JPH03283464A - Thin film laminated type electronic device - Google Patents

Abstract

PURPOSE:To prevent a thin film formed on a board from being separated off from the board due to the inner stress induced in the film concerned so as to improve a device in characteristics and reliability by a method wherein the inner stress is absorbed by a part of the thin film where the main component elements of the film are small in number per unit area or which is small in density. CONSTITUTION:An amorphous silicon hydride film 2 is deposited as thick as 1mum or so on a glass board 1, an amorphous silicon hydride film small in silicon density is formed as thick as required but 200Angstrom or below, and an amorphous silicon hydride film 4 is deposited thereon as thick as 1mum or so under the same conditions with the silicon film 2. When the amorphous silicon films 2 and 4 are formed in succession, the films are separated off, since they are formed under the conditions where inner stress exceeds 4X10<9>dyne/cm<2>, but in this case, as the structure 3 small in silicon density absorbs inner stress, films are prevented from being separated off.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to a thin film laminated electronic device having a structure in which a thin film is formed on a substrate, and in particular aims to reduce the internal stress of the thin film. The present invention relates to a thin film stacked electronic device.

(Prior Art) In recent years, research and development has been progressing on contact image sensors, solid-state image sensors, thin film transistors, etc. that use non-single crystal thin films such as amorphous silicon as active layers. Of these,
Contact image sensors and solid-state imaging devices are devices that utilize carriers generated by light in non-single-crystal thin films, while thin-film transistors are devices that control current flowing in non-single-crystal thin films using electric field effects. In any device, it is essential to improve the quality of the active layer's non-single-crystal thin film, typically a hydrogenated amorphous silicon film, in order to obtain good device characteristics, especially the carrier mobility in the thin film. It is important to make the

By the way, hydrogenated amorphous silicon film has 5t-H bonds and 5
Although it has r-H2 bonds, it is known that the electron mobility in a hydrogenated amorphous silicon film depends on the ratio of these bonds. Specifically, by increasing the proportion of 5i-H bonds and decreasing the proportion of 5i-H2 bonds, the electron mobility in the hydrogenated amorphous silicon film is increased.

However, by increasing the proportion of S+-H bonds, 5t-1(
Reducing the proportion of 2 bonds generally increases the internal stress of the thin film. That is, as shown as an example in FIG. 9, when the electron mobility of the hydrogenated amorphous silicon film increases, the internal stress of the film inevitably increases. For this reason, when attempting to improve device characteristics by depositing a thin film having high electron mobility, there is a problem in that the film tends to peel off from the substrate.

(Problem to be solved by the invention) As described above, conventional contact image sensors have used a non-single crystal thin film such as amorphous silicon as an active layer.

BACKGROUND ART In electronic devices such as solid-state imaging devices and thin film transistors, when a thin film with high electron mobility is deposited, there is a problem that the thin film easily peels off due to the large inherent internal stress. As a result, in order to reduce the internal stress, it is necessary to deposit a thin film whose electron mobility is smaller than the desired value, resulting in the problem that good device characteristics cannot be obtained.

Note that the above problem is not limited to electronic devices using non-single crystal thin films, but in electronic devices having a structure in which a thin film is deposited on a substrate, there is a risk that the thin film will peel off from the substrate due to internal stress of the thin film. This has been a factor in reducing the reliability of electronic devices.

The present invention has been made in consideration of the above circumstances, and its purpose is to prevent peeling of a thin film deposited on a substrate due to internal stress, and improve device characteristics and reliability. An object of the present invention is to provide a thin film stacked electronic device that can improve the performance.

In particular, the present invention can be applied to contact image sensors, solid-state image sensors, thin film transistors, etc. that do not cause peeling of the film even when depositing a non-single-crystal thin film with high electron mobility, and have excellent transport properties for electrons and holes. The purpose of the present invention is to provide a thin film stacked electronic device.

