The Countdown to Maker Faire Bay Area is On!


Children’s toy inspires a cheap, easy production method for high-tech diagnostic chips

Children’s toy inspires a cheap, easy production method for high-tech diagnostic chips

 Files 32463 0909-Khine-B X600
A children’s toy inspires a cheap, easy production method for high-tech diagnostic chips (Michelle Khine tR35 winner)… Microfluidic chips cost more than $100,000 –

Racking her brain for a quick-and-dirty way to make microfluidic devices, Khine remembered her favorite childhood toy: Shrinky Dinks, large sheets of thin plastic that can be colored with paint or ink and then shrunk in a hot oven. “I thought if I could print out the [designs] at a certain resolution and then make them shrink, I could make channels the right size for micro­fluidics,” she says.

To test her idea, she whipped up a channel design in AutoCAD, printed it out on Shrinky Dink material using a laser printer, and stuck the result in a toaster oven. As the plastic shrank, the ink particles on its surface clumped together, forming tiny ridges. That was exactly the effect Khine wanted. When she poured a flexible polymer known as PDMS onto the surface of the cooled Shrinky Dink, the ink ridges created tiny channels in the surface of the polymer as it hardened. She pulled the PDMS away from the Shrinky Dink mold, and voilà: a finished microfluidic device that cost less than a fast-food meal.

Khine began using the chips in her experiments, but she didn’t view her toaster-oven hack as a breakthrough right away. “I thought it would be something to hold me over until we got the proper equipment in place,” she says. But when she published a short paper about her technique, she was floored by the response she got from scientists all over the world. “I had no idea people were going to be so interested,” Khine says.

20 thoughts on “Children’s toy inspires a cheap, easy production method for high-tech diagnostic chips

  1. machright says:

    How many great inventions have come out of childhood experiences!

    Maybe this could be useful in the make community?

  2. Jason says:

    Laser printers don’t leave “ink particles”. Is it really too much to ask that tech writers at least have some basic level of intelligence?

    I thought one reason microfluidic chips were so expensive is the tolerances were so precise, and I think I read here a while ago that the shrinkage is not very exact depending on time, temperature and a lot of criteria. I’d like to see a link to details like how she got around that.

    1. ehrichweiss says:

      I think you’re overthinking this and being a bit pedantic to boot.

      Toner may not equal “ink” in your book but to many people(including here) it likely does at least in a general sense and I’ll bet that 99% of the people reading this, yourself possibly included, do not know the fundamental difference between an inkjet printer and a laser printer besides the obvious. An inkjet prints blobs of mixed color, where a laser printer prints discrete dots of CMYK that look like the intended color from a distance; look at the color orange printed by both under a magnifying glass and that becomes clear. That is beside the point somewhat but worth noting.

      As for shrinkage and tolerances, well she apparently produced something easily enough that she was able to use it and others found it useful as well. Maybe this is more about how inexpensive one could produce something on the lower range, than how tight the tolerances were for her particular implementation.

  3. alandove says:

    The comment thread on has a pretty good synopsis of this. For details of the technique, you can check out Khine’s web site, and/or see one of the many earlier print articles about it (including the one I wrote for Nature Medicine in 2008, which subscribers can read at

  4. Jason says:

    @ehrich – In this particular instance, the physical differences between ink and toner make a huge difference and not for the “obvious” reasons you mention. Fools will think “well ink and toner give about the same result on paper so it’s the same thing here.”

    As far as tolerances — she is obviously a brilliant scientist who must have had to deal with those pesky little details, where the tech writer is a fool who glosses over it.

    Anyway, won’t a laser printer fuser shrink the shrinky dink film and destroy the printer when the media becomes small and hard? At least that’s what I read on this blog when they were discussed, and it certainly seems logical.

    1. AndyL says:

      Perhaps the fuser’s heat is too momentary. According to the paper they’re using A “Hewlett-Packard LaserJet 2200D” but they mention trying a couple of different laser printers, and there’s no discussion of the printer’s heat damaging the Shrinky Dink.

    2. alandove says:

      You clearly enjoy harshing on tech writers, which shows that you’ve never tried condensing a difficult topic into a fixed word count on a tight deadline (usually for mediocre pay). There is never enough space in a general-audience article to detail every procedure at the molecular level, so yes, some simplifications are inevitable. As for the specific difference between ink and toner, ehrichweiss hits it on the head: it may matter to you, and it may matter for someone who actually wants to replicate this process, but it really, truly doesn’t matter for a meta-discussion about what makes the technique interesting.

      Rather than taking potshots from your armchair, why not look up Khine’s original papers and/or her web site for more details on her protocols. Or call her and ask. Or just pick up a few sheets of Shrinky-Dink material at the toy store and try it out yourself. Empirically, the heat from the laser printer isn’t a problem – the technique works.

      As Thoreau said, “It is remarkable how long men will believe in the bottomlessness of a pond without taking the trouble to sound it.”

      1. Sean says:

        of why some research gets held up. Too many people who say it can’t be done for some series of reasons.

        And thankfully, every so often, someone who’s willing to do it despite “known facts”, possibly fail a few times trying, but in the end come up with a new idea or method that hasn’t been done before.

        Think Goodyear and all the failed attempts before he discovered that sulphur would encourage polymer crosslinking that stabilized natural rubber in a process he named “Vulcanization”.

      2. Andy L says:

        Besides, it’s perfectly clear what you mean when you talk about the “ink” used by a “laser printer”. So long as you specify laser printer, there’s no chance of confusion.

        It’s like calling graphite “lead”. So long as you mention pencils you will communicate effectively and precisely even though you’re using the “wrong” word.

        You can nitpick, but it serves no purpose, because no one was confused.

        1. Miguel says:

          Sorry but I feel stupid but I was confused a little on that. When they said ink I wasn’t sure if there was a mistake in calling it a laser jet rather than an ink jet printer.
          As for ink vs. carbon it would make sense that it would be an ink jet since you would color the shrinky dinks with markers as I remember.
          I wonder if using ink would create those same channels or if the laser printer is special since the carbon has to build up and can’t melt into the plastic.
          I guess the point is that it works and lets move one.

Comments are closed.

Discuss this article with the rest of the community on our Discord server!

current: @adafruit - previous: MAKE, popular science, hackaday, engadget, fallon, braincraft ... howtoons, 2600...

View more articles by Phillip Torrone
Maker Faire Bay Area 2023 - Mare Island, CA

Escape to an island of imagination + innovation as Maker Faire Bay Area returns for its 15th iteration!

Buy Tickets today! SAVE 15% and lock-in your preferred date(s).