CO oxidation by CeO₂–Al₂O₃–CeAlO₃ hybrid oxides

Abstract

A modified solution combustion technique was successfully used to synthesize sub-10 nm crystallites of hybrid CeO₂–Al₂O₃–CeAlO₃. The fuel in the solution combustion was tuned to obtain mixed oxides and solid solutions of the compound. The compounds were characterized by X-ray diffraction, transmission electron microscopy and X-ray photoelectron spectroscopy. XRD and TEM analysis showed the substitution of Al³⁺ ions in the CeO₂ matrix when a combination of glycine, urea, hexamine and oxalyl dihydrazide was used as fuel for the synthesis. The compounds showed high activity for CO oxidation and the activity of the compounds was dependent upon the composition of the oxide.

---

Introduction

Phase, size, microstructure and morphology are important characteristics of a material governing its various properties, especially the catalytic activity. The phase dependence of photocatalytic properties of TiO₂ is well reported.¹ Similarly, differences in the catalytic activity of different phases of ZrO₂ and Al₂O₃ have been also reported.^2,3 The effect of crystallite size on the catalytic activity of the material has been well established both by experimental as well as theoretical calculations.⁴ The influence of microstructure of the material on the properties has been reported by various experimental techniques like electron microscopy.⁵ The effect of crystallite morphology at the nanoscale for the catalytic activity of CeO₂-based materials has been investigated for several gas and vapor phase reactions.^6–8 These studies have shown that it is important to devise or modify the synthesis techniques that allow suitable tuning of the properties of the catalyst.

The solution combustion technique was introduced as an improvement over the established self-propagating high-temperature synthesis following which it was possible to obtain nanocrystalline oxide materials.^9–11 Numerous efforts have been made for synthesizing nanocrystalline CeO₂ using a large number of techniques including hydrothermal method, solvothermal method, microemulsion methods and sol–gel methods. The various techniques used and the particle size have been discussed in detail¹² and it is indeed possible to obtain CeO₂ crystallites with an average size of 10 nm. However, further reduction in the size of the crystallite is difficult and the synthesis of sub-10 nm crystallites is very difficult. Such small CeO₂ crystallite sizes cannot be obtained by using the conventional solution combustion technique.

Tuning of the properties using hybrid oxides is an established technique. Such approaches have proved to be useful especially with reference to the synthesis of ceria-based compounds, which are prevalent in exhaust catalytic applications. The introduction of a second component has been reported to improve the thermal and mechanical stability of the material.^13–15 The thermal stability and dispersion of CeO₂ supported on Al₂O₃ are greatly improved by insertion of ZrO₂ into the CeO₂ lattice.¹⁶ Similarly, CeO₂–ZrO₂ mixed oxides show improved redox properties as compared to CeO₂ which make them important innovative materials for three-way catalysts.¹⁷ By impregnating γ-Al₂O₃ with cerium/zirconium citrate solutions, nanostructured mixed oxides supported on Al₂O₃ are obtained, which exhibit high oxygen storage.¹⁸ Mixed oxides of the general formula La_0.5Sr_xCe_yFeO_z were found to exhibit high activity for NO reduction.¹⁹ Similarly, synergistic effects of crystal phases and mixed valences in La–Sr–Ce–Fe–O mixed oxidic/perovskitic solids on catalytic activity have been reported.²⁰ Synthesis of CeAlO₃ is relatively difficult but can be accomplished using cerium and aluminium by a modified solution combustion technique.²¹

Thus it has been established that development of sub-10 nm materials and hybrid materials is desirable for high catalytic activity. Therefore, in this study, our objectives are to synthesize hybrid materials with crystallite sizes below 10 nm and demonstrate the superior catalytic activity. We report a modified solution combustion technique for the synthesis of CeO₂–Al₂O₃–CeAlO₃ hybrid oxide materials with crystallite size as small as 2–3 nm. A thorough structural characterization of the compounds, including XRD, TEM and XPS, was carried out. We have explored the CO oxidation activity of these hybrid oxides and established the relation between the structure and the activity of the compounds for CO oxidation.

Experimental

Synthesis of hybrid oxides

The compounds were synthesized using the modified solution combustion technique. Unlike the conventional solution combustion technique, which utilizes only one fuel for combustion, this technique utilized a combination of fuels during the combustion step resulting in different phases and sizes of the final compounds. The use of fuel mixture reduces the flame temperature and thereby decreases the crystallite sizes, as demonstrated for the synthesis of zirconia toughened alumina preparation.⁹ When urea alone was used as the fuel,¹¹ the flame temperature measured was 1100 °C and when a mixture of fuels (glycine and urea) was used, the flame temperature was 900 °C.

