JPH11168217A - Thin film transistor, manufacture thereof and liquid crystal display device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To conduct simultaneously an activation of implanted impurity ions and a recovery of damage generated due to an impurity implantation and to reduce the resistances of the source and drain regions of a TFT, the contact resistance between the source and drain electrodes of the TFT and the sheet resistivities of the source and drain regions. SOLUTION: A base insulating film 11, a P-type Si film 13, a gate insulating film 14 and a gate electrode 15 are formed in order on a glass substrate 10 and impurities are ion-implanted in the film 13 using the electrode 15 as a mask. After that, the lower part, which is located in the vicinity of the junction part between at least channel and source regions in the film 13 and in the vicinity of the junction part between at least the source region and a drain region in the film 13, of the substrate 10 is etched away into a recessed type as seen from the rear of the substrate 10, an excimer laser beam is applied through the rear of the substrate 10 and an activation of implanted impurity ions, a recovery of implantation damage and the like are simultaneously conducted.

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 (TFT) having high performance and high reliability, a method of manufacturing the same, and a liquid crystal display (LCD) having the thin film transistor. It is about.

[0002]

2. Description of the Related Art TFTs are used as switching elements and drivers for pixels in an active matrix type LCD.
Circuit or image sensor, and SRAM (St
atic Random Access Memory). In particular, active matrix LCDs have recently been developed as a semiconductor material using polycrystalline silicon (Polycrystallin).
The use of e silicon (p-Si for short) has led to a trend toward incorporating pixel driver circuits on the same substrate. In particular, if the maximum temperature of the process is 450 ° C or less,
The development of low-temperature p-Si TFT-LCD technology that enables the use of inexpensive glass substrates with a large area has been actively developed, and some of them have been put into practical use. Generally, p-Si
TFTs have a higher carrier mobility than amorphous silicon (a-Si) TFTs. In addition, by forming source / drain regions in a self-aligned manner with gate electrodes, miniaturization and inter-gate / drain connections can be achieved. , It is possible to form a high-speed driver circuit. Therefore, the pixel and the driver circuit can be integrally formed, so that the cost reduction and the high definition of the LCD can be realized.

Hereinafter, an example of a conventional method for manufacturing a p-Si TFT will be described with reference to the drawings. FIG. 9 shows p-Si
FIG. 4 is a process cross-sectional structure diagram illustrating a method for manufacturing a TFT.

First, a p-Si13 film is formed on a translucent glass substrate 10 having a strain point of about 600 ° C., and is patterned in an island shape. Next, a gate insulating film 14 is formed on the p-Si 13, and a conductive metal thin film serving as the gate electrode 15 is formed on the gate insulating film 14. Pattern. Thereafter, using the gate electrode 15 as a mask, impurity ions such as phosphorus and boron are implanted into the p-Si 13 by ion doping to form a channel region 13a, a source region 13b, and a drain region 13c in the p-Si 13. (FIG. 9A).

[0005] Next, using an apparatus such as an annealing furnace, 300
By performing a thermal annealing process or the like for several hours to several tens of hours in an atmosphere at a temperature of from about 600 ° C. to about 600 ° C., the implanted impurities are replaced (activated) at the lattice positions (FIG. 9B).

Next, an interlayer insulating film 16 is formed, a contact hole 17 is opened, and a source / drain electrode 18 is formed.
A conductive metal thin film is formed and is patterned into a predetermined shape. Finally, hydrogenation treatment is performed on the p-Si 13 at 350 ° C. for 2 hours in a hydrogen plasma atmosphere to complete the p-Si TFT (FIG. 9C).

[0007]

As described above, conventionally, after impurity ions are implanted into the source and drain regions of p-Si, carriers (electrons and holes) that contribute to electric conduction from the implanted impurities are conventionally used. To release
The substitution (activation) of the impurity at the lattice position is performed.

However, when impurity ions are implanted into p-Si, the interface between the p-Si and the gate insulating film and the crystal grain boundaries of the p-Si, particularly the source and drain ends, that is, the channel region and the source region are formed. In addition, p-Si is greatly damaged because many traps, levels, dangling bonds, and the like are formed near the junction of the drain region. This damage, especially damage that has entered the vicinity of the junction between the channel region and the source region and the drain region, has the problem of causing deterioration of the initial characteristics of the TFT and deterioration of the reliability.

In a conventional method of manufacturing a TFT, p-Si1
3 is implanted with impurity ions,
The impurity ions implanted by thermal annealing at a relatively low temperature of 0 ° C. are activated. Generally, heat treatment at a high temperature close to 1000 ° C. is necessary to activate impurities. However, when an inexpensive glass substrate is used, heat treatment cannot be performed at a temperature equal to or higher than the strain point of the substrate. There is a problem that it takes a long time for activation of the device and the throughput is poor.

