Semiconductor manufacturing apparatus and semiconductor manufacturing method

By measuring and etching the oxide film on the surface of semiconductor materials, and combining it with high-frequency ion flow bonding technology, the problem of insufficient bonding strength of semiconductor wafers was solved, achieving higher bonding strength and a more stable process.

CN123124675APending Publication Date: 2026-11-17SIHE MICRO TECHNOLOGY (SHANGHAI) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611284427.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2025-08-25
Filing Date
2026-08-24
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing semiconductor manufacturing processes, poor control of the oxide film on the surface of semiconductor wafers leads to insufficient bonding strength and the presence of particles and voids, affecting the stability and reliability of wafer bonding.

Method used

The oxide film on the surface of semiconductor materials is measured under high vacuum. The oxide film is removed by isotropic etching or plasma etching using etching solution. High-frequency power supply is used to polarize ions on the surface of semiconductor materials. Combined with high-frequency ion flow bonding technology, the processing parameters are optimized to improve the bonding strength.

Benefits of technology

Effective control of the oxide film on the surface of semiconductor materials improves the bonding strength and interface quality of semiconductor materials, ensures the stability and reproducibility of the process, and reduces thermal stress and material damage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123124675A_ABST
    Figure CN123124675A_ABST
Patent Text Reader

Abstract

A semiconductor manufacturing apparatus and a semiconductor manufacturing method are available that can improve the bonding strength between materials by obtaining a natural oxide film formed on the surface of the semiconductor material during the transport process of the semiconductor material. The device comprises: a measurement step (S1000) in which a first oxide film formed on the surface of semiconductor materials (W1, W2) is measured; a stripping step (S1300) in which the first oxide film is stripped away by isotropic etching or plasma etching based on the measurement result of the first oxide film; a transport step (S1400) in which semiconductor materials (W1, W2) are transported in a predetermined environment; and a bonding step in which, after the surface of semiconductor materials (W1, W2) is activated by plasma treatment, ions are polarized on the surface of semiconductor materials (W1, W2) under a high vacuum state for a predetermined time, and a high-frequency power supply is applied to semiconductor materials (W1, W2) to remove the second oxide film formed on the surface of semiconductor materials (W1, W2) during the transport process (S1400), and ions move on the surface of semiconductor materials (W1, W2) to perform bonding.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a semiconductor manufacturing apparatus and a semiconductor manufacturing method, wherein the semiconductor manufacturing apparatus includes a bonding device for bonding semiconductor materials to each other in a semiconductor manufacturing process by, for example, high-frequency ion flow bonding (RFIFB). Background Technology

[0002] As one of the existing semiconductor manufacturing processes, hybrid bonding includes processes such as plasma treatment, cleaning, drying and bonding.

[0003] In the plasma treatment process, for example, atmospheric pressure plasma or vacuum plasma is applied to the semiconductor wafer to activate its surface. In the cleaning process, for example, the surface of the semiconductor wafer is cleaned by dripping a cleaning solution (pure water or chemical solution) while the wafer is rotated around a central axis. In the drying process, for example, water droplets on the surface of the semiconductor wafer are dispersed while the wafer is rotated at high speed around a central axis. In the bonding process, for example, two semiconductor wafers, after undergoing the plasma treatment and drying processes, are placed face-to-face and bonded using a bonding device.

[0004] Here, in the existing bonding apparatus, semiconductor wafers are placed on planar workpiece stages arranged in an opposing manner, and a pair of opposing semiconductor wafers are bonded to each other by a surface active bonding (SAB) method (hereinafter appropriately referred to as the "SAB method"), which is a room temperature bonding process.

[0005] Surface activated bonding refers to the process of ionizing an inert gas such as argon to create an ion beam, which is then irradiated onto the surface of the materials to be bonded. This removes oxide films and contaminant layers present on the surface of the materials and activates the surface, thereby achieving bonding at low to room temperature.

[0006] Here, in the SAB method, the entire surface of the semiconductor wafer is uniformly activated in one go. When removing the oxide film, the irradiation area of ​​the ion beam is small. Therefore, the surface of the semiconductor wafer is processed by sliding the semiconductor wafer or the beam source of the ion beam.

[0007] However, the SAB method has the following problems: poor surface uniformity of the semiconductor wafers, with roughness on some parts of the wafer surface, which reduces the bonding strength between the semiconductor wafers. Furthermore, there are problems such as the generation of particles and voids due to the movement of the semiconductor wafers or beam caused by point irradiation with the Ar ion beam.

[0008] Therefore, a semiconductor wafer bonding apparatus is known that, through a simple structure, can improve the bonding strength between wafers in a semiconductor manufacturing process. In this bonding apparatus, semiconductor wafers are bonded to each other in a semiconductor manufacturing process, for example, by high-frequency ion flow bonding (RF Ion Flow Bonding, (RFIFB)).

[0009] Specifically, the semiconductor wafer bonding apparatus is as follows: after plasma treatment to activate the surface of the semiconductor wafer, a high vacuum state is set for a predetermined time to polarize the ions on the surface of the semiconductor wafer. The oxide film formed on the surface of the semiconductor wafer is removed by applying two high-frequency power supplies of different frequencies to each of the opposing semiconductor wafers. The ions move on the surface of the semiconductor wafers to improve the bonding strength between the semiconductor wafers. (Refer to Patent Document 1 below)

[0010] Prior art literature Patent documents Patent Document 1: Japanese Patent Publication No. 7645524 Summary of the Invention The technical problem that the invention aims to solve However, in the apparatus for bonding semiconductor materials together by the aforementioned high-frequency ion flow bonding, controlling the oxide film formed on the surface of the semiconductor material is a technical problem in order to improve the bonding strength between the semiconductor materials.

[0011] Therefore, in view of the above-mentioned technical problems, the present invention aims to provide a semiconductor manufacturing apparatus and a semiconductor manufacturing method that can improve the bonding strength between semiconductor materials in a semiconductor manufacturing process by controlling, for example, the natural oxide film formed on the surface of the semiconductor material during the transport of the semiconductor material.

[0012] The means used to solve technical problems The first invention is a semiconductor manufacturing apparatus that uses a semiconductor material bonding device for bonding a pair of semiconductor materials. A first oxide film formed on the surface of the semiconductor material is measured. Based on the measurement result of the first oxide film, the first oxide film is removed by isotropic etching or plasma etching based on an etching solution. For the semiconductor material that has undergone a transport process in a high-vacuum environment, an inert environment, or an atmospheric pressure environment, the bonding device performs plasma treatment to activate the surface of the semiconductor material. Within a predetermined time, ions are polarized on the surface of the semiconductor material under a high-vacuum state. A high-frequency power supply is applied to the semiconductor material, thereby removing a second oxide film formed on the surface of the semiconductor material during the transport process. The ions move on the surface of the semiconductor material and bond the semiconductor materials together.

[0013] Preferably, the processing parameters related to the processing performance of the isotropic etching, the plasma etching, and the plasma treatment are optimized based on the measurement results of the first oxide film.

[0014] Preferably, if there is an anomaly in the measurement results of the first oxide film, at least a portion of the first oxide film on the surface of the semiconductor material is removed and planarized before peeling off the first oxide film.

[0015] Preferably, after the semiconductor material is bonded, the bonding strength or porosity of the semiconductor material is checked.

[0016] Preferably, the adjustment of the processing parameters is performed when the bonding strength or porosity of the semiconductor material does not meet a predetermined benchmark.

[0017] Preferably, a film thickness gauge based on X-ray fluorescence is used to select at least 13 points on the surface of the semiconductor material to measure the first oxide film.

[0018] Preferably, in the plasma treatment of the bonding device, the emission spectrum is monitored in real time during plasma activation to obtain the concentration of active particles and detect the plasma activity.

[0019] Preferably, in the inspection of the bonding strength or porosity of the semiconductor material, a scanning acoustic microscope is used to examine the bonding interface of the semiconductor material, and the acoustic signal is analyzed to quantitatively evaluate the bonding strength or porosity of the semiconductor material.

[0020] Preferably, the processing parameters include at least one element: etching time or etchant concentration in the isotropic etching or plasma etching, gas flow rate in the plasma processing or applied output of the high-frequency power supply, bonding temperature or bonding time of the semiconductor material, and heat treatment temperature or heat treatment time for the semiconductor material.

[0021] Preferably, the predetermined time is 0.1 to 10 seconds.

[0022] Preferably, the semiconductor materials that are bonded together are held by an electrostatic chuck that serves as a parallel plate plasma electrode.

[0023] The second invention comprises: a measurement step in which a first oxide film formed on the surface of a semiconductor material is measured; a stripping step in which the first oxide film is stripped away based on isotropic etching or plasma etching using an etching solution, according to the measurement result of the first oxide film; a transport step in which the semiconductor material is transported in a high-vacuum environment, an inert environment, or an atmospheric pressure environment; and a bonding step in which, after the semiconductor material has been plasma-treated to activate its surface, ions are polarized on the surface of the semiconductor material under a high-vacuum state within a predetermined time, and a high-frequency power supply is applied to the semiconductor material to remove the second oxide film formed on the surface of the semiconductor material during the transport process, and the ions move on the surface of the semiconductor material to bond the semiconductor materials together.

[0024] Preferably, the process includes an optimization step that optimizes the processing parameters responsible for the processing performance related to the isotropic etching, the plasma etching, and the plasma treatment based on the measurement results of the first oxide film.

[0025] Preferably, it includes a planarization process that, in the event of an abnormality in the measurement process, removes and planarizes at least a portion of the first oxide film on the surface of the semiconductor material before the stripping process.

[0026] Preferably, the assembly includes an inspection step after the bonding process to check the bonding strength or porosity of the semiconductor material.

[0027] Preferably, there is an adjustment step that adjusts the processing parameters when the bonding strength or porosity of the semiconductor material does not meet a predetermined benchmark during the bonding process.

[0028] Preferably, in the measurement process, a film thickness gauge based on X-ray fluorescence is used to select at least 13 points on the surface of the semiconductor material for measurement.

