INVESTIGATION OF ADIABATIC WAVEGUIDE MODES MODEL FOR SMOOTHLY IRREGULAR INTEGRATED OPTICAL WAVEGUIDES Текст научной статьи по специальности «Медицинские технологии»

Discrete & Continuous Models

#& Applied Computational Science 2022, 30 (2) 149-159

ISSN 2658-7149 (online), 2658-4670 (print) http://journals-rudn-ru/miph UDC 535:535.3:681.7

DOI: 10.22363/2658-4670-2022-30-2-149-159

Investigation of adiabatic waveguide modes model for smoothly irregular integrated optical waveguides

Anton L. Sevastyanov

Higher School of Economics, 11, Pokrovsky Bulvar, Moscow, 109028, Russian Federation

(received: March 18, 2022; revised: April 18, 2022; accepted: April 19, 2022)

Abstract. The model of adiabatic waveguide modes (AWMs) in a smoothly irregular integrated optical waveguide is studied. The model explicitly takes into account the dependence on the rapidly varying transverse coordinate and on the slowly varying horizontal coordinates. Equations are formulated for the strengths of the AWM fields in the approximations of zero and first order of smallness. The contributions of the first order of smallness introduce depolarization and complex values characteristic of leaky modes into the expressions of the AWM electromagnetic fields. A stable method is proposed for calculating the vertical distribution of the electromagnetic field of guided modes in regular multilayer waveguides, including those with a variable number of layers. A stable method for solving a nonlinear equation in partial derivatives of the first order (dispersion equation) for the thickness profile of a smoothly irregular integrated optical waveguide in models of adiabatic waveguide modes of zero and first orders of smallness is described. Stable regularized methods for calculating the AWM field strengths depending on vertical and horizontal coordinates are described. Within the framework of the listed matrix models, the same methods and algorithms for the approximate solution of problems arising in these models are used. Verification of approximate solutions of models of adiabatic waveguide modes of the first and zero orders is proposed; we compare them with the results obtained by other authors in the study of more crude models.

Key words and phrases: smoothly irregular thin-film dielectric waveguides, adiabatic waveguide modes, regularized methods for calculating field strengths

1. Introduction

The adiabatic waveguide propagation of optical radiation was previously described in optical fibers using the method of cross sections in the papers by B. Z. Katsenelenbaum [1], V. V. Shevchenko [2], M. V. Fedoruk [3], and in integrated optical waveguides using the method of adiabatic waveguide modes — in the papers by A. A. Egorov, L. A. Sevastyanov and their coauthors [4]-[6]. In the papers by A. L. Sevastyanov [7], [8], the model of adiabatic waveguide modes was substantiated.

It should be noted that in the last decade there has been an interest in the adiabatic waveguide propagation of electromagnetic radiation for the study

© Sevastyanov A.L., 2022

This work is licensed under a Creative Commons Attribution 4.0 International License http://creativecommons.org/licenses/by/4.0/

of coherent quantum effects in atomic, molecular or condensed matter systems. These effects are difficult to investigate because of dephasing effects or fast temporal dynamics. Optical Bloch oscillations [9], quantum-mechanical analogy of dynamic mode stabilization and radiation loss suppression [10], quantum enhancement and suppression of tunneling in directional optical couplers [11], [12], as well as Landau-Zener tunneling in coupled waveguides [13] can serve as optical models of coherent quantum effects. An interesting example is the three-level system with stimulated Raman adiabatic passage (STIRAP), which vividly illustrates counterintuitive quantum effects [14]-

[19].

2. Model of adiabatic waveguide modes in a multilayer

waveguide

Let us specify the class of integrated optical waveguides to be considered and the electromagnetic radiation propagating through them.

1. Electromagnetic radiation is polarized, monochromatic with a given wavelength A e [380; 780], nm.

2. The thickness of the guiding layer of the base thin-film waveguide is comparable to the wavelength of the propagating monochromatic electromagnetic radiation d ~ A.

