Shrinky dinks for science.

November 8, 2009 3:52 PM

Michelle Khine makes high tech out of children's toys. Doing medical nanotech with Shrinky Dinks and a toaster oven.
posted by idiopath (10 comments total)

I'm amazed that the nonlinearities of the shrinking didn't completely screw up the channels. You'd think with all the Brownian noise and uneven heating, you'd have a big noise issue.

What I'm more interested in, aside from how clever and awesome this is, is the last paragraph:
Khine recently printed metal patterns on Shrinky Dinks. As the plastic shrinks, the metal buckles to form shallow wells, which Khine thinks may concentrate sunlight; the discovery could help make solar cells more efficient. "We haven't come close to pushing the limits of this technology yet

I guess they're thinking that something could be placed over the cells to concentrate the photons; what's not clear is how the metal wells would actually redirect the particles, because it's unlikely they're shrinking the metal small or precise enough to form waveguides.

In any case, I love reading about all the science that can be done on shoestring budgets. Hell, this is something you could probably do at home if you had access to a decent microscope and a laser printer.
posted by spiderskull at 4:06 PM on November 8, 2009

I knew Dean Venture was on to something.
posted by arruns at 6:05 PM on November 8, 2009 [2 favorites]

Kudos. I never would have associated Shrinky Dinks with a girl like that.
posted by weapons-grade pandemonium at 6:13 PM on November 8, 2009

I'm not sure you'd even need a microscope, spiderskull. (OTOH, I do have a decent microscope and a laser printer at home, and I've been wondering if I could do something nifty with this technique. The fact that I know almost no chemistry or biology is a stumbling block.)

I think the bit about solar cell concentrators is talking about using them like classical-optical reflecting mirrors. Why this would be any better than using an old fashioned macro-scale focusing mirror, I don't know.
posted by hattifattener at 6:46 PM on November 8, 2009

I like news like this that makes me feel proud for people I don't even know. Resourcefulness is up there with the list of character traits I appreciate the most in other folks.
posted by invitapriore at 7:01 PM on November 8, 2009

Let the internet take note: MeFi user mhjb fell eternally in love at exactly 7:47, November 8 2009.
posted by mhjb at 7:47 PM on November 8, 2009

This is good. It's nice to be reminded that the earth is seeded with ideas. Even the most mundane things like a shit one-note toy can be someone's important first step.
posted by kuujjuarapik at 7:50 PM on November 8, 2009

I guess I have the same problem with this that I do with every other microfluidic diagnostic application. Khine wants to "diagnose HIV and other infections at the bedside." Well, guess what? We're already really, really good at diagnosing infection. Microfluidic diagnostics are a solution to a problem that simply doesn't exist.

If you have access to a full diagnostic laboratory, you can run as many enzyme-linked assays as you like, easy-peasy, and get the best, cleanest, most repeatable results possible. Why would you want to complicate this well developed, incredibly reliable technology with a new and kludgey set of devices?

On the other hand, if you're in the field, you'll have lateral flow and dip tests. Durable, simple, and with a human-readable output that even a non-expert can interpret. No syringe pumps, electroosmotic power supplies, fluorescence microscopes, etc. If working in the field is your priority, why in the world would you sacrifice the simplicity of lateral flow for a microfluidic system? You already have something that works.

I mean, it's a clever idea and all, but the applications really need to be thought through. This entire field is hanging on based on a decade-and-a-half worth of speculative applications in the literature that have already been outpaced by the real world.
posted by mr_roboto at 10:20 PM on November 8, 2009 [1 favorite]

mhjb: and you'd say the same thing if she looked a little different?
posted by artifarce at 5:17 PM on November 9, 2009

well.. this would have been cooler if I hadn't heard of George M. Whitesides at Harvard doing similar microstamps designs with pdms...
posted by mhh5 at 3:33 PM on November 10, 2009