Noble metal ions in CeO₂ and TiO₂: synthesis, structure and catalytic properties

Abstract

In the past four decades, CeO₂ has been recognized as an attractive material in the area of auto exhaust catalysis because of its unique redox properties. In the presence of CeO₂, the catalytic activity of noble metals supported on Al₂O₃ is enhanced due to higher dispersion of noble metals in their ionic form. In the last few years, we have been exploring an entirely new approach of dispersing noble metal ions on CeO₂ and TiO₂ matrices for redox catalysis. In this study, the dispersion of noble metal ions by solution combustion as well as other methods over CeO₂ and TiO₂ resulting mainly in Ce_1−xM_xO_2−δ, Ce_1−x−yTi_xM_yO_2−δ, Ce_1−x−ySn_xM_yO_2−δ, Ce_1−x−yFe_xM_yO_2−δ, Ce_1−x−yZr_xM_yO_2−δ and Ti_1−xM_xO_2−δ (M = Pd, Pt, Rh and Ru) catalysts, the structure of these materials, their catalytic properties toward different types of catalysis, structure–property relationships and mechanisms of catalytic reactions are reviewed. In these catalysts, noble metal ions are incorporated into a substrate matrix to a certain limit in a solid solution form. Lower valent noble metal-ion substitution in CeO₂ and TiO₂ creates noble metal ionic sites and oxide ion vacancies that act as adsorption sites for redox catalysis. It has been demonstrated that these new generation noble metal ionic catalysts (NMIC) have been found to be catalytically more active than conventional nanocrystalline noble metal catalysts dispersed on oxide supports.

---

Parthasarathi Bera

Parthasarathi Bera was born and raised at Moyna in the Purba Medinipur district of West Bengal, India. He obtained his B.Sc. and M.Sc. in Chemistry from Jadavpur University, Kolkata and received his Ph.D. in heterogeneous catalysis from the Indian Institute of Science, Bangalore under the supervision of Professor M. S. Hegde. He worked as a Post-Doctoral Fellow at the University of Washington, Seattle, USA with Professor Charles T. Campbell and the University of Pennsylvania, Philadelphia, USA with Professor John M. Vohs. He was a Marie Curie Fellow at the Instituto de Catálisis y Petroleoquímica, CSIC, Madrid, Spain under Professor Arturo Martínez-Arias. He joined CSIR – National Aerospace Laboratories, Bangalore in January 2010. Presently, he is a Senior Scientist in CSIR – National Aerospace Laboratories. His main research interests include heterogeneous catalysis for green energy and environment, in situ catalysis, surface science and materials science.

M. S. Hegde

M. S. Hegde received his Ph.D. from the Indian Institute of Technology, Kanpur in 1976 and joined the Solid State and Structural Chemistry Unit, Indian Institute of Science, Bangalore in 1977. In the initial stages of his career, he worked mainly on surface science employing XPS, UPS, AES and thermal desorption techniques. Then, he moved to areas such as solid state chemistry, epitaxial oxide thin films and finally to heterogeneous catalysis. During the last 15 years, he and his colleagues have developed ‘Noble Metal Ionic Catalysts’ where noble metal ions substituted in reducible oxide supports, creating oxide ion vacancies, are observed to show higher catalytic activities compared to supported noble metal nanoparticles. Presently, he is an Emeritus Professor in the Solid State and Structural Chemistry Unit, Indian Institute of Science and is engaged in training science teachers from schools, colleges and universities in the newly created Talent Development Centre of the Indian Institute of Science in its second campus at Kudapura.

---

1. Introduction

Catalysis has widespread applications from environment clean-up to manufacturing various chemicals, energy production, food processing and biological processes. The production of most of the useful commercial bulk and fine chemicals involves catalysis.^1–9 The science and technology of heterogeneous catalysis is one of the rapidly moving frontiers in the area of catalysis as well as chemical sciences. The synthesis of ammonia, sulfuric acid and nitric acid, the production of hydrogen, the control of exhaust emissions, ammoxidation, methanol synthesis and many other important reactions of the present day are associated with heterogeneous catalysis. Metals, metal clusters, alloys, oxides, sulfides, solid acids and bases have been used as catalysts for several catalytic reactions since the early days of heterogeneous catalysis.¹ Among these catalysts, supported and unsupported metals have been regarded as efficient catalysts for their usefulness in different catalytic reactions. Extensive research and investigations, such as metal–support interactions, structure–catalytic property relations, metal dispersion, redispersion, and synergistic effects, have been carried out over these catalysts for decades.

Since the early days of catalysis, it has been established that metals, in general, are active sites for the adsorption of reactant molecules and subsequent catalysis, and therefore most catalysts contain nanocrystalline metals, especially noble metals dispersed on supports like Al₂O₃ or SiO₂.^1,10,11 However, in these catalysts, noble metal atoms present on the surfaces of the metal nanoparticles in their zero valent state serve as active sites for both oxidizing and reducing molecules and only 1/4 or 1/5 of the total number of noble metals are utilized for adsorption and subsequent catalytic conversion in 4–6 nm metal particles. For example, if a 5 nm Pt particle is considered, then the total number of Pt atoms in a 5 nm cubic Pt metal crystal is 8365, taking four atoms per unit cell of Pt (fcc) metal. Again, the total number of metal atoms on the 6 surfaces of a 5 nm Pt metal cube is 1962 with two atoms per unit cell surface. Pt metal particles are anchored on a solid substrate surface. Therefore, five faces of a Pt metal crystal with 1635 surface atoms are exposed. This indicates that 1/5th of the total number of Pt atoms are available for adsorption and catalysis when reactant molecules are exposed on the catalyst surface and 4/5th of the atoms inside the metal nanocrystal are not available for adsorption or they are wasted. If all the Pt atoms are utilized, then the reaction rate would be 5 times higher, and thus catalytic activity would increase with the increase in metal dispersion. However, the atomic dispersion of Pt and other noble metals on traditional supports, such as Al₂O₃ and SiO₂, is difficult because metal atoms sinter into metal particles due to metal–metal bonding. Are there alternative ways to utilize all the Pt atoms present in the catalyst? One possible way to keep the Pt metal atoms apart is to convert Pt atoms (Pt⁰) into Pt ions (Pt²⁺) and keep them separated from O²⁻ ions. Pt²⁺ ions cannot come closer to form Pt crystals. If Pt²⁺ ions act as adsorption centers as well as Pt atoms, then in principle, all the metal atoms become available for adsorption.

Auto exhaust catalysts are mainly dispersed noble metals, such as Pt, Pd and Rh, on oxide supports like Al₂O₃ in the form of nanocrystalline metal particles.¹² A huge increase in the dispersion of noble metals has been observed with the addition of CeO₂ to Al₂O₃, leading to an increase in oxygen storage capacity (OSC) and auto exhaust catalytic activity.^12–18 Even though oxidized Pt, Pd and Rh are found on the Al₂O₃/CeO₂ surface, the reasons for higher dispersion of noble metals in the form of ions in the presence of CeO₂ have not been understood. In the early 1990s, Sayle and coworkers demonstrated the segregation of noble metal ions, such as Pd²⁺, Pt²⁺ and Rh³⁺, at the surface of CeO₂ with oxygen vacancies by computer simulation studies.¹⁹ Theoretically, noble metal ions can be stable in a CeO₂ matrix, leading to the formation of defects on the surface. Therefore, complete dispersion of noble metals can be achieved in terms of ions within reducible oxide supports such as CeO₂ and TiO₂. In this regard, entirely new generation catalysts that can be useful for multipurpose applications in different catalytic reactions have been developed in our laboratory in the last 15 years. It has been found that an increase in the dispersion of noble metals is possible through the oxidation of noble metals into their ions, which are dispersed on the reducible oxide surface, while simultaneously reducing the oxide support.^20–24 In other words, direct substitution of noble metal ions in reducible supports, such as CeO₂ and TiO₂, results in uniform solid catalysts in the form of their solid solutions. Thus, a new idea in heterogeneous catalysis has been conceptualized wherein dispersed metal ions act as catalytically active sites for heterogeneous catalysis. In the last few years, significant studies on these types of ionically dispersed catalysts have been done by other groups. Research studies on supported gold catalysts by Corma's and Gates' groups have demonstrated that gold is mainly present in its ionic form in these catalysts. They have reviewed the significance of cationic gold in different matrices with respect to several organic reactions and CO oxidation.^25,26 Flytzani-Stephanopoulos and Gates have discussed key findings of cationic noble metals as catalytic sites on solid supports.²⁷

Noble metal oxides, such as PtO, PtO₂, PdO and Rh₂O₃, are known, and therefore it should be possible to synthesize solid solutions between CeO₂ or TiO₂ and noble metal oxides. Noble metal loading for several catalytic reactions is only to the extent of 1–2 at%, and therefore substitution of only 1–2 at% of noble metal ions in CeO₂ is sufficient to develop a catalyst. The underlying principle of doping aliovalent metal ions into a CeO₂ or TiO₂ lattice is to retain their parent structures. In this sense, new-age advanced catalysts, such as Ce_1−xM_xO_2−δ (M = Pd, Pt, Rh, Cu, Ag and Au), Ce_1−x−yA_xM_yO_2−δ (A = Ti, Zr, Sn and Fe; M = Pt and Pd) and Ti_1−xM_xO_2−δ (M = Pd, Pt, Rh and Ru), that retained their parent fluorite and anatase structures have been prepared in our laboratory, employing a solution combustion method.^20–23 Thus, the synthesis of these single-phase oxides is a new concept that is the basis of metal ion catalysts. In these catalysts, the noble metal ions and corresponding oxide ion vacancies are supposed to be the active sites. Because the adsorption sites in these catalysts are noble metal ions, such as Pd²⁺, Pt²⁺, Rh³⁺ and Ru⁴⁺, we have named these materials as noble metal ionic catalysts (NMIC). In the last few years, we have been studying several catalytic exhaust reactions, such as NO reduction, CO and hydrocarbon oxidation, selective catalytic reduction of NO, three-way catalytic reactions, water gas shift (WGS) reactions, preferential oxidation of CO (CO-PROX), H₂ + O₂ recombination reactions, hydrogenation and Heck reaction, over these noble metal ionic catalysts and it has been demonstrated that these catalysts are much more active for several catalytic reactions compared to other conventional supported metal catalysts. In this review, at first, we mainly discuss the background, synthesis and characterization of these noble metal ionic catalysts (NMIC). Then, we emphasize on our detailed study related to exhaust catalysis and comparing activities with several catalysts reported in the literature. Furthermore, we write about our studies on the water gas shift reaction, the CO-PROX reaction, the H₂ + O₂ reaction, hydrogenation, Heck reaction and unique reaction mechanisms over these catalysts. Metal–support interactions and synergistic effects of the interactions that lead to the superiority of these catalysts over conventional catalysts are also discussed. We also focus on our density functional theory (DFT) studies for high oxygen storage capacity (OSC) and the catalytic properties of several CeO₂ and TiO₂-based materials. Finally, we conclude our review by focusing on the future prospects of these noble metal ionic catalysts.

2. Synthesis of noble metal ionic catalysts

2.1. Conventional preparation methods of catalysts

The dispersion of metals over suitable supports is one of the important issues for higher catalytic activities. Higher the dispersion higher is the catalytic activity. Dispersed noble metals and transition metal ions on a particular support are the active sites for adsorption of CO, NO, hydrocarbons and oxygen. In this sense, the method of catalyst preparation can influence the dispersion characteristics of metals on the supports and their catalytic behavior. The main objective of the preparation of a supported metal catalyst is to obtain the maximum dispersion of catalytically active species over the support material from its precursor at a particular concentration. The catalysts for fundamental studies are usually prepared by coprecipitation, sol–gel, deposition, impregnation, ion exchange, anchoring–grafting and spreading–wetting methods.^28,29 In general, the dispersion of active components is carried out over the oxide support in a single-step preparation, such as a coprecipitation or sol–gel process, whereas dispersion has been done over an independently prepared oxide support by impregnation, deposition, or incipient wetness methods. Some of these preparation procedures are briefly described below.

(a) Coprecipitation – this involves a reaction between a solution of two or more metal salts in the presence of a base such as hydroxide, alkaline carbonate and bicarbonate. For example, a Ni/Al₂O₃ catalyst is prepared from Ni(NO₃)₂ and AlCl₃ in the presence of NaHCO₃.²⁹

(b) Sol–gel – this process describes the transition of a system from a mostly colloidal liquid into a solid or a gel phase. This is based on the polymerization of molecular precursors such as metal alkoxides. Hydrolysis and condensation of these alkoxides lead to the formation of oxopolymers, which are then transformed into a homogeneous oxide network. Pd/SiO₂, Pt/SiO₂ and Pd/Al₂O₃ catalysts are prepared by this method.³⁰

(c) Deposition – this describes the addition of a catalytically active component to a separately produced support. There are two types of deposition.

(i) Deposition–precipitation – a precipitating agent is added to a suspension of the support in an active metal salt solution. Accordingly, the active metal species is adsorbed onto the surface of the support. A Ni/Al₂O₃ catalyst is prepared by the slow addition of NaOH to a suspension of Al₂O₃ in NiCl₂ solution. This process is used to synthesize nickel supported on silica, alumina, magnesia, titania, thoria, ceria, zinc oxide and chromium oxide.³¹ This process is also used for the preparation of Pd(OH)₂/C, Rh(OH)₃/C and Ru(OH)₃/C.

(ii) Deposition–reduction – addition of a reducing agent to a suspension of the support in a solution of the active metal salt provides supported reduced metal catalysts. Pd/C and Pt/C can be prepared by this method. In general, hydrogen, formaldehyde, hydrazine and borohydride are used as the reducing agents.²⁹

(d) Impregnation – this involves introducing a precursor solution into the pore space of a support that is filled with the same solvent as the precursor solution. Herein, the concentration gradient is the driving force. It is also called wet or diffusional impregnation. Rh/TiO₂, Rh/Al₂O₃, Pt/C and Ru/MgO are prepared by this method.²⁹

(e) Incipient wetness – this is a type of impregnation process in which a precursor solution is introduced into contact with the previously dried support. It is also called dry or pore-volume impregnation. It is an exothermic process and it depends on capillary pressure developed in the pores and the rate of the impregnation process. Ni/Al₂O₃, Pt/SiO₂, Pt/Al₂O₃, Ni/SiO₂, Ni/C, Ru/SiO₂, Rh/SiO₂ and Ru/C are prepared by this method.²⁹

(f) Ion exchange – in this method, an ion in an electrostatic interaction with the surface of a support is replaced by another ionic species. For example, zeolites, clays and silicates are cation exchangers, whereas hydrotalcites are anion exchangers. The overall charge of a zeolite is negative and is distributed over the oxygen atom; this charge is neutralized by various cations such as Na⁺ and K⁺.²⁹

(g) Microemulsion – this is a physical mixture of two immiscible liquids along with a surfactant. Reverse micelles formed can act as micro-reactors. These reactors are utilized for the synthesis of nanoparticles. Nanocrystalline CeO₂ was prepared by water-in-oil micro-emulsion methods.³² CeO₂–ZrO₂ ultrafine particles were also prepared by the micro-emulsion method.³³

(h) Mechanical milling – this is a solid-to-solid synthetic method. Mechanical milling has been used extensively to synthesize CeO₂–MO_x (M = Zr, Hf, Tb, Cu and Mn) mixed oxide catalysts.^34,35 The main feature of this method is to obtain a powder consisting of small crystallites with a size of a few nanometers along with a high concentration of lattice defects. Mechanical alloying has also been applied to prepare mixed oxides containing CeO₂ to enhance the catalysis. High-energy mechanical alloying of pure CeO₂ and ZrO₂ at room temperature results in the formation of a single-phase fluorite-structured solid solution of Ce_1−xZr_xO₂ in all the examined composition ranges.

In general, most of the abovementioned conventional preparation methods are quite involved and require long processing times. The post preparation steps include washing, drying, H₂ reduction and removal of Cl⁻ ions by heating in steam.

2.2. Solution combustion method

In the last few years, there has been a trend of novel chemical routes of synthesis that lead to ultra-fine and high surface area catalysts for heterogeneous catalysis. The solution combustion method has been found to be unique for obtaining nanocrystalline oxide materials.^36,37 In general, combustion is a chemical reaction that is accompanied by heat and light. Many exothermic, non-catalytic, solid–solid or solid–gas reactions liberate enough heat to allow them to self-propagate after being ignited. The reactions, ignited locally by an external energy source with short term service, may propagate throughout the sample at a certain rate and occur in a narrow zone, which separates the starting substances and reactions. In the literature, a variety of names have been given to redox reactions employed for the synthesis of different materials. Some of the fashionable names are furnaceless or fire synthesis, self-propagating chemical decomposition (SCD), self-propagating high temperature synthesis (SHS) and the combustion method. Merzhanov has introduced the process of SHS wherein he has employed metals and organic fuels to produce oxides, carbides and sulfides.³⁸ Patil and coworkers have developed a solution combustion method involving metal salts, mainly nitrates and organic fuels.^39,40 Metal salts are easily available, and therefore this method provides a better alternative for the synthesis of a variety of complex oxides. Thus, the solution combustion method is useful for preparing high-surface-area oxides.

The solution combustion method involves rapid heating at a particular temperature of an aqueous redox mixture containing stoichiometric amounts of the corresponding metal salts and hydrazine-based fuels and the product is formed within 5 min. Redox compounds or mixtures containing hydrazide groups has been found useful in the solution combustion synthesis of oxide materials. An ideal fuel should be water-soluble and has a low ignition temperature (<500 °C). It should be compatible with the transition metal nitrate so that the combustion can be controlled, carried out smoothly and does not lead to explosion. The merits of the solution combustion technique are as follows: (a) being a solution process, it has all the advantages of a wet chemical process such as control of stoichiometry, the doping of a desired amount of impurity ions and the formation of nano-sized particles (b) it is a low-temperature initiated process, (c) it is highly exothermic due to the redox reaction, (d) it is self-propagating, (e) it involves a transient high temperature, (f) it involves the production of a huge amount of gases, (g) it is simple, fast and economical and (h) it involves the formation of a high surface area, voluminous and homogeneous product. Using this method, a variety of oxide materials, such as α-Al₂O₃, ZnO, Y₂O₃, CeO₂, γ-Fe₂O₃, Cr₂O₃, MgFe₂O_4, MnFe₂O₄, CoFe₂O₄, PbTiO₃, PbZrO₃ and BaFe₁₂O₁₉, have been prepared. The solution combustion method not only yields nano-sized TiO₂, ZrO₂ and hexaferrites, but it also yields metastable phases such as γ-Fe₂O₃, t-ZrO₂ and anatase TiO₂. Several supported, unsupported and doped oxide catalysts have been synthesized by this method. In a recent review, González-Cortés and Imbert have discussed current approaches and future possibilities for developing advanced solid catalysts using solution combustion synthesis.⁴¹ Fig. 1 shows a typical sequence of the formation of a catalyst by the solution combustion process.

Fig. 1 Sequences of solution combustion synthesis: initial solution of precursors, flame during combustion and final product. Reprinted from ref. 21 with the permission of American Chemical Society.

The substitution of noble metal ions in CeO₂, achieved for the first time by the solution combustion method, was an accidental discovery.²⁰ When an aqueous solution containing stoichiometric amounts of Al(NO₃)₃, H₂PtCl₆ and urea (CON₂H₄) is heated rapidly at 500 °C, the solution boils, froths and burns with a flame temperature of 1500 °C, yielding nano-sized Pt metal particles dispersed on α-Al₂O₃.⁴² In contrast, Pt metal particles are not formed when a similar attempt has been made to disperse 1–2 at% Pt metal on CeO₂ by the combustion of ceric ammonium nitrate [(NH₄)₂Ce(NO₃)₆], chloroplatinic acid (H₂PtCl₆) and oxalyl dihydrazide (C₂H₆N₄O₂, ODH) heated at 350 °C. Instead of nano-sized Pt metal particles dispersed on CeO₂, 25–30 nm Ce_1−xPt_xO_2−δ nanocrystallites have been formed, wherein Pt has been observed to be present in the +2 and +4 oxidation states. A typical reaction for the preparation of a Ce_1−xPt_xO_2−x catalyst is given below:

5(1 − x)(NH₄)₂Ce(NO₃)₆ + 5xH₂PtCl₆ + 12(1 − x)C₂H₆N₄O₂ → 5Ce_1−xPt_xO_2−x + 24(1 − x)CO₂ + 44(1 − x)N₂ + (56 − 66x)H₂O + 10xHCl.

Similarly, Pd, Rh, Ag and Au-substituted Ce_1−xM_xO_2−δ catalysts have been prepared by this method using PdCl₂, RhCl₃, AgNO₃ and HAuCl₄ as respective metal precursors.^20,43–46 Ce_1−x−yTi_xM_yO_2−δ (M = Pt and Pd), Ce_1−x−ySn_xPd_yO_2−δ, Ce_1−x−yFe_xPd_yO_2−δ, Ce_1−x−yZr_xPd_yO_2−δ and Ce_1−x−yHf_xPd_yO_2−δ catalysts have also been prepared with a combustion mixture of (NH₄)₂Ce(NO₃)₆ with TiO(NO₃)₂ or SnC₂O₄ or Fe(NO₃)₃ or ZrO(NO₃)₃ or Hf(NO₃)₄, H₂PtCl₆, PdCl₂ and C₂H₅NO₂.^47–52

The combustion of TiO(NO₃)₂, noble metal salts and glycine (C₂H₅NO₂) fuel has been employed for the preparation of noble metal ion-doped TiO₂ catalysts in the form of a Ti_1−xM_xO_2−x (M = Pd, Pt, Rh and Rh) solid solution.^53–56 A typical reaction for the preparation of Ti_1−xPd_xO_2−x catalyst is given below:

9(1 − x)TiO(NO₃)₂ + 10(1 − x)C₂H₅NO₂ + 9xPdCl₂ → 9Ti_1−xPd_xO_2−x + 20(1 − x)CO₂ + 14(1 − x)N₂ + (25 − 34x)H₂O + 18HCl.

