24小时热门版块排行榜

返回列表

ansysabaqus

新虫 (正式写手)

应助: 1 (幼儿园)
金币: 156.8
散金: 75
红花: 2
帖子: 449
在线: 176.4小时
虫号: 3021830
注册: 2014-03-06
性别: GG
专业: 金属结构材料

[交流] 研究扩散，2000个基体原子和125个扩散原子合适吗？nvt弛豫50ps，npt100ps 已有3人参与

算出来的结果可靠吗？会不会原子选的少了，或者时间设少了

回复此楼

» 猜你喜欢

1楼 2014-04-29 15:45:36

已阅回复此楼关注TA 给TA发消息送TA红花 TA的回帖

月只蓝

主管区长 (职业作家)

专家经验: +1059
模拟EPI: 1
应助: 1712 (讲师)
贵宾: 8.888
金币: 68121.7
散金: 1938
红花: 443
沙发: 4
帖子: 4373
在线: 3291.4小时
虫号: 1122189
注册: 2010-10-14
专业: 宇宙学
管辖: 计算模拟区

★
小木虫: 金币+0.5, 给个红包，谢谢回帖

不是简单地说合适不合适。
1.模型原子数的多少，与模拟盒子的大小有直接关系，如果你使用了cutoff的方法，cutoff的半径对盒子的大小有要求。
2.具体要弛豫多久，体系越大弛豫的时间越长。
参见如下：
It is rarely the case that the starting conditions of a simulation (the particle coordinates and velocities) correspond to a configuration that is consistent with the equilibrium distribution of the system. Often, in the construction of the system, an optimization of the geometry with respect to a forcefield has been performed in order to remove unphysical interactions, but this has the consequence of placing the system in a configuration that is a potential minimum.

If particle velocities are assigned at random from a Maxwell-Boltzmann distribution and an integration is then performed in the microcanonical ensemble (i.e. internal energy is conserved), a decline in temperature is observed as some of the kinetic energy is transferred to potential energy. It is thus important that the simulation is run for a period before any analysis is conducted, in order to ensure that it is properly at equilibrium. In many cases the equilibrium can be facilitated by the application of a thermostat.

There are a number of methods for determining when a system is at equilibrium: approximate methods include plotting the energy or temperature against time, at equilibrium these quantities will fluctuate around constant averages. More sophisticated methods include calculating order parameters (see Allen and Tildesley, 1987).