3DOM La0.6Sr0.4MnO3: Highly active nanocatalysts for the oxidation of carbon monoxide and toluene

Journal of Catalysis 305 (2013) 146–153

Contents lists available at SciVerse ScienceDirect

Journal of Catalysis journal homepage: www.elsevier.com/locate/jcat

Au/3DOM La0.6Sr0.4MnO3: Highly active nanocatalysts for the oxidation of carbon monoxide and toluene Yuxi Liu a, Hongxing Dai a,⇑, Jiguang Deng a, Xinwei Li a, Yuan Wang a, Hamidreza Arandiyan b, Shaohua Xie a, Huanggen Yang a, Guangsheng Guo a a Laboratory of Catalysis Chemistry and Nanoscience, Department of Chemistry and Chemical Engineering, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing 100124, China b State Key Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), School of Environment, Tsinghua University, Beijing 100084, China

a r t i c l e

i n f o

Article history: Received 21 February 2013 Revised 27 April 2013 Accepted 29 April 2013 Available online 12 June 2013 Keywords: Supported gold catalyst Three-dimensionally ordered macroporous perovskite-type oxide Strontium-substituted lanthanum manganite Carbon monoxide oxidation Toluene oxidation

a b s t r a c t Three-dimensionally ordered macroporous (3DOM) La0.6Sr0.4MnO3 (LSMO) and its supported gold (xAu/ LSMO, x = 3.4–7.9 wt%) catalysts were prepared using the polymethyl methacrylate-templating and gasbubble-assisted polyvinyl alcohol-protected reduction methods, respectively. There were good correlations of surface-adsorbed oxygen species concentration and low-temperature reducibility with the catalytic activity of the sample for CO and toluene oxidation. Among the LSMO and xAu/LSMO samples, 6.4Au/ LSMO performed the best, giving T50% and T90% values of 19 and 3 °C for CO oxidation and 150 and 170 °C for toluene oxidation, respectively. The apparent activation energies (31–32 and 44–48 kJ/mol) obtained over xAu/LSMO were much lower than those (45 and 59 kJ/mol) obtained over LSMO for the oxidation of CO and toluene, respectively. It is concluded that higher oxygen adspecies concentration, better low-temperature reducibility, and strong interaction between Au and LSMO are responsible for the excellent catalytic performance of 6.4Au/LSMO. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction Catalytic oxidation of carbon monoxide and volatile organic compounds (VOCs) is of considerable interest in the control of environmental pollutants. Since their discovery by Haruta et al., who observed extraordinary catalytic activity of Au/TiO2 for CO low-temperature oxidation [1], supported gold catalysts have attracted tremendous attention. Apart from the Au particle size, the choice of support also plays a key role in the development of active Au nanocatalysts [2]. To obtain high-performance Au catalysts, some metal oxides (e.g., TiO2, Fe2O3, and CeO2) are chosen as supports [3–5]. Recently, investigations on gold catalysts have also been focused on the oxidation of VOCs [6–8]. Several authors reported that the activities of the Co3O4- and Fe2O3-supported gold catalysts were higher than that of Pt/Al2O3 for VOC oxidation [9,10], which was due to the enhancement in oxygen mobility of the active support [10,11]. Compared to other precious metals (e.g., Pt, Rh, and Pd), Au is much less expensive and more abundant. Therefore, the use of Au in the removal of CO and VOCs is continuously explored as a substitute for more expensive noble metals.

⇑ Corresponding author. Fax: +86 10 6739 1983. E-mail address: [email protected] (H. Dai). 0021-9517/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcat.2013.04.025

Perovskite-type oxides (ABO3) are generally considered efficient catalysts for the oxidation of CO and VOCs [12]. They offer the advantages of low cost, high activity, good anti-poisoning ability, and thermal stability over single-metal-oxide catalysts. Since diffusion is often a problem that limits the overall performance of a bulk catalyst, making a catalyst with a porous structure is expected to greatly increase the number of accessible active sites and ultimately enhance the catalytic efficiency. To the best of our knowledge, only a few reports on the preparation of ABO3-supported Au catalysts have been seen in the literature. For example, Zhao and co-workers adopted polymethyl methacrylate (PMMA) as a template to generate three-dimensionally ordered macroporous (3DOM) LaFeO3-supported Au (surface area = ca. 32 m2/g) catalysts and found that Au/3DOM LaFeO3 showed superior performance for the oxidation of soot [13]. Jia et al. observed 100% CO conversion over the nanosized LaCoO3-supported Au (surface area = ca. 14 m2/g) catalyst at 90 °C and 15,000 mL/(g h) [14]. Up to now, however, there have been no reports on the preparation and use for catalytic VOC oxidation of ABO3-supported Au nanocatalysts. Hence, it is highly desirable to establish an effective method for the controlled fabrication of Au supported on porous ABO3 and explore its applications in catalyzing the oxidation of CO and VOCs. Herein, we report a strategy for preparing 3DOM-structured La0.6Sr0.4MnO3 and its supported Au nanocatalysts, which was

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

based on the use of colloidal crystal PMMA as a hard template to generate a 3DOM structure and the gas-bubble-assisted polyvinyl alcohol (PVA)-protected reduction route to obtain supported Au nanomaterials. The aim of this work was to investigate the preparation and catalytic behavior of xAu/3DOM-structured La0.6Sr0.4MnO3 (x = 3.4–7.9 wt%) for the oxidation of CO and toluene. In addition, the effect of water on the performance of the typical catalyst was also examined. 2. Experimental 2.1. Catalyst preparation Well-arrayed colloid crystal template PMMA microspheres with an average diameter of ca. 300 nm were synthesized according to procedures described elsewhere [15]. A 3DOM-structured La0.6Sr0.4MnO3 sample was prepared using the surfactant-assisted PMMA-templating strategy. In a typical procedure, stoichiometric amounts of La(NO3)36H2O, Sr(NO3)2, and Mn(NO3)2 (50 wt% aqueous solution) were dissolved in 3 mL of poly(ethylene glycol) (PEG, MW = 400 g/mol) and 12 mL of water at 50 °C under stirring for 2 h to obtain a transparent solution. A quantity of 1.0 g of L-lysine was dissolved in a HNO3 aqueous solution (5 mol/L), and the pH value of this solution was adjusted to ca. 6 for avoiding the formation of metal hydroxide precipitates in the following steps. Then, the L-lysine-containing solution was mixed with the metal nitrate-containing transparent solution under stirring for 1 h to obtain a uniform precursor solution, to which a certain amount of methanol was added to achieve a total metal concentration of 2 mol/L. Circa 2.0 g of the PMMA hard template was soaked in the above precursor solution for 4 h. After being filtered, the obtained wet PMMA template was dried in air at room temperature (RT) for 48 h and then transferred to a ceramic boat, which was placed in a tubular furnace. The obtained powders were subsequently heated in N2 (200 mL/min) at 300 °C for 3 h, cooled to 50 °C in the same atmosphere, and finally calcined in air (100 mL/min) at 750 °C for 4 h to remove the template, thus generating the 3DOM-structured La0.6Sr0.4MnO3 (denoted as LSMO) catalyst. The LSMO-supported gold (xAu/LSMO) catalysts were prepared via a gas-bubble-assisted PVA-protected reduction method [16]. The typical preparation procedure is as follows: A desired amount of PVA (MW = 10,000 g/mol) was added to a 100 mg/L HAuCl4 aqueous solution (Au/PVA mass ratio = 1.5:1) at RT under vigorous bubbling for 10 min. After rapid injection of an aqueous solution of 0.1 mol/L NaBH4 (Au/NaBH4 molar ratio = 1:5), one could obtain a dark orange-brown solution (so-called gold sol). A desired amount (theoretical Au loading = 5, 8, or 10 wt%) of the LSMO support was then added to the gold sol, and the obtained suspension was subjected to sonication (60 kHz) for 30 s. A gas-bubble-assisted stirring operation with three bubble outlets in solution was used to further agitate the system, and the suspension was vigorously bubbled with N2 for 6 h. The solid was collected by filtration, followed by washing with 2 L of deionized water and drying at 80 °C for 12 h, thus obtaining the xAu/LSMO catalysts. The results of inductively coupled plasma atomic emission spectroscopic (ICP-AES) investigation reveal that the real Au loading was 3.4, 6.4, and 7.9 wt% for the Au-loaded samples, respectively. All of the above steps were carried out by covering all of the containers with a layer of aluminum foil. For comparison purposes, the bulk La0.6Sr0.4MnO3 (denoted as bulk LSMO) and 6.2Au/bulk LSMO catalysts were also prepared via the citric acid-complexing [17] and gas-bubble-assisted PVAprotected reduction routes, respectively. All of the chemicals (A.R. in purity) were purchased from Beijing Chemical Reagent Company and used without further purification.

147

2.2. Catalyst characterization Physicochemical properties of the bulk LSMO, LSMO, xAu/LSMO, and 6.2Au/bulk LSMO catalysts were characterized by means of techniques such as X-ray diffraction (XRD), inductively coupled plasma atomic emission spectroscopy (ICP-AES), N2 adsorption– desorption (BET), scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and hydrogen temperature-programmed reduction (H2 TPR). The detailed procedures are described in the Supplementary data. 2.3. Catalyst evaluation The catalytic activity was evaluated with the sample charged in a continuous-flow fixed-bed U-type quartz microreactor (i.d. = 4 mm). To minimize the effect of hot spots, the sample (50 mg, 40–60 mesh) was diluted with 0.25 g quartz sands (40– 60 mesh). Prior to the test, the sample was treated in O2 (30 mL/ min) at 250 °C for 1 h. After being cooled to a given temperature, the reactant gas containing CO or toluene was passed through the sample bed for 1 h to completely purge the O2. For CO oxidation, the reactant feed was 1 vol% CO + 20% O2 + N2 (balance), giving a space velocity (SV) of ca. 10,000 mL/(g h). In the case of water vapor introduction, certain concentrations (0.5, 2.5, and 5.0 vol%) of H2O were introduced by passing the feed stream through a water saturator in an isothermal bath at 10, 25, and 50 °C, respectively. Catalytic activities of the samples for CO oxidation at low temperatures (below RT) were measured by immersing the microreactor in an ethanol–liquid N2 mixture with certain volumetric ratios. Reactants and products were analyzed on line by a gas chromatograph (GC-14C, Shimadzu) equipped with a thermal conductivity detector (TCD), using a 13 X column. For toluene combustion, the total flow rate of the reactant mixture (1000 ppm toluene + O2 + N2 (balance)) was 16.7 mL/min, giving a toluene/O2 molar ratio of 1/400 and a SV of ca. 20,000 mL/(g h). The 500-, 1000-, and 5000-ppm toluene was generated by passing a N2 flow through a bottle containing pure toluene chilled in an isothermal bath at 15, 0, and 45 °C, respectively. For the change of SV, we altered the mass of catalyst. Reactants and products were analyzed on line by a gas chromatograph (GC-2010, Shimadzu) equipped with a flame ionization detector (FID) and a thermal conductivity detector (TCD), using a Stabilwax@-DA column (30 m in length) for VOC separation and a 1/8 in Carboxen 1000 column (3 m in length) for permanent gas separation. The balance of carbon throughout the investigation was estimated to be 99.5%. 3. Results and discussion 3.1. Real gold content, crystal-phase composition, morphology, pore structure, and surface area The actual Au content of the xAu/LSMO catalysts was determined by the ICP-AES technique; the data are shown in Table 1. It can be seen that the actual Au content of the samples was lower than the initial theoretical values during the preparation processes, indicating that most of the Au in the aqueous solution were deposited on the surface of the LSMO support via the gas-bubble-assisted PVA-protected reduction route. Fig. 1 shows the XRD patterns of the LSMO and supported gold samples. Compared to the XRD pattern of the LSMO support, the loading of Au did not lead to any changes in perovskite structure. The XRD patterns clearly reveal that the crystal structures of all of the samples could be indexed to the rhombohedral perovskite structure (JCPDS PDF# 82-1152). The calculated grain sizes of LSMO in LSMO and xAu/LSMO were 26–27 nm, while those in bulk

