JP2014213420A - Method of manufacturing micro device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method of manufacturing a micro device which is excellent in flatness and can improve yields and reliability by suppressing both of generation of a residue on a substrate and fixation between a movable structure and the substrate.SOLUTION: A sacrifice insulation film 4 formed of oxide containing at least phosphorous and having a through hole, a column part 5f embedded in the through hole 6, and a conductor film 5a for a movable part positioned on the sacrifice insulation film 4 and the column part 5f are formed on a substrate 1A. The sacrifice insulation film 4 is removed by liquid phase treatment, and the conductor film 5a for a movable part is supported by the column part 5f to the substrate 1A. The column part 5f is removed by gas phase treatment. The conductor film 5a for a movable part is floated from the substrate 1A in a region where the column part 5f is formed.

Description

  The present invention relates to a method for manufacturing a micro device, and more particularly to a method for manufacturing a micro device such as an acceleration sensor, an angular velocity sensor, or a micro mirror.

  As a method for detecting the inertial force in the direction perpendicular to the substrate, there is a method for detecting a change in electrostatic force accompanying the inertial force. An example of an acceleration sensor by this method is described in Patent Document 1 and Non-Patent Document 1, for example. In this acceleration sensor, a detection frame is rotatably supported by the substrate via a torsion beam, and an inertial mass body is supported by the detection frame via a link beam so as to be displaceable in the thickness direction of the substrate so as to face the detection frame. A detection electrode is formed on the substrate. In this acceleration sensor, when acceleration is applied in a direction perpendicular to the substrate, the inertial mass body is displaced in the direction perpendicular to the substrate. The displacement of the inertial mass body is transmitted to the detection frame via the link beam, so that the detection frame rotates about the torsion beam. By this rotation, the distance between the detection frame and the detection electrode changes, and the capacitance between the detection frame and the detection electrode changes. The acceleration is detected by converting the capacitance into a voltage proportional to the acceleration by the capacitance-voltage conversion circuit.

  The method for manufacturing the acceleration sensor is described in FIGS. 8 to 12 of Patent Document 1 and FIG. 6 of Non-Patent Document 1. After the sacrificial layer film is deposited on the fixed electrode on the substrate and the movable structure (conductive film for the movable part) made of polycrystalline silicon is formed on the sacrificial layer film, the sacrificial layer film is selectively removed and moved. The structure can be displaced.

  In such a micro device that is displaced in the out-of-plane direction, the smaller the gap between the movable part and the substrate, the higher the sensitivity, and the performance tends to be improved, such as a broad band. In the case of forming a structure having a small gap, for example, in Patent Document 2, after forming a basic structure of a structure having a movable electrode and a fixed electrode in order to form a lateral inertial force detection structure, the movable electrode and the fixed electrode are fixed. A method is described in which an electric current is passed through a fixing member that fixes an electrode, and the fixing member is melted and manufactured.

  However, if the gap is reduced, the electrode of the movable part and the electrode of the substrate are liable to stick, which may lead to a decrease in yield and reliability. For this reason, it is necessary to prevent the movable part from adhering to the substrate. For example, Patent Document 3 describes a method for preventing sticking by etching a sacrificial layer film with vapor-phase hydrofluoric acid.

International Publication No. 2009/125510 Japanese Patent No. 3966155 Japanese Patent No. 3577766

Hirata et al., “Improvement of collision speed detection by mechanical displacement conversion type Z-axis acceleration sensor using damping effect”, IEEJ Transactions E, 2012, Vo.132, No.9, pp.296-302 W. I. Jang et al., "Fabrication of MEMS devices by using anhydrous HF gas-phase etching with alcoholic vapor", J. Micromech. Microeng., Vol.12 (2002), pp.297-306

  It is known that the sacrificial layer film formed between the electrode of the movable part and the electrode of the substrate is selectively removed by the vapor hydrofluoric acid described in Patent Document 3 as described above. On the other hand, in the microsensor manufacturing method using polycrystalline silicon as a movable structure disclosed in Patent Document 1 and Non-Patent Document 1, as a sacrificial layer film, phosphate glass (which has excellent flatness by high-temperature reflow after film formation) Phosphorus Silicon Glass (PSG) is used. However, when PSG is selectively removed by dry etching using vapor-phase hydrofluoric acid, residues are generated on the substrate due to the reaction with hydrofluoric acid and cannot be removed, as shown in FIG. 6). Therefore, when a PSG sacrificial layer film is used, removal by vapor hydrofluoric acid cannot be used, and removal by liquid hydrofluoric acid needs to be used. Removal by liquid hydrofluoric acid requires water washing after hydrofluoric acid etching, and there is a problem that the movable structure and the substrate are fixed in the drying step after water washing, resulting in a decrease in yield.

  The present invention has been made in view of the above-described problems, and its object is to easily reduce the gap between the substrate and the movable structure, and to generate residues on the substrate, and the movable structure and It is an object of the present invention to provide a method of manufacturing a microdevice that can suppress both the fixation of a substrate.

The manufacturing method of the microdevice of the present invention includes the following steps.
A sacrificial insulating film made of an oxide containing at least phosphorus and having a through-hole, a post portion embedded in the through-hole, and a sacrificial insulating film and a movable portion conductor film located on the post portion are formed on the substrate. Is done. The sacrificial insulating film is removed by a liquid phase process, and the conductive film for the movable part is supported by the support column with respect to the substrate. The support portion is removed by the vapor phase process, and the movable portion conductive film is lifted from the substrate in the region where the support portion is formed.

  According to the present invention, since the oxide film containing phosphorus having excellent flatness is used as the sacrificial insulating film, it is easy to narrow the gap between the substrate and the movable structure. In addition, since the sacrificial insulating film is removed by the liquid phase treatment, generation of residues can be suppressed. In addition, the conductive film for the movable part is supported by the support during the liquid phase process, and the support is removed by the vapor phase process after the liquid process, thereby preventing adhesion between the conductive film for the movable part and the substrate. be able to.

It is a top view which shows roughly the structure of the acceleration sensor which is a micro device in Embodiment 1 of this invention. It is a schematic sectional drawing in alignment with the II-II line of FIG. It is a schematic sectional drawing in alignment with the III-III line of FIG. FIG. 3 is a schematic cross-sectional view corresponding to regions A and B of FIG. 2 showing a first step of the method for manufacturing a microdevice in the first embodiment of the present invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 2nd process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 3rd process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 4th process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 5th process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 6th process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 7th process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 8th process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a schematic sectional drawing corresponding to the area | regions A and B of FIG. 2 which shows the 9th process of the manufacturing method of the micro device in Embodiment 1 of this invention. It is a top view corresponding to the process of FIG. 9 which shows the planar position of a support | pillar part. It is an expansion schematic sectional drawing of the area | region corresponded to the area | region XIV enclosed with the dotted line of FIG. 11, which shows the aspect by which the support | pillar part in Embodiment 2 of this invention is thermally oxidized. FIG. 13 is a schematic cross-sectional view corresponding to regions A and B in FIG. 2 showing a process corresponding to FIG. 12 in the microdevice manufacturing method according to the second embodiment of the present invention. FIG. 8 is a schematic cross-sectional view corresponding to a region A in FIG. 2 showing a process corresponding to FIG. 7 in the method for manufacturing a microdevice in the third embodiment of the present invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the process corresponding to FIG. 7 of the manufacturing method of the micro device in Embodiment 4 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 1st process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 2nd process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 3rd process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 4th process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 5th process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 6th process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 7th process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 8th process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a schematic sectional drawing corresponding to the area | region A of FIG. 2 which shows the 9th process of the manufacturing method of the micro device in Embodiment 5 of this invention. It is a top view which shows roughly the structure of the angular velocity sensor which is the other example of the microdevice of this invention. It is a schematic sectional drawing which follows the XXVIII-XXVIII line | wire of FIG. It is a schematic sectional drawing in alignment with the XXIX-XXIX line | wire of FIG. It is a top view which shows roughly the structure of the micromirror which is the other example of the microdevice of this invention. It is a schematic sectional drawing which follows the XXXI-XXXI line | wire of FIG.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(Embodiment 1)
First, the configuration of the acceleration sensor will be described with reference to FIGS. 1 to 3 as the configuration of the microdevice of the present embodiment. For convenience of explanation, an X direction, a Y direction, and a Z direction are introduced. In FIG. 1, the X direction is the direction in which the two link beams 5b extend. The Y direction is a direction orthogonal to the X direction, and is a direction in which the eight holes 5e are arranged. The Z direction is a direction orthogonal to both the X direction and the Y direction, and is a vertical direction orthogonal to the surface of the support substrate 1 (a direction in which the substrate 1A and the movable body face each other). Note that the Z direction coincides with the acceleration direction to be measured by the acceleration sensor of the present embodiment.

  Referring to FIGS. 1 to 3, acceleration sensor 100 of the present embodiment mainly includes substrate 1A, detection frame 5a, link beam 5b, and support portion 5c.

  The substrate 1A has a support substrate 1, an insulating film 2, and a fixed electrode 3. An insulating film 2 is formed on the support substrate 1, and a fixed electrode 3 is formed on the insulating film 2. For example, a silicon substrate can be used as the support substrate 1. As the insulating film 2, for example, a low stress silicon nitride film or silicon oxide film can be used. For the fixed electrode 3, for example, a polycrystalline silicon film having conductivity can be used. The fixed electrode 3 is formed in a region overlapping with a region of a part of the detection frame 5a (about 3/4 on the upper side of the detection frame 5a in FIG. 1) in plan view, and has a rectangular shape.

  The detection frame 5a (movable part conductive film) has a rectangular shape in plan view. The detection frame 5a is disposed on the opposite side (upper side) of the fixed electrode 3 from the substrate 1A so as to face the fixed electrode 3 in the Z direction. The fixed electrode 3 and the detection frame 5a immediately above in the Z direction are arranged to be spaced from each other in the Z direction, in other words, with a space between the fixed electrode 3 and the detection frame 5a. A plurality of holes 5e are formed in the detection frame 5a so as to be arranged in a matrix at intervals in the X direction and the Y direction (for example, in FIG. 1, four rows in the X direction and eight columns in the Y direction). . The hole 5e penetrates in the Z direction from one main surface of the detection frame 5a to the other main surface facing the one main surface in the Z direction. In FIG. 1, the hole 5e has a rectangular shape in plan view, but is not limited thereto, and may be, for example, a polygonal shape, a circular shape, or a line shape.

  The link beam (beam conductor film) 5b is connected to two places outside the rectangular shape of the detection frame 5a. The link beam 5b is supported by the detection frame 5a so that it can be twisted around an axis extending in the X direction of the link beam 5b. The axis extending in the X direction of the link beam 5b is shifted by a distance e with respect to the center of gravity G of the detection frame 5a and the Y direction.

  The support part (support part conductive film) 5c has a frame shape so as to surround the periphery of the detection frame 5a with a gap 5d in plan view. The support portion 5c is fixedly supported on the substrate 1A around the detection frame 5a. The support portion 5c is connected to the link beam 5b. Thereby, the support part 5c is connected to the detection frame 5a via the link beam 5b. As described above, the detection frame 5a is fixedly supported on the substrate 1A via the link beam 5b and the support portion 5c. Further, at least a part of the support portion 5c has conductivity.

  The detection frame 5a, the link beam 5b, and the support portion 5c are formed of an integral conductor film 5, and are made of, for example, a conductive polycrystalline silicon film. In the present embodiment, the detection frame 5a is a movable portion conductive film that moves relative to the substrate 1A.

  Next, the operation principle of the acceleration sensor as the micro device of this embodiment will be described with reference to FIGS.

  Referring to FIGS. 1 to 3, in this acceleration sensor 100, when acceleration is applied in a direction perpendicular to the surface of substrate 1A (Z direction), detection frame 5a that also functions as an inertial mass body is formed on substrate 1A. Is displaced in the vertical direction (Z direction). Due to this displacement, the detection frame 5a rotates around the axis of the link beam 5b extending in the X direction. By this rotation, the distance between the detection frame 5a and the fixed electrode 3 changes, and the electrostatic capacitance between the detection frame 5a and the fixed electrode 3 changes. The acceleration is detected by converting the capacitance into a voltage proportional to the acceleration by the capacitance-voltage conversion circuit.

  Next, a method for manufacturing an acceleration sensor as a method for manufacturing a microdevice according to the present embodiment will be described with reference to FIGS.

  Referring to FIG. 4, conductor film 3 made of, for example, polycrystalline silicon is formed on the surface of support substrate 1 made of, for example, silicon, with insulating film 2 interposed. The insulating film 2 and the conductor film 3 are preferably formed by, for example, a CVD (Chemical Vapor Deposition) method. The conductive film 3 is patterned using a photolithography technique and an etching technique, whereby the fixed electrode 3 is formed. As a result, a substrate 1A composed of the support substrate 1, the insulating film 2, and the patterned fixed electrode 3 is formed.

  Next, a first protective insulating film 4a, a sacrificial insulating film 4, and a second protective insulating film 4b are formed in this order on the substrate 1A so as to cover the fixed electrode 3, for example, by the CVD method. The first protective insulating film 4a and the second protective insulating film 4b are preferably formed of a material that is not easily thermally oxidized, and is preferably made of, for example, silicon nitride. The sacrificial insulating film 4 is preferably formed of a material that is excellent in coverage and can flatten the surface by reducing the level difference of the surface covered with the sacrificial insulating film 4 by a so-called reflow process. The insulating film 4 is preferably made of an oxide containing at least phosphorus, such as phosphate glass (PSG). However, the sacrificial insulating film 4 may be formed of, for example, glass (Boron Phosphorus Silicon Glass: BPSG) in which boron is added to phosphate glass instead of phosphate glass (PSG).

  The first protective insulating film 4a, the sacrificial insulating film 4 and the second protective insulating film 4b are preferably formed as thin as possible. For example, the thickness T1 (FIG. 4) of the first protective insulating film 4a and the thickness T3 (FIG. 4) of the second protective insulating film 4b are each 500 nm, and the thickness T2 (FIG. 4) of the sacrificial insulating film 4 is 1 μm. .

  After the first protective insulating film 4a and the sacrificial insulating film 4 are formed, it is preferably heated to about 1000 ° C. by a reflow process. In this way, the surface roughness of the upper surface of the sacrificial insulating film 4 can be made to have a very excellent shape such as, for example, 1 nm or less. As a result, the flatness of the upper surface of the second protective insulating film 4b can also be improved.

  Referring to FIG. 5, first protective insulating film 4a, sacrificial insulating film 4 and second protective insulating film 4b are patterned using a photolithography technique and an etching technique. Accordingly, the region A in FIG. 2, that is, the region where the detection frame 5a is to be formed, penetrates the first protective insulating film 4a, the sacrificial insulating film 4 and the second protective insulating film 4b to reach the fixed electrode 3. A plurality of holes 6 are formed at regular intervals. This interval is preferably substantially equal to the pitch of the holes 5e formed so as to be arranged in a plurality in the detection frame 5a formed later. In the region B of FIG. 2, that is, the region where the support portion 5c (FIGS. 1 to 3) is to be formed, the insulating film 2 of the substrate 1A is exposed.

  Referring to FIG. 6, conductor film 5 made of, for example, polycrystalline silicon is formed by, for example, a CVD method so as to cover second protective insulating film 4b. The conductor film 5 is formed so as to fill the through hole 6. The conductor film 5 filled in the through-hole 6 directly below the detection frame 5a is formed as a support portion 5f. The conductor film 5 is formed to have a thickness T4 (FIG. 6) of about 8 μm, for example.

  Referring to FIG. 7, the conductive film 5 is patterned by performing deep reactive ion etching (DRIE) or the like on the conductive film 5. At this time, the second protective insulating film 4b functions as an etching stopper.

  The detection frame 5a and the support portion 5c are formed from the conductor film 5 by the above-described deep etching. Although not shown, a link beam 5b (FIGS. 1 and 3) for connecting the detection frame 5a and the support portion 5c is also formed. Only by performing this one-time deep etching, the detection frame 5a as the conductive film for the movable portion and the detection frame 5a are supported from the integral conductive film 5 on the sacrificial insulating film 4 and the substrate 1A. The support portion 5c as the support portion conductor film can be formed at the same time.

  The column portion 5f is formed so as to fill the inside of the through hole 6 extending in the Z direction with the second protective insulating film 4b, the sacrificial insulating film 4 and the first protective insulating film 4a from the lower surface of the detection frame 5a to the upper surface of the fixed electrode 3. Has been. In this way, it is possible to form the column portion 5f for supporting the detection frame 5a with respect to the substrate 1A so that the detection frame 5a does not adhere to the substrate 1A when the sacrificial insulating film 4 is removed as will be described later.

  A plurality of holes 5e are formed in the detection frame 5a at regular intervals so as to penetrate from the upper surface to the lower surface. The detection frame 5a is connected to the fixed electrode 3 via a support portion 5f directly below the detection frame 5a.

  Referring to FIG. 8, thin film 7 as a third protective insulating film is formed so as to cover the surface (upper surface and side surface) of conductive film 5 such as detection frame 5a and support portion 5c. The thin film 7 is formed of a silicon nitride film or the like that prevents thermal oxidation of polycrystalline silicon when the conductor film 5 is made of, for example, polycrystalline silicon. The thin film 7 is formed in anticipation of the pattern change of the conductor film 5 due to oxidation, but is unnecessary when the pattern change of the conductor film 5 due to oxidation can be ignored.

  After the thin film 7 is formed, the second protective insulating film 4b immediately below the hole 5e is removed by fluorocarbon-based reactive ion etching (RIE) or the like to form the opening 5g.

  Referring to FIG. 9, sacrificial insulating film 4 is etched away by a liquid phase treatment, that is, a hydrofluoric acid solution, starting from the region where sacrificial insulating film 4 is exposed by opening 5g. By the etching removal described above, a gap is generated between the upper surface of the substrate 1A and the lower surface of the conductive film 5 such as the detection frame 5a and the link beam 5b (not shown). Further, by removing the sacrificial insulating film 4, the side surface of the column portion 5f is exposed to the outside. However, since the fixed electrode 3 and the detection frame 5a are connected by the support column 5f, the detection frame 5a is fixed to the substrate 1A by the support column 5f at this stage.

  The plurality of support columns 5f that support the detection frame 5a are arranged, for example, in a matrix in a plan view as shown in FIG. Each of the plurality of support columns 5f arranged in a matrix is formed at an intermediate point between a pair of holes 5e adjacent in the diagonal direction among the plurality of holes 5e arranged in a matrix of the detection frame 5a. However, it is not limited to this. Moreover, although the support | pillar part 5f becomes a rectangular shape in planar view in FIG. 13, not only this but polygonal shape, circular shape, and line shape may be sufficient, for example.

  As described above, as shown in FIG. 9, the movable portion conductor film (detection frame 5a) disposed between the substrate 1A and the substrate 1A is supported on the substrate 1A by each of the plurality of support columns 5f. A support structure is formed.

  Referring to FIG. 10, thermal oxidation is performed thereafter. Thereby, the column part 5f exposed to the outside is thermally oxidized from the surface (side surface), and the thermal oxide film 8 is formed. When the support 5f, that is, the conductor film 5 is made of polycrystalline silicon, the thermal oxide film 8 is made of silicon oxide.

The thermal oxidation is performed by chemically altering silicon by flowing a gas containing oxygen at a high temperature of 900 ° C. or higher, for example. Examples of the thermal oxidation method include dry O ₂ oxidation, wet O ₂ oxidation, steam oxidation, hydrogen combustion oxidation (pyrogenic oxidation), oxygen partial pressure oxidation, and hydrochloric acid oxidation.

  Referring to FIG. 11, this thermal oxide film 8 grows at a constant rate on the surface side and in-film side of support column 5f. The thermal oxidation is performed so that the entire width direction (Y direction) perpendicular to the vertical direction of the support column 5f is oxidized in at least a part of the support column 5f in the vertical direction (Z direction). In other words, thermal oxidation is performed until the entire inside of the column portion 5f is transformed into the thermal oxide film 8 and the detection frame 5a and the fixed electrode 3 opposed thereto are supported only by the thermal oxide film 8.

  The thermal oxide film 8 generated by the alteration of the column portion 5f is removed by etching with, for example, vapor-phase hydrofluoric acid. Further, the first protective insulating film 4a, the second protective insulating film 4b, and the third protective insulating film 7 are etched away by, for example, plasma etching using a fluorocarbon gas.

  Referring to FIG. 12, by removing the thermal oxide film 8 by etching, the movable portion conductor film (detection frame 5a) is lifted from the substrate 1A at the position where the support column portion 5f is formed. In this state, the movable portion conductive film (detection frame 5a) is supported with respect to the substrate 1A through the link beam 5b and the support portion 5c (see FIG. 3).

  Further, by removing the thin film 7 by etching, the surfaces of the fixed electrode 3 and the conductor film 5 (the detection frame 5a and the support portion 5c) are exposed, and the movable portion conductor film can be displaced in the Z direction. Thereby, the acceleration sensor as the micro device of the present embodiment shown in FIGS. 1 to 3 is manufactured.

  When the first protective insulating film 4a, the second protective insulating film 4b, and the third protective insulating film 7 are removed by etching with a chemical solution having a high etching selectivity with respect to an oxide film such as hot phosphoric acid, the gas phase It is preferable that the chemical solution is etched away before the thermal oxide film 8 is removed by hydrofluoric acid. In other words, it is preferable to remove the thermal oxide film 8 by vapor-phase hydrofluoric acid after etching with a chemical solution (for example, as the last step).

  In this case, the remaining chemical solution that may be generated when the removal of the first protective insulating films 4a, 4b and the like using the chemical solution is performed after the etching removal by the gas phase (for example, as the last step). For example, a problem that the lower surface of the detection frame 5a and the upper surface of the fixed electrode 3 are fixed to each other due to the surface tension can be suppressed.

Next, the effect of this Embodiment is demonstrated.
According to the present embodiment, the sacrificial insulating film 4 made of an oxide containing at least phosphorus such as phosphate glass (PSG) is used as the conductive film (polycrystalline silicon thin film) between the fixed electrode 3 and the detection frame 5a. It is formed so as to be sandwiched between. The upper surface of the sacrificial insulating film 4 can be planarized by performing a high temperature reflow process.

  For example, the surface roughness of a fixed electrode thin film having a thickness of 500 nm formed of polycrystalline silicon is usually about several nm. However, when phosphate glass (PSG) is formed as a sacrificial insulating film on the fixed electrode thin film and a high temperature reflow process is performed at about 1000 ° C., the surface roughness of the upper surface of the phosphate glass is set to 1 nm or less. be able to. This is due to the excellent step coverage (function to cover and flatten the step) and flattening of phosphate glass. As a result, the gap between the fixed electrode 3 and the detection frame 5a can be narrowed, and a microdevice (capacitance acceleration sensor) having high sensitivity and a wide band (with improved performance) can be realized.

  Further, the sacrificial insulating film 4 made of, for example, phosphate glass is removed by liquid phase processing. For this reason, for example, the residue containing phosphorus, which is generated when the sacrificial insulating film 4 made of phosphate glass is removed in the gas phase, is not generated by the removal by the liquid phase treatment. Therefore, the yield of products can be improved.

  In addition, the sacrificial insulating film 4 is removed in a state where the detection frame 5a is supported by the column portion 5f formed so as to be embedded in the through-hole 6 of the sacrificial insulating film 4 with respect to the fixed electrode 3 of the substrate 1A. Therefore, even if the sacrificial insulating film 4 is removed by the liquid phase, the fixed electrode 3 and the detection frame 5a are not fixed in the drying process.

  Moreover, the support | pillar part 5f is removed by a gaseous phase. For this reason, the drying process like the case where it removes by a liquid phase becomes unnecessary. Therefore, the sticking in the drying process does not occur at the time of removing the column portion 5f.

  As described above, since the supporting column 5f is removed not by the liquid phase process but by the vapor phase process in the final process for floating the detection frame 5a from the substrate 1A, the detection frame 5a and the fixed electrode 3 are not separated during the chemical liquid drying process. Sticking can be suppressed.

  The surface of the sacrificial insulating film 4 on the substrate side (lower side) is covered with the first protective insulating film 4a, and the surface on the opposite side (upper side) of the substrate side is covered with the second protective insulating film 4b. Since these protective insulating films 4a and 4b are formed of a material (for example, silicon nitride) different from the phosphate glass of the sacrificial insulating film 4, they are not easily etched when the sacrificial insulating film 4 in FIG. 9 is removed by etching. This is due to the high etching selectivity between the protective insulating films 4 a and 4 b and the sacrificial insulating film 4. Therefore, the protective insulating films 4a and 4b can protect the substrate 1A and the detection frame 5a.

  Further, the first protective insulating film 4a is formed of silicon nitride or the like which is not easily thermally oxidized, and prevents the substrate 1A (fixed electrode 3) from being oxidized in the step of oxidizing the column portion 5f of FIGS. It is formed so as to cover the upper surface of the fixed electrode 3. The second protective insulating film 4b is also formed of silicon nitride or the like which is not easily thermally oxidized, and prevents the lower surface of the detection frame 5a from being oxidized in the steps of FIGS. 10 and 11 (detection frame 5a (conductive It is formed so as to cover the lower surface of the body membrane 5). For this reason, it is possible to prevent the substrate 1A, the detection frame 5a, and the like from being etched or oxidized both when the sacrificial insulating film 4 is etched and when the column portion 5f is oxidized.

  Therefore, a change in the gap between the fixed electrode 3 and the detection frame 5a can be suppressed.

  Similar to the protective insulating films 4a and 4b, the thin film 7 as the third protective insulating film covers the upper surface and side surfaces of the detection frame 5a so as to prevent the upper surface and side surfaces of the detection frame 5a from being oxidized. Is formed. For this reason, deterioration of the performance as a micro device due to oxidation of the upper surface and side surfaces of the detection frame 5a can be suppressed.

(Embodiment 2)
Referring to FIG. 14, in the step of thermally oxidizing pillar portion 5 f shown in FIG. 11 of Embodiment 1 to become thermal oxide film 8, the width direction (Y direction) perpendicular to the vertical direction from the side surface of pillar portion 5 f is obtained. Concave portions 8 a and 8 b may be formed in the thermal oxide film 8 at a portion where the traveling thermal oxide films collide with each other (a central portion in the Y direction of the thermal oxide film 8 in FIG. 14). The sizes of the recesses 8a and 8b increase as the width of the column portion 5f in the X direction or the Y direction increases (as the width increases). The first embodiment shows a case where the concave portions 8a and 8b are small enough to be ignored. In the following embodiments, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof will not be repeated.

  If the thermal oxide film 8 of FIG. 14 is removed by vapor-phase hydrofluoric acid as in the step shown in FIG. 12 of the first embodiment, as shown in FIG. 15, the substrate 1A is formed at the place where the column portion 5f is formed. The movable portion conductor film (detection frame 5a, inertial mass body) is in a floating state, and the projections 9a and 9b are formed at locations where the fixed electrode 3 and the detection frame 5a face each other in the vertical direction (Z direction). And are formed.

  Thereby, a convex portion 9a is formed on the surface of the detection frame 5a facing the fixed electrode 3 of the substrate 1A, and a convex portion 9b is formed on the surface of the fixed electrode 3 of the substrate 1A facing the detection frame 5a. Has been. A convex portion 9b of the fixed electrode 3 is formed at a position facing the convex portion 9a of the detection frame 5a in the vertical direction (Z direction).

  According to the present embodiment, after the column portion 5f is oxidized as shown in FIG. 14 to form the thermal oxide film 8, the thermal oxide film 8 is removed as shown in FIG. Protrusions 9a and 9b are formed at locations where 5f has been connected.

  At this time, as shown in FIG. 14, when the polycrystalline silicon of the conductor film 3 and the conductor film 5 is thermally oxidized, the thermally oxidized portion is not oxidized any more. For this reason, if the oxidation time becomes longer, the shape of the protrusions 9a and 9b remains unchanged, and the thermal oxide film 8 is formed to be thicker in the vertical direction. Therefore, even if the time of thermal oxidation becomes long, the shape of the convex portions 9a and 9b is maintained, and the variation in the shape of the convex portions 9a and 9b is reduced. Therefore, a highly reliable microdevice (electrostatic) that has a small amount of sticking during manufacture and use of the substrate 1A and the conductive film for the movable part (detection frame 5a, inertial mass body) and that can easily make the sticking prevention effect constant. (Capacitive acceleration sensor) can be realized.

(Embodiment 3)
It is preferable that the above-mentioned column part 5f has as small a lateral dimension as possible to the extent that it can support the detection frame 5a. Referring to FIG. 16, a plurality of (for example, six) through-holes 6 and support columns 5f may be formed immediately below each region of detection frame 5a partitioned by hole 5e. In the following drawings, the illustration regarding the region B in FIG. 2 in the diagram showing each step of the first embodiment is omitted, but the diagram concerning the region A in FIG. 2 in the diagram showing each step in the first embodiment. Illustrated.

  If each region of the detection frame 5a partitioned by the hole 5e is supported by a plurality of support columns 5f in this way, each region is supported by a single support column 5f as in the first embodiment, for example. The support strength can be increased compared to the case.

  As shown in FIG. 16, the structure supported by the plurality of support columns 5 f is used when patterning the first protective insulating film 4 a, the sacrificial insulating film 4, and the second protective insulating film 4 b to form the through holes 6. This can be realized by setting the lateral dimension to the smallest dimension that can be processed by photolithography.

  Thus, if the horizontal dimension of the column part 5f is reduced, the time for thermally oxidizing the column part 5f and the time required for removing the thermal oxide film that has been thermally oxidized by the column part 5f can be shortened. Further, the residual stress caused by the films such as the detection frame 5a, the protective insulating films 4a and 4b, and the fixed electrode 3 can be dispersed, and the stress concentration on the column portion 5f can be reduced.

(Embodiment 4)
In the present embodiment, for example, as shown in FIG. 5 of the first embodiment (through the first protective insulating film 4a, the sacrificial insulating film 4 and the second protective insulating film 4b), the lateral dimension (X direction) And dimensions in the Y direction) are different from those of the first embodiment.

  Specifically, referring to FIG. 17, in the present embodiment, the through hole 6 has a through hole 6a in the second protective insulating film 4b, a through hole 6b in the sacrificial insulating film 4, and a first protective insulation. And a through hole 6c of the film 4a. The lateral dimension W2 of the through hole 6b of the sacrificial insulating film 4 is smaller than the lateral dimension W1 of the through hole 6a of the second protective insulating film 4b. The through hole 6 is formed so that the lateral dimension W3 of the through hole 6c of the first protective insulating film 4a is smaller than the lateral dimension W2 of the through hole 6b of the sacrificial insulating film 4. In this way, when the entire through hole 6 is viewed macroscopically, the through hole 6 has a tapered shape that narrows toward the substrate 1 side.

  The through holes 6 (through holes 6a, 6b, 6c) shown in FIG. 17 are formed as follows. First, the second protective insulating film 4b is patterned using a photolithography technique and an etching technique with a first mask pattern having a relatively large lateral dimension (having the dimension W1 in FIG. 17). Next, the sacrificial insulating film 4 is patterned in the same manner as described above with a second mask pattern whose lateral dimension is smaller than that of the first mask pattern (having the dimension W2 in FIG. 17). Next, the first protective insulating film 4a is patterned in the same manner as described above with a third mask pattern smaller than the second mask pattern (having the dimension W3 in FIG. 17).

  However, both the second protective insulating film 4b and the sacrificial insulating film 4 may be simultaneously patterned by the photolithography technique and the etching technique using the first mask pattern. In this way, the taper angle of the through hole 6 becomes steeper (that is, the difference between W1, W2, and W3 becomes smaller) than in the embodiment of FIG. 17 formed using the first to third mask patterns. .

  According to the present embodiment, for example, only the column portion 5f in the through hole 6c closest to the substrate 1A can be selectively thermally oxidized and removed. In this way, by leaving the column portion 5f in the through holes 6a and 6b, this can be used as the convex portion 9a shown in FIG. 15 of the second embodiment. Therefore, as described above, the supporting column portion 5f remaining in the through holes 6a and 6b functions as the convex portion 9a, so that the fixing between the detection frame 5a and the fixed electrode 3 can be suppressed as in the second embodiment. .

  Further, for example, if the lateral dimension W1 is increased, the lateral dimension of the base (upper side in FIG. 17) of the convex part of the support part 5f fixed to the detection frame 5a can be increased. The mechanical strength of the part 5f can be increased, and a highly reliable micro device (capacitance acceleration sensor) can be realized. In this case, it is preferable that the taper angle (the angle formed between the lower surface of the detection frame 5a and the side surface of the column portion 5f viewed macroscopically) is 45 ° or less.

(Embodiment 5)
Also in this embodiment, the final form is the same as that of the acceleration sensor 100 of the first embodiment shown in FIGS. 1 to 3, but the manufacturing method is partially different. Hereinafter, a method for manufacturing an acceleration sensor as a method for manufacturing a microdevice according to the present embodiment will be described with reference to FIGS.

  Referring to FIG. 18, first, a substrate 1A composed of a support substrate 1, an insulating film 2, and a patterned fixed electrode 3 is formed as in FIG. Next, a first protective insulating film 4a and a conductor film 5 are formed in this order on the substrate 1A so as to cover the fixed electrode 3. The thickness T5 (FIG. 18) of the conductor film 5 here is the sum of the thickness T2 (FIG. 4) of the sacrificial insulating film 4 of the first embodiment and the thickness T3 (FIG. 4) of the first protective insulating film 4a. It is preferable to make them approximately equal.

  Referring to FIG. 19, conductive film 5 in FIG. 18 is patterned using a photolithography technique and an etching technique. Thereby, the column part 5f similar to Embodiment 1 is formed.

  Referring to FIG. 20, sacrificial insulating film 4 is formed on first protective insulating film 4a so as to cover column portion 5f. Thereafter, the upper surface side of the sacrificial insulating film 4 is polished and removed by, for example, CMP (Chemical Vapor Deposition) so that the thickness of the sacrificial insulating film 4 (in the vertical direction in FIG. 20) becomes substantially equal to the thickness of the column portion 5f. As a result, the sacrificial insulating film 4 is formed as the same layer as the support column 5f so as to cover the entire side surface of the support column 5f of FIG. If the viewpoint is changed, the column portion 5f is arranged so as to bury the through hole of the sacrificial insulating film 4. The flatness of the sacrificial insulating film 4 can be improved by the above CMP.

  Referring to FIG. 21, a second protective insulating film 4b and a conductor film 5 are formed in this order so as to cover the upper surfaces of the sacrificial insulating film 4 and the column portions 5f.

  Referring to FIG. 22, similarly to the process of FIG. 7, deep etching (DRIE) or the like is performed on the conductive film 5 to pattern the conductive film 5, and the detection frame 5 a and the illustrated figure are formed from the conductive film 5. Unsupported support portions 5c, link beams 5b, and the like are formed. At this time, the second protective insulating film 4b functions as an etching stopper. A hole 5e is also formed.

  Referring to FIG. 23, similarly to the process of FIG. 8, a thin film 7 as a third protective insulating film is formed so as to cover the surface of the conductor film 5 such as the detection frame 5a, and the hole 5e. The second protective insulating film 4b immediately below is removed, and an opening 5g is formed.

  Referring to FIG. 24, as in the step of FIG. 9, sacrificial insulating film 4 is etched away by a liquid phase process, that is, a hydrofluoric acid solution.

  Referring to FIGS. 25 and 26, as in the steps of FIGS. 10 and 11, the entire support column portion 5f is thermally oxidized, whereby the thermal oxide film 8 is formed.

  Thereafter, the thermal oxide film 8 is removed by vapor-phase hydrofluoric acid as in the first embodiment. Further, as in the first embodiment, the first protective insulating film 4a, the second protective insulating film 4b, and the third protective insulating film 7 are removed by etching, so that a mode similar to FIG. 12 is obtained.

  In the present embodiment, the support column 5f is formed as a separate body rather than integrated with the detection frame 5a. Therefore, as compared with the case where the support column 5f and the detection frame 5a are integrated as in the first embodiment, detection immediately above the support column 5f during the thermal oxidation of the support column 5f shown in FIGS. The shape change of the frame 5a can be reduced. For this reason, a more reliable microdevice (capacitance type acceleration sensor) is realizable.

(Other)
In the first to fifth embodiments, the acceleration sensor has been described as the micro device. However, the micro device may be another device. For example, the present invention may be applied to the angular velocity sensor shown in FIGS. 27 to 29, the micromirror shown in FIGS.

First, the angular velocity sensor will be described.
27 to 29, the angular velocity sensor 200 is configured such that each component is disposed on a semiconductor substrate 11 on which a thin insulating film 12 is formed, and the detection frame according to the first embodiment. It has an inertial mass body 15a having a function similar to that of 5a. The inertia mass body 15a is connected via a link beam 15b extending in the Y-axis direction, and is held parallel to the surface of the substrate 11 above the semiconductor substrate 11, as can be seen from FIG.

  The support portion 15c has a frame shape so as to surround the periphery of the inertial mass body 15a with a gap 15d in plan view, and is fixedly supported by the semiconductor substrate 11 around the inertial mass body 15a. The support portion 15c is connected to the link beam 15b. Thereby, the support part 15c is connected to the inertia mass body 15a via the link beam 15b. As described above, the inertial mass body 15a is fixedly supported on the semiconductor substrate 11 with the link beam 15b and the support portion 15c interposed therebetween. Further, at least a part of the support portion 15c has conductivity.

  The inertia mass body 15a, the link beam 15b and the support portion 15c are made of, for example, polycrystalline silicon, the semiconductor substrate 11 is made of silicon, and the insulating film 12 is made of, for example, a low stress silicon nitride film or silicon oxide film. Has been.

  A detection electrode 13 is provided on the insulating film 12 corresponding to the inertia mass body 15a. The inertia mass body 15a and the detection electrode 13 are electrically independent with a small gap in the Z direction. Further, a hole 15e may be formed in the inertial mass body 15a in the same manner as the detection frame 5a.

  The operating principle of this angular velocity sensor 200 is as follows. Inertial mass body 15a vibrates on semiconductor substrate 11 by a driving mechanism (not shown) (inertial mass bodies 15a on the left side and right side of gap 15d at the center in the X direction in FIG. 27 vibrate in opposite phases with respect to the X direction) When the angular velocity around the axis in the Y direction in the figure is input, the inertial mass body 15a has its velocity vibration vector direction and a direction orthogonal to the angular velocity vector direction (Z direction orthogonal to the semiconductor substrate 11). In addition, it receives inertial force, that is, Coriolis force. As a result, the inertial mass body 15a receives an inertial force outward from the semiconductor substrate 11 having opposite phases, and rotational vibration around the Y-direction central axis is induced. Since the displacement of this rotational vibration is proportional to the angular velocity, this displacement is converted into an electrical output by a CV converter through a capacitance change between the inertial mass body 15a and the detection electrode 13 provided therebelow. Thus, the angular velocity signal can be acquired.

  The capacitance change formed between the detection electrode 13 and the inertial mass body 15a becomes a differential change at the time of angular velocity detection. By adopting a circuit configuration capable of detecting this differential change, a disturbance that changes in the same phase. A configuration that is not affected by vibration (displacement vibration of the inertial mass body 15a generated by acceleration in the X direction or Z direction) can be realized.

  This angular velocity sensor 200 may be formed by the same manufacturing method as the acceleration sensors of the first to fifth embodiments.

Next, the micromirror will be described.
Referring to FIGS. 30 and 31, in micromirror 300, mirror portion 25e is formed on one surface of mirror forming substrate 25a, and is formed of an aluminum thin film, a gold thin film, or the like. The mirror forming substrate 25a is configured to rotate about the center of gravity G at the center. The link beam 25b is formed on the central extension of the mirror forming substrate 25a.

  The support portion 25c has a frame shape so as to surround the periphery of the mirror formation substrate 25a with a gap 25d in plan view, and is fixedly supported by the substrate 21A around the mirror formation substrate 25a. The support portion 25c is connected to the link beam 25b. Thus, the support portion 25c is connected to the mirror forming substrate 25a via the link beam 25b. As described above, the mirror forming substrate 25a is fixedly supported on the substrate 21A via the link beam 25b and the support portion 25c. Further, at least a part of the support portion 25c has conductivity.

  The mirror forming substrate 25a, the link beam 25b, and the support portion 25c are made of, for example, polycrystalline silicon, and the substrate 21A is made of silicon, glass, or the like.

  Two drive electrodes 23 are formed on the substrate 21A. The two drive electrodes 23 are portions that are formed at a gap g0 from the mirror forming substrate 25a and to which a voltage is applied when the mirror forming substrate 25a is driven by electrostatic attraction.

  The operating principle of this micromirror is as follows. When a voltage is applied to one of the drive electrodes 23, an electrostatic attractive force corresponding to the potential difference and the capacitance between the mirror forming substrate 25a and the drive electrode 23 is generated, and the mirror forming substrate 25a is centered on the central portion. The mirror unit 25e is rotated and tilted by an angle (scanning angle) θs. For example, by applying a predetermined bias DC voltage to the two drive electrodes 23 and further applying a predetermined AC voltage to the drive electrodes 23 so that the phases of the two drive electrodes 23 are 180 degrees different from each other, the mirror portion 25e. Can be rotated and vibrated. In this way, the angle of the mirror unit 25e is controlled based on the applied voltage, and the light beam is scanned.

  This micro mirror 300 may be formed by the same manufacturing method as the acceleration sensors of the first to fifth embodiments.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  1A, 21A substrate, 1 support substrate, 2,12 insulating film, 3 fixed electrode, 4 sacrificial insulating film, 4a first protective insulating film, 4b second protective insulating film, 5 conductor film, 5a detection frame, 5b 15b, 25b Link beam, 5c, 15c, 25c Support part, 5d, 15d, 25d Clearance, 5e, 15e Hole part, 5f Post part, 5g Opening part, 6, 6a, 6b, 6c Through hole, 7 Thin film, 8 Thermal oxide film, 8a, 8b concave portion, 9a, 9b convex portion, 11 semiconductor substrate, 13 detection electrode, 15a inertial mass, 23 drive electrode, 25a mirror forming substrate, 25e mirror portion, 100 acceleration sensor, 200 angular velocity sensor, 300 Micro mirror.

Claims (9)

A sacrificial insulating film made of an oxide containing at least phosphorus and having a through-hole on a substrate, a post portion embedded in the through-hole, and the sacrificial insulating film and the conductor film for movable part located on the post portion Forming a process; and
Removing the sacrificial insulating film by a liquid phase process, and supporting the conductive film for the movable portion with the support portion with respect to the substrate;
A step of removing the support column by vapor phase processing and floating the conductive film for the movable part from the substrate in a region where the support column is formed. The step of removing the column portion by the vapor phase treatment,
Thermally oxidizing the support column;
The method for manufacturing a microdevice according to claim 1, further comprising a step of removing the thermally oxidized column portion. Before the step of removing the sacrificial insulating film by liquid phase treatment,
Forming a first protective insulating film for preventing the substrate from being oxidized so as to cover the surface of the sacrificial insulating film on the substrate side;
Forming a second protective insulating film for preventing the lower surface of the conductive film for the movable part from being oxidized so as to cover the surface of the sacrificial insulating film opposite to the substrate side. The manufacturing method of the microdevice of Claim 2. The through hole is formed to penetrate the first and second protective insulating films,
The through hole of the sacrificial insulating film has a smaller lateral dimension than the through hole of the second protective insulating film, and the through hole of the first protective insulating film is lateral to the through hole of the sacrificial insulating film. The manufacturing method of the microdevice of Claim 3 with a small dimension. The step of forming the sacrificial insulating film and the column portion includes
Forming the support column on the substrate;
5. The method of manufacturing a microdevice according to claim 1, further comprising: forming the sacrificial insulating film so as to cover an entire side surface of the support column after forming the support column. The step of forming the sacrificial insulating film and the column portion includes
Forming the sacrificial insulating film on the substrate;
Forming the through hole penetrating the sacrificial insulating film;
The method for manufacturing a microdevice according to claim 1, further comprising a step of forming the support column so as to embed the inside of the through hole. The step of forming the movable part conductor film,
Forming a conductor film on the sacrificial insulating film;
Forming the conductive film for the movable part and the conductive film for the supporting part supported by the substrate and supporting the conductive film for the movable part from the conductive film by patterning the conductive film. The manufacturing method of the microdevice of any one of Claims 1-6 containing these.   The step of removing the column portion by the vapor phase treatment is performed by floating the conductive film for the movable portion from the substrate at the position where the column portion is formed, by removing the thermally oxidized column portion, The method according to claim 2, further comprising forming a convex portion on at least one of a surface of the movable portion conductive film facing the substrate and a surface of the substrate facing the movable portion conductive film. Device manufacturing method.   5. The method includes a step of forming a third protective insulating film for preventing the movable part conductor film from being oxidized so as to cover an upper surface and a side surface of the movable part conductor film. A method for producing a microdevice as described in 1.

Images