The oxidizing role of CO₂ at mild temperature on ceria-based catalysts

Abstract

For thermodynamic reasons, CO₂ has always been considered as inert at mild reaction temperatures (∼300 °C). In this study, we show that CO₂ may be used as a valuable compound for the catalytic combustion of methane (CCM), if ceria-based materials are used as support for the palladium active phase. Adding CO₂ in the feed significantly improves performances of ceria-zirconia supported catalysts. On the contrary, catalytic performances are inhibited on Pd/γ-Al₂O₃. Inhibition can be avoided by mixing the Pd/γ-Al₂O₃ catalyst with some CeO₂ evidencing cooperation phenomena between both catalysts. In situ DRIFTS experiments show that the inhibition of the alumina-supported catalyst is not due to formation of carbonates species. After an in situ reducing pre-treatment, pure CO₂ is able to rapidly oxidize reduced Pd/Ce_0.21Zr_0.79O₂ catalyst at 300 °C. Dissociation of CO₂ on Ce_0.21Zr_0.79O₂ would be responsible for the oxidation process. Thus, CO₂ helps in replenishing the O reservoir (OSC) of the Ce-Zr-O support which is normally consumed by reductants such as CH₄, H₂ or other HC's. XPS experiments show enrichment in oxygen species bound to Ce (Low BE O1s) on the surface of ceria-zirconia when working in the presence of CO₂. Implications of these results on the behavior of ceria-containing catalysts can be important for practical applications, e.g., in automotive exhaust catalysis.

Introduction

One of the major challenges of this century is certainly the lowering of the emission levels of noxious and/or greenhouse effect gases in the atmosphere. One way to reach this goal is to develop the catalytic combustion (CC) process, which is known to produce heat and energy with much lower emission levels of CO, NO_x and hydrocarbons (HC's) than thermal combustion [1], [2], [3], [4], [5], [6], [7]. A debated issue concerning the catalytic combustion of methane (CCM) is the influence of the reaction products upon the rate of the reaction. Many authors have investigated this point so far but their conclusions vary in a more or less extended way [7], [8], [9], [10], [11], [12], [13], [14], [15]. They all came to the conclusion that water produced during the reaction has an inhibiting effect on the reaction, which is reversible, since it disappears when water is removed from the feed [7], [10], [11], [12], [13], [14], [15]. Nonetheless, it has been shown that the amount of water effectively produced in real CCM conditions is too low to significantly influence the reaction rate, as it is for CO₂, the other major product of the reaction [12].

The influence exerted by CO₂ is more debated than the effect of water [7], [11], [13], [15]. Indeed, some authors consider CO₂ as a simple inert gas towards the CC reaction rate [15] while others consider it to have an inhibiting effect which can only be observed in the absence of water [7], [11], [13]. This dependence of the inhibiting effect of CO₂ on the presence of water logically means that the inhibition of the reaction exerted by water is stronger than that attributed to CO₂. This statement can be correlated to the experimental fact that CO₂ desorbs before water [8], [10]. Most of these studies were carried out on Pd/Al₂O₃ and/or Pd/ZrO₂ catalysts. In all these experiments, CO₂ has been considered as inert. This is logical because thermodynamics disfavors the conversion of CO₂ to O₂ and CO at mild temperature. For instance, the dissociation constant (K_p) is only about 10⁻¹⁷ at 427 °C. Thus, at such low temperatures, it is assumed that the dissociation of gaseous CO₂ is completely excluded and consequently, CO₂ is traditionally considered as inert. However, recent investigations in our laboratory showed that it is possible to dissociatively activate CO₂ at low temperatures [16], [17]. CO₂ could dissociate on the surface of the catalysts yielding highly oxidative O species following the surface reaction: CO₂(g) → CO(ads) + O(a) [16]. The resulting oxygen species produced are suggested to play an important role in oxidation reactions concerned with both selective oxidation and combustion processes. In addition, we have recently shown that the effective influence of the presence of CO₂ in the gaseous atmosphere actually depends on the type of support of the palladium catalyst [18], [19]. In particular, the commonly used Pd/γ-alumina catalyst is inhibited by the presence of CO₂, whereas the Pd/Ce-Zr-O catalyst is not weakened under those conditions and can even display a better behavior [18], [19], [20]. Some activating effects of CO₂ can be observed, as stated above, when using Pd/Ce_xZr_1−xO₂ catalysts [18], [19], [20]. More precisely, results indicated that the Ce_0.21Zr_0.79O₂ composition gives rise to the best ceria-zirconia based catalysts and, what is more, that the activity of this catalyst (like most of the ceria-zirconia supported palladium catalysts) can be enhanced by the addition of some CO₂ in the reactant feed [19]. In this paper, we investigated more deeply the causes and origin of the effective influence of CO₂ on the CCM process using a Pd/γ-Al₂O₃, a Pd/CeO₂ and a Pd/Ce_0.21Zr_0.79O₂ catalyst, CeO₂ and Ce₂(CO₃)₃ materials and a 50/50 (w/w) mixture of Pd/γ-Al₂O₃ and CeO₂. For this purpose, they were tested under reaction conditions with and without CO₂ in the CCM feed and characterized by BET, XPS and XRD techniques. Moreover, in situ DRIFTS experiments were also carried out to study the formation and stability of potential adsorbed species.