JPH05275671A - Phototransistor and image sensor provided therewith - Google Patents

Abstract

PURPOSE:To restrain a signal voltage from being attenuated by the emitter follower action of a buffer transistor by a method wherein the buffer transistor is inserted between the emitter electrode of a phototransistor and an access FET. CONSTITUTION:An image sensor is equipped with phototransistors 21a-21d, buffer transistors 22a-22d, and access FETs 23a-23d. A reset FET is turned ON by reset signals Y2 to reset the emitter electrodes of the phototransistors and the base electrodes of the buffer transistors to the voltage of a reset power supply. When the reset FET is turned OFF and put into a storage time, charge stored between the collector and the base of the phototransistor is discharged as a photocurrent, and the base and the emitter are made to rise in voltage. After a storage time elapses, the access FET 23a is turned ON by access signals Y1, and the emitter voltage of a phototransistor is guided to an output line 28 through the intermediary of a buffer transistor, so that a signal voltage can be restrained from being attenuated.

Description

Detailed Description of the Invention

[0001]

[Industrial application] With the progress of image processing and image communication equipment, the need for high-performance image sensors is increasing. The present invention relates to a phototransistor for forming an image sensor and an image sensor that enables high-speed and high-quality reading of document information.

[0002]

2. Description of the Related Art An image sensor using an integrated circuit uses a photodiode or a phototransistor as a light detecting element. Photocurrent is about 1 in photodiode
nA / lx / mm ² , about 100n for phototransistor
A / lx / mm ² . The phototransistor has a large photocurrent because the sensing element itself has an amplification function. Further, the phototransistor is regarded as a composite element of a photodiode and a transistor, and the signal voltage generated in the photodiode can be output with low impedance by the emitter follower operation of the transistor. As the resolution of the image sensor is increased, the light receiving area becomes smaller and the photocurrent becomes smaller. In addition, in order to improve sensitivity, various image sensors employ a drive system in a charge storage mode in which photocurrent is stored for a certain period of time and is intensively output at a signal output timing. For example, CC
The D image sensor simultaneously guides the optical signal charges accumulated in the photodiodes to the potential well of the CCD at the read timing, and then transfers them to the output amplifier side along the potential well to obtain an image signal. The MOS image sensor charges a photodiode or a phototransistor with a certain amount of electric charge, then discharges it with a photocurrent until the next reading timing, and obtains an accumulated image signal by the next recharging. A bipolar image sensor charges a phototransistor with a certain amount of electric charge, then discharges it with a photocurrent until the next reading timing, amplifies it with a transistor function at the next recharging timing, and then obtains an accumulated image signal. ..

The sensitivity of the photodetector is highest in the storage system using a phototransistor. Planar and cross-sectional structures of a conventional phototransistor are shown in FIGS. This is a structure in which a base diffusion layer 2 having a transistor amplifying function and a photoelectric conversion function is formed on an n-type Si substrate 1, and an n-type emitter diffusion layer 4 is formed on the surface thereof. The diffusion layer is widened to form a photoelectric conversion unit, and the collector-base capacitance is large. FIG. 4 shows an equivalent circuit of a pixel portion of an image sensor using a phototransistor. The photocurrent generated at the base-collector junction discharges the electric charge stored in the base-collector capacitance, and as a result, the collector-base voltage decreases. This voltage change is led from the emitter electrode of the phototransistor 12 to the image output line 19 by conducting the access switch 13 to output an image signal, and subsequently the reset switch 14 is conducted to make the emitter potential of the phototransistor 11 zero. Reset to potential. This method has a simple circuit and is quite sensitive, but has a problem that high-speed reading is difficult because of a large time constant for recharging. Therefore,
Although it can be used as a fax for G3 class fax, the sensitivity and afterimage performance are degraded for G4 class fax due to the delay time from access to signal output and incomplete recharging.

[0004]

Since the phototransistor has a photoelectric conversion portion, the junction capacitance Cbc between the base and collector becomes larger than that of a normal transistor.
Therefore, the sensitivity of the signal voltage represented by photocurrent × accumulation time / Cbc becomes small. Further, in the operation in the charge storage mode, as shown in FIG. 2, Cbc is charged with the base electrode in a floating state. However, a base-emitter junction is interposed in series in the charging circuit, which becomes a non-linear resistance. Therefore, the charging time to the junction capacitance increases extremely with Cbc. Therefore, during high-speed scanning in which the charging time cannot be taken long, recharging is incomplete and the afterimage performance is deteriorated.
Attempts have been made to reduce Cbc by forming a phototransistor on a high-resistance Si substrate, but there is a limit to further increase the resistance of the substrate in terms of manufacturing technology.

