24小时热门版块排行榜    

查看: 1408  |  回复: 4

stractor

金虫 (著名写手)

[交流] 【求助】VASP正则系综模拟问题

我从来没做过第一性原理分子动力学模拟。最近想在这方面尝试一下。我模拟的是NVT系综。现在对SMASS这个关键字的设置不是很清楚。
手册上说了,对于SMASS>=0,就是选择正则系综,而且模拟采用Nose质量控制温度波动。SMASS的值就对应Nose质量。

手册上这样描述的:The Nose massshould be set so that the induced temperature fluctuation show approximately the same frequencies as the typical phono frequencies …… The frequency of the approximate temperature fluctuations induced by the Nose thermostat is written to the OUTCAR file.
我的问题是,这个质量应该设置多少?
是不是需要尝试以些SMASS的值,再在OUTCAR中找温度波动的频率?
OUTCAR中这个波动频率在什么位置?

[ Last edited by stractor on 2010-9-2 at 14:33 ]
回复此楼
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

cenwanglai

荣誉版主 (知名作家)

老和山猥琐派九段

优秀版主

aylayl08:不守版规? 2010-09-08 08:37:59


关注vasp的分子动力学模拟。沙发学习。

[ Last edited by cenwanglai on 2010-9-8 at 08:56 ]
2楼2010-09-02 14:29:53
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

freer007

木虫 (正式写手)

同样想知道。
3楼2010-09-02 15:27:49
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

gavinliu7390

木虫 (著名写手)

叶落鹰飞


zzy870720z(金币+1):谢谢指教 2010-09-03 08:25:15
stractor(金币+2):谢谢交流。 2010-09-03 09:48:04
一般来说不用,  影响实际不大。
真理是一点点接近的!
4楼2010-09-03 08:11:42
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

stractor

金虫 (著名写手)

★ ★ ★
aylayl08(金币+3):谢谢分享 2010-09-08 08:38:29
我发邮件问了我所看的一篇做BOMD模拟的老外,很快他就回答了我的问题,
下面是他的回答:
The nose mass is related to the average vibrational frequency in the system (at least the way that VASP implements it). The reason for this is you are trying to make the 'frequency' of the thermostat match that of the system you are simulating. If the frequency is too large, you will get unphysical excitations, if it is too low your system will take a very long time to equilibrate.


Unfortunately, in the case of something like Li4BN3H10, there is a wide range of vibrational frequencies (from the heavy nitrogen to the very light hydrogen). So, what I ended up doing is taking a system average vibrational frequency based on some earlier phonon calculations (I forget what this frequency was, though) then doing a series of shortruns with VASP to see what the nose frequency was based on its output (it prints out the nose frequency). Since the mass is related to the frequency by f ~ sqrt(1/m), I was able to zero in on a 'close enough' value (it ended up being 5.0 for my calculations). You can probably also go into the source code to see how they set up the default SMASS to make this process even easier.
Hope that helps,
5楼2010-09-08 08:21:48
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖
相关版块跳转 我要订阅楼主 stractor 的主题更新
普通表情 高级回复(可上传附件)
信息提示
请填处理意见