JPH01155663A - Amorphous silicon thin-film transistor - Google Patents

Abstract

PURPOSE:To reduce the volume of a photocurrent because of the thin film thickness of an amorphous Si film by a method wherein an a-Si film is formed with a protrusion built therein so that a region thereof to be exposed to a light beam coming out of a substrate rear side may be formed thin. CONSTITUTION:An a-Si layer 4 is formed, provided with a protrusion. That is, Cr is deposited by spattering on a glass substrate 1 for the formation of a gate electrode 2. Next, an SiN film 3 is formed by plasma CVD, and then a non-dope a-Si layer 4 is formed, in that order, in contact with each other. The a-Si layer 4 is subjected to photoetching or the like whereby the region for a channel upper section is formed thick, and the region for source.drain electrodes thin. This design, with the region of the a-Si layer 4 to be exposed to a light beam coming out of the substrate rear surface being formed thin, a photocurrent may be reduced in volume.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an amorphous silicon thin film transistor (hereinafter abbreviated as a5iTFT) used in liquid crystal display panels, etc. The present invention relates to an a-S i TFT suitable for reducing current.

[Conventional technology]

The conventional a-S i T F T for liquid crystal display panels is
As described in JP-A-58-148458, a-8i
(iM) has a film thickness of about 2,000 people. This a-5i
When light is irradiated from the back side of the substrate on which the TFT is constructed, the light hits the a-8i near the gate electrode, holes are generated, and a photocurrent flows. When the film thickness of @a-si becomes 500 Å or more, the photocurrent increases. is about 1O-10A or more, and a
-The ○N/○FF ratio of S i T F T becomes small, and it becomes impossible to perform a sufficient function as a switching element.

[Problem that the invention seeks to solve]

In the above-mentioned conventional technology, a light shielding film is provided to block light from the upper side, but no consideration is given to the lower light. Therefore, a −
A large photocurrent is generated in S i T F T .

FIG. 2 shows a cross-sectional view of a conventional a5i TFT.

A light shielding film 8 made of a metal film is provided against light from the upper part of the substrate on which the TPT element is formed.

However, no effective measures have been taken against light from the back surface of the substrate (a backlight such as a fluorescent lamp is used for liquid crystal display). For this reason, the light 9 that has passed near the gate electrode 6 is irradiated onto the a-S i 4, which causes the aa-8i (i layer) film thickness, which generates a photocurrent, to be 200 mm.
For 0 people, a photocurrent of about 10-9 A is generated. As a method to reduce this generated photocurrent, a -S
It is also conceivable to make the i-film thin (500 Å or less).

FIG. 3 shows the relationship between the a-Si film thickness and the generated photocurrent IO. It can be seen that when the film thickness of a-8i becomes about 200, the photocurrent becomes about 1O-11A.

Although it is possible to reduce the generation of photocurrent by reducing the a-Si film thickness to 500 Å or less, the selectivity of the n/i layer is low in the process of removing the n layer above the channel ( In dry etching, the etching distribution, film thickness distribution, etc. are interrelated, and the i-layer is over-etched at the same time as the n-layer is removed, making it extremely difficult to remove the n-layer. To avoid this, the n layer and i
A method of etching by inserting a stopper layer between the layers has been put into practical use. After forming the i-layer, a stopper layer is formed using a silicon nitride film on the top of the channel.
An n layer is formed thereon. Since the selectivity between the n-layer and silicon nitride is high, the silicon nitride film serves to protect the underlying i-layer from etching when the n-layer is removed. Therefore, even in a large area, the thickness of one layer can be reduced. However, this method has problems such as etching of the a-Si when patterning the silicon nitride on the a-Si and an increase in the number of plasma CVD steps, which complicates the process.

An object of the present invention is to provide an a-S i T F T that is effective in reducing this photocurrent and can be easily formed.

[Means for solving problems]

The above purpose is achieved by forming the a-8i film in a convex shape.
That is, this problem can be solved by making the lower layer a-8i, which corresponds to the source and drain electrode portions of the a-S i T F T, thinner, and by forming the a-8i film, which corresponds to the upper part of the channel, thicker.

[Effect]

As described above (if the a-Si film is formed in a convex shape, when light is irradiated from the back side of the substrate, the a-
Since the Si film is thin, the photocurrent can be suppressed to a small level. On the other hand, the thick a-8i film prevents the etching distribution from being uneven in the upper part of the channel, which is blocked by the gate metal and is not exposed to light from the back side. It will not be partially lost due to the influence of shifting or the like.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG.

A Cr film is deposited on a glass substrate 1 by sputtering to form a gate electrode 2. Next, a silicon nitride film (SiN) 3° non-doped a-8i (
The a-8i (i-layer) 4 is connected and deposited in this order (Fig. 1(a)), and the a-8i (i-layer) 4 is made thicker in the part corresponding to the upper part of the channel and thinner in the part corresponding to the source and drain electrodes by photoetching. Process and form (Fig. 1(b)).

sauce.

It is sufficient that the thickness of the a-8i (i-layer) in the portion corresponding to the drain electrode is around 200 in view of charging current characteristics.

The a-S i (i layer) film thickness corresponding to the upper part of the channel is:
Film thickness distribution of a-8i (i, n).

Approximately 500 people are sufficient considering the dry etching distribution (when removing CnM), etc. However, regarding the upper part of the channel, the pA thickness of a-8i may be slightly thicker due to the presence of a light shielding film. Next, approximately 200 layers of phosphorus-doped a-Si (n layer) 5 are deposited on top of this a-3i (i layer) 4 using plasma CV.
D is deposited to form a-3i4.5 in a dot shape. A two-layer film of Cr-AQ is deposited by sputtering or vapor deposition to form the upper electrodes 6 of the source and drain (
Figure 1 (C)). In this state, the n layer above the channel is removed by dry etching (FIG. 1(d)).

In this etching, a large area (e.g. 220 x 180
m), an etching distribution of about ±20% generally occurs at the center and edges. Further, the a-8i deposited film thickness distribution is generally around ±10%. Taking both of these into consideration, the dry etching process completely removes 200 layers per layer.
i (i layer) remains. At this time, the a-8i remaining after etching
The (i-layer) preferably has a maximum thickness of 500 Å or less, which is taken into consideration from the photocurrent characteristics. After removing the n layer, a silicon nitride film is formed as a passivation film to complete the process (FIG. 1(e)).

〔Effect of the invention〕

In the a-8i TFT of the present invention, when light is irradiated from the back side of the substrate, the a-8i (i layer) is thin, so the photocurrent is suppressed to a small level (about 10-11 A). Since the 8i (i-layer) film is formed thickly in consideration of dry etching characteristics, accidents such as partial disappearance due to the influence of etching distribution etc. do not occur. Further, since there is no need to provide a stopper layer between the n-layer and the i-layer, there is an advantage that the process can be shortened.

[Brief explanation of the drawing]

FIG. 1 shows a -S i T F T of one embodiment of the present invention.
Figure 2 shows the manufacturing process diagram of the conventional a-5iTF.
FIG. 3 is a cross-sectional view of T, which shows the thickness of one layer of a-S i T F T and the photocurrent characteristics generated by light irradiation. 1...Glass substrate, 2...Gate electrode, 3...S
iN film, 4-=a-8i (i layer), 5-a-8i (n layer), 6'$l layer (b) (dl 2nd layer)

Claims (1)

[Claims] 1. An amorphous silicon thin film transistor comprising a gate electrode, a gate insulating film, an amorphous silicon thin film, and a source/drain electrode formed on a transparent substrate, wherein the amorphous silicon thin film substantially A first amorphous silicon thin film containing no impurities and a second amorphous silicon thin film containing impurities.
an amorphous silicon thin film, wherein the first amorphous silicon thin film is thin at a lower portion of the source/drain electrode and thick at a channel portion. .