Self-aligned polysilicon single-crystalline silicon hybrid MEMS vertical electrodes and manufacturing method thereof

Abstract

The present invention discloses a self-aligned polysilicon single-crystal hybrid MEMS vertical electrode and a manufacturing method thereof. By etching a plurality of deep grooves on a single-crystal wafer, oxidizing the single-crystal wafer, and depositing polysilicon, a polysilicon single-crystal hybrid MEMS vertical electrode is formed. The single crystal silicon has a perfect lattice and good physical properties, and can be used to manufacture movable electrodes and functional structures; the polysilicon is fixed on the single-crystal fixing posts as fixed electrodes. The single-crystal movable electrode and the polysilicon fixed electrode are determined by the pattern of one-time photolithography / etching to achieve self-alignment. The electrode spacing in the horizontal direction between the electrodes is determined by the thickness of the silicon dioxide layer. The polysilicon single-crystal hybrid MEMS vertical electrode of the present invention has the advantages of high dimensional accuracy, good processing repeatability, good performance consistency, and simple processing technology. Moreover, since it does not require the use of expensive SOI wafers, the cost is also reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of microelectronic wafer processing, and particularly relates to a self-aligned polysilicon single-crystal silicon hybrid MEMS vertical electrode and a manufacturing method thereof. Background Art

[0002] MEMS (Micro-Electro-Mechanical System) chips usually have movable structures, fixed structures and spring structures for supporting the movable structures, fixed electrodes, and cavities for providing free movement space for the movable structures. Some MEMS chip structures require driving structures to provide motive power for the MEMS movable structures. These motive powers include electrostatic force, magnetic force, fluid pressure, piezoelectric force, etc. Among them, the most widely used is the electrostatic force. For example, MEMS oscillators, MEMS gyroscopes, resonant MEMS accelerometers, MEMS micromirrors, MEMS optical attenuators, resonant MEMS pressure sensors, MEMS actuators, etc. all use electrostatic force mechanical driving structures. Generally, the electrostatic force is provided by the voltage difference between the fixed electrode and the movable electrode. The corresponding electrode on the movable structure is driven through parallel plate electrodes or interdigital electrodes to make the movable structure generate controllable movement. The movement direction of the movable structure can be horizontal or vertical according to the functional requirements of the MEMS device. The horizontal driving structure is relatively easy to process, while the vertical driving structure has a complex processing technology. The parallel plate electrode has a small driving distance, poor linearity, and low driving efficiency per unit area. Therefore, some MEMS devices, such as optical attenuators, micromirrors, and certain three-axis gyroscopes, require a vertical comb-tooth electrode driving structure. In addition, the signals of some MEMS sensors are sensed through the displacement of the movable structure. For example, MEMS gyroscopes, MEMS accelerometers, etc. also require a vertical electrode structure in their chip designs.

[0003] Figure 1 The shown vertical electrode (hereinafter, the vertical electrode refers to the electrode structure in which the MEMS movable electrode moves in a direction perpendicular to the bottom plate of the MEMS chip) structure is the easiest to process. The movable electrode 173 and the fixed electrode 171 form a height difference D at the upper part and have no height difference at the lower part. The electrode spacing is W0. The movable electrode 173 can move in the vertical direction. When this structure is used for driving, the movable electrode can only move upward in a single direction; when this structure is used for detection, it can only detect signals in a single direction, and the linearity is poor. Figure 2The vertical electrode structure shown is widely used in products such as MEMS optical attenuators and MEMS micromirrors. The movable electrode 273 is located above the fixed electrode 271, and there is no overlap between them in the horizontal direction. The spacing difference in the vertical direction is D. Since this structure is formed by a secondary lithography alignment / etching process or by a double single-crystalline silicon wafer bonding process, considering the alignment accuracy, the electrode spacings W1 and W2 in the horizontal direction are not necessarily equal and are relatively large. The driving force is inversely proportional to the electrode spacing, resulting in a small driving force. Moreover, the manufacturing process of this structure is complex. The movable electrode 273 can move in the vertical direction. When this structure is used for driving, the movable electrode can only move downward in a single direction, with poor linearity and cannot be used for precise detection. The MEMS device structures and processing methods described in patents CN11718906, CN103086316, CN113820851, CN113820852, US10268037B2, US10551613B2, etc. form the Figure 2 structure shown.

