Information processing apparatus, information processing method, storage medium, and computer program product

Abstract

The invention relates to an information processing apparatus, an information processing method, a storage medium, and a computer program product. Provided is a technique that facilitates the suppression of wear between a substrate and a holding section that holds the substrate. An information processing device outputs a control parameter relating to the supply of a gas between an object and a holding unit that holds the object when the object is separated from the holding unit. The information processing device is provided with: an output unit that outputs the control parameter on the basis of an inference model; and a machine learning unit that performs machine learning of the reasoning model on the basis of a result of separation between the object and the holding unit when the gas is supplied on the basis of the control parameter.

Description

Technical Field

[0001] The present invention relates to an information processing device, an information processing method, a storage medium and a program product. Background Art

[0002] A substrate processing device used in the manufacturing process of semiconductor devices, liquid crystal display devices, etc. may include a substrate chuck (substrate holding portion) for holding a substrate. The smoothness of holding and releasing the substrate in the substrate chuck is related not only to productivity but also to the wear resistance of the substrate and the substrate chuck.

[0003] Patent Document 1 describes that, when a substrate is unloaded from a support table, gas is supplied to a gap between a base surface of the support table and the substrate through a gas flow opening formed in the support table.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application No. 2019-505841 Summary of the invention

[0007] Problem that the invention aims to solve

[0008] The substrate chuck chucks the substrate, and even if the chuck is released afterwards, an adhesion force still acts between the substrate and the substrate chuck. If the amount of gas supplied when separating the substrate from the substrate chuck is too large relative to the adhesion force, the substrate will float from the substrate chuck and the substrate will slide laterally. On the other hand, if the amount of gas supplied when separating the substrate from the substrate chuck is too small relative to the adhesion force, the substrate will not be separated from the substrate chuck, causing wear of the substrate chuck. Since the adhesion force depends on the characteristics of the substrate and the substrate chuck, it is preferred that the supply of gas used to separate the substrate from the substrate chuck is appropriately set each time according to the adhesion force.

[0009] In addition, for example, if the substrate stage is moved at a high speed when separating the substrate from the substrate chuck, the substrate and the substrate chuck slide, thereby causing the substrate chuck to wear. This sliding is also caused by the adhesion between the substrate and the substrate chuck. In order to reduce the influence of the adhesion and separate the substrate from the substrate chuck, it is preferred that the drive of the substrate stage is also appropriately set each time.

[0010] The present invention provides a technology that is advantageous in suppressing wear of a substrate and a holding portion that holds the substrate.

[0011] Solutions for solving problems

[0012] According to the first aspect of the present invention, there is provided an information processing device, characterized in that a control parameter is output, wherein the control parameter is a control parameter related to the supply of gas between the object and the holding portion when the object is separated from the holding portion holding the object, wherein the information processing device comprises: an output unit that outputs the control parameter based on an inference model; and a machine learning unit that performs machine learning of the inference model based on the result of separation of the object and the holding portion when the gas is supplied based on the control parameter.

[0013] According to a second aspect of the present invention, there is provided a computer-readable storage medium storing a program for causing a computer to function as each unit of the information processing apparatus according to the first aspect.

[0014] According to a third aspect of the present invention, there is provided a computer program product including a program for causing a computer to function as each unit of the information processing apparatus according to the first aspect.

[0015] According to the fourth aspect of the present invention, there is provided an information processing method, characterized in that a control parameter is output, wherein the control parameter is a control parameter related to the supply of gas between the object and the holding portion when the object is separated from the holding portion holding the object, wherein the information processing method comprises: an output step of outputting the control parameter based on an inference model; and a machine learning step of performing machine learning of the inference model based on the result of separation of the object and the holding portion when the gas is supplied based on the control parameter.

[0016] Effects of the Invention

[0017] According to the present invention, it is possible to provide a technique which is advantageous in suppressing wear of a substrate and a holding portion that holds the substrate. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram showing the structure of an exposure device.

[0019] Figure 2 It is a diagram showing the structure of a substrate mounting table.

[0020] Figure 3 It is a diagram showing the functional structure of the control unit.

[0021] Figure 4 is a top view of the substrate chuck.

[0022] Figure 5 This is a diagram for explaining the control of gas supply.

[0023] Figure 6 It is a diagram for explaining the transfer operation of the substrate.

[0024] Figure 7 is a top view of the substrate chuck.

[0025] Figure 8 It is a diagram showing a driving curve of the fine motion stage.

[0026] Fig. 9 It is a diagram showing the time transition of the Z deviation of the fine movement stage.

[0027] Fig.10 This is a diagram showing the time change of pressure during gas supply.

[0028] Fig.11 It is a diagram showing the structure of a neural network.

[0029] Fig.12 It is a flowchart of the learning phase.

[0030] Fig.13 This is a flow chart of the estimation phase.

[0031] Description of Reference Numerals

[0032] 1: substrate mounting table, 2: substrate, 3: substrate chuck, 5: vacuum pump, 6: positive pressure tank, 8: fine motion mounting table, 10: gas flow path, 13: coarse motion mounting table, 100: control unit, 700: information processing device, 701: processing unit. DETAILED DESCRIPTION

[0033] Hereinafter, the embodiments are described in detail with reference to the accompanying drawings. It should be noted that the following embodiments do not limit the invention involved in the claims. A plurality of features are described in the embodiments, but these plurality of features are not limited to being all necessary features for the invention, and a plurality of features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same or similar structures are marked with the same reference numerals, and repeated descriptions are omitted.

[0034] <First Embodiment>

[0035] The present disclosure relates to an information processing device that provides control parameters of a holding unit to a control unit that controls a holding unit that holds an object. The object as an object can be a flat or gently curvatured object such as a substrate, an original plate, an optical element, etc. in a substrate processing device that processes a substrate. In the present disclosure, the substrate processing device can be, for example, a photolithography device that forms a pattern on a substrate such as an exposure device and an imprinting device. Alternatively, the substrate processing device can be an inspection device such as an overlay inspection device, a CD inspection device, a defect inspection device, an electrical characteristic inspection device, etc. Alternatively, the substrate processing device can be a processing device such as an etching device and a film forming device. Below, a case where the substrate processing device in the present disclosure is an exposure device as an example of a photolithography device is described. As described in detail below, the substrate processing device can have an original plate holding unit that holds an original plate as an object, and a substrate holding unit that holds a substrate as an object.

[0036] Figure 1 : is a schematic diagram of the exposure device 200 in the embodiment. In this specification and the drawings, directions are shown in an XYZ coordinate system in which a horizontal plane is set as an XY plane. Usually, the substrate 2 is placed on the substrate chuck 3 of the substrate mounting table 1 in a manner such that its surface is parallel to the horizontal plane (XY plane). Therefore, the directions orthogonal to each other in a plane along the surface of the substrate 2 are set as the X-axis and the Y-axis, and the direction perpendicular to the X-axis and the Y-axis is set as the Z-axis. In addition, the directions parallel to the X-axis, the Y-axis, and the Z-axis in the XYZ coordinate system are referred to as the X-direction, the Y-direction, and the Z-direction, respectively, and the rotation direction around the X-axis, the rotation direction around the Y-axis, and the rotation direction around the Z-axis are referred to as the θX direction, the θY direction, and the θZ direction, respectively.

[0037] In the exposure device 200, exposure light guided from a light source (not shown) via an illumination optical system 202 is irradiated onto a master plate 203. The master plate is also called an intermediate mask, a mask, etc. The exposure light that has passed through the master plate 203 is irradiated onto a substrate 2 via a projection optical system 206. As a material for the substrate, for example, glass, ceramics, metal, semiconductor, resin, etc. can be used. If necessary, a member made of a material different from the substrate can also be provided on the surface of the substrate. For example, the substrate is a silicon substrate, a compound semiconductor substrate, or quartz glass.

[0038] The original plate stage 204 has an original plate chuck (not shown) as an original plate holding portion. The original plate stage 204 uses the original plate chuck to hold the original plate 203. The original plate stage 204 can move linearly in the X direction, the Y direction, and the Z direction, and can rotate in the θX direction, the θY direction, and the θZ direction, and these axes can be controlled independently.

[0039] A laser interferometer (not shown) is configured to accurately measure the position of the plate mounting table 204. The position of the plate mounting table 204 can be determined based on the amount of change in the measurement value obtained by the laser interferometer.

[0040] The original plate conveying system 214 can include an original plate manipulator 215, an original plate pre-alignment stage 216, an original plate conveying robot 217, and a storage section 218. The original plate conveying robot 217 is a multi-jointed robot. The original plate conveying robot 217 has a manipulator for holding the original plate 203, and can move the manipulator to any position in the XYZ space. The storage section 218 stores the original plate 203. The original plate 203 can be placed in the storage section 218 from outside the device. The original plate 203 placed in the storage section 218 is placed on the original plate pre-alignment stage 216 by the original plate conveying robot 217 which is a multi-jointed robot.

