The ABINIT Users Mailing List ( CLOSED )

[zhangting <zhangting@pku.edu.cn>]

Shtogun, Yaroslav,您好！

        Hi
   Thanks very much for your help! By using default rprim 
(1,0,0;0,1,0;0,0,1), the code can find a P1 symmetry,and the memory 
requirement decreases 90%, thanks a lot! your suggestion is very useful.
   My k-point setting is from CASTEP..for I'm also a CASTEP user. In CASTEP's 
settings,this cell(about 15Ax15Ax2.4A) need a 2x2x10's mesh,for the k-space 
grid density probably being 0.03~0.04 (1/Angstrom).In fact I don't clearly 
know whether there are differences between these two codes in k-space grid 
representation,
and maybe I don't exactly understand your question yet.Do you mean my k-point 
mesh density is too high or too low? or my unit cell is too small? 
   
                                                                              
   Zhang Ting
                                                                              
   Apr.7th 2006
                                                                              
   Peking Unv.

======= 2006-04-06 12:33:00 您在来信中写道：=======

>Hi!
> I also try to optimize the geometry 7_0 CNT but in my case I use default 
> value of rprim (1 0 0; 0 1 0; 0 0 1) see my input file and sketch of unit 
> cell in attachment. In this case abinit calculates the space group base on 
> atoms position. I can advise you use default value of rprim and see what 
> happened.  Also, use less value of toldff, your calculation will be faster, 
> then you specify more precisely. 
>
> On the other hand I am confused with ngkpt? 
>   Can you explain to my how you choose ngkpt? Why is it 2 2 10, what 
> criteria for this?
> 
> 
>
>Yaroslav Shtogun
>*********************************
>College Art & Science
>University of South Florida
>4202 East Fowler Ave
>Tampa, FL 33620-5700
>USA
> 
>Cell: (813) 401-0016
>Tel: (813) 974-8489
>Fax: (813) 974-5813
> 
>
> 
> 
>
>-----Original Message-----
>From: 张 [mailto:zhangting@pku.edu.cn] 
>Sent: Thursday, April 06, 2006 10:51 AM
>To: forum@abinit.org
>Cc: zhangting@ele.pku.edu.cn
>Subject: [abinit-forum] Why the code can't find a symmetry automatically?
>
>Dear Abinit user:
>   
>    I'm a beginner of abinit codes. I'm going to do some electron-
>phonon coupling calculation of CNTs with the code.When I want to 
>optimize the geometric configuration of (5,5) CNT,the program send me 
>a warning that it can't find the cell's space group. As a result,the 
>optimization job continues with maybe P1 symmetry, and consumes quite 
>large memory and time. Now the job is still 
>running, I'd like to know hwo to let the code find a space group 
>automaticaly? and how to impose a symmetry in the cell manually? Does 
>the cell have a P21/M (number 11) space group? I've set the keyword 
>spgroup = 11,but it doesn't work...  I'm also afraid that if the 
>single point energy is calculated with no symmetry, whether ot not the 
>next phonon calculation will be reliable? For there are often 
>imaginary frequencies near Gamma point in carbon systems(graphite,CNTs)
>    My input file is attached.If there are any other mistakes or 
>parameters unsuitable in the input file, don't hesitate to point 
>out.thanks a lot 
>
>                                                                    
>Zhang Ting
>                                                                    
>Apr.6th 2006

= = = = = = = = = = = = = = = = = = = =
                        

　　　　　　　　致
礼！
 
                                 
　　　　　　　　zhangting
　　　　　　　　zhangting@pku.edu.cn
　　　　　　　　　　2006-04-07