3. The surface of the additional guiding layer (x = h(y,z)) satisfies the

following restrictions 4. The integrated optica

dh dh

dy' dz

hkn

« 0

Ap

kn «û ■

2n

waveguide is a material medium consisting of dielectric subregions, which together fill the entire three-dimensional space.

5. The permittivities of the subregions are different and real-valued, and the permeability is everywhere equal to that of vacuum.

6. There are no external currents and charges. Therefore, in the absence of foreign currents and charges, the induced currents and charges are zero.

7. The Cartesian coordinate system is introduced as follows: the interfaces between the dielectric media of the basic three-layer waveguide are parallel to the yOz plane. The subdomains of the space corresponding to the cover and substrate layers are infinite; the additional guiding layers are asymptotically parallel to the yOz plane. Therefore, e = e(x).

In Cartesian coordinates associated with the geometry of the substrate (or a three-layer planar dielectric waveguide underlying a smoothly irregular integrated optical waveguide), with the introduced restrictions taken into account, the Maxwell equations have the form

dHz dHy e dEx dEz dEy pdHx

dy dz c dt ' dy dz c dt

dHx dHz edEy dEx dEz ßdHy

dz dx c dt ' dz dx c dt

dHy dHx edEz dEy dEx ^ dHz

dx dy c dt ' dx dy c dt

Note that variable x is fast, and variables y, z are slow with respect to the small dimensioned parameter 1/u. The approximate solutions to the Maxwell equations (1) within the asymptotic method [20], [21], with the separation of slow and fast variables taken into account are sought in the form

T^r \ \—^ Es (x; y, z) r. , . >

E(x, y, z,t) = 2_^ , . exp {iut — ik0 ip(y, z)} ,

H(x,y,z,t) = ^

=0 (—iu) Hs (x;y,z)

=0 (-M7+s

exp [iut — ik0p(y, z)} .

(2) (3)

Keeping in the solution (2), (3) the terms of the zero and first order of smallness leads to the model of adiabatic waveguide modes (AWMs) that describes the guided-wave propagation of a polarized optical radiation through irregular segments of smoothly irregular (multilayer) optical waveguides. In regular parts, the adiabatic waveguide modes become normal modes of a regular planar optical waveguide.

In the notation Es (x;y,z), Hs (x;y,z), the separation by a semicolon means the following assumptions:

and

dÊs (x; y, z) dÊs(x;y,z) 1 dÊs (x; y, z)

dy , dz u dx

dHs ( x;y,z) dHs (x; y, z) 1 dHs(x;y,z)

dy , dz u dx

(4)

(5)

for each s, where |||| is the Hilbert norm of functions of x, and u is the circular frequency of the propagating monochromatic electromagnetic radiation.

2.1. AWM model equations in the zero-order approximation

In Ref. [7] it was shown that the zero-order approximation (within the asymptotic approach) of the waveguide solution to the Maxwell equations is given by the following relations:

E(x,y,z,t) H(x,y,z,t)

Eo (x;y,z) H0 (x;y,z)

exp [iut — iip(y, z)} ,

(6)

with

=-0 (I

dx

Hy — ik

qz I u0 lK0

= lk d x = 1 k0

(7)

and

as well as

^ (^)(SH + * H*)2)* (9)

№ = {">-(%))* ^ (M)(%)E* (10) * = + & ■ 'H)

»i = & (12) + (li(!J'z))2 = "2ff (!/'z)- (13)

For a thin-film multilayer waveguide consisting of optically homogeneous layers, the conditions for matching the electromagnetic field at the interfaces between the media are valid, namely

nxE- + nxÊ+ =0, (14)

nxH- + nxH + = 0. (15)

In addition, the asymptotic conditions

E°y, E°z, H0, H°z -► 0 (16)

are fulfilled.

The system of Eqs. (7)-(10), (16) for any fixed (y, z) defines the problem of

—*■ 2 ■ T

finding eigenvalues (Vp) (y,z) and eigenfunctions (EJy,EJZ,Hl,HJz) (y,z), normalized to unity:

TO TO

I \E3y|2dx = l, I \H3y\2dx = l. (17)

2.2. AWM model equations in the first approximation

We continue to apply the approach based on the small parameter expansion and arrive at the system of equations in the first approximation of the method:

dEî ik0 dip (dip ^z dip y] y

+ (iZ-Hï -ifM ) + lkorn =

dx e dz \ dy dz

dEg iu dip (dHf a™*

dz e dz

= ^^(-i-^), (18)

dE\

г%д<р ( chp HZ _ dp Hy

£ ду \ ду 1 д z 1

) + г k0fi Щ =

= _ги

öeq

ду

( дЩ _ дЩ )

\ дz ду J '

(19)

дЩ

+ г%д<р

_ ^Ez dz 1 ду 1

_ гк0еЕ\ =

=

дЩ

гшдр (дЕ%

д z \ д z

дЩ ду

У

(20)

^ _10 ' _%Е î Ь ^ =

= _ . дЩ_ iwdipi dEl д Eq

*_ m (22)

), (21)

I (&PHZ — fy jjv) = (— dHçz

£ \ dy 1 dz 1 ) £ k0 \ dz dy

*—gm—f)- (23)

The system of zero order equations (7)-(12) coincides with the system of equations (18)-(23), if in the latter we put zero into the right-hand sides (the contributions with zero-order quantities).

Substituting the solutions of system (7)-(12) into the right-hand sides of equations (18)-(23) leads to the following form of expressions for electromagnetic fields in the first (plus zero) approximation

E(x;y,z) = E{0(x;y, z) + -Ex(x;y,z),

H(x; y, z) = H0(x; y, z) + -Hi(x; y, z).

u

These fields are necessarily complex-valued. Thus, the contributions of the first order of smallness introduce into the expressions for the AWM electromagnetic fields the characteristic features of leaky modes.

3. Implementation of numerical experiment

In Ref. [22], an hierarchy of mathematical models for the adiabatic waveguide propagation of optical radiation in integrated optical waveguides was proposed. The AWM model consists in representing the electromagnetic field in the form (6). The dependences of the field strengths on the fast variable have the form (7)-(12) in the zero approximation and (18)-(23) in the

first approximation. Of course, the rigging conditions (13)-(17) of the AWM mathematical model are assumed to be fulfilled.

3.1. Algorithm for calculating the AWM electromagnetic field

A. Stage 1: reconstructing the dependence of the AWM electromagnetic field on the fast variable at fixed values of the slow variables

1. Solve the system (7)-(12) for E0, H0 describing the AWM model in the zero order of smallness in 1/u, rigged with (6), (18)-(23) using the method, asymptotic with respect to 5, to obtain systems for contributions of different orders of smallness with respect to 5.

2. Solve the system (13)-(17) for E1, H1 describing the AWM model in the first order of smallness in 1/u, rigged with (6), (18)-(23) using the method, asymptotic with respect to 5, to obtain systems for contributions of different orders of smallness with respect to 5.

B. Stage 2: reconstructing the dependence of the AWM electromagnetic field on the slow variables.

In Ref. [7] it is shown how the general solutions of the system of ODEs (7)-(12) and (13)-(17), represented in the form of expansion in the

fundamental system of solutions with indefinite coefficients (A, B) , can be reduced to a homogeneous system of linear algebraic equations (SLAE) with respect to these indefinite coefficients using the conditions (14)-(16).

3. Implement stable methods of approximate solutions of the homogeneous SLAE

M0 [(z, y), h(z, y), ip(z, y), V<p(z, y)](A0(z, y), B0 (z, y))T = (0, of, (24) satisfying the conditions

detjM0 } [(z, y), h(z, y), <p(z, y), V<p(z, y)] = 0. (25)

4. Implement stable methods of approximate solutions of the homogeneous SLAE

M1 [(z,y),h(z,y),<p(z,y),V<p(z,y)] (A1 (z,y),B1 (z,y))T = (0,0) T (26) satisfying the conditions

detjM1} [(z,y),h(z,y),ip(z,y),Vip(z,yj] =0. (27)

In both cases, the solution for the field strengths depending on the fast variable x for a fixed value of the slow variables y, z makes it possible, using the rigging (6), (18)-(23), to find the dependence of the AWM electromagnetic field for all values of the slow variables (see, e.g., Ref. [8]).

Homogeneous systems of linear algebraic equations (24) and (26) are uniquely solvable under conditions (25) and (27). In both cases, these equations with respect to the derivative Vp(z, y) are partial differential equations

of the form

F0 (v<p(z, y); h(z, y), Vh(z, y)) = 0 (28)

and

F1 (Vp(z, y); h(z, y), Vh(z, y)) = 0. (29)

5. Solve Eqs. (28) and (29) numeric-symbolically using the Cauchy method (see, e.g. [23], [24]).

6. For each V^(z,y) calculate (z,y,V^(z,y)) ,B0 (z,y,V^(z,y))) using the Tikhonov regularization method, which consists in minimizing the Nelder-Mead functional:

T

F0 (0) = M0 \(z, y), h(z, y), <p(z, y), V<p(z, y)\ (A0 (z, y), B0 (z, y)

+a

(Â0 (z, y)-Â0 (z -Az,y- Ay)) ,(B0 (z, y)-B() (z - Az,y - Ay)))

+

T

^ Stage 3: verifying the obtained numerical results and AWM models of the first and zero orders of smallness.

The validation of the asymptotic method of constructing AWM models is carried out by comparing solutions E1, H1 and E0, H0.

The formulation of the third condition from the set of conditions 1-7 implic-

lA^I

itly implies the presence of the second small parameter ô = max —¡-^—¡- < 1

y>z k0 |V<p|

(see the beginning of the first section).

To verify the obtained approximate solutions of the zero-order model of adiabatic modes, we compare them with the results obtained by other authors using more crude models:

— matrix model of adiabatic modes in the approximation of horizontal boundary conditions (a stepped set of plates for a Luneburg thin-film generalized waveguide lens)

Such configurations are impossible in optical fibers and can be implemented in the case of adiabatic waveguide propagation of a nonparallel (converging or diverging) 2D beam of rays, normal to a nonplanar (2D) wave front.

— matrix model of comparison waveguides (passing to the horizontal boundary conditions + replacement — 0, — 0).

Thus, three levels of making the AWM model cruder were used.

2

4. Discussion and conclusion

In the paper, we consider three levels of making the AWM model cruder:

— replacing the first-order AWM model with the zero-order one;

— replacing the tangential boundary conditions with the horizontal ones — the matrix model still having no name;

— replacing the tangential boundary conditions with the horizontal ones and —y 0, — P — the matrix model of comparison waveguides.

Two latter approximations have been used by other authors.

Within the listed matrix models, similar methods and algorithms are used for the approximate solution of problems, arising in the models. The method of studying the matrix model of adiabatic waveguide modes in the zero and first approximation of a smoothly irregular multilayer integrated optical waveguide is proposed for the first time. It allows to grade the crudeness of the approximate models used by other authors and approximate solutions in the adiabatic mode models of different order of smallness.

References

[1] B. Z. Katsenelenbaum, Theory of irregular waveguides with slowly varying parameters [Teoriya neregulyarnyh volnovodov s medlenno menyayushchimisya parametrami]. Moscow: Akad. Nauk SSSR, 1961, in Russian.

[2] V. V. Shevchenko, Continuous transitions in open waveguides [Plavnyye perekhody v otkrytykh volnovodakh]. Moscow: Nauka, 1969, in Russian.

[3] M. V. Fedoryuk, "Justification of the method of cross-sections for an acoustic waveguide with inhomogeneous filling", USSR Computational Mathematics and Mathematical Physics, vol. 13, no. 1, pp. 162-173, 1973. DOI: 10.1016/0041-5553(74)90012-3.

