forum@abinit.org
Subject: The ABINIT Users Mailing List ( CLOSED )
List archive
- From: Yusuke Iguchi <iguchi@coral.t.u-tokyo.ac.jp>
- To: forum@abinit.org
- Subject: Re: [abinit-forum] Si(100): success and questions
- Date: Mon, 14 Jun 2004 19:38:11 +0900
Dear, Prasenjit Sen
I met a similar difficulty in convergece at the calculation of Si surface before.
At that time, it was converged by using a simple mixing.
Generally, the accelleration method such as Broyden mixing is fast but unstable since it supposes a smooth behavior of the energy in a given space.
Therefore, I always switch to a more stable simple mixing method when the convergence cannot be reached by another smart method.
Although it costs several times of time as you may guessed, it is faster than unconverged calculations.
I show my input file for Si(100) P(2x1) .
iscf=2 represents the simple mixing method and diemix controls the degree of mixing.
I hope you find it informative.
Best regards,
Yusuke IGUCHI.
---------------beginning of input file--------------
occopt 4
tsmear 0.01
ecut 6
ecutsm 0.1
kptopt 1
ngkpt 3 6 1
ionmov 5
iprcfc 1
ntime 40
tolmxf 5.0d-4
prtgeo 1
iscf 2
diemix 0.5
nstep 60
toldfe 1.0d-6
ntypeat 1
znucl 14
natom 24
typat 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
acell 14.514225 7.257112 41.052427
rprim 1.000000000000 0.000000000000 0.000000000000
0.000000000000 1.000000000000 0.000000000000
0.000000000000 0.000000000000 1.000000000000
xred 1.5535586130E-01 0.0000000000E+00 3.1181419291E-01
4.3938108714E-01 0.0000000000E+00 2.7982668639E-01
1.1993405498E-02 5.0000000000E-01 3.4460880224E-01
4.8389604800E-01 5.0000000000E-01 3.4319546729E-01
2.5533986209E-01 5.0000000000E-01 4.1503262880E-01
7.4715548326E-01 5.0000000000E-01 4.0090292677E-01
2.4554742920E-01 0.0000000000E+00 4.7581021353E-01
7.5197587707E-01 0.0000000000E+00 4.6465791321E-01
9.9047399686E-01 0.0000000000E+00 5.3382398078E-01
5.0687513737E-01 0.0000000000E+00 5.3031443544E-01
-5.8234745565E-03 5.0000000000E-01 5.9621456457E-01
5.0583752379E-01 5.0000000000E-01 5.9186372714E-01
2.5582347455E-01 5.0000000000E-01 6.5378543543E-01
7.4416247620E-01 5.0000000000E-01 6.5813627286E-01
2.5952600314E-01 0.0000000000E+00 7.1617601922E-01
7.4312486263E-01 0.0000000000E+00 7.1968556456E-01
4.4525707976E-03 0.0000000000E+00 7.7418978647E-01
4.9802412293E-01 0.0000000000E+00 7.8534208679E-01
9.9466013791E-01 5.0000000000E-01 8.3496737120E-01
5.0284451673E-01 5.0000000000E-01 8.4909707323E-01
2.3800659450E-01 5.0000000000E-01 9.0539119776E-01
7.6610395199E-01 5.0000000000E-01 9.0680453271E-01
9.4644138695E-02 0.0000000000E+00 9.3818580709E-01
8.1061891285E-01 0.0000000000E+00 9.7017331361E-01
-----------------end of input file --------------
- Si(100): success and questions, Prasenjit Sen, 06/14/2004
- Re: [abinit-forum] Si(100): success and questions, verstraete, 06/14/2004
- Re: [abinit-forum] Si(100): success and questions, Yusuke Iguchi, 06/14/2004
Archive powered by MHonArc 2.6.16.