“Storage-oxidation” cycling process for indoor benzene removal at room temperature

Catalysis Today xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Catalysis Today journal homepage: www.elsevier.com/locate/cattod

“Storage-oxidation” cycling process for indoor benzene removal at room temperature ⁎

Yidi Wanga,b, Bingbing Chena,b, , Bo Wua,b, Limei Yua, Xiaobing Zhub, Chuan Shia,b, a b

⁎⁎

State Key Laboratory of Fine Chemicals, School of Chemistry, Dalian University of Technology, Dalian 116024, China Laboratory of Plasma Physical Chemistry, Dalian University of Technology, Dalian 116024, China

A R T I C L E I N F O

A B S T R A C T

Keywords: AgCu/TS-1 zeolite Benzene Non-thermal air plasma “storage-oxidation” cycling process

Non-thermal air plasma assisted “storage-oxidation” cycling process was applied to indoor benzene removal at room temperature and realized over TS-1 zeolite supported AgCu catalysts. AgCu/TS-1 catalyst showed not only promising C6H6 capacity under humid condition in the storage phase, but also high oxidation efficiency of absorbed benzene with the assistance of non-thermal plasma (NTP) at a moderate discharge power of 6 W. Due to the generation of increased amounts of Cu+ and active surface oxygen species resulting from the Ag–Cu interaction, the AgCu/TS-1 catalyst showed higher benzene storage and oxidation capacity than the other catalysts. Compared with the thermal regeneration, the good humidity tolerance, non-benzene release, and low energy cost was achieved in the non-thermal air plasma regeneration, which providing a promising way for indoor benzene purification.

1. Introduction Physical adsorption and catalytic oxidation have been considered as the most promising methods for indoor C6H6 removal [1–3]. However, it has been shown that traditional physical adsorption by activated carbon (AC) is facing the problem of hard to regenerate [4,5]. And in terms of catalytic oxidation, even on noble metals like Pt, Pd and Au catalysts, the temperature required for the complete oxidation of benzene (ca. 100–4000 ppm C6H6, GHSV = 10,000–60,000 h−1) is still higher than 150 °C [6–9], which result in the consequences of operational difficulties and high energy consumption. In our previous work, we reported a novel energy-efficient catalytic removal of benzene by tandem temperature-pulse oxidation (TTPO) process over Pt/HZ(200) [10]. Benzene is firstly gathered at room temperature and stored in the channel of HZSM-5; after saturation the catalyst could be regenerated in situ by heating, the stored species being completely oxidized to CO2 and H2O by Pt. The key issue for such a cycling process is the enhancement of benzene storage capacity under humid condition in order to prolong the catalyst's useful life for each cycle, as well as solving the problem of benzene release during temperature-rising stage, in which precious metal is needed (reaching the aim of benzene oxidation before desorption) and H2O existing must be avoid (huge amount of benzene will desorb in moisture containing regeneration).

⁎

Herein, as invoked by the characteristic of non-thermal plasma (NTP) [11], we attempted to substitute “temperature-pulse” oxidation by “NTP-pulse” regeneration to enable the whole “storage-oxidation” cycling process operated at room temperature. Since the oxidation regeneration by NTP is undertaken at room temperature, the possible release of benzene during “temperature-rising” stage could be avoided. Moreover, due to the reactive species such as O(3p) and OH generated by NTP in air, the stored benzene could be oxidized at room temperature without the catalysis by noble metals such as Pt [12]. TS-1 zeolite was selected as substrates for benzene storage based on its two characteristics: (i) 10-ring elliptical character of the channels renders TS-1 accessible for diffusion of C6H6 (kinetic diameter of 5.8 Å); (ii) the excellent hydrophobic properties of TS-1 as compared with HZSM-5, enables it to store benzene under humid conditions [13]. Meanwhile, it is noticed that the unique characteristics of the d orbitals in Ag/Cu or Ag+/Cu+ ions enable them to form bonds with unsaturated hydrocarbons in a nonclassical manner, which is broadly referred to as πcomplexation. This π-complexation has been proved to be able to enhance HCHO and C6H6 storage capacity in previous study [14–17]. Therefore, in this study, we propose a tandem “NTP-pulse” regeneration process which combines with benzene storage on AgCu/TS1 catalyst for indoor benzene removal at room temperature, as depicted in Scheme 1. It is a cyclic process of enrichment of low concentration of C6H6 on the TS-1 zeolite and a followed oxidation of the stored benzene

Corresponding author. Tel.: +86 41184986083. Corresponding author. Tel.: +86 41184986083. E-mail addresses: [email protected] (B. Chen), [email protected] (C. Shi).

