JP3331828B2 - Optical transmission / reception module - 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 communication system for performing two-way communication using light of two wavelengths so that light of two wavelengths can be separated by combining two photodiodes and a light source. The present invention relates to an optical transmitting and receiving module.

[0002] First, two-way optical communication will be described. The transmission loss of the optical fiber is reduced, and the semiconductor laser (hereinafter referred to as LD)
) And the characteristics of the semiconductor light receiving element (sometimes abbreviated as PD) are improved. For this reason, communication attempts using optical signals have been actively conducted. This is called optical communication. In particular, research on communication using optical signals using long-wavelength light of 1.3 μm and 1.55 μm has been vigorously conducted. For example, it is an attempt to transmit a signal of a telephone, a facsimile, a television, or the like by an optical signal.

[0003] Optical communication also has various aspects. In particular, recently, it is possible to use a single optical fiber to simultaneously perform low-speed digital communication such as telephone and facsimile that exchanges signals in two directions, and high-speed analog communication of optical CATV that transmits analog TV signals by wire. Sex is being considered. The advantage of this method is that only one optical fiber is required.

FIG. 1 is a principle diagram of such two-way communication. The base station 1 amplifies a telephone or TV signal, sends it out with one optical fiber, transmits it as it is near home, and divides it into, for example, 16 signals at the branching device 2.
Each home (subscriber terminal 3) is connected from the branch unit 2 by one optical fiber. This is to reduce the number of optical fibers of the base station 1 and the equipment cost.
Such a system for transmitting a signal to each home by light is called an optical subscriber system.

[0005] The optical signal transmitted in this manner is:
It is converted into an electric signal at a subscriber terminal (called ONU: an abbreviation for OPTICAL NETWORK UNIT). The conversion mechanism will be described. In the wavelength demultiplexer 4 of the subscriber terminal 3, 1.3 μm light carrying a digital signal and 1.5 μm carrying an analog signal
Separate 5 μm light. The separated 1.55 μm light is a photodiode (PD) that accurately reproduces analog signals
5 is incident. Optical signals are converted to electrical signals. Various signal processing is performed in the signal processing unit 6 and the TV set 7
It is led to.

The 1.3 μm light carrying digital signals is:
The light is guided to a single multiplexer / demultiplexer (referred to as an optical coupler 8). This is converted into an electric signal by the digital PD 9. This is guided to the telephone / facsimile 11 through the processing by the signal processing unit 10. The signal transmitted from the base station is received by a home telephone or facsimile.

Since telephone and facsimile are two-way communication, a function of transmitting a signal from a terminal to a station is also required. A transmission signal from a telephone or a facsimile is processed by the signal processing unit 10 and converted into an optical signal by the digital LD 12. This optical signal passes through the optical coupler 8 and the wavelength demultiplexer 4 and enters the optical fiber.
Sent to. The optical coupler 8 has such a 1.3 μm
It is inserted to enable two-way communication by light.

In summary, in such an ONU system, a wavelength demultiplexer 4 for separating 1.3 μm light and 1.55 μm light, an optical coupler 8 for bidirectional communication,
PD5 and PD9 for light / electricity conversion and LD12 for light / electricity conversion are required. A combination of these is called an ONU optical module. The present invention relates to an improvement in the configuration of the ONU optical module.

[0009]

2. Description of the Related Art First, the prior art of an ONU optical module will be outlined. As described above, the ONU optical module requires a wavelength demultiplexer for separating or combining two wavelengths of light and an optical coupler for transmitting and receiving the same wavelength light. Several proposals have been made for these devices. All of these are merely proposals, and none have been put to practical use. Each has its advantages, but it still has its own drawbacks.

[Structure example 1 of optical module for ONU 1: mirror type wavelength demultiplexer / optical coupler] FIG. 2 shows an example of an optical module using a mirror type wavelength demultiplexer / optical coupler. A mirror manufactured by depositing a dielectric multilayer film on a glass substrate is used as a wavelength demultiplexer and an optical coupler. Thin films of two kinds of materials having different refractive indexes are alternately laminated and cut at an angle of 45 °, and the end face is at an angle of 45 ° to the multilayer film. The product of the thickness of the thin film, the refractive index, and the cosine of the tilt angle gives the effective optical path length. Light reflection and refraction occur at the boundaries of different films. Light having a wavelength such that the multiple reflected lights strengthen each other is reflected more strongly.

In the case of light having a wavelength such that the multiple reflected lights cancel each other, most of the light energy is transmitted. Therefore, the mirror type wavelength demultiplexer of FIG. 2 is combined with a thin film so that light of 1.3 μm is transmitted and light of 1.55 μm is reflected. This allows 1.3 μm light and 1.55 μm
Light can be separated. The 1.3 μm light travels straight and enters the mirror type optical coupler 8.

Similarly, the optical coupler 8 is also formed of a multilayer film in which two kinds of dielectrics having different refractive indexes are stacked. The membrane surface and the end surface are 45
゜. Light of a specific wavelength is partially transmitted and partially reflected. The optical fiber is connected to the optical connector 13. Thereafter, the light propagates in free space. However, since the distance from the optical connector 13 to the digital PD 9, LD 12, and analog PD is long, the light is collimated by the collimator lens 14 and then guided to the mirror-type wavelength demultiplexer 4. Correspondingly, the lenses 17, 1 are used to convert the parallel light into a focused light.
8, 19 are installed.

A system using such a mirror-type wavelength demultiplexer or a coupler is described in, for example, IEEJ Spring Meeting, 1990, Conference No. C-203 (P4-25).
8), Proposal Lecture No. C-222 (p4-264), 1990 Autumn Meeting of the Institute of Electronics, Information and Communication Engineers.

A disadvantage of this method is that it is difficult to separate 1.3 μm light and 1.55 μm light by a mirror type wavelength demultiplexer. The optical path a through which analog signals pass must not receive 1.3 μm light, and the optical path b through which digital signals pass must not contain 1.55 μm light. If the separation ratio between the two is poor, noise will enter the telephone or television. In each optical path, the noise cannot be ignored unless the energy of the other party's light is less than 10 ^-4 .
Image quality and sound quality will be impaired. A desired separation ratio cannot be obtained only by the mirror type wavelength demultiplexer.

Therefore, a contrivance is made to completely block 1.3 μm light by inserting the multilayer filter 15 in the optical path a. The multi-layer filter 16 that blocks 1.55 μm light is also inserted into the optical path b.

[Configuration Example 2 of Optical Module for ONU: Waveguide Type Wavelength Demultiplexer / Optical Coupler] FIG. 3 shows the configuration of a waveguide type optical module. With a silica-based planar optical waveguide,
The wavelength demultiplexer and the optical coupler are configured. This can reduce the number of parts. Promising because of the potential for downsizing the module. For a planar optical waveguide (referred to as PLC), a quartz cladding layer 23 is deposited on a Si substrate 22, and an impurity (eg, Ge) that increases the refractive index is linearly doped into the cladding layer 23 to form an optical waveguide. It was done. The waveguide 24 extends straight from the end face, and f
At this point, the waveguide is divided into two waveguides 25 and 26. This corresponds to the optical coupler 8.

These waveguides are bent in the middle to become parallel waveguides 27 and 28. Further, at point g, another waveguide 29 is formed near the waveguide 24. This is waveguide 3
It leads to 0 and 31. An end face of the single mode optical fiber 32 is joined to the front end of the waveguide 24. The waveguides 24 and 29 in the vicinity of the point g function as a wavelength demultiplexer 4 for separating 1.3 μm light and 1.55 μm light. The example of FIG.
TT R & D vol. 42, no. 7, 1993, p
903-912.

Signal light enters the waveguide 24 from the optical fiber 32 and is separated into 1.3 μm light and 1.55 μm light at point g. The 1.3 μm light travels straight and is distributed to the waveguides 25 and 26 by the optical coupler 8. The light passing through the waveguide 25 passes through the multilayer filter 34 and passes through the digital PD 9.
Incident on. This is the propagation of the received light. The light of the digital laser LD 12, which is the transmission light, is converged by the condenser lens 33, enters the waveguide 28, passes through the waveguides 26 and 24, and enters the optical fiber 32.

At the branch point g, the 1.55 μm light travels to the waveguide 30, travels along the optical path a, passes through the multilayer filter 35, and enters the analog output single mode fiber 36. The provision of the multilayer filters 35 and 34 in the optical path a and the optical path b is because unnecessary 1.3 μm light
This is to completely block 1.55 μm light. In this method, the wavelength demultiplexer and the optical coupler can be manufactured on a single quartz substrate by bringing the waveguides close to each other (point g) or branching them (point f). Such a waveguide type optical module is disclosed in NTT R & D vol. 43, no. 1
1, 1994, pp. 1273-1280.

In this example, there is no condenser lens between the waveguides 27 and 31 and the digital PD 9 and analog fiber 36. This is because in the case of the fiber 36, the fiber 36 is directly bonded to the waveguide 23. This is because in the case of the PD 9, the light is not spread because it is sufficiently close to the waveguide 23. A lens can also be omitted between the digital LD 12 and the waveguide 28. However, when it is desired to transmit light into the waveguide 28 with high efficiency, the lens 33 is provided.

The advantage of this method is that it can be manufactured entirely on one substrate and can be made compact. There is also an advantage that the number of parts can be reduced. Further, since the planar waveguide is made of the same material as the optical fiber and has almost the same size, it is said that the flat waveguide is well adapted to the optical fiber.

[Configuration Example 3 of Optical Module for ONU: Optical Fiber Type Wavelength Demultiplexer / Optical Coupler] A wavelength demultiplexer using an optical fiber has already been put to practical use. Optical fiber WDM
Called. Couplers using optical fibers have also been put to practical use. This is called an optical coupler. Further, an optical fiber filter in which a multilayer filter is inserted between two optical fibers is also manufactured. An ONU optical module can also be configured using components using these optical fibers. FIG. 4 shows an optical fiber type ONU optical module.

The input single mode fiber 39 is connected to an ONU terminal by an optical connector 40. Optical fiber 4
Reference numerals 1 and 42 denote optical fiber wavelength demultiplexers 43
Is configured. The wavelength demultiplexer 43 has an effect of evanescently coupling the two optical fibers and exchanging light energy depending on the wavelength or not. In this example,
The 1.55 μm light moves to the optical fiber 42 on the branch side,
The 1.3 μm light travels through the optical fiber 41 as it is. This passes through the connector 44 and leads to the optical fiber 45. The optical fiber 45 branches into equivalent optical fibers 46 and 47.
This branch becomes the optical fiber coupler 48. Coupler 4
Reference numeral 8 designates a structure in which two optical fibers are brought close to each other or twisted to shorten the distance between cores so that energy can be exchanged.

The optical fiber 46 for receiving light passes through the multilayer filter 59, passes through the optical fiber 50, and passes through the digital PD.
The light enters the module 51. The optical signal is converted to an electric signal, which leads to a telephone or the like. A digital signal from a telephone or the like passes from a digital LD module 53 to an optical fiber 54, and from an optical connector 52 to an optical fiber 47 and a coupler 48.
Passes through the wavelength demultiplexer 43 and exits to the optical fiber 39.

On the other hand, the 1.55 μm light is separated into the optical fiber 42 by the optical fiber wavelength demultiplexer 43, passes through the multilayer film 55 and the optical connector 56, and
It reaches the analog PD module 58, where it is photoelectrically converted. In this example, both the wavelength demultiplexer and the coupler are constituted only by optical fibers. Since the optical fiber can be bent freely, analog and digital circuits can be arranged at electrically convenient positions. In addition, since the components can be joined by fusing the optical fiber, a module can be freely configured. Functions can be distributed.

[0026]

An example of an ONU optical module of a mirror type, an optical waveguide type and an optical fiber type has been described. All of these required a wavelength demultiplexer for separating 1.3 μm light and 1.55 μm light, and a coupler for integrating the transmission light and the reception light into one path. For this reason,
An extremely large number of parts had to be provided. Therefore, the optical module for ONU becomes complicated and expensive.
If the wavelength demultiplexer or the coupler can be omitted, an inexpensive optical transceiver module with a simple structure should be able to be obtained. An object of the present invention is to provide an optical transceiver module having a simpler structure without a wavelength demultiplexer.

[0027]

SUMMARY OF THE INVENTION The present inventor has determined why an optical module for ONU has such a complicated configuration. Can't be further simplified? I thought. As a result, the following was found. All modules are prejudiced by the fact that the 1.3 μm light and the 1.55 μm light must be separated in advance, which has forced a complicated configuration. In order to make the structure simpler, it is necessary to remove the wavelength demultiplexer and the coupler. If 1.3 μm light and 1.55
If it is not necessary to separate μm light, the wavelength demultiplexer can be omitted. This is natural. So why was it necessary to separate light of both wavelengths in the past? Consider this.

This is because there is actually a problem in the structure of the light receiving element. Unexpectedly, the light receiving element complicates the ONU optical module. The inventor came to this point for the first time. Since it is difficult to understand, the structure of the conventional light receiving element will be described.

FIG. 5 is a sectional view of a photodiode chip used in a conventional module. n-type InP substrate 6
0, an n-type InP buffer layer 61 and an n-type InGaAs
The s light receiving layer 62 (light absorbing layer) and the n-type InP window layer 63 are formed by an epitaxial growth method. InP window layer 63
A passivation film 66 of an insulator is provided on the outer peripheral portion of the upper part. Zinc (Zn) as a p-type impurity is diffused from the center of the window layer 63 to the middle of the n-type InGaAs light receiving layer 62. A pn junction is formed in the middle of the InGaAs layer.

A p-electrode 65 is provided around the zinc diffusion region 64.
Are provided for ohmic connection. It is an annular electrode. Light enters a portion 69 surrounded by the annular electrode 65. The light incident portion 69 is covered with an antireflection film 67. An n-electrode 68 is provided on the bottom surface of the n-type InP substrate 60 on the opposite side.
Is formed. The electrode covers the entire surface because no light enters from this side.

The p-electrode 65 is biased negative and the n-electrode 68 is biased positive. When the incident light enters the InGaAs layer, it is absorbed there and generates electron-hole pairs. Electrons flow toward the n-electrode and holes flow toward the p-electrode. This is the photocurrent. InGaAs is used for the light absorption layer because it has a high sensitivity to infrared light.

FIG. 6 shows the light receiving sensitivity of this light receiving element. The horizontal axis is the wavelength (μm). The vertical axis is the sensitivity (A / W). It has high sensitivity over a wide wavelength range from 1.0 μm to 1.6 μm. That is, this light receiving element has a high sensitivity to both 1.3 μm and 1.55 μm. The wavelength range having high sensitivity is determined by the material of the light absorbing layer. In this case, it is determined by the material characteristics of InGaAs.

The lower limit λa of the wavelength at which the light receiving element has high sensitivity is determined by the absorption characteristics of the window layer immediately above the InGaAs absorption layer 62. Since the window layer is InP in this case,
Λa is determined from the band gap. The upper limit λb of the wavelength having sensitivity is determined by the absorption layer. The reason will be described. Semiconductors are characterized by a band gap Eg. The energy hc / λ of the incident light has a band gap E
If it is smaller than g (λ> hc / Eg), this light is transmitted as it is because an electron-hole pair cannot be generated. Ideally, the transmittance is 100%. The semiconductor is transparent to this light.

On the contrary, when the energy hc / λ of the incident light is larger than the band gap Eg (λ <hc / E
g), this light creates an electron-hole pair and disappears. That is, this light is absorbed. The wavelength corresponding to the band gap is hc
/ Eg. This is expressed as λg. The band gap Eg is also called a basic absorption edge. λg can be said to be a wavelength corresponding to the fundamental absorption edge. It can be simply called the absorption edge wavelength λg. In summary, when the wavelength λ of the incident light is shorter than the absorption edge wavelength λg (λ <λg), it is completely absorbed, and when it is longer than λg (λ> λg), it is completely transmitted. The transition from absorption to transmission is quite steep. However, the curve becomes slightly dull depending on the temperature.

If the absolute value is 0 ° (0K), there are no holes in the valence band and no electrons in the conduction band, so that the absorption is divided into total absorption and total transmission at λg. But at finite temperature, due to heat,
Some holes are excited in the valence band and some electrons are excited in the conduction band. For this reason, a transmission curve or an absorption curve draws a curve near λg. The light receiving element is made by combining two types of semiconductors, a window layer and an absorption layer. Incident light passes through the window layer,
Since it enters the absorption layer, the band gap Eg of the window layer
1 must be wider than the band gap Eg2 of the absorption layer (light-receiving layer). Eg1> Eg2. The reason is as follows.

Assuming that the energy of the incident light is Ep = hc / λ, it is necessary that Ep <Eg to pass through the window layer (Eg1).
Must be 1. In order for this to be absorbed by the absorption layer (Eg2) (for sensitivity), Ep> Eg
Must be 2. In order for this to be compatible, Eg1
> Eg2. That is, it is necessary to use a material having a band gap larger than the band gap of the absorption layer for the window layer. If the window layer is made of InP, it is necessary to use a semiconductor having a narrower band gap for the absorption layer. For this purpose, InGaAs is used for the absorption layer.

This light receiving element can detect only the light that passes through the window layer and reaches the absorption layer and is absorbed therein. That is, only light that satisfies the inequality Eg1>Ep> Eg2 is detected by this detector. Expressing the same by wavelength, if λg1 <λ <λg2, this light is detected by the light receiving element. The lower limit λa = λg1 and the upper limit λb = λg2 of the high sensitivity region. InG
A light receiving element having aAs as an absorption layer and InP as a window layer has a wavelength of λ.
g1 <1.3 μm <1.55 μm <λg2,
It has high sensitivity to both 1.3 μm and 1.55 μm light. I
The absorption edge wavelength λg1 of the nP window layer is 0.92 μm, InGa
The absorption edge wavelength λg2 of the As absorption layer is 1.67 μm.

In FIG. 1, a light receiving element having the same combination of semiconductors is used for the analog PD 5 and the digital PD 9. That is, a light receiving element having sensitivity to both 1.3 μm and 1.55 μm is used for both. For this reason, when 1.3 μm light is mixed with the analog PD 5, this light receiving element becomes 1.
Even 3 μm light is detected. Therefore, 1.3 μm light becomes noise. Conversely, 1.55 μm for digital PD9
When the light is mixed, the light receiving element detects this and causes noise. The same applies to the conventional examples shown in FIGS.

Similarly, the photodiodes for digital signals (1.3 μm light) and analog signals (1.55 μm light) have 1.3 μm and 1.55 μm.
Since a photodiode having sensitivity to μm is used,
The 3 μm and 1.55 μm optical paths must be separated. Wavelength demultiplexers are required for optical path separation. However, all existing wavelength demultiplexers have insufficient separation ratios. It is impossible to prevent the light of the other party's wavelength from being mixed. For this purpose, the multilayer filters 15, 1
6, 34, 35, 59 and 55 had to be interposed in each optical path. Since the optical paths of the 1.3 μm light and the 1.55 μm light are to be separated, such a complicated structure is obtained.

