JP2551658B2 - Photoelectric conversion device - Google Patents

Description

The present invention relates to a photoelectric conversion device for reducing sensor noise.

[Problems to be Solved by the Related Art and Invention] An amplification type sensor is preferably used for a sensor used in a solid-state imaging device or the like in order to increase an output signal level.

The amplification type sensor is composed of transistors of MOS type, SIT type, FET type, bipolar type, etc., and amplifies the electric charge accumulated in those control electrodes and outputs it from the main electrode. . For example, Japanese Patent Publication Sho 55-
28456 discloses an example of an amplification type sensor. One of the problems of such an amplification type sensor is that the sensor noise is large.

There are two types of sensor noise: fixed pattern noise that appears in a fixed manner (hereinafter referred to as FPN), and random noise that is introduced into the control electrode when the control electrode is reset.

Of the sensor noise, the FPN appears statically, so it can be completely removed by subtracting the dark output of the sensor from the optical signal output of the sensor. The dark output can be obtained by reading the accumulation time to almost zero, that is, immediately after the sensor is reset.

On the other hand, in order to remove the random noise introduced into the control electrode, the sensor output (optical signal) after the accumulation may be subtracted from the sensor output (sensor noise) immediately after the accumulation starts. As a photoelectric conversion device capable of such subtraction processing, Japanese Patent Application No. Sho 63-4749 filed previously by the present applicant has been proposed.
There is one disclosed in No. 2. This photoelectric conversion device is provided with a means for storing an optical signal and a storage means for storing sensor noise, and takes the difference between the optical signal stored in both storage means and the sensor noise.

Such a photoelectric conversion device is very practical when used for a line sensor, but when used for an area sensor, the chip area of the storage means for storing the sensor noise becomes large. Even when the storage means is provided outside the image sensor, a field memory or frame memory is required, which is a cost problem.
Practical application is difficult.

Further, there is one in which a photoelectric conversion region is laminated on an amplification transistor in order to increase the sensitivity of the sensor and relatively reduce the influence of sensor noise. As such a solid-state imaging device, there is one disclosed in Japanese Patent Application No. 1-9089 filed by the present applicant earlier.

In this solid-state imaging device, when the photoconductive film is laminated, the aperture ratio of the photoelectric conversion region becomes very large, so that the sensor output (S) also becomes large. However, in order to achieve a higher S / N ratio, it is necessary to reduce the sensor noise (N).

In view of the above problems, an object of the present invention is to provide a photoelectric conversion device suitable for removing sensor noise.

[Means for Solving the Problems] A photoelectric conversion device of the present invention includes a storage unit that stores photoexcited charges, an amplification unit that amplifies charges of a control electrode,
Switch means for connecting the storage means and the control electrode, a first read means for reading the output of the amplifying means as a first signal when the switch means is not conducting, and a switch means for conducting the switch means. Transfer means for transferring the charge of the storage means to the control electrode, second read means for reading the output of the amplifying means as a second signal after the transfer of the charge, the first signal and the second signal And subtraction processing means for performing the subtraction processing of.

[Operation] In the photoelectric conversion device of the present invention, the switch means is provided between the accumulating means for accumulating the photoexcited electric charge and the amplifying means for amplifying the electric charge of the control electrode. The noise can be read out independently, and it is possible to remove not only the FPN but also the dark current component of the amplification element and random noise.

It is to be noted that when a storage area serving as storage means and an amplification area serving as amplification means are formed on a substrate, a photoelectric conversion area is stacked on the storage area, and the photoelectric conversion area and the amplification area are connected to each other, the aperture ratio of the socket Can be increased and the S / N ratio can be increased.

EXAMPLES Examples of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a partial circuit diagram of a first embodiment of a photoelectric conversion device of the present invention.

FIG. 2 is a schematic plan view of a pixel of the photoelectric conversion device.

FIG. 3 is a sectional view taken along line XX ′ of the photoelectric conversion device of FIG.

FIG. 4 is a cross-sectional view of the photoelectric conversion device of FIG. 2 taken along the line YY ′.

As shown in FIG. 1, one pixel includes a PMOS transistor Tr _A for resetting the pixel, a capacitance Co _X for controlling transient reset of the pixel and signal accumulation / reading,
A photo diode D which is a storage means for accumulating the photoexcited electric charge, an amplifying transistor Tr _C for amplifying a signal from the photo diode D, and a photo diode D.
Transistor Tr _C for amplifying charges generated in the photoelectric conversion portion
It is composed of a PMOS transistor Tr _B which is a switch element for controlling transfer to a base which is a control electrode of the. φ _VR ,
φ _TB is a pulse for controlling the PMOS transistors Tr _A and Tr _B , respectively.

The emitter of the amplifying transistor Tr _C is a MOS transistor
It is connected to the storage capacitors C ₁ and C ₂ via T ₁ and T ₂ . φ _T1 , φ
_T2 is a pulse for controlling the MOS transistors T ₁ and T ₂ , respectively. T _VC is a MOS transistor that resets the vertical signal line, and is set to the potential V _VC by controlling the pulse φ _VC .

As shown in FIGS. 2 and 3, each pixel is separated by a PMOS transistor Tr _A (shown by a broken line in FIG. 3).
On a part of the base B of the amplifying transistor Tr _C, a capacitor Co _X is formed by the poly layer L ₁ provided on SiO ₂ . The emitter E is connected by the Al layer L ₂ as a vertical output line above the pixel separation. As shown in FIGS. 2 and 4, the photodiode D, which is a photoelectric conversion unit, and the base B of the amplification transistor Tr _C are formed by a switch element composed of a PMOS transistor Tr _B (shown by a broken line in FIG. 4). It is separated. It should be noted that, except for the photodiode D, it is normally shielded from light, but is omitted in FIGS.

The operation of the photoelectric conversion device will be described below with reference to FIGS. 1 to 4.

FIG. 5 is a timing chart for explaining the operation of the photoelectric conversion device.

In FIG. 5, first, a period T ₁ is a reset period (f ₀ ) of the horizontal pixel column, and the pulse φ _VR and the pulse φ _TB are set to a negative potential (−V) to set the PMOS transistor Tr _A and the switching element Tr. _{With B} turned on, the horizontal pixel column for which reading has been completed is reset.

Period T ₂ is a horizontal pixel row transient reset period,
When the pulse φ _{VR is set} to a positive potential (+ V), the PMOS transistor Tr _A is turned off, and the potential of the base of the amplifying transistor Tr _C and the photodiode D via the capacitance Co _X.
The electric potential of is decreased. At the same time, the pulse φ _{VC is set} to the high level to turn on the MOS transistor T _VC, and the emitter of the amplifying transistor Tr _C is set to the predetermined potential V _VC . At this time, the residual charges at the base of the photodiode D and the base of the amplifying transistor Tr _C are removed by recombination with the emitter current (this is called transient reset). After that, the pulse φ _{VC is set} to the low level to turn off the MOS transistor T _VC ,
When the pulse φ _VR is changed from the positive potential to the ground potential (GND), the potential of the base of the amplifying transistor Tr _C and the potential of the photodiode D also decrease via the capacitance Co _X. further,
When the pulse _φTB is raised to the ground potential (GND), PMO
The S-transistor Tr _B is turned off, and the photodiode D starts accumulating.

A period T ₃ is a vertical scanning switching period (f ₀ → f ₁ ) and vertical scanning is performed by a vertical shift register (not shown).
The next horizontal pixel column is selected by controlling the pulses φ _V1 and φ _V2 .

The period T ₄ is a residual charge clearing period of the storage capacitor C ₁ ,
When pulse φ _VC and pulse φ _T1 are set to high level, MOS
The transistor T _VC and the MOS transistor T _T1 are turned on, and the residual charge on the storage capacitor C ₁ is cleared. After that, the pulse φ _{VC is set} to the low level, and the MOS transistor T _VC is OF
Set to F state.

The period T ₅ is a period for reading out the sensor noise, and the pulse φ _{VR is set} to a positive potential (+ V), and the amplification transistor is used.
The base of Tr _C is raised, and reading drive is performed from the amplifying transistor Tr _C.

The base potential at this time is the output potential in the dark, and the output voltage that depends on the characteristics of the amplifying transistor Tr _C appears at the emitter. This is commonly called the offset voltage,
Since a plurality of amplifying transistors have slightly different parameters, the emitter output voltage, that is, the offset voltage also differs. Here, the difference between these offset potentials is called sensor noise. This sensor noise is stored in the storage capacitor C ₁ .

The period T ₆ is a period for clearing the residual charge of the storage capacitor C ₂ ,
When pulse φ _VC and pulse φ _T2 are set to high level, MOS
The transistor T _VC and the MOS transistor T _T2 are turned on, and the residual charge on the storage capacitor C ₂ is cleared. After that, the pulse φ _{VC is set} to the low level, and the MOS transistor T _VC is OF
Set to F state.

The period T ₇ is a period for transferring the photocharge of the photodiode D to the base of the amplifying transistor Tr _C , and when the pulse φ _TB is lowered to a negative potential (−V),
The PMOS transistor Tr _B is turned on, and the photocharges accumulated in the photo diode D are transferred to the base of the amplifying transistor Tr _C.

The period T ₈ is a read period of the optical signal, and the pulse φ _VR
The When a positive potential (+ V), the potential of the base of the amplifying transistor Tr _C rises through the capacitor Co _X, the optical signal output is read from the amplifying transistor Tr _C. This optical signal contains the sensor noise described above, and the storage capacitance C ₂
Is accumulated in

The column of the storage capacitor C ₁ that stores the sensor noise and the column of the storage capacitor C ₂ that stores the optical signal are driven by the horizontal shift register (not shown), and the horizontal output line from the horizontal transfer switch (not shown) is selected. Transferred to. Since the output terminal of the horizontal output line is connected to the differential amplifier, as a result, the sensor noise is subtracted from the optical signal, and only the optical signal can be obtained.

The sensor noise subtraction process will be described in detail below.

According to our experimental results, the FPN after transient reset is
The base potential V _B varies by ΔV after the end of the transient reset, depending on the difference in parameters such as h _FE of the amplifying transistor Tr _C. This variation ΔV is due to the read operation and Is done. Here, C _B is a combined capacitance of the base-collector capacitance C _bC , the base-emitter capacitance C _be , and the capacitance C _OX .

Random noise components include kTC noise at the time of reset, which depends on the combined capacitance C _B in the period T ₁ , random noise during the read operation of the sensor transistor, and periods T ₇ , T
₈ is random noise at the time of transferring the photocharge of the photodiode D in FIG. KTC noise at reset is a little smaller due to transient reset However, it is removed by the above subtraction processing.

Random noise during photocharge transfer is band-limited by the base input time constant and is small enough to be ignored.

Further, random noise in a read operation base capacitance C _B, the load capacitance C _V and the storage capacitor C _T as viewed from the emitter, although depends on the current amplification factor h _FE, non-destructive to increase the current amplification factor h _FE It was found that if the degree is increased, it can be made much smaller than the random noise at the time of reset. This means that the FPN and kTC noise of the sensor can greatly improve the S / N ratio by the subtraction processing of the optical signal and the sensor noise. In CCD, the kTC noise due to the reset transistor is removed by the correlated double sampling method after the charge / voltage conversion output amplifier, but the amplifier noise is dominant because of the high speed driving.

The one in which each pixel is composed of the amplifying element as in the present invention is a low speed scan for every 1H, and therefore the amplifier noise, that is, the read noise is extremely small.

Further, the dark current component of the amplifying transistor is also removed when the sensor noise and the optical signal are subtracted.

FIG. 6 is a partial circuit diagram of a second embodiment of the photoelectric conversion device of the present invention.

The characteristic part of this embodiment is that the PMOS transistors Tr _A and Tr _B are connected in series, and the photodiode D is arranged at the connection point. Such an arrangement is preferably used when the horizontal pixel size is larger than the vertical pixel size.

Pixels are reset by setting the pulses φ _VR1 and φ _VR2 to a negative potential and _turning on the PMOS transistors Tr _A and Tr _B.

In the transient reset, the pulse φ _{VR2 is set} to a positive potential to turn off the PMOS transistor Tr _B , the potential of the base of the amplifying transistor Tr _C is raised via the capacitance C _OX, and the emitter of the amplifying transistor Tr _C is switched on. It is performed by setting a predetermined potential.

To store the charge, _connect the pulses φ _VR1 and φ _VR2 to the ground potential (GND).
Then, the PMOS transistors Tr _A and Tr _B are turned off, and charges are accumulated in the photodiode D.

After reading the sensor noise, pulse φ
_{By making VR2} a positive potential and turning off the PMOS transistor Tr _B , raising the base potential of the amplifying transistor Tr _C via the capacitance C _OX , and reading the sensor noise from the emitter of the amplifying transistor Tr _C. Done.

To read the optical signal, pulse φ _VR2 to ground potential (GND)
To turn off the PMOS transistor Tr _B and turn on the pulse φ
_{VR1 is set} to a negative potential, the PMOS transistor Tr _A is turned on, the photocharge of the photodiode D is transferred to the base of the amplification transistor Tr _C , and the pulse φ _{VR1 is set} to the ground potential (GN
D) to turn off the PMOS transistor Tr _A , set the pulse φ _VR2 to a positive potential to turn off the PMOS transistor Tr _B , and turn on the potential of the base of the amplification MOS transistor Tr _C via the capacitor C _OX. Then, the light signal is read out from the emitter of the MOS transistor Tr _C for amplification.

Although the operation method of this embodiment is slightly different from that of the first embodiment, the sensor noise can be removed similarly to the first embodiment.

FIG. 7 is a partial circuit diagram of a third embodiment of the photoelectric conversion device of the present invention.

The characteristic part of this embodiment is that the amplifying element is a MOS transistor.

In the figure, the MOS transistor Tr _A ′ is a reset MOS transistor for clearing the residual charge in the gate region which is the control electrode of the MOS transistor Tr _C ′.
In the first embodiment, since the amplifying element is composed of the bipolar transistor, the transient reset is performed. However, in this embodiment, since the amplifying element is the MOS transistor, the transient reset operation is unnecessary. Random noise when transferring the photocharge accumulated in the photo diode D to the gate and reading the optical signal from the MOS transistor of the amplifying element is the control electrode at the time of reading if MOS or FET is used as the amplifying element of the sensor MOS. Since the gate charge is not destroyed, the correlation between the sensor noise and the sensor noise included in the optical signal output becomes very high, and as a result, a high S / N sensor can be obtained.

FIG. 8 is a sectional view of a sensor according to the fourth embodiment of the present invention.

In this embodiment, the photoelectric conversion region is a laminated photoconductive film.

As shown in the figure, a pixel electrode made of a photoconductive film is connected to the photodiode portion of FIG. 4 which is the first embodiment of the photoelectric conversion device.

In the figure, 1 is a photoconductive film, 2 is a transparent electrode, 3 is a pixel electrode,
4 is an insulating layer, the photoconductive film 1 is through the pixel electrode 3,
It is connected to the p-type semiconductor region that constitutes the photodiode portion.

According to the present embodiment, the photoconductive film can be provided also on the amplification element portion, so that the aperture ratio of the receptacle and the S / N ratio can be increased.

FIG. 9 is a schematic configuration diagram of a solid-state imaging device to which the present invention is applied.

In the figure, the vertical scanning unit 202 and the horizontal scanning unit 203 perform television scanning on the image sensor 201 in which optical sensors are arranged in an area.

The signal output from the horizontal scanning unit 203 is output as a standard television signal through the processing circuit 204.

Driving pulse φ _{HS for} the vertical and horizontal scanning units 202 and 203,
φ _H1 , φ _H2 , φ _VS , φ _V1 , φ _V2, etc. are supplied by the driver 205. The driver 205 is also limited by the controller 206.

[Effects of the Invention] As described in detail above, according to the photoelectric conversion device of the present invention, the switch means is provided between the storage means for accumulating the photoexcited charge and the amplification means for amplifying the charge of the control electrode. As a result, the sensor noise of the amplification element can be independently read regardless of the operation of the storage unit, and not only the FPN but also the dark current component and random noise of the amplification element can be removed, and the amplification element can be driven at low speed. It is possible to realize an image pickup device with a high S / N ratio, with very small noise during reading.

If a storage area serving as storage means and an amplification area serving as amplification means are formed on the substrate, and a photoelectric conversion area is stored thereon, and the photoelectric conversion area and the amplification area are connected to each other, the aperture ratio of the socket And the S / N ratio can be increased.

[Brief description of drawings]

FIG. 1 is a partial circuit diagram of a first embodiment of a photoelectric conversion device of the present invention. FIG. 2 is a schematic plan view of a pixel of the photoelectric conversion device. FIG. 3 is a sectional view taken along line XX ′ of the photoelectric conversion device of FIG. FIG. 4 is a cross-sectional view of the photoelectric conversion device of FIG. 2 taken along the line YY ′. FIG. 5 is a timing chart of the photoelectric conversion device. FIG. 6 is a partial circuit diagram of a second embodiment of the photoelectric conversion device of the present invention. FIG. 7 is a partial circuit diagram of a third embodiment of the photoelectric conversion device of the present invention. FIG. 8 is a sectional view of a sensor according to the fourth embodiment of the present invention. FIG. 9 is a schematic configuration diagram of a solid-state imaging device to which the present invention is applied. Tr _A , Tr _B : PMOS transistor, Co _X : Capacitance, D: Photo diode D, Tr _C : Transistor for amplification, φ _VR , φ _TB , φ _T1 ,
φ _T2 , φ _VC : pulse, T _VC , T ₁ , T ₂ : MOS transistor, C ₁ ,
C _2: capacitor.

Claims (2)

(57) [Claims] 1. Storage means for storing photoexcited charges,
Amplifying means for amplifying the electric charge of the control electrode, a switch means for connecting the accumulating means and the control electrode, and a first readout for reading out the output of the amplifying means of the switch means as a first signal. Means, a transfer means for electrically connecting the switch means to transfer the charge of the storage means to the control electrode, a second read means for reading the output of the amplifying means as a second signal after the transfer of the charge, A subtraction processing unit that performs a subtraction process between a first signal and the second signal, and a photoelectric conversion device. 2. A storage region serving as the storage unit and an amplification region serving as the amplification unit are formed on a substrate, and a photoelectric conversion region is stacked on the storage region, and the photoelectric conversion region and the amplification region are electrically connected to each other. The photoelectric conversion device according to claim 1, wherein the photoelectric conversion device is connected.