Source

Yttria-stabilized zirconia supported platinum catalysts (Pt/YSZs) for CH4/CO2 reforming Although YSZs has received extensive attention with respect to oxygen sensors and fuel cells [8], the role of catalysis has seldom been addressed. Ekerdt and co-workers [9–11] found that certain hydrocarbons can be synthesized over ZrO2 and YSZs in CO/H2. They also proposed that the oxygen vacancies of these oxides are the active sites for CO hydrogenation. Meanwhile, Metcalfe et al. [12,13] found that YSZs as support can enhance the activities of Pt and Rh catalysts in CO oxidation and NO reduction. Dow and Huang also observed a similar enhancing effect for copper oxide catalysts supported on YSZs [14], and attributed it to the oxygen vacancies [15,16]. We also found that oxygen vacancies can enhance the reduction of carbonyl groups (C=O) of crotonaldehyde over Pt/YSZs early in reactions at lower reaction temperatures, and this promoting effect could be duplicated after treating at above 400◦C [17]. We hypothesized that the oxygen vacancies could be used to activate CO2 for CO2 reforming of CH4 at higher temperature (>600◦C). This work used YSZs as support for platinum catalysts to investigate the effect of YSZs in CH 4/CO2 reforming. Recently, CH4/CO2 reforming has attracted interest for environmental and industrial reasons. Both CO2 and CH4 are considered as harmful greenhouse gases [18–21] and this reaction provides a means of disposing of or recycling these gases. The CH4/CO2 reforming also provides an important route to produce CO-rich synthesis gas or extremely pure CO for the synthesis of chemicals such as acetic acid, dimethyl ether and alcohols [22]. Numerous authors have investigated various metals like Ni, Ru, Rh, Pt, Ir and Pd [21,23–27] for CH4/CO2 reforming. Most authors preferred Rh catalysts, owing to their good activity and stability. Meanwhile, Ni catalysts are commercially more interesting than noble metals but suffer the disadvantage of a high rate of coke formation [19,21,28]. From the perspective of relative price and good availability, Pt is a reasonable compromise. Several works [26,29,30] have confirmed that ZrO2 is a very effective support for the Pt catalysts; such combinations not only are more active but also more stable than Pt supported on other materials [26]. We hypothesized that the YSZs might also exhibit some properties resembling the ZrO2 for the Pt catalysts. In accordance with the above discussion, this study investigates the influence of YSZs on CH 4/CO2 reforming. YSZs supports and Pt/YSZs catalysts were prepared and characterized, and the

efficiencies of these catalysts were demonstrated by CH4/CO2 reforming. Additionally, the role of YSZs was revealed by pulse analysis of CO2 and CH4 at the reaction temperature of CH4/CO2 reforming.

Catalytic properties of yttria doped bismuth oxide ceramics for oxidative coupling of methane Much work was reported in the past decade on catalytically oxidative coupling of methane (OCM), a promising process for direct conversion of natural gas into C2 (C2H4 and C2H6) products [1-4]. Due to deep oxidation reactions in the gas phase as well as on the catalyst surface, the C 2 yield of OCM achieved so far on any catalysts operated in conventional packed-bed reactor is less than 25% when C2 selectivity is higher than 50%, far below the requirement for making OCM economically attractive (>30-40%) [4]. In order to improve the C2 yield, some researchers have recently shifted their focus from searching for better catalysts to developing new types of reactors [5-9]. Due to their several unique properties, ceramic membrane reactors offer potential for improving the C2 yield by genetically minimizing the deep oxidation reactions [10-17]. Most applications of membrane reactors permit the integration of the reaction and separation steps and increase of the yield beyond nominal equilibrium values by selectively removing the desired products from reaction medium. Different from these applications, the OCM membrane reactors are employed to distribute or transport oxygen into reaction medium in a well-controlled manner in order to maintain an optimum oxygen to methane ratio in the reactor. Among different types of ceramic membrane reactors, those made of dense ionic-conducting ceramics appear to be most promising for obtaining higher C2 yield. Lin and his co-workers [18,19] recently pointed out that OCM catalytic properties of the membrane materials are critical to the success of an OCM dense membrane reactor. Three well-known groups of oxygen semipermeable ionic/mixed-conducting inorganic materials are zirconia-based, bismuth oxide-based and perovskite-type ceramics [20,21]. It is known that zirconia-based ceramics are not good catalysts for OCM unless they are promoted with alkali metal compounds (e.g., Na +-ZrO2--Cl-) [22]. The OCM properties of two highly oxygen semipermeable perovskite-type ceramics, La0.2Sr0.8CoO3 and SrCo0.8Fe0.2O3, were recently studied in our group [19] by using both steady-state (cofeed mode) and unsteady-state (cyclic mode) methods. In the steady-state study, La0.2Sr0.8CoO3 exhibited good OCM catalytic properties in terms of C2 yield (>14%), selectivity (>50%) and space-time yield (>5

~tmol/g s) while Sreo0.sFeo.203 showed very poor OCM catalytic properties. However, the C2 selectivity of Lao.2Sro.8CoO 3 decreased substantially (<30%) when operated in cyclic mode with a reducing environment similar to that in a dense membrane reactor. This suggests that Lao.2Sro.8CoO 3 may not possess desired OCM catalytic properties when used in an OCM dense membrane reactor.

Effect of chlorine on TPR and TPO behavior of an YSZ/T-Al2O3 supported copper oxide catalyst Many catalyst studies have increasingly focused on oxygen-ion conducting oxides [1-8]. Previous studies indicated that using such materials (a) as supports could significantly enhance the catalytic activity of Pt [1], Rh [4] and CuO [5,7] for CO oxidation and NO reduction, or (b) as catalysts might enhance the selectivity of CO hydrogenation [2,3] and the decomposition rate of formic acid [8]. These improvements in catalytic activity have been attributed to the effects of oxygen ionic transport [1,4,5,7], interfacial metal-support interaction (IMSI) [1,5,7] or surface oxygen vacancies of these oxygen-ion conducting oxides [2,3,7,8]. Our previous studies [5-7] have demonstrated that the oxygen-ion conducting support, yttriastabilized zirconia (YSZ), can markedly enhance the reducibility of supported copper oxide and improve the catalytic activity of a copper oxide catalyst. More specifically, the YSZ support can improve the light-off behavior of a copper oxide catalyst for CO oxidation [7], which is unexpected since the base-metal oxides generally have poorer light-off behavior than that of precious metals [9]. This activity enhancement of YSZ supported copper oxide catalysts has been ascribed to the modification of the reaction mechanism and the synergistic effect caused by the interfacial metal oxide-support interaction (IMOSI) [7]. Hence, extending the interfacial boundary line between the supported copper oxide and the YSZ support should significantly enhance the copper oxide activity for CO oxidation. Dispersing YSZ crystallites on γ-alumina support could be a feasible alternative for application purposes. Such an action cannot only extend the interfacial boundary line but may also reduce the usage requirement of YSZ. However, if the precursor metal salts of supported YSZ are chlorides such as YCl3 and ZrCl4, it should first be considered whether chlorine is a contaminant, since many studies have reported that the residual chlorine coming from the precursor chloride salts can affect the catalysts' physical and chemical properties. For instance, chlorine can

(a) improve the dispersion of supported metals [10-12], (b) change the catalysts' activity and selectivity [13,14], and (c) affect the crystallization of the ZrO2 support [15]. Moreover, chlorine can significantly reduce both the adsorption rate and capacity of a metal surface [16,17]. In fact, Cl can lead to a more serious poisoning effect than S and P owing to its stronger electronegativity [16,17]. Therefore, this work examines (a) the effects of chlorine, which comes from the precursor metal salts of YSZ supported on γ-alumina, on the reduction and reoxidation behaviors of supported copper oxide and (b) the feasibility of preparing the YSZ/γ-alumina support.

Study of ceria-supported nickel catalyst and effect of yttria doping on carbon dioxide reforming of methane Oxygen-ion conducting oxides such as ceria, samaria-doped ceria, and yttria-doped ceria (YDC), when used as supports, are known to impart metal–support interactions to enhance catalytic performances, due to the oxygen storage/transport characteristics of the support [18,23] or to the generation of active centers at the interface between metal and support [24–27]. Thus, to take advantage of the metal–support synergistic effect in the treatment of CO 2 reforming of CH4, supporting nickel on ceria appears to be appropriate to develop a catalyst that is resistant to carbon deposition and that exhibits stable operation for extended periods of time. The aim of this work is to elucidate the role of Ce3+ in metal–support interaction on the activity enhancement and coking resistivity of nickel supported on ceria and YDC, and to further examine the effects brought about by the doping of yttria. For this purpose, Ni/γ-Al2O3 and Ni/YDC catalysts with various yttria loadings in ceria were prepared and characterized using X-ray diffraction, electron paramagnetic resonance, and temperature-programmed reduction and hydrogenation. Activity tests were carried out to evaluate the performance of these catalysts.

Oxygen vacancies contributing to intragranular electrical conduction of yttria-stabilized zirconia (YSZ) ceramics

Although the physical and chemical aspects of YSZ ceramics have been investigated in detail, this material still has several issues that require experimental clarification, including (i) estimation of energy values for EM and EO and (ii) identification of the trapping center for oxygen vacancies. For oxides in which electrical conduction involves dielectric relaxation processes, dielectric measurements can be used to estimate EM and EO, because the relaxation processes resulting from O2 -migration appear in the dielectric properties, and their intensities are proportional to the number of migrating O2 ions, i.e., the density of the oxygen vacancies. The previous results for other oxide electrolytes [2,3,5,6] suggest that either Y 3+ or Zr4+ must act as a trapping center for oxygen vacancies in YSZ at low temperatures. Some trials have been carried out in order to identify the trapping center. An extended X-ray absorption fine structure (EXAFS) study indicates that an oxygen vacancy is sited adjacent to Zr4+ [7]. However, the molecular dynamics analysis shows that oxygen vacancies are trapped at the second-neighbor positions to Y3+ [8]. From a very different standpoint, the present study attempts to identify the trapping center. Some knowledge about the trapping center would be obtained if Y3+ or Zr4+ ions were partially replaced by other trivalent or tetravalent ionic species, because such replacements would change the bonding nature between the trapping center and an oxygen vacancy. However, this is just the scenario within the grains. The grain boundary effect is also important for the electrical conduction in a ceramic electrolyte because a high resistivity at the boundary deteriorates the foremost function of electrolytes, i.e., high ionic conductivity [3,6,9]. In particular, the interaction between oxygen vacancies and the space charge layers is very important for reducing the resistance at the boundary [9]. However, it is difficult to treat the grain boundary quantitatively because of its complicated behavior. Considering these factors for YSZ, the present study has elucidated the intragranular electrical conduction of YSZ in terms of the dynamic behavior of oxygen vacancies. The illustrative examples treated here are oxides based on the (ZrO2)0.92(Y2O3)0.08 system, since this composition enhances electrical conduction [10,11]. In the present study, non-doped (ZrO2)0.92(Y2O3)0.08 as well as doped specimens, in which Zr4+ or Y3+ are partially replaced with Hf4+ or Sc3+, were prepared and the dielectric properties were measured. The utility of dielectric measurements on YSZ has already been proved [12].

During the past year the research involved yttria-stabilized zirconia (CZA) in the chemical synthesis

