Fick+Reaction Solver
Online simulation of motion of chloride ions through a concrete sample. The numerical solution of Fickian diffusion and binding according to Freundlich isotherm.
Online simulation of transport of chloride ion through a concrete sample in contact with chloride solution. Fickian diffusion and binding according to Freundlich isotherm is assumed inside the sample. Method of Lines with uniform grid is used to transform partial differential equations (PDEs) to ordinary differential equations (ODEs). Resulting set of ODEs is then solved using explicit Euler's method.
Time-dependent diffusion-reaction model - basic equations
Diffusion and binding
After including the information about porisity of concrete and chloride binding in solid phase, the mass ballance equation takes the form:
The reaction term, `r` is calculated using Freundlich isotherm:
The density of the solid phase in concrete, `\rho_s` is calculated using the weighted sum:
Initial and boundary conditions
We assume the constant concentrations at the boundaries (Dirichlet boundary conditions):
At the beginning of the process, there is no chlorides in the sample:
The total concentration of chlorides
The total concentration of chlorides (rescaled to `kg`/`kg` of concrete) is a sum of free and bound chlorides:
Diffusion-reaction model - Numerical solution
Numerical solution of the above equations is based on the Method of Lines (MoL) and on the explicit Euler method.
Method of Lines
We divide the considered segment into `n+1` sub-segments, each having the length `h`, to obtain the discrete points `x_0` to `x_{n+1}`. Let's then define `c_i \equiv c(x_i)` and `C_{b,i} \equiv C_b(x_i)`. The second derivative can be now substituted with the finite difference. That leads to the following set of equations:
Eulers Method
If `c_{i}` and `C_{b,i}` denote the concentrations at time `t_j`, while `\overline c_{i}` and `\overline C_{b,i}` denote the chloride concentrations at time `t_{j+1}` then:
After inserting the time derivative, one obtains: