Microfluidics “Frogger” Is A Game Changer For DIY Biology


See those blue and green dots in the GIF? Those aren’t pixels on an LCD display. Those are actual drops of liquid moving across a special PCB. The fact that the droplets are being manipulated to play a microfluidics game of “Frogger” only makes OpenDrop v 2.0 even cooler.

Lab biology is mainly an exercise in liquid handling – transferring a little of solution X into some of solution Y with a pipette. Manual pipetting is tedious, error prone, and very low throughput, but automated liquid handling workstations run into the hundreds of thousands of dollars. This makes [Urs Gaudenz]’s “OpenDrop” microfluidics project a potential game changer for the nascent biohacking movement by offering cheap and easy desktop liquid handling.

Details are scarce on the OpenDrop website as to exactly how this works, but diving into the literature cited reveals that the pads on the PCB are driven to high voltages to attract the droplets. The PCB itself is covered with a hydrophobic film – Saran wrap that has been treated with either peanut oil or Rain-X. Moving the droplets is a simple matter of controlling which pads are charged. Splitting drops is possible, as is combining them – witness the “frog” getting run over by the blue car.

There is a lot of cool work being done in microfluidics, and we’re looking forward to see what comes out of this open effort. We’ve covered other open source efforts in microfluidics before, but this one seems so approachable that it’s sure to capture someone’s imagination.

Thanks to [Thomas] for the tip.

30 thoughts on “Microfluidics “Frogger” Is A Game Changer For DIY Biology

  1. There is also a device called “Voltrax” by Oxford Nanopore Technologies (https://nanoporetech.com/products/voltrax). But there the drops move inside a glas enclosure, and they use additives to make the drops move better. The primary use is to prepare samples for their USB drive sized DNA Sequencer MinION, and its big brother, the PromethION.

    Considering OpenDrop, it’s pretty cool. I wonder how practical it actually is. For example the coating seems to take a bit of work and a source for problems. In order to save hands-on time, the device would need to be used for multiple samples.

    Now the million dollar question: If multiple drops cross the same cell at different times, how much material to they transfer? In certain applications, especially PCR or microbiology, a tiny amount of transfer would ruin the whole workflow.

    1. I suppose you’d need to give different paths for drops that aren’t meant to combine. You wouldn’t have to use a pure grid layout for the electrodes, and since PCB etching isn’t difficult, you could have a specific PCB for each job. It says “cartridge 1” on the system in the picture, so it looks like he’s thought of that.

      Are there home PCB printers? Something that prints out a resist, either on a PCB, or transfers it, then drops it into an etching tank? Actually that should be no problem for chemists and biologists, etching is basically chemistry.

      This is a great idea, cheap for teaching students, or a little more complex for actual experiments and developing new drugs etc.

      1. If you have to build something for every different workflow, then the flexibility of this approach isn’t necessary.

        Creating PCBs at home or at a hackerspace is entirely possible. Messy, but possible. There is PCB etching using toner transfer or photosensitive materials, and also CNCs can be coopted to mill them.

        For students this is useless, I think, because they will learn more when they actually have to use the usual manual tools. Probably easier to learn, for most of them, too.

        1. Not for every workflow. But if some need particular paths to remain uncontaminated, then a bit of custom or semi-custom design might help, where some needs overtake others. Might turn out that a few flexible designs for PCBs will cover 90% of cases.

          But for education, where certain experiments are done over and over again each year, a library of particular boards might be nice.

          Point is, it only needs a bit of PCB etching, so it’s not a great burden, if circumstances require it. Which means this device is more or less unlimited. You can choose which limitations to put up with for the freedoms you require.

        1. The problem is that cleaning or even replacing the surface layer is manual labor and cuts into the benefits of automation.

          The cleaner may be problematic on its own, for example acid if you need a particular Ph level, or detergents with biological material.

          1. I think this is a toy for experiments and education. If you actually want full-scale production you’d get out the glassware and pumps, maybe consider building a few reactor towers. This isn’t meant to be fully automated. But it can get students thinking about how they’d design a full-scale automated system, what stages would be needed, what pipelines and pumps and mixers. Having to replace a bit of cling-film hardly matters.

            Even for on-site production of particular chemicals, it’d be used in small batches. Automation is only needed when you’re doing large volume.

    1. It’s both traditional and obligatory to port Tetris to any new machine with a low-res display. Using different coloured water reservoirs will give a nice effect when they mix at the bottom.

  2. « but automated liquid handling workstations run into the hundreds of thousands of dollars. »

    Hey ! You guys forgot about OpenTrons ! https://opentrons.com/
    I think you covered them in an article before even.

    It’s Open-Source, it’s only like $3k, and there’s even a Smoothieboard in it :)

    This new tech is really awesome though, I want one.

        1. I’ve often thought conceiving children ought to be difficult and painful. As compensation I’d make childbirth easier. Sex, I’d keep the same. Just divorce sex from reproduction, as a default. So only people who really wanted to, would go through with it. Maybe even make it slightly life-threatening, for both parents.

      1. My DIYbio experiments have turned out pretty well thus far, surprisingly labor intensive on the initiation time though, but after a thorough incubation and a little more manual labor, have produced quite impressive results, even when taken from a non-biased standpoint.
        Although, after introduction of Sample 2, Sample 1 appears to be reverting to a previous state. Further external environmental adjustment doesn’t seem to rectify or even affect the issue…
        Further research is required…

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.