JPH0463473A - Solid state image sensor - Google Patents

Abstract

PURPOSE:To obtain an image having high sensitivity and excellent color reproducibility by one solid state image sensor chip by separately providing photodiodes as photoelectric converters in a depth direction. CONSTITUTION:A plurality of vertical CCDs 20 are disposed in parallel, and a plurality of photoelectric converters 30 are arranged along the CCDs 20. The photoelectric converters 30 are formed of three photodiodes separated in a depth direction, the photodiode 31 of an uppermost layer is formed in a depth for absorbing blue, and a p<-> type layer 34 of a burr layer is formed thereunder. Similarly, a second photodiode 32 is formed in a depth for absorbing green light, and the photodiode 33 of an uppermost layer is formed in a depth for absorbing red light. In this case, signal charges stored in the respective photodiodes are calculated to obtain R, G and B components, and a color image can be imaged. Thus, an image having high sensitivity and excellent color reproducibility is obtained by one solid state image sensor chip.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a solid-state imaging device, and particularly to a solid-state imaging device for capturing color images.

(Prior Art) Conventionally, there are two main types of solid-state imaging devices that capture color images: The first method is to color-separate the incident light using an optical system, and image the three color-separated lights using three solid-state imaging device chips. The second is
This is a method in which a color separation filter is formed in each pixel of a solid-state image sensor chip, and color signals are obtained with one solid-state image sensor chip.

The first method is the method used in broadcast TV left cameras that require particularly high image quality, and usually uses red (R), green (G), and blue (B) colors that enter the lens. The light is divided into three primary colors, and three solid-state image sensor chips are provided corresponding to each of the three primary colors. This method has low loss of incident light and high sensitivity, and because it separates the light into three primary colors,
Excellent color reproducibility on TV monitors. However, since three optical spectroscopic systems and three solid-state image pickup chips are required each, there is a drawback that the left TV camera becomes large and expensive.

The second method has the advantage of being small and inexpensive for the TV left camera, since a color image can be obtained using one solid-state image sensor chip. However, the color separation filter formed on the solid-state image sensor chip is usually formed as a bandpass filter, bypass, or low-pass filter on each pixel, and therefore the efficiency of utilizing incident light is extremely poor and the sensitivity is low.

Note that in consideration of sensitivity, a complementary color filter configuration may be used, but in this case color reproducibility will be poor.

In addition to these methods, there is also a structure in which photoconductive materials with different sensitivities depending on the wavelength are laminated with a transparent insulating layer in between, for example, the top layer is a material that is sensitive only to blue colors and transmits green and red (Zn5e). A device has been proposed in which the next layer is a material (Cd5) that is sensitive to blue and green and transmits red light, and the bottom layer is a material that is sensitive to red (Cd5e).

However, this structure requires three layers of different materials to be laminated by epitaxial growth or the like, which is almost impossible to achieve with current technology.

(Problems to be Solved by the Invention) Conventionally, when an optical spectroscopic system is used to capture color images, the device configuration becomes larger and the cost increases.
Further, when a color separation filter is used, there is a problem that the sensitivity is lowered or the color reproducibility is lowered.

The present invention has been made in consideration of the above circumstances, and its purpose is to achieve high sensitivity and good color reproducibility with a single solid-state image sensor chip without using color separation filters or optical spectroscopy systems. An object of the present invention is to provide a solid-state imaging device that can obtain images.

[Structure of the Invention] (Means for Solving the Problems) The gist of the present invention is that blue, green, red, etc. light is absorbed in a semiconductor (particularly Si) of a photoelectric conversion element (photodiode, etc.). The purpose of this method is to take advantage of the fact that the regions in the depth direction are different, and to perform color separation by separating the conventionally used photodiodes in the depth direction.

That is, the present invention provides a solid-state imaging device in which a plurality of photoelectric conversion sections are provided on a semiconductor substrate and a signal charge readout section that reads out signal charges converted and accumulated in these photoelectric conversion sections, respectively. At least 3 in the depth direction
The signal charge readout section is formed of a readout transistor independently connected to each photodiode of the photoelectric conversion section.

Here, the arrangement of the photoelectric conversion units may be one-dimensional or two-dimensional (
matrix). Further, the read signal charges may be transferred by a CCD or taken out to metal wiring. In addition, when a photoelectric conversion section consisting of three photodiodes is arranged two-dimensionally and the read signal charges are transferred by a CCD, three horizontal CCDs are provided after the vertical CCD, and the signal charges are transferred from the vertical CCD to the horizontal CCD. When reading out, the signal charges of the photodiodes at the same depth position are transferred to the same horizontal CC.
Supply to D.

(Function) According to the present invention, since the regions in the depth direction in which blue, green, red, etc. Each component is detected. Therefore, by performing arithmetic processing based on the signal charges accumulated in each photodiode, the R, G, and B components can be determined, thereby making it possible to capture a color image.

(Example) First, before describing an example, the basic principle of the present invention will be explained.

FIG. 6 shows a cross-sectional structural diagram of a photoelectric conversion section according to the present invention.

This photoelectric conversion section consists of a plurality of photodiodes, and its material is Si, which is generally widely used in semiconductor devices. In the figure, reference numerals 4, 5, and 6 are barrier layers that separate the photodiodes in the depth direction, and the photodiodes 1 and 2 are separated by barrier layers.
゜3 is separated. 7 is a layer that prevents the surface from being depleted.

Light incident on the photoelectric conversion unit from the front side is absorbed by photodiodes 1 and 2°3 separated in the vertical direction and converted into an electrical signal. The light comes from the top layer photodiode 1
The photodiode 2 absorbs mainly blue light, which is short wavelength light (some of the green and red light is also absorbed), and the subsequent photodiode 2 absorbs the green light (some of the red light is also absorbed). Red light is absorbed by the photodiode 3 in the bottom layer. The intensity of the incident light is I. Bind, IOゞl BO+ I Go + I RQ
... Expressed as (1). Here, IBO,
Ico, Iyo. are each blue.

It represents the intensity of light in the green and red regions. As shown in Figure 7, the transmission of light through a substance is as follows. It can be expressed as 'e-at, (2). t is the depth from the surface, α is the absorption coefficient, and ■ is the intensity of light at the depth t. If the absorption coefficients of blue, green, and red are αB and α0°α7, respectively, the light intensity in the depth direction for each color is IBmlBoee−α”−(3)1(
, -1(, oee "t-(4)IR-IRo-e-
It is expressed as aRt-(5). Here, since the relationship is αB>αG>αR, as shown in FIG. 7, blue and green.

The regions in the depth direction where red light and other light are absorbed in Si are different.

Now, in FIG. 6, if the depth from the surface to the barrier 4 is T1, the depth from the surface to the barrier 5 is T2, and the depth from the surface to the barrier 6 is T3, as shown in FIG. , photodiode 1 almost absorbs blue light,
Up to photodiode 2 absorbs green light, and up to photodiode 3 absorbs red light. If the quantum efficiency of each photodiode is assumed to be 1 for simplicity, then for photodiode 1, −αGT −
αRT 11- IBO+IGO(1e li+Lo(le
(6) For photodiode 2, 12 = Ico (e'-aGT'-αGT
2) + IRQ (e-αRT1-ee αRT2) For photodiode 3, T3 = IRQ (e ′-aRT2” T3)
-(8) e light is absorbed and a charge corresponding to the light is generated.

Therefore, from these three equations, ... (10) represents the light intensity for each of the incident blue, green, and red components. As a result, by processing the electrical signals obtained from each photodiode 1, 2.3 in an external circuit, blue (
R), green (G), and red (B) signals can be obtained.

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a plan view showing one pixel configuration of a solid-state imaging device according to an embodiment of the present invention, and FIGS. 2(a) to (c) are arrow views AA, BB, and C in FIG. 1. -C sectional view, and FIG. 3 is a sectional view taken along arrow DD in FIG.

As shown in FIG. 1, a photoelectric conversion section 30 is arranged adjacent to the vertical CCD 20. Although this figure shows one pixel portion, a plurality of vertical CCDs 20 are arranged in parallel, and a plurality of photoelectric conversion units 30 are arranged along each vertical CCD 20.

As shown in FIG. 2, the photoelectric conversion section 30 has a depth of 3
It is formed from separate photodiodes. The photodiode 31 in the uppermost layer is formed to a depth that absorbs blue light, and a p- layer 34 as a barrier layer is formed below it. The second photodiode 32 is formed to a depth capable of absorbing green light, and a p- layer 34, which is also a barrier layer, is formed below it. Furthermore, the photodiode 33 in the bottom layer is formed to a depth that allows it to absorb red light.

Here, the photodiode is formed by a junction of an n-layer and a p-layer, and the n-layer essentially functions as a storage diode. And these photodiodes 31, 3
An n-layer (CCD channel) 12 is formed adjacent to the photodiodes 31, 32, and 33, and a portion of each photodiode 31, 32, and 33 is in contact with a channel (p layer) of a readout transistor, which will be described later, at different positions.

Note that in order to realize the above configuration, a p-well 11 is formed on an n-type Si substrate 10, and a p+ element isolation layer 1B and an n-CCD channel 1 are formed by ion implantation or the like into this p-well 11.
2. N-photodiodes 31, 32, 33 and p-barrier layer 34 may be formed. Furthermore, instead of ion implantation, each of the layers 31 to 34 may be formed by epitaxial growth.

As shown in FIG. 3, the vertical CCD 20 has transfer electrodes formed on the CCD channel 12 via a gate oxide film 14, with nine transfer electrodes 21, 22, 23 per pixel.
I can see that.

Three electrodes are made to correspond to each of the three photodiodes 31, 32 and 33. That is, the photodiode 31 that absorbs blue light has transfer electrodes 211 and 221.
．． 23□ corresponds to the transfer electrode 221, which extends to above the photodiode 31 and also serves as the gate of the readout transistor. Similarly, the photodiode 32 that absorbs green light has a transfer electrode 21□.

222.23□ corresponds to 213.223 to the photodiode 33 that absorbs red light.

233 is compatible. And these transfer electrodes 21
, 22.23 are three-phase clocks φ1゜φ2. φ3 is applied, and the signal charges of the photodiodes 31, 32, and 33 are simultaneously read out to the vertical CCD 20.

FIG. 4 shows the overall configuration of the device of this embodiment. Vertical CCD2
The signal charge transferred by 0 is transferred to three horizontal CCDs 41
, 42.43. Here, the signal charge from the photodiode 31 transferred by the vertical CCD 20 is read out to the horizontal CCD 41, the signal charge from the photodiode 32 is read out to the horizontal CCD 42, and the signal charge from the photodiode 33 is read out to the horizontal CCD 43. Ru. The signal charges transferred by the horizontal CCDs 41, 42 and 43 are amplified by amplifiers and then sent to signal processing circuits 51, 52.
53, and further supplied to the arithmetic processing circuit 60. Here, the signal processing circuits 51, 52, and 53 perform noise reduction processing, sample hold, and the like. The arithmetic processing circuit 60 performs the arithmetic operations shown in equations (9) to (11) above, and calculates R, G.

It outputs a signal corresponding to B.

In such a configuration, when the photoelectric conversion unit 30 is irradiated with light, the photodiode 31 absorbs the blue light and a portion of the green and red light, and a signal charge is accumulated according to the amount of absorption. Ru. Similarly, the photodiode 32 absorbs green light and a portion of red light, and accumulates signal charges corresponding to the amount of absorption. Further, the photodiode 33 absorbs red light, and accumulates signal charges corresponding to the amount of red light absorbed. Each photodiode 31, 32.3
The signal charge accumulated in 3 turns on the readout transistor.
By doing so, the signals are simultaneously read out to the vertical CCD 20.

Here, the signal readout method is not limited to simultaneous readout, but may be such that the signal charges of each photodiode 31, 32, 33 are sequentially read out.

However, when one readout transistor is turned on,
There is a possibility that other transistors are turned on by the voltage applied to the read gate, and signal charges that should not be read are read. Therefore, when reading signal charges sequentially, the readout transistors are turned on from the upper photodiode to the lower photodiode. In the read transistor, the lower transistor requires a higher voltage to turn on than the upper transistor. Therefore, when the upper transistor is turned on, the lower transistor will not be turned on. When the lower transistor is turned on, the signal charge in the upper photodiode has already been read out and is empty, so the upper transistor is turned on.
However, no equivalence problem arises.

Now, the signal charge transferred by the vertical CCD 20 is 3
Each signal charge read out to the horizontal CCD 41, 42.43 of the book or read out and transferred from the photodiode 31°32.33 at this time is read out to the horizontal CCD 41, 42.43, respectively. Therefore, the horizontal CCD 41 receives the signal charge expressed by the above equation (6), the horizontal CCD 42 receives the signal charge expressed by the above equation (7), and the horizontal CCD 43 receives the signal charge expressed by the above equation (8). The charge will be read out. The signal charges transferred by the horizontal CCD 41.42.43 are supplied to the arithmetic processing circuit 60 via the signal processing circuits 51, 52.53. Then, by performing the calculations shown in equations (9), (10), and (11) in the calculation processing circuit 60, signals corresponding to R, G, and B are output.

Thus, according to this embodiment, the photoelectric conversion section 30 is formed by the photodiodes 31, 32, and 33 divided into three parts in the depth direction, and each signal charge accumulated in these photodiodes 31, 32, and 33 is By performing the above-mentioned arithmetic processing based on this, the magnitudes of the R, G, and B components of the input light can be detected, and a color image can be captured. In this case, since a color image can be captured with one solid-state image sensor chip, an optical spectroscopic system or the like is not required, and the overall configuration can be simplified. Furthermore, unlike devices using color filters, there is no loss of light, and it is possible to increase light utilization efficiency and improve sensitivity and color reproducibility. In addition, since the structure is made by stacking photodiodes made of the same material (Si) instead of stacking different materials, there is no need to use special manufacturing methods, and it can be easily manufactured using existing manufacturing methods. This has a great practical advantage.

FIG. 5 is a plan view showing a pixel configuration of another embodiment of the present invention. Note that the same parts as in FIG. 1 are given the same reference numerals, and detailed explanation thereof will be omitted.

This embodiment differs from the previously described embodiment in that the CCD
The signal charge read out from the photoelectric conversion section is directly extracted without using the photoelectric conversion section. That is, the photodiodes 31, 32 and 33 have readout gates 81 and 82.
．． Transistors 83 are connected to each other, and signal charges read out by these transistors are taken out to metal wirings 71, 92, and 73 through impurity diffusion layers 91, 92, and 93, respectively. Even with such a configuration, the same effects as in the previous embodiment can be obtained.

Note that the present invention is not limited to the embodiments described above. In the embodiment, photoelectric conversion units made of photodiodes are arranged two-dimensionally, but it is also possible to apply the present invention to a line image sensor in which photoelectric conversion units are arranged one-dimensionally. Further, the depth position of the photodiode constituting the photoelectric conversion section may be set as appropriate depending on the absorption depths of R, G, and B shown in FIG. 7. Furthermore, the number of photodiodes constituting the photoelectric conversion section is not limited to three, and may be divided into more than three. In addition, various modifications can be made without departing from the gist of the present invention.

[Effects of the Invention] As described in detail above, according to the present invention, the photodiode as a photoelectric conversion unit can be moved in the depth direction by utilizing the fact that the light component absorbed in the semiconductor differs in the depth direction. R, G, and H components can be measured by providing separate photodiodes and performing predetermined arithmetic processing based on the detection signals of each separated photodiode. Therefore, a color image can be captured, and an image with high sensitivity and good color reproducibility can be obtained with a single solid-state image sensor chip without using a color separation filter or an optical spectroscopy system.

[Brief explanation of drawings]

FIG. 1 is a plan view showing one pixel configuration of a solid-state imaging device according to an embodiment of the present invention, FIGS. 2 and 3 are cross-sectional views showing the main part configuration of FIG. 1, and FIG. FIG. 5 is a schematic diagram showing the overall configuration of an example; FIG. 5 is a plan view showing a single pixel configuration of another embodiment of the present invention; FIGS. Figure 6 is a cross-sectional view showing photodiodes separated in the depth direction, and Figure 7 is a cross-sectional view showing R and G in the depth direction.
, B is a characteristic diagram showing changes in transmittance of three colors. DESCRIPTION OF SYMBOLS 10... N-type Si substrate, 11... P well, 12... CCD channel, 13... Element isolation layer, 14... Gate oxide film, 20... Vertical CCD. 21.22.23...Transfer electrode, 30...Photoelectric conversion unit, 31.32.33...Photodiode, 34...
Barrier layer, 41.42.43...Horizontal CCD. 60...71°81. 91° Arithmetic processing circuit, 72.73... Metal wiring, 82.83... Read gate, 92.93... Impurity diffusion layer.

Claims (3)

[Claims] (1) In addition to providing a plurality of photoelectric conversion parts on a semiconductor substrate,
In a solid-state imaging device provided with a signal charge readout section that reads out signal charges converted and accumulated in these photoelectric conversion sections, the photoelectric conversion section is composed of at least three photodiodes separated in the depth direction, A solid-state imaging device characterized in that the signal charge readout section includes a readout transistor independently connected to each photodiode of the photoelectric conversion section. (2) A photoelectric conversion unit consisting of a plurality of photodiodes arranged in a matrix on a semiconductor substrate and each separated into three in the depth direction; A signal charge readout section consisting of three readout transistors that read out the signal charges accumulated in the photoelectric conversion section for each photodiode, and a plurality of transistors that vertically transfer the signal charges read out by these signal charge readout sections. a vertical charge transfer section, a horizontal charge transfer section that horizontally transfers the signal charges transferred by these vertical charge transfer sections, and a horizontal charge transfer section that transfers R, G, and B signals based on the signal charges transferred by the horizontal charge transfer section. What is claimed is: 1. A solid-state imaging device comprising: an arithmetic processing unit that obtains three primary color signals. (3) Three horizontal charge transfer sections are formed corresponding to the three photodiodes of the photoelectric conversion section, and each horizontal charge transfer section transfers the signal charge transferred by the vertical charge transfer section to the same depth. 3. The solid-state imaging device according to claim 2, wherein the solid-state imaging device is supplied separately for each photodiode at a different position.