[0029] Preferably, in the bonding process, the emission spectrum is monitored in real time during plasma activation to obtain the concentration of active particles and detect the plasma activity.

[0030] Preferably, in the inspection process, a scanning acoustic microscope is used to inspect the bonding interface of the semiconductor material, and the acoustic signal is analyzed to quantitatively evaluate the bonding strength or porosity of the semiconductor material.

[0031] Preferably, the processing parameters include at least one element: etching time or etchant concentration in the isotropic etching or plasma etching, gas flow rate in the plasma processing or applied output of the high-frequency power supply, bonding temperature or bonding time of the semiconductor material, and heat treatment temperature or heat treatment time for the semiconductor material.

[0032] Preferably, the predetermined time is 0.1 to 10 seconds.

[0033] Preferably, the semiconductor materials that are bonded together are held by an electrostatic chuck that serves as a parallel plate plasma electrode.

[0034] Invention Effects According to the present invention, by controlling the natural oxide film formed on the surface of the semiconductor material during the transport of the semiconductor material, the bonding strength between the semiconductor materials can be improved. Attached Figure Description

[0035] Figure 1 This is a structural diagram of a semiconductor wafer bonding apparatus constituting a semiconductor manufacturing apparatus according to an embodiment of the present invention.

[0036] Figure 2 This is a structural diagram of the plasma processing and bonding process performed by the bonding apparatus of a semiconductor wafer constituting an embodiment of the present invention.

[0037] Figure 3 This is a process diagram of plasma processing and bonding processing performed by a semiconductor wafer bonding apparatus constituting an embodiment of the present invention.

[0038] Figure 4 A flowchart illustrating an example of a high-frequency ion flow bonding process is provided.

[0039] Figure 5 This is a top view of a semiconductor manufacturing system in which a bonding apparatus for semiconductor wafers, constituting a semiconductor manufacturing apparatus according to an embodiment of the present invention, is assembled.

[0040] Figure 6This is a side view of a semiconductor manufacturing system in which a bonding apparatus for semiconductor wafers, constituting a semiconductor manufacturing apparatus according to an embodiment of the present invention, is assembled.

[0041] Figure 7 The diagram shows the structure of a unipolar electrostatic chuck.

[0042] Figure 8 The diagram shows the structure of a bipolar electrostatic chuck.

[0043] Figure 9 This is a structural diagram showing the state in which semiconductor wafers are bonded together using a bipolar electrostatic chuck disposed on the workpiece stage of a bonding apparatus constituting a semiconductor wafer.

[0044] Figure 10 This is a block diagram illustrating various parts of a semiconductor manufacturing apparatus according to one embodiment of the present invention.

[0045] Figure 11 This is a flowchart illustrating the manufacturing process of a semiconductor manufacturing method according to an embodiment of the present invention. Detailed Implementation

[0046] First, a semiconductor manufacturing apparatus and a semiconductor manufacturing method according to one embodiment of the present invention will be described. Although the present invention utilizes semiconductor materials and relates to the bonding of semiconductor materials with each other, semiconductor wafers, for example, can also be used as semiconductor materials. Examples of semiconductor wafers include, for example, Si wafers; however, they can also be made of glass or the like. Furthermore, the invention is not limited to semiconductor wafers; any material suitable for semiconductors, such as ceramics, is acceptable. In the following embodiments, for ease of explanation, a "semiconductor wafer" is used as an example of a semiconductor material; however, this is not intended to limit the present invention to semiconductor wafers.

[0047] like Figure 10 As shown, a semiconductor manufacturing apparatus 300 according to one embodiment of the present invention includes, for example, a control unit 310, a measurement unit 320, a planarization unit 330, a stripping unit 340, a transport unit 350, a bonding unit 360, an inspection unit 370, and an adjustment unit 380.

[0048] The semiconductor manufacturing apparatus 300 includes a semiconductor material bonding apparatus 34, which uses a plasma chamber or a vacuum chamber to perform bonding of semiconductor materials to each other. The plasma chamber or vacuum chamber may also be referred to as a bonding chamber, bonding apparatus, or bonding device.

[0049] Alternatively, the plasma chamber or vacuum chamber may have a pair of holding portions positioned opposite each other on the workpiece stage. When the semiconductor material is adsorbed by the holding portions, at least one of the holding portions can be moved by applying plasma between the semiconductor materials, thereby enabling the opposite semiconductor materials to be bonded together.

[0050] Preferably, before bonding the opposing semiconductor materials to each other, a silicon electrode is used on one side of the semiconductor materials, and a semiconductor wafer is used on the other side of the semiconductor materials. After performing the first plasma treatment on the silicon electrode and the other semiconductor wafer, the silicon electrode is replaced with a bonding wafer that is a semiconductor material, and a second plasma treatment is performed on the bonding wafer and the semiconductor wafer.

[0051] Alternatively, between the first plasma treatment and the second plasma treatment, the oxide film (second oxide film, native oxide film) formed on the surfaces of the opposing semiconductor materials can be removed by supplying etching gas.

[0052] The technical concept of the present invention described above is not limited to semiconductor material bonding devices, but is also applicable to semiconductor material bonding methods having the same features as described above.

[0053] The control unit 310 of the semiconductor manufacturing apparatus 300 controls each component. The measurement unit 320 performs the measurement process S1000 (described later). The planarization unit 330 performs the planarization process S1200 (described later). The stripping unit 340 performs the stripping process S1300 (described later). The transport unit 350 performs the transport process S1400 (described later). The bonding unit 360 performs the bonding process S1500 (described later). The inspection unit 370 performs the inspection process S1600 (described later). The adjustment unit 380 performs the adjustment process S1800 (described later).

[0054] [Technical Concept of the Invention] The present invention relates to a semiconductor manufacturing apparatus or method that uses a semiconductor material bonding device to bond a pair of opposing semiconductor materials together in a vacuum environment (or a plasma environment) by means of RF high-frequency ion flow and Coulomb force (also known as "electrostatic force", hereinafter the same) inside a vacuum chamber where plasma processing is performed on semiconductor materials, thereby improving the bonding strength between them, and includes the following semiconductor manufacturing process.

[0055] Specifically, such as Figure 11 As shown, the semiconductor manufacturing process of this embodiment includes, for example, the following steps.

[0056] Measurement step S1000 is used to measure the first oxide film formed on the surface of a semiconductor material.

[0057] If an abnormality occurs in the measurement process S1000 (S1100: Yes), a planarization process S1200 is performed to remove and planarize at least a portion of the first oxide film on the surface of the semiconductor material before proceeding to the stripping process S1300.

[0058] If there are no abnormalities in the measurement process S1000 (S1100: No), the first oxide film is peeled off in the stripping process S1300 by isotropic etching or plasma etching based on the measurement results of the first oxide film.

[0059] S1400 is a process for transporting semiconductor materials in a high-vacuum environment, an inert environment, or an atmospheric pressure environment.

[0060] After the surface of the semiconductor material is activated by plasma treatment, within a predetermined time, ions are polarized on the surface of the semiconductor material in a high vacuum state or plasma environment. Two high-frequency power supplies 110 of different frequencies are applied to each of the opposing semiconductor materials, thereby removing the second oxide film formed on the surface of the semiconductor material in the transport process S1400. The ions move on the surface of the semiconductor material to bond the semiconductor materials together in the bonding process S1500.

[0061] After the bonding process S1500, there is an inspection process S1600 to check the bonding strength or porosity of the semiconductor material.

[0062] If the bonding strength or porosity of the semiconductor material in the inspection process S1600 does not meet the predetermined benchmark (S1700: No), an adjustment process S1800 is performed to adjust the processing parameters responsible for the processing performance of the stripping process S1300 or the bonding process S1500.

[0063] [Measurement Procedure] In the measurement step S1000, the first oxide film that has been formed on the surface of a pair of semiconductor materials is measured.

[0064] Here, preferably, a film thickness gauge based on X-ray fluorescence is used to measure the first oxide film at at least 13 points on the surface of the semiconductor material. Fluorescent X-rays are generated by irradiating the semiconductor material, which serves as a substrate, with these fluorescent X-rays being classified and analyzed to measure the thickness of the film, coating, etc. Specifically, when X-rays irradiate a substance (atoms), electrons located inside the substance are ejected due to the energy of the X-rays and fly to the outside of the substance. Electron vacancies are created inside the substance, and the atoms are in an unstable excited state, thus attracting electrons from the outside to the inside. At this time, energy is released as X-rays (fluorescent X-rays). Therefore, by measuring the wavelength and energy of these fluorescent X-rays, the type of atoms and the thickness of the coating can be detected and measured. During the detection and classification of fluorescent X-rays, the thicker the film, the stronger the detection intensity of the film components and the weaker the detection intensity of the substrate components. Furthermore, the measurement system is automated, allowing the operator to check the film thickness with high accuracy and speed by setting the measurement conditions and measurement positions.

[0065] The measurement step S1000 includes an optimization step that optimizes the processing parameters responsible for improving the processing performance of the subsequent stripping step S1300 or bonding step S1500 based on the measurement results of the first oxide film. Hereinafter, isotropic etching and plasma etching in the subsequent stripping step S1300 and plasma treatment in the subsequent bonding step S1500 are performed according to the optimized processing parameters. Furthermore, it is preferable that the optimized processing parameters are set by the control unit 310, which controls each part of the semiconductor control device 300; however, it is not limited to this.

[0066] Here, the processing parameters represent the processing functions of each part of the semiconductor manufacturing apparatus 300, and include at least one element such as etching time in isotropic etching or plasma etching, etching solution concentration, gas flow rate in plasma processing, applied output of high-frequency power supply 110, bonding temperature or bonding time of semiconductor material, and heat treatment temperature or heat treatment time of semiconductor material.

[0067] [Planarization process] If an abnormality exists in the measurement process S1000 (S1100: Yes), in other words, if the surface condition of the semiconductor material is poor (the first oxide film is thick), it is not transferred to the stripping process S1300, but instead the planarization process S1200 is performed.

[0068] In the planarization process S1200, as a preceding process, at least a portion (e.g., wall thickness in the thickness direction) of the first oxide film formed on the surface of the semiconductor material is removed by chemical mechanical polishing (CMP) (a technique that uses a chemical substance containing an abrasive and a grinding stone to polish and planarize the surface of a semiconductor material), thereby planarizing the surface. For example, the semiconductor material is held in place by a component called a carrier, and while a slurry containing chemicals and abrasive particles is flowing, the semiconductor material is brought into contact with and rotated on a polishing pad, thereby grinding the surface of the semiconductor material to a flat surface.

[0069] In CMP, grinding is performed using a slurry containing chemicals and abrasive particles, employing both chemical and mechanical action.

[0070] Chemical action (chemistry) Abrasive-based mechanical grinding is assisted by using chemicals to degrade and dissolve the grinding surface.

[0071] Mechanical action (machinery) The surface of a semiconductor material is polished by the physical action of abrasive particles contained in the slurry.

[0072] When the planarization process S1200 ends, the process proceeds again to the measurement process S1000, where the first oxide film formed on the surface of the semiconductor material is measured. If there is no abnormality in the measurement result of the first oxide film (S1100: No), the process proceeds to the stripping process S1300. Alternatively, if there is an abnormality in the measurement result of the first oxide film (S1100: Yes), the planarization process S1200 is repeated until the abnormality disappears.

[0073] [Stripping Process] In the stripping process S1300, the first oxide film is stripped away by isotropic etching or plasma etching based on an etching solution, using processing parameters optimized based on the measurement results of the first oxide film formed on the surface of the semiconductor material.

[0074] Isotropic etching is a wet etching process, in which the surface of the semiconductor material is cleaned using at least one etching solution. Cleaning is preferably performed using hydrofluoric acid (HF), but is not limited to this method. Preferably, after cleaning with the etching solution, the surface of the semiconductor material is cleaned with pure water.

[0075] Plasma etching uses a gas that has become a plasma in a vacuum chamber to remove the first oxide film from the surface of a semiconductor material. Plasma etching is a processing technique that uses a gas in a plasma state and its active particles to remove material.

[0076] [Conveying Process] In the transport process S1400, semiconductor materials are transported in a high-vacuum atmosphere, an inert atmosphere, or an atmospheric pressure atmosphere. "High-vacuum atmosphere" refers to, for example, an atmosphere with a pressure below 10... -1 Pa or 10 -6 A pressure environment of Pa or higher. "Inert environment gas" refers to a purified environment containing inert gases (N2 (nitrogen), Ar (argon)). "Atmospheric pressure environment gas" refers to an atmospheric pressure environment in which the semiconductor material is exposed to air. Therefore, during the transport of semiconductor materials in a high-vacuum environment gas or an inert environment gas, the thickness of the second oxide film (natural oxide film, thin film) formed on the surface of the semiconductor material can be controlled in a predetermined manner. Furthermore, although a second oxide film (natural oxide film, thin film) of a predetermined thickness is formed on the surface of the semiconductor material during the transport of semiconductor materials in an atmospheric pressure environment gas, the second oxide film (natural oxide film, thin film) can be removed as described later.

[0077] [Jointing Process] In the bonding process S1500, after introducing gas into a vacuum device or vacuum chamber or plasma chamber and performing plasma treatment on the semiconductor material to activate the surface, within a predetermined time, ions are polarized on the surface of the semiconductor material under a high vacuum state. Two high-frequency power supplies 110 of different frequencies are applied to each of the opposing semiconductor materials, thereby removing the second oxide film (natural oxide film, thin film) formed on the surface of the semiconductor material in the transport process S1400. Ions move on the surface of the semiconductor material and bond the semiconductor materials together.

[0078] The bonding apparatus 34, which bonds semiconductor materials together using high-frequency ion flow bonding (RFIFB), performs the bonding process S1500; however, details of the bonding apparatus 34 will be described later. The bonding apparatus 34 has the function of a bonding portion 360 and performs the bonding process S1500.

[0079] [Inspection Procedure] In inspection step S1600, after bonding step S1500, the bonding strength or porosity of the semiconductor material is inspected. For example, preferably, a scanning acoustic microscope is used to inspect the bonding interface of the semiconductor material, and the acoustic signal is analyzed to quantitatively evaluate the bonding strength or porosity of the semiconductor material. By quantitatively evaluating the bonding strength or porosity of the semiconductor material, the bonding accuracy of the semiconductor material can be confirmed.

[0080] In the bonding process S1500, it is preferable to monitor the emission spectrum in real time during plasma activation to obtain the concentration of active particles and detect the plasma activity. By detecting the plasma activity, the accuracy of the plasma treatment can be confirmed.

[0081] In the bonding process S1500, the bonding surfaces of at least two opposing semiconductor materials are bonded together in close contact. The bonding temperature at this time is preferably room temperature (e.g., 0–35°C) or 100–200°C.

[0082] Preferably, heat treatment is performed on the interconnected semiconductor materials. Types of heat treatment include, for example, batch-type, sheet-type RTA (sheet-type rapid thermal annealing), and laser annealing. Batch-type refers to the simultaneous heat treatment of multiple semiconductor materials at once, where the semiconductor materials are arranged in a quartz furnace core and heated from the outside by a heater. Sheet-type RTA refers to the method of rapidly heating semiconductor materials one sheet at a time using infrared lamps. Laser annealing refers to the method of heat-treating only the surface of the semiconductor materials by irradiating them with an ultraviolet laser.

[0083] [Adjustment Process] In the adjustment process S1800, if the quantitative evaluation of the bonding strength or porosity of the semiconductor material in the inspection process S1600 does not meet the predetermined criteria (S1700: No), the processing parameters responsible for the processing performance of the stripping process S1300 or the bonding process S1500 are adjusted. Specifically, the optimized processing parameters are adjusted. The adjusted processing parameters are, for example, stored in the control unit 310. Preferably, the components are adjusted according to the adjusted processing parameters. Figure 10 The outputs of each part of the semiconductor manufacturing apparatus 300 are controlled by feedback.

[0084] As described above, the semiconductor manufacturing apparatus and method of this embodiment provide an improved bonding process that combines wet etching and plasma activation. By performing steps such as measuring the oxide film thickness of the semiconductor material, wet etching or plasma etching, plasma activation, bonding, interface inspection of the semiconductor material, and feedback control of processing parameters, it is possible to simultaneously remove contaminants from the surface of the semiconductor material and activate the surface, thereby improving process stability and reproducibility.

[0085] The bonding method based on this embodiment, which combines wet etching and plasma activation, has the following beneficial effects.

[0086] 1. It can efficiently remove oxide film and contaminants from the material surface while improving surface activity.

[0087] 2. Wet etching removes contaminants and oxide layers through selective reactions, thereby maintaining active groups (such as amino groups) and improving bonding strength.

[0088] 3. Enhanced plasma activation increases the surface hydroxyl density and promotes water molecule diffusion, thereby enabling low-temperature bonding. As a result, it can reduce thermal stress and material deformation and damage caused by heat.

[0089] 4. By measuring the oxide film thickness, bonding strength, and porosity, the process parameters are adjusted based on feedback to ensure the stability and reproducibility of the quality.

[0090] 5. It achieves interfaces with high bonding strength and low defect rate, and can withstand post-processing such as grinding and CMP. Therefore, it can be applied to the bonding of materials with large widths, such as Si, quartz glass, SiC, and Cu.

[0091] The following describes in detail the semiconductor manufacturing apparatus and the bonding apparatus used in the semiconductor manufacturing method of this embodiment. The bonding apparatus 34 bonds semiconductor materials together by, for example, high-frequency ion flow bonding (RF Ion Flow Bonding, (RFIFB)). As an example of semiconductor material, a semiconductor wafer is described.

[0092] Semiconductor material that has undergone measurement process S1000, planarization process S1200, stripping process S1300, and transport process S1400 is fed into bonding apparatus 34. In the semiconductor material bonding apparatus 34, at least bonding process S1500 is performed.

[0093] [Structure of semiconductor wafer bonding device] like Figure 1 As shown, the semiconductor wafer bonding apparatus 34 of this embodiment bonds various wafers together. Furthermore, although the semiconductor wafers and bonding wafers are, for example, silicon wafers, they are not limited to silicon wafers. Semiconductor wafers of different materials currently used can also be used, or semiconductor wafers of different materials can be bonded together.

[0094] like Figure 1 As shown, the semiconductor wafer bonding apparatus 34 includes a chamber 14 serving as a housing, a first workpiece stage 36 located on the upper side of the chamber, and a second workpiece stage 38 located on the lower side of the chamber. A first holding portion 102 capable of holding a semiconductor wafer or an electrode (e.g., a silicon electrode) is disposed on the first workpiece stage 36. A second holding portion 104 capable of holding a semiconductor wafer or an electrode (e.g., a silicon electrode) is disposed on the second workpiece stage 38.

[0095] In addition, the chamber 14, which serves as the housing, or the connecting device 34, can also be referred to as a "vacuum chamber".

[0096] Here, the first workpiece stage 36 and the second workpiece stage 38, or the first holding part 102 and the second holding part 104, use electrostatic chucks or mechanical clamps, for example. Furthermore, electrostatic chucks 78 and 80 can also be used, for example. Figures 7 to 9 The electrostatic chuck shown is illustrated below. It will be described later. Figures 7 to 9 The structure of the electrostatic chuck 78 and 80.

[0097] In addition, such as Figure 2 As shown, a first holding part 102 and a first replacement device 106 are disposed on or near the first workpiece stage 36 on the upstream side of the direction of gravity. The first replacement device 106 is used to replace the semiconductor wafer W1 or silicon electrode G1 held by the first holding part 102 with semiconductor wafers or silicon electrodes held by other holding parts. Furthermore, it is preferable that the replacement process of silicon electrode G1 and semiconductor wafer W1 is performed in groups of silicon electrode G1 and the first holding part 102 holding silicon electrode G1, and in groups of semiconductor wafer W1 and the first holding part 102 holding semiconductor wafer W1, but it is not limited to this method.

[0098] The silicon electrode G1 is well known, and is also referred to as a dummy wafer or bare wafer.

