UCLA

Glasses can be made of virtually all the elements of the periodic table, provided that a melt is cooled fast enough from the liquid state. The number of possible glass compositions is virtually infinite. Although such a large compositional space offers limitless opportunities to develop novel glasses with improved functionalities, it also comes with some challenges, since the large number of possible compositions render traditional “trial and error” Edisonian approaches poorly efficient. As a goal of this thesis, overcoming the limit of empirical approaches of glass design requires the development of accurate and transferable predictive models linking glasses’ composition and structure to their macroscopic property, is crucially important to the glass science community.