Many students have reported numbers as they are given by the program, this is not sensible. Remember that the accuracty to which you report a number tells the reader the relative error level. For the methods you are using,
the energy will have an error of ≈ 10 kJ/mol, so how much is this in atomic units (which are hartree)?
the dipole moment will be accurate to about 2 decimal places, ie 0.01 Deby.
frequencies in wavenumbers are by convention reported with no decimal places and for the methods you are using in this course you need to know that there is a known systematic error of around 10%, this is because we use a harmonic approximation for vibrations that are actually anharmonic. Note that 10% is around 10cm-1 for vibrations in the order of several 100cm-1 but for very high energy vibrations like C-H stretches this can be as much as 300 cm-1 . (the situtation is actually more complex, and is covered in my fourth year course "Computational Inorganic Chemistry")
intensities are rounded to the nearest whole integer and in fact the accuracy is much less that this, but this is convention
bond distances are accurate to ≈ 0.01 Å
bond angles are accurate to ≈ 0.1°
important Have you reported values consistently and to the correct accuracy?
Energies
Many students have reported energies incorrectly, because the energy is HIGHLY dependent on the method, basis set and details of a calculation they total enery of molecules cannot be compared, this is a very important point!
the total energy for a system is reported only for the purpose of ensuring others can repeate your calculations, and hence reference to this energy.
you should report your total energies in atomic units, but make no attempt to interprete or compare them
relative energies are sensible things to calculate and compare. For example the energy difference between two isomers
report relative energies in kJ/mol
relative energies can only be computed for systems with exactly the same number and type of atoms
all components MUST be computed with exactly the same method/basis set/computational details (eg things like scf=tight)
relative energies for reactions can be reported but they MUST be balanced
for example to determine the energy of association of the acid base pair
NH3 + BH3 -> H3B:NH3 compute E(adduct)-[E(NH3)+E(BH3)]