[0005]

A phototransistor according to a first embodiment of the present invention has a first amplification function on a Si substrate.
A second base diffusion layer having a photoelectric conversion function is formed with the base diffusion layer, an emitter diffusion layer is formed on the first base diffusion layer, and the surface of the second base diffusion layer is connected to the substrate or is connected to the substrate or the diffusion layer. This is a structure in which a thin diffusion layer of the same conductivity type as that of the connected substrate is formed, and the impurity concentration of the second base diffusion layer is made small so that a second voltage is applied between the collector and the emitter in an operating bias state. Completely deplete the base layer. In the phototransistor according to the second embodiment of the present invention, a first base diffusion layer having an amplifying function and a second base diffusion layer having a photoelectric conversion function are formed on an n-type Si substrate, and an emitter diffusion layer is formed in the first base diffusion layer. And a thin film of SiO ₂ , SiO _x N _1-x , and SiN are formed in this order on the surface of the second base diffusion layer, and the impurity concentration of the second base layer is reduced to between the collector and the emitter. The second base layer is completely depleted in an operating bias state in which a voltage of several V is applied. An image sensor is composed of an array of phototransistors according to the present invention, a buffer transistor, a field effect transistor for access and reset, and a shift register for scanning.

[0006]

The capacitance Cbc of the base-collector junction of the phototransistor of the present invention is the sum of the capacitance of the first base-collector junction and the capacitance of the second base-collector junction, but the second base region is completely depleted. Since it is in the state, its junction capacitance is almost 0, and Cbc is only the junction capacitance due to the first base layer and the collector region, which is extremely small. As a result, the signal voltage sensitivity is high and the recharge speed is high. Therefore, the afterimage performance is not deteriorated even at the time of high-speed reading while having high sensitivity. The image sensor of the present invention can take out the signal voltage appearing at the emitter electrode of each phototransistor to the output line with little attenuation by the emitter follower action of the buffer transistor.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 4 is a well-known equivalent circuit of the pixel portion when the phototransistor is operated in the charge storage mode. The phototransistor 12, the access field effect transistor (hereinafter FET) 13, and the reset FET 1 are shown.
4 and power supply 15. Here, the phototransistor 12 is the photocurrent source 16 and the collector-base capacitance 1
7 and transistor 18. By applying a reset pulse to the gate electrode of the reset FET 14 at an accumulation time interval, the emitter voltage of the phototransistor 12 is reset to 0 or a desired voltage, and an access pulse is applied to the gate electrode of the access FET 13 immediately before each reset. When applied, the signal voltage appearing at the emitter of the phototransistor is output to the output line 19. When the photocurrent generated at the collector-base junction of the phototransistor is IL, the storage time T, and the storage capacitance Cbc, the output voltage Vo becomes IL × T / Cbc. Since the signal voltage decreases as Cbc increases, it is necessary to reduce Cbc to increase the sensitivity of the signal voltage.

A cross section and a planar structure of a phototransistor according to the first embodiment of the present invention are shown in FIGS. an n-type Si substrate 1, a p-type first base diffusion layer 2 formed close to the emitter on the Si substrate, a p-type second base diffusion layer 3 formed so as not to contact the emitter and having a photoelectric conversion action, The n-type emitter diffusion layer 4 is formed on the first base diffusion layer, and the n-type thin diffusion layer 5 is formed on the second base diffusion layer. The thin diffusion layer 5 includes the n-type substrate 1 and its diffusion layer. Connect through. 6 is a protective oxide film, and 7
Is a collector diffusion layer. The second base diffusion layer 3 has a sandwich structure of an n-type substrate 1 and a thin diffusion layer 5 extending from the surface. If the impurity concentration of the second base diffusion layer 3 is lowered, the substrate side and The second base region 3 can be completely depleted by the expansion of the depletion layer from the front surface side. Impurity concentration of n-type Si substrate is 9 × 1
0 ¹⁴ / cm ³ , the impurity concentration of the emitter diffusion layer is 9 × 1
For various impurity concentrations of the base diffusion layer as 0 ¹⁸ / cm ³ , the thickness of the depletion layer in the base diffusion layer of the surface side junction and the thickness of the depletion layer in the base diffusion layer of the substrate side junction are set to a bias voltage of 2V, The results of analysis in the case of 4 V are shown in (Table 1) and (Table 2), respectively. Base impurity concentration 1 ×
When it is 10 ¹⁶ / cm ³ or less, the sum of the thicknesses of the depletion layers of the front surface side junction and the substrate side junction becomes 0.8 μm or more. Therefore,
It is shown that the base layer can be completely depleted by setting the thickness of the base diffusion layer to 0.8 μm or less.

