Intelligent Machine Learning: Tailor-Making Macromolecules

Abstract

Nowadays, polymer reaction engineers seek robust and effective tools to synthesize complex macromolecules with well-defined and desirable microstructural and architectural characteristics. Over the past few decades, several promising approaches, such as controlled living (co)polymerization systems and chain-shuttling reactions have been proposed and widely applied to synthesize rather complex macromolecules with controlled monomer sequences. Despite the unique potential of the newly developed techniques, tailor-making the microstructure of macromolecules by suggesting the most appropriate polymerization recipe still remains a very challenging task. In the current work, two versatile and powerful tools capable of effectively addressing the aforementioned questions have been proposed and successfully put into practice. The two tools are established through the amalgamation of the Kinetic Monte Carlo simulation approach and machine learning techniques. The former, an intelligent modeling tool, is able to model and visualize the intricate inter-relationships of polymerization recipes/conditions (as input variables) and microstructural features of the produced macromolecules (as responses). The latter is capable of precisely predicting optimal copolymerization conditions to simultaneously satisfy all predefined microstructural features. The effectiveness of the proposed intelligent modeling and optimization techniques for solving this extremely important ‘inverse’ engineering problem was successfully examined by investigating the possibility of tailor-making the microstructure of Olefin Block Copolymers via chain-shuttling coordination polymerization.

Translating Simulated Chain Length and Molar Mass Distributions in Chain‐Growth Polymerization for Experimental Comparison and Mechanistic Insight

“…In this way, a large ensemble of initial molecules can be easily used (e.g., above 10 6 ), guaranteeing a correct sampling of (stochastic) reaction probabilities and, hence, an accurate description of the polymerization kinetics. [98,110,111,[118][119][120] In the present contribution, the aim is to raise awareness in both the theoretical and experimental research community that several CLD and MMD representations exist and a correct translation from one representation to the other should be performed with care. Specific translation procedures are outlined and applied for polymerization recipes with basic and more complex reaction schemes.…”

Toward Olefin Multiblock Copolymers with Tailored Properties: A Molecular Perspective

“…The microstructural characteristics of OBCs can be classified into two categories: (i) topology‐related molecular variables, including the number of linkage points per chain (LP), the average degree of polymerization of soft segments (), the average degree of polymerization of hard segments (), ethylene sequence length of soft segments (), and ethylene sequence length of hard segments (); and (ii) property‐related molecular variables, including the comonomer content of the soft segments (), comonomer content of the hard segments (), the average longest ethylene sequence length of the soft blocks (), the average longest ethylene sequence length of the hard blocks (), and hard block percentage (HB%). [ 18 ] The topology‐related characteristics were comprehensively discussed in a previous publication by patterning macromolecular landscape of OBCs. [ 19 ] Nevertheless, there is sufficient evidence that physical, thermal, rheological, and mechanical properties of OBCs are mostly controlled by the characteristics of the second group.…”

Machine Learning-Based Detection of Graphene Defects with Atomic Precision

“…Compared to elliptical holes and cracks which can be shed light on using traditional fracture mechanics, randomly distributed atomic vacancies have a much more implicit but not necessarily less profound impact on the mechanical properties of graphene. Emerging machine learning approaches offer solutions for learning patterns from complex data and have been extensively applied in material design and discovery problems [31][32][33][34][35][36][37][38][39][40]. The power of machine learning-based approaches can be fully utilized with a rational selection of features.…”