Matt Nowell, EBSD Product Manager, Gatan/EDAX
My son Parker graduated with a degree in materials science and engineering last May, and we are fortunate to enjoy discussing materials, microstructures, and characterization together as a shared interest. About a month ago, he sent me a video of someone showing a 2D-grain growth example using BBs moving between two pieces of plexiglass. He expressed an interest in trying to do this together. During his recent visit home during the holiday season, we tried to replicate this work.
To build this, we decided to make the plexiglass casing using the 3D printer we have at home. I purchased this years ago to encourage my boys to learn about technology and because of my interest in additive manufacturing. While I’m used to analyzing 3D printed metallic materials with electron backscatter diffraction (EBSD), we printed using a polylactic acid (PLA) filament, a recyclable thermoplastic.
We had to make a 3D drawing of our design. I haven’t done 3D CAD work in a long time, but we were able to hack a base and a lid design together in Blender. This lid is shown in Figure 1.
Figure 1. 3D model of the lid of our design.
Printing wasn’t as easy as I had hoped, but it was a learning experience. We learned that our printer is better if printing directly from an SD card rather than communicating with a PC. We learned you shouldn’t keep PLA filament for years, as it becomes brittle and breaks during long prints. We learned that our printer had a maximum printing size close to our design’s dimensions. We learned that sometimes, when upgrading the firmware and software to fix a problem, it will introduce new issues that then need to be resolved. In the end, though, and after a few iterations, we were able to print a working design. Figure 2 shows the printing of the plexiglass frame. And yes, my 3D printer is made by AnyCubic, which seems appropriate for my EBSD interests.
Figure 2. The printing of the plexiglass frame.
Once printed, assembled, and filled with BBs, you can set this model on its side, and the BBs arrange themselves in a 2D lattice arrangement, as shown in Figure 3. Figure 3a shows the initial distribution. Some areas are organized into ordered regions, which are analogous to 2D grains. Some stacking defects are also observed within some of these grains. There are also regions that are not ordered, which would be comparable to amorphous materials.
Figure 3. a) The initial distribution and b-d) the evolution of the 2D model structure as energy is input into the system through tapping and shaking the model.
We then proceeded to tap and shake the model gently. This is essentially input energy into the system, like what thermal energy would be in a real material. Figures 3b-3d show the evolution of the 2D model structure. Grains coalesce and then grow. Eventually, only a few grains remain, with some twinning-like defects present. A video of this process is shown in Figure 4.
Figure 4. Eventually, only a few grains remain, with some twinning-like defects present.
Of course, this model isn’t perfect, and we will continue to spend some time working with it. It’s easy to get templated growth from the sides aligning the BBs, and we have to be extra careful not to spill them everywhere, or we will both be in trouble. It does remind me, though, of the in-situ grain growth experiments I’ve done with EBSD. Figure 5 shows a video of an orientation map of recrystallization and grain growth in aluminum.
Figure 5. A video of an orientation map of recrystallization and grain growth in aluminum.
I like how models can help us understand the physical phenomena that are actually occurring, and I like being able to discuss them with Parker.