"Thermodust: Computational modeling of thermal properties of 2D-metal composite materials"
Organization:
Funded by: |
|
|  |
|  |
|   |
Collaboration: | Trinity College Dublin: Jožef Stefan Institute: Polytechnic University of Milan: University of Barcelona: |
Thermodust project, funded within the Pathfinder program of the EU (project number 101046835)Description:
Background
This research project is part of Thermodust, an EU-funded Pathfinder program focusing on novel materials for thermal management problems, e.g., electronics cooling. Advances in manufacturing technologies have opened new opportunities for designing multi-materials with unique structures and properties. This enables the exploration of countless configurations of printed composites explicitly tailored to their applications. We focus on creating modeling techniques to predict the material properties of novel 2D-metal matrix composites (2DMMC). Particular focus is on the effects on the thermal conductivity of printed parts derived from adding nano-scale structures such as graphene.
Targets
Our research within Thermodust focuses on enhancing current computational modeling methods. In particular, we focus on preserving the underlying geometric structure of the governing equations. By conserving geometric features and quantities such as symmetry, symplecticity, and energy, we expect to obtain improved predictions, clarifying heat transport near interfaces. We focus on two areas:
- Molecular dynamics. Accurately predicting mechanical and thermal characteristics of 2D-metal matrix materials remains a challenge with current methods. The fundamental phenomena impacting mechanical properties and heat transfer stem from nano-scale structures and interactions. Detailed modeling based on molecular dynamics methods offers a comprehensive approach - we address challenges regarding the reliability of MD models for 2DMMCs.
- Upscaling. The quantitative and qualitative correspondence of nano-scale outcomes to properties at micro or milli-scales poses a second major challenge in this research. Homogenization and reduced order modeling for upscaling the molecular-scale predictions is a second driving element in this research.
Pictures:
