Need MATLAB program for 2DFATMIC code for 1D reactant solute transport with distance dépend flow and dispersion and a pulse injection

I would like to applied the 2DFATMIC code to one-dimensional advection-dispersion transport with variable velocity flow and dispersion coefficient. In my problem (single-well push-test applying to PCE reactant solute) the inlet boundary condition is assumed to be a pulse type (first type), the contamination is assumed to be injected until t = 5 days. I want to obtain the concentration distribution during the injection phase and after. But i don't have the 2DFATMIC code. And i wanted to use this code un MATLAB. Thanks !!! I have the following transport équation:

R*dC/dt = d/dx*(D(x)*dC/dx-v(x)*C)-mu*C The velocity is express like this v(x) = 30/(2*pi*10*0.1*x); thé dispersion coefficient is equal to D(x)= 0.1*v(x)

mu is the first order reaction coefficient (mu = 0.075 day^(-1)) Initial condition C(x,t=0) = 0

Boundaries conditions

C(x=0,t) = 1 , 0=<t<t0 and 0 , t> t0 where t0 = 5 days dC(x=L, t) = 0

L is the length of the domain L = 10 m

If it is not possible to obtain the 2DFMTMIC code i wanted as well as possible the cranck Nicholson code or other schème code.

[Torsten]

Use "pdepe".

[Thomas TJOCK-MBAGA]

I used it but the résultats are very differents to my analytical reusults and those present in the litterature. If possible help me in that way

[Torsten]

If you include the pdepe code, I can try to compare with your PDE from above.

I don't think that the software is responsible for differences between your analytical results and those obtained by using pdepe because pdepe should be suited to solve your problem. Maybe some settings are not correct.

As far as I can see, 2DFATMIC is FORTRAN code available here:

https://www.epa.gov/water-research/two-dimensional-subsurface-flow-fate-and-transport-microbes-and-chemicals-2dfatmic

But I don't understand why you want to use 2d-code for your case.

[Thomas TJOCK-MBAGA]

Ok,i will do it soon

[Thomas TJOCK-MBAGA]

I wanted to use it because several authors in thé past and the date used it for such problem

[Torsten]

If you want: try it. The MINGW compiler should be able to compile the FORTRAN code provided in the link.

[Thomas TJOCK-MBAGA]

Here is the code but I couldn't introduce the pulse type condition. I instead put a continuous entry condition, but the curves do not give with those of the analytical results. And you you can help me introduce this condition I will be very proud of it

[Thomas TJOCK-MBAGA]

Here is the code but I couldn't introduce the pulse type condition. I instead put a continuous entry condition, but the curves do not give with those of the analytical results. And you you can help me introduce this condition I will be very proud of it

[Torsten]

I get

pr = 10*ur
qr = (2*pi*10*0.1*x)/3

if you want to set

dc/dx = 0

at x=L.

[Thomas TJOCK-MBAGA]

I run the code with this outlet boundary condition but i obtain the same distribution as before. I dont know why

[Torsten]

Then this is the solution to your equation. I don't see other errors in your code.