JPS61183958A - Solid-state photo detector - Google Patents

Abstract

PURPOSE:To obtain the titled device sensitive in every wavelength region by a method wherein the first photo receiving cell means is formed from the main surface to a depth corresponding to the first wavelength region, and the second photo receiving cell means from the main surface to a position deeper than the first region by corresponding to the second wavelength region. CONSTITUTION:A short-wavelength light absorption region 10a is formed as a MOS structure on one main surface of a p-type single crystal Si substrate 12 and consists of a relatively shallow n<+> region 14 and a p<+> region 16 thereunder: these regions form a photo diode generating photo carriers in response to incident light. A long-wavelength light absorption region 10c basically has the same structure as that of said region 10a and consists of an n<+> region 44 formed at a relatively deep junction depth and a p<+> region 46 thereunder. The p<+> chip region 46 includes a LOCOS region, and elements are isolated by reducing the cross-talk to adjacent image pickup cells by shortening the lifetime of minority carriers.

Description

DETAILED DESCRIPTION OF THE INVENTION TECHNICAL FIELD The present invention relates to a solid-state photodetection device, and particularly to a solid-state photodetection device in which a photosensitive region, that is, a light-receiving cell portion has a pn junction diode structure.

Background Art In particular, solid-state photodetectors with a photodiode structure that use an n◆p junction in the light receiving section are used to improve sensitivity to short wavelength light, that is, blue sensitivity, and to expand the dynamic range by increasing the capacitance of the junction. In order to
Conventionally, measures have been taken such as forming a relatively shallow junction from the surface of a semiconductor structure or providing a p-type region in the vicinity of the junction region.

However, along with this, the junction breakdown voltage decreases, and when such a solid-state photodetector is used as an imaging device, white scratches may appear on the image due to junction leakage current due to an increase in impurity concentration in the medium p region. There was a drawback that it appeared.

In addition, with regard to light in the long wavelength region, crosstalk may occur in which the generated photocarriers diffuse into adjacent pixel areas, and this may cause the problem of incomplete color separation in the case of color images. Ta.

Purpose The present invention eliminates the drawbacks of the prior art, and provides a method to achieve this without substantially reducing the dynamic range or junction breakdown voltage.
An object of the present invention is to provide a solid-state photodetector having high sensitivity in each wavelength region.

DISCLOSURE OF THE INVENTION A solid-state photodetection device according to the present invention includes a semiconductor substrate of one conductivity type, a first light receiving cell means formed on the main surface of the substrate, and generating photocarriers by incident light in a first wavelength range. ,
A second wavelength region formed on the main surface and having a wavelength longer than the first wavelength range.
a second light-receiving cell means for generating optical carriers by incident light in a wavelength range, and the first light-receiving cell means is formed from the main surface to a depth corresponding to the first wavelength range, The light-receiving cell means 752 includes a first region of the other conductivity type forming a junction that excites optical carriers by incident light in a wavelength range of . It includes a second region of the other conductivity type that is formed to a deeper depth and forms a junction that excites optical carriers by incident light in the second wavelength range.

DESCRIPTION OF EMBODIMENTS Next, with reference to the accompanying drawings, a solid-state optical detector 11fj according to the present invention will be described.
An embodiment of the device will be described in detail.

Referring to FIGS. 1A to 1C, in this embodiment.

One solid-state photodetector is formed with three regions, that is, light receiving cells IOa, 10b, and 10c. These regions 10a, 10b, and 10c correspond to three wavelength regions of visible light in this embodiment, namely, short wavelength, intermediate wavelength, and long wavelength regions. For example, when using a water device as a color imaging device in which many of these regions are arranged in a two-dimensional array, region 1
0a is a pixel cell (W
), the region 10b can be used as a pixel cell (Ye) sensitive to a wavelength region longer than green light, and the region 10c can be used as a pixel cell (R) sensitive to a wavelength region of red light. Or areas 10a and 1
If appropriate long wavelength light cut filters are used for 0b, the region 10a will become a pixel cell (B) sensitive to the wavelength region of dark light, and the region 10b will become a pixel cell (B) sensitive to the wavelength region of green light. G) can be used. An example in which this embodiment is applied to a color imaging device will be described below.

As shown in FIG. 1A, the short wavelength light absorption region 10a is formed as a MOS structure on one main surface of a p-type crystalline silicon substrate 12. As shown in FIG. This region 10a consists of a relatively shallow n+ region 14 that constitutes the source of a MOS transistor and contains As or P ion-implanted as an impurity, and a p÷ region 16 below it, which responds to incident light. This forms a photodiode that generates optical carriers. A region 1 is located on the main surface of the substrate 12 opposite to this.
An n+ region 18 is formed which is separated from the MO
S) It constitutes the drain of the transistor. Note that the p medium region 16 has a mechanism that increases the capacitance of the 1+p junction formed between the n medium region 14 and the p type substrate 12, and as a result increases the saturation amount of light incident on the cell 10a. However, it is not necessarily necessary.

A silicon dioxide film 20 is formed to cover these n medium regions 14 and 18, and a gate electrode 22 is formed above the silicon dioxide film 20 at a position between the circular regions 14 and 18. In this embodiment, the gate electrode 22 is made of polycrystalline silicon doped with phosphorus or As, for example. Such a short wavelength light absorption region, that is, the imaging cell 10a has a field oxide film 24 such as silicon dioxide, which is different from other imaging cells.
Elements are isolated by a p region (not shown) below.

Shallow p+ chip region 16 formed under n medium region 14
The impurity concentration is set according to the desired lower limit of the reverse breakdown voltage of the n+p junction. When using the short wavelength light absorption region 10a as a region that responds only to short wavelength light, the region 10a
An optical filter that substantially removes light with a wavelength longer than that of light in the blue region may be disposed above the blue region. In order to avoid complicating the diagram, electrical connection elements such as electrodes,
Elements that are not directly relevant to understanding the present invention, such as protective films such as passivation layers, are not shown.

As shown in FIG. 1B, the intermediate wavelength light absorption region 10b basically has the same structure as the long wavelength light absorption region 10a,
An O8 structure is formed on one main surface of the same silicon substrate 12. The difference is that the region 10b is composed of an n-medium region 34 formed at a junction depth corresponding to the wavelength region to be absorbed therein, and a p-medium region 36 below the n-medium region 34. The p-middle region 36 has the same function as the p-middle region 16 of the imaging cell 10a.

The depth of this junction is deeper than the n-middle region 14 of the imaging cell 10a, and is determined by controlling the acceleration energy when ion-implanting As ions or P ions into the n-middle region 34. In the following figures, similar elements are designated with the same reference numerals.

The width and impurity concentration of the p+ chip region 36 are set as large as possible in consideration of the desired lower limit value of the reverse breakdown voltage of the 1+p junction and the gate dark value voltage of the MOS transistor. When the intermediate wavelength light absorption region 10b is used as a region that responds only to intermediate wavelength light, a filter that absorbs light with a wavelength longer than the intermediate wavelength may be provided above the region 10b.

As shown in FIG. 1C, the long wavelength light absorption region 10c basically has the same structure as the short wavelength light absorption region 10a, and is formed as a MOS structure on one main surface of the same silicon substrate 12. . The difference is that region 10c is composed of an n+ region 44 formed at a relatively deep junction depth and a p-medium region 46 below. p middle region 46
has the same Ia ability as the p-middle region 16 of the imaging cell 10a. The depth of this junction is determined by the depth of the n-middle region 34 of the imaging cell 10b.
It is formed deeper by implanting phosphorus ions in addition to As ions into the n-middle region 34, and in this embodiment is configured to absorb light with a wavelength longer than about 800 nm. Note that the p+ chip region 4B also includes a LOGOS region, and by shortening the life of minority carriers, crosstalk to adjacent imaging cells is reduced and element isolation is achieved.

Here, the conditions for forming the p-middle regions 18, 38 and 46 will be explained in detail. A blue light absorption cell 10a having a shallow n-medium region 14 does not require a large junction capacitance. This is because visibility to blue is low, and therefore its light intensity may also be low. Therefore, the concentration of the p-type impurity in the p-type region 16 may be lower than that of the green light absorption cell 10b. Similarly, the junction capacitance of red light absorption cell 10c may be smaller than that of green light absorption cell 10b.

It is important that the p+ region 48 in the red light absorption cell 10c be formed as deep as the n+ region 44. For this purpose, high acceleration energy is required during impurity ion implantation. Therefore, B+ that passes through a thick oxide film such as the field oxide film 24 with high implantation energy.
The presence of implanted ions, such as ions, serves to prevent charges generated by long-wavelength light from leaking to adjacent pixel cells.Therefore, in this case, setting the impurity concentration of the p4 region 46 only increases the junction capacitance. It is also intended to improve crosstalk.

