From Nuts and Bolts to the Biggest Mystery in the Universe: How to Build a Dark Matter Detector
By Zhiheng Sheng, Northwestern University
Imagine building a custom car from scratch. You’d contemplate the engine designs, meticulously weld the frame, and painstakingly route the thousands of meters of cable. And then comes the moment of truth: you take it out onto the road for a first drive. You feel the knot in your stomach as you turn the key, half-expecting it to fall apart despite all your planning and calculations.
The mix of careful engineering and gut-wrenching uncertainty is exactly what it’s like to build a new dark matter detector. We weren’t just building a machine based on a blueprint; we were building a prototype, a pathfinder to test our new ideas. Along the journey, we ran into problems. Sometimes, the “proper” fix was too slow or expensive, forcing us to invent a clever workaround just to push forward. For example, an expensive part that requires a year of lead time cannot be easily replaced. It would be much easier to put together a contraption that will work with said part. Every setback became a valuable lesson for the next time, but for now, all we can do is to keep going.
The Hunt for Dark Matter
Dark matter is one of the biggest mysteries in our universe. From almost a century ago, scientists have had convincing evidence that it exists, but we’ve only seen its gravitational effects on the stuff we can see, like stars and galaxies.
In the 1960s, astronomer Vera Rubin noticed something strange about the way galaxies rotate. The stars on the edges were spinning just as fast as stars were closer to the center. This contradicts the laws of gravity, unless they are also affected by gravity from a huge amount of invisible “stuff.” This invisible halo, or dark matter, permeates from the very center of the galaxies to the most distant star and beyond. Yet, after decades of searching through numerous experiments, not even one experiment has managed to directly detect a dark matter particle.
Building the Detector
Our experiment, the Scintillating Bubble Chamber, aims to do just that, but will look for a new set of signals. Dark matter experiments have been using both bubble chambers and scintillating liquids like argon and xenon. The Scintillating Bubble Chamber is functionally a bubble chamber. But instead of using freon used like the previous PICO experiment, it is filled with argon, which will create a tiny flash of light when a particle comes in and interacts with it, hence the name “scintillation”. Imagine a super-pure tank of liquid argon, chilled to a frosty -180°C (-300°F), and held between two clear quartz jars. The jars are surrounded by an array of sensors that are so sensitive that they can detect the faintest of light and the tiniest of sounds from a bubble less than a micron across, less than one hundredth of the width of a human hair. When a dark matter particle comes across the detector and strikes an argon nucleus, it will create a tiny bubble. Three cameras situated on top can capture the bubble formation in less than five hundredths of a second. When a dark matter particle interacts with the argon, it will have a chance to create a bubble. However, a list of other particles is also able to create bubbles. In those cases, the scintillation light that accompanies the bubble can be used for differentiation.
For the last few years, our small team at Fermilab — a mix of postdocs, grad students, and professors — have been hard at work to bring this detector to life. We test thousands of instruments and kilometers of cables. We did multiple rounds of simulated operation, varying the temperature and pressure in our detector to the target values. Finally, in May 2024, we are ready to move the completed detector from our assembly and testing lab on the surface to its final home — the underground cavern that previously housed the MINOS experiment.
The tunnel is 100 meters (300 ft) underground, the deepest spot in Fermilab. The thick overburden of rocks shields our detector from a constant shower of cosmic ray particles. For an experiment as sensitive as ours, reducing such backgrounds is crucial. Otherwise, we will be overwhelmed by bubbles from particles that we are not interested in.
The detector was loaded on a flatbed truck, driven four kilometers across Fermilab, and then a crane slowly lowered it down a shaft into the tunnel, where a forklift drives it to its final place. After years of work, this was our moment. The detector was safe in its cradle, ready to begin its mission. I was happy that night, full of hope, completely unaware of the disaster brewing within…
The Catastrophe
Underground, we went to work again, starting another round of commissioning tests to control the temperature and pressure of the liquid argon. Small problems are always expected: a disconnected sensor, a bug in the control software. But one troubling problem slowly arises. There seems to be a leak through the jars. When we measure the leak again after a few days, it was worse. After weeks of measurements and discussions, we decided that we cannot start data taking in these conditions, and we need to take the whole detector apart and know for sure what’s going on.
It took a month of coordination and planning to start the disassembly. On an early morning, the riggers arrived with the forklift and the crane, and we followed the step-by-step procedure of taking the detector apart without damaging any of the cables. Once the final shield was lifted, the mood suddenly got tense. It wasn’t a small scratch on the seal, like we assumed. The whole jar was in pieces. We all took a step back and stared at each other speechlessly. It’s true that we have discussed this possibility before. But until we all laid eyes on the shattered glass, it was just that — a possibility. The worst has really happened, and now, we must find a way to fix it and make sure it doesn’t happen again.
From the Ashes
The next two months were like a forensic investigation. We were in a clean room every day, suited up in gloves, hair nets, and shoe covers. We carefully removed the shards of quartz, disconnected the sensors, and made precise measurements along the way. We took hundreds of photos and filled out pages of spreadsheets while our Canadian collaborators, who designed the jars, did the investigation.
The culprit was the mounting system. As the detector cooled 200 degrees Celsius, it was unable to provide enough support, allowing the jars to shift and shatter under uneven stress. They returned with not only spare jars, but a redesigned mounting system — an elegant fix to a devastating problem.
Yes, it was a major setback. And yes, it added months to our timeline. But now, armed with a new set of jars and the knowledge of how to protect them, we resumed our work. The detector is once again fully assembled, more robust than ever before. Seven years after the project was first proposed, we are on the verge of finally letting it listen for the quietest of whispers from the cosmos. After another day of work, I gaze at the detector and realize: the journey of curiosity is never a plain sail, but a never-ending winding trail of trials, errors, and persistence.
Acknowledgements: This essay was written for the Science Policy & Advocacy for Research Competition (SPARC) at the Universities Research Association.
