The ABINIT Users Mailing List ( CLOSED )

["Shtogun, Yaroslav" <yshtogun@cas.usf.edu>]

Hi! "Bobryi Den' Usim" 
- Thank you to everybody for help. I just get convergence very fest with
Pierre-Matthieu Anglade' advice - "iprcel 49" just 13 steps.
 - I used default value for tsmear - 0.04.
 - I didn't check jet convergence according k-points mesh, but I did
this for (7,0) CNT and 1x1x4 was the best, but I will look at it. Also,
I investigate convergence according ecut and I decide to take 32Ha
because the difference between 32 and 34 Ha was very small. The size of
unit cell is big for my future calculation; I need size of unit cell
where I can neglect interaction between nearest nanotubes - 12R is good.


Thank you,

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: Masayoshi Mikami [mailto:mmikami@rc.m-kagaku.co.jp] 
Sent: Thursday, June 15, 2006 9:12 PM
To: forum@abinit.org
Subject: Re: [abinit-forum] Convergence

Dear Yaroslav Shtogun,

Hello. ("dobry den" in Ukrainian ? Just a guess, if not, sorry.)
I have just noticed something in your input.
You should use tsmear, not "ecutsm",
along with occopt (3, 4, 5, 6, or 7) for such a "metallic" case.

Please consult with the help files.
http://www.abinit.org/Infos_v4.6/varbas.html#occopt
http://www.abinit.org/Infos_v4.6/vargs.html#tsmear

In passing, "ecutsm" is used for geometry optimization
including cell volume.
http://www.abinit.org/Infos_v4.6/varrlx.html#ecutsm

If you are not familiar with these variables,
just have fun with Tutorials/Test_* in ABINIT tar.gz ball.
Have a nice weekend.

Good continuation,
Masayoshi

P.S. Thanks for following the Netiquette. (i.e. input attachment !)
Speaking of Netiquette, not a few ABINITioners here on this ML
might also want to read  the following WEB page by Eric S. Raymond.
   "How To Ask Questions The Smart Way"
     http://www.catb.org/~esr/faqs/smart-questions.html
There are not a few translations available there. Enjoy !

On 2006/06/15, at 23:52, Anglade Pierre-Matthieu wrote:

