Structure Prediction of Organic/Inorganic Interfaces from First Principles

Functional hybrid inorganic-organic systems(HIOSs), especially molecular thin films on inorganic substrates material have critical applications in electronic and optical devices. Therefore, understanding the structure and electronic properties of HIOSs are highly important. Molecule thin film as an essential component of HIOs often possess similar properties to its bulk crystal form and therefore making it also important to understand the structure-property relationship for bulk molecular crystals. Here in the thesis, we present four pieces of work – two related to molecular crystal structure prediction and the other two related to molecular interface structure prediction. Altogether, they help to promote the understanding of interactions in molecular crystals and molecular interfaces, potentially guide the synthesis efforts in molecule- related devices in promising directions. For molecular crystal prediction, Two pieces of code developed before, Genarris 2.0 and GAtor, are used to predict the structure of three energetic materials - TATB, DATB and the chiral arene, 4,5- dimethylphenanthrene. This is the first time that the crystal structure prediction workflow is applied to the structures with distinct nitro group interactions, and packing in particularly dense crystal structures. For structure prediction of HIOs, we extend the Genarris 2.0 for molecular crystals to Genarris 4.0 – a random structure generator for organic/inorganic interfaces. A structure generation and down-selection workflow is developed such that the experimental structure of three representative HIO systems - TCNE/Au(111), PTCDA/Ag(111), and naphthalene/Cu(111) are successfully generated. Ogre, a code we developed for molecular surface generation, is also incorporated in the workflow. For all three cases, the generated interface structures show good agreement with experimental STMs. In addition, the adsorption site, electronic properties including charge transfer and frontier molecular level alignment are also identified.