JPH01117067A - Thin-film element - Google Patents

Abstract

PURPOSE:To utilize the feature of Mo without deteriorating the accuracy of a pattern by bringing the thickness of a diffusion layer, into which Mo is diffused, to the thickness or less of a doping layer and laminating an ohmic layer composed of a conductive material except Mo and Al and an Al film. CONSTITUTION:A gate electrode 11, a gate insulating film 12, an active layer 13 and a doping layer 14 are formed in succession. Mo is shaped onto the whole surface on a substrate 10, and thermally treated. Mo is diffused to the doping layer 14 at that time, and the thickness of a diffusion layer 15 formed is made smaller than that of the doping layer 14. An ohmic layer 16 composed of Cr and an Al film 17 are formed successively, and the pattern of a conductive electrode 18 is shaped.

Description

[Detailed Description of the Invention] [Objective of the Invention] (Industrial Application Field) The present invention relates to thin film elements used for driving, switching, or optical sensors of active matrix liquid crystal devices, contact image sensors, etc. Gf thin film transistor
, TFT) and thin film diode (Thin Film
Diode, TFD).

(Prior art) Semiconductor thin films using amorphous silicon (a-3i) as a base material are made of hydrogenated amorphous silicon (a-3i) containing hydrogen (H2).
-3i:H), which can be produced by a relatively low-temperature process of about 200 to 300°C and exhibits good semiconductor properties, has been tried to be applied to many electronic devices. In these electronic devices, for example, TPT,
A functional element using a-3i as a TFD or simply a photoconductive film is incorporated into a device, and one of the important factors that determines the performance of the element is the electrode formation method.

This is because the ohmic characteristics and blocking characteristics required of a functional element depend not only on material control including valence electron control of a-3i, but also on the electrode material and its manufacturing method. In addition, in these electronic devices, the reduction in element size and miniaturization of wiring due to improvements in device performance are
Since the manufacturing process is becoming more complex, it is necessary to design the process and optimize the process conditions, taking selective etching properties into consideration, especially with respect to photolithography technology.

FIG. 6 is a schematic cross-sectional view of an a-3i TFT which is an example of an addressing nonlinear element of an active matrix liquid crystal device. In the figure, on a glass substrate (1), there is a gate electrode (2) made of, for example, chromium (Cr), a gate insulating film (3), and an active layer (made of, for example, a-31).
4) is formed, and a metal electrode (7) composed of a first layer (5) made of molybdenum (MO) and a second layer (6) made of aluminum (A1) is formed, for example, of n-type or It is connected to the active layer (4) via a p-type a-sr doping layer (8), forming source and drain electrodes. Here, MO as the first layer (5) is used for the metal electrode (7) to obtain good ohmic characteristics with the doping layer (8), and furthermore, MO as the second layer (6) is used as the doping layer (8). 8) Prevents it from spreading inside. AI
A-3i is cheap and has low electrical resistance, and is most commonly used as a wiring material in the silicon semiconductor technology field.
It diffuses into the interior at a relatively low temperature of 200 to 300°C, causing defects in ohmic properties or blocking properties. Therefore, in order to prevent heat diffusion during the process, Mo is mixed with AI and a
-3i. Furthermore, MO exhibits good ohmic characteristics with respect to both n-type and p-type a-31, and these ohmic characteristics are not impaired even after undergoing a heat treatment process at about 300°C.

(Problems to be Solved by the Invention) By the way, when forming source and drain electrodes made of A1 using photolithography technology, etching of A1 is carried out using - nitric acid, phosphorous, etc. A mixture of acid, acetic acid and water is used. At this time, MO is also etched away at the same time, but side etching (9) as shown in FIG. 6 occurs because the etching rate of Mo is higher than that of A1. The amount of this side etch (9) depends on the composition and temperature of the liquid mixture, but is about 0.5 to 2 μm. The size of a-srTFT is often used with a channel length, that is, a distance between the source and drain electrodes of about 5 to 10 μm, but the pattern transfer accuracy deteriorates due to the amount of side etching, which directly affects the TPT characteristics. Affect.

In addition, the variation in side etching amount within the substrate is
This results in variations in characteristics. For example, if the channel length at the time of design is 10 μm, the variation in side etching amount is 0.5~
If it is 2 μm, the variation in TPT characteristics is 10 to 30%.
It will also extend to When the channel length is shortened as devices become smaller, this variation becomes even larger.

As described above, although the combination of MO and AI has excellent functional properties as an electrode forming material for a-3i, the conventional etching method had the problem of significantly deteriorating pattern transfer accuracy. .

This invention provides a new structure that takes advantage of the characteristics of MO and does not cause deterioration in pattern accuracy, and further increases the degree of freedom in selecting electrode materials.

[Structure of the Invention] (Means for Solving the Problems) This invention provides a method in which a conductive electrode is in contact with an active layer made of i-type a-3i via a doping layer made of n-type or p-type a-3i. It is a thin film element made of
The thickness of the diffusion layer formed by the diffusion of O does not exceed the thickness of the doping layer, and the conductive electrode connects the A1 film and the ohmic layer made of a conductive material other than MO and A1 from the side closer to the doping layer. It has a layered structure. Furthermore, in terms of the manufacturing process, this invention includes a step of forming a diffusion layer of MO and a-3i, and a step of removing unreacted Mo. , titanium (T i >), tungsten (W), nickel (N+), tantalum (Ta), and combinations therebetween, using readily available conventional selective etching solutions to form conductive electrodes.

(Operation) The principle of this invention will be explained here using a-3i:)l produced by the high-frequency glow discharge decomposition method of s r 1-14.

First, as a sample, add about 0.25% P on a-8i:H.
A doped n-type a-3i was also produced by the high-frequency glow discharge decomposition method, and its thickness was set to 0.05 μm. Subsequently, 0.05 μm of Mo was formed on this n-type a-3i by sputtering, and then heat treatment was performed to remove unreacted MO on the surface using a mixed solution consisting of phosphoric acid, acetic acid, nitric acid, and water. After that, the diffusion depth of MO into the n-type a-3i:l-1 layer was determined from Auger electron spectroscopy (AES>depth profile.
This is an example of the AES spectrum of this sample after 0 minutes of heat treatment; Mo is na-3i:) (approximately 0.0
It can be seen that the particles are diffused to a depth of 2 μm. At this time, the concentration of MO is several percent or more. The solid line corresponding to MO in FIG. 3 is a plot of this diffusion depth against the heat treatment temperature, and it can be seen that even in heat treatment at 300° C., the diffusion depth hardly changes and remains constant. On the sample with the MO diffusion layer formed on its surface, 1.05 μm of Cr was formed by sputtering, heat treated in the same way, and the diffusion was examined by AES.
As shown by the broken line in the figure, cr was detected only near the surface. Metal materials other than Cr, such as Tr, W, Nr
, Ta and their alloys T+w, Ni Cr, MoTa
Furthermore, similar results have been obtained for an indium-tin oxide film known as a transparent conductive film.

Moreover, the sheet resistance of this Mo diffusion layer is in the order of Ω/gate, and considering that the specific resistance of the na-3i:H layer is ~100 Ωcm, the conductivity is good by 3 to 4 orders of magnitude. From this, the effect on the electrical characteristics of the TFT or TFD can be considered as follows.

If the device is designed in advance so that the thickness of the doping layer, for example, the na-3i:H layer, is thicker than the thermal diffusion depth of MO, which is approximately 0.02 μm, the na-3i:H layer will be formed in this conductive MO diffusion layer. Since this conductive MO diffusion layer makes ohmic contact with the electrode metal material, it is essentially not affected by the metal material. Therefore, the same device characteristics as when using an electrode made of MO should be obtained.

As an example of using a metal material other than MO, Cr etc. may actually be used, and in the case of a combination of Cr and AI,
Each can be selectively etched using a known etching solution. However, in the case of Cr, it was confirmed through the following experiment that there is a problem in heat resistance. That is, a-3
i:H doped with P on n-type a-3i:l-1 with 0.
0.05 μm of Cr was grown on the substrate by sputtering, and then 0.05 μm of Cr was grown at 250°C for 30 μm.
The depth of diffusion of Cr into the n-type a-3i:H layer after heat treatment for 30 minutes was investigated by AES, and the result was as shown in FIG. As before, the solid line corresponding to Cr in Figure 3 is the diffusion depth plotted against the heat treatment temperature, and as can be seen from the figure, especially at a high temperature of 300°C, the diffusion depth is It reaches as much as 0.16 μm. Comparing FIG. 2 and FIG. 4, it is clear that the diffusion of Or is different, and when an MO diffusion layer is formed in advance, Cr does not diffuse into this MO diffusion layer. As evidenced by this fact, in the a-31TFT that uses a laminated structure of Or and AI for the electrode, as shown by the broken line connecting the black circles in Figure 5, as the heat treatment temperature increases, the TPT characteristics deteriorate. An increase in threshold voltage Vth was observed.

For this reason, when using Cr, in order to prevent deterioration of TFT characteristics, process restrictions such as thickening the H layer in n-type A-5 or lowering the heat treatment temperature during the process will be added. , causing inconvenience.

(Example) The details of the present invention will be described below with reference to the drawings, taking as an example the case where the thin film element is a TFT.

FIG. 1 is a cross-sectional view showing one embodiment of a thin film element, and this will be explained according to the manufacturing process. In the figure, a gate electrode (11) made of Cr is etched into a predetermined shape on a substrate (10) made of alkali-free glass, and then a silicon nitride film (S i Nx ) gate insulating film (12) consisting of
, a 0.3 μm thick TPT active layer (13) made of I-type a-3i, and a 0.05 μm thick film doped with P, for example.
m n-type a-3i:H doping layer (14)
are formed sequentially. After that, this board (10)
MO is formed on the entire upper surface by sputtering to a thickness of 0.05 μm, and after heat treatment, unreacted MO is removed. At this time, MO is diffused into the doping layer (14), and the thickness of the diffusion layer (15) thus formed is 0.02 μm, which does not exceed the thickness of the doping layer (14). Next, after forming islands of the active layer (13), sputtering is performed to form an ohmic layer (16) made of, for example, Cr.
) and A1 film (17) are sequentially formed, and a pattern of a conductive electrode (18) is formed by photolithography.

At this time, a mixed solution consisting of phosphoric acid, acetic acid, nitric acid, and water was used to etch the A1 film (17), and the ohmic layer (17)
For the etching step 6), a mixed solution consisting of ceric ammonium nitrate and an aqueous hydrogen peroxide solution was used. Both of these mixed solutions are generally known etching solutions and have selectivity to each other. Here, the conductive electrode (18) is TP
A source electrode (19) and a drain electrode (20) of T, which are in contact with the active layer (13) via a doping layer (14) and a diffusion layer (15).

Then, by removing the doping layer (14) and the diffusion layer (15) existing between them, a desired TPT can be obtained.

In this example, M
Ohmic layer (16) made of conductive material other than O and AI
The conductive electrode (1) has a structure in which the A1 film (17) and
8) and the doped layer (14), there is a diffusion layer (15) in which Mo is diffused. TFTs can be miniaturized without special consideration. Actually, the effect of heat treatment on the transfer characteristics of the TPT thus produced was evaluated. Figure 5 shows TPT determined from transfer characteristics.
FIG. 3 is a diagram showing the heat treatment effect on the threshold voltage (Vth) of FIG. In the same figure, the case where Cr is used without using MO is shown by a broken line connecting black circles for comparison, and it is clearly observed that as the heat treatment temperature increases, Vth increases, that is, TFT characteristics deteriorate. On the other hand, in the TPT using Cr electrodes as the MO diffusion layer and the ohmic layer, as shown by the solid line connecting the white circles in FIG. 5, there is not much characteristic deterioration. In addition, the graph shown with this solid line is MO
The same applies to a TPT made using only a diffusion layer.

Although an example of application to an inverted staggered electrode structure TPT has been described here, it is clear that the present invention can also be applied to a planar electrode structure, and can also be used as an electrode formation method not only for TFTs but also for TFDs. is clear.

Furthermore, considering the case where the main charge is holes, it goes without saying that the conductivity type of the doped layer (14) is not limited to n-type but may be p-type.

[Effects of the Invention] This invention makes the original electrical contact with the doped layer Mo
This method takes advantage of the fact that the depth of this diffusion layer does not change depending on the heat treatment temperature.If the thickness of the doping layer is made equal to or greater than the thickness of the MO diffusion layer, the upper metal electrode material It has the advantage of being able to select from a wide variety of metals. This is not just a problem of selective etching;
This increases the degree of design freedom in material selection, and its effects are significant.

[Brief explanation of the drawing]

Fig. 1 is a cross-sectional view showing an embodiment of the present invention, and Figs. 2 and 4 show the relationship between the concentration (%) and depth (μm) of each component contained in the layer with a-3i as the matrix. Figure 3 shows Or
and the diffusion depth of MO into the a-3i matrix layer (μm)
Figure 5 shows the relationship between the temperature and heat treatment temperature (°C), and Figure 5 shows the threshold voltage (V) when using only a Cr electrode or a laminated structure of an MO diffusion layer and a Cr electrode as the conductive electrode of a thin film element.
FIG. 6 is a cross-sectional view showing an example of a conventional thin film element. (13)... Active layer (14)... Doping layer (15)... Diffusion layer (16)... Ohmic layer (17)... ... Aluminum film (18) ... Conductive electrode

Claims (1)

[Claims] A conductive electrode is formed in the active layer made of i-type amorphous silicon.
In a thin film element in which molybdenum is diffused into the doped layer and the thickness of the diffusion layer formed is equal to the thickness of the doped layer. 2. A thin film element characterized in that the conductive electrode has a structure in which an ohmic layer made of a conductive material other than molybdenum and aluminum and an aluminum film are laminated from the side closer to the doped layer.