Experimental Design in Polymer Chemistry—A Guide towards True Optimization of a RAFT Polymerization Using Design of Experiments (DoE)

Abstract

Despite the great potential of design of experiments (DoE) for efficiency and plannability in academic research, it remains a method predominantly used in industrial processes. From our perspective though, DoE additionally provides greater information gain than conventional experimentation approaches, even for more complex systems such as chemical reactions. Hence, this work presents a comprehensive DoE investigation on thermally initiated reversible addition–fragmentation chain transfer (RAFT) polymerization of methacrylamide (MAAm). To facilitate the adaptation of DoE for virtually every other polymerization, this work provides a step-by-step application guide emphasizing the biggest challenges along the way. Optimization of the RAFT system was achieved via response surface methodology utilizing a face-centered central composite design (FC-CCD). Highly accurate prediction models for the responses of monomer conversion, theoretical and apparent number averaged molecular weights, and dispersity are presented. The obtained equations not only facilitate thorough understanding of the observed system but also allow selection of synthetic targets for each individual response by prediction of the respective optimal factor settings. This work successfully demonstrates the great capability of DoE in academic research and aims to encourage fellow scientists to incorporate the technique into their repertoire of experimental strategies.

A Brief Introduction to Chemical Reaction Optimization

“…As DoE builds a statistical model, it has only an empirical meaning rather than physical, therefore, there is no ingrained physicochemical information about the optimized process within the model; this means that model responses are only considered to be accurate within the explored bounds of the experimental factors. , For example, if the reaction time is explored as a factor as part of a DoE study between the bounds of 5–30 min, extrapolating the model to predict responses after 60 min would likely result in inaccuracies, and further study must be conducted to predict these outputs. Furthermore, the number of experiments that are required to be performed in parallel for some DoE studies may be large, depending on the amount of reaction material or time required to conduct these experiments, this may be prohibitive in some circumstances .…”

CIME4R: Exploring iterative, AI-guided chemical reaction optimization campaigns in their parameter space

“…3D). For example, all values of the acquisition function for each cycle are visualized in one heatmap that changes colors corresponding to the timesteps along the x-axis 4 . The progression over time is thus visualized in a compact way and users can investigate iterative data (T2).…”

“…The fundamental challenge of RO lies in its expansive search space, which consists of a multitude of categorical (e.g., choice of reagent, base, catalyst, methods) and continuous (e.g., temperature, concentration, reagent equivalent) parameters. Some common approaches [1] to solving such a task are high-throughput experimentation (HTE) [2], one factor at a time (OFAT) [3], design of experiment (DoE) [4], and, more recently, AI-guided optimization [5][6][7]. All approaches share the ultimate goal of identifying one or more optimal solutions within the high-dimensional parameter space.…”

CIME4R: Exploring iterative, AI-guided chemical reaction optimization campaigns in their parameter space

“…The fundamental challenge of RO lies in its expansive search space, which consists of a multitude of categorical (e.g., choice of reagent, base, catalyst, methods) and continuous (e.g., temperature, concentration, reagent equivalent) parameters. Some common approaches [1] to solving such a task are high-throughput experimentation (HTE) [2], one factor at a time (OFAT) [3], design of experiment (DoE) [4], or, more recently, AI-guided optimization [5,6,7]. All approaches have the ultimate goal in common to identify optimal solution(s) within the high-dimensional parameter space.…”