As a major source of renewable resources, the valorization of lignocellulosic biomass is one pillar of the future economy. Lignin consists of aryl ether units connected by ether and alkyl bonds, whose cleavage yields monomeric phenols and methoxy phenols. Thus, the pyrolysis of lignin produces oxygenated bio-oils (20-45%) with poor fuel efficiency, such as phenols, aromatics, aldehydes, ketones, esters, acids and alcohols. Hence, those bio-oils have to be upgraded, mainly through hydrodeoxygenation (HDO), which is a thermal catalytic process under H2 pressure leading to the production of oxygen-free molecules and water as by-product. Our objective is to elaborate new catalytic formulations for the HDO process by combining Density Functional Theory (DFT) calculations with experimental studies (synthesis, characterization, and catalytic tests). HDO of phenolic compounds includes two deoxygenation routes: (i) the hydrogenation of aromatic ring before C-O bond cleavage (HYD) or (ii) the direct C-O bond cleavage - Direct DeOxygenation (DDO). The present work aims to promote the second route in order to increase the production of aromatics under lower H2 pressure. As described in the literature, Fe@silica catalysts show a good activity and selectivity toward aromatics, with being rather resistant towards deactivation. Thus, our work focused on designing silica-supported metallic Fe (or bimetallic Fe-Cu) catalysts with specific properties allowing higher performances.
The adsorption energies of phenol (as a model molecule of bio-oils) and inhibitors (CO and water) over several amorphous and crystalline silica surfaces, with various silanol densities and types, were computed by DFT. Three interaction modes were investigated for phenol: the “perpendicular O-interaction”, the “flat π-interaction”, and the “flat O-interaction”. Results show that crystalline surfaces promote the HYD route since the “flat π-interaction” dominates. For amorphous silica, the highest adsorption energies were found for the “flat O-interaction” mode, and a specific interaction of 120 kJ/mol (including a C-Si bond and phenol deformation) was observed for surfaces with a silanol density between 2 and 3.3 OH/nm2. The CO competitive adsorption is negligible for all silica surfaces, which make them more attractive than conventional sulfide catalysts. Water shows an inhibiting effect for all surfaces whose silanol density is higher than 5 OH/nm2. Hence, these results motivated the synthesis of silica-supported catalysts with a silanol density between 2 and 4 OH/nm2.
Single iron atom catalysts (SACs) supported on mesoporous silica were elaborated following the Sol-gel mechanism using non-ionic (P123) and metallic (CTAF) surfactants as porogens. Iron distribution within mixed P123/CTAF micelles, which is used as template for the silica framework polymerization, allows the fine dispersion of those atoms within silica mesopores. Tuning the P123/CTAF molar ratio enables to control the iron load, as well as the silica structural properties.
Résumé - Abstract
Therefore, different micellar solutions were prepared and analyzed (critical micellar concentration CMC, dynamic light scattering DLS, small angle neutron scattering SANS, magnetic studies) in order to understand their structuration depending on this ratio. The lack of iron clusters and nanoparticles was confirmed by synchrotrons - pair distribution function PDF, scanning transmission electron microscopy STEM mapping, magnetic and nuclear magnetic resonance NMR measurements. Both experimental results and DFT calculations confirmed that those atoms are predominately present as Fe(III) with high spin. Obtaining single metallic iron atoms increases the number of active sites, which improves the catalytic performance. However, the results of catalytic tests were unsatisfactory because the reduction of Fe(III) appeared to be difficult.
Silica-supported metallic (Fe & Cu) and bimetallic (Fe-Cu) catalysts were synthesized by simple impregnation, and Fe catalysts were elaborated following the co-precipitation method with urea decomposition under heating, enabling obtaining Fe(II) species. Catalysts were tested for guaiacol (as a model molecule of bio-oils) HDO conversion and results proved that bimetallic Fe-Cu have a better performance (90% conversion, 70% phenol selectivity) than iron-based catalysts since incorporation of Cu facilitates the reduction of Fe(III) species into Fe(0). Similarly, catalysts prepared by co-precipitation with urea decomposition proved a better performance than those prepared by impregnation since the reduction of iron species was easier.