We're increasingly seeing 3D-printed metal components being used in fields such as aviation, where failures could be catastrophic. It's therefore very important to check those items for structural flaws before they're installed, and the best way of doing so may involve freezing them in ice.
Because such parts are manufactured (in most cases) by depositing or melting successive layers of metal, the inspection of the solid material can only be performed once the parts have been made. By contrast, when parts are more traditionally machined out of a solid block of metal, it's relatively easy to first check that block for cracks or other faults by running ultrasound waves through it – the waves will bounce off of any irregularities within the block.
Unfortunately, if you try running ultrasound waves through a 3D-printed metal part, those waves will bounce off the curves and angles of its surface, masking any flaws within. What's needed is to immerse the part in a material that's similar in density to the metal, so that the ultrasound waves can travel unimpeded through both that material and the 3D-printed part, reflecting only off of flaws.
Francesco Simonetti, an aerospace engineering professor at the University of Cincinnati, discovered that ice can serve as that surrounding material (known as a coupling medium). In lab tests, he successfully utilized a rig that he built from "things bought on Amazon" to first freeze 3D-printed metal objects within a cylinder of ice, and then find any faults within them by running ultrasound waves through that cylinder.
The ice, however, can't contain any cracks or bubbles that the sound waves might detect as irregularities – it has to be crystal clear.
For their part, cracks typically occur as water freezes from the outside of a vessel inwards, resulting in an ice shell surrounding a liquid core. As that core in turn freezes and expands, it exerts pressure against the shell, causing cracks.
To keep that from happening, Simonetti first placed the 3D-printed items within the empty cylinder, then filled it with water, and then used a metal plate on the bottom of the cylinder to chill that water from the bottom up. This caused the water to freeze starting at the bottom of the cylinder – not the sides – expanding to the container's open and non-restraining top. There was no internal pressure, so no cracks formed.
In order to keep bubbles from forming, the water was mechanically stirred as it was chilling. This kept the dissolved air within the water from forming into bubbles along the "freeze front," which is the area where the forming ice meets the still-liquid water.
Although the resulting ultra-clear ice works well as a coupling medium, it's still not perfect. To that end, Simonetti plans on experimenting with adding suspensions of nanoparticles to the water, resulting in denser, heavier, stronger ice that's even more similar to metal.
A paper on the new technology, which has been dubbed cryoultrasonics, was recently published in the journal NDT& E International.
Source: University of Cincinnati