Description | Abstract The increasing adoption of data-science, machine learning, and artificial intelligence within chemical engineering and materials science promises to drastically accelerate materials discovery and technology translation. In order to continue advancing data-driven materials developments, high- throughput experimentation (HTE) and automation throughout complete laboratory workflows must be developed and widely adopted by both new and established researchers. In this talk I will outline examples and experiences showcasing how researchers in our group are developing and adapting hardware and software infrastructure to accelerate the pace of materials discovery in soft-matter systems (i.e. colloids, polymers, complex fluids and nanomaterials). The talk will highlight recent research examples related to the implementation of HTE for colloidal formulation/synthesis and electrolyte design. It will also highlight outstanding challenges that emerge when transitioning from traditional ‘wet-laboratory’ practices to HTE, and AI-driven experimentation. These include adapting specialized experimental methods to HTE, developing key skills within the research workforce, adoption of good data stewardship practices, financial and infrastructure obstacles, needs for autonomous data treatment, algorithms for automatic modeling and analysis, and others. Conversely, I will also highlight the numerous opportunities that emerge for enhancing virtual collaboration, enabling open data/hardware/software sharing, tackling challenging irreducible problems (e.g. optimization of complex formulations), and the promise of implementing autonomous ‘self-driving’ laboratories. |
---|