JPH02262128A - Polarization-independent optical switch - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a polarization-independent optical switch using a directional coupler, and particularly to a polarization-independent optical switch that has a high light confinement effect in a leading waveguide and reduces polarization dependence. This invention relates to a polarization-independent optical switch.

[Conventional technology]

Conventionally, this type of polarization-independent optical switch has a directional coupler forming the main part as shown in the schematic configuration diagram of FIG. 5. In the figure, the conventional polarization-independent optical switch is made of lithium niobate (hereinafter, L+NbOs
), two waveguides (2) and (3) are formed on the substrate (1), and the negative portions of the waveguides (2) and (3) approach each other to form a coupling portion (6). It is.

Intensity P from the left incident end face of the waveguide (2) in -. When the light of
Lights A and PB are emitted from the right output end face of each waveguide (2) and (3). The initial length L at which the relationship of PA/(PA+PR) to 0 is established for the emitted multiple lights of intensities PA and Pi is called the complete coupling length. The relationship between the emitted light PA and Pa regarding this complete bond length is shown in FIG. 6 (8). Further, the complete coupling length I generally takes a different length depending on the polarization state of the incident light, that is, whether it is in the TE mode or the TM mode. Here, the TE mode refers to a polarization state in which the electric field component is parallel to the substrate (1), and the TM mode refers to a polarization state in which the electric field component is parallel to the substrate (1).
) refers to the state of polarization perpendicular to Note that the two waveguides (2) and (3) have the same structure.

The directional coupler of the conventional polarization-independent optical switch described above is configured to have a uniform Δβ type by providing a pair of electrodes (not shown) in the coupling part (6), and by applying an electric field, the electro-optic effect can be achieved. The incident light can be switched based on.

This switching state is shown in FIG. 6(B) as a -like Δβ switching diagram. In the figure, in the e state (bar state), the intensity of the incident light is p, and the intensity of the emitted light is PA.
This indicates a state where the intensity of the other emitted light is p, = o, and in the ■ state (cross state), the intensity of the incident light P0 is the intensity of the emitted light PB.
This indicates a state where the intensity PA of the other emitted light is 0. Further, when an electric field (ΔβL/π) is applied, the above-mentioned θ state occurs on a plurality of circular arcs.

Next, the ratio of the bond length to the complete bond length (L/L) is T
rlJ (L/
FIG. 7 shows an all-mode switching diagram for both TE and TM modes when an electric field is applied with jl=1). In the same figure, when an electric field is applied and (ΔβL/π) is increased, the TE mode becomes the e state at a cape of (ΔβL/π) 5.2 (V), and when an electric field is further applied, (ΔβL/π) ), the TM mode becomes (ΔβL/π)45.9
At (V), the state becomes e. In this way, the incident light can be switched by switching from all TE/TM modes to the e state.

In addition, ELEC is another conventional polarization-independent optical switch.
TRONIC5LETTER58th 0ctorb
er 1987 Vo123, No. 21 No. 11
There is something described on pages 67 to 1168, which is shown in FIG. In the figure, the conventional polarization-independent optical switch has two waveguides (2) and (3) formed on a substrate (1) of NbO3, and the two waveguides (2)
, (3) a pair of electrodes (4), (5) on the top and another pair of electrodes (45) on the outside of the pair of electrodes (4), (5),
(55) is installed every evening.

In the above configuration, the electrodes (4), (5), (4
5), (55), and the conditions for all modes of both TE and TM are L, Nb.
It is determined by the electro-optic coefficients γ13 and γ33 of the O3 substrate (1). FIG. 9 shows all mode talk in both TE and TM modes versus Δβ (corresponding to applied voltage) in this case. In the same figure, α=ΔβTE/ΔβTM (ΔβT
E% 68TM is the phase mismatch of the directional coupler for the TE and TM modes, respectively). This α is 0.25≦α≦0
．． 34, the crosstalk can be reduced to -20 dB or less in both modes by adjusting the applied voltage. The actual measured value at a wavelength of around 1.3 μm is α = 0.29,
Since the above conditions are satisfied, the voltage at which the θ state is reached is T
This means that both E and TM modes can be used.

