搜索

x
中国物理学会期刊

基于蒙特卡罗-离散纵标方法的氘氚激光等离子体聚变反应率数值模拟

CSTR: 32037.14.aps.68.20190440

Numerical simulation of deuterium-tritium fusion reaction rate in laser plasma based on Monte Carlo-discrete ordinate method

CSTR: 32037.14.aps.68.20190440
PDF
HTML
导出引用
  • 惯性约束聚变(ICF)是实现受控热核聚变可能途径之一. 聚变中子源项是氘氚激光等离子体物理设计与分析的重要参数之一, 其准确性直接影响分析结果的可靠性. 目前国内外对于ICF氘氚聚变反应产生的中子源项研究主要基于解析公式法, 在温度和反应类型等方面适用范围有限. 本文采用粒子云概念对氘、氚粒子云团开展了随机抽样与时空网格划分, 然后基于麦克斯韦速率分布律对氘氚聚变反应开展了多普勒能量展宽效应分析与微分截面温度修正工作, 耦合蒙特卡罗方法和离散纵标方法, 开展了激光等离子体中D-T粒子云团聚变反应率的数值模拟工作. 研究结果显示, 与原核数据库截面相比, D-T, D-D, T-D截面经修正后多普勒温度效应显著. 在20—100 keV的等离子体温度范围内, 相较传统的解析公式法, 本文模拟结果更符合最新的ENDF核数据库的氘氚反应截面数据, 且与解析公式法结果在低能区存在较大误差, 可能是计算方法不同与低温截面差异过大原因导致.

     

    Inertial confinement fusion (ICF) is one of the possible ways to realize controlled thermonuclear fusion. The fusion neutron source term is one of the important parameters in the physical design and analysis of laser plasma. The accuracy of the fusion neutron source term directly affects the reliability of the analysis results. At present, the neutron source term of deuterium-tritium fusion reaction in ICF is mainly based on formula method. It has limited applications in temperature and reaction type. Because of a large quantity of data, it is impossible to simulate the fusion reaction of each particle. In this paper, the concept of particle cloud is introduced, that is, the collection of the like particles with the same position and speed, and it is considered that the action of particle cloud is the same reaction. Because the particles should satisfy the Maxwell velocity distribution at a certain temperature and the direction is all around the circumference angle, the collision cross sections between the incident particle and different target particles are different. Therefore, the design program takes all the possible velocities, polar angles and direction angles, reads the collision cross sections between deuterium and tritium and makes corrections, and obtains the multi-temperature differential correction cross sections of deuterium and tritium fusion with Doppler energy broadening. On these bases, Monte Carlo method and discrete ordinate method method are used. A numerical simulation program for the fusion rate of D-T particles in laser plasma is developed in this paper. It is found that there are significant differences between the DT, DD, TD cross sections and the original cross sections after Doppler broadening. In a range of plasma temperature between 20 keV and 100 keV, the simulation results are more consistent with the cross section data of ENDF/B-VI and ENDF/B-VII databases of deuterium-tritium fusion reaction than those from the analytical formula method. There is a large error between the numerical simulation results and the analytical formula method in the low energy region. It may be caused by the difference of calculation methods and too big difference among the used fusion cross sections at low temperature.

     

    目录

    /

    返回文章
    返回