Therefore, as shown in FIG. 8A, the p<0> region 16 of the blue light absorption cell 10a may be locally formed in a part of the bottom region of the yl+p junction, and the p<0> region 3B of the green light absorption cell 10b may be formed locally in a part of the bottom region of the yl+p junction. As shown in FIG. 9B, it is formed wider than that of the blue light absorption cell 10a in order to increase the junction capacitance. Furthermore, the p+ region 46 of the red light absorbing cell 10c is formed to extend to the region below the field oxide film 24, as shown in FIG. 9C.

The impurity concentrations of the n+ regions 14, 34 and 44 and the p÷ regions 18, 36 and 46 will now be described.

Advantageously, the n+ region 14 of the blue light absorbing cell 10a has a lower concentration of n+ impurities than in the green light absorbing cell tab. This is to avoid reducing the sensitivity of the element to blue light by reducing the lifetime of carriers generated in this region.

It is also advantageous for the n÷region 34 of the green light absorbing cell 10b to be high to increase junction capacitance. The n region 44 of the red light absorption cell 10c is formed with a depth of n
A large amount of + ions is implanted at high energy. However, since p + ions are used, the diffusion coefficient is large and the concentration may end up being lower than in the region where As + ions are implanted. Therefore, the n♦ region 44 of the red light absorption cell 10c may have a lower impurity concentration than that of the green light absorption cell 10b.

The junction capacitance of the blue light absorption cell 10a may be smaller than that of the green light absorption cell 10b, as described above.

On the contrary, it is advantageous to increase the junction capacitance of the p+ region 3B of the green light absorption cell lOb in order to obtain a large amount of saturated light. Therefore, it is advantageous to set the concentration of p÷impurity in the p+ region 16 of the blue light absorption cell 10a to be lower than that in the p+ region 3B of the green light absorption cell 10b.

Furthermore, the main purpose of the p+ region 4B of the red light absorption cell 10c is to be formed deeply, that is, to be ion-implanted with high energy, and it is not necessarily necessary to make the junction capacitance larger than that of the green light absorption cell lOb. deep p ten area 4
It is sufficient to consider the depth of diffusion for forming B. Therefore, the impurity concentration of the p/region 46 of the red light absorption cell 10c may be lower than that of the p medium region 36 of the green light absorption cell 10b. As can be seen, the green light absorption cell 10b has the highest amount of saturated light compared to the cells 10a and 10c of other colors.

An example of a method for manufacturing the photodetecting device according to this embodiment will be described with reference to FIGS. 2 to 7C. In the figure, 1 is used for simplicity.
Although an imaging cell corresponding to one pixel is shown, in this embodiment, an imaging cell array in which a large number of such cells are arranged in a two-dimensional array on a single semiconductor substrate is manufactured. In this example, the process up to the formation of the polycrystalline silicon gate electrode 22 is the same as the manufacturing process of a normal n-channel MOS device.
For example, a dielectric isolation method using LOGOS is advantageously applied.

First, as shown in FIG. 2, a single-crystal p-type silicon substrate 12 is prepared, and a 5i02 layer 100 is formed on one main surface of the substrate.
and Si3N, forming layer 102. Next, as shown in FIG. 3, a region constituting an imaging cell and an element isolation region are formed by LOIII; O9. The imaging cell area is masked with photoresist 104 and the remaining area is plasma etched. B+ ions are implanted from the exposed portion of the silicon dioxide layer 100, and a field oxide film 24 is further grown. FIG. 4 shows a state in which the photoresist 104 and the nitride film 102 have been removed.

Next, a gate oxide film is grown on this structure to form a MOS
A gate oxide film 20 is formed, and ions for controlling the threshold voltage of the gate electrode 22 are implanted. A polycrystalline silicon layer 10B is deposited to the north of this (FIG. 5).

A region where the gate electrode 22 is to be formed is masked, the remaining region is removed by plasma etching, and As ions are implanted. The acceleration voltage for ion implantation is, for example, about 50~
200KeV, impurity density is about 0.1~~1xlO18
cm-2. As a result, a shallow n-medium region 14 and a drain region 18 are formed (FIG. 6). The depth of the n-medium region 14, that is, the depth of the junction, is 0.11Lrs to 0.4Lrs.
JLtm level is advantageous.

When forming the short wavelength light absorption region 10a on the main surface of the semiconductor structure prepared in this way, as shown in FIG. Then, boron ions B+ are implanted. The accelerating voltage for ion implantation is, for example, about 3
0 to 200KeV, impurity concentration approximately Oo1 to 3x101
There is 3C11-2. As a result, a shallow p+ region 16 is formed below the n+ region 14. Note that the boundary between the n♦ region 14 and the field oxide film 24 has the greatest risk of leakage from the pn junction.

It is advantageous to avoid this portion and implant the p-doped impurity. When the photoresist 108 is removed, the state shown in FIG. 1A is obtained.

When forming the intermediate wavelength light absorption region 10b on the main surface of the semiconductor structure, as shown in FIG. 7B, a portion of the main surface where the p+ chip region 3B will be formed is opened and the rest is covered with photoresist 11O, and boron Ions B÷ or B÷+ are implanted. The acceleration voltage for ion implantation is, for example, about 200
~500KeV, impurity concentration approximately 0.1~8Klo13
It is layer 0-2. By this, p of the imaging cell loa ÷
A medium p region 36 having a higher impurity concentration than the region 16 is formed.

Next, P (phosphorus) ions are implanted. The accelerating voltage for ion implantation is, for example, about 10 to 200 KeV, the impurity concentration is about 0.1 to 2xlO16c++-2, and the ion implantation is performed at a depth of about 0.4 to 0.8 l in the direction perpendicular to the substrate 12. .

This completes the n♦ region 34. When the photoresist 110 is removed, the state shown in FIG. 1B is obtained.

The long wavelength light absorption region 10c is formed as follows.

As shown in FIG. 7C, a portion of the main surface where the p+ chip region 46 will be formed is opened, the rest is covered with a photoresist 112, and boron ions B+1 are implanted. In this case p
The middle region 4B does not need to be localized in the region directly under the n+ region 44 as in conventional elements, but is intended to prevent charges excited by long wavelength light from laterally diffusing to adjacent pixel cells. The acceleration voltage for ion implantation, which may be distributed also in the boundary between adjacent pixel cells, that is, in the region under the field oxide film 24, is, for example, approximately 400 KeV to IMe.
V, the impurity concentration is about 0.01 to 1×10 15 cm −2 te. As a result, a p-medium region 4B that is deeper in the vertical direction than the p-medium region 36 of the imaging cell 10b is formed.

Next, P (phosphorus) ions are implanted. The acceleration voltage for ion implantation is, for example, about 200 to GOOKeV, and the impurity density is about 0.1 to 2xlO"'cm-". As a result, the deep n-medium region 44 is completed. Photoresist 112
When is removed, the state shown in FIG. 1C is obtained.

The subsequent steps may be the same as normal device manufacturing steps.

Electrical connection contact holes are opened, and processes such as PSG deposition, annealing, aluminum deposition, etching, passivation film deposition, bonding, and packaging are performed to complete the device.

In this way, each imaging cell 10a, 10b and 1
0c are formed on the substrate 12 in a two-dimensional array. When this is used, for example, as a Bayer array imaging cell array as shown in FIG. 8A, a mosaic color filter is arranged above the light receiving area of each imaging cell 10a, 1Ob, and 1Oc. In this example, the imaging cell 10a is used as a blue (B) cell, the imaging cell 10b is used as a green (G) cell, and the imaging cell 10c is used as a red (R) cell. Therefore, those and
B, G and R color filter segments 120 respectively
B, 120G and 120R are deposited, respectively.

Note that it is advantageous to arrange an infrared (IR) cut filter above each of these imaging cells. In that case, the red imaging cell lOc does not necessarily need to be provided with an R filter.

The cross section seen from the dashed line B-8 in FIG. 8A is schematically shown in FIG.
As shown in Figure B, below the filter segment 120B of B is a relatively shallow 1+ region 14 of the short wavelength absorption region 10a.
is located, and below the G filter segment 120G, the n medium region 34 of the intermediate wavelength absorption region 10b is located, and this arrangement is alternately repeated. In addition, in the same figure, the bottoms of n+ regions 14, 34 and 44, that is, fi+
The location of the p-junction is indicated conceptually by line 122, and other parts are omitted.

Similarly, as shown schematically in FIG. 8C, a cross section taken along the dashed line CC in FIG. 8A, the R filter segment 1
A relatively deep n-middle region 44 of the long wavelength absorption region 10b is located below 20R, and an n-middle region 34 of the intermediate wavelength absorption region 10b is located below the G filter segment 120G, and this arrangement is alternately arranged. Repeated. In this way, in this example, a solid-state imaging device is configured using a Bayer array pixel array. Note that when color filters are not used, each pixel cell 10a, 10b, and 10c may be arranged in W, Ye, and H, respectively.

As described above, in this embodiment, the junction is formed with a depth and size depending on the wavelength range of the incident light. Therefore, as a photodetecting device, it is possible to particularly improve blue sensitivity, and it is possible to correspond to color wavelength regions and set a dynamic range. In addition, since the area of the P+ chip region and the depth and concentration of impurity distribution can be set for each pixel cell according to the absorption wavelength, a junction capacitance corresponding to each spectral wavelength intensity can be obtained, and the junction breakdown voltage of cell separation can be reduced. The decline can be prevented. Therefore, it is not necessary to increase the impurity concentration in the p/region region of the pixel cell region corresponding to the wavelength range where the light intensity is relatively low, so that the junction leakage current is small and the occurrence of white scratches in one image is prevented. Further, by deepening the vertical distribution of the p-domain, crosstalk in long wavelength light is improved.

The embodiments described here are for illustrating the present invention, and the present invention is not necessarily limited thereto, and modifications and variations that can be made by those skilled in the art without departing from the spirit of the present invention. Modifications are within the scope of this invention. For example, in the embodiment, a cell 10a corresponding to three types of wavelength ranges,
10b and 10c were provided, but it is not necessary to do this, and the visible light region may be divided into two types, short wavelength and long wavelength, and two types of light receiving cells corresponding to both may be provided.

Effects As described above, the solid-state photodetector according to the present invention has high sensitivity in a desired wavelength range without substantially reducing the dynamic range or junction breakdown voltage. Therefore, if this solid-state photodetection device is applied to an imaging device, a reproduced image with good sensitivity even in a short wavelength region and with few white scratches and good color separation can be obtained.

[Brief explanation of the drawing]

1A to 1C are cross-sectional views showing configuration examples of the light receiving area in an embodiment of the solid-state photodetecting device according to the present invention, and FIGS. 8A is a cross-sectional view showing step-by-step an example of a method for manufacturing a solid-state photodetection device having the configuration shown in FIGS. An explanatory diagram conceptually illustrating an example; Figures 8B and 8C are diagrams schematically showing cross sections taken from dashed-dotted lines B-8 and CC in Figure 8A, respectively; Figures 8A and 9B; and FIG. 8C is a graph showing an example of setting the impurity concentration of the junction region. Explanation of symbols of main parts 10a, IOb, , imaging cell 14, 34, 44, n+ region ilt, 38.4B, p+ region 1B, , , drain region 20, , gate oxide film 22, - , , Gate electrode patent applicant Fuji Photo Film Co., Ltd. Book 1A Figure 8B 8 CUD cap Inuhan Yam, Clam Zero Mai, Reinan C Katsuo Rin Proceedings Amendment Document April 15, 1985

Claims (1)

[Claims] 1. A semiconductor substrate of one conductivity type; a first light-receiving cell means formed on the main surface of the substrate and generating optical carriers by incident light in a first wavelength range; a second light-receiving cell means formed on a surface and generating optical carriers by incident light in a second wavelength range having a longer wavelength than the first wavelength range, the first light-receiving cell means comprising: is formed to a depth corresponding to the first wavelength range,
The second light-receiving cell means includes a first region of the other conductivity type forming a junction that excites optical carriers by at least incident light in a first wavelength range, and the second light-receiving cell means A solid-state photodetection device comprising a second region of the other conductivity type that is formed to a deeper depth than the first region and forms a junction that excites optical carriers by incident light in a second wavelength range. 2. The device according to claim 1, wherein the device is formed on the main surface and drives optical carriers by incident light in a third wavelength range between the first wavelength range and the second wavelength range. The third light receiving cell means is formed from the main surface to a depth deeper than the first region and shallower than the second region according to the third wavelength range, and at least Third
A solid-state photodetection device comprising a third region of the other conductivity type forming a junction that excites optical carriers by incident light in a wavelength range of . 3. In the device according to claim 1 or 2, at least one of the first to third light receiving cell means
A solid-state photodetection device characterized in that it has a fourth region of one conductivity type that contains more impurities of one conductivity type than the substrate below the corresponding first to third regions. 4. In the device according to claim 3, the impurity concentration in the fourth region formed in the third light receiving cell means is higher than the impurity concentration in the fourth region formed in the first light receiving cell means. A solid-state photodetecting device characterized in that the impurity concentration in the fourth region formed in the second light receiving cell means is higher than the impurity concentration in the fourth region formed in the third light receiving cell means. 5. In the device according to any one of claims 1 to 4, at least one of the first and third light receiving cell means has a wavelength longer than the corresponding first or third wavelength range. A solid state photodetection device comprising optical filter means for substantially removing incident light.