[Pw_forum] Total forces of relaxing metallic system decreases then increases gradually

[Hind Almisbahi]

Dear experts of quantum espresso,

I am doing relaxation and the problem is that the total force is decreasing
until the 4th scf, then start to increase gradually. The system does not
converge. Actually, I need to calculate the band structure and see the band
gap of Armchair graphene ribbon. The ribbon width is composed of 17 carbon
atoms (N=3p+2 family), so it is metallic. I have one unit cell of the
ribbon. I read in this mailing list many useful discussions about suitable
input for metallic systems and I followed them as I understand. The version
of quantum espresso is 6.1.

I tried many options of the input script but all the runs have the same
problem, total force is decreasing until the 4th scf then start to increase
gradually. This is a complete version of my input:



&CONTROL

  calculation='relax',

  outdir='.\Work',

  prefix='Arm17',

  pseudo_dir='C:\qe-6.1\pseudo',

  verbosity='high',

  etot_conv_thr= 1.0D-3,

  forc_conv_thr=1.0d-2,

/

&SYSTEM

  ibrav=8,

  a=41.1089d0,

  b=4.26258d0,

  c=20.0d0,

  nat=34,

  ntyp=1,

  ecutwfc=65,

  ecutrho=520,

  nbnd=76

  occupations='smearing',

  smearing='marzari-vanderbilt',

  degauss=0.001d0,

/



&ELECTRONS

  conv_thr=1d-06,

  mixing_beta=0.7d0,

  mixing_mode='local-TF'

  mixing_ndim=12

  diagonalization='cg'

/



&IONS

/



ATOMIC_SPECIES

  C 12.010700d0 C.pbe-rrkjus.UPF



ATOMIC_POSITIONS {crystal}

   C   0.0336555229d0   0.6666666956d0   0.5000000000d0

   C   0.0919486003d0   0.8333333913d0   0.5000000000d0

   C   0.0336555229d0   0.3333333044d0   0.5000000000d0

   C   0.1502416777d0   0.6666666956d0   0.5000000000d0

   C   0.2085347551d0   0.8333333913d0   0.5000000000d0

   C   0.0919486003d0   0.1666666087d0   0.5000000000d0

   C   0.1502416777d0   0.3333333044d0   0.5000000000d0

   C   0.2668277852d0   0.6666666956d0   0.5000000000d0

   C   0.3251206731d0   0.8333333913d0   0.5000000000d0

   C   0.2085347551d0   0.1666666087d0   0.5000000000d0

   C   0.2668277852d0   0.3333333044d0   0.5000000000d0

   C   0.3834141768d0   0.6666666956d0   0.5000000000d0

   C   0.4417072068d0   0.8333333913d0   0.5000000000d0

   C   0.3251206731d0   0.1666666087d0   0.5000000000d0

   C   0.3834141768d0   0.3333333044d0   0.5000000000d0

   C   0.5000002369d0   0.6666666956d0   0.5000000000d0

   C   0.5582932669d0   0.8333333913d0   0.5000000000d0

   C   0.4417072068d0   0.1666666087d0   0.5000000000d0

   C   0.5000002369d0   0.3333333044d0   0.5000000000d0

   C   0.6165862969d0   0.6666666956d0   0.5000000000d0

   C   0.6748793269d0   0.8333333913d0   0.5000000000d0

   C   0.5582932669d0   0.1666666087d0   0.5000000000d0

   C   0.6165862969d0   0.3333333044d0   0.5000000000d0

   C   0.7331723570d0   0.6666666956d0   0.5000000000d0

   C   0.7914653870d0   0.8333333913d0   0.5000000000d0

   C   0.6748793269d0   0.1666666087d0   0.5000000000d0

   C   0.7331723570d0   0.3333333044d0   0.5000000000d0

   C   0.8497584170d0   0.6666666956d0   0.5000000000d0

   C   0.9080514470d0   0.8333333913d0   0.5000000000d0

   C   0.7914653870d0   0.1666666087d0   0.5000000000d0

   C   0.8497584170d0   0.3333333044d0   0.5000000000d0

   C   0.9663444771d0   0.6666666956d0   0.5000000000d0

   C   0.9080514470d0   0.1666666087d0   0.5000000000d0

   C   0.9663444771d0   0.3333333044d0   0.5000000000d0



K_POINTS {automatic}

1       50  1  0  0  0



These are some of the many options that I have explored:

The above potential is ultrasoft, I used also norm conserving
(C.pbe-mt_gipaw.UPF).

Smearing: I used methfessel-paxton and marzari-vanderbilt

Degause: I used 0.01 and 0.001

mixing_beta: I used 0.1 and 0.7

mixing_mode: I used local-TF and TF

mixing_ndim: I used 8 and 12

diagonalization: I used cg and david

ecutwfc: I used 65 and 75

ecutrho: with norm conserving (ecutwfc*4 and ecutwfc*10), with ultrasoft
(ecutwfc*8)

nbnd=76 (I have 34 atoms, so 136 valence electrons which need 68 bands,
because it is metallic I added 8 extra bands)

Previously I have successfully calculated the band structure and band gap
of semiconductor ribbon but I could not do the calculation of metallic
ribbon. I appreciate any help to solve my problem, why the forces always
increase after the 4th scf?

​

Best Regards,

Hind

Lecturer, IT department

King Abdulaziz University

-- 
Disclaimer: The information contained in this message is intended for the 
addressee only and may contain classified information. If you are not the 
addressee, please delete this message and notify the sender; you should not 
copy or distribute this message or disclose its contents to anyone. Any 
views or opinions expressed in this message are those of the individual(s) 
and not necessarily of the university. No reliance may be placed on this 
message without written confirmation from an authorized representative of 
its contents. No guarantee is implied that this message or any attachment 
is virus free or has not been intercepted and amended. 

إخلاء مسؤليه:  ان جميع المعلومات المتضمنة في هذه الرسالة تخص المستلم، وربما 
تتضمن معلومات سرية، واذا لم تكن انت المستلم فالرجاء الغاء الرسالة واشعار 
المرسل، كما يمنع نسخ او توزيع هذه الرسالة او افشاء محتوياتها لاخرين. ويجدر 
بالعلم ان جميع الآراء ووجهات النظر الواردة في هذه الرسالة تعتبر شخصية وليست 
بالضرورة منسوبة للجامعة. كما لايمكن ان يعتمد على هذه الرسالة ما لم تحصل على 
موافقة خطية من المسئول عن محتوياتها. كما لايوجد ضمان بان هذه الرسالة أو 
مرفقاتها خالية من الفيروسات أو انها لم يتم اعتراضها وتعديلها

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20180218/4093025b/attachment.html>