# Automated closed-loop continuous flow block copolymer synthesizer

## TL;DR

This paper introduces an automated system for making diblock copolymers using continuous flow chemistry and machine learning.

## Contribution

The novel contribution is an automated system integrating flow chemistry, machine learning, and in-line FTIR monitoring for efficient BCP synthesis.

## Key findings

- An in-line FTIR method achieved 2% accuracy in monomer conversion compared to NMR.
- A library of 95 BCPs was created with minimal human intervention.
- The system enabled high-throughput synthesis of BCPs with varying hydrophilicities and degrees of polymerization.

## Abstract

A fully automated continuous flow synthesizer for diblock copolymer (BCP) synthesis was constructed comprising elements of flow chemistry, automation, machine learning and in-line monitoring. A new method using in-line FTIR spectroscopic analysis for accurate determination of monomer conversion (with an error as low as 2% relative to an NMR spectroscopic baseline) is presented, thereby generating a reliable feedback system for reaction self-optimisation using the platform. By employing reversible addition–fragmentation chain transfer (RAFT) polymerization at 100 °C, acrylates and acrylamides of different hydrophilicities (namely methyl acrylate, ethyl acrylate, butyl acrylate, 2-ethylhexyl acrylate, 2-hydroxyethyl acrylate, ethylene glycol methyl ether acrylate, diethylene glycol ethyl ether acrylate, 2-(dimethylamino)ethyl acrylate, acrylamide & N,N-dimethylacrylamide) were polymerized to make mixed BCPs, targeting different degrees of polymerization (15 to 100). Samples were collected automatically, and a BCP material library comprising 95 diblock copolymers (7 sets of double hydrophobic, 7 sets of amphiphilic and 3 sets of double hydrophilic monomer systems) with Mn ranging from 1800 g mol−1 to 14 700 g mol−1, was obtained in a high-throughput manner, with minimal human intervention throughout the entire process.

A fully automated continuous flow synthesizer for diblock copolymer (BCP) synthesis was constructed comprising elements of flow chemistry, automation, machine learning and in-line monitoring.