[0041] On the original plate pre-alignment stage 216, the mark on the original plate 203 is observed by a microscope (not shown), thereby measuring the position of the original plate 203 on the original plate pre-alignment stage 216. After the measurement, the original plate robot 215 holds the original plate 203 on the original plate pre-alignment stage 216 and drives along a guide (not shown) in the +Y direction to a position where the original plate 203 is handed over to the original plate mounting stage 204. After the drive is completed, the original plate robot 215 drives in the -Z direction and hands over the original plate 203 to the original plate mounting stage 204. The original plate mounting stage 204 vacuum absorbs and holds the original plate 203. The displacement of the original plate 203 measured by the original plate pre-alignment stage 216 in the X direction, the Y direction, and the θZ direction relative to the original plate pre-alignment stage 216 can be corrected by changing the position where the original plate mounting stage 204 receives the original plate 203.

[0042] The substrate conveying system 210 may include a supply robot 211, a recovery robot 212, a storage section 213, a pre-alignment unit 219, a conveying robot 220, and a temporary placement table 221. The substrate 2 is stored in the storage section 213. The conveying robot 220 is a multi-joint robot. The conveying robot 220 has a robot for holding the substrate 2, and the robot can be moved to any position in the XYZ space. The conveying robot 220 takes out the substrate 2 from the storage section 213 and places the substrate 2 on the pre-alignment unit 219. In the pre-alignment unit 219, the outer periphery of the substrate 2 is irradiated with measurement light, and the reflected light is measured by a sensor, thereby measuring the outer shape position of the substrate 2 relative to the pre-alignment unit 219.

[0043] The supply robot 211 is a conveying mechanism that conveys the substrate 2 from the pre-alignment unit 219 to the substrate stage 1, specifically, to the pin member 9 of the coarse motion stage 13. The supply robot 211 can hold the substrate 2 by vacuum adsorption. The recovery robot 212 is a conveying mechanism that recovers the substrate 2 from the substrate stage 1 and conveys it to the temporary placement stage 221. The recovery robot 212 can hold the substrate 2 by vacuum adsorption.

[0044] The temporary placement table 221 is a table for temporarily placing the substrate 2 collected from the substrate mounting table 1 by the collection robot 212 . Since the transfer robot 220 cannot directly take out the substrate from the collection robot 212 , the substrate 2 is transferred via the temporary placement table 221 .

[0045] The substrate stage 1 can include a coarse motion stage 13 and a fine motion stage 8 mounted on the coarse motion stage 13. A substrate chuck 3 is mounted on the fine motion stage 8. The substrate chuck 3 is a substrate holding device that holds the substrate 2. The substrate chuck 3 can be a vacuum chuck that adsorbs the lower surface of the substrate 2 to the holding surface by vacuum suction. Alternatively, the substrate chuck 3 can also be an electrostatic chuck that adsorbs the lower surface of the substrate 2 to the holding surface by electrostatic action. The following describes the case where the substrate chuck 3 is a vacuum chuck. The fine motion stage 8 is configured to be able to move linearly in the X direction, the Y direction, and the Z direction, and to be able to rotate in the θX direction, the θY direction, and the θZ direction. The substrate stage 1 has a driving mechanism (not shown) that drives the fine motion stage 8 on these axes, and these axes can be controlled independently.

[0046] A laser interferometer (not shown) is configured to accurately measure the position of fine movement stage 8. The position of fine movement stage 8 can be determined based on the displacement amount of the measurement value obtained by the laser interferometer.

[0047] The pin member 9 is used when receiving the substrate 2 from the supply robot 211 and when delivering the substrate 2 to the recovery robot 212. The pin member 9 can adsorb and hold the substrate 2. In order to stably support the substrate 2, at least three pin members can be arranged at the vertices of the triangle as large as possible or in the vicinity thereof in the holding surface. The pin member 9 passes through the holes formed in the fine motion stage 8 and the substrate chuck 3, respectively, and the longitudinal axis of the pin member is arranged in a manner parallel to the direction perpendicular to the holding surface of the substrate chuck 3 (holding part) (i.e., the Z direction). The lower end of the pin member 9 is fixed to the coarse motion stage 13, and the upper end is configured to be able to appear and disappear relative to the holding surface of the substrate chuck 3. The pin member 9 acts as a supporting member for supporting the substrate 2 by protruding upward more than the holding surface. The appearance and disappearance of the pin member 9 relative to the holding surface of the substrate chuck 3 can be performed by at least any one of the extension and contraction of the pin member 9, the up-and-down movement of the coarse motion stage 13, and the up-and-down movement of the fine motion stage 8.

[0048] The coarse movement stage 13 can move linearly in the X direction and the Y direction and can rotate in the θZ direction. For example, a capacitance sensor is used to measure the distance between the coarse movement stage 13 and the fine movement stage 8. Based on the measurement result, an actuator such as a linear motor is used to control the position of the coarse movement stage 13 so as to follow the position of the fine movement stage 8. The pin member 9 is fixed to the coarse movement stage 13 and thus moves together with the coarse movement stage 13.

[0049] The off-axis scope 209 is capable of measuring alignment marks on the substrate 2 .

[0050] The control unit 100 controls the operation of the exposure device 200. The control unit 100 can include functions of a driver and a controller for controlling sensors, actuators, and the like in the exposure device 200.

[0051] The information processing device 700 is connected to the control unit 100 that controls the exposure device 200 in a communicable manner. The information processing device 700 can provide the control parameters of the holding unit to the control unit 100 that controls the exposure device 200 (i.e., the holding unit that holds the object). The information processing device 700 can include a processing unit 701, a display unit 705, and an input unit 706. The processing unit 701 can be composed of a computer including a processor such as a CPU (Central Processing Unit), a memory, and a GPU (Graphical Processing Unit). The processing unit 701 can include an acquisition unit 702, a calculation unit 703, a supply unit 704, and a storage unit 707. The acquisition unit 702 acquires information from the exposure device 200. The calculation unit 703 calculates the control parameters based on the information acquired by the acquisition unit 702. The control parameters can include parameters related to gas supply and parameters related to the drive control of the micro-motion stage. The supply unit 704 supplies (sends, provides) the control parameters to the control unit 100 of the exposure device 200. The storage unit 707 stores information acquired from the exposure device 200. The acquisition unit 702 and the supply unit 704 can also be understood as a unit constituting a communication unit that communicates with the control unit 100 that controls the exposure device 200 to transmit and receive information.

[0052] The following is a description of the recovery sequence of the substrate 2 in the present embodiment. The recovery sequence of the substrate 2 may include an action of handing the substrate 2 from the substrate chuck 3 to the pin member 9 after the exposure is completed. This action can be performed by relatively driving the substrate chuck 3 and the pin member 9 in the Z direction. Here, the adhesion force acts between the substrate 2 and the substrate chuck 3. In this specification, the adhesion force refers to the force required to separate the substrate 2 and the substrate chuck 3 by applying force to the substrate 2 and the substrate chuck 3 in the direction of separation from each other. That is, the adhesion force is related to the ease of separation of the substrate 2 relative to the substrate chuck 3. Even if the substrate 2 is unloaded, that is, even if the substrate chuck 3 releases the holding of the substrate 2, the adhesion force may remain depending on the characteristics of the holding surface of the substrate chuck 3, the characteristics of the substrate 2, the characteristics of the exposure process, the influence of foreign matter mixed between the substrate 2 and the substrate chuck 3, etc. In addition, even if the vacuum adsorption is released when the substrate is unloaded, it will not immediately become atmospheric pressure, and the pressure that cannot be released may become adhesion force. In addition, in the case where the substrate chuck 3 is an electrostatic chuck, the adhesion force may also remain due to the electrostatic chuck and the charging of the substrate when the power is turned off. If the adhesive force is strong, the adhesive force may be overcome to pull the substrate 2 away from the substrate chuck 3 , thereby damaging the substrate 2 or the substrate chuck 3 , or both.

[0053] Therefore, in the present embodiment, when the substrate 2 is released from the substrate chuck 3, gas is supplied between the substrate 2 and the substrate chuck 3 to facilitate the separation of the two. However, if the amount of gas supplied when separating the substrate 2 from the substrate chuck 3 is too large relative to the adhesion force, the substrate 2 floats from the substrate chuck 3 and the substrate 2 slides laterally. On the other hand, if the amount of gas supplied when separating the substrate 2 from the substrate chuck 3 is too small relative to the adhesion force, the substrate 2 is not separated from the substrate chuck 3, causing wear of the substrate chuck 3. Since the adhesion force is caused by the characteristics of the substrate 2, the characteristics of the substrate chuck 3, the characteristics of the exposure process, etc., it is necessary to appropriately set the supply of gas for separating the substrate 2 from the substrate chuck 3 each time according to the adhesion force.

[0054] In addition, for example, if the fine motion stage 8 is moved at a high speed when separating the substrate 2 from the substrate chuck 3, the substrate chuck 3 may be worn due to the sliding between the substrate 2 and the substrate chuck 3. This sliding is also caused by the adhesion between the substrate 2 and the substrate chuck 3. In order to reduce the influence of the adhesion and separate the substrate 2 from the substrate chuck 3, the driving of the fine motion stage 8 also needs to be appropriately set each time.

