Kinetic Monte Carlo Simulations as a Tool for Unraveling the Impact of Solvent and Temperature on Polymer Topology for Self‐Initiated Butyl Acrylate Radical Polymerizations at High Temperatures

Abstract

High‐temperature butyl acrylate polymerizations in bulk and in solution are investigated experimentally and by kinetic Monte Carlo (kMC) simulations. The experimental data comprise conversion‐time data, molar mass distributions, and branching levels per polymer chain derived from size‐exclusion chromatography with a multiangle laser light scattering detector. A kMC model is established, which allows for the description of the impact of solvent and temperature on molar mass distribution as well as type and content of macromonomers. Within the study kinetic coefficients for transfer to solvent and the thermal self‐initiation of the monomer are determined according to the Metropolis Hastings algorithm. The kMC simulations provide information, which are otherwise not accessible, for example, the number of branch points per molecule as a function of molar mass or the molar mass distribution of various macromonomer species. Moreover, molar ratios of mid‐chain and chain‐end radicals are at hand for temperatures up to 160°C, which are important for the interpretation of the experimentally and via simulation‐derived polymer topology as a function of molar masses.

Reverse Engineering of Radical Polymerizations by Multi-Objective Optimization

“…The search for candidate solutions will be performed with the help of genetic algorithms. Moreover, the consideration of more complex microstructural details, e.g., such as branching in the high-temperature polymerization of acrylate [36], may require the number of objectives in MOO to be increased.…”

AI-Based Forecasting of Polymer Properties for High-Temperature Butyl Acrylate Polymerizations

“…19 mcPolymer has a proven track record in modeling various polymerization processes, including reversible deactivation radical polymerizations, 19,56 acrylate polymerizations, 57 and self-initiated butyl acrylate polymerizations. 20 Previously, the kMC simulator was successfully used for generating training data for ML methods 32 and for solving a reverse engineering problem by multiobjective optimization. 58 While kMC simulations generate highly detailed information on polymer microstructure, the raw data are not directly usable for training ML models.…”

“…Due to the complex microstructure of the resulting polymers the memory demand increases and can easily reach 100 GiB and more. 20 Therefore, it is favorable to combine kMC simulations with AI-based forecasting techniques that are much faster after they are trained. AI offers exciting possibilities for advancements in polymer research.…”

A Signal‐To‐Noise‐Ratio‐Based Automated Algorithm to accelerate Kinetic Monte Carlo Convergence in Basic Polymerizations

“…Other typical discriminants are the chemical composition (e.g., comonomer content) or topological features (e.g., the number of short-chain branches). [12,[29][30][31][47][48][49][50] In PRE, the kMC method has been successfully applied for both polymerization (e.g., homogeneous free radical, [11,15,[51][52][53] nitroxide mediated, [54][55][56] atom transfer radical polymerization, [57][58][59][60][61] and dispersed phase thus heterogeneous polymerization [62][63][64][65][66][67][68] ), polymer modification [27,69,70] as well as polymer recycling, both via chemical and mechanical means. [71][72][73][74] As opposed to conditional Monte Carlo, [75][76][77] which deals with a posteriori mimicking of chemical structures simplifying the natural flow of reaction kinetics, the kMC method is based on the stochastic simulation algorithm (SSA), as proposed in the seminal work of Gillespie for reactions between elemental species.…”

“…Other typical discriminants are the chemical composition (e.g., comonomer content) or topological features (e.g., the number of short‐chain branches). [ 12,29–31,47–50 ] In PRE, the k MC method has been successfully applied for both polymerization (e.g., homogeneous free radical, [ 11,15,51–53 ] nitroxide mediated, [ 54–56 ] atom transfer radical polymerization, [ 57–61 ] and dispersed phase thus heterogeneous polymerization [ 62–68 ] ), polymer modification [ 27,69,70 ] as well as polymer recycling, both via chemical and mechanical means. [ 71–74 ]…”