Further, in the above method, even though the implanted impurity ions can be activated, p-type ions can be activated by ion bombardment during the implantation.
There is a problem that the damage entering Si cannot be completely removed. In particular, the vicinity of the junction between the channel region 13a, the source region 13b, and the drain region 13c has a structure in which the junction coincides with the end of the gate electrode. Therefore, local heating is performed using an energy beam such as an excimer laser. Therefore, even if activation is carried out, there is a problem that it is shielded by the gate electrode 15 and it is very difficult to remove damage.

Further, in the conventional method of manufacturing a TFT, in the step of implanting impurity ions, the channel region 13a into which the impurity ions are not implanted maintains the crystallinity after crystallization, but the source region into which the impurity ions are implanted. Area 1
The 3b and the drain region 13c have a problem that the periodicity is disturbed by the ion bombardment, and particularly the upper layer becomes amorphous. Since the crystallinity of the amorphous portion is not recovered by activation by conventional thermal annealing,
Source region 13b and drain region 13c
In some cases, the contact resistance between the drain electrodes 18 and the sheet resistance of the source region 13b and the drain region 13c increase.

The present invention solves the above-mentioned conventional problems. First, the present invention simultaneously activates impurity ions implanted at a temperature lower than the strain point of the substrate and recovers damage caused by the impurity implantation. Second, the contact resistance between the source and drain regions of the TFT and the source / drain electrodes and the sheet resistance of the source and drain regions can be reduced, thereby simplifying the process and improving the throughput. It is another object of the present invention to provide a TFT having high performance and high reliability, a manufacturing method thereof, and a liquid crystal display device.

[0013]

SUMMARY OF THE INVENTION In order to activate implanted impurity ions and recover implantation damage at the same time,
The present inventors have studied various methods while performing a life test on a TFT. Firstly, we considered that an energy beam such as an excimer laser was irradiated from the substrate surface after impurity ion implantation. In this method, however, the junction between the channel region and the source and drain regions shielded by the gate electrode was examined. Because the energy beam is not irradiated,
It was found that injection damage such as traps, levels, and dangling bonds generated near the junction could not be recovered, and reliability deterioration such as hot carrier deterioration in which carriers were captured was not improved. Therefore, an energy beam such as an excimer laser was irradiated from the back surface (the other surface) of the substrate after the impurity ions were implanted. However, when an energy beam such as a XeCl excimer laser (wavelength: 308 nm) is irradiated from the back surface of the Corning # 7059 glass substrate that the present inventors experimented, the laser light is transmitted by less than 50%, which is extremely low. ineffective. Therefore, the present inventors have further devised, and in order to more efficiently irradiate the energy beam from the back surface of the substrate to the p-Si film or the like, a channel region such as p-Si formed on one surface of the substrate is formed. The present inventors have found a method for irradiating an energy beam from the back surface of the substrate after etching the substrate below the vicinity of the junction between the source region and the drain region in a concave shape as viewed from the back surface of the substrate, and have reached the present invention. The TFT obtained according to the present invention can simultaneously activate an implanted impurity and recover implantation damage, can lower the sheet resistance and the like of the source region and the drain region, and can achieve a much higher activation rate than the conventional one. At the same time, the time required for the carrier mobility to be reduced to 50% of the initial value in a TFT life test was improved by about two orders of magnitude.

The solution of the present invention relates to a thin film transistor, a method of manufacturing the same, and a liquid crystal display device.

The first thin film transistor of the present invention is a thin film transistor having at least a light transmitting glass substrate and a non-single-crystal silicon thin film formed on the light transmitting glass substrate, wherein the light transmitting glass substrate Wherein the thickness of at least the lower region near the junction between the channel region and the source and drain regions of the non-single-crystal silicon thin film is thinner than the other portions.
According to the present invention, energy efficiency when an energy beam is irradiated from the other surface of the translucent glass substrate is improved, activation of implanted impurity ions and implantation damage which could not be recovered conventionally, particularly non-implantation. Recovery of implantation damage that has entered the vicinity of the junction between the channel region and the source and drain regions of the single crystal silicon thin film can be performed at the same time, improving the activation rate of implanted impurity ions and deteriorating hot carriers where carriers are captured. TFT with high performance and high reliability by improving the resistance to reliability deterioration etc.
Can be obtained.

In a first method of manufacturing a thin film transistor according to the present invention, a step of forming a polysilicon thin film on one surface of a light transmitting glass substrate, and a step of forming a gate insulating film on the light transmitting glass substrate and the polysilicon thin film. Forming a film;
Forming a gate electrode on the gate insulating film; selectively injecting a P-type or N-type impurity into the polysilicon thin film using the gate electrode as a mask; A step of making the thickness of the lower region near the junction between the channel region and the source region and the drain region of the polysilicon thin film thinner than the other portions, and irradiating an energy beam from the other surface of the translucent glass substrate, Simultaneously activating the implanted P-type or N-type impurities and recovering damage caused by the implantation of the P-type or N-type impurities. According to the present invention, energy efficiency when an energy beam is irradiated from the other surface of the translucent glass substrate is improved, activation of implanted impurity ions and implantation damage which could not be recovered conventionally, particularly non-implantation. Recovery of implantation damage that has entered the vicinity of the junction between the channel region and the source and drain regions of the single crystal silicon thin film can be performed at the same time, improving the activation rate of implanted impurity ions and deteriorating hot carriers where carriers are captured. Thus, a TFT with high performance and high reliability can be obtained by improving the resistance against the reliability deterioration.