⁎⁎

http://dx.doi.org/10.1016/j.cattod.2017.04.054 Received 25 October 2016; Received in revised form 6 March 2017; Accepted 25 April 2017 0920-5861/ © 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Wang, Y., Catalysis Today (2017), http://dx.doi.org/10.1016/j.cattod.2017.04.054

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

Transmission electron microscopy (TEM) images of the samples were obtained on a JEOL JEM-2000EX microscope operated at 200 kV. Diluted suspensions of samples in ethanol were prepared and drop dried on carbon coated copper TEM grids. 2.3. Catalytic activity measurements The catalytic activity evaluations include two processes: the storage process and the oxidation process. Both were carried out in a continuous flow fixed bed quartz microreactor at atmospheric pressure. 0.10 g catalyst (40–60 mesh) was sandwiched between quartz wool layers in the tube reactor. Before performing the experiment, the catalysts were pretreated in the N2 flow at 450 °C for 2 h. The storage process was carried out at room temperature, and 14 ppm C6H6/21% O2/1.564% H2O (relative humidity of 50%, at 25 °C) was introduced into the reactor at a total flow rate of 100 ml min−1, corresponding to a gas hourly space velocity (GHSV) of 30,000 h−1. Gaseous H2O was carried into the gas stream by passing N2 through a bubbler in a water bath at room temperature. The amount of water, expressed as the relative humidity (RH) at 25 ̊C, was controlled by adjusting the flow rate of N2, while keeping the total flow unchanged. The inlet and outlet C6H6 concentrations were measured by converting it to CO2 in a homemade C6H6-to-CO2 converter (CuMn/γ-Al2O3 catalyst) at 450 °C and determining the amount of CO2 formed by IRAS (SICK-MAIHAKS710, Germany) [18]. In thermal regeneration, the same simulated air stream without benzene was switched into the reactor, with the temperature ramped to 500 °C at 10 °C·min−1 and held there for 10 min. In plasma regeneration, air without benzene was switched to the catalystpacked DBD, with discharge at input power of 6 W. The C6H6 storage capacities were determined at 25 °C. The examined parameter, C6H6 storage capacity (NC6H6 , mmol·gcat−1) are defined as follow:

Scheme 1. Schematic diagram of the proposed “storage-oxidation” cycling process for C6H6 removal.

by assistance of NTP. The “NTP-pulse” only takes 1/10 time of storage time, providing an energy-efficient method for the synergy of plasma and catalysis. 2. Experimental details 2.1. Catalyst preparation Ag and Cu catalysts with nominal metal loading of 1 wt% were prepared by conventional incipient wetness impregnation method using TS-1 (commercial samples) as support. AgNO3 and Cu(NO3)2 were used as Ag and Cu precursors, respectively. The desired amount of AgNO3 was dissolved in needed deionized water. After impregnation, the sample was aged at room temperature overnight, followed by drying at 110 °C for 16 h and then calcined at 450 °C for 4 h in air. The obtained catalyst was denoted as Ag/TS-1 and Cu/TS-1. 0.5 wt%Ag0.5 wt%Cu/TS-1 (AgCu/TS-1) bimetal catalyst was prepared by the co-impregnation method. In a typical synthesis, first, desired AgNO3 and Cu(NO3)2 were co-dissolved in a certain amount of deionized water. Then, the mixed solution was added dropwise to the TS-1 support. The impregnated samples were aged overnight at ambient temperature in darkness followed by drying at 110 °C for 12 h and calcined in air at 450 °C for 4 h.

CCin6H6 F1 t1 ⎛ ⎞ NC6H6 ⎜mmol/g cat⎟ = Wcat ⎝ ⎠

(1)

, where CCin6H6 is the inlet C6H6 concentration in the feed gas, F1 is the total flow rate, t1 is the breakthrough time and Wcat is the catalyst weight. Breakthrough time was defined as the time when outlet C6H6 concentration reached 2% of feed concentration. A capillary-sampled quadrupole mass spectrometer with two-stage differential pumping (Omnistar™, Pfeiffer Vacuum, Germany) was used to examine the benzene desorption and the possible by-products at the discharge stage. The examined parameters, stored-C6H6 to COx oxidation efficiency (EC6H6 → COx , %), carbon balance (Bc, %), CO2 Selectivity (SCO2 , %), CO selectivity (SCO, %), and energy cost for remedying 1m3 air (Ec, kWh·m−3)

2.2. Characterization techniques Before characterization test, the samples were calcined in air at 450 °C for 4 h. Specific surface areas (SBETs) of the catalysts were calculated from a multipoint Brunauer–Emmett–Teller (BET) analysis of the nitrogen adsorption and desorption isotherms at −196 °C measured in a Quantachrome QUADRASORB SI gas adsorption analyzer. Prior to the measurement, the samples were degassed in vacuum at 300 °C for 10 h. X-ray powder diffraction (XRD) analysis was conducted using a D/ Max-2400 (Rigaku) instrument with CuKα radiation (λ = 0.1542 nm), operating at 40 kV and 100 mA; phase identification was achieved through comparison of XRD patterns to those of the Joint Committee on Powder Diffraction Standards (JCPDS). X-ray photoelectron spectroscopy (XPS) measurements were performed on a Thermo Scientific ESCALAB250 spectrometer using AlKα radiation (1486.6 eV) as the X-ray source. The beam was monochromatized by a twin crystal monochromator, yielding a focused X-ray spot with a size of 500 μm, at 10 mA × 5 kV. Curving fitting of the Ag 3d, Cu 2p and O 1s peaks were performed using XPSPEAK software. Scanning electron microscopy (SEM) images were obtained on a QUANTA 450 instrument with an acceleration voltage of 3 kV.

EC6H6 → COx (%) =

BC (%) =

6 × nCstored 6 H6

× 100%, (2)

produced produced nCO + nCO + 6 × nCdesorption 2 6 H6

SCO2 (%) =

SCO (%) =

out nCO x

6 × nCstored 6 H6 produced nCO 2 produced produced + nCO nCO2 produced nCO produced produced nCO2 + nCO

Ec (kW⋅h⋅m−3) =

2

(3)

× 100%, (4)

× 100%, (5)

P × t2 × c , nCstored × EC6H6 → COx 6 H6

produced = C(C6H6)a F1 t1, nCO where nCstored 6 H6 2 t2 t2 desorption F C d t n = F C ; 2 (C H ) 2 CO C H 6 6 d dt , 6 6 0 0

∫

× 100%,

∫

(6) t2 0

= ∫ F2 CCO2 dt ,

produced nCO

=

C(C6H6)a and C(C6H6)d means

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

2p3/2 component of Cu2+, while that at 930.4 eV to the Cu+ [22]. By comparison the ratio of Cu+/Cu2+ as shown in Table 2, a higher amount of Cu+ was obtained for AgCu/TS-1 sample, indicating that coexistence of Ag is in favor of the reduction of Cu2+ to Cu+ [23,24]. More importantly, Cu+ is believed to be more conductive to benzene adsorption [14,15,25]. The XPS spectra of O 1s of TS-1, Ag/TS-1, Cu/TS-1, AgCu/TS-1 catalysts are also shown in Fig. 2. There are two kinds of oxygen species, the one with O 1s binding energy at 532.4 eV is assigned to surface adsorbed oxygen (Oads), while that at 530.8 eV is due to surface lattice oxygen (Olat) [26–28]. The results reveal that the ratio of Oads/Olat increased upon Cu, Ag and AgCu doping to the TS-1 zeolite, with the largest portion of Oads was observed over AgCu/TS-1 sample. As it is known that these surface oxygen species always play crucial roles in catalytic oxidation reactions [29,30]. SEM images show that the average particles size of TS-1 zeolite are ca. 300–500 nm, introducing Ag and Cu into the zeolite did not change the morphology and particle size (Fig. 3). TEM images display the particle size distribution of the doped metals. As can be seen in Fig. 4(A), nano-sized silver particles were dispersed on TS-1 zeolite with an average particle size of ca. 6∼7 nm. For Cu/TS-1 sample (Fig. 4(B)), there were no particles dispersed on TS-1 substrates observed, which should be due to the poor contrast ratio between CuO and TS-1 zeolite [31,32]. While highly dispersed silver particles were clearly observed over AgCu/TS-1 sample (Fig. 4(C)), with an average particle size of ca. 6∼7 nm. BET, XRD and SEM results indicated that introducing Ag, Cu or AgCu into the TS-1 zeolite did not change the structure and morphology of the TS-1 itself. Moreover, Ag and/or Cu nanoparticles were highly dispersed on the TS-1 zeolite as shown in the TEM images. Meanwhile after loading with Ag, Cu and AgCu, the surface adsorbed oxygen species was increased, especially for the AgCu/TS-1 sample. The surface oxygen species of AgCu/TS-1 was nearly 2.5 times than TS-1 sample, which should be a key factor for the catalytic oxidation ability of AgCu/ TS-1 catalyst. In addition, the XPS results showed that co-existence of AgCu led to a higher ratio of Cu+ in the sample. From the results of molecular orbital calculation [33], Cu+ cation would form stronger πcomplexation bonds with benzene, which may contribute to the superior capacity of AgCu/TS-1 for benzene adsorption [34].

Table 1 The nominal metal content and specific surface area of the samples. Sample

Surface area (m2·g−1)

Metal content (wt%)

SiO2 HZ(25) HZ(50) HZ(200) TS-1 Cu/TS-1 Ag/TS-1 AgCu/TS-1

Ag

Cu

– – – – – – 1 0.5

– – – – – 1 – 0.5

304 349 350 351 357 346 343 356

benzene concentration in the gas stream during the storage phase and in the gas stream of the discharge phase, respectively. CCO and CCO2 are the concentrations of CO and CO2 in the gas stream of a discharge stage; F1 and t1 are the total flow rate (m3·h−1) and the storage period (h) at the storage stage, respectively; F2, t2 and P are the total flow rate (m3·h−1), discharge period (h) and discharge power (kW) at the discharge stage, respectively.

3. Results and discussion 3.1. Physicochemical properties of the catalysts Table 1 summarizes the surface area of the as-prepared Ag/TS-1, Cu/TS-1, AgCu/TS-1 catalysts together with the TS-1 and HZ support. The surface areas of these samples were 343, 346, 356, 357 and 351 m2·g−1, respectively, illustrated that the addition of 1 wt% Ag and/or Cu by co-impregnation method did not block or fill the zeolite channel. XRD patterns of TS-1 and metal loaded TS-1 catalysts in Fig. 1 correspond to the powder pattern of crystalline MFI zeolite. No diffraction peaks attributed to Ag, Cu metal and/or oxide were observed, due to the lower metal loading (≤1 wt%). Surface chemical state of the TS-1 supported Cu, Ag and AgCu catalysts were analyzed by XPS, results being shown in Fig. 2. The Cu 2p, Ag 3d and O 1s core level spectra of the catalysts were deconvoluted into several components and the results were listed in Table 2. Ag 3d spectra of Ag/TS-1 and AgCu/TS-1 both exhibit a doublet with a splitting of 6.0 eV. The binding energy (BE) of Ag 3d5/2 was at 368.4 eV, being ascribed to Ag0, indicating that Ag0 is the predominant species on the Ag/TS-1 and AgCu/TS-1 catalysts [19–21]. By means of curve-fitting, two kinds of copper species were detected both for Cu/TS1 and AgCu/TS-1 catalysts. The peak at BE of 932.8 eV is assigned to Cu

3.2. Benzene storage capacity Benzene storage capacities over HZSM-5 zeolites with different Si/ Al ratios, TS-1 zeolite, Ag, Cu or Ag-Cu doped TS-1 catalysts, as well as SiO2 were evaluated under simulated air containing C6H6 (14 ppm)

Intensity

Fig. 1. XRD patterns of TS-1, Cu/TS-1, Ag/TS-1 and AgCu/TS-1 catalysts.

AgCu/TS-1 Ag/TS-1 Cu/TS-1 TS-1 PDF#43-0055

5

10

15

20

25

30

35

40

45

50

55

2θ / 3

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

Fig. 2. Ag 3d, Cu 2p, O 1s spectra of the TS-1, Ag/TS-1, Cu/TS-1 and AgCu/TS-1 catalysts.

enhancement of hydrophobicity, TS-1 exhibited the largest benzene storage capacity. The extraordinary resistance to the competitive adsorption of water can be perceived in the inner pattern of Fig. 5. The storage capacity decreased by 60% over HZ, while only 10% over TS-1 catalyst in the case of introducing humidity (RH = 50%, RT) to feed gas. These results demonstrated that hydrophobic zeolite is preferred to C6H6 adsorption under humid conditions. When loaded with Ag, it was evident that the storage amount enhanced. The benzene storage capacity further increased on AgCu/TS-1 catalyst, reaching 0.0980 mmol g−1. The improvement in storage amount from TS-1 possibly related to the π-interaction of benzene with the anchored metal center, which can significantly strengthen the interaction of C6H6 on the catalyst. πcomplexation is a subclass of chemical complexation. The unique characteristics of the d orbitals in Ag or Cu enable them to form bonds with unsaturated hydrocarbons in a nonclassical manner, which is broadly referred to as π-complexation [39,40]. The formation of these complexes was due to the two-way donor-acceptor interactions. One is the charge donation from the filled π orbital of the ligand into the

Table 2 XPS data for the TS-1, Ag/TS-1, Cu/TS-1 and AgCu/TS-1 catalysts. Catalyst

BE/eV Cu

TS-1 Ag/TS-1 Cu/TS-1 AgCu/TS-1

2+

– – 932.8 932.7

surface atomic ratio +

0

Cu

Ag

– – 930.9 930.4

– 368.6 – 368.4

Olat

Oads

Cu+/ Cu2+

Oads/ Olat

530.8 530.8 530.6 530.6

532.4 532.6 532.4 532.4

– – 0.261 0.577

0.641 1.307 1.1 1.574

with presence or absence of humidity (RH = 50%, RT), results were shown in Fig. 5 (SiO2 and HZSM-5 were commercial materials). SiO2 showed the lowest storage capacity. The lack of channel structure accounts for its poor property. With increase of Si/Al ratio of HZSM-5 zeolites from 25 to 200, benzene storage capacity enhanced. The excellent benzene adsorption ability of HZSM-5(200) catalysts derives from the hydrophobic property which could be attributed to the weaken of electrostatic field in the zeolite cage and the decrease of polarization ability of high-silica zeolite [35–38]. Due to the 4

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

0.08

0.06

Dry gas RH=50%

Storage capacity/(mmol/g)

C6H6 Storage Capacity/(mmol/g)

0.10

HZSM-5(200)

TS-1

) (25 HZ

0) ) (20 (50 HZ HZ

0.04

0.02

0.00 SiO

2

TS

-1

Ag

/TS

-1

Cu

/TS

-1

1

S-

u/T

C Ag

Fig. 5. The C6H6 storage capacity of HZSM-5, TS-1, SiO2 and supported TS-1 under humid and dry conditions.

temperature over the TS-1, Cu/TS-1, Ag/TS-1 and AgCu/TS-1 catalysts. The TS-1 sample shows the lowest C6H6 conversion in the temperature range investigated, the C6H6 conversion is only 16% at 420 °C. When doped Ag and Cu into the TS-1 sample, the C6H6 conversion increases monotonously with rise of reaction temperature from 150 to 250 °C, and complete oxidation of C6H6 occurs at 305 and 325 °C, respectively. In contrast, the 100% conversion over AgCu/TS-1 catalyst is obtained at 270 °C. Compared with the Ag/TS-1 and Cu/TS-1 catalysts, the remarkable catalytic behavior shown by AgCu/TS-1 may depends on forming more surface adsorbed oxygen species by the synergistic effect of Ag and Cu, on the basis of XPS results.

3.4. Comparison on thermal regeneration and plasma regeneration After C6H6 adsorption, the catalyst saturated with C6H6 was regenerated in situ either by thermal or NTPs in 21% O2/1.56% H2O/N2, results being shown in Fig. 7. During heating-up, obvious desorption of C6H6 was observed. It is notable that TS-1 zeolite exhibited a major C6H6 desorption peak at low temperature (120 °C). While on Ag and Cu modified TS-1 catalysts, besides this, there was another C6H6 desorption peak appeared at higher temperatures (263 °C). It is clear that C6H6 desorption at low temperature (LT) should come from the weak adsorption on TS-1 zeolite, while that at higher temperature (HT) is attributed to π-complexation of C6H6 with Ag, Cu and AgCu [10]. The higher portion of C6H6 desorption at HT than LT over AgCu/TS-1 catalysts indicates that C6H6 is apt to be stored on AgCu than on TS-1 zeolite. Meanwhile, there is no CO2 (g) detected over TS-1, indicating that all the stored C6H6 is desorbed before it can be oxidized into CO2.

Fig. 3. SEM images of TS-1 and AgCu/TS-1 samples.

vacant s orbital of the metal and the other is the back-donation from the d orbital of the metal to the vacant π* orbital of ligand [16,17]. 3.3. Benzene oxidation capacity Fig. 6 compares the C6H6 conversion to CO2 as a function of

Fig. 4. TEM images of Ag/TS-1, Cu/TS-1 and AgCu/TS-1 samples.