2.3. Hydrothermal method

The term ‘hydrothermal’ is of geological origin. Many materials, including zeolites, were formed in the Earth's crust under high temperature and high pressure. The hydrothermal method has been introduced by Barrer to synthesize zeolites.⁵⁷ The hydrothermal method utilizes water under pressure and at temperatures above its boiling point as a means of speeding up the reaction between solids. Under hydrothermal conditions, the reaction occurs at a relatively low temperature (∼150–200 °C); otherwise, a much higher temperature is required. Because the hydrothermal reaction must be carried out in a closed vessel, the temperature–pressure relationship is very important. Among all these preparation methods, hydrothermal synthesis always offers an extra merit in preparing highly crystallized nanomaterials with controlled shape, size and orientation. In this method, metal salts are dissolved in water and mixed with structure-directing reagents, generally, amines that form a gel. Gel is used in a vessel called a hydrothermal bomb, which is a thick-walled steel vessel with a hermetic seal. Inside the vessel, containers made of Teflon are filled with the gel to the extent of 75%. The pressure vessel is closed and heated in an oven to 200 °C for 24–48 h. Nanostructural CeO₂ and TiO₂ were prepared via an alkaline hydrothermal route.^58,59 The hydrothermal method is one of the most popular techniques for preparing metal-doped CeO₂ materials such as Ce_1−xM_xO_2−δ (M = Ca and Sm).⁶⁰ Several Ce_1−xM_xO_2−δ (M = Zr, Ti, Pr, Y, Fe, Cr and Ru) catalysts have been prepared using the hydrothermal method in our laboratory.^61–63

2.4. Sonochemical method

Ultrasound radiation is used mostly for the diagnosis of diseases, whereas focused ultrasound radiation has been used to burn cancer cells. Ultrasonic radiation is utilized to break chemical bonds to produce nanoparticles. Gedanken has pioneered the sonochemical method to prepare a number of oxide nanomaterials such as CeO₂, ZnO and RVO₄ (R = La, Ce, Nd, Sm, Eu and Gd).^64–66 Ceria nanoparticles embedded in poly methyl methacrylate (PMMA) have been synthesized using the sonochemical method.⁶⁷ Pt-supported TiO₂@C core–shell composites have been synthesized by a sonochemical route for methanol electrooxidation applications.⁶⁸ This method is especially suitable for making Fe and noble metal ion-substituted CeO₂ and TiO₂ catalysts. In our laboratory, Ce_1−xPt_xO_2−δ, Ce_1−xFe_xO_2−δ and Ce_1−x−yFe_xPd_yO_2−δ catalysts have been synthesized by the sonochemical method.^69,70

Noble as well as transition metal ions have been substituted into CeO₂ and TiO₂ matrices by solution combustion, hydrothermal and sonochemical methods successfully. Table 1 summarizes the metal ion-substituted CeO₂ and TiO₂ materials synthesized in our laboratory.^{20,43–56,61–63,69–75}

Table 1 Various metal ion-substituted CeO₂ and TiO₂ catalysts prepared in our laboratory. Reprinted from ref. 24 with the permission of Elsevier B. V.

Metal ions	Support oxides
	CeO2	Ce1−xTixO2	Ce1−xSnxO2	Ce1−xFexO2−δ	Ce1−xZrxO2	TiO2
Pd^2+	Ce1−xPdxO2−δ (x = 0.01, 0.02) (20, 43)	Ce0.75−xTi0.25PdxO2−δ (x = 0.01, 0.02) (48)	Ce0.78Sn0.2Pd0.02O2−δ (49, 50)	Ce1−x−yFexPdyO2−δ (x = 0.1–0.45, y = 0.02) (51, 70)	Ce1−x−yZrxPdyO2−δ (x = 0.25, y = 0.02) (52)	Ti1−xPdxO2−δ (x = 0.01, 0.03) (54, 55)
Rh^3+	Ce1−xRhxO2−δ (x = 0.005, 0.01) (44)					Ti1−xRhxO2−δ (x = 0.01) (54)
Pt^2+, Pt^4+	Ce1−xPtxO2−δ (x = 0.01, 0.02) (20, 69, 71, 72)	Ce0.85−xTi0.15PtxO2−δ (x = 0.01, 0.02) (47)				Ti1−xPtxO2−δ (x = 0.01) (54)
Pt^2+, Rh^3+	Ce1−xRhx/2Ptx/2O2−δ (x = 0.01, 0.02) (73)
Ru^4+	Ce1−xRuxO2−δ (x = 0.05, 0.1) (63)					Ti1−xRuxO2−δ (x = 0.01) (54)
Cu^2+	Ce1−xCuxO2−δ (x = 0.01–0.1) (74, 75)
Ag^+	Ce1−xAgxO2−δ (x = 0.01) (45)
Au^3+	Ce1−xAuxO2−δ (x = 0.01) (46)
Cr^4+, Cr^6+	Ce1−xCrxO2+δ (x = 0.01–0.2) (62)

---

3. Structure of Ce_1−xM_xO_2−δ and Ti_1−xM_xO_2−δ-based catalysts

Several physical characterization techniques, such as X-ray diffraction (XRD), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and X-ray absorption fine structure spectroscopy (XAFS), have been employed to study the crystal structure, morphology, electronic structure, local structure and oxide ion vacancies of Ce_1−xM_xO_2−δ and Ti_1−xM_xO_2−δ-based catalysts. The main findings of these studies are discussed in the following subsections.

3.1. XRD studies

High resolution XRD data (Rigaku-2000, PANalytical X'Pert PRO) demonstrate that CeO₂ crystallizes in the fluorite structure in all CeO₂-based catalysts, where Ce⁴⁺ ions are cubic close-packed with all tetrahedral sites occupied by oxygen. There are no impurity peaks related to any platinum oxides in the Ce_0.99Pt_0.01O_2−δ catalyst. However, a small, broad hump at 2θ = 39° due to Pt(111) is observed in the diffraction pattern, indicating the presence of trace amounts of Pt in the catalyst. Comparing the intensity ratio of Pt(111) in Ce_0.99Pt_0.01O_2−δ with that in a 1% Pt + CeO₂ physical mixture, it has been evaluated that at least 92 at% of the platinum used in the preparation of Ce_0.99Pt_0.01O_2−δ is incorporated into the CeO₂ lattice.^71,72 The total oxygen content in Ce_0.99Pt_0.01O_2−δ is 1.883 and that in pure CeO₂ is 1.934 and a decrease in the lattice parameter is observed in Ce_0.99Pt_0.01O_2−δ compared to pure CeO₂. Similarly, diffraction lines are indexed to the fluorite structure in Ce_0.99Pd_0.01O_2−δ and Ce_0.98Pd_0.02O_2−δ, and lines corresponding to Pd or PdO are not observed in their XRD patterns. Even a slow scan in the Pd(111) region (2θ = 40° ± 5°) does not show any indication of a Pd metal peak.⁴³ On the other hand, a 1% Pd/CeO₂ catalyst prepared by the impregnation method shows diffraction lines due to Pd metal. The lattice parameter decreases from 5.4113(2) Å in pure CeO₂ to 5.4107(3) Å in Ce_0.98Pd_0.02O_2−δ, confirming the substitution of Pd²⁺ ions (0.84 Å) for Ce⁴⁺ ions (0.99 Å) in CeO₂. In a Ce_0.99Rh_0.01O_2−δ catalyst, diffraction lines corresponding to Rh metal, Rh₂O₃ and RhO₂ are not detected. The total oxygen content in a Ce_0.99Rh_0.01O_2−δ catalyst is 1.87, whereas it is 1.934 for pure CeO₂.⁴⁴ Therefore, a decrease in the oxygen content in the catalyst compared with pure CeO₂ signifies the creation of an oxide ion vacancy. Thus, detailed XRD studies demonstrate the formation of Ce_1−xM_xO_2−δ (M = Pd, Rh, Pt, Ru, Cu, Ag and Au), Ce_1−x−yTi_xM_yO_2−δ (M = Pd and Pt), Ce_1−x−ySn_xPd_yO_2−δ, Ce_1−x−yFe_xPd_yO_2−δ and Ce_1−x−yZr_xPd_yO_2−δ solid solution phases.^{43–52,63,69–75} Similarly, solution combustion-synthesized Ti_1−xM_xO_2−δ (M = Pd, Rh, Ru and Pt) catalysts crystallize in the anatase structure.^53–56 All the catalysts show broad lines in their XRD patterns. A rutile contribution has been observed to the extent of 10% only for Ti_1−xRu_xO_2−δ. No diffraction lines due to noble metals or their oxides are observed in Ti_1−xM_xO_2−δ. Rietveld refinement of these catalysts reveals a slight increase in cell volume for Ti_1−xPt_xO_2−δ due to a slight increase in the ionic radius of Pt; however, there is almost no change in cell volume or cell parameters for Pd²⁺, Rh³⁺ and Ru⁴⁺ ion-doped TiO₂. The particle sizes measured from the Scherrer formula are in the range of 10–15 nm for the four catalysts. Typical Rietveld-refined XRD patterns of Ce_0.99Pt_0.01O_2−δ and Ti_0.98Pd_0.02O_2−δ are shown in Fig. 2.

Fig. 2 Rietveld refined XRD patterns of (a) Ce_0.99Pt_0.01O_2−δ and (b) Ti_0.99Pd_0.01O_2−δ. Reprinted from ref. 71 with the permission of American Chemical Society and from ref. 142 with the permission of Elsevier B. V.

XRD patterns of hydrothermally prepared Ce_1−xM_xO_2−δ (M = Zr, Ti, Pr, Y, Fe and Cr) catalysts are indexed to the fluorite structure and there are no traces of the respective oxides in their patterns.^61,62 Similarly, diffraction lines associated with RuO₂ and Ru metal are not observed in the XRD patterns of Ce_1−xRu_xO_2−δ (0 ≤ x ≤ 0.1) prepared by the hydrothermal method and they show the fluorite structure.⁶³ Ru and RuO₂ impurity peaks are observed in the XRD pattern of Ce_1−xRu_xO_2−δ prepared with x = 0.15. Thus, only up to 10% Ru⁴⁺ ions can be substituted for Ce⁴⁺ in CeO₂ by this method.

Diffraction lines related to Pt and Pd metals or PtO₂ and PdO impurities are not observed in the XRD patterns of Ce_1−xM_xO_2−δ (M = Pt and Pd; x = 0.02, 0.05 and 0.10) synthesized by the sonochemical method, whereas all the peaks are indexed to the cubic fluorite lattice, indicating the substitution of Pt²⁺ and Pd²⁺ ions into the CeO₂ lattice to a certain extent.⁷⁰

Li and coworkers have demonstrated the absence of Pd metal up to 2.5 wt% metal loading in the XRD pattern of mesoporous Pd/CeO₂.⁷⁶ Pt metal is not found in combustion-synthesized 1 at% Pt/CeO₂, as shown by Tang et al.⁷⁷ Diffraction peaks related to Pd metal are not observed in sol–gel synthesized 1 at% Pd/CeO₂ in the study done by Wang and coworkers.⁷⁸ Pt metal and its oxides are absent in the XRD pattern of solution combustion-synthesized 2 at% Pt/CeO₂, as shown by Bisht and coauthors.⁷⁹ Pd metal is not observed in the XRD pattern of a 1 wt% Pd/CeO₂–SiO₂ catalyst.⁸⁰ Misch et al. have demonstrated the formation of a single phase Ce_1−xPd_xO_2−δ (x = 0.025, 0.05, 0.075, and 0.1) solid solution with a fluorite structure.⁸¹

3.2. TEM studies

High resolution TEM (FEI Technai 20) images of noble metal ion-substituted CeO₂ and TiO₂ catalysts have been obtained to understand their morphology and lattice fringes. CeO₂-based catalysts show the cubic morphology of CeO₂ crystallites with average sizes of 5–30 nm. Noble metal particles are not observed on the surface of CeO₂ crystallites. A HRTEM image and the electron diffraction (ED) pattern of solution combustion synthesized Ce_0.73Ti_0.25Pd_0.02O_2−δ is displayed in the top panel of Fig. 3. The fringe spacing at ∼3.12 Å corresponds to the (111) layers of Ce_0.75Ti_0.25O₂. The ED pattern clearly demonstrates the crystalline nature of the catalyst and it is indexed to the fluorite structure. Neither a lattice fringe in the HRTEM image nor a ring in the ED pattern due to Pd metal particles at 2.25 Å is observed. In contrast, the Pd(111) ring is visible as indicated by an arrow in the ED pattern of 2% Pd/Ce_0.75Ti_0.25O₂ prepared by the impregnation method, shown in Fig. 3C of top panel. The absence of a diffraction ring due to Pd metal in the solution combustion-synthesized material shows substitution of Pd in the lattice, leading to a Ce_0.73Ti_0.25Pd_0.02O_2−δ solid solution.⁴⁸ The average particle size of CeO₂ crystallites obtained from the low magnification image is ∼20 nm. Similarly, lattice fringes that are 3.2 Å apart, corresponding to the (111) plane of Ce_0.85Ti_0.15O₂ have been observed in a high resolution image of Pt ion-substituted Ce_0.85Ti_0.15O₂ and fringes related to Pt metal particles at 2.30 Å are absent. Energy dispersive X-ray spectroscopy (EDXS) confirms the presence of 1% Pt in the catalyst and Pt ion substitution in Ce_0.85Ti_0.15O₂.⁴⁷ Lattice fringes in the HRTEM image of Ce_0.89Fe_0.1Pd_0.01O_2−δ are 3.1 Å apart, corresponding to its (111) plane.⁵¹ On the other hand, a large number of nano-sized fine Pt metal particles can be dispersed on α-Al₂O₃ by the solution combustion method.^42,71

Fig. 3 Top panel: HRTEM image of Ce_0.73Ti_0.25Pd_0.02O_2−δ (A) and ED patterns of Ce_0.73Ti_0.25Pd_0.02O_2−δ (B) and impregnated 2 at% Pd/Ce_0.75Ti_0.25O₂ (C); bottom panel: (a) bright field TEM image of Ti_0.97Pd_0.03O_2−δ and its ED pattern (inset), (b) HRTEM image of Ti_0.97Pd_0.03O_2−δ and (c) EDXS of Ti_0.97Pd_0.03O_2−δ from the image. Reprinted from ref. 48 and 55 with the permission of American Chemical Society.

An HRTEM image and the corresponding ED pattern of a Ti_0.97Pd_0.03O_2−δ catalyst are shown in the bottom panel of Fig. 3. Lattice fringes in the Ti_0.97Pd_0.03O_2−δ catalyst, shown in the bottom panel of Fig. 3 (Fig. 3b) correspond to the (101) planes of TiO₂. Lattice fringes due to PdO or Pd metals are not detected in the high resolution lattice images from the Ti_0.97Pd_0.03O_2−δ catalyst. The EDX spectrum of this image shows the presence of Pd. The ED pattern is indexed to polycrystalline TiO₂ in the anatase structure in both catalysts and there is no signature of Pd metal or PdO phases in their ED patterns.⁵⁵ Therefore, the absence of Pd or PdO in Ti_0.97Pd_0.03O_2−δ in the TEM image and the presence of Pd in the EDX spectrum confirm substitution of the Pd²⁺ ion in the TiO₂ lattice. In an another study, the Pd metal diffraction line is observed in the ED pattern of 1% Pd/TiO₂ prepared by impregnation.⁵⁴ The particle sizes measured from the image are in the range of 8–15 nm, which agrees well with the XRD studies. Other noble metal-doped TiO₂ catalysts show similar type of characteristics.

An HRTEM image of Ce_0.9Ru_0.1O_2−δ synthesized by the hydrothermal method shows the absence of isolated RuO₂ and Ru metal and the average crystallite size is 10 nm. The ring-type ED pattern is indexed to the fluorite structure.⁶³

3.3. XPS studies

Oxidation states of Pt, Pd, Rh, Ru, Cu, Ag, Au, Ce and Ti present in the CeO₂ and TiO₂-based catalysts studied herein can be obtained from XPS (ESCA-3 Mark II, Thermo Fisher Scientific). Typical Pt4f and Rh3d core level spectra in reference compounds and CeO₂-based catalysts are shown in Fig. 4. In Ce_0.99Pt_0.01O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ catalysts, Pt is present in both the +2 and +4 oxidation states, along with a small amount of Pt metal.^20,71,72,47 XPS of the Pt4f core level peaks of Pt metal, PtO₂, Ce_0.99Pt_0.01O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ is displayed in the left panel of Fig. 4. Rh is in the +3 oxidation state in the Ce_1−xRh_xO_2−δ catalyst as observed in the Rh3d core level spectra.⁴⁴ In the right panel of Fig. 4, Rh3d core level spectra in Rh, Rh₂O₃, RhCl₃ and Ce_1−xRh_xO_2−δ are displayed. Similarly, a core-level Pd3d_5/2 peak is observed at 337.6 eV in Ce_0.98Pd_0.02O_2−δ, which is 2.4 and 0.9 eV higher than in Pd metal and PdO.⁴³ The binding energies of Pd3d peaks in Ce_0.98Pd_0.02O_2−δ are close to the Pd3d peaks of PdCl₂ and Pd(NO₃)₂, clearly indicating that Pd²⁺ ions are much more ionic in the CeO₂ matrix than in PdO. Pd is observed to be in the high binding energy region in Ce_0.98Pd_0.02O_2−δ. Pd is also observed to be in the +2 state in Ce_0.73Ti_0.25Pd_0.02O_2−δ, Ce_0.78Sn_0.2Pd_0.02O_2−δ, Ce_0.89Fe_0.1Pd_0.01O_2−δ and Ce_0.73Zr_0.25Pd_0.02O_2−δ.^48–52 Therefore, XPS of Ce_1−xPd_xO_2−δ-based catalysts indicates that Pd is present in the highly ionic Pd²⁺ state. Similarly, XPS results demonstrate that Cu in Ce_1−xCu_xO_2−δ, Ag in Ce_1−xAg_xO_2−δ and Au in Ce_1−xAu_xO_2−δ are fully dispersed in the +2, +1 and +3 oxidation states, respectively.^45,46,74,75 XP spectra of Ce3d with characteristic satellites in all catalysts show that Ce is mainly present in the +4 oxidation state. Ru in Ce_1−xRu_xO_2−δ prepared by the hydrothermal method, Pt and Pd in Ce_1−xM_xO_2−δ (M = Pt and Pd) and Pd in doped Ce_1−xFe_xO_2−δ prepared by sonochemical method are in oxidized states.^63,69,70

Fig. 4 Left panel: XPS of Pt4f core levels of (a) Pt metal, (b) (NH₃)₄Pt(NO₃)₂, (c) PtO₂, (d) Ce_0.99Pt_0.01O_2−δ and (e) Ce_0.84Ti_0.15Pt_0.01O_2−δ; right panel: XPS of Rh3d core levels of (a) Rh metal, (b) Rh₂O₃, (c) Ce_0.99Rh_0.01O_2−δ and (d) Ce_0.98Rh_0.02O_2−δ. Reprinted from ref. 24 with the permission of Elsevier B. V. and from ref. 44 with the permission of American Chemical Society.

Pd metal and highly ionic Pd²⁺ species are observed in Pd/CeO₂ prepared by wet impregnation method.⁷⁶ Pd3d_5/2,3/2 core level peaks observed at 336.5 and 338.0 eV in Pd/CeO₂ catalysts prepared by the coprecipitation method correspond to PdO_x and highly ionic Pd²⁺ species of the Ce_1−xPd_xO_2−δ solid solution.⁸² Li and coworkers have demonstrated a Pd3d_5/2 core level peak at 336.7 eV in a coprecipitated PdCeO_x solid solution.⁸³ XPS studies have shown that Pt, Pd and Ru are in their ionic states in M/Ce_0.72Zr_0.18Pr_0.1O₂ (M = Pt, Pd and Ru) catalysts.⁸⁴ There are several recent reports where Pt and Pd are present in their ionic states in a CeO₂ matrix.^85–89 Pt4f_7/2 core level peaks are observed at 71.3, 72.3 and 73.9 eV in 3.7 wt% Pt–Ce(La)O_x catalyst synthesized by the coprecipitation method.⁹⁰ A combustion-synthesized Pt/CeO₂ catalyst contains Pt⁰, Pt²⁺ and Pt⁴⁺ species as demonstrated by XPS.⁷⁷ Pt⁰, Pt²⁺ and Pt⁴⁺ species are found in Pt/CeO₂ catalysts prepared by the deposition–reduction method.⁹¹ Ru is in the +4 oxidation state in Ru/CeO₂ catalysts.^92,93 Au⁰, Au⁺ and Au³⁺ species are present in 4.7 wt% Au–Ce(La)O_x and 1% Au/CeO₂ prepared by coprecipitation and photoreduction methods.^90,94 Au is in Au⁰, Au⁺ and Au³⁺ oxidation states in 2.8 wt% Au/CeO₂ obtained from the deposition–precipitation method.⁹⁵ Ag3d_5/2 core level peaks at 368.3 and 367.8 eV observed in an Ag/CeO₂ catalyst prepared by the impregnation method are attributed to Ag⁰ and Ag⁺ species.⁹⁶

XPS studies of Ti_0.99M_0.01O_1.99 (M = Pd, Pt, Rh and Ru) catalysts also show the ionic nature of substituted noble metals in a TiO₂ matrix.⁵⁴ Core level XPS of Pd3d in Ti_0.99Pd_0.01O_1.99, Pt4f in Ti_0.99Pt_0.01O_1.99, Rh3d in Ti_0.99Rh_0.01O_1.99 and Ru3p in Ti_0.99Ru_0.01O_1.99 are presented in Fig. 5. A Pd3d_5/2 peak at 337.2 eV is attributed to Pd²⁺ ion substitution in the TiO₂ support. Pd3d peaks in Ti_0.99Pd_0.01O_1.99 are observed at higher binding energies compared to PdO, suggesting that Pd²⁺ ions in TiO₂ are in a more ionic state than Pd²⁺ in PdO. Pt4f in Ti_0.99Pt_0.01O_1.99 shows a broad spectrum of mixed valence states of Pt²⁺ and Pt⁴⁺. Rh3d_5/2 at 310.0 eV indicates clearly that Rh is in the +3 oxidation state in Ti_0.99Pt_0.01O_1.99. A Ru3p_3/2 peak observed at 463 eV indicates that Ru is in the +4 state in Ti_0.99Ru_0.01O_1.99 and the peak position is similar to that in RuO₂. In all the cases, core level Ti2p_3/2,1/2 peaks are observed at 459.0 and 464.0 eV, indicating that Ti is in the +4 oxidation state.

