significant differences (~ 8-9%) in selectivity were observed across the transition metal catalysts and support, indicating early catalyst deactivation under CH4-CO2 plasma conditions. XPS analysis revealed distinct deactivation mechanisms for Ni- and Cu-based systems (Figs. 42C-42D). For Ni / Al2O3, changes in the O Is region were minimal, suggesting that oxygen accumulation was not the primary deactivation pathway. Instead, a 50% increase in normalized coke deposition-calculated from the C Is to Al 2p intensity ratio-was observed in the spent sample compared to the reference. This was accompanied by an increase of 62% in the normalized C-C / C=C peak area and 33% in the C-O / C=O component, indicating carbon buildup in both graphitic and oxygenated forms. Operando DRIFTS further confirmed this mechanism, showing a rapid accumulation of mono- and bidentate carbonates on Ni / Al₂O₃ (Fig. 42F) over the first 7 minutes of plasma exposure, after which the carbonate coverage plateaued (Figs. 42G-42H). The basicity of the Al2O3support stabilized CCF-derived surface species on Ni, facilitating carbonate formation and promoting carbonaceous coverage of the active sites. In contrast, Cu / Al₂O₃ showed a smaller increase (~ 20%) in coke deposition relative to the reference sample but exhibited significant changes in the O ls region. The normalized intensity of surface oxygen ad-atoms increased from 33% to 49% (~ 50% increase), while contributions from lattice and bulk oxygen decreased. These shifts indicate oxygen accumulation on the surface, consistent with oxygen poisoning as the dominant deactivation pathway. DRIFTS measurements for Cu / Al₂O₃ showed no substantial buildup of carbonates or other carbon- containing species, showing that Cu deactivates primarily through oxidative site blocking rather than coking (Figs. 42E, 42G, and 42H). These experimentally observed trends directly mirror DFT-predicted surface energetics, where CHx* stabilization is favored on Ni and O* stabilization is favored on Cu (Fig. 41D), providing a consistent mechanistic explanation for the distinct deactivation pathways.

[0449] These experimental and DFT results establish that plasma activation accelerates the buildup of surface intermediates but does not inherently prevent surface poisoning. This raises a key unresolved question: which plasma-generated species are responsible for populating the surface and driving these deactivation pathways? Sustained plasma-catalyticAttorney Docket No. 10046-649W01

[0450] performance depends not only on the ability of plasma to activate reactant molecules, but on which plasma-generated species can actually reach and react with catalyst surfaces under operating conditions. Surface deactivation and regeneration arise from competing surface reactions initiated by plasma-activated species, and these outcomes depend sensitively on whether vibrationally excited molecules or radicals couple to surface. If radical -driven pathways govern the surface deactivation mechanism, surface poisoning may be unavoidable. If vibrationally excited molecules govern the deactivation mechanism, surface reactions may instead be controlled by adsorbate-mediated chemistry that can, in principle, be controlled. Establishing the physically relevant plasma-surface coupling pathways is therefore a prerequisite for interpreting deactivation behavior and for identifying conditions under which regeneration can be achieved in situ.

[0451] To establish this prerequisite, we develop a coupled plasma kinetic and DFT-informed surface microkinetic model (MKM) to identify which plasma-surface coupling pathways govern the reaction mechanism under non-equilibrium conditions. We first quantify how the plasma- activated species participate in surface chemistry. A plasma-phase kinetic model was solved as a function of electron temperature (Ts= 1-6 eV) to compute number densities of electrons, positive ions, neutral radicals, stable molecules, and vibrationally excited states of CH4, and CO2. The plasma model explicitly accounts for electron-impact vibrational excitation, dissociation, ionization, dissociative recombination, vibrational relaxation, charge transfer, and neutral -radical recombination processes. The resulting steady-state densities of long-lived plasma-generated species that can reach the catalyst surface are then used as gas-phase inputs to a coupled plasma- surface microkinetic model on Ni(H l). Other physicochemical interactions such as electric field effects, surface charging, and ion bombardment are not considered within the model for the rest of this manuscript.

[0452] The surface microkinetic model is initialized using these plasma-derived species densities (-7.5 x 1019total initial molecules) with a clean catalyst surface containing a fixed number of active sites (-4.2 x 1019). The surface state vector tracks coverages of adsorbed CHXspecies, C2 intermediates, O, OH, H, CO, oxygenated intermediates, and vacant sites, while gas-phase pools track CH₄, CO₂, H₂, CO, H2O, O2, C2 hydrocarbons, and plasmagenerated radicals. Plasma-activated species are converted through adsorption and surface reaction terms, while the plasma-phase production terms remain active throughout the simulation. All simulations were performed at 1 atm and Tgas= Tsur= 473 K, and surface reaction rates were evaluated up to 3% CH₄ conversion to ensure comparison underAttorney Docket No. 10046-649W01

[0453] identical surface states and gas compositions. This coupled framework enables a direct comparison of methane activation and CH4-CO2 coupling pathways arising from vibrationally excited Langmuir-Hinshelwood reactions, radical-assisted Langmuir-Hinshelwood reactions, and selected Langmuir-Rideal reactions. By holding surface coverage and conversion fixed, differences in computed rates directly reflect differences in the plasma-generated species populations and their coupling mechanisms, allowing quantitative assessment of how vibrational excitation and radical chemistry contribute to surface activation under CH4 / CO2 plasma conditions (Fig. 43A). This assessment allows us to filter plasma-surface coupling pathways quantitatively using kinetic rate and lifetime comparisons, and to use this reduced mechanism to interpret all subsequent experiments.

[0454] Results from coupled microkinetic simulation for CH₄:CO₂ = 1:1 show methane activation proceeds through dissociative adsorption of vibrationally excited CH₄ across the entire Terange examined (1-6 eV)(Figs. 43B-43C). At Te= 1 eV, the rate of CH₄(v) → CH₃* + H* exceeds the radical-assisted pathway by ~ 108times (Fig. 43B). Even at elevated Te= 4 eV, where gas-phase radical production increases, vibrationally enabled L-H activation remains dominant by at least two to three orders of magnitude on Ni( 111).

[0455] Consequently, methane activation under CH4 / CO2 plasma conditions occurs exclusively through dissociative adsorption rather than gas phase dissociation and subsequent adsorption, establishing a vibrationally controlled surface-reaction regime under atmospheric DBD operation. Similarly, it is found that CO₂ activation proceeds through CO₂(v) → CO₂* first, then it splits on the surface as CO* and O* (Fig. 43C). These rate trends are consistent with the characteristic timescales of plasma-generated species.

[0456] Vibrationally excited CH₄(v > 0) relaxes via vibrational -translational (V-T) relaxation within 1.54-3.08 qs (k = 855 s^Torr'1), ( '()-• v. > relaxes within ~ 3.44 us (k = 1.88 x IO’14cm3s ) at p = 1 atm and Tsur= 473 K. CH4 dissociation to CH3* and H* (Ea = 1.89 eV) initiates hydrocarbon chemistry and further hydrocarbon buildup on the clean catalytic sites, while CO2* dissociation to CO* and O* (Ea = 0.56 eV) provides surface oxygen species that can inhibit reactions leading to sustained oxygenate formation. The presence of surface carbonates from the DRIFTS results provides direct evidence for reactions involving vibrationally excited CO2 interacting with the catalyst surface, since carbonate (CO32-) formation is most consistent with molecular CO2 adsorption / activation pathways instead of CO2 dissociation to CO or atomic O. In contrast, radicals exhibit short lifetimes due to rapid recombination. CH₃ radicals recombine quickly (CH₃· + CH₃· → C₂H₆, k = 2.49 x 10-10cm3s-1), yielding estimated lifetimes of - 0.4 qs at number densities of 1021m-3), consistentAttorney Docket No. 10046-649W01

[0457] with population densities in a typical DBD reactor (Maitre, P. A., et al. Plasma-Catalysis of Nonoxidative Methane Coupling: A Dynamic Investigation of Plasma and Surface Microkinetics over Ni(111). J Phys Chem C 126, 19987-20003 (2022)). As a result, the steady-state flux of vibrationally excited molecules to the surface exceeds that of hydrocarbon radicals under the same plasma conditions.

[0458] This microkinetic analysis clarifies which plasma-surface interactions matter under mild operating conditions. The results show that the surface poisoning observed experimentally is not caused by direct adsorption from gas-phase radicals. Instead, deactivation arises from the buildup of surface-bound intermediates formed after vibrationally excited reactants adsorb and dissociate on the catalyst. This distinction is important because it shows that surface poisoning under plasma conditions is not unavoidable, but depends on whether adsorbed species can be removed once they form. By identifying vibrationally assisted L-H pathways as the primary mode of plasma-surface coupling, the analysis defines a regime in which surface regeneration can be achieved through controlled surface reactions rather than by adjusting plasma parameters in the design space.

[0459] Together, the results from DPT, experiments, and microkinetic modelling highlight that as these surface species saturate the active metal sites, the catalytic behavior becomes indistinguishable from that of the inert Al2O3support, which lacks metal sites to enable dissociative chemisorption followed by stabilization of the reactive intermediates for desired products. Although plasmas increase the kinetics of adsorption reactions (by changing the population fractions of vibrationally active reactants), they fail to influence the kinetics of desorption, leading to active site saturation and catalytic deactivation. This rapid deactivation highlights a fundamental limitation of CH4-CO2plasma-catalysis: without an in situ regeneration mechanism, the catalytic advantage of transition metals can be lost quickly under continuous plasma operation.

[0460] Plasma-Enabled Active Site Regeneration During CH4-CO2 Conversion Building on the observed accumulation of surface-bound species (e.g., CHx*, O*), which block active sites and suppress selective product formation even as the rate kinetics of reactant chemisorption are increased, a need emerges for their continuous removal to sustain catalytic activity under steady-state CH4-CO2conversion. While traditional regeneration strategies require periodic thermal cycling, plasma systems offer an opportunity to regenerate active sites continuously in situ by leveraging the steady-state generation of short-lived reactive species. We investigate how plasma-enabled active siteAttorney Docket No. 10046-649W01

[0461] regeneration, achieved through co-reactant excitation (e.g., H2), alters catalyst surface chemistry during CH4-CO2 conversion across different metal -adsorbate binding regimes (Fig. 44A). Because CH4, CO2, and H2are all dissociative! y chemisorbed by the plasma upstream of surfaces that are maintained at near ambient conditions, the surface is continuously supplied with species that can both react and self-clean.

[0462] To demonstrate the idea of plasma-enabled active site regeneration, a series of experiments was conducted using Ni / Al2O3, C11 / Al2O3, and Al2O3catalysts in a dielectric barrier discharge (DBD) reactor (p - 1 atm, Tgas~ 473 K, Tvib~ 4200 K, V=4-5kV, Fig. 44B). Two testing conditions were employed: with and without H 2 co-feeding. In the “without H2” condition, the experiment consisted of three sequential phases: (a) a purge phase where the reactor was flushed with a CH4 / CO2 mixture (1:1 ratio) at a total flow rate of 20 seem for 45 minutes, during which baseline water and hydrocarbon levels were recorded; (b) a plasma-catalytic phase from t = 45 to 90 minutes under CH4 / CO2 plasma, during which time-resolved quantification of water and hydrocarbon production was performed; and (c) a regeneration phase, where CH4 / CO2 flow was replaced with pure H2(20 seem), and the evolution of product species was monitored. For the “with H2” condition, the three-phase sequence followed a similar structure. During the purge phase, a CH4 / CO2 / H2 mixture (1:1:4 ratio) was introduced at 20 seem for 45 minutes to establish baseline product concentrations. This H2 concentration reflects realistic levels in recycled or unutilized feedstock streams, such as from upstream reforming or syngas processes, where residual hydrogen can be activated by plasma to selectively remove surface-bound poisons without altering equilibrium product desorption kinetics appreciably. More broadly, this represents a strategy to repurpose components of waste streams as targeted co-reactants for in situ catalyst regeneration. A lower bound CH4 / CO2 / H2 ratio of 1:1:1 represents the stoichiometric minimum (considering CH4+CO2+H2→CH3OH+H2O) required to eliminate all surface-bound oxygen by H2O formation and maintain catalytic activity.

[0463] Plasma-catalytic operation was then carried out between t = 45 and 90 minutes under the same ternary gas mixture. In the final phase, CH4and CO2were removed from the feed while maintaining a flow of pure H2(20 seem), allowing direct assessment of regeneration behavior under plasma-excited 112 alone.

[0464] Experimental results from the “without H2” case show that II2O and hydrocarbon signals remained constant during both phase a (CH4 / CO2 purge) and phase b (CH4 / CO2 plasma) across all catalysts, indicating minimal catalytic activity or surface changes in the absence of hydrogen (Fig. 44B). However, during phase c, when H2plasma was introduced,Attorney Docket No. 10046-649W01

[0465] an increase in water production was observed specifically on CU / Al2O3. The water signal steadily rose from t = 90 to 120 minutes (up to ~ 0.5%), plateaued between t = 120 and 150 minutes, and then gradually returned to baseline. This trend indicates that surface -bound oxygen, accumulated on Cu during phase b, was removed through reaction with plasma- excited hydrogen, forming gas -phase H2O. Assuming that each H2O molecule corresponds to the removal of one surface oxygen atom, thereby liberating one active site on CU / Al2O3, the number of regenerated sites was quantified by integrating the cumulative H2O signal during phase c (Fig. 44C). Similarly, on Ni / AbOs, site regeneration was estimated by integrating the cumulative hydrocarbon (C1+C2) signals, based on the assumptions that each CH4molecule desorbing corresponds to one active site regenerated and each C2H6molecule corresponds to two active sites regenerated (Fig. 44C). These calculations enabled a quantitative comparison of catalyst reactivation dynamics driven by plasma-excited hydrogen.

[0466] It was observed that the total number of active sites regenerated during phase c for both Cu / AhOs and N1 / Al2O3approached the independently measured number of active sites from the pulsed chemisorption experiments present in the reactor prior to the onset of plasma-catalytic operation (t = 0). This confirms true active recovery rather than transient surface cleaning. In the “with 11 “ case, where hydrogen was co-fed throughout all three phases, C11 / Al2O3exhibited a steady rise in water production during the plasma-catalytic phase (phase b), reaching a maximum of ~ 0.15%, followed by an immediate return to baseline upon switching to H2-only plasma in phase c (Fig. 44B). This sustained increase in water formation during phase b indicates that plasma-excited H2continuously removed surface oxygen as it accumulated, effectively maintaining the availability of active sites and preventing deactivation. While this calculation relies on a stoichiometric mapping between desorbed products and freed sites, the agreement in magnitude with independent chemisorption site counts supports the interpretation of active site recovery. In contrast, Ni / Al2O3showed a slight drop in water levels below baseline during phase b, while simultaneously exhibiting a gradual increase in C1+ C2hydrocarbon production (Fig. 44B). This behavior is consistent with suppression of carbon deposition and reactivation of CHXcoupling pathways facilitated by in situ hydrogenation.

