JP3053966B2 - Directional coupler filter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical filter used for optical communication, optical exchange, optical information processing, optical recording, etc., for selecting light of a specific wavelength from incident light, and more particularly to a plurality of waveguides. The present invention relates to a directional coupler filter with a grating that performs waveguide mode conversion between layers.

[0002]

2. Description of the Related Art Recently, optical communication and optical information processing using a wavelength division multiplexing system based on an optical integrated circuit have been actively studied. One of the important devices for realizing the above system is a wavelength filter. In particular, when an optical integrated circuit is considered, an element such as a directional coupler filter which is formed of a waveguide and can connect different elements is very attractive. However, the directional coupler filter generally has poor wavelength selectivity, except for the backward coupling type, which is difficult to use because the input port and the output port exist in the same direction, and thus are not suitable for dense wavelength multiplexing. And has been.

However, recently, a grating forward coupling filter using a semiconductor laminated waveguide as disclosed in Japanese Patent Application Laid-Open No. 2-239209 has attracted attention because of its sharp wavelength selectivity. Further, regarding the transmission side lobe which causes the crosstalk between the channels to deteriorate, the duty ratio or the coupling coefficient of the grating formed in the vicinity of the two waveguides proposed in Japanese Patent Application No. 1-314841 is also known. The suppression can be realized by a method of distributing in a tapered shape. This is based on the principle that the transmission response of the filter is proportional to the Fourier transform of the coupling coefficient.

[0004]

However, in the conventional method, the taper function for tapering the coupling coefficient is limited to a positive function, and the type of the taper function is limited. Therefore, the transmission side lobe cannot be suppressed sufficiently effectively, and the specifications of the channel spacing and the crosstalk suppression ratio cannot be sufficiently widened.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a directional coupler filter with a grating having a structure capable of expanding the degree of freedom of applying a taper function of a grating that distributes coupling coefficients.

[0006]

According to the present invention, different phases are introduced by shifting the phase of the grating in the middle of the coupling region between a plurality of waveguides, and the type of the taper function for suppressing transmission side lobes is introduced. Is spreading. Since the coupling coefficient can be partially negative (this principle will be described later), for example, a sink function (sin (x) /
x) can be applied as a taper function.

FIG. 1 shows the principle of operation of a directional coupler filter according to the present invention. The device is composed of two layers of asymmetric waveguides 1 and 2 (different in thickness, composition, etc.), and the layer thicknesses of these waveguides 1 and 2 have two vertical transverse modes (in FIG. Mode and an odd mode optical field intensity distribution are shown). Lower waveguide 1
Is propagated as an odd mode having a central intensity to the lower waveguide 1, and in the interaction region, at the phase matching wavelength determined by the grating 3, the upper waveguide 2.
As an even mode. Wavelengths apart from the phase matching wavelength remain in the lower waveguide 1 and propagate as they are. Here, the phase matching wavelength λ satisfies the following first-order Bragg condition. β _e (λ) -β _o (λ) = 2π / Λ (1) Here, β _e and β _o represent the propagation constants of the even mode and the odd mode, respectively. Λ is the period of the grating 3.

In this device, even mode and
Each of the odd-mode electric fields is E_e, E_oIt represents. Two layers
Due to the strong asymmetry between 1 and 2, the electric field of the guided mode is
Mode for guiding each waveguide when the waveguides exist independently
Very similar to the electric field of Therefore, mode coupling is
It may be considered that it depends on the diffraction of rating 3. Both
The coupling coefficient κ between the modes is represented by the following equation. κ (z) = (ωε₀/ 8) ∫₀ ^gE_eΔn _g ^TwoE_odx (2) g is the depth of the grating. ω is the angular frequency of light,
ε₀Is the dielectric constant of vacuum, x is the depth direction of grating 3
And z are coordinates in the traveling direction of the guided light.

Here, Δn _g ^TwoIs in grating 3
Is the perturbation of the refractive index due to
Of the upper waveguide layer 2 constituting the grating 3
Rate n_Two, The refractive index of the cladding layer 5 is n₁As below
Is represented as − (N _Two ^Two-N ₁ ^Two) / Πsin (πw / Λ): Δn in the grating layer _g ^Two= ｛(3) 0: Other than the above.

