Restructuring Transition Metal Oxide Nanorods for 100% Selectivity in

Jun 3, 2013 - Bharat Kumar , Gyandshwar Kumar Rao , Soumen Saha , Ashok K. Ganguli. ChemPhysChem 2016 17 (10.1002/cphc.v17.1), 155-161 ...
0 downloads 0 Views 3MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Restructuring Transition Metal Oxide Nanorods for 100% Selectivity in Reduction of Nitric Oxide with Carbon Monoxide Shiran Zhang,† Junjun Shan,† Yuan Zhu,† Luan Nguyen,† Weixin Huang,† Hideto Yoshida,‡ Seiji Takeda,‡ and Franklin (Feng) Tao*,† †

Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana, 46556, United States Institute of Scientific and Industrial Research, Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan

‡

S Supporting Information *

ABSTRACT: Transition metal oxide is one of the main categories of heterogeneous catalysts. They exhibit multiple phases and oxidation states. Typically, they are prepared and/ or synthesized in solution or by vapor deposition. Here we report that a controlled reaction, in a gaseous environment, after synthesis can restructure the as-synthesized transition metal oxide nanorods into a new catalytic phase. Co3O4 nanorods with a preferentially exposed (110) surface can be restructured into nonstoichiometric CoO1−x nanorods. Structure and surface chemistry during the process were tracked with ambient pressure X-ray photoelectron spectroscopy (AP-XPS) and environmental transmission electron microscopy (ETEM). The restructured nanorods are highly active in reducing NO with CO, with 100% selectivity for the formation of N2 in temperatures of 250−520 °C. AP-XPS and E-TEM studies revealed the nonstoichiometric CoO1−x nanorods with a rock-salt structure as the active phase responsible for the 100% selectivity. This study suggests a route to generate new oxide catalysts. KEYWORDS: Transition metal oxide, in situ study, restructuring, catalytic selectivity, carbon monoxide, X-ray photoelectron spectroscopy

M

modified wet chemistry method reported in literature.18,22 The as-synthesized Co3O4 nanorods have dimensions of approximately 6 nm × 6 nm × 100−200 nm (Figure 1). The crystallization of Co3O4 is evidenced by XRD diffraction peaks

ost transition metal oxides are reducible. Many of them exist in different phases and exhibit multiple oxidation states.1−3 Some of them, such as VOx,2,4 Fe3O4,5−7 and CeO2,8−10 are catalysts or active components in heterogeneous catalysis. Similar to bimetallic nanocatalysts,11−13 a reducible transition metal oxide can be restructured under reaction conditions. Here we report the restructuring of pure Co3O4 in gaseous environments. We found that a catalytic phase, nonstoichiometric rock-salt cobalt monoxide CoO1−x, can be formed through a restructuring of Co3O4 nanorods under reaction conditions. The restructuring of Co3O4 and formation of nonstoichiometric rock-salt CoO1−x under reaction conditions were monitored through in situ studies using ambient pressure X-ray photoelectron spectroscopy (AP-XPS) and environmental transmission electron microscopy (E-TEM). The nonstoichiometric rock-salt CoO1−x is highly active in the reduction of nitric oxide (NO) with carbon monoxide (CO) for producing N2 with 100% selectivity. This catalyst, formed through restructuring, exhibits an activity similar to Rh-based catalysts.14−17 The 100% selectivity for production of N2 remains constant at 520 °C for at least 96 h. In situ studies of this catalyst show that the surface of cobalt monoxide nanorods is nonstoichiometric with ∼20%−25% oxygen vacancies. Co3O4 nanorods are active for CO oxidation, even at −77 °C, due to the high reactivity of surface lattice oxygen atoms.18−21 We synthesized Co3O4 nanorods through a © 2013 American Chemical Society

Figure 1. TEM images of as-synthesized Co3O4 nanorods. (a) Largescale image. (b) High-resolution TEM image of a representative Co3O4 nanorod. Received: April 26, 2013 Revised: May 29, 2013 Published: June 3, 2013 3310

dx.doi.org/10.1021/nl4015292 | Nano Lett. 2013, 13, 3310−3314

Nano Letters

Letter

pure CO or H2 at temperatures below 400 °C was identified in Figure S10 as well. The restructuring of bulk phase Co3O4 nanorods in reactive environment was observed by in situ studies of E-TEM. The phase transformation performed at 250−350 °C was clearly identified with in situ electron diffraction in E-TEM (Figure 3a

(Figure S1 of the Supporting Information). In photoemission studies of Co3O4 nanorods at room temperature (Figure 2a1),

Figure 2. Evolution of photoemission features of Co 2p (a) and Co 2p 3/2 (b) of Co3O4 in the mixtures of CO and NO at different reaction temperatures. The evolution from low temperatures to high temperatures shows restructuring of spinel Co3O4 to rock-salt CoO1−x under reaction conditions. The composition of reactant gases over the catalyst during data acquisition is 1 Torr NO and 3 Torr CO.

Co 2p3/2 at 779.8 eV and Co 2p1/2 at 795.1 eV can be clearly identified. These are consistent with those of nanoparticles23 and single crystals of Co3O4.19 Co 2p3/2 (or Co 2p1/2) can be deconvoluted into two asymmetric peaks (Figure S8) which are assigned to Co3+ of an octahedral site and Co2+ of a tetrahedral site of Co3O4, respectively;24 a very weak, broad shoulder on the high binding energy side at 789.6 eV is attributed to the satellite peak of Co3+ in octahedral sites of Co3O423 (Figure S8). In situ studies of Co3O4 nanorods under reaction conditions using AP-XPS showed that the surface region of Co3O4 nanorods restructures into nonstoichiometric CoO1−x in reactive environments, such as a mixture of CO and NO (Figure 2), pure CO (Figure S10a), or pure H2 (Figure S10b). These reduction processes were monitored with AP-XPS in pressure ranges of 0.1−10 Torr. AP-XPS studies of Co3O4 in a 3:1 mixture of CO and NO (Figure 2) and 1:1 (Figure S12) show there is no significant change of photoemission features of Co 2p in the temperature regime of 25 °C−250 °C in CO + NO. Thus, Co3O4 is not restructured at temperatures lower than 250 °C during catalysis. At temperatures higher than 250 °C, Co3O4 starts to restructure into CoO1−x. It is evidenced by the appearance of satellite peaks of Co 2p from Co2+ in an octahedral coordination with oxygen atoms in rock-salt CoO1−x marked in Figure 2a (black dashed line) and Figures S10 and S12. The satellite peaks of Co 2p of CoO1−x were identified at 786.4 and 803.0 eV for Co 2p3/2 and Co 2p1/2, respectively, which are about 6.6 and 7.9 eV higher than the corresponding main photoemission features. The photoemission features of Co 2p are consistent with those of CoO in literature.19,23 At 400 °C in CO + NO (Figure 2), Co3O4 is completely reduced to nonstoichiometric CoO1−x. Notably, there is no photoemission feature of metallic cobalt at 778.3 eV25 in Figure 2. Similar restructuring of Co3O4 to nonstoichiometric CoO1−x in

Figure 3. Diffraction patterns of Co3O4 at 250 °C (a) and 350 °C (b) and electron energy loss spectra at room temperature (25 °C) (c) and 350 °C (d) in a reactive environment.

and b). This restructuring was revealed in situ by an electron energy loss spectroscopy (EELS) study of E-TEM as well. The intensity of the first peak in the O−K edge of Co3O4 at room temperature (indicated by an arrow in Figure 3c) decreases drastically at 350 °C in reactants. In Figure 3d, this peak is missing. According to the previous studies,26−28 the decreased intensity of the first peak of O−K edge marked with an arrow in Figure 3c is caused by oxygen deficiency and a structural transition. The in situ EELS spectra obtained above 350 °C agreed well with those obtained from CoO particles in previous studies with ultrahigh vacuum.26,28,29 Catalytic performance was measured in a fixed-bed flow reactor. Figure 4 presents the evolution of catalytic performance of Co3O4 and the restructured catalyst, CoO1−x in the reduction of NO with CO as a function of reaction temperature. The new catalyst phase, nonstoichiometric rocksalt CoO1−x, is highly active in the reduction of NO with CO. 3311

dx.doi.org/10.1021/nl4015292 | Nano Lett. 2013, 13, 3310−3314

Nano Letters

Letter

Figure 5. Schematic of Co3O4 nanorods and the CoO formed through restructuring from Co3O4 under reaction conditions.

