Jump to content
RockWare Support Forum
Marek Svab

Transport modelling (sorption) - breakthrough curves

Recommended Posts

Dear GWB team, I would like to share with you our results of transprt modelling, which are hard for me to explain. I do not know if I do something wrongly or if there i some other mistake or problem.

I wanted to calculate the sorption filter 100x100x200 cm filled by sawdust. Sorbed substance is naphthalene. A attach script, compressed thermodataset (extended by naphthalene by us), dataset with langmuir data and two figures - outputs from calculation including my drawing. Thus, you can look directly at the results and if you would like to, also re-calculate our task.

Let describe the problem. Looking on the Figure 1 (concentration profile of naphthalene within the layer during the time): my opinion is that shape of the last two curves (right side) should be as I marked it by blue dots. Simply, the shape of the curves should not substantially change except its termination with the end of the layer. I see no reason why the curves should change their shape at the end of the layer, especially when boundary condition is set on "outlet" on the right side. Looking on the figure 2 (concentrations of naphthalene in various positions at the layer during the time): again the shape of the breakthrough curve (in the last node - last curve on right side) differs from shape of other curves. Why? It is obvious that the sheerer shape of the curve starts just at the end of the layer (it is clear also from the figure 1 - if we think about it). It seems, that there is other source of adsorbate at the end of layer, but boundary condition is "outlet" - it should be ok. Furthermore, the breakthrough curve (right curve in the figure 2) is "unreal". Its shape should be similar to for example first two curves in the same figure. I was playing a lot with this, but I can not explain it and I have no idea what to do with this problem. Finally, I found similar problem also in frame of other sorption calculations (advanced sorption modelling), I added these results because they are simplest for disclosing here. I would be very glad to have an advice what is wrong. I really think that shapes of curves are incorrect, but if you explained me if its correct, I will be glad. Thank you!

Marek Svab, Dekonta,a.s.

thermo.com.v8.r6__MS_KS_PRB_v2.zip

post-624-1215703633_thumb.jpg

post-624-1215703642_thumb.jpg

sorption_example_svab.x1t

Langmuir_piliny.zip

Share this post


Link to post
Share on other sites

Marek,

I believe you are observing the effect of saturating your column's sorption sites. I checked your example input and if you increase your exhange capacity (you can do this directly in the Sorbing Surface panel) to a higher value you can see how the curves assume the shapes you expected. Hope this helps.

Geoffrey Thyne

post-619-1217352341_thumb.jpg

Share this post


Link to post
Share on other sites
Marek,

I believe you are observing the effect of saturating your column's sorption sites. I checked your example input and if you increase your exhange capacity (you can do this directly in the Sorbing Surface panel) to a higher value you can see how the curves assume the shapes you expected. Hope this helps.

Geoffrey Thyne

Geoffrey, thank you for your response. However, I have to say I am not much sastisfied with your answer. Of course, If you increase exchange capacity, you delay breakthrough of sorbate through layer. It is nothing surprising and shape of the curves inside of layer before breakthrough is ok also in my contribution. Finally, you only showed me, how it can be delayed (this is what I was playing with a lot). In your picture, it is not breakthrough. I agree, it is ok.

But my question related to breakthrough situation, where the shape of the curves is not correct, I think. Please, read my first contribution carefully. Could you really try to explain me such behaviour of the breakthrough curves? I believe the situation is deeply described in first contribution. Many thanks. Also, could you, my be, give me contact on somebody who would be better for such type of problem?

Looking forward on your answer.

Have a nice day!

Regards,

Marek Svab

Share this post


Link to post
Share on other sites
Geoffrey, thank you for your response. However, I have to say I am not much sastisfied with your answer. Of course, If you increase exchange capacity, you delay breakthrough of sorbate through layer. It is nothing surprising and shape of the curves inside of layer before breakthrough is ok also in my contribution. Finally, you only showed me, how it can be delayed (this is what I was playing with a lot). In your picture, it is not breakthrough. I agree, it is ok.

But my question related to breakthrough situation, where the shape of the curves is not correct, I think. Please, read my first contribution carefully. Could you really try to explain me such behaviour of the breakthrough curves? I believe the situation is deeply described in first contribution. Many thanks. Also, could you, my be, give me contact on somebody who would be better for such type of problem?

Looking forward on your answer.

Have a nice day!

Regards,

Marek Svab

Share this post


Link to post
Share on other sites

Marek,

The shape of the curve is due to the dispersivity value when solving the ADE. The dispersivity value incorporates both diffusion and physical dispersion. The more vertical the curve, the smaller the dispersion, the more horizontal the curve, the greater the dispersion around the center of mass. The center of mass (center of the curve) is exactly the advective flow velocity. In plug flow you would have misible displacement and essentially a vertical front with the injected fluid displaying the initial fluid. There might be a slight spreading (the curve would not be perfectly vertical) due to diffusion. You might wish to try the simulation without any sorption first to confirm this.

In your example I believe the curves are inclined upward because the center of mass has passed through the column in the last two time steps you plotted. In effect your solute has accelerated near the column's end because there is no longer any retardation (sorptive capacity overwhelmed -you can verfiy this by plotting sorped fraction) and you are seeing plug flow with a slight reduction in inlet concentration due to mixing between inlet and initial fluids. If you kept running the inlet fluid through the column it would eventually completely displace the initial fluid and the solute concentration at the end of the column would be equal to the front of the column (new inlet fluid). It may help to think of the last two curves as the trailing edges of a plug flow example with very limited reduction in the solute concentration by sorption.

My solution was to increase the sorptive capacity to maintain the curve shape meaning the shape was a function purely of the dispersive term in the ADE. Of course, this also has the effect of slowing the solute transport since the retardation is now constant throughout your simulation time and may have confused the issue. It may be helpful for you to add a conservative tracer to your inlet solution so you can compare the transport of both and see the difference.

Sorry if I did not make this clear in my first reply. As to a better person to explain this to you - you can try Dr. Bethke or perhaps Dr Fetter, although I would review Fetter's Contaminant Hydrogeology text before asking him. You can also try an analytical solution to compare with the model.

Geoff

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×