Equal moles of a cerium precursor (ceric ammonium nitrate, Merck, India) and an aluminium precursor (aluminium nitrate, Merck, India) were used along with different fuels in different amounts. Glycine, urea, oxalyl dihydrazide (ODH), and hexamethylenetetramine (HMT) (all from Merck, India) were used as fuels. A number of hybrid oxides, pure CeO₂ and CeAlO₃ were synthesized. The moles of various fuels used and metal precursor moles used for the synthesis are given in Table 1. All experiments were repeated at least three times and the average variation in the crystallite size obtained was less than ±1 nm.

Table 1 Stoichiometry for the synthesis of various CeO₂–Al₂O₃–CeAlO₃ hybrid oxides using a modified solution combustion technique and the crystallite size obtained

Compound	Precursors/mole						Crystallite size/nm	Surface area/m^2 g^−1
	Aluminium nitrate	Ceric ammonium nitrate	Glycine	Urea	Hexamine	Oxalyl dihydrazide
C-1	0.04	0.04	0.165	0.039	—	—	11	12
C-2	0.04	0.04	0.165	0.019	—	—	10	13
C-3	0.023	0.023	0.109	0.023	—	—	22	8
C-4	0.133	0.133	0.576	—	—	—	10	13
C-5	0.033	0.033	0.044	0.005	0.004	0.003	3	19
CeO2–A		0.033	0.044	0.005	0.004	0.003	16	18
CeO2		0.033	—	—	—	0.079	38	12
CeAlO3							48	10

---

In a typical experiment, the precursor nitrates were dissolved in a minimum amount of water along with a fuel or a mixture of fuels. The clear solution obtained on dissolving the metal precursors and the fuels was slowly heated on a hot plate to increase the temperature. The hot solution was then transferred to the preheated muffle furnace. The clear solution was then heated in an alumina crucible in a muffle furnace, preheated to 500 °C. The solution boiled with evaporation of water and the leftover solid was observed to catch fire. The muffle furnace was kept in a fume hood in order for the gases of combustion to be moved out of the chamber quickly without any hazard. The mode of firing was dependent upon the fuel used. When oxalyl dihydrazide was used, the solids were found to catch fire spontaneously liberating a large volume of gases. With urea and glycine as fuels, the solution was found to froth during the vaporization of water. The solids obtained after the combustion were the required compounds which were characterized and used without further treatment.

Characterization

All the compounds were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), and X-ray photoelectron spectroscopy (XPS). XRD patterns were recorded using an Xpert Pro diffractometer. The diffractometer utilized CuKα radiation (λ_avg = 1.5418 Å) with a Ni filter. The patterns were recorded in a 2θ range of 20 to 65° in steps of 0.017°. Rietveld analysis of selected samples was carried out using the Rietveld analysis computer program FullProf and the widths of the diffraction peaks were obtained using WinPLOTR. TEM images were recorded on a Technai F-30 machine. The samples were suspended in acetone and mounted over carbon-coated Cu grids. XPS of the samples were recorded on a Multilab 2000 ECSA machine of Thermo Fisher instruments. The samples were added with a small amount of graphite and mounted in an ultrahigh vacuum chamber. The surface areas of the materials were measured by the BET nitrogen sorption method at 77 K with a Belsorp instrument (Japan).

CO oxidation experiments

CO oxidation was carried out in continuous flow reactors made up of glass having 9 mm internal diameter. Approximately 100 mg of the catalyst in powder form was placed between ceramic wool plugs for a bed length of 1 mm to ensure a bed volume of 63.6 mm³. An inlet CO flow rate of 2 ml min⁻¹ and O₂ flow rate of 2 ml min⁻¹ were maintained and the flow was diluted with N₂ for a total flow rate of 100 cm³ min⁻¹ and was maintained by control valves. Thus the space velocity for all the experiments was maintained at 26.2 s⁻¹. A thermocouple was placed in the catalyst bed and all experiments were conducted in the isothermal mode.

In order to carry out reactions under isothermal conditions, a PID temperature controller was used to maintain the temperature within ±1 °C. The temperature of the reactor was set and the gases were allowed to flow through the reactor continuously. The readings from the gas chromatogram were taken after ensuring that the gases went through the catalyst bed at constant temperature for at least 15 minutes. Six readings at a constant temperature were taken and the temperature was then increased to the next high temperature. The reported conversions are the average of the six readings and the sample standard deviation of the reported conversions was less than 2%.

The gas mixture from the outlet of the reactor consisted of a mixture of CO, CO₂, O₂ and N₂. Three distinct peaks in the gas chromatogram were observed corresponding to CO, CO₂ and a combined N₂–O₂ peak. The peaks were identified by following the retention times obtained by injecting pure gases. CO and CO₂ were detected using a flame ionization detector (FID) while N₂ and O₂ were detected using a thermal conductivity detector. The gas chromatographic system was equipped with a methanator to convert CO to CH₄. The use of methanator was essential due to a weak signal of CO in the FID. The methanator used Ru catalyst and was maintained at 350 °C. Separation and quantification of N₂–O₂ was not possible. However, CO and CO₂ were quantified using a calibration gas mixture having a predetermined concentration of the gases. The conversions reported for CO oxidation are based on the formation of CO₂.

In order to ensure the stability of the catalyst, experiments were conducted similar to our previous study.²² The oxidation of CO was carried out continuously for 4 h over the samples at 450 °C and the spectra of the spent catalyst were recorded. Both the XRD and XPS spectra of the catalysts did not show any large change in the oxidation state indicating the stability of the materials. The conversions of CO were also monitored and the compounds were found to retain their activity.

Results and discussion

Synthesis

The compounds were synthesized using the modified solution combustion technique wherein a mixture of fuels was chosen to obtain a range of phases and sizes of the compounds. The maximum temperature attained during combustion is determined by the composition of the fuel and this can be a governing factor for size and phase of the final compound. When the furnace temperature was maintained below 350 °C, ignition of the redox mixture was not observed to occur resulting in no or improper combustion. For accomplishing the product formation from such a route, post-synthesis calcination is required which may result in larger particle sizes. The composition of the fuel actually determined the maximum temperature attained during combustion and is an important factor for determining the size and phase of the final material.

The reason for the choice of the different fuels is that these are the most common fuels used in the synthesis of nanocrystalline oxides.⁹ The fuel and fuel mixtures were taken such that the total number of moles of gases evolved during combustion was almost the same for all the reactions (∼70 moles). In our previous work¹¹ on synthesis of the CeO₂–Al₂O₃ system using a mixture of fuels, we observed that a non-stoichiometric mixture of Ce : Al in a 4 : 6 atomic ratio gave rise to nanocrystalline CeO₂ (with Al₂O₃ being in the amorphous state) when urea alone was used as fuel but gave a nanocomposite of CeO₂–CeAlO₃ when a mixture of urea and glycine was used as fuel.¹¹ To study the effect of the mixture of fuels used in the combustion of equimolar Ce : Al precursors on the properties of powders, a set of experiments with different fuel mixtures were conducted. In the case of C-5, with the intention of obtaining smaller particles, fuels like ODH, HMT with lower heat of combustion were used. Small differences in the color of the compounds were observed depending upon the composition; CeO₂ was yellow and CeAlO₃ was green.

Structural characterization

X-Ray diffraction. The XRD patterns of all the synthesized compounds were recorded to determine the different phases present in the compound. The XRD of C-1, C-2, C-3 and C-4 are shown in Fig. 1(a–d). All the compounds showed nanocrystallinity exhibiting broad peaks and the crystallite sizes of the compounds were determined using the Scherrer formula. It can be seen from Table 1 that with the use of both glycine as well as urea, a decrease in the crystallite size occurred. However, the crystallite size did not depend significantly on the amount of urea used. But, the crystallite size decreased with a decrease in the amount of glycine in the solution. Reflections of four phases corresponding to CeO₂, CeAlO₃, α-Al₂O₃ and δ-Al₂O₃ were observed. All the phases are marked in the figure. The first intense peak at 22° was used for the Scherrer formula for determining the crystallite size. Al₂O₃ is known to undergo a series of crystal structural transformations starting from γ-Al₂O₃ to the most stable form α-Al₂O₃ as γ → δ → θ → α. This possibility cannot be discarded owing to different combustion temperatures attained with different fuel combinations.

Fig. 1 XRD patterns of the synthesized CeO₂–Al₂O₃–CeAlO₃ hybrid oxides. (a) C-1, (b) C-2, (c) C-3, (d) C-4.

The Rietveld refined XRD pattern of CeO₂ synthesized using oxalyl dihydrazide is shown in Fig. 2a. All the lines in the pattern could be indexed exactly to the cubic structure with a space group of Fm3m. The details of the refinement parameters are given in Table 2. The cell parameter was found to be 5.4135(3) Å, which is close to that reported for CeO₂.²¹ The crystallite size for CeO₂ was found to be nearly 38 nm. This was further confirmed by TEM analysis (discussed later). Although the crystallites were indeed nanosized, this size was actually much larger compared to the sizes obtained for other compounds as well as those reported for CeO₂ following microemulsion techniques. However, the method bears an advantage of short synthesis time and no requirement of post-processing for catalytic applications.

Fig. 2 Rietveld refined XRD pattern of (a) CeO₂ (b) C-5 and (c) CeAlO₃. The symbols show the experimental data; the continuous line shows the calculated pattern; the difference pattern is shown at the bottom.

Table 2 Lattice parameters obtained after Rietveld refinement

Compound	Space group	a/Å	c/Å	χ ^2
C-5	Fm3m	5.397(2)	—	0.9
CeO2	Fm3m	5.4135(3)	—	1.69
CeAlO3	I4/mcm	5.3278(1)	7.5717(3)	3.72

---

Very broad peaks were observed in the XRD of C-5 (Fig. 2b). Only four clear reflections corresponding to the ones marked in the figure could be observed. A very small peak corresponding to (2 2 2) reflection of cubic CeO₂ structure could also be observed in the pattern as a shoulder of (3 1 1) reflection. Rietveld refinement of the pattern was carried out for cubic structure of CeO₂ and satisfactory reliability parameters were obtained on refinement (Table 2). This showed the crystallization of the compound in cubic structure and the lattice parameter for this compound was found to be 5.397(2) Å. This was relatively smaller compared to that of CeO₂. However, due to very wide peaks, the possibility of merger of Al₂O₃ reflections in the pattern also existed. Therefore, TEM analysis was used for the structural characterization of this particular compound using the electron diffraction pattern (discussed later). Further, the crystallite size of the compound from Scherrer's formula was found to be nearly 2–3 nm, which was also confirmed using TEM analysis.

It is interesting to note that the CeAlO₃ phase was not present in the above compound. With the presence of both cerium as well as aluminium salts in the precursor solution, the possibility of co-formation of CeO₂, Al₂O₃ and CeAlO₃ existed. This was indeed observed in the case of compounds C-1 to C-4 (Fig. 1(a–d)). Damyanova et al.²³ have reported the formation of CeAlO₃ type phases in CeO₂–Al₂O₃ systems and temperatures above 525 °C were reported to favor CeAlO₃ formation. The use of different fuels results in the attainment of different temperatures during the synthesis. Owing to this, partial formation of different phases including their interconversion is possible. However, suitable tuning of the fuel can result in phase selectivity as well as large reduction in the crystallite size. Therefore, the synthesis of pure CeAlO₃ was also possible. This was the case when a mixture of glycine and urea was used as fuel. The Rietveld refined XRD pattern for the compound is shown in Fig. 2c. All the peaks in the compound could be indexed to a single phase tetragonal structure. The ratio of glycine and urea had to be optimized to obtain this single phase compound. The crystallite size of the compound was found to be nearly 48 nm, which was higher compared to the crystallite size of any other compound synthesized in this study.

Transmission electron microscopy. TEM images along with the electron diffraction pattern for CeO₂ are shown in Fig. 3a. The polydispersity in the sample could be observed with a wide distribution of both small as well as large particles. Particles as small as 10 nm could be observed along with bigger particles of size around 40 nm. A crystallite size of 38 nm was obtained with XRD analysis. The results from TEM analysis are in good agreement with the XRD results. The electron diffraction pattern was also used to determine the d-spacing. Using the (1 1 1) reflection from the image, the d-spacing was found to be 3.26 nm which was close to a value of 3.2 nm for CeO₂.

Fig. 3 TEM images and electron diffraction pattern of (a) CeO₂ and (b) C-5 and (c) SEM image of CeAlO₃.

High resolution TEM images for C-5 are shown in Fig. 3b. Crystallites of size 2–3 nm can be clearly observed. Therefore, broad lines in the XRD of C-5 (Fig. 2b) were indeed due to highly nanocrystalline nature of the compound. The polydispersity in the sample was negligible and the crystallite size was mainly restricted to 2–3 nm.