[0467] To confirm the regeneration of transition metal active sites during the plasma- enabled active site regeneration, XPS measurements were performed on the Ni / AhCh, C11 / Al2O3, and Al2O3catalysts (Figs. 45A-45D). High-resolution spectra were acquired for the 0 Is, C Is, and Al 2p regions for all catalysts, while the Cu 2p and Ni 2p regions wereAttorney Docket No. 10046-649W01

[0468] additionally collected for Cu / Al2O3and Ni / Al2O3, respectively, to probe the oxidation state and surface composition of the active metals. Each catalyst was analyzed under three distinct conditions: a) reference samples-H2plasma-treated catalysts prior to CH4-CO2 exposure; b) w / o H2-catalysts exposed to CH4 / CO2 plasma at a 1:1 ratio (Q = 20 seem) for 200 minutes without hydrogen; and c) w / H2-catalysts exposed to CH4 / CO2 / H2 plasma at a 1:1:4 ratio (Q = 20 seem) for 200 minutes. This comparative framework as confirmed by DPT computations (Fig. 44D) enabled the direct assessment of surface chemical changes associated with catalyst deactivation and reactivation with the help of a co-reactant at mild thermodynamic conditions. We note that in typical thermal catalytic conditions, H addition often shifts the equilibrium unfavorably to undesired products like CH4(favoring methanation (CO2+ 4H2→ CH4+ 2H2O) over oxygenate formation) and does not selectively remove surface poisons (Ye, R.-P. et al. CO2 hydrogenation to high-value products via heterogeneous catalysis. Nat Commun 10, 5698 (2019)).

[0469] The addition of hydrogen as a co-reactant on the strongly binding Ni / Al2O3catalyst resulted in a reduction in surface carbon accumulation (Figs. 45A-45D). XPS analysis revealed that the normalized C Is / Al 2p intensity ratio decreased from 0.8 in the w / o H sample to 0.4 in the w / II2 sample-a 50%’ reduction in surface carbon-while the H2 plasma- treated reference sample exhibited a ratio of 0.2 (Fig. 45A). This progressive decrease suggests that plasma-excited H2 facilitates the removal of carbonaceous species during CH4-CO2 conversion, leading to active site regeneration and restoring the catalyst surface closer to its original state. Furthermore, deconvolution of the C Is spectra showed proportional decreases in the sp’-hybridized and sp iybridized carbon peaks, indicating the removal of both types of coke deposits from Ni active sites (Fig. 45A). On the weakly binding Cu / Al2O3catalyst, no significant change in the normalized C Is / Al 2p intensity ratio was observed between the w / o H2and w / H2samples, indicating that carbon deposition was not a dominant deactivation mechanism (Fig. 45B). However, substantial differences were observed in the oxygen coverage (O Is) and copper oxidation state (Cu 2p) between these samples (Figs. 45C-45D). Deconvolution of the O Is spectra revealed that the normalized surface oxygen adatom fraction decreased from 50% in the w / o H2sample to 33% in the w / H2sample, aligning with the value observed in the H2plasma-treated reference. This reduction indicates that the original chemical state of the Cu surface is largely preserved during plasma-catalysis when H2is co-fed, due to continuous removal of adsorbed oxygen. The decrease in surface oxygen ad-atom signal was accompanied by a proportional increase in contributions from surface-bound and bulk lattice oxygen (due to the presence of Al2O3Attorney Docket No. 10046-649W01

[0470] support). In parallel, the Cu 2p spectra showed a shift in the main Cu 2p3 / 2peak from 932.0 eV to 931.8 eV with H2co-feeding, along with suppression of the Cu2+satellite peak (Fig.

[0471] 45D). These changes indicate a partial reduction of the copper surface. Quantitatively, the fraction of Cu° increased from ~ 50% in the w / o H2sample to over 80% in the w / H2sample, confirming effective removal of oxygen species from the Cu surface under plasma- excited hydrogen, thereby maintaining the catalyst in its reduced, active state (Fig. 45D). The restored Cu° fraction and reduced O* coverage are consistent with the low barriers for O* hydrogenation predicted by DPT (Fig. 44D).

[0472] These observations show that hydrogen does not introduce a new activation pathway for CH4or CO2but instead reacts with surface-bound species formed during plasma- assisted dissociative chemisorption. Its role under plasma conditions, therefore, differs fundamentally from that in thermal catalysis, where H2addition typically promotes methanation and suppresses oxygenate formation. In contrast, under nonthermal plasma operation, hydrogen functions as a selective scavenger that restores access to metal active sites without altering the plasma-driven activation of the reactants. This behavior arises from the way nonthermal plasmas activate molecular co-reactants. Electron impact dissociation and vibrational excitation continuously generate reactive hydrogen species, which, unlike in thermal systems, are replenished under steady plasma operation and can interact directly with surface-bound poisons. As a result, hydrogen can participate in surface reactions that remove accumulated intermediates at mild temperatures, rather than shifting overall reaction equilibria.

[0473] To determine which surface pathways enable this regeneration and why it is effective under plasma conditions, we next analyze co-reactant-assisted surface reactions using the coupled plasma-surface microkinetic framework. Microkinetic simulations were performed for CH4: CO2: H2= 1:1:4 to quantify how hydrogen interacts with poisoned surfaces under low-temperature plasma conditions (Figs. 46A-46C). At low Te= 1 eV, hydrogen activation occurs primarily through dissociative adsorption of vibrationally excited H 2 (relaxes within 200-390 ps (k = 5000-2500 s’1atm’1)(Fig. 46A), producing surface H*. These H* species react with surface O* to form OH*, followed by OH* hydrogenation to H2O*, which subsequently desorbs as H2O(g). This L-H pathway H* + O* → OH* + * proceeds at a higher rate (~102times) than the direct radical reaction H(g) + O* —> OH*, and simultaneously regenerates a vacant surface site (Fig. 46B). Subsequent OH* + H* → H2O* reactions further reduce surface oxygen coverage (Fig. 46B). Together, these steps provide a surface-mediated route for oxygen removal under CH4 / CO2 / H2 conditions.Attorney Docket No. 10046-649W01

[0474] Although gas-phase O and H radicals persist longer than hydrocarbon radicals, their contribution to poison removal remains limited At number densities of 1020-1021m-3, recombination timescales for O· + O· → O2(k = 1.22 x 10-13cm3s-1) and H· + H· → H2(k = 1.37 x 10-13cm3s-1) are on the order of 7-80 ms. These timescales are longer than those of CHs radicals and allow O and H radicals to participate in surface reactions when present. Nevertheless, their contribution to overall surface reaction rates remains lower (at least by 1 order) than that of vibrationally enabled site-mediated pathways within the electron¬ temperature range consistent with experimentally measured vibrational populations (Chen, T. Y. et al. Time-resolved characterization of plasma properties in a CH 4 / He nanosecond-pulsed dielectric barrier discharge. J Phys D Appl Phys 52, 18LT02 (2019)). In the coupled simulations, we also observe that adding H2 does not change the dominant CH4(v) > and CO2(v) dissociative adsorption pathways.

[0475] Together, experimental observations and microkinetic analysis show that plasma-enabled regeneration is governed by reactions between vibrationally excited molecules and surface -bound species. This result explains why plasma catalysis can sustain activity despite rapid surface poisoning: regeneration is possible when surface-bound intermediates are selectively removed as they form. The interaction of plasma-excited H2with catalyst surfaces is found to be highly dependent on metal binding strength. On Ni, which strongly stabilizes CHx* species, surface hydrogenation can enable removal of carbon through desorption as C2H6 (Fig. 44A). On Cu, which preferentially accumulates O*, plasma-excited H2can facilitate the removal of surface-bound oxygen by forming H2O (Fig. 44A). These site-specific regeneration reactions occur in situ and under reaction conditions without the need for thermal cycling or oxidative pulses. This regeneration process could restore access to the metal surface for continuous dissociative chemisorption of CH4 and CO2 and subsequent formation of key surface intermediates required for selective product formation. Together, these findings demonstrate that co-reactant excitation enables site¬ specific active site recovery through distinct mechanistic routes depending on catalyst binding strength.

[0476] Promoting Intermediates for Selective Liquid Fuel Formation with Plasma- Enabled Active Site Regeneration

[0477] With the regeneration of active sites, the reactants can dissociatively chemisorb on the catalytic surfaces continuously without poisoning the active sites. This system can further be tuned to promote and stabilize actively selective surface intermediates, unlike thermal conditions, where the surface coverage of reaction intermediates is governed by aAttorney Docket No. 10046-649W01

[0478] single gas-phase temperature and strong coupling between adsorption and desorption rates. Once surface poisoning is removed, the excited reactants can undergo dissociative chemisorption, forming intermediates such as CH3*, CH3O*, or COOH*, depending on the catalytic surface properties and the presence of co-reactants like H2(Fig. 47A). To establish the reaction mechanism in such catalytic systems, in situ DRIFTS measurements of a modified DBD reactor were performed on CU / Al2O3(weakly binding) and Ni / AFOs (strongly binding) catalysts. The catalysts were exposed to a CH4: CO2: H2 feed mixture (1:1:4) under dielectric barrier discharge (DBD) plasma at ambient temperature and atmospheric pressure. Spectra were collected continuously at every - 2 min (till 30 min) under plasma-on conditions to capture real-time evolution of surface-bound species.

[0479] Operando DRIFTS spectra collected under CH4-CO2 plasma conditions revealed distinct differences in gas-phase CO evolution between C11 / Al2O3and Al2O3catalysts. On Al2O3, a sharp gas-phase CO signal at ~ 2243 cm-1appears immediately upon plasma ignition (t = 2 min), indicating CO formation through gas-phase dissociation (Meng, H. et al. Designing Cu0-Cu+ dual sites for improved C-H bond fracture towards methanol steam reforming. Nat Commun 14, 7980 (2023)). In contrast, on CU / Al2O3, no CO signal is detected for the first 5 minutes of plasma operation (Fig. 47 A). A dual -peak CO feature emerges only after t = 5 minutes and gradually intensifies with time, suggesting that Cu loses its catalytic activity rapidly within minutes of exposure to CH4-CO2 plasma. This delayed onset of CO formation implies that surface deactivation on Cu-due to oxygen poisoning prevents dissociative chemisorption of CO2 and CH4 (Fig. 47 A). As a result, CO2dissociation shifts to the gas phase, consistent with our prior results and shifting of selectivity towards CO (an undesired product) from desired liquid fuels. Under CH4-CO2-H2 plasma conditions, the suppression of gas-phase CO formation on C11 / Al2O3suggests that surface reaction pathways are dominant and that CO2 dissociation is now taking place primarily at the catalyst interface (Fig. 47 A). This shift enables the buildup of surfacebound intermediates, as evidenced by the gradual increase in OH* stretching bands (3200-3700 cm-1) and δ(CH3) bending modes at 1460 and 1480 cm-1, which are characteristic of methoxy (CH3O*) species (Figs. 47B-47C). The presence of OH* corresponds to water formation from surface hydrogenation, while the formation of CH3O* indicates that the regenerated Cu sites are actively participating in methanol-related pathways. In contrast, DRIFTS analysis on Ni / Al2O3shows negligible growth in these intermediate signals, despite regeneration of active sites through coke removal (Figs. 47D-47E). This suggests that, unlike Cu, the regenerated Ni surface does not stabilize CH3O* or OH* species andAttorney Docket No. 10046-649W01

[0480] instead could favor C-C coupling reactions.

[0481] Together, the time-resolved DRIFTS data demonstrate that plasma-enabled regeneration does not only restore activity but also selectively stabilizes different surface intermediates depending on metal binding strength, linking regeneration chemistry directly to potential selectivity enhancement. On Cu, regeneration enables stabilization of oxygenated intermediates like CH₃O* and promotes methanol formation, while on Ni, reactivation is linked to C-C coupling and hydrocarbon production. This highlights the ability of plasma-co-reactant systems to tune intermediate speciation and product selectivity by coupling surface dynamics with catalyst identity.

[0482] Plasma-Catalytic Design Space for Liquid Fuel Synthesis

[0483] Plasma-catalytic CH4-CO2 conversion operates within distinct catalytic regimes defined by the metal surface’s binding characteristics and how they interact with plasma- excited species. The multidimensional design space can be understood as the intersection of three primary control knobs: plasma properties (e.g., electron temperature, vibrational excitation), catalyst surface properties (e.g., carbon and oxygen binding energies), and gas¬ phase parameters (e.g., pressure and co-reactant composition) (Fig. 48A). Within this plasma-enabled active site regeneration framework, two distinct catalytic regimes emerge: (i) a weakly binding regime, represented by Cu / Al2O3, which intrinsically favors oxygenate formation but suffers from surface oxygen accumulation that deactivates active sites; however, the introduction of plasma-excited hydrogen enables continuous removal of surface oxygen, thereby sustaining catalytic activity and enhancing methanol selectivity, (ii) A strongly binding regime, exemplified by Ni / Al2O3, which favors CH4activation but readily accumulates carbonaceous species, leading to coke formation and deactivation. With co-reactant addition, surface carbon is hydrogenated and removed, restoring activity, but the product distribution shifts away from oxygenates toward C2 hydrocarbons due to favorable CHx* coupling pathways on the regenerated Ni surface. The time evolution of oxygenate selectivity is studied pre and post active site regeneration on Cu / Al2O3, Ni / Al2O3, and their trends are compared to the baseline trends of Al2O3(Fig. 48B). Under H2 co-feeding, CU / Al2O3exhibits a distinct recovery in oxygenate selectivity (~ 25%), indicating that hydrogen reacts with O* to regenerate active Cu° sites. In contrast, Ni / Al2O3does not show similar recovery in oxygenate formation, suggesting a shift in pathway rather than full reactivation. These findings show that regeneration pathways are catalyst-specific and critically depend on the metal-adsorbate interaction strengths.

