Hi all,
I am using CXTFIT to estimate parameters of metal transport through homogenous soil columns. I have started with an experiment and I already have several doubts.
My experiment is as follow:
A packed soil column where 1g of contaminated soil (1000 mg/g Cu) was added at the top. EDTA solution was applied from a Mariott system at 2 cm constant pounded level on top of soil surface. Volume of leachates was collected and mg/L of Cu measured.
The question are:
- Type of boundary conditions: Is Dirac delta input ?
- Mass of dirac input: Is it calculated as (1gx1000mg g-1)/(column area)?
- Concentration in dimensionless unit: can I used mg(t)/mg0 ?
- If I start fluxing the EDTA solution through a dry soil column, what boundary conditions should be used? Is it possible to divide the estimation in two parts: one before steady state and another after? If no, could you give me some hints?
Thank you a lot
cxtfit
Yes, Lola,
You can use the dirac delta function, as long as you specify the mass correctly. You can use dimensionless concentrations if sorption is linear. That is something you need to investigate (batch sorption experiments?), i.e., if the isotherm is linear or not.
Use a numerical solution (HYDRUS-1D?) when sorption is nonlinear. You may have to use numerics anyway if you pond and the soil is initially dry, as I gather from your message. Then it becomes a combined unsaturated flow and solute transport problem (use Hydrus-1D). This unless the time to saturate the column is relatively short and you are doing a long-time leaching experiment. But in that case you might as well bring the column first to saturation (e.g., slowly from below).
Hope this helps. Regards. --Rien van G.
You can use the dirac delta function, as long as you specify the mass correctly. You can use dimensionless concentrations if sorption is linear. That is something you need to investigate (batch sorption experiments?), i.e., if the isotherm is linear or not.
Use a numerical solution (HYDRUS-1D?) when sorption is nonlinear. You may have to use numerics anyway if you pond and the soil is initially dry, as I gather from your message. Then it becomes a combined unsaturated flow and solute transport problem (use Hydrus-1D). This unless the time to saturate the column is relatively short and you are doing a long-time leaching experiment. But in that case you might as well bring the column first to saturation (e.g., slowly from below).
Hope this helps. Regards. --Rien van G.