In the above structure, before selectively implanting P-type or N-type impurities into the polysilicon thin film using the gate electrode as a mask, the gate insulating film on the source region and the drain region of the polysilicon thin film is removed. When the step of performing the step is performed, the impurity ions can be implanted at a lower acceleration voltage, so that implantation damage can be reduced.

In the above structure, before irradiating the energy beam from the other surface of the translucent glass substrate,
If a step of removing the gate insulating film on the source region and the drain region of the polysilicon thin film is included, the energy beam reaching the polysilicon thin film is diffused into the gate insulating film on the source region and the drain region of the polysilicon thin film. Without being absorbed by the polysilicon thin film
Energy efficiency is improved.

Next, according to a second method of manufacturing a thin film transistor of the present invention, a step of forming an amorphous silicon thin film on one surface of a light transmitting glass substrate, and a step of forming a gate on the light transmitting glass substrate and the amorphous silicon thin film. Forming an insulating film, forming a gate electrode on the gate insulating film, selectively implanting a P-type or N-type impurity into the amorphous silicon thin film using the gate electrode as a mask, Reducing the thickness of at least the lower region near the junction between the channel region, the source region, and the drain region of the amorphous silicon thin film in the translucent glass substrate to be thinner than other portions, and the other surface of the translucent glass substrate Irradiating an energy beam from the crystallization of the amorphous silicon thin film, the implanted P-type or Activation of type impurity, and a step of simultaneously performing the recovery of damage caused by the injection of the P-type or N-type impurities. According to the present invention, the energy efficiency when the energy beam is irradiated from the other surface of the translucent glass substrate is improved, and the crystallization of the amorphous silicon thin film and the activation of the implanted impurity ions,
Recovery of implantation damage that could not be recovered conventionally, particularly implantation damage that has entered the vicinity of the junction between the channel region and the source and drain regions of the non-single-crystal silicon thin film, can be simultaneously performed, and the activity of the implanted impurity ions can be reduced. High performance, by improving the conversion rate and improving the resistance to reliability degradation such as hot carrier degradation in which carriers are captured.
A highly reliable TFT can be obtained. Further, since the amorphous silicon thin film is simultaneously crystallized, the activation rate and the resistance to reliability deterioration such as hot carrier deterioration are further improved as compared with the above-described first manufacturing method of the thin film transistor of the present invention.

In the above structure, before selectively implanting P-type or N-type impurities into the amorphous silicon thin film using the gate electrode as a mask, the gate insulating film on the source region and the drain region of the amorphous silicon thin film is removed. When the step of performing the step is performed, the impurity ions can be implanted at a lower acceleration voltage, so that implantation damage can be reduced.

In the above structure, before irradiating the energy beam from the other surface of the translucent glass substrate,
If a step of removing the gate insulating film on the source region and the drain region of the amorphous silicon thin film is included, the energy beam reaching the amorphous silicon thin film is diffused into the gate insulating film on the source region and the drain region of the amorphous silicon thin film. Since it is absorbed by the amorphous silicon thin film without being absorbed, the energy efficiency is improved.

Since the liquid crystal display device of the present invention has at least the thin film transistor of the present invention, a liquid crystal display device having high performance and high reliability can be obtained.

[0023]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to FIGS. 1 and 6 to 8 are sectional structural views for explaining the first embodiment, FIGS. 2 and 3 are process sectional structural views for explaining the first and second embodiments, and FIG. FIG. 5 is a characteristic diagram for explaining the first embodiment. FIG. 5 is a characteristic diagram for explaining the first embodiment, and the same reference numerals denote the same objects throughout the drawings.

(Embodiment 1) A TFT used in an active matrix type liquid crystal display device as an embodiment of the present invention will be described with reference to FIG.

In the figure, reference numeral 10 denotes, for example, a Corning # 7059 glass substrate (thickness: 1.1 mm) having a strain point of 590 ° C.
The lower portion of the p-Si semiconductor layer 13 formed on one surface of the glass substrate 10, which includes at least the vicinity of the junction 13 d of the channel region 13 a and the source region 13 b and the drain region 13 c, is formed of the glass substrate. It is etched away in a concave shape when viewed from the other surface (back surface). On the glass substrate 10, for example, SiO2
A base insulating film layer 11 made of is deposited. On the base insulating film layer, a semiconductor layer 13 made of, for example, a 500-nm-thick p-Si thin film is patterned into a predetermined shape, and this p-Si semiconductor layer 13 is formed by, for example, laser crystallization, and a channel region is formed. A source region 13b and a drain region 13c are formed on both sides of the source region 13a. The p-Si 13 contains impurity ions such as phosphorus and boron implanted by the ion doping method. Further, on the p-Si13, for example insulating film made of SiO ₂ having a thickness of 1000 Å (gate - gate insulating film) 14 is deposited, on top thereof, for example, film thickness 300
A gate electrode 15 made of 0 ° Al is patterned in a predetermined shape. Then, on the gate electrode 15,
For example, an interlayer insulating film 1 made of SiO ₂ having a thickness of 4000 °
The gate electrode 15 is covered with the interlayer insulating film 16. Further, the gate insulating film 14 and the interlayer insulating film 1
In FIG. 6, contact holes 17 are formed so as to reach the source region 13b and the drain region 13c of p-Si 13, respectively. And the Al film are patterned into a predetermined shape, and the contact hole 17 is filled with the Ti film and the Al film to form the source region 1.
3b and the source / drain electrode 18 for making contact with the drain region 13c are formed.
The FT is configured.

Next, an example of a method of manufacturing the above-described TFT will be described with reference to FIG. First, for example, as a translucent glass substrate, for example, SiO _{2 is} placed on a Corning # 7059 glass substrate (thickness: 1.1 mm) 10 having a strain point of 590 ° C.
Is formed at 450 ° C. to a thickness of 2000 ° by a normal pressure CVD method.
i: H is deposited to a thickness of 500 ° by a plasma CVD method. Next, a-Si: H at 450 ° C., 6
A 0 minute dehydrogenation treatment is performed. This step is intended to prevent occurrence of ablation of the Si film due to desorption of hydrogen during crystallization described later. After dehydrogenation, for example, an XeCl excimer laser having a wavelength of 308 nm is irradiated from one surface of the glass substrate 10 by a pulse laser annealing method to crystallize a-Si. -Si13 (FIG. 2A).

Next, an insulating film (gate insulating film) 14 made of, for example, SiO _{2 is} formed on the p-Si 13 to a thickness of 1000 ° at 450 ° C. by a normal pressure CVD method. The film is sputtered to a thickness of 3000 °. Then, the Al film is subjected to wet etching using an Al etchant for about 1 minute, and is patterned into a predetermined shape to form a gate electrode 15 (FIG. 2).
(B)).

Next, using the gate electrode 15 as a mask, p
Impurities such as phosphorus and boron are ion-implanted into the -Si 13 by ion doping, for example, and a source region 13b and a drain region 13c are formed on both sides of the p-Si 13 with the channel region 13a interposed therebetween (FIG. 2 ( c)).

Next, a resist film 19 is applied as an etching preventive film on both surfaces (front and back surfaces) of the glass substrate 10 and only the resist film on the back surface is patterned into a predetermined shape by photolithography.
In a BHF (buffered hydrofluoric acid) solution (a mixture of HF and NH ₄ F at a predetermined concentration) to form a p-Si semiconductor layer 13.
At least the lower portion of the glass substrate 10 near the junction 13d between the channel region 13a, the source region 13b, and the drain region 13c is etched away in a concave shape when viewed from the back surface. Thereby, the thickness of at least the lower region near the junction between the channel region, the source region, and the drain region of the non-single-crystal silicon thin film in the glass substrate 10 becomes thinner than other portions. The glass substrate 10 is rapidly etched, having a high surface property of about 0.15 mm in about 120 minutes, about 0.30 mm in about 240 minutes, and about 0.6 mm in 360 minutes. The present inventors performed etching for 360 minutes to improve the transmittance of laser light described below, and a part of the glass substrate 10 was 0.6 m thick.
m was removed by etching. Since the etching rate depends on the solution concentration, the desired etching rate can be obtained by changing the solution concentration (FIG. 2D).

Next, for example, a XeCl excimer laser having a wavelength of 308 nm is irradiated from the back surface of the glass substrate 10 by a pulse laser annealing method to replace (activate) implanted impurities such as phosphorus and boron at lattice positions (activation), and Recovery of damage due to ion bombardment near the region 13b, the drain region 13c, and the junction 13d is simultaneously performed.

Incidentally, according to the confirmation experiment of the present inventors, when a # 7059 glass substrate having a thickness of 1.1 mm is used, 3
The laser light having a wavelength of 08 nm transmits only about 50% of the laser light.
If the back of 0 is etched away in a concave shape when viewed from the back,
The laser light transmittance of the etched part is improved,
60% at 0.8mm thickness, 70% at 0.5mm thickness
It is known that the above laser light is transmitted. In order to increase the transmittance of the laser beam and more efficiently activate and recover from the injection damage, the thinner the substrate, the more desirable it is. The present inventors removed a part of the back surface of the # 7059 glass substrate 10 having a thickness of 1.1 mm by etching by 0.6 mm to have a thickness of 0.5 mm.
The above laser light is transmitted (see FIG. 2).
(E)).