The crystal structure of C-5 was also confirmed from the TEM images and electron diffraction pattern. As mentioned earlier, broad peaks often make the structure determination difficult. From XRD, only the CeO₂ phase was found to be present in spite of the presence of alumina precursors in the solution. TEM analysis was, therefore, used for determining the crystal structure. From the high resolution images, the d-spacing corresponding to the CeO₂ (1 1 1) plane was determined. The average d-spacing for the crystallites, determined from five different crystallites was found to be 3.1 nm. Nearly the same spacing was observed in all the regions of the sample. This showed that the sample mainly contained nanocrystallites of CeO₂. The d-spacing obtained from the electron diffraction pattern was found to be 3.2 Å, which was also in agreement with that for CeO₂. Therefore, it was possible to obtain sub-10 nm CeO₂ particles using a mixture of fuels, as shown in Table 1. Though the aluminium precursor was indeed present in the solution, only purely crystalline CeO₂ structure was observed both by XRD and TEM. One possibility is that Al₂O₃ formed during combustion was amorphous, which has been reported by various techniques.^24–26 However, a more likely reason would be the substitution of Al³⁺ ions in Ce⁴⁺ sites. Owing to this, the reflections can be seen neither by XRD nor by TEM. Diffusion of Ce³⁺ ions into Al₂O₃ from the reduction of CeO₂ is possible only at high temperatures.²² However, cation substitution in CeO₂ is possible and has been reported by several techniques.^27–29 We have previously reported the use of the solution combustion technique for the synthesis of solid solutions of CeO₂ with TiO₂, ZrO₂ and SnO₂.^30–32 Therefore, in principle, the solution combustion technique can indeed be used for the synthesis of ionically substituted Al in CeO₂. However, this could be realized only with the use of a combination of fuels and the stoichiometric ratios given in Table 1. Further, the color of the compound was yellow. This was also an indication of the formation of the CeO₂ phase as the formation of CeAlO₃ resulted in green colored compounds. CeAlO₃ crystallized with larger size crystallites. The SEM image of CeAlO₃ is shown in Fig. 3c and agglomerates of nanoparticles can be clearly seen in the image.

X-Ray photoelectron spectroscopy. The XPS of all the samples were recorded to determine the oxidation state of the elements. Binding energy of C1s was used as the reference to determine the binding energy of other elements. The spectra of Ce3d for all the samples are shown in Fig. 4. Ce3d characteristic peaks for +3 and +4 states were located in the spectra.³³ It can be seen from the figure that Ce was present completely in +4 only in the case of CeO₂ (Fig. 4f). Characteristic Ce⁴⁺ peaks along with the satellites of the +4 state can be observed in Fig. 4f. Pure CeO₂ was observed from the XRD of the compound and correspondingly, Ce in the +4 state could be observed from the XPS. In CeAlO₃, Ce was mainly in the +3 oxidation state (Fig. 4g). However, a weak Ce⁴⁺ satellite peak could be also observed, indicating the presence of a small amount of Ce in the +4 state.

Fig. 4 XPS of Ce3d. (a) C-1, (b) C-2, (c) C-3, (d) C-4, (e) C-5, (f) CeO₂, (g) CeAlO₃.

Mixed Ce³⁺–Ce⁴⁺ oxidation states were observed for rest of the compounds. Peaks characteristic of both +3 as well as +4 states were present, as can be observed form Fig. 4(a–e). Both Ce³⁺ and Ce⁴⁺ states are marked in the figure. In most of the compounds, predominant Ce³⁺ peaks can be observed. From the XRD of the compounds, the presence of CeAlO₃ was confirmed in which Ce was present in the +3 state. Ce in the +4 state was present when the formation of CeO₂ took place. The XPS observations are in agreement with the XRD results, conveying the formation of mixed oxides constituting CeO₂–Al₂O₃–CeAlO₃ components. In the current study, we have only qualitatively shown the presence of both Ce³⁺ and Ce⁴⁺ states in the compounds. Detailed curve decomposition for obtaining peaks corresponding to Ce3d_5/2–3/2 can be accomplished by a careful procedure³³ for determining the relative amounts of +3 and +4 states, however, in this study, we have not attempted to quantify the different oxidation states.

It is to be noted that both Ce³⁺ and Ce⁴⁺ states were observed in the compound C-5 also. It was concluded that C-5 mainly consisted of CeO₂ from XRD as well as TEM studies. Aluminium in the compound was found to be present in substituted form for Ce⁴⁺ sites. This proposition can be rationalized with the help of XPS. Al was substituted in CeO₂ in the +3 state. Since the oxidation of Ce in CeO₂ is +4, the formation of oxide ion vacancies takes place in the compound for electrostatic neutrality. Therefore, the compound was actually a solid solution of the form Ce_1−x⁴⁺Al_x³⁺O_2−δ where δ is the reduction in oxygen occupancy of the compound due to the substitution of Al in the +3 state. Due to the presence of anionic vacancies in the compound, and the synthesis conditions, there indeed can occur electron transfers and oxygen exchange in the compound leading to a partial reduction of Ce from the Ce⁴⁺ to the Ce³⁺ state as Ce_1−x⁴⁺Al_x³⁺O_2−δ ↔ Ce_1−x−y⁴⁺Ce_y³⁺Al_x³⁺O_2−δ′ depending upon the oxygen rich or oxygen lean conditions at high temperatures of synthesis.

