Bader and voronoi analysis discrepancy

Bader charge density analysis

Moderator: moderators

Post Reply
mykdia
Posts: 5
Joined: Sun Aug 02, 2020 2:19 am

Bader and voronoi analysis discrepancy

Post by mykdia »

Hi,

I am trying to extract partial charges for an isolated electrically neutral fixed molecule in various states with the gaussian cubes of size 701*501*501 pts. I tried to compare the default bader and voronoi analysis and while the electron density count is accurate and same for both cases (375.4897), the partial charges significantly deviate. For example, the Bader partial charge (ACF.dat) on an sp2 carbon atom (#63) is -4

# X Y Z CHARGE MIN DIST ATOMIC VOL
63 -14.356371 0.576009 -0.086491 10.035034 1.205927 1824.423711

but for Voronoi analysis (voronoi_analysis_2.dat) the charge is close to -2

# X Y Z CHARGE ATOMIC VOL
63 -14.3564 0.5760 -0.0865 8.2390 206.6259

Is there any way to fine tune the results for Bader analysis or some way to avoid this issue? The cube file size seems sufficiently large and for the ground state cube file I get the same partial charge values +/-0.5 au for both analysis. I have attached the files and an xyz file for the coordinates.
Attachments
t_0030.zip
Bader and Voronoi data+ XYZ coordinates
(8.18 KiB) Downloaded 3744 times
graeme
Site Admin
Posts: 2289
Joined: Tue Apr 26, 2005 4:25 am
Contact:

Re: Bader and voronoi analysis discrepancy

Post by graeme »

The Voronoi charges are not to be taken seriously. They just provide a sanity check on the charge distribution in a system.

You should rely on the Bader partitioning for your charge analysis.
mykdia
Posts: 5
Joined: Sun Aug 02, 2020 2:19 am

Re: Bader and voronoi analysis discrepancy

Post by mykdia »

Hi Prof. Graeme,

Thanks for your reply. Then how should I modify the Bader partitioning procedure? In this case Bader analysis is providing a significantly deviant result for the partial charges. For example for the ground state:

# X Y Z CHARGE MIN DIST ATOMIC VOL
---------------------------------------------------------------------------------------------
1 14.107481 1.575699 -0.242380 6.027102 1.181888 986.427993
2 12.971614 0.707368 -2.427548 5.553176 0.894044 576.205849
3 12.964707 1.429778 2.101149 5.567943 0.897286 504.657490
4 10.525832 -0.180885 -2.585597 8.171197 1.516140 1316.059640
5 14.278843 0.393874 -4.791294 6.072730 1.155209 933.577730
6 10.518731 0.631409 2.520666 8.180869 1.515300 1465.306890
7 14.265313 1.866911 4.448823 6.078544 1.176348 1079.977427
8 10.308359 -1.074625 -5.022495 5.540484 0.900934 468.650577
9 12.639400 -0.770844 -6.388833 6.087980 1.180326 846.297705
10 16.216294 0.958499 -5.126838 0.928643 0.648450 2282.948001
11 10.295388 0.535771 5.114062 5.534987 0.934177 466.196259
12 12.622215 1.252104 6.324628 6.076040 1.167589 786.391239
13 16.201443 2.510573 4.597427 0.926839 0.638957 2770.002234
14 8.091449 -1.938065 -6.097394 6.062736 1.183374 739.359170
15 5.390370 -0.993342 -2.449399 8.179173 1.556749 577.441592
16 5.384865 -0.191796 2.631321 8.176003 1.578977 744.016446
17 12.926279 -1.341292 -8.332329 0.923127 0.643444 805.661049
18 8.077135 0.040623 6.398446 6.042119 1.189210 589.060163
19 12.904183 1.312041 8.349896 0.921399 0.660855 583.539973
20 5.773597 -1.784733 -4.904771 5.537232 0.880391 370.350255
21 8.140010 -2.592353 -8.040081 0.927808 0.641181 417.929743
22 2.793572 -0.883196 -2.187588 5.564343 0.932261 74.561433
23 5.762680 -0.189749 5.211889 5.534982 0.926929 426.449079
24 2.788918 -0.177555 2.342880 5.561659 0.936921 74.367083
25 8.121919 0.017522 8.448315 0.945939 0.701655 292.121760
26 3.410479 -2.199870 -6.177006 6.078614 1.167766 556.186496
27 1.557973 -1.575355 -4.515547 6.054076 1.169917 206.036340
28 1.506610 -0.390225 0.054405 5.978978 1.206977 61.653673
29 3.396835 -0.206199 6.545164 6.071164 1.180508 519.911704
30 1.547740 -0.130897 4.768274 6.086654 1.199819 208.498755
31 3.232747 -2.832686 -8.114175 0.921964 0.635120 615.035698
32 -0.465101 -1.617632 -4.786750 0.924900 0.649085 287.622232
33 3.214745 -0.215655 8.582680 0.929105 0.677961 460.545207
34 -0.476116 -0.098933 5.034988 0.897604 0.638477 278.671116
35 7.954361 -0.182835 0.029077 26.653945 1.601926 2717.023700
36 -1.332130 -0.235346 0.030221 6.017892 1.202139 64.889925
37 -2.415960 2.165826 -0.316967 5.547405 0.905026 77.534017
38 -2.710334 -2.484198 0.356969 5.548769 0.934374 68.624686
39 -4.964293 2.676372 -0.388184 8.227594 1.556882 448.143081
40 -1.002426 4.485148 -0.654816 6.074488 1.163496 225.966616
41 -5.303641 -2.673889 0.384461 8.216760 1.510399 389.523213
42 -1.602622 -4.960708 0.716558 6.074064 1.181415 180.254184
43 -5.195146 5.238201 -0.758668 5.542272 0.929889 312.248977
44 -2.723297 6.378647 -0.926461 6.082105 1.179497 488.147776
45 1.035373 4.624915 -0.678242 0.894706 0.630760 358.130170
46 -5.857871 -5.186871 0.748619 5.544047 0.898343 309.919692
47 -3.550470 -6.625296 0.957679 6.073079 1.187518 507.520915
48 0.401276 -5.352702 0.772611 0.904431 0.644833 389.024381
49 -7.468738 6.556641 -0.947189 6.072019 1.190848 694.918111
50 -10.434649 3.031187 -0.438851 8.297217 1.565372 723.013358
51 -10.776068 -2.346705 0.334019 8.248774 1.536329 718.236881
52 -2.369178 8.373221 -1.215591 0.924899 0.636906 1237.292472
53 -8.280868 -6.212308 0.895159 6.053407 1.203991 699.852336
54 -3.452184 -8.648136 1.250933 0.939244 0.660354 1328.855973
55 -9.897417 5.542871 -0.799925 5.498478 0.858793 325.537987
56 -7.331133 8.584088 -1.239275 0.933522 0.648805 901.433524
57 -13.022690 2.826301 -0.409024 5.497132 0.875009 314.418896
58 -10.561675 -4.905150 0.703753 5.515150 0.899671 327.036102
59 -13.317721 -1.821972 0.257657 5.525886 0.922963 313.203560
60 -8.401852 -8.240662 1.188477 0.939329 0.671023 927.463477
61 -12.215754 6.970220 -1.004019 6.075382 1.181655 806.185146
62 -14.151544 5.287181 -0.761798 6.065485 1.180134 855.965325
63 -14.356371 0.576009 -0.086491 6.054571 1.205927 763.224063
64 -13.042577 -6.033064 0.864759 6.081808 1.164083 831.610022
65 -14.749610 -4.122968 0.588412 6.074409 1.192295 863.561174
66 -12.325794 8.992883 -1.294423 0.926925 0.671026 2019.020605
67 -16.163620 5.656800 -0.814258 0.934256 0.650518 3289.361659
68 -16.405271 0.703479 -0.104551 0.951671 0.693891 1172.538268
69 -13.408560 -8.025855 1.151879 0.930938 0.647732 2645.124216
70 -16.792644 -4.239589 0.604081 0.932864 0.676329 3037.347923
71 16.048186 2.229733 -0.343528 0.937042 0.650540 1398.209148
72 -7.873056 0.172492 -0.027341 28.611587 1.765004 891.629535
--------------------------------------------------------------------------------
VACUUM CHARGE: 0.0000
VACUUM VOLUME: 0.0000
NUMBER OF ELECTRONS: 375.4897

C63 electron count is 6.054571.
graeme
Site Admin
Posts: 2289
Joined: Tue Apr 26, 2005 4:25 am
Contact:

Re: Bader and voronoi analysis discrepancy

Post by graeme »

You will have to tell me more about what you mean by 'deviant'. What is it in the analysis that you recognize as being wrong?

I don't know enough about your calculation to know what makes sense. If you have included all of the electrons for C, then a Bader charge of ~6 would give a neutral atom, which is reasonable. If you included only the valence electrons, then I agree that a net charge of -2 is strange. But anyway, I just don't have sufficient information to understand if there is a problem with the calculation or the Bader analysis.

It would help to know where the cube file came from, and also why the total number of electrons is not an integer.
mykdia
Posts: 5
Joined: Sun Aug 02, 2020 2:19 am

Re: Bader and voronoi analysis discrepancy

Post by mykdia »

By deviant I mean the charge assigned to the C atom. I am dealing with a time dependent valence excitation and relaxation process in the molecule and I assumed for this process I would expect the bader charge to vary by +/-2 for the C atom, i.e, equal or less than the oxidation number. So when I got 10 a.u. bader charge on the C atom, I was a bit suspicious of the result.

The ground state cube file was generated directly using NWChem using all-electron CAMB3lYP calculation. The time dependent process was tracked using NWChem MOs with the same grid size. I cross checked the total electron count which remains same for the entire TD process. I had read elsewhere that the electron integration for the cube file is highly dependent on the number of grid pts, hence I had to define a highly dense 3D grid to get the total electron count (=376) within +/-0.5 au for the molecule.
graeme
Site Admin
Posts: 2289
Joined: Tue Apr 26, 2005 4:25 am
Contact:

Re: Bader and voronoi analysis discrepancy

Post by graeme »

From the data you sent for the ground state, the charge of 6.054571 on C 63 looks reasonable.

I'm not sure how much to trust Bader charges for a highly excited electronic state - I just don't have any experience with that.
mykdia
Posts: 5
Joined: Sun Aug 02, 2020 2:19 am

Re: Bader and voronoi analysis discrepancy

Post by mykdia »

Then, is it possible to somehow use constraints to force charge values within a certain reasonable range? In my case, the partial charges themselves are not crucial, as long as they can represent the potential due to the changing electronic density.
Post Reply