On the contrary, if the optical paths are not divided, the structure should be greatly simplified. It suffices if 1.3 μm light and 1.55 μm light travel along the same optical path and can be separately detected. Is such a magic trick possible? That is possible. In the first place, the conventional example spatially separates 1.3 μm light and 1.55 μm light, that is, separates the optical paths because the light receiving element has no wavelength selectivity. 1.35m light is 1.55
Since the light of μm is also sensed, it is necessary to separate the optical paths.

If there is a light-receiving element that absorbs and senses only 1.3 μm light and transmits 1.55 μm light, there is no need to spatially separate the optical paths. Such a light receiving element itself is a novel one. If such a light receiving element is present, a light receiving element for receiving 1.55 μm light is provided on the same optical path as the light receiving element and at a position behind the light receiving element to provide 1.3 μm light and 1.55 μm light. Light can be detected independently without any influence of others. Here is the gist of the present invention.

In the light receiving system of the present invention, the first light receiving element and the second light receiving element are arranged in series on the same optical path, and the first light receiving element is 1.3.
absorbs all the 1.3 μm light and detects 1.55 μm
m light is all transmitted. The second light receiving element is 1.5
5 μm light is perceived. After all, there are two important points of the present invention. One is that two light receiving elements are arranged in series. The other is that the first light receiving element absorbs all the 1.3 μm light, has no sensitivity to the 1.55 μm light,
This means that all μm light is transmitted. The first light receiving element must have a special property. To list its properties,

Sensing 1.3 μm light. Absorb all 1.3μm light. Not have sensitivity to 1.55μm light. Pass all 1.55μm light.

Is as follows. This is a new property in itself. It can be said that a light receiving element satisfying such a condition has not existed so far. The light receiving element itself is new. The light transmitting / receiving module of the present invention is a combination of a novel light receiving element and a light receiving element capable of detecting 1.55 μm light in series.

The relationship between the semiconductor band gap and light absorption has been described above. The above is inherently different in nature. Therefore, four parameters are required to satisfy all conditions. However, in the case of a semiconductor, the properties described above can be satisfied at the same time only by specifying one parameter. The present inventor has first noticed such specialty of a semiconductor.

As described above, the semiconductor absorbs light having energy equal to or higher than the band gap, and transmits all light having energy lower than the band gap. Therefore, the upper limit wavelength λb of the sensitivity region of the light receiving element is determined by the absorption edge wavelength λg2 of the absorption layer (light reception layer) (λb = λg2), and the lower limit wavelength λa is determined by the absorption edge wavelength λg1 of the window layer (λa = λg1).

The upper limit wavelength λb is further reduced to 1.3 μm
By making it longer and smaller than 1.55 μm,
Condition can be satisfied. That is, the band gap E of the absorption layer
g2 is not less than 1.55 μm light energy and 1.3 μm
By setting the energy to be equal to or less than the energy of light, a light receiving element that receives light of 1.3 μm and does not feel light of 1.55 μm can be obtained. This is because 1.55 μm light is not absorbed at all. Expressed by the absorption edge wavelength λg of the absorption layer,

1.3 μm <λg2 <1.55 μm (1)

That is to say. This most clearly expresses the present invention. Further, by making the thickness appropriate, it is possible to absorb all 1.3 μm light and prevent it from leaking. By making a hole in the center of the n-electrode on the back of the substrate, all of the 1.55 μm light can be transmitted. This condition can be satisfied. The material of the window layer is also a problem. The band gap Eg1 of the window layer is 1.3 μm
Make it shorter than light. Since a wavelength shorter than 1.3 μm must not be incident, the band gap of the window layer is set to be slightly higher than this. This is easy

Λg1 <1.3 μm (2)

Can be expressed as follows. Thus, 1.3
The first light receiving element that senses μm, absorbs all 1.3 μm light, and transmits 1.55 μm light, receives the 1.55 μm light in the second light receiving element.
An optical receiving module according to the present invention includes light receiving elements connected in series. This is an improvement of the receiving system. With this improvement, it is not necessary to separate the optical paths of the 1.3 μm light and the 1.55 μm light, so that a wavelength demultiplexer becomes unnecessary. However, a laser for transmitting digital light is separately required. The configuration in which the coupler introduces laser light into the optical fiber remains unchanged.

The first advantage of the present invention is that the wavelength demultiplexer can be omitted. Since the configuration is simple, miniaturization and cost reduction are possible. It can greatly promote the development of optical subscriber systems. However, not only that, since the light receiving element itself has wavelength selectivity, 1.55 μm light does not interfere with the 1.3 μm light signal.
The 1.3 μm light does not enter the 55 μm light. That is, there is an advantage that noise can be greatly reduced and high-quality images and sounds can be received.

[0053]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The most important point of the present invention lies in a first photodiode. This is a light receiving element that absorbs all 1.3 μm light, detects 1.3 μm light, is insensitive to 1.55 μm light, and transmits 1.55 μm light. The structure of such a photodiode will be described. FIG. 7 is a sectional view of such a photodiode. n-type InP substrate 70
, An n-type InP buffer layer 71 and an n-type InGaAs
P light receiving layer (absorption layer) 72, n-type InGaAsP window layer 73
Are formed by epitaxial growth. InG
The absorption edge wavelength λg2 of the aAsP light receiving layer 72 is 1.42 μm.
It is. The absorption edge wavelength λg1 of the InGaAsP window layer 73 is 1.15 μm.

The InGaAsP window layer 73 and the InGaAsP
At the center of the light receiving layer 72, a p-type region 79 is formed by zinc diffusion. In the p-type region 79 on the upper surface of the InGaAsP window layer 73, an annular p-electrode 74 is provided so as to make ohmic connection. The outside of the ring electrode 74 is covered with a passivation film 75. The inside of the annular p-electrode 74 is covered with an antireflection film 76. Incident light 80 passes through antireflection film 76 and enters p-type region 79.

The rear surface of the n-type InP substrate 70 is formed such that the annular n-electrode 77 is in ohmic connection. This does not cover the entire bottom surface of the substrate. It is formed in an annular shape only in the peripheral portion. There is no electrode in the center and light passes through. An antireflection film 78 is formed in a portion through which light passes.

Such a light receiving element differs from the conventional light receiving element in FIG. 5 in the following three points. The window layer is changed from InP (λg = 0.92 μm) to InGa
It has changed to AsP (λg = 1.15 μm). The light receiving layer (absorption layer) is made of InGaAs (λg = 1.67 μm).
m) to InGaAsP (λg = 1.42 μm).

There is a wide opening 81 at the bottom of the substrate that allows 1.55 μm light to pass through. FIG. 8 is a graph showing the light transmittance of the absorption layer (1.3 μm band selective epitaxial layer). The horizontal axis is the wavelength (μm), and the vertical axis is the transmittance (relative%). Since λg of the absorption layer is 1.42 μm, all light having a wavelength longer than this is transmitted. Light of shorter wavelength is absorbed. 1.3 μm light is absorbed by the light receiving layer (absorption layer 72),
1.55 μm light is transmitted. However, 1.
In order to absorb all the light of 3 μm, the thickness of the absorbing layer must be somewhat large.

The upper limit of the wavelength of light that can be detected by the light receiving element is:
The lower limit is determined by λg2 (1.45 μm) of the absorption layer, and the lower limit is determined by λg1 (1.15 μm) of the window layer. This light receiving element can sense only light having a wavelength in a narrow range (1.15 μm to 1.42 μm). FIG. 9 is a graph showing the wavelength sensitivity characteristics of this photodiode. 1.15 μm to 1.42 μ
Since the light is felt only in the range of m, the 1.55 μm light is not detected at all. Although the 1.55 μm light passes through this photodiode, there is no sensitivity, so the 1.55 μm light does not become noise. Conversely, 1.3 μm light can be detected.

From the transmittance curve of FIG. 8, it can be seen that 1.55 μm light can be completely transmitted through this light receiving element.
These properties make the upper limit λb of the sensitivity region 1.55 μm
Obtained by moving it down. The above gives this property. Λg2 of the absorption layer is generally
It may be any value between 1.3 μm and 1.55 μm. Here, 1.42 μm is selected as an example. It is just an intermediate value.

The λg1 of the window layer generally needs to be smaller than 1.3 μm. Here, 1.15 μm is selected. However, λg of InP = 0.92 μm may be used. Further, the opening 80 on the back surface of the substrate is indispensable for extracting 1.55 μm light. Only 1.55 μm light exits the aperture, and all 1.3 μm light is absorbed by the absorbing layer.

The 1.3 μm light selective light receiving element of the present invention
A quaternary mixed crystal of nGaAsP is used. This is because a ternary mixed crystal cannot provide a band gap freely. 4
Based mixed crystal can be written as _{In 1-x Ga x As 1} -y P y. There are two parameters that determine the composition. x and y. The buffer layer, the light absorption layer, and the window layer are all substrates of In.
This gives one constraint, since it must be lattice matched to P. However, since there is another degree of freedom, a band gap can be freely provided. For the composition of mixed crystals, band gaps, lattice constants, etc., see Tetsuji Imai et al. "Compound Semiconductor Device (I)"
It is described in P56, P87, 1984, issued by the Industrial Research Institute Co., Ltd. The band gap can also be represented by wavelength (Eg = hc / λg).

