Uv Absorption Of Zro2

Ce Ce

C: 3640 cm-1 SCHEME 11.1. Surface OH groups

Ce I Ce Ce

(Scheme 11.1).30 IR peak positions of different hydroxyl groups for CeO2 sample are also given in Scheme 11.1. After passing hydrogen and then water at 873 K, the sample was evacuated and the IR spectrum was measured. The intensity of the OH groups on different supports increased in the order: Al2O3 < CeO2-ZrO2/Al2O3 < NiO(20)-Al2O3(20)-CeO2(6)-ZrO2(54). Significant amount of surface OH groups were present under reaction conditions. In addition to increasing the total concentration of OH groups on ceria, ZrO2 also increases the relative concentration of the mono-coordinated, basic OH groups. The availability of higher amount of OH groups during the reaction conditions is, perhaps, the cause for the lower deactivation of the CeO2-ZrO2 catalysts.

Diffuse Reflectance UV-Visible Spectroscopy: Samples 1A-6A (activated at 473 K) showed an intense, asymmetric diffuse reflectance UV band with a maximum at 240 nm and a shoulder around 310 nm. The former dominated in samples with higher Ni-content, indicating its origin from O^Ni charge transitions.31 The shoulder is due to O ^ Ce/Zr transitions.32'33 Additionally, weak d-d bands of oc-tahedrally coordinated Ni2+ were also observed at 724, 649, and 400 nm.31'34 The intensity of these bands increased with increasing Ni-content. The Ce/Zr ratio considerably affected the DRUV-visible spectra (Fig. 11.5). Sample 8A (activated at 473 K) showed an additional, asymmetric broad band in the region 400 to 650 nm attributable to the d-d transitions of Ni+ ions.34 The DRUV-VIS data also support the conclusions drawn earlier from XPS, namely, that ceria promotes the reduction of NiO to Ni metal. On treating with dry H2, the intensity of the charge transfer band was reduced considerably and the d-d bands disappeared. The samples became black and the reflectance was considerably lower. A broad absorption beyond 350 nm appeared. All these features suggest that Ni2+ ions in the original samples were reduced to metallic nickel.35'36

EPR Spectroscopy-Defect Sites and Oxygen—Centered Radicals: EPR spectra of pure CeO2 reduced with dry hydrogen revealed the presence of two types of paramagnetic species, A, due to Ce3+ (gi = 1.965and g|| = 1.940) and B, due to O2- ions23-25 (g1 = 2.030, g2 = 2.016, and g3 = 2.011), the concentration of which depended on the reduction temperature (7r). Pure ZrO2 did not form these paramagnetic species when reduced in hydrogen. CeO2-ZrO2 behaved in a manner similar to pure CeO2. Upon reoxidation by exposure to air at 298 K, the signals due to species A disappeared and of B decreased and broadened. Specie A and B could be generated, repeatedly, in several cycles indicating the facile redox behavior of CeO2 and CeO2-ZrO2 supports.

Vis Doped Ceo2

FIGURE 11.5. Diffuse reflectance UV-visible spectra of NiO-CeO2-ZrO2.22

Wavelength (nm)

FIGURE 11.5. Diffuse reflectance UV-visible spectra of NiO-CeO2-ZrO2.22

Temperature-programmed reduction experiments also indicated that the reduction of ceria is facilitated in the presence of zirconia. The reduction of ceria occurred at 620 K in CeO2-ZrO2 compared to 658 K in CeO2. In NiO-containing samples, signals due to O- (g = 2.008) and a broad feature due to Ni+ species37'38 (g = 2.21 for 1A-6A and 2.36 for 8A) were observed additionally. Interestingly, sample 7A containing less CeO2 did not show this signal, indicating that CeO2 promotes the reduction. Samples 7C and 10D, prepared with a template showed broader features (g = 2.24, AHpp = 907 G for 7C and g = 2.16, AHpp = 1140 G for 10D) than 7B or 9B prepared without a template (g = 2.22, AHpp = 352 G for 7B and g = 2.22, AHpp = 436 G for 9B). Nickel ions are probably better dispersed in materials prepared with template (7C and 10D) and the signals are dipolar-broadened.

EPR of Ni-Reduction with Hydrogen, Hydrocarbons, and Alcohols: Sample 1A, reduced in dry hydrogen (for 1 h at elevated temperatures) showed two types of Ni signals: Species I was characterized by sharp signals at g|| = 2.08 and = 2.009 whose intensity did not change with the reduction temperature (Tr). Species II was characterized by a broad signal (A Hpp = 340 G) at g = 2.20 whose intensity increased with the reduction temperature (Fig. 11.6a). Sample 2A when reduced in hydrogen showed three types of species (species III in addition to species I and II observed in sample 1A). The spectra of 2A at different reduction temperatures are shown in Fig. 11.6b. Species III was characterized by a very broad signal on which the signals due to species I and II were superimposed. Species I was observed at 573 to 798 K, species II (with significant intensity) was seen above 748 K and species III from 673 K onward. The signal intensity of III did not change significantly beyond 773 K, but that of II increased continuously with the reduction temperature.

2000 2500 3000 3500 4000 4500 Magnetic Field (G)

2000 2500 3000 3500 4000 4500 Magnetic Field (G)

FIGURE 11.6. EPR spectra of (a) NiO(1)-CeO2(49.5)-ZrO2(49.5) and (b) NiO(5)-CeO2(47.5)-ZrO2(47.5) reduced in hydrogen at different temperatures. Signals due to species I, II, and III are marked.22

2000 2500 3000 3500 4000 4500 Magnetic Field (G)

FIGURE 11.6. EPR spectra of (a) NiO(1)-CeO2(49.5)-ZrO2(49.5) and (b) NiO(5)-CeO2(47.5)-ZrO2(47.5) reduced in hydrogen at different temperatures. Signals due to species I, II, and III are marked.22

The spectra of samples 6A, 7A, and 8A (containing 40 wt.% NiO) reduced in hydrogen were mainly due to species III. Species II was not present. Species I was present, but the signals were weak and masked by the intense, broad signal of species III. The CeO2-ZrO2 composition affected both the signal position (g value) and the line width at Tr = 798 K; for example: Sample 6A — g value = 3.1 (line width = 1660 G), 7A — 3.1 (2130 G), and 8A — 2.4 (1850 G) respectively. The shift in the EPR signal with Tr was smaller (100 G) in sample 7A (containing more ZrO2 than CeO2) than in 6A (320 G; Ce:Zr = 1:1) and 8A (600 G; containing only CeO2). Reduction of Ni (species III) occurred at lower temperatures (573 K) in CeO2 containing samples. In other words, CeO2 facilitated the reduction of Ni.

When Ce4+ (coordination number = 8 in the CeO2 lattice) is substituted for a six-coordinated Zr4+ ion in the ZrO2 lattice, defect centers are created. Also, doping ceria with zirconia promotes formation of Ce3+ ions due to the smaller size of Zr4+ cations that take part in removing the strain associated with the increase of ionic size accompanying the Ce4+ to Ce3+ transition. These defects in solid CeO2-ZrO2 superlattice and the redox properties of Ce and Zr ions at temperatures above 873 K are responsible for the superior oxygen storage capacity and the ion transport properties. EPR spectroscopy has identified three types of paramagnetic defects in CeO2-ZrO2 oxides viz., electrons, "quasi-free" or trapped in anion vacancies (signals at g = 1.965 and 1.943), O- ions (signal at g = 2.008) and O- ions (signals at g = 2.083, 2.033, and 2.001).23-25'37 These species arise by the following interactions:

The oxygen storage capacity of CeO2-ZrO2 catalysts was correlated in the past to the concentration of these oxide ion species.23-25 Samples 8A and 6A showed signals at g = 2.36 and 2.21, respectively, due to a small amount of Ni substituted in the

Ce3+ + O2 ^ Ce4+(O2-^) Ni+ + O2 ^ Ni2+(O2~*) Ce4+ + O2- ^ Ce3+ (O-)

CeO2-ZrO2 solid lattice. The latter is reduced to a Ni+ species:

Ce4+-O2--Zr4+ ^ Ce3+-O--Zr4+ Ce3+ + Ni2+ ^ Ce4+ + Ni+ Ni+urf + O2 ^ Ni(O2-^)

Part of the Ni+ ions, subsequently, react with oxygen and formNi2+ (O2-^) species.

The reduction experiments have revealed at least three types of Ni-species in our NiO-CeO2-ZrO2 samples: (I) Ni substituted in the lattice forming Ni(O2-^) ions (signals at g|| = 2.028 and g± = 2.009),37 (II) nanosize crystallites of NiO, forming superparamagnetic metallic Ni (g = 2.195; linewidth = 340 G)39, and (III) larger particles ofNiO containing ferromagnetic Ni.40 Sample 1 (1 wt.%Ni) contained species I and II, sample 2 (5 wt.% Ni) contained all the three species, while samples 6A-8A (40 wt.% Ni) contained species I and III. The reducibility of these species decreased in the order: I > III > II. Reduction experiments at steam reforming conditions in the presence of both H2 and hydrocarbons have shown an increased reducibility of nickel ions in the presence of hydrocarbons and long-chain alcohol.

11.2.1.4. Catalytic Activity of NiO-CeO2-ZrO2—Influence ofNi Crystallite Size and Support Composition

Hydrogenation of Benzene: Benzene hydrogenation is a "facile" reaction and is indicative of the number of surface Ni atoms available for catalysis. The activity of the catalysts in the benzene hydrogenation reaction was investigated at 453 K.22 Cyclohexane was the only product of the reaction. Benzene conversion (Table 11.4) increased with increasing Ni content (samples 2A-4A) up to 20 wt.%.

A marked effect of the CeO2/ZrO2 composition (in samples containing 40 wt.% NiO) on the catalytic activity was noticed. The catalysts with Ce:Zr = 1:1 (6A) were not only more active (than 7A and 8A) but were also stable during the reaction. Sample 8A containing no zirconia in the support showed a low activity. The NiO crystallite size (Table 11.2) in these compositions varied in the order: 7A < 6A < 8A. It may be recalled that on ceria-based catalysts the crystallite size of nickel metal was similar to that of NiO. The higher activity for 6A than 7A indicates that in addition to accessibility of

TABLE 11.4. Benzene hydrogenation3

Benzene

Sample No. Catalyst Conversion (wt.%)b TOFc

aReaction conditions: catalyst = 1 g, benzene : H2 = 1:4, temperature = 453 K, time = 1 h. b Reaction with thiophene (20 ppm)-containing benzene feed.

cT0F = turnover frequency (number of moles of benzene converted per mole ofNi per hour).

TABLE 11.5. Steam reforming of n-octane

Catalyst (Relative) Dry Gas CO/CO2

nickel, the reducibility-reoxidizability as well as the oxygen storage-release capacity of the support plays crucial roles determining catalytic activity.

Steam Reforming ofn-Decane: Sample 7C, prepared by hydrothermal synthesis in the presence of CTABr and TAOH templates, has the highest catalytic activity (H2 yield) in the steam reforming of n-decane (Table 11.5). Even though samples 7B and 7C have identical bulk composition (Table 11.1), a similar surface area, and porosities (Table 11.2), accessible nickel metal on the surface is much higher on 7C (Table 11.3) accounting for the higher catalytic activity of the latter. Template molecules present during the synthesis of NiO-CeO2-ZrO2 reduce the particle size and enhance the accessibility of nickel. It is interesting to note that the concentration of CO2 (produced by the water gas shift) follows the reverse trend. The crystallites of Ni in 7C (larger than those in 7B) are less active, as expected, in the water-gas shift reactions. 10D having a smaller concentration of surface nickel is also less active. Consequently, catalyst deactivation decreased in the order: 10D > 7B > 7C.

Steam Reforming ofEthanol: Even though all the samples containing 20 wt.% or above of Ni exhibited similar activity and selectivity (Table 11.6), their stability over a 500 h duration run varied; sample 6A, with 40% Ni and a CeO2/ZrO2 ratio of 1 was the most stable. The Ce/Zr ratio has a marked effect on the catalytic stability of these materials.

11.2.2. Au andAu-Pt Bimetallic Nanoparticles in MCM-41 for Preferential Oxidation of CO in the Presence ofH2

Supported noble metals such as Au and Pt are active for CO oxidation. Since Pt-based catalysts operate at relatively higher reaction temperatures (403 to 473 K), they are not very selective for CO oxidation in hydrogen-rich streams.41 Au catalysts,

TABLE 11.6. Steam reforming of ethanola

Sample

Was this article helpful?

0 0

Post a comment