The Nottingham team’s experiment does not take up a whole table—it has a volume of 0.15 cubic meter, which makes it slightly bigger than a stack of 10 large pizza boxes. “It is very, very small. We reduced the size by about 70 percent, compared to a conventional setup,” says Somaya Madkhaly, a graduate student at Nottingham and the study’s first author. To build it, she and her colleagues engaged in something like a very customizable game of Lego. Instead of buying parts, they assembled their setup out of blocks that they 3D-printed to be shaped exactly like they wanted.

Instead of machining the vacuum chamber from sturdy but heavy metals, the team printed it out of a more lightweight aluminium alloy. Instead of building a sprawling maze of lenses and mirrors, they slotted them into a holder they printed out of a polymer. This rectangular piece, only 5 inches long, 4 inches wide, and very sturdy, replaced the delicate optics labyrinth that is usually many feet long.

Importantly, the miniaturized setup worked. The team loaded 200 million rubidium atoms into their vacuum chamber and passed laser light through all the optics components, making the light collide with the atoms. The atoms formed a sample colder than –450 Fahrenheit—exactly as scientists have done with the more conventional kind of apparatus for the past 30 years.

“I think building a cold-atom system like this is a huge step. Only individual components have been 3D-printed before,” says Aline Dinkelaker, a physicist at the Leibniz Institute for Astrophysics Potsdam who was not involved with the study. If previous experiments were sort of like buying a special Lego kit that lets you build a predesigned spaceship, the Nottingham team’s approach was more like designing the spaceship first, then 3D-printing the blocks that make it up.

A big benefit of using 3D printing is that you can custom-design every component, Dinkelaker notes. “Sometimes you have just a small weirdly-shaped component or a weirdly shaped space. Here, 3D printing can be a great solution,” she says.

Lucia Hackermuller, another coauthor on the paper, says that making each piece according to their own specifications allowed them to optimize. “We want to have the best design possible, and the problem is that normally we have construction constraints,” she says. “But if you use 3D printing methods, you can basically print anything that you can think of.” As part of this optimization process, the team used a computer algorithm that they developed to find the best placement for their magnets. They also worked through 10 or so iterations of their 3D-printed components until they fully finessed them.

The new study is a step forward in making this tool for fundamental physics research more affordable and accessible. “I hope this will accelerate—and also to a degree democratize—standard ultracold-atoms experiments by making them cheaper and much faster to set up,” Cooper says. He speculates that if he were stranded on a desert island with just some lenses and mirrors, rubidium atoms, and a 3D printer, he could go from zero to a fully functional device in about a month—five or six times faster than usual. For Madkhaly, starting from scratch may not be just an imaginary scenario. After she graduates, she says, she may return to her home country of Saudi Arabia and use 3D printing to jump-start new ultracold-atom research. “This is a very new field there,” she adds.