共聚物
动力学蒙特卡罗方法
蒙特卡罗方法
材料科学
热力学
动能
聚合物
反应速率
微观结构
统计物理学
粒子(生态学)
灵敏度(控制系统)
工作(物理)
降冰片烯
产品(数学)
蒙特卡罗算法
反应机理
链条(单位)
光滑粒子流体力学
反应速率常数
粒子群优化
三合会(社会学)
平滑度
序列(生物学)
生物系统
动态蒙特卡罗方法
化学
粒径
最终产品
蒙特卡罗分子模拟
计算机科学
作者
Guoyan Wang,Tianlei Wang,Guilin Liu,Zhiping Wang,Xin Fang Feng
标识
DOI:10.1021/acs.iecr.6c03895
摘要
Abstract A reaction kinetic model involving a mechanism controlled by chain initiation and catalyst deactivation as well as an accelerated kinetic Monte Carlo (kMC) method predicting copolymer microblocks were developed for the metallocene-catalyzed ethylene–norbornene copolymerization. The outcomes demonstrated excellent agreement with the published experimental data and revealed that the vast majority of polymer chains are formed by initiated ethylene. By combining code optimization with the Numba module, our kMC simulation calculated the intricate copolymer microstructure without conventional quasi-steady-state assumption while optimizing system size to balance computational efficiency against the smoothness of concentration profiles. Found by a systematic sensitivity analysis for all rate constants, the key reactions and their rate constants were finely tuned using a particle swarm optimization (PSO) framework. The integrated kMC-PSO approach effectively captured macroscopic experimental metrics including norbornene conversion and molecular weights. Furthermore, while the model accurately reflects the overall composition, discrepancies in norbornene-containing triad distributions demonstrate the inherent physical boundaries of classical penultimate models in capturing complex stereochemical sequence effects under high feed ratios, providing critical mechanistic insights for advanced copolymerization modeling.
科研通智能强力驱动
Strongly Powered by AbleSci AI