JP2003282859A - Solid state imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enlarge a dynamic range not depending on an enlargement of a light-receiving area or an enlargement of an electric charge transfer width. <P>SOLUTION: There are provided a light-receiving part 1 for generating a signal electric charge by a photoelectric transfer; a horizontal transfer register 5 for horizontally transferring the signal electric charge photoelectrically transferred by the light-receiving part 1; an overflow barrier 7 provided in the horizontal transfer register 5 for dividing the horizontal transfer register 5 into a plurality of parts; and FD amplifiers 9A, 9B connected respectively independently of individual horizontal transfer registers 5A, 5B divided by this overflow barrier 7. This overflow barrier 7 is formed so as to determine a handling electric charge amount of the horizontal transfer register 5A on a side of the light-receiving part. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a CCD (Charge Couple) for photoelectrically converting an optical signal and outputting the signal.
The present invention relates to a solid-state image sensor suitable for being applied to an ed device). In detail, each charge transfer unit divided into a plurality of potential barrier units is provided with a charge-voltage converter independently connected to each other, and the potential barrier unit determines the amount of charge handled by the charge transfer unit on the light receiving unit side. In this way, when a large amount of light is incident on the light receiving part, the signal charge corresponding to the normal amount of light can be transferred from the charge transfer part on the light receiving part side to one of the charge-voltage conversion parts. The signal charge exceeding the amount of charge handled by each unit can be independently transferred from the charge transfer unit on the non-light-receiving unit side to the other charge-voltage conversion unit.

[0002]

2. Description of the Related Art In recent years, image pickup devices such as digital cameras and digital video have become more and more popular. Along with this, there is an increasing demand for improvement in image quality of these image pickup devices, and how to enhance the image pickup function of the solid-state image pickup device mounted in the image pickup device has been a problem.

FIG. 8 is a CCD solid-state image pickup device 9 according to a conventional example.
It is a conceptual diagram which shows the structural example of 0. This solid-state image sensor 90
The imaging operation of is as follows. First, individual light receiving parts (pixels: Unit Cell) arranged in a matrix.
Light is incident on 91. Then, a signal charge corresponding to the amount of incident light is generated in each light receiving portion 91 (photoelectric conversion). next,
These signal charges are read out from the individual light receiving portions 91, and the floating diffusion (FD) amplifier 9 is read by the vertical transfer register 92 and the horizontal transfer register 93.
Up to 4 are sequentially transferred. Then, the FD amplifier 94 converts the signal charge into a signal voltage and outputs the signal voltage from a predetermined terminal.

The image pickup function of such a solid-state image pickup device 90,
For example, in order to enhance the color (contrast of colors) of an image, the dynamic range of the solid-state image sensor 90 may be expanded. This dynamic range is expressed by a formula. D _range = 10 × log (V _max / (V _noise ² / V _max )) ... D _range : Dynamic range [dB] V _max : Saturation signal amount (maximum signal amount that can be output) [mV] V _noise : As is clear from the noise level [mV] equation, the dynamic range can be expanded by increasing V _max .

Therefore, in order to increase this V _max (to expand the dynamic range), the solid-state image pickup device 9
At 0, the area of the light receiving unit 91 is expanded to
It was designed to increase the amount of charge that can be stored in the. Further, according to this solid-state image pickup device 90, the dimensional width (register width) of the vertical transfer register 92 and the horizontal transfer register 92 is widened to increase the amount of charge handled. In addition, in order to expand the dynamic range, a method of combining the signal outputs (packets) by causing the same light receiving unit or a plurality of light receiving units to output signal voltages with different storage times is also under study.

