JPH0794749A - Method of fabrication of thin film transistor - Google Patents

Abstract

PURPOSE:To form a microcrystal silicon film not including an amorphous layer on a gate insulating layer interface at a speed high enough to form a device by specifying a hydrogen concentration ratio upon formation of a silicon layer in the vicinity of the gate insulating layer interface and after the foregoing formation of the silicon layer. CONSTITUTION:A metal film such as Ta or an MoTa alloy is deposited on a transparent insulating substrate 1, and is patterned by plasma etching using CF4 and O2 for example as reaction gases to form an electrode 2. Then, a gate insulating film 3, a first microcrystal silicon film 4a as an active layer, and a second microcrystal silicon film 4b are deposited in order to form a microcrystal silicon layer 4. The ratio of hydrogen concentration of stock gas upon the first microcrystal silicon film 4a being formed to the hydrogen concentration of the stock gas upon the first microcrystal silicon film 4a being formed is 1. Thus, the microcrystal silicon layer can be formed from the neighourhood of the gate insulating film interface to ensure a TFT corresponding to p-Si in a low temperature process.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of manufacturing a thin film transistor.

[0002]

2. Description of the Related Art Display devices such as electroluminescence, light emitting diode, plasma fluorescent display, and liquid crystal are
The display unit can be made thinner, and there is an increasing demand for use as a display device such as office equipment and computers or a special display device. Among them, a liquid crystal display device (hereinafter abbreviated as TFT-LCD) in which a thin film transistor (hereinafter abbreviated as TFT) is arranged as a switching element on a matrix has a high display quality and consumes less power, and therefore has been developed. It is being actively conducted.

A TFT used as a pixel switching element of a liquid crystal display device can be roughly divided into two types depending on the state of silicon which is its active layer. One is a-Si: HTFT using hydrogenated amorphous silicon (hereinafter abbreviated as a-Si: H) and the other is p-S using polycrystalline silicon (hereinafter abbreviated as p-Si).
iTFT. These two types of TFTs have advantages and disadvantages in their formation process and device characteristics. For example, since a-Si: HTFT uses a plasma CVD method capable of forming a film at a low temperature of about 300 ° C., it is possible to use inexpensive glass as a substrate and it is relatively easy to increase the size. There are problems such as low degree. On the other hand, the p-Si TFT requires a high temperature process of at least about 600 ° C. because it is necessary to polycrystallize the silicon film. Therefore, an inexpensive glass substrate cannot be used, and an expensive substrate such as a heat-resistant quartz substrate is required. Further, it is difficult to increase the size of the quartz substrate, and thus it is also difficult to increase the screen size as compared with the LCD configured by a-Si: HTFT. However, the device characteristics are a-Si: H.
The mobility of 10 times or more is obtained.

Microcrystalline silicon is an active layer material having both of these good characteristics. That is, p-SiT can be formed at a low temperature like an a-Si: H film and has a crystal layer.
Device characteristics similar to those of FT can be obtained.

[0005]

However, the microcrystalline silicon film has the following problems in its manufacturing method. The microcrystalline silicon film is generally formed by diluting SiH ₄ which is a source gas with hydrogen by using a plasma CVD method or the like. However, the hydrogen deposition diminishes the deposition rate. The film formation rate is an important factor that affects the tact time in forming a device, and when the film formation rate is reduced, the tact time cannot be shortened, which is not practical. .

On the other hand, when the hydrogen concentration is decreased and the film formation rate is increased, the microcrystalline silicon film is not formed from the substrate interface, and an amorphous region exists near the interface. As a result, in the inverted stagger type TFT, there is a problem that the characteristics of the microcrystalline silicon film are not fully exerted in the device characteristics because the area becomes the channel area of the TFT.

The present invention has been made to solve such a problem, and it is possible to form a microcrystalline silicon film having no amorphous layer at the interface of a gate insulating layer at a film forming speed sufficient for device formation. It is an object to provide a method for manufacturing a TFT.

[0008]

A method of manufacturing a TFT according to the present invention comprises a step of forming a gate electrode layer on an insulating substrate, and a silicon layer formed on the gate electrode layer via a gate insulating layer by vapor phase epitaxy. In the method of manufacturing a TFT, which comprises a step of forming a silicon oxide layer and a step of forming a source and drain electrode layer that makes a substantially ohmic contact with the silicon layer, the step of forming the silicon layer includes It is characterized in that the ratio of the hydrogen concentration of the source gas at the time of layer formation to the hydrogen concentration of the source gas after the silicon layer is formed near the interface of the gate insulating layer is 1 or more.

Various studies have been made on the formation process of the microcrystalline silicon film. The results will be described below. First, the relationship between the hydrogen concentration of the source gas and the film formation rate will be described with reference to FIG.
FIG. 4 is a diagram showing how the film formation rate depends on the hydrogen concentration of the source gas. In the present invention, the hydrogen concentration of the source gas is defined as a value obtained by dividing the hydrogen gas flow rate by the sum of the source gas flow rate containing silicon atoms and the hydrogen gas flow rate.
From FIG. 4, it can be seen that the film formation rate decreases as the hydrogen concentration of the source gas increases, and particularly when the hydrogen concentration of the source gas is 70% or more. Similarly, the relationship between the hydrogen concentration of the source gas and the film crystallization will be described with reference to FIG. FIG. 5 is a diagram showing how the X-ray peak intensity, which is an index of film crystallization, depends on the hydrogen concentration of the source gas. From FIG. 5, it can be seen that the X-ray peak intensity increases and the film is crystallized at a hydrogen concentration of 70% or more, at which the hydrogen concentration of the source gas increases and the film formation rate further decreases. As described above, the film formed is microcrystallized as the hydrogen concentration in the source gas increases, but at the same time, the film formation rate decreases.

FIG. 6 shows the result of measuring the hydrogen concentration in the produced silicon film. Hydrogen concentration in source gas is about
The hydrogen concentration in the film shows 8 × 10 ²¹ atom / cm ^{3 up} to about 70%, but it can be seen that the amount of hydrogen in the film decreases sharply after that. Therefore, the hydrogen concentration of the source gas that produces the microcrystals is 8 × 10 in the silicon film.
It is preferable to control to a range of less than ²¹ atom / cm ³ .

Next, the microcrystal growth process was investigated in detail in the hydrogen concentration region of the raw material gas for forming the microcrystalline silicon film. When forming a silicon film of about 200 nm on the glass substrate 9, silane 20 sc for hydrogen 80 sccm
Cross-sectional TEM of the silicon film formed when a mixed gas of cm (source gas has a hydrogen concentration of 80%) and a mixed gas of silane 5 sccm with respect to 95 sccm of hydrogen (source gas has a hydrogen concentration of 95%) are used. A schematic diagram of the observation is shown in FIG. Figure 7
7A shows the case where the source gas has a hydrogen concentration of 80%, and FIG. 7B shows the case where the source gas has a hydrogen concentration of 95%. When the source gas had a hydrogen concentration of 80%, an amorphous region 10 appeared in a film having a thickness of about 5 nm immediately above the substrate, and a microcrystalline layer 11 was formed on the amorphous region 10. When an amorphous region appears just above the substrate 9, in an inverted staggered TFT in which the gate electrode and the source / drain electrodes are on the opposite side to the active layer, the amorphous region formed at the initial stage of film formation is the channel region. Therefore, a TFT using microcrystalline silicon as an active layer
The original characteristics cannot be obtained. On the other hand, when the source gas had a hydrogen concentration of 95%, the microcrystalline layer 11 was formed immediately above the substrate.

Further, FIG. 8 shows the result of investigation on the correlation between the hydrogen concentration in the source gas and the mobility (μ) of the TFT. As can be seen, when the hydrogen concentration in the source gas is low, the TFT mobility is constant at about 0.5 cm ² / vs, but when it exceeds about 90%, the TFT mobility increases. Further, within the range where crystallization occurs, when the hydrogen concentration in the source gas where an amorphous region appears in the layer immediately above the substrate is 80%, the high mobility characteristic of microcrystalline silicon is not obtained. Therefore, it is preferable that the hydrogen concentration of the source gas is controlled within a range in which the TFT mobility is 0.5 cm ² / vs or more when the silicon layer is formed near the interface of the gate insulating layer. Note that once the microcrystalline silicon film is formed in the vicinity of the interface, the hydrogen concentration of the source gas can be reduced and the deposition rate can be increased. That is, the hydrogen concentration of the source gas at the time of forming the silicon layer near the interface of the gate insulating layer is controlled within the above range, and the hydrogen concentration of the source gas at the time of forming the silicon layer near the interface of the gate insulating layer and the subsequent silicon layer formation. By setting the ratio to the hydrogen concentration of the source gas at that time to 1 or more, a microcrystalline silicon active layer having a TFT mobility of 0.5 cm ² / vs or more can be obtained. In the present invention, the vicinity of the interface of the gate insulating layer means a layer of 20 nm or less on the silicon layer side from the interface of the gate insulating layer. In addition, if the ratio of the hydrogen concentration of the source gas during the silicon layer formation in the vicinity of the gate insulating layer interface to the hydrogen concentration of the source gas during the subsequent silicon layer formation is 1 or more, the hydrogen concentration of the source gas is It may be changed intentionally or in an inclined manner.

As described above, in the initial stage of forming the microcrystalline silicon film which is the active layer of the TFT, the film forming rate is low, but under the film forming conditions in which sufficient microcrystals can be formed, that is, the hydrogen concentration of the source gas is higher than in the latter half of the film formation. The active layer is formed under the film forming conditions that are likely to become crystalline silicon, and the amorphous layer is not formed. After that, the active layer is formed on the crystalline film under the condition that the hydrogen concentration of the source gas is decreased so that the film forming rate can be increased in the latter stage of the film formation where the crystalline film is easily formed.

In this way, a TFT having no amorphous layer near the interface can be formed with high throughput. As the raw material gas for obtaining the microcrystalline silicon active layer according to the present invention can be used silane gas such as SiH _4, Si ₂ H _6, gas was further halogenating a gas having a SiH bond, for example, SiH ₂ F ₂
But it's okay. As a vapor phase growth method for forming a silicon layer, a plasma CVD method, a photo CVD method, an ECR / CVD method or the like can be used.

The term “substantially ohmic contact” means that no matter what layer or region exists between the silicon layer and the source / drain electrode, as a result, ohmic characteristics are exhibited between the two layers and the electrode.

[0016]

[Operation] By increasing the hydrogen concentration of the source gas,
No amorphous layer is formed in the initial stage of film formation. At the same time, the amount of hydrogen in the microcrystalline silicon film decreases, and the film formation rate also decreases. That is, when the hydrogen concentration of the source gas is increased, the crystallinity in the vicinity of the interface is increased, but a sufficient film formation rate cannot be obtained. Therefore, if a film near the interface is formed under conditions where the film formation rate is low but crystallinity is good to improve the crystallinity near the interface, and the film formation speed is high except in the vicinity of the interface, the total film formation time is reduced. A TFT having high mobility can be formed without increasing the number. In addition, a TFT formed by the method of the present invention
Even if a low temperature process of about 300 ° C. is used, a TFT having the same device characteristics as when using a high temperature process of 600 ° C. or higher can be obtained.

[0017]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The details of the present invention will be described below with reference to the drawings.
FIG. 1 is a sectional view of a thin film transistor according to this embodiment. This will be described according to the manufacturing process. First, a metal film of Ta or MoTa alloy having a thickness of about 200 nm is deposited on the translucent insulating substrate 1 made of a glass substrate by a sputtering method or the like. This metal film is patterned by plasma etching using, for example, CF ₄ and O ₂ as reaction gases to form the gate electrode 2. Next, the gate insulating film 3 is deposited to a thickness of 300 nm, the first microcrystalline silicon film 4a, which is an active layer, is deposited to a thickness of 10 nm, and the second microcrystalline silicon film 4b is deposited to a thickness of 40 nm by plasma CVD or the like. A 50 nm microcrystalline silicon layer 4 is formed. S that is the channel protection layer
An iNx: H film is continuously formed to 200 nm. Then, the laminated film is subjected to a lithography process to form the protective layer 5. Then, an ohmic contact layer 6 is formed on the entire surface of the microcrystalline silicon layer including the protective layer 5 to have a thickness of n ^{+ of} 50 nm.
An a-Si: H film is formed. Next, a metal film to be an electrode is formed and patterned to form source / drain electrodes 7 and 8. Finally, n ⁺ a-S remaining on the protective film
The i: H film is processed by, for example, reactive ion etching (RIE).
The TFT is completed by removing it by etching or the like.

A mixed gas of 95 sccm of hydrogen and 5 sccm of silane was used as a raw material gas at the time of forming the above-mentioned first microcrystalline silicon film 4a. About 10 in that condition
nm was formed. Next, a mixed gas of the second microcrystalline silicon film 4b was switched to a mixed gas of 80 sccm of hydrogen and 20 sccm of silane to form a film having a thickness of about 40 nm. Both conditions are conditions under which microcrystals can be obtained in experiments with a single-layer film, but there is a difference in terms of crystallinity from the vicinity of the substrate. That is, under the film forming conditions of hydrogen: silane = 95: 5, although the crystallites are formed from the substrate interface, the film forming rate is low. On the other hand, under the film forming conditions of hydrogen: silane = 80: 20, the film forming rate is high, but an amorphous region exists near the interface with the substrate.

FIG. 2 shows an evaluation of the hydrogen concentration distribution in the film thickness direction of the microcrystalline silicon layer which is the active layer of the TFT thus formed. In the figure, the horizontal axis represents the film thickness of microcrystalline silicon, and the vertical axis represents the amount of hydrogen in the film. It can be seen that the amount of hydrogen is small near the gate insulating interface.

As in this example, in the initial stage of film formation, the condition that the crystallinity from the interface is good is used, and the film formation rate is high at the stage when microcrystals are formed to some extent. By switching to the microcrystalline film forming conditions to be formed, it is possible to obtain the active layer of microcrystalline silicon in which the microcrystalline film grows from the interface of the gate insulating film without increasing the total film forming time. The gate insulating film is a SiNx film or a SiOx film formed by a CVD method, or a laminated film or a mixed film SiNx thereof.
Alumina obtained by anodizing not only the Oy film but also the gate electrode such as aluminum can be used. Further, an oxide film is formed on it.

The present invention can be applied not only to the formation of microcrystalline silicon near the interface with the gate electrode but also to the interface with the protective film. In other words, the bonding portion between the insulating film and the microcrystalline silicon film can be variously modified and used.

Further, the hydrogen concentration distribution may be changed continuously instead of being changed stepwise as in the above embodiment. That is, as shown in FIG. 3, the hydrogen concentration distribution in the microcrystalline silicon may be set so that the concentration becomes lower from the surface toward the gate insulating film.

[0023]

According to the TFT manufacturing method of the present invention, the ratio of the hydrogen concentration of the source gas at the time of forming the silicon layer in the vicinity of the gate insulating layer interface to the hydrogen concentration of the source gas at the time of forming the silicon layer thereafter is 1 or more. Therefore, a microcrystalline silicon layer can be formed near the interface of the gate insulating film, and a p-Si-like TFT can be obtained by a low temperature process. In addition, since the film forming conditions are switched, the total film forming time does not increase significantly, thus improving the throughput.

[Brief description of drawings]

FIG. 1 is a cross-sectional view of a thin film transistor according to this embodiment.

FIG. 2 is a diagram showing a hydrogen concentration distribution in a film thickness direction of a microcrystalline silicon layer.

FIG. 3 is a diagram when the hydrogen concentration distribution is continuously changed.

FIG. 4 is a diagram showing a relationship between a hydrogen concentration of a source gas and a film forming rate.

FIG. 5 is a diagram showing a relationship between hydrogen concentration of a source gas and film crystallization.

FIG. 6 is a diagram showing measurement of hydrogen concentration in a silicon film.

FIG. 7 is a schematic diagram of cross-sectional TEM observation of a silicon film.

FIG. 8: Hydrogen concentration in source gas and TFT mobility (μ)
It is a figure which shows the correlation of.

[Explanation of symbols]

1 ... Transparent insulating substrate, 2 ... Gate electrode, 3 ...
... gate insulating film, 4 ... microcrystalline silicon layer, 5 ... protective layer, 6 ... ohmic contact layer, 7 ... source electrode, 8 ... drain electrode, 9 ... substrate, 1
0 ... Amorphous region, 11 ... Microcrystalline layer.

Claims (1)

[Claims] 1. A step of forming a gate electrode layer on an insulating substrate, a step of forming a silicon layer on the gate electrode layer via a gate insulating layer by a vapor phase epitaxy method, and substantially forming the silicon layer on the silicon layer. In the method of manufacturing a thin film transistor, which includes a step of forming a source and drain electrode layer that makes ohmic contact, the step of forming the silicon layer includes the hydrogen concentration of a source gas at the time of forming the silicon layer in the vicinity of the interface of the gate insulating layer, A method of manufacturing a thin film transistor, characterized in that the ratio to the hydrogen concentration of the source gas after the silicon layer is formed near the interface of the gate insulating layer is 1 or more.