If the absorption layer is λg = 1.42 μm, the composition is determined as x = 0.34 and y = 0.24. In _0.66 G
a _0.34 As _0.76 P _0.24 is the composition of the absorbing layer. In this case, the 1.55 μm light is transmitted without being absorbed. 1.3 μm light is completely absorbed. Assuming that the window layer λg = 1.15 μm, the composition is determined to be x = 0.18 and y = 0.60. The composition of the window is In _0.82 Ga _0.18 As _0.40 P _0.60 . It transmits both 1.3 μm and 1.55 μm light. It is natural because it is a window. However, this prevents light of shorter wavelengths from entering.

A method of manufacturing the PD chip shown in FIG. 7 will be described.
On a 350 μm thick InP substrate, a 2.5 μm thick InP buffer layer, 4.5 μm InGaAsP (λg
= 1.42 μm) Light receiving layer, 1.5 μm thick InGaAs
An sP (λg = 1.15 μm) window layer is formed in this order by a liquid phase epitaxial method. FIG. 8 is a graph showing the results of measuring the light transmittance of the epitaxial wafer thus formed. Light having a wavelength of 1.42 μm or less is totally absorbed, and light having a wavelength longer than 1.42 μm has a high transmittance. At 1.42 μm, the transmittance changes sharply.

The PN junction is formed by diffusing zinc (Zn) into the wafer. Further, a passivation film, an antireflection film, a p-electrode, an n-electrode, and the like are formed by photolithography. However, an n-electrode is formed annularly on the back surface of the chip. The central part is an opening through which light passes. Further, an antireflection film (for example, SiO ₂ ) for 1.55 μm light is coated on the center of the back surface surrounded by the n-electrode. On the surface:
An anti-reflection film for both 3 μm light and 1.55 μm light is formed (for example, a multilayer film of SiO ₂ and SiN).

The PD of the present invention thus produced was assembled in a package, and a voltage of 5 V was applied as a reverse bias to measure the wavelength sensitivity characteristics. FIG. 9 shows the measurement results. 1.
It can be seen that it has high sensitivity to 3 μm light and transmits 1.55 μm light without absorbing it. The above-mentioned light receiving element,
The thin film layer is made by liquid phase epitaxial method,
In addition, a chloride VPE method can be used.

[0066]

【Example】

[Example 1: When using a mirror-type coupler: Part 1]
Next, two embodiments using a mirror type coupler will be described. FIG.
Reference numeral 0 indicates an embodiment in which a mirror type coupler is used. 1.3 μm light transmitted from the base station / 1.
After passing through the optical connector 91, the 55 μm light passes through the free space c.
And is converted into parallel light by the collimator lens 92, and passes through or is reflected by the mirror type optical coupler 93. The reflected light goes to the LD 94, but this light has no role. The coupler 93 is for passing the reception light and the transmission light through the same optical path. Does not have wavelength selectivity.

A digital transmission signal from a telephone or a facsimile is converted from electric to light by a digital LD 94 and converted into a parallel light d by a condenser lens 95. This is reflected by the mirror-type coupler 93 and enters the optical path c. This enters the optical fiber via the optical connector 91 and is transmitted to the base station.

The received optical signal travels along the optical path e, is converged by the condenser lens 96, and enters the 1.3 μm light selection PD 97. The 1.3 μm light selective PD has been described with reference to FIGS. 7, 8 and 9. All 1.3 μm light is absorbed and detected, and all 1.55 μm light is transmitted. The PD 97 detects a 1.3 μm light digital signal.

The 1.55 μm light is transmitted through the PD 97, passes through the optical path f, and is incident on an analog PD 98 for 1.55 μm light located on the rear side on the axis. The PD 98 detects the intensity of 1.55 μm light. As the analog PD 98 for 1.55 μm light, a conventional photodiode can be used. For example, a conventional photodiode using InGaAs as an absorption layer is:
It has sensitivity over a wide range of 1 μm to 1.6 μm.
However, since 1.3 μm light has already been removed in the optical path f, only 1.55 μm light enters the PD 98.

As described above, the PD97 for 1.3 μm light and the PD98 for 1.55 μm light are arranged in series on the same axis, and the 1.3 μm light for digital is detected by the former and the 1.55 μm light for analog is detected by the latter. I am trying to do it. Conventionally, the two have been separated into an optical path a and an optical path b by a wavelength demultiplexer. In the present invention, this can be put together in the optical path e. Since there is no wavelength demultiplexer and the optical path is simplified, the structure is extremely simple.

1.3 μm light / 1.5 by a wavelength demultiplexer
When the light of 5 μm is separated, the separation ratio becomes insufficient and a dielectric multilayer film is required. However, the present invention does not require a dielectric multilayer film.
This is because the 1.3 μm light selective PD can absorb all 1.3 μm light. This also simplifies the configuration. The number of parts can be reduced, and a small and low-cost ONU optical module 99 can be obtained. This is one in which an analog PD is built in a module.

[Embodiment 2: In the case of using a mirror type coupler: part 2] Another embodiment using a mirror type coupler is shown in FIG. The fact that the transmission light of the digital LD 94 is reflected by the mirror type optical coupler 93 and guided to the optical fiber, and the reception light from the optical fiber passes through the coupler 93 and is guided to two PDs 97 and 98 arranged in series is different from the previous example. Absent. The difference from the previous example is the position of the analog PD. 1.
The optical fiber 100 is located at the position where the 55 μm optical analog PD 98 was located, and 1.55 μm light is sent to a device at another position, and is detected by the 1.55 μm optical PD provided here. Such a device is suitable when there is a device that detects an analog signal outside the ONU optical module 101.

Both the first and second embodiments have no dielectric multilayer filter. This is a 1.3 μm light PD as described above.
97 absorbs all 1.3 μm light, and 1.3 μm
This is because there is no m light. However, 1.3μm optical PD
If there is 1.3 μm light behind 97
A filter for cutting 1.3 μm light may be inserted immediately after D97.

[Embodiment 3: In the case of using a planar optical waveguide:
Part 1] FIG. 12 shows an embodiment in which a planar optical waveguide is used as an optical coupler. An optical signal from a single mode optical fiber 110 connected to a base station enters an optical waveguide 111 formed on a quartz substrate, and is split into two by an optical coupler 120 formed here. Waveguide 112 leads to branch waveguides 113 and 114. Waveguides 112 to 114
Is the same as that shown in FIG. In the quartz clad layer, a portion having a slightly higher refractive index is formed on a continuous line by impurity doping.

The received light that has passed through the waveguide 113 is arranged outside the optical waveguide 111 in series, 1.3.
The light is incident on the μm light selection PD 115. Here, the 1.3 μm light is subjected to light / electric conversion and becomes an electric signal. Only 1.55 μm light is 1.55 μm behind the 1.3 μm light selection PD 115.
It enters the μm light PD 116. The 1.55 μm light is photoelectrically converted by this PD. This means that the 1.3 μm light photodiode 115 also has 1.5 μm inside the package 119.
It accommodates both the photodiodes 116 for 5 μm light.

The transmission system is the same as in the conventional example. Telephone and facsimile digital signals are subjected to electrical / optical conversion by the laser LD 118. The transmission signal is transmitted to the condenser lens 11
7, the light is incident on the end face of the waveguide 114, and is guided from the waveguide 112 to the optical fiber 110. This is sent to the base station. Of the received light, the light that has proceeded to the waveguide 114 reaches the LD 118, which is merely absorbed and has no role. This ONU optical module does not require a wavelength demultiplexer and has a small number of products. Therefore, the structure is simple and the manufacturing cost is reduced. Since the coupler is constituted by the optical waveguide, further miniaturization is promoted.

[Embodiment 4: In the case of using a planar optical waveguide:
Part 2] An embodiment using another planar optical waveguide will be described with reference to FIG. This is one in which no analog PD is provided in the package and 1.55 μm light is extracted to the outside by an optical fiber. Other configurations are the same as those in FIG. Photodiode 1 for 1.3 μm light
A condenser lens 122 is provided behind the optical fiber 15, thereby condensing 1.55 μm light and entering the optical fiber 121.
The 1.55 μm light is transmitted to an analog PD provided at a different location, and the 1.55 μm light is detected there.

[Embodiment 5: Embodiment using an optical fiber coupler] An embodiment using an optical coupler for coupling transmitted and received light will be described with reference to FIG. The input single mode optical fiber 130 is coupled to the ONU optical module by the optical connector 131. Optical fiber 1 for optical connector
32 are connected. The optical fiber 132 is connected to the optical fibers 133 and 134 in the optical fiber coupler 135.
Has branched to. The optical fiber 133 has 1.3 μm light /
It is connected to the 1.55 μm optical PD module 136. This is a photodiode module in which a 1.3 μm light PD is provided in a preceding stage and a 1.55 μm light PD is provided in a subsequent stage.

The other optical fiber 134 is connected to a 1.3 μm optical LD module 137 for sending out a digital signal. Half of the light from input fiber 130 enters optical fiber 134, which is not received and is useless. Since this embodiment is constituted by optical fibers, the positions of the parts are arbitrary. High degree of freedom in arranging parts.

Example 6: 1.3 μm / 1.55 μmP
D Module] The overall configuration of the ONU optical module has been described above. Important components used in the module are also described. An example of the configuration of a PD module used in the optical fiber coupler system of FIG. 14 will be described with reference to FIG. Unlike a normal light receiving element module, two light receiving elements are built in. Circular header 140
, Four lead pins 141, 142, 143 are attached. Transparent PD on ridge 145 of header
Submount 146 is fixed. On top of this is 1.3
The μm light wavelength selection PD chip 147 is fixed.

