Xiang Huang1 and Shenghong Ju1, 2, a)1- China-UK Low Carbon College, Shanghai Jiao Tong University, Shanghai, 201306, China2- Materials Genome Initiative Center, School of Material Science and Engineering, Shanghai Jiao Tong University, Shanghai, 201306, Chinaa) Author to whom correspondence should be addressed: shenghong.ju@sjtu.edu.cnABSTRACTDesigning polymers with high intrinsic thermal conductivity (TC) is critically important for the thermal management of organic electronics and photonics. However, this is a challenging task owing to the diversity of the chemical space and the barriers to advanced synthetic experiments/characterization techniques for polymers. In this Tutorial, the fundamentals and implementation of combining classical molecular dynamics simulation and machine learning (ML) for the development of polymers with high TC are comprehensively introduced. We begin by describing the core components of a universal ML framework, involving polymer datasets, property calculators, feature engineering and informatics algorithms. Then, the process of constructing interpretable regression algorithms for TC prediction is introduced, aiming to extract the underlying relationships between microstructures and TCs for polymers. We also explore the design of sequence-ordered polymers with high TC using lightweight and mainstream active learning algorithms. Lastly, we conclude by addressing the current limitations and suggesting potential avenues for future research on this topic.
[ * ]n1c(=O)c2cc3c(cc2c1=O)c(=O)n([ * ])c3=O 15 [ * ]c1nc2ccc([ * ])cc2[nH]1 31 [ * ]N1C(=O)c2cc3cc4Cc5cc6cc7C(=O)N([ * ]) C(=O)c7cc6cc5Cc4cc3cc2C1=O
id: 45af22cfd3e003ed54cb62c2fc720e36 - page: 24
A. Establishment of benchmark datasets for triblock polymers Inspired by the knowledge of the interpretable DNN model outcomes in Section III B and the seen high TC polymers, we constructed a library containing 32 polymer motifs, as listed in Table VI. To ensure the uniqueness of the identification for each fragment, these motifs were binary coded from to . Theoretically, we can construct an infinite number of multi-block polymers by controlling the number and order of fragments. However, in consideration of the computational cost and hardware capabilities, we built a benchmark dataset of triblock polymers. Fig. 7(a) depicts the process of producing a triblock polymer, which was characterized by a binary sequence of 15 bits in length. The 15-bit binary sequence was divided equally into three equal parts, each one corresponding 24 / 48 to a fragment. The composition of the polymer fragments is directionless, i.e., two polymers consisting
id: 14515aa310c8acc5ed0b495188f44865 - page: 24
The entire benchmark dataset contains 16896 triblock polymers, which were classified into 13 categories referring to the same classification method as PoLyInfo, such as polyolefins, polyethers and polyimides (in Fig.7(c)). The TC of emerging polymers was estimated by a high-fidelity DNN model trained in Section III B and ranged from 0.16 to 1.03 W m-1K-1, with 42.6% of the TC greater than 0.40 W m-1K-1 (in Fig.7(d)). Moreover, the synthesizability of these polymers was evaluated by the synthetic accessibility (SA) score.217 The SA score, which ranges from 1 (easy) to 10 (hard), is calculated by considering both fragment contribution and complexity penalties. This score is used to evaluate the synthesizability of molecules or polymer repeating units. The SA scores for polymers within a range of 2.28 to 6.21, with 6.3% of them having
id: 815ce7600061a0c67ab4e57070c67697 - page: 25
0. Out of all 16,896 generated polymers, only 4.5% meet the predefined criteria for both TC and SA, which are known as ideal polymers. FIG. 7. Construction of triblock polymers dataset. (a) Example of the generation of a triblock polymer. (b) SA score versus TC of all 16896 triblock polymers, where stars indicate candidates at the Pareto front. (c) and (d) Distributions of the TC and SA for the whole triblock polymers. The gray strips highlight the statistics of polymers with TC > 0.4 W m-1K-1 or SA < 3.0.213 B. Single-objective optimization trials 25 / 48
id: 54b5f21ff75f01e6d74bfd6bb9535cb6 - page: 25