JP7364163B2 - MEMS element and its manufacturing method - Google Patents

Description

The present invention relates to a MEMS element and a method for manufacturing the same, and more particularly to a capacitive MEMS element used as a microphone, various sensors, switches, etc., and a method for manufacturing the same.

Conventionally, in MEMS (Micro Electro Mechanical Systems) devices using semiconductor processes, capacitive MEMS devices have been widely used, which have a structure that includes a fixed electrode and a movable electrode arranged to face the fixed electrode with an air gap interposed therebetween. Are known. In this type of MEMS element, a fixed electrode and a movable electrode are formed with an intermediate film called a sacrificial layer sandwiched therebetween, and then a part of the sacrificial layer is removed by etching to form an air gap.

Generally, the sacrificial layer is removed by wet etching using a solution that selectively etches the sacrificial layer, or by dry etching using an etching gas. Among these methods, in order to form an inexpensive MEMS element, it is preferable to select the wet etching method, which has a low manufacturing cost.

However, in the wet etching method, when etching the sacrificial layer, a problem arises in that the fixed electrode film including the fixed electrode and the movable electrode film including the movable electrode stick together. In order to solve this problem, a method has been adopted in which small protrusions are formed on opposing surfaces of the fixed electrode film and the movable electrode film.

FIG. 3 is an explanatory diagram of a manufacturing method for forming protrusions previously proposed by the applicant of the present invention (Patent Document 1). First, as shown in FIG. 3(a), a thermal oxide film 2 is formed on a substrate 1 made of a silicon substrate, and a movable electrode 3 made of conductive polycrystalline silicon (polysilicon) is formed on the thermal oxide film 2. do. Further, a sacrificial layer 4 made of a USG (Undoped Silicate Glass) film is laminated on the movable electrode 3, and a fixed electrode 5 made of conductive polycrystalline silicon is laminated on this sacrificial layer 4.

A through hole 6 is formed in the fixed electrode 5, and the surface of the sacrificial layer 4 exposed inside the through hole 6 is etched. Thereafter, a nitride film 7 is formed over the entire surface. Since this nitride film 7 becomes a protrusion to be described later, it has a structure in which it protrudes from the back side of the fixed electrode 5 by the depth of etching the surface of the sacrificial layer 4 (FIG. 3b).

To form the protrusion, at least a portion of the nitride film 7 connected to the side wall of the through hole 6 of the fixed electrode 5 is left as a protrusion 8, and the nitride film 7 at the bottom of the through hole 6 is etched (FIG. 3c).

Thereafter, the extraction electrodes 9 connected to the movable electrode 3 and the fixed electrode 5, respectively, are formed according to a general manufacturing process of a capacitive MEMS element. Thereafter, a part of the substrate 1 is removed from the back side of the substrate 1 until the thermal oxide film 2 is exposed, thereby forming a back chamber 10. Further, by removing a portion of the sacrificial layer 4 through the through hole 6 formed in the fixed electrode 5, an air gap 11 is formed between the movable electrode 3 and the fixed electrode 5. When the sacrificial layer 4 is removed, the nitride film 7 remains without being etched, so a protrusion 8 is formed on the surface of the fixed electrode 5 on the air gap 11 side to protrude toward the movable electrode 3 side, as shown in FIG. 3(d). can do.

Patent registration No. 6151541

The protrusion 8 previously proposed by the applicant of the present application is formed by processing the nitride film 7 within the through hole 6. Therefore, compared to the conventional manufacturing method of MEMS elements, the process of processing the nitride film 7, that is, the process of applying and patterning a photoresist on the nitride film 7 according to the usual photolithography method, and the process of etching and removing the nitride film. , it is necessary to add a step of removing the photoresist after that, resulting in an increase in manufacturing costs. An object of the present invention is to provide a MEMS element and a manufacturing method thereof that can form protrusions that prevent fixed electrodes and movable electrodes from sticking together while reducing such additional steps.

In order to achieve the above object, the invention according to claim 1 of the present application provides a MEMS device including a fixed electrode and a movable electrode arranged with an air gap formed by a spacer, in which at least one of the fixed electrode and the movable electrode is provided. A projection protruding toward the opposing movable electrode or the fixed electrode is provided on one surface on the air gap side, the spacer is made of polycrystalline silicon dioxide, and the fixed electrode and the movable electrode are made of polycrystalline silicon. , the protrusion is made of a silicon-rich compound formed by bonding silicon in the polycrystalline silicon and oxygen in the polycrystalline silicon dioxide .

The invention according to claim 2 of the present application is a method for manufacturing a MEMS device comprising a fixed electrode and a movable electrode arranged with an air gap formed by a spacer, the fixed electrode made of polycrystalline silicon, a sacrificial layer, and a movable electrode. a step of forming a laminated film of the movable electrode made of polycrystalline silicon, heating the laminated film, and protruding toward the opposing sacrificial layer side on each surface of the fixed electrode and the movable electrode in contact with the sacrificial layer; , a step of growing a protrusion made of an insulating compound in which a substance constituting the fixed electrode and the movable electrode is combined with a substance constituting the sacrificial layer; and a step of growing a protrusion made of an insulating compound in which the substance constituting the fixed electrode and the movable electrode is combined with the substance constituting the sacrificial layer, and the protrusion is left between the fixed electrode and the movable electrode. etching away a part of the sacrificial layer, forming an air gap with a spacer made of a part of the sacrificial layer that remains unetched, and arranging the fixed electrode and the movable electrode facing each other. In the method for manufacturing an element, the step of forming polycrystalline silicon dioxide as the sacrificial layer by a low pressure plasma CVD method using a gas containing tetraethoxysilane and oxygen as a raw material gas, and forming the polycrystalline silicon dioxide as the insulating protrusion. The method is characterized by including the step of growing a protrusion made of a silicon-rich compound in which silicon in the polycrystalline silicon dioxide and oxygen in the polycrystalline silicon dioxide are bonded.

The step of forming a protrusion according to the manufacturing method of the present invention requires only adding a step of heating these laminated films after forming a laminated film of a fixed electrode, a sacrificial layer, and a movable electrode. The number of additional steps can be significantly reduced, and manufacturing costs can be reduced.

In order to form protrusions according to the manufacturing method of the present invention, a sacrificial layer made of a different material from that of the conventional manufacturing method is used, but the materials used are those commonly used in the manufacturing process of semiconductor devices. It can be easily formed with good controllability.

By appropriately changing the heating conditions, the MEMS element formed by the manufacturing method of the present invention can be configured to have protrusions made of an insulating compound of a desired size with good controllability. It is possible to reliably prevent the fixed electrode and the movable electrode from sticking together when etching the sacrificial layer, and it is also possible to reliably prevent them from sticking together during use.

1 is a diagram illustrating a method for manufacturing a MEMS device according to a first embodiment of the present invention; FIG. It is a figure explaining the manufacturing method of the MEMS element of 2nd Example of this invention. It is a figure explaining the manufacturing method of the conventional MEMS element.

The present invention heat-treats a sacrificial layer, and a fixed electrode and a movable electrode laminated in contact with the sacrificial layer, so that the material constituting the sacrificial layer and the electrode The protrusion is formed by growing a compound with the substance that constitutes the protrusion. Examples of the present invention will be described in detail below.

Hereinafter, the MEMS device of the present invention according to the first embodiment will be explained according to its manufacturing process. First, a substrate 1 made of a silicon substrate is prepared, and a thermal oxide film 2 is formed on the surface of the substrate 1. Thereafter, a movable electrode 3 made of conductive polycrystalline silicon is laminated on the thermal oxide film 2.

Next, a sacrificial layer 4a is formed on the movable electrode 3. Here, as the sacrificial layer 4a, polycrystalline silicon dioxide is formed by a low pressure plasma CVD method using TEOS (tetraethoxysilane) and a gas containing oxygen such as oxygen, ozone, or nitrous oxide as raw material gases. Further, a fixed electrode 5 made of conductive polycrystalline silicon is laminated on the sacrificial layer 4a. A nitride film 7 is formed on the entire surface of the fixed electrode film 5.

