(141e) The Effect of Ca2+ and Al3+ Additions On the Stability of Potassium Disilicate (K2Si2O5) As a Soot Oxidation Catalyst

The high activity of potassium-containing catalysts for soot oxidation has attracted much attention over the past decade. Unfortunately, potassium containing catalysts usually tend to degrade after repeated thermal cycles due to the loss of potassium. This is especially the case for K₂CO₃ or KNO₃.  This deactivation is thought to occur due to loss of potassium through sublimation. Thus the challenge in using potassium and other alkali metal based catalysts is minimizing, if not eliminating, the loss of the potassium or alkali metal. One approach often used is to incorporate K into a more stable structure (e.g. perovskites)

In many cases attempts to stabilize K were pursued by impregnating potassium species into a support material. This approach has been common over the past decade, with examples such as potassium catalyzed soot oxidation supported by TiO₂, ZrO₂ , Al₂O₃, SiO₂  etc. Among these support materials, Al₂O₃ and SiO₂ do not show any catalytic effect while ZrO₂ and TiO₂ show very low activity in soot oxidation. The potassium-TiO₂ interaction and the improvement of contact conditions between soot and the catalyst, studied by means of thermal programmed reduction (TPR) and X-ray diffraction(XRD), were offered as an explanation for the activity differences for soot oxidation compared to catalysts supported on alumina.  K/ZrO₂, with a K/Zr ratio=0.14, synthesized from potassium nitrates was reported as providing the highest activity.  The reversible transformation of a bridged NO³⁻ and a monodentate NO³⁻ was suggested as a possible mechanism. Studies on the application of K on silica supports was only recently reported.

Ogura et al. tried to prevent potassium losses by using silica-alumina and zeolites as support materials and impregnating potassium as the active catalytic center. They found that the second test cycle of soot oxidation on K/sodalite showed higher activity than the first run. It appears that the state of potassium was changed under thermal conditions with a minor loss of the crystal structure of sodalite resulting from interaction of Al sites and K in aluminosilicate. However, no extended stability test was performed to reveal the relatively long-term catalytic performance in this study. Relatively extended stability testing was performed by López-Suárez et al.  when they studied the effect of copper on potassium stability in K/SrTiO₃ catalyzed soot oxidation. According to their 6-cycle repeated TPR results, significant degradation was found between the first and second TPR cycle for all potassium catalysts. Even though the less active Cu/SrTiO₃ catalysts do not degrade, the addition of copper into K/SrTiO₃ structure does not affect the potassium stability.

Research results show that catalysts with potassium can lead to high activity for soot combustion. Nevertheless, such catalysts can be readily degraded due to the loss of active potassium. Conversely, an investigation showed that catalysts with immobile K ions that are rigidly bound in the lattice give lower activity and higher soot oxidation temperatures. Neither case is ideally suited for the application on DPF.  As an alternative one can consider the use of potassium containing glasses as catalysts. This approach relies on the slow passive release of potassium from a glass to provide for renewal of the catalytic surface activity of the glass. The purpose is to mitigate the effect of loss of the active potassium species by providing new ions over time.

The slow release of ions desired in a catalytic glass is the opposite of what is desired for a durable, weather resistant glass.  Several approaches to modeling glass durability have been described, primarily for predicting the durability of glasses for nuclear waste storage. In that application, the goal is to minimize leaching from, or corrosion of, the glass, with researchers hoping to predict glass behavior over thousands of years. The approaches usually involve considering a glass as a mixture of species with known free energies of hydration. Glass durability is then predicted based upon thermodynamic hydration equations, with the contribution of the various species scaling with mole percentage of components present in the glass.

To design durable glasses, glass reactant species are selected based on hydration reactions that are expected to occur between the glass and an aqueous solution (acidic or basic). This is based on expectations as to whether cations in the glass will anionically complex with silica or other oxides. This is determined from their relative anionic force, which reflects their relative field strength. Cation species that might be incorporated into a glass can be classified as to whether they are network formers (i.e. ions with high atomic field strengths (F), calculated as the atomic charge (Z) divided by the square of the ionic radius  ( r )), network modifiers (i.e. ions with low atomic field strengths), or intermediate cations, that can act as either network formers or network modifiers. Network modifier cations are oxide species which are highly anionically associated with [SiO₄ ]^-4  tetrahedra. Potassium and sodium are examples of network modifiers that have low field strengths, so are susceptible to leaching.

The field strengths are considered along with the relative partial molar free energies of the hydration reactions (DG_i) of the cation species that can occur in an aqueous environment.   A chemically and electrically balanced hydration reaction can be written for potential components, and the partial molar hydration free energy for each reaction be calculated (DG_i =DG_(products)- DG _(reactants) ) for the expected hydration reactions in the environment (e.g. aqueous oxidized basic environment). Jantzen showed  that the thermodynamic free energies of hydration (DG_hyd) of the silicate and oxide glass components are correlated with their ionic field strengths (F).

The results of these studies provide a starting point for selecting ions for incorporation into a glass with poor weathering characteristics, as is desired for DPF application. Such a glass should incorporate relatively high levels of low field strength, highly negative free energy of hydration ion species (i.e. alkali ions). For the novel application of glasses as soot combustion catalysts it is necessary to balance two characteristics of the glass: melting point and ion delivery.  Not only is easy ion exchange desired, but the glass should have a melting point high enough to withstand the diesel exhaust environment without substantial melting or particle coarsening.  During soot combustion, it is possible that local temperatures can reach 700-800C for brief periods. Thus a desirable glass possesses a relatively high melting point as well as facile ion exchange.

For designing a catalytic glass, a good starting point is provided by the alkali disilicates (or “phyllosilicates”), of the type M₂Si₂O₅, where M is one or more alkali metal ions. The alkali disilicates are of particular interest because while some of them are very susceptible to degradation by moisture, others are unaffected.  For example, K₂Si₂O₅ is very hygroscopic, while Li₂Si₂O₅ is unaffected by moisture.   Potassium disilicate (K₂O.2Si₂O₂) glass thus can provide a starting point for examining mixed ion effects in soot oxidation catalysts based on glass. In the present study we examine if such mixed ion effects can be observed in terms of soot oxidation activity.