Next, an interlayer insulating film 16 made of, for example, SiO _{2 is} formed by atmospheric pressure CVD at 450 ° C. so as to have a thickness of 4000 °. Thereafter, the contact hole 17 is opened, and for example, a Ti film and an Al film
Sputtering to Å and 7000Å. Then, the Ti film and the Al film are patterned into a predetermined shape by dry etching using a BCl ₃ / Cl ₂ gas to form a source / drain electrode 18 (FIG. 2F).

Finally, when a plasma hydrogenation treatment is performed at 350 ° C. for 2 hours, for example, the p-Si TFT is completed. This step is intended to inactivate dangling bonds at the p-Si13 grain boundaries and defects existing at the interface between the p-Si13 and the gate insulating film 14 (FIG. 2 (g)).

When a TFT is manufactured by the above-described manufacturing method, the energy efficiency when an energy beam is irradiated from the back surface of the glass substrate 10 is improved. Therefore, the activation of the implanted impurity ions and the conventional recovery cannot be performed. Recovery of implantation damage, particularly implantation damage that has entered the vicinity of the junction between the channel region and the source and drain regions of the non-single-crystal silicon thin film (specifically, a polycrystalline silicon thin film) can be performed simultaneously. . As a result, the activation rate of the implanted impurity ions and the resistance to reliability deterioration such as hot carrier deterioration in which carriers are captured are improved, and the initial characteristics and reliability of the TFT are improved, and high performance and high reliability are achieved. A TFT having the same can be obtained. The inventors of the present invention have performed a life test on a TFT and confirmed that the TFT manufactured by the above-described method of simultaneously performing the activation and the recovery of the implantation damage has a longer TFT life than the TFT manufactured by the conventional method. Was improved by about two orders of magnitude (FIG. 5).

Since the XeCl excimer laser used in the present embodiment is a pulse laser, the irradiation time is very short, and the time required for activation is significantly reduced as compared with the conventional activation method by thermal annealing. it can. Therefore,
Throughput can be significantly improved.

Although impurity ions are implanted into the p-Si 13 through the gate insulating film 14 in this embodiment, the impurity ions are implanted into the source region 13b and the drain region 13c of the p-Si 13 as shown in FIG. Upper gate insulating film 14
May be removed before impurity implantation. in this case,
70 ke for impurity ion implantation through the gate insulating film 14
The acceleration voltage, which is required to be about V, can be reduced to 10 keV or less, which is preferable, and the implantation damage can be reduced. In order to reduce injection damage, the lower the acceleration voltage is, the more preferable it is.
The range is 8 keV.

In this embodiment, the energy beam is irradiated from the back surface of the glass substrate 10 in a state where the gate insulating film 14 is deposited on the p-Si 13, but this is applied to the source of the p-Si 13 as shown in FIG. The energy beam may be irradiated after the gate insulating film 14 on the region 13b and the drain region 13c is removed. In this case, p-Si13
The energy beam reaching the gate insulating film 1 on the source region 13b and the drain region 13c of p-Si 13
4 is absorbed by the p-Si 13 without being diffused, and thus has an advantage that energy efficiency is improved as compared with the case where irradiation is performed with the gate insulating film 14 deposited.

Further, in the present embodiment, the back surface of the glass substrate 10 is etched away from the vicinity of the junction 13d of the p-Si 13 to the source region 13b and the drain region 13c, but as shown in FIG. 13
It suffices that at least the back surface of the glass substrate 10 only near d is removed by etching. However, in this case, p-Si1
3, the transmittance of the laser light to the source region 13b and the drain region 13c is reduced.
Before or after the step of irradiating the laser beam from the back surface (FIG. 2E), the step of irradiating the laser beam from the front surface of the glass substrate 10 may be combined. In addition to this, also in the present embodiment, the step of irradiating the laser light from the surface of the glass substrate 10 before or after the step of irradiating the laser light from the back surface of the glass substrate 10 (FIG. 2E). When combined, the activation rate of the implanted impurity ions and the resistance to reliability deterioration such as hot carrier deterioration in which carriers are captured are further improved.

In this embodiment, a XeCl excimer laser is used as an energy beam. However, similar effects can be obtained by using various lasers such as an argon laser, strong light such as infrared light, or an electron beam. .

(Embodiment 2) As another embodiment of the present invention, a method of manufacturing a TFT used in an active matrix type liquid crystal display device will be described with reference to FIG. This embodiment is different from the first embodiment in that the crystallization process for forming the source / drain region and the channel region of the thin film transistor is performed simultaneously with the activation of the source / drain region.

First, for example, as a translucent glass substrate,
On a Corning # 7059 glass substrate (thickness: 1.1 mm) 10 having a strain point of 590 ° C., a base insulating film 11 made of, for example, SiO _{2 is} formed at 450 ° C. by atmospheric pressure CVD at a film thickness of 20.
After the film is formed to have a thickness of 00 °, for example, a-Si: H is formed to have a thickness of 500 ° by a plasma CVD method, and is islanded into a predetermined shape to obtain a-Si: H12 (FIG. 3). (A)).

Next, on the a-Si: H12, for example, SiO
After an insulating film (gate insulating film) 14 made of 2 is formed at 450 ° C. to a thickness of 1000 ° by normal pressure CVD, an Al film is sputtered to a thickness of 3000 °, for example. Then, the Al film is subjected to wet etching using an Al etchant for about 1 minute, and patterned into a predetermined shape to form a gate electrode 15 (FIG. 3B).

Next, using the gate electrode 15 as a mask, a
-Si: phosphorus is added to H12 by, for example, an ion doping method,
Impurities such as boron are ion-implanted, and a-Si: H12
The source region 1 on both sides of the channel region 12a.
2b and the drain region 12c are formed (FIG. 3
(C)).

Next, a resist film 19 is applied as an etching preventive film on the entire surface (front and back surfaces) of both surfaces of the glass substrate 10, and only the resist film on the back surface is patterned into a predetermined shape by photolithography.
In a BHF (buffered hydrofluoric acid) solution (a mixed solution of HF and NH 4 F at a predetermined concentration) to form at least a channel region 12 a and a source region 12 of the a-Si: H semiconductor layer 12.
The glass substrate 10 in the lower part near the junction 12d of the drain region 12b and the drain region 12c is etched and removed in a concave shape when viewed from the back surface. Thereby, the thickness of at least the lower region near the junction between the channel region, the source region, and the drain region of the amorphous silicon thin film in the glass substrate 10 becomes thinner than other portions. Note that the etching rate and the like are the same as those in the first embodiment, and a description thereof will be omitted (FIG. 3D).

Next, 450 to a-Si: H12
A dehydrogenation treatment is performed at 60 ° C. for 60 minutes. This step is intended to prevent occurrence of ablation of the Si film due to desorption of hydrogen during crystallization described later. Thereafter, for example, a XeCl excimer laser having a wavelength of 308 nm is irradiated from the back surface of the glass substrate 10 by a pulsed laser annealing method to crystallize a-Si: H12 and replace implanted impurities such as phosphorus and boron at lattice positions (activity). And the recovery from damage due to ion bombardment near the source region 12b, the drain region 12c, and the junction 12d. Thus, a source region 13b and a drain region 13c are formed on both sides of the p-Si 13 with the channel region 13a interposed therebetween. Note that the relationship between the thickness of the glass substrate 10 and the transmittance of the laser beam is the same as in the first embodiment, and a description thereof will be omitted (FIG. 3E).

Next, FIGS. 2F to 2G of the first embodiment.
The contact hole 1 is formed by a process similar to the process shown in FIG.
7, the source / drain electrode 18 and the plasma hydrogenation process complete the p-Si TFT (FIG. 3).
(F)-(g)).

When a TFT is manufactured by the above-described manufacturing method, energy efficiency when an energy beam is irradiated from the back surface of the glass substrate 10 is improved.
12, activation of the implanted impurity ions, and implantation damage that could not be recovered conventionally, particularly implantation near the junction between the channel region and the source and drain regions of the non-single-crystal silicon thin film. You can recover all damage at the same time. Thus, the activation rate of the implanted impurity ions and the resistance to reliability degradation such as hot carrier degradation in which carriers are captured are significantly improved as compared with the case of using the TFT manufacturing method described in the first embodiment. As a result, the initial characteristics and reliability of the TFT are greatly improved, and a TFT having high performance and high reliability can be obtained. According to the confirmation by the present inventors, in a TFT manufactured by the method of simultaneously performing the above-described crystallization, activation, and recovery of implantation damage, recovery of the above implantation damage is performed in a state where the semiconductor layer is amorphous. Therefore, the damage can be recovered more easily than in the first embodiment, so that the TFT life is improved by about three digits or more as compared with the TFT manufactured by the conventional method.

Since the XeCl excimer laser used in the present embodiment is a pulse laser, the irradiation time is very short, and the time required for activation is significantly reduced as compared with the conventional activation method by thermal annealing. it can. Moreover,
Since crystallization of a-Si can be performed at the same time, not only can throughput be significantly improved,
It is also possible to simplify the process.

In this embodiment, a-Si: H12
Was implanted through the gate insulating film 14 in the same manner as in the first embodiment, but after removing the gate insulating film 14 on the source region 12b and the drain region 12c of a-Si: H12, Even if the injection is performed, the same effect as in the first embodiment can be obtained.

In this embodiment, a-Si: H12
The glass substrate 10 with the gate insulating film 14 deposited on
The energy beam was irradiated from the back surface of the substrate, but the energy beam was irradiated after removing the gate insulating film 14 on the source region 12b and the drain region 12c of a-Si: H12 as in the first embodiment. The same effect as in the first embodiment can be obtained.

Further, in this embodiment, a-Si: H1
2 from the vicinity of the junction 12d to the source region 12b and the drain region 12c, the back surface of the glass substrate 10 was removed by etching.
It is sufficient that at least the back surface of the glass substrate 10 in the vicinity of the joint 12d is removed at least by etching. However, in this case, the laser light is irradiated from the back surface of the glass substrate 10 as in the first embodiment in order to cover a decrease in the transmittance of the laser light to the source region 12b and the drain region 12c of a-Si: H12. Before or after the step (FIG. 2E), a step of irradiating a laser beam from the surface of the glass substrate 10 may be combined. In addition, not only this, but also in the above-described embodiment, before or after the step of irradiating the laser light from the back surface of the glass substrate 10 (FIG. 2E).
By combining the steps of irradiating the laser light from the surface of No. 0, the same effect as in the first embodiment can be obtained.

In this embodiment, a XeCl excimer laser is used as an energy beam. However, similar effects can be obtained by using various lasers such as an argon laser, strong light such as infrared light, or an electron beam. Needless to say.

(Embodiment 3) As another embodiment of the present invention, an active matrix type liquid crystal display device will be described with reference to FIG.

The active matrix type liquid crystal display device of the present invention has a structure in which a pair of glass substrates 40 and 41 are arranged to face each other, and a liquid crystal layer 42 is sealed in the gap. On a glass substrate 40 serving as a TFT array substrate,
The scanning lines 43 and the signal lines 44 arranged in a matrix and the TFTs 4 arranged at the intersections of the scanning lines 43 and the signal lines 44.
5a, a scanning line driving circuit 46 and a signal line driving circuit 47 which are peripheral driving circuits are formed. The TFT 45a is
The TFT shown in FIG. 1 described in the first embodiment is connected to the pixel electrode 48, and a pixel signal is transmitted to the pixel electrode 48.
It functions as a switching element for writing to the memory.
In addition, the scanning line driving circuit 46 and the signal line driving circuit 47, which are peripheral driving circuits, are the same as those shown in FIG.
And a CMOS inverter 46b formed by combining the TFTs shown in FIG. The TFTs 45a and 4
5b are all the above-described Embodiment 1 or Embodiment 2.
Are manufactured according to the manufacturing method shown in FIG. On the other hand, on a glass substrate 41 serving as a counter substrate, a counter electrode 49 and a color filter 50 are formed on the liquid crystal layer 42 side.
The color filter 49 is divided into red (R), green (G), and blue (B) segments corresponding to each pixel electrode 48. A desired image display can be obtained by sandwiching the active matrix type liquid crystal display device having the above configuration between two polarizing plates 51 and 52 and allowing light to enter.

In the active matrix type liquid crystal display having the above structure, the p-type TFTs 45a and 45b
According to the present invention, at least the lower portion of the glass substrate 40 near the junction between the channel region and the source region and the drain region is etched and removed in a concave shape as viewed from the back surface, so that implantation damage that could not be recovered conventionally, In particular, it is possible to recover implantation damage that has entered the vicinity of the junction between the p-Si channel region and the source and drain regions. As a result, the activation rate of the implanted impurity ions and the resistance to reliability deterioration such as hot carrier deterioration in which carriers are captured are improved, and a driver circuit requiring high reliability can be formed on the same substrate by p-Si TFTs. Can be built in. In addition, the initial characteristics and reliability of the TFT are improved, and a p-Si TFT-LCD with a built-in driver circuit having high performance and high reliability can be realized.

In the first to third embodiments,
Although the TFT used for the liquid crystal display device and the manufacturing method thereof have been described, the present invention is not limited to this, and the TFT of the present application can be used for various thin film integrated circuits such as an image sensor and a semiconductor memory (SRAM). is there.

[0057]

By using the thin film transistor of the present invention and the method of manufacturing the same, an energy beam can be efficiently irradiated from the back surface of the glass substrate, and the implanted impurity ions cannot be activated and cannot be recovered conventionally. Recovery of implantation damage, especially implantation damage that has entered the vicinity of the junction between the channel region and the source and drain regions of the non-single-crystal silicon thin film, etc., could be performed simultaneously,
The activation rate of the implanted impurity ions and the resistance to reliability deterioration such as hot carrier deterioration in which carriers are captured are improved. As a result, the initial characteristics and reliability of the TFT were improved, and a TFT having high performance and high reliability could be obtained. Also, by using an energy beam such as a XeCl excimer laser, the time required for activation can be greatly reduced as compared with the conventional activation method by thermal annealing, and the throughput can be greatly improved. In addition, by manufacturing a liquid crystal display device using the thin film transistor of the present invention and the manufacturing method thereof, a driver circuit requiring high reliability can be incorporated on the same substrate, and high performance and high reliability can be achieved. The liquid crystal display device having the above was realized.

[Brief description of the drawings]

FIG. 1 is a cross-sectional view illustrating a structure of a thin film transistor according to Embodiment 1 of the present invention.

FIG. 2 is a sectional view showing a manufacturing process of the thin film transistor according to Embodiment 1 of the present invention;

FIG. 3 is a cross-sectional view illustrating a manufacturing process of the thin film transistor according to the second embodiment of the present invention.

FIG. 4 is a cross-sectional view illustrating a structure of a thin film transistor according to Embodiment 3 of the present invention.

FIG. 5 is a diagram showing a life test result of the thin film transistor according to Embodiment 1 of the present invention.

FIG. 6 is a cross-sectional view illustrating a structure of a thin film transistor according to Embodiment 1 of the present invention.

FIG. 7 is a cross-sectional view illustrating a structure of a thin film transistor according to Embodiment 1 of the present invention.

FIG. 8 is a cross-sectional view illustrating a structure of a thin film transistor according to Embodiment 1 of the present invention.

FIG. 9 is a sectional view showing the structure of a conventional thin film transistor.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 Glass substrate 11 Base insulating film 12 a-Si: H semiconductor layer 12a Channel region (a-Si) 12b Source region (a-Si) 12c Drain region (a-Si) 12d Junction 13 Semiconductor layer (p- Si) 13a Channel region (p-Si) 13b Source region (p-Si) 13c Drain region (p-Si 13d junction) 14 Gate insulating film (SiO ₂ ) 15 Gate electrode 16 Interlayer insulating film 17 Contact hole 18 Source / drain electrodes 19 resist film

Claims (10)

[Claims] 1. A thin film transistor having at least a light-transmitting glass substrate and a non-single-crystal silicon thin film formed on the light-transmitting glass substrate, wherein at least the non-single-crystal silicon in the light-transmitting glass substrate is provided. A thin film transistor, wherein a thickness of a lower region near a junction between a channel region and a source region and a drain region of a thin film is thinner than other portions. A step of forming a polysilicon thin film on one surface of the light-transmitting glass substrate; a step of forming a gate insulating film on the light-transmitting glass substrate and the polysilicon thin film; Forming a gate electrode thereon, selectively implanting a P-type or N-type impurity into the polysilicon thin film using the gate electrode as a mask, and forming at least the polysilicon thin film in the translucent glass substrate. Reducing the thickness of the lower region near the junction between the channel region and the source region and the drain region to be smaller than the other portion; and irradiating the energy beam from the other surface of the light-transmitting glass substrate to form the implanted P region. Simultaneously activating the N-type or N-type impurities and recovering the damage caused by the implantation of the P-type or N-type impurities. Method for producing a register. 3. The step of reducing the thickness of at least the lower region near the junction between the channel region and the source region and the drain region of the polysilicon thin film in the translucent glass substrate than the other portions, 3. The light-transmissive glass substrate below a region near a junction between a source region and a drain region is etched in a concave shape when viewed from the other surface of the light-transmissive glass substrate. 4. Method for manufacturing thin film transistor. 4. The method according to claim 1, wherein the P-type or N-type impurities are selectively implanted into the polysilicon thin film using the gate electrode as a mask.
3. The method according to claim 2, further comprising the step of removing a gate insulating film on a source region and a drain region of the polysilicon thin film. 5. The thin film transistor according to claim 2, further comprising a step of removing a gate insulating film on a source region and a drain region of the polysilicon thin film before irradiating the energy beam from the other surface of the translucent glass substrate. Production method. 6. A step of forming an amorphous silicon thin film on one surface of a light-transmitting glass substrate, a step of forming a gate insulating film on the light-transmitting glass substrate and the amorphous silicon thin film, and Forming a gate electrode thereon, selectively injecting a P-type or N-type impurity into the amorphous silicon thin film using the gate electrode as a mask, and forming at least the amorphous silicon thin film in the translucent glass substrate. A step of reducing the thickness of the lower region near the junction between the channel region and the source region and the drain region, and irradiating an energy beam from the other surface of the translucent glass substrate with the amorphous silicon thin film Crystallization, activation of the implanted P-type or N-type impurities, and
Simultaneously recovering damage caused by the implantation of N-type or N-type impurities. 7. The step of making a thickness of at least a lower region near a junction between a channel region, a source region, and a drain region of the amorphous silicon thin film in a translucent glass substrate thinner than other portions. 7. The light-transmitting glass substrate below the vicinity of the junction between the channel region and the source region and the drain region is etched in a concave shape when viewed from the other surface of the light-transmitting glass substrate. 3. The method for manufacturing a thin film transistor according to item 1. 8. A step of removing a gate insulating film on a source region and a drain region of the amorphous silicon thin film before selectively implanting P-type or N-type impurities into the amorphous silicon thin film using the gate electrode as a mask. The method for manufacturing a thin film transistor according to claim 6. 9. The thin film transistor according to claim 6, further comprising a step of removing a gate insulating film on a source region and a drain region of the amorphous silicon thin film before irradiating the energy beam from the other surface of the translucent glass substrate. Production method. 10. A light-transmitting glass substrate, and at least a non-single-crystal silicon thin film formed on the light-transmitting glass substrate, wherein at least a channel of the non-single-crystal silicon thin film in the light-transmitting glass substrate is provided. A liquid crystal display device including at least a thin film transistor in which a thickness of a lower region near a junction between a region, a source region, and a drain region is thinner than other portions.