In addition, as other conventional polarization-independent optical switches, ELE
CTRONIC5LETTER515th Septe
amber 1988V()1.24. NO, 19th 11th
This conventional polarization-independent optical switch is described on pages 98 to 1200, and in the case of an optical crystal such as NbO3, the electro-optic effect acts equally on all modes in both TE and TM modes. Polarization independence was ensured by using an X-cut crystal.

(Problem to be Solved by the Invention) Since the conventional polarization-independent optical switch is configured as described above, the problem regarding mode switching between TE mode and TM mode (ΔβL /π), and there was a problem that polarization independence could not be achieved completely. Other conventional polarization-independent optical switches can achieve complete matching of the coupling lengths of both TE and TM modes relatively close to the cutoff condition, but because the light confinement effect in the waveguide is weak, the waveguide propagation This has the problem that loss and radiation loss in the waveguide section tend to increase, and the electro-optic effect that can be used with other conventional optical switches has a weak effect, so the switching voltage required for optical switching increases. There was a problem with this.

The present invention was made to solve this problem, and an object of the present invention is to obtain a polarization-independent optical switch that can ensure an extremely high optical confinement effect of a waveguide and reduce crosstalk.

[Means to solve the problem]

The polarization-independent optical switch according to the present invention has two waveguides disposed close to each other, and the light intensity is increased between the two waveguides. The coupling length of the directional coupler to be converted is TE path (2
), (3), and the length of the waveguides (2) and (3) is approximately 3 yards of the complete coupling length of the TE mode.
The parts approach to form a joint part (6), and the joint part (6) is doubled, and the TM mode '
The coupling length of the TE mode is set to approximately one time the complete bond length of the TE mode, and the above-mentioned light intensity is set to approximately three times the complete bond length of the TE mode, and the conversion of the TM mode is uniformly driven in the Δβ type. The bond length is approximately 1 times the complete bond length of the TM mode, which is 8°.
and the above joint part (6
) forming waveguides (2) and (3) (effect)
The polarization-independent optical switch of the present invention has an electrode (4) on which a switching voltage is applied (5).

The coupling length of the tropic coupler is the complete coupling length of TE mode.
Next, the operation of the optical switch of this embodiment based on the above configuration will be described with the length of the optical switch being three times as long as the length of the optical switch of this embodiment and approximately one time as the complete coupling length of the TM mode. First, by forming the electrodes (4) and (5), in both the TE and TM modes, in the state where the switching voltage V is not applied, the transition from the state to the e state is achieved at an applied voltage value that is very close to that shown in Figure 2. (Δβ
Perform in TE mode (L/π).
ItLr3) is in ■ state, and TM mode (L/(Example)
J241) is■
state. Here, an embodiment of the present invention will be described below with reference to FIGS. 1 to 2, where V is applied and gradually increased under the switch gummy tool, as shown in FIG. This figure 1 shows the L/u of both TE/TM mode in this embodiment.
43, L/β~1 Schematic configuration diagram of a polarization-independent optical switch, FIG. 2 is translated to the right side. This parallel movement straight line indicates the switching diaphragm of this embodiment. In each of the above figures of each switching dieger for both TE and TM modes, the points that intersect with the polarization-independent light beam curve according to this embodiment correspond to the θ state, switching, and switching, respectively.

This indicates that the The switching point to this e state is TE
/TM both (ΔβL/π) = 7.94
This is a state in which a switching voltage V of (V) is applied.

Furthermore, the operation of the optical switch of the above embodiment will be explained in detail with respect to the channel output intensity based on FIG. FIG. 3 is a light intensity graph showing the relationship between the channel output intensity in the ■ state and the complete bond length. In the same figure, a numerical range is defined as 2.7≦L/ItTM≦3.3 for TE mode and 0.7≦L/ItTM≦1.3 for TM mode, and this numerical range corresponds to the insertion loss. This is determined based on the viewpoint. When each complete bond length is 2.7≦L/uA, ≦3.3.0.7≦L/J2TM≦1.3, the channel output light intensity in state ■ is approximately 75%.
The lower limit can be set as the lower limit. Since this channel output light intensity of 75% results in an insertion loss of approximately 1.25 dB, it is acceptable even if the insertion loss is increased within a practical range. Therefore, if the insertion loss is in the range of approximately -2 dB, the complete coupling length of both TE and TM modes is 2.6≦L/
ItT, ≦3.4.0.6≦L/itTM≦1.
It can also be set to 4.

Next, a description will be given in relation to a manufacturing method of an embodiment of the polarization-independent optical switch according to the above embodiment.

Generally, a waveguide (2), (
As a method for forming 3), the TI diffusion method is most commonly used. The waveguide formed by this T1 diffusion method (
In 2) and (3), it is easier to make the refractive index change Δn8 for extraordinary rays larger than the refractive index change Δn0 for ordinary rays. Namely, Z-cut crystal glue.

In NbO3 substrate (1), the complete bond length of TM mode is TE
It is easy to make the length longer than the complete bond length of the mode.

By appropriately setting the width of the waveguides (2) and (3), the spacing between the two waveguides, and the TI diffusion conditions, the length of the coupling section of the directional coupler can be adjusted to suit both the TE and TM modes. It is easily possible to make the length approximately three times and one time the perfect coupling length, respectively, and such a waveguide has a sufficiently large light confinement effect.

This configuration is shown in FIG. 4, in which the electrode (4),
Since (5) is made of metal, the waveguides (2), (3) and electrodes (4), (
The structure is such that there are 51 and 5I02 (approximately 4,000 people) layers.

[Effects of the Invention] As explained above, in the polarization-independent optical switch according to the present invention, the coupling length of the directional coupler is approximately three times the complete coupling length of the TE mode for the TE mode, and the TM By adopting a configuration in which the mode's perfect coupling length is approximately one time and the directional coupler is driven and controlled in a uniform Δβ type, it is possible to ensure an extremely high optical confinement effect of the waveguide, and also to ensure both TE and TM modes. Since the transition from the ■ state to the e state can be performed with extremely close applied voltage values, it is possible to reduce crosstalk.

[Brief explanation of drawings]

Fig. 1 is a schematic configuration diagram of a polarization-independent optical switch according to an embodiment of the present invention, Fig. 2 is a switching diagram in the embodiment shown in Fig. 1, and Fig. 3 is a channel output in the ■ state with respect to the complete coupling length. 4 is a partial cross-sectional configuration diagram of the embodiment shown in FIG. 1, FIG. 5 is a schematic configuration diagram of a conventional polarization-independent optical switch, and FIG. 6 (A
) is the relationship graph of the emitted light intensity with respect to the complete bond length, the sixth
Figure (B) is a switching diagram of -like Δβ, the seventh
The figure shows the switching diagram for both TE and TM modes.
FIG. 8 is a schematic diagram of another conventional polarization-independent optical switch, and FIG. 9 is a graph showing the relationship between crosstalk and switching voltage in TE mode. (1)...L, Nbo3 substrate (2), (3)...waveguide (4), (5)...electrode (6)...coupling part Note that the same reference numerals in each figure indicate Indicates the same or equivalent part. □l/] V snoring '1' end 207- V]

Claims (1)

[Claims] Polarization-independent light is composed of a directional coupler that arranges two waveguides close to each other and converts the light intensity between the two waveguides, and drives and controls the conversion of the light intensity in a uniform Δβ type. In the switch, the coupling length of the directional coupler is approximately three times the complete coupling length of the TE mode for the TE mode, and
1. A polarization-independent optical switch characterized in that the TM mode has a length approximately one times the complete coupling length of the TM mode.