PHOTOELECTRIC CONVERSION ELEMENT

Abstract

A photoelectric conversion element comprising: a) a first zone of a first conduction type, the first zone being a semiconductor base body (11, 21, 30) itself; b) a dot-shaped second region of a second conductivity type different from the first conductivity type, the dot-shaped second region being formed by diffusing an impurity or by growing an epitaxial layer on a surface of the first region within a light-receiving region; c) a light-blocking member (50) for blocking light incident on the point-shaped second zone from light incident on the light-receiving area on an entrance surface receiving light from outside; and d) a wiring part (15, 27, 45) having a first branch and a second branch arranged parallel to each other within the light receiving area, wherein the point-shaped second zone generates a photodetection signal by collecting charge carriers from a surrounding area which are generated in the first zone by the light incident on the first zone in the light receiving area, wherein a plurality of the point-shaped second zones are arranged in a mutually separated form within the light receiving area to obtain a single photodetection signal, wherein the plurality of second zones are interconnected so as to sum photodetection signals generated by the individual point-shaped second zones, and wherein the wiring part (15, 27, 45) connects the plurality of point-shaped second zones via the first branch and the second branch, and wherein the charge carriers collected in the plurality of point-shaped second zones flow externally as photocurrent.

Description

TECHNICAL FIELD

[0001] The present invention relates to a photoelectric conversion element for receiving light and converting it into electrical signals. The term "photoelectric conversion element" used herein includes photodiodes, phototransistors, Darlington phototransistors, photothyristors, phototriacs, and other similar elements. The photoelectric conversion element may be constructed as a standalone element, or it may be a portion of an integrated circuit (IC) or a large-scale integrated circuit (LSI) in which various other functions are provided. STATE OF THE ART

[0002] A photodiode typically comprises an n-type semiconductor region and a p-type semiconductor region formed by selectively diffusing an impurity (e.g., boron) into the n-type semiconductor region. The p-type semiconductor region and the n-type semiconductor region form a pn junction. When light of a suitable intensity arrives at the photodiode, electron-hole pairs are generated throughout the body of the photodiode, i.e., in the depletion layer near the junction of the p-type and n-type semiconductor regions, in the p-type semiconductor region, and in the n-type semiconductor region. Normally, electrons and holes are accelerated to the n-type semiconductor region and the p-type semiconductor region, respectively, due to the action of an electric field in the depletion layer.Among the electron-hole pairs generated in the n-type semiconductor region, the electrons remain in the n-type semiconductor region along with the electrons transferred from the n-type semiconductor region, while the holes in the n-type semiconductor region diffuse to the depletion layer. Upon reaching the depletion layer, the holes are accelerated by the electric field and collected in the p-type semiconductor region. In this way, holes and electrons are collected in the p-type and n-type semiconductor regions, respectively, and flow through an externally connected load as a photocurrent.

[0003] Even though the conductivity types of the semiconductors are opposite to those described previously, the structure and operation of the device are basically the same.

[0004] In commonly used photodiodes, the p-type semiconductor region formed by selective diffusion almost completely covers the light-receiving surface that receives incident light, thus allowing the light to reach the entire junction between the p-type and n-type semiconductor regions. Fig. 15A is a schematic sectional view of a typical photodiode, and Fig. 15B is its top view. In the present example, the base body 11 itself serves as the n-type semiconductor region, and the p-type semiconductor region 12 is formed by selective diffusion over almost the entire area such as the light-receiving surface 10 on the surface of the base body 11. The contact part 13, which consists of a conductor formed in contact with the base body 11, is the cathode terminal (C), while the contact part 14, which consists of a conductor formed in contact with the p-type semiconductor region 12, is the anode terminal (A).

[0005] To achieve a high level of photodetection sensitivity in such a photodiode, the light-receiving surface 10 should preferably have a large area. If the light-receiving surface 10 is enlarged, the p-type semiconductor region 12 must also be enlarged for the reason described above. However, increasing the area of ​​the p-type semiconductor region 12 causes an increase in the junction capacitance, which in turn increases the noise level in elements such as the amplifier connected to convert the photocurrent generated by the photodiode into voltage. Consequently, the signal-to-noise (SN) ratio of the photodetection signal decreases, necessitating a reduction in the frequency bandwidth of the amplifier or a reduction in its gain.

[0006] In other words, to reduce high-frequency noise in a photodiode, the junction capacitance must be reduced (see, for example, JP H11-312823 A). However, if the area of ​​the p-type semiconductor region formed by selective diffusion is reduced to reduce the junction capacitance, the photodetection sensitivity may be reduced.

[0007] To reduce the junction capacitance of the pn junction, a conventional photodiode described in JP S59-12034 B uses a structure in which multiple island-like p-type diffusion layers are formed on the surface of an n-type substrate, and the same number of electrodes as the island-like diffusion layers are provided in an interconnected manner. The spacing of the island-like p-type diffusion layers is designed to be equal to or smaller than the distance over which minority carriers diffuse ("minority carrier diffusion length").JP S59-12034 B claims that such a structure provides the photodiode with a larger light-receiving area, effectively operating in both the horizontal and vertical directions from each island-like p-type diffusion layer. The photocurrent corresponding to the incident light can also be obtained in the regions between the p-type diffusion layers ("non-diffusion layer regions"), so that the photodetection sensitivity hardly deteriorates compared to the case where the same diffusion layer is formed over the regions between the island-like p-type diffusion layers. It is also claimed that the junction capacitance becomes lower because the area of ​​the pn junction is reduced by an amount corresponding to the non-diffusion layer regions. CITATION LISTPATENT LITERATURE Patent Literature 1: JP H11-312823 A Patent Literature 2: JP S59-12034 B Patent Literature 3: JP 2010-102387 A Patent Literature 4: US 6,150,704 A BRIEF DESCRIPTION OF THE INVENTION TECHNICAL PROBLEM

[0008] In fact, the structure of the element described in JP S59-12034 B does not significantly reduce the junction capacitance, partly due to the large area of ​​the diffusion zone. The performance improvement due to the reduced junction capacitance is also less than expected. Accordingly, a photoelectric conversion element with a novel structure capable of achieving an even greater reduction in junction capacitance while preventing deterioration in photodetection sensitivity has been demanded. For example, in a photoelectric switch that uses modulated light to reduce noise susceptibility to ambient light, when the modulation frequency is set to a high value, it is important to reduce the pn junction capacitance and thereby achieve the highest possible SN ratio.Accordingly, for a photoelectric conversion element used as a photodetector or a similar element in a photoelectric switch, it is particularly important to reduce the pn junction capacitance.

[0009] In a photodiode with a commonly used structure, the wavelength sensitivity characteristics can be controlled to some extent by the thickness of the diffusion layer, the concentration of impurities, or other parameters. However, significant sensitivity adjustment is difficult. Therefore, when it is necessary to reduce the influence of short-wavelength light, such as in the case of optical measurement with a large amount of ultraviolet light, an optical filter is usually used with the photodiode to block light in an undesirable wavelength range. However, integrating such an optical filter with the photodiode into a single unit requires an additional manufacturing process, whereas using a separate optical filter with the photodiode increases the number of parts. In either case, the manufacturing cost increases dramatically.Accordingly, a photodiode or other types of photoelectric conversion elements that can block short wavelength light without using an additional structural component have been demanded.

[0010] The present invention has been developed with such problems in mind. Its primary objective is to provide a photoelectric conversion element in which the junction capacitance of a pn junction (or other types of junctions) can be smaller than the conventional level while preventing a reduction in photodetection sensitivity, thus reducing the noise level dependent on the junction capacitance and thereby improving the SN ratio of the photodetection signal.

[0011] Another object of the present invention is to provide a photoelectric conversion element having an easily implemented structural device that can effectively block light at shorter wavelengths within the wavelength range of spectral sensitivity obtained by the photoelectric conversion element portion, while ensuring a high level of photodetection sensitivity at longer light wavelengths. PROBLEM SOLVING

[0012] The photoelectric conversion element according to the present invention, which has been developed to solve the above-described problem, comprises the features of claim 1.

[0013] The photoelectric conversion element according to the present invention comprises a plurality of the second zones arranged in a spaced-apart manner within the light receiving area to obtain a single photodetection signal, the plurality of second zones being interconnected so as to sum photodetection signals generated from the individual second zones.

[0014] Examples of the photoelectric conversion element according to the present invention include photodiodes, phototransistors, Darlington phototransistors, photothyristors, phototriacs, and other similar elements. The photoelectric conversion element according to the present invention may be configured as a standalone element, or it may be a component in a larger device (such as a photodiode array in which multiple photodiodes are arranged), or it may be a portion of an IC or LSI in which various other functions are provided.

[0015] In the photoelectric conversion element according to the present invention, either the combination of n-type as the first conductivity type and p-type as the second conductivity type or conversely p-type as the first conductivity type and n-type as the second conductivity type is possible.

[0016] In conventional and commonly used photoelectric conversion elements, including the one described in JP S59-12034 B, not only the first region but also the second region formed within the light-receiving area by diffusing an impurity has the function of generating charge carriers in response to incident light. In comparison, in the photoelectric conversion element according to the present invention, the very small second regions are shielded from light by the light-blocking member and do not significantly contribute to photoelectric conversion. Their primary function is to collect the charge carriers generated by photoelectric conversion in the surrounding area, i.e., the first region.This means that when charge carriers are generated by light incident on the first zone, they move by drift into the depletion layer formed around the second zone or by diffusion into the first zone outside the depletion layer. The second zone collects the charge carriers arriving in this zone after they have been generated in the surrounding regions. Thus, the charge carriers are reflected in the photocurrent. Since the second zone only needs to be able to collect the charge carriers coming from the surrounding regions, its area can be significantly reduced, resulting in a very small junction capacitance at a second zone.

[0017] Light of a longer wavelength penetrates into a deeper portion in the first region and generates charge carriers in that region. Conversely, light of a shorter wavelength generates charge carriers in a near-surface region of the first region. The charge carriers generated in the near-surface region easily reach the surface during their self-diffusion movement and undergo surface recombination, causing a significant loss of photocurrent. Therefore, charge carriers generated by the light with longer wavelengths are more likely to reach the second region, while those generated by the light with shorter wavelengths are less likely to reach the second region. In the photoelectric conversion element according to the present invention, the generation of charge carriers in the second region is almost completely prevented by shielding the second region from light by the light-blocking member.Therefore, the charge carriers that would normally be generated in the second zone due to the shorter wavelengths of light are practically negligible. Consequently, the photodetection sensitivity to light of shorter wavelengths, or in particular to light with wavelengths of 400-450 nm or shorter, is reduced without reducing the photodetection sensitivity to light of longer wavelengths. In other words, the influence of ultraviolet light, which has shorter wavelengths, is reduced, allowing the light of interest to be accurately detected.

[0018] The dot-shaped second region in the photoelectric conversion element according to the present invention has a reasonably smaller size than the diffusion region in the conventional photoelectric conversion element of this type. For example, this is a region whose area is equal to or smaller than 1% of the entire light-receiving area and can be regarded as a dot compared to the entire light-receiving area. In particular, in the case of a photodiode used in a photoelectric switch (which normally has a light-receiving area of ​​approximately 10,000 µm 2 up to 1,000,000 µm 2 the area of ​​the second zone should preferably be equal to or less than 100 µm 2 (equal to or smaller than the 10 µm square if the zone has a square shape) or, more preferably, equal to or smaller than 25 µm 2 (equal to or smaller than the 5 µm square if the zone has a square shape).

[0019] However, if multiple second zones are arranged within the light-receiving region, even if the area of ​​each zone is small, the total area of ​​the second zones (the sum of the areas of the individual second zones) per light-receiving region increases with the number of second zones. The larger their total area, the smaller the area that effectively contributes to photoelectric conversion and the larger the junction capacitance in the light-receiving region.

[0020] Accordingly, as an object in the photoelectric conversion element according to the present invention, the total area of ​​the plurality of second zones within the light receiving area should preferably be equal to or smaller than 5% of the area of ​​the light receiving area.

[0021] That is, as a preferred mode of the photoelectric conversion element according to the present invention, the second region having a small size, as just mentioned, may preferably be arranged, for example, in the form of a large number of dots in a suitable distribution density on the surface of the first region within the light-receiving region, wherein the total area of ​​the large number of second regions is equal to or less than about 5% of the area of ​​the light-receiving region, or more preferably, equal to or less than about 1%. That is, in the case of the above-mentioned photodiode used in a photoelectric switch, the total area of ​​the second regions should preferably be equal to or less than about 500 µm 2 up to 50,000 µm 2 be.

[0022] With this design, it is possible to lower the junction capacitance of the junction between the first and second regions in the light-receiving region, thus reducing the noise level dependent on the junction capacitance, while efficiently collecting the charge carriers generated in the first region by the light arriving in the light-receiving region to obtain an appropriate signal strength level. Consequently, the signal's SN ratio is improved.

[0023] Among the previously described carrier transfer modes, the transfer of carriers occurs by diffusion in various directions, during which the carrier is directed toward the second region by drift. Therefore, the latter mode provides a higher carrier collection efficiency. Accordingly, to maximize the signal strength corresponding to the amount of received light, it is preferable to reduce the gap between the depletion layers formed around the second regions. To this end, the gap between the adjacent second regions must not be too large. Taking this into account, the distribution density of the plurality of second regions or the number of second regions per unit area can preferably be determined in addition to the percentage of the total area of ​​the second regions in the light-receiving surface.

[0024] The above-mentioned JP S59-12034 B argues that the junction capacitance of the pn junction depends on the area of ​​the p-type diffusion region in the n-type region. Indeed, the electrostatic capacitance of a junction (typically a pn junction) between a first region and an island-like second region formed in the first region is expressed as a function not only of the area of ​​the second region, but also of the peripheral length of the second region, the depth of the depletion layer, and other parameters. The depth of the depletion layer depends on the impurity concentration in the first and second regions, as well as on the (reverse) bias applied between the two regions. Therefore, it is possible to consider the junction capacitance as dependent on the area of ​​the second region and the peripheral length of the second region, provided that the impurity concentration and the reverse bias are equal (see, for example, JP 2010-102387 A).

[0025] Accordingly, as a mode of the photoelectric conversion element according to the present invention, the number of second regions and the area and circumferential length of each second region should preferably be determined such that the sum of the junction capacitances between the first region and the plurality of second regions is smaller than a junction capacitance observed at the junction between the first region and the second region when the second region is formed over the entire light-receiving area. Particular attention should be paid to the fact that providing a large number of second regions causes the junction capacitance to be significantly influenced by the circumferential length of each second region.

[0026] This design makes it possible to reliably make the junction capacitance of the pn junction lower than that of conventional photoelectric conversion elements. Consequently, the noise level dependent on the junction capacitance is reduced, and the SN ratio of the photodetection signal is improved.

[0027] The photoelectric conversion element according to the present invention may further comprise: a plurality of contact parts provided respectively above the plurality of second zones, each contact part being electrically connected to the second zone located thereunder; and the wiring part, which is an electrical conductor, for connecting the plurality of contact parts to each other so as to connect the plurality of second zones within the light receiving area in parallel, wherein all second zones within the light receiving area are located either directly below the contact parts and / or the wiring part to configure the contact parts and / or the wiring part to function as the light blocking part.

[0028] In a typical semiconductor process, the second region is formed by diffusing an impurity, and therefore extends slightly beyond the area where the impurity is injected. On the other hand, the contact portion is formed slightly inside the junction boundary to prevent short circuits with the first region. Due to such design factors, the outer edge of the second region is slightly outside the outer edge of the contact portion, even if the contact portion and the wiring portion are formed to cover the second region from above. Even in this case, the contact portion can satisfactorily function as the light-blocking portion.

[0029] According to this configuration, the contact part and / or the wiring part for connecting the plurality of second zones to each other can be used as the light-blocking part. Therefore, no special manufacturing process is required for providing the light-blocking part.

[0030] In the photoelectric conversion element according to the present invention, the wiring part located above a depletion layer spreading region formed in the first region surrounding the second region should preferably have a smaller line width than the contact part and the wiring part located above the second region.

[0031] It is essential that at least a portion of the wiring connecting the contact parts crosses the first region. However, in the above-mentioned configuration, the percentage of light blocked by the wiring part is very small or almost negligible, allowing a larger amount of photocurrent to be obtained.

[0032] Essentially, the second zone can have any shape when viewed from above, for example, a square, a rectangular, a polygonal, an elliptical, and a circular shape. However, provided the zone has the same area, a shape with a shorter perimeter is more advantageous for reducing the junction capacitance. Therefore, the second zone, viewed from above, should preferably have a roughly square shape rather than a rectangular shape, more preferably a polygonal shape, and ideally a circular shape.

[0033] In the case where the plurality of second zones are arranged in the form of dots in the manner described above, the second zones may be arranged at the intersection points of a rectangular grid, or this arrangement may be modified so that the position of the second zones in every second row or column is shifted by one half of the pitch of the second zones along the extending direction of the row or column.

[0034] Compared to the arrangement in which the second zones are located at the intersection points of a rectangular grid, the latter configuration requires a smaller number of second zones to be arranged, provided that each point within the light-receiving area has the nearest second zone within a specific distance. As a result, the junction's junction capacitance is further reduced while maintaining photodetection sensitivity at approximately the same level.

[0035] Further, in the case where a large number of second zones in the form of dots are created in the manner described above, it is preferable to arrange a plurality of second zones in straight lines, and connect the plurality of second zones in each straight line by the wiring member having a linear shape, wherein the pitch of the plurality of second zones in the extending direction of the wiring member is smaller than the pitch of the plurality of second zones in the direction perpendicular to the extending direction of the wiring member. This configuration allows the wiring members to be arranged with a larger pitch under the condition that each dot within the light-receiving surface has the nearest second zone within a specific distance, so that the percentage of light blocked by the wiring members will be low.As a result, a greater amount of light is received and the signal strength is increased.

[0036] In the photoelectric conversion element according to the present invention, a highly concentrated region having a predetermined shape when viewed from above, for example, a dot-like shape or a linear shape, may be formed outside the depletion layer region located between the plurality of second regions discretely arranged in the first region. The highly concentrated region has a higher concentration of first conductivity type impurities than the surrounding first region. Diffusion of this highly concentrated region can be achieved not only by surface diffusion but also by implant-type diffusion.

[0037] This design creates an impurity concentration gradient within the first zone from the position of the high-concentration zone to the second zone, thereby forming a potential gradient to promote charge carrier transfer. Consequently, charge carriers are transferred more efficiently to the second zone, achieving higher photodetection sensitivity and higher operating speed.

[0038] In the photoelectric conversion element according to the present invention, as already explained, the area of ​​the second region should be as small as possible in terms of junction capacitance, provided that certain specific requirements, such as secure and reliable electrical connection with the contact part, are met. On the other hand, the gap between the adjacent second regions should preferably be equal to or smaller than the width of the drift region in which the charge carriers generated by light incident in the first region can drift, and also equal to or smaller than the distance at which the charge carriers can reach the second regions with a reasonably high level of efficiency (i.e., at which the decrease in photodetection sensitivity is sufficiently small).

[0039] However, it is difficult to increase the movement speed of charge carriers by diffusion in the first region outside the depletion layer to a level comparable to the movement speed of charge carriers by drift within the depletion layer. Accordingly, if high-speed operation is particularly important, the second regions should preferably be arranged so that no gap is left between the depletion layer spreading regions formed around the second regions. This configuration allows almost all charge carriers generated in response to the incident light to reach the second regions by drift, thus achieving a high response speed. The efficiency of charge carrier collection will also be improved.Of course, in this case too, the gap between the second regions should preferably be as large as possible to reduce the junction capacitance of the entire light-receiving region. Accordingly, as one mode of the photoelectric conversion element according to the present invention, the plurality of second regions should be arranged at the intersection points of a honeycomb pattern as viewed from above, as much as possible, so that the second regions are arranged without leaving any gap between the depletion layer spreading regions while minimizing the overlapping area of ​​the depletion layer spreading regions.

[0040] In order to simultaneously realize both a high-speed response and a large area of ​​the light receiving region, the plurality of second regions may preferably be arranged separately from each other so as to avoid overlap of the depletion layer spreading regions formed in the first region around the respective second regions, and a light-blocking member for blocking light incidence may be provided over a portion or the entirety of the area outside the depletion layer spreading region around every second region within a light receiving region.

[0041] With this configuration, even when the light-receiving region has a large area, almost all the carriers that respond to the incident light are generated within the depletion layer spreading region around every other region. Therefore, the carriers quickly reach the second regions, thereby achieving high-speed response. In addition, even when the light-receiving region has a large area, the total number of second regions can be comparatively small, so that the junction capacitance can also be comparatively low. On the other hand, the large area of ​​the light-receiving region enables efficient reception of the incident light incident over a wide area, as well as reliable reception of the incident light incident at various positions within a wide range. This increases the allowable range for adjusting the optical axis of the incident light.

[0042] If the second zones are created in the form of dots, the wiring connecting the second zones is inevitably provided within the light-receiving surface. If electromagnetic noise enters this wiring, the noise level of the photodetection signal will be significantly increased.

[0043] Accordingly, the photoelectric conversion element according to the present invention may further comprise: an insulating film formed to cover the contact part and the wiring part; and a conductor part provided on the insulating film over the contact part and the wiring part, wherein a predetermined fixed electrical potential is applied to the conductor member to configure the conductor member to function as an electrical shield.

[0044] With this configuration, the conductor part produces the electromagnetic shielding effect to prevent external noise from entering the wiring part in the lower layer, thereby reducing the influence of external noise on the photodetection signal.

[0045] In the photoelectric conversion element according to the present invention, it is even more preferable that a plurality of conductive columnar parts extending from the conductor part located above the wiring part into the insulating film positioned between the conductor part and the wiring part are formed in such a manner as to surround the lateral sides of the wiring part or that they form conductive linear parts extending substantially parallel to the wiring part in such a manner as to be placed on both sides of the wiring part, and wherein the columnar parts or the linear parts are designed to function as an electrical shield together with the conductor part.

[0046] According to this configuration, not only the conductor part has an electromagnetic shielding effect; the multiple columnar parts or the linear parts located on the lateral sides of the wiring part also have an electromagnetic shielding effect. Therefore, noise signals incident obliquely on the wiring parts can also be blocked, thereby achieving an even higher level of electromagnetic shielding effect.