[0099] The semiconductor wafer W1 held by the first holding part 102 includes, for example, a silicon electrode G1, a bonding wafer G2, etc.

[0100] The first changing device 106 is not limited to a structure disposed on the side of the first workpiece stage 36, but can be another structure of the joining device 34, or it can be disposed outside the joining device 34.

[0101] Alternatively, a second holding portion 104 and a second replacement device 108 may be disposed on or near the second workpiece stage 38 on the downstream side of the direction of gravity. The second replacement device 108 is used to replace the semiconductor wafer W2 or silicon electrode held by the second holding portion 104 with the semiconductor wafer W2 or silicon electrode held by other holding portions. Furthermore, it is preferable that the replacement process of the silicon electrode (not shown) and the semiconductor wafer W2 is performed in groups of silicon electrode (not shown) and the second holding portion 104 holding the silicon electrode (not shown), and in groups of semiconductor wafer W2 and the second holding portion 104 holding the semiconductor wafer W2, but it is not limited to this method.

[0102] The second replacement device 108 is not limited to a structure disposed on the side of the second workpiece stage 38, but can be another structure of the engagement device 34, or it can be disposed outside the engagement device 34.

[0103] Furthermore, it is preferred that the first replacement device 106 and the second replacement device 108 be configured with at least one of them.

[0104] Here, a robotic arm is used as an example of the first replacement device 106 and the second replacement device 108, but it is not limited to this.

[0105] The semiconductor wafer W1 held by the first holding portion 102 and the semiconductor wafer W2 held by the second holding portion 104 are, for example, Si wafers or SiC wafers, but are not limited to these.

[0106] like Figure 1 As shown, the high-frequency power supply 110 is connected to the first workpiece stage 36 and the second workpiece stage 38, respectively. The high-frequency power supply 110 has the capability to generate a high-frequency electric field (13.56MHz, 200W). By applying the high-frequency power supply 110, the electric field promotes the flow and diffusion of atoms, thereby improving the adhesion of the bonding surfaces of the semiconductor wafers W1 and W2.

[0107] Furthermore, in this embodiment or example, the structure in which the high-frequency power supply 110 is connected to both the first workpiece stage 36 and the second workpiece stage 38 envisions applying high-frequency voltages to both the semiconductor wafer W1 held on the first workpiece stage 36 side and the semiconductor wafer W2 held on the second workpiece stage 38 side. However, it is also possible to use the structure in which the high-frequency power supply 110 is connected to both the first workpiece stage 36 and the second workpiece stage 38 to apply a high-frequency voltage to only one of the semiconductor wafers. However, in this case, it is also possible to provide the high-frequency power supply 110 only for either the first workpiece stage 36 or the second workpiece stage 38.

[0108] The semiconductor wafer bonding apparatus 34 is equipped with an automatic pressure control device 112 (APC) capable of controlling the pressure inside the housing. The automatic pressure control device 112 controls the pressure inside the housing.

[0109] The semiconductor wafer bonding device 34 is equipped with a turbomolecular pump 114 (TMP). The turbomolecular pump 114 is, for example, composed of a rotor (moving blade) and a stator (fixed blade) with turbine-type blades.

[0110] Turbomolecular pump 114 is at 1×10 -6 High-performance pumps with high exhaust velocity under high vacuum conditions, such as Pa.

[0111] [The effect of semiconductor wafer bonding devices] The bonding between the individual wafers is further strengthened.

[0112] When semiconductor wafers with different plasma processing times are bonded together, the plasma processing time can be shortened by replacing the semiconductor wafer with the one requiring a shorter plasma processing time as the bonding wafer and replacing it with the silicon electrode. This enables short-time plasma processing for the bonding wafer. As a result, the quality degradation of semiconductor products (semiconductor devices, etc.) formed by bonding individual wafers can be prevented in a way that the surface of the bonding wafer does not deteriorate due to the long plasma processing time.

[0113] [Execution of the first plasma treatment] like Figure 1 as well as Figure 2 As shown, the semiconductor wafer bonding apparatus 34 has a parallel plate electrode structure. The interior of the housing is set to a vacuum. With a silicon electrode G1 (e.g., also called a "silicon wafer electrode") held by a first holding part 102 and a semiconductor wafer W2 (e.g., a silicon wafer) held by a second holding part 104, a first plasma treatment based on argon plasma is performed. Here, argon plasma is used inside the vacuum chamber 14 to sputter silicon from the silicon electrode G1, and the surface of the semiconductor wafer W2 held by the second holding part 104 is activated by the plasma, forming a silicon-based vapor-deposited film X on the surface of the semiconductor wafer W2. Furthermore, in the plasma-based surface activation of the semiconductor wafer W2, gases such as argon (Ar) or nitrogen (N2) are used, and the oxide film (second oxide film, native oxide film) and contaminants on the surface of the semiconductor wafer W2 are removed. The same effect is achieved in the second plasma treatment of the semiconductor wafer W2 surface activation.

[0114] Here, a high-frequency power supply 110 is used, thus reducing energy consumption and minimizing plasma damage to the surface of the semiconductor wafer W2. Furthermore, a thin vapor-deposited film X is formed on the semiconductor wafer W2 by sputtering a silicon electrode G1, which serves as a silicon wafer electrode.

[0115] The first plasma treatment can also be omitted (the following implementation methods and examples are the same). In other words, instead of using the upper silicon electrode (semiconductor wafer), the upper and lower semiconductor wafers can be directly grouped together, and plasma can be generated by applying a high-frequency power supply for the first time, thereby bonding the upper and lower semiconductor wafers together. In particular, when the oxide film of the semiconductor wafer is removed by wet etching before being placed in the device, the oxide film can be removed by a single plasma treatment without dividing the plasma treatment into two parts.

[0116] Isotropic etching Furthermore, between the first plasma treatment and the subsequent second plasma treatment, an isotropic etching process can be performed to remove the oxide film formed on the surface of the semiconductor wafer W2 by supplying an etching gas such as SF6. After the isotropic etching process, the plasma gas is switched (gas replacement) to argon gas, and wafer surface activation is performed in the second plasma treatment.

[0117] [Execution of the second plasma treatment] Next, the silicon electrode G1 is replaced with a bonding wafer G2, which is a semiconductor wafer W1. Although the bonding wafer G2 refers to a semiconductor wafer such as a silicon wafer, it is not limited to silicon. The replacement process of the silicon electrode G1 and the bonding wafer G2 is performed, for example, by replacing a component that is a group of silicon electrode G1 and a first holding part 102 that holds the silicon electrode G1 with another component that is a group of bonding wafer G2 and a first holding part 102 that holds the bonding wafer G2. Alternatively, the first holding part 102 can be used in common in this state, and the bonding wafer G2 can be held in the first holding part 102 after the silicon electrode G1 has been separated from the first holding part 102. During and after the replacement process, plasma treatment is performed using a gas such as argon (Ar) or nitrogen (N2) to activate the surfaces of the bonding wafer G2 held by the first holding part 102 and the semiconductor wafer W2 held by the second holding part 104. The oxide films (second oxide film, native oxide film) and contaminants on the surfaces of the bonding wafer G2 held by the first holding part 102 and the semiconductor wafer W2 held by the second holding part 104 are removed respectively. Furthermore, while the application of the high-frequency power supply 110 continues, the supply of gases such as argon (Ar) or nitrogen (N2) is stopped, and the automatic pressure control device 112 is driven by high output to achieve a vacuum level (e.g., 1×10⁻⁶) inside the chamber 14 within approximately 10 seconds. -6 Pa (pascal)). At this time, although the plasma disappears, the surface ions of the bonding wafer G2 held by the first holding part 102 and the semiconductor wafer W2 held by the second holding part 104 are polarized.

[0118] The time required to achieve a vacuum inside chamber 14 is not limited to approximately 10 seconds; it can be set to, for example, a range of 1 to 10 seconds. The shorter the time required to achieve a vacuum inside chamber 14, the better.

[0119] Here, the time to achieve a vacuum inside chamber 14 can be appropriately set within the range of 0.1 seconds to 30 seconds, such as 0.1 seconds to 10 seconds.

[0120] [Execution of wafer bonding processes] In this state, for example Figure 3 As shown, the first workpiece stage 36 and / or the second workpiece stage 38 move in a manner that brings them closer together, thereby shortening the separation distance between the bonding wafer G2 held by the first holding portion 102 and the semiconductor wafer W2 held by the second holding portion 104. Due to the Coulomb force (electrostatic force) generated on their surfaces, the interatomic distance between the wafers becomes closer, eventually leading to contact. At this time, the first workpiece stage 36 can also be fixed, and the second workpiece stage 38 can be moved in a manner that brings it closer to the side of the first workpiece stage 36. Alternatively, the second workpiece stage 38 can be fixed, and the first workpiece stage 36 can be moved in a manner that brings it closer to the side of the second workpiece stage 38. Although the initial bonding force between the bonding wafer G2 held by the first holding portion 102 and the semiconductor wafer W2 held by the second holding portion 104 becomes stronger, by continuing to apply the high-frequency power supply 110 after the opposing wafers G2 and W2 have made contact, the RF high frequency flows between the surfaces of each wafer G2 and W2, and ion flow (atomic diffusion) is generated due to the electric field, thereby making the bonding even stronger.

[0121] Here, the time from the attainment of a vacuum level inside chamber 14 to the rapid bonding process following the cessation of plasma is preferably in the range of 0.1 seconds to 30 seconds. More preferably, it can also be set in the range of 0.1 seconds to 10 seconds. Here, 0.1 seconds to 10 seconds means bonding the semiconductor wafers together while the plasma is applied and the process is performed under high vacuum. In this case, it includes bonding the semiconductor wafers together while maintaining the plasma state.

[0122] In addition, the total time from the time until the vacuum level is achieved inside chamber 14 and the time from the time the vacuum level is achieved until the semiconductor wafers are bonded to each other can be set to a range of 0.1 seconds to 30 seconds, for example, 0.1 seconds to 10 seconds.

[0123] The semiconductor wafer manufacturing method based on this embodiment is an industry-first manufacturing method, called high-frequency ion flow bonding (RF Ion Flow Bonding, RFIFB).

[0124] Furthermore, although the plasma processing for the bonding wafer G2 held by the first holding part 102 was performed only once, the plasma processing for the semiconductor wafer W2 held by the second holding part 104 was performed only twice.

[0125] Furthermore, it is preferable to replace the first holding portion 102 and the silicon electrode G1 held by the first holding portion 102 with another first holding portion 102 and a bonding wafer G2 held by the first holding portion 102, and then bond the bonding wafer G2 held by the first holding portion 102 and the semiconductor wafer W2 held by the second holding portion 104. However, this is not a limitation. Furthermore, it is preferable that the first holding portion 102 is located upstream (upper) in the direction of gravity, and the second holding portion 104 is located downstream (lower) in the direction of gravity.

[0126] On the other hand, as an existing technology, there is atomic diffusion bonding (ADB). Although atomic diffusion bonding refers to a technology called room temperature activation bonding, there are known technologies that utilize the activation of semiconductor wafer surfaces based on Ar beams or Si sputtering. However, the bonding of the activated semiconductor wafers to each other requires several minutes.

[0127] Generally speaking, the activation state of the semiconductor wafer surface deteriorates in a short period of time, therefore, rapid vacuuming and bonding are required.

[0128] Therefore, according to this embodiment, after the wafer surface is activated by plasma treatment, for example, after about 10 seconds, a vacuum is applied by a turbomolecular pump 114 until pressure is reached, and the individual wafers are bonded together, thereby enabling atomic diffusion bonding that reduces the surface activation decay after wafer surface activation to the limit.

[0129] That is, in this embodiment, a structure (e.g., electrostatic chuck, mechanical clamp, etc.) is used to hold the wafer in a vertical position by means of a parallel plate plasma electrode (e.g., vertically oriented). One of the electrodes is configured to be replaceable with the bonded wafer.

[0130] According to this embodiment, microparticles appear in the parallel-plate wafer plasma due to the roughness of the electrode surface caused by sputtering of the silicon electrode G1, forming voids. The surface of the silicon electrode G1 is roughened by this sputtering, and further, silicon is sputtered again and grows, peeling off from the surface of the silicon electrode G1. Although the peeled silicon particles undergo Brownian motion and continue to float, contributing to the formation of void defects in the wafer, in this embodiment, since the silicon electrode G1 (e.g., a bare Si wafer) is used in the initial stage instead of the bonding wafer G2, damage caused by voids in the bonding wafer G2 can be reduced. This avoids quality degradation and deterioration of the semiconductor product obtained due to the bonding of the bonding wafer G2 and the semiconductor wafer W2. Furthermore, it is preferable to replace the silicon electrode G1 each time or periodically over multiple times.

[0131] As described above, due to the surface activation of the wafer used on one or both sides of the parallel plate and the soft sputtering generated by the application of the high-frequency power supply 110 to the silicon electrode G1, an evaporation film X, which is an extremely thin high-purity Si film, is formed on the processed surface of the semiconductor wafer W2 opposite to the silicon electrode G1.

[0132] In a plasma environment, the surface of a semiconductor wafer W2 is knocked by ions, exposing a clean surface. After the surface of the semiconductor wafer W2 is cleaned, a high-frequency power supply 110 of a different frequency is applied to the upper silicon electrode G1 (e.g., the silicon wafer). Hereby, tiny Si particles fly out from the silicon electrode G1 and deposit Si on the surface of the lower semiconductor wafer W2, thereby forming a vapor-deposited film X as a thin film formed by vapor deposition.

[0133] Furthermore, this step can be omitted when a thin film formed by vapor deposition is not required.

[0134] Next, during the bonding of the bonding wafer G2 and the semiconductor wafer W2, with the bonding wafer G2 fixed to, for example, an upper (upper side) electrode, the surface of the bonding wafer G2 is activated using a gas such as argon (Ar) or nitrogen (N2). At this time, the surface of the semiconductor wafer W2, which is disposed at the lower (lower side), is also activated. Thus, the surface activation of the bonding wafer G2 is performed only once, and the surface activation of the semiconductor wafer W2 is performed only twice.

[0135] Inside chamber 14, which serves as a housing, plasma pressure can be applied between opposing electrodes, and the system is designed to maintain plasma pressure under high vacuum to activate the surfaces of the two wafers (bonded wafer G2 and semiconductor wafer W2). As an electrode structure, magnetrons, ICP coils, etc., can be added to maintain plasma under high vacuum.

[0136] Furthermore, it is preferable that the separation distance between the bonding wafer G2, positioned on the upper side after plasma surface activation, and the semiconductor wafer W2, positioned on the lower side, is narrow. Furthermore, it is preferable that the separation distance between the silicon electrode G1, positioned on the upper side, and the semiconductor wafer W2, positioned on the lower side, is narrow. As such, the separation distance can, for example, be in the range of 10 μm or more and 50 μm or less when pressure changes are performed in a high vacuum after plasma processing between electrodes capable of uniformly applying plasma with minimal intervals. This is because by narrowing the separation distance, it is difficult for particles or the like to intrude between the electrodes from the outside, thus preventing particle mixing during the wafer bonding process and shortening the time from vacuum attainment to bonding.

[0137] After plasma treatment of the bonding wafer G2 and the semiconductor wafer W2, and activation of their respective surfaces, the supply of gases such as argon (Ar) or nitrogen (N2) is stopped, and the automatic pressure control device 112 is activated to maintain a vacuum pressure until, for example, the turbomolecular pump 114 (or cryo pump) reaches a vacuum pressure within approximately 10 seconds. During this time, the application of the high-frequency power supply 110 continues, but is not limited to this operation.

[0138] In this way, the oxide film (second oxide film, native oxide film) formed on the wafer surface during the transfer process of bonding wafer G2 and semiconductor wafer W2 is removed, and polarized ions (electrons) remain on the respective surfaces of bonding wafer G2 and semiconductor wafer W2. When bonding wafer G2 and semiconductor wafer W2 are joined in a high vacuum, the atomic motion is increased due to the charged ion potential on the respective surfaces of bonding wafer G2 and semiconductor wafer W2, and atoms easily move temporarily on the respective surfaces of bonding wafer G2 and semiconductor wafer W2. Furthermore, even with the application of the high-frequency power supply 110 after bonding wafer G2 and semiconductor wafer W2, atomic motion is increased, and atoms easily move temporarily on the respective surfaces of bonding wafer G2 and semiconductor wafer W2. Through the above synergistic effect, atoms are reconfigured to a more stable energy state. In particular, when the respective surfaces of bonding wafer G2 and semiconductor wafer W2 are activated, atoms are easily reconfigured to a further optimal position. As a result, the energy on the surfaces of the bonding wafer G2 and the semiconductor wafer W2 is minimized, thereby improving the bonding strength between the bonding wafer G2 and the semiconductor wafer W2.

[0139] Generally speaking, regarding the time from the plasma treatment of the semiconductor wafers to the bonding of the semiconductor wafers, when the transfer time of moving the semiconductor wafers from the plasma treatment chamber to the bonding chamber, the plasma treatment time for the second semiconductor wafer, and the transport time are added together, a waiting time of several minutes is generated. During this period, the surface of the semiconductor wafer deteriorates by about 10% to 50% of its activity immediately after the plasma treatment (attenuation rate 10% to 50%).

[0140] In contrast, in this embodiment, instantaneous bonding is performed immediately after the plasma treatment of the bonding wafer G2 and the semiconductor wafer W2. Therefore, the bonding wafer G2 and the semiconductor wafer W2 are bonded while the activation state of the respective surfaces of the bonding wafer G2 and the semiconductor wafer W2 is maintained at a decay rate of 1% or less. As a result, the bonding between the bonding wafer G2 and the semiconductor wafer W2 becomes strong.

[0141] voidless bonding The voids are mostly composed of Si sputtered flakes and particles within the chamber. Frequent replacement of the silicon electrode G1 eliminates the flakes and particles formed by sputtering, reducing the voids. Furthermore, in vacuum plasma, organic matter that serves as seeds for forming voids between the bonding wafer G2 and the semiconductor wafer W2 is removed by the plasma. Additionally, plasma is used to prevent the intrusion of organic matter from the outside into the wafer-to-wafer space. This achieves void-free bonding.

[0142] [Characteristics of high-frequency ion flow bonding] High-frequency ion flow bonding has the following characteristics.

[0143] (1) It can realize the surface activation and cleaning of wafers by plasma treatment and instantaneous vacuum degree management and control.

[0144] (2) It can enhance the initial bonding force between wafers that utilize Coulomb force (electrostatic force).

[0145] (3) The use of high-frequency power supply (RF high frequency) promotes the flow and diffusion of atoms, thereby improving the bonding strength between wafers.

[0146] [High-frequency ion flow bonding process] An example of a process for high-frequency ion flow bonding treatment is described.

[0147] like Figure 4 As shown, for example, a wafer is removed from the wafer cassette (S100) and wafer alignment is performed (S200). A wafer cleaning process is performed (S300), and the wafer is transported to a load-locking chamber (S400). Plasma treatment is performed on the wafer (S500), and the wafers are bonded together (S600). Then, the semiconductor product formed by bonding the wafers together is removed from the load-locking chamber (S700), and the semiconductor product is inspected (S800).

[0148] Here, the time required from the end of the plasma processing step (S500) for the wafer to the end of the bonding step (S600) is preferably, for example, 10 seconds or less. In the plasma processing step (S500) for the wafer, the oxide film (second oxide film, native oxide film) formed on the surface of the wafer by the wafer transport step (e.g., S400, etc.) is removed.

[0149] As described above, for example, auxiliary bonding can be controlled by using the Coulomb force of the ESC chuck to prevent misalignment when the opposing wafers are in contact with each other, and high-strength bonding can be achieved through high-frequency ion flow bonding technology.

[0150] (Example) Next, an example of high-frequency ion flow bonding will be described.

[0151] [Overview] (1) Two Si wafers are arranged in a vacuum by means of a parallel plate electrode structure.

[0152] (2) Use a replaceable silicon wafer electrode on one side and activate the surface of the silicon wafer electrode with argon plasma.

[0153] (3) Sputtering silicon that flies out from the surface of the silicon wafer electrode onto the surface of the semiconductor wafer opposite to the silicon wafer electrode.

[0154] [Process] (1) Surface activation based on silicon wafer electrodes Silicon is sputtered from the silicon wafer electrode by activating the surfaces of the silicon wafer electrode and the semiconductor wafer opposite the silicon wafer electrode through plasma treatment (first time) using argon plasma.

[0155] The silicon wafer electrode is replaced with a bonding wafer that becomes the bonding target.

[0156] An argon plasma treatment (second time) is performed on the bonding wafer and the semiconductor wafer, and after the gas supply is stopped, a vacuum is achieved, for example, within about 10 seconds, using a turbomolecular pump or the like.

[0157] Although the plasma disappears due to vacuuming, ions remain in a polarized state on the surfaces of the opposing laminated wafers and semiconductor wafers.

[0158] This removes the oxide film (second oxide film, native oxide film) formed on the surface of the semiconductor wafer during the semiconductor wafer transport process.

[0159] (2) Joining wafers together The Coulomb force (electrostatic force) is generated by the polarized ions that remain on the surfaces of the bonding wafer and the semiconductor wafer, causing the atoms of the wafers to come closer together and make contact.

[0160] RF high frequencies flow across the surface of each wafer, causing ion flow (atomic diffusion) due to the electric field, which enables strong bonding between the wafers.

[0161] Furthermore, although in wafer bonding, a high vacuum can be rapidly established after plasma application to achieve wafer bonding, it is also possible to bond wafers while plasma is being applied (by evacuating the vacuum during plasma excitation). In other words, wafers can be bonded in a state where the vacuum level does not change until the area where the plasma does not stand upright. The area where the plasma does not stand upright is generally around 1 Pa in the case of parallel plate plasma. In the case of magnetron and ICP plasma, it is 1 × 10⁻⁶ Pa. -2 Approximately. In the case of an argon ion beam, it is 1 × 10⁻� -6 about.

[0162] [Specific Structure] (1) Parallel plate plasma electrode structure (1-1) Wafer holding mechanism The structure is designed such that each wafer is held between the upper electrode and the lower electrode by using an electrostatic chuck or mechanical clamp as each holding part.

[0163] At least one of the electrodes on the upper and lower sides is configured to be replaceable with a wafer-mounted electrode. Preferably, the replacement operation is performed by a robotic arm or the like. Furthermore, it is preferable to replace the upper electrode with a wafer-mounted electrode. This helps to suppress the generation of particles caused by the roughness of the electrode surface.

[0164] (1-2) Electrode sputtering As electrodes, silicon electrodes made of SI bare wafers are used to control particles in the plasma process. This helps to suppress voids during the bonding of the laminated wafer and the semiconductor wafer.

[0165] (1-3) Surface activation of wafers on one or both sides facing each other By applying a high-frequency power supply to each electrode to perform soft sputtering, silicon ejected from the electrodes deposits on the surface of an opposing wafer, thereby forming a thin, high-purity Si film. Furthermore, the process of forming the high-purity Si film (silicon film) can be omitted if necessary.

[0166] Si (silicon) can also be sputtered by applying high-frequency power supplies of different frequencies to the upper electrodes.

[0167] (2) Fixing of the bonded wafer (2-1) Activation of wafer surfaces using gas Ar (argon) or N2 (nitrogen) gas is used to activate the surface of the wafer.

[0168] Plasma processing is performed simultaneously on two wafers positioned on the top and bottom sides in a single room.

[0169] (2-2) Plasma treatment Plasma processing can be performed in a high vacuum. Magnetrons or ICP coils can be used.

[0170] The structure between the electrodes is a uniform plasma structure that is not affected by external factors such as dust intrusion.

[0171] Plasma treatment can remove the oxide film (second oxide film, native oxide film) formed on the surface of a semiconductor wafer during the semiconductor wafer transport process.

[0172] (2-3) Vacuuming after plasma treatment Stop supplying argon and nitrogen to the room, activate the automatic pressure control device (APC), and use a turbomolecular pump to bring the room to a vacuum pressure in, for example, about 10 seconds.

[0173] The application of high-frequency power to the upper and lower electrodes can be continued depending on the product, or it can be stopped.

[0174] (2-4) Vacuum bonding (vacuum joining) Wafers with charged surfaces are brought close together and contacted using Coulomb force (electrostatic force). The use of Coulomb force increases the initial bonding strength between the wafers. Furthermore, the use of Coulomb force increases atomic movement, making it easier for atoms residing on the surfaces of each wafer to move.

[0175] By applying a high-frequency power supply to each electrode, atoms are reconfigured to a stable energy state, forming a strong bond between wafers.

[0176] (3) Instantaneous bonding of the plasma-treated wafers to each other (instantaneous bonding) After activation on the surface of each wafer, bonding is performed, for example, in about 10 seconds, thereby minimizing the degradation of the active state (activity) on the wafer surface.

[0177] In contrast, in conventional room-temperature activation bonding or atomic diffusion bonding, the bonding of wafers requires time. Specifically, if a waiting time of several minutes occurs, the active state is reduced by approximately 20%. Although this results in a temperature that degrades the quality of semiconductor wafer products, this temperature is resolved through the high-frequency ion flow bonding of this embodiment.

[0178] [In high vacuum (1×10)] -6 [Pascal's surface activation decay rate] The decay rate after 10 seconds in a vacuum: A(10 seconds) ≈ 99.83% A(10 seconds) = 99.83% = 0.17% decay rate. The decay rate after 2 minutes in a vacuum: A(2 minutes) ≈ 81.87% A(2 minutes) = 81.87% = 18.13% decay rate. As can be seen from the above, rapid vacuuming and wafer bonding are extremely important for maintaining the surface activation state, resulting in a strong bond between the wafers.

[0179] Next, the diffusion of ions in the high-frequency ion flow bonding process is analyzed. In this process, after plasma treatment and surface activation of silicon wafers in a vacuum, the wafers are bonded together while the gas is stopped and the surface activity is maintained at high speed in a vacuum, and a high-frequency power supply is applied.

[0180] [High-frequency ion flow bonding process] (1) Plasma surface activation Organic contaminants and oxides on the wafer surface are chemically removed by using plasma etching with fluorine-based gases.

[0181] Argon sputtering removes minute contaminants and uneven chemical residues that may remain.

[0182] While removing and cleaning the oxide film (second oxide film, natural oxide film) on the wafer surface through activation and plasma etching, the wafer surface is physically struck by argon sputtering, thereby further activating the atoms on the wafer surface and improving the bonding efficiency between wafers.

[0183] By purifying and activating the wafer surface to allow atoms to diffuse, a higher bonding strength between wafers can be achieved.

[0184] (2) Gas stoppage and high-speed vacuuming After plasma treatment of the wafer, the supply of argon and nitrogen is stopped, and a turbomolecular pump is used to achieve a high vacuum (1×10⁻⁶) at high speed. -6 Control is performed using the Pascal method.

[0185] By achieving a vacuum level within approximately 10 seconds, the active state of the wafer surface is maintained.

[0186] (3) The bonding of wafers to each other The wafers are joined together while polarized ions remain on the surface of each wafer.

[0187] Due to Coulomb force (electrostatic force), the distance between atoms on the surfaces of the wafers being approached becomes closer, thus strengthening the contact between the wafers.

[0188] (4) Application of high-frequency power supply After the wafers are joined together, a high-frequency power supply is applied to generate an electric field on the surface of each wafer.

[0189] The electric field causes ions to flow on the wafer surface, strengthening the bonds between atoms.

[0190] [Investigation of ion diffusion] (1) The state of ions on the wafer surface Activation based on plasma treatment puts the ions remaining on the wafer surface into a polarized state.

[0191] Because polarized ions are strongly adsorbed onto the wafer surface, they are easy to move.

[0192] In a high vacuum, the electrodes become insulated, and ions do not ionize (discharge) but remain on the wafer surface.

[0193] (2) High vacuum state In a high vacuum, the active state of the wafer surface is maintained, and recontamination of the wafer surface is suppressed.

[0194] (3) Effects of high-frequency power supply The high-frequency electric field (13.56MHz, 200W) has the following effects on the wafer surface and interface.

[0195] (3-1) Variation of surface potential An alternating electric field is generated on the wafer surface by using a high-frequency electric field. The resulting potential variation on the wafer surface causes the following effects.

[0196] Surface potential variation: A potential of ±200V varies at 13.56MHz, causing the movement of electrons and ions on the wafer surface.

[0197] The effects of high-frequency electric fields: the atoms and ions on the wafer surface are reconfigured through high-frequency electric fields (alternating electric fields).

[0198] (3-2) Promotion of atomic diffusion at the interface High-frequency electric fields have the following effects on the wafer surface.

[0199] Ion movement: Ions on the wafer surface can be easily moved by an alternating electric field. Interatomic changes: The interatomic forces between wafers change through alternating electric fields, resulting in stronger bonds.

[0200] The reduction in activation energy: The alternating electric field reduces the energy required for atomic diffusion, thus promoting the bonding between atoms.

[0201] [Mechanism of ion diffusion] The diffusion of ions is induced by the application of a high-frequency power source through the following mechanism.

[0202] Furthermore, by applying the high-frequency power supply only to one of the opposing wafers, it is possible to achieve all the effects described below: (1) ion movement induced by the electric field, (2) surface diffusion, and (3) enhancement of interatomic bonding. Therefore, by applying the high-frequency power supply to at least one of the opposing wafers, ion flow phenomena can be generated.

[0203] (1) Movement of ions caused by electric field induction Ions are attracted or expelled from the wafer surface by a high-frequency electric field. In this way, the ions move uniformly across the entire wafer surface.

[0204] (2) Surface diffusion Atoms are rearranged by the movement of ions on the wafer surface.

[0205] A more stable energy state is formed by reconfiguring the atoms on the wafer surface.

[0206] (3) Strengthening of interatomic bonds By reconfiguring the atoms on the wafer surface, the bonding between atoms is strengthened. This makes the bonding between wafers more robust.

[0207] After isotropic etching of the oxide films (first oxide film, initial oxide film) on the wafer surface, a high-frequency ion flow bonding process is employed, specifically involving a chemical reaction of the silicon wafer in a vacuum, activating the wafer surface through argon-based soft sputtering. Then, an oxide film (second oxide film, native oxide film) is formed on the wafer surface during wafer transport. However, by maintaining the activated state of the wafer surface in a high-vacuum environment, for example, at a high speed of approximately 10 seconds, the wafers are bonded together. Furthermore, by continuing to apply a high-frequency power supply between the wafers, the oxide films (second oxide film, native oxide film) are removed from the wafer surface through ion diffusion, and the movement of ions on the wafer surface strengthens the interatomic bonding. This results in a strong wafer-to-wafer bond. Thus, high-frequency ion flow bonding technology is highly effective in wafer bonding, simultaneously improving both the bonding strength and bonding quality between wafers.

[0208] Furthermore, frequent automatic electrode replacement and the use of bare wafers help reduce flakes and particles formed by Si sputtering, achieving void-free bonding. This significantly reduces the frequency of indoor cleaning, contributing to reduced maintenance time and increased system uptime.

[0209] [An example of a semiconductor manufacturing system] An example of how the semiconductor wafer bonding apparatus of this embodiment and the embodiment is assembled in a semiconductor manufacturing system will be described.

[0210] like Figure 5 as well as Figure 6 As shown, the semiconductor wafer bonding device 34 can be assembled into the semiconductor manufacturing system 200 to replace the plasma chamber 10 of the existing system. For example... Figure 5 as well as Figure 6 The semiconductor manufacturing system 200 shown is equipped with a semiconductor wafer bonding device 34, a vacuum bonding chamber 204, an atomic bonding chamber 206, a rotary cleaning chamber 208, a standby chamber 210, and a load locking chamber 212 around a central chamber 202 housing a robotic arm. A wafer transport mechanism 214 and multiple wafer cassettes 216 are arranged near the load locking chamber 212.

[0211] [An example of a wafer holding mechanism] Next, an example of a wafer holding mechanism used to hold each wafer will be described.

[0212] Regarding the semiconductor wafer bonding apparatus and bonding method of this embodiment, an electrostatic chuck 80 (see reference) is used, which serves as a wafer holding mechanism and is disposed on the first worktable 36 and the second worktable 38 for holding semiconductor wafers W1 and W2. Figures 7 to 9 The structure can also be configured to replace the electrostatic chuck 80 (see reference). Figures 7 to 9 The structure of the first holding part 102 and the second holding part 104.

[0213] Figure 7 as well as Figure 8 The diagram illustrates the structures of a unipolar electrostatic chuck 78 and a bipolar electrostatic chuck 80. Each electrostatic chuck 78 and 80 is made of an insulating material such as aluminum oxide (Al2O3) or aluminum nitride (AlN). This results in a structure with electrodes embedded within the insulator, allowing the adsorption of semiconductor wafers W1, W2, etc., by applying a voltage to these electrodes.

[0214] Figure 7The unipolar electrostatic chuck 78 shown is unipolar and has a base plate 82 and an internal electrode (also called an electrode sheet or polyimide film electrode layer) 84 disposed on the base plate 82. The base plate 82 applies a voltage, either positive or negative, to the internal electrode 84. For example, when a positive voltage is applied to the internal electrode 84, negative charges move to the surface of the object to be attracted 86, and the object to be attracted 86 is attracted to the unipolar electrostatic chuck 78. Conversely, when a negative voltage is applied to the internal electrode 84, positive charges move to the surface of the object to be attracted 86, and the object to be attracted 86 is attracted to the unipolar electrostatic chuck 78.

[0215] Figure 8 The electrostatic chuck 80 shown is bipolar and has a base plate 88 and internal electrodes (also called electrode sheets or polyimide film electrode layers) 90 disposed on the base plate 88. The base plate 88 applies both positive and negative voltages to the internal electrodes 90. For example, negative charges move to the surface of the object 92 opposite to the internal electrode 90 to which a positive voltage is applied, and positive charges move to the surface of the object 92 opposite to the internal electrode 90 to which a negative voltage is applied, and the object 92 is attracted to the bipolar electrostatic chuck 80.

[0216] An example is given of the use of bipolar electrostatic chucks in semiconductor wafer bonding apparatus. For example... Figure 9 As shown, preferably, the electrostatic chucks 80 are respectively built into the first workpiece stage 36 and the second workpiece stage 38. For ease of explanation, the electrostatic chuck 80 built into the first workpiece stage 36 is referred to as the first electrostatic chuck 80A, and the electrostatic chuck 80 built into the second workpiece stage 38 is referred to as the second electrostatic chuck 80B. Each electrostatic chuck 80A and 80B is, for example, bipolar. The first workpiece stage 36 and the second workpiece stage 38 are arranged opposite each other, so that each electrostatic chuck 80A and 80B is arranged in a mutually opposite position, thereby forming a pair of electrostatic chucks.

[0217] The first electrostatic chuck 80A has a first base plate 88A and a first internal electrode 90A disposed on the first base plate 88A. The first internal electrode 90A is grounded. The first base plate 88A controls the voltage applied to the first internal electrode 90A. The first internal electrode 90A is composed of two or more adjacent first unit electrodes 91A, and positive (positive) and negative (negative) voltages that are reversed are applied to each other on adjacent first unit electrodes 91A.

[0218] The second electrostatic chuck 80B is positioned opposite the first electrostatic chuck 80A. The second electrostatic chuck 80B has the same structure as the first electrostatic chuck 80A, including a second base plate 88B and a second internal electrode 90B disposed on the second base plate 88B. The second internal electrode 90B is grounded. The second base plate 88B controls the voltage applied to the second internal electrode 90B. The second internal electrode 90B is composed of two or more adjacent second unit electrodes 91B, and positive (positive) and negative (negative) voltages that are reversed are applied to each of the adjacent second unit electrodes 91B.

[0219] Here, the first unit electrode 91A on the first electrostatic chuck 80A side and the second unit electrode 91B on the second electrostatic chuck 80B side opposite to the first unit electrode 91A are controlled by the first base plate 88A and the second base plate 88B to apply voltages of opposite polarities (positive and negative), respectively. Therefore, the first unit electrode 91A in the first internal electrode 90A of the first electrostatic chuck 80A, where a negative voltage is applied, is positioned opposite to the second unit electrode 91B in the second unit electrode 90B of the second electrostatic chuck 80B, where a positive voltage is applied. Similarly, the first unit electrode 91A in the first internal electrode 90A of the first electrostatic chuck 80A, where a positive voltage is applied, is positioned opposite to the first unit electrode 91B in the second unit electrode 90B of the second electrostatic chuck 80B, where a negative voltage is applied.

[0220] As a general principle of the bipolar electrostatic chuck 80, with semiconductor wafer W1 placed on the first electrostatic chuck 80A and semiconductor wafer W2 placed on the second electrostatic chuck 80B, positive and negative voltages are applied to the first internal electrode 90A and the second internal electrode 90B, respectively. This causes the positive and negative charges on the semiconductor wafers W1 and W2 to move (inductively polarize) by attracting each other relative to the respective internal electrodes 90A and 90B opposite to the aforementioned components. As a result, an adsorption force is generated between the first internal electrode 90A and the semiconductor wafer W1, and between the second internal electrode 90B and the semiconductor wafer W2, thereby fixing the semiconductor wafers W1 and W2 in a structure.

[0221] like Figure 7As shown, in a typical unipolar electrostatic chuck 78, a voltage is applied between the chuck (semiconductor wafer, etc.) 86 and the internal electrode 84 (also called a chuck or holding device), thereby generating a charge on the surface of the chuck (semiconductor wafer, etc.) 86. Specifically, if the internal electrode 84 is positively charged, the surface of the chuck (semiconductor wafer, etc.) 86 opposite it is negatively charged; conversely, if the internal electrode 84 is negatively charged, the surface of the chuck (semiconductor wafer, etc.) 86 opposite it is positively charged. This principle is... Figure 8 The same applies to the bipolar electrostatic chuck 80 shown.

[0222] Typically, semiconductor wafers W1 and W2, after being activated by plasma treatment, have a negative (or sometimes positive) surface charge. Therefore, when semiconductor wafers W1 and W2 with negative (or positive) surfaces are to be bonded together, they repel each other. In this state, when semiconductor wafers W1 and W2 are forcibly bonded together, they repel each other, causing positional misalignment and resulting in reduced bonding precision.

[0223] like Figure 9 As shown, by making the surface of one semiconductor wafer W1, which is facing each other, positively (negatively) charged, and the surface of the other semiconductor wafer W2, negatively (positively) charged, a state is intentionally formed in which the semiconductor wafers to be bonded are charged with opposite polarities. This, for example, utilizes the van der Waals forces (intermolecular forces) and Coulomb forces (electrostatic forces) generated between them to enhance the adsorption force and assist in the bonding of the semiconductor wafers W1 and W2.

[0224] Generally speaking, when objects with negative static electricity (or objects with positive static electricity) are brought together, they will repel each other. This is one of the fundamental properties of static electricity. Charges of the same sign (e.g., two objects with negative charges) repel each other, while charges of opposite polarities (e.g., one is positive and the other is negative) attract each other.

[0225] Semiconductor wafers W1 and W2, charged with opposite polarities, are subjected to surface voltage control that does not result in discharge. Furthermore, through contact between the positively charged (or W2) semiconductor wafer W1 and the negatively charged (or W1) semiconductor wafer W2, electrons move from the positively charged (or W2) to the negatively charged (or W1) semiconductor wafer, thereby neutralizing the charge. This weakens the effect of electrostatics while increasing the bonding and adsorption forces during bonding.

[0226] In a typical bipolar electrostatic chuck, the surface of a semiconductor wafer is kept in electrical equilibrium by applying the same voltage (absolute value) to both the positive (positive) and negative (negative) sides.

