NORMALLY, creating a universe isn’t the job of the Large Hadron Collider (LHC). Most of the back-breaking science—singling out and tracking Higgs bosons, for example—from the world’s largest particle accelerator happens when it launches humble protons at nearly the speed of light.

But for around a month near the end of each year, LHC switches its ammunition from protons to bullets that are about 208 times heavier: lead ions.

When the LHC crashes those ions into each other, scientists can—if they have worked everything out properly—glimpse a fleeting droplet of a universe like the one that ceased to exist a few millionths of a second after the big bang.

This is the story of quark-gluon plasma. Take an atom, any atom. Peel away its whirling electron clouds to reveal its core, the atomic nucleus. Then, finely dice the nucleus into its base components, protons and neutrons.

When physicists first split an atomic nucleus in the early 20th century, this was as far as they got. Protons, neutrons, and electrons formed the entire universe’s mass—well, those, plus dashes of short-lived electrically charged particles like muons. But calculations, primitive particle accelerators, and cosmic rays striking Earth’s atmosphere began to reveal an additional menagerie of esoteric particles: kaons, pions, hyperons, and others that sound as if they’d give aliens psychic powers.

It seemed rather inelegant of the universe to present so many basic ingredients. Physicists soon figured out that some of those particles weren’t elementary at all, but combinations of even tinier particles, which they named with a word partly inspired by James Joyce’s Finnegans Wake: quarks.

Quarks come in six different “flavors,” but the vast majority of the observable universe consists of just two: up quarks and down quarks. A proton consists of two up quarks and one down quark; a neutron, two down and one up. (The other four, in ascending order of heaviness and elusiveness: strange quarks, charm quarks, beauty quarks, and the top quark.)

CERN particle accelerator
The ALICE experiment measures heavy-ion collisions (and their aftermath) with the world’s longest particle accelerator, hosted at CERN. Wladyslaw Henryk Trzaska / CERN

At this point, the list of ingredients ends. You can’t ordinarily chop a proton or neutron into quarks in our world; in most cases, quarks can’t exist on their own. But by the 1970s, physicists had come up with a workaround: heating things up. At a point that scientists call the Hagedorn temperature, those subatomic particles are reduced to a high-energy soup of quarks and the even tinier particles that glue them together: gluons. Scientists dubbed that soup quark-gluon plasma (QGP).

It’s a tantalizing recipe because, again, quarks and gluons can’t normally exist on their own, and reconstructing them from the larger particles they build is challenging. “If I give you water, it’s very difficult to tell the properties of [hydrogen and oxygen atoms],” says Bedangadas Mohanty, a physicist at India’s National Institute of Science Education and Research and at CERN. “Similarly, I can give you protons, neutrons, pions…but if you really want to study properties of quarks and gluons, you need them in a box, free.”

This isn’t a recipe you can test in a home oven. In units of the everyday world, the temperature in a hadronic system is about 3 trillion degrees Fahrenheit—100 thousand times hotter than the center of the sun. The best appliance for the job is a particle accelerator. 

But not just any particle accelerator will do. You need to boost your particles with sufficient energy. And when scientists set out to create QGP, LHC was no more than a dream of a distant future. Instead, CERN had an older collider only about a quarter of LHC’s circumference: the Super Proton Synchrotron (SPS).

As its name suggests, SPS was designed to crash protons into fixed targets. But by the end of the 1980s, scientists had decided to try swapping out the protons for heavy ions—lead nuclei—and see what they could manage. In experiment after experiment across the 1990s, CERN researchers thought they saw something happening to the nuclei. 

“Somewhat to our surprise, already at these relatively low energies, it looked like we were creating quark-gluon plasma,” says Marco van Leeuwen, a physicist at Dutch National Institute for Subatomic Physics and at CERN. In 2000, his team claimed they had “compelling evidence” of the achievement.

For the brief flickers for which the quantum matter exists in the world, physicists can watch the plasma materialize in what they call “little bangs.”

Across the Atlantic, CERN’s counterparts at Long Island’s Brookhaven National Laboratory had been trying their hands with equal parts optimism and uncertainty. The uncertainty faded around the turn of the millennium, when Brookhaven switched on the Relativistic Heavy Ion Collider (RHIC), a device designed specifically to create QGP.

“RHIC turned on, and we were deeply within quark-gluon plasma,” says James Dunlop, a physicist at Brookhaven National Laboratory.

So there are two major QGP factories in the world today: CERN and Brookhaven. With this pair of colliders, for the brief flickers for which the quantum matter exists in the world, physicists can watch the plasma materialize in what they call “little bangs.”

helmeted person stands inside inner workings at CERN
At ALICE’s heart lies a 39-foot-long solenoid maganet, coiled around a thermal shield and a number of fast-trigger detectors. Julien Marius Ordan / Maximillien Brice / CERN

Going back and forth in time

The closer in time to the big bang that you travel, the less the universe resembles your familiar one. As of this writing, the James Webb Space Telescope has possibly observed galaxies from around 320 million years after the big bang. Go farther back, and you’ll reach a very literal Dark Ages—a time before the first stars, when there was little to illuminate the universe except the cosmic background.

In this shadowy age, astronomy steadily gives way to subatomic physics. Go even farther back, to just 380,000 years after the big bang, and electrons are just joining their nuclei to form atoms. Keep going back; the universe is ever smaller, denser, hotter. Seconds after the big bang, protons and neutrons haven’t joined together to form nuclei more complex than hydrogen. 

Go back even farther—around a millionth of a second after the big bang—and the universe is hot enough that quarks and gluons stay split apart. It’s a miniature version of this universe that physicists seek to create.

Physicists puzzle over that universe in office blocks like the exquisitely modernist one overlooking CERN’s visitors center. Look out this building’s window, and you might see the terminus of a Geneva tram line. Cornavin, the city’s main railway station, is only 20 minutes away.

CERN physicists Urs Wiedemann and Federico Antinori meet me in their office. Wiedemann is a theoretical physicist by background; Antinori is an experimentalist, presiding over heavy-ion collision runs. Studying QGP requires the talents of both.

“The existence of quark-gluon plasma we have established,” says Antinori. “What is most interesting is understanding what kind of animal it is.”

For instance, their colleagues who first created QGP expected to find a sort of gas. Instead, QGP behaves like a liquid. QGP, in fact, behaves like what’s called a perfect liquid, one with almost no viscosity. (Yes, the early universe may have been, very briefly, a sort of superheated ocean. Many creation myths might find a distant mirror inside a particle accelerator.)

Both Antinori and Wiedemann are especially interested in watching the liquid come into being, watching atomic nuclei rend themselves apart. Some scientists call the process a “phase transition,” as if creating QGP is like melting snow to create liquid water. But turning protons and neutrons into QGP is far more than melting ice; it’s creating a transition into a very different world with fundamentally different laws of physics. “The symmetries of the world we live in change,” Wiedemann says.

This transition happened in reverse in the very early universe as it cooled down past the Hagedorn temperature. The quarks and gluons clumped together, forming the protons and neutrons that, in turn, form the atoms we know and love today.

But physicists struggle to understand this process with mathematics. They come closer by examining QGP collisions in the lab.

scintillator array at CERN
Central detector components, like the VZERO scintillator array, were built to handle the “ultra-relativistic energies” of the LHC. Julien Marius Ordan / CERN

QGP is also a laboratory for the strong nuclear force. One of the four fundamental forces of the universe—alongside gravity, electromagnetism, and the weak nuclear force that governs certain radioactive processes—the strong nuclear force is what holds particles together at the hearts of atoms. The gluons in QGP’s name are the strong nuclear force’s tools. Without them, charged particles would electromagnetically repel each other and atoms would rip themselves apart.

Yet while we know quite a lot about gravity and electromagnetism, the inner workings of the strong nuclear force remain a secret. Moreover, scientists want to learn more about the role the strong nuclear force plays.

“You can say, ‘I understand how an electron interacts with a photon,’” says Wiedemann, “but that doesn’t mean that you understand how a laser functions. That doesn’t mean that you know why this table doesn’t break down.”

Again, to understand such things, they’ve got to crash heavy ions together.

With the likes of SPS, scientists could look at droplets of QGP and confirm they existed. But if they wanted to actually peer inside and see their properties at work—to examine them—they’d need something more powerful.

“It was clear,” says Antinori, “that one had to go to higher energies than were available at the SPS.”

The universe-faking machine

Crossing from CERN’s campus into France, it’s impossible to tell that this green and pleasant vale—under the grace of the Jura Mountains—sits atop a 17-mile-long ring of superconducting magnets and steel. Scattered around that ring are different experiments and detectors. The search for QGP is headquartered in one such detector.

The road there passes through the glistening hamlet of Saint-Genis-Pouilly, where many of CERN’s staff live. On the pastoral outskirts sits a cluster of industrial cuboids and cooling towers.

Apart from a mural on the corrugated metal facade overlooking a parking lot, the complex doesn’t really advertise that this is where scientists look for QGP—that one of these warehouselike buildings is the outer cocoon of a large ion collider experiment called, well, A Large Ion Collider Experiment (ALICE).

inner workings at CERN
To date, more than 2,000 physicists from 40 different countries have been involved with the decades-long experiment. Jan Hosan / CERN / Fotogloria Agency

CERN physicist Nima Zardoshti greets me beneath that mural: ALICE’s detector, the QGP-watcher, depicted in a pastel-colored mural. Zardoshti leads me inside, past a control room that wouldn’t look out of place in a moon-landing documentary, around a corner covered in sheet metal, and out to a precipice. A concrete shield caps it, several stories below. “This concrete is what stops radiation,” he explains.

Beneath it, occluded from sight, sits the genuine article, a machine the size of a small building that weighs nearly the same as the Eiffel Tower. The detector sits more than 180 feet beneath the ground, accessible by a mine lift. No one is allowed to go down there while the LHC is running, save for CERN’s fire department, which needs to move in quickly if any radioactive or hazardous materials combust.

The heavy ions that collide inside that machine don’t originate in this building. Several miles away sits the old SPS, transformed into LHC’s first steppingstone. SPS accelerates bunches of lead nuclei up to very near the speed of light. Once they’re ready, the shorter collider unloads them into the longer one.

But unlike SPS, LHC doesn’t do fixed-target experiments. Instead, ALICE creates a magnetic squeeze that goads lead beams, racing in opposite directions, into violently crashing head-on.

Lead ions make fine ingredients. A lead-208 ion has 82 protons and 126 neutrons, and both of those are “magic numbers” that help make the nuclei as spherical as nuclei can become. Spherical nuclei create better collisions. (Across the Atlantic, Brookhaven’s RHIC uses gold ions.)

ALICE’s detector isn’t a camera; QGP isn’t like a ball of light that you can “see.” When these lead ions collide at high energies, they erupt into a flash of QGP, which dissipates into a perfect storm of smaller particles. Instead of watching for light, the detector watches the particles as they cascade away. 

A proton-proton collision might produce a few dozen particles—maybe a hundred, if physicists are lucky. A heavy-ion collision produces several thousand.

When heavy ions collide, they create a flash of QGP and spiky jets of more “normal” particles: often combinations of heavy quarks, like charm and beauty quarks. The jets pierce through the QGP before they reach the detector. Physicists can reconstruct what the QGP looked like by examining those jets and how they changed as they passed through.

First those particles crash through silicon chips not unlike the pixels in your smartphone. Then the particles pass through a time projection chamber: a cylinder filled with gas. Still streaking at high energy, they shoot through the gas atoms like meteors through the upper atmosphere. They knock electrons free of their atoms, leaving brilliant trails that the chamber can pick up.

inner workings at CERN
After completing major upgrades in 2021, the ALICE team is ready for Run 3, where they aim to increase the number of particle collisions they sample by 50 times. Jan Hosan / CERN / Fotogloria Agency

For fans of particle physics equipment, the time projection chamber makes ALICE special. “It’s super useful, but the downside of it, and why other experiments don’t use it, is it’s very slow,” says Zardoshti. “The process takes, I think, roughly something on the order of a millionth of a second.”

ALICE creates about 3.5 terabytes of data—around the equivalent of three full-length feature films—each second. Physicists process that data to reconstruct the QGP that produced the particles. Much of that data is processed right here, but much of it is also processed by a vast global network of computers.

From particle accelerators to neutron stars

Particle physics is a field that always has one foot extended decades into the future. While ALICE kicked into operation in 2010, physicists had already begun sketching it out in the early 1990s, years before scientists had even detected QGP at all. 

One of their current big questions is whether they can make QGP by smashing ions smaller than lead or gold. They’ve already succeeded with xenon; later this year, they want to try with an even scanter substance like oxygen. “We want to see: Where is the transition where we can make this material?” says Zardoshti. “Is oxygen already too light?” They expect the life-giving element to work. But in particle physics, there’s no knowing for certain until after the fact.

In the longer term, ALICE’s stewards have big plans. After 2025, the LHC will shut off for several years for maintenance and upgrades, which will boost the collider’s energy. Alongside those upgrades will come a wholesale renovation of ALICE’s detector, scheduled for installation as early as 2033. All of this is planned out precisely many years in advance.

CERN’s stewards are daring to draft a device for an even more distant future, a Future Circular Collider that would be more than three times the LHC’s size and wouldn’t be online till the 2050s. No one is sure yet if it will pan out; if it does, it will require securing an investment of more than 20 billion euros.

ALICE project's inner workings at CERN
ALICE’s inner tracking system holds the record for the biggest pixel system ever built. Felix Reidt / Jochen Klein / CERN

Higher energies, larger colliders, and more sensitive detectors all make for stronger tools in QGP-watchers’ arsenals. The particles they’re seeking are tiny and incredibly short-lived, and they need those tools to see more of them.

But while particle physicists have spent billions of euros and decades of effort bringing fragments of the very early universe back into reality, some astrophysicists think the universe might have been showing the same zeal.

Instead of a particle accelerator, the universe can avail itself of a far more powerful appliance: a neutron star. 

When an immense star, far larger than the mass of our sun, ends its life in a spectacular supernova, the shard of a core that remains begins to cave in. The core can’t be too large, or else it will collapse into a black hole. But if the mass is just right, the core will reach pressures and temperatures that might just tear atomic nuclei apart into quarks. It’s like the ALICE experiment at scale in a more natural setting—the unruly universe, where it all began.

Read more PopSci+ stories.