I have calculated the atomic volumes of the actinoids from the EOS of bcc-cells, non-magnetic, with and without SOC ( primary to test pseudopotentials I have created.). However, I am quite surprised that beyond Am the SOC results deviate so strongly from the scalar-relativistic calculations (see file attached). All EOS curves look smooth. I have fixed the MT-radius to the smallest used volume. My PAW calculations show a similar trend, however deviate more strongly above Am from the Elk calculations. Unfortunately, all references I can find in the web are dealing with the light actinoides. So I don't know if my results are reliable or if I have missed something in the input file. Does anyone has an idea?
Hi, Malte,
I think these peculiarities in your plot are related to the complicated nature of 5f states in actinides, which are believed to change their behavior from transition metals-like to rare earth metals-like exactly at Am - you can have a look at this review http://dx.doi.org/10.1103/RevModPhys.81.235
Actinides are also known to be systems, where SOC is important (because these are heavy elements and jj-type coupling dominates), so there is no wonder your scalar relativistic calculations and the ones including SOC are different.
In addition, one should remember that 5f electrons in actinide metals are strongly correlated and e.g. DFT+U corrections should be considered when calculating the ground state of an actinide.
Best regards,
Anton F.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear all,
I have calculated the atomic volumes of the actinoids from the EOS of bcc-cells, non-magnetic, with and without SOC ( primary to test pseudopotentials I have created.). However, I am quite surprised that beyond Am the SOC results deviate so strongly from the scalar-relativistic calculations (see file attached). All EOS curves look smooth. I have fixed the MT-radius to the smallest used volume. My PAW calculations show a similar trend, however deviate more strongly above Am from the Elk calculations. Unfortunately, all references I can find in the web are dealing with the light actinoides. So I don't know if my results are reliable or if I have missed something in the input file. Does anyone has an idea?
Best regards,
Malte
Here is an example of an input file:
tasks
0
avec
0.5 0.5 0.5
-0.5 0.5 0.5
-0.5 -0.5 0.5
scale1
7.11378660898
scale2
7.11378660898
scale3
7.11378660898
scale
1.06
xctype
20
sppath
'./'
rgkmax
9.5
gmaxvr
18
lmaxapw
10
lmaxo
8
lradstp
2
msmooth
4
nempty
15
autolinengy
.true.
stype
1
swidth
0.01
mixtype
3
broydpm
0.3 0.12
epspot
1.E-06
epsengy
1.E-05
spinpol
.false.
spinorb
.true.
atoms
1 : nspecies
'Md.in' : spfname
1 : natoms
0.0 0.0 0.0 0.0 0.0 0.0 : atposl, bfcmt
ngridk
15 15 15
Sorry: in the key the SR and SOC results have to be changed!
Hi, Malte,
I think these peculiarities in your plot are related to the complicated nature of 5f states in actinides, which are believed to change their behavior from transition metals-like to rare earth metals-like exactly at Am - you can have a look at this review http://dx.doi.org/10.1103/RevModPhys.81.235
Actinides are also known to be systems, where SOC is important (because these are heavy elements and jj-type coupling dominates), so there is no wonder your scalar relativistic calculations and the ones including SOC are different.
In addition, one should remember that 5f electrons in actinide metals are strongly correlated and e.g. DFT+U corrections should be considered when calculating the ground state of an actinide.
Best regards,
Anton F.