as Environmental Catalyst: Purification - ACS Publications

Subscriber access provided by UCL Library Services

Energy, Environmental, and Catalysis Applications 1-x

x

3-#

Pd/SrFe TiO as Environmental Catalyst: Purification of Automotive Exhaust Gases Kosuke Beppu, Akito Demizu, Saburo Hosokawa, Hiroyuki Asakura, Kentaro Teramura, and Tsunehiro Tanaka ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b05169 • Publication Date (Web): 12 Jun 2018 Downloaded from http://pubs.acs.org on June 12, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Pd/SrFe1-xTixO3-δ as Environmental Catalyst: Purification of Automotive Exhaust Gases Kosuke Beppu,† Akito Demizu,† Saburo Hosokawa,*,†,‡ Hiroyuki Asakura,†,‡ Kentaro Teramura,†,‡ Tsunehiro Tanaka*,†,‡ †

Department of Molecular Engineering, Graduate School of Engineering, Kyoto University, Kyotodaigaku Katsura, Nishikyo-ku, Kyoto 615-8510, Japan ‡ Elements Strategy Initiative for Catalysts & Batteries (ESICB), Kyoto University, Kyotodaigaku Katsura, Nishikyo-ku, Kyoto 615-8245, Japan ABSTRACT: Environmental catalysts are required to operate highly efficiently under severe conditions in which they are exposed to reductive and oxidative atmospheres at high temperatures. This study demonstrates that SrFe1-xTixO3-δ-supported Pd catalysts exhibit high catalytic activities for NO reduction with C3H6 and CO in the presence of O2, which is a model reaction for the purification of automotive exhaust gases. Catalytic activity is enhanced with increasing Ti content, and the highest activity is observed for Pd/SrFe0.2Ti0.8O3-δ among the examined catalysts. The state of the loaded Pd species can be controlled by the Fe/(Fe+Ti) ratio in SrFe1-xTixO3-δ, and highly active PdO nanoparticles are properly anchored on the SrFe0.2Ti0.8O3-δ. The Ti-rich Pd/SrFe0.2Ti0.8O3-δ shows significantly higher catalytic activity and is more thermally stable than the conventional Pd/Al2O3, which has a high surface area. Since Fe-rich SrFe1-xTixO3-δ has high the oxygen storage capacity, its response capabilities to atmospheric fluctuations were evaluated by changing the oxygen concentration during NO reduction. As a result, Fe-rich Pd/SrFe0.8Ti0.2O3-δ retains its high NO-reduction activity for longer times even under oxidative conditions, when compared to SrFeO3-δ or Ti-rich Pd/SrFe1-xTixO3-δ. The oxygen storage properties of Pd/SrFe0.8Ti0.2O3-δ allow it to effectively act as an oxygen buffer for NO reduction. These properties ensure that the SrFe 1-xTixO3-δ support, with both high thermal stability and oxygen storage capacity, is a very useful environmental-catalyst material. KEYWORDS: NO reduction, thermal stability, anchoring effect, oxygen storage capacity, SrFe1-xTixO3-δ

1. Introduction Unlike catalysts used to promote general reactions regarding energy and resource conversion, environmental catalysts are often used under severely restrictive conditions.1-3 For example, for catalysis leading to high valueadded organic compounds, the reaction temperature and/or atmospheric conditions, under which the catalyst performs most effectively, are controllable.4,5 On the other hand, environmental catalysts are required to operate under extremely harsh conditions; that is, automotive threeway catalysts (TWCs) are exposed to atmospheres containing a large amounts of CO2 and H2O at high temperatures above 1073 K.1,6-9 Under such severe conditions, the TWC must completely convert specific pollutants, such as hydrocarbons, CO, and NOx, into N2, CO2, and H2O. Therefore, the development of environmental catalysts that are thermally and chemically stable is essential for the purification of automotive exhaust gases. The catalyst support plays a very important role in improving the thermal stability of the catalyst material. 10-21 Usually, TWCs consist of platinum-group-metal (PGM) species, such as Pd, Rh, and Pt species, loaded on Al2O3 and/or CeO2-ZrO2 supports. If the PGM species are extremely aggregated by the high temperature, the catalytic activity dramatically decreases due to decreases in the numbers of effective active site. Catalyst supports effectively act to prevent the

aggregation of PGM species under the reaction conditions. Among various properties of the catalyst support, the anchoring effect is well known to overcome this unfavorable aggregation.22-27 This effect immobilizes PGM nanoparticles and inhibits aggregation through the formation of bonds between the PGM species and the support. CeO 2-based supports have been reported to exhibit suitable anchoring properties, in which PGM species are strongly linked to the support. Nagai et al. have reported that, for Pt catalysts, the electron densities of the oxygens in the support affect the sintering behavior of the Pt particles, and that Pt-O-Ce bonds, which are formed in CeO2-based supports having a relatively high electron densities, inhibit Pt migration.22,23 The AlPO4 support has also been reported to show suitable anchoring properties; hence, Rh/AlPO4 is an effective TWC.24-26 Subtle control over the interactions between the catalyst support and the loaded metal species has generally been considered to be difficult for catalytic reactions at high temperatures above 1073 K. Hence, catalyst design that takes the anchoring effect into account is evidently desired for the development of environmental catalysts. Considering that the oxygen concentrations of automotive exhaust gases vary depending on the driving mode, catalyst supports capable of storing/releasing oxygen are preferable, because oxygen storage supports can adjust the oxygen levels on the surface of TWCs to provide optimal

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

concentrations.28-34 In other words, the support adsorbs oxygen under excess oxygen conditions, and releases oxygen in an oxygen-deficient atmosphere. As a result, such TWCs efficiently purify automotive exhaust gases under a wide range of oxygen concentrations. We reported that Sr3Fe2O7δ, with a perovskite-derived structure, exhibits comparable oxygen storage capacity to CeO2-based materials, and that PGM catalysts on Sr3Fe2O7-δ supports exhibit activities for NO reduction under a wide range of oxygen concentrations.35-38 Recently, the partial substitution of Fe-sites with Ti species in SrFeO3-δ, with a high oxygen storage ability, has been found to enhance the oxygen-release rate of SrFeO3-δ and its chemical stability against CO2.38 The observed improvement in CO2 resistance suggests that Ti doping into the perovskite support may control the electron densities of the surface-oxygen species, which correlates with the anchoring effect. In this context, this paper demonstrates for the first time that the state of the Pd species is delicately tunable by controlling the ratio of Fe/(Fe+Ti) in SrFe1-xTixO3-δ even under severe conditions, and that SrFe1-xTixO3-δ-supported Pd catalysts exhibit both high catalytic activities and thermal stabilities for the reduction of NO, which is a model reaction for the purification of automotive exhaust gases. Furthermore, the response capabilities of Pd/SrFe1-xTixO3-δ to atmospheric fluctuation have been evaluated by switching between reductive and oxidative conditions; consequently, we successfully clarified the oxygen storage functions of SrFe1-xTixO3-δ, which can act as environmental catalysts under practical conditions. 2. Experimental Section Preparation of Pd catalysts: All reagents were purchased from Wako Pure Chemical Industries Ltd., Osaka, Japan. SrFe1-xTixO3-δ was synthesized by a polymerized-complex method, as described in our previous report.38 Citric acid (400 mmol) was dissolved in deionized water (180 mL) at room temperature. Strontium carbonate (10 mmol), iron nitrate nonahydrate (10(1-x) mmol), and titanium (IV) tetraisopropoxide (10x mmol) were then added to the solution, which was stirred at 353 K for 2 h; x is the fraction of titanium in SrFe1-xTixO3-δ (x = 0 – 1). Ethylene glycol (400 mmol) was added to this solution, and the mixture was stirred at 403 K for 4 h to form a gelatinous solution, after which the gel was heated in a mantle heater at 623 K for 3 h. The powder obtained in this manner was calcined at 1273 K for 2 h. Pd catalysts were synthesized by an impregnation method. Pd acetate was dissolved in acetone, after which the support (SrFe1-xTixO3-δ or γ-Al2O3 (JRC-ALO-7)) was added. The solution was evaporated at 353 K. The powder obtained following drying, which contained the Pd precursor and the support, was calcined at various temperatures for 5 h. The amount of loaded Pd species was 1.0 wt% based on the metal. NO reduction with C3H6 and CO in the presence of O2: Catalytic reactions were carried out using a fixed-bed flow reactor at atmospheric pressure, as described in the previous report.37 A 200 mg sample of the catalyst (25/50 mesh) was placed in a tubular reactor. The total gas flow rate was 100 mL min1. The catalyst was pretreated by heating to 773 K under a flow of He (30 mL min-1) for 1 h. The gas composition was

analyzed by gas chromatography (Shimadzu GC8A) with MS-5A and Porapak-Q columns. The temperature-programmed reaction was carried out through the stepwise increase in temperature (50 K intervals) from 473 K under stoichiometric conditions: NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; and O2, 1125 ppm. The long-term reaction at 773 K was performed under stoichiometric conditions. The catalyst was heated to 773 K under He and maintained at that temperature for 1 h, after which the flow gas was changed to the reaction gas, and the outlet gases were monitored every 30 min. The λ value is defined as follows: the number of oxygen atoms in the reaction system ([NO] + [CO] + [O2] × 2) was divided by the number of oxygen atoms under stoichiometric conditions ([NO] + [CO] + [O2] × 2 = 4250 ppm). Lean-rich cycle testing for NO reduction was performed by alternating the atmospheric conditions between lean (λ = 1.15; NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1444 ppm) and rich (λ = 0.85; NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 806 ppm). The catalyst was heated to 773 K under lean conditions, after which the reaction conditions were alternated between lean and rich every 1 h. The outlet gases were detected using a micro gas chromatograph (Agilent 490 Micro GC) fitted with MS-5A and PoraPLOT Q columns. Characterization: X-ray powder diffraction patterns (Rigaku UltimaIV) were recorded using CuKα radiation. X-ray absorption fine structure (XAFS) spectra at the Pd K-edge were measured on the BL01B1 beam line of the SPring-8 facility. The XAFS spectra were recorded in fluorescence mode at room temperature using a Si(311) double-crystal monochromator. In-situ XAFS measurements were carried out under 5% H2/He at 50 mL min-1. The samples (200 mg) were placed in an in-situ cell and heated under a flow of hydrogen. The in-situ XAFS spectra were recorded at various temperatures. SEM images were obtained by fieldemission scanning electron microscopy (FE-SEM, SU-8220, Hitachi High-Technologies) augmented with an energydispersive X-ray spectroscoy (EDS). 3. Results and Discussion 3.1. NO reduction over Pd/SrFe1-xTixO3-δ Figure 1 shows the results of NO reduction under the following stoichiometric condition (λ = 1.00): 4CO + C3H6 + 4NO + 9/2O2 → 2N2 + 7CO2 + 3H2O The Fe/(Fe+Ti) ratio clearly affects the NO reduction and the CO + C3H6 oxidation activities. Oxidation activity was observed from 473 K over Pd/SrTiO3 and Pd/SrFe0.2Ti0.8O3δ, and from 523 K over Pd/SrFeO 3-δ and Pd/SrFe0.8Ti0.2O3-δ. Reduction activity was observed above 523 K over Pd/SrTiO3 and Pd/SrFe0.2Ti0.8O3-δ, and above 573 K over Pd/SrFeO3-δ and Pd/SrFe0.8Ti0.2O3-δ. The reproducibility of the data over Pd/SrFe0.2Ti0.8O3-δ was confirmed, and the accuracy was sufficiently reliable (Figure S1). These results suggest that the Ti-containing catalysts are more active for NO reduction than Pd/SrFeO3-δ. To evaluate the effect of the proportion of Ti in the catalyst on NO reduction, we compared the T50 values for all Pd/SrFe1-xTixO3-δ catalysts; T50 is defined as the temperature at which the NO or CO + C3H6 conversion reaches 50%. Figure S2 displays the T50


Page 2 of 9

Page 3 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Fourier transformed EXAFS spectra of Pd catalysts supported on SrFe1-xTixO3-δ prepared at 1073 K. (A) Ti-rich samples; (B) Fe-rich samples.

Figure 1. (A) NO conversions and (B) CO + C3H6 conversions over Pd catalysts supported on SrFe1-xTixO3-δ prepared at 1073 K. The reaction gas was composed of 1000 ppm NO, 1000 ppm CO, 250 ppm C3H6, 1125 ppm O2, with He as the balance. *The concentration (1750ppm) of CO2 produced by complete oxidation of the introduced CO (1000ppm) and C3H6 (250ppm) is defined as 100%.

values for the NO reduction over Pd/SrFe1-xTixO3-δ; the oxidation and reduction activities were observed to improve with increasing amount of the Ti dopant. The optimal amount of Ti was determined to be 80 mol% (Fe/(Fe+Ti) = 0.8). The X-ray diffraction (XRD) pattern of 1 wt% Pd/SrFe1-xTixO3-δ was essentially identical to that of SrFe1xTixO3-δ, and peaks corresponding Pd species were not observed (Figure S3), indicating that the Pd species are highly dispersed on SrFe1-xTixO3-δ. Pd K-edge XAFS spectra of Pd/SrFe1-xTixO3-δ were acquired. Figures 2 and S4 show the results of Pd K-edge extended X-ray absorption fine structure (EXAFS) analyses. The Fourier transformed EXAFS spectra of Ti-rich Pd/SrFe1-xTixO3-δ were also analogous to that of PdO. However, the EXAFS oscillations and Fourier transformed EXAFS spectra of Fe-rich Pd/SrFe1-xTixO3-δ were different to those of PdO. Curve-fitting of the EXAFS results for Pd/SrFeO3-δ (Table S1) confirmed the presence of Pd–O–Sr and Pd–O–Fe bonds, in addition to Pd-O-Pd bonds, indicating that a PdOx cluster had formed through Pd–O–Sr and Pd–O–Fe bonding on the surface of the SrFeO3δ. To examine the difference in the reducibilities of the Pd species on SrFeO3-δ and SrTiO3, Pd K-edge in-situ XAFS experiments were carried out under a flow of hydrogen. Figure 3 shows the in-situ X-ray absorption near edge

Figure 3. Pd K-edge in-situ XANES spectra of (A) Pd/SrTiO3 and (B) Pd/SrFeO3-δ under hydrogen flow. (C) Pd K-edge XANES spectra of PdO (broken line) and Pd metal (solid line). (D) shows the results of XANES linear combination fitting (LCF), which was performed by using as-synthesized Pd catalyst and Pd foil. The fraction for as-synthesized catalyst was assigned as 1, and that for Pd foil as 0.

structure (XANES) spectra. The observed XANES spectra were analyzed by XANES linear combination fitting (LCF), which was performed using the as-synthesized Pd catalyst and Pd foil (the fraction of as-synthesized catalyst was assigned to be 1). The Pd species on SrTiO3 were completely reduced at 348 K. On the other hand, the Pd species on SrFeO3-δ began to be reduced at 373 K, and the reduction was almost complete at 423 K. These results indicate that the Pd species on the Ti-rich SrFe1-xTixO3-δ are reduced at lower temperatures than those on the Fe-rich analogues. Therefore, the Pd species on the Fe-rich SrFe1-xTixO3-δ are strongly anchored to the support and are difficult to reduce. On the other hand, the anchoring effect is moderately weakened with decreasing Fe/(Fe+Ti) ratio, eventually resulting in the formation of highly dispersed PdO nanoparticles that are easily reduced under a flow of hydrogen.



Figure 5. XRD patterns and SEM images of Pd catalysts supported on (A and C) SrFe0.2Ti0.8O3-δ and (B and D) Al2O3 calcined at 1273 K. The arrowed lines show Pd species.

Figure 4. (A) NO conversions and (B) CO + C3H6 conversions over Pd catalysts supported on SrFe0.2Ti0.8O3-δ and Al2O3 calcined at 1273 K. *The concentration (1750ppm) of CO2 produced by complete oxidation of the introduced CO (1000ppm) and C3H6 (250ppm) is defined as 100%.

These results reveal that the Fe/(Fe+Ti) ratio in SrFe1-xTixO3-δ can systematically control the anchoring effect; that is, PdOx clusters are strongly linked to the Fe-rich SrFe1-xTixO3-δ through Pd–O–Sr or Pd–O–Fe bonds, and PdO nanoparticles are moderately anchored to Ti-rich SrFe1xTixO3-δ without the requirement for bonds involving Sr or Fe ions. Taking into account the observation that the catalytic activity of Pd/SrFe0.2Ti0.8O3-δ was the highest among the examined catalysts, the PdO nanoparticles on this material contribute to its catalytic activity. Such control of the anchoring effect demonstrates for the first time that PdO nanoparticles moderately anchored on a support exhibit high catalytic activities, rather than PdOx species strongly-anchored onto the support. Pd/Al2O3, with a high surface area, is a well-known representative catalyst for the purification of automotive exhaust gas under stoichiometric conditions.9, 39 Therefore, the catalytic activities of Pd/SrFe0.2Ti0.8O3-δ calcined at various temperatures were compared with those of Pd/Al2O3 (Figures 4 and S5). Thermal stability was also evaluated from the T50 values for NO reduction over Pd/SrFe0.2Ti0.8O3-δ and Pd/Al2O3 calcined at various temperatures (Figure S6). A threshold in the catalytic performance of Pd/Al2O3 was observed between 1073 and 1173 K. The catalytic activity of Pd/Al2O3 dramatically deteriorated when calcined at temperatures above the threshold. On the other hand, the deterioration was suppressed in the case of Pd/SrFe0.2Ti0.8O3-δ. The catalytic activity of Pd/SrFe0.2Ti0.8O3δ prepared at 1273 K, which exhibited high thermal stability,

was stable for at least 6 h in the long-term reaction at 773 K (Figure S7). To investigate the factors responsible for the observed differences in the thermal stabilities of Pd/Al2O3 and Pd/SrFe0.2Ti0.8O3-δ, we examined the structures of these catalysts by XRD patterns (Figure S8). Peaks corresponding to perovskite-type structures were only observed in the XRD patterns of Pd/SrFe0.2Ti0.8O3-δ at various calcination temperatures, and peaks corresponding to Pd species were not detected even in the sample calcined at 1373 K. However, the peaks in the pattern of Pd/Al2O3 changed with increasing calcination temperature; peaks corresponding to PdO were observed at 1073 K, and intensified and sharpened with increasing calcination temperature. The XRD peak due to Pd metal was expected to be detected for the Pd/Al2O3 catalyst calcined above 1073 K because of the equilibrium involving PdO dissociation (at ~1073 K). However, since PdO on Al2O3 has been reported to decompose to Pd metal at temperatures above 973 K during heating in air, and to reoxidize to PdO at around 923 K during cooling,40, 41 the Pd metal on Al2O3 formed by calcination at high temperature in this study must reoxidize to PdO during the cooling process. Furthermore, an Al2O3 phase transition was observed; γAl2O3 apparently transformed into - and δ-Al2O3 up to 1173 K. The sintering of PdO particles is considered to be largely promoted by the reconstruction of the surface structure on Al2O3 accompanied with the phase transition. These results indicate that PdO-particle sintering on Al2O3 proceeds more easily than that on SrFe0.2Ti0.8O3-δ, resulting in the dramatic deterioration of the activity of the Pd/Al2O3. SrFe0.2Ti0.8O3-δ support, which does not cause phase transition, is desirable for a highly stabile catalyst support. When the samples calcined at 1273 K, which showed significant difference in catalytic activity (Figure 4), were investigated by SEM, EDS images and XAFS spectra (Figures 5, S9, S10 and S11), bulk PdO species with the particle size around 100 nm were clearly observed only in Pd/Al2O3. These results are in good agreement with those obtained by XRD patterns. Although the Brunauer–Emmett–Teller (BET) surface area of Pd/SrFe0.2Ti0.8O3-δ calcined at 1273 K (9 m2/g) was much lower than that of


Page 4 of 9

Page 5 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces Pd/Al2O3 (99 m2/g), the sintering of the PdO species was apparently suppressed using the SrFe0.2Ti0.8O3-δ support. The anchoring effect due to interactions between the Pd and the catalyst support is more important for the suppression of PdO-particle sintering than the effect of surface area. Therefore, the high thermal stability of Pd/SrFe 0.2Ti0.8O3-δ indicates that the highly dispersed PdO nanoparticles, which exhibit high catalytic activities, are appropriately anchored to the support even at high temperatures. These results suggest that strong anchoring, such as in Fe-rich Pd/SrFe1xTixO3-δ, is disadvantageous for NO reduction, and that weak anchoring, such as in Pd/Al2O3, is also disadvantageous due to the aggregation and sintering of the Pd species. From this viewpoint, SrFe0.2Ti0.8O3-δ, with both high activity and thermal stability, exhibits the most suitable anchoring properties. 3.2. NO reduction during lean-rich cycle testing over Pd/SrFe1-xTixO3-δ Since SrFe1-xTixO3-δ supports are reported to have unique oxygen storage capacities,38 we expect that Pd/SrFe1-xTixO3-δ is able to adjust the oxygen concentration on the surface of the catalyst during NO reduction at various oxygen concentrations. Hence, we examined the effect of modulating the oxygen concentration during NO reduction over Pd/SrFe1-xTixO3-δ. Figures 6 and S12 show the results for four oxidative (lean condition: λ = 1.15) and three reductive (rich condition: λ = 0.85) cycles during the NO reduction over Pd/SrFeO3-δ and Pd/SrFe0.8Ti0.2O3-δ. Under the first set of lean conditions, each catalyst maintained a NO reduction activity of 25%, and the CO + C3H6 oxidation activities reached 100%. When the reaction conditions were changed from lean to rich, the reduction activity of each catalyst

Figure 6 Time courses of N2 production for the NO reduction during lean-rich cycle test over Pd/SrFe0.8Ti0.2O3δ and Pd/SrFeO3-δ prepared at 1073 K. The lean-rich cycle test was carried out by alternating the atmospheric conditions between lean (λ = 1.15; NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 1444 ppm) and rich (λ = 0.85; NO, 1000 ppm; CO, 1000 ppm; C3H6, 250 ppm; O2, 806 ppm). The λ value was defined as follows: the number of oxygen atoms in the reaction system ([NO] + [CO] + [O2] × 2) was divided by the number of oxygen atoms under the stoichiometric condition ([NO] + [CO] + [O2] × 2 = 4250 ppm).

Figure 7. Time courses for the NO reduction during lean-rich cycle testing over Pd/SrFe1-xTixO3-δ prepared at 1073 K during the third-rich and fourth-lean cycles.

moderately decreased in the first cycle, with a concomitant gradual decrease in oxidation activity. Interestingly, after the second cycle, Pd/SrFeO3-δ and Pd/SrFe0.8Ti0.2O3-δ exhibited different catalytic behavior. The reduction activity over Pd/SrFeO3-δ was observed to immediately decrease during the second lean cycle, and oxidation activity over Pd/SrFeO3-δ was maintained at 100%. On the other hand, the deterioration in the reduction activity of Pd/SrFe0.8Ti0.2O3-δ was more moderate during the second set of lean conditions than that of Pd/SrFeO3-δ, although the oxidation behavior of Pd/SrFe0.8Ti0.2O3-δ was almost the same as that observed for Pd/SrFeO3-δ. During the second rich cycle, the reduction and oxidation activities of Pd/SrFe0.8Ti0.2O3-δ also moderately decreased, compared to those of Pd/SrFeO3-δ. The oxidation and reduction activities during subsequent lean-rich cycles were repeats of those observed during the second lean-rich cycle for each catalyst. Figure 7 displays the results from the fourth set of lean conditions during the lean-rich cycle testing for NO reduction over Pd/SrFe1-xTixO3-δ with various Ti contents. The reduction behavior of Pd/SrFe0.8Ti0.2O3-δ deteriorated the slowest, and increasing rates of deterioration were observed with increasing Ti content. The oxygen storage capacity of SrFe1-xTixO3-δ had been reported to decrease with decreasing Fe/(Fe+Ti) ratio using the H2-O2 titration method at 773 K.38 If the oxygen storage capacity is related to the deterioration in NO-reduction activity, then, among these catalysts, Pd/SrFeO3-δ should exhibit higher catalytic activity for longer times than Pd/SrFe0.8Ti0.2O3-δ. However, the activity of Pd/SrFeO3-δ deteriorated immediately, indicating that oxygen storage capacities estimated using the H2-O2 titration method do not directly correlate with NOreduction behavior under lean-rich cycling. Pd/SrFeO3-δ and Pd/SrFe0.8Ti0.2O3-δ, before and after reaction, were then examined by XRD (Figure 8), where “after reaction” refers to the catalyst after the fourth lean cycle. The XRD peaks corresponding to Pd/SrFe0.8Ti0.2O3-δ, after reaction, were assigned to the perovskite type structure with the I4/mmm space group, as were those of the as-synthesized Pd/SrFe0.8Ti0.2O3-δ. The



Figure 8. XRD patterns of (a, b) Pd/SrFe0.8Ti0.2O3-δ and (c, d) Pd/SrFeO3-δ prepared at 1073 K. Red: as-synthesized; blue: after reaction (lean-rich cycle test).

XRD peaks of Pd/SrFe0.8Ti0.2O3-δ after reaction were slightly shifted to lower angles than those of the material before reaction. The peak shifts are ascribable to topotactic lattice oxygen release in SrFe0.8Ti0.2O3-δ; 37, 38 therefore, a fraction of the oxygen-vacancy sites was recovered by excess oxygen under lean conditions after most of the lattice oxygen had been released under rich conditions. Such behavior was also observed for the other Pd/SrFe1-xTixO3-δ catalysts (Figure S13). On the other hand, although Pd/SrFeO3-δ, after reaction, was exposed to excess oxygen (lean conditions), its XRD peaks did not correspond to a perovskite type structure. Rather, they corresponded to SrFeO2.5 with a brownmillerite-type structure with the Ima2 space group. Ti4+ ions are generally more stable under reductive condition than Fe4+ ions. In fact, SrFe1-xTixO3-δ (Fe/(Fe+Ti) ≥ 0.2) samples have been reported to retain their original perovskite structures even after reduction treatment under a flow of H2 at 773 K.38 The phase transition from the perovskite-type SrFeO3-δ to the brownmillerite-type SrFeO2.5 must take place during the first set of rich conditions. The brownmillerite structure is stable and did not return to the perovskite structure even under lean conditions. As a result, Pd/SrFeO3-δ exhibited a dramatic deterioration in activity after the second lean-rich cycle despite the high amount of oxygen storage. On the other hand, Ti-containing catalysts can reversibly release/store lattice oxygen while retaining their perovskite-type structures during lean-rich cycling. Therefore, the SrFe0.8Ti0.2O3-δ support, which has the highest oxygen storage capacity among the Ti-containing samples, act as an oxygen buffer for excess O2 during NO reduction, resulting in the maintenance of high catalytic activity for long times even under lean conditions. 4. Conclusion Ti-rich Pd/SrFe0.2Ti0.8O3-δ shows the highest activity for NO reduction at a stoichiometric oxygen concentration (λ = 1) among the examined catalysts. The Pd species on SrFe0.2Ti0.8O3-δ exhibit PdO structures, and these species contribute to NO reduction. Furthermore, Pd/SrFe0.2Ti0.8O3δ is more thermally stable than Pd/Al 2O3. The anchoring effect, which involves interactions between the Pd species

and SrFe0.2Ti0.8O3-δ, stabilizes the highly dispersed PdO species on SrFe0.2Ti0.8O3-δ and prevents their sintering. When the oxygen concentration during NO reduction varies from oxidative conditions (lean conditions, λ > 1) to reductive conditions (rich conditions, λ < 1), the Fe-rich Pd/SrFe0.8Ti0.2O3-δ maintains its catalytic activity for the longest time under oxidative conditions among the examined catalysts. The Ti-containing catalysts preserved their perovskite type structures during this reaction, whereas Pd/SrFeO3-δ is transformed from the perovskite to the brownmillerite structure under reductive conditions and does not return to its perovskite structure under lean conditions. The reversible oxygen intake/release of the Ti-containing catalyst with such a robust perovskite structure contributes to the retention of high catalytic activity during lean-rich cycling. Therefore, Ti-rich SrFe1-xTixO3-δ supports with suitable anchoring properties form highly dispersed PdO nanoparticles that have both excellent catalytic activities and high thermal stabilities, despite their limited oxygen storage capacities. On the other hand, Pd catalysts supported on Fe-rich SrFe1-xTixO3-δ supports, with high oxygen storage capacities, show high resistance to atmospheric fluctuations between reductive and oxidative conditions. These observations indicate that the anchoring functions and oxygen storage properties of Pd/SrFe1-xTixO3-δ, which is highly thermally stable, can be controlled by adjusting the Fe/(Fe+Ti) ratio. We believe that SrFe1-xTixO3-δ is an extremely promising support for environmental catalysts. ASSOCIATED CONTENT Supporting Information. The Supporting Information is available free of charge on the ACS Publications website at DOI: NO conversion and CO + C3H6 conversion with error bars over Pd/SrFe0.2Ti0.8O3-δ, T50 as functions of Ti proportion (Fe/(Fe+Ti)), XRD patterns of Pd/SrFe1-xTixO3-δ, Pd K-edge XAFS spectra of Pd/SrFe1-xTixO3-δ, Results of Pd K-edge EXAFS curve fitting of Pd/SrFeO3-δ and Pd/SrTiO3, NO conversion and CO and C3H6 conversions and T50 values over Pd/SrFe0.2Ti0.8O3δ and Pd/Al2O3, XRD patterns and EDS mappings of Pd/SrFe0.2Ti0.8O3-δ and Pd/Al2O3, XAFS spectra of Pd/Al2O3, CO2 productions over Pd/SrFe0.8Ti0.2O3-δ and Pd/SrFeO3-δ, and XRD patterns of Pd/SrFe0.8Ti0.2O3-δ and Pd/SrFe0.2Ti0.8O3-δ before and after reaction.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] (S.H) [email protected] (T.T)

ORCHID Kosuke Beppu: 0000-0002-1913-2033 Saburo Hosokawa: 0000-0003-1251-3543 Hiroyuki Asakura: 0000-0001-6451-4738 Kentaro Teramura: 0000-0003-2916-4597 Tsunehiro Tanaka: 0000-0002-1371-5836

Notes Any additional relevant notes should be placed here.


Page 6 of 9

Page 7 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ACKNOWLEDGMENT This study was supported by the Program for Element Strategy Initiative for Catalysts & Batteries (ESICB). XAFS studies were performed with the approval of SPring-8 (Proposal No. 2017A1419). We would like to express our thanks to Mr. Kenya Onishi of Kyoto University for experimental support about long time reaction.

REFERENCES (1) Kašpar, J.; Fornasiero, P.; Hickey, N., Automotive Catalytic Converters: Current Status and Some Perspectives. Catal. Today 2003, 77 (4), 419-449. (2) Song, C., An Overview of New Approaches to Deep Desulfurization for Ultra-Clean Gasoline, Diesel Fuel and Jet Fuel. Catal. Today 2003, 86 (1-4), 211-263. (3) Imanaka, N.; Masui, T., Advances in Direct NOx Decomposition Catalysts. Appl. Catal. A: General 2012, 431-432, 1-8. (4) Tamura, M.; Tomishige, K., Redox Properties of CeO2 at Low Temperature: the Direct Synthesis of Imines from Alcohol and Amine. Angew. Chem. Int. Ed. Engl. 2015, 54 (3), 864-867. (5) Tateyama, K.; Wada, K.; Miura, H.; Hosokawa, S.; Abe, R.; Inoue, M., Dehydrogenative Synthesis of Benzimidazoles under Mild Conditions with Supported Iridium Catalysts. Catal. Sci. Technol. 2016, 6 (6), 1677-1684. (6) Jones, J. H.; Kummer, J. T.; Otto, K.; Shelef, M.; Weaver, E. E., Selective Catalytic Reaction of Hydrogen with Nitric Oxide in the Presence of Oxygen. Environ. Sci. Technol. 1971, 5, 790-798. (7) Kummer, J. T., Catalysts for Automobile Emission Control. Prog. Energy Combust. Sci. 1980, 6, 177-199. (8) Shelef, M.; McCabe, R. W., Twenty-Five Years after Introduction of Automotive Catalysts: What Next? Catal. Today 2000, 62, 35-50. (9) Gandhi, H. S.; Graham, G. W.; McCabe, R. W., Automotive Exhaust Catalysis. J. Catal. 2003, 216 (1-2), 433-442. (10) Ozawa, M.; Kimura, M.; Isogai, A., The Application of CeZr Oxide Solid Solution to Oxygen Storage Promoters in Automotive Catalysts J. Alloys Compd. 1993, 193, 73-75. (11) Kenevey, K.; Valdivieso, F.; Soustelle, M.; Pijolat, M., Thermal Stability of Pd or Pt Loaded Ce0.68Zr0.32O2 and Ce0.50Zr0.50O2 Catalyst Materials under Oxidising Conditions. Appl. Catal. B: Environ. 2001, 29, 93-101. (12) Wu, X.; Yang, B.; Weng, D., Effect of Ce–Zr Mixed Oxides on the Thermal Stability of Transition Aluminas at Elevated Temperature. J. Alloys Compd. 2004, 376 (1-2), 241-245. (13) Morikawa, A.; Suzuki, T.; Kanazawa, T.; Kikuta, K.; Suda, A.; Shinjo, H., A New Concept in High Performance Ceria–Zirconia Oxygen Storage Capacity Material with Al2O3 as a Diffusion Barrier. Appl. Catal. B: Environ. 2008, 78 (3-4), 210-221. (14) Morikawa, A.; Kikuta, K.; Suda, A.; Shinjo, H., Enhancement of Oxygen Storage Capacity by Reductive Treatment of Al 2O3 and CeO2–ZrO2 Solid Solution Nanocomposite. Appl. Catal. B: Environ. 2009, 88 (3-4), 542-549. (15) Papavasiliou, A.; Tsetsekou, A.; Matsouka, V.; Konsolakis, M.; Yentekakis, I. V.; Boukos, N., Development of a Ce–Zr–La Modified Pt/γ-Al2O3 TWCs’ Washcoat: Effect of Synthesis Procedure on Catalytic Behaviour and Thermal Durability. Appl. Catal. B: Environ. 2009, 90 (1-2), 162-174. (16) Morikawa, A.; Yamamura, K.; Suda, A.; Tanabe, T.; Nobukawa, T.; Chiba, A.; Takahashi, N.; Goyal, A., Phase Transformation Behavior of a Pyrochlore-Type CeO2-ZrO2 Binary Compound. J. Am. Ceram. Soc. 2015, 98 (2), 659-662. (17) Nagao, Y.; Nakahara, Y.; Sato, T.; Nakano, S.; Machida, M., TWC Performance of Honeycomb Catalysts Coated with Pd-Supported 10Al2O3· 2B2O3 and Its Cation-Substituted Compounds. Emiss. Control Sci. Technol. 2016, 2 (2), 57-65. (18) Alikin, E. A.; Vedyagin, A. A., High Temperature Interaction of Rhodium with Oxygen Storage Component in Three-Way Catalysts. Top. Catal. 2016, 59 (10-12), 1033-1038.

(19) Vedyagin, A. A.; Volodin, A. M.; Kenzhin, R. M.; Stoyanovskii, V. O.; Shubin, Y. V.; Plyusnin, P. E.; Mishakov, I. V., Effect of Metal-Metal and Metal-Support Interaction on Activity and Stability of Pd-Rh/alumina in CO Oxidation. Catal. Today 2017, 293-294, 73-81. (20) Vedyagin, A. A.; Volodin, A. M.; Kenzhin, R. M.; Stoyanovskii, V. O.; Rogov, V. A.; Kriventsov, V. V.; Mishakov, I. V., The Role of Chemisorbed Water in Formation and Stabilization of Active Sites on Pd/Alumina Oxidation Catalysts. Catal. Today 2018, 307, 102-110. (21) Vedyagin, A. A.; Stoyanovskii, V. O.; Plyusnin, P. E.; Shubin, Y. V.; Slavinskaya, E. M.; Mishakov, I. V., Effect of Metal Ratio in Alumina-Supported Pd-Rh Nanoalloys on its Performance in Three Way Catalysis. J. Alloys Compd. 2018, 749, 155-162. (22) Nagai, Y.; Hirabayashi, T.; Dohmae, K.; Takagi, N.; Minami, T.; Shinjoh, H.; Matsumoto, S., Sintering Inhibition Mechanism of Platinum Supported on Ceria-Based Oxide and Pt-Oxide–Support Interaction. J. Catal. 2006, 242 (1), 103-109. (23) Nagai, Y.; Dohmae, K.; Ikeda, Y.; Takagi, N.; Tanabe, T.; Hara, N.; Guilera, G.; Pascarelli, S.; Newton, M. A.; Kuno, O.; Jiang, H.; Shinjoh, H.; Matsumoto, S., In situ Redispersion of Platinum Autoexhaust Catalysts: an On-Line Approach to Increasing Catalyst Lifetimes? Angew. Chem. Int. Ed. Engl. 2008, 47 (48), 9303-9306. (24) Machida, M.; Minami, S.; Ikeue, K.; Hinokuma, S.; Nagao, Y.; Sato, T.; Nakahara, Y., Rhodium Nanoparticle Anchoring on AlPO4 for Efficient Catalyst Sintering Suppression. Chem. Mater. 2014, 26 (19), 5799-5805. (25) Machida, M.; Minami, S.; Hinokuma, S.; Yoshida, H.; Nagao, Y.; Sato, T.; Nakahara, Y., Unusual Redox Behavior of Rh/AlPO4 and Its Impact on Three-Way Catalysis. J. Phys. Chem. C 2014, 119 (1), 373-380. (26) Machida, M.; Eidome, T.; Minami, S.; Buwono, H. P.; Hinokuma, S.; Nagao, Y.; Nakahara, Y., Tuning the Electron Density of Rh Supported on Metal Phosphates for Three-Way Catalysis. J. Phys. Chem. C 2015, 119 (21), 11653-11661. (27) Buwono, H. P.; Eidome, T.; Minami, S.; Hinokuma, S.; Nagao, Y.; Nakahara, Y.; Machida, M., Rh Supported on LaPO 4/SiO2 Nanocomposites as Thermally Stable Catalysts for TWC Applications. Emiss. Control Sci. Technol. 2015, 1, 284-291. (28) Yao, H. C.; Yao, Y. F. Y., Ceria in Automotive Exhaust Catalysts J. Catal. 1984, 86, 254-265. (29) Nunan, J. G.; Robota, H. J.; Cohn, M. J.; Bradley, S. A., Physicochemical Properties of Ce-Containing Three-Way Catalysts and the Effect of Ce on Catalyst Activity J. Catal. 1992, 133, 309-324. (30) Rao, G. R.; Fornasiero, P.; Monte, R. D.; Kašpar, J.; Vlaic, G.; Balducci, G.; Meriani, S.; Gubitosa, G.; Cremona, A.; Graziani, M., Reduction of NO over Partially Reduced Metal-Loaded CeO2–ZrO2 Solid Solutions. J. Catal. 1996, 162, 1-9. (31) Fornasiero, P.; Rao, G. R.; Kašpar, J.; L'Erario, F.; Graziani, M., Reduction of NO by CO over Rh/CeO2–ZrO2 Catalysts. J. Catal. 1998, 175 (2), 269-279. (32) Kašpar, J.; Fornasiero, P.; Graziani, M., Use of CeO2-Based Oxides in the Three-Way Catalysis. Catal. Today 1999, 50, 285-298. (33) Sugiura, M.; Ozawa, M.; Suda, A.; Suzuki, T.; Kanazawa, T., Development of Innovative Three-Way Catalysts Containing Ceria– Zirconia Solid Solutions with High Oxygen Storage/Release Capacity. Bull. Chem. Soc. Jpn. 2005, 78 (5), 752-767. (34) Rink, J.; Meister, N.; Herbst, F.; Votsmeier, M., Oxygen Storage in Three-Way-Catalysts is an Equilibrium Controlled Process: Experimental Investigation of the Redox Thermodynamics. Appl. Catal. B: Environ. 2017, 206, 104-114. (35) Beppu, K.; Hosokawa, S.; Teramura, K.; Tanaka, T., Oxygen Storage Capacity of Sr3Fe2O7−δ Having High Structural Stability. J. Mater. Chem. A 2015, 3 (25), 13540-13545. (36) Beppu, K.; Hosokawa, S.; Shibano, T.; Demizu, A.; Kato, K.; Wada, K.; Asakura, H.; Teramura, K.; Tanaka, T., Enhanced OxygenRelease/Storage Properties of Pd-Loaded Sr3Fe2O7−δ. Phys. Chem. Chem. Phys. 2017, 19 (21), 14107-14113.



(37) Beppu, K.; Hosokawa, S.; Asakura, H.; Teramura, K.; Tanaka, T., Role of Lattice Oxygen and Oxygen Vacancy Sites in Platinum Group Metal Catalysts Supported on Sr3Fe2O7−δ for NOSelective Reduction. Catal. Sci. Technol. 2018, 8 (1), 147-153. (38) Demizu, A.; Beppu, K.; Hosokawa, S.; Kato, K.; Asakura, H.; Teramura, K.; Tanaka, T., Oxygen Storage Property and Chemical Stability of SrFe1–xTixO3−δ with Robust Perovskite Structure. J. Phys. Chem. C 2017, 121 (35), 19358-19364. (39) Muraki, H.; Shinjoh, H.; Fujitani, Y., Effect of Lanthanum on the NO Reduction over Palladium Catalyst. Appl. Catal. 1986, 2, 325-335

(40) Datye, A. K.; Bravo, J.; Nelson, T. R.; Atanasova, P.; Lyubovsky, M.; Pfefferle, L., Catalyst Microstructure and Methane Oxidation Reactivity During the Pd-PdO Transformation on Alumina Supports. Appl. Catal. A: General 2000, 198, 179-196. (41) Chen, X.; Schwank, J. W.; Fisher, G. B.; Cheng, Y.; Jagner, M.; McCabe, R. W.; Katz, M. B.; Graham, G. W.; Pan, X., Nature of the Two-Step Temperature-Programmed Decomposition of PdO Supported on Alumina. Appl. Catal. A: General 2014, 475, 420-426.


Page 8 of 9

Page 9 of 9

Table of Contents C3H6+NO+CO+O2

CO2+N2+H2O Pd/SrFe1-xTixO3-d

80

Ti-rich

100

Excess O 2 condition

80

Fe-rich

at 773K

(PdO) N2 production (%)

100

NO conversion to N 2 + N2O (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


60 40 20

Fe-rich (PdOx cluster)

0 500

600 700 Temperature / K

60 40 20

Ti-rich

0 21000

22000 23000 Time / sec


24000

9

as Environmental Catalyst: Purification - ACS Publications

Recommend Documents