I am in the process of manufacturing a microfluidic device for use in research, and would like to do a CFD simulation of the interior of the device to see if an oxygen gradient is formed.

Device diagram

The idea is to run a CFD (computational fluid dynamics) simulation of gas mixtures in a small space (33mm x 15mm x 0.1mm) to see if there is an oxygen gradient formed prior to manufacturing the device.

The simulation space would have two inputs and two outputs, all are 1mm. There is also a repeating square structure in the simulation space [interior of the microfluidics device], although it is not totally necessary it would be a plus to have it in the simulation.

enter image description here

1 atm air and 1 atm of a pre-mixed oxygen/air mixture would be supplied into the device at a rate of 1ml/min.

  • What would be some CFD simulation software to use to accomplish this? I am proficient with Linux command line and basic scripting but also have access to a windows system.
  • Would the software be able to take into account different gasses e.g. if it was an air to nitrogen gradient that was to be formed?
  • Would it be able to simulate the interior of the device with the square structure?
  • Would increasing the height of the device modify the gradient if it is formed?
  • $\begingroup$ Have you chosen a CFD package? Have you tried example simulations - usually part of the help? Have you run sequential simulations changing only the height? $\endgroup$
    – Solar Mike
    May 31, 2021 at 6:20
  • $\begingroup$ caelinux.com/CMS3 is a nice complete package. It is free and built on Ubuntu 18.04. I've gone through the complete tutorial on the piston simulation at home, but while working at HP, I always had access to alternate packages on high end machines. They do have a fluids component. $\endgroup$
    – Jim Clark
    May 31, 2021 at 13:41
  • $\begingroup$ Is there any fluid in this? Anyway, if you want a gradient, you would need to precisely balance the air (or N2) and O2 ... Suggest modeling any imbalance, so you get an idea of how sensitive you are to that. A neat "1D" horizontal gradient might be hard to achieve, but not sure $\endgroup$
    – Pete W
    May 31, 2021 at 14:46
  • $\begingroup$ Both gasses and liquids are considered fluids. $\endgroup$
    – Jim Clark
    May 31, 2021 at 19:47
  • $\begingroup$ Yes I will use ANSYS + Autodesk Fusion 360 to export the part, it looks easiest to use and is free for students. There is no fluid in the device, however the gasses will be delivered vas via gas-syringes (en.wikipedia.org/wiki/Gas_syringe) and syringe pumps. $\endgroup$
    – mangoTango
    Jun 1, 2021 at 18:59

1 Answer 1

  • I've decided to go with ANSYS student edition to do the CFD simulation and Autodesk Fusion360 to model the part (microfluidics device), as the part is very simple just a box with two holes in it.

  • ANSYS student edition and Autodesk products are both free for student use. Autodesk also offers a CFD package but it doesn't include the course along with it, and as a non-engineer trying to do a relatively simple simulation I'll be going with what is better documented.

  • It looks like ANSYS put up a course on EdX on how to use their product with CFD and also FEA which is what I'll be using. Course is CornellX: ENGR2000X A Hands-on Introduction to Engineering Simulations

The one other question would then be if ANSYS has a PDMS (DOW SYLGARD™ 184 Silicone Elastomer Kit) material? If it were possible to do a FEA of a PDMS device to see stress points would be helpful to run before manufacturing the photolithography mask and wafer, although we have not had too many problems in this regard in the past.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.