Challenges and opportunities of polymer design with machine learning and high throughput experimentation
- PDF / 924,239 Bytes
- 8 Pages / 612 x 792 pts (letter) Page_size
- 88 Downloads / 147 Views
rtificial Intelligence Prospective
Challenges and opportunities of polymer design with machine learning and high throughput experimentation Jatin N. Kumar , Institute of Materials Research & Engineering, 2 Fusionopolis Way, #08-03, 138634, Singapore Qianxiao Li, and Ye Jun, Institute of High-Performance Computing, 1 Fusionopolis Way, #16-16, 138632, Singapore Address all correspondence to Jatin N. Kumar at [email protected] and [email protected] (Received 18 January 2019; accepted 17 April 2019)
Abstract In this perspective, the authors challenge the status quo of polymer innovation. The authors first explore how research in polymer design is conducted today, which is both time consuming and unable to capture the multi-scale complexities of polymers. The authors discuss strategies that could be employed in bringing together machine learning, data curation, high-throughput experimentation, and simulations, to build a system that can accurately predict polymer properties from their descriptors and enable inverse design that is capable of designing polymers based on desired properties.
Introduction Polymers are ubiquitous in their use spanning a vast range of materials due to their highly tunable physical and chemical properties. They are critical enablers of many modern and emerging technologies today, ranging from structural material used to build modern aircraft, to flexible electronics, and even emulsifiers in personal care products—arguably making them one of the most important classes of material that exist. The properties that govern their specific use are influenced by the architecture or structure,[1] made up of the following four parameters: topology, composition, functionality, and size (Fig. 1). Fine control of polymer architecture and a deep understanding of its relationship with physiochemical properties is what allows for the innovation of materials with novel properties with important end-applications.[2–7] Polymer properties are influenced by either polymer dynamics or chemo-functionality or both. Polymer dynamics are governed by the entanglement and flexibility of the polymer chain, which in turn influence physical properties such as rheology, glass transition temperature, and mechanical properties. Chemo-functionality affords the polymer chains’ chemical response, biologic activity, and electro-chemical activity. A combination of the two results in interesting morphological and phase characteristics, as well as stimuli-responsive characteristics, which have been thoroughly exploited in drug delivery.[8] Apart from understanding structure–property relationships, the mechanism, and kinetics of polymerization, the process of polymer synthesis is also a critical area of study of importance when preparing tailored material to fine-tune polymer properties.
For instance, copolymerization compositional relationships between your reaction feed and final product is governed by the Mayo–Lewis equation, which relates the two monomers with their reactivity ratio, a function of their propagation rate constants.
Data Loading...