[0484] The observed product selectivity over C11 / Al2O3and Ni / AbOs is dictated byAttorney Docket No. 10046-649W01

[0485] branching mechanisms governed by surface reaction energetics (Fig. 48C). Product selectivity over Cu / Al2O3is determined by surface reaction energetics that favor oxygenate formation while suppressing C-C coupling. On Cu (111), the dominant methanol formation pathway proceeds via CH3* + O* → CH3O*, which is both kinetically accessible (Ea= 0.23 eV) and thermodynamically favorable (AE = -0.53 eV). The subsequent hydrogenation of CH₃O* to CH3OH* and desorption of CH3OH(g) completes the pathway efficiently. In contrast, the alternative route via CH3* + OH* → CH3OH* is less likely to occur due to its higher activation barrier (Ea= 1.95 eV). Under CH4-CO2-H2 plasma, H - dissociation to H* on Cu is facile (Ea = 0.33 eV, AE = -0.50 eV), leading to a high surface concentration of H*. This promotes selective removal of surface oxygen via two competing pathways. First, O can be hydrogenated to OH* (H* + O* → OH*, Ea = 0.72 eV, AE = -0.55 eV), followed by OH* + H* → H2O* (Ea = 1.05 eV, AE = -0.12 eV), and H2O subsequently desorbs (AE = -0.40 eV). This two-step sequence provides an efficient route for continuous O scavenging via H2O formation, thereby regenerating active Cu° sites. Second, O* can directly react with CH 3* to form CH3O*. further coupling site regeneration with methanol synthesis. In contrast, OH* is not favored as a reactant for CH3* due to its higher coupling barrier.

[0486] Cu surfaces bind CHx* intermediates weakly, limiting further dehydrogenation and C-C coupling. The conversion of CH3* to CH 2* requires a high activation energy of 1.31 eV, making this pathway kinetically unfavorable. This energetic barrier constrains hydrocarbon formation and allows surface CH * to be selectively converted into oxygenates. DFT calculations show that formation of higher oxygenates such as ethanol (CII3* + CH2O* CH3CH2O*, Ea = 0.91 eV) and acetic acid (CII3* + COOH* --> CII3COOII*, Ea = 1.94 eV) are also accessible but require more energy than the methanol pathway. CH4activation on Cu (CH4* + * → CH3* + H*, Ea = 1.47 eV) is the rate -limiting step under thermal conditions but becomes more favorable under plasma due to vibrational excitation of CH 4. This leads to high surface coverage of CH3*, and in the presence of plasma-excited H2, accumulated O* is removed as H2O, regenerating active sites. The restored surface then promotes CH3* + O* → CH3OH* pathway, leading to higher methanol selectivity, as confirmed by the carbon balance plot (Fig. 48D). These mechanistic insights are consistent with experimental observations, where C11 / Al2O3exhibits the largest increase in methanol production under CH4-CO2-H2plasma conditions.

[0487] On Ni (111), the addition of H2under plasma conditions fails to enhance liquid oxygenate production due to high kinetic barriers and poor thermodynamic driving forcesAttorney Docket No. 10046-649W01

[0488] associated with oxygenate-forming pathways. Methanol synthesis via CH3* + OH* → CH3OH* has a high activation energy of 1.94 eV and a mildly endothermic reaction energy (AE = +0.40 eV), while the alternative route CH3O* + H* → CH3OH* also remains kinetically hindered (Ea= 1.43 eV) and thermodynamically neutral (AE = +0.42 eV). These barriers make CH3OH formation unfavorable under standard surface temperatures.

[0489] Similarly, ethanol synthesis through CH3* + CH2O* → CH3CH2O* and subsequent hydrogenation (CH3CH2O* + H → CH3CH2OH*, Ea = 1.47 eV) also faces kinetic limitations. Although acetic acid formation via CH3* + COOH* → CH3COOH* is kinetically accessible (Ea = 0.68 eV), this pathway is secondary and does not benefit from co-fed H2.

[0490] In contrast, C-C bond formation and hydrocarbon release are highly favorable on Ni. The dominant coupling pathway proceeds through CH3* + CH2* → CH3CH2* (Ea = 0.61 eV, AE = -0.35 eV), which is readily accessible at plasma-modified surfaces. The subsequent hydrogenation to ethane (CH3CH2* + H* → C2H6↑) is strongly exothermic (AE = -0.95 eV) and requires only 1.20 eV, a barrier much lower than for any oxygenate¬ forming step. While CH3* + CH3* → C2H6↑ is theoretically possible, its activation barrier (Ea = 3.11 eV) renders it inaccessible. Instead, CH3* dehydrogenates to CH2* (Ea = 0.72 eV), which then couples with CH3* more favorably, confirming the dominance of the CH2* + CH3* route. Additionally, desorption barriers for hydrocarbons such as C2H4 are low (Ea = 0.11 eV), ensuring rapid removal from the surface and preventing site blocking.

[0491] Conclusion

[0492] Our results show that non-equilibrium plasmas can decouple bond-breaking dissociative adsorption (usually rate limiting step) from surface -bound chemistry, transforming CH4 + CO2 conversion at 473 K and 1 atm from essentially inert to highly selective through plasma-enabled active-site regeneration that exploits multiple plasma- excited co-reactants. The nonthermal nature of plasma provides access to vibrationally or electronically excited species and radicals that accelerate dissociative adsorption steps¬ circumventing the activation barriers that limit conventional thermal catalysis. These conclusions are supported by quantitative plasma kinetics, microkinetic rate comparisons, operando spectroscopy, and surface chemical analysis. In the absence of plasma, CH4and CO2remain essentially inert on both weakly binding (Cu) and strongly binding (Ni) catalysts at 473 K and 1 atm; only in presence of non-equilibrium plasmas the energetic electrons drive CH4and CO2dissociation, enabling subsequent surface chemistry. This decoupling, where dissociation rates are governed by plasma properties (electron energy,Attorney Docket No. 10046-649W01

[0493] density) and surface reaction rates by adsorption energies and surface temperature, simultaneously (1) activates weakly binding, earth-abundant catalysts such as Cu that would otherwise be inactive, (2) tailors reaction pathways via selective generation of CHx* and O*, and (3) mitigates surface poisoning through hydrogenation by vibrationally excited H₂. Our results show that plasma-excited co-reactants like H can enable in situ removal of poisoning intermediates-such as O* on Cu or CHx* on Ni-thus sustaining catalytic activity and shifting product selectivity. Surface temperatures and intrinsic metal-adsorbate interaction strengths govern whether regeneration restores oxygenate-producing pathways (Cu) or promotes CHXcoupling and hydrocarbon formation (Ni). 'These mechanisms were elucidated through experimental selectivity trends and supported by DFT calculations of activation and reaction energies, which quantitatively explain why certain pathways, such as CH₃* + O* → CH₃O* on Cu or CH₂* + CH₃* → C₂H₅* on Ni are favored under plasma conditions. This work expands the functional space of heterogeneous catalysis, offering new opportunities for low-temperature, energy-efficient conversion of inert molecules like CH4 and CO2 into value-added fuels and chemicals. These insights provide a foundation for designing plasma-assisted processes across a range of environmental and synthetic applications, including carbon utilization, distributed chemical manufacturing, and sustainable fuel production.

[0494] Materials and Methods

[0495] Reactor Design

[0496] A plasma-catalytic reactor was designed to study the conversion of CH4 and CO2 into liquid fuels. The reactor consisted of an aluminum frame with O-rings (dash number 022) and removable transparent windows (quartz and acrylic, 2 mm thick) for optical access. Compression lids ensured a gas-tight seal (Fig. 37). To maintain an inert reaction zone, a polytetrafluoroethylene (PTFE) sleeve was inserted inside the frame. The reactor was operated at atmospheric pressure. Plasma was generated by introducing electrical energy into the gas stream using two electrical feedthroughs (KJL Part EFT0513055, 1 / 8” NFF fitting). The reactor was configured for continuous flow operation, with reactants metered using a digital mass flow controller (MFC, Brooks Instrument, Model GF40). 'The residence time (τresidence= 1 s - 120 s) was controlled by varying the flow rate (1.875 seem - 50 seem) to optimize conditions for selective liquid oxygenate formation. Before initiating plasma, the reactor was purged for 15 - 20 minutes, and online gas chromatography-mass spectrometry (GC / MS) analysis was performed to establish a stable baseline composition. Gas-phase products were quantified by bypassing the cold trap and sampling the reactorAttorney Docket No. 10046-649W01

[0497] effluent directly to the online GC / MS. In this configuration, the outlet line was heated to prevent condensation, and the GC sampling loop received an undisturbed representation of the gas stream exiting the discharge. This mode was used to quantify stable gaseous species and light hydrocarbons, including CH4, CO2, CO, H2, and C2 products, and to establish carbon and hydrogen balances under operating conditions. Volatile liquid oxygenates were quantified by repeating the experiment under identical plasma power, temperature, pressure, and inlet flow conditions, but with a 6 mL ice- water cold trap placed immediately downstream of the reactor outlet. The trap was held at -4 °C to condense and retain oxygenated products while allowing permanent gases to pass through. After a prescribed collection period, the cold trap was isolated, and the condensed liquid was recovered for analysis to determine methanol, ethanol, and acetic acid production.

[0498] Plasma Generation

[0499] A dielectric barrier discharge (DBD) plasma was generated between two 1 mm-thick quartz dielectric plates using aluminum electrodes (dimensions: 10 mm x 5 mm x 1 mm) (Fig. 37). Electrical characteristics were monitored using a high-voltage probe (Tektronix 6015A), a current probe (Pearson Model 6585), and a mixed-signal oscilloscope (Tektronix MSO44, 200 MHz bandwidth). The discharge gap was maintained at 1 mm, and gas flow rates ranging from 1.875 to 50 seem with various CH4 / CO2 / H2 mixtures were used to modulate the specific energy input (SEI). The applied discharge voltage (V) was varied between 5 and 7 kV at a frequency of ~ 20 kHz and a duty cycle of ~ 30%, producing peak currents around 70 mA. Changes in flow rate had a minimal effect on the plasma power, resulting in less than 2% variation in SEI (Q = 1.875-50 seem).

[0500] Powder Catalysts

[0501] Plasma-catalytic experiments were conducted using Al2O3support powder and two supported catalysts: 20 wt% Ni / Al₂O₃ and 20

[0502]

[0503] C11 / Al2O3. The catalysts consisted of polycrystalline metal nanoparticles (~ 2-50 nm, dispersed on Al2O3particles with sizes ranging from ~ 10 to 500 pm. The metal dispersion ranged from 1% to 8%, and the corresponding active site densities were measured between 60 and 110 μmol / gcat.

[0504] Table 14. Active site characteristics of powder catalysts from pulsed chemisorption measurements.

[0505] Metal Dispersion (% bulk atoms), Loading (% wt), Active Sites (μmol / gcatalyst)

[0506] Catalyst Loading (% wt) Active Sites ( )

[0507] (% bulk atoms) S catalyst Ni / Al₂O₃ 2.1 20 68

[0508]

[0509] C11 / Al2O31.9 20 86Attorney Docket No. 10046-649W01

[0510] Catalytic Loading

[0511] Approximately 100 mg of catalyst was loaded into the reactor between two 1 mm-thick quartz dielectric plates. The catalyst bed was structured into four alternating layers, each separated by ~ 10 mg of quartz wool, to maintain uniform distribution and consistent spacing within the discharge region (~ 55% porosity). This layered configuration ensured even plasma exposure and reproducible catalytic performance throughout the reactor.

[0512] Analytical Characterization

[0513] Gas Chromatography

[0514] Gas-phase and liquid-phase products were analyzed using an Agilent 8890 gas chromatography-mass spectrometry (GC-MS) system equipped with a thermal conductivity detector (TCD) and a mass spectrometer operating in electron ionization (El) mode. The TCD was used to quantify permanent gases including CH4, CO2, H2, and CO, while all hydrocarbon and oxygenate products (e.g., C₂H₆, C₂H₄, C₂H₂, CH₃OH, C₂H₅OH, CH₃COOH) were detected and quantified using the mass spectrometer. Gas-phase samples were introduced using automated 250 μL gas sampling valves and manually using a gas¬ tight Hamilton syringe (1700 series) fitted with a Chaney adapter. Liquid-phase samples were collected downstream and manually injected into the MS for off-line analysis. Helium (99.999% purity) served as the carrier gas at 300 mL / min with a 500:1 split ratio. A HayeSep column was used upstream of the TCD and a Porabond Q column was used for product separation prior to MS detection. The GC oven temperature was initially held at 45 °C, ramped at 10 °C / min to a final temperature of 160 °C (0 min hold time). Product identification and quantification were performed by comparing retention times and fragmentation patterns against calibration standards.

[0515] Characterization of catalysts

[0516] CO pulse chemisorption experiments were conducted on Ni / Al2O3using a Micromeritics AutoChem III. Additionally, N2O pulse chemisorption experiments were performed on C11 / Al2O3using a Micromeritics AutoChem II 2920. For Ni / ALOs (~ 100 mg samples), initial reduction was performed at 773 K under a 10% H₂ / Ar gas mixture flowing at 50 ml / min for 30 minutes to clean and activate the metal sites. After reduction, samples were cooled to 300 K under a continuous He flow (99.999%, 50 ml / min), ensuring complete purging of residual H₂ / Ar until achieving a stable TCD baseline. CO chemisorption was conducted by injecting pulses of 10% CO / He mixture (injection volume ~ 0.54 ml) until consecutive peaks were consistent or until a maximum of 10 pulses was reached. For CU / Al2O3(~ 50 mg), samples were similarly reduced at 773 K with a 10% IL / Ar flow (50Attorney Docket No. 10046-649W01

[0517] ml / min) for 30 minutes. Cooling to the analysis temperature of 353 K was accomplished under a continuous He flow. N₂O pulse chemisorption was carried out alongside mass spectrometry to track N2O consumption. Pulses of N2O were administered repeatedly until peak consistency was reached or up to 50 pulses. Mass spectrometry data were captured at intervals of 0.1 seconds. Post-experiment, the samples were re- weighed. Mass spectrometer data were analyzed with MicroActive software, integrating signals at m / z = 28 and m / z = 44 to quantify the number of active sites.

[0518] Surface Diagnostics

[0519] Transmission Electron Microscopy (TEM) with Energy Dispersive X-Ray Analysis (EDX)

[0520] High-angle annular dark field scanning transmission electron microscopy (HAADF- STEM) and energy-dispersive X-Ray spectroscopy (EDX) were performed using a probe-aberration-corrected JEOL NEOARM TEM operated at 200 kV with an average beam current of 10 pA. Imaging was conducted with a dwell time of 10 ps, and EDX mapping was carried out using a JEOL large-angle silicon drift detector (0.96 sr). HAADF-STEM images were processed and analyzed using the Gatan Microscopy Suite, and particle size distributions were quantified using ImageJ. For sample preparation, 10 mg of H₂ plasma- treated catalysts were dispersed in 10 mL of high-purity methanol (99.99%). This mixture was further diluted to a concentration of 0.1 mg / mL by adding 1 mL of the suspension to an additional 10 mL of methanol. A 5 pL sample of the final solution was drop-cast onto ultrathin carbon-coated copper TEM grids (Ted Pella, 400 mesh, lacey carbon support) and allowed to dry for 5 minutes. The prepared grids were then stored under high vacuum (~ 1 × 10-7Torr) for a minimum of 24 hours prior to analysis.

