Magnetic fields are indispensable for controlling quantum phenomena in materials. Recently, they have been integrated directly into ARPES (magneto-ARPES), but trajectory distortions turned out to be significant even in small fields. Related studies include magnetic patterns in thin films, spin compasses, and magnetic resonance in chromium oxide. All seek ways to control spins and electronic structure, but the challenge here is different: not to create a field, but to correct its effect on the measurement itself.
We used laser ARPES with a photon energy of 6 eV and a solenoid around the sample producing a dipole field up to 10 mT. Photoelectron trajectories were modeled with finite elements, and equations of motion were solved numerically. Post-processing involved mapping measured angles back to initial angles using correction matrices. The Bi2Se3 sample was grown by a modified Bridgman method, and measurements were taken at 30 K.
We recovered the Fermi surface and dispersion of topological surface states of Bi2Se3 with an accuracy of ±1° in angle for fields up to 8.6 mT. The momentum broadening did not exceed 2Δθ0 (where Δθ0 = 0.8° is the intrinsic linewidth at zero field). We showed that the dipole field acts as an electron lens with focal points where angular resolution is lost, but between them measurements are possible. Circular dichroism reveals the spin-orbit texture and does not change with field, confirming the quality of the correction.
This work differs from neighboring approaches in that it does not attempt to eliminate field distortions with hardware (e.g., multipole configurations) but corrects them numerically. This allows the use of a simple solenoid while maintaining high resolution. For the first time, it is shown that magneto-ARPES can operate in a varying field with accurate recovery of the spectral function. Scaling laws from simulations provide a roadmap for transferring the method to other setups.
The method can be extended to larger fields by avoiding focal points and using higher photon energies or a bias voltage. Combining it with ferromagnetic cores or multipole field configurations will increase the range. It could also be applied to other materials with strong spin-orbit coupling.
It will impact research on topological insulators, superconductors, and magnetic materials, where knowing the electronic structure in a field is crucial.
Test the method on materials with more pronounced Zeeman splitting and implement measurements in fields above 10 mT.
This connects to an unresolved problem in condensed matter physics: how to observe field-induced splitting of the Dirac point in topological insulators without broadening artifacts. The work shows that smaller beam spots and larger fields are needed.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The paper develops a numerical method to correct distortions in photoelectron trajectories in magneto-ARPES. On the topological insulator Bi2Se3, it demonstrates the recovery of the Fermi surface and dispersion with an accuracy of ±1° at fields up to 8.6 mT. It is shown that the dipole field acts as an electron lens, and measurements are possible between focal points.
Where its strength is
The strength lies in the fact that the correction is performed numerically, which allows the use of a simple solenoid while maintaining high resolution without complex magnetic optics. For the first time, the possibility of magneto-ARPES in a varying field with accurate recovery of the spectral function is demonstrated.
Where it could go next
Worth trying
Why this may matter
The method paves the way for routine magneto-ARPES measurements in laboratories that lack specialized magnetic lenses, and it could become a standard for studying topological materials in a magnetic field.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.55, against a usual 0.64. It looks like few have drilled nearby.
The solenoid heats up by only 4 K at maximum current, and its field at a distance of 1.35 mm from the sample is attenuated by 99.6%.