Discussion Forum

Hello,
I am trying to model a microfluidic gradient generator, like this:

www.elveflow.com/wp-content/gallery/common/gradient-on-a-microfluidic-chip-a.png

While it was not difficult to put together the stationary study using the laminar flow and transport of diluted species modules, what I would like to do is add an additional time-dependent step where after a few seconds the flow from both inlets is stopped, and the existing gradient that was generated by flow is allowed to decay inside the channel.

What would be the best way to do this? I tried using a stationary study and then a time-dependent study with the stationary study results as input, but was not able to get it to work (no concentration profile generated--the object was just solid red).

I also tried setting a global definition of a time-dependent variable with this expression:

(1-(1/(1+(2.71828182846^((-2)*100000*(t-30))))))[1/s]

To give a very steep step function that goes from 1 to 0 at 30 seconds.

I then multiplied the flow rates for both inlets by this variable, so that after 30 seconds the flow rate went from whatever it was defined as to 0. However, I had a lot of issues with the number of Newtonian iterations being too high, and was unable to fix this after playing around with it quite a bit (usually getting an error message of "Repeated error test failures, may have reached a singularity). Are there simpler ways of doing this?

Thanks for any suggestions!