Abstract
Supramolecular control (e.g., co-crystallisation or co-amorphisation) for pharmaceuticals holds the opportunity to enhance the therapeutic potential of an active pharmaceutical ingredient as well as other physical and biochemical properties (such as permeability, bioavailability, etc.). By applying mechanochemical synthetic methods for supramolecular control, it is possible to forgo the use of hazardous solvents while enabling unseen reactivities. The mechanochemical design of pharmaceutical co-crystals and co-amorphous mixtures offers high flexibility in terms of the possible structures that can be created through the combination of two or more molecules. This design flexibility has shown the potential to generate innovative drugs in an environmentally friendly manner. The literature has shown random studies that have merely demonstrated the potential benefits of mechanochemistry; however, predicting reactivity and a mechanistic understanding of mechanochemical reactions remains the biggest challenge in the field. To harness the potential of mechanochemistry, this Ph.D. project aimed at developing data-driven models (e.g., machine learning) capable of predicting the mechanochemical supramolecular control and gaining insights into the rules behind solid-state reactions. The developed models demonstrate the accurate prediction of fast-occurring solid-state reactions, that exceed random guessing by far and open the possibility of more efficient and economic pharmaceutical manufacturing.Additionally, novel green methods (i.e., concurrent antisolvent coaxial electrospray and metal oxides as additives during crystallisation) for supramolecular control were explored to shape the way sustainable chemistry is conducted within the industry.
| Date of Award | 28 Jun 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Bernardo Castro Dominguez (Supervisor) & Chris Chuck (Supervisor) |