C.J. Hanson, S. Jeffrey, Z. Hu, H. Htoon, J. Hollingsworth
Los Alamos National Laboratory,
Keywords: automated synthesis, quantum dots, stable emitters, non blinking, combinatorial chemistry, high-throughput synthesis, scale up
Summary:Semiconducting nanocrystals, also known as quantum dots (QDs), that emit light with near-unity quantum yield and that are extremely photostable are attractive options as down-conversion and direct electricity-to-light materials for a variety of applications, including solid-state lighting, display technologies, bio-imaging and optical tracking. However, to speed their utilization in demanding, large-volume applications, fast methods for materials discovery and optimization are needed. Furthermore, with an optimized material in hand, large-volume synthetic methods are required to meet demand. Thus, the ability to scale-up benchtop chemistry is also critical, as is maintaining batch-to-batch and within-batch consistency. It has been estimated that the development of a hypothetical new nanomaterial requires ~100 reactions to be performed, or 3-12 months assuming reaction times of 1 h and the ability to conduct 2 reactions per day (E. M. Chan Chem. Soc. Rev., 2015, 1653). The situation for the development of heterostructured (multicomponent) nanomaterials is somewhat more grave. Such “complex” nanocrystals are technologically relevant, as they often afford the closest-to-ideal performance properties. For example, we have developed the so-called “giant” QD (gQD), which is an ultra-stable photon source both at the ensemble and single-QD level, i.e., yielding 100% blinking suppression and no photobleaching when exposed at room-temperature or elevated temperatures to high-power laser excitation for extended periods (at least 1 hour) as “bare” solid-state nanocrystals. gQDs further exhibit significantly suppressed non-radiative Auger recombination and minimal self-reabsorption. However, to realize these properties, the optimal synthetic ‘workflow’ (reaction process) requires many separate reagent additions and long anneal times, such that reactions can take ~5-7 days to complete. This is in large part due to interruptions in the workflow resulting from reliance on human operators for manual addition of shell precursors, which is ‘paused’ outside the hours of the typical workday. Despite such inefficiencies, it is still possible to make significant progress in the optimization and development of functional gQDs (e.g., J. Am. Chem. Soc.: 2008, 130, 5026; 2012, 134, 9634; 2015, 137, 3755; 2017, 139, 11081), since gQD workflows are not simply the result of trial-and-error but benefit from rational design. Nevertheless, to address needs for more rapid discovery and optimization, as well as synthesis scale-up, we have developed a fully-automated batch reactor system (FABRS). Utilizing combinatorial chemistry, we can now screen many parameters that may influence nanomaterial performance, including gQD brightness and stability. By automating shell-precursor addition and aliquot sampling, reaction times are shortened by 300%, while in situ probes allow real-time analysis of reaction turbidity, photoluminescence and absorption. FABRS also affords facile and precise control over the rate of precursor addition, stirring speed, and reaction and precursor temperatures. Here, we describe the results of a series of experiments for which various reaction variables were systematically varied and correlated with resulting gQD structural and performance (ensemble and single-QD optical) properties. We demonstrate a rapid process for moving from a large parameter space to a narrow and more ideal parameter space facilitated by novel nanomaterials synthesis automation.