i don't think this is basically possible to do because we even don't know all cyclic subgroups of no formula is known for the maximum order of elemens of however, there's a result due

to Landau that if is the maximum order of elements of then it's not surprising that there's no closed form for because is related to the partition function.