JPH04342178A - Pyroelectric solid-state imaging device - Google Patents

Abstract

PURPOSE:To improve the yield of a pyroelectric type solid-state imaging device by arranging each read voltage setting element in such a manner that a pair of charges are transferred to transfer units, which are adjacent to each other or which are distant from each other with several units therebetween, in the same signal charge transfer element. CONSTITUTION:In order to develop a short circuit across both ends of a pyroelectric element which, in turn, causes the element to be set to an initial voltage 'Vinit', reset transistors MR1 and MR2 are connected. Electric charges stored in diodes D1 and D2, which act as a capacitive element, are transferred to transfer units 20-3 to 20-1 in a CCD by way of read transistors MT1 and MT2, respectively. Here, the transfer units 20-1 and 20-3, in which the charges read from the diodes are initially stored, may be disposed adjacently to each other. In such a case, the charges should be scanned in such a way that the charges, read from the diode D1, are transferred from 20-3 to an empty unit 20-4, and the charges, read from the diode D2, are then transferred from 20-1 to 20-3, thereby improving the yield of a pyroelectric type solid-state imaging device.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a solid-state imaging device, and more particularly to a solid-state imaging device for capturing infrared images.

0002

2. Description of the Related Art Unlike a visible light sensor, a pyroelectric infrared sensor forms an infrared image using an optical lens made of an infrared optical material such as Ge. Infrared rays are absorbed as heat by the pyroelectric film, creating a temperature change distribution on the film surface that corresponds to the infrared image. Since the polarization of the pyroelectric film decreases as the temperature rises, the pyroelectric effect converts the temperature change distribution into the surface charge distribution of the film. Since the pyroelectric effect is caused by temperature changes over time in a dielectric material, it is necessary to provide a chopper in front of the solid-state image sensor that cuts off the infrared light over time in order to image a stationary object. When the temperature of the pyroelectric film changes by ΔT due to temporal interruption of infrared light, the amount of charge ΔQ induced on the film surface becomes PT·ΔT, where PT is the pyroelectric coefficient. Since ΔQ is due to an increase or decrease in polarization, the absolute value of the amount of charge induced on both sides of the pyroelectric film is equal and the signs are opposite.

FIG. 6 is a diagram equivalently showing a pixel section of a conventional pyroelectric solid-state imaging device for capturing infrared images. A pyroelectric element is represented by a capacitor and a current source. Reset transistors MR1 and MR2 are connected to short-circuit both ends of the pyroelectric element to the initial voltage Vinit. Diode D integrates the signal as a capacitive element
1, the charge accumulated in D2 is transferred to the readout transistor MY
The signals are transferred to the CCD via the coupling section under the control of the gates YG1 and YG2 of 1 and MY2. 20-1, 20-2 are CC
It shows the transfer unit of D.

FIG. 8 is a cross-sectional view along the reset transistor MR1, diode D1, readout transistor MY1, coupling portion MC, and CCD connected to one end of the pyroelectric element.

FIG. 9 is a sectional view of a similar structure at the other end of the pyroelectric element.

As described above, this conventional example has two signal transfer mechanisms for each pixel in order to read out signals from the pyroelectric element. 1 is a p- type semiconductor substrate, 2 is an n- diffusion layer for the CCD channel, 3-1 is an n+ diffusion layer for the reset drain RD which also serves as wiring, and 3-3 is a signal integration layer connected to the lower surface of the pyroelectric film 12. n+ type diffusion layer for forming diode D1,
3-4 is an n+ type diffusion layer for forming a signal integrating diode D2 connected to the upper surface of the pyroelectric film 12. 4-1 and 4-2 respectively connect the lower and upper surfaces of the pyroelectric film 12 to the initial voltage Vin.
Reset transistor MR1 for setting it to
The channel P-type diffusion layers 9-1 and 9-2 of MR2 are gate electrodes of MR1 and MR2, respectively. 5 is a p+ type diffusion layer for channel stopper, 6 is SiO2 for green separation, 8
-1 and 8-2 are the read voltages Vr of D1 and D2, respectively.
Read gates YG1, YG2 for setting ead
, 21 is a coupling gate CG for combining two signal flows from diodes D1 and D2 and transferring them to CCD channel 2, and is a Y-shaped coupling gate CG that slightly overlaps with YG1, YG2 and CCD gate 7. This is the gate electrode. 11-4 and 11-5 are the first Al layers connected to YG1 and YG2, respectively, 10 and 14 are interlayer insulating films such as SiO2 or polyimide, 13 is a heat absorption film such as nichrome, and 15 is a wiring and This is the second Al layer for light shielding.

FIG. 10 is a pulse timing diagram explaining the operation of the conventional example, and FIGS. 11 and 12 are pulse timing diagrams for explaining the operation of the conventional example.
This is the surface potential corresponding to FIGS. 8 and 9 at t5. The shaded area indicates that it is filled with electrons. Below, we will discuss the case where polarization treatment is performed so that when the chopper is opened and incident infrared light is applied to the element surface and the temperature of the pyroelectric film rises, negative charges are induced on the bottom surface of the pyroelectric film 12 by the pyroelectric effect. Explain the operation.

As shown at time t1 in FIG. 10, while the chopper changes to the open state during the vertical blanking period (V-BLK), voltage VGH is applied to the reset gate (RG) to turn on MR1 and MR2. Then, both sides of the pyroelectric film 12 are short-circuited and the voltage is set to Vinit. When voltage VGL is applied to RG to turn MR1 and MR2 into the OFF state, the channels under the gates of YG1 and YG2 are in the OFF state, so the signal integration diodes D1 and D2, which are in the floating state, do not integrate the signal. Start. Since negative charges are induced on the lower surface of the pyroelectric film and positive charges are induced on the upper surface due to temperature rise, at time t2 in FIG. 10, the potential of D1 decreases and the potential of D2 increases as shown in FIGS. 11 and 12. Time t
3, Y is used to transfer charge from D1 to CCD20-1.
Apply VYH and VCH to G1 and CG, respectively, to open the channel under the gate. The amount of charge Qread at this time is Vi
Bias charge amount Qb determined by nit and Vread
ias and the amount of charge Qsig induced on the lower surface of the pyroelectric film 12, resulting in Qbias+Qsig. Vread is determined by the gate applied voltage VYH and the threshold voltage of YG1. When the voltage of the diode becomes Vread and the charge transfer ends, VYL and YG1 and CG are respectively applied.
Apply VCL and close the channel under the gate. Next, C
Drive the CD to shift by 1 bit, and charge Qread
moves from the CCD 20-1 to the adjacent transfer unit 20-2 in FIG. At time t4, the channels of YG2 and CG are opened, and charges are transferred from diode D2 to CCD 20-1. The amount of charge Q'read at this time becomes Qbias-Q'sig due to the amount of charge Q'sig induced on the upper surface of the pyroelectric film. After closing the channels of YG2 and CG, at time t5, MR1 and MR2 are turned on as at time t1, and the voltages of diodes D1 and D2 are reset to Vinit. When the chopper is closed to block infrared light, the temperature of the pyroelectric film decreases, and a charge of the opposite sign is induced on the surface of the pyroelectric film. During this signal integration, the charges transferred to the CCD are sequentially scanned and taken out from the image sensor. At time t3, YG1 and CG channels were opened, but at time t6, YG
2. Open the CG channel and read out the charge of diode D2 first. Since the chopper is closed, the amount of charge Q"read at this time is the amount of negative charge Q"sig on the top surface of the pyroelectric film.
is induced, resulting in Qbiast+Q″sig.The CCD is shifted by 1 bit, and at time t7, the channels of YG1 and CG are opened and the charges are read out.The amount of charge at this time is Q′″rea
d becomes Qbias-Q'''sig due to the amount of positive charge Q'''sig induced on the lower surface of the pyroelectric film. In this way, if a field that closes the YG1 channel first and a field that opens the YG2 channel first are provided in correspondence with the cycle of the chopper during signal readout, the image sensor can detect the Qbia
s+Qsig or Qbias+Q″sig, followed by Qbias−Q′sig or Qbias−Q′″s, respectively.
A signal string corresponding to ig can be output. Therefore, by subtracting the subsequently outputted Q″read or Q′″read from Qread or Q′read by an external circuit, Q
sig+Q'sig or Q''sig+Q'''sig is obtained, and the bias charge Qbias can be removed for each pixel.

The structure and driving method of such a pyroelectric solid-state imaging device are described in, for example, Japanese Patent Application No. 1-242016 and Japanese Patent Application No. 2-117588.

0010

[Problems to be Solved by the Invention] In this conventional example, both the charges induced on the upper and lower surfaces of the pyroelectric film are used as signals, and the signal output from the image sensor is first output from the next one, regardless of the opening/closing period of the chopper. Although it has the advantage that the bias charge can be canceled for each unit pixel by pulling the signal output from Since the structure is complicated, the yield is reduced. An object of the present invention is to provide a solid-state imaging device in which the structure and operation of a pixel section are simplified.

[0011]

[Means for Solving the Problems] In order to achieve the above object, the present invention stores charges induced on the upper and lower surfaces of a pyroelectric film in separate capacitive elements, and sets an initial voltage setting element for each capacitive element. , and a read voltage setting element,
Each read voltage setting element is provided so that the charges forming the pair are transferred to transfer units adjacent to each other or separated by a plurality of units of the same signal charge transfer element.

0012

[Operation] As a result, the coupling part MC as shown in FIG. 7 is no longer necessary, and the readout gate does not need to be divided into YG1 and YG2, so the structure of the pixel part and the readout operation are simplified.
Yield is improved.

[0013]

EXAMPLES The present invention will be explained below using examples. FIG. 1 is an equivalent circuit diagram of a unit pixel of a pyroelectric solid-state imaging device according to the present invention. By shorting both ends of the pyroelectric element, the initial voltage Vin
Reset transistor MR1, MR to make it
2 are connected. Diode D1 as a capacitive element
, D2 is transferred to the readout transistor MT1.
, MT2 to CCD transfer unit 20-3, 20, respectively.
-1. There is another transfer unit 20-2 between the CCDs 20-1 and 20-3.

FIG. 2 is a cross-sectional view along the reset transistor MR1, diode D1, readout transistor MT1, and CCD20-3 connected to one end of the pyroelectric element. FIG. 3 is a cross-sectional view of a similar structure at the other end of the pyroelectric element. That is, in order to read out signals from the pyroelectric element, each pixel has two signal transfer mechanisms. 3-1 is an n+ diffusion layer for reset/drain RD which also serves as wiring, 3-3 and 3-4 are pyroelectric films 1, respectively.
Signal integrating diodes D1 and D connected to the bottom and top surfaces of 2
This is an n+ diffusion layer for 2. 4-1 and 4-2 are P-type diffusion layers for channels of transistors MR1 and MR2, respectively;
is the gate electrode of the CCD, and the second layer of polycrystalline silicon that slightly overlaps the first layer of polycrystalline silicon makes the CCD
constitutes a transfer unit. 8 and 9 are third layer polycrystalline silicon for transfer gate TG and reset gate RG, respectively. FIG. 4 is a pulse timing diagram for explaining the operation, and FIGS. 5 and 6 are time tA, tB, tc, t in FIG.
These are the surface potentials corresponding to FIGS. 2 and 3 in D.

Hereinafter, when the chopper is opened and incident infrared light is applied to the element surface and the temperature of the pyroelectric film rises, polarization occurs such that negative charges (electrons) are induced on the lower surface of the pyroelectric film 12 due to the pyroelectric effect. The operation when processing will be explained.

As shown at time tA in FIG. 4, the voltage VGH is applied to the reset gate RG while the chopper changes from the closed state to the open state during the vertical blanking period (V.BLK).
is applied to turn on transistors MR1 and MR2, and both sides of the pyroelectric film are short-circuited to increase the voltage Vinit as shown in FIG.
Set to . Next, voltage VGL is applied to reset gate RG to turn off transistors MR1 and MR2, and the channel under the transfer gate is also turned off by applying voltage VTL.
Due to the FF state, the diodes D1 and D2 are in a floating state, and at time tB in FIG. 3, signal integration starts as shown in FIG. 5. Since negative charges are induced on the lower surface of the pyroelectric film and positive charges are induced on the upper surface, the potential of the diode D1 decreases and the potential of the diode D2 increases. Next vertical retrace period (V
At time tc of (BLK), VTH is applied to the transfer gate TG to transfer charges from the diodes D1 and D2 to the CCD transfer units 20-3 and 20-1, respectively. The read charge amount Qread from the diode D1 is Vinit and V
Qbia is determined by the bias charge amount Qbias determined by read and the charge amount Qsig induced on the lower surface of the pyroelectric film.
s+Qsig, and the read charge amount Qread' from the diode D2 similarly becomes Qbias-Qsig'. Vread is determined by the voltage VTH applied to the TG and the threshold voltage of the read transistor. When the voltage of the diode becomes Vread and signal reading is completed, VTL is applied to TG to close the channel and the next time tD
Then, transistors MR1 and MR2 are turned on and set to the initial voltage Vinit again. When the chopper is closed, the temperature decreases because the incident infrared light disappears, and a charge with a sign opposite to that at time tB is induced on the surface of the pyroelectric film. During this signal integration, the charges of CCD 20-3 and 20-1 are increased by 20
-4, 20-1, etc., and then taken out from the image sensor.

The external circuit detects this amount of charge and calculates Qread-Q'read for each unit pixel for the charge accumulated during the open period of the chopper, and Q'read-Qread for the opposite period. By doing so, the bias charge Qbias is canceled and a signal Qsig+Q'sig from the unit pixel can be obtained.

Although the present invention has been described above based on embodiments, it can be modified without departing from the gist thereof. For example, in FIGS. 2 and 3, the wiring for the reset drain RD is also made of an n+ diffusion layer, but it may also be made of the first Al layer. All the conductivity types of the semiconductors may be reversed, or the polarization direction of the pyroelectric film may be reversed. The CCD 20-1 and the CCD 20-3, which store the read charges first, may be adjacent to each other. In this case, from CCD20-3 to empty CCD
After transferring Qread to CCD 20-4, charges may be scanned to transfer Q'read from CCD 20-1 to CCD 20-3. In addition, the transfer gate TG is a CCD
The first or second polycrystalline silicon layer of the gate can be extended and used instead. In this case, a P-type diffusion layer for threshold voltage control is formed in the transfer gate region, and a voltage higher than the normal CCD drive voltage is applied only during signal readout to increase the channel potential of the transfer gate to Vre.
All you have to do is make it ad. Therefore, in this case TG
This eliminates the need for additional wiring.

[0019]

[Effects of the Invention] As explained above, conventionally, it is necessary to drive the three gates YG1, YG2, and CG independently for signal readout, and the readout operation and the structure of the pixel section are complicated, resulting in a reduction in yield. However, according to the present invention, the structure is simplified because charge is transferred to the CCD without going through a coupling part, and the operation is simpler as shown in FIG. 4 compared to FIG. 7. This makes it possible to improve yield.

[Brief explanation of drawings]

FIG. 1 is an equivalent circuit diagram of a unit pixel of a pyroelectric solid-state imaging device of the present invention.

FIG. 2 is a cross-sectional view of the pyroelectric solid-state imaging device of the present invention.

FIG. 3 is a cross-sectional view of the pyroelectric solid-state imaging device of the present invention.

FIG. 4 is a pulse timing diagram for explaining the operation of the present invention.

FIG. 5 is a surface potential diagram for explaining the operation of the present invention.

FIG. 6 is a surface potential diagram for explaining the operation of the present invention.

FIG. 7 is an equivalent circuit diagram of a unit pixel of a conventional pyroelectric solid-state imaging device.

FIG. 8 is a cross-sectional view of a conventional pyroelectric solid-state imaging device.

FIG. 9 is a cross-sectional view of a conventional pyroelectric solid-state imaging device.

FIG. 10 is a pulse timing diagram for explaining conventional operation.

FIG. 11 is a surface potential diagram for explaining conventional operation.

FIG. 12 is a surface potential diagram for explaining conventional operation.

[Explanation of symbols]

MR1 reset transistor MR2 reset transistor Vinit initial voltage D1 Diode D2 Diode 20-1 CCD transfer unit 20-2 CCD transfer unit 20-3 CCD transfer unit 20-4 CCD transfer unit MT1 Read transistor

Claims (1)

[Claims] Claim 1: Consists of a group of capacitive elements that store optical information on the main surface of a semiconductor substrate of a first conductivity type, an initial voltage setting element for the capacitive elements, and a pyroelectric element formed of a dielectric material exhibiting pyroelectricity. In a solid-state imaging device in which a signal charge transfer element that forms a group of photoelectric conversion elements and sequentially transfers signal charges accumulated in the capacitive element is focused, the charge induced on the upper surface of the film of the pyroelectric element and the Charges induced on the lower surface of the film of the pyroelectric element are stored in separate capacitive elements, and each capacitive element is provided with an initial voltage setting element and a read voltage setting element, and each read voltage setting element is connected to the pair described above. 1. A pyroelectric solid-state image pickup device, characterized in that the charge generated by the signal charge transfer device is transferred to transfer units adjacent to each other or separated by a plurality of units of the same signal charge transfer device.