Mechanistic study of catalytic CO2 hydrogenation in a plasma by operando DRIFT spectroscopy

(1)

Mechanistic study of catalytic CO2 hydrogenation in a plasma

by operando DRIFT spectroscopy

Citation for published version (APA):

Parastaev, A., Kosinov, N., & Hensen, E. J. M. (2021). Mechanistic study of catalytic CO2 hydrogenation in a plasma by operando DRIFT spectroscopy. Journal of Physics D: Applied Physics, 54(26), [264004].

https://doi.org/10.1088/1361-6463/abeb96

DOI:

10.1088/1361-6463/abeb96 Document status and date: Published: 23/04/2021

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

providing details and we will investigate your claim.

(2)

Journal of Physics D: Applied Physics

PAPER • OPEN ACCESS

Mechanistic study of catalytic CO

2 hydrogenation in a plasma by

operando DRIFT spectroscopy

To cite this article: A Parastaev et al 2021 J. Phys. D: Appl. Phys. 54 264004

View the article online for updates and enhancements.

(3)

Journal of Physics D: Applied Physics J. Phys. D: Appl. Phys. 54 (2021) 264004 (9pp) https://doi.org/10.1088/1361-6463/abeb96

Mechanistic study of catalytic CO

₂

hydrogenation in a plasma by operando

DRIFT spectroscopy

A Parastaev



, N Kosinov



and E J M Hensen

∗



Laboratory of Inorganic Materials and Catalysis, Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, Eindhoven 5600 MB, The Netherlands

E-mail:e.j.m.hensen@tue.nl

Received 7 January 2021, revised 23 February 2021 Accepted for publication 3 March 2021

Published 23 April 2021 Abstract

Plasma-enhanced heterogeneous catalysis offers a promising alternative to thermal catalysis for many industrially relevant processes. There is only limited mechanistic understanding about the relation between the interactions of highly energetic electrons and excited molecules with heterogeneous catalysts in a plasma and their catalytic performance. Herein, a novel operando infrared spectroscopy cell is presented allowing the investigation of surface intermediates upon exposure of a catalyst to plasma. The polyether ether ketone cell enclosure embedding a quartz reactor is operated at atmospheric pressure and can be heated to 250◦C. A case study involved the characterization of surface intermediates during CO2hydrogenation on a Co/CeZrO4 catalyst. The temperature was monitored using online UV–Vis spectroscopy. This combined approach offers new experimental insights into plasma-catalyst synergy. The most important one is the demonstration of CO2methanation at the catalyst surface at room temperature in a plasma. Keywords: plasma catalysis, operando DRIFTS, CO2hydrogenation, cobalt, ceria-zirconia (Some figures may appear in colour only in the online journal)

1. Introduction

Plasma-enhanced heterogeneous catalysis differs from con-ventional thermal gas-phase catalysis by the interaction of the catalyst surface with a plasma. The plasma-activated gas con-tains not only molecules but also highly reactive species such as radicals, excited species, ions and electrons [1,2]. There remains a lack of mechanistic understanding about the influ-ence of such reactive species on the action of heterogeneous catalysts. Evaluating catalyst performance in plasma is com-plicated by the intricate nature of the plasma-catalyst system:

∗

Author to whom any correspondence should be addressed.

Original content from this work may be used under the terms of theCreative Commons Attribution 4.0 licence. Any fur-ther distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

the properties of the plasma depend on the catalyst material (e.g. an increase of the dielectric constant enhances the elec-tric field) [3] and vice versa (e.g. permittivity and polarization of a catalyst are determined by the plasma parameters). Mech-anistic studies of catalysts in a plasma are, unlike in conven-tional heterogeneous catalysis, hampered by the fact that the reactions occur both at the catalyst surface and in the plasma. Several new experimental and theoretical approaches have already been applied to gain a better understanding of the interaction of solid catalysts with plasma. Among others, iso-topic labeling in catalytic experiments [4], electron paramag-netic resonance spectroscopy [5], chemical looping [6–10], computational methods [11], and IR spectroscopy have been used for this purpose. IR spectroscopy is an important tool in heterogeneous catalysis, because it allows to probe reac-tion intermediates at the surface and, in principle, above the surface, in an in situ and even operando manner. However, there are various challenges associated with carrying out IR

(4)

J. Phys. D: Appl. Phys. 54 (2021) 264004 A Parastaev et al

spectroscopy experiments in, for instance, a dielectric barrier discharge (DBD) plasma. The major problem is how to prop-erly insulate the high-voltage electrode from the IR dome and other conductive elements. Several approaches have already been developed to tackle this issue. Stere et al designed a dedicated diffuse reflectance infrared Fourier transform spec-troscopy (DRIFTS) cell with an embedded plasma jet [12]. This approach is indeed suitable for post-situ plasma cata-lysis (PPC) studies, where the catalyst is placed in the down-stream of plasma reactor [13]. In a few IR studies focusing on

in situ plasma catalysis (IPC), sub-atmospheric pressure was

employed and mostly no heating was applied [13–15]. Another way to overcome the insulation challenge is to manufacture a cell out of dielectric materials such as polytetrafluoroethylene (PTFE) [16], polyether ether ketone (PEEK), or quartz. Quartz IR cells for PPC and IPC were reported by Azzolina-Jury et al, although these cells were operated at low pressure resulting in a glow discharge plasma (not DBD plasma) due to the config-uration of the electrodes [17,18].

Another relevant operando characterization technique is x-ray absorption spectroscopy (XAS), which allows investigat-ing the structure and the oxidation state of the active phase under plasma-catalysis conditions. Gibson et al developed an operando plasma-catalysis XAS cell to address a well-known challenge of plasma-catalysis—overheating of the catalyst upon exposure to plasma. The mean-squared thermal disorder parameter (σ2_{), obtained from the operando extended x-ray} absorption fine structure analysis of a Pd/Al2O3catalyst during methane oxidation reaction, was used to determine the actual temperature of the metal nanoparticles [19]. Due to the gen-eral low availability of the synchrotron radiation this method is not often used to estimate temperature of plasma-catalysts. Other methods to monitor the temperature of plasma catalysts are IR thermography in combination with a sapphire DBD reactor [20], the use of an ethanol thermometer [21] and meas-urements with a thermocouple directly after switching off the plasma.

In the present study, a novel DRIFTS cell manufactured out of PEEK with embedded quartz DBD reactor is demonstrated. The configuration of the cell allowed operando IR, UV–Vis and temperature measurements upon exposure catalytic mater-ials to plasma at atmospheric pressure. The developed plasma DRIFTS cell was used to study the mechanism of plasma-assisted CO2 hydrogenation in the presence of a Co/CeZrO4 (CoCZ) catalyst. The operando spectroscopic data revealed some of the unique aspects of plasma-catalytic reactions for this system. In addition, accurate temperature measurements of the catalyst bed showed that spectroscopically observed res-ults can be attributed to the plasma-induced phenomena rather than just energy dissipation.

2. Experimental 2.1. Cell design

A novel DRIFTS cell was developed for operando measure-ments in a plasma environment. In order to implement the high-voltage circuit in an FTIR spectrometer, a (Spectra-Tech

Figure 1. Design of the operando DRIFTS cell for operation in a plasma environment incorporated in a modified Praying Mantis™ diffuse reflection accessory. 1—High voltage electrode;

2—insulation between high voltage electrode and ellipsoidal mirrors; 3—ellipsoidal mirror; 4—holder; 5—operando DRIFTS cell; 6—dielectric barrier discharge quartz reactor.

Inc.) Praying Mantis™ diffuse reflection accessory was ified. A schematic presentation of the Praying Mantis™ mod-ified for the high-voltage environment is shown in figure1.

The insulator (2) made of PEEK was introduced to pre-vent plasma discharges between the high-voltage electrode (1) and the ellipsoidal mirrors (3). The high-voltage electrode (1) was a stainless-steel rod with a diameter of 3 mm, which was connected to a home-made microsecond-pulsed power supply (discharge frequency up to 1 kHz with variable pulse duration, voltage up to 20 kV, identical to the power supply used in a previous work [6]) connected via a flexible coaxial cable with double shielding. The holder (4) was extended due to the lar-ger size of the plasma DRIFTS cell than a conventional high-temperature DRIFTS cell. The modified Praying Mantis™ cell was embedded in a Faraday cage for electromagnetic shield-ing. The cell compartment was purged with nitrogen prior to experiments to limit the contribution of H2O and CO2 gases to the IR spectra. A detailed design of the DRIFTS cell (5) is shown in figure2.

The cell was made of PEEK and consisted of a base (9), a cylindrical dome (6), and a quartz DBD reactor (7). The dome was equipped with two CaF2windows (3), allowing IR and UV–Vis measurements and also suitable for IR temper-ature sensors operating in the NIR and MIR range. A quartz tubular DBD reactor (7) was incorporated inside the base (9). The outer diameter of the reactor was 8 mm, while the inner diameter was 4 mm (figure3). These dimensions allowed the insertion of a rod-like heater inside the reactor. The top of the reactor was a 19 mm quartz plate with a thickness of 2 mm. The ground electrode (5) was placed directly underneath the plate. The gas mixture was introduced from the bottom of the DBD reactor.

(5)

Figure 2. Design of the operando DRIFTS cell for operation in a plasma environment. 1—High-voltage electrode; 2—IR beam; 3—CaF2window; 4—catalyst powder; 5—ground electrode;

6—dome of the cell; 7—quartz DBD reactor; 8—gas outlet; 9—base.

Figure 3. Drawings of quartz DBD reactor for DRIFTS plasma-cell: top (a) and bottom (b) parts; (c) photograph of the reactor with the inserted heater.

The DBD zone was located between the high-voltage elec-trode and the ground elecelec-trode placed under a dielectric plate. The catalyst powder (4) can be placed on top of the quartz grid, which is underneath the high-voltage electrode. The gap between the high-voltage electrode and the catalyst powder was kept constant (1 mm). The catalyst sample was in the

Figure 4. Temperature maps of the quartz DBD reactor upon conventional heating acquired by an IR temperature sensor: (a) side view and (b) top view.

form of powder (<125 µm) slightly pressed into the sample holder in order to avoid its spreading under the gas flow and plasma. Prior to IR measurements, the catalyst sample was reduced at 500◦C in hydrogen and passivated in 1% O2 in He in a conventional plug-flow reactor. Then, it was trans-ferred to the DRIFTS cell and re-reduced in H2 at 250 ◦C. DBD plasma parameters for the DRIFTS cell were kept the same for all experiments, i.e. a pulse frequency of 1000 Hz, a pulse length of 50 µs, and a voltage of 8 kV. Due to the spe-cific geometry, it was not possible to insert voltage and current probes close enough to the cell in order to measure the power density.

A Nicolet NEXUS 670/870 FTIR spectrometer equipped with a liquid nitrogen cooled mercury cadmium telluride detector was used to record the IR spectra in the 4000– 1000 cm−1range with a resolution of 4 cm−1. Spectra were acquired with and without plasma at room temperature. All the experiments were carried out at atmospheric pressure. 2.2. Temperature monitoring

The current design has the advantage that it is possible to thermally treat the catalyst sample. A DRIFTS cell for oper-ando DBD plasma measurements described by Rodrigues et al [16] made of PTFE did not allow the pretreatment of samples at elevated temperature. Here, the use of a quartz reactor inside a PEEK cell allows heating the catalyst sample to 250 ◦C (figure 4) without degradation of the cell. The temperature in the absence of plasma was controlled by a thermocouple placed on top of the dielectric plate.

An IR temperature sensor was used to monitor the cata-lyst temperature in the plasma. However, the spatial resolution of the IR temperature sensor was too low. Therefore, another method was employed to measure the catalyst temperature in order to exclude the possibility of local overheating of the sample with the plasma turned on. We exploited the thermo-chromic properties of rutile titanium dioxide [22,23]. First, the rutile TiO2 powder was dehydrated in a nitrogen flow at 250◦C. Then, the thermally induced changes of the absorption intensity at 420 nm in the UV–Vis spectrum were measured without plasma (figure5), followed by similar measurements in a plasma.

(6)

Figure 5. UV–Vis spectra of rutile TiO2acquired at different

temperature in absence of plasma.

Figure 6. Temperature in the DBD reactor as a function of time measured by thermocouple and the new UV–Vis approach based on the thermochromic properties of rutile TiO2. Conditions: 1000 Hz,

90 µs pulse, 18 kV energy per pulse 18 mJ.

This approach was also applied for the DBD reactor used in the previous work [6] and compared with the temperature pro-files obtained using a thermocouple connected directly after switching off the plasma. The UV–Vis method led to similar results as obtained with a thermocouple (figure6).

The advantage of the novel approach based on the thermo-chromic material are (a) the possibility to measure the catalyst temperature in situ in the plasma and (b) the absence of addi-tional elements in the discharge zone (e.g. thermocouple) that could affect the electromagnetic field within the DBD reactor. This method is also easier to implement as compared to previ-ously reported synchrotron-based temperature-measurement technique [19].

2.3. Catalyst preparation and characterization

The preparation and basic physico-chemical properties of the 10% CoCZ catalyst can be found elsewhere [24]. A commer-cial CZ support (Sigma-Aldrich, surface area 54 m2g−1) was used to prepare the CoCZ (10 wt.% cobalt) catalyst by wet impregnation using a hexaamminecobalt (III) [Co(NH3)6]3+ complex in a strong electrostatic adsorption mode.

CO chemisorption measurements were carried out using a Micromeritics ASAP 2010C instrument. Before measurement, the sample was dried in vacuum at 110◦C. Sample was sub-sequently heated in flowing H2 with a rate of 10◦C min−1 to the final reduction temperature of 500 ◦C. A reduction time of 4 h was used, after which the samples were evacu-ated for 60 min at 520◦C. The CO adsorption isotherms were measured at 30◦C. Particle size estimations are based on the assumption of hemispherical geometry, assuming a CO/Cos adsorption stoichiometry of 1.5 (Cosreferring to metallic Co surface sites). The cobalt particle size was 10 nm as determ-ined by CO chemisorption after reduction in H2at 500◦C.

2.4. Tubular DBD reactor

The catalytic activity measurements were carried out in a tubu-lar DBD reactor used in the previous work [6]. The outer dia-meter of the reactor was 14 mm and the inner diadia-meter 10 mm. The discharge zone in the reactor was 80 mm long and the dis-charge gap was 3 mm. A stainless-steel mesh wrapped around the reactor served as the ground electrode. The inner electrode was a stainless steel rod with a diameter of 4 mm. Plasma para-meters were monitored by an oscilloscope (Tektronix TDS 2024C) equipped with a high-voltage probe (PVM-2, North Star) and a current probe (Model 6600, Pearson Electronics). A volume of 5 ml catalyst (125–250 µm fraction) was packed in the discharge region and held by quartz wool. External heat-ing of 150 ◦C was used during experiments in DBD reactor in order to avoid water condensation. The effluent gas was analyzed by online gas chromatography using Shimadzu GC-2014 instrument equipped with TCD and FID detectors. 3. Results and discussion

The developed DBD plasma DRIFTS cell was applied to study the mechanism of CO2hydrogenation by a 10% CoCZ catalyst, for which a synergetic plasma-catalytic effect was observed in a larger tubular DBD plasma reactor (figure7).

First, in order to distinguish the temperature- and plasma-induced effects in the DRIFTS cell, UV–Vis temperature measurements using rutile as a thermochromic material were carried out in a CO2flow without external heating (figure8(a)). The catalyst holder was filled with pure rutile powder slightly pressed into the holder. The measurements demonstrated a slight increase of the absorbance at 280 nm compared to the spectra measured upon conventional heating. This absorp-tion feature is assigned to ozone formaabsorp-tion [25]. The form-ation of ozone by plasma-induced CO2 dissociation in the absence of hydrogen has been reported before [26]. The absorption feature at 420 nm decreased only slightly during 4

(7)

Figure 7. Plasma-catalytic CO2hydrogenation in DBD reactor in

the presence of 10% CoCZ (external heating—150◦C, 0–1000 Hz, 90 µs pulses, 18 kV, 18 mJ per pulse, CO2/H2= 1/4, total flow

25 ml min−1).

Figure 8. Temperature measurements of the DRIFTS cell using the UV–Vis method with TiO2as the thermochromic material (a) and

using an IR temperature sensor (b) in a CO2flow (5 ml min−1) in

the presence of plasma (1000 Hz, 50 µs pulses, 8 kV).

plasma operation, pointing to a small temperature differ-ence. A nearly negligible increase of the temperature inside the plasma reactor was confirmed by the IR temperature sensor (temperature 31◦C), while the dielectric plate remained unheated (figure8(b)). Therefore, the spectroscopic changes observed in our operando plasma DRIFTS cell can be exclus-ively attributed to plasma-induced phenomena and there will be no influence of overheating of the catalyst bed and the DBD reactor on the catalytic performance.

For referencing purposes, KBr powder was first exposed to a plasma under CO2 hydrogenation conditions (figure9(a)). The plasma was switched on at room temperature and the dif-ference spectra showed that upon plasma exposure CO2was converted into CO, which is typical for plasma-induced non-catalytic CO2 hydrogenation [6]. Other regions of the spec-tra remained unchanged. When KBr was replaced by the 10% CoCZ passivated sample (reduction at 500◦C, passivation in

1% O2in He at room temperature), which was first re-reduced at 250 ◦C, consumption of CO2 and formation of a larger amount of CO was observed upon exposure to plasma. Dur-ing conventional thermocatalytic CO2hydrogenation over the same 10% CoCZ catalyst, surface carbonyls were observed on the metallic cobalt nanoparticles [27]. These carbonyl bands were not present during the plasma-catalytic measurements (figure 9(b)). The absence of carbonyl groups upon plasma exposure could be related to facile CO dissociation on the cata-lyst surface or plasma-induced CO desorption [6].

After the plasma was switched off, a broad feature appeared at 2046 cm−1, which is due to CO adsorbed on metallic cobalt. Likely conversion of CO (formed in the gas phase during CO2 dissociation) on the cobalt surface was immedi-ately terminated when plasma was switched off, while resid-ual CO molecules still present in the gas phase gave rise to the observed surface carbonyls. Gas-phase CO was fully purged after 2 min after the plasma was switched off.

Activation of CO2 can also take place on interfacial sites or the CZ support, leading to the formation of formate spe-cies (HCOO) [28,29]. There were no changes in the intens-ity of formate species in CH region of the IR spectra (2800– 3000 cm−1), indicating that formate species are not involved in the conversion of CO2. Sharp features at 3150 cm−1observed in plasma likely represent an artifact of the plasma discharges rather than surface adsorbed species. The location of this arti-fact was demonstrated to be dependent on the high voltage fre-quency and the interferogram sampling frefre-quency [30].

In our previous work [6], it was demonstrated by isotope exchange that, upon plasma exposure at room temperature, the formation of methane involves the dissociation of CO2to CO in the gas phase, followed by hydrogenation of CO to CH4 on the metallic cobalt surface. In turn, CO2 pre-adsorbed on the CZ support in the form of carbonate (CO32−) and form-ate species was not involved in the catalytic cycle, unless the temperature of the sample was increased substantially. The obtained operando DRIFTS results are in line with these pre-vious conclusions, demonstrating that plasma-assisted CO2 hydrogenation at low temperature follows the carbonyl path-way. Clearly, the formate pathway requires a higher reaction temperature (figure 9(c)). The occurrence of plasma-assisted CO/CO2hydrogenation on the cobalt surface at room temper-ature is a remarkable result, which cannot be obtained in con-ventional thermal catalysis. We attribute this to high-energy electrons, the electric field, surface charge and radicals on the surface during the plasma exposure [2,31,32]. It has been speculated before that electron-induced dissociation of the C– O bond can occur [33]. This would promote CO/CO2 hydro-genation at the catalyst surface rather than in the gas phase. We should mention here that in actual DBD reactors, where a higher energy density is applied, overheating is usually more substantial than in the plasma DRIFTS cell. Such overheating can accelerate plasma-assisted formation of methane via the formate route as well.

The CZ support and Co–CZ interface plays an important role in CO2 hydrogenation [24]. Strong metal-support inter-actions and facile oxygen vacancy formation in CoCZ lead

(8)

Figure 9. Plasma CO2hydrogenation with (a) a reference KBr and (b) 10% CoCZ catalyst followed by IR spectroscopy in the absence and

presence of plasma (1000 Hz, 50 µs pulses, 8 kV, CO2/H2= 1/4, total flow 12.5 ml min−1); (c) schematic comparison of carbonyl (green)

and formate (blue) mechanisms.

to higher CO2 hydrogenation activity via the formate path-way in which generation and healing of oxygen vacancies play an important role. Therefore, the redox properties of the 10% CoCZ catalyst and CZ support were also studied in a plasma using in situ IR and UV–Vis spectroscopy and compared to the temperature-dependent evolution of oxygen vacancy form-ation (figures10–12). The presence of oxygen vacancies can be probed by the broad IR feature at 2115–2140 cm−1due to the2_F

5/2→ 2F7/2electronic transition of Ce3+[34].

Exposure of the 10% CoCZ sample to a hydrogen plasma for 30 min at room temperature resulted in the formation of oxygen vacancies at the ceria surface (figure10). The form-ation of oxygen vacancies in samples that were calcined at 400 ◦C to remove most of the carbonates occurred at tem-peratures higher than 150◦C in hydrogen [24]. In the plasma DRIFTS cell, such carbonates cannot be removed due to tem-perature limitations (250◦C). The presence of carbonates on the surface hampers oxygen vacancy formation (figure 11). Nonetheless, plasma-assisted oxygen vacancy formation can still be observed at room temperature for the non-calcined sample as shown in figure10.

Oxygen vacancies are not formed on the CZ support without cobalt upon plasma exposure. However, UV–Vis spec-troscopy showed that a hydrogen plasma treatment affected the support without cobalt (figure 12). For these measurements, the bare CZ support was investigated due to the high absorb-ance of cobalt in the UV–Vis region. For the bare CZ, a new absorption feature around 450 nm developed during the hydro-gen plasma treatment. This phenomenon has been observed before for ceria exposed to hydrogen peroxide, leading to the formation of surface radicals [35,36]. It is therefore likely that hydroxyl radicals are formed on the CZ surface in a hydrogen DBD plasma.

It has been speculated that OH radicals can be an important intermediate in plasma-assisted CO2 hydrogenation [37–39]. OH radicals can react with CO and form COOH, followed by dissociation into H∗and CO2. High coverage of OH radicals can be beneficial for the water formation [38]. OH radicals can also be a precursor for methanol formation [39]. Methanol was not observed in the presence of CoCZ. Nevertheless, the herein developed characterization approach can be useful for the ana-lysis of catalysts active in methanol formation in plasma.

(9)

Figure 10. Redox behavior of CoCZ upon hydrogen plasma exposure for 30 min at room temperature followed by IR spectroscopy (1000 Hz, 50 µs pulses, 8 kV, H2flow 5 ml min−1).

Figure 11. Temperature-dependent evolution of oxygen vacancy formation in the absence of plasma for 10% CoCZ sample followed by IR spectroscopy. VOindicates oxygen vacancy.

To summarize, electron-induced CO/CO2dissociation can contribute to the unusual activity of 10% CoCZ in CO2 hydrogenation at low temperature in plasma. Combined with earlier results, we speculate that the dominant pathway of CO2 methanation is CO2 dissociation in the gas phase fol-lowed by plasma-promoted CO dissociation on metallic cobalt particles. The promotion is most likely due to electron impact on the catalyst. Facile formation of oxygen vacancies and hydroxyl radicals at the surface observed in hydrogen plasma can contribute to the formate mechanism at elevated temperatures.

Figure 12. Radical formation on CZ upon hydrogen plasma exposure for 30 min at room temperature followed by UV–Vis spectroscopy (1000 Hz, 50 µs pulses, 8 kV, H2flow 5 ml min−1).

4. Conclusions

An in situ/operando DRIFTS cell suitable to study catalytic materials under plasma condition was developed. The design is a modification of a Praying MantisTMDRIFTS cell, imple-menting a DBD configuration using quartz and PEEK mater-ials to prevent plasma discharges between the high-voltage electrode and the essential parts of the cell. The assembly was placed in a Faraday cage. The cell allows thermal and reductive pretreatment of the catalysts up to 250◦C prior to plasma-catalytic testing. Operando IR and UV–Vis spectro-scopic measurements during plasma exposure were demon-strated. The temperature of the powder catalyst was monitored as well. The new cell was employed to study the mechanism of plasma-induced CO2hydrogenation at the surface of a well-reduced 10% CoCZ catalyst.

Most profoundly, it was observed that cobalt nanoparticles on a CZ support can hydrogenate CO2 in plasma at ambi-ent conditions. The following observations were made: (a) ambient CO2 hydrogenation involves CO as an intermediate on metallic cobalt particles, (b) Ce3+ formation, pointing to ceria oxygen vacancy generation, in a hydrogen plasma, and (c) formation of radicals, likely hydroxyls. In situ temperature monitoring shows that these effects can be attributed to the plasma and not to overheating. The findings indicate that CO2 is activated in the gas phase yielding CO. After adsorption on the reduced cobalt surface, CO is dissociated and hydrogen-ated on the metal surface, which is only possible in the plasma at room temperature. Facile Ce3+and radicals formation might promote plasma-assited CO2methanation at elevated temper-ature via formate pathway.

The use of a modified DRIFTS cell allows for com-bined DRIFTS, UV–Vis and temperature measurements to investigate the nature of plasma-catalyst interactions, essential

(10)

to further development of efficient plasma-enhanced catalytic processes.

Data availability statement

All data that support the findings of this study are included within the article (and any supplementary files).

Acknowledgments

This research was supported by the Applied and Engineer-ing Sciences division of the Netherlands Organization for Sci-entific Research through the Alliander (now Qirion) Perspect-ive program on Plasma Conversion of CO2.

ORCID iDs

A Parastaevhttps://orcid.org/0000-0001-7135-7240

N Kosinovhttps://orcid.org/0000-0001-8520-4886

E J M Hensenhttps://orcid.org/0000-0002-9754-2417

References

[1] Whitehead J C 2016 Plasma–catalysis: the known knowns, the known unknowns and the unknown unknowns J. Phys. D: Appl. Phys.49 243001

[2] Neyts E C and Bogaerts A 2014 Understanding plasma catalysis through modelling and simulation—a review J. Phys. D: Appl. Phys.47 224010

[3] Van Laer K and Bogaerts A 2017 Influence of gap size and dielectric constant of the packing material on the plasma behaviour in a packed bed DBD reactor: a fluid modelling study Plasma Process. Polym.14 1600129

[4] Kim -H-H, Ogata A, Schiorlin M, Marotta E and Paradisi C 2011 Oxygen isotope (18O2) evidence on the role of oxygen

in the plasma-driven catalysis of VOC oxidation Catal. Lett. 141 277–82

[5] Roland U, Holzer F, Pöppl A and Kopinke F-D 2005 Combination of non-thermal plasma and heterogeneous catalysis for oxidation of volatile organic compounds: part 3. electron paramagnetic resonance (EPR) studies of plasma-treated porous alumina Appl. Catal. B58 227–34 [6] Parastaev A, Hoeben W F L M, Van Heesch B E J M,

Kosinov N and Hensen J M 2018 Temperature-programmed plasma surface reaction: an approach to determine

plasma-catalytic performance Appl. Catal. B239 168–77 [7] Sivachandiran L, Thevenet F and Rousseau A 2015

Isopropanol removal using MnxOypacked bed non-thermal plasma reactor: comparison between continuous treatment and sequential sorption/regeneration Chem. Eng. J. 270 327–35

[8] Sivachandiran L, Thevenet F, Gravejat P and Rousseau A 2013 Isopropanol saturated TiO2surface regeneration by

non-thermal plasma: influence of air relative humidity Chem. Eng. J.214 17–26

[9] Sivachandiran L, Thevenet F and Rousseau A 2014

Regeneration of isopropyl alcohol saturated MnxOysurface: comparison of thermal, ozonolysis and non-thermal plasma treatments Chem. Eng. J.246 184–95

[10] Song Y-H, Kim S-J, Choi K-I and Yamamoto T 2002 Effects of adsorption and temperature on a nonthermal plasma process for removing VOCs J. Electrostat.55 189–201 [11] Somers W, Bogaerts A, Van Duin A C T and Neyts E C 2012

Plasma species interacting with nickel surfaces: toward an atomic scale understanding of plasma-catalysis J. Phys. Chem. C116 20958–65

[12] Stere C E, Adress W, Burch R, Chansai S, Goguet A, Graham W G and Hardacre C 2015 Probing a non-thermal plasma activated heterogeneously catalyzed reaction using in situ DRIFTS-MS ACS Catal.5 956–64

[13] Jia Z and Rousseau A 2016 Sorbent track: quantitative monitoring of adsorbed VOCs under in-situ plasma exposure Sci. Rep.6 31888

[14] Jia Z, Wang X, Thevenet F and Rousseau A 2017 Dynamic probing of plasma-catalytic surface processes: oxidation of toluene on CeO2Plasma Process. Polym.14 e1600114

[15] Rivallan M, Fourré E, Aiello S, Tatibouët J-M and

Thibault-Starzyk F 2012 Insights into the mechanisms of isopropanol conversion on γ-Al2O3by dielectric barrier

discharge Plasma Process. Polym.9 850–4

[16] Rodrigues A, Tatibouët J M and Fourré E 2016 Operando DRIFT spectroscopy characterization of intermediate species on catalysts surface in VOC removal from air by non-thermal plasma assisted catalysis Plasma Chem. Plasma Process.36 901–15

[17] Azzolina-Jury F, Bento D, Henriques C and

Thibault-Starzyk F 2017 Chemical engineering aspects of plasma-assisted CO2hydrogenation over nickel zeolites

under partial vacuum J. CO2Util.22 97–109

[18] Azzolina-Jury F and Thibault-Starzyk F 2017 Mechanism of low pressure plasma-assisted CO2hydrogenation over

Ni-USY by microsecond time-resolved FTIR spectroscopy Top. Catal.60 17091721

[19] Gibson E K et al 2017 Probing the role of a non-thermal plasma (NTP) in the hybrid NTP catalytic oxidation of methane Angew. Chem. Int. Ed.56 9351–5

[20] Nozaki T, Hiroyuki T and Okazaki K 2006 Hydrogen enrichment of low-calorific fuels using barrier discharge enhanced Ni/γ-Al2O3bed reactor: thermal and nonthermal

effect of nonequilibrium plasma Energy Fuels20 339–45 [21] Trinh Q H, Lee S B and Mok Y S 2015 Removal of ethylene

from air stream by adsorption and plasma-catalytic oxidation using silver-based bimetallic catalysts supported on zeolite J. Hazard. Mater.285 525–34

[22] Kuznetsov V N, Ryabchuk V K, Emeline A V, Mikhaylov R V, Rudakova A V and Serpone N 2013 Thermo- and

photo-stimulated effects on the optical properties of rutile titania ceramic layers formed on titanium substrates Chem. Mater.25 170–7

[23] Toyoda T and Yabe M 1983 The temperature dependence of the refractive indices of SrTiO3and TiO2J. Phys. D: Appl.

Phys.16 L251–5

[24] Parastaev A, Muravev V, Huertas Osta E, van Hoof A J F, Kimpel T F, Kosinov N and Hensen E J M 2020 Boosting CO2hydrogenation via size-dependent metal–support

interactions in cobalt/ceria-based catalysts Nat. Catal. 3 526–33

[25] Aoyagi Y, Takeuchi M, Yoshida K, Kurouchi M, Araki T, Nanishi Y, Sugano H, Ahiko Y and Nakamura H 2012 High-sensitivity ozone sensing using 280 nm deep ultraviolet light-emitting diode for detection of natural hazard ozone J. Environ. Prot.3 695–9

[26] Xu S, Whitehead J C and Martin P A 2017 CO2conversion in a

non-thermal, barium titanate packed bed plasma reactor: the effect of dilution by Ar and N2Chem. Eng. J.327 764–73

[27] Parastaev A 2020 Insights into conventional and

plasma-assisted catalysis for carbon dioxide hydrogenation

(11)

PhD Thesis Technische Universiteit Eindhoven, Eindhoven, The Netherlands

[28] Wang F, He S, Chen H, Wang B, Zheng L, Wei M, Evans D G and Duan X 2016 Active site dependent reaction

mechanism over Ru/CeO2catalyst toward CO2methanation

J. Am. Chem. Soc.138 6298–305 [29] Aldana P A U, Ocampo F, Kobl K, Louis B,

Thibault-Starzyk F, Daturi M, Bazin P, Thomas S and Roger A C 2013 Catalytic CO2valorization into CH4on Ni-based

ceria-zirconia. Reaction mechanism by operando IR spectroscopy Catal. Today215 201–7

[30] Milaniak N, Audet P, Griffiths P R, Massines F and Laroche G 2020 Interpretation of artifacts in Fourier transform infrared spectra of atmospheric pressure dielectric barrier discharges: relationship with the plasma frequency between 300 Hz and 15 kHz J. Phys. D: Appl. Phys. 53 015201

[31] Holzer F, Roland U and Kopinke F-D 2002 Combination of non-thermal plasma and heterogeneous catalysis for oxidation of volatile organic compounds: part 1. accessibility of the intra-particle volume Appl. Catal. B 38 163–81

[32] Jafarzadeh A, Bal K M, Bogaerts A and Neyts E C 2020 Activation of CO2on copper surfaces: the synergy between

electric field, surface morphology, and excess electrons J. Phys. Chem. C124 6747–55

[33] Lee J, Sorescu D C and Deng X 2011 Electron-induced dissociation of CO2on TiO2(110) J. Am. Chem. Soc.

133 10066–9

[34] Binet C, Badri A and Lavalley J-C 1994 A spectroscopic characterization of the reduction of ceria from electronic transitions of intrinsic point defects J. Phys. Chem. 98 6392–8

[35] Gaynor J D, Karakoti A S, Inerbaev T, Sanghavi S,

Nachimuthu P, Shutthanandan V, Seal S and Thevuthasan S 2013 Enzyme-free detection of hydrogen peroxide from cerium oxide nanoparticles immobilized on

poly(4-vinylpyridine) self-assembled monolayers J. Mater. Chem. B1 3443–50

[36] Ornatska M, Sharpe E, Andreescu D and Andreescu S 2011 Paper bioassay based on ceria nanoparticles as colorimetric probes Anal. Chem.83 4273–80

[37] Liao Y, Zhong W, Qian M, Liu S, Zhang J and Wang D 2020 Numerical study on the reaction mechanism of CO2

hydrogenation in atmospheric-pressure dielectric barrier discharge J. Appl. Phys.128 233303

[38] Michiels R, Engelmann Y and Bogaerts A 2020 Plasma catalysis for CO2hydrogenation: unlocking new pathways

toward CH3OH J. Phys. Chem. C124 25859–72

[39] De Bie C, Van Dijk J and Bogaerts A 2016 CO2hydrogenation

in a dielectric barrier discharge plasma revealed J. Phys. Chem. C120 25210–24