[0055] Hereinafter, the holding of the substrate 2 by the substrate chuck 3 and its release, and the control of the supply of gas between the substrate 2 and the substrate chuck 3 will be described.

[0056] Figure 21 is a schematic diagram showing the structure of the substrate stage 1. The substrate chuck 3 is configured to hold the substrate 2 by vacuum adsorbing the substrate 2 to the holding surface by reducing the pressure in the space between the substrate 2 and the holding surface of the substrate chuck 3. The substrate chuck 3 is held on the fine movement stage 8 by vacuum adsorption. The fine movement stage 8 is mounted on the coarse movement stage 13.

[0057] A gas flow path 10 is arranged in the coarse motion stage 13. The gas flow path 10 is arranged so that gas is sucked from the vacuum pump 5 (vacuum source) as a negative pressure generating member through an opening formed on the surface of the substrate chuck 3 facing the substrate 2. In addition, the gas flow path 10 is arranged so that gas from the positive pressure tank 6 as a positive pressure generating member (gas supply source) is supplied through the opening. In addition, in the coarse motion stage 13, a pressure sensor 20, regulators 41, 61, 62, 63, and electromagnetic valves 31, 51, 52, 53 are provided on the gas flow path 10.

[0058] A plurality of openings 7 are formed on the surface of the substrate chuck 3 that faces the substrate 2 (see Figure 4 ). The gas flow path 10 is branched and connected to each opening via the fine motion stage 8. In addition, the gas flow path 10 is connected to the vacuum pump 5 via the electromagnetic valve 31 and the regulator 41, and is connected to the positive pressure tank 6 via the electromagnetic valves 51, 52, 53 and the regulators 61, 62, 63. Usually, the vacuum pump 5 and the positive pressure tank 6 are configured as factory equipment and are therefore provided outside the substrate stage 1.

[0059] The pressure inside the gas flow path 10 is measured by the pressure sensor 20. In one example, the pressure sensor 20 is configured to measure the pressure at a position in the gas flow path 10 that is closer to the vacuum pump than the branch of the gas flow path 10 and closer to the substrate chuck 3 than the solenoid valves 31, 51, 52, and 53. This is to accurately measure the pressure of the space between the substrate 2 and the substrate chuck 3 using the pressure sensor 20.

[0060] The regulator 41 and the electromagnetic valve 31 are arranged between the vacuum pump 5 and the substrate chuck 3. The regulator 41 is configured to be able to adjust the pressure in the gas flow path 10 to the desired vacuum pressure. The electromagnetic valve 31 is arranged on the substrate chuck 3 side relative to the regulator 41. This is to supply the pressure adjusted by the regulator 41 to the substrate chuck 3. When the electromagnetic valve 31 is turned on, the vacuum pump 5 adjusts the pressure of the space between the substrate 2 and the substrate chuck 3. When the electromagnetic valve 31 is turned off, the pressure of the space between the substrate 2 and the substrate chuck 3 is not adjusted.

[0061] The regulators 61, 62, 63 and the solenoid valves 51, 52, 53 are arranged between the positive pressure tank 6 and the substrate chuck 3. The regulators 61, 62, 63 are constructed to be able to adjust the supply pressure from the positive pressure tank 6 in the gas flow path 10. Being able to adjust the supply pressure means that the supply flow rate can also be adjusted. The solenoid valves 51, 52, 53 are arranged on the substrate chuck 3 side relative to the regulators 61, 62, 63 on the gas flow path 10. When the solenoid valves 51, 52, 53 are turned on, the positive pressure tank 6 adjusts the pressure of the space between the substrate 2 and the substrate chuck 3. When the solenoid valves 51, 52, 53 are turned off, the pressure of the space between the substrate 2 and the substrate chuck 3 is not adjusted. It should be noted that in Figure 2 In the figure, three sets of regulators and electromagnetic valves are shown between the positive pressure tank 6 and the substrate chuck 3, but the number of sets is not limited to a specific number. It can be one set or four or more sets. In addition, a servo valve capable of freely setting the opening and closing degree can be formed to replace the combination of the electromagnetic valve and the regulator.

[0062] When the substrate 2 is vacuum-adsorbed by the substrate chuck 3, the solenoid valve 31 is turned on, and the solenoid valves 51, 52, and 53 are turned off. Thus, the pressure of the space between the substrate 2 and the substrate chuck 3 can be reduced by the vacuum pump 5. When the substrate 2 is released from the substrate chuck 3, the solenoid valve 31 is turned off, and at least one of the solenoid valves 51, 52, and 53 is turned on. Thus, the space between the substrate 2 and the substrate chuck 3 can be pressurized by the gas from the positive pressure tank 6.

[0063] By changing the settings of the regulators 61, 62, and 63, respectively, when any one of the electromagnetic valves 51, 52, and 53 is turned on, the flow rate and pressure of the gas filling the space between the substrate 2 and the substrate chuck 3 can be changed. In one example, the regulators 61, 62, and 63 are set in such a way that the supply flow rate of the gas increases in the order of the regulators 61, 62, and 63. When the substrate 2 is released from the substrate chuck 3, the control unit 100 controls the supply of the gas through the gas flow path 10 and the opening 7 based on the control parameters provided from the information processing device 700. The control parameters include information on the ease of separation of the substrate 2 from the holding surface. For example, the control unit 100 determines which of the electromagnetic valves 51, 52, and 53 to turn on and how to determine the supply time of the gas based on the information on the ease of separation of the substrate 2 from the holding surface.

[0064] The control unit 100 can be constituted by, for example, a general-purpose or special-purpose processor having a program programmed in a memory. Alternatively, the control unit 100 can also be constituted by a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). Alternatively, the control unit 100 can also be constituted by a combination of all or part of a processor, a PLD, and an ASIC. It should be noted that the control unit 100 can be constituted either inside the exposure device 200 or outside the exposure device 200. In the case where the control unit 100 is constituted outside the exposure device 200, the functions of the control unit 100 are installed in, for example, a server device connected to the exposure device 200 in a manner capable of communicating. The functions of the control unit 100 can also be implemented by the information processing device 700.

[0065] exist Figure 3 , the functional structure of the control unit 100 is shown. The control unit 100 can include an indication unit 110, a storage unit 120, and a processing unit 130. The indication unit 110 indicates the control amount of each driving unit including the fine-motion stage 8. The storage unit 120 stores various data and programs related to the control of exposure. The processing unit 130 performs various processes related to the control of exposure. For example, the processing unit 130 calculates the drive target positions of the substrate stage 1 and the original plate stage 204. In addition, the processing unit 130 executes the exposure sequence, the supply and recovery sequence of the substrate according to the control program stored in the storage unit 120.

[0066] The functions of the processing unit 130 will be described later. Figure 3 In FIG. 1 , an example is shown in which the control unit 100 includes the storage unit 120 and the processing unit 130 , but the storage unit 120 and the processing unit 130 may also be provided relatively independently from the control unit 100 .

[0067] Figure 4 60 is a top view of the substrate chuck 3 viewed from the +Z direction. The annular sealing portion 60 is provided along the outer circumference of the substrate chuck 3. However, the presence, number, and arrangement of the sealing portion 60 are not limited to Figure 4 For example, a sealing portion may not be provided on the substrate chuck 3. Alternatively, a plurality of sealing portions may be provided.

[0068] The substrate chuck 3 has a plurality of protrusions (not shown) on its surface, and the upper surfaces of these protrusions can serve as holding surfaces for holding the substrate 2. These protrusions are referred to as chuck pins. The substrate chuck 3 has an opening 7 connected to the gas flow path 10. Figure 4 12 openings 7 are shown in FIG. 1 , but the arrangement and number of the openings 7 are not limited to Figure 4 Example shown.

[0069] The substrate 2 is vacuum-adsorbed by the substrate chuck 3 by exhausting the gas in the space between the substrate 2 and the substrate chuck 3 from the opening 7. The sealing portion 60 is provided between the substrate 2 and the substrate chuck 3 to prevent the air from flowing in from the surroundings and enable vacuum adsorption.

[0070] The action of releasing the substrate 2 from the substrate chuck 3 is performed by supplying gas between the substrate 2 and the substrate chuck 3 from the opening 7. Here, if the amount of gas supplied (pressure, flow rate) and the time of gas supply are too large relative to the adhesion force, the substrate 2 will float from the substrate chuck 3 and the substrate 2 will slide laterally. On the other hand, if the amount of gas supplied when the substrate 2 and the substrate chuck 3 are separated and the time of gas supply are too small relative to the adhesion force, the substrate 2 will not be separated from the substrate chuck 3, and the substrate chuck will be worn when the substrate 2 is lifted. The detailed description of the adhesion force and the method of optimizing the amount of gas supplied relative to the adhesion force will be described later.

[0071] Reference Figure 6 , a method of transferring the substrate 2 from the substrate chuck 3 to the pin member 9 will be described. Figure 6 Schematic cross-sectional view of substrate stage 1. Fine motion stage 8 moves in Z direction along the length direction of pin member 9. It should be noted that fine motion stage 8 can move in Z direction while tilting according to the Z position relationship of three pin members 9. Pin member 9 is arranged on coarse motion stage 13 which can be driven in XY direction. The Z direction position of pin member 9 can be measured using, for example, an interferometer, an electrostatic capacitance sensor, etc. (not shown).

[0072] The pin member 9 is connected to the vacuum pump 5 via a gas flow path 70. A regulator 73 and an electromagnetic valve 72 are provided between the vacuum pump 5 and the pin member 9. The regulator 73 is configured to be adjustable to a desired vacuum pressure. The electromagnetic valve 72 is arranged at a position closer to the pin member 9 than the regulator 73. This is to supply the pressure adjusted by the regulator 73 to the substrate chuck 3. When the electromagnetic valve 72 is turned on, the vacuum pump 5 adjusts the pressure of the space between the substrate 2 and the substrate chuck 3, and when the electromagnetic valve 72 is turned off, the pressure of the space between the substrate 2 and the substrate chuck 3 is not adjusted. In addition, in order to monitor the pressure between the pin member 9 and the substrate 2, a pressure sensor 71 is provided. The pressure sensor 71 needs to be arranged at a position closer to the pin member 9 than the electromagnetic valve 72. Figure 8 In the embodiment, the pressure sensor 71, the electromagnetic valve 72, and the regulator 73 are arranged outside the coarse movement stage 13, but they may be arranged inside the coarse movement stage.

[0073] exist Figure 6 In (a), the fine-motion stage 8 is located on the +Z direction side, and the substrate chuck 3 vacuum-adsorbs the substrate 2. Figure 6 In (b), the micro-motion stage 8 moves in the -Z direction, the pin member 9 contacts the substrate 2, and the micro-motion stage 8 stops moving. Thereafter, the control unit 100 releases the adsorption and holding of the substrate 2 by the substrate chuck 3. At this time, the control unit 100 controls the supply of gas through the gas flow path 10 and the opening 7 based on information on the ease of separation of the substrate 2 from the holding surface (for example, the adhesion between the substrate 2 and the substrate chuck 3).

[0074] After that, the electromagnetic valve 72 is turned on, whereby the pin member 9 adsorbs and holds the substrate 2. Figure 6 As shown in (c) of FIG. 1 , the fine movement stage 8 is further moved in the −Z direction while the pin member 9 is holding the substrate 2 by suction. In this way, the substrate 2 is transferred from the substrate chuck 3 to the pin member 9 .

[0075] exist Figure 6 In the example shown, the substrate 2 is transferred from the substrate chuck 3 to the pin member 9 by the movement of the fine motion stage 8, but any method is sufficient as long as the relative position of the substrate chuck 3 and the pin member 9 can be changed. For example, the entire pin member 9 or the front end thereof may be moved in the +Z direction. Changing the relative position of the substrate chuck 3 and the pin member 9 is synonymous with relatively moving the substrate chuck 3 and the pin member 9.

[0076] Figure 7 It is a top view of the substrate chuck 3 observed from the +Z direction. A plurality of protrusions (chuck pins) not shown are provided on the surface of the substrate chuck 3, and the upper surfaces of these plurality of protrusions can serve as holding surfaces for holding the substrate 2. An annular sealing portion 60 is provided on the outer periphery of the substrate chuck 3 so that air does not flow out to the outside of the substrate chuck 3 when the substrate 2 is vacuum adsorbed. The substrate chuck 3 has an opening 7 connected to the gas flow path 10. The substrate chuck 3 vacuum adsorbs the substrate 2 via the opening 7. The space between the substrate 2 and the chuck pins becomes a vacuum state, and the substrate 2 can be adsorbed and held with a uniform force.

[0077] The substrate chuck 3 and the fine movement stage 8 are provided with holes 14 for passing the pin member 9 in the Z direction, so that the fine movement stage 8 can be driven in the Z direction without interfering with the pin member 9. Figure 7 In the figure, the hole 14 through which the pin member 9 passes is shown. Figure 7Although the structure in which three pin members 9 are provided is shown, the number of the pin members 9 is not limited as long as the substrate 2 can be stably held depending on the size, material, and mass of the substrate 2 .

[0078] exist Figure 6 In the example shown, an example is shown in which the pin member 9 adsorbs and holds the substrate 2 by moving the fine motion stage 8, but it is sufficient as long as the relative position of the substrate chuck 3 and the pin member 9 can be changed. For example, the entire pin member 9 or the front end portion may be moved in the +Z direction. Changing the relative position of the substrate chuck 3 and the pin member 9 is synonymous with relatively moving the substrate chuck 3 and the pin member 9. Alternatively, a structure may be adopted in which the pin member 9 adsorbs and holds the substrate 2 not by moving the fine motion stage 8 but by extending the pin member 9. In this way, the substrate stage 1 can include a driving unit for driving the fine motion stage 8 and / or driving or deforming the pin member 9 for performing the action of separating the substrate 2 from the substrate chuck 3.

[0079] exist Figure 8 2 is a graph showing a driving curve of the fine-motion stage 8. Figure 8 The relationship between time and the Z direction position of the fine movement stage 8 in the delivery operation of the substrate 2 from the substrate chuck 3 to the pin member 9 is shown. It should be noted that the change amount of the Z direction position of the fine movement stage 8 is synonymous with the change amount of the Z direction position of the substrate chuck 3.

[0080] In the first section, the indicator 110 indicates the first position stored in the storage unit 120 as the target position to the fine motion stage 8 in such a way that the fine motion stage 8 moves in the -Z direction at a high speed. When the fine motion stage 8 is driven in the -Z direction, the first position is set to be the same position as the Z direction position of the fine motion stage 8 when the substrate 2 and the pin member 9 are in contact, or a position closer to the +Z direction side. Here, as Figure 6 In this way, an example of moving the fine motion stage 8 in the -Z direction is shown. However, in the case of a structure in which the pin member 9 is moved in the +Z direction to perform substrate delivery, only the moving direction is different, so the -Z direction and the +Z direction can be replaced for application. Regarding the following figures and descriptions, the -Z direction and the +Z direction can also be replaced for application in the case of moving the pin member 9 in the +Z direction. In addition, in the method of driving both the fine motion stage 8 and the pin member 9, moving the fine motion stage 8 in the Z direction can also be replaced with the description of moving the fine motion stage 8 and the pin member 9 relative to each other.

[0081] In the second section, the control unit 100 moves the fine motion stage 8 in the -Z direction at a low speed and monitors the driving current value during the driving of the fine motion stage 8. By monitoring the driving current value when the fine motion stage 8 is moved in the -Z direction, the driving current value that changes when the substrate 2 contacts the pin member 9 can be detected, and the contact between the substrate 2 and the pin member 9 can be detected. The reason for moving the fine motion stage 8 at a low speed in the second section is that if the substrate 2 contacts the pin member 9 at a high speed, there is a possibility that the substrate 2 and the pin member 9 will be damaged. If the control unit 100 detects that the pin member 9 contacts the substrate 2 by monitoring the driving current value, the movement of the fine motion stage 8 in the -Z direction is stopped, and the second section is ended. It should be noted that the control unit 100 can also update the value of the first position stored in the storage unit 120 with the value of the Z direction position of the fine motion stage 8 when the contact between the pin member 9 and the substrate 2 is detected in the second section. By updating the first position in this way, the time required for the second section can be reduced.

[0082] In addition, when the coordinates of the fine motion stage 8 when the substrate 2 is in contact with the pin member 9 are the same as the first position, the substrate 2 is in contact with the pin member 9 at the time point of moving to the first position. In this case, since the contact between the substrate 2 and the pin member 9 is detected by the driving current value at the same time as the start of the second section, the fine motion stage 8 is not moved at a low speed in the second section, but is transferred to the third section. It should be noted that, when it is known that the coordinates of the fine motion stage 8 when the substrate 2 is in contact with the pin member 9 are the same as the first position, the second section may be omitted, and the transfer to the third section may be made after the first section.

[0083] In the third section, the control unit 100 stops the suction of the gas between the substrate 2 and the substrate chuck 3, and in parallel, supplies the gas between the substrate 2 and the substrate chuck 3. At this time, by supplying the gas according to the adhesion force, the phenomenon of the substrate 2 floating from the substrate chuck 3 and sliding laterally is prevented. In addition, in the fourth section described later, the substrate 2 is forcibly separated from the substrate chuck 3 in a state where the adhesion force remains, thereby suppressing the wear of the substrate chuck 3 caused by the sliding of the substrate 2 and the substrate chuck 3. In the third section, the pin member 9 also starts to suck the gas between the substrate 2 and the pin member 9. In the third section, the control unit 100 can estimate the adhesion force based on the characteristics of the substrate 2, the characteristics of the substrate chuck 3, the characteristics of the exposure process, etc., optimize the gas supply parameters, and supply the gas. The estimation of the adhesion force and the supply of the gas will be described later.

[0084] In the fourth section, the control unit 100 moves the fine motion stage 8 in the -Z direction at a low speed to separate the substrate 2 from the substrate chuck 3 by a small distance. If the fine motion stage 8 is moved in the -Z direction at a high speed in the state where the adhesion force exists to forcibly separate the substrate 2 and the substrate chuck 3, the substrate 2 will slip laterally on the substrate chuck 3, thereby causing the substrate chuck 3 to wear. The wear of the substrate chuck 3 affects the subsequent adsorption and retention of the substrate 2. Therefore, in order to reduce the wear of the substrate chuck 3, in the third section, gas is supplied between the substrate 2 and the substrate chuck 3 according to the adhesion force.

[0085] In the fifth section, the fine movement stage 8 is moved in the -Z direction at a high speed to the final position in the Z direction.

[0086] If the adhesion between the substrate 2 and the substrate chuck 3 can be reduced by supplying gas in the third section, the moving speed of the fine motion stage 8 in the fourth section can be increased. For example, the fifth section described later can be omitted and the substrate 2 can be driven at high speed in the -Z direction to the final position in the Z direction.

[0087] The method for estimating adhesion is described. As one of the indicators indicating the ease of separation of the substrate, there is the adhesion between the substrate 2 and the substrate chuck 3 (holding surface). That is, the information on the ease of separation of the substrate can include the information on the adhesion. The adhesion is the resultant force of the adsorption pressure acting between the substrate 2 and the substrate chuck 3 and the bonding force between the substrate 2 and each chuck pin of the substrate chuck 3. The adhesion can be estimated based on the characteristics of the substrate 2, the characteristics of the substrate chuck 3, the characteristics of the exposure process, etc. In one example, the adhesion can be estimated based on at least any one of the characteristics of the substrate 2 and the characteristics of the substrate chuck 3.

[0088] The characteristics of the substrate 2 are based on at least one of, for example, the amount of warping of the substrate 2, the roughness of the surface (back side) of the substrate 2 in contact with the holding surface of the substrate chuck 3, the type of film formed on the substrate 2, the raw material of the substrate 2, and the thickness of the substrate 2. For example, if the substrate 2 is warped, the adhesion between the substrate 2 and the substrate chuck 3 becomes smaller. This is because due to the warping of the substrate 2, stress acts in the direction in which the substrate 2 leaves the substrate chuck 3, and the close contact is reduced. On the other hand, when the substrate 2 is not warped, it becomes a state of high close contact and strong adhesion. The roughness of the back side of the substrate 2 also affects the adhesion. For example, when the back side of the substrate 2 is not rough, the substrate 2 and the substrate chuck 3 are firmly in contact. This is a state of strong adhesion. On the other hand, when the back side of the substrate 2 is rough, the substrate 2 and the substrate chuck 3 are not in contact, so it becomes a state of weak adhesion. In addition, when film is formed on the back side of the substrate, the type of film formation may affect the adhesion. Furthermore, the raw material and thickness of the substrate 2 also affect the weight of the substrate 2. The heavier the substrate, the more firmly the substrate 2 and the substrate chuck 3 adhere to each other. This is a state of strong adhesion.

[0089] The characteristics of the substrate chuck 3 are based on at least one of the uniformity of the height of the chuck pin (flatness of the holding surface), the surface coating of the chuck pin (surface roughness of the holding surface), the thickness of the chuck pin, the material, etc. The more uniform the height of the chuck pin, the more firmly the substrate 2 and the substrate chuck 3 are in close contact, so that the adhesion is strong. In addition, the thicker the chuck pin, the larger the contact area with the substrate, and the adhesion is strong. In the case where the material of the chuck pin, such as particles, is composed of fine substances, the substrate 2 and the substrate chuck 3 are firmly in close contact, and the adhesion is strong.

[0090] The characteristics of the exposure process can include exposure time, exposure amount, the presence or absence of θ correction drive, etc. For example, if the exposure time is long, the vacuum adsorption time of the substrate 2 and the substrate chuck 4 becomes longer, and the adhesion between the substrate 2 and the substrate chuck 4 becomes firm. If the exposure amount is large, the substrate 2 undergoes thermal expansion. During exposure, the substrate 2 is vacuum adsorbed on the substrate chuck 3, so the back of the substrate 2 does not expand although internal stress acts on it. However, if the vacuum adsorption is released, the internal stress is released and the substrate 2 expands. The substrate 2 is slightly offset laterally on the substrate chuck 3, thereby weakening the adhesion between the substrate 2 and the substrate chuck 3. In addition, in the exposure sequence, when the alignment mark is observed using an off-axis observer, the substrate 2 sometimes deviates in the θZ direction on the substrate chuck 3. In order to rotate the substrate 2 in the θZ direction on the substrate stage 1, an action called θ correction drive as follows is sometimes performed. The case where the substrate 2 is rotated by 200 μrad is described as an example. The control unit 100 rotates the fine-motion stage 8 by 200 μrad from a position of 0 μrad. Next, the control unit 100 transfers the substrate 2 from the substrate chuck 3 to the pin member 9. In this state, the control unit 100 rotates the fine-motion stage 8 by -200 μrad to return to the position of 0 μrad. The substrate 2 on the pin member 9 is not affected by this rotation. In this state, the control unit 100 transfers the substrate 2 to the substrate chuck 3 again. As a result, the substrate 2 is rotated by 200 μrad relative to the substrate chuck 3. This method can correct the offset in the θZ direction. Since the substrate chuck 3 releases the hold on the substrate 2 once and performs vacuum adsorption again, this is synonymous with a change in exposure time. In this way, the adhesion also changes through the exposure process.

[0091] Thus, if the characteristics of the substrate 2, substrate chuck 3, and exposure process are known, the force acting between the substrate 2 and substrate chuck 3, namely, the adhesion force can be estimated. The adhesion force can be estimated by calculation using a simulator (not shown) that models the substrate 2 and substrate chuck 3.

[0092] Next, the simulation of the change of the adhesion force when the gas supply amount and the gas supply time are changed is described. The larger the gas supply amount, the weaker the adhesion force between the substrate 2 and the substrate chuck 3. In addition, even if the gas supply amount is the same, the longer the supply time, the weaker the adhesion force between the substrate 2 and the substrate chuck 3. The gas supply amount and the gas supply time parameters are determined in such a way that the adhesion force approaches zero after the gas supply. This is used as the optimal parameter for the gas supply. In this way, information on the ease of separation of the substrate 2 relative to the holding surface of the substrate chuck 3 can be obtained in advance.

[0093] Reference Figure 5 (a) of the present invention will explain the control of gas supply. Figure 5 (a) shows an example of the time transition of the supply flow rate of the gas when the substrate 2 is released from the substrate chuck 3 . Figure 5 (a) shows that the supply flow rate of gas is constant when each electromagnetic valve is turned on. Here, since the supply flow rate of gas is set to increase in the order of regulators 61, 62, and 63, the supply flow rate of gas supplied to the substrate 2 and the substrate chuck 3 increases in the order in which the electromagnetic valves 51, 52, and 53 are turned on.

[0094] Figure 5 (a) is a diagram assuming that the adhesion between the substrate 2 and the substrate chuck 3 is large. Based on the results of the simulation with the characteristics of the substrate 2 and the characteristics of the substrate chuck 3 given in advance, the gas supply amount and the gas supply time are determined. The determined gas supply amount and gas supply time are stored in the storage unit 120. Figure 5 In (a), the control unit 100 (instruction unit 110) turns on the electromagnetic valve 51 according to the supply amount of gas stored in the storage unit 120. The length of time the electromagnetic valve 51 is turned on also follows the supply time of the gas stored in the storage unit 120.

[0095] exist Figure 5 (b) is exemplified in Figure 5 The change in the adhesion between the substrate 2 and the substrate chuck 3 during the supply of gas under the control of the gas supply shown in (a). During the supply of gas, the adhesion gradually weakens. If the adhesion weakens to Figure 5 At the optimum point shown in (b), the substrate 2 does not slide laterally on the substrate chuck 3, and the substrate chuck 3 is not worn. The point where the adhesion is 0 is not set as the optimum point in order to slightly leave adhesion so that the substrate 2 does not slide laterally.

[0096] By changing the electromagnetic valve that is turned on, the amount of gas supplied between the substrate 2 and the substrate chuck 3 can be controlled. In addition, the electromagnetic valve can be turned off after a predetermined time has passed since the start of gas supply to stop the gas supply. As a result, the substrate 2 will not slide laterally on the substrate chuck 3, and a state can be formed in which the substrate chuck 3 will not be worn.

[0097] The following is a method for measuring adhesion based on Z deviation. As described above, the fine-motion stage 8 can be driven in the Z direction (the first direction orthogonal to the holding surface) by a driving mechanism (driving unit) not shown in the figure. In addition, the Z direction position of the fine-motion stage 8 at this time can be measured by a laser interferometer (measuring unit) not shown in the figure.

