Chemical Engineering Journal - Nanjing University...showed superior sulfur resistance to that of the...

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

The dual effects of ammonium bisulfate on the selective catalytic reductionof NO with NH3 over Fe2O3-WO3 catalyst confined in MCM-41

Kai Guoa, Yuxiang Zhua, Zhen Yanb, Annai Liua, Xiangze Duc, Xin Wanga, Wei Tana, Lulu Lid,Jingfang Sune, Qing Tonge, Changjin Tanga,e,⁎, Lin Donga,d,e,⁎

a School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023, ChinabDepartment of Biomedical Engineering, University of Science and Technology of China, Hefei 230031, Chinac School of Chemistry, Sichuan University, Chengdu 610000, Chinad School of Environment, Nanjing University, Nanjing 210023, Chinae Jiangsu Key Laboratory of Vehicle Emission Control, Center of Modern Analysis, Nanjing University, Nanjing 210023, China

H I G H L I G H T S

• The dual effects of ABS on NH3-SCRperformance of Fe2O3-WO3/MCM-41is observed and illustrated for the firsttime.

• The formation of surface metal sul-fates sustainably promotes the cata-lytic performance.

• Nearly 90% of sulfates in ABS areimmobilized after NH4+ consumption.

• Both the acidity and redox propertiesof catalyst are greatly improved aftermetal sulfates formation.

G R A P H I C A L A B S T R A C T

A R T I C L E I N F O

Keywords:NH3-SCRAmmonium bisulfateMetal sulfateFe2O3-WO3 catalystSulfur-resistant

A B S T R A C T

Ammonium bisulfate (ABS) has long been considered as a main poisoning material in the selective catalyticreduction by NH3 (NH3-SCR) due to the inevitable coverage of sticky ABS on catalytic active sites in sulfur-containing atmospheres, which severely hinders the achievement of stable and highly active NH3-SCR catalysts.In the present study, we report a novel observation of the dual effects of ABS on the NH3-SCR reaction. That is,ABS inhibits the NH3-SCR performance of Fe2O3-WO3/MCM-41 catalyst at low temperatures (50–200 °C) butshows apparent and sustainable reaction promotion when the temperature surpasses 250 °C. X-ray photoelectronspectroscopy (XPS) and NO probing adsorption confirmed that when ABS is deposited on the catalyst surface, apartial interaction between ABS and the Fe2O3-WO3 component occurs, resulting in blocking of the active sitesand an obvious loss of catalytic activity. With increasing reaction temperature, the ammonium in ABS can befacilely consumed by NO/O2, thus inducing the disintegration of poisoning species. The sulfate group transformsinto metal sulfate and survives at high temperatures. Importantly, owing to the strong inductive effect of sulfatespecies, both the acidity and redox properties of the catalyst are greatly improved, which contributes to theenhanced activity. The results of the present study expand our knowledge of the role of ABS in the NH3-SCRreaction and will be useful for designing high-performing NH3-SCR catalysts.

https://doi.org/10.1016/j.cej.2020.124271Received 4 October 2019; Received in revised form 25 January 2020; Accepted 28 January 2020

⁎ Corresponding authors at: School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023, China.E-mail addresses: [email protected] (C. Tang), [email protected] (L. Dong).

Chemical Engineering Journal 389 (2020) 124271

Available online 30 January 20201385-8947/ © 2020 Elsevier B.V. All rights reserved.

T

http://www.sciencedirect.com/science/journal/13858947https://www.elsevier.com/locate/cejhttps://doi.org/10.1016/j.cej.2020.124271https://doi.org/10.1016/j.cej.2020.124271mailto:[email protected]:[email protected]://doi.org/10.1016/j.cej.2020.124271http://crossmark.crossref.org/dialog/?doi=10.1016/j.cej.2020.124271&domain=pdf

1. Introduction

Nitric oxides (NOx), a major kind of air pollutant originated fromburning fossil fuel, have gained much attention due to the increasinglyharmful impacts on the environment and human health [1–3]. Foryears, the most effective technology for NO emission control has beenselective catalytic reduction by NH3 (NH3-SCR) [4]. Although the V2O5-WO3/TiO2 catalyst has been used commercially for decades, its low N2selectivity at high temperature, narrow reaction temperature windowand toxicity of vanadia are still not satisfactory [5]. Therefore, thedevelopment of new catalysts with high activity at low temperature isurgently demanded. By far, there are numerous examples of transitionmetal and rare earth (e.g. Mn, Fe, Ti, Cu, Ce and Sm) oxide catalystsexhibit excellent activity between 80 and 300 °C under ideal reactionconditions [6–8]. Nevertheless, due to the presence of SO2 in flue gas,these catalysts are exceptionally susceptible to deactivation at low-temperature operation. The mixture of SO2 with NH3 and H2O couldeasily produce NH4HSO4 (ammonium bisulfate, ABS) with the assis-tance of the NH3-SCR catalyst, which barricades the active sites fromreactants adsorbing onto the catalyst surface, thus deactivates the NH3-SCR catalysts [9]. Thus, the deposition of ABS is always regarded as themain poisoning factor when the catalysts are operated in flue gas withSO2 at low temperature.

In recent years, the increasing attention has been devoted to un-derstand the behavior of ABS on the catalyst surface to alleviate thepoisoning effect for developing novel sulfur-resistant NH3-SCR cata-lysts. Zhu et al. found that the ABS could react with gaseous NO and itsdecomposition was promoted by activated carbon (AC) due to the re-ducible property of carbon [10]. A recent study by Chen at al. showedthat decomposition of deposited ABS could be promoted by inter-calating NH4+ into the layer structure of MoO3 [11]. Wang et al. re-ported that the V2O5-WO3/TiO2 catalyst deactivated by in situ depositedABS could be rapidly recovered by the reaction between NH4+ and aNO/NO2/O2 mixture [12]. Moreover, the pore structure of the catalystis also shown to have a significant impact on the thermal behavior ofABS [13]. In our previous study, we have reported that the decom-position temperature of ABS could be simply tailored by adjusting thepore size of the support. The model catalyst with larger pores certainlyshowed superior sulfur resistance to that of the catalyst with smallerpores, which confirmed that the effect of pore expansion could accel-erate ABS decomposition [14,15]. The studies shown above providevarious effective strategies to promote the decomposition and reactivityof ABS thus alleviating the deactivation effects. On the other hand, thesulfation of metal components, which usually occurs simultaneouslywith the formation of ABS, also needs to be considered. It reported thatthe sulfation of ceria-based catalysts prevents the redox cycle of Ce3+

and Ce4+, thus playing a negative role in the NH3-SCR reaction [16].However, there are also reports showing the metal oxide sulfation canenhance the Brønsted acidity of the catalyst, which in turn promotes NOreduction efficiency to a large extent [17–19]. The seemingly con-troversial phenomenon reminds us that the sulfation of metal compo-nents may not always be a disadvantage, and the unique properties ofmetal sulfates might be utilized in sulfur-resistant catalyst develop-ment.

It is well acknowledged that the ABS deposition is unavoidable atlow temperatures and causes deactivation of NH3-SCR catalyst, thus theinteraction between ABS and the catalyst has become a primary ques-tion that needs to be addressed. Ye et al. reported that after the de-position of ABS on the V2O5/TiO2 catalyst surface, HSO4− transformedinto bidentate SO42−, and an increase in ABS deposition led to forma-tion of the (NH4)2TiO(SO4)2, which diversely affected the ABS decom-position and reactivity behaviors [20]. Unfortunately, this kind of in-teraction strongly affects NH3-SCR performance but has yet beenillustrated so far. Thus, in this work, ABS is intentionally deposited onan environment friendly Fe2O3-WO3/MCM-41 catalyst to investigatethe effect of ABS on NH3-SCR performance [21]. Because of that in our

previous report, Fe2O3 supported mesoporous silica catalyst exhibitsgood stability in presence of SO2 due to its ABS decomposition pro-motion effect [14]. However, the deficiency in surface acidity makesthe catalyst activity is not that promising. As a typical kind of acidoxide, WO3 is widely used as a promoter in NH3-SCR, which couldprovide high thermal stability and sufficient acidity. Hence, WO3 isadded to compensate the weak acidity of FeOx and construct Fe2O3-WO3/MCM-41 catalyst. Interestingly, the dual effects of ABS on deNOxefficiency, in which ABS slightly inhibited the catalyst performance attemperatures lower than 250 °C, but largely promoted the reactionabove 250 °C, are discovered for the first time. The real-time behaviorof ABS on the catalyst is evaluated by temperature-programmed surfacereaction (TPSR) and inductively coupled plasma (ICP) analysis. Thefresh and used catalysts are characterized by X-ray photoelectronspectroscopy (XPS), temperature-programmed desorption of ammonia(NH3-TPD), temperature-programmed reduction by hydrogen (H2-TPR), Fourier transform infrared (FT-IR) spectroscopy, and thermo-gravimetric (TG) analysis to illustrate how ABS interacts with catalysts.Finally, the reaction mechanisms of fresh and used samples are studiedby in situ diffuse reflectance infrared Fourier transform spectroscopy (insitu DRIFTS). The results of this work might provide new insights on theregeneration of ABS deactivated NH3-SCR catalysts, which will benefitthe development of long-lived catalysts for practical applications.

2. Experiments

2.1. Sample preparation

MCM-41 was purchased from the supplier and calcined at 750 °C for5 h before use. Fe2O3-WO3 oxides supported on MCM-41 were syn-thesized by a conventional wetness impregnation method. In a typicalpreparation, calculated amounts of Fe(NO3)3·9H2O (0.257 g) and(NH4)10W12O41 (0.052 g, nFe:nW = 3:1) were dissolved in 30 ml dis-tilled water. MCM-41 (0.5 g) was added, and the mixture was stirred for2 h. After evaporating excess water at 80 °C oil bath, the sample wasdried at 100 °C overnight and calcined at 450 °C for 3 h. The obtainedcatalyst was labeled as FeW/MCM-41.

Deposition of ammonium bisulfate (ABS) was performed by thewetness impregnation method. Typically, 10 wt% ABS was dissolvedand mixed with FeW/MCM-41, and the suspension was vigorouslystirred for 2 h. Then, excess water was evaporated, and the collectedsample was further dried overnight at 100 °C. For simplicity, the cat-alyst was denoted as ABS/FeW/MCM-41

2.2. Sample characterizations

The specific surface area and pore size distribution of the sampleswere measured by N2 physisorption technique at −196 °C on aMicromeritics ASAP-3020 analyzer. Before measurement, the samplewas vacuum-pretreated at 300 °C (120 °C for ABS deposited samples)for 3 h and pore size distribution results were obtained by BJH (Barrett-Joyner-Halenda) method. XPS measurements were conducted using aPHI 5000 Versa Probe system with a monochromatic Al Kα radiation(1486.6 eV, 15 kW). All binding energies were calibrated relative to theadventitious C 1s (284.6 eV) to compensate for surface charge effects.FT-IR spectra of the samples were recorded on a Nicolet Is10 FT-IRspectrometer in ATR pattern from 400 to 4000 cm−1. The backgroundis collected and subtracted for every sample. The number of scans was32 at resolution of 4 cm−1. Inductively coupled plasma atomic emissionspectroscopy (ICP-AES) was used to investigate the soluble species onthe sample surface by using a Perkin Elmer Optima 5300DV instrumentwith a radiofrequency power of 1300 W. The sample is dispersed in20 ml deionized water and treated by ultrasonic for 2 h, then the so-lution is filtered, liquid phase is kept for ICP tests. The concentrations ofanions in leached solutions were measured by Ion Chromatography (IC,DIONEX ICS-1100AR, Thermofisher scientific, India). The system

K. Guo, et al. Chemical Engineering Journal 389 (2020) 124271

2

consisted of one IC pump, an injection valve, column heater, an ion-exchange analytical column (AS19), and a guard column (AG19).Samples were loaded onto a 100 μL sample loop with a syringe. Allcolumns, tubing, and fittings were made of poly-ether-ether-ketone.During the IC measurement, 25 mM KOH eluent, 0.38 ml/min flow rate,and 24 mA applied current were maintained.

H2-TPR was carried out in a quartz U-tube reactor connected to athermal conductivity detector (TCD) with a H2/Ar mixture (7% H2 byvolume) as a reductant. 20 mg of each sample is used for measure-ments. Before introducing the sample to the H2/Ar stream, the samplewas pretreated in N2 stream at 200 °C for 1 h. The H2 consumptionprofile was collected from room temperature to 900 °C at a rate of10 °C/min. The acidic properties of the catalysts were determined fromNH3-TPD by using a Micromeritics Autochem II 2920 instrument.Sample was pretreated by helium (25 ml/min) at 200 °C for 1 h, thencooled to 100 °C and exposed to 10% NH3/He (25 ml/min) for 4 h.Thereafter, the sample was purged with helium at 100 °C for 1 h. AllNH3-TPD tests were carried out by increasing the temperature from 100to 800 °C at a rate of 10 °C/min. The signal of released NH3 was ana-lyzed by a TCD. TPSR was carried out with an activity test system.Briefly, 100 mg sample was placed in a quartz tube and heated to450 °C at a heating rate of 2 °C/min with blew of 100 ml/min 500 ppmNO/Ar flow. In situ DRIFTS was performed on a Nicolet Nexus 5700 FT-IR spectrometer equipped with a highly sensitive mercury cadmiumtelluride detector with 32 scans at a resolution of 4 cm−1, and a diffuse

reflectance reaction cell (HARRICK) equipped with KBr window wasused. The sample was pretreated in a flowing N2 stream at 400 °C for1 h to eliminate physisorbed water and other impurities, and the samplebackground was collected at various target temperatures during thecooling process. For the experiments of NH3 or NO + O2 ad-sorption–desorption, the reaction cell with sample was saturated withNH3/N2 (1% NH3 by volume) or NO/N2 + O2/N2 (500 ppm NO and 5%O2 by volume) for steady state at room temperature, then the samplewas purged by N2 for 30 min and the spectra were recorded at targettemperatures. The decomposition behaviors of various samples weremeasured by TG-DTA (Thermogravimetry-differential thermal ana-lysis). The measurement was carried out in flowing of nitrogen atmo-sphere (60 ml/min) on a NETZSCH STA-449-F5 and the heating rate isset at 5 °C/min. The reference curve for DTA was generated frommeasurements of an empty crucible.

2.3. Catalytic activity test

The NH3-SCR performance of the catalysts was evaluated in a fixed-bed quartz reactor (8 mm internal diameter). The feed stream is fixedwith 500 ppm NO, 500 ppm NH3, 5% O2, and Ar in balance with a gasflow of 200 ml/min (Weight hourly space velocity, WHSV, was60000 ml·g−1·h−1, Gas hourly space velocity, GHSV, was approxi-mately 30000 h−1 because the density of the sample varied). The cat-alyst (200 mg) was sieved to 40–60 mesh. Before the test, the sample

Fig. 1. (a) NO conversion, (b) N2 selectivity of FeW/MCM-41, ABS/FeW/MCM-41 and ABS/FeW/MCM-41-U catalysts and (c) stability test of ABS/FeW/MCM-41-Uat 350 °C. Reaction conditions: 500 ppm NO, 500 ppm NH3, 5% O2, WSHV of 60000 ml·g−1·h−1.


3

was pretreated in a purified Ar stream at 150 °C for 1 h to avoidingsurface impurities. Then, the mixed gases were switched on, and theactivity data were collected at every target temperature after stabilizingfor 30 min in the range of 50–400 °C. The concentration of effluentgases was continuously analyzed on an IS10 FT-IR spectrometerequipped with a 2 m path-length gas cell (2 L volume). The NO con-version and N2 selectivity were calculated based on the followingequations:

=−

×NO NO NONO

conversion(%) [ ] [ ][ ]

100%in outin (1)

=− + − − −

− + −

×

NNO NO NH NH NO N O

NO NO NH NH

selectivity(%)[ ] [ ] [ ] [ ] [ ] 2[ ]

[ ] [ ] [ ] [ ]

100%

in out in out out out

in out in out

2

3 3 2 2

3 3

(2)

3. Results and discussion

3.1. Reaction features of fresh and ABS-deposited FeW/MCM-41 in NH3-SCR

The influence of ABS deposition on the catalytic performance ofFeW/MCM-41 in NH3-SCR is displayed in Fig. 1. In general, the freshFeW/MCM-41 catalyst shows poor NO conversion efficiency, particu-larly at low temperatures. It is clear from Fig. 1a that no activity isdetected for FeW/MCM-41 until 200 °C and the NO conversion reachesa saturated value at 78% when the reaction temperature rises to 350 °C.The absence of a consecutive activity increment at high temperatures isassumed to be related to the competitive oxidation of ammonia fromoxygen, which reduces the availability of NH3 for NO reduction [22].This conclusion is supported by the decreasing N2 selectivity at hightemperatures (Fig. 1b).

After ABS deposition, the reaction characteristic of the catalyst isnot greatly changed. The activity of ABS/FeW/MCM-41 increases from200 °C, and a plateau is observed at 350 °C, which is the same as that ofFeW/MCM-41. Nevertheless, by comparing the detailed NO conversionsbetween the two samples, differences can be clearly discriminated. It isfound that at reaction temperature lower than 250 °C, the efficiency ofABS-deposited sample is obviously restrained, demonstrating the ap-parent inhibition effect caused by ABS deposition. This finding con-forms to the general impact of ABS on NH3-SCR catalysts [23]. How-ever, the activity order is reversed when the reaction temperaturesurpasses 250 °C. The activity of the ABS-deposited sample reveals a NOconversion much higher than that of the fresh sample, with almost100% NO conversion achieved in the range of 350–400 °C. This phe-nomenon is peculiar and has yet been reported. This also allow us toutilize the poisoning species (ABS) to promote NH3-SCR performance.Intuitively, by considering the limited thermal stability of ABS at ele-vated temperatures [24], we expect unique temperature-dependentreaction behavior that can be related to composition/structure mod-ification of ABS/FeW/MCM-41 during the reaction. To simply confirmthis hypothesis, another round of activity test was performed, and theused catalyst was denoted as ABS/FeW/MCM-41-U. For ABS/FeW/MCM-41-U, the reaction curve at temperatures higher than 250 °C es-sentially overlaps with that of ABS/FeW/MCM-41. Combining the sta-bility test at 350 °C in Fig. 1c, the results both suggest that the

promotion effect is sustainable. Moreover, the used catalyst exhibitspromising sulfur resistance (Fig. S1). The activity profile in the low-temperature range (50–250 °C) is no longer the same as that of ABS/FeW/MCM-41, implying that composition or structural variation of thecatalyst indeed occurred.

The above results clearly show that deposition of ABS on the cata-lyst surface, which is usually considered detrimental to NH3-SCR ap-plication, may not always be negative to the reaction. For FeW/MCM-41, a distinct dual effect is exhibited. That is, ABS degrades the catalystperformance at low temperatures but largely enhances the NO con-version at higher temperatures. Additionally, the repeated activity testsuggests that modification of the catalyst composition or structure oc-curs around the breakpoint of 250 °C. To obtain more details about thechanges of catalysts and the nature of the interaction between ABS andthe active component, a series of characterizations was carried out.

3.2. The dispersion of surface species

Table 1 summarizes the textual properties of various FeW/MCM-41catalysts. The pore sizes are comparable for different samples. How-ever, the surface area and pore volume are largely changed. For pristineMCM-41, the surface area and pore volume are 1025 m2/g and0.93 m3/g, respectively. After the introduction of FeW oxides, theparameters decrease to 845 m2/g and 0.69 m3/g, and decreases of thesurface area and pore volume are observed when ABS is deposited.These results suggest that most ABS and active components are insertedinto the pore channels of MCM-41 resulting in the decrease in surfacearea and pore volume.

The X-ray diffraction (XRD) patterns of fresh and ABS-depositedFeW/MCM-41 samples are shown in Fig. S2. Only a broad peak relatedto amorphous SiO2 can be observed [25], demonstrating well dispersionof the guest species (Fe2O3-WO3 and ABS) on MCM-41. This result isfurther supported by the transmission electron microscopy (TEM)images (Fig. S3), since no bulk particles are found on the surface ofmesoporous silica framework for both fresh and used samples. Theseresults collectively indicate that the deviation of catalytic performancecannot be attributed to variation in surface dispersion of active com-ponents.

3.3. The chemical states of surface species

To explore the chemical states of surface species, XPS character-ization was performed, and the results are shown in Fig. 2. For FeW/MCM-41, the binding energies at 35.9 eV and 38.0 eV are characteristicof W species in the +6 oxidation state [26,27]. For Fe species, thesignal at 711.8 eV and 725.7 eV, indicates the presence of Fe3+ [28].After ABS deposition, perturbation of Fe and W signals is observed,demonstrating that certain chemical interactions between ABS and FeWcomponents. In particular, important information can be distilled byexamining the peak shifts. The W 4f 5/2 signal deviates approximately0.3 eV to the lower binding energy after ABS deposition, suggesting anincreased electron cloud density of W atoms upon interaction with ABS.Thus, it is deduced that W6+ is dominantly bonded with NH4+ whichexhibits an electron donation effect. This conclusion is reasonable be-cause WO3 is well known as an acidic oxide and can easily bond withNH4+ [29]. The Fe 2p signal shifts to a higher binding energy sug-gesting Fe species mainly interact with sulfate group due to the ap-parent electron inductive effect of the S = O covalent bond. Thisfinding is consistent with literature report indicating that when Fespecies are sulfated, an obvious shift in the Fe XPS spectra to higherbinding energy is observed [19].

Based on the XPS results, we propose that the ammonium and sul-fate groups in ABS show preferential contact with Fe-W component inFeW/MCM-41. That is, the majority of NH4+ interacts with WO3 whilethe remaining sulfates mainly deposit around the Fe species. Moreover,by inspecting the XPS spectra of ABS-deposited samples after the

Table 1The textual properties of MCM-41 and various catalysts.

Sample SBET (m2/g) Pore volume (cm3/g) Pore size (nm)

MCM-41 1025 0.93 2.85FeW/MCM-41 845 0.69 2.74ABS/FeW/MCM-41 634 0.54 2.75


4

reaction (ABS/FeW/MCM-41-U), it is shown that the W 4f peak shiftsback to a high binding energy, even higher than that of pristine FeW/MCM-41. Considering the strong electron inductive effect of sulfategroup, we suppose that the NH4+ connected to WO3 is replaced bySO42− after the NH3-SCR reaction, which may be due to NH4+ con-sumption. Further evidence for this conclusion can be found from theTPSR and FT-IR results described in the following sections.

3.4. Evolution of ABS during the reaction

To disclose the ABS evolution during the reaction, the TPSR of ABS/FeW/MCM-41 with NO + O2 was carried out, and the result is presentin Fig. 3. After initial saturated adsorption, the level of NO becomesconstant (500 ppm). The outlet NO concentration starts to decreaseonly when the temperature exceeds 200 °C, which also suggests that theNH4+ in ABS is unreactive at low temperatures. The largest NO con-sumption occurs at approximately 300 °C. Since the adsorption ofNO + O2 is saturated, the decrease in NO concentration above 200 °Creflects the chemical reaction between deposited ammonium ions andNO/O2 [10,30]. Moreover, the recovery of NO concentration to thesaturated value (500 ppm) near 400 °C suggests that most of the am-monium ions are consumed during the reaction. On the other hand, it isclear from Fig. 3 that no obvious SO2 evolves, implying that the sulfatesare thermostable on the catalyst surface during the reaction.

FT-IR is sensitive to variations of sulfate and ammonium ions on the

catalyst surface and is thus a suitable tool to explore the surface species.Fig. 4a is the FT-IR spectra of various catalysts. Typically, bands at1047 cm−1, 958 cm−1 and 795 cm−1 are observed and attributed tovibrations of silica framework from MCM-41 [31,32]. After ABS de-position, the NH4+ signal (1434 cm−1) appears [33]. For the reactedcatalyst, this band is still present, but the intensity is much weaker,implying that ammonium ions on the catalyst surface are largely con-sumed, which supports the TPSR result.

It is noted that the signals of sulfate-containing species are maskedby the Si-O vibrations from MCM-41 [14,31,34]. To discriminate thesignals, differential IR spectra obtained by subtracting the spectrum ofFeW/MCM-41 are depicted and displayed in Fig. 4b. In the differentialspectra, the difference between ammonium ions (1436 cm−1) in ABS/FeW/MCM-41 and ABS/FeW/MCM-41-U can still be observed. More-over, the signals attributed to sulfate species appear. The bands at960 cm−1 and 1096 cm−1 represent the asymmetrical stretching ofSeO, and the band at 1174 cm−1 is attributed to the symmetricalstretching of the S]O bond [12,20]. An additional band at 1040 cm−1

can be assigned to the v3 vibration of bidentate SO42− [35]. The in-tensity of these bands of two samples is comparable, suggesting thatmost sulfur-containing species are preserved on the catalyst surface,most likely in the form of metal sulfates by considering its relativelyhigh thermal stability [19]. Apart from FT-IR, XPS spectra of N 1 s and S2p in Fig. S4 also reveal the same conclusion.

The possible transformation of ABS to metal sulfates is furtherverified by thermal analysis by taking advantage of the distinctly dif-ferent thermal stabilities of ABS and metal sulfates. Fig. 5 shows the TGprofiles of the three catalysts. For FeW/MCM-41, the major weight lossthat occurred at step 1 (40–200 °C) is attributed to the release of phy-sically adsorbed H2O on the catalyst surface. For ABS/FeW/MCM-41,three weight loss steps can be detected. The signal attributed to ad-sorbed H2O is enhanced, indicating the presence of ABS can promotewater adsorption on catalyst surface. The process of ABS decompositionis represented by a notable weight loss in step 2 (200–450 °C). In ad-dition, the obvious weight loss in step 3 (450–800 °C) is attributed tothe decomposition of metal sulfates [36,37], indicating partial reactionbetween deposited ABS and metal oxide during the heating process[38,39]. The remarkable weight loss of ABS/FeW/MCM-1-U in step 2 ismuch less than that of ABS/FeW/MCM-41, which further supports theconsumption of ammonium ions during the NH3-SCR reaction process.Furthermore, the signal represents weight loss of metal sulfates in-creases dramatically, implying that most of the deposited ABS has beentransformed into metal sulfates.

From the TG results, we recognize that most sulfate groups arebonded with ammonium ions (in the form of ABS) in pristine ABS/FeW/

Fig. 2. W 4f and Fe 2p XPS spectra for various samples.

Fig. 3. The TPSR profile and evolved SO2 signal of ABS/FeW/MCM-41 in thestream of 500 ppm NO and 5% O2.


5

MCM-41, then they evolve as metal sulfates after a cycle of the NH3-SCRreaction. Nevertheless, the key information regarding the quantitativeaspect of the various sulfate species (ABS and metal sulfates) is stillunder discover. To elucidate the variation in the surface species duringABS evolution, the ABS/FeW/MCM-41 sample was treated at

designated temperatures (200, 300 and 400 °C), and a leaching processin distilled water was the performed to obtain solutions containingsoluble surface species. The extent of metal sulfation can be estimatedfrom the solution by measuring the soluble metal ions. Fig. 6 shows theresult of the detected concentration of Fe3+. The solution from pristineABS/FeW/MCM-41 contains minor Fe3+ species (4.57 mg/L), verifyingthe partial chemical interaction between the deposited ABS and Fe2O3-WO3 components. The concentration of Fe3+ reaches 6.68 mg/L for thesample treated at 200 °C. This value is close to the data for pristineABS/FeW/MCM-41, implying that the majority of deposited ABS isstable enough at this temperature and that the formation of metalsulfate does not vastly occur. For the sample treated at 300 °C, it isreasonable to deduce that the sulfate species are largely combined withiron oxides, resulting in a drastic increase in the Fe3+ concentration to18.6 mg/L. The residual sulfate species in deposited ABS can con-tinually react with metal oxide at the 400 °C, finally leading to23.1 mg/L Fe3+ in the solution. In addition to the concentration ofFe3+, the concentration of SO42− in solutions is also analyzed and re-sults are depicted in Fig. 6. The SO42− from the samples treated below300 °C showed nearly identical values and only a small decrease wasobserved for the sample treated at 400 °C, indicating the negligible lossof sulfates from the catalyst during the heating process. From theleaching results, we can quantitatively conclude that 90% of sulfates inABS/FeW/MCM-41 remain as metal sulfates on the catalyst surfaceafter a cycle of activity testing.

Base on the characterizations shown above, the existence states andevolution of ABS on the FeW/MCM-41 surface under reaction condi-tions are clarified. As shown in Scheme 1, when ABS is deposited, itshows a strong chemical interaction with the active Fe2O3-WO3 com-ponent. The ammonium ions are partially bonded with WO3 and sulfategroups are connected with Fe2O3 species. Due to the combined effectsof physical coverage and chemical interaction, the adsorption proper-ties of FeW/MCM-41 are greatly restrained (Fig. S5), leading to obviousactivity loss. However, when the reaction temperature increases to200 °C, the chemical reaction between the deposited ammonium ionswith NO + O2 begins. With the depletion of ammonium ions, the WO3component is liberated and promotes the chemical interaction betweenWO3 and SO42−, which induces a significant shift of the W 4f signal to ahigher binding energy. From the TG results, we know that in compar-ison with ABS, the metal sulfates are more thermally stable. Hence,most of the sulfate groups can be stored on the catalyst surface in theform of metal sulfates after one cycle of reaction.

The inhibitory effect on the catalytic performance of the FeW/MCM-41 catalyst at low temperatures can be well explained by the coverageof ABS on the FeW components that remarkably reduce the exposed

Fig. 4. (a) FT-IR spectra of FeW/MCM-41, ABS/FeW/MCM-41 and ABS/FeW/MCM-41-U. (b) Differential spectra of ABS/FeW/MCM-41 and ABS/FeW/MCM-41-U.

Fig. 5. TG profiles of FeW/MCM-41, ABS/FeW/MCM-41 and ABS/FeW/MCM-41-U.

Fig. 6. Concentrations of Fe3+ and SO42− in the leached solution of ABS/FeW/MCM-41 treated at different temperatures.


6

active sites to reactants (Fig. S5) and the inertness of NH4+ in ABS.However, the promotion effect at high temperatures is still not wellunderstood. Thus, the acid and redox properties of the catalyst arecharacterized.

3.5. Surface acidity of catalysts (NH3-TPD and NH3-DRIFTS)

Fig. 7a displays the NH3-TPD profiles of FeW/MCM-41 and ABS/FeW/MCM-41-U. For FeW/MCM-41, three well resolved peaks at173 °C, 350 °C and 469 °C are detected, demonstrating the simultaneouspossession of weak, medium and strong acid sites on the catalyst surface[21,40]. Moreover, the comparable intensity of the three peaks suggeststhat these different acid sites are evenly distributed. After ABS de-position and a cycle activity test, the intensity of the desorption peakshows an obvious increase, revealing that the amount of NH3 adsorp-tion on the ABS/FeW/MCM-41-U catalyst is significantly increased. Inparticular, in comparison with the signals at low and medium tem-peratures, the desorption peak at 433 °C, typical of a strong acid,

exhibits a sharp increase. The signal at temperatures above 600 °C, isabnormal, which is attributed to the evolution of SO2 caused by metalsulfate decomposition, which has been proven by the TG results. Fromthe results above, it is obvious that the presence of metal sulfates de-rived from the interaction between the deposited ABS and metal oxidecomponents can significantly enhance the catalyst acidity. The im-proved surface acid property of ABS/FeW/MCM-41-U catalyst shouldmostly account for the promoted NH3-SCR activity at high tempera-tures.

To ascertain the nature of the acid sites on the catalyst surface, theDRIFT spectra of NH3 adsorption are shown in Fig. 7b. The bands at1638 cm−1 and 1478 cm−1 correspond to NH3 adsorbed on Brønstedacid sites [41], while the bands at 1610 cm−1 and 1276 cm−1 are at-tributed to NH3 adsorbed on Lewis acid sites [42,43]. The strongestpeak observed at 1478 cm−1 suggests that both catalysts are dominatedby Brønsted acid sites. Correlating with the NH3-TPD results, it issupposed that the strong acid sites are represented by Brønsted acid.Importantly, the distinct increase in both Brønsted and Lewis acid sites

Scheme 1. Schematic illustration of the evolution of ABS on the Fe2O3-WO3/MCM-41 catalyst at different temperatures.

Fig. 7. (a) NH3-TPD profiles of FeW/MCM-41 and ABS/FeW/MCM-41-U. (b) NH3 adsorption DRIFTS spectra of FeW/MCM-41 and ABS/FeW/MCM-41-U.


7

on ABS/FeW/MCM-41-U compared to those on FeW/MCM-41 is shown.According to previous studies, the increase in Brønsted acid sites isthought to originate from the generation of metal sulfates, while theincrease in Lewis acid sites may result from the electron-withdrawingeffect of sulfates, making the metal atoms more affable to lone pairelectrons in gaseous NH3 [44]. The obtained results indicate that thereare still both Lewis and Brønsted acid sites on the surfaces of fresh andused samples. Furthermore, the strength of NH3 chemisorption on theacid sites of ABS/FeW/MCM-41-U sample is reinforced as well, asevidenced by the in situ DRIFT results shown in Fig. S6. It is clear thatthe NH3 molecules can still adsorb on the ABS/FeW/MCM-41-U surfaceat temperatures higher than 400 °C. On the other hand, for FeW/MCM-41, the NH3 completely desorbs from the acid sites at approximately250 °C. The stable adsorption of NH3 at high temperatures is believed tobe conducive to continuous and effective reduction of NO.

3.6. Redox properties of catalysts (H2-TPR, NH3 and NO oxidation,NO + O2 DRIFTS)

H2-TPR characterization was performed to obtain the redox profilesof the catalysts, and the results are given in Fig. S7. Unfortunately, thereduction profile of ABS/FeW/MCM-41-U is dominated by H2 con-sumption at 507 °C from sulfate species [17], which covers the signalfrom the active component. To avoid this issue, NO and NH3 oxidationtests were performed to evaluate the redox properties of the catalysts,and the results are depicted in Fig. 8. For FeW/MCM-41, negligible NOoxidation (< 1%) is observed over the whole test range (50–400 °C),suggesting the poor efficiency of FeW/MCM-41 in converting NO toNO2. This explains the poor NH3-SCR performance at low temperatures,since the absence of NO2 could not promote the reaction via “fast SCR”pathway. Additionally, the poor redox property is reflected by the in-ferior NH3 oxidation performance at low temperatures. Compared toFeW/MCM-41, ABS/FeW/MCM-41-U achieves better NO and NH3oxidation performances, demonstrating that the redox properties ofFeW/MCM-41 can be enhanced by ABS deposition and a subsequentNH3-SCR reaction [45].

The improved redox property of ABS/FeW/MCM-41-U can be fur-ther evidenced by the enhanced adsorption of NO + O2 on catalystsurface. Fig. 9 shows the DRIFT spectra of NOx adsorption over FeW/MCM-41 and ABS/FeW/MCM-41-U catalysts. For FeW/MCM-41, whenNO+ O2 mixed gases are introduced, several bands appear in the rangeof 1200–1800 cm−1. The bands at 1675 and 1607 cm−1 can be at-tributed to cis-HNO2 and NO2 species [46,47], and the bands at 1625and 1575 cm−1 are assigned to different vibration modes of the brid-ging bidentate nitrate [40,48,49]. In comparison, the signal is obviously

enhanced for ABS/FeW/MCM-41-U. Moreover, the configurations ofabsorbed species are more complex than those of FeW/MCM-41 cata-lyst. New bands at 1740 and 1287 cm−1 assigned to N2O4 and nitrocompounds are present. These phenomena might originate from theinductive effect of the S]O bond in the formed metal sulfate resultingin a much stronger affinity and leading to an enhancement of theelectrophilicity of iron and tungsten species that provide more ad-sorption sites for NO + O2 [50].

3.7. Reaction mechanism of NH3-SCR over fresh and ABS depositedcatalysts

To determine whether the formation of surface sulfate species onthe catalysts alters the SCR reaction pathway, a transient reaction wasperformed by in situ DRIFT. For FeW/MCM-41, after NH3 adsorption for1 h at 250 °C, the weak peak at 1451 cm−1 assigned to the NH4+

species appears (Fig. 10a). When NO + O2 is introduced, NH4+ speciesare consumed, accompanied by the gradual augmentation of the signalattributed to gaseous NO vibrations (1600 and 1628 cm−1). Subse-quently, NH3 was injected into the catalyst with pre-adsorbed NO+ O2,and the resulting spectra are shown in Fig. 10b. After the adsorption ofNO + O2, bands assigned to various vibration modes of nitrates at1577, 1524 and 1349 cm−1 emerge [46,51]. However, along with theappearance of the NH4+ vibration bands, the intensity of nitrate bandsis hardly affected, indicating that the chemisorbed nitrate species is

Fig. 8. (a) NO oxidation and (b) NH3 oxidation tests of FeW/MCM-41 and ABS/FeW/MCM-41-U.

Fig. 9. The DRIFT spectra of NO + O2 adsorption over FeW/MCM-41 and ABS/FeW/MCM-41-U catalysts.


8

inert to gaseous ammonia or surface ammonium species. Therefore, itcan be deduced that the SCR reaction simply proceeded via the Eley-Rideal (E-R) mechanism over the FeW/MCM-41 catalyst.

As for the ABS/FeW/MCM-41-U catalyst, the spectra of NH3 in-troduced into the catalyst with NO + O2 pre-adsorption are illustratedin Fig. 10c. No infrared bands attributed to nitrate species are detected,implying that although nitrate species are abundant at low tempera-tures, they are not stable enough to stay on the catalyst surface at250 °C. Since the absorbed nitrates are inert, the absence of nitrates onABS/FeW/MCM-41-U at 250 °C is expected to provide more adsorptionsites for NH3, which promotes the activation of ammonia and enhancesthe NO removal efficiency. After NH3 injection, only the IR band at1451 cm−1 assigned to NH4+ appears. Fig. 10d shows the in situ DRIFTexperiment of the reaction between NO + O2 and pre-adsorbed NH3species. After NH3 adsorption, an intense band ascribed to ionic NH4+

at 1451 cm−1 shows up, suggesting that only NH3 adsorbed onBrønsted acid sites is stable enough on the catalyst surface, similar tothe result of the FeW/MCM-41 sample. The subsequent introduction ofNO + O2 results in an evident diminishing of the NH4+ IR band and noappearance of nitrate species bands. The above results reveal that afterthe formation of metal sulfates, the NH3-SCR reaction pathway is notchanged, and thus, the reaction can proceed more effectively becausemany more NH3 species are adsorbed and activated on ABS/FeW/MCM-41, which improves the NO reduction efficiency.

Lastly, we would like to present a short discussion on the possiblemodulation of the breakpoint of the dual effect for improved low-temperature performance. In view of the abundant acidity and suitableredox properties made by sulfate groups, it is reasonable to expect thatthe dual effect of ABS may be of certain universality. Thus, how tocontrol the behavior of ABS to shift the turning point to lower tem-peratures becomes an important issue. This is directly related to the lowtemperature requirement of NH3-SCR catalysts. It is known from theabove results that the formation of metal sulfates originates from the

interaction between ABS and metal oxides, a process that is accom-panied by the disintegration of ammonium and bisulfate groups. Thus,release of NH3 from ABS may be a crucial step in accelerating metalsulfate formation, which in turn determines the onset temperature ofthe promotion effect. As a piece of evidence, we find that by depositingABS onto the surface of bulk FeW oxide (without MCM-41 support), thedual effect is still present, but the turning point shifts to 300 °C (Fig.S8). At the same time, we find that the NH3 evolution signal is alsopostponed by approximately 50 °C (Fig. S9). This phenomenon in-dicates that if we could accelerate the release of NH3 in ABS [12,14], itwould be possible to shift the promotion effect to ideal lower tem-peratures.

4. Conclusion

In summary, we initiated this research by intentionally depositingABS on an environmental friendly NH3-SCR catalyst (Fe2O3-WO3/MCM-41). The dual effects of ABS on the catalytic performance over FeW/MCM-41 are shown and the reason for that unique behavior is explored.At low temperatures, the NH4+ is not reactive and ABS mainly serves asa physical blocking layer on catalyst surface, preventing the adsorptionand activation of reactants. Nevertheless, the case is changed whenreaction temperature elevates. The deposited ABS is disintegrated byNO/O2 consuming ammonium ions at reaction temperature higher than200 °C. As such, the blocking effect of ABS is removed and more im-portantly, the sulfate species can be preserved on catalyst surface in theform of the metal sulfates. The immobilization of the residual sulfateafter NH4+ consumption provides sufficient strong acid sites for NH3adsorption. Apart from acidity, the ABS/FeW/MCM-41-U also revealsimproved NO oxidation ability, thus prompts activity enhancement viathe “fast NH3-SCR” pathway. Moreover, the reacted catalyst showsnegligible adsorption of inactive surface nitrates at high temperatures,giving out more adsorption sites for NH3. In terms of the simply E-R

Fig. 10. In situ DRIFT spectra of the transient reactions at 250 °C as a function of time between (a) NO + O2 and pre-adsorbed NH3 and (b) NH3 and pre-adsorbedNO + O2 over FeW/MCM-41 and (c) NH3 and pre-adsorbed NO + O2 and (d) NO + O2 and pre-adsorbed NH3 as a function of time over ABS/FeW/MCM-41-U.


9

mechanism over FeW/MCM-41 catalyst, the enhanced surface acidityand improved ability to produce NO2 ensure the reaction goes throughmore effectively over ABS/FeW/MCM-41-U sample. This work preciselyillustrates the impact of ABS on NH3-SCR performance, which inspires anew strategy to the regeneration of ABS deactivated NH3-SCR catalystand benefit the development of long-lived catalysts for the practicalapplications.

Declaration of Competing Interest

The authors declare that they have no known competing financialinterests or personal relationships that could have appeared to influ-ence the work reported in this paper.

Acknowledgements

Financial support from the National Natural Science Foundation ofChina (21773106, 21707066, 21806077, 21976081), the MajorScientific and Technological Project of Bingtuan (2018AA002) and theEnvironmental Protection Department of Jiangsu Province (2016048)are gratefully acknowledged.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cej.2020.124271.

References

[1] J.-A. James, S. Sung, H. Jeong, O.A. Broesicke, S.P. French, D. Li, J.C. Crittenden,Impacts of combined cooling, heating and power systems, and rainwater harvestingon water demand, carbon dioxide, and nox emissions for atlanta, Environ. Sci.Technol. 52 (2018) 3–10.

[2] A. Richter, J.P. Burrows, H. Nüß, C. Granier, U. Niemeier, Increase in troposphericnitrogen dioxide over China observed from space, Nature 437 (2005) 129–132.

[3] G.C. Borillo, Y.S. Tadano, A.F.L. Godoi, S.S.M. Santana, F.M. Weronka,R.A. Penteado Neto, D. Rempel, C.I. Yamamoto, S. Potgieter-Vermaak,J.H. Potgieter, R.H.M. Godoi, Effectiveness of selective catalytic reduction systemson reducing gaseous emissions from an engine using diesel and biodiesel blends,Environ. Sci. Technol. 49 (2015) 3246–3251.

[4] C. Tang, H. Zhang, L. Dong, Ceria-based catalysts for low-temperature selectivecatalytic reduction of NO with NH3, Catal. Sci. Technol. 6 (2016) 1248–1264.

[5] L. Chen, J. Li, M. Ge, The poisoning effect of alkali metals doping over nanoV2O5–WO3/TiO2 catalysts on selective catalytic reduction of NOx by NH3, Chem.Eng. J. 170 (2011) 531–537.

[6] H. Liu, Z. Fan, C. Sun, S. Yu, S. Feng, W. Chen, D. Chen, C. Tang, F. Gao, L. Dong,Improved activity and significant SO2 tolerance of samarium modified CeO2-TiO2catalyst for NO selective catalytic reduction with NH3, Appl. Catal. B: Environ. 244(2019) 671–683.

[7] P.G. Smirniotis, D.A. Peña, B.S. Uphade, Low-temperature selective catalytic re-duction (SCR) of NO with NH3 by using Mn, Cr, and Cu oxides supported onhombikat TiO2, Angew. Chem. Int. Ed. 40 (2001) 2479–2482.

[8] R.Q. Long, R.T. Yang, R. Chang, Low temperature selective catalytic reduction(SCR) of NO with NH3 over Fe–Mn based catalysts, Chem. Commun. (2002)452–453.

[9] S. Matsuda, T. Kamo, A. Kato, F. Nakajima, T. Kumura, H. Kuroda, Deposition ofammonium bisulfate in the selective catalytic reduction of nitrogen oxides withammonia, Ind. Eng. Chem. Prod. Res. Dev. 21 (1982) 48–52.

[10] Z. Zhu, H. Niu, Z. Liu, S. Liu, Decomposition and reactivity of NH4HSO4 on V2O5/AC catalysts used for NO reduction with ammonia, J. Catal. 195 (2000) 268–278.

[11] Y. Chen, C. Li, J. Chen, X. Tang, Self-prevention of well-defined-facet Fe2O3/MoO3against deposition of ammonium bisulfate in low-temperature NH3–SCR, Environ.Sci. Technol. 52 (2018) 11796–11802.

[12] X. Wang, X. Du, L. Zhang, Y. Chen, G. Yang, J. Ran, Promotion of NH4HSO4 de-composition in NO/NO2 contained atmosphere at low temperature over V2O5-WO3/TiO2 catalyst for NO reduction, Appl. Catal. A 559 (2018) 112–121.

[13] J. Yu, F. Guo, Y. Wang, J. Zhu, Y. Liu, F. Su, S. Gao, G. Xu, Sulfur poisoning resistantmesoporous Mn-base catalyst for low-temperature SCR of NO with NH3, Appl.Catal. B 95 (2010) 160–168.

[14] K. Guo, G. Fan, D. Gu, S. Yu, K. Ma, A. Liu, W. Tan, J. Wang, X. Du, W. Zou, C. Tang,L. Dong, Pore size expansion accelerates ammonium bisulfate decomposition forimproved sulfur resistance in low-temperature NH3-SCR, ACS Appl. Mater.Interfaces 11 (2019) 4900–4907.

[15] K. Ma, K. Guo, L. Li, W. Zou, C. Tang, L. Dong, Cavity size dependent SO2 resistancefor NH3-SCR of hollow structured CeO2-TiO2 catalysts, Catal. Commun. 128 (2019)105719.

[16] H. Chang, X. Chen, J. Li, L. Ma, C. Wang, C. Liu, J.W. Schwank, J. Hao,

Improvement of activity and SO2 tolerance of Sn-modified MnOx-CeO2 catalysts forNH3-SCR at low temperatures, Environ. Sci. Technol. 47 (2013) 5294–5301.

[17] L. Zhang, W. Zou, K. Ma, Y. Cao, Y. Xiong, S. Wu, C. Tang, F. Gao, L. Dong, Sulfatedtemperature effects on the catalytic activity of CeO2 in NH3-selective catalytic re-duction conditions, J. Phys. Chem. C 119 (2015) 1155–1163.

[18] L. Zhang, L. Li, Y. Cao, X. Yao, C. Ge, F. Gao, Y. Deng, C. Tang, L. Dong, Gettinginsight into the influence of SO2 on TiO2/CeO2 for the selective catalytic reductionof NO by NH3, Appl. Catal. B 165 (2015) 589–598.

[19] F. Liu, K. Asakura, H. He, W. Shan, X. Shi, C. Zhang, Influence of sulfation on irontitanate catalyst for the selective catalytic reduction of NOx with NH3, Appl. Catal. B103 (2011) 369–377.

[20] D. Ye, R. Qu, H. Song, X. Gao, Z. Luo, M. Ni, K. Cen, New insights into the variousdecomposition and reactivity behaviors of NH4HSO4 with NO on V2O5/TiO2 cata-lyst surfaces, Chem. Eng. J. 283 (2016) 846–854.

[21] C. Li, Z. Huang, Y. Chen, X. Liu, J. Chen, W. Qu, Z. Ma, X. Tang, Optimizing se-lective catalytic reduction of NO with NH3 on Fe2O3/WO3 via redox-acid synergy,ChemCatChem 10 (2018) 3990–3994.

[22] P.R. Ettireddy, N. Ettireddy, S. Mamedov, P. Boolchand, P.G. Smirniotis, Surfacecharacterization studies of TiO2 supported manganese oxide catalysts for lowtemperature SCR of NO with NH3, Appl. Catal. B Environ. 76 (2007) 123–134.

[23] E. García-Bordejé, J.L. Pinilla, M.J. Lázaro, R. Moliner, NH3-SCR of NO at lowtemperatures over sulphated vanadia on carbon-coated monoliths: effect of H2Oand SO2 traces in the gas feed, Appl. Catal. B 66 (2006) 281–287.

[24] R. Kiyoura, K. Urano, Mechanism kinetics, and equilibrium of thermal decom-position of ammonium sulfate, Ind. Eng. Chem. Process Des. Dev. 9 (1970)489–494.

[25] R. Huang, B. Lan, Z. Chen, H. Yan, Q. Zhang, J. Bing, L. Li, Catalytic ozonation of p-chlorobenzoic acid over MCM-41 and Fe loaded MCM-41, Chem. Eng. J. 180 (2012)19–24.

[26] H. Wang, Z. Qu, S. Dong, H. Xie, C. Tang, Superior performance of Fe1–xWxOδ forthe selective catalytic reduction of NOx with NH3: interaction between Fe and W,Environ. Sci. Technol. 50 (2016) 13511–13519.

[27] D. Ye, R. Qu, H. Song, C. Zheng, X. Gao, Z. Luo, M. Ni, K. Cen, Investigation of thepromotion effect of WO3 on the decomposition and reactivity of NH4HSO4 with NOon V2O5–WO3/TiO2 SCR catalysts, RSC Adv. 6 (2016) 55584–55592.

[28] Y. Yu, C. Chen, M. Ma, M. Douthwaite, C. He, J. Miao, J. Chen, C. Li, SO2 promotedin situ recovery of thermally deactivated Fe2(SO4)3/TiO2 NH3-SCR catalysts: fromexperimental work to theoretical study, Chem. Eng. J. 361 (2019) 820–829.

[29] W. Shan, F. Liu, H. He, X. Shi, C. Zhang, Novel cerium–tungsten mixed oxide cat-alyst for the selective catalytic reduction of NOx with NH3, Chem. Commun. 47(2011) 8046–8048.

[30] D. Ye, R. Qu, C. Zheng, K. Cen, X. Gao, Mechanistic investigation of enhanced re-activity of NH4HSO4 and NO on Nb- and Sb-doped VW/Ti SCR catalysts, Appl.Catal. A:Gen. 549 (2018) 310–319.

[31] T. Kang, Y. Park, K. Choi, J.S. Lee, J. Yi, Ordered mesoporous silica (SBA-15) de-rivatized with imidazole-containing functionalities as a selective adsorbent of pre-cious metal ions, J. Mater. Chem. 14 (2004) 1043–1049.

[32] T.P.B. Nguyen, J.-W. Lee, W.G. Shim, H. Moon, Synthesis of functionalized SBA-15with ordered large pore size and its adsorption properties of BSA, MicroporousMesoporous Mater. 110 (2008) 560–569.

[33] M.A. Larrubia, G. Ramis, G. Busca, An FT-IR study of the adsorption of urea andammonia over V2O5–MoO3–TiO2 SCR catalysts, Appl. Catal. B 27 (2000)L145–L151.

[34] J. Yang, J. Zhang, L. Zhu, S. Chen, Y. Zhang, Y. Tang, Y. Zhu, Y. Li, Synthesis ofnano titania particles embedded in mesoporous SBA-15: characterization andphotocatalytic activity, J. Hazard. Mater. 137 (2006) 952–958.

[35] B.Q. Jiang, Z.B. Wu, Y. Liu, S.C. Lee, W.K. Ho, DRIFT study of the SO2 effect on low-temperature SCR reaction over Fe−Mn/TiO2, J. Phys. Chem. C 114 (2010)4961–4965.

[36] L. Han, M. Gao, J.-Y. Hasegawa, S. Li, Y. Shen, H. Li, L. Shi, D. Zhang, SO2-tolerantselective catalytic reduction of NOx over Meso-TiO2@Fe2O3@Al2O3 metal-basedmonolith catalysts, Environ. Sci. Technol. 53 (2019) 6462–6473.

[37] S. Cai, H. Hu, H. Li, L. Shi, D. Zhang, Design of multi-shell Fe2O3@MnOx@CNTs forthe selective catalytic reduction of NO with NH3: improvement of catalytic activityand SO2 tolerance, Nanoscale 8 (2016) 3588–3598.

[38] L. Xu, C. Wang, H. Chang, Q. Wu, T. Zhang, J. Li, New insight into SO2 poisoningand regeneration of CeO2–WO3/TiO2 and V2O5–WO3/TiO2 catalysts for low-tem-perature NH3–SCR, Environ. Sci. Technol. 52 (2018) 7064–7071.

[39] L. Song, J. Chao, Y. Fang, H. He, J. Li, W. Qiu, G. Zhang, Promotion of ceria fordecomposition of ammonia bisulfate over V2O5-MoO3/TiO2 catalyst for selectivecatalytic reduction, Chem. Eng. J. 303 (2016) 275–281.

[40] L. Kang, L. Han, J. He, H. Li, T. Yan, G. Chen, J. Zhang, L. Shi, D. Zhang, improvedNOx reduction in the presence of SO2 by using Fe2O3-promoted halloysite-supportedCeO2–WO3 Catalysts, Environ. Sci. Technol. 53 (2019) 938–945.

[41] N. Apostolescu, B. Geiger, K. Hizbullah, M.T. Jan, S. Kureti, D. Reichert, F. Schott,W. Weisweiler, Selective catalytic reduction of nitrogen oxides by ammonia on ironoxide catalysts, Appl. Catal. B Environ. 62 (2006) 104–114.

[42] F. Liu, H. He, C. Zhang, Z. Feng, L. Zheng, Y. Xie, T. Hu, Selective catalytic re-duction of NO with NH3 over iron titanate catalyst: catalytic performance andcharacterization, Appl. Catal. B Environ. 96 (2010) 408–420.

[43] F. Liu, H. He, Selective catalytic reduction of NO with NH3 over manganese sub-stituted iron titanate catalyst: Reaction mechanism and H2O/SO2 inhibition me-chanism study, Catal. Today 153 (2010) 70–76.

[44] H. Wang, Z. Qu, S. Dong, C. Tang, Mechanistic investigation into the effect ofsulfuration on the FeW catalysts for the selective catalytic reduction of NOx withNH3, ACS Appl. Mater. Interfaces 9 (2017) 7017–7028.


10

https://doi.org/10.1016/j.cej.2020.124271https://doi.org/10.1016/j.cej.2020.124271http://refhub.elsevier.com/S1385-8947(20)30262-X/h0005http://refhub.elsevier.com/S1385-8947(20)30262-X/h0005http://refhub.elsevier.com/S1385-8947(20)30262-X/h0005http://refhub.elsevier.com/S1385-8947(20)30262-X/h0005http://refhub.elsevier.com/S1385-8947(20)30262-X/h0010http://refhub.elsevier.com/S1385-8947(20)30262-X/h0010http://refhub.elsevier.com/S1385-8947(20)30262-X/h0015http://refhub.elsevier.com/S1385-8947(20)30262-X/h0015http://refhub.elsevier.com/S1385-8947(20)30262-X/h0015http://refhub.elsevier.com/S1385-8947(20)30262-X/h0015http://refhub.elsevier.com/S1385-8947(20)30262-X/h0015http://refhub.elsevier.com/S1385-8947(20)30262-X/h0020http://refhub.elsevier.com/S1385-8947(20)30262-X/h0020http://refhub.elsevier.com/S1385-8947(20)30262-X/h0025http://refhub.elsevier.com/S1385-8947(20)30262-X/h0025http://refhub.elsevier.com/S1385-8947(20)30262-X/h0025http://refhub.elsevier.com/S1385-8947(20)30262-X/h0030http://refhub.elsevier.com/S1385-8947(20)30262-X/h0030http://refhub.elsevier.com/S1385-8947(20)30262-X/h0030http://refhub.elsevier.com/S1385-8947(20)30262-X/h0030http://refhub.elsevier.com/S1385-8947(20)30262-X/h0035http://refhub.elsevier.com/S1385-8947(20)30262-X/h0035http://refhub.elsevier.com/S1385-8947(20)30262-X/h0035http://refhub.elsevier.com/S1385-8947(20)30262-X/h0040http://refhub.elsevier.com/S1385-8947(20)30262-X/h0040http://refhub.elsevier.com/S1385-8947(20)30262-X/h0040http://refhub.elsevier.com/S1385-8947(20)30262-X/h0045http://refhub.elsevier.com/S1385-8947(20)30262-X/h0045http://refhub.elsevier.com/S1385-8947(20)30262-X/h0045http://refhub.elsevier.com/S1385-8947(20)30262-X/h0050http://refhub.elsevier.com/S1385-8947(20)30262-X/h0050http://refhub.elsevier.com/S1385-8947(20)30262-X/h0055http://refhub.elsevier.com/S1385-8947(20)30262-X/h0055http://refhub.elsevier.com/S1385-8947(20)30262-X/h0055http://refhub.elsevier.com/S1385-8947(20)30262-X/h0060http://refhub.elsevier.com/S1385-8947(20)30262-X/h0060http://refhub.elsevier.com/S1385-8947(20)30262-X/h0060http://refhub.elsevier.com/S1385-8947(20)30262-X/h0065http://refhub.elsevier.com/S1385-8947(20)30262-X/h0065http://refhub.elsevier.com/S1385-8947(20)30262-X/h0065http://refhub.elsevier.com/S1385-8947(20)30262-X/h0070http://refhub.elsevier.com/S1385-8947(20)30262-X/h0070http://refhub.elsevier.com/S1385-8947(20)30262-X/h0070http://refhub.elsevier.com/S1385-8947(20)30262-X/h0070http://refhub.elsevier.com/S1385-8947(20)30262-X/h0075http://refhub.elsevier.com/S1385-8947(20)30262-X/h0075http://refhub.elsevier.com/S1385-8947(20)30262-X/h0075http://refhub.elsevier.com/S1385-8947(20)30262-X/h0080http://refhub.elsevier.com/S1385-8947(20)30262-X/h0080http://refhub.elsevier.com/S1385-8947(20)30262-X/h0080http://refhub.elsevier.com/S1385-8947(20)30262-X/h0085http://refhub.elsevier.com/S1385-8947(20)30262-X/h0085http://refhub.elsevier.com/S1385-8947(20)30262-X/h0085http://refhub.elsevier.com/S1385-8947(20)30262-X/h0090http://refhub.elsevier.com/S1385-8947(20)30262-X/h0090http://refhub.elsevier.com/S1385-8947(20)30262-X/h0090http://refhub.elsevier.com/S1385-8947(20)30262-X/h0095http://refhub.elsevier.com/S1385-8947(20)30262-X/h0095http://refhub.elsevier.com/S1385-8947(20)30262-X/h0095http://refhub.elsevier.com/S1385-8947(20)30262-X/h0100http://refhub.elsevier.com/S1385-8947(20)30262-X/h0100http://refhub.elsevier.com/S1385-8947(20)30262-X/h0100http://refhub.elsevier.com/S1385-8947(20)30262-X/h0105http://refhub.elsevier.com/S1385-8947(20)30262-X/h0105http://refhub.elsevier.com/S1385-8947(20)30262-X/h0105http://refhub.elsevier.com/S1385-8947(20)30262-X/h0110http://refhub.elsevier.com/S1385-8947(20)30262-X/h0110http://refhub.elsevier.com/S1385-8947(20)30262-X/h0110http://refhub.elsevier.com/S1385-8947(20)30262-X/h0115http://refhub.elsevier.com/S1385-8947(20)30262-X/h0115http://refhub.elsevier.com/S1385-8947(20)30262-X/h0115http://refhub.elsevier.com/S1385-8947(20)30262-X/h0120http://refhub.elsevier.com/S1385-8947(20)30262-X/h0120http://refhub.elsevier.com/S1385-8947(20)30262-X/h0120http://refhub.elsevier.com/S1385-8947(20)30262-X/h0125http://refhub.elsevier.com/S1385-8947(20)30262-X/h0125http://refhub.elsevier.com/S1385-8947(20)30262-X/h0125http://refhub.elsevier.com/S1385-8947(20)30262-X/h0130http://refhub.elsevier.com/S1385-8947(20)30262-X/h0130http://refhub.elsevier.com/S1385-8947(20)30262-X/h0130http://refhub.elsevier.com/S1385-8947(20)30262-X/h0135http://refhub.elsevier.com/S1385-8947(20)30262-X/h0135http://refhub.elsevier.com/S1385-8947(20)30262-X/h0135http://refhub.elsevier.com/S1385-8947(20)30262-X/h0140http://refhub.elsevier.com/S1385-8947(20)30262-X/h0140http://refhub.elsevier.com/S1385-8947(20)30262-X/h0140http://refhub.elsevier.com/S1385-8947(20)30262-X/h0145http://refhub.elsevier.com/S1385-8947(20)30262-X/h0145http://refhub.elsevier.com/S1385-8947(20)30262-X/h0145http://refhub.elsevier.com/S1385-8947(20)30262-X/h0150http://refhub.elsevier.com/S1385-8947(20)30262-X/h0150http://refhub.elsevier.com/S1385-8947(20)30262-X/h0150http://refhub.elsevier.com/S1385-8947(20)30262-X/h0155http://refhub.elsevier.com/S1385-8947(20)30262-X/h0155http://refhub.elsevier.com/S1385-8947(20)30262-X/h0155http://refhub.elsevier.com/S1385-8947(20)30262-X/h0160http://refhub.elsevier.com/S1385-8947(20)30262-X/h0160http://refhub.elsevier.com/S1385-8947(20)30262-X/h0160http://refhub.elsevier.com/S1385-8947(20)30262-X/h0165http://refhub.elsevier.com/S1385-8947(20)30262-X/h0165http://refhub.elsevier.com/S1385-8947(20)30262-X/h0165http://refhub.elsevier.com/S1385-8947(20)30262-X/h0170http://refhub.elsevier.com/S1385-8947(20)30262-X/h0170http://refhub.elsevier.com/S1385-8947(20)30262-X/h0170http://refhub.elsevier.com/S1385-8947(20)30262-X/h0175http://refhub.elsevier.com/S1385-8947(20)30262-X/h0175http://refhub.elsevier.com/S1385-8947(20)30262-X/h0175http://refhub.elsevier.com/S1385-8947(20)30262-X/h0180http://refhub.elsevier.com/S1385-8947(20)30262-X/h0180http://refhub.elsevier.com/S1385-8947(20)30262-X/h0180http://refhub.elsevier.com/S1385-8947(20)30262-X/h0185http://refhub.elsevier.com/S1385-8947(20)30262-X/h0185http://refhub.elsevier.com/S1385-8947(20)30262-X/h0185http://refhub.elsevier.com/S1385-8947(20)30262-X/h0190http://refhub.elsevier.com/S1385-8947(20)30262-X/h0190http://refhub.elsevier.com/S1385-8947(20)30262-X/h0190http://refhub.elsevier.com/S1385-8947(20)30262-X/h0195http://refhub.elsevier.com/S1385-8947(20)30262-X/h0195http://refhub.elsevier.com/S1385-8947(20)30262-X/h0195http://refhub.elsevier.com/S1385-8947(20)30262-X/h0200http://refhub.elsevier.com/S1385-8947(20)30262-X/h0200http://refhub.elsevier.com/S1385-8947(20)30262-X/h0200http://refhub.elsevier.com/S1385-8947(20)30262-X/h0205http://refhub.elsevier.com/S1385-8947(20)30262-X/h0205http://refhub.elsevier.com/S1385-8947(20)30262-X/h0205http://refhub.elsevier.com/S1385-8947(20)30262-X/h0210http://refhub.elsevier.com/S1385-8947(20)30262-X/h0210http://refhub.elsevier.com/S1385-8947(20)30262-X/h0210http://refhub.elsevier.com/S1385-8947(20)30262-X/h0215http://refhub.elsevier.com/S1385-8947(20)30262-X/h0215http://refhub.elsevier.com/S1385-8947(20)30262-X/h0215http://refhub.elsevier.com/S1385-8947(20)30262-X/h0220http://refhub.elsevier.com/S1385-8947(20)30262-X/h0220http://refhub.elsevier.com/S1385-8947(20)30262-X/h0220

[45] H. Wang, Z. Qu, H. Xie, N. Maeda, L. Miao, Z. Wang, Insight into the mesoporousFexCe1−xO2−δ catalysts for selective catalytic reduction of NO with NH3: regulablestructure and activity, J. Catal. 338 (2016) 56–67.

[46] M. Kantcheva, E.Z. Ciftlikli, FTIR spectroscopic characterization of NOx speciesadsorbed on ZrO2 and ZrO2−SO42−, J. Phys. Chem. B 106 (2002) 3941–3949.

[47] L. Li, L. Zhang, K. Ma, W. Zou, Y. Cao, Y. Xiong, C. Tang, L. Dong, Ultra-low loadingof copper modified TiO2/CeO2 catalysts for low-temperature selective catalyticreduction of NO by NH3, Appl. Catal. B 207 (2017) 366–375.

[48] Z. Liu, Y. Liu, Y. Li, H. Su, L. Ma, WO3 promoted Mn–Zr mixed oxide catalyst for theselective catalytic reduction of NOx with NH3, Chem. Eng. J. 283 (2016)

1044–1050.[49] Y. Li, Y. Wan, Y. Li, S. Zhan, Q. Guan, Y. Tian, Low-temperature selective catalytic

reduction of NO with NH3 over Mn2O3-doped Fe2O3 hexagonal microsheets, ACSAppl. Mater. Interfaces 8 (2016) 5224–5233.

[50] L. Han, M. Gao, C. Feng, L. Shi, D. Zhang, Fe2O3–CeO2@Al2O3 nanoarrays on Al-Mesh as SO2-tolerant monolith catalysts for NOx reduction by NH3, Environ. Sci.Technol. 53 (2019) 5946–5956.

[51] M. Kantcheva, A.S. Vakkasoglu, Cobalt supported on zirconia and sulfated zirconiaI.: FT-IR spectroscopic characterization of the NOx species formed upon NO ad-sorption and NO/O2 coadsorption, J. Catal. 223 (2004) 352–363.


11

http://refhub.elsevier.com/S1385-8947(20)30262-X/h0225http://refhub.elsevier.com/S1385-8947(20)30262-X/h0225http://refhub.elsevier.com/S1385-8947(20)30262-X/h0225http://refhub.elsevier.com/S1385-8947(20)30262-X/h0230http://refhub.elsevier.com/S1385-8947(20)30262-X/h0230http://refhub.elsevier.com/S1385-8947(20)30262-X/h0235http://refhub.elsevier.com/S1385-8947(20)30262-X/h0235http://refhub.elsevier.com/S1385-8947(20)30262-X/h0235http://refhub.elsevier.com/S1385-8947(20)30262-X/h0240http://refhub.elsevier.com/S1385-8947(20)30262-X/h0240http://refhub.elsevier.com/S1385-8947(20)30262-X/h0240http://refhub.elsevier.com/S1385-8947(20)30262-X/h0245http://refhub.elsevier.com/S1385-8947(20)30262-X/h0245http://refhub.elsevier.com/S1385-8947(20)30262-X/h0245http://refhub.elsevier.com/S1385-8947(20)30262-X/h0250http://refhub.elsevier.com/S1385-8947(20)30262-X/h0250http://refhub.elsevier.com/S1385-8947(20)30262-X/h0250http://refhub.elsevier.com/S1385-8947(20)30262-X/h0255http://refhub.elsevier.com/S1385-8947(20)30262-X/h0255http://refhub.elsevier.com/S1385-8947(20)30262-X/h0255

The dual effects of ammonium bisulfate on the selective catalytic reduction of NO with NH3 over Fe2O3-WO3 catalyst confined in MCM-41IntroductionExperimentsSample preparationSample characterizationsCatalytic activity test

Results and discussionReaction features of fresh and ABS-deposited FeW/MCM-41 in NH3-SCRThe dispersion of surface speciesThe chemical states of surface speciesEvolution of ABS during the reactionSurface acidity of catalysts (NH3-TPD and NH3-DRIFTS)Redox properties of catalysts (H2-TPR, NH3 and NO oxidation, NO + O2 DRIFTS)Reaction mechanism of NH3-SCR over fresh and ABS deposited catalysts

Conclusionmk:H1_15AcknowledgementsSupplementary dataReferences

Chemical Engineering Journal - Nanjing University...showed superior sulfur resistance to that of the...

Documents

Transcript of Chemical Engineering Journal - Nanjing University...showed superior sulfur resistance to that of the...