固态金属中声子热传递的分子动力学模拟研究

固态金属中的热传递是声子和自由电子共同作用的结果。自由电子引起的热导率可以通过电导率,利用Wiedemann-Franz定律得到,声子引起的热导率目前仍然不能进行实验测量,只能借助其他方法来研究。本文采用非平衡分子动力学(NEMD)方法,用镶嵌原子方法(EAM)势能模型,模拟计算了不同厚度(1.760-10.56nm)金属镍薄膜中由于声子-声子作用引起的热导率。然后根据纳米厚度金属薄膜的热导率借助关联式推到宏观尺度下由于声子-声子作用引起的热导率。结果表明,对于纳米厚度金属薄膜,由于声子-声子作用引起的热导率比块体金属镍的热导率小一个数量级;薄膜厚度越小,声子-声子作用引起的热导率越小;对于块体金属镍,由于声子-声子作用引起的热导率约占其总热导率的33.0%左右。     

液态Co物性的平衡及非平衡分子动力学模拟

本文采用EAM作用势,通过平衡分子动力学(EMD)模拟的方法计算了Co熔体的自扩散系数、剪切粘度等物理性质.同时采用非平衡分子动力学(NEMD)方法计算了Co的剪切粘度.研究表明有关传输性质的计算是可与实验比较的,能够反映出液态Co典型的动力学特性.     

流体粘度的计算机模拟

对流体粘度的计算机模拟作了简单回顾;并利用非平衡分子动力学方法,将在三相点附近的氩看成具有Lennard-Jones势能函数的球形分子,用自编程序对其进行了计算机模拟。     

氩晶体薄膜法向热导率的分子动力学模拟

结合卫星“微型核”的特点，研究电介质薄膜中的导热机理以及薄膜厚度对导热系数的影响.以结构较为简单、具有可靠势能函数，实验数据较为丰富和可靠的氩的(fcc)晶体为模型,采用平衡分子动力学方法(EMD)和各向异性非平衡分子动力学方法(NEMD)计算了氩晶体及其法向薄膜的热导率,并与实验结果进行比较.模拟结果表明，氩晶体纳米薄膜的热导率显著小于对应大体积晶体的实验值，具有明显的尺寸效应.在氩薄膜厚度为2.124—5.310nm的模拟范围内,薄膜的法向热导率随着薄膜厚度的增加而呈近似线性增加. 关键词： 热导率 纳米薄膜 尺寸效应 平衡分子动力学 非平衡分子动力学     

Nonequilibrium molecular dynamics simulation for studying the effect of pressure difference and periodic boundary conditions on water transport through a CNT membrane

A NEMD simulation system is constructed to simulate at two-dimensional (2D) periodic boundary conditions (PBCs) and to create two different pressures on two sides of the carbon nanotube (CNT) membrane. The simulation results show that water permeation through the same CNT membrane driven by different pressure differences exhibit similar transport phenomenon including unusually fast water permeation and a periodic (non-parabolic) radial velocity distribution unlike the parabolic form characteristic of continuum flow in the CNT membrane. A three-dimensional (3D) PBC system is also constructed to simulate water permeation through the same CNT membrane at the same pressure differences, to show the effect of PBC and simulation methodologies on transport phenomenon. The two systems both show that the forward/backward water flux increases/decreases with increasing the pressure difference from 1.0 MPa to 8.0 MPa. However, the net flux is higher for the 3D PBC system, especially at higher pressure difference is high. In general, the NEMD simulation method using the 2D PBC system is shown to be a feasible and valuable tool for studying pressure-driven permeation processes such as nanofiltration through these studies with model CNT membrane.     

Performance investigation of nanoscale thermal cloak by the perforated silicon film

In recent years, nanoscale thermal cloak, as a representative of nanoscale heat flux regulation devices, has attracted a lot of attention from researchers. However, the existing design methods are relatively complicated and all adopt constant temperature boundaries, the temperature changes constantly in the real environment, which greatly hinders its engineering applications. In this paper, inspired by phonon localization theory, we construct a nanoscale thermal cloak by a perforated silicon membrane and evaluate its cloaking performance and dynamic response. Results show that when the perforated area is fixed, the more the number of holes, the better the cloaking performance. In addition, the nanoscale thermal cloak still exhibits good cloaking performance in the dynamic environment. Finally, the cloaking mechanism is analyzed by calculating the phonon density of states (PDOS) and mode participation rate (MPR), and the reduction of thermal conductivity in the functional region is attributed to phonon localization.     

Model hysteresis dimer molecule II: deductions from probability profile due to system coordinates

The hysteresis dimer reaction of the first sequel is applied to test the Gibbs density-in-phase hypothesis for a canonical distribution at equilibrium. The probability distribution of variously defined internal and external variables is probed using the algorithms described, in particular the novel probing of the energy states of a labeled particle where it is found that there is compliance with the Gibbs’ hypothesis for the stated equilibrium condition and where the probability data strongly suggests that an extended equipartition principle may be formulated for some specific molecular coordinates, whose equipartition temperature does not equal the mean system temperature and a conjecture concerning which coordinates may be suitable is provided. Evidence of violations to the mesoscopic nonequilibrium thermodynamics (MNET) assumptions used without clear qualifications for a canonical distribution for internal variables are described, and possible reasons outlined, where it is found that the free dimer and atom particle kinetic energy distributions agree fully with Maxwell–Boltzmann statistics but the distribution for the relative kinetic energy of bonded atoms does not. The principle of local equilibrium (PLE) commonly used in nonequilibrium theories to model irreversible systems is investigated through NEMD simulation at extreme conditions of bond formation and breakup at the reservoir ends in the presence of a temperature gradient, where for this study a simple and novel difference equation algorithm to test the divergence theorem for mass conservation is utilized, where mass is found to be conserved from the algorithm in the presence of flux currents, in contradiction to at least one aspect of PLE in the linear domain. It is concluded therefore that this principle can be a good approximation at best, corroborating previous purely theoretical results derived from the generalized Clausius Inequality, which proved that the PLE cannot be an exact principle for nonequilibrium systems.