Volatile organic compounds (VOCs) removal in non

Journal of Hazardous Materials 347 (2018) 317–324

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Volatile organic compounds (VOCs) removal in non-thermal plasma double dielectric barrier discharge reactor

T

Muhammad Farooq Mustafaa,b, Xindi Fua, Yanjun Liua, Yawar Abbasa, Hongtao Wanga,b, ⁎ Wenjing Lua,b, a

School of Environment, Tsinghua University, Beijing, 100084, PR China Key Laboratory for Solid Waste Management and Environment Safety (Tsinghua University), Ministry of Education of China, Tsinghua University, Beijing, 100084, PR China b

A R T I C L E I N F O

A B S T R A C T

Keywords: Non-thermal plasma Double dielectric barrier discharge Volatile organic compounds Plasma-catalyst reaction Energy efficiency

Non-thermal plasma (NTP) an emerging technology to treat volatile organic compounds (VOCs) present in unhygienic point source air streams. In present study, double dielectric barrier discharge (DDBD) reactors were used for the first time to evaluate the removal efficiency of VOCs mixture of different nature at constant experimental conditions (input power 16–65.8 W, VOCs mixture feeding rate 1–6 L/min, 100–101 ppm inlet concentration of individual VOC). Reactor A and B with discharge gap at 6 mm and 3 mm respectively, were used in current study. When treated at an input power of 53.7 W with gas feeding rate of 1 L/min in DDBD reactor A, removal efficiency of the VOCs were: tetrachloroethylene (100%), toluene (100%), trichloroethylene (100%), benzene (100%), ethyl acetate (100%) and carbon disulfide (88.30%); whereas in reactor B, the removal efficiency of all VOCs were 100%. Plasma-catalyst (Pt-Sn/Al2O3, BaTiO3 and HZSM-5) synergistic effect on VOCs removal efficiency was also investigated. Highest removal efficiency i.e 100% was observed for each compound with BaTiO3 and HZSM-5 at an input power 65.8 W. However, integrating NTP with BaTiO3 and HZSM-5 leads to enhanced removal performance of VOCs mixture with high activity, increase in energy efficiency and suppression of unwanted byproducts.

1. Introduction Volatile organic compounds (VOCs) released from waste treatment facilities (composting units/plants, incineration plants and landfills) into atmosphere has become an important issue in that they are not only harmful to human health but also destructive to the urban environment [1]. Potential reported VOCs from the waste treatment facilities are aromatics, hydrocarbons, sulfur compounds, aldehydes, alcohols, acids, halogenated compounds, terpenes etc [2]. Many VOCs have been proved to be carcinogenic and/or mutagenic. Moreover, once emitted into the atmosphere, VOCs could act as a predecessor for photochemical smog and aerosols formation [3]. Though optimization of process is a viable approach for reduction of VOCs, but still an effective and economical abatement technique is entailed to meet the progressively strict emission regulations around the world. So far, many technologies includes thermal decomposition, catalytic decomposition, biofilteration and oxidation are developed to decompose or reduce VOCs emissions [4]. All these techniques revealed to be unsuitable for the treatment of large air feedings at relatively low concentration of

⁎

contaminants (< 100 ppm), due to their impacts in terms of material and chemicals required, energy consumption, generation of waste products and related costs [5]. In recent years, non-thermal plasma (NTP) decomposition of VOCs under low temperature and pressure has got great attention and is considered to be green and potentially effective synthetic strategy for air pollutant remediation [6,7]. NTP mainly functions to oxidize and decompose VOCs through extremely active electrons and chemical dynamic species produced from high discharge voltage [7]. The investigations of NTP decomposition of VOCs have been made globally by different types of discharges like spark, corona, gliding arc or dielectric barrier discharges (DBD) [8–12]. DBD plasma discharge type has gain more importance among others to decompose VOCs in a cost-effective way [13–15]. DBD plasma reactor has the advantages of high decomposition efficiency especially for low concentration VOCs, simple experimental setup [16], stable and reproducible plasma conditions [17]. DBD reactors are not only appropriate for laboratory-scale research but also can be upgraded for industrial applications due to easy handling merit [18]. Furthermore, catalysts can be combined with DBD plasma to generate a hybrid

Corresponding author at: Division of Solid Waste Management, School of Environment, Tsinghua University, Beijing 100084, PR China. E-mail address: [email protected] (W. Lu).

https://doi.org/10.1016/j.jhazmat.2018.01.021 Received 8 October 2017; Received in revised form 30 December 2017; Accepted 9 January 2018 Available online 09 January 2018 0304-3894/ © 2018 Elsevier B.V. All rights reserved.


