JPH0462471B2 - - Google Patents

Description

[Detailed description of the invention]

[Industrial Application Field] The present invention relates to a long image sensor unit used in facsimile machines, digital copying machines, and the like. [Prior Art] It is generally well known that photo sensors are used as photoelectric conversion elements in image information processing devices such as facsimile machines, digital copying machines, and character reading devices. In particular, in recent years, photo sensors have been arranged one-dimensionally to form a long photo sensor array, and this has been assembled with a light source array for illuminating the document to be read and an imaging array for forming an image of the document on the photo sensor array. Highly sensitive image reading has also been carried out using a long image sensor unit. As an example of a photo sensor conventionally used for such image reading, there is a sandwich type photo sensor in which a pair of electrode layers are formed on both sides of a photoconductive layer containing amorphous silicon (hereinafter referred to as a-Si) or the like. I can give an example of what is called. However, since this photo sensor is of a type in which the primary photocurrent generated in the photoconductive layer by incident light is extracted as a signal, the signal output is relatively small. Further, since this photo sensor has a structure in which electrode layers are located on both sides of the photoconductive layer, shoots will occur if a pinhole is formed in the photoconductive layer during manufacturing. Therefore, in recent years, a photo sensor called a planar type has been used, which has a pair of electrodes formed on the same surface of a photoconductive layer containing a-Si or the like with a gap that forms at least a part of the light receiving part. It's coming. Since this photo sensor is of a type that extracts the secondary photocurrent in the photoconductive layer as a signal, it has the advantage that the signal output is larger than that of the sand-gerch type photo sensor. By the way, plasma CVD is a method for manufacturing a-Si that constitutes such a planar type photo sensor.
method, reactive sputtering method, ion plating method, etc., and in all of them, the reaction is promoted by glow discharge. However, in any case, in order to obtain a high-quality a-Si film with high photoconductivity, it is necessary to form the film with a relatively low discharge power. However, the photoconductive layer obtained by film formation using such a low discharge power does not have sufficient adhesion to the substrate made of glass or ceramic, and may be difficult to adhere to during the subsequent photolithography process during electrode formation. There was a problem that the film was easily peeled off. Therefore, conventionally, in order to prevent film peeling, a method has been adopted in which a-Si is deposited after roughening the surface of the substrate. That is, the surface of the substrate is chemically etched with, for example, hydrofluoric acid, or physically rubbed with, for example, a brush. However, such a method has the following drawbacks. (1) When using chemicals such as hydrofluoric acid, the equipment in the cleaning line becomes complicated and expensive. (2) It is difficult to control the degree of unevenness on the substrate surface. (3) Microscopic defects are likely to occur when the substrate surface is roughened, and because the characteristics of the a-Si film deposited on the microscopic defects are different, variations in characteristics are likely to occur. Therefore, when constructing a long image sensor unit using the above-mentioned photo sensors, since the variations in each bit signal are large, it is necessary to attach a correction circuit to compensate for the variations, which causes an increase in costs. It was becoming. [Objective] In view of the above-mentioned prior art, the present invention improves the uniformity of the characteristics of each photo sensor in a long photo sensor array using photo sensors, thereby reducing signal dispersion between bits, and improving the correction circuit. The purpose of the present invention is to provide a low-cost long image sensor unit that does not require [Means for Solving the Problems] According to the present invention, the above object is such that the refractive index of the photoconductive layer of each photo sensor changes continuously in the thickness direction in at least a part of the photoconductive layer. This is achieved by a long image sensor unit characterized in that the photoconductive layer near the substrate surface has a refractive index of 3.2 or less for light with a wavelength of 6328 Å. [Example] Hereinafter, specific examples of the present invention will be described. In this specification, a layer of the photoconductive layer that is close to the substrate surface is sometimes referred to as an a-Si subbing layer, and one or more layers thereon are sometimes referred to as an a-Si layer. In the photoconductive layer of the present invention, it is preferable that a portion where the refractive index changes continuously in the film thickness direction is formed between the a-Si subbing layer and the layer immediately above it. Further, it is preferable that the a-Si layer includes a layer with high photoconductivity. In the photo sensor of the present invention, the photoconductive layer has a refractive index that changes continuously in at least a portion thereof in the thickness direction, and the refractive index near the substrate surface of the photoconductive layer is 6328 Å. 3.2 or less, but such a photoconductive layer can be used under conditions such as discharge power, substrate temperature, raw material gas composition, and raw material when performing glow discharge in methods such as plasma CVD, reactive sputtering, and ion plating. It can be formed by appropriately setting gas pressure and the like. FIG. 1 is a partial plan view of a photo sensor in one embodiment of the elongated image sensor unit of the present invention, and FIG. 2 is a sectional view thereof. In the figure, 1 is a base. 2 is an a-Si subbing layer, 3 is an a-Si layer, and these constitute a photoconductive layer. 4 is an n ⁺ layer which is an ohmic contact layer. 5 is a common electrode, and 6 is an individual electrode. In addition, in FIG. 2, the a-Si subbing layer 2 and the
Although the boundary with the -Si layer 3 is clearly illustrated, in reality, this boundary is a layer whose refractive index changes continuously and has properties intermediate between both layers. As the substrate 1, glass such as #7059 manufactured by Corning, #7740 manufactured by Corning, SCG manufactured by Tokyo Ohka Co., Ltd., quartz glass, ceramic such as partially glazed ceramic, etc. can be used. The photoconductive layer is made of an amorphous material containing a-Si as a main component, and the a-Si subbing layer 2 has a refractive index of 3.2 or less. Further, the a-Si layer 3 has a refractive index greater than 3.2, preferably about 3.4. The photoconductive layer is plasma
It is formed by methods such as CVD, reactive sputtering, and ion plating, especially plasma CVD. In the photoconductive layer thus formed, stress is generated due to hydrogen taken in during layer formation, and if this stress is too large, the adhesion with the substrate deteriorates and the film is likely to peel off. The magnitude of stress in the photoconductive layer can be controlled by appropriately setting the conditions during layer formation, such as glow discharge discharge power, substrate temperature, source gas composition, and source gas pressure. and a- adjacent to the base body 1;
Good adhesion to the substrate 1 can be maintained as the Si subbing layer 2 by forming a layer with low stress, for example, by performing glow discharge with a relatively large discharge power. On the other hand, stress in the photoconductive layer has a strong correlation with the refractive index of the layer, and it is generally known that the lower the stress, the lower the refractive index. It has also been found that in order to improve the photoconductivity of the photoconductive layer, it is necessary to perform glow discharge at a relatively low discharge power. Therefore, glow discharge is first performed with a relatively large discharge power on the substrate 1 to form an a-Si subbing layer 2 having a relatively small refractive index, for example, a refractive index of 3.2 or less.
After forming, the deposition is continued while the discharge power is gradually reduced to a relatively small power to form a part where the refractive index continuously changes in the film thickness direction, and then the discharge power is further reduced to a relatively low power. Glow discharge is performed to generate a material with a relatively large refractive index,
It is preferable to form an a-Si layer 3 having a high photoconductivity of about 3.4. The common electrode 5 and the individual electrodes 6 are made of a conductive film such as Al. Hereinafter, the present invention will be explained in detail with reference to Examples. Example 1: A double-sided polished glass substrate (#7059 manufactured by Corning Incorporated) was subjected to ordinary cleaning using a neutral detergent or an organic alkaline detergent. Next, the glass substrate 1 was covered with a mask of a desired pattern and set in a capacitively coupled glow discharge decomposition apparatus, and a 1×
It was maintained at 230°C under an evacuation vacuum of 10 ^-6 Torr. Then, epitaxial grade pure _SiH4 is placed in the device.
Gas (manufactured by Komatsu Electronics) was introduced at a flow rate of 10 SCCM, and the gas pressure was set at 0.07 Torr. after that,
Using 13.56MHz high frequency power supply, input voltage 2.0kV,
Glow discharge was performed for 2 minutes at an RF (Radio Frequency) discharge power of 120 W to form an a-Si subbing layer 2 with a thickness of about 400 Å. Next, the input voltage was gradually lowered and set to 0.3 kV after 5 minutes. Then the input voltage
Glow discharge was performed at 0.3 kV and discharge power of 8 W for 4.5 hours to form an a-Si layer 3 with a thickness of about 0.85 μm. followed by _SiH4 diluted to 10% with _H2 and _H2
Using a gas mixed with PH ₃ diluted to 100ppm at a mixing ratio of 1:10 as a raw material, the n ⁺ layer (approximately 0.15μ thick), which is an ohmic contact layer, is formed at a discharge power of 30W.
was deposited. Next, Al
was deposited to a thickness of 0.3μ to form a conductive layer. Next, a positive photoresist (manufactured by Shipley) was applied.
After forming a photoresist pattern in the desired shape using AZ-1370), phosphoric acid (85% aqueous solution by volume), nitric acid (60% by volume aqueous solution), glacial acetic acid, and water were added.
A liquid mixed at a volume ratio of 16:1:2:1 (hereinafter referred to as
The exposed portions of the conductive layer were removed using etching solution 1) to form a common electrode 5 and individual electrodes 6. Next, using a plasma etching method using a parallel plate type device, an RF discharge power of 120 W and a gas pressure of 120 W were applied.
Dry etching was performed using CF ₄ gas at 0.07 Torr to remove the exposed portion of the n ⁺ layer, thereby forming an n ⁺ layer 4 having a desired pattern. Next, the photoresist was peeled off. On the other hand, for comparison, the surface of the same glass substrate as above was treated with hydrofluoric acid (49 volume% aqueous solution), nitric acid (60 volume%
A planar type photosensor (hereinafter referred to as ``substrate'') was treated in the same manner as above except that an a-Si subbing layer was not formed. A photo sensor with acid treatment and without subbing layer was manufactured. When the photocurrent values obtained for the above two types of photosensors were compared under the same conditions when light of λ _nax =565 nm was incident from the glass substrate 1 side, almost the same values were obtained for both. This shows that the presence of the a-Si subbing layer 2 in the photo sensor of the present invention does not deteriorate the photocurrent characteristics. Next, the above two types of photo sensors were subjected to a heat cycle durability test under the same conditions, and it was found that the films did not peel off and had sufficient adhesion. Example 2: In the manufacturing process of the photo sensor of the present invention in Example 1, when forming the a-Si subbing layer 2, the discharge power (hereinafter abbreviated as "discharge power 1") initially set and the discharge power 1 The same steps as in Example 1 were carried out, except that glow discharge was performed using the following combinations of discharge times.

[Table] As a result, when the discharge power was 80W and 50W, it was possible to obtain a photo sensor without film peeling, but when the discharge power was 30W, 8W and 4W, it was possible to obtain a photo sensor using photoresist AZ-1370. Film peeling occurred during the lithography process (including cleaning with an ultrasonic cleaner), making it impossible to obtain the intended good photo sensor. Example 3: After forming the a-Si subbing layer 2 in the same manner as in Examples 1 and 2, the substrate 1 was taken out and the substrate 1
The refractive index of the a-Si subbing layer 2 formed thereon was measured. Discharge power of glow discharge and a-Si subbing layer 2
The relationship between the refractive index and the refractive index is shown in FIG. The adhesion between the substrate and the photoconductive layer is related to the discharge power of the glow discharge during film formation, and film peeling is caused by the difference in the coefficient of thermal expansion between the intrinsic stress induced depending on the internal structure of the thin film and the substrate. It is thought that this is due to the total stress due to the combination with the internal stress depending on . Therefore, the total stress of the a-Si subbing layer 2 formed on the substrate 1 was measured. The relationship between the discharge power 1 of glow discharge and the total stress of the a-Si subbing layer 2 is shown in FIG. Stress appears as compressive stress and reaches its maximum value when discharge power 1 is around 10W;
The stress decreases as the value increases. The reason why the stress decreases as the discharge power 1 increases is considered to be mainly because the increasing number of voids in the film generates tensile stress, which offsets the compressive stress. As mentioned above, the photoconductivity of the photoconductive layer is related to the discharge power during film formation, and in order to obtain the desired photoconductive properties it is necessary to perform the deposition at a relatively low discharge power, and therefore The a-Si layer 3 in Examples 1 and 2 was deposited at relatively low discharge power. From the above, it is clear that the a-Si subbing layer 2 of the photo sensor in the image sensor unit of the present invention has the function of a stress relaxation layer and exhibits the effect of improving the adhesion between the substrate and the photoconductive layer. Ru. In addition, in this photo sensor, the base 1
When used by irradiating light from the side, it is preferable that the thickness of the a-Si subbing layer 2 is not too thick in order to obtain good photoconductive properties, and preferably, for example, 1000 Å or less. Note that when light is incident from the side opposite to the substrate 1 side, there is no need to consider the influence of light absorption in the a-Si subbing layer 2 on the photoconductive properties.
The Si sublayer 2 may be quite thick. Example 4: In the photo sensor manufacturing process of the present invention in Example 1, except that after the formation of the a-Si layer 3, the discharge power was increased to 80 W, glow discharge was performed for 25 minutes, and an a-Si layer was further formed. Then, the same steps as in Example 1 were performed. FIG. 5 is a partial sectional view of the planar type photo sensor thus obtained, showing the same portion as FIG. 2. In FIG. 5, members similar to those in FIG. 2 are given the same reference numerals, and 3' is an a-Si layer.
The thickness of the a-Si layer 3' is 0.3μ, and the formation rate per unit thickness of this layer is increased by increasing the discharge power.
- significantly higher than the formation rate per unit thickness of the Si layer 3; In the photo sensor obtained in this example, a photoconductive layer is constituted by the a-Si subbing layer 2, the a-Si layer 3, and the a-Si layer 3'. According to the photo sensor of this embodiment, the obtained photocurrent is larger than that of the first embodiment due to the increased thickness of the a-Si layer. Example 5: In the photo sensor manufacturing process of the present invention in Example 1, the substrate temperature was adjusted during the formation of the a-Si subbing layer 2.
The same steps as in Example 1 were carried out, except that the temperature was maintained at 70° C., the discharge power 1 was 8 W, and glow discharge was performed for 15 minutes. When the a-Si subbing layer 2 was formed under the same conditions, the substrate 1 was taken out and the refractive index of the a-Si subbing layer 2 was measured and found to be 3.10. The photo sensor obtained in this example was as good as the photo sensor of the present invention obtained in Example 1. Example 6: In the photo sensor manufacturing process of the present invention in Example 1, SiH ₄ diluted to 5% with H 2 was used as the raw material gas when forming the a-Si subbing layer 2, and the discharge power ₁ was set to 30 W. The same process as in Example 1 was performed except for glow discharge for 10 minutes. When the a-Si subbing layer 2 was formed under the same conditions, the substrate 1 was taken out and the refractive index of the a-Si subbing layer 2 was measured and found to be 3.02. The photo sensor obtained in this example was as good as the photo sensor of the present invention obtained in Example 1. Example 7: In the photo sensor manufacturing process of the present invention in Example 1, gas pressure was applied during the formation of the a-Si subbing layer 2.
The same steps as in Example 1 were carried out, except that the glow discharge was performed at 0.30 Torr, discharge power 1 was 50 W, and glow discharge was performed for 5 minutes. When the a-Si subbing layer 2 was formed under the same conditions, the substrate 1 was taken out and the refractive index of the a-Si subbing layer 2 was measured and found to be 3.12. The photo sensor obtained in this example was as good as the photo sensor of the present invention obtained in Example 1. Example 8: By the same method as in the case of the photo sensor of the present invention in Example 1, 864 photo sensors were manufactured by arranging them in an array on the same substrate. This can be easily done by appropriately setting a mask during the photolithography process. A schematic partial plan view of the long photo sensor array thus obtained is shown in FIG. In FIG. 6, 11 is an individual electrode, and 12 is a common electrode. The density of this long photo sensor array is 8 bits/mm,
It has the width of an A6 page. The uniformity of photocurrent and dark current between bits of the photo sensor array obtained in this example was measured. The results are shown in FIG. On the other hand, for comparison, the same substrate was prepared using the method described in Example 1 with the base acid treatment and without the subbing layer.
The uniformity of photocurrent and dark current between bits of a long photosensor array manufactured by arranging 864 photosensors was measured. The results are shown in FIG. A comparison between FIG. 7 and FIG. 8 shows that the photo sensor of the present invention has no microscopic defects on the substrate, and the a-Si subbing layer acts as a stress relaxation layer, so the photoconductive It can be seen that the uniformity of characteristics is extremely good. Example 9: An attempt was made to drive an 864-bit long photo sensor array as obtained in Example 8 by dividing it into 27 blocks of 32 bits each in a matrix. That is, after manufacturing a long photo sensor array using the same process as in Example 8, a polyimide resin (PIQ manufactured by Hitachi Chemical Co., Ltd.) was coated on the entire surface and baked, and then a negative type photoresist (manufactured by Tokyo Ohka Co., Ltd.) was applied.
After forming a pattern in the desired shape using OMR-83), remove unnecessary portions of PIQ with a polyimide resin etching solution (PIQ etchant manufactured by Hitachi Chemical Co., Ltd.), peel off OMR-81, and heat at 300℃. It was cured for 1 hour in a nitrogen atmosphere to form an insulating layer and through holes for matrix wiring.
Next, Al was deposited to a thickness of 2 μm by electron beam evaporation, and the upper electrode of the matrix wiring was formed using a positive photoresist AZ-1370 and etching solution 1. A schematic partial plan view of the matrix wiring portion of the long photo sensor array thus obtained is shown in FIG. 9, and a sectional view taken along line X--X is shown in FIG. In FIGS. 9 and 10, 21 is a base body, and 22
is an a-Si subbing layer, 23 is an a-Si layer, 24 is an n ⁺ layer, 25 is a common electrode,
26 is an individual electrode, 27 is an insulating layer, 2
8 is a through hole, and 29 is an upper electrode of the matrix wiring. FIG. 11 shows a drive circuit diagram for matrix driving the long photo sensor array of 8 bits/mm and A6 width obtained in this way. In FIG. 11, 31 is the photoconductive layer of the photo sensor;
2 is a block selection switch, 33 is a common switch, and 34 is an amplifier. FIG. 12 shows a partially cutaway perspective view of the image sensor unit used in this example. 1st
Figure 3 is a sectional view thereof. In the figure, 41 is the base of the photo sensor array. Below the base body 41 is a fiber lens array 42.
is provided, and light emitting diode (LED) arrays 43 are provided on both sides thereof. 44 is a driving IC, and the IC 44 is electrically connected to the matrix wiring section on the base 41 by a flexible conductive material 45. 46 is a document to be read, and 47 is a feeding roller thereof. Also, 48
are heat dissipation fins, and 49 is a heat dissipation plate. drive
The IC is thermally connected to a heat sink 49. still,
Photo sensor array and fiber lens array 4
The arrangement directions of LED array 2 and LED array 43 are parallel to each other. When the image sensor unit is driven in matrix as described above, voltage is applied for 100μsec.
The uniformity of the output photocurrent between bits was then measured. The results are shown in FIG. As can be seen from FIG. 14, the output photocurrent of each bit exhibits extremely good uniformity, indicating that signal readout is sufficiently possible with matrix driving. Example 10: A long image sensor unit was constructed using the photo sensor array obtained in Example 9. FIG. 15 is its circuit diagram. In FIG. 15, photo sensors E1 to E9
Three blocks constitute one block, and three blocks constitute a photo sensor array. Capacitors C1~ corresponding to photo sensors E1~E9 respectively
The same applies to C9 and switching transistors T1 to T9. One electrode (common electrode) of each of the photo sensors E1 to E9 is connected to a power source 101, and the other electrode (individual electrode) is grounded via a capacitor C1 to C9, respectively. Further, the individual electrodes having the same order within each block of the photo sensors E1 to E9 are connected to the common line 10 through the switching transistors T1 to T9, respectively.
2 to 104. In detail, the first switching transistors T1, T4, T7 of each block are connected to the common line 102.
Then, the second switching transistors T2, T5, T8 of each block are connected to the common line 103, and the third switching transistors T of each block are connected to the common line 103.
3, T6, and T9 are connected to the common line 104, respectively. Common lines 102-104 are connected to amplifier 105 via switching transistors T10-T12, respectively.
It is connected to the. Further, the gate electrodes of the switching transistors T1 to T9 are connected in common for each block, and are connected to parallel output terminals of the shift register 106, respectively. Since high level signals are sequentially output from the parallel output terminals of the shift register 106 at predetermined timing, the switching transistors T1 to T
9 is turned on sequentially for each block. In addition, switching transistors T10 to T1
Each gate electrode of the transistors T10 to T12 is connected to a parallel output terminal of the shift register 107, and a high level is sequentially output from the parallel output terminal at a predetermined timing, thereby sequentially turning on the switching transistors T10 to T12. Furthermore, switching transistors T10 to T
The twelve commonly connected terminals are grounded via a discharging switching transistor T13, and the gate electrode of the switching transistor T13 is connected to the terminal 108. The operation of the elongated image sensor unit having such a configuration will be briefly explained. When light enters the photo sensors E1 to E9, the power supply 101 connects the capacitors C1 to C1 depending on the intensity of the light.
Charge is accumulated in C9. Subsequently, the shift registers 106 and 107 sequentially output a high level at respective timings, and it is now assumed that a high level is output from the first parallel output terminals of both registers. Then, the switching transistor T10 connected to the switching transistors T1 to T3 of the first block and the common line 102 is turned on, and the charge accumulated in the capacitor C1 is transferred to the switching transistor T1, the common line 102, and Through the switching transistor T10,
It is input to the amplifier 105 and output as image information. When the charge stored in the capacitor C1 is read out, a high level is applied to the terminal 108,
Switching transistor T13 is turned on. Thereby, the residual charge in the capacitor C1 is transferred to the switching transistor T1, the common line 10
2, completely discharged through the switching transistor T10 and the switching transistor T13. Next, while keeping the first parallel output of shift register 106 at high level, shift register 1
07 to turn on the switching transistors T11 and T12 in turn. As a result, the above reading and discharging operations are performed on the capacitors C2 and C3, and the information stored therein is sequentially read out. When the reading of the information of the first block is completed in this way, the shift register 106 is sequentially shifted, and the reading of the information of the second and third blocks is performed in the same manner as described above. In this way, the information stored in the capacitors C1 to C9 is serially read out and output from the amplifier 105 as image information. Since the long photo sensor array shown in FIG. 15 has a capacitor for charge storage, it is possible to increase the output signal. Also, photo sensors E1 to E9, capacitor C
1 to C9 and switching transistor T1 to
When T9 is formed using a thin film semiconductor on the same substrate, it has advantages such as being able to reduce the number of connection points with external circuits. Example 11: A long image sensor unit was constructed using the photo sensor array obtained in Example 9. FIG. 16 is its circuit diagram. However, in this embodiment, photo sensors E1 to E
9, capacitors C1 to C9, switching transistors T1 to T12, and shift register 10
6, 107, etc. are the same as those shown in FIG. 15, so their explanation will be omitted. In FIG. 16, the individual electrodes of the photo sensors E1 to E9 are each connected to a switching transistor ST1.
- Grounded via ST9. That is, each of the switching transistors ST1 to ST9 is
Connected in parallel with capacitors C1 to C9. The gate electrodes of switching transistors ST1 to ST9 are connected to the gate electrodes of switching transistors T1 to T9.
Similarly to the gate electrodes of the gate electrodes, they are commonly connected for each block, and are connected to the parallel output terminals of the shift register 201 for each block. Therefore, depending on the shift timing of the shift register 201, the switching transistor T
1 to T9 are turned on for each block. Next, the operation of this embodiment having such a configuration will be explained using the switching transistor T shown in FIG.
This will be explained using timing charts of 1 to T12 and ST1 to ST9. First, when light enters the photo sensors E1 to E9, charges are accumulated in the capacitors C1 to C9 from the power source 101 according to the intensity of the light. Then, first, a high level is output from the first parallel terminal of the shift register 106, and the switching transistors T1 to T3 are turned on.
Figure 7a]. In the meantime, the shift register 107 shifts and the switching transistors T10 to T12 are sequentially turned on [FIG. 17 d to f]. That is, the optical information stored in the capacitors C1 to C3 of the first block is sequentially read out. When the information of the last capacitor C3 of the first block is read, the shift register 106 is shifted and a high level is output from the second parallel terminal.
Switching transistors T4 to T6 are turned on [FIG. 17b]. At the same time, a high level is output from the first parallel terminal of the shift register 201, and the switching transistors ST1 to ST3 are turned on.
The residual charges in capacitors C1-C3 are completely discharged [Fig. 17g]. In parallel with this discharging operation, while the switching transistors T4 to T6 are in the on state, the switching transistors T10 to T12 are sequentially turned on by the shift of the shift register 107, and the capacitors C4 to C6 of the second block are turned on. The accumulated light information is read out sequentially [first
7 d-f]. Next, in parallel with the read operation of the third block [Fig. 17c], the capacitor C4 of the second block is
-C6 discharge is performed [FIG. 17h], and the above operation is repeated for each block. In this way, the capacitor of the block whose readout has been completed can be discharged in parallel with the readout of the next block, and the overall operating time can be shortened. FIG. 18 shows another embodiment of the present invention, in which only portion A in FIG. 16 is different. That is, amplifiers 202 to 204 are connected to common lines 102 to 104, respectively, and amplifiers 202 to 20
4 are connected to parallel input terminals of shift register 205. And shift register 2
Image information is serially output from the serial output terminal 05. Therefore, in this configuration, one block of information is simultaneously input to the shift register 205, and then serial image information is output by shifting the shift register 205. In this embodiment as well, when one block's worth of information is output from the shift register 205, the capacitor of that block can be discharged and the next block can be read out in parallel. In addition, switching transistors ST1 to ST9
Similarly to the switching transistors T1 to T9, thin film transistors may be used, and in that case, they can be formed on the same substrate as other elements. Even if thin film transistors are used as the switching transistors ST1 to ST9, the capacitor of one block can be discharged and the next block can be read out in parallel, so the overall readout time is shorter than that of Example 10. is shortened. In the above embodiment, an a-Si subbing layer 2 having a constant refractive index and an a-Si subbing layer 2 having a constant refractive index are used.
Although an example has been shown in which a layer with a refractive index that continuously changes in the film thickness direction is formed between the Si layer 3 and the Si layer 3, in the photo sensor of the long image sensor unit of the present invention, without forming an a-Si subbing layer 2 having a constant refractive index at a thickness of
A layer having a refractive index that gradually and continuously changes in the thickness direction from the surface of the substrate 1 may be formed. [Effects of the Invention] According to the elongated image sensor unit of the present invention as described above, the characteristics of the photo sensor can be made uniform, so that costs can be reduced without using a correction circuit. Furthermore, in the long image sensor unit of the present invention, since the refractive index of the photoconductive layer of the photo sensor changes continuously in the film thickness direction, stress relaxation at the layer interface is achieved and adhesion is good.
Further, during use, reflection at the layer interface becomes extremely small, thereby preventing loss of light quantity.

[Brief explanation of the drawing]

FIG. 1 is a partial plan view of the photo sensor of the image sensor unit of the present invention, and FIG.
FIG. FIGS. 3 and 4 are graphs showing the characteristics of the a-Si subbing layer. FIG. 5 is a partial sectional view of the photo sensor of the image sensor unit of the present invention. FIG. 6 is a partial plan view of the photo sensor array, and FIGS. 7 and 8 are graphs showing the characteristics of the photocurrent and dark current. FIG. 9 is a partial plan view of the matrix wiring section, and FIG. 10 is a cross-sectional view taken along line XX. FIG. 11 is a matrix drive circuit diagram. FIG. 12 is a partially cutaway perspective view of the long image sensor unit, and FIG.
The figure is a sectional view thereof. FIG. 14 is a graph of output photocurrent. 15 and 16 are circuit diagrams of a long image sensor unit,
FIG. 17 is a timing chart, and FIG. 18 is a diagram showing a partial modification of FIG. 16. 1...Substrate, 2...a-Si subbing layer, 3...a
-Si layer, 4...n ⁺ layer, 5...common electrode, 6...individual electrode, 41...photo sensor array substrate, 42
...Fiber lens array, 43...LED array, 44...Drive IC, 46...Reading document.

Claims (1)

[Scope of Claims] 1. A photoelectric conversion layer mainly composed of amorphous silicon and a pair of electrodes electrically connected to the photoelectric conversion layer, wherein the photoelectric conversion layer has a refractive index in the layer thickness direction. A plurality of photoelectric conversion elements each having a continuously changing region and having a refractive index of 3.2 or less in a region located closest to the substrate of the photoelectric conversion layer at a wavelength of 6328 Å,
A photo sensor array provided on a substrate, a light source for illuminating a document to be read, and an imaging array for forming an image of the document on the photo sensor array. Image sensor unit. 2 Thickness of the photoelectric conversion layer located closest to the substrate
The image sensor unit according to claim 1, wherein the refractive index in a region of 1000 Å or less is 3.2 or less for light with a wavelength of 6328 Å.