Three-Dimensionally Ordered Macroporous La0.6Sr0.4MnO3

Apr 11, 2014 - Supported Ag Nanoparticles for the Combustion of Methane ... Beijing University of Technology, Beijing 100124, People,s Republic of Chi...
2 downloads 0 Views 4MB Size
Article pubs.acs.org/JPCC

Three-Dimensionally Ordered Macroporous La0.6Sr0.4MnO3 Supported Ag Nanoparticles for the Combustion of Methane Hamidreza Arandiyan,† Hongxing Dai,*,‡ Jiguang Deng,‡ Yuan Wang,‡,§ Hongyu Sun,∥ Shaohua Xie,‡ Bingyang Bai,† Yuxi Liu,‡ Kemeng Ji,‡ and Junhua Li*,† †

State Key Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), School of Environment, Tsinghua University, Beijing 100084, People’s Republic of China ‡ Laboratory of Catalysis Chemistry and Nanoscience, Department of Chemistry and Chemical Engineering, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing 100124, People’s Republic of China § National Research Center for Geoanalysis, Beijing 100037, People’s Republic of China ∥ Beijing National Center for Electron Microscopy, School of Materials Science and Engineering, The State Key Laboratory of New Ceramics and Fine Processing, Key Laboratory of Advanced Materials (MOE), Tsinghua University, Beijing 100084, People’s Republic of China S Supporting Information *

ABSTRACT: A series of Ag nanoparticles (NPs) supported on three-dimensionally ordered macroporous (3DOM) La0.6Sr0.4MnO3 (yAg/3DOM La0.6Sr0.4MnO3; y = 0, 1.57, 3.63, and 5.71 wt %) were successfully prepared with high surface areas (38.2−42.7 m2/g) by a facile novel reduction method using poly methacrylate colloidal crystal as template in a dimethoxytetraethylene glycol (DMOTEG) solution. Physicochemical properties of these materials were characterized by means of numerous techniques, and their catalytic activities were evaluated for the combustion of methane. It is shown that the yAg/3DOM La0.6Sr0.4MnO3 materials possessed unique nanovoid-like 3DOM architectures, and the Ag NPs were well dispersed on the inner walls of macropores. Among the La 1−x Sr x MnO 3 (x = 0.2, 0.4, 0.6, 0.8) and yAg/3DOM La0.6Sr0.4MnO3 (y = 0, 1.57, 3.63, and 5.71 wt %) samples, 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 performed the best, giving T10%, T50%, and T90% (temperatures corresponding to methane conversion =10, 50, and 90%) of 361, 454, and 524 °C, respectively, and the highest turnover frequency (TOFAg) value of 1.86 × 10−5 (mol/molAg s) at 300 °C. The apparent activation energies (39.1−37.5 kJ/mol) of the yAg/3DOM La0.6Sr0.4MnO3 samples were much lower than that (91.4 kJ/mol) of the bulk La0.6Sr0.4MnO3 sample. The effects of water vapor and sulfur dioxide on the catalytic activity of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample were also examined. It is concluded that its super catalytic activity was associated with its high oxygen adspecies concentration, good low-temperature reducibility, large surface area, and strong interaction between Ag and La0.6Sr0.4MnO3 as well as the unique nanovoid-walled 3DOM structure. silica), into a face-centered close-packed array.6 Manganese oxides are among the various catalytic materials which show fairly good activity in flameless methane combustion.7,8 Their heat resistance can be improved when manganese oxides are incorporated into the ABO3 structure, e.g., lanthanum-based perovskite LaMnO3, which is one of the most active perovskite catalysts for the combustion of methane.9,10 The rare earth metals in the perovskite structure provide the thermal stability of transition metal oxides.11,12 Lanthanum oxide alone also

1. INTRODUCTION Perovskite-type oxides (ABO3) have attracted a lot of attention in the past few decades.1,2 They are a kind of versatile catalyst for applications in CO oxidation3 and photocatalysis.4 This is due to their unique crystal structures, nonstoichiometric chemistry, and adsorption properties.1,2 The major drawback of traditional ABO3 is its low surface area, which limits its catalytic application. Therefore, it is highly desirable to develop an effective strategy for generating porous ABO3. One approach to achieve this goal is the surfactant-assisted colloidal crystal templating strategy.5 The colloidal crystal template can be fabricated by ordering monodispersed microspheres (e.g., polystyrene (PS), poly(methyl methacrylate) (PMMA) or © 2014 American Chemical Society

Received: March 5, 2014 Revised: April 2, 2014 Published: April 11, 2014 14913

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

The Journal of Physical Chemistry C

Article

Scheme 1. Schematic Illustration of Loading Ag NPs on 3DOM La0.6Sr0.4MnO3 Catalysts

tion activities of 3DOM La0.6Sr0.4MnO3 with nanovoid-like or mesoporous skeletons and its supported Ag NP nanocatalyst (yAg/3DOM La0.6Sr0.4MnO3; y = 0, 1.57, 3.63, and 5.71 wt %) via the dimethoxytetraethylene glycol (DMOTEG) assisted gas bubbling reduction route with well-arrayed PMMA microspheres as the hard template. It was found that the DMOTEGmediated reduction strategy not only produced size-controlled Ag NPs, but also stabilized them against conglomeration without the need for additional stabilizers. By using this novel method, Ag NPs could be highly dispersed on 3DOM La0.6Sr0.4MnO3, and the obtained catalysts showed excellent performance and stability for methane combustion.

shows catalytic activity in the complete oxidation of methane, though much lower than manganese oxide.13 Previously, we found that chemisorbed oxygen and the active cobalt species, especially Co3+, could play a major role in the oxidation reaction in vanadium-doped CoCr2−xVxO4 (x = 0−0.20) which could lead to increases in Co3+ concentration and oxygen vacancy density.14 For (La0.9K0.1CoO3)x/nmCeO2 catalysts consisting of comparably sized La0.9K0.1CoO3 active component and CeO2 nanosized support performed well in the combustion of volatile organic compounds.15 It has been demonstrated that doping of perovskites such as LaCoO3,16 LaFeO3, LaFe0.95Pd0.05O3,17 or LaFe0.95Pd0.05O318 with Ag increases the catalytic activities in methane combustion. Gardner et al.19 studied the manganese oxide catalysts with low Ag loadings (≤1%) for low-temperature CO oxidation and observed good catalytic activities. Moreover, Ag-doped LaCoO3 showed no deactivation in methane combustion during 50 h of on-stream reaction at 600 °C.16 Even greater activity enhancement in CO oxidation was observed over Ag/MnOx/perovskite composites as compared to over LaMnO3.12 To the best of our knowledge, however, there have been no reports on the successful preparation of Ag nanoparticles (NPs) supported on threedimensionally ordered macroporous La0.6Sr0.4MnO3 with mesoporous walls using the PMMA-templating method and their applications in catalyzing the combustion of methane. In the past several years, our group has extended attention to the synthesis and physicochemical property characterization of three-dimensionally ordered macroporous (3DOM) structured materials (e.g., Co3O4/3DOM La0.6Sr0.4CoO3 (surface area = 29−32 m2/g),20 yCrOx/3DOM InVO4 (surface area = 41.3− 52.3 m2/g),21 3DOM InVO4 (surface area = 35−52 m2/g),22 xAu/3DOM LaMnO3 (surface area = 29.8−32.7 m2/g),23 Au/ 3DOM LaCoO3 (surface area = 24−29 m2/g),24 and Au/ 3DOM La0.6Sr0.4MnO3 (surface area = 31.1−32.9 m2/g)25 by surfactant-assisted PMMA-templating approaches, and observed that some of the 3DOM materials showed excellent catalytic performance in the combustion of toluene and methane.26−28 Herein, we report for the first time the preparation, characterization, and catalytic methane combus-

2. EXPERIMENTAL METHODS 2.1. Synthesis of Monodisperse PMMA Microspheres. Monodisperse PMMA microspheres (average diameter: 161 nm) and well-arrayed PMMA spheres were synthesized according to the methods reported in the literature.5,6,28 Potassium peroxydisulfate (K2S2O8) (0.40 g, 3.0 mmol) and deionized water (1500 mL) were stirred at 400 rpm, heated to 70 °C, and degassed with flowing N2 in a separable four-neck 2000-mL round-bottom flask. After equilibrating at 70 °C, methyl methacrylate (115 mL) was poured into the flask, and the resulting suspension was stirred at 70 °C for 1 h. The PMMA colloidal crystal template was prepared by centrifugation (4100 rpm) of the colloidal suspension (ca. 10.0 g) in a 25mL centrifugation tube for 75 min. When the water was evaporated in a water bath (80 °C), the obtained solid was first dried at room temperature (RT) for 48 h. 2.2. Synthesis of yAg/3DOM La0.6Sr0.4MnO3 Catalysts. The La0.4Sr0.6MnO3-supported silver catalysts (yAg/3DOM La0.4Sr0.6MnO3) were prepared via a DMOTEG-assisted reduction method. During the homogeneous catalysis of methane combustion by Ag NPs supported on 3DOM La0.4Sr0.6MnO3, we observed that DMOTEG was used as a multifunctional reagent for the formation of Ag NPs. It is found that the DMOTEG-mediated route not only produced sizecontrolled Ag NPs, but also stabilized them against conglomeration without the need for additional stabilizers. 14914

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

14915

PMMA templating

in situ PMMA templating

in situ PMMA templating

in situ PMMA templating

3DOM La0.6Sr0.4MnO3

1.58 wt % Ag/3DOM La0.6Sr0.4MnO3

3.63 wt % Ag/3DOM La0.6Sr0.4MnO3

5.71 wt % Ag/3DOM La0.6Sr0.4MnO3

PMMA/ (DMOTEGPEG400)

PMMA/ (DMOTEGPEG400)

PMMA/ (DMOTEGPEG400)

PMMA/ (DMOTEGPEG400)

−/−

hard template/ soft template

→300 °C, air (50 mL/min), 1 h →800 °C, air (50 mL/min), 4 h

→300 °C, air (50 mL/min), 1 h →800 °C, air (50 mL/min), 4 h 300 °C, N2 (50 mL/min), 3 h

→300 °C, air (50 mL/min), 1 h →800 °C, air (50 mL/min), 4 h 300 °C, N2 (50 mL/min), 3 h

→300 °C, air (50 mL/min), 1 h →800 °C, air (50 mL/min), 4 h 300 °C, N2 (50 mL/min), 3 h

850 °C, air (50 mL/min), 4 h 300 °C, N2 (50 mL/min), 3 h

calcination condition

6.0

4.0

5.71

3.63

1.58

−/−

−/−

2.0

−/−

−/−

−/−

3.5

3.2

3.3

−/−

Ag particle sizeb (nm)

RV

TV

rhombohedral

rhombohedral

rhombohedral

rhombohedral

rhombohedral

crystal phase

23.9

24.6

24.2

23.5

61.4

crystallite sizec (nm)

XRD result

7.6

7.4

7.1

30.6

34.1

35.6

34.8

38.2

41.5

42.7

42.1

2.6

−/−

−/− 7.3

total

mesopore ( 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3 (1.83) > 5.71 wt % Ag/3DOM La0.6Sr0.4MnO3 (1.69) > 3DOM La0.6Sr0.4MnO3 (1.22) > bulk La0.6Sr0.4MnO3 (0.81). The rise in surface active oxygen species concentration would give rise to enhanced catalytic performance of 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 for the combustion of methane. 3.6. Reducibility. H2-TPR experiments were conducted to investigate the reducibility of the bulk La0.6Sr0.4MnO3, 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 samples, and their profiles are illustrated in Figure 6 and Table 3. As shown in Figure 6a, two reduction peaks at 456 and 723 °C were observed for the 3DOM La0.6Sr0.4MnO3 sample. Since La3+ and Sr2+ are both nonreducible under the H2-TPR conditions adopted in the present study, the observed reduction peaks were due to the reduction of Mnn+ species.25,28 When Ag NPs were loaded on the La0.6Sr0.4MnO3 surface, all of the reduction peaks shifted to lower temperatures. For the 3.63 wt % Ag/ 3DOM La0.6Sr0.4MnO3 sample, the high-temperature reduction peak at 721 °C leading to the breakdown of the perovskite phase was almost unchanged in position and shape, as compared to that of the Ag-free 3DOM La0.6Sr0.4MnO3 sample (Figure 6a,b). However, the downshift of the low-temperature reduction peaks and the increment of their peak area were both apparent. The above XPS results (Figure 5B) show that there still existed an appreciable quantity of Agδ+ species on the surface of yAg/3DOM La0.6Sr0.4MnO3, and that the modification of Ag resulted in formation of a certain quantity of Mn4+. Thus, the small peak at 304 °C might be assigned to the reduction of Agδ+ species.39 The first reduction peak at 415 °C of the 3.63 wt % Ag/La0.6Sr0.4MnO3 sample was due to a single electron reduction of Mn3+ locating in coordination-unsaturated microenvironments and/or due to the reduction of the chemically adsorbed oxygen species on the highly dispersed Ag NPs (i.e., from Ag−Ox to Ag) or the interface between Ag NPs and La0.6Sr0.4MnO3 support (i.e., from Mn−Ox−Ag to Ag), which might be related to the weakening of the Mn−O bond induced by Ag atom.38,39 The high-temperature reduction peak at 721 °C most likely corresponded to the reduction of the remaining Mn3+ to Mn2+, which led to the breakdown of the perovskite phase and formation of the discrete oxide phases, such as La2O3, SrO, and MnO (see Figure 6a,b). By comparing the reduction profiles of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 and 3DOM La0.6Sr0.4MnO3 samples (Figure 6b), one can see that the porous structure promoted the dispersion of Ag NPs, thus favoring the reduction of Mn species. For the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 and bulk La0.6Sr0.4MnO3 samples, values of the H2 consumption at low temperatures (550 °C) were 1.40−146 and 1.52 mmol/g, respectively (Table 2 and Figure 6c). Values of the total H2 consumption of the 3DOM La0.6Sr0.4MnO3 and yAg/3DOM La0.6Sr0.4MnO3 samples were in the range 2.95− 3.38 mmol/g. The reduction of La0.6Sr0.4MnO3 usually proceeds via the sequence of Mn4+ → Mn3+ → Mn2+. If manganese ions in La0.6Sr0.4MnO3 were Mn4+ and Mn3+ and reduced to Mn2+, values of the H2 consumption would be 4.14 and 2.07 mmol/g, respectively. It has been reported that there

