[PHD DEFENSE] Nabil Nassar - ModSpeC

Dynamic behavior investigation of actives species involved in catalytic reaction is still highly challenging since a catalytic pathway is a combination of rapid adsorption and desorption of reactants/intermediates/products on ill-defined active surface sites. The spectroscopic signal arising from non-active sites (spectators) is frequently stronger than that of actives sites, overwhelming key information on catalysis. One way to mitigate this problem is to apply Modulation Excitation Spectroscopy (MES) which consists in a periodic perturbation of the state of the system by the variation of external parameter such as temperature, pressure, concentration, isotopic exchange... This rapid periodic perturbation of the system state will influence only the concentration profile of some species (e.g. active species) which will oscillate at the frequency of the periodic excitation but with a phase delay. The concentration profile of species not responding to the periodic excitation (i.e. spectator species) will remain constant making possible their removal from the global signal by a post data acquisition mathematical treatment known as PSD (Phase Sensitive Detection). A new data treatment methodology is introduced to overcome the limitations of the approach currently applied. The conventional method focuses primarily on the fundamental frequency in phase-sensitive detection, which retains only part of the information contained in the time-domain dataset. In contrast, the proposed methodology will not be limited to the analysis of the fundamental frequency. Instead, it will extend to higher-frequency components, enabling a more comprehensive interpretation of the data.

The first reaction is carbon monoxide oxidation. CO oxidation has been carried out using an alumina-supported platinum catalyst. The isotopic-IR experiment was performed by switching the reactant from ¹²CO to its labelled counterpart ¹³CO at the same concentration, pressure and temperature. PSD helps to clarify the IR spectrum by reducing noise and isolating only the peaks corresponding to the species that react with the system perturbation. The kinetics of exchange between various CO isotope peaks is analyzed by applying the PSD calculation with different harmonics. The analysis of the PSD signal also clearly shows that the exchange between ¹²CO and ¹³CO is a simple reaction when adsorbed in a bridging position between two platinum whereas the situation is much more complex when adsorbed on single atom. The signal of the latter is superposition of different peaks representing different species that evolve at different speed. The interpretation of this region is not limited to PSD analysis, but also incorporates numerical simulation for the treatment of different concentration profiles. This approach provides a clearer understanding of the CO reaction mechanism by simplifying the analysis and interpretation of the active sites involved in the reaction. This first set of experiments allowed to build a robust methodology to treat and interpret the demodulated spectra with an original approach.

The second reaction is the oxidation reaction of methanol with an unsupported catalyst, which is mixed oxide iron molybdate catalyst: Fe₂(MoO₃)₄ - MoO₃ (FeMo). In this work, different catalysts with Mo/Fe ratios were prepared. The synthesized catalysts are characterized using different techniques such as XAS, XRD, ICP and Raman, in order to identify its composition and properties. After identification and characterization, catalytic tests were performed on the different catalysts. The most efficient catalyst was further investigated in the methanol oxidation under isotope ME condition (¹²C methanol and ¹³C-labeled methanol) combined with infrared spectroscopy. This experiment proved the feasibility of such experiments on bulk catalysts and allowed the identification of absorbed methanol on FeMo under Operando conditions.