Multiwfn official website: http://sobereva.com/multiwfn. Multiwfn forum in Chinese: http://bbs.keinsci.com/wfn
You are not logged in.
Dear Saeed,
I have a relevant blog article http://sobereva.com/134. Briefly speaking, you need to use main function 5 of Multiwfn to respectively export cube file for the two NBOs, and then load them into VMD, and properly set graphical representation to show their isosurfaces.
Best regards,
Tian
This depends on the employed color scale. Usually there should be evident attractive vdW interaction, but may be marginally more repulsive than common attractive vdW interaction.
Dear Saeed,
You can simultaneously plot the corresponding two orbitals in VMD. See J. Comput. Chem., 40, 1488 (2019) for example (this paper directly plotted two NBOs in main function 0 of Multiwfn).
Best,
Tian
Dear Marcos,
Yes, it is already available in the precompiled binary version of Multiwfn.
Best,
Tian
em... I think some critical computational details were not explicitly mentioned in their work. Perhaps the best way is contacting the authors.
By the way, you may also try to use different SCF convergence algorithms, such as SCF=QC. According to my experience, SCF=QC has a higher probability than default algorithm to converge to a unstable wavefunction, perhaps the "unstable" wavefunction is what you actually need.
You may try constraint DFT (CDFT) via NWChem or CP2K. First performing geometry optimization with CDFT to constraint Fe to the expected atomic charge and spin population, and then use the wavefunction at final geometry step as initial guess for subsequent calculations without CDFT.
I have a blog article about CDFT: http://sobereva.com/271 (written in Chinese)
guess=only only perform guessing, while the orbitals may not be saved to .chk file. Perhaps you should use guess(only,save), and after that you can use guess=read to read information from it.
You can use the method described in Section "3.100.8 Generate Gaussian input file with initial guess combined from
fragment wavefunctions" of Multiwfn manual to generate initial guess. According to Table S2 of the paper, Fe in (III) and (IV) states have evidently different net charge and spin population, which may guide you to choose a proper state for calculation of isolated Fe atom.
This depends on how to use Multiwfn. If you follows "Skill 1: Accelerating calculation of TrEsp by making use of cubegen utility" in Section 4.A.9 of Multiwfn manual, then you can use .fch/fchk as input file of Multiwfn, as shown in the manual (this example utilized cubegen, but it is not absolutely necessary)
Dear Marcos,
I just updated Multiwfn today. A new option -45 has been added to basin analysis module. Description of this option in the updated manual:
-45 Export attractor information and cube file of present grid data: This option is available when attractors and basins have been generated. By this option you can export attractor and basin information as basinana.txt in current folder and export present grid data as basinana.cub in current folder. When you perform the same basin analysis next time, after choosing option 1 to try to generate basins, if the two files are found in current folder, Multiwfn will ask you if directly loading them; if you choose y, the attractor&basin information as well as grid data will be loaded, and hence the cost for regeneration of them is fully avoided.
Best regards,
Tian
Dear Marcos,
Simply printing the located attractors on screen doesn't notably slow down the calculation, the most time-consuming part is generating the attractors/basins.
Perhaps, what you need is an option, by which you can directly load the information of previously located attractors and attribution of grids to basins, so that you can avoid regenerating the attractors/basins in the second run? If yes, I can add it.
Best,
Tian
Dear Dr. Daniel Glossman-Mitnik,
Currently Multiwfn can only calculate hardness in terms of finite difference.
Best regards,
Tian Lu
In Gaussian, RESC can only represent scalar relativistic effect, SOC is not incorporated. In addition, relativistic Hamiltonian cannot be used in conjunction with pseudopotential.
If you want to incorporate SOC effect into DFT ground state calculation, you can consider to use spin-orbit DFT feature in NWChem. I have a blog article written in Chinese, you may use Google translator to understand the content: http://sobereva.com/368
Great!
I know noting about JANPA.
If you have Gaussian, you can freely use the NBO 3.1 embedded in it. In addition, licence of NBO 7.0 is merely 100 USD.
The two factors are independent of each other. Standard definition of MCBO is incompatible with diffuse functions, irrespective of the choice of theoretical method.
Please read Section 3.11.10 of Multiwfn manual, the theoretical background has been carefully introduced.
For a large ring, please only use multi-center bond order (alternatively, use AV1245), do not use multicenter DI (only supports up to 10 atoms). Your Multiwfn version is new enough.
Please do not use very old version. Due to my special algorithm for Multicenter Bond Index (see "Appendix: The extremely efficient implementation of MCBO in Multiwfn" in Section 3.11.2 of latest Multiwfn manual), current version of Multiwfn is able to calculate Multicenter Bond Index for arbitrarily large rings (even more than 100 atoms!)
Frankly speaking, I don't have any knowledge about TSCT and TBCT. The author Cheng Zhong had developed a script and shared in http://bbs.keinsci.com/thread-18597-1-1.html, which can realize this analysis. Please contact him via E-mail for more information about the script, or use Google translator to try to understand content of his post.
I don't know why you intend to use a very small delta value, perhaps you only want to consider the contribution from the "most" frontier MO? If yes, you can gradually decrease the delta value from the default value 0.1 a.u., each time you can choose "5 Print current orbital weights used in orbital-weighted (OW) calculation" to check the weights.
Taking examples\benzene.fch as example, if the delta is decreased to 0.01 a.u., you will see the following output after choosing option 5
10 Highest weights in orbital-weighted f+
Orbital 22 (LUMO ) Weight: 50.00 % E_diff: 3.410 eV
Orbital 23 (LUMO+1) Weight: 50.00 % E_diff: 3.410 eV
Orbital 24 (LUMO+2) Weight: 0.00 % E_diff: 5.656 eV
Orbital 25 (LUMO+3) Weight: 0.00 % E_diff: 7.147 eV
Orbital 26 (LUMO+4) Weight: 0.00 % E_diff: 7.147 eV
Orbital 27 (LUMO+5) Weight: 0.00 % E_diff: 7.789 eV
Orbital 28 (LUMO+6) Weight: 0.00 % E_diff: 8.202 eV
Orbital 29 (LUMO+7) Weight: 0.00 % E_diff: 8.202 eV
Orbital 30 (LUMO+8) Weight: 0.00 % E_diff: 8.389 eV
Total weight of above listed orbitals: 100.00 %
10 Highest weights in orbital-weighted f-
Orbital 21 (HOMO ) Weight: 50.00 % E_diff: -3.410 eV
Orbital 20 (HOMO-1) Weight: 50.00 % E_diff: -3.410 eV
Orbital 19 (HOMO-2) Weight: 0.00 % E_diff: -5.924 eV
Orbital 18 (HOMO-3) Weight: 0.00 % E_diff: -5.924 eV
Orbital 17 (HOMO-4) Weight: 0.00 % E_diff: -6.498 eV
Orbital 16 (HOMO-5) Weight: 0.00 % E_diff: -8.011 eV
Orbital 15 (HOMO-6) Weight: 0.00 % E_diff: -8.011 eV
Orbital 14 (HOMO-7) Weight: 0.00 % E_diff: -8.639 eV
Orbital 13 (HOMO-8) Weight: 0.00 % E_diff: -9.084 eV
Orbital 12 (HOMO-9) Weight: 0.00 % E_diff: -10.752 eV
Total weight of above listed orbitals: 100.00 %Clearly, now only the degenerate two HOMOs and the two LUMOs contribute to orbital weighted Fukui function or dual descriptor.
Best,
Tian
Hello,
I don't exactly know which data you are talking about, could you provide a screenshot?
Best,
Tian
Dear Asdrubal Lozada,
Very nice, I will mentioning your script in Multiwfn manual in the corresponding part of Section 2.7 of manual.
Because the patch may or may not work for other releases of Multiwfn source code, I think it is best to also upload the Multiwfn source code package (version 3.8(dev) Update [2024-Nov-13]) onto https://github.com/aslozada/Stress_tensor/, so that other people can always use your script successfully.
Best regards,
Tian
Dear Sedsil,
Just put the three files in current folder, boot up Multiwfn and load any one of them, then enter main function 22, choose "2 Calculate various quantitative indices" to calculate various indices including condensed Fukui function, or choose "3 Calculate grid data of Fukui function, dual descriptor and related functions" to visualize Fukui function.
Best,
Tian
Fractional occupation number weighted electron density (FOD) proposed by Grimme is an important and very convenient method to visually detect and quantify static correlation, this method has been supported by Multiwfn since the version updated on 2024-Nov-13. Please check Section 4.A.7 of new Multiwfn manual for introduction and how to perform the analysis.
Note only for isolated systems, but also for periodic systems this method can be used, molden file produced by CP2K should be used in the latter case.
Sir, as you previously mentioned, I followed 'Section 3.21.1' and 'Section 4.18.1' for plotting the Photo-Induced Electron Transfer (PET) analysis diagram. However, this procedure only provides a map and does not include the energy state transition figure as illustrated in the attached diagram. I would greatly appreciate it if you could specify which software is used to generate such diagrams.
Additionally, I have another query regarding the Charge Decomposition Analysis (CDA) diagram in Multiwfn. These diagrams often include energy levels indicated in blue. Is it possible to plot the CDA diagram without displaying these energy values?
https://i.postimg.cc/BXfpBBTq/Screenshot-2024-07-30-210852.jpg
https://i.postimg.cc/8sdGHw24/dislin.png
The energy level diagram was plotted manually by the authors (using such as powerpoint), not generated by Multiwfn.
Choose option "6 Disable labelling orbital indices" then replot.
Dear Saeed,
Unfortunately, it is not possible to directly derive the method and basis set used for generate the .wfn file.
Best,
Tian
Dear Camps,
When Multiwfn asks you to select a way to set up grid, you can choose option "8 Use grid setting of another cube file" and input path of an existing cube file, then the newly calculated grid data will share exactly the same grid setting with the existing cube file.
Best,
Tian
Multiwfn has been updated today. Stress tensor ellipticity has been added as the 118th user-defined function.
Dear Duc,
The atom indices in the two fragments are not contigouous. Please look at atom 198, it is in the large molecule, but it was set to fragment 2 (which should completely correspond to the small molecule).
In addition, implicit solvation model should not be used when performing sobEDA. The solvent contribution should be manually calculated, please check the paragraph "Solvation effect: influence of solvation effect on ..." in page 7025 of sobEDA original paper. In other words, sobEDA.sh can only be directly used to carry out energy decomposition analysis in gas phase.
Best,
Tian