148

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

Table 1 BET surface areas, pore volumes, LSMO crystallite sizes (DLSMO), Au particle sizes, and real Au content of the LSMO and xAu/LSMO samples. Sample

BET surface area (m2/g)

Pore volume (cm3/g)

DLSMOa (nm)

Au particle size (nm)b

Au content (wt%)c

LSMO 3.4Au/LSMO 6.4Au/LSMO 7.9Au/LSMO Bulk LSMO 6.2Au/bulk LSMO

32.1 32.9 31.1 31.5 4.2 4.8

0.082 0.076 0.074 0.081 – –

26 27 27 26 139 132

– 2–5 2–5 2–5 – 3–8

– 3.4 6.4 7.9 – 6.2

(110) (104)

a Data determined based on the XRD results according to the Scherrer equation using the FWHM of the (1 1 0) line of LSMO. b Estimated according to the TEM images. c Determined by the ICP-AES technique.

(220) (208)

60

70

(134) (128)

(300) (214) (018)

Intensity (a.u.)

(012)

(111) (202)

6.2Au/bulk-LSMO

(024)

Au

bulk-LSMO

7.9Au/LSMO

6.4Au/LSMO

3.4Au/LSMO

LSMO

10

20

30

40

50

80

2 Theta (Deg.) Fig. 1. XRD patterns of the LSMO and xAu/LSMO samples.

LSMO and 6.2Au/bulk LSMO were in the range 132–139 nm (Table 1). Weak diffraction of Au (1 1 1) was detected at 2h = 38.5°. This result confirms the formation of cubic-phased Au (JCPDS PDF# 04-0784) nanoparticles (NPs) on the support surface. The

(a) LSMO

(b) 3.4Au/LSMO

1 µm

(e) 6.4Au/LSMO

(d) 6.4Au/LSMO

(c) 3.4Au/LSMO

1 µm

(f)7.9Au/LSMO

100 nm

discrepancy in diffraction peak intensity of the bulk and porous LSMO and xAu/LSMO samples indicates the presence of difference in crystallinity, a result due to different preparation methods. It is known that an excessive growth of crystallites would destroy the porous structure of material. By using the PMMA-templating method, the growth of crystallites was confined by the hard template even at a higher temperature; thus, the 3DOM structure of LSMO was retained. Fig. 2 shows SEM images of the LSMO and xAu/LSMO samples. In addition to the submicrosized 6.2Au/bulk LSMO sample, the other samples displayed a 3DOM structure with a macropore diameter of ca. 140 nm and a wall thickness of ca. 25 nm. This result indicates that the loading of Au by gas-bubble-assisted PVAprotected reduction method did not induce significant alteration in 3DOM architecture. Fig. 3 shows the TEM images and the SAED patterns of the LSMO and xAu/LSMO samples as well as the size distribution of Au NPs in the 6.4Au/LSMO sample. It is clearly seen from Fig. 3a and b that the LSMO sample possessed a high-quality 3DOM structure that was composed of interconnected macropores with nanocrystal skeletons, in good agreement with its SEM observations. After loading Au, one can observe uniform Au NPs highly dispersed on the LSMO surface. The size of Au NPs was in the range of 1–6 nm. After statistical analysis of the sizes of more than 200 Au NPs in the TEM images of the 6.4Au/LSMO sample, one can know that its mean diameter was 3.2 nm (Fig. 3l). From the high-resolution TEM (HRTEM) images of the xAu/LSMO samples, the intraplanar spacings (d values) were measured to be ca. 0.27 nm and 0.23 nm (Fig. 3e, h, and k), in good consistency with that of the (1 1 0) crystal plane of the standard LaMnO3 sample (JCPDS PDF# 82-1152) and that of the (1 1 1) crystal plane of the standard Au sample (JCPDS PDF# 040784). The recording of multiple bright electron diffraction rings in the SAED patterns of the LSMO and xAu/LSMO samples (insets of Fig. 3e, h, and k) suggests that the LSMO and xAu/LSMO samples were polycrystalline. Fig. S1 of the Supplementary material shows the N2 adsorption–desorption isotherms and pore-size distributions of the xAu/LSMO samples. Apparently, the N2 adsorption–desorption isotherms were characteristic of typical macropores, as confirmed by their pore-size distributions. Surface areas of the xAu/LSMO samples were 31–33 m2/g, whereas those of the bulk LSMO and 6.2Au/bulk LSMO samples were 4–5 m2/g; pore volumes of the LSMO and xAu/LSMO samples were in the range of 0.074–0.082 m2/g (Table 1).

100 nm

(g) 7.9Au/LSMO

1 µm

1 µm

(h) 6.2Au/bulk-LSMO

100 nm

Fig. 2. SEM images of the LSMO and xAu/LSMO samples.

200 nm

149

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

(a) LSMO

(b) LSMO

(c) 3.4Au/LSMO

20 nm

60 nm

(f) 6.4Au/LSMO

(e) 3.4Au/LSMO

(d) 3.4Au/LSMO

60 nm

20 nm

(h) 6.4Au/LSMO

(g) 6.4Au/LSMO

0.23 nm

Au (111)

Au (111)

LaMnO3 (110)

0.23 nm 0.27 nm 3 nm

20 nm

60 nm

5 nm 45

(j) 7.9Au/LSMO

(i) 7.9Au/LSMO

40

(k) 7.9Au/LSMO

(l) 6.4Au/LSMO

Frequency (%)

35

Au (111)

0.23 nm

30 25 20 15 10 5

60 nm

20 nm

5 nm

0 1

2

3

4

5

6

Au particle size (nm)

Fig. 3. TEM images and SAED patterns (insets) as well as the Au size distribution of the LSMO and xAu/LSMO samples.

