Thank you for your answer. Well? To solve the problem, the idea is to solve, first, the thermo-electrical problem by considering the seawater as a solid. Then, the temperature at each node of the seawater is transferred as initial condition to solve the fluid problem. And, finally and again, the seawater temperature after solving the fluid problem is transferred as initial condition to the solid problem. nSome time ago, I tried to use workbench to solve it using FSI, but I failed to transfer volumetric temperatures, i.e. seawater temperatures, from a physics to other. I only got to transfer temperatures at the surfaces. As far as I know, before it could be done using FLOTRAN, but currently I don't know how to do it using APDL, and I haven't found a way to do it using workbench either.nn