[4] A. A. Egorov and L. A. Sevast'yanov, "Structure of modes of a smoothly irregular integrated optical four-layer three-dimensional waveguide", Quantum Electronics, vol. 39, no. 6, pp. 566-574, 2009. DOI: 10.1070/ QE2009v039n06ABEH013966.

[5] A. A. Egorov et al., "Simulation of guided modes (eigenmodes) and synthesis of a thin-film generalised waveguide Luneburg lens in the zero-order vector approximation", Quantum Electronics, vol. 40, no. 9, pp. 830-836, 2010. DOI: 10.1070/QE2010v040n09ABEH014332.

[6] A. A. Egorov, L. A. Sevastianov, and A. L. Sevastianov, "Method of adiabatic modes in research of smoothly irregular integrated optical waveguides: zero approximation", Quantum Electronics, vol. 44, no. 2, pp. 167-173, 2014. DOI: 10.1070/QE2014v044n02ABEH015303.

[7] A. L. Sevastianov, "Asymptotic method for constructing a model of adiabatic guided modes of smoothly irregular integrated optical waveguides", Discrete and Continuous Models and Applied Computational Science, vol. 20, no. 3, pp. 252-273, 2020. DOI: 10.22363/2658-4670-2020-283-252-273.

[8] A. L. Sevastianov, "Single-mode propagation of adiabatic guided modes in smoothly irregular integral optical waveguides", Discretee and Continuous Models and Applied Computational Science, vol. 28, no. 4, pp. 361377, 2020. DOI: 10.22363/2658-4670-2020-28-4-361-377.

[9] G. Lenz, I. Talanina, and C. M. de Sterke, "Bloch oscillations in an array of curved optical waveguides", Physical Review Letters, vol. 83, no. 5, pp. 963-966, 1999. DOI: 10.1103/PhysRevLett.83.963.

[10] S. Longhi, D. Janner, M. Marano, and P. Laporta, "Quantum-mechanical analogy of beam propagation in waveguides with a bent axis: dynamic-mode stabilization and radiation-loss suppression", Physical Review E, vol. 67, no. 3, p. 036 601, 2003. DOI: 10.1103/PhysRevE. 67.036601.

[11] I. Vorobeichik et al., "Electromagnetic realization of orders-of-magnitude tunneling enhancement in a double well system", Physical Review Letters, vol. 90, p. 176 806, 17 2003. DOI: 10.1103/PhysRevLett.90.176806.

[12] S. Longhi, "Coherent destruction of tunneling in waveguide directional couplers", Physical Review A, vol. 71, p. 065 801, 6 2005. DOI: 10.1103/ PhysRevA.71.065801.

[13] R. Khomeriki and S. Ruffo, "Nonadiabatic Landau-Zener tunneling in waveguide arrays with a step in the refractive index", Physical Review Letters, vol. 94, p. 113 904, 11 2005. DOI: 10.1103/PhysRevLett. 94. 113904.

[14] K. Bergmann, H. Theuer, and B. W. Shore, "Coherent population transfer among quantum states of atoms and molecules", Reviews of Modern Physics, vol. 70, pp. 1003-1025, 3 1998. DOI: 10.1103/RevModPhys.70. 1003.

[15] F. T. Hioe and J. H. Eberly, "N-Level coherence vector and higher conservation laws in quantum optics and quantum mechanics", Physical Review Letters, vol. 47, pp. 838-841, 12 1981. DOI: 10.1103/PhysRevLett.47. 838.

[16] J. Oreg, F. T. Hioe, and J. H. Eberly, "Adiabatic following in multilevel systems", Physical Review A, vol. 29, pp. 690-697, 2 1984. DOI: 10.1103/ PhysRevA.29.690.

[17] C. E. Carroll and F. T. Hioe, "Three-state systems driven by resonant optical pulses of different shapes", Journal of the Optical Society of America B: Optical Physics, vol. 5, no. 6, pp. 1335-1340, 1988. DOI: 10.1364/J0SAB.5.001335.

