Open Access Article
Liangpang
Xu†
a,
Yi
Xie†
a,
Bichao
Wu
a,
Qian
Lu
a,
Lu
Wang
b and
Ying
Wang
*a
aDepartment of Chemistry, The Chinese University of Hong Kong, Hong Kong S. A. R., China. E-mail: ying.b.wang@cuhk.edu.hk
bSchool of Science and Engineering, The Chinese University of Hong Kong (Shenzhen), Shenzhen, China
First published on 12th March 2026
Non-thermal plasma (NTP) is an emerging technology for the conversion of CO2 and CH4 under mild conditions. This mini review systematically summarizes recent advances in NTP catalysis for the direct conversion of CO2 and CH4 into value-added oxygenates, with a focus on two key aspects: catalyst design and reactor optimization. The metal active sites (e.g., Cu, Ni, Co) and their properties (valence state, dispersion) are critical in directing reaction pathways towards specific oxygenates like alcohols or acids, while the support material modulates performance by influencing the local electric field and stabilizing intermediates. Dielectric barrier discharge (DBD) reactors are predominant, and innovations in reactor structure, electrode design (e.g., water electrodes, surface microdischarge), and configuration (e.g., plasma bubble reactors) are crucial for enhancing efficiency and selectivity, even enabling long-chain hydrocarbon formation. Despite progress, challenges in selectivity and energy efficiency remain. Future efforts should focus on rational catalyst design and advanced reactor integration to advance the industrial application of NTP for greenhouse gas valorization.
Keywords: Non-thermal plasma; Plasma catalysis; CO2 and CH4 conversion; Oxygenates.
Traditional thermal catalysis is typically indirect and relies on a two-step sequence: (1) dry reforming of methane (DRM) at high temperatures (>700 °C) to generate syngas (CO and H2),7,8 and (2) high-pressure conversion of syngas to fuels and chemicals.9 Despite technological maturity, this route is energy intensive, depends on fossil heat and H2, and causes significant CO2 emissions. Biological approaches rely on the precise action of enzymes, but suffer from low volumetric productivity, large land footprint, and high water and nutrient consumption.10 Recently, chemical methods powered by renewable energy have attracted increasing attention for carbon-neutral conversion. Photo- and electro-catalytic approaches have been widely reported for CH4 oxidation or CO2 reduction to various oxygenates.11,12 Despite their potential, the conversion efficiency is still highly restrained due to the fundamental challenges in breaking strong C–H bonds in CH4 and activating linear CO2 molecules.11,13,14
Non-thermal plasma (NTP) has emerged as a promising approach for direct co-conversion of CO2 and CH4 into oxygenates under mild conditions.15–19 The highly energetic electrons (typically at 1–10 eV) can simultaneously activate CO2 and CH4via excitation, ionization, and dissociation at near-ambient gas temperatures.20 The resulting intermediates such as radicals and ions then freely recombine to form various products including oxygenates. Although NTP shows considerable promise, significant challenges remain for efficient conversion of CO2 and CH4 into oxygenates. Uncontrolled recombination of reactive species restricts overall oxygenate selectivity (typically <15%) and impedes the targeted formation of specific products.21,22 In addition, energy efficiency is low because a substantial portion of input energy is consumed in generating high-energy electrons and sustaining background plasma processes rather than being effectively utilized for reactant activation. Integrating catalysts with NTP offers a viable strategy to address these issues by lowering apparent activation barriers, tuning surface energetics, and directing reaction pathways toward desired products.23,24 Reactor design further plays a critical role in affecting discharge characteristics and reaction mechanisms.25 Considerable efforts have been devoted to rational catalyst and reactor engineering to advance CO2/CH4 co-conversion.25,26 A comprehensive review of these developments is therefore essential to consolidate current achievements, identify remaining gaps, and guide future research toward industrial implementation.
While previous reviews have extensively covered the fundamentals and broad applications of plasma catalysis,27–29 a dedicated analysis targeting the direct and selective one-step synthesis of liquid oxygenates from CO2 and CH4 – encompassing integrated advances in catalyst design, reactor engineering, and interface mechanisms – remains insufficiently addressed. Here, we provide a systematic review of NTP technology for the conversion of CO2 and CH4 into oxygenates, focusing on both catalyst design and reactor optimization. The mechanisms of NTP-catalyzed reactions are first elucidated. We further analyze the two key components of catalysts – metal active sites and the support environment, to highlight the structure–activity relationships under plasma conditions to outline fundamental principles for catalyst design. In addition, a detailed analysis of reactor design is presented, covering structural configuration, electrode material selection, and reactor setup. Finally, we outline future perspectives for the co-conversion of CO2 and CH4 into oxygenates via NTP technology, addressing key challenges in selectivity and energy efficiency, with the goal of advancing the application of NTP in greenhouse gas conversion in the future.
O (Ediss = 5.5 eV) and C–H (Ediss = 4.5 eV). Besides, electron density (Ne) and Te are essential for describing NTP activation. Specifically, under representative atmospheric-pressure dielectric barrier discharge (DBD) operating conditions, the electron density typically ranges from 1018–1021 m−3.33
As shown in Fig. 1a, CO2 dissociation can proceed via direct electronic excitation at high electron energies (>7 eV) or stepwise vibrational excitation.24 In the latter route, CO2 climbs the “vibrational ladder” through successive low-energy excitations, eventually reaching the dissociation limit at an overall activation energy of approximately 5.5 eV. Because the input energy closely matches the C
O bond energy and is channeled into vibrational modes rather than electronic excitation, this pathway can be highly energy-efficient and is considered a key channel in warm and certain non-thermal discharges. The practical significance of vibrational excitation depends on plasma parameters. The vibrational excitation pathway is especially efficient in plasmas like gliding arc, microwave, and certain radio-frequency discharges, where energy can be selectively channeled into vibrational modes.24 In DBD, both direct electronic excitation and vibrational mechanisms can contribute, with their relative importance influenced by specific operating conditions (e.g., specific energy input, gas composition).24 CO2 fragmentation predominantly produces ·CO and ·O radicals or vibrationally excited CO2 molecules. These reactive species trigger a cascade of radical chain reactions and secondary recombination step.
![]() | ||
| Fig. 1 (a) Energy-level diagram for CO2 electronic and vibrational modes.24 Copyright 2017, Royal Society of Chemistry; (b) overview diagram illustrating the potential reaction routes for the formation of oxygenate products in CH4 and CO2 plasma. The numbers in (b) represent the reported rate coefficient, m3 per molecule per s.15 | ||
By contrast, under typical atmospheric-pressure DBD plasma for CH4 and CO2 conversion, the vibrational activation of CH4 is inefficient due to its complex vibrational spectrum and unfavourable energy coupling with the electron energy distribution.35 The stretching and bending modes of the C–H bonds do not couple effectively with typical plasma electron energies, and CH4 vibrationally excited states exhibit very short lifetimes, undergoing rapid vibrational-to-translational relaxation.36 As a consequence, vibrational energy is quickly converted into heat rather than being accumulated stepwise, and CH4 cannot effectively undergo ladder-climbing dissociation. Electron-impact dissociation of CH4 therefore proceeds mainly via direct electronic excitation, yielding abundant ·CH3 radicals together with smaller amounts of ·CH2 and ·CH species. These reactive species trigger a cascade of radical chain reactions and secondary recombination steps, underpinning the complex chemistry of NTP-driven CO2/CH4 conversion. It should be noted, however, that CH4 can also undergo substantial vibrational excitation in appropriately designed plasma environments, particularly under low-pressure or microwave discharge conditions – where pronounced vibrational nonequilibrium behavior has been experimentally observed.35 Furthermore, vibrationally excited species such as CO2(V), as well as plasma-activated catalyst surfaces, can transfer energy to CH4 through Eley–Rideal interactions or enhanced vibrational–vibrational coupling, thereby opening alternative activation pathways in plasma-catalytic systems.37
Various oxygenated products can be produced in NTP-driven CO2/CH4 systems, including methanol (CH3OH), formic acid (HCOOH), ethanol (CH3CH2OH), acetic acid (CH3COOH), etc. (Fig. 1b, Table 1).23,24 Methanol is the dominant C1 oxygenate in most plasma-catalytic systems, formed through reactions between ·CH3 radicals and O-containing intermediates (e.g. ·O, ·OH species). Alternatively, methanol may arise via a CO2 reduction pathway involving successive hydrogenation of CO2 through formate-type intermediates (CO2 + ·H → HCOO· → CH3O· → CH3OH). Ethanol formation requires C–C coupling between C1 fragments, typically through reactions between ·CH3 and ·CH2OH species. Acetic acid is generally produced via the coupling of ·CH3 with ·COOH or bicarbonate (HCO3−) intermediates.
| Catalyst | Experimental conditions | Conversion (%) | Energy efficiency | Main liquid products and selectivity (%) | Liquid selectivity (%) | Ref. | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| CO2/CH4 | Plasma | Reactor | Power (W) | CO2 | CH4 | |||||
| Cu/γ-Al2O3 | 1 | DBD | 3 mm gap, 45 mm length | 10 | 15.4 | 18.3 | — | CH3COOH (40.2), CH3OH, C2H5OH, C3H6O | 59.1 | 15 |
| Fe–SiO2 | 2 : 1 |
DBD | 3 mm gap, 45 mm length | ∼10 | ∼34.5 | ∼45.5 | ∼20.5 (SIE) | CH3OH (31), CH3COOH (12.4), C2H5OH | ∼40.0 | 16 |
| Co–SiO2 | ∼33.5 | ∼47.5 | ∼21 (SIE) | CH3OH (20.5), CH3COOH (17.9), C2H5OH | ||||||
| — | 1 : 1 |
DBD | Temperature-controlled water electrode | 15 | 29.0 | ∼32.0 | 0.83 mol kWh−1 (CH3OH) | CH3OH (43) | 60.0 | 40 |
| Cu/Al(OH)3 | 1 : 1 |
DBD | 3 mm gap, 45 mm length | 5 | ∼17 | ∼7 | — | CH3OH (22), C2H5OH (14), C3H8O | ∼44% | 41 |
| K-Fe/C-SiO2 | 1 : 1 |
DBD | 3.5 mm gap, 150 mm length | 2.37 | 5.17 | 15.34 | 0.58 mmol kJ−1 | CH3COOH (12.97), C3+ acid (7.35), CH3OH, C2H5OH | 21.76 | 42 |
| NiGa/NF | 1 : 1 |
DBD | 2.25 mm gap, 50 mm length | 25 | 8.5 | 16.0 | 0.168 mmol kJ−1 | CH3COOH (17.8), CH3OH (∼6.2), C2H5COOH | ∼34 | 43 |
| Ni/HZSM-5 | 1 : 1 |
DBD | 3 mm gap, 45 mm length, water electrode | 18 | ∼2.3 | ∼4 | — | Alcohols (C1–C4), CH3COOH, C3H6O, aldehyde | 45 | 44 |
| Ni–Fe/Al2O3 | — | DBD | 2 mm gap | 23 | 19.7 | 34.8 | — | CH3COOH (23.9), CH3OH (19.1), CH3CH2OH (14.5), C3H8O (15.7), C2H5COOH (16.3), HCHO (10.5) (in liquid) | 38.6 | 45 |
| CuSA-CN | 1 : 1 |
DBD | — | 10 | 7.31 | 11.58 | 0.28 mmol kJ−1 | CH3COOH (11.85), CH3OH (5.08), C2H5OH, C3H7OH, C2H5COOH | — | 46 |
| Cu/5A | 1 : 1 (with trace water) |
DBD | 2 mm gap, 60 mm length | 36 | ∼23 | ∼40 | 73.6 kJ L−1 | CH3OH (∼18%), C2H5OH (5.08), CH3COOH, C3H8O | 26 | 47 |
| Cu/UiO-66-NH2 | 1 : 1 |
DBD | — | 20 | 21.6 | 23.5 | — | CH3OH (20.9), C2H5OH (18.4) C3H6O(8.6), CH3COOH (3.3), CH3CHO (2.3) | 53.4 | 48 |
| Fe2O3/g-C3N4 | 1 : 1 |
DBD | 3 mm gap, 50 mm length | 20 | 18 | 22 | 0.35 mmol kJ−1 | CH3OH (24.4%), C2H5OH (5.08) | 40.1 | 49 |
| Au-HZSM-5 | 1 : 1 |
DBD | 3 mm gap, 40 mm length | 12 | ∼12 | ∼20 | — | R-COOH, R-OH, R-CO-R, R-CHO | ∼32 | 50 |
| P-CoMgAl/NF | 1 : 1 |
DBD | 50 mm length | — | ∼10.5 | ∼16 | 0.16 mmol kJ−1 | C2H5OH, CH3OH, C3H6O2, C2H5OH | 40 | 31 |
In plasma-only systems, the short lifetimes of radicals (micro- to milliseconds) and the absence of defined surface sites lead to largely random gas-phase interactions, multiple parallel pathways and extensive oxidation or cracking.24 As a result, plasma alone usually yields broad product distributions and modest energy efficiency. Coupling NTP with heterogeneous catalysis (plasma catalysis) can effectively overcome these limitations. In such hybrid systems, the plasma rapidly activates CO2 and CH4, whereas the catalyst offers active sites that stabilize key intermediates and steer the reaction toward desired products. On the catalyst surface, plasma-generated radicals, ions and excited species predominantly react via Eley–Rideal (E–R) and Langmuir–Hinshelwood (L–H) mechanisms.38,39 In the E–R mechanism, a gas-phase radical directly collides with and reacts with an adsorbed species, forming products through a low-barrier pathway. In the L–H mechanism, two adsorbed intermediates diffuse on the surface and react with each other. The interplay between these plasma-driven gas-phase reactions and surface catalytic steps is crucial for achieving selective and energy-efficient CO2/CH4 conversion to oxygenates.
