Borrowing from nothing
A field at zero is still restless
The grid is a quantum field in its lowest-energy state: the vacuum. Its average value is zero, but it keeps fluctuating. Pairs of particles flicker into existence, one +1 and one −1, then meet and vanish.
The hill and the pit they raise in the field cancel exactly. The total never moves off zero, so no energy is created. Bigger fluctuations borrow more energy and must pay it back sooner.
+1 + −1 = 0
Waiting for a pair…
Pairs alive now0
Net field charge0
Pairs so far0
Energy created0
Energy × lifetime stays roughly constant for every pair, a picture of the energy–time uncertainty relation ΔE·Δt ≳ ħ/2.
Picture, not measurement. "Virtual pairs" are a way to visualize terms in the calculation, not events anyone has watched happen. See Notes.
Gravity, time, and a leak
Split a pair at the event horizon
When a pair appears right at the horizon, one partner can fall in before they rejoin. The other escapes. From far away, an otherwise dark black hole seems to glow faintly: Hawking radiation.
Each escaping particle carries off energy that came from the hole's gravity, so the hole loses mass. Smaller holes are hotter and radiate faster, so evaporation speeds up until the end.
Mass left100%
Hawking temp.1.00 T₀
Horizon radius2.40
Particles escaped0
Clock rates near the hole
Gravity slows time. A clock at distance r ticks at √(1 − rₛ/r) of a distant clock. As the hole shrinks, the gold clocks catch up.
The pair-at-the-horizon story is Hawking's own heuristic. The actual calculation compares how the quantum field's vacuum looks to observers near and far from the horizon. In the pair picture, the partner that falls in carries negative energy, which is why the hole shrinks.
The same mathematics describes an accelerating object in empty space (the Unruh effect). Step 3 uses that link.
The proposal · Evans & Schützhold, 2026
Shake a nucleus, shake loose axions
Two heavy ions fly past each other without touching, a so-called ultraperipheral collision. Each carries an electromagnetic field: an electric field pointing outward (gold spokes) and a magnetic field circling its path (teal rings). When the fields overlap, the ions push each other off course. The electric repulsion does most of the pushing.
The dark-matter candidate is the axion. Inside a nucleus the strong force's vacuum (its QCD condensate) is weakened, which lowers the axion's effective mass there. So each nucleus is a small bubble of different material for the axion field (the violet sheath), the way glass is for light.
The fields never touch the axions. They only turn the ions. Turning a nucleus sharply shakes its axion bubble, and shaking a medium fast enough turns vacuum fluctuations into real, entangled pairs: the dynamical Casimir effect. Seen from the nucleus itself, this is the Unruh effect: the empty space around it looks warm and full of axions.
Nuclei touch at this distance. That is a collision, not a near miss.
Closest approach–
Turn–
Peak acceleration–
Unruh temp.–
Axion pairs0
Background e⁺e⁻0
Real turn at LHC energy–
What the detector reportsWaiting for a pass
Electron–positron pairs are real and common in these collisions, but they come from a different process: the two ions' clouds of photons colliding. They are the background any axion search has to beat. They curl in the detector's magnetic field and hit the barrel. Axions are neutral and pass straight through. They are drawn here, joined by a dashed thread because each pair is entangled, but a detector would not see them.
Not to scale. At the LHC, lead ions move at 99.99999% of light speed and turn by about two millionths of a radian. The turn here is exaggerated about 100,000× so you can see it. The acceleration and Unruh readouts are rough non-relativistic estimates, and the pair rates are made up for illustration.
Sources and caveats
What's solid, and what's a sketch
This lab grew from a Reddit comment explaining a ScienceAlert article, which reports on a theory paper. Each retelling simplifies. Below is where the layers agree, where they drift apart, and where this lab itself cuts corners.
Sources
- S. Evans and R. Schützhold, “Axion pair creation by accelerated nuclei,” Physical Review D (2026). doi:10.1103/4nc7-cw6l
- S. Evans and R. Schützhold, “Axion-induced Casimir force between nuclei and dynamical axion pair creation” (2024). arXiv:2404.12996
- “A Particle Accelerator Could Create Dark Matter From Empty Space,” ScienceAlert.
- A Reddit comment summarizing the article, checked claim by claim below.
The paper in four sentences
Inside nuclear matter the QCD condensate is reduced, so the axion's effective mass is lower there and a nucleus acts like a dielectric for the axion field. Two nuclei at rest should feel a weak Casimir-type attraction through that field. A nucleus that is accelerated hard, as in an ultraperipheral collision, can create entangled axion pairs out of the vacuum, by analogy with the dynamical Casimir effect. The same process can be read as a signature of the Unruh effect.
The paper is theoretical. Nothing has been observed. I could not verify predicted rates or whether current accelerators could detect the effect, so this lab makes no such claims.
Checking the Reddit comment
“A vacuum in quantum mechanics is just a field with zero value.”
Simplified
The vacuum is the field's lowest-energy state. Its average value is zero, but it is not a field sitting still at zero. It keeps fluctuating, which is what Step 1 draws.
“Pairs of particles spontaneously appear and disappear… you go from 0 to a +1 and a −1, then back to 0… no energy was created.”
Hand-wave
A popular and useful picture, not an observed process. “Virtual pairs” stand for terms in the calculation. The energy–time relation is not a loan with a due date. Step 1 animates the picture.
“A pair… appears on the edge of a black hole. One falls into the event horizon and one escapes.”
Simplified
This is Hawking's own heuristic. It leaves out that the infalling partner carries negative energy, which is what actually shrinks the hole.
“This is allowed because the acceleration from the black hole imparts energy to the particles.”
Simplified
The energy comes from the hole's gravitational field, not a push on the pair. The real link to acceleration is that Hawking radiation and the Unruh effect share the same mathematics.
“If you accelerate heavy particles in the right way you can impart energy to the zero value field… [which] can then manifest as a pair of particles.”
Holds up
This is the paper's core claim: accelerated nuclei create pairs, via the dynamical Casimir effect or, equivalently, the Unruh effect.
“It's possible this created pair could be a pair of dark matter particles.”
Narrower
Specifically axions, a hypothetical particle that is one dark-matter candidate. The mechanism depends on nuclear matter changing the axion field, so it doesn't apply to dark matter in general. The paper also stresses that the pairs are entangled.
“It's not the magnetic fields interacting with dark matter… Their magnetic fields would interact… It's this acceleration that creates the dark matter particles.”
Mostly holds up
Right that the fields don't touch the axions and that acceleration is the key. Two refinements: the whole electromagnetic field turns the ions, mostly through electric repulsion, and the axions couple to the nucleus's strong-force structure, not to its charge.
Read the full comment
A vacuum in quantum mechanics is just a field with zero value. It's possible for pairs of particles to spontaneously appear and disappear in this field on very short timescales. Essentially you go from 0 to a +1 and a -1, then back to 0. Overall the sum was always 0 so no energy was created. The most famous example of this is Hawking Radiation. A pair of these particles spontaneously appears on the edge of a black hole. One falls into the event horizon and one escapes. This would be seen as radiation from an otherwise inert black hole. Because each radiated particle takes a little energy with it, over huge timescales the black hole evaporates. This is allowed because the acceleration from the black hole imparts energy to the particles.
This is where this proposed experiment comes in. If you accelerate heavy particles in the right way you can impart energy to the zero value field that the particles are traveling through. This energy can then manifest as a pair of particles which could then be detected. And it's possible this created pair could be a pair of dark matter particles.
The important thing here is that it's not the magnetic fields interacting with dark matter. They're proposing to do extremely near misses with charged ions. Their magnetic fields would interact with each other during the near miss to suddenly change the path of the ions. It's this acceleration that creates the dark matter particles.
Where this lab cuts corners
- Axion pairs appear at a rate proportional to acceleration squared. That is a stand-in chosen for illustration, not the paper's formula.
- The axion-mass slider's falloff is a plausible shape, not a calculation.
- Background e⁺e⁻ rates are scaled by Z⁴ like the real process, but the numbers are made up. Real rates are far higher.
- The ions' turn is exaggerated about 100,000×, and time is slowed about 10²³×.
- Acceleration and Unruh temperature use a non-relativistic estimate.
- Axions are drawn so you can follow them. A detector would not see them.
- “Ride with a nucleus” is a cartoon of a frame-dependent idea. No one inside the nucleus sees glowing dots.
- Steps 1 and 2 animate the popular pair picture, which physicists treat as a heuristic.