Reac Kinet Mech Cat DOI 10.1007/s11144-015-0962-9
Redox reaction of nitric oxide and carbon monoxide over Fe2O3 and Co3O4 phases L. A. Flores-Sanchez1 • J. M. Quintana-Melgoza1 R. Valdez2 • A. Olivas3 • M. Avalos-Borja4
•
Received: 14 August 2015 / Accepted: 29 November 2015 Ó Akade´miai Kiado´, Budapest, Hungary 2015
Abstract Fe2O3 and Co3O4 phases are synthesized by the thermal treatment in an air-flow of hydrated salts based on Fe and Co. The materials are characterized by X-ray diffraction, energy dispersive spectroscopy, scanning electron microscopy, thermogravimetric analysis and surface area measurements. Fe2O3 and Co3O4 were tested for nitric oxide reduction with carbon monoxide in a 1:5 gas phase ratio and showed reaction rates in a range from 8.7 9 10-11 to 5.9 9 10-10 mol s-1 g-1 with an activation energy interval from 50.2 to 54.4 kJ mol-1. Fe2O3 and Co3O4 achieved (100 and 98) % NO conversion with (96 and 80) % selectivity to N2 at (275 and 350) °C. Fe2O3 and Co3O4 have the advantages such as facile synthesis, low-cost, good thermal stability, high activity and selective to N2. These results regarding activity and selectivity are better or similar to some catalysts based on noble metals (Rh, Pt and Pd) currently used for air pollutants control. Keywords selectivity
Oxides catalysts NO reduction CO oxidation High activity and
& M. Avalos-Borja
[email protected] 1
Facultad de Ciencias Quı´micas e Ingenierı´a, UABC, Calzada Universidad, 14418, Parque Industrial Internacional Tijuana, 22390 Tijuana, BC, Mexico
2
Centro de Investigacio´n y Desarrollo Tecnolo´gico en Electroquı´mica, Unidad Tijuana, Km. 26.5 Carretera Libre Tijuana-Tecate, esq. Blvd Nogales Parque Industrial el Florido, 22444 Tijuana, BC, Mexico
3
Centro de Nanociencias y Nanotecnologı´a-UNAM, Km. 107 Carr. Tijuana-Ensenada, 22860 Ensenada, BC, Mexico
4
Divisio´n de Materiales Avanzados, Instituto Potosino de Investigacio´n Cientı´fica y Tecnolo´gica, San Luis Potosı´, SLP, Mexico
123
Reac Kinet Mech Cat
Introduction Research into materials with activity in the simultaneous decomposition of NO and CO is very important as both polluting gases represent 72 % of the six most common air pollutants listed by the USEPA [1]. Currently, the catalysts in use for controlling automobile exhaust emissions (Rh, Pd and Pt) have disadvantages, such as high costs of precursors (Rh(NO3)3xH2O Aldrich 83750: 444,000.00 USD kg-1, Pd(NO3)22H2O Aldrich 76070: 54,800.00 USD kg-1 and PtCl2 Aldrich 206091: 154,500.00 USD kg-1), scarcity and high demand [2–5]. Therefore, research into alternative materials based on transition metal oxides (Fe and Co) of low-cost (Fe(NO3)39H2O Aldrich 216828: 305.00 USD kg-1 and Co(NO3)26H2O Aldrich 230375: 375.00 USD kg-1), high activity and selectivity to reaction redox of NO and CO to obtain harmless or less toxic products (N2, CO2) is highly interesting in this field [6–9]. However, many materials based on Rh, Pd, Pt, Ru, Ag and Au need high temperatures (300–800) °C, or use a long reaction time (2–12 h) to obtain chemically the catalytic systems according to reports in the literature [10–21]. Therefore, it is of interest to prepare low-cost catalysts with high thermal stability and high activity to reduce NO with CO. The aim of this work is to synthesize Fe2O3 and Co3O4 catalysts by a procedure involving less time and lower temperatures and test them in the decomposition of NO and CO. The synthesized materials were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), thermogravimetric analysis-differential scanning calorimetric (TGA-DSC), surface area and crystal size.
Experimental Synthesis of materials Fe2O3 and Co3O4 were synthesized by direct decomposition of precursors salts: iron nitrate nonahydrate (Aldrich 216828, Fe(NO3)39H2O) and cobalt nitrate hexahydrate (Aldrich 239267, Co(NO3)26H2O), respectively. Air-flow at atmospheric pressure was used as a carrier and oxidant (120 cm3 min-1) at 600 °C (heating rate 42 °C min-1) for 1 h of reaction time. Characterization The elemental composition and morphological structure of the materials synthesized were analyzed by EDS and SEM in a FEI Quanta 200 (20–25 kV) SEM. XRD was carried out in a Bruker D8 Advance diffractometer using Cu Ka radiation (40 kV, 30 mA) with a wavelength of 0.154 nm. Fe2O3 and Co3O4 crystal size were estimated from the half width of (104) and (311) peaks, respectively, using the Scherrer equation [22]. TGA-DSC was performed using a TA Instruments SDTQ600. Samples were heated into an alumina pan from 30 °C up to 1000 °C (heating rate 10 °C min-1) under a continuous flow of dry nitrogen (100 cm3 min-1).
123
Reac Kinet Mech Cat
Surface area measurements were made using a Quantachrome Instruments Autosorb-1. The surface area was determined from N2 adsorption isotherm using the BET method [23]. Prior to measurement, the samples were degassed under vacuum at 200 °C until a constant weight was achieved. Catalytic activity Fe2O3 and Co3O4 were tested in NO reduction with CO, this reaction was studied in a temperature range of 100–500 °C using a tubular continuous reactor at atmospheric pressure. The reaction conditions were as follows: 0.2 g of catalytic mass was placed in a quartz micro-reactor and was given a pretreatment at 400 °C for 1 h with air-flow at 20 cm3 min-1. Following this treatment, the catalytic system was cooled to 50 °C prior to the NO and CO reaction. The catalytic system was fed for 8 h with a mixture of NO ? CO (ratio NO/CO = 1/5), Praxair UHP gases, NO flow (20 cm3 min-1 NO 2 vol% balance He) and CO flow (40 cm3 min-1 CO 5 vol% balance He). Reactants and products were analyzed in a gas chromatograph GOW-MAC (TCD) with a packed column carbosphere 80 9 100 (ALLTECH). The gas flow was kept in contact with the catalyst for 15 min to achieve a steady state for the reaction. Subsequently, the gas samples were taken at every 25 °C in the 100–600 °C range. The reactants (NO, CO) and products (N2, N2O, CO2) were injected into a gas chromatograph with an automatic injection valve (2 cm3). The concentrations of reactants (NO, CO) and products (N2, CO2, N2O) were calculated following standard procedures [24]. The reaction rate was calculated using Eq. 1 (more details about this equation in Ref. [25]). 1x n P n r½NO ¼ k ð1Þ 1þx RT Here r[NO] is the reaction rate, x is the fraction NO conversion, n is the reaction order, P is the partial pressure of NO in flux, R is the gas constant = 8.3145 (J mol-1 K-1), T is the temperature (K). Rate constants (k) were calculated using Eq. 2 (more details about this equation in Ref. [25]). Zx Fzo FVf n 1þx n k¼ dx 1x w Fzo
ð2Þ
0
Here w is the catalyst weight (g), F is the mole of NO by fed per second (mol s-1), zo is the molar fraction of NO in the feed (Fzo is the 5.5 9 10-7 mol s-1), Vf is the feeding volume per mole (cm3 mol-1). Therefore, FVf is the total flow powered = 1 cm3 s-1. The reaction orders were estimated according to the methodology reported by Oh et al. [26].
123
Reac Kinet Mech Cat
Results and discussion Analysis by XRD Fig. 1 shows a comparison between XRD patterns of synthesized materials (label E) and patterns from JCPDS-ICDD database cards (label T); Fe2O3 (33-0664) and Co3O4 (42-1467) [27]. In each catalyst, only one crystalline phase was identified and the experimental patterns correspond to the following proposed phases: Fe2O3 and Co3O4. The Scherrer equation [22] was used to calculate the crystal size of materials, the results were 92 and 100 nm for Fe2O3 and Co3O4 respectively (see Table 1). Analysis by SEM–EDS SEM micrographs (Fig. 2) exhibit a different morphology and elongated particles of average size (1.6 ± 0.7 9 1.0 ± 0.6) lm for Fe2O3, (Figs. 2a and 2b) and
Co3O4 (T) 533 622
440
511
422
400
222
220
311
Intensity (Arb. Units)
Co3O4 (E)
20
30
40
50
10 10
214
60
300
018
116
024
Fe2O3 (T) 113
110
012
104
Fe2O3 (E)
70
80
2θ Fig. 1 X-ray patterns of the material synthesized by thermal decomposition at 600 °C for 1 h reaction time in air-flow at 120 cm3 min-1. Label ‘‘E’’ refers to experimentally obtained diffractograms and ‘‘T’’ to reported diffractograms in the JCPDS-ICDD [27] database. The corresponding card numbers are shown in parenthesis: Fe2O3 (33-0664) and Co3O4 (42-1467)
123
Reac Kinet Mech Cat Table 1 Physical properties of catalysts synthesized in this work: average particle size (PS) and its standard deviation (r), crystal size calculate by Scherrer equation (CS) [22], surface area (SA) [23] CS nm (hkl)
SA (r) (m2 g-1)
Catalyst
PS l 9 b (r) (lm)
Fe2O3
1.6 ± 0.7 9 1.0 ± 0.6
92 (104)
7 ± 0.35
Co3O4
2.5 ± 0.8 9 2.0 ± 0.7
100 (311)
4 ± 0.20
Fig. 2 SEM micrographs of materials synthesized by thermal decomposition at 600 °C for 1 h in airflow 120 cm3 min-1. Images of Fe2O3 (a, b) exhibit fibrous agglomerates (1.6 ± 0.7 9 1.0 ± 0.6) lm, and images of Co3O4 (c, d) show elongated agglomerates (2.5 ± 0.8 9 2.0 ± 0.7) lm
(2.5 ± 0.8 9 2.0 ± 0.7) lm for Co3O4, (Figs. 2a, 2c and 2d) (average and standard deviations from at least 100 particles per sample). The direct decomposition enabled a single product to be obtained from nitrate salts under established conditions (Table 2). EDS spectra (Fig. 3) show the characteristic signals expected for each sample, this corresponds with results from XRD (Fig. 1), where only one phase per sample was found without remnants contamination. The decomposition of the
123
Reac Kinet Mech Cat Table 2 Synthesis conditions of chemical reactions for thermal decomposition method: nitrate salts of Fe and Co as precursors, temperature, air-flow, product identified by XRD, atomic percent (at.%) of transition metal (M) and oxygen (O) determined by EDS, standard deviation (r) and chemical reaction yield (CRY) Precursor
Temperature (°C)
Air-flow (cm3 min-1)
Product
at.% EDS (M:O) (r)
CRY (%)
Fe(NO3)39H2O
600
120
Fe2O3
(34.24:65.87) ± 3.2
93
Co(NO3)26H2O
600
120
Co3O4
(34.36:65.63) ± 6.8
91
O Kα
Co K α
Co L α , β
Counts (Arb. Units)
Co 3 O 4 Co K β
Fe K α
O Kα
Fe 2 O 3
Fe L α , β Fe K β
0
1
2
3
4
5 Energy (keV)
6
7
8
9
10
Fig. 3 EDS of synthesized oxides by thermal decomposition at 600 °C for 1 h reaction time in air-flow at 120 cm3 min-1. There is no evidence of external contamination
precursor salts and reaction due to products have been studied by different authors finding that an oxide is produced under non-reducing conditions [28–30]. Similarly, the thermal oxidative decomposition of metal nitrates has an oxide as a final product, according to previously reported work [25]. Analysis by TGA-DSC Fig. 4 shows typical thermograms for Fe2O3 and Co3O4 [31–33]. Fig.4a shows a TGA-DSC plot similar to hematite (Fe2O3), which is thermally stable up to 1000 °C
123
Reac Kinet Mech Cat
(a)
0.0
60 Weight loss
40
Heat flow
20 Fe2O3
0 -20
-0.8
-40 Exo up
Heat flow (mW)
Weight loss (%)
-0.4
-60
-1.2
-80
200
(b)
400 600 o Temperature ( C)
Co3O4
0
Weight loss Heat flow
-40
Heat flow (mW)
-20
-4
-6
-100 1000 20
0
-2
Weight loss (%)
800
Co3O4 to CoO -8
-10
-60
Exo up
200
400
600
800
-80 1000
o
Temperature ( C) Fig. 4 TGA-DSC of iron (a) and cobalt (b) oxide samples synthesized by thermal decomposition at 600 °C for 1 h reaction time in air-flow at 120 cm3 min-1. The thermograms confirm a high thermal stability
[31]. Fig. 4b exhibits an endothermic peak at 830 °C, which is characteristic of the Co3O4 to CoO transformation [32, 33]. We also noticed a 1.10 % weight loss for Fe2O3 (Fig. 4a) in the interval from 115 to 830 °C and a 1.07 % weight loss for Co3O4 (Fig. 4b) in the interval from 121 to 715 °C; this weight loss could be
123
Reac Kinet Mech Cat
attributed to excess oxygen adsorbed on the surface of the materials, and may explain the excess in atomic oxygen as measured by EDS (Table 2). Catalytic activity Fig. 5 shows the reaction of NO with CO over Fe2O3 and Co3O4 catalysts. The reaction without catalyst achieved a NO conversion of 4 % at most (at 600 °C). Fe2O3 is the best catalyst with NO conversion of 100 % at 275 °C and is maintained until 600 °C. The activity curves for Co3O4 exhibit a light off and lift off in % NO conversion at temperature range from 350 to 600 °C. At 350 °C, cobaltosic oxide achieves 97 % NO conversion and subsequently light-off to 83 % at 400 °C, followed by a significant lift-off in conversion upon exceeding 400 °C to achieve 100 % at 500 °C. The light-off of NO conversion from 350 to 400 °C by Co3O4 could be caused by CO2 adsorption, which saturates the catalytic surface, obstructing adsorption of the reactants NO and CO, in agreement with previous works [17, 34, 35]. Reactivation upon exceeding 400 °C is due to the desorption of CO2 from the surface. Fig. 6 shows the formation of N2 and N2O during NO reduction with CO as a function of temperature. Selectivity to N2 at 275 °C resulted
100
Conversion NO (%)
80
Fe2O3
60
Co3O4
40
20 NC
0 100
200
300 400 Temperature (°C)
500
600
Fig. 5 Catalytic activity for NO reduction with CO over Fe2O3 (stars), and Co3O4 (circles) catalysts, (squares) indicates no catalyst (NC) present. Reaction conditions: NO flow: 20 cm3 min-1 (NO 2 vol%, balance He); CO flow: 40 cm3 min-1 (CO 5 vol%, balance He); total flow of reactants: 60 cm3 min-1; catalyst weight: 0.2 g
123
Reac Kinet Mech Cat 100
Selectivity (%)
80
60
Fe2O3
40 Co3O4
20 Fe2O3
Co3O4
0 100
200
300 o Temperature ( C)
400
500
Fig. 6 Selectivity to N2 and N2O during NO reduction with CO over Fe2O3 (N2 filled stars, N2O open stars), and Co3O4 (N2 filled circles, N2O open circles). Reaction conditions: NO flow: 20 cm3 min-1 (NO 2 vol%, balance He); CO flow: 40 cm3 min-1 (CO 5 vol%, balance He); total flow of reactants: 60 cm3 min-1; catalyst weight: 0.2 g
in 96 and 30 % by Fe2O3 and Co3O4, respectively, (see Table 3). Lower selectivity to N2O by both catalysts (1 and 5 % at 200 °C by Fe2O3 and Co3O4, respectively) is observed, and tends to be zero when N2 production is 100 %, suggesting decomposition into molecular nitrogen and atomic oxygen, according to the mechanism proposed by Athanasios et al. [36]. The light-off of NO reduction (350–400) °C may be due to adsorbed carbon dioxide (CO2) and carbonates (CO3)2-, occupying active sites on CoII and CoIII of Co3O4 catalyst, in agreement with literature reports [25, 37, 38], decreasing N2 selectivity by 12 %. The reduction of NO with CO is carried out by reduction–oxidation of catalytic surface to oxidize the CO to CO2 and reduce NO to N. The reduction–oxidation states of metal are expressed as follows MIII ? MII ? MIII, where M is the metal Fe or Co [37–41]. Fe2O3 presents a higher reaction order (n = 0.67) as compared to Co3O4 (n = 0.33), the order of reaction of NO with CO was determined according to Oh et al. [26], the results are congruent with the high reaction rate (r = 5.9 9 10-10 mol s-1 g-1) of Fe2O3, which is 6.7 times faster than Co3O4 (r = 8.7 9 10-11 mol s-1 g-1). Table 3 shows the kinetic characteristics of unsupported catalysts based on iron and cobalt prepared in this work and those reported in the literature for supported catalysts based on rhodium, palladium and platinum. Some authors have reported methodologies with long reaction time (1–4 h) and high temperature
123
123
0.20
0.20
0.30
0.10–0.30
0.05
0.20
0.10
0.02
Fe2O3
Co3O4
Pd/a-Al2O3
Pd/Ce0.6Zr0.4O2/Al2O3
Rh/Ce0.6Zr0.4O2
Pt/c-Al2O3
Rh/Al2O3
Rh/CeO2
UD unreported data
Cw (g)
Catalyst
86,000
UD
4500
50,000
UD
17,500
109,000
94,000
GHSV (h-1)
0.33
0.67
Order (n)
8.7 9 10
3.3 9 10-10
2.2 9 10-8
5.9 9 10-10 -11
k (mol g-1 s-1)
r (mol s-1 g-1)
98 (350 °C) 90 (150 °C) 100 (357 °C) 100 (290 °C) 25 (500 °C) 100 (325 °C) 98 (410 °C)
10-90 (200 °C) 30–70 (357) °C UD-68 (290 °C) 19-UD (500 °C) UD-23 (400 °C) 82–18 (300 °C)
100 (275 °C)
Conversion NO (%)
80–0 (350 °C)
96–0 (275 °C)
SN2–SN2O (%)
UD
UD
UD
67–86 (200–257) °C
62.8 (197–227) °C
50.2 (145–286) °C
54.4 (125–250) °C
50.2 (125–250) °C
Ea (kJ mol-1)
[46]
[43]
[10]
[45]
[44]
[14]
This paper
This paper
Reference number
Table 3 Kinetic characteristics of catalysts prepared via thermal decomposition and other reported materials base on Pd, Rh and Pt: Catalyst weight (Cw), gas hour space velocity (GHSV), reaction order (n), reaction rate (r) (Eq. 1), rate constant (k) from Eq. 2, selectivity (SN2 and SN2O), NO conversion and apparent activation energy (Ea)
Reac Kinet Mech Cat
Reac Kinet Mech Cat
(500–900 °C) to synthesize Fe2O3 and Co3O4 phases [37, 39–42]. However, these studies of catalytic activity use different experimental conditions [37, 40], present less conversion of NO (30 % at 200 °C) [42], need high temperature (310, 700 and 900) °C to convert 100 % of NO [39, 41] or are theoretical studies [38]. We can see that the results obtained by low-cost materials such as Fe2O3 and Co3O4 are similar or better than the supported catalysts based on precious metals as Rh, Pt and Pd [10, 14, 43–46].
Conclusions We propose that the thermal decomposition methodology allows synthesis of unsupported, low-cost catalysts with high activity and selectivity to N2 in reaction redox of NO and CO. Fe2O3 achieved (at 275 °C) 100 % NO conversion, 96 % selectivity to N2 with an activation energy of 50.2 kJ mol-1. Fe2O3 has advantages in nitric oxide reduction when compared to some high-cost materials as Rh, Pt and Pd. Furthermore, it is an easy catalytic material to synthesize and exhibit high thermal stability. Therefore, Fe2O3 may be an important alternative to the simultaneous decomposition of main gaseous pollutants such as NO and CO to obtain harmless products as N2 and CO2. Acknowledgments We would like to thank CONACYT for a scholarship to LAFS and financial support through grants CB-151551 and LINAN-0260860. We are very grateful to Gladis J. Labrada, Beatriz A. Rivera, Antonio Go´mez, Eric Flores, Graham M. Tippett and Carlos A. Olivas for technical assistance. We also thank MyDCI-FCQI-UABC, LINAN-IPICyT, IF-UNAM and CNyN-UNAM, for providing laboratory facilities.
References 1. National Emissions Invetory (2008) version 2 U.S. Environmental Protection Agency 2012, North Carolina 2. Diwell AF, Rajaram RR, Shaw HA, Truex TJ (1991) Stud Surf Sci Catal 71:139–152 3. Takahashi N, Shinjoh H, Iijima T, Suzuki T, Yamazaki K, Yokota K, Suzuki H, Miyoshi N, Matsumoto S, Tanizawa T, Tanaka T, Tateishi S, Kasahara K (1996) Catal Today 27:63–69 4. Steel MCF (1991) Stud Surf Sci Catal 71:105–114 5. Studt F, Abild-Pedersen F, Bligaard T, Sørensen RZ, Christensen CH, Nørskov JK (2008) Science 320:1320–1322 6. Yu Q, Wu X, Yao X, Liu B, Gao F, Wang J, Dong L (2011) Catal Commun 12:1311–1317 7. Reichert D, Finke T, Atanassova N, Bockhorn H, Kureti S (2008) Appl Catal B 84:803–812 8. Pasel J, Ka¨ßner P, Montanari B, Gazzano M, Vaccari A, Makowski W, Lojewski T, Dziembaj R, Papp H (1998) Appl Catal B 18:199–213 9. Zhang F, Zhang S, Guan N, Schreier E, Richter M, Eckelt R, Fricke R (2007) Appl Catal B 73:209–219 10. Meng M, Lin P, Fu Y (1997) Catal Lett 48:213–222 11. Patel A, Shukla P, Rufford T, Wang S, Chen J, Rudolph V, Zhu Z (2011) Appl Catal A 409–410:55–65 12. Ferna´ndez-Garcı´a M, Martı´nez-Arias A, Belver C, Anderson JA, Conesa JC, Soria J (2000) J Catal 190:387–395 13. Hungrı´a AB, Calvino JJ, Anderson JA, Martı´nez-Arias A (2006) Appl Catal B 62:359–368 14. Pisanu AM, Gigola CE (1999) Appl Catal B 20:179–189 15. Hornung A, Muhler M, Ertl G (2000) Top Catal 11/12:263–270
123
Reac Kinet Mech Cat 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.
Ueda A, Haruta M (1999) Gold Bull 32(1):3–11 Spassova I, Khristova M, Panayotov D, Mehandjiev D (1999) J Catal 185:43–57 Radev L, Khristova M, Mehandjiev D, Samuneva B (2006) Catal Lett 112(3–4):181–186 Xiaoyuan J, Yanrong J, Hua HP, Xiaoming Z (2005) Catal Lett 104(3–4):169–175 Lv Y, Zhang H, Cao Y, Dong L, Zhang L, Yao K, Gao F, Dong L, Chen Y (2012) J Colloid Interface Sci 372:63–72 Peter SD, Garbowski E, Perrichon V, Primet M (2000) Catal Lett 70:27–33 Patterson AL (1939) Phys Rev 56(15):978–982 Brunauer S, Emmett PH, Teller E (1938) J Am Chem Soc 60:309–319 ´ valos M (2002) React Kinet Catal Lett 76(1):131–140 Quintana JM, Go´mez A, A Flores-Sanchez LA, Quintana-Melgoza JM, Olivas A, Avalos-Borja M (2015) Reac Kinet Mech Cat 114:597–609 Oh SH, Fisher GB, Carpenter JE, Goodman DW (1986) J Catal 100:360–376 Joint Committee on Powder Diffraction Standards (JCPDS)-International Centre for Diffraction Data (ICDD) (2012) 33-0664, 42-1467 Stern KH (1972) J Phys Chem 1(3):747–772 Wieczorek K, Kozak AJ (1999) J Therm Anal Calorim 58:647–651 Ehrhardt C, Gjikaj M, Brockner W (2005) Thermochim Acta 432:36–40 Woo JA, Hyo SJ, Hyoung JC (2015) R Soc Chem 5:23094–23100 Aguilar JA, Esneider MA, Herna´ndez MB, Pech MI, Shaji S (2013) J Alloy Compd 574:278–282 Tang Chih-Wei, Wang Chen-Bin, Chien Shu-Hua (2008) Thermochim Acta 473:68–73 Dunne JA, Rao M, Sircar S, Gorte RJ, Myers AL (1996) Langmuir 12:5896–5904 Horiuchi T, Hidaka H, Fukui T, Kubo Y, Horio M, Suzuki K, Mori T (1998) Appl Catal A 167:195–202 Ladavos AK, Pomonis PJ (1993) Appl Catal B 2:27–47 Jansson J, Skoglundh M, Fridell E, Thorma¨hlen P (2001) Top Catal 16/17(1–4):385–389 Xie Y, Dong F, Heinbuch S, Rocca JJ, Bernstein ER (2010) Phys Chem Chem Phys 12:947–959 Hayhurst AN, Lawrence AD (1997) Combust Flame 110:351–365 Randall H, Doepper R, Renken A (1997) Ind Eng Chem Res 36:2996–3001 Randall H, Doepper R, Renken A (1998) Appl Catal B 17:357–369 Pavani Sreekanth M, Panagiotis G, Smirniotis G (2008) Catal Lett 122:37–42 Li M, Xiaodong W, Cao Y, Liu S, Weng D, Ran R (2013) J Colloid Interface Sci 408:157–163 Di Monte R, Kaspar J, Fornasiero P, Graziani M, Paze´ C, Gubitosa G (2002) Inorg Chim Acta 334:318–326 Fornasiero P, Ranga Rao G, Kasˇpar J, L’Erario F, Graziani M (1998) J Catal 175:269–279 Cho BK (1994) J Catal 148:697–708
123