Structural Effects on Optical Properties—Insights from Correlative Optical and Electron Microscopies

Synthetic control over quantum dot (QD) nucleation and growth generates highly crystalline particles with narrow size distributions while subsequent shelling passivates the QD surface.^1,2 Together, these techniques have yielded particles with near-unity photoluminescence quantum yield (PLQY), high photostability, and tailored surfaces.³ These advances have directly contributed to the commercialization of QDs in display technologies. However, such optimizations target ensemble observables, hindering microscopic insight into the underlying structural factors that give rise to these superior properties. Here, single-particle techniques offer an avenue to build ensemble statistics from individual observations. Specifically, correlating between particle structure and photoluminescence at the single-dot level will provide a direct link between the anatomy of a QD and its resulting optical properties, thereby providing specific targets for synthetic optimization.⁴<br/>Here, we combine single particle optical spectroscopy and scanning transmission electron microscopy (STEM) to reveal the link between QD structure and optical properties with high spatial and temporal resolution. Using fiduciary markers to reproducibly observe the same QDs in both optical and electron microscopes, we find that QDs with ON fractions &gt;90% are observably defect-free and show regular, uninterrupted crystallinity at the atomic level. In addition, these particles are routinely well-passivated by shells and lack exposed core facets. On the other hand, QDs with ON fractions &lt;80% exhibit characteristic structural defects such as under-passivated core facets, stacking faults, or irregular shell growth. Monitoring the dynamics of the exciton in these QDs, we find that the lifetime and linewidth of the excitonic emission is correlated with the QD structure. Specifically, defected particles show shorter exciton recombination lifetimes and broader linewidths, consistent with these defects serving as parasitic recombination sites. Finally, we identify a permanently non-emissive sub-ensemble of QDs which are only directly observed <i>via</i> electron microscopy but nonetheless contribute to photon absorption. Though we find that a subset of these “dark” particles suffer from exacerbated structural defects, a separate population intriguingly does not show clear structural anomalies yet remains non-emissive. Importantly, these results demonstrate that non-structural considerations such as the identity and extent of ligand coverage also play a decisive role during photon emission.<br/><br/>Our results offer a microscopic view of the ensemble and demonstrate that both the structure of the core and shell should be highly crystalline and free of defects and that this condition is necessary but insufficient to guarantee efficient photon emission. Now, we are exploring the effects of photoinduced structural dynamism, particularly at the particle surface.^5,6 Photoexcitation has been shown to induce anisotropic lattice displacement in QDs which can persist on timescales up to exciton recombination lifetimes. Using an electron microscope equipped with an optical fiber, we image QDs under laser irradiation. Our preliminary results suggest that some fraction of photoexcitations become trapped, leading to structural disorder which is captured in electron diffraction. Taken together, our technique provides for a direct method to correlate between the structure of an individual QD and its optical properties. Our methods offer an avenue to build ensemble statistics from single particle observations, thereby circumventing ensemble averaging techniques and offering microscopic insight into the structure-function relationships of emissive nanomaterials.<br/>1) https://doi.org/10.1021/jp9530562.<br/>2) https://doi.org/10.1021/ja970754m.<br/>3) https://doi.org/10.1126/science.aac5523.<br/>4) https://doi.org/10.1021/nn506420w.<br/>5) https://doi.org/10.1021/acs.jpcc.1c07064.<br/>6) https://doi.org/10.1038/s41467-021-22116-0.