JPH0338070A - Semiconductor photodetector - Google Patents

Abstract

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

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a semiconductor light-receiving device, and particularly to a photodiode (hereinafter abbreviated as PD) capable of high-speed response.

[Conventional technology]

Figure 2 (aL (b)) shows, for example, the conventional InGaAs planar type P described in 1985 IEICE General Conference 1%51978 (pages 4 to 149).
FIG. 2(a) is a sectional view taken along line BB' of the plan view shown in FIG. 2(b) during operation. In these figures, 1 is an n''-InP substrate, 2 is an n-InP layer with 77 layers, and 3 is an n--InP substrate. G
The aAs light absorption layer, 3a is a depletion layer, 4 is an n--InP window layer, 5 is a p+ region, and the conductivity type is reversed by diffusion of impurities such as Zn. 10 is an n-electrode (hereinafter referred to as a cathode), 11 is a p-electrode (hereinafter referred to as an anode), 20 is a surface protective film (transparent to light) such as a silicon nitride film formed by a method such as plasma CVD, and 31 is a Anode wire, 40 is incident light, 50 is electron, and 51 is hole.

Next, the operation will be explained.

Generally, a crystal is grown on an InP substrate, and the lattice constant is I
Bandgap wavelength λ8 of InGaAs layer suitable for nP
is λg: 1.67 μm, and on the other hand, in InP, λ,
;0.93 μm, so InGaAs-P in FIG.
The wavelength sensitivity of D is in the ^=1.0 to 1.6 μm band.

Therefore, the operation will be described in the case where the wavelength of the incident light 40 is 1.3 μm.

Since the PD is generally used in an unbiased or reverse biased state, an applied voltage of zero or a negative voltage (−5 to −IOV) is applied to the 1 node 11 with respect to the cathode 10 . n-InP: The carrier concentration of 9 layer 4 is I x 10
The carrier concentration of the n--InGaAs light absorption layer 3, which has a m-3 layer of 1 degree, is about 5X101Seffi-, and the carrier concentration of the n--InGaAs light absorption layer 3 is smaller, so in the reverse bias state, In particular, the depletion layer 3a mainly spreads within the n--InGaAs light absorption layer 3. When light of λ≦1.3 μm is incident on the light receiving surface, this light is λ, ==
It is not absorbed by the n-I n P window layer 4 of 0.92, and n-
-Absorbed by the InGaAs light absorption layer 3, 50. A pair of holes 51 occurs. The carriers generated in the depletion layer 3a in the n'--InGaAs light absorption layer 3 are the depletion JI
It is observed in the external circuit as a drift current due to the electric field within 3a. The holes outside the depletion layer are diffused into the depletion layer 3a.
, which contributes to the drift l-current. n--I
The carrier concentration of the nGaAs light absorption layer 3 is lowered, and the depletion layer 3 is
By widening the width of a, it is possible to increase the proportion of carriers contributing to the drift current, increase the value of photocurrent (that is, improve sensitivity), and increase breakdown voltage.

[Problem to be solved by the invention]

However, in the conventional InGaAs planar PD as described above, the entire light receiving surface other than the anode 11 is 1.3
Since the surface protection 111!20 is transparent to μm light, when the incident light 40 spreads over the entire PD chip, the depletion occurs as shown in FIG. 2(a). Electron and hole pairs are generated even at a location away from j13a.

Among these holes, holes 51 from the diffusion layer to the depletion layer 3a contribute to the photocurrent, but because the diffusion is slow, the time at which the photocurrent contributes is delayed compared to carriers generated in or near the depletion layer 3a. Therefore, when the incident light 40 is in the form of a pulse, the waveform of the light-receiving current has a tail compared to the waveform of the incident light 40. Figures 3(a) and 3(b) are diagrams explaining this situation. Figure 3(a) is a pulse waveform showing the temporal change in the incident light intensity, and Figure 3(b) is the response of the photocurrent. It is a waveform. As is clear from these figures, the fall time tet is longer than the waveform of the incident light 40, that is, it is delayed by the time tfl-tt2, and the waveform cannot respond to the rapidly changing incident light 40. This problem arises.

The present invention was made in order to solve the above-mentioned problems, and an object of the present invention is to obtain a semiconductor light-receiving element that can sufficiently respond to incident light that fluctuates at high speed.

[Means to solve the problem]

The semiconductor light-receiving device according to the present invention includes an InP substrate having a first conductivity type, and a conductivity type for absorbing the InPn substrate having the first conductivity type or directly on the InP substrate. InGaA for
s InP having the first conductivity type layered on the light absorption layer
At least I from a window layer and a portion on this InPfi layer.
A conductivity type inversion region having a second conductivity type formed to reach the nGaAs light absorption layer, and a conductivity type inversion region formed on the InP window layer except for the window portion on the conductivity type inversion region and the window portion on the InPg layer. a first electrode electrically connected to the conductivity type inversion region; and an InP window layer separated from the first electrode by Ti electrical insulation and having no conductivity type inversion region formed therein. a metal light-shielding film formed on the surface through a surface protection film 5 and electrically connected to the InP window layer;
It is composed of an InP substrate and a second electrode that is electrically conductive.

[Effect]

In this invention, the metal light-shielding film prevents light from entering other than the conductivity type inversion region, and carriers that conventionally occur at a sufficient distance from the depletion layer of the light-receiving section and contribute to the photocurrent with a time delay can be eliminated. In addition to suppressing the occurrence, when the first electrode and the metal light-shielding film are electrically connected, a significant change appears in the voltage/f4 current characteristics.

〔Example〕

Figure 1 (a)? (b) is a diagram showing the structure of one embodiment of the semiconductor light-receiving element of the present invention, and FIG. It is. In these figures, FIG. 2(a).

The same reference numerals as in (b) indicate the same or corresponding parts, 6 is a contact hole processed and formed in the surface protection film 20;
7 is a metal light-shielding film newly formed according to this invention,
It is electrically and spatially insulated and separated from the anode 11 as the first electrode, and electrically connected to the n--InP window layer 4 through the contact hole 6 formed in a part of the surface protection 11420. Furthermore, the conductivity type inversion region which becomes the light receiving portion is also formed in such a manner as to cover almost the entire surface of the extra plane other than the p+ region 5.

As described in the explanation of the operation of the conventional PD in FIG.
A negative voltage is applied with respect to 0.

However, in this state, 1.3μ spread throughout the PD chip.
Even if light having a wavelength of m is incident, the light is not incident on areas other than the light receiving portion due to the metal light shield 1II7 formed according to the present invention. Therefore, carriers that contribute with a time delay, which was a problem in the characteristics of the conventional structure, are no longer generated. This results in Figure 3 (aL (b
The fall time 111 shown in ) becomes dramatically smaller than the value of the conventional PD, the tail of the pulse response waveform is improved, and the high speed of the PD is improved.

Next, the metal light-shielding film 7 formed according to the present invention has a surface-protecting layer X1ll! ! The reason why the contact hole 6 newly formed in 20 is configured to be electrically connected to the n--InP window layer 4 will be described. Even when the contact hole 6 is not formed, the metal light shielding film 7
It is considered that the target efficiency is fully achieved. In this case, metal shading #! 47 is electrically floating, but in actual device manufacturing, the anode 11 and the metal light shielding film 7 are
It is thought that the possibility that electrical conduction may occur due to various causes such as foreign matter pattern defects during the process, wire bond position defects during assembly, etc. is not completely zero, but is at least possible. If the anode 11 and the metal light-shielding film 7 are electrically connected, the dark current detected from the anode 11, that is, the current value in the absence of incident light, will be structurally different from the case without the above-mentioned defect. 1 node 11
It does not seem to have any adverse effect on the voltage/current characteristics between the and quad 10, and there seems to be no problem, but the electrode area on the AND 11 side is relatively wider by t2 by the area of the metal light shielding film 7. Therefore, the electrode capacitance increases and the response speed decreases significantly. C)l capacity is C! If L is the other capacitance, Cx is the :a fault wave number, and resistance is R, then
- It is a well-known fact that the relationship within the shell holds true.

Therefore, it is necessary to measure the capacitance between the anode 11 and the cathode 10 on all of them to eliminate the above-mentioned problems, but in reality, measuring the total capacitance is time-consuming and labor-intensive, and it is costly in terms of device production. It's not a good idea.

However, according to the structure of the present invention, the contact hole 6
Since the metal light-shielding film 7 is electrically connected to the n--InP window layer 4, even if the above-mentioned problem occurs, one node 11 is connected to the metal light-shielding film 7. n--InP window layer 4.
n--I nGaAs light absorption layer 3. n-InP buffer-r layer 2. Because it is electrically connected to the cathode 10 through the n''-InP substrate 1 with the same conductivity type, there is a large difference in voltage and current characteristics from those without the above defects, which was discovered during the voltage and current characteristics inspection stage. It can be easily detected and removed using a conventional inspection flow.

〔Effect of the invention〕

As explained above, this invention includes an InP substrate having a first conductivity type, and an
an InGaAs light absorption layer for an InPn flange 9F operation having a conductivity type, an InP window layer having a first conductivity type laminated on the InGaAs light absorption layer, and a part on the InP window layer. At least InGaA
A conductivity type inversion region having a second conductivity type formed to reach the light absorption layer, and a window portion on the conductivity type inversion region and the window portion on the InP window layer except for the window portion formed on the InP window layer. Surface preservation 3! a first electrode electrically connected to the film and the conductivity type inversion region; and electrically insulated from the first electrode.

Separately, the I
Since it is composed of a metal light-shielding film that is electrically conductive to the nP window layer and a second electrode that is electrically conductive to the InP substrate, there is a conductive film that acts as a light receiving part even for spatially spread incident light. Since no light enters areas other than the mold reversal area, and there is no generation of carriers that act with a time delay in areas other than the light receiving area, this has the effect of enabling high-speed response, as well as eliminating carriers generated during the process. Even if electrical continuity occurs between the first electrode and the metal light-shielding film due to foreign objects, pattern defects, wire bond position problems, etc., it can be removed at the time-saving and easy voltage/current characteristics inspection stage. , there is no increase in manufacturing costs.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the structure of an embodiment of a semiconductor light receiving element of the present invention, FIG. 2 is a diagram showing the structure of an example of a conventional semiconductor light receiving element, and FIG. 3 is a diagram showing the relationship between incident light and detected current. It is a diagram. Figure 1: n”-InP substrate, 2: n-InPn
Brim 9 Fufu convergence layer, 3a is depletion layer, 4 is n--InP window layer, 5 is p+
The area 6 is a contact hole, 7 is a metal light shielding film, 10 is a cathode, 11 is an anode, 20 is a surface protective film, 40 is incident light, 50 is an electron, and 51 is a hole. In addition, the same reference numerals in each figure represent the same or corresponding parts.

Claims (1)

[Claims] An InP substrate having a first conductivity type, and an InGaAs optical absorption layer having a first conductivity type laminated on the InP substrate directly or via an InP buffer layer having the first conductivity type. an InP window layer having a first conductivity type laminated on the InGaAs light absorption layer;
From a part of this InP window layer to at least the InGaA
s A conductivity type inversion region having a second conductivity type formed to reach the light absorption layer, a window portion on the conductivity type inversion region, and a window portion on the InP window layer except for the window portion on the InP window layer. a first electrode electrically connected to the conductivity type inversion region; and at least the conductivity type inversion region is formed electrically insulated and separated from the first electrode. A metal light-shielding film formed on the InP window layer that is not in contact with the surface through the surface protection film and electrically conductive to the InP window layer, and a second electrode electrically conductive to the InP substrate. A semiconductor light receiving element characterized by the following.