Fig. 5 XPS of (a) Pd3d in Ti_0.99Pd_0.01O_2−δ, (b) Pt4f in Ti_0.99Pt_0.01O_2−δ, (c) Rh3d in Ti_0.99Rh_0.01O_2−δ and (d) Ru3p in Ti_0.99Ru_0.01O₂. Reprinted from ref. 54 with the permission of Elsevier B. V.

The stabilization of noble metal ions in CeO₂ or TiO₂ can be substantiated from the relative positions of metal valence levels with respect to valence levels of CeO₂ and TiO₂. Typical valence band XPS of CeO₂, Ce_0.98Pd_0.02O_1.98, Pd metal, PdO, TiO₂ and Ti_0.97Pd_0.03O_2−δ are shown in Fig. 6. The valence band of CeO₂ consists of the O2p band spread over ∼3–9 eV that is shown in the left panel of Fig. 6. An empty Ce4f level is located at ∼2 eV below the Fermi level (E_F), where binding energy is referenced to zero. Pt, Pd and Rh metals have a high electron density at the E_F and the valence bands of these metals even extend up to 6 eV below the E_F. When these metals are oxidized, Mⁿ⁺d bands shift to higher binding energies with the effect that the Mⁿ⁺d bands are located at about 2.5–3.5 eV below E_F as in PdO in the figure. In Ce_1−xM_xO_2−δ, the Mⁿ⁺d band lies between the Ce4f and O2p bands; therefore, Ce in the compound remains mostly in the +4 state. When the metal ion in Ce_1−xM_xO_2−δ is reduced to the metal in the lattice, the metal valence band is shifted towards the E_F that is above the empty Ce4f band. Therefore, electron transfer from the metal to the Ce⁴⁺ ion (M⁰ + 2Ce⁴⁺4f⁰ → M²⁺ + 2Ce³⁺4f¹) becomes facile and the noble metals remain ionic in CeO₂.²¹ For example, the difference in the valence band XPS of CeO₂ and Ce_0.98Pd_0.02O_1.98 shows a small density of states at ∼3.2 eV from E_F. This density of states corresponds to the Pd²⁺4d⁸ occupied state that is higher than that of the Ce4f level.⁵¹ Similarly, Pt²⁺ and Pd²⁺ ions are stabilized in TiO₂, because the Pt²⁺5d and Pd²⁺4d bands are situated between Ti³⁺3d (0.9 eV) and O2p bands.^21,53,97 The right panel of Fig. 6 shows the band positions of TiO₂ and Ti_0.97Pd_0.03O_2−δ.

Fig. 6 Left panel: valence band XPS of CeO₂, Ce_0.9Pd_0.1O_2−δ, PdO, Pd metal and the difference spectrum of Ce_0.9Pd_0.1O_2−δ and CeO₂ valence bands showing the Pd²⁺4d band position. Reprinted from ref. 23 with the permission of Springer; right panel: valence band XPS of TiO₂, Ti_0.97Pd_0.03O_2−δ and the difference spectrum of Ti_0.97Pd_0.03O_2−δ and TiO₂ valence bands showing the Pd²⁺4d band position.

3.4. XAFS studies

Substitution of metal ions into the CeO₂ lattice has been confirmed by XAFS studies (SPring-8, Japan). XAFS analysis shows unique Ce–Pd, Ce–Rh and Ce–Pt distances of 3.31, 3.16 and 3.28 Å, respectively, whereas the Ce–Ce distance is observed to be 3.84 Å, demonstrating the substitution of Pd²⁺, Rh³⁺ and Pt²⁺ ions into the CeO₂ lattice.^43,44,72 Similar distances related to Ce–Pt and Ce–Pd interactions are also observed in Ce_1−x−yTi_xPt_yO_2−δ, Ce_1−x−yTi_xPd_yO_2−δ and Ce_1−x−ySn_xPd_yO_2−δ catalysts.^50,98 It is important to mention that these unique interactions are found in neither CeO₂ nor noble metals or their oxides such as PdO, Rh₂O₃ and PtO₂. Typical XAFS of Pt and Rh ion-substituted CeO₂ catalysts along with Pt metal, PtO₂, Rh metal and Rh₂O₃ are shown in Fig. 7. Lower coordination numbers around metal ions compared to the Ce ion indicate that the oxide ion vacancy is due to lower valent metal ion substitution.

Fig. 7 Left panel: Fourier transformed XAFS of Pt metal, PtO₂, Ce_0.99Pt_0.01O_2−δ and Ce_0.98Pt_0.02O_2−δ, reprinted from ref. 23 with the permission of Springer; right panel: Rh metal, Rh₂O₃, Ce_0.99Rh_0.01O_2−δ and Ce_0.98Rh_0.02O_2−δ.

3.5. DRIFTS studies

The DRIFTS technique has been used to understand the interaction of CO and NO with noble metal ionic catalysts, such as Ce_0.98Pd_0.02O_2−δ and Ce_0.73Sn_0.25Pd_0.02O_2−δ, and the nature of Pd species present on their surface through CO and NO adsorption. Adsorption studies are also carried out with Pd/Al₂O₃, where Pd is in the metallic state. It has been demonstrated that CO and NO are molecularly adsorbed on Pd/Al₂O₃, whereas NO is dissociatively adsorbed on Ce_0.98Pd_0.02O_2−δ and Ce_0.73Sn_0.25Pd_0.02O_2−δ catalysts.⁹⁹ In contrast, a linear Pd^δ+–CO adsorption band along with a bridging CO band has been observed over CeO₂-based catalysts when CO is adsorbed first, whereas a band related to Pd²⁺–CO interaction is noticed on preadsorbed NO catalysts. As CO is a reducing molecule, partial surface reduction of Pd²⁺ occurs only when CO is adsorbed on the Ce_0.98Pd_0.02O_2−δ and Ce_0.73Sn_0.25Pd_0.02O_2−δ catalysts. However, Pd²⁺ remains intact after CO adsorption in the presence of an oxidizing molecule such as NO indicating the presence of Pd²⁺ species on the surface. Again, a Pd²⁺–NO band is observed only when NO is adsorbed on their surfaces. Therefore, CO and NO adsorption studies demonstrate that Pd is in a +2 oxidation state in Ce_0.98Pd_0.02O_2−δ and Ce_0.73Sn_0.25Pd_0.02O_2−δ catalysts. Typical DRIFT spectra of (a) NO adsorption and (b) followed by CO adsorption over Ce_0.98Pd_0.02O_2−δ are shown in Fig. 8. Similarly, a Pt²⁺–CO band has been observed when CO is passed over the Ce_0.99Pt_0.01O_2−δ catalyst.⁷¹ In a recent study, a Cu²⁺–CO band was observed to be present in Ce_0.93Cu_0.07O_2−δ and Ce_1−x−yM_xCu_yO_2−δ (M = Zr, Hf and Th) catalysts with CO exposure.¹⁰⁰

Fig. 8 DRIFT spectra of (a) NO adsorption and (b) followed by CO adsorption over Ce_0.98Pd_0.02O_2−δ. Reprinted from ref. 99 with the permission of Elsevier B. V.

Based on extensive characterization studies employing XRD, TEM, XPS, XAFS and DRIFTS, it has been demonstrated that noble metals are ionically dispersed over CeO₂-based catalysts and noble metal ions are incorporated into the CeO₂ matrix in the form of a Ce_1−xM_xO_2−δ solid solution phase in combustion-synthesized M/CeO₂ catalysts. In addition to this, oxide ion vacancies are created due to aliovalent ionic substitutions for Ce⁴⁺; there seem to be additional oxide ion vacancies to the extent of 3.5% in nano CeO₂. Similarly, a Ti_1−xM_xO_2−δ type of solid solution phase has been formed in M/TiO₂ catalysts by substituting noble metal ions into a TiO₂ matrix.

4. Exhaust catalysis

Energy for modern civilization comes from the burning of fossil fuels such as coal, natural gas and oil. Most fossil fuels are constituted of carbon, nitrogen, hydrogen and to some extent sulfur-containing compounds. The energy stored in the fossil fuels, biomass, industrial and domestic waste is released mostly by complete as well as partial combustion of these constituents. Partial combustion leads to emission of nitrogen oxides (NO_x), carbon monoxide (CO), sulfur oxides (SO_x), NH₃ and unburned hydrocarbons (HC) into the atmosphere, which is the main reason for environmental pollution. These pollutants are emitted every day from the exhausts of mobile and stationary sources and have harmful effects on the Earth's atmosphere, the ecological system and also on human health.^22,101,102 Emissions from automobiles and aeroplanes are the mobile sources, whereas industrial processes, power plants, combustion of waste and biomass and domestic burning are the stationary sources. In the early 1950s, air pollution and automobiles were first correlated by Californian scientists who established that a huge amount of traffic was to blame for the smoggy skies over the Los Angeles area.^103–107 Over the years, international concern about environmental pollution caused by polluting sources has been increasing as it becomes a global environmental issue. Consequently, several developed and developing countries have enacted legislation for the control of pollutant gases in the exhaust stream of automobiles. Each of these countries has set particular emission standards by law for different types of vehicles according to their own country's situation. Emission standards set specific limits to the amount of pollutants that can be released into the environment. It is a limit that sets threshold amounts above which different types of emission control technology must be used. Standards generally regulate the emissions of NO_x, CO, HCs, SO_x and particulate matter (PM) or soot. Therefore, controlling these pollutants to achieve a clean environment is an important issue in the present day.

There are several technologies for controlling pollutants in exhaust emission. Among the available technologies, one of the best ways of controlling exhaust emission is to reduce or convert these pollutant gases by using a catalyst at the source itself. Therefore, the conversion of environmentally unacceptable gases, such as NO_x, CO and HC to N₂, CO₂ and H₂O using catalysts, is a challenging task, and therefore this area of heterogeneous catalysis has been developing immensely in the last four decades. Within this context, environmental catalysis related to auto exhaust has received much more attention as exhausts from automobiles have significantly contributed to global environmental pollution. Extensive studies on the scientific research and technological development in this area have been carried out in the last few years. A number of reviews, articles and books regarding various aspects of exhaust catalysis have been published to update the status of catalytic science and technological development in this area from time to time.^108–115

4.1. Reactions involved in exhaust catalysis

In the last few years, enormous efforts have been made to advance scientific and technological development for the controlling of exhaust emission. Catalytic converters have been found to be the best way to control exhaust emission from automobiles among all the types of technologies developed so far.¹¹⁶ A catalytic converter is a device that usually reduces the toxicity of exhaust emission. A catalyst housed inside a catalytic converter performs chemical reactions by which combustion by-products of pollutants are converted to less polluting or benign substances. Since 1981, three-way catalytic converters have been used in vehicle emission control systems of roadgoing vehicles in North America and many other countries. The catalyst used in a three-way catalytic converter is called a three-way catalyst (TWC). A TWC has three simultaneous tasks, namely (i) reduction of NO_x to nitrogen and oxygen, (ii) oxidation of CO to CO₂ and (iii) oxidation of unburned hydrocarbons to CO₂ and H₂O, provided that the air to fuel (A/F) ratio in the exhaust stream is constantly kept at the stoichiometric point (14.7), where concentrations of oxidizing and reducing gases are equal.

The essential requirement for a catalyst employed in present-day catalytic converters for a gasoline engine is to carry out the conversion of NO_x, CO and HCs present in the auto exhaust gases to inert and harmless N₂, CO₂ and H₂O. The overall catalytic reactions that are important for controlling auto exhaust emissions are given by the following stoichiometric equations:

(a) Reduction reactions

2NO + 2CO → N₂ + 2CO₂

2NO + 2H₂ → N₂ + 2H₂O

(b) Oxidation reactions

2CO + O₂ → 2CO₂

2H₂ + O₂ → 2H₂O

(c) Three-way catalytic reaction

(d) Decomposition reaction

2NO → N₂ + O₂

(e) Selective catalytic reduction

4NO + 4NH₃ + O₂ → 4N₂ + 6H₂O

6NO₂ + 8NH₃ → 7N₂ + 12H₂O

(f) Secondary reactions with NO

2NO + CO → N₂O + CO₂

2NO + H₂ → N₂O + H₂O

2NO + 5H₂ → 2NH₃ + 2H₂O

4NH₃ + 5O₂ → 4NO + 6H₂O

4NH₃ + 4NO + 3O₂ → 4N₂O + 6H₂O

2NH₃ + 2O₂ → N₂O + 3H₂O

2NO + O₂ → 2NO₂

(g) Secondary reactions with SO₂

2SO₂ + O₂ → 2SO₃

SO₂ + 3H₂ → H₂S + 2H₂O.

All the abovementioned reactions required some heat or temperature on the catalyst surface for the reaction to occur. In the case of auto exhaust catalysis, when an automobile first starts, both the engine and the catalyst are cold. After start up, the heat of combustion is transferred from the engine and the exhaust piping begins to heat up. Finally, a temperature is reached within the catalyst that initiates the catalytic reactions. This light-off temperature depends on the chemistry of the catalyst.

4.2. Importance of OSC in auto exhaust catalysis and role of CeO₂

In the early 1980s, the discovery of a catalytic converter with a three way catalyst (TWC) was considered as a major breakthrough in the development of devices for the control of auto exhaust emissions. Pollutant gases, such as CO, HC and NO_x, are converted into benign CO₂, H₂O and N₂ simultaneously and efficiently over TWCs with a constant A/F stoichiometric ratio (14.7) in the exhaust. A high efficiency of conversion is achieved over the TWC in a very narrow A/F window that requires highly efficient control of the exhaust composition, particularly in terms of the residual oxygen concentration. Strong deviations or oscillations of A/F from the stoichiometric value lead to widening of the operating window, resulting in a below average performance of the TWC. To widen the operating window of the TWC, the exhaust gas composition on the catalyst must be controlled by storing oxygen in oxygen-rich conditions and by releasing oxygen in oxygen-lean conditions. This can be achieved by adding oxygen storage material, which shows high oxygen storage capacity (OSC), to the TWC. In this sense, OSC is defined as the extent of reversibly exchanged oxygen of an oxygen storage material under rich and lean fluctuations of the exhaust gas composition. This oxygen storage material readily undergoes reduction–oxidation cycles that provides oxygen for CO and HC oxidation in the rich region and the reduced state can remove oxygen from the gas phase when the exhaust gas runs into the lean region. Thus, an oxygen storage material not only widens the A/F ratio window, but also promotes oxidation activity. Therefore, materials for TWC applications should have redox behavior capable of releasing/uptaking rapidly as much oxygen as possible, leading to a high OSC. Utilization of lattice oxygen from reducible oxides under reducing conditions and thereafter replenishment of oxygen under an oxidizing atmosphere has played a critical role in various oxidation reactions. Therefore, a reducible oxide with redox characteristics can act as both an oxidizing and reducing agent to various reactants, depending on the reaction conditions. In this regard, CeO₂ has been found to be an efficient oxygen storage material for wide application in TWCs according to the following reversible reactions:

CeO₂ → CeO_2−δ + δ/2O₂

CeO_2−δ + δ/2O₂ → CeO₂

where the first reaction is related to the oxygen-lean region, the second one is for the oxygen-rich region and the value for δ is between 0 and 0.25. However, the experimentally achieved δ is ∼0.05 in the temperature range up to 600 °C.

To understand the reducibility of oxygen species as well as the metal–support interaction in CeO₂ and related materials, H₂-TPR is extensively employed where the volume of H₂ consumed by the reduction of an oxide is measured. The OSC of CeO₂ for oxidation/reduction reactions was estimated for the first time in a H₂-TPR experiment on CeO₂ by Yao and Yao.¹³ A simple mechanism for the OSC from H₂-TPR can be written as follows:

CeO₂ + δH₂ → CeO_2−δ + δH₂O

CeO_2−δ + δ/2O₂ → CeO₂.

H₂-TPR of CeO₂ shows primarily two peaks at approximately 500 °C and 825 °C.^{13,103,117,118} The low temperature peak is due to the reduction of the most easily reducible surface oxygen of CeO₂, whereas the removal of bulk oxygen is suggested as the cause of the high temperature peak at 825 °C. The first peak is dependent on the surface area and is responsible for higher catalytic activity at lower temperature. Thus, the unique redox property of CeO₂ has made it a special component in three-ways catalysts, because it can store oxygen during lean (net oxidizing, excess of O₂) conditions and release it during rich (net reducing, deficient of O₂) conditions as given below:¹¹⁹

CeO₂ + δCO → CeO_2−δ + δCO₂

CeO₂ + C_xH_y → CeO_{2−(2x+1/2y)} + xCO₂ + 1/2yH₂O

CeO₂ + δH₂ → CeO_2−δ + δH₂O

CeO_2−δ + δNO → CeO₂ + δ/2N₂

CeO_2−δ + δH₂O → CeO₂ + δH₂

CeO_2−δ + 1/2δO₂ → CeO₂.

It has also been observed that substitution of isovalent cations, such as Hf⁴⁺, Pr⁴⁺ and Tb⁴⁺, into the CeO₂ lattice, forming a Ce_1−xM_xO_2−δ solid solution, enhances the OSC.^117,118 Nonreducible ZrO₂ does not show redox properties as there is no peak in its respective H₂-TPR profile, whereas significant OSC has been observed in Ce_1−xZr_xO₂ samples.^117,118 Higher OSC has also been noticed when precious metals are doped with CeO₂.¹¹⁷ In general, a high surface area, oxygen vacancies and movement of oxide ions from tetrahedral sites to vacant octahedral sites are the possibilities for enhancing the OSC. It is important to note that Ce_1−xZr_xO₂ and such oxygen storage materials are not the catalysts themselves for real applications. Noble metals, such as Pt, Pd, Rh, and bimetallic Pt–Rh, are dispersed mostly in the form of nanocrystalline metal particles on such OSC materials and they are used as catalysts.

4.3. Conventional catalysts for exhaust catalysis

In auto exhaust catalysis, a high efficiency for the conversions of NO_x, CO and hydrocarbons can be achieved over a TWC in a very narrow A/F window that requires highly efficient control of exhaust composition, particularly in terms of residual oxygen concentration. This has been achieved by adding oxygen storage material to the TWC, which shows high OSC. Therefore, formulating suitable catalysts for TWC is a challenging task in auto exhaust catalysis. Within this context, studies on individual reactions for controlling exhaust emission, such as NO reduction, CO and hydrocarbon oxidation and SCR of NO by NH₃, CO and hydrocarbons in the presence of O₂, are also important. Therefore, extensive research work has been going on in the last few decades to find efficient catalysts for these reactions.

In general, in the early years of exhaust catalysis, metal, alloy and metal oxide catalysts were used for catalytic reactions to control exhaust emission, especially NO_x removal.¹¹⁹ Basic studies on metal catalysts involve single crystal or polycrystalline forms of bulk metal such as powder, wire, foil or ribbons. Prominent metals used are Pt, Pd, Rh, Ru and Ir.^120–123 CuO, Fe₂O₃, TiO₂, V₂O₅, MoO₃, WO₃, Cr₂O₃ and Al₂O₃ are employed as oxide catalysts.¹²⁴ Transition metal ion-exchanged zeolites have also been developed for exhaust catalysis.¹²⁵ Copper is one of the promising elements, which has the ability to increase the SCR activity of NO with NH₃. Cu²⁺ ion-exchanged Y zeolite, mordenite, ZSM-5 and MFI-ferrisilicate have been found to be effective SCR catalysts for NO.^119,126,127 Other metal ions, such as Fe³⁺, Co²⁺, Ce⁴⁺ and Ag⁺-exchanged zeolites, also show good SCR activity for NO.^{119,128–130} Noble metal exchanged zeolites are active for the removal of NO.¹³¹ The catalytic reduction of NO and the oxidation of CO and hydrocarbons have also been carried out over pure and cation-substituted perovskite oxides such as LaFeO₃, LaCoO₃, LaMnO₃, Ln_1−xPb_xMnO₃ (Ln = La, Pr and Nd), and LaMn_1−xCu_xO₃ and La_1−xSr_xMnO₃.^132–134 They have different cationic sites in the crystal structure, which are highly amenable to substitution. Usually, metals in high oxidation states can be stabilized in the perovskite structure. Similarly, a large number of spinels, AB₂O₄ (A = Mg²⁺, Co²⁺, Ni²⁺, Cu²⁺ and Zn²⁺ and B = Al³⁺, Cr³⁺, Mn³⁺, Fe³⁺ and Co³⁺), are also employed for NO reduction and CO and hydrocarbon oxidation.^119,135

In early 1980s, supported metal catalysts came into the scenario of auto exhaust catalysis.^136–138 Metal catalysts with low metal loadings are usually dispersed on the surface of a support. Highly dispersed catalysts provide a maximum surface area. As the metal loading increases, the crystallites come close together, resulting in larger particle sizes. In general, the more expensive noble metal catalysts have low metal loadings and are highly dispersed, whereas catalysts containing less expensive base metals have higher metal loadings, usually 20–40% by weight or atom. Supports, such as Al₂O₃, SiO₂, TiO₂, CeO₂, ZrO₂, SnO₂, WO₃ and MgO, are used for the dispersion of metals. In this sense, in the search for a suitable material to satisfy several requirements for exhaust catalysis, CeO₂ has emerged as a unique and active oxide support in the last few years. However, a catalyst in rigorous automotive exhaust conditions should have (a) intrinsic reactivity, (b) poison resistance ability and (c) durability and it soon has become apparent that the base metal oxides of Ni, Co, Mn and Cr do not have these properties. In particular, modern-day automotive exhaust catalysts revolve around three noble metals, Pt, Pd and Rh, which have been dispersed, stabilized, promoted, alloyed and segregated in sophisticated ways to fulfill the required performance and γ-Al₂O₃ has been found to be the main support along with a CeO₂ additive. It is observed that the addition of CeO₂ increases the activity of the Al₂O₃ catalyst towards three-way catalytic activities. The promoting effect of CeO₂ is largely attributed to the enhancement of metal dispersions, OSC, increase in oxygen mobility due to oxide ion defects and thermal stability. CeO₂ is able to act as an efficient oxygen storage material by releasing/storing oxygen because of its capability to undergo effective reduction and reoxidation under fuel-rich (A/F < 14.7) and fuel-lean (A/F > 14.7) conditions. The reversible oxygen storage/release feature coupled with chemical stability in adverse conditions is a unique feature that makes CeO₂ a prime material in TWC design.^117,136,139

4.4. Noble metal ionic catalysts for exhaust catalysis

From the fundamental basis of TWC it is necessary to develop materials that can provide good OSC as well as active sites for redox-type catalytic exhaust reactions, because both these properties are correlated. Realizing the importance of CeO₂ as an oxygen storage material as well as an active catalyst support for redox-type catalytic exhaust reactions, we have synthesized Ce_1−xM_xO_2−δ (M = Pd, Pt, Rh, Ru, Cu, Ag and Au), Ce_1−x−yA_xM_yO_2−δ (A = Ti, Zr, Sn and Fe; M = Pt and Pd) and Ti_1−xM_xO_2−δ (M = Pd, Pt, Rh and Ru) catalysts by solution combustion, hydrothermal and sonochemical methods, where metal ions act as active sites for adsorption as well as reaction. Promoting the action of CeO₂ and TiO₂, the metal–support interaction, high OSC, hydrogen spillover and synergistic interaction are attributed to the interaction of Mⁿ⁺/M⁰, Ce⁴⁺/Ce³⁺ and Ti⁴⁺/Ti³⁺ redox couples in the noble metal ion-substituted reducible oxides.²¹ In the following sections, we discuss several aspects of exhaust catalysis over these novel materials.

4.4.1. OSC of CeO₂ and TiO₂-based materials. To understand the OSC, the reducibility of oxygen species and metal–support interactions in CeO₂ and TiO₂ and related materials, H₂-TPR and CO-TPR are extensively employed and the volume of H₂ or CO consumed by the reduction of an oxide is measured. Pure CeO₂, TiO₂, PdO, Rh₂O₃, CuO and SnO₂ show H₂ uptake peaks at higher temperatures in comparison with metal-substituted CeO₂. ZrO₂ does not show redox properties as there is no peak in its H₂-TPR profile, whereas significant OSC has been observed in Ce_1−xZr_xO₂ samples. As a specific example, H₂-TPR profiles of CeO₂, Pt doped CeO₂ and Ce_1−xTi_xO₂ compounds are shown in Fig. 9. There are three regions of H₂ uptake. The first one extends from −50 to 120 °C, the second region extends from ∼120 to 300 °C and the third region is above 300 °C. The low-temperature region corresponds to hydrogen uptake related to Pt species in the compound.⁴⁷ The second and third regions of H₂ uptake are attributed to the reduction of the Ti⁴⁺ and Ce⁴⁺ states, respectively. Reduction of both Ti⁴⁺ and Ce⁴⁺ ions occurs at a lower temperature upon Pt substitution and the extent of reduction of Ce⁴⁺ increases. The H₂ uptake curve for Ce_0.99Pt_0.01O_2−δ is also given for comparison in the figure. The H₂/Pt molar ratio in the first region of 1 at% Pt/Ce_0.85Ti_0.15O₂ is 12.0, which is 3 times higher than that observed over Ce_0.99Pt_0.01O_2−δ. The ratio is 7.5 in the first region for 1 at% Pt/Ce_0.85Ti_0.15O₂. H₂ uptake starts at ∼30 °C in Ce_0.98Pd_0.02O_2−δ with a hydrogen adsorption peak at ∼65 °C and a small peak at 450 °C due to CeO₂ reduction is observed. Taking the low temperature H₂ peak at 65 °C, the ratio of H/Pd is four in Ce_0.98Pd_0.02O_2−δ. In the case of Ce_0.73Ti_0.25Pd_0.02O_2−δ, the H/Pd ratio is 17.⁴⁸ The hydrogen consumption peak around 100 °C for Ce_1−xRh_xO_2−δ is attributed to reduction of Rh³⁺ species and a second broad peak centered at ∼270 °C corresponds to the reduction of CeO₂.⁴⁴ Thus, Ce_1−xM_xO_2−δ, Ce_1−x−yTi_xM_yO_2−δ, Ce_1−x−ySn_xM_yO_2−δ, Ce_1−x−yFe_xM_yO_2−δ and Ce_1−x−yZr_xM_yO_2−δ (M = Pd, Pt, Rh and Ru) catalysts have been observed to show high oxygen storage capacities.^{44,47,49–52,62,63,70} There is one sharp peak at low temperatures and a broad peak in the higher temperature region in all the substituted TiO₂ catalysts.⁵⁴ Unsubstituted TiO₂ shows very little H₂-uptake and reduction starts above 400 °C. Therefore, the higher temperature broad peak for the substituted catalysts is attributed to the reduction of TiO₂. On the other hand, the low temperature peak corresponds to the reduction of Pd²⁺ and Pt²⁺ ions in TiO₂. The oxygen storage capacities of Ti_0.97Pd_0.03O_2−δ, Ti_0.97Pt_0.03²⁺O_1.97 and Ti_0.97Pt_0.03⁴⁺O_1.97 are 5100, 2010 and 2315 μmol g⁻¹, respectively.^55,56 Oxygen storage capacities of a large number of CeO₂ and TiO₂-based materials studied in our laboratory are summarized in Table 2.^{44,47,49–52,55,56,62,63,70,140,141}

Fig. 9 H₂-TPR profiles of (a) Ce_0.99Pt_0.01O_2−δ, (b) Ce_0.84Ti_0.15Pt_0.01O_2−δ and (c) Ce_0.83Ti_0.15Pt_0.02O_2−δ.

Table 2 Oxygen storage capacities (OSC) of CeO₂ and TiO₂-based materials studied in our laboratory

Materials	OSC (μmol g^−1)	References
CeO2	174	52
Ce0.98Rh0.02O1.99	244	44
Ce0.75Zr0.25O2	1900	140
Ce0.6Zr0.4O2	2429	52
Ce0.5Zr0.5O2	2900	140
Ce0.9Ti0.1O2	740	47
Ce0.85Ti0.15O2	880	47
Ce0.8Ti0.2O2	1110	47
Ce0.75Ti0.25O2	1340	141
Ce0.7Ti0.3O2	1590	47
Ce0.6Ti0.4O2	2000	47
Ce0.99Pt0.01O1.99	290	47
Ce0.84Ti0.15Pt0.01O1.99	1570	47
Ce0.83Ti0.15Pt0.02O1.98	2320	47
Ce0.73Ti0.25Pd0.02O1.98	1286	52
Ce0.75Hf0.25O2	714	52
Ce0.73Hf0.25Pd0.02O1.98	1121	52
Ce0.58Hf0.4Pd0.02O1.98	804	52
Ce0.98Pd0.02O1.98	232	52
Ce0.83Zr0.15Pd0.02O1.98	638	52
Ce0.73Zr0.25Pd0.02O1.98	1509	52
Ce0.58Zr0.4Pd0.02O1.98	1612	52
Ce0.78Zr0.2Pd0.02O1.98	620	50
Ce0.9Sn0.1O2	980	49
Ce0.8Sn0.2O2	1650	49
Ce0.7Sn0.3O2	2100	49
Ce0.6Sn0.4O2	2545	49
Ce0.5Sn0.5O2	2780	49
Ce0.78Sn0.2Pd0.02O1.98	1650	50
Ce0.5Sn0.48Pd0.02O1.98	1330	50
Ce0.95Ru0.05O1.97	1335	63
Ce0.9Ru0.1O1.94	2513	63
Ce0.9Fe0.1O1.88	665	51
Ce0.89Fe0.1Pd0.01O1.84	780	51
Ce0.67Cr0.33O2.11	2513	62
TiO2	250	55
Ti0.99Pd0.01O1.99	1240	55
Ti0.98Pd0.02O1.98	2220	55
Ti0.97Pd0.03O1.97	5100	55
Ti0.97Pt0.03^2+O1.97	2010	56
Ti0.97Pt0.03^4+O1.97	2315	56

---

4.4.2. Exhaust catalysis over noble metal ionic catalysts. Noble metal ionic catalysts have been examined for various important catalytic reactions, such as CO oxidation, NO reduction by NH₃ and CO, hydrocarbon oxidation, three-way catalytic reactions, the water-gas shift reaction (WGSR), preferential oxidation of CO (CO-PROX), H₂–O₂ recombination, hydrogenation and Heck reaction, which are discussed in the following sections. It has been observed that the catalytic activities of noble metal ionic catalysts toward these reactions are much higher than those of supported noble metal catalysts.
4.4.2.1. CO oxidation by O₂. Among Pt²⁺, Pd²⁺ and Rh³⁺ ions, Pd²⁺ ion-substituted CeO₂, Ce_1−xTi_xO₂ and TiO₂ show the highest rate of CO conversion and the lowest activation energy toward CO oxidation. Within this context, it is important to note that Pd is the cheapest among the noble metals. It has also been observed that activation energy decreases with an increase in the effective charge on the Pd²⁺ ion in these oxides.⁵² It has to be noted that the rates with metal ionic catalysts are higher by 20–30 times compared with the same amount of metal-impregnated catalysts. Pt and Pd ion-substituted Ce_1−xTi_xO₂, Ce_1−xSn_xO₂ and Ce_1−xFe_xO_2−δ catalysts show very high CO conversion.^47–50 In the top panel of Fig. 10, CO conversion over different Pt²⁺ substituted catalysts is presented. The Ti_0.99Pt_0.01O_2−δ catalyst is observed to show the lowest temperature 100% CO conversion. Complete conversions of CO to CO₂ are achieved below 105 and 135 °C over hydrothermally prepared Ce_0.9Ru_0.1O_2−δ and Ce_0.95Ru_0.05O_2−δ catalysts, respectively, in the presence of feed oxygen.⁶³ In another type of experiment, CO is oxidized to CO₂ by extracting activated lattice oxygen in the absence of feed oxygen. On exposure to oxygen, the lattice oxygen is replenished. This means that the feed oxygen is adsorbed and incorporated into the oxide ion vacancy, indicating the presence of two independent sites on the catalyst that adsorb reducing and oxidizing molecules.^48,49,63 A Ce_0.65Fe_0.33Pd_0.02O_2−δ catalyst prepared by sonochemical method shows very low temperature CO oxidation with an activation energy of 38 kJ mol⁻¹.⁷⁰ Low temperature CO oxidation over Ti_1−xPd_xO_2−x catalysts has been demonstrated by our group.¹⁴² It is important to note that CO oxidation over Pt/Al₂O₃ and Pd/Al₂O₃ synthesized by the solution combustion method occurs at higher temperatures compared to CeO₂ and TiO₂-based catalysts.⁴²

Fig. 10 Top panel: CO oxidation over Pt substituted CeO₂ and TiO₂-based catalysts; bottom panel: CO oxidation over Rh metal and oxidized Rh species. Reprinted from ref. 47 with the permission of American Chemical Society and from ref. 145 with the permission of Springer.

In recent times, there have been reports about the CO oxidation activities of supported ionic noble metals and it has been found that these oxidized noble metals are more active compared to noble metals. Grass and coauthors have demonstrated the formation of a thin, active RhO_x over-layer on polymer-stabilized Rh nanoparticles, which is responsible for low-temperature CO oxidation.¹⁴³ Active RhO_x phase dispersed on CeO₂ has been found to be highly active for CO oxidation as reported by Ligthart et al.¹⁴⁴ Manuel and coworkers have shown that an oxidized Rh species (Rh^x+) on Ce_0.75Zr_0.25O₂ exhibits a significant 25 times higher turnover rate in CO oxidation than the corresponding zero valent one (Rh⁰).¹⁴⁵ Herein, the Ce_0.75Zr_0.25O₂ support stabilizes Rh atoms as ions. A comparison of CO oxidation over Rh^x+/Ce_0.75Zr_0.25O₂ and Rh⁰/Ce_0.75Zr_0.25O₂ is displayed in the bottom panel of Fig. 10. A solution combustion-synthesized Au/CeO₂ catalyst, where Au is in the +3 oxidation state, shows a very high CO oxidation rate at low temperatures.⁴⁶ Carrettin et al. have demonstrated the stabilization of Au³⁺ in nanocrystalline CeO₂, which shows very high CO oxidation activity.¹⁴⁶ A strong interaction between ionic gold and CeO₂ is attributed to an exceptionally high CO oxidation rate over an Au/CeO₂ catalyst prepared by the deposition–precipitation method.¹⁴⁷ In Table 3, a comparison of CO conversion rates and activation energies (E_a) for the CO + O₂ reaction over noble metal ion-substituted CeO₂ and TiO₂ catalysts with supported nanometal catalysts is presented.^{47–49,51,55,56,63,70,82,84,87,142,148–153}

Table 3 Comparison of CO conversion rates and activation energies (E_a) for the CO + O₂ reaction over noble metal ionic catalysts with supported nanometal catalysts

Catalysts	Rate (μmol g^−1 s^−1)	Ea (kJ mol^−1)	References
5 wt% Ru/SiO2	1.0 (110 °C)	94	148
5 wt% Rh/SiO2	0.0251 (110 °C)	103	148
5 wt% Pd/SiO2	0.316 (143 °C)	103	148
5 wt% Ir/SiO2	0.26 (180 °C)	105	148
5 wt% Pt/SiO2	0.32 (115 °C)	56	148
Pd/CeO2/Al2O3	38 (250 °C)	84	149
0.014 wt% Rh/Al2O3	0.6 (196 °C)	115.5	150
0.014 wt% Rh/9 wt% Ce/Al2O3	0.95 (196 °C)	90.3	150
0.5 wt% Pd/CeO2–ZrO2	0.7 (220 °C)	175	151
1.53% Pd/SiO2	0.155 (140 °C)	—	152
25Au75Pd/SiO2	0.128 (140 °C)	—	152
50Au50Pd/SiO2	0.06 (140 °C)	—	152
5% Au/CeO2	0.77 (300 °C)	80.3	153
5% Pd/CeO2	0.76 (300 °C)	76.9	153
Ce0.98Pd0.02O2−δ	7.0 (130 °C)	67.2	48
1 at% Pd/Al2O3 (impregnated)	2.73 (200 °C)	86.88	142
1 at% Pd/TiO2 (impregnated)	0.758 (100 °C)	75.25	142
Ce0.99Pt0.01O2−δ	0.6 (155 °C)	55.4	47
Ce0.84Ti0.15Pt0.01O2−δ	28.0 (155 °C)	82.3	47
Ce0.73Ti0.25Pd0.02O2−δ	18.0 (120 °C)	54.6	48
2% Pd/Ce0.75Ti0.25O2−δ	—	132.3	48
Ce0.78Sn0.2Pd0.02O2−δ	1.9 (50 °C)	84	49
Ce0.95Ru0.05O2−δ	2.05 (100 °C)	94.5	63
Ce0.9Ru0.1O2−δ	3.3 (100 °C)	44.8	63
Ce0.89Fe0.1Pd0.01O2−δ	—	52.5	51
Ce0.9Fe0.1O2−δ	—	46.2	51
Ce0.99Pd0.01O2−δ	—	63	51
Ce0.65Fe0.33Pd0.02O2−δ	4.1 (80 °C)	38	70
0.74% Pd/CeO2	—	42	82
2% Pt/Ce0.72Zr0.18Pr0.1O2	1.6 (61 °C)	—	84
2% Pd/Ce0.72Zr0.18Pr0.1O2	1.6 (130 °C)	—	84
2% Ru/Ce0.72Zr0.18Pr0.1O2	1.6 (150 °C)	—	84
2% Pt/CeO2	2.89 (40 °C)	51	87
Ti0.99Pd0.01O1.99	13.83 (120 °C)	53.13	142
Ti0.97Pd0.03O1.97	2.75 (60 °C)	17	55
Ti0.97Pt0.03^2+O1.97	0.76 (60 °C)	33.6	56
Ti0.97Pt0.03^4+O1.97	0.08 (60 °C)	16.8	56

---

4.4.2.2. NO reduction by CO. High activity for the NO + CO reaction with 100% N₂ selectivity has been observed over Ce_0.98Pd_0.02O_2−δ.^154,155 The decomposition rate of adsorbed NO is faster over this catalyst, leading to higher N₂ selectivity. For the NO + CO + O₂ reaction, N₂ selectivity of 100% is observed at higher temperatures over this catalyst. Ce_0.98Pd_0.02O_2−δ shows lower temperature NO reduction by CO in relation to Ce_0.98Pt_0.02O_2−δ and Ce_0.98Rh_0.02O_2−δ catalysts.¹⁵⁵ N₂ selectivity is highest in Ce_0.98Pd_0.02O_2−δ among all these three catalysts in the temperature range of 125–350 °C; however, N₂ selectivities are more or less the same over all catalysts at higher temperatures (>300 °C). Complete NO reduction by CO occurs over Ce_0.99Pt_0.01O_2−δ and Ce_0.84Ti_0.15Pt_0.01O_2−δ at 180 °C.⁴⁷ 100% N₂ selectivity is observed at 240 °C for NO reduction by CO over a Ce_0.73Ti_0.25Pd_0.02O_2−δ catalyst.⁴⁸ Ce_0.78Sn_0.2Pd_0.02O_2−δ exhibits 70% N₂ selectivity at 200 °C, whereas 100% N₂ selectivity is obtained at 245 °C over this catalyst.⁴⁹ Complete NO conversion into N₂ is observed at 200 and 250 °C over Ce_0.95Ru_0.05O_2−δ and Ce_0.9Ru_0.1O_2−δ, respectively, where N₂O formation occurs at low temperatures.⁶³ NO reduction by CO over a Ti_0.99Pd_0.01O_1.99 catalyst shows high rates of NO conversion for the NO + CO reaction with high N₂ selectivity. The rate of N₂O reduction by CO is very high over this catalyst.¹³⁷ N₂ selectivity is 80% or more at all temperatures over this catalyst for the NO + CO + O₂ reaction. However, it is important to mention that complete NO reduction by CO over 1% Pt/Al₂O₃ and 1% Pd/Al₂O₃ catalysts is observed at 400 and 350 °C, respectively.⁴² Comparisons of NO conversion rates, activation energies and N₂ selectivities for the NO + CO reaction over noble metal ion-substituted CeO₂ and TiO₂ catalysts with supported nanometal catalysts are presented in Table 4.^{47–49,51,63,142,154,156–161}

Table 4 Comparison of NO conversion rates, activation energies (E_a) and selectivities (S_N2 and S_N2O) for the NO + CO reaction over noble metal ionic catalysts with supported nanometal catalysts

Catalysts	Rate (μmol g^−1 s^−1)	Ea (kJ mol^−1)	SN2	SN2O	References
1 wt% Pt/Al2O3	0.1 (300 °C)	92.4	27	73	156
0.2 wt% Rh/Al2O3	7.2 (300 °C)	184.8	32	68	157
1 wt% Pt–0.02 wt% Rh/Al2O3	0.58 (300 °C)	126	39	61	158
Pt–Rh/Al2O3–CeO2	0.54 (300 °C)	84	33	67	158
Rh/Al2O3	0.58 (227 °C)	86.6	45	55	159
Pd/Rh/Al2O3	0.11 (227 °C)	100	43	57	159
Pd/Al2O3	0.83 (227 °C)	59.8	36	64	159
Pd/La2O3/Al2O3	4.3 (227 °C)	100	30	70	159
1% Pd/γ-Al2O3	0.75 (287 °C)	158	34	66	160
Pd8Mo/Al2O3	2.0 (300 °C)	84	39	61	161
Ce0.98Pd0.02O2−δ	0.242 (175 °C)	70.2	75	25	154
Ce0.73Ti0.25Pd0.02O2−δ	2.0 (180 °C)	52.9	—	—	48
Ce0.99Pt0.01O2−δ	0.2 (135 °C)	100.8	—	—	47
Ce0.84Ti0.15Pt0.01O2−δ	0.2 (135 °C)	96.6	—	—	47
Ce0.78Sn0.2Pd0.02O2−δ	5.0 (150 °C)	105.4	100	—	49
Ce0.89Fe0.1Pd0.01O2−δ	3.8 (220 °C)	—	56	44	51
Ce0.95Ru0.05O2−δ	3.8 (200 °C)	41	100	—	63
Ce0.9Ru0.1O2−δ	3.4 (200 °C)	53.2	100	—	63
Ti0.99Pd0.01O2−δ	1.47 (175 °C)	64.1	70	30	142
1% Pd/Al2O3 (impregnated)	0.05 (240 °C)	86.6	59	41	142

