JP2001085729A - Semiconductor light receiving element - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent leaking transmission light or the stray light of transmission light from coming in due to the incompleteness of WDM by installing the first absorbing layer of an InGaAsP on one side of a substrate and installing the second absorbing layer of the InGaAsP above the substrate and just below a light receiving layer. SOLUTION: An InGaAsP layer 67 is formed at the rear face of an n+-type InP substrate 57 as a first absorbing layer and a second n+-type InGaAsP light absorbing layer 58 and an n-type InGaAs light receiving layer 59 are epitaxially grown on the surface side of the substrate. The reception light A of λ2 is made to pass through the absorbing layer 67, the substrate 57 and the absorbing layer 58 from the rear face and it reaches the depletion layer 62 of the light receiving layer 59. The depletion layer 62 absorbs it and a pair of electronic holes are generated. Most of leaking light B of λ1, which enters from the rear face, is absorbed by the first n-InGaAsP absorbing layer 67. The second absorbing layer 58 absorbs the stray light C of λ1 from obliquely below.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention
_{1 relates} to a structure of a light receiving element capable of satisfactorily blocking the effect of the transmission light in an optical transceiver module to transmit and receive the single optical fiber using lambda _2. When one optical fiber is used for both transmission and reception, an LD that emits transmission light and a PD that receives reception light are provided on the same container or on the same substrate. Alternatively, two or more PDs are provided on the same substrate in the case of transmitting in one direction using light of two or more wavelengths with one optical fiber. PD is a highly sensitive element. LDs have intense power to carry signals far away. Although the wavelengths are different, the light receiving element has sensitivity to the transmitted light, so that the light is sensed.

[0002] The same terminal has an LD and a PD, and the PD is L
Sensing the transmitted light of D is called optical crosstalk. Sensing the transmitted light makes it difficult to detect the received light. Therefore, it is necessary to suppress crosstalk as small as possible. There are two types of interaction between the transmitter and the receiver due to the electromagnetic coupling between the electrical circuits, electrical crosstalk and this optical crosstalk. Both are difficult issues to overcome. Here, only optical crosstalk is considered.

Various types of transmitters and receivers for performing transmission and reception using one optical fiber have been devised according to a method of separating transmission light and reception light. The most common one uses a wavelength demultiplexer to split a transmission optical path and a reception optical path. Such spatial separations have relatively minor optical crosstalk problems. As a more specific transmission / reception module, there is a module in which PDs and LDs are arranged on one straight line and transmission / reception paths are almost the same. If such transmission and reception paths are the same, the problem of optical crosstalk becomes more serious.

[0004]

2. Description of the Related Art A typical example of a transmission / reception module is shown in FIG.
(Wavelength multiplex bidirectional communication). On the office side, a signal is generated by the LD 1 and transmitted to the subscriber side via the optical fiber 1, the splitter 2, the optical fiber 3, the splitter 4, and the optical fiber 5. The subscriber generates a signal by the LD 2 and the optical fiber 6,
The signal is sent to the station-side PD 1 via the splitter 4, the optical fiber 3, the splitter 2, and the optical fiber 7. A single optical fiber can transmit signals in both directions simultaneously. The wavelength demultiplexer 2 separates the upstream and downstream signals into two different optical fibers 1 depending on the difference in wavelength.
7 is separated. The wavelength of the downstream signal is λ ₂ , and the wavelength of the upstream signal is λ ₁ . Each of them is sent vertically by one optical fiber 3. On the subscriber side, the transmission light and the reception light are separated by the wavelength demultiplexer 4 into different optical fibers 5 and 6. Receiving light lambda ₂ at the subscriber side receives the PD2.
Transmitting light lambda ₁ generates by LD2. PD2, LD
Although there is an electric circuit ahead of 2, the illustration is omitted here.
The term "transmission / reception" is opposite on the station side and the subscriber side, but in the present invention, transmission / reception is from the viewpoint of the subscriber. Upstream optical lambda ₁ is transmitted light, the downstream optical lambda ₂ is the received light. In this example, since the optical path is branched by the wavelength demultiplexer 4, the PD 2 and the LD 2 are spatially separated.

FIG. 2 shows a case where two wavelengths of light λ ₁ and λ ₂ are transmitted in one direction from a station side to a subscriber side (wavelength multiplexing one-way communication). On the station side, the transmission light λ of different wavelength
₁ and λ ₂ are united. The two signals are sent on one optical fiber. On the subscriber side, the wavelength is demultiplexed by the wavelength demultiplexer 4. Lights of λ ₁ and λ ₂ are received by different PDs ₁ and ₂ . Also in this case, optical crosstalk is a problem in each of PD1 and PD2.

FIG. 3 shows a PD module according to a conventional example which can be used as a receiver in an optical communication system in which the optical paths are separated as shown in FIGS. A PD chip 12 is attached to the center of a circular metal stem 10 having lead pins 9 via a submount 11. A cylindrical cap 14 having a lens 13 is aligned and welded onto the stem 10. Furthermore, a cylindrical sleeve 15 is provided thereon. A ferrule 16 is inserted into a shaft hole of the sleeve 15. The ferrule 16 holds the tip of the optical fiber 17. The tip of the ferrule 16 is polished obliquely. A bend limiter 18 is placed over the sleeve to protect the optical fiber 17. 1 and 2 further includes a transmitter. It is not described because it is obtained by replacing the PD in FIG. 3 with an LD.

The present invention can also be applied to a system including a wavelength demultiplexer as shown in FIGS. 1 and 2, and FIGS. 1 and 2 have been described because they include a receiver as shown in FIG. The receiver uses a metal case and the optical fiber arrangement is three-dimensional. Although high performance, alignment is indispensable, the production cost is high, and it is difficult in terms of practical use.

[0008] A surface-mount type light receiving element module has been studied as a cheaper light receiving element module. FIG. 4 shows a back-illuminated P
1 shows a conventional example of a surface mount type module using D. Since the present invention can be applied to a surface mount type light receiving element, it will be described in advance. A vertical V-groove 2 in the center of a rectangular Si bench 19
0 is formed by etching. At the end of the V-groove 20, there is an inclined mirror surface 21. This is also produced simultaneously by etching. A PD chip 23 is fixed just above the end of the V-groove 20. This is a back-illuminated type and has a light receiving unit 24 on the upper surface. The light emitted from the optical fiber 22 is S
The light propagates in the V groove parallel to the i bench surface, is reflected upward by the mirror surface 21, enters the PD 23 from the bottom surface, and reaches the light receiving unit 24. The surface mount type module is easy to manufacture because there is no centering portion.

3 and 4 can be used to detect the received light separated by the wavelength demultiplexer shown in FIGS. 1 and 2. The wavelength demultiplexer can be manufactured, for example, by forming a wavelength-selective branch waveguide on a Si substrate. However, there are various wavelength demultiplexers, and a prism type wavelength demultiplexer as shown in FIG. 5 can be used.
In FIG. 5, the transparent triangular prism glass blocks 25 and 26 have a dielectric multilayer film 27 laminated on the hypotenuse surface, which has wavelength selectivity. Selectivity is such that light of different wavelengths emitted from the optical fiber 28 is reflected on one side and transmitted on the other side. However, here, selectivity for combining transmission light and reception light is provided. Because of the reciprocity of light, only the same selectivity is used for different applications. The reception light λ ₂ output from the optical fiber 28 is reflected by the multilayer film 27 and guided to the PD 30. The transmission light λ ₁ emitted from the LD 29 passes through the multilayer film 27 and enters the fiber 28. The present invention has been preliminarily described because it can be used for such a transmitting / receiving module.

However, the present invention is most suitably applied to a transmission / reception module in which the path is not separated by a wavelength demultiplexer. In order to distinguish it from what has been described above, let us tentatively call it a path non-separable type. It is a module originally proposed by the present inventors for the first time, and can be said to be still a special module. The non-separable path type has the advantage that the structure is simple because the PD and LD are on the extension of the optical fiber. On the other hand, since the optical axes are the same, crosstalk becomes serious. FIG. 6 shows a non-separable path type. As a use of the present invention, the non-separable type shown in FIG. 6 is a favorite.

There is a Si bench inside the housing 31, but is not shown here. An optical fiber 32 is provided in the vertical direction. LD to face the end of optical fiber
33 is attached. WDM in the middle of optical fiber 32
There is a filter 35 for wavelength separation.
The PD 34 is located immediately above the WDM 35. LD33 output λ
_One transmission light is, for example, as strong as 1 mW. This goes out through the optical fiber 32 to the outside. The received light λ ₂ transmitted from the outside through the optical fiber is reflected by the WDM 35, enters the back surface of the PD 34, and is detected by the light receiving unit 36. The transmission light λ ₁ is strong light, and the reception light λ ₂ is weak light. The route to the WDM in the optical fiber 32 is the same. Although the directions are opposite and separated by WDM, some of the transmitted light may enter the PD. This causes optical crosstalk. Even a small percentage,
Since the original transmission light power is large and the reception light is weak, it becomes a large noise that cannot be ignored for the reception light.

When the PD according to the conventional example as shown in FIG. 7 is used, it is not simply noise. The transmission light noise exceeds the reception light itself and the S / N ratio (the larger the signal / noise ratio is better) becomes a value much smaller than 1. The manufacturing of the InP-based light receiving element of FIG.
On a P substrate 37, an n-InP buffer layer 38, n-I
Starting from an epitaxial wafer on which the nGaAs light receiving layer 39 and the n-InP window layer 40 are stacked. A p-type region 41 is formed by Zn diffusion, and an annular p-electrode 42 is formed thereon. An anti-reflection film 43 is placed on the inside of the p-electrode 42 and a passivation film 44 is placed on the outside. N electrode 45 on the back
To form In the case of this figure, the incident light is incident from the upper surface (upper surface incident type). In the case of a back-illuminated type, the n-electrode is formed in a ring shape, and the p-electrode is an electrode having no opening and having a uniform thickness.

FIG. 8 shows the sensitivity characteristics of the light receiving element of FIG.
The portion of P where the sensitivity decreases on the short wavelength side corresponds to the band gap of the InP window layer. Shorter wavelength light is less than InP
Not absorbed because it is absorbed by the window layer. The portion of R where the sensitivity decreases on the long wavelength side corresponds to the band gap of the InGaAs light receiving layer. Light having a longer wavelength than that, that is, light having lower energy (here, the energy hν of photons) cannot be perceived because the energy is less than the band gap of the light-receiving layer. This light receiving element has sensitivity in a wide range Q from the band gap P of the InP window layer to the band gap R of the InGaAs light receiving layer. It has sufficient sensitivity in both the 1.3 μm band and the 1.55 μm band.

Therefore, the light receiving element shown in FIG. 7 is most generally used for long wavelength light used in optical communication. In the PD having the conventional structure shown in FIG. 7, 1 μm to 1.65 μm as shown in FIG.
There is sensitivity up to. Having such a wide range of sensitivity is an advantage in that the same light receiving element can be used for both 1.3 μm and 1.55 μm. However, on the other hand, there is a drawback that when used in a transmission / reception module, transmission light is sensed and optical crosstalk is caused.

If the transmitted light λ ₁ has lower energy than the received light λ ₂ (λ ₁ > λ ₂ ), the crosstalk can be reduced by devising the light receiving layer of the PD. The PD absorbs light having an energy higher than the band gap of the light receiving layer and converts the light into a photocurrent. However, the PD transmits light having an energy lower than the band gap and cannot sense it. If a light receiving layer having an energy band gap between transmission light energy and reception light energy is selected, a PD that can selectively receive only reception light can be obtained.

On the contrary, when the transmitted light λ ₁ has higher energy than the received light λ ₂ (λ ₁ <λ ₂ ), such means cannot be used. By devising the band gap of the light receiving layer, crosstalk cannot be suppressed. The present invention is directed to such cases. That is, in the present invention, the transmission light wavelength is shorter than the reception light wavelength (λ ₁ <λ ₂ ). For example, λ ₁ = 1.3 μm transmission, λ ₂ =
This is the case with 1.55 μm reception. 1 and 2
WD even when using a wavelength demultiplexer having a Y-branch such as
If the performance of M is insufficient, the crosstalk will increase, and in the case of the non-separable path type as shown in FIG. 6, the crosstalk will be even more remarkable.

The reason why optical crosstalk occurs in the transmitting / receiving module shown in FIG. 6 will be described. In the device of FIG. 6, actually, the strong transmission light (λ ₁ ) from the LD arranged on the Si platform (Si bench) does not entirely enter the optical fiber, but a part of the transmission light does not reach the optical fiber.
The i-platform hits the resin and is scattered. This is because the light emitted from the LD spreads at a considerable angle. Si bench with respect to the transmission light lambda ₁ is transparent. This is the origin of the problem in the first place. LD and the like are covered with a transparent resin. S
The transmission light λ ₁ entering the i-bench or the space of the transparent resin is transmitted through Si, reflected and scattered. Various scattered lights are generated depending on the distribution of the resin, the shape of the Si platform, and the element arrangement. The trajectory of the reflected scattered light of λ ₁ is complicated. The whole of the Si platform when viewed from the PD is as if shining brightly at λ _1. Thus, λ ₁ from other than the path is called scattered light or stray light.

Although it is not visible light, it is not perceived by human eyes but is visible by PD. Λ ₁ from various directions from various directions
Is incident on the PD. P from back, top and side
Enter D. Therefore, transmission light that does not pass through the optical fiber enters the PD and causes crosstalk. Crosstalk caused by the λ ₁ scattered light which does not pass through such a WDM can not be suppressed no matter how pulling the WDM performance. The only WDM as the output of the LD increases have become not completely suppress the leakage of λ ₁ light passing through the optical fiber. Light that is reflected and refracted from the LD by the WDM and reaches the PD is called leakage light. For λ _1, it is easily explained better to distinguish between the leakage light and the scattered light (stray light).

Therefore, the present inventor has proposed an absorption layer having a wavelength selectivity that absorbs only the transmission light λ ₁ and allows the reception light λ ₂ to pass through the PD.
Invented a PD with a clever structure that is provided inside a chip. JP-A-11-83619 (Applicant: Sumitomo Electric Industries, Ltd., Inventors: Miki Kohara, Yumi Nakanishi, Hitoshi Terauchi, 199
(Filed on September 3, 1995). FIG. 9 shows a basic structure of a PD having such an absorption layer. An n-type InGaAs light receiving layer 47 is provided on an n-type InP substrate 46. n-type InGa
In the center of the As light receiving layer 47, the p-type region 4 is formed by Zn diffusion.
8 is formed. An i layer 49 (depletion layer) is formed in contact with the pn junction. A p-type electrode 50 is provided on the p-type region 48. p
A passivation film 51 (eg, SiN) is provided outside the electrodes so as to cover the pn junction. n-type InP substrate 46
The n-type InGaAsP absorption layer 52 is provided on the back surface of the substrate. Further below this, a ring-shaped n-electrode 53 is formed. The central portion of the n-electrode 53 becomes the entrance window. The entrance window is an anti-reflection film 5
4 coated.

FIG. 10 shows a more specific structure of a PD.
Actually, in order to improve the crystallinity, the n-type InP substrate 46 is formed.
An n-type InP buffer layer 56 is provided between the substrate and the InGaAs light receiving layer 47. Further, the IGaAs
An nP window layer 55 is provided. From the window layer 55 to the light receiving layer 47, a p region 48 is formed in the center by Zn diffusion.
The p electrode 50 is formed on the p region 48. As described above, since the InP substrate 46 and the InP buffer layer 56 are provided, if the InGaAsP absorption layer 52 is not provided, the sensitivity curve has a wide sensitivity (Q) as shown in FIG. .

This is a back-illuminated type and is used for a module as shown in FIG. The InGaAsP absorption layer 52 is a novel point of this PD. Since it is a quaternary compound, the band gap and lattice constant can be freely selected. About semiconductor,
The insulator can absorb light having an energy larger than the band gap and cannot absorb light having an energy smaller than the band gap. Therefore, if the band gap of the absorption layer 52 is selected to be intermediate between the energy of the transmission light λ _{1 and the} energy of the reception light λ ₂ , it should be possible to selectively absorb only λ ₁ .

For example, if the transmission light is 1.3 μm and the reception light is 1.55 μm, the band gap wavelength of the absorption layer is preferably about 1.4 μm. This absorption layer is 1 × 10
¹⁸ cm ^- the ^third carrier concentration (electron concentration), the thickness is set to 5μm, for example. It is a rather thick absorbent layer. Also, the electron concentration is made considerably high. 1.3 μm thick absorption layer
This is for sufficiently absorbing light. 1.3 μm of this absorption layer
The attenuation constant α for m light is α = 10 ⁴ cm ⁻¹ .
When the film thickness is 5 μm, the attenuation is exp (−αd) = 0.007.
And the transmittance becomes sufficiently small at 0.7%. One reason for the high electron concentration is that the forward resistance of the PD does not become too high. The other is that the absorption layer absorbs 1.3 μm light to form electron-hole pairs, but this also means that holes are easily recombined. Since it is n-type, there are many electrons from the beginning and there is no problem. Since the absorption layer absorbs the transmission light (1.3 μm) and selectively transmits only the reception light (1.55 μm), it is extremely effective in preventing optical crosstalk between the transmission and reception light.

FIG. 11 shows the wavelength dependence of the transmittance of the InGaAsP absorption layer 52. Bandgap wavelength λ _g =
The mixed crystal ratio of InGaAsP is selected so as to be 1.42 μm. Therefore, wavelengths shorter than 1.42 μm are absorbed and wavelengths longer than 1.42 μm are transmitted. It states that the sensitivity of the PD was 1 A / W for 1.55 μm light and 0.01 A / W or less for 1.3 μm light. Extinction ratio is 1
/ 100 or less (20 dB).

Since this is a back-illuminated PD, λ₁
And λ₂Is incident, the absorption layer 52 becomes λ ₂Through as it is,
λ₁Absorb. In other words, the transmitted light is absorbed here.
Do not go to a light-receiving layer. So the LD transmitting light enters the PD
No optical crosstalk occurs. This is shown in FIG.
This is extremely effective in a path separation type optical system. Diagram of path separation
1, Even in the case of Fig. 2, can not be separated by WDM filter
1.3 μm light can be dropped by the absorption layer
You.

This is a very clever and excellent invention. n-type In
An n-InGaAs light receiving layer 47 is provided on the upper side of the P substrate 46, and an InGaAsP absorption layer 52 is provided on the lower side. Both are formed by epitaxial growth. Since the absorption layer and the light receiving layer are on the opposite sides of the substrate, there is an advantage that electron-hole pairs generated by absorption disappear without any influence on the light receiving layer. Therefore, when considering the thickness, only the photoelectric conversion exp (−αd) of 1.3 μm light needs to be considered. It just needs to be small enough. Since it is possible to remove the disturbing 1.3 μm light, it is considered that the crosstalk between the transmitted and received light that has not been removed by the WDM filter can be completely removed.

[0026]

The invention of the present invention is very effective for 1.3 μm leak light incident straight from the back surface. This is because it always passes through the absorption layer. However, the light from the LD in the transmission / reception module is not limited to such a simple one (leakage light). An optical fiber end coming out of the LD,
There is also a resin (stray light) that performs complicated scattering and irregular reflection with a resin or the like. Since the LD has a short resonator length, the emission angle is wide in the first place.
Since no lens is used, the emitted light spreads up, down, left, and right. LD
Considerable power of light does not enter the optical fiber. As shown in FIG. 6, in a non-separable type in which a PD and an LD are placed on a fiber extension using a Si bench, light from the LD illuminates the entire Si bench brightly. This was unexpected. λ
_{Since the} Si bench is transparent, the scattered light of λ ₁ is generated from the Si bench and the transparent resin. In other words, in addition to removing the leaked light caused by the imperfections of the WDM by the absorbing layer, scattered light (stray light) which exits the LD and repeats the scattered reflection
Need to be removed as well.

[Lambda] _1, which does not enter the fiber and becomes transmission light after exiting the LD, is called stray light. It can be said that the stray light fills the housing of FIG. Since it is a back-illuminated type, λ ₁ does not always come from the back. The InP substrate has a thickness of as much as 200 μm. So there is stray light coming from the side of the board. Some stray light enters from the top. Even if the InGaAsP absorption layer 52 is below the substrate, it cannot absorb stray light (λ ₁ ) entering from the top surface or the side surface of the PD. Λ entered from the side
The stray light ₁ reaches the depletion layer 49 and the p region 48 in FIG. 9 and generates a photocurrent. That is, the absorption layer 52 on the bottom surface is ineffective at all against stray light from the side surface and the top surface.

It is also possible to widen the p-electrode which is a blind plate against stray light from above, and to cover portions other than the p-electrode with an opaque material. However, since the PD is manufactured by a wafer process, it is difficult to cut the wafer into a large number of chips, form the PD chips, and then process the side surfaces. Chip side thickness is 200μ
m, but this remains exposed. It is difficult to cover the sides to prevent stray light from entering the sides.

The wavelength separation ratio in a WDM filter or a WDM coupler is 15 dB to 20 dB. On the other hand, the entire receiver needs a wavelength separation ratio of at least 30 dB, preferably 40 dB. For example, if the LD is 1 mW (= 0 d
Bm), the minimum reception limit of the reception light exiting from the optical fiber is −
It is assumed that the power ratio is 30 dBm, and the power ratio of the transmission light incident on the PD to the reception light must be −10 dB.
As a whole, the wavelength separation requires 40 dB (30 + 10). Assuming that the ratio of the transmitted light and the received light is -15 dB (about 1/30) and the minimum receiving sensitivity is -35 dBm, a wavelength separation ratio of 50 dB is required.

In a device for transmitting and receiving a long-wavelength received light λ ₂ and a short-wavelength transmitted light λ ₁ , a structure of a PD in which neither leakage light of transmission light due to imperfections of WDM nor stray light of transmission light enters. It is an object of the present invention to provide.

[0031]

According to the present invention, one side of a substrate (hereinafter referred to as a back surface) is provided with a .lambda.
Provided with absorbing layer InGaAsP system (first absorber layer) for _1, just below the light receiving layer above the substrate to remove the scattering lambda _1, I selectively absorbing transmission light lambda ₁
An nGaAsP-based absorption layer (second absorption layer) is provided. Not than providing lambda ₁ absorbing layer only on the back surface of the substrate as.
The present invention is provided an absorbent layer of lambda ₁ on both the back surface and the surface of the substrate. The provision of lambda ₁ absorption layer on both sides of the substrate is a feature of the present invention. Since there are two lambda ₁ absorbing layer, the absorbing layer of the back side first absorption layer, the absorption layer on the surface side of the second absorbent layer. The surface is the side on which the light receiving layer is provided. Not in the direction of the incident surface. The first absorbing layer on the back side compensates for imperfections of WDM and removes WDM leakage light. The second absorbing layer on the front side is W
Of course, it has the function of removing the DM leakage light, but also has other functions. The second absorption layer absorbs all scattered light (stray light) of λ ₁ (1.3 μm light) not only from directly below, but also from the side and obliquely below. In other words, the roles can be written as follows: the first absorbing layer (back side) = removal of leaked light. The second absorbing layer (front side) = removal of stray light + removal of leaked light.

The transmission light λ ₁ is absorbed and annihilated by the first and second absorption layers and does not reach the light receiving layer, so that optical crosstalk is significantly reduced. Bandgap wavelength lambda _g is lambda ₁ and lambda ₂ of the intermediate values of InGaAsP quaternary mixed crystal (λ ₁ <λ _g <
The mixed crystal ratio of the absorption layer is determined so as to be λ ₂ ). The thickness d is 5
The thickness is preferably about μm, and is 3 μm to 10 μm. If it is too thin, 1.3 μm light cannot be completely absorbed. If the thickness is too large, the epitaxial growth time is increased and the material cost is increased. In addition, since the absorption layer is a quaternary system, if the thickness is 5 μm or more, the crystallinity decreases. From the viewpoint of crystallinity, it is preferable that both the first and second absorption layers do not have a thickness of about 5 μm or more. The thickness will be described later in detail.

The carrier (electron) concentration of the absorption layer is considerably high.
Is necessary. 10¹⁸cm ^-3Desirable degree
No. If the carrier concentration of the absorption layer is low, the forward resistance of PD
Becomes too large and reconstitutes holes made of 1.3μm light
It takes time to match. Occurs in the first absorption layer on the back side
Holes passed through a thick n-type substrate (thickness of 200 μm or more)
It hardly escapes. Generated in the second absorbing layer on the surface side
Since the holes do not pass through the substrate, they may travel to the light receiving layer.
Therefore, the hole disappears inside the second absorption layer.
desired. This will also be described later.

Further, the peripheral portion p is formed so as to surround the central p region.
When an area (referred to as a diffusion shielding layer) is provided, the carrier due to light from the side disappears here.

It is more preferable to form an InP window layer on the InGaAs light receiving layer. If an InP window layer is provided, pn
The surface portion of the junction can be stabilized by the passivation film, the dark current can be reduced, and long-term reliability can be ensured.

It is also effective to provide a lightly doped InP buffer layer between the absorption layer and the light receiving layer in order to improve the crystallinity. If the impurity concentration is high, the lattice structure may be disturbed and crystallinity may be deteriorated. Since the absorption layer has a high dopant concentration, the buffer layer restores the crystallinity.

The thickness d of the absorbing layer will be described. I've mentioned it a few times, but here's a more detailed one. The absorption coefficient of the absorption layer is α, the thickness of the absorption layer is d, and the transmittance is T. If there is no reflection on the front and back,

T = exp (−αd) (1)

Is as follows. α depends on the crystal composition,
For example, in one example of the present invention, α = 1 × 10 ⁴ cm ⁻¹ when the composition has an absorption edge of 1.42 μm. FIG. 16 shows this relationship. Since 1.55 μm light is hardly absorbed by the absorption layer, T for 1.3 μm light can be regarded as a number indicating the effect as a filter (through 1.55 μm light and excluding 1.3 μm light).

Filter effect = −10 log T = 4.343αd (2) (assuming α ⁻¹ = 1 μm)

FIG. 16 shows that the filter effect of 10 dB (10
%), It is necessary that d = 2.3 μm or more. To obtain a filter effect (1%) of 20 dB, a thickness of d = 4.6 μm is required. Taking into account the variation in the thickness of the epitaxial growth layer, to secure a filter effect (1%) of 20 dB, d = 4 to 6
μm, that is, a thickness of about 5 μm is preferable.

