JP2915893B1 - Diffusion wafer manufacturing method and diffusion wafer - Google Patents

Abstract

An object of the present invention is to provide a method for manufacturing a diffusion wafer to which a two-piece slice technique can be applied even when forming a plurality of orientation flats or notches while suppressing autodoping from the outer peripheral portion of the diffusion wafer, and a diffusion wafer thereof. SOLUTION: The present invention is characterized in that after performing impurity diffusion on both sides of a wafer having a diameter slightly larger than a standard diameter, the wafer is divided into two parts, and a diffusion layer on an outer peripheral portion of the divided wafer is removed. A wafer comprising a layer and a non-diffusion layer is formed. When a plurality of orientation flats are formed on the outer peripheral edge of the wafer, a second orientation flat or notch is simultaneously formed in the outer peripheral portion removing step of the divided wafer.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method for manufacturing a diffusion wafer having a two-layer structure of an impurity diffusion layer and a non-diffusion layer. More specifically, the present invention relates to a wafer having an impurity diffusion layer in which impurities are diffused on both surfaces. The present invention relates to a method for producing a diffusion wafer formed by dividing the impurity diffusion layer and the non-diffusion layer by dividing the wafer into two along the center line of the thickness width, and particularly to a single crystal silicon wafer suitable as a substrate for a discrete element. The present invention relates to a method for producing a diffusion wafer having a required concentration or conductivity type, which is diffused by a vapor phase diffusion method.

[0002]

2. Description of the Related Art Conventionally, as a method of manufacturing a diffusion wafer used for a substrate for a discrete element (individual element) such as a transistor or a diode, Japanese Unexamined Patent Publication No. 1-293613 discloses a method of manufacturing a wafer having an impurity diffusion layer in which impurities are diffused on both sides. There is disclosed a technique of forming a wafer composed of the impurity diffusion layer and the non-diffusion layer by dividing the wafer into two along a center line of the width.

[0003] The outer periphery of these diffusion wafers is usually provided with an orientation flat (hereinafter referred to as an orientation flat) or a notch portion called a notch for the purpose of crystal orientation or alignment.

[0004] The diffusion wafer manufacturing technique will be described with reference to a flow chart of FIG. 8 by taking as an example a case where there are two orientation flats. (S1): FZ (Float Zoning) method or CZ (Czochralsk)
y)), a substantially cylindrical single crystal silicon ingot having an n-type low impurity concentration (n ⁻ ) is manufactured. (S2): The outer periphery of the ingot is processed to a predetermined standard diameter by a grinder. Here, _two orientation flats OF ₁ and OF ₂ are also provided at different positions by 90 ° by the outer peripheral grinding machine.

(S3): A slicer slices the wafer 1 'into a wafer 1' having a predetermined thickness larger than twice the final thickness (finished thickness of the diffusion wafer). (FIG. 4A) (S4): The slice distortion on the wafer surface is removed by a lapping machine to obtain a predetermined thickness.

(S5): Then, the wafer 1 'is etched by cleaning or chemical treatment (for example, a total of 4
μm or less). (S6): Impurity diffusion is performed using a diffusion source such as POCl ₃ called “deposition”, and the diffusion source is uniformly diffused at a predetermined concentration in the wafer surface.

(S7): After the completion of the deposition, the diffusion layer formed by the deposition is diffused to a predetermined depth by heat treatment, and a diffusion layer n containing a high concentration of P (phosphorus) is formed.
Form a ⁺ . Thereby wafer 1 'inside the sectional structure, the wafer having a non-diffusion layer impurity diffusion layer impurity is diffused around both sides of the ^{n-n +} is formed. (Figure 2
(A)) (S8): wafer 1A comprised a wafer from the two-layer structure of the impurity diffusion layer n ⁺ a non-diffusion layer ^n-by bisection along the center line of the thickness of the width by slicer ', 1
B ′ is formed (FIG. 2B). Thereafter, the non-diffusion layer ^n-
After the above-mentioned division into two parts, the sliced surface is processed using a surface grinder and a polishing machine.