[0006]

However, the method of expanding the dynamic range according to the conventional example has the following problems. In order to expand the dynamic range, the light receiving unit 9
It is necessary to enlarge the area of No. 1 or to widen the vertical transfer register 92 and the horizontal transfer register 93, which hinders the miniaturization of the solid-state imaging device 90. When the narrowing angle of the horizontal output (HOG) gate section between the horizontal transfer register 93 and the FD amplifier 94 is θ ′, this narrowing angle θ ′ is the horizontal transfer register 9
A wide angle of 3 results in a wide angle. The widening of the narrowing-down angle θ ′ causes the transfer deterioration of the horizontal transfer register 93. With the method of outputting signal voltages with different storage times from the same light receiving unit and combining the signal outputs, signal charges with different storage times cannot be stored in the light receiving unit at the same timing, so especially when capturing a moving image. It becomes an unnatural picture. In the method of outputting signal voltages with different storage times from a plurality of light receiving units and synthesizing the signal outputs, the resolution is deteriorated as compared with the method of using the same light receiving unit. Incidentally, in order to obtain the same resolution as the method using the same light receiving section, it is necessary to halve the area of the light receiving section. The general use of this method is not preferable in terms of manufacturing technology and manufacturing cost.

Therefore, the present invention solves such a problem, and provides a solid-state image pickup device capable of expanding the dynamic range without depending on the expansion of the light receiving area and the expansion of the charge transfer width. With the goal.

[0008]

SUMMARY OF THE INVENTION The above-mentioned problems are solved by a light receiving section for generating a signal charge by photoelectric conversion, a charge transfer section for transferring the signal charge photoelectrically converted by the light receiving section in a predetermined direction, and this charge transfer. A potential barrier section provided in the charge transfer section to divide the section into a plurality of sections, and a charge-voltage conversion section independently connected to each charge transfer section divided by the potential barrier section, This potential barrier section is solved by a solid-state imaging device characterized in that it determines the amount of charges handled by the charge transfer section on the light receiving section side.

According to the solid-state image pickup device of the present invention, when a large amount of light is incident on the light receiving part, the signal charge corresponding to the normal amount of light can be transferred from the charge transfer part on the light receiving part side to one of the charge-voltage conversion parts. Moreover, the signal charges exceeding the amount of charges handled by the charge transfer unit on the light receiving unit side can be independently transferred from the charge transfer unit on the non-light receiving unit side to the other charge voltage conversion unit.

Further, for example, with respect to the charge transfer section on the light receiving section side, the sensitivity of the charge transfer section on the non-light receiving section side is lowered, and the signal voltages output from the respective charge voltage conversion sections are weighted and combined. It As a result, the dynamic range can be greatly expanded.

[0011]

DETAILED DESCRIPTION OF THE INVENTION A solid-state image sensor according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a conceptual diagram showing a configuration example of a solid-state image sensor 100 according to the embodiment of the present invention. In this embodiment, a charge-voltage conversion unit independently connected to each charge transfer unit divided into a plurality of potential barrier units is provided, and the amount of charges handled by the charge transfer unit on the light receiving unit side is set to the potential barrier unit. When a large amount of light is incident on the light receiving unit, the signal charge corresponding to the normal amount of light can be transferred from the charge transfer unit on the light receiving unit side to one of the charge-voltage conversion units, and the charge transfer on the light receiving unit side can be performed. It is possible to independently transfer the signal charges exceeding the amount of charge handled by one unit from the charge transfer unit on the non-light-receiving unit side to the other charge-voltage conversion unit, and at the same time, a plurality of signals corresponding to the received light amount from one light-receiving unit. The voltage can be output at the same time.

The solid-state image sensor 100 shown in FIG. 1 is an area-type CCD solid-state image sensor. As shown in FIG. 1, the solid-state imaging device 100 includes a plurality of light receiving units 1 arranged in a matrix. The light receiving unit 1 is a so-called pixel that generates (photoelectrically converts) a predetermined amount of signal charge according to the amount of incident light.

Further, in the solid-state image pickup device 100, the signal charges read from the light receiving portions 1 are vertically (see FIG. 1).
Vertical transfer register 3 for transfer in the direction of the dashed arrow shown in
Is equipped with. The vertical transfer register 3 is, for example, a four-phase drive type CCD. A plurality of the vertical transfer registers 3 are provided along one side of the column of the light receiving units 1.

The solid-state image pickup device 100 further includes a horizontal transfer register 5 as an example of a charge transfer unit for horizontally transferring the signal charges vertically transferred by the vertical transfer register 3. This horizontal transfer register 5
Is a two-phase drive type CCD, for example. As shown in FIG. 1, the horizontal transfer register 5 is arranged in the horizontal direction so as to be connected to each of the plurality of vertical transfer registers 3 arranged in the vertical direction.

By the way, as shown in FIG. 1, in the solid-state image pickup device 100, an overflow barrier 7 which is an example of a potential barrier portion is provided in the horizontal transfer register 5. The horizontal transfer register 5 is divided by the overflow barrier 7 into the light receiving portion 1 side and the non-light receiving portion 1 side.

Hereinafter, the light receiving portion side of the horizontal transfer register 5 divided by the overflow barrier 7 is referred to as a first horizontal transfer register 5A, and the non-light receiving portion side is referred to as a second horizontal transfer register 5B. This first horizontal transfer register 5
A is for transferring the signal charges corresponding to the normal light amount in the horizontal direction (the direction of the two-dot chain line arrow shown in FIG. 1). The second horizontal transfer register 5B is for horizontally transferring the signal charge (a large amount of light signal charge) exceeding the amount of charge handled by the first horizontal transfer register 5A. The transfer functions of the first and second horizontal transfer registers 5A and 5B will be described later.

Further, the solid-state image pickup device 100 includes a first floating diffusion amplifier (FD amplifier) 9A which is an example of one charge voltage transfer section. The first FD amplifier 9A is connected to the first horizontal transfer register 5A, and the first horizontal transfer register 5A
It has a function of converting the signal charge transferred by the above into a signal voltage and outputting it. In addition, this solid-state imaging device 10
0 includes a second FD amplifier 9B which is an example of the other charge voltage transfer unit. The second FD amplifier 9B is connected to the second horizontal transfer register 5B, and has a function of converting the signal charge transferred by the second horizontal transfer register 5B into a signal voltage and outputting the signal voltage. In this way, the first and second FD amplifiers 9A and 9B are independently provided.

FIG. 2 is a sectional view taken along arrow X1-X2 showing an example of the structure of the first horizontal transfer register 5A. As shown in FIG. 2, the horizontal transfer register 5A for horizontally transferring the signal charges of the normal light quantity is provided on the N-type semiconductor substrate 11 by the P
-Type well layer 13, N-type buried channel layer 15A selectively provided in the well layer 13, a gate insulating film provided on the buried channel layer 15A and the well layer 13, and this gate insulating film It is composed of electrode portions 17A and 19A for signal charge transfer, which are provided above the well layer 13 and the buried channel layer 15A, respectively.

Of these, the semiconductor substrate 11 is, for example, single crystal silicon. The gate insulating film is a silicon oxide film. When the two-phase horizontal transfer clock pulses φH1 and φH2 are applied to the electrode portions 17A and 19A, the signal charges of the normal light amount are transferred to the left in FIG.

Further, as shown in FIG. 2, the FD amplifier 9A for converting the signal charge transferred by the horizontal transfer register 5A into a voltage and outputting the voltage is an N-type floating diffusion (which is provided in the P-type well layer 13). FD) Layer 2
3A, an N-type reset drain (RD) layer 25A, a gate insulating film provided on the well layer 13 between the FD layer 23A and the RD layer 25A, and a reset gate provided on the gate insulating film ( RG) electrode 27 and FD layer 23
And an output source follower 29A connected to A.

The FD layer 23A is an impurity diffusion layer that functions as a capacitor that stores the signal charges transferred from the horizontal transfer register 5A to the FD amplifier 9A. The output source follower 29A outputs the signal charges accumulated in the FD layer 23A. Furthermore, the RG electrode 2
The 7A and RD layers 25A are for resetting the FD layer 23A after outputting the signal charges accumulated in the FD layer 23A.

As shown in FIG. 2, a horizontal output gate (HOG) 21A is provided between the FD amplifier 9A and the horizontal transfer register 5A. The horizontal output gate 21A includes a horizontal transfer register 5A and an FD amplifier 9A.
It is for controlling the flow of signal charges between them.

FIG. 3 is a sectional view taken along arrow X3-X4 showing an example of the configuration of the second horizontal transfer register 5B. As shown in FIG. 3, the second horizontal transfer register 5B that transfers a large amount of signal charge has the same structure as the first horizontal transfer register 5A shown in FIG. That is, the electrode parts 17B and 1
By applying the two-phase horizontal transfer clock pulses φH1 and φH2 to 9B, a large amount of signal charge is transferred to the left in FIG.

The N-type impurity concentration of the buried channel layer 15B of the second horizontal transfer register 5B is set lower than that of the buried channel layer 15A (see FIG. 2) of the first horizontal transfer register 5A. There is. Also,
The N-type impurity concentration of the FD layer 23B and the RD layer 25B is F
The impurity concentration is lower than that of the D layer 23A and the RD layer 23A (see FIG. 2).

4A and 4B are sectional views taken along the arrows Y1-Y2 and Y3-Y4 showing an example of the structure of the overflow barrier 7. As shown in FIG. 4A, the first horizontal transfer register 5A
The overflow barrier 7 that divides the second horizontal transfer register 5B and the second horizontal transfer register 5B is, for example, a P-type impurity diffusion layer provided in the well layer 13. The P-type impurity concentration of the overflow barrier 7 is made lower than the P-type impurity concentration of the well layer 13. That is, the potential of the overflow barrier 7 is higher than that of the well layer 13.

By adjusting the impurity concentration (potential) of the overflow barrier 7, the formation position, etc., the first horizontal transfer register 5A can be used as a transfer path for signal charges of normal light quantity. In other words, the overflow barrier 7 determines the amount of charge handled by the horizontal transfer register 5A.

Therefore, a large amount of light is incident on the light receiving portion 1, and a signal charge (normal amount of light signal charge + large amount of signal charge) exceeding the amount of charge to be handled is transferred to the buried channel layer 15A of the horizontal transfer register 5A. In such a case, only a large amount of signal charges can be retransferred to the channel layer 15B of the horizontal transfer register 5B by passing over the overflow barrier 7.

Further, the electrode portion 1 for charge transfer shown in FIG. 4A.
7A and 17B and the electrode parts 19A and 19 shown in FIG. 4B.
2 phase horizontal transfer clock pulses φH1 and φ
By applying H2, the signal charge of normal light quantity and
A large amount of light signal charges can be transferred simultaneously in the FD layers 23A and 23B.

Then, the signal charge of the normal light quantity transferred to the FD layer 23A is converted into a signal voltage by the output source follower 29A shown in FIG. 5 and output from the terminal 31A. At the same time, the large amount of light signal charge transferred to the FD layer 23B is converted into a signal voltage by the output source follower 29B and output from the terminal 31B.

FIG. 6 is a diagram showing the relationship between the amount of incident light and the signal output of the solid-state image sensor 100. The X-axis of FIG.
It is the amount of light incident on (see FIG. 1). The Y axis in FIG. 6 is the signal voltage output from the terminals 31A and 31B (see FIG. 4). As shown in FIG. 6, when the normal light amount is incident on the light receiving portion, only the signal voltage (V _OUT1 ) having the normal light amount is output.

On the other hand, when a large amount of light is incident on the light receiving portion, V _OUT1 is output and at the same time, a large amount of signal voltage (V _OUT2 ) is output. As described above, in the solid-state imaging device 100, the output of the normal light amount and the output of the large light amount can be obtained at the same time, so that the signal voltage (signal output) can be easily combined.

Further, in FIG. 6, the dynamic range can be greatly expanded by reducing the slope of the signal voltage with respect to the amount of incident light of V _OUT2 and further weighting and combining V _OUT2 and V _OUT1 according to this slope. . This V
_In order to reduce the inclination of _OUT2 , the gain of the FD amplifier 9B (see FIG. 1) may be set lower than that of the FD amplifier 9A (see FIG. 1). Change in output of FD amplifier ΔV
_OUT is represented by an expression. ΔV _out = Q / C _FD ... Q: Signal charge amount C _FD : Floating diffusion (FD) capacitance

Therefore, as is clear from the equation, in order to lower the gain of the FD amplifier 9B, it is sufficient to reduce Q (a large amount of light signal charge) and increase C _FD (capacity of the FD layer 23B). . In order to lower the Q, for example, the sensitivity of the second horizontal transfer register is made lower than that of the first horizontal transfer register. That is, as shown in FIG. 4A, the N-type impurity concentration of the buried channel layer 15B is made lower than that of the buried channel layer 15A, and the transfer efficiency of the horizontal transfer register 5B is suppressed to a low level. In order to increase C _FD , for example, the FD layer 23B (see FIG. 3) has an impurity concentration lower than that of the FD layer 23A (see FIG. 2).

As described above, the sensitivity of the horizontal transfer register 5B is lowered to suppress the Q and the impurity concentration of the FD layer 23B is made thin, so that the slope of the signal voltage with respect to the incident light amount of V _OUT2 shown in FIG. Can be made smaller. In addition, when the signal processing of the external circuit is performed using these signal voltages V _OUT1 and V _OUT2 , the amplifier gain on the external circuit side may be adjusted and the signal output may be combined in the external circuit. . Also in this case, the dynamic range of the solid-state imaging device 100 can be expanded. The synthesis of the signal outputs may be performed inside the solid-state imaging device 100 or may be taken out to the outside.

By the way, the solid-state image pickup device 10 shown in FIG.
Since the first horizontal transfer register 5A of 0 only needs to handle the signal of the normal light amount, the conventional solid-state image sensor 90 that transfers both the signal of the normal light amount and the signal of the large light amount by one horizontal transfer register is used. In comparison, the register width can be narrowed.

As a result, the horizontal transfer registers 5A and FD
When the narrowing angle of the HOG gate between the amplifiers 9A is θ, this narrowing angle θ is the conventional solid-state imaging device 9
The angle can be narrower than 0 (see FIG. 8). Therefore, the margin for the transfer deterioration of the horizontal transfer register 5A can be widened.

Further, the first horizontal transfer register 5A
The register width of is set narrower than the register width of the second horizontal transfer register. This can further improve the transfer efficiency of signal charges in a small signal.

Thus, the solid-state image pickup device 1 according to the present invention
According to 00, the horizontal transfer registers 5A and 5B divided by the overflow barrier 7 are provided with FD amplifiers 9A and 9B, which are independently connected to each other. It is determined by the overflow barrier 7.

Therefore, when a large amount of light is incident on the light receiving portion 1, the horizontal transfer register 5A on the light receiving portion side to the FD amplifier 9
The signal charge corresponding to the normal light amount can be transferred to A, and the signal charge exceeding the handling charge amount of the horizontal transfer register 5A can be transferred independently from the horizontal transfer register 5B on the non-light-receiving side to the FD amplifier 9B.

With this, the signal voltage of the normal light quantity and the signal voltage of the large light quantity can be simultaneously output, so that the signal voltage (signal output) can be easily combined. Therefore, the dynamic range can be greatly expanded.

Further, as compared with the conventional method, the signal charge of the normal light quantity and the signal charge of the large light quantity can be simultaneously stored in one light receiving portion. Therefore, a natural image can be obtained even when a moving image is captured.

In this embodiment, the case where the horizontal transfer register 5 is provided with one overflow barrier has been described, but the number of overflow barriers is not limited to one. For example, an overflow barrier (drain) may be further provided in the second horizontal transfer register 5B, and the signal charges overflowing in the second horizontal transfer register 5B may be transferred using the third horizontal transfer register. By converting the signal charges transferred by the first, second and third horizontal transfer registers into charge-voltage and synthesizing the outputs,
It is possible to further improve the dynamic range.

The overflow barrier 7 that determines the amount of charge handled by the first horizontal transfer register 5A is not limited to the P-type impurity diffusion layer that serves as a barrier for the N-type buried channel layers 15A and 15B. . For example, as shown in FIG. 7A, horizontal transfer registers 5A and 5A
A gate insulating film and an independent gate electrode 51 may be provided on the semiconductor substrate 11 between B, and an electrostatic potential (for example, −5 V) may be applied to this gate electrode 51 to lower the potential of this portion. Also in this case, the buried channel layer 15A
And between 15B and the P-type well layer 13 shown in FIG. 7B,
It is possible to form a potential barrier that determines the amount of charge handled by the first horizontal transfer register 5A.

Further, in this embodiment, an area type C
The case of the CD solid-state image sensor has been described, but the present invention is not limited to this, and may be, for example, a facsimile machine (fax).
It can also be applied to a line-type CCD solid-state image pickup device used in.

[0045]

As described above, according to the solid-state image pickup device 100 of the present invention, the charge-voltage converters independently connected to the respective charge transfer units divided by the potential barrier unit are provided. The charge amount handled by the charge transfer unit on the light receiving unit side is determined by the potential barrier unit.

With this configuration, when a large amount of light is incident on the light receiving portion, the signal charge corresponding to the normal amount of light can be transferred from the charge transfer portion on the light receiving portion side to one of the charge-voltage conversion portions, and the light receiving portion side. The signal charges exceeding the amount of charge handled by the charge transfer unit can be independently transferred from the charge transfer unit on the non-light receiving unit side to the other charge voltage conversion unit.

Therefore, as compared with the conventional method, a plurality of signal voltages corresponding to the amount of received light can be simultaneously output from one light receiving portion, so that the signal voltages can be easily combined. As a result, the dynamic range can be greatly expanded.

The present invention is extremely suitable when applied to a CCD solid-state image pickup device which photoelectrically converts an optical signal and outputs the signal.

[Brief description of drawings]

FIG. 1 is a conceptual diagram showing a configuration example of a solid-state image sensor 100.

FIG. 2 is an X diagram showing a configuration example of a first horizontal transfer register 5A.
It is a 1-X2 arrow sectional view.

FIG. 3 is an X diagram showing a configuration example of a second horizontal transfer register 5B.
FIG. 3 is a sectional view taken along line 3-X4.

4A and 4B are cross-sectional views taken along arrows Y1-Y2 and Y3-Y4 showing a configuration example of the overflow barrier 7.

FIG. 5 is a circuit diagram showing a configuration example of FD amplifiers 9A and 9B.

6 is a diagram showing the relationship between the amount of incident light and the signal output of the solid-state image sensor 100. FIG.

7A and 7B are cross-sectional views showing another configuration example of the overflow barrier 7.

FIG. 8 is a conceptual diagram showing a configuration example of a solid-state imaging device 90 according to a conventional example.

[Explanation of symbols]

1 ... Light receiving part, 3 ... Vertical transfer register, 5 ...
Horizontal transfer register (charge transfer unit), 7 ... Overflow barrier (potential barrier unit), 9A, 9B ... FD amplifier (charge voltage conversion unit), 100 ... Solid-state image sensor

Claims (5)

[Claims] 1. A light receiving section for generating signal charges by photoelectric conversion, a charge transfer section for transferring signal charges photoelectrically converted by the light receiving section in a predetermined direction, and a charge transfer section for dividing the charge transfer section into a plurality of parts. A potential barrier section provided in the charge transfer section; and a charge-voltage conversion section independently connected to each charge transfer section divided by the potential barrier section, wherein the potential barrier section is on the light receiving section side. A solid-state image sensor, wherein the amount of charges handled by the charge transfer unit is determined. 2. The solid-state image sensor according to claim 1, wherein the respective signal voltages output from the plurality of charge-voltage converters are weighted and combined. 3. The solid-state image sensor according to claim 2, wherein the signal voltages are combined inside or outside the solid-state image sensor. 4. The solid-state image sensor according to claim 1, wherein a plurality of potential barrier sections are provided in the charge transfer section. 5. The individual charge transfer sections divided by the potential barrier section are characterized in that the charge transfer section on the light receiving section side has a smaller dimension width than the charge transfer section on the non-light receiving section side. The solid-state image sensor according to claim 1.