From the viewpoint of the filter effect, it is better to increase the thickness of the absorbing layer. However, if the absorption layer is too thick, the crystallinity decreases. Since the absorption layer has a high dopant concentration and is a quaternary mixed crystal, a thick layer deteriorates crystallinity. From the viewpoint of crystallinity, about 5 μm is the best. However, even if the crystallinity is slightly deteriorated, the crystallinity is recovered to some extent when an InP buffer layer is inserted thereon. Therefore, a thickness of about 10 μm is acceptable.

[0043] First, the leakage light lambda ₁ by WDM imperfections can be removed almost completely by the first absorption layer of the back surface (InGaAsP). Since the holes generated here cannot pass through the substrate and disappear, they do not become photocurrents and cause no problem. The second absorbing layer placed near the light receiving layer mainly to remove scattered light (stray light) still poses a problem. The second absorber layer has the problem of early carrier recombination. Unnecessary λ ₁ (1.3 μm light) is absorbed by the absorption layer to form an electron-hole pair.
μm light is included in the signal as noise. It must be recombined early without this becoming a photocurrent. This means that the absorption layer also needs a thickness for recombination. Since the absorption layer is made of n-type InGaAsP, when electron-hole pairs are formed, the electrons are majority carriers and many conduction electrons already exist, so there is no particular problem.

Holes are minority carriers in the n-type absorption layer. Since the n-type absorption layer has a high carrier concentration, no electric field works (electric field = 0). The holes do not move toward the light receiving layer due to the electrostatic field. However, there is a hole concentration gradient toward the p region. Holes diffuse toward the p region due to the concentration gradient. During diffusion, they interact with a large number of electrons to recombine and disappear. To a length which diffusion before recombination disappearance of the diffusion length L _d. In practice the hole lifetime tau, diffusion length L _dh is defined as the square root of time (d.tau) the diffusion coefficient was set to D. L _dh = (Dτ) ^1/2 . Diffusion is not a one-way movement but a whole movement that goes back and forth, so D
It is itself defined by the square of the displacement x divided by the time t, the limit of x ² / t.

Generally, electron diffusion in InGaAsP crystals
Length L_deIs long and the hole diffusion length L_dhIs short. If p-type
Λ using the absorption layer₁If electrons are absorbed,
Since it acts as a rear and has a long electron diffusion length, photocurrent is easily generated. This
Since this is a problem, minority carriers are made holes using the n-type absorption layer.
To The diffusion length depends on the crystal composition, but that alone
Absent. Is the collision less as the crystal purity is higher even with the same composition?
The diffusion length becomes longer. High purity (n = 10^Fifteencm^-3)
Electron diffusion length L in InGaAsP crystals_de= 6.0μ
m, hole diffusion length L_dh= 1.6 μm. Low purity
The crystals have a shorter diffusion length. More carriers increase
This is because the recombination cross section increases (τ decreases). one
In general, the hole diffusion length L_dhIs the carrier concentration n of the n-type region.
Decreases in proportion to the square root. Significantly high concentration of InGa
AsP absorption layer (n = 10¹⁸cm ^-3), So high
About 1/30 lower than pure InGaAsP
Down. That is, hole diffusion in the n-InGaAsP absorption layer
Length is L_dh= 0.05 μm. This is 0.05μm
Means that the hole concentration decreases to 1 / e.

Even if it says absorption, light does not disappear immediately. That is, light is converted into electron-hole pairs. Since α is large, light of 1.3 μm almost becomes an electron-hole pair in the first part of the absorption layer. Holes begin to diffuse from the first part, but if the thickness of the absorbing layer is as large as d = 5 μm, exp (−5 /
0.05) ≒ 10 ^-44 . The reason why the carrier concentration n of the absorption layer is considerably increased is that the hole diffusion length L _dh
It is also to shorten. Therefore, the lower limit of the carrier concentration of the absorbing layer also depends on the thickness d of the absorbing layer. If d is large, the hole diffusion length may be somewhat longer, so the concentration n may be slightly smaller.

Next, one having only an absorption layer on the back side ()
And the difference between the present invention having absorption layers on both sides.
In short, it means that the solid angle Ω at which the absorption layer is desired from the light receiving layer of the PD is large. The width of the PD chip is W (300 μm to 500 μm). The distance from the light receiving layer to the absorption layer is g. The solid angle Ω facing the absorption layer from one point of the light-receiving layer is within the approximate range

Ω = 2π [1-g / ｛(W / 2) ² + g ² ｝ ^1/2 ] (3)

Is as follows. In the case of the present invention, the distance g between the absorption layer and the light receiving layer is as short as about 2 μm to 10 μm. But W / 2
Is as large as about 200 μm. Therefore, in the present invention, approximately Ω ~
2π. However, in this case, since the thickness of the substrate is included in g (200 μm to 300 μm), the thickness is about the same as W / 2, so that it is about Ω to π. The solid angle facing the absorption layer from one point of the light receiving layer is about twice as large in the present invention. This is a difference in the effect on stray light (scattered light). Further, the present invention is useful for WDM leak light. The WDM leak light enters from the back surface, but cannot reach the light receiving layer unless it passes through the first absorption layer and the second absorption layer twice. Most of the leak light disappears in the first absorption layer. However, there is some leakage light passing through it. The leaked light is completely extinguished by the second absorption layer. As described above, the present invention has an absorption effect on both scattered light (stray light) and leaked light, and has an excellent effect on optical crosstalk removal.

[0050]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 (Back-illuminated type; basic type) A PD in which the present invention is applied to a back-illuminated PD will be described with reference to FIG. After forming an InGaAsP layer 67 as a first absorption layer on the back surface of the n ⁺ type InP substrate 57,
A second n ⁺ -type InGaAsP light absorbing layer 58 and an n-type InGaAs light receiving layer 59 are epitaxially grown on the front surface side of the substrate. Thus, the simple feature of the present invention is to use the absorption layer-substrate-absorption layer-light-receiving layer. The n ⁺ -InP substrate 57 has a thickness of, for example, 200 μm and a carrier (electron) concentration of n = 3 × 10 ¹⁸ cm ⁻³ . First and second n ⁺ −
Each of the InGaAsP absorption layers has a thickness of, for example, 5 μm and a carrier concentration of n = 1 × 10 ¹⁸ cm ⁻³ . n-In
The GaAs light receiving layer 59 has, for example, a thickness of 3 to 4 μm and a carrier concentration of n = 1 × 10 ¹⁵ cm ⁻³ .

The center of the n-type InGaAs light receiving layer 59 has a Z
A p ⁺ region 60 is formed by n diffusion. The carrier concentration p is not constant because it is manufactured by diffusion, but n-type I
The pn junction 61 is where the carrier concentration becomes equal to that of the nGaAs light receiving layer 59 (p = n). Since the carrier concentration n is low, the light-receiving layer 59 has a thick depletion layer (i-layer) 62 formed of p.
It occurs continuously below the n-junction. A wide p electrode 63 is formed on p region 60. Since it is a back-illuminated type, the p-electrode may cover almost the p-region. A passivation film 64 is formed on the periphery. The passivation film is, for example, a SiN film and protects the termination of the pn junction 61. On the back surface of the n-InP substrate 57, an annular n-electrode 65 is formed. Light enters from the back side, so n-electrode 65
Becomes ring-shaped. The anti-reflection film 66 is made of the n-InP substrate 5
7 is coated on the back center.

Incident light (received light) A of λ ₂ entering from the back side
Passes through the absorption layer 67, the substrate 57, and the absorption layer 58,
To the depletion layer 62. The reception light A is absorbed by the depletion layer 62 to form an electron-hole pair. The depletion layer 62 is a region where light is sensed. Due to the reverse bias, electrons travel to the n-type region and holes travel to the p-type region to flow a photocurrent. This is a normal PD photoelectric conversion operation.

Since the WDM is incomplete, the transmitted light λ ₁
(Leakage light) B is incident. Λ entered from the back
Most of the leaked light B of ₁ (transmission light wavelength) is the first n−
It is absorbed by the InGaAsP absorption layer 67. Even if the leaked light is intense and a part of the light passes through the first absorption layer, it is absorbed by the second absorption layer 58 above. Leakage light B of back incidence λ ₁
Does not reach the light receiving layer 59. That is not all. Obliquely from below lambda ₁ of the incident light (scattered light, stray light) C is absorbed by the second absorption layer 58 as entering the PD. The obliquely incident light C does not reach the light receiving layer 59 either. PD is oblique incident light C
Is not detected. The PD of the present invention is insensitive to the obliquely incident light C. That two lambda ₁ light B (light leakage), are both insensitive to C (scattered light). For the leaked light (light traveling straight from the back surface), the first and second absorbing layers work in duplicate to completely eliminate the leaked light. Since the second absorption layer is near the light receiving layer, oblique incident light (scattered light) C can also be removed. So this PD does not have the sensitivity to the transmission light wavelength λ _1.

FIG. 12 shows a basic form, and the structure operates as it is. However, we can add some more ideas. For example, on the n-InGaAs light receiving layer 59, I
More preferably, an nP window layer is inserted. In addition, the InP substrate 57
Inserting a buffer layer between the InGaAsP absorption layer 58 and the InGaAsP absorption layer 58 is effective for improving crystallinity. These measures are also described below.

Embodiment 2 (Back-illuminated Type; Diffusion Shielding Layer) FIG. 13 shows another example of application of the present invention to a back-illuminated type. A diffusion shield layer is provided around the p-electrode. On the back surface of the n ⁺ -type InP substrate 57, n ⁺ -type InGa
An AsP absorption layer 67 (first absorption layer) is epitaxially grown. On the substrate surface side, an n ⁺ -type InGaAsP light absorbing layer (second absorbing layer) 58, an n-type InGaAs light receiving layer 59,
An n-InP window layer 69 is epitaxially grown. The feature of the present invention is to form an absorption layer-substrate-absorption layer-light-receiving layer. Here, an InP window layer 69 is further added. The use of the InP window layer 69 having a wide band gap has the effect of stabilizing the passivation of the pn junction and reducing the dark current. Further, the light from the outside is given the absorption edge P in FIG. That is, the InP window layer 69 has three excellent effects.

The n ⁺ -InP substrate has, for example, a thickness of 200 μm and a carrier concentration of n = 3 × 10 ¹⁸ cm ⁻³ . (First and second) n ⁺ -InGaAsP absorption layers 6 on both sides of the substrate
7, 58 have a thickness of, for example, 5 μm and a carrier concentration of n = 10.
¹⁸ cm ^-3 . The total thickness of the absorbing layer is 10 μm. The thickness of the n-InGaAs light receiving layer 59 is, for example, 3 to
At 4 μm, the carrier concentration is n = 10 ¹⁵ cm ⁻³ . The n-type InP window layer 69 has, for example, a thickness of 2 μm and a carrier concentration of n = 2 × 10 ¹⁵ cm ⁻³ . n-InP
The center of the window layer 69 and the n-type InGaAs light receiving layer 59 is Zn
A p ⁺ region 60 is formed by diffusion. A pn junction 61 is generated. A thick depletion layer (i-layer) 62 is continuously generated below the pn junction 61. Wide p-electrode 6 on p-region 60
3 is formed. In addition, Zn diffusion is simultaneously performed on the peripheral portion of the upper surface to form the p ⁺ region 70. The present inventor calls this p ⁺ region 70 a “diffusion shielding layer”. Rather than preventing the diffusion of electrons, it means that the p-region is created by Zn diffusion and the influence of peripheral incident light is shielded. Although not necessarily a proper term, the inventors have long used such words to describe the surrounding p-region. A pn junction 71 is generated immediately below the peripheral p region 70, and a depletion layer 72 is generated therebelow. The portions where the pn junctions 61 and 71 are exposed on the surface are protected by the SiN passivation film 64. On the back surface of the n-InP substrate 57, a ring-shaped n
An electrode 65 and an antireflection film 66 are formed.

In the back-illuminated type, the λ ₂ incident light A from immediately below reaches the depletion layer 62 in the center to form an electron-hole pair, which becomes a photocurrent. Transmission light (lambda ₁₎ B incident from immediately below is absorbed by the first absorbent layer 67 of the substrate backside, lambda ₁ light C from diagonally below is absorbed by the ^{n +} -type InGaAsP light absorbing layer 58. Lambda ₁ incident light D which came from the side in addition to also not sensitive. The reason that D is not perceived is not due to the absorption layer so far. The side incident light D is transmitted through the diffusion shielding layer 70.
A pair of electrons and holes is generated in the depletion layer 72, but the holes enter the peripheral p region 70 and disappear, so that the current does not become a current of the central p electrode. The influence of the side incident light is canceled by the diffusion shielding layer 70.

[Embodiment 3 (Back-illuminated type; diffusion shielding layer +
FIG. 14 shows a back-illuminated type and a diffusion shielding layer.
In addition, a reflective film is further provided to transmit λ₁(Stray
Shows what is trying to prevent light E) from entering. n⁺Type InP
The first n⁺Type InGaAsP light absorption
The layer 67 is epitaxially grown, and further on the substrate surface side.
The second n⁺Type InGaAsP light absorbing layer 58, n-type In
The GaAs light receiving layer 59 and the n-InP window layer 69 are formed by epitaxy.
To grow. n-InP window layer 69, n-type InGaAs
In the center of the s light-receiving layer 59, p ⁺Area 60
To form A pn junction 61 is generated. Thick depletion layer
(I-layer) 62 occurs below the pn junction 61. p region 6
A wide p-electrode 63 is formed on 0.

At the same time, Zn diffusion is also performed on
^{A +} region (diffusion shielding layer) 70 is formed. Peripheral p region 70
A pn junction 71 is generated directly below the pn junction, and a depletion layer 72 is generated therebelow. The portions where the pn junctions 61 and 71 are exposed on the surface are SiN
It is protected by a passivation film 64. The center of the upper surface of the PD is covered with a metal p-electrode 63, but the outside is entirely covered with a reflective film 73. Looking at PD from above, p
The whole is covered with the electrode 63 and the reflection film 73.
The transmission light of 1.3 μm (λ ₁ ) may have a lower incidence B, an oblique lower incidence C, a side incidence D, an upper incidence E, and the like. B to D can also be prevented in the second embodiment. In addition, in this embodiment, the upper incidence E can be repelled by the reflection film. By doing so, no matter where the stray light (λ ₁ ) comes from, it will not reach the central depletion layer 62 of the PD and will not be perceived.

Embodiment 4 (Back-illuminated type; diffusion shielding layer +
(Reflection film; buffer layer)] When the high concentration InGaAsP absorption layer 58 is provided, the crystallinity may be disturbed. In that case, crystallinity can be improved by inserting a buffer layer having a low impurity concentration. The buffer layer can prevent crystal defects and impurities in the absorption layer 58 from affecting the light receiving layer. FIG.
In the figure, an n-InP buffer layer 74 is provided between the absorption layer 58 and the light receiving layer 59. n ⁺ -type on the back side of the InP substrate 57 is provided an ^{n +} -type InGaAsP light absorbing layer 67, InGaAsP light absorbing layer 58 of ^{n +} -type on the surface side, n-InP
Buffer layer 74, n-type InGaAs light receiving layer 59, n-I
The nP window layer 69 is epitaxially grown. n-InP
The center of the window layer 69 and the n-type InGaAs light receiving layer 59 is Zn
A p ⁺ region 60 is formed by diffusion. A pn junction 61 is formed, under which a depletion layer 62 is generated. At the same time, Zn diffusion is also performed on the periphery of the upper surface to form ap ⁺ region (diffusion shielding layer)
70 is manufactured. A pn junction 71 immediately below the peripheral p region 70
However, a depletion layer 72 is generated thereunder.

A p electrode 63 is attached to the central p region 60.
The pn junction 61 by the SiN passivation film 64,
71 are protected. The center of the PD upper surface is a metal p electrode 6
Covered with 3. The outside is entirely covered with a reflective film 73. The buffer layer 74 has a thickness of, for example, 2 μm to 4 μm, and the carrier concentration is about n = 10 ¹⁵ cm ⁻³ .
This is simply because the buffer layer 74 is inserted between the absorption layer 58 and the light receiving layer 59, and can be applied to the embodiments shown in FIGS. The property of not receiving the incident light B, C, D and E of λ ₁ is the same as in the previous embodiment.

With this configuration, 500 μm square and 200 μm thick
Was prepared, and the sensitivity was measured. High sensitivity of 1.0 A / W to 1.55 μm light incident from the back surface,
It was confirmed that the sensitivity was as low as 0.001 A / W or less for 1.3 μm light. For 1.3 μm light (leakage light) incident on the back surface, two absorption layers work effectively,
An extinction ratio (-30 dB) as large as 0 times can be obtained. In addition, 1.3μm light (stray light) was irradiated from the surroundings,
The sensitivity was as low as 0.01 A / W or less for these lights. Only the second absorption layer contributes to the absorption of stray light from the surroundings, and the sensitivity is higher than in the case of light leaked from the back surface which is doubly absorbed. Next, when this PD was integrated with a 1.3 μm light-LD in the configuration of FIG. 6, the transmission power was 1 mW (0 dB).
m), the receiving sensitivity of 1.55 μm light is −35 d.
High sensitivity of Bm (in combination with amplifier) was realized without crosstalk.

[0063]

According to the present invention, there is provided a transmitting / receiving module which uses a short wavelength λ ₁ for transmission and a long wavelength λ ₂ for reception.
An absorption layer that absorbs ₁ is provided on the back surface of the substrate of the light receiving element (first absorption layer) and between the light reception layer and the substrate (second absorption layer).
Strong λ ₁ is emitted from the LD and is scattered and filled in the module (scattered light and stray light), but does not reach the light receiving layer of the light receiving element because of the second absorption layer. Thereby, the stray light of the transmission light can be eliminated from the light receiving element. Further, since the first absorption layer is provided below (backside) the substrate, it is possible to thoroughly prevent strong transmission (leakage light) light that cannot be removed by WDM. That is, the second absorption layer works effectively for stray light, and the first and second absorption layers work effectively for leaked light. Nevertheless, the structure provided with only the second absorption layer of 10 μm and the structure provided with the first absorption layer of 5 μm and the second absorption layer of 5 μm look the same for leaked light. But it is not. Since the absorption layer has a high carrier density and is a quaternary system, the crystallinity deteriorates when the absorption layer is thickly stacked. Considering the deterioration, the absorption layer of 10 μm is just the thickness that can be tolerated. Providing two 5 μm absorption layers with more crystallinity than that is more effective in blocking leakage light. Furthermore, the roles of the second and first absorption layers are not exactly the same with respect to the leakage light entering from the back surface. The holes formed in the first absorption layer are always eliminated due to the presence of the substrate, and there is no possibility of generating a photocurrent. Therefore, it is useful to absorb strongly in the first absorption layer. Since the second absorption layer is adjacent to the light receiving layer, it is possible that the long-lived holes survive to the light receiving layer. That is, the first absorbing layer is superior in the power for eliminating λ ₁ . Therefore, the combination of the 5 μm first absorption layer + 5 μm second absorption layer of the present invention is superior to the 10 μm second absorption layer. The present invention can thus severely inhibit the transmission light from entering the light receiving layer, and can significantly reduce optical crosstalk. In particular, it is extremely effective in a transmission / reception module of a non-separable optical path type as shown in FIG. Of course, the present invention can also be applied to an optical path separation type transmission / reception module as shown in FIGS.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of wavelength division multiplexing bidirectional communication as one application target candidate of a light receiving element of the present invention.

FIG. 2 is a schematic diagram of wavelength-division multiplex one-way communication as one application target candidate of the light receiving element of the present invention.

FIG. 3 is a vertical perspective view showing an example of a light receiving element according to a conventional example used for optical communication.

FIG. 4 is a surface-mounted light receiving module using a back-illuminated PD.

FIG. 5 is a schematic diagram of a transmission / reception module using a dielectric multilayer film WDM.

FIG. 6 is a schematic configuration diagram of a path non-separable type optical transceiver module in which PDs and LDs are provided on a fiber extension line.

FIG. 7 is a cross-sectional view of an InGaAs-based photodiode according to a conventional example.

FIG. 8 is a sensitivity characteristic diagram of an InGaAs-based photodiode according to a conventional example.

FIG. 9 is a schematic cross-sectional view of a light receiving element proposed by the present inventor to drop 1.3 μm light and receive only 1.55 μm light.

FIG. 10 is a cross-sectional view more specifically illustrating a light-receiving element that has been proposed by the inventor before and which receives 1.3 μm light and receives only 1.55 μm light.

FIG. 11 is a graph showing the wavelength dependence of the light transmittance of an InGaAsP absorption layer epitaxially grown on the back surface of a substrate as a 1.3 μm light absorption layer.

FIG. 12 is a cross-sectional view illustrating a back-illuminated light-receiving element of the present invention.

FIG. 13 shows a back-illuminated type with Zn around a central p region.
FIG. 3 is a cross-sectional view of the light receiving element of the present invention provided with a diffusion shielding layer by diffusion.

FIG. 14 is a cross-sectional view of a light-receiving element according to the present invention which is a back-illuminated type and has a diffusion shielding layer provided around a central p region and a p-electrode covered with a reflective film.

FIG. 15 is a cross-sectional view of a light-receiving element of the present invention which is a back-illuminated type, in which a diffusion shielding layer is provided around a central p region, a p-electrode is covered with a reflective film, and a buffer layer is further provided.

FIG. 16 is a graph showing the relationship between the thickness d of the InGaAsP absorption layer and the transmittance T for 1.3 μm light.

[Explanation of symbols]

1 optical fiber 2 wavelength demultiplexer 3 optical fiber 4 wavelength demultiplexer 5 optical fiber 6 optical fiber 7 optical fiber 8 wavelength multiplexer 9 lead pin 10 stem 11 submount 12 PD chip 13 lens 14 cap 15 sleeve 16 ferrule 17 optical fiber 18 bend limiter 19 Si bench 20 V groove 21 mirror surface 22 optical fiber 23 PD chip 24 light receiving unit 25 glass block 26 glass block 27 dielectric multilayer film 28 optical fiber 29 LD 30 PD 31 housing 32 optical fiber 33 LD 34 PD 35 WDM 36 light receiving unit 37 n-InP substrate 38 n -InP buffer layer 39n-InGaAs light receiving layer 40n-InP window layer 41p type region 42p electrode 43 antireflection film 44 passivation film 45n electrode 46n type InP substrate 47n type InGaAs s light receiving layer 48p type region 49 depletion layer 50p electrode 51 passivation film 52InGaAsP absorption layer 53n electrode 54 antireflection film 55InP window layer 56n type InP buffer layer 57n ⁺ type InP substrate 58InGaAsP light absorption layer 59n type InGaAs light absorption layer 60p ⁺ region 61pn junction 62 depletion layer 63p electrode 64 passivation film 65n electrode 66 antireflection film 67n ⁺ -InGaAsP light absorption layer 69n-InP window layer 70p ⁺ region 71pn junction 72 depletion layer 73 reflection film 74n-InP buffer layer

Continued on the front page (72) Inventor Yasuhiro Inoguchi 1-3-1 Shimaya, Konohana-ku, Osaka-shi, Osaka Sumitomo Electric Industries, Ltd. Osaka Works F-term (reference) 5F049 MA02 MB07 NA04 NA09 NA10 NB01 NB10 QA03 QA06 QA17 QA18 RA07 SE05 SS04 SS07 SZ01 WA01

Claims (12)

[Claims] A first absorption layer including a semiconductor substrate and a light-receiving layer provided on the semiconductor substrate and having a pn junction and having a shorter absorption edge wavelength than the light-receiving layer, the first absorption layer being closer to the light-receiving layer than the semiconductor substrate; A semiconductor light receiving element, wherein at least one is provided, and at least one second absorption layer having a shorter absorption edge wavelength than the light receiving layer is provided closer to the light receiving layer than the semiconductor substrate. 2. The light-receiving layer and the absorption layers on both sides of the substrate are formed by epitaxial growth.
4. The semiconductor light receiving element according to claim 1. 3. The wavelength of light to be detected is λ _2, and light λ _{1 of a} shorter wavelength exists as noise, and an absorption layer for the wavelength λ ₁ to be blocked is provided on both sides of the semiconductor substrate. 3. The semiconductor light receiving device according to claim 1, wherein a light receiving layer is provided on the absorption layer. 4. The wavelength λ ₂ to be received is in the 1.5 μm to 1.6 μm band, and the wavelength λ ₁ to be blocked is 1.2 μm.
The semiconductor light-receiving device according to claim 1, wherein the wavelength is in a range of 1.3 μm to 1.3 μm. 5. On both sides of a substrate, first and second absorption layers for a wavelength λ ₁ to be blocked are provided, a light receiving layer is provided on one of the absorption layers, and a wavelength λ to be blocked is further provided thereon. 5. The semiconductor light receiving device according to claim _{1, further comprising} an absorption layer for the light receiving device. 6. The first and second In layers on both sides of an InP substrate.
5. A pn junction is formed in the InGaAs light-receiving layer by epitaxially growing an InGaAs light-receiving layer on the GaAsP (absorption edge wavelength: about 1.4 .mu.m) absorption layer and on one of the absorption layers. The semiconductor light receiving device according to any one of the above. 7. The semiconductor light receiving device according to claim 6, wherein an InP buffer layer is grown between the InGaAsP absorption layer and the InGaAs light receiving layer. 8. The semiconductor light receiving device according to claim 7, wherein a diffusion shielding layer composed of a p region is formed around the pn junction of the InGaAs light receiving layer. 9. The semiconductor light-receiving element according to claim 1, wherein an InP window layer is provided on the InGaAs light-receiving layer, and a pn junction is provided between the InGaAs light-receiving layer and the InP window layer. 10. An InGaAs light-receiving layer on which InGaAs is formed.
An sP window layer (absorption edge wavelength: about 1.4 μm) was provided, and InGa
9. The semiconductor light receiving device according to claim 1, wherein a pn junction is provided between the As light receiving layer and the InGaAsP window layer. 11. A back-illuminated type having a ring-shaped electrode on the first absorption layer on the substrate side, an opaque metal electrode on the light-receiving layer, and most of the light-receiving layer covered with the opaque electrode. The semiconductor light receiving device according to claim 1, wherein: 12. All around a diffusion shielding layer and an opaque electrode.
Wavelength λ to block ₁Covered with a dielectric reflective film
The semiconductor light receiving device according to claim 8, wherein: