[Pw_forum] [Fwd: Re: Contraint relaxation and SIC in version 3.1.1]

[Diep Quang Vinh]

Dear all,

The following suggestions (set tstress=.false.) fixed my problems

Thank you very much

Vinh

----------------------------------- Original Message -----------------------------------
Subject: Re: Contraint relaxation and SIC in version 3.1.1
From:    "Carlo Cavazzoni" <c.cavazzoni at cineca.it>
Date:    Mon, September 4, 2006 3:26 pm
To:      "Diep Quang Vinh" <vdiep at ictp.it>
----------------------------------------------------------------------------------------

dear Vinh

let's try to fix 3.1.1.
First of all you have to exclude the computation of stress:
tstress = .false.

since stress with SIC has not been implemented, neither in the 3.0 the stress was
implemented, it was siply wrong, now we have added the message. Then you have to check
the type of costraint.
Distance should be implemented, then there is an error in the input or a bug in the code.

In conclusion try first 3.1.1 without tstress,
then we will check for the constraint type

ciao
carlo

----- Original Message -----
From: "Diep Quang Vinh" <vdiep at ictp.it>
To: "Carlo Cavazzoni" <c.cavazzoni at cineca.it>
Sent: Monday, September 04, 2006 3:18 PM
Subject: Contraint relaxation and SIC in version 3.1.1


> Dear all,
> My problems are:
>
> 1. I need include the Self-Interaction-Correction (SIC) in the calculation. Actually,
I did it in the version 3.0 and the results are good.
> 2. Now i want to do some constraint (distance) relaxations (with SIC correction). I
think i should change to version 3.1.1??
>
> Here are my run:
>
>
> If i use the version 3.1 of QE the code give the the following error:
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>     from  module setup  : error #         1
>      Stress is not yet implemented with SIC
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> ERROR: 0031-250  task 0: Terminated
>
> It seems to me that the code don't understand the SIC. But if I use the previous
version of QE like 3.0 I meet the following error. In this case the code don't
understant the constrain.
>
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>     from init_constraint : error #         1
>     constrain type not implemented
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> ERROR: 0031-250  task 0: Terminated
>
> Both of them have the same input like the following:
>
>
> -------------
> &CONTROL
>  calculation = 'fpmd',
> title = 'resconstructed and ee',
>  prefix='RECON-EE-CON'
>  ndr = 60,
>  ndw = 61,
>  restart_mode = 'restart'
>  nstep  = 200
>  iprint = 20
>  isave  = 20
>  tstress = .TRUE.,
>  tprnfor = .TRUE.,
>  dt    = 4.d0
>  etot_conv_thr = 1.d-5,
>  pseudo_dir= '/sfs/sanfs/home/userictp/ictpcm01/UPF/'
> /
>
> &SYSTEM
>    ibrav=  14,
>     celldm(1) = 16.972000,
>    celldm(2) = 1.,
>    celldm(3) = 4.9733070,
>    celldm(4) = 0.0,
>    celldm(5) = 0.0,
>    celldm(6) = -0.5,
>    nat=  101,
>    ntyp= 4,
>    nspin = 2,
>    ecutwfc =80.0,
>    ecfixed = 68.0,
>    q2sigma = 8.0,
>    nbnd=145,
>    nelec=289,
>    nelup = 145.d0,
>    neldw = 144.d0,
>    sic = 'sic_mac',
>    sic_epsilon = 1.d0,
>    sic_alpha = 1.d0,
>    force_pairing = .TRUE.,
>    xc_type = 'BLYP',
> /
>
> &ELECTRONS
>  emass = 200.d0,
>  emass_cutoff = 2.5d0,
>  orthogonalization = 'ortho',
>  ortho_eps = 5.d-8,
>  ortho_max = 28,
>  electron_dynamics = 'damp'
>  electron_damping=0.1d0
>  electron_velocities = 'default'
>  electron_temperature = 'not_controlled',
> /
>
> &IONS
>  ion_dynamics = 'sd'
>  !on_damping=0.2d0
>  ion_velocities='zero'
>  ion_positions='default'
>  !ion_temperature='nose'
>  !fnosep=30.d0
>  !tempw=150.d0
>   ion_radius(1) = 0.8d0,
>  ion_radius(2) = 0.8d0,
>  ion_radius(3) = 0.8d0,
>  ion_radius(4) = 0.8d0,
>  sic_rloc = 1.d0,
>
> /
>
> &CELL
>  cell_dynamics = 'none',
> /
>
> ATOMIC_SPECIES
> Cl  35.4d0 Cl.blyp-mt.UPF
> O  16.0d0 O.blyp-mt.UPF
> C  12.0d0 C.blyp-mt.UPF
> H   1.0d0 H.blyp-mt.UPF
>
> ATOMIC_POSITIONS (bohr)
> Cl      2.61430   7.78152 -34.88870  1 1 1 1
> Cl      3.35627   4.05314 -31.36570  1 1 1 1
> Cl      6.74509   3.51397 -35.12750  1 1 1 1
> Cl      7.44480   8.10866 -32.96470  1 1 1 1
> C       5.03399   5.90311 -33.57770  1 1 1 1
> H      15.45910  -3.29036 -22.36830  1 1 1 1
> ...................
> CONSTRAINTS
> 1
> 'distance' 1 5
>
> --------------
>
> If you ommit the these last lines(CONSTRAINTS), The version 3.0 work very well (no
error at all) but not the version 3.1.1.
>
>
> So it seems to me that the version 3.1.1 don't incoporate the SIC correction ? Am i
right?
>
>
> Thank you very much
>
> Best regards.
>
> Vinh
>
>
>
>
>
>
>
>
>
>
>
>