process have focused on two issues: confirmation that anion vacancies are the active slues for CO activation and hydrogenation, and the influence of additives on the reactions that occur during isosynthesis. The interaction of CO and CO/H2 mixtures with ZrO2 has been studied, and the species which formed and how they transformed are understood [1,2,5]. Our study was directed at establishing the role of the oxide surface in the activation of CO and in the formation of the C1 fragments which are involved in methane and methanol formation at 1 atm [2] and in isosynthesis at 35 atm [4]. The active site for formate formation and reduction to methoxlde had been suggested to be a surface anion vacancy [2]. Molecular probes were selected as the means to identify the site over ZrO2. Carbon monoxide was expected to form a formate, CO/H2 adsorption followed by hydrolysis of methoxide to methanol was expected to identify the methoxide surface concentration, and SO3 was expected to react with anion vacancies and form a sulfate. Sulfur trioxide was selected as an anion vacancy titrant for ZrO2 on the basis of studies with SO3 over other oxides and on the ability to form sulfate species over ZrO2. Infrared studies over Fe2O3 by Yamaguchi et el. [5] showed that sulfur trioxide adsorbed as a sulfate species with a dioxo structure, (FeO)2S(-O)2. This was proposed to occur at oxygen anion vacancy sites. A sulfate species could also be formed over Fe2O3 from SO2 in excess O2 or by calcining (NH4)2SO4 and Fe(OH)3 [5]. A similar sulfate structure with dioxo ligands was proposed by Jin et el. [6] following calcination of (NH 4)2SO4 and Zr(OH)4 at 600 °C. A recent infrared study by Bensitel et el. [7] found that adsorption of H 2S or SO2 in excess O2 over ZrO2 at 450 °C resulted in the (ZrO)3S-O sulfate species. This same (ZrO)3S-O sulfate structure was also reported following impregnation of ZrO2 with either H2SO4, (NH4)2SO4 or Zr(SO4)2 followed by evacuation at 450 °C [7]. The oxides of zirconium form three different phases, monoclinic, tetragonal and cubic [8-10]. The monoclinic phase is stable below 1200 oC. The tetragonal phase is normally stable above 1200 °C, but can he obtained in a metastable condition at much lower temperatures and is the crystal structure of zirconia particles under 300 angstrom in diameter [11]. The cubic phase is formed at temperatures above 2280 oC, but can be stabilized at room temperature by the addition of other oxides such as Y2O3 and CaO. The uptake studies were performed over pure zirconia and over yttria doped zirconia. It was not possible to make pure ZrO 2 which was all one phase and of sufficient area (at least 5 m2 g-l) for the gravimetric experiments. Only mixtures of monoclinic and tetragonal zirconia formed. Yttria doping of ZrO2 was used to generate a single phase (cubic) and to introduce

known concent cations of anion vacancies. This latter aspect results since stabilization of the cubic structure is accomplished by direct substitution of trivalent yttria cations for the host lattice Zr +4 cation. Since the dopant cation is of lower valence than the host cation, oxygen vacancies are created to preserve lattice neutrality. Spectroscopic studies have shown that the trivalent Y and tetravalent Zr cations are statistically distributed [12] and that the Zr cations are nearest neighbors to the anion vacancies [13]. These vacancies increase the electrical conductivity, with diffusing oxygen ions being the primary charge carrier [14-16]. This diffusion of oxygen ions has been associated with lattice vacancy migration [14,15]. A series of catalysts was prepared for this study. The composition and phases are presented in Table 1. The uptake results are presented in Table 2. Fourier-transform infrared spectroscopy was used to establish that CO adsorbed as a formate (as expected [3]). Therefore, the CO uptake results represent the amount of formate which adsorbed over the catalyst. The highest in the 9.1% YSZ sample and significantly depressed in the other YSZ samples. We reasoned that SO 3 reacted with the vacancies; because of their mobility in the 9.1% sample the vacancies could iigrate to the surface. We proposed that the uptake studies over YSZ confirm that SO 3 reacted with anion Vacancies. Methanol was produced in a noncatalytic reaction to determine the number of methoxide species which formed. Methanol can be produced in a catalytic reaction over ZrO 2 at atmospheric pressure [2,18] and over both ZrO2 and yttria stabilized ZrO2 at 35 atm [4]. Methoxide is the precursor to methanol [2]. Catalyst samples which contained different amounts of monoclinic, tetragonal and cubic zirconia are represented in Figure 2. Similar correlations (not shown) were found when the amount of CO adsorbed as formate was plotted versus the amount of SO 3 adsorbed or when CO adsorbed was plotted versus methanol formed. These correlations support the proposal made earlier [2] that CO hydrogenation to methanol proceeds over anion vacancy sites. Figure 3 presents the mechanism for CO interaction with ZrO2, which was presented in an earlier publication [2], and a representation of SO3 reacting with a vacancy site on the ZrO 2 surface. Hydroxyl groups are not shown in Figure 3a because our studies did not indicate any role of hydroxyl groups in sulfate formation. (Hydroxyl groups were present during all of the experimerits.) Desoz~tton in argon or oxygen, during the r~eated dose experiments,

should have prevented repopulation of hydroxyl groups. If hydroxyl groups play a role in the uptake of SO3, this would have led to a continuous decrease in SO3 uptake with continued dosing/desorptlon cycles. 'Bensitel et al. [7] also did not find any infrared evidence for hydroxyl interaction with SO2 during sulfate formation. The chemistry for hydrocarbon synthesis over ZrO2 was investigated and reported [2,4]. The reactions which lead to higher weight products are presented in Figures 4 and 5. We have been interested in establishing the surface characteristics of metal oxides which give rise to and influence the reactions presented in these figures. Our approach has been to investigate how acid and base additives and vacancy concentration influence the relative rates of CO insertion and condensation. We have prepared a variety of doped zirconias and have examined the effect of dopant type and concentration on the selectivity at 35 atm and 425 °C to identify trends with dopant level. These catalysts have been analyzed by X-ray diffraction and XPS. Figures 6 and 7 are representative of the activity and selectivity patterns of the vacancy level (Figure 6) and basicity (Figure 7). Causes for these trends can be established through the use of carbon-13 labeled acetone and propionaldehyde which will permit us to determine, in an absolute manner, the effect of additives on the rates of CO insertion and condensation for linear and branched propagating species. Carbon labelling was used no establish the mechanisms [4]. We will also investigate secondary alcohol dehydration selectivity to 1- and 2-olefins because the key intermediate in dehydration, an enolaue, is also a key intermediate (VIII and X) in the CO hydrogenation mechanisms shown in Figures 4 and 5.

Activation of O2 and CH4 on yttrium-stabilized zirconia for the partial oxidation of methane to synthesis gas Yttrium-stabilized zirconia (YSZ) appeared to be a promising catalyst for catalytic oxidation of methane to synthesis gas, despite its insufficient reforming activity, which needs to be compensated for with a reforming catalyst in a dual-bed system, as proposed in our previous work [1]. In general, it is known that defects, such as oxygen vacancies, are important in the surface chemistry and catalysis of metal oxides [11]. The number of oxygen vacancies in ZrO 2 can be increased significantly by doping with lower valence metal ions, such as Y 3+ and Ca2+. Compared with ZrO2, the improved catalytic performance of YSZ in oxidation catalysis has been attributed to

a high concentration of oxygen vacancies [3,12]. Lattice oxygen ions are often involved in reactions over oxide catalysts. Most of the partial oxidation reactions proceed via a Mars–van Krevelen mechanism, in which lattice oxygen ions are incorporated into the products [13]. The cycle for catalytic partial oxidation is closed via replenishment of the extracted lattice oxygen ions through the dissociative adsorption of molecular oxygen at the surface [14]. In our previous work [2], formaldehyde and formate were both shown to be reaction intermediates for catalytic partial oxidation of methane over YSZ. Based on the results of in situ FTIR and both steady-state and transient experiments, a reaction scheme was proposed. We also investigated the effect of the surface composition of YSZ catalysts on the catalytic performance in catalytic partial oxidation of methane. It was postulated that oxygen vacancies are most likely involved in catalytic partial oxidation of methane [3]. However, the mechanistic details, particularly the role of oxygen vacancies and that of lattice oxygen ions in the partial oxidation of methane, are still not clear. Moreover, the nature of the oxygen species, such as surface lattice oxygen ions or adsorbed oxygen, in the activation of methane is still a matter of controversy [15, 16]. In this work we identify the active sites for the activation of both oxygen and methane on the surfaces of YSZ and ZrO2. CPOM is studied over ZrO2-based catalysts in transient experiments. Isotopic oxygen 18O2 exchange with the catalysts is investigated in both the absence and presence of methane and under reaction conditions for CPOM.