Nobody has made dark matter in a laboratory, at least not that we can confirm. But physicists are actively trying. The most realistic approach uses particle colliders to smash protons together at extreme energies, hoping to produce dark matter particles in the debris. The challenge isn’t just creating them. It’s proving you did, since dark matter’s defining trait is that it barely interacts with anything we can detect.
Dark matter makes up roughly 27% of the universe’s total energy content, roughly five times more than all visible matter combined. We know it exists because galaxies rotate in ways that only make sense if unseen mass is holding them together. Yet after decades of searching, no one has caught a dark matter particle in the act. Here’s what scientists are doing to change that.
What Dark Matter Probably Is
Before you can make something, you need to know what you’re making, and that’s the core problem. Physicists have two leading candidates, each requiring a completely different production strategy.
The first and most heavily pursued candidate is the WIMP, or weakly interacting massive particle. WIMPs would be heavy enough to exert significant gravitational pull but would only interact with normal matter through the weak nuclear force, the same force responsible for certain types of radioactive decay. That combination of mass and near-invisibility makes them a natural fit for what we observe in the cosmos.
The second candidate is the axion, a much lighter particle originally proposed to solve an unrelated problem in nuclear physics. Unlike WIMPs, axions would interact with matter through electromagnetism and gravity rather than the weak force. Their extremely low mass means they’d behave less like individual bullets and more like a pervasive field filling space. Each candidate demands its own detection method and, hypothetically, its own creation method.
How Colliders Try to Produce Dark Matter
The Large Hadron Collider at CERN is the most powerful tool available for this work. It accelerates protons in opposite directions around a 27-kilometer ring and crashes them together at energies of 13.6 trillion electronvolts, a record set when the collider began its third run of data collection in 2022. That run will continue for close to four years.
The idea is straightforward in principle. Einstein’s equation linking energy and mass means that enough collision energy can produce new particles, including ones that don’t normally exist in everyday matter. If dark matter particles have masses within the collider’s energy reach (up to around 1 teraelectronvolt for certain interaction types), proton collisions should occasionally produce them.
The problem is what happens next. A dark matter particle created in a collision would pass straight through every detector without leaving a trace. You can’t see it, photograph it, or measure it directly.
Finding What You Can’t See
Physicists get around this by looking for what’s missing. In any collision, the total energy and momentum before and after must balance perfectly. If a dark matter particle flies away undetected, it carries energy with it, and that energy shows up as a gap in the books. Scientists call this “missing transverse energy,” and it’s the primary signature they hunt for.
In practice, this means looking for collisions that produce a single visible jet of particles (or a single photon, or a single heavier particle) recoiling against nothing. These are called “mono-X” searches: mono-jet plus missing energy, mono-photon plus missing energy, and so on. If a visible particle flies off to the left but nothing visible flies off to the right, something invisible carried away the balancing momentum. The ATLAS and CMS detectors at the LHC are specifically designed to catch these lopsided events.
The difficulty is that known particles called neutrinos also escape detectors and create missing energy. Distinguishing a potential dark matter signal from the neutrino background requires enormous statistical power, which is why the LHC needs years of continuous collisions to accumulate enough data.
Why You Can’t Bottle It
Even if the LHC does produce dark matter particles, there’s no way to collect, store, or contain them. Dark matter is, by definition, nearly non-interacting. It doesn’t stick to surfaces, it doesn’t get trapped by electric or magnetic fields (at least not WIMPs), and it passes through solid matter as if it weren’t there. Experiments searching for naturally occurring dark matter are buried deep underground to shield them from cosmic rays, and they still detect essentially nothing. The LUX-ZEPLIN experiment, currently the most sensitive direct detection effort, found zero WIMP interactions in its first science run, setting an upper limit on how often dark matter could possibly bump into a normal atom.
That limit is staggeringly small. At the most sensitive mass point (around 30 times the mass of a proton), the chance of a dark matter particle interacting with a single nucleus is less than 5.9 × 10⁻⁴⁸ square centimeters. For context, that’s roughly a trillion times smaller than the cross-section of a neutrino interaction, and neutrinos are already famously ghostly. Containing something this reluctant to interact with matter is not a materials science challenge. It’s a fundamental physics impossibility with current understanding.
The Axion Approach
Axion experiments take a completely different path. Instead of smashing particles together, they exploit a phenomenon called the Primakoff effect: in the presence of a powerful magnetic field, photons can theoretically convert into axions, and axions can convert back into photons. This mixing between light and axions is the basis for several experiments that aim to “make” axions by shining a laser through a strong magnet and looking for photons that reappear on the other side of a wall.
These “light shining through a wall” experiments are elegantly simple in concept. A photon enters a magnetic field region, converts to an axion, passes through a barrier that would block any normal photon, enters a second magnetic field region, and converts back to a detectable photon. The conversion probability is extraordinarily low, which is why these experiments require the most powerful magnets available and the most sensitive photon detectors ever built.
How the Universe Already Did It
The universe produced all the dark matter we observe today within the first fractions of a second after the Big Bang, when temperatures and energies were far beyond anything a laboratory can replicate. The leading explanation is a process called freeze-out. In the ultra-hot early universe, dark matter particles were constantly being created and destroyed in collisions with normal matter. As the universe expanded and cooled, those collisions became too weak to produce new dark matter, but also too infrequent to destroy what already existed. The remaining dark matter “froze out” at a fixed abundance that persists today.
An alternative mechanism called freeze-in works in the opposite direction. Instead of starting abundant and settling to a relic level, dark matter particles were never in equilibrium with normal matter. They were produced slowly, one by one, through rare interactions until the universe cooled too much for even those feeble processes to continue. The two mechanisms predict different interaction strengths, which is part of why experiments like the LHC and LUX-ZEPLIN are so important. The strength of the interaction, if ever measured, would tell physicists which origin story is correct.
Where Things Stand
The LHC’s current run is pushing into new territory, with collision energies and data volumes that make previous searches look preliminary. For certain types of dark matter interactions, particularly those involving gluons (the particles that bind quarks inside protons), collider searches are already more sensitive than underground direct detection experiments for dark matter masses up to about 1,000 times the proton’s mass.
Proposed next-generation colliders, such as the Future Circular Collider, would dramatically expand this reach. With higher energies and greater precision, they could probe interaction strengths five orders of magnitude below current thresholds.
The honest summary: no one has made dark matter, and even if someone does, they won’t be able to hold onto it. The goal isn’t to produce a usable substance. It’s to prove that a specific particle exists, measure its properties, and finally solve a puzzle that has defined physics for nearly a century.