mirror of https://github.com/abinit/abinit.git
100 lines
3.4 KiB
Plaintext
100 lines
3.4 KiB
Plaintext
#test silicon linear chain and finite oscillating electric field
|
|
ndtset 5
|
|
|
|
getwfk1 0
|
|
nstep1 30
|
|
|
|
qprtrb2 0 0 1
|
|
vprtrb2 100.0 0.0
|
|
|
|
qprtrb3 0 0 1
|
|
vprtrb3 10.0 0.0
|
|
|
|
qprtrb4 0 0 1
|
|
vprtrb4 1.0 0.0
|
|
|
|
qprtrb5 0 0 1
|
|
vprtrb5 -10.0 0.0
|
|
|
|
#Common data
|
|
acell 2*10.00 50.00
|
|
diecut 1.20
|
|
dielam 0.5
|
|
diegap 0.2
|
|
ecut 2.00
|
|
getwfk 1
|
|
iprcel 45
|
|
ixc 3
|
|
kptopt 0
|
|
|
|
kpt
|
|
0.00000 0.00000 0.500
|
|
natom 8 nband 16
|
|
ngfft 2*16 64 nkpt 1
|
|
nstep 15
|
|
nsym 1 ntypat 1
|
|
occopt 1
|
|
rprim 1.0 0.0 0.0
|
|
0.0 1.0 0.0
|
|
0.0 0.0 1.0
|
|
symrel 1 0 0 0 1 0 0 0 1
|
|
xred 0.0 0.0 0.0
|
|
0.0 0.0 0.05
|
|
0.0 0.0 0.25
|
|
0.0 0.0 0.30
|
|
0.0 0.0 0.50
|
|
0.0 0.0 0.55
|
|
0.0 0.0 0.75
|
|
0.0 0.0 0.80
|
|
tnons 3*0.0
|
|
typat 8*1
|
|
tolwfr 1.e-22
|
|
wtk 1
|
|
znucl 14
|
|
|
|
pp_dirpath "$ABI_PSPDIR"
|
|
pseudos "14si.Hamann_mod"
|
|
|
|
#%%<BEGIN TEST_INFO>
|
|
#%% [setup]
|
|
#%% executable = abinit
|
|
#%% [files]
|
|
#%% files_to_test =
|
|
#%% t01.abo, tolnlines = 2, tolabs = 1e-8, tolrel = 1e-3
|
|
#%% [paral_info]
|
|
#%% max_nprocs = 1
|
|
#%% [extra_info]
|
|
#%% authors = Unknown
|
|
#%% keywords =
|
|
#%% description =
|
|
#%% Chain of Silicon diatomic molecules (4 Si2 molecules in the cell)
|
|
#%% Freeze oscillatory perturbations with different wavelengths and intensities,
|
|
#%% thanks to the qprtrb and vprtrb input variables.
|
|
#%% This should be linked with the computation of the dielectric constant,
|
|
#%% test v2#05, that uses directly the RF capabilities of ABINIT,
|
|
#%% for one diatomic molecule.
|
|
#%% For dataset 1, one reproduces the results obtained in Tv2#05, multiplied by 4.
|
|
#%% The total energy is consistent up to more than 10 digits :
|
|
#%% -6.6499924738006 Ha for Tv2#05, -26.599969895203 Ha for the present calculation.
|
|
#%% For dataset 2, the perturbation qprtrb 0 0 1 is frozen in, with vprtrb 100.
|
|
#%% The total energy is -26.600317638775 Ha. The difference wrt the unperturbed situation is
|
|
#%% 0.000348743572 Ha.
|
|
#%% For dataset 3, a much smaller perturbation (10 times smaller) is taken,
|
|
#%% giving total energy -26.599973367786 Ha. The difference wrt the unperturbed situation is
|
|
#%% 0.3472583 microHa.
|
|
#%% For dataset 4, an even smaller perturbation (100 times smaller) is taken,
|
|
#%% giving total energy -26.599969929928 Ha. The difference wrt the unperturbed situation is
|
|
#%% 0.000034725 microHa. With datasets 3 and 4, we are in the linear regime.
|
|
#%% The previous amplitude is better for such studies.
|
|
#%% Dataset 5 is the same as 3, with reversed amplitude. Results are similar to dataset 3.
|
|
#%% I had no sufficient time to analyze these data correctly and make the connection with the
|
|
#%% results of Tv2#05, unfortunately. The following (also test 02 below) gives some more data, and raise questions.
|
|
#%% There might be some problem with the use of qprtrb and vprtrb.
|
|
#%% For dataset 2, the group of the four lowest eigenenergies (each corresponding to a different molecule) is :
|
|
#%% -0.47198 -0.46381 -0.46091 -0.45266 , whose spread is 0.01932 Ha.
|
|
#%% One might think that the maximum and minimum of the potential are separated roughly by 0.02 Ha.
|
|
#%% The value vprtrb 100 corresponds to a cosine wave whose amplitude is 100, divided by the volume
|
|
#%% of the cell, that is 5000 Bohr^3 : 0.02 Ha. The maximum
|
|
#%% and minimum of the potential should thus be separated by 0.04 Ha. There seems to be a factor of 2 off.
|
|
#%%<END TEST_INFO>
|