FIG. 2 shows the duty ratio of the grating 3.
(Grating tooth width / cycle: w / Λ) is changed
Represents the change of the coupling coefficient κ. Grating 3
When the duty ratio is changed, from equation (3), Δn _g 
 ^TwoChanges to change the coupling coefficient κ. In this case,
To change the effective thickness of the upper waveguide 2, the electric field distribution type
Both the shape and the propagation constant change. Therefore, the phase matching
The condition changes, and the phase matching wavelength shifts. So du
The period Λ is controlled in accordance with the change in the
Operation to keep the phase matching wavelength constant over the whole range
There is. The curve with period Λ in FIG.
is there. In this element, the duty ratio is
By operating, the coupling coefficient κ can be controlled.

Next, it is explained that the coupling coefficient κ can be made negative.
I will tell. In order to perform the distribution of the coupling coefficient, the coupling area (length
L) is divided into N regions (the length of each region is l) as shown in FIG.
¹, L^Two, ..., l^N-1, L^NAnd each area
Within, assume that the parameters are constant. Para of each area
When the meter is represented by an F-matrix, the electric field E at the input end_e
(0), E_o(0) and the electric field E at the output end_e(L), E
_o(L) is as follows. However, | is arranged vertically
And [] represent a matrix. | E_e(0) || E_e(L) | | | = [F] | | (4) | E_o(0) || E_o(L) | [F] represents the characteristics through the coupling region.
F matrix of area^kThen, [F] = Π _{k = 1} ^NF^k (5).

Each element of the F-matrix is as follows. _{F 11 = [cosh (γL)} + iΔβLsinh (γL) / γL)] exp (iβ B L), F 12 = -κLsinh (γL) exp [-i (β B L + φ)] / (γL), F 21 = - κLsinh (γL) exp [i ( β B L + φ)] / (γL), (6) F 22 = [cosh (γL) -iΔβLsinh (γL) / γL)] exp (-iβ B L). Here, Δβ and γ are as follows. Δβ = β−β _B , (7) γ ² = κ ² − (Δβ) ² , (8).

Here, β represents the propagation constant of the propagation wavelength, and β _B represents the propagation constant at the phase matching wavelength. φ represents the phase of the grating 3. If φ is 0,
This is the case where the phase of the grating 3 is formed continuously, and the case of π corresponds to a shift of the grating 3 by a half wavelength at the boundary as shown in FIG.
That is, F ₁₂ and F ₂₁ including the coupling coefficient κ are expressed as exp (−
Since iπ) =-1, the sign of the coupling coefficient κ is reversed in sign. In this manner, by shifting the phase φ by π at the boundary, a negative region can be formed in the distribution of the coupling coefficient κ.

[0014]

Embodiment 1 Hereinafter, an embodiment will be described. Japanese Patent Application No. 1-31
As shown in US Pat. No. 4841, several taper functions have been studied for the coupling coefficient κ. Therefore, a combination of the taper function and a sink function Sinc (απz / L) that makes the Fourier transform spectrum a square wave will be described. α is a constant. The combined taper functions are listed below. z indicates the light propagation direction.

Sync Hamming function: f (z) = Sinc (3πz / L) [1 + 0.852 cos (2πz / L)] (9) Sink raised cosine function: f (z) = Sinc (3πz / L) [ 1 + cos (2πz / L)] (10) Sink Blackman function: f (z) = Sinc (3πz / L) [1 + 1.19 cos (2πz / L) + 0.19cos (4πz / L)] (11) Sink Kaiser function: f (z) = Sinc (3πz / L) [γ / sinh (γ)] I ₀ ｛γ [1- (2z / L) ² ] ^1/2 (12).

Where I₀Is a first-order variant of order 0
Bessel function, γ is the shaping parameter, where
It is assumed to be 10. The above taper function is normalized by the bond length L.
And represented as follows: ∫ _{-L / 2} ^{2 / L}f (z) dz = 1 (13) From the above, the coupling coefficient κ is expressed as follows. κ (z) = κ₀f (z) / L (14).

K ₀ is a value obtained by integrating the coupling coefficient κ over the interaction region. FIG. 5B shows the distribution of the coupling coefficient κ tapered by the sink Hamming function. In order to obtain this distribution, changes in the duty ratio and the period are obtained from FIG. FIG. 5A shows the distribution of the duty ratio and the period of the grating 3 for realizing the distribution of the sink Hamming function. The dotted line in the figure is the phase-shifted boundary, and the shaded area is the region of the negative coupling coefficient κ. As a result, a distribution of the coupling strength κ along the sink Hamming function as shown in FIG. 5B is obtained. In a region surrounded by the phase-shifted boundary, the coupling coefficient κ is negative.

FIGS. 6 to 9 show transmission spectra when tapered by the functions (9) to (12) (only one side is shown). The structure is as shown in FIG. In FIGS. 6 to 9, the dotted line indicates the case where no combination with the sync function is performed, and the solid line indicates the case where the combination is performed. In each case, transmission side lobes can be further suppressed as compared with the case where the sink function is not combined, and the waveform approaches a square wave.

[0019]

Embodiment 2 A directional coupler filter according to the present invention is Ga
An example manufactured using an As-based semiconductor will be described with reference to FIGS. An AlGaAs cladding layer 6, a lower waveguide 1 composed of GaAs (well) / AlGaAs (barrier) -multiple quantum well (MQW), and an AlGaAs are sequentially formed on a GaAs substrate by MBE (molecular beam epitaxy) from the bottom.
An s cladding layer 5, an upper waveguide 2 made of GaAs / AlGaAs-MQW, and an AlGaAs grating layer 7 are formed. Next, the grating 3 is formed by photomask exposure and reactive ion beam etching (RIBE).
Is formed, and then CBE (Chemical Beam Epitaxy)
Alternatively, the AlGaAs cladding layer 8 is formed by an LPE (liquid phase epitaxy) method. The photomask used here is produced by electron beam exposure under computer control, and can easily form the grating 3 having a desired period and duty ratio.

In this embodiment, the sink Gaussian function f
(Z) = Sinc (πz / L) exp [− (2πz /
L) The taper of the coupling coefficient κ according to 2] was used. FIG. 10 shows a transmission spectrum from the upper waveguide 2 when the light of the variable wavelength light source is input from the lower waveguide 1. It can be seen that the transmission side lobe is sufficiently suppressed.

[0021]

Embodiment 3 FIG. 11 shows an example in which a directional coupler filter according to the present invention is integrated with a photodetector. The integrated device is composed of a filter region 11 and a detector region 12, and in the detector region 12, the i-G
A light absorption layer 28 made of aAs is formed. i-G
The lower layers of the aAs layer 28 are all formed of n-type.
The upper layer is formed of p-type. Therefore, detector area 1
In 2, a pin photodetector is configured.

In the fabrication, the entire detector region 12 is grown in the first growth, only the filter region 11 is etched until the upper waveguide 25 is exposed, and a grating 26 having a phase shift region is formed. AlGaAs
The cladding layer 27 is regrown. In this first growth, the cladding layer 22 and the lower waveguide layer 2 are sequentially formed on the substrate 21.
3, cladding layer 24, upper waveguide layer 25, light absorbing layer 2
8, the cladding layer 29 and the contact layer 30 are laminated. 11, 31 is a p-type electrode on the contact layer 30, and 32 is an n-type electrode on the back surface of the substrate 21.

At a desired wavelength, light input from the lower waveguide 23 of the filter region 11 is coupled to the upper waveguide 25 and strongly coupled to the absorption layer 28 of the detector region 12. Light of other wavelengths propagates through the lower waveguide 23, and in the detector region 12, since the electric field intensity distribution hardly reaches the absorption layer 28, the light is hardly absorbed and continues to travel through the lower waveguide 23. By providing this element having the same structure as that of FIG. 11 having a different grating period at the subsequent stage, it is possible to simultaneously detect a plurality of wavelengths. In the detector region 12, it is of course possible to improve the frequency response by applying a bias as necessary.

[0024]

Embodiment 4 The present directional coupler filter can be used as a cut filter for a spontaneous emission light generated from an optical amplifier or an active optical filter. According to this element, it is easy to integrate the semiconductor optical amplifier and the active optical filter on the same substrate, and it becomes an effective noise cut filter.

