Show simple item record

dc.contributor.advisorHe, Jianying
dc.contributor.advisorZhang, Zhiliang
dc.contributor.advisorKristiansen, Helge
dc.contributor.advisorLervik, Anders
dc.contributor.authorSeljevold, Sandra Sæther
dc.date.accessioned2024-06-28T06:37:20Z
dc.date.available2024-06-28T06:37:20Z
dc.date.issued2024
dc.identifier.isbn978-82-326-8143-3
dc.identifier.issn2703-8084
dc.identifier.urihttps://hdl.handle.net/11250/3136415
dc.description.abstractThe never-ending need for faster, more complex devices has started a trend of miniaturization that has dominated the electronic industry since the 60s. This trend has led to the characteristic dimensions of components going from the microscale to the nanoscale over the past two decades. This poses significant challenges to the thermal management of these electronic devices. The high heat fluxes and non-uniform temperature distribution arising from this miniaturization leads to the reduction of lifetime and reliability of electronic devices. Overheating is a major challenge for the efficient operation and service life of electronic components, consequently, accumulated heat is the critical bottleneck in the advancement of microelectronics. In this thesis, groundwork for studying thermal transport across a metal-polymer interface has been done. The focus has been on (I) establishing confidence in the modelling method, and (II) understanding individual factors that should be taken into account before attempting to simulate such a challenging system. This work was done by utilizing Molecular Dynamics simulations and can be divided into three main results. Firstly, the choice between different system representations of the polymer was investigated. For the OPLS-AA and OPLS-UA force field chosen, it was discovered that the system representation both did, and did not, matter for thermal transport. When going from an all-atom representation to a united-atom representation, clear structural changes were observed, as a result of lacking hydrogen-hydrogen repulsion in the united-atom system. When these structural differences were taken into account, however, the thermal conductivities were comparable. The SHAKE algorithm was also tested and did not affect thermal transport. This insight enables an informed decision when selecting the force field to represent the polymer system and allows one to confidently choose less computationally expensive methods, as long as the structural awareness is there. Secondly, the focus was put on the interfaces within these materials themselves. Both metal and semicrystalline polymer systems have internal interfaces, between grains of different orientation for metals, and chains with different degree of order for polymers. These were investigated, and their effect quantified. The resistance at the interfaces is naturally very structure dependent, but nevertheless the order of magnitude obtained for the interfaces gives an indication of whether or not these effects need to be considered, depending on the overall conductance needed for the intended use cases. Thirdly, the influence of strain rates on the thermal transport in semicrystalline polymers was investigated. It was found that the thermal conductivity initially drops when a strain is applied to the polymer system. This effect is more pronounced at lower strain rates. As the strain rates used in MD simulations are artificially high, it suggests that the impact of strain rates in flexible electronics could be of crucial significanceen_US
dc.language.isoengen_US
dc.publisherNTNUen_US
dc.relation.ispartofseriesDoctoral theses at NTNU;2024:276
dc.titleAtomistic scale thermal transport in metal and polymeren_US
dc.typeDoctoral thesisen_US
dc.subject.nsiVDP::Technology: 500::Building technology: 530en_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record