CO oxidation activity of hybrid oxides

CO oxidation was carried out over all the synthesized compounds. The variation of CO concentration with temperature over the different compounds is shown in Fig. 5. All the compounds were found to be catalytically active for the conversion of CO to CO₂ and more than 90% conversions were observed over most of the compounds within 500 °C. The activity of the compounds, based on the conversions achieved at given temperatures, followed the order C-5 > CeO₂ > C-3 > CeAlO₃ > C-1 > C-4 ≈ C2.

Fig. 5 Variation of CO concentration with temperature during CO oxidation reaction over the different compounds.

The variation of the rate of reaction with temperature is shown in Fig. 6. The rates of the reaction are the highest for C-5 followed by CeO₂ and the general trend discussed above can also be observed in this figure. The intermediate activity of CeAlO₃ compared to the different mixed oxides can be observed in Fig. 6a. Another important observation from Fig. 6a is the sigmoidal nature of temperature dependence of the rate of reaction. In the case of C-5, which showed the highest activity, the exponential increase in the rate of reaction with temperature was observed for a small range of temperature followed by saturation. Similar trends were observed for the other compounds also but at much higher temperatures (not shown in the figure). From the temperature dependence of CO conversion as well as the reaction rate, a clear trend in the activity of the compound was established. The Arrhenius plot showing the variation of natural log of the initial rate of reaction with the inverse of absolute temperature is shown in Fig. 6b. The plot was drawn for only a limited range of temperature as the ordinate in the figure is the initial rate and not the rate constant as in the case of a conventional Arrhenius plot. The linearity in the plot for limited lower conversions can be seen from Fig. 6b. Deviation from linearity (not shown in the figure) was observed which is a typical behavior of heterogeneous reaction and the non-linearity at higher conversions and temperatures is due to a change in the rate control regimes and mass transfer limitations. Therefore, we have concentrated on the linear portion of the plot. The activation energies over the complete range of conversion can be obtained only by a detailed kinetic model and fitting of the rate parameters. The activation energies calculated from Fig. 6b show only a qualitative trend because they are based on the initial rates. However, it is clear that the activation energy is the lowest for C-5 and high for C-2, C-4 and CeAlO₃. This plot also establishes the highest activity (and lowest activation energy) of the C-5 compound compared to the other compounds investigated in this study. The reason behind the differences in the activity of different compounds and the order of activity depends upon several factors. We describe each of them by classifying the compounds into mainly three categories.

Fig. 6 Temperature dependence of the rate of reaction. (a) Variation rate of CO conversion with temperature. (b) Arrhenius plot.

CeO₂ shows the activity for CO oxidation owing to the reducibility of the compound. Owing to the oxygen storage capacity of the compound, a small fraction of oxygen atoms in the compound can be exchanged from the stream resulting in the formation of CO₂ and reduced CeO₂. The oxidation of CeO₂ takes place by the stream oxygen. The following steps can be written for the reaction

CeO₂ + δCO → CeO_2−δ + δCO₂ (1)

CeO_2−δ + δ/2O₂ → CeO₂ (2)

δ in the above set of reactions represents the amount of labile oxygen in the compound. Higher amount of labile oxygen implies higher activity of the catalyst. Al₂O₃ is not reducible and, hence, the above reactions are not possible over Al₂O₃. Therefore, the working component in a CeO₂–Al₂O₃ hybrid oxide is CeO₂. We use the activity of CeO₂ as a reference to compare the activity of the other compounds.

The activity of C-5 was found to be higher than the activity of CeO₂. In C-5, Al was substituted in CeO₂ and this increases the lability of oxygen in the compound and also induces vacancy formation. The enhancement in the reducibility of the compound on ionic substitution has been established using density functional theory calculations for several metal–CeO₂ systems.^34,35 Further, the anionic vacancies aid oxygen dissociation and, hence, increase the activity of the compound. The mechanism of involvement of anionic vacancies for CO oxidation has been established in our previous study.³⁶ Therefore, with Al substituted in ionic form in CeO₂, C-5 shows the highest activity for the reaction. Small crystallite size and higher surface area of C-5 also contributes to this effect. However, at very high temperature, over reduction of the compound may take place owing to smaller crystallite size and higher reducibility. This is detrimental for the activity of the catalyst and as a result, the activity of the compound was found to be lesser than the activity of pure CeO₂ above 400 °C.

The activity of CeAlO₃ can be explained on the basis of interconversion of the compound to hybrid CeO₂–Al₂O₃ oxides. As mentioned earlier, the active species in the hybrid oxide is CeO₂. The surface formation of CeO₂–Al₂O₃ can take place following the reaction given below.

2δCeAlO₃ + δ/2O₂ → (CeO₂)_2δ(Al₂O₃)_δ (3)

The above reaction represents the formation of surface hybrid oxides by oxidation of CeAlO₃. The oxidized compound can again get reduced and the original compound can be restored by the interaction with the stream CO as follows.

(CeO₂)_2δ(Al₂O₃)_δ + δCO → 2δCeAlO₃ + δCO₂ (4)

However, it has to be noted that there is a fundamental difference between the mechanism which takes place during CO oxidation over CeO₂-based compounds and CeAlO₃-based compounds. Whereas the reaction is initiated by the reduction of the compound by lattice oxygen abstraction by CO in the case of CeO₂-based compounds, the phenomena over CeAlO₃ are initiated by O₂ resulting in the formation of CeO₂ species. This is followed by re-reduction of the support back to CeAlO₃ from CO in the stream. Surface oxidation of the compound can be expected to be a function of the crystallite size and, therefore, due to higher crystallite size, the activity of the compound was lesser than that of CeO₂.

To the best of our knowledge, CO oxidation activity of CeAlO₃ has not been reported previously. Prakash et al.³⁷ have studied phase evolution and thermal stability of CeO₂/γ-Al₂O₃ systems. The CeAlO₃ phase was observed on thermal treatment and interconversions of CeO₂/γ-Al₂O₃ and CeAlO₃/γ-Al₂O₃ were observed by surface oxidation–reduction steps. Complete transformation from one phase to another phase required temperatures as high as 700 °C. The presence of mixed Ce³⁺/Ce⁴⁺ states in CeAlO₃ at lower temperatures showed the surface oxidation of the compound. This is in support of our current observation and the redox processes during CO oxidation over CeAlO₃ can be expected to take place via the surface oxidation mechanism shown by eqn (3). CeAlO₃ phase formation in CeO₂–ZrO₂–Al₂O₃ systems has been reported by Liotta et al.³⁸ The formation of CeAlO₃ was found to be detrimental to the catalytic activity of the compound. Ce³⁺–Ce⁴⁺ transformations were observed in their studies also corresponding to the interconversions of CeAlO₃ and CeO₂. A similar mechanism can be expected over the hybrid oxides that have CeAlO₃, CeO₂ and Al₂O₃. One of the parameters governing the activity of the catalyst can be the relative amounts of the three components. C-1, C-2 and C-4 had similar crystallite sizes, surface areas and they also showed similar activity. To summarize, the activity of the compounds for CO oxidation is a result of the relative contributions of the crystallite size, surface area and composition of the hybrid oxide.

Conclusions

CeO₂–Al₂O₃–CeAlO₃ hybrid oxides were synthesized using the modified solution combustion technique. A mixture of fuels approach was adopted to obtain sub-10 nanometre crystallites. Pure phases, mixed oxides and solid solutions could be synthesized selectively by tuning the ratio of the fuels in the solution. A solid solution of CeO₂ with Al substitution could be obtained using a mixture of glycine, urea, hexamine and oxalyl dihydrazide with 2–3 nm crystallite size. All the compounds showed activity for CO oxidation and the activity of the compound was found to depend on the surface area and the composition of the hybrid oxide.

Acknowledgements

PAD gratefully acknowledges Bristol-Myers Squibb for graduate fellowship. GM gratefully acknowledges Department of Science and Technology, India, for the Swarnajayanti fellowship. The authors thank Dr. Nagesh Kini of Thermax India for the Rietveld analysis.

References

1. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Chem. Rev., 1995, 95, 69.
2. J.-N. Park, J. Noh, J.-S. Chang and S.-E. Park, Catal. Lett., 2000, 65, 75.
3. C. Meephoka, C. Chaisuk, P. Samparnpiboon and P. Praserthdam, Catal. Commun., 2008, 9, 546.
4. B. Hvolbaek, T. V. W. Janssens, B. S. Clausen, H. Falsig, C. H. Christensen and J. K. Norskov, Nano Today, 2007, 2, 14.
5. P. B. Amama, C. L. Pint, S. M. Kim, L. McJilton, K. G. Eyink, E. A. Stach, R. H. Hauge and B. Mayurama, ACS Nano, 2010, 4, 895.
6. N. Yi, R. Si, H. Saltsburg and M. Flytzani-Stephanopoulos, Energy Environ. Sci., 2010, 3, 831.
7. N. Yi, R. Si, H. Saltsburg and M. Flytzani-Stephanopoulos, Appl. Catal., B, 2010, 95, 87.
8. R. Si and M. Flytzani-Stephanopoulos, Angew. Chem., Int. Ed., 2008, 47, 2884.
9. S. T. Aruna and K. S. Rajam, Mater. Res. Bull., 2004, 39, 157.
10. M. S. Hegde, G. Madras and K. C. Patil, Acc. Chem. Res., 2009, 42, 704.
11. S. T. Aruna, N. S. Kini and K. S. Rajam, Mater. Res. Bull., 2009, 44, 728.
12. H. Gu and M. D. Soucek, Chem. Mater., 2007, 19, 1103.
13. W. J. Stark, J. D. Grunwaldt, M. Maciejewski, S. E. Pratsinis and A. Baiker, Chem. Mater., 2005, 17, 3352.
14. T. Yuzhakova, V. Rakic, C. Giomon and A. Auroux, Chem. Mater., 2007, 19, 2970.
15. M. Kim, T. R. Hinklin and R. M. Laine, Chem. Mater., 2008, 20, 5154.
16. R. Di Monte, P. Fornasiero, S. Desinan, J. Kašpar, J. M. Gatica, J. J. Calvino and E. Fonda, Chem. Mater., 2004, 16, 4273.
17. R. Di Monte, P. Fornasiero, M. Graziani and J. Kašpar, J. Alloys Compd., 1998, 275, 877.
18. R. Di Monte, P. Fornasiero, J. Kašpar and M. Graziani, Stud. Surf. Sci. Catal., 2001, 140, 229.
19. V. C. Belessi, T. V. Bakas, C. N. Costa, A. M. Efstathiou and P. J. Pomonis, Appl. Catal., B, 2000, 28, 13.
20. V. C. Belessi, C. N. Costa, T. V. Bakas, T. Anastasiadou, P. J. Pomonis and A. M. Efstathiou, Catal. Today, 2000, 59, 347.
21. S. T. Aruna, N. S. Kini, S. Shetty and K. S. Rajam, Mater. Chem. Phys., 2010, 119, 485.
22. P. A. Deshpande, S. Polisetti and G. Madras, AIChE J., 2011 DOI:10.1002/aic.12636.
23. S. Damyanova, C. A. Perez, M. Schmal and J. M. C. Bueno, Appl. Catal., A, 2002, 234, 271.
24. J. M. Wu, Mater. Lett., 2001, 48, 324–330.
25. A. Ishihara, H. Negura, T. Hashimoto and H. Nasu, Appl. Catal., A, 2010, 388, 68–76.
26. S. K. Lee, S. Y. Park, Y. S. Yi and J. Moon, J. Phys. Chem. C, 2010, 114, 13890.
27. S. Rossignol, F. Gerard and D. Duprez, J. Mater. Chem., 1999, 9, 1615.
28. C. Liang, Z. Ma, H. Lin, L. Ding, J. Qiu, W. Fradensen and D. Su, J. Mater. Chem., 2009, 19, 1417.
29. S. Song, R. O. Fuentes and R. T. Baker, J. Mater. Chem., 2010, 20, 9760.
30. P. A. Deshpande, M. S. Hegde and G. Madras, Appl. Catal., B, 2010, 96, 83.
31. S. Sharma, P. A. Deshpande, M. S. Hegde and G. Madras, Ind. Eng. Chem. Res., 2009, 48, 6535.
32. T. Baidya, A. Gupta, P. A. Deshpande, G. Madras and M. S. Hegde, J. Phys. Chem. C, 2009, 113, 4059.
33. E. Paparazzo, Chem. Eng. J., 2011, 170, 342–343.
34. T. Baidya, G. Dutta, M. S. Hegde and U. V. Waghmare, Dalton Trans., 2009, 455–464.
35. G. Dutta, U. V. Waghmare, T. Baidya, M. S. Hegde, K. R. Priolkar and P. R. Sarode, Catal. Lett., 2006, 108, 165–172.
36. S. Roy, A. Marimuthu, M. S. Hegde and G. Madras, Appl. Catal., B, 2007, 71, 23.
37. A. S. Prakash, C. Shivakumara and M. S. Hegde, Mater. Sci. Eng., B, 2007, 139, 55–61.
38. L. F. Liotta, A. Longo, G. Pantaleo, G. Di Carlo, A. Martorana, S. Cimino, G. Russo and G. Deganello, Appl. Catal., B, 2009, 90, 470–477.

---