guest ::
%0 Thesis %A Nevolianis, Thomas %T Harnessing the power of quantum chemistry and machine learning for improved synthesis and characterization of functional microgels; 1. Auflage %V 58 %I Rheinisch-Westfälische Technische Hochschule Aachen %V Dissertation %C Aachen %M RWTH-2025-10126 %@ 978-3-95886-552-5 %B Aachener Beiträge zur technischen Thermodynamik %P 1 Online-Ressource : Illustrationen %D 2026 %Z Druckausgabe: 2026. - Auch veröffentlicht auf dem Publikationsserver der RWTH Aachen University. - Schreibfehler im Übersetzungstitel des Dokuments: funktionelen %Z Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2025 %X Microgels are versatile soft polymer networks with applications in catalysis, drug delivery, and enzyme immobilization. Their functionality and behavior are closely linked to synthesis conditions, which determine the conditions under which they swell and the extent of that swelling. This thesis advances the predictive modeling of microgel synthesis by integrating quantum chemical calculations and machine learning, providing quantitative, model-based guidance grounded in computational and experimental data for designing microgels with tailored swelling behavior and other desired properties, thereby enhancing their utility in both industrial and research settings. First, it is demonstrated that the monomer distribution within N-vinylcaprolactam-based microgels functionalized with glycidyl methacrylate can be tuned using an existing synthesis model that integrates quantum chemical calculations with experimental measurements, enabling precise control over the microgel’s structure and properties. Second, deuteration-induced changes in the Flory interaction parameter are quantified using quantum chemical calculations to predict swelling behavior in deuterated microgels, and these predictions are validated against experimental data, providing insights into the differences in responsiveness, structure, and softness between deuterated and hydrogenated microgels. Third, new methods are developed that significantly improve the accuracy of solvation free energy predictions for ionic solutes; crucial for modeling reaction kinetics and thermochemistry in ionic microgel domains and thereby facilitating the design of functional microgels with tailored ionic domains. Finally, multi-fidelity learning approaches are introduced that leverage quantum chemical and experimental data to improve the predictability of toluene/water partition coefficients; these approaches are transferable to other solvent systems, crucial for designing microgels with hydrophobic domains. Together, this work demonstrates how integrating computational chemistry with experimental validation can optimize microgel synthesis and functionality, paving the way for advances in various applications. %F PUB:(DE-HGF)11 ; PUB:(DE-HGF)3 %9 Dissertation / PhD ThesisBook %R 10.18154/RWTH-2025-10126 %U https://publications.rwth-aachen.de/record/1022617