FIG. 12 shows an example. The same numbers are assigned to the same functional units as in FIG. Optical amplifier 13 in the center
Of the active layer 38 coincides with the upper waveguide 25 of the filter region 11. In the first-stage filter region 11, only the wavelength of the used amplification band is shifted from the lower waveguide 23 to the upper waveguide 25, and is amplified by the optical amplifier 13 into which a current slightly smaller than the threshold is injected. Then, the light is returned to the lower waveguide 23 again and output. Therefore, noise light generated by another optical amplifier can be cut before entering the optical amplifier 13 of the present element, and the noise light generated by itself does not leak as output light. The above operation is extremely effective when semiconductor optical amplifiers are connected in multiple stages.

[0026]

According to the present invention, the coupling coefficient between the plurality of waveguides can be made negative by the phase shift of the grating which causes the coupling of the waveguide mode between the plurality of waveguides. The degree of freedom of applying the taper function for giving a distribution can be expanded. Therefore, effective transmission side lobe suppression can be achieved, and the filter transmission spectrum can be made closer to a more ideal square wave.

Further, the directional coupler filter according to the present invention can easily design a coupling coefficient, and can widely conform to the specifications of the channel (waveguide) interval and the crosstalk suppression ratio. Further, it is extremely suitable for integration with other elements such as a photodetector, an optical amplifier, etc., and present and future optical information transmission systems, optical communication systems, optical LANs, optical computing, optical recording, optical recording, and the like. It can be widely applied to exchange and the like.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the operation principle of a directional coupler filter according to the present invention.

FIG. 2 is a diagram illustrating a relationship between a duty ratio and a period of a grating and a coupling coefficient.

FIG. 3 is a diagram illustrating a method of analyzing a directional coupler filter according to the present invention.

FIG. 4 is a diagram for explaining the principle of the phase shift of the grating in the directional coupler filter according to the present invention.

FIG. 5 is a diagram illustrating a distribution of a duty ratio and a period of a grating and a distribution of a coupling strength based on the distribution.

FIG. 6 is a diagram illustrating a transmission response when a coupling coefficient is tapered by a sink Hamming function.

FIG. 7 is a diagram illustrating a transmission response when a coupling coefficient is tapered by a sync raised cosine function.

FIG. 8 is a diagram illustrating a transmission response when a coupling coefficient is tapered by a Synch Blackman function.

FIG. 9 is a diagram illustrating a transmission response when a coupling coefficient is tapered by a Synch Kaiser function.

FIG. 10 is a diagram illustrating a transmission response when a coupling coefficient is tapered by a sink-Gaussian function.

FIG. 11 is a diagram illustrating an embodiment in which a directional coupler filter and a photodetector according to the present invention are integrated.

FIG. 12 is a diagram illustrating an embodiment in which a directional coupler filter and an optical amplifier according to the present invention are integrated.

[Explanation of symbols]

 1, 23 Lower waveguide 2, 25 Upper waveguide 3, 26 Grating 5, 6, 8, 22, 24, 29 Cladding layer 7 Grating layer 11 Filter region 12 Detector region 13 Optical amplifier 21 Substrate 27 High resistance cladding layer 28 Light absorbing layer 30 Contact layer 31, 32 Electrode 38 Active layer

Claims (9)

(57) [Claims] In a directional coupler filter for selecting a wavelength using a grating, the duty ratio of the grating changes in a tapered shape over a coupling region, and the phase of the grating is at least one in the middle.
A directional coupler filter characterized by being shifted in places. 2. The directional coupler filter according to claim 1, wherein the period itself changes in accordance with a change in the duty ratio of the grating. 3. The directional coupler filter according to claim 1, wherein the directional coupler filter comprises a semiconductor waveguide laminated. 4. The directional coupler filter according to claim 1, wherein the filter is integrated with a photodetector. 5. The directional coupler filter according to claim 1, wherein the directional coupler filter is integrated with an optical amplifier. 6. The directional coupling according to claim 1, comprising a lower waveguide, a cladding layer, and an upper waveguide, wherein a grating is formed in one of the lower and upper waveguides and a peripheral region thereof. Filter. 7. The directional coupler filter according to claim 5, wherein the photodetector has a pin structure. 8. The directional coupler filter according to claim 5, wherein the optical amplifier has a semiconductor laser structure. 9. The method according to claim 1, wherein the duty ratio of the grating is tapered in accordance with one of a sync Hamming function, a sync raised cosine function, a sync blackman function, a sync Kaiser function, and a sync Gaussian function. The directional coupler filter according to claim 1, wherein: