Corrigendum to ‘Pore shape evolution by solution transfer: thermodynamics and mechanics’

The chemical potential of a component of the solid in solution is given by the equilibrium condition between the stressed solid and its solution. This condition was first established by Gibbs (1877) for a plane interface and then generalized to any curved interface. It was rederived later by Lehner & Bataille (1985) and Mullins & Sekerka (1985). Gibbs (1877) chose to have the solid and the fluid phases enclosed within rigid walls, thus preventing any exchange of mechanical work between the system (solid + fluid) and the environment. However, in the experience described in our reference paper, samples are subjected to constant stress conditions, which induce a deformation of the samples and thus an exchange of work with the environment. One may wonder if Gibbs’ classical equation is still valid under these conditions which are the usual conditions for creep of rocks by pressure solution. We tried to give an answer to this question by using an approach similar to the Griffith’s crack approach. Our derivation led us to the conclusion that Gibbs’ equation had to be modified to take into account the deformability of the solid. We thus proposed to add an elastic energy term to the classical Gibbs’ equation. The question arose as to whether or not Gibbs’ and Griffith’s approaches are compatible. In this corrigendum we want to correct our previous derivation by showing that it contained some omissions leading to an incorrect conclusion. In the reference paper, we proposed to write the variation A U of the internal energy of the system (solid+fluid) as follows (equation 9 of the reference paper): A U = AQ + AW = AU, + AU, + (uf us) 6n (9)