[0521] X-Ray Photoelectron Spectroscopy (XPS)

[0522] X-Ray photoelectron spectroscopy (XPS) was performed using a VersaProbe 4 microprobe system equipped with a monochromatic, microfocused Al Kα X-ray source (hν = 1486.5 eV). To prevent surface contamination and oxidation, all samples were handled under inert conditions. Immediately after plasma-catalytic experiments, the reactor chamber was purged with an Ar / H₂ mixture, and the samples were transferred into an argon-filled glovebox. Samples were then sealed inside an airtight XPS transfer vessel and loaded into the spectrometer under ultra-high vacuum (UHV) conditions (base pressure ~ IO’9Torr). A charge neutralizer filament was employed to mitigate surface charging effects. Survey scans were collected with a 2 eV step size and a 10-minute acquisition time to determine the elemental surface composition. High-resolution spectra were acquired over ~ 250 pm x 250Attorney Docket No. 10046-649W01

[0523] pm areas using a 10 eV pass energy, 0.5 eV step size, and 45-minute sweep per region. Peak fitting and deconvolution were conducted using CasaXPS software, applying mixed Gaussian-Lorentzian (GL(30)) functions and a Shirley-type background with a variable offset for transition metal peaks.

[0524] Chemical state identification was performed by deconvolution of the C 1s and O 1s spectra based on established binding-energy assignments. In the C 1s region, components were assigned to surface -bound hydrocarbon fragments and oxygenated species, including CH3(285.1-285.2 eV), CH2(284.1-284.3 eV), CH (283.1-283.4 eV), graphitic or sp² carbon (284.4–284.6 eV), and carbonyl or carbonate species (C=O / O–C=O) at 288.0–288.4 eV. Oxygen-containing functional groups, including alcohols, ethers, and carbonates, were assigned within the broader C-0 and C=O envelope spanning 286.0-289.5 eV, consistent with prior reports for hydrocarbon and oxygenate species on transition-metal surfaces. The O 1s spectra were deconvoluted into contributions from adsorbed oxygen adatoms (~530.0 eV), surface lattice oxygen or metal–oxygen species (~531.3–531.7 eV), and higher binding energy components associated with bulk oxygen or chemisorbed carbonate species (~533.0–533.5 eV). These assignments enable discrimination between reactive oxygen species involved in surface oxidation chemistry and accumulated oxygen-containing surface groups. Quantification of surface functional groups was performed by integrating fitted peak areas following background subtraction and normalizing by elemental sensitivity factors provided by the instrument manufacturer. Relative surface coverages of carbonaceous and oxygen-containing species were obtained from the normalized C 1s and O 1s peak areas, enabling comparison of surface speciation across different plasma and co-reactant conditions.

[0525] X-Ray Diffraction (XRD)

[0526] Powder X-ray diffraction (XRD) measurements were performed to characterize the crystallographic structure of the Al2O3, C11 / Al2O3, and Ni / Al2O3catalysts. The data were collected using a Rigaku MiniFlex diffractometer equipped with a Cu Kα radiation source (λ = 1.5406 Å) operated at 40 kV and 15 mA. Scans were recorded in the 20 range of 10° to 80° at a step size of 0.02° and a scan rate of 2° / min. Catalyst powders were gently ground and evenly spread onto a low-background sample holder prior to measurement. Phase identification was carried out by comparing the diffraction patterns to standard reference data from the ICDD (International Centre for Diffraction Data) database. The XRD analysis was used to confirm the presence of crystalline phases and to assess metal dispersion or crystallite size differences across the catalysts.Attorney Docket No. 10046-649W01

[0527] Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) Operando diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was conducted on Al2O3, C11 / Al2O3, and Ni / Al2O3catalysts to investigate surface intermediate evolution during CH4-CO2 and CH4-CO2-H2 plasma-catalytic reactions. Experiments were performed using a DRIFTS cell (DiffusIR, PIKE Technologies) coupled to a JASCO 6800 FTIR spectrometer equipped with a liquid nitrogen-cooled mercury-cadmium-telluride (MCT) detector. The DRIFTS cell accommodated the plasma reactor. Spectra were collected before plasma ignition (t < 0 min) and continuously during plasma operation, with the catalyst exposed to a dielectric barrier discharge (DBD) plasma (V ≈ 15 kV, p = 1 atm, duty cycle = 30%). The total gas flow rate was maintained at 20 seem for both binary and ternary mixtures. Plasma was applied for 30 minutes under steady-state conditions, and post-reaction spectra were collected after shutdown. All spectra were baseline-corrected and converted using the Kubelka-Munk transformation. Surface species were assigned based on characteristic infrared bands: CH₃O* (δ(CH₃) at 1460–1480 cm-1), OH* (3200–3700 cm-1), monodentate carbonates (MC, ~ 1418 cm-1), and bidentate carbonates (BC, ~ 1544 cm-1)71,72Quantitative comparisons across catalysts and feed compositions were made by integrating normalized absorbance within these spectral regions.

[0528] Process Evaluation

[0529] Plasma Input Energy

[0530] Voltage and current waveforms were measured using a high-voltage probe (Tektronix P6015A, 75 MHz bandwidth) and a Pearson current probe (Model 6585, 250 MHz bandwidth), respectively. These signals were recorded with a Tektronix MSO44 mixed-signal oscilloscope (200 MHz bandwidth). The average power (P) input to the system was calculated using the time-averaged integral of voltage and current,

[0531] P̄ = (∫t₁t₂V(t)·I(t) dt) / (t₂ - t₁)

[0532]

[0533] t₂ - t₁, where V(t) is the applied voltage, I(t) is the measured current and t₂-t₁ is the time window used for averaging. A minimum of 10 discharge cycles within this time interval was used to quantify P to ensure statistically consistent power measurements. Plasma discharge power was also measured using the Lissajous method with a 470 pF capacitor placed in series with the reactor. The voltage across the capacitor, VCap, was measured using a Tektronix voltage probe (10× attenuation, 200 MHz bandwidth), while the voltage across the entire plasma reactor, VApp, was measured with a Tektronix P6015A high-voltage probe (1000× attenuation,Attorney Docket No. 10046-649W01

[0534] 75 MHz bandwidth). The charge on the capacitor was calculated as,

[0535] QCap= CCap× VCap. (2) The average discharge power was determined from the area enclosed by the resulting Lissajous figure,

[0536] ^avg duty ■ f V / jpp dQca ’ (2) where f is the frequency and "duty" refers to the duty cycle of the applied waveform. The resulting power (~ 1.8 W) agreed within 10% of values obtained from voltage-current waveforms (~ 2.0 W).

[0537] Yield and Selectivity Calculation

[0538] Based on gas chromatography measurements (Nallapareddy, C. R. & Underwood, T. C. Tailoring Vibrational Excitation Pathways for High-Yield Oxidation of Methane to Methanol. ACS Sustain Chem Eng 12, 9144-9155 (2024)), the conversion of reactants is calculated as

[0539] XCH₄= (moles of CH₄ converted / moles of initial CH₄) × 100%, (4)

[0540] and

[0541] moles of C02converted

[0542] = - - -. — x 100%. (5) “ moles of initial C O2

[0543] Carbon conversion was evaluated on a total-carbon basis using the combined consumption of CH4 and CO2, each containing one carbon atom per molecule,

[0544] (moles of CH4converted) + (moles of CO2converted)

[0545] XC= [(moles of CH₄ converted) + (moles of CO₂ converted)] / [(moles of initial CH₄) + (moles of initial CO₂)] × 100%. (6)

[0546] (moles of initial CH4) + (moles of initial CO2)

[0547] Carbon-based selectivity to product i was calculated from the fraction of converted carbon appearing in that product,

[0548] Si= νc,i(moles of i produced) / [(moles of CH₄ converted) + (moles of CO₂ converted)] × 100%,

[0549] where vc>iis the number of carbon atoms in product i.

[0550] Carbon-based yield to product i was defined as the fraction of inlet carbon appearing in that product,

[0551] Yi= νC,i(moles of i produced) / [(moles of initial CH₄) + (moles of initial CO₂)] × 100% (8) = Xc· Si / 100

[0552] Fuel production efficiency (r? / ,j) (Nallapareddy, C. R. & Underwood, T. C. What is “efficiency” in plasma chemical processes? iScience 28, 112297 (2025)) is defined as, Ytx LHV,

[0553] V

[0554]

[0555] f.t = SEIAttorney Docket No. 10046-649W01

[0556] Computational Evaluation

[0557] DFT Computations

[0558] Density functional theory (DFT) calculations were performed using the Vienna ab initio simulation package (VASP, version 6.4.3) with projector-augmented plane-wave pseudopotentials, using the Perdew-Burke-Ernzerhof generalized gradient approximation exchange-correlation functional (Kresse, G. & Furthmiiller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys Rev B 54, 11169— 11186 (1996); Perdew, J. P., et al. Generalized Gradient Approximation Made Simple. Phys Rev Lett 77, 3865–3868 (1996)). A plane-wave basis set is used to describe the valence electrons with an energy cutoff of 400 eV, and a 2x2x1 Monkhorst-Pack k-point grid was used for the integration of the reciprocal space. Spin polarization was especially taken into account with specifying initial magnetic moments as ferromagnetic for Ni and Cu atoms, while lighter atoms (O, N, C, H) were assigned zero initial magnetic moments (Wu, R. & Freeman, A. J. Spin-orbit induced magnetic phenomena in bulk metals and their surfaces and interfaces. J Magn Magn Mater 200, 498-514 (1999)). Ni and Cu bulk are both in Fm- 3m space group and the slab structures are made by cutting from fee 111 lattice plane. The slab models are at least 8 A in thickness with 4 layers of metal atoms separated by a vacuum layer of 12 A. For calculations regarding surface reactions, the bottom two layers in the slab were frozen to make sure it represents a bulk structure.

[0559] The geometry relaxation of each system was carried out to investigate various reaction rate pathways, focusing on the dissociation and partial oxidation of methane. All the possible configurations for adsorbates binding on surface were optimized until the forces on each atom were less than 0.01 eV / A, and electronic self-consistency was achieved to within 10’6eV. Transition state searches were performed using the nudged elastic band (NEB) method, through VASP transition state theory (VTST) tools package as implemented in VASP (Henkelman, G. & Jonsson, H. Improved tangent estimate in the nudged elastic band method for finding minimum energy paths and saddle points. J Chem Phys 113, 9978- 9985 (2000); Henkelman, G., et al. Climbing image nudged elastic band method for finding saddle points and minimum energy paths. J Chem Phys 113, 9901–9904 (2000); Sheppard, D., et al. A generalized solid-state nudged elastic band method. J Chem Phys 136, (2012)). A set of 5 images was generated between initial and final states (for shorter reaction pathways, we used 3 intermediate images), and these images were optimized simultaneously to map out the minimum -energy path (MEP), with a force convergence criterion of 0.02 eV / A. The climbing image NEB (CI-NEB) scheme was also used to refineAttorney Docket No. 10046-649W01

[0560] the highest-energy image, ensuring an accurate transition-state geometry.

[0561] The Gibbs free energy is calculated with gas correction by the equation,

[0562] ΔG = ΔE + ΔZPE − TsurΔS, (10) in which AE is the potential energy difference between the reactant and product molecules in each reaction step; AS is the change in entropy of each step and AZPE is the zero-point energy correction to the Gibbs free energy. Only the entropy of gaseous species were included in the calculations. All structures and energy profiles were visualized and analyzed using the atomic simulation environment's graphical user-interface (ASE-GUI) and visualization for electronic and structural analysis (VESTA) (Hjorth Larsen, A. et al. The atomic simulation environment--a Python library for working with atoms. J Phys Condensed Matter 29, 273002 (2017); Momma, K. & Izumi, F. VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data.. J Appl Crystallogr 44, 1272–1276 (2011)).

[0563] Evaluating Plasma Generated Species

[0564] The plasma phase was modeled using a zero-dimensional, transient kinetic framework to quantify gas-phase species densities relevant to plasma-catalyst coupling. The model solves coupled balance equations for electrons, neutral molecules, vibrationally excited states, radicals, and ions under atmospheric pressure conditions (1 atm, Tgas= 473K) for CH₄ / CO₂ and CH₄ / CO₂ / H₂ mixtures. Plasma gas heating is not modelled in this work. The number densities of all species in this model as obtained by solving reactions in time until they reach steady-state (~ ps). Number densities of molecular species nsare obtained from species balance equations of the form,

[0565] dns / dt = Σ csjRj(11)

[0566] where csj is the net stoichiometric coefficient of species s in reaction j, and Rj is the reaction rate. Electron density is computed from an analogous balance,

[0567] dne / dt = Σ cejRj(12)

[0568] with cejaccounting for electron production and loss through ionization and dissociative recombination, respectively. Reaction rates are evaluated using mass-action kinetics,

[0569] Rj= kj(Te, Tgas) ∏mnma(13)

[0570]

[0571] _ _

[0572] where kj is the rate coefficient and amis the reaction order of reactant m.

[0573] Electron -impact rate coefficients were computed from the electron energy distribution function obtained using BOLSIG+ with a complete and consistent CH4, CO2,Attorney Docket No. 10046-649W01

[0574] and H₂ cross-section set from the LXCat database (Hagelaar, G. J. M. et al. Solving the Boltzmann equation to obtain electron transport coefficients and rate coefficients for fluid models. Plasma Sources Sci Technol 14, 722-733 (2005)). The reaction set includes electron-impact vibrational excitation (A1-A6), ionization (A7-A16), dissociation (A17-A25), dissociative recombination (A26-A29), charge transfer (A30), vibrational- translational (V-T, A31-A36), and radical recombination reactions (A37-A43). Nonelectron rates for V-T and radical recombination are calculated at Tgas= 473 K. Calculations were performed over Te= 1-6 eV to capture the transition from vibrational excitation regimes to dissociation and ionization. The resulting steady-state densities of electrons, radicals, ions, and vibrationally excited molecules serve as gas-phase inputs to the surface microkinetic model.

[0575] Microkinetic Model

[0576] Surface chemistry was described using a microkinetic model constructed on Ni(111) that explicitly couples Langmuir-Hinshelwood and Langmuir-Rideal pathways for CII4, CO2, and H2 conversion. The mechanism tracks methane activation and sequential dehydrogenation, C-C coupling and C₂ hydrocarbon formation, hydrogen recombination, CO2 adsorption and dissociation to CO* and ()*, carbonate and bicarbonate formation, oxygenate formation, and oxygen-hydrogen chemistry leading to OH*, H₂O*, and O₂. Plasma-generated radicals are incorporated as gas-phase pools and participate through adsorption steps and a reduced set of direct radical -adsorbate reactions identified as kinetically relevant under the conditions studied. Surface coverages θiand dimensionless gas pools ñievolve according to,

[0577] d6ty-’ J dn, \ >

[0578] (15)

[0579]

[0580] j where rjis the net rate of reaction j. Each elementary step is written in mass-action form,

[0581]

[0582] rj= kj,f∏sθsc∏sñsc− kj,r∏sθsc∏sñscForward rate constants are computed using transition-state theory,

[0583]

[0584] Reverse rate constants are obtained from detailed balance using equilibrium constants derived from reaction enthalpies and zero-point energy corrections. Vibrational coupling is incorporated by the vibrational state populations computed from the plasma model for CH₄, CO₂, and H₂. The surface model is initialized using steady-state plasmaAttorney Docket No. 10046-649W01

[0585] species densities, which are rescaled to match the total number of molecules in the reactor volume and converted to ñi= Ni / nsite. After the initialization of surface mkm, the plasma species equations continue to be solved self-consistently, with their original plasmaproduction terms retained. This formulation reflects conditions in which the plasma has equilibrated (~ 0.1 ps) before contacting the surface, but once in contact, gas-phase and surface reactions proceed concurrently. Surface coverages are initialized with vacant sites, and site conservation is enforced through θ* = 1 −

[0586]

[0587] 0f. The coupled stiff ODE system is solved using an implicit variable-step solver, and simulations are terminated when either CID or CO2 conversion reaches 3%. Physicochemical coupling effects such as electric fields, surface charging, and ion bombardment are neglected to isolate the chemical consequences of plasma-generated species interacting with the catalytic surface.

[0588] Statistical Analysis

[0589] Error Quantification in Analytical Characterization

[0590] Experiments were repeated at least three times to quantify the random variations (standard deviation) in the data (N > 3).

[0591] Error Quantification in XPS

[0592] Uncertainty quantification of XPS peak intensities employed Monte Carlo simulations, assuming Poisson-distributed noise for pulse-counted data. The standard deviation (o) for each data bin was calculated by σ =

[0593]

[0594] , where N represents counts per bin. Monte Carlo simulations were used to generate multiple synthetic data sets, each varying according to the expected statistical noise distribution (with N > = 200). From these simulations, standard deviations and 68.3% confidence intervals for peak intensities were derived.

[0595] Uncertainty in the percentage areas of XPS peak intensities was calculated by considering both variances and covariances of peak areas. For two peak intensities, A and B, uncertainty was calculated using,

[0596] >

[0597] σ{A+B}= √{σA2+ σB2+ 2Cov(A, B)}, (18)

[0598]

[0599] where oAand oBare standard deviations for peaks A and B, respectively, and Cov(A, B) is their covariance. The covariance matrix, which accounts for correlations between measurements, was defined as,

[0600] Cov(A, B) =

[0601]

[0602]

[0603] where and are individual measured intensities, A and B represent their respectiveAttorney Docket No. 10046-649W01

[0604] mean values, and n is the total number of observations.

[0605] Example 3:

[0606] Deactivation Caused by Oxygen Coverage

[0607] Initial flow reactor experiments were conducted to investigate how plasma excitations influence catalytic performance during N2O abatement. These experiments were designed to probe plasma-catalyst interactions and product yields under steady-state conditions and to elucidate mechanisms by which plasma activation modifies catalytic behavior within the reactor environment. A bare Al2O3sample as well as polycrystalline Cu supported on Al2O3sample were selected to isolate the role of the transition metal surface and to demonstrate the potential for plasma activation to enable the use of weakly binding catalysts (E 0,1-0 < 3 eV) under mild thermodynamic conditions (with binding energies ECU-N = 4.98 eV, cu-o = 2.97 eV). A dielectric barrier discharge (DBD) plasma, operated at 10 kV, with a frequency of 25 kHz, and average power of 10 W, was used with pure N2O (99.99%) as the feed gas. Gas flow rates were varied from 3-30 seem with a fixed pressure of p = 1 atm to modulate both residence time and specific energy input from the plasma phase. Conversion of N2O was measured after 30 min of testing using a GC-MS. Optical emission spectroscopy revealed N2C3Πu→ B3Πgrotational and vibrational temperatures of 330- 380 K and 2500-3000 K, respectively under these conditions, indicative of a state of vibrational nonequilibrium within the plasma.

[0608] Experiments found that the Cu / Al2O3catalyst did not show enhanced N?. O conversion compared to bare Al2O3across all tested SEI values (up to 30 eV / molecule). Both materials exhibited a similar linear increase in conversion with increasing SEI, corresponding to an energy cost of ~ 150 eV per N2O molecule converted (Fig. 50A).

[0609] Product distributions confirmed N2 and O2 were the predominant products in a 2:1 ratio, with trace NO and NO2 detected at levels below 0.05 and 0.1%, respectively. These findings suggest that Cu surface sites did not contribute significantly to N2O activation under plasma conditions, highlighting that plasma activation alone does not enhance catalytic performance universally across all catalyst types and reaction conditions. This trend aligns with previous reports that found that most supported transition metals showed minimal improvement over bare supports, with Ru as a notable exception. The lack of improvement in N2O conversion between the Cu catalyst and the Al2O3control sample suggests possible deactivation on the transition metal surface. To investigate catalyst deactivation, short-duration batch-mode experiments were conducted. Before exposure to an N2O plasma, catalysts were cleaned in an H2plasma (4 kV, 1.3 W, 30 min) to remove surface-boundAttorney Docket No. 10046-649W01

[0610] oxygen. The plasma was applied in five consecutive 30 s intervals, with N2O purged and replenished between each exposure. N2O conversion was measured after each interval over a total duration of 2.5 min using the same catalyst without intermediate surface cleaning. During these exposures, CU / Al2O3exhibited a rapid decline in activity, with conversion dropping by approximately 50% within 90 s, ultimately reaching levels comparable to the bare Al2O3support (Fig. 50B). In contrast, Al2O3alone maintained a stable (low) performance and level of conversion.

[0611] These results provide an explanation for similar levels of conversion that were measured in batch mode experiments (Fig. 50A), where C11 / Al2O3and Al2O3were compared after operating in a reactor for 30 min. Results also emphasize that while plasma excitation can enhance dissociative chemisorption by activating gas phase N2O, the sustained catalytic turnover is governed by the properties of the catalytic surface (Fig. 51A). The observed deactivation of Cu under plasma exposure indicates that once plasma-activated species reach the catalyst, subsequent reaction pathways must be considered.

[0612] A combination of XPS analysis and DFT calculations were performed to identify the mechanism responsible for the observed catalytic deactivation of Cu. XPS measurements were performed on a reference Cu / Al2O3sample and a sample exposed to N2O plasma for 30 min under steady-state conditions (Figs. IB and 1C). During plasma exposure, the flow' rate of 2O was held constant at 10 seem (10 kV, 10 W), and the specific energy input (SEI) was maintained at 15 eV / molecule. Analysis of the Cu 2p3 / 2 spectra revealed a nearcomplete loss of the Cu(0) / Cu(I) species following plasma exposure, with 99% of the signal corresponding to the Cu(II) oxidation state, compared to approximately 60% Cu(II) in the untreated reference sample. This increase in surface oxidation and Cu(II) content, indicative of high O* coverage, aligns with our earlier flow' reactor experiments showing no catalytic enhancement from Cu. Together, these findings confirm that extended plasma exposure leads to surface oxidation and deactivation, suppressing the catalytic role of Cu under reaction conditions.

[0613] DFT calculations were performed to elucidate the reaction mechanism responsible for the oxygen accumulation observed experimentally on Cu surfaces (Fig. 51 A). Using a model Cu(111) surface, the primary facet identified by XRD analysis, the N2O decomposition pathway was modeled through three elementary steps (Scheme 1):

[0614] Scheme 1: N2O plasma-assisted dissociation on catalyst surface and oxygen recombination to regenerate active sitesAttorney Docket No. 10046-649W01

[0615] N2Ovib+ * → N2O → N2(g) + O*

[0616]

[0617] 2 O* → O2* → O2(g) + 2*

[0618] (1) dissociative adsorption of N2O to form N2and adsorbed oxygen (O*), (2) surface recombination of two O* to form O2*, and (3) desorption of O2. The calculated activation energies (Ea) and free energy changes (AG) for each step were (0.47, -1.83 eV), (2.34, 2.33 eV), and (0.87, 0.87 eV), respectively. These results reveal that while N2O readily dissociates, the high energy barriers for both O* recombination and O2desorption render oxygen removal kinetically unfavorable at mild thermodynamic conditions within plasma reactors (TGas= 360 K, p = 1 atm). This promotes the accumulation of stable surface-bound O*, consistent with the XPS observations of Cu(II) enrichment after plasma exposure. Moreover, even when O2* is formed, its desorption is inefficient, and reoxidation of the surface is more favorable than regeneration, with a negligible reverse barrier of only 0.01 eV, indicating that the desorption of O* can be described as a single step with an even higher barrier of 3.2 eV. There is some uncertainty of Cu surfaces after long exposure times and their comparison to the calculated DFT energies and reaction processes, with significant coverage of O*, binding energies may be different, and reactions may be limited to E-R type or radical reactions, not considered in this work.

[0619] These findings clarify the mechanism of rapid catalytic deactivation during N2O conversion under non-thermal plasma conditions. While plasma excitation can enhance the kinetics of dissociation reactions, particularly the initial activation of N2O, catalytic turnover cannot be sustained without favorable surface thermodynamics that enable efficient removal of reaction intermediates. While plasmas can enable the use of weaker binding catalysts like Cu through excitation of reactant molecules at low surface temperatures, their performance is constrained by how efficiently intermediates can be removed. In the case of N2O, oxygen atoms accumulate on Cu surfaces due to high barriers for O* recombination and O2 desorption, effectively deactivating the catalyst. 'This underscores the need to codesign both plasma conditions and catalytic materials. Plasma properties must be tuned to deliver reactive species to the catalyst surface through mechanisms like vibrational excitation or selective energy coupling, while the catalyst must be capable of stabilizing, transforming, and ultimately releasing intermediates to regenerate active sites. Without this synergy, plasma may successfully activate reactants, but the catalytic surface may act as a sink rather than a site of sustained reaction.

[0620] Clearing Surface Species via Co-Reactant Addition of Hydrogen Carriers A series of co-reactant experiments were performed to investigate strategies forAttorney Docket No. 10046-649W01

[0621] removing O* and maintaining catalytic activity during plasma-assisted N2O conversion. Motivated by prior inactivation studies (Figs. 50A-51B), which showed that precleaned Cu / Al2O3achieved nearly twice the conversion of bare Al2O3at an SEI of 7 eV / molecule, these experiments aimed to sustain activity of a Cu transition metal surface. H2 and CH4 were separately combined with N2O to facilitate surface reactions that promote O* removal through the formation and desorption of II2O (Scheme 2).

[0622] Scheme 2: Alternate O* removal pathways accessed by the incorporation of hydrogen into the gas stream

[0623] H2vib+ 2* → 2H*

[0624] H* + O* → OH* + *

[0625] OH* + H* → H2O* + *

[0626] H2O* → H2O (g) + *

[0627] All experiments were conducted under the previous mild thermodynamic conditions (p = 1 atm, TGas≈ 360 K), where thermal activation alone is not sufficient to drive appreciable surface reactions. Under these conditions, the nonthermal plasma activates otherwise inert co-reactants, increases the probability of adsorption, and allows them to participate in reaction pathways on surfaces. This synergy allows a broader design strategy in heterogeneous catalysis where co-reactants can be added at low gas temperatures and excited selectively in plasmas to promote desirable pathways to remove undesirable intermediates that are bound to surfaces.

[0628] Hydrogen as Carrier

[0629] Mixtures of N2O and H2 (outside of flammability limits, <5% or >75% II2 by volume) were used to evaluate the effect of hydrogen addition to catalyst performance under constant plasma conditions (10 kV, 10 W, SEI = 7 eV / molecule). The addition of hydrogen to the Al2O3support alone had no positive effect on N2O conversion and, in some cases, slightly decreased performance (Fig. 52A). In contrast, the C11 / Al2O3catalyst exhibited a clear enhancement with increasing hydrogen concentration, displaying a near- linear increase in conversion up to 4.8% H2 by volume, 95% higher than the highest Al2O3only conversion. At this concentration, the system achieved a maximum of 90 eV per converted N2O molecule, nearly halving the energy cost relative to a pure N2O stream. At higher H ■ concentrations (>80%), conversion increased but energy efficiency plateaued and declined to ~130 eV / molecule (9 g / kWh) (Fig. 52B), indicating that while H2 enhances Cu- catalyzed performance, another step likely kinetic or availability of N2 becomes limiting.Attorney Docket No. 10046-649W01

[0630] However, while a saturation in performance gain was measured at high concentrations of H2, these conditions are generally unattractive for practical applications due to the cost and storage complexity of II 2. These results demonstrate that the addition of H at low concentrations can improve the plasma catalytic abatement efficiency with weakly binding catalysts like Cu / AhOs at near ambient temperatures.

[0631] Methane as Carrier

[0632] Methane was investigated as an alternative hydrogen source for removing O* on catalysts in plasma environments, motivated by its high hydrogen-to carbon (H / C) ratio, low cost, and potential to generate value-added hydrocarbons under nonthermal conditions. In contrast to thermal catalysis where methane in SCR typically forms only CO2 plasma systems operating at 90 °C and 1 atm have been shown to favor the formation of C2products, reducing both coking and CO2 emissions. Methane was introduced into the gas stream at molar fractions of 50%, 65%, and 80% onto Cu / AhOs, following a protocol similar to the hydrogen experiments. The trends in N2O conversion mirrored those observed with hydrogen, with energy efficiency increasing significantly as methane concentration rose. Notably, the highest energy efficiency observed with methane exceeded that of hydrogen, reaching values over 100 eV per converted N2O molecule (16 g / kWh) (Fig. 53A).

[0633] In addition to evaluating N2O conversion, the carbon balance of the system was analyzed to quantify the products of methane decomposition. Over 90% of the converted methane was accounted for, with ethane and carbon dioxide being the dominant carbon- containing products, representing approximately 41% and 28% of the total carbon output, respectively (Fig. 53B). Minor products included ethylene and acetylene. The formation of C2 hydrocarbons such as ethane and ethylene highlights the added value of methane as a coreactant in plasma catalytic systems with weakly binding catalysts, conditions where subsequent decomposition of CH3* on surfaces is limited. Similar to the behavior observed with hydrogen, energy efficiency gains with increasing methane concentration eventually plateaued, suggesting that the removal of 0* is no longer the rate-limiting step at high coreactant levels.

[0634] These findings underscore the role that carrier gas selection plays in determining the performance and efficiency of N2O decomposition systems. Among the co-reactants tested, methane proved slightly more effective than hydrogen when used with Cu / AhC in an NTP, achieving lower energy costs for N2O conversion, around 16 g / kWh while the use of H2 as a co-reactant reached a maximum of 15.5 g / kWh and at higher concentrations, maintained energy costs at 10 g / kWh. This enhanced performance can be attributed to complementaryAttorney Docket No. 10046-649W01

[0635] O* removal pathways, namely, the formation of both CO2and H2O as well as the concurrent conversion of CH4 into value-added C2 hydrocarbons. In contrast, previous studies have often employed oxidizing carrier gases such as O2, N2, or N2 / O2 mixtures, which may inhibit net conversion by replenishing surface oxygen or promoting reverse reactions. The strategic use of reductive co-reactants like hydrogen and methane in plasma environments not only facilitates the removal of inhibitory oxygen species but also improves energy efficiency. Additionally, in the case of methane, the potential to produce economically valuable hydrocarbons further strengthens its appeal. Even inert gases such as argon, frequently used to dilute N2O streams, may influence surface conditions through physical mechanisms such as sputtering or energy transfer, highlighting the need for further study into the broader effects of carrier gas composition.

[0636] Mechanism for H2 / Co-Reactant Addition

[0637] Although NTPs excite N2O and drive its dissociation under mild conditions, the overall reaction rate is limited by surface-mediated steps, most critically, the removal of O*. Our time-on-stream measurements confirmed that oxygen accumulates on catalyst surfaces during operation, leading to rapid deactivation, particularly on Cu / AhOs. To address this limitation, we combined experimental and computational approaches, including time-on¬ stream conversion measurements, surface characterization, and DFT calculations, to investigate how co-reactants such as H? and CH4 promote alternative pathways for O* removal. L-H type reaction mechanisms were considered as the main pathway with radical-assisted reactions and (E-R) pathways excluded from consideration in this work. This was based on reported lifetime comparisons. However, further investigation is required for the relative contribution of these vibrational and radical-assisted pathways to plasma-assisted catalysis. Time on stream experiments were performed to probe the generation of water from O* deactivated Cu surfaces after extended exposure to N2O plasmas (Figs. 54A-54C). Following over 1 h of N2O plasma treatment, the reactor was purged with H2, after which a low-power H2plasma (4 kV, 1.33 W) was ignited. Water formation was monitored over a 2 h period to determine the extent of O* removal from the catalyst surface. Since no external oxygen source was introduced during this stage, any detected H2O was attributed to residual oxygen on the support, O* on Cu, or oxygen stored within the Al2O3matrix, as indicated by prior XPS and DFT analyses. Distinct differences in H2O production were observed between the two catalysts. C11 / Al2O3exhibited a pronounced transient increase in water signal, followed by a gradual decline, whereas Al2O3showed only a minimal increase above baseline (Fig. 54A). Integration of the H2O peak area for Cu / AhCh revealed a total of ~0.95Attorney Docket No. 10046-649W01

[0638] μmol H2O produced, and 0.15 μmol assigned to the alumina support (Fig. 54B). The resulting water corresponding solely to the Cu sites was approximately one third of the total estimated active sites on the catalyst (2.5− 2.8 μmol). This result provides strong support for the hypothesis that surface oxygen accumulation deactivates the Cu sites, and that hydrogen plasma treatment can remove adsorbed oxygen and produce H2O.

[0639] A combination of surface characterization experiments and DFT computations were performed to support the proposed regeneration mechanism. Samples were collected before and after N2O plasma exposure, both in the presence and absence of the H2co-reactant (Figs. 55A-55B). XPS analysis revealed that an unexposed reference and post H2plasma samples (following 30 min plasma exposure) showed very similar Cu oxidation state distributions, with 35-45% of surface sites as Cu(0 / I) and 55-65% of sites displayed Cu(ll) states. H2 plasma-treated samples restored a small fraction of Cu(0 / I) states that were lost under pure N2O plasma 99% Cu(II), 1 % Cu(0 / I). The appearance of Cu(OH)2 features in XPS indicated hydroxyl formation evidence of an active water-forming pathway during the regeneration process (Figs. 55B-55C). Additionally, N Is spectra showed nitrogen¬ containing species on the surface, specifically nitride formation, suggesting the presence of alternative reaction pathways such as N-N bond scission, which are typically inaccessible under purely thermal conditions. These species (e.g., N*, NO*) likely arise from radical production in the plasma, which can facilitate bond cleavage mechanisms distinct from the conventional O-N bond dissociation. Moreover, plasma studies have observed NO as a product, though its yield is often suppressed when catalytic or even inert surfaces are present in the discharge zone, shifting selectivity toward N2 and O2. In our experiments, both gas-phase NO signals (GC-MS) (<0.05%) and NO related XPS peaks were low (0.5% of O Is signal), indicating that while N-N scission is plasma accessible, O-N cleavage remains the dominant decomposition pathway under the studied conditions.

[0640] Additional evidence for O* removal was obtained by comparing gaseous H2O and O2 signals from flow reactor experiments across varying II2 concentrations. When the II2 content exceeded the stoichiometric ratio required for complete conversion of N2O to N2 and H2O, O2 formation ceased entirely, and H2O became the dominant product (Fig. 55D). A similar trend was observed when CH4was used as the hydrogen carrier, as confirmed by GC-MS analysis, indicating that both H2and CH4can facilitate surface oxygen removal via hydrogenation pathways.

[0641] DFT calculations were carried out to evaluate the role of adsorbed hydrogen (H*) in promoting ()* removal and gain mechanistic insight into these observations. The computedAttorney Docket No. 10046-649W01

[0642] activation energy for H2adsorption on Cu(111) was 0.34 eV, supporting experimental findings from time on stream tests that H2 must be excited (e.g., by plasma) to remove surface bound oxygen effectively under mild thermodynamic conditions (Fig. 55A). Once both O* and H* species are present on the surface, DFT results show that water formation becomes the most favorable pathway for regenerating active sites. While H* recombination to form H2has the lowest activation barrier (Ea = 0.87 eV), it is thermodynamically uphill (AG = 0.53 eV), making it a less favorable route. In contrast, stepwise formation of OH* and then H2O* exhibits higher activation energies (0.72 and 1.05 eV, respectively) but leads to more thermodynamically stable intermediates and products. Notably, the energy barrier for the rate-limiting step is reduced by more than half compared to the O* recombination pathway in the absence of hydrogen (from 2.34 to 1.05 eV), demonstrating how co-reactant addition alters surface reaction energetics fundamentally. These results explain the observed increase in N2O conversion with the addition of hydrogen carriers. Plasma activation enables the formation of reactive H* species, which in turn promote O* removal through water forming pathways.

[0643] Analysis was also performed with CII4 as the co-reactant, showing similar results to that of H2. XPS samples of Cu post N2O / CH4 plasmas (1:1, 30 min) showed a smaller fraction of Cu (II) states compared to the pure N2O plasma stream and higher than the reference sample, showing the reducing effect of CH4 as a co-reactant and its ability to generate new active sites. The fraction of Cu(OH)2 detected in these samples was significantly lower (~3%) than in those treated with N2O / H2 plasma (~20%), further indicating that the Cu surface was more strongly shifted toward the reduced Cu(0 / I) state (Figs. 56B-56C). The smaller contribution of the hydroxide state can be attributed to other pathways for O* removal, like CO2 generation. Carbon Is XPS samples were also investigated, fitted with three different possible surface carbon states: C-C (adventitious carbon and hydrocarbon fragments), C-O-C (intermediates for CO2 formation), and O-C_O (final state before desorption) (Figs. 56D-56E). The reference Cu / AhCh sample showed predominantly C-C species (-55%), typical of ambient contamination or adventitious carbon. After N2O plasma exposure, a marked increase in oxygenated carbon species was observed, with C-O-C and 0-C..0 accounting for 43% and 33%’ of the total signal, respectively. This is expected with the lack of new carbon sources and the continuous presence of oxygen in the system from decomposing N2O, driving toward possible CO2 formation. Upon addition of CII4 to the plasma, the C is spectrum shifted again, this time toward a higher proportion of C-C species, approaching the levels of theAttorney Docket No. 10046-649W01

[0644] oxygenated species. This shift correlates well with gas phase product analysis, which showed the formation of CO?, as well as light hydrocarbons such as C He and other C? fragments, and whose intermediates are shown on the XPS spectra.

[0645] Analysis was also performed with CH4 as the co-reactant, showing similar results to that of IF. XPS samples of Cu post N2O / CH4 plasmas showed a smaller fraction of Cu (II) states compared to the pure N? O plasma stream and higher than the reference sample, showing the reducing effect of CH4 as a coreactant and its ability to generate new active sites, perhaps even better than H? (Figs. 56B-56C). The presence of Cu(OH)? sites, 3% of total signal was much smaller than that of the N2O / H2 samples (20%), much more fa voring the Cu (0 / 1) state (Figs. 56B-56C). The smaller contribution of the hydroxide state can be attributed to other pathways for O* removal, like CO? generation. Carbon Is XPS spectra were also examined, fitted with three different possible surface carbon states, C-C, carbon, featuring adventitious carbons and C2 states, C-O--C, intermediates for CO2 formation, and O-C=(), final state CO2 before desorption (Figs. 56D-56E). The fitted states for the reference sample (Cu / AhCh pre plasma) showed a majority of states associated with adventitious carbon on the catalyst surface (55%) with smaller amounts of the oxygenated states. The post N2O sample showed a shift in the XPS spectra toward the C-O-C and O-C-0 states, at 43% and 33%, respectively. This was expected with continuous delivery of surface bound oxygen in the system from decomposing N2O and no C* species actively delivered to the surface, thus driving toward possible CO2 formation. Wien CH4 was added to the gas stream, the C Is spectra displayed a small change, with an increase in the C-C portion of the peak, to similar levels of the carbon oxygenate species. This result correlated to the experimental results in the steady state reactions where both CO2, C dL. and other C2 hydrocarbons are produced, and whose intermediates are shown on the XPS spectra. Results from DFT computations align well with observations from surface characterization and steady state experiments, revealing multiple reaction pathways and product distributions. For the N2O / CH4 system, two pathways overlap with those found in the N2O / H2 reaction, producing water and hydrogen, but additional pathways involving carbon containing products _such as C? hydrocarbons and CO2 are also identified (Fig. 56A). The calculations show that the most thermodynamically stable state for adsorbed CH3* favors ethane formation on Cu(111), with a AG of —1.29 eV, however, the activation energy (Ea = 1.59 eV) for this pathway is higher than that for CH?* oxidation to CH 2* and H* (Ea = 1.37 eV). This lower Ea and higher AG value for C2H4compared to C2H6, while having very different experimental selectivities indicates that these are not the sole determining factors forAttorney Docket No. 10046-649W01

[0646] product distribution. Considering the individual steps within the reaction, the formation of CH₂* from CH₃* is unfavorable to C₂H₆ due to the final reaction energy. Thus, this difference in selectivity was attributed with the concentration and stability of being CH₃* higher than that of CH₂* as CH₃* is the initial adsorbed carbon state in a more stable state than CH₂*.

[0647] As the adsorbed CI Lr intermediates become progressively more oxidized along the reaction coordinate, their relative energies increase, with the carbonaceous (coked) C* state representing the highest energy intermediate. This supports previous predictions that under mild (ambient) conditions with plasma catalysis, the system favors C₂ hydrocarbon formation. However, at elevated temperatures with CH4 present, coking becomes a more significant deactivation pathway, with CO? formation becoming the primary mechanism for coke removal, shifting the balance away from simultaneous water and CO? formation toward CO? dominated surface cleaning.

[0648] The removal of inactivating surface species during N₂O abatement can be mitigated by introducing co-reactants into the gas stream. These co-reactants, also excited in the plasma, scrub strongly bound O* off the catalyst surface by allowing new mechanistic pathways with more favorable thermodynamics and kinetics to remove these O* atoms. By¬ employing weakly binding catalyst surfaces, like Cu, alongside plasma excitation of both reactants and co-reactants, it is possible to overcome the limitations of strong O* adsorption. Plasmas can be used to promote dissociation adsorption reactions, catalysts can be chosen to promote desirable surface pathways, and coreactants can be integrated and excited to mitigate inactivation. Such synergy opens the door to using inexpensive, weakly-binding catalysts with cost-effective co-reactants for efficient, long-term N₂O abatement. These conditions also enable the utilization and upgrading of broader types of co-reactants, such as CH4, that are not feasible in thermal conditions.

[0649] Conclusions

[0650] This work identifies surface oxygen accumulation as a critical bottleneck in plasma assisted N₂O decomposition and demonstrates that targeted co-reactant strategies can effectively mitigate this form of catalyst deactivation. Despite the enhanced dissociation kinetics enabled by non thermal plasma, the overall reaction mechanism remains constrained by surface chemistry, specifically, the recombination and desorption of adsorbed oxygen (O*), which limits long-term catalytic activity.

[0651] By introducing hydrogen or methane as co-reactants, we unlock alternative, lower energy pathways for O* removal via water or carbon-containing products. These co-Attorney Docket No. 10046-649W01

[0652] reactants, activated in the plasma alongside N2O, regenerate active sites and enhance conversion, while also shifting product selectivity toward benign or value-added species. Hydrogen addition was shown to improve energy efficiency, reducing the energy cost by nearly half at optimal loadings, though conversion eventually plateaued at high H2 concentrations. Methane performed even better, enabling comparable N2O removal while coproducing valuable C2 hydrocarbons such as ethane and ethylene under mild conditions, offering both environmental and economic advantages. Mechanistic insights from optical diagnostics, XPS, and DFT reveal that these coreactants facilitate O* removal through the formation of transient surface-bound intermediates that evolve into water or CO2 and hydrocarbons.

[0653] More broadly, this study illustrates that plasma catalysis offers a multidimensional design space for chemical processes, where plasma properties, catalyst surface chemistry, and gas phase composition can be co-optimized. Plasma excitation accelerates dissociative steps and enables the use of weakly binding, earth-abundant catalysts like Cu that are otherwise inactive under mild thermal conditions. Surface temperatures and catalytic properties govern the fate of adsorbed intermediates, while co-reactants provide a mechanism to mitigate pathways that lead to inactivation through the accumulation of surface intermediates. These findings demonstrate how plasma catalysis offers a framework to augment heterogeneous catalysis, offering new opportunities for efficient, low- temperature catalytic transformations in environmental and synthetic applications.

[0654] Materials and methods

[0655] Plasma Catalytic Reactor Setup

[0656] The abatement of N2O was investigated in a plasma catalytic reactor (Fig. 49). Gas mixtures, including combinations of N2O (99.99%, Matheson), H₂, and CH₄, were introduced into the reactor using mass flow controllers (Brooks GF40) and a mixing manifold. Total flow rates ranging from 3-40 standard cubic centimeters per minute (seem) were used during experiments, allowing control over the residence time (2.3-31.3 s) (i.e., the ratio of plasma volume to volumetric flow rate) in the region where plasmas and catalysts were placed. The reactor pressure was maintained at atmospheric pressure by connecting the outflow of the cell to a snorkel with a fixed exhaust pressure of 0.99 atm. The outflowing tubing was heated to 320 K to facilitate desorption of condensable species from tubing walls and to prevent product loss. Plasma was generated inside the reactor test section using a dielectric barrier discharge (DBD). The DBD featured a discharge gap of 0.1 in. (2.5 mm), enclosed by two quartz dielectric plates of 1 mm thickness, creating a reactionAttorney Docket No. 10046-649W01

[0657] volume of 0.094 cu in (1.55 cm3). Aluminum electrodes were positioned behind each dielectric to prevent direct contact with the reactive gas stream. A 25 kHz resonant power supply was used to drive the discharge, applying voltages in the range of 5—10 kV and producing peak currents between 15-25 mA.

[0658] The reactor was designed with a modular configuration to support two operational modes: (1) flow mode, in which reactants were continuously purged through the system at a controlled flow rate, and (2) batch mode, in which a fixed volume of reactant gas was introduced and sampled after a defined exposure time. The flow mode was used predominantly to study performance in a process environment and to provide information over extended durations ranging from 0.5-2 h. The batch mode was used to study short- time scale phenomena, such as catalyst deactivation, by sampling directly from the static gas volume within the plasma region after exposure periods of less than 5 min. The reactor was operated by placing packed beds of catalysts in the DBD plasma volume between dielectric plates. Catalysts (350 mg) were placed between the dielectric plates (Fig. 49), suspended by glass wool. Before plasma ignition, gases were purged through the reactor for at least 15 min to ensure a fixed reactant composition within the reactor, confirmed via gas chromatography. Additional diagnostics were used to characterize the plasma environment near the catalysts, including current and voltage probes as well as Lissajous power measurements to quantify the electrical energy input and optical emission spectroscopy (through optically accessible windows on top of the reactor) to determine gas phase properties such as rotational and vibrational temperatures. This ensured consistent plasma characteristics and minimized altering properties within the plasma phase. Lower voltages and powers (4 kV, 1.33 W) were applied separately for catalyst precleaning.

[0659] Experiments were performed to quantify the conversion of N2O (referred to as destructive removal efficiency (DRE)) for a given energy input to the plasma. The energy cost of the process was quantified through the specific energy input (SEI) to the reactant molecules where SEI can be presented as eV / molecule and then can be converted into an energy cost per converted molecule by dividing by the conversion percentage.

[0660] SEI = power / flow rate (1) Residence time as a function of flow rate was also monitored, calculated as volume / flow rate, representing the average time a molecule spends inside the plasma volume. The SEI was normalized by the total number of reactant molecules in the process, including both N2O and co-reactants. This normalization is important, as definitions of SEI vary across theAttorney Docket No. 10046-649W01

[0661] literature with some references reporting SEI relative to only the primary reactant (e.g., N₂O), with recent reviews discussing a need for consistent metrics for plasma chemistry.

[0662] Catalyst Materials and Packing

[0663] Plasma-assisted catalytic abatement experiments were conducted using polycrystalline Cu deposited on Al2O3powder (20 wt %) and Al2O3powder catalysts, synthesized by incipient wetness impregnation. The C11 / Al2O3catalyst was used to evaluate the role of Cu in enhancing N2O conversion, while Al2O3served as a baseline to distinguish contributions from the catalyst support and the plasma phase. Each experiment used 350 mg of catalyst and 40 mg of glass wool with mean particle sizes of 7.30 nm in diameter. Active site density of the catalysts was quantified using temperature -programmed pulsed chemisorption coupled with mass spectrometry. Copper site density was measured to be 85 μmol Cu / g catalyst. Experiments were performed by dry loading catalysts and supports into a wool matrix with a packing porosity of 90% to maximize exposure to the plasma discharge and minimize pressure losses within the packed bed. Selected catalyst samples underwent surface cleaning using a hydrogen plasma treatment (4 kV, 1.33 W, 30 min) before exposure to N2O, while others were used without pretreatment.

[0664] Gas Chromatography

[0665] Reactants and product concentrations were measured with an Agilent 8890 gas chromatography-mass spectrometer (GCMS) system. A thermal conductivity detector (TCD) was used in tandem with the mass spectrometer (Electron Ionization, El) to measure the concentration of H₂. The instrument was equipped with gas sampling valves for automated injection of gas samples (250 μL). Manual injections were also performed for volumes ranging from 0.2−250 μL using a gastight syringe (Hamilton Gaslight 1700 Series) that was fitted with a Chaney adapter. A Haysep column was used with the TCD, and a Porabond Q column was paired with the MS. The GC oven temperature was set initially to 45 °C and ramped at 10 °C / min to a final temperature of 105 °C. Helium was used as the carrier gas in a 500:1 split ratio, with a 300 mL / min flow rate. Calibration curves were generated by varying gas composition through the sampling valves and by adjusting manual injection volumes. Compounds were identified by comparing chromatographic retention times and mass fragmentation patterns to those of calibration standards, which were analyzed alongside both reaction and reference samples.

[0666] Optical Emission Spectroscopy

[0667] Optical emission spectroscopy was used to measure the rotational and vibrational temperatures of N2 within the plasma. The rotational temperature (TROI) was used as anAttorney Docket No. 10046-649W01

[0668] approximation of the bulk gas temperature (T_Gas ≈ T_Rot), while the vibrational temperature provided insight into the state of vibrational non equilibrium within the plasma.

[0669] Measurements of optical emission from the plasma were recorded using a Horiba iHR 550 imaging spectrometer with an EMCCD camera. The spectrometer was configured with an 1800 grooves / mm diffraction grating (blaze wavelength: 500 nm, angle: 26°45').

[0670] Measurements were taken with a slit width of 100 μm, 90 s exposure time, and 40 accumulations to achieve a sufficient signal-to-noise ratio. Analysis focused on the N₂ second positive system (C³Πu→ B³Πg) in the 360-400 nm range due to its well- characterized structure and strong emission intensity. We used a state -by- state fitting of each individual state in the transition was fitted in a linear combination to match the measured spectra via a least sq fit. The state-by-state fit outputs a relative population of states for each rovibrational level, which were then fit to estimate the rotational and vibrational temperatures. The rotational and vibrational temperature of N2 were measured to be 360 ± 20 K and 2650 ± 250 K respectively for our DBD plasma that was operated at atmospheric pressure with a peak voltage of 10 kV, average power of 10 W, and frequency of 2.5 kHz. Rotational temperature was used to estimate the translational and thus gas temperature for the nonequilibrium plasma. This assumption is due to the rotational energy levels being extremely small compared to mean thermal energy so that T_Rot ≈ T_Gas.

[0671] X-ray Photoelectron Spectroscopy

[0672] X-ray photoelectron spectroscopy was used to analyze surface chemical composition and oxidation states of species on Cu / Al₂O₃ samples following plasma exposure. The measurements used a VersaProbe 4 system that was equipped with a monochromatic Al Ka X-ray source and a 180° hemispherical energy analyzer. The instrument was configured for high sensitivity, spatially resolved surface analysis. Both survey and high-resolution spectra were acquired for polycrystalline Cu / Al2O3samples after exposure to (1) N2O, (2) N2O / H2, and (3) N₂O / CH₄ for 0.5 h. Survey spectra were taken with a pass energy of 140 eV with a single pass and a step of 0.5 eV. High-resolution scans were taken at a pass energy of 27 eV and energy step of 0.1 eV for 8 scans. Spectral fitting and peak deconvolution were conducted to infer the presence of Cu (II), Cu (I), and Cu(OH)₂, and other additional states, using CasaXPS (Figs. 51A-51D, 55A-56E), with binding energies referenced to the Al 2p peak to account for charging effects. All samples were handled in an oxygen and moisture free environment to minimize surface contamination and oxidation. Following plasma treatment, the reaction chamber was purged with an Ar / H₂ gas mixture and samples were transferred into an inert atmosphere glovebox. Samples were then sealed in an airtight XPSAttorney Docket No. 10046-649W01

[0673] transfer holder and introduced into the analysis chamber under ultrahigh vacuum conditions.

[0674] Plasma Mediated Surface Mechanism

[0675] Plasma assisted catalysis involves the generation of reactive or excited species in the plasma phase that interact with heterogeneous catalysts. These interactions can alter surface reaction pathways and modify the physicochemical state of catalytic surfaces, including effects such as localized charging, transient heating, and structural changes. In DBD plasmas, vibrationally excited molecules, electronically excited molecules, radicals, free electrons, and ions are produced under thermal nonequilibrium conditions. These species can participate in catalytic reactions through both L–H and E–R mechanisms, depending on their lifetimes, surface accessibility, and reactivity. For instance, vibrationally excited reactant molecules can be generated within a plasma and transported to an available catalytic active site to increase the rate kinetics of dissociative adsorption reactions selectively (L-H). Alternatively, radicals such as atomic oxygen can adsorb onto catalytic surfaces and react with co-adsorbed species (L-H) or react upon collision with surface¬ bound intermediates (E-R). In contrast, ions and electrons, despite their high energies, often have a limited direct impact on surface reactions due to short residence times, low surface fluxes caused by plasma sheath effects, and rapid recombination. Similarly, electronically excited states of N2O tend to relax rapidly through collisions and radiation, making them less likely to persist long enough to interact meaningfully with the catalyst surface. The mechanism by which plasma-activated reactants couple with catalysts depends on their rate of generation and relaxation, only those species that can be transported to catalysts before they relax contribute to surface reaction mechanisms. Nonetheless, further studies are required to fully decouple and quantify the effects of these various excitation modes. Within a plasma, N2O can be vibrationally excited across three distinct modes: the symmetric N–O stretch (1285 cm⁻¹), the degenerate bending mode (590 cm⁻¹), and the asymmetric N–N stretch (2224 cm⁻¹). The population and lifetime of these excited states depend on the rates of vibrational–vibrational (V–V, τ = 1 / k_V-V[N₂O] = 1.84 μs) and vibrational–translational (V–T τ ≈ 1 / k_V-T[N₂O] = 1.6 μs) relaxation processes (k ≈ 4 × 10⁻¹⁴ cm³s⁻¹), which in turn determine their likelihood of reaching the catalyst surface before deactivation. In addition, N2O plasmas generate reactive radicals, particularly O and N», whose concentrations and surface reactivity are governed by recombination rates (k ≈ 1 × 10⁻¹⁰ cm³ s⁻¹), wall losses, and co-reactant concentrations. Introducing species such as methane further enriches the plasma chemistry, producing additional vibrationally excited molecules (e.g., CH '1’’, H2vlb), methyl radicals (CH3 •), and atomic hydrogen (H»). These radicals are extremely reactiveAttorney Docket No. 10046-649W01

[0676] and may have higher adsorption probabilities compared to vibrationally excited species, but with higher reaction rates for gas-phase reformation. For atomic H radicals, the chance of interaction with surfaces is higher, due to gas-phase reaction rates lOOOx smaller than those of the other radical species possibly generated, but still greater than the molecular vibrational H species. These lifetimes are ultimately dependent on radical concentration and, with limited concentration compared to vibrationally excited molecules, may contribute to surface reactions. These species may initiate new reaction pathways either in the gas phase or on the surface via L-H or E-R mechanisms. The net contribution of each species to surface reactivity ultimately depends on both its chemical reactivity and its transport and relaxation dynamics. This example focuses on identifying how plasma excitation alters the design space of heterogeneous catalysts for N2O abatement, using a Langmuir-Hinshelwood framework to describe gas-surface interactions and interpret experimental trends. In this approach, reaction pathways are assumed to be modified by the selective excitation of reactant molecules in the plasma, which enhances the kinetics of dissociative adsorption by increasing the likelihood that these excited species reach and interact with the catalyst surface. The subsequent surface reaction rates are governed by catalyst properties and the surface temperature of a model surface, both of which are examined using design tools informed by DFT. Radical-assisted reactions and (E-R) pathways are excluded from consideration in this work. This decision is based on reported lifetime comparisons, which show that vibrationally excited species exhibit lifetimes of ~1.8 μs at ambient conditions, making them likely to contribute to catalytic reactions before relaxation. Nonetheless, the relative contributions of vibrational and radical pathways remain an open question and a compelling direction for future investigation.

[0677] Density Functional Theory Computations

[0678] Density functional theory (DFT) calculations were carried out to infer the energetics and mechanisms of surface reactions on Cu(111) surfaces. All computations were performed using the Vienna Ab initio Simulation Package (VASP) with projector-augmented wave (PAW) pseudopotentials and the Perdew-Burke-Ernzerhof (PBE) generalized gradient approximation (GGA) for the exchange-correlation functional.⁴³'⁴⁴ A plane-wave basis set is used to describe the valence electrons with an energy cutoff of 400 eV, and a 2 x 2 x 1 Monkhorst- Pack k-point grid was used for the integration of the reciprocal space. Based on the diamagnetic properties of Cu and the lightatom adsorbates (O, N, C, H) on Cu(111) yields closed-shell (paired -electron) species with no net magnetic moment, so spin polarization was disabled.45 Cu bulk are both in the Fm3̄m space groupAttorney Docket No. 10046-649W01

[0679] and the slab structures are made by cutting from a fee (111) lattice plane. The slab models are at least 8 A in thickness with 4 layers of metal atoms separated by a vacuum layer of 12 A. The bottom two metal layers were fixed during relaxation to approximate bulk-like behavior, while the top layers and adsorbates were relaxed fully. Gibbs energies for each reaction state were calculated to obtain activation energies and overall reaction energies on Cu(111) slabs, corrected for zero-point energy (ZPE) values of adsorbed states. Transition states were identified using the Nudged Elastic Band (NEB) method, as implemented in the VTST tools package within VASP. For each reaction pathway, five intermediate images were interpolated between the initial and final states; for shorter pathways, three intermediate images were sufficient. These images were optimized simultaneously to trace the minimum- energy path (MEP), with a force convergence criterion of 0.02 eV / A. The climbing image NEB (CI-NEB) scheme was used to refine the transition state by elevating the highest-energy image to the saddle point, thereby ensuring accurate determination of the transition-state geometry. These energies were used to construct reaction coordinate diagrams, providing activation barriers and overall reaction energies that support experimental observations.

[0680] Other advantages which are obvious and which are inherent to the invention will be evident to one skilled in the art. It will be understood that certain features and sub¬ combinations are of utility and may be employed without reference to other features and sub-combinations. This is contemplated by and is within the scope of the claims. Since many possible embodiments may be made of the invention without departing from the scope thereof, it is to be understood that all matter herein set forth or shown in the accompanying drawings is to be interpreted as illustrative and not in a limiting sense.

Claims

Attorney Docket No. 10046-649W01CLAIMSWhat is claimed is:

1. A method of selectively producing a target carbon-containing product during the production of hydrogen gas from a methane-containing gas, the method comprising: a. forming a non-thermal plasma in the presence of the methane-containing gas and a metal catalyst,i. wherein the methane-containing gas has a gas temperature,ii. wherein the metal catalyst has a surface temperature, andiii. wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature, and that dissociatively adsorbs on the metal catalyst; andb. adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, and / or selecting a different metal catalyst,thereby producing hydrogen gas and the target carbon-containing product, wherein the target carbon-containing product is produced in a greater amount than non-target carbon- containing products.

2. The method of claim 1, wherein the target carbon-containing product is C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, or fullerenes.

3. The method of claim 1, wherein the target carbon-containing product is:a. C2H6 and the non-target carbon-containing products are C₂H₄, C₂H₂, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;b. C2H4 and the non-target carbon-containing products are C2H6, C2H2, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;c. C2II2 and the non-target carbon-containing products are C2II6, C2H4,, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;d. carbon black and the non-target carbon-containing products are C2H6, (' ll;.C₂H₂, graphite, graphene, carbon nanotubes, and fullerenes;e. graphite and the non-target carbon-containing products are C₂H₆, C₂H₄, C₂H₂, carbon black, graphene, carbon nanotubes, and fullerenes;f. graphene and the non-target carbon-containing products are C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, carbon nanotubes, and fullerenes;Attorney Docket No. 10046-649W01 g. carbon nanotubes and the non-target carbon-containing products are C₂H₆, C₂H₄, C₂H₂, carbon black, graphite, graphene, and fullerenes; orh. fullerenes and the non-target carbon -containing products are C2H6, C2H4, C2H2, carbon black, graphite, graphene, and carbon nanotubes.

4. The method of claim 1, wherein the non-thermal plasma is formed by a dielectric barrier discharge, diffused plasma discharge, corona discharge, RF coupled plasma, microwave reactor, or plasma jets.

5. The method of claim 1, wherein the non-thermal plasma is formed with gas temperatures ranging from 300 K to 800K and vibrational temperatures greater than 1500 K.

6. 'The method of claim 1, wherein the non-thermal plasma is formed with surface temperatures ranging from 373K to 1000K and vibrational temperatures ranging from 800 K to 6000K.

7. The method of claim 1, wherein the non-thermal plasma is formed at 1 atm.

8. The method of claim 1, wherein the methane-containing gas is liquified natural gas.

9. The method of claim 1, wherein the methane-containing gas is from a distributed methane source.

10. The method of claim 9, wherein the distributed methane source is from wetland emissions, termite activity, ocean release, natural gas reserves, dumps, pump jacks, or tail gas streams from an industrial process.

11. The method of claim 1, wherein the metal catalyst is a transition metal catalyst.

12. The method of claim 1, wherein the metal catalyst is a metal oxide.

13. The method of claim 1, wherein the metal catalyst comprises a mixture of more than one different metal.

14. The method of claim 1, wherein the metal catalyst is a bimetallic catalyst.

15. The method of claim 1, wherein the metal catalyst is on a support.

16. The method of claim 15, wherein the support is a dielectric support selected from the group consisting of Al2O3. SiCh, TiCh, zeolites, activated carbon, glass, and ceramic.Attorney Docket No. 10046-649W01 17. The method of claim 1, wherein the metal catalyst has a binding energy of carbon to a metal atom in the metal catalyst of greater than -0.5 eV.

18. The method of claim 1, wherein the metal catalyst is Cu, Ag, Au, or any combination thereof.

19. The method of claim 1, wherein the metal catalyst is Re, Pt, Rh, Ru, Pd, Ni, Ni, Co, Zn, or any combination thereof.

20. The method of claim 1, wherein the surface temperature of the metal catalyst is adjusted to from 300K to 1000 K.

21. The method of claim 1, wherein the gas temperature is adjusted to from 300K to 800K.

22. The method of claim 1, wherein the method produces less than 1% of CO2.

23. The method of claim 1, wherein the metal catalyst is Cu, the surface temperature is greater than 973 K, the gas temperature is from 300K to 800K, and the target carbon-containing product is CH4.

24. The method of claim 1, wherein the metal catalyst is Cu, the surface temperature is from 573 to 973 K, the gas temperature is from 300K to 800K, and the target carbon- containing product is C H 4.

25. The method of claim 1, wherein the metal catalyst is Cu, the surface temperature is less than 573 K, the gas temperature is from 300K to 800K, and the target carbon-containing product is C2H6.

26. The method of claim 1, wherein the metal catalyst is Ni, the surface temperature is greater than 640 K, the gas temperature is from 300K to 800K, and the target carbon-containing product is CH4.

27. The method of claim 1, wherein the metal catalyst is Ni, the surface temperature is from 473 to 640 K, the gas temperature is from 300K to 800K, and the target carbon-containing product is C2H4.

28. The method of claim 1, wherein the metal catalyst is Ni, the surface temperature is less than 473 K, the gas temperature is from 300K to 800K, and the target carbon-containing product is C2H6..Attorney Docket No. 10046-649W01 29. The method of claim 1, wherein the metal catalyst is Cu and it is not coked or deactivated at a surface temperature of less than 500 K and gas temperature of greater than 2000K for up to 840 minutes.

30. The method of claim 1, wherein the target carbon -containing product is C2H6, the vibrational temperature is from 100 K to 6000K, the surface temperature is from 350K to 450 K, and the catalyst binding energy of from 0.1 eV to 1 eV.

31. The method of claim 1, wherein the target carbon-containing product is CH4, the vibrational temperature is from 800 K to 6000K, the surface temperature is from 500K to 1000 K, and the catalyst binding energy of from -1.2 eV to -0.8 eV.

32. The method of claim 1, wherein the target carbon-containing product is C4H10. the surface temperature is from 600 K to 1000 K, and the catalyst binding energy is from -1.4 eV to -0.2 eV.

33. The method of claim 1, wherein H2 is produced at least at 14 %, the vibrational temperature is from 800 K to 6000K, the surface temperature is from 550 K to 1000 K, and the catalyst binding energy of from -1.5 eV to 0 eV.

34. The method of claim 1, wherein electrical energy is used to heat the methane-containing gas, the metal catalyst, and / or make the non-thermal plasma.

35. A method of selectively producing a target carbon -containing product during the production of a liquid fuel from a methane-containing gas and a vibrationally active oxygen-containing co-reactant, the method comprising:a. forming a non-thermal plasma in the presence of the methane-containing gas, the vibrationally active oxygen-containing co-reactant, and a metal catalyst; i. wherein the methane-containing gas has a gas temperature,ii. wherein the metal catalyst has a surface temperature, andiii. wherein the non-thermal plasma produces vibrationally excited methane having a methane vibrational temperature and vibrationally excited oxygen- containing co-reactant having an oxygen-containing co-reactant vibrational temperature, and that each dissociatively adsorbs on the metal catalyst; and b. adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the methane vibrational temperature, adjusting the oxygen-containing co-reactant vibrational temperature, and / or selecting a different metal catalyst,Attorney Docket No. 10046-649W01 thereby producing a liquid fuel and the target carbon-containing product, wherein the target carbon-containing product is produced in a greater amount than non-target carbon- containing products.

36. The method of claim 35, wherein the vibrationally active oxygen-containing co-reactant is CO?, CO, NO?, and / or O2.

37. The method of claim 35, further comprising forming the non-thermal plasma in the presence of a hydrogen-containing co-reactant.

38. The method of claim 37, wherein the hydrogen-containing co-reactant is H2 and / or H2O.

39. The method of claim 35, wherein the target carbon-containing product is methanol, ethanol, acetic acid, C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, or fullerenes.

40. The method of claim 35, wherein the target carbon-containing product is:a. methanol and the non-target carbon-containing products are ethanol, acetic acid C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, and fullerenes;b. ethanol and the non-target carbon-containing products are methanol, acetic acid C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, and fullerenes: orc. acetic acid and the non-target carbon-containing products are methanol, ethanol, C2H6, C2H4, C2H2, carbon black, graphite, graphene, carbon nanotubes, and fullerenes.

41. The method of claim 35, wherein the non-thermal plasma is formed by a dielectric barrier discharge, diffused plasma discharge, corona discharge, RF coupled plasma, micro wave reactor, or plasma jets.

42. The method of claim 35, wherein the non-thermal plasma is formed with gas temperatures ranging from 300 K to 800K and vibrational temperatures greater than 1500 K.

43. The method of claim 35, wherein the non-thermal plasma is formed at 1 atm.

44. The method of claim 35, wherein the methane-containing gas is liquified natural gas.Attorney Docket No. 10046-649W01 45. The method of claim 35, wherein the methane-containing gas is from a distributed methane source.

46. The method of claim 45, wherein the distributed methane source is from wetland emissions, termite activity, ocean release, natural gas reserves, dumps, pump jacks, or tail gas streams from an industrial process.

47. The method of claim 35, wherein the metal catalyst is a transition metal catalyst.

48. The method of claim 35, wherein the metal catalyst is a metal oxide.

49. The method of claim 35, wherein the metal catalyst comprises a mixture of more than one different metal.

50. The method of claim 35, wherein the metal catalyst is a bimetallic catalyst.

1. The method of claim 35, wherein the metal catalyst is on a support.

52. The method of claim 51, wherein the support is a dielectric support selected from the group consisting of Al2O3, SiO?, TiCh, zeolites, activated carbon, glass, and ceramic.

53. The method of claim 35, wherein the metal catalyst has a binding energy of carbon to a metal atom in the metal catalyst of greater than -0.5 eV.

54. The method of claim 35, wherein the metal catalyst is Cu, Ag, Au, or any combination thereof.

55. The method of claim 35, wherein the metal catalyst is Re, Pt, Rh, Ru, Pd, Ni, Ni, Co, Zn, or any combination thereof.

56. The method of claim 35, wherein the surface temperature of the metal catalyst is adjusted to from 300K to 1000 K.

57. The method of claim 35, wherein the gas temperature is adjusted to from 300K to 800K.

58. The method of claim 35, wherein the method further produces water and / or hydrogen.

59. The method of claim 35, wherein electrical energy is used to heat the methane- containing gas, the metal catalyst, and / or make the non-thermal plasma.Attorney Docket No. 10046-649W01 60. The method of claim 35, wherein the target carbon-containing product is methanol, the metal catalyst is Cu, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

61. The method of claim 35, wherein the target carbon-containing product is ethanol, the metal catalyst is Cu, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

62. The method of claim 35, wherein the target carbon-containing product is C2H6, the metal catalyst is Cu, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

63. The method of claim 35, wherein the target carbon-containing product is ethanol, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

64. The method of claim 35, wherein the target carbon-containing product is acetic acid, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

65. The method of claim 35, wherein the target carbon-containing product is ethane, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

66. The method of claim 35, wherein the target carbon-containing product is ethene, the metal catalyst is Ni, the surface temperature is 473K, and the vibrational temperature is from 2000 K to 5000 K.

67. The method of claim 35, wherein ratio of CH4 to CO2 co-reactant is 1:3.

68. The method of claim 35, wherein the ratio of CH4 to CO2 and H2 co-reactants is from 1:1:4 to 1:1:1.

69. A device for producing a product from a gaseous reactant, comprising:a. a reaction chamber having an inlet for the gaseous reactant and outlet for the product;b. a valve connected to the inlet that controls flow rate of gaseous reactant into the reaction chamber and a valve connected to the outlet that controls flow rate ofAttorney Docket No. 10046-649W01 product out of the chamber;c. two electrodes, wherein the reaction chamber is positioned between the two electrodes,d. an electrical energy supply connected to the two electrodes to supply sufficient electrical energy to produce a non-thermal plasma from the gaseous reactants; e. a metal catalyst positioned inside the reaction chamber; andf. an optional electrical heater connected to the device to heat the reaction chamber, heat the metal catalyst, and / or heat the gaseous reactant.

70. The device of claim 69, wherein the gaseous reactant is from a distributed source that is in close proximity to the device.

71. The device of claim 69, wherein the electrical energy supply is from a renewable energy source.

72. The device of claim 69, wherein the catalyst is on a packed bed.

73. The device of claim 69, wherein the catalyst is on a film.

74. A method of removing nitrous oxide from a gas stream, the method comprising:a. forming a non-thermal plasma in the presence of a nitrous oxide-containing gas, an additional co-reactant, and a metal catalyst,i. wherein the nitrous oxide-containing gas has a gas temperature, ii. wherein the metal catalyst has a surface temperature, andiii. wherein the non-thermal plasma produces vibrationally excited nitrous oxide having a nitrous oxide vibrational temperature, a vibrationally excited co¬ reactant having a co-reactant vibrational temperature, and that each dissociatively adsorbs on the metal catalyst; andb. adjusting the gas temperature, adjusting the surface temperature of the metal catalyst, adjusting the nitrous oxide vibrational temperature, adjusting the co-reactant vibrational temperature, and / or selecting a different metal catalyst; thereby removing nitrous oxide from the gas stream, and producing nitrogen and oxygen gases.

75. The method of claim 74, wherein the co-reactant is CH4, CO2, and / or H2.

76. The method of claim 74, wherein the co-reactant is a 1:1:4 mixture of H2, CO2, and H2.

77. The method of claim 74, wherein the method additionally produces C2H6, C2H4, C2H2.Attorney Docket No. 10046-649W01 carbon black, graphite, graphene, carbon nanotubes, or fullerenes.

78. The method of claim 74, wherein the non-thermal plasma is formed with gas temperatures ranging from 300 K to 800K and vibrational temperatures greater than 1500 K.

79. The method of claim 74, wherein the non-thermal plasma is formed at 1 atm.

80. The method of claim 74, wherein the metal catalyst is a transition metal catalyst.

81. The method of claim 74, wherein tire metal catalyst is a metal oxide.

82. The method of claim 74, wherein the metal catalyst comprises a mixture of more than one different metal.

83. The method of claim 74, wherein the metal catalyst is a bimetallic catalyst.

84. The method of claim 74, wherein the metal catalyst is on a support.

85. The method of claim 84, wherein the support is a dielectric support selected from the group consisting of Al2O3, SiCh, TiCh, zeolites, activated carbon, glass, and ceramic.

86. The method of claim 74, wherein the metal catalyst is Cu, Ag, Au, or any combination thereof.

87. The method of claim 74, wherein the metal catalyst is Re, Pt, Rh, Ru, Pd, Ni, Ni, Co, Zn, or any combination thereof.

88. The method of claim 74, wherein the surface temperature of the metal catalyst is adjusted to from 300K to 1000 K.

89. The method of claim 74, wherein the metal catalyst is Cu, the surface temperature is greater than 640 K, and the gas temperature is from 300K to 800K.

90. The method of claim 74, wherein the metal catalyst is Ni, the surface temperature is less than 573 K, and the gas temperature is from 300K to 800K.

Images