[0227] In contrast, instead of applying the same magnitude (absolute value) of voltage to the positive and negative sides, different magnitudes of voltage are intentionally applied to deliberately disrupt the balance between positive and negative. In other words, from the perspective of the magnitude of the applied voltage, a non-equilibrium state is created between the positive and negative voltages.

[0228] like Figure 9 As shown, for example, a voltage of -400 volts is applied to the first unit electrode 91A on one side of the first internal electrode 90A constituting the first electrostatic chuck 80A, and a voltage of +500 volts is applied to the first unit electrode 91A on the other side. Simultaneously, a voltage of +400 volts is applied to the second unit electrode 91B on one side of the second internal electrode 90B constituting the second electrostatic chuck 80B, and a voltage of -500 volts is applied to the second unit electrode 91B on the other side.

[0229] Here, the first unit electrode 91A of the first internal electrode 90A of the first electrostatic chuck 80A, on the side where a voltage of -400 volts is applied, and the second unit electrode 91B of the second internal electrode 90B of the second electrostatic chuck 80B, on the side where a voltage of +400 volts is applied, are positioned opposite each other. Although the polarities of the voltages are different, the absolute values ​​of the voltages applied are the same.

[0230] Furthermore, the first unit electrode 91A of the first internal electrode 90A of the first electrostatic chuck 80A, on the other side where a voltage of +500 volts is applied, and the second unit electrode 91B of the second internal electrode 90B of the second electrostatic chuck 80B, on the other side where a voltage of -500 volts is applied, are positioned opposite each other. Although the polarities of the voltages are different, the same magnitude of voltage is applied as an absolute value.

[0231] In this way, the surface of the semiconductor wafer W1, which is opposite to the first electrostatic chuck 80A, becomes charged with a difference of +500 volts and -400 volts, i.e., +100 volts, from the first electrostatic chuck 80A. Similarly, the surface of the semiconductor wafer W2, which is opposite to the second electrostatic chuck 80B, becomes charged with a difference of +400 volts and -500 volts, i.e., -100 volts, from the second electrostatic chuck 80B.

[0232] At this time, floating charges are generated on the surfaces of semiconductor wafers W1 and W2. By applying opposite voltages to the opposing internal electrodes 90A and 90B (or unit electrodes), a strong Coulomb force attracts and bonds the semiconductor wafers W1 and W2 together. As a result, the generation of bubbles during the bonding of semiconductor wafers W1 and W2 can be suppressed.

[0233] Preferably, semiconductor wafers W1 and W2, held by a pair of electrostatic chucks 80A and 80B, are placed in an environment with a separation distance between them of 10 μm and 50 μm (microspace). In the microspace, semiconductor wafers W1 and W2 are bonded together while maintaining their planar orientation.

[0234] Furthermore, this embodiment and examples illustrate one aspect of the invention, and the invention is not limited thereto. Differences in the degree of design modifications to this embodiment and examples are naturally included within the scope of the technical concept of the invention.

[0235] Symbol Explanation Room 14 16 Workpiece Stage 34. Semiconductor wafer bonding apparatus 36 First workpiece stage 38 Second workpiece stage 78 Unipolar electrostatic chuck 80 Bipolar electrostatic chuck 80A First Electrostatic Chuck 80B Second Electrostatic Chuck 82 Base plate 84 Internal Electrodes 86 Adsorbed body 88 base plate 88A First Base Plate 88B Second Base Plate 90 Internal Electrode 90A First Internal Electrode 90B Second Internal Electrode 91A First Unit Electrode 91B Second Unit Electrode 92 Adsorbed body 102 First Maintenance Section 104 Second Maintenance Section 106 First Replacement Device 108 Second Replacement Device 110 High-Frequency Power Supply 112 Automatic pressure control device 114 turbomolecular pump 200 Semiconductor Manufacturing Systems 202 Central Room 204 Vacuum Joint Chamber 206 atomic bonding chamber 208 Rotary Washing Chamber 210 Waiting Room 212 Load Locking Chamber 214 Wafer Delivery Mechanism 216 wafer box 300 Semiconductor Manufacturing Equipment 310 Control Department 320 Measurement Department 330 Planarization section 340 Peeling Section 350 Conveying Unit 360° joint 370 Inspection Department 380 Adjustment Section G1 Silicon wafer (semiconductor material) G2 bonding wafer (semiconductor material) W1 Semiconductor wafer (semiconductor material) W2 Semiconductor wafer (semiconductor material).

Claims

1. A semiconductor manufacturing apparatus, using a semiconductor material bonding device for bonding a pair of semiconductor materials, wherein, The first oxide film formed on the surface of the semiconductor material was measured. Based on the measurement results of the first oxide film, the first oxide film is removed by isotropic etching or plasma etching. For the semiconductor material that has undergone a transport process in a predetermined environment, the bonding device performs plasma treatment to activate the surface of the semiconductor material, polarizes the ions on the surface of the semiconductor material under a high vacuum state within a predetermined time, applies a high-frequency power supply to the semiconductor material, thereby removing the second oxide film formed on the surface of the semiconductor material during the transport process, and the ions move on the surface of the semiconductor material to bond the semiconductor materials together.

2. The semiconductor manufacturing apparatus as claimed in claim 1, wherein, Based on the measurement results of the first oxide film, the processing parameters related to the processing performance of the isotropic etching, the plasma etching, and the plasma treatment will be optimized responsibly.

3. The semiconductor manufacturing apparatus as claimed in claim 1 or 2, wherein, If there is an anomaly in the measurement results of the first oxide film, at least a portion of the first oxide film on the surface of the semiconductor material is removed and planarized before peeling off the first oxide film.

4. The semiconductor manufacturing apparatus as claimed in claim 2, wherein, After the semiconductor materials are bonded, the bonding strength or porosity of the semiconductor materials is checked.

5. The semiconductor manufacturing apparatus as claimed in claim 4, wherein, If the bonding strength or porosity of the semiconductor material does not meet a predetermined benchmark, the processing parameters are adjusted.

6. The semiconductor manufacturing apparatus as claimed in claim 1 or 2, wherein, The first oxide film was measured using a film thickness gauge based on X-ray fluorescence, selecting at least 13 points on the surface of the semiconductor material.

7. The semiconductor manufacturing apparatus as claimed in claim 1 or 2, wherein, In the plasma treatment of the bonding device, the emission spectrum is monitored in real time during plasma activation to obtain the concentration of active particles and detect the plasma activity.

8. The semiconductor manufacturing apparatus as claimed in claim 4 or 5, wherein, In the inspection of the bonding strength or porosity of the semiconductor material, a scanning acoustic microscope is used to examine the bonding interface of the semiconductor material, and the acoustic signal is analyzed to quantitatively evaluate the bonding strength or porosity of the semiconductor material.

9. The semiconductor manufacturing apparatus as claimed in claim 2, wherein, The processing parameters include at least one element: etching time or etchant concentration in the isotropic etching or plasma etching, gas flow rate in the plasma processing or applied output of the high-frequency power supply, bonding temperature or bonding time of the semiconductor material, and heat treatment temperature or heat treatment time for the semiconductor material.

10. The semiconductor manufacturing apparatus as claimed in claim 1 or 2, wherein, The predetermined time is 0.1 to 10 seconds.

11. The semiconductor manufacturing apparatus as claimed in claim 1 or 2, wherein, The semiconductor materials that are bonded together are held by an electrostatic chuck that serves as a parallel plate plasma electrode.

12. A semiconductor manufacturing method, wherein, have: The measurement process includes measuring the first oxide film formed on the surface of the semiconductor material. In the stripping process, the first oxide film is stripped away by isotropic etching or plasma etching based on the measurement results of the first oxide film. A conveying process in which the semiconductor material is conveyed under a predetermined environment; In the bonding process, after the semiconductor material is activated by plasma treatment, ions are polarized on the surface of the semiconductor material under high vacuum within a predetermined time. A high-frequency power supply is applied to the semiconductor material to remove the second oxide film formed on the surface of the semiconductor material during the transport process. The ions move on the surface of the semiconductor material and bond the semiconductor materials together.

13. The semiconductor manufacturing method as described in claim 12, wherein, The measurement process includes an optimization process that optimizes the processing parameters responsible for the processing performance related to the isotropic etching, the plasma etching, and the plasma treatment based on the measurement results of the first oxide film.

14. The semiconductor manufacturing method as described in claim 12 or 13, wherein, A planarization process is provided that, in the event of an abnormality in the measurement process, at least a portion of the first oxide film on the surface of the semiconductor material is removed and planarized prior to the stripping process.

15. The semiconductor manufacturing method as described in claim 13, wherein, It includes an inspection step that checks the bonding strength or porosity of the semiconductor material after the bonding process.

16. The semiconductor manufacturing method as described in claim 15, wherein, An adjustment process is included to adjust the processing parameters when the bonding strength or porosity of the semiconductor material in the bonding process does not meet a predetermined benchmark.

17. The semiconductor manufacturing method as described in claim 12 or 13, wherein, In the measurement process, a film thickness gauge based on X-ray fluorescence is used to select at least 13 points on the surface of the semiconductor material for measurement.

18. The semiconductor manufacturing method as claimed in claim 12 or 13, wherein, In the bonding process, the emission spectrum is monitored in real time during plasma activation to obtain the concentration of active particles and detect the plasma activity.

19. The semiconductor manufacturing method as described in claim 15 or 16, wherein, In the inspection process, a scanning acoustic microscope is used to examine the bonding interface of the semiconductor material, and the acoustic signals are analyzed to quantitatively evaluate the bonding strength or porosity of the semiconductor material.

20. The semiconductor manufacturing method of claim 13, wherein, The processing parameters include at least one element: etching time or etchant concentration in the isotropic etching or plasma etching, gas flow rate in the plasma processing or applied output of the high-frequency power supply, bonding temperature or bonding time of the semiconductor material, and heat treatment temperature or heat treatment time for the semiconductor material.

21. The semiconductor manufacturing method as described in claim 12 or 13, wherein, The predetermined time is 0.1 to 10 seconds.

22. The semiconductor manufacturing method as described in claim 12 or 13, wherein, The semiconductor materials that are bonded together are held by an electrostatic chuck that serves as a parallel plate plasma electrode.