3.2. Surface composition, metal oxidation state, and oxygen species XPS is an effective technique for gaining information related to the surface element compositions, metal oxidation states, and adsorbed species of a solid material. Fig. S2 (Supplementary material) illustrates the Mn2p3/2, O1s, and Au4f XPS spectra of the samples. It can be observed from Fig. S2A that the asymmetrical Mn2p3/2 XPS spectrum of each of the samples could be decomposed into three components at BE = 641.3, 642.8, and 644.8 eV, assignable to the surface Mn3+ and Mn4+ species and satellite Mn3+ species, respectively [18]. As summarized in Table 2, the surface Mn4+/Mn3+ molar ratio (1.03) of the LSMO sample was

Table 2 Surface element compositions and H2 consumption of the LSMO and xAu/LSMO samples.

a

Sample

Mn4+/Mn3+ molar ratio

Aud+/Au0 molar ratio

Oads/Olatt molar ratio

H2 consumption (mmol/g)a <550 °C

P550 °C

LSMO 3.4Au/LSMO 6.4Au/LSMO 7.9Au/LSMO Bulk LSMO 6.2Au/bulk LSMO Used 6.4Au/LSMOb Used 6.4Au/LSMOc

1.03 0.52 0.49 0.48 0.66 0.45 0.51 0.53

– 0.25 0.29 0.27 – 0.23 0.28 0.30

1.27 1.49 2.43 1.86 1.02 1.51 2.40 2.41

2.41 2.52 2.74 2.60 1.72 1.83 – –

1.18 1.14 1.16 1.17 1.46 1.41 – –

The data were estimated by quantitatively analyzing the H2 TPR profiles. The 6.4Au/LSMO sample after 96 h of on-stream reaction for CO oxidation under the conditions of CO concentration = 1 vol%, CO/O2 molar ratio = 1/20, SV = 10,000 mL/(g h), and temperature = 8 °C. c The 6.4Au/LSMO sample after 96 h of on-stream reaction for toluene oxidation under the conditions of toluene concentration = 1000 ppm, toluene/O2 molar ratio = 1/400, SV = 20,000 mL/(g h), and temperature = 170 °C. b

higher than that (0.66) of the bulk counterpart. With the introduction of gold, however, the xAu/LSMO and 6.2Au/bulk LSMO samples decreased in surface Mn4+/Mn3+ molar ratio (0.45–0.52). For the asymmetrical O1s spectrum of each of the samples, one can decompose it into three components at BE = 529.0, 531.4, and 533.3 eV (Fig. S3B), ascribable to the surface lattice oxygen (Olatt), 2  adsorbed oxygen (O 2 ; O2 , or O ), and carbonate species, respectively [19,20]. The LSMO sample exhibited a surface Oads/Olatt molar ratio (1.27) much higher than that (1.02) of the bulk LSMO sample, indicating that a higher surface area was beneficial for enhancement in surface-adsorbed oxygen species, in consistency with the result observed in our previous study [21]. After the loading of Au, the Oads/Olatt molar ratio increased remarkably, with the 6.4Au/LSMO sample showing the highest Oads/Olatt molar 2  ratio (2.43). The rise in surface-active oxygen (O 2 ; O2 , or O ) species concentration could give rise to enhanced performance of the catalyst for the deep oxidation of VOCs [22,23]. After loading of Au, the Au4f XPS spectrum of either xAu/LSMO or 6.2Au/bulk LSMO could be decomposed into four components at BE = 83.4, 87.1, 84.6, and 88.2 eV (Fig. S2C). The former two were attributable to the surface Au0 species, whereas the latter two were assignable to the surface Aud+ species [24]. Therefore, there were metallic Au0 (in the majority) and ionic Aud+ (in the minority) species on the surface of the xAu/LSMO and 6.2Au/bulk LSMO samples, with the Aud+/Au0 molar ratios (0.25–0.29) of xAu/LSMO being higher than that (0.23) of 6.2Au/bulk LSMO. When a metal phase is deposited on a support surface, a strong metal–support interaction (SMSI) will exist. In the case of xAu/LSMO, the Au atoms could be oxidized to Aud+ by the surface Mn4+ (to Mn3+) species, causing the Aud+/Au0 molar ratio to rise and the Mn4+/Mn3+ molar ratio to drop. The XPS results reveal that the loading of gold resulted in a rise in Mn3+ content, indicating that there was a strong interaction between Au NPs and LSMO.

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

3.3. Reducibility H2 TPR experiments were carried out to investigate the reducibility of the LSMO and its supported gold samples, and the results are illustrated in Fig. 4. The LSMO sample showed a low-temperature reduction peak at 281 °C with a shoulder at 365 °C (Fig. 4A(a)). The reduction peak at 281 °C was due to the reduction of Mn4+ to Mn3+ as well as the removal of nonstoichiometric oxygen and adsorbed oxygen species, and the shoulder at 365 °C was due to the single-electron reduction of Mn3+ located in a coordination-unsaturated microenvironment, whereas the reduction peak above 550 °C was due to the reduction of the remaining Mn3+ to Mn2+ [25,26]. The two reduction temperatures (437 and 678 °C) of the bulk LSMO sample (Fig. 4A(e)) were much higher than those (281 and 653 °C) of the porous counterpart, indicating that formation of a porous structure facilitated the reduction of LSMO [21]. Quantitative analyses on the H2 TPR profiles can give the H2 consumption of the samples, as summarized in Table 2. For the bulk LSMO and porous LSMO samples, the H2 consumption at low temperatures (<550 °C) was 1.72 and 2.41 mmol/g, while that at high temperatures (P550 °C) was 1.46 and 1.18 mmol/g, respectively. It is apparent that the low-temperature H2 consumption of the LSMO sample was much larger than that of the bulk counterpart, suggesting that the LSMO sample possessed a greater amount of Mn4+, which was confirmed by the XPS results. When Au was loaded onto the LSMO surface, all of the reduction peaks shifted to lower temperatures. This result might be due to the weakening of the Mn–O bonds in LSMO located near the vicinity of gold, leading to higher lattice oxygen mobility [10,27]. The first reduction peak could be attributed to the reduction of the chemically adsorbed oxygen species on the highly dispersed Au NPs (i.e., from Au–Ox to Au) or the interface between Au NPs and LSMO support (i.e., from Mn–Ox–Au to Au) [28]. The shift toward lower temperature was more pronounced for the samples with higher gold loadings. This result implies that Au had a positive effect on the promotion in reduction of manganese species. By comparing the reduction profiles of xAu/LSMO and 6.2Au/bulk LSMO samples, one can see that the porous structure promoted the dispersion of Au NPs, thus favoring the reduction of manganese species. For the xAu/LSMO and 6.2Au/bulk LSMO samples, the H2 consumption at low temperatures (<550 °C) was 2.52–2.74 and 1.83 mmol/g, while that at high temperatures (P550 °C) was 1.14–1.17 and 1.41 mmol/g, respectively. It has generally been accepted that the low-temperature reducibility of a catalyst can be conveniently evaluated using the initial

(A)

267

649

-

6.2Au/bulk LSMO

(f) 678

H 2 consumed (a.u.)