M.F. Mustafa et al.

diluted in Nitrogen (N2). Process parameters including input power, feed gas-mixture feeding rate as well as plasma discharge gap distance on VOCs removal efficiency have been studied in a DDBD reactor. Moreover, combination of NTP with different catalysts has also been investigated to scrutinize the effect on complex mixture of VOCs decomposition.

plasma-catalytic process, which has great potential to produce synergy, thereby enhancing the decomposition of pollutants, it also helps in enhancing the energy efficiency of the process [18]. Number of studies has been conceded on the application of NTP to remove single VOCs from air. Ye et al. carried out both a laboratory scale and at pilot scale study to eliminate benzene from air in a single DBD (SDBD) plasma reactor [19] ; Schiorlin et al. applied different kinds of corona discharges to abate toluene in air, providing possible mechanisms of oxidation [20]; Ragazzi et al. removed methyl ethyl ketone from dry synthetic air by applying a DBD in a closed hydraulic circuit [17]; Schmidt et al. adopted a DBD reactor to remove toluene both from dry and wet synthetic air [21]. Only few, studies on VOC removal with NTPs were carried out on mixtures of compounds. Subrahmanyam et al. worked on a plasma driven catalyst (PDC) system based on a DBD to remove toluene, iso-propanol and trichloroethylene from air by testing different catalysts [22]; Schiavon et al. applied a DBD to remove two mixtures of VOCs: ethanol and ethyl acetate, in a first experiment, and toluene, benzene and n-octane in another experiment [5]. However, SDBD reactor cannot completely mineralize the contaminants in environment but also produces noxious byproducts. Meanwhile, solid byproducts can easily deposit on the reactor’s inner surface and electrodes, which further influence the performance of SDBD reactor[23]. For gas removal applications, double dielectric barrier discharge (DDBD) configured plasma reactor is more suitable as it isolates both electrodes from the plasma reaction chamber as well as to protect the inner electrode from carbon deposition and generation of by-products during chemical reactions [24]. Hongbo et al. developed a DDBD reactor contains two plasma discharge regions to treat styrene in air [23]. They found that, styrene having low bond energy groups gets destroyed first in outer discharge region by weak plasma discharge energy and the groups with high bond energy like aromatic rings abolished in inner plasma discharge region by high plasma discharge energy. A knowledge gap is still exists about mixture of different nature of VOCs treatment in DDBD reactor. For that reason, complex mixtures of six toxic and odorous VOCs were chosen in present study to represent the stripping air of composting facility. These VOCs includes tetrachloroethylene (C2Cl4), toluene (C7H8), trichloroethylene (C2HCl3), benzene (C6H6), ethyl acetate (C4H8O2) and carbon disulfide (CS2). A novel DDBD reactor in a cylindrical configuration is proposed for the first time to decompose low concentrated complex mixture of VOCs

2. Materials and methods 2.1. Plasma rector The experimental set-up (Fig. 1) comprised of a reaction gas supply system, a DDBD plasma reactor, AC power supply and analytical instrumentation (GC–MS). Cylindrical DDBD reactor was made in quartz glass and connected to the plasma generator source in a sequential manner. The inner tube (outer diameter 8 mm, thickness 1 mm, length 360 mm) is adjacent to the ground electrode; the outer surface of outer tube (outer diameter 25 mm, thickness 2.5 mm, length 360 mm) is contacted with the high voltage discharge electrode. The discharge electrode made of a steel mesh (length 205 mm) and is connected to plasma generator power supply system. The copper metal rod was used as ground electrode inside internal dielectric quartz tube. VOCs gas mixture diluted in Nitrogen (N2) were fed into reactor. Two DDBD reactors (A and B) with different plasma discharge gap distances have been adopted to scrutinize the effect on VOCs mixture removal efficiency. Reactor A has 6 mm plasma discharge gap between two quartz dielectrics while that of reactor B is 3 mm. The discharge gap for DDBD reactors is adjusted by changing the diameter of outer quartz tube only. For reactor A and B, the dimensions of inner quartz tube remain the same. Reactor B has 3 mm discharge gap, the diameter of its outer tube is 19 mm; while reactor A has 6 mm discharge gap, and the diameter of its outer tube is 25 mm. Different catalysts in powder form were introduced individually in the centre of single plasma discharge zone. 2.2. Input gas and controlled parameters Alternating current plasma generator (Nanjing Suman Electronics Corp. CTP-2000K) was used to generate plasma discharge. The target VOCs in the current study were C2Cl4, C7H8, C2HCl3, C6H6, C4H8O2 and CS2, which are all typical VOCs from MSW composting plant [2]. Initial

Fig. 1. Schematic illustration of DDBD reactor setup for VOCs decomposition.

318


M.F. Mustafa et al.

concentration of the each VOC was 100 ppm approx balanced in Nitrogen (N2) cylinder and VOCs gas-mixture feeding rate was then controlled by the mass flow controllers (MFCs) (MC Co Ltd., China). Nine different levels of input power including 16, 23.1, 32.6, 39.3, 44.3, 49.5, 53.7, 57.6 and 65.8 W; six different feeding rates involving 1, 2, 3, 4, 5 and 6 L/min were chosen to analyze the effect of input power and gas feeding rate on VOCs removal efficiency under atmospheric pressure. Applied voltage is recorded by a digital oscilloscope (Tektronix DPO 4034). The applied voltage are 6.52, 6.62, 6.83, 6.74, 7.04, 6.63, 7.03, 6.74 and 6.96 kV across input powers (W) of 16, 23.1, 32.6, 39.3, 44.3, 49.5, 53.7, 57.6 and 65.8 W respectively. 2.3. Gas analysis The exhaust gaseous samples from each process trails were analyzed by a gas chromatography equipped with mass detector (Agilent 7860 A) and argon was used as the carrier gas. These analyses were performed in the Environmental Analysis and Observation Laboratory in Tsinghua University (Beijing, China) as per method of United States Environmental Protection Agency (USEPA, Compendium Method TO 15). Fig. 2. DDBD reactor with catalyst placed in the centre of plasma discharge zone.

2.4. Plasma- catalyst system

Energy efficiecny (μmol/kJ ) Removal efficiency of VOCs mixture was investigated at plasma alone conditions as well as plasma-catalysts (Pt-Sn/Al2O3, HZSM-5 and BaTiO3) combined system. The Pt-Sn/Al2O3 catalyst used in this study was prepared by sequential impregnation of Al2O3 support with ‘Sn’ first, followed by Pt using the 1 wt% Pt and 0.5 wt% Sn [25]. The X-ray diffraction measurements were obtained and presented in supplementary data (Fig. S1). Commercial HZSM-5 zeolite and BaTiO3 was supplied by Catalytic Factory of Nankai University (Tianjing, China) and Aladdin Industrial Corporation (Shanghai, China), respectively. The specific surface area of Pt-Sn/Al2O3, HZSM-5 and BaTiO3 is 139.21, 202.82 and 3.06 m2/g, respectively. HZSM-5 has the largest surface area among the tested catalysts. According to the theory, employment of material with large specific surface area in plasma active zone can improve catalyst surface reaction. Whereas, barium titanate (BaTiO3) is selected in this study as catalyst to combine with the NTP system, because it is the most widely used ferroelectric material for environmental purposes, and the dielectric constant of BaTiO3 is 2000 < ε < 10,000 [26]. In the DDBD reactor, each type of catalyst in fine powder form was placed inside the single plasma discharge zone between two dielectric quartz tubes as shown in Fig. 2. Two grams of each type of catalyst was used and the height of catalyst bed as well as plasma discharge gap distance was fixed at 6 mm. The tests were performed at different input power with a constant VOCs gas feeding rate i.e. 4 L/min in DDBD reactor A. All experiments were conducted at least 3 times (n ≥ 3) and data reported are average values.

No of (C 2Cl 4 + C 7H 8 + C 2HCl3 + C 6H 6 + C 4H 8O2 =

The present study has been aimed at the removal of VOCs mixture of different nature in DDBD reactor. In order to understand removal behavior of VOCs mixture in DDBD plasma alone condition, initial experiments were carried out as a function of input power and gas-mixture feeding rate in two DDBD reactors (A and B) with different discharge gapes. 3.1. Influence of input power on VOCs removal efficiency Input power is an important factor to influence VOCs removal in DDBD reactor. Measurements on removal efficiency of VOCs mixture as a function of input-power (from 16 W up to 65.8 W) were taken at a constant frequency of 9 KHz and a gas feeding rate at 1 L/min. Fig. 3 shows the VOCs removal efficiency in reactor A and B. At 16 W, the removal efficiency for the C2Cl4, C7H8, C2HCl3, C6H6, C4H8O2 and CS2 in reactor A was 90.46, 88.07, 86.17, 85.46, 83.45 and 67.70%, respectively, whereas, in reactor B it was 100, 100, 96.62, 94.61, 100 and 86.32%, respectively. The removal efficiency of VOCs mixture in both reactors A and B increases proportionally with an increase in input power. At 53.7 W, the removal efficiencies in reactor A were reached 100% for each VOC except CS2 i.e. 88.30%, whereas in reactor B, it was observed 100% for all six VOCs. With further increase in input power up to 65.8 W, the removal efficiency for each VOC was remained stable at 100% and increase in removal efficiency for CS2 from 88.30 to 93.82% was noticed in reactor A, while in reactor B, it remained stable at 100% for all VOCs mixture. Several possible explanations of this observation is that, first, reactive species generated during plasma reaction is directly proportional to input power applied to the system, leading to high removal efficiency at high power than at low power [27]; Second, reactor A has greater discharge gap distance consequently has less reactive surrounding which decreases the collision possibility between VOCs molecules and species that are active in a discharge zone, led to lower removal efficiency. Third, reactor B has less volume

VOCs mixture removal efficiency was measured by comparing the VOCs concentration in exhaust gas with the initial concentration in inlet, as expressed in following Eq. (1):

[VOCs]in − [VOCs]out × 100 [VOCs]in

(1)

Moreover, the input power and energy efficiency was computed by following Eqs. 2 and 3:

∫ (V (t ) × I (t )) dt

Input power = f

(3)

3. Results and discussion

2.5. Data processing

Removal efficiency (%) =

+ CS 2) moles converted input power × 1000

(2) 319


M.F. Mustafa et al.

Fig. 3. VOCs removal efficiency as a function of input power in DDBD reactors. a) Reactor A with 6 mm discharge gap; b) reactor B with 3 mm discharge gap.

than reactor A, consequently increases the electric filed strength and also electron density, contributes high removal efficiency in this reactor. Furthermore, increase in input plasma power at a constant excitation frequency with fixed feeding rate may effectively enhance the electric field, electron density and also provides more energy to electrons consequently leads to high VOCs removal efficiency [28]. Karuppiah et al. conducted a study on removal efficiency of individual VOC (toluene, benzene, chlorobenzene) and mixture of VOCs in SDBD reactor. Hence concluded that DBD decomposition method is a better approach for VOCs mixture removal than an individual VOC, possibly due to reactions occurs chemically with partially decomposed molecules and/or excited species in the gas mixture [29]. Treatment of VOCs mixture may enhance the utilization efficiency of these active species.

