Microfluidic Work
Random musings on my experience with microfluidic projects
I have spent several years micromachining microfluidic devices for customers, both in an educational setting as well as commercial setting. I have done extensive work in developing microfluidic devices for my research on microbubbles.
All this was done using a variety of micromachining tools, including UV and IR lasers and Focused Ion Beam Micromachining. I learned a lot from this experience, both from a prototyping as well as a production viewpoint.
Although I had a variety of micromachining tools to carry out my work, the strange mix of available material and process compatibility meant that I had to stick to a few select processes.
I am putting some thoughts down in the hopes that it would help others who are just setting up microfluidic test-stations. Reader beware: A lot of this relates to 2 phase flows (gas-liquid), and the critical dimensions were usually in the 10 micron range, unlike mainstream commercial microfluidics where the CDs are much higher. Your mileage may vary.
Some Background
A spanking new Institute for Micromanufacturing was coming up at our university, and several graduate courses in Microfluidiics were offered. None of them prepared me for the real-world microfluidic situations. It is just not the device engineering aspects that I had to master, I also needed to be extra friendly to the machinists, Buyer, and vendors!
Fluidic Interconections
Even before I could start my experiments, I faced acute and often depressing difficulties in interconnects and line-in line-outs. I had issues with the test-bed trough, fluid inlets, plexiglass joinings, high pressure gas delivery fittings (compression and Viton O-ring) fittings, miniature QDCs, epoxies with varying drying times, UV curing epoxies.
Over time, there was an evolution of my design, implementation, fixturing, optics, lighting etc. As the test station matured, I ended up developing protocols for my projects- and it became easier to setup test-beds. These protocols become specific to my particular needs. The next set of experimental runs became a whole lot easier.
I should hasten to add that fabricating non-standard parts was not a smart thing to do. It took me several years to come to that conclusion! For example, I no more employ non-standard thread sizes. Instead, I change the design so that I can go for the next (standard) size up or down. A thing to remember: Packaging and interconnect might become trickier than the fluidic part.
Microscopy
Often, in microscopy setups for flow visualization, even ELWD objective lenses might just not cut it. Space constraints will have to solved by refining the size and shape of test components and fixturing.
Lighting can become an issue. Good illumination can be used advantageously. I would suggest using both transmitted as well as reflected light, with a sub-stage condenser and iris, but forget about getting perfect Koehler illumination! For most microfluidic work, metallurgical microscopes with ELWD lenses are the preferred way to go.
Fouling of objective lens: A microscope glass slide held in place with rubber bands on the objective lens seem to work best. Anytime the field view gets bad, push the glass slide just a bit. The trade off is reduction in resolution.
When I was a student just beginning my grad program, I thought a 100x lens was the best thing for my work. A professor in the Biology department had tried explaining to me that a 40x objective would suffice for my work, but I chose to not understand his logical reasoning. I did eventually understand the essence of it all: Field of View, Magnification, Working Distance, Depth of focus- all are important factors when working in a real-world microfluidic experiments- not just magnification. Too much magnification can be a bad thing.
Ergonomics
I sit for several hours observing microbubbles. Instead of suffering through the discomfort of badly designed tables and chairs, I designed three of my own a few years back. Thick wooden (3×3 inch) legs, 1 inch plywood with one inch thick granite tops that were mounted using rubber adhesives. No vibration, perfect rigidity, and I could spend hours without breaking my back. Who would have thought furniture design would be part of experiments in two phase microfluidic investigation ???
So, what is the best chair/table design for a video microscopy system that requires extensive and continuous use by one person? Well you will have to pay me to give you the design details, but let me throw you a bone here- it mainly depends on two factors: 1) The distance between your heel and top of your upper thigh, and, 2) The height of the eyepiece of your trinocular microscope. If you want to know more, get in touch with me! You’ll thank me some day!
Feature size versus microfabrication system mass
The smaller the feature or phenomenon you want to observe, the bigger (and more massive) the fabrication and/or visualization system, both mass-wise and thermal-mass wise. And I am talking about features less than 5 or 10 microns, which I love.