JP2003273368A - Manufacturing method of semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent a movable portion from sticking on a portion around itself, in a manufacturing method of a semiconductor provided with a laminate consisting of a first semiconductor layer and a second semiconductor layer laminated thereon through an insulating layer, and a movable section formed on the second semiconductor layer and being displaceable in response to application of dynamic quantity. <P>SOLUTION: In this manufacturing method of semiconductor device, to suppress sticking of a movable section 20 and fixed electrodes 32, 42, ashing is executed, provided that the surface of an oxide film 13 is not charged, after executing a movable section forming process shown in (d), thereby providing a mask member 50 on a second semiconductor layer 12. In this way, in the movable section forming process, even if sections 20, 30, 40 and 13a are temporarily charged, the charges on theses sections are removed by electrons respectively. Thus, the section 20 and the electrodes 32, 42 can be prevented from sticking (adhering) to the portion around themselves. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a laminated body formed by laminating a second semiconductor layer on a first semiconductor layer with an insulating layer interposed between the laminated body and the second semiconductor layer. The present invention relates to a method of manufacturing a semiconductor device including a movable part that can be displaced.

[0002]

2. Description of the Related Art As this type of semiconductor device, for example, the one described in Japanese Patent Laid-Open No. 11-274142 has been proposed.

This is a laminated body in which a second semiconductor layer is laminated on the first semiconductor layer with an insulating layer interposed therebetween.
It uses an OI substrate.

Then, after forming a trench for defining the movable portion by dry etching so as to reach the insulating layer from the surface of the second semiconductor layer, further dry etching is performed to laterally extend the bottom of the trench. By applying etching ions to the second semiconductor layer located to remove the second semiconductor layer located in the lateral direction, a movable portion as a second semiconductor layer separated from the insulating layer is formed. There is.

[0005]

When the manufacturing method as described in the above-mentioned conventional publication is used, when the trench is formed and further dry etching is performed, the insulating layer at the bottom of the trench is charged.

Then, due to the charging, the etching ions of the dry etching repel, and the etching ions bend from the depth direction of the trench to the lateral direction orthogonal to the depth direction and are located at the lateral direction of the bottom of the trench. Second
Etching ions hit the semiconductor layer of.

Therefore, the second semiconductor layer located laterally at the bottom of the trench is etched and removed to form the movable portion.

Here, in the above-mentioned conventional publication, in order to detect the etching depth, a movable portion dedicated to the etching depth detection is formed, and by utilizing the charging, an adjacent movable portion dedicated to the etching depth detection is formed. I try to stick each other.

However, according to the studies by the present inventors, in the above-mentioned conventional publication, the insulating layer at the bottom of the trench is charged by dry etching performed after the formation of the trench. 5 (a) and FIG. 5 (a) and FIG. 5 (b).
As shown in (b), it has been found that there is a problem that sticking (adhesion) occurs between the respective parts due to the uneven distribution of charges between these parts.

As described above, when the movable portion used for the original function of the device causes sticking due to the charging, it is apparent that the original movable portion is not moved and the device characteristics are adversely affected.

Therefore, in view of the above problems, an object of the present invention is to form a second semiconductor layer and a laminated body in which a second semiconductor layer is laminated on a first semiconductor layer with an insulating layer interposed therebetween. In a method of manufacturing a semiconductor device including a movable portion that can be displaced according to the application of a mechanical amount, it is possible to prevent the movable portion from sticking to a peripheral portion thereof.

[0012]

According to a first aspect of the present invention, there is provided a method of manufacturing a semiconductor device comprising: a laminated body formed by laminating a second semiconductor layer on a first semiconductor layer with an insulating layer interposed therebetween; In a method of manufacturing a semiconductor device, which comprises a movable part formed in the semiconductor layer and displaceable in response to the application of a mechanical amount, a stacked body is prepared, and the insulating layer is reached from the surface of the second semiconductor layer.
After performing the trench forming step of forming a trench for defining the movable portion and the trench forming step, etching is performed to charge the surface of the insulating layer at the bottom of the trench,
A movable part forming step of forming a movable part by applying etching ions to a second semiconductor layer laterally located at the bottom of the trench and removing the second semiconductor layer laterally located; An ashing step of performing ashing under the condition that the surface of the insulating layer is not charged after performing the part forming step.

According to the first aspect of the present invention, the surface of the insulating layer is charged in the movable portion forming step, so that the movable portion or a portion of the second semiconductor layer facing the movable portion or the insulating portion is insulated. Even if the portion of the layer facing the movable portion is charged, the ashing step is performed after the movable portion forming step is performed under the condition that the surface of the insulating layer is not charged, so that the charge of each portion is removed. Therefore, it is possible to prevent the movable portion from sticking to the surrounding portion.

It should be noted that the removal of the electric charge of each of the above-mentioned parts is not limited to the complete removal of the electric charge, but at least the electric charge may be removed to such an extent that the movable part does not stick to the surrounding parts.

The method of manufacturing a semiconductor device according to a second aspect of the present invention is characterized in that the frequency applied when performing the ashing step is set to 5 MHz or less, preferably 600 Hz or less under the condition that the surface of the insulating layer is not charged. There is.

According to the second aspect of the invention, the frequency applied at the time of performing the ashing step is set to 5 MHz or less, preferably 600 Hz or less, so that the ashing is performed without charging the surface of the insulating layer. Therefore, the manufacturing method according to claim 1 can be realized. The ashing performed under the condition that the surface of the insulating layer is not charged can be realized by pulse oscillation using a low frequency RF power source as described in claim 3.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION In this embodiment, the present invention is applied to a differential capacitance type semiconductor acceleration sensor as a semiconductor device.

FIG. 1 shows an overall plan structure of a semiconductor type acceleration sensor S1 of this embodiment, and FIG. 2 shows a chain line Q in FIG.
It is a figure which shows the typical partial cross-section structure expressed in the state which combined the cross-section structure in each part shown by 1, Q2, and Q3.

The semiconductor type acceleration sensor S1 can be applied to, for example, an automobile acceleration sensor or a gyro sensor for controlling the operation of an airbag, ABS, VSC or the like.

As shown in FIG. 2, the laminated body constituting the semiconductor type acceleration sensor S1 has an oxide film 13 as an insulating layer on a first silicon layer 11 as a first semiconductor layer.
The second silicon layer 12 as the second semiconductor layer via the
Is a rectangular SOI (Silicon On Insulator) substrate 10 formed by stacking.

Further, as shown in FIG. 1, by forming a trench 14 in the second silicon layer 12, a beam having a comb-tooth shape composed of a movable portion 20 and fixed portions 30 and 40. A structure is formed.

The movable portion 20 is composed of a rectangular weight portion 21 and spring portions 22 formed at both ends of the weight portion 21, and the anchor portions 23 a and 23 are interposed via the spring portions 22.
It is integrally connected to b.

Here, the anchor portions 23a and 23b are
It is fixed to the oxide film 13 located immediately below it (see FIG. 2).
The movable portion 20 (the weight portion 21 and the spring portion 22) located between the anchor portions 23a and 23b is located away from the oxide film 13 located immediately below the movable portion 20 (the weight portion 21 and the spring portion 22).

That is, the movable portion 20 has the weight portion 2 between the anchor portions 23a and 23b fixed to the oxide film 13.
1 and the spring portion 22 are suspended on the oxide film 13.

Further, the spring portion 22 has a rectangular shape in which two beams are connected at both ends thereof, and has a spring function of displacing in a direction orthogonal to the longitudinal direction of the beams.

Then, the spring portion 22 displaces the weight portion 21 in the arrow X direction when receiving the acceleration including the component in the arrow X direction in FIG. 1, and returns to the original state according to the disappearance of the acceleration. It is designed to be restored.

As described above, the movable portion 20 is the anchor portion 2
With 3a and 23b as fixed points, they can be displaced in the above-mentioned arrow X direction in response to the application of acceleration.

Further, the weight 21 is integrally formed in a comb-teeth shape in opposite directions from both side surfaces of the weight 21 in a direction orthogonal to the displacement direction of the spring 22 (direction of arrow X). A plurality of movable electrodes 24 are provided. In FIG. 1, three movable electrodes 24 are formed on each of the left and right sides of the weight portion 21 so as to project.

Each movable electrode 24 is formed in a beam shape having a rectangular cross section, and is separated from the oxide film 13 (for example, several μ).
m)) (see FIG. 2).

As described above, each movable electrode 24 has a movable portion 20.
Is formed integrally with the spring portion 22 and the weight portion 21, and can be displaced together with the weight portion 21 in the displacement direction of the spring portion 22.

The fixing portions 30 and 40 are supported by another pair of opposing side portions of the oxide film 13 where the anchor portions 23a and 23b are not supported.

Here, the fixed portions 30 and 40 are the weight 21
1, two fixing parts 30 and 40 are provided with the first fixing part 30 located on the left side in FIG. 1 and the second fixing part 40 located on the right side in FIG. Are electrically independent of each other.

The fixing portions 30 and 40 are connected to the wiring portions 31 and 4, respectively.
1 and the fixed electrodes 32 and 42. The wiring parts 31 and 41 are fixed to the oxide film 13 located immediately below the wiring parts 31 and 41, respectively.
Supported by.

The fixed electrodes 32 and 42 are arranged in the respective wiring portions 31, in the direction orthogonal to the displacement direction of the spring portion (direction of arrow X).
41 from the side surface toward the weight portion 21 in a comb-teeth shape,
It is arranged so as to mesh with the comb teeth of the movable electrode 24.
In addition, in FIG. 1, the fixed electrodes 32 and 42 are connected to the wiring portions 3 respectively.
Three units are provided integrally with Nos. 1 and 41.

Each of the fixed electrodes 32 and 42 is formed in a beam shape having a rectangular cross section, and is separated from the oxide film 13 while being cantilevered by the wiring portions 31 and 41 (for example, several μm).
m)) (see FIG. 2).

The individual fixed electrodes 32 and 42 are
The side surface is arranged to face the corresponding side surface of each movable electrode 24 in parallel with a predetermined detection interval.

Further, each wiring portion 3 of each fixing portion 30, 40
Fixed electrode pads 31a and 41a for wire bonding are respectively formed in predetermined areas on Nos. 1 and 41, and movable electrode pads 20a for wire bonding are formed in predetermined areas on one anchor portion 23b. ing. The electrode pads 20a, 31a, 41a are
For example, it is made of aluminum.

Although not shown, the semiconductor type acceleration sensor S1 of this embodiment is fixed to the package on the back surface (surface opposite to the oxide film 13) of the first silicon layer 11 with an adhesive or the like. Has been done.

This package contains circuit means, and this circuit means and the above-mentioned electrode pads 20a, 3 are provided.
1a and 41a are electrically connected by gold or aluminum wire bonding or the like.

In such a structure, the first fixing portion 3
The fixed electrode 32 on the 0 side is the first fixed electrode and the second fixed portion 4
When the 0-side fixed electrode 342 is the second fixed electrode, the first
A first capacitor CS1 is formed in the detection interval between the fixed electrode 32 and the movable electrode 24, and a second capacitor CS2 is formed in the detection interval between the second fixed electrode 42 and the movable electrode 24.

Then, when acceleration is applied to the semiconductor type acceleration sensor S1, the spring function of the spring section 22 causes the movable section 20 as a whole to move integrally in the arrow X direction with the anchor sections 23a and 23b as fulcrums. The detection interval changes according to the displacement of the electrode 24, and the capacitors CS1 and CS2 change.

The movable electrode 24 and the fixed electrodes 32, 4
The applied acceleration is detected on the basis of the change in the differential capacitance (CS1-CS2) due to 2.

Next, with reference to FIG. 3, a method of manufacturing the semiconductor type acceleration sensor S1 of this embodiment will be described.
Incidentally, this FIG. 3 is a schematic cross section corresponding to FIG.
It is process explanatory drawing which shows the state of the workpiece | work in the middle of this manufacture.

First, as shown in FIG. 3A, the first
An SOI substrate (laminate) 10 in which a second silicon layer (second semiconductor layer) 12 is laminated on a silicon layer (first semiconductor layer) 11 via an oxide film (insulating layer) 13 prepare.

In this SOI substrate 10, for example, a silicon single crystal having a surface orientation of (100) is used as the first silicon layer 11 and the second silicon layer 12, and both silicon layers 11 and 12 are It is possible to use a structure in which the oxide film 13 made of a silicon oxide film (SiO ₂ ) having a film thickness of about 1 μm is bonded to the substrate.

Further, Al (aluminum) is used, for example,
Electrode pads 20a, 31a and 41a for taking out signals are formed by vapor deposition of about μm, photo and etching. The electrode pad 20a is not shown in FIG.

Subsequently, as shown in FIG. 3B, etching is performed under the etching condition that the surface of the oxide film 13 is not charged to define the first trench (the spring portion 22 and the movable electrode 24) having a wide opening width. Trenches 14a are formed so as to reach the oxide film 13 from the surface of the second silicon layer 12, and further, etching is performed under the same etching conditions as shown in FIG. Narrow second
14b (trench that defines the frame hollow portion of the spring portion 22) is formed so as to reach the oxide film 13 from the surface of the second silicon layer 12 (trench forming step).
The etching performed under the etching condition in which the surface of the oxide film 13 is not charged can be realized by pulse oscillation using a low frequency RF power source.

As a result, the movable portion 20 and the fixed portion 30,
A trench 14 for defining 40 can be formed from the surface of the second silicon layer 12 to reach the oxide film 13.

More specifically, the second silicon layer 1
The beam structures 20, 30, 4 having a comb-tooth shape on the surface of 2.
A mask material 50 for forming a pattern corresponding to 0 is formed by a resist or the like using a photolithographic technique,
A trench shape is formed vertically up to the oxide film 13 by dry etching such as plasma etching.

For dry etching, CF ₄ or SF _{6 is used.}
ICP (inductively coupled plasma) using an etching gas such as, or RIE using the same etching gas as above
An etching method such as (reactive ion etching) can be adopted.

Subsequently, as shown in FIG. 3D, after the trench forming step, dry etching is performed under the etching condition that the surface of the oxide film 13 at the bottom of the trench 14 is charged, and the movable electrode 24 is included. Weight 21 and spring 2
2 forms the movable portion 20 released from the oxide film 13 (movable portion forming step). The etching performed under the etching condition that the surface of the oxide film 13 is charged is high frequency RF.
It can be realized by continuous oscillation using a power supply.

In this movable part forming step, the fixed part 3
The fixed electrodes 32 and 42 of 0 and 40 are released from the oxide film 13.

3D, the movable portion 24 of the movable portion 20, the anchor portion 23b, the spring portion 22, and the fixed portion 3 are shown.
The wiring portions 31 and 41 of 0 and 40 and the second fixed electrode 42 are shown.

By the dry etching in the process of forming the movable portion, the oxide film 1 at the bottom of the trench 14 is caused by etching ions (CF ₄ and SF ₆ are converted into plasma).
3 The surface is charged (usually positively charged).

Then, the etching ions receive a repulsive force on the surface of the charged oxide film 13 and are bent laterally as shown by the arrow Y direction in FIG. 3C.

As a result, as shown in FIG. 3D, the etching ions hit the second silicon layer 12 located in the lateral direction (the arrow Y direction) of the bottom of the trench 14 and the lateral direction (the arrow Y). The second silicon layer 12 located in the direction) is etched and removed.

Then, the movable portion 20 and the fixed electrodes 32 and 42 floating from the oxide film 13 are formed, and then the mask material 50 is removed by ashing under the condition that the surface of the oxide film 13 is not charged. The semiconductor type acceleration sensor S1 as shown in FIGS.
Is completed.

As described above, in the present embodiment, the trench forming step performs etching under the etching condition that does not charge the surface of the oxide film 13, and the movable portion forming step performs etching under the etching condition that charges the surface of the oxide film 13. It is characterized by that.

Thereby, FIG. 3 (b) and FIG. 3 (c)
In the trench forming step as shown in FIG. 6, even if the etching time until the second trench 14b having the narrowest opening reaches the oxide film 13 is set to the entire etching time of the trench etching, the trench forming step is performed. Since the etching is performed under the etching condition that the surface of the film 13 is not charged, the first trench 1 having a wide opening width is formed.
In 4a, it is possible to prevent the occurrence of a notch phenomenon in which the bottom portion is side-etched.

As a result, as shown in FIG.
After the trench forming step, etching is performed under etching conditions in which the surface of the oxide film 13 is charged to release the weight portion 21 including the movable electrode 24, the spring portion 22, and the fixed electrodes 32 and 42 from the oxide film 13. Even if the above procedure is performed, the trench angle and the gap between the oxide film 13 and the hollow portion of the spring portion 22 and the movable electrode 24 do not vary.

Therefore, it is possible to reduce the processing variation of the spring portion 22 and obtain a uniform spring function, so that stable sensor characteristics can be exhibited.

The trench forming step is not limited to the etching condition in which the surface of the oxide film 13 is charged, but at least the etching condition in which the notch phenomenon does not occur in the spring portion 22 may be performed.

By the way, in the dry etching in the movable portion forming step, the surface of the oxide film 13 at the bottom of the trench 14 is charged by etching ions. At this time, for example, the movable portion 20 or the movable portion 20 via the trench 14 is charged. The second silicon layer 12 (that is, the fixed portions 30 and 40) and the oxide film 13 facing each other are also easily charged.

When this charging occurs, due to the uneven distribution of charges between the above-mentioned respective parts, for example, arrows K1 and K in FIG.
2, K3, the movable electrode 24 and the fixed electrode 3
Sticking (adhesion) may occur between 2, 42 or the oxide film 13 and between the fixed electrode 42 and the oxide film 13.

When sticking occurs in the movable portion 20 used for the original function of the device, the displacement characteristic of the movable portion 20 changes and the facing area between the movable electrode 24 and the fixed electrodes 32 and 42 (the detection capacitance is changed). Change) and adversely affect the characteristics of the sensor.

Therefore, in the present embodiment, in order to suppress the sticking of the movable portion 20 and the fixed electrodes 32 and 42 due to this charging, the movable portion forming step shown in FIG. 3D is executed in the above manufacturing method. After that, ashing is performed under the condition that the surface of the oxide film 13 is not charged, and the second semiconductor layer 1
The mask material 50 provided on 2 is removed.

As described above, by performing the ashing under the condition that the surface of the oxide film 13 is not charged, it becomes possible to sufficiently supply the negative electrons to the inside of the trench 14, so that the movable portion 20 or the second portion. Portions 30, 40 of the silicon layer 12 facing the movable portion 20 or the oxide film 13
It is possible to remove the electric charges generated by positively charging the oxide film 13 in the movable portion forming step from the portions of the movable portion 20 facing the movable portion 20 and the fixed electrodes 32 and 42.

As a result, even if the respective portions 20, 30, 40, 13a are temporarily charged in the movable portion forming step, the electric charges in these respective portions are removed by the electrons.
It is possible to prevent the movable part 20 and the fixed electrodes 32 and 42 from sticking to the surrounding parts.

In removing the electric charges in the above respective parts, the electric charges may be completely removed, but at least to the extent that the movable part 20 and the fixed electrodes 32 and 42 do not stick to the surrounding parts. Just remove it.

By the way, in the dry etching in the movable portion forming step, the etching step for removing the second silicon layer 12 and the protective film forming step for forming the protective film on the inner wall of the trench 14 are alternately repeated. In this protective film forming step, the protective film formed on the back surface side of the movable portion 20 serves as a mask, and in the etching step, the etching ions repelled on the surface of the charged oxide film 13 cause the back surface side of the movable portion 20 (the oxide film). (Part facing 13)
By hitting the movable part 2 as shown in FIG.
A needle-like protrusion 60 is formed on the back surface side of 0.

When the protrusion 60 is formed on the back surface side of the movable portion 20, the distance between the opposing movable portion 20 and the oxide film 13 becomes narrow, or the protrusion 60 peels off from the movable portion 20 due to a change with time. The movable portion 20 may be sandwiched between the movable electric portion 20 and the oxide film 13 or the like so that the movable portion 20 may be sandwiched between the opposed portions.
Therefore, the oxide film 13 is likely to stick.

As described above, when the movable portion 20 used for the original function of the device causes sticking, the displacement characteristic of the movable portion 20 is changed and the facing area between the movable electrode 24 and the fixed electrodes 32 and 42 is changed (the detection capacitance is changed). Change) and adversely affect the characteristics of the sensor.

Therefore, in the present embodiment, as described above,
After the movable portion forming step shown in FIG. 3D is performed, ashing is performed under a condition that the surface of the oxide film 13 is not charged, so that the needle-shaped needles formed on the back surface side of the movable portion 20 in the movable portion forming step. Since the protrusion 60 can be removed, sticking between the movable portion 20 and the oxide film 13 can be suppressed.

Here, the etching conditions under which the surface of the oxide film 13 is not charged in the ashing performed after the dry etching in the trench forming step and the movable portion forming step and the surface of the oxide film 13 in the movable portion forming step are charged. Regarding the etching conditions to be used, the method described in US Pat. No. 6,187,685 can be used. The contents described in this publication are shown below.

First, when the frequency applied during etching is set to 5 MHz or higher, preferably 10 MHz or lower, of the etching ions and electrons supplied during dry etching, the electrons that neutralize the charge follow the electric field direction. Then, the advancing direction changes, but the advancing direction of the etching ions does not change. Therefore, a larger amount of etching ions than electrons are supplied to the bottom of the trench, and the surface of the oxide film can be charged.

When the frequency applied during etching is set to 5 MHz or less, preferably 600 Hz or less, not only the electron traveling direction but also the etching ion traveling direction changes following the electric field direction. Etching ions and electrons are supplied to the
The supplied electrons can alleviate the charge on the surface of the oxide film. The etching for reducing the charge on the surface of the oxide film is realized by pulse oscillation using a low frequency RF power supply.

Therefore, in the ashing performed after the dry etching and the movable portion forming step of the trench forming step shown in FIGS. 3B and 3C, the frequency applied at the time of etching is 5 MHz or less, preferably. 6
If the frequency is set to 00 Hz or less, it is possible to realize etching in which the surface of the oxide film 13 is not charged, and in the movable part forming step shown in FIG. 3D, the frequency applied during etching is 5 MHz or more, preferably 10 Hz or less. When set to, etching in which the surface of the oxide film 13 is charged can be realized.

The present invention is not limited to the above embodiment, but can be applied to various aspects.

For example, although the present invention is applied to the acceleration sensor in the above embodiment, the present invention is not limited to this.
It can be applied to semiconductor devices such as angular velocity sensors and pressure sensors.

The fixed electrodes 32 and 42 are made of the oxide film 13
It may be in a state of being connected to the oxide film 13 without being released from. For example, the width of the beam in the fixed electrodes 32, 42 may be made wider than that of the movable electrode 24 so that the remaining portion connected to the oxide film 13 is formed in the fixed electrode at the end of the dry etching in the movable portion forming step. Good.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall planar configuration of a semiconductor type acceleration sensor according to an embodiment of the present invention.

FIG. 2 is a schematic partial cross-sectional view of the semiconductor type acceleration sensor shown in FIG.

3A to 3E are process explanatory views showing a method for manufacturing a semiconductor acceleration sensor according to an embodiment of the present invention.

FIG. 4 is a diagram showing a state in which needle-like protrusions are formed on the back surface side of a movable portion of the semiconductor acceleration sensor.

FIG. 5 is a diagram showing sticking in a conventional semiconductor acceleration sensor.

[Explanation of symbols]

S1 ... Semiconductor type acceleration sensor, 10 ... SOI substrate (laminated body), 11 ... a first silicon layer (first semiconductor layer), 12 ... second silicon layer (second semiconductor layer), 13 ... Oxide film (insulating layer), 14 ... trench, 14a ... the first trench, 14b ... second trench, 20 ... movable part, 20a ... movable electrode pad, 21 ... Weight part, 22 ... Spring part, 23a, 23b ... Anchor part, 24 ... movable electrode, 30 ... the 1st fixed part, 31, 41 ... Wiring part, 31a, 41a ... Fixed electrode pad, 32 ... a first fixed electrode, 40 ... a second fixing portion, 42 ... a second fixed electrode, 50 ... Mask material, 60 ... protrusion, CS1 ... the first capacity, CS2 ... Second capacity.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Kazuhiko Kano             1-1, Showa-cho, Kariya city, Aichi stock market             Inside the company DENSO F-term (reference) 4M112 AA02 BA07 CA21 CA22 CA24                       CA25 DA03 DA15 DA18 EA03                       EA06 EA11 FA20                 5F004 AA16 BA04 BA20 BB11 BD01                       BD03 DA01 DA18 DB01 EA23                       EB08

Claims (3)

[Claims] 1. A second semiconductor layer formed on the first semiconductor layer with an insulating layer interposed therebetween.
In a method for manufacturing a semiconductor device, comprising: a laminated body formed by laminating semiconductor layers, and a movable portion that is formed in the second semiconductor layer and that can be displaced in accordance with the application of a mechanical amount, preparing the laminated body, A trench forming step of forming a trench for defining the movable portion so as to reach the insulating layer from the surface of the second semiconductor layer; and after performing the trench forming step, etching is performed to form the trench of the trench. The surface of the bottom insulating layer is charged, and the etching ions are applied to the second semiconductor layer located laterally of the bottom of the trench to remove the second semiconductor layer located laterally. By
Manufacturing of a semiconductor device comprising: a movable part forming step of forming the movable part; and an ashing step of performing ashing under a condition that the surface of the insulating layer is not charged after the movable part forming step is performed. Method. 2. The conditions under which the surface of the insulating layer is not charged are:
The frequency applied when performing the ashing process is 5 MHz.
The method for manufacturing a semiconductor device according to claim 1, wherein the frequency is set to z or less, preferably 600 Hz or less. 3. The method of manufacturing a semiconductor device according to claim 1, wherein the ashing performed under the condition that the surface of the insulating layer is not charged is performed by pulse oscillation using a low-frequency RF power source.