by Laura Jane Brooks (Bowman group, Cambridge) Labtimes 02/2016
Looking for a high performance microscope that’s small enough to slip under a fume hood or inside a biosafety cabinet? One so cheap to produce that it’s practically disposable? 3D printing can provide the answer.
Richard Bowman's research group at the NanoPhotonics Centre in Cambridge, UK, purchased their first 3D printer around three years ago. Very quickly, simple printed gadgets such as clips and holders found their way into many of our set-ups. But Bowman, who made his PhD in optical tweezers, had something rather more ambitious in mind. He set himself the challenge of designing a lab-worthy 3D-printed microscope, using as few extra parts as practically possible.
The task is a tricky one. Ask any microscopist what makes the difference between a good microscope and a bad one, and they are sure to tell you that stability and accurate sample positioning are absolutely crucial. But can it really be possible to produce a high performance instrument from extruded plastic? Richard thought so, and after many months of careful tinkering, a zoo of printed prototypes and countless cups of coffee, he and his colleagues proved that it could be done. Better still, their innovative design is completely open source, so you’re free to use it and modify it for yourself.
Attempting to replicate conventional metal components in plastic can lead to quite disappointing results, producing flimsy printed pieces that look decidedly shoddy next to their machined metal counterparts. Clearly, a different approach is needed – one that is adapted to the properties of the materials at hand.
Unlike metals, polymers excel at reversible deformation – the ability to bend when stress is applied and return to their original shape when the load is removed. Realising this, Richard turned the natural compliance of plastic to his advantage, by designing a flexure-based mechanism for positioning and focussing.
The microscope body is printed as a single piece, to enhance its stiffness. Photo: Bowman group
Flexures are thinned regions of material that bend to allow movement. Acting as hinges between rigid linkers, these flexures create a system of levers that translates the microscope’s sample stage along its x and y axes. This arrangement provides remarkably precise motion control, producing movements accurate to less than a micron. Using a similar mechanism for the vertical axis, the microscope objective moves up and down to focus accurately on the sample and, what’s more, each axis can either be operated by hand, or fitted with stepper motors for automated control.
So far, so good but surely the same pliability that makes printed plastic effective for flexures must cost a great deal in terms of mechanical stability? How can you make a stable microscope out of a material with such notoriously low stiffness?
As it turns out, a bit of thoughtful design can go a long way to minimising these problems. To start with, the entire stage prints in one single piece, which greatly increases the rigidity of the overall structure. The fact that the whole mechanism is formed of the same material also gives it a strong advantage, since the effects of thermal expansion are largely cancelled out across the object. Meanwhile, the stage’s compact, table-like shape makes it far stiffer than conventional designs and its small size means that there is simply less material to bend or expand.
As a result, both the focal height and the lateral position of the sample drift very little over time, making it suitable for time-lapse measurements over many hours, days, or even weeks. One of the beauties of Richard’s design is that it needs very few non-printed parts and they are easy to get your hands on for not a lot of money. The optics and electronics can be sourced, for example, from an inexpensive Raspberry Pi camera, providing an optical resolution of about 2 microns, a value comparable with a good 20x objective in a standard lab microscope. A simple LED is all that is required to provide suitable illumination, with just a tiny handful of nuts, bolts, washers and elastic bands beeing needed to fix the parts together to complete the assembly.
Thanks to the open hardware movement, makers are now able to invent, share and develop their designs more easily than ever before. You can freely download Richard’s source code and assembly instructions for the microscope here: www.docubricks.com/projects/openflexure-microscope
The OpenSCAD code can be easily adapted to suit your own requirements and should print on pretty much any extrusion-based printer, from the basic kits costing just a few hundred pounds, right up to more expensive machines. However, the optimal flexure thickness can vary, depending on the printer and the filament material, so a test object is helpfully included in the instructions to make it easy to correctly adjust your machine.
If you would like to read more about the flexure-based design, you can find all the details reported in the journal Review of Scientific Instruments (Sharkey et al., 2016, 87, 025104).
Cheap and easy to manufacture, lightweight and portable – the printed microscope enables sensitive time-lapse experiments without the weight and expense of a conventional research microscope. This makes it perfect for use in tight spaces, like fume hoods and incubators, as a disposable device in containment facilities, or even out in the field.
Last Changed: 11.04.2016