JP2023004877A - Susceptor and manufacturing method thereof - Google Patents

Abstract

To provide a susceptor with reduced contamination, which can improve the uniformity of the film thickness of a silicon carbide film formed on a base material, and suppress variations in thermal conductivity, and is made of a carbon composite material in which the surface of the substrate made of a carbon material is covered with a thin film of silicon carbide (SiC), and a manufacturing method thereof.SOLUTION: In a susceptor 1 having a substrate 2 made of a carbon material and having one main surface on which a silicon wafer is placed and another main surface facing the one main surface, the entire surface of the substrate is covered with a thin film 3 made of silicon carbide, the variation in emissivity in the one main surface is within 3%, and the ratio of the average emissivity of the other main surface facing the one main surface is 1:1 to 1:0.8.SELECTED DRAWING: Figure 2

Description

The present invention relates to a susceptor and its manufacturing method, and more particularly to a susceptor for holding a wafer in an epitaxial film deposition apparatus and its manufacturing method.

In an epitaxial deposition apparatus, which is one of semiconductor manufacturing apparatuses, a carbon composite material in which a carbon material (called a carbon substrate) is covered with silicon carbide (SiC) is used as a susceptor, which is a member that holds a silicon wafer. ing. The susceptor has a pancake type, a barrel type, a single substrate type, etc., depending on its shape, and a plurality of types are used depending on the apparatus and processing method.
When manufacturing the susceptor, regardless of the type, the susceptor is placed in a predetermined coating furnace in the form of a carbon base material, and a film of silicon carbide (SiC) is formed on the surface of the carbon base material by a CVD method or the like. A susceptor of carbon composite material is obtained.

By the way, when a silicon carbide (SiC) thin film is formed on the surface of a carbon base material by the CVD method, the silicon carbide film does not adhere to the contact portion between the jig supporting the carbon base material and the carbon base material. Become.
In order to address such a problem, Patent Document 1 discloses that after the first film formation process, the carbon base material is once removed from the furnace, and the second and subsequent film formation processes are performed by changing the position at which the carbon base material and the jig contact each other. processing is described. By doing so, a carbon composite material in which the entire surface is covered with silicon carbide (SiC) can be obtained.

As described above, film formation processing is performed multiple times while shifting the contact position. will be exposed to Therefore, there is a problem that the surface of the silicon carbide film may be contaminated. If it is contaminated, a new silicon carbide film is laminated on the contaminated layer, and if this carbon composite material is used as a susceptor, it will cause contamination of the silicon wafer in the epitaxial process.

Therefore, in the invention described in Patent Document 1, after first forming a film of silicon carbide, the carbon composite material is taken out once to change the support position in order to eliminate the support trace, and the surface of the carbon composite material is changed. A purification treatment (halogen gas blowing) is performed to reduce surface contamination, and silicon carbide film formation is performed again in the furnace.

JP 2008-174841 A

However, in the method disclosed in Patent Document 1, the possibility of contamination cannot be eliminated because the carbon base material is once removed from the furnace. In addition, since it is necessary to perform the film formation process a plurality of times with an interval of time, there is a problem that the number of man-hours and costs increase, which is not desirable.

Furthermore, the portion of silicon carbide that is in contact with the jig that holds the carbon base material has a thinner film thickness than other portions, and in the case of two-time film formation, the film thickness is about half. Therefore, in the epitaxial film formation process, there is a possibility that carbon, which is the base material, may be exposed due to consumption of the silicon carbide film.
In addition, non-uniformity in the thickness of the silicon carbide film is also a factor in producing film thickness variations in the epitaxial film formation process on the silicon wafer. When the film thickness of silicon carbide varies greatly, there is a problem that it is difficult to obtain a uniform epitaxial film due to the difference in thermal conductivity.

Further, if the silicon carbide film covering the carbon substrate is non-uniform, the emissivity varies in the temperature range where the susceptor is used. If the emissivity varies greatly, temperature unevenness occurs in the susceptor, resulting in variations in the wafer temperature and, in turn, variations in the thickness of the epitaxial film.

The present invention has been made under the above circumstances, and provides a susceptor made of a carbon composite material in which the surface of a base material made of a carbon material is covered with a thin film of silicon carbide (SiC), wherein carbonization formed on the base material It is an object of the present invention to provide a susceptor with suppressed contamination, which can improve the uniformity of the film thickness of a silicon film and suppress variations in thermal conductivity, and a method of manufacturing the same.

