JPH0214790B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to the manufacture of a light-receiving element comprising a lower electrode formed on a substrate, a photoconductive film made of an amorphous material mainly composed of silicon and containing hydrogen, and a transparent electrode formed by sputtering. It is about the method. For example, manufacturing a solid-state image sensor in which a photoconductor layer and a transparent electrode made of an amorphous material mainly composed of silicon and containing hydrogen (hereinafter referred to as amorphous hydrogenated silicon) are laminated on a scanning Si-IC substrate. It is useful to apply the method. Of course, it can also be applied to other light receiving elements.

The method of the present invention is useful after forming a photoconductor layer and transparent electrode of amorphous hydrogenated silicon.

In the example of the solid-state imaging device described above, a plurality of solid-state elements having a photoelectric conversion function and a signal accumulation function are arranged, and each solid-state element corresponds to one pixel to form an imaging surface.
It is a solid-state imaging device that converts external video information into electrical signals by sequentially scanning this imaging surface, and in particular, the photoconductor layer that forms the imaging surface covers a scanning IC board on which switches, scanning circuits, etc. are formed. It is formed like this.

A solid-state imaging device in which a photoconductor layer forming an imaging surface is formed so as to cover a semiconductor substrate on which switches, scanning circuits, etc. are formed has been reported in, for example, Japanese Patent Laid-Open No. 10715/1983. There is. below,
This technology will be briefly explained. As shown in Figure 1
A scanning circuit, a switch circuit, etc. are integrated on a Si substrate 1, and a photoconductive film 8 that plays the role of photoelectric conversion is attached to the Si-IC.
It is deposited on the substrate. The operating principle will be explained with reference to FIG. 1. Incident light 10 reaches photoconductive film 8 through transparent electrode 9. The light is absorbed here, producing electron-hole pairs, and these carriers are accumulated in the metal electrode 7 by the bias voltage V _T . The accumulated carriers are switched by an insulated gate field effect transistor (MOSFET) formed on a semiconductor substrate 1 and consisting of a source 2, a drain 3, and a gate 4, and taken out to the outside through a signal line 5. 6 is an insulating film. In this structure, the scanning circuit and photoelectric conversion section are separated, so not only does it not cause a decrease in resolution or photosensitivity, but it also has the feature that blooming is less likely to occur because light does not reach the Si substrate.

Furthermore, a solid-state imaging device shown in FIG. 1 has been proposed in which amorphous hydrogenated silicon having excellent photoconductive properties is used as a photoconductive film.

However, after forming a photoconductive film made of amorphous hydrogenated silicon on a scanning Si-IC substrate, a transparent electrode made of indium oxide-tin oxide or a translucent electrode made of platinum or the like is formed on top of it by sputtering. This resulted in a drawback that the photoresponse characteristics of the photoconductive film deteriorated.

Forming a transparent electrode of indium oxide-tin oxide based metal oxide or a translucent metal electrode such as gold and platinum on a photoconductive film by sputtering is a method for forming a photoconductive film made of amorphous hydrogenated silicon. This is to improve adhesiveness. This problem is especially required in color solid-state imaging devices. It is also possible to form transparent oxide electrodes or translucent metal electrodes by vacuum evaporation, but films formed by evaporation generally have poorer adhesion to the underlying film than films formed by sputtering. There is. When the solid-state image sensor whose picture element section is shown in cross-section in Figure 1 is used as a color solid-state image sensor, it is necessary to form a color filter layer on top of the transparent electrode that transmits only light in a predetermined wavelength range. There is. When carrying out the step of forming this color filter layer, if the adhesiveness between the photoconductive film 8 and the transparent electrode 9 is weak, a problem often occurs in which the transparent electrode 9 peels off. In this respect, it is preferable to form the transparent electrode 9 by a sputtering method rather than by a vacuum evaporation method. In addition, indium oxide-tin oxide based transparent electrodes can be fabricated using CVD using indium-tin halides or organometallic salts.
It is also known to use a chemical vapor deposition (Chemical Vapor Deposition) method. However, with this method, the specific resistance is low, there is no change in resistance over time, and
In order to obtain a film with good adhesion to the base film, the substrate temperature must be 300°C or higher. On the other hand, when a photoconductive film made of amorphous hydrogenated silicon is heated to 300° C. or higher, its photosensitivity in the visible light region decreases significantly. Therefore, transparent electrodes for solid-state imaging devices using amorphous hydrogenated silicon as a photoconductive film are
It cannot be created by CVD method.

In the solid-state image sensor shown in FIG.
A method (referred to as an accumulation operation method) is used in which data is read out through a signal line 5 using a built-in MOSFET switch within an extremely short period of time. The light-receiving element shown in FIG. 2 is a test light-receiving element for measuring optical response characteristics. It consists of a lower electrode 12 provided on a substrate 11, a photoconductive film 13 made of amorphous hydrogenated silicon, and a transparent electrode 14, and a constant voltage V _T is always applied to the photoconductive film. light pulse 15
The photocharge generated on the photoconductive layer 13 by the ammeter 1
6 can be read directly. An example of the photoresponse characteristics of a light-receiving element in which a transparent electrode is formed by sputtering is as shown in FIG. 3. This light-receiving element is formed by a normal manufacturing method,
After forming a transparent electrode by a sputtering method under a substrate temperature condition of about 80 to 220°C, heating was stopped and left as it was. In Figure 3, characteristic a is the incident light pulse, and curves b and c are the optical response characteristics when the transparent electrode side is positively biased (generally V _T = 0 to 21 V is used). The photoresponse characteristics when biased negatively (generally V _T =0 to −21 V is used) are shown. As can be seen from the characteristic curve in FIG. 3, the photoresponse characteristics are significantly inferior, especially when a negative bias is applied to the transparent electrode side. In other words, in Figure 3, when a light pulse is irradiated with the transparent electrode side negative, a phenomenon in which negative charges are injected from the transparent electrode (also called secondary photocurrent) occurs, and even after the light is turned off, the attenuation current continues. This shows that a large amount of current flows over a long period of time, and that it does not return to the dark current level easily. This phenomenon occurs in solid-state imaging devices, where an image that has been projected once remains as an afterimage even if the light is blocked, or even becomes burned-in and cannot be removed. Such a phenomenon in a solid-state image sensor is an extremely serious drawback in practical use.

The present invention is extremely effective in obtaining an amorphous hydrogenated silicon thin film solid-state imaging device that eliminates the above-mentioned drawbacks.

In order to achieve the above object, the present invention
- After forming an amorphous photoconductive film mainly made of silicon containing hydrogen on an IC substrate by a reactive sputtering method or a glow discharge CVD method, a transparent electrode is formed on the photoconductive film by a sputtering method. . Thereafter, the present solid-state image sensor is heat-treated in a temperature range of 170°C to 250°C to improve the deterioration of the photoresponse characteristics of the present solid-state image sensor that occurred due to the transparent electrode being formed on the photoconductive film by the sputtering method. It is something. According to the present invention, it is possible to obtain an element that has excellent resolution and spectral sensitivity in the visible light region, which are the advantages of the present solid-state image sensor, and is less likely to cause the blooming phenomenon. For the reactive sputtering method of the photoconductive film, a general sputtering device or a magnetron-type high-speed sputtering device can also be used. Polycrystalline silicon is placed as a sputtering target on one cathode (target-side electrode) of the counter electrodes in the sputtering device, and a scanning Si-IC substrate is placed on the other anode (substrate-side electrode). After degassing the sputtering chamber by heating it to 250 to 300℃ while keeping the sputtering chamber under a high vacuum of 1×10 ^-5 Torr or less, a mixed gas of hydrogen and a rare gas such as argon is used as the discharge gas. The film is introduced into a sputtering chamber and high-frequency sputtering is performed at 13.56 MHz to deposit an amorphous photoconductive film mainly composed of silicon containing hydrogen on the scanning Si-IC substrate. The substrate temperature during film formation is 100 to 350° C., the pressure of the discharge gas is 2×10 ⁻³ Torr to 5×10 ⁻² Torr, and the composition of hydrogen gas in the discharge gas is within the range of 10 to 60 mol %.

