Catalytic activity of perovskite-type doped La0.08Sr0.92Ti1−xMxO3−δ (M = Mn, Fe, and Co) oxides for methane oxidation

Catalytic activity of perovskite-type doped La0.08Sr0.92Ti1−xMxO3−δ (M = Mn, Fe, and Co) oxides for methane oxidation

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2 Available online at www.sciencedirect.com S...

870KB Sizes 8 Downloads 31 Views

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

Catalytic activity of perovskite-type doped La0.08Sr0.92Ti1LxMxO3Ld (M [ Mn, Fe, and Co) oxides for methane oxidation Jong Seol Yoon a, Ye-Sol Lim a, Byung Hyun Choi b, Hae Jin Hwang a,* a b

Division of Materials Science and Engineering, Inha University, Republic of Korea Korea Institute of Ceramic Engineering and Technology, Republic of Korea

article info

abstract

Article history:

Transition metal doped La0.08Sr0.92M0.20Ti0.80O3d (M ¼ Mn, Fe, and Co) perovskite oxides

Received 10 December 2013

were synthesized by the Pechini method. The methane oxidation behavior and the po-

Received in revised form

larization resistance of the solid oxide fuel cells (SOFCs) with the perovskite oxides as

8 February 2014

anode was subsequently measured as a function of operation temperature. Surface atomic

Accepted 1 March 2014

concentrations of the perovskite oxides were evaluated using X-ray photoelectron spec-

Available online 13 April 2014

troscopy (XPS) and their relationship to the catalytic activity were discussed with respect to the transition metal dopant. The complete oxidation of methane was predominant in the

Keywords:

low-temperature region, while the partial oxidation of methane occurred at high temper-

Methane oxidation

atures. Fe- and Co-doped perovskites showed better catalytic activity for the methane

Perovskite oxides

oxidation reaction than Mn-doped powder. This phenomenon could be explained by the

Solid oxide fuel cells

high atomic concentration with low oxidation states and the resulting high oxygen va-

Oxide anodes

cancy concentration in the Fe- and Co-doped perovskite powder samples.

Doped strontium titanate

Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

Introduction Recently, there has been a growing interest in developing oxide anode materials for direct hydrocarbon solid oxide fuel cells (SOFCs) [1,2]. At the anode, the main reaction is the oxidation of the fuels. Therefore, electronic conductivity, oxygen ionic conductivity, and catalytic activity are very important properties for SOFC anode. Traditionally, a nickelyttria-stabilized zirconia (Ni-YSZ) cermet has been used for the SOFC anode; however, some problems are evidenced with the use of the Ni-YSZ cermet when hydrocarbons containing

sulfur impurities are used as a fuel. For instance, the effective electrode area reduction by carbon deposition or sulfur poisoning is the most frequently encountered issue [3]. In addition, microstructural changes in the Ni-YSZ anode can occur during the redox cycles [4]. To overcome the above-mentioned limitations of the NiYSZ anode, oxide anode materials are being considered as alternative anode materials for SOFCs. Among the potential candidates discussed in the literature, perovskite-type strontium titanates (SrTiO3), which are doped by rare earth elements, are the most promising candidates for SOFC anodes from the viewpoint of ability to maintain phase stability and

* Corresponding author. Division of Materials Science and Engineering, Inha University, 253 Yonghyun-dong, Nam-gu, Incheon 402-751, Republic of Korea. Tel.: þ82 32 860 7521, Fax : þ82 32 862 4482. E-mail address: [email protected] (H.J. Hwang). http://dx.doi.org/10.1016/j.ijhydene.2014.03.008 0360-3199/Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

