Page 2 of 2

Re: Bader Charge analysis

Posted: Thu Dec 12, 2019 8:38 pm
by graeme
Not nessecarily. You can check convergence of the atomic charge values in the ACF.dat with respect to the grid density.

Re: Bader Charge analysis

Posted: Thu Dec 12, 2019 9:37 pm
by Chaure
Dear Professor,

When i issued bader -b weight CHGCAR -ref CHGCAR_sum
I got the value of charge on ACF.dat file as follows. As you said -b weight method is more accurate.

Now let's calculate charge transfer on each atom. We have the formula to calculate charge transfer as:
Charge transfer or Net bader charge or Oxidation state = ZVAL - CHARGE(on ACF.dat file)
so, for atom:
1 : 12-12.271232 = -0.271232(For fisrt Zn)
2: 12- 12.322490 = -0.32249 (For 2nd Zn), and so on
.
.
.
9: 5 - 4.728565 = 0.271435 (For 1st Sb atom)
10: 5 - 4.695027 = 0.304973 (For 2nd Sb atom), and so on
.
.
.
However , the reported literature value of charge transfer for Zn and Sb in this compound are +0.27 and -0.27. But we are getting just opposite sign of it.

Please let me know, where i did mistake.
Sincerely








# X Y Z CHARGE MIN DIST ATOMIC VOL
--------------------------------------------------------------------------------
1 2.886798 0.832153 7.178721 12.271232 1.046396 24.450991
2 3.396977 6.987807 1.046296 12.322490 1.046396 24.718661
3 0.255090 6.987807 3.066212 12.312711 1.075365 24.510344
4 6.028685 0.832153 5.158805 12.257772 1.075365 24.252951
5 3.396977 4.742133 5.158805 12.324135 1.075365 24.640583
6 2.886798 3.077827 3.066212 12.277368 1.075365 24.423437
7 6.028685 3.077827 1.046296 12.279998 1.046396 24.630925
8 0.255090 4.742133 7.178721 12.312083 1.046396 24.676707
9 0.890713 0.648736 0.899899 4.728565 1.259563 26.217818
10 5.393062 7.171224 7.325118 4.695027 1.187734 26.023708
11 2.251175 7.171224 5.012408 4.703118 1.190693 26.056856
12 4.032600 0.648736 3.212609 4.731395 1.212093 26.091965
13 5.393062 4.558716 3.212609 4.694148 1.095826 25.884409
14 0.890713 3.261244 5.012408 4.715281 1.098770 25.891255
15 4.032600 3.261244 7.325118 4.695485 1.259563 25.833837
16 2.251175 4.558716 0.899899 4.679194 1.187734 25.863586
--------------------------------------------------------------------------------
VACUUM CHARGE: 0.0000
VACUUM VOLUME: 0.0000
NUMBER OF ELECTRONS: 136.0000

Re: Bader Charge analysis

Posted: Thu Dec 12, 2019 9:57 pm
by graeme
I don't know about the literature values. I don't see any problems with your calculation.

Re: Bader Charge analysis

Posted: Fri Dec 13, 2019 4:25 am
by skm
graeme wrote:
> If you are analyzing the CHGCAR file, the sum of the valence charge will
> always be correct. Increasing the fft grid density can be used to check
> convergence of the Bader charges on each atom.

As I am recently started doing Bader analysis, I think i cant understand fully, unless you give a sequence of steps, we need to do with description for 'What is called convergence and what is not", or kindly give a link if its already explained clearly in this forum elsewhere. I have already seen the Main Bader web page, and its not clear from there.

1. What do you mean by " analysing CHGCAR only?
2. I always use the command "bader CHGCAR -ref CHGCAR_sum" and check the ACF total electrons with that from OUTCAR.

KINDLY clear the doubt on What is meant by CONVERGENCE in this case.

Regards

Re: Bader Charge analysis

Posted: Fri Dec 13, 2019 4:35 am
by skm
Chaure wrote:
> Dear Professor,
>
> When i issued bader -b weight CHGCAR -ref CHGCAR_sum
> I got the value of charge on ACF.dat file as follows. As you said -b weight
> method is more accurate.
>
> Now let's calculate charge transfer on each atom. We have the formula to
> calculate charge transfer as:
> Charge transfer or Net bader charge or Oxidation state = ZVAL - CHARGE(on
> ACF.dat file)
> so, for atom:
> 1 : 12-12.271232 = -0.271232(For fisrt Zn)
> 2: 12- 12.322490 = -0.32249 (For 2nd Zn), and so on
> .
> .
> .
> 9: 5 - 4.728565 = 0.271435 (For 1st Sb atom)
> 10: 5 - 4.695027 = 0.304973 (For 2nd Sb atom), and so on
> .
> .
> .
> However , the reported literature value of charge transfer for Zn and Sb in
> this compound are +0.27 and -0.27. But we are getting just opposite sign of
> it.
>
> Please let me know, where i did mistake.
> Sincerely
>
>
>
>
>
>
>
>
> # X Y Z CHARGE MIN DIST ATOMIC
> VOL
>
> --------------------------------------------------------------------------------
> 1 2.886798 0.832153 7.178721 12.271232 1.046396
> 24.450991
> 2 3.396977 6.987807 1.046296 12.322490 1.046396
> 24.718661
> 3 0.255090 6.987807 3.066212 12.312711 1.075365
> 24.510344
> 4 6.028685 0.832153 5.158805 12.257772 1.075365
> 24.252951
> 5 3.396977 4.742133 5.158805 12.324135 1.075365
> 24.640583
> 6 2.886798 3.077827 3.066212 12.277368 1.075365
> 24.423437
> 7 6.028685 3.077827 1.046296 12.279998 1.046396
> 24.630925
> 8 0.255090 4.742133 7.178721 12.312083 1.046396
> 24.676707
> 9 0.890713 0.648736 0.899899 4.728565 1.259563
> 26.217818
> 10 5.393062 7.171224 7.325118 4.695027 1.187734
> 26.023708
> 11 2.251175 7.171224 5.012408 4.703118 1.190693
> 26.056856
> 12 4.032600 0.648736 3.212609 4.731395 1.212093
> 26.091965
> 13 5.393062 4.558716 3.212609 4.694148 1.095826
> 25.884409
> 14 0.890713 3.261244 5.012408 4.715281 1.098770
> 25.891255
> 15 4.032600 3.261244 7.325118 4.695485 1.259563
> 25.833837
> 16 2.251175 4.558716 0.899899 4.679194 1.187734
> 25.863586
>
> --------------------------------------------------------------------------------
> VACUUM CHARGE: 0.0000
> VACUUM VOLUME: 0.0000
> NUMBER OF ELECTRONS: 136.0000


1. Can you give the Literature Reference?
2. If you getting same numerical value but the opposite sign, they might have deducted the other way i.e. instead of [valence minus bader charge] they did opposite way, i think. why dont you contact the author?

Regards

Re: Bader Charge analysis

Posted: Mon Dec 16, 2019 10:23 pm
by Chaure
Dear Professor,
Is it also necessary to get total(Valence+core)electrons equal by doing

bader CHGCAR_sum

only in ACF.dat and sum of all (valence + core) electrons on all the atoms. At beginning i get confused with this step that i tried to match number of electrons in this way and keep on increasing FFT grid until i get total electrons equal but never achieved.
Sincerely,

Re: Bader Charge analysis

Posted: Mon Dec 16, 2019 10:39 pm
by graeme
No, you do not have to converge the total charge - that is near impossible given the cusps at the atomic cores. Just look for convergence of the Bader charges.

Re: Bader Charge analysis

Posted: Tue Dec 17, 2019 12:27 am
by Chaure
Dear Professor,
Thank you so much for making me clear on bader charge analysis .
Sincerely

Re: Bader Charge analysis

Posted: Tue Dec 17, 2019 1:07 am
by Chaure
Dear Professor,
I have two different values of bader charge while issuing two commands on the same compound:
1. bader -b weight CHGCAR

# X Y Z CHARGE MIN DIST ATOMIC VOL
--------------------------------------------------------------------------------
1 2.886798 0.832153 7.178721 11.776657 1.123510 18.910567
2 3.396977 6.987807 1.046296 11.776603 1.123510 18.910293
3 0.255090 6.987807 3.066212 11.776598 1.123510 18.909196
4 6.028685 0.832153 5.158805 11.776543 1.123510 18.908922
5 3.396977 4.742133 5.158805 11.776536 1.123510 18.908648
6 2.886798 3.077827 3.066212 11.776591 1.123510 18.908922
7 6.028685 3.077827 1.046296 11.776603 1.123510 18.910293
8 0.255090 4.742133 7.178721 11.776657 1.123510 18.910567
9 0.890713 0.648736 0.899899 5.219155 1.302479 31.589058
10 5.393062 7.171224 7.325118 5.227640 1.332531 31.632639
11 2.251175 7.171224 5.012408 5.219159 1.302479 31.589881
12 4.032600 0.648736 3.212609 5.227644 1.332531 31.633461
13 5.393062 4.558716 3.212609 5.227651 1.332531 31.633736
14 0.890713 3.261244 5.012408 5.219166 1.302479 31.590155
15 4.032600 3.261244 7.325118 5.227640 1.332531 31.632639
16 2.251175 4.558716 0.899899 5.219155 1.302479 31.589058
--------------------------------------------------------------------------------
VACUUM CHARGE: 0.0000
VACUUM VOLUME: 0.0000
NUMBER OF ELECTRONS: 136.0000

2. bader -b weight CHGCAR -ref CHGCAR_sum

# X Y Z CHARGE MIN DIST ATOMIC VOL
--------------------------------------------------------------------------------
1 2.886798 0.832153 7.178721 12.271232 1.046396 24.450991
2 3.396977 6.987807 1.046296 12.322490 1.046396 24.718661
3 0.255090 6.987807 3.066212 12.312711 1.075365 24.510344
4 6.028685 0.832153 5.158805 12.257772 1.075365 24.252951
5 3.396977 4.742133 5.158805 12.324135 1.075365 24.640583
6 2.886798 3.077827 3.066212 12.277368 1.075365 24.423437
7 6.028685 3.077827 1.046296 12.279998 1.046396 24.630925
8 0.255090 4.742133 7.178721 12.312083 1.046396 24.676707
9 0.890713 0.648736 0.899899 4.728565 1.259563 26.217818
10 5.393062 7.171224 7.325118 4.695027 1.187734 26.023708
11 2.251175 7.171224 5.012408 4.703118 1.190693 26.056856
12 4.032600 0.648736 3.212609 4.731395 1.212093 26.091965
13 5.393062 4.558716 3.212609 4.694148 1.095826 25.884409
14 0.890713 3.261244 5.012408 4.715281 1.098770 25.891255
15 4.032600 3.261244 7.325118 4.695485 1.259563 25.833837
16 2.251175 4.558716 0.899899 4.679194 1.187734 25.863586
--------------------------------------------------------------------------------
VACUUM CHARGE: 0.0000
VACUUM VOLUME: 0.0000
NUMBER OF ELECTRONS: 136.0000

Where ZVAL for atoms 1, 2,...8 is 12
and ZVAL for atoms 9, 10, ....16 is 5
which one should i choose for accuracy.
Probably this may be my last question.

Sincerely

Re: Bader Charge analysis

Posted: Tue Dec 17, 2019 2:30 am
by graeme
use the later calculation

Re: Bader Charge analysis

Posted: Wed Dec 18, 2019 1:26 am
by skm
Dr. Graeme,

How are the following analysis commands vary computationally?

$ bader CHGCAR -ref CHGCAR_sum

and

$ bader -b weight CHGCAR -ref CHGCAR_sum

Regards

Re: Bader Charge analysis

Posted: Wed Dec 18, 2019 1:36 am
by Chaure
Dear Professor,

This is what striking me for a long time. As you see I have three different values of bader charges for three different commands for the same compound and for the outfiles of same calculation.
1. bader -b weight CHGCAR
2. bader CHGCAR -ref CHGCAR_sum
3. bader -b weight CHGCAR -ref CHGCAR_sum
Sincerely

Re: Bader Charge analysis

Posted: Wed Dec 18, 2019 2:21 am
by graeme
Ignore the results of the first command.

For the second two, you should see convergence with respect to the grid density, with the -b weight method converging faster than the regular method.