At the center of the header 140, an inclined surface 148 is provided.
, And the submount 149 is fixed here. 1.55 μm optical PD chip 15 on submount 149
0 is attached. The incident surface of the PD chip 150 is inclined with respect to the beam line. This is to prevent the reflected light from returning to the original path. Header 140
A cap 151 having a U-shaped cross section is welded to the upper surface of. The cap 151 surrounds the 1.55 μm optical PD chip 150 and the 1.3 μm optical PD chip 147. A ball lens 152 is fixed to the center of the cap 151. The center axis of the lens 152, the center of the 1.3 μm PD chip 147, and the center of the 1.55 μm PD chip 150 are on the same axis.

Further, a cylindrical ferrule holder-153 is welded to the header 140 at a position surrounding the cap. The distal end of the ferrule holder is reduced in diameter, and a through hole 154 is formed in the narrow diameter portion in the axial direction. A ferrule 156 for holding the tip of the optical fiber 155 is inserted into the through hole 154 and fixed at an appropriate height. The end face 157 is cut so as to form an angle of 8 ° with the plane perpendicular to the axis. This is to prevent the reflected light from the end face from returning inside the optical fiber and returning to the light source.

A bend limiter 158 made of a conical elastic material is attached to the small diameter portion of the ferrule holder-153. This has the effect of preventing excessive bending at the distal end of the holder of the optical fiber 155. The characteristics of this light receiving element module are 1.3 μm light PD 147 and 1.55 μm
The optical PDs 150 are arranged in a straight line. The two light receiving elements 147 and 150 are coupled to an optical fiber 155 by a lens 152.

The reason why the 1.55 μm light PD 150 in the latter stage is inclined is to prevent reflected light from returning to the light source. In this example, the 1.3 μm light PD 147 is not inclined with respect to the axis. However, the 1.3 μm light PD can also be tilted. The top surface of the table 145 may be inclined.
Even in this case, it is possible to make the transmitted light incident on the 1.55 μm light PD. Next, a method of manufacturing the PD module having the above configuration will be described.

As the header 140, iron, Kovar,
A metal material such as copper tungsten is used. Here, a Kovar header 140 having four lead pins is used. An inclined surface 148 is formed in the center of the surface without the pin in advance. Ceramic (Alumina Al ₂ O
₃ ) The submount is soldered to the inclined surface 148 using a soldering agent. For example, the solder includes gold tin (AuSn) solder, tin lead (SnPb) solder, gold germanium (AuGe) solder, and the like. Here, gold tin solder is used.

Further, a photodiode having a sensitivity in the 1.55 μm band (the light receiving layer is InGaAsP) is soldered on the submount 149. 1.55μ by gold wire
The n-electrode and p-electrode of the m-light PD 150 are electrically connected to respective lead pins. Externally through lead pin
The electric signal of D can be extracted. One of the four lead pins 141 of the header is a common ground terminal. The above steps are the same as the conventional photodiode manufacturing steps. There are many types of submount, solder, and header materials. There are various electrical connection methods. What is shown here is only an example.

What will now be described relates to the assembly of new structural parts. The header having the ridge 145 is already used for a laser or the like. A submount 146 made of alumina (or aluminum nitride (AlN) or the like) may be mounted on the ridge 145. The submount 146 is metallized on the entire surface and then gold-plated thereon. The submount has a cutout in the center for transmitting 1.55 μm light, and is therefore generally U-shaped. The width of the cutout portion is slightly larger than the light receiving area of the 1.3 μm light (wavelength selective) PD chip.

In this example, the light receiving diameter of 200 μm is 1.3.
A μm optical PD chip 147 is used. Therefore, the width of the notch of the submount is set to 250 μm. The depth of the notch is 500 μm, which is the same as the size of the chip. Here, a sub-mount that is entirely metallized is used, but a partially-metallized sub-mount may be used. The cut-out portion does not have to be square, and may have a circular cut-out portion.

The thickness of each of the submounts 146 and 149 is 500 μm in this example. The submount 146 having such a U-shaped notch is fixed to the top of the raised portion 145 such that the notch is located on the center axis. On the submount 146 shown in FIG.
The 3 μm light wavelength selection PD chip 147 is positioned so that the light receiving surface of the chip overlaps the notch, and soldered with gold tin solder. The submount 146 is mounted on the header 1 so that the PD chip 147 is positioned on the center axis.
On top of the 40 ridges (poles) 145, tin lead (SnP)
b) Soldering with solder.

Since the sub-mount metallized on the entire surface (front and back) is used, the sub-mount is soldered to the raised portion of the header and, at the same time, is electrically connected to the ground on the n-electrode side. The positioning of the submount 146 at this time is easily performed by making the outer corner of the raised portion 145 flush with the corner of the submount. p
The electrodes are connected by gold wires to lead pins 143 that extend to approximately the same height as the ridges 145. Three pins other than the ground pin 141 are fixed to the through holes of the header 140 by the insulator 160.

Next, the cap 1 having the spherical lens 152
51 is pressed against the header 140 and an inert gas (for example,
(N2, Argon, etc.) and weld the cap to the header. The inside of the cap is hermetically sealed by completely welding the periphery of the cap. In this example, a spherical lens is made using BK-7 glass and fixed to a Kovar cap. Such a cap with a spherical lens has been often used conventionally when assembling a PD. Such a cap can be secured to the header by electrode welding. Also in this embodiment, the cap is fixed to the header by electric welding. In addition, YAG welding may be performed.

Since the PD chip has a sufficiently wide light receiving surface, there is no need for positioning in a plane perpendicular to the optical axis. Such a light receiving element module can be used as the 1.3 / 1.55 μm optical PD module 136 of the embodiment of FIG. Such cylindrical PD modules are already widely used, but all have only one light receiving element. The present invention requires a light receiving element module in which two light receiving elements are arranged in series. Since such a light receiving element module itself is new, it has been described in detail here.

[Embodiment 7: Case of Using Half Mirror Type Wavelength Selective Coupler] In the embodiments described above, each of the optical couplers divides incident light into 50%: 50%. The 1.3 μm light is also 1.55 for the optical coupler.
Each of them contains the light of μm and is divided into two equal parts. This coupler is indispensable for the ONU to transmit and receive 1.3 μm light for bidirectional communication.

However, when viewed from the 1.55 μm light, the existence of the coupler is merely a hindrance to halving the incident light. 1.
It only receives analog signals of 55 μm light. TV
It is inconvenient for couplers to reduce power by half for 1.55 μm light that only transmits signals and the like. 50
The% loss is a -3 dB loss. When converted into the loss of the optical fiber, it corresponds to a distance close to 10 km. This loss is caused by the insertion of the coupler.

In some cases, there may be a demand for receiving 1.55 μm light without reducing the light intensity as much as possible. In particular, it seems that such a demand is strong when a TV signal is transmitted to a remote place. In response to such a demand, an embodiment in which a 1.55 μm optical signal can be received even at a long distance will be described below.

FIG. 16 shows the structure of this embodiment. The basic configuration is the same as the embodiment of FIG. However, the difference is that a half mirror having wavelength selectivity is used instead of a simple half mirror. In FIG. 16A, a wiring pattern is printed on a ceramic substrate. The wiring pattern is a metal coating for fixing the element, and is also an electric wiring.

On the substrate 162, a half mirror 163,
A condenser lens 164, an LD 165 for 1.3 μm light, a condenser lens 167, a 1.3 μm light PD 168, and a 1.55 μm light PD 169 are attached. LD16 for 1.3μm light
5 has a monitor PD 176 at the back. An LD / monitor PD lead pin 166 is provided near the LD 165 around the substrate 162. LD165 which is a light emitting element
Since the light emission amount must not fluctuate with time, a monitoring PD is provided behind the LD to monitor the light emission amount.

A single mode optical fiber 171 connected to a base station is optically coupled to a half mirror 163.
The digital transmission signal exits the laser 165, is reflected by the half mirror 163, and enters the optical fiber 171. 1.3 μm light from optical fiber 171, 1.55 μm
The m light is transmitted through the half mirror, is stopped down by the lens 167, and is detected by the 1.3 μm light PD 168 and the 1.55 μm light PD 169. The half mirror 163 is, as shown in FIG. 16B, the multilayer film 1 from the side closer to the optical fiber.
73, a transparent substrate 174, a 1.3 μm / 1.55 μm antireflection film 175 having a triple structure.

The multilayer film 173 on the input side transmits 1.55 μm light almost completely, and 1.3 μm light is transmitted and reflected at a finite ratio. What is required here is that the light does not reflect 1.55 μm light. The multilayer film is formed by alternately and repeatedly depositing, for example, two types of dielectric thin films having different refractive indices. However, by appropriately selecting the refractive index and the thickness, light of any wavelength can be totally reflected or half-reflected. Or, it can be made to transmit all. Here, the material and thickness of the thin film may be designed under the condition that the reflection of 1.55 μm light is reduced.

The reflection of 1.55 μm light is reduced to 0, 1.3 μm
It seems that the design of 1: 1 light makes it difficult to design and manufacture. However, it is possible to design a half mirror under such conditions. Further, it is not really necessary to split 1.3 μm light at a ratio of 1: 1. In many cases, the intensity of the received light is sufficient for 1.3 μm light, and the intensity of the transmitted light is often insufficient. In this case, the branching ratio of 1.3 μm light is defined as PD
It is more desirable to set 20% on the side and 80% on the LD side. In other words, since conditions relating to 1.3 μm light have a wide range, it is easy to find a condition that meets the purpose of transmitting 1.55 μm light entirely.

Looking back, if the 1.3 μm light and the 1.55 μm light are the same as described above, 50%: 5
It is rather difficult to satisfy the condition of transmitting and reflecting at 0%. This is because two severe conditions are imposed. In practice, it is very difficult to make such a structure using a dielectric multilayer film. Rather 1.55
It is much easier to form a mirror that transmits all μm light with a dielectric multilayer film.

As the material of the multilayer film, SiO ₂ , Al ₂
A dielectric such as O ₃ or TiO ₂ is used. An antireflection film for 1.3 μm light and 1.55 μm light is formed on the back surface of the half mirror. When such a multilayer half mirror is used, 1.55 μm light can pass through the half mirror with almost no loss, and a 1.55 μm light P
D. The noise is reduced and the TV image becomes sharper.

As shown in FIG. 16C, the 1.3 μm light PD 168 is soldered to a ceramic (Al ₂ O ₃ , AlN, etc.) submount 178 having a light transmission window 179 at the center. The submount is provided with a metallized pattern of gold (Au) and is connected to a wiring pattern (not shown) on the substrate 162 that fixes the entire optical system. This pattern is composed of a lead wire for PD 1
70. 1.55 μm light PD1
Similarly, each electrode of the 69 μm and 1.3 μm optical LD 165 also has lead pins 166,
170 and the like.

The 1.3 μm optical LD 165 includes an LD itself for generating an optical signal and a monitoring PD for monitoring the LD. The LD and the monitor PD are set on a heat sink.
This is made of, for example, SiN having good heat conductivity. The heat sink may be metal. However, in the case of metal, the electrodes of the monitoring PD and the LD must be insulated, so that the monitoring PD must be mounted on a submount of an insulator such as Al ₂ O ₃ . The LD generates a large amount of current and generates a large amount of heat. Therefore, heat is dissipated by the heat sink.

The lenses 164 and 167 are fixed to a holder, and the holder is fixed to a substrate. The package supporting all of these components is made of metal. For example, iron, Kovar, brass and the like. This embodiment corresponds to the conventional example shown in FIG. Similarly, a mirror is used for the coupler, but the embodiment of the present invention has a much simplified configuration. The number of parts is reduced, the structure is simple, and it can be manufactured at low cost. Moreover, the performance is not different from that of FIG. Why can such a big effect be obtained? This is because anyone who had believed that 1.3 μm light and 1.55 μm light separate the optical path had been designed from the completely opposite idea.

[Embodiment 8: Using a Wavelength-Selecting Planar Waveguide] An embodiment in which an optical coupler is constituted by a wavelength-selective planar waveguide will be described with reference to FIG. The couplers in FIGS. 12 and 13 use a planar waveguide, but have no wavelength selectivity because they use a simple Y-branch. As described in the seventh embodiment, it is expected that the coupler is required to have wavelength selectivity. In this case, the present embodiment is extremely effective. The branching of the waveguide itself has wavelength selectivity.

The substrate 180 is a rectangular plate made of, for example, ceramic. A quartz optical waveguide 181 is mounted in front of this. It has a structure similar to that shown in the sectional view of FIG. Quartz (S
There is a cladding layer of iO ₂ ) in which a portion (core) having a linearly high refractive index is formed. This becomes a waveguide for guiding light.

In the latter half of the substrate 180, the submount 18
2 is fixed. The submount is formed by metalizing a wiring pattern, a component mounting pattern, and the like on a ceramic plate. On the submount 182, a condenser lens 184,
The 1.3 μm light LD 185 and the monitor PD 186 are fixed on a straight line. Furthermore, 1.3 μm light PD187, 1.
55 μm optical PDs 188 are arranged before and after. Substrate 180
Are provided with lead pins 189, 190, and 191 to connect the electrodes of each element to an external circuit. The end of the single mode optical fiber 192 connected to the base station is connected to a straight waveguide 193 in order to provide the silica-based optical waveguide 181 with received light. In addition, a bent waveguide 195 is formed on the quartz cladding layer.

This waveguide 195 has a part JH approaching the above-mentioned straight waveguide. The waveguide 19
3, 195 combine. Therefore, this part is the coupler 19
It becomes 4. But that's not all. The coupling between the waveguides for the light having the wavelength λ can be arbitrarily specified by the length L and the interval d of the proximity portion JH. That is, the waveguide 193
, The power of the light traveling to 193 itself or to 195 can be determined by the wavelength λ, the length L, and the interval d. The entrance and exit of the straight waveguide are 1 and 2, and the entrance and exit of the bent waveguide are 4 and 3. Transition power from 1 to 2 - F a transition ratio of the ratio from the F _31, 3 to 4 - transition power to the F _12, 1 to 3 ratio of - the ratio of the transition power from F _13, 3 to 1 I will write ₃₄ etc.

Sum rule F ₁₂ + F ₁₃ = 1, F ₂₁ + F ₂₄ =
1, such as F ₃₁ + F ₃₄ = 1 is satisfied. Due to the reversibility of the optical path, there is symmetry such as Fmk = Fkm. The ratio Fmk is a function of the light wavelength λ, the coupler spacing d, and the length L.
That is, it can be generally written as Fmk (λ, d, L). Since it is desirable that 1.55 μm light does not enter the waveguide 195,

F ₁₃ (1.55 μm, d, L) = 0

Is required. Also, 1.3 μm light is 1: 1.
It is better to distribute more to the LD path 195 than to the distribution ratio. Then

F ₁₃ (1.3 μm, d, L) ≧ F ₁₂ (1.
3 μm, d, L)> 0

The following inequality is required. Thus, the length L and the interval d of the coupling portion of the waveguide can be determined. FIG. 17C shows a Mach-Zehnder type coupler. The two waveguides are close at two points A and C. In the middle B, the waveguides are separated to such an extent that the wave functions of light do not overlap. The path ABC is slightly curved and therefore longer than a straight path. The optical path difference Δ is constant. If this is an integer multiple of the wavelength of the light, the coupling at A and C will reinforce each other. If it is a half-integer multiple of different light, the bonds at A and C cancel each other out.

For this reason, the transition probability from the straight road to the curved road ABC differs depending on the wavelength. That is, wavelength selectivity is provided. Since the number of parameters is larger than in the case of (b), the Mach-Zehnder coupler inhibits the transition of 1.55 μm light (F ₁₃ (1.55 μm).
m) = 0) 1.3 μm light is more strongly transitioned (F ₁₃
(1.3 μm)> F ₁₂ (1.3 μm)).

This example has the advantages that it is well compatible with the fiber and can be miniaturized, and that a large number of chips can be simultaneously produced on a Si substrate by photolithography. This is because a quartz waveguide is formed on a Si substrate. Such features are combined with the advantage of the aforementioned wavelength selectivity.

A configuration example of a waveguide type wavelength selective coupler which is important for this embodiment will be described in more detail. 3
One example is shown in FIG. Each shows a high refractive index path (core) formed by continuously doping a high refractive index impurity into the cladding. In this case, the LD wavelength in the 1.3 μm band is 1.31 μm, which is suitable for long-distance transmission.

FIG. 18A shows a symmetric directional coupling type coupler. It is composed of a straight core 200 and a core 201 partially bent in the vicinity of the core. The two cores have the same sectional shape and the same refractive index. The parallel cores thus form a directional coupler. Since the wave function of the light propagating through the core oozes over the core, the power of the light moves to the adjacent core as the light propagates through the core. When two parallel cores have the same cross section and the same refractive index, one core 200
, The power gradually shifts to the other core 201 and eventually completely shifts to the other core 201. This time, the power starts to move from the core 201 to the first core. In this way, light travels alternately through the two cores. If the length L and the gap d are appropriately determined, the optical power P0 can be distributed to the cores 200 and 201 at an arbitrary ratio P1: P2.
The distribution ratio depends on L, d, and λ.

Therefore, P0 = 1.55 μm light
When the power of 1 is input to the end 202 of the core 200,
P1 = 1, P2 = 0, and for 1.31 μm light, when power of P0 = 1 is applied to the end 202, P1 = 0.5 and P2 = 0.5 are equally distributed. L and d are determined as described above.

FIG. 19 is a graph showing the wavelength dependence of the insertion loss of P1 and P2 when L and d are determined as described above. ● represents the insertion loss of P1. A zero insertion loss means that all passes. 1.55 μm light is P1
However, there is an infinite loss for 0 dB and P2.
That is, P1 = 1 and P2 = 0. 1.31
In the μm light, P1 and P2 have the same loss, so P1: P2
= 1: 1. Of course, it is not essential that the ratio be completely 1: 1. As explained earlier, it may be more convenient to couple strongly to the sender. in this case,
P1 / P2 <0.5.

As described above, 1.31 μm light and 1.55 μm
The power of the light can be controlled separately based on the principle that the propagation constant of the light in the core is different when the wavelength is different. The transfer of power between two parallel waveguides is described in, for example, Hiroshi Nishihara et al., "Optical Integrated Circuit" Ohmsha, p. 55-63, published in February 1985.

