I am given to understand that the energy is stored in the molecule in the form of changes to bond angles. With water, the bond angle between the hydrogen atoms changes by a degree or so during melting.
Here, the bond angle is described as well as the dipole moment of torque on being rotated by an electromagnetic wave. Fun fact: Apparently the moment is calculated using the Shrodinger equation. Whatever that is:
But the problem is that the rotation of the molecule is described as being not so different than typical oscillation which shows up as sensible heat.
So what's the deal? Is the rapid rotation generating enough centrifugal force to deform the bond angle? I have a hard time with this. It seems more likely that the energy of heat is absorbed directly by the molecule's bonds, thus changing the angles and resulting in no sensible heat or oscillation.