437

-

(e)

bulk LSMO

195

614

7.9Au/LSMO

107

(d)

180

603

(c)

6.4Au/LSMO

244

638

(b)

3.4Au/LSMO

281

50

150

250

365

350

LSMO

450

550

Temperature (oC)

653

650

(a) 750

(where less than 25% oxygen in the sample was removed for the first reduction peak) H2 consumption rate [29,30]. Fig. 4B shows the initial H2 consumption rate as a function of the inverse temperature of the LSMO and xAu/LSMO samples. It is clearly seen that the initial H2 consumption rates of the samples decreased in the order bulk LSMO < 6.2Au/bulk LSMO  LSMO < 3.4Au/LSMO < 7.9Au/ LSMO < 6.4Au/LSMO. This change trend in low-temperature reducibility was in good agreement with that in catalytic performance (shown below). 3.4. Catalytic performance Fig. 5A and B show the catalytic performance of the LSMO and xAu/LSMO samples for the oxidation of CO and toluene, respectively. It can be clearly observed that the LSMO and xAu/LSMO samples performed much better than the bulk LSMO and 6.2Au/ bulk LSMO samples. This result suggests that the formation of a porous structure was beneficial for enhancement of the catalytic performance of the sample for CO and toluene oxidation. It is worth pointing out that toluene was completely oxidized to CO2 and H2O over the as-prepared xAu/LSMO catalysts, and there was no detection of incomplete oxidation products, as confirmed by the good carbon balance (ca. 99.5%) in each run. It is convenient to compare the catalytic activities of the samples by adopting the reaction temperatures T10%, T50%, and T90% (corresponding to CO or toluene conversion = 10%, 50%, and 90%), as summarized in Table 3. For the 6.4Au/LSMO sample, the T50% and T90% values were 19 and 3 °C for CO oxidation and 150 and 170 °C for toluene combustion, respectively, whereas the T50% and T90% values obtained over the 6.2Au/bulk LSMO sample were 51 and 80 °C for CO oxidation and 219 and 243 °C for toluene combustion, respectively. For all of the samples, the change trend in CO or toluene consumption rate versus temperature was similar to that in CO or toluene conversion versus temperature (Fig. S3 of the Supplementary material). The turnover frequency (TOFAu) was calculated on the basis of the number of Au atoms exposed on the sample surface. It is found that the TOFAu value increased in the order 6.4Au/ LSMO > 3.4Au/LSMO > 7.9Au/LSMO > 6.2Au/bulk LSMO (Table 3). The effects of SV and toluene concentration on the catalytic performance of the 6.4Au/LSMO sample are shown in Figs. S4 and S5 of the Supplementary material, respectively. As expected, the catalytic activity decreased with the rise in SV and toluene concentration. Even at a higher toluene concentration (5000 ppm), the LSMO-supported gold catalyst still exhibited good activity for toluene oxidation. To examine the catalytic stability, we carried

Initial H 2 consumption rate (10-4 mol/(mol Mn s))

150

16

(B)

LSMO

14

3.4Au/LSMO 6.4Au/LSMO

12

7.9Au/LSMO bulk-LSMO

10

6.2Au/bulk-LSMO

8 6 4 2 0 1.20

1.40

1.60

1.80

2.00

2.20

2.40

2.60

1000/T (K -1)

Fig. 4. (A) H2 TPR profiles and (B) initial H2 consumption rate as a function of inverse temperature of the LSMO and xAu/LSMO samples.

151

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

100

100

(A)

CO conversion (%)

80

Toluene conversion (%)

LSMO 3.4Au/LSMO 6.4Au/LSMO 7.9Au/LSMO bulk-LSMO 6.2Au/bulk-LSMO

60

40

20

0 -140 -100 -60

(B) LSMO 3.4Au/LSMO 6.4Au/LSMO 7.9Au/LSMO bulk-LSMO 6.2Au/bulk-LSMO

80

60

40

20

0 -20

20

60

80

100 140 180

120

Temperature ( oC)

160

200

240

280

Temperature ( oC)

Fig. 5. (A) CO and (B) toluene conversion as a function of temperature over the LSMO and xAu/LSMO catalysts under the conditions CO concentration = 1 vol%, CO/O2 molar ratio = 1/20, and SV = 10,000 mL/(g h) or toluene concentration = 1000 ppm, toluene/O2 molar ratio = 0 1/400, and SV = 20,000 mL/(g h).

Table 3 Catalytic activities, TOFAu values, and apparent activation energies (Ea) of the LSMO and xAu/LSMO samples for CO and toluene oxidation. Sample

LSMO 3.4Au/LSMO 6.4Au/LSMO 7.9Au/LSMO Bulk LSMO 6.2Au/bulk LSMO

Toluene oxidation T90% (°C)

TOFAu ( 103 s1)a

Ea (kJ/mol)

T50% (°C)

T90% (°C)

TOFAu ( 103 s1)a

Ea (kJ/mol)

124 1 19 12 181 51

152 37 3 17 200 80

– 3.7 6.9 4.2 – 0.54

45 32 31 31 57 34

203 172 150 153 243 219

219 197 170 177 276 243

– 0.34 0.43 0.28 – 0.090

59 48 44 45 74 55

The TOFAu values at low conversions under a kinetically controlled regime at 20 °C for CO oxidation or at 120 °C for toluene oxidation.

110

110

H2 O off

(A) H O on

100

o

CO conversion (%)

(B)

2

Toluene conversion (%)

a

CO oxidation T50% (°C)

T =3 C

90 H2 O on

80

H2O off

70 o

T = 140 C

60

LSMO 6.4Au/LSMO

50

100

H2O on

90

o

T = 220 C

80 o

T = 170 C

70 LSMO 6.4Au/LSMO

60

40

H2 O off

50 0

40

80

120

160

200

240

On-stream reaction time (min)

0

40

80

120

160

200

240

On-stream reaction time (min)

Fig. 6. Effect of water vapor on (A) CO and (B) toluene conversions over LSMO and 6.4Au/LSMO in the presence of 2.5 vol% H2O under conditions of CO concentration = 1 vol%, CO/O2 molar ratio = 1/20, SV = 10,000 mL/(g h) or toluene concentration = 1000 ppm, toluene/O2 molar ratio = 1/400, and SV = 20,000 mL/(g h).

out t96-h on-stream reaction experiments over the best-performing 6.4Au/LSMO sample at 8 °C for CO oxidation and at 170 °C for toluene oxidation, and the results are shown in Fig. S6 (Supplementary material). Obviously, no significant loss in catalytic activity was observed either for CO oxidation or for toluene oxidation. The XRD pattern (Fig. S7 of the Supplementary material) of the used sample was rather similar to that of the fresh one. Furthermore, the surface Mn4+/Mn3+, Aud+/Au0, and Oads/Olatt molar ratios of the used sample after 96 h of CO or toluene oxidation (Table 2 and Fig. S8) were rather close to those of the fresh sample. These results demonstrate that the 6.4Au/LSMO sample was catalytically durable.

To examine the effect of moisture on the catalytic activity, we conducted on-stream CO and toluene oxidation over the LSMO and 6.4Au/LSMO samples at different temperatures. When the catalytic activity reached a steady value, 2.5 vol% of water vapor in the feed was introduced to the reaction system. As can be seen from Fig. 6, the addition of water vapor decreased the T90% value of the LSMO sample by 10% but increased the T90% value of the 6.4Au/ LSMO sample by 9% for CO oxidation; in the case of toluene oxidation, the introduction of water vapor decreased the T90% values of the LSMO and 6.4Au/LSMO samples by 6 and 2%, respectively. Over the LSMO and 6.4Au/LSMO samples, the deactivation due to water vapor addition was reversible. When water vapor was cut off,

152

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

catalytic activities were restored to the initial values. The influence of water vapor concentration on the activity of the 6.4Au/LSMO catalyst for toluene oxidation is shown in Fig. S9A of the Supplementary material. The catalytic activity was not significantly affected at a lower water vapor concentration (0.5 vol%), whereas introduction of a higher water vapor concentration (5.0 vol%) decreased the T90% value by ca. 5.0%. Fig. S9B of the Supplementary material shows the toluene conversion vs. on-stream reaction time over the 6.4Au/LSMO catalyst in the presence of 2.5 vol% water vapor in the feedstock. An activity deactivation in the presence of water vapor and a recovery in the case of cutting off water vapor were observed. Furthermore, the XRD pattern (Fig. S10A of the Supplementary material) of the used catalyst was rather similar to that of the fresh one, and the surface area (30.3 m2/g) of the former was close to that (31.1 m2/g) of the latter. The TEM image (Fig. S10B of the Supplementary material) of the used sample further reveals that the gold NPs were well stabilized on the porous LSMO surface. These results show that the 6.4Au/LSMO catalyst was 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. It has been reported that in the presence of water, water was adsorbed onto the metal oxide support rather than onto the precious metal [31]. Over the 6.4Au/LSMO sample for CO oxidation, the positive effect of water vapor might be due to the fact that under ambient pressure, the active interface sites were gradually covered with the inert carbonate species accumulated during the reaction, which decreased the adsorption of CO and its reaction rate, but the deactivated sample could be regenerated via the addition of moisture in the feed gas mixture [32]. For the oxidation of toluene, the activity inhibition of the sample by water vapor was a result of the competitive adsorption of H2O and toluene on the surface of 6.4Au/LSMO. The catalytic activity of the 6.4Au/LSMO sample for the oxidation of CO and toluene was further compared with those of other supported gold samples. The specific reaction rates (mmol/(gAu s)) normalized per gram of gold and the turnover frequencies (TOFAu) normalized per single gold atom exposed on the surface are listed in Table S1 (Supplementary material). Under similar reaction conditions, 6.4Au/LSMO was the most active supported Au sample for toluene oxidation so far reported in the literature [10,23,33–36]. For CO oxidation, however, 6.4Au/LSMO showed a catalytic activity inferior to the TiO2-, CeO2-, and Mn2O3-supported gold samples and porous Co3O4 [37–40], but it was more active than the SiO2-, nanosized LaCoO3-supported gold, and nanosized Mn2O3 samples [14,38,41].

3.5. Kinetic parameters According to the Weisz–Prater criterion, when the effectiveness factor g P 0.95 and reaction order n = 1, the dimensionless Weisz– Prater parameter (NW–P) value is less than 0.3, which can be considered a sufficient condition for the absence of significant mass diffusion limitations [42]. At CO or toluene conversion 620%, we carried out the Weisz–Prater analysis and calculated the NW–P values. The NW–P value was 1.26  102 for CO oxidation and 1.63  103 for toluene oxidation, which were much less than 0.3. Therefore, no significant mass transfer problems were present in our catalytic system. In the past years, there have been reports on the kinetics of catalytic oxidation of CO and VOCs. For example, Jia et al. claimed that the oxidation of CO over the CuO/Ce1xCuxO2d catalyst was first-order toward CO concentration and zero toward oxygen concentration [43]. Wong et al. pointed out that the oxidation of butyl acetate over AgZSM-5 was first-order toward butyl acetate concentration and zero toward oxygen concentration [44]. By assuming first-order kinetics with respect to toluene concentration and zero-order kinetics with respect to oxygen concentration, Alifanti et al. obtained good linear Arrhenius plots for the oxidation of toluene over the ceria–zirconia-supported LaCoO3 catalysts [45]. Therefore, it is reasonable to suppose that the oxidation of CO and toluene in the presence of excess oxygen (CO/O2 molar ratio = 1/20 and toluene/O2 molar ratio = 1/400) would obey a first-order reaction mechanism with respect to CO or toluene concentration (c),

r ¼ kc ¼ ðA expðEa =RTÞÞc; where r, k, A, and Ea are the reaction rate (mol/s), rate constant (s1), pre-exponential factor, and apparent activation energy (kJ/mol), respectively. The Arrhenius plots for CO and toluene oxidation over the LSMO and xAu/LSMO samples are shown in Fig. 7, and their apparent activation energies are summarized in Table 3. It can be observed that the Ea value for either CO oxidation or toluene oxidation decreased in the sequence bulk LSMO > LSMO > 6.2Au/bulk LSMO > 3.4Au/LSMO > 6.4Au/LSMO  7.9Au/LSMO, with the lower Ea values (31–32 and 44–48 kJ/mol for CO and toluene oxidation, respectively) being achieved over the xAu/LSMO samples. This result suggests that CO and toluene oxidation might proceed more readily over the porous LSMO-supported gold samples. The big discrepancy in Ea value might be related to the difference in the total number of active sites and the presence of a strong Au–LSMO interaction. 0

-1

(A) CO oxidation

(B) Toluene oxidation -1

-2 -2

ln k

ln k

-3 -4

-3

LSMO

-5

LSMO

3.4Au/LSMO

-4

6.4Au/LSMO 7.9Au/LSMO

-6

-5

bulk-LSMO 6.2Au/bulk-LSMO

-7 1.4

1.8

2.2

3.4Au/LSMO 6.4Au/LSMO

2.6

3.0

7.9Au/LSMO bulk-LSMO 6.2Au/bulk-LSMO

3.4

1000/T (K-1)

3.8

4.2

4.6

-6 1.9

2.1

2.3

2.5

2.7

2.9

1000/T (K-1)

Fig. 7. The Arrhenius plots for the oxidation of (A) CO and (B) toluene over the LSMO and xAu/LSMO catalysts under the conditions CO concentration = 1 vol%, CO/O2 molar ratio = 1/20, and SV = 10,000 mL/(g h) or toluene concentration = 1000 ppm, toluene/O2 molar ratio = 1/400, and SV = 20,000 mL/(g h).

Y. Liu et al. / Journal of Catalysis 305 (2013) 146–153

As we know, the presence of a 3DOM network within the LSMO sample allows better diffusion of the reactants (CO, toluene, and oxygen), hence leading to easy accessibility of active sites for the reactant molecules. This can consequently increase the oxidation rates of the reactants, and catalytic performance will be improved as a result. The oxidation of CO and toluene over xAu/LSMO could proceed via a suprafacial mechanism. After loading of gold, partial surface Mn4+ species were reduced to Mn3+ species by Au0 via the strong SMSI between Au NPs and LSMO, resulting in the formation of surface oxygen vacancies. The gas-phase O2 molecules were then activated in the oxygen vacancies near the Au–LSMO interface to active oxygen species. On the other hand, addition of Au could significantly promote the adsorption of CO and toluene, enhancing the migration of chemisorbed CO and toluene to the Au–LSMO interface. Hence, the oxidation of CO and toluene would occur between the chemisorbed CO or toluene and active oxygen species at the Au–LSMO interface.

References

4. Conclusions

[14] [15] [16]

The LSMO and xAu/LSMO samples could be prepared using the PMMA-templating and gas-bubble-assisted PVA-protected reduction methods, respectively. There were good correlations of Oads concentration and low-temperature reducibility with catalytic activity of the sample for the oxidation of CO and toluene. The 6.4Au/LSMO sample showed the best catalytic performance (T50% and T90% values were 19 and 3 °C for CO oxidation at SV = 10,000 mL/(g h) and 150 and 170 °C for toluene oxidation at SV = 20,000 mL/(g h), respectively). The effect of water vapor on catalytic activity was positive over 6.4Au/LSMO for CO oxidation, whereas it was negative over LSMO for CO oxidation and over LSMO and 6.4Au/LSMO for toluene oxidation. The deactivation due to moisture addition was reversible. The apparent activation energies obtained over xAu/LSMO were 31–32 and 44–48 kJ/mol for the oxidation of CO and toluene, respectively. We believe that the excellent catalytic performance of 6.4Au/LSMO might be associated with its higher Oads concentration, better low-temperature reducibility, and strong interaction between Au NPs and LSMO.

[1] [2] [3] [4]

[5] [6] [7] [8] [9] [10] [11] [12] [13]

[17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]

Acknowledgments The work presented above was financially supported by the NSF of China (Grants 21077007 and 20973017), the Discipline and Postgraduate Education (Grant 005000541212014), and the Funding Project for Academic Human Resources Development in Institutions of Higher Learning under the Jurisdiction of Beijing Municipality (Grant PHR201107104). We also thank Professor Chak Tong Au (Department of Chemistry, Hong Kong Baptist University) and Jianping He (State Key Laboratory of Advanced Metals and Materials, University of Science & Technology Beijing) for doing the XPS and SEM analyses, respectively. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcat.2013.04.025.

[33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45]

153

M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. 2 (1987) 405–408. B.K. Min, C.M. Friend, Chem. Rev. 107 (2007) 2709–2724. T.A. Ntho, J.A. Anderson, M.S. Scurrell, J. Catal. 261 (2009) 94–100. G.J. Hutchings, M.S. Hall, A.F. Carley, P. Landon, B.E. Solsona, C.J. Kiely, A. Herzing, M. Makkee, J.A. Moulijn, A. Overweg, J.C. Fierro-Gonzalez, J. Guzman, B.C. Gates, J. Catal. 242 (2006) 71–81. S. Carrettin, P. Concepción, A. Corma, J.M. López Nieto, V.F. Puntes, Angew. Chem. Int. Ed. 43 (2004) 2538–2540. S. Scirè, L.F. Liotta, Appl. Catal. B 125 (2012) 222–246. T. Barakat, J.C. Rooke, E. Genty, R. Cousin, S. Siffert, B.L. Su, Energy Environ. Sci. 6 (2013) 371–391. C.Y. Ma, D.H. Wang, W.J. Xue, B.J. Dou, H.L. Wang, Z.P. Hao, Environ. Sci. Technol. 45 (2011) 3628–3634. M. Haruta, Catal. Today 36 (1997) 153–166. S. Minicò, S. Scirè, C. Crisafulli, R. Maggiore, S. Galvagno, Appl. Catal. B 28 (2000) 245–251. D. Andreeva, T. Tabakova, L. Llieva, A. Naydenov, D. Mehanjiev, M.V. Abrashev, Appl. Catal. A 209 (2001) 291–300. M.A. Peña, J.L.G. Fierro, Chem. Rev. 101 (2001) 1981–2018. Y.C. Wei, J. Liu, Z. Zhao, Y.S. Chen, C.M. Xu, A.J. Duan, G.Y. Jiang, H. He, Angew. Chem. Int. Ed. 50 (2011) 2326–2329. M.L. Jia, Menggentuya, Z. Bao, Y.N. Shen, Catal. Lett. 134 (2010) 87–92. H.N. Li, L. Zhang, H.X. Dai, H. He, Inorg. Chem. 48 (2009) 4421–4434. M. Comotti, W.C. Li, B. Spliethoff, F. Schüth, J. Am. Chem. Soc. 128 (2006) 917– 924. P. Ciambelli, S. Cimino, S.D. Rossi, M. Faticanti, L. Lisi, G. Minelli, I. Pettiti, P. Porta, G. Russo, M. Turco, Appl. Catal. B 24 (2000) 243–253. S. Ponce, M.A. Peña, J.L.G. Fierro, Appl. Catal. B 24 (2000) 193–205. A. Machocki, T. Ioannides, B. Stasinska, W. Gac, G. Avgouropoulos, D. Delimaris, W. Grzegorczyk, S. Pasieczna, J. Catal. 227 (2004) 282–296. J.G. Deng, H.X. Dai, H.Y. Jiang, L. Zhang, G.Z. Wang, H. He, C.T. Au, Environ. Sci. Technol. 44 (2010) 2618–2623. Y.X. Liu, H.X. Dai, Y.C. Du, J.G. Deng, L. Zhang, Z.X. Zhao, C.T. Au, J. Catal. 287 (2012) 149–160. B. Solsona, T. García, S.H. Taylor, G.J. Hutchings, M. Makkee, Appl. Catal. A 365 (2009) 222–230. H. Wu, L. Wang, J. Zhang, Z. Shen, J. Zhao, Catal. Commun. 12 (2011) 859–865. M.M. Han, X.J. Wang, Y.N. Shen, C.H. Tang, G.S. Li, R.L. Smith, J. Phys. Chem. C 114 (2010) 793–798. F. Teng, W. Han, S.H. Liang, B. Gaugeu, R.L. Zong, Y.F. Zhu, J. Catal. 250 (2007) 1–11. Z. Sarshar, F. Kleitz, S. Kaliaguine, Energy Environ. Sci. 4 (2011) 4258–4269. L.C. Wang, Q. Liu, X.S. Huang, Y.M. Liu, Y. Cao, K.N. Fan, Appl. Catal. B 88 (2009) 204–212. Y.C. Wei, J. Liu, Z. Zhao, A.J. Duan, G.Y. Jiang, J. Catal. 287 (2012) 13–29. K.D. Chen, S.B. Xie, A.T. Bell, E. Iglesia, J. Catal. 198 (2001) 232–242. H.X. Dai, A.T. Bell, E. Iglesia, J. Catal. 221 (2004) 491–499. R. Rajasree, J.H.B.J. Hoebink, J.C. Schouten, J. Catal. 223 (2004) 36–43. M. Daté, M. Okumura, S. Tsubota, M. Haruta, Angew. Chem. Int. Ed. 43 (2004) 2129–2132. S. Scirè, P.M. Riccobene, C. Crisafulli, Appl. Catal. B 101 (2010) 109–117. M. Hosseini, S. Siffert, H.L. Tidahy, R. Cousin, J.F. Lamonier, A. Aboukais, A. Vantomme, M. Roussel, B.L. Su, Catal. Today 122 (2007) 391–396. V. Blasin-Aubé, J. Belkouch, L. Monceaux, Appl. Catal. B 43 (2003) 175–186. L.F. Liotta, M. Ousmane, G. Di Carlo, G. Pantaleo, G. Deganello, A. Boreave, A. Giroir-Fendler, Catal. Lett. 127 (2009) 270–276. Tana, F.G. Wang, H.J. Li, W.J. Shen, Catal. Today 175 (2011) 541–545. L.C. Wang, X.S. Huang, Q. Liu, Y.M. Liu, Y. Cao, H.Y. He, K.N. Fan, J.H. Zhuang, J. Catal. 259 (2008) 66–74. G.R. Bamwenda, S. Tsubota, T. Nakamura, M. Haruta, Catal. Lett. 44 (1997) 83– 87. R. Yu, Z. Ma, L.P. Qian, S. Dai, H.Y. He, P.G. Bruce, Catal. Lett. 131 (2009) 146– 154. H.G. Zhu, Z. Ma, J.C. Clark, Z.W. Pan, S.H. Overbury, S. Dai, Appl. Catal. A 326 (2007) 89–99. P.B. Weisz, C.D. Prater, Adv. Catal. Relat. Subj. 6 (1954) 143–196. A.P. Jia, G.S. Hu, L. Meng, Y.L. Xie, J.Q. Lu, M.F. Luo, J. Catal. 289 (2012) 199–209. C.T. Wong, A.Z. Abdullah, S. Bhatia, J. Hazard. Mater. 157 (2008) 480–489. M. Alifanti, M. Florea, S. Somacescu, V.I. Pârvulescu, Appl. Catal. B 60 (2005) 33–39.