FIG. 18B shows an asymmetric directional coupling type coupler. As in the previous example, the straight waveguide 210
, A bent waveguide 211 is provided. Proximity part 2
The width S ₁ of the core is smaller in the 12. In this case, the power of light entering from 213 is all
Does not shift to 1. Part of the power remains in the original core even when the maximum shift is made. The maximum power that the 211 can obtain is determined by the difference between the respective propagation constants when each waveguide exists alone. Since the propagation constant is determined by the cross-sectional shape of the waveguide and the refractive index difference, the maximum power obtained by the 211 can be arbitrarily adjusted by appropriately selecting the widths S ₁ and S ₂ of each waveguide. Therefore, for 1.3 μm light, the ratio of the maximum power that shifts to 211 when entering from the end 213 is 50.
% For 1.55 μm light so that it does not shift to 211 at all when incident from the end 213.

That is, as in the previous example, when P0 = 1,
For 1.31 μm light, P1: P2 = 0.5:
0.5 and 1.55 μm light, P1 = 1, P1
That is, 2 = 0. FIG. 20 is a graph of the insertion loss of the asymmetric directional coupling type coupler with respect to P1 and P2.
For 1.31 μm light, P1: P2 = 1: 1, but since the insertion loss of P1 is smaller regardless of the wavelength, the more energy is distributed to P1. Become.

The asymmetric directional coupling type coupler (b) is
Since λ = 1.31 μm and the insertion loss takes an extreme value for both P1 and P2, 1.31 μm is calculated as P1: P2 =
There is an advantage that the wavelength range of 0.5: 0.5 can be broadened. On the other hand, the symmetrical directional coupling type coupler (a) has an advantage that a wavelength range in which P2 = 0 can be set wider for 1.55 μm. This is because P2 takes an extreme value at 1.55 μm. Depending on the purpose, symmetrical and asymmetrical directional junction type couplers are used properly.

FIG. 18C shows a Mach-Zehnder type coupler, which is also shown in FIG. 17C. In this case, the cores exchange power as shown in FIGS. 18A and 18B in the vicinity. Further, there are two proximity parts (directional coupling parts) 218 and 219, and the distance between these parts is
6 and 217, these become new parameters L and L + Δl, and the degree of freedom in design increases.

[Embodiment 9: Embodiment using wavelength-selective optical coupler] Embodiment 8 uses a coupler that provides wavelength selectivity by a planar waveguide. A coupler having the same wavelength selectivity can be constituted by an optical fiber. An embodiment using an optical fiber type wavelength selective coupler will be described with reference to FIG.

A single mode optical fiber 231 connected to a base station is connected to an optical connector 230 provided at the entrance of the ONU optical module. The optical fiber type wavelength selective coupler 232 has a configuration in which a part of two optical fibers 233 and 234 is installed at a certain proximity position 235 so that light propagating through the optical fiber exchanges its power. . It is called wavelength selectivity because the exchange ratio depends on the wavelength. The principle is the same as the coupling between the cores formed in the planar waveguide, and will not be described here.

The distribution ratio may be arbitrarily determined according to the purpose. Here, the distribution ratio of 1.3 μm light is set to 50%: 50%. The 1.55 μm light has a distribution ratio of 100%: 0%. The optical fiber 233 is a 1.3 μm optical PD / 1.
This leads to a PD module 236 containing a 55 μm optical PD. It has lead pins 237 and is connected to an external power supply circuit, amplifier circuit and the like. The other optical fiber 234 is connected to a 1.3 μm optical LD module 238 for transmission. A transmission signal from a telephone or a facsimile is supplied from a lead pin 239 to a laser diode, where it becomes a 1.3 μm light signal. This is an optical fiber 23
4 through the coupler 232 to the single mode optical fiber 231 to reach the base station.

The distribution ratio of 1.3 μm light is not limited to 50:50, and an appropriate distribution ratio should be given according to the output of the laser module 238. If the received signal has a sufficient strength, it is preferable to increase the coupling of the optical fiber 234 so as not to impose a load on the transmission side, such as 30:70 or 20:80. In this embodiment, two light receiving elements are accommodated in the PD module. However, otherwise, the 1.55 μm light PD is moved outside and the 1.3 μm light P
The transmitted light of D may be made incident on an external 1.55 μm light PD via an optical fiber.

[0130]

According to the present invention, 1.3 μm light and 1.55 μm
The optical module for ONU which does not separate the optical paths of 1.3 μm light and 1.55 μm light is provided for the first time, overturning the conventional common sense that the light must be detected after being separated.
1.3 μm so that 1.55 μm light is transmitted without being perceived.
By devising the m-light PD, 1.3 μm light and 1.55 μm light can be independently received without optical path separation.

An element such as a wavelength demultiplexer becomes unnecessary. In addition, since the capability of the wavelength demultiplexer is incomplete, it has conventionally been necessary to provide a dielectric multilayer film to cut off light of unnecessary wavelengths.
Since the present invention gives the 1.3 μm light PD itself a function of absorbing 1.3 μm light, such a multilayer film is unnecessary. The present invention can provide an inexpensive optical transmitting and receiving module that can be easily assembled with a much smaller number of components than conventionally proposed modules. This will greatly contribute to the spread of optical subscriber systems.

[Brief description of the drawings]

FIG. 1 shows a digital optical signal of 1.3 μm light and 1.55 μm.
FIG. 2 is a diagram illustrating a principle configuration of bidirectional communication using an optical analog optical signal.

FIG. 2 is a schematic configuration diagram of a mirror type ONU optical module according to a conventional example using a mirror for a wavelength demultiplexer and a coupler.

FIG. 3 is a schematic configuration diagram of an ONU optical module according to a conventional example using a quartz-based planar waveguide for a wavelength demultiplexer and a coupler.
3A is an overall configuration diagram, and FIG. 3B is a cross-sectional view taken along a line AB in FIG.

FIG. 4 is a schematic configuration diagram of an ONU optical module according to a conventional example in which a wavelength demultiplexer and a coupler are manufactured using optical fibers.

FIG. 5 is a cross-sectional view of a near-infrared InP substrate photodiode chip according to a conventional example. The light absorption layer is InGaAs;
The window layer is InP.

FIG. 6 is a graph showing the wavelength dependency of the light receiving sensitivity of a near-infrared InP substrate photodiode according to a conventional example.

FIG. 7 is a cross-sectional view of a 1.3 μm band selective photodiode chip having sensitivity to only 1.3 μm light and allowing 1.55 μm light to pass through the bottom surface used in the present invention.

FIG. 8 shows InGaAsP (λg = λ) forming a light absorption layer of a 1.3 μm band selective photodiode used in the present invention.
The graph which shows the wavelength dependence of the light transmittance of a 1.42 micrometer layer.

FIG. 9 is a graph showing the wavelength dependence of the light receiving sensitivity of a 1.3 μm band selection photodiode used in the present invention.

FIG. 10: 1.55 μm optical PD using a mirror-type coupler
FIG. 1 is a schematic configuration diagram of an ONU optical module according to a first embodiment of the present invention, which incorporates an optical module.

FIG. 11 is a schematic configuration diagram of an ONU optical module according to a second embodiment of the present invention in which 1.55 μm light is guided to an external PD by an optical fiber using a mirror-type coupler.

FIG. 12: 1.55 μm using a silica-based planar waveguide coupler
ONU according to the third embodiment of the present invention incorporating m-optical PD
FIG. 1 is a schematic configuration diagram of an optical module for use.

FIG. 13 shows 1.55 μm using a silica-based planar waveguide coupler.
FIG. 13 is a schematic configuration diagram of an ONU optical module according to a fourth embodiment of the present invention in which m light is guided to an external PD by an optical fiber.

FIG. 14: Optical fiber coupler and 1.3 μm light / 1.5
FIG. 11 is a schematic configuration diagram of an ONU optical module according to a fifth embodiment of the present invention using a 5 μm optical PD module.

FIG. 15 shows 1.3 μm light / light according to a sixth embodiment of the present invention.
Sectional perspective view of a 1.55 μm optical PD module.

FIG. 16 is a configuration diagram of an ONU optical module according to a seventh embodiment of the present invention using a half mirror type wavelength selective coupler. (A) shows the entire structure, (b) shows the structure of the half mirror, and (c) shows the mounting structure of the 1.3 μm light selective PD.

FIG. 17 is a schematic configuration diagram of an ONU optical module according to an eighth embodiment of the present invention using a waveguide type wavelength selective coupler. (A) is an overall view, (b) is a plan view of a portion near a waveguide, and (c) is a plan view of a Mach-Zehnder type coupler.

FIG. 18 is a structural example diagram of a waveguide type wavelength selective coupler.
(A) shows a symmetric directional coupling type, (b) shows an asymmetric directional coupling type, and (c) shows a Mach-Zehnder type wavelength selective coupler.

FIG. 19 is a graph showing an example of wavelength selection characteristics of a symmetric coupler type waveguide type wavelength selection coupler. The horizontal axis is wavelength (μm),
The vertical axis is the insertion loss (dB). Black circles indicate the output P1, and white circles indicate the output P2. P1 has a minimum value only at 1.55 μm.

FIG. 20 is a graph showing an example of wavelength selection characteristics of an asymmetric directional coupler type waveguide type wavelength selection coupler. The horizontal axis is the wavelength (μ
m) and the vertical axis is the insertion loss (dB). Black circle indicates output P1
, And the white circle indicates the output P2. P1 has a minimum value at 1.55 μm and a maximum value at 1.31 μm. P2 is 1.31μ
Take the minimum value at m.

FIG. 21 is a schematic configuration diagram of an ONU optical module according to a ninth embodiment of the present invention using an optical fiber type wavelength selective coupler.

[Explanation of symbols]

Reference Signs List 1 base station 2 branching device 3 subscriber side terminal 4 wavelength demultiplexer 5 analog PD 6 signal processing unit 7 TV 8 optical coupler 9 digital PD 10 signal processing unit 11 telephone / facsimile 12 digital LD 13 optical connector 14 collimator lens 15 multilayer Film filter 16 multilayer film filter 17 condenser lens 18 condenser lens 22 Si substrate 23 quartz cladding layer 24 waveguide 29 waveguide 30 waveguide 31 waveguide 39 input single mode optical fiber 40 optical connector 43 optical fiber wavelength demultiplexer 44 Optical connector 48 Optical coupler 51 Digital PD module 53 Digital LD module 58 Analog PD module 60 n-type InP substrate 61 n-type InP buffer layer 62 n-type InGaAs light receiving layer (light absorption layer) 63 n-type InP window layer 64 Zn diffusion region 65 p Pole 66 passivation film 67 antireflection film 68 n electrode 69 incident region 70 n-type InP substrate 71 n-type InP buffer layer 72 n-type InGaAsP light receiving layer (λg = 1.42μ
m) 73 n-type InGaAsP window layer (λg = 1.15 μm) 74 p-electrode 75 passivation film 76 anti-reflection film 77 n-electrode 78 anti-reflection film 79 p-type region 80 incident light 81 opening 91 optical connector 93 mirror-type optical coupler 94 Digital LD 97 1.3 μm light selection PD 98 1.55 μm light analog PD 100 output optical fiber 110 input single mode optical fiber 111 optical waveguide 115 1.3 1.3 μm light selection PD 116 1.55 μm light analog PD 118 1.3 1.3 μm light LD 119 Package 120 Optical coupler 121 Output optical fiber 130 Input single mode optical fiber 131 Optical connector 135 Optical fiber coupler 136 1.3 μm light / 1.55 μm optical PD module 137 1.3 1.3 μm optical LD module 140 Header 141 Lead pin 142 Lead pin 143 Lead pin 145 Ridge 146 Submount 147 1.3 μm optical wavelength selection PD chip 148 Inclined surface 149 Submount 150 1.55 μm optical PD chip 151 Cap 152 Ball lens 153 Ferrule holder 154 Through hole 155 Single mode optical fiber 156 Ferrule 157 Inclined end face 158 Bend limiter 163 Half mirror 164 Condensing lens 165 1.3 μm light LD 166 LD / monitor PD lead pin 167 Condensing lens 168 1.3 μm light PD 169 1.55 μm light PD 170 Lead lead 171 input Single mode optical fiber 172 Package 173 Multilayer film 174 Substrate 175 Antireflection film 178 Submount 179 Aperture 180 Substrate 181 Quartz light Wave path 182 Submount 185 1.3 μm optical LD 186 Monitor PD 187 1.3 μm optical PD 188 1.55 μm optical PD 192 Input single mode optical fiber 194 Waveguide type wavelength selective coupler 230 Optical connector 231 Input single mode optical fiber 232 Optical fiber Type wavelength selection coupler 236 1.3 μm light / 1.55 μm light PD module 238 1.3 μm light LD module

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Hiroo Kanamori 1st Taya-cho, Sakae-ku, Yokohama-shi, Kanagawa Pref. Inspector in Yokohama Works, Sumitomo Electric Industries Co., Ltd. Eiichi Yoshida (56) JP-A-3-103804 (JP, A) JP-A-3-179404 (JP, A) JP-A-5-127028 (JP, A) JP-A-6-102429 (JP, A) JP-A Sho 60-175024 (JP, A) JP-A-60-251678 (JP, A) JP-A-61-144888 (JP, A) JP-A-62-121409 (JP, A) (58) Fields investigated (Int. Cl. ^7, DB name) G02B 6/42 H01L 31/02 H01L 31/10

Claims (16)

(57) [Claims] 1. Two wavelength bands λ1, λ2 (λ1 <λ2)
An optical transmitting and receiving module used for performing bidirectional communication using light having a sensitivity in a first wavelength band λ1 to be received and transmitting a light in a second wavelength band λ2 longer than the first wavelength band λ1.
A photodiode, a second photodiode that receives the wavelength band λ2 transmitted therethrough, a semiconductor laser that emits light in the first wavelength band λ1, and an optical coupler that combines transmission light and reception light into one optical path. The bandgap energy Eg1 at the basic absorption edge of the semiconductor layer constituting the first photodiode is larger than the energy of the light of the long wavelength λ2 and smaller than the energy of the light of the short wavelength λ1, and the light is received. An optical coupler, the first photodiode, and the second photodiode are further disposed in this order along the traveling direction of the light to be received, and all the light of the short wavelength λ1 to be received is absorbed by the first photodiode. The light having the long wavelength .lambda.2 to be received passes through the first photodiode and enters the second photodiode. Optical transceiver module, characterized in that the emitted light is arranged to be transmitted through the optical coupler. 2. Two wavelength bands λ1, λ2 (λ1 <λ2).
An optical transmitting and receiving module used for performing bidirectional communication using light having a sensitivity in a first wavelength band λ1 to be received and transmitting a light in a second wavelength band λ2 longer than the first wavelength band λ1.
A photodiode, an optical fiber for transmitting light of the wavelength band λ2 transmitted therethrough, a semiconductor laser for emitting light in the first wavelength band λ1, and a device for combining transmission light and reception light into one optical path. And the bandgap energy Eg1 at the fundamental absorption edge of the semiconductor layer constituting the first photodiode is larger than the energy of the light of the long wavelength λ2 and smaller than the energy of the light of the short wavelength λ1. An optical coupler, the first photodiode, and the optical fiber are further disposed behind the optical coupler and the first photodiode in order along the traveling direction of the light to be received, and all the light of the short wavelength λ1 to be received is received by the first photodiode. The light having the long wavelength λ2 to be absorbed and received passes through the first photodiode and enters the optical fiber, and the light emitted from the semiconductor laser is further cut off by the optical coupler. Optical transceiver module, characterized in that arranged to be sent through la. 3. The optical transceiver module according to claim 1, wherein a mirror-type coupler formed by forming a multilayer film on a glass substrate is used as the optical coupler. 4. The mirror type optical coupler distributes the light of the first wavelength band λ1 at a predetermined ratio and adjusts the second wavelength to approximately one.
The optical transceiver module according to claim 3, wherein a mirror-type wavelength selective coupler that transmits 00% is used. 5. The optical coupler according to claim 1, wherein an optical coupler formed in a planar optical waveguide is used as the optical coupler.
The optical transceiver module according to item 1. 6. The optical waveguide optical coupler uses a waveguide type wavelength selective coupler which distributes a first wavelength at a predetermined ratio and transmits almost 100% of a second wavelength. The optical transceiver module according to claim 5. 7. The optical transceiver module according to claim 1, wherein an optical coupler formed by an optical fiber is used as the optical coupler. 8. The optical fiber type optical coupler, wherein the first wavelength is distributed at a predetermined ratio, and the second wavelength is substantially 100%.
The optical transceiver module according to claim 7, wherein an optical fiber type wavelength selective coupler that transmits light is used. 9. A package having lead pins, a first photodiode fixed on a central axis of the package, and a second photodiode fixed behind the first photodiode on a central axis of the package. A cap which has a transparent window or lens, is attached to the package and hermetically seals the two photodiodes, and a ferrule holder which surrounds the cap and is fixed to the package and fixes the optical fiber and the ferrule. 9. The optical transmitting and receiving module according to claim 7, wherein a light receiving module including a light-receiving module and a wire connecting the electrode of the second photodiode and the lead pin is used. 10. The first receiving photodiode,
It has sensitivity in the 1.3 μm band, has no sensitivity in the 1.55 μm band, and transmits the light, the second receiving photodiode has a sensitivity in 1.55 μm, and the semiconductor laser has a sensitivity in the 1.55 μm band. The light emitting device emits light in a 1.3 μm band.
10. The optical transceiver module according to any one of items 9. 11. The optical transceiver module according to claim 10, wherein the first receiving photodiode chip has a light receiving layer of InGaAsP (λg = 1.42 μm). 12. A first receiving photodiode chip includes an InP buffer layer and InGaAs on an InP substrate.
P (λg = 1.42 μm) light-receiving layer, InGaAsP (λ
g = 1.15 μm) window layer or InP (λ = 0.92 μm)
The optical transceiver module according to claim 10, wherein m) a window layer is formed, and the p / n electrode is formed in a portion except a central portion so as not to hinder light transmission. 13. An anti-reflection film for transmitting light of 1.3 μm to 1.55 μm and preventing reflection is formed on a light incident surface of the first receiving photodiode chip, and 1.55 μm is formed on a substrate surface.
13. The optical transceiver module according to claim 12, wherein an anti-reflection film is formed to transmit m and prevent reflection. 14. A symmetric directional coupling in which an optical waveguide type wavelength selective coupler is formed by bringing two cores having the same cross-sectional area close to each other in parallel so that the energy of propagating light can be mutually exchanged. 6. A container according to claim 5,
Or the optical transceiver module according to 6. 15. An asymmetric directional coupling in which the configuration of an optical waveguide type wavelength selective coupler is such that two cores having different cross-sectional areas are formed close to each other in parallel so that the energy of propagating light can be mutually exchanged. 6. A container according to claim 5,
Or the optical transceiver module according to 6. 16. The configuration of an optical waveguide type wavelength selective coupler is a Mach-Zehnder type coupler in which two cores are formed close to each other in parallel so that the energy of propagating light can be mutually exchanged. The optical transceiver module according to claim 5, wherein the optical transceiver module is provided.