[0098] Z deviation refers to the control deviation of the drive unit in the Z direction. Specifically, Z deviation refers to the difference between the target position of the substrate chuck 3 in the Z direction and the current position of the substrate chuck 3 in the Z direction measured by the laser interferometer (measuring unit). This is equivalent to the difference between the target position of the fine motion stage 8 in the Z direction indicated by the indicating unit 110 and the position of the fine motion stage 8 in the Z direction measured by the laser interferometer (not shown). In addition, this is equivalent to the difference between the target position of the pin member 9 in the Z direction and the position of the pin member 9 in the Z direction measured by the laser interferometer (not shown).

[0099] Adhesion can be determined according to the fourth interval ( Figure 8 ) is calculated. Fig. 9 is a graph showing the time course of the Z deviation. Fig. 9 In the figure, D1 represents the start time of the 4th interval (the end time of the 3rd interval), and D2 represents the end time of the 4th interval (the start time of the 5th interval). In the present embodiment, the control unit 100 controls the fine-motion stage 8 based on the pre-acquired Z deviation (the first condition). As described above, when the fine-motion stage 8 is moved in the -Z direction at a low speed in the 4th interval, the substrate 2 and the substrate chuck 3 may be in a state where they are difficult to separate due to adhesion. In this case, the fine-motion stage 8 cannot move to the target position indicated by the indication unit 110 (the substrate chuck 3 or the pin member 9 cannot be moved to the target position), and the target position and the actual position in the Z direction become different positions, resulting in a Z deviation. The processing unit 130 calculates the Z deviation, which is the difference between the target position of the fine-motion stage 8, the substrate chuck 3, or the pin member 9 in the Z direction and the actual position in the Z direction measured by a sensor not shown. The Z deviation is, for example, Fig. 9 The waveform 17 indicates the Z deviation. The waveform 18 indicates the Z deviation serving as a reference. The waveform 18 serving as a reference is a reference waveform indicating the Z deviation when the fine movement stage 8 is moved without holding the substrate 2.

[0100] In the present embodiment, the absolute value of the value of the Z deviation when the Z deviation is the maximum value or the minimum value is used as the representative value of the Z deviation. In addition, the representative value of the Z deviation that serves as the reference when the micro-motion stage 8 is moved without placing the substrate 2 on the substrate chuck 3 is set as the Z deviation reference value. The maximum value or minimum value of the Z deviation is the maximum value or minimum value among the representative values ​​of the Z deviation from when the substrate 2 contacts the pin member 9 to when the entire surface of the substrate 2 is separated from the substrate chuck 3. The processing unit 130 obtains the absolute value of the value of the Z deviation when the Z deviation is the maximum or the minimum in the waveform of the obtained Z deviation. The absolute value is used as the representative value of the Z deviation. The adhesion is obtained based on the difference between the representative value of the Z deviation and the Z deviation reference value.

[0101] In the present embodiment, an inference model (learning model) is used to infer parameters (gas supply parameters) related to the supply of gas between the substrate 2 and the substrate chuck 3. The operation unit 703 and the supply unit 704 can function as an output unit that outputs the control parameters to be provided to the control unit 100 for controlling the exposure device 200 according to the inference model. In addition, the information processing device 700 can have a machine learning unit that generates an inference model through machine learning. The operation unit 703 can function as a machine learning unit. The machine learning unit performs machine learning of the inference model based on the result of the separation of the substrate 2 and the holding surface of the substrate chuck 3 performed in the substrate chuck 3 while the control unit 100 controls the supply of gas based on the control parameters. As an inference method, existing methods such as neural networks, Bayesian networks, support vector machines, linear regression, and logistic regression can be used. Below, an example of determining gas supply parameters by machine learning using a neural network as an inference model (learning model) is described. By using a neural network, even without using a complex calculation formula, the optimal parameter can be determined based on the causal relationship between input and output. Furthermore, by performing learning using the results of gas supply by the exposure apparatus, it is possible to provide optimal parameters for unloading the substrate with higher accuracy than simulation.

[0102] exist Fig.11 91 is a structure example of a neural network 91 generated as an inference model. The neural network 91 includes an input layer 910, one or more intermediate layers 911, and an output layer 912. The values ​​of the input data 915 in the input layer 910 are represented by A1, A2, ..., A N In response to the input, the output data 917 of the output layer 912 is determined via one or more intermediate layers 911. Fig.11 In the example, there are two intermediate layers 911, and each value in each intermediate data 916 is represented by M 11 、M 12 , …, M 1M and M 21 、M 22 , …, M2P In addition, the values ​​of the output data 917 in the output layer 912 are represented by O1, O2, ..., O O The calculations of each layer of the neural network 91 are performed as follows.

[0103] The calculation method in the process of transforming input data into data of the intermediate layer is shown. First, the input data is linearly transformed. Here, if the intermediate data is set as U, the intermediate data U is represented as follows.

[0104] [Number 1]

[0105] U 11 =W 11 *A1+W 12 *A2+···+W 1N *A N +B1

[0106] U 12 =W 21 *A1+W 22 *A2+···+W 2N *A N +B2

[0107] …

[0108] U 1M =W M1 *A1+W M2 *A2+···+W MN *A N +B M

[0109] If the above mathematical formula is expressed as a matrix expression, it will be as follows.

[0110] [Number 2]

[0111]

[0112] In addition, if it is expressed by vector expression, it is as follows.

[0113] [Number 3]

[0114] U=WA+B

[0115] Here, the vector W is called a weight matrix, and the vector B is called a bias vector.

[0116] The intermediate data U is nonlinearly transformed using an activation function to obtain the intermediate layer data M. For example, when ReLU is used as the activation function, the intermediate layer data M is obtained by the following formula.

[0117] [Number 4]

[0118] M = max(0, U)

[0119] Here, the max function takes the maximum value between 0 and U. It should be noted that an activation function such as LeakyReLU can also be used.

[0120] In this way, by performing linear transformation and nonlinear transformation, each data of the neural network can be calculated and obtained. Here, an example of calculating intermediate data 916 based on input data 915 is shown, but the calculation from intermediate data 916 to intermediate data 916 can also be performed in the same way. In the calculation from intermediate layer 911 to output layer 912, tanh function can be used as an activation function. Finally, output layer 912 has an expected value standardized to between -1 and 1.

[0121] The operation of the neural network can be performed by the operation unit 703 of the information processing device 700. The parameters of the neural network 91 need to be adjusted in advance by some method. As an adjustment method, for example, a network parameter adjustment method based on reinforcement learning can be used. The network parameter adjustment method here is called "learning". The following is an explanation of the information processing method performed by the information processing device 700.

[0122] In the learning phase (learning process), the inference model for outputting the control parameters of the substrate chuck 3 to be provided to the control unit 100 is learned. In the learning phase, supervised learning is used. Based on the data obtained by the exposure device 200, the parameters of the neural network are adjusted so that the output obtained for a certain input is the same as the output calculated using the neural network 91 and the output obtained by the exposure device.

[0123] The calculation unit 703 follows Fig.12 , implements learning of the neural network 91. In S1701, the operation unit 703 changes the neural network parameters to initial values. Specifically, the operation unit 703 sets the vector W (weight matrix) and the bias vector B to zero.

[0124] In S1702, the acquisition unit 702 of the information processing device 700 acquires data when the substrate 2 is transferred from the substrate chuck 3 to the pin member 9 from the exposure device 200. The data acquired from the exposure device 200 is called learning data. The learning data is composed of input data A'1, A'2, ..., A' N and training data O'1, O'2, ..., O' O The learning data is used for learning the neural network 91. The learning data is stored in the storage unit 707. In the calculation of the neural network 91, it is desirable that the values ​​of the input layer or the output layer are standardized between -1 and +1. Therefore, in one example, the learning data can be set to a value obtained by dividing by the maximum value that the device can take.

[0125] In one example, input data of the learning data may include at least any one of the following.

[0126] A'1: estimated value of adhesion,

[0127] A'2: The third interval ( Figure 8 ) is the gas supply amount,

[0128] A'3: Gas supply time in the third interval,

[0129] A'4: Driving speed of the fine movement stage 8 in the Z direction in the fourth section

[0130] In addition, the training data of the learning data can include the following.

[0131] O'1: the amount of change in the measured value of the pressure sensor 20 (ie, the pressure in the space between the substrate 2 and the substrate chuck 3) between the third section and the fourth section,

[0132] O'2: The difference between the Z deviation representative value and the Z deviation reference value in the fourth interval

[0133] The above input data and training data are regarded as one set of data. It should be noted that the acquired data set can be one or multiple sets. When multiple sets are acquired, the learning accuracy is improved.

[0134] Provide additional information on input and output data.

[0135] Adhesion force acts between the substrate 2 and the substrate chuck 3. The adhesion force can be estimated using at least one of the characteristics of the substrate 2, the characteristics of the substrate chuck 3, and the characteristics of the exposure process. The estimated value as information on the adhesion force is given as A'1.

[0136] A'2 and A'3 are data related to the supply of gas in the third interval for reducing the adhesion between the substrate 2 and the substrate chuck 3. A'2 is data related to the supply amount of gas. The data related to the supply amount of gas can be, for example, the flow rate of the gas or the pressure value of the gas. Since the pressure and flow rate of the gas are proportional, both can be data related to the supply amount of the gas. A'3 is data related to the supply time of the gas. A'4 is data related to the speed when the substrate 2 and the substrate chuck 3 are separated.

[0137] O'1 and O'2 as training data represent information related to the separation result of the substrate 2 and the substrate chuck 3. O'1 can be an indicator indicating whether the amount of gas supplied between the substrate 2 and the substrate chuck 3 is excessive. In the case where the gas supply is excessive, Fig.10As shown in FIG. 1 , the substrate 2 is separated from the substrate chuck 3, so the rising pressure drops sharply, and O'1 shows a value close to atmospheric pressure. In addition, if the supply amount shown in A'2 is too small or the supply time shown in A'3 is too short relative to A'1 indicating the adhesion, the pressure drop disappears. In this way, from the perspective of the supply amount of gas, there is a correlation between A'1, A'2, A'3 and O'1.

[0138] O'2 is information about the control deviation of the drive unit and is an indicator of the residual adhesion after the gas is supplied. The greater the residual adhesion, the more likely the substrate chuck 3 is to be worn. When the data A'2 and A'3 related to the gas supply are small relative to the adhesion, the residual adhesion becomes larger. At this time, if the driving speed shown in A'4 is fast, the value of O'2 becomes larger. If the driving speed shown in A'4 is slow, the value of O'2 becomes smaller. In this way, there is a correlation between A'1, A'2, A'3, A'4 and O'2.

[0139] The driving speed of the fine motion stage 8 in the Z direction in the fourth interval is set as A'4 because, in the case of residual adhesion between the substrate 2 and the substrate chuck 3 after gas supply, the driving speed has the greatest influence on the wear of the substrate chuck 3. The influence is small, but other parameters related to the drive, such as acceleration and jerk, may also be set as input data after A'5.

[0140] As O'1, the change in the measured value of the pressure sensor 71 (i.e., the pressure between the substrate 2 and the pin member 9) may be used. This is because when the gas is oversupplied, the substrate 2 moves away from the pin member 9, and the pressure sensor 71 shows a value close to the atmospheric pressure, which indicates oversupply.

[0141] As output data, only one of O'1 and O'2 may be used. Although the accuracy of the control parameter when the substrate is unloaded is reduced, the calculation time and learning time of the neural network can be shortened.

[0142] It should be noted that, in S1702 described above, the information processing device 700 acquires the learning data from the exposure device 200 , but the learning data to be acquired may be accumulated in advance in the storage unit 707 and the learning data may be acquired from the storage unit 707 .

[0143] In S1703, the operation unit 703 performs reward calculation based on the learning data obtained in S1702. For example, the operation unit 703 randomly selects a set of data from the data obtained in S1702. The operation unit 703 inputs the learning data obtained in S1702 to the input data 915 of the neural network 91. Specifically, A'1 is input to A1, A'2 is input to A2, A'3 is input to A3, and A'4 is input to A4. The operation unit 703 performs calculations on the neural network and calculates the output data O1 and O2. The reward is obtained based on the difference between the outputs O1 and O2 obtained by the calculation of the neural network and the training data O'1 and O'2 obtained from the exposure device 200. The smaller the difference, the higher the reward.

[0144] In S1704, the calculation unit 703 changes the neural network 91 (that is, updates the parameters of the neural network 91) so that the reward becomes higher. Specifically, the vector W (weight matrix) and the bias vector B are updated.

[0145] By repeating S1703 and S1704 , the results of O1 and O2 calculated when the input data A1 to A4 are given to the neural network 91 can be made close to the training data O′ 1 and O′ 2 obtained from the exposure device 200 .

[0146] In S1705, the operation unit 703 determines whether the number of learning data used in the learning has reached a specified number. The specified number can be input via the input unit 706 at the start of the learning, or can be pre-set in the processing unit 701. In one example, the specified number is the same as the number of learning data obtained from the exposure device 200. In another example, the specified number is a number greater than the number of learning data obtained from the exposure device 200. In yet another example, the specified number is a number less than the number of learning data obtained from the exposure device 200. When the specified number is set to a number greater than the number of learning data obtained from the exposure device 200, in S1703, learning is repeatedly performed using the same learning data. Generally, when the learning data is increased, the accuracy of the learning result becomes higher. In S1705, when the number of learning data used in the learning has not yet reached the specified number, the processing returns to S1703 and the learning is repeated. In S1705, when the number of learning data used in the learning reaches the designated number, the operation unit 703 stores the parameters of the neural network 91 in the storage unit 707, thereby ending the learning phase.

[0147] In addition, in S1702, the evaluation data can be obtained relatively independently from the learning data, and the evaluation process can be implemented after the number of learning data used for learning reaches the specified number in S1705. In the evaluation process, the operation unit 703 determines whether the learning is qualified or not. In the evaluation process, the operation unit 703 calculates the reward based on the output obtained by putting the evaluation data into the neural network 91 and the output data for evaluation. If the reward obtained is higher than the predetermined value, it is determined that the learning is successful. If the reward obtained is below the predetermined value, it is determined that the learning should continue.

[0148] Next, the estimation (inference) stage using the learned neural network is described. In the estimation stage, the information processing device 700 uses the learned neural network based on the input data from the exposure device 200 to calculate the control parameters when the substrate is unloaded, and provides the calculated parameters to the exposure device 200.

[0149] The following is an overview of the estimation stage. The information processing device 700 obtains A1, which is the input data of the neural network 91, from the exposure device 200. The calculation unit 703 sets A2, A3, and A4 to temporary values, performs calculations on the neural network 91, and obtains O1 and O2, which are output data of the neural network 91. The calculation unit 703 searches for a combination of A2, A3, and A4 that produces the most preferred results for the output data O1 and O2.

[0150] Reference Fig.13 , the process in the estimation phase is described. In the estimation phase, the control parameters when the substrate is unloaded are optimized using the neural network learned in the learning phase.

[0151] In S1801, the calculation unit 703 initializes the control parameters. Specifically, A2, A3, and A4, which are input data of the neural network 91, are initialized to zero.

[0152] In S1802, the acquisition unit 702 acquires parameters related to the adhesion of A1 put into the neural network 91 from the exposure device 200. The information of A1 acquired from the exposure device 200 may be information learned in the past or information not learned.

[0153] In S1803, the calculation unit 703 updates the control parameter. Specifically, the calculation unit 703 changes at least one of A2, A3, and A4.

[0154] In S1804, the calculation unit 703 performs calculations of the neural network 91 based on the input data A1 to A4, and calculates O1 and O2.

[0155] By repeatedly performing S1803 and S1804, the parameter values ​​of A2, A3, and A4 are changed to obtain output data O1 and O2. At this time, the parameter value is searched in such a way that the output data O1 and O2 are close to 0. O1 represents the reduction in the measured value of the pressure sensor 20 in the 3rd interval and the 4th interval (i.e., the pressure in the space between the substrate 2 and the substrate chuck 3). The reduction of 0 indicates that there is no lateral slippage between the substrate 2 and the substrate chuck 3. In addition, O2 is the difference between the Z deviation representative value and the Z deviation reference value in the 4th interval. The closer to 0, the smaller the adhesion and the smaller the wear of the substrate chuck 3.

[0156] By finding O1 and O2 as described above, it is possible to prevent the substrate 2 from sliding laterally relative to the substrate chuck 3 and prevent the substrate 2 from being separated from the substrate chuck 3 and causing the substrate chuck 3 to wear. Figure 5 The effect of the optimal point shown in (b) of FIG. 1 is shown. By using a neural network, even for unknown adhesion that is not calculated by the simulator, it is possible to provide the optimal parameters when the substrate is unloaded. The parameters provided are obtained by learning actual data, so the reliability of the parameters is also high.

[0157] As a result of the search, when a plurality of parameter value groups of A2, A3, and A4 are found in which O1 and O2 are equal, a parameter value group that has the highest productivity of the exposure device 200 may be selected from them. A3 and A4 contribute to productivity. Therefore, A3 with a short supply time may be selected, or A4 with a fast drive speed may be selected. Alternatively, the time of both A3 and A4 may be combined to select a combination with the shortest substrate unloading time.

[0158] In S1805, the operation unit 703 determines whether the parameter search for A2, A3, and A4 is completed. If the parameter search is not completed, the process returns to S1803 and the search is repeated. In the case where the parameter search is completed, the process proceeds to S1806. In S1806, the supply unit 704 supplies the determined parameters A2, A3, and A4 to the exposure device 200 as control parameters.

[0159] <Second Embodiment>

[0160] In the first embodiment described above, an estimated value of the adhesion between the substrate 2 and the substrate chuck 3 is given as an input to the neural network 91. In the second embodiment, instead of the estimated value of the adhesion, characteristics of the substrate 2, characteristics of the substrate chuck 3, and characteristics of the exposure process, which are factors of the adhesion, are given as input. In the second embodiment, the input data of the learning data can include at least any one of the following.

[0161] A1: The warpage of substrate 2,

[0162] A2: Roughness of the back side of the substrate 2 (the side opposite to the holding side of the substrate chuck), A3: Thickness of the substrate 2,

[0163] A4: The area or diameter of the front end surface of the chuck pin (holding member) of the substrate chuck 3 that contacts the substrate,

[0164] A5: Material of substrate chuck 3,

[0165] A6: Exposure time,

[0166] A7: Exposure

[0167] A8: The amount of gas supplied in the third zone.

[0168] A9: Gas supply time in the third section,

[0169] A 10 : The driving speed of the fine-motion stage 8 in the Z direction in the fourth section

[0170] In addition, the training data of the learning data can include the following.

[0171] O1: the amount of change in the measured value of the pressure sensor 20 (i.e., the pressure in the space between the substrate 2 and the substrate chuck 3) between the third section and the fourth section,

[0172] O2: Difference between Z deviation representative value and Z deviation reference value in the 4th interval

[0173] The above input data and training data are regarded as 1 set of data.

[0174] Similar to the first embodiment, the neural network 91 is learned in the learning phase, and the control parameters for unloading the substrate are estimated using the learned neural network 91 in the estimation phase.

[0175] By setting the characteristics of the substrate 2, the characteristics of the substrate chuck 3, and the characteristics of the exposure process, which are factors of the adhesion force, as inputs to the neural network 91, rather than the estimated value of the adhesion force, the adhesion force estimation process can be included in the neural network. As a result, the control parameters when the substrate is unloaded can be optimized. In addition, even in a combination of different characteristics that have not been processed by the exposure device 200, the control parameters when the substrate is unloaded can be optimized.

[0176] In the first and second embodiments, the configuration in which the gas flow path for supplying gas is one system has been described, but the gas flow path may be two systems. By having the gas flow path be two systems, the supply of gas can be controlled more finely.

[0177] Furthermore, the processing unit 701 may display a list of input parameters and output results on the display unit 705 and allow the user to select parameters to be supplied to the exposure device 200 via the input unit 706. At this time, the processing unit 701 may cause the display unit 705 to highlight the recommended settings.

[0178] In addition, the information processing device 700 can also be connected to multiple exposure devices in a communicative manner, obtain learning data from each of the multiple exposure devices, and supply optimal parameters for substrate unloading to each of the multiple exposure devices. By collecting data from multiple exposure devices, learning can be performed based on a large amount of data. The characteristics of the substrate chuck are different for each device, and information on substrate chucks under various conditions can be included in the learning data. The substrate chuck is sometimes replaced after wear. Even if the exposure device is equipped with a new substrate chuck, due to the existence of learning data in other exposure devices, the optimal control parameters for substrate unloading can be provided to the exposure device.

[0179] In a mass production factory of semiconductor devices, a specific number of substrates (e.g., 25 substrates) that have undergone the same manufacturing process are usually managed as one batch. Therefore, the exposure processing task is put into batch units. The information processing device 700 can also provide the optimal parameters for substrate unloading in batch units. This is because it is assumed that the characteristics of the substrate 2 used in the estimation of adhesion, the characteristics of the substrate chuck 3, and the characteristics of the exposure process are roughly the same within the batch.

[0180] The information processing device 700 may also be connected to an external storage unit (not shown) of a larger capacity to store the learning data acquired from the exposure device 200. Thus, more information is stored in the external storage unit, and the information processing device 700 can use a large amount of information for learning. As a result, the performance of the generated (updated) neural network 91 is improved.

[0181] <Implementation Method of Article Manufacturing Method>

[0182] The article manufacturing method in the embodiment of the present invention is suitable for manufacturing micro devices such as semiconductor devices, components with fine structures and other articles. The article manufacturing method of this embodiment may include: a formation step, using the above-mentioned substrate processing device to form an original pattern on a substrate; and a processing step, processing the substrate with a pattern formed in the formation step. In addition, the article manufacturing method may include other well-known steps (oxidation, film formation, evaporation, doping, flattening, etching, resist stripping, cutting, bonding, packaging, etc.). Compared with previous methods, the article manufacturing method of this embodiment is advantageous in at least one of the performance, quality, productivity, and production cost of the article.

[0183] Other embodiments

[0184] The embodiments of the present invention may also be implemented by a computer of a system or device, which reads and executes computer executable instructions (e.g., one or more programs) recorded on a storage medium (which may also be more comprehensively referred to as a "non-volatile computer-readable storage medium") to perform one or more functions of the above-mentioned embodiments and / or includes one or more circuits (e.g., application-specific integrated circuits (ASICs)) for performing one or more functions of the above-mentioned embodiments, and by a method executed by a computer of a system or device, which, for example, reads and executes computer executable instructions from a storage medium to perform one or more functions of the above-mentioned embodiments and / or controls one or more circuits to perform one or more functions of the above-mentioned embodiments. The computer may include one or more processors (e.g., a central processing unit (CPU), a microprocessing unit (MPU)), and may include a network of multiple independent computers or independent processors for reading and executing computer executable instructions. The computer executable instructions may be provided to the computer, for example, from a network or a storage medium. The storage medium may include, for example, one or more hard disks, random access memory (RAM), read-only memory (ROM), storage of a distributed computing system, optical disks (e.g., compact disks (CDs), digital versatile disks (DVDs), Blu-ray disks (BDs)) and other storage media. TM ), flash memory devices, memory cards, etc.

[0185] Other embodiments

[0186] The embodiments of the present invention may also be implemented by providing software (program) for performing the functions of the above-described embodiments to a system or device via a network or various storage media, and a computer or a central processing unit (CPU) or a microprocessing unit (MPU) of the system or device reads and executes the program.

[0187] The present invention is not limited to the above-described embodiments, and various changes and modifications can be made without departing from the spirit and scope of the invention. Therefore, the appended claims are intended to disclose the scope of the invention.

Claims

1. An information processing device, characterized in that: outputting a control parameter, the control parameter being a control parameter related to supply of gas between the object and the holding portion when the object is separated from the holding portion holding the object, wherein the information processing device has: an output unit, which outputs the control parameter based on the inference model; and The machine learning unit performs machine learning of the inference model based on a result of separation of the object and the holding unit when the gas is supplied based on the control parameter.

2. The information processing device according to claim 1, characterized in that The machine learning unit uses information related to the adhesion between the object and the holding unit and information related to the supply of the gas as input data, and performs machine learning using information related to the separation result as training data.

3. The information processing device according to claim 2, characterized in that: The information related to the supply of the gas includes information on at least one of the supply amount of the gas and the supply time of the gas.

4. The information processing device according to claim 1, characterized in that The machine learning unit uses the characteristics of the object and the characteristics of the holding unit as input data and performs machine learning using information related to the separation result as training data.

5. The information processing device according to claim 4, characterized in that The characteristic of the object includes information on at least any one of a warpage amount of the object, roughness of a surface of the object, and thickness of the object.

6. The information processing device according to claim 4, characterized in that The holding portion includes a support member configured to be able to appear and disappear relative to a holding surface holding the object, and to support the object by protruding upward from the holding surface. The characteristic of the holding portion includes information on at least one of an area or a diameter of a front end surface of the supporting member that comes into contact with the object, and a material of the holding portion.

7. The information processing device according to claim 4, characterized in that: The holding portion is configured to hold the substrate as the object in an exposure device that exposes the substrate. The input data also include characteristics of the exposure process in the exposure device.

8. The information processing device according to claim 7, characterized in that: The characteristics of the exposure process include information on at least one of exposure time and exposure amount of the substrate.

9. The information processing device according to claim 2, characterized in that: The holding portion includes a support member configured to be able to appear and disappear relative to a holding surface holding the object, and to support the object by protruding upward from the holding surface. The information related to the result of the separation as the training data includes information on the amount of change in the pressure between the object and the supporting member.

10. The information processing device according to claim 2, characterized in that: The input data further includes a driving speed of a driving unit that performs the separation operation, and the information related to the result of the separation as the training data includes information on a control deviation of the driving unit.

11. The information processing device according to claim 1, characterized in that: The inference model is generated by a neural network. 12 . A computer-readable storage medium storing a program for causing a computer to function as each unit of the information processing device according to claim 1 . 13 . A computer program product comprising a program for causing a computer to function as each unit of the information processing device according to claim 1 .

14. An information processing method, characterized in that: Outputting a control parameter, the control parameter being a control parameter related to supply of gas between the object and the holding portion when the object is separated from the holding portion holding the object, wherein the information processing method comprises: An output step of outputting the control parameters based on the inference model; and The machine learning step performs machine learning of the inference model based on a result of separation of the object and the holding portion when the gas is supplied based on the control parameter.

Images