In addition, the glow discharge CVD (Chemical
There are two types of vapor deposition methods: the RF coil method and the bipolar discharge method. In both cases, a mixture of a silane gas such as SiH ₄ and a rare gas such as argon is used as the discharge gas, and glow discharge is performed to generate hydrogen-containing silicon on the scanning IC substrate through a decomposition reaction of the silane gas. This is a method of depositing a primarily amorphous photoconductive film, and is distinguished from the reactive sputtering method, which uses a reaction of adding hydrogen to silicon. The RF coil method places a reaction chamber inside an RF coil, applies a high frequency of 13.56 MHz to the RF coil, and causes a glow discharge of a mixed gas of SiH ₄ and argon introduced into the reaction chamber. This method involves depositing an amorphous photoconductive film mainly composed of silicon containing hydrogen on a commercial IC substrate. In addition, the bipolar discharge method uses a normal sputtering device, and a 13.56MHz
By applying _a high frequency of This method involves depositing a film. The substrate temperature during film formation is 100 to 300℃, the pressure of the discharge gas is higher than that of the reactive sputtering method, from 5 × 10 ^-2 Torr to 2 Torr, and the composition of SiH ₄ gas in the discharge gas is within the range of 5 to 40 mol%. be.

After a photoconductive film made of amorphous hydrogenated silicon is formed on the scanning Si-IC by the above method, a transparent electrode is formed on the photoconductive film by sputtering.
As this transparent electrode, a transparent electrode whose main component is (1) one selected from indium oxide, tin oxide, and a mixture thereof is used. (2) A translucent metal electrode whose main component is one selected from the group consisting of gold, platinum, tantalum, molybdenum, aluminum, chromium, nickel, and mixtures thereof can also be used.

In order to form the transparent electrode (1), there is a method of performing reactive RF sputtering in argon gas containing oxygen gas using an indium-tin metal as a target.
A method is used in which RF sputtering is performed using a tin oxide-based sintered target in a rare gas such as argon gas. In this case, an indium oxide-tin oxide based sintered body is installed as a sputtering target on one cathode (target side electrode) of the opposing electrodes in the sputtering device, and an amorphous material is installed on the other anode (substrate side electrode). A scanning Si-IC substrate on which a photoconductive film made of silicon hydride is deposited is installed. After evacuating the sputtering chamber to a high vacuum of 5×10 ^-6 Torr or less, a rare gas such as argon is introduced into the sputtering chamber as a discharge gas, and high frequency sputtering at 13.56MHz is performed to form a predetermined sputtering layer on the photoconductive film. An indium oxide-tin oxide based transparent electrode with a pattern of Substrate temperature during film formation is 80℃~220℃
℃, the pressure of the discharge gas is 3×10 ^-3 Torr to 5×
10 ^-2 Torr. In this way, a solid-state imaging device as shown in FIG. 4, a cross-section of the pixel portion thereof, is obtained.

In the figure, 20 is a semiconductor substrate, and 26 and 27 are diffusion regions formed therein to form a source or a drain. 25 is a gate electrode, 29,3
10 is a drain electrode and a source electrode, respectively; 2
1, 22, 28, and 30 are insulating layers. Note that the source electrode is formed of two layers: a metal layer 31 provided on the source region 26 and a metal layer 31 further provided above this. The layer 32 is a photoconductive film formed by the aforementioned sputtering method or glow discharge method. 33 is a transparent electrode formed by the above-mentioned sputtering method. Regarding the transparent electrode (2), gold, platinum, tantalum, molybdenum,
If a metal whose main component is one selected from the group consisting of aluminum, chromium, nickel, and mixtures thereof is set as a sputtering target, a translucent shape can be formed by the sputtering method similar to the transparent electrode in (1) above. Metal electrodes can be deposited. In this case, in order to improve light transmittance, the thickness of the semi-transparent metal electrode must be made as thin as possible within a range where there is no disconnection between each picture element of the solid-state image sensor.
Usually, the film thickness is 400 Å or less. After forming a transparent electrode by the sputtering method under the substrate temperature condition of 80 to 220° C. in this manner, heating is stopped and the substrate is left to stand. The method for manufacturing a light receiving element according to the prior art is completed above.

The solid-state image sensor obtained by the method described above is
As explained in the figure, this is an element with deteriorated photoresponse characteristics. In particular, when a negative bias voltage V _T is applied to the transparent electrode 33 in FIG. 4, the afterimage and burn-in become large. However, this element is 170
When heat-treated at a temperature between 150° C. and 250° C. for about 15 minutes to several hours, the afterimage and burn-in characteristics are improved to such an extent that they do not cause problems at all. As described above, the present invention solves the problems that have not been solved in light receiving elements manufactured by the conventional conventional manufacturing method, that is, the manufacturing method in which transparent electrodes are formed under heating conditions and then left to stand. When comparing the conventional method in which transparent electrodes are left to stand after formation, and the method of the present invention in which heat treatment is performed after formation, there is a noticeable difference in their effects. This improvement can be expressed by the photoresponse characteristics of the light receiving element shown in FIG. 2, and an example thereof is shown in FIG. In FIG. 5, the characteristic a is the incident optical pulse,
Curves d and e are the photoresponse characteristics when each transparent electrode side is positively biased (generally V _T = 0 to 21 V is used), and the transparent electrode side is negatively biased (generally
The photoresponse characteristics are shown when V _T =0 to -21V).
As is clear from FIG. 5, it can be seen that the photoresponse characteristics are significantly improved when the transparent electrode side is negatively biased. That is, the secondary photocurrent in which negative charges are injected from the transparent electrode side is suppressed, and the decay current after the light is turned off quickly drops to the same level as the dark current. Another major improvement is that regardless of whether the bias applied to the transparent electrode side is positive or negative, photosensitivity can be achieved at a relatively low voltage V _T value compared to the characteristics before heat treatment. This phenomenon is observed in exactly the same way in the solid-state imaging device shown in FIG.

In the solid-state image sensor shown in FIG. 4, the relationship between the heat treatment temperature and the afterimage 50 ms after the light was turned off was as shown in FIG. However, the heat treatment time was 20 minutes. As is clear from Figure 6, when the heat treatment temperature is gradually raised from room temperature,
The afterimage gradually becomes larger, reaching its maximum value between 100 and 120℃, and then rapidly decreasing from around 150℃.
It shows a minimum value between 170℃ and 250℃, and also tends to increase. The heat treatment time is 20 to 40 minutes at each temperature, and the saturation value of the afterimage is approximately reached at that temperature. Therefore, there is no particular point in carrying out heat treatment for a longer time than necessary. Heat treatment is normally performed in the atmosphere, but similar effects were confirmed even when heat treatment is performed in a rare gas such as argon gas or an inert gas such as nitrogen. A general imaging device can be used sufficiently if the afterimage after 50 ms is 1% or less. From Figure 6, the effect starts to appear at least above 140℃, but from 170℃
If heat treatment is performed in the range of 250° C., the solid-state imaging device shown in FIG. 4 will have an afterimage of 1% or less after 50 ms, and can be used extremely conveniently as an imaging device.

The effects of the present invention shown in FIGS. 5 and 6 are solely due to the photoconductive film and the transparent electrode produced by depositing the transparent electrode by sputtering on the photoconductive film made of amorphous hydrogenated silicon. This improves the problem of electrical contact between the two. Immediately after amorphous hydrogenated silicon is deposited using the above-mentioned reactive sputtering method or glow discharge method, it is kept in a photoconductive film deposition apparatus at 220 to 270°C in vacuum for the purpose of greatly improving photosensitivity. This is a different technology from heat treatment.

Further, the present invention applies not only to the solid-state imaging device shown as an example in FIG. In principle, it is also effective for all light receiving elements having the configuration shown in FIG. For example, it can be applied to one-dimensional close-contact line sensors or solar cells. In addition, CCD (Charge) is used as a scanning circuit for solid-state imaging devices.
Of course, the present invention can also be applied to a device using a coupled device (coupled device) transfer area.

The present invention will be explained in detail below with reference to Examples.

Embodiment 1 FIG. 7 shows the principle of a solid-state imaging device. The picture elements 44 are arranged in a matrix and read out one by one using the XY addressing method. Selection of each picture element is performed by a horizontal scanning signal generator 41 and a vertical scanning signal generator 42. 43 is a switch section connected to each picture element, and 45 is an output end.

FIG. 8 to FIG. 12 are cross-sectional views of a picture element portion showing a method of manufacturing a solid-state imaging device according to the present invention. The switch circuit, scanning circuit section, etc. formed on the semiconductor substrate are manufactured using normal semiconductor device processes. A thin SiO ₂ film of about 800 Å is formed on the p-type silicon substrate 20, and a Si ₃ N ₄ film of about 1400 Å is formed at a predetermined position on this SiO ₂ film. The SiO ₂ film is made using the normal CVD method, and the Si ₃ N ₄ film is made using SiH ₄ , NH ₄ , N ₂
The CVD method was used. A p-diffusion region is formed from the top of the silicon substrate by ion implantation. This diffusion region 21 was provided to better isolate each element. Then H ₂ :
Locally oxidize silicon in an O ₂ =1:8 atmosphere,
A SiO ₂ layer 22 is formed (FIG. 8). This method is a local oxidation method of silicon for element isolation, generally called LOCOS. Once, the aforementioned Si ₃ N ₄
The film is removed, and the gate insulating film of the MOS transistor is formed using a SiO ₂ film. Next, a gate portion 25 made of polysilicon and n-type diffusion regions 26, 2 are formed.
7 is formed, and a SiO ₂ film 28 is further formed on top of this. Then, openings for the electrodes of the source 26 and drain 27 are opened in this film by etching (FIG. 9). Al is deposited to a thickness of 6000 Å as the drain electrode 29 and source electrode 310. Furthermore, SiO ₂
A film 30 is formed to a thickness of 7500 Å, and then a source electrode 31 is formed.
Then, evaporate Al to a thickness of 2500Å. FIG. 10 is a sectional view showing this state. Note that the electrode 31 is located in the region 2.
6, 27 and the gate portion. This is because if light enters the signal processing region between the element isolation diffusion layers 21, it will cause blooming, which is undesirable.

An amorphous photoconductive film 3 mainly made of silicon containing hydrogen is deposited on the semiconductor IC substrate prepared in this manner.
2 was deposited to a thickness of 3 μm by reactive sputtering. At this time, polycrystalline silicon is used as a sputtering target by placing it at the cathode. Using hydrogen and argon mixed gas (H ₂ :Ar=20:80) as the discharge gas, the temperature was 3×10 ^-3 Torr.
High frequency sputtering was performed at 13.56MHz with a discharge gas pressure of . The state after the photoconductive film is formed is as shown in FIG. In ₂ O ₃ − on top of this photoconductive film
SnO ₂ -based transparent electrode with a thickness of 1000Å by sputtering method
Deposits to a film thickness of . At this time, an In ₂ O ₃ sintered body containing 5 mol % of SnO ₂ is used as a sputtering target by placing it on the cathode. Ar gas was used as the discharge gas at a gas pressure of 8×10 ^-3 Torr.
High frequency sputtering of 13.56MHz was used. After forming the transparent electrodes, an amorphous solid-state imaging device is obtained as shown in FIG. The photoresponse characteristics of the above-mentioned element result in an afterimage of 10% or more, and image burn-in is also significant. Next, when this element is heat-treated in air at 240°C for 20 minutes, an amorphous solid-state image sensor with a small afterimage of 1% or less and no burn-in phenomenon is obtained. Note that a second electrode is normally provided on the back surface of the semiconductor substrate 20 and is generally grounded. A color filter layer with predetermined spectral transmission characteristics is formed so as to correspond to each pixel electrode on this element, so that it can be used as a single-plate color amorphous solid-state image sensor at the junction interface between the photoconductive film and the transparent electrode. No peeling phenomenon occurred.

Embodiment 2 As in Embodiment 1, a switch circuit, a scanning circuit, etc. are formed on a predetermined semiconductor substrate. 10th
The figure is a sectional view of the substrate showing this state. However, the metal electrode 31 is a Ta electrode formed to a thickness of 3000 Å by sputtering.

An amorphous photoconductive film 3 mainly made of silicon containing hydrogen is deposited on the semiconductor IC substrate prepared in this manner.
2 was deposited to a thickness of 3 μm by glow discharge CVD. At this time, as a discharge gas (SiH ₄ 10mol% +
Using a mixed gas (Ar90 mol%), a high frequency discharge of 13.56 MHz was generated between opposing electrodes at a discharge gas pressure of 6 × 10 ^-2 Torr, and a decomposition reaction of SiH ₄ gas caused
Amorphous silicon containing hydrogen was deposited on the IC base placed on the cathode side and heated to 250°C. The state after the photoconductive film is formed is the same as in the above example.
The result will be as shown in Figure 1. pt on the top of this photoconductive film
A semitransparent electrode is deposited to a thickness of 200 Å by sputtering. At this time, a PT plate was installed on the cathode, and Ar gas was applied at 13.56MHz at a gas pressure of 5 × 10 ^-3 Torr.
High frequency sputtering was performed to obtain an amorphous solid-state image sensor as shown in FIG. The photoresponse characteristics of the above element have an afterimage of 15% or more, and the image is large. Next, this device was heated at 225℃ in an Ar gas atmosphere.
After heat treatment for 30 minutes, an element with an afterimage of about 0.5% and no burn-in phenomenon was obtained.

As explained using the above embodiments, if the manufacturing method of the solid-state imaging device of the present invention is used, the photoresponse of the amorphous solid-state imaging device generated by depositing the transparent electrode on the top of the photoconductive film by the sputtering method Deterioration of properties can be improved, afterimage and burn-in are both extremely small, and photoconductive properties are good.

Furthermore, similar effects can be obtained by using the various metals described above as the transparent electrode.

[Brief explanation of drawings]

Figure 1 is a cross-sectional view showing the basic structure of a solid-state imaging device, Figure 2 is a cross-sectional view of a typical light-receiving element, and Figure 3 is the optical response of the light-receiving element when a transparent electrode is formed by sputtering. Diagram showing an example of characteristics, No. 4
The figure is a cross-sectional view of one pixel of the amorphous solid-state image sensor manufactured according to the present invention, and FIG. Figure 6 is a diagram showing an example of response characteristics, Figure 6 is a diagram showing the effect of the present invention in relation to heat treatment temperature and afterimage after 50ms, Figure 7 is a diagram showing the principle of a solid-state image sensor, Figure 8 FIG. 12 is a sectional view of the main parts showing the manufacturing process of the solid-state imaging device of the present invention. 10...Incoming light, 1,20...Semiconductor substrate,
2, 3, 26, 27... Diffusion area, 4, 25...
Gate electrode, 6, 22, 28, 30...insulating layer,
7, 31, 310... source electrode, 5, 29...
Drain electrode, 8, 32...Photoconductive thin film, 9, 3
3...Transparent electrode, 37...Anodized film, 21...
Diffusion region, 11... Substrate, 12... Lower electrode, 1
3... Photoconductive film, 14... Transparent electrode, 15... Light pulse, 16... Ammeter, 41... Horizontal scanning signal generator, 42... Vertical scanning signal generator, 43...
Switch section, 44...picture element, 45...output end.

Claims (1)

[Claims] 1. A step of forming a photoconductive film made of an amorphous material mainly composed of silicon and containing hydrogen on a desired substrate, and forming a transparent conductive film on the photoconductive film by a sputtering method. In the method for manufacturing a light-receiving element, the light-receiving element is heated in a temperature range of 170°C to 250°C for 15 minutes after forming the transparent conductive film.
1. A method of manufacturing a light-receiving element, comprising a step of heating for more than 1 minute. 2. Claim 1, wherein the substrate is a semiconductor substrate having at least two-dimensionally arranged switches and a scanning element that transfers photocharges corresponding to an optical image taken out through the switches. A method of manufacturing the light-receiving element described above. 3. Claim 1, wherein the transparent conductive film is a transparent conductive film formed by a sputtering method and whose main component is one selected from indium oxide, tin oxide, and a mixture thereof. 2. Method for manufacturing the light receiving element described in Section 1. 4. A translucent metal film whose main component is one selected from the group consisting of gold, platinum, tantalum, molybdenum, aluminum, chromium, nickel, and mixtures thereof, which is formed by the sputtering method of the transparent conductive film described above. A method of manufacturing a light-receiving element according to claim 1, characterized in that: