基于Monte Carlo方法改进的化学反应动力学计算机模拟

在Monte Carlo方法及其改进方法的基础上,进一步改进了化学反应动力学方程计算机模拟算法.改进后的算法延续了原方法编制模拟程序方便等优点,并提高了计算机模拟的稳定性、精确性,克服了由于Monte Carlo方法是一种概率方法而导致模拟结果具有随机性的缺点.结合工业应用中常见的平行、连续、对峙和综合化学反应进行了模拟计算,模拟结果与这些化学反应的解析解相对误差小于0.1%.     

高分子构象模拟研究中的Monte Carlo方法

介绍了Monte Carlo方法及其特点,进而分析了Monte Carlo用于高分子模拟的优势,并描述了两类模拟模型。论文重点综述了近年来Monte Carlo方法在高分子构象模拟中的一些研究与应用,并展望了Monte Carlo方法在高分子构象模拟中的发展趋势和前景。     

加速Monte Carlo分子模拟收敛的能量选择取样法

为提高Monte Carlo分子模拟的取样效率,加速模拟过程的收敛,本文提出了根据粒子能量进行取样的方法。通过对凝固点附近的L-J流体的模拟,将能量选择取样法与Metropolis取样法以及力偏倚取样法的收敛性和模拟结果作了比较,表明能量选择取样法的收敛速度显著高于Metropolis取样法,与力偏倚取样法相当,而计算量比力偏倚取样法小。     

催化裂解过程分子尺度反应动力学模型研究

以工业数据为基础,运用结构导向集总方法来构造烃分子,对催化裂解原料油进行了分子尺度上的Monte Carlo模拟.在深入研究催化裂解反应机理的基础上,以原料油模拟产生的分子矩阵作为反应物分子,将结构导向集总方法与Monte Carlo方法相结合,建立了催化裂解过程分子尺度的反应动力学模型.结果表明:Monte Carlo方法可以在分子尺度上实现对催化裂解原料较好的模拟,产品产率和汽油组成与实际值能较好地拟合.且随着虚拟分子数的增加,对原料油性质和反应结果的模拟计算精度提高;模型具有较好的适应性和外推性.     

气溶胶在毛细管中输运Monte Carlo模拟

气溶胶在管道中的输运广泛应用于气溶胶采样、监测和气溶胶的危害评价中,前人基于对流扩散方程、流体速度分布规律,结合初始条件和边界条件,对气溶胶粒子在管道中的扩散特性进行了理论分析,取得了一定成功．Monte Carlo方法作为一种统计实验方法,在模拟中子或光子在介质中的Markoff输运过程中取得了巨大成功．另外,Monte Carlo方法也成功用于吸附质在多孔介质中的吸附和解吸特征研究．气溶胶粒子在气体介质中,由于受到气体分子的随机碰撞,     

Ozawa方程处理PET非等温结晶动力学过程研究

采用Monte Carlo法计算机模拟在预先成核条件和不同降温速率下PET非等温结晶曲线,将模拟的结果用Ozawa方程及改进的Ozawa方程进行处理。结果表明,用Ozawa方程解析的Avrami指数与理论值相近,用改进的Ozawa方程可直接得到结晶速率常数,求出的线生长速率与理论值基本吻合。     

NPT系综Monte Carlo分子模拟方法的改进

通过调节体系密度使得体系的瞬时压力保持恒定,从而实现 NPT 系综的 Monte Carlo 分子模拟。对 Lennard-Jones 流体的模拟结果表明,该方法与现行的 NPT 系综模拟方法相比具有更快的收敛速度。本文应用改进的方法对 Ar-Kr 混合物的热力学性质进行了模拟。     

丁二酰亚胺无灰分散剂作用机理的分子模拟研究

通过对30种丁二酰亚胺类无灰分散剂在水和石墨表面作用的Monte Carlo Docking模拟,得到其在水和石墨表面的低能量吸附构象及结合能.QSAR分析和Monte Carlo Docking模拟的结果表明分散剂在水表面的吸附行为,与分散剂分子的极性基团的数量、分子正负电荷的分布、分子在水中的溶解自由能有关;而分散剂在石墨表面的吸附,与分散剂分子的链长度和支链数目有关.在此基础上合成了9个丁二酰亚胺无灰分散剂,对其分散烟炱(碳黑)的性能进行了模拟评定,通过对这些分散剂作用的计算机模拟,验证了模拟方法的可行性、可操作性.     

MATLAB和Monte Carlo法在连续反应动力学中的应用

基于MATLAB语设计编写了Monte Carlo法处理连续化学反应的动力学计算及其计算机模拟的应用程序,将所编程序用于模拟3-丁烯腈在盐酸催化剂溶液中的水解反应,模拟结果与按照常规化学反应动力学公式的计算结果相比较,表明Monte Carlo法模拟对于预测反应动力学过程具有较高准确性。     

Monte Carlo法模拟球形-线性Lennard-Jones流体混合物

本文用Monte Carlo方法对球形-线性Lennard-Jones流体混合物进行了计算机模拟。推导出含有该类线性分子的体系能量、压力和化学位的长程校正公式,并用于NPT系综模拟,研究了组成对体系物性的影响,为进一步模拟链状分子及聚合物体系打下基础。     

A two-dimensional Monte Carlo model for pore densification in a bi-crystal via grain boundary diffusion: Effect of diffusion rate,initial pore distance,temperature, boundary energy and number of pores

A two-dimensional Monte Carlo (MC) model is introduced for simulating the evolution of the pore on a bi-crystal grain boundary via grain boundary diffusion. Simulated pore shrinkage kinetics is found to be consistent with previously reported results over variable grain boundary diffusion rates and initial pore distances while the essential characteristics of the microstructural evolution are simultaneously realized. The influence on the pore densification kinetics of grain boundary motion, boundary energy ratio, simulation temperature and pore interactions in an array is found such that pore shrinkage rate increases as the grain boundary motion, the simulation temperature and the grain boundary energy increase. The interactions of the pores are found to hinder the pore densification. The body of results signify that the more elongated the pore shape and the shaper the pore tip region are favored for the faster pore shrinkage kinetics during the simulated densification process via grain boundary diffusion.     

Computational Design for Grain-Oriented Microstructure of Functional Ceramics Prepared by Templated Grain Growth

We have developed a computational model using Monte Carlo simulation in order to design grain-oriented microstructures of ceramics processed by templated grain growth for functional materials with anisotropic grain growth due to anisotropy in interfacial energies. The current model with plate-like template particles expressed the development of grain-oriented microstructures with the low interfacial energy planes well faceted. In addition, the proposed model gave a guiding principle for optimizing the initial parameters (i.e., ratio of templates to total solid particles, initial size and aspect ratio of template particles, and amount of liquid phase), leading to highly grain-oriented ceramics with improved performances.     

Monte Carlo simulation of microstructure evolution in biphasic-systems

Over the past few decades, a variety of models have been proposed in order to investigate the grain growth kinetics and the development of crystallographic textures in polycrystalline materials. In particular, a full understanding of the microstructure evolution is a key issue for ceramic systems, since their mechanical or thermal behaviour is intimately related to their microstructure. Moreover, the development of appropriate simulative tools is crucial to reproduce, control and finally optimize the solid-state sintering process of ceramics. Monte Carlo simulations are particularly attractive because of their ability to reproduce the statistical behaviour of atoms and grain boundaries with time. However, Monte Carlo simulations applied to two-phase materials, such as many ceramic systems, result complex because both grain growth and diffusion processes should be taken into account. Here the Monte Carlo Potts model, which is widely used to investigate the crystallization kinetics for monophasic systems, is modified and extended to biphasic ones. The proposed model maps the microstructure onto a discrete lattice. Each lattice element contains a number representing its phase and its crystallographic orientation. The grain formation and growth are simulated by appropriate switching and reorientation attempts involving the lattice elements. The effect of temperature is also discussed.     

Monte Carlo Simulation of Kinetics of Ammonia Oxidative Decomposition over the Commercial Propylene Ammoxidation Catalyst (Mo-Bi)

Monte Carlo method is applied to investigate the kinetics of ammonia oxidative decomposition over the commercial propylene ammoxidation catalyst(Mo-Bi). The simulation is quite in agreement with experimental results. Monte Carlo simulation proves that the process of ammonia oxidation decomposition is a two-step reaction.     

Direct simulation Monte Carlo for simultaneous nucleation, coagulation, and surface growth in dispersed systems

A Monte Carlo simulation technique is developed to describe dispersed phase systems. The method is formulated for simultaneous coagulation, nucleation and surface growth, but can be extended to include other processes. These processes are considered in an initially constant simulation volume. The changes in the particle ensemble are determined by a random choice procedure, while the particle number in the simulation volume changes according to the chosen events. Every time, when the particle number in the simulation volume increases or decreases by a factor of two of its initial value, the simulation volume is halved or doubled, respectively. Therefore, this method is called a stepwise constant-volume Monte Carlo simulation. It allows to use only several thousands simulation particles, even if the particle number concentration experiences changes of several orders of magnitude. The simulations are validated through a comparison with the exact mathematical solutions for several simple cases. An example of simultaneous nucleation and coagulation in the free-molecular region demonstrates, that the stepwise constant-volume Monte Carlo simulations lead to more accurate results than the constant-number Monte Carlo simulations.     

基于蒙特卡洛模拟法的钻井经济评价

对石油钻井进行投资决策时,要考虑各方面的风险因素和投资成本,为了降低石油钻井中的投资风险,在钻井之前就能预测到可能的成本,可以采用蒙特卡洛模拟法对钻井投资进行风险预估。本文通过对不同的地层分别运用蒙特卡洛模拟法,在不同的地层建立蒙特卡洛模的模型,并应用matlab模拟软件处理数据解释分析结果,最终得出钻井费用的预测值和误差,为投资决策提供依据。     

The use of Monte Carlo simulation to evaluate the optical properties of polyester fabric treated with titanium dioxide nanopigments

The current study utilises Monte Carlo simulation and Mie scattering theory to estimate the reflectance spectra of fabric coated with titanium dioxide nanopigments of various diameters and concentrations. Image processing was carried out and experimental data were gathered to evaluate the performance of Monte Carlo simulation. The distribution and location of the nanopigments on the surface of fabric were determined using the Monte Carlo method. Reflection of the fabric was calculated based on Monte Carlo simulation with the partitive mixing method and Mie theory. According to the experimental and simulation results, the reflectance of coated samples was increased by increasing the concentration and number of titanium dioxide nanoparticles. There was a good match between the results obtained by Monte Carlo simulation and the experimental results. For coated samples (dTiO₂: 500 nm), the root mean square error between measured and predicted reflectance by the Monte Carlo and partitive mixing method and by Monte Carlo and Mie theory was 0.022 and 0.0078, respectively. The results indicate that the performance of the Monte Carlo and Mie method was better than that of the Monte Carlo and partitive mixing method. According to t-test analysis, there was no statistically significant difference between the experimental data and Monte Carlo simulation.     

Simulation of Grain Growth and Pore Migration in a Thermal Gradient

The Potts Monte Carlo simulation was used to simulate microstructural evolution in uranium dioxide fuel rods. During service, grain growth, pore migration, and thermal segregation of the pores and UO₂ occur in the rods in a thermal gradient. In this investigation, we developed a model which simulates simultaneous grain growth, pore migration, and thermal segregation of the pores and UO₂ in a temperature gradient. Grain growth in a thermal gradient was simulated using the Monte Carlo Potts model technique developed by Anderson, Srolovitz, and co-workers. Pore migration was simulated using conserved dynamics with minimum-energy exchanges at a finite temperature. A temperature gradient was introduced into the model via interfacial mobility gradient. Finally, thermal segregation of the pores and UO₂ was achieved by introducing a heat of migration term, Δ E _t, which biased the motion of porosity to the high-temperature region. The development of this model is described and the incorporation of the proper physics of pore migration and thermal segregation is discussed.     

Effect of surface pressurization on the growth of α-Fe₂O₃ nanostructures

By suitably pressurizing iron substrates under different conditions, the resulting α-Fe(2)O(3) nanostructures, formed by its direct thermal oxidation, can gradually change in succession from nanowires to nanoleaves and to micropillars as the pressure is increased. The inter-relation between the pressure conditions and the resulting nanostructure is studied by density functional calculations using ultrasoft pseudopotentials with a plane-wave basis method and with the generalized gradient approximation (GGA). It is shown that the shape of the formed nanostructures is primarily determined by the anisotropic activation energy and, as the latter is lowered, there is a shape change from wire to pillar. A simulation model of diffusion using the Monte Carlo method is applied in the 3-D (dimensional) case to show how the anisotropic activation energy influences the growth process of the α-Fe(2)O(3) nanostructure. The present study provides a way to control the shape of the nanostructures grown by the thermal-oxidation method.     

Original Monte Carlo Methodology Devoted to the Study of Sintering Processes

A Monte Carlo methodology different from others based on the Potts model has been developed to solve capillary-driven mass transport involved in the sintering process. The addition of a cohesive energy term to the energetic model enables quantification of the stress gradients within grains, as well as the induced mass fluxes. The study of two-particle sintering involving volume diffusion has been chosen as a first example. The morphological evolution of the two-particle system to a single particle is followed. Real-time-dependent neck growth and shrinkage rates are defined. The rheological behavior of the particles induced by the Monte Carlo procedure is discussed.