7956

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

electronic conductivity under a reducing environment. It is well known that substituting divalent Sr2þ with trivalent atoms such as Y3þ or La3þ can improve the electronic conduction of SrTiO3 under reduction conditions [5,6]. When transition metals such as Mn, Fe, and Co are replaced in the Ti sub-lattice, both, the oxygen ionic conductivity related to catalytic activity and oxygen vacancy formation, are improved [7,8]. There are some studies investigating the catalytic activity of perovskite-type oxide materials for methane oxidation [9,10]. According to these studies, the catalytic activity depends on the oxidation state of the B-site transition metal elements and oxygen non-stoichiometry that is closely related to oxygen vacancy formation [11,12]. Therefore, it is expected that the catalytic activity of the oxide anode materials can be modified by the partial substitution of strontium and titanium using lanthanum and transition metal elements, respectively. In this study, 8 mol% of La- and 20 mol% of transition metals such as Mn-, Fe-, or Co-doped SrTiO3 were synthesized by the Pechini method. Electrolyte supported solid oxide fuel cells with the perovskite oxides as anode were fabricated. The CH4 conversion rate on the perovskite oxide powders and the polarization resistance of the cell with the perovskite oxide anode were measured as a function of temperature. The effect of surface atomic concentration and oxidation state of the perovskite oxides on the catalytic activity for methane oxidation was also investigated in terms of the surface condition of the B-site dopants.

Experimental procedure Powder synthesis Three perovskite-type oxide powders with formulas of La0.08Sr0.92Mn0.20Ti0.80O3d (LSMTO), La0.08Sr0.92Fe0.20Ti0.80O3d (LSFTO), and La0.08Sr0.92Co0.20Ti0.80O3d (LSCTO) were synthesized by the Pechini method. Strontium nitrate (98.0%, Aldrich), lanthanum nitrate (99.0%, Aldrich), titanium isopropoxide (97.0%, Aldrich), cobalt nitrate (99.9%, Junsei), manganese nitrate (99.9%, Kanto Chemical Co., Inc.), iron nitrate (99.9%, Aldrich), citric acid (99.5%, Junsei), and ethylene glycol (99.5%, Junsei) were used as the starting materials. The molar ratio of metal:citric acid:ethylene glycol was 1:2:8. After the titanium isopropoxide was dissolved in ethylene glycol, citric acid and other metal nitrates were added sequentially. The whole process was carried out in a glove box under deactivated nitrogen gas atmosphere. To make a homogeneous precursor, the mixture was stirred and heated on a hot plate at 80  C for 1 h. Then, the obtained precursor solution was further heated at 120  C for 8 h to remove the water produced by the polyesterification between ethylene glycol and citric acid. During heating at 120  C, the solution transformed into a viscous polymeric resin. The resin was dried at 250  C for 12 h in an oven and was subsequently calcined at 800, 850, and 900  C for 5 h.

Fig. 1 e XRD patterns of La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO) calcined at 800  C for 5 h in air.

single cells (La0.6Sr0.4Co0.2Fe0.8O3d (LSCF)-Gd0.1Ce0.9O2 (GDC, cathode)/GDC (buffer)/ScSZ (electrolyte)/Anodes). The commercially available ScSZ (89 mol% ZrO2-10 mol% Sc2O31 mol% Al2O3, Daich Kigenso Kagau Kogyo Co. Ltd., Japan) powder was pressed into a pellet and sintered at 1500  C for 5 h in air. A ScSZ disc with a diameter of 20 mm and a thickness of 0.2 mm was obtained. Anode (LSMTO, LSFTO, and LSCTO), LSCF (Fuel Cell Materials, USA)-GDC (gadolinium-doped ceria, Anan Kasei, Japan) cathode, and GDC powders were mixed with an organic vehicle (Fuel Cell Materials, USA) to form pastes. The anode pastes were then applied on the ScSZ electrolyte using screen-printing method, which followed by sintering at 1000  C for 2 h. The GDC buffer layer was deposited on the opposite side of the ScSZ electrolyte using screenprinting method using a GDC paste, which was then fired at 1400  C for 2 h in air. Then, the LSCF-GDC was applied on the GDC buffer layer by screen-printing and sintering at 1000  C for 2 h.

Characterization For phase characterization, X-ray diffraction (XRD, RU-200B, Rigaku Co. Ltd.) was performed using Ni-filtered CuKa radiation. The microstructure of the electrode was observed by field emission scanning electron microscopy (FESEM,

Table 1 e BET surface areas of powder samples calcined at 800, 850, and 900  C, respectively. BET surface area (m2/g)

Cell fabrication The electrochemical performance of the LSMTO, LSFTO, and LSCTO anodes were measured using an electrolyte supported

(La0.08Sr0.92)(Mn0.20Ti0.80)O3d (La0.08Sr0.92)(Fe0.20Ti0.80)O3d (La0.08Sr0.92)(Co0.20Ti0.80)O3d

Calcination temperature,  C 800

850

900

15.96 14.86 22.44

12.06 12.33 6.72

6.73 7.49 2.65

7957

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

Fig. 2 e SEM images of (a) La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), (b) La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and (c) La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO) powder samples; LSMTO and LSFTO were calcined at 900  C and LSCTO was calcined at 850  C.

Raw Base line Fitted line Olatt.

30000

20000

0 530

532

534

Binding energy (eV)

-1

Oads. 20000

0 528

Counts (S )

30000

10000

10000

40000

Raw Base line Fitted line Olatt.

-1

Oads.

(b) Counts (S )

-1

Counts (S )

40000

40000

(a)

(c)

528

530

532

534

Binding energy (eV) Raw Base line Fitted line Olatt.

30000

Oads.

20000

10000

0 528

530

532

534

Binding energy (eV)

Fig. 3 e XPS spectra of the O 1s state in (a) La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), (b) La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and (c) La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO) powder samples.

7958

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

HITACHI, S-4200). The surface area was determined by BrunauereEmmetteTeller (BET) analysis from the amount of N2 gas adsorbed at various partial pressures (0.01 < p/p0 < 1) (ASAP 2010, Micrometrics, USA). The surface composition and oxidation state of the perovskite oxides were analyzed using the XPS technique. In this study, we focused on the oxygen and B-site transition metal species. The XPS spectra were calibrated with respect to the C 1s binding energy of 285.0 eV, a result of atmospheric contamination; C 1s can always be expected from a solid surface [13]. The calibrated XPS curves were deconvoluted by the standard GaussianeLorentzian deconvolution method. The methane oxidation was measured in a fixed-bed quartz reactor (inner diameter: 18 mm; length: 760 mm). 100 mg of perovskite powders was diluted with 1 g of pure quartz sands (50e70 mesh) to avoid sintering between perovskite particles at high temperatures. The diluted powder samples were inserted into the middle of a quartz tube between two quartz wools. Catalytic oxidation of methane was carried out by feeding the reactor with a total flow rate of 50 SCCM of high-purity gas mixture of 20 cc of 10% CH4 and 10 cc of 10% O2 balanced by argon. The gas-flow rate was controlled using a mass-flow controller (SEC-400, Horiba). The reactant gas concentrations were analyzed by on-line micro gas chromatography (CP-4900, Varian) which was connected to the vent line. The electrochemical properties of the LSMTO, LSFTO, and LSCTO anodes were measured for anode/ScSZ interface using AC impedance spectroscopy (IM6e, Zaner) in a 3-electrode

6000

Fig. 1 shows the XRD patterns of the LSMTO, LSFTO, and LSCTO powders calcined at 800  C for 5 h. All peaks were assigned to a cubic perovskite structure. There were no peaks present that indicated secondary phases or starting materials, suggesting that the lanthanum and transition metal dopants are completely dissolved in the perovskite structure. In addition, it appears that transition metal doping doesn’t lead to a change in the peak shape or intensity significantly. A slight shift to higher angles, observed in the XRD pattern of the LSFTO sample, may be attributed to the large ionic radii of Fe3þ ions compared to the Mn3þ or Co3þ ions, which will be described later. Table 1 shows the BET surface areas of LSMTO, LSFTO, and LSCTO powder samples calcined at 800, 850, and 900  C, respectively, for 5 h. For all samples, the specific surface areas (SSAs) decreased with increasing calcination temperature. The SSAs for the Mn- and Fe-doped powder samples were almost the same values, while those for the Co-doped powder sample were found to be much smaller than the other two samples and this tendency was much more pronounced at the

4000

(b)

Raw Base line Fitted curve Curve for Fe Curve for Fe

3000

4000 -1

Counts (s )

-1

Results

Raw Base line Fitted curve Curve for Mn Curve for Mn

(a)

5000

Counts (s )

configuration under wet hydrogen and wet 50% CH4/Ar mixture feed gas. The impedance spectra were obtained in a frequency range of 1 MHze0.01 Hz with applied AC voltage amplitude of 20 mV at 800  C. Fuel gas with 3% water vapor was fed to anode side, and the dry air was fed to cathode side.

3000 2000

2000

1000

1000 0

0 639

640

641

642

643

644

Binding energy (eV) 7000 6000

(c)

-1

709

710

711

712

713

714

715

Binding energy (eV) Raw Base line Fitted curve Curve for Co Curve for Co

5000

Counts (s )

708

4000 3000 2000 1000 0 778

780

782

784

Binding energy (eV) Fig. 4 e XPS spectra of (a) the Mn 2p state in La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), (b) the Fe 2p state in La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and (c) the Co 2p state in La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO) powder samples.

7959

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

high calcination temperature. This phenomenon suggests that doping the perovskite structure with cobalt can lead to an increase in the sinterability and shrinkage of the powder sample. SEM images of the LSMTO, LSFTO, and LSCTO powder samples are shown in Fig. 2. Fig. 2(a) and (b) shows the LSMTO and LSFTO sample, respectively, calcined at 900  C, and Fig. 2(c) shows the LSCTO powder sample calcined at 850  C. As can be seen in Fig. 2, all powder samples exhibited a similar particle size, particle morphology, and degree of agglomeration. The particle size of the powder samples was estimated to be approximately 100 nm. To investigate the surface structure of the three perovskite powder samples, XPS analysis was carried out. The XPS spectra for the O 1s, Ti 2p, Mn 2p, Fe 2p, and Co 2p states are represented in Figs. 3e6, respectively. Three O 1s spectra for the three samples are shown in Fig. 3. Two peaks were observed at 529.0 and 531.0e531.4 eV in the spectra. The peak at 531.3 eV is attributed to the surface-adsorbed oxygen from hydroxide [14,15] or carbonate compounds [16,17], while the peak at 529.0 eV corresponds to the lattice oxygen in the perovskite structure [18,19]. The atomic surface concentrations calculated from the peak intensity ratio for the O 1s spectra are shown in Table 2. Generally, the lattice oxygen concentration is closely related to the oxygen vacancy concentration [20]. Doping by transition metals such as Mn, Fe, and Co can result in the formation of oxygen vacancies and thereby causing a decrease

(a)

in the lattice oxygen concentration. As can be seen in Table 2, the lattice oxygen concentration of LSMTO (48.5%) was greater than that of LSFTO (40.84%) or LSCTO (39.31%). This result means that more oxygen vacancies were produced in the perovskite structure in the LSFTO and LSCTO samples. XPS spectra of Mn 2p, Fe 2p, and Co 2p are shown in Fig. 4(a)e(c) for the LSMTO, LSFTO, and LSCTO sample, respectively. For the LSMTO sample, the Mn 2p spectrum consists of two peaks at 641.5 and 642.9 eV that can be assigned to Mn3þ and Mn4þ ions [21]. Similar to the LSMTO, two peaks were confirmed for the LSFTO and LSCTO samples. In Fig. 4(b), the peaks at 710.5 and 712.6 eV indicate the presence of Fe3þ and Fe4þ ions, respectively [13]. In addition, for the Co 2p spectrum in Fig. 4(c), the peaks at 779.3 and 780.3 eV indicate the presence of Co3þ and Co4þ ions, respectively [22]. Consequently, it can be confirmed the transition metal dopants have two oxidation states in the perovskite structure. For the LSMTO sample, the ratio of Mn3þ to Mn4þ was calculated to be approximately 1.5. On the other hand, the ratio increased to 3.0 in the LSFTO and LSCTO samples, which suggests that iron or cobalt are easily reduced compared to manganese in the perovskite structure. On the other hand, the titanium species that is a host for the B-site sub-lattice in the perovskite structure is represented by one peak of Ti4þ in the XPS spectra as can be seen in Fig. 5. Accordingly, it can be considered that the transition metal dopants such as Mn, Fe, and Co can play an important role as the active site for the methane oxidation reaction.

