The use of energy and co-energy for the evaluation of forces in non-linear, anisotropic dielectric solids

In this paper, in spite of the wide diffusion of the method that makes use of suitable variations of the electric energy (or co-energy) for computing forces in non-linear, anisotropic dielectric solids, it is shown that the arguments generally developed to justify this approach are, in reality, unsound. A rigorous proof of the correctness of the method is provided. It is shown also that the method is frequently used in cases where it fails; to this end, the limitations of the method itself are pointed out. Received 13 December 1999