5

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

C 6H 6 Conversion to CO 2 / %

Fig. 6. Complete oxidation of C6H6 to CO2 over the TS-1, Ag/TS-1, Cu/TS-1 and AgCu/TS1.

AgCu/TS-1 Ag/TS-1 Cu/TS-1 TS-1

100

80

60

40

20

0 100

150

200

250

300

350

400

450

Temperature / º C release of C6H6 [18]. Herein, it is found that by employing NTPs instead of thermal regeneration, the stored benzene can be completely oxidized into to CO2 at room temperature without release, EC6H6 → CO2 is above 99%. This should be owing to the reactive and oxidative species such as O (3P) and OH generated in O2 plasma which oxidize the stored benzene to CO/CO2 at RT. This provides a new strategy to avoid the VOC release during thermal regeneration of the saturated materials. But it is unwilling to see that there was CO produced, during air-plasma regeneration, the least amount appeared over AgCu/TS-1 catalyst and the selectivity of CO2 reaches 85.6%. This should be ascribed to the better

Doping Ag, Cu and AgCu on TS-1 improves the catalytic oxidation activity, the largest amount of CO2 generated appeared over AgCu/TS-1 catalyst. But its regeneration efficiency (which is defined as Eq. 2) is only 16.9%, with large portion of the stored C6H6 being released during temperature-rising phase. From the above results, it is clear that C6H6 adsorbed on the materials should be strong enough to make sure that before oxidation it is still bonded to the catalysts instead of being released. Therefore, in our previous study, we employed Pt to catalyze the oxidation of the stored C6H6 and lower the oxidation temperature, and therefore minimize the

Fig. 7. COx evolution with time (temperature) in thermal regeneration (A); COx evolution with discharge time in plasma regeneration (B) over TS-1, Cu/TS-1, Ag/TS-1, AgCu/TS-1 catalysts.

6

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

100 dry air

80

thermal regeneration plasma regeneration

(%)

60 40

E

dry air

20 0

0

1 2 H2O concentration (%)

3

100

2

SCO /%

80

60

40

Fig. 9. Five times “cycled-storage-discharge” removal of benzene on AgCu/TS-1.

20

3.6. “Storage-Oxidation” cycles process 0

0.0

0.5

1.0

1.5

2.0

H2O concentration/%

2.5

3.0

Seeing that AgCu/TS-1 catalyst exhibited good performances both in storage process and the regeneration process, we conducted a consecutive cycling test over AgCu/TS-1 catalyst to examine the catalyst's stability and carbon balance during each cycle. As shown in Fig. 9, benzene was adsorbed and stored by AgCu/TS-1 catalyst, with no benzene release detected in the outlet gas. During the discharge phase, a large amount of CO2 was generated without the release of benzene. The conversion of stored benzene in the oxidation reaction was calculated to be over 99%. Moreover, no NOx was observed in the emission, implying that there is no secondary air pollutants generated during the plasma assisted regeneration process. Assume that benzene in simulated air is 10 times to the standard (0.026 ppm), to deal with simulated air containing 0.26 ppm benzene with 50% RH in the feed (100 ml min−1) in a “storage-oxidation” cycling process, the energy cost was as low as 2.36 × 10−4 kW·h·m−3 over 1 g AgCu/TS-1 catalyst.

Fig. 8. Effect of H2O concentration on EC6H6 (A) and SCO2 (B) over AgCu/TS-1 in regeneration.

oxidative ability of AgCu/TS-1 catalyst, which possesses the highest ratio of active surface adsorbed oxygen species, which are regarded to be essential for catalytic oxidation reactions [41]. In comparison with catalysis only regeneration, EC6H6 → COx of plasma regeneration is much higher (around 100%), indicating that plasma regeneration operated at room temperature solved the problem of benzene release during thermal regeneration.

3.5. Effect of relative humidity on oxidation of the adsorbed benzene To study the influence of relative humidity on the efficiency of stored-C6H6 conversion to COx (EC6H6 → COx ) and on CO2 selectivity SCO2 , H2O content in the feed gas was varied from 0 to 2.81%, results being shown in Fig. 8(A). It can be seen that the stored-benzene conversion to COx in thermal regeneration decreased from 16.9 to 1.2% with rise of H2O content from 0 to 2.81%. This is in consistent with our previous study that the presence of H2O is apt to enhance the release of C6H6 due to the competitive adsorption between C6H6 and H2O on the zeolite. In contrast, in air plasma regeneration, the efficiency of stored-C6H6 conversion to COx remains at 100% during the whole tested range. Plasma regeneration shows better humidity tolerance. As shown in Fig. 8(B), the CO2 selectivity remains above 85% for the entire H2O contents of the test (H2O concentration of 0∼2.81%). It is observed that the water concentration has little effect on the SCO2 over the catalysts.

3.7. Structure-activity correlations Correlated the characterization results with activity measurement, it may become easier to understand the performance of the catalysts. First of all, TS-1 zeolite itself could exhibit an excellent benzene storage capacity, which shows the highest storage capacity compared with the HZ-zeolite and SiO2 in the humidity condition. Next, it is found that doping AgCu into TS-1 zeolite could observably increase the C6H6 storage capacity. As shown in the XPS, the ratio of Cu+ is higher in the AgCu/TS-1 sample, indicating that co-existence of Ag is in favor of the reduction of Cu2+ to Cu+, and π-back-donation of benzene was strong with Cu+. Due to the unique (n-1)d10ns0 electron configuration, Cu+ or Ag is easier to accept electron and donate electron at the same time, which is apt to form more stable π-complexes [16,17,25]. Thus, this 7

Catalysis Today xxx (xxxx) xxx–xxx

Y. Wang et al.

[9] T. Tabakova, L. Ilieva, P. Petrova, A.M. Venezia, G. Avdeev, R. Zanella, Y. Karakirova, Chem. Eng. J. 260 (2015) 133–141. [10] Y. Wang, C. Dai, B. Chen, Y. Wang, C. Shi, X. Guo, Catal. Today 258 (2015) 616–626. [11] E. Moreau, J. Phys. D: Appl. Phys. 40 (2007) 605–636. [12] S. Yao, S. Weng, Y. Tang, C. Zhao, Z. Wu, X. Zhang, S. Yamamoto, S. Kodama, Vacuum 126 (2016) 16–23. [13] S.K. Wirawan, D. Creaser, Microporous Mesoporous Mater. 91 (2006) 196–205. [14] S.L. Zhang, L. Liu, Y. Fu, Q.X. Guo, J. Mol. Struc. Theochem 757 (2005) 37–46. [15] P. Chaquin, D. Costa, C. Lepetit, M. Che, J. Phys. Chem. A 105 (2001) 4541–4545. [16] A. Takahashi, F.H. Yang, R.T. Yang, Ind. Eng. Chem. Res. 39 (2000) 3856–3867. [17] H.Y. Huang, J. Padin, R.T. Yang, J. Phys. Chem. B 103 (1999) 3206–3212. [18] Y. Wang, S. Yao, M. Crocker, X. Zhu, B. Chen, J. Xie, C. Shi, D. Ma, Catal. Sci. Technol. 5 (2015) 4968–4972. [19] B. Bai, J. Li, ACS Catal. 4 (2014) 2753–2762. [20] H. Dong, G. Chen, J. Sun, C. Li, Y. Yu, D. Chen, Appl. Catal. B 134–135 (2013) 46–54. [21] Z. Huang, X. Gu, Q. Cao, P. Hu, J. Hao, J. Li, X. Tang, Angew. Chem. Int. Ed. Engl. 51 (2012) 4198–4203. [22] W. Tang, X. Wu, S. Li, X. Shan, G. Liu, Y. Chen, Appl. Catal. B 162 (2015) 110–121. [23] H.K. Wang, C.Y. Yi, L. Tian, W.J. Wang, J. Fang, J.H. Zhao, W.G. Shen, J. Nanomater. 2012 (2012) 8. [24] X. Zheng, Q. Zhang, Y. Guo, W. Zhan, Y. Guo, Y. Wang, G. Lu, J. Mol. Catal. A Chem. 357 (2012) 106–111. [25] J. Liu, X. Zhang, M.E.I. Dong, J. Wang, J. Theor. Comput. Chem. 10 (2011) 133–145. [26] A.S. Poyraz, J. Huang, S. Cheng, D.C. Bock, L. Wu, Y. Zhu, A.C. Marschilok, K.J. Takeuchi, E.S. Takeuchi, Green Chem. 18 (2016) 3414–3421. [27] X. Xu, X. Sun, H. Han, H. Peng, W. Liu, X. Peng, X. Wang, X. Yang, Appl. Surf. Sci. 355 (2015) 1254–1260. [28] S. Ameen, M.S. Akhtar, H.-K. Seo, Y.S. Kim, H.S. Shin, Chem. Eng. J. 187 (2012) 351–356. [29] Y. Li, C. Zhang, H. He, J. Zhang, M. Chen, Catal. Sci. Technol. 6 (2016) 2289–2295. [30] Y. Wang, A. Zhu, B. Chen, M. Crocker, C. Shi, Catal. Commun. 36 (2013) 52–57. [31] Y. Huang, L. Wei, Z. Crandall, J. Julson, Z. Gu, Fuel 150 (2015) 656–663. [32] N.V. Beznis, B.M. Weckhuysen, J.H. Bitter, Catal. Lett. 138 (2010) 14–22. [33] W. Liu, V.G. Ruiz, G.-X. Zhang, B. Santra, X. Ren, M. Scheffler, A. Tkatchenko, New J. Phys. 15 (2013) 053046. [34] F.A.K. Franc Meyer, P.B. Armentrou, J. Am. Chem. Soc. 117 (1995) 9740–9748. [35] D.Z. Zhao, X.S. Li, C. Shi, H.Y. Fan, A.M. Zhu, Chem. Eng. Sci. 66 (2011) 3922–3929. [36] Y. Liu, X.S. Li, J.L. Liu, J. Wu, D. Ye, A.M. Zhu, Catal. Sci. Technol. 6 (2016) 3788–3796. [37] S.W. Baek, J.R. Kim, S.K. Ihm, Catal. Today 93–95 (2004) 575–581. [38] H.Y. Fan, C. Shi, X.S. Li, D.Z. Zhao, Y. Xu, A.M. Zhu, J. Phys. D: Appl. Phys. 42 (2009) 225105. [39] W.S.W. Ho, G. Doyle, D.W. Savage, R.L. Pruett, Ind. Eng. Chem. Res. 27 (1988) 334–337. [40] R.B. Eldridge, Ind. Eng. Chem. Res. 32 (1993) 2208–2212. [41] C. Shi, Y. Wang, A. Zhu, B. Chen, C. Au, Catal. Commun. 28 (2012) 18–22.

should be the reason for the AgCu/TS-1 sample shows higher storage capacity than the other samples. In addition, comparing the surface adsorption oxygen species of AgCu with the others, there are higher concentration of surface adsorbed oxygen species in the AgCu/TS-1 sample, which might facilitate the activities of C6H6 oxidation to CO2. 4. Conclusions For the first time non-thermal plasma assisted “storage-oxidation” cycling process over AgCu/TS-1 catalyst was applied to indoor benzene removal at room temperature. TS-1 zeolite was found to be the optimal choice for benzene storage due to its channel structure and hydrophobicity. The addition of AgCu enhanced the storage capacities of the catalysts. The pulsed air plasma significantly increased the efficiency of benzene oxidized into COx, without benzene release during regeneration process, providing a possible way for indoor benzene purification, which operated at room temperature without using any precious metals, achieving excellent benzene removal efficiency. Acknowledgment The work was supported by the National Natural Foundation of China (Nos. 21373037, 21577013), and by China Postdoctoral Science Foundation of MOE (2014M560201) as well as by the Fundamental Research Funds for the Central Universities (No. DUT15TD49, DUT16RC(4)01 and DUT16ZD224). References [1] S. Li, H. Wang, W. Li, X. Wu, W. Tang, Y. Chen, Appl. Catal. B 166–167 (2015) 260–269. [2] H.B. Liu, B. Yang, N.D. Xue, J. Hazard. Mater. 318 (2016) 425–432. [3] S. Mo, S. Li, W. Li, J. Li, J. Chen, Y. Chen, J. Mater. Chem. A 4 (2016) 8113–8122. [4] A. Veksha, E. Sasaoka, M.A. Uddin, Carbon 47 (2009) 2371–2378. [5] M.A. Sidheswaran, H. Destaillats, D.P. Sullivan, S. Cohn, W.J. Fisk, Build. Environ. 47 (2012) 357–367. [6] F. Liu, S. Zuo, C. Wang, J. Li, F.S. Xiao, C. Qi, Appl. Catal. B 148–149 (2014) 106–113. [7] Y. Liu, H. Dai, J. Deng, S. Xie, H. Yang, W. Tan, W. Han, Y. Jiang, G. Guo, J. Catal. 309 (2014) 408–418. [8] Z. Wang, M. Yang, G. Shen, H. Liu, Y. Chen, Q. Wang, J. Nanopart. Res. 16 (2014).

8