Raw Base line Curve for Ti

15000

-1

Counts (s )

-1

Counts (s )

30000

20000

Raw Base line Curve for Ti

10000

5000

10000

0

0 456

457

458

459

456

Binding energy (eV)

(c)

457

458

459

Binding energy (eV) Raw Base line Curve for Ti

20000

-1

Counts (s )

(b)

10000

0 456

457

458

459

Binding energy (eV) Fig. 5 e XPS spectra of the Ti 2p state in (a) La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), (b) La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and (c) La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO) powder samples.

7960

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

100

100

(a)

La0.08Sr0.92Ti0.80Mn0.20O3-δ

80

La0.08Sr0.92Ti0.80Mn0.20O3-δ

80

La0.08Sr0.92Ti0.80Co0.20O3-δ

H2 selectivity (%)

CH4 conversion (%)

La0.08Sr0.92Ti0.80Co0.20O3-δ 60

40

(b)

La0.08Sr0.92Ti0.80Fe0.20O3-δ

La0.08Sr0.92Ti0.80Fe0.20O3-δ

60

40

20

20

0

0 300

400

500

600

o

700

800

300

900

400

500

600

o

700

800

900

Temperature ( C)

Temperature ( C)

Fig. 6 e (a) Methane conversion rate and (b) hydrogen selectivity of La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO) samples.

Fig. 6 shows the methane conversion rate and hydrogen selectivity of the LSMTO, LSFTO, and LSCTO samples as a function of operating temperature. It appears that the methane conversion rate exhibited quite a complicated behavior as a function of temperature. First, the methane conversion rate gradually increases with increasing temperature, shows a plateau at the intermediate temperature region, and thereafter increases again at higher temperatures. On the other hand, the hydrogen selectivity was below 5% for the three samples at low temperatures and started to increase moderately at 700  C. The methane conversion rate of the LSFTO sample was slightly higher than that of the LSCTO sample; it was maintained at almost the same value below 600  C, while it was higher in LSFTO than in LSCTO above 700  C. Among the three perovskite powder samples, LSMTO exhibited the lowest methane conversion rate across the entire temperature region. For the hydrogen selectivity, a similar trend was observed as is evident in Fig. 6(b). The hydrogen selectivity increased in the order LSFTO > LSCTO > LSMTO. The impedance spectra of three single cells with different anodes measured in H2 and CH4 atmospheres are shown in Fig. 7. All the data were obtained at 800  C under open circuit voltage (OCV).

From the impedance spectra, the polarization resistances of LSMTO, LSFTO, and LSCTO anodes in H2 were calculated to be 9.0, 2.5, and 3.6 U cm2, respectively. Each impedance spectrum appeared to consist of three arcs at high (w104 Hz), intermediate (w102 Hz), and low frequency (w101 Hz), respectively that were increased when the cells were operated with CH4. Compared with the high and intermediate arcs, the increase in the low frequency arc is much more significant, suggesting that the low frequency arc is attributed to the oxidation of methane on the perovskite oxide anodes. Both in H2 and CH4, the LSFTO anode showed the lowest polarization resistance. This result suggests that the catalytic activity for hydrogen and methane oxidation reactions of the LSFTO anode is better than those of the LSMTO and LSCTO anodes. A large low frequency arc observed in the LSMTO anode might be associated with poor catalytic activity for the methane oxidation reaction.

Discussions From the methane conversion rate and hydrogen selectivity results in Fig. 6, it was considered that methane is oxidized

Table 2 e Surface atomic concentrations calculated from XPS spectra for the La0.08Sr0.92M0.20Ti0.80O3Ld (M [ Mn, Fe, and Co) powder samples. Elements

La 3d Sr 3d Ti 2p (Ti4þ) O 1s (Latt.) Mn 2p (Mn4þ) Mn 2p (Mn3þ) Fe 2p (Fe4þ) Fe 2p (Fe3þ) Co 2p (Co4þ) Co 2p (Co3þ) Total

Binding energy (eV)

834.9 133.5 457.7e8 529.0 642.5 641.6 712.5 710.5 781.5 779.6 e

Atomic concentrations (%) La0.08Sr0.92Mn0.20Ti0.80O3d

La0.08Sr0.92Fe0.20Ti0.80O3d

La0.08Sr0.92Co0.20Ti0.80O3d

1.20 21.77 8.17 48.50 0.82 1.21 e e e e 99.99

0.93 20.92 7.47 40.84 e e 0.54 1.52 e e 99.99

0.92 21.06 7.81 39.31 e e e e 0.49 1.48 99.94

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

(a)

