JPH09293878A - Manufacture of thin-film transistor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for manufacturing a thin-film transistor which can provide respective self-aligned source, drain and channel regions, improve its operational speed, remarkably shorten its necessary processing time, eliminate any possibility of impurities mixed into a substrate, and reduce a time necessary for hydrogenation. SOLUTION: A photoresist film is subjected to a light exposure from the rear side of a glass substrate 10 (rear exposure). At this time, a gate electrode 11 is used as a mask and thus only the photoresist film 1 having the same width as a gate electrode 11 is left. The substrate is subjected to a plasma doping process with use of a photoresist film 1a on the photoresist film 15 extended in its pepriperal edge direction as a mask to introduce n-type impurities 16 into an amorphous silicon thin film 14. Therefore, source, drain and channel regions 14a, 14b and 14c are self-aligned respectively. Thereafter, the film 14 is irradiated with a laser beam for crystallization.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a thin film transistor having a source, drain and channel regions formed in a thin polycrystalline silicon film.

[0002]

2. Description of the Related Art In a liquid crystal display, a thin film transistor (TFT) integrated on a glass substrate for controlling liquid crystal.
Thin Film Transistor) is used.

Conventionally, this type of thin film transistor has a structure as shown in FIG. 5, for example. That is, a gate electrode 101 made of molybdenum tantalum (MoTa) is formed on a glass substrate 100, and this gate electrode 101 is formed.
An oxide film (Ta ₂ O ₅ ) 102 is formed thereon. On the glass substrate 100 including the oxide film 102, a silicon nitride (SiN _x ) film 103 and a silicon dioxide film (S
A gate insulating film made of iO ₂ ) 104 is formed, and a thin polycrystalline silicon film 105 is formed on the silicon dioxide film 104. This polycrystalline silicon film 105
The source region 105 is formed therein by introducing, for example, an n-type impurity.
a and the drain region 105b are formed respectively. On the polysilicon film 105, a silicon dioxide film (SiO ₂ ) 106 is selectively formed corresponding to the channel region 105c of the polysilicon film 105.
Polycrystalline silicon film 105 and silicon dioxide film 106
Is formed on the n ⁺ -doped polycrystalline silicon film 107, a source electrode 108 is formed on the n ⁺ -doped polycrystalline silicon film 107 so as to face the source region 105a, and a drain electrode 109 is formed so as to face the drain region 105b. Has been done.

Incidentally, this conventional thin film transistor is manufactured by the following method. That is,
Molybdenum tantalum (MoT)
a) After forming the film, the molybdenum tantalum film is patterned into a predetermined shape by etching to form a gate electrode 1
01 is formed. Thereafter, the gate electrode 101 is anodized to form an oxide film 102 on its surface.

Next, PECVD (Plasma Enhanced Chem)
A silicon nitride film 103, a silicon dioxide film 104 and an amorphous silicon thin film are continuously formed on the entire surface of the oxide film 102 by the ical vapor deposition method.

Next, by irradiating the amorphous silicon thin film with a laser beam from, for example, an excimer laser, the amorphous silicon thin film is once melted, and thereafter,
Cool to room temperature and crystallize. As a result, the amorphous silicon thin film becomes the polycrystalline silicon film 105. Subsequently, a silicon dioxide film 106 having a shape corresponding to the channel region is selectively formed on the portion of the polycrystalline silicon film 105 which will be the channel region, and then an n-type impurity such as phosphorus (P) or arsenic (As) is included. Then, an amorphous silicon thin film is formed, and the impurities are electrically activated as the n ⁺ -doped amorphous silicon film 107 by irradiating a laser beam with an excimer laser again.

Next, argon (A) is used as a sputtering gas.
After the aluminum (Al) film is formed on the entire surface by the sputtering method using r), the aluminum film and the n ⁺ -doped polycrystalline silicon film 107 are each patterned into a predetermined shape by etching, and the source region 1
05a and the source electrode 10 on the drain region 105b.
8 and the drain electrode 109 are formed. Subsequently, the channel region 105c is hydrogenated by hydrogen radicals and atomic hydrogen that are exposed to hydrogen and pass through the silicon dioxide film 106 to inactivate dangling bonds and the like. Through the above process, the conventional thin film transistor shown in FIG. 5 can be obtained.

[0008]

However, the above-mentioned conventional method has the following problems. First
Since a stopper (silicon dioxide film 106) is used when forming the source region 105a and the drain region 105b, a mask is also required to form the silicon dioxide film 106. And the process becomes complicated, and the production yield decreases. Also, with this method,
The source region 105a and the drain region 105b cannot be formed in a self-aligned manner with respect to the gate electrode 101, respectively. For this reason, it is necessary to provide a margin for mask alignment in consideration of an error in mask alignment, which increases the dimensions of the device and hinders high integration.
The gate electrode 101 and the source region 105a and the drain region 105b can be excessively overlapped, and this portion acts as a capacitor (parasitic capacitance), which hinders high-speed operation of the transistor.

Secondly, according to the conventional method, two amorphous silicon thin films are formed in order to form the polycrystalline silicon film 105 and the n ⁺ -doped polycrystalline silicon film 107, and each is irradiated with a laser beam to be crystallized. Because it has become 2
The laser beam has to be irradiated once, and the crystallization process takes a long time.

Third, since it is necessary to separately deposit a silicon dioxide film 106 as a stopper in addition to the silicon dioxide film 104 as a gate insulating film,
There is a problem that two steps are required, which also increases the time required for the process.

Fourth, in the conventional method, the silicon dioxide film 106 serving as a stopper is patterned by wet etching. However, the glass substrate 100 having a high etching rate is also removed during the wet etching. Sodium (N
Impurities such as a) may be mixed into the device and have an adverse effect.

Fifth, when hydrogenating the polycrystalline silicon film 105, the metal film (aluminum film) and the n ⁺ -doped polycrystalline silicon film 1 are formed on both ends of the channel region 105c.
There is a problem that the area through which the hydrogen passes and which is covered by 07 becomes narrow, and it takes a considerable time for sufficient hydrogenation.

The present invention has been made in view of the above problems, and an object thereof is to form source and drain regions formed in a polycrystalline silicon film in a self-aligned manner.
Provided is a method for manufacturing a thin film transistor, which has an improved operation speed, can significantly shorten the time required for the process, has no risk of impurities in the substrate being mixed, and can also shorten the time required for hydrogenation. Especially.

[0014]

According to a method of manufacturing a thin film transistor according to the present invention, a gate electrode is formed on a surface of a substrate, and a gate insulating film and an amorphous silicon film are formed in this order on the substrate including the gate electrode. Then, a step of forming a photoresist film on the amorphous silicon film, and a region corresponding to the gate electrode by exposing and selectively removing the photoresist film from the back surface of the substrate using the gate electrode as a mask. The photoresist film as a mask, a step of selectively introducing impurities into the amorphous silicon film by using the photoresist film as a mask to form a source region, a drain region, and a channel region, respectively. After the removal, a step of crystallizing the amorphous silicon film by irradiating with a laser beam is included. Here, it is preferable to introduce impurities into the amorphous silicon film by plasma doping, ion doping, or ion implantation.

In this method of manufacturing a thin film transistor, the photoresist film is exposed from the back surface of the substrate by using the gate electrode as a mask, so as to correspond to the region corresponding to the gate electrode, that is, the region where the channel is to be formed in the amorphous silicon film. The photoresist film remains only in the region. Then, by using this photoresist film as a mask, plasma doping or the like is performed from the surface side of the substrate to selectively introduce impurities into the amorphous silicon film, thereby forming a source region,
The drain region and the channel region are formed in a self-aligned manner.

[0016]

Embodiments of the present invention will be described below in detail with reference to the drawings.

1 (a) to 1 (d) and FIG. 2 (a),
(B) shows a method for manufacturing a thin film transistor according to an embodiment of the present invention in the order of steps. First, as shown in FIG. 1A, an aluminum (Al) film 1 is formed on the entire surface of a substrate, for example, a glass substrate 10 by a sputtering method using argon (Ar) as a sputtering gas.
A gate electrode 11 composed of 1a and an aluminum oxide (Al ₂ O ₃ ) film 11b is formed. Then, a silicon nitride (SiN _x ) film 1 is formed on the entire surface by, for example, a sputtering method using helium (He) as a sputtering gas.
After forming 2, the silicon dioxide (SiO ₂ ) is continued.
The film 13 is formed. Next, an amorphous silicon thin film 14 having a film thickness of, for example, 30 nm is formed on the silicon dioxide film 13 by a sputtering method using, for example, helium (He) gas as a sputtering gas, and subsequently, for example, 250 to 400.
By performing a heat treatment at a temperature of ° C, the helium contained in the amorphous silicon thin film 14 during film formation is released to the outside.

Next, a photoresist film is formed on the amorphous silicon thin film 14. Subsequently, as shown in FIG. 1B, the photoresist film is exposed from the back surface side of the glass substrate 10 by, for example, g-line (wavelength 436 nm) (back surface exposure). At this time, the gate electrode 11 serves as a mask, and only the photoresist film 15 having the same width as the gate electrode 11 remains. Subsequently, the substrate is heated to cause reflow, so that the photoresist film 15 is extended in the peripheral direction larger than the width of the gate electrode 11 to form a photoresist film 15a that spreads over the top of the amorphous silicon thin film 14.

After that, as shown in FIG.
Plasma doping is performed on the amorphous silicon thin film 14 using the plasma processing apparatus 20 shown in FIG.

In this plasma processing apparatus 20, the object to be processed 23 to be doped (that is, the process of FIG. 1B) is finished on the mounting table 22 in the processing container 21 which is evacuated by a pump (not shown). After accommodating the substrate 10), a reaction gas containing phosphorus (P) such as phosphine (P
H ₃ ), and in the case of p-type impurities, for example, boron (B)
A reaction gas containing, for example, diborane (B ₂ H ₆ ) is introduced, and a high frequency is introduced into the processing container 21 by a high frequency power supply (RF) 25 to generate plasma of n-type impurities or p-type impurities and to be processed. Impurities are introduced into the object 23.

In this embodiment, the temperature is lower than the heat resistant temperature of the photoresist film (eg, 150 ° C.), for example, a low temperature of 90 ° C., the inside of the processing container 21 is evacuated to 200 mTorr, and the reaction gas is 1% in argon, for example. Plasma doping using a gas containing phosphine (PH ₃ ) is performed for 5 minutes to use the photoresist film 15a as a mask to form n-type impurities 16 in the amorphous silicon thin film 14.
To introduce. Thereby, the source region 14a, the drain region 14b, and the channel region 14c are formed in a self-aligned manner. Since the photoresist film 15a is formed larger than the width of the gate electrode 11 by thermal reflow, the offset region 19 is provided between each of the source region 14a and the drain region 14b and the channel region 14c.
Are formed in a self-aligned manner. After that, the photoresist film 15a is removed.

Next, as shown in FIG. 1D, the amorphous silicon thin film 14 is irradiated with a laser beam by, for example, an excimer laser, so that the amorphous silicon thin film 14 has a source region 14a and a drain region. 14b and channel region 14c are once melted, and then cooled to room temperature to be crystallized. As a result, a thin polycrystalline silicon film 14A having the source region 14a, the drain region 14b and the channel region 14c is formed.

A beam having a wavelength absorbed by the amorphous silicon thin film 14, for example, a laser beam is used as the energy beam, and a pulse laser beam by an excimer laser is particularly preferable. As the excimer laser, a pulse laser beam (wavelength: 308 nm) using a XeCl excimer laser, a pulse laser beam (wavelength: 350 nm) using a XeF excimer laser, or the like is used.

Next, the polycrystalline silicon film 14A is patterned into a predetermined shape by etching, and then, as shown in FIG.
As shown in (a), on the source region 14a of the polycrystalline silicon film 14A, on the source electrode 18a made of, for example, aluminum and on the drain region 14b by, for example, a sputtering method using argon (Ar) as a sputtering gas. Drain electrode 1 also made of aluminum
8b are formed respectively. After that, as shown in FIG. 2B, plasma hydrogenation is performed in hydrogen plasma to hydrogenate the channel region 14c in the polycrystalline silicon film 14 and inactivate dangling bonds.

As described above, according to the method of manufacturing the thin film transistor according to the present embodiment, as shown in FIG. 1B, when the source region 14a, the drain region 14b and the channel region 14c are formed, the gate electrode 11 is formed. Is used as a mask to form a photoresist film 15 having the same width as that of the gate electrode 11, and then the resist film 15A obtained by extending the photoresist film 15 is used as a mask to selectively perform plasma doping. The source region 14a, the drain region 14b, and the channel region 14c including the offset regions 19 at both ends can be formed in the crystalline silicon thin film 14 in a self-aligned manner. Therefore, there is no need to provide a margin for mask alignment unlike the related art, and unnecessary parasitic capacitance is not formed due to the formation of the source and drain regions, and the high-speed operation of the transistor is not hindered. Further, the electric field can be weakened by the offset regions 19 provided at both ends of the channel region 14c, which is effective in reducing the off current.

Further, as shown in the step of FIG. 1D, the source region 14a, the drain region 14b and the channel region 14c in the amorphous silicon thin film 14 can be crystallized by one laser irradiation. The time required for the crystallization process can be shortened to half as compared with the conventional method. In addition, the P
It is not necessary to form a silicon oxide thin film pattern by the ECVD method, and the time required for this can be shortened.

Further, in the conventional method, the silicon dioxide film 102 as a mask at the time of forming the source / drain regions.
There is a risk that impurities contained in the glass substrate may be mixed in when patterning (FIG. 5) by wet etching, but since this method does not require the wet etching process as described above, such a risk may occur. It disappears.

Further, in this method, when hydrogenation is performed, both ends of the channel region 14c in the amorphous silicon thin film 14 are not covered with the metal film and the polycrystalline silicon film unlike the conventional method. The time required for hydrogenation can be significantly reduced.

Further, in the conventional method, a large number (6 sheets) in all steps.
However, in this method, the number of masks required for forming the gate electrode, the polycrystalline silicon film, and the source / drain electrodes is three, so the process is simplified.

Further, in this method, the source region 14a and the drain region 14b are formed in the amorphous silicon film 14 respectively.
Since it is formed by performing plasma doping inside, a polycrystalline silicon film 14A having a lower resistance than the conventional method can be obtained as shown in the following experimental example (FIG. 4).

[Experimental Example]

FIG. 4 shows a doped amorphous silicon film (A to A) formed by a conventional method (normal sputtering method).
FIG. 3 shows the relationship between the amount of energy when crystallized by laser beam irradiation and the sheet resistance for each of C) and the plasma-doped amorphous silicon film (D) formed by plasma doping according to the present invention. Here, the plasma-doped amorphous silicon film (D) according to the present invention is formed under the plasma-taping conditions shown in the above-mentioned embodiment. Further, as a conventional doped amorphous silicon film for comparison, a film thickness of 2
Three types of 0 nm (A), 30 nm (B), and 60 nm (C) were prepared. An excimer laser beam (wavelength 308 nm) was used as the laser.

As is clear from this figure, the sheet resistance of the plasma-doped amorphous silicon film (D) according to the present invention is significantly smaller than that of the conventional doped amorphous silicon film (A to C). Has become. In each sample, it was visually confirmed that the regions with low sheet resistance were crystallized well. The doped amorphous silicon film (B) having the same film thickness (30 nm) to be compared is 90 ° C.
When the film is formed at a low temperature, the film is amorphized again with a low energy of 240 mJ / cm ² , and is destroyed when the energy is further increased. Therefore, it is not possible to manufacture a crystal having a large crystal grain size, which increases the sheet resistance. On the other hand, the plasma-doped amorphous silicon film (D) of the present invention having the same thickness has a thickness of 280 mJ.
Even if it is irradiated with a high energy beam of / cm ² or more, it is not destroyed, and the sheet resistance is about 4.6 × 10 ^−4.
It showed a low resistance value of Ω · cm.

The present invention has been described with reference to the examples.
The present invention is not limited to the above-mentioned embodiment, but can be variously modified. For example, in the above embodiment, the plasma doping method is used when introducing impurities into the amorphous silicon film 14, but a doping method other than plasma may be used. In this case, the sheet resistance has a problem as compared with the case of the plasma doping method, but the above-mentioned other effects can be obtained. Further, in the above-mentioned embodiment, the gate electrode 11 is formed by the sputtering method, but other vapor deposition methods or C
It may be formed by using a VD (Chemical Vapor Deposition) method.

[0035]

As described above, according to the method of manufacturing a thin film transistor of the present invention, the photoresist film is exposed from the back surface side of the substrate using the gate electrode as a mask so that the photoresist is formed only in the region corresponding to the gate electrode. Since the film is left and the impurities are selectively introduced into the amorphous silicon film by performing plasma doping or the like from the surface side of the substrate using the photoresist film as a mask, the source region, the drain region and the channel region are formed. Can be formed in a self-aligned manner. Therefore, the operation speed of the device is improved, and the time required for the process can be significantly reduced. In addition, since wet etching is not required, there is no risk of impurities in the substrate being mixed, and the time required for hydrogenation is also reduced. It has the effect that it can be done.

[Brief description of drawings]

FIG. 1 is a cross-sectional view showing each step of a method for manufacturing a thin film transistor according to an embodiment of the present invention.

FIG. 2 is a cross-sectional view illustrating a process following the process in FIG.

3 is a cross-sectional view showing a configuration of a plasma processing apparatus used in the method shown in FIG.

FIG. 4 shows a case where a commonly used doped amorphous silicon film (A to C) and a plasma doped amorphous silicon film (D) formed by plasma doping used in the present invention are crystallized by laser beam irradiation. It represents the relationship between the amount of energy and the sheet resistance.

FIG. 5 is a cross-sectional view for explaining a structure and a manufacturing method of a conventional thin film transistor.

[Explanation of symbols]

10 ... Glass substrate, 11 ... Gate electrode, 12 ... Silicon nitride film, 13 ... Silicon dioxide film, 14 ... Amorphous silicon film, 14A ... Polycrystalline silicon film, 14a ... Source region, 14b ... Drain region, 14c ... Channel Area, 1
5, 15a ... Photoresist film, 16 ... N-type impurities, 1
7 ... Excimer laser beam, 18a ... Source electrode, 18
b ... Drain electrode, 19 ... Offset region

 ──────────────────────────────────────────────────の Continuing from the front page (72) Inventor Setsuo Usui 6-7-35 Kita-Shinagawa, Shinagawa-ku, Tokyo Inside Sony Corporation

Claims (3)

[Claims] 1. A gate electrode is formed on a surface of a substrate, a gate insulating film and an amorphous silicon film are formed in this order on the substrate including the gate electrode, and then a photolithographic film is formed on the amorphous silicon film. A step of forming a resist film, and a step of leaving the photoresist film only in a region corresponding to the gate electrode by exposing and selectively removing the photoresist film from the back surface of the substrate using the gate electrode as a mask. A step of selectively introducing impurities into the amorphous silicon film by using the photoresist film as a mask to form a source region, a drain region and a channel region, respectively, and a laser beam after removing the photoresist film. And a step of crystallizing the amorphous silicon film by irradiating the thin film transistor. 2. The method of manufacturing a thin film transistor according to claim 1, wherein the impurity is introduced into the amorphous silicon film by plasma doping. 3. The method of manufacturing a thin film transistor according to claim 1, wherein the plasma doping is performed at a temperature lower than a heat resistant temperature of the photoresist film.