A perspective on the data-driven design of polymer nanodielectrics
A perspective on the data-driven design of polymer nanodielectrics
复制标题
DOI：
10.1088/1361-6463/ab8b01
复制
发表时间：
2020-04
期刊：
影响因子：
--
通讯作者：
L. Schadler;L. Brinson;Wei Chen;R. Sundararaman;P. Gupta;Prajakta Prabhune;Akshay Iyer;Yixing Wang;Abhishek Shandilya
中科院分区：
文献类型：
--
作者：
L. Schadler;L. Brinson;Wei Chen;R. Sundararaman;P. Gupta;Prajakta Prabhune;Akshay Iyer;Yixing Wang;Abhishek Shandilya
聚合物纳米电介质是一类新兴的材料，具有有趣的性能组合。它们可以应用于从能量存储到高压电力传输和能量产生的所有领域。这篇文章的重点是绝缘纳米电介质。然而，在所有情况下，控制性能的复杂参数集使得验证模型和开发设计方法变得困难。本文展示了一种最近的方法来开发基于物理模型和实验校准的数据驱动的设计方法。具体而言，它结合了介电常数和损耗函数的有限元建模与载流子跳跃的蒙特卡罗多尺度模拟来预测击穿强度预测。在这两种情况下，填料的分散和界面特性明确考虑到计算目标函数的理想纳米电介质绝缘体。使用高斯过程的元建模和多目标优化的聚苯乙烯-二氧化硅复合材料的这些计算预测，本文确定了帕累托边界相对于负载和分散的纳米填料，最大限度地提高击穿强度和最小化的介电常数和损耗角正切。
Polymer nanodielectrics are an emerging class of materials with intriguing combinations of properties. They have application in everything from energy storage to high voltage electrical transmission, and energy generation. This article focuses on insulating nanodielectrics. In all cases, however, the complex set of parameters controlling the properties have made it difficult to both validate models and develop a design methodology. This paper demonstrates a recent approach to developing a data driven design methodology grounded in physics-based models and experimental calibration. Specifically, it combines finite element modeling of dielectric constant and loss functions with a Monte Carlo multi-scale simulation of carrier hopping to predict break down strength predictions. In both cases, the filler dispersion and interface properties are explicitly taken into to account to compute objective functions for ideal nanodielectric insulators. Using a Gaussian process for meta-modeling and multi-objective optimization of these computational predictions for polystyrene-silica composites, this paper identifies the Pareto frontiers with respect to loading and dispersion of nanofillers for maximizing breakdown strength and minimizing the dielectric constant and loss tangents.