Physicists Unveil a Broader Version of the Aharonov-Bohm Effect

In one of quantum physics’ most celebrated demonstrations, electrons can be visibly altered by a magnetic field they never pass through. Now a new theoretical study has extended that phenomenon — the Aharonov-Bohm effect — to the case in which the hidden magnetic flux changes with time, an open problem that has divided theorists for more than three decades. In a paper published on 22 November 2025 in Foundations of Physics, Shan Gao, a physicist at the Research Center for Philosophy of Science and Technology at Shanxi University in Taiyuan, China, derives a general expression for the phase shift acquired by a charged particle moving through a region that contains no magnetic field but is threaded by a time-dependent vector potential. Working with the semiclassical WKB method, he shows that the textbook result survives in a modified and, at first sight, surprising form: for circular paths the potential-driven shift follows the time-averaged flux, while the full observable phase is fixed solely by the flux present at the moment the particle sets out.

First anticipated in 1949, when W. Ehrenberg and R. E. Siday analyzed the refractive index in electron optics and glimpsed its consequences, and formalized a decade later by Yakir Aharonov and David Bohm, the effect remains the cleanest demonstration that electromagnetic potentials — the scalar and vector quantities from which electric and magnetic fields are mathematically derived — carry direct physical significance in quantum theory. In the canonical arrangement, a coherent electron beam is split and routed on either side of a long solenoid, or of a magnetically shielded torus. The magnetic field is confined entirely within the solenoid, so both partial beams travel through strictly field-free regions; classically, nothing distinguishes their journeys. Quantum mechanically, however, the vector potential outside the solenoid is nonzero, and the two waves acquire a relative phase equal to eΦ/ħ, where e is the particle’s charge and Φ the enclosed magnetic flux. The interference fringes shift even though no force ever touches the electrons. Robert Chambers confirmed the displacement in 1960, and Akira Tonomura’s 1986 experiment, using a toroidal ferromagnet wholly enclosed by a superconducting shield, eliminated lingering doubts that a completely shielded flux moves fringes.

That static result, however, says nothing about what happens when the flux itself is made to change. The moment the current in the solenoid is ramped up or down, Faraday’s law of induction comes into play: a changing magnetic flux generates a circulating electric field in the space around the solenoid — precisely the nominally field-free region in which the electrons travel. That induced field exerts real forces on the particles, accelerates them, and alters their kinetic phases, blurring the boundary between a ‘pure’ potential-driven effect and an ordinary field-driven one. Since the early 1990s a series of analyses has produced partly conflicting answers: B. Lee, E. Yin, T. K. Gustafson and R. Chiao computed the effect of time-dependent vector potentials in 1992; D. Singleton and E. C. Vagenas gave a covariant treatment in 2013; J. Macdougall and Singleton connected the problem to Stokes’ theorem in 2014; J. Jing and colleagues re-examined it in 2017; S. R. Choudhury and S. Mahajan recalculated it directly in 2019; and M. Wakamatsu revisited the controversy in 2024 and 2025. Authors have disagreed over whether the time-dependent phase is a topological flux effect, an induced-field effect, or an inseparable mixture of the two.

Gao’s new paper cuts through the dispute with a direct dynamical calculation rather than a symmetry argument. He applies the WKB, or semiclassical, approximation to the Schrödinger equation for a charged particle moving in a time-dependent magnetic vector potential of the standard solenoid form, in which the azimuthal component falls off as the inverse distance from the axis and is proportional to the instantaneous flux. In this scheme the phase accumulated along a trajectory is written as the line integral of the particle’s canonical momentum, which separates naturally into two contributions: one carried by the vector potential itself, and one generated by the particle’s own motion, which changes because the induced electric field does work on it and reshapes its de Broglie wavelength. The calculation assumes the quasistatic regime, in which the flux varies slowly compared with the time light needs to cross the apparatus, so retardation effects can be neglected. Gao then verifies explicitly that this widely used gauge choice remains consistent with Maxwell’s equations in that limit.

For the simplest and most symmetric case — an electron held on a circular path around the flux — the outcome is a compact formula. The potential-dependent part of the phase shift, the generalized Aharonov-Bohm phase, is proportional not to the instantaneous flux but to its average over the traversal time: it equals one over the period T of the motion, times the integral from zero to T of eΦ(t), where e is the charge and Φ(t) the enclosed flux at time t. In plain terms, a flux that ramps steadily during the electron’s loop produces the shift corresponding to the flux midway through the journey. The paper’s most striking result, however, concerns the total phase. When the kinetic contribution accumulated through the induced electric field is added, the sum collapses to eΦ(0) — the charge multiplied by the flux at the instant the electron begins its circuit. The interference pattern, the quantity an experimenter actually records, therefore depends only on the initial flux: the entire history of the ramp is cancelled, term by term, by the kinetic phase picked up from the induced field. The paper works throughout in natural units, with the reduced Planck constant and the speed of light set to one.

As soon as the restriction to circular paths is dropped, that clean cancellation disappears. For general, non-circular trajectories, Gao finds that the phase depends both on the flux history and on the geometry of the path, because the induced electric field does different amounts of work on different segments of the trajectory, and the kinetic phase no longer mirrors the potential phase in any simple way. The time-dependent Aharonov-Bohm effect, on this analysis, is irreducibly hybrid: part of the phase is carried by the gauge potential, and part is generated by local forces exerted by the induced electric field. The conclusion bears directly on a long-running argument about whether the effect demonstrates a purely nonlocal influence of hidden flux on particles that never encounter the field, or whether induced fields supply a complete local account. Neither extreme is correct once the flux moves in time; potentials and induced fields share the work, and any interpretation must account for both.

A recurrent criticism of the calculations widely used for the time-dependent case is that the solenoid gauge, with its inverse-radius vector potential and its accompanying induced electric field, is at best an approximation that may sit uneasily with exact electrodynamics. Gao devotes part of the paper to testing it against Maxwell’s equations, drawing on exact analyses of the ideal infinite solenoid, including John Templin’s 1995 solution and T. E. Parker’s 2024 treatment of a solenoid driven by a time-dependent surface current, as well as the classic 1985 analysis by T. A. Abbott and David Griffiths of how a solenoid can change its flux without radiating. Those works show that a change of flux launches an electromagnetic disturbance that propagates outward at the speed of light, and that only after the wavefront has passed a given radius does the field there settle into the quasistatic pattern assumed in AB calculations. Gao’s derivation is accordingly restricted to that regime, and he argues that within it the gauge choice common to much of the literature is fully consistent.

The implications reach into one of the deepest interpretive questions in quantum theory: whether gauge potentials are physically real or mere mathematical conveniences. Aharonov and Bohm themselves argued that their effect reveals the significance of potentials, not only of fields, and a 2025 oral-history interview with Aharonov and G. Hetzroni revisits the theoretical discovery, the confirming experiments and the controversy that followed. Gao, who has also published separately this year on the reality of gauge potentials, argues that the generalized results sharpen that debate. Because the observable fringe shift for a closed circular path is locked to the initial flux, while the internal accounting splits between a potential term and a kinetic term driven by an actual electric field, the effect cannot be described as either purely local or purely nonlocal. In his reading, the topological role of the potential persists, but its observable content in time-dependent situations is inseparable from the induced fields that accompany any real change of flux.

The paper then pushes the generalization further, to situations in which an external magnetic field is present in the region of motion, so that the electrons are no longer in strictly field-free space. There the relevant flux is no longer that of the hidden solenoid alone but the flux enclosed by the actual electron trajectory. For circular paths of radius R threading an external field, Gao again finds that the Aharonov-Bohm phase is proportional to the time average of the enclosed flux, and that the total phase, kinetic contribution included, depends only on the initial enclosed flux. For non-circular paths the picture changes qualitatively: the external magnetic field acts on the electrons through the Lorentz force, bending their trajectories and modifying their speeds, so that phase accumulation acquires an additional dependence on the precise path taken. The framework even accommodates external constraining forces that hold a particle on a prescribed loop without influencing its tangential dynamics, a situation relevant to ring-shaped interferometer geometries.

Because the study is entirely theoretical — the paper states that no datasets were generated or analysed — its conclusions await experimental confrontation. But the predictions are concrete enough to be tested. An interferometer built around a pulsed solenoid, with electron beams routed along well-defined paths while the flux is ramped, should display a fringe shift determined by the flux at the start of the loop rather than by its final value, a signature that would directly distinguish Gao’s formula from earlier proposals in the debate. Beyond foundational questions, the work may matter for quantum technologies that already exploit controlled fluxes — superconducting quantum interference devices, flux qubits and interferometric sensors, in which time-modulated magnetic fluxes are routine and phase bookkeeping is everything. As Gao summarizes, the findings clarify how gauge-dependent potentials and induced fields jointly shape the generalized Aharonov-Bohm effect, offering new theoretical insights and, potentially, new control knobs for quantum engineering.