No energy minimum lattice in some Cd and Hg alloys

Posted on : June 30, 2018 (Sat) 16:40:54

by Taewon Jin

Dear professors

I tried to calculate the Cd and Hg alloys with various volume to find optimized volume

Here is the input file I used:

c---------------------FeCo----------------------------------
go ./data/mag/FFFF-bcc-28-80-25-5.6
c------------------------------------------------------------
c brvtyp a c/a b/a alpha beta gamma
bcc 5.6, , , , , ,
c------------------------------------------------------------
c edelt ewidth reltyp sdftyp magtyp record
0.001 1.0 sra gga91 mag init
c------------------------------------------------------------
c outtyp bzqlty maxitr pmix
update 11 500 0.01
c------------------------------------------------------------
c ntyp
1
c------------------------------------------------------------
c type ncmp rmt field mxl anclr conc
FeCo 2 0 0.0 2
28 75
80 25
c------------------------------------------------------------
c natm
1
c------------------------------------------------------------
c atmicx(in the unit of a) atmtyp
0a 0b 0c FeCo
c------ The following types of inputs and their combination
c are also allowed.
c------ In those cases a, b, b mean primitive unit vectors
c and x, y, z mean shortest unit cell vectors along
c x, y, and z axses.
c 1/2 1/2 1/2 Fe
c 0a 0b 0c Fe
c 0x 0y 0z Fe
c 1/2a 1/2b 1/2c Fe
c------------------------------------------------------------


and here is the total energy output of various volume.

some alloys shows that there is no minimum data with large volume
(5.6 ~ 7.0 are lattice parameter of Ni0.75Hg0.25)

as volume increase, total energy decrease linearly and it seems there is no minimum data.

bcc-28-75-80-25-5.6: total energy= -12073.3683839
bcc-28-75-80-25-5.7: total energy= -12073.6325132
bcc-28-75-80-25-5.8: total energy= -12073.8746239
bcc-28-75-80-25-5.9: total energy= -12074.1209851
bcc-28-75-80-25-6.0: total energy= -12074.3473338
bcc-28-75-80-25-6.1: total energy= -12074.5518777
bcc-28-75-80-25-6.2: total energy= -12074.7496060
bcc-28-75-80-25-6.3: total energy= -12074.9459123
bcc-28-75-80-25-6.4: total energy= -12075.1374990
bcc-28-75-80-25-6.5: total energy= -12075.3348339
bcc-28-75-80-25-6.6: total energy= -12075.5210637
bcc-28-75-80-25-6.7: total energy= -12075.7155416
bcc-28-75-80-25-6.8: total energy= -12075.8961307
bcc-28-75-80-25-6.9: total energy= -12076.0913719
bcc-28-75-80-25-7.0: total energy= -12076.2649733

Some Cd alloy shows similar result.

Some Cd & Hg alloy originally not prefer alloying? or

Do I need to fix parameter to find energy minimum?

[Re:04] No energy minimum lattice in some Cd and Hg alloys

Posted on : July 02, 2018 (Mon) 20:40:01

by Hitoshi GOMI

Dear Taewon Jin,

Did you try this?
http://kkr.issp.u-tokyo.ac.jp/bbs/thread.php?id=191

[Re:03] No energy minimum lattice in some Cd and Hg alloys

Posted on : July 02, 2018 (Mon) 16:57:35

by Taewon Jin

Dear Hitoshi GOMI

Thank you for your advice

I solve the problem in most materials using ewidth=1.5 Ry

However, for Os alloy, their convergence total energy value is too different as volume increase, and it makes hard to find minimum volume.

Here is the input file


c---------------------FeCo----------------------------------
go ./data/mag/FFFF-bcc-28-76-25-5.6
c------------------------------------------------------------
c brvtyp a c/a b/a alpha beta gamma
bcc 5.6, , , , , ,
c------------------------------------------------------------
c edelt ewidth reltyp sdftyp magtyp record
0.005 1.3 sra gga91 mag init
c------------------------------------------------------------
c outtyp bzqlty maxitr pmix
update 11 500 0.01
c------------------------------------------------------------
c ntyp
1
c------------------------------------------------------------
c type ncmp rmt field mxl anclr conc
NiOs 2 0 0.0 2
28 75
76 25
c Vc 1 0.2 0.0 0 0 100
c------------------------------------------------------------
c natm
1
c------------------------------------------------------------
c atmicx(in the unit of a) atmtyp
0a 0b 0c NiOs
c 0.5x Vc
c------ The following types of inputs and their combination
c are also allowed.
c------ In those cases a, b, b mean primitive unit vectors
c and x, y, z mean shortest unit cell vectors along
c x, y, and z axses.
c 1/2 1/2 1/2 Fe
c 0a 0b 0c Fe
c 0x 0y 0z Fe
c 1/2a 1/2b 1/2c Fe
c------------------------------------------------------------

And here is the total energy of NiOs

bcc-28-75-76-25-5.6: total energy= -10909.3673844
bcc-28-75-76-25-5.7: total energy= -10909.3698674
bcc-28-75-76-25-5.8: total energy= -10909.3705111
bcc-28-75-76-25-5.9: total energy= -10913.5422117
bcc-28-75-76-25-6.0: total energy= -10913.5368512
bcc-28-75-76-25-6.1: total energy= -10913.5290111
bcc-28-75-76-25-6.2: total energy= -10913.5194895
bcc-28-75-76-25-6.3: total energy= -10913.5089888
bcc-28-75-76-25-6.4: total energy= -10913.4977273
bcc-28-75-76-25-6.5: total energy= -10913.4860305
bcc-28-75-76-25-6.6: total energy= -10913.4741285
bcc-28-75-76-25-6.7: total energy= -10913.4621590
bcc-28-75-76-25-6.8: total energy= -10913.4501856
bcc-28-75-76-25-6.9: total energy= -10913.4374223
bcc-28-75-76-25-7.0: total energy= -10913.4252915
bcc-28-75-76-25-7.1: total energy= -10913.4125691
bcc-28-75-76-25-7.2: total energy= -10913.3993411
bcc-28-75-76-25-7.3: total energy= -10913.2400327
bcc-28-75-76-25-7.4: total energy= -10913.2410568
bcc-28-75-76-25-7.5: total energy= -10913.2355950


Sincerely
Taewon Jin

[Re:02] No energy minimum lattice in some Cd and Hg alloys

Posted on : July 01, 2018 (Sun) 02:22:56

by Hitoshi GOMI

Dear Taewon Jin,

I got the energy minimum for Ni0.75Hg0.25 with ewidth=1.5 Ry.
Maybe ewidth=1.0 is too small.

P.S. You had better to use the exact chemical formula, at least in this BBS, to avoid confusions. Initially, you wrote "the Cd and Hg alloys", therefore I (and possibly administrator) think that you want to calculate the Cd-Hg alloys. But in you input, you use "FeCo" for the name of a site. Furthermore, the atomic numbers indicate NiHg...

5.6 -12093.500777588
5.7 -12093.511469699
5.8 -12093.516626013
5.9 -12093.520453600
6.0 -12093.519590076
6.1 -12093.517866629
6.2 -12093.512589122
6.3 -12093.506225466
6.4 -12093.499628281
6.5 -12093.490923955
6.6 -12093.484210413
6.7 -12093.474396396
6.8 -12093.467190239
6.9 -12093.459533942
7.0 -12093.453083208


c---------------------NiHg----------------------------------
go data/bccNiHg
c------------------------------------------------------------
c brvtyp a c/a b/a alpha beta gamma
bcc 5.6, , , , , ,
c------------------------------------------------------------
c edelt ewidth reltyp sdftyp magtyp record
0.001 1.5 sra gga91 mag 2nd
c------------------------------------------------------------
c outtyp bzqlty maxitr pmix
update 4 500 0.01
c------------------------------------------------------------
c ntyp
1
c------------------------------------------------------------
c type ncmp rmt field mxl anclr conc
NiHg 2 0 0.0 2
28 75
80 25
c------------------------------------------------------------
c natm
1
c------------------------------------------------------------
c atmicx(in the unit of a) atmtyp
0 0 0 NiHg
c------------------------------------------------------------

[Re:01] No energy minimum lattice in some Cd and Hg alloys

Posted on : June 30, 2018 (Sat) 17:41:17

by Administrator

Dear Taewon Jin,

are you trying to calculate NiHg?
The atomic number 28 is nickel.

Best regards,

KKR administrator