JPH10219462A - Oxide deposition method - Google Patents

Abstract

(57) [Summary] [PROBLEMS] In the conventional sol-gel method, organic residues or the like remain in a thin film, the abundance ratio of oxygen in an oxide film becomes lower than that of a crystal system to be obtained, or a thick film is formed. Difficulties were issues. SOLUTION: A sol using an organometallic compound as a raw material is applied to a substrate, dried, irradiated with ultraviolet rays, and then heated to a high temperature. The coating, drying, ultraviolet irradiation, and heating steps are repeated as necessary. Alternatively, they are laminated and then heated at a high temperature. The atmosphere during the process is controlled as needed. By annealing these oxides, an oxide crystal thin film is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for forming a thin film on a substrate by using a sol-gel method. The present invention can be used for an electronic device using an oxide, particularly an oxide thin film having high crystallinity.

[0002]

2. Description of the Related Art The sol-gel method has excellent composition controllability,
The feature is that a uniform thin film can be formed in a plane with good reproducibility. As a method for forming an oxide film using a sol-gel method,
A method is known in which a sol is applied, dried and heated at a high temperature, and the above steps are repeated as necessary. In order to prevent the occurrence of cracks, a method of adding a solvent-soluble polymer to the sol or a method of adding a chelating agent to secure the stability of the sol has been adopted.

On the other hand, as a film forming method using ultraviolet irradiation, a method of obtaining an oxide at room temperature or a low temperature of 200 ° C. or less is known.

[0004]

However, the conventional sol-gel method has a problem that organic residues remain in the thin film or that the abundance ratio of oxygen in the oxide film is lower than that of the desired crystal system. . Further, there is a problem in the process that it is difficult to increase the film thickness and the crystallization ratio does not increase due to the presence of organic residues.

On the other hand, the method of forming a film at a low temperature using ultraviolet irradiation has a problem that the densification is insufficient, the residual hydroxyl groups are large, and the application is extremely limited. In particular, it has been difficult to use it for electronic device applications.

Accordingly, the present invention is to solve such a problem, and an object of the present invention is to provide a method for easily forming an oxide thin film having no organic residue with good productivity. . A good oxide thin film which can be made thicker, has high crystallinity, and has no oxygen deficiency can be obtained.

[0007]

According to the method for forming an oxide film of the present invention, a sol made of an organometallic compound is applied to a substrate,
The sol-gel method in which a gel is dried and then heated to a high temperature to form an oxide thin film is characterized by irradiating ultraviolet rays after drying and then heating to a high temperature. Further, it is characterized in that the steps of coating, drying and ultraviolet irradiation are repeated to form a thick film by laminating and heating to a high temperature after laminating, or by repeating the steps from heating to high temperature. Further, it is characterized in that the atmosphere during the process is controlled as required.
An oxide crystal thin film is obtained by annealing these oxides.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS By the above-mentioned irradiation with ultraviolet rays, it is possible to cut chemical bonds of organic compounds present in a target sample before a heating step. Therefore, organic substances can be removed by irradiation with ultraviolet rays, or organic decomposition products can be easily removed by subsequent heating.

Hereinafter, the present invention will be described in detail with reference to examples.

(Example 1) Ten platinum plates (1 cm x 1)
cm) were prepared and used as substrates for the following experiments.

An organosol prepared by dissolving 10 parts by weight of tetraethoxysilane in ethyl alcohol is prepared, applied to each of the ten platinum substrates by spin coating under the same conditions, and then heated in a 200 ° C. oven for 10 minutes. Dried. Five of these 10 samples were irradiated with ultraviolet light of 172 nm (half-width 14 nm) for 2 minutes by a dielectric barrier discharge Xe2 excimer lamp (output 200 W) (samples 1 to 5). On the other hand, the remaining five sheets were not irradiated with ultraviolet rays as before (Comparative Examples 1 to 5).
Samples 1 to 5 and Comparative Examples 1 to 5 were further heated in a 400 ° C. oven for 30 minutes in the atmosphere, and then cooled to room temperature to obtain a silicon dioxide film on a platinum substrate.

In order to examine the presence or absence of residual hydroxyl groups (OH groups) in the obtained silicon dioxide thin film, an infrared absorption spectrum with a wave number of 3600 cm -1 due to OH stretching vibration was measured. Table 1 shows the results. As shown in Table 1, in the case where the ultraviolet irradiation was performed between the drying step and the heating step (Samples 1 to 5), the presence of hydroxyl groups was not recognized, and a good oxide was obtained. However, in the conventional method in which the ultraviolet irradiation was not performed, it was clearly found that the residual hydroxyl groups were present as impurities (Comparative Examples 1 to 5).

From this example, it has been clarified that the irradiation of ultraviolet rays between the drying step and the heating step is effective in obtaining a good oxide film. Here, Xe is used as the ultraviolet light source.
2 An example using an excimer lamp was given, but the irradiation effects of widely used ultraviolet light sources, such as low-pressure or high-pressure mercury lamps and halogen lamps, as well as other excimer lamps such as Ar and Kr as ultraviolet light sources, were investigated. An effect similar to that of the present example was confirmed under optimal conditions. The same effect as in the present example was also confirmed for other oxides by selecting optimum ultraviolet light sources and irradiation conditions.

[0014]

[Table 1]

Example 2 Eight platinum plates (1 cm × 1c)
m) were prepared and used as substrates in the following experiments.

An organosol prepared by dissolving 10 parts by weight of tetraethoxysilane in ethyl alcohol is prepared, applied to each of the eight platinum substrates by spin coating under the same conditions, and then placed in a 200 ° C. oven for 10 minutes. Dried. For seven of these eight samples, ultraviolet irradiation was performed under predetermined conditions using different light sources (samples 1 to 7). On the other hand, the remaining one was not irradiated with ultraviolet rays as before (Comparative Example 1). Samples 1 to 7 and Comparative Example 1 were further heated in a 400 ° C. oven for 30 minutes in the atmosphere, and then cooled to room temperature to obtain a silicon dioxide film on a platinum substrate.

In order to examine the presence or absence of residual hydroxyl groups (OH groups) in the obtained silicon dioxide thin film, an infrared absorption spectrum at a wave number of 3600 cm -1 due to OH stretching vibration was measured. Table 2 shows the results. From Table 2, it was found that, when ultraviolet irradiation was performed between the drying step and the heating step (samples 1 to 7), the presence of hydroxyl groups was not recognized when the irradiation conditions were appropriately selected. However, in order to obtain a good oxide film, it can be seen that there are differences in the ultraviolet irradiation conditions or margins depending on the light source (Table 2).

The ultraviolet light source is an excimer lamp (Xe 2,
Kr2, Ar2), mercury lamp, halogen lamp, metal halide lamp (samples 1, 2,
(Corresponding to 3, 4, 5, and 6) have a relatively large margin of irradiation time optimal for obtaining a good oxide film. In addition, over a wide area, the result that the dependence on the place where the sample is placed was relatively small was also obtained. That is, large area uniform processing can be performed. Above all, the Xe2 excimer lamp has the shortest irradiation time, has a large optimum time margin, and has obtained the best results (Sample 1). The one using an excimer laser (sample 7) has a small margin of irradiation time and is slightly inferior in operability and controllability.
Furthermore, there is a variation in the energy density of the laser on the sample setting surface, which is not suitable for large area uniform processing. In particular, in the case of a laser, the irradiation energy density is higher than that of other light sources, and therefore, attention must be paid to damage to the sample. On the other hand, in the conventional method in which the ultraviolet irradiation was not performed, it was clearly found that the residual hydroxyl group was present as an impurity (Comparative Example 1).

As described above, when ultraviolet irradiation is performed between the drying step and the heating step, any of an excimer lamp, a mercury lamp, a halogen lamp, and a metal halide lamp is suitable as a light source, and among them, a Xe 2 excimer lamp may be used. It turned out to be desirable. In the present embodiment, the case of a silicon dioxide film was shown as a target, but an experiment on an ultraviolet light source was tried with another oxide in the same process, and the same result as in the present embodiment was obtained.

[0020]

[Table 2]

(Embodiment 3) Here, an embodiment of a method for increasing the thickness of an oxide film will be described. In order to obtain a thick film by the sol-gel method, the application of the sol is repeated a plurality of times. The details of the experiment conducted this time are shown below.

Two platinum plates (1 cm × 1 cm) were prepared and used as substrates for the following experiments.

An organosol prepared by dissolving 10 parts by weight of tetraethoxysilane in ethyl alcohol was prepared, applied to each of the two platinum substrates by spin coating under the same conditions, and then placed in a 200 ° C. oven for 10 minutes. Dried. One of these two samples was irradiated with ultraviolet rays having a wavelength of 172 nm (half-width 14 nm) for 2 minutes using a Xe2 excimer lamp (output 200 W). Further, a series of steps of the above-mentioned application of sol, drying and irradiation of ultraviolet rays are performed in 10 steps.
(Sample 1). On the other hand, application and drying of the sol were repeated 10 times for the remaining one without performing ultraviolet irradiation as before (Comparative Example 1). Thereafter, Sample 1 and Comparative Example 1 were heated in a 400 ° C. oven for 30 minutes in the air, and then cooled to room temperature, whereby a silicon dioxide film having a thickness of 1 μm was formed on a platinum substrate. The following evaluation was performed for both. The results are summarized in Table 3.

Observation of the obtained silicon dioxide thin film with an optical microscope showed that Sample 1 had a flat and uniform surface, whereas Comparative Example 1 had a rough surface and many cracks were found throughout. . Next, in order to examine the presence or absence of residual hydroxyl groups (OH groups) in the film, a wave number of 3600 cm due to OH stretching vibration was used.
The infrared absorption spectrum of -1 was measured. Table 3 shows that the sample irradiated with ultraviolet rays between the drying step and the heating step (Sample 1) did not show the presence of hydroxyl groups, and a good oxide was obtained. However, in the conventional method without irradiation with ultraviolet rays, it was clearly found that the residual hydroxyl groups were present as impurities (Comparative Example 1). Further, Sample 1 and Comparative Example 1 were each sandwiched between two plate electrodes, and a potential of 10 V was applied between the electrodes. In Sample 1, there was no current leakage, and the obtained silicon dioxide film had good insulating properties. However, in Comparative Example 1, current leakage was observed (Table 3).

From this example, it was clarified that a good oxide film having a desired film thickness can be obtained by repeatedly performing the drying step and the irradiation of ultraviolet rays and then heating the same. Here, an example in which a Xe2 excimer lamp is used as the ultraviolet light source has been described. Also, the same effect as in the present example was confirmed under the optimum conditions. Further, when the same process was tried for other oxides, the same result as that of the present example was obtained.

[0026]

[Table 3]

(Embodiment 4) An embodiment of a method for increasing the thickness of an oxide film will be described. In order to obtain a thick film by the sol-gel method, the application of the sol is repeated a plurality of times. The details of the experiment conducted this time are shown below.

Two platinum plates (1 cm × 1 cm) were prepared and used as substrates for the following experiments.

An organosol prepared by dissolving 10 parts by weight of tetraethoxysilane in ethyl alcohol was prepared, applied to each of the two platinum substrates by spin coating under the same conditions, and then placed in a 200 ° C. oven for 10 minutes. Dried. One of these two samples was subjected to ultraviolet irradiation at a wavelength of 172 nm (half-width 14 nm) for 2 minutes using a Xe2 excimer lamp (output 200 W), and then 400 ° C.
Heated in oven for 30 minutes in air. Further, a series of steps of the above-described application of the sol, drying, ultraviolet irradiation, and heating was repeated 10 times (sample 1). On the other hand, a series of sol application, drying, and heating steps was repeated 10 times for the remaining one without performing UV irradiation as before (Comparative Example 1). In the above steps, both are 1 μm thick on a platinum substrate.
Was obtained. The following evaluation was performed for both. The results are summarized in Table 4.

Observation of the obtained silicon dioxide thin film with an optical microscope showed that Sample 1 had a flat and uniform surface, whereas Comparative Example 1 had a rough surface and many cracks were found throughout. . Next, in order to examine the presence or absence of residual hydroxyl groups (OH groups) in the film, a wave number of 3600 cm due to OH stretching vibration was used.
The infrared absorption spectrum of -1 was measured. From Table 4, it was found that the sample subjected to ultraviolet irradiation between the drying step and the heating step (Sample 1) did not show the presence of hydroxyl groups, and a good oxide was obtained. However, in the conventional method without irradiation with ultraviolet rays, it was clearly found that the residual hydroxyl groups were present as impurities (Comparative Example 1). Further, Sample 1 and Comparative Example 1 were each sandwiched between two plate electrodes, and a potential of 10 V was applied between the electrodes. In Sample 1, there was no current leakage, and the obtained silicon dioxide film had good insulating properties. In contrast, in Comparative Example 1, current leakage was observed (Table 4).

From this example, it has been clarified that a favorable oxide film can be obtained with a desired film thickness by repeatedly performing the operations of coating, drying, irradiating ultraviolet rays and heating the sol.
Here, an example in which a Xe2 excimer lamp is used as the ultraviolet light source has been described. Also, the same effect as in the present example was confirmed under the optimum conditions. Further, when the same process was tried for other oxides, the same result as that of the present example was obtained.

[0032]

[Table 4]

Example 5 Five nickel plates (1 cm ×
1 cm) were prepared and used as substrates in the following experiments.

Zirconium acetylacetonate was dissolved in acetic acid and ethyl carbitol, and a small amount of triethylene glycol was added. The obtained organosol was applied on each of the five nickel substrates by spin coating under the same conditions, and then dried in a 200 ° C. oven for 10 minutes. Further, ultraviolet irradiation by a high-pressure mercury lamp was performed on each of the five samples under different atmospheres for a predetermined time (samples 1 to 5).
At this time, the results of measuring the presence or absence of ozone generation in the atmosphere are shown in Table 5. Samples 1 to 5 are further placed in a 400 ° C. oven
After heating in the air for 0 minutes and cooling to room temperature, a zirconia film was obtained on the substrate. To check for the presence of residual hydroxyl groups (OH groups) in the obtained zirconia film, OH
An infrared absorption spectrum at a wave number of 3600 cm -1 due to stretching vibration was measured. Table 5 shows the results.

The following has been clarified in the above embodiment.

When the ultraviolet irradiation is performed under reduced pressure or a nitrogen atmosphere, the amount of ozone generated in the atmosphere can be suppressed to be lower than that in the atmosphere (samples 1 to 3). Since ozone is harmful to the human body, it is desirable that the amount of ozone generated is small. Samples 1 to 4 were all under the same conditions except for the atmosphere at the time of ultraviolet irradiation.
Only in the case of (atmospheric UV treatment), residual hydroxyl groups (OH groups) were observed. This is probably because the energy of ultraviolet rays is absorbed or scattered in the atmosphere, so that the efficiency of energy transfer to the target sample is reduced. Therefore, irradiation energy must be increased in order to obtain a good oxide film by irradiating with ultraviolet light in the atmosphere (sample 5), which is not preferable in terms of efficiency and ozone generation.

The above example shows the superiority of performing ultraviolet irradiation under reduced pressure or in a nitrogen atmosphere.

[0038]

[Table 5]

Example 6 Two nickel plates (1 cm ×
1 cm) were prepared and used as substrates in the following experiments.

Titanium isopropoxide was dissolved in acetic acid and ethyl carbitol, and a small amount of triethylene glycol was added. The obtained organosol was applied on the two nickel substrates by spin coating under the same conditions, and then dried in a 200 ° C. oven for 10 minutes. Further, ultraviolet irradiation by a metal halide lamp was performed for a predetermined time in a nitrogen atmosphere. Then, for one of the two samples,
After heating in an oxygen atmosphere 400 ° C. oven for 30 minutes and then cooling to room temperature, titania (TiO 2)
2) A film was obtained (Sample 1). On the other hand, the other sheet was heated in an air oven at 400 ° C. for 30 minutes and then cooled to room temperature to obtain a titania (TiO 2) film on the substrate (Comparative Example 1). The stoichiometric ratio between titanium and oxygen (theoretically 1: 2) in the obtained titania film was determined by plasma emission analysis (I
CP method) and the coulometric method. Table 6 shows the results.

As shown in Table 6, in Sample 1, the stoichiometric ratio of titanium to oxygen was 1: 2, which is the theoretical value, and an ideal oxide film was obtained. On the other hand, in Comparative Example 1, the stoichiometric ratio between titanium and oxygen was 1: 1.6, and it was found that oxygen was lower than the theoretical value.

From the above, it is apparent that performing the heating step in an oxygen atmosphere eliminates oxygen deficiency (deficiency) of the obtained oxide and provides an oxide having an ideal stoichiometric composition. became. In addition, by performing the heating step in an oxygen atmosphere for other oxides, the same effect as in this example was confirmed.

[0043]

[Table 6]

Example 7 Two platinum plates (1 cm × 1c)
m) were prepared and used as substrates in the following experiments.

After dissolving titanium isopropoxide in diethanolamine and butyl cellosolve, 2 moles of lead acetate trihydrate and an equimolar amount of zirconium acetylacenate were added to titanium and dissolved therein. The resulting organosol is applied on the substrate by spin coating,
Dried at 200 ° C. One of these two samples was irradiated with ultraviolet light having a wavelength of 172 nm (half-width 14 nm) for 2 minutes by a dielectric barrier discharge Xe2 excimer lamp (output 200 W) (sample 1). Meanwhile, the remaining one
No UV irradiation was performed on the sheets as before (Comparative Example 1). Sample 1 and Comparative Example 1 were further heated in a 400 ° C. oven for 30 minutes in the atmosphere, then heated to 800 ° C. in an oxygen atmosphere in a lamp heating rapid temperature annealing furnace (RTA) and held for 1 minute. X-ray diffraction confirmed that a perovskite-type lead zirconate titanate (PZT) crystal film was obtained in both samples by the above steps.

Table 7 shows the results of evaluating the crystallinity of the obtained PZT by the half value width of the X-ray diffraction peak from the (100) plane (relative value). From this, it was found that the sample irradiated with ultraviolet rays (Sample 1) had a smaller half-width of the (100) peak and better crystallinity than the conventional method (Comparative Example 1). As a result, it was found that the relative permittivity and the piezoelectric characteristics (d31) also improved as a result of the measurement (Table 7).

In this embodiment, an example of PZT is shown as an object. However, even in the case of another oxide film having an ultraviolet irradiation step, improvement in crystallinity was recognized. Furthermore, as shown in the previous Examples 1 to 6, it is possible to obtain a good oxide crystal film corresponding to various ultraviolet light sources, thickening, or atmosphere control in the ultraviolet irradiation step / heating step. An experiment was performed on the subject, which was confirmed.

[0048]

[Table 7]

[0049]

As described above, according to the method for forming an oxide film of the present invention, a sol made of an organometallic compound is applied to a substrate, gelled and dried, and then heated to a high temperature to form an oxide thin film. The sol-gel method is characterized in that ultraviolet rays are irradiated after drying to break chemical bonds of organic compounds present in the gel thin film. This makes it possible to more easily remove the organic decomposition products during the subsequent high-temperature heating, so that an oxide thin film having no organic residue can be easily formed with good productivity. In addition, since it is possible to respond to thickening and obtain a good oxide thin film with high crystallinity and no oxygen deficiency after annealing, it can be applied to all fields including electronic device applications and micro machine devices. The effect brought is great.

Claims (8)

[Claims] In a sol-gel method, a sol containing an organometallic compound as a raw material is applied to a substrate, gelled and dried, and then heated to a high temperature to form an oxide thin film. A method for forming an oxide film, comprising heating. 2. A sol-gel method in which a sol containing an organometallic compound as a raw material is applied to a substrate, gelled and dried, irradiated with ultraviolet rays, and then heated to a high temperature to form an oxide thin film,
2. The oxide film forming method according to claim 1, wherein the ultraviolet light source is any one of a mercury lamp, a halogen lamp, an excimer lamp, and a metal halide lamp. 3. A sol-gel method in which a sol made of an organometallic compound is applied to a substrate, gelled and dried, irradiated with ultraviolet rays, and then heated to a high temperature to form an oxide thin film.
3. The method according to claim 1, wherein the ultraviolet light source is a Xe2 excimer lamp. 4. A sol-gel method in which a sol containing an organometallic compound as a raw material is applied to a substrate, gelled and dried, irradiated with ultraviolet rays, and then heated to a high temperature to form an oxide thin film.
A method of forming an oxide film, comprising repeating the steps of coating, drying, and irradiating ultraviolet rays, laminating the layers, and heating the layers to a high temperature. 5. A sol-gel method in which a sol containing an organometallic compound as a raw material is applied to a substrate, gelled and dried, irradiated with ultraviolet light, and then heated to a high temperature to form an oxide thin film.
A method for forming an oxide, comprising repeating a process of heating to a high temperature and stacking to obtain a thick film. 6. A sol-gel method in which a sol containing an organometallic compound as a raw material is applied to a substrate, gelled and dried, irradiated with ultraviolet rays, and then heated to a high temperature to form an oxide thin film,
6. The method according to claim 1, wherein the ultraviolet irradiation is performed under reduced pressure or in a nitrogen atmosphere. 7. A sol-gel method in which a sol using an organometallic compound as a raw material is applied to a substrate, gelled and dried, irradiated with ultraviolet rays, and then heated to a high temperature to form an oxide thin film,
7. The method according to claim 1, wherein the high-temperature heating is performed in an oxygen atmosphere. 8. A sol-gel method in which a sol containing an organometallic compound as a raw material is applied to a substrate, gelled and dried, irradiated with ultraviolet rays, and then heated to a high temperature to form an oxide thin film,
8. The method for forming an oxide film according to claim 1, wherein the oxide formed on the substrate is annealed together with the substrate at an appropriate temperature to obtain a crystalline oxide crystal thin film.