[SOLVED] MIXing coefficients for ALCHemical potentials

[Mutta]

Dear all
How do I set the input file for Pb{Zr_xTi_1-x}O3 : computation of the total energy
Zr/Ti=20/80 The convergence in acell. I want to doped Zr/Ti = 0.2/0.8 ratio
but in the input files
Example 1. Suppose that we want to describe Ba(0.25) Sr(0.75) Ti O3.
The input variables related to the construction of the alchemical Ba(0.25) Sr(0.75) potential will be :
npsp 4
znucl 8 40 56 38
ntypat 3
ntypalch 1
mixalch 0.25 0.75 for doped Ba/Sr
so my input files how to I set in this
This is my input files
# Pb{Zr_xTi_1-x}O3 : computation of the total energy
# Zr/Ti=20/80.The convergence in acell
ndtset 2
#Definition of the atomic positions without relaxing the lattice parameters
optcell1 0
ionmov1 2
#dilatmx 2.5
#ecutsm 0.5

#Definition of the basis of the geometry determined by dataset 1 (so using getxcart 1)
getxcart2 1
optcell2 3
ionmov2 2
dilatmx2 2.7
ecutsm2 0.5

# COMMON INPUT DATA

#Definition of the unit cell
acell 7.336 7.336 7.844
rprim 1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
spgroup 99
brvltt -1

#Definition of the atom types
npsp 4
znucl 82 40 22 8
ntypat 3
ntypalch 1
mixalch 0.2 0.8

#Definition of the atoms
natom 5
typat 1 2 3 3 3
xred
0.0 0.0 0.00
0.5 0.5 0.30
0.5 0.5 0.10
0.0 0.5 0.55
0.5 0.0 0.55
natfix 1
iatfix 1

#Definition of the planewave basis set
ecut 20.0
pawecutdg 35
pawovlp -1

# find the optimal total energy
ntime 50
tolmxf 1.0d-5
toldff 1.0d-6

#Definition of the k-point grid
kptopt 1
ngkpt 4 4 4
nshiftk 1
shiftk
0.5 0.5 0.5

#Definition of the SCF procedure
nstep 100
#toldfe 1.0d-6
diemac 1000.0

Best.
Abdulmutta

[gonze]

Dear Mutta,

You write :
"The input variables related to the construction of the alchemical Ba(0.25) Sr(0.75) potential will be :
npsp 4
znucl 8 40 56 38
ntypat 3
ntypalch 1
mixalch 0.25 0.75 for doped Ba/Sr"

This is right, but

npsp 4
znucl 82 40 22 8
ntypat 3
ntypalch 1
mixalch 0.2 0.8

will not give you Pb{Zr_xTi_1-x}O3 , it will give you PbZr{Ti_x_O1-x}3 .

For Pb{Zr_xTi_1-x}O3, you have to use
npsp 4
znucl 82 8 40 22
ntypat 3
ntypalch 1
mixalch 0.2 0.8

with
#Definition of the atoms
natom 5
typat 1 3 2 2 2
xred
0.0 0.0 0.00
0.5 0.5 0.30
0.5 0.5 0.10
0.0 0.5 0.55
0.5 0.0 0.55

Best wishes,
X

[Mutta]

Dear Gonze
How do I used PAW pp. for mixing Zr/Ti .
I ran to find optimized structure of Pb{Zr_xTi_1-x}O3 with Zr/Ti = 0.2/0.8 this structure is tetragonal
I used Norm pp. the lattice parameter not accurate
so I want to switch to PAW pp. when I ran found that the code comment is
================================================================================
-P-0000 leave_test : synchronization done...

chkinp: machine precision is 2.2204460492503131E-16

chkinp: Checking input parameters for consistency, jdtset= 1.

chkint: ERROR -
Context : the value of the variable pspcod(at is 7.
The value of the input variable ntypalch is 1, while it must be
equal to 0
Action : you should change the input variables ntypalch or pspcod(at.

chkinp: Checking input parameters for consistency, jdtset= 2.

chkint: ERROR -
Context : the value of the variable pspcod(at is 7.
The value of the input variable ntypalch is 1, while it must be
equal to 0
Action : you should change the input variables ntypalch or pspcod(at.
-P-0000
-P-0000 leave_new : decision taken to exit ...

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
so how I going please

Best
Mutta

[gonze]

Dear Mutta,

Alchemical mixing, as coded in ABINIT, cannot be used with PAW.

That's life ...
X

[Mutta]

Dear Prof. Gonze
I used norm-conserving to find optimized structure of Pb{Zr_0.2Ti_0.8}O3
the result not agree with exp. data .When I used PAW to test with PbTiO3
the result agree with exp. data. but PAW can not mixing pp.
Can you comment , suggestion to my group

Best
Mutta

[gonze]

Dear Mutta,

Alchemical mixing of pseudopotentials is a very primitive technique to try simulate
alloys. For certain pairs of pseudopotentials, the result is quite decent (even in case of heterovalent
species, which is quite surprising), but for some other pairs of pseudopotentials, the results
are deceiving. So, if you tried it for your case and you observed that it does not work,
it is better to forget about alchemical mixing.

You should simply switch to supercells in which you put real atoms of the two kinds,
and explore the different possibilities to place the Zr and Ti atoms in such a supercell ...

Best wishes,
X