| Complex networks are ubiquitous in natural and man-made systems.In recent years,network science has attracted great attention,but few people pay attention to their effects on heat conduction.Studies on heat conduction are so far mainly focused on regular systems such as the one-dimensional(1D)and two-dimensional(2D)lattices where atoms are regularly connected.However,realistic systems such as the nanotube/nanowire networks are not regular but heterogeneously structured,and their heat conduction remains largely unknown.Based on the rapid development of nanonetworks and their importance in engineering and technology,it is necessary to understand how heat conducts in these real nano-networks and the influence of network topology on heat conduction.In this thesis,we construct a physical network model,and then study the influence of network structure on heat conduction.Firstly,our study shows that heat transfer in physical networks is not a direct extension of one-dimensional and two-dimensional lattice systems,but is significantly affected by network structure.On the other hand,we interestingly find that the main heat fluxes of network are always localized in a subnetwork.By introducing a transmission diagram,a skeleton network is found to control the heat transfer in the network.Secondly,we find that when the temperature of the heat source is fixed,the total heat flow of the network will increase with the increase of the source nodes’ degree,but it is not expected to increase monotonously.When the degree of the source nodes is large enough,it will inhibit the heat flow.Moreover,the network density also affects the heat transport on the network.The results show that the sparse network is more conducive to heat conduction.Next,we theoretically prove that the phonon spectrum of nodes in the network relates to the degree of nodes.According to this conclusion,we explain the previous results perfectly from the point of view of phonon scattering.Thirdly,based on the relationship between phonon spectrum and node degree,we realize the thermal rectification effect by changing the degree of the heat source node on the network.The results show that the rectification effect will be enhanced significantly with the increase of source nodes’ degree difference.In addition,we construct an asymmetric network structure from two-dimensional lattice network by means of reconnection,which enormously promotes the rectification effect.Fourthly,we change the topology of complex networks by cross-changing edges,and induce the network to change from negative degree correlation to positive degree correlation.Our research shows that positive correlation network will enhance heat transport,while negative correlation network will weaken heat transport,that is,the heat transport of network is positively correlated with the degree correlation.Meanwhile,we also find that the degree of source nodes can control the temperature distribution on the network.Finally,we study and compare the energy transport of heat and electricity in reconnected physical networks based on two-dimensional lattices.It is found that the network can be transferred from a poor conductor to good conductor depending on the rewired network structure and coupling scheme.Two interesting phenomena are discovered:(a)thermal siphon effect-namely the heat flux goes from low temperature node to higher temperature node;(b)In some specific network structures,lower thermal conductance and higher electrical conductance are found.These discoveries reveal the potential applications of network structured materials in thermal energy management and thermal electric energy conversion.So far,a variety of nano-networks can be fabricated easily in the laboratory,and researchers start to pay more attention to their physical properties.Our research provides new ideas and guidance for understanding the law of heat conduction in complex networks. |
