Autonomous optimization of liquid-handling parameters for accurate transfer of viscous liquids using automated pipette robots Pablo Quijano Velasco 1 , Jayce Jian Wei Cheng 1 , Chan Jie Leong 1 , Aniket Chitre 2,3 , Riko I Made 1 , Kedar Hippalgaonkar 1,4 1 Institute of Materials Research and Engineering, A*STAR (Agency for Science, Technology and Research), Singapore 2 Department of Chemical Engineering and Biotechnology, University of Cambridge, Philippa Fawcett Drive, Cambridge CB3 0AS, United Kingdom 3 Cambridge Centre for Advanced Research and Education in Singapore, CARES Ltd. 1 CREATE Way, CREATE Tower #05-05, Singapore 138602, Singapore 4 Department of Materials Science and Engineering, Nanyang Technological University, Singapore 117575, Singapore Materials Accelerated Platforms (MAPs) are an emerging paradigm for the discovery of materials that combine high-throughput experimentation, automation and artificial intelligence in order to achieve autonomous experimentation 1 . Although there is many commercial equipment that can be used for the development of MAPs, the number of materials that can be processed are limited since most tools have not been designed for the automation of experiments that are specific for Materials Chemistry 2 . For example, most liquid-handling robots for automated solution-based synthesis and processing of materials are engineered to pipette aqueous solutions and are not calibrated for the accurate transfer of viscous liquids. Although accurate handling parameters can be achieved for the transfer of viscous liquids, the process of obtaining these parameters is time consuming and must be performed by a human 3 . To make liquid-handling robots compatibility with autonomous experimentation more robust, new automated methods that calibrate the robots to handle liquids with a wider spectrum of viscosities is required. Herein we present an approach that is data-driven and automated to obtain liquid-handling parameters to accurately transfer viscous liquids using automated pipettes. Our workflow starts with the estimation of the unknown values for the handling parameters of a viscous liquid through a model that correlates a proxy measurement of viscosity with the aspiration and dispense rates for the accurate transfer of such liquid. Afterwards, the liquid-handling parameters are fine-tuned via an iterative gravimetric pipette calibration method performed by the liquid-handling robot autonomously, which is guided by an optimization algorithm. Through this method we can achieve the transfer of viscous liquids with viscosities in the order of 10 to 10 3 cP within a 5 % mass error. We believe that the incorporation of this workflow into MAPs will accelerate the development of autonomous experimentation for Materials Chemistry. References 1. M. Seifrid, J. Hattrick-Simpers, A. Aspuru-Guzik, T. Kalil and S. Cranford, Matter, 2022 ,5, 1-5. 2. M. Christensen, L. P. E. Yunker, P. Shiri, T. Zepel, P. L. Prieto, S. Grunert, F. Bork and J. E. Hein, Chem. Sci., 2021, 12, 15473–15490. 3. How to handle viscous liquids in Protocol Designer, https://support.opentrons.com/s/article/How-to-handle-viscous-liquids-in- Protocol-Designer, (accessed January 23, 2023).
DD09
© The Author(s), 2023
Made with FlippingBook Learn more on our blog