Surprises in the theory of elasticity: pairwise atomic interactions, central forces, and Cauchy relations

Although the theory of elasticity has been well known for many years there are still some commonly accepted misconceptions that continue to be taught in solid-state physics courses. In this paper we review some of these misconceptions through a detailed study of the relationship between macroscopic and microscopic theories of elasticity in a simple-cubic crystal with one atom per primitive cell. The analysis is performed at different stages of complexity. In the first-neighbor approximation we find that the crystal is simply unstable. If pairwise interaction up to second neighbors is considered the crystal is stable but the elastic constants present additional symmetries known as Cauchy relations which are not observed experimentally. The reasons for this behavior are analyzed, pointing out that the invariance of the potential energy with respect to rotations implies necessarily that pairwise interactions must be central. Only if the atomic interactions involve the participation of 3-bodies or more, the forces are non-central and the resulting elastic tensor may be acceptable.