I’m using Avogadro to gather information about the properties of a series of different compounds. For each one, I am optimising energy with the UFF force field (4 steps per update, steepest descent, stopping when dE = 0). After optimization, Avogadro reports an AutoOpt E = 601.221 kJ/mol. When I go to Extensions → Molecular Mechanics → Calculate Energy, it gives me a value of -441.066 kJ/mol.
Could someone please clarify why these two energy values differ, and which one would be more appropriate to use when comparing the relative stability or properties of different molecules?
First off, it sounds like you’re using Avogadro 1.2. We cannot support that release anymore. I highly suggest installing a recent release or nightly build from Install — Avogadro 1.102.1 documentation
The reason for your difference is clearly that you used UFF for the AutoOpt, but a different force field is being used for the Molecular Mechanics method. (I’d guess MMFF94)
Force fields are useful for cleaning up geometries and relative energy differences (e.g., chair vs. boat cyclohexane) are generally okay. But they are not reliable for “properties of a series of different compounds.”
If you want to calculate thermochemical properties, you want to use a quantum chemical method in a program like ORCA, Gaussian, GAMESS, etc.