T-matrix method

{{Short description|Technique for computing light scattering by nonspherical particles}}

The Transition Matrix Method (T-matrix method, TMM) is a computational technique of light scattering by nonspherical particles originally formulated by Peter C. Waterman (1928–2012) in 1965.{{cite journal | last=Waterman | first=P.C. | title=Matrix formulation of electromagnetic scattering | journal=Proceedings of the IEEE | publisher=Institute of Electrical and Electronics Engineers (IEEE) | volume=53 | issue=8 | year=1965 | issn=0018-9219 | doi=10.1109/proc.1965.4058 | pages=805–812}}{{cite journal |last1=Waterman |first1=Peter C. |title=Symmetry, unitarity, and geometry in electromagnetic scattering |journal=Physical Review D |date=1971 |volume=3 |issue=4 |pages=825–839 |doi=10.1103/PhysRevD.3.825|bibcode=1971PhRvD...3..825W }}

The technique is also known as null field method and extended boundary condition method (EBCM).{{cite journal | last1=Mishchenko | first1=Michael I. | last2=Travis | first2=Larry D. | last3=Mackowski | first3=Daniel W. | title=T-matrix computations of light scattering by nonspherical particles: A review | journal=Journal of Quantitative Spectroscopy and Radiative Transfer | publisher=Elsevier BV | volume=55 | issue=5 | year=1996 | issn=0022-4073 | doi=10.1016/0022-4073(96)00002-7 | pages=535–575}} In the method, matrix elements are obtained by matching boundary conditions for solutions of Maxwell equations. It has been greatly extended to incorporate diverse types of linear media occupying the region enclosing the scatterer.{{cite book |last=Lakhtakia |first=Akhlesh |title=The Ewald–Oseen Extinction Theorem and the Extended Boundary Condition Method, in: The World of Applied Electromagnetics |date=2018 |publisher=Springer |location=Cham, Switzerland |isbn=978-3-319-58403-4|doi=10.1007/978-3-319-58403-4_19 }}

T-matrix method proves to be highly efficient and has been widely used in computing electromagnetic scattering of single and compound particles.{{cite book |last1=Mishchenko |first1=Michael I.| last2=Travis | first2=Larry D. | last3=Lacis | first3=Andrew A. |title=Scattering, Absorption, and Emission of Light by Small Particles |date=2002 |publisher=Cambridge University Press |location=Cambridge, UK |isbn=9780521782524}}

Definition of the T-matrix

The incident and scattered electric field are expanded into spherical vector wave functions (SVWF), which are also encountered in Mie scattering. They are the fundamental solutions of the vector Helmholtz equation and can be generated from the scalar fundamental solutions in spherical coordinates, the spherical Bessel functions of the first kind and the spherical Hankel functions. Accordingly, there are two linearly independent sets of solutions denoted as $\mathbf{M}^1,\mathbf{N}^1$ and $\mathbf{M}^3,\mathbf{N}^3$ , respectively. They are also called regular and outgoing SVWFs, respectively. With this, we can write the incident field as

: $\mathbf{E}_{inc}= \sum_{n=1}^\infty \sum_{m=-n}^n\left( a_{mn} \mathbf{M}^1_{mn}+ b_{mn} \mathbf{N}^1_{mn}\right).$

The scattered field is expanded into radiating SVWFs:

: $\mathbf{E}_{scat}= \sum_{n=1}^\infty \sum_{m=-n}^n\left( f_{mn} \mathbf{M}^3_{mn}+ g_{mn} \mathbf{N}^3_{mn}\right).$

The T-matrix relates the expansion coefficients of the incident field to those of the scattered field.

: $\begin{pmatrix} f_{mn}\\ g_{mn}\end{pmatrix} = T \begin{pmatrix} a_{mn} \\ b_{mn} \end{pmatrix}$

The T-matrix is determined by the scatterer shape and material and for a given incident field allows one to calculate the scattered field.

Calculation of the T-matrix

The standard way to calculate the T-matrix is the null-field method, which relies on the Stratton–Chu equations.{{cite journal | last1=Stratton | first1=J. A. | last2=Chu | first2=L. J. | title=Diffraction Theory of Electromagnetic Waves | journal=Physical Review | publisher=American Physical Society (APS) | volume=56 | issue=1 | date=1939-07-01 | issn=0031-899X | doi=10.1103/physrev.56.99 | pages=99–107| bibcode=1939PhRv...56...99S }} They basically state that the electromagnetic fields outside a given volume can be expressed as integrals over the surface enclosing the volume involving only the tangential components of the fields on the surface. If the observation point is located inside this volume, the integrals vanish.

By making use of the boundary conditions for the tangential field components on the scatterer surface,

: $\mathbf{n} \times (\mathbf{E}_{scat} + \mathbf{E}_{inc}) =\mathbf{n} \times \mathbf{E}_{int}$

and

: $\mathbf{n} \times (\mathbf{H}_{scat} + \mathbf{H}_{inc}) = \mathbf{n} \times \mathbf{H}_{int}$ ,

where $\mathbf{n}$ is the normal vector to the scatterer surface, one can derive an integral representation of the scattered field in terms of the tangential components of the internal fields on the scatterer surface. A similar representation can be derived for the incident field.

By expanding the internal field in terms of SVWFs and exploiting their orthogonality on spherical surfaces, one arrives at an expression for the T-matrix. The T-matrix can also be computed from far field data.{{Cite journal|last1=Ganesh|first1=M.|last2=Hawkins|first2=Stuart C.|date=2010|title=Three dimensional electromagnetic scattering T-matrix computations|journal=Journal of Computational and Applied Mathematics|volume=234|issue=6|pages=1702–1709|doi=10.1016/j.cam.2009.08.018|doi-access=free}} This approach avoids numerical stability issues associated with the null-field method.{{Cite journal|last1=Ganesh|first1=M.|last2=Hawkins|first2=Stuart C.|date=2017|title=Algorithm 975: TMATROM - A T-matrix Reduced Order Model Software|journal=ACM Transactions on Mathematical Software|volume=44|pages=9:1–9:18|doi=10.1145/3054945|s2cid=24838138}}

Several numerical codes for the evaluation of the T-matrix can be found online [http://www.scattport.org/index.php/light-scattering-software/t-matrix-codes/list] [https://www.ugr.es/~aquiran/codigos.htm] [https://www.giss.nasa.gov/staff/mmishchenko/t_matrix.html].

The T matrix can be found with methods other than null field method and extended boundary condition method (EBCM); therefore, the term "T-matrix method" is infelicitous.

Improvement of traditional T-matrix includes Invariant-imbedding T-matrix Method (IITM) by B. R. Johnson.{{Cite journal |last=Johnson |first=B. R. |date=1988-12-01 |title=Invariant imbedding T matrix approach to electromagnetic scattering |url=https://opg.optica.org/ao/abstract.cfm?uri=ao-27-23-4861 |journal=Applied Optics |language=EN |volume=27 |issue=23 |pages=4861–4873 |doi=10.1364/AO.27.004861 |pmid=20539668 |issn=2155-3165|url-access=subscription }} The numerical code of IITM is developed by Lei Bi, based on Mishchenko's EBCM code.{{Cite journal |last1=Bi |first1=Lei |last2=Yang |first2=Ping |last3=Kattawar |first3=George W. |last4=Mishchenko |first4=Michael I. |date=2013-02-01 |title=Efficient implementation of the invariant imbedding T-matrix method and the separation of variables method applied to large nonspherical inhomogeneous particles |url=https://www.sciencedirect.com/science/article/pii/S0022407312005201 |journal=Journal of Quantitative Spectroscopy and Radiative Transfer |language=en |volume=116 |pages=169–183 |doi=10.1016/j.jqsrt.2012.11.014 |issn=0022-4073|hdl=2060/20140010884 |s2cid=11722624 |hdl-access=free |url-access=subscription }} It is more powerful than EBCM as it is more efficient and increases the upper limit of particle size during the computation.

References

Category:Computational physics

Category:Electromagnetism

Category:Electrodynamics

Category:Scattering, absorption and radiative transfer (optics)

Category:Computational electromagnetics