If you kept cutting a magnet down to its last atom, would you ever hold a lone north pole?
Welcome to FreeAstroScience. We went back to Dirac’s 1931 paper and CERN’s 2022 empty-handed search to trace the one particle physics wants but cannot catch. What we found is a hunt kept alive by a promise: a single monopole, anywhere, would explain electric charge itself.
No experiment has ever isolated a north or a south magnetic pole: every magnet ever tested, down to the single electron, carries both at once, an absence written directly into Maxwell’s equations. Paul Dirac showed in 1931 that just one magnetic monopole in the universe would explain why electric charge comes in fixed units. The strongest collider search so far, by the MoEDAL collaboration at CERN in 2022, excluded monopoles up to 75 gigaelectronvolts in mass and found nothing.
During five runs totaling 151 days, a superconducting loop in Blas Cabrera’s Stanford laboratory registered a single jump, sized exactly to the step a magnetic monopole carrying one Dirac unit of charge would have produced. One candidate event. Cabrera published it in May 1982 with a physicist’s caution, and no detector anywhere has repeated it since. Our whole story lives inside that lonely blip, because the particle that would make the deepest sense of electromagnetism keeps declining every invitation to exist.
What Maxwell’s equations forbid
Electricity happily hands you a lone charge. An electron carries pure negative charge, a proton pure positive, and either can drift through space alone. Magnetism never grants the matching freedom. Every magnetic field we have mapped closes back on itself, and Maxwell’s equations state that refusal as a built-in rule: no isolated poles, anywhere.
Cut a bar magnet in half and you hold two complete magnets, each with its own north and south. Halve them again, and again, down past dust and molecules to single atoms, and the outcome never changes. Even the electron acts as a tiny dipole: its magnetism rises from spin, since nature builds every magnetic field out of electric currents and spinning particles.
A lone pole has simply never shown up.
Why did Dirac want one anyway?
In 1931, P. A. M. Dirac published a short paper in the Proceedings of the Royal Society of London with something strange inside. If even one magnetic monopole exists anywhere in the universe, his math showed, quantum mechanics then forces electric charge to come in fixed units, whole-number multiples of a basic step.
Look at any measured charge and you see exactly that pattern. Without a monopole, though, the quantization of charge stays an observed fact with no explanation attached, a rule nature follows for no reason we can name.
We find that the most seductive unfinished argument in physics.
Where did the Big Bang’s monopoles go?
The standard story says inflation diluted them into near nonexistence. Grand unified theories, the attempts to weld the strong, weak, and electromagnetic forces into one structure, predict that monopoles formed in enormous quantities during the first fractions of a second after the Big Bang.
None survive anywhere we can look. Cosmologists call that mismatch the monopole problem. Guth’s 1981 paper in Physical Review D answered it with supercooling: let the infant universe hang billions of times below its critical temperature, and space inflates so fast that any surviving monopoles end up spread too thin to ever cross our path. His abstract argues the same stretch cures the horizon and flatness problems as well — the two puzzles we unpacked in our piece on why the Big Bang needs cosmic inflation, which left this monopole chapter untold.
Guth also flagged the catch himself: the 1981 abstract concedes the scenario “seems to lead to some unacceptable consequences, so modifications must be sought.” Few field-defining papers ship with their flaw printed right in the abstract. We trust the idea more because of that sentence.
The impostors living in spin ice
In 2008, C. Castelnovo, R. Moessner, and S. L. Sondhi argued in Nature that spin ices can host excitations that move and interact like isolated magnetic poles. Spin ices are frozen magnetic crystals, with Dy2Ti2O7 and Ho2Ti2O7 the two they named. Their claim, in their own words: “the dipole moment of the underlying electronic degrees of freedom fractionalises into monopoles” (their spelling, and we are keeping it). The picture even explained a known phase change in these crystals as a liquid-gas transition of monopoles.
Be careful with the headlines, though. When a report says a laboratory created magnetic monopoles, what actually happened is subtler and, we would argue, better. Inside the crystal, a collective motion mimics a monopole perfectly, yet it can never leave its host material. Dirac’s particle would be fundamental, something you could in principle carry across a room. The spin-ice version is neither of those things, and blurring that line oversells a result that needs no exaggeration.
Condensed-matter groups keep mining crystals for particle stand-ins, and the craft carries over: the same computation-first culture produced the machine-learning search that turned up two new superconductors.
Why has no magnetic monopole ever been detected?
Effort was never the problem.
Cabrera’s loop covered 20 square centimeters and listened for 151 days. Take his single candidate at face value and the arithmetic implies roughly 1,200 monopole crossings per square meter every year, which means a detector the size of a door should catch about six a day. Decades of far larger experiments caught none at all. So we will say it plainly: the 1982 event was almost certainly noise, and keeping it romantically alive does the search no favors.
Today’s benchmark belongs to the MoEDAL collaboration at CERN. Smashing lead nuclei together produces, in the 2022 paper’s words, “the strongest known magnetic fields in the current Universe,” fields so intense that light-enough monopoles would simply boil out of empty space through the Schwinger mechanism. None appeared. MoEDAL’s analysis excluded monopoles carrying Dirac charges 1 to 3 with masses up to 75 gigaelectronvolts, at the 95 percent confidence level. In betting terms, those are odds of about 19 in 20 that nothing in that range slipped through. Seventy-five gigaelectronvolts, for scale, is the energy an electron would collect falling through seventy-five billion volts. As of August 2026, that exclusion is still the floor to beat.
Table 1 — Landmark monopole results and what they settled (primary papers, 1931 to 2022)
| Year | Result | Verdict |
|---|---|---|
| 1931 | Dirac ties one monopole to charge quantization | theory, still unclaimed |
| 1982 | one candidate on Cabrera’s superconducting loop | never repeated |
| 2008 | monopole-like excitations in spin ice | emergent, not fundamental |
| 2022 | MoEDAL excludes charges 1 to 3 below 75 GeV | no detection |
Two honest limits sit under the MoEDAL number. A monopole born in a grand unified phase transition would be enormously heavier than anything the LHC can produce. None of the papers we worked from prints the predicted mass, so we will not quote one from memory. Exclusions below 75 gigaelectronvolts say very little about the heavyweight monopoles cosmology actually expects, which could sail past every instrument we own.
We are also leaving out the gauge-theory machinery that makes monopoles compulsory in grand unified models. That construction deserves an article of its own, and skipping it changes nothing on today’s scoreboard. Patience like this has siblings, too. The same negative-result craft drives the axion program, and our explainer on why axion dark matter behaves like a classical field shows the identical discipline aimed at dark matter instead of magnetism.
What 95 years of absence have bought
Count from Dirac’s 1931 paper and the ledger is short: one promise about electric charge still unclaimed, and one candidate blip in Cabrera’s 151 days that never returned. Everything since has widened the silence, from the emergent impostors in Dy2Ti2O7 to MoEDAL’s clean sweep below 75 gigaelectronvolts.
We wrote this page at FreeAstroScience because you deserved more than the one-line answer, and turning stubborn physics into words you can carry on a train ride is the work we choose every day. Keep questioning whatever sounds settled, since a mind allowed to doze is exactly where unreason gets comfortable. Check our door-detector arithmetic, and argue with our verdict on Cabrera if you read his 1982 paper differently. And come back the day any experiment moves that 75 gigaelectronvolt floor, because we will rewrite it the same week. Gerd Dani, FreeAstroScience, Rimini.
Sources
- Dirac, P. A. M. (1931). Quantised Singularities in the Electromagnetic Field. Proceedings of the Royal Society of London Series A, 133, 60. DOI: 10.1098/rspa.1931.0130. https://ui.adsabs.harvard.edu/abs/1931RSPSA.133…60D/abstract
- Guth (1981). Inflationary universe: A possible solution to the horizon and flatness problems. Physical Review D, 23, 347. Published January 1981. https://ui.adsabs.harvard.edu/abs/1981PhRvD..23..347G/abstract
- Cabrera, B. (1982). First Results from a Superconductive Detector for Moving Magnetic Monopoles. Physical Review Letters, 48, 1378. Published May 17, 1982. DOI: 10.1103/PhysRevLett.48.1378. https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.48.1378
- Castelnovo, C., Moessner, R., and Sondhi, S. L. (2008). Magnetic monopoles in spin ice. Nature, 451, 42-45. Published January 3, 2008. DOI: 10.1038/nature06433. https://www.nature.com/articles/nature06433
- Acharya, B., Alexandre, J., et al. (MoEDAL Collaboration) (2022). Search for magnetic monopoles produced via the Schwinger mechanism. Nature, 602, 63-67. Published February 2, 2022. DOI: 10.1038/s41586-021-04298-1. https://www.nature.com/articles/s41586-021-04298-1
