A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

saeed_E · 2023-04-10 10:52:48

Dear Tian,

***Please simply ignore all my emails regarding CDA and only reply to this forum, if possible. Some questions have been forgotten to be included in emails while all them are included here. Many thanks to the editable character of the forum posts.****

1- For a CDA, I have generated Gaussian input files regarding single-point calculations (to generate .fch files) as follows (the complex geometry was optimized at MP2(full)/aug-cc-pVTZ):

For complex:
#PBE1PBE/cc-pVTZ nosymm pop=full iop(3/33=1)
You have recommended to avoid using diffuse functions, hence, the "aug" prefix is removed.

For both fragments in the complex geometry:
#PBE1PBE/cc-pVTZ nosymm pop=full

Please let me know if the employed methodology is quite correct.

2- To avoid any problem and have complete consistency with manual and the order of printed results, I always define electron-donor (Lewis base) as fragment 1 and electron-acceptor (Lewis acid) as fragment 2. What does a negative d or b value mean in CDA and how should such negative value be chemically interpreted? Is it true to state that a negative donation (d) or a negative back donation (b) is quite meaningless (and should be ignored) even corresponding value to be considerable (e.g., d=-0.2358 or b= -0.2847)?

For a complex namely HCN...GaF3, the complex geometry was firstly optimized at MP2(full)/aug-cc-pVTZ. Then, Single-point Gaussian files were generated and I exactly used the methodology described in step 1 to do so. The CDA results are as follows:

Orb. Occ. d b d - b r
1 2.000000 -0.000000 -0.000001 0.000001 -0.000001
2 2.000000 -0.000042 0.000303 -0.000345 0.000169
3 2.000000 -0.000000 -0.000005 0.000005 -0.000003
4 2.000000 -0.000000 -0.000000 0.000000 -0.000000
5 2.000000 -0.000000 -0.000000 0.000000 -0.000000
6 2.000000 -0.000000 0.000000 -0.000000 0.000000
7 2.000000 -0.000000 -0.000000 0.000000 0.000001
8 2.000000 -0.000000 -0.000000 0.000000 0.000001
9 2.000000 0.000399 -0.000001 0.000400 0.000038
10 2.000000 -0.000066 -0.000001 -0.000065 -0.000002
11 2.000000 -0.000026 -0.001229 0.001203 -0.000745
12 2.000000 -0.000015 -0.000548 0.000533 -0.000386
13 2.000000 -0.000000 -0.000043 0.000043 -0.000004
14 2.000000 -0.000000 -0.000043 0.000043 -0.000004
15 2.000000 0.000075 -0.002470 0.002545 0.001618
16 2.000000 -0.000001 -0.000016 0.000016 0.000111
17 2.000000 -0.000001 -0.000016 0.000015 0.000110
18 2.000000 0.004235 0.000056 0.004180 0.007580
19 2.000000 0.000961 -0.001541 0.002502 0.016795
20 2.000000 0.000017 0.000495 -0.000477 0.001110
21 2.000000 0.000017 0.000495 -0.000477 0.001111
22 2.000000 0.000003 0.000016 -0.000013 0.000076
23 2.000000 0.000003 0.000016 -0.000013 0.000076
24 2.000000 0.012701 -0.000284 0.012984 0.004926
25 2.000000 0.135396 -0.024979 0.160375 0.181747
26 2.000000 0.008959 0.000255 0.008704 0.022258
27 2.000000 0.008962 0.000255 0.008707 0.022242
28 2.000000 0.044012 0.007505 0.036507 -0.153321
29 2.000000 0.001341 0.002769 -0.001428 -0.019889
30 2.000000 0.001339 0.002769 -0.001430 -0.019884
31 2.000000 0.029733 -0.006341 0.036074 -0.200001
32 2.000000 0.000130 0.001295 -0.001165 -0.008365
33 2.000000 0.000132 0.001296 -0.001163 -0.008322
34 2.000000 -0.000482 -0.000715 0.000233 -0.004575
35 2.000000 -0.000484 -0.000716 0.000232 -0.004605
36 2.000000 -0.000000 0.000000 -0.000000 -0.000000
37 0.000000 0.000000 0.000000 0.000000 0.000000
38 0.000000 0.000000 0.000000 0.000000 0.000000
39 0.000000 0.000000 0.000000 0.000000 0.000000
40 0.000000 0.000000 0.000000 0.000000 0.000000
......
-------------------------------------------------------------------
Sum: 72.000000 0.247300 -0.021423 0.268723 -0.160136

The net electrons obtained by frag. 2 = CT( 1-> 2) - CT( 2-> 1) = 0.3777

Occupation number of orbital 25 of the complex: 2.00000000
Orbital 4 of fragment 1, Occ: 2.00000 Contribution: 2.42 %
Orbital 5 of fragment 1, Occ: 2.00000 Contribution: 64.61 %
Orbital 18 of fragment 2, Occ: 2.00000 Contribution: 1.09 %
Orbital 21 of fragment 2, Occ: 2.00000 Contribution: 19.40 %
Orbital 24 of fragment 2, Occ: 2.00000 Contribution: 3.45 %
Orbital 30 of fragment 2, Occ: 0.00000 Contribution: 7.24 %
Sum of values shown above: 98.21 %

Occupation number of orbital 25 of the complex: 2.00000000
FragA Orb(Occ.) FragB Orb(Occ.) d b d - b r
3( 2.0000) 21( 2.0000) 0.000000 0.000000 0.000000 -0.008233
3( 2.0000) 30( 0.0000) -0.011730 0.000000 -0.011730 0.000000
4( 2.0000) 21( 2.0000) 0.000000 0.000000 0.000000 -0.008355
5( 2.0000) 18( 2.0000) 0.000000 0.000000 0.000000 -0.025181
5( 2.0000) 21( 2.0000) 0.000000 0.000000 0.000000 0.175760
5( 2.0000) 24( 2.0000) 0.000000 0.000000 0.000000 0.058671
5( 2.0000) 30( 0.0000) 0.117119 0.000000 0.117119 0.000000
5( 2.0000) 31( 0.0000) 0.013532 0.000000 0.013532 0.000000
5( 2.0000) 38( 0.0000) 0.011786 0.000000 0.011786 0.000000
5( 2.0000) 52( 0.0000) 0.007743 0.000000 0.007743 0.000000
14( 0.0000) 21( 2.0000) 0.000000 -0.010944 0.010944 0.000000
16( 0.0000) 21( 2.0000) 0.000000 -0.005755 0.005755 0.000000

As can be seen, in the MO 25 of complex, the primary source of donor-acceptor happens during which an electron value by d= 0.135 is donated from fragment 1(HCN, Lewis base) to fragment 2 (GaF3, Lewis acid) but, at the same time, a back-donation by b= -0.0250 e happens from LA to LB in MO 25 of complex. But this back-donation is negative and made me confused chemically. Why this value is negative and how should be explained? The sum value of b is also negative which is not understandable and explainable; why?
There are some cases in which, EVEN, the value of a given donation (d) is also negative!

3- Within a CDA, orbital interaction diagram can also be plotted. In this diagram, two fragment orbitals any of which has a contribution of 10% or more build an orbital of complex by a red link. Is there any way to reduce this threshold value so that FO orbitals with a contribution less than 10% (e.g., 5%) are also linked to a complex orbital? Indeed, in a case, a complex orbital is constructed by 80% of FO on one fragment and 5.5% of FO of other fragment but the latter does not show a red link to the complex orbital due to threshold issue.

Sincerely yours,
Saeed

sobereva · 2023-04-11 09:25:12

Dear Saeed,

1 correct

2 Just ignore the very small negative value, and simply view it as zero. CDA algorithm is not perfect, non-negative d and b terms cannot be guaranteed.

3 See Multiwfn manual:

By default, all FOs and complex MOs are plotted, and if the contribution of a FO of fragment A or B to a complex MO is larger than 10% then they will be connected. For large systems, usually there are too many bars and linking lines in the diagram, and it is hence difficult to identify the orbital interaction mode based on the diagram. In these cases, you should properly use the option "4 Set the rule for connecting and drawing orbital bars" to manually set up the rule for connecting and plotting the orbital bars. See the prompt shown on the screen on how to use this option.

Clearly you can customize the rule of plotting linking lines.

Best,

Tian

saeed_E · 2023-04-11 11:24:59

Dear Tian,

Many many thanks for your very kind attention and so informative comments.

You have stated that the very small negative value of b or d should be ignored (should be viewed as zero). Now, the question is that what value should be taken into account as very small negative value for d or b. In other words, we can claim that:

If d equals or less than (-X) and b equals or less than (-Y), then, these values should be ignored and simply viewed as zero. Please provide a quite reasonable value for X and Y so that I have a threshold at hand by which a reasonable and logical decision could be adopted when I am going to ignore a given d or b value. Should values such as d(or b)= -0.2584, -0.6589, -0.4583, -0.8925 be taken as very small negative and ignorable values? If no, how should their negative sign be interpreted? What about for values such as d(or b)= -0.0258, -0.0358, -0.0681, -0.0852?

Once again, too many thanks for your highly valuable time and energy.

Sincerely,
Saeed

Last edited by saeed_E (2023-04-11 14:09:02)

sobereva · 2023-04-12 03:28:28

Dear Saeed,

There is no definitive threshold. In your present instance, |b| is less than |d| by one order of magnitude, so the small negative b never notably affects the analysis result, and thus you can safely ignore it.

Best,

Tian

saeed_E · 2023-04-12 04:32:15

Dear Tian,
Too many thanks for your kind attention and very nice comments.
To make this discussion much more obvious, please let's consider data presented for the Triel-bonded complex HCN...GaF3 at the first message. As can be seen, in this complex, the primary donor-acceptor interaction is played through MO 25 for which d=0.135396 indicating that there is a slight electron donation by 0.135396 electrons from HCN to GaF3. At the same time, a back-donation by b= -0.024979 takes place from GaF3 to HCN. Since this b value is negative and is smaller than corresponding d by one order of magnitude, we can safely ignore that meaning that back-donation through MO 25 does not significant role within this complexation process. In addition, the other b values with positive sign are highly smaller than can be taken into account. Thus, in this complexation, back-donation never plays a significant effect. Please let me know whether you are quite agree with my statement.

Now, please consider a case for which the d and b value for the primary interaction to be 0.4256 and -0.2789, respectively. Please let me know your analysis about this negative (but not small) b value which is also in the same order of magnitude with d.

Sincerely,
Saeed

sobereva · 2023-04-12 06:51:31

Dear Saeed,

1 It is correct

2 If b really equals to -0.2789, the CDA analysis must be problematic and the result cannot be accepted. For example, diffuse functions were employed.

Best,

Tian

saeed_E · 2023-04-12 07:45:04

Dear Tian,

Many Many thanks for your highly kindness to prompt reply with so nice comments.
The given b value is only an example, not a true value observed in a CDA. Consequently, based on your comments, the b value should always be within a few hundredths. If so, then its sign is never important because it should be ignored in any case (due to its small value). Consequently, talking about b is always meaningless.

Please let me ask in other way: when a b value is important and should be discussed? Could you please mention an example (even a hypothetical one)?

Once again, too many thanks for your valuable time and highly informative comments.

Sincerely yours,
Saeed

sobereva · 2023-04-12 07:57:42

Dear Saeed,

For example, it is well-known that in some transition-metal complex, there are donation effect and back-donation effect between ligand and metal atom. In this case, both d and b are important in analyzing the nature of the interaction.

Best,

Tian

saeed_E · 2023-04-12 15:31:12

Dear Tian,
Thank you very very much, my nice friend.
Is it true to state any given b value MUST be: 1-positive 2- not very small to be accepted?

Sincerely,
Saeed

Last edited by saeed_E (2023-04-12 16:34:17)

saeed_E · 2023-04-12 16:36:17

Dear Tian,
Thank you very very much, my nice friend.
Is it true to state any given b value MUST be: 1-positive 2- not very small to be accepted?

Sincerely,
Saeed

sobereva · 2023-04-12 17:05:02

Dear Tian,

Yes

Sincerely,
Saeed

saeed_E · 2023-04-12 17:25:36

Dear Tian,

Your kind attention to prompt reply with highly valuable confirmation is extremely appreciated.

Sincerely yours,
Saeed

Multiwfn forum

#12023-04-10 10:52:48

A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#22023-04-11 09:25:12

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#32023-04-11 11:24:59

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#42023-04-12 03:28:28

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#52023-04-12 04:32:15

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#62023-04-12 06:51:31

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#72023-04-12 07:45:04

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#82023-04-12 07:57:42

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#92023-04-12 15:31:12

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#102023-04-12 16:36:17

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#112023-04-12 17:05:02

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

#122023-04-12 17:25:36

Re: A problem regarding CDA in Multiwfn (an urgest and a crucial ask)

Board footer