| dc.description |
The constant technological advances in integrated circuits and electronic systems experienced over the last few years have resulted in large temperature gradients. These can damage electronic devices. Current cooling methods are unable to cope with highly demanding applications such as military systems. Furthermore, for applications in which failure is not an option, a lack of sufficient thermal management can be a limiting factor in the design and addition of functionality.
The aim of this research project is to provide possible solutions to the overheating of
electronics. Following an in depth review of the state-of-the-art in cooling technologies,
we have identified nanofluidics and nanofluids as promising candidates for thermal management. However, systems characterised by such small dimensions are governed by surface phenomena. Sometimes, continuum computational methods such as Computational Fluid Dynamics (CFD) are inadequate in providing a detailed description of such effects. Instead, molecular methods, such as Molecular Dynamics (MD), study systems at a higher resolution and can potentially provide a more accurate understanding of such systems.
This thesis uses MD to understand how the thermodynamic properties of liquids and
nanofluids are modified by spatial restrictions. An important finding is that heat is transferred
differently in confined and unconfined liquids. Following this realisation, an analysis
of the system parameters is carried out to understand how to optimise the heat conductance of such systems. We also consider confined nanofluids. Different materials are modelled and compared with respect to their possible practical use as thermal management agents.
The thermodynamic behaviour discovered has not been described elsewhere and has
potentially high practical importance. Although in its infancy, we believe that it can eventually provide a framework for the design of efficient cooling devices. |
|