Potential energy curve of diatomics
In this exercise we will do potential energy surface scans of diatomics. Outputfiles can be opened in Chemcraft to visualize the geometries. A table labelled "The Calculated Surface using the 'Actual Energy'", near the bottom of the ORCA outputfile will contain the scan parameter (in Angstrom) and Energies.
Inputfile template:
! HF-3c pal8 noautostart
%paras
R= 0.3,2.0,20
end
* xyz 0 1
X 0 0 0
X 0 0 {R}
*
Dihydrogen
1. Run the scan for the H2 molecule in a singlet state.
2. Run the scan for the H2 molecule in the triplet state. Here you are actually calculating an excited state wavefunction for the H2 molecule. Excited states are easy to calculate when the spin multiplicity is different (special methodology required for excited states of the same spin multiplicity, such as CIS, TDHF, TDDFT etc.).
3. Do a geometry optimization of the H2 molecule in the singlet state.
4. Do a geometry optimization of the H2 molecule in the triplet state.
Plot both potential energy curves on 1 plot using an absolute energy scale. Compare the potential energy curves. You can change the scan parameters to different values if you prefer.
Questions:
Why does the energy rise for short bond distances? Why does it rise for long bond distances?
Why is the triplet curve so different?
If one would experimentally take an H2 molecule and somehow excite it into the triplet state what would happen based on your potential energy curves?
Dinitrogen
1. Change the scan parameters and use this inputfile instead.
! HF-3c pal8 noautostart
%paras
R= 0.6,2.0,30
end
* xyz 0 1
N 0 0 0
N 0 0 {R}
*
2. Run the scan for the N2 molecule in a singlet state.
3. Run the scan for the N2 molecule in the triplet state. Here you are actually calculating an excited state wavefunction for the N2 molecule. Excited states are easy to calculate when the spin multiplicity is different.
Plot both potential energy curves on 1 plot using an absolute energy scale. Compare the potential energy curves.
Questions:
Why is the triplet curve for N2 closer in shape to the singlet N2 curve than for H2?
Why would it not matter whether you do an unrelaxed scan or a relaxed scan for a diatomic molecule?
Note:
Potential energy curves of diatomic molecules like H2 and N2 can actually be difficult to get completely right as correlation effects (that is not included in Hartree-Fock) play in important role in correctly breaking up the chemical bond at long bond distances. One can even see weird discontinuities in the energy. We will talk more about this later.