ABINIT Discussion Forums

Hello,

I'm trying to calculate the positron lifetimes for vacancy defects in a transition metal oxide. When calculating the bulk positron lifetime, by employing positron=1 and a zero positron density limit exchange-correlation functional (whether LDA or GGA), the calculations give the desired results. But, when trying to use the automated procedure to obtain the positron lifetimes for the mentioned defects (because I use the PAW Psp) the calculations fail when entering the 3rd Scf cycle. TC-DFT STEP 1 - ELECTRONIC GROUND-STATE CALCULATION and TC-DFT STEP 2 - POSITRONIC GROUND-STATE CALCULATION IN PRESENCE OF ELECTRONS AND IONS converges and ends after the toldfe parameter is satisfied. The TC-DFT STEP 3 - ELECTRONIC GROUND-STATE CALCULATION IN PRESENCE OF A POSITRON ends abruptly without any error message. For this I've employed the ixcpositron=2: LDA electron-positron correlation, due to the nature of the calculation - defects. I thought that this problem could be a bug related to the ixpositron, but even switching to any other zero positron density limit Exc, this problem still occurs. I'm also using the same criteria convergence for postoldfe as I do in toldfe (10^-8).

I would appreciate some help regarding this matter. Meanwhile I'm running the manual procedure to see the respective outcome.
Regards
elora

Hello elora,

I use ixcpositron=2 and it works well (at least in the 6.10.3 version), so it is probably not a bug. Was the 3rd step started? Maybe I could take a look at your input file? What value did you use for positron?

As for the convergence, I think that it is better to set the postoldfe to 10*toldfe, but it shouldn't stop the calculation.

Regards,
Julia

Hello Julia,

Thanks for your reply. I'm using Version 6.12.3 of ABINIT. I have tried several input parameters to see if I could find out what the problem was. Below I copied an example of a run, where I define the postoldfe tolerance (when using this tolerance I define postoldfe=1.0d-8 but I also increase the value up to 1.0d-6)

# Y203: bulk
#
#Definition of the unit cell
acell 10.67800 10.67800 10.67800 angstrom
rprim 1.0 0.0 0.0 #
0.0 1.0 0.0 #
0.0 0.0 1.0 #
chkprim 0 #CHecK whether the cell is PRIMitive
#Definition of the atom types
ntypat 2
znucl 39 8


#Definition of the atoms
natom 79
typat 31*1 48*2
xred
2.5000000000E-01 7.5000000000E-01 7.5000000000E-01
7.5000000000E-01 2.5000000000E-01 7.5000000000E-01
7.5000000000E-01 7.5000000000E-01 2.5000000000E-01
7.5000000000E-01 7.5000000000E-01 7.5000000000E-01
7.5000000000E-01 2.5000000000E-01 2.5000000000E-01
2.5000000000E-01 7.5000000000E-01 2.5000000000E-01
2.5000000000E-01 2.5000000000E-01 7.5000000000E-01
9.6716863427E-01 -5.2179045880E-04 2.5088586146E-01
2.5088586146E-01 9.6716863427E-01 -5.2179045880E-04
-5.2179045880E-04 2.5088586146E-01 9.6716863427E-01
3.1839245018E-02 4.9853261748E-01 2.5243467542E-01
2.5243467542E-01 3.1839245018E-02 4.9853261748E-01
4.9853261748E-01 2.5243467542E-01 3.1839245018E-02
3.2284779812E-02 -6.0980987871E-04 7.5064328717E-01
7.5064328717E-01 3.2284779812E-02 -6.0980987871E-04
-6.0980987871E-04 7.5064328717E-01 3.2284779812E-02
9.6724287261E-01 4.9971509262E-01 7.4758616993E-01
7.4758616993E-01 9.6724287261E-01 4.9971509262E-01
4.9971509262E-01 7.4758616993E-01 9.6724287261E-01
4.6771522019E-01 5.0060980988E-01 7.4935671283E-01
7.4935671283E-01 4.6771522019E-01 5.0060980988E-01
5.0060980988E-01 7.4935671283E-01 4.6771522019E-01
5.3275712739E-01 2.8490738100E-04 7.5241383007E-01
7.5241383007E-01 5.3275712739E-01 2.8490738100E-04
2.8490738100E-04 7.5241383007E-01 5.3275712739E-01
5.3283136573E-01 5.0052179046E-01 2.4911413854E-01
2.4911413854E-01 5.3283136573E-01 5.0052179046E-01
5.0052179046E-01 2.4911413854E-01 5.3283136573E-01
4.6816075498E-01 1.4673825177E-03 2.4756532458E-01
2.4756532458E-01 4.6816075498E-01 1.4673825177E-03
1.4673825177E-03 2.4756532458E-01 4.6816075498E-01
4.0987346640E-01 1.2970544875E-01 3.9411734671E-01
3.8975227112E-01 8.4326250528E-01 1.1956647110E-01
1.0929059200E-01 1.5182584168E-01 6.2418659378E-01
6.0843028942E-01 3.4877381453E-01 3.7980228967E-01
3.9411734671E-01 4.0987346640E-01 1.2970544875E-01
1.1956647110E-01 3.8975227112E-01 8.4326250528E-01
6.2418659378E-01 1.0929059200E-01 1.5182584168E-01
3.7980228967E-01 6.0843028942E-01 3.4877381453E-01
1.2970544875E-01 3.9411734671E-01 4.0987346640E-01
8.4326250528E-01 1.1956647110E-01 3.8975227112E-01
1.5182584168E-01 6.2418659378E-01 1.0929059200E-01
3.4877381453E-01 3.7980228967E-01 6.0843028942E-01
6.0864897014E-01 8.4861975306E-01 6.2091027766E-01
6.1101385346E-01 1.5258680374E-01 8.8004462689E-01
8.8867794957E-01 8.4796478867E-01 3.8025443342E-01
3.9200374538E-01 6.5212964643E-01 6.2099837476E-01
6.2091027766E-01 6.0864897014E-01 8.4861975306E-01
8.8004462689E-01 6.1101385346E-01 1.5258680374E-01
3.8025443342E-01 8.8867794957E-01 8.4796478867E-01
6.2099837476E-01 3.9200374538E-01 6.5212964643E-01
8.4861975306E-01 6.2091027766E-01 6.0864897014E-01
1.5258680374E-01 8.8004462689E-01 6.1101385346E-01
8.4796478867E-01 3.8025443342E-01 8.8867794957E-01
6.5212964643E-01 6.2099837476E-01 3.9200374538E-01
8.9135102986E-01 6.5138024694E-01 8.7908972234E-01
8.8898614654E-01 3.4741319626E-01 6.1995537311E-01
6.1132205043E-01 6.5203521133E-01 1.1974556658E-01
1.0799625462E-01 8.4787035357E-01 8.7900162524E-01
8.7908972234E-01 8.9135102986E-01 6.5138024694E-01
6.1995537311E-01 8.8898614654E-01 3.4741319626E-01
1.1974556658E-01 6.1132205043E-01 6.5203521133E-01
8.7900162524E-01 1.0799625462E-01 8.4787035357E-01
6.5138024694E-01 8.7908972234E-01 8.9135102986E-01
3.4741319626E-01 6.1995537311E-01 8.8898614654E-01
6.5203521133E-01 1.1974556658E-01 6.1132205043E-01
8.4787035357E-01 8.7900162524E-01 1.0799625462E-01
9.0126533596E-02 3.7029455125E-01 1.0588265329E-01
1.1024772888E-01 6.5673749472E-01 3.8043352890E-01
3.9070940800E-01 3.4817415832E-01 8.7581340622E-01
8.9156971058E-01 1.5122618547E-01 1.2019771033E-01
1.0588265329E-01 9.0126533596E-02 3.7029455125E-01
3.8043352890E-01 1.1024772888E-01 6.5673749472E-01
8.7581340622E-01 3.9070940800E-01 3.4817415832E-01
1.2019771033E-01 8.9156971058E-01 1.5122618547E-01
3.7029455125E-01 1.0588265329E-01 9.0126533596E-02
6.5673749472E-01 3.8043352890E-01 1.1024772888E-01
3.4817415832E-01 8.7581340622E-01 3.9070940800E-01
1.5122618547E-01 1.2019771033E-01 8.9156971058E-01
#Definition of the planewave basis set
ecut 18.5
#Definition of the k-point grid
kptopt 1 # Option for the automatic generation of k points, taking
# into account the symmetry
ngkpt 4 4 4
nshiftk 1 #
#
shiftk 0.5 0.5 0.5
#Optimization of the lattice parameters
ionmov 0

#Definition of the SCF procedure
nstep 5000 # Maximal number of SCF cycles
toldfe 1.0d-8 # Will stop when, twice in a row, the difference
# between two consecutive evaluations of total energy
# differ by less than toldfe (in Hartree)
# This value is way too large for most realistic studies of materials

diemac 15.0 # 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.


optforces 2 #calculation of forces at the end of the SCF iterations (like the stresses)

ixc 11 #GGA, Perdew-Burke-Ernzerhof GGA functional

usexcnhat 1 #USE eXchange-Correlation with NHAT (compensation charge density)
pawecutdg 40
pawnhatxc 0 #Flag for exact computation of gradients of NHAT density in eXchange-Correlation
pawsushat 0 #only plane-wave contribution to suscep. matrix is computed
pawstgylm 0 #g_l(r).Y_lm(r) are not stored in memory and recomputed

#Positron calculation
positron -10
ixcpositron 2 #LDA electron-positron correlation provided by Puska, Seitsonen, and Nieminen
postoldfe 1.0d-6 #POSitron calculation: TOLerance on the DiFference of total Energy
posnstep 5000 #POSitron calculation: max. Number of STEPs for the two-component DFT


As I thought it could be memory shortcomes I also added some paw variables in order to decrease these requirements and I've also included some other keywords related to the compensation charge density.

The last line of the output file is: TC-DFT STEP 3 - ELECTRONIC GROUND-STATE CALCULATION IN PRESENCE OF A POSITRON; but no step is started. The output error file gives me a segmentation fault. I've also tried the manual procedure and after the first 2 steps are calculated (positron=0 and positron=1) and defining positron=2, the job also crashes.

Thank-you for your help
Kind regards
elora

Dear elora,

You can try setting optforces=1 and optstress=0. Other choices for these variables usually make the automatic positron calculations crash.

Julia

Hello Julia,

I've been trying out some tests with the variables you've suggested and even so the jobs keep on crashing. In order to see if it is a bug problem or not, in our local clusters we still have the Abinit version 6.8.1...I've started a job there and the calculation has already entered the TC-DFT STEP 3, even when defining the variable optforces 2 (whereas in the former jobs the calculation crashes when starting this cycle). Can it be a bug of the new version of Abinit?

Regards
elora