Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Spatiotemporal coupled-mode theory of guided-mode resonant gratings

Open Access Open Access

Abstract

In this paper, we develop spatiotemporal coupled-mode theory to describe optical properties of guided-mode resonant gratings. We derive partial differential equations that describe both spatial and temporal evolution of the field inside the grating. These equations describe the coupling of two counter-propagating grating modes, revealing the structure’s “dark” and “bright” resonances at normal incidence of light. Moreover, the proposed theory allows us to obtain a simple approximation of the transmission and reflection coefficients taking into account both light’s frequency and angle of incidence. This approximation can be considered as the generalization of the Fano line-shape. The approximation is in good agreement with the rigorous computations based on the Fourier modal method. The results of the paper will be useful for design and analysis of guided-mode resonant filters and other photonic devices.

© 2015 Optical Society of America

1. Introduction

The resonance phenomenon in optical structures drew much attention in the past decade [1–3]. Resonant photonic structures found numerous application as optical filters, sensors, laser resonators, nonlinear optical devices, (magneto-)optical modulators and switches [1]. Optical properties of such structures are usually characterized by the resonance lineshape representing a dependence of the scattering amplitude on the light frequency. The coupled-mode theory provides an intuitive approach for deriving the resonance lineshapes.

Different coupled-mode theories were developed to describe various optical structures including coupled waveguides and fibers [4–6], Bragg reflectors in waveguides [4, 5], coupled resonators [4, 5], waveguide-coupled resonators [7–9], gratings [1, 8, 10], individual scatterers [11–13]. The referenced papers use either the spatial formulation of the coupled-mode theory [4–6] or the temporal one [4, 5, 7–13]. Spatial theories describe spatial evolution of the propagating mode, while temporal coupled-mode theories deal with temporal evolution of optical field. A generalization of these two approaches is the time-dependent coupled-mode theory [14–16] or the spatiotemporal coupled-mode theory [17]. This theory, considering both spatial and temporal evolution of optical fields, was used to investigate parallel waveguides [14,16] and various photonic-crystal devices [15].

In the current paper we develop a new formulation of the spatiotemporal coupled-mode theory to describe resonant light scattering by guided-mode resonant gratings. The paper is organized as follows. Following the Introduction, in Section 2 we derive partial differential equations that describe both temporal and spatial field evolution inside the grating. In Section 3 we use these equations to obtain approximation of the transmission and reflection coefficients of the structure as a function of both frequency and in-plane wave vector. In Section 4 we compare the results of the coupled-mode theory with the rigorous computations. Besides, we show that the proposed approach describes a number of important optical effects: mode coupling, emergence of a photonic band gap, emergence of even and odd modes at normal incidence of light.

2. Spatiotemporal coupled-mode theory

Let us consider an optical pulse diffraction by a guided-mode resonant grating (Fig. 1). We assume that the structure and the incident field are invariant along the y-axis. For the sake of simplicity we consider the symmetric grating with yOz symmetry plane. We also assume that the grating period is subwavelength, i.e. the grating supports only zeroth diffraction orders (reflected and transmitted). However, inside the waveguide layer several diffraction orders exist.

 figure: Fig. 1

Fig. 1 Diffraction of a pulse on the guided-mode resonant grating (waveguide with a grating on its top) supporting two quasiguided modes.

Download Full Size | PDF

We start our analysis from replacing a grating with a slab waveguide (with an effective refractive index). In this case the field inside a waveguide is a superposition of two uncoupled counter-propagating modes. We suppose that at the angular frequency ω0 the mode wave number is kx0. Moreover, we suppose the following linear dispersion law for this mode:

ω=ω0+vg(kxkx0),
where vg = dω/dkx is a mode group velocity.

Now let us consider a mode wave packet (pulse). Its field distribution can be written as

u(x,t)=P(xvgt)ei(kx0xωot),
where P is a wave packet envelope and ω0 is a wave packet central (carrier) angular frequency. By direct substitution one can easily verify that Eq. (1) satisfies the following partial differential equation:
ut=vgux+i(kx0vgω0)u.

Note that the first part of this equation (ut=vgux) is a simple unidirectional wave equation for a wave packet envelope P. While the last term in Eq. (2) takes account of the phase of the wave packet.

Let us consider two wave packets with opposite directions [see Fig. 2(a)]:

{ut=vgux+i(kx0vgω0)u;vt=vgvx+i(kx0vgω0)v,
where u = u(x,t) and v = v(x,t) are the amplitudes of the mode wave packets that propagate in the positive and negative directions of the x-axis.

 figure: Fig. 2

Fig. 2 (a) Two counter-propagating modes in a slab waveguide. (b) Modes excitation by ±m-th diffraction orders. (c) Mode leakage into −m-th diffraction order and mode coupling by means of 2m-th diffraction order.

Download Full Size | PDF

Now let us add a grating on the top of the waveguide (Fig. 1). Moreover, we assume that grating period d satisfies the mode excitation condition

kx0=2πm/d.

For the incident pulse central frequency ω0 this condition provides matching of the propagation constants of the mode and of the ±m-th diffraction orders. In this case a number of new effects will arise. First, due to the diffraction on the grating grooves, at every moment of time an α0-fraction of the mode will scatter away from the waveguide. Second, an extra energy can be introduced into the waveguide from the incident field. Third, a coupling between two counter-propagating modes will take place (i.e. energy from a mode will be transferred to the counter-propagating one). All these effects are sketched in Figs. 1 and 2.

The first two effects will modify Eqs. (3) in the following way:

{ut=vgux+i(kx0vgω0)uα0u+qf(x,t)eikxox;vt=vgvx+i(kx0vgω0)uα0v+qf(x,t)eikx0x,
where f (x,t) is the incident field distribution at the upper interface of the grating; q is the free-space-to-mode coupling coefficient; e±ikx0x is a multiplier that takes account of the incident field scattering into the ±m-th diffraction orders [see Fig. 2(b)] [4].

The third effect caused by the grating, coupling between modes, can be taken into account by introducing two coupling coefficients, c1 and c2:

{ut=vguxα1uc1u+c2e2ikx0xv+qf(x,t)eikx0x;vt=vgvxα1vc1v+c2e2ikx0xu+qf(x,t)eikx0x;
where α1 = α0 i(kx0vg −ω0). The exponential terms in Eq. (5) are of the same nature as in Eq. (4): the mode inside the waveguide couples to the counter-propagating mode by means of diffraction into 2m-th diffraction order [see Fig. 2(c)].

The energy leaked out of the grating forms the resonant contribution to the reflected and transmitted fields. The non-resonant part of the scattered field is the field transmitted/reflected directly through the structure without the mode excitation (see Fig. 1). Hence the transmitted and reflected field can be defined as

fT(x,t)=t˜f(x,t)+γ1s(x,t);fR(x,t)=r˜f(x,t)+γ2s(x,t),
where r˜ and t˜ are complex non-resonant reflection and transmission coefficients; γ1,2 are mode-to-free-space coupling coefficients; γ1,2·s(x,t) is the scattered field distribution, where
s=ueikx0x+veikx0x.

The exponential terms here take account of the fact that the modes are scattered out of the waveguide by means of ∓m-th diffraction orders [see Fig. 2(c)].

Equations (5)(7) define the spatiotemporal coupled-mode theory for guided-mode resonant gratings.

3. Transmission and reflection coefficients

In what follows we will analyze Eqs. (5)(7) to obtain the analytical expressions for the reflection and transmission coefficients of the grating. To do this, we need to solve Eq. (5) with respect to the total scattered field (7). We first rewrite Eq. (5) in terms of functions u˜=ueikx0x and v˜=veikx0x:

{u˜t=vgu˜xαu˜+c2v˜+qf(x,t);v˜t=vgv˜xαv˜+c2u˜+qf(x,t),
where α = α0 +c1 +iω0. Then we consider the sum and the difference of equations (8) in terms of s=u˜+v˜ and w=u˜v˜:
st=vgwx(αc2)s+2qf(x,t);
wt=vgsx(α+c2)w.

Finally, we eliminate wx from Eq. (9) and substitute it into Eq. (10) differentiated with respect to x. This leads us to the following partial differential equation for s:

2st2+2αst=vg22sx2(α2c22)s+g(x,t),
where the inhomogeneous part in the right-hand side of the equation is given by
g(x,t)=2q[f(x,t)t+(α+c2)f(x,t)].

Equations (6) and (11) relate the incident and the scattered fields. These equation are formulated in spatiotemporal domain. Let us rewrite these equations in the frequency domain. To do this, we define the incident field spectrum as the Fourier transform of the incident field:

F(kx,ω)=f(x,t)ei(kxxωt)dxdt.

The reflected and transmitted field spectra, FR and FT, are defined in a similar way. Besides, we define the scattered field spectrum S as the Fourier transform of s(x,t).

By applying the Fourier transform to Eq. (11) we obtain the following relation between the incident and scattered field spectra:

ω2S2iωαS=vg2kx2S(α2c22)S+2q[iω+(α+c2)]F.

Fourier transform of Eq. (6) gives

FT=t˜F+γ1S;FR=r˜F+γ2S.

Let us consider the incident field to be a monochromatic plane wave with angular frequency ω and in-plane wave vector kx. In this case the complex transmission coefficient is the ratio of the transmitted and the incident field spectra [T(kx) = FT(kx)/F(kx)]. A similar expression is valid for the reflection coefficient. According to Eqs. (13) and (12), the transmission coefficient T and the reflection coefficient R are given by.

T=FTF=t˜+γ1SF;R=FRF=r˜+γ2SF,
where
SF=2iqω+i(α+c2)vg2kx2(ω+iα)2c22.

By substituting Eq. (15) into Eq. (14) we obtain the following form of the reflection and transmission coefficients:

T(kx,ω)=t˜vg2kx2(ωωt)(ωωp2)vg2kx2(ωωp1)(ωωp2);R(kx,ω)=r˜vg2kx2(ωωr)(ωωp2)vg2kx2(ωωp1)(ωωp2)
where ωp1 = ω0 i(α0 + c1 − c2), ωp2 = ω0 i(α0 + c1 + c2), ωt=ω0i(α0+c1c2+2qγ1t˜1), ωr=ω0i(α0+c1c2+2qγ2r˜1). According to Eq. (16) there are eight independent parameters that define the spatiotemporal transmission and reflection spectra. Three parameters, vg, ω0 and α0 + c1, are non-negative real numbers, while c2,1,2, r˜, t˜ are complex numbers. The number of independent parameters can be reduced if the structure is lossless or has additional symmetries [4,10].

Equation (16) can be considered as a generalization of the Fano lineshape. Indeed, at fixed ω the transmission and reflection coefficients take the form of two-resonance Fano lineshape [18] a + b/(kx − kp) − b/(kx + kp). At fixed kx Eq. (16) can be rewritten in a similar two-resonance form [8]: a + b/(ω −ω1) + c/(ω −ω2). Note that at kx = 0 (at normal incidence) the transmission coefficient has only one resonance with frequency ωp1. The physical origin of this effect is that the second mode with frequency ωp2 cannot be excited due to the symmetry conditions [10]. At normal incidence this mode has an odd (antisymmetric) field distribution with respect to the plane yOz and cannot be excited by the symmetric incident plane wave. Note that the second resonance with frequency ωp1 corresponds to the even (symmetric) mode that can be excited.

4. Numerical validation

To check the validity of the proposed theory we considered guided-mode resonant filter (Fig. 1) with the following parameters: grating period d = 320nm, grating fill-factor a = 70/d, grating height h1 = 50nm, waveguide layer height h2 = 200nm, superstrate refractive index ns = 1, grating and waveguide layer refractive index n2 = 1.7, substrate refractive index n1 = 1.5. The presented parameters are typical for guided modes resonant filters. This filter was designed to exhibit pronounced peaks in the visible spectrum range for TE-polarized light.

We used the Fourier modal method [19, 20] to simulate the diffraction of a plane wave on the considered structure. Figure 3(a) shows the transmission coefficient as a function of both ω and kx. Let us note that due to the grating symmetry, T(ω,+kx) = T(ω, −kx). For the sake of comparison, the simulated spectrum is presented on the left-hand side of the figure (at kx < 0) while the approximating spectrum obtained from Eq. (16) is presented on the right (at kx > 0). Figure 3(b) presents the simulated and approximating transmission spectra at different angles of incidence. Circles in Fig. 3(b) presents the transmission coefficient obtained from Eq. (16) with kx = (ω/c)ns sinθ.

 figure: Fig. 3

Fig. 3 (a) Transmission coefficient of the grating, |T|2, vs. incident light’s kx and ω: rigorous simulation (left part, at kx < 0) and analytical formula (16) (right part, at kx > 0). The dashed lines specify the light lines for θ = 0.2°,1°. (b) Transmission coefficient of the grating, |T|2, for different angles of incidence (θ = 0°,0.2°,1°): rigorous simulation (solid lines) and analytical formula (16) (circles).

Download Full Size | PDF

We used the following parameters in Eq. (16): vg = 1.768 1017 nms1, ω0 = 3.751×1015 s1, α0 +c1 = 8.694×1011 s1, c2 = −8.694×1011+6.968×1012 is1,1 = (−2.023+8.181i)×1011,2 = (−5.408−3.706i)×1011, r˜=0.16200.1923i, t˜=0.28270.9257i. To estimate these parameters we used the following approach. First, we calculated ωp1 and ωp2 as the pole of the scattering matrix at kx = 0 using the scattering matrix approach [21]. Then, we calculated ωr and ωt as the zeros of the reflection and transmission coefficients at kx = 0 using the Fourier modal method [19,20]. Using the calculated parameters we obtained c2, ω0, α0 +c1, qγ1t˜1, qγ2r˜1. Finally, the remaining parameters (vg, r˜, t˜) were estimated using the conjugate gradient method aimed at minimizing the root-mean-square deviation between the rigorously computed transmission spectrum [left part of Fig. 3(a)] and approximate relation (16).

Figure 3 shows pronounced transmission minimum governed by the quasi-guided mode excitation. According to Fig. 3(a), the dispersion law of the mode resembles a hyperbola. The latter’s equation can be obtained from analytical Eq. (16) by equating to zero its denominator. As we discussed in the previous section, in the case of oblique incidence there are two resonances at fixed angle of incidence. However, in the case of normal incidence only one resonance is present in Fig. 3(a) and Fig. 3(b). The second resonance corresponds to the antisymmetric mode and cannot be excited by a normally incident plane wave.

According to Fig. 3, the spatiotemporal coupled-mode theory provides a good approximation of the transmission spectrum at different angles of incidence. For the considered frequency range and |θ| < 0.2° the approximation error |T1 − T2|2 is less than 0.017; when |θ| < 1° and |θ| < 2° the error does not exceed 0.018 and 0.035, respectively.

5. Conclusion

We presented the spatiotemporal coupled-mode theory for guided-mode resonant grating. The theory describes resonant light scattering using two coupled partial differential equations (unidirectional wave equations for dispersive medium). By solving these equations we obtained analytical approximation of the transmission and refection spectra of the grating. This approximation takes into account both frequency and the in-plane wave vector of the incident light. The obtained formulae are in good agreement with rigorous simulations, describing such grating-theory phenomena as mode coupling, emerging of “dark” and “bright” resonances at normal incidence of light. The theory can be generalized to the case of conical mount, as well as to the case of crossed grating. In the latter case four coupled modes can be excited simultaneously.

The presented analytical expressions for the reflection and transmission coefficients can be used for the design of a wide range of photonic devices, such as filters, lasers, sensors. Besides, the developed theory provides theoretical description of the general class of spatiotemporal transformations of optical fields that can be implemented by resonant periodic structures.

Acknowledgments

The work was funded by the Russian Science Foundation grant 14-19-00796.

References and links

1. W. Zhou, D. Zhao, Y.-C. Shuai, H. Yang, S. Chuwongin, A. Chadha, J.-H. Seo, K. X. Wang, V. Liu, Z. Ma, and S. Fan, “Progress in 2D photonic crystal Fano resonance photonics,” Prog. Quantum Electron 38, 1–74 (2014). [CrossRef]  

2. A. E. Miroshnichenko, S. Flach, and Y. S. Kivshar, “Fano resonances in nanoscale structures,” Rev. Mod. Phys. 82, 2257–2298 (2010). [CrossRef]  

3. S. Collin, “Nanostructure arrays in free-space: optical properties and applications,” Rep. Prog. Phys. 77, 126402 (2014). [CrossRef]   [PubMed]  

4. H. A. Haus, Waves and Fields in Optoelectronics (Prentice Hall, 1984).

5. H. Haus and W. Huang, “Coupled-mode theory,” Proc. IEEE 79, 1505–1518 (1991). [CrossRef]  

6. A. W. Snyder, “Coupled-mode theory for optical fibers,” J. Opt. Soc. Am. 62, 1267–1277 (1972). [CrossRef]  

7. C. Manolatou, M. Khan, S. Fan, P. R. Villeneuve, H. Haus, and J. Joannopoulos, “Coupling of modes analysis of resonant channel add-drop filters,” IEEE J. Quantum Electron. 35, 1322–1331 (1999). [CrossRef]  

8. W. Suh, Z. Wang, and S. Fan, “Temporal coupled-mode theory and the presence of non-orthogonal modes in lossless multimode cavities,” IEEE J. Quantum Electron. 40, 1511–1518 (2004). [CrossRef]  

9. E. Waks and J. Vuckovic, “Coupled mode theory for photonic crystal cavity-waveguide interaction,” Opt. Express 13, 5064–5073 (2005). [CrossRef]   [PubMed]  

10. S. Fan, W. Suh, and J. D. Joannopoulos, “Temporal coupled-mode theory for the Fano resonance in optical resonators,” J. Opt. Soc. Am. A 20, 569–572 (2003). [CrossRef]  

11. Z. Ruan and S. Fan, “Temporal coupled-mode theory for light scattering by an arbitrarily shaped object supporting a single resonance,” Phys. Rev. A 85, 043828 (2012). [CrossRef]  

12. L. Verslegers, Z. Yu, Z. Ruan, P. B. Catrysse, and S. Fan, “From electromagnetically induced transparency to superscattering with a single structure: A coupled-mode theory for doubly resonant structures,” Phys. Rev. Lett. 108, 083902 (2012). [CrossRef]   [PubMed]  

13. L. Verslegers, Z. Yu, P. B. Catrysse, and S. Fan, “Temporal coupled-mode theory for resonant apertures,” J. Opt. Soc. Am. B 27, 1947–1956 (2010). [CrossRef]  

14. V. R. Shteeman, I. Nusinsky, and A. A. Hardy, “Time-dependent coupled mode analysis of parallel waveguides,” J. Opt. Soc. Am. B 27, 735–741 (2010). [CrossRef]  

15. V. Shteeman and A. A. Hardy, “Analysis of advanced photonic devices with time-dependent coupled mode equations,” Opt. Eng. 51, 054001 (2012). [CrossRef]  

16. E. Smith, V. Shteeman, and A. A. Hardy, “Time-dependent coupled mode analysis of advanced photonic micro devices,” in Proceedings of IEEE 27th Convention of Electrical & Electronics Engineers in Israel (IEEE, 2012).

17. B. Dana, L. Lobachinsky, and A. Bahabad, “Spatiotemporal coupled-mode theory in dispersive media under a dynamic modulation,” Opt. Commun. 324, 165–167 (2014). [CrossRef]  

18. D. A. Bykov, L. L. Doskolovich, E. A. Bezus, and V. A. Soifer, “Optical computation of the Laplace operator using phase-shifted Bragg grating,” Opt. Express 22, 25084–25092 (2014). [CrossRef]   [PubMed]  

19. M. G. Moharam, E. B. Grann, D. A. Pommet, and T. K. Gaylord, “Formulation for stable and efficient implementation of the rigorous coupled-wave analysis of binary gratings,” J. Opt. Soc. Am. A 12, 1068–1076 (1995). [CrossRef]  

20. L. Li, “Formulation and comparison of two recursive matrix algorithms for modeling layered diffraction gratings,” J. Opt. Soc. Am. A 13, 1024–1035 (1996). [CrossRef]  

21. S. G. Tikhodeev, A. L. Yablonskii, E. A. Muljarov, N. A. Gippius, and T. Ishihara, “Quasiguided modes and optical properties of photonic crystal slabs,” Phys. Rev. B 66, 045102 (2002). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (3)

Fig. 1
Fig. 1 Diffraction of a pulse on the guided-mode resonant grating (waveguide with a grating on its top) supporting two quasiguided modes.
Fig. 2
Fig. 2 (a) Two counter-propagating modes in a slab waveguide. (b) Modes excitation by ±m-th diffraction orders. (c) Mode leakage into −m-th diffraction order and mode coupling by means of 2m-th diffraction order.
Fig. 3
Fig. 3 (a) Transmission coefficient of the grating, |T|2, vs. incident light’s kx and ω: rigorous simulation (left part, at kx < 0) and analytical formula (16) (right part, at kx > 0). The dashed lines specify the light lines for θ = 0.2°,1°. (b) Transmission coefficient of the grating, |T|2, for different angles of incidence (θ = 0°,0.2°,1°): rigorous simulation (solid lines) and analytical formula (16) (circles).

Equations (20)

Equations on this page are rendered with MathJax. Learn more.

ω = ω 0 + v g ( k x k x 0 ) ,
u ( x , t ) = P ( x v g t ) e i ( k x 0 x ω o t ) ,
u t = v g u x + i ( k x 0 v g ω 0 ) u .
{ u t = v g u x + i ( k x 0 v g ω 0 ) u ; v t = v g v x + i ( k x 0 v g ω 0 ) v ,
k x 0 = 2 π m / d .
{ u t = v g u x + i ( k x 0 v g ω 0 ) u α 0 u + q f ( x , t ) e i k x o x ; v t = v g v x + i ( k x 0 v g ω 0 ) u α 0 v + q f ( x , t ) e i k x 0 x ,
{ u t = v g u x α 1 u c 1 u + c 2 e 2 i k x 0 x v + q f ( x , t ) e i k x 0 x ; v t = v g v x α 1 v c 1 v + c 2 e 2 i k x 0 x u + q f ( x , t ) e i k x 0 x ;
f T ( x , t ) = t ˜ f ( x , t ) + γ 1 s ( x , t ) ; f R ( x , t ) = r ˜ f ( x , t ) + γ 2 s ( x , t ) ,
s = u e i k x 0 x + v e i k x 0 x .
{ u ˜ t = v g u ˜ x α u ˜ + c 2 v ˜ + q f ( x , t ) ; v ˜ t = v g v ˜ x α v ˜ + c 2 u ˜ + q f ( x , t ) ,
s t = v g w x ( α c 2 ) s + 2 q f ( x , t ) ;
w t = v g s x ( α + c 2 ) w .
2 s t 2 + 2 α s t = v g 2 2 s x 2 ( α 2 c 2 2 ) s + g ( x , t ) ,
g ( x , t ) = 2 q [ f ( x , t ) t + ( α + c 2 ) f ( x , t ) ] .
F ( k x , ω ) = f ( x , t ) e i ( k x x ω t ) d x d t .
ω 2 S 2 i ω α S = v g 2 k x 2 S ( α 2 c 2 2 ) S + 2 q [ i ω + ( α + c 2 ) ] F .
F T = t ˜ F + γ 1 S ; F R = r ˜ F + γ 2 S .
T = F T F = t ˜ + γ 1 S F ; R = F R F = r ˜ + γ 2 S F ,
S F = 2 i q ω + i ( α + c 2 ) v g 2 k x 2 ( ω + i α ) 2 c 2 2 .
T ( k x , ω ) = t ˜ v g 2 k x 2 ( ω ω t ) ( ω ω p 2 ) v g 2 k x 2 ( ω ω p 1 ) ( ω ω p 2 ) ; R ( k x , ω ) = r ˜ v g 2 k x 2 ( ω ω r ) ( ω ω p 2 ) v g 2 k x 2 ( ω ω p 1 ) ( ω ω p 2 )
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.