A susceptor according to the present invention, which has been made to solve the above problems, has a substrate made of a carbon material, and has one main surface on which a silicon wafer is placed and another main surface facing the one main surface. wherein the entire surface of the base material is covered with a thin film made of silicon carbide, the variation in emissivity between the one main surface is within 3%, and the one main surface and the one main surface It is characterized in that the ratio of the average emissivity of the other main surface facing to is 1:1 to 1:0.8.
In addition, it is desirable that the variation in the emissivity of the other main surface facing the one main surface is within 3%.
Further, the ratio of the thickness of the thin film formed on the other main surface to the thickness of the thin film formed on the one main surface is 0.7 or more and 1.2 or less, and The film thickness difference between the central portion and the outer edge portion is 40% or less of the average value of the film thickness of the thin film formed on the one main surface, and the maximum film thickness and the minimum film thickness of the outer edge portion of the one main surface. It is desirable that the difference between the thickness and the thickness is 40% or less of the average value of the thickness of the thin film formed on the one main surface.
Moreover, it is desirable that the thickness of the thin film made of silicon carbide formed on the entire surface of the base material is at least 60 μm.

According to such a configuration, the uniformity of the thin film formed on the surface of the substrate is improved, and the uniformity of heat conduction on the one main surface is improved. As a result, a uniform epitaxial film can be obtained in the epitaxial film forming process on the silicon wafer using the susceptor.

A method of manufacturing a susceptor according to the present invention, which has been made to solve the above-mentioned problems, is a method of manufacturing the susceptor, in which a substrate made of a carbon material is moved in a chamber in a supporting position with respect to the substrate. It is characterized in that a thin film of silicon carbide is formed on the entire surface of the base material by supporting the base material while supporting the base material and supplying the raw material gas so that the supply direction is parallel to the main surface of the base material.
According to such a method, the susceptor with suppressed contamination can be obtained.

According to the present invention, in a susceptor made of a carbon composite material in which the surface of a substrate made of a carbon material is covered with a thin film of silicon carbide (SiC), the thickness uniformity of the silicon carbide film formed on the substrate is measured. It is possible to provide a susceptor with reduced contamination and a method of manufacturing the same, which can increase the height and suppress variations in thermal conductivity.

FIG. 1 is a cross-sectional view of a susceptor according to the invention. FIG. 2 is a cross-sectional view enlarging a part of the susceptor of FIG. FIG. 3 is a cross-sectional view schematically showing a CVD apparatus used when manufacturing the susceptor of FIG. 4 is a plan view of the CVD apparatus of FIG. 3. FIG.

An embodiment of a susceptor and a method of manufacturing the same according to the present invention will be described below with reference to FIGS. 1 to 4. FIG. The drawings are schematic or conceptual, and the relationship between the thickness and width of each part, the size ratio between parts, etc. are not shown accurately.

As shown in FIG. 1, the susceptor 1 has a disk-shaped carbon substrate 2 made of a carbon material. The entire surface of the carbon base material 2 is covered with a thin film 3 of silicon carbide having a predetermined thickness (for example, 60 μm or more).
That is, the thin film 3 consists of a thin film 3F made of silicon carbide covering one principal surface F1 of the susceptor 1, which is the wafer mounting surface, and another principal surface F2, which is the back surface facing the one principal surface F1. and a thin film 3S made of silicon carbide covering the outer peripheral surface of the carbon substrate 2. As shown in FIG.

Further, the susceptor 1 is a so-called single wafer type susceptor in which one main surface F1 is formed with one recessed counterbore 4 for mounting a semiconductor substrate thereon.
The counterbore portion 4 is formed in a circular shape in a plan view, and has a cylindrical concave portion 4a formed in the center. Further, the susceptor 1 has circular symmetry with a rotation axis L passing through the center O thereof. At this time, assuming that the depth of the deepest portion (central portion O) of the counterbore portion 4 is To, the average depth Td is To/2.

A ratio (T/Td) between the thickness T of the susceptor 1 and the average depth Td is preferably 6≦T/Td≦30. Preferably, the ratio (T/To) of the thickness T of the susceptor 1 to the depth To is 3≦T/To≦13.
As described above, the counterbore portion 4 is formed such that the ratio (T/Td) of the thickness T of the susceptor 1 to the average depth Td is 6≦T/Td≦30. effect can be obtained.

Here, when the ratio (T/Td) of the thickness T of the susceptor 1 to the average depth Td is less than 6, the counterbore is too deep with respect to the thickness of the susceptor 1 and the outer periphery of the wafer It is not preferable because the film formation may become defective. Further, when the ratio (T/Td) of the thickness T of the susceptor 1 to the average depth Td exceeds 30, the susceptor becomes thick and the influence of the rigidity of the carbon base material 2 cannot be ignored. This is not preferable because it makes it difficult to control the amount of warpage in a thin film.

As described above, the carbon substrate 2 is made of a carbon material that can be used as a susceptor for semiconductors, and the thin film 3 is made of silicon carbide. The thin film 3 is formed on the entire surface of the carbon base material 2 to prevent dust generation from the carbon base material 2 and outward diffusion of impurities, or to protect the entire surface of the carbon base material 2 . It has a role of suppressing the warpage of.

Here, the ratio of the average thickness t2 of the thin film 3B formed on the main surface F2 of the susceptor 1 shown in FIG. 2 to the average thickness t1 of the thin film 3F formed on the main surface F1 of the susceptor 1 is 0.7. It is preferably formed between ∼1.2.
If the ratio is less than 0.7, it is not preferable because a difference in thermal conductivity may occur in the epitaxial film formation process using the susceptor, making it difficult to obtain a uniform epitaxial film.
On the other hand, if the ratio is more than 1.2, in addition to the difference in thermal conductivity caused by the film thickness variation of the thin film 3, the susceptor is likely to warp and the epitaxial film becomes non-uniform, which is not preferable.

Further, on the main surface F1 of the susceptor 1, the film thickness difference d1 between the central portion O and the outer edge portion F1a is formed to be 40% or less of the average thickness t1 of the thin film 3F formed on the main surface F1. preferably.
Further, on the main surface F1 of the susceptor 1, the film thickness difference d2 between the maximum film thickness and the minimum film thickness of the outer edge portion F1a is 40% or less of the average film thickness t1 of the thin film 3F formed on the main surface F1. It is preferably formed like this.
If the film thickness difference d1 or d2 is 40% or less of the average film thickness t1, the uniformity of heat conduction on the main surface F1 is improved, and a uniform epitaxial film is obtained in the epitaxial film forming process using the susceptor. be able to.
On the other hand, if the film thickness difference is larger than 40% of the average film thickness t1, spots are likely to occur, and heat conduction on the main surface F1 becomes nonuniform, which may make it impossible to obtain a uniform epitaxial film. .

In the susceptor 1, the variation in the emissivity of the wafer mounting surface (main surface F1) is within 3% in the temperature range (900 to 1300° C.) in the epitaxial film formation process, and It is formed so that the average emissivity ratio of (another main surface F2) is 1:1 to 1:0.8.
Moreover, it is desirable that the back surface (other main surface F2) of the wafer mounting surface is also formed so that the variation in emissivity within the same surface is within 3%.
By setting the emissivity of the susceptor 1 to such a value, variations in the thermal conductivity of the susceptor are suppressed and temperature unevenness does not occur. Therefore, the temperature of the mounted wafer is made uniform, and variations in the thickness of the epitaxial film are prevented. be able to.

The susceptor 1 as described above can be manufactured by using, for example, a CVD apparatus 5 as shown in FIG.
The CVD apparatus 5 shown in FIG. 3 includes a chamber 10 forming a processing space, a gas inlet 11 provided on the side of the chamber 10 for supplying a carrier gas (hydrogen gas) into the chamber 10, and a gas inlet 11 at the inlet 11. and a gas outlet 12 provided on the side of the chamber 10 on the opposite side.

In addition, support portions 20 for supporting the lower surface side of the carbon base material 2 of the susceptor 1 in the chamber 10, and a plurality of support portions 20 arranged around the carbon base material 2 to support the side peripheral surface (periphery) of the carbon base material 2. and a columnar guard member 13 that is slidably supported.
The support portion 20 has a plurality of support legs 20a to 20d arranged so that a roller 22, which is rotatable at a constant speed by a motor 21, rotates along the circumferential direction of the carbon base material 2. As shown in FIG. The rollers 22 of the support legs 20a to 20d are in contact with the periphery of the back surface of the carbon base material 2, and each roller 22 rotates in one direction, thereby supporting the carbon base material 2 around the center O. configured to rotate. The rotation operations (rotation start, stop, rotation direction, rotation speed) of the rollers 22 of the support legs 20a to 20d are controlled by a control unit (not shown) so as to be synchronized.
Further, as shown in FIG. 3, heater units 15 are provided above and below the chamber 10 so as to raise the temperature inside the furnace to a predetermined temperature.

When the susceptor 1 is manufactured using this CVD apparatus 5, the carbon substrate 2 made of a carbon material and having a circular counterbore formed in advance is placed on the support legs 20a to 20d in the chamber 10. As shown in FIG.
Next, a controller (not shown) starts rotating the rollers 22 of the support legs 20a to 20d at a predetermined rotational speed. As a result, the carbon base material 2 rotates around the center O at a predetermined speed (for example, 0.1 rpm).

Further, the heater unit 15 is driven to heat the inside of the chamber 10 to, for example, 500° C., and the inside of the chamber 10 is sucked from the gas outlet 12 to create a vacuum state.
Next, a carrier gas (H ₂ ) is introduced into the chamber 10 through the gas inlet 11 at a predetermined flow rate. After that, the inside of the chamber 10 is heated to, for example, 1300° C., and the raw material gases (SiCl ₄ , C ₃ H ₈ ) are introduced together with the carrier gas for a predetermined time. The raw material gas concentration in the chamber 10 at the start of introduction is, for example, 15% to 20%.

Here, the raw material gas flows along the upper and lower surfaces of the carbon substrate 2 by the carrier gas and is discharged from the gas outlet 12 .
In addition, since the carbon base material 2 is rotated around the center O by a plurality of rotationally driven rollers 22 provided to support the lower surface side peripheral edge part of the carbon base material 2 , the lower surface side peripheral edge part of the carbon base material 2 is rotated. The support position in the 1 is not the same (it is not fixed and changes), and the uniformity of the film thickness of the formed film is improved.

The raw material gas is supplied into the chamber 10 for a predetermined time (eg, 14 hours) so that the formed film has a predetermined thickness (eg, 60 μm or more).
Then, in the final stage of the raw material gas supply process (for example, 5 to 60 minutes before the end), the raw material gas is diluted stepwise to 1/2 to 1/4 of the normal concentration.
As a result, the raw material gas is made leaner than usual, the film formation rate is lower than usual, and the crystal grains are formed in a uniform size. As a result, the amount of film formed in the plane tends to be uniform, and the emissivity in the same plane can be made more constant. Specifically, the variation in emissivity on the wafer mounting surface (and its back surface) is within 3%, and the ratio of the average emissivity between the wafer mounting surface and its back surface is 1:1 to 1:0.8. adjusted to
When a preset supply time of the raw material gas elapses, the supply of the raw material gas is stopped, and the rotation of the roller 22 is stopped after a predetermined time (for example, one hour) has passed.

By these treatments, a thin film 3 made of silicon carbide is formed on the carbon substrate 2, and the susceptor 1 of the present invention is manufactured. This thin film 3 is formed with high uniformity in film thickness because the position at which the carbon base material 2 is supported in the chamber 10 is constantly changing while the carbon base material 2 is exposed to the raw material gas. That is, in the obtained susceptor 1, the ratio of the average thickness t2 of the thin film 3B formed on the main surface F2 to the average thickness t1 of the thin film 3F formed on the main surface F1 is 0.7. ∼1.2. In addition, on the main surface F1 of the susceptor 1, a film thickness difference d1 between the central portion O and the outer edge portion F1a and a film thickness difference d2 between the maximum film thickness and the minimum film thickness of the outer edge portion F1a are formed on the main surface F1. It is formed so as to be 40% or less of the average film thickness t1 of the thin film 3F.

As described above, according to the embodiment of the present invention, in the susceptor 1, the thin film 3 formed on the carbon substrate 2 has a highly uniform film thickness. The variation is within 3%, and the average emissivity ratio of the wafer mounting surface and the back surface thereof is set to 1:1 to 1:0.8.
Further, the ratio of the average thickness t2 of the thin film 3B formed on the main surface F2 of the susceptor 1 to the average thickness t1 of the thin film 3F formed on the main surface F1 of the susceptor 1 is 0.7 to 1.2. In the main surface F1 of the susceptor 1, the film thickness difference d1 between the central portion O and the outer edge portion F1a or the film thickness difference d2 between the maximum film thickness and the minimum film thickness of the outer edge portion F1a is formed on the main surface F1. It is formed so as to be 40% or less of the average film thickness t1 of the formed thin film 3F.
As a result, the uniformity of the thin film 3 formed on the surface of the carbon substrate 2 is improved, temperature unevenness does not occur in the susceptor 1, and the uniformity of heat conduction on the main surface F1 is improved.
As a result, a uniform epitaxial film can be obtained in the epitaxial film forming process on the silicon wafer using the susceptor.
Further, when a thin film made of silicon carbide is formed on the carbon base material 2 made of a carbon material by CVD, a uniform thin film can be formed on the entire carbon base material 2 by not fixing the support position with respect to the carbon base material 2. can be done. Further, it is not necessary to take out the carbon substrate 2 from the chamber in the middle of thin film formation as in the conventional method, and a single-layer thin film can be formed in which contamination is suppressed.

In the above embodiment, as a method of suppressing variations in emissivity within the same plane of the wafer mounting surface (and back surface), the source gas is diluted at the final stage of the source gas supply process. , but not limited to that example.
Further, in the above-described embodiment, the susceptor having the counterbore portion was described as an example, but the present invention is not limited to this form, and can be applied to a susceptor having no counterbore portion. be able to.
Further, when the counterbore portion is provided, the present invention can be applied not only to the cylindrical counterbore portion shown in the figure, but also to a susceptor having a concavely curved counterbore portion, for example.

The susceptor and the manufacturing method thereof according to the present invention will be further described based on examples.
[Experiment 1]
In Experiment 1, isotropic graphite was used as the material of the base material of the susceptor, and a plurality of carbon base materials with counterbore portions were prepared. Using the CVD apparatus shown in FIG. 3, a silicon carbide film was formed on the substrate surface under a plurality of film thickness forming conditions.
In the CVD apparatus, a carbon substrate was placed in a chamber, and after evacuation, the temperature of the chamber was raised to 500° C., and carrier gas (H ₂ ) was introduced into the chamber. Next, the inside of the chamber is heated to 1300° C., the substrate is rotated at a rotation speed of 0.1 rpm so as not to fix the support position of the carbon substrate, and the raw material gas (SiCl _{4 , C3H8} ) was fed. After a predetermined time (14 hours), the supply of the raw material gas was stopped, and after 1 hour the rotation of the carbon substrate was stopped to form a silicon carbide thin film with a thickness of 70 μm on the surface of the substrate.

Here, in the final stage (0.2 hours before the end) of the raw material gas supply process (14 hours), the concentration of the raw material gas is diluted to dilute the wafer mounting surface and the back surface of the susceptor formed. to suppress variations in emissivity.
The emissivity variation was set as a condition for Examples 1 to 4 and Comparative Examples 1 to 3, and the emissivity variation was adjusted by the source gas concentration. The emissivity of the wafer mounting surface and its back surface was measured at 120° intervals on concentric circles located at the center of the wafer mounting surface of the carbon substrate and on the outer peripheral side from the center at a radius of 50% of the mounting surface. FTIR (Fourier transform infrared spectroscopy) manufactured by Thermo Fisher Co., Ltd. and a method using an integrating sphere were used at a total of four points in three locations. The average emissivity was calculated at the above four points, and the emissivity variation was calculated at the above four points by (maximum value−minimum value)÷average value. The back surface of the wafer mounting surface was also measured and calculated at the same measurement position as the wafer mounting surface.

In Example 1, the emissivity variation of the wafer mounting surface of the carbon substrate is 1%, the emissivity variation of the back surface of the wafer mounting surface is 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface is 1. :1.
In Example 2, the emissivity variation of the wafer mounting surface of the carbon substrate is 2%, the emissivity variation of the back surface of the wafer mounting surface is 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface is 1. : 0.9.
In Example 3, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 4, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 5, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 4%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.

In Comparative Example 1, the emissivity variation of the wafer mounting surface of the carbon substrate was 4%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Comparative Example 2, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.7.

Using the susceptors manufactured in Examples 1 to 5 and Comparative Examples 1 and 2, a process for forming an epitaxial film on a silicon wafer was performed.
The results of Experiment 1 are shown in Table 1. Each condition shown in Table 1 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

From the results of Experiment 1 above, when a silicon carbide thin film having a thickness of 70 μm is formed on the substrate surface, the variation in the emissivity of the wafer mounting surface (surface) should be within 3%, and the wafer mounting surface and It was confirmed that the uniformity of the epitaxial film formed on the silicon wafer was improved by setting the average emissivity ratio of the back surface to 1:1 to 1:0.8.

[Experiment 2]
In Experiment 2, the thickness of the silicon carbide thin film on the substrate surface was changed to 30 μm, and other conditions were the same as those in Experiment 1 for evaluation.

In Example 6, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. :1.
In Example 7, the emissivity variation of the wafer mounting surface of the carbon substrate is 2%, the emissivity variation of the back surface of the wafer mounting surface is 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface is 1. : 0.9.
In Example 8, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 9, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 10, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 4%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.

In Comparative Example 3, the emissivity variation of the wafer mounting surface of the carbon substrate was 4%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Comparative Example 4, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.7.

Using the susceptors manufactured in Examples 6 to 10 and Comparative Examples 3 and 4, processing for forming epitaxial films on silicon wafers was performed.
The results of Experiment 2 are shown in Table 2. Each condition shown in Table 2 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

According to the results of Experiment 2 shown in Table 2, when a silicon carbide thin film having a thickness of 30 μm is formed on the substrate surface, the variation in the emissivity of the wafer mounting surface (surface) should be within 3%, and the wafer mounting surface should be within 3%. It was confirmed that the uniformity of the epitaxial film formed on the silicon wafer was improved by setting the ratio of the average emissivity between the placement surface and the back surface to 1:1 to 1:0.8.

[Experiment 3]
In Experiment 3, the thickness of the silicon carbide thin film on the substrate surface was changed to 60 μm, and other conditions were the same as those in Experiment 1 for evaluation.

In Example 11, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. :1.
In Example 12, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Example 13, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 14, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 15, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 4%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.

In Comparative Example 5, the emissivity variation of the wafer mounting surface of the carbon substrate was 4%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Comparative Example 6, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.7.

Using the susceptors manufactured in Examples 11 to 15 and Comparative Examples 5 and 6, a process for forming epitaxial films on silicon wafers was performed.
The results of Experiment 3 are shown in Table 3. Each condition shown in Table 3 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

According to the results of Experiment 3 shown in Table 3, when a silicon carbide thin film having a thickness of 60 μm is formed on the substrate surface, the variation in the emissivity of the wafer mounting surface (surface) should be within 3%, and the wafer mounting surface should be within 3%. It was confirmed that the uniformity of the epitaxial film formed on the silicon wafer was improved by setting the ratio of the average emissivity between the placement surface and the back surface to 1:1 to 1:0.8.

[Experiment 4]
In Experiment 4, the thickness of the silicon carbide thin film on the substrate surface was changed to 140 μm, and other conditions were the same as those in Experiment 1 for evaluation.

In Example 16, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. :1.
In Example 17, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Example 18, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 19, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 20, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 4%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.

In Comparative Example 7, the emissivity variation of the wafer mounting surface of the carbon substrate was 4%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Comparative Example 8, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.7.

Using the susceptors manufactured in Examples 16 to 20 and Comparative Examples 7 and 8, processing for forming epitaxial films on silicon wafers was performed.
The results of Experiment 4 are shown in Table 4. Each condition shown in Table 4 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

According to the results of Experiment 4 shown in Table 4, when a silicon carbide thin film having a thickness of 140 μm is formed on the substrate surface, the variation in the emissivity of the wafer mounting surface (surface) should be within 3%, and the wafer mounting surface should be within 3%. It was confirmed that the uniformity of the epitaxial film formed on the silicon wafer was improved by setting the ratio of the average emissivity between the placement surface and the back surface to 1:1 to 1:0.8.

[Experiment 5]
In Experiment 5, the thickness of the silicon carbide thin film on the substrate surface was changed to 200 μm, and the other conditions were the same as those in Experiment 1 for evaluation.

In Example 21, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. :1.
In Example 22, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Example 23, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 1%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 24, the emissivity variation of the wafer mounting surface of the carbon substrate was 1%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.
In Example 25, the emissivity variation of the wafer mounting surface of the carbon substrate was 2%, the emissivity variation of the back surface of the wafer mounting surface was 4%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.8.

In Comparative Example 9, the emissivity variation of the wafer mounting surface of the carbon substrate was 4%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.9.
In Comparative Example 10, the emissivity variation of the wafer mounting surface of the carbon substrate was 3%, the emissivity variation of the back surface of the wafer mounting surface was 3%, and the ratio of the average emissivity between the wafer mounting surface and the back surface was 1. : 0.7.

Using the susceptors manufactured in Examples 21 to 25 and Comparative Examples 9 and 10, processing for forming an epitaxial film on a silicon wafer was performed.
The results of Experiment 5 are shown in Table 5. Each condition shown in Table 5 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

According to the results of Experiment 5 shown in Table 5, when a silicon carbide thin film having a thickness of 200 μm is formed on the substrate surface, the variation in the emissivity of the wafer mounting surface (surface) should be within 3%, and the wafer mounting surface should be within 3%. It was confirmed that the uniformity of the epitaxial film formed on the silicon wafer was improved by setting the ratio of the average emissivity between the placement surface and the back surface to 1:1 to 1:0.8.

From the results of Experiments 1 to 5 above, regardless of the thickness of the silicon carbide thin film formed on the substrate surface, the variation in the emissivity of the wafer mounting surface (surface) is within 3%. and the ratio of the average emissivity between the wafer mounting surface and the back surface thereof is 1:1 to 1:0.8, the uniformity of the epitaxial film formed on the silicon wafer is improved.
More preferably, it was confirmed that the uniformity of the epitaxial film formed on the silicon wafer was improved by keeping the variation in the emissivity of the back surface of the susceptor within 3%.

[Experiment 6]
In Experiment 6, a suitable film thickness of the silicon carbide film formed on the surface of the carbon substrate was examined. In Examples 26-30 and Comparative Examples 11-13, the film thickness was adjusted by the supply time of the raw material gas. Furthermore, in Experiment 6, the raw material gas was diluted in the final stage of the raw material gas supply process, and the dispersion of the emissivity of the obtained wafer mounting surface and the back surface of the susceptor was kept within 3%. The average emissivity ratio of the placement surface and its back surface was adjusted to be in the range of 1:1 to 1:0.8.

Using the obtained susceptor, the epitaxial film formation process and the cleaning process were repeated to verify whether a predetermined life time (4000 hours of continuous operation) could be achieved.
In Example 26, the film thickness of the silicon carbide film on the main surface (wafer mounting surface) was 42 μm. The film thickness was 55 μm in Example 27, 58 μm in Example 28, 61 μm in Example 29, and 66 μm in Example 30.
The film thickness was 70 μm in Example 31, 80 μm in Example 32, and 100 μm in Example 33.
The evaluation of Experiment 6 is shown in Table 6.

As shown in Table 6, a susceptor with a silicon carbide film of less than 60 μm could not obtain the required life. Therefore, it was confirmed that the silicon carbide film preferably has a thickness of 60 μm or more.

[Experiment 7]
In Experiment 7, isotropic graphite was used as the material for the base material of the susceptor, and a carbon base material having a counterbore was prepared. Using the CVD apparatus shown in FIG. 3, a silicon carbide film was formed on the substrate surface under a plurality of film thickness forming conditions.
Next, using the susceptor formed under each condition, the silicon wafer was processed to form an epitaxial film.

In the manufacture of the susceptor, the CVD apparatus shown in FIG. 3 was used to adjust the film thickness by increasing or decreasing the processing time when forming the silicon carbide film on the surface of the substrate of the susceptor. Furthermore, in Experiment 7, the raw material gas was diluted in the final stage of the raw material gas supply process, and the variation in the emissivity of the obtained wafer mounting surface and the back surface of the susceptor was controlled to within 3%. The average emissivity ratio of the placement surface and its back surface was adjusted to be in the range of 1:1 to 1:0.8.
After the formation of the susceptor, the average thickness of the silicon carbide film formed on the other main surface (wafer non-mounting surface) relative to the average thickness of the silicon carbide film formed on the main surface (wafer mounting surface) The ratio of The average value of the film thickness of the silicon carbide film formed on the principal surface (wafer mounting surface) and the film thickness of the silicon carbide film formed on the other principal surface (wafer non-mounting surface) was determined by emissivity measurement. Cross sections at the same positions as the points were measured using an optical microscope, and their average was calculated.

As shown in Table 7, the ratio is 0.5 for Example 34, 0.6 for Example 35, 0.7 for Example 36, and 0.8 for Example 37; In Example 38 it was 0.9 and in Example 39 it was 1.0.
The ratio was 1.1 in Example 40, 1.2 in Example 41, 1.3 in Example 42, and 1.4 in Example 43.

In Examples 34 to 43, on the main surface (wafer mounting surface) of the susceptor, the ratio ( %) was set to 30%. Furthermore, on the main surface (wafer mounting surface) of the susceptor, the ratio (%) of the film thickness difference between the maximum film thickness and the minimum film thickness at the outer edge to the average film thickness of the thin film formed on the main surface is Both were 30%.
The results of Experiment 7 are shown in Table 7. Each condition shown in Table 7 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

From the results of Experiment 7, the thickness of the silicon carbide film formed on the other main surface (wafer non-mounting surface) relative to the average thickness of the silicon carbide film formed on the main surface (wafer mounting surface) of the susceptor. It was confirmed that if the average ratio is in the range of 0.7 to 1.2, the film thickness uniformity of the epitaxial film is improved.

[Experiment 8]
In Experiment 8, the CVD apparatus shown in FIG. 3 was used in the same manner as in Experiment 7, and a silicon carbide film was formed on the substrate surface under a plurality of film thickness forming conditions.
Next, using the susceptor formed under each condition, the silicon wafer was processed to form an epitaxial film.

In the manufacture of the susceptor, the CVD apparatus shown in FIG. 3 was used to adjust the film thickness by increasing or decreasing the processing time when forming the silicon carbide film on the surface of the substrate of the susceptor. Then, in the final stage of the raw material gas supply process, the raw material gas is diluted, and the variation in emissivity between the wafer mounting surface and the back surface of the obtained susceptor is set to within 3%. The average emissivity ratio was adjusted to be in the range of 1:1 to 1:0.8.
Ratio (%) of the film thickness difference between the center and the outer edge of the main surface (wafer mounting surface) of the susceptor taken out from the CVD apparatus after the completion of the thin film formation, to the average film thickness of the thin film formed on the main surface. asked for

Said percentages were 0% for Example 44, 10% for Example 45, 20% for Example 46, 30% for Example 47 and 40% for Example 48. The ratio was 50% in Example 49 and 60% in Example 50.

In Examples 44 to 50, the average thickness of the silicon carbide film formed on the main surface (wafer mounting surface) of the susceptor was formed on the other main surface (wafer non-mounting surface). The average ratio of the film thickness of each thin film was set to 1.0. Furthermore, on the main surface (wafer mounting surface) of the susceptor, the ratio (%) of the film thickness difference between the maximum film thickness and the minimum film thickness at the outer edge to the average film thickness of the thin film formed on the main surface is as follows: Both were 30%.
The results of Experiment 8 are shown in Table 8. Each condition shown in Table 8 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer as in Experiment 7. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

From the results of Experiment 8, it was found that the film thickness difference between the center and the outer edge of the main surface (wafer mounting surface) of the susceptor was within the range of 0% to 40% of the average film thickness of the thin film formed on the main surface. It was confirmed that the film thickness uniformity of the epitaxial film was improved.

[Experiment 9]
In Experiment 9, the CVD apparatus shown in FIG. 3 was used in the same manner as in Experiment 7, and a silicon carbide film was formed on the substrate surface under a plurality of film thickness forming conditions.
Next, using the susceptor formed under each condition, the silicon wafer was processed to form an epitaxial film.

In the manufacture of the susceptor, the CVD apparatus shown in FIG. 3 was used to adjust the film thickness by increasing or decreasing the processing time when forming the silicon carbide film on the surface of the substrate of the susceptor. Then, in the final stage of the raw material gas supply process, the raw material gas is diluted, and the variation in emissivity between the wafer mounting surface and the back surface of the obtained susceptor is set to within 3%. The average emissivity ratio was adjusted to be in the range of 1:1 to 1:0.8.
After the completion of thin film formation, on the main surface (wafer mounting surface) of the susceptor taken out from the CVD apparatus, the difference in film thickness between the maximum film thickness and the minimum film thickness at the outer edge, which is the thickness of the thin film formed on the main surface. A ratio (%) to the average was obtained.

Said percentage was 0% for Example 51, 10% for Example 52, 20% for Example 53, 30% for Example 54 and 40% for Example 55. The ratio was 50% in Example 56 and 60% in Example 57.

In Examples 51 to 57, the average thickness of the silicon carbide film formed on the main surface (wafer mounting surface) of the susceptor was formed on the other main surface (non-wafer mounting surface). The average ratio of the film thickness of the silicon carbide film was set to 1.0. Furthermore, on the main surface (wafer mounting surface) of the susceptor, the ratio (%) of the film thickness difference between the center and the outer edge to the average film thickness of the thin film formed on the main surface was set to 30%. .
The results of Experiment 9 are shown in Table 9. Each condition shown in Table 9 was evaluated based on the uniformity of the epitaxial film formed on the silicon wafer as in Experiments 7 and 8. The in-plane distribution of the film thickness of the epitaxial film was evaluated as ◯ when it was ±5% or less, Δ when it exceeded ±5% to 7%, and × when it exceeded ±7%.

From the results of Experiment 9, on the main surface (wafer mounting surface) of the susceptor, the film thickness difference between the maximum film thickness and the minimum film thickness at the outer edge is 0% to 0% of the average film thickness of the thin film formed on the main surface. It was confirmed that the film thickness uniformity of the epitaxial film was improved within the range of 40%.

Reference Signs List 1 susceptor 2 carbon substrate 3 thin film 4 counterbore 5 CVD device 10 chamber 11 gas inlet 12 gas outlet 20 support leg

Claims (5)

A susceptor having a substrate made of a carbon material and having one main surface on which a silicon wafer is placed and another main surface facing the one main surface,
The entire surface of the base material is coated with a thin film made of silicon carbide,
Variation in emissivity of the one main surface is within 3%, and the ratio of average emissivity of the one main surface and the other main surface facing each other is 1:1 to 1:0.8. A susceptor characterized by: 2. The susceptor according to claim 1, wherein a variation in emissivity between said one principal surface and the other principal surface is within 3%. The ratio of the thickness of the thin film formed on the other main surface to the thickness of the thin film formed on the one main surface is 0.7 or more and 1.2 or less, and the one main surface has a central portion and the outer edge portion is 40% or less of the average value of the film thickness of the thin film formed on the one main surface, and the maximum thickness and the minimum thickness of the outer edge portion of the one main surface 2. The susceptor according to claim 1, wherein the film thickness difference between the two is 40% or less of the average film thickness of the thin film formed on the one main surface. 2. The susceptor according to claim 1, wherein the thin film of silicon carbide formed on the entire surface of the substrate has a thickness of at least 60 [mu]m. A method for manufacturing the susceptor according to any one of claims 1 to 4,
supporting a substrate made of a carbon material in a chamber while moving a supporting position with respect to the substrate;
Manufacture of a susceptor characterized by comprising a step of forming a thin film made of silicon carbide on the entire surface of the substrate by supplying the material gas so that the supply direction is parallel to the one main surface of the substrate. Method.

Images