Hi Andrei, To be a ferroelectric, the material must have at least two different states of polarization, so first you would need to identify those different stable structural configurations. In typical ferroelectric perovskites (like BaTiO3 or PbTiO3) the different polarization states are related by symmetry and can be trivially identified, but I guess that's not your case. Once you find the different stable configurations, a switching path would be a continuous structural transformation that brings the crystal from one state to the other, ideally through the lowest energy barrier (note that this is not how real ferroelectrics actually switch, this is how they would switch if they did it as a whole, infinitely symmetric, single crystal). To get the change in P during switching though, you don't need to use the ideal switching path, just a reasonable one. One thing that you do have to take care of is the possible jump between polarization branches along the path.
Depending on your familiarity with modern theory of polarization and ferroelectrics, the paper I mentioned in my first email is acctually a good starting point (at the end, for instance, it discusses the problem of the different polarization branches and the switching path). A more detailed study you might find useful is PRB 84, 115107 (2011). Best regards, Pablo
