A novel geometric phase for optical beams

In this paper, we provide an accurate description of geometric phases emerging in simple optical systems – nanostructures (metaatoms) interacting with vortex beams. We show that this interaction leads to a new class of geometric phase for optical beams, which is different from the geometric phases commonly discussed in structured-light optics. We compare this setting to the usual description of geometric phases for beams and show that the underlying geometry is different.

💡 Research Summary

The manuscript “A novel geometric phase for optical beams” investigates a previously unreported class of geometric phases that arise when vortex beams interact with nanostructured meta‑atoms. While the Pancharatnam‑Berry (PB) phase—originating from the rotation of the polarization state on the Poincaré sphere—is well‑understood and can be described by an SU(2) transformation, the authors demonstrate that vortex beams acquire an additional, fundamentally different geometric phase that depends on the total angular momentum (orbital + spin) of the beam.

The paper begins with a concise review of the conventional PB phase. The authors recast the Jones vector of a polarized plane wave in the circular basis, introduce the complex ratio (z=E_R/E_L), and map it onto the Poincaré sphere via stereographic projection. A waveplate (or a nanostructure with 2‑fold rotational symmetry) is shown to act as an SU(2) matrix (U(\beta,\delta)=\mathbb{I}\cos(\delta/2)+i(\cos2\beta,\sigma_x+\sin2\beta,\sigma_y)\sin(\delta/2)), where (\beta) is the rotation angle of the optic axis and (\delta) the retardation. For half‑wave retardation ((\delta=\pi)) the transformation reduces to a pure rotation on the sphere, and the accumulated phase equals half the solid angle subtended by the closed path, i.e. the familiar PB phase.

To place this in a differential‑geometric framework, the authors introduce the Hopf fibration (S^3\rightarrow S^2). The total space (S^3) (the set of normalized Jones vectors) fibers over the base (S^2) (the Poincaré sphere) with fibre (S^1) (the overall phase). A connection 1‑form (A=2i g,\mathrm{Im}(\bar z_1 dz_1+\bar z_2 dz_2)) is defined; its holonomy along a lifted curve yields precisely the PB phase. The curvature 2‑form (F=dA) and the associated Chern class are identified as the geometric source of the phase.

The core contribution follows: when a vortex beam—specifically a Laguerre‑Gaussian mode (\mathrm{LG}_{p}^{\ell}) with spin (m=\pm1)—illuminates a meta‑atom, the phase shift induced by rotating the meta‑atom is no longer simply (\beta)‑dependent. Instead, the phase scales as ((\ell+m)\beta). The authors justify this by moving beyond the simple two‑dimensional polarization space to a higher‑dimensional mode space. They employ two complementary representations: (i) the Hermite‑Gauss sphere, which treats the real and imaginary parts of the transverse field as coordinates on a sphere, and (ii) the Majorana representation, where an (N)-photon (or multimode) state is mapped to a constellation of points on the Bloch sphere. In both pictures the state space is effectively a complex projective space (\mathbb{CP}^{N-1}), whose fibre bundle structure differs from the Hopf bundle. Consequently, the connection acquires an extra factor proportional to the total angular momentum, and the holonomy becomes (\Phi = (\ell+m)\Omega/2), where (\Omega) is the solid angle traced by the effective “mode‑sphere” trajectory.

Numerical simulations support the theory. A rectangular nanoparticle with (D_{2h}) symmetry is rotated in steps of (\beta) while being illuminated by a right‑circularly polarized fundamental Gaussian beam. The cross‑polarized (left‑circular) component of the scattered field exhibits a phase (\phi = 2\beta), whereas the co‑polarized component remains invariant, confirming the ((\ell+m)) scaling for (\ell=0). The authors argue that for higher‑order vortex beams the phase modulation would be proportionally larger, enabling mode‑selective phase control.

In the discussion, the authors reflect on terminology, emphasizing that “geometric phase” should be reserved for holonomies of connections on appropriate bundles. They note that while the PB phase lives on the Hopf bundle (base (S^2), fibre (S^1)), the vortex‑beam phase lives on a more intricate bundle whose base is a higher‑dimensional sphere (or projective space) and whose curvature encodes orbital angular momentum. Potential applications are highlighted: metasurfaces that simultaneously manipulate polarization and orbital angular momentum, multiplexed free‑space communication channels, and quantum information protocols where mode‑dependent phases are essential.

The conclusion restates that the work extends the geometric‑phase paradigm from pure polarization to full spatial‑mode structure, providing a rigorous differential‑geometric description and suggesting practical metasurface designs for mode‑selective phase engineering. Limitations include the lack of experimental validation beyond numerical models and the complexity of fabricating meta‑atoms with the required symmetry and low loss. Future work is suggested in experimental demonstration, loss mitigation, and exploration of nonlinear regimes where the geometric phase may couple to intensity‑dependent refractive indices.

Overall, the paper offers a substantial theoretical advance, linking fiber‑bundle mathematics, Chern classes, and optical vortex physics, and opens a promising route toward multifunctional metasurfaces that exploit both spin and orbital degrees of freedom.