In this manner, the sacrificial layer 4a is laminated on the movable electrode 3 and the fixed electrode 5 is laminated on the sacrificial layer 4a, in other words, the movable electrode 3 and the fixed electrode 5 are in contact with the sacrificial layer 4a, and then heated. Perform processing. This heat treatment is performed, for example, at a temperature of 800° C. or higher in a nitrogen atmosphere. Through this heat treatment, the surfaces of the movable electrode 3 and the fixed electrode 5 made of polycrystalline silicon that are in contact with the sacrificial layer 4a are made rich in silicon, in which silicon in the polycrystalline silicon and oxygen in the polycrystalline silicon dioxide are combined. A chemical compound (a compound of silicon and oxygen) is formed (Fig. 1a). This compound grows at the grain boundaries of polycrystalline silicon, and forms protrusions 12 that prevent fixed electrode 5 and movable electrode 3 from sticking together in the process of forming a spacer, which will be described later. Here, the size of the protrusion 12 can be controlled by controlling the heat treatment temperature and time. Furthermore, by controlling the film quality of polycrystalline silicon, the location where protrusions occur (crystal grain boundaries) can also be controlled. Therefore, if a normal semiconductor device manufacturing method is adopted, it will be a manufacturing method with good controllability.

Thereafter, a through hole 6 is formed in the nitride film 7 and the fixed electrode 5, and the sacrificial layer 4a is exposed within the through hole 6 (FIG. 1b), according to a general manufacturing process of a capacitive MEMS element.

After forming extraction electrodes 9 connected to the movable electrode 3 and the fixed electrode 5, a part of the substrate 1 is removed from the back side of the substrate 1 until the thermal oxide film 2 is exposed, and a back chamber 10 is formed. Further, by removing a portion of the sacrificial layer 4a through the through hole 6 formed in the fixed electrode 5, an air gap 11 is formed between the movable electrode 3 and the fixed electrode 5. When removing the sacrificial layer 4a, since the protrusions 12 are made of a silicon-rich compound, the etching rate is slower than that of normal polycrystalline silicon dioxide, making it possible to selectively remove the sacrificial layer 4a while leaving the protrusions 12. As a result, as shown in FIG. 1(c), protrusions 12 protruding toward the air gap 11 are formed on the surface of the movable electrode 3 on the air gap 11 side and on the surface of the fixed electrode 5 on the air gap 11 side, respectively. Can be done.

As explained above, according to the manufacturing method of this embodiment, the protrusion 12 is formed by forming the movable electrode 3 or the fixed electrode 5 made of polycrystalline silicon so as to be in contact with the sacrificial layer 4a made of polycrystalline silicon dioxide, and heating it. It will only be processed. This heat treatment does not require sheet-based treatment like the photolithography method described in the conventional method, and is a very simple method because it can be placed in a heating furnace, for example.

Next, a MEMS element according to a second embodiment will be explained according to its manufacturing process. In the first embodiment, the method of forming the protrusions 12 on both the movable electrode 3 and the fixed electrode 5 was described. However, in the present invention, it is possible to form the protrusion 12 only on either the movable electrode 3 or the fixed electrode 5. Hereinafter, a method for forming the protrusion 12 on the movable electrode 3 will be explained.

As in the above embodiment, first, a substrate 1 made of a silicon substrate is prepared, and a thermal oxide film 2 is formed on the surface of the substrate 1. Thereafter, a movable electrode 3 made of conductive polycrystalline silicon is laminated on the thermal oxide film 2.

Next, a sacrificial layer 4a is formed on the movable electrode 3. Here, polycrystalline silicon dioxide is formed as the sacrificial layer 4a by a low pressure plasma CVD method using TEOS (tetraethoxysilane) and a gas containing oxygen such as oxygen, ozone, or nitrous oxide as raw material gases.

Next, in this embodiment, a thin nitride film 7a is formed on the sacrificial layer 4a, a fixed electrode 5 made of conductive polycrystalline silicon is formed on this nitride film 7a, and a nitride film 7 is formed on the entire surface.

In this way, the sacrificial layer 4a is laminated on the movable electrode 3, and the fixed electrode 5 is laminated on the sacrificial layer 4a with the nitride film 7a in between, in other words, the movable electrode 3 is in contact with the sacrificial layer 4a, and is fixed. The heat treatment is performed while the electrode 5 is not in contact with the sacrificial layer 4a. This heat treatment is performed, for example, at a temperature of 800° C. or higher in a nitrogen atmosphere. By this heat treatment, a silicon-rich compound (a combination of silicon and oxygen) in which silicon in polycrystalline silicon and oxygen in silicon dioxide are combined is formed on the surface of movable electrode 3 made of polycrystalline silicon that is in contact with sacrificial layer 4a. compound) is formed (Fig. 2a). This compound grows at the grain boundaries of polycrystalline silicon, and forms protrusions 12 that prevent fixed electrode 5 and movable electrode 3 from sticking together in the process of forming a spacer, which will be described later. Here, the size of the protrusion 12 can be controlled by controlling the heat treatment temperature and time. Furthermore, by controlling the film quality of polycrystalline silicon, the location where protrusions occur (crystal grain boundaries) can also be controlled. Therefore, if a normal semiconductor device manufacturing process is adopted, a manufacturing method with good controllability can be obtained. In this embodiment, the fixed electrode 5 is not in contact with the sacrificial layer 4a, so no protrusion 12 is formed on the fixed electrode 5.

Thereafter, a through hole 6 is formed in the nitride film 7, the fixed electrode 5, and the nitride film 7a, and the sacrificial layer 4a is exposed in the through hole 6 (FIG. 2b) according to a general manufacturing process of a capacitive MEMS element.

After forming extraction electrodes 9 connected to the movable electrode 3 and the fixed electrode 5, a part of the substrate 1 is removed from the back side of the substrate 1 until the thermal oxide film 2 is exposed, and a back chamber 10 is formed. Further, by removing a portion of the sacrificial layer 4a through the through hole 6 formed in the fixed electrode 5, an air gap 11 is formed between the movable electrode 3 and the fixed electrode 5. When removing the sacrificial layer 4a, since the protrusions 12 are made of a silicon-rich compound, the etching rate is slower than that of normal polycrystalline silicon dioxide, making it possible to selectively remove the sacrificial layer 4a while leaving the protrusions 12. As a result, as shown in FIG. 2(c), a protrusion 12 protruding toward the air gap 11 can be formed on the surface of the movable electrode 3 on the air gap 11 side.

As explained above, according to the manufacturing method of this embodiment, the protrusion 12 is formed by simply forming the movable electrode 3 made of polycrystalline silicon so as to be in contact with the sacrificial layer 4a made of polycrystalline silicon dioxide, and then heat-treating the movable electrode 3 made of polycrystalline silicon. becomes. This heat treatment does not require sheet-based treatment like the photolithography method described in the conventional method, and is a very simple method because it can be placed in a heating furnace, for example. Furthermore, if the formation of the protrusion 12 is not necessary, it is sufficient to form a nitride film so that polycrystalline silicon and polycrystalline silicon dioxide do not come into direct contact with each other, which is a very simple method.

Next, a MEMS element according to a third example will be explained. In the second embodiment, the method of forming the protrusions 12 on the surface of the movable electrode 3 has been described. However, in the present invention, it is also possible to form the protrusion 12 on the fixed electrode 5 instead of on the movable electrode 3.

Unlike the second embodiment described above, a sacrificial layer 4a is laminated on the movable electrode 3 with a nitride film 7a interposed therebetween, and a fixed electrode 5 is laminated on the sacrificial layer 4a. Heat treatment is performed with the fixed electrode 5 in contact with the sacrificial layer 4a without contacting the layer 4a. This heat treatment is carried out under conditions of, for example, 800° C. or higher in a nitrogen atmosphere, as in the above embodiments. By this heat treatment, a silicon-rich compound (a combination of silicon and oxygen) in which silicon in polycrystalline silicon and oxygen in silicon dioxide are combined is formed on the surface of fixed electrode 5 made of polycrystalline silicon that is in contact with sacrificial layer 4a. compound) is formed. This compound grows at the grain boundaries of polycrystalline silicon and forms protrusions 12 that prevent fixed electrode 5 and movable electrode 3 from sticking together in the process of forming a spacer. Here, the size of the protrusion 12 can be controlled by controlling the heat treatment temperature and time. Furthermore, by controlling the film quality of polycrystalline silicon, the location where protrusions occur (crystal grain boundaries) can also be controlled. Therefore, if a normal semiconductor device manufacturing process is adopted, a manufacturing method with good controllability can be obtained. In this embodiment, the movable electrode 3 is not in contact with the sacrificial layer 4a, so no protrusion 12 is formed on the movable electrode 3.

The MEMS element of this example can be formed by the same manufacturing method as the first example above, except for the above manufacturing process.

As explained above, according to the manufacturing method of this embodiment, the protrusion 12 is formed by simply forming polycrystalline silicon dioxide on the movable electrode 3, further forming the fixed electrode 5, and performing heat treatment. This heat treatment does not require sheet-based treatment like the photolithography method described in the conventional method, and is a very simple method because it can be placed in a heating furnace, for example.

It goes without saying that the present invention is not limited to the above embodiments. For example, the protrusions formed on the movable electrode 3 and the fixed electrode 5 can be formed by depositing, growing, etc. a sacrificial layer and an insulating substance that can be selectively etched on the movable electrode 3 and the fixed electrode 5 using a desired combination of materials. If possible, it is not limited to silicon-rich compounds of silicon and oxygen.

1: Substrate, 2: Thermal oxide film, 3: Movable electrode, 4, 4a: Sacrificial layer, 5: Fixed electrode, 6: Through hole, 7, 7a: Nitride film, 8: Protrusion, 9: Extracting electrode, 10: Back chamber, 11: Air gap, 12: Protrusion

Claims (2)

In a MEMS element including a fixed electrode and a movable electrode arranged with an air gap formed by a spacer,
A projection protruding toward the opposing movable electrode or the fixed electrode is provided on the air gap side surface of at least one of the fixed electrode and the movable electrode,
the spacer is made of polycrystalline silicon dioxide;
The fixed electrode and the movable electrode are made of polycrystalline silicon,
A MEMS device characterized in that the protrusion is made of a silicon-rich compound formed by bonding silicon in the polycrystalline silicon and oxygen in the polycrystalline silicon dioxide. A method for manufacturing a MEMS device comprising a fixed electrode and a movable electrode arranged with an air gap formed by a spacer, the method comprising:
forming a laminated film of the fixed electrode made of polycrystalline silicon, a sacrificial layer, and the movable electrode made of polycrystalline silicon;
The laminated film is heated, and a material constituting the fixed electrode and the movable electrode and the sacrificial layer are added to the surfaces of the fixed electrode and the movable electrode in contact with the sacrificial layer, protruding toward the opposing sacrificial layer side. A process of growing protrusions made of an insulating compound combined with constituent substances;
A part of the sacrificial layer between the fixed electrode and the movable electrode is removed by etching, leaving the protrusion, and an air gap is formed by a spacer made of a part of the sacrificial layer that remains unetched, and the part of the sacrificial layer is removed between the fixed electrode and the fixed electrode. In a method for manufacturing a MEMS device, the method includes the step of arranging the movable electrodes to face each other,
forming polycrystalline silicon dioxide as the sacrificial layer by a low pressure plasma CVD method using a gas containing tetraethoxysilane and oxygen as a source gas;
Manufacturing a MEMS device, comprising the step of growing a protrusion made of a silicon-rich compound in which silicon in the polycrystalline silicon and oxygen in the polycrystalline silicon dioxide are combined as the insulating protrusion. Method.

Images