Recently, I was searching for Xmas gifts ideas and came up with a crystallography toy called [Atomix] which use thousands of steel balls to simulate atomic motions. The original toy was created by the French Canadian designer Francois Dallegret back in 1966. After watching the very well explained introduction video covered by Steve Mould [Self organising steel balls explain metal heat treatment], I was quite fascinated that so many complicated materials science concepts can be visualized by such a simple but elegant design. My internal mat-sci nerdiness makes me wanting to have one of this toy so badly. However, after hours of exhaustive googling, it turns out the original toy is no longer in production and the only place sells the replicas is from an overseas private ebay seller with a pretty pricey postage cost. Therefore, I redesign and made an upgrade version of the Atomix crystallography toy myself.

The first prototype was made by laser cuting acrylic panels, 3d printing spacer and sandwiching stainless steel balls (a perfect combination of traditional subtractive and additive manufacturing methods. There is a total of 11127 steel balls (I’ll reveal later how I got this number). Although I’m calling it a prototype, but all the parts assembled together beautifully. After hours of playing with the toy, I found it could be a perfect educational tool for many critical materials science concepts and analytical techniques. So here is a brief summary of what I’ve learnt:
Simulation of different phases of materials
- Quickly shaking Atomix will scramble up the steel balls and spread them into sparsely dispersed “gas” form.
- Slightly tilt the plate of dispersed “gas” will bring the “atoms” more closer to each other under the gravity to form the “liquid” form .
- Tilt the plate straight up will pack the “atoms” more closely into the “solid” form.
- “Solid” form is composed of small ordered hexagonal close-packing domains that resemble the grain structure in polycrystalline materials.
- Shaking the steels balls (or charging the toy by a plastic rod after rubbed on wool) can generate a lot of static charges to cause the atoms to “evaporate”.


Polymerization in materials

Metallography and recrystallization of metallic materials
- Shaking and “condensing” the atoms into a fine grain polycrystal “solid”.
- Use a vortex shaker or a massage gun to vibrate the Atomix plate in the up right orientation. The fine grains will “recrystallize” into a coarse grain polycrystal that resembles the process of annealing in metallic materials.
- Grain boundary and various other crystal defects such as vacancies, stacking faults, dislocations are frequently found in the close-packing crystals.
- During vibration, crystal defects can be found moving rapidly within the crystals, which has been demonstrated in Steve Mould’s video.

Particle analysis technique with FIJI-ImageJ
The total number of steel balls can be easily counted by using the open source image analysis tool [FIJI-ImageJ].




Electron microscopy analysis



And MORE
Of course, there are always more ways that the Atomix toy can be integrated with the materials science education and play. For example, it can be used to generate small angle scattering data demonstrated in the video by youtuber [drheaddamage]; or the crystal orientation can be mapped by simulating a virtual probe to generate datasets that are similar to the Electron BackScattered Diffraction (EBSD) or 4 Dimension – Scanning Transmission Electron Microscopy (4D-STEM), etc. If you have any cool idea, please don’t hesitate to post them in the comment below.
Hope you enjoy the reading.
Monash Materials Special Edition
This design and make was more like a reflection of my previous study and work in the Monash Materials Engineering Department years ago. Therefore, I also created a Monash Materials Special Edition with the “M” logo!


This is great stuff. As a materials professor I enjoyed the simplicity that bubble raft model can provide, and appreciated it more since becoming an educator! Is it possible to purchase this, I think it will make my teaching in crystalline defects much much more clear. Thanks!
Best regards,
Bo
LikeLike
Hi Bo,
Thank you for your kind comment and your interests in this project. Yes, I still have some materials available for making a few more. I have just setup an Etsy store just so that the purchase can be processed easily and securely.
https://www.etsy.com/au/listing/1843536933/atomix-by-francois-dallegret-remixed
I have also made a list of my other recent designs in the page below which I haven’t got time to post on Etsy. Please feel free to check it out.
https://wordpress.com/page/zxmake.dev/1550
LikeLiked by 1 person
I would be very interested in purchasing your material science demonstration, “Atomix by Francois Dallegret Remixed. I have been trying to find a source for this demo for several years now to use in my course on deformation of earth materials (crystals and rocks). Your ebay and epsy links are no longer active, so I am hoping my request makes it to you via your comment page.
Best,
Mark
LikeLike
Hi Mark,
My apologies for the very delayed response! It’s been an incredibly busy year, and unfortunately, I overlooked the comments on my website.
I still have one available. If you’re still interested, I’d be more than happy to offer it to you at a discounted price as a small gesture of apology for the long delay.
Thanks for your patience, and please let me know if you’re still interested!
LikeLike
Nice to hear from you, and thanks for the discount offer. In the meantime, I purchased an Atomix demonstration from another source, so no need for another
Best,
Mark
LikeLike
徐老师您好!我是清华大学行健书院(理论与应用力学专业)的一名大一本科生。
我最近在暑期科研中接触到了力学超材料和二维自组织颗粒介质领域。 在寻找实体物理实验的对照参考时,我非常幸运地看到了您在 zxmake 上的这篇博文。您的 Atomix Remixed 完美地复刻了 1966 年原版的物理精髓,且通过亚克力与精密垫片展示二维晶格缺陷、晶界位错的设计非常优雅,这正是我实验中梦寐以求的简易物理模型。
我目前正计划将圆球颗粒扩展为更复杂的非球形几何形状,以探索它们在二维振动下的自组织堆积规律并进行仿真对比。遗憾的是,我看到您的 Etsy 商店目前似乎已经售罄下架了。
请问您近期是否还有计划组装少量的 Atomix 经典版成品?如果有的话,我很希望能够向您购买一个用于前期的上手观察。另外,如果您的课题组或个人方便,是否可以向我透露一些关于腔体厚度与钢珠直径配比的设计心得?这对于我们自行搭建非球形颗粒自组织实验装置将会有巨大的帮助。非常期待您的回复,谢谢您!
LikeLike
实在抱歉。这段时间太忙,一直没有时间上来查看评论。
我目前手上还有一套Atomix,不过前面Mark有评论也想购买。所以我可能会先等他回复。
其实你自己搭建也不太难,我用的名义上1mm直径的钢珠,腔体厚度可能得需要你根据钢珠实际大小来调整。太大的话,钢珠会偏离六方密排结构,太小的话钢珠又没法自由移动。
LikeLike