[Structure of the Invention] (Means for Solving the Problems) The gist of the present invention is to provide a portion in which the number of main constituent elements of the thin film per unit volume, that is, density is small, inside the thin film layer, and to apply stress in this portion. The purpose is to absorb it.

That is, the present invention provides a thin film stacked electronic device in which a thin film of amorphous silicon or the like is formed on a substrate having steps, and a structure in which the density of the main constituent elements of the thin film is lower than that in other parts of the thin film. The parts are arranged in a linear or planar shape with a desired width.

(Function) According to the present invention, since the low-density structure arranged inside the thin film layer plays the role of absorbing stress, a non-single crystal thin film such as amorphous silicon having high electron mobility can be peeled off from the film. can be deposited without any By appropriately selecting the width of the linear or planar low-density portion within a desired range, there is no problem with the adverse effect this portion has on device characteristics. Therefore, it is possible to form a thin film laminated type electronic device with good characteristics, in which the mobility of carriers such as electrons and holes is better than in the past, without any process problems such as film peeling.

Although the effect has been mainly described above on amorphous silicon-based thin films, the present invention is not limited to this, but has similar effects on metal layers, and furthermore, insulating layers such as silicon nitride films and silicon oxide films. Even if the thin film is formed under conditions of high stress, it is possible to make it difficult for the thin film to peel off.

(Example) Hereinafter, details of the present invention will be explained with reference to illustrated embodiments.

First, FIGS. 1 and 2 show examples in which peeling of the film does not occur when an amorphous silicon film with a large internal stress is formed on a glass substrate. In both figures, schematic diagrams of cross-sectional transmission electron micrographs are shown so that the magnitude of density can be clearly seen.

In FIG. 1, after depositing a hydrogenated amorphous silicon film 2 of about 1 μm on a glass substrate 1, a hydrogenated amorphous silicon film 3 having a low silicon density is formed to a desired thickness of 200 Å or less, Thereafter, a hydrogenated amorphous silicon film 4 of about 1 μm was deposited under the same conditions as the silicon film 2. In addition, in the actual photograph, the part 3 in the figure is compared to the thin film 2 and 4.
It appears whitish because its density is low and electrons can easily pass through it.

When the amorphous silicon film 2.4 is formed continuously,
The film peeled off because it was formed using conditions where the internal stress was 4 x 1 (19 dyn/Cll12 or more), but
In the embodiment shown in FIG. 1, the structure 3 having a low silicon density absorbs the stress, so that no peeling of the film occurs.

Here, the method for forming the hydrogenated amorphous silicon film is as follows:
The conventionally well-known plasma CVD method in which silane gas is divided by plasma or the photoCVD method in which it is decomposed by light energy may be used. Furthermore, in order to form a hydrogenated amorphous silicon film with a low silicon density, for example, the formation substrate temperature Ts may be lowered to increase the amount of hydrogen contained in the thin film. In this example, 2 and 4 are Ts=250
℃, the structure 3 with low silicon density has Ts=1
Formed at 50°C.

FIG. 2 shows a structural part 3 of a hydrogenated amorphous silicon film with a low silicon density that is not parallel to the glass substrate 1 but oblique.
This is an example in which it is formed into a planar shape.

In this figure, 2 is a hydrogenated amorphous silicon film,
Structures 3 were intentionally formed and placed during the deposition of 2. In addition,
3 is linear in the figure, but in reality it is arranged in a planar manner.

In the structure 3 having a low silicon density, the line width is set to a desired value in the range of 10 to 200 people, or does not need to be constant. Furthermore, the structural portion 3 may be constructed of two or more lines or two or more surfaces. The structure portion 3 can be formed arbitrarily by appropriately selecting the three-dimensional structure of the surface of the substrate 1.

In order to confirm that structure 3 actually achieves a small silicon density, the present inventors conducted an energy-dispersive
Evaluation was performed using a line analysis method (EDX method). Diameter 10 people φ
An electron beam focused on the sample was scanned in the direction 5 and irradiated onto the cross section, and the amount of silicon characteristic X-rays (X-ray peak intensity) emitted from the cross section of the sample was measured. As a result, as shown in FIG. 3, it was found that the silicon density in structure portion 3 was clearly reduced. Similarly, energy loss spectroscopy (EELS)
It was also confirmed that the silicon density in the structure 3 was reduced by the method (method).

As an effect of this example, when the structure part 3 was not formed, the hydrogenated amorphous silicon film 2 peeled off from the substrate 1, but in this example where the structure part 3 was formed, no peeling of the film occurred. It turns out there isn't.

Next, a second embodiment in which the present invention is applied to an actual electronic device will be described.

Figure 4 shows a hydrogenated amorphous silicon thin film photoelectric conversion layer.
FIG. 2 is a diagram showing a cross-sectional structure of a solid-state image sensor having a structure in which a scanning circuit and a current collecting electrode are stacked on a CCD substrate. In the figure, 10 is a Si substrate, 11 is a storage diode, 1
2 is a vertical CCD, 13 is a transfer electrode, 14.16 is an insulating film, 15 is an extraction electrode, and 17 is a pixel electrode.
A CCD image sensor substrate is formed from 0 to 17.

Further, 21 is a hole blocking layer, for example, a hydrogenated amorphous carbide layer, 22 is a hydrogenated amorphous silicon film, which is the main photoconductive layer, and 23 is an electron blocking layer, for example, a hydrogenated amorphous carbide layer.
21 to 23 form a hydrogenated amorphous silicon thin film type photoelectric conversion layer. The total film thickness of this photoelectric conversion layer is about 1 to 2 μm. Finally, a transparent electrode 24 that can transmit light is formed.

Furthermore, in order to absorb the inherent internal stress of the hydrogenated amorphous silicon film 22, a planar (linear in the figure) structural part 25 having a low silicon density is arranged in the photoelectric conversion layer.

In this example, the hydrogenated amorphous silicon film 2 has a thin film having a large internal stress of 4×1 (19 dyn/cm2 or more) and a large electron mobility of 0.45 cm2/V or more. However, due to the effect of Z stress absorption in the structure portion 25, no peeling of the film occurred, and a good element with high optical response could be formed.

On the other hand, the three-dimensional structure of the CCD image sensor substrate, which has a step on the surface due to the formation of the pixel electrode 17, is not appropriate;
In the case where the structural portion 25 was not provided, the film easily peeled off when similar lamination was performed, and no element was formed. Note that it is desirable that the line width of the structural portion 25 is 200 Å or less when viewed in cross section, and typically a value in the range of 20 to 100 is sufficient to obtain the effects of the present invention. Furthermore, it goes without saying that the structural portion 25 is not limited to a planar shape, but may be effective even if it is linear.

Next, in a two-story high-sensitivity solid-state imaging device in which hydrogenated amorphous silicon-based thin films are laminated as shown in FIG. Various methods for doing so will be explained in detail.

The first method is to change the thickness of the pixel electrode 17. The thickness of the pixel electrode 17 is 500 and 1000, respectively.
Figures 5(a) to 5(c) outline the layout of the structure 25 with low silicon density when the number of people is changed to 2,000.
) is schematically shown.

In the case of FIG. 5(a), the structure 25 having a low silicon density is hardly formed as observed in the cross-sectional TEM photograph. As the thickness of the pixel electrode 17 increases from FIG. 5(b) to FIG. 5(C), the structure portion 25 can be clearly observed.

That is, within the scope of this example, as the pixel electrode 17 becomes thicker, the structure portions 25 with lower density can be arranged more reliably. As a result, this stacked solid-state image sensor 100
The number of raw chips with film peeling per chip was as shown in the table below.

Here, as the film thickness of the pixel electrode 17 becomes thicker, the number of chips with film peeling decreases.
It goes without saying that this is an effect of ensuring that the structure 25 is placed and that the internal stress of the film is absorbed by the structure 25. In this example, photoelectric conversion films 21, 22, 23 made of amorphous silicon-based thin films are used.
The hydrogenated amorphous silicon film 22 used had an internal stress of 4 x 10<9>dyn/cm<2>. In addition, in this example, a photoelectric conversion film was formed by a mercury-sensitized photoCVD method using a low-pressure mercury lamp as an ultraviolet light source and silane gas as a raw material gas.
Similar effects can be obtained even if the film is formed by other film forming methods such as plasma CVD. For example, in the mercury-sensitized photoCVD method, in order to obtain a hydrogenated amorphous silicon film with an internal stress of 4 x 109 dyn/cm2 or more, the substrate temperature must be 230°C.
As described above, the silane gas pressure is set to 0.2 Torr or less, and the flow rate of silane, the amount of mercury contained in the silane gas, and the ultraviolet light intensity of the low-pressure mercury lamp may be set to appropriate conditions.

In this embodiment, a Ti electrode can be used as the pixel electrode 17, and in this case, a reactive ion etching (RIE) method, a wet etching method, or the like can be used as a patterning method for the Ti electrode. Preferably, there is less worry of undercutting the Ti pixel electrode.
It is better to use the RIE method, which has the characteristics of anisotropic etching.

As the RIE etching conditions for the Ti electrode, the following conditions can typically be used. Etching gas is CO
is 20-50scc11. B C13 is 30-60
scca+, Cl2 is 20~5osccI111 diluted He is 1000~SOOOsccm, the total gas pressure is 0.5~2Torr, and the discharge power is 20
What is necessary is just to set it in the range of 0-500W. By adjusting the above etching conditions, etched Ti
The Taber angle of the side wall portion of the pixel electrode can be controlled, and the electrode end portion 26 where the side wall portion and SiO2 are in contact can be provided with a curvature.

Here, the structure portion 25 having a low silicon density is a type of crystal defect, and this crystal defect portion 25 may come into contact with the pixel electrode 17 depending on the case. In this case, pixel electrode 1
7 and the transparent electrode 24 or between the adjacent pixel electrodes 17 via the crystal defect portion 25, which is undesirable. Therefore, by forming the end portion 26 of the pixel electrode 17 with a smooth continuous curve, crystal defects will not occur from the end portion, and the above leakage current can be prevented. As a method for forming this end portion smoothly, it is sufficient to appropriately select etching and sowing conditions in the etching, ie, wet etching or dry etching, when forming the pixel electrode.

Another method of arranging the structure 25 with low silicon density will be described. This is a hydrogenated amorphous silicon film 22
This is a method in which the layers are sequentially formed using different film formation methods.

FIG. 6 shows a hydrogenated amorphous silicon film 22 formed by plasma CVD.
This is an example of a combination of a lower layer formed by the D method and an upper layer formed by the mercury-sensitized CVD method. In FIG. 6, 221 is a hydrogenated amorphous silicon film formed by plasma CVD, and 22□ is a mercury-sensitized CVD film.
This is a hydrogenated amorphous silicon film formed by a method.

In this example, 13.56MHz is used as the plasma CVD method.
A method of decomposing silane gas and forming a film using high-frequency discharge was used. - For boat fishing, the temperature of the board installed on the ground electrode should be 200 to 300℃, the high frequency power should be 5 to 100W, the gas pressure should be 0.1 to 2 Torr, and the flow rate of silane introduced should be 5 to 1
A thin film may be formed at a speed of 0.00 sec.

In this example, the mercury-sensitized photoCVD method involves introducing ultraviolet light from a low-pressure mercury lamp into a silane gas containing a trace amount of mercury through a quartz window into a chamber, and decomposing the silane gas to deposit a thin film. The conditions are that the substrate temperature is 100 to 300℃, and the silane flow rate is 5 to 20℃.
0sec.

The gas pressure may be set to 0.05 to 2 Torr. In FIG. 6, 25. 252 is a structure with a low silicon density disposed within the thin film 221 formed by plasma CVD, and 252 is a structure with low silicon density disposed within the thin film 222 formed by mercury-sensitized CVD. - 27 indicated by a dotted chain line is a thin film 22. and 222
However, the silicon density does not necessarily decrease at that interface. As shown in this figure, with the boundary surface 27 as a boundary, a structure portion 2 with a low silicon density
5 is arranged differently. The reason for this will be explained below.

In general, according to the plasma CVD method, in addition to neutral species, ions mainly participate in film formation, and a distribution of seed potential occurs on the substrate, so there is anisotropy in growth, and the layering direction is It has a characteristic that the deposition rate is high, but the deposition rate is low in the direction perpendicular to the deposition direction, that is, in the direction parallel to the substrate.

For this reason, structures 25. with low silicon density are formed at an angle of 45° or more with respect to the horizontal direction of the substrate. is placed. On the other hand, the mercury-sensitized photoCVD method using a low-pressure mercury lamp is characterized in that the film formation is approximately isotropic because neutral species are involved in film formation.

Therefore, the structure 252 having a low silicon density can be arranged at an angle of approximately 45'' with respect to the horizontal plane of the substrate.

Next, as shown in FIG. 7, a hydrogenated amorphous silicon film 222 formed by mercury-sensitized photoCVD is formed as a lower layer, and a hydrogenated amorphous silicon film 222 formed by plasma CVD is formed as an upper layer.
An example of forming 2 is shown below. From the above principle, the seventh
In the figure, the manner in which the structural portion 25 is formed is reversed from that in FIG.

As described above, as shown in FIGS. 6 and 7, the arrangement of the structures with low silicon density can be changed as desired by using different film forming methods or a combination of different film forming methods. In these examples, two layers are laminated using two different film-forming methods, but the method is not limited to this, and two or more film-forming methods may be used, for example, in addition to the above-mentioned film-forming method, ECR plasma CVD
method and magnetron sputter link method.

Even if two layers are laminated using a combination of ion blating method, ion gun CVD method, direct excitation light CVD method, etc.
Likewise, the arrangement of the low silicon density portions can be varied as desired in each case. moreover,
It goes without saying that by changing the arrangement, the way the internal stress within the thin film is absorbed differs, and the way the film peels off also changes.

According to the research conducted by the present inventors, a case where thin films having the same internal stress are laminated is compared. The film is more difficult to peel off when the part with low silicon density is placed at an angle close to 45'' with respect to the horizontal direction of the substrate surface.For example, the same 4
When forming a hydrogenated amorphous silicon film having an internal stress of X 109 dyn/cm2, peeling of the film was less likely to occur when formed by mercury-sensitized photoCVD than by plasma CVD. In addition, a method of changing the growth rate of the thin film is also effective in order to change the arrangement method of a low-density portion, or to change the density itself of that portion.

Next, a third embodiment in which the present invention is applied to a thin film transistor will be described. FIG. 8 is a cross-sectional view showing a schematic structure of an inverted staggered TPT, and similarly to the previous example, a cross-sectional TEM photograph is schematically shown. After forming a gate electrode 31 on a glass substrate 30 and forming an insulating film 32 such as a silicon oxide film or a silicon nitride film by about 3,000 people, a hydrogenated amorphous silicon film 33 as an active layer is formed by, for example, 500 people. For example, 500 n+ type ice-hydrogenated amorphous silicon 834 for ohmic contacts are deposited, and source electrodes and drain electrodes are provided at 35 and 36.

In this embodiment, in addition to this structure, a structure with a low silicon density, indicated by 37 in the figure, is arranged in a plane (in a line in the figure). The structural parts 37 are not necessarily arranged in all the stack parts 32, 33, 34 and 36; in this embodiment, the structure part 37 is
．． 33 and 34.

In this example, as in the previous example, the hydrogenated amorphous silicon film 33 has an internal stress of 4 x 109 dyn/cm.
2 or more, and the electron mobility is also 0.45Cmaro2/
Although a thin film having a large value of V, S or more was used, it was possible to obtain good device characteristics without peeling of the film and having a sufficiently large value of field effect mobility. That is, it was possible to realize high-speed switching characteristics with a large on-current. On the other hand, it goes without saying that when the structural portion 37 was not provided, the film was more likely to peel off. Of course,
By arranging the structure portion 37, peeling of the insulating film 32 can be prevented.

Note that the present invention is not limited to the embodiments described above. In the example, an example was described in which an amorphous silicon film was used as the thin film, but the present invention is not limited to this and can be applied to non-single crystal thin films formed on a substrate, and can also be applied to insulating layers and metal layers. Can be applied. Moreover, it is also possible to apply it to the formation of an oxide superconductor thin film. Further, the present invention is not limited to solid-state image sensors and thin film transistors, but can be applied to contact type image sensors, and furthermore, can be applied to electronic devices having a structure in which a thin film is deposited on a substrate having steps. In addition, various modifications can be made without departing from the gist of the present invention.

[Effects of the Invention] As detailed above, according to the present invention, a structure having a small number of main constituent elements of the thin film per unit volume, that is, a small density, is provided inside a thin film deposited on a substrate having steps. Because there are
This structure can absorb stress. Therefore,
It is possible to prevent peeling of a thin film deposited on a substrate due to internal stress, and it is possible to realize a thin film stacked electronic device that can improve device characteristics and reliability. In particular, when applied to a solid-state image sensor, it is possible to achieve high-speed photoresponse characteristics, and when applied to a thin film transistor, it is possible to achieve high-speed switching characteristics with a large on-current.

[Brief explanation of the drawing]

1 and 2 are structural cross-sectional views showing a first embodiment of the present invention, FIG. 3 is a characteristic diagram showing the relationship between scanning position and peak intensity, and FIG. 4 is a structural cross-sectional view showing a first embodiment of the present invention. This is a cross-sectional view showing the schematic structure of a stacked solid-state image sensor, which is for explaining an example. FIG. 5 is a cross-sectional view showing the relationship between the pixel electrode thickness and the state of structure formation. FIG. 8 is a cross-sectional view showing an example in which a crystalline silicon film is formed in two layers. FIG. 8 is a cross-sectional view for explaining the third embodiment of the present invention and shows a schematic structure of a thin film transistor. FIG. FIG. 2 is a characteristic diagram showing the relationship between internal stress and electron mobility of a hydrogenated amorphous silicon film for explaining the problem. 1...Glass substrate, 2.4...Hydrogenated amorphous silicon film, 3.25.37...Structure section, 17...Pixel electrode, 21.23...Hydrogenated amorphous carbide Layer, 22...
- Hydrogenated amorphous silicon film, 24... Transparent electrode, 26... Electrode end, 27... Boundary surface, 30... Glass substrate, 31... Gate electrode, 32... Insulation Film, 33.34...Hydrogenated amorphous silicon film.

Claims (3)

[Claims] (1) In a thin film stacked electronic device formed by forming a thin film on a substrate having steps, a structure portion having a desired width is formed inside the thin film in which the density of the main constituent elements of the thin film is lower than in other parts. 1. A thin film stacked electronic device characterized by being arranged in a linear or planar manner. (2) The thin film stacked electronic device according to claim 1, wherein the thin film is an amorphous silicon film. (3) Steps are formed on the surface of the substrate depending on the presence or absence of a conductive layer, and the structure portion in which the density of the main constituent elements is smaller than others is formed in the thickness direction of the thin film due to the step difference on the substrate surface, and 2. The thin film stacked electronic device according to claim 1, wherein the side is not in contact with the conductive layer on the surface of the substrate.