Quantum dots (QDs) are composed of a semiconducting core coated in an organic ligand shell and are of interest for next-generation sensing, photovoltaic, and computing devices. The ligand shell plays a critical role in determining the properties and processability of QDs. For example, ligands impart colloidal stability to QD dispersions, mediate the self-assembly of QDs into solid structures, passivate surface trap states on optically active QDs, and influence chemical reactivity. Despite its importance, the ligand shell is notoriously difficult to characterize using traditional methods such as electron microscopy and small-angle X-ray scattering due to the sensitivity of these techniques to atomic mass, which renders the light, organic shell invisible compared to the heavy, inorganic core. By comparison, small-angle neutron scattering (SANS) offers a uniquely sensitive and tunable platform to study the organic ligand shell of QDs.
In my thesis, I’ve applied SANS to quantify how the ligand shell thickness of lead sulfide QDs varies with core radius and solvent as well as to confirm the presence of a thin epitaxial shell on the QD surface. This structural information accelerates materials understanding and design. Yet despite the rich information content of SANS data, SANS methods are still nascent in the field of colloidal QDs because the experiments require careful design and time-consuming data analysis. Thus, in order to make this powerful technique accessible to a broader community, there is a need for a theory-based framework for SANS experimental design.
I addressed this need by developing a statistics-based framework to aid in SANS experimental design. With this framework, optimal experimental parameters such as sample composition and measurement range are chosen to maximize the statistical information of the data. When applied to my previously studied system of lead sulfide QDs, the design framework predicts that two sample variations – the points of maximum and minimum particle contrast – are sufficient to confidently determine QD structure, consistent with experimental observations. However, we find that data requirements increase non-intuitively with added material complexity, such that multi-component systems require a multi-dimensional design space. The distilled learnings from this statistical framework are generalizable to many types of materials and lower the barrier to entry for new SANS users.
In summary, my thesis has revealed the organic ligand shell structure of lead sulfide QDs, enabled by statistically optimized SANS methods which are applicable to a broad range of materials.