当前位置: X-MOL 学术ACS Appl. Polym. Mater. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Stabilizer Formulation Based on High-Throughput Chemiluminescence Imaging and Machine Learning
ACS Applied Polymer Materials ( IF 4.4 ) Pub Date : 2020-06-29 , DOI: 10.1021/acsapm.0c00442
Toshiaki Taniike 1 , Taishi Kitamura 1 , Koyuru Nakayama 1 , Ken Takimoto 1 , Naoki Aratani 1 , Toru Wada 1 , Ashutosh Thakur 1 , Patchanee Chammingkwan 1
Affiliation  

The combination of synergistic stabilizers is a basic strategy for prolonging the lifetime of polymeric materials, but exploration of combinations has been minimally accomplished due to certain problems. Here, we report a highly efficient exploration of stabilizer formulations based on high-throughput chemiluminescence imaging (HTP-CLI) and machine learning. Different formulations were generated by selecting 10 kinds of stabilizers from a library, and their performance in stabilizing polypropylene (PP) was evaluated based on HTP-CLI measurements. Formulations were evolved through a genetic algorithm to elongate the lifetime of PP. A demonstrative implementation up to the fifth generation successfully identified performant formulations, in which mutually synergistic combinations of stabilizers played a pivotal role.

中文翻译:

基于高通量化学发光成像和机器学习的稳定剂配方

协同稳定剂的组合是延长聚合物材料寿命的基本策略,但是由于某些问题,对组合的探索已很少完成。在这里,我们报告了基于高通量化学发光成像(HTP-CLI)和机器学习的稳定剂配方的高效探索。通过从库中选择10种稳定剂来生成不同的配方,并根据HTP-CLI测量评估其稳定聚丙烯(PP)的性能。通过遗传算法改进了配方,以延长PP的使用寿命。直到第五代的示范实施成功地确定了性能配方,其中稳定剂的相互协同组合起着关键作用。
更新日期:2020-08-14
down
wechat
bug