Solar-Enhanced Plasma-Catalytic Oxidation of Toluene over a Bifunctional Graphene Fin Foam Decorated with Nanofin-like MnO2.

In this work, we propose a hybrid and unique process combining solar irradiation and post-plasma catalysis (PPC) for the effective oxidation of toluene over a highly active and stable MnO2/GFF (bifunctional graphene fin foam) catalyst. The bifunctional GFF, serving as both the catalyst support and light absorber, is decorated with MnO2 nanofins, forming a hierarchical fin-on-fin structure. The results show that the MnO2/GFF catalyst can effectively capture and convert renewable solar energy into heat (absorption of >95%), leading to a temperature rise (55.6 °C) of the catalyst bed under solar irradiation (1 sun, light intensity 1000 W m-2). The catalyst weight (9.8 mg) used in this work was significantly lower (10-100 times lower) than that used in previous studies (usually 100-1000 mg). Introducing solar energy into the typical PPC process via solar thermal conversion significantly enhances the conversion of toluene and CO2 selectivity by 36-63%, reaching ∼93% for toluene conversion and ∼83% for CO2 selectivity at a specific input energy of ∼350 J L-1, thus remarkably reducing the energy consumption of the plasma-catalytic gas cleaning process. The energy efficiency for toluene conversion in the solar-enhanced post-plasma catalytic (SEPPC) process reaches up to 12.7 g kWh-1, ∼57% higher than that using the PPC process without solar irradiation (8.1 g kWh-1), whereas the energy consumption of the SEPPC process is reduced by 35-52%. Moreover, the MnO2/GFF catalyst exhibits an excellent self-cleaning capability induced by solar irradiation, demonstrating a superior long-term catalytic stability of 72 h at 1 sun, significantly better than that reported in previous works. The prominent synergistic effect of solar irradiation and PPC with a synergistic capacity of ∼42% can be mainly attributed to the solar-induced thermal effect on the catalyst bed, boosting ozone decomposition (an almost triple enhancement from ∼0.18 gO3 g-1 h-1 for PPC to ∼0.52 gO3 g-1 h-1 for SEPPC) to generate more oxidative species (e.g., O radicals) and enhancing the catalytic oxidation on the catalyst surfaces, as well as the self-cleaning capacity of the catalyst at elevated temperatures driven by solar irradiation. This work opens a rational route to use abundant, renewable solar power to achieve high-performance and energy-efficient removal of volatile organic compounds.

Introduction

Air pollution remains a global environmental problem and has significant harmful effects on human health and the environment. Exposure to air pollution is estimated to have contributed to more than four million deaths in 2015 because of increased mortality and morbidity from cardiovascular and respiratory disease, as well as lung cancer.[1] Volatile organic compounds (VOCs) are an important source of air pollutants and play a significant role in the formation of photochemical smog and haze.[2−6] Plasma-catalysis, the combination of nonthermal plasma (NTP) and heterogeneous catalysis, has been demonstrated as a promising technology for the effective removal of VOCs in low concentrations at low temperatures.[7,8] In a plasma-catalysis system, catalysts can be packed inside the plasma zone (in-plasma catalysis, IPC) or placed downstream of the plasma zone (post-plasma catalysis, PPC).[9−11] The coupling of NTP with suitable catalysts can generate a plasma-catalysis synergy,[12−14] enhancing the conversion of VOCs, the selectivity to CO2, and the energy efficiency of the hybrid process, while reducing the formation of undesirable byproducts.[15,16] However, catalyst stability remains a key barrier in the development of a plasma-catalytic gas cleaning process for the removal of VOCs, especially when using IPC because of complex physicochemical interactions between plasma and the catalyst and limited available knowledge of designing highly active and stable catalysts effective in the IPC processes.[3,9] In contrast, PPC integrates the two relatively independent processes of NTP and well-developed catalytic oxidation, where plasma-generated ozone can be effectively converted to more active oxygen atoms on catalyst surfaces to achieve superior conversion of VOCs and CO2 selectivity, showing great potential to be used on a commercial scale.[17,18] However, further increasing the energy efficiency for VOC conversion is of primary importance to make the plasma-catalytic gas cleaning technology more attractive and economically competitive.

The overall energy efficiency of a PPC process can be manipulated by tuning either the plasma process, the catalytic process, or both.[17] Increasing the plasma energy input in a PPC process can enhance the conversion of VOCs while keeping the initial concentration of VOCs constant.[19−21] However, this usually results in a decreased energy efficiency of PPC processes. Catalysts have been demonstrated to play an important role in PPC processes. Thus, designing innovative, highly active, and stable catalytic materials as well as alternative efficient catalytic routes is critical to tackling these challenges. Nanomaterials have shown significant potential in heterogeneous catalysis for environmental and energy applications with superior catalytic performance compared to bulk catalysts because of their unique physical and chemical properties. However, limited efforts have been dedicated to the integration of NTP with catalytic nanomaterials for the removal of VOCs in plasma-catalytic processes. In addition to the judicious design of high-performance catalysts,[22−24] using external heating to elevate the temperature of the catalyst bed in PPC processes has been commonly used as an effective way to enhance catalyst activities for the oxidation of VOCs.[22,25−27] Increasing the temperature of the catalyst bed can facilitate the activation of lattice oxygen and promote ozone conversion on the catalyst surfaces, thus significantly enhancing catalytic oxidation of VOCs.[28−30] Rezaei et al. investigated the oxidation of toluene by ozone with a temperature-controlled MnO/γ-Al2O3 catalyst bed.[31−33] They found that efficient toluene conversion (>90%) and ozone conversion (∼100%) can be achieved by raising the temperature of the catalyst bed from room temperature to 80–100 °C. Compared to the catalyst bed without heating, heating the catalyst bed to >65 °C enhances the stability of the catalyst. However, using external heating to elevate the temperature of the catalyst bed to achieve the enhanced oxidation of VOCs consumes additional energy. Therefore, achieving high-performance removal of VOCs and ultra-low energy consumption simultaneously in PPC processes remains a significant challenge.

Solar energy, the largest renewable energy source, has attracted significant interest in emerging fields such as chemical synthesis, water splitting, CO2 conversion, solar thermal conversion, and so on.[34,35] It is, therefore, a rational strategy to exploit and use low-cost renewable solar energy to drive catalytic reactions. Solar thermal conversion is a promising approach to use solar energy, where the incident light is harvested by light absorbers and effectively converted into heat.[36] Therefore, combining renewable and sustainable solar power with a PPC process via solar thermal conversion has significant potential to open a new route for the solar-driven effective catalytic conversion of VOCs with low energy costs.

The rational design of the catalysts is crucial for efficiently utilizing solar irradiation to drive catalytic reactions. Graphene has been widely used as a catalyst support.[37−39] Graphene fin foam (GFF) with a three-dimensional (3D) interconnected architecture has been investigated for a range of energy applications including energy storage and solar-steam generation.[40,41] The unique advantages of graphene, including inherent broad spectral absorption, high thermal conductivity, and excellent solar thermal conversion capacity, make this material an ideal light absorber.[42] In addition, GFF has a non-agglomerated interconnected morphology, an enlarged accessible surface area, and sharp exposed edges, all of which are favorable in enhancing catalyst dispersion and catalytic activity.[43] MnO2 has been demonstrated as an efficient catalyst in PPC processes as it can effectively convert ozone into active oxygen species.[44] Therefore, the integration of GFF and MnO2 has excellent potential to achieve remarkable solar thermal conversions and efficient plasma catalytic reactions.

In this work, we propose for the first time a novel solar-enhanced post-plasma catalytic (SEPPC) process for the oxidation of toluene, a model VOC, over a nanofin-like MnO2 catalyst decorated on a bifunctional GFF in a coaxial dielectric barrier discharge (DBD) plasma reactor. GFF, as both a catalyst support and a light absorber, was synthesized using plasma-enhanced chemical vapor deposition (PECVD). Nanofin-like MnO2 was then decorated on GFF as a high-performance catalyst via a facile redox reaction, generating a hierarchical fin-on-fin structure. To the best of our knowledge, such a novel catalytic nanomaterial with a hierarchical fin-on-fin structure has not been investigated in plasma-catalytic processes before. The influence of the catalyst and solar illumination (1 sun) on the oxidation of toluene has been investigated in terms of the conversion of toluene, CO2 selectivity, ozone conversion, and energy efficiency. Moreover, excellent catalyst stability toward toluene conversion and CO2 selectivity has been demonstrated. The contributions of the PPC, solar thermal catalysis (STC), and their synergistic effect to the SEPPC oxidation of toluene have been discussed.

Experimental Section

Synthesis of GFF

GFF was prepared using the PECVD method in a temperature-controlled radio-frequency inductively coupled plasma (ICP) system. Nickel powders with an average particle size of 5 μm were spread on a quartz slide (120 mm × 40 mm) and placed in the plasma chamber (quartz tube). The chamber was pumped to a low pressure of 3 Pa before the synthesis. When the plasma system was switched off, the sample (Ni powders) was heated to 900 °C and cooled down to 700 °C. The ICP system was then switched on, generating plasma in the quartz tube using mixed hydrogen (5 mL min–1) and methane (5 mL min–1) at a constant pressure of 28 Pa and a fixed power of 250 W for the growth of graphene fins (GF). After the synthesis, the plasma chamber was cooled to room temperature under vacuum conditions. The obtained GF/Ni foam was functionalized by 240 ppm moist ozone (humidified by passing through a water bubble bottle) in air (250 mL min–1) for 5 min. After dissolving Ni scaffolds with a 3 M HCl solution at 80 °C for 12 h, the samples were freeze-dried for 8 h.

MnO2 Deposition on GFF

MnO2 was loaded on GFF by the facile redox reaction between graphene and KMnO4. GFF was immersed into a KMnO4 solution (1067 mg of KMnO4 dissolved in 160 mL of deionized water) at 80 °C for 18 h in an oven. Afterward, the residual KMnO4 was removed by washing the sample several times using deionized water. The freeze-dried sample was annealed at 400 °C for 4 h under vacuum conditions to obtain MnO2/GFF.

Synthesis of MnO2

Pure MnO2 was synthesized via the reduction of KMnO4 using a similar method for the synthesis of MnO2/GFF. Sulfuric acid (2 mL, 98 wt %) was slowly dropped into a KMnO4 solution (1067 mg of KMnO4 dissolved in 160 mL of deionized water). Cu foil (8 × 8 cm2) was then immersed in the KMnO4 solution and heated at 80 °C for 18 h in an oven. After that, the precipitate was filtered, washed using deionized water, and dried at 80 °C to obtain pure MnO2.

