DU
D.M. Ullersma
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
This master thesis looks at gallium Supported Catalytically Active Liquid Metal (SCALM) systems which show great promise in different catalytic applications, specifically propane dehydrogenation is taken as its main focus. Machine learning interatomic potential models (MLIP) provide a way to accurately simulate large atomic systems at long time scales. This new and improved speed provides the possibility to investigate dynamic systems. A selection of the best currently available models were benchmarked on computational efficiency and accuracy to make an informed choice of MLIP model architecture. Models were finetuned further to improve accuracy with DFT data. Using the best model, MD simulations were performed on gallium nanoparticles of different sizes and solutes. Pt, Pd and Ag were observed to equilibrate in the subsurface as was seen in previous literature. Breaking from this trend, gold was observed to reside in the surface and bismuth was seen to migrate to the surface of the nanoparticle. Clustering MACE descriptors showed it could accurately discern 4 groups: solute atoms, surface atoms, high and low coordination bulk. Furthermore, propane was added to a GaPt SCALM system to capture the dynamic site formation using MD, but none were recorded. The entire computational workflow was designed in Python and can serve as a basis for (multi-architectural) MLIP nanoparticle computational workflows.
...
This master thesis looks at gallium Supported Catalytically Active Liquid Metal (SCALM) systems which show great promise in different catalytic applications, specifically propane dehydrogenation is taken as its main focus. Machine learning interatomic potential models (MLIP) provide a way to accurately simulate large atomic systems at long time scales. This new and improved speed provides the possibility to investigate dynamic systems. A selection of the best currently available models were benchmarked on computational efficiency and accuracy to make an informed choice of MLIP model architecture. Models were finetuned further to improve accuracy with DFT data. Using the best model, MD simulations were performed on gallium nanoparticles of different sizes and solutes. Pt, Pd and Ag were observed to equilibrate in the subsurface as was seen in previous literature. Breaking from this trend, gold was observed to reside in the surface and bismuth was seen to migrate to the surface of the nanoparticle. Clustering MACE descriptors showed it could accurately discern 4 groups: solute atoms, surface atoms, high and low coordination bulk. Furthermore, propane was added to a GaPt SCALM system to capture the dynamic site formation using MD, but none were recorded. The entire computational workflow was designed in Python and can serve as a basis for (multi-architectural) MLIP nanoparticle computational workflows.
A large portion of plastic waste is burned or put into a landfill, which are unsustainable practices. Recycling is a good solution to increase circularity, but currently a significant part of plastic is recycled mechanically. Mechanical recycling reduces the quality of the plastic, consequently plastic can only be recycled a few times before it needs to be discarded. This significant downside can be solved by chemical recycling aided by catalysis. The investigated zirconium catalyst (BuCp2ZrH – OSi) is supported on an amorphous silica surface. The ISE group at the TU Delft has developed a configurational space exploration algorithm that has found states of this supported catalyst, which are far more thermodynamically stable compared to a chemical guess. Many stable states are the result of the hydride transfer from Zr to Si. The thermodynamic stability does not only determine if the state can be accessible at the reaction conditions (80∘C), the kinetics must also be considered as well. To determine the kinetic accessibility, the hydride transfer activation energies were estimated for various amorphous silica surfaces as well as perfect beta-cristobalite. The Cristobalite and minimum strain silica surface showed relatively high energy reaction barriers (≈96-107 kJ/mol) compared to the higher strain surfaces (≈15-54 kJ/mol). This shows that the reaction barrier is highly dependent on the surface structure. The found barriers show that the configurational space exploration algorithm can find kinetically accessible states at reaction conditions.
...
A large portion of plastic waste is burned or put into a landfill, which are unsustainable practices. Recycling is a good solution to increase circularity, but currently a significant part of plastic is recycled mechanically. Mechanical recycling reduces the quality of the plastic, consequently plastic can only be recycled a few times before it needs to be discarded. This significant downside can be solved by chemical recycling aided by catalysis. The investigated zirconium catalyst (BuCp2ZrH – OSi) is supported on an amorphous silica surface. The ISE group at the TU Delft has developed a configurational space exploration algorithm that has found states of this supported catalyst, which are far more thermodynamically stable compared to a chemical guess. Many stable states are the result of the hydride transfer from Zr to Si. The thermodynamic stability does not only determine if the state can be accessible at the reaction conditions (80∘C), the kinetics must also be considered as well. To determine the kinetic accessibility, the hydride transfer activation energies were estimated for various amorphous silica surfaces as well as perfect beta-cristobalite. The Cristobalite and minimum strain silica surface showed relatively high energy reaction barriers (≈96-107 kJ/mol) compared to the higher strain surfaces (≈15-54 kJ/mol). This shows that the reaction barrier is highly dependent on the surface structure. The found barriers show that the configurational space exploration algorithm can find kinetically accessible states at reaction conditions.