a

697 498 399 1.45 1.92 1.69 530.7

524 454 361 1.44 1.94 2.01 530.9

661 565 436 1.40 1.92 1.83 530.8

661 566 437 1.49 1.86 1.22 530.8

749 672 562 1.52 1.43 0.81 530.9

T90% T50% T10% 550−950 °C 50−550 °C Oads/ Olat2− Oadsd

methane combustionf (°C) H2 consumptione (mmol/g) molar ratio BE

Article

O Mn La Ag

Table 2. Surface Element Compositions, H2 Consumption, and Catalytic Activities of Bulk La0.6Sr0.4MnO3, 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 Samples

The Journal of Physical Chemistry C

14921

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

The Journal of Physical Chemistry C

Article

Figure 6. (a, b) H2-TPR profiles (i, 3DOM La0.6Sr0.4MnO3; ii, 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3; iii, 5.71 wt % Ag/3DOM La0.6Sr0.4MnO3; vi, 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3) and (c) initial H2 consumption rate as a function of inverse temperature of bulk La0.6Sr0.4MnO3, 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 samples.

was 35% Mn4+ and 65% Mn3+ (which led to the presence of nonstoichiometric oxygen in La0.6Sr0.4MnO3), corresponding to a H2 consumption of 2.80 mmol/g if Mn4+ and Mn3+ were reduced to Mn2+.40 In our case, the total H2 consumption of 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 was 3.38 mmol/g, which was higher than the above value (Table 2). It is known that the reducibility of a sample can be effectively evaluated by using the initial (where less than 25% oxygen in the sample was consumed for the first reduction peak) H2 consumption rate.14 The initial H2 consumption rate was calculated according to the H2 amount consumed per mole of Mn per second, which corresponds to the initial 25% area of the first reduction peak where no phase transformation of the sample occurs. Figure 6c shows the initial H2 consumption rate as a function of inverse temperature of the bulk La 0 . 6 Sr 0 . 4 MnO 3 , 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 samples. It is clearly seen that the initial H2 consumption rate decreased in the order 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 < 5.71 wt % Ag/3DOM La0.6Sr0.4MnO3 < 3DOM La0.6Sr0.4MnO3 < 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3 < bulk La0.6Sr0.4MnO3. The trend in low-temperature reducibility was in good agreement with the sequence in catalytic activity shown below. Through the above comparison, we believe that the loading of Ag NPs greatly improved the low-temperature reducibility of the La0.6Sr0.4MnO3 support, remarkably increasing the content of reducible Mnn+ species at low temperatures. 3.7. Catalytic Performance. In the blank experiment (only quartz sands were loaded in the microreactor), we did not detect a significant conversion of methane below 776 °C. This result demonstrates that no homogeneous reactions took place below 776 °C. Figure 7 and Figure S14 in the Supporting Information show the catalytic performances of the samples for the combustion of methane in the temperature range 300−900 °C under the conditions of 2% CH4 + 20% O2 + 78% N2 (balance) and GHSV of ca. 30 000 mL/(g h) over 20 mg of each catalyst. Figure S9 in the Supporting Information shows the catalytic activities of La1−xSrxMnO3 (x = 0.2, 0.4, 0.6, and

0.8) for methane combustion. As seen from Figure 7a, Table 2, and Figure S14b in the Supporting Information, the catalytic activities of the bulk La 0.6 Sr0.4 MnO 3 and yAg/3DOM La0.6Sr0.4MnO3 samples increased with the rise in temperature. It can be clearly observed that the yAg/3DOM La0.6Sr0.4MnO3 samples performed much better than the 3DOM La0.6Sr0.4MnO3 and bulk La0.6Sr0.4MnO3 samples. This result suggests that the porous materials outperformed the nonporous bulk counterpart for methane oxidation (Figure 7a and Figure S14b in the Supporting Information). Preparing the nanocatalysts with a considerable specific surface area leads the catalyst to performing excellently. As shown in Figure S14a in the Supporting Information, the catalytic activity decreased at elevated GHSV values. At GHSV = 30 000 mL/(g h), the T50% and T90% (temperatures required for methane conversion of 50 and 90%) were 454 and 524 °C, respectively, which were 87 and 113 °C lower than those achieved at GHSV = 40 000 mL/ (g h). A further rise in GHSV from 40 000 to 45 000 mL/(g h) resulted in a drop in catalytic activity. It is worth noticing that the 5.71 wt % Ag/3DOM La0.6Sr0.4MnO3 and 1.58 wt % Ag/ 3DOM La0.6Sr0.4MnO3 samples showed catalytic performances much inferior to that of 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 for methane oxidation, which might be due to the larger sizes of Ag NPs in the former than those in the latter, as confirmed by our high-resolution TEM images (Figure 3h,m,r) of the samples. The Ag particle size was in the range 1−6 nm with a narrow distribution and the mean diameters of the yAg/ 3DOM La0.6Sr0.4MnO3 (y = 1.58, 3.63, and 5.71 wt %) samples were 3.4, 3.3, and 3.5 nm (Figure 3i,n,s), respectively. This result suggests that Ag NPs were the active sites for the combustion of methane, and catalytic activity was related to the particle size and loading of Ag NPs. It is convenient to compare the catalytic activities of the samples using the T10%, T50%, and T90%, as summarized in Table 2. It can be found that the catalytic performance decreased in the sequence 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 > 5.71 wt % Ag/3DOM La 0.6 Sr 0.4 MnO 3 > 1.58 wt % Ag/3DOM 14922

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

Data before the “/” symbol are the k values, whereas those after the “/” symbol are the r values. bReaction conditions: 2% CH4 + 20% O2 + 78% N2 (balance); total flow = 41.6 mL/min; GHSV of ca. 30 000 mL/(g h).

Article

Figure 7. (a) Methane conversion and corresponding reaction rate versus reaction temperature over bulk La0.6Sr0.4MnO3, 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 catalysts. (b) Catalytic stability of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 catalyst at GHSV = 30 000 mL/(g h) under the conditions of 2% CH4 + 20% O2 + 78% N2 (balance) and total flow of 41.6 mL/min.

La 0 . 6 Sr 0 . 4 MnO 3 > 3DOM La 0 . 6 Sr 0 . 4 MnO 3 > bulk La0.6Sr0.4MnO3, coinciding with the order in low-temperature reducibility (i.e., the initial H2 consumption rate (Figure 5C) of the samples). Obviously, the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample with a surface area of 41.5 m2/g performed the best, giving T10%, T50%, and T90% of 361, 454, and 524 °C, respectively, which were much lower than those (T10% = 562 °C, T50% = 672 °C, and T90% = 749 °C) achieved over the bulk La0.6Sr0.4MnO3 sample with a surface area of 2.6 m2/g by 201, 216, and 225 °C, respectively. According to the activity data and moles of Mn in the La0.6Sr0.4MnO3 and yAg/3DOM La0.6Sr0.4MnO3 samples, we calculated the turnover frequencies (TOFAg and TOFMn) and reaction rates (μmol/(g s)) according to the single surface Ag site (or moles of Mn) and the Ag weight in the yAg/ La0.6Sr0.4MnO3 samples, respectively. The TOFAg (mol/molAg s) was calculated according to the number of CH4 molecules converted by single surface Ag site per second. The dispersion of silver was estimated according to the reported procedure25 and the assumption that the Ag NPs displayed a spherical or hemispherical shape (as confirmed by our high-resolution TEM images (Figures S5−S7 in the Supporting Information) of the samples. The TOFAg and TOFMn of the samples were calculated under the conditions of CH4/O2 molar ratio = 1/ 10, GHSV = 30 000 mL/(g h), and temperature = 300, 350, and 400 °C, as summarized in Table 3. It is observed that,

a

0.9991 0.9990 0.9995 0.9992 0.9998 91.4 56.6 39.1 37.5 38.2 0.42/0.92 2.64/4.71 37.8/58.7 41.0/61.8 38.2/59.3 0.35/0.76 1.31/2.63 10.7/17.7 13.1/21.0 11.7/15.3 0.25/0.55 0.95/1.97 2.19/4.08 5.86/8.65 2.23/4.69 0.16/0.37 0.67/1.43 1.14/2.56 1.57/3.51 1.33/2.42 −/− −/− 7.28 × 10−5 11.8 × 10−5 6.14 × 10−5 −/− −/− 4.32 × 10−5 5.68 × 10−5 2.62 × 10−5 −/− −/− 1.23 × 10−5 1.86 × 10−5 1.31 × 10−5 bulk La0.6Sr0.4MnO3 3DOM La0.6Sr0.4MnO3 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 5.71 wt % Ag/3DOM La0.6Sr0.4MnO3

1.73 4.17 5.07 5.56 5.05

× × × × ×

10 10−7 10−7 10−7 10−7

5.73 6.22 7.14 9.03 8.11

× × × × ×

10 10−7 10−7 10−7 10−7

8.40 7.92 11.1 16.7 13.2

× × × × ×

10 10−7 10−7 10−7 10−7

R2 Ea (kJ/mol) 550 °C 500 °C 400 °C 400 °C 350 °C 300 °C catalyst

300 °C

−8

350 °C

−8

400 °C

−8

300 °C

kinetic params k (×10−3 s−1)/ra (×10−4 μmol/(g s))b TOFMn (mol/molMn s) TOFAg (mol/molAg s)

Table 3. Turnover Frequencies (TOFs), Rate Constants (k), Reaction Rates (r), Activation Energies (Ea), and Correlation Coefficients (R2) of Bulk La0.6Sr0.4MnO3, 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 Samples for Methane Combustion at Different Temperatures

The Journal of Physical Chemistry C

14923

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

The Journal of Physical Chemistry C

Article

Figure 8. (A) Effect of water vapor on methane conversion at different reaction temperatures over the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 catalyst (H2O concentration = 3.0 vol %). (B) Effect of water introduction or cutting off in the feedstock over 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 and La0.6Sr0.4MnO3 at 600 °C (H2O concentration = 3.0 vol %). (C) Effect of water vapor concentration and (D) effect of SO2 on methane conversion over the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 catalyst under the conditions of GHSV = 30 000 mL/(g h) and SO2 concentration = 40 ppm.

yAg/3DOM La0.6Sr0.4MnO3 were much higher than that (14.2 μmol/(g s)) over 3.6 wt % Au/LaCoO3,41 but inferior to that (74.3 μmol/(g s)) over 8 wt % Au/MnO2.42 In past years, the oxidation of methane and volatile organic compounds (VOCs) has been studied over various catalysts, as summarized in Table S1 in the Supporting Information. It is worth pointing out that, under similar reaction conditions, the catalytic activity (T50% = 454 °C and T90% = 524 °C) over our best-performing 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample was much better than those over La0.9Cu0.1MnO3,43 20 wt% LaMnO3/MgO,44 La0.5Sr0.5MnO3,36 and La2CuO4 nanorods,45 but inferior to that over 1 wt% Pd/ZrO246 (Table S1 of the Supporting Information) In addition, the catalytic stability of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample was tested in a consecutive reaction experiment (Figure 7b), in which the sample experienced first two runs of catalytic tests, then 10 h of on-stream reaction at 500 °C and 30 000 mL/(g h), and finally two runs of catalytic tests. Each run lasted ca. 5 h. The overall test time was about 30 h. The result reveals that no obvious activity loss was detected within 30 h of the reaction (Figure S10C in the Supporting Information). Furthermore, the Mn 2p3/2 and O 1s XPS spectra (Figure S8 in the Supporting Information) and also the XRD pattern of the used sample were rather similar to those of the fresh sample. The surface Mn4+/Mn3+ (1.46−1.50) and Oads/Olat (1.65−2.02) molar ratios of the used sample were also similar to those (1.52 and

compared to the nonporous bulk La0.6Sr0.4MnO3 sample at the same temperature, the TOFMn values of the porous 3DOM La0.6Sr0.4MnO3 and yAg/3DOM La0.6Sr0.4MnO3 samples were much higher. For instance, the TOFMn value (5.56 × 10−7) of 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 was approximately 1.5 times as much as that (4.17 × 10−7) of 3DOM La0.6Sr0.4MnO3 for methane combustion at 300 °C. With the rise in Ag NP loading from 1.58 to 3.63 wt %, the obtained 1.58 wt % Ag/ 3DOM La 0 .6 Sr 0. 4 MnO 3 and 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 samples exhibited the TOFMn values of 5.07 × 10−7 and 5.56 × 10−7 mol/(molMn s) at 300 °C, and 7.14 × 10−7 and 9.03 × 10−7 mol/(molMn s) at 350 °C, respectively, but they decreased to 5.05 × 10−7 mol/(molMn s) at 300 °C and 8.11 × 10−7 mol/(molMn s) at 400 °C with a further rise in Ag NP loading from 3.63 to 5.71 wt %. In the meantime, from Table 3, one can also observe that the TOFAg values (1.86 × 10−5 mol/(molAg s) at 300 °C and 11.8 × 10−5 mol/(molAg s) at 400 °C) of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample was much higher than those (1.23 × 10−5 mol/(molAg s) at 300 °C and 7.28 × 10−5 mol/(molAg s) at 400 °C) of the 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3 sample. This result indicates that there was a strong metal−support interaction between the metal (Ag NPs) and the support (La0.6Sr0.4MnO3), which gave rise to an enhanced catalytic performance of yAg/3DOM La0.6Sr0.4MnO3 for methane combustion. The reaction rates (58.7−61.8 μmol/(g s)) for methane oxidation at 550 °C over 14924

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

The Journal of Physical Chemistry C

Article

Figure 9. Arrhenius plots for methane combustion over (a) bulk La0.6Sr0.4MnO3, 3DOM La0.6Sr0.4MnO3, and yAg/3DOM La0.6Sr0.4MnO3 samples at GHSV = 30 000 mL/(g h), and (b) 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 at different GHSV values.

% water vapor did not affect the catalytic activity of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample if the reaction temperature was higher than 600 °C, but it decreased the catalytic activity when the reaction temperature was below 550 °C. Meanwhile, the influence of water vapor concentration (1.0, 3.0, or 5.0 vol %) on the activity of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample for methane oxidation is shown in Figure 8C. The catalytic activity was not significantly affected at a lower water vapor concentration (1.0 vol %), whereas introduction of a higher water vapor concentration (5.0 vol %) decreased the T90% value by ca. 12%. Furthermore, the XRD pattern (Figure S12 in the Supporting Information) of the used sample in the case of 3.0 vol % water vapor addition was rather similar to that of the fresh one, and the surface area (40.6 m2/ g) of the former was close to that (41.5 m2/g) of the latter. The HRSEM and HRTEM images (Figure S13 in the Supporting Information) of the used sample further reveal that the Ag NPs were well stabilized on the surface of 3DOM La0.6Sr0.4MnO3. These results show that the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample was catalytically stable and the introduction of water vapor at a higher concentration into the reaction system gave rise to a negative effect on its catalytic performance. The drastic decrease in catalytic activity during water vapor addition was in agreement with the previous studies.23,25 It is concluded that the presence of extra water vapor had more or less effects on the metal oxide support, rather than on the precious metal.47 It might be due to the competitive adsorption of water and methane as well as oxygen molecules on the surface of 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 where the adsorption of water was stronger than that of oxygen. On the other hand, at a relatively high temperature, hydroxyl accumulation on the La0.6Sr0.4MnO3 support was associated with catalytic deactivation, due to inhibition of lattice oxygen mobility for activation of oxygen molecules on the sample surface.48,49 In order to examine the effect of SO2 on catalytic activity of the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample, we carried out methane oxidation in the presence of 40 ppm SO2 in the

2.01) of the fresh sample (Table 2), respectively. The surface area (42.7 m2/g) of fresh 3.63 wt% Ag/3DOM La0.6Sr0.4MnO3 was close to that (41.2 m2/g) of used 3.63 wt% Ag/3DOM La0.6Sr0.4MnO3 (Table 1). The SEM image (Figure S10A,B in the Supporting Information) of the used sample further reveals that the structure and morphology of 3DOM La0.6Sr0.4MnO3 were well stabilized. In the present study, the higher activity of the yAg/3DOM La0.6Sr0.4MnO3 samples could be not only due to the better dispersion of Ag NPs, but also due to the smaller size (1−6 nm) of Ag NPs, as compared to the 3.63 wt % Ag/ 3DOM La0.6Sr0.4MnO3 sample (Figure S6 in the Supporting Information). Comparison of the activity data and characterization results, we conclude that the good catalytic performance of 3.63 wt % Ag/3DOM La 0.6Sr0.4MnO3 for methane combustion was associated with its higher surface area, higher surface oxygen species concentration, and better low-temperature reducibility as well as the good-quality nanovoid-walled 3DOM structure. 3.8. Effects of Water Vapor and Sulfur Dioxide. The effects of water vapor and SO2 have been studied in the switching experiments. We conducted the oxidation of CH4 in the presence of 1.0, 3.0, or 5.0 vol % water vapor or 40 ppm SO2 over the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 and La0.6Sr0.4MnO3 samples at different temperatures, and the results are shown in Figure 8. Feed-gas-containing water vapor was introduced into a catalyst bed at a temperature of 550, 600, or 650 °C for 1 h, and then water-free gas was fed for 1 h. This alternating cycle was repeated four times. When the catalytic activity reached a steady value, 3.0 vol % water vapor in the feed was introduced to the reaction system. It is observed from Figure 8A that the addition of water vapor at 550 or 600 °C decreased the methane conversion by ca. 15−20%. The deactivation due to water vapor addition was reversible, and a recovery in the case of cutting off water vapor was observed (catalytic activities were restored to the initial values). Methane conversion after on-stream reaction at 550 °C for 10 h was 66.4%, which was lower than that (77.9%) after on-stream reaction at 550 °C for 10 min. Therefore, the addition of 3.0 vol 14925

dx.doi.org/10.1021/jp502256t | J. Phys. Chem. C 2014, 118, 14913−14928

The Journal of Physical Chemistry C

Article

La0.6Sr0.4MnO3 catalysts were much higher than that (39.1 kJ/mol) of the 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3 catalyst; with the rise in Ag loading from 1.58 to 3.63 wt %, the Ea value decreased from 39.1 to 37.5 kJ/mol, but a further rise in Ag loading to 5.71 wt % led to a slight increase in Ea value to 38.2 kJ/mol. The Ea value for methane combustion decreased in the sequence bulk La0.6Sr0.4MnO3 > 3DOM La0.6Sr0.4MnO3 > 1.58 wt % Ag/3DOM La0.6Sr0.4MnO3 > 5.71 wt % Ag/3DOM La0.6Sr0.4MnO3 > 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3, with the lower Ea values (37.5−39.1 kJ/mol) being achieved over the yAg/3DOM La0.6Sr0.4MnO3 samples (Table 3). Such a result suggests that methane oxidation might take place more readily over the porous yAg/3DOM La0.6Sr0.4MnO3 samples. The striking difference in Ea could be likely related to the difference in the total number of active sites, which was directly related to the extent of exposed Ag and La0.6Sr0.4MnO3 surfaces and the presence of a strong Ag NPs−La0.6Sr0.4MnO3 interaction. As can be seen from Table 3, the apparent activation energy (37.5 kJ/mol) obtained over 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 was close to or slightly lower than those (73−89 kJ/mol) over MxFe3−xO4 (M = Ni, Mn; x = 0.50−0.65)52 and that (62 kJ/ mol) over Pt/Ce0.64Zr0.15Bi0.21O1.895/Al2O3,53 much lower than those (120−144 kJ/mol) over CuO/Al2O3 and MnO/Al2O3,54 and also lower than those (51−79 kJ/mol) over 10−20 wt % LaCoO3/Ce1−xZrxO2 (x = 0−0.2).55 Therefore, the results of kinetic investigations confirm that the yAg/3DOM La0.6Sr0.4MnO3 samples showed excellent catalytic performance for the combustion of methane, and they are promising catalytic materials for the combustion of methane in practical applications.

feedstock and then switched the SO2-containing atmosphere to SO2-free atmosphere at 550 °C. This temperature was chosen because most samples exhibited high reaction rates around this temperature where the conversion vs temperature curve was very steep, and the results are shown in Figure 8D. However, no significant deactivation due to SO2 introduction was observed over the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample: after this sample was treated in 40 ppm SO2 at 550 °C for 3 h, the catalytic activity decreased by ca. 5%. This result indicates that the 3.63 wt % Ag/3DOM La0.6Sr0.4MnO3 sample was good in SO2 resistance. Such a small loss in catalytic activity was a result due to the strong adsorption of SO2 on the active sites of La0.6Sr0.4MnO3. According to the results of the present investigation and those reported previously, we believe that silver could also improve the resistance to sulfur poisoning, mainly by increasing the concentration of the acidic Mn4+ ions and weakening the SO2 adsorption on the sample. However, the SO 2 introduced could adsorb on the surface of La0.6Sr0.4MnO3 to generate a small amount of sulfate that led to a small decrease in catalytic activity. This performance makes yAg/3DOM La0.6Sr0.4MnO3 an attractive material and worthy of further investigation. 3.9. Activation Energy. There are several reports related to the catalytic kinetic behaviors of VOC combustion in the literature. For instance, Landi et al. claimed that the combustion kinetics of CH4, H2, and CO over a LaMnO3based catalyst was proven to be first order toward methane concentration and zero order toward oxygen concentration.50 Good linear Arrhenius plots were obtained over the supported Pt and Pt−Pd catalysts for methane oxidation when the reaction was first order toward methane concentration.46,50 Therefore, it is reasonable to suppose that the combustion of methane in the presence of excess oxygen (CH4/O2 molar ratio = 1/10) would obey a first-order reaction mechanism with respect to methane concentration (c): r = −kc = (−A exp(−Ea/ RT))c, where r, k, A, and Ea are the reaction rate μmol/(g s)), rate constant (s−1), pre-exponential factor, and apparent activation energy (kJ/mol), respectively. Figure 9 shows the Arrhenius plots for methane combustion at CH4 conversion of