[18] J. Oreg, K. Bergmann, B. W. Shore, and S. Rosenwaks, "Population transfer with delayed pulses in four-state systems", Physical Review A, vol. 45, pp. 4888-4896, 7 1992. DOI: 10.1103/PhysRevA.45.4888.

[19] N. V. Vitanov and S. Stenholm, "Analytic properties and effective two-level problems in stimulated Raman adiabatic passage", Physical Review A, vol. 55, pp. 648-660, 1 1997. DOI: 10.1103/PhysRevA.55.648.

[20] V. M. Babich and V. S. Buldyrev, Asymptotic methods in short-wavelength diffraction theory (Alpha Science Series on Wave Phenomena), English. Harrow, UK: Alpha Science International, 2009.

[21] Y. A. Kravtsov and Y. I. Orlov, Geometrical optics of inhomogeneous media. Berlin: Springer-Verlag, 1990.

[22] A. L. Sevastyanov, "Single-mode waveguide spread of light in a smooth irregular integral optical waveguide [Komp'yuternoe modelirovanie polej napravlyaemyh mod tonkoplenochnoj obobshchennoj volnovodnoj linzy Lyuneberga]", in Russian, Ph.D. dissertation, Peoples' Friendship University of Russia, Moscow, 2010.

[23] M. D. Malykh, "On integration of the first order differential equations in a finite terms", Journal of Physics: Conference Series, vol. 788, p. 012 026, 2017. DOI: 10.1088/1742-6596/788/1/012026.

[24] A. D. Polyanin and V. E. Nazaikinskii, Handbook of linear partial differential equations for engineers and scientists, 2nd ed. Boca Raton, London: CRC Press, 2016. DOI: 10.1201/b19056.

For citation:

A. L. Sevastyanov, Investigation of adiabatic waveguide modes model for smoothly irregular integrated optical waveguides, Discrete and Continuous Models and Applied Computational Science 30 (2) (2022) 149-159. DOI: 10.22363/2658-4670-2022-30-2-149-159.

Information about the authors:

Sevastyanov, Anton L. — PhD in Physical and Mathematical Sciences, Deputy head of department: Department of Digitalization of Education (e-mail: [email protected], phone: +7(495)772-95-90 (28571), ORCID: https://orcid.org/0000-0002-0280-485X)

УДК 535:535.3:681.7

DOI: 10.22363/2658-4670-2022-30-2-149-159

Исследование модели адиабатических волноводных мод для плавно-нерегулярных интегрально-оптических волноводов

А. Л. Севастьянов

Национальный исследовательский университет «Высшая школа экономики», Покровский бульвар, д. 11, Москва, 109028, Россия

Аннотация. Проведено исследование модели адиабатических волноводных мод плавно-нерегулярного интегрально-оптического волновода. В модели явно учтена зависимость от быстропеременной поперечной координаты и от медленно-переменных горизонтальных координат. Сформулированы уравнения для напряженностей полей АВМ в приближениях нулевого и первого порядка малости. Вклады первого порядка малости вносят в выражения электромагнитных полей АВМ деполяризацию и комлекснозначность, т.е. характерные черты вытекающих мод. Предложен устойчивый метод вычисления вертикального распределения электромагнитного поля направляемых мод регулярных многослойных волноводов, в том числе с переменным числом слоев. Описан устойчивый метод решения нелинейного уравнения в частных производных первого порядка (дисперсионного уравнения) для профиля толщины плавно-нерегулярного интегрально-оптического волновода в моделях адиабатических волноводных мод нулевого и первого порядков малости. Описаны устойчивые ре-гуляризованные методы вычисления напряженностей полей АВМ в зависимости от вертикальных и горизонтальных координат. В рамках перечисленных матричных моделей используются одинаковые методы и алгоритмы приближенного решения задач, возникающих в этих моделях. Предложена верификация приближенных решений моделей адиабатических волноводных мод первого и нулевого порядков; проведено сравнение их с результатами других авторов, полученных при исследовании более грубых моделей.

Ключевые слова: модели квантовых измерений, возмущение дискретного спектра, комплексные собственные значения, пучки операторов