The notion that magnetism arises from isolated poles dates to the earliest investigations of natural magnets. Pierre de Maricourt, writing in 1269, first systematically described the behaviour of lodestones and identified the concept of magnetic poles. However, every attempt to isolate a single pole — by breaking a magnet, for instance — invariably produced two new dipoles.

By the 19th century, the development of classical field theory by Faraday and Maxwell established that magnetic fields in nature are always solenoidal: the magnetic flux through any closed surface is precisely zero. This is expressed mathematically as ∇ · B = 0, one of Maxwell's four equations.

The four Maxwell equations, in their standard formulation, encode the empirical observation that no magnetic charges exist. The divergence of the magnetic field vanishes everywhere:

Were magnetic monopoles to exist, this equation would be modified to:

where ρ_m denotes the magnetic charge density. The resulting “symmetric” Maxwell equations would exhibit a complete duality between electric and magnetic phenomena.

At the classical level, the introduction of magnetic charge density ρ_m and magnetic current density J_m into Maxwell's equations yields a fully symmetric set of field equations. The Lorentz force law must also be modified to include the force on a magnetic charge moving through electric and magnetic fields.

The symmetrised equations are elegantly expressed in terms of the field-strength tensor and its dual, and the equations of motion exhibit complete electric-magnetic duality. Classical solutions include the Coulomb-like radial magnetic field of a point monopole: B = (g/4πr²) r̂, where g is the magnetic charge.

The modern mathematical description of the monopole utilises the language of principal fibre bundles. The Dirac monopole corresponds to a nontrivial U(1) bundle over the two-sphere surrounding the monopole. The transition functions between the northern and southern hemispheres encode the magnetic charge, and the Dirac quantization condition emerges as the requirement that these transition functions be well-defined.

This topological perspective reveals that magnetic charge is fundamentally a topological invariant — the first Chern number of the U(1) bundle — and is therefore necessarily quantized in integer multiples of a basic unit.

Dirac considered the quantum mechanics of an electrically charged particle in the field of a magnetic monopole. The vector potential of such a configuration necessarily contains a singularity along a semi-infinite line extending from the monopole — the so-called Dirac string. For the string to be physically unobservable (as it must be, since it is merely a gauge artifact), the wave function of any electrically charged particle must acquire a phase of exactly 2πn when transported around the string.

This requirement yields the Dirac quantization condition:

where e is the electric charge, g is the magnetic charge, and n is any integer. This condition implies that if even a single monopole exists in the universe, all electric charges must be integer multiples of ℏc/2g.

Julian Schwinger generalised the Dirac quantization condition to systems containing both electric and magnetic charges. For two particles with charges (e₁, g₁) and (e₂, g₂), the Schwinger-Zwanziger condition reads:

This more general condition allows for dyons — particles carrying both electric and magnetic charge — and constrains the allowed spectrum of charges in any theory containing monopoles.

In 1974, Gerard 't Hooft and Alexander Polyakov independently discovered that spontaneous symmetry breaking in non-Abelian gauge theories necessarily produces magnetic monopoles as finite-energy, topologically stable soliton solutions. Unlike Dirac's point-like monopole, these are smooth, regular field configurations with a definite core radius of order r ~ 1/(M_X), where M_X is the symmetry-breaking scale.

The 't Hooft–Polyakov monopole has a mass of approximately M ~ M_X/α, where α is the gauge coupling constant. For a GUT scale of 10¹⁶ GeV, this yields monopole masses of roughly 10¹⁷ GeV/c² — about 10⁻⁸ grams, a truly macroscopic mass for a single particle.

Standard Big Bang cosmology combined with GUTs predicts the copious production of magnetic monopoles during the grand unification phase transition in the very early universe, approximately 10⁻³⁶ seconds after the Big Bang. The expected density would vastly exceed observational bounds — this is the celebrated “monopole problem.”

The monopole problem was one of the primary motivations for the development of cosmic inflation by Alan Guth in 1981. An epoch of inflationary expansion would dilute the primordial monopole density to negligible levels, explaining their apparent absence from the observable universe.

A magnetic monopole passing through a superconducting loop would induce a persistent current proportional to the magnetic charge. The most famous such experiment was performed by Blas Cabrera in 1982 at Stanford University, using a superconducting quantum interference device (SQUID) coupled to a four-turn loop.

On February 14, 1982, Cabrera's detector recorded a single event consistent with the passage of a monopole carrying one Dirac magnetic charge. The current jump was precisely 2Φ₀ (two flux quanta), as predicted for a monopole passing through a four-turn loop. Despite extensive subsequent searches with larger and more sensitive detectors, no second event has ever been recorded.

Particle colliders have searched for monopole pair production in high-energy collisions. The MoEDAL experiment at the Large Hadron Collider (LHC) at CERN is specifically designed for this purpose, using plastic nuclear track detectors and aluminium trapping volumes to capture and identify monopoles produced in proton-proton collisions.

To date, no collider experiment has observed magnetic monopole production. Current limits from the LHC exclude monopoles with masses below approximately 1–4 TeV/c² for charges up to 5g_D (five Dirac charges), depending on the production model assumed.

Large-area detectors have searched for monopoles in cosmic rays. The MACRO experiment at Gran Sasso, the AMANDA/IceCube neutrino telescopes at the South Pole, and the Baikal Deep Underwater Neutrino Telescope have all placed stringent limits on the flux of cosmic monopoles. Current bounds constrain the monopole flux to below approximately 10⁻¹⁶ cm⁻² sr⁻¹ s⁻¹ for relativistic monopoles — the Parker bound, derived from the requirement that monopoles not drain energy from the Galactic magnetic field faster than it can be regenerated.

During symmetry-breaking phase transitions in the early universe, topological defects — including magnetic monopoles — are inevitably produced by the Kibble mechanism. As causally disconnected regions of space independently choose vacuum states, the mismatch at their boundaries creates defects. The density of monopoles produced is determined by the correlation length at the time of the phase transition, which is bounded by the causal horizon.

For the GUT phase transition at temperature T ~ 10¹⁶ GeV, this mechanism predicts roughly one monopole per horizon volume, resulting in a present-day monopole density comparable to the baryon density — far exceeding observational limits by many orders of magnitude.

Cosmic inflation resolves the monopole problem by positing an epoch of exponential expansion occurring after (or during) the GUT phase transition. This expansion dilutes the monopole density by a factor of at least e^3N, where N ≥ 60 is the number of e-folds. For N = 60, this dilution factor exceeds 10⁷⁸, reducing the expected monopole density in the observable universe to essentially zero.

The success of inflation in resolving the monopole problem — simultaneously addressing the horizon and flatness problems — is regarded as strong circumstantial evidence for inflationary cosmology, though it does not prove the existence of the monopoles whose overproduction motivated the theory.

In 2009, Castelnovo, Moessner, and Sondhi demonstrated that the frustrated magnetic material spin ice hosts emergent quasi-particles that behave as magnetic monopoles. In these materials — notably dysprosium titanate (Dy₂Ti₂O₇) and holmium titanate (Ho₂Ti₂O₇) — the magnetic moments on a pyrochlore lattice obey an “ice rule” analogous to the proton-disorder rule in water ice.

Excitations above the ice-rule ground state take the form of pairs of defects that carry effective magnetic charge and interact via a magnetic Coulomb law. These emergent monopoles have been experimentally observed through neutron scattering, muon spin rotation, and specific heat measurements.

The Montonen–Olive conjecture (1977) proposes an exact strong-weak duality in certain gauge theories: a theory with gauge coupling g is equivalent to a dual theory with coupling 1/g, in which the roles of electrically charged particles and magnetic monopoles are exchanged. This S-duality has been rigorously established in N = 4 supersymmetric Yang–Mills theory and plays a central role in the modern understanding of quantum field theory and string theory.

The Seiberg–Witten solution of N = 2 supersymmetric gauge theory demonstrated that monopole condensation is responsible for confinement in the dual description, providing a concrete realization of the long-suspected dual-superconductor mechanism for quark confinement.

The search for fundamental magnetic monopoles continues with increasing sophistication. The MoEDAL experiment at the LHC continues to extend mass and charge limits. Proposed future colliders, including the Future Circular Collider (FCC), would probe monopole masses up to tens of TeV. Meanwhile, advanced SQUID-based induction detectors and large-volume neutrino telescopes continue to improve flux limits for cosmic monopoles.

Novel search strategies include the examination of lunar regolith samples for trapped monopoles, the use of quantum sensors for ultra-sensitive magnetic charge detection, and proposals to search for monopole-catalysed proton decay in large underground detectors — the Rubakov–Callan effect.