7961

As was described in Table 2, the atomic concentrations of Fe3þ in LSFTO or Co3þ in LSCTO were greater than that of Mn3þ in LSMTO. The good catalytic activity observed in the LSFTO and LSCTO samples is partially due to the high concentration of transition metal ions with a low oxidation state and the resulting high concentration of oxygen vacancies. On the other hand, the reason why the LSFTO sample exhibited better catalytic activity than the LSCTO sample is not clear. It may be associated with the binding energy difference between oxygen and the metal ions.

Conclusions

(b)

Fig. 7 e AC impedance spectra of the La0.08Sr0.92Mn0.20Ti0.80O3Ld (LSMTO), La0.08Sr0.92Fe0.20Ti0.80O3Ld (LSFTO), and La0.08Sr0.92Co0.20Ti0.80O3Ld (LSCTO)/ScSZ interface at 800  C; (a) wet hydrogen, (b) wet 50% methane/Ar mixture.

to carbon dioxide (CO2) and water vapor (H2O) on the perovskite powder sample via a complete oxidation reaction, i.e., CH4 þ 2O2 / 2H2O þ CO2 below 700  C. The lack of hydrogen production suggests the complete oxidation of methane at low temperatures. The plateau observed in the methane conversionerate curve might be associated with oxygen depletion, i.e., all the oxygen is consumed by the methane oxidation reaction since the methane-to-oxygen ratio was 0.5 in this study. On the other hand, partial oxidation of methane occurs at temperatures above 650  C. The methane is oxidized to carbon monoxide (CO) and hydrogen (H2) through the reaction CH4 þ (1/2) O2 / CO þ 2H2. The catalytic activity for the oxidation of methane depends on the oxidation state of the host or foreign ions, the oxygen vacancy concentration and its oxygen binding energy in the perovskite structure. As is evident in Figs. 3e5, the transition metal doping does not lead to a variation in the oxidation state of the titanium ion, while a significant amount of Mn3þ, Fe3þ, and Co3þ was confirmed in the LSMTO, LSFTO, and LSCTO powder samples, respectively. In general, Mn, Fe, and Co ions, which are substituted for Ti, serve as active sites for the methane oxidation reaction. Especially, it is known that the metal ion having a low oxidation state shows a better catalytic activity because the methane or oxygen molecules can be easily ionized by the electrons which come from the reduced metal ions.

Single-phase perovskite-type La0.08Sr0.92MxTi1xO3d (M ¼ Mn, Fe, or Co) oxides were synthesized by the Pechini method. It was found that complete methane oxidation was predominant below 650  C, while partial oxidation of methane was predominant at the temperature higher than 650  C. It was considered that the oxidation state of Mn, Fe, or Co in the perovskite structure plays an important role in the catalytic methane oxidation reactions. From the methane conversion rate and the polarization resistance of SOFC cell, Fe- or Co-doped perovskite samples showed better catalytic activity than the Mn-doped sample and this phenomenon may be associated with the high ratio of Fe3þ/Fe4þ or Co3þ/Co4þ compared to Mn3þ/Mn4þ. In addition, the enhanced catalytic activity was partially a result of the high surface oxygen vacancy concentration in the LSFTO and LSCTO samples.

Acknowledgments This research was supported by a grant from the Fundamental R&D Program for Core Technology of Materials funded by the Ministry of Knowledge Economy, Republic of Korea. A part of this work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (No.2010-0010744).

references

[1] Gorte RJ, Vohs JM. Novel SOFC anodes for the direct electrochemical oxidation of hydrocarbons. J Catal 2003;216:477e86. [2] Gorte RJ, Kim H, Vohs JM. Novel SOFC anodes for the direct electrochemical oxidation of hydrocarbon. J Power Sources 2002;106(1):10e5. [3] Singhal SC, Kendall K, editors. High temperature solid oxide fuel cells. Oxford, UK: Elsevier Ltd; 2003. p. 149. [4] Irvine JTS, Sauvet A. Improved oxidation of hydrocarbons with new electrodes in high temperature Fuel Cells. Fuel Cells 2001;1(3e4):205e10. [5] Hui S, Petric A. Evaluation of yttrium-doped SrTiO3 as an anode for solid oxide fuel cells. J Euro Ceram Soc 2002;22:1673e81. [6] Yoo KB, Choi GM. Performance of La-doped strontium titanate (LST) anode on LaGaO3-based SOFC. Solid State Ion 2009;180:867e71.

7962

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 7 9 5 5 e7 9 6 2

[7] Cimino S, Lisi L, De Rossi S, Faticanti M, Port P. Methane combustion and CO oxidation on LaAl1xMnxO3 perovskitetype oxide solid solutions. Appl Catal B Environ 2003;43:397e406. [8] Wei HJ, Cao Y, Ji WJ, Au CT. Lattice oxygen of La1xSrxMO3 (M ¼ Mn, Ni) and LaMnO3aFb perovskite oxides for the partial oxidation of methane to synthesis gas. Catal Commun 2008;9:2509e14. [9] Arai H, Yamada T, Eguchi K, Seiyama T. Catalytic combustion of methane over various perovskite-type oxides. Appl Catal 1986;26(1e2):265e76. [10] Saracco G, Geobaldo F, Baldi G. Methane combustion on Mgdoped LaMnO3 perovskite catalysts. Appl Catal B Environ 1999;20(4):277e88. [11] Nakamura T, Misono M, Yoneda Y. Reduction-oxidation and catalytic properties of perovskite-type mixed oxide catalysts (La1-xSrxCoO3). Chem Lett 1981;10(11):1589e92. [12] Ferri D, Forni L. Methane combustion on some perovskitelike mixed oxides. Appl Catal B Environ 1998;16(2):119e26. [13] Ghaffari M, Shannon M, Hui H, Tan OK, Irannejad A. Preparation, surface state and band structure studies of SrTi(1  x)Fe(x)O(3  d) (x ¼ 0e1) perovskite-type nano structure by X-ray and ultraviolet photoelectron spectroscopy. Surf Sci 2012;606:670e7. [14] Briggs D, Seah MP, editors. Practical Surface Analysis by Auger and X-ray PhotoelectronSpectroscopy. Chichester: John Willey&Sons; 1983. 1983, xivþ 533.

[15] Wu Q, Liu M, Jaegermann W. X-ray photoelectron spectroscopy of La0.5Sr0.5MnO3. Mater Lett 2005;59:1480e3. [16] Liu B, Zhang Y, Tang L. X-ray photoelectron spectroscopic studies of Ba0.5Sr0.5Co0.8Fe0.2O3d cathode for solid oxide fuel cells. Int J Hydrogen Energy 2009;34:435e9. [17] Young V, Otagawa T. XPS studies on strontium compounds. Appl Surf Sci 1985;20:228e48. [18] Jurado JR, Figueiredo FM, Gharbage B, Frade JR. Electrochemical permeability of Sr0.7(Ti,Fe)O3d materials. Solid State Ion 1999;118:89e95. [19] Yan A, Margou V, Arico A, Cheng M, Tsiakaras P. Investigation of a Ba0.5Sr0.5Co0.8Fe0.2O3d based cathode SOFC: II. The effect of CO2 on the chemical stability. Appl Catal B Environ 2007;76:320e7. [20] Mukai D, Tochiya S, Murai Y, Imori M, Hashimoto T, Sugiura Y, et al. Role of support lattice oxygen on steam reforming of toluene for hydrogen production over Ni/ La0.7Sr0.3AlO3d catalyst. Appl Catal A Gen 2013;453:60e70. [21] Oku M, Hirokawa K, Ikeda S. X-ray photoelectron spectroscopy of manganesedoxygen systems. J Electron Spectrosc Relat Phenom 1975;7:465e73. [22] Gao Z, Wang R. Catalytic activity for methane combustion of the perovskite-type La1xSrxCoO3d oxide prepared by the urea decomposition method. Appl Catal B Environ 2010;98:147e53.