Beyond classical radical stabilization, plasma-catalytic systems benefit from additional electronically driven surface processes. Plasma emission or external illumination can generate localized electrons on semiconductor catalysts such as TiO2 and ZnO, which then interact with plasma-generated radicals and lower activation barriers.54,55 Functionalized nanocarbons (e.g., doped graphene, CNTs) can accumulate transient surface charge under plasma exposure, creating strong local electric fields that polarize adsorbates and modify reaction trajectories.56 Additionally, plasmonic catalysts such as ordered Au or Ag nanoparticles on TiO2 offer unique near-field enhancement, hot-electron injection, and photothermal activation pathways. Experiments in plasma-illuminated Au/TiO2 systems demonstrate light-plasma synergy that enhances CH4 activation and alters microdischarge behavior.57 Meanwhile, plasmon-enabled resonance energy transfer mechanisms (e.g., PC-RET) provide a valid conceptual basis for coupling plasmonic fields with reactive plasma species.58,59
For example, Guo et al. reported that exposure of manganese oxide catalysts to DBD plasma reduced granularity and increased dispersion, thereby enlarging the specific surface area and creating more active sites (e.g., vacancies, edges, and corner atoms), which enhances catalytic reactivity.61 Similarly, plasma can modify the oxidation state of catalysts. In plasma-assisted DRM, studies by Gallon et al.62 and Tu et al.63 observed the reduction of NiO to metallic Ni under plasma treatment. The plasma-enabled formation of highly dispersed, ultrasmall Ni nanoparticles significantly improves catalytic activity at low temperatures, enhances CO2 adsorption, mitigates coking through size effects, and ensures exceptional long-term stability.64
In plasma-assisted co-conversion of CH4 and CO2, the highly reactive environment can significantly alter Ni- or Cu-based catalysts, inducing surface oxidation, reduction, sintering, or phase transformations. For instance, Ni0 can be partially oxidized to NiO, while Cu may cycle between Cu0 and Cu+/Cu2+ states. Such changes can either deactivate the catalyst by blocking active sites or, conversely, enhance activity by promoting redox dynamics and improving carbon resistance. The net effect on catalytic performance depends on plasma parameters, catalyst structure, and gas composition. Therefore, the dynamic structural evolution of metal nanoparticles in CO2/CH4 plasma catalysis is an important aspect worthy of deeper investigation.
More intriguingly, when these vibrationally excited molecules interact with nanocatalysts exhibiting localized surface plasmon resonance (LSPR) (e.g., Au, Ag), additional synergy may arise: plasmon decay produces hot electrons or strong localized fields that promote adsorbate polarization and, via electron-vibration coupling, excite vibrational modes.69 If gas-phase vibrational energy matches surface excited states, resonant energy transfer may occur, enhancing efficiency and selectivity.70,71 Although this “vibration-plasmon” synergy remains entangled with thermal and radical effects, it represents a core non-thermal pathway, offering new strategies for designing low-energy plasma-assisted catalytic systems.
There are several primary methods and related metrics used to assess energy efficiency:
![]() | (1) |
![]() | (2) |
The standard reaction enthalpy for pure CO2 dissociation is 283 kJ mol−1 (equivalent to 2.93 eV per molecule), whereas for the DRM reaction, it is 247 kJ mol−1 (or 2.56 eV per molecule). The SEI and P are the specific energy input and plasma power (W), respectively.
Energy consumption (EC) refers to the amount of energy required to carry out the reaction and is typically expressed in terms of energy per mole or per molecule converted (in units of kJ mol−1 or eV per molecule):
![]() | (3) |
To convert energy consumption into eV, the following relationship is used:
![]() | (4) |
Compared with conventional thermal catalysis, NTP-catalysis still faces key challenges, particularly limited energy efficiency and product selectivity (Table S1), and thus is not a direct replacement but rather a complementary technology suited for specific scenarios. Its unique niche stems from the ability to activate inert molecules (such as CO2, N2, and CH4) under mild, near-ambient conditions, enabling reaction pathways inaccessible to thermal approaches. This advantage positions NTP for promising applications, including coupling with intermittent renewable energy sources to enhance economic viability, enabling modular and distributed small-scale chemical production (e.g., gas upgrading), and potentially revolutionizing the synthesis of high-value chemicals where process flexibility and tunability outweigh raw conversion efficiency. Realizing this potential requires focused efforts to deepen understanding of plasma–catalyst interactions, selectively channel energy into desired reaction channels, and overcome scale-up barriers such as discharge uniformity and thermal management through advanced diagnostics and modelling. Bridging these gaps is essential to transition NTP-catalysis from laboratory research toward industrial implementation.
Within this plasma-specific framework, the structural design of catalysts becomes critical for enhancing the adsorption and reaction of plasma-derived active species involved in CH4 and CO2 activation-processes that ultimately govern oxygenate selectivity.72 Currently, catalysts employed in NTP technology are predominantly supported catalysts, which primarily consist of two key components: the metal active sites and the support (Table 1). The selection of metals (such as Cu, Ni, Co, Fe metals) primarily determines the reaction pathway, while their structural properties including valence state, particle size, and dispersion also significantly influence the reaction selectivity and activity. The support, beyond the basic role of metal loading, can also alter the plasma discharge characteristics, modulate the electronic structure of active species, and provide essential acid–base sites. This synergistic interaction between the metal active sites and support collectively creates the “active center” for oxygenate generation within the plasma environment. In this section, we systematically introduce widely studied metals for CO2/CH4 conversion, including Cu-, Ni-, and Co-based systems, as well as other representative metals. Furthermore, we discussed the influences of support in modulating the metal active sites and its impact on the plasma-catalytic performance.
![]() | ||
| Fig. 2 (a) The mole ratio of R-COOH/R-OH (CH3COOH/C1–C4OH) as a function of Cu+ percentage; (b) possible reaction pathways for plasma-catalytic conversion of CO2 and CH4 to CH3OH and CH3COOH.73 Copyright 2022, Elsevier; (c) selectivity of oxygenates and relative selectivity of acid to alcohol for Cu-SA-CN; (d) calculated free-energy diagram for CH4 and CO2 to generate CH3COOH on CN and Cu-SA.46 Copyright 2024, Elsevier; (e) effect of water addition on the selectivity for methanol and acetic acid and effect of discharge power on the selectivity for methanol and acetic acid in the (f) Cu/Al2O3 + plasma case and (g) plasma-only case.75 Copyright 2025, American Chemical Society. | ||
For instance, in Cu/Al(OH)3 catalysts calcined at 540 °C with 5 wt% Cu loading, the strong redox capacity and abundant Cu2+ species led to high alcohol selectivity of >38%.41 In contrast, Cu/HZSM-5 with a high proportion of Cu+ species significantly enhanced acetic acid selectivity, aided by the presence of Brønsted acid sites (BAS) in HZSM-5 supports which facilitate CO2 protonation and C–C coupling.73
The coordination structure of Cu active sites also influences catalytic performance. Single-atom Cu catalysts (CuSA) with a Cu–N4 coordination structure anchored on carbon nitride (CuSA-CN) not only accelerate the activation of CH4 and CO2 but also lower the energy barriers for CH4 dissociation and C–C coupling, thereby promoting acetic acid formation (Fig. 2c).46 DFT calculations revealed that the energy barrier for the key step of *CH3 and *CO2 coupling to form *CH3COO is significantly reduced on CuSA-CN compared to the bare support without Cu active sites (Fig. 2d), leading to a higher selectivity toward acetic acid.
Moreover, external conditions such as water addition and discharge power can further tailor the product distribution. The introduction of water vapor enhances methanol selectivity by participating in methanol formation and facilitating its desorption from the catalyst surface (Fig. 2e), while increasing the discharge power can shift the dominant product from methanol to acetic acid when using Cu/Al2O3, which is not significant for the plasma only system (Fig. 2f and g).75 These findings underscore the versatility of Cu-based catalysts in plasma-catalytic systems and highlight the importance of catalyst design and reaction conditions in steering the conversion of CH4 and CO2 toward desired oxygenates.
Li et al. designed a series of structured Ni-foam (NF) supported catalysts to elucidate the role of Ni active sites.43 As shown in Fig. 3a–c, the Ni0 in Ni/NF and NiGa/NF catalysts showed high total liquid selectivity (>30%) with CH3COOH as the dominant product (>15% selectivity). In contrast, the Ni2+ in NiO/NF yielded comparable selectivity for CH3OH (8.9%) and CH3COOH (9.6%). Surprisingly, NiAl-layered double hydroxide (LDH)/NF, rich in surface –OH groups, exhibited the highest CH3OH selectivity (12.3%) and negligible carbon deposition. An “–OH reservoir” mechanism was proposed (Fig. 3d), where the abundant –OH groups on ultrathin LDH nanosheets readily react with
radicals to form methanol, while the inferior electron affinity of the –OH rich surface for carbon atoms endows the catalyst with high coke resistance.
![]() | ||
| Fig. 3 (a) Ni 2p XPS spectra of NiAl-LDH/NF, NiO/NF, Ni/NF and NiGa/NF; (b) the selectivity of liquid products; (c) absolute concentration of the oxygenates; (d) the possible reaction pathways for the formation of liquid products on different catalysts via the plasma-catalytic approach.43 Copyright 2023, Elsevier; (e) oxygenate generation from CH4/CO2 over Ni/HZSM-5 catalysts packed in the DBD reactor with Al foil as the ground electrode.44 Copyright 2023, Springer Nature. | ||
Catalyst preparation, loading amount, and reactor design strongly influence the performance of Ni catalysts. Fan et al. used the solid ion exchange method (SIE) to prepare Ni/HZSM-5 and achieved ∼45% oxygenate selectivity (Fig. 3e).44 High Ni loading promotes alcohol formation via stronger Lewis acid sites but suppresses acetic acid. In contrast, low Ni loading (1 wt%) with the liquid ion-exchange method (LIE) or SIE favors acetic acid due to abundant Brønsted acid sites and framework Ni2+. Using an Al foil electrode instead of water cooling significantly enhances alcohol selectivity, likely due to differences in CO electronic states affecting alcohol vs. acetic acid formation.
A key feature of Co catalysts is the complementary role played by metallic Co sites and oxygen vacancies (Ov) on the support. Ov in reducible oxides (e.g., CeO2, TiO2) are well-known for activating CO2 in thermal catalysis via electron donation. Under plasma, however, their role expands: Ov act as electron traps that stabilize anionic plasma intermediates such as CO2− or ·CO2−, which are otherwise short-lived in the gas phase. This stabilization enables subsequent hydrogenation or coupling reactions that bypass high-barrier thermal pathways. For example, Li et al. demonstrated that using Co/SiO2 aerogel catalysts under ambient plasma conditions increased total liquid product selectivity to about 40%, with methanol and acetic acid as major products.16 Traces of long-chain compounds such as C6–C7 hydrocarbons and esters were also detected, suggesting the intrinsic chain-growing ability on Co active sites. In subsequent work, Li et al. combined Co/SiO2 with HZSM-5 zeolite in a composite bed, which further enhanced the formation of C5–C8 products including heptane, pentanol, and C6 esters, highlighting the synergy between Co-mediated C–C coupling and zeolite-assisted oligomerization.76
Further insight into the Co–support interaction was provided by Dou et al., who designed a Co–MgAlO–Ov structured catalyst.31 The catalyst delivers a total oxygenate selectivity of 40%, attributed to the synergistic effect between Ov in MgAlO supports and the small size of Co nanoparticles (Fig. 4a). Moreover, distinct distributions of liquid products are observed across different Co-based catalysts (Fig. 4b). Through combined experimental and computational studies, they identified metallic Co as the main site for acetic acid formation, where it strongly adsorbs CH3 and COOH radicals, enabling coupling via a low-barrier E–R mechanism (Fig. 4c). In contrast, Ov on the MgAlO support preferentially captured CO2 and oxygenated radicals (e.g., O and OH), shifting the product towards methanol with a selectivity of ∼9% (Fig. 4d). This bifunctional mechanism allows product distribution to be tuned by tailoring the catalyst's surface composition.
![]() | ||
| Fig. 4 (a) Overall selectivity for major products and (b) selectivity toward liquid oxygenates over different Co-based catalysts; the Ov-assisted reaction pathways (red arrows) for the formation of (c) CH3COOH and (d) CH3OH on Co-based catalysts.31 Copyright 2022, Elsevier. | ||
![]() | ||
| Fig. 5 (a) Selectivity to gaseous hydrocarbons and (b) selectivity to liquid oxygenates over different Fe-based catalysts; (c) reaction mechanisms proposed for plasma-catalytic conversion of CO2 and CH4.42 Copyright 2023, Elsevier; selectivity to liquid oxygenates for (d) Au/HZSM-5, HZSM-5-ST, Au/HZSM-5-ST and Au/S-1 catalysts, and (e) 13X, Au/13X, 13X-ST and Au/13X-ST in plasma-catalytic CH4/CO2 conversion; (f) reaction pathways for acetic acid formation over BAS in plasma-catalytic CH4/CO2 conversion on HZSM-5 zeolite.50 Copyright 2024, Elsevier. | ||
Au-based catalysts present unique catalytic effects towards higher selectivity for acetic acid. Wang et al. designed a series of Au-zeolite catalysts to elucidate the role of Au species and acid sites.50 The Au/HZSM-5 catalyst, containing both Au3+ and BAS, exhibited the highest selectivity toward CH3COOH (Fig. 5d). In contrast, Au/HZSM-5-ST, which lost both BAS and Au3+ after steam treatment, showed a dramatic decrease in acetic acid selectivity and a shift toward alcohols. Similarly, Au/13X, with only Lewis acid sites (LAS) and a small amount of Au3+, still favors R-OH production, though the presence of Au3+ slightly enhanced acetic acid formation (Fig. 5e). A “Au3+-BAS synergy” mechanism was proposed, where Au3+ species located in the H+ sites of HZSM-5 enhance CO2 activation and facilitate the bicarbonate pathway toward acetic acid, while the poor dispersion and low loading of Au limit its overall catalytic activity in the plasma environment (Fig. 5f).
This synergy is exemplified by BAS in zeolitic supports such as HZSM-5, whose function diverges markedly between thermal and plasma environments. In conventional thermal catalysis, BAS protonates olefins to form carbenium-like intermediates that drive C–C coupling. Under NTP conditions, however, olefins are scarce; instead, BAS act as proton donors to plasma-generated anionic species such as CO2− or ·CO2−, yielding bicarbonate-like surface intermediates.50 These species readily undergo radical-surface reactions with gas-phase ·CH3 to form oxygenates like acetic acid or ethanol. Consequently, the “acidity” that matters in plasma catalysis is not oriented toward carbocation stabilization, but rather toward facilitating proton transfer to and stabilization of charged plasma-generated intermediates. This paradigm shift underscores that support design in plasma catalysis must be guided by plasma-specific reaction mechanisms, not thermal analogues.