[0047] It is also possible to provide conductor parts above and below at least a portion of the wiring part with an insulating film between the wiring part and each of the upper and lower conductor parts, and to form a plurality of vias connecting the upper and lower conductor parts as the columnar parts.

[0048] As already explained, if the total area of ​​the second zones is sufficiently smaller than the area of ​​the light receiving zone, the primary effect of reducing the junction capacitance-dependent noise level can be achieved without the light-blocking part.

[0049] Accordingly, the photoelectric conversion element according to the second present invention, which has been developed to solve the above-described problem, comprises the features of independent claim 2. ADVANTAGEOUS EFFECTS OF THE INVENTION

[0050] In the photoelectric conversion element according to the present invention, the junction capacitance of the photoelectric conversion junction is reduced, thereby reducing high-frequency noise dependent on the junction capacitance and improving the SN ratio of the photodetection signal. Furthermore, in the photoelectric conversion element according to the present invention, the selective diffusion area for forming the second regions is smaller than in conventional devices. Therefore, for example, the pn junction area is also smaller, so that deterioration of characteristics or failure due to crystal defects or other factors is less likely to occur.

[0051] In the photoelectric conversion element according to the present invention, the second regions are shielded from light by the light-blocking part, such as the contact part or the wiring part. Therefore, it is possible to achieve both a reduction in the junction capacitance for photoelectric conversion and a simple structure for implementing a photodetection sensitivity characteristic with a reduced sensitivity to short-wavelength light. Therefore, even if the photoelectric conversion element is used under conditions where a significant amount of ultraviolet light is present as a noise component, the influence of this noise is reduced, and the photodetection signal with high SN ratios can be obtained.Since the light-blocking part is formed by a normal manufacturing process for photoelectric conversion elements, the above-mentioned effects can be obtained without any particular increase in device costs. SHORT DESCRIPTION OF THE DRAWING Fig. 1A, Fig. 1B and Fig. 1C are a schematic sectional view, a top plan view, and an equivalent circuit diagram of a photodiode as an embodiment of the photoelectric conversion element according to the present invention, respectively. Fig. 2 is an enlarged sectional view of an area around an n-type semiconductor region in the photodiode of the present embodiment. Fig. 3A and Fig. 3B are enlarged top plan views of an area around an n-type semiconductor region in the photodiode of the present embodiment. Fig. Figure 4 is a top view of a photodiode in one variant. Fig. 5A and Fig. 5B are a top view and a schematic cross-sectional view of a photodiode in another variant, respectively. Fig. 6 is a schematic sectional view of a photodiode in yet another variant. Fig. Figure 7 is a top view of a photodiode in yet another variant. Fig. Figure 8 is a top view of a photodiode in yet another variant. Fig. Figure 9 is a top view of a photodiode in yet another variation. Fig. 10A and Fig. 10B are an enlarged sectional view and a schematic sectional view of a variant in which a shielding wiring part is added. Fig. 11 is a schematic sectional view of another variant in which a shielding wiring part is added. Fig. 12A is a schematic sectional view of a conventional and commonly used phototransistor, and Fig. 12B is a schematic sectional view of a phototransistor as an embodiment of the photoelectric conversion element according to the present invention. Fig. 13 is a schematic sectional view of a phototransistor on an integrated circuit as an embodiment of the photoelectric conversion element according to the present invention. Fig. 14 is a schematic sectional view of a Darlington phototransistor as an embodiment of the photoelectric conversion element according to the present invention. Fig. 15A and Fig. 15B are a schematic sectional view and a top plan view of a conventional and commonly used phototransistor, respectively. DESCRIPTION OF EMBODIMENTS

[0052] Hereinafter, embodiments of the photoelectric conversion element according to the present invention will be described with reference to the accompanying drawings. [First embodiment]

[0053] Fig. 1A is a schematic sectional view of a photodiode as an embodiment of the photoelectric conversion element according to the present invention, Fig. 1B is a top view, and Fig. 1C is an equivalent circuit diagram. Fig. Fig. 2 is an enlarged sectional view of an area around an n-type semiconductor region in the photodiode of the first embodiment. Fig. 3A and Fig. 3B are enlarged top plan views of an area around an n-type semiconductor region in the photodiode of the first embodiment.

[0054] As this is Fig. 1B, the photodiode in the present embodiment comprises a base body 11 (corresponding to the “first zone” in the present invention) made of a p-type silicon (Si) semiconductor and a plurality of (16 in the present example) n-type semiconductor zones 121 arranged in the form of mutually separated points within a light receiving area 10 on the surface (in Fig. 1A on the upper surface) of the base body 11. Each n-type semiconductor region 121 is a very small n-type semiconductor diffusion layer with a rectangular shape when viewed from above. The n-type semiconductor regions 121 are formed by selective diffusion of an impurity, such as phosphorus, or by epitaxial growth.

[0055] In the present embodiment, an n-type semiconductor region 121 has a square shape with a size d×d when viewed from above. The gap between the adjacent n-type semiconductor regions 121 is L. For example, the length d of one side of the n-type semiconductor regions 121 is 1 µm, and the gap L of the n-type semiconductor regions 121 is 5-20 times as large as d, that is, approximately 5-20 µm. Regarding the area within the light-receiving surface 10, an n-type semiconductor region 121 with an area of ​​approximately 1 µm 2 25 - 400 µm each 2 of the light-receiving surface. That is, the percentage of the sum of the areas of all n-type semiconductor regions 121 arranged in the form of dots within the light-receiving surface 10 is very small, amounting to no more than 5% of the area of ​​the light-receiving surface 10.

[0056] The n-type semiconductor regions 121 are regularly arranged at the intersection points of a rectangular grid when viewed from above. For each n-type conductor region 121, a contact part 14, made of metal or another type of conductor, is formed in contact with the n-type conductor region. As shown in Fig. 2, an insulating film 16 made of SiO2 (or other materials) is formed so as to cover the surface of the base body 11 (which is the p-type semiconductor) and the n-type semiconductor region 121. A portion of the insulating film 16 over the n-type semiconductor region 121 is removed by etching (or other methods) to form a contact hole (or via hole) 17. The contact hole 17 (which is in the Fig. 2 and Fig. 3 has a square shape s×s) is slightly smaller in size than the n-type semiconductor region 121. Through this contact hole 17, the contact part 14 is formed in contact with the n-type semiconductor region 121. In the present embodiment, the contact part 14 has a rectangular shape, the size of which is approximately the same as or slightly larger than the boundary line of the underlying n-type semiconductor region 121. In other words, the size of a contact part 14 is also approximately d×d.

[0057] When the n-type semiconductor region 121 is formed by impurity diffusion, the diffusion region is often extended slightly beyond the intended boundary line during the manufacturing process. Fig. In Figure 2, such an n-type semiconductor region formed by a slightly expanded diffusion is designated by reference numeral 121a. This n-type semiconductor region 121a has a size of D×D (where D > d). Thus, the position of the boundary line of the n-type semiconductor region varies to some extent. Therefore, if all contact parts 14 are formed in the same size, a situation may arise in which one contact part 14 covers the entire n-type semiconductor region 121, as shown in Figure 2. Fig. 3B, while another contact part 14 leaves the peripheral edge of the n-type semiconductor region 121 partially uncovered, as shown in Fig. 3A is shown.

[0058] The contact parts 14, which are provided for all n-type semiconductor regions 121 within the light-receiving surface 10, are connected to each other by a wiring part 15 made of a metal or other type of conductor, which is formed simultaneously with the contact parts 14. This wiring part 15 serves as the cathode terminal (C) of the photodiode. On the other hand, the common contact part 13, which is formed in contact with the base body 11, serves as the anode terminal (A) of the photodiode. As shown in the Fig. 3A and Fig. 3B, the wiring part 15 in the present embodiment has a width equal to the length of one side of the contact part 14, although this is not essential.

[0059] The photodiode of the present embodiment having such a configuration can be regarded as equivalent to a circuit in which the same number of smallest photodiodes are connected in parallel as the n-type semiconductor regions 121, as shown in Fig. 1C is shown.

[0060] The junction capacitance due to the n-type semiconductor region 121 depends on the contact area between the n-type semiconductor region 121 (which has an approximately flat rectangular shape) and the surrounding p-type semiconductor base body 11. Since the depth of the n-type semiconductor region 121 is almost uniform, the junction capacitance can be calculated as a function of the area and the circumferential length of the n-type semiconductor region 121, provided that the ambient temperature, reverse bias, and other relevant conditions are the same. Thus, the junction capacitance C in this case is expressed as: C∝A×[Area of ​​the n-type semiconductor region 121]+B×[Circumferential length of the n-type semiconductor region 121] where A and B are specific constants.

[0061] The values ​​of the constants A and B vary depending on the impurity concentration distribution, the diffusion depth, and other values ​​related to the diffusion layer. For the case of the base body 11, which is made of a p-type semiconductor, in which the impurities have a uniform concentration of approximately 1 × 10 14 atoms / cm 3 For example, if the components are diffused to a depth of approximately 2 µm, the ratio of A to B is approximately 0.5 to 2. For the case of a point-like diffusion zone whose circumferential length value is larger than its area value, the percentage of component capacitance increases proportionally to the circumferential length in the junction capacitance with an increase in the number of diffusion zones.

[0062] The junction capacitance of the photodiode in the present embodiment is equal to the sum of the junction capacitances of the large number of very small photodiodes connected in parallel. Accordingly, by increasing the gap L between adjacent n-type semiconductor regions 121 so as to reduce the number of n-type semiconductor regions 121 included in the light-receiving surface 10 while reducing the length d of one side of each n-type semiconductor region 121, it is possible to make the sum of the junction capacitances smaller than the junction capacitance of the conventional photodiode. Therefore, in the present example, the length d of one side of the n-type semiconductor region 121 is set to 1 µm, and the gap is set to 5-20 µm. These values ​​result in a smaller junction capacitance than that of the conventional photodiode.As a result, peaking of gain on the frequency characteristics of an amplifier connected to the output side of the photodiode according to the present embodiment is prevented, and the high-frequency noise in the amplifier is thereby reduced, so that the bandwidth can be widened.

[0063] Compared with the conventional photodiode, the photodiode in the present embodiment has a smaller area of ​​the n-type semiconductor regions 121 for the same area of ​​the light-receiving surface 10, resulting in a smaller area of ​​the depletion layer regions formed around the n-type semiconductor regions 121. Although carriers can be generated outside the depletion layer regions within the p-type semiconductor regions constituting the main body 11, the photoelectric conversion efficiency in this region is lower than in the depletion layer spreading region. Furthermore, the wiring part 15 formed on the p-type semiconductor region blocks a small but certain amount of incident light, thereby reducing the amount of light that can contribute to photoelectric conversion. Due to these factors, the photodetection signal will inevitably be smaller than that of the conventional photodiode.However, in the photodiode of the present embodiment, since the degree of reduction of the high-frequency noise achieved by reducing the junction capacitance is larger than the degree of reduction of the photodetection signal, the SN ratio of the photodetection signal will be larger than in the conventional case.

[0064] In commonly used photodiodes, the pn junction plane, where the n-type semiconductor region is in contact with the p-type semiconductor region, is formed almost horizontally within the base body. The light entering the base body reaches this pn junction plane and generates charge carriers. In comparison, in the photodiode of the present embodiment, as can be seen from Fig. 2, the pn junction plane is hardly affected, since the area above the pn junction plane within the main body 11 is almost completely covered by the contact part 14. Therefore, it can be deduced that the charge carriers contributing to the photocurrent in the photodiode of the present embodiment are mostly generated within the main body 11 made of the p-type semiconductor, and hardly within the n-type semiconductor region 121 or at the pn junction plane. Based on this fact, it is possible to consider that the very small n-type semiconductor regions 121, which are discretely arranged in the form of dots within the light-receiving surface 10 in the photodiode of the present embodiment, function specifically as a collector of the charge carriers generated in the surrounding regions (the depletion layer spreading region and its surrounding regions) rather than as a charge carrier generator.

[0065] In the photodiode of the present embodiment, the area above the n-type semiconductor region 121 is almost completely covered by the contact part 14. This has the following effect: The contact part 14, which is made of a metal or other type of non-transparent conductor, functions to block light. Therefore, within the light-receiving surface 10, light hardly falls on the n-type semiconductor regions 121; only the base body 11 receives light. Therefore, when light falls on the light-receiving surface 10, charge carriers are hardly generated in the n-type semiconductor regions 121; most of the charge carriers reflected in the photocurrent are generated within the base body 11.

[0066] As is well known, light of a longer wavelength penetrates into a deeper region of the bulk body 11 and generates charge carriers in this region. Therefore, for example, light of a shorter wavelength, such as ultraviolet (near-ultraviolet) light or visible light near the ultraviolet range, generates charge carriers within a shallow region near the surface of the bulk body 11. Although some of these charge carriers move to the n-type semiconductor regions 121 due to free diffusion, most of them will eventually reach the surface of the bulk body 11 (the interface with the insulating film 16) and be lost.Therefore, the probability that the carriers generated in response to the shorter wavelength light reach the n-type semiconductor regions 121 and are reflected in the photocurrent is drastically lower than the probability that the carriers generated in response to the longer wavelength light reach the n-type semiconductor regions 121 and are reflected in the photocurrent. As a result, the photodetection sensitivity of the photodetector of the present embodiment to the shorter wavelength light is drastically lower than its photodetection sensitivity to the longer wavelength light. Thus, an effect practically the same as removing shorter wavelength light with an optical filter can be achieved.In general, the photodetection sensitivity for light with wavelengths of 400-450 nm or shorter can be drastically reduced depending on the impurity concentration, the DC reverse bias, and other factors. The photodiode structure of the present embodiment also reduces the voltage dependence of the junction capacitance.

[0067] Thus, in the photodiode in the present embodiment, the junction capacitance-dependent noise level can be reduced without significantly lowering the photodetection sensitivity, compared to the conventional and commonly used photodiode in which the n-type semiconductor region is formed over the entire light-receiving surface 10. The photodetection sensitivity to longer wavelengths of light can be relatively improved by lowering the photodetection sensitivity to shorter wavelengths of light. These features make the photodiode in the present embodiment particularly useful in the case where the target signal generated by visible light or near-infrared light needs to be detected with high sensitivity in the presence of a large amount of ultraviolet light as a noise component. [Variant of the first embodiment]

[0068] In the photodiode of the first embodiment, a number of n-type semiconductor regions 121 having a square shape when viewed from above are provided within a light-receiving surface 10. The shape of each n-type semiconductor region 121 is not limited to a square when viewed from above. For example, it may be an elongated rectangular shape, a circular shape, or a polygonal shape. However, as understood from equation (1), decreasing the area of ​​the n-type semiconductor region does not always decrease the junction capacitance if the peripheral length of the region increases. Therefore, it is essential to find a balance between the area and the peripheral length of each n-type semiconductor region so that the junction capacitance is reduced. The number of n-type semiconductor regions arranged within the light-receiving surface 10, that is, the distribution density, must also be controlled in the same way.

[0069] Fig. 4 is a variant in which the n-type semiconductor regions 124 having a substantially square shape, similar to the first embodiment, are arranged in such a way that the arrangement position of the n-type semiconductor regions 124 in every other column is shifted by approximately L / 2. Shifting the arrangement of the n-type semiconductor regions 124, as shown in Fig. 4 has the following advantage.

[0070] In the case of the n-type semiconductor regions 121, which, as shown in Fig. 1B, the point located in the middle of the four n-type semiconductor regions 121 that are adjacent to each other in the longitudinal and lateral directions is the point furthest from the four n-type semiconductor regions 121. The distance to this point is approximately 0.7 L. In comparison, for the case of the n-type semiconductor regions 124 that, as shown in Fig. 4, the gap (lateral distance) of the columns may be greater than L under the condition that an n-type semiconductor region 124 is located at a distance from any point within the area surrounded by the three adjacent n-type semiconductor regions 124 that is equal to or less than 0.7 L. Compared with the gap shown in Fig. Therefore, compared to the arrangement shown in FIG. 1B, the present arrangement requires a smaller number of n-type semiconductor regions to cover the light-receiving surface 10 so that carriers can be collected from the entire light-receiving surface 10. Consequently, the junction capacitance is further reduced. The present arrangement also allows the spacing of the wiring members 15 to be increased, thus reducing the percentage of light blocked by the wiring member 15 and thereby increasing the photodetection signal.

[0071] In the photodiode of the first embodiment, the contact part 14 covers almost the entire area above the very small n-type semiconductor region 121 created by diffusion. It is also possible to reduce the width of the wiring part 15 so as to cover only a portion of the n-type semiconductor region 121, although this structure reduces the effect of lowering the photodetection sensitivity to shorter wavelengths of light.

[0072] In the photodiode of the first embodiment, the p-type semiconductor region, which occupies the largest portion of the light-receiving surface 10, has a substantially uniform impurity concentration. However, based on the fact that the potential gradient in the p-type semiconductor region depends on the impurity concentration gradient, an impurity concentration gradient can be created within the p-type semiconductor region to efficiently transfer the charge carriers generated in the p-type semiconductor region to the n-type semiconductor region 121. For this purpose, for example, as shown in Fig. 5, a highly concentrated p-type diffusion region 20 in the shape of a dot, created by diffusing a highly concentrated p-type impurity, may be provided within a region outside the depletion layer regions between the adjacent n-type semiconductor regions 121 and at approximately equal distances from the surrounding n-type semiconductor regions 121. This creates an impurity concentration gradient from a region near this highly concentrated p-type diffusion region 20 to the n-type semiconductor region 121, forming a slight potential gradient. This promotes the transfer of the charge carriers generated in the p-type bulk device 11, thereby improving photosensitivity and operating speed.

[0073] The highly concentrated p-type diffusion zones 20 do not have to be shaped like dots, but may have any other shape, such as a linear shape. Fig. 7 shows an example in which the highly concentrated p-type diffusion regions 20 are shaped like a substantially hexagonal wire (honeycomb shape) when viewed from above, with the n-type semiconductor regions 121 located near the center of each highly concentrated p-type diffusion region 20. This configuration creates an impurity concentration gradient surrounding each n-type semiconductor region 121 and sloping from all directions toward the n-type semiconductor region 121. Consequently, a slight potential gradient is formed for each n-type semiconductor region 121 from the entire area around this region 121 toward the same region, so that the charge carriers are collected more efficiently and quickly in the n-type semiconductor region 121.

[0074] In the examples in the Fig. 5 and Fig. 7, the potential gradient is formed to promote charge carrier transfer in the horizontal (lateral) direction. It is also possible to form a potential gradient in the vertical (depth) direction. Fig. 6 is a schematic sectional view showing such an example. In this example, the upper surface of the base body 11, which is made of a p-type semiconductor with a relatively high impurity concentration, is covered with a low-concentration p-type layer 11a having the same conductivity type (in this case, p-type) but a relatively low impurity concentration. The n-type semiconductor regions 121 are formed in the near-surface region of this low-concentration p-type layer 11a. In this configuration, the impurity concentration gradient is formed upward from the base body 11, that is, toward the n-type semiconductor regions 121. Accordingly, when a reverse bias voltage is applied from the base body 11, a slight potential gradient is formed to promote the upward transfer of charge carriers.

[0075] Of course, it is possible to use the horizontal impurity concentration gradient and the vertical impurity concentration gradient together.

[0076] In the first embodiment and its variants described so far, not only the charge carriers generated within the depletion layer regions around the n-type semiconductor regions 121 but also the charge carriers generated outside these regions and transferred to the n-type semiconductor regions 121 by diffusion are used as the photocurrent. However, for an application that particularly requires a high-speed response, it may be preferable to use only the charge carriers generated within the depletion layer regions as the photocurrent. For this purpose, the spacing of the adjacent n-type semiconductor regions 121 should be determined such that the depletion layer regions around the adjacent n-type semiconductor regions 121 maximally overlap each other with their edge portions.However, as explained above, in order to reduce the sum of the junction capacitances, the gap between the n-type semiconductor regions 121 should preferably be widened to reduce the distribution density of the n-type semiconductor regions 121. In order to arrange the n-type semiconductor regions 121 without a gap between the adjacent depletion layer regions while suppressing the sum of the junction capacitances to the lowest possible level, an arrangement such as that shown in FIG. Fig. 8, in which the n-type semiconductor regions 121 are arranged in a honeycomb pattern, each region being given an edge region with an approximately hexagonal shape of a predetermined size when viewed from above. The region indicated by the reference numerals 121b in Fig. The regions indicated in Figure 8 are depletion layer regions formed around the n-type semiconductor regions 121. Such an arrangement of the n-type semiconductor regions 121 makes it possible to eliminate the gap between the adjacent depletion layer regions 121b while maximally avoiding unnecessary overlap of the depletion layer regions 121b.

[0077] Under the condition that the total number of n-type semiconductor regions 121 arranged in the light-receiving surface 10 is the same, if the n-type semiconductor regions 121 are arranged in the previously described manner to eliminate the gap between the depletion layer regions 121b, the light-receiving surface 10 will naturally become relatively small. However, in the case, for example, where the incident light reaches a large area, the light-receiving surface 10 should preferably have a large area to efficiently receive the incident light. Even in the case where the incident light only reaches a small area, if the position of this area can change greatly within a certain range, the light-receiving surface 10 should preferably also have a large area to reliably receive the incident light. Accordingly, a configuration such as that shown in FIG. Fig. 9 can be applied to achieve both a large size of the light receiving area 10 and a high-speed response.

[0078] Instead of narrowing the spacing of the n-type semiconductor regions to eliminate the gap between the depletion layer regions, in this photodiode, a light-blocking member 50 is provided to prevent incident light from reaching the base body (p-type semiconductor region) 11 over almost the entire area outside the depletion layer regions 121b surrounding the n-type semiconductor regions 121, which are spaced apart to some extent, so as to prevent the overlap of their depletion layer regions. For example, the light-blocking member 50 may be formed using a metal (or a similar material) used for the wiring part. Although the light-receiving area 10 is large, carriers are hardly generated outside the depletion layer regions 121b, that is,within the region where carriers are transferred by diffusion, since the light reaching the area other than the depletion layer region 121b around each n-type semiconductor region 121 is blocked by the light-blocking member 50. Therefore, most of the carriers generated by the incident light are quickly transferred to the n-type semiconductor regions 121 by drift and reflected in the photocurrent. Thus, high-speed response is achieved. Meanwhile, since the area of ​​the light-receiving surface 10 itself is large (although the regions under the light-blocking member 50 do not effectively contribute to light detection), it can efficiently receive incident light reaching a large area, as well as detect incident light hitting any portion of the large light-receiving surface 10.If the light-receiving surface 10 has a small area, it may be necessary to precisely adjust the optical axis so that the incident light reliably hits the light-receiving surface. Using a light-receiving surface 10 with a large area, as in the present example, is advantageous in that incident light can be reliably detected even if the optical axis has not been adjusted with high precision (in some cases, even if no optical axis adjustment is performed).

[0079] Compared to the conventional photodiode, the photodiode of the first embodiment inevitably requires a larger number of wiring parts 15 to be provided within the light-receiving surface 10. If external noise enters the wiring parts 15, the noise will appear in the photocurrent. To reduce the amount of external noise entering the wiring parts 15, a shielding wiring part may be additionally provided. Fig. 10A, Fig. 10B and Fig. 11 are sectional views showing examples of the structure provided with such a shielding wiring member.

[0080] In the example from the Fig. 10A and Fig. In FIG. 10B, the wiring part 15 connecting a plurality of n-type semiconductor regions 121 is formed on an insulating film (first insulating film) 16 formed on the surface of the base body 11. Additionally, a second insulating film 18 is formed so as to cover the insulating film 16 and the wiring part 15. A shielding wiring part 19 is formed on this second insulating film 18 and only within the area above the wiring part 15 and the contact part 14. The shielding wiring part 19 extends to the outside of the light-receiving surface 10 and is connected, for example, to a feeder line having a fixed electric potential (e.g., ground potential). This shielding wiring part 19 intercepts most of the noise that would enter the wiring part 15 if the shielding wiring part 19 were not present. Consequently, the amount of noise entering the wiring part 15 in the lower layer is reduced.

[0081] In the example in Fig. 11, a lower-layer shielding wiring part 19a is formed on an insulating film (first insulating film) 16 formed on the surface of the base body 11, and only within the area below the wiring part 15 (described later). A second insulating film 18a is formed to cover the lower-layer shielding wiring part 19a. The wiring part 15 connecting a plurality of n-type semiconductor regions 121 is formed on this second insulating film 18a. A third insulating film 18b is provided to cover the second insulating film 18a and the wiring part 15. An upper-layer shielding wiring part 19b is formed on this third insulating film 18b and only within the area above the wiring part 15 and the contact part 14.The lower layer shielding wiring part 19a and the upper layer shielding wiring part 19b are connected to each other by a plurality of vias 19c which vertically penetrate the second and third insulating films 18a and 18b and which are regularly arranged along the extending direction of the wiring part 15 on both sides of this part.

[0082] The vias 19c serve to ensure electrical conduction between the lower-layer shielding wiring part 19a and the upper-layer shielding wiring part 19b; they are not electrically connected to the wiring part 15. Either the lower-layer shielding wiring part 19a or the upper-layer shielding wiring part 19b extends to the outside of the light-receiving surface 10 and is connected, for example, to a feeder line with a fixed electrical potential (e.g., ground potential). Accordingly, both the lower-layer shielding wiring part 19a and the upper-layer shielding wiring part 19b, which are connected by the vias 19c, as well as the vias 19 themselves, are at the same electrical potential.In the present embodiment, the wiring part 15 is positioned vertically between the lower-layer shielding wiring part 19a and the upper-layer shielding wiring part 19b, and is also surrounded by the vias 19c on its lateral sides. Such a configuration further reduces the amount of external noise entering the wiring part 15 compared to that shown in FIGS. Fig. 10A and Fig. 10B shown configuration.

[0083] In the Fig. In the example shown in Fig. 11, the lower-layer shielding wiring part 19a is provided below the wiring part 15. However, the shielding effect hardly changes if this lower-layer shielding wiring part 19a is not provided, because the base body 11 is at the ground potential in many cases, and therefore the electric potential in the area below the wiring part 15 is practically at a fixed potential even if the lower-layer shielding wiring part 19a is not provided.In the case where the lower layer shielding wiring part 19a is omitted, the vias 19c may be formed in the form of columnar parts with only their upper ends connected to the upper layer shielding wiring part 19b, or the lower ends of the vias 19c may be connected to the base body 11 (in this case, a highly concentrated p-type region should preferably be formed at the contact portions on the base body 11), so that the base body 11, the vias 19c, and the upper layer shielding wiring part 19b will be at the ground potential.

[0084] Instead of the columnar vias 19c surrounding the wiring part 15 on both sides, side-shielding wiring parts may be provided extending substantially parallel to the wiring part 15 so as to protect the same wiring part 15 from both sides, the side-shielding wiring parts being set to the same electric potential as the lower-layer shielding wiring part 19a and the upper-layer shielding wiring part 19b.

[0085] The first embodiment is an example in which the photoelectric conversion element according to the present invention is applied to a silicon photodiode. The present invention can be applied to various elements that perform photoelectric conversion. [Second embodiment]

[0086] Fig. 12A is a schematic sectional view of a conventional and commonly used phototransistor, and Fig. 12B is a schematic sectional view of a phototransistor as an embodiment of the present invention. In the present example, the base body 21, made of a p-type semiconductor, is the collector region, the n-type semiconductor region 22 (or 221) created by diffusion on the surface of the base body 11 is the base region, the p-type semiconductor region 23 created by diffusion in the n-type semiconductor region 22 (or 221) is the emitter region, and the junction between the collector region and the base region is the photoelectric conversion region. The contact part 24 formed in contact with the base body 21 serves as the collector terminal (C), the contact part 26 (or 261) formed in contact with the n-type semiconductor region 22 (or 221) serves as the base terminal (B), and the contact part 25 formed in contact with the p-type semiconductor region 23 serves as the emitter terminal (E).

[0087] As this is Fig. 12B, the phototransistor of the present embodiment has a plurality of very small n-type semiconductor regions 221, wherein the contact parts 261 are formed in contact with the respective n-type semiconductor regions 221 and are connected to each other by the wiring part 27. This structure is similar to Fig. 1C, equivalent to a multi-junction circuit of the parallel-connected collector and base regions. Similar to the photodiode in the first embodiment, the present structure allows the size and spacing of the n-type semiconductor regions 221 to be appropriately determined, thus reducing the junction capacitance of the pn junctions to a lower level than in the conventional case and thereby reducing the noise level without significantly degrading the photodetection sensitivity. Furthermore, the photodetection sensitivity to shorter wavelengths of light can be drastically reduced by providing the contact part connected to the n-type semiconductor region 221 serving as the base region in such a manner that the entire area above the n-type semiconductor region 221 is covered by the contact part.

[0088] In the phototransistor of the second embodiment, the base body 21 itself serves as the collector region. In the case where the phototransistor is used as one of the elements in an integrated circuit, the configuration as shown in Fig. 13, where a well 31, which is realized from a p-type semiconductor created in the base body 30, serves as the collector region and several n-type semiconductor regions 221 are formed within this well 31 in a similar manner as in Fig. 12B are provided. [Third Embodiment]

[0089] Fig. 14 is a schematic sectional view of a Darlington phototransistor as an embodiment of the present invention. In the present Darlington phototransistor, the contact part 25, which is formed in contact with the p-type semiconductor region 23 serving as the emitter region of the first transistor, is connected by a wiring part 45 to the contact part 43, which is formed in contact with the n-type semiconductor region 41 serving as the base region of the second transistor. A p-type semiconductor region 42, which serves as the emitter region of the second transistor, is provided in this n-type semiconductor region 41. The contact part 44, which is formed in contact with this p-type semiconductor region 42, serves as the emitter terminal. The other basic structures are substantially the same as those shown in Fig. Phototransistor shown in Figure 12B.

[0090] Of course, the photoelectric conversion element according to the present invention can also be used in devices such as a phototransistor integrated into an optocoupler. Some optocouplers have a triac output or a thyristor output for controlling an AC load or for other purposes. It should be appreciated that the present invention can also be applied to a phototriac or a photothyristor used in such an element. LIST OF REFERENCE SYMBOLS 10 Light receiving surface 11, 21, 30 basic body 121, 121a, 122, 123, 124, 221, 41 n-type semiconductor region 13, 14, 24, 25, 261, 43, 44 contact part 15, 27, 45 wiring part 16, 18, 18a, 18b Insulating film 17 contact hole 19, 19a, 19b shielding wiring part 19c via 20 highly concentrated p-type diffusion zone 23, 42 p-type semiconductor region 31 tub 50 light-blocking part

Claims

[1] A photoelectric conversion element comprising: a) a first zone of a first conduction type, the first zone being a semiconductor base body (11, 21, 30) itself; b) a dot-shaped second region of a second conductivity type different from the first conductivity type, the dot-shaped second region being formed by diffusing an impurity or by growing an epitaxial layer on a surface of the first region within a light-receiving region; c) a light-blocking member (50) for blocking light incident on the point-shaped second zone from light incident on the light-receiving area on an entrance surface receiving light from outside; and d) a wiring part (15, 27, 45) having a first branch and a second branch arranged parallel to each other within the light receiving area, wherein the point-shaped second zone generates a photodetection signal by collecting charge carriers from a surrounding area which are generated in the first zone by the light incident on the first zone in the light receiving area, wherein a plurality of the point-shaped second zones are arranged in a mutually separated form within the light receiving area to obtain a single photodetection signal, wherein the plurality of second zones are interconnected so as to sum photodetection signals generated by the individual point-shaped second zones, and wherein the wiring part (15, 27, 45) connects the plurality of point-shaped second zones via the first branch and the second branch, and wherein the charge carriers collected in the plurality of point-shaped second zones flow externally as photocurrent. [2] A photoelectric conversion element comprising: a) a first zone of a first conductivity type, the first zone being either a semiconductor base body (11, 21, 30) itself or a zone formed in the semiconductor base body (11, 21, 30) and different from the semiconductor base body (11, 21, 30); and b) a plurality of dot-shaped second zones of a second conductivity type different from the first conductivity type, arranged in a separate form within a light-receiving region to obtain a single photodetection signal, wherein the plurality of dot-shaped second zones are interconnected so as to sum photodetection signals generated by the individual dot-shaped second zones, wherein the dot-shaped second zones are formed by diffusing an impurity or by growing an epitaxial layer on a surface of the first zone within the light-receiving region, wherein the respective point-shaped second zone generates a photodetection signal by collecting charge carriers from a surrounding area which are generated in the first zone by the light incident on the light receiving area, and wherein the respective point-shaped second zone is a zone whose area is equal to or less than 1% of the total light receiving area. [3] A photoelectric conversion element according to claim 1, further comprising: a plurality of contact parts (13, 14, 24, 25, 261, 43, 44) provided respectively above the plurality of point-shaped second zones, each contact part (13, 14, 24, 25, 261, 43, 44) being electrically connected to the point-shaped second zone located thereunder, wherein the wiring part (15, 27, 45) is an electrical conductor for connecting the plurality of contact parts (13, 14, 24, 25, 261, 43, 44) to each other, and wherein all point-shaped second zones within the light receiving area are located either directly below the contact parts (13, 14, 24, 25, 261, 43, 44) and / or the wiring part (15, 27, 45), so that the contact parts (13, 14, 24, 25, 261, 43, 44) and / or the wiring part (15, 27, 45) form the light-blocking part (50). [4] A photoelectric conversion element according to claim 3, wherein the wiring part (15, 27, 45) located above a depletion layer spreading region formed in the first region surrounding the dot-shaped second region has a smaller line width than the contact part (13, 14, 24, 25, 261, 43, 44) and the wiring part (15, 27, 45) located above the dot-shaped second region. [5] A photoelectric conversion element according to claim 3, further comprising: an insulating film (16, 18, 18a, 18b) formed to cover the contact part (13, 14, 24, 25, 261, 43, 44) and the wiring part (15, 27, 45); and a conductor part provided on the insulating film (16, 18, 18a, 18b) above the contact part (13, 14, 24, 25, 261, 43, 44) and the wiring part (15, 27, 45), wherein a predetermined fixed electrical potential is applied to the conductor member to configure the conductor member to function as an electrical shield. [6] The photoelectric conversion element according to claim 5, wherein a plurality of conductive columnar members extending from the conductor member located above the wiring member (15, 27, 45) into the insulating film (16, 18, 18a, 18b) positioned between the conductor member and the wiring member (16, 18, 18a, 18b) are formed in such a manner as to surround lateral sides of the wiring member (15, 27, 45), or wherein conductive linear members extending substantially parallel to the wiring member (15, 27, 45) are formed in such a manner as to protect the wiring member (15, 27, 45) from both sides, and wherein the columnar members or the linear members are configured to function as an electrical shield together with the conductor member. [7] A photoelectric conversion element according to claim 1, wherein the plurality of dot-shaped second regions are arranged separately from each other so as to avoid overlap of depletion layer spreading regions formed in the first region around the respective dot-shaped second regions, and wherein the light-blocking member (50) is provided over a portion or a whole of an area outside the depletion layer spreading region around each dot-shaped second region within the light receiving area. [8] The photoelectric conversion element according to claim 1 or 2, wherein a total area of ​​the plurality of dot-shaped second zones arranged within the light receiving area is equal to or smaller than 5% of an area of ​​the light receiving area. [9] A photoelectric conversion element according to claim 8, wherein the dot-shaped second region has a square shape and wherein a gap (L) between adjacent dot-shaped second regions is approximately 5 µm to 20 µm. [10] A photoelectric conversion element according to claim 1 or 2, wherein a number of the dot-shaped second regions and an area and a circumferential length of each dot-shaped second region are determined such that a sum of junction capacitances of junctions between the first region and the plurality of dot-shaped second regions is smaller than a junction capacitance observed at the junction between the first region and the dot-shaped second region in a case where the dot-shaped second region is formed over a complete light receiving area. [11] A photoelectric conversion element according to claim 1 or 2, wherein the plurality of dot-shaped second regions are formed in the form of dots arranged in rows or columns as viewed from above, and wherein a position of the dot-shaped second regions in every second row or column is shifted by a half of a pitch (L) of the dot-shaped second regions along an extending direction of the row or column. [12] A photoelectric conversion element according to claim 1 or 2, wherein the plurality of dot-shaped second regions are arranged such that depletion layer spreading regions formed around the respective dot-shaped second regions overlap each other without leaving any gap. [13] A photoelectric conversion element according to claim 1 or 2, wherein a highly concentrated region (20) having a predetermined shape when viewed from above is formed between the plurality of dot-shaped second regions discretely arranged in the first region, the highly concentrated region (20) having a higher concentration of the impurities of the first conductivity type than the surrounding first region. [14] A photoelectric conversion element according to claim 2, wherein the plurality of dot-shaped second regions are arranged separately from each other so as to avoid overlap of depletion layer spreading regions formed in the first region around the respective second dot-shaped regions, and wherein a light-blocking member (50) for blocking incident light is provided over a portion or a whole of an area outside the depletion layer spreading region around each dot-shaped second region within the light receiving area.

Images