Novel Sn catalysis for the production of acrylic acid

Instigated by the growing societal demand for a more sustainable and circular economy, the chemical sector needs to abandon fossil-based carbon and material disposal, and shift to renewable bio-based carbon and material recycling. However, in contrast to fossil resources, bio-based substrates typically possess more functional groups, which increase their polarity and reactivity. It is therefore preferred to process such reactive polar molecules in liquid phase at moderate temperature rather than in traditional gas-phase processes at high temperature, which are currently used in existing petrorefineries. Moreover, to enable the transition towards bio-based chemistry, it is of utmost importance to develop new liquid-phase catalytic technologies.

In this regard, Lewis acidic Snβ zeolites are one of the most promising catalysts for future biorefineries. Not only can they selectively interact with oxygen-rich functionalities, but their hydrophobic nature allows liquid-phase processing in polar solvents. Snβ zeolites are known to catalyze actively and selectively a variety of reactions including the isomerization and epimerization of mono- and disaccharides, the conversion of sugars to lactate esters and the coupling of carbon-carbon bonds. Also, they display excellent activities in the Baeyer-Villiger oxidations (BVO) and Meerwein-Ponndorf-Verley reactions (MPV).

In recent years, to enhance the industrial relevance of Snβ catalysts, research has focused on ameliorating the synthetic procedure and on augmenting the Sn loading to maximize the space-time-yield, which is a key industrial benchmark. Whereas industrial implementation of the conventional hydrothermal synthesis is rather difficult due to long processing times, toxic agent use (e.g., HF), and low final Sn loadings (<2 wt.%), the post-synthetic solid-state incorporation of Sn lends itself more to industrial scale-up since it is fast, easy and yields high Sn loadings (up to 10 wt.%). Unfortunately, however, high Sn loadings irrevocably coincide with the existence of large amounts of inactive SnO2 clusters, which drastically reduces the catalytic activity on Sn atom basis. Therefore, to resolve the formation of inactive Sn oxide clusters at high Sn loadings this PhD dissertation unraveled a novel strategy to obtain highly active Snβ catalyst – even at high Sn loadings – by exploiting the redox chemistry of Sn.

While previous findings reported that post-synthetic Snβ catalysts with higher activity can be obtained when the standard calcination in air is preceded by pyrolysis under inert atmosphere (pre-pyrolysis), up to now, the exact mechanism is not clear. Hence, the first part of this work scrutinized the exact mechanism by which the Sn atoms are incorporated into the dealuminated β zeolite during solid-state incorporation. More specifically, the decisive role of the heating atmosphere on generating active Sn sites in Snβ zeolites was linked to the formation of Sn(II)O species created during pre-pyrolysis of the Sn(II) acetate precursor before calcination. The main evidence for this result was deduced from in-situ characterizations via 119Sn Mossbauer spectroscopy, TPDE-MS and TGA. It is hypothesized that the higher Sn dispersion is facilitated by the higher volatility, and hence mobility, of Sn(II)O species compared to Sn(IV)O2 species, which are formed in standard air-calcination.

Next, prompted by this knowledge, it was investigated if inactive Sn(IV) oxide clusters can be (re)converted into intermediate mobile in-situ Sn(II)O species via reduction in hydrogen to enable Sn redispersion as well as Sn reactivation. Interestingly, the catalytic BVO and MPV activity of Snβ samples with a large amount of inactive SnO2 was indeed increased via consecutive reduction–reoxidation. Surprisingly, it was even possible to surpass a previously observed threshold BVO activity – present at high Sn loadings – by 50%. Depending on the Sn content, two explanations were proposed. At low Sn loading (< 5 wt.%), Sn dispersion increased through the formation of mobile Sn(II)O species, while at higher Sn loading, even though the total amount of active sites decreased, the catalytic activity raised by re-speciation of the active sites towards pentacoordinated Sn sites. These Sn sites are suggested to be strong Lewis acidic Sn sites which more efficiently form the required substrate orientation. Overall, this implies that inferior Snβ catalysts can be reactivated by redox activation of inactive SnO2 clusters.

In the final part of this doctoral dissertation, the research findings and general conclusions were framed within the existing Snβ research field together with some future perspectives.