The physical properties of the support significantly affect the plasma environment. The dielectric constant (εr) primarily determines the discharge behavior. Materials such as BaTiO3 with a high εr act as macroscopic dipoles, which enhances the local field and the formation of intense micro-discharges. Mei et al. found that BaTiO3 beads could double the average electric field and mean electron energy of the CO2 discharge (Fig. 6a and b), thus boosting CO2-to-CO conversion (Fig. 6c).78 Conversely, the support (e.g., SiO2, Al2O3) with low-εr induces a weaker polarization effect, leading to a more diffuse and uniform discharge (e.g., glow-like), which can be beneficial for uniform surface reactions. Furthermore, porous supports such as zeolites and MOFs exhibit nano-confinement effects.79 By extending the residence time of active species within pores, secondary reactions can be promoted. For example, ZSM-5 zeolite can enhance the selectivity of CH4 conversion to C2 hydrocarbons via CHx intermediate coupling.80
![]() | ||
| Fig. 6 Effect of packing materials (i.e. glass beads and BaTiO3) on (a) average electric field strength, (b) mean electron energy, and (c) CO2 conversion in the CO2 DBD at different discharge powers.78 Copyright 2014, IOP Publishing; (d) gas conversion and total liquid selectivity for different catalysts in CO2/CH4 conversion; (e) proposed mechanism for absorption and activation of CO2 and CH4, and the role of Cu in enhancing *COOH conversion.48 Copyright 2024, American Chemical Society. | ||
Tailoring the electronic structure of metals is a primary function of supports, which directly governs reaction pathways and product selectivity. This is well-illustrated by Cu-based catalysts in the plasma-driven conversion of CO2/CH4. Basic supports like Al(OH)3 and Mg(OH)2 enhance the formation of Cu2+ species, directing the reaction toward production of methanol and other alcohols (Fig. 2a).73 In contrast, the HZSM-5 zeolite support can stabilize Cu+ species and provides Brønsted acid sites, thus facilitating CO2 protonation and C–C coupling. This synergistic effect contributes to higher selectivity toward C2 oxygenates, such as acetic acid. Meanwhile, a balanced Cu2+/Cu+ ratio was observed when dispersed on neutral supports such as SiO2 and TiO2, contributing to both alcohol and acid formation with moderate selectivity.
Furthermore, the surface acidity or basicity of the support critically influences the CO2/CH4 adsorption and activation. Qi et al. used UiO-66-NH2 as a support for Cu in plasma-catalytic CO2/CH4 conversion. Th synergistic effect of Cu and UiO-66-NH2 realizes an overall liquid product selectivity of 53.4%, with 60.8% C2+ oxygenates in the liquid products (Fig. 6d). The UiO-66-NH2 support with a high surface area and porosity offers abundant adsorption sites, while its –NH2 groups serve as Lewis basic sites that enhance CO2 chemisorption and promote *COOH intermediate formation (Fig. 6e). The organic framework can also facilitate CH4 activation to obtain stable *CH3 and *CH2 species. Strong electronic interaction with Cu further promotes C–H and C
O bond cleavage, accelerating the conversion of *COOH into *CO and other intermediates.
![]() | ||
| Fig. 7 (a) Expanded view of a parallel plate DBD reactor.87 Copyright 2021, Elsevier; (b) design of a water electrode as the ground electrode in the DBD reactor.15 Copyright 2017, Wiley-VCH; (c) water-electrode surface microdischarge (SMD) reactor using nickel foam as the high-voltage electrode.88 Copyright 2025, Elsevier; (d) plasma bubble reactor configuration.19 Copyright 2024, American Chemical Society. | ||
Wang et al. designed an innovative electrode structure as shown in Fig. 7b.15 They realized the one-step reforming of CO2 and CH4 into liquid fuels and chemicals, such as acetic acid, methanol, and ethanol, at room temperature and atmospheric pressure, achieving a total oxygenate selectivity of 50–60%, with acetic acid selectivity reaching 40%. Notably, the synthesis of acetic acid from CO2 and CH4 proceeded with 100% atom economy, a reaction considered thermodynamically unfavorable and nearly unattainable via conventional thermal catalysis. The water electrode enhances energy efficiency through a dual functionality: it promotes uniform plasma discharge via superior charge dissipation and electric field redistribution, and effectively removes joule heat to maintain near-ambient reaction temperatures. This combination not only suppresses energy-wasting parasitic reactions (e.g., total oxidation or cracking) but also stabilizes key reactive intermediates (such as ·CH3 and ·COOH), thereby steering electron-driven chemistry toward selective oxygenate formation. By simultaneously optimizing plasma homogeneity and thermal management, the water electrode outperforms conventional solid metal electrodes as a strategic enabler of energy-efficient CO2/CH4 valorization.
In conventional DBD reactors, the plasma region typically spans the entire discharge gap, which tends to cause re-decomposition of generated products, resulting in low conversion efficiency. To address this, Feng et al.88 developed a novel surface microdischarge (SMD) reactor utilizing metal-loaded nickel foam (NF) as the high-voltage electrode (Fig. 7c). By spatially confining the plasma to the dielectric surface, this design enables synthesized NH3 to rapidly diffuse through the porous NF electrode into a plasma-free “product-safe zone”, effectively suppressing energy-wasting reverse reactions. This strategic separation of reaction and product zones directly translates into superior energy efficiency: under identical conditions (SEI = 10 kJ L−1), the SMD system with Ni–Co/NF achieved an NH3 energy yield of 0.79 g kWh−1, a 38.6% improvement over the plasma-only baseline (0.57 g kWh−1). Thus, the SMD electrode architecture not only preserves products but also significantly enhances the fraction of electrical energy channeled into useful chemical output.
Gregory and co-workers developed an Ir/TiO2 thin-film electrode for plasma-assisted CO2 hydrogenation to CH4, demonstrating that the thin-film structure not only stabilized the plasma discharge but also promoted CO2 activation and hydrogenation on Ir sites, achieving superior methane selectivity and energy efficiency compared with particulate catalysts.93 Specifically, this configuration delivered a fuel production efficiency of 3.5%, a key metric of energy utilization, and exhibited a CH4 formation rate two orders of magnitude higher per unit catalyst mass than conventional packed-bed systems. The enhancement stems from the intimate coupling between plasma and the catalyst: by confining the discharge directly over the ultrathin catalytic layer (<2 μm), reactive species (e.g., ·H) are generated in immediate proximity to active Ir sites, minimizing transport losses and avoiding energy-wasting gas-phase side reactions. Moreover, the separation of H2 dissociation (driven by plasma) and CO2 hydrogenation (catalyzed by Ir) reduces the overall energy demand, as the catalyst bypasses the need for plasma-mediated H2 activation. Similarly, in CO2–CH4 co-conversion systems, thin-film electrodes based on Cu, Ni, or Co are expected to leverage intensified local electric fields and directed intermediate transport to enhance the formation of oxygenated products.92,94 Thin-film electrodes are particularly compatible with microdischarge or surface-discharge reactors, which help suppress decomposition of gas products and provide model platforms for mechanistic studies of plasma-catalyst synergy.
A central finding is that catalyst design for plasma environments must move beyond thermal catalysis paradigms. As discussed in sections 2 and 3, the optimal active sites (e.g., Cu2+ for alcohols, Cu+ for acids) and support functions (e.g., Brønsted acid sites for protonating plasma-generated anions) are defined by their ability to capture and steer transient plasma-generated species (radicals, ions, vibrationally excited molecules) rather than just activate stable reactants. Furthermore, energy efficiency remains a paramount challenge. As analyzed in section 4, while innovative reactor designs (e.g., water electrodes, surface microdischarges) can enhance efficiency by improving discharge uniformity and creating product-safe zones, the energy cost of breaking stable C–H and C
O bonds via electron impact is still high. Critically, overcoming the energy efficiency bottleneck requires reactor architectures that not only localize energy input to targeted molecular activation but also maximize plasma-catalyst synergy while minimizing all parasitic energy losses, including those from gas heating, non-productive electron scattering, and unintended side reactions.
Therefore, future research should pursue integrated strategies that co-optimize catalysts and reactors to address both selectivity and energy efficiency. Three grounded directions emerge from the present analysis:
(1) Rational catalyst design for enhancing conversion and selectivity. Future catalysts should be engineered to directly leverage the non-equilibrium energy carriers in plasma. This involves designing “vibrationally-coupled” catalysts (e.g., using oxides like ZrO2 or CeO2 with tailored phonon spectra) that can resonantly accept energy from plasma-generated vibrationally excited CO2 (CO2(V)), a key species discussed in section 2.3.3. This resonant energy transfer could promote surface reactions at lower overall energy input, moving beyond reliance on high-energy electron collisions alone. Furthermore, elevating the C2+ product selectivity is essential for synthesis of high-value chemicals. It's necessary to design catalysts with spatially organized bifunctional sites that can co-adsorb and orient key intermediates for selective C–C coupling. For instance, the catalyst surface with adjacent metallic sites (e.g., Co, Fe) and acidic sites (e.g., HZSM-5) is promising for synergistically adsorbing ·CH3 radicals and activating CO2 into ·COOH, facilitating acetic acid formation via a low-barrier E–R mechanism.
(2) Reactor optimization should prioritize energy localization, intensified mass transfer, and product preservation. Building on the design principles outlined in section 4, a promising path forward is the development of “zoned synergistic plasma reactors”. In this advanced configuration, CO2 and CH4 activation are decoupled and individually optimized in separate plasma zones (e.g., microwave for CO2, DBD for CH4), with the generated reactive intermediates then channeled into a common catalytic zone for selective conversion. This approach enhances energy efficiency by matching each molecule's activation profile to the most suitable plasma source, while also improving overall process control. Furthermore, reactor architecture must be closely integrated with catalyst design. Practical strategies include directly coating catalysts on electrodes (section 4.4), fabricating 3D-printed structured electrodes, and using ordered high-dielectric fillers (e.g., BaTiO3 monoliths) to replace random particle packings.
(3) To elucidate the complex mechanisms in plasma catalysis, future research should integrate advanced in situ diagnostics (e.g., OES, LIF, DRIFTS) with isotopically labelled experiments and targeted probing. Coupling these experimental insights with multiscale theoretical simulations, from DFT to kinetic modelling, is essential. This synergistic approach will enable the unambiguous identification of transient species, disentangle reaction pathways, and distinguish chemical effects from physical transport, thereby advancing the fundamental understanding and rational design of plasma-catalytic processes.
Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5im00347d.
Footnote |
| † These authors contributed equally to this work. |
| This journal is © Institute of Process Engineering of CAS 2026 |