---

4.4.2.3. NO reduction by H₂. NO can also be reduced by H₂, which is also present in the mobile and stationary sources. 100% N₂ selectivity and high reaction rates at low temperature are observed over the Ce_0.98Pd_0.02O_1.98 catalyst, therefore making it superior to other existing catalysts reported in the literature.¹⁶² 100% NO conversions are observed at 160, 180 and 370 °C over Ce_0.89Fe_0.1Pd_0.02O_2−δ, Ce_0.99Pd_0.01O_2−δ and Ce_0.9Fe_0.1O_2−δ, respectively.⁵¹ Solution combustion-synthesized Ti_0.99M_0.01O_2−δ (M = Pd, Pt, Rh and Ru) catalysts have been tested for the NO + H₂ reaction.⁵⁴ The rate of NO reduction by H₂ follows the order: Ti_0.99Pt_0.01O_2−δ > Ti_0.99Pd_0.01O_2−δ > Ti_0.99Rh_0.01O_2−δ > Ti_0.99Ru_0.01O_2−δ. The formation of the lowest amount of N₂O occurs over Ti_0.99Pd_0.01O_2−δ and no NH₃ formation is traced during the reaction, whereas a significant amount of N₂O and NH₃ formation can be observed over the Ti_0.99Pt_0.01O_2−δ catalyst. The experimental observations demonstrate that Ti_0.99Pd_0.01O_2−δ is the best catalyst among the four catalysts investigated. In the presence of excess O₂, N₂ selectivity increases over the Ti_0.99Pd_0.01O_2−δ catalyst during the NO + H₂ + O₂ reaction (lean conditions).
4.4.2.4. Selective catalytic reduction (SCR) of NO by NH₃. SCR of NO by NH₃ has been used extensively to clean up emissions from stationary exhausts. Good SCR catalysts should show low temperature activity, high N₂ selectivity and a wide SCR window. A series of metal ion-substituted TiO₂ catalysts have been synthesized by the solution combustion method and their SCR activities have been tested to understand the structure–property relationship and the reaction mechanism. The SCR reaction occurs at the lowest temperature over Ti_0.9Mn_0.1O_2−δ among the Ti_0.9M_0.1O_2−δ (M = Cr, Mn, Fe, Co and Cu) catalysts but N₂ selectivity is found to be highest over Ti_0.9Fe_0.1O_2−δ.¹⁶³ In this sense, the Ti_0.9Mn_0.05Fe_0.05O_2−δ catalyst, where both Mn and Fe are substituted in TiO₂, has been synthesized to optimize low temperature SCR activity and N₂ selectivity. Indeed, the SCR reaction occurs at low temperatures with a high N₂ selectivity over this catalyst. SCR activities over Ti_0.9Mn_0.1O_2−δ, Ti_0.9Fe_0.1O_2−δ and Ti_0.89Mn_0.1Pd_0.01O_2−δ have also been investigated to understand the role of base metal and noble metal ionic catalysts.¹⁶⁴ From H₂-TPR and SCR activity studies it has been observed that a more reducible catalyst shows better NH₃ oxidation and poor SCR activity. NO reduction follows the order: Ti_0.9Mn_0.1O_2−δ > Ti_0.89Mn_0.1Pd_0.01O_2−δ > Ti_0.9Fe_0.1O_2−δ and ammonia oxidation follows the reverse order, which agrees well with the reducibility of the catalyst. The Ti_0.89Mn_0.1Pd_0.01O_2−δ catalyst, where a base metal and a noble metal are substituted in TiO₂, does not show any synergetic effect. Therefore, the catalyst should have less reducibility to achieve better SCR activity. Thus, noble metal ion (Pd²⁺)-substituted TiO₂, which has better reducibility than the base metal ion (Mn³⁺)-substituted TiO₂, performs poorly for SCR.
4.4.2.5. NO reduction over NMIC coated on cordierite monolith. In recent years, monolith supports with honeycomb structures have been attractive alternative carriers for catalysts in heterogeneous catalysis and have been extensively used as catalyst supports inside catalytic converters for removal of exhaust gases in the automobile industry.^165–167 In general, the monolith is first coated with a high-surface-area oxide such as γ-Al₂O₃ and the process is called washcoating. The active catalyst phase is dispersed over the monolith support after washcoating. Several methods of washcoating and coating of the active catalyst phase on monoliths are available in the literature.^168,169 Due to advances in monolith technology, simple catalyst mounting methods, flexibility in reactor design, low pressure drop and high heat and mass transfer rates, monolithic supports dominate the entire automotive market as the preferred catalyst support. Ceramic monolith materials are leaders in the market among different monolith supports and the preferred material is cordierite.

Noble metal ionic catalysts are coated on cordierite monoliths by a single-step solution combustion method and three-way catalytic reactions have been carried out over these catalysts. The coating process involves growing of γ-Al₂O₃ on cordierite by the solution combustion method and then further growing of the active catalyst phase containing noble metal ionic catalysts on γ-Al₂O₃-coated cordierite, again by the same method. Due to the closeness of unit cell lattice parameters of cordierite and γ-Al₂O₃ as well as γ-Al₂O₃ and Ce_1−xM_xO_2−δ, polycrystalline γ-Al₂O₃ on cordierites and Ce_0.98Pd_0.02O_2−δ, Ce_0.98Pt_0.02O_2−δ and Ce_0.9Cu_0.1O_2−δ over γ-Al₂O₃ coated cordierites have been grown by the solution combustion method following the principle of epitaxial growth.^170,171

It has been observed that Ce_0.98Pd_0.02O_2−δ coated on cordierite shows much more catalytic activity toward CO oxidation, NO reduction and three-way catalysis in comparison with Ce_0.98Pt_0.02O_2−δ and Ce_0.9Cu_0.1O_2−δ-coated cordierite catalysts. Within this context, a detailed investigation on a Ce_0.98Pd_0.02O_2−δ-coated cordierite catalyst has been carried out. 100% CO and NO conversions are achieved at around 80 and 200 °C for CO + O₂ and NO + CO reactions, respectively, at a space velocity of 880 h⁻¹. At the same space velocity, a three-way catalytic reaction over a Ce_0.98Pd_0.02O_2−δ-coated monolith has been carried out with a gas mixture of 10 000 ppm CO, 2000 ppm NO, 2000 ppm acetylene (C₂H₂) and 7000 ppm O₂. It has been observed that 100% conversion of all the pollutants occurs below 220 °C and CO and C₂H₂ conversion is observed before NO conversion.¹⁷⁰ All pollutant gases are converted to N₂, CO₂ and H₂O below 225 °C, even with 15% excess oxygen. CO + O₂, NO + CO and three-way catalytic reactions over Ce_0.98Pd_0.02O_2−δ coated on a cordierite monolith are displayed in Fig. 11. In another study, three-way catalysis over a noble metal ionic catalyst coated on a monolith is carried out in the presence of NO, CO, O₂ and a mixture of hydrocarbons, such as acetylene (C₂H₂), ethylene (C₂H₄), propene (C₃H₆) and propane (C₃H₈), with a GHSV of 4523 h⁻¹. All the pollutant gases are fully converted to N₂, CO₂ and H₂O within 340 °C. NO, CO and C₂H₂ are converted at almost same temperature of 260 °C, whereas complete conversions of C₂H₄, C₃H₆ and C₃H₈ occur at 290, 315 and 340 °C, respectively.¹⁷¹ Thus, the Ce_0.98Pd_0.02O_2−δ catalyst coated on a cordierite monolith by the novel solution combustion method demonstrates good low-temperature three-way catalytic activity. In this regard, it is to be noted that the solution combustion method provides a new technique of washcoating a high-surface-area oxide, such as γ-Al₂O₃ and dispersing an active catalyst phase, containing noble metal ions over a monolith.

Fig. 11 Catalytic reactions over Ce_0.98Pd_0.02O_2−δ coated on cordierite monolith: (a) CO + O₂, (b) NO + CO and (c) three-way catalysis. Reprinted from ref. 170 with the permission of Springer.

4.4.2.6. Hydrocarbon oxidation. Complete CH₄ oxidation occurs over Ce_0.99Pt_0.01O_2−δ and Ce_0.99Pd_0.01O_2−δ at 400 and 330 °C, respectively, whereas complete oxidation of C₃H₈ over these catalysts has been observed at 110 and 230 °C, respectively.²⁰ On the other hand, these reactions occur at much higher temperatures over 1% Pt/Al₂O₃ and 1% Pd/Al₂O₃ catalysts.¹⁷² Similarly, 100% conversion of CH₄ and C₃H₈ has been noticed at much lower temperatures over Ce_0.99Ag_0.01O_2−δ and Ce_0.99Au_0.01O_2−δ catalysts in comparison with 1% Ag/Al₂O₃ and 1% Au/Al₂O₃.^45,46 The rate of C₂H₄ oxidation over bimetallic ionic Ce_1−xPt_x/2 Rh_x/2O_2−δ is observed to be higher than the corresponding monometallic ionic Ce_1−xPt_xO_2−δ and Ce_1−xRh_xO_2−δ catalysts.⁷³ Ce_0.84Ti_0.15Pt_0.01O_2−δ shows a higher catalytic activity toward the oxidation of C₃H₈, C₂H₄ and C₂H₂.⁴⁷ Activation energies of oxidation of C₂H₄ and C₂H₂ over Ce_0.98Pd_0.02O_2−δ and Ti_0.99Pd_0.01O_2−δ are found to be lower in comparison with previous studies in the literature.^142,154 Oxidation reactions of several hydrocarbons over Ce_1−xRu_xO_2−δ occur at lower temperatures and complete conversion temperatures depend on the amount of Ru.⁶³ Hydrocarbon oxidation temperatures over Ti_1−xPd_xO_2−δ catalysts decrease with an increase in Pd content and room temperature C₂H₂ oxidation occurs over a Ti_0.97Pd_0.03O_2−δ catalyst.⁵⁵

5. Water gas shift reaction

The reaction of CO with steam, leading to H₂ and CO₂, is generally called the water gas shift reaction (WGSR). It has been used as an important step in the production of H₂ as well as bringing down the CO level in the steam reformate introduced into the polymer electrolyte fuel cells (PEMFC). Fu and Flytzani-Stephanopoulos have shown that the water gas shift reaction occurs at significantly low temperatures over leached Pt/CeO₂ and Au/CeO₂ containing ionic Pt and Au, compared to unleached catalysts that contain metallic species.⁹⁰ Ionic Pt and Au species that are strongly associated with surface cerium–oxygen groups are responsible for the high activity. CO conversion has been found to be maximized at 200 °C over Ce_1−xPt_xO_2−δ catalysts without any methanation.¹⁷³ However, Ti_0.99Pt_0.01O_2−δ shows the highest activity among Ti_0.99Pt_0.01O_2−δ, Ce_0.83Ti_0.15Pt_0.02O_2−δ and Ce_0.98Pt_0.02O_2−δ catalysts without deactivation.¹⁷⁴ There is no carbonate formation over Ti_0.99Pt_0.01O_2−δ due to the highly acidic nature of Ti⁴⁺ in the catalyst and consequently, no deactivation is observed even after prolonged reaction. Pt²⁺-substituted ZrO₂ shows the lowest temperature WGS reaction among Pt²⁺ and Pd²⁺-substituted ZrO₂ and Ce_0.85Zr_0.15O₂ catalysts.¹⁷⁵ The formation of Brønsted acid–base pairs is the main cause for the high activity of ZrO₂ catalysts. Ce_0.78Sn_0.2Pt_0.02O_2−δ shows excellent CO conversion due to the presence of Ce⁴⁺/Ce³⁺ and Sn⁴⁺/Sn²⁺ redox couples.¹⁷⁶ Much higher activity for the WGS reaction is noticed over a Ce_0.65Fe_0.33Pt_0.02O_2−δ catalyst, compared to Ce_0.67Fe_0.33O_2−δ, due to the synergistic interaction of Pt⁴⁺ with Ce⁴⁺ and Fe³⁺ ions.¹⁷⁷ Ce_0.95Ru_0.05O_2−δ prepared by the hydrothermal method shows a low activation energy and high conversion rates at low temperatures and 100% H₂ selectivity is observed over this catalyst.¹⁷⁸ A sonochemically synthesized Ce_0.65Fe_0.33Pd_0.02O_2−δ catalyst is highly active for water gas shift reaction with reaction rate of 27.2 μmol g⁻¹ s⁻¹ and activation energy of 46.4 kJ mol⁻¹.⁷⁰

6. CO-PROX reaction

Preferential oxidation of CO (CO-PROX) is widely used to reduce trace levels of CO in a H₂-rich gas stream from fuel reformate that is fed to polymer electrolyte fuel cells (PEMFC).¹⁷⁹ A CO content above 10 ppm is present in the H₂-rich gas stream from fuel reformate that is fed to deactivated platinum-based electrodes in PEMFCs. An effective CO-PROX catalyst shows high activity and selectivity for CO oxidation. It should be active in two temperature ranges, 170–230 °C, which is the outlet temperature for the low temperature water gas shift reaction (WGSR) and 80–100 °C, which is the operational temperature for PEMFCs. Therefore, a wide temperature range is required to avoid precise temperature controls. There have been reports of a Pt/CeO₂ catalyst for the PROX reaction, where Pt is present in the metallic state.¹⁸⁰ Pt²⁺ in CeO₂ in the form of a Ce_1−xPt_xO_2−δ solid solution is a more efficient catalyst than supported Pt metal catalysts and just 15% of Ti substitution in CeO₂ improves the overall PROX activity.¹⁸¹ CO conversion of 100% is observed over a Ce_0.83Ti_0.15Pt_0.02O_2−δ catalyst at 55 °C, which represents a much higher conversion than that over Ce_0.98Pt_0.02O_2−δ. Thus, the CO-PROX temperature window is larger than that of the Ce_0.98Pt_0.02O_2−δ catalyst. The higher activity of Ce_0.83Ti_0.15Pt_0.02O_2−δ is because of the higher reducibility of this catalyst in comparison with Ce_0.98Pt_0.02O_2−δ. Lattice oxygen becomes activated because of Ti substitution and can easily be removed under the reaction conditions. Herein, reducibility is an important factor as the reaction conditions are highly reducing (35% H₂), making Ce_0.83Ti_0.15Pt_0.02O_2−δ more active than Ce_0.98Pt_0.02O_2−δ. On the other hand, Ti_0.98Pt_0.02O_2−δ is not a good PROX catalyst for CO as it shows a good activity for the H₂ + O₂ recombination reaction.

7. H₂ + O₂ reaction

H₂ + O₂ recombination is technologically a very important reaction. Keeping the H₂ concentration lower than 4.1% in nuclear power reactors is crucial for safety. Furthermore, the catalytic combustion of H₂ by O₂ is the cleanest source of energy. Hydrogen fuel cells also make use of a Pt/C catalyst for the overall recombination to produce power. Ever since Michael Faraday demonstrated the H₂ + O₂ recombination reaction over platinum metal plates, Pt nanoparticles have remained the only room-temperature recombination catalyst for this reaction. Alternative approaches to achieving higher recombination rates are: (1) to design a catalyst possessing two different adsorption sites for H₂ and O₂, unlike the single Pt⁰ site in Pt nanoparticles for both the molecules, (2) dissociative chemisorptions of H₂ and O₂ so that the activation energy of recombination is brought down and (3) making the catalyst more tolerant to CO to avoid CO poisoning in fuel cell applications. We have designed new catalysts, such as Ce_1−xPt_xO_2−δ and Ti_0.99Pd_0.01O_2−δ, where Pt and Pd are present in their ionic state and are highly active for H₂ + O₂ reaction.^182,183 It has been demonstrated that the Ti_0.99Pd_0.01O_2−δ catalyst not only shows higher rates for the H₂ + O₂ recombination reaction, but is also more tolerant to CO than Ce_0.98Pt_0.02O_2−δ.¹⁸³ Ce_0.98Pt_0.02O_2−δ has also been found to be an excellent hydrogen and oxygen recombinant catalyst at room temperature for the recovery of water in sealed lead-acid batteries.¹⁸⁴

8. Photocatalysis

Photocatalysis is more economical with regard to low energy consumption and operating costs. It is an alternative method of NO reduction and decomposition that has been carried out over noble metal ionic catalysts in the presence of UV light in our laboratory. The optimum Pd²⁺ concentration has been found to be 1 at% for NO reduction by CO and NO decomposition over Pd²⁺ ion-substituted TiO₂ containing Ti⁴⁺, Pd²⁺, O²⁻ and the oxide ion vacancy sites.⁵³ Measureable NO conversion is not observed over pure TiO₂. With an increase in the Pd²⁺ ion content in Ti_1−xPd_xO_2−δ, NO conversion increases with values of x from 0.005 to 0.01, and then it decreases with an increase in the Pd²⁺ concentration and for 3 at% Pd there is no NO conversion. The highest NO conversion of 80% was observed for Ti_0.99Pd_0.01O_2−δ. This shows that there is an optimum Pd²⁺ ion concentration for NO conversion. The Pd²⁺4d band becomes broader as the Pd²⁺ ion concentration increases and the photoluminescence decreases significantly at 2 and 3%, leading to lower conversion. A decrease in NO conversion over Ti_0.98Pd_0.02O_2−δ and Ti_0.97Pd_0.03O_2−δ catalysts can also be correlated with an increase in electron density due to the Pd²⁺4d band in the band gap region (0–3 eV) of pure TiO₂. With an increase in the Pd²⁺ ion concentration, an increase in the h⁺–e⁻ pair recombination (shorting effect) should occur that lowers the rate of the photocatalytic reaction. Again, NO is adsorbed over both pure TiO₂ and Ti_0.99Pd_0.01O_2−δ, but photodissociation of NO happens only on Ti_0.99Pd_0.01O_2−δ, not on pure TiO₂. Therefore, photodissociation of NO requires oxide ion vacancies, which are present on Ti_0.99Pd_0.01O_2−δ, and therefore NO adsorbed on oxygen vacancy sites is dissociative NO. Therefore, Pd²⁺ ions for adsorption, oxide ion vacancies and UV light are essential requirements for the photo dissociation of NO.

NO conversion of 45% is observed when NO decomposition is carried out in the absence of CO over a Ti_0.99Pd_0.01O_2−δ catalyst. In contrast, there is no NO conversion over pure TiO₂. NO decomposition decreases over Ti_0.99Pd_0.01O_2−δ in light on/off mode. It has been demonstrated that NO₂ formed due to a partial reaction between NO and evolved O₂ from NO decomposition blocks the oxide ion vacancy sites. However, in the presence of CO, this O₂ is utilized to form CO₂, and therefore a high NO conversion of 80% is observed over this catalyst.

9. Hydrogenation

The hydrogenation of p-chloronitrobenzene (p-CNB) to p-chloroaniline (p-CAN) is an important reaction, because chloroanilines are key intermediates in industry for the synthesis of herbicides, pesticides, pigments, pharmaceuticals, cosmetics and other organic fine chemicals. Recently, Mistri and coworkers have demonstrated the hydrogenation of p-chloronitrobenzene to p-chloroaniline over Ce_0.98Pd_0.02O_2−δ, Ce_0.98Ru_0.02O_2−δ and Ce_0.98Pd_0.01Ru_0.1O_2−δ catalysts.¹⁸⁵ It has been observed that hydrogenation is inactive over monometallic ion-substituted catalysts. On the other hand, hydrogenation is completed beyond 75 min over Ce_0.98Pd_0.01Ru_0.1O_2−δ. A little higher activity with a lower selectivity is noticed with an increase in temperature from 35 to 80 °C. The presence of Ru⁴⁺, together with Pd²⁺ on CeO₂ plays a crucial role towards the hydrogenation of p-CNB without dechlorination, indicating remarkable Ru⁴⁺ promotion in the bimetallic ion-substituted CeO₂ catalyst.

Hydrogenation of benzene to cyclohexane is an important reaction from an industrial point of view. A 42% conversion of benzene to cyclohexane with 100% selectivity is observed at 100 °C over Ce_0.98Pt_0.02O_2−δ catalysts.¹⁸⁶ The turnover frequency is observed to be an order of magnitude higher than those of other Pt catalysts.

10. Heck reaction

Heck coupling reaction, catalyzed by Pd, is well known in organic synthesis.^187,188 It is known for its high tolerance of functional groups and general applicability.¹⁸⁹ The Pd-catalyzed coupling reaction of organic/aromatic halides with olefins allows a one-step synthesis of aromatic olefins that are used extensively as biologically active compounds, natural products, pharmaceuticals and precursors of conjugated polymers.^190,191 Heck reaction proceeds in the presence of homogeneous as well as heterogeneous Pd catalysts. Ce_0.98Pd_0.02O_1.98 synthesized by the solution combustion method catalyzes C–C cross-coupling of substituted iodobenzene, bromobenzene and bromoheteroaryls with alkyl acrylates and styrenes with good yields. There is no conversion when iodobenzene with methyl acrylate is passed over pure CeO₂, indicating that Pd²⁺ ions in Ce_0.98Pd_0.02O_1.98 are the active sites for Heck reaction.¹⁹² It has been observed that C–X (X = Cl, Br, and I) bond strength influences the activity towards Heck coupling reaction as the coupling reaction follows the order: C–I > C–Br > C–Cl. Temperature also affects the conversion for the coupling of iodobenzene and methyl acrylate that is presented in Fig. 12. Around 100% conversion is observed in 30 min at 150 °C, whereas the reaction requires nearly 1 h when temperature is 125 °C. However, the reaction is very slow at 80 °C, where conversion is only 60% even after 7 h. The Ce_0.98Pd_0.02O_1.98 catalyst can also be recycled without significant loss of catalytic activity, which indicates that the catalyst is stable under the reaction conditions. The observed turnover frequency (TOF) value for the reaction of iodobenzene and methyl acrylate is higher than for other Pd-substituted ligandless heterogeneous catalysts for C–C cross coupling reactions reported in the literature.

Fig. 12 Conversion efficiency of the reaction of iodobenzene and methyl acrylate over Ce_0.98Pd_0.02O_1.98 at different temperatures. Reprinted from ref. 192 with the permission of Springer.

11. Electrocatalysis and redox coupling between noble metal ions and CeO₂

Electrocatalysis has been studied by several electrochemical methods with regard to oxygen evolution, oxidation of formic acid and methanol oxidation over noble metal ionic catalysts, providing a better understanding of electronic interactions between noble metal ions and CeO₂ through electron exchange among redox ions, rather than gas–solid interactions. In a potentiostatic experiment (chronoamperometry) combined with extensive XPS studies, a pure CeO₂ electrode shows almost complete (70%) reduction when it is subjected to an oxygen evolution potential (1.2 V) for 1000 s.¹⁹³ The envelope of the Ce3d core level spectrum of the CeO₂ electrode looks more similar to Ce₂O₃ after the chronoamperometry experiment. In an another experiment under potentiodynamic conditions (cyclic voltammetry), when the CeO₂ electrode is cycled in the potential range of 0–1.2 V, Ce⁴⁺ is reduced to Ce³⁺, where the Ce³⁺ content is 75%. Though the fluorite structure is retained after the electrochemical experiment, as demonstrated by XRD patterns, this reduced CeO₂ does not oxidize back under any electrochemical conditions. On the other hand, the Ce_0.98Pt_0.02O_2−δ electrode shows partial reduction (38%) of Ce⁴⁺ to Ce³⁺ in a chronoamperometry experiment at an oxygen evolution potential (1.2 V) for 1000 s and Pt remains in the +2 and +4 oxidation states. The Ce⁴⁺ species in the Ce_0.98Pt_0.02O_2−δ electrode material is reduced to 50% Ce³⁺ after stopping the cycle at 1.2 V in a cyclic voltammetry experiment. However, it returns to its initial status when the Ce_0.98Pt_0.02O_2−δ electrode is cycled under potentiodynamic conditions in the positive potential range of 0–1.2 V with a particular scan rate. A similar phenomenon is also observed for Pt component species present in the Ce_0.98Pt_0.02O_2−δ electrode. There is not much difference between the Ce and Pt component ratios with an increase in scan rates. Thus, the Ce_0.98Pt_0.02O_2−δ electrode is stable to redox cycles in the potential range of 0–1.2 V. Similar evidence of redox reactions is observed in Ce_0.98Pd_0.02O_2−δ.¹⁹⁴ The catalytic redox properties of Ce_0.98Pt_0.02O_2−δ and Ce_0.98Pd_0.02O_2−δ electrodes is associated with the interaction between Pt⁴⁺/Pt²⁺/Pt⁰ and Ce⁴⁺/Ce³⁺, Pd²⁺/Pd⁰ and Ce⁴⁺/Ce³⁺ redox couples within the oxide catalysts. The redox ability of Cu²⁺/Cu⁺ and Cu⁺/Cu⁰ couples in Cu/CeO₂ catalysts has been studied using cyclic voltammetry to understand the catalytic redox reactions.^195–197 The catalytic activities of Ti_0.99M_0.01O_2−δ (M = Pd, Pt, Rh and Ru) catalysts are correlated with their cyclic voltammetry studies.⁵⁴

Substitution of the Pt²⁺ and Pd²⁺ ions in CeO₂ by sonochemical method activates the lattice oxygen and hydrogen spillover or a higher H/Pt ratio ∼ 8.1 and a H/Pd ratio ∼ 4.2 is observed. Ce_0.95Pt_0.05O_2−δ reversibly releases 0.203 mol of [O] per mol of compound below −15 °C and Ce_0.95Pd_0.05O_2−δ releases 0.105 mol of [O] per mol of compound below 55 °C. The reversible nature of the higher oxygen storage capacity or the higher H/Pt and H/Pd ratios is due to the interaction of redox couples such as Pt⁴⁺/Pt²⁺ (0.91 V), Pt²⁺/Pt⁰ (1.18 V), Pd²⁺/Pd⁰ (0.92 V), and Ce⁴⁺/Ce³⁺ (1.61 V). Because of the participation of lattice oxygen through the reduction of Ce⁴⁺ to Ce³⁺, Ce_0.95Pt_0.05O_2−δ and Ce_0.95Pd_0.05O_2−δ have shown higher electrooxidation of methanol compared to the same molar quantities of Pt in 5% Pt/C.⁶⁹

It has been observed that Pt²⁺ ion substitution in CeO₂ and Ce_1−xTi_xO₂ improves the electrocatalytic activity of the oxygen evolution reaction.¹⁹⁸ In a cyclic voltammetry experiment, bubbles can be observed on the Ce_0.98Pt_0.02O_2−δ electrode surface above 1.1 V, which confirms the oxygen evolution. The CeO₂ electrode shows bubbling only above 1.4 V, which confirms the catalytic effect of Pt²⁺ in CeO₂. Steady state current density values obtained from an 1800 s chronoamperometry experiment are 0.09, 0.15 and 0.31 mA cm⁻² for CeO₂, Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ, respectively. From XPS studies, it has been demonstrated that oxygen evolution is directly related to the extent of Ce⁴⁺ reduction and the stability of CeO₂. The oxygen evolution current increases with a decrease in Ce⁴⁺ reduction. Pt ions in CeO₂ and Ce_0.83Ti_0.15O₂ control the lattice oxygen in a particular way to impart the activity in Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ.

Electrooxidation of formic acid has been investigated over Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ catalysts and the results are compared with the Pt/C electrocatalyst, where Pt is in the metallic state. In cyclic voltammetry experiments, a CO poisoning effect is observed on Pt/C, which is eliminated in Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ catalysts. In chronoamperometry experiments, a gain in the oxidation current in the ionic catalysts is 10 times more than the corresponding Pt/C catalysts for the same amount of Pt metal.¹⁹⁸ Pt²⁺/Pt⁴⁺ states remain the same after cyclic voltammetry and chronoamperometry experiments in Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ, as demonstrated by XPS studies of the electrode materials. The extents of reduction of Ce⁴⁺ are 50% and 25% in Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ, respectively, indicating that lattice oxygen present in the ionic catalysts is utilized initially and a steady state is reached with equilibrium concentrations of Pt²⁺/Pt⁴⁺ and Ce⁴⁺/Ce³⁺ states. The involvement of lattice oxygen during electrooxidation of formic acid prevents CO poisoning of the catalysts. Similar behavior by Ce_0.98Pt_0.02O_2−δ and Ce_0.83Ti_0.15Pt_0.02O_2−δ catalysts toward methanol oxidation has been observed and their activities are much higher than the same amount of Pt⁰ in Pt/C, which is attributed to interaction of the Pt²⁺ ion with CeO₂ and Ce_0.83Ti_0.15O₂ activating their lattice oxygen.¹⁹⁸

12. Mechanism of redox reactions over noble metal ionic catalysts

Employing several characterization techniques, it has been demonstrated that there are two independent sites on the noble metal ionic catalysts, which adsorb reducing and oxidizing molecules during redox-type reactions such as CO + O₂, NO + CO, NO + H₂, NO + NH₃, WGS reaction, CO-PROX reaction, H₂ + O₂ recombination, photocatalytic reduction of NO and Heck reaction. A dual site mechanism for redox reactions, such as CO + O₂ and NO + CO, based on the structure of Ce_1−xM_xO_2−δ and Ti_1−xM_xO_2−δ-related catalysts has been proposed that is different from Langmuir–Hinshelwood mechanism on a noble metal surface. In Fig. 13, the CO + O₂ reaction mechanism over a Pt metal surface is compared with that over Ce_1−xM_xO_2−δ. It is well established that the CO + O₂ reaction over Pt follows the Langmuir–Hinshelwood mechanism, where both CO and O₂ are adsorbed on the Pt metal surface. CO is a polar molecule. CO has a triple bond as it is isoelectronic to nitrogen. The third bond is a co-ordinate-covalent bond due to lone pair donation from oxygen. Thus, the carbon end of CO is relatively negative. Therefore, carbon can donate electrons to the Pt metal. Bonding of CO with Pt is described by σ-donation to the Pt metal atom from its 5σ orbital and back-donation of the Pt metal d electron to the antibonding orbital of CO. A weak chemical bond is formed between CO and Pt metal and therefore CO adsorption is a chemisorption process. Thus, adsorption of CO makes Pt metal atoms negative as a whole. The electrons received by the Pt atoms on the surface from CO adsorption are shared by all the atoms in the particle; these are referred to as delocalized electrons in modern scientific terms. On the other hand, oxygen being electronegative, it tends to acquire electrons. The O=O bond length is known to be 1.21 Å. Unlike CO, O₂ is adsorbed horizontally on the Pt surface. The Pt–Pt distance in Pt metal is 2.74 Å, and when an oxygen atom aligns with the top of adjacent Pt atoms, the molecule is stretched from 1.21 to 2.71 Å, leading to weakening of the O=O bond of molecular O₂. Lengthening of O–O is favored when the anti-bonding orbital of the O₂ molecule receives electrons from Pt atoms. Transfer of electrons by Pt is possible because the Pt atoms have received electrons from CO. In this sense, the O₂ molecule is dissociatively adsorbed on the Pt metal surface. This eventually leads to dissociation of O₂ into two O atoms or partially negative ions. Once the O atom is created, it combines with the C end of CO, forming CO₂. In contrast, on the Pt ionic catalyst surface, CO is adsorbed on the electron-deficient metal ions, whereas O₂ is adsorbed onto the oxide ion vacancy because the vacancy site is activated by the electron-rich environment. The size of the oxide ion vacancy is ∼2.8 Å, which can accommodate an oxygen molecule of diameter 2.42 Å. Thus, there are two distinct sites, one for reducing and another for oxidizing molecules in the Pt metal ionic catalysts, which is different from the mechanism involving Pt metal atoms. Electron transfer from a reducing molecule to oxygen is facilitated by the lattice coupling between accessible Pt²⁺/Pt⁰ and Ce⁴⁺/Ce³⁺ redox couples. The enhanced activity is due to the creation of redox sites leading to site-specific adsorption and electronic interaction between the noble metal ions and the lattice. In addition to this, the lattice oxygen is activated in the catalyst with long M–O and Ce–O bonds and CO can be oxidized through a Mars–van Krevelen mechanism wherein the oxide ion is continuously consumed and formed. The difference in the rate of CO oxidation over different noble metal ion-substituted catalysts is due to differences in their redox properties.

Fig. 13 CO + O₂ reaction: (a) Langmuir–Hinshelwood mechanism on Pt metal and (b) dual site mechanism on Ce_1−xPt_xO_2−δ. Reprinted from ref. 21 with the permission of American Chemical Society.

A similar explanation can be given for the NO + CO reaction over noble metal ionic catalysts. NO is an electron donor as well as an acceptor molecule. Therefore, on the ionic catalysts, NO can be molecularly adsorbed on the noble metal ions and dissociatively chemisorbed on the vacant oxide ion sites. Adsorption of NO over noble metal ions is facilitated by the donation of the 5σ orbital electron of NO to the metal ion, leading to molecular adsorption. On the other hand, the anti-bonding 2π* orbital of NO obtains an electron from the oxide ion vacancy site where it is adsorbed, leading to the dissociation of NO. On the contrary, CO is specifically adsorbed on noble metal ion sites by donating its 5σ orbital electron to the metal ions. Molecularly adsorbed NO on the metal ion sites leads to N₂O formation, whereas N₂ is formed when NO is dissociatively adsorbed on the vacant oxide ion sites. Therefore, N₂ selectivity is governed by dual site adsorption of NO on the catalyst. Because both NO and CO compete for metal ion sites and NO is adsorbed on vacant oxide ion sites, overall NO reduction results in N₂ and N₂O formation. However, N₂O formed on the metal ion sites can move to the vacant oxide ion sites, making way for CO adsorption on metal ion sites and the reaction, N₂O + CO → N₂ + CO₂, takes over, leading to N₂ formation. Based on dual site adsorption (molecular adsorption on noble metal ion and dissociative adsorption on the vacant oxide ion), N₂ and N₂O formation along with CO₂ over NMICs such as the Ce_1−xPd_xO_2−δ catalyst during the NO + CO reaction are schematically shown in Fig. 14. A similar mechanism can be applied to the NO + CO reaction over Ti_0.99M_0.01O_2−δ catalysts. Again, it has been demonstrated that Ce_0.98Pd_0.02O_2−δ shows higher N₂ selectivity compared to Ce_0.98Rh_0.02O_2−δ.¹⁵⁵ Higher N₂O formation over Ce_0.98Rh_0.02O_2−δ at lower temperatures seems to be due to higher NO adsorption on Rh³⁺ sites. Furthermore, the Rh ion being in the +3 state, there are fewer oxide ion vacancies in Rh³⁺-substituted CeO₂ compared to Pd²⁺-substituted CeO₂. Therefore, the NO → N₂O conversion is facilitated as NO is more molecularly adsorbed on Rh³⁺ sites and less dissociatively adsorbed on the vacant sites of Ce_0.98Rh_0.02O_1.99 at low temperatures. However, at higher temperatures, NO as well as N₂O dissociation is increased and NO → N₂ takes place. In contrast, NO → N₂ conversion predominates on the Ce_0.98Pd_0.02O_1.98 catalyst over the entire range of temperatures.

Fig. 14 Mechanism of N₂, N₂O and CO₂ formation during the NO + CO reaction over Ce_1−xPd_xO_2−δ. Reprinted from ref. 23 with the permission of Springer.

For the NO + H₂ reaction over Ti_0.99M_0.01O_2−δ (M = Pd, Pt, Rh and Ru) catalysts, both NO and H₂ are adsorbed on noble metal ion sites and NO is also adsorbed on the oxide ion vacancy sites.⁵⁴ The formation of N₂O occurs because of the close proximity of the N atom of adsorbed NO and the N atom of adsorbed NO on oxide ion vacancy sites. The hydrogen atom from the noble metal ion site reacts with the oxygen in the vacant oxide ion site to form water. The N₂O molecule dissociates directly into N₂ and an O atom in oxide ion vacancy sites. NH₃ can also be formed by the combination of an adsorbed H atom with adsorbed NO in oxygen vacancy sites.

NH₃-TPD demonstrates that low temperature SCR activity of Ti_0.9Mn_0.1O_2−δ is due to the highest Brønsted acidity observed over this catalyst, whereas the highest Lewis acidity in Ti_0.9Fe_0.1O_2−δ corresponds to a wide SCR window.¹⁶³ Therefore, NH₃ can be adsorbed on the surface as NH₄⁺ or as molecular NH₃ in the cases of Brønsted acid sites and Lewis acid sites, respectively. NO can be adsorbed molecularly on the transition metal ion or it can dissociatively chemisorb in the oxide ion vacancy of the catalysts. The adsorbed NH₃ or NH₄⁺ can form an adduct with the adsorbed NO and the NH₂NO adduct dissociates into N₂ and H₂O. This path of dissociation dominates in the cases of Ti_0.9Fe_0.1O_2−δ and Ti_0.9Mn_0.05Fe_0.05O_2−δ. However, NH₂NO dissociating into N₂O is more likely to be favored over Ti_0.9Mn_0.1O_2−δ. Therefore, a “dual site-redox” mechanism occurs on these catalysts' surfaces. However, if the adsorbed NH₃ reacts with the oxygen to form a NH₂OO adduct, it produces NO and the SCR window becomes narrow. This seems to be the mechanism for Ti_0.9Cu_0.1O_2−δ. Thus, it has been shown that the SCR reaction occurs at the lowest temperature with the highest N₂ selectivity over the catalysts that have Brønsted and Lewis acid sites for NH₃ adsorption as well as oxide ion vacancies for NO dissociation.

With regard to the WGS reaction, it has been proposed that CO is adsorbed on noble metal ionic sites, whereas H₂O is dissociatively adsorbed over oxide ion vacancy sites.¹⁹⁹ The surface reaction between two adsorbed species results in the formation of products that restore the catalyst upon desorption. From kinetic rate expression, it has been demonstrated that the WGS reaction over noble metal ionic catalysts does not follow the Eley–Rideal and Langmuir–Hinshelwood mechanisms. Apparent activation energies are found to be less, considering the dual site mechanism for the WGS reaction over NMICs.

In case of photocatalytic reduction of NO over Ti_0.99Pd_0.01O_1.98, the presence of Pd²⁺ ion sites for adsorption of CO, oxide ion vacancy sites and UV light are the primary requirements for the photodecomposition of NO. A detailed mechanism accounts for the fact that there is competitive adsorption of NO and CO on Pd²⁺ sites and a high rate of NO photodissociation on the vacant oxide ion site present on the catalyst surface enhances NO reduction and decomposition activity.⁵³

The Ce_0.98Pd_0.02O_1.98 catalyst has two distinct sites: Pd²⁺ ions and oxide ion vacancies. Electron-deficient Pd²⁺ sites are useful for the adsorption of donor molecules, whereas vacant oxide ion sites next to Pd²⁺ ions act as sites for the adsorption of acceptor molecules. In Heck reaction over this catalyst, olefin is adsorbed on Pd²⁺ ions, whereas oxide ion vacancies are nucleophilic sites that are ideal for the accommodation of –Br and –I ions of aryl halides.¹⁹² The elimination of HBr facilitated by the base leads to C–C coupling.

13. DFT studies of noble metal ionic catalysts for higher OSC, adsorption and catalytic properties

In the last few decades, an enormous amount of studies on the synthesis, structure and catalytic properties of CeO₂-related materials with different aspects have been carried out. It has been found that the OSC and catalytic activities of CeO₂-based materials are enhanced when CeO₂ is substituted with several divalent, trivalent and tetravalent metal ions. The substitution of noble metal ions into CeO₂ further augments these properties. In this sense, there should have been changes in the atomic environment of the parent CeO₂ structure after substitution of metal ions, leading to enhancement of the OSC and catalytic activity. The OSC of CeO₂ is observed to be enhanced by the substitution of a Zr⁴⁺ ion for the Ce⁴⁺ site in CeO₂.^{117,118,140,200} As ZrO₂ itself cannot be reduced by H₂, this observation is surprising. OSC studies demonstrate that the reducibility of CeO₂ is increased by Ti⁴⁺ substitution in Ce_1−xTi_xO_2−δ.⁴⁷ Again, Pt ion substitution in Ce_1−xTi_xO₂ enhances its reducibility further compared to Ce_1−xTi_xO₂ and Ce_1−xPt_xO_2−δ. The rate of CO conversion in the CO + O₂ reaction over Pt ion-substituted Ce_1−xTi_xO₂ is observed to be much higher than with Ce_1−xPt_xO_2–δ.⁴⁷ Therefore, several questions can be raised about these findings. In this regard, what is the structural and chemical origin of the increased OSC in transition-metal-ion-substituted CeO₂? And what is the effect of noble metal substitution on the structural and chemical origins of the higher OSC and the high catalytic activities of CeO₂? A simulation of the crystal structure of CeO₂-based catalysts and probe reactions with these catalysts employing DFT calculations, has provided answers to these questions.^{50,51,201–223}

The first conception of the activation of lattice oxygen of CeO₂ by the substitution of a smaller ion originated from the substitution of a Zr⁴⁺ ion into a CeO₂ lattice in Ce_1−xZr_xO₂ (x = 0.0–0.5) solid solutions.¹⁴⁰ The structure can be simulated by DFT calculations, considering a lattice containing 32Ce⁴⁺ ions and 64O²⁻ ions in their crystal positions in a fluorite lattice. CeO₂ crystallizes in the fluorite structure, where the Ce⁴⁺ ion is inside a cube formed by eight oxide ions or the coordination number of Ce⁴⁺ is 8. Each O²⁻ is coordinated with four Ce⁴⁺ ions. Because of the smaller ionic radius of Zr⁴⁺ (0.84 Å) compared to Ce⁴⁺ (0.97 Å), the lattice parameter decreases when a Zr⁴⁺ ion is substituted into the CeO₂ lattice. Ce⁴⁺ and Zr⁴⁺ ions occupy alternate cubes in the Ce₁₆Zr₁₆O₆₄ ideal lattice. When the smaller Zr⁴⁺ ion occupies the place of the Ce⁴⁺ ion, the Zr⁴⁺ ion does not have complete contact with the 8 oxide ions. Accordingly, in the [ZrO₈] cube, the alternate four oxide ions come closer to the Zr⁴⁺ ion, forming a tetrahedron with a shortened Zr–O distance and the other four oxide ions also form a tetrahedron with longer Zr–O distances. Distortion in the ZrO₈ cube will induce distortion in the CeO₈ cube, also leading to 4 + 4 coordination that creates long Ce–O and short Ce–O bonds compared to 2.34 Å in pure CeO₂. The longer Ce–O bond is naturally weaker. There is a distribution of bond lengths from the mean Ce–O values when the Zr⁴⁺ ion is substituted.²⁴ Longer Ce–O bonds are indeed weaker and therefore they can be more easily extracted by H₂, leading to a higher OSC.¹⁴⁰ Similarly, XAFS studies and DFT calculations show the presence of short and long Ce–O and Ti–O bonds in Ce_1−xTi_xO₂, giving strongly and weakly bound oxygens.¹⁴¹ The stronger bonds have a valency greater than 2 and the weaker ones have a valency less than 2 and weakly-bound oxygen has been found to be responsible for a higher OSC in these mixed oxides.

It has been demonstrated that CeO₂ doped with several transition, noble and rare earth metals undergoes structural distortion that results in a higher oxygen storage capacity.^{50,51,201–203} On the other hand, this type of situation cannot be found in pure CeO₂ and TiO₂, therefore these pure oxides show a lower OSC compared to mixed oxides. For example, structural simulations of CeO₂, Ce_0.8Sn_0.2O₂ and Ce_0.78Sn_0.2Pd_0.02O_2−δ have been carried out by DFT calculations, using their respective Ce₄O₈, Ce₂₆Sn₆O₆₄ and Ce₂₅Sn₆Pd₁O₆₃ supercells.⁵⁰ Bond length distributions of Ce₄O₈, Ce₂₆Sn₆O₆₄ and Ce₂₅Sn₆Pd₁O₆₃ supercells are shown in Fig. 15. The optimized Ce₄O₈ supercell has all the Ce–O distances equal to 2.34 Å. In contrast, Ce–O bonds are distributed in shorter and longer bonds over the range of 2.16–2.54 Å in the Ce₂₆Sn₆O₆₄ supercell. Sn–O and Pd–O bonds become longer in Ce_0.78Sn_0.2Pd_0.02O_2−δ with oxygen vacancies where oxygen sites are highly activated and can easily be removed from the lattice, leading to a higher OSC. With an increase in the Sn content, a larger number of longer Sn–O bonds is obtained in the Ce₁₆Sn₁₆O₆₄ supercell related to Ce_0.5Sn_0.5O₂ in comparison with the Ce₂₆Sn₆O₆₄ supercell, thus, Ce_0.5Sn_0.5O₂ shows a higher OSC in relation to Ce_0.8Sn_0.2O₂. With the introduction of Pd²⁺ ions in Ce_0.5Sn_0.5O₂, Ce–O and Sn–O bonds exhibit a wider distribution in the Ce₁₆Sn₁₅Pd₁O₆₃ supercell and here Pd–O bonds are longer than in the Ce₂₅Sn₆Pd₁O₆₃ supercell. Similarly, substitution of Pd²⁺ in Ce_1−xFe_xO_2−δ enhances the OSC and the catalytic activity.⁵¹ It has also been observed that transition metal ion substitution in CeO₂ greatly enhances its reducibility in Ce_1−xM_xO_2−δ solid solution, leading to a higher OSC due to longer Ce–O and M–O bonds, whereas rare earth metal ion substitution has little effect on it.²⁰² The substitution of noble metal ions further enhances the OSC due to the presence of still longer bonds. It has also been demonstrated that Pt²⁺, Pd²⁺ and Cu²⁺ ions are stabilized in the CeO₂ matrix and CO is adsorbed only on Pt²⁺, Pd²⁺ and Cu²⁺ sites present on the CeO₂ surface.²⁰⁴

Fig. 15 Bond length distributions in (a) Ce₄O₈, (b) Ce₂₆Sn₆O₆₄ and (c) Ce₂₅Sn₆Pd₁O₆₃ supercells from DFT calculations. Reprinted from ref. 50 with the permission of American Chemical Society.

Both structural and electronic factors help to make CeO₂ more reducible in Pd²⁺, Pt²⁺ and Rh³⁺-doped CeO₂ than in pure CeO₂ and Zr⁴⁺-doped CeO₂, as shown by Yang et al.²⁰⁵ Wang and coworkers have proposed a model to understand the oxygen vacancy formation in a series of Ce_1−xZr_xO₂ materials, which consists of bond energy (E_bond) and relaxation energy (E_relax).²⁰⁶ It has been found that the relaxation energy plays a vital role in affecting the oxygen vacancy formation energy and Ce_0.5Zr_0.5O₂ possesses the lowest oxygen vacancy formation energy, which is the origin of its excellent OSC performance. Metiu and coworkers have demonstrated that doping of Pt in a CeO₂ surface weakens the bonds of oxygen atoms in the neighborhood by lowering the formation energy of the oxygen vacancy. The easier specific oxygen atoms are to remove, the more reactive they are when exposed to a reductant.⁷⁷ This works in a similar way for other doped CeO₂ materials.²⁰⁹ Au³⁺-substituted CeO₂ is more active for CO oxidation than a single Au⁺ species supported on stoichiometric CeO₂, as demonstrated by Camellone and Fabris with DFT studies.²¹⁰ This higher catalytic activity is due to the formation of oxygen vacancies and the interplay between reversible Au³⁺/Au⁺ and Ce⁴⁺/Ce³⁺ reductions. Chen et al. have found that substitution of Fe, Ru, Os, Sm, and Pu in CeO₂ results in activated oxygen in Ce_1−xM_xO₂ compared to pure CeO₂, due to its structural and electronic modifications and oxygen vacancy formation is lowered by doping of noble metals.²¹¹ Watson's group has shown with DFT calculations that increased OSC in Pt and Pd-doped CeO₂ is due to a large displacement of dopant ions from the Ce lattice site. Pd²⁺ and Pt²⁺ with d⁸ configuration move by ∼1.2 Å to adopt a square planar coordination with four lattice oxygens due to the crystal field effect.²¹² This large lattice distortion leaves three oxygen atoms under-coordinated or weakly bound in the vicinity of the dopant, which are more facile to remove than an oxygen atom in pure CeO₂. Similarly, several divalent dopant metal ions, such as Be²⁺, Mg²⁺, Ca²⁺, Ni²⁺, Cu²⁺ and Zn²⁺, distort to adopt the coordination of their own binary oxide, instead of cubically coordinated Ce in CeO₂. Depending on the electronic structure of the dopants, the different coordinations can create weakly or under-coordinated oxygen ions that are more easily removed than in pure CeO₂.²¹³ Recently, Nolan has shown that divalent Pd and Ni in CeO₂(111) and CeO₂(110) surfaces distort the local atomic environment, and the most stable structure for both dopants arises through compensation of the dopant +2 valency by oxygen vacancy formation.²¹⁴ Both Pd²⁺ and Ni²⁺ dopants lower the formation energy of the oxygen vacancy. In another study, it is found that strong interactions exist between the Pt and the surfaces, and Pt adsorption is stronger on the reduced surface than that on the unreduced surface.²¹⁵ A Pt–O–Ce bond may be formed between the Pt adatom and the CeO₂ surface. Thus, DFT calculations provide local structures, metal–oxygen distances, and density of valence states to understand OSC, adsorption and catalytic properties.

The OSC of Ti_0.97Pd_0.03O_1.97 is as high as 5100 μmol g⁻¹, which is more than 20 times higher than that of pure TiO₂. Oxygen is extracted by CO, resulting in the formation of CO₂ in the absence of feed oxygen, even at room temperature. The rate of CO oxidation is 2.75 μmol g⁻¹ s⁻¹ at 60 °C over Ti_0.97Pd_0.03O_1.97. Such a high catalytic activity is due to activated lattice oxygen created by the substitution of a Pd²⁺ ion, as observed from DFT calculations with 96 atom supercells of Ti₃₂O₆₄, Ti₃₁Pd₁O₆₃, Ti₃₀Pd₂O₆₂ and Ti₂₉Pd₃O₆₁ related to TiO₂, Ti_0.99Pd_0.01O_2−δ, Ti_0.98Pd_0.02O_2−δ and Ti_0.97Pd_0.03O_2−δ, respectively.⁵⁵ Ti–O bond distribution in the Ti₃₂O₆₄ supercell has two types of Ti–O bonds at 1.93 and 1.97 Å. The bond length of 1.93 Å corresponds to a Ti ion that is coordinated with four oxygen atoms in a planar geometry and other one is related to a Ti ion coordinating with two axial oxygen atoms in addition to 4 planar oxygens. Upon substitution of Pd²⁺, there is a distortion in the structure, resulting in metal–oxygen bonds with long and short distances compared to pure TiO₂. Four Pd–O bonds in the three Pd²⁺ ion-substituted compounds are in nearly square planar geometry with Pd–O short bonds between 1.97 and 2.0 Å. Ti–O bonds are distorted with reference to pure TiO₂ by the creation of longer and shorter Ti–O bonds. Similarly, by substitution of a Pt²⁺ ion in TiO₂, there is a large change in bond distribution, with short and long metal–oxygen bond lengths compared to Ti₃₂O₆₄.⁵⁶ In the Ti₃₁Pt₁O₆₃ supercell, the Pt²⁺ ion is coordinated with four oxygens in a square planar geometry having two Pt–O bonds of 1.98 Å and two Pt–O bonds of 2.00 Å. As a result of Pt²⁺ substitution, two shorter Ti–O bonds of 1.81 Å and one longer Ti–O bond of 2.18 Å occur, coordinating to the same oxygen atom. In Ti₃₀Pt₂O₆₂, there is a slight increase in the bond length distribution compared to that in Ti₃₁Pt₁O₆₃. Bond length distributions in the two different configurations of Ti₂₉Pt₃O₆₁ show a similar trend and these have slightly wider distributions compared to that in Ti₃₀Pt₂O₆₂. Chrétien and Metiu have demonstrated that doping TiO₂ with Cu, Ag, Au, Ni, Pd and Pt weakens the bonds of surface oxygens and oxides, making them better oxidation catalysts.²²⁴ Strong electron transfer from Ru to TiO₂(101) makes Ru particles positively charged, lowering the activation energy for CO dissociation, resulting in high catalytic activity toward CO₂ methanation.²²⁵

It has been established from our studies that noble metal ion-substituted CeO₂ and TiO₂ catalysts show a higher OSC as well as catalytic activity towards several important reactions. DFT calculations have been carried out to understand the causes behind the higher OSC and catalytic properties in CeO₂ and TiO₂-based materials. From our DFT studies, along with TPR and XAFS, it has been found that a change of structure at the atomic level is the reason for higher OSC in these mixed oxides. Because of the smaller size of substituent cations, surrounding oxygen ions are displaced from their original position and the oxygens at a larger distance are easily removed by reducing gases.

14. Conclusions and future outlook

The dispersion of noble metal ions over CeO₂ and TiO₂ by the solution combustion method, resulting in Ce_1−xM_xO_2−δ, Ce_1−x−yTi_xM_yO_2−δ, Ce_1−x−ySn_xM_yO_2−δ, Ce_1−x−yFe_xM_yO_2−δ and Ce_1−x−yZr_xM_yO_2−δ and Ti_1−xM_xO_2−δ (M = Pd, Pt, Rh and Ru) catalysts, the structure of these materials and their catalytic properties are documented here. In these materials, noble metals are present as ions and they are incorporated into reducible oxide matrices, such as CeO₂ and TiO₂, to a certain limit as the solution combustion method is a redox-reaction-type of preparation procedure. Hydrothermal and sonochemical methods have also been adopted to substitute aliovalent and isovalent noble metal ions in CeO₂ and TiO₂ matrices. Extensive experimental and theoretical studies done by our group and other research groups have established that these ionically dispersed noble metal catalysts are much more active than conventional finely-dispersed noble metal particles toward several important reactions such as CO oxidation, NO reduction by CO, H₂ and NH₃, selective catalytic reduction of NO, hydrocarbon oxidation, water gas shift reaction, CO-PROX reaction, H₂–O₂ reaction and hydrogenation. The idea of dispersing noble metals as ions can be extended to other reducible transition metal oxide supports such as SnO₂, V₂O₅, WO₃ and MoO₃. These catalysts can be probed for several catalytic reactions, obtaining more insights into metal–support interactions, redox properties and synergistic effects. In general, extensive studies on these NMIC materials for several heterogeneous catalytic reactions that are relevant to present-day requirements, other than exhaust catalytic reactions, are a worthwhile area to investigate on the basis of solid state chemistry view point.

Acknowledgements

We thank our past and present laboratory members and collaborators for their significant contributions. We thank the Department of Science and Technology (DST), Government of India, for financial support. MSH is grateful to the Council of Scientific and Industrial Research (CSIR), Government of India, for an Emeritus Professor Fellowship.

References

1. J. T. Richardson, Principles of Catalyst Development, Plenum Press, New York, 1989.
2. J. M. Thomas and W. J. Thomas, Principles and Practice of Heterogeneous Catalysis, Wiley-VCH, Weinheim, 2nd edn, 2015.
3. I. Chorkendorff and J. W. Niemantsverdriet, Concepts of Modern Catalysis and Kinetics, Wiley-VCH, Weinheim, 2003.
4. J. A. Anderson and M. F. García, Supported Metals in Catalysis, Imperial College Press, London, 2005.
5. J. Hagen, Industrial Catalysis, Wiley-VCH, Weinheim, 2006.
6. G. Rothenberg, Catalysis: Concepts and Green Applications, Wiley-VCH, Weinheim, 2008.
7. J. R. H. Ross, Heterogeneous Catalysis: Fundamentals and Applications, Elsevier, Oxford, 2012.
8. R. A. Sheldon and H. van Bekkum, Fine Chemicals through Heterogeneous Catalysis, Wiley-VCH, Weinheim, 2001.
9. Nanoparticles and Catalysis, ed. D. Astruc, Wiley-VCH, Weinheim, 2008.
10. G. C. Bond, Catalysis by Metals, Academic Press, New York, 1962.
11. J. R. Anderson, Structure of Metallic Catalysts, Academic Press, London, 1975.
12. H. S. Gandhi, G. W. Graham and R. W. McCabe, J. Catal., 2003, 216, 433.
13. J. C. Summers and S. A. Ausen, J. Catal., 1979, 58, 131.
14. H. C. Yao, H. S. Gandhi and M. Shelef, Metal–Support and Metal–Additive Effects in Catalysis, ed. B. Imelik, C. Naccache, G. Coudurier, H. Praliaud, P. Meriaudeau, P. Gallezot, G. A. Martin and J. C. Vedrine, Elsevier, Amsterdam, 1982, p. 159.
15. H. C. Yao and Y. F. Y. Yao, J. Catal., 1984, 86, 254.
16. G. Munuera, A. Fernandez and A. R. Gonzalez-Elipe, Stud. Surf. Sci. Catal., 1991, 71, 207.
17. J. G. Nunan, H. J. Robota, M. J. Cohn and S. A. Bradley, Stud. Surf. Sci. Catal., 1991, 71, 221.
18. J. Kašpar, P. Fornasiero and N. Hickey, Catal. Today, 2003, 77, 419.
19. T. X. T. Sayle, S. C. Parker and C. R. A. Catlow, J. Phys. Chem., 1994, 98, 13625.
20. P. Bera, K. C. Patil, V. Jayaram, G. N. Subbanna and M. S. Hegde, J. Catal., 2000, 196, 293.
21. M. S. Hegde, K. C. Patil and G. Madras, Acc. Chem. Res., 2009, 42, 704.
22. P. Bera and M. S. Hegde, J. Indian Inst. Sci., 2010, 90, 299.
23. P. Bera and M. S. Hegde, Catal. Surv. Asia, 2011, 15, 181.
24. M. S. Hedge and P. Bera, Catal. Today, 2015, 253, 40.
25. A. Corma and H. Garcia, Chem. Soc. Rev., 2008, 37, 2096.
26. J. C. Fierro-Gonzalez and B. C. Gates, Chem. Soc. Rev., 2008, 37, 2127.
27. M. Flytzani-Stephanopoulos and B. C. Gates, Annu. Rev. Chem. Biomol. Eng., 2012, 3, 545.
28. F. J. Janssen, Handbook of Heterogeneous Catalysis, ed. G. Ertl, H. Knözinger and J. Weitkamp, VCH, Weinheim, 1997, vol. 1, p.191.
29. R. L. Augustine, Heterogeneous Catalysis for the Synthetic Chemist, Marcel Dekker, Inc., New York, 1996, p. 267.
30. M. A. Cauqui and J. M. Rodríguez-Izquierdo, J. Non-Cryst. Solids, 1992, 147–148, 724.
31. P. Turlier, H. Praliaud, P. Moral, G. A. Martin and J. A. Dalmon, Appl. Catal., 1985, 19, 286.
32. A. Hadi and I. I. Yaacob, Mater. Lett., 2007, 61, 93.
33. T. Masui, K. Fujiwara, Y. Penga, T. Sakatab, K. Machida, H. Mori and G. Adachi, J. Alloys Compd., 1998, 269, 116.
34. A. Suda, T. Kandori, N. Terao, Y. Ukyo, H. Sobukawa and M. Sugiura, J. Mater. Sci. Lett., 1998, 17, 89.
35. S. Enzo, F. Delogu, R. Frattini, A. Primavera and A. Trovarelli, J. Mater. Res., 2000, 15, 1538.
36. K. C. Patil, M. S. Hegde, T. Rattan and S. T. Aruna, Chemistry of Nanocrystalline Oxide Materials: Combustion Synthesis, Properties and Applications, World Scientific, Singapore, 2008.
37. W. Wen and J.-M. Wu, RSC Adv., 2014, 4, 58090.
38. A. G. Merzhanov, SHS Research and Development Handbook, Chernogolovka, Russia, Russian Academy of Sciences, 1999.
39. K. C. Patil, S. T. Aruna and S. Ekabaram, Curr. Opin. Solid State Mater. Sci., 1997, 2, 158.
40. K. C. Patil, S. T. Aruna and T. Mimani, Curr. Opin. Solid State Mater. Sci., 2002, 6, 507.
41. S. L. González-Cortés and F. E. Imbert, Appl. Catal., A, 2013, 452, 117.
42. P. Bera, K. C. Patil, V. Jayaram, M. S. Hegde and G. N. Subbanna, J. Mater. Chem., 1999, 9, 1801.
43. K. R. Priolkar, P. Bera, P. R. Sarode, M. S. Hegde, S. Emura, R. Kumashiro and N. P. Lalla, Chem. Mater., 2002, 14, 2120.
44. A. Gayen, K. R. Priolkar, P. R. Sarode, V. Jayaram, M. S. Hegde, G. N. Subbanna and S. Emura, Chem. Mater., 2004, 16, 2317.
45. P. Bera, K. C. Patil and M. S. Hegde, Phys. Chem. Chem. Phys., 2000, 2, 3715.
46. P. Bera and M. S. Hegde, Catal. Lett., 2002, 79, 75.
47. T. Baidya, A. Gayen, M. S. Hegde, N. Ravishankar and L. Dupont, J. Phys. Chem. B, 2006, 110, 5262.
48. T. Baidya, A. Marimuthu, M. S. Hegde, N. Ravishankar and G. Madras, J. Phys. Chem. C, 2007, 111, 830.
49. T. Baidya, A. Gupta, P. A. Deshpandey, G. Madras and M. S. Hegde, J. Phys. Chem. C, 2009, 113, 4059.
50. A. Gupta, M. S. Hegde, K. R. Priolkar, U. V. Waghmare, P. R. Sarode and S. Emura, Chem. Mater., 2009, 21, 5836.
51. A. Gupta, A. Kumar, U. V. Waghmare and M. S. Hegde, Chem. Mater., 2009, 21, 4880.
52. T. Baidya, G. Dutta, M. S. Hegde and U. V. Waghmare, Dalton Trans., 2009, 455.
53. S. Roy, M. S. Hegde, N. Ravishankar and G. Madras, J. Phys. Chem. C, 2007, 111, 8153.
54. S. Roy, M. S. Hegde, S. Sharma, N. P. Lalla, A. Marimuthu and G. Madras, Appl. Catal., B, 2008, 84, 341.
55. B. D. Mukri, G. Dutta, U. V. Waghmare and M. S. Hegde, Chem. Mater., 2012, 24, 4491.
56. B. D. Mukri, U. V. Waghmare and M. S. Hegde, Chem. Mater., 2013, 25, 3822.
57. R. M. Barrer, Hydrothermal Chemistry of Zeolites, Academic Press, London, 1982.
58. W.-Q. Han, L. Wu and Y. Zhu, J. Am. Chem. Soc., 2005, 127, 12814.
59. Z. Y. Zhong, T.-P. Ang, J. Z. Luo, H.-C. Gan and A. Gedanken, Chem. Mater., 2005, 17, 6814.
60. W. Hung, P. Shuk and M. Greenblatt, Chem. Mater., 1997, 9, 2240.
61. P. Singh and M. S. Hegde, J. Solid State Chem., 2008, 181, 3248.
62. P. Singh, M. S. Hegde and J. Gopalakrishnan, Chem. Mater., 2008, 20, 7268.
63. P. Singh and M. S. Hegde, Chem. Mater., 2009, 21, 3337.
64. L. Yin, Y. Wang, G. Pang, Y. Koltypin and A. Gedanken, J. Colloid Interface Sci., 2002, 246, 78.
65. S. Bhattacharyya and A. Gedanken, Microporous Mesoporous Mater., 2008, 110, 553.
66. R. K. Selvan, A. Gedanken, P. Anilkumar, G. Manikandan and C. Karunakaran, J. Cluster Sci., 2009, 20, 291.
67. V. K. Rangari, D. N. Srivastava and A. Gedanken, Mater. Lett., 2006, 60, 3766.
68. S. Shanmugam and A. Gedanken, J. Phys. Chem. C, 2009, 113, 18707.
69. P. Singh and M. S. Hegde, Cryst. Growth Des., 2010, 10, 2995.
70. P. Singh and M. S. Hegde, Dalton Trans., 2010, 39, 10768.
71. P. Bera, A. Gayen, M. S. Hegde, N. P. Lalla, L. Spadaro, F. Frusteri and F. Arena, J. Phys. Chem. B, 2003, 107, 6122.
72. P. Bera, K. R. Priolkar, A. Gayen, P. R. Sarode, M. S. Hegde, S. Emura, R. Kumashiro, V. Jayaram and G. N. Subbanna, Chem. Mater., 2003, 15, 2049.
73. A. Gayen, T. Baidya, K. Biswas, S. Roy and M. S. Hegde, Appl. Catal., A, 2006, 315, 135.
74. P. Bera, S. T. Aruna, K. C. Patil and M. S. Hegde, J. Catal., 1999, 186, 36.
75. P. Bera, K. R. Priolkar, P. R. Sarode, M. S. Hegde, S. Emura, R. Kumashiro and N. P. Lalla, Chem. Mater., 2002, 14, 3591.
76. G. Li, L. Li, Y. Yuan, J. Shi, Y. Yuan, Y. Li, W. Zhao and J. Shi, Appl. Catal., B, 2014, 158–159, 341.
77. W. Tang, Z. Hua, M. Wang, G. D. Stucky, H. Metiu and E. W. McFarland, J. Catal., 2010, 273, 125.
78. B. Wang, D. Weng, X. Wu and J. Fan, Catal. Today, 2010, 153, 111.
79. A. Bisht, B. P. Gangwar, T. Anupriya and S. Sharma, J. Solid State Electrochem., 2014, 18, 197.
80. S.-Y. Wang, N. Li, R.-M. Zhou, L.-Y. Jin, G.-S. Hu, J.-Q. Lu and M.-F. Luo, J. Mol. Catal. A: Chem., 2013, 374–375, 53.
81. L. M. Misch, J. A. Kurzman, A. R. Derk, Y.-I. Kim, R. Seshadri, H. Metiu, E. W. McFarland and G. D. Stucky, Chem. Mater., 2011, 23, 5432.
82. E. M. Slavinskaya, R. V. Gulyaev, A. V. Zadesenets, O. A. Stonkus, V. I. Zaikovskii, Y. V. Shubin, S. V. Korenev and A. I. Boronin, Appl. Catal., B, 2015, 166–167, 91.
83. G. Li, L. Li and D. Jiang, J. Phys. Chem. C, 2015, 119, 12520.
84. A. V. Malyutin, A. I. Mikhailichenko, Y. V. Zubavichus, V. Y. Murzin, A. G. Koshkin and I. V. Sokolov, Kinet. Catal., 2015, 56, 89.
85. W. Miao, S. Meiqing, W. Jianqiang, W. Guangxi, L. Hang and W. Jun, J. Rare Earths, 2014, 32, 1114.
86. R. V. Gulyaev, E. M. Slavinskaya, S. A. Novopashin, D. V. Smovzh, A. V. Zaikovskii, D. Yu. Osadchii, O. A. Bulavchenko, S. V. Korenev and A. I. Boronin, Appl. Catal., B, 2014, 147, 32.
87. H.-H. Liu, Y. Wang, A.-P. Jia, S.-Y. Wang, M.-F. Luo and Z.-Q. Lu, Appl. Surf. Sci., 2014, 314, 725.
88. R. V. Gulyaev, D. Yu. Osadchii, S. V. Koscheev and A. I. Boronin, J. Struct. Chem., 2015, 56, 566.
89. F. Pilger, A. Testino, M. A. Lucchini, A. Kambolis, M. Tarik, M. E. Kazzi, Y. Arroyo, M. D. Rossell and C. Ludwig, J. Nanosci. Nanotechnol., 2015, 15, 3530.
90. Q. Fu, H. Saltsburg and M. Flytzani-Stephanopoulos, Science, 2003, 301, 935.
91. X. Tang, B. Zhang, Y. Li, Y. Xu, Q. Xin and W. Shen, Catal. Lett., 2004, 97, 163.
92. J. Lin, L. Zhang, Z. Wang, J. Ni, R. Wang and K. Wei, J. Mol. Catal. A: Chem., 2013, 366, 375.
93. M. Kurnatowska, W. Mista, P. Mazur and L. Kepinski, Appl. Catal., B, 2014, 148–149, 123.
94. L. Ge, T. Chen, Z. Liu and F. Chen, Catal. Today, 2014, 224, 209.
95. A. Karpenko, R. Leppelt, V. Plzak and R. J. Behm, J. Catal., 2007, 252, 231.
96. L. Wang, H. He, Y. Yu, L. Sun, S. Liu, C. Zhang and L. He, J. Inorg. Biochem., 2014, 135, 45.
97. M. S. Chen and D. W. Goodman, Top. Catal., 2007, 44, 41.
98. T. Baidya, K. R. Priolkar, P. R. Sarode, M. S. Hegde, K. Asakura, G. Tateno and Y. Koike, J. Chem. Phys., 2008, 128, 124711.
99. T. Baidya, P. Bera, B. D. Mukri, S. K. Parida, O. Kröcher, M. Elsener and M. S. Hegde, J. Catal., 2013, 303, 117.
100. T. Baidya and P. Bera, Catal. Struct. React.,  DOI:10.1179/2055075815Y.0000000004.
101. C. T. Driscoll, K. M. Driscoll, M. J. Mitchell and D. J. Raynal, Environ. Pollut., 2008, 123, 327.
102. K. Marilena and E. Castanas, Environ. Pollut., 2008, 151, 362.
103. A. J. Haagen-Smit, Ind. Eng. Chem., 1952, 44, 1342.
104. P. L. Magill and R. W. Benoliel, Ind. Eng. Chem., 1952, 44, 1347.
105. A. J. Haagen-Smit and M. M. Fox, Ind. Eng. Chem., 1956, 48, 1484.
106. F. E. Littman, H. W. Ford and N. Endow, Ind. Eng. Chem., 1956, 48, 1492.
107. E. R. Stephens, P. L. Hanst, R. C. Doerr and W. E. Scott, Ind. Eng. Chem., 1956, 48, 1498.
108. A. Troveralli, Catalysis by Ceria and Related Materials, Imperial College Press, London, 2002.
109. A. Troveralli, Catalysis by Ceria and Related Materials, Imperial College Press, London, 2nd edn, 2013.
110. M. V. Twigg, Philos. Trans. R. Soc., A, 2005, 363, 1013.
111. F. Klingstedt, K. Arve, K. Eränen and D. Y. Murzin, Acc. Chem. Res., 2006, 39, 273.
112. E. S. J. Lox, Handbook of Heterogeneous Catalysis, ed. G. Ertl, H. Knözinger, F. Schüth and J. Weitkamp, Wiley-VCH, Weinheim, 2nd edn, 2008, vol. 5, p. 2274.
113. P. Gabrielsson and H. G. Pedersen, Handbook of Heterogeneous Catalysis, ed. G. Ertl, H. Knözinger, F. Schüth and J. Weitkamp, Wiley-VCH, Weinheim, 2nd edn, 2008, vol. 5, p. 2345.
114. M. Bowker, Chem. Soc. Rev., 2008, 37, 2204.
115. P. Granger and V. I. Parvulescu, Chem. Rev., 2011, 111, 3155.
116. R. J. Farrauto and R. M. Heck, Catal. Today, 1999, 51, 351.
117. A. Troveralli, Catal. Rev.: Sci. Eng., 1996, 38, 439.
118. A. Troveralli, Comments Inorg. Chem., 1999, 20, 263.
119. P. Bera, Ph. D. thesis, Indian Institute of Science, Bangalore, India, 2002.
120. T. P. Kobylinski and B. W. Taylor, J. Catal., 1974, 33, 376.
121. G. L. Bauerle, S. C. Wu and K. Nobe, Ind. Eng. Chem. Prod. Res. Dev., 1975, 14, 123.
122. J. T. Kummer, J. Phys. Chem., 1986, 90, 4747.
123. H. Bartholomew, Ind. Eng. Chem. Prod. Res. Dev., 1975, 14, 29.
124. G. L. Bauerle and K. Nobe, Ind. Eng. Chem. Prod. Res. Dev., 1974, 13, 185.
125. M. Iwamoto, Stud. Surf. Sci. Catal., 1994, 84, 1395.
126. T. Seiyama, T. Arakawa, T. Matsuda, Y. Takita and N. Yamazoe, J. Catal., 1977, 48, 1.
127. T. Komatsu, M. Nunokawa, I. S. Moon, T. Takahara, S. Namba and T. Yashima, J. Catal., 1998, 148, 427.
128. R. T. Yang, T. J. Pinnavaia, W. Li and W. Zhang, J. Catal., 1997, 172, 488.
129. E. Ito, Y. J. Mergler, B. E. Nieuwenhuys, H. P. A. Calis, H. van Bekkum and C. M. van den Bleek, J. Chem. Soc., Faraday Trans., 1996, 92, 1799.
130. J. A. Martens, A. Cauvel, A. Francis, C. Hermans, F. Jayat, M. Remy, M. Keung, J. Lievens and P. A. Jacobs, Angew. Chem., Int. Ed., 1998, 37, 1901.
131. W. Fu, X.-H. Li, H.-L. Bao, K.-X. Wang, X. Wei, Y.-Y. Cai and J.-S. Chen, Sci. Rep., 2013, 3, 2349.
132. W. F. Libby, Science, 1971, 176, 499.
133. R. J. H. Voorhoeve, J. P. Remeika, P. E. Freeland and B. T. Matthias, Science, 1973, 177, 353.
134. R. Zhang, A. Villanueva, H. Alamdari and S. Kaliaguine, J. Catal., 2006, 237, 368.
135. F. Severino, J. L. Brito, J. Laine, J. L. G. Fierro and A. L. Agudo, J. Catal., 1998, 177, 82.
136. R. K. Herz, J. B. Klela and J. A. Sell, Ind. Eng. Chem. Prod. Res. Dev., 1983, 22, 387.
137. M. C. Kung, K. A. Bethke, J. Yan, J.-H. Lee and H. H. Kung, Appl. Surf. Sci., 1997, 121–122, 261.
138. R. Burch, Pure Appl. Chem., 1996, 68, 377.
139. G. Kim, Ind. Eng. Chem. Prod. Res. Dev., 1982, 21, 267.
140. G. Dutta, U. V. Waghmare, T. Baidya, M. S. Hegde, K. R. Priolkar and P. R. Sarode, Catal. Lett., 2006, 108, 165.
141. G. Dutta, U. V. Waghmare, T. Baidya, M. S. Hegde, K. R. Priolkar and P. R. Sarode, Chem. Mater., 2006, 18, 3249.
142. S. Roy, A. Marimuthu, M. S. Hegde and G. Madras, Appl. Catal., B, 2007, 73, 300.
143. M. E. Grass, Y. Zhang, D. R. Butcher, J. Y. Park, Y. Li, H. Bluhm, K. M. Bratlie, T. Zhang and G. A. Somorjai, Angew. Chem., Int. Ed., 2008, 47, 8893.
144. D. A. J. M. Ligthart, R. A. van Santen and E. J. M. Hensen, Angew. Chem., Int. Ed., 2011, 50, 5306.
145. I. Manuel, C. Thomas, C. Bourgeois, H. Colas, N. Matthess and G. Djéga-Mariadassou, Catal. Lett., 2001, 77, 193.
146. S. Carrettin, A. Corma, M. Iglesis and F. Sánchez, Appl. Catal., A, 2005, 291, 247.
147. A. M. Venezia, G. Pantaleo, A. Longo, G. Di Carlo, M. P. Casaletto, F. L. Liotta and G. Deganello, J. Phys. Chem. B, 2005, 109, 2821.
148. N. W. Cant, P. C. Hicks and B. S. Lennon, J. Catal., 1978, 54, 372.
149. Y.-F. Yu Yao, J. Catal., 1984, 87, 152.
150. S. H. Oh and C. C. Eickel, J. Catal., 1988, 112, 543.
151. E. Bekyarova, P. Fornasiero, J. Kašpar and M. Graziani, Catal. Today, 1998, 45, 179.
152. A. M. Venezia, L. F. Liotta, G. Pantaleo, V. L. Parola, G. Deganello, A. Beck, Zs. Koppány, K. Frey, D. Horváth and L. Guczi, Appl. Catal., A, 2003, 251, 359.
153. Y. Park, S. K. Kim, D. Pradhan and Y. Sohn, React. Kinet., Mech. Catal., 2014, 113, 85.
154. S. Roy, A. Marimuthu, M. S. Hegde and G. Madras, Appl. Catal., B, 2007, 71, 23.
155. S. Roy and M. S. Hegde, Catal. Commun., 2008, 9, 811.
156. P. Granger, C. Dathy, J. J. Lecomte, L. Leclercq, M. Prigent, G. Mabilon and G. Leclercq, J. Catal., 1998, 173, 304.
157. P. Granger, J. J. Lecomte, C. Dathy, L. Leclercq and G. Leclercq, J. Catal., 1998, 175, 194.
158. P. Granger, L. Delannoy, J. J. Lecomte, C. Dathy, H. Praliaud, L. Leclercq and G. Leclercq, J. Catal., 2002, 207, 202.
159. J. H. Holles, M. A. Switzer and R. J. Davis, J. Catal., 2000, 190, 247.
160. D. R. Rainer, S. M. Vesecky, M. Koranne, W. S. Oh and D. W. Goodman, J. Catal., 1997, 167, 234.
161. M. Schmal, M. A. S. Baldaza and M. A. Vannice, J. Catal., 1999, 185, 138.
162. S. Roy, A. Marimuthu, M. S. Hegde and G. Madras, Catal. Commun., 2008, 9, 101.
163. S. Roy, B. Viswanath, M. S. Hegde and G. Madras, J. Phys. Chem. C, 2008, 112, 6002.
164. S. Roy, A. Marimuthu, P. A. Deshpande, M. S. Hegde and G. Madras, Ind. Eng. Chem. Res., 2008, 47, 9240.
165. M. V. Twigg, Appl. Catal., B, 2007, 70, 2.
166. P. Avila, M. Montes and E. E. Miró, Chem. Eng. J., 2005, 109, 11.
167. V. Tomašić, Catal. Today, 2007, 119, 106.
168. T. A. Nijhuis, A. E. W. Beers, T. Vergunst, I. Hoek, F. Kapteijn and J. A. Moulijn, Catal. Rev.: Sci. Eng., 2001, 43, 345.
169. P. Jiang, G. Lu, Y. Guo, Y. Guo, S. Zhang and X. Wang, Surf. Coat. Technol., 2005, 190, 314.
170. S. Sharma and M. S. Hegde, Catal. Lett., 2006, 112, 69.
171. S. Sharma, M. S. Hegde, R. N. Das and M. Pandey, Appl. Catal., A, 2008, 337, 130.
172. P. Bera, K. C. Patil and M. S. Hegde, Phys. Chem. Chem. Phys., 2000, 2, 273.
173. P. Bera, S. Malwadkar, A. Gayen, C. V. V. Satyanarayana, B. S. Rao and M. S. Hegde, Catal. Lett., 2004, 96, 213.
174. S. Sharma, P. A. Deshpande, M. S. Hegde and G. Madras, Ind. Eng. Chem. Res., 2009, 48, 6535.
175. P. A. Deshpande, M. S. Hegde and G. Madras, Appl. Catal., B, 2010, 96, 83.
176. A. Gupta and M. S. Hegde, Appl. Catal., B, 2010, 99, 279.
177. N. Mahadevaiah, P. Singh, B. D. Mukri, S. K. Parida and M. S. Hegde, Appl. Catal., B, 2011, 108–109, 117.
178. P. Singh, N. Mahadevaiah, S. K. Parida and M. S. Hegde, J. Chem. Sci., 2011, 123, 577.
179. A. Martínez-Arias, D. Gamarra, M. Fernández-García, A. Hornés, P. Bera, Z. Koppány and Z. Scahy, Catal. Today, 2009, 143, 211.
180. D. Teschner, A. Wootsch, O. Pozdnyakova-Tellinger, J. Kröhnert, E. M. Vass, M. Hävecker, S. Zafeiratos, P. Schnörch, P. C. Jentoft, A. Knop-Gericke and R. Schlögl, J. Catal., 2007, 249, 318.
181. S. Sharma, A. Gupta and M. S. Hegde, Catal. Lett., 2010, 134, 330.
182. P. Bera, M. S. Hegde and K. C. Patil, Curr. Sci., 2001, 80, 1576.
183. S. Sharma and M. S. Hegde, ChemPhysChem, 2009, 10, 637.
184. B. Hariprakash, P. Bera, S. K. Martha, S. A. Gaffoor, M. S. Hegde and A. K. Shukla, Electrochem. Solid-State Lett., 2001, 4, A23.
185. R. Mistri, J. Llorca, B. C. Ray and A. Gayen, J. Mol. Catal. A: Chem., 2013, 376, 111.
186. K. Nagaveni, G. Sivalingam, A. Gayen, G. Madras and M. S. Hegde, Catal. Lett., 2003, 88, 73.
187. A. de Meijere and F. E. Meyer, Angew. Chem., Int. Ed. Engl., 1995, 33, 2379.
188. A. P. Beletskaya and V. Cheprakov, Chem. Rev., 2000, 100, 3009.
189. G. T. Crisp, Chem. Soc. Rev., 1998, 27, 427.
190. J. G. de Vries, Can. J. Chem., 2001, 79, 1086.
191. T. R. Burke, D. G. Liu and Y. J. Gao, J. Org. Chem., 2000, 65, 6288.
192. S. R. Sanjaykumar, B. D. Mukri, S. Patil, G. Madras and M. S. Hegde, J. Chem. Sci., 2011, 123, 47.
193. S. Sharma and M. S. Hegde, J. Chem. Phys., 2009, 130, 114706.
194. S. Sharma, B. D. Mukri and M. S. Hegde, Dalton Trans., 2011, 40, 11480.
195. P. Bera, S. Mitra, S. Sampath and M. S. Hegde, Chem. Commun., 2001, 927.
196. X. Tang, B. Zhang, Y. Li, Y. Xu, Q. Xin and W. Shen, Catal. Today, 2004, 93–95, 191.
197. L. Li, Y. Zhan, Q. Zheng, Y. Zheng, X. Lin, D. Li and J. Zhu, Catal. Lett., 2007, 118, 91.
198. S. Sharma, P. Singh and M. S. Hegde, J. Solid State Electrochem., 2011, 15, 2185.
199. P. A. Deshpande, M. S. Hegde and G. Madras, AIChE J., 2010, 56, 1315.
200. R. Di Monte, P. Fornasiero, M. Graziani and J. Kašpar, J. Alloys Compd., 1998, 275, 877.
201. G. Dutta, U. V. Waghmare, T. Baidya and M. S. Hegde, Chem. Mater., 2007, 19, 6430.
202. A. Gupta, U. V. Waghmare and M. S. Hegde, Chem. Mater., 2010, 22, 5184.
203. A. Gupta, A. Kumar, U. V. Waghmare and M. S. Hegde, J. Chem. Phys., 2010, 132, 194702.
204. G. Dutta, A. Gupta, U. V. Waghmare and M. S. Hegde, J. Chem. Sci., 2011, 123, 509.
205. Z. Yang, G. Luo, Z. Lu, T. K. Woo and K. Hermansson, J. Phys.: Condens. Matter, 2008, 20, 035210.
206. H.-F. Wang, X.-Q. Gong, Y.-L. Guo, Y. Guo, G. Z. Lu and P. Hu, J. Phys. Chem. C, 2009, 113, 10229.
207. V. Shapovalov and H. Metiu, J. Catal., 2007, 245, 205.
208. Z. Hu and H. Metiu, J. Phys. Chem. C, 2011, 115, 17898.
209. E. W. McFarland and H. Metiu, Chem. Rev., 2013, 113, 4391.
210. M. F. Camellone and S. Fabris, J. Am. Chem. Soc., 2009, 131, 10473.
211. H.-L. Chen, J.-G. Chang and H.-T. Chen, Chem. Phys. Lett., 2011, 502, 169.
212. D. O. Scanlon, B. J. Morgan and G. W. Watson, Phys. Chem. Chem. Phys., 2011, 13, 4279.
213. A. B. Kehoe, D. O. Scanlon and G. W. Watson, Chem. Mater., 2011, 23, 4464.
214. M. Nolan, J. Mater. Chem., 2011, 21, 9160.
215. Y. Zhao, B.-T. Teng, X.-D. Wen, Y. Zhao, L.-H. Zhao and M.-F. Luo, Catal. Commun., 2012, 27, 63.
216. H.-T. Chen, J. Phys. Chem. C, 2012, 116, 6239.
217. Z. Yang, Z. Lu and G. Luo, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 075421.
218. Z. Lu and Z. Yang, J. Phys.: Condens. Matter, 2010, 22, 475003.
219. A. Bruix, K. M. Neyman and F. Illas, J. Phys. Chem. C, 2010, 114, 14202.
220. W. Song, C. Popa, A. P. J. Jansen and E. J. M. Hensen, J. Phys. Chem. C, 2010, 116, 22904.
221. J. Paier, C. Penschke and J. Sauer, Chem. Rev., 2013, 113, 3949.
222. X.-Q. Gong, L.-L. Yin, J. Zhang, H.-F. Wang, X.-M. Cao, G. Lu and P. Hu, Adv. Chem. Eng., 2014, 44, 1.
223. A. D. Mayernick and M. J. Janik, J. Catal., 2011, 278, 16.
224. S. Chrétien and H. Metiu, Catal. Lett., 2006, 107, 143.
225. F. Wang, S. Zhang, C. Li, J. Liu, S. He, Y. Zhao, H. Yan, M. Wei, D. G. Evans and X. Duan, RSC Adv., 2014, 4, 10834.

---