JPH09326480A - Solid state image sensor and driving method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a good image by injecting bias charges into a region for storing charges through photoelectric conversion after reset and then storing signal charges and reading out a pixel signal thereby improving fluctuation in the I/O characteristics of the pixel. SOLUTION: After signal charges are stored (for an interval T1 A) in a first frame, a pixel signal is read out (at a moment of time T1 B) and then the charges stored in a pixel are reset entirely (at a moment of time T1 C). Subsequently, bias charges are injected into the pixel (at a moment of time T1 D) in order to compensate for the fluctuation of potential well thus bringing about a uniform potential state in the breadthwise direction of the gate. A signal representative of the absence of the signal charge in the pixel, i.e., a reference value due to the bias charges, is then read out (at a moment of time T1 E). Thereafter, storage of signal charges in a second frame is started and a similar operation is repeated. The difference between a pixel signal read out after storing the signal charges and a reference value based on the bias charges of the pixel is taken out as an output signal.

Description

Detailed Description of the Invention

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a solid-state imaging device and a driving method thereof.

[0002]

2. Description of the Related Art In recent years, in accordance with a demand for higher resolution of a solid-state image pickup device, an amplification type solid-state image pickup device having no smear and capable of realizing fine pixels has been developed. This amplification type solid-state imaging device is provided with a transistor that amplifies with an optical signal for each pixel, is configured to accumulate electric charge by photoelectric conversion under the gate electrode of the pixel transistor, and take out potential modulation by this electric charge as a signal. There is. For example CM
Imaging device structures such as D (charge modulation device) are known. There are various types of circuits for extracting signals, but basically, this is done by turning on a transistor and flowing a current through the channel.

[0003]

By the way, in this type of amplification type solid-state image pickup device, for example, in the amplification type solid-state image pickup device having MOS type transistors as pixels, the potential under the gate electrode is uniform in the gate width direction. Is ideal. However, in reality, there is a maximum unevenness of about 100 mV. What kind of problems will occur when this unevenness exists will be described with reference to the conceptual diagram 15.

As a pixel of an amplification type solid-state image pickup device, FIG.
The pixel MOS transistor shown in is used. This pixel MO
In the S-transistor 1, a second conductivity type, that is, an n-type semiconductor region, that is, an overflow barrier region 3 and a p-type semiconductor well region 4 are formed on a silicon semiconductor substrate 2 of a first conductivity type, for example, p-type. S on the semiconductor well region 4
A ring-shaped gate electrode 6 capable of transmitting light is formed through a gate insulating film 5 of iO ₂ or the like, and the gate electrode 6 is masked in the p-type semiconductor well region 4 corresponding to the inside and the outside of the ring-shaped gate electrode 6. And each selfa line
A mold source region 7 and a drain region 8 are formed. The annular gate electrode 6 is made of a thin or transparent material so that light is not absorbed as much as possible, and for example, a thin polycrystalline silicon is used. In the amplification type solid-state imaging device, a plurality of pixel MOS transistors are arranged in a matrix.

In the pixel MOS transistor 1, as shown in the figure, the light transmitted through the ring-shaped gate electrode 6 is photoelectrically converted in silicon to generate an electron-hole, and one of these charges, in this example, is generated. The holes h are accumulated as signal charges in the potential well (see FIG. 14) formed in the p-type semiconductor well region 4 below the annular gate electrode 6, and the modulation of the substrate bias due to the charges is taken out as a signal. . That is, when a high voltage is applied to the gate electrode 6 through the vertical selection line and the pixel MOS transistor 1 is turned on, a channel current (so-called drain current) Id flows in the surface channel, and this channel current Id is modulated by the signal charge h. Therefore, this channel current is output through the signal line, and the amount of change is output as a signal.

The above example is an n-channel transistor, but the same can be said for a p-channel transistor and a CMD.

By the way, when the amount of fixed charges in the gate insulating film 5 is uneven in the gate width (W) direction (so-called gate width in the circumferential direction), or in the impurity concentration in the gate width (W) direction. 15 is uneven, or due to the asymmetry of the pattern (source and drain contact positions, etc.) of the pixel MOS transistor in the gate width (W) direction, the surface potential (or sensor potential) as shown in the three-dimensional potential diagram of FIG. ) Shows unevenness in the gate width (W) direction. In addition, in FIG.
The solid line shows an ideal surface potential without unevenness when local unevenness occurs.

Since the unevenness of the potential in the gate width (W) direction is different among pixels, there is a problem that the input / output characteristics of the pixels are varied and the image is adversely affected. As described above, there are many causes of potential unevenness, and it is very difficult to make them uniform by the effort of the wafer process.

In view of the above points, the present invention provides a solid-state image sensor and a method of driving the same that can improve the input / output variability of pixels and obtain good images.

[0010]

According to the present invention, after resetting, a bias charge is injected into a region for accumulating charges by photoelectric conversion, signal charges are accumulated, and a pixel signal is read out.

The variation of the potential well is filled by the injection of the bias charge, and the signal charge is accumulated in the uniform potential well state. By reading out a signal based on this signal charge, variations in the input / output characteristics of the pixel are improved.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION The solid-state image pickup device according to the present invention is configured such that after resetting, bias charges are injected into a region for accumulating charges by photoelectric conversion, signal charges are accumulated, and pixel signals are read out.

According to the present invention, in the above solid-state image pickup device, bias charges are injected to read a reference value, and then signal charges are accumulated.

According to the present invention, in the above solid-state image pickup device, the readout of pixel signals is nondestructive readout.

According to the present invention, in the above solid-state image pickup device, the difference between the reference value read immediately after the injection of the bias charge and the pixel signal read after the accumulation of the signal charge is used as the output signal.

According to the present invention, in the above solid-state image pickup device, the difference between the reference value due to the bias charge injected after reading the pixel signal and the pixel signal is output as an output signal.

According to the present invention, in the above-mentioned solid-state image pickup device, the difference between the reference value by the bias charge injected before reading the pixel signal and the pixel signal is output as an output signal.

According to the present invention, in the above solid-state imaging device, a charge injection electrode is provided to inject a bias charge into a region for accumulating charges by photoelectric conversion.

According to the present invention, in the above solid-state image pickup device, a bias charge is injected into a region for accumulating charges by photoelectric conversion using avalanche.

According to the present invention, in the above solid-state image pickup device, the bias charges are injected from the substrate side into the region for accumulating charges by photoelectric conversion.

According to the present invention, in the above solid-state imaging device, the bias charge injection amount is controlled by performing incomplete reset after the bias charge is injected.

A method of driving a solid-state image pickup device according to the present invention comprises:
After resetting, bias charges are injected into a region for accumulating charges by photoelectric conversion, signal charges are accumulated, and pixel signals are read out.

According to the present invention, in the method for driving a solid-state image pickup device, a signal charge is stored after injecting a bias charge and reading a reference value.

According to the present invention, in the method for driving a solid-state image pickup device, pixel signals are read out nondestructively.

According to the present invention, in the method for driving a solid-state image pickup device, the difference between the reference value read immediately after the injection of the bias charge and the pixel signal read after the accumulation of the signal charge is used as the output signal.

According to the present invention, in the method for driving a solid-state image pickup device, a difference between a reference value due to a bias charge injected after reading a pixel signal and the pixel signal is used as an output signal.

According to the present invention, in the method for driving a solid-state image pickup device, a difference between a reference value due to a bias charge injected before reading a pixel signal and the pixel signal is used as an output signal.

According to the present invention, in the method for driving a solid-state image pickup device, a charge injection electrode is provided to inject a bias charge into a region where charges are accumulated by photoelectric conversion.

According to the present invention, in the method for driving a solid-state image pickup device, bias charge is injected into a region for accumulating charges by photoelectric conversion using avalanche.

According to the present invention, in the method for driving a solid-state image pickup device, bias charges are injected from the substrate side into a region for accumulating charges by photoelectric conversion.

According to the present invention, in the method for driving a solid-state image pickup device, after the bias charge is injected, an incomplete reset is performed to control the bias charge injection amount.

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

1 and 2 show an example of an amplification type solid-state image pickup device applied to the present invention. This amplification type solid-state imaging device 21, as described above, has a second conductivity type, that is, an n-type semiconductor layer, that is, an overflow barrier region on a first conductivity type, for example, a p-type silicon semiconductor substrate 22, as shown in FIG. 23
And a p-type semiconductor well region 24 are formed, and a ring-shaped gate electrode 26 capable of transmitting light is formed on the p-type semiconductor well region 24 through a gate insulating film 25 made of SiO ₂ or the like. Gate electrodes 26 are formed in the p-type semiconductor well regions 24 corresponding to the inside and the outside of the gate electrode 26, respectively.
An n-type source region 27 and a drain region 28 are formed by an ion implantation method using self-alignment so as to sandwich the above region, and a pixel MOS transistor 29, which constitutes one pixel, is formed therein.

The ring-shaped gate electrode 26 is made of a thin material or a transparent material so as not to absorb light as much as possible. For example, polycrystalline silicon, tungsten polycide, tungsten silicide or the like can be used. In this example, a thin film of polycrystalline silicon having good transparency is used.

This pixel MOS transistor 29 is shown in FIG.
As shown in FIG. 3, a plurality of source regions 2 of the pixel MOS transistors 29 are arranged in a matrix and correspond to each column.
7 is connected to a common signal line 31 made of, for example, the first layer Al formed along the vertical direction. The second layer 7 is arranged at a position corresponding to each row of the pixel MOS transistors 29 so as to be orthogonal to the signal line 31. A vertical selection line 32 of Al is formed along the horizontal direction.

Then, two pixels M adjacent in the horizontal direction
A U-shaped wiring layer, that is, a contact buffer layer is formed to extend to each of the annular gate electrode 26 and the vertical selection line 32 of the OS transistor 29, and this contact buffer layer 33 and two pixel MOS transistors 29 are formed.
And the vertical selection line 32.

Further, between the pixel MOS transistors 29 that do not straddle the contact buffer layer 33, a drain power supply line 34 made of, for example, the first layer Al connected to the drain region 28 is formed. Reference numeral 35 denotes a drain contact portion between the drain power supply line 34 and the drain region 28, 36 denotes a source contact portion between the source region 27 and the signal line 31, and 37 denotes a contact portion between the contact buffer layer 33 and the vertical selection line 32.

The basic idea of this embodiment is to bias the potential well below the gate electrode 26 (so-called sensor portion) of the pixel MOS transistor 29 in the amplification type solid-state image pickup device 21 in which the signal charge is to be accumulated. Injecting charges to eliminate the unevenness of the potential well.

Next, the timing for injecting the bias charge will be described with reference to FIGS. 3 and 4 showing the operation phase for reading a signal from the amplification type solid-state image pickup device.

FIG. 3 shows an example when the electronic shutter is not used. In this example, after the signal charge is accumulated in the first frame (period T _1A ), the pixel signal is read out (time T
_1B ), then all the charges accumulated in the pixel are reset (time point T _1C ), and then bias charges are injected into the pixel (time point T _1D ), the unevenness of the potential well is filled, and a uniform potential state in the gate width direction is obtained. And Next, the signal in the state where there is no signal charge in the pixel, that is, the reference value based on the bias charge is read (time T _1E ).

After reading the reference value, the accumulation of the signal charges in the next second frame is started, and after the accumulation of the signal charges is completed (period T _2A ), the pixel signal is read out (time T _2B ) as described above. The charge of the pixel is reset (time T _2C ), the bias charge is injected into the pixel (time T _2D ), and the reference value is read out by the bias charge (time T _2E ). Hereinafter, the same operation is repeated.

In this example, the pixel signal (that is, the signal including the bias charge and the signal charge) read after the accumulation of the signal charge and the reference value (that is, the state in which there is no signal charge) based on the bias charge of this pixel. Signal) in the output signal. As an output signal of the difference between the two, for example, considering a pixel signal read at time T _2B in the second frame, this pixel signal and a reference value based on the bias charges injected before reading the pixel signal, that is, When the difference from the reference value read in the period T _1E of the previous frame (first frame) is used as the output signal (first combination (A)), or the time T _2B in the second frame is read. The difference between the pixel signal and the reference value due to the bias charge injected after the reading of the pixel signal, that is, the reference value read at time T _2E can be used as the output signal (second combination (B)).

In the case of the output signal of the difference due to the former combination (A), noise due to the variation of the bias charge injection amount can be eliminated, but a frame memory for storing the memory until the pixel signal of the next frame is read out. Will be needed. On the contrary, in the case of the difference output signal due to the latter combination (B), the frame memory can be eliminated. However, noise is generated due to the bias charge injection.

FIG. 4 shows an example when the electronic shutter is used.
In this example, the electronic phase is further added to the operation phase of FIG.
The operation to perform the setting is added. That is, for example, in the second frame
Accumulation period T_2AAt a desired point in time (eg period T _2AaEnd of
The pixel charge is reset at the end (time T)_2G),
A bias is then injected into the pixel (time T_2H), This bike
Read the reference value based on the as-charge (time T_2I). That
After that, the remaining accumulation period T_2AbDepending on the signal charge accumulated in
Read pixel signal is read out (at time T_2B). Then the same
Reset the pixel charge to (at time T_2C), Note of bias charge
On (Time T_2D), Reading the reference value with this bias charge
Out (at time T_2E) Is done. Amplification type solid-state image sensor
When resetting the select line,
The focal plane electronic shutter
Has been realized. This focal plane electronic sha
Even if the electronic shutter is not used,
It is possible to drive at the timing.

That is, even when this electronic shutter is used, the difference between the pixel signal and the reference value due to the bias charge of the pixel is used as the output signal. As an output signal of the difference between the two, for example, considering a pixel signal read in the period T _2B in the second frame, this pixel signal and the bias charge injected in the electronic shutter operation before reading the pixel signal are considered. When a difference from the reference value (reference value read at the time point T _2I ) is used as the output signal (first combination (A)), or the pixel signal read in the period T _2B in the second frame The output signal can be the difference between the reference value due to the bias charge injected after the reading of the pixel signal, that is, the reference value read during the period T _2E (second combination (B)).

In the case of the output signal of the difference due to the former combination (A), noise accompanying bias charge injection can be eliminated. However, as in the above example, a frame memory for storing the reference value until reading the pixel signal is required. Further, in the case of the output signal of the difference due to the latter combination (B), the frame memory is not required. However, noise is generated due to the bias charge injection.

Next, an example of a bias charge injection method will be described. 5 and 6 show an example (example (I)) of a bias charge injection method. In this example, a semiconductor region serving as a charge injection source is formed so as to be adjacent to the gate electrode 26 of the pixel MOS transistor 29, which is a p-type semiconductor region 41 in this example.
And a bias charge injecting electrode 44 for applying a charge injecting bias voltage V _prb is ohmic-connected to the p-type semiconductor region 41, and the p-type semiconductor region 41 and the gate electrode 26 are connected to each other. Gate insulating film 4 between
A bias charge injection control gate electrode 43 for controlling the injection of bias charge is formed by way of FIG.

In the bias charge injection operation phase, as shown in the potential diagram of FIG. 6, in the state of the potential φ ₁ when there is no accumulated charge under the gate electrode 26, the charge injection source of the charge injection source is supplied through the bias charge injection electrode 44. A predetermined charge injection bias voltage V is applied to the p-type semiconductor region 41.
_The control gate voltage V _gc is applied to the bias charge injection control gate electrode 43 by applying _prb , the control gate is opened, and bias charge (holes) is applied to the so-called sensor potential below the gate electrode 26 from the p-type semiconductor region 41. 45 is injected to set the sensor potential to the bias potential V _prb . That is, φ ₂ is the potential after bias charge injection.

FIG. 7 shows the drive timing in the example (I) in which the bias charge injection electrode 44 is provided. In the figure, V
_{g is the} pixel gate voltage, V _gc is the gate voltage of the control gate electrode 43, and V _sub is the substrate voltage. Here, the gate voltage V _g of the pixel MOS transistor during storage
Is high level H (dashed line) and low level L (solid line)
Indicates the type. This is because the gate voltage V _g may be high level H or low level L during storage.
Regarding the gate voltage V _g during this accumulation, the following example (I
The same applies to I) and Example (III).

FIG. 8 shows another example of the bias charge injection method (example (II)). This example is a method of using the electric charge h generated in the imbalance as the bias electric charge. In the bias charge injection phase, the pixel MOS transistor 2
A voltage higher than that in the read operation is applied to the drain region 28 of 9 to cause the avalanche 46, and the bias charge 47 is injected by the holes (h) generated by the avalanche.

FIG. 9 shows the drive timing in the example (II) in which the charge generated by the avalanche is used as the bias charge. In the figure, V _g is the gate voltage of the pixel, V _d is the drain voltage applied to the drain region, and V _sub is the substrate voltage.

10 and 11 show still another example (example (III)) of the bias charge injection method. This example is a method of injecting a bias charge by injection from the substrate. That is, in the bias charge injection phase, the gate electrode 2
6, a sufficiently low voltage is applied to bias bias charge 48 from the substrate.
Is injected (see potential 49 in FIG. 11). In this case, on the contrary, the same can be done by applying a high voltage to the substrate.

FIG. 12 shows the drive timing in the example (III) of injecting the bias charge by utilizing the injection from this substrate. In the figure, V _g selection is the gate voltage of the pixel on the selected line, V _g non-selection is the gate voltage of the pixel on the non-selected line, and V _sub is the substrate voltage. At the timing shown by the solid line, bias charge injection is performed only by V _g selection. The timing shown by the broken line is for V _g selection, V _g non-selection, and bias charge injection at V _sub .

However, in the case of a device such as CMD in which one line is selected and the reading operation is performed, the line selected for reading is selected so as not to inject charges from the substrate into pixels of other lines. It is necessary to selectively inject charges from the substrate. This problem does not occur in the method of applying a sufficiently low potential to the gate electrode. In the case of the method of increasing the substrate voltage, it is necessary to make the gate voltage of the line to be injected lower than the other lines in order to make the injected line have selectivity.

When it is difficult to control the injection amount as in the method of injecting the bias charge by the avalanche,
By performing the incomplete pixel charge reset again after the injection (that is, a part of the bias charges is discarded while the bias charges are slightly over-injected), the controllability of the injection amount of the bias charges is increased.

According to the above-described embodiment, even if the sensor potential of the pixel MOS transistor has unevenness in the gate width direction in the manufacturing process of the solid-state image pickup device, the bias charge is injected to eliminate the potential unevenness and the signal charge is accumulated. By using the difference between the pixel signal read after that and the reference value based only on the bias charge as the output signal, it is possible to eliminate the variation in the characteristics of the incident light amount vs. the output signal between pixels, which adversely affects the image. There is no.

The above-mentioned amplification type solid-state image pickup device is only an example, and it can be applied to other amplification type solid-state image pickup device such as CMD. Further, the present invention can be applied to other solid-state image pickup devices.

[0058]

According to the present invention, by injecting the bias charge after the reset, it is possible to eliminate the unevenness of the potential well in the region where the charge is accumulated by photoelectric conversion.
By accumulating the signal charges with the bias charges injected and reading the pixel signals, it is possible to reduce or eliminate the variation in the input / output characteristics (that is, the characteristics of the incident light amount vs. the output signal) between the pixels. .

By using the difference between the reference value of the bias charge read immediately after the injection of the bias charge and the pixel signal read after the accumulation of the signal charge as the output signal, only the signal corresponding to the signal charge can be output. Therefore, it is possible to reduce or eliminate variations in input / output characteristics between pixels.

When the difference between the reference value due to the bias charge injected after reading the pixel signal and the pixel signal is used as the output signal, noise due to variations in the injection amount of the bias charge can be eliminated.

When the difference between the reference value due to the bias charge injected before the reading of the pixel signal and the pixel signal is used as the output signal, the memory for the reference value can be eliminated.

When the bias charge is injected using the charge injection electrode, the avalanche method, or the injection from the substrate, the bias charge should be injected easily and with good control. You can After the bias charge is injected, an incomplete reset, that is, a part of the injected bias charge is reset, so that the bias charge injection amount can be controlled more accurately.

[Brief description of drawings]

FIG. 1 is a plan view showing an example of an amplification type solid-state imaging device according to the present invention.

FIG. 2 is a sectional view of a pixel MOS transistor of the amplification type solid-state imaging device of FIG. 1;

FIG. 3 is an explanatory diagram showing an operation phase when the electronic shutter is not used in the amplification type solid-state imaging device according to the present invention.

FIG. 4 is an explanatory diagram showing an operation phase when an electronic shutter is used in the amplification type solid-state imaging device according to the present invention.

FIG. 5 is a configuration diagram showing an example (example (I)) of a bias charge injection method.

FIG. 6 is a potential diagram in the method of FIG.

7 is a driving timing chart in the bias charge injection method of FIG.

FIG. 8 is a configuration diagram showing another example (example (II)) of the bias charge injection method.

9 is a driving timing chart in the bias charge injection method of FIG.

FIG. 10 is another example of the bias charge injection method (example (III))
It is a block diagram which shows.

11 is a potential diagram in the method of FIG.

12 is a driving timing chart in the bias charge injection method of FIG.

FIG. 13 is a cross-sectional view showing an example of a pixel of an amplification type solid-state imaging device.

14 is a potential diagram in the depth direction of the pixel of FIG.

FIG. 15 is a three-dimensional potential diagram of a pixel for explaining the concept of potential unevenness.

[Explanation of symbols]

21 amplifying solid-state imaging device, 22 p-type semiconductor substrate, 2
3 n-type overflow barrier region, 24 p-type semiconductor well region, 25 gate insulating film, 26 gate electrode,
27 source region, 28 drain region, 29 pixel M
OS transistor, 31 signal line, 32 vertical select line,
34 Drain power line

Claims (20)

[Claims] 1. A solid-state image pickup device, comprising: after resetting, bias charges are injected into a region for accumulating charges by photoelectric conversion, and signal charges are accumulated to read pixel signals. 2. The solid-state image sensor according to claim 1, wherein the signal charge is accumulated after injecting the bias charge and reading a reference value. 3. The solid-state imaging device according to claim 1, wherein the readout of the pixel signal is nondestructive readout. 4. The solid according to claim 1, wherein a difference between the reference value read out immediately after the injection of the bias charge and the pixel signal read out after the accumulation of the signal charge is used as an output signal. Image sensor. 5. The solid-state image sensor according to claim 1, wherein a difference between a reference value based on the bias charge injected after the pixel signal is read and the pixel signal is output. 6. The solid-state imaging device according to claim 1, wherein a difference between a reference value based on the bias charge injected before the reading of the pixel signal and the pixel signal is output. 7. The solid-state image pickup device according to claim 1, wherein a charge injection electrode is provided to inject a bias charge into the region. 8. The solid-state image sensor according to claim 1, wherein a bias charge is injected into the region by utilizing avalanche. 9. The solid-state image pickup device according to claim 1, wherein a bias charge is injected from the substrate side into the region. 10. The solid-state image sensor according to claim 1, wherein the bias charge injection amount is controlled by performing an incomplete reset after the bias charge is injected. 11. A method for driving a solid-state image pickup device, comprising: after resetting, injecting bias charges into a region for accumulating charges by photoelectric conversion, accumulating signal charges, and reading pixel signals. 12. The method of driving a solid-state image sensor according to claim 11, wherein the signal charge is accumulated after the bias charge is injected to read a reference value. 13. The method for driving a solid-state image sensor according to claim 11, wherein the readout of the pixel signal is nondestructive readout. 14. The solid according to claim 11, wherein a difference between the reference value read out immediately after the injection of the bias charge and a pixel signal read out after the accumulation of the signal charge is used as an output signal. Driving method of image sensor. 15. The method of driving a solid-state image sensor according to claim 11, wherein a difference between a reference value due to a bias charge injected after reading the pixel signal and the pixel signal is output. 16. The method of driving a solid-state image sensor according to claim 11, wherein a difference between a reference value due to a bias charge injected before reading the pixel signal and the pixel signal is output. 17. The method for driving a solid-state image sensor according to claim 11, wherein a charge injection electrode is provided to inject a bias charge into the region. 18. The method for driving a solid-state image pickup device according to claim 11, wherein a bias charge is injected into the region by utilizing avalanche. 19. The method for driving a solid-state image pickup device according to claim 11, wherein a bias charge is injected from the substrate side into the region. 20. The method for driving a solid-state imaging device according to claim 11, wherein the bias charge injection amount is controlled by performing incomplete reset after the bias charge is injected.