Characterization

The morphologies of GFF and MnO2/GFF were analyzed using a SU-70 scanning electron microscope (SEM, Hitachi). The elemental distribution (Mn, O, and C) was investigated by the energy-dispersive spectroscopy (EDS, Oxford Instruments) technique. The dispersion of MnO2 on GFF and the corresponding elemental mapping were determined by using transmission electron microscopy (TEM, JEM-2100; JEOL, Japan) equipped with an Oxford Instruments X-MAX EDS system. The crystal structures of GFF and MnO2/GFF were characterized by X-ray diffraction (XRD) using a PANalytical X’Pert diffractometer equipped with Ni β-filtered Cu Kα radiation. The surface chemical states of GFF and MnO2/GFF were measured by X-ray photoelectron spectroscopy (XPS, VG Escalab Mark II) with a monochromatic Mg Kα X-ray source (hν = 1253.6 eV, West Sussex). The binding energies of the spectra were calibrated with the C 1s photoemission peak at 284.6 eV. Raman spectra of both samples were recorded using a LabRAM HR Evolution Raman spectrometer (Horiba Jobin Yvon). Infrared spectra of GFF and MnO2/GFF were measured using a Fourier transform infrared (FTIR) spectrometer (Thermo Fisher, Nicolet 5700). Diffuse reflectance UV–vis–infrared absorption of the samples was performed on a spectrophotometer (Shimadzu UV-3150). An infrared (IR) thermographic camera (FLIR ThermaCAM S65, USA) was used to record the IR images of GFF and MnO2/GFF under simulated solar irradiation. The window of the catalytic reactor was removed for the IR measurements. The weight loading of Mn in MnO2/GFF was 17.4 wt %, measured by inductively coupled plasma–mass spectrometry (XSENIES, Thermo Electron Corporation). The hydrogen temperature-programmed reduction (H2-TPR) analysis of the catalyst was carried out using a chemisorption analyzer (Micrometrics, Autochem II 2920). The Brunauer–Emmett–Teller (BET) surface area of the samples was determined via N2 adsorption–desorption measurements at −196 °C (Micromeritics TriStar II). Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) analysis of the fresh (before reaction) and spent catalysts after the SEPPC and PPC reactions was carried out on an FTIR spectrometer (Thermo Fisher, Nicolet 460) equipped with a deuterated triglycine sulfate detector. The organic intermediates deposited on the catalyst surfaces (PPC reaction for 8 h) were extracted using a methanol solution under ultrasonic vibration for 1 h and then analyzed by gas chromatography–mass spectrometry (GC–MS, QP2010SE, Shimadzu).

Experimental Setup

Figure shows the schematic diagram of the SEPPC system for gas clean-up, consisting of a typical post-plasma catalytic reactor coupled with a solar irradiation unit for the catalyst bed. A 2 mm-long stainless-steel mesh was wrapped around a quartz tube (external diameter 7 mm and inner diameter 5 mm) and used as a ground electrode, whereas a stainless-steel rod with a diameter of 4 mm was placed in the quartz tube, acting as the high-voltage electrode. The DBD reactor was connected to a high voltage alternating-current (ac) power supply (CTP-2000K, Suman Plasma Technology). Air (zero grade, Gingergas) was used as the carrier gas with a fixed flow rate of 250 mL min–1. A high-resolution syringe pump (LSP01-1BH, Longer Precision Pump) was used to inject liquid toluene into a toluene generator (FD-PG, Friend Laboratory Equipment) to form a toluene vapor of 260 ppm in air. The electrical signals of the DBD were monitored using a digital oscilloscope (MDO 3034, Tektronix). A solar-driven catalytic reactor with a quartz window was placed downstream of the DBD reactor. To reduce heat loss during the solar irradiation, a piece of the MnO2/GFF catalyst (diameter of 35 mm and weight of 9.8 mg) was covered on thermal insulation and was placed on a porous tray in the solar-driven catalytic reactor. For comparison, the catalytic activities of pure MnO2 (powder, 9.8 mg) and GFF (35 mm in diameter) were also investigated. Instead of using conventional heating for the catalyst bed, renewable solar power was used to thermally activate the catalyst bed in this work. A xenon lamp (Microsolar300A, Perfectlight Technology) was placed over the catalytic reactor and used as a light source to produce an AM1.5G (air mass 1.5, global) spectrum between 320 and 2500 nm. The light intensity was maintained at 1000 W m–2 (corresponding to 1 sun) by controlling the light current (12 A) and the distance between the xenon lamp and the catalyst bed (37.3 cm). The oxidation of toluene was investigated under different process modes: NTP-only, PPC (without solar irradiation), SEPPC, and catalytic oxidation with and without solar irradiation. The plasma oxidation of toluene (NTP-only) was carried out at room temperature and atmospheric pressure in the DBD reactor, whereas no external heating was used in the PPC without the use of solar irradiation. The solar-enhanced PPC experiment was tested under the solar irradiation (1 sun). In addition, a catalyst stability test of 72 h for the PPC and SEPPC processes was carried out at a constant specific input energy (SIE) of 364 J L–1. Catalytic oxidation of toluene in air, in both the dark and under sun radiation (1 sun) was also carried out with plasma off to better understand the role of ozone and the solar-induced temperature effect on this reaction. The operating conditions (catalyst amount, toluene concentration, and total gas flow) used in this reaction were kept the same as those used in NTP-only, PPC, and SEPPC processes. In addition, no ozone was involved in the catalytic oxidation of toluene.

Figure 1

Schematic diagram of the SEPPC system.

The concentrations of toluene and CO2 were measured using gas chromatography (GC9790Plus, Fuli Instruments) equipped with two flame ionization detectors and a reformer furnace. Ozone was analyzed with an ozone monitor (Models 106-MH, 2B Technology) before and after the solar thermal catalytic reactor to understand the contributions of the catalyst to the conversion of ozone. The formation of byproducts was monitored using a portable FTIR gas analyzer (Gasmet Dx4000, Finland). The outlet gas was fed into the Gasmet gas cell. The FTIR spectra of the gas phase byproducts were recorded in the wavenumber range of 600–4250 cm–1.

The regeneration of deactivated catalysts with ozone under solar irradiation was investigated. In this test, a fresh MnO2/GFF catalyst was used for toluene oxidation in the PPC process without solar irradiation (dark) for 2 h, followed by an oxidative regeneration with ozone (250 ppm) under solar irradiation (1 sun) for 1 h. The ozone was generated by NTP using the same DBD reactor. The flow rate of air was maintained at 250 mL min–1 in the process.

The SIE is defined as the energy dissipated per unit volume of the gas flow aswhere the discharge power (P) of the DBD reactor is determined using the Lissajous method,[45] and Q is the total gas flow rate.

The ozone conversion (ηO), toluene conversion (ηC), CO2 selectivity (SCO), CO selectivity (SCO), carbon balance, energy efficiency (ηEY), ozone decomposition capacity (ODC), and synergistic capacity (SC) of the SEPPC process are defined aswhere [O3]0 is the concentration of ozone measured before the catalyst bed; [O3]out is the concentration of ozone at the outlet of the catalyst bed; [C7H8]in and [C7H8]out are the inlet and outlet concentrations of toluene, respectively; [CO2]out and [CO]out are the concentrations of generated CO2 and CO, respectively; M and MO are the molar mass of toluene and ozone, respectively; ηSEPPC, ηPPC, and ηSTC are the toluene conversion of SEPPC, PPC, and STC processes, respectively.

Results and Discussion

Characterization of the MnO2/GFF Catalyst

Figure a illustrates the hierarchical fin-on-fin structure of the MnO2/GFF catalyst. Bifunctional GFF, serving as the catalyst support and light absorber, can efficiently trap the incident light and convert the absorbed solar energy to heat, thus elevating the temperature of the catalytic reaction. Meanwhile, the nanofin-like MnO2 catalyst can effectively turn ozone into active oxygen species, thus enhancing the oxidation of toluene on the catalytic surface.

Figure 2

(a) Schematic illustration of a hierarchical fin-on-fin catalyst for light harvest and toluene oxidation. (b–d) SEM images of the MnO2/GFF catalyst.

Figure b–d shows the surface morphologies of GFF and MnO2/GFF. GFF presents a 3D interconnected macroporous structure, which is beneficial for gas transport in the oxidation of toluene (Figure b). A close-up view in Figure c indicates that GFF is composed of dense fins with exposed sharp edges, in a vertical orientation, and with a non-agglomerated morphology. Because of the multiple reflections in the dense fins, GFF can trap and almost fully absorb the incident light, effectively converting the absorbed solar energy into heat.[40,41] In addition, GFF provides an abundance of sites for uniform dispersion of the nanofin-like MnO2 on its surface, forming a hierarchical fin-on-fin structure (Figure d). The open channels and non-agglomerated morphology of MnO2 nanofins facilitate gas diffusion and enlarge the accessible surface area in the catalytic reaction.[46] As shown in Figure S1 (Supporting Information), the element mapping of the catalyst indicates the presence of a graphene framework and a homogeneous distribution of MnO2 on GFF. The photos of the samples in the different steps of the synthesis can be found in Figure S2 (Supporting Information). Compared to the pristine nickel foam, GFF and MnO2/GFF are black and exhibit an excellent absorption within visible wavelengths. Both samples retain a freestanding and monolithic structure after HCl etching and MnO2 deposition, which can be ascribed to the formation of a densely interconnected network of GFF. In addition, the size of GFF and MnO2/GFF can be extended by using a larger nickel foam as a template for harvesting more solar energy in practical applications.

Figure a,b presents the surface chemical states of MnO2, GFF, and MnO2/GFF. The full-range XPS analysis of GFF shows the presence of carbon in GFF, whereas the existence of manganese oxide after the deposition of MnO2 onto GFF can be observed (Figure a). The XPS spectrum of pure MnO2 is similar to that of MnO2/GFF. A spin-energy separation of 11.8 eV between Mn 2p1/2 and Mn 2p3/2 further confirms the presence of MnO2 in MnO2/GFF (Figure b).[47] The average oxidation state (AOS) of MnO2 and MnO2/GFF was estimated based on the Mn 3s spectra using the equation: AOS = 8.956 – 1.126 × ΔEs, where ΔEs represents the difference of peak energies between the doublet Mn 3s peaks.[48] The AOS of MnO2 and MnO2/GFF was 3.60 and 3.21, respectively (Figure S3, Supporting Information), suggesting that both MnO2 and MnO2/GFF have an intermediate oxidation state between Mn4+ and Mn3+. The loading of MnO2 on GFF can reduce the average oxidation state of Mn, which could facilitate the conversion of ozone to enhance the oxidation of toluene in the PPC process.[32] The binding energy of the Mn 2p3/2 peak in the MnO2/GFF catalyst shifts to a lower value (639.6 eV) compared to that of the as-prepared MnO2 (641.8 eV), which can be attributed to the incorporation of MnO2 into the graphene resulting in an increased chargeability of Mn atoms.[49,50] The downshift phenomenon can be ascribed to an increased π electron cloud density because of the introduction of graphene. The hybridization between graphene and MnO2 at the interface reduced the electronic localization and extended the π conjugated systems. As a result, the interfacial charge transfer from graphene promoted the chargeability of Mn atoms. The XRD patterns of both samples exhibit a dominant characteristic peak at 26°, which correlates with the (002) diffraction of graphitic carbon in GFF and MnO2/GFF (Figure c),[51,52] whereas the two peaks at 2θ = 36 and 41° in the XRD of MnO2/GFF are associated with the (400) and (420) lattice planes of standard α-MnO2 (JCPDS 44-0141), respectively.[53,54] The magnified XRD patterns of the samples at 30–50° are presented in Figure S4 (Supporting Information). Moreover, the broad and weak MnO2 peaks present in the XRD pattern of MnO2/GFF suggest the formation of a low crystallinity of MnO2, resulting in high dispersion of MnO2 on GFF,[55] which can be further validated by the TEM images of MnO2/GFF. In addition, the characteristic peaks of MnO2 can be indexed to standard α-MnO2 (JCPDS 44-0141), indicating that the as-prepared MnO2 has high purity and crystallinity. As shown in Figure S5 (Supporting Information), MnO2 nanoparticles with a diameter of 10–15 nm are uniformly dispersed on the surface of GFF. The lattice spacings of 0.248 and 0.214 nm for MnO2, correspond to the (400) and (420) lattice planes of standard α-MnO2 (Figure S5c,d, Supporting Information), respectively, which agrees well with the XRD results. As shown in Figure S6 (Supporting Information), the elemental mapping has demonstrated the presence of well-defined dispersion of Mn and O on the graphene layer, which is consistent with the distribution of nanoparticles. Figure d shows the structural and electronic properties of both samples using Raman spectra. The peaks at ∼630 cm–1 in the spectra of MnO2 and MnO2/GFF can be ascribed to Mn–O lattice vibration in the [MnO6] octahedral.[56,57] The H2-TPR profile of MnO2/GFF is given in Figure S7 (Supporting Information). Two peaks of H2 consumption at 296 and 353 °C are associated with the reduction of MnO2 to Mn3O4 and the reduction of Mn3O4 to MnO, respectively.[58] The presence of low reduction temperatures in the TPR profile of MnO2/GFF indicates the low temperature reducibility of the catalyst and the formation of high-mobility oxygen species on the catalyst surfaces, as confirmed in previous works.[59−61] The BET surface area of GFF and MnO2/GFF was 29.2 and 26.7 m2 g–1, respectively.

Figure 3

Characterization of MnO2, GFF, and MnO2/GFF. (a) XPS survey spectra, (b) Mn 2p spectra, (c) XRD patterns, and (d) Raman spectra.

The light absorption ability of MnO2, GFF, and MnO2/GFF was evaluated using UV–vis–infrared analysis. As shown in Figure a, the absorption of MnO2 significantly decreases when the spectrum changes from the UV and visible region to the infrared region, indicating that pure MnO2 does not have an efficient light absorption capability (≤60%) in the entire solar spectrum. By contrast, both GFF and MnO2/GFF show an ultrahigh absorption of >95% across the whole solar spectrum, which significantly improves the light utilization efficiency, especially in the infrared region. Figure b shows the IR images of MnO2, GFF, and MnO2/GFF under solar irradiation (1 sun). The steady-state surface temperatures of MnO2, GFF, and MnO2/GFF were 57.6, 71.0, and 72.6 °C, respectively, demonstrating that both GFF and MnO2/GFF have a high solar-to-thermal conversion efficiency (13.4–15 °C higher than that of MnO2). Moreover, the time-dependent variation of temperature under solar illumination reveals that MnO2/GFF exhibits a faster temperature, and an increased response compared with pure MnO2 (Figure S8, Supporting Information). Thus, GFF not only traps and converts the solar irradiation but also transfers heat from the graphene support to MnO2 because of its excellent thermal conductivity. The maximum surface temperature change (54.0–55.6 °C at 1 sun) is more significant than that reported in previous works using graphene foam-based nanomaterials (33.5–39 °C at 1 sun).[40,41] The solar-thermal conversion efficiency ηst was estimated to be ∼62.2% based on the energy balance between the converted thermal energy Q̇in and the heat dissipation Q̇surr (see the Supporting Information & Figure S8 for details).

Figure 4

(a) UV–vis–infrared spectra and (b) IR images of MnO2, GFF, and MnO2/GFF at 1 sun.

Toluene Oxidation under Different Process Modes

Figure shows the influence of SIE on the oxidation of toluene using different processes (NTP-only, PPC, and SEPPC). Increasing the SIE from 144 to 382 J L–1 enhances the conversion of toluene and CO2 selectivity but decreases the energy efficiency for toluene conversion in all three processes. At a fixed SIE, the performance of the PPC process is better than that using NTP-only, whereas the combination of PPC oxidation with solar irradiation (1 sun) further enhances the conversion of toluene and CO2 selectivity, as well as the energy efficiency of the hybrid process. For instance, compared to the PPC process, the conversion of toluene and CO2 selectivity in the SEPPC process were enhanced by 63 and 36%, respectively, reaching up to ∼93 and ∼83% at an SIE of ∼350 J L–1.

Figure 5

Comparison of the catalytic performance and energy efficiency in the NTP-only, PPC, and SEPPC processes: (a) toluene conversion, (b) CO2 selectivity, and (c) energy efficiency.

Compared to the NTP-only and PPC processes, the coupling of solar radiation with the PPC process shifts the SIE window. As shown in Figure a,b, the SIE required to achieve the same toluene conversion and CO2 selectivity in the SEPPC process is about 52 and 35% lower than that of PPC and NTP-only processes, respectively, suggesting that low energy consumption can be achieved in the SEPPC process. This finding can also be confirmed by the significantly enhanced energy efficiency in Figure c. Upon solar illumination, the energy efficiency of the SEPPC process is increased by ∼57% at an SIE of 152 J L–1 in comparison to the PPC process, achieving a superior efficiency of 12.7 g kWh–1 for toluene conversion, which is among the best in the state-of-the-art studies on plasma oxidation of toluene with and without a catalyst.[18−21]

The enhanced energy efficiency achieved in the SEPPC oxidation of toluene over the MnO2/GFF catalyst is mainly attributed to the remarkable solar thermal effect on the catalyst. First, the elevated temperature of the catalyst bed because of the heating effect resulted from the solar irradiation promotes the conversion of ozone to active oxygen species including more active O radicals over the MnO2/GFF catalyst (Figure a), which can be confirmed by the decreased ozone concentration measured at the outlet of the SEPPC process (Figure b). Although the catalyst weight (9.8 mg only) used in this work is 10–100 times lower than that used in most of the previous works,[44,62,63] the maximum O3 decomposition capacity over the MnO2/GFF catalyst in the PPC process (without solar radiation) reaches up to ∼0.18 gO g–1 h–1. Moreover, the conversion of ozone (maximum O3 decomposition capacity) in the SEPPC process is ∼0.52 gO g–1 h–1, almost triple that of the PPC without the solar radiation counterpart operated at room temperature, which is significantly higher than that in a previous work (∼0.30 gO g–1 h–1).[20] The active oxygen species formed on the surface of MnO2/GFF, especially O radicals, have higher oxidative activity compared to ozone,[18] which significantly contributes to the enhanced oxidation of toluene and the reaction intermediates, resulting in higher CO2 selectivity, higher energy efficiency, and lower CO selectivity (Figure S9a, Supporting Information). Moreover, the carbon balance of the SEPPC process is much higher than that of the PPC process, indicating that the coupling of solar illumination with the PPC process substantially reduced the formation of organic byproducts (Figure S9b, Supporting Information).[64] Note that the concentration of residue ozone after the SEPPC process is significantly lower than that of the PPC process without solar irradiation and the NTP-only process because of superior ODC over the MnO2/GFF catalyst assisted by solar irradiation (Figure b).

Figure 6

Catalytic performance of the MnO2/GFF catalyst: (a) catalytic conversion of ozone in PPC and SEPPC processes; (b) ozone concentration in NTP-only, PPC, and SEPPC processes, and (c) catalytic conversion of toluene in the dark and the STC process without NTP.

To get new insights into the contributions of the solar thermal effect and ozone to the SEPPC oxidation of toluene, the catalytic oxidation of toluene in air, with and without solar radiation, was carried out in the absence of ozone. Figure c shows that the MnO2/GFF catalyst has no catalytic activity toward toluene oxidation without solar irradiation (dark). Upon solar irradiation, the elevated temperature of the catalyst bed induced by the solar thermal effect can thermally activate the MnO2/GFF catalyst, enabling the spontaneous solar thermal catalytic (STC) reaction to proceed. However, the conversion of toluene was only 8% with a CO2 selectivity of 60%, which can be attributed to the absence of ozone in this reaction and the insufficient activation of lattice oxygen at the relatively low temperature (72.6 °C) induced by the solar irradiation. These results clearly show that the conversion of toluene in the SEPPC is significantly higher than the sum of the toluene conversion using STC and PPC processes, which demonstrates that a clear synergistic effect of the SEPPC process resulted from the coupling of solar irradiation and PPC has been achieved with a SC of ∼42% in this work.

The roles of pure MnO2 and GFF in the SEPPC oxidation of toluene were further investigated. The catalytic performance of pure MnO2 in the SEPPC and PPC processes is given in Figure S10 (Supporting Information). Compared with MnO2/GFF, pure MnO2 exhibits lower toluene conversion, CO2 selectivity, carbon balance, and energy efficiency in the PPC process. The residual ozone concentration over pure MnO2 is higher than that of MnO2/GFF, indicating that the ozone conversion capability of pure MnO2 is weaker than that of MnO2/GFF. Under solar irradiation, the catalytic activity of MnO2 is enhanced because of the heating of the catalyst bed. However, the toluene conversion, CO2 selectivity, carbon balance, and energy efficiency in the SEPPC process using MnO2 are still 21, 23, 19, and 15% lower than that of MnO2/GFF at an SIE of 360 J L–1, respectively, demonstrating that pure MnO2 (without GFF) is not efficient for the oxidation of toluene in the SEPPC process. As shown in Figure S11 (Supporting Information), the toluene conversion over pure GFF in the SEPPC and PPC processes is ∼46.8 and ∼46.2%, respectively, at an SIE of 401 J L–1, which is almost the same as that achieved in the NTP-only process (∼47.3%). Therefore, pure GFF without MnO2 loading exhibits no catalytic activity toward toluene oxidation in the SEPPC and PCC processes. These results clearly show that the bifunctional GFF, designed as the catalyst support and light absorber, plays a crucial role in the oxidation of toluene in the SEPPC process. Bifunctional GFF can provide abundant sites for the dispersion of MnO2 and enhance the absorption of light, elevating the temperature of MnO2, significantly contributing to the enhanced catalytic activity of the MnO2/GFF catalyst.

Catalyst Stability

Catalyst stability is the key to the using plasma-catalytic gas cleaning technology on a commercial scale. Figure shows a comparison of the catalyst stability in the PPC and SEPPC processes. The results clearly show that the MnO2/GFF catalyst suffers from serious deactivation in the PPC process at room temperature without solar illumination. Specifically, the conversion of toluene decreases from 56 to 45% after the 12 h test at an SIE of 364 J L–1. Similarly, the selectivity of CO2 in the PPC process decreases significantly after running the reaction for 12 h. In contrast, the MnO2/GFF catalyst exhibits outstanding stability in the SEPPC process after the 72 h-test at 1 sun (activity loss of <3%), significantly superior to those reported in previous works (e.g., loss of ∼16% after 0.5 h[23] and loss of ∼50% after 2.5 h[32]). Furthermore, the surface states of the MnO2/GFF catalyst before and after the 72 h-reaction were analyzed using the C 1s spectra (Figure S12, Supporting Information). The peaks at 284.6, 286.5, 287.6, and 289.1 eV correspond to C=C/C–C, C–O, C=O, and O=C–O bonds, respectively.[65] Although weak oxidation of GFF is detected after the exposure of the catalyst to the ozone flow in the SEPPC and PPC processes for 72 h, the content of the formed C=O and O=C–O bonds is only 3.9 and 3.1%, respectively. In addition, the intensity of C=C/C–C and C–O peaks is almost unchanged, indicating that GFF has excellent structural stability. Therefore, we have successfully demonstrated the steady-state toluene conversion at 1 sun via the exploitation of abundant, renewable solar energy, significantly reducing the energy consumption and avoiding the use of external heating devices in the conventional PPC process.

Figure 7

Comparison of the catalytic stability in PPC and SEPPC processes: regarding (a) toluene conversion and (b) CO2 selectivity.

To elucidate the underlying mechanism contributing to the achieved superior stability of the MnO2/GFF catalyst in the SEPPC oxidation of toluene, a solar-induced regeneration of the deactivated catalyst was carried out using ozone. The characteristics of the fresh, spent, and regenerated catalysts were investigated using FTIR and XPS analysis. Compared with the fresh catalyst, the FTIR spectrum of the spent catalyst exhibits a new absorption peak corresponding to C–O–H bending at 1390 cm–1 (Figure a), indicating the deposition of organic intermediates on the catalyst surface.[55] Without solar irradiation, the deposited reaction intermediates on the catalyst cannot be completely removed or oxidized because of the ozone decomposition to more oxidative species on the catalyst being ineffective at room temperature, resulting in the deactivation of the MnO2/GFF catalyst in the PPC process.[62] Interestingly, this characteristic peak disappears after the solar-induced thermal regeneration of the spent catalyst. The elevated temperature of the catalyst because of solar irradiation promotes the decomposition of ozone, generating more active oxygen species, which significantly contribute to the highly efficient oxidation of intermediates on the catalyst surface.

Figure 8

(a) FTIR, (b) Mn 2p, and (c) O 1s spectra of fresh, spent, and regenerated MnO2/GFF catalysts.

The change of chemical states on the catalyst surface was examined using XPS, as plotted in Figure b,c. The Mn 2p spectra can be split into Mn3+ (641.0–641.4 eV) and Mn4+ (642.1–642.6 eV) peaks,[49,66,67] whereas the spectra of O 1s fit the Olatt (lattice oxygen, 529.2 eV) and Oads (surface adsorbed oxygen, 531.6–531.9 eV) peaks.[59] Higher Mn3+/Mn4+ and Oads/Olatt ratios represent the presence of more oxygen vacancies, as well as adsorbed oxygen species with a high mobility on the catalyst surface, respectively.[30,31,60,68] Compared to the fresh catalyst, both the Mn3+/Mn4+ and Oads/Olatt ratios on the spent catalyst after the reaction decrease from 1.10 and 1.31 to 0.87 and 1.19, respectively. In contrast, both ratios on the surface of the solar-induced regenerated catalyst increase and reach 1.04 and 1.28, respectively, demonstrating the self-cleaning capability of the MnO2/GFF catalyst, regenerating oxygen vacancies and adsorbed oxygen species on the catalyst surface. These promising findings show the unique solar-driven self-cleaning ability of the MnO2/GFF catalyst in the reaction, which contributes to the superior long-term catalytic stability in the SEPPC oxidation of toluene.

Reaction Mechanism

Figure a shows the formation of byproducts under different process modes. Benzaldehyde, benzoic acid, maleic anhydride, and nitrophenol are identified as the byproducts. To be specific, the absorption bands at 3600–3400 and 1600–1450 cm–1 correspond to the stretching and bending vibrations of OH groups in organic byproducts (e.g., phenol, carboxylic acid, etc.).[69] The absorption bands at 1900–1800 and 1700–1680 cm–1 can be ascribed to the stretching vibrations of C=O groups in maleic anhydride and benzaldehyde, respectively.[70,71] The presence of benzoic acid is confirmed by the bands at 1650–1610, 1550, and 1350–1300 cm–1,[72,73] whereas the absorption peaks at 1400–1350 cm–1 are related to the C–N stretching vibration.[70] In addition, NO2 and N2O are also detected (the absorption bands at 1621 and 2224 cm–1, respectively).[20,58]

Figure 9

(a) FTIR spectra of toluene oxidation using NTP-only, PPC, and SEPPC. (b) DRIFTS analysis of fresh and spent catalysts (after SEPPC and PPC reactions). (c) GC–MS analysis of a spent catalyst (after the PPC reaction).

Figure b shows the DRIFTS analysis of the fresh and spent catalysts (after the SEPPC and PPC reactions). The bands at 1340 cm–1 correspond to the CH2 deformation vibration of benzyl species because of the formation of benzyl alcohol.[74] The absorption bands at 1540 cm–1 can be assigned to the COO– stretching vibration of benzoic acid.[75] The characteristic bands at 1700 and 1760 cm–1 are ascribed to C=O stretching vibration of benzaldehyde and maleic anhydride, respectively.[75,76] As shown in Figure c, different organic intermediates are measured on the surfaces of the spent catalyst using GC–MS (see the Supporting Information & Table S1 for details).

The formation of byproducts under different process modes provides direct evidence to elucidate the plausible dominant reaction pathways in the oxidation of toluene using SEPPC (Figure ). The oxidation of toluene in the SEPPC process can be attributed to both the gas-phase reaction in the DBD plasma and solar-enhanced catalytic oxidation on MnO2/GFF.[77] It has been well established that excited nitrogen species (e.g., excited N2 (C3Πu–B3Πg)) and oxidative species (e.g., O and OH radicals) generated in air plasmas play critical roles in the stepwise decomposition and oxidation of toluene and its intermediates (Figure a).[23,78] Benzyl generated by the H-abstraction of methyl can be oxidized by O or OH to form benzaldehyde, which can be further oxidized to benzoic acid.[79] Benzyl radicals can also be oxidized to produce benzyl alcohol.[80] Benzoic acid could react with methyl radicals and electrons to form benzoic acid, methyl ester, and further generate benzaldehyde dimethyl acetal. In addition, the C–C bond between methyl and the benzene ring can be fractured by excited nitrogen species and electrons to generate methyl and phenyl radicals. Phenyl radicals can react with OH, H, and NO2 to form phenol, benzene, and nitrobenzene, respectively.[21] Phenol can be further oxidized to 1,4-dioxan-2-ol, as confirmed by the analysis of GC–MS. After that, aromatic intermediates collide with energetic electrons and reactive species, resulting in the rupture of aromatic rings and stepwise oxidation into CO, CO2, and H2O.[81]

Figure 10

Plausible reaction pathways for toluene conversion in the SEPPC process: (a) NTP induced gas-phase reactions in the plasma zone and (b) solar-enhanced catalytic reactions on the surfaces of MnO2/GFF.