[0004] Patents US10077184B2 and US20050013087A1 describe adding materials with different coefficients of thermal expansion to a spring, causing the spring to warp in the vertical direction and forming a height difference between different electrodes in the vertical direction. The processing accuracy of this type of technology has poor repeatability, and the temperature characteristics of the fabricated devices are poor. Patents US7469588B2 and US9493344B2 form a fully suspended vertical induction electrode through the electrical isolation of interdigitated electrodes in the vertical direction and cannot be used for vertical driving. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a self-aligning polysilicon single-crystalline silicon hybrid MEMS vertical electrode and its manufacturing method. The single-crystalline silicon movable electrode and the polysilicon fixed electrode are formed by a single lithography / etching process, with precise control of the electrode spacing W, good processing repeatability, good performance, and a simple processing technology.

[0006] To solve the above technical problems, the present invention provides a self-aligned polysilicon single-crystal hybrid MEMS vertical electrode, which is composed of a single-crystal movable electrode, a polysilicon fixed electrode, a cavity, a single-crystal substrate, a single-crystal fixing column, and an edge bonding region. The polysilicon fixed electrode is fixed on the single-crystal fixing column through a silicon dioxide layer and is electrically connected to the single-crystal fixing column through a contact hole; the single-crystal fixing column is bonded to the single-crystal substrate, and there is an insulating layer between the single-crystal fixing column and the single-crystal substrate; the single-crystal movable electrode is connected to the MEMS functional structure; there is a horizontal distance W between the single-crystal movable electrode and the polysilicon fixed electrode, the bottom of the single-crystal movable electrode is lower than the bottom of the polysilicon movable electrode, and the height difference is D1, and the top of the polysilicon fixed electrode is higher than the single-crystal movable electrode, and the height difference is D2; there is a cavity between the single-crystal movable electrode and the polysilicon fixed electrode and the single-crystal substrate, providing space for the free movement of the single-crystal movable electrode.

[0007] The MEMS functional structure is a mirror, a mass block, or an oscillator.

[0008] There is silicon dioxide remaining between the polysilicon fixed electrode and the single-crystal fixing column, which plays a fixing role.

[0009] The self-aligned polysilicon single-crystal hybrid MEMS vertical electrode of the present invention can be used to detect the displacement signal of the MEMS functional structure in the vertical direction and can also be used to drive the MEMS functional structure to move in the vertical direction. When used as a detection electrode, the displacement of the MEMS functional structure in the vertical direction is measured by the change in the capacitance signal between the polysilicon fixed electrode and the single-crystal movable electrode; when used as a driving electrode, when different polar voltages are applied between the polysilicon fixed electrode and the single-crystal movable electrode, the single-crystal movable electrode moves upward; when the same polar voltage is applied, the single-crystal movable electrode moves downward. Moreover, the electrode distance W of the present invention is determined by the thickness of the oxide layer (silicon dioxide layer), that is, for MEMS structure layers of the same thickness, the electrode distance of the present invention can be made smaller. When the vertical electrode of the present invention is used as the driving electrode of the MEMS structure, the driving force in the vertical direction is inversely proportional to the electrode distance W. The smaller the electrode distance W, the greater the driving force; similarly, when the vertical electrode of the present invention is used as the detection electrode of the MEMS structure, the sensitivity in the vertical direction is inversely proportional to the electrode distance W. The smaller the electrode distance W, the greater the sensitivity; the greater advantage is that during the detection process, the electrode distance W remains unchanged, and only the overlapping area between the single-crystal movable electrode and the polysilicon fixed electrode changes, with good linearity and a large detection range; in addition, the height differences D1 and D2 between the single-crystal movable electrode and the polysilicon fixed electrode in the vertical direction are formed by etching, and there are height differences at both the upper and lower ends, and this height difference can be adjusted according to design requirements. Considering the edge effect, the driving force is greater than that of the electrode with a single-end height difference. Therefore, the self-aligned polysilicon single-crystal hybrid MEMS vertical electrode of the present invention has a strong driving force, a large driving stroke, and good linearity.

[0010] To solve the above technical problems, the present invention also provides a manufacturing method for a self-aligned polysilicon single-crystal hybrid MEMS vertical electrode, including the following steps:

[0011] (1) Etch the single-crystal silicon on the surface of the single-crystal silicon wafer through a photoresist mask to form a first cavity, an edge bonding region, and a fixed bonding column;

[0012] (2) Etch a deep trench in the first cavity through a photoresist mask. The single-crystal silicon wafer is divided into a single-crystal silicon movable electrode column, a single-crystal silicon fixed column, and an edge region in the horizontal direction. A first cavity step and a fixed bonding column are etched on the single-crystal silicon fixed column, and an edge step is etched in the edge region; in the vertical direction, it is divided into a substrate layer and a MEMS structure layer;

[0013] (3) Oxidize the single-crystal silicon wafer processed in step (2) to generate a silicon dioxide layer, and then etch the silicon dioxide layer through a photoresist mask on the first cavity step to form a contact hole;

[0014] (4) Deposit a polysilicon layer on the surface of the single-crystal silicon wafer processed in step (3) by in-situ doped CVD method and anneal;

[0015] (5) Back-etch the polysilicon layer without a mask to remove a part of the polysilicon layer to form a second polysilicon surface, exposing the first surface of the silicon dioxide layer of the single-crystal silicon fixed column and the edge region. The second polysilicon surface is lower than the first surface of the silicon dioxide layer;

[0016] (6) Cover the single-crystal silicon fixed column with a photoresist mask and etch the polysilicon layer. The polysilicon under the protection of the photoresist mask is not etched off to form a polysilicon fixed region. The polysilicon on the edge step is etched off to expose the silicon dioxide. The polysilicon in the deep trench is partially etched off to form a third polysilicon surface. The third polysilicon surface is lower than the first surface of the single-crystal silicon movable electrode column. The remaining polysilicon in the deep trench constitutes the fixed electrode of the MEMS structure layer;

[0017] (7) Corrode the silicon dioxide layer on the surface of the single-crystal silicon wafer processed in step (6) to expose the first surface of the single-crystal silicon movable electrode column and the first surface of the edge bonding region; then remove the photoresist mask and corrode and remove the silicon dioxide layer on the surface of the fixed bonding column to expose the first surface of the fixed bonding column to form a structure wafer;

[0018] (8) Bond the structure wafer in step (7) to the bottom plate wafer to form a bonded wafer. The bottom plate wafer is composed of a single-crystal silicon substrate and an insulating layer;

[0019] (9) Remove the substrate layer to expose the silicon dioxide layer on the back of the deep trench. Cover the single-crystalline silicon fixing posts and the edge bonding area of the bonding wafer with a photoresist mask, and etch the single-crystalline silicon. The single-crystalline silicon movable electrode posts are etched to form a second surface of the single-crystalline silicon movable electrode posts, and the second surface of the single-crystalline silicon movable electrode posts is lower than the second surface of the silicon dioxide layer on the back of the deep trench.

[0020] (10) Remove the photoresist on the bonding wafer, etch away the silicon dioxide layer between the polysilicon layer and the single-crystalline silicon movable electrode posts to release the single-crystalline silicon movable electrode posts, forming single-crystalline silicon movable electrodes, and the polysilicon in the deep trench forms polysilicon fixed electrodes.

[0021] In step (3), the single-crystalline silicon wafer can also be thermally oxidized first to form a thin silicon dioxide layer on all surfaces. After etching away the thin silicon dioxide layer, the single-crystalline silicon wafer is thermally oxidized again to form a silicon dioxide layer.

[0022] Step (5) can also be: first thin the polysilicon layer until just the first cavity step is exposed, then form contact holes, then deposit polysilicon layer twice and anneal, and finally thin the polysilicon layer deposited for the second time.

[0023] In step (7), the etching amount of the silicon dioxide between the single-crystalline silicon movable electrode posts and the polysilicon fixed area is greater than the etching amount of the silicon dioxide between the polysilicon fixed area and the single-crystalline silicon fixing posts.

[0024] In step (10), there is silicon dioxide remaining between the polysilicon fixed electrode and the single-crystalline silicon fixing post.

[0025] The manufacturing method of the present invention forms a polysilicon and single-crystalline silicon hybrid vertical comb electrode by etching multiple deep grooves on a single-crystalline silicon wafer, oxidizing the single-crystalline silicon wafer, and depositing polysilicon. The single-crystalline silicon has a perfect lattice and good physical properties, and is used to manufacture movable electrodes and functional structures. The polysilicon is fixed on the single-crystalline silicon fixing posts as polysilicon fixed electrodes. The single-crystalline silicon movable electrodes and the polysilicon fixed electrodes are determined by the pattern of one-time photolithography / etching to achieve self-alignment. The electrode spacing W in the horizontal direction between the electrodes (hereinafter referred to as the electrode spacing) is determined by the thickness of the silicon dioxide layer. The self-aligned polysilicon single-crystalline silicon hybrid MEMS vertical electrode manufactured by the method of the present invention has the advantages of high dimensional accuracy, good processing repeatability, good performance consistency, and simple processing technology. Moreover, since it does not require expensive SOI wafers, the cost is also low.

