Polymorphous transformation of rod-shaped iron oxides and their catalytic properties in selective reduction of NO by NH₃

Abstract

Polymorphous transformation of rod-shaped γ-Fe₂O₃ was applied to fabricate Fe₃O₄/Fe₂O₃ nanorods. Hydrogen reduction of γ-Fe₂O₃ nanorods at 350 °C yielded Fe₃O₄ nanorods with similar size; re-oxidation of the resulting Fe₃O₄ nanorods produced γ-Fe₂O₃ at 500 °C and α-Fe₂O₃ nanorods at 600 °C. When applied to catalyze selective reduction of NO with NH₃, the activity followed the order γ-Fe₂O₃ > γ-Fe₂O₃-500 > α-Fe₂O₃ > Fe₃O₄, which was well correlated with their crystalline structures. The superior performance of γ-Fe₂O₃ nanorods was attributed to the simultaneous exposure of Fe³⁺ and O²⁻, which favoured the adsorption and activation of NH₃ and NO molecules.

---

Introduction

Nanostructured iron oxides, mostly Fe₂O₃ and Fe₃O₄, have attracted increasing attention in catalysis due to their tunable size, morphology, and crystal phase on the nanometer level.^1–4 Among the four polymorphs of Fe₂O₃, α- and γ-phases are the most extensively studied cases;⁵ their size and shape altered the catalytic performance considerably. α-Fe₂O₃ nanorods had a higher activity for CO oxidation than cubic nanocrystals, primarily because of the enrichment of active iron ion on the {110} facets.⁶ Polyhedral γ-Fe₂O₃ particles were more active and selective for styrene oxidation to benzaldehyde than spherical particles; and this was ascribed to the exposed {110} planes that provided more Fe ion for the activation of H₂O₂.⁷ Moreover, the crystal phase of Fe₂O₃ affected their catalytic properties as well. For example, γ-Fe₂O₃ nanorods catalyzed selective reduction of NO with NH₃ more efficiently than α-Fe₂O₃ nanorods.^8,9 The γ-phase preferentially exposed the {110} and {100} facets that facilitated the adsorption and activation of NH₃ and NO molecules.

The crystalline structure of iron oxides could be described as close-packed arrays of iron cation and oxygen anion with flexible configurations. Polymorphous transformation is one of the most effective routes to mediate the crystal phases among iron oxides; it often involves thermal treatment of a proper precursor at suitable temperature and under oxidizing or reducing atmospheres.^5,10,11 For example, calcination of γ-Fe₂O₃ particles at 370–600 °C in air yielded α-type particles;^5,10 oxidation of Fe₃O₄ particles produced α- or γ-Fe₂O₃ particles at 200–700 °C.¹⁰ While hydrogen reduction of α-Fe₂O₃ nanorings at 360 °C led to the formation of Fe₃O₄ nanorings, which could be further oxidized into γ-Fe₂O₃ nanorings by air at 240 °C.¹¹ Following such a strategy, octahedral γ-Fe₂O₃ nanoparticles,¹² hierarchical γ-Fe₂O₃ dendrite,¹³ Fe₃O₄ disc¹⁴ and rods¹⁵ have been synthesized. In this work, we studied polymorphous transformation of rod-shaped γ-Fe₂O₃ to Fe₃O₄ and α-Fe₂O₃ nanorods. The influence of crystal phase of these rod-shaped iron oxides on their catalytic performances in selective reduction of NO with NH₃ was examined.

