节点文献

流体输运特性和物态转变的分子动力学研究

A Study on Transport Properties and Phase Transition of Fluids by Molecular Dynamical Simulation Method

【作者】 刘娟芳

【导师】 曾丹苓;

【作者基本信息】 重庆大学 , 工程热物理, 2005, 博士

【摘要】 本论文旨在运用近年发展起来的分子动力学模拟方法系统地研究工程上具有重要意义的流体的输运系数,包括简单流体、复杂流体和混合物,特别是纳米流体的输运系数,探讨模拟方法,提高模拟精度。同时,根据我们所承担的中俄合作研究项目,对玻璃的物态转变、输运特性及微观结构进行了分子动力学模拟,取得了一些有意义的结果。众所周知,描述物质宏观输运特性的输运系数是物质重要的物性参数,也是工程应用上最基本、最重要的特性,但由于其在测量上的困难,很难用实验的方法精确测量和确定。无论是简单流体还是复杂流体,特别是新型工质,这方面的数据均很匮乏。因此,流体输运系数的确定成为研究者们十分关注的问题。近年来,已被广泛应用的分子动力学模拟方法为获得不同状态下流体的输运系数提供了一种可行的、有效的手段和一条新的途径。本文首先以简单流体作为研究对象,利用分形理论和分子动力学模拟方法,计算了实际流体Ar在不同温度下的分维数,提出了实际流体分子的随机运动可用分数Brown函数来描写的假说,并构造了相应的位移、速度、速率的分数布朗函数,对所提出的假设进行了论证。其次,以简单流体、复杂流体和混合物作为研究对象,主要利用平衡时间相关函数理论对其扩散、粘性及导热系数进行了MD模拟。在论文中,主要从各输运系数对应流的自相关系数随时间变化的曲线形式上分析了不同热力学状态下的输运特性,并研究了它们随温度和密度的变化,对用MD方法研究输运特性的可行性进行了论证。利用MD方法研究水的输运特性很少见诸报导,这可能是因为其分子结构和运动的复杂性所致。在论文中,对于复杂流体选用极性水分子作为研究对象。将简单流体的Green-Kubo公式进行了发展,导出了水的相应计算式,并对其扩散系数和导热系数进行了计算。从微观尺度上分析了质量、动量和能量的传递过程和流体的输运特性。近年来,纳米技术受到广泛关注。纳米流体作为一种新型工质,其物理特性不同于纯流体,但纳米流体的物性研究却十分缺乏。本论文在简单流体和复杂流体输运特性研究的基础上,计算了纳米流体的相关热物性参数。论文中先选取LJ势函数对液体氩和金属钠粒子所组成的纳米流体的输运特性进行了探索性研究,主要从不同时间平均数、不同截断半径和不同纳米粒子含量百分数的时间相关函数进行了研究,然后选取复杂势函数对水和金属铂粒子组成的纳米流体的导热系数进行了计算,这些探索性的研究为今后不同纳米流体输运系数的研究提供了新

【Abstract】 The main purpose of the present dissertation is to perform systematically Molecular Dynamics (MD) simulation of the transport properties for fluids, which include simple fluids, complex fluids and mixture, in particular, nano-fluids. The MD technique has been widely used in various fields in the past years. In the study, the detailed simulation techniques are discussed in order to improve the precision of the calculation. In addition, according to the project in cooperation with Russia, the phase transition, transport properties and the micro-structure of the glasses are investigated by using the molecular dynamics simulation method. Some significant results were obtained.The knowledge of the transport coefficients for describing the macroscope transport properties is of considerable technological and academic importance, and it is also the most principal physical parameters for substances. However, the transport coefficients are very difficult to be precisely determined by experiments, so that the corresponding physical data are deficient. To search for a direct and accurate simulation method for determining the transport coefficients is an area of active research. The MD technique provides a promising and effective way to study the transport properties of fluids in different thermodynamical states.In the dissertation, firstly, molecular dynamics simulations based on the fractal theory are carried out to calculate the fractional dimensionality for the real fluid of Argon at the different temperatures. A hypothesis was proposed which suggested that the random motion of the molecules for real fluids can be described by the Fractional Brownian function. At the same time, the corresponding FB functions of displacement, velocity and speed are constructed. The results obtained by the functions were compared with the results obtained directly by molecular dynamics simulation. They agree pretty well, thus the reliability of the hypothesis is verified.Secondly, liquid Ar, water, and mixtures are chosen as the working substances to calculate their diffusion coefficients, shearing viscosities and thermal conductivities by MD methods. We analyzed mainly the shape of the curves of Auto-correlation function to reflect the transport properties in different thermodynamical states. For liquid water, different potential models were applied and compared. The results show that the potential model has strong influence on the transport coefficients. In the study, the calculation formula of thermal conductivity was deduced from that of the simple fluid.

  • 【网络出版投稿人】 重庆大学
  • 【网络出版年期】2007年 01期
  • 【分类号】O414.2
  • 【被引频次】18
  • 【下载频次】1397
节点文献中: 

本文链接的文献网络图示:

本文的引文网络