In electromagnetic scattering, the so-called T matrix encompasses the optical response of a scatterer for any incident excitation and is most commonly defined using the basis of multipolar fields. It can therefore be viewed as a generalization of the concept of polarizability of the scatterer. We calculate here the series expansion of the T matrix for a spheroidal particle in the small-size, long-wavelength limit, up to third lowest order with respect to the size parameter, ̃ \~X, which we will define rigorously for a nonspherical particle. T is calculated from the standard extended boundary condition method with a linear system involving two infinite matrices P and Q, whose matrix elements are integrals on the particle surface. We show that the limiting form of the P and Q matrices, which is different in the special case of spheroid, ensures that this Taylor expansion can be obtained by considering only multipoles of order 3 or less (i.e., dipoles, quadrupoles, and octupoles). This allows us to obtain self-contained expressions for the Taylor expansion of T( ̃ \~X). The lowest order is O( ̃ \~X³) and equivalent to the quasistatic limit or Rayleigh approximation. Expressions to order O( ̃ \~X⁵) are obtained by Taylor expansion of the integrals in P and Q followed by matrix inversion. We then apply a radiative correction scheme, which makes the resulting expressions valid up to order O( ̃ \~X⁶). Orientation-averaged extinction, scattering, and absorption cross sections are then derived. All results are compared to the exact T-matrix predictions to confirm the validity of our expressions and assess their range of applicability. For a wavelength of 400 nm, the new approximation remains valid (within 1% error) up to particle dimensions of the order of 100--200 nm depending on the exact parameters (aspect ratio and material). These results provide a relatively simple and computationally friendly alternative to the standard T-matrix method for spheroidal particles smaller than the wavelength, in a size range much larger than for the commonly used Rayleigh approximation.
No takes yet. Share an insight, caveat, or question.
Majić et al. (2019) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: