Subscribe to Make Magazine Today!

According to the hacker-proprietor of, shining a 250 mW green laser through a hanging drop of scummy plant water produces a kind of “laser microscope” that will project shadows of all the little microflora on the wall so you can watch them flitter about with the naked eye. The embedded video is nothing short of amazing. I felt just like that cat.

Anybody out there ever tried this with a blue or violet laser? Is the resolution any better?


Sean Michael Ragan

Sean Michael Ragan

I am descended from 5,000 generations of tool-using primates. Also, I went to college and stuff. I am a long-time contributor to MAKE magazine and My work has also appeared in ReadyMade, c’t – Magazin für Computertechnik, and The Wall Street Journal.

  • cogworkz

    Leeuwenhoek, way back in the 1700s, experimented with a primitive microscope that used a drop of water as a lens.

    In this case, rather than looking through the drop to magnify it, the drop instead is being used as a beam spreader and the organisms are acting as occlusions in the lens which appear as shadows.

    As far as improving resolution – you might find some kind of answer in the field of laser holography. I haven’t looked – just suggesting a direction.

  • Alan

    Every science teacher should now rush to the classroom and duplicate this experiment. You can use it to explain a variety of concepts, from optical physics to microbial ecology. And your students will remember it the rest of their lives.

    As for using shorter wavelengths to improve the resolution, I doubt you’d see much difference. This is relatively low magnification (for a microscope), and wavelength doesn’t really become significant until you’re in the oil-immersion range, if I recall correctly.

    It would be really cool to put some GFP-expressing microbes into a drop and display them this way. Call it Leeuwenhoek’s Laser Lightshow.

  • cogworkz

    Had a weird thought about this last night.

    Have you considered focusing 1 red laser and 1 blue laser such that they intersect inside the water drop?

    It dawned on me that if you positioned the lasers correctly and knew the distance to the wall, you could find the angle that the projection would create a red/blue anaglyph shadow. So not only would you have Leeuwenhoek’s laser lightshow, it would be in 3D (if you wore red/blue 3D glasses).

    Also, have a look at this:

  • Sean Michael Ragan

    Had no idea such a thing existed. Assuming that the polarizations of the two beams it produces are actually crossed (it doesn’t say that specifically, that I can tell), then yes, definitely. I note that this particular beam splitter is optimized at 473 nm, which is blue. And since splitting the beam must split the beam power, and you want 250 mW in each beam, I’m guessing you’d want a 500mW blue laser to use this particular beam splitter in this particular application. A 500 mW blue laser diode is very expensive, as I understand it. But if you can find a beamsplitter cube optimized for 555nm or so (in the green), it looks like you can score a 500 mW green laser diode these days for less than $100.

    Also, I think Alan is absolutely correct: My instinct that the color of the light would appreciably affect the resolution of the image was dead wrong. Objects have to be on the order of hundreds of nanometers before the color of the light you use makes a difference, I think, where as these little bugs are probably hundreds of micrometers. Point being: You could probably go ahead and use the cheapest visible laser color (i.e. red) and expect this to work just as well. Assuming that you can find a polarizing beam-splitter cube optimized for the red. If you go the beam-splitter route, you’ll also need mirrors, I think, to aim the split beams, and some kind of mounting system to hold the laser, cube, and two mirrors.

    It might be worthwhile to compare the total cost of the beamsplitting system to the cost of just buying two separate laser diodes, slapping polarizing film over the lenses, and aiming them at the drop separately.

    • KYourke

      Wonderful thread! To achieve a polarized projection, the screen- or wall- you will be projecting onto will have to be able to maintain the polarization of the light as it bounces off the surface of the screen so each of your eyes see only their single image. Most surfaces scramble the polarization- although some plastics and old style silver screens might maintain the organization of the polarized light, allowing you to achieve this cool 3-D effect.

      If you do slap polarizing filters over the lasers they will need to be at the same angle as the glasses, usually opposing directions at 45 degrees, but you can test the match.

      • Sean Michael Ragan

        Thanks for pointing that out. The requirement for a screen that will not scramble the polarizations makes cogworkz’ original red/blue idea rather more appealing. Since the images are monochromatic to begin with, there’s no particular reason to favor the crossed polarizations approach unless it’s cheaper. Some basic investigation confirms that even blue lasers in the 100-200 mW power range can now be had for less than $100, so in terms of maintaining the simple elegance of the original idea, I think one red 200 mW laser + one blue 200 mW laser carefully aimed at the same drop, and viewed with red/blue 3D glasses, would be the way to go.

    • Dileep

      Actually, if I look closely, I see diffraction patterns along the edges of every object in the projection. So this image can’t be made any clearer by focussing. I’d bet the diffraction fringe width is of the order of 0.5 microns (wavelength of green laser), which would put the microbes at 1-10 microns, which are the largest prokaryotes get.

  • cogworkz

    The polarized films would be the better approach for a multitude of reason.

    Here’s some food for though – fluorescent microscopy stains + near UVlaser = something like:


  • Sean Michael Ragan

    …as the laser can only excite fluorescence, which, when it occurs, will be quite weak by comparison, won’t be coherent or polarized, and will not be visible outside the drop of water itself.

  • cogworkz

    Certainly not as a projection! :)

    But what if you went the other way around – the beam hits a luminescent surface (surface painted with luminescent paint) which serves as a plate to expose the variations in the light/shadow. (So you’d turn the laser on for a duration and then kill the lights to see the ghost images captured by the luminescent paint).

    Idea comes in a round about away from this:

    Long exposure with low power lasers might provide superior results. High power may overexpose the details.


    I tried this last night, using a 5mW laser and my bedroom wall :)

    My pictures and videos are here:

    Note that capturing was very hard; even with the ISO boosted to 1600, I couldn’t shoot less than a 1/50s exposure, and there was too much movement. In person, though, the diffraction was pretty intense; at least 5 visible “rings” around everything. The video at gives a good impression of what it was like.

    Although the pics aren’t great, the actual experience was very similar to what Teravolt’s video showed. I also learned how _unstable_ my concrete floors are; jumping 5 meters away was enough to shake the picture beyond recognition, and even walking made it hard to see clearly [the drop was wobbling, not the laser].

    Any thoughts on why Teravolt had so much less visible diffraction than I did?

  • Dileep

Related Supplies at Maker Shed