[0026] In the prior art, the electrode spacing is generally formed by a deep silicon etching process. The existing mass-produced deep silicon etching technology has an aspect ratio of 30:1. The so-called aspect ratio is the ratio of the thickness of the MEMS structure layer to the horizontal spacing between the MEMS structures. Therefore, the electrode spacing of the vertical electrodes produced by the prior art is subject to the thickness of the MEMS structure. For example, the electrode spacing of a 60μm thick MEMS structure layer cannot be less than 2μm. In the present invention, because the electrode spacing W is determined by the thickness of the oxide layer (silicon dioxide layer), it does not depend on the thickness of the MEMS structure. Taking a 60μm thick MEMS structure layer as an example, the electrode spacing of the present invention can be as low as 0.5μm. When the vertical electrode is used as the driving electrode of the MEMS structure, the driving force in the vertical direction is inversely proportional to the electrode spacing W. The smaller the electrode spacing W, the greater the driving force. Similarly, when the vertical electrode is used as the detection electrode of the MEMS structure, the sensitivity in the vertical direction is inversely proportional to the electrode spacing W. The smaller the electrode spacing W, the greater the sensitivity. A greater advantage is that during the detection process, the electrode spacing W remains unchanged, and only the overlapping area of the single-crystal silicon movable electrode and the polycrystalline silicon fixed electrode changes, resulting in good linearity and a large detection range. In addition, the height difference D1 and D2 in the vertical direction between the single-crystal silicon movable electrode and the polycrystalline silicon fixed electrode is formed by etching, and there is a height difference between the upper and lower ends. The height difference can be adjusted according to design requirements. Taking into account the edge effect, the driving force is greater than that of the electrode with a single-end height difference. Therefore, the self-aligned polycrystalline silicon and single-crystal silicon hybrid MEMS vertical electrode manufactured by the method of the present invention has a strong driving force, a large driving stroke, and good linearity. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Schematic diagram of an existing vertical electrode structure.

[0028] Figure 2 Schematic diagram of another existing vertical electrode structure.

[0029] Figure 3 — Figure 14 This is a flow chart of a method for manufacturing a self-aligned polysilicon and single crystal silicon hybrid MEMS vertical electrode according to the first embodiment.

[0030] Figure 15 yes Figure 14 Enlarged view of the dotted line portion.

[0031] Figure 16 The diagram is a schematic diagram of a vertical electrode unit used as a core unit in a MEMS device.

[0032] Figure 17 It is a cross-sectional view of the MEMS structure wafer after the oxide layer between the electrodes is removed in Example 2. DETAILED DESCRIPTION

[0033] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0034] Embodiment 1

[0035] A manufacturing method of a self-aligned polysilicon single-crystal hybrid MEMS vertical electrode includes the following steps:

[0036] (1), Take a heavily doped single-crystal silicon wafer 12 as the single-crystal silicon layer of the MEMS structure, with a resistivity of 0.001 - 0.1 Ω*CM and a wafer thickness between 300 - 800 μm. Through processing procedures such as coating, exposure, development, etching, stripping, and cleaning, a first cavity 21, an edge bonding area 23, and a fixed bonding post 25 are formed on the surface of the single-crystal silicon wafer 12. The depth of the first cavity 21 is between 0.5 - 50 μm, and this depth determines the movable space of the single-crystal silicon movable electrode in the vertical direction, as Figure 3 shown.

[0037] (2), Through processing procedures such as coating, exposure, development, deep silicon etching, stripping, and cleaning, a deep trench 27 is formed in the first cavity 21, with a depth between 1 - 200 μm. At this time, the single-crystal silicon wafer 12 is divided into a single-crystal silicon movable electrode column 32, a single-crystal silicon fixed column 34, and an edge area 36 in the horizontal direction. A first cavity step 21a and a fixed bonding post 25 are formed on the single-crystal silicon fixed column 34, and an edge step 21b is formed on the edge area 36; the fixed bonding post 25 is located on the single-crystal silicon fixed column 34, and the edge bonding area 23 is located on the edge area 36. The single-crystal silicon wafer 12 is divided into a single-crystal silicon substrate layer 15 and a MEMS structure layer 17 in the vertical direction, as Figure 4 shown.

[0038] (3), Thermally oxidize Figure 4 the etched single-crystal silicon wafer with the deep trench 27 as shown, and form a uniform silicon dioxide layer 39 with a thickness between 0.5 - 5 μm on the first surface 17a of the MEMS structure layer 17, as Figure 5 shown.

[0039] To improve the quality of the sidewall of the MEMS structure layer 17, for example, to reduce the wavy roughness formed by the Bosch deep silicon etching process, the single-crystal silicon wafer 12 can also be thermally oxidized first to form a silicon dioxide layer with a thickness of 1 - 5 μm on all surfaces, and then this silicon dioxide layer is etched off with HF acid, and then the single-crystal silicon wafer 12 is thermally oxidized to form the silicon dioxide layer 39.

[0040] (4), On Figure 5 the surface of the single-crystal silicon wafer as shown, through processing procedures such as spraying glue, exposure, development, etching the silicon dioxide layer 39, stripping, and cleaning, a contact hole 38 is formed on the first cavity step 21a of the single-crystal silicon fixed column 34, as Figure 6As shown. The pattern of the contact holes 38 can be a rectangular or circular hole chain composed of multiple round holes, or a rectangular or circular hole chain composed of multiple square holes; it can also be a circular or square ring.

[0041] (5)、On the surface of the single-crystal silicon wafer shown, a polysilicon layer 40 doped in-situ is deposited by CVD (Chemical Vapor Deposition) method, and a flat first surface 40a of the polysilicon is formed through a CMP (Chemical Mechanical Polishing) process. At this time, the polysilicon layer 40 fills the deep trenches 27, the first cavity 21, and the contact holes 38; the polysilicon layer 40 covers all the silicon dioxide layers 39, and the first surface 40a of the first polysilicon layer is significantly higher than the first surface 39a of the silicon dioxide layer, as Figure 6 shown. Figure 7 As shown.

[0042] (6)、Using a maskless re-etching process, such as a plasma or reactive ion etching process of gases such as CF4 and SF6, a part of the polysilicon layer 40 is uniformly removed to form a second surface 40b of the polysilicon, as Figure 8 shown. At this time, the polysilicon layer 40 still covers the first cavity step 21a, the edge step 21b, and the first surface 32a of the single-crystal silicon movable electrode column. The contact holes 38 and the single-crystal silicon movable electrode column 32 are also covered by the polysilicon layer 40. The polysilicon layer 40 above the edge bonding region 23 and the fixed bonding column 25 is etched away, exposing the first surface 39a of the silicon dioxide layer. The second surface 40b of the polysilicon is significantly lower than the first surface 39a of the silicon dioxide layer.

[0043] (7)、On the surface of the single-crystal silicon wafer shown, processes such as spin coating, exposure, and development are carried out to form a photoresist mask 42, whose size is significantly larger than that of the single-crystal silicon fixed column 34 in at least one horizontal direction. The polysilicon layer 40 is etched to form a third surface 40c of the polysilicon, as Figure 8 shown. At this time, the polysilicon under the protected area of the mask 42 is not etched away, forming a polysilicon fixed area 45. The polysilicon on the edge step 21b is etched away, exposing the silicon dioxide layer 39. The polysilicon 40 in the deep trenches 27 is partially etched away to form the third surface 40c of the polysilicon. The third surface 40c of the polysilicon is significantly lower than the edge step 21b and the first surface 32a of the single-crystal silicon movable electrode column. The remaining polysilicon 40 in the deep trenches 27 constitutes the fixed electrode of the MEMS structure layer 17. Figure 9 As shown.

[0044] (8)、Etch with HF acid solution or gaseous HF acid Figure 9The single-crystalline silicon wafer shown is processed by controlling the etching time to etch away the silicon dioxide layer 39 outside the mask 42 that is not covered by the polysilicon 40, as well as a part of the silicon dioxide 39 between the polysilicon layer 40 and the single-crystalline silicon movable electrode column 32. The purpose is to facilitate the removal of all the silicon dioxide layers 39 in subsequent processes, release the movable structure, and expose the first surface 32a of the single-crystalline silicon movable electrode column and the first surface 23a of the edge bonding region. Then, the photoresist mask 42 is removed, and HF acid solution or gaseous HF acid is continuously used for etching to remove the silicon dioxide layer 39 on the surface of the fixed bonding column 25, exposing the first surface 25a of the fixed bonding column. The silicon dioxide 39 between the polysilicon layer 40 and the single-crystalline silicon fixed column 34 still exists, thus forming a structural wafer 50 as shown in Figure 10 shown.

[0045] (9). Take a bottom wafer 60 with the same crystal orientation as the single-crystalline silicon wafer 12. The bottom wafer 60 consists of a single-crystalline silicon substrate 62 and an insulating layer 64. The material of the insulating layer 64 is thermally grown silicon dioxide with a thickness between 0.5 and 3 μm. The first surface 17a of the MEMS structure layer of the structural wafer 50 is Si-SiO2 Fusion bonded to the insulating layer 64 of the bottom wafer 60, bonding the bottom wafer 60 and the structural wafer 50 together, and forming a bonding surface 55 between the first surface 25a of the fixed bonding column and the first surface 23a of the edge bonding region and the insulating layer 64, as shown in Figure 11 shown.

[0046] (10). Remove the single-crystalline silicon substrate layer 15 of the bonded wafer shown in Figure 11 through processes such as grinding, CMP, and maskless etching to form the second surface 17b of the MEMS structure layer, exposing the back surface 27a of the deep trench. The second surface 39b of the silicon dioxide layer on the back surface 27a of the deep trench blocks the active reaction components during the etching process, protecting the polysilicon 40 in the deep trench, as shown in Figure 12 shown.

[0047] (11). Through processing steps such as spin coating, exposure, development, etching, stripping, and cleaning, etch the bonded wafer shown in Figure 12 to form a second cavity 66, as shown in Figure 13 shown. The photoresist 68 covers the edge bonding region 23 and the fixed electrode column 34, and these two regions are not etched. Since the second surface 39b of the silicon dioxide layer blocks the active reaction components during the etching process, the polysilicon 40 in the deep trench 27 is not etched. At this time, the single-crystalline silicon movable electrode column 32 is etched to form the second surface 32b of the single-crystalline silicon movable electrode column, and the second surface 32b of the single-crystalline silicon movable electrode column is significantly lower than the back surface 27b of the deep trench.

[0048] (12). Remove Figure 13The photoresist 68 on the bonding wafer is cleaned, and then the silicon dioxide layer 39 is etched with HF solution or gaseous HF until the silicon dioxide 39 between the polysilicon layer 40 and the single crystal silicon movable electrode column 32 is completely removed and the two are separated. In this way, the single crystal silicon movable electrode column 32 is released and can move freely to form a single crystal silicon movable electrode 71; the polysilicon 40 in the deep trench 27 forms a polysilicon fixed electrode 73, as shown. Figure 14 As shown; the distance between the polycrystalline silicon fixed electrode 73 and the single crystal silicon movable electrode 71 is equal to the thickness of the silicon dioxide layer 39; a portion of silicon dioxide 39c remains between the polycrystalline silicon layer 40 and the single crystal silicon fixed column 34, which plays a fixing role; the insulating layer (silicon dioxide) 64 in the non-bonding area is also corroded away during the HF solution or gaseous HF etching process, exposing the single crystal silicon substrate 62; the insulating layer 64 remaining after etching, the single crystal silicon substrate 62 and the first concave cavity 21 together form a chamber 75, providing a movable space for the single crystal silicon movable electrode 71.

[0049] In order to more clearly illustrate the vertical comb electrode structure unit, Figure 14 The dotted line frame part (vertical electrode unit 100) is enlarged, as shown in FIG. Figure 15 As shown, there is a gap W between the single crystal silicon movable electrode 71 and the adjacent polycrystalline silicon fixed electrode 73, which is the electrode spacing in the horizontal direction; in the vertical direction, the second surface 32b of the single crystal silicon movable electrode 71 is lower than the upper surface of the polycrystalline silicon fixed electrode 73, which is the fourth surface 40d of the polycrystalline silicon, with a height difference of D1; the first surface 32a of the single crystal silicon movable electrode 71 is lower than the lower surface of the polycrystalline silicon fixed electrode 73, which is the third surface 40c of the polycrystalline silicon, with a height difference of D2; D1 and D2 can be equal or unequal, and the single crystal silicon movable electrode 71 can move freely in the vertical direction.

[0050] The vertical electrode unit 100 can be used as a core unit in different MEMS device structures, such as MEMS gyroscopes, accelerometers, microscopes, resonators or actuators, etc. Figure 16 As shown, a plurality of polycrystalline silicon fixed electrodes 73 form a fixed comb-tooth electrode group 83, which is fixed on a single crystal silicon fixed column 34. The surface 34a of the single crystal silicon fixed column can be bonded to a cover plate for exporting signals by Si-metal bonding or Si-Si bonding. A plurality of single crystal silicon movable electrodes 71 form a movable comb-tooth electrode group 81, which is connected to a movable functional structure 110 of a MEMS device. The movable functional structure 110 can be a reflector, a mass block or an oscillator, etc. The dotted lines around the single crystal silicon movable electrode 71 indicate that its surface 32b is vertically lower than the surface 34a of the single crystal silicon fixed column, the second surface 23b of the edge bonding area and the surface of the movable functional structure 110. The edge bonding area 23 can be used for bonding a cover plate to protect the MEMS structure, or it can be not bonded to any structure.

[0051] The self-aligned polysilicon single-crystalline silicon hybrid MEMS vertical electrode fabricated in this embodiment, as shown in Figure 14 , Figure 15 , is composed of a single-crystalline silicon movable electrode 71, a polysilicon fixed electrode 73, an insulating layer 64, a single-crystalline silicon substrate 62, a chamber 75, a silicon dioxide layer 39, a single-crystalline silicon fixed column 34, and an edge bonding region 23; the polysilicon fixed electrode 73 is fixed on the single-crystalline silicon fixed column 34 through a polysilicon fixed region 45 and a residual silicon dioxide 39c, and is electrically connected to the single-crystalline silicon fixed column 34 through a contact hole 38; the single-crystalline silicon fixed column 34 is fixed on the insulating layer 64 through a bonding process and has no electrical connection with the single-crystalline silicon substrate 62; the single-crystalline silicon movable electrode 71 is connected to the movable functional structure 110 of the MEMS device, such as a reflector, a driving mass, a detecting mass, or an oscillator, etc.; there is a gap W between the single-crystalline silicon movable electrode 71 and the polysilicon fixed electrode 73, the bottom 32a of the single-crystalline silicon movable electrode 71 is lower than the bottom 40c of the polysilicon fixed electrode 73, and the height difference is D1; the top 32b of the single-crystalline silicon movable electrode 71 is lower than the top 40d of the polysilicon fixed electrode 73, and the height difference is D2; there is a cavity 75 between the vertical electrode unit 100 and the single-crystalline silicon substrate 62, providing a free movement space for the movable electrode 71; the edge bonding region 23 provides protection for the MEMS structure.

[0052] Embodiment 2

[0053] After the bonding wafer shown in Figure 13 is formed, the photoresist 68 is retained, and then the silicon dioxide layer 39 is etched with an HF solution until the silicon dioxide 39 between the polysilicon layer 40 and the single-crystalline silicon movable electrode column 32 is completely removed, and the two are separated. Then, the photoresist 68 is removed and then cleaned, as shown in Figure 17 . In this way, the single-crystalline silicon movable electrode column 32 is released and can move freely to form the single-crystalline silicon movable electrode 71; the polysilicon 40 in the deep trench 27 forms the polysilicon fixed electrode 73; the spacing W between the polysilicon fixed electrode 73 and the single-crystalline silicon movable electrode 71 is equal to the thickness of the silicon dioxide layer 39; there is still a residual silicon dioxide 39d between the polysilicon 40 and the single-crystalline silicon fixed column 34. Since during the HF solution etching process, the photoresist 68 prevents the vertical etching of the silicon dioxide layer 39 covered by it, but there will still be some silicon dioxide etched horizontally by the HF solution. Obviously, the residual silicon dioxide 39d in Embodiment 2 is more than the residual silicon dioxide 39c in Embodiment 1, and the bonding mechanical strength between the polysilicon fixed electrode 73 and the single-crystalline silicon fixed column 34 is higher than that in Embodiment 1; the non-bonding region insulating layer (silicon dioxide) 64 is also etched away during the HF solution etching process, exposing the single-crystalline silicon substrate 62; the remaining insulating layer 64, the single-crystalline silicon substrate 62, and the first cavity 21 after etching together enclose a chamber 75, providing a movement space for the single-crystalline silicon movable electrode 71.

[0054] The above is only the best implementation mode of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several deformations or equivalent replacements can be made to the technical solution of the present invention. For example: on Figure 5 the silicon single crystal wafer shown, deposit an in-situ doped polysilicon layer 40, thin the polysilicon layer 40 through the CMP process and the back etching process until the first cavity step 21a is just exposed, and then form a contact hole 38 through processing steps such as coating, exposure, development, etching the silicon dioxide layer 39, removing the glue, and cleaning. Then deposit a polysilicon layer for the second time, anneal, and thin the polysilicon layer deposited for the second time through the CMP process and the back etching process to form Figure 8 the wafer structure shown; this method can avoid the large height difference between the deep trench 27 and the first cavity step 21a when forming the contact hole 38, and reduce the difficulty of the lithography process. These can also achieve the technical effects of the present invention and should also be regarded as belonging to the protection scope of the present invention.

Claims

1. A manufacturing method of a self-aligned polysilicon single-crystal hybrid MEMS vertical electrode, characterized in that, It includes the following steps: (1) Etch single-crystal silicon on the surface of a single-crystal silicon wafer through a photoresist mask to form a first concave cavity, an edge bonding area, and fixed bonding columns; (2) Etch deep trenches in the first concave cavity through a photoresist mask. The single-crystal silicon wafer is divided into single-crystal silicon movable electrode columns, single-crystal silicon fixed columns, and an edge area in the horizontal direction. A first concave cavity step and fixed bonding columns are etched on the single-crystal silicon fixed columns, and an edge step is etched in the edge area; in the vertical direction, it is divided into a substrate layer and a MEMS structure layer; (3) Oxidize the single-crystal silicon wafer processed in step (2) to generate a silicon dioxide layer, and then etch the silicon dioxide layer on the first concave cavity step through a photoresist mask to form contact holes; (4) Deposit a polysilicon layer on the surface of the single-crystal silicon wafer processed in step (3) by in-situ doped CVD method and anneal; (5) Back-etch the polysilicon layer without a mask to remove a part of the polysilicon layer to form a second polysilicon surface, exposing the first surface of the silicon dioxide layer of the single-crystal silicon fixed column and the edge area. The second polysilicon surface is lower than the first surface of the silicon dioxide layer; (6) Cover the single-crystal silicon fixed column with a photoresist mask and etch the polysilicon layer. The polysilicon under the protection of the photoresist mask is not etched away to form a polysilicon fixed area. The polysilicon on the edge step is etched away to expose the silicon dioxide. The polysilicon in the deep trench is partially etched away to form a third polysilicon surface. The third polysilicon surface is lower than the first surface of the single-crystal silicon movable electrode column. The remaining polysilicon in the deep trench constitutes the fixed electrode of the MEMS structure layer; (7) Corrode the silicon dioxide layer on the surface of the single-crystal silicon wafer processed in step (6) to expose the first surface of the single-crystal silicon movable electrode column and the first surface of the edge bonding area; then remove the photoresist mask and corrode away the silicon dioxide layer on the surface of the fixed bonding column to expose the first surface of the fixed bonding column to form a structure wafer; (8) Bond the structure wafer in step (7) to a bottom plate wafer to form a bonded wafer. The bottom plate wafer is composed of a single-crystal silicon substrate and an insulating layer; (9) Remove the substrate layer to expose the silicon dioxide layer on the back of the deep trench. Cover the single-crystal silicon fixed column and the edge bonding area of the bonded wafer with a photoresist mask and etch the single-crystal silicon. The single-crystal silicon movable electrode column is etched to form a second surface of the single-crystal silicon movable electrode column. The second surface of the single-crystal silicon movable electrode column is lower than the second surface of the silicon dioxide layer on the back of the deep trench; (10) Remove the photoresist on the bonded wafer and corrode away the silicon dioxide layer between the polysilicon layer and the single-crystal silicon movable electrode column to release the single-crystal silicon movable electrode column to form a single-crystal silicon movable electrode. The polysilicon in the deep trench forms a polysilicon fixed electrode.

2. The manufacturing method of the self-aligned polysilicon single-crystalline silicon hybrid MEMS vertical electrode according to claim 1, wherein: In step (7), the corrosion amount of the silicon dioxide between the single-crystal silicon movable electrode column and the polysilicon fixed area is greater than the corrosion amount of the silicon dioxide between the polysilicon fixed area and the single-crystal silicon fixed column.

3. The manufacturing method of the self-aligned polysilicon single-crystal silicon hybrid MEMS vertical electrode according to claim 1, wherein: In step (10), silicon dioxide remains between the polysilicon fixed electrode and the single-crystal silicon fixed column.

4. The manufacturing method of the self-aligned polysilicon single-crystal silicon hybrid MEMS vertical electrode according to claim 1, wherein: The horizontal distance between the single-crystal silicon movable electrode and the polysilicon fixed electrode is determined by the thickness of the silicon dioxide layer generated in step (3).

Images