As shown in Figure b, in the solar-induced thermal catalytic reaction, unconverted toluene and reaction intermediates in the plasma gas-phase reaction can be absorbed onto the surface of the MnO2/GFF catalyst and be further oxidized by active atomic oxygen (O*) species produced over the oxygen vacancies (VO) from the decomposition of ozone according to the following processes (R1–R3).[30,31,82] The presence of Mn3+ in MnO2/GFF indicates the formation of oxygen vacancies on the catalyst surface because of the electrostatic balance.[83] The transformation of Mn4+ to Mn3+ in the catalytic reaction can supply oxygen vacancies. In addition, the consumed oxygen vacancies could be replenished by gas-phase oxygen.[23]

The conversion of ozone depends on the decomposition rate of the peroxide species (O2*).[84] Jia et al. reported that the decomposition of peroxide species could be accelerated by increasing the reaction temperature.[30] Thus, the solar-induced thermal effect enhances the catalytic decomposition of ozone on the surface of MnO2/GFF, generating more atomic oxygen species, which make a major contribution to the enhanced oxidation of toluene and reaction intermediates. The finding can be confirmed by the significantly higher ozone decomposition (an triple enhancement) in the SEPPC process compared to that of the PPC (without the solar irradiation) counterpart operated at room temperature. Furthermore, the O3 decomposition capacity over the MnO2/GFF catalyst with solar irradiation is significantly higher than that reported in previous studies.[20] The concentration of residue ozone after the SEPPC process is considerably lower than that of the PPC process without the solar irradiation and NTP-only process. Note that the catalyst weight (9.8 mg only) used in this work is 10–100 times lower than that used in most of the previous works.[44,62,63] Therefore, the concentration of residue ozone can be further reduced by tuning the weight of the catalyst.

In addition, the Mars-van Krevelen (MVK) mechanism has been generally considered as a major pathway in thermal catalytic oxidation of toluene over MnO.[50,85] Lattice oxygen (Olatt) on the catalyst surface can oxidize toluene and reaction intermediates but often requires relatively high temperatures to ensure the effective surface mobility of lattice oxygen.[85] In this study, we find that the conversion of toluene was only 8% in the solar thermal catalytic oxidation of toluene on MnO2 by air (without O3), which can be ascribed to the weak or insufficient activation of lattice oxygen at the relatively low temperature (72.6 °C) induced by the solar irradiation. Thus, the STC oxidation of toluene and reaction intermediates via the MVK mechanism has made a limited contribution to the superior performance achieved in the SEPPC process. Figure shows the plausible reaction mechanisms in the SEPPC oxidation of toluene over the MnO2/GFF catalyst.

Figure 11

Mechanism of toluene conversion over the hierarchical fin-on-fin MnO2/GFF catalyst in the SEPPC process.

Conclusions

In this study, we demonstrated that the SEPPC process offers a promising solution for highly efficient oxidation of toluene over the MnO2/GFF catalyst with a hierarchical fin-on-fin structure. The results show that the bifunctional MnO2/GFF catalyst with vertical orientation and non-agglomerated morphology can minimize the reflection and transmission of incident light (absorption of >95%), effectively converting the absorbed solar energy into heat, which yields a remarkable temperature rise of 55.6 °C of the catalyst bed under solar irradiation at 1 sun. The amount of MnO2/GFF catalyst (9.8 mg) was substantially lower (10–100 times lower) than that used in previous works (usually 100–1000 mg). Compared to the PPC process, the coupling of solar illumination with the PPC process significantly enhances the toluene conversion and CO2 selectivity by 36–63%, reaching up to ∼93 and ∼83%, respectively, at an SIE of ∼350 J L–1, while reducing the energy consumption of the hybrid process by 35–52%. As a result, the SEPPC process exhibits a high energy efficiency of 12.7 g kWh–1 for toluene conversion (∼57% higher than that using the PPC process), which is among the highest in the previously reported works. In addition, the MnO2/GFF catalyst exhibits a superior long-term catalytic stability (activity loss <3%) after running the reaction for 72 h at 1 sun because of its excellent self-cleaning capability induced by solar irradiation. The synergistic effect of solar irradiation and PPC with a SC of ∼42% can be mainly attributed to the solar-induced thermal effect on the catalyst bed, promoting ozone conversion (enhanced by ∼182% compared to PPC) into more oxidative species (e.g., O radicals) and enhancing the catalytic oxidation of toluene and reaction intermediates on the catalyst surface, as well as the self-cleaning capacity of the catalyst at elevated temperatures. By contrast, the thermal catalytic oxidation of toluene and reaction intermediates via the MVK mechanism has a limited contribution to the superior performance achieved in the SEPPC. This work opens a new route to use abundant, renewable solar energy to achieve high-performance and energy-efficient removal of VOCs using the emerging plasma catalytic process. Getting a fundamental understanding of such a promising solar-enhanced plasma-catalytic gas cleaning process using a model VOC compound is the key and the foundation for the further R&D and process optimization toward future industrial applications.