Electrostatic Potential map is showing the wrong colors

I’m having an issue with Avogadro that I’m unsure how to fix when typing to make electrostatic potential maps.

**Please see this dropbox for photos:

Environment Information

Avogadro version: 1.2.0
Operating system and version: Mac OS Sonoma 14.2.1

Expected Behavior

  • I want the electrostatic potential map to look like the first photo in the Dropbox, with the red indicating more electron density and the blue indicating less.

Actual Behavior

  • However, when I try the electrostatic potential map I end up with the other photo where it is just all one color.

Steps to Reproduce

  • I start with my structure then go to “extensions”, “create surfaces”, surface type: “van der waals”, color by: “electrostatic potential”, and iso type: “0”.

I’m not quite sure why you’re seeing all blue, but the molecule itself seems a bit weird. It looks like:

(i.e., a substituted porphyrin with benzoic acid substituents) … but your graphic shows four oxygens with extra hydrogens, e.g. \ce{C(=O)OH2+} which seems highly unusual.

For example, version 1.98 shows this electrostatic potential for the protonated form I have above:

Note that the \ce{CO2H} groups all have negative charges and there’s a general white region around the aromatic porphryn.

In short, I guess I’d want to know:

  • what’s the actual molecular structure you’re trying to map (e.g., can you post a link to the molecule itself?)
  • why you think there should be red color around the porphryn (e.g., why the desired graphic is correct)


The molecule I am trying to recreate is this: link to molecule image

I am somewhat new to making organic compounds like this so excuse the oxygens on the end, they shouldn’t have the hydrogens on them.

I am also not sure what I am looking for in the electrostatic potential map but it would seem like there should be some areas of more electron density.

The electrostatic potential map is a reflection of positive and negative partial charge across the molecule. (It’s the difference between the electron density and the positive charge of the protons in each atom.)

For a neutral molecule, there will likely be some regions of more positive and more negative charge.

The first molecule you posted had four “extra” hydrogens on the oxygens. (The molecule structure you just linked had neutral \ce{CO2H} groups off each “arm.”

So the first molecule has a total molecular charge of +4 … so there’s a lot less electron density than a neutral molecule and you’ll see positive charges.

The molecule image you just posted has no hydrogens on those oxygens, so each has negative charge and a -4 overall molecular charge. There will be a lot more electron density and probably everything is red.

But neither looks to be “right” based on the molecular structure you say is your target. That has neutral molecular charge (as in the images I posted) and some regions of positive charge (e.g., H atoms) and some regions of negative charge (around the oxygen atoms).

Now in solution, some of those carboxylic acids will probably deprotonate, but I doubt it will be all four of them.