Continuous flow synthesis of colloidal semiconductor nanocrystals: towards autonomous experimentation for accelerated material discovery

Vikram, Ajit

Permalink

https://hdl.handle.net/2142/110820 Copy

Description

Title

Continuous flow synthesis of colloidal semiconductor nanocrystals: towards autonomous experimentation for accelerated material discovery

Author(s)

Vikram, Ajit

Issue Date

2021-04-18

Director of Research (if dissertation) or Advisor (if thesis)

Kenis, Paul J. A.

Doctoral Committee Chair(s)

Kenis, Paul J. A.

Committee Member(s)

• Seebauer, Edmund G.
• Shim, Moonsub
• Diao, Ying

Department of Study

Chemical & Biomolecular Engr

Discipline

Chemical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Quantum Dots
• Artificial Intelligence
• Autonomous Experimentation

Abstract

Colloidal semiconductor nanocrystals (NC) with tunable optical and electronic properties are opening up exciting opportunities for high-performance optoelectronics, photovoltaics, and bioimaging applications. While significant advancements have been made in this area in the past two decades, the vast majority of work has focused on synthesis of semiconductor NCs containing toxic heavy-metal elements such as cadmium and lead. The inherent toxicity of heavy-metal-based colloidal NCs and the associated regulatory constraints have largely impeded their widespread implementation. Therefore, the discovery of efficient synthesis routes to design high quality heavy-metal-free semiconductor NCs is critical to enable their application for various technologies such as solar cells, infrared sensors, and display technologies. Identifying the optimal synthesis and screening of syntheses recipe for these complex NCs, however, remains one of the major bottlenecks for discovery of colloidal NCs. Conventional batch rector-based synthesis screening strategies are often guided by limited understanding of the underlying growth mechanisms. Such approaches are expensive in both time and resources, and thus significantly impede the overall material discovery process. This dissertation focuses on developing an autonomous flow synthesis approach that enables accelerated screening of synthesis recipes, while also providing key kinetic insights into the underlying chemistry. In Chapter 2, a modular millifluidic reactor that leverages precise control over reaction conditions for a fully continuous multi-step synthesis of high quality InP/ZnSeS core-shell NCs exhibiting PL QYs up to 67% is reported. Using the synthesis insights gained from this work, a new pathway for synthesis of InP NCs that exhibit a very rare size-focused growth mechanism is developed in Chapter 3. Molecular insights using NMR and UV-Vis spectroscopy revealed that by addition of trace amounts of water to the reaction mixture, the reactivity of the precursors can be controlled to synthesize NCs with unprecedent control over their size distribution. In Chapter 4, an automated reconfigurable flow reactor platform integrated with inline UV-Vis and PL spectroscopy is designed to develop an efficient shell growth strategy for core/shell NCs. Feeding individual precursors into the reactor channel in a sequential fashion combined with real-time reaction monitoring enabled precise control over layer-by-layer shell passivation of the InP NCs. Further investigation using FTIR, liquid NMR, and solid NMR revealed the origin of interfacial defects and their impact on optical properties of InP-based NCs. In Chapter 5, an Artificial Intelligence-based decision-making feedback module was integrated with the automated flow reactor to develop an autonomous experimentation platform, specifically designed for accelerating screening and discovery of colloidal QDs. The autonomous platform learns the synthesis parameter space through self-driven experiments and synthesizes QDs with user-specified band gap and polydispersity without-any-prior-knowledge of the synthesis chemistry. Using a closed-loop iterative framework, it executed a minimal number of self-driven iterative experiments (28 experiments) through continuous operation (44 hours) to learn the entire synthesis parameter space for predicting the reaction outcomes for more than 100,000 different combinations of synthesis conditions. Finally, in Chapter 6, a summary of insights gained in this dissertation along with a comprehensive proposal for future research directions arising from this dissertation, is provided. Overall, the studies reported in this dissertation provide insights into flow synthesis of indium phosphide nanocrystals as well as provide an artificial-intelligence-guided autonomous experimentation approach that can be utilized to accelerate the discovery of colloidal nanocrystals in future.

Graduation Semester

2021-05

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/110820

Copyright and License Information

Copyright 2020 Ajit Vikram

Owning Collections