Selectivity to CO2 is 100% since CO2 is the only product in reduction of NO with CO. The turnover frequency (TOF) of N2 production at 250 °C is ∼0.08 N2 molecules per cobalt atom per second (see Section 3 in the Supporting Information), similar to Rh-based catalysts.16,17 In addition, the nonstoichiometric rock-salt CoO1−x exhibits a very similar catalytic performance in reactant mixtures of different CO and NO compositions (Figure 2 and Figures S11, S12, and S13). Oxygen vacancies are important structural defects for reducible transition metal oxides and active sites of heterogeneous catalysis. A measurement of O/M (M stands for metal) atomic ratio can reflect the existence and evolution of surface oxygen vacancies. O/Co atomic ratios of cobalt oxides during catalysis were calculated at different temperatures with the following method based on the in situ AP-XPS studies of an oxide catalyst during catalysis. First, the area ratio of O 1s to Co 2p of Co3O4 at room temperature was calculated. A factor can be obtained by using the area ratio of O 1s to Co 2p and dividing it by the stoichiometric ratio of O to Co in Co3O4, 4/ 3. Oxygen vacancies on Co3O4 could be generated during calcination (one of the experimental steps in preparation of Co3O4 nanorods) since oxygen atoms are probably lost during the calcination. In the process of cooling to room temperature in air upon calcination, H2O molecules in air typically adsorb and dissociate on oxygen vacancies. OH groups fill in oxygen vacancies. Therefore, the total number of surface oxygen atoms of calcinated Co3O4 nanorods at room temperature is almost identical to that of a stoichiometric surface of Co3O4 without oxygen vacancies. Thus, it is reasonable to assume a 1.33 O/Co ratio of Co3O4 at room temperature before catalysis as OH groups fill into oxygen vacancies. The area ratio (AO1s/ACo2p) at other temperatures is divided by this factor, giving the atomic ratio of oxygen to cobalt at this temperature. Figure 6 presents the O/Co atomic ratio as a function of reaction temperature. The atomic ratio of oxygen to cobalt of the catalyst, in reduction of NO with CO, exhibits a great deal of variation along the restructuring of Co3O4 to nonstoichiometric rock-salt CoO1−x (Figure 6). The atomic ratio continually decreases with an increase of reaction temperature. Notably, the measured O/ Co atomic ratio during catalysis, in the temperature regimes from 420 to 480 °C (the first heating cycle) and from 480 °C to 225 °C (the first cooling cycle upon the first heating cycle), is ∼75% (Figure 6). It is ∼20% lower than the stoichiometric ratio, 1:1 of rock-salt CoO. This suggests that the catalytically active surface of the rock-salt cobalt monoxide is nonstoichiometric. Notably, the low O/Co ratio of nonstoichiometric CoO1−x during catalysis at 400−480 °C (Figure 6) does not result from any over-reduction of Co3O4 to metallic Co since there are no photoemission features of Co 2p3/2 of

Figure 4. Catalytic activity (the left Y-axis, black line in figure) and selectivity (the right Y-axis, red line in figure) of Co3O4 and CoO1−x (formed through a restructuring at a temperature above 250 °C) at different temperatures in the mixture gas of NO (5%) and CO (15%) followed by cooling down to 300 °C and then 100 °C.

Its catalytic performance is significantly different from that of Co3O4. At 110 °C, Co3O4 exhibits a detectable conversion of NO (17%), but only N2O is produced. There is no selectivity for the production of N2. The conversion of NO at 150 °C is 85% with a low selectivity for production of N2 (9.0%); at 170 °C the conversion of NO reaches 100%. In situ studies using the AP-XPS suggest that the Co3O4 remains at surface region at 170 °C. In addition, AP-XPS studies showed that there is no significant change in the atomic ratio of C/Co at 170 °C in contrast to temperatures between 25 and 150 °C. It suggests the increase of conversion rate of NO from 25 to 170 °C does not result from a cleaning process of trace amounts of carbon species on the surface of Co3O4. Selectivity for the production of N2 on Co3O4 is only 0−20% at temperatures ≤250 °C. It increases rapidly in the temperature regime from 300 to 420 °C and reaches 100% at 420 °C and remains at this level through 480 °C or even higher. In the temperature regime from 420 to 480 °C, the catalytic surface is cobalt monoxide (Figure 2). Thus, the correlation between catalytic selectivity and surface chemistry of catalysts shows that the rock-salt CoO1−x is active for the reduction of nitric oxide with CO with a 100% selectivity for N2 production. To confirm that CoO1−x is the active phase and responsible for the 100% selectivity for N2 production, the nonstoichiometric CoO1−x catalyst formed during the first heating cycle was cooled down to 225 °C and further down to 100 °C. Notably, selectivity to the production of N2 on the nonstoichiometric CoO1−x at 300 and 250 °C is still 100% (Figure 4). Corresponding to the 100% selectivity to production of N2 in the temperature regime from 480 to 250 °C in the cooling process, in situ studies of AP-XPS showed the preservation of CoO1−x surface phases in the cooling process to 250 °C (Figures 2a12). The preservation of the surface phase of CoO1−x and selectivity for production of N2 at 100% in the temperature regime from 420 to 480 °C (first heating cycle) and from 480 to 250 °C (the first cooling cycle after the first heating cycles) definitely confirmed that CoO1−x is the active phase for reduction of NO with CO into N2 (Figure 5). Conversion of CO follows the stoichiometry of catalytic reaction 2CO + 2NO = N2 + 2CO2 toward production of N2 when selectivity to production of N2 is 100% (Figure S13). 3312

dx.doi.org/10.1021/nl4015292 | Nano Lett. 2013, 13, 3310−3314

Nano Letters

Letter

Figure 6. Atomic ratio of oxygen to cobalt of Co3O4 and CoO1−x (formed through a restructuring of Co3O4 at a temperature above 250 °C) during catalysis of CO + NO (1 Torr NO and 3 Torr CO) in the same temperature regime as catalytic measurements in Figure 4. The atomic ratios of O to Co at different temperatures were calculated through a strict implementation of the same set of XPS analytical parameters of O 1s and Co 2p photoemission features to all spectra obtained at different temperatures for minimization of a potential error.

Figure 7. Catalytic selectivity for production of N2 in reduction of NO with CO on CoO1−x at 520 °C.

conditions. This catalyst exhibits a much better catalytic performance than its original Co3O4. The correlation between catalytic selectivity and its corresponding surface chemistry was built. This nonstoichiometric rock-salt CoO1−x formed through restructuring Co3O4 is highly active for reducing NO with CO. It exhibits a TOF at 250 °C close to Rh-based catalysts reported at 250 °C.16,17 Although a comparison of TOF does not necessarily make a lot sense, the similarity in TOF still suggests the high activity of CoO1−x in reduction of NO with CO. It exhibits 100% selectivity to the production of N2 in the temperature regime of 250 °C−520 °C and a durability of 100% catalytic selectivity for production of N2 for 96 h at 520 °C .

metallic cobalt at 778.3 eV, even at a temperature as high as 480 °C in a gas mixture of CO and NO. Comparable studies of Co 2p photoemission feature of Co3O4 in the mixture of CO and NO and pure CO showed that Co3O4 is readily reduced to metallic cobalt in pure CO at 400 °C (Figure 2 versus Figure S10a) and only reduced to nonstoichiometric rock-salt CoO1−x at 400 °C in the mixture gas of CO and NO (Figure 2). Thus, the difference in restructuring behaviors of Co3O4, resulting from different reactive environments, shows that an in situ study is necessary for identification of restructured phases and searches of an appropriate reaction condition for preserving a new catalytic phase. The measured O/Co ratio of nonstoichiometric CoO1−x under reaction conditions of a mixture of CO and NO (Figure 6), suggests that the surface has ∼25% oxygen vacancies during the catalysis with 100% selectivity to N2. Thus, the new active phase for NO reduction with CO with 100% selectivity is termed nonstoichiometric cobalt monoxide, CoO1−x (x = ∼0.25). Oxygen vacancies play an important role in heterogeneous catalysis.8,9,30−32 In many cases, they act as oxygen storage.8,9 In the case of nonstoichiometric CoO1−x during reduction of NO with CO, oxygen vacancies are probably formed through the reaction of CO with surface lattice oxygen atoms to form CO2. Oxygen vacancies on nonstoichiometric CoO1−x could then be filled with oxygen atoms from NO molecules through dissociative adsorption of NO on Co2+. Two nitrogen atoms formed from the dissociation of two NO molecules probably couple to form a N2 molecule. We expect that theoretical calculations will help elucidate the reaction pathway on it. Industrial tests of durability of three-way catalysts33,34 were performed at 520 °C in mixtures of CO and NO with a ratio of 1:1 or 3:1. The stability of our CoO1−x catalyst for the reduction of nitric oxide with carbon monoxide was tested by measurements of catalytic performance on 5 mg of Co3O4 nanorods at 520 °C in a gas mixture in 5% NO + 15% CO + 80% Ar or 5% NO + 5% CO + 90% Ar with a flow rate of 50 mL min−1 (Figure 7). Our tests showed there is no change in selectivity for production of N2 within the first 96 h. In summary, we have experimentally demonstrated the formation of a restructured oxide catalyst through the restructuring of a transition metal oxide under reaction

■

ASSOCIATED CONTENT

* Supporting Information S

Synthesis of Co3O4 nanorods; measurements of catalytic activity and selectivity; calculation of turnover frequency; ambient pressure X-ray photoelectron spectroscopy and in situ studies of surface chemistry of catalysts during catalysis and under reaction conditions; photoemission features of Co3O4 and CoO1−x; restructuring of Co3O4 nanorods in different gases; catalytic performances in reactant mixture with different gas compositions; catalytic conversion of CO; AP-XPS studies of cobalt oxide in reactant mixture with different compositions; oxygen vacancies on the surface of CoO1−x; AP-XPS studies of cobalt oxides in reactant mixture of CO and NO with different pressures; representation of the origin of satellite peaks of Co3+ of Co3O4 and Co2+ of CoO 1−x; potential reaction mechanisms for nitric oxide reduction with carbon monoxide on cobalt oxides. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, Office of Science, U.S. Department of Energy under the grant DE-FG02-12ER1635. 3313

dx.doi.org/10.1021/nl4015292 | Nano Lett. 2013, 13, 3310−3314

Nano Letters

■

Letter

REFERENCES

(1) Wachs, I. E.; Routray, K. ACS Catal. 2012, 2, 1235. (2) Wachs, I. E. Appl. Catal., A 2011, 391, 36. (3) Wachs, I. E.; Roberts, C. A. Chem. Soc. Rev. 2010, 39, 5002. (4) Zhou, W.; Doura, K. F.; Watanabe, M.; Herzing, A. A.; Okunishi, E.; Ross-Medgaarden, E. I.; Wachs, I. E.; Kiely, C. J. ChemCatChem 2011, 3, 1045. (5) Merte, L. R.; Peng, G.; Bechstein, R.; Rieboldt, F.; Farberow, C. A.; Grabow, L. C.; Kudernatsch, W.; Wendt, S.; Laegsgaard, E.; Mavrikakis, M.; Besenbacher, F. Science 2012, 336, 889. (6) Edwards, M. A.; Whittle, D. M.; Rhodes, C.; Ward, A. M.; Rohan, D.; Shannon, M. D.; Hutchings, G. J.; Kiely, C. J. Phys. Chem. Chem. Phys. 2002, 4, 3902. (7) Rhodes, C.; Williams, B. P.; King, F.; Hutchings, G. J. Catal. Commun. 2002, 3, 381. (8) Campbell, C. T.; Peden, C. H. F. Science 2005, 309, 713. (9) Esch, F.; Fabris, S.; Zhou, L.; Montini, T.; Africh, C.; Fornasiero, P.; Comelli, G.; Rosei, R. Science 2005, 309, 752. (10) Farmer, J. A.; Campbell, C. T. Science 2010, 329, 933. (11) Tao, F.; Grass, M. E.; Zhang, Y.; Butcher, D. R.; Renzas, J. R.; Liu, Z.; Chung, J. Y.; Mun, B. S.; Salmeron, M.; Somorjai, G. A. Science 2008, 322, 932. (12) Tao, F.; Salmeron, M. Science 2011, 331, 171. (13) Zafeiratos, S.; Piccinin, S.; Teschner, D. Catal. Sci. Technol. 2012, 2, 1787. (14) Choi, K. I.; Vannice, M. A. J. Catal. 1991, 131, 1. (15) Yao, Y. F. Y. J. Catal. 1984, 87, 152. (16) Yoon, K. J.; Kang, H. K.; Yie, J. E. Korean J. Chem. Eng. 1997, 14, 399. (17) Oh, S. H.; Eickel, C. C. J. Catal. 1991, 128, 526. (18) Xie, X.; Li, Y.; Liu, Z.-Q.; Haruta, M.; Shen, W. Nature 2009, 458, 746. (19) Petitto, S. C.; Marsh, E. M.; Carson, G. A.; Langell, M. A. J. Mol. Catal A: Chem. 2008, 281, 49. (20) Petitto, S. C.; Langell, M. A. J. Vac. Sci. Technol., A 2004, 22, 1690. (21) Langell, M. A.; Kim, J. G.; Pugmire, D. L.; McCarroll, W. J. Vac. Sci. Technol., A 2001, 19, 1977. (22) Xie, X.; Shang, P.; Liu, Z.; Lv, Y.; Li, Y.; Shen, W. J. Phys. Chem. C 2010, 114, 2116. (23) Chuang, T. J.; Brundle, C. R.; Rice, D. W. Surf. Sci. 1976, 59, 413. (24) Kim, K. S. Phys. Rev. B 1975, 11, 2177. (25) J. F. Moulder, W. F. S., P. E. Sobol, K. D. Bomben Handbook of X-ray Photoelectron Spectroscopy; Perkin-Elmer Corporation: Waltham, MA, 1992. (26) Wang, Z. L.; Yin, J. S.; Mo, W. D.; Zhang, Z. J. J. Phys. Chem. B 1997, 101, 6793. (27) Zhang, Z. Ultramicroscopy 2007, 107, 598. (28) Zhao, Y.; Feltes, T. E.; Regalbuto, J. R.; Meyer, R. J.; Klie, R. F. J. Appl. Phys. 2010, 108, 063704. (29) Li, P.; Liu, J.; Nag, N.; Crozier, P. A. Appl. Catal., A 2006, 307, 212. (30) Wen, C.; Zhu, Y.; Ye, Y.; Zhang, S.; Cheng, F.; Liu, Y.; Wang, P.; Tao, F. ACS Nano 2012, 6, 9305. (31) Schaub, R.; Wahlstrom, E.; Ronnau, A.; Laegsgaard, E.; Stensgaard, I.; Besenbacher, F. Science 2003, 299, 377. (32) Wahlstrom, E.; Vestergaard, E. K.; Schaub, R.; Ronnau, A.; Vestergaard, M.; Laegsgaard, E.; Stensgaard, I.; Besenbacher, F. Science 2004, 303, 511. (33) Westerholm, R.; Egeback, K. E. Environ. Health Perspect. 1994, 102, 13. (34) Harrison, R. T. Catalytic converter exhaust system temperature tests; Forest Services, U.S. Department of Agriculture, Equipment Development Center, San Dimas, CA, 1977.

3314

dx.doi.org/10.1021/nl4015292 | Nano Lett. 2013, 13, 3310−3314