[biao]

I found the problem

I used the FINDSYM to make a new cif and used the cif2cell to make an input file, the input file looks like this:

&CONTROL
calculation=’scf’,
outdir=’.’,
prefix=’FeAl96′,
pseudo_dir=’/home/max/workQE/my_pseudo’,
verbosity=’low’,
tprnfor=.true.,
tstress=.true.,
/
&SYSTEM
ibrav = 0
A = 5.3
nat = 8
ntyp = 2
ecutwfc=70,
ecutrho=700,
input_dft=’pbe’,
occupations=’smearing’,
smearing=’mv’,
degauss=0.005d0,
/
&ELECTRONS
conv_thr=1d-08,
mixing_beta=0.7d0,
/
CELL_PARAMETERS {alat}
1.000000000000000 0.000000000000000 0.000000000000000
0.000000000000000 1.060667434561135 0.000000000000000
0.000000000000000 0.000000000000000 0.612376980364380
ATOMIC_SPECIES
Fe 55.84500 Fe.pbe-spn-kjpaw_psl.0.2.1.UPF
Al 26.98100 Al.pbe-n-kjpaw_psl.1.0.0.UPF
ATOMIC_POSITIONS {crystal}
Fe 0.250000000000000 0.083330000000000 0.250000000000000
Fe 0.750000000000000 0.583330000000000 0.750000000000000
Fe 0.750000000000000 0.916670000000000 0.750000000000000
Fe 0.250000000000000 0.416670000000000 0.250000000000000
Al 0.250000000000000 0.583330000000000 0.750000000000000
Al 0.750000000000000 0.083330000000000 0.250000000000000
Al 0.750000000000000 0.416670000000000 0.250000000000000
Al 0.250000000000000 0.916670000000000 0.750000000000000
K_POINTS {automatic}
6 4 4 0 0 0
This input gives the above results

Then I used the cif2cell directly convert the cif provided and the input looks like this:
&CONTROL
calculation=’scf’,
outdir=’.’,
prefix=’FeAl44_p1′,
pseudo_dir=’/home/max/workQE/my_pseudo’,
verbosity=’low’,
tprnfor=.true.,
tstress=.true.,
/

&SYSTEM
ibrav = 0
A = 2.5121
nat = 4
ntyp = 2
ecutwfc=70,
ecutrho=700,
input_dft=’pbe’,
occupations=’smearing’,
smearing=’mv’,
degauss=0.005d0,
/

&ELECTRONS
conv_thr=1d-08,
mixing_beta=0.7d0,
/

CELL_PARAMETERS {alat}
1.000000000000000 0.000000000000000 0.000000000000000
0.000000000000000 1.632981042829164 0.000000000000000
0.000000000000000 0.000000000000000 1.732049813384575

ATOMIC_SPECIES
Fe 55.84500 Fe.pbe-spn-kjpaw_psl.0.2.1.UPF
Al 26.98100 Al.pbe-n-kjpaw_psl.1.0.0.UPF

ATOMIC_POSITIONS {crystal}
Fe 0.750000000000000 0.750000000000000 0.916666667000000
Fe 0.250000000000000 0.250000000000000 0.083333333000000
Al 0.250000000000000 0.750000000000000 0.416666667000000
Al 0.750000000000000 0.250000000000000 0.583333333000000

K_POINTS {automatic}
6 4 4 0 0 0

The new input works very well

[biao]

I had the same problem with cif files from Kunz et al., 1991. But I modified the cif file a bit: remove the “a” in the name of elements and the modified cif file should work with cif2cell. I hope this trick will help you.

The warning looks like this:
***Warning: Si2a is not a chemical element.
***Warning: Si3a is not a chemical element.
***Warning: Si4a is not a chemical element.
***Warning: O1a is not a chemical element.
***Warning: O2a is not a chemical element.
***Warning: O3a is not a chemical element.
***Warning: O4a is not a chemical element.
***Warning: O5a is not a chemical element.
***Warning: O6a is not a chemical element.
***Warning: O7a is not a chemical element.
***Warning: O8a is not a chemical element.
***Warning: O9a is not a chemical element.
***Warning: O10a is not a chemical element.
***Warning: O11a is not a chemical element.
***Warning: O12a is not a chemical element.
***Warning : Site occupancies not found, assuming all occupancies = 1.
***Error: Chemical composition of the generated cell differs from that given
by _chemical_formula_sum. Use –force to generate a cell anyway.

[biao]

The Born-Oppenheimer approximation is only valid if the electron state is sufficiently separated from the other electron states. The Born-Oppenheimer approximation fails when the electronic states cross or approach.

[biao]

i like your explanation

[biao]

good explanation

[biao]

great explanation

[biao]

good explanation

[biao]

i like your explanation

[biao]

i like your explanation

[biao]

i like your answer

[biao]

They are the sample

[biao]

I like the explanation about the third question

[biao]

I like your answer to the third question!

[biao]

I think that the cost depended on the complexity of your material, some small and simple system can be calculated with your own PC but some large system will need HPC. There are many different models can be used in calculations and they are always compared with the results from experiments. Some models are proved to be correct in some aspect and not so correct in other aspect. And it is possible to do computation with your own PC if you want to study simple system

[Author]

Posts