> Hi,
>
> Your preconditioning is not correct. There is two problem with it:
> your nanotube is metallic so you must expect a higher value for diemac
> and your cell is not homogeneous (lots of vacuum). You'd better use
> "iprcel 49" instead of "diemac 12" it will certainly helps a lot.
>
> regards
>
> PMA
>
> On 6/15/06, yshtogun@cas.usf.edu <yshtogun@cas.usf.edu> wrote:
>> Convergence
>>  Dear ABINIT users!
>>   I try to calculate Etotal for metal carbon nanotubes, but I was  
>> not able to reach convergence. I saw that in my output file Etotal  
>> starts wobbling around some value, I looked at lesson for Al  and  
>> did the same way.  There is my out put file:
>>
>>  ETOT 81  -138.61248625690    -1.621E-02 7.759E-02 4.898E+03  
>> 7.247E-02 6.434E-02
>>  ETOT 82  -138.59572056047     1.677E-02 8.067E-02 4.905E+03  
>> 7.189E-02 6.463E-02
>>  ETOT 83  -138.59789426457    -2.174E-03 9.536E-02 4.916E+03  
>> 7.272E-02 6.541E-02
>>  ETOT 84  -138.59854520570    -6.509E-04 8.775E-02 4.917E+03  
>> 7.350E-02 6.541E-02
>>  ETOT 85  -138.59404342836     4.502E-03 9.851E-02 4.921E+03  
>> 7.384E-02 6.486E-02
>>  ETOT 86  -138.59218749179     1.856E-03 9.981E-02 4.921E+03  
>> 7.315E-02 6.471E-02
>>  ETOT 87  -138.58897216161     3.215E-03 1.020E-01 4.922E+03  
>> 7.301E-02 6.429E-02
>>  ETOT 88  -138.59681906813    -7.847E-03 1.188E-01 4.918E+03  
>> 7.280E-02 6.450E-02
>>  ETOT 89  -138.60192171587    -5.103E-03 1.160E-01 4.911E+03  
>> 7.209E-02 6.342E-02
>>  ETOT 90  -138.63243709379    -3.052E-02 1.452E-01 4.891E+03  
>> 7.288E-02 6.528E-02
>>  ETOT 91  -138.67069938233    -3.826E-02 1.283E-01 4.855E+03  
>> 7.464E-02 6.543E-02
>>  ETOT 92  -138.65744598321     1.325E-02 9.016E-02 4.839E+03  
>> 7.518E-02 6.583E-02
>>  ETOT 93  -138.69142100515    -3.398E-02 1.003E-01 4.840E+03  
>> 7.397E-02 6.380E-02
>>  ETOT 94  -138.67800496752     1.342E-02 1.362E-01 4.844E+03  
>> 7.331E-02 6.517E-02
>>  ETOT 95  -138.78701676235    -1.090E-01 1.661E-01 4.777E+03  
>> 7.236E-02 5.931E-02
>>  ETOT 96  -138.81659394107    -2.958E-02 8.866E-02 4.738E+03  
>> 7.003E-02 6.284E-02
>>  ETOT 97  -138.76127302580     5.532E-02 1.127E-01 4.780E+03  
>> 7.247E-02 6.457E-02
>>  ETOT 98  -138.83699500452    -7.572E-02 1.895E-01 4.735E+03  
>> 7.125E-02 6.081E-02
>>  ETOT 99  -138.84379116069    -6.796E-03 8.810E-02 4.714E+03  
>> 7.024E-02 6.172E-02
>>  ETOT100  -138.78397012486     5.982E-02 1.138E-01 4.761E+03  
>> 7.426E-02 6.483E-02
>>
>> My input file:
>> # (6,6) Carbon Nanotube : computation of the total energy
>> # bond legth a = 1.419 angstrom 1 unit cell
>> # the potential files - 6c.pspnc LDA (Troullier - Martins)
>>
>>
>> #Definition of the unit cell
>> acell 48.7816266774 48.7816266774 2.4577800959  angstrom    # The  
>> size of unit cell in angstrom
>>
>>
>> #Definition of the atom types
>> ntypat 1          # There is only one type of atom
>> znucl 6           # The keyword "znucl" refers to the atomic  
>> number of the
>>                   # possible type(s) of atom. The pseudopotential(s)
>>                   # mentioned in the "files" file must correspond
>>                   # to the type(s) of atom. Here, the only type is  
>> Silicon.
>>
>>
>> #Definition of the atoms
>> natom 24          # There are 24 atoms
>> typat 24*1
>>                   # They all are of type 1, that is, Carbon.
>>
>> xangst            # This keyword indicate that the location of the  
>> atoms
>>                   # will follow, one triplet of number for each atom
>>                   # Triplet giving the cartesian coordinates of  
>> all atoms, in Angstrom
>> 28.4559489387  24.3908133387  0.0000000000
>> 28.2107912387  25.7811715387  0.0000000000
>> 26.4233811387  27.9113240387  0.0000000000
>> 25.0967167387  28.3941903387  0.0000000000
>> 22.3582455387  27.9113240387  0.0000000000
>> 21.2767388387  27.0038321387  0.0000000000
>> 20.3256777387  24.3908133387  0.0000000000
>> 20.5708354387  23.0004551387  0.0000000000
>> 22.3582455387  20.8703026387  0.0000000000
>> 23.6849099387  20.3874363387  0.0000000000
>> 26.4233811387  20.8703026387  0.0000000000
>> 27.5048878387  21.7777945387  0.0000000000
>> 27.9113240387  26.4233811387  1.2288900000
>> 27.0038321387  27.5048878387  1.2288900000
>> 24.3908133387  28.4559489387  1.2288900000
>> 23.0004551387  28.2107912387  1.2288900000
>> 20.8703026387  26.4233811387  1.2288900000
>> 20.3874363387  25.0967167387  1.2288900000
>> 20.8703026387  22.3582455387  1.2288900000
>> 21.7777945387  21.2767388387  1.2288900000
>> 24.3908133387  20.3256777387  1.2288900000
>> 25.7811715387  20.5708354387  1.2288900000
>> 27.9113240387  22.3582455387  1.2288900000
>> 28.3941903387  23.6849099387  1.2288900000
>>
>> #Definition of the planewave basis set
>> ecut  20.0         # Maximal kinetic energy cut-off, in Hartree
>>
>> #Exchange-correlation functional
>> ixc 1             # LDA Teter Pade parametrization
>> #iscf 2
>>
>> ecutsm 0.5
>>
>> #Definition of the k-point grid
>> kptopt 1          # Option for the automatic generation of k  
>> points, taking
>>                   # into account the symmetry
>> ngkpt 1 1 4       # This is a 1x1x4 grid based on the primitive  
>> vectors
>> nshiftk 1
>> shiftk  0 0 0.5
>>
>>
>> #Definition of the SCF procedure
>> nstep 100         # Maximal number of SCF cycles
>> toldfe 1.0d-6     # Will stop when, twice in a row, the difference
>>                   # between two consecutive evaluations of total  
>> energy
>>                   # differ by less than toldfe (in Hartree)
>> diemac 12.0       # Although this is not mandatory, it is worth to
>>                   # precondition the SCF cycle. The model dielectric
>>                   # function used as the standard preconditioner
>>                   # is described in the "dielng" input variable  
>> section.
>>                   # Here, we follow the prescription for carbon.
>>
>>                Should I specify some other parameters?
>>         I appreciate all suggestion and comments.
>>
>>   Thank you in advance!
>>
>>   Yaroslav!
>>
>>
>
>
> -- 
> Pierre-Matthieu Anglade
>