Data centric modeling and design of polymer nanocomposites for mechanical and dielectric properties
Abstract
Building reliable surrogate models based on accurate and comprehensive process-structure-property relationships is essential for advancing a self-driven, accelerated materials design and manufacturing framework. Polymer nanocomposites have immense potential as they exhibit many desirable structural and functional properties that can be honed for variety of applications. However, owing to their heterogeneous and disordered nature, they present a unique set of challenges for the successful development of such an autonomous framework. Our research, in principle, helps in two ways - by developing more realistic and accurate computational models and then using these models to establish structure-property and process-property relationships. The first part is achieved by incorporating experimental measurements of local nanoscale, molecular behaviour into macroscale material models in mechanical and dielectric property space. The second part is accomplished by combining these material models with computational design of experiments (DOE).
While building material models and designing experiments, a central emphasis is placed on the crucial role of microstructure and interfacial properties in tailoring material behavior and generating guidelines for optimal designs with the help of machine learning and optimization techniques. Within microstructure space, we especially focus on particle dispersion states as experimental data on quantifying its effects on material properties is scarce. Such experimental characterization is limited by many hurdles, like process control, narrow windows of characterization as well as exploration within realistic time frames and plausible efforts. One area of focus involves the strategic manipulation of composition, morphologies, and interface trap densities in polymer nanodielectrics for capacitive energy storage, a challenging material design problem where high permittivity and breakdown strength are needed to achieve high energy density and low loss. This aims to simultaneously enhance permittivity and breakdown strength while minimizing energy loss. A continuum scale model for permittivity and loss calculations was combined with first principle-based calculations for interface trap densities. A key finding was that particle aspect ratio and extrinsic functionalization can be leveraged to improve stored energy density and intrinsic interface design variable can be used to tune loss characteristic, independently with adversely affecting the former. Dielectric loss component is most challenging to minimize due to coupled mechanistic nature of electric polarization. However, designing intrinsic interface may hold a key to overcome this challenge.
Complementing this functional property design work, another investigation delves into process-structure-property (PSP) relationships for mechanical properties of nanocomposites. It can provide guidelines for process and constituents’ design for load bearing, structural applications. This work includes developing an interfacial material model, informed by atomic force microscopy and fluorescence microscopy, to represent the local decay of elastoplastic (Young’s modulus and yield strength) properties in the vicinity of particle interface. It also sheds light on the impact of twin-screw extrusion driven dispersion processes, erosion and rupture, on nanocomposite mechanical yielding process by developing a spatial, linear property map called yield resistance map. This map acts as a blueprint for yield progression within the a material. Yield resistance map indicates the low and high resistance domains of the material based on the microstructures and interfacial parameters. A future study for developing a predictive model for nanocomposite yield strength using resistance maps is also proposed.
Furthermore, viscoelastic behaviour, a fundamental property of all polymeric materials is examined through the simulation of a high-dimensional design space. For many polymeric materials, viscoelastic property characterization can be used as a measure of glass transition phenomena and hence molecular scale mobility of polymer bulk and interfacial regions. This approach assesses the effects of particle dispersion, agglomerations, and interfacial characteristics on material mobility. It is accomplished by comparing the effectiveness dispersion processes, mentioned above, rupture and erosion of agglomerates in terms of viscoelastic properties. It was observed that erosion facilitates larger interphase percolation and gives rise to wider mobility spectra.
In summary, by connecting processing techniques to the resulting properties, and large datasets generated through simulation, this integrated approach seeks to provide guidelines for the rational design of high-performance polymer nanocomposites as well as lay the groundwork for future research into accelerated design and autonomous fabrication platforms for next generation mutli-scale, multi-functional, disordered, heterogeneous materials.
Type
Department
Description
Provenance
Subjects
Citation
Permalink
Citation
Prabhune, Prajakta V (2025). Data centric modeling and design of polymer nanocomposites for mechanical and dielectric properties. Dissertation, Duke University. Retrieved from https://hdl.handle.net/10161/32689.
Collections
Except where otherwise noted, student scholarship that was shared on DukeSpace after 2009 is made available to the public under a Creative Commons Attribution / Non-commercial / No derivatives (CC-BY-NC-ND) license. All rights in student work shared on DukeSpace before 2009 remain with the author and/or their designee, whose permission may be required for reuse.