removal efficiencies for all VOCs in reactor A were 100% except CS2 which is 93.82% at constant input power 65.8 W. When the gas feeding rate was increased to 3 L/min, the removal efficiency of VOCs mixture remained 100% for all the VOCs except for CS2 that dropped from 93.82 to 84.84%. With a further increase in gas feeding rate upto 6 L/min, the removal efficiency of C2Cl4, C7H8, C2HCl3, C6H6, C4H8O2 and CS2 dropped to 91.37, 94.46, 92.89, 92.79, 90.27 and 67.83% respectively, showed that VOCs mixture feeding rate has negative correlation with removal efficiency. Magureanu et al. have found similar decreasing trend on toluene oxidation in SDBD for a gas feeding rate between 0.11–0.33 L/min [30]. The reason for this trend could be filamentary discharge, since a fraction of the gas can evade the filaments of plasma at higher feeding rates hence leave the plasma discharge reaction chamber un-reacted [31]. In addition, increase in feeding rate of VOCs mixture has negative effect on reactant retention time and energy density (defined as ratio of input power over feeding rate). Once the geometry of DDBD plasma reactor is fixed, VOCs mixture residence time will have an effect on reaction time and will also influence the energy density all the way through the plasma discharge region. When increase the VOCs gas mixture feeding rate from 1 to 6 L/min, substantial dependency on feeding rate was observed for the removal

3.2. Effect of feeding rate Removal efficiency of VOCs as a function of gas feeding rate (from 1 to 6 L/min) were observed at a constant frequency of 9 KHz and an input power at 65.8 W. Fig. 4 illustrate that there is an inverse relationship between the VOCs mixture feeding rate and its removal efficiency in both reactors A and B. At gas feeding rate of 1 L/min, the

Fig. 4. VOCs removal efficiency as a function of gas feeding rate in DDBD reactor at fixed input power of 65.8 W. a) reactor A with 6 mm discharge gap; b) reactor B with 3 mm discharge gap.

320


M.F. Mustafa et al.

efficiency of CS2, which decreased from 93.82 to 67.83% in reactor A. At higher feeding rate, residence time of CS2 molecules and energetic particles in plasma will be curtailed. Consequently, the dissociation of CS2 due to high-energy electrons collision is less prone to occur, which will make the reaction less thorough and hence escort to low CS2 removal efficiency. Furthermore, this behavior is also expected due to reduction in energy density which consequently generates less reactive species and radicals for CS2 removal. Similar decreasing trend (100–85.55%) with increase in gas mixture feeding rate from 1 to 6 L/ min on CS2 removal efficiency was observed in reactor B as well. On the other hand, at 1 L/min, the removal efficiencies for other VOCs in reactor B were 100%. With an increase in gas feeding rate upto 6 L/min, the removal efficiency stabled at 100% for all other VOCs except CS2. Removal of VOCs may also depend on composition of specific compound being treated and the reactive species generated during plasma reaction. Generation of H, Cl and OH radicals could lead to high removal efficiency of C2Cl4, C7H8, C2HCl3, C6H6, and C4H8O2 except CS2. Introduction of oxygen content in the feed gas mixture could generate more OH radicals which may increase the removal efficiency of CS2 in applied experimental conditions. Xiao et al. have found that highly active short living oxygen species (such as O and OH radicals) might be vital to CS2 oxidation through direct attacks towards CS2 molecules [32].

the case of HZSM-5 combine with plasma. HZSM-5 has an adsorption property that extends the retention time of VOCs in the reactor, consequently collision probability increased between plasma generated active species and VOCs molecules lead to enhancement in removal efficiency of VOCs [26]. Moreover plasma-catalyst synergetic effect with BaTiO3 on VOCs gas-mixture also showed complete removal efficiency i.e. 100% for C2Cl4, C7H8, C2HCl3, C4H8O2 except C6H6 and CS2 i.e 98.92, and 89.05%, respectively, at low input power (16 W). The presence of BaTiO3 can be as an alternative approach for the enhancement of electric field in plasma discharge zone due to its ferroelectric properties, resulted in enrichment of mean electron energy. Therefore, the energetic electrons have tendency to generate more energetic species by means of ionization and dissociation, instead of forming less active species during rotational and vibrational excitation [26]. Overall, HZSM-5 and BaTiO3 combined with plasma showed better synergistic effect on removal efficiency of VOCs mixture even at very low input power. This might be because of surface reaction phenomena that might incorporate direct interaction with high energetic electrons and active radicals in plasma discharge zone and have reactions indirectly with surface radicals produced after decomposition and adsorption at active surface sites.

3.3. Combined effect of plasma and catalyst

DDBD plasma reactor seems a very promising development for a treatment of complex mixture of VOCs. However, undesired by-products may form in DDBD plasma reactor during complex chemical reactions. In general, these by-products can be either in form of solid deposition or in gaseous statues. Treatment of VOCs mixture results in a solid deposit forming on the outer surface of inner dielectric barrier as well as on the inner walls of outer dielectric barrier both in plasma alone and plasma catalyst combine system. Karatum et al. observed dark brown solid deposits inside the SDBD plasma reactor in the treatment of toluene and ethylbenzene [35]. Guo et al. observed a solid deposit product which is yellow in coloration during the treatment of toluene in SDBD, and expressed it as an aromatic polymer [36]. Few researches exist on solid deposit formation when treated VOCs with NTP alone and plasma catalyst hybrid systems, and explained the deposits as polymeric substances or carbonaceous deposits [3,37]. As showed in Fig. 6, our results are quite similar to these studies in both plasma alone and plasma–catalyst combine system. Solid deposit formed both on the outer surface of inner dielectric barrier (Fig. 6a) and on the inner walls of outer dielectric barrier (Fig. 6b). Deposits were yellow-brown, dark brown and black particles, most probably are carbonaceous deposits. In current study, the deposits were analyzed by the X-ray energy-dispersive spectrum (EDS). It showed strong peaks of carbonaceous compounds (Fig. S2, Supplementary data) which account for 79.79% of the deposited material. The rest were oxygenated compounds (11.50%), sulfur containing compounds (2.17%) and chlorinated compound (1.26%). Weight percentage of elements deposits on DDBD reactor walls during treatment of VCs is presented in (Table S1 Supplementary data). However, solid deposits can be detached by passing air at certain input voltage (6 kV) through the reactor by numerous times [26]. Although, plasma-catalyst combine system improves the removal efficiency of VOCs but at the same time, there might be some chlorine deposition on catalysts surface due to the decomposition of C2Cl4, and C2HCl3 in VOCs mixture. Magureanu et al. reported a deposition of chlorine on the surface of catalyst during C2HCl3 treatment in plasma–catalytic DBD reactor [38] At the same time, some gaseous by-products left the NTP reactor via the exhaust gas instead of being deposited in reactors. Table 1 shows gaseous by-products during removal of VOCs mixture in DDBD reactor conducted at 65.8 W with 4 L/min in compared with plasma alone and plasma-catalyst combined system. The gaseous by-products generated includes carbon dioxide (CO2), sulfur dioxide (SO2), ethylene oxide

3.4. Byproduct analysis

As aforementioned, another aim of the study is the removal performance of VOCs mixture in plasma-catalysts combine system. Experiments were carried out in plasma reactor combined with different types of catalysts, i.e. Pt-Sn/Al2O3, HZSM-5 and BaTiO3. The tests were performed at different input power with a constant VOCs gas feeding rate i.e. 4 L/min in DDBD reactor A. Removal efficiency of VOCs mixture were compared (Fig. 5). Removal efficiency obtained in plasma alone system for C2Cl4, C7H8, C2HCl3, C6H6, C4H8O2 and CS2 was 81.81, 81.36, 75.44, 76.73, 73.76 and 59.14%, respectively at 16 W, as shown in Fig. 5. Whereas in plasma-catalyst combine system, use of catalyst not only be able to enhances reaction surface area, but also maintain the non-uniform characteristics of gas discharge. Moreover, it can function as a dielectric source in the plasma discharge zone. In other words, catalyst particles polarization in the active plasma discharge zone may stimulates strong electric field more or less around each particle, subsequently results in micro-discharges between catalysts particles. Micro-discharges generates more energetic electrons, which is vital for plasma discharge sustainability [27]. However, in comparison of plasma alone system, plasma- catalyst combine system leads to high removal efficiency of VOCs mixture. In plasma which combined with bi-metallic catalyst PtSn/Al2O3, removal efficiency for C2Cl4, C7H8, C2HCl3, C6H6, C4H8O2 and CS2 was 74.03, 84.59, 70.96, 65.60, 93.90 and 53.05%, respectively at 16 W. This variation in performance of catalyst might depend on the nature and chemical composition of VOCs being treated. Sedjame et al. used a Pt with CeO2-Al2O3 for the oxidation of acetic acid (CH3COOH) and n-butanol (C4H10O), where catalyst showed enrichment in the oxidation of CH3COOH, at the same time there was not significant oxidation of C4H10O, indicates that VOCs nature influence the catalyst performance [33]. In addition, mixture of molecules may also inhibit the obliteration of VOCs because of contest between VOCs species and byproducts generated during plasma chemical reaction for adsorption sites. Ordonez et al. showed that aromatic compounds when present together retard each other such as n-hexane had no effect on toluene and benzene removal, moreover their presence inhibits hexane decomposition [34]. In addition, presence of chlorinated compounds may also influence the performance of catalysts due to the likelihood of poisoning of catalyst andadsorption of HCl and Cl2 on the active sites of catalyst. Furthermore, removal efficiency for all VOCs mixture were achieved 100% except CS2 i.e. was 80.18% at same low input-power in 321


M.F. Mustafa et al.

Fig. 5. VOCs removal efficiency as a function of input power in DDBD plasma alone and DDBD plasma-catalysts combine system.

be changed into surface streamers adjacent to catalyst surface in plasma-catalyst combine system [39]. These surface streamers showed increased ionization, resulted in promotion of removal efficiency and complete mineralization of VOCs. However, comprehensive mineralization of VOCs is significantly associated with the surface reactivity of the catalysts [40]. In addition, hydrogen was observed as a major gaseous by-product in both (NTP + BaTiO3) and (NTP + HZSM-5) cases; however it is not quantified in this study. Plasma-catalyst combine system enhanced performance mechanisms and serves as an effective solution in the reduction of unwanted byproducts, instead of working independently. Overall, integrating non-thermal plasma with HZSM-5 and BaTiO3 leads to enhanced removal performance of mixture

(C2H4O), ethyl chloride (C2H5Cl), cyclohexane (C6H12), pentadecane (C15H32) benzenonitrile (C7H5N), benzenediamine (C6H8N2), methane,isocyanato (C2H3NO), 2,2-dimethyltetradecane (C16H34), 1-propene1-thiol (C3H6S). Improvement in reduction of undesired byproducts was observed in plasma-catalysts combine DDBD reactor and followed a pattern of NTP alone < NTP + Pt-Sn/Al2O3 < NTP + HZSM5 < NTP + BaTiO3. Some by-products includes C6H12, C15H32, C7H5N, C16H34 and C3H6S were not detected in NTP + BaTiO3 and NTP + HZSM-5 cases. It implies that BaTiO3 and HZSM-5 promote the mineralization process towards total removal of VOCs. Introduction of a catalyst inside the plasma discharge zone can alter the nature of the discharge itself. Malik et al. demonstrated that volume streamers might 322


M.F. Mustafa et al.

Fig. 6. Solid deposit forming (a) on the outer surface of inner dielectric barrier (b) on the inner walls of outer dielectric barrier (c) cross-sectional view of solid deposits on the inner wall of outer dielectric tube.

Table 1 Gaseous by-products generated in DDBD reactors at 65.8 W with 4 L/min: plasma alone vs. plasma-catalyst combined system. Compound name

Plasma alone

Plasma + Pt − Sn/ Al2O3

Plasma + BaTiO3

Plasma + HZSM-5

H2 CO2 SO2 C2H5Cl C2H4O C6H12 C15H32 C7H5N C6H8N2 C2H3NO C16H34 C3H6S

+ + + + + + + + + + + +

+ − + + + + − + + − + +

+ + − + + − − − + − − −

+ + + + + − − − + + − − Fig. 7. Energy efficiency as a function of input power in a DDBD reactor with and without catalysts.

‘+’: detected and ‘−’: not detected.

adsorption and plasma discharge is responsible for enhancement in energy efficiency [26]. Plasma-catalyst combined system is favorable from the point of view of energy efficiency, as the activation energy required for VOCs decomposition is lower. Moreover, placement of catalyst in plasma discharge zone also serves as a beneficial route to enhance energy efficiency because synergistic effect tends to form more useful active species rather than forming less useful species for complete mineralization of VOCs. The study indicates that the removal efficiency and energy efficiency are the key parameters and need to be considered both during the assessment of NTP treatment. Plasma-catalyst combined system represents the future for full application of NTP technology in VOCs emission reduction (Fig. 7).

of VOCs with high activity and suppression of unwanted byproducts. 3.5. Energy efficiency Energy efficiency as a function of input power in plasma alone and plasma catalyst combined system were evaluated and compared at 4 L/ min in reactor A. In the case of plasma alone, maximum energy efficiency observed was 8.44 × 10−5 μmol/kJ at 16 W while minimum energy efficiency 2.63 × 10−5 μmol/kJ was observed at 65.8 W. Clearly, energy efficiency in DDBD system decreases with an increase of the input power, despite of high VOCs conversion efficiency at high input-power as shown in Fig. 7. This is because when input power increases more energy there might be more lost due to vibrational and electronic excitation of VOC molecules, which means less input energy is spent in dissociation of mixture of VOCs at high input power. However, transfer of energy and kinetic interaction in-between non-thermal plasma generated radicals and excited molecules remains uncertain [6]. Improvement in energy efficiency was observed with introduction of catalysts in DDBD reactor and followed a pattern of NTP alone < NTP + Pt-Sn/Al2O3 < NTP + BaTiO3 < NTP + HZSM-5. Plasma combine with BaTiO3 and HZSM-5 not only outcompete plasma alone system in terms of the performance of VOCs mixture removal efficiency, but also increased energy efficiency by 23.87 and 29.45% respectively, at input power of 16 W. In plasma combined with BaTiO3, maximum energy efficiency obtained was 11.09 × 10−5 μmol/kJ, which is 23.87% higher than that in plasma alone system. BaTiO3 serve as alternative way to enhance energy efficiency due to its ferroelectric properties that leads to increase the electric field in plasma discharge zone. In addition, synergistic effect of adsorption of VOCs on HZSM-5 trailed by surface disintegration and plasma reaction might be reason for enhancement in VOCs removal efficiency. Recurring operation of

4. Conclusion The removal of VOCs mixture of different nature was investigated in DDBD non-thermal plasma reactor with and without catalysts at constant operating conditions. Input power and discharge gap in-between two electrodes in DDBD reactor has direct effect on VOCs removal efficiency. With 100 ppmv approx concentration of selected individual VOC and at input power of 53.7 W in DDBD reactor B, the removal efficiency of selected VOCs were 100% at 1 L/min. Moreover, BaTiO3 and HZSM-5 combined with plasma showed better synergistic effect on removal efficiency of VOCs mixture even at very low input power. Solid deposits from treating VOCs mixture were detected. Plasma-catalyst combine system improved plasma reaction by increasing VOCs removal efficiency and by eliminating unwanted byproducts. Furthermore, improvement in energy efficiency by catalyst was observed following a pattern of NTP alone < NTP + Pt-Sn/ Al2O3 < NTP + BaTiO3 < NTP + HZSM-5.

323


M.F. Mustafa et al.

Acknowledgements

68–78. [19] Z. Ye, Y. Zhang, P. Li, L. Yang, R. Zhang, H. Hou, Feasibility of destruction of gaseous benzene with dielectric barrier discharge, J. Hazard. Mater. 156 (2008) 356–364. [20] M. Schiorlin, E. Marotta, M. Rea, C. Paradisi, Comparison of toluene removal in air at atmospheric conditions by different corona discharges, Environ. Sci. Technol. 43 (2009) 9386–9392. [21] M. Schmidt, M. Schiorlin, R. Brandenburg, Studies on the electrical behaviour and removal of toluene with a dielectric barrier discharge, Open Chem. 13 (2015) 477–483. [22] C. Subrahmanyam, A. Renken, L. Kiwi-Minsker, Novel catalytic non-thermal plasma reactor for the abatement of VOCs, Chem. Eng. J. 134 (2007) 78–83. [23] H. Zhang, K. Li, C. Shu, Z. Lou, T. Sun, J. Jia, Enhancement of styrene removal using a novel double-tube dielectric barrier discharge (DDBD) reactor, Chem. Eng. J. 256 (2014) 107–118. [24] Sazal K. Kundu, Eric M. Kennedy, Vaibhav V. Gaikwad, Thomas S. Molloy, Bogdan Z. Dlugogorski, Experimental investigation of alumina and quartz as dielectrics for a cylindrical double dielectric barrier discharge reactor in argon diluted methane plasma, Chem. Eng. J. 180 (2012) 178–189. [25] Z. Nawaz, F. Wei, Pt–Sn-based catalyst’s intensification using Al2O3–SAPO-34 as a support for propane dehydrogenation to propylene, J. Ind. Eng. Chem. 17 (2011) 389–393. [26] A.M. Vandenbroucke, R. Morent, N. De Geyter, C. Leys, Non-thermal plasmas for non-catalytic and catalytic VOC abatement, J. Hazard. Mater. 195 (2011) 30–54. [27] O. Khalifeh, H. Taghvaei, A. Mosallanejad, M.R. Rahimpour, A. Shariati, Extra pure hydrogen production through methane decomposition using nanosecond pulsed plasma and Pt–Re catalyst, Chem. Eng. J. 294 (2016) 132–145. [28] Hoang Hai Nguyen, K. Kyo-Seon, Combination of plasmas and catalytic reactions for CO2reforming ofCH4by dielectric barrier discharge process, Catal. Today 256 (2015) 88–95. [29] J. Karuppiah, E. Linga Reddy, P. Manoj Kumar Reddy, B. Ramaraju, R. Karvembu, C. Subrahmanyam, Abatement of mixture of volatile organic compounds (VOCs) in a catalytic non-thermal plasma reactor, J. Hazard. Mater. 237–238 (2012) 283–289. [30] M. Magureanu, N.B. Mandache, E. Gaigneaux, C. Paun, V.I. Parvulescu, Toluene oxidation in a plasma-catalytic system, J. Appl. Phys. 99 (2006). [31] C.A. Aggelopoulos, A. Gkelios, M.I. Klapa, C. Kaltsonoudis, P. Svarnas, C.D. Tsakiroglou, Parametric analysis of the operation of a non-thermal plasma reactor for the remediation of NAPL-polluted soils, Chem. Eng. J. 301 (2016) 353–361. [32] X. Yan, Y. Sun, T. Zhu, X. Fan, Conversion of carbon disulfide in air by non-thermal plasma, J. Hazard. Mater. 261 (2013) 669–674. [33] H.-J. Sedjame, C. Fontaine, G. Lafaye, J. Barbier Jr, On the promoting effect of the addition of ceria to platinum based alumina catalysts for VOCs oxidation, Appl. Catal. B Environ. 144 (2014) 233–242. [34] S. Ordóñez, L. Bello, H. Sastre, R. Rosal, F.V. Dı́ez, Kinetics of the deep oxidation of benzene, toluene, n-hexane and their binary mixtures over a platinum on γ-alumina catalyst, Appl. Catal. B Environ. 38 (2002) 139–149. [35] O. Karatum, M.A. Deshusses, A comparative study of dilute VOCs treatment in a non-thermal plasma reactor, Chem. Eng. J. 294 (2016) 308–315. [36] Y.-F. Guo, D.-Q. Ye, K.-F. Chen, J.-C. He, W.-L. Chen, Toluene decomposition using a wire-plate dielectric barrier discharge reactor with manganese oxide catalyst in situ, J. Mol. Catal. A Chem. 245 (2006) 93–100. [37] M. Magureanu, D. Piroi, N.B. Mandache, V.I. Pârvulescu, V. Pârvulescu, B. Cojocaru, C. Cadigan, R. Richards, H. Daly, C. Hardacre, In situ study of ozone and hybrid plasma Ag–Al catalysts for the oxidation of toluene: evidence of the nature of the active sites, Appl. Catal. B Environ. 104 (2011) 84–90. [38] M. Magureanu, N.B. Mandache, V.I. Parvulescu, C. Subrahmanyam, A. Renken, L. Kiwi-Minsker, Improved performance of non-thermal plasma reactor during decomposition of trichloroethylene: optimization of the reactor geometry and introduction of catalytic electrode, Appl. Catal. B Environ. 74 (2007) 270–277. [39] M.A. Malik, Y. Minamitani, K.H. Schoenbach, Comparison of catalytic activity of aluminum oxide and silica gel for decomposition of volatile organic compounds (VOCs) in a plasmacatalytic reactor, IEEE Trans. Plasma Sci. 33 (2005) 50–56. [40] A. Ogata, K. Yamanouchi, K. Mizuno, S. Kushiyama, T. Yamamoto, Oxidation of dilute benzene in an alumina hybrid plasma reactor at atmospheric pressure, Plasma Chem. Plasma Process. 19 (1999) 383–394.

This work was financially supported by Special Fund of Environmental Protection Research for Public Welfare, Ministry of Environmental Protection, P. R. China (No. 201209022) and the National Natural Science Foundation of China (Grant No. 41101225). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jhazmat.2018.01.021. References [1] Z.Z. Noor, R.O. Yusuf, A.H. Abba, M.A. Abu Hassan, M.F. Mohd Din, An overview for energy recovery from municipal solid wastes (MSW) in Malaysia scenario, Renew. Sustain. Energy Rev. 20 (2013) 378–384. [2] M.F. Mustafa, Y. Liu, Z. Duan, H. Guo, S. Xu, H. Wang, W. Lu, Volatile compounds emission and health risk assessment during composting of organic fraction of municipal solid waste, J. Hazard. Mater. 327 (2017) 35–43. [3] H.L. Chen, H.M. Lee, S.H. Chen, M.B. Chang, S.J. Yu, S.N. Li, Removal of volatile organic compounds by single-stage and two-stage plasma catalysis systems: a review of the performance enhancement mechanisms, current status, and suitable applications, Environ. Sci. Technol. 43 (2009) 2216–2227. [4] H.F. Abbas, W.W. Daud, Hydrogen production by methane decomposition: a review, Int. J. Hydrogen Energy 35 (2010) 1160–1190. [5] M. Schiavon, M. Scapinello, P. Tosi, M. Ragazzi, V. Torretta, E.C. Rada, Potential of non-thermal plasmas for helping the biodegradation of volatile organic compounds (VOCs) released by waste management plants, J. Clean. Prod. 104 (2015) 211–219. [6] T. Nozaki, K. Okazaki, Non-thermal plasma catalysis of methane: principles, energy efficiency, and applications, Catal. Today 211 (2013) 29–38. [7] N. Tsolas, K. Togai, R.A. Yetter, Non-equilibrium plasma-assisted flow reactor studies of highly diluted reactive mixtures, 53rd AIAA Aerosp. Sci. Meet. (2015) 0159. [8] C. Hoffmann, C. Berganza, J. Zhang, Cold atmospheric plasma: methods of production and application in dentistry and oncology, Med. Gas Res. 3 (2013) 1. [9] W.-J. Liang, H.-P. Fang, J. Li, F. Zheng, J.-X. Li, Y.-Q. Jin, Performance of nonthermal DBD plasma reactor during the removal of hydrogen sulfide, J. Electrostat. 69 (2011) 206–213. [10] J.D. Holladay, J. Hu, D.L. King, Y. Wang, An overview of hydrogen production technologies, Catal. Today 139 (2009) 244–260. [11] A. Indarto, J.-W. Choi, H. Lee, H.K. Song, Effect of additive gases on methane conversion using gliding arc discharge, Energy 31 (2006) 2986–2995. [12] T. Matsumoto, D. Wang, H. Akiyama, T. Namihira, INTECH open access publisher, Non-Thermal Plasma Technic for Air Pollution Control, (2012). [13] S.K. Kundu, E.M. Kennedy, V.V. Gaikwad, T.S. Molloy, B.Z. Dlugogorski, Experimental investigation of alumina and quartz as dielectrics for a cylindrical double dielectric barrier discharge reactor in argon diluted methane plasma, Chem. Eng. J. 180 (2012) 178–189. [14] R. Valdivia-Barrientos, J. Pacheco-Sotelo, M. Pacheco-Pacheco, J. Benítez-Read, R. López-Callejas, Analysis and electrical modelling of a cylindrical DBD configuration at different operating frequencies, Plasma Sources Sci. Technol. 15 (2006) 237. [15] M. Abdel-Salam, A. Hashem, A. Yehia, A. Mizuno, A. Turky, A. Gabr, Characteristics of corona and silent discharges as influenced by geometry of the discharge reactor, J. Phys. D Appl. Phys. 36 (2003) 252. [16] J. Hu, N. Jiang, J. Li, K. Shang, N. Lu, Y. Wu, A. Mizuno, Discharge characteristics of series surface/packed-bed discharge reactor diven by bipolar pulsed power, Plasma Sci. Technol. 18 (2016) 254–258. [17] M. Ragazzi, P. Tosi, E.C. Rada, V. Torretta, M. Schiavon, Effluents from MBT plants: plasma techniques for the treatment of VOCs, Waste Manage. 34 (2014) 2400–2406. [18] D. Mei, X. Tu, Conversion of CO2 in a cylindrical dielectric barrier discharge reactor: effects of plasma processing parameters and reactor design, J. CO2 Util. 19 (2017)

324