[0009]

[Table 1]

[0010]

[Table 2]

The area of the base layer having a photoelectric conversion function is 70
For a conventional phototransistor of 00 μm ² , Cbc = 1.2
Although it was pF, Cbc = 0.13 pF was obtained in the phototransistor of the same size according to the present invention. Even if the base layer is completely depleted, the photocurrent generated by photoexcitation is almost the same as when it is not depleted, and the output signal voltage is theoretically expressed by IL × Tint / Cbc. Therefore, IL
Is photocurrent and Tint is storage time. Even if the base layer is completely depleted, the photocurrent generated by photoexcitation is considered to be almost the same. As a result of the experiment, the output signal sensitivity of the conventional phototransistor is about 11V /
Although it was Lx.s, the output signal sensitivity of the phototransistor of this example was about 83 V / Lx.s, which was extremely high sensitivity. The afterimage performance was about 7% for the conventional phototransistor, but about 2% for the phototransistor of the present invention.

Although FIG. 1 shows the case of the n-type substrate, the same action and effect can be exhibited even if the conductivity type is completely reversed. That is, after forming the first base diffusion layer and the second base diffusion layer formed of the n-type diffusion layer on the p-type substrate and forming the p-type emitter diffusion layer on the first base diffusion layer,
By forming a thin p-type diffusion layer on the second base diffusion layer, the second base diffusion layer can be completely depleted in the operation bias state. As a result, the effective collector-base capacitance can be significantly reduced as in the phototransistor of FIG.

FIG. 2 shows a cross section and a planar structure of a phototransistor according to the second embodiment of the present invention. This phototransistor is formed on an n-type Si substrate 1, a p-type first base diffusion layer 2 formed on the substrate, a p-type second base diffusion layer 3 acting as a photoelectric conversion part, and a first base diffusion layer 2. The n-type emitter diffusion layer 4, the SiO ₂ layer 8, the SiO _x N _{1 -x} layer 9, and the SiN layer 10 formed on the second base diffusion layer 3 are formed as insulating thin films. When the thickness of SiO ₂ is set to several tens nm and SiN is formed on the surface, SiO _x N _1-x is automatically formed at the SiO ₂ / SiN interface. With this structure, the insulating thin film or the insulating thin film and the p-type second
A large amount of positive charges (> 10 ¹² / cm ² ) are present at the interface of the base diffusion layer 3 as compared with the case of a normal MOS transistor, and when the impurity concentration on the surface of the base diffusion layer 3 is low,
Due to this positive charge, the inversion layer 11 is formed on the surface of the base diffusion layer 3.
Is formed. In order to easily form the inversion layer 11, the impurity concentration of the base layer must be 1 × 10 ¹⁶ / cm ² or less. The second base diffusion layer 3 has a sandwich structure of an n-type substrate and an inversion layer 11. If the impurity concentration is lowered, depletion from the substrate side and the surface side is only required by applying a voltage of several V or more between the collector and the substrate. The second base layer 3 can be completely depleted by the spread of the layer. Compared to the first embodiment, the present embodiment exists in the insulating film or at the interface between the insulating film and the Si, and since the inversion layer 11 is formed by the positive charge determined by the composition or the surface treatment state, the junction is shallow and uniform. .. Therefore, the sensitivity of the phototransistor at a short wavelength is significantly improved as compared with that of the first embodiment.

FIG. 5 is an equivalent circuit of an image sensor formed by using a phototransistor according to the present invention, which is characterized in that a signal voltage generated by the phototransistor is taken out to an output line with reduced attenuation. Each pixel includes a phototransistor 21a to 21d, a buffer transistor 22a to 22d, an access FET 23a,
.About.23d and reset FETs 24a to 24d. Note that 25a to 25d are parasitic capacitances associated with the emitter electrode of the phototransistor and the base electrode of the buffer transistor. 26 is a reset power supply, 2
Reference numeral 7 is a scanning shift register, 28 is an output line, 29 is a reset FET for the output line 28, and 30 is a positive power supply line. The scanning shift register operates by a start signal ST and a clock signal CK from the outside, and generates scanning signals Y1 to Y4 and Y5 at its parallel output terminals. Note that Ext is a scanning end signal terminal, and when a plurality of image sensor chips are continuously connected to form a multi-chip image sensor, the Ext signal of the preceding chip is used as the ST signal of the succeeding chip. Scanning signals Y1 to Y4 are sequentially applied to the gate electrodes of the access FETs,
Sequential scanning signal Y is applied to the gate electrode of each reset FET.
2-Y5 are applied. The operation of the leftmost pixel will be described. The reset signal Y2 causes the reset FET to become conductive, and resets the emitter electrode of the phototransistor and the base electrode of the buffer transistor to the voltage value of the reset power supply. Next, when the reset FET is made non-conductive to enter the accumulation time, the electric charge accumulated between the collector and the base of the phototransistor is discharged by the photocurrent, and the voltage of the base electrode and the emitter electrode of the phototransistor rises. After the storage time, the access FET 23a is turned on by the access signal Y1 and the emitter voltage of the phototransistor is guided to the output line 28 via the buffer transistor. In general, since the output line is commonly connected to all pixels, the line capacitance becomes tens to several tens of pF, and when the signal voltage appearing at the emitter electrode of the phototransistor is taken out to the output line through the direct access FET, Due to the large capacitance of the line, the output signal voltage is greatly attenuated and the rising speed of the signal waveform is also reduced. According to the present invention, by inserting the buffer transistor between the emitter electrode of the phototransistor and the access FET, the attenuation of the signal voltage is minimized by the emitter follower action of the buffer transistor, and the slew rate of the output signal voltage is increased. it can.

[0015]

The phototransistor of the present invention can completely deplete the base layer that performs photoelectric conversion in the operating bias state, and can significantly reduce the base-collector capacitance. Therefore, the sensitivity and slew rate of the signal voltage output to the emitter electrode of the phototransistor and the afterimage performance are significantly improved. Further, the image sensor including the phototransistor array, the buffer transistor, the switching FET, and the shift register has a high signal voltage sensitivity, and a high S / N can be obtained even with a low exposure amount. Therefore, the present invention is useful as a photodetector operating in the charge storage mode and an image sensor for reading a document, and its industrial effect is great.

[Brief description of drawings]

FIG. 1 is a plan view and a sectional view of a phototransistor according to a first embodiment of the present invention.

FIG. 2 is a plan view and a sectional view of a phototransistor according to a second embodiment of the present invention.

FIG. 3 is a plan view and a sectional view of a phototransistor in a conventional example.

FIG. 4 is an equivalent circuit diagram of a phototransistor operating in a charge storage mode.

FIG. 5 is an equivalent circuit diagram of an image sensor using a phototransistor according to the present invention.

[Explanation of symbols]

1 Si substrate 2 First base diffusion layer 3 Second base diffusion layer 4 Emitter diffusion layer 5 Thin diffusion layer 6 Protective oxide film 7 SiO ₂ layer 8 SiO _x N _1-x layer 9 SiN layer 10 Inversion layer 11 Phototransistor 12 Access FET 13 for reset 14 Positive power supply 15 Photocurrent 16 Collector-base capacitance 17 Transistor 18 Output line 21a-21d Phototransistor 22a-22b Buffer transistor 23a-23b Access FET 24a-24b Reset FET 27 Scan shift register 28 Image signal output line

Claims (6)

[Claims] 1. A phototransistor comprising a Si substrate, a substrate serving as a collector, a base layer formed on the substrate by a diffusion method, and an emitter layer formed on the base layer by a diffusion method,
The base layer is composed of a first base diffusion layer having an amplifying function around the emitter and a second base diffusion layer having a photoelectric conversion function,
In addition, a thin diffusion layer having the same conductivity type as the substrate electrically connected to the substrate is added to the surface of the second base diffusion layer, and the second base layer is completely depleted in the operation bias state. Transistor. 2. The Si substrate is n-type and the first base diffusion layer is
Pure substance concentration 10¹⁷-10 ¹⁸/ Cm³Second p-type layer of
Base diffusion layer has an impurity concentration of 10¹⁴-10¹⁶/ Cm³P
The phototransistor according to claim 1, which is a mold layer.
Star. 3. The Si substrate is p-type and the first base diffusion layer is not
Pure substance concentration 10¹⁷-10 ¹⁸/ Cm³The n-type layer of the second
Base diffusion layer has an impurity concentration of 10¹⁴-10¹⁶/ Cm³N
The phototransistor according to claim 1, which is a mold layer.
Star. 4. A p-type base in a phototransistor comprising an n-type Si substrate, a p-type base layer formed on the substrate by a diffusion method, and an n-type emitter layer formed on the base layer by a diffusion method using the substrate as a collector. The layer is composed of a first base diffusion layer having an amplifying function around the emitter and a second base diffusion layer having a photoelectric conversion function, and SiO 2 is formed on the surface of the second base diffusion layer.
₂ , SiN _X O _1-X and SiN thin films are provided, and the inversion layer is formed on the second base surface by the positive charges existing in the thin films and at the Si / thin film interface, and the second base layer is completely depleted under the operating bias condition. A phototransistor characterized by being put into a state. 5. The average impurity concentration of the second base layer is 1 × 10.
¹⁴~ 3 x 10¹⁵/ Cm ³Claims characterized in that
4 phototransistor. 6. An array of phototransistors according to claim 1 or 4, a buffer transistor in which a base electrode is connected to an emitter electrode of each phototransistor,
It consists of an access field effect transistor connected between the emitter of each transistor and the output line, a reset field effect transistor for resetting the emitter potential of the phototransistor, a reset power supply and a scanning shift register. An image sensor which outputs an image signal from an output line while reducing the attenuation of the signal voltage appearing in the.