[0008]

[SUMMARY OF THE INVENTION Now in the wafer diffusion layer formation step, the impurity diffusion diffusion layer n ⁺ both sides not only the wafer outer periphery over the entire periphery of the wafer will be formed, which two Since the wafer structure formed by division has the diffusion layer n ^{+ at the} outer peripheral edge of the wafer as shown in FIG. 2, this causes a quality abnormality called auto-doping during the device manufacturing process. Further, at the time of the above-mentioned two-division slicing, chips, chips, and the like are apt to be generated on the outer peripheral edge portion of the wafer, which causes a decrease in yield.

In the wafer, as shown in FIG. 4, an orientation flat O is used to distinguish the orientation and the type.
Of F ₁ and OF ₂ , OF ₂ is distributed to a 180 ° left-right symmetrical position, and as a result, one wafer 1
The A ′ or 1B ′ orientation flats are formed in reverse, making it impossible to apply the two-piece wafer slicing technique as a premise of the present technique.

In view of the above technical problems, an object of the present invention is to provide a method for manufacturing a diffusion wafer which can apply a two-piece wafer slicing technique when a plurality of orientation flats are formed on the outer peripheral edge of a wafer. . Another object of the present invention is to provide a method for manufacturing a diffusion wafer which does not cause an abnormal quality called auto-doping during a device manufacturing process when a two-piece wafer slicing technique is applied, and a diffusion wafer. To provide. Still another object of the present invention is to provide a method for manufacturing a diffusion wafer which can easily repair a chip or the like at the outer peripheral edge of the wafer during the two-slicing.

[0011]

According to the first aspect of the present invention,
A method for manufacturing a diffusion wafer, comprising forming a wafer comprising the impurity diffusion layer and the non-diffusion layer by dividing a wafer having an impurity diffusion layer in which impurities are diffused on both sides along a center line of a thickness width, After diffusing impurities on the entire surface of a wafer having a diameter slightly larger than the standard diameter, the wafer is divided into two parts, and the outer periphery of the divided wafer is removed to remove the wafer composed of the impurity diffusion layer and the non-diffusion layer. It is characterized by forming. According to this invention, after diffusing impurities on a wafer having a diameter slightly larger than the standard diameter, the wafer outer peripheral portion is removed to make the standard diameter, whereby impurities on the wafer outer peripheral edge formed at the time of the diffusion step are removed. The diffusion layer is removed, and as a result, there is no possibility that a quality abnormality called auto-doping occurs during the device manufacturing process.

Further, according to the present invention, even if chips or chips are generated at the outer peripheral edge of the wafer during the above-mentioned two-division slicing, the wafer can be easily repaired by removing the outer peripheral part of the wafer after the two-division.

In this case, the standard diameter is the final finished diameter of the diffusion wafer, and the "slightly larger diameter than the standard diameter" is obtained by adding the allowance for removing the outer peripheral portion of the wafer to the standard diameter. Means the caliber. Since the concentration of the dopant in the diffusion layer becomes higher near the back surface of the wafer, the effect of suppressing the auto doping can be obtained even if the allowance is very small. Also, the allowance is at least impure over the entire circumference of the wafer.
By setting the thickness of the material diffusion layer and completely removing it, auto-doping can be reliably prevented. On the other hand, if the diameter is too large, the processing efficiency is lowered or the setting diameter of the processing device or the like needs to be changed, which is not appropriate. Therefore, the maximum value of the "diameter slightly larger than the standard diameter" is a diameter that can be processed with the processing apparatus and the like with the standard diameter set,
Specifically, it is desirable to set the diameter to 1 mm or less from the standard aperture.

Further, the step of removing the outer peripheral portion of the wafer divided into two parts according to claim 2 includes a step of forming at least one notch or orientation flat, so that a diffusion wafer requiring a plurality of notches or orientation flats is provided. Even the two-slice technique can be applied.

A third aspect of the present invention is to prepare a wafer having a first notch or orientation flat and having a diameter slightly larger than a standard diameter, and diffusing impurities on both surfaces of the wafer. Removing the outer peripheral portions of two diffusion wafers obtained by dividing the wafer along the center line thereof, and forming at least one other notch or orientation flat at the same position on each of the diffusion wafers. A method for manufacturing a diffusion wafer characterized by the above feature, whereby even in the case of a diffusion wafer in which a plurality of notches or orientation flats are formed, the two-piece slice technique can be reliably applied.

According to a fifth aspect of the present invention, in the step of removing the outer peripheral portion of the wafer, the wafer is chamfered by a mechanical chamfering machine.
By performing the step together with the cutting, the step of removing the outer peripheral portion and the step of chamfering can be performed simultaneously or continuously, and the steps can be simplified. According to a sixth aspect of the present invention, there is provided a diffusion wafer in which an impurity diffusion layer and a non-diffusion layer form a layer structure in a thickness direction of the wafer, wherein a boundary between the diffusion layer and the non-diffusion layer is chamfered by the wafer. By using the diffusion wafer formed in the portion, auto doping during the device manufacturing process can be suppressed.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described in detail with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention unless otherwise specified, and are merely illustrative examples. Only.

A method for manufacturing a diffusion wafer according to an embodiment of the present invention will be described. First, in the present embodiment, a non-diffusion layer n ⁻ is 170 μm, and an impurity diffusion layer (hereinafter referred to as a diffusion layer).
That) a diffusion wafer used as the n ⁺ 200μm discrete elements (discrete devices) substrate standard diameter 150mm, the flowchart of FIG. 7 a method for manufacturing a wafer orientation flat OF _1, OF ₂ is provided two in 90 ° different positions Description will be made based on the drawings.

(S1): FZ (Float Zoning) method or CZ
A substantially cylindrical single crystal silicon ingot having an n-type low impurity concentration is manufactured by the (Czochralsky) method. (S2): The outer periphery of the ingot is processed by a grinder into a diameter (R + 2α) slightly larger than the standard diameter R. Here, α ≧ β is set. (Β: diffusion layer n ⁺ thickness) In this embodiment, the standard aperture is 150 mm and 2β is 4
Since the diameter is 00 μm, the aperture (R + 2α): 150 mm +
Set to 700 μm. At this time, as shown in FIG. 3 (A), first the orientation flat OF ₁ only, simultaneously formed by the outer peripheral grinder.

(S3): 2 mm of final thickness by slicer
The wafer 1 is processed into a wafer 1 having a predetermined thickness larger than twice. (S4): The slice distortion on the wafer surface is removed by a lapping machine to obtain a predetermined thickness. (S5): Then, the wafer 1 is subjected to etching by cleaning or chemical treatment (for example, an etching allowance of 4 μm or less in total on both sides). (S6): Impurity diffusion is performed using a diffusion source such as POCl ₃ called “deposition”, and the diffusion source is uniformly diffused at a predetermined concentration in the surface of the wafer.

(S7): After the completion of the deposition, the diffusion layer n ⁺ formed by the deposition is diffused to a predetermined depth (maximum β (200 μm in this embodiment)) by a heat treatment. As a result, as shown in FIG.
Wafer 1 inside the section structure, wafer 1 having a non-diffusion layer impurity diffusion layer impurity is diffused around both sides of the ^{n-n +} is formed. (S8): FIG. 1 (B) and FIG. 3 are obtained by dividing the wafer into two along the center line of the thickness width by a slicer.
The wafers 1A and 1B shown in FIG.

(S9): As shown in FIG. 1 (C), the wafer 1 is divided into two parts by chamfering with a mechanical chamfering machine.
The outer peripheral portions of A and 1B have a thickness of α (350 μm in this embodiment).
m) is removed in a ring shape to form a diffusion layer n on the outer periphery of the wafer.
Remove ⁺ . This makes it possible to suppress autodoping during the device manufacturing process, and at the same time, to easily repair any chipping or the like at the outer peripheral edge of the wafer at the time of the above-mentioned two-division slicing and to deteriorate the outer peripheral shape of the wafer which occurs during the diffusion manufacturing process. There are modified, the impurity diffusion layer n ⁺ a non-diffusion layer n ^over a two-layer structure consisting of standard diameter R of the wafer 1A, 1B is formed. 1 (B), (C), FIG. 3 (B),
As shown in (C), the second orientation flat OF ₂ is applied to the wafers 1A and 1B by a mechanical chamfering machine simultaneously with the removal of the diffusion layer n ⁺ .

[0023] (S10): Subsequently, (in this embodiment 170 [mu] m) the non-diffusion layer n ^over a predetermined thickness in order to, using a surface grinding machine and polishing machine, processing the sliced surface.

[0024] According to such an embodiment, after the bisected, by performing the second orientation flat processing of the orientation flat OF ₂ by a mechanical chamfering machine, without being limited to the position or the number of the orientation flat OF _1, OF _2, prepared Becomes possible.

Next, the difference between the present invention and the prior art will be described with reference to FIGS. As shown in FIG. 2, in the conventional technique in which a wafer 1 'having a standard aperture is divided into wafers 1A' and 1B 'after diffusion processing, impurity diffusion is performed not only on both the front and back surfaces of the wafer but also on the entire periphery of the wafer. Layer n ⁺
Exist at the outer peripheral edge of the wafer as shown in FIGS. Because exposed to the outer periphery surface of the words Kaere if non diffusion layer ^n-, which causes, for example, in the p-type boron diffusion and oxide film device manufacturing process in such form, referred to as autodoping quality abnormality occurs .

[0026] That is, by using the (n ⁺ / n ^over) the substrate wafer shown in FIG. 2 (B), the cleaning step, the oxide film with process, photolithography process, after the opening step, the base diffusion using boron diffusion source In this case, as shown in FIG. 6, the n ⁺ diffusion layer on the outer peripheral portion of the wafer is exposed in the opening step after the photolithography step, and phosphorus is out-diffused by the thermal diffusion in the base diffusion step, and the base portion ( In this case, the n ^- type collector portion performs not only p-type diffusion but also n-type diffusion. (Auto dope)

On the other hand, according to the present invention, in order to remove the diffusion layer n ⁺ on the outer peripheral portion of the wafer 1 with a mechanical chamfering machine after the diffusion manufacturing step and the completion of the two divisions with the diameter of the ingot set to be relatively large, FIG. (B) and the non-diffusion layer ^n-without diffusion layer n ⁺ is present on the outer periphery such as shown in FIG. 6, Build the wafer having a two-layer structure as shown in FIG. 1 (C). Therefore, when individual semiconductors such as transistors are manufactured as shown in FIG. 5 using the wafers 1A and 1B manufactured by such a method, the occurrence of the auto-doping can be suppressed.

On the other hand, the conventional method of FIG. 4, to describe the differences between the present invention of FIG. 3, the conventional method of FIG. 4, for processing the second orientation flat OF ₂ is to be already processed when grinding the outer periphery of the ingot , The position of the second orientation flat OF ₂ is the first orientation flat OF
_In contrast to the above, since the wafer is located 90 ° clockwise, when the wafer is divided into two parts after diffusion, one of the two divided wafers 1A ′ and 1B ′ is the second wafer 1A ′. to position the first orientation flat oF ₁ orientation flat oF _2, will be formed at a position shifted 90 ° in a counterclockwise direction, the two wafer 1A of different specifications ', 1B' is completed.
(See FIG. 4B)

On the other hand, in the embodiment of the present invention, the processing of the first orientation flat OF ₁ is performed before the two divisions (FIG. 3).
(A)) is processed in the second orientation flat OF ₂ because performed after bisection (FIG 3 (C)), can two sheets wafers having the same specification.

[0030]

As described above, according to the present invention, when the two-piece wafer slicing technique is applied, it is possible to suppress an abnormal quality called auto-doping during the device manufacturing process. Even if chipping or the like occurs at the outer peripheral edge of the wafer at the time of slicing, it can be easily repaired, so that the yield is improved. According to the second and third aspects of the present invention, the two-piece wafer slicing technique can be applied even when a plurality of notches or orientation flats are formed.

[Brief description of the drawings]

FIG. 1 shows a manufacturing procedure of a diffusion wafer according to an embodiment of the present invention, particularly showing a manufacturing procedure focusing on a cross-sectional structure of a diffusion layer and a non-diffusion layer.

FIG. 2 shows a manufacturing procedure of a diffusion wafer according to a conventional technique, and particularly shows a manufacturing procedure focusing on a cross-sectional structure of a diffusion layer and a non-diffusion layer.

FIG. 3 shows a manufacturing procedure of a diffusion wafer according to an embodiment of the present invention, and particularly shows a manufacturing procedure focusing on the formation of an orientation flat.

FIG. 4 shows a manufacturing procedure of a diffusion wafer according to the prior art, and particularly shows a manufacturing procedure focusing on the formation of an orientation flat.

FIG. 5 shows a cross-sectional structure of a wafer in base diffusion using a diffusion wafer according to an embodiment of the present invention.

FIG. 6 shows a cross-sectional structure of a wafer in base diffusion using a diffusion wafer according to the prior art.

FIG. 7 is a flowchart illustrating a procedure for manufacturing a diffusion wafer according to the embodiment of the present invention.

FIG. 8 is a flowchart showing a procedure for manufacturing a diffusion wafer according to the conventional technique.

[Explanation of symbols]

1, 1 'Wafer before splitting 1A, 1B, 1A', 1B 'Wafer after splitting OF ₁ First orientation flat OF ₂ Second orientation flat n ⁺ Diffusion layer n ^- Non-diffusion layer

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Shozaburo Goto 1-4-2 Marunouchi, Chiyoda-ku, Tokyo Shin-Etsu Semiconductor Co., Ltd. (72) Katsumi Ichimura 1-4-2 Marunouchi, Chiyoda-ku, Tokyo Shin-Etsu (58) Investigated field (Int.Cl. ⁶ , DB name) H01L 21/02 H01L 21/223

Claims (6)

(57) [Claims] 1. A diffusion wafer formed by dividing a wafer having an impurity diffusion layer in which impurities are diffused on both sides along a center line of a thickness width into a wafer including the impurity diffusion layer and a non-diffusion layer. In the manufacturing method, the impurity diffusion is performed on the entire surface of a wafer having a diameter slightly larger than the standard diameter, and then the two divisions are performed. A method for producing a diffusion wafer, comprising forming a wafer having a standard diameter consisting of layers. 2. The method for manufacturing a diffusion wafer according to claim 1, wherein the step of removing an outer peripheral portion of the divided wafer includes a step of forming at least one notch or an orientation flat (hereinafter, referred to as an orientation flat). 3. A wafer having a first notch or orientation flat and having a diameter slightly larger than a standard diameter is prepared, and impurities are diffused to the entire surface of the wafer. 2 obtained by dividing into two
A method for manufacturing a diffusion wafer, comprising: a step of removing an outer peripheral portion of a plurality of diffusion wafers; and a step of forming at least one other notch or orientation flat at the same position on each of the diffusion wafers. 4. An aperture slightly larger than the standard aperture,
4. The method for producing a diffusion wafer according to claim 1, wherein the diameter is a diameter obtained by adding at least twice the thickness of the impurity diffusion layer to the standard diameter. 5. The step of removing the outer peripheral portion of the wafer is performed on a mechanical surface.
4. The method for producing a diffusion wafer according to claim 1, wherein the method is performed together with wafer chamfering by a chamfering machine . 6. A diffusion wafer in which an impurity diffusion layer and a non-diffusion layer form a layer structure in a thickness direction of the wafer, wherein a boundary between the diffusion layer and the non-diffusion layer is formed in a chamfered portion of the wafer. Diffusion wafer characterized in that: