[QE-users] vc-relax

[Giuseppe Mattioli]

Dear Sabike
I see that you are using the BLYP functional, and I suppose that this  
is what you want to use. But 50 Ry for a norm-conserving  
Martins-Troullier C pseudopotential is probably a poor cutoff that can  
affect convergence. Moreover, there is no need to specify ecutrho for  
NC pseudopotentials, as it should be automatically fixed at 4*ecutwfc  
(check your output file). Finally, there are also a few parameters  
that can be adjusted to improve convergence, but it would be useful to  
know what you mean when you say "it doesn't converge". Does your  
calculation reach the maximum electron_maxstep or nstep?
HTH
Giuseppe

Stefano Baroni <baroni at sissa.it> ha scritto:

> Sorry, Chitta, but this does not make much sense. k-point sampling  
> is intimately related to periodicity: no periodicity, No k-point  
> sampling. The dependence of any quantity of a non periodic system on  
> the number of k point simply shows that the size of the supercell  
> may be too small. Using k-point sampling may in this case speed up  
> convergence towards the infinite-size limit, but in doing so one  
> should know what one is doing and why, which this discussion shows  
> not to be necessarily the case. The rule of thumb is: if your result  
> depend too much on the number of k point, increase the size of the  
> cell until it does not. The number of monolayers has nothing to do  
> with the rate of convergence. The vacuum in between pairs of  
> monolayers does. Regards — Stefano B
>
>
>> On 9 Oct 2018, at 06:34, Sai Pavan Chitta <chitta.sai at iitgn.ac.in> wrote:
>>
>> Dear Sabike,
>>
>> In my experience of working with monolayers, I found that the  
>> number of k-points in the k-mesh in z direction aalso affects the  
>> convergence. The energy starts converging the moment I increase the  
>> number of k-points in z-direction. It also makes sense as you are  
>> using two monolayers in your model and thus requires more k-points  
>> in the z-direction.
>>
>> Moreover, you can make the convergence threshold more stringent to  
>> about 1e-8 for a proper convergence.
>>
>> Regards
>> Chitta Sai Pavan
>> IIT Gandhinagar
>>
>>
>> On Tue, 9 Oct 2018, 08:18 Sabike Ghasemi, <s.ghasemi at du.ac.ir  
>> <mailto:s.ghasemi at du.ac.ir>> wrote:
>> dear all
>> i want to relax atomic coordinates and lattic constants for  
>> graphene bilayer. but it dosen't converge.
>> input file is:
>>
>>
>>  &CONTROL
>>                        title = GRAPHENEBILAYER ,
>>                  calculation = 'vc-relax' ,
>>                 restart_mode = 'from_scratch' ,
>>                   wf_collect = .true. ,
>>                       outdir = './out' ,
>>                       wfcdir = './out' ,
>>                   pseudo_dir = './' ,
>>                       prefix = 'GBI' ,
>>                    verbosity = 'high' ,
>>                etot_conv_thr = 1.0D-4 ,
>>                forc_conv_thr = 1.0D-4 ,
>>                      tstress = .true. ,
>>                      tprnfor = .true. ,
>>  /
>>  &SYSTEM
>>                        ibrav = 4,
>>                    celldm(1) = 4.648726,
>>                    celldm(3) = 9.512195384,
>>                          nat = 4,
>>                         ntyp = 1,
>>                      ecutwfc = 50 ,
>>                      ecutrho = 500 ,
>>                        nosym = .false. ,
>>                  occupations = 'smearing' ,
>>                      degauss = 0.002 ,
>>                     smearing = 'gaussian' ,
>>  /
>>  &ELECTRONS
>>             electron_maxstep = 200,
>>                     conv_thr = 1.0D-6 ,
>>                  mixing_beta = 0.3 ,
>>              diagonalization = 'david' ,
>>  /
>>  &IONS
>>                 ion_dynamics = 'bfgs' ,
>>  /
>>  &CELL
>>                cell_dynamics = 'bfgs' ,
>>                  cell_factor = 2.D0 ,
>>                  cell_dofree = '2Dxy' ,
>>  /
>> ATOMIC_SPECIES
>>     C   12.00000  C.blyp-mt.UPF
>> ATOMIC_POSITIONS crystal
>>     C      0.333333330    0.666666660    0.572649573
>>     C      1.000000000    0.000000000    0.572649753
>>     C      0.666666660    0.333333330    0.427350427
>>     C      1.000000000    0.000000000    0.427350427
>> K_POINTS automatic
>>   16 16 1   0 0 0
>>
>>
>> best,
>> sabike
>> _______________________________________________
>> users mailing list
>> users at lists.quantum-espresso.org <mailto:users at lists.quantum-espresso.org>
>> https://lists.quantum-espresso.org/mailman/listinfo/users  
>> <https://lists.quantum-espresso.org/mailman/listinfo/users>_______________________________________________
>> users mailing list
>> users at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/users
>
> —
> Stefano Baroni - Trieste —  http://stefano.baroni.me  
> <http://stefano.baroni.me/>



GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.mattioli at ism.cnr.it>