Answers are grounded in this paper and cite their sources.
In crystals with degenerate phonons, Raman optical activity (ROA), the difference in Raman scattering intensity between circular polarization channels, can arise when the scattering distinguishes components carrying opposite crystal angular momenta. In crystals it has long been associated with structural chirality and more recently with ferroaxial order, which are the electric toroidal multipoles of rank zero and rank one, but whether it can probe higher ranks has remained unknown. Here we report ROA in pyrite FeS₂, a centrosymmetric and nonmagnetic crystal whose Tₕ symmetry forbids the electric toroidal multipoles of ranks zero, one, and two, so that the lowest-rank one it admits is the octupole of rank three. Using circularly polarized Raman scattering in a backscattering geometry, we observe a pronounced circular intensity difference exclusively for the doubly degenerate Eg phonon, reaching about $85%$ of the mean intensity at an excitation wavelength of 633 nm. The difference reverses sign reproducibly between neighboring {111} facets and is strongly suppressed on {001}. Group theory shows that this difference transforms as the antisymmetric part of the self-product of the mode representation, which contains the octupolar A2g representation for the Eg doublet but not for the Tg triplets and that it must vanish for propagation along 001. First-principles resonant Raman calculations, evaluated at the three experimental photon energies within the electric-dipole approximation, reproduce the Eg selectivity and the reversal of sign with excitation energy. These results extend ROA from rank-one to rank-three axial multipolar symmetry and establish it as a symmetry-selective, facet-resolved optical probe of symmetries that produce neither macroscopic polarization nor magnetization.
No takes yet. Share an insight, caveat, or question.
Suganuma et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: