JP6356512B2 - MEMS element - Google Patents

Description

  The present invention relates to a MEMS element, and more particularly to a capacitive MEMS element used as a microphone, various sensors, switches, and the like.

  Conventionally, a MEMS (Micro Electro Mechanical Systems) element using a semiconductor process has a fixed electrode, a sacrificial layer (insulating film), and a movable electrode formed on a semiconductor substrate, and then a part of the sacrificial layer is removed to form a spacer. An air gap (hollow) structure is formed between the fixed electrode fixed via the movable electrode and the movable electrode.

  For example, in a condenser microphone that is a MEMS element, a fixed electrode having a plurality of through-holes that allow sound pressure to pass through and a movable electrode that vibrates in response to sound pressure are arranged facing each other, and vibrates in response to sound pressure. The displacement of the movable electrode is detected as a change in capacitance between the electrodes.

  Generally, in order to increase the sensitivity of a condenser microphone, it is necessary to increase the displacement of the movable electrode by sound pressure. Therefore, a method of softening the spring of the movable electrode and a method of increasing the volume of the back chamber are employed.

  FIG. 11 is an explanatory diagram of a conventional MEMS device in which the volume of the back chamber is increased. As shown in FIG. 11, a movable electrode 3 is formed on a silicon substrate 1 with a thermal oxide film 2 interposed. A fixed electrode 5 and a nitride film 6 are formed on the movable electrode 3 via a spacer 4, and a through hole 7 is formed in the fixed electrode 5.

  In the MEMS element shown in FIG. 11, the volume of the back chamber is increased by removing the silicon substrate 1 stepwise (Patent Document 1). Further, the volume of the back chamber can be further increased by increasing the number of stages.

  By the way, when forming such a step-like back chamber, it is necessary to repeat the formation of the etching mask and the etching process a plurality of times. A specific manufacturing process is shown in FIG. First, after forming the thermal oxide film 2, the movable electrode 3, the fixed electrode 5, etc. on the surface side of the silicon substrate 1, the first mask film 9 a used when etching the silicon substrate 1 on the back side of the silicon substrate 1. And a second mask film 9b. The first mask film 9a and the second mask film 9b are a combination of films that can be selectively removed. Specifically, the first mask film 9a is a metal film, and the second mask film 9b is a photoresist. Next, using the second mask film 9b as an etching mask, the exposed surface of the silicon substrate 1 is anisotropically etched by the RIE method. As a result, the first recess 10a is formed as shown in FIG.

  The second mask film 9b is selectively removed to expose the first mask film 9a. The first mask film 9a is patterned so as to open at a position recessed from the second mask film 9a, and a part of the back surface side of the silicon substrate 1 remaining without being etched is exposed ( FIG. 12b).

  Thereafter, using the first mask film 9a as an etching mask, the exposed surface of the silicon substrate 1 is anisotropically etched by RIE to expose the thermal oxide film 2. As a result, as shown in FIG. 12C, a first recess 10a and a second recess 10b are formed, thereby forming a stepped back chamber. Here, when a back chamber having a larger number of stages is formed, the formation of the etching mask and the etching are repeated.

  Thereafter, by removing a part of the spacer film 4a from the through hole 7, the MEMS element shown in FIG. 11 can be formed.

JP 2008-17395 A

  By the way, the conventionally proposed method has a problem that the etching depth is difficult to control because the etching with different opening diameters proceeds simultaneously when the back chamber is formed. Further, in the second etching, there is a problem that the polymer deposited on the side wall of the first recess 10a during the first etching remains without being removed. In order to solve such a problem, an object of the present invention is to provide a MEMS element that can easily increase the volume of a back chamber.

In order to achieve the above object, according to the first aspect of the present invention, an air gap is formed by arranging a substrate having a back chamber and a fixed electrode and a movable electrode on the substrate with a spacer interposed therebetween. In the MEMS element, the movable electrode is sandwiched between an insulating member and the spacer, and the insulating member has one end sandwiched between the substrate and the spacer, and the other end from the end of the substrate to the back chamber. And the movable electrode is sandwiched between the other end of the insulating member and the spacer , a circular or polygonal ring-shaped insulating member, or a cylindrical or prismatic shape inside the ring shape. It is characterized by comprising an insulating member where a part of the substrate remains .

  The MEMS element of the present invention has a configuration in which the movable electrode is sandwiched between the insulating members, so that the substrate can be removed more widely than before, and its volume becomes larger than a general back chamber, and the sensitivity of the MEMS element is improved. Can be achieved.

  In addition, the MEMS element of the present invention can be configured to sandwich the movable electrode as a structure in which a part of the plurality of divided substrates remains, and stress relaxation between the insulating member and the movable electrode in the sandwiching portion is achieved. It becomes possible.

  Furthermore, in the MEMS element of the present invention, the process of forming the insulating member only needs to add the process of forming the groove part and the process of flattening the groove part to the manufacturing process of the normal MEMS element, and the controllability is very good. Further, since it is only necessary to process the substrate surface, the method is very simple and has good controllability.

It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the manufacturing process of the MEMS element of this invention. It is a figure explaining the MEMS element of this invention. It is a figure explaining another MEMS element of the present invention. It is a figure explaining the structure of the insulation member of another MEMS element of this invention. It is explanatory drawing of this kind of conventional MEMS element. It is a figure explaining the manufacturing process of this kind of conventional MEMS element.

  Since the MEMS element according to the present invention supports the movable electrode, instead of sandwiching the end of the movable electrode between the substrate and the spacer, the insulating member is sandwiched between the spacer and the insulating member is sandwiched between the substrate and the spacer. By adopting such a structure, the substrate is retracted from the conventional structure, and the volume of the back chamber is increased. Hereinafter, examples will be described in detail.

The first embodiment of the present invention will be described according to the manufacturing process, taking as an example the case where the insulating member for holding the movable electrode is formed in a circular or polygonal ring shape. First, a thermal oxide film 2 (SiO ₂ ) having a thickness of about 1 μm is formed on a silicon substrate 1 having a crystal orientation (100) plane of 420 μm, and a part of the thermal oxide film 2 is formed by a normal photolithography method. Etching is removed, and the exposed surface of the silicon substrate 1 is removed by etching to form a ring-shaped groove 11 (FIG. 1). The groove portion 11 is filled with a desired material to form an insulating member that sandwiches a movable electrode described later. Therefore, when the shape is a ring shape, a movable electrode is stacked on the inner end of the ring, and the silicon substrate remains on the outer end of the ring after the back chamber is formed. It is necessary to have a shape that is intruded. Therefore, if these conditions are satisfied, the shape is not limited to a circle but can be a polygon.

  Next, the silicon substrate 1 exposed on the surface of the groove 11 is thermally oxidized, and the surface is covered with the thermal oxide film 2 (FIG. 2). The thermal oxide film 2 serves as an etching stopper when a back chamber is formed as described later.

  A nitride film 12 having a thickness of 0.2 μm and a polysilicon film 13 are laminated and planarized on the entire surface by LPCVD (Low Pressure Chemical Vapor Deposition) method (FIG. 3). The nitride film 12 was formed to serve as an etching stopper when the spacer 4 described later is formed, and the polysilicon film 13 was selected to fill and flatten the relatively deep groove 11. The polysilicon film 13 becomes an insulating material when non-doped silicon is used. However, even if doped polysilicon is used, any structure can be used as long as the surface can be insulated and separated from the movable electrode as described later. No problem.

  Thereafter, at least the polysilicon film 13 on the surface of the silicon substrate 1 is removed by etching back, and the trench 11 is filled with the polysilicon film 13. Here, when the nitride film 12 remains on the surface other than the groove, the nitride film 12 in the region surrounded by the groove and where the movable electrode is to be formed is removed, and the thermal oxide film 2 is exposed. Thereafter, the surface of the polysilicon film 13 is thermally oxidized for insulation, and the surface of the polysilicon film 13 filled in the groove 11 is covered with a thermal oxide film 14 (FIG. 4). The reason why the thermal oxide film 14 is formed here is to make the surface of the material filled in the groove 11 insulative. Therefore, when the polysilicon film 13 is an insulating polysilicon or when another insulating material is used, it is not necessary to form the thermal oxide film 14.

  A conductive polysilicon film having a thickness of 0.2 to 1.0 μm is laminated and formed on the entire surface by a CVD (Chemical Vapor Deposition) method. Next, patterning is performed by a normal photolithography method to form the movable electrode 3 (FIG. 5).

  Further, a sacrificial layer 4a made of a USG (Undoped Silicate Glass) film having a thickness of about 2.0 to 4.0 μm is stacked on the entire surface, and further, a thickness of about 0.1 to 1.0 μm is formed on the sacrificial layer 4a. A conductive polysilicon film is stacked. Next, patterning is performed by a normal photolithography method to form the fixed electrode 5, and the nitride film 6 is laminated on the fixed electrode 5. Thereafter, a part of the nitride film 6 and the fixed electrode 5 is removed by etching to form a through hole 7, and a part of the sacrificial layer 4a is exposed (FIG. 6).

  Thereafter, the silicon substrate 1 is removed from the back surface side of the silicon substrate 1 until the thermal oxide film 2 is exposed, and a back chamber 8 is formed (FIG. 7). The thermal oxide film 2 serves as an etching stopper. Here, the silicon substrate 1 to be removed can be removed beyond the size of the movable electrode 5. That is, conventionally, since the silicon substrate 1 had to support the movable electrode 5, the silicon substrate 1 located at the end of the movable electrode 5 had to remain. On the other hand, in the present invention, since the end of the movable electrode 5 is supported by the polysilicon film 13 or the like filled in the groove 11, it can be largely removed so that the bottom of the groove 11 is exposed. It was.

  Thereafter, a part of the sacrificial layer 4a is removed through the through hole 7 to form the spacer 4, whereby the movable electrode 3 and the fixed electrode 5 are opposed to each other. Simultaneously with the removal of the sacrificial layer 4a, a part of the exposed thermal oxide film 2 in the back chamber 8 is also removed to complete the MEMS element (FIG. 8).

  As shown in FIG. 8, in the MEMS element of the present invention, the movable electrode 5 is composed of a thermal oxide film 14, a polysilicon film 13 and a nitride film 12 (these are called insulating members) filled in the groove 11 and a spacer 4. It is sandwiched and supported. Further, since the insulating member is sandwiched and supported by the silicon substrate 1 and the spacer 4 that form the side wall of the back chamber, the mechanical strength can be sufficiently maintained.

  As described above, according to the present invention, the structure is such that the movable electrode is sandwiched and supported by the insulating member protruding into the back chamber, so that the silicon substrate can be removed larger than before, and the volume of the back chamber is increased. The sensitivity of the MEMS element can be improved.

  Next, a second embodiment will be described. In the first embodiment, the structure in which a circular or polygonal ring-shaped groove is formed and filled with a polysilicon film or the like when the insulating member is formed has been described. However, the present invention is not limited to this. Is not to be done. For example, as shown in FIG. 9, it is also possible to make the groove part 11 into a multiple structure and leave a part of the silicon substrate 1 between them. Here, when the size of the adjacent groove portion 11 becomes small, the silicon substrate 1 has silicon between the groove portions 11 during the etching for forming the back chamber 8 even in the same manufacturing process as in the first embodiment. The etching rate of the substrate 1 becomes slow and remains without being etched. Therefore, there is no problem even if the silicon substrate 1 has a shape that does not remain as long as sufficient strength can be maintained as in a groove portion of a honeycomb structure described later.

  Examples of the groove structure having a multiple structure include, for example, a method in which circular ring-shaped groove portions 11 are stacked as shown in FIG. 10 (a), and a groove shape 11 in a honeycomb shape as shown in FIG. 10 (b). It can be formed by a method of forming (a plurality of hexagonal substrates remain). Furthermore, instead of the hexagonal aggregate as shown in FIG. 10 (b), a structure in which a plurality of circular substrates remain, a structure in which a plurality of other polygonal substrates remain, or a regular assembly is not possible. It can also be a regular set.

  In any case, it is needless to say that the movable electrode can be sandwiched and has a shape that does not hinder the displacement of the movable electrode. In addition, the shape of the groove part shown in FIG. 10 is shown typically, and the cross-sectional shape and dimension of the MEMS element shown in FIG. 9 do not match.

1: silicon substrate, 2: thermal oxide film, 3: movable electrode, 4a: sacrificial layer, 4: spacer, 5: fixed electrode, 6: nitride film, 7: through-hole, 8: back chamber, 9: first Mask film, 10: second mask film, 11: groove, 12: nitride film, 13: polysilicon film, 14: thermal oxide film

Claims (1)

In a MEMS element in which an air gap is formed by arranging a substrate having a back chamber and a fixed electrode and a movable electrode on the substrate with a spacer interposed therebetween,
The movable electrode is sandwiched between an insulating member and the spacer,
One end of the insulating member is sandwiched between the substrate and the spacer, the other end projects from the end of the substrate into the back chamber, and the movable electrode is sandwiched between the other end of the insulating member and the spacer. And an insulating member having a circular or polygonal ring shape, or an insulating member in which a part of the columnar or prismatic substrate remains inside the ring shape .

Images