Experimental

Synthetic procedure

All chemical reagents were commercially available in analytical purity and used without further purification. Rod-shaped β-FeOOH was prepared by aqueous-phase precipitation of ferric salt with Na₂CO₃.^8,9 An aqueous solution containing 5.38 g FeCl₃·6H₂O, 50 mL polyethylene glycol (PEG), and 150 mL water was gradually heated to 120 °C. 200 mL 0.2 M Na₂CO₃ aqueous solution was added through a syringe pump at a rate of 5.5 mL min⁻¹. The mixture was further aged at 120 °C for 1 h. The precipitate was collected by filtration, washed with water and ethanol, and dried at 50 °C for 6 h in vacuum. γ-Fe₂O₃ nanorods were obtained by refluxing the β-FeOOH precursor in PEG. A slurry mixture of 5.0 g β-FeOOH nanorods and 500 mL PEG was gradually heated to 200 °C and refluxed for 24 h under nitrogen flow. The resulting solid was washed with water and ethanol and dried at 50 °C for 12 h in vacuum. Fe₃O₄ nanorods were obtained by reducing the γ-Fe₂O₃ nanorods at 350 °C for 3 h with a 5% H₂/N₂ mixture. Re-oxidation of Fe₃O₄ nanorods with a 6% O₂/N₂ atmosphere at 500 °C for 0.5 h yielded γ-Fe₂O₃-500 nanorods; while re-oxidizing the Fe₃O₄ nanorods with a 20% O₂/N₂ atmosphere at 600 °C for 4 h produced α-Fe₂O₃ nanorods.

Analysis and measurements

X-ray diffraction (XRD) patterns were recorded on a Rigaku D/Max2500V/PC powder diffractometer with a CuKα radiation source that was operated at 40 kV and 200 mA. Nitrogen adsorption–desorption isotherms were measured on a Nova 4200e instrument at −196 °C. Before the measurement, the sample was degassed at 300 °C for 6 h under vacuum. Micro-Raman spectra were recorded on a Renishaw inVia spectro-meter at room temperature by using a 532 nm solid laser as the excitation source. X-ray photoelectron spectra (XPS) were recorded with a VG ESCALAB MK2 spectrometer using an Al Kα radiation operated at an accelerating voltage of 12.5 kV. The γ-Fe₂O₃ sample was pressed into a thin disc and mounted on a sample rod placed in a pre-treatment chamber; and it was heated to 350 °C and kept at that temperature for 3 h under flowing a 5% H₂/N₂ mixture. After being cooled down to room temperature, the sample was evacuated and transferred into the analysis chamber where the spectra of Fe2p and O1s were recorded. The charging effect was corrected by adjusting the binding energy of C1s to 284.6 eV. Transmission electron microscopy (TEM) images were taken on a Philips Tecnai G² Spirit microscope (120 kV). High-resolution TEM (HRTEM) images were recorded on a G² F30S-Twin (300 kV) and JEOL 2100 microscope (200 kV). Field-emission scanning electron microscopy (FESEM) images were taken on a Philips Fei Quanta 200F instrument operated at 20 kV. Temperature-programmed reduction of hydrogen (H₂-TPR) was conducted with an AutoChem II 2920 instrument (Micromeritics) and analyzed with a thermal conductor detector (TCD). 50 mg samples were pre-treated with He (30 mL min⁻¹ for Fe₃O₄) or a 3.0 vol% O₂/He mixture (30 mL min⁻¹ for α-Fe₂O₃, γ-Fe₂O₃ and γ-Fe₂O₃-500) at 400 °C for 0.5 h. After being purged with He for 1 h at room temperature, the sample was heated to 900 °C at a ramp of 5 °C min⁻¹ under flowing a 5.0 vol% H₂/N₂ mixture (30 mL min⁻¹).

SCR of NO by NH₃

Selective reduction of NO with NH₃ was conducted with a continuous-flow quartz tubular reactor under atmospheric pressure. 100 mg catalysts (40–60 mesh) were treated with a 3.0 vol% O₂/He mixture (60 mL min⁻¹, α-Fe₂O₃, γ-Fe₂O₃ and γ-Fe₂O₃-500) or He (60 mL min⁻¹, Fe₃O₄) at 400 °C for 0.5 h. The feed gas contained 1000 ppm NO, 1000 ppm NH₃, and 3 vol% O₂ balanced with He (120 mL min⁻¹). The effluent from the reactor was analyzed by an on-line mass spectrometer (Omini-Star) for the analysis of N₂ and N₂O. A NO/NO₂/NO_x analyzer (Thermo Environmental Instruments Inc.) was used to monitor the concentrations of NO and NO₂. The conversion of NO and the selectivity of N₂ were calculated according to the following equations.

NO conversion% = ([NO]_in − [NO]_out) × 100/[NO]_in

N₂ selectivity = [N₂]_out/([N₂]_out + [N₂O]_out) × 100%

Results and discussion

Fig. 1 shows XRD pattern of the γ-Fe₂O₃ nanorods. Typical diffraction lines of γ-Fe₂O₃ were clearly identified. Compared with the standard pattern of γ-Fe₂O₃, however, the intensities of the (400) and (440) lines enhanced remarkably, indicating the preferential exposure of the {110} and {100} planes. The micro-Raman spectrum (Fig. 1b) showed the typical band of A_1g mode at 700 cm⁻¹, the band of E_g mode at 500 cm⁻¹, and the band of T_2g mode at 350 cm⁻¹,¹⁶ confirming the crystalline structure of γ-Fe₂O₃. The XPS profile (Fig. 2) showed binding energies of Fe2p at 710.9 and 724.4 eV and a satellite peak at 719.2 eV, suggesting the presence of Fe³⁺ species on the surface.¹⁷ There were two types of oxygen species in the O1s spectrum; the binding energy at 530 eV represented the lattice oxygen (O_α) while the binding energy at 531.6 eV referred to the deficient oxygen (O_β);¹⁸ the concentration of O_β on the γ-Fe₂O₃ nanorods was 38%. SEM and TEM images (Fig. 3) verified that the γ-Fe₂O₃ nanorod had an average diameter of 20 nm and a mean length of 130 nm. When viewed along [211] direction, the interplanar distances of 0.29 and 0.49 nm with an interfacial angle of 90° was assigned to the (022) and (111) planes. The lattice spacings of 0.29 nm with a dihedral angle of 60°, viewed along [111] direction, suggested the anisotropic growth of the nanorod along [110] direction (Fig. 3). Taking the rectangular tip into account, the γ-Fe₂O₃ nanorod was proposed to expose {110} and {001} side planes and {110} top planes. According to the atomic configuration of γ-Fe₂O₃,¹⁹ the {110} and {001} facets are simultaneously terminated with iron cation and oxygen anion.

Fig. 1 XRD patterns (a) and Raman spectra (b) of the Fe₂O₃ and Fe₃O₄ nanorods.

Fig. 2 Fe2p (a) and O1s (b) XPS profiles of the iron oxides nanorods.

Fig. 3 SEM and TEM images of the γ-Fe₂O₃ nanorods.

Hydrogen reduction of the γ-Fe₂O₃ nanorods at 350 °C yielded Fe₃O₄ nanorods. Although the obtained sample exhibited similar diffraction lines to the γ-Fe₂O₃ nanorods (Fig. 1a), micro-Raman spectra discriminated its crystal phase. The band of A_1g mode of Fe₃O₄ appeared at 665 cm⁻¹ while bands of T_2g mode appeared at 540 and 311 cm⁻¹ (Fig. 1b).²⁰ Moreover, the XPS profile (Fig. 2) showed typical bonding energies of Fe2p_3/2 at 710 eV and Fe2p_1/2 at 724 eV; quantitative analysis verified that the Fe³⁺/Fe²⁺ ratio was about 1.95 that is very close to the stoichiometric value of 2.0 in Fe₃O₄,^21,22 evidencing that γ-Fe₂O₃ was entirely converted into Fe₃O₄. The Fe₃O₄ nanorod had a diameter of about 20 nm and a mean length of 130 nm; and it was enclosed with {110} and {001} side planes and {110} top planes (Fig. 4a–c). By far, there is no general consensus on the atomic configurations of these polar surfaces on Fe₃O₄.^23–25 Fe₃O₄ consists of tetrahedral Fe³⁺ site and octahedral site that was equally occupied by Fe²⁺ and Fe³⁺;¹⁰ the facets might be terminated by tetrahedral iron cation, oxygen anion or octahedral iron cation.

Fig. 4 TEM images of the Fe₃O₄ (a–c), γ-Fe₂O₃-500 (d–f), and α-Fe₂O₃ (g–i) nanorods.

Re-oxidization of the Fe₃O₄ nanorods at 500 °C yielded γ-Fe₂O₃-500 nanorods with characteristic XRD pattern (Fig. 1a) and Raman bands at 700, 500, and 350 cm⁻¹ (Fig. 1b).¹⁶ The XPS profile was similar to that of the original γ-Fe₂O₃ nanorods with the presence of only Fe³⁺ on the surface, but the concentration of O_β species lowered to 19% (Fig. 2). This might be caused by the reconstruction of the porous nanostructure during the polymorphous transformation of the rod-shaped iron oxides. The γ-Fe₂O₃-500 nanorod (Fig. 4d–f) exposed similar crystal facets as the original γ-Fe₂O₃ nanorod, although the exposure of the {100} and {001} facets slightly decreased from 58% to 40%.

Further increasing the temperature of oxidation up to 600 °C, α-Fe₂O₃ nanorods were produced, as being evidenced by the diffraction lines of α-Fe₂O₃ (Fig. 1a) and the micro-Raman spectrum (Fig. 1b). A_1g mode in α-Fe₂O₃ appeared at 221 and 497 cm⁻¹ and E_g modes presented at 292, 406 and 611 cm⁻¹.¹⁶ TEM analysis indicated that the size of the α-Fe₂O₃ nanorod (Fig. 4g) was almost the same as the Fe₃O₄ nanorods. HRTEM observation verified that the α-Fe₂O₃ nanorods mainly exposed the (210) plane, which is terminated by ferric ion only. The surface areas of the rod-shaped iron oxides were 111 m² g⁻¹ for γ-Fe₂O₃, 138 m² g⁻¹ for Fe₃O₄, 92 m² g⁻¹ for γ-Fe₂O₃-500, and 56 m² g⁻¹ for α-Fe₂O₃.

Fig. 5 shows H₂-TPR profiles of the rod-shaped iron oxides. The two γ-Fe₂O₃ samples showed an intense reduction peak at 200–400 °C and two broad reduction peaks at 400–900 °C. The total amount of hydrogen consumed approximately equalled to the stoichiometric amount (18.75 mmol H₂ per g) for the conversion of Fe₂O₃ to Fe. Since the amount of hydrogen consumed at 200–400 °C was about 13% of the total amount, this low-temperature reduction could be assigned to the reduction of Fe₂O₃ to Fe₃O₄. The two broad reduction peaks at high-temperature range corresponded to the further reduction of Fe₃O₄ to Fe.^26,27 The reduction of Fe₃O₄ nanorods consisted of two steps: Fe₃O₄ to FeO at 400–650 °C and FeO to Fe at 650–900 °C. The total amount of H₂ consumed was 17.53 mmol H₂ per g, equalling to the theoretical value (17.24 mmol H₂ per g) for the reduction of Fe₃O₄ to Fe. The ratio of the amounts of hydrogen consumed during the two steps was 0.25, confirming a sequential process from Fe₃O₄ to Fe through FeO.²⁷ α-Fe₂O₃ nanorods showed a similar reduction pattern as γ-Fe₂O₃ but the reduction of Fe₂O₃ to Fe₃O₄ occurred at relatively a higher temperature, about 400 °C, suggesting that γ-Fe₂O₃ was more reducible than α-Fe₂O₃ with similar size and shape. This is linked to the dominant exposure of the {110} and {100} facets on the γ-Fe₂O₃ nanorod, which are simultaneously terminated by Fe³⁺ and O²⁻ sites. This kind of surface atomic configuration caused the Fe–O bond more reducible and reactive^8,9 and also favored the activation of hydrogen molecule.²⁸

Fig. 5 H₂-TPR profiles of the Fe₂O₃ and Fe₃O₄ nanorods.

Polymorphous transformation of Fe₂O₃ to Fe₃O₄ starts with the removal of surface oxygen ion by hydrogen, followed by the diffusion of Fe²⁺ to the bulk. The crystal structure of Fe₂O₃ determines the rearrangements of iron cation and oxygen anion.²⁹ Since γ-Fe₂O₃ and Fe₃O₄ share a very similar cubic structure,²³ the phase transformation is actually a topotactic process without crystallographic change, making the migration of Fe²⁺ is more energetically favourable. Therefore, γ-Fe₂O₃ nanorods were easily reduced into Fe₃O₄ nanorods. On the other hand, re-oxidation of Fe₃O₄ to Fe₂O₃ is initialized by the adsorption of oxygen molecule, followed by electron transfer from Fe²⁺ to Fe³⁺, requiring the diffusion of bulk iron cation toward the surface.³⁰ Because of the similar cubic structure between Fe₃O₄ and γ-Fe₂O₃, the topotactic conversion only involved rearrangement of ferric ions inside the unit cell but without requiring nucleation species.^30,31 As a consequence, Fe₃O₄ nanorods were converted into γ-Fe₂O₃ nanorods at 500 °C. On the other hand, the oxygen layer in α-Fe₂O₃ is in a hexagonal-close-packed pattern with a two-layer periodicity, differing from the face-center-cubic stacking manner in Fe₃O₄, re-oxidation of Fe₃O₄ to α-Fe₂O₃ requires a rearrangement of oxygen layers and a nucleation species.^32,33 Therefore, the conversion of Fe₃O₄ nanorods to α-Fe₂O₃ nanorods occurred at a higher temperature, 600 °C.

Fig. 6a compares NO conversion on the rod-shaped iron oxides in selective reduction of NO with NH₃. Both the conversion of NO below 350 °C and the temperature window for 80% NO conversion followed the order γ-Fe₂O₃ > γ-Fe₂O₃-500 > α-Fe₂O₃ > Fe₃O₄. At 225 °C, the conversion of NO on γ-Fe₂O₃ nanorods approached 80%, whereas it was 38% on γ-Fe₂O₃-500, 13% on α-Fe₂O₃, and only 2% on Fe₃O₄. This result evidences the superior activity of the γ-Fe₂O₃ nanorods. In addition, the selectivity of N₂ was more than 95% while the selectivity of N₂O was less than 5% in the temperature region investigated (Fig. 6b). Long-term stability test at 300 °C indicated that the conversion of NO over the γ-Fe₂O₃ nanorods retained at 90% for 100 hours; and the spent catalyst kept the original crystal phase, size and shape (Fig. 6c), demonstrating the outstanding stability of rod-shaped γ-Fe₂O₃ under the reaction conditions. This prominent performance of the γ-Fe₂O₃ nanorods is closely associated with the simultaneously exposed iron and oxygen ions on the {100} and {001} facets. The reaction pathway of NH₃-SCR on iron oxides generally involves the initial adsorption and activation of NH₃ on the Lewis acidic Fe³⁺ site and H-removal by the neighbouring oxygen anion, followed by its reaction with the weakly adsorbed NO to release N₂.^34,35 In this context, the reducibility of iron oxides plays a major role in determining the activity. For the γ-Fe₂O₃ nanorods, the surface atomic configuration favored a superior redox feature with the Fe–O bond more reducible and reactive.²⁸ The relatively lower activity of the γ-Fe₂O₃-500 nanorods might be due to the less exposing fraction of reactive {110} and {100} facets (58% vs. 40%) and the decreased amount of surface deficient oxygen species. The lower activity of the α-Fe₂O₃ nanorods was caused by the dominant exposure of the (210) planes that is terminated by ferric ion only. This surface provides adsorption site of NH₃ but lacks neighbouring oxygen anion for the activation of ammonia molecule.⁸ The much lower activity of the Fe₃O₄ nanorods was ascribed to the surface coordination environment;²⁵ there was at least 1/3 iron species (Fe²⁺) in the unit cell, which are almost inert for the NH₃-SCR reaction.

Fig. 6 (a) NO conversion, (b) N₂/N₂O selectivities as a function of temperature in NH₃-SCR over the Fe₂O₃ and Fe₃O₄ nanorods and (c) long-term test of the γ-Fe₂O₃ nanorods at 300 °C. Reaction conditions: 100 mg catalyst, 1000 ppm NO/1000 ppm NH₃/3.0 vol% O₂/He, 72 000 mL g⁻¹ h⁻¹; the temperature was increased to 500 °C at a rate of 3 °C min⁻¹.

Conclusions

Polymorphous transformation of rod-shaped γ-Fe₂O₃ to Fe₃O₄ and α-Fe₂O₃ was achieved at proper temperatures and under reducing and/or oxidative atmospheres. Since γ-Fe₂O₃ and Fe₃O₄ had a similar crystalline structure, their polymorphous transformation followed a typical topotactic process under relatively mild conditions; whereas the phase transfer from Fe₃O₄ to α-Fe₂O₃ required a higher temperature because of lattice reconstruction. The γ-Fe₂O₃ nanorods showed superior catalytic activity toward NH₃-SCR, primarily because of the preferentially exposed {110} and {100} facets that were co-terminated by iron and oxygen ions.

Acknowledgements

We acknowledge financial support for this research work from the National Natural Science Foundation of China (21025312, 21303193, and 21321002) and the National Key Basic Research Program of China (2013CB933100).

Notes and references

1. J. Ouyang, J. Pei, Q. Kuang, Z. Xie and L. Zheng, ACS Appl. Mater. Interfaces, 2014, 6, 12505–12514.
2. X. Mou, X. Wei, Y. Li and W. Shen, CrystEngComm, 2012, 14, 5107–5120.
3. H. Liang, W. Chen, X. Jiang, X. Xu, B. Xu and Z. Wang, J. Mater. Chem. A, 2014, 2, 4340–4346.
4. W. Cheng, J. He, T. Yao, Z. Sun, Y. Jiang, Q. Liu, S. Jiang, F. Hu, Z. Xie, B. He, W. Yan and S. Wei, J. Am. Chem. Soc., 2014, 136, 10393–10398.
5. L. Machala, J. Tucek and R. Zboril, Chem. Mater., 2011, 23, 3255–3272.
6. X. Liu, J. Liu, Z. Chang, X. Sun and Y. Li, Catal. Commun., 2011, 12, 530–534.
7. N. Zhao, W. Ma, Z. Cui, W. Song, C. Xu and M. Gao, ACS Nano, 2009, 3, 1775–1780.
8. X. Mou, B. Zhang, Y. Li, L. Yao, X. Wei, D. S. Su and W. Shen, Angew. Chem., Int. Ed., 2012, 51, 2989–2993.
9. X. Mou, Y. Li, B. Zhang, L. Yao, X. Wei, D. S. Su and W. Shen, Eur. J. Inorg. Chem., 2012, 14, 2684–2690.
10. R. M. Cornell and U. Schwertmann, The iron oxides: structure, properties, occurrence and uses, Wiley-VCH, Weinheim, Germany, 2nd edn, 2003.
11. C. J. Jia, L. D. Sun, F. Luo, X. D. Han, L. J. Heyderman, Z. G. Yan, C. H. Yan, K. Zheng, Z. Zhang, M. Takano, N. Hayashi, M. Eltschka, M. Klaui, U. Rudiger, T. Kasama, L. Cervera Gontard, R. E. Dunin-Borkowski, G. Tzvetkov and J. Raabe, J. Am. Chem. Soc., 2008, 130, 16968–16977.
12. W. X. Li, B. L. Lv and Y. Xu, J. Nanopart. Res., 2013, 15, 2114.
13. G. Sun, B. Dong, M. Cao, B. Wei and C. Hu, Chem. Mater., 2011, 23, 1587–1593.
14. Y. Yang, X. Liu, Y. Lv, T. S. Herng, X. Xu, W. Xia, T. Zhang, J. Fang, W. Xiao and J. Ding, Adv. Funct. Mater., 2015, 25, 812–820.
15. L. Hou, Q. Zhang, F. Jérôme, D. Duprez, H. Zhang and S. Royer, Appl. Catal., B, 2014, 144, 739–749.
16. X. Zhang, Y. Niu, X. Meng, Y. Li and J. Zhao, CrystEngComm, 2013, 15, 8166–8172.
17. D. Li, S. J. Li, Y. Zhang, J. J. Jiang, W. J. Gong, J. H. Wang and Z. D. Zhang, Mater. Res. Innovations, 2015, 19, 58–62.
18. A. P. Grosvenor, B. A. Kobe and N. S. McIntyre, Surf. Sci., 2004, 572, 217–227.
19. X. L. Hu and J. C. Yu, Adv. Funct. Mater., 2008, 18, 880–887.
20. J. M. Li, A. C. H. Huan, L. Wang, Y. W. Du and D. Feng, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 6876–6878.
21. T. Fujii, F. M. F. de Groot, G. A. Sawatzky, F. C. Voogt, T. Hibma and K. Okada, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 3195–3202.
22. G. Bhargava, I. Gouzman, C. M. Chun, T. A. Ramanarayanan and S. L. Bernasek, Appl. Surf. Sci., 2007, 253, 4322–4329.
23. R. Jansen, V. Brabers and H. van Kempen, Surf. Sci., 1995, 328, 237–247.
24. B. Stanka, W. Hebenstreit, U. Diebold and S. Chambers, Surf. Sci., 2000, 448, 49–63.
25. N. Spiridis, J. Barbasz, Z. Łodziana and J. Korecki, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 74, 155423.
26. S. J. Hug, J. Colloid Interface Sci., 1997, 188, 415–422.
27. J. Zieliński, I. Zglinicka, L. Znak and Z. Kaszkur, Appl. Catal., A, 2010, 381, 191–196.
28. Q. Han, Z. Liu, Y. Xu, Z. Chen, T. Wang and H. Zhang, J. Phys. Chem. C, 2007, 111, 5034–5038.
29. U. Colombo, F. Gazzarrini and G. Lanzavecchia, Mater. Sci. Eng., 1967, 2, 125–135.
30. K. J. Gallagher, W. Feitknecht and U. Mannweiler, Nature, 1968, 217, 1118–1121.
31. F. Jiao, J. C. Jumas, M. Womes, A. V. Chadwick, A. Harrison and P. G. Bruce, J. Am. Chem. Soc., 2006, 128, 12905–12909.
32. U. Colombo, F. Gazzarrini, G. Lanzavecchia and G. Sironi, Science, 1965, 147, 1033.
33. S. Nie, E. Starodub, M. Monti, D. A. Siegel, L. Vergara, F. El Gabaly, N. C. Bartelt, J. De la Figuera and K. F. McCarty, J. Am. Chem. Soc., 2013, 135, 10091–10098.
34. P. Granger and V. I. Parvulescu, Chem. Rev., 